ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

HIP/CUDA: set the paramerter value in maintain_cuda_graph instead of replaceing it. (#12209)

Browse source 
This avoids conflict with internal cuda/hip runtimes memory managment behavior.
This commit is contained in:
uvos 2025-03-06 08:20:52 +01:00 • committed by GitHub
parent 57b6abf85a
commit e721c05c93
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml/src/ggml-cuda/ggml-cuda.cu
View file

@ -2571,7 +2571,7 @@ static void maintain_cuda_graph(ggml_backend_cuda_context * cuda_ctx, std::vecto
for (size_t i = 0; i < cuda_ctx->cuda_graph->num_nodes; i++) {
if(count(ggml_cuda_cpy_fn_ptrs.begin(), ggml_cuda_cpy_fn_ptrs.end(), cuda_ctx->cuda_graph->params[i].func) > 0) {
char ** updated_kernel_arg_ptr = cuda_ctx->cuda_graph->updated_kernel_arg.at(k++);
cuda_ctx->cuda_graph->params[i].kernelParams[1] = updated_kernel_arg_ptr;
*(void**)cuda_ctx->cuda_graph->params[i].kernelParams[1] = *(void**)updated_kernel_arg_ptr;
CUDA_CHECK(cudaGraphKernelNodeSetParams(cuda_ctx->cuda_graph->nodes[i], &cuda_ctx->cuda_graph->params[i]));
}
}