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_TEMP_ALL_GATHER_MESH
      12              : #define HCCLV2_TEMP_ALL_GATHER_MESH
      13              : 
      14              : #include "string_util.h"
      15              : 
      16              : #include "alg_template_base_v2.h"
      17              : #include "connected_link_mgr.h"
      18              : 
      19              : namespace Hccl {
      20              : 
      21              : class TempAllGatherMesh : public AlgTemplateBase {
      22              : public:
      23              :     explicit TempAllGatherMesh(const RankId virtualRank, const u32 tempRankSize,
      24              :                                const std::vector<std::vector<RankId>> &tempVTopo,
      25              :                                const std::map<RankId, u32>            &tempVirtRankMap);
      26              :     ~TempAllGatherMesh() override;
      27              : 
      28            0 :     std::string Describe() const override
      29              :     {
      30            0 :         return StringFormat("Template of all gather mesh with tempRankSize [%u].", tempRankSize_);
      31              :     }
      32              : 
      33              :     HcclResult GenPrimQue(const TempFuncs &tempFuncs, const RankSliceInfo &sliceInfoVec, const BuffInfo &buffInfo,
      34              :                           const ResLinks &tempLinks, std::vector<PrimQuePtr> &tempPrimQues) override;
      35              :     using AlgTemplateBase::CalcSliceInfo;
      36              :     HcclResult CalcSliceInfo(const AllignInfo &allignInfo, const u64 dataSize, RankSliceInfo &sliceInfoVec) override;
      37              : 
      38              :     using AlgTemplateBase::CalcRes;
      39              :     HcclResult CalcRes(AlgTempResReq &tempResReq) override;
      40              : 
      41              :     using AlgTemplateBase::CalcResDetour;
      42              :     HcclResult CalcResDetour(const RankGraph *rankGraph, AlgTempResReq &tempResReq) override;
      43              :     HcclResult CalcResDetour(ConnectedLinkMgr *linkMgr, AlgTempResReq &tempResReq) override;
      44              : 
      45              : private:
      46              :     HcclResult PreCopyOffload(const RankSliceInfo &sliceInfoVec, const bool forAllReduce,
      47              :                               std::vector<PrimQuePtr> &tempPrimQues);
      48              : 
      49              :     HcclResult RunMesh(const u32 myAlgRank, const std::vector<RankId> &vTopo, const RankSliceInfo &sliceInfoVec,
      50              :                        const ResLinks &tempLinks, std::vector<PrimQuePtr> &tempPrimQues);
      51              : 
      52              :     HcclResult RunIndividualPeerDetour(const RankId neighborRank, const SliceInfo &sendSlice,
      53              :                                        const SliceInfo &recvSlice, const ResLinks &tempLinks,
      54              :                                        std::vector<PrimQuePtr> &detourPrimQues);
      55              :     HcclResult RunIndividualPeer(const RankId neighborRank, const LinkData &neighborLinkData,
      56              :                                  const SliceInfo &sendSlice, const SliceInfo &recvSlice, PrimQuePtr currQue);
      57              :     HcclResult GetSendRecvLinks(const RankId neighborRank, const ResLinks &tempLinks,
      58              :                                 std::vector<std::vector<LinkDataIterator>> &sendRecvLinks) const;
      59              : 
      60              :     std::unique_ptr<PrimGroup> RunSendRecv(const RankId neighborRank, const LinkData &sendLinkData,
      61              :                                            const LinkData &recvLinkData, const SliceInfo &currSendSlice,
      62              :                                            const SliceInfo &currRecvSlice) const;
      63              : 
      64              :     u32 majorQueNum_       = 0;
      65              :     u32 queNumPerNeighbor_ = 1;
      66              : };
      67              : 
      68              : } // namespace Hccl
      69              : 
      70              : #endif // !HCCLV2_TEMP_ALL_GATHER_MESH