change 'fwd' inc file name to 'infer'

example/52_flash_atten_bias/batched_gemm_multihead_attention_bias_infer.cpp

@@ -157,6 +157,6 @@ using ReferenceGemm1Instance = ck::tensor_operation::host::ReferenceBatchedGemm<
                                                                                 B1ElementOp,
                                                                                 CElementOp>;
 
-#include "run_batched_multihead_attention_bias_forward.inc"
+#include "run_batched_multihead_attention_bias_infer.inc"
 
 int main(int argc, char* argv[]) { return run(argc, argv); }

example/52_flash_atten_bias/batched_gemm_multihead_attention_infer.cpp

@@ -157,6 +157,6 @@ using ReferenceGemm1Instance = ck::tensor_operation::host::ReferenceBatchedGemm<
                                                                                 B1ElementOp,
                                                                                 CElementOp>;
 
-#include "run_batched_multihead_attention_forward.inc"
+#include "run_batched_multihead_attention_infer.inc"
 
 int main(int argc, char* argv[]) { return run(argc, argv); }

example/52_flash_atten_bias/grouped_mutihead_attention_bias_infer.cpp

@@ -156,6 +156,6 @@ using ReferenceGemm1Instance = ck::tensor_operation::host::ReferenceBatchedGemm<
                                                                                 B1ElementOp,
                                                                                 CElementOp>;
 
-#include "run_grouped_multihead_attention_bias_forward.inc"
+#include "run_grouped_multihead_attention_bias_infer.inc"
 
 int main(int argc, char* argv[]) { return run(argc, argv); }

example/52_flash_atten_bias/run_batched_multihead_attention_bias_forward.inc → example/52_flash_atten_bias/run_batched_multihead_attention_bias_infer.inc

@@ -5,7 +5,7 @@ int run(int argc, char* argv[])
 {
     bool do_verification = true;
     int init_method      = 1;
-    bool time_kernel     = true;
+    bool time_kernel     = false;
 
     // GEMM shape for A/B0/B1/C
     // C_g_m_o = A_g_m_k * B0_g_k_n * B1_g_n_o

example/52_flash_atten_bias/run_batched_multihead_attention_forward.inc → example/52_flash_atten_bias/run_batched_multihead_attention_infer.inc

@@ -5,7 +5,7 @@ int run(int argc, char* argv[])
 {
     bool do_verification = true;
     int init_method      = 1;
-    bool time_kernel     = true;
+    bool time_kernel     = false;
 
     // GEMM shape for A/B0/B1/C
     // C_g_m_o = A_g_m_k * B0_g_k_n * B1_g_n_o

example/52_flash_atten_bias/run_grouped_multihead_attention_bias_forward.inc → example/52_flash_atten_bias/run_grouped_multihead_attention_bias_infer.inc

@@ -5,7 +5,7 @@ int run(int argc, char* argv[])
 {
     bool do_verification = true;
     int init_method      = 1;
-    bool time_kernel     = true;
+    bool time_kernel     = false;
 
     bool input_permute  = false;
     bool output_permute = true;