From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759492Ab1FWONd (ORCPT ); Thu, 23 Jun 2011 10:13:33 -0400 Received: from ironport2-out.teksavvy.com ([206.248.154.183]:36706 "EHLO ironport2-out.pppoe.ca" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1759360Ab1FWONc (ORCPT ); Thu, 23 Jun 2011 10:13:32 -0400 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApIBAFxJA05Ld/sX/2dsb2JhbAAMRoRJ3BOQe4Erg3iBCgSiHA X-IronPort-AV: E=Sophos;i="4.65,413,1304308800"; d="scan'208";a="116988993" Message-ID: <4E034A09.6050107@teksavvy.com> Date: Thu, 23 Jun 2011 10:13:29 -0400 From: Mark Lord User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-GB; rv:1.9.2.18) Gecko/20110616 Thunderbird/3.1.11 MIME-Version: 1.0 To: Wim Van Sebroeck CC: Alan Cox , Arnd Bergmann , LKML , Linux Watchdog Mailing List Subject: Re: [PATCH 7/10 v2] Generic Watchdog Timer Driver References: <20110618172537.GH3441@infomag.iguana.be> <201106182107.45682.arnd@arndb.de> <20110619110328.39c0f5e1@lxorguk.ukuu.org.uk> <201106191325.17181.arnd@arndb.de> <20110619151909.1403205d@lxorguk.ukuu.org.uk> <4DFE31DC.4080201@teksavvy.com> <20110622201346.GE26745@infomag.iguana.be> In-Reply-To: <20110622201346.GE26745@infomag.iguana.be> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 11-06-22 04:13 PM, Wim Van Sebroeck wrote: > Hi Mark, > >>>>> Then you'd need an additional interface to specify which watchdog as soon >>>>> as we support multiple watchdogs. >>>> >>>> You can always have multiple ways of setting nowayout -- hardware requirements, >>>> global module option, local module option, and a new ioctl command -- but >>>> what is being used is then the logical OR of all of them. >>> >>> An ioctl for it would make a lot of sense as watchdogs are often compiled >>> in so currently there isn't a good way to runtime set this. >> >> I wouldn't mind a kernel parameter to enable a hardware watchdog timer at boot. >> Currently, there's a window at startup where the watchdog is not enabled, >> and the system could lock up and die in there without it being triggered. > > This is another tricky thing were developers will always discuss about. > What you don't want to happen is that the watchdog reboots your system when it does > an fsck at bootup (for instance because the system rebooted by the watchdog and left > the filesystem in a dirty state...). > > So it's more complex if you look at the overal system... Sure, but that's got little to do with wanting a kernel parameter to OPTIONALLY enable a hardware watchdog timer at boot. Filesystem checks are a separate issue, easily worked around in practice.