fix format

microsoft · Oct 12, 2023 · 6b213eb · 6b213eb
1 parent e393299
commit 6b213eb
Showing 1 changed file with 4 additions and 4 deletions.
diff --git a/onnxruntime/contrib_ops/cuda/bert/flash_attention/flash_fwd_launch_template.h b/onnxruntime/contrib_ops/cuda/bert/flash_attention/flash_fwd_launch_template.h
@@ -3,7 +3,7 @@
  ******************************************************************************/
 #pragma once
 
-#include<stdexcept>
+#include <stdexcept>
 
 #include "contrib_ops/cuda/bert/flash_attention/static_switch.h"
 #include "contrib_ops/cuda/bert/flash_attention/flash.h"
@@ -113,7 +113,7 @@ void run_flash_splitkv_fwd(Flash_fwd_params& params, cudaStream_t stream) {
 
 template <typename T, int Headdim>
 void run_mha_fwd_splitkv_dispatch(Flash_fwd_params& params, cudaStream_t stream) {
-  #if defined(__CUDA_ARCH__) && __CUDA_ARCH__ >= 800
+#if defined(__CUDA_ARCH__) && __CUDA_ARCH__ >= 800
   bool is_sm8x = params.dprops->major == 8 && params.dprops->minor > 0;
   constexpr int kBlockM = 64;  // Fixed for all head dimensions
   if (!is_sm8x) {              // A100, H100
@@ -125,11 +125,11 @@ void run_mha_fwd_splitkv_dispatch(Flash_fwd_params& params, cudaStream_t stream)
     constexpr int kBlockN = Headdim <= 64 ? 256 : (Headdim <= 128 ? 128 : 64);
     run_flash_splitkv_fwd<Flash_fwd_kernel_traits<Headdim, kBlockM, kBlockN, 4, false, false, T>>(params, stream);
   }
-  #else
+#else
   (void)params;
   (void)stream;
   throw std::runtime_error("FlashAttention is only implemented for SM>=80");
-  #endif
+#endif
 }
 
 template <typename T>