Product	Supported (√/x)
Ascend 950PR/Ascend 950DT	√
Atlas A3 training products/Atlas A3 inference products	√
Atlas A2 training products/Atlas A2 inference products	√
Atlas training products	√
Atlas inference products	√
Atlas 200I/500 A2 inference products	√

Obtains the execution end time of an event (time elapsed since the AI processor system is started).

This API must be used together with other key APIs. The API call sequence is as follows:

Call the acl.rt.create_event or acl.rt.create_event_with_flag API to create an event.
Call the acl.rt.record_event API to record the event in the stream.
Call the acl.rt.synchronize_stream API to block the running of the application until all tasks in the specified stream are complete.
Call acl.rt.event_get_timestamp to obtain the execution time of the event.

C Prototype

aclError aclrtEventGetTimestamp(aclrtEvent event, uint64_t *timestamp)

Python Function

timestamp, ret = acl.rt.event_get_timestamp(event)

Parameter	Description
event	Int, event to be queried.

Return Value	Description
timestamp	Int, time when the event execution ends, in microseconds.
ret	Int, 0 on success; else, failure.

Function: event_get_timestamp