Function: get_overflow_status

Applicability

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product/Atlas A3 inference product	√
Atlas A2 training product/Atlas A2 inference product	√
Atlas training product	√
Atlas inference product	√
Atlas 200I/500 A2 inference product	√

Description

Obtains the overflow status of tasks on all streams in the current device and copies the status value to the device memory allocated by the user. This API is asynchronous.

Prototype

C Prototype

aclError aclrtGetOverflowStatus(void *outputAddr, size_t outputSize, aclrtStream stream)

Python Function

ret = acl.rt.get_overflow_status(outputAddr, outputSize, stream)

Parameters

Parameter	Description
outputAddr	Int, user-allocated device memory, which needs to be allocated by calling acl.rt.malloc. To view data on the host, call acl.rt.memcpy or acl.rt.memcpy_async to transfer data from the device to the host.
outputSize	Int, size of the device memory to be allocated, in bytes. The fixed size is 64 bytes.
stream	Int, pointer address of a stream to deliver an overflow status query task.

Return Value

Return Value	Description
ret	Int, error code. 0 on success; else, failure.

Restrictions

This API is asynchronous. The API call delivers a task rather than executes a task. After this API is called, call the synchronization API (for example, acl.rt.synchronize_stream) to ensure that the task is complete.

For the Atlas A2 training product/Atlas A2 inference product, the overflow status queried by calling this API is at the process level.

For the Atlas A3 training product/Atlas A3 inference product, the overflow status queried by calling this API is at the process level.

Parent topic: Device Management