On Thu, Mar 14, 2002 at 12:59:31PM -0500, ozan s. yigit wrote: > in a related tangential note, i recently found (out of sheer irritation) > in less than an hour that several (including the latest) versions of GCC > -O and -O2 failed the paranoia test in different ways, to wit: > > gcc -o paranoia paranoia.c > > The number of DEFECTs discovered = 1. > The number of FLAWs discovered = 1. > > gcc -O2 -o paranoia paranoia.c > > The number of FAILUREs encountered = 4. > The number of SERIOUS DEFECTs discovered = 4. > The number of DEFECTs discovered = 2. > The number of FLAWs discovered = 2. > > i assume everyone knows about kahan and paranoia. if not see netlib. >
Add the -ffloat-store flag to your compilation flags (or add -msoft-float). No failures, defects nor flaws have been discovered. Rounding appears to conform to the proposed IEEE standard P754, except for possibly Double Rounding during Gradual Underflow. The arithmetic diagnosed appears to be Excellent! END OF TEST. -- Steve To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
