Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

CUDA: fix overflow in MMA kernel without stream-k (#17939)

Browse Source
This commit is contained in:
Johannes Gäßler
2025-12-12 17:43:58 +01:00
committed by GitHub
parent 7bed317f53
commit 482211438d
2 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-cuda/fattn-common.cuh
+3 -3
View File
@@ -642,8 +642,8 @@ static __global__ void flash_attn_stream_k_fixup(
const int iter_k = (ne11 + (nbatch_fa - 1)) / nbatch_fa;
const int iter_j = (ne01 + (ncols1 - 1)) / ncols1;
const int kbc0 = (bidx0 + 0)*(iter_k*iter_j*(ne02/ncols2)*ne03) / gridDim.x;
const int kbc0_stop = (bidx0 + 1)*(iter_k*iter_j*(ne02/ncols2)*ne03) / gridDim.x;
const int kbc0 = int64_t(bidx0 + 0)*(iter_k*iter_j*(ne02/ncols2)*ne03) / gridDim.x;
const int kbc0_stop = int64_t(bidx0 + 1)*(iter_k*iter_j*(ne02/ncols2)*ne03) / gridDim.x;
const bool did_not_have_any_data = kbc0 == kbc0_stop;
const bool wrote_beginning_of_tile = kbc0 % iter_k == 0;
@@ -679,7 +679,7 @@ static __global__ void flash_attn_stream_k_fixup(
int bidx = bidx0 - 1;
int kbc_stop = kbc0;
while(true) {
const int kbc = bidx*(iter_k*iter_j*(ne02/ncols2)*ne03) / gridDim.x;
const int kbc = int64_t(bidx)*(iter_k*iter_j*(ne02/ncols2)*ne03) / gridDim.x;
if (kbc == kbc_stop) { // Did not have any data.
bidx--;
kbc_stop = kbc;