From owner-freebsd-hackers  Tue Aug 22 08:36:55 1995
Return-Path: hackers-owner
Received: (from majordom@localhost)
          by freefall.FreeBSD.org (8.6.11/8.6.6) id IAA01428
          for hackers-outgoing; Tue, 22 Aug 1995 08:36:55 -0700
Received: from ref.tfs.com (ref.tfs.com [140.145.254.251])
          by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id IAA01422
          for <hackers@freebsd.org>; Tue, 22 Aug 1995 08:36:53 -0700
Received: (from julian@localhost) by ref.tfs.com (8.6.11/8.6.9) id IAA27053; Tue, 22 Aug 1995 08:35:37 -0700
From: Julian Elischer <julian@ref.tfs.com>
Message-Id: <199508221535.IAA27053@ref.tfs.com>
Subject: Re: Using space in a DOS filesystem
To: bde@zeta.org.au (Bruce Evans)
Date: Tue, 22 Aug 1995 08:35:36 -0700 (PDT)
Cc: bde@zeta.org.au, msmith@atrad.adelaide.edu.au, hackers@freebsd.org
In-Reply-To: <199508221526.BAA13856@godzilla.zeta.org.au> from "Bruce Evans" at Aug 23, 95 01:26:05 am
X-Mailer: ELM [version 2.4 PL24]
Content-Type: text
Content-Length: 1328      
Sender: hackers-owner@freebsd.org
Precedence: bulk

> 
> >> >But if the slab _is_ the boot device, how do we read the tools from it?
> >> 
> >> With difficulty.  Don't boot from it.
> 
> >That more or less rules that approach out completely.  We're trying to
> >do an ersatz partition that can do everything a "real" partition does.
> >That's going to have to include booting.
> 
> The standard boot loader isn't going to work on it unless it is partition
> (within another partition :-().

for a demo that doesn't matter because you would be booting from DOS 
anyhow....
just have a kernel that uses the dosfs code to find it's partition, instead of
using the disklable code to do the same thing.

> 
> >> It would be relatively easy to check the partition once you have located
> >> it.  You might be able to boot from the DOS file system, run some
> 
> >I can understand that.  How would you feel about being able to recognise
> >one of these ersatz partitions if the info were passed in to the kernel 
> >at bootstrap time?
> 
> The mount of root would probably fail if the partition was invalid.  I
> guess you could put the partition boundaries in the data for the program
> replaces the standard boot loader.  How are you going to select where
> to boot from?  Standard boot managers should be able to handle the
> partition within a partition method (:-().
> 
> Bruce
>