linux-kernel - Re: duplicate requests on host side while streaming via uvcvideo gadget

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: <170731108881.252503.6282895768429052406@ping.linuxembedded.co.uk>
Date: Wed, 07 Feb 2024 13:04:48 +0000
From: Kieran Bingham <kieran.bingham@...asonboard.com>
To: Michael Grzeschik <mgr@...gutronix.de>, Thinh Nguyen <Thinh.Nguyen@...opsys.com>
Cc: linux-kernel@...r.kernel.org <linux-kernel@...r.kernel.org>, linux-usb@...r.kernel.org <linux-usb@...r.kernel.org>
Subject: Re: duplicate requests on host side while streaming via uvcvideo gadget

Hi Michael,

Quoting Michael Grzeschik (2024-02-06 21:30:17)
> On Tue, Feb 06, 2024 at 03:23:17AM +0000, Thinh Nguyen wrote:
> >On Tue, Feb 06, 2024, Michael Grzeschik wrote:
> >> Hi Thinh
> >>
> >> I found some strange situation while streaming via uvc-gadget to some
> >> usb host. It happens when some requests are missed due to higher load on
> >> the gadget machine. In some cases some requests will reach the host
> >> twice. In my special case, I added the following changes [1] for the
> >> host and gadget side.
> >
> >Does this only happen to some specific hosts?
> 
> >Are all the data of the duplicate requests matching or just some bits of
> >the transfer? Were you able to confirm from some usb analyzer/sniffer
> >that the data out the wire is actually duplicate?
> 
> Turns out, this duplicates are just misinterpretations.

I'm glad there's no deeper issue to worry about here.

> The linux uvcvideo driver will parse the uvc header payload twice. (If
> the FID was incremented inbetween). This led to those double misleading
> outputs. Although this means that there is a bug in
> uvc_video_stats_decode for incrementing the error count.

Do you plan/are you able to submit a patch to fix this?  Hopefully that
would prevent anyone else following the same rabbit-hole.

--
Kieran


> 
> Anyway, so just ignore this whole thread and be thanked for even having any
> thoghts on this.
> 
> Regards,
> Michael
> 
> -- 
> Pengutronix e.K.                           |                             |
> Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
> 31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
> Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |