ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-06-08 23:38:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

use remove_cvref_t

Browse Source
This commit is contained in:
Chao Liu
2022-05-04 03:00:16 +00:00
parent e86f3769b8
commit 7b4de77570
10 changed files with 22 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
include/ck/config.hpp
View File

@@ -26,7 +26,7 @@
#endif
#endif
// buffer resourse
// buffer resource
#ifndef __HIP_DEVICE_COMPILE__ // for host code
#define CK_BUFFER_RESOURCE_3RD_DWORD -1
#elif defined(__gfx803__) || defined(__gfx900__) || defined(__gfx906__) || defined(__gfx908__) || \

4
include/ck/tensor_operation/gpu/block/blockwise_tensor_slice_transfer_v4r1.hpp
View File

@@ -56,8 +56,8 @@ struct BlockwiseTensorSliceTransfer_v4r1
dst_element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<SrcDesc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<SrcDesc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&
nDim == SrcDimAccessOrder::Size() && nDim == DstDimAccessOrder::Size(),

4
include/ck/tensor_operation/gpu/block/blockwise_tensor_slice_transfer_v5r1.hpp
View File

@@ -45,8 +45,8 @@ struct BlockwiseTensorSliceTransfer_v5r1
src_desc, make_zero_multi_index<nDim>(), dst_desc, make_zero_multi_index<nDim>())
{
static_assert(nDim == remove_reference_t<remove_cv_t<SrcDesc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<SrcDesc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == BlockSliceLengths::Size() && nDim == ThreadSliceLengths::Size() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&

4
include/ck/tensor_operation/gpu/block/blockwise_tensor_slice_transfer_v6r1.hpp
View File

@@ -48,8 +48,8 @@ struct BlockwiseTensorSliceTransfer_v6r1
element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<SrcDesc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<SrcDesc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&
nDim == DimAccessOrder::Size(),

6
include/ck/tensor_operation/gpu/block/blockwise_tensor_slice_transfer_v6r2.hpp
View File

@@ -55,9 +55,9 @@ struct BlockwiseTensorSliceTransfer_v6r2
element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<Src0Desc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<Src1Desc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<Src0Desc>::GetNumOfDimension() &&
nDim == remove_cvref_t<Src1Desc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&
nDim == DimAccessOrder::Size(),

8
include/ck/tensor_operation/gpu/block/blockwise_tensor_slice_transfer_v6r3.hpp
View File

@@ -62,10 +62,10 @@ struct BlockwiseTensorSliceTransfer_v6r3
element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<Src0Desc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<Src1Desc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<Src2Desc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<Src0Desc>::GetNumOfDimension() &&
nDim == remove_cvref_t<Src1Desc>::GetNumOfDimension() &&
nDim == remove_cvref_t<Src2Desc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&
nDim == DimAccessOrder::Size(),

4
include/ck/tensor_operation/gpu/block/thread_group_tensor_slice_transfer_v4r1.hpp
View File

@@ -56,8 +56,8 @@ struct ThreadGroupTensorSliceTransfer_v4r1
dst_element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<SrcDesc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<SrcDesc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&
nDim == SrcDimAccessOrder::Size() && nDim == DstDimAccessOrder::Size(),

4
include/ck/tensor_operation/gpu/block/thread_group_tensor_slice_transfer_v6r1.hpp
View File

@@ -46,8 +46,8 @@ struct ThreadGroupTensorSliceTransfer_v6r1
element_op)
{
static_assert(nDim == remove_reference_t<remove_cv_t<SrcDesc>>::GetNumOfDimension() &&
nDim == remove_reference_t<remove_cv_t<DstDesc>>::GetNumOfDimension() &&
static_assert(nDim == remove_cvref_t<SrcDesc>::GetNumOfDimension() &&
nDim == remove_cvref_t<DstDesc>::GetNumOfDimension() &&
nDim == ThreadClusterLengths::Size() &&
nDim == ThreadClusterArrangeOrder::Size() &&
nDim == DimAccessOrder::Size(),

2
include/ck/tensor_operation/gpu/device/device_gemm_xdl_c_shuffle_bias_2d.hpp
View File

@@ -289,7 +289,7 @@ struct DeviceGemmXdl_C_Shuffle_Bias_2d
arg.N01_))
{
throw std::runtime_error(
"wrong! GridwiseGemm_km_kn_m0m1n0n1_xdlops_v3r2 has invalid setting");
"wrong! GridwiseGemm_k0mk1_k0nk1_mn_xdlops_v3r2 has invalid setting");
}
const index_t grid_size = GridwiseGemm::CalculateGridSize(arg.c_grid_desc_m_n_);

6
include/ck/utility/tuple.hpp
View File

@@ -22,7 +22,7 @@ struct TupleElement
__host__ __device__ constexpr TupleElement() = default;
template <typename T,
typename enable_if<!is_same<remove_reference_t<remove_cv_t<T>>, TupleElement>::value,
typename enable_if<!is_same<remove_cvref_t<T>, TupleElement>::value,
bool>::type = false>
__host__ __device__ constexpr TupleElement(T&& v) : mData(std::forward<T>(v))
{
@@ -60,7 +60,7 @@ struct TupleImpl<Sequence<Is...>, Xs...> : TupleElement<TupleElementKey<Is>, Xs>
template <typename Y,
typename enable_if<sizeof...(Is) == 1 && sizeof...(Xs) == 1 &&
!is_same<remove_reference_t<remove_cv_t<Y>>, TupleImpl>::value,
!is_same<remove_cvref_t<Y>, TupleImpl>::value,
bool>::type = false>
__host__ __device__ constexpr TupleImpl(Y&& y)
: TupleElement<TupleElementKey<Is>, Xs>(std::forward<Y>(y))...
@@ -102,7 +102,7 @@ struct Tuple : detail::TupleImpl<typename arithmetic_sequence_gen<0, sizeof...(X
template <typename Y,
typename enable_if<sizeof...(Xs) == 1 &&
!is_same<remove_reference_t<remove_cv_t<Y>>, Tuple>::value,
!is_same<remove_cvref_t<Y>, Tuple>::value,
bool>::type = false>
__host__ __device__ constexpr Tuple(Y&& y) : base(std::forward<Y>(y))
{