Hi, This bug report is specially targeted at Georg Archer. It contains extensive debugging information.
Some time ago (last november) we had a thread about usb-uhci and usb-storage hanging my machine, while uhci and usb-storage worked. In his last reply Matthew advised me to contact the maintainers of uhci and usb-uhci, because the log I sent him contained some errors from the HCD. (see the attached message) It took me a while to find the time to produce new test data. Sorry for that. Here it is. This mail message contains an attachment with a gzipped tar file, containing information and logging regarding my tests. The testing comprises the following steps: 1. hook up my Ricoh RDC5300 digital still camera to the USB bus 2. turn the camera on 3. mount the internal and external memory banks (vfat filesystems) 4. copy a JPEG image from the camera to the Linux host 5. umount the memory banks 6. turn the camera off I repeated the steps using the USB-UHCI module and the UHCI module. With the UHCI module all steps in the test succeed. (it works) With the USB-UHCI module the first mount at step 3 hangs indefinitely. This is different from my earlyer tests back in november, where the whole machine would hang solid. I used the pristine kernel source release 2.4.18-rc4 from www.kernel.org with the rmap patch from rik van Riel applied. The uname -a output is: Linux roach 2.4.18-rc4-rmap12f #13 Sun Mar 17 21:52:41 GMT+1 2002 i686 unknown The kernel was compiled with the option CONFIG_USB_DEBUG, and I had the syslogd daemon log everything (*.*). The SCSI-driver was intentionally compiled with the option "probe all LUN's", because the camera demands this (it exports the internal and external memory banks as two logical units on the same SCSI-device). The Linux machine is a Dell Inspiron 8100 laptop. The tar file contains the following files: uhci-commandlog.txt - shell commands I issued with the uhci module uhci-debug-messages.txt - debug messages with the uhci module uhci-lsmod.txt - lsmod output with the uhci module usb-uhci-debug-messages.txt - debug messages with the usb-uhci module usb-uhci-devices.txt - output of the cmd 'cat /proc/bus/usb/devices' usb-uhci-lsmod.txt - lsmod output with the usb-uhci module usb-uhci-scsi-usb-storage.txt - output of the command 'cat /proc/scsi/usb-storage-0/0' usb-uhci-scsi.txt - output of the command 'cat /proc/scsi/scsi' Maybe you could take a look and find out why the mount command is hanging with the usb-uhci module? Regards, Toon. -- /"\ | Windows XP: \ / ASCII RIBBON CAMPAIGN | "Sorry Dave... X AGAINST HTML MAIL | I'm afraid I can't do that." / \
>From [EMAIL PROTECTED] Mon Nov 12 00:35:23 2001
Return-Path: <[EMAIL PROTECTED]>
Received: (from uucp@localhost)
by vdpas.hobby.nl (8.11.6/8.11.2) with UUCP id fABNYNa03012
for [EMAIL PROTECTED]; Mon, 12 Nov 2001 00:34:23 +0100
Received: from usw-sf-list1.sourceforge.net (usw-sf-fw2.sourceforge.net
[216.136.171.252])
by hgatenl.hobby.nl (8.11.6/8.11.5) with ESMTP id fABNQfI28596
for <[EMAIL PROTECTED]>; Mon, 12 Nov 2001 00:26:42 +0100 (CET)
(envelope-from [EMAIL PROTECTED])
Received: from localhost ([127.0.0.1] helo=usw-sf-list1.sourceforge.net)
by usw-sf-list1.sourceforge.net with esmtp (Exim 3.31-VA-mm2 #1 (Debian))
id 1633li-0006RT-00; Sun, 11 Nov 2001 15:12:02 -0800
Received: from con-64-133-52-190-ria.sprinthome.com ([64.133.52.190]
helo=ziggy.one-eyed-alien.net)
by usw-sf-list1.sourceforge.net with esmtp (Exim 3.31-VA-mm2 #1 (Debian))
id 1633kz-0006A3-00
for <[EMAIL PROTECTED]>; Sun, 11 Nov 2001 15:11:17 -0800
Received: (from mdharm@localhost)
by ziggy.one-eyed-alien.net (8.9.3/8.9.3) id PAA22943;
Sun, 11 Nov 2001 15:11:13 -0800
From: Matthew Dharm <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: Re: [linux-usb-devel] Hang with usb-storage (using usb-uhci in 2.4.15pre1)
Message-ID: <[EMAIL PROTECTED]>
Mail-Followup-To: [EMAIL PROTECTED],
[EMAIL PROTECTED]
References: <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
<[EMAIL PROTECTED]>
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-md5;
protocol="application/pgp-signature"; boundary="MfFXiAuoTsnnDAfZ"
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <[EMAIL PROTECTED]>; from [EMAIL PROTECTED] on Sun,
Nov 11, 2001 at 12:34:59PM +0100
Organization: One Eyed Alien Networks
X-Copyright: (C) 2001 Matthew Dharm, all rights reserved.
Sender: [EMAIL PROTECTED]
Errors-To: [EMAIL PROTECTED]
X-BeenThere: [EMAIL PROTECTED]
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: <mailto:[EMAIL PROTECTED]?subject=help>
List-Post: <mailto:[EMAIL PROTECTED]>
List-Subscribe: <https://lists.sourceforge.net/lists/listinfo/linux-usb-devel>,
<mailto:[EMAIL PROTECTED]?subject=subscribe>
List-Id: <linux-usb-devel.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/linux-usb-devel>,
<mailto:[EMAIL PROTECTED]?subject=unsubscribe>
List-Archive: <http://www.geocrawler.com/redir-sf.php3?list=linux-usb-devel>
X-Original-Date: Sun, 11 Nov 2001 15:11:13 -0800
Date: Sun, 11 Nov 2001 15:11:13 -0800
Status: RO
X-Status: A
Content-Length: 2730
Lines: 79
--MfFXiAuoTsnnDAfZ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
I suggest you get in contact with both the uhci.o and usb-uhci.o folks --
the "working" log you sent me contained some errors from the HCD. Search
for the word "link" and you'll see them, along with some pointers and other
low-level debugging information.
Matt
On Sun, Nov 11, 2001 at 12:34:59PM +0100, [EMAIL PROTECTED] wrote:
> On Sun, Nov 11, 2001 at 11:36:42AM +0100, [EMAIL PROTECTED] wrote:
> >=20
> > [...]
> > Much to my disappointment it took far too long for the cp command
> > to complete. During the cp I could hear the head of the disk drive
> > move up-and-down the platter over a long distance (definitely not
> > a small move-to-the-next-track movement).
> >=20
> > What is happening here?
> > It isn't a problem with de IDE disk (/tmp is located on an IDE disk),
> > because it doesn't happen when I cp a file from a regular filesystem
> > to /tmp. Is something being reset before/after each IO operation?
>=20
> Well, call me an amateur. <blush>
> Forget about the bad-performance-because-of-disk-trashing story.
> I had turned on a lot of debugging, and the syslog daemon was
> configured to sync after every debugging log record.
> This will garantee that every last record hits the disk, but
> it will also trash your disk in the process, 100% garanteed. :-)
>=20
> So it works.
> Throughput: 54044 bytes/second (3 files, 1987072 bytes total.
> Not really great, but much faster than via an asynchronous serial port.
>=20
> Thanks for your assistance.
> If you want me to investigate the problems with USB-UHCI any further,
> just ask.
>=20
> Regards,
> Toon.
> --=20
> /"\ | Windows XP:
> \ / ASCII RIBBON CAMPAIGN | "Sorry Dave...
> X AGAINST HTML MAIL | I'm afraid I can't do that."
> / \
--=20
Matthew Dharm Home: mdharm-usb@one-eyed-alien.=
net=20
Maintainer, Linux USB Mass Storage Driver
It's not that hard. No matter what the problem is, tell the customer=20
to reinstall Windows.
-- Nurse
User Friendly, 3/22/1998
--MfFXiAuoTsnnDAfZ
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE77wWRz64nssGU+ykRAnKbAJ0QuRD6HvxcseVk2yaJRshsW9bLwQCfceES
dGczUNuCA/69uh5raO7yX2g=
=Aeh6
-----END PGP SIGNATURE-----
--MfFXiAuoTsnnDAfZ--
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
usb-uhci-hanging-mount-bug.tar.gz
Description: GNU Zip compressed data
