4.20-stable review patch. If anyone has any objections, please let me know.
------------------
[ Upstream commit c764da98a600a4b068d25c77164f092f159cecec ]
The video device release() callback for video-i2c driver frees the whole struct video_i2c_data. If there is no user left for the video device when video_unregister_device() is called, the release callback is executed.
However, in video_i2c_remove() some fields (v4l2_dev, lock, and queue_lock) in struct video_i2c_data are still accessed after video_unregister_device() is called.
This fixes the use after free by moving the code from video_i2c_remove() to the release() callback.
Fixes: 5cebaac60974 ("media: video-i2c: add video-i2c driver")
Reviewed-by: Matt Ranostay matt.ranostay@konsulko.com Signed-off-by: Akinobu Mita akinobu.mita@gmail.com Acked-by: Sakari Ailus sakari.ailus@linux.intel.com Signed-off-by: Hans Verkuil hverkuil@xs4all.nl Signed-off-by: Mauro Carvalho Chehab mchehab+samsung@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/media/i2c/video-i2c.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/media/i2c/video-i2c.c b/drivers/media/i2c/video-i2c.c index 4d49af86c15e..ec0758dca2fc 100644 --- a/drivers/media/i2c/video-i2c.c +++ b/drivers/media/i2c/video-i2c.c @@ -510,7 +510,12 @@ static const struct v4l2_ioctl_ops video_i2c_ioctl_ops = {
static void video_i2c_release(struct video_device *vdev) { - kfree(video_get_drvdata(vdev)); + struct video_i2c_data *data = video_get_drvdata(vdev); + + v4l2_device_unregister(&data->v4l2_dev); + mutex_destroy(&data->lock); + mutex_destroy(&data->queue_lock); + kfree(data); }
static int video_i2c_probe(struct i2c_client *client, @@ -608,10 +613,6 @@ static int video_i2c_remove(struct i2c_client *client) struct video_i2c_data *data = i2c_get_clientdata(client);
video_unregister_device(&data->vdev); - v4l2_device_unregister(&data->v4l2_dev); - - mutex_destroy(&data->lock); - mutex_destroy(&data->queue_lock);
return 0; }