]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 19f4dec)
vulkan: Use larger workgroups for mul_mat_vec when M is small (#15355)
authorJeff Bolz <redacted>
Sun, 17 Aug 2025 16:08:57 +0000 (11:08 -0500)
committerGitHub <redacted>
Sun, 17 Aug 2025 16:08:57 +0000 (18:08 +0200)
commit21c17b5befc5f6be5992bc87fc1ba99d388561df
tree3d3a2fc68f9b7916a222cdec18f9f85c8c64b246tree
parent19f4decae0ead52debe56095ba8d693b4f14e4dfcommit | diff
vulkan: Use larger workgroups for mul_mat_vec when M is small (#15355)

* vulkan: Use larger workgroups for mul_mat_vec when M is small

Also use subgroup instructions for (part of) the reduction when supported.
Without this, the more expensive reductions would eat into the benefits of
the larger workgroups.

* update heuristic for amd/intel

Co-authored-by: 0cc4m <redacted>
---------

Co-authored-by: 0cc4m <redacted>
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
ggml/src/ggml-vulkan/vulkan-shaders/mul_mat_vec_base.comp diff | blob | history
ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp diff | blob | history
Packaging of ggml-org/llama.cpp