SyTen

◆ cuda_transpose_recursive_impl()

template<Rank rank, typename Scalar >
void syten::CudaDenseTensorImpl::cuda_transpose_recursive_impl ( CudaConstSpan< Scalar >  inp,
CudaMutSpan< Scalar >  out,
CudaMutSpan< Scalar >  workspace,
ConstSpan< IndexNumber perm,
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(), cuda_transpose_recursive_impl(), syten::RepRegister::dim(), syten::LimVec< Type, max, MaxType, type >::end(), if_constexpr, std::is_sorted(), syten::rank(), syten::Deparallelise::rows(), and SYTEN_ASSERT.

Referenced by cuda_transpose_recursive_impl().

+ Here is the call graph for this function:
+ Here is the caller graph for this function: