linux-kernel - Re: [PATCH] media: uvcvideo: Support borderline implementation of hw timestamping

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] [day] [month] [year] [list]

Date:   Fri, 20 Aug 2021 01:54:11 +0300
From:   Laurent Pinchart <laurent.pinchart@...asonboard.com>
To:     Ricardo Ribalda <ribalda@...omium.org>
Cc:     Mauro Carvalho Chehab <mchehab@...nel.org>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] media: uvcvideo: Support borderline implementation of hw
 timestamping

Hi Ricardo,

Thank you for the patch.

On Tue, Aug 17, 2021 at 06:12:02PM +0200, Ricardo Ribalda wrote:
> Some cameras do not set pts and src if there is no camera data in the
> buffer. They do this without clearing the PTS and SCR bits of the
> header. This is probably done due to a strict/borderline interpretation
> of the standard:
> 
> "STC must be captured when the first video data of a video frame is put
> on the USB bus."
> 
> Eg:
> buffer: 0xa7755c00 len 000012 header:0x8c stc 00000000 sof 0000 pts 00000000
> buffer: 0xa7755c00 len 000012 header:0x8c stc 00000000 sof 0000 pts 00000000
> buffer: 0xa7755c00 len 000668 header:0x8c stc 73779dba sof 070c pts 7376d37a
> 
> Support those cameras by only decoding the clock if there is camera data
> in the buffer.

Which cameras do you know to be affected ?

> Signed-off-by: Ricardo Ribalda <ribalda@...omium.org>
> ---
>  drivers/media/usb/uvc/uvc_video.c | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
> 
> diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
> index e16464606b14..6d0e474671a2 100644
> --- a/drivers/media/usb/uvc/uvc_video.c
> +++ b/drivers/media/usb/uvc/uvc_video.c
> @@ -490,6 +490,18 @@ uvc_video_clock_decode(struct uvc_streaming *stream, struct uvc_buffer *buf,
>  	if (len < header_size)
>  		return;
>  
> +	/*
> +	 * Some cameras when there is no camera data in a buffer, do not
> +	 * handle properly the pts and scr. This can be due to an borderline
> +	 * interpretation of the standard: "STC must be captured when the
> +	 * first video data of a video frame is put on the USB bus."
> +	 * Due to their internal design, their firmware cannot clear the
> +	 * UVC_STREAM_PTS and UVC_STREAM_SCR bits. Which forces us to use the
> +	 * length of the buffer to decide if pts and scr are valid or not.
> +	 */
> +	if (len == header_size)
> +		return;
> +

Won't this prevent using timestamp data provided by cameras that behave
properly ?

>  	/* Extract the timestamps:
>  	 *
>  	 * - store the frame PTS in the buffer structure

-- 
Regards,

Laurent Pinchart