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

fix out-of-bounds write

Browse source
This commit is contained in:
Johannes Gäßler 2024-06-03 22:10:28 +02:00
parent 158e3d3e4f
commit bd8422dbe7
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda/mmq.cuh
View file

@ -1124,7 +1124,7 @@ static __global__ void mul_mat_q(
for (int i0 = 0; i0 < mmq_y; i0 += WARP_SIZE) {
const int i = blockIdx.x*mmq_y + i0 + threadIdx.x;
if (need_check && i > ne0) {
if (need_check && i >= ne0) {
continue;
}