From: Dinghao Liu <dinghao....@zju.edu.cn>

[ Upstream commit 5afb6d203d0293512aa2c6ae098274a2a4f6ed02 ]

When watchdog_kworker is NULL, we should free wd_data
before the function returns to prevent memleak.

Fixes: 664a39236e718 ("watchdog: Introduce hardware maximum heartbeat in 
watchdog core")
Signed-off-by: Dinghao Liu <dinghao....@zju.edu.cn>
Reviewed-by: Guenter Roeck <li...@roeck-us.net>
Link: https://lore.kernel.org/r/20200824024001.25474-1-dinghao....@zju.edu.cn
Signed-off-by: Guenter Roeck <li...@roeck-us.net>
Signed-off-by: Wim Van Sebroeck <w...@linux-watchdog.org>
Signed-off-by: Sasha Levin <sas...@kernel.org>
---
 drivers/watchdog/watchdog_dev.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c
index 1c322caecf7f1..1e4921f89fb52 100644
--- a/drivers/watchdog/watchdog_dev.c
+++ b/drivers/watchdog/watchdog_dev.c
@@ -944,8 +944,10 @@ static int watchdog_cdev_register(struct watchdog_device 
*wdd)
        wd_data->wdd = wdd;
        wdd->wd_data = wd_data;
 
-       if (IS_ERR_OR_NULL(watchdog_kworker))
+       if (IS_ERR_OR_NULL(watchdog_kworker)) {
+               kfree(wd_data);
                return -ENODEV;
+       }
 
        device_initialize(&wd_data->dev);
        wd_data->dev.devt = MKDEV(MAJOR(watchdog_devt), wdd->id);
-- 
2.25.1



Reply via email to