]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
CUDA: refactor mma data loading for AMD (#22051)
authorJohannes Gäßler <redacted>
Sun, 19 Apr 2026 16:26:59 +0000 (18:26 +0200)
committerGitHub <redacted>
Sun, 19 Apr 2026 16:26:59 +0000 (18:26 +0200)
commit4eac5b45095a4e8a1ff1cce4f6d030e0872fb4ad
tree1f141933d4ac0fcd72f20918c3e933748489900c
parentd5b780a676b755ec8f550e9b633256d2dea8b8c5
CUDA: refactor mma data loading for AMD (#22051)

* CUDA: refactor mma data loading for AMD

* fix CDNA MMQ occupancy

* fix CDNA3 mma

* fix RDNA3 compile
ggml/src/ggml-cuda/common.cuh
ggml/src/ggml-cuda/fattn-mma-f16.cuh
ggml/src/ggml-cuda/mma.cuh
ggml/src/ggml-cuda/mmq.cuh