[PATCH] D111453: Delay trailing constraing evaluation so we short-circuit

erichkeane abandoned this revision. erichkeane added a comment. Looking into my other issue, I think I'm reasonably convinced this is wrong. I was looking into the PR44833 blocker for 'ranges' which hits the VisitCXXRecordMethod part, and removing THAT breaks the constraint entirely, so I thin

[PATCH] D111453: Delay trailing constraing evaluation so we short-circuit

erichkeane added a comment. So debugging a DIFFERENT issue, I found that this is likely incomplete since we likely have a similar problem in `VisitCXXMethodDecl` below this. I can extend the tests here/fix easily enough, if we decide this is the right way around this. CHANGES SINCE LAST ACTI

[PATCH] D111453: Delay trailing constraing evaluation so we short-circuit

erichkeane added a comment. The implementation seems to have tried checking this constraint 2x, and the first time required fully instantiating the constraint. The only condition in the check is PartialOrdering here: https://github.com/llvm/llvm-project/blob/main/clang/lib/Sema/SemaTemplateDed

[PATCH] D111453: Delay trailing constraing evaluation so we short-circuit

erichkeane created this revision. erichkeane added reviewers: rsmith, hubert.reinterpretcast, saar.raz, arthur.j.odwyer, aaron.ballman, changyu, nwilson, cor3ntin. erichkeane requested review of this revision. As reported as a part of PR47508 (and somewhat confirmed in PR44833), our behavior to f