From owner-freebsd-questions  Wed Nov 15 02:24:24 1995
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id CAA15486
          for questions-outgoing; Wed, 15 Nov 1995 02:24:24 -0800
Received: from rf900.physics.usyd.edu.au (rf900.physics.usyd.edu.au [129.78.129.109])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id CAA15386
          ; Wed, 15 Nov 1995 02:22:59 -0800
Received: (from dawes@localhost) by rf900.physics.usyd.edu.au (8.6.11/8.6.9) id VAA29801; Wed, 15 Nov 1995 21:21:24 +1100
From: David Dawes <dawes@rf900.physics.usyd.edu.au>
Message-Id: <199511151021.VAA29801@rf900.physics.usyd.edu.au>
Subject: Re: FreeBSD 2.0.5R hanging inexplicably...
To: bde@zeta.org.au (Bruce Evans)
Date: Wed, 15 Nov 1995 21:21:23 +1100 (EST)
Cc: darrylo@hpnmhjw.sr.hp.com, freebsd-hackers@freebsd.org,
        freebsd-questions@freebsd.org
In-Reply-To: <199511150821.TAA25998@godzilla.zeta.org.au> from "Bruce Evans" at Nov 15, 95 07:21:08 pm
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 1802      
Sender: owner-questions@freebsd.org
Precedence: bulk

>>> > The ET4000 (and particularly the W32 variants) XFree server is _very_fragil
>>> e_.
>>> > 
>>> > The cards themselves are no great shakes, and the behaviour of the server
>>> > leads me to _strongly_ discourage people from using them. (Especially given
>>> > how cheap S3-based cards are these days).
>>> 
>>> The et4000/w32p server has been rock solid under Linux for over a year...
>>> just a data point...
>
>>     It would be interesting to see if the people who are having
>>problems with ET4000/W32p-based video cards are using syscons or pcvt.
>>There have been a few posts recently regarding the "better robustness"
>>of pcvt (vs syscons), and I just recently found a "bug" in syscons that
>>was causing all my syscons fonts to be corrupted (after using
>>X-Windows).
>
>Don't blame syscons or pcvt for bugs in the X server.  The W32 server
>apparently does incomplete initialization so it sometimes fails to switch
>out of modes that are adequate for syscons and pcvt.  I've reported the
>following problems for a W32i:
>
>XF3.1.2: W32 server always crashes when starting up time trying to
>         access through a null pointer to the accelerator registers.
>XF3.1.1: W32 always hangs waiting for bit 1 in the Accelerator Status
>         Register when switching back to X from syscons console.
>         It sometimes hangs in the same place when starting up.

Yes, we're aware of these two problems.  The first has been fixed.  The
second has been worked around for the time being.  Although I'm told it
didn't show up with older Linux versions, it seems to show up with 1.3.x
Linux kernels.  The W32 server is probably the flakiest of the XFree86
servers.  It will work fine on some hardware, but there is a lot of
hardware where it doesn't.  Some work is being done on this.

David