NAME IEEEDPAbs -- compute absolute value of IEEE double precision argument SYNOPSIS x = IEEEDPAbs( y ); d0/d1 d0/d1 double x,y; FUNCTION Take the absolute value of argument y and return it to caller. INPUTS y -- IEEE double precision floating point value RESULT x -- IEEE double precision floating point value BUGS SEE ALSO