Walter Bright:
> The reason the optimizer doesn't replace floating divides with floating 
> multiplies of the reciprocal is that one gets slightly different answers 
> (due to roundoff).

Right. Because FP numbers are just approximations of the Real field. GCC has 
command line arguments (like -ffast-math -funsafe-math-optimizations and 
others) that force similar unsafe optimizations.

Bye,
bearophile

Reply via email to