Function: set_op_execute_timeout_v2

Applicable Products

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	√

Description

Sets the timeout interval (in microseconds) for operator execution. If a timeout interval is carried during operator delivery, the timeout interval takes precedence over the timeout interval set by this API.

Prototype

C Prototype

aclError aclrtSetOpExecuteTimeOutV2(uint64_t timeout,  uint64_t *actualTimeout)

Python Function

actual_time_out, ret = acl.rt.set_op_execute_timeout_v2(timeout)

Parameters

Parameter	Description
timeout	Int, timeout interval, in microseconds. 0 means to use the maximum timeout interval.

Return Values

Return Value	Description
actual_time_out	Int, actual timeout, in microseconds.
ret	Int, error code. 0 on success; else, failure.

Restrictions

For the product models below, if the interval obtained by calling the aclrtGetOpTimeoutInterval API is less than 100,000 microseconds and the timeout parameter is set to 0, AI Core operators will never time out. In this case, the value of the actualTimeout parameter is the maximum value of uint64_t.
Ascend 950PR/Ascend 950DT

Atlas A3 training products/Atlas A3 inference products

Atlas A2 training products/Atlas A2 inference products
For the following product models, only the timeout interval for AI Core operator execution can be set by calling this API:
Atlas 200I/500 A2 inference products

Atlas inference products

Atlas training products

Parent topic: Execution Control