[Sc-devel] -0 bug

[Ryan Brown]

> as Brian W. has mentioned before, there is a bug where:
>
> 0.0 * -1.0
>
> returns -0.

-0.0 is part of the floating point spec and should be allowed in SC.  
The original issue that was brought up with (60.0 % 12) returning -0.0  
on PPC and 0.0 on i386 is a bug though. I spent a while trying to  
figure out why this is...and I have no idea. Some kind of bizarre  
floating point error. Just putting this line before sc_mod returns  
will fix the problem on PPC:

fprintf(stderr, "sc_mod dbl: %f", hi*floor(in/hi));

Using fmod rather than sc_mod also fixes it, and this is what we  
should probably do.

Ryan