sammccall added a comment.

Yeah, so I guess the point of marking decls invalid is in large part to stop 
recovery using it.
It's not surprising 1 has recovery regressions. 3 should have fewer (basically 
because you'll get the dependent bits too), but likely still some.
2 definitely seems the way to go given that this is currently inconsistent 
between different invalid ways of initializing a variable. Thanks for D77395 
<https://reviews.llvm.org/D77395>!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76831/new/

https://reviews.llvm.org/D76831



_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
  • [PATCH] D76831: [... Sam McCall via Phabricator via cfe-commits
    • [PATCH] D768... Haojian Wu via Phabricator via cfe-commits
    • [PATCH] D768... Sam McCall via Phabricator via cfe-commits
    • [PATCH] D768... Richard Smith - zygoloid via Phabricator via cfe-commits

Reply via email to