rorth wrote: When comparing the `clang` invocations between Solaris/amd64 (test `PASS`es) and Solaris/sparcv9 (test `FAIL`s), I find this crucial difference: ``` -"-funwind-tables=2" -"-target-cpu" -"x86-64" -"-tune-cpu" -"generic" +"-mfloat-abi" +"hard" ``` Looking closer, `-target-cpu` is only printed in `clang/lib/Driver/ToolChains/Clang.cpp` (`Clang::RenderTargetOptions`) iff `!CPU.empty()`. However, in `clang/lib/Driver/ToolChains/Arch/Sparc.cpp` (`sparc::getSparcTargetCPU`) returns the empty string for a default SPARC V9 compilation.
So expecting `-target-cpu` to always be present is a fallacy if not specifying a specific target. If doing so (and this should be some `x86_64` target obviously), the tests needs to ascertain that the `clang` under test is built with `x86_64` support using `REQUIRES: x86-registered-target`. https://github.com/llvm/llvm-project/pull/125646 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
