From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755062Ab1GVTSv (ORCPT ); Fri, 22 Jul 2011 15:18:51 -0400 Received: from mailrelay009.isp.belgacom.be ([195.238.6.176]:63782 "EHLO mailrelay009.isp.belgacom.be" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754859Ab1GVTSt (ORCPT ); Fri, 22 Jul 2011 15:18:49 -0400 X-Belgacom-Dynamic: yes X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Av0EAJLMKU5tgwKe/2dsb2JhbABGDadJeIh8ArtBDoMWgxsEo08 Date: Fri, 22 Jul 2011 21:18:47 +0200 From: Wim Van Sebroeck To: Wolfram Sang Cc: LKML , Linux Watchdog Mailing List , Alan Cox , Arnd Bergmann , Andrew Morton Subject: Re: [PATCH 0/11 v3] Generic Watchdog Timer Driver Message-ID: <20110722191847.GI31767@infomag.iguana.be> References: <20110711135056.GT29245@infomag.iguana.be> <20110711213145.GB5811@pengutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20110711213145.GB5811@pengutronix.de> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Wolfram, > > Changes since V2: > > * all "flags" are unsigned > > * timeout values are also unsigned > > * removed unnececessary debugging > > * clean-up comments > > * the core will be in drivers/watchdog/ and not in drivers/watchdog/core/ > > Also (just in case someone started converting drivers): > > * some error-codes changed > * .name has been removed from watchdog_device in favor of watchdog_info->identity Yup, I forgot these. Thanks. Status update: I incorporated your comments and Lars-Peter Clausen's comments and created V4. The changes since V3 are: * removed overdocumentation * removed MODULE_SUPPORTED_DEVICE and +MODULE_ALIAS_MISCDEV * removed return -ENOTTY after switch (cmd) because we have the default allready * s/sent/send/ * I'd suggest to swap the logic here (branch taken on error) * no_llseek removal squashed into patch 1 * provide getter/setter methods for driver specific data so that we can easily do the conversion towards the device driver model * removed parent because this would change when we will start using the device driver model > I already started converting a few drivers and will post the patches as RFC > tomorrow. In general, converting went quite well and despite a few comments > (see later mails), I think this could go into 3.1, so that converted drivers > can go into 3.2 together with (probably) a few updates to the framework. There > is potential for a bit more consolidation, yet I think the basic stuff could > go in now. > > Thanks to Wim and Alan for their work and CELF/LF for supporting the final > steps :) And for your assistance and the feedback from others also. Kind regards, Wim.