float to double in Dynamic C?

I need to do computations that require double precision in Dynamic C rather than just floats. Anyone know how I can do that?