This escape occurs when a floating point error is detected using the software floating point routines.
These routines come into play when numbers of type single or double are manipulated with conventional
infix notation. If you are using the math coprocessor this floating point escape is not used.