Line data Source code
1 : /**
2 : * Copyright (c) 2026 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 : #include "uboe_endpoint.h"
12 : #include "endpoint_mgr.h"
13 : #include "log.h"
14 : #include "hccl/hccl_res.h"
15 : #include "urma_mem.h"
16 : #include "proc_reged_mem_mgr_cache.h"
17 : #include "adapter_rts_common.h"
18 :
19 : namespace hcomm {
20 :
21 13 : UboeEndpoint::UboeEndpoint(const EndpointDesc &endpointDesc)
22 13 : : UboeUbgEndpointHelper(endpointDesc)
23 : {
24 13 : }
25 :
26 26 : UboeEndpoint::~UboeEndpoint() noexcept
27 : {
28 13 : ProcRegedMemMgrCache::GetInstance().Release(cacheKey_);
29 26 : }
30 :
31 13 : HcclResult UboeEndpoint::Init()
32 : {
33 13 : HCCL_INFO("[%s] localEndpoint protocol[%d]", __func__, endpointDesc_.protocol);
34 :
35 13 : Hccl::IpAddress ipAddr{};
36 13 : CHK_RET(CommAddrToIpAddress(endpointDesc_.commAddr, ipAddr));
37 :
38 : s32 deviceLogicId;
39 : u32 devPhyId;
40 11 : CHK_RET(hrtGetDevice(&deviceLogicId));
41 11 : CHK_RET(hrtGetDevicePhyIdByIndex(deviceLogicId, devPhyId));
42 11 : endpointDesc_.loc.device.devPhyId = devPhyId;
43 :
44 11 : Hccl::HccpHdcManager::GetInstance().Init(deviceLogicId);
45 11 : auto &rdmaHandleMgr = Hccl::RdmaHandleManager::GetInstance();
46 11 : Hccl::IpAddress eidAddress{};
47 11 : rdmaHandleMgr.UboeIpv4ToEid(ipAddr, eidAddress, devPhyId);
48 11 : EXCEPTION_CATCH(ctxHandle_ = static_cast<void *>(rdmaHandleMgr.GetByIp(endpointDesc_.loc.device.devPhyId, eidAddress)),
49 : return HCCL_E_PARA);
50 11 : CHK_PTR_NULL(ctxHandle_);
51 11 : HCCL_INFO("%s success, devPhyId[%u], eidAddress[%s], ctxHandle[%p]",
52 : __func__, devPhyId, eidAddress.Describe().c_str(), ctxHandle_);
53 :
54 11 : cacheKey_ = MemMgrCacheKey{devPhyId, COMM_PROTOCOL_UBC_CTP, ipAddr, LocTypeToPortType(endpointDesc_.loc.locType)};
55 11 : auto &cache = ProcRegedMemMgrCache::GetInstance();
56 13 : EXCEPTION_CATCH(regedMemMgr_ = cache.GetOrCreate(cacheKey_, [this]() {
57 : auto m = std::make_shared<UbRegedMemMgr>();
58 : m->rdmaHandle_ = this->ctxHandle_;
59 : return m;
60 : }), return HCCL_E_INTERNAL);
61 :
62 11 : return HcclResult::HCCL_SUCCESS;
63 : }
64 :
65 : }
66 :
|