Star550
758
代码
代码
Issues84
Pull Requests156
流水线
讨论
Wiki
分析
Star550
758
cann-robotcann-robot新增A5 CAPI接口
fc352ac6创建于 1月30日历史提交
/**
* 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.
*/

#ifndef INCLUDE_C_API_REG_COMPUTE_REG_LOAD_H
#define INCLUDE_C_API_REG_COMPUTE_REG_LOAD_H

#include "impl/c_api/instr_impl/npu_arch_3510/vector_datamove_impl.h"

// ==========源操作数为起初地址，目的数为寄存器=========
// ==========asc_gather(u8/s8/half/u16/s16/float/u32/s32/bf16/e4m3/e5m2)=========
__simd_callee__ inline void asc_gather(vector_int16_t& dst, __ubuf__  int8_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_uint16_t& dst, __ubuf__  uint8_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_int16_t& dst, __ubuf__  int16_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_uint16_t& dst, __ubuf__  uint16_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_int32_t& dst, __ubuf__  int32_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_uint32_t& dst, __ubuf__  uint32_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_half& dst, __ubuf__  half* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_float& dst, __ubuf__  float* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_bfloat16_t& dst, __ubuf__  bfloat16_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_fp8_e4m3fn_t& dst, __ubuf__  fp8_e4m3fn_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_fp8_e5m2_t& dst, __ubuf__  fp8_e5m2_t* src,
    vector_uint16_t index, vector_bool mask);

__simd_callee__ inline void asc_gather(vector_fp8_e8m0_t& dst, __ubuf__  fp8_e8m0_t* src,
    vector_uint16_t index, vector_bool mask);

// ==========源操作数和目的数都为寄存器=========
// ==========asc_gather(u8/s8/half/u16/s16/u32/s32/bf16/e4m3/e5m2)=========
__simd_callee__ inline void asc_gather(vector_int8_t& dst, vector_int8_t src,
    vector_uint8_t index);

__simd_callee__ inline void asc_gather(vector_uint8_t& dst, vector_uint8_t src,
    vector_uint8_t index);

__simd_callee__ inline void asc_gather(vector_int16_t& dst, vector_int16_t src,
    vector_uint16_t index);

__simd_callee__ inline void asc_gather(vector_uint16_t& dst, vector_uint16_t src,
    vector_uint16_t index);

__simd_callee__ inline void asc_gather(vector_int32_t& dst, vector_int32_t src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather(vector_uint32_t& dst, vector_uint32_t src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather(vector_half& dst, vector_half src,
    vector_uint16_t index);

__simd_callee__ inline void asc_gather(vector_bfloat16_t& dst, vector_bfloat16_t src,
    vector_uint16_t index);

__simd_callee__ inline void asc_gather(vector_fp8_e4m3fn_t& dst, vector_fp8_e4m3fn_t src,
    vector_uint8_t index);

__simd_callee__ inline void asc_gather(vector_fp8_e5m2_t& dst, vector_fp8_e5m2_t src,
    vector_uint8_t index);

__simd_callee__ inline void asc_gather(vector_fp8_e8m0_t& dst, vector_fp8_e8m0_t src,
    vector_uint8_t index);

// ==========有mask参数=========
// ==========asc_gather_datablock(u8/s8/half/u16/s16/float/u32/s32/u64/s64/bf16/e4m3/e5m2/e2m1/e1m2)=========
__simd_callee__ inline void asc_gather_datablock(vector_int8_t& dst, __ubuf__  int8_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_uint8_t& dst, __ubuf__  uint8_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_int16_t& dst, __ubuf__  int16_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_uint16_t& dst, __ubuf__  uint16_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_int32_t& dst, __ubuf__  int32_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_uint32_t& dst, __ubuf__  uint32_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_int64_t& dst, __ubuf__  int64_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_uint64_t& dst, __ubuf__  uint64_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_half& dst, __ubuf__  half* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_float& dst, __ubuf__  float* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_bfloat16_t& dst, __ubuf__  bfloat16_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_fp8_e4m3fn_t& dst, __ubuf__  fp8_e4m3fn_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_fp8_e5m2_t& dst, __ubuf__  fp8_e5m2_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_fp4x2_e2m1_t& dst, __ubuf__  fp4x2_e2m1_t* src,
    vector_uint32_t index, vector_bool mask);

__simd_callee__ inline void asc_gather_datablock(vector_fp4x2_e1m2_t& dst, __ubuf__  fp4x2_e1m2_t* src,
    vector_uint32_t index, vector_bool mask);

// ==========无mask参数=========
// ==========asc_gather_datablock(u8/s8/half/u16/s16/float/u32/s32/s64/bf16/e4m3/e5m2/e2m1/e1m2)=========
__simd_callee__ inline void asc_gather_datablock(vector_int8_t& dst, __ubuf__  int8_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_uint8_t& dst, __ubuf__  uint8_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_int16_t& dst, __ubuf__  int16_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_uint16_t& dst, __ubuf__  uint16_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_int32_t& dst, __ubuf__  int32_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_uint32_t& dst, __ubuf__  uint32_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_int64_t& dst, __ubuf__  int64_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_half& dst, __ubuf__  half* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_float& dst, __ubuf__  float* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_bfloat16_t& dst, __ubuf__  bfloat16_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_fp8_e4m3fn_t& dst, __ubuf__  fp8_e4m3fn_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_fp8_e5m2_t& dst, __ubuf__  fp8_e5m2_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_fp4x2_e2m1_t& dst, __ubuf__  fp4x2_e2m1_t* src,
    vector_uint32_t index);

__simd_callee__ inline void asc_gather_datablock(vector_fp4x2_e1m2_t& dst, __ubuf__  fp4x2_e1m2_t* src,
    vector_uint32_t index);

// ==========asc_get_mask_spr==========
__simd_callee__ inline vector_bool asc_get_mask_spr_b16();

__simd_callee__ inline vector_bool asc_get_mask_spr_b32();

// ==========asc_loadalign(int8_t/uint8_t/fp4x2_e2m1_t/fp4x2_e1m2_t/fp8_e8m0_t/fp8_e5m2_t/fp8_e4m3fn_t/int16_t/uint16_t/half/bfloat16_t/int32_t/uint32_t/float/int64_t/uint64_t)==========
__simd_callee__ inline void asc_loadalign(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign(vector_int32_t& dst, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadalign(vector_uint32_t& dst, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadalign(vector_float& dst, __ubuf__ float* src);

__simd_callee__ inline void asc_loadalign(vector_int64_t& dst, __ubuf__ int64_t* src);

__simd_callee__ inline void asc_loadalign(vector_uint64_t& dst, __ubuf__ uint64_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_brc(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_int32_t& dst, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_uint32_t& dst, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadalign_brc(vector_float& dst, __ubuf__ float* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_upsample(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_downsample(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_int32_t& dst, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_uint32_t& dst, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadalign_unpack(vector_float& dst, __ubuf__ float* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_int8_t& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_uint8_t& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_int32_t& dst, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_uint32_t& dst, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_float& dst, __ubuf__ float* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_int16_t& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_uint16_t& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_half& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_int32_t& dst, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_uint32_t& dst, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_float& dst, __ubuf__ float* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_int8_t& dst0, vector_int8_t& dst1, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_uint8_t& dst0, vector_uint8_t& dst1, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp4x2_e2m1_t& dst0, vector_fp4x2_e2m1_t& dst1, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp4x2_e1m2_t& dst0, vector_fp4x2_e1m2_t& dst1, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp8_e8m0_t& dst0, vector_fp8_e8m0_t& dst1, __ubuf__ fp8_e8m0_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp8_e5m2_t& dst0, vector_fp8_e5m2_t& dst1, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp8_e4m3fn_t& dst0, vector_fp8_e4m3fn_t& dst1, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_int16_t& dst0, vector_int16_t& dst1, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_uint16_t& dst0, vector_uint16_t& dst1, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_half& dst0, vector_half& dst1, __ubuf__ half* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_bfloat16_t& dst0, vector_bfloat16_t& dst1, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_int32_t& dst0, vector_int32_t& dst1, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_uint32_t& dst0, vector_uint32_t& dst1, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadalign_deintlv(vector_float& dst0, vector_float& dst1, __ubuf__ float* src);

__simd_callee__ inline void asc_loadalign(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_int64_t& dst, __ubuf__ int64_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign(vector_uint64_t& dst, __ubuf__ uint64_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_v2(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v2(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_v3(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_int8_t& dst0, vector_int8_t& dst1, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_uint8_t& dst0, vector_uint8_t& dst1, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp4x2_e2m1_t& dst0, vector_fp4x2_e2m1_t& dst1, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp4x2_e1m2_t& dst0, vector_fp4x2_e1m2_t& dst1, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp8_e8m0_t& dst0, vector_fp8_e8m0_t& dst1, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp8_e5m2_t& dst0, vector_fp8_e5m2_t& dst1, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_fp8_e4m3fn_t& dst0, vector_fp8_e4m3fn_t& dst1, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_int16_t& dst0, vector_int16_t& dst1, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_uint16_t& dst0, vector_uint16_t& dst1, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_half& dst0, vector_half& dst1, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_bfloat16_t& dst0, vector_bfloat16_t& dst1, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_int32_t& dst0, vector_int32_t& dst1, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_uint32_t& dst0, vector_uint32_t& dst1, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv(vector_float& dst0, vector_float& dst1, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int64_t& dst, __ubuf__ int64_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint64_t& dst, __ubuf__ uint64_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_upsample_postupdate(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_downsample_postupdate(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_unpack_postupdate_v2(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_int8_t& dst, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_uint8_t& dst, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_fp8_e8m0_t& dst, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v2(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_int16_t& dst, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_uint16_t& dst, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_half& dst, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_int32_t& dst, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_uint32_t& dst, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_brc_postupdate_v3(vector_float& dst, __ubuf__ float* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_int8_t& dst0, vector_int8_t& dst1, __ubuf__ int8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_uint8_t& dst0, vector_uint8_t& dst1, __ubuf__ uint8_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_fp4x2_e2m1_t& dst0, vector_fp4x2_e2m1_t& dst1, __ubuf__ fp4x2_e2m1_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_fp4x2_e1m2_t& dst0, vector_fp4x2_e1m2_t& dst1, __ubuf__ fp4x2_e1m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_fp8_e8m0_t& dst0, vector_fp8_e8m0_t& dst1, __ubuf__ fp8_e8m0_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_fp8_e5m2_t& dst0, vector_fp8_e5m2_t& dst1, __ubuf__ fp8_e5m2_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_fp8_e4m3fn_t& dst0, vector_fp8_e4m3fn_t& dst1, __ubuf__ fp8_e4m3fn_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_int16_t& dst0, vector_int16_t& dst1, __ubuf__ int16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_uint16_t& dst0, vector_uint16_t& dst1, __ubuf__ uint16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_half& dst0, vector_half& dst1, __ubuf__ half* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_bfloat16_t& dst0, vector_bfloat16_t& dst1, __ubuf__ bfloat16_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_int32_t& dst0, vector_int32_t& dst1, __ubuf__ int32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_uint32_t& dst0, vector_uint32_t& dst1, __ubuf__ uint32_t* src, int32_t offset);

__simd_callee__ inline void asc_loadalign_deintlv_postupdate(vector_float& dst0, vector_float& dst1, __ubuf__ float* src, int32_t offset);

// repeat stride模式读入
__simd_callee__ inline void asc_loadalign(vector_int8_t& dst, __ubuf__ int8_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_uint8_t& dst, __ubuf__ uint8_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_int16_t& dst, __ubuf__ int16_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_uint16_t& dst, __ubuf__ uint16_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_int32_t& dst, __ubuf__ int32_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_uint32_t& dst, __ubuf__ uint32_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_int64_t& dst, __ubuf__ int64_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_bfloat16_t& dst, __ubuf__ bfloat16_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_half& dst, __ubuf__ half* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign(vector_float& dst, __ubuf__ float* src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int8_t& dst, __ubuf__ int8_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint8_t& dst, __ubuf__ uint8_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int16_t& dst, __ubuf__ int16_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint16_t& dst, __ubuf__ uint16_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int32_t& dst, __ubuf__ int32_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_uint32_t& dst, __ubuf__ uint32_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_int64_t& dst, __ubuf__ int64_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_bfloat16_t& dst, __ubuf__ bfloat16_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp8_e4m3fn_t& dst, __ubuf__ fp8_e4m3fn_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp8_e5m2_t& dst, __ubuf__ fp8_e5m2_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp4x2_e1m2_t& dst, __ubuf__ fp4x2_e1m2_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_fp4x2_e2m1_t& dst, __ubuf__ fp4x2_e2m1_t*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_half& dst, __ubuf__ half*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

__simd_callee__ inline void asc_loadalign_postupdate(vector_float& dst, __ubuf__ float*& src_align32b, uint16_t block_stride, uint16_t repeat_stride, vector_bool mask);

// ==========asc_loadunalign_pre(int8_t/uint8_t/fp4x2_e2m1_t/fp4x2_e1m2_t/fp8_e5m2_t/fp8_e4m3fn_t/int16_t/uint16_t/half/bfloat16_t/int32_t/uint32_t/float/int64_t)==========
__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int8_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ uint8_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp4x2_e2m1_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp4x2_e1m2_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp8_e5m2_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp8_e4m3fn_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int16_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ uint16_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ half* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ bfloat16_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int32_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ uint32_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ float* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int64_t* src);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int8_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ uint8_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp4x2_e2m1_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp4x2_e1m2_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp8_e5m2_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ fp8_e4m3fn_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int16_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ uint16_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ half* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ bfloat16_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int32_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ uint32_t* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ float* src, iter_reg offset);

__simd_callee__ inline void asc_loadunalign_pre(vector_load_align& dst, __ubuf__ int64_t* src, iter_reg offset);

#endif