Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!samsung!munnari.oz.au!ariel!ucsvc.ucs.unimelb.edu.au!lu!cchd
From: CCHD@lure.latrobe.edu.au (Huw Davies - La Trobe University Computer Centre)
Newsgroups: comp.unix.aux
Subject: Re: A/UX concerns
Message-ID: <5103@lure.latrobe.edu.au>
Date: 26 Feb 91 16:13:00 GMT
References: <1991Feb22.175718.6395@agate.berkeley.edu> <49580@apple.Apple.COM> <5097@lure.latrobe.edu.au> <49611@apple.Apple.COM>
Organization: VAX Cluster, Computer Centre, La Trobe University
Lines: 27

In article <49611@apple.Apple.COM>, ksand@Apple.COM (Kent Sandvik) writes:
> 
> Someone correct me if I'm wrong - I'm no DECStation expert - but
> wasn't there some fuzz with byte ordering on the DECStation that
> separates the file system design from the MIPS implementation. I'll
> remember reading something about LSB byte ordering switching so the
> DECStation would be compatible with the rest of the VAX systems.

Yes - the DECstation is little-endian (like a VAX!) whilst the MIPS
systems are configured big-endian (like almost everything else :-).

I have just completed testing an IBM RS/6000 system with NFS, using
the (patented :-) test of compiling and attempting to run the xlock 
application. Well, it failed too. The score so far:

A/UX to Mips 120/5 (Risc/OS 4.5.1) OK
        DECstation-2100 (Ultrix V4.0) No
        IBM RS/6000 (AIX 3.1) No

Given that the DECstation and the IBM are different-endian, I assume
(hope :-) that endianism is not the problem. I wonder if there are
subtle problems with the A/UX implementation?

I am _not_ having a go at Apple nor the A/UX team. In my opinion they
have done a much better job at their first port of Unix than other
companies I could name.... I would just like to see A/UX get better
and better.