Absolute include path (#281)

* ad gelu and fast_gelu

* added GeLU and fast GeLU

* clean up

* add gemm+fastgelu example

* add gemm+gelu instances

* update profiler

* clean up

* clean up

* adding gemm+bias+activation

* clean

* adding bias

* clean

* adding gemm multiple d

* debugging

* add gemm bias add fastgelu

* rename, clean

* refactoring; add readme

* refactor

* refactor

* refactor

* refactor

* refactor

* refactor

* fix

* fix

* update example

* update example

* rename

* update example

* add ckProfiler

* clean

* clean

* clean

* clean

* add client app example

* update readme

* delete obselete files

* remove old client app

* delete old file

* cleaning

* clean

* remove half

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path for all examples

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path

* fix header path

* revert client app example

* clean build

* fix build

* temporary disable client test on Jenkins

* clean

* clean

* clean

include/ck/tensor_operation/gpu/device/device_conv_fwd_bias_activation_add.hpp

-#ifndef DEVICE_CONV_FWD_BIAS_ACTIVATION_ADD_HPP
-#define DEVICE_CONV_FWD_BIAS_ACTIVATION_ADD_HPP
+#pragma once
 
+#include <vector>
 #include <iostream>
-#include "device_base.hpp"
+
+#include "ck/tensor_operation/gpu/device/device_base.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -47,4 +48,3 @@ using DeviceConvFwdBiasActivationAddPtr =
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_convnd_backward_weight_xdl_c_shuffle_nhwc_kyxc_nhwk.hpp

@@ -2,16 +2,17 @@
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_base.hpp"
-#include "device_conv_backward_weight.hpp"
-#include "convolution_backward_weight_specialization.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_bwd_weight.hpp"
-#include "gridwise_unary_elementwise_1d.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_conv_backward_weight.hpp"
+#include "ck/tensor_operation/gpu/device/convolution_backward_weight_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_bwd_weight.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_unary_elementwise_1d.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_convnd_bwd_data_xdl_ndhwc_kzyxc_ndhwk.hpp

-#ifndef DEVICE_CONVND_BWD_DATA_XDL_NDHWC_KZYXC_NDHWK_HPP
-#define DEVICE_CONVND_BWD_DATA_XDL_NDHWC_KZYXC_NDHWK_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_base.hpp"
-#include "device_conv_bwd_data.hpp"
-#include "convolution_backward_data_specialization.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v2r3.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_conv_bwd_data.hpp"
+#include "ck/tensor_operation/gpu/device/convolution_backward_data_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -1546,4 +1546,3 @@ struct DeviceConvndBwdDataXdl_Input_N_Di_Hi_Wi_C_Weight_K_Z_Y_X_C_Output_N_Do_Ho
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_convnd_fwd_xdl_nhwc_kyxc_nhwk.hpp

@@ -6,16 +6,15 @@
 #include <numeric>
 #include <sstream>
 
-#include "device.hpp"
-#include "device_prop.hpp"
-#include "device_base.hpp"
-#include "device_conv_fwd.hpp"
-#include "convolution_forward_specialization.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v2r3.hpp"
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_conv_fwd.hpp"
+#include "ck/tensor_operation/gpu/device/convolution_forward_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_bias.hpp

 #pragma once
+
 #include <iostream>
-#include "device_base.hpp"
+
+#include "ck/tensor_operation/gpu/device/device_base.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_bias_activation.hpp

-#ifndef DEVICE_GEMM_BIAS_ACTIVATION_HPP
-#define DEVICE_GEMM_BIAS_ACTIVATION_HPP
+#pragma once
 
 #include <iostream>
-#include "device_base.hpp"
+
+#include "ck/tensor_operation/gpu/device/device_base.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -40,4 +40,3 @@ using DeviceGemmBiasActivationPtr = std::unique_ptr<
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_gemm_bias_add_reduce_xdl_cshuffle.hpp

 #pragma once
+
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_gemm_reduce.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_bias_add_reduce_xdl_cshuffle_v1.hpp"
-#include "gemm_specialization.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_reduce.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_bias_add_reduce_xdl_cshuffle_v1.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_dl.hpp

@@ -3,17 +3,15 @@
 #include <iostream>
 #include <sstream>
 
-#include "device.hpp"
-#include "device_prop.hpp"
-#include "device_base.hpp"
-#include "device_gemm.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gemm_specialization.hpp"
-#include "element_wise_operation.hpp"
-#include "gridwise_gemm_dl_v1r3.hpp"
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_dl_v1r3.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_multiple_d_xdl_cshuffle.hpp

@@ -3,15 +3,15 @@
 #include <iostream>
 #include <sstream>
 
-#include "device.hpp"
-#include "device_gemm_multiple_d.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_multiple_d_xdl_cshuffle.hpp"
-#include "gemm_specialization.hpp"
-#include "device_prop.hpp"
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_multiple_d.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 

include/ck/tensor_operation/gpu/device/device_gemm_reduce_xdl_cshuffle.hpp

 #pragma once
+
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_gemm_reduce.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_reduce_xdl_cshuffle_v1.hpp"
-#include "gemm_specialization.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_reduce.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_reduce_xdl_cshuffle_v1.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_xdl.hpp

@@ -2,16 +2,16 @@
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_prop.hpp"
-#include "device_base.hpp"
-#include "device_gemm.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v2r3.hpp"
-#include "gemm_specialization.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_xdl_c_shuffle_bias_2d.hpp

 #pragma once
+
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_gemm_bias.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v3r2.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_bias.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r2.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_xdl_c_shuffle_bias_activation.hpp

-#ifndef DEVICE_GEMM_XDL_C_SHUFFLE_BIAS_ACTIVATION_HPP
-#define DEVICE_GEMM_XDL_C_SHUFFLE_BIAS_ACTIVATION_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_gemm_bias_activation.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v3r2.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_bias_activation.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r2.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -513,4 +515,3 @@ struct DeviceGemmXdl_C_Shuffle_Bias_Activation
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_gemm_xdl_c_shuffle_bias_activation_add.hpp

-#ifndef DEVICE_GEMM_XDL_C_SHUFFLE_BIAS_ACTIVATION_ADD_HPP
-#define DEVICE_GEMM_XDL_C_SHUFFLE_BIAS_ACTIVATION_ADD_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_gemm_bias_activation_add.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v3r3.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm_bias_activation_add.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r3.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -573,4 +575,3 @@ struct DeviceGemmXdl_C_Shuffle_Bias_Activation_Add
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp

 #pragma once
+
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_gemm.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdl_cshuffle_v1.hpp"
-#include "tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "device_prop.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v1.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk.hpp

-#ifndef DEVICE_GEMM_SPLITK_XDL_HPP
-#define DEVICE_GEMM_SPLITK_XDL_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_base.hpp"
-#include "device_gemm.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v2r4.hpp"
-#include "gemm_specialization.hpp"
-#include "device_prop.hpp"
-
-#ifndef CK_RUN_KERNEL_AND_TIME
-#define CK_RUN_KERNEL_AND_TIME 1
-#endif
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -639,4 +634,3 @@ struct DeviceGemmXdlSplitK
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk_c_shuffle.hpp

-#ifndef DEVICE_GEMM_XDL_SPLITK_C_SHUFFLE_HPP
-#define DEVICE_GEMM_XDL_SPLITK_C_SHUFFLE_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_base.hpp"
-#include "device_gemm.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v2r4r2.hpp"
-#include "gemm_specialization.hpp"
-
-#ifndef CK_RUN_KERNEL_AND_TIME
-#define CK_RUN_KERNEL_AND_TIME 1
-#endif
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4r2.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -641,4 +637,3 @@ struct DeviceGemmXdlSplitKCShuffle
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_grouped_gemm_xdl.hpp

-#ifndef DEVICE_GROUPED_GEMM_XDL_HPP
-#define DEVICE_GROUPED_GEMM_XDL_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device.hpp"
-#include "device_base.hpp"
-#include "device_gemm.hpp"
-#include "common_header.hpp"
-#include "tensor_layout.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "gridwise_gemm_xdlops_v2r3.hpp"
-#include "gemm_specialization.hpp"
+
+#include "ck/utility/common_header.hpp"
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
+#include "ck/tensor_operation/gpu/device/device_gemm.hpp"
+#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -638,4 +638,3 @@ struct DeviceGroupedGemmXdl
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_pool2d_fwd.hpp

-#ifndef DEVICE_POOL2D_FWD_HPP
-#define DEVICE_POOL2D_FWD_HPP
+#pragma once
 
 #include <iostream>
 #include <array>
-#include "device_base.hpp"
-#include "reduction_enums.hpp"
+
+#include "ck/tensor_operation/gpu/device/device_base.hpp"
+#include "ck/utility/reduction_enums.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -35,4 +35,3 @@ using DevicePool2dFwdPtr = std::unique_ptr<DevicePool2dFwd<ReduceOpId>>;
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif

include/ck/tensor_operation/gpu/device/device_pool2d_fwd_nhwc_nhwc.hpp

-#ifndef DEVICE_POOL2D_FWD_NHWC_NHWC_HPP
-#define DEVICE_POOL2D_FWD_NHWC_NHWC_HPP
+#pragma once
 
 #include <iostream>
 #include <sstream>
-#include "device_pool2d_fwd.hpp"
-#include "tensor_descriptor.hpp"
-#include "tensor_descriptor_helper.hpp"
-#include "reduction_operator_mapping.hpp"
-#include "gridwise_2d_reduction_threadwise.hpp"
+
+#include "ck/tensor_description/tensor_descriptor.hpp"
+#include "ck/tensor_description/tensor_descriptor_helper.hpp"
+#include "ck/tensor_operation/gpu/device/reduction_operator_mapping.hpp"
+#include "ck/tensor_operation/gpu/device/device_pool2d_fwd.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_2d_reduction_threadwise.hpp"
+#include "ck/device_utility/device_prop.hpp"
+#include "ck/device_utility/kernel_launch.hpp"
 
 namespace ck {
 namespace tensor_operation {
@@ -315,9 +317,8 @@ struct DevicePool2dFwd_Input_N_Hi_Wi_C_Output_N_Ho_Wo_C : public DevicePool2dFwd
 
         return str.str();
     }
-}; // namespace device
+};
 
 } // namespace device
 } // namespace tensor_operation
 } // namespace ck
-#endif