Skip to content

CUDA: fix FA tg at long context for CC >= 8.9 #13852

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
May 28, 2025
Merged

CUDA: fix FA tg at long context for CC >= 8.9 #13852

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions ggml/src/ggml-cuda/fattn-common.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -623,8 +623,8 @@ static __global__ void flash_attn_combine_results(
__builtin_assume(tid < D);

extern __shared__ float2 meta[];
if (tid < 2*parallel_blocks) {
((float *) meta)[threadIdx.x] = ((const float *)VKQ_meta) [blockIdx.z*(2*parallel_blocks) + tid];
for (int i = tid; i < 2*parallel_blocks; i += D) {
((float *) meta)[i] = ((const float *)VKQ_meta) [blockIdx.z*(2*parallel_blocks) + i];
}

__syncthreads();
Expand Down
Loading