CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-11 10:36:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

CUDA: fix MMQ stream-k fixup ne1 indices (#17089)

Browse Source
This commit is contained in:
Johannes Gäßler
2025-11-08 08:26:18 +01:00
committed by GitHub
parent 647b960bd8
commit e14e842e87
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/mmq.cuh
View File

@@ -3494,7 +3494,7 @@ static __global__ void mul_mat_q_stream_k_fixup(
const int col_diff = col_high - col_low; const int col_diff = col_high - col_low;
for (int j = threadIdx.y*warp_size + threadIdx.x; j < mmq_x; j += nwarps*warp_size) { for (int j = threadIdx.y*warp_size + threadIdx.x; j < mmq_x; j += nwarps*warp_size) {
ids_dst_shared[j] = ids_dst[col_low + j]; ids_dst_shared[j] = ids_dst[col_low + jt*mmq_x + j];
} }
__syncthreads(); __syncthreads();