Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 7 Dec 2001 10:47:02 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 7 Dec 2001 10:46:52 -0500 Received: from sushi.toad.net ([162.33.130.105]:5033 "EHLO sushi.toad.net") by vger.kernel.org with ESMTP id ; Fri, 7 Dec 2001 10:46:43 -0500 Subject: Re: devfs unable to handle permission: 2.4.17-pre[4,5] / From: Thomas Hood To: linux-kernel@vger.kernel.org Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Evolution/1.0 (Preview Release) Date: 07 Dec 2001 10:47:37 -0500 Message-Id: <1007740060.2031.2.camel@thanatos> Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Some devfs permission problems may have arisen because of the fact that devfs now notifies devfsd of the creation of directories. Many people have devfsd configured to set permissions to all devices matching a certain regular expression --- e.g., all devices with "sound" in their pathname. The problem is that the "sound" directory itself matches this regular expression, and so will have its perm bits set exactly like the device files' perm bits---e.g., with the eXamine bit cleared. The solution is to edit the devfsd config so that it excludes the directory. E.g., instead of: REGISTER sound PERMISSIONS root.audio 0664 (which worked before but won't any more) do: REGISTER ^sound/.* PERMISSIONS root.audio 0664 or something similar. Roman Zippel wrote: > Option 3: > Turn a user generated entry into a kernel generated > one and return 0. Prepopulating devfs was a valid > option so far, you cannot simply change this during > a stable kernel release. As Richard has pointed out, devfs is still marked "experimental", so it's not unreasonable to make changes if they are improvements. -- Thomas Hood - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/