Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add warnings for explicit choose and old VC gen #1336

Merged
merged 1 commit into from
Nov 17, 2022

Conversation

mario-bucev
Copy link
Collaborator

No description provided.

@vkuncak
Copy link
Collaborator

vkuncak commented Nov 14, 2022

Another thing that might help users understand trusted proof base would be if we could print a record of the pipeline invoked in a given verification run, indicating in one line concise string whether any of the options that may increase trusted proof base were invoked in a given verification run:

  • whether --admit-vcs was used (empty string if not)
  • whether termination was turned off
  • whether cache was used (empty if not)
  • whether antialiasing and imperative elimination were used (empty if not)
  • which smt solvers were invoked (including internal ones and princess, if used)
  • whether type encoding did anything (empty string if not)
  • whether non-library extern functions were used
  • whether batch mode was on

@mario-bucev
Copy link
Collaborator Author

I like this idea. I'll check what should be done to integrate this feature, ideally within this PR.

@mario-bucev mario-bucev marked this pull request as draft November 15, 2022 05:28
@mario-bucev mario-bucev force-pushed the embarrassing-features branch from 825402d to a6c4932 Compare November 17, 2022 07:01
@mario-bucev
Copy link
Collaborator Author

I've pushed what I hope corresponds to the above comment. Here is the concise summary printed for ScalaList (after the verification report):

[  Info  ] Verification pipeline summary:
[  Info  ]   cache, anti-aliasing, type encoding, choose injection, nativez3

With --extended-summary, we have extra information:

[  Info  ] Verification pipeline summary:
[  Info  ]   Cache used
[  Info  ]   Anti-aliasing phase transformed the following functions:
[  Info  ]     length
[  Info  ]   Type-encoding transformed the following:
[  Info  ]     -Functions:
[  Info  ]       ++, ::, head, length, map, tail
[  Info  ]     -Classes:
[  Info  ]       Cons, List, Nil
[  Info  ]   Choose injected in the following functions:
[  Info  ]     -User:
[  Info  ]       ++, ::, head, length, map, tail
[  Info  ]   Solvers used: nativez3
[  Info  ]   Processing mode: partial

The choose injected in functions ++, etc. are for their copies, not the original ones. How could we convey such information? (I'm thinking of something like "++ (derived from original ++)")

@mario-bucev mario-bucev force-pushed the embarrassing-features branch from a6c4932 to eb8ba8c Compare November 17, 2022 07:16
@vkuncak
Copy link
Collaborator

vkuncak commented Nov 17, 2022

If the function was copied, that's similar from user point of view as if it was modified, so no need for that.

@vkuncak vkuncak marked this pull request as ready for review November 17, 2022 08:47
@vkuncak vkuncak merged commit bc481e2 into epfl-lara:main Nov 17, 2022
@mario-bucev mario-bucev deleted the embarrassing-features branch November 17, 2022 09:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants