Xazax-hun wrote: > I don't have strong opinion, I just want to fix annoying false positives with > ctu =)
Thanks for working on this, I really appreciate it! :) I would even say this is not related to ctu, because users could run into this behavior within a single translation unit. I just want to make sure we exhausted other options before considering something that might not compose. > it seems very unlikely... I usually find it hard to predict how a certain codebase would evolve. In case someone wants to generalize this checker to non-fuchsia handles (and now suddenly we evalcall way more functions), or fuchsia starts to add support for mixing handles (e.g., converting from a C handle to a fuchsia one), and suddenly this becomes a likely scenario that users could hit. To be clear, I also don't expect this to happen, at least not any time soon, but I don't want to avoid picking a solution that is not extensible in certain directions unless we absolutely have to. https://github.com/llvm/llvm-project/pull/111588 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
