Received: with ECARTIS (v1.0.0; list linux-xfs); Wed, 25 Sep 2002 13:10:30 -0700 (PDT)
Received: from mail.cs.tu-berlin.de (root@mail.cs.tu-berlin.de [130.149.17.13])
	by oss.sgi.com (8.12.5/8.12.5) with SMTP id g8PKAOtG008665
	for <linux-xfs@oss.sgi.com>; Wed, 25 Sep 2002 13:10:25 -0700
Received: from bolero.cs.tu-berlin.de (daemon@bolero.cs.tu-berlin.de [130.149.19.1])
	by mail.cs.tu-berlin.de (8.9.3/8.9.3) with ESMTP id WAA12314;
	Wed, 25 Sep 2002 22:07:32 +0200 (MET DST)
Received: (from doko@localhost)
	by bolero.cs.tu-berlin.de (8.11.6+Sun/8.9.3) id g8PK7WG16852;
	Wed, 25 Sep 2002 22:07:32 +0200 (MEST)
From: Matthias Klose <doko@cs.tu-berlin.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15762.6020.364002.585622@gargle.gargle.HOWL>
Date: Wed, 25 Sep 2002 22:07:32 +0200
To: Chris Wedgwood <cw@f00f.org>
Cc: linux-xfs@oss.sgi.com, Harald Wagener <hwagener@hamburg.fcb.com>
Subject: Re: XFS "freezes" on AMD systems
In-Reply-To: <20020925075844.GA26750@tapu.f00f.org>
References: <15761.27018.659158.237109@gargle.gargle.HOWL>
	<20020925075844.GA26750@tapu.f00f.org>
X-Mailer: VM 7.03 under 21.4 (patch 6) "Common Lisp" XEmacs Lucid
X-archive-position: 581
X-ecartis-version: Ecartis v1.0.0
Sender: linux-xfs-bounce@oss.sgi.com
Errors-to: linux-xfs-bounce@oss.sgi.com
X-original-sender: doko@cs.tu-berlin.de
Precedence: bulk
X-list: linux-xfs

Chris Wedgwood writes:
> On Wed, Sep 25, 2002 at 09:45:14AM +0200, Matthias Klose wrote:
> 
> > When unpacking large tarballs or copying data bigger than 20MB on a
> > XFS file system, the system seems to freeze for about three to five
> > seconds
> 
> [...]
> 
> > but only on Athlon/Duron based systems, both using IDE drives. I
> > don't see this type of freeze on a Celeron based system.
> 
> At a guess...
> 
> What IDE chipset?  I would guess the chipset used on the Duran/Athlon
> systems isn't using DMA for IDE transfer but rather PIO.
> 
> hdparm(8) may help here, or you may beed to recompile the kernel with
> different IDE chipset support[1].

Thanks for the pointer! That fixed my setup.