clean up

example/01_gemm/gemm_xdl_bf16.cpp

@@ -101,11 +101,7 @@ int main(int argc, char* argv[])
     ck::index_t StrideB = 4096;
     ck::index_t StrideC = 4096;
 
-    if(argc == 1)
-    {
-        // use default case
-    }
-    else if(argc == 4)
+    if(argc == 4)
     {
         do_verification = std::stoi(argv[1]);
         init_method     = std::stoi(argv[2]);

example/01_gemm/gemm_xdl_fp16.cpp

@@ -69,11 +69,7 @@ int main(int argc, char* argv[])
     ck::index_t StrideB = 4096;
     ck::index_t StrideC = 4096;
 
-    if(argc == 1)
-    {
-        // use default case
-    }
-    else if(argc == 4)
+    if(argc == 4)
     {
         do_verification = std::stoi(argv[1]);
         init_method     = std::stoi(argv[2]);

example/01_gemm/gemm_xdl_int8.cpp

@@ -96,11 +96,7 @@ int main(int argc, char* argv[])
     ck::index_t StrideB = 4096;
     ck::index_t StrideC = 4096;
 
-    if(argc == 1)
-    {
-        // use default case
-    }
-    else if(argc == 4)
+    if(argc == 4)
     {
         do_verification = std::stoi(argv[1]);
         init_method     = std::stoi(argv[2]);

example/21_gemm_gelu/gemm_gelu_xdl_fp16.cpp

@@ -3,7 +3,7 @@
 #include <initializer_list>
 #include <cstdlib>
 #include <stdlib.h>
-#include <half.hpp>
+
 #include "check_err.hpp"
 #include "config.hpp"
 #include "device.hpp"

library/src/tensor_operation_instance/gpu/gemm_gelu/device_gemm_gelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instance.cpp

 #include <stdlib.h>
+
 #include "config.hpp"
 #include "device_gemm_xdl_cshuffle.hpp"
 #include "element_wise_operation.hpp"

library/src/tensor_operation_instance/gpu/gemm_gelu/device_gemm_gelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instance.cpp

 #include <stdlib.h>
+
 #include "config.hpp"
 #include "device_gemm_xdl_cshuffle.hpp"
 #include "element_wise_operation.hpp"

library/src/tensor_operation_instance/gpu/gemm_gelu/device_gemm_gelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instance.cpp

 #include <stdlib.h>
+
 #include "config.hpp"
 #include "device_gemm_xdl_cshuffle.hpp"
 #include "element_wise_operation.hpp"

library/src/tensor_operation_instance/gpu/gemm_gelu/device_gemm_gelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instance.cpp

 #include <stdlib.h>
+
 #include "config.hpp"
 #include "device_gemm_xdl_cshuffle.hpp"
 #include "element_wise_operation.hpp"

profiler/src/profiler.cpp

@@ -7,7 +7,6 @@
 #include "profile_convnd_fwd.hpp"
 
 int profile_gemm(int, char*[]);
-#if 0
 int profile_gemm_bias_2d(int, char*[]);
 int profile_gemm_bias_relu(int, char*[]);
 int profile_gemm_bias_relu_add(int, char*[]);
@@ -22,7 +21,6 @@ int profile_convnd_bwd_data(int, char*[], int);
 int profile_reduce(int, char*[]);
 int profile_conv_bwd_weight(int, char*[]);
 int profile_batched_gemm_reduce(int, char*[]);
-#endif
 int profile_gemm_gelu(int, char*[]);
 
 static void print_helper_message()
@@ -60,7 +58,6 @@ int main(int argc, char* argv[])
     {
         return profile_gemm(argc, argv);
     }
-#if 0
     else if(strcmp(argv[1], "gemm_bias_2d") == 0)
     {
         return profile_gemm_bias_2d(argc, argv);
@@ -125,7 +122,6 @@ int main(int argc, char* argv[])
     {
         return profile_conv_bwd_weight(argc, argv);
     }
-#endif
     else if(strcmp(argv[1], "gemm_gelu") == 0)
     {
         return profile_gemm_gelu(argc, argv);