dblaikie added a comment. Hi - sorry about the stalled opaque pointer types effort.
For my money - ideally - if someone comes across a bug caused by incorrect pointee types, ideally that would be fixed by adjusting whatever piece of code was depending on that pointee type being correct to not depend on that information. Though, yes, there are likely cases where a small bug in the type information exposes vast swathes of LLVM code - where the argument might reasonably made that the cleanup would take too long to leave things broken. Worth seeing what that looks like, though, before making the call - hopefully. https://reviews.llvm.org/D49403 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
