[PATCH 7/8] watchdog: core: Fix unbalanced module_put() in watchdog_open()

[Guenter Roeck <linux@roeck-us.net>]

If the hardware watchdog is running (hw_running is true), try_module_get()
is skipped. However, if watchdog_start() then fails, the code jumps to
out_mod and unconditionally calls module_put().

This can cause a premature driver module unload while hardware timers or
interrupts are active by dropping a reference that was never acquired.

Only call module_put() if the hardware watchdog is not running to fix the
problem.

Fixes: ee142889e32f ("watchdog: Introduce WDOG_HW_RUNNING flag")
Assisted-by: LLM
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
 drivers/watchdog/watchdog_dev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c
index 31567ffbfc23..24488c20539a 100644
--- a/drivers/watchdog/watchdog_dev.c
+++ b/drivers/watchdog/watchdog_dev.c
@@ -937,7 +937,8 @@ static int watchdog_open(struct inode *inode, struct file *file)
 	return stream_open(inode, file);
 
 out_mod:
-	module_put(wd_data->wdd->ops->owner);
+	if (!hw_running)
+		module_put(wd_data->wdd->ops->owner);
 out_clear:
 	clear_bit(_WDOG_DEV_OPEN, &wd_data->status);
 	mutex_unlock(&wd_data->lock);
-- 
2.45.2