http://bugzilla.opensuse.org/show_bug.cgi?id=1127564 http://bugzilla.opensuse.org/show_bug.cgi?id=1127564#c3 Richard Biener <rguenther@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |IN_PROGRESS See Also| |https://gcc.gnu.org/bugzill | |a/show_bug.cgi?id=89573 --- Comment #3 from Richard Biener <rguenther@suse.com> --- Hmm, this is a strange case of excess-precision and resulting differences (x87 doubles have 80bit internal precision). Usually the solution is -fexcess-precision=standard but somehow that doesn't work for the multiplication-and-conversion-to-integer sequence. The "older" -ffloat-store workaround works fine though. Filed upstream. -- You are receiving this mail because: You are on the CC list for the bug.