On Fri, Sep 30, 2016 at 11:54 PM, Rui Miguel Silva <rmfrfs@xxxxxxxxx> wrote:
> Even though we trust leds core that the pointers should be valid, we are
> safer to check delay_{on|off} before use.
As these routines are getting called from core, I am not sure if it will
be right to have such checks.
Though, I saw the core code after writing above and the core doesn't check
these at all :)
> Also, this avoid a smatch warning:
> drivers/staging/greybus/light.c:484 gb_blink_set()
> warn: variable dereferenced before check 'delay_on' (see line 476)
>
> Signed-off-by: Rui Miguel Silva <rmfrfs@xxxxxxxxx>
> ---
> drivers/staging/greybus/light.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/staging/greybus/light.c b/drivers/staging/greybus/light.c
> index f3cd485..80dc4a9 100644
> --- a/drivers/staging/greybus/light.c
> +++ b/drivers/staging/greybus/light.c
> @@ -463,6 +463,9 @@ static int gb_blink_set(struct led_classdev *cdev, unsigned long *delay_on,
> if (channel->releasing)
> return -ESHUTDOWN;
>
> + if (!delay_on || !delay_off)
> + return -EINVAL;
> +
> mutex_lock(&channel->lock);
> ret = gb_pm_runtime_get_sync(bundle);
> if (ret < 0)
Acked-by: Viresh Kumar <viresh.kumar@xxxxxxxxxx>
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
