Disable the constant folding output-size limit by default - #32150
Closed
pragyam32 wants to merge 1 commit into
Closed
Disable the constant folding output-size limit by default#32150pragyam32 wants to merge 1 commit into
pragyam32 wants to merge 1 commit into
Conversation
The output-size guard added in microsoft#28055 skips any node whose output size cannot be estimated before execution. Ops that have no ONNX shape inference function, such as the GreaterOrEqual and LessOrEqual functions below opset 16, reach the guard with no inferred output shape, so since 1.27 they are never constant folded and the entire downstream constant chain stops folding with them. No positive cap value avoids this because the bail is inside the guard. Default the cap to 0 so folding behaves as it did in 1.26, and make the mitigation opt-in through optimization.constant_folding_max_output_size_in_bytes. Fixes microsoft#32130 Co-authored-by: Cursor <cursoragent@cursor.com>
|
Azure Pipelines: There may be pipelines that require an authorized user to comment /azp run to run. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The output-size guard added in #28055 skips any node whose output size cannot be estimated before execution. Ops that have no ONNX shape inference function, such as the GreaterOrEqual and LessOrEqual functions below opset 16, reach the guard with no inferred output shape, so since 1.27 they are never constant folded and the entire downstream constant chain stops folding with them. No positive cap value avoids this because the bail is inside the guard.
Default the cap to 0 so folding behaves as it did in 1.26, and make the mitigation opt-in through
optimization.constant_folding_max_output_size_in_bytes.
Fixes #32130