Files
enginex-ascend-910-vllm/csrc/common/include/kernel/mma.h
Sun Ruoxi 7f8a1b1f7a init v0.23.0
Signed-off-by: Sun Ruoxi <sunruoxi@4paradigm.com>
2026-08-27 15:11:51 +08:00

79 lines
3.2 KiB
C++
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

/**
 * Copyright (c) 2025 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 mma.h
* \brief
*/
#ifndef INCLUDE_MMA_H
#define INCLUDE_MMA_H
#include "hardware.h"
#include "kernel_tensor.h"
template <ArchType ArchTag, typename ElementA, typename ElementB, typename AccDTypeC, bool IsTransposeA>
struct mmad {
__aicore__ mmad(AscendC::LocalTensor<AccDTypeC> l0cTensor,
AscendC::LocalTensor<ElementA> l0aTensor,
AscendC::LocalTensor<ElementB> l0bTensor,
uint32_t mTileActual,
uint32_t nTileActual,
uint32_t kPartActual,
bool initC) {};
__aicore__ mmad(AscendC::LocalTensor<AccDTypeC> l0cTensor,
AscendC::LocalTensor<ElementA> l0aTensor,
AscendC::LocalTensor<ElementB> l0bTensor,
uint64_t biasBt,
uint32_t mTileActual,
uint32_t nTileActual,
uint32_t kPartActual,
bool initC) {};
};
// Partial specialization for V220, int8_t, not_vector_A, not TransposeA
template <ArchType ArchTag, typename AccDTypeC, typename ElementA, typename ElementB>
struct mmad<ArchTag, ElementA, ElementB, AccDTypeC, false> {
__aicore__ mmad(AscendC::LocalTensor<AccDTypeC> l0cTensor,
AscendC::LocalTensor<ElementA> l0aTensor,
AscendC::LocalTensor<ElementB> l0bTensor,
uint32_t mTileActual,
uint32_t nTileActual,
uint32_t kPartActual,
bool initC)
{
AscendC::Mmad(l0cTensor,
l0aTensor,
l0bTensor,
AscendC::MmadParams(mTileActual, nTileActual, kPartActual, 0, false, initC));
};
__aicore__ mmad(AscendC::LocalTensor<AccDTypeC> l0cTensor,
AscendC::LocalTensor<ElementA> l0aTensor,
AscendC::LocalTensor<ElementB> l0bTensor,
uint64_t biasBt,
uint32_t mTileActual,
uint32_t nTileActual,
uint32_t kPartActual,
bool initC)
{
AscendC::LocalTensor<ElementA> biasTensor;
biasTensor.InitBuffer(biasBt, mTileActual);
biasTensor.address_.logicPos = static_cast<uint8_t>(AscendC::TPosition::C2);
AscendC::Mmad(l0cTensor,
l0aTensor,
l0bTensor,
biasTensor,
AscendC::MmadParams(mTileActual, nTileActual, kPartActual, 0, false, initC));
};
};
#endif