]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 3025b62)
vulkan: coopmat2 mul_mat optimizations (#14934)
authorJeff Bolz <redacted>
Sat, 2 Aug 2025 09:21:37 +0000 (04:21 -0500)
committerGitHub <redacted>
Sat, 2 Aug 2025 09:21:37 +0000 (11:21 +0200)
commit4cb208c93c1c938591a5b40354e2a6f9b94489bc
tree2f3ed812c0a13fc0fd1be3a6d60957005a25cbbftree
parent3025b621d12a6931ff5e9775d4f644719980ad91commit | diff
vulkan: coopmat2 mul_mat optimizations (#14934)

- Increase tile size for k-quants, to match non-k-quants
- Choose more carefully between large and medium tiles, considering how it
  interacts with split_k
- Allow larger/non-power of two split_k, and make the splits a multiple of 256
- Use split_k==3 to when >1/2 and <=2/3 of the SMs would hae been used
ggml/src/ggml-vulkan/ggml-vulkan.cpp diff | blob | history
Packaging of ggml-org/llama.cpp