Moved mscclpp_ncclGetUniqueId call into ncclCommInitRankFunc (#1332)

* Moved call to `mscclpp_ncclGetUniqueId` into `ncclCommInitRankFunc` to avoid setting up transport early in environments where MSCCL++ isn't valid.

* Checking `mscclEnabled` for the process and the topology to gate MSCCL++.

* Allowed `mscclForceEnable` to enable MSCCL++.

[ROCm/rccl commit: 853a0586b4]
Dieser Commit ist enthalten in:
corey-derochie-amd
2024-09-16 16:41:40 -06:00
committet von GitHub
Ursprung 31b576b969
Commit cf48e57bd9
3 geänderte Dateien mit 38 neuen und 33 gelöschten Zeilen
@@ -84,6 +84,12 @@ static bool mscclCommCompatible(ncclComm_t comm) {
return true;
}
#ifdef ENABLE_MSCCLPP
bool mscclppCommCompatible(ncclComm_t comm) {
return mscclCommCompatible(comm);
}
#endif
const char *mscclFuncNames[] = {
"mscclFuncReduce",
"mscclFuncBroadcast",