Floating-point support not loaded (CV1048)
 
     An attempt was made to access the math processor registers in a
     program that does not use floating-point arithmetic.
 
     Several situations can cause this error:
 
        ■ The math processor registers can only be accessed through the
          floating-point library code. This code is not loaded if the
          program does not perform floating-point calculations.
 
        ■ If the program does not use floating-point instructions, this
          error may occur when attempting to access the math processor
          before any floating-point instructions have been performed.
          The C run-time library includes a floating-point instruction
          near the beginning so that the math processor registers are
          always accessible.
 
        ■ If a floating-point instruction occurs in an assembly-
          language routine before such an instruction occurs in the C
          code that calls the routine, this error occurs.
                                    -♦-