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

CUDA: fix shared memory access condition for mmv

Browse source
This commit is contained in:
Johannes Gäßler 2024-12-09 19:29:22 +01:00
parent c37fb4cf62
commit 6768787dd3
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/mmv.cu
View file

@ -57,7 +57,7 @@ static __global__ void mul_mat_vec(
if (block_size > WARP_SIZE) {
buf_iw[tid/WARP_SIZE] = sumf;
__syncthreads();
if (tid > WARP_SIZE) {
if (tid >= WARP_SIZE) {
return;
}
sumf = buf_iw[tid];