[80/98] hwmon: (fschmd) Fix a memleak on multiple opens of /dev/watchdog

From: Greg KH
Date: Tue Jan 26 2010 - 18:48:55 EST


2.6.32-stable review patch. If anyone has any objections, please let us know.

------------------

From: Hans de Goede <hdegoede@xxxxxxxxxx>

commit c453615f77aa51593c1c9c9031b4278797d3fd19 upstream.

When /dev/watchdog gets opened a second time we return -EBUSY, but
we already have got a kref then, so we end up leaking our data struct.

Signed-off-by: Hans de Goede <hdegoede@xxxxxxxxxx>
Signed-off-by: Jean Delvare <khali@xxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>

---
drivers/hwmon/fschmd.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)

--- a/drivers/hwmon/fschmd.c
+++ b/drivers/hwmon/fschmd.c
@@ -767,6 +767,7 @@ leave:
static int watchdog_open(struct inode *inode, struct file *filp)
{
struct fschmd_data *pos, *data = NULL;
+ int watchdog_is_open;

/* We get called from drivers/char/misc.c with misc_mtx hold, and we
call misc_register() from fschmd_probe() with watchdog_data_mutex
@@ -781,10 +782,12 @@ static int watchdog_open(struct inode *i
}
}
/* Note we can never not have found data, so we don't check for this */
- kref_get(&data->kref);
+ watchdog_is_open = test_and_set_bit(0, &data->watchdog_is_open);
+ if (!watchdog_is_open)
+ kref_get(&data->kref);
mutex_unlock(&watchdog_data_mutex);

- if (test_and_set_bit(0, &data->watchdog_is_open))
+ if (watchdog_is_open)
return -EBUSY;

/* Start the watchdog */


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/