Revert "Use device_id in dist init to reduce NCCL communicator warmup & creation overhead" (#5786)
This commit is contained in:
@@ -1055,11 +1055,6 @@ def init_distributed_environment(
|
|||||||
world_size=world_size,
|
world_size=world_size,
|
||||||
rank=rank,
|
rank=rank,
|
||||||
timeout=timeout,
|
timeout=timeout,
|
||||||
device_id=torch.device(
|
|
||||||
f"cuda:{torch.cuda.current_device()}"
|
|
||||||
if hasattr(torch, "cuda") and torch.cuda.is_available()
|
|
||||||
else None
|
|
||||||
), # Allow NCCL to eagerly init communicator
|
|
||||||
)
|
)
|
||||||
|
|
||||||
# set the local rank
|
# set the local rank
|
||||||
|
|||||||
Reference in New Issue
Block a user