CUDA Runtime API :: CUDA Toolkit Documentation

4.4. Stream Management

This section describes the stream management functions of the CUDA runtime application programming interface.

Typedefs

typedef void(CUDART_CB* cudaStreamCallback_t )( cudaStream_t stream, cudaError_t status, void* userData )

Functions

__host__ cudaError_t cudaStreamAddCallback ( cudaStream_t stream, cudaStreamCallback_t callback, void* userData, unsigned int flags ): Add a callback to a compute stream.
__host__ cudaError_t cudaStreamAttachMemAsync ( cudaStream_t stream, void* devPtr, size_t length = 0, unsigned int flags = cudaMemAttachSingle ): Attach memory to a stream asynchronously.
__host__ cudaError_t cudaStreamCreate ( cudaStream_t* pStream ): Create an asynchronous stream.
__host__ __device__ cudaError_t cudaStreamCreateWithFlags ( cudaStream_t* pStream, unsigned int flags ): Create an asynchronous stream.
__host__ cudaError_t cudaStreamCreateWithPriority ( cudaStream_t* pStream, unsigned int flags, int priority ): Create an asynchronous stream with the specified priority.
__host__ __device__ cudaError_t cudaStreamDestroy ( cudaStream_t stream ): Destroys and cleans up an asynchronous stream.
__host__ cudaError_t cudaStreamGetFlags ( cudaStream_t hStream, unsigned int* flags ): Query the flags of a stream.
__host__ cudaError_t cudaStreamGetPriority ( cudaStream_t hStream, int* priority ): Query the priority of a stream.
__host__ cudaError_t cudaStreamQuery ( cudaStream_t stream ): Queries an asynchronous stream for completion status.
__host__ cudaError_t cudaStreamSynchronize ( cudaStream_t stream ): Waits for stream tasks to complete.
__host__ __device__ cudaError_t cudaStreamWaitEvent ( cudaStream_t stream, cudaEvent_t event, unsigned int flags ): Make a compute stream wait on an event.

Typedefs

void(CUDART_CB* cudaStreamCallback_t )( cudaStream_t stream, cudaError_t status, void* userData ): Type of stream callback functions.

Functions

__host__ cudaError_t cudaStreamAddCallback ( cudaStream_t stream, cudaStreamCallback_t callback, void* userData, unsigned int flags )

Add a callback to a compute stream.

Parameters

stream: - Stream to add callback to
callback: - The function to call once preceding stream operations are complete
userData: - User specified data to be passed to the callback function
flags: - Reserved for future use, must be 0

Returns

cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorNotSupported

Description

Adds a callback to be called on the host after all currently enqueued items in the stream have completed. For each cudaStreamAddCallback call, a callback will be executed exactly once. The callback will block later work in the stream until it is finished.

The callback may be passed cudaSuccess or an error code. In the event of a device error, all subsequently executed callbacks will receive an appropriate cudaError_t.

Callbacks must not make any CUDA API calls. Attempting to use CUDA APIs will result in cudaErrorNotPermitted. Callbacks must not perform any synchronization that may depend on outstanding device work or other callbacks that are not mandated to run earlier. Callbacks without a mandated order (in independent streams) execute in undefined order and may be serialized.

For the purposes of Unified Memory, callback execution makes a number of guarantees:

The callback stream is considered idle for the duration of the callback. Thus, for example, a callback may always use memory attached to the callback stream.
The start of execution of a callback has the same effect as synchronizing an event recorded in the same stream immediately prior to the callback. It thus synchronizes streams which have been "joined" prior to the callback.
Adding device work to any stream does not have the effect of making the stream active until all preceding callbacks have executed. Thus, for example, a callback might use global attached memory even if work has been added to another stream, if it has been properly ordered with an event.
Completion of a callback does not cause a stream to become active except as described above. The callback stream will remain idle if no device work follows the callback, and will remain idle across consecutive callbacks without device work in between. Thus, for example, stream synchronization can be done by signaling from a callback at the end of the stream.

Note:

This function uses standard default stream semantics.
Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamAttachMemAsync ( cudaStream_t stream, void* devPtr, size_t length = 0, unsigned int flags = cudaMemAttachSingle )

Attach memory to a stream asynchronously.

Parameters

stream: - Stream in which to enqueue the attach operation
devPtr: - Pointer to memory (must be a pointer to managed memory)
length: - Length of memory (must be zero, defaults to zero)
flags: - Must be one of cudaMemAttachGlobal, cudaMemAttachHost or cudaMemAttachSingle (defaults to cudaMemAttachSingle)

Returns

cudaSuccess, cudaErrorNotReady, cudaErrorInvalidValue cudaErrorInvalidResourceHandle

Description

Enqueues an operation in stream to specify stream association of length bytes of memory starting from devPtr. This function is a stream-ordered operation, meaning that it is dependent on, and will only take effect when, previous work in stream has completed. Any previous association is automatically replaced.

devPtr must point to an address within managed memory space declared using the __managed__ keyword or allocated with cudaMallocManaged.

length must be zero, to indicate that the entire allocation's stream association is being changed. Currently, it's not possible to change stream association for a portion of an allocation. The default value for length is zero.

The stream association is specified using flags which must be one of cudaMemAttachGlobal, cudaMemAttachHost or cudaMemAttachSingle. The default value for flags is cudaMemAttachSingle If the cudaMemAttachGlobal flag is specified, the memory can be accessed by any stream on any device. If the cudaMemAttachHost flag is specified, the program makes a guarantee that it won't access the memory on the device from any stream on a device that has a zero value for the device attribute cudaDevAttrConcurrentManagedAccess. If the cudaMemAttachSingle flag is specified and stream is associated with a device that has a zero value for the device attribute cudaDevAttrConcurrentManagedAccess, the program makes a guarantee that it will only access the memory on the device from stream. It is illegal to attach singly to the NULL stream, because the NULL stream is a virtual global stream and not a specific stream. An error will be returned in this case.

When memory is associated with a single stream, the Unified Memory system will allow CPU access to this memory region so long as all operations in stream have completed, regardless of whether other streams are active. In effect, this constrains exclusive ownership of the managed memory region by an active GPU to per-stream activity instead of whole-GPU activity.

Accessing memory on the device from streams that are not associated with it will produce undefined results. No error checking is performed by the Unified Memory system to ensure that kernels launched into other streams do not access this region.

It is a program's responsibility to order calls to cudaStreamAttachMemAsync via events, synchronization or other means to ensure legal access to memory at all times. Data visibility and coherency will be changed appropriately for all kernels which follow a stream-association change.

If stream is destroyed while data is associated with it, the association is removed and the association reverts to the default visibility of the allocation as specified at cudaMallocManaged. For __managed__ variables, the default association is always cudaMemAttachGlobal. Note that destroying a stream is an asynchronous operation, and as a result, the change to default association won't happen until all work in the stream has completed.

Note:

Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamCreate ( cudaStream_t* pStream )

Create an asynchronous stream.

Parameters

pStream: - Pointer to new stream identifier

Returns

cudaSuccess, cudaErrorInvalidValue

Description

Creates a new asynchronous stream.

Note:

Note that this function may also return error codes from previous, asynchronous launches.

__host__ __device__ cudaError_t cudaStreamCreateWithFlags ( cudaStream_t* pStream, unsigned int flags )

Create an asynchronous stream.

Parameters

pStream: - Pointer to new stream identifier
flags: - Parameters for stream creation

Returns

cudaSuccess, cudaErrorInvalidValue

Description

Creates a new asynchronous stream. The flags argument determines the behaviors of the stream. Valid values for flags are

cudaStreamDefault: Default stream creation flag.
cudaStreamNonBlocking: Specifies that work running in the created stream may run concurrently with work in stream 0 (the NULL stream), and that the created stream should perform no implicit synchronization with stream 0.

Note:

Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamCreateWithPriority ( cudaStream_t* pStream, unsigned int flags, int priority )

Create an asynchronous stream with the specified priority.

Parameters

pStream: - Pointer to new stream identifier
flags: - Flags for stream creation. See cudaStreamCreateWithFlags for a list of valid flags that can be passed
priority: - Priority of the stream. Lower numbers represent higher priorities. See cudaDeviceGetStreamPriorityRange for more information about the meaningful stream priorities that can be passed.

Returns

cudaSuccess, cudaErrorInvalidValue

Description

Creates a stream with the specified priority and returns a handle in pStream. This API alters the scheduler priority of work in the stream. Work in a higher priority stream may preempt work already executing in a low priority stream.

priority follows a convention where lower numbers represent higher priorities. '0' represents default priority. The range of meaningful numerical priorities can be queried using cudaDeviceGetStreamPriorityRange. If the specified priority is outside the numerical range returned by cudaDeviceGetStreamPriorityRange, it will automatically be clamped to the lowest or the highest number in the range.

Note:

Note that this function may also return error codes from previous, asynchronous launches.
Stream priorities are supported only on GPUs with compute capability 3.5 or higher.
In the current implementation, only compute kernels launched in priority streams are affected by the stream's priority. Stream priorities have no effect on host-to-device and device-to-host memory operations.

__host__ __device__ cudaError_t cudaStreamDestroy ( cudaStream_t stream )

Destroys and cleans up an asynchronous stream.

Parameters

stream: - Stream identifier

Returns

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

Description

Destroys and cleans up the asynchronous stream specified by stream.

In case the device is still doing work in the stream stream when cudaStreamDestroy() is called, the function will return immediately and the resources associated with stream will be released automatically once the device has completed all work in stream.

Note:

This function uses standard default stream semantics.
Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamGetFlags ( cudaStream_t hStream, unsigned int* flags )

Query the flags of a stream.

Parameters

hStream: - Handle to the stream to be queried
flags: - Pointer to an unsigned integer in which the stream's flags are returned

Returns

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

Description

Query the flags of a stream. The flags are returned in flags. See cudaStreamCreateWithFlags for a list of valid flags.

Note:

This function uses standard default stream semantics.
Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamGetPriority ( cudaStream_t hStream, int* priority )

Query the priority of a stream.

Parameters

hStream: - Handle to the stream to be queried
priority: - Pointer to a signed integer in which the stream's priority is returned

Returns

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

Description

Query the priority of a stream. The priority is returned in in priority. Note that if the stream was created with a priority outside the meaningful numerical range returned by cudaDeviceGetStreamPriorityRange, this function returns the clamped priority. See cudaStreamCreateWithPriority for details about priority clamping.

Note:

Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamQuery ( cudaStream_t stream )

Queries an asynchronous stream for completion status.

Parameters

stream: - Stream identifier

Returns

cudaSuccess, cudaErrorNotReady, cudaErrorInvalidResourceHandle

Description

Returns cudaSuccess if all operations in stream have completed, or cudaErrorNotReady if not.

For the purposes of Unified Memory, a return value of cudaSuccess is equivalent to having called cudaStreamSynchronize().

Note:

This function uses standard default stream semantics.
Note that this function may also return error codes from previous, asynchronous launches.

__host__ cudaError_t cudaStreamSynchronize ( cudaStream_t stream )

Waits for stream tasks to complete.

Parameters

stream: - Stream identifier

Returns

cudaSuccess, cudaErrorInvalidResourceHandle

Description

Blocks until stream has completed all operations. If the cudaDeviceScheduleBlockingSync flag was set for this device, the host thread will block until the stream is finished with all of its tasks.

Note:

This function uses standard default stream semantics.
Note that this function may also return error codes from previous, asynchronous launches.

__host__ __device__ cudaError_t cudaStreamWaitEvent ( cudaStream_t stream, cudaEvent_t event, unsigned int flags )

Make a compute stream wait on an event.

Parameters

stream: - Stream to wait
event: - Event to wait on
flags: - Parameters for the operation (must be 0)

Returns

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

Description

Makes all future work submitted to stream wait until event reports completion before beginning execution. This synchronization will be performed efficiently on the device. The event event may be from a different context than stream, in which case this function will perform cross-device synchronization.

The stream stream will wait only for the completion of the most recent host call to cudaEventRecord() on event. Once this call has returned, any functions (including cudaEventRecord() and cudaEventDestroy()) may be called on event again, and the subsequent calls will not have any effect on stream.

If cudaEventRecord() has not been called on event, this call acts as if the record has already completed, and so is a functional no-op.

Note:

This function uses standard default stream semantics.
Note that this function may also return error codes from previous, asynchronous launches.