In SPMD partitioner, preprocess the sharding on singleton dimensions (dimensions whose size is 1). #84908
+88
−4
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Outdated suggestions cannot be applied.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.
In SPMD partitioner, preprocess the sharding on singleton dimensions (dimensions whose size is 1).
It is meaningless to partition a dimension whose size is 1. Redundant padding and unpadding may be inserted. To avoid this, we replicate the sharding on these dimensions as a pre-processing.
Take the following input as example
Previous result with redundant instructions
Result with this improvement
Reverts a7703e7
FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#21273 from nvcastet:ncclCommInitRankScalable dd6362af36a1f4d22532ad15b2007527898b5fa1