update include path

library/include/ck/library/reference_tensor_operation/cpu/reference_conv_backward_weight.hpp

@@ -7,7 +7,7 @@
 #include <sstream>
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_conv_bwd_data.hpp

@@ -8,7 +8,7 @@
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_conv_fwd_bias_activation.hpp

@@ -7,7 +7,7 @@
 #include <sstream>
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_conv_fwd_bias_activation_add.hpp

@@ -7,7 +7,7 @@
 #include <sstream>
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_gemm.hpp

@@ -7,7 +7,7 @@
 #include <sstream>
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_gemm_bias_2d.hpp

@@ -7,7 +7,7 @@
 #include <sstream>
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_gemm_bias_activation.hpp

@@ -8,7 +8,7 @@
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_gemm_bias_activation_add.hpp

@@ -8,7 +8,7 @@
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/reference_tensor_operation/cpu/reference_softmax.hpp

@@ -9,8 +9,8 @@
 #include <algorithm>
 
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 
 namespace ck {
 namespace tensor_operation {

library/include/ck/library/utility/host_reduction.hpp

@@ -11,8 +11,8 @@
 #include "ck/utility/reduction_enums.hpp"
 #include "ck/utility/reduction_common.hpp"
 #include "ck/utility/reduction_functions_accumulate.hpp"
-#include "ck/library/host_tensor/host_common_util.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/host_common_util.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 template <int NDim>
 static void get_all_indexes(const std::array<size_t, NDim>& dimLengths,

library/include/ck/library/utility/op_instance_engine.hpp

@@ -16,8 +16,8 @@
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 
 #include "ck/library/utility/check_err.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
 
 namespace ck {
 namespace utils {

library/src/tensor_operation_instance/gpu/CMakeLists.txt

@@ -75,7 +75,7 @@ target_include_directories(device_operations PUBLIC
     $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/tensor_operation/gpu/warp>
     $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/tensor_operation/gpu/thread>
     $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/tensor_operation/gpu/element>
-    $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/library/host_tensor>
+    $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/library/utility>
     $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/library/tensor_operation_instance>
     $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/library/tensor_operation_instance/gpu>
     $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}/ck/library/tensor_operation_instance/gpu/reduce>

profiler/CMakeLists.txt

@@ -5,45 +5,44 @@ include_directories(BEFORE
 # ck_profiler
 set(PROFILER_SOURCE
     src/profiler.cpp
-#   src/profile_gemm.cpp
-#   src/profile_gemm_splitk.cpp
-#   src/profile_gemm_bilinear.cpp
-#   src/profile_gemm_bias_add_reduce.cpp
-#   src/profile_gemm_add_add_fastgelu.cpp
-#   src/profile_gemm_reduce.cpp
-#   src/profile_batched_gemm.cpp
-#   src/profile_batched_gemm_reduce.cpp
-#   src/profile_grouped_gemm.cpp
+    src/profile_gemm.cpp
+    src/profile_gemm_splitk.cpp
+    src/profile_gemm_bilinear.cpp
+    src/profile_gemm_bias_add_reduce.cpp
+    src/profile_gemm_add_add_fastgelu.cpp
+    src/profile_gemm_reduce.cpp
+    src/profile_batched_gemm.cpp
+    src/profile_batched_gemm_reduce.cpp
+    src/profile_grouped_gemm.cpp
     src/profile_conv_fwd.cpp
-#   src/profile_conv_fwd_bias_relu.cpp
-#   src/profile_conv_fwd_bias_relu_add.cpp
-#   src/profile_convnd_fwd.cpp
-#   src/profile_convnd_bwd_data.cpp
-#   src/profile_conv_bwd_weight.cpp
-#   src/profile_convnd_bwd_weight.cpp
-#   src/profile_reduce.cpp
-#   src/profile_normalization.cpp
+    src/profile_conv_fwd_bias_relu.cpp
+    src/profile_conv_fwd_bias_relu_add.cpp
+    src/profile_convnd_bwd_data.cpp
+    src/profile_conv_bwd_weight.cpp
+    src/profile_convnd_bwd_weight.cpp
+    src/profile_reduce.cpp
+    src/profile_normalization.cpp
 )
 
 add_executable(ckProfiler ${PROFILER_SOURCE})
 
 target_link_libraries(ckProfiler PRIVATE utility)
-#target_link_libraries(ckProfiler PRIVATE device_gemm_instance)
-#target_link_libraries(ckProfiler PRIVATE device_gemm_splitk_instance)
-#target_link_libraries(ckProfiler PRIVATE device_gemm_bilinear_instance)
-#target_link_libraries(ckProfiler PRIVATE device_gemm_add_add_fastgelu_instance)
-#target_link_libraries(ckProfiler PRIVATE device_gemm_reduce_instance)
-#target_link_libraries(ckProfiler PRIVATE device_gemm_bias_add_reduce_instance)
-#target_link_libraries(ckProfiler PRIVATE device_batched_gemm_instance)
-#target_link_libraries(ckProfiler PRIVATE device_batched_gemm_reduce_instance)
-#target_link_libraries(ckProfiler PRIVATE device_grouped_gemm_instance)
+target_link_libraries(ckProfiler PRIVATE device_gemm_instance)
+target_link_libraries(ckProfiler PRIVATE device_gemm_splitk_instance)
+target_link_libraries(ckProfiler PRIVATE device_gemm_bilinear_instance)
+target_link_libraries(ckProfiler PRIVATE device_gemm_add_add_fastgelu_instance)
+target_link_libraries(ckProfiler PRIVATE device_gemm_reduce_instance)
+target_link_libraries(ckProfiler PRIVATE device_gemm_bias_add_reduce_instance)
+target_link_libraries(ckProfiler PRIVATE device_batched_gemm_instance)
+target_link_libraries(ckProfiler PRIVATE device_batched_gemm_reduce_instance)
+target_link_libraries(ckProfiler PRIVATE device_grouped_gemm_instance)
 target_link_libraries(ckProfiler PRIVATE device_conv1d_fwd_instance)
 target_link_libraries(ckProfiler PRIVATE device_conv2d_fwd_instance)
 target_link_libraries(ckProfiler PRIVATE device_conv3d_fwd_instance)
-#target_link_libraries(ckProfiler PRIVATE device_conv2d_fwd_bias_relu_instance)
-#target_link_libraries(ckProfiler PRIVATE device_conv2d_fwd_bias_relu_add_instance)
-#target_link_libraries(ckProfiler PRIVATE device_convnd_bwd_data_instance)
-#target_link_libraries(ckProfiler PRIVATE device_conv2d_bwd_weight_instance)
-#target_link_libraries(ckProfiler PRIVATE device_convnd_bwd_weight_instance)
-#target_link_libraries(ckProfiler PRIVATE device_normalization_instance)
-#target_link_libraries(ckProfiler PRIVATE device_reduce_instance)
+target_link_libraries(ckProfiler PRIVATE device_conv2d_fwd_bias_relu_instance)
+target_link_libraries(ckProfiler PRIVATE device_conv2d_fwd_bias_relu_add_instance)
+target_link_libraries(ckProfiler PRIVATE device_convnd_bwd_data_instance)
+target_link_libraries(ckProfiler PRIVATE device_conv2d_bwd_weight_instance)
+target_link_libraries(ckProfiler PRIVATE device_convnd_bwd_weight_instance)
+target_link_libraries(ckProfiler PRIVATE device_normalization_instance)
+target_link_libraries(ckProfiler PRIVATE device_reduce_instance)

profiler/include/profile_batched_gemm_impl.hpp

@@ -13,9 +13,9 @@
 #include "ck/library/tensor_operation_instance/gpu/batched_gemm.hpp"
 
 #include "ck/library/utility/check_err.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_batched_gemm.hpp"
 
 namespace ck {

profiler/include/profile_batched_gemm_reduce_impl.hpp

@@ -11,9 +11,9 @@
 
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/conv_util.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_batched_gemm.hpp"
 
 namespace ck {

profiler/include/profile_conv_bwd_weight_impl.hpp

@@ -10,9 +10,9 @@
 
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/conv_util.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_conv_backward_weight.hpp"
 
 namespace ck {

profiler/include/profile_conv_fwd_bias_relu_add_impl.hpp

@@ -9,9 +9,9 @@
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/check_err.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_conv_fwd_bias_activation_add.hpp"
 
 namespace ck {

profiler/include/profile_conv_fwd_bias_relu_impl.hpp

@@ -9,9 +9,9 @@
 #include "ck/tensor_operation/gpu/device/device_conv_fwd_bias_activation.hpp"
 
 #include "ck/library/utility/check_err.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_conv_fwd_bias_activation.hpp"
 
 namespace ck {

profiler/include/profile_convnd_bwd_data_impl.hpp

@@ -9,9 +9,9 @@
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/conv_util.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_conv_bwd_data.hpp"
 
 using F16  = ck::half_t;

profiler/include/profile_convnd_bwd_weight_impl.hpp

@@ -7,9 +7,9 @@
 
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/conv_util.hpp"
-#include "ck/library/host_tensor/device_memory.hpp"
-#include "ck/library/host_tensor/host_tensor.hpp"
-#include "ck/library/host_tensor/host_tensor_generator.hpp"
+#include "ck/library/utility/device_memory.hpp"
+#include "ck/library/utility/host_tensor.hpp"
+#include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_conv_backward_weight.hpp"
 
 using F16  = ck::half_t;