From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750726AbWGUNVw (ORCPT ); Fri, 21 Jul 2006 09:21:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750727AbWGUNVw (ORCPT ); Fri, 21 Jul 2006 09:21:52 -0400 Received: from smtp.net4india.com ([202.71.129.67]:16568 "EHLO smtp.net4india.com") by vger.kernel.org with ESMTP id S1750726AbWGUNVv (ORCPT ); Fri, 21 Jul 2006 09:21:51 -0400 Message-ID: <44C0D5B0.4000008@designergraphix.com> Date: Fri, 21 Jul 2006 18:55:04 +0530 From: Kaiwan N Billimoria Reply-To: kaiwan@designergraphix.com Organization: Designer Graphix User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.13) Gecko/20060417 X-Accept-Language: en-us, en MIME-Version: 1.0 To: linux-kernel@vger.kernel.org CC: Tim Waugh , Philip Blundell Subject: parport: small addition to Documentation/parport-lowlevel.txt Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org A small note inserted, regarding usage of the parport_unregister_driver interface. Only a single file Documentation/parport-lowlevel.txt is affected (pulled from 2.6.18-rc2 : http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;h=8f2302415eff90b0df7aa655f132f9567ff9ccd7;hb=82d6897fefca6206bca7153805b4c5359ce97fc4;f=Documentation/parport-lowlevel.txt ) Came across this "issue" while writing a parport-based driver. Does this make sense? - Kaiwan. --- --- parport-lowlevel.vanilla 2006-07-21 18:25:24.000000000 +0530 +++ parport-lowlevel 2006-07-21 18:34:52.000000000 +0530 @@ -258,6 +258,18 @@ ... } +Note- +If you are using this interface in the "usual" way, you are probably first +registering your driver with parport_register_driver and unregistering with +parport_unregister_driver (as shown in the "lp_driver" example about a page +up from here). You will have attach and detach function callbacks. +The point is: when your driver is removed from the kernel, your +cleanup_module function is invoked, which typically invokes +parport_unregister_driver. This will cause the detach function to kick in. +Now, this guy should _not_ be (re)invoking parport_unregister_driver, but +just parport_unregister_device as necessary. + + SEE ALSO parport_register_driver, parport_enumerate