* Copyright (c) 2025-2026 Huawei Technologies Co., Ltd.
* This program is free software, you can redistribute it and/or modify it under the terms and conditions of
* CANN Open Software License Agreement Version 2.0 (the "License").
* Please refer to the License for details. You may not use this file except in compliance with the License.
* THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
* INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
* See LICENSE in the root of the software repository for the full text of the License.
*/
* \file lin_space_small_shape.h
* \brief
*/
#ifndef LINSPACE_SMALL_SHAPE_H
#define LINSPACE_SMALL_SHAPE_H
#include "lin_space_base.h"
namespace LinSpace {
using namespace AscendC;
template <typename T>
class LinSpaceSmallShape : public LinSpaceBase<T> {
public:
__aicore__ inline LinSpaceSmallShape(){};
__aicore__ inline void Init(
GM_ADDR start, GM_ADDR stop, GM_ADDR num, GM_ADDR output, GM_ADDR workspace,
const LinSpaceTilingData* tilingData);
__aicore__ inline void Process();
private:
__aicore__ inline void Compute();
__aicore__ inline void CopyOut();
constexpr static int32_t bufferNum = 1;
constexpr static int64_t HALF_NUM = 2;
constexpr static int32_t outSize = 64 * 4;
constexpr static int64_t blockSize = 32;
constexpr static int64_t elementPerBlock = blockSize / sizeof(T);
private:
TPipe pipe;
TQue<QuePosition::VECOUT, bufferNum> outQueue;
GlobalTensor<T> outputGm;
int32_t blockIdx = 0;
T blockOffset = 0;
LinSpaceTilingData m_tilingData;
};
template <typename T>
__aicore__ inline void LinSpaceSmallShape<T>::Init(
GM_ADDR start, GM_ADDR stop, GM_ADDR num, GM_ADDR output, GM_ADDR workspace, const LinSpaceTilingData* tilingData)
{
blockIdx = GetBlockIdx();
outputGm.SetGlobalBuffer((__gm__ T*)output);
this->ParseTilingData(tilingData, m_tilingData);
pipe.InitBuffer(outQueue, bufferNum, outSize);
}
template <typename T>
__aicore__ inline void LinSpaceSmallShape<T>::Process()
{
if (m_tilingData.num == 0 || blockIdx >= m_tilingData.realCoreNum) {
return;
}
Compute();
CopyOut();
}
template <typename T>
__aicore__ inline void LinSpaceSmallShape<T>::Compute()
{
LocalTensor<T> outLocal = outQueue.AllocTensor<T>();
int64_t halfway = m_tilingData.num / HALF_NUM;
if (m_tilingData.num == 1) {
outLocal.SetValue(0, m_tilingData.start);
} else {
for (int64_t idx = 0; idx < m_tilingData.num; idx++) {
if (idx < halfway) {
outLocal.SetValue(idx, m_tilingData.start + m_tilingData.scalar * idx);
} else {
outLocal.SetValue(idx, m_tilingData.stop - m_tilingData.scalar * (m_tilingData.num - idx - 1));
}
}
}
outQueue.EnQue(outLocal);
}
template <typename T>
__aicore__ inline void LinSpaceSmallShape<T>::CopyOut()
{
LocalTensor<T> outLocal = outQueue.DeQue<T>();
if constexpr (IsDataCopyPadSupport()) {
DataCopyExtParams copyParams = {1, 0, 0, 0, 0};
copyParams.blockLen = m_tilingData.tailNum * sizeof(T);
DataCopyPad(outputGm, outLocal, copyParams);
} else {
int64_t tailAlignNum = this->CeilDiv(m_tilingData.tailNum, elementPerBlock) * elementPerBlock;
DataCopy(outputGm, outLocal, tailAlignNum);
}
outQueue.FreeTensor(outLocal);
}
}
#endif