[PATCH 09/10] watchdog: core: Clear WDOG_HW_RUNNING before calling the stop function

Maxim Yu, Osipov

From: Guenter Roeck <linux@...>

commit 3c10bbde10fe4dca52726e246cefa6b0a1dfbd3e upstream.

WDOG_HW_RUNNING indicates that the hardware watchdog is running while the
watchdog device is closed. The flag may be set by the driver when it is
instantiated to indicate that the watchdog is running, and that the
watchdog core needs to send heartbeat requests to the driver until the
watchdog device is opened.

When the watchdog device is closed, the flag can be used by the driver's
stop function to indicate to the watchdog core that it was unable to stop
the watchdog, and that the watchdog core needs to send heartbeat requests.
This only works if the flag is actually cleared when the watchdog is
stopped. To avoid having to clear the flag in each driver's stop function,
clear it in the watchdog core before calling the stop function.

Reported-by: Rasmus Villemoes <rasmus.villemoes@...>
Fixes: ee142889e32f ("watchdog: Introduce WDOG_HW_RUNNING flag")
Signed-off-by: Guenter Roeck <linux@...>
Signed-off-by: Wim Van Sebroeck <wim@...>
[mosipov@... backported to 4.4.y]
Signed-off-by: Maxim Yu. Osipov <mosipov@...>
drivers/watchdog/watchdog_dev.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c
index 263d7fa0ef31..1edc3d78636b 100644
--- a/drivers/watchdog/watchdog_dev.c
+++ b/drivers/watchdog/watchdog_dev.c
@@ -241,10 +241,12 @@ static int watchdog_stop(struct watchdog_device *wdd)
goto out_stop;

- if (wdd->ops->stop)
+ if (wdd->ops->stop) {
+ clear_bit(WDOG_HW_RUNNING, &wdd->status);
err = wdd->ops->stop(wdd);
- else
+ } else {
set_bit(WDOG_HW_RUNNING, &wdd->status);
+ }

if (err == 0) {
clear_bit(WDOG_ACTIVE, &wdd->status);

Join {cip-dev@lists.cip-project.org to automatically receive all group messages.