root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-26 03:14:35 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

added missing CUDA_CHECKs

Browse Source
This commit is contained in:
Alan Gray 2024-04-24 02:37:57 -07:00
parent df4719ec7e
commit c3d4ead136
1 changed files with 2 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
ggml-cuda.cu
View File

@ -2440,12 +2440,11 @@ GGML_CALL static enum ggml_status ggml_backend_cuda_graph_compute(ggml_backend_t
// kernel parameters which need updated in the graph for each token // kernel parameters which need updated in the graph for each token
void* ggmlCudaCpyFn = nullptr; void* ggmlCudaCpyFn = nullptr;
if(cudaGraph.count==0){ if(cudaGraph.count==0){
cudaDeviceProp prop; cudaDeviceProp prop;
int device; int device;
cudaGetDevice(&device); CUDA_CHECK(cudaGetDevice(&device));
cudaGetDeviceProperties(&prop, device); CUDA_CHECK(cudaGetDeviceProperties(&prop, device));
if (prop.major < 8){ if (prop.major < 8){
cudaGraph.disableDueToGpuArch=true; cudaGraph.disableDueToGpuArch=true;
} }