cuda_transpose_recursive_impl()

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.

References syten::LimVec< Type, max, MaxType, type >::begin(), syten::Cuda::CudaMutSpan< T >::begin(), syten::Deparallelise::cols(), syten::Cuda::copy(), cuda_transpose_r2(), syten::RepRegister::dim(), syten::LimVec< Type, max, MaxType, type >::end(), if_constexpr, std::is_sorted(), syten::rank(), syten::Deparallelise::rows(), and SYTEN_ASSERT.

Here is the call graph for this function: