>     Hi Clarence,
> 
> you wrote in <199602021306.IAA20237@hopf2.math.purdue.edu>:
> > My experiment was on NS 3.3 install and developer cdroms. The results
> > on my sparc running NEtBSD 1.1 ( -current ) was that mount_ffs reported
> > misplaced superblocks  on these.
> 
> Are you sure that this an endian problem? MC680x0 (NeXT stations) and
> SPARC CPUs use AFAIK the same byte order (at least the both use not the
> same byte order as Intel x86 CPUs).

We're veering away from relevance, but here's my $0.02.

>From NeXTAnswers:

"Unfortunately, it's very difficult to support booting a computer from a 
Rock Ridge CD-ROM.  Because the ability to boot is critical in order to 
allow a CD-ROM to function as a backup for system software, Release 3 will 
also support booting from and reading CD-ROMs formatted using the NeXT UNIX 
file system.  The Release 3 CD-ROM product will use this format and will 
allow a computer to boot with a minimal operating system environment in 
order to repair or reinstall the system software on the hard disk.  NeXT 
will be one of the first vendors to offer this advanced CD-ROM booting 
capability."

NeXT filesystems are 4.3 BSD all the way.  The same CD is used to install
both the Intel _and_ Motorola platforms (?!).  However, under NetBSD-Amiga
(big-endian), _and_ NetBSD-x86 (little):

bard@tiamat> mount -o ro -t ffs /dev/cd0a /cdrom
ufs: /dev/cd0a on /cdrom: incorrect super block

I'm pretty sure it's just a funky format that only NEXTSTEP has hooks for.

> Matthias Scheler

David Hopper (bard@tiamat.chaldea.com)
Anthropology Department, Stanford University
NetBSD 1.1 (tiamat.chaldea.com) : NEXTSTEP 3.3 (marduk.chaldea.com)
"Two pale drops of fire guttering in the vast consuming darkness" --V. Price