cor3ntin wrote: > CC @AaronBallman @cor3ntin Do we want to make this a warning that defaults > to an error instead? Imo it’s probably fine as-is because I don’t imagine too > many people would write code that abuses us not diagnosing this...
It looks fine to me as is. Code that would do something like `static_cast<BaseReturnType>(derived->f())` (which I think is perfectly reasonable) would explode, despite the standard guarantees it should work. If people scream we have plenty of time to reconsider making it a warning https://github.com/llvm/llvm-project/pull/111856 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
