xla
Fix: Ensure checkpoint deletion only occurs on rank 0
#7296

Merged

Fix: Ensure checkpoint deletion only occurs on rank 0 #7296

JackCaoG merged 2 commits into pytorch:master from huzama:chkpt_mngr_delete_chkpt

Fix: only delete chkpt at step from rank 0

55aa2b55

JackCaoG requested a review from

jonb377 1 year ago

jonb377 approved these changes on 2024-06-17

Add rank check to `_delete_chkpt_at_step` to ensure only rank 0 delet…

8b1039fc

huzama requested a review from

jonb377 1 year ago

jonb377 approved these changes on 2024-06-21

JackCaoG merged 85143432 into master 1 year ago

huzama deleted the chkpt_mngr_delete_chkpt branch 1 year ago

Reviewers

jonb377

Assignees

No one assigned

Labels

None yet

Milestone

No milestone