]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 3b15924)
vulkan: Use larger loads in scalar/coopmat1 matmul (#15729)
authorJeff Bolz <redacted>
Sun, 7 Sep 2025 16:53:07 +0000 (11:53 -0500)
committerGitHub <redacted>
Sun, 7 Sep 2025 16:53:07 +0000 (18:53 +0200)
commit267e99867f09bec8bcc2e424ad9bcddd6cccf9d0
treebb7e766ff89351367b2a37df4554db770a639d6ctree
parent3b15924d71237a43bb5ad71f5b885ee66a821342commit | diff
vulkan: Use larger loads in scalar/coopmat1 matmul (#15729)

I think glslang will translate an access like x[i][1].z to
OpAccessChain ... x, i, 1, 2
OpLoad float16_t ...

rather than loading all of x[i] in a single OpLoad. Change the
code to explicitly load the vector/matrix.
ggml/src/ggml-vulkan/vulkan-shaders/mul_mm.comp diff | blob | history
ggml/src/ggml-vulkan/vulkan-shaders/types.comp diff | blob | history
ggml/src/ggml-vulkan/vulkan-shaders/vulkan-shaders-gen.cpp diff | blob | history
Packaging of ggml-org/llama.cpp