From owner-freebsd-hackers  Sat Nov  1 20:46:46 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id UAA27978
          for hackers-outgoing; Sat, 1 Nov 1997 20:46:46 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from metronet.com (pgilley@fohnix.metronet.com [192.245.137.2])
          by hub.freebsd.org (8.8.7/8.8.7) with SMTP id UAA27969
          for <hackers@FreeBSD.ORG>; Sat, 1 Nov 1997 20:46:40 -0800 (PST)
          (envelope-from pgilley@metronet.com)
Received: from localhost by metronet.com with SMTP id AA25031
  (5.67a/IDA1.5hp for <hackers@FreeBSD.ORG>); Sat, 1 Nov 1997 22:47:20 -0600
Date: Sat, 1 Nov 1997 22:47:16 -0600 (CST)
From: Phil Gilley <pgilley@metronet.com>
To: David Greenman <dg@root.com>
Cc: hackers@FreeBSD.ORG, Doug White <dwhite@resnet.uoregon.edu>
Subject: Re: Problem with ed driver in 2.2.5 
In-Reply-To: <199711010758.XAA28986@implode.root.com>
Message-Id: <Pine.HPP.3.95.971101224336.24297A-100000@fohnix.metronet.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Fri, 31 Oct 1997, David Greenman wrote:

>    Hmmm. Not sure how to deal with this. The reason why 0WS was turned on
> was to 'fix' a serious ISA shared-memory performance problem that a lot of
> newer motherboards have - the 8K RAM cards are almost useless without it.
> It turned out to cause problems with reading the EEPROM on the '790 based
> cards, so I killed the option for those prior to the 2.2.5 release...I'm
> surprised to hear that you're having troubles with a '690 based board.
> It shouldn't be a problem on most systems - this might indicate that your
> ISA bus speed is set too fast.

I couldn't figure out how to change the ISA bus speed on that PC so I
tried a different 486/66 and SMC WD8013EPC.  This other machine has a
very different motherboard and the combo version of the ethernet card.
With this hardware the default 2.2.5 ed driver usually causes a "Fatal
trap 12: page fault while in kernel mode" at random addresses on FTP
transfers.  When it does work I get transfer rates around 40 KB/s.
If I disable the 0WS option in the ed driver things work fine.  Am I
the only person experiencing this or am I just the only person still
running old hardware?

Phil Gilley
pgilley@metronet.com