]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: f28af0d)
cuda : revert CUDA pool stuff (#3944)
authorslaren <redacted>
Sun, 5 Nov 2023 07:12:13 +0000 (08:12 +0100)
committerGitHub <redacted>
Sun, 5 Nov 2023 07:12:13 +0000 (09:12 +0200)
commit48ade94538fa509465d71023e49d07aab0ec8cd5
tree84e51bff37eed34cf68528d14e0792c69e2fa1e9tree
parentf28af0d81aa1010afa5de74cf627dcb04bea3157commit | diff
cuda : revert CUDA pool stuff (#3944)

* Revert "cuda : add ROCM aliases for CUDA pool stuff (#3918)"

This reverts commit 629f917cd6b96ba1274c49a8aab163b1b189229d.

* Revert "cuda : use CUDA memory pool with async memory allocation/deallocation when available (#3903)"

This reverts commit d6069051de7165a4e06662c89257f5d2905bb156.

ggml-ci
ggml-cuda.cu diff | blob | history
Packaging of ggml-org/llama.cpp