>* Does the arm software floating point emulation use IEEE-754?
>
>    In particular, I need to implement a function that turns a double number
>    into its IEEE 754 floating-point "double format" bit layout, where
>    Bit 63 represents the sign of the floating-point number.  Bits 62-52 
>    represent the exponent.  Bits 51-0 represent the significand (sometimes 
>    called the mantissa) of the floating-point number.

Almost.  On the ARM the layout of double precision numbers is strange; the two 
words are the opposite way around to what you'd expect.  I think this is an 
artifact from using a big-endian floating point unit on a little-endian 
processor.

Actually, if you use soft-float emulation there's no real need to stick with 
this format.

p.