Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Graphbolt] [NFC] Remove default backend torch.distributed.init_process_group calls #7870

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

[Graphbolt] [NFC] Remove default backend torch.distributed.init_process_group calls #7870

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2349c70
Remove default backend torch.distributed.init_process_group calls
tpopp Feb 22, 2025
b4654bc
Merge branch 'master' into dev/tpopp/torch-2.6
tpopp Feb 22, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions examples/graphbolt/pyg/multigpu/node_classification.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -390,7 +390,8 @@
def run(rank, world_size, args, dataset):
# Set up multiprocessing environment.
torch.cuda.set_device(rank)
dist.init_process_group(

Check warning on line 393 in examples/graphbolt/pyg/multigpu/node_classification.py

View workflow job for this annotation

GitHub Actions / lintrunner

UFMT format 

Run `lintrunner -a` to apply this patch.
backend='cpu:gloo,cuda:nccl',
init_method="tcp://127.0.0.1:12345",
rank=rank,
world_size=world_size,
Expand Down
1 change: 1 addition & 0 deletions tests/python/pytorch/graphbolt/test_dataloader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,7 +88,8 @@
if platform == "win32"
else "tcp://127.0.0.1:12345"
)
thd.init_process_group(

Check warning on line 91 in tests/python/pytorch/graphbolt/test_dataloader.py

View workflow job for this annotation

GitHub Actions / lintrunner

UFMT format 

Run `lintrunner -a` to apply this patch.
backend='cpu:gloo,cuda:nccl',
init_method=init_method,
world_size=1,
rank=0,
Expand Down
Loading