Re: SuperH port

[Ulrich Mueller]

>>>>> On Sat, 18 Oct 2008, Andreas Schwab wrote:

>> (To add some confusion, endian.h then also defines _all_ four of
>> __LITTLE_ENDIAN, LITTLE_ENDIAN, __BIG_ENDIAN, and BIG_ENDIAN,
>> regardless of the machine's byte sex. Go figure.)

> There is nothing wrong with that.  They define the valid values for
> BYTE_ORDER.

I just noticed that m/xtensa.h is testing for __LITTLE_ENDIAN to
determine byte order. Either this is not right, or it's a
compiler-defined symbol there ...

Ulrich