Awaiter that establishes or changes a CUDA task's device affinity. More...

#include <task_awaiters.hpp>

Inheritance diagram for uni20::async::CudaDeviceAwaiter:

Collaboration diagram for uni20::async::CudaDeviceAwaiter:

Public Member Functions
constexpr bool	await_ready () const noexcept

constexpr void	await_resume () const noexcept

bool	await_suspend (std::coroutine_handle< CudaTaskPromise > handle) const

constexpr	CudaDeviceAwaiter (int device) noexcept

Detailed Description

Awaiter that establishes or changes a CUDA task's device affinity.

The scheduler-established current CUDA device determines whether selection needs another activation. A different device resubmits through the recorded scheduler; an already-current device resumes immediately after recording affinity.

Constructor & Destructor Documentation

◆ CudaDeviceAwaiter()

constexpr uni20::async::CudaDeviceAwaiter::CudaDeviceAwaiter ( int device )

inlineexplicitconstexprnoexcept

Member Function Documentation

◆ await_ready()

constexpr bool uni20::async::CudaDeviceAwaiter::await_ready ( ) const

inlineconstexprnoexcept

◆ await_resume()

constexpr void uni20::async::CudaDeviceAwaiter::await_resume ( ) const

inlineconstexprnoexcept

◆ await_suspend()

bool uni20::async::CudaDeviceAwaiter::await_suspend ( std::coroutine_handle< CudaTaskPromise > handle ) const

inline

The documentation for this class was generated from the following file:

cuda/task_awaiters.hpp