Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ggml: allow passing alloc dependencies in graph_optimize (#27301)

Browse Source 
* ggml: allow passing alloc dependencies in graph_optimize

* add alloc dep tests

* add TODO about using flat array
This commit is contained in:
Aman Gupta
2026-08-30 11:34:20 +08:00
committed by GitHub
parent c589f0ed10
commit 57291f2644
8 changed files with 151 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/ggml-cuda.cu
+3 -1
View File
@@ -4328,7 +4328,9 @@ static void ggml_backend_cuda_event_wait(ggml_backend_t backend, ggml_backend_ev
}
}
static void ggml_backend_cuda_graph_optimize(ggml_backend_t backend, ggml_cgraph * cgraph) {
static void ggml_backend_cuda_graph_optimize(ggml_backend_t backend, ggml_cgraph * cgraph, ggml_backend_graph_optimize_params * params) {
GGML_UNUSED(params);
ggml_backend_cuda_context * cuda_ctx = (ggml_backend_cuda_context *) backend->context;
#ifdef USE_CUDA_GRAPH