re-enable delta optimizer test and fix a bug in adasum communicator init logic #3379
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
re-enable delta optimizer test and fix a bug in adasum communicator init logic
Signed-off-by: TJ tix@uber.com
Checklist before submitting
Description
This addresses issue 3314. Torch's delta_optimizer test was disabled due to a deadlock issue with torch 1.10, this is to enable the test with a barrier at the end to properly synchronize. While the main root cause was due to the python GIL which was fixed by this pr, another bug with adasum reduction comm init logic was discovered during debugging, adasum was always initialized relative to mpi_world, the correct relative communicator should be mpi_context->mpi_comm.
Fixes # (issue).
#3314
Review process to land