2.20.3-1

Add support for alternating rings, allow for cross-nic rings without cross-rail communication. Add support for user buffer registration for network send/recv. Optimize aggregated operations to better utilize all channels. Add flattening for BCM PCI gen5 switches. Add support for inter-node NVLink communication Add support for port fusion in NET/IB. Add support for ReduceScatter and AllGather using Collnet. Update net API to v8. Fix hang during A2A connection. [ROCm/rccl commit: b6475625fb]
2024-02-05 05:06:02 -08:00
parent 382ae08419
commit 2ab8a3a750
74 changed files with 4632 additions and 2165 deletions
@@ -18,6 +18,7 @@ struct ncclShmemCollBuff {
  volatile size_t *cnt[2];
  volatile void *ptr[2];
  int round;
+  size_t maxTypeSize;
 };

 ncclResult_t ncclShmemAllgather(struct ncclComm *comm, struct ncclShmemCollBuff *shmem, void *sendbuff, void *recvbuff, size_t typeSize);