public inbox for linux-arm-kernel@lists.infradead.org
 help / color / mirror / Atom feed
From: Freddy.Hsin <freddy.hsin@mediatek.com>
To: Guenter Roeck <linux@roeck-us.net>
Cc: linux-watchdog@vger.kernel.org, wsd_upstream@mediatek.com,
	chun-hung.wu@mediatek.com, kuohong.wang@mediatek.com,
	linux-kernel@vger.kernel.org, linux-mediatek@lists.infradead.org,
	Matthias Brugger <matthias.bgg@gmail.com>,
	chang-an.chen@mediatek.com,
	Wim Van Sebroeck <wim@linux-watchdog.org>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v1 1/1] driver: watchdog: Add mtk_wdt_init for mediatek watchdog hw init
Date: Sat, 29 Aug 2020 14:26:02 +0800	[thread overview]
Message-ID: <1598682362.14037.28.camel@mtkswgap22> (raw)
In-Reply-To: <a63e9613-d448-d784-0c60-0fd8d1b9b937@roeck-us.net>

On Fri, 2020-08-28 at 07:26 -0700, Guenter Roeck wrote:
> On 8/28/20 12:49 AM, Freddy Hsin wrote:
> > 1. add a hw initialization function
> > 2. enable/disable the watchdog depends on the original hw setting
> > 3. set WDOD_HW_RUNNING in start function in order to start
> >    kicker after driver probe and clear the bit in stop function
> > 
> > Change-Id: I25aa797f3b88288f26984455e499e599e27f09fa
> > Signed-off-by: Freddy Hsin <freddy.hsin@mediatek.com>
> 
> The subject is misleading - what this really does it to honor
> that/if a watchdog is already running. However, WDOG_HW_RUNNING
> should not generally be set/cleared in the start/stop function.
> Also, calling mtk_wdt_stop is pointless if WDT_MODE_EN is not set,
> since clearing WDT_MODE_EN is all it does. On top of that,
> setting WDOG_HW_RUNNING in the probe function requires that
> max_hw_heartbeat_ms is set (instead of max_timeout).
> 
> So if you really want to honor that the watchdog is already running
> at boot, you would have to initialize max_hw_heartbeat_ms, set
> WDOG_HW_RUNNING explicitly, and possibly call the ping function
> before changing the timeout.
> 
> Thanks,
> Guenter

Hi Guenter,

Thanks for you suggestion. I will submit a new patch according to
your advice.

The purpose of this patch is to control WDT_MODE_EN in the first stage
of preloader bring up, and it should not be disabled afterward to
prevent the system freeze and cannot be reset, so remove the
mtk_wdt_stop function and keep the original state set in preloader.

> 
> > ---
> >  drivers/watchdog/mtk_wdt.c |   22 +++++++++++++++++++++-
> >  1 file changed, 21 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/watchdog/mtk_wdt.c b/drivers/watchdog/mtk_wdt.c
> > index d6a6393..59b5061 100644
> > --- a/drivers/watchdog/mtk_wdt.c
> > +++ b/drivers/watchdog/mtk_wdt.c
> > @@ -57,6 +57,9 @@
> >  static bool nowayout = WATCHDOG_NOWAYOUT;
> >  static unsigned int timeout;
> >  
> > +static int mtk_wdt_start(struct watchdog_device *wdt_dev);
> > +static int mtk_wdt_stop(struct watchdog_device *wdt_dev);
> > +
> >  struct mtk_wdt_dev {
> >  	struct watchdog_device wdt_dev;
> >  	void __iomem *wdt_base;
> > @@ -148,6 +151,19 @@ static int toprgu_register_reset_controller(struct platform_device *pdev,
> >  	return ret;
> >  }
> >  
> > +static int mtk_wdt_init(struct watchdog_device *wdt_dev)
> > +{
> > +	struct mtk_wdt_dev *mtk_wdt = watchdog_get_drvdata(wdt_dev);
> > +	void __iomem *wdt_base;
> > +
> > +	wdt_base = mtk_wdt->wdt_base;
> > +
> > +	if (readl(wdt_base + WDT_MODE) & WDT_MODE_EN)
> > +		mtk_wdt_start(wdt_dev);
> > +	else
> > +		mtk_wdt_stop(wdt_dev);
> > +}
> > +
> >  static int mtk_wdt_restart(struct watchdog_device *wdt_dev,
> >  			   unsigned long action, void *data)
> >  {
> > @@ -206,6 +222,8 @@ static int mtk_wdt_stop(struct watchdog_device *wdt_dev)
> >  	reg |= WDT_MODE_KEY;
> >  	iowrite32(reg, wdt_base + WDT_MODE);
> >  
> > +	clear_bit(WDOG_HW_RUNNING, &wdt_dev->status);
> > +
> >  	return 0;
> >  }
> >  
> > @@ -225,6 +243,8 @@ static int mtk_wdt_start(struct watchdog_device *wdt_dev)
> >  	reg |= (WDT_MODE_EN | WDT_MODE_KEY);
> >  	iowrite32(reg, wdt_base + WDT_MODE);
> >  
> > +	set_bit(WDOG_HW_RUNNING, &wdt_dev->status);
> > +
> >  	return 0;
> >  }
> >  
> > @@ -274,7 +294,7 @@ static int mtk_wdt_probe(struct platform_device *pdev)
> >  
> >  	watchdog_set_drvdata(&mtk_wdt->wdt_dev, mtk_wdt);
> >  
> > -	mtk_wdt_stop(&mtk_wdt->wdt_dev);
> > +	mtk_wdt_init(&mtk_wdt->wdt_dev);
> >  
> >  	watchdog_stop_on_reboot(&mtk_wdt->wdt_dev);
> >  	err = devm_watchdog_register_device(dev, &mtk_wdt->wdt_dev);
> > 
> 

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

      reply	other threads:[~2020-08-29  6:38 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-28  7:49 [PATCH 0/1] Refine mtk wdt driver init flow Freddy Hsin
2020-08-28  7:49 ` [PATCH v1 1/1] driver: watchdog: Add mtk_wdt_init for mediatek watchdog hw init Freddy Hsin
2020-08-28 14:26   ` Guenter Roeck
2020-08-29  6:26     ` Freddy.Hsin [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1598682362.14037.28.camel@mtkswgap22 \
    --to=freddy.hsin@mediatek.com \
    --cc=chang-an.chen@mediatek.com \
    --cc=chun-hung.wu@mediatek.com \
    --cc=kuohong.wang@mediatek.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=linux-watchdog@vger.kernel.org \
    --cc=linux@roeck-us.net \
    --cc=matthias.bgg@gmail.com \
    --cc=wim@linux-watchdog.org \
    --cc=wsd_upstream@mediatek.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox