]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
overview / pkg / ggml / sources / ggml / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 02f5fdf)
cuda: Add cublasLt_static linking when GGML_STATIC is enabled (llama/15622)
authormatiaslin <redacted>
Thu, 28 Aug 2025 00:32:36 +0000 (17:32 -0700)
committerGeorgi Gerganov <redacted>
Fri, 5 Sep 2025 09:54:06 +0000 (12:54 +0300)
commit1a88e98363f1d8e502cbc34365b802a6a3e5888e
treead353f0da3802b46b62373af179ba939487912ectree
parent02f5fdf0059a0c741adf02fff456038467ced7f1commit | diff
cuda: Add cublasLt_static linking when GGML_STATIC is enabled (llama/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.
src/ggml-cuda/CMakeLists.txt diff | blob | history
Packaging of ggml-org/ggml