HcclGetCommAsyncError

Applicability

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

For Atlas A2 training product/Atlas A2 inference product, only the Atlas 800T A2 training server, Atlas 900 A2 PoD cluster basic unit, and Atlas 200T A2 Box16 heterogeneous subrack are supported.

For the Atlas inference products, only the Atlas 300I Duo inference card is supported.

Function

If the cluster information shows that the communication link of the device network port is unstable or network congestion occurs, "error cqe" is printed in the device log. This error is called "RDMA ERROR CQE".

In the current version, this API can only be used to check whether the "RDMA ERROR CQE" error exists in the communicator.

This API is a synchronous API. That is, after this API is called, you need to wait for the returned result.

Prototype

HcclResult HcclGetCommAsyncError(HcclComm comm, HcclResult *asyncError)

Parameters

Parameter	Input/Output	Description
comm	Input	Communicator for which you want to check whether error information exists. For details about the definition of the HcclComm type, see HcclComm.
asyncError	Output	0: No error occurs in the communicator. 21: An RDMA ERROR CQE error occurs in the communicator.

Parameter

Input/Output

Description

comm

Input

Communicator for which you want to check whether error information exists.

For details about the definition of the HcclComm type, see HcclComm.

asyncError

Output

0: No error occurs in the communicator.
21: An RDMA ERROR CQE error occurs in the communicator.

Returns

For details, see the HcclResult type. In the current version, only the HCCL_E_REMOTE error type is returned.

Constraints

This API can be called only after a communicator is created.
This API cannot be called after the communicator is destroyed.

Parent topic: C APIs