mozempk/ggml-org-llama-cpp-mirror
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-10-11 07:20:33 +02:00
Code Issues Packages Projects Releases Wiki Activity

Avoid races in fp8_fallback

Browse Source
This commit is contained in:
Oliver Simons
2026-10-09 21:11:12 +02:00
parent d34e22aac6
commit b757f0d5cf
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/fp8.cu
+1
View File
@@ -24,6 +24,7 @@ static __global__ void mul_mat_fp8_fallback(
if (threadIdx.x == 0) { if (threadIdx.x == 0) {
*(float *) (dst + i0*sizeof(float) + i1*nb1 + i2*nb2 + i3*nb3) = sum; *(float *) (dst + i0*sizeof(float) + i1*nb1 + i2*nb2 + i3*nb3) = sum;
} }
__syncthreads();
} }
} }