HasIdleBuffer

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 AI Core	√
Atlas inference product Vector Core	x
Atlas training product	√

Function Usage

Queries whether there are free buffers in the queue.

Prototype

      
           __aicore__ inline bool HasIdleBuffer()

Parameters

None

Constraints

This API does not support inplace tensor operations. This means depth of TQue is set to 0.

Returns

true: indicates that free memory exists in the queue.
false: No free memory exists in the queue.

Example

      
           // Four buffers have been allocated in the current queue.
AscendC::TPipe pipe;
AscendC::TQue<AscendC::TPosition::VECOUT, 1> que;
int num = 4;
int len = 1024;
pipe.InitBuffer(que, num, len);
bool ret = que.HasIdleBuffer(); // AllocTensor is not called, and the return is true.
AscendC::LocalTensor<half> tensor1 = que.AllocTensor<half>();
ret = que.HasIdleBuffer(); // One buffer has been allocated by AllocTensor, and the return is true.
AscendC::LocalTensor<half> tensor2 = que.AllocTensor<half>();
AscendC::LocalTensor<half> tensor3 = que.AllocTensor<half>();
AscendC::LocalTensor<half> tensor4 = que.AllocTensor<half>();
ret = que.HasIdleBuffer(); // Four buffers have been allocated by AllocTensor. Currently, there is no idle buffer, and the return is false. An error is reported when AllocTensor is called.

Parent topic: TQue