tf::cudaStream Class Reference

‍** More...

#include <cuda_stream.hpp>

Public Member Functions
	cudaStream (cudaStream_t stream)
	constructs an RAII-styled object from the given CUDA stream
	cudaStream ()
	constructs an RAII-styled object for a new CUDA stream
	cudaStream (const cudaStream &)=delete
	disabled copy constructor
	cudaStream (cudaStream &&rhs)
	move constructor
	~cudaStream ()
	destructs the CUDA stream
cudaStream &	operator= (const cudaStream &)=delete
	disabled copy assignment
cudaStream &	operator= (cudaStream &&rhs)
	move assignment
void	reset (cudaStream_t stream=nullptr)
	replaces the managed stream
	operator cudaStream_t () const
	implicit conversion to the native CUDA stream (cudaStream_t)
void	synchronize () const
	synchronizes the associated stream
void	begin_capture (cudaStreamCaptureMode m=cudaStreamCaptureModeGlobal) const
	begins graph capturing on the stream
cudaGraph_t	end_capture () const
	ends graph capturing on the stream
void	record (cudaEvent_t event) const
	records an event on the stream
void	wait (cudaEvent_t event) const
	waits on an event

Detailed Description

‍**

‍** ‍**

class to create a CUDA stream in an RAII-styled wrapper

A cudaStream object is an RAII-styled wrapper over a native CUDA stream (cudaStream_t). A cudaStream object is move-only.

Constructor & Destructor Documentation

cudaStream() [1/2]

tf::cudaStream::cudaStream ( cudaStream_t  stream )

inlineexplicit

constructs an RAII-styled object from the given CUDA stream

Constructs a cudaStream object which owns stream.

cudaStream() [2/2]

tf::cudaStream::cudaStream ( )

inline

constructs an RAII-styled object for a new CUDA stream

Equivalently calling cudaStreamCreate to create a stream.

Member Function Documentation

begin_capture()

void tf::cudaStream::begin_capture ( cudaStreamCaptureMode  m = cudaStreamCaptureModeGlobal ) const

inline

begins graph capturing on the stream

When a stream is in capture mode, all operations pushed into the stream will not be executed, but will instead be captured into a graph, which will be returned via cudaStream::end_capture.

A thread's mode can be one of the following:

• cudaStreamCaptureModeGlobal: This is the default mode. If the local thread has an ongoing capture sequence that was not initiated with cudaStreamCaptureModeRelaxed at cuStreamBeginCapture, or if any other thread has a concurrent capture sequence initiated with cudaStreamCaptureModeGlobal, this thread is prohibited from potentially unsafe API calls.
• cudaStreamCaptureModeThreadLocal: If the local thread has an ongoing capture sequence not initiated with cudaStreamCaptureModeRelaxed, it is prohibited from potentially unsafe API calls. Concurrent capture sequences in other threads are ignored.
• cudaStreamCaptureModeRelaxed: The local thread is not prohibited from potentially unsafe API calls. Note that the thread is still prohibited from API calls which necessarily conflict with stream capture, for example, attempting cudaEventQuery on an event that was last recorded inside a capture sequence.

end_capture()

cudaGraph_t tf::cudaStream::end_capture ( ) const

inline

ends graph capturing on the stream

Equivalently calling cudaStreamEndCapture to end capture on stream and returning the captured graph. Capture must have been initiated on stream via a call to cudaStream::begin_capture. If capture was invalidated, due to a violation of the rules of stream capture, then a NULL graph will be returned.

operator cudaStream_t()

tf::cudaStream::operator cudaStream_t ( ) const

inline

implicit conversion to the native CUDA stream (cudaStream_t)

Returns the underlying stream of type cudaStream_t.

record()

void tf::cudaStream::record ( cudaEvent_t  event ) const

inline

records an event on the stream

Equivalently calling cudaEventRecord to record an event on this stream, both of which must be on the same CUDA context.

reset()

void tf::cudaStream::reset ( cudaStream_t  stream = nullptr )

inline

replaces the managed stream

Destructs the managed stream and resets it to the given stream.

synchronize()

void tf::cudaStream::synchronize ( ) const

inline

synchronizes the associated stream

Equivalently calling cudaStreamSynchronize to block until this stream has completed all operations.

wait()

void tf::cudaStream::wait ( cudaEvent_t  event ) const

inline

waits on an event

Equivalently calling cudaStreamWaitEvent to make all future work submitted to stream wait for all work captured in event.

---

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

• cuda_stream.hpp