https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107490
--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> --- Created attachment 53811 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=53811&action=edit testcase, -Os -fno-trapping-math I changed the testcase so -fsingle-precision-constant is no longer needed; just add the f to the end of the constants.
