nilness: bound generated state machine analysis - #1728
Closed
zbarsky-openai wants to merge 1 commit into
Closed
Conversation
Dense nilness dataflow can spend more than 263 seconds analyzing a single generated go-json state machine. Return the existing conservative unknown-result facts only for generated functions with more than 1,000 IR blocks; leave handwritten functions and smaller generated functions unchanged. Construct the oversized generated state machine in a focused test instead of committing a large fixture.
Owner
|
Thank you, but the number of blocks is a poor proxy for CFG complexity. There is also a better way to fix this, by improving the data flow analysis runner. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Dense nilness dataflow can spend more than 263 seconds analyzing one generated go-json encoder, even after the source carries the standard generated-file marker. The observed case was the vm_color_indent nogo action in a focused downstream build; the full invocation was interrupted after 286 seconds.
For functions with more than 1,000 IR blocks, check that the function belongs to a generated file before using the existing conservative unknown-return bailout. All handwritten functions and smaller generated functions retain their current nilness analysis. The new focused regression test constructs the large generated state machine at runtime instead of committing a large fixture.
This is intentionally separate from the recover fix in #1726.