linux-kernel - Re: uvcvideo: Unknown video format,00000032-0002-0010-8000-00aa00389b71

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: <1521603539.27691.5.camel@ndufresne.ca>
Date:   Tue, 20 Mar 2018 23:38:59 -0400
From:   Nicolas Dufresne <nicolas@...fresne.ca>
To:     Laurent Pinchart <laurent.pinchart@...asonboard.com>
Cc:     Paul Menzel <pmenzel+linux-media@...gen.mpg.de>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
        it+linux-media@...gen.mpg.de
Subject: Re: uvcvideo: Unknown video
 format,00000032-0002-0010-8000-00aa00389b71

Le mardi 20 mars 2018 à 20:04 +0200, Laurent Pinchart a écrit :
> Hi Nicolas,
> 
> On Tuesday, 20 March 2018 19:45:51 EET Nicolas Dufresne wrote:
> > Le mardi 20 mars 2018 à 13:20 +0100, Paul Menzel a écrit :
> > > Dear Linux folks,
> > > 
> > > 
> > > On the Dell XPS 13 9370, Linux 4.16-rc6 outputs the messages below.
> > > 
> > > ```
> > > [    2.338094] calling  uvc_init+0x0/0x1000 [uvcvideo] @ 295
> > > [    2.338569] calling  iTCO_wdt_init_module+0x0/0x1000 [iTCO_wdt] @ 280
> > > [    2.338570] iTCO_wdt: Intel TCO WatchDog Timer Driver v1.11
> > > [    2.338713] iTCO_wdt: Found a Intel PCH TCO device (Version=4,
> > > TCOBASE=0x0400)
> > > [    2.338755] uvcvideo: Found UVC 1.00 device Integrated_Webcam_HD
> > > (0bda:58f4)
> > > [    2.338827] iTCO_wdt: initialized. heartbeat=30 sec (nowayout=0)
> > > [    2.338851] initcall iTCO_wdt_init_module+0x0/0x1000 [iTCO_wdt]
> > > returned 0 after 271 usecs
> > > [    2.340669] uvcvideo 1-5:1.0: Entity type for entity Extension 4 was
> > > not initialized!
> > > [    2.340670] uvcvideo 1-5:1.0: Entity type for entity Extension 7 was
> > > not initialized!
> > > [    2.340672] uvcvideo 1-5:1.0: Entity type for entity Processing 2 was
> > > not initialized!
> > > [    2.340673] uvcvideo 1-5:1.0: Entity type for entity Camera 1 was not
> > > initialized!
> > > [    2.340736] input: Integrated_Webcam_HD: Integrate as
> > > /devices/pci0000:00/0000:00:14.0/usb1/1-5/1-5:1.0/input/input9
> > > [    2.341447] uvcvideo: Unknown video format
> > > 00000032-0002-0010-8000-00aa00389b71
> > 
> > While the 0002 is suspicious, this is pretty close to a color format.
> > I've recently come across of similar format using D3DFORMAT instead of
> > GUID. According to the vendor*, this camera module includes an infrared
> > camera (340x340), so I suspect this is to specify the format it
> > outputs. A good guess to start with would be that this is
> > D3DFMT_X8L8V8U8 (0x32).
> 
> Isn't 0x32 D3DFMT_L8, not D3DFMT_X8L8V8U8 ?

You are right, sorry about that, I totally miss-translate. It felt
weird. This is much more likely yes. So maybe it's the same mapping
(but with the -00002- instead) as what I added for the HoloLense
Camera.
> 
> > To test it, you could map this
> > V4L2_PIX_FMT_YUV32/xRGB and see if the driver is happy with the buffer
> > size.
> 
> VideoStreaming Interface Descriptor:
>         bLength                            30
>         bDescriptorType                    36
>         bDescriptorSubtype                  5 (FRAME_UNCOMPRESSED)
>         bFrameIndex                         1
>         bmCapabilities                   0x00
>           Still image unsupported
>         wWidth                            340
>         wHeight                           340
>         dwMinBitRate                 55488000
>         dwMaxBitRate                 55488000
>         dwMaxVideoFrameBufferSize      115600
>         dwDefaultFrameInterval         166666
>         bFrameIntervalType                  1
>         dwFrameInterval( 0)            166666
> 
> 340*340 is 115600, so this should be a 8-bit format.

Indeed, that matches.

> 
> > Then render it to make sure it looks some image of some sort. A
> > new format will need to be defined as this format is in the wrong
> > order, and is ambiguous (it may mean AYUV or xYUV). I'm not sure if we
> > need specific formats to differentiate infrared data from YUV images,
> > need to be discussed.
> 
> If the format is indeed D3DFMT_L8, it should map to V4L2_PIX_FMT_GREY (8-bit 
> luminance). I suspect the camera transmits a depth map though.

I wonder if we should think of a way to tell userspace this is fnfrared
data rather then black and white ?

> 
> > *https://dustinweb.azureedge.net/media/338953/xps-13-9370.pdf
> > 
> > > [    2.341450] uvcvideo: Found UVC 1.00 device Integrated_Webcam_HD
> > > (0bda:58f4)
> > > [    2.343371] uvcvideo: Unable to create debugfs 1-2 directory.
> > > [    2.343420] uvcvideo 1-5:1.2: Entity type for entity Extension 10 was
> > > not initialized!
> > > [    2.343422] uvcvideo 1-5:1.2: Entity type for entity Extension 12 was
> > > not initialized!
> > > [    2.343423] uvcvideo 1-5:1.2: Entity type for entity Processing 9 was
> > > not initialized!
> > > [    2.343424] uvcvideo 1-5:1.2: Entity type for entity Camera 11 was
> > > not initialized!
> > > [    2.343472] input: Integrated_Webcam_HD: Integrate as
> > > /devices/pci0000:00/0000:00:14.0/usb1/1-5/1-5:1.2/input/input10
> > > [    2.343496] usbcore: registered new interface driver uvcvideo
> > > [    2.343496] USB Video Class driver (1.1.1)
> > > [    2.343501] initcall uvc_init+0x0/0x1000 [uvcvideo] returned 0 after
> > > 5275 usecs
> > > ```
> > > 
> > > Please tell me, what I can do to improve the situation.
> 
>