ldexp()                  Convert Mantissa and Exponent to Floating Point
 
 #include   <math.h>
 
 double     ldexp(m,exp);
 double     m;                           Floating-point value
 int        exp;                         Integer exponent
 
    ldexp() returns the result of the calculation:
 
                        'm' * 2^ 'exp'
 
              where ^ means "raised to the power of".
 
    Returns:    'm' * 2^ 'exp'.  On overflow, 'errno' (defined in
                <stdlib.h>) is set to ERANGE (defined in <math.h>), and
                the value HUGE (+ or -, depending on the sign of 'm') is
                returned. (HUGE is defined in math.h.)  On underflow 0.0
                is returned ('errno' is not set on underflow).
 
      Notes:    frexp() breaks a floating-point value into mantissa and
                exponent.
 
                ldexp() does not call matherr() on error.
 
  -------------------------------- Example ---------------------------------
 
    The following statements calculate the values of 1.0 * 2^24, 6.5 *
    2^-10, -2.0 * 2^0, and 9.9e99 * 2^999 (the last calculation
    produces an OVERFLOW error):
 
         #include <math.h>      /* for ldexp() and ERANGE */
         #include <stdio.h>     /* for printf() */
         #include <stdlib.h>    /* for errno */
 
         main()
         {
             double a, b, c, d;
 
             a = ldexp(1.0, 24);     /* a = 16777216.0 */
             b = ldexp(6.5, -10);    /* b = 0.0063477 */
             c = ldexp(-2.0, 0);     /* c = -2.0 */
             errno = 0;
             d = ldexp(9.9e99, 999); /* d = HUGE, causes an OVERFLOW error */
             if (errno == ERANGE)
                 printf("OVERFLOW error in ldexp()\n");
         }

Seealso:


 frexp() 
 modf()