Revert "Fix race in the CUDA DeepSeek FA kernel (#406)"

This reverts commit 36e6e888b75ae93fb5aac212bb0e147d8379ae23. I should have tested. We get NaNs.
2025-05-11 12:22:19 +03:00 · 2025-05-11 12:22:19 +03:00 · 8f7bd74afb
parent 0abcf0749e
commit 8f7bd74afb
1 changed files with 0 additions and 2 deletions
--- a/ggml/src/ggml-cuda/fattn-new-mma.cu
+++ b/ggml/src/ggml-cuda/fattn-new-mma.cu
@ -898,8 +898,6 @@ static __device__ __forceinline__ void flash_attn_ext_f16_process_tile(
            KQ_crs += __shfl_xor_sync(0xFFFFFFFF, KQ_crs, offset, WARP_SIZE);
        }

-        __syncthreads();
-
        // Write back combined meta data:
 #pragma unroll
        for (int imeta = 0; imeta < nmeta; ++imeta) {