-
Notifications
You must be signed in to change notification settings - Fork 24.3k
[tp] propagate src_data_rank kwarg in TP API #144005
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
Conversation
as titled, this PR propagates the src_data_rank in the TP API, so that module level APIs could leverage the flexibility to choose src_data_rank, and avoid the communication if it does not need to [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/144005
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit cd708c2 with merge base d88a8c4 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
as titled, this PR propagates the src_data_rank in the TP API, so that module level APIs could leverage the flexibility to choose src_data_rank, and avoid the communication if it does not need to cc H-Huang awgu kwen2501 fegin fduwjj wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. One nit is that the test only covers 1D device mesh case, but the code is supposed to work under n-D.
The TP API is suppose to work with 1D DeviceMesh though, but I think we can test n-D one with the |
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Stack from ghstack (oldest at bottom):
as titled, this PR propagates the src_data_rank in the TP API, so that
module level APIs could leverage the flexibility to choose
src_data_rank, and avoid the communication if it does not need to
cc @H-Huang @awgu @kwen2501 @fegin @fduwjj @wz337 @wconstab @d4l3k @c-p-i-o