remove .hip extension

src/include/blockwise_4d_tensor_op.hip.hpp → src/include/blockwise_4d_tensor_op.hpp

 #pragma once
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "threadwise_tensor_slice_op.hpp"
 
 template <index_t BlockSize, class Float, class DstDesc, class F>
 __device__ void

src/include/blockwise_batched_gemm.hip.hpp → src/include/blockwise_batched_gemm.hpp

 #pragma once
-#include "threadwise_gemm.hip.hpp"
+#include "threadwise_gemm.hpp"
 
 template <index_t BlockSize,
           class BlockMatrixA,

src/include/blockwise_gemm.hip.hpp → src/include/blockwise_gemm.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "threadwise_gemm.hip.hpp"
+#include "common.hpp"
+#include "threadwise_gemm.hpp"
 
 // if following number are power of 2, index calculation shall be greatly reduced:
 //    MPerThreadSubC, NPerThreadSubC, MLevel0Cluster, NLevel0Cluster, MLevel1Cluster, NLevel1Cluster

src/include/blockwise_generic_tensor_slice_op.hip.hpp → src/include/blockwise_generic_tensor_slice_op.hpp

 #pragma once
-#include "threadwise_tensor_slice_op.hip.hpp"
+#include "threadwise_tensor_slice_op.hpp"
 
 // slice a (normal or merged) tensor, and copy it into another (normal or merged) tensor
 // memory layout (ordering of dimensions) can be different between src and dst

src/include/blockwise_tensor_slice_op.hip.hpp → src/include/blockwise_tensor_slice_op.hpp

 #pragma once
-#include "threadwise_tensor_slice_op.hip.hpp"
+#include "threadwise_tensor_slice_op.hpp"
 
 template <index_t BlockSize,
           class Float,

src/include/common.hip.hpp

-#pragma once
-#include "base.hip.hpp"
-#include "vector_type.hip.hpp"
-#include "integral_constant.hip.hpp"
-#include "Sequence.hip.hpp"
-#include "Array.hip.hpp"
-#include "functional.hip.hpp"
-#include "functional2.hip.hpp"
-#include "functional3.hip.hpp"
-
-#if USE_AMD_INLINE_ASM
-#include "amd_inline_asm.hip.hpp"
-#endif

src/include/common.hpp

+#pragma once
+#include "base.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 USE_AMD_INLINE_ASM
+#include "amd_inline_asm.hpp"
+#endif

src/include/conv_common.hip.hpp → src/include/conv_common.hpp

 #pragma once
-#include "ConstantTensorDescriptor.hip.hpp"
+#include "ConstantTensorDescriptor.hpp"
 
 // this is ugly, only for 4d
 template <class InDesc, class WeiDesc>

src/include/functional.hip.hpp → src/include/functional.hpp

 #pragma once
-#include "integral_constant.hip.hpp"
-#include "Sequence.hip.hpp"
+#include "integral_constant.hpp"
+#include "Sequence.hpp"
 
 struct forwarder
 {

src/include/functional2.hip.hpp → src/include/functional2.hpp

 #pragma once
-#include "functional.hip.hpp"
-#include "Sequence.hip.hpp"
+#include "functional.hpp"
+#include "Sequence.hpp"
 
 template <class>
 struct static_for_impl;

src/include/functional3.hip.hpp → src/include/functional3.hpp

 #pragma once
-#include "functional.hip.hpp"
-#include "functional2.hip.hpp"
-#include "Sequence.hip.hpp"
-#include "Array.hip.hpp"
+#include "functional.hpp"
+#include "functional2.hpp"
+#include "Sequence.hpp"
+#include "Array.hpp"
 
 // RemainLengths: Sequence<...>
 template <class RemainLengths>

src/include/gridwise_convolution_direct_v2_nchw_kcyx_nkhw.hip.hpp → src/include/gridwise_convolution_direct_v2_nchw_kcyx_nkhw.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_4d_tensor_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_direct_convolution.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_4d_tensor_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_direct_convolution.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r1_chwn_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r1_chwn_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_4d_tensor_op.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_4d_tensor_op.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r2_chwn_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r2_chwn_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_3d_tensor_op.hip.hpp"
-#include "blockwise_4d_tensor_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_3d_tensor_op.hpp"
+#include "blockwise_4d_tensor_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r2_nchw_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r2_nchw_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_3d_tensor_op.hip.hpp"
-#include "blockwise_tensor_slice_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_3d_tensor_op.hpp"
+#include "blockwise_tensor_slice_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r3_chwn_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r3_chwn_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_4d_tensor_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_4d_tensor_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r3_lds_double_buffer_chwn_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r3_lds_double_buffer_chwn_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_4d_tensor_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_4d_tensor_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r3_lds_double_buffer_nchw_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r3_lds_double_buffer_nchw_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_tensor_slice_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_tensor_slice_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r3_lds_double_buffer_nchw_cyxk_nkhw.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r3_lds_double_buffer_nchw_cyxk_nkhw.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_tensor_slice_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_tensor_slice_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,

src/include/gridwise_convolution_implicit_gemm_v1r3_nchw_cyxk_khwn.hip.hpp → src/include/gridwise_convolution_implicit_gemm_v1r3_nchw_cyxk_khwn.hpp

 #pragma once
-#include "common.hip.hpp"
-#include "ConstantTensorDescriptor.hip.hpp"
-#include "ConstantMatrixDescriptor.hip.hpp"
-#include "blockwise_2d_tensor_op.hip.hpp"
-#include "blockwise_tensor_slice_op.hip.hpp"
-#include "threadwise_tensor_slice_op.hip.hpp"
-#include "threadwise_4d_tensor_op.hip.hpp"
-#include "blockwise_batched_gemm.hip.hpp"
+#include "common.hpp"
+#include "ConstantTensorDescriptor.hpp"
+#include "ConstantMatrixDescriptor.hpp"
+#include "blockwise_2d_tensor_op.hpp"
+#include "blockwise_tensor_slice_op.hpp"
+#include "threadwise_tensor_slice_op.hpp"
+#include "threadwise_4d_tensor_op.hpp"
+#include "blockwise_batched_gemm.hpp"
 
 template <index_t GridSize,
           index_t BlockSize,