Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Guenter Roeck <groeck7@gmail.com>
Date:   Wed, 16 Jun 2021 10:57:35 -0700
From:   Guenter Roeck <linux@roeck-us.net>
To:     Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Cc:     wim@linux-watchdog.org, shawnguo@kernel.org,
        linux-watchdog@vger.kernel.org,
        linux-arm-kernel <linux-arm-kernel@lists.infradead.org>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/2] watchdog: introduce watchdog_dev_suspend/resume
Message-ID: <20210616175735.GA1111788@roeck-us.net>
References: <20210615123904.2568052-1-grzegorz.jaszczyk@linaro.org>
 <20210615123904.2568052-2-grzegorz.jaszczyk@linaro.org>
 <20210615141803.GA957871@roeck-us.net>
 <CAMxfBF7EZRaCcrCYmY1RkWytWWHiH7xATfzKO68Hqtf3M--L_Q@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAMxfBF7EZRaCcrCYmY1RkWytWWHiH7xATfzKO68Hqtf3M--L_Q@mail.gmail.com>
Precedence: bulk

On Wed, Jun 16, 2021 at 03:59:23PM +0200, Grzegorz Jaszczyk wrote:
> On Tue, 15 Jun 2021 at 16:18, Guenter Roeck <linux@roeck-us.net> wrote:
> >
> > On Tue, Jun 15, 2021 at 02:39:03PM +0200, Grzegorz Jaszczyk wrote:
> > > The watchdog drivers often disable wdog clock during suspend and then
> > > enable it again during resume. Nevertheless the ping worker is still
> > > running and can issue low-level ping while the wdog clock is disabled
> > > causing the system hang. To prevent such condition introduce
> > > watchdog_dev_suspend/resume which can be used by any wdog driver and
> > > actually cancel ping worker during suspend and restore it back, if
> > > needed, during resume.
> > >
> >
> > I'll have to look into this further, but I don't think this is the correct
> > solution. Most likely the watchdog core needs to have its own independent
> > suspend/resule functions and suspend the high resolution timer on
> > suspend and restore it on resume. This may require an additional flag
> > to be set by drivers to indicate that the timer should be stopped on
> > suspend.
> 
> That makes sense - thank you for your suggestion. I think I could
> register a pm notifier in the watchdog core when the new e.g.
> WDOG_STOP_PING_ON_SUSPEND status flag will be set by the driver and
> actually call watchdog_dev_suspend/resume from the notifier callback.
> Please let me know if you see any other issue with this solution, if
> not I will post v2.
> 
Go for it.

Thanks,
Guenter