Jack Howarth <[email protected]> writes: Torbjörn, Are you just using...
./configure Yes. or are you explicitly passing -fstack-check to clang on CFLAGS and CXXFLAGS? I did notice one extra oddity to this bug. When the default stack checking in Xcode 11 clang is used, the t-powm test case fails. If you explicitly pass -fstack-check then it doesn't. Odd. Perhaps you drop the various other options GMP chooses for CFLAGS? These Catalina problems are real, but even with an Apple engineer pointing fingers at us, I am not going to isolate yet another bug with clang. Sorry. Of the many issues I've encountered over the years with clang and GMP, all thus far were in clang. I've used so much time isolating bugs with clang to last a lifetime. Sure, this may be a bug with GMP, but I strongly doubt that. If history teaches us anything, this problem will also be in clang. If somebody isolates the problem and if it is in GMP, I will happily work on resolving the problem as soon as I have the time. -- Torbjörn Please encrypt, key id 0xC8601622 _______________________________________________ gmp-bugs mailing list [email protected] https://gmplib.org/mailman/listinfo/gmp-bugs
