]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: fbef0fa)
cuda: Add cublasLt_static linking when GGML_STATIC is enabled (#15622)
authormatiaslin <redacted>
Thu, 28 Aug 2025 00:32:36 +0000 (17:32 -0700)
committerGitHub <redacted>
Thu, 28 Aug 2025 00:32:36 +0000 (02:32 +0200)
commit5a0e3ef6f00c658fbae53797f02d5a360ebf8fec
tree2aaa9d00949973798169c647793645e50457f53btree
parentfbef0fad7a7c765939f6c9e322fa05cd52cf0c15commit | diff
cuda: Add cublasLt_static linking when GGML_STATIC is enabled (#15622)

Prior to this change, we faced undefined cublasLt references when
attempting to compile 'llama-cli' with GGML_STATIC=ON on Linux.

We add linking with CUDA::cublasLt_static when CUDA version is greater
than 10.1.
ggml/src/ggml-cuda/CMakeLists.txt diff | blob | history
Packaging of ggml-org/llama.cpp