ab added a comment. In D82699#2118313 <https://reviews.llvm.org/D82699#2118313>, @arphaman wrote:
> @t.p.northover @ab I noticed that the use of "apple-a12" doesn't infer the > right target features when we're passing in a11 or older, so that's why my > test file has the `INFER-A12` separate line. Do you think this is a bug? This > is decided here: > > else if ((A = Args.getLastArg(options::OPT_mcpu_EQ))) > success = getAArch64ArchFeaturesFromMcpu(D, A->getValue(), Args, > Features); > > > and CPU value isn't checked here. Interesting. Does it even pass the newer CPU then? If so I imagine the features are still enabled in the backend because of the cpu, so this would only affect frontend behavior (e.g., preprocessor macros, which might in turn affect some intrinsic headers, etc..). Either way, I agree it's a bug But the apple-a12 floor isn't a hard requirement here (unlike e.g., arm64e where we'll fail to compile without certain features), so I'm tempted to avoid these shenanigans entirely and let users specify what they want. What do you think? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D82699/new/ https://reviews.llvm.org/D82699 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
