Meinersbur wrote: >From the summary?
> This has created a nowhere to go situation, the only remaining option is to > make the -stdlib flag visible to flang and silently ignored. Can you clarify? I think it is not ignored, otherwise the test would fail since it requires linking to the `<iostream>` implementation. It must add `-l(std)c++` to the linker command line. > > (and there isn't even a warning that the option is unused) > > @pawosm-arm Wasn't this ever an issue? Should those to be consistent? > > Interesting, clang would do the same as flang-new patched with my patch. > Maybe addition of the warning would be a good idea, if it isn't an overkill. AFAICS your patch only makes the `-stdlib` recognized by Flang. `lc++` being added must have been inherited from the C++ driver, but seemingly missing from the C driver path. A lot of flags are purposefully by the driver because build systems typically pass the same flags to every build phase which would otherwise cause errors ([example](https://github.com/llvm/llvm-project/issues/88173)). A warning usually only annoys people that compile with `-werror`. > what I've noticed is that using -DLLVM_ENABLE_LIBCXX=True CMake flag has > unexpected effect when it comes to C++/Fortran mixed code. What effects? https://github.com/llvm/llvm-project/pull/110598 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
