linux-kernel - Re: [PATCH 1/1] media: v4l: Don't turn on privacy LED if streamon fails

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <b4960c23-4354-427b-8201-6b7531f1343e@redhat.com>
Date: Wed, 10 Apr 2024 13:37:13 +0200
From: Hans de Goede <hdegoede@...hat.com>
To: Sakari Ailus <sakari.ailus@...ux.intel.com>
Cc: linux-media@...r.kernel.org, tomi.valkeinen@...asonboard.com,
 Mauro Carvalho Chehab <mchehab@...nel.org>, Hans Verkuil
 <hverkuil@...all.nl>, Laurent Pinchart <laurent.pinchart@...asonboard.com>,
 Umang Jain <umang.jain@...asonboard.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/1] media: v4l: Don't turn on privacy LED if streamon
 fails

Hi,

On 4/10/24 1:35 PM, Sakari Ailus wrote:
> On Wed, Apr 10, 2024 at 11:31:48AM +0000, Sakari Ailus wrote:
>> Hi Hans,
>>
>> Thanks for the review.
>>
>> On Wed, Apr 10, 2024 at 12:12:33PM +0200, Hans de Goede wrote:
>>> Hi Sakari,
>>>
>>> Thank you for fixing this.
>>>
>>> On 4/10/24 12:03 PM, Sakari Ailus wrote:
>>>> Turn on the privacy LED only if streamon succeeds. This can be done after
>>>> enabling streaming on the sensor.
>>>>
>>>> Fixes: b6e10ff6c23d ("media: v4l2-core: Make the v4l2-core code enable/disable the privacy LED if present")
>>>> Signed-off-by: Sakari Ailus <sakari.ailus@...ux.intel.com>
>>>> ---
>>>>  drivers/media/v4l2-core/v4l2-subdev.c | 19 ++++++++++---------
>>>>  1 file changed, 10 insertions(+), 9 deletions(-)
>>>>
>>>> diff --git a/drivers/media/v4l2-core/v4l2-subdev.c b/drivers/media/v4l2-core/v4l2-subdev.c
>>>> index 4c6198c48dd6..acb7c3003ab6 100644
>>>> --- a/drivers/media/v4l2-core/v4l2-subdev.c
>>>> +++ b/drivers/media/v4l2-core/v4l2-subdev.c
>>>> @@ -412,15 +412,6 @@ static int call_s_stream(struct v4l2_subdev *sd, int enable)
>>>>  	if (WARN_ON(!!sd->enabled_streams == !!enable))
>>>>  		return 0;
>>>>  
>>>> -#if IS_REACHABLE(CONFIG_LEDS_CLASS)
>>>> -	if (!IS_ERR_OR_NULL(sd->privacy_led)) {
>>>> -		if (enable)
>>>> -			led_set_brightness(sd->privacy_led,
>>>> -					   sd->privacy_led->max_brightness);
>>>> -		else
>>>> -			led_set_brightness(sd->privacy_led, 0);
>>>> -	}
>>>> -#endif
>>>>  	ret = sd->ops->video->s_stream(sd, enable);
>>>>  
>>>>  	if (!enable && ret < 0) {
>>>> @@ -431,6 +422,16 @@ static int call_s_stream(struct v4l2_subdev *sd, int enable)
>>>>  	if (!ret)
>>>>  		sd->enabled_streams = enable ? BIT(0) : 0;
>>>>  
>>>> +#if IS_REACHABLE(CONFIG_LEDS_CLASS)
>>>> +	if (!IS_ERR_OR_NULL(sd->privacy_led) && !(enable && ret < 0)) {
>>>
>>> There already is a:
>>>
>>> 	if (!enable && ret < 0)
>>> 		ret = 0;
>>>
>>> block above to ignore stream-off errors, so you can just test for:
>>>
>>> 	if (!ret && !IS_ERR_OR_NULL(sd->privacy_led)) {
>>> 		...
>>> 	}
>>>
>>> And then you can move the whole block to inside the existing:
>>>
>>> 	if (!ret)
>>> 		sd->enabled_streams = enable ? BIT(0) : 0;
>>>
>>> block, so you get something like this:
>>>
>>> 	if (!ret) {
>>> 		sd->enabled_streams = enable ? BIT(0) : 0;
>>> #if IS_REACHABLE(CONFIG_LEDS_CLASS)
>>> 		if (!IS_ERR_OR_NULL(sd->privacy_led)) {
>>> 			if (enable)
>>> 				led_set_brightness(sd->privacy_led,
>>> 						   sd->privacy_led->max_brightness);
>>> 			else
>>> 				led_set_brightness(sd->privacy_led, 0);
>>> 		}
>>> #endif
>>> 	}
>>
>> Seems reasonable, I'll send v2.
> 
> I was a bit too fast to hit y. ret will be non-zero if streamon fails
> (enable non-zero), in which case the LED needs to be turned off. So you
> can't have !ret condition for all LED control.

But if stream-on fails then AFAICT the LED was never turned on in the first
place. stream on / off calls must always be balanced, right ?

IOW stream-on is not supposed to get called if the device is already
streaming and if the device is not already streaming then the LED will
already be off.

Regards,

Hans



> 
>>
>>>
>>> Which then cleans up nicely by Tomi's patch
>>> introducing the privacy LED helper functions,
>>> to something like this:
>>>
>>> 	if (!ret {
>>> 		if (enable) {
>>> 			sd->enabled_streams = BIT(0);
>>> 			v4l2_subdev_enable_privacy_led(sd);
>>> 		} else {
>>> 			sd->enabled_streams = 0;
>>> 			v4l2_subdev_disable_privacy_led(sd);
>>> 		}
>>> 	}
>>>
>>
>