From owner-freebsd-stable@FreeBSD.ORG  Mon Apr 28 08:45:13 2003
Return-Path: <owner-freebsd-stable@FreeBSD.ORG>
Delivered-To: freebsd-stable@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP
	id BF3AC37B401; Mon, 28 Apr 2003 08:45:13 -0700 (PDT)
Received: from mail.imp.ch (mail.imp.ch [157.161.1.2])
	by mx1.FreeBSD.org (Postfix) with ESMTP
	id 77CCA43FBF; Mon, 28 Apr 2003 08:45:12 -0700 (PDT)
	(envelope-from mb@imp.ch)
Received: from cvs.imp.ch (cvs.imp.ch [157.161.4.9])
	by mail.imp.ch (8.12.6p2/8.12.3) with ESMTP id h3SFj3wv059040;
	Mon, 28 Apr 2003 17:45:09 +0200 (CEST)
	(envelope-from Martin.Blapp@imp.ch)
Date: Mon, 28 Apr 2003 17:45:03 +0200 (CEST)
From: Martin Blapp <mb@imp.ch>
To: Soeren Schmidt <sos@spider.deepcore.dk>
In-Reply-To: <200304230721.h3N7LIoi037105@spider.deepcore.dk>
Message-ID: <20030428173710.G52034@cvs.imp.ch>
References: <200304230721.h3N7LIoi037105@spider.deepcore.dk>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
cc: freebsd-stable@FreeBSD.ORG
cc: patg@imp.ch
cc: sos@FreeBSD.ORG
Subject: Re: Deadlock with ATA disk on FreeBSD 4.8 Stable
X-BeenThere: freebsd-stable@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Production branch of FreeBSD source code
	<freebsd-stable.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>,
	<mailto:freebsd-stable-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-stable>
List-Post: <mailto:freebsd-stable@freebsd.org>
List-Help: <mailto:freebsd-stable-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-stable>,
	<mailto:freebsd-stable-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Mon, 28 Apr 2003 15:45:14 -0000


Hi,

> One thing that could explain this is that the disk stops working in the
> midst of a busmaster DMA transfer, that will effectively lockup the
> system in most cases. I'm suspecting bad HW here, as it makes no sense
> you can access the rest of the disk without problems.. If you can spare
> the data, I'd do a dd if=/dev/zero of=/dev/adN bs=1m and try to get the
> disk to remap all bad sectors...

We removed the disk and used another ata-controller. There it works.

ad1s1e: hard error reading fsbn 438711007 of 219355472-219355631 (ad1s1 bn
438711007; cn 27308 tn 126 sn 49) trying PIO mode
ad1: DMA problem fallback to PIO mode

but why does it stay at PIO mode ? Shouldn't the driver enable DMA again
after we got a block error anyway ?

Here are the settings from the controller where it crashes:

ATA channel 0:
    Master: acd0 <FX4830T/R02E> ATA/ATAPI rev 0
    Slave:       no device present
ATA channel 1:
    Master:  ad1 <ST3120020A/3.30> ATA/ATAPI rev 6
    Slave:       no device present

ad1 runs at DMA33

atapci0: <ServerWorks ROSB4 ATA33 controller> port 0xb800-0xb80f at device 15.1
on pci0
ata0: at 0x1f0 irq 14 on atapci0
ata1: at 0x170 irq 15 on atapci0

Martin