Bytes order and words order

[Gabriel Dos Reis]

Richard Earnshaw <rearnsha@arm.com> writes:

| > OK, thanks.  I was actually thinking of the word-order of
| > floating-point numbers not that of integral types.  I think I should
| > name it  __TARGET_FLOAT_WORDS_ORDER__ ?  
| > Is it reasonable to assume that the byte order of integral and
| > floating point coincide? 
| 
| I'm not aware of any machines where the byte ordering doesn't match that 
| for integers -- at least in the sense that the 'sign bit' of a floating 
| point number lives in the 'most significant' byte of a word, etc.  

Thanks.

| But I'm not sure if IEEE-754 actually requires this.

No, it doesn't require that

| R.
| 
| BTW, if you are picking apart a float like this, does it mean that you are 
| assuming IEEE-754 format numbers?

No, I'm not assuming IEEE-754 -- even though I'm testing on such a
plateform.  The idea is to provide support for behaviour mandated by
the C++ library (numeric_limits<>), preliminary support for LIA and
IEEE-754 if possible (compatibility with C99).  

-- Gaby