reorginze files

src/include/gridwise_convolution_implicit_gemm_v3_nchw_cyxk_nkhw_lds_double_buffer.hpp → include/composable_kernel/kernel_algorithm/gridwise_convolution_implicit_gemm_v3_nchw_cyxk_nkhw_lds_double_buffer.hpp

 #ifndef CK_GRIDWISE_CONVOLUTION_IMPLICIT_GEMM_V3_NCHW_CYXK_NKHW_LDS_DOUBLE_BUFFER
 #define CK_GRIDWISE_CONVOLUTION_IMPLICIT_GEMM_V3_NCHW_CYXK_NKHW_LDS_DOUBLE_BUFFER
 
-#include "common.hpp"
-#include "ConstantTensorDescriptor.hpp"
-#include "ConstantMergedTensorDescriptor.hpp"
-#include "ConstantMatrixDescriptor.hpp"
-#include "blockwise_generic_tensor_slice_copy.hpp"
-#include "blockwise_gemm.hpp"
+#include "composable_kernel/utility/common.hpp"
+#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMatrixDescriptor.hpp"
+#include "composable_kernel/tensor_operation/blockwise_generic_tensor_slice_copy.hpp"
+#include "composable_kernel/tensor_operation/blockwise_gemm.hpp"
 
 namespace ck {
 

src/include/gridwise_convolution_implicit_gemm_v4_nchw_kcyx_nkhw.hpp → include/composable_kernel/kernel_algorithm/gridwise_convolution_implicit_gemm_v4_nchw_kcyx_nkhw.hpp

 #ifndef CK_GRIDWISE_CONVOLUTION_IMPLICIT_GEMM_V4_NCHW_KCYX_NKHW
 #define CK_GRIDWISE_CONVOLUTION_IMPLICIT_GEMM_V4_NCHW_KCYX_NKHW
 
-#include "common.hpp"
-#include "ConstantTensorDescriptor.hpp"
-#include "ConstantMergedTensorDescriptor.hpp"
-#include "ConstantMatrixDescriptor.hpp"
-#include "blockwise_generic_tensor_slice_copy.hpp"
-#include "blockwise_gemm.hpp"
-#include "threadwise_generic_tensor_slice_copy.hpp"
+#include "composable_kernel/utility/common.hpp"
+#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMatrixDescriptor.hpp"
+#include "composable_kernel/tensor_operation/blockwise_generic_tensor_slice_copy.hpp"
+#include "composable_kernel/tensor_operation/blockwise_gemm.hpp"
+#include "composable_kernel/tensor_operation/threadwise_generic_tensor_slice_copy.hpp"
 
 namespace ck {
 

src/include/gridwise_convolution_implicit_gemm_v4_nchw_kcyx_nkhw_lds_double_buffer.hpp → include/composable_kernel/kernel_algorithm/gridwise_convolution_implicit_gemm_v4_nchw_kcyx_nkhw_lds_double_buffer.hpp

 #ifndef CK_GRIDWISE_CONVOLUTION_IMPLICIT_GEMM_V4_NCHW_KCYX_NKHW_LDS_DOUBLE_BUFFER
 #define CK_GRIDWISE_CONVOLUTION_IMPLICIT_GEMM_V4_NCHW_KCYX_NKHW_LDS_DOUBLE_BUFFER
 
-#include "common.hpp"
-#include "ConstantTensorDescriptor.hpp"
-#include "ConstantMergedTensorDescriptor.hpp"
-#include "ConstantMatrixDescriptor.hpp"
-#include "blockwise_generic_tensor_slice_copy.hpp"
-#include "blockwise_gemm.hpp"
-#include "threadwise_generic_tensor_slice_copy.hpp"
+#include "composable_kernel/utility/common.hpp"
+#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMatrixDescriptor.hpp"
+#include "composable_kernel/tensor_operation/blockwise_generic_tensor_slice_copy.hpp"
+#include "composable_kernel/tensor_operation/blockwise_gemm.hpp"
+#include "composable_kernel/tensor_operation/threadwise_generic_tensor_slice_copy.hpp"
 
 namespace ck {
 

src/include/gridwise_direct_convolution_2_vectorized_nchw_kcyx_nkhw.hpp → include/composable_kernel/kernel_algorithm/gridwise_direct_convolution_2_vectorized_nchw_kcyx_nkhw.hpp

 #pragma once
-#include "common.hpp"
-#include "ConstantTensorDescriptor.hpp"
-#include "blockwise_2d_tensor_op.hpp"
-#include "blockwise_4d_tensor_op.hpp"
-#include "blockwise_direct_convolution.hpp"
-#include "threadwise_4d_tensor_op.hpp"
-#include "threadwise_direct_convolution.hpp"
+#include "composable_kernel/utility/common.hpp"
+#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "composable_kernel/tensor_operation/blockwise_2d_tensor_op.hpp"
+#include "composable_kernel/tensor_operation/blockwise_4d_tensor_op.hpp"
+#include "composable_kernel/tensor_operation/blockwise_direct_convolution.hpp"
+#include "composable_kernel/tensor_operation/threadwise_4d_tensor_op.hpp"
+#include "composable_kernel/tensor_operation/threadwise_direct_convolution.hpp"
 
 namespace ck {
 

src/include/gridwise_implicit_gemm_convolution_1_chwn_cyxk_khwn_padded.hpp → include/composable_kernel/kernel_algorithm/gridwise_implicit_gemm_convolution_1_chwn_cyxk_khwn_padded.hpp

 #pragma once
-#include "common.hpp"
-#include "ConstantTensorDescriptor.hpp"
-#include "ConstantMatrixDescriptor.hpp"
-#include "blockwise_4d_tensor_op.hpp"
-#include "blockwise_2d_tensor_op.hpp"
-#include "threadwise_4d_tensor_op.hpp"
-#include "blockwise_gemm.hpp"
+#include "composable_kernel/utility/common.hpp"
+#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMatrixDescriptor.hpp"
+#include "composable_kernel/tensor_operation/blockwise_4d_tensor_op.hpp"
+#include "composable_kernel/tensor_operation/blockwise_2d_tensor_op.hpp"
+#include "composable_kernel/tensor_operation/threadwise_4d_tensor_op.hpp"
+#include "composable_kernel/tensor_operation/blockwise_gemm.hpp"
 
 namespace ck {
 

src/include/ConstantMatrixDescriptor.hpp → include/composable_kernel/tensor_description/ConstantMatrixDescriptor.hpp

 #ifndef CK_CONSTANT_MATRIX_DESCRIPTOR_HPP
 #define CK_CONSTANT_MATRIX_DESCRIPTOR_HPP
 
-#include "common.hpp"
+#include "composable_kernel/utility/common.hpp"
 
 namespace ck {
 

src/include/ConstantMergedTensorDescriptor.hpp → include/composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp

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

src/include/ConstantTensorDescriptor.hpp → include/composable_kernel/tensor_description/ConstantTensorDescriptor.hpp

 #ifndef CK_CONSTANT_TENSOR_DESCRIPTOR_HPP
 #define CK_CONSTANT_TENSOR_DESCRIPTOR_HPP
 
-#include "common.hpp"
+#include "composable_kernel/utility/common.hpp"
 
 namespace ck {
 

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

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

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

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

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

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

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

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

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

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

src/include/blockwise_generic_tensor_slice_copy.hpp → include/composable_kernel/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 "threadwise_generic_tensor_slice_copy.hpp"
+#include "composable_kernel/tensor_operation/threadwise_generic_tensor_slice_copy.hpp"
 
 namespace ck {
 

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

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

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

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

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

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

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

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

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

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

src/include/threadwise_generic_tensor_slice_copy.hpp → include/composable_kernel/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 "ConstantTensorDescriptor.hpp"
-#include "ConstantMergedTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantTensorDescriptor.hpp"
+#include "composable_kernel/tensor_description/ConstantMergedTensorDescriptor.hpp"
 
 namespace ck {