MaskRay wrote: > I'm not so sure that it is that simple. `-idirafter` and `-isystem-after` > have slightly different semantics. They impact ordering and warning > suppression. If there are alternative flags that can be used to control > those, that might be reasonable.
Do we know the users of `*-windows-itanium` triples? Right now `-isystem-after` unimplemented for other targets causes confusion due to [clang.llvm.org/docs/ClangCommandLineReference.html#cmdoption-clang-isystem-after-directory](https://clang.llvm.org/docs/ClangCommandLineReference.html#cmdoption-clang-isystem-after-directory). Does the order matter? In addition, https://sourcegraph.com/search?q=context:global+isystem-after+-f:clang/&patternType=keyword&sm=0&filters=%5B%5B%22repo%22%2C%22github.com%2Fmit-ll%2Fgaleed%22%2C%22repo%3A%5Egithub%5C%5C.com%2Fmit-ll%2Fgaleed%24%22%5D%5D says there is no -isystem-dirafter user. https://github.com/llvm/llvm-project/pull/84121 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
