-
Notifications
You must be signed in to change notification settings - Fork 24.9k
[DTensor] Fix aten.all strategy with min instead of sum as the reduce_op #155420
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.
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
base: main
Are you sure you want to change the base?
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/155420
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (7 Unrelated Failures)As of commit e456fea with merge base 706bc41 ( FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
@pytorchbot label "topic: not user facing" |
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
For
aten.all
, when all_reduce withmin
, if one rank returns torch.Tensor([False]), the reduction result will be torch.Tensor([False]). Only when all ranks return torch.Tensor([True]) will return torch.Tensor([True]). the behavior matches the aten.all.cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta