https://github.com/statham-arm commented:
I've done as much of the review as I can. The syntax in the `REQUIRES` lines looks sensible to me, and I see why _most_ of these tests would depend on clang being configured to target its host Darwin platform by default: * `apple-arm64-arch.c`, `compilation_database_multiarch.c`, `darwin-ld-platform-version-macos-requires-darwin.c` and `xros-driver-requires-darwin-host.c` specifically check for `apple` or `macos` in the compiler output without having provided one in the input command line, so it makes sense that they wouldn't behave like that if clang's host was Darwin but its default target was something else. I'm less sure I understand `mtargetos-darwin.c`, because I don't know what `-mtargetos=...` means. I guess that's supposed to be a modification to the _detail_ of the OS component of the target triple, and doesn't have the side effect of setting it to Darwin in the first place? And I know I don't understand what `pch-from-libclang.c` has to do with Darwin host _or_ target, but the FIXME in the comment suggests that nobody else knows that either, so perhaps that doesn't block the commit. Can you confirm my understanding of `-mtargetos`? https://github.com/llvm/llvm-project/pull/130138 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
