linux-watchdog.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Guenter Roeck <linux@roeck-us.net>
To: Wolfram Sang <wsa+renesas@sang-engineering.com>,
	linux-watchdog@vger.kernel.org
Cc: linux-renesas-soc@vger.kernel.org
Subject: Re: [PATCH] watchdog: renesas_wdt: don't sleep in atomic context
Date: Mon, 21 Dec 2020 08:27:00 -0800	[thread overview]
Message-ID: <63c0b5ba-fe18-b667-d346-880287061c46@roeck-us.net> (raw)
In-Reply-To: <20201219173415.21848-1-wsa+renesas@sang-engineering.com>

On 12/19/20 9:34 AM, Wolfram Sang wrote:
> In the restart handler, we hit multiple OOPSes. One because of
> usleep_range() and one because of RPM. So, instead of re-using
> rwdt_start(), we implement an atomic version of it in the restart
> handler. As a little bonus, reboot will occur sooner because we can now
> use the smallest divider in the custom restart routine.
> 
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Reviewed-by: Guenter Roeck <linux@roeck-us.net>

> ---
> 
> It is problematic to add a Fixes: tag because we would need three of
> them. So, I'd think we add a stable tag and as long as the patch
> applies, all is fine. Other opinions?
> 
>  drivers/watchdog/renesas_wdt.c | 30 ++++++++++++++++++++++++------
>  1 file changed, 24 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/watchdog/renesas_wdt.c b/drivers/watchdog/renesas_wdt.c
> index 47fce4de0110..d2b5074bca65 100644
> --- a/drivers/watchdog/renesas_wdt.c
> +++ b/drivers/watchdog/renesas_wdt.c
> @@ -9,6 +9,7 @@
>  #include <linux/clk.h>
>  #include <linux/delay.h>
>  #include <linux/io.h>
> +#include <linux/iopoll.h>
>  #include <linux/kernel.h>
>  #include <linux/module.h>
>  #include <linux/of.h>
> @@ -50,6 +51,7 @@ struct rwdt_priv {
>  	struct watchdog_device wdev;
>  	unsigned long clk_rate;
>  	u8 cks;
> +	struct clk *clk;
>  };
>  
>  static void rwdt_write(struct rwdt_priv *priv, u32 val, unsigned int reg)
> @@ -125,13 +127,30 @@ static unsigned int rwdt_get_timeleft(struct watchdog_device *wdev)
>  	return DIV_BY_CLKS_PER_SEC(priv, 65536 - val);
>  }
>  
> +/* needs to be atomic - no RPM, no usleep_range, no scheduling! */
>  static int rwdt_restart(struct watchdog_device *wdev, unsigned long action,
>  			void *data)
>  {
>  	struct rwdt_priv *priv = watchdog_get_drvdata(wdev);
> +	u8 val;
> +
> +	clk_prepare_enable(priv->clk);
> +
> +	/* Stop the timer before we modify any register */
> +	val = readb_relaxed(priv->base + RWTCSRA) & ~RWTCSRA_TME;
> +	rwdt_write(priv, val, RWTCSRA);
> +	/* Delay 2 cycles before setting watchdog counter */
> +	udelay(DIV_ROUND_UP(2 * 1000000, priv->clk_rate));
>  
> -	rwdt_start(wdev);
>  	rwdt_write(priv, 0xffff, RWTCNT);
> +	/* smallest divider to reboot soon */
> +	rwdt_write(priv, 0, RWTCSRA);
> +
> +	readb_poll_timeout_atomic(priv->base + RWTCSRA, val,
> +				  !(val & RWTCSRA_WRFLG), 1, 100);
> +
> +	rwdt_write(priv, RWTCSRA_TME, RWTCSRA);
> +
>  	return 0;
>  }
>  
> @@ -191,7 +210,6 @@ static int rwdt_probe(struct platform_device *pdev)
>  {
>  	struct device *dev = &pdev->dev;
>  	struct rwdt_priv *priv;
> -	struct clk *clk;
>  	unsigned long clks_per_sec;
>  	int ret, i;
>  	u8 csra;
> @@ -207,13 +225,13 @@ static int rwdt_probe(struct platform_device *pdev)
>  	if (IS_ERR(priv->base))
>  		return PTR_ERR(priv->base);
>  
> -	clk = devm_clk_get(dev, NULL);
> -	if (IS_ERR(clk))
> -		return PTR_ERR(clk);
> +	priv->clk = devm_clk_get(dev, NULL);
> +	if (IS_ERR(priv->clk))
> +		return PTR_ERR(priv->clk);
>  
>  	pm_runtime_enable(dev);
>  	pm_runtime_get_sync(dev);
> -	priv->clk_rate = clk_get_rate(clk);
> +	priv->clk_rate = clk_get_rate(priv->clk);
>  	csra = readb_relaxed(priv->base + RWTCSRA);
>  	priv->wdev.bootstatus = csra & RWTCSRA_WOVF ? WDIOF_CARDRESET : 0;
>  	pm_runtime_put(dev);
> 


  reply	other threads:[~2020-12-21 18:19 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-19 17:34 [PATCH] watchdog: renesas_wdt: don't sleep in atomic context Wolfram Sang
2020-12-21 16:27 ` Guenter Roeck [this message]
2020-12-22  8:26 ` Geert Uytterhoeven
2020-12-27 14:11 ` Niklas Söderlund

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=63c0b5ba-fe18-b667-d346-880287061c46@roeck-us.net \
    --to=linux@roeck-us.net \
    --cc=linux-renesas-soc@vger.kernel.org \
    --cc=linux-watchdog@vger.kernel.org \
    --cc=wsa+renesas@sang-engineering.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;
as well as URLs for NNTP newsgroup(s).