cuda_transpose_r2() [4/4]

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.

If conj is true and the scalar type is complex, entries are complex-conjugated.

References conj(), syten::Cuda::get_handle(), syten::Cuda::handle_error(), syten::Cuda::host_device(), syten::Cuda::CudaStream::sync_destroy(), and syten::transpose_r2().

Here is the call graph for this function: