[mpmd] Bypass max_clones limit for trivial ops in mesh inference.#1360
Merged
Conversation
226dda1 to
93e02a0
Compare
21ded3c to
7b61f51
Compare
When a meshless op has more fragment users than max_clones, and has operands, it gets wrapped in a new fragment per mesh instead of being cloned into each consumer fragment. For trivial ops (<=1 operand, 1 result, 0 regions, pure) like sdy.sharding_constraint, this wrapping creates millions of tiny fragments that are expensive to merge later. This change bypasses the max_clones limit for trivial ops since they are always cheap to clone. PiperOrigin-RevId: 918944215
7b61f51 to
b38eded
Compare
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
[mpmd] Bypass max_clones limit for trivial ops in mesh inference.
When a meshless op has more fragment users than max_clones, and has operands, it gets wrapped in a new fragment per mesh instead of being cloned into each consumer fragment. For trivial ops (<=1 operand, 1 result, 0 regions, pure) like sdy.sharding_constraint, this wrapping creates millions of tiny fragments that are expensive to merge later.
This change bypasses the max_clones limit for trivial ops since they
are always cheap to clone.