https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105570
--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> --- Oops, I was just testing the backports on x86_64-linux. The fix itself is generic, and thus IMHO worth the backport, so I think the options are kill the testcase on the branch, or just stop using -mcpu=power9 there.
