Use One Slice per Basic Primitive for AllReduce, ReduceScatter, AllGather (#1681) for Single Node on Some GFX9 Systems

Using a single slice rather than the typical two provides about 5% speedup (sometimes more or less) on some GFX9 systems for single node.

[ROCm/rccl commit: 2f6b20c00a]

projects/rccl/src/init.cc

@@ -1353,6 +1353,8 @@ static ncclResult_t initTransportsRank(struct ncclComm* comm, struct ncclComm* p
     // Multi-node MI300A
     int managed = 0;
     CUDACHECK(hipDeviceGetAttribute(&managed, hipDeviceAttributeDirectManagedMemAccessFromHost, 0));
+    // RCCL: Only use one slice per primitive on some single node gfx9xx systems
+    comm->rcclUseOneSlice = !managed && nNodes == 1;
     if (managed && nNodes > 1) {
       // This forces the minimum channels to 24
       allGather3Data[rank].nc = 6;