* 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 kernel_operator_mm_impl.h
* \brief
*/
#if !defined(__ASCENDC_INCLUDE_INTERNAL_HEADERS__)
#pragma message("impl/basic_api/dav_m200/kernel_operator_mm_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 \"basic_api/kernel_tensor.h\"\" and use public functions or variables defined in interface headers files.")
#define __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#define __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_OPERATOR_MM_IMPL_H__
#endif
#ifndef ASCENDC_MODULE_OPERATOR_MM_IMPL_H
#define ASCENDC_MODULE_OPERATOR_MM_IMPL_H
#include "kernel_struct_mm.h"
namespace AscendC {
* LoadData 2dv2 *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void CheckLoadData2DType()
{
ASCENDC_ASSERT((SupportType<T, uint8_t, int8_t, uint16_t, int16_t, half, int4b_t>()), {KERNEL_LOG(KERNEL_ERROR, "Failed to "
"check dtype in LoadData with LoadData2DParams, current api support dtype combination is src and dst both: "
"uint8_t, int8_t, uint16_t, int16_t, half.");});
}
template <typename T>
__aicore__ inline void LoadData2DL12L0ACal(__ca__ T* dst, __cbuf__ T* src, const LoadData2DParams& loadDataParam)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData from A1 to A2");
#else
CheckLoadData2DType<T>();
if constexpr (IsSameType<T, int4b_t>::value) {
load_cbuf_to_ca_s4((__ca__ void *)dst, (__cbuf__ void *)src, loadDataParam.startIndex,
loadDataParam.repeatTimes, loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, 0, inc);
} else {
using U = typename Conditional<SupportType<T, uint16_t, int16_t>(), half, T>::type;
if (loadDataParam.ifTranspose) {
load_cbuf_to_ca((__ca__ U*)dst, (__cbuf__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, 1, inc);
} else {
load_cbuf_to_ca((__ca__ U*)dst, (__cbuf__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, 0, inc);
}
}
#endif
}
template <typename T>
__aicore__ inline void LoadData2DL12L0BCal(__cb__ T* dst, __cbuf__ T* src, const LoadData2DParams& loadDataParam)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData from B1 to B2");
#else
CheckLoadData2DType<T>();
if constexpr (IsSameType<T, int4b_t>::value) {
load_cbuf_to_cb_s4((__cb__ void *)dst, (__cbuf__ void *)src, loadDataParam.startIndex,
loadDataParam.repeatTimes, loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, 0, inc);
} else {
using U = typename Conditional<SupportType<T, uint16_t, int16_t>(), half, T>::type;
if (loadDataParam.ifTranspose) {
load_cbuf_to_cb((__cb__ U*)dst, (__cbuf__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, 1, inc);
} else {
load_cbuf_to_cb((__cb__ U*)dst, (__cbuf__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, 0, inc);
}
}
#endif
}
template <typename T>
__aicore__ inline void LoadData2DGM2L0ACal(__ca__ T* dst, __gm__ T* src, const LoadData2DParams& loadDataParam)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData from GM to A2");
#else
CheckLoadData2DType<T>();
using U = typename Conditional<SupportType<T, uint16_t, int16_t>(), half, T>::type;
load_gm_to_ca((__ca__ U*)dst, (__gm__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, (addr_cal_mode_t)0);
#endif
}
template <typename T>
__aicore__ inline void LoadData2DGM2L0BCal(__cb__ T* dst, __gm__ T* src, const LoadData2DParams& loadDataParam)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData from GM to B2");
#else
CheckLoadData2DType<T>();
using U = typename Conditional<SupportType<T, uint16_t, int16_t>(), half, T>::type;
load_gm_to_cb((__cb__ U*)dst, (__gm__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, (addr_cal_mode_t)0);
#endif
}
template <typename T>
__aicore__ inline void LoadData2DGM2L1Cal(__cbuf__ T* dst, __gm__ T* src, const LoadData2DParams& loadDataParam)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData from GM to A1 / B1");
#else
CheckLoadData2DType<T>();
using U = typename Conditional<SupportType<T, uint16_t, int16_t>(), half, T>::type;
if (loadDataParam.addrMode == 0) {
load_gm_to_cbuf((__cbuf__ U*)dst, (__gm__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, inc);
} else {
load_gm_to_cbuf((__cbuf__ U*)dst, (__gm__ U*)src, loadDataParam.startIndex, loadDataParam.repeatTimes,
loadDataParam.srcStride, loadDataParam.dstGap, loadDataParam.sid, dec);
}
#endif
}
template <typename T>
__aicore__ inline void LoadData2DL12L0ACal(__ca__ T *dst, __cbuf__ T *src, const LoadData2DParamsV2 &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData2DParamsV2");
}
template <typename T>
__aicore__ inline void LoadData2DL12L0BCal(__cb__ T *dst, __cbuf__ T *src, const LoadData2DParamsV2 &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData2DParamsV2");
}
template <typename T>
__aicore__ inline void LoadData2DGM2L0ACal(__ca__ T *dst, __gm__ T *src, const LoadData2DParamsV2 &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData2DParamsV2");
}
template <typename T>
__aicore__ inline void LoadData2DGM2L0BCal(__cb__ T *dst, __gm__ T *src, const LoadData2DParamsV2 &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData2DParamsV2");
}
template <typename T>
__aicore__ inline void LoadData2DGM2L1Cal(__cbuf__ T *dst, __gm__ T *src, const LoadData2DParamsV2 &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData2DParamsV2");
}
template <typename T>
__aicore__ inline void LoadData2DL12L0ATransposeCal(__ca__ T *dst, __cbuf__ T *src,
const LoadData2dTransposeParams &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadDataWithTranspose");
}
template <typename T>
__aicore__ inline void LoadData2DL12L0BTransposeCal(__cb__ T *dst, __cbuf__ T *src,
const LoadData2dTransposeParams &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadDataWithTranspose");
}
template <typename T>
__aicore__ inline void LoadData2DL12L0BTransposeCal(__cb__ T *dst, __cbuf__ T *src,
const LoadData2dTransposeParamsV2 &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadDataWithTranspose");
}
* LoadData 3dv1 *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void LoadData3DV1L12L0ACal(__ca__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV1<T>& loadDataParams)
{
if (loadDataParams.cSize == 0) {
img2col_cbuf_to_ca(dst, src, loadDataParams.fetchFilterW, loadDataParams.fetchFilterH, loadDataParams.leftTopW,
loadDataParams.leftTopH, loadDataParams.c1Index, loadDataParams.strideW, loadDataParams.strideH,
loadDataParams.filterW, loadDataParams.filterH, loadDataParams.dilationFilterW,
loadDataParams.dilationFilterH, loadDataParams.jumpStride, loadDataParams.repeatMode,
loadDataParams.repeatTime, CSIZE0);
} else {
img2col_cbuf_to_ca(dst, src, loadDataParams.fetchFilterW, loadDataParams.fetchFilterH, loadDataParams.leftTopW,
loadDataParams.leftTopH, loadDataParams.c1Index, loadDataParams.strideW, loadDataParams.strideH,
loadDataParams.filterW, loadDataParams.filterH, loadDataParams.dilationFilterW,
loadDataParams.dilationFilterH, loadDataParams.jumpStride, loadDataParams.repeatMode,
loadDataParams.repeatTime, CSIZE1);
}
}
template <typename T>
__aicore__ inline void LoadData3DV1L12L0BCal(__cb__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV1<T>& loadDataParams)
{
if (loadDataParams.cSize == 0) {
img2col_cbuf_to_cb(dst, src, loadDataParams.fetchFilterW, loadDataParams.fetchFilterH, loadDataParams.leftTopW,
loadDataParams.leftTopH, loadDataParams.c1Index, loadDataParams.strideW, loadDataParams.strideH,
loadDataParams.filterW, loadDataParams.filterH, loadDataParams.dilationFilterW,
loadDataParams.dilationFilterH, loadDataParams.jumpStride, loadDataParams.repeatMode,
loadDataParams.repeatTime, CSIZE0);
} else {
img2col_cbuf_to_cb(dst, src, loadDataParams.fetchFilterW, loadDataParams.fetchFilterH, loadDataParams.leftTopW,
loadDataParams.leftTopH, loadDataParams.c1Index, loadDataParams.strideW, loadDataParams.strideH,
loadDataParams.filterW, loadDataParams.filterH, loadDataParams.dilationFilterW,
loadDataParams.dilationFilterH, loadDataParams.jumpStride, loadDataParams.repeatMode,
loadDataParams.repeatTime, CSIZE1);
}
}
template <typename T>
__aicore__ inline void LoadData3DV1L12UBCal(__ubuf__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV1<T>& loadDataParams)
{
if (loadDataParams.cSize == 0) {
img2col_cbuf_to_ub(dst, src, loadDataParams.fetchFilterW, loadDataParams.fetchFilterH, loadDataParams.leftTopW,
loadDataParams.leftTopH, loadDataParams.c1Index, loadDataParams.strideW, loadDataParams.strideH,
loadDataParams.filterW, loadDataParams.filterH, loadDataParams.dilationFilterW,
loadDataParams.dilationFilterH, loadDataParams.jumpStride, loadDataParams.repeatMode,
loadDataParams.repeatTime, CSIZE0);
} else {
img2col_cbuf_to_ub(dst, src, loadDataParams.fetchFilterW, loadDataParams.fetchFilterH, loadDataParams.leftTopW,
loadDataParams.leftTopH, loadDataParams.c1Index, loadDataParams.strideW, loadDataParams.strideH,
loadDataParams.filterW, loadDataParams.filterH, loadDataParams.dilationFilterW,
loadDataParams.dilationFilterH, loadDataParams.jumpStride, loadDataParams.repeatMode,
loadDataParams.repeatTime, CSIZE1);
}
}
* LoadData 3dv2 *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void LoadData3DV2L12L0ACal(__ca__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV2<T>& loadDataParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2 from A1 to A2");
#else
if constexpr (IsSameType<T, int4b_t>::value) {
img2colv2_cbuf_to_ca_s4((__ca__ void *)dst, (__cbuf__ void *)src, loadDataParams.kExtension, loadDataParams.mExtension, loadDataParams.kStartPt,
loadDataParams.mStartPt, loadDataParams.strideW, loadDataParams.strideH, loadDataParams.filterW,
loadDataParams.filterH, loadDataParams.dilationFilterW, loadDataParams.dilationFilterH,
loadDataParams.enTranspose, loadDataParams.channelSize);
} else {
img2colv2_cbuf_to_ca(dst, src, loadDataParams.kExtension, loadDataParams.mExtension, loadDataParams.kStartPt,
loadDataParams.mStartPt, loadDataParams.strideW, loadDataParams.strideH, loadDataParams.filterW,
loadDataParams.filterH, loadDataParams.dilationFilterW, loadDataParams.dilationFilterH,
loadDataParams.enTranspose, loadDataParams.channelSize);
}
#endif
}
template <typename T>
__aicore__ inline void LoadData3DV2L12L0BCal(__cb__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV2<T>& loadDataParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2 from B1 to B2");
#else
if constexpr (IsSameType<T, int4b_t>::value) {
img2colv2_cbuf_to_cb_s4((__cb__ void *)dst, (__cbuf__ void *)src, loadDataParams.kExtension, loadDataParams.mExtension, loadDataParams.kStartPt,
loadDataParams.mStartPt, loadDataParams.strideW, loadDataParams.strideH, loadDataParams.filterW,
loadDataParams.filterH, loadDataParams.dilationFilterW, loadDataParams.dilationFilterH,
loadDataParams.enTranspose, loadDataParams.channelSize);
} else {
img2colv2_cbuf_to_cb(dst, src, loadDataParams.kExtension, loadDataParams.mExtension, loadDataParams.kStartPt,
loadDataParams.mStartPt, loadDataParams.strideW, loadDataParams.strideH, loadDataParams.filterW,
loadDataParams.filterH, loadDataParams.dilationFilterW, loadDataParams.dilationFilterH,
loadDataParams.enTranspose, loadDataParams.channelSize);
}
#endif
}
template <typename T>
__aicore__ inline void LoadData3DV2L12UBCal(__ubuf__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV2<T>& loadDataParams)
{
if constexpr (!IsSameType<T, int4b_t>::value) {
img2colv2_cbuf_to_ub(dst, src, loadDataParams.kExtension, loadDataParams.mExtension, loadDataParams.kStartPt,
loadDataParams.mStartPt, loadDataParams.strideW, loadDataParams.strideH, loadDataParams.filterW,
loadDataParams.filterH, loadDataParams.dilationFilterW, loadDataParams.dilationFilterH,
loadDataParams.enTranspose, loadDataParams.channelSize);
}
}
* LoadData 3dv2Pro *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void LoadData3DV2L12L0ACal(__ca__ T *dst, __cbuf__ T *src,
const LoadData3DParamsV2Pro& loadDataParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2Pro from A1 to A2");
#else
if constexpr (IsSameType<T, int4b_t>::value) {
img2colv2_cbuf_to_ca_s4((__ca__ void *)dst, (__cbuf__ void *)src, loadDataParams.extConfig, loadDataParams.extConfig >> LOAD_M_EXTENSION,
loadDataParams.extConfig >> LOAD_K_START_POSITION, loadDataParams.extConfig >> LOAD_M_START_POSITION,
loadDataParams.filterConfig, loadDataParams.filterConfig >> LOAD_STRIDE_H,
loadDataParams.filterConfig >> LOAD_FILTER_W, loadDataParams.filterConfig >> LOAD_FILTER_H,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_W,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_H, loadDataParams.enTranspose,
loadDataParams.channelSize);
} else {
img2colv2_cbuf_to_ca(dst, src, loadDataParams.extConfig, loadDataParams.extConfig >> LOAD_M_EXTENSION,
loadDataParams.extConfig >> LOAD_K_START_POSITION, loadDataParams.extConfig >> LOAD_M_START_POSITION,
loadDataParams.filterConfig, loadDataParams.filterConfig >> LOAD_STRIDE_H,
loadDataParams.filterConfig >> LOAD_FILTER_W, loadDataParams.filterConfig >> LOAD_FILTER_H,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_W,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_H, loadDataParams.enTranspose,
loadDataParams.channelSize);
}
#endif
}
template <typename T>
__aicore__ inline void LoadData3DV2L12L0BCal(__cb__ T *dst, __cbuf__ T *src,
const LoadData3DParamsV2Pro& loadDataParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2Pro from B1 to B2");
#else
if constexpr (IsSameType<T, int4b_t>::value) {
img2colv2_cbuf_to_cb_s4((__cb__ void *)dst, (__cbuf__ void *)src, loadDataParams.extConfig, loadDataParams.extConfig >> LOAD_M_EXTENSION,
loadDataParams.extConfig >> LOAD_K_START_POSITION, loadDataParams.extConfig >> LOAD_M_START_POSITION,
loadDataParams.filterConfig, loadDataParams.filterConfig >> LOAD_STRIDE_H,
loadDataParams.filterConfig >> LOAD_FILTER_W, loadDataParams.filterConfig >> LOAD_FILTER_H,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_W,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_H, loadDataParams.enTranspose,
loadDataParams.channelSize);
} else {
img2colv2_cbuf_to_cb(dst, src, loadDataParams.extConfig, loadDataParams.extConfig >> LOAD_M_EXTENSION,
loadDataParams.extConfig >> LOAD_K_START_POSITION, loadDataParams.extConfig >> LOAD_M_START_POSITION,
loadDataParams.filterConfig, loadDataParams.filterConfig >> LOAD_STRIDE_H,
loadDataParams.filterConfig >> LOAD_FILTER_W, loadDataParams.filterConfig >> LOAD_FILTER_H,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_W,
loadDataParams.filterConfig >> LOAD_DILATION_FILTER_H, loadDataParams.enTranspose,
loadDataParams.channelSize);
}
#endif
}
template <typename T>
__aicore__ inline void LoadData3DV2L12UBCal(__ubuf__ T* dst, __cbuf__ T* src,
const LoadData3DParamsV2Pro& loadDataParams)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2Pro from L1 to UB");
}
template <>
__aicore__ inline void LoadData3DV2L12L0BCal(__cb__ int8_t* dst, __cbuf__ int8_t* src,
const LoadData3DParamsV2Pro& loadDataParams)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2Pro from B1 to B2 with type int8_t");
}
template <>
__aicore__ inline void LoadData3DV2L12L0BCal(__cb__ uint8_t* dst, __cbuf__ uint8_t* src,
const LoadData3DParamsV2Pro& loadDataParams)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadData with LoadData3DParamsV2Pro from B1 to B2 with type uint8_t");
}
* Mmad *
* ************************************************************************************************* */
template <typename T, typename U, typename S>
__aicore__ inline void MmadCal(__cc__ T* c, __ca__ U* a, __cb__ S* b, const MmadParams& mmadParams)
{
ASCENDC_ASSERT(mmadParams.cmatrixSource == 0,
{ KERNEL_LOG(KERNEL_ERROR, "the C matrix source doesn't support BT buffer on current device"); });
bool cmatrixInitVal = mmadParams.cmatrixInitVal && (!mmadParams.isBias);
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "Mmad");
#else
ASCENDC_ASSERT((SupportType<Tuple<T, U, S>,
Tuple<int32_t, int8_t, int8_t>, Tuple<int32_t, uint8_t, int8_t>, Tuple<int32_t, uint8_t, uint8_t>,
Tuple<half, half, half>, Tuple<float, half, half>, Tuple<int32_t, int4b_t, int4b_t>>()), {KERNEL_LOG(KERNEL_ERROR,
"Failed to check dtype in Mmad, current api support dtype combination is Dst: int32_t, src0: int8_t, "
"src1: int8_t; Dst: int32_t, src0: uint8_t, src1: int8_t; Dst: int32_t, src0: uint8_t, src1: uint8_t; "
"Dst: half, src0: half, src1: half; Dst: float, src0: half, src1: half"
"Dst: int32_t, src0: int4b_t, src1: int4b_t");});
if constexpr ((IsSameType<U, int4b_t>::value) && (IsSameType<S, int4b_t>::value)) {
ASCENDC_ASSERT(mmadParams.k % 2 == 0,
{KERNEL_LOG(KERNEL_ERROR, "When src0 and src1 are int4b_t, k must be even number.");});
mad_s4(c, a, b, mmadParams.m,mmadParams.k, mmadParams.n, cmatrixInitVal);
} else {
mad(c, a, b, mmadParams.m, mmadParams.k, mmadParams.n, cmatrixInitVal);
}
#endif
}
template <typename T, typename U, typename S>
__aicore__ inline void MmadCal(__cc__ T* c, __ca__ U* a, __cb__ S* b, uint64_t bias,
const MmadParams& mmadParams, bool cmatrixSource)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "Mmad with biasLocal");
}
__aicore__ inline void MmadSpCal(__cc__ int32_t *c, __ca__ int8_t *a, __cb__ int8_t *b, const MmadParams &mmadParams)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "MmadWithSparse");
}
template <typename T = int8_t, typename U = uint8_t,
typename std::enable_if<IsSameType<PrimT<T>, int8_t>::value, bool>::type = true,
typename std::enable_if<IsSameType<PrimT<U>, uint8_t>::value, bool>::type = true>
__aicore__ inline void LoadDataWithSparseCal(const LocalTensor<T> &dst, const LocalTensor<T> &src,
const LocalTensor<U> &idx, const LoadData2dParams &loadDataParam)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadDataWithSparse");
}
template <typename T = int8_t, typename std::enable_if<IsSameType<PrimT<T>, int8_t>::value, bool>::type = true>
__aicore__ inline void LoadUnzipIndexCal(const GlobalTensor<T> &src, uint32_t numOfIndexTabEntry)
{
#if ASCENDC_CPU_DEBUG
const __gm__ int64_t *srcAddr = reinterpret_cast<const __gm__ int64_t *>(src.GetPhyAddr());
constexpr uint32_t unitOfLength = 128;
for (int i = 0; i < numOfIndexTabEntry; i++) {
uint32_t compressedLength = *(srcAddr + i) & 0x1fff;
SetUnzipCompressedLen(compressedLength * unitOfLength);
}
constexpr uint32_t maxSize = 32 * 1024;
ASCENDC_ASSERT((src.GetSize() % VALUE_512 == 0 && src.GetSize() <= maxSize), {KERNEL_LOG(KERNEL_ERROR,
"Failed to check src size in LoadUnzipIndex, it must be less than 32KB and divisible by 512, current "
"value is %u.", src.GetSize());});
ASCENDC_CHECK_VALUE_RANGE(numOfIndexTabEntry, 1, UINT32_MAX, "numOfIndexTabEntry", "LoadUnzipIndex");
#endif
load_unzip_index_from_gm((__gm__ void *)src.GetPhyAddr(), numOfIndexTabEntry, 0);
}
* BroadCastVecToMM *
* ************************************************************************************************* */
template <typename T, typename U>
__aicore__ inline void BroadCastVecToMMCal(__cc__ T* dst, __ubuf__ U* src, const int32_t blockCount,
const uint8_t blockLen, const uint8_t srcGap, const uint8_t dstGap)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "BroadCastVecToMM");
#else
ASCENDC_CHECK_VALUE_RANGE(blockCount, 1, 255, "blockCount", "BroadCastVecToMM");
ASCENDC_CHECK_VALUE_RANGE(blockLen, 1, 255, "blockLen", "BroadCastVecToMM");
ASCENDC_ASSERT((SupportType<Tuple<T, U>, Tuple<int32_t, int32_t>, Tuple<float, float>, Tuple<half, half>>()),
{KERNEL_LOG(KERNEL_ERROR, "Failed to check dtype in BroadCastVecToMM, current api support dtype combination is "
"src and dst both: half / float / int32_t.");});
broadcast_ub_to_cc(dst, src, blockCount, blockLen, srcGap, dstGap);
#endif
}
* LoadData 3dv1 *
* ************************************************************************************************* */
__aicore__ inline void Load3DSetFMatrixCal(uint16_t l1H, uint16_t l1W, const uint8_t padList[4])
{
uint64_t regFMatrix = 0;
regFMatrix |= uint64_t(l1W & 0xFFFF);
uint32_t l1HShiftBit = 16;
regFMatrix |= uint64_t(l1H & 0xFFFF) << l1HShiftBit;
uint32_t padNumber = 4;
uint32_t padListShiftBit = 8;
uint32_t padListShiftBase = 32;
for (uint32_t i = 0; i < padNumber; i++) {
regFMatrix |= uint64_t(padList[i] & 0xFF) << (padListShiftBase + i * padListShiftBit);
}
set_fmatrix(regFMatrix);
}
__aicore__ inline void Load3DSetFMatrixBCal(uint16_t l1H, uint16_t l1W, const uint8_t padList[4])
{
ASCENDC_REPORT_NOT_SUPPORT(false, "SetFMatrix with fmatrixMode = FMATRIX_RIGHT");
}
template <typename T>
__aicore__ inline void Load3DSetPaddingCal(const T padValue)
{
uint16_t paddingValue = 0;
uint16_t padValueShiftBit = 8;
if constexpr (sizeof(T) == B16_BYTE_SIZE) {
paddingValue = static_cast<uint16_t>(GetScalarBitcodeValue((T)padValue));
} else if constexpr (sizeof(T) == B32_BYTE_SIZE) {
ASCENDC_ASSERT(false, {KERNEL_LOG(KERNEL_ERROR, "Failed to check dtype in SetLoadDataPaddingValue, current "
"api support dtype combination is: int8_t / uint8_t / half / int16_t / uint16_t.");});
} else {
paddingValue = ((static_cast<uint16_t>(padValue)) << padValueShiftBit) | static_cast<uint16_t>(padValue);
}
set_padding(paddingValue);
}
* InitL1Buffer *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void InitL1BufferCal(__cbuf__ T *dst, const InitConstValueParams<T> &initConstValueParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "InitConstValue in A1 / B1");
#else
ASCENDC_CHECK_VALUE_RANGE(initConstValueParams.repeatTimes, 0, UINT8_MAX, "repeatTimes", "InitConstValue");
int64_t repeatBit = initConstValueParams.repeatTimes;
if constexpr (IsSameType<T, half>::value) {
create_cbuf_matrix((__cbuf__ void *)dst, repeatBit, (half)initConstValueParams.initValue);
} else if constexpr (IsSameType<T, int16_t>::value || IsSameType<T, uint16_t>::value) {
create_cbuf_matrix((__cbuf__ void *)dst, repeatBit,
GetScalarBitcodeToHalf(initConstValueParams.initValue));
} else {
ASCENDC_ASSERT(false, {KERNEL_LOG(KERNEL_ERROR, "Failed to check dtype in InitConstValue, current api "
"support dtype combination is dst: half, int16_t, uint16_t");});
}
#endif
}
* InitL0ANzMatrix *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void InitL0ANzMatrixCal(__ca__ T *dst, const InitConstValueParams<T> &initConstValueParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "InitConstValue in A2");
#else
ASCENDC_CHECK_VALUE_RANGE(initConstValueParams.repeatTimes, 0, UINT8_MAX, "repeatTimes", "InitConstValue");
int64_t repeatBit = initConstValueParams.repeatTimes;
if constexpr (IsSameType<T, half>::value) {
create_ca_matrix((__ca__ void *)dst, repeatBit, (half)initConstValueParams.initValue);
} else if constexpr (IsSameType<T, int16_t>::value || IsSameType<T, uint16_t>::value) {
create_ca_matrix((__ca__ void *)dst, repeatBit, GetScalarBitcodeToHalf(initConstValueParams.initValue));
} else {
ASCENDC_ASSERT(false, {KERNEL_LOG(KERNEL_ERROR, "Failed to check dtype in InitConstValue, current api "
"support dtype combination is dst: half, int16_t, uint16_t");});
}
#endif
}
* InitL0BNzMatrix *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void InitL0BNzMatrixCal(__cb__ T *dst, const InitConstValueParams<T> &initConstValueParams)
{
#if defined(__DAV_M200_VEC__)
ASCENDC_REPORT_NOT_SUPPORT(false, "InitConstValue in B2");
#else
ASCENDC_CHECK_VALUE_RANGE(initConstValueParams.repeatTimes, 0, UINT8_MAX, "repeatTimes", "InitConstValue");
int64_t repeatBit = initConstValueParams.repeatTimes;
if constexpr (IsSameType<T, half>::value) {
create_cb_matrix((__cb__ void *)dst, repeatBit, (half)initConstValueParams.initValue);
} else if constexpr (IsSameType<T, int16_t>::value || IsSameType<T, uint16_t>::value) {
create_cb_matrix((__cb__ void *)dst, repeatBit, GetScalarBitcodeToHalf(initConstValueParams.initValue));
} else {
ASCENDC_ASSERT(false, {KERNEL_LOG(KERNEL_ERROR, "Failed to check dtype in InitConstValue, current api "
"support dtype combination is dst: half, int16_t, uint16_t");});
}
#endif
}
* SetLoadDataRepeat *
* ************************************************************************************************* */
__aicore__ inline void SetLoadDataRepeatCal(const LoadDataRepeatParam& repeatParams)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "SetLoadDataRepeat");
}
* SetLoadDataBoundary *
* ************************************************************************************************* */
__aicore__ inline void SetLoadDataBoundaryCal(uint32_t boundaryValue)
{
set_l1_3d_size(static_cast<uint64_t>(boundaryValue));
}
* LoadImageToLocalCal *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void LoadImageToLocalCal(__cbuf__ T *dst, const LoadImageToLocalParams &loadDataParams)
{
ASCENDC_ASSERT((SupportType<T, uint8_t, int8_t, half>()), {KERNEL_LOG(KERNEL_ERROR, "Failed to check dtype in "
"LoadImageToLocal, current api support dtype combination is dst: uint8_t / int8_t / half.");});
load_image_to_cbuf(dst, static_cast<uint16_t>(loadDataParams.horizSize - 1),
static_cast<uint16_t>(loadDataParams.vertSize - 1), loadDataParams.horizStartPos, loadDataParams.vertStartPos,
static_cast<uint16_t>(loadDataParams.srcHorizSize - 1), loadDataParams.topPadSize, loadDataParams.botPadSize,
loadDataParams.leftPadSize, loadDataParams.rightPadSize, loadDataParams.sid);
}
* LoadDataUnzip *
* ************************************************************************************************* */
template <typename T>
__aicore__ inline void LoadDataUnzipToL1Cal(__cbuf__ T *dst, __gm__ T *src)
{
load_gm_to_cbuf_unzip((__cbuf__ void *)dst, (__gm__ void *)src);
}
template <typename T>
__aicore__ inline void LoadDataUnzipToL0BCal(__cb__ T *dst, __gm__ T *src)
{
load_gm_to_cb_unzip((__cb__ void *)dst, (__gm__ void *)src);
}
template <typename T>
__aicore__ inline void LoadDataUnzipToL0ACal(__ca__ T *dst, __gm__ T *src)
{
ASCENDC_REPORT_NOT_SUPPORT(false, "LoadDataUnzip to A2");
}
}
#endif
#if defined(__UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_OPERATOR_MM_IMPL_H__)
#undef __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#undef __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_OPERATOR_MM_IMPL_H__
#endif
