From owner-freebsd-hackers  Tue Oct  1 22:45:38 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id WAA07316
          for hackers-outgoing; Tue, 1 Oct 1996 22:45:38 -0700 (PDT)
Received: from Pescadero.DSG.Stanford.EDU (Pescadero.DSG.Stanford.EDU [171.64.79.10])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id WAA07311
          for <hackers@freebsd.org>; Tue, 1 Oct 1996 22:45:36 -0700 (PDT)
Received: from localhost (Pescadero.DSG.Stanford.EDU [171.64.79.10]) by Pescadero.DSG.Stanford.EDU (8.7.4/8.6.9) with ESMTP id WAA24656; Tue, 1 Oct 1996 22:44:44 -0700
Message-Id: <199610020544.WAA24656@Pescadero.DSG.Stanford.EDU>
To: Chris G Demetriou <Chris_G_Demetriou@ux2.sp.cs.cmu.edu>
cc: "Kevin P. Neal" <kpneal@pobox.com>,
        Poul-Henning Kamp <phk@critter.tfs.com>,
        Jason Thorpe <thorpej@nas.nasa.gov>,
        "Justin T. Gibbs" <gibbs@freefall.freebsd.org>,
        James Graham <greywolf@siva.captech.com>, hackers@freebsd.org,
        tech-kern@netbsd.org, jonathan@DSG.Stanford.EDU
Subject: Re: VPS mailing list, BSD interest? 
In-reply-to: Your message of "Wed, 02 Oct 1996 01:14:55 EDT."
             <122.844233295@ux2.sp.cs.cmu.edu> 
Date: Tue, 01 Oct 1996 22:44:43 -0700
From: Jonathan Stone <jonathan@DSG.Stanford.EDU>
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

In message <122.844233295@ux2.sp.cs.cmu.edu>Chris G Demetriou writes
>
>Sane architectures typically read the first N blocks off the disk,
>or read something in a well-known format from the first block of the
>disk that indicates what other blocks to read.


>Yes, on a sun3 or DECstation, you're going to lose, but that's not
>that big a deal, at least to me.

Wrong, at least on DECstations.  They're ``sane'' by your description;
very like the TC alphas, in fact.

The lossage on DECstations is that there's no good way to set
anything in the PROM that indicates  to the kernel which
partition to use as root. 

We could always invent  a syntax for encoding root partitions
for  kernels that prompt for a root partition. It's getting
autoboot to DTTR, or specifying it to the PROM instead
of a generic-style kernel, that's problematic.

(The other lossage is that I haven't discovered a realiable, 
 sufficiently low-pain installation procedure, but that's another story).