From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from bh-25.webhostbox.net ([208.91.199.152]:35942 "EHLO bh-25.webhostbox.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933635AbbHDPB2 (ORCPT ); Tue, 4 Aug 2015 11:01:28 -0400 Message-ID: <55C0D3C2.6020709@roeck-us.net> Date: Tue, 04 Aug 2015 08:01:22 -0700 From: Guenter Roeck MIME-Version: 1.0 To: =?windows-1252?Q?Uwe_Kleine-K=F6nig?= CC: linux-watchdog@vger.kernel.org, Wim Van Sebroeck , linux-kernel@vger.kernel.org, Timo Kokkonen , linux-doc@vger.kernel.org, Jonathan Corbet Subject: Re: [PATCH 0/8] watchdog: Add support for keepalives triggered by infrastructure References: <1438654414-29259-1-git-send-email-linux@roeck-us.net> <20150804112431.GK9999@pengutronix.de> In-Reply-To: <20150804112431.GK9999@pengutronix.de> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 8bit Sender: linux-watchdog-owner@vger.kernel.org List-Id: linux-watchdog@vger.kernel.org Hi Uwe, On 08/04/2015 04:24 AM, Uwe Kleine-König wrote: > Hello Guenter, > > On Mon, Aug 03, 2015 at 07:13:26PM -0700, Guenter Roeck wrote: >> This patch set does not solve all limitations of the watchdog subsystem. >> Specifically, it does not add support for the following features. >> >> - It is desirable to be able to specify a maximum early timeout, >> from booting the system to opening the watchdog device. >> - Some watchdogs may require a minimum period of time between >> heartbeats. Examples are DA9062 and possibly AT91SAM9x. > > Other things that come to my mind: > - move handling of nowayout into the core > - some drivers use a reboot notifier. Not sure what they are intended > to do there. For i.MX2x it triggers a reboot; for gpio-wdt it tries > to stop the watchdog. > Be careful - those are different functions. The imx2 driver installs a _restart_ handler, which indeed is supposed to restart the system. The gpio wdt driver installs a _reboot_ notifier, which is supposed to stop the watchdog. Other drivers install a shutdown callback in the platform driver structure to do the same; not really sure what is better. > Of course this doesn't mean this should addressed here. I like the > series in general. > The list wasn't supposed to be exhaustive. Otherwise adding "support devm" should have been there as well. I even have a patch hanging around somewhere to do that. Guess I should have said something like "does not solve all timeout related limitations ...". Thanks, Guenter