Skip to content

Commit a29dba1

Browse files
committed
debug
1 parent 132abd4 commit a29dba1

File tree

1 file changed

+5
-0
lines changed

1 file changed

+5
-0
lines changed

torch/distributed/run.py

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -713,6 +713,11 @@ def determine_local_world_size(nproc_per_node: str):
713713
else:
714714
num_proc = os.cpu_count()
715715
device_type = "cpu"
716+
### CI ERROR DEBUG CODE BEGIN ###
717+
print(f"nproc_per_node=auto: {device_type = }, {num_proc = }, {os.cpu_count() = }")
718+
if torch.cuda.is_available():
719+
print(f"{torch.cuda.device_count() = }")
720+
### CI ERROR DEBUG CODE END ###
716721
else:
717722
raise ValueError(
718723
f"Unsupported nproc_per_node value: {nproc_per_node}"

0 commit comments

Comments
 (0)