]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: ba42794)
ggml : Q2k interleaving implementation - x86/x64 SIMD (#14373)
authorSrihari-mcw <redacted>
Fri, 1 Aug 2025 06:20:33 +0000 (11:50 +0530)
committerGitHub <redacted>
Fri, 1 Aug 2025 06:20:33 +0000 (09:20 +0300)
commitbaad94885df512bb24ab01e2b22d1998fce4d00e
treeaa749ff9368c2bf5fbf2e4fba1aa746b3f2619ebtree
parentba42794c9ead96ad52311ba1b23eefcbf3d6f63dcommit | diff
ggml : Q2k interleaving implementation - x86/x64 SIMD (#14373)

* Initial Q2_K Block Interleaving Implementation

* Addressed review comments and clean up of the code

* Post rebase fixes

* Initial CI/CD fixes

* Update declarations in arch-fallback.h

* Changes for GEMV Q2_K in arch-fallback.h

* Enable repacking only on AVX-512 machines

* Update comments in repack.cpp

* Address q2k comments

---------

Co-authored-by: Manogna-Sree <redacted>
ggml/src/ggml-cpu/arch-fallback.h diff | blob | history
ggml/src/ggml-cpu/arch/x86/repack.cpp diff | blob | history
ggml/src/ggml-cpu/repack.cpp diff | blob | history
ggml/src/ggml-cpu/repack.h diff | blob | history
Packaging of ggml-org/llama.cpp