]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
CUDA: batch out_prod inner loop with cublasSgemmStridedBatched (#22651)
authorleonardHONG <redacted>
Thu, 7 May 2026 19:59:29 +0000 (03:59 +0800)
committerGitHub <redacted>
Thu, 7 May 2026 19:59:29 +0000 (21:59 +0200)
commit05ff59cb57860cc992fc6dcede32c696efea711c
tree98b789f0283b2afa3509e6f8e31012142b5ea49b
parentaaf4a4d5e08222404a8e92ddf4147615957b67f7
CUDA: batch out_prod inner loop with cublasSgemmStridedBatched (#22651)

* CUDA: batch out_prod inner loop with cublasSgemmStridedBatched

* CUDA: batch out_prod inner loop with cublasSgemmStridedBatched

* CUDA: add cublasSgemmStridedBatched mapping for HIP and MUSA backends
ggml/src/ggml-cuda/out-prod.cu
ggml/src/ggml-cuda/vendors/hip.h
ggml/src/ggml-cuda/vendors/musa.h
tests/test-backend-ops.cpp