]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 43e8995)
ggml: bypass code incompatible with CUDA < 11.1 (whisper/2020)
authorSlava Primenko <redacted>
Thu, 4 Apr 2024 12:49:24 +0000 (14:49 +0200)
committerGeorgi Gerganov <redacted>
Sun, 7 Apr 2024 14:05:40 +0000 (17:05 +0300)
commitf77261a7c525fa1fa47b18a3d78cd308ae41cafc
tree1299ee567a8ddde7e3f723b0996c284fafa98f94tree
parent43e8995e754b8e04642e92822055d193a3272b37commit | diff
ggml: bypass code incompatible with CUDA < 11.1 (whisper/2020)

`cudaHostRegisterReadOnly` parameter was only introduced in CUDA 11.1

See this issue for more details:
https://github.com/ggerganov/examples/whisper/whisper.cpp/issues/2007
ggml-cuda.cu diff | blob | history
Packaging of ggml-org/llama.cpp