This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Endian[n]ess, language-specific problems and help with egcs.
- To: smith at vaj5184 dot tu dot hac dot com, toon at moene dot indiv dot nluug dot nl
- Subject: Re: Endian[n]ess, language-specific problems and help with egcs.
- From: N8TM at aol dot com
- Date: Tue, 18 Aug 1998 09:01:46 EDT
- Cc: egcs at cygnus dot com
In a message dated 8/18/98 4:54:49 AM Pacific Daylight Time,
smith@vaj5184.tu.hac.com writes:
> SGI have some kind of support for both BE and LE.
>
> I've also seen somewhere that VAX seemed to use neither BE
> ('1234') or LE ('4321'), but instead '2143' or something
> like that. Is that right?
If you can present the situation in a comprehensible fashion, you will have
done a service.
Yes, the MIPS CPU's internally use a byte order like the one of the VAX
architecture. But, SGI Fortran, at least, uses binary data files in big
endian order for compatibility with other Unix architectures. As you point
out, the internal byte order is big-endian for 16 bit groups within floating
point numbers. But are the integers fully little-ended?
In a case like this, the binary files used by a gnu compiler may not match
those used by proprietary compilers.