]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: e920c52)
CUDA: refactor MMQ kernel configuration (#24127)
authorJohannes Gäßler <redacted>
Mon, 13 Jul 2026 16:37:57 +0000 (18:37 +0200)
committerGitHub <redacted>
Mon, 13 Jul 2026 16:37:57 +0000 (18:37 +0200)
commit6eddde06a4f25d55d538b5d15628dcc2b6882147
tree68891d630bb44d6720bba9e2a0da9a72333bd0c9tree
parente920c523e3b8a0163fe498af5bf90df35ff51d25commit | diff
CUDA: refactor MMQ kernel configuration (#24127)

* CUDA: refactor MMQ kernel configuration

* fix Blackwell config

* remove legacy code
ggml/src/ggml-cuda/mmq-config-ampere.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-config-blackwell.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-config-cdna.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-config-pascal.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-config-rdna2.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-config-rdna4.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-load-tiles.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq-vec-dot.cuh [new file with mode: 0644] blob
ggml/src/ggml-cuda/mmq.cu diff | blob | history
ggml/src/ggml-cuda/mmq.cuh diff | blob | history
Packaging of ggml-org/llama.cpp