On Wed, Nov 13, 2019 at 02:36:56PM +0800, Chuhong Yuan wrote:
> The driver forgets to deal with work and workqueue in remove like what
> is done when probe fails.
> Add the missed operations to fix it.
Is it really possible for the work to still be pending when fully
registered device is properly unregistered? I thought we'd wait for
successful data acquisition in rmi_f54_buffer_queue() before unregister
can complete.
>
> Signed-off-by: Chuhong Yuan <hslester96@xxxxxxxxx>
> ---
> drivers/input/rmi4/rmi_f54.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/input/rmi4/rmi_f54.c b/drivers/input/rmi4/rmi_f54.c
> index 710b02595486..2c0cde5c775c 100644
> --- a/drivers/input/rmi4/rmi_f54.c
> +++ b/drivers/input/rmi4/rmi_f54.c
> @@ -730,6 +730,9 @@ static void rmi_f54_remove(struct rmi_function *fn)
>
> video_unregister_device(&f54->vdev);
> v4l2_device_unregister(&f54->v4l2);
> + cancel_delayed_work_sync(&f54->work);
> + flush_workqueue(f54->workqueue);
> + destroy_workqueue(f54->workqueue);
> }
>
> struct rmi_function_handler rmi_f54_handler = {
> --
> 2.23.0
>
Thanks.
--
Dmitry
