WaitIterateBatch

Function Description

Waits for the return of the asynchronous IterateBatch API and continuously outputs data to the Global Memory.

Prototype

__aicore__ inline void WaitIterateBatch()

Parameters

Parameter	Input/Output	Description
None	None	N/A

Returns

None

Availability

Precautions

This API is used together with the asynchronous IterateBatch API.
Data can only be continuously output to the Global Memory.

Example

matmul::Matmul<aType, bType, cType, biasType> mm;
mm.SetTensorA(queryGm[tensorACoreOffset]);
mm.SetTensorB(keyGm[tensorBCoreOffset + sInnerStart * singleProcessSInnerSize *
      tilingData->attentionScoreOffestStrideParams.matmulHead], true);
mm.SetTail(singleProcessSOuterSize, mmNNum);
mm.template IterateBatch<false>(workspaceGm[tmp_block_idx * mmResUbSize * sInnerLoopTimes],batchA, batchB, false);
// do some others compute
mm.WaitIterateBatch(); // Wait for IterateBatch to complete.
DataCopy(dstUB, GM); // Copy data from GM to UB.

Parent topic: Matmul