Found a 64bit double implementation library, but I cannot compile it

[el_supremo]

x -= x;

This will set x to zero, not negate it??
Shouldn't it be:

x = -x;

Pete