Subject: Re: Booting question
To: Scott Lipcon <slipcon@hops.cs.jhu.edu>
From: Bill Studenmund <wrstuden@loki.stanford.edu>
List: port-mac68k
Date: 10/03/1996 13:13:33
> 
> On Thu, 3 Oct 1996 briggs@puma.macbsd.com wrote:
> 
> > I've given serious thought to a native BSD disklabel, but I think it
> > only makes sense when you're formatting the whole disk.
> 
> This make sense.  I'm curious though if this would allow some sort of 
> "native" boot, instead of having to go through the MacOS first.  Why 
> isn't this possible now?

Actually, having a BSD disklabel and a direct boot w/o MacOS are
two different questions. Someone's working on NetBSD bootblocks,
but for them to work, the Mac booter has to see a MacOS partition
table (or at least enough of one to be happy about the boot blocks).

Also, as I understand it, the booter querries MacOS for a lot of info
about the machine we're running on. I don't think we know how to find all
that info on our own. For example, AFAIK we use the Gestalt manager for
a number of things, and it's not in ROM on early machines, so we'd
have to duplicate a lot of fact-finding.

Another reason to boot MacOS first is that A/UX did it. True, just because
your friends jump off a cliff doesn't mean you have to too. But it
seems to work well. Also, given that there are large functionality gaps
in the mac68k port (mounting MacOS directories, Color X, SCSI on
Quadras, ADB on many machines, Ethernet on Quadras, ...), efforts
seem to be focused elsewhere.

Take care,

Bill