Hi Ryan; Here is my solution to my Pluto gmp library problem (after researching vi Gemini AI):
#!/bin/bash clang -mmacosx-version-min=26.0 -c -fpic ./pluto-gmp.c clang -mmacosx-version-min=26.0 -O3 -fpic -shared -o ../libpluto-gmp.dylib ./pluto-gmp.o -lgmp rm ./pluto-gmp.o Hmmm, yes, but I don't recall seeing a successful build of the target (the library) even though there were only warnings produced. The new version does create the library. But this has, apparently, no direct relation to my Ada linkage problem. Looking forward to having the fix from the bug report you referenced, as it looks like it solves the problem that I am experiencing with my Ada object linking. Thanks, Ken Wolcott On Thu, Sep 25, 2025 at 3:49 PM Ryan Carsten Schmidt < [email protected]> wrote: > On Sep 24, 2025, at 17:15, Kenneth Wolcott wrote: > > > > I'm getting the same kind of error when trying to compile a gcc (C) > binding for GMP to the Pluto language: > > I see no error below, only warnings. > > > > ./compile_gmp_support_for_pluto.bash > > clang: warning: overriding deployment version from '16.0' to '26.0' > [-Woverriding-deployment-version] > > ld: warning: building for macOS-16.0, but linking with dylib > '/opt/local/lib/libgmp.10.dylib' which was built for newer version 26.0 > > > > cat ./compile_gmp_support_for_pluto.bash > > #!/bin/bash > > > > # Steps needed to compile this file (pluto-gmp.c) on Linux using GCC. > > # gcc -c -fpic pluto-gmp.c > > # gcc -O3 -fpic -shared -o pluto-gmp.so pluto-gmp.o -lgmp > > /opt/local/bin/gcc-mp-15 -c -fpic ./pluto-gmp.c > > /opt/local/bin/gcc-mp-15 -O3 -fpic -shared -o ./libpluto-gmp.dylib > ./pluto-gmp.o -lgmp > > > > So maybe something has to change with both my gnat compilation AND my > gcc compilation usage? > > > > How do I verify that I don't have version 16 stuff hanging around > confusing things and that I do have version 26 stuff? > > > > I've successfully compiled the llvm-21 packages (today), so I think I > have the version 26 stuff installed... > > There is no such thing as macOS 16. Apple skipped from macOS 15 (Darwin > 24) directly to macOS 26 (Darwin 25). > > This is a bug in gcc: > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120645 > > gcc computes the macOS deployment target from the Darwin version. The gcc > version you are using was released before it was known that Apple would > change the macOS versioning scheme again. > > A link to the commit fixing the problem is in the bug report. We could > apply it to our gcc ports. > > >
