* Copyright (c) 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 kernel_operator_block_sync_intf_impl.h
* \brief
*/
#if !defined(__ASCENDC_INCLUDE_INTERNAL_HEADERS__)
#pragma message( \
"impl/basic_api/kernel_operator_block_sync_intf_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_operator_block_sync_intf.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_BLOCK_SYNC_INTF_IMPL_H__
#endif
#ifndef ASCENDC_MODULE_OPERATOR_BLOCK_SYNC_INTERFACE_IMPL_H
#define ASCENDC_MODULE_OPERATOR_BLOCK_SYNC_INTERFACE_IMPL_H
#include "kernel_event.h"
#include "../../include/basic_api/kernel_tensor.h"
#include "../../include/basic_api/kernel_tpipe.h"
#if !defined(__ASCENDC_INCLUDE_INTERNAL_HEADERS__)
#define __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#define __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_TPIPE_H__
#endif
#include "../../include/basic_api/kernel_struct_mm.h"
#include "../../include/basic_api/kernel_operator_data_copy_intf.h"
#if !defined(__ASCENDC_INCLUDE_INTERNAL_HEADERS__)
#define __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#define __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_OPERATOR_DATA_COPY_INTF_H__
#endif
#include "../../include/basic_api/kernel_operator_sys_var_intf.h"
#if __NPU_ARCH__ == 1001
#include "dav_c100/kernel_operator_sync_impl.h"
#elif __NPU_ARCH__ == 2002
#include "dav_m200/kernel_operator_sync_impl.h"
#elif __NPU_ARCH__ == 2201
#include "dav_c220/kernel_operator_sync_impl.h"
#elif __NPU_ARCH__ == 3002
#include "dav_m300/kernel_operator_common_impl.h"
#elif __NPU_ARCH__ == 3003
#include "dav_l300/kernel_operator_sync_impl.h"
#elif __NPU_ARCH__ == 3102
#include "dav_m310/kernel_operator_common_impl.h"
#elif __NPU_ARCH__ == 3510
#include "dav_3510/kernel_operator_sync_impl.h"
#elif (__NPU_ARCH__ == 5102)
#include "dav_m510/kernel_operator_sync_impl.h"
#elif (__NPU_ARCH__ == 3113)
#include "dav_l311/kernel_operator_sync_impl.h"
#elif (__NPU_ARCH__ == 5101) || (__NPU_ARCH__ == 5161) || (__NPU_ARCH__ == 5165) || (__NPU_ARCH__ == 5163)
#include "dav_5161/kernel_operator_sync_impl.h"
#endif
namespace AscendC {
template <HardEvent event>
__aicore__ inline void SetFlag(int32_t eventID)
{
if ASCEND_IS_AIC {
if constexpr (
event == HardEvent::MTE2_V || event == HardEvent::V_MTE2 || event == HardEvent::MTE3_V ||
event == HardEvent::V_MTE3 || event == HardEvent::V_V || event == HardEvent::S_V ||
#if defined(__NPU_ARCH__) && (__NPU_ARCH__ == 3510)
event == HardEvent::V_S || event == HardEvent::MTE2_MTE3 || event == HardEvent::MTE3_MTE2 ||
event == HardEvent::MTE3_S || event == HardEvent::S_MTE3) {
#else
event == HardEvent::V_S) {
#endif
return;
}
}
if ASCEND_IS_AIV {
if constexpr (
(event == HardEvent::MTE2_MTE1) || (event == HardEvent::MTE1_MTE2) || (event == HardEvent::MTE1_M) ||
(event == HardEvent::M_MTE1) || (event == HardEvent::M_FIX) || (event == HardEvent::FIX_M)) {
return;
}
}
SetFlagImpl<event>(eventID);
}
template <HardEvent event>
__aicore__ inline void WaitFlag(int32_t eventID)
{
if ASCEND_IS_AIC {
if constexpr (
event == HardEvent::MTE2_V || event == HardEvent::V_MTE2 || event == HardEvent::MTE3_V ||
event == HardEvent::V_MTE3 || event == HardEvent::V_V || event == HardEvent::S_V ||
#if defined(__NPU_ARCH__) && (__NPU_ARCH__ == 3510)
event == HardEvent::V_S || event == HardEvent::MTE2_MTE3 || event == HardEvent::MTE3_MTE2 ||
event == HardEvent::MTE3_S || event == HardEvent::S_MTE3) {
#else
event == HardEvent::V_S) {
#endif
return;
}
}
if ASCEND_IS_AIV {
if constexpr (
(event == HardEvent::MTE2_MTE1) || (event == HardEvent::MTE1_MTE2) || (event == HardEvent::MTE1_M) ||
(event == HardEvent::M_MTE1) || (event == HardEvent::M_FIX) || (event == HardEvent::FIX_M)) {
return;
}
}
WaitFlagImpl(event, eventID);
}
template <pipe_t pipe>
__aicore__ inline void PipeBarrier()
{
PipeBarrierImpl<pipe>();
}
#if defined(__NPU_ARCH__) && \
((__NPU_ARCH__ == 2201) || (__NPU_ARCH__ == 3002) || (__NPU_ARCH__ == 3102) || (__NPU_ARCH__ == 3510) || \
(__NPU_ARCH__ == 5102) || (__NPU_ARCH__ == 3003) || (__NPU_ARCH__ == 3103) || (__NPU_ARCH__ == 3113))
template <MemDsbT arg0>
__aicore__ inline void DataSyncBarrier()
{
DataSyncBarrierImpl<arg0>();
}
#endif
* @ingroup：IBSet, IBWait
* @brief：Set the flag bit of a core
* @param [in] gmWorkspace GlobalTensor to store core state
* @param [in] ubWorkspce LocalTensor for current core
* @param [in] blockIdx the idx number waiting for the core
* @param [in] eventID Set and wait events
*/
template <bool isAIVOnly>
__aicore__ inline void IBSet(
const GlobalTensor<int32_t>& gmWorkspace, const LocalTensor<int32_t>& ubWorkspace, int32_t blockIdx,
int32_t eventID)
{
int32_t blockNum = GetBlockNum();
#if (__NPU_ARCH__ == 2201) || (__NPU_ARCH__ == 3510) || (__NPU_ARCH__ == 5102)
#if (__NPU_ARCH__ != 5102)
if ASCEND_IS_AIC {
return;
}
#endif
if (!isAIVOnly) {
blockNum = GetBlockNum() * 2;
}
#endif
#if __NPU_ARCH__ == 2201 || __NPU_ARCH__ == 2002
__ib_set_stub(blockIdx, eventID, isAIVOnly);
#endif
auto localSyncGM = gmWorkspace[blockNum * 8 * eventID + blockIdx * 8];
pipe_barrier(PIPE_ALL);
while (true) {
DataCopy(ubWorkspace, localSyncGM, ONE_BLK_SIZE / sizeof(int32_t));
event_t eventIdMte2ToS = static_cast<event_t>(GetTPipePtr()->FetchEventID(HardEvent::MTE2_S));
SetFlag<HardEvent::MTE2_S>(eventIdMte2ToS);
WaitFlag<HardEvent::MTE2_S>(eventIdMte2ToS);
if (ubWorkspace.GetValue(0) == 0) {
ubWorkspace.SetValue(0, 1);
event_t eventIdSToMte3 = static_cast<event_t>(GetTPipePtr()->FetchEventID(HardEvent::S_MTE3));
SetFlag<HardEvent::S_MTE3>(eventIdSToMte3);
WaitFlag<HardEvent::S_MTE3>(eventIdSToMte3);
DataCopy(localSyncGM, ubWorkspace, ONE_BLK_SIZE / sizeof(int32_t));
break;
}
}
pipe_barrier(PIPE_ALL);
#if __NPU_ARCH__ == 2201 || __NPU_ARCH__ == 2002
__ib_set_stub(blockIdx, eventID, isAIVOnly);
#endif
}
template <bool isAIVOnly>
__aicore__ inline void IBWait(
const GlobalTensor<int32_t>& gmWorkspace, const LocalTensor<int32_t>& ubWorkspace, int32_t blockIdx,
int32_t eventID)
{
int32_t blockNum = GetBlockNum();
#if (__NPU_ARCH__ == 2201) || (__NPU_ARCH__ == 3510) || (__NPU_ARCH__ == 5102)
#if (__NPU_ARCH__ != 5102)
if ASCEND_IS_AIC {
return;
}
#endif
if (!isAIVOnly) {
blockNum = GetBlockNum() * 2;
}
#endif
#if __NPU_ARCH__ == 2201 || __NPU_ARCH__ == 2002
__ib_wait_stub(blockIdx, eventID, isAIVOnly);
#endif
auto localSyncGM = gmWorkspace[blockNum * 8 * eventID + blockIdx * 8];
pipe_barrier(PIPE_ALL);
while (true) {
DataCopy(ubWorkspace, localSyncGM, ONE_BLK_SIZE / sizeof(int32_t));
event_t eventIdMte2ToS = static_cast<event_t>(GetTPipePtr()->FetchEventID(HardEvent::MTE2_S));
SetFlag<HardEvent::MTE2_S>(eventIdMte2ToS);
WaitFlag<HardEvent::MTE2_S>(eventIdMte2ToS);
if (ubWorkspace.GetValue(0) == 1) {
ubWorkspace.SetValue(0, 0);
event_t eventIdSToMte3 = static_cast<event_t>(GetTPipePtr()->FetchEventID(HardEvent::S_MTE3));
SetFlag<HardEvent::S_MTE3>(eventIdSToMte3);
WaitFlag<HardEvent::S_MTE3>(eventIdSToMte3);
DataCopy(localSyncGM, ubWorkspace, ONE_BLK_SIZE / sizeof(int32_t));
break;
}
}
pipe_barrier(PIPE_ALL);
#if __NPU_ARCH__ == 2201 || __NPU_ARCH__ == 2002
__ib_wait_stub(blockIdx, eventID, isAIVOnly);
#endif
}
* @ingroup：SyncALL
* @brief：Set flag bits of all cores
* @param [in] gmWorkspace GlobalTensor to store core state
* @param [in] ubWorkspce LocalTensor for current core
*/
template <bool isAIVOnly>
__aicore__ inline void SyncAll(
const GlobalTensor<int32_t>& gmWorkspace, const LocalTensor<int32_t>& ubWorkspace, const int usedCores)
{
#if ASCENDC_CPU_DEBUG
SoftSyncAllImpl<false>(
(__gm__ int32_t*)gmWorkspace.GetPhyAddr(), (__ubuf__ int32_t*)ubWorkspace.GetPhyAddr(), usedCores);
#else
SoftSyncAllImpl<isAIVOnly>(
(__gm__ int32_t*)gmWorkspace.GetPhyAddr(), (__ubuf__ int32_t*)ubWorkspace.GetPhyAddr(), usedCores);
#endif
}
#if (__NPU_ARCH__ == 3510) || (__NPU_ARCH__ == 5102)
template <bool isAIVOnly, const SyncAllConfig& config>
__aicore__ inline void SyncAll()
{
SyncAllImpl<isAIVOnly, config.triggerPipe, config.waitPipe>();
}
#else
template <bool isAIVOnly>
__aicore__ inline void SyncAll()
{
SyncAllImpl<isAIVOnly>();
}
#endif
template <uint8_t modeId, pipe_t pipe>
__aicore__ inline void CrossCoreSetFlag(uint16_t flagId)
{
NotifyEventImpl<modeId, pipe>(flagId);
}
template <uint8_t modeId, pipe_t pipe>
__aicore__ inline void CrossCoreWaitFlag(uint16_t flagId)
{
WaitEventImpl<modeId, pipe>(flagId);
}
#if (__NPU_ARCH__ == 3510) || (__NPU_ARCH__ == 5102)
template <pipe_t pipe, uint8_t subBlockSyncMode = 2>
__aicore__ inline void NotifyEvent(uint16_t flagId)
{
NotifyEventImpl<subBlockSyncMode, pipe>(flagId);
}
template <pipe_t pipe = PIPE_S, uint8_t mode = 0>
__aicore__ inline void WaitEvent(uint16_t flagId)
{
WaitEventImpl<mode, pipe>(flagId);
}
#else
template <pipe_t pipe>
__aicore__ inline void NotifyEvent(uint16_t flagId)
{
constexpr uint8_t subBlockSyncMode = 0x02;
NotifyEventImpl<subBlockSyncMode, pipe>(flagId);
}
template <pipe_t pipe = PIPE_S>
__aicore__ inline void WaitEvent(uint16_t flagId)
{
constexpr uint8_t mode = 0;
WaitEventImpl<mode, pipe>(flagId);
}
#endif
}
#endif
#if defined(__UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_OPERATOR_BLOCK_SYNC_INTF_IMPL_H__)
#undef __ASCENDC_INCLUDE_INTERNAL_HEADERS__
#undef __UNDEF_ASCENDC_INCLUDE_INTERNAL_HEADERS_KERNEL_OPERATOR_BLOCK_SYNC_INTF_IMPL_H__
#endif
