[PATCH] watchdog: iTCO_wdt: Report error if timeout configuration fails

Ziyan Fu posted 1 patch 3 months ago
drivers/watchdog/iTCO_wdt.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
[PATCH] watchdog: iTCO_wdt: Report error if timeout configuration fails
Posted by Ziyan Fu 3 months ago
From: Ziyan Fu <fuzy5@lenovo.com>

The driver probes with the invalid timeout value when
'iTCO_wdt_set_timeout()' fails, as its return value is not checked. In
this case, when executing "wdctl", we may get:

Device:        /dev/watchdog0
Timeout:       30 seconds
Timeleft:      613 seconds

The timeout value is the value of "heartbeat" or "WATCHDOG_TIMEOUT", and
the timeleft value is calculated from the register value we actually read
(0xffff) by masking with 0x3ff and converting ticks to seconds (* 6 / 10).

Add error handling to return the failure code if 'iTCO_wdt_set_timeout()'
fails, ensuring the driver probe fails and prevents invalid operation.

Signed-off-by: Ziyan Fu <fuzy5@lenovo.com>
---
 drivers/watchdog/iTCO_wdt.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/watchdog/iTCO_wdt.c b/drivers/watchdog/iTCO_wdt.c
index 9ab769aa0244..4ab3405ef8e6 100644
--- a/drivers/watchdog/iTCO_wdt.c
+++ b/drivers/watchdog/iTCO_wdt.c
@@ -577,7 +577,11 @@ static int iTCO_wdt_probe(struct platform_device *pdev)
 	/* Check that the heartbeat value is within it's range;
 	   if not reset to the default */
 	if (iTCO_wdt_set_timeout(&p->wddev, heartbeat)) {
-		iTCO_wdt_set_timeout(&p->wddev, WATCHDOG_TIMEOUT);
+		ret = iTCO_wdt_set_timeout(&p->wddev, WATCHDOG_TIMEOUT);
+		if (ret != 0) {
+			dev_err(dev, "Failed to set watchdog timeout (%d)\n", WATCHDOG_TIMEOUT);
+			return ret;
+		}
 		dev_info(dev, "timeout value out of range, using %d\n",
 			WATCHDOG_TIMEOUT);
 		heartbeat = WATCHDOG_TIMEOUT;
-- 
2.27.0
Re: [PATCH] watchdog: iTCO_wdt: Report error if timeout configuration fails
Posted by Guenter Roeck 3 months ago
On Fri, Jul 04, 2025 at 03:35:18PM +0800, Ziyan Fu wrote:
> From: Ziyan Fu <fuzy5@lenovo.com>
> 
> The driver probes with the invalid timeout value when
> 'iTCO_wdt_set_timeout()' fails, as its return value is not checked. In
> this case, when executing "wdctl", we may get:
> 
> Device:        /dev/watchdog0
> Timeout:       30 seconds
> Timeleft:      613 seconds
> 
> The timeout value is the value of "heartbeat" or "WATCHDOG_TIMEOUT", and
> the timeleft value is calculated from the register value we actually read
> (0xffff) by masking with 0x3ff and converting ticks to seconds (* 6 / 10).
> 
> Add error handling to return the failure code if 'iTCO_wdt_set_timeout()'
> fails, ensuring the driver probe fails and prevents invalid operation.
> 
> Signed-off-by: Ziyan Fu <fuzy5@lenovo.com>

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

You might want to figure out why this fails in the first place, and address
the root cause in addition to its consequences. A timeout of 30 seconds
should never be invalid.

Guenter

> ---
>  drivers/watchdog/iTCO_wdt.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/watchdog/iTCO_wdt.c b/drivers/watchdog/iTCO_wdt.c
> index 9ab769aa0244..4ab3405ef8e6 100644
> --- a/drivers/watchdog/iTCO_wdt.c
> +++ b/drivers/watchdog/iTCO_wdt.c
> @@ -577,7 +577,11 @@ static int iTCO_wdt_probe(struct platform_device *pdev)
>  	/* Check that the heartbeat value is within it's range;
>  	   if not reset to the default */
>  	if (iTCO_wdt_set_timeout(&p->wddev, heartbeat)) {
> -		iTCO_wdt_set_timeout(&p->wddev, WATCHDOG_TIMEOUT);
> +		ret = iTCO_wdt_set_timeout(&p->wddev, WATCHDOG_TIMEOUT);
> +		if (ret != 0) {
> +			dev_err(dev, "Failed to set watchdog timeout (%d)\n", WATCHDOG_TIMEOUT);
> +			return ret;
> +		}
>  		dev_info(dev, "timeout value out of range, using %d\n",
>  			WATCHDOG_TIMEOUT);
>  		heartbeat = WATCHDOG_TIMEOUT;
> -- 
> 2.27.0
>