https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107021
--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> --- (In reply to Martin Liška from comment #5) > > > > so I'd call that a testsuite issue. > > > > Yep, how should I fix the verifier? Just skip it if the expected value is > infinite? No, don't build it with -fno-finite-math-only?