Uni20
Asynchronous tensor and runtime library.
Loading...
Searching...
No Matches
Namespaces | Functions
copy_task.hpp File Reference

Coroutine kernel-dispatch customization for CUDA Tensor copies. More...

#include <uni20/async/cuda_task.hpp>
#include <uni20/backend/cuda/task_awaiters.hpp>
#include <uni20/linalg/async/kernel_task.hpp>
#include <uni20/linalg/backends/cuda/copy.hpp>
Include dependency graph for copy_task.hpp:

Go to the source code of this file.

Namespaces

namespace  uni20
 
namespace  uni20::linalg
 

Functions

template<uni20::MutableMdspecLike OutputMdspan, uni20::MdspecLike InputMdspan>
requires (detail::cuda_reference::SupportedCopyMdspans<OutputMdspan, InputMdspan> && detail::cuda_reference::is_raw_cuda_mdspec<OutputMdspan> && detail::cuda_reference::is_supported_cuda_mdspan<InputMdspan>)
auto uni20::linalg::try_make_kernel_task (CudaReferenceBackend, copy_op const &, OutputMdspan &output, InputMdspan &input) -> KernelTaskAttempt< async::CudaTask >
 Create a CUDA copy task from normalized mdspecs.
 

Detailed Description

Coroutine kernel-dispatch customization for CUDA Tensor copies.