From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 2 Jan 2002 19:14:44 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 2 Jan 2002 19:14:37 -0500 Received: from relais.videotron.ca ([24.201.245.36]:18379 "EHLO VL-MS-MR001.sc1.videotron.ca") by vger.kernel.org with ESMTP id ; Wed, 2 Jan 2002 19:13:38 -0500 Message-ID: <3C33A22F.40906@videotron.ca> Date: Wed, 02 Jan 2002 19:13:35 -0500 From: Roger Leblanc Organization: General DataComm User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.4) Gecko/20010914 X-Accept-Language: en-us MIME-Version: 1.0 To: Greg KH CC: linux-kernel@vger.kernel.org Subject: Re: Deadlock in kernel on USB shutdown Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org ----- Original Message ----- From: "Greg KH" To: "Roger Leblanc" Cc: Sent: Wednesday, January 02, 2002 9:40 AM Subject: Re: Deadlock in kernel on USB shutdown >> On Tue, Jan 01, 2002 at 04:48:41PM -0500, Roger Leblanc wrote: > >>> > Hi, >>> > >>> > I just compiled version 2.4.17 of the Linux kernel for my Pentium III. >>> > It is compiled with modular USB support so I can run my USB scanner (an >>> > Epson Perfection 1200U). >>> > >>> > The scanner works fine but the system freeses when I shut it down. I >>> > investigated a bit and found that in the file: >>> > /drivers/usb/usb.c >>> > in function: >>> > usb_disconnect(struct usb_device **pdev) >>> > >>> > there is a call to function: >>> > usbdevfs_remove_device(dev) >>> > at line 2423. >>> > >>> > That is the exact point where it freeses. If I comment out that line, >>> > everything goes fine. I know! This is not the proper way to fix it! But >>> > at least, it fixes my problem. Since I'm not a kernel expert, I will >>> > leave it to you to find the right way to fix it. >> >> >> Does the system lock up when you unload the usbcore module by hand >> without shutting the system down? >> >> Are your init scripts unmounting the usbdevfs filesystem properly before >> trying to unload the usbcore module? > It doesn't get that far. The first thing my init script (or Mandrake 8.1 script) does at shutdown is to run modprobe -r on modules usb-ohci, usb-uhci and uhci. The system freeses when it gets to usb-uhci. It does it also if I run these commands on the command line. Thanks Roger