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]
This commit is contained in:
@@ -481,6 +481,7 @@ struct ncclComm {
|
||||
|
||||
int node;
|
||||
int nNodes;
|
||||
int rcclUseOneSlice; // RCCL: true if this comm is using one slice per primitive
|
||||
int localRank;
|
||||
int localRanks;
|
||||
int maxLocalRanks;
|
||||
|
||||
Reference in New Issue
Block a user