From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Date: Wed, 26 Aug 2015 04:49:17 +0530 From: Pratyush Anand To: Guenter Roeck Cc: linux-watchdog@vger.kernel.org, dyoung@redhat.com, dzickus@redhat.com, open list: ABI/API , open list , Wim Van Sebroeck Subject: Re: [PATCH RFC] watchdog: Add watchdog device control through sysfs attributes Message-ID: <20150825231917.GD13340@dhcppc13.redhat.com> References: <20150825164559.GC13340@dhcppc13.redhat.com> <55DCA3C1.5050307@roeck-us.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <55DCA3C1.5050307@roeck-us.net> List-ID: On 25/08/2015:10:20:01 AM, Guenter Roeck wrote: > On 08/25/2015 09:45 AM, Pratyush Anand wrote: > >Hi Guenter, > > > >On 21/08/2015:11:18:12 PM, Pratyush Anand wrote: > >>This patch adds following attributes to watchdog device's sysfs interface. > > > >Please see if you can review it. > > > >Does this patch look fine to you? If yes then do I need to resend it by removing > >RFC tag. > > > > I didn't have time to review your patch in detail. Sorry, but please keep in mind > that this is all volunteer work for me. It's understandable. Sorry for my impatience. Please take your time. > > I don't feel comfortable with modifying the watchdog state through sysfs. > That changes the ABI substantially, since it is no longer necessary to have > an application to activate the watchdog. It also messes up internal state keeping, > since there is now no longer a well defined state indicating that the kernel > expects user space to send heartbeats, or if the kernel needs to do it (in the > driver or, with the patch set I submitted a couple of weeks ago, in the watchdog > core). I will be OK with removing 'start'. > > Given that, and I really don't have time to look into it more deeply, > I put the whole thing on a back burner for now. > No issue, will wait :-) ~Pratyush