-- Michael D. Berger m.d.ber...@ieee.org http://www.rosemike.net/
> -----Original Message----- > From: gcc-ow...@gcc.gnu.org [mailto:gcc-ow...@gcc.gnu.org] On > Behalf Of Tim Prince > Sent: Sunday, May 08, 2011 11:38 > To: gcc@gcc.gnu.org > Subject: Re: numerical results differ after irrelevant code change [...] > > How does the extra precision lead to the variable result? > > Also, is there a way to prevent it? It is a pain in > regression testing. > > > If you don't need to support CPUs over 10 years old, consider > -march=pentium4 -mfpmath=sse or use the 64-bit OS and gcc. > Note the resemblance of your quoted differences to > DBL_EPSILON from <float.h>. That's 1 ULP relative to 1.0. I > have a hard time imagining the nature of real applications > which don't need to tolerate differences of 1 ULP. > I made the changes you suggest. While I was previously getting -1.16e-16 and -1.03e-16 depending presence of an "extra" class member, I now get 1.11e-16 (sic not -). But it is now independent of the extra class member, which is excellent. Hopefully, things will stay that way. Thanks, Mike.
