[SPMD] Support reduce-scatter in manual sharding#7231
Merged
alanwaketan merged 9 commits intomasterfrom Jun 11, 2024
Merged
Conversation
Collaborator
Author
|
Why the hell do we run spmd tests on cpu? |
Collaborator
|
You can skip your test for CPU. For the most sharding related test they actually passed on CPU. CPU test is the only CI that upstream runs against us. |
JackCaoG
reviewed
Jun 10, 2024
| return result_tuple; | ||
| }); | ||
| m.def( | ||
| "_xla_spmd_reduce_scatter", |
Collaborator
There was a problem hiding this comment.
so the only difference is this one does not have token?
Collaborator
Author
There was a problem hiding this comment.
That's one difference. Others are mentioned in the description.
JackCaoG
approved these changes
Jun 11, 2024
Collaborator
Author
|
The GPU test failure doesn't seem to be related. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary:
This PR is to add experimental support of cc ops in manual sharding zones. This one adds reduce-scatter as the initial step. The key here is to add channel_id, replica_groups, and use_global_device_ids in the lowering.
Test Plan:
PJRT_DEVICE=TPU XLA_USE_SPMD=1 python test/spmd/test_xla_sharding.py -v -k test_spmd_reduce_scatter