https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88074
--- Comment #28 from Jakub Jelinek <jakub at gcc dot gnu.org> --- Created attachment 45676 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=45676&action=edit gcc9-pr88074.patch Incremental patch to tweak norm2 simplification. If e->value.real**2 would overflow or be close to that, scale everything down (this and all further operands by 2**norm2_scale, result by 4**norm2_scale) and at the end up again.
