Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 16 Jan 2023 23:01:09 +0200
From:   Sakari Ailus <sakari.ailus@iki.fi>
To:     Aleksandr Burakov <a.burakov@rosalinux.ru>
Cc:     Mauro Carvalho Chehab <mchehab@kernel.org>,
        linux-media@vger.kernel.org, linux-kernel@vger.kernel.org,
        lvc-project@linuxtesting.org
Subject: Re: [PATCH] media: v4l2-flash: fix NULL dereference in
 v4l2_flash_s_ctrl()
Message-ID: <Y8W7FaNouVxmeWox@valkosipuli.retiisi.eu>
References: <20221207141808.22922-1-a.burakov@rosalinux.ru>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20221207141808.22922-1-a.burakov@rosalinux.ru>
Precedence: bulk

Hi Alexandr,

On Wed, Dec 07, 2022 at 05:18:07PM +0300, Aleksandr Burakov wrote:
> The NULL check added for fled_cdev before dereference.
> 
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
> 
> Signed-off-by: Aleksandr Burakov <a.burakov@rosalinux.ru>
> Fixes: 42bd6f59ae90 ("media: Add registration helpers for V4L2 flash sub-devices")
> ---
>  drivers/media/v4l2-core/v4l2-flash-led-class.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/media/v4l2-core/v4l2-flash-led-class.c b/drivers/media/v4l2-core/v4l2-flash-led-class.c
> index 355595a0fefa..36cc46e80eea 100644
> --- a/drivers/media/v4l2-core/v4l2-flash-led-class.c
> +++ b/drivers/media/v4l2-core/v4l2-flash-led-class.c
> @@ -291,12 +291,16 @@ static int v4l2_flash_s_ctrl(struct v4l2_ctrl *c)
>  		 * No conversion is needed as LED Flash class also uses
>  		 * microseconds for flash timeout units.
>  		 */
> +		if (!fled_cdev)
> +			return -EINVAL;
>  		return led_set_flash_timeout(fled_cdev, c->val);
>  	case V4L2_CID_FLASH_INTENSITY:
>  		/*
>  		 * No conversion is needed as LED Flash class also uses
>  		 * microamperes for flash intensity units.
>  		 */
> +		if (!fled_cdev)
> +			return -EINVAL;
>  		return led_set_flash_brightness(fled_cdev, c->val);
>  	}
>  

fled_cdev won't be NULL above, this is checked elsewhere in the V4L2 flash
LED class code. I guess the question then is whether doing such a check is
meaningful. It would require a bug elsewhere in the code to happen.

-- 
Kind regards,

Sakari Ailus