Line data    Source code

       1              : /**
       2              :  * Copyright (c) 2025 Huawei Technologies Co., Ltd.
       3              :  * This program is free software, you can redistribute it and/or modify it under the terms and conditions of
       4              :  * CANN Open Software License Agreement Version 2.0 (the "License").
       5              :  * Please refer to the License for details. You may not use this file except in compliance with the License.
       6              :  * THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
       7              :  * INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
       8              :  * See LICENSE in the root of the software repository for the full text of the License.
       9              :  */
      10              : 
      11              : #ifndef HCCLV2_CCU_TEMP_ALL_REDUCE_MESH_DETOUR_1D_H_
      12              : #define HCCLV2_CCU_TEMP_ALL_REDUCE_MESH_DETOUR_1D_H_
      13              : 
      14              : #include <vector>
      15              : #include <map>
      16              : #include <string>
      17              : #include <hccl/hccl_types.h>
      18              : #include "ccu_instruction_all_reduce_mesh1d_detour.h"
      19              : #include "reduce_op.h"
      20              : #include "hccl/base.h"
      21              : #include "types/types.h"
      22              : #include "string_util.h"
      23              : #include "data_type.h"
      24              : #include "template_utils.h"
      25              : #include "ccu_alg_template_base.h"
      26              : 
      27              : namespace Hccl {
      28              : 
      29              : class CcuTempAllReduceMeshDetour1D : public CcuAlgTemplateBase {
      30              : public:
      31              :     explicit CcuTempAllReduceMeshDetour1D(const RankId virtualRank, const u32 tempRankSize,
      32              :                                     const std::vector<std::vector<RankId>> &tempVTopo,
      33              :                                     const std::map<RankId, u32>            &tempVirtRankMap);
      34              :     ~CcuTempAllReduceMeshDetour1D() override;
      35              : 
      36            0 :     std::string Describe() const override
      37              :     {
      38            0 :         return StringFormat("Template of All Reduce ccu mesh 1D detour with tempRankSize [%u].", tempRankSize_);
      39              :     }
      40              : 
      41              :     HcclResult Run(const TempFuncs &tempFuncs, const RankSliceInfo &sliceInfoVec, const BuffInfo &buffInfo,
      42              :                          const ResLinks &tempLinks, std::vector<InsQuePtr> &tempInsQues) override;
      43              :     HcclResult CalcResDetour(const RankGraph *rankGraph, AlgTempResReq &tempResReq) override;
      44              :     HcclResult CalcResDetour(ConnectedLinkMgr *linkMgr, AlgTempResReq &tempResReq) override;
      45              :     HcclResult CalcSliceInfo(const AllignInfo &allignInfo, const u64 dataSize, RankSliceInfo &sliceInfoVec) override;
      46              :     // init reduceInfo
      47              :     void InitReduceInfo(const ReduceOp &reduceOp, const DataType &dataType);
      48              : 
      49              : private:
      50              :     void GetAddrInfo(const TempFuncs &tempFuncs, uint64_t &inputAddr, uint64_t &outputAddr);
      51              :     void CalcDetourOffset(uint64_t sliceSize, uint64_t &tailOffset, uint64_t &tailSize, uint64_t &iterNum);
      52              :     void ProcessLinks(std::vector<LinkData> &links, const ResLinks &tempLinks) const;
      53              :     ReduceOp reduceOp_;
      54              :     DataType dataType_;
      55              :     uint64_t detourPathNum_{0};  // 到每个对端有几个绕路路径
      56              :     uint64_t pathNumPerPeer_{0};
      57              :     std::vector<uint64_t> lengths_;
      58              :     uint64_t singleTransportSize_{0};
      59              : };
      60              : }
      61              : #endif // HCCLV2_CCU_TEMP_ALL_REDUCE_MESH_DETOUR_1D_H_