vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-10-03 11:35:46 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

CUDA: add a reserve to avoid spurious warning on older GCC builds (#29317)

Browse source
This commit is contained in:
Aman Gupta 2026-09-24 00:52:40 +08:00 • committed by GitHub
parent bddf8263c3
commit 66fba63af1
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 2 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

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

@ -3470,6 +3470,7 @@ static int ggml_cuda_try_fuse(ggml_backend_cuda_context * cuda_ctx, ggml_cgraph
ggml_cuda_topk_moe_args args;
const bool can_fuse = ggml_cuda_topk_moe_fusion(cgraph, i, args);
std::vector<ggml_op> ops;
ops.reserve(13); // max ops; avoids gcc -Wstringop-overflow false positive
if (can_fuse) {
const ggml_tensor * logits = node->src[0];
@ -4539,6 +4540,7 @@ static void ggml_backend_cuda_graph_optimize(ggml_backend_t backend, ggml_cgraph
ggml_cuda_topk_moe_args args;
const bool can_fuse = ggml_cuda_topk_moe_fusion(cgraph, i, args);
std::vector<ggml_op> ops;
ops.reserve(13); // max ops; avoids gcc -Wstringop-overflow false positive
const ggml_tensor * node = cgraph->nodes[i];