cuda_transpose.h File Reference

Transposition functions for CUDA tensors. More...

#include "inc/dense/cuda.h"
#include "inc/dense/cuda_transpose_impl.h"
#include "inc/util/cuda_span.h"
#include "inc/util/limvec.h"
#include "inc/util/pair.h"

Include dependency graph for cuda_transpose.h:

This graph shows which files directly or indirectly include this file:

Namespaces
	syten
	Syten namespace.
	syten::CudaDenseTensorImpl
	Implementation namespace for CUDA dense tensors.

Functions
template<Rank rank, typename Scalar , TransposeMethod method = TransposeMethod::Default, Rank... Ranks>
void	syten::CudaDenseTensorImpl::cuda_transpose_impl (CudaConstSpan< Scalar > inp, CudaMutSpan< Scalar > out, ConstSpan< Index > in_perm, ConstSpan< Index > in_dim, Conj do_conj=Conj::n(), bool do_checks=true, std::index_sequence< Ranks... >={})
	Implementation of the CUDA tensor transposition routines, to be called only from cuda_transpose(). More...
void	syten::CudaDenseTensorImpl::cuda_transpose_r2 (CudaConstSpan< double > inp, CudaMutSpan< double > out, Index inp_rows, Index inp_cols, Conj const conj=Conj::n())
	Transposition of a rank-2 CUDA matrix from range inp into range out in row-major format with inp_rows and inp_cols respectively. More...
void	syten::CudaDenseTensorImpl::cuda_transpose_r2 (CudaConstSpan< float > inp, CudaMutSpan< float > out, Index inp_rows, Index inp_cols, Conj const conj=Conj::n())
	Transposition of a rank-2 CUDA matrix from range inp into range out in row-major format with inp_rows and inp_cols respectively. More...
void	syten::CudaDenseTensorImpl::cuda_transpose_r2 (CudaConstSpan< std::complex< double > > inp, CudaMutSpan< std::complex< double > > out, Index inp_rows, Index inp_cols, Conj const conj=Conj::n())
	Transposition of a rank-2 CUDA matrix from range inp into range out in row-major format with inp_rows and inp_cols respectively. More...
void	syten::CudaDenseTensorImpl::cuda_transpose_r2 (CudaConstSpan< std::complex< float > > inp, CudaMutSpan< std::complex< float > > out, Index inp_rows, Index inp_cols, Conj const conj=Conj::n())
	Transposition of a rank-2 CUDA matrix from range inp into range out in row-major format with inp_rows and inp_cols respectively. More...
template<Rank rank, typename Scalar >
void	syten::CudaDenseTensorImpl::cuda_transpose_recursive (CudaConstSpan< Scalar > inp, CudaMutSpan< Scalar > out, ConstSpan< IndexNumber > in_perm, ConstSpan< Index > in_dim, Conj do_conj=Conj::n())
	Entry point for the CUDA recursive tensor transposition implementation, to be called only from cuda_transpose_impl(). More...
template<Rank rank, typename Scalar >
void	syten::CudaDenseTensorImpl::cuda_transpose_recursive_impl (CudaConstSpan< Scalar > inp, CudaMutSpan< Scalar > out, [[maybe_unused]] CudaMutSpan< Scalar > workspace, [[maybe_unused]] ConstSpan< IndexNumber > perm, [[maybe_unused]] ConstSpan< Index > dim)
	Recursive implementation of the CUDA tensor transposition. More...

Detailed Description

Transposition functions for CUDA tensors.