Merge remote-tracking branch 'nccl/master' into develop

[ROCm/rccl commit: 84081064a0]
Этот коммит содержится в:
BertanDogancay
2024-10-02 09:29:22 -05:00
родитель 74aa13afbe 792c33598f
Коммит 9059445acb
54 изменённых файлов: 2158 добавлений и 964 удалений
+8 -6
Просмотреть файл
@@ -211,9 +211,10 @@ ncclResult_t netSendSetup(struct ncclComm* comm, struct ncclTopoGraph* graph, st
req.netDev = -1;
int proxyRank = myInfo->rank;
if (connIndex == NCCL_CONN_IDX_P2P_NET) NCCLCHECK(ncclTopoGetIntraNetDev(comm->topo, myInfo->rank, graph, channelId, 1, &req.netDev));
if (req.netDev < 0) NCCLCHECK(ncclTopoGetNetDev(comm, myInfo->rank, graph, channelId, peerInfo->rank, &req.netDev, &proxyRank));
NCCLCHECK(ncclTopoCheckGdr(comm->topo, myInfo->busId, req.netDev, 1, &req.useGdr));
int64_t netId;
if (connIndex == NCCL_CONN_IDX_P2P_NET) NCCLCHECK(ncclTopoGetIntraNetDev(comm->topo, myInfo->rank, graph, channelId, 1, &netId, &req.netDev));
if (req.netDev < 0) NCCLCHECK(ncclTopoGetNetDev(comm, myInfo->rank, graph, channelId, peerInfo->rank, &netId, &req.netDev, &proxyRank));
NCCLCHECK(ncclTopoCheckGdr(comm->topo, myInfo->busId, netId, 1, &req.useGdr));
if (proxyRank == myInfo->rank) {
INFO(NCCL_INIT|NCCL_NET,"Channel %02d/%d : %d[%lx] -> %d[%lx] [send] via NET/%s/%d%s%s", channelId, connIndex, myInfo->rank, myInfo->busId, peerInfo->rank, peerInfo->busId, comm->ncclNet->name, req.netDev,
@@ -238,9 +239,10 @@ ncclResult_t netRecvSetup(struct ncclComm* comm, struct ncclTopoGraph* graph, st
// Use myInfo->rank as the receiver uses its own NIC
int proxyRank = myInfo->rank;
if (connIndex == NCCL_CONN_IDX_P2P_NET) NCCLCHECK(ncclTopoGetIntraNetDev(comm->topo, myInfo->rank, graph, channelId, 0, &req.netDev));
if (req.netDev < 0) NCCLCHECK(ncclTopoGetNetDev(comm, myInfo->rank, graph, channelId, myInfo->rank, &req.netDev, &proxyRank));
NCCLCHECK(ncclTopoCheckGdr(comm->topo, myInfo->busId, req.netDev, 0, &req.useGdr));
int64_t netId;
if (connIndex == NCCL_CONN_IDX_P2P_NET) NCCLCHECK(ncclTopoGetIntraNetDev(comm->topo, myInfo->rank, graph, channelId, 0, &netId, &req.netDev));
if (req.netDev < 0) NCCLCHECK(ncclTopoGetNetDev(comm, myInfo->rank, graph, channelId, myInfo->rank, &netId, &req.netDev, &proxyRank));
NCCLCHECK(ncclTopoCheckGdr(comm->topo, myInfo->busId, netId, 0, &req.useGdr));
INFO(NCCL_INIT|NCCL_NET,"Channel %02d/%d : %d[%lx] -> %d[%lx] [receive] via NET/%s/%d%s%s", channelId, connIndex, peerInfo->rank, peerInfo->busId, myInfo->rank, myInfo->busId, comm->ncclNet->name, req.netDev,
req.useGdr ? "/GDRDMA" : "", req.shared ? "/Shared" : "");