From owner-freebsd-current  Thu Jul 20 11:00:18 1995
Return-Path: current-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.11/8.6.6) id LAA17174
          for current-outgoing; Thu, 20 Jul 1995 11:00:18 -0700
Received: from cs.weber.edu (cs.weber.edu [137.190.16.16])
          by freefall.cdrom.com (8.6.11/8.6.6) with SMTP id LAA17168
          ; Thu, 20 Jul 1995 11:00:16 -0700
Received: by cs.weber.edu (4.1/SMI-4.1.1)
	id AA02515; Thu, 20 Jul 95 11:52:44 MDT
From: terry@cs.weber.edu (Terry Lambert)
Message-Id: <9507201752.AA02515@cs.weber.edu>
Subject: Re: Hmmmm!  New error encountered with cpio while building root.flp
To: bde@zeta.org.au (Bruce Evans)
Date: Thu, 20 Jul 95 11:52:43 MDT
Cc: bde@zeta.org.au, phk@freefall.cdrom.com, current@freefall.cdrom.com,
        jkh@time.cdrom.com
In-Reply-To: <199507201102.VAA22942@godzilla.zeta.org.au> from "Bruce Evans" at Jul 20, 95 09:02:31 pm
X-Mailer: ELM [version 2.4dev PL52]
Sender: current-owner@FreeBSD.org
Precedence: bulk

> Now I can think of some more places: for `cpio -i', 32-bit rdevs in the
> input may be silently truncated if ints or rdevs in the target system
> are less than 32 bits.  I think there are similar bugs in nfs.

The "bug" is that device file systems should be accessed locally through
devfs only for BSD-to-BSD and that device nodes are an antiquated concept
that should be present only fodiskless support of legacy systems.

If there's going to be a hatchet-job on minor numbers, then it ought to
be done all the way instead of half-heartedly.


					Terry Lambert
					terry@cs.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.