Description

   The mrand48 function generates pseudorandom numbers using the
   linear congruential algorithm and 48-bit integer arithmetic.

   It returns signed long integers uniformly distributed over the
   range of y values such that -231 
   Before you call the mrand48 function, use either srand48,
   seed48, or lcong48 to initialize the random-number generator.
   You must initialize the mrand48 function prior to invoking it,
   because it stores the last 48-bit Xi generated into an internal
   buffer. (Although it is not recommended, constant default
   initializer values are supplied automatically if the drand48,
   lrand48, or mrand48 functions are called without first calling an
   initialization function.)

   The function works by generating a sequence of 48-bit integer
   values, Xi, according to the linear congruential formula:

          Xn+1 = (aXn+c)mod m        n >= 0

   The argument m equals 248, so 48-bit integer arithmetic is

   performed. Unless you invoke the lcong48 function, the multiplier
   value a and the addend value c are:

         a = 5DEECE66D16 = 2736731631558

         c = B16 = 138

   The values returned by the mrand48 function is computed by
   first generating the next 48-bit Xi in the sequence. Then the
   appropriate bits, according to the type of returned data item,
   are copied from the high-order (most significant) bits of Xi and
   transformed into the returned value.

   See also drand48, lrand48, lcong48, seed48, and srand48.