LLM-Inference/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-08-12 22:29:39 +04:00
Code Issues Packages Projects Releases Wiki Activity

Don't overstep

Browse Source
This commit is contained in:
Kawrakow
2026-08-08 16:54:28 +03:00
parent 764ec4ffe5
commit 50318aa980
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/indexer_topk.cu
+1
View File
@@ -121,6 +121,7 @@ void ggml_cuda_op_indexer_topk(ggml_backend_cuda_context & ctx, ggml_tensor * ds
for (int istep = 0; istep < nstep; ++istep) {
int first_row = max_rows*istep;
if (first_row >= int(q->ne[2])) break;
int last_row = std::min(first_row + max_rows, int(q->ne[2]));
int nrows = last_row - first_row;