From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262965AbTJEDyI (ORCPT ); Sat, 4 Oct 2003 23:54:08 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262968AbTJEDyI (ORCPT ); Sat, 4 Oct 2003 23:54:08 -0400 Received: from mail.kroah.org ([65.200.24.183]:14766 "EHLO perch.kroah.org") by vger.kernel.org with ESMTP id S262965AbTJEDyE (ORCPT ); Sat, 4 Oct 2003 23:54:04 -0400 Date: Sat, 4 Oct 2003 20:48:17 -0700 From: Greg KH To: reg@dwf.com Cc: linux-kernel@vger.kernel.org, reg@orion.dwf.com Subject: Re: trying to get udev running with 2.6.0-test6 Message-ID: <20031005034816.GA9384@kroah.com> References: <20031004213909.GA8566@kroah.com> <200310050209.h9529gSm002941@orion.dwf.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200310050209.h9529gSm002941@orion.dwf.com> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Oct 04, 2003 at 08:09:42PM -0600, reg@dwf.com wrote: > > I did a couple of greps of the code, found that there was some debugging > support, changed the syslog to print this stuff out, and ran your two tests. > > My problem is in the same place in both, probably 80 lines or so in, where > I get the sequence: > > Oct 4 19:41:52 orion udev: udev_init: sysfs_path = /sysfs > Oct 4 19:41:52 orion udev: get_class_dev: looking at > /sysfs/class/tty/ttyUSB0 > Oct 4 19:41:52 orion udev: get_class_dev: sysfs_open_class_device failed > > And, yes your code is smart enough to find the sysfs, nomatter whare it is > mounted, > be that /sys or /sysfs. For the other test the middle line is replaced with > > Oct 4 19:57:35 orion udev: get_class_dev: looking at /sysfs/block/sda > > In neither case does the LAST component of the name exist in the real /dev > filesystem. Then the test wouldn't have helped you out any, as it needs those devices present. Those tests were what I used in debugging, I need to build them up into some real test cases. > Oct 4 20:06:35 orion kernel: hub 3-0:1.0: new USB device on port 2, > assigned address 4 > Oct 4 20:06:35 orion udev: main: looking at /devices/pci0000:00/0000:00:0d > .0/usb3/3-2 > Oct 4 20:06:35 orion udev: main: not block or class > Oct 4 20:06:35 orion udev: main: looking at /devices/pci0000:00/0000:00:0d > .0/usb3/3-2/3-2:2.0 > Oct 4 20:06:35 orion udev: main: not block or class Hm, does this USB device actually work on your machine? Do you have the usb-storage driver and the scsi stuff loaded? You need to have that loaded for udev to be able to work. > But just a REALLY dumb question, one that should be answered in the README or > INSTALL or something. If this *were* working, where would I expect to > see the new fs entry appear? in /udev, in /dev, somewhare else? Ive > been looking everywhere, but with the above error, havent seen > anything yet. You should see the device node created in /udev. But it should also show up in sysfs too, and it doesn't seem like that is hapening, so udev can't really work yet. Thanks for putting up with some very rough code. Hope this helps, greg k-h