reorginzed files

include/composable_kernel/tensor_operation/blockwise_2d_tensor_op.hpp → composable_kernel/include/tensor_operation/blockwise_2d_tensor_op.hpp

 #ifndef CK_BLOCKWISE_2D_TENSOR_OP_HPP
 #define CK_BLOCKWISE_2D_TENSOR_OP_HPP
 
-#include "composable_kernel/utility/common.hpp"
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/blockwise_3d_tensor_op.hpp → composable_kernel/include/tensor_operation/blockwise_3d_tensor_op.hpp

 #ifndef CK_BLOCKWISE_3D_TENSOR_OP_HPP
 #define CK_BLOCKWISE_3D_TENSOR_OP_HPP
 
-#include "composable_kernel/utility/common.hpp"
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/blockwise_4d_tensor_op.hpp → composable_kernel/include/tensor_operation/blockwise_4d_tensor_op.hpp

 #ifndef CK_BLOCKWISE_4D_TENSOR_OP_HPP
 #define CK_BLOCKWISE_4D_TENSOR_OP_HPP
 
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
-#include "composable_kernel/tensor_operation/threadwise_tensor_slice_copy.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "threadwise_tensor_slice_copy.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/blockwise_batched_gemm.hpp → composable_kernel/include/tensor_operation/blockwise_batched_gemm.hpp

 #ifndef CK_BLOCKWISE_BATCHED_GEMM_HPP
 #define CK_BLOCKWISE_BATCHED_GEMM_HPP
 
-#include "composable_kernel/tensor_operation/threadwise_gemm.hpp"
+#include "common_header.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "threadwise_gemm.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/blockwise_gemm.hpp → composable_kernel/include/tensor_operation/blockwise_gemm.hpp

 #ifndef CK_BLOCKWISE_GEMM_HPP
 #define CK_BLOCKWISE_GEMM_HPP
 
-#include "composable_kernel/utility/common.hpp"
-#include "composable_kernel/tensor_operation/threadwise_gemm.hpp"
+#include "common_header.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "threadwise_gemm.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/blockwise_generic_tensor_slice_copy.hpp → composable_kernel/include/tensor_operation/blockwise_generic_tensor_slice_copy.hpp

 #ifndef CK_BLOCKWISE_GENERIC_TENSOR_SLICE_COPY_HPP
 #define CK_BLOCKWISE_GENERIC_TENSOR_SLICE_COPY_HPP
 
-#include "composable_kernel/tensor_operation/threadwise_generic_tensor_slice_copy.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMergedTensorDescriptor.hpp"
+#include "threadwise_generic_tensor_slice_copy.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/blockwise_tensor_slice_copy.hpp → composable_kernel/include/tensor_operation/blockwise_tensor_slice_copy.hpp

 #ifndef CK_BLOCKWISE_TENSOR_SLICE_COPY_HPP
 #define CK_BLOCKWISE_TENSOR_SLICE_COPY_HPP
 
-#include "composable_kernel/tensor_operation/threadwise_tensor_slice_copy.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "threadwise_tensor_slice_copy.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/threadwise_4d_tensor_op.hpp → composable_kernel/include/tensor_operation/threadwise_4d_tensor_op.hpp

 #ifndef CK_THREADWISE_4D_TENSOR_OP_HPP
 #define CK_THREADWISE_4D_TENSOR_OP_HPP
 
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/threadwise_direct_convolution.hpp → composable_kernel/include/tensor_operation/threadwise_direct_convolution.hpp

 #ifndef CK_THREADWISE_DIRECT_CONVOLUTION_HPP
 #define CK_THREADWISE_DIRECT_CONVOLUTION_HPP
 
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
-#include "composable_kernel/tensor_operation/threadwise_tensor_slice_copy.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "threadwise_tensor_slice_copy.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/threadwise_gemm.hpp → composable_kernel/include/tensor_operation/threadwise_gemm.hpp

 #ifndef CK_THREADWISE_GEMM_HPP
 #define CK_THREADWISE_GEMM_HPP
 
-#include "composable_kernel/utility/common.hpp"
-#include "composable_kernel/tensor_description/ConstantMatrixDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantMatrixDescriptor.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/threadwise_generic_tensor_op.hpp → composable_kernel/include/tensor_operation/threadwise_generic_tensor_op.hpp

 #ifndef CK_THREADWISE_GENERIC_TENSOR_OP_HPP
 #define CK_THREADWISE_GENERIC_TENSOR_OP_HPP
 
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
-#include "composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMergedTensorDescriptor.hpp"
 
 namespace ck {
 template <class Float, class TDesc>

include/composable_kernel/tensor_operation/threadwise_generic_tensor_slice_copy.hpp → composable_kernel/include/tensor_operation/threadwise_generic_tensor_slice_copy.hpp

 #ifndef CK_THREADWISE_GENERIC_TENSOR_SLICE_COPY_HPP
 #define CK_THREADWISE_GENERIC_TENSOR_SLICE_COPY_HPP
 
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
-#include "composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMergedTensorDescriptor.hpp"
 
 namespace ck {
 

include/composable_kernel/tensor_operation/threadwise_tensor_slice_copy.hpp → composable_kernel/include/tensor_operation/threadwise_tensor_slice_copy.hpp

 #ifndef CK_THREADWISE_TENSOR_SLICE_COPY_HPP
 #define CK_THREADWISE_TENSOR_SLICE_COPY_HPP
 
-#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "common_header.hpp"
+#include "ConstantTensorDescriptor.hpp"
 
 namespace ck {
 

include/composable_kernel/utility/Array.hpp → composable_kernel/include/utility/Array.hpp

 #ifndef CK_ARRAY_HPP
 #define CK_ARRAY_HPP
 
-#include "composable_kernel/utility/Sequence.hpp"
-#include "composable_kernel/utility/functional2.hpp"
+#include "Sequence.hpp"
+#include "functional2.hpp"
 
 namespace ck {
 

include/composable_kernel/utility/Sequence.hpp → composable_kernel/include/utility/Sequence.hpp

 #ifndef CK_SEQUENCE_HPP
 #define CK_SEQUENCE_HPP
 
-#include "composable_kernel/utility/integral_constant.hpp"
-#include "composable_kernel/utility/functional.hpp"
+#include "integral_constant.hpp"
+#include "functional.hpp"
 
 namespace ck {
 

include/composable_kernel/utility/amd_inline_asm.hpp → composable_kernel/include/utility/amd_inline_asm.hpp

 #ifndef CK_AMD_INLINE_ASM_HPP
 #define CK_AMD_INLINE_ASM_HPP
 
-#include "composable_kernel/utility/vector_type.hpp"
+#include "vector_type.hpp"
 
 #define NO_VM_WAIT 0
 #define NO_LGKM_WAIT 0

composable_kernel/include/utility/common_header.hpp

+#ifndef CK_COMMON_HPP
+#define CK_COMMON_HPP
+
+#include "config.hpp"
+#include "utility.hpp"
+#include "vector_type.hpp"
+#include "integral_constant.hpp"
+#include "Sequence.hpp"
+#include "Array.hpp"
+#include "functional.hpp"
+#include "functional2.hpp"
+#include "functional3.hpp"
+
+#if CK_USE_AMD_INLINE_ASM
+#include "amd_inline_asm.hpp"
+#endif
+
+#endif

include/composable_kernel/utility/config_amd.hpp.in → composable_kernel/include/utility/config_amd.hpp.in

-#ifndef CK_CONFIG_HPP
-#define CK_CONFIG_HPP
+#ifndef CK_CONFIG_AMD_HPP
+#define CK_CONFIG_AMD_HPP
 
 #cmakedefine01 CK_DEVICE_BACKEND_AMD
 

include/composable_kernel/utility/config_nvidia.hpp.in → composable_kernel/include/utility/config_nvidia.hpp.in

-#ifndef CK_CONFIG_CUDA_HPP
-#define CK_CONFIG_CUDA_HPP
+#ifndef CK_CONFIG_NVIDIA_HPP
+#define CK_CONFIG_NVIDIA_HPP
 
 #cmakedefine01 CK_DEVICE_BACKEND_NVIDIA
 

include/composable_kernel/utility/functional.hpp → composable_kernel/include/utility/functional.hpp

 #ifndef CK_FUNCTIONAL_HPP
 #define CK_FUNCTIONAL_HPP
 
-#include "composable_kernel/utility/integral_constant.hpp"
-#include "composable_kernel/utility/Sequence.hpp"
+#include "integral_constant.hpp"
+#include "Sequence.hpp"
 
 namespace ck {