From owner-freebsd-current  Wed Mar 27 22:15:37 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id WAA03177
          for current-outgoing; Wed, 27 Mar 1996 22:15:37 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id WAA03167
          for <current@freebsd.org>; Wed, 27 Mar 1996 22:15:26 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id RAA14635; Thu, 28 Mar 1996 17:14:14 +1100
Date: Thu, 28 Mar 1996 17:14:14 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199603280614.RAA14635@godzilla.zeta.org.au>
To: bde@zeta.org.au, swallace@ece.uci.edu
Subject: Re: 2.2-960323-SNAP: ibcs2 panic
Cc: current@freebsd.org
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>As I explained when I wrote the spx "support" it is truely a hack.
>Look at the code:
>...
>It doesn't matter what device it opens, just as long as it doesn't exist.

You get to implement /dev/nonexistent or /dev/spx_hack :-).

>Then it returns ENXIO so it checks the path.  This is for speed so
>not all opens check the path.  Since the old socksys support was deprecated,
>the old spx device that no longer exists is a good choice for this hack!

Major 41 is still reserved, so it is a good choice for /dev/nonexistent.

Bruce