* 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 sinh_common_impl.h
* \brief
*/
#if !defined(__ASCENDC_INCLUDE_INTERNAL_HEADERS__)
#pragma message( \
"impl/adv_api/detail/math/sinh/sinh_common_impl.h is an internal header file and must not be used directly. Functions or variables defined in this file may be removed in the future. Please use \"#include \"adv_api/math/sinh.h\"\" and use public functions or variables defined in interface headers files.")
#define __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#define __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_MATH_SINH_SINH_COMMON_IMPL_H__
#endif
#ifndef IMPL_MATH_SINH_SINH_COMMON_IMPL_H
#define IMPL_MATH_SINH_SINH_COMMON_IMPL_H
#include "kernel_basic_intf.h"
#include "kernel_tensor.h"
#include "kernel_pop_stack_buffer.h"
#include "../../common/check.h"
#include "../math_common_impl.h"
#ifdef ASCENDC_CPU_DEBUG
#include "../../api_check/kernel_check/math/sinh/sinh_check.h"
#endif
#include "../../api_check/kernel_api_check.h"
namespace AscendC {
constexpr float SINH_NEG_LN_TWO = -0.69314718055994530941723212145818;
constexpr float SINH_NEG_ONE = -1.0;
constexpr float SINH_POINT_FIVE = 0.5;
constexpr uint32_t SINH_HALF_CALC_PROC = 4;
constexpr uint32_t SINH_FLOAT_CALC_PROC = 1;
template <typename T>
__aicore__ inline void SinhCompute(
const LocalTensor<T>& dst, const LocalTensor<T>& src, const LocalTensor<uint8_t>& tmpBuffer, uint32_t offset)
{
UnaryRepeatParams unaryParams;
BinaryRepeatParams binaryParams;
const LocalTensor<T>& tmpFloatBuffer1 = tmpBuffer.ReinterpretCast<T>();
Muls<T, false>(tmpFloatBuffer1, src, static_cast<T>(SINH_NEG_ONE), MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Adds<T, false>(tmpFloatBuffer1, tmpFloatBuffer1, static_cast<T>(SINH_NEG_LN_TWO), MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Exp<T, false>(tmpFloatBuffer1, tmpFloatBuffer1, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Adds<T, false>(dst, src, static_cast<T>(SINH_NEG_LN_TWO), MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Exp<T, false>(dst, dst, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Sub<T, false>(dst, dst, tmpFloatBuffer1, MASK_PLACEHOLDER, 1, binaryParams);
PipeBarrier<PIPE_V>();
}
template <>
__aicore__ inline void SinhCompute(
const LocalTensor<half>& dst, const LocalTensor<half>& src, const LocalTensor<uint8_t>& tmpBuffer, uint32_t offset)
{
UnaryRepeatParams unaryParams;
BinaryRepeatParams binaryParams;
const LocalTensor<float>& tmpFloatBuffer1 = tmpBuffer.ReinterpretCast<float>();
const LocalTensor<float>& tmpFloatBuffer2 = tmpFloatBuffer1[offset];
Cast<float, half, false>(
tmpFloatBuffer1, src, RoundMode::CAST_NONE, MASK_PLACEHOLDER, 1,
{1, 1, DEFAULT_REPEAT_STRIDE, HALF_DEFAULT_REPEAT_STRIDE});
PipeBarrier<PIPE_V>();
Adds<float, false>(tmpFloatBuffer2, tmpFloatBuffer1, SINH_NEG_LN_TWO, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Exp<float, false>(tmpFloatBuffer2, tmpFloatBuffer2, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Muls<float, false>(tmpFloatBuffer1, tmpFloatBuffer1, SINH_NEG_ONE, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Adds<float, false>(tmpFloatBuffer1, tmpFloatBuffer1, SINH_NEG_LN_TWO, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Exp<float, false>(tmpFloatBuffer1, tmpFloatBuffer1, MASK_PLACEHOLDER, 1, unaryParams);
PipeBarrier<PIPE_V>();
Sub<float, false>(tmpFloatBuffer2, tmpFloatBuffer2, tmpFloatBuffer1, MASK_PLACEHOLDER, 1, binaryParams);
PipeBarrier<PIPE_V>();
Cast<half, float, false>(
dst, tmpFloatBuffer2, RoundMode::CAST_NONE, MASK_PLACEHOLDER, 1,
{1, 1, HALF_DEFAULT_REPEAT_STRIDE, DEFAULT_REPEAT_STRIDE});
PipeBarrier<PIPE_V>();
}
template <typename T, bool isReuseSource = false>
__aicore__ inline void SinhImpl(
const LocalTensor<T>& dstTensor, const LocalTensor<T>& srcTensor, const LocalTensor<uint8_t>& sharedTmpBuffer,
const uint32_t calCount)
{
if ASCEND_IS_AIC {
return;
}
CHECK_FUNC_HIGHLEVEL_API(Sinh, (T, isReuseSource), (dstTensor, srcTensor, sharedTmpBuffer, calCount));
uint32_t tmpBufferSize = sharedTmpBuffer.GetSize();
uint32_t splitCount = tmpBufferSize / sizeof(T);
if constexpr (sizeof(T) == sizeof(half)) {
splitCount = splitCount / SINH_HALF_CALC_PROC / ONE_BLK_SIZE * ONE_BLK_SIZE;
} else {
splitCount = splitCount / SINH_FLOAT_CALC_PROC / ONE_BLK_SIZE * ONE_BLK_SIZE;
}
CheckTmpBufferSize(splitCount, 0, tmpBufferSize);
const uint32_t loopCount = calCount / splitCount;
const uint32_t calcTail = calCount % splitCount;
SetMaskCount();
SetVectorMask<T, MaskMode::COUNTER>(0, splitCount);
for (uint32_t i = 0; i < loopCount; ++i) {
SinhCompute(dstTensor[i * splitCount], srcTensor[i * splitCount], sharedTmpBuffer, splitCount);
}
if (calcTail > 0) {
SetVectorMask<T, MaskMode::COUNTER>(0, calcTail);
SinhCompute(dstTensor[loopCount * splitCount], srcTensor[loopCount * splitCount], sharedTmpBuffer, splitCount);
}
SetMaskNorm();
ResetMask();
}
template <typename T, bool isReuseSource = false>
__aicore__ inline void SinhImpl(
const LocalTensor<T>& dstTensor, const LocalTensor<T>& srcTensor, const uint32_t calCount)
{
if ASCEND_IS_AIC {
return;
}
LocalTensor<uint8_t> sharedTmpBuffer;
bool ans = PopStackBuffer<uint8_t, TPosition::LCM>(sharedTmpBuffer);
ASCENDC_ASSERT((ans), { KERNEL_LOG(KERNEL_ERROR, "PopStackBuffer Error!"); });
SinhImpl<T, isReuseSource>(dstTensor, srcTensor, sharedTmpBuffer, calCount);
}
}
#endif
#if defined(__UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_MATH_SINH_SINH_COMMON_IMPL_H__)
#undef __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#undef __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_MATH_SINH_SINH_COMMON_IMPL_H__
#endif