]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 8eff955)
CUDA: add set rows for f32 and f16 (#14551) upstream/0.0.5882
authorAman Gupta <redacted>
Sat, 12 Jul 2025 13:31:38 +0000 (21:31 +0800)
committerGitHub <redacted>
Sat, 12 Jul 2025 13:31:38 +0000 (16:31 +0300)
commit7de5c7cab61d4da4387ed9b216f88b96297bcc2d
treeca2ffcf34bf44e2c398c93bb717aedd8a91e99fftree
parent8eff95544e817704d44bec20f9fc956ce76a33becommit | diff
CUDA: add set rows for f32 and f16 (#14551)

* CUDA: add set rows for f32 and f16

* Review: change kernel params, use strides from host

* Use 1-d kernel

* Review: use int64_t for blockDim.x, rename nb->s for clarity
ggml/src/ggml-cuda/ggml-cuda.cu diff | blob | history
ggml/src/ggml-cuda/set-rows.cu [new file with mode: 0644] blob
ggml/src/ggml-cuda/set-rows.cuh [new file with mode: 0644] blob
Packaging of ggml-org/llama.cpp