Fix goal-tactic association in typeclass search#22037
Open
Janno wants to merge 6 commits into
Open
Conversation
coqbot-app
Bot
added
the
needs: full CI
Contributor
Author
|
@coqbot run full ci |
coqbot-app
Bot
removed
the
needs: full CI
Member
|
@coqbot bench |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The code was written by GPT-5.5 based on the findings from my debugging attempts in #21044. I have not changed the LLM's code yet. I hope the change is small enough to be reviewable. If the mechanism is acceptable, I will polish the code.
This change introduces a notion of search "generations" to
search_fixpoint. This allows us to track precisely which goals have been stuck for how long and which ones need to be retried when we make any kind of progress. Then, to make sure that goals and tactics stay properly aligned in the presence of goal reorderings, we look up the evar's index in the proofview goals and focus that index explicitly.The lookup is currently a linear search through the list of goals. I have previously tried to work with goal state instead. My idea was to keep an index in the goal that maps to a tactic in some kind of array/map. But Class_tactics drops goal state in several places and I could not make the change work. The LLM might have a better chance so I am happy to try that if the linear search is a problem and there is no easy workaround.
The change has been tested on skylabs code where it brings a very modest 0.1% overall speedup with no significant slowdowns. Full report
Fixes / closes #21044
make doc_gram_rsts.