vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

limit to GGML_ALLOW_CUDA_GRAPHS defined in llama.cpp cmake

Browse source
This commit is contained in:
Alan Gray 2024-04-25 00:51:48 -07:00
parent 408759687f
commit 0640427f7b
2 changed files with 2 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda.cu
View file

@ -2405,7 +2405,7 @@ GGML_CALL static void ggml_backend_cuda_synchronize(ggml_backend_t backend) {
GGML_UNUSED(backend);
}
#if (CUDART_VERSION >= 12000)
#if (CUDART_VERSION >= 12000) && defined(GGML_ALLOW_CUDA_GRAPHS)
#define USE_CUDA_GRAPH
#endif