Re: [PATCH] Input: rmi4 - prevent null pointer dereference in f30

From: Dmitry Torokhov
Date: Fri Mar 10 2017 - 13:23:51 EST


On Fri, Mar 10, 2017 at 07:05:52PM +0100, Benjamin Tissoires wrote:
> If the platform data has f30_data.disable set, f30 in rmi_f30_config()
> might be null. Prevent a kernel oops by checking for non-null f30.
>
> Signed-off-by: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx>

Applied, thank you.

> ---
> drivers/input/rmi4/rmi_f30.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/drivers/input/rmi4/rmi_f30.c b/drivers/input/rmi4/rmi_f30.c
> index d8ff285..a0e4030 100644
> --- a/drivers/input/rmi4/rmi_f30.c
> +++ b/drivers/input/rmi4/rmi_f30.c
> @@ -170,6 +170,10 @@ static int rmi_f30_config(struct rmi_function *fn)
> rmi_get_platform_data(fn->rmi_dev);
> int error;
>
> + /* can happen if f30_data.disable is set */
> + if (!f30)
> + return 0;
> +
> if (pdata->f30_data.trackstick_buttons) {
> /* Try [re-]establish link to F03. */
> f30->f03 = rmi_find_function(fn->rmi_dev, 0x03);
> --
> 2.9.3
>

--
Dmitry