NAME
        IEEESPFix -- convert IEEE single float to integer

   SYNOPSIS
        x   = IEEESPFix(  y  );
        d0               d0

        long    x;
        float   y;

   FUNCTION
        Convert IEEE single precision argument to a 32 bit signed integer
        and return result.

   INPUTS
        y -- IEEE single precision floating point value

   RESULT
        if no overflow occured then return
                x -- 32 bit signed integer
        if overflow return largest +- integer
                For round to zero

   BUGS

   SEE ALSO
        ieeespflt()