https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88630
--- Comment #8 from Oleg Endo <olegendo at gcc dot gnu.org> --- Created attachment 46987 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=46987&action=edit Trial patch The ST-40 manual describes the fneg and fabs instructions separately for single and double precision. I take it as a hint that they are indeed implemented differently than on Hitachi/Renesas cores. The attached patch is an experiment for the fneg instruction. Can you try it out please and see if it fixes the issue on your hardware?
