AaronBallman wrote: > > > > That's how this already works in this patch: > > > > AH! I missed that this was doing that. Awesome! Though I think the > > > > 'no' flag is the one that makes sense in a "turn it on for all BUT > > > > these". The flag as typed there isn't particularly useful (that is, > > > > ONLY warn me for the ones that I know about). > > > > > > > > > > I think what Corentin is asking for is a similar (follow-up?) feature > > > > that allows you to say "warn me about unknown attributes, but not of > > > > unknown attributes." that can take attributes with or without a > > > > namespace. > > > > > > > > > I'm not hearing a difference? > > > > > > Wow. My fingers are obviously not listening to my brain very well. :-D > > I mean "warn me about unknown attribute, but not these particular unknown > > attributes." e.g., (fake warning group name) > > ``` > > // -Wwarn-about-unknown-attributes-except-these=foo,bar::baz > > [[foo, bar::baz, quux]] int x; // warning: unknown attribute 'quux' > > ``` > > Ah, hrm. I see much less value in that. How likely is it that you have an > attribute in a 'known' namespace that is 'unknown' but valid?
For example, we know about the `gnu` namespace, but there are a ton of attributes GCC supports but we do not. https://github.com/llvm/llvm-project/pull/120925 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
