njames93 added a comment. In D82825#2123186 <https://reviews.llvm.org/D82825#2123186>, @aaron.ballman wrote:
> I can see arguments either way on this. Personally, I would not want the > check to diagnose this code because that would encourage people to widen the > scope and lifetime of `X` or require them to introduce a new compound scope > to get the same behavior, and I think that's more problematic than the `else` > following a `return`. I am not certain if others feel the same way though. I'll leave this for a bit before merging and see if anyone else wants a weigh in on the default llvm behaviour as this will likely have an effect on other contributors Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D82825/new/ https://reviews.llvm.org/D82825 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
