]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
overview / pkg / ggml / sources / ggml / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 65f7889)
CUDA: faster non-contiguous concat (llama/10760)
authora3sh <redacted>
Thu, 12 Dec 2024 18:09:50 +0000 (02:09 +0800)
committerGeorgi Gerganov <redacted>
Tue, 17 Dec 2024 17:23:40 +0000 (19:23 +0200)
commit57b412fce0f2b3808939ae832601a6c8a0377374
tree845385f5b060269fd4189935594f678d0b9818a5tree
parent65f788911b1630857108694c7a6fe5bb5fa732d7commit | diff
CUDA: faster non-contiguous concat (llama/10760)

* faster uncontiguous concat

* Use a lambda to avoid code duplication

Co-authored-by: Diego Devesa <redacted>
* Update ggml/src/ggml-cuda/concat.cu

* add constexpr  and static assert

---------

Co-authored-by: Diego Devesa <redacted>
src/ggml-cuda/concat.cu diff | blob | history
Packaging of ggml-org/ggml