rmm::cuda_stream_view Class Reference

Strongly-typed non-owning wrapper for CUDA streams with default constructor. More...

#include <cuda_stream_view.hpp>

Public Member Functions
	cuda_stream_view (cuda_stream_view const &)=default
	Default copy constructor.
	cuda_stream_view (cuda_stream_view &&)=default
	Default move constructor.
cuda_stream_view &	operator= (cuda_stream_view const &)=default
	Default copy assignment operator. More...
cuda_stream_view &	operator= (cuda_stream_view &&)=default
	Default move assignment operator. More...
	cuda_stream_view (int)=delete
	cuda_stream_view (std::nullptr_t)=delete
	cuda_stream_view (cudaStream_t stream) noexcept
	Constructor from a cudaStream_t. More...
	cuda_stream_view (cuda::stream_ref stream) noexcept
	Implicit conversion from stream_ref. More...
cudaStream_t	value () const noexcept
	Get the wrapped stream. More...
cudaStream_t	get () const noexcept
	Get the wrapped stream. More...
	operator cudaStream_t () const noexcept
	Implicit conversion to cudaStream_t. More...
	operator cuda::stream_ref () const noexcept
	Implicit conversion to stream_ref. More...
bool	is_per_thread_default () const noexcept
	true if the wrapped stream is the CUDA per-thread default stream More...
bool	is_default () const noexcept
	true if the wrapped stream is explicitly the CUDA legacy default stream More...
void	synchronize () const
	Synchronize the viewed CUDA stream. More...
void	sync () const
	Synchronize the viewed CUDA stream. More...
void	synchronize_no_throw () const noexcept
	Synchronize the viewed CUDA stream. Does not throw if there is an error. More...

Detailed Description

Strongly-typed non-owning wrapper for CUDA streams with default constructor.

This wrapper is simply a "view": it does not own the lifetime of the stream it wraps.

Deprecated:

Use cuda::stream_ref instead.

Constructor & Destructor Documentation

cuda_stream_view() [1/2]

rmm::cuda_stream_view::cuda_stream_view ( cudaStream_t  stream )

noexcept

Constructor from a cudaStream_t.

Parameters

stream

The underlying stream for this view

cuda_stream_view() [2/2]

rmm::cuda_stream_view::cuda_stream_view ( cuda::stream_ref  stream )

noexcept

Implicit conversion from stream_ref.

Parameters

stream

The underlying stream for this view

Member Function Documentation

get()

cudaStream_t rmm::cuda_stream_view::get ( ) const

noexcept

Get the wrapped stream.

Returns

cudaStream_t The underlying stream referenced by this cuda_stream_view

is_default()

bool rmm::cuda_stream_view::is_default ( ) const

noexcept

true if the wrapped stream is explicitly the CUDA legacy default stream

Returns

true if the wrapped stream is explicitly the CUDA legacy default stream

is_per_thread_default()

bool rmm::cuda_stream_view::is_per_thread_default ( ) const

noexcept

true if the wrapped stream is the CUDA per-thread default stream

Returns

true if the wrapped stream is the CUDA per-thread default stream

operator cuda::stream_ref()

rmm::cuda_stream_view::operator cuda::stream_ref ( ) const

noexcept

Implicit conversion to stream_ref.

Returns

stream_ref The underlying stream referenced by this cuda_stream_view

operator cudaStream_t()

rmm::cuda_stream_view::operator cudaStream_t ( ) const

noexcept

Implicit conversion to cudaStream_t.

Returns

cudaStream_t The underlying stream referenced by this cuda_stream_view

operator=() [1/2]

cuda_stream_view& rmm::cuda_stream_view::operator= ( cuda_stream_view &&  )

default

Default move assignment operator.

Returns

cuda_stream_view& Reference to the assigned object

operator=() [2/2]

cuda_stream_view& rmm::cuda_stream_view::operator= ( cuda_stream_view const &  )

default

Default copy assignment operator.

Returns

cuda_stream_view& Reference to the assigned object

sync()

void rmm::cuda_stream_view::sync

(

)

const

Synchronize the viewed CUDA stream.

Calls cudaStreamSynchronize().

Exceptions

rmm::cuda_error

if stream synchronization fails

synchronize()

void rmm::cuda_stream_view::synchronize

(

)

const

Synchronize the viewed CUDA stream.

Calls cudaStreamSynchronize().

Exceptions

rmm::cuda_error

if stream synchronization fails

synchronize_no_throw()

void rmm::cuda_stream_view::synchronize_no_throw ( ) const

noexcept

Synchronize the viewed CUDA stream. Does not throw if there is an error.

Calls cudaStreamSynchronize() and asserts if there is an error.

value()

cudaStream_t rmm::cuda_stream_view::value ( ) const

noexcept

Get the wrapped stream.

Returns

cudaStream_t The underlying stream referenced by this cuda_stream_view

---

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

• cuda_stream_view.hpp