em28xx issues

em28xx issues

[Paul Thomas]

Hello,

I'm trying to get a Empia EM2860 board working (specifically the "DVD
Maker USB 2.0"). Everything seems OK to start with, when I plug in the
board I get a reasonable dmesg (posted at the bottom).

I then try to use cheese or camstream and all I get is a black screen
or a black screen with a green stripe at the bottom. I've tested it
with a Windows box so I know there's nothing wrong with the video
input.

Are cheese & camstream the appropriate v4l2 frontends for testing
with. Is there something more low level I could test with?

I using Fedora 10, my kernel version is 2.6.27.9-159.fc10.i686.

Thanks in advance.

-Paul

usb 1-2: new high speed USB device using ehci_hcd and address 20
usb 1-2: configuration #1 chosen from 1 choice
em28xx new video device (eb1a:2860): interface 0, class 255
em28xx Doesn't have usb audio class
em28xx #0: Alternate settings: 8
em28xx #0: Alternate setting 0, max size= 0
em28xx #0: Alternate setting 1, max size= 0
em28xx #0: Alternate setting 2, max size= 1448
em28xx #0: Alternate setting 3, max size= 2048
em28xx #0: Alternate setting 4, max size= 2304
em28xx #0: Alternate setting 5, max size= 2580
em28xx #0: Alternate setting 6, max size= 2892
em28xx #0: Alternate setting 7, max size= 3072
em28xx #0: chip ID is em2860
saa7115' 1-0025: saa7113 found (1f7113d0e100000) @ 0x4a (em28xx #0)
em28xx #0: found i2c device @ 0x4a [saa7113h]
em28xx #0: Your board has no unique USB ID.
em28xx #0: A hint were successfully done, based on i2c devicelist hash.
em28xx #0: This method is not 100% failproof.
em28xx #0: If the board were missdetected, please email this log to:
em28xx #0: 	V4L Mailing List  <video4linux-list@redhat.com>
em28xx #0: Board detected as PointNix Intra-Oral Camera
em28xx #0: Registering snapshot button...
input: em28xx snapshot button as
/devices/pci0000:00/0000:00:1d.7/usb1/1-2/input/input16
em28xx #0: V4L2 device registered as /dev/video0 and /dev/vbi0
em28xx-audio.c: probing for em28x1 non standard usbaudio
em28xx-audio.c: Copyright (C) 2006 Markus Rechberger
em28xx #0: Found PointNix Intra-Oral Camera
usb 1-2: New USB device found, idVendor=eb1a, idProduct=2860
usb 1-2: New USB device strings: Mfr=0, Product=0, SerialNumber=0

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Markus & Douglas,

Thanks for your help. Do I need more than just the kernel headers
installed? Both the full v4l-dvb and em28xx-new start to compile, but
generate errors an error before finishing.

thanks,
Paul

On Wed, Dec 31, 2008 at 10:53 AM, Douglas Schilling Landgraf
<dougsland@gmail.com> wrote:
> Hello Paul,
>
>  I'm working to put SilverCrest webcam to work with em28xx module
> (the same module that yours work). I will keep you update when I finish
> it okay?
>
> Let me: Did you test with lastest v4l-dvb tree?
>
> If not, please remove your webcam from your computer and restart it.
>
> Download lastest:
>  hg clone http://linuxtv.org/hg/v4l-dvb
>
> Compile:
>
>  cd v4l-dvb
>  make
>  make install
>
> Plug your device again and try to use with your favorite application.
>
> Fell free to ping me or email-me okay?
>
> Thanks,
> Douglas
>
>
>
> On Tue, 30 Dec
> 2008 17:46:34 -0700 "Paul Thomas" <pthomas8589@gmail.com> wrote:
>
>> Hello,
>>
>> I'm trying to get a Empia EM2860 board working (specifically the "DVD
>> Maker USB 2.0"). Everything seems OK to start with, when I plug in the
>> board I get a reasonable dmesg (posted at the bottom).
>>
>> I then try to use cheese or camstream and all I get is a black screen
>> or a black screen with a green stripe at the bottom. I've tested it
>> with a Windows box so I know there's nothing wrong with the video
>> input.
>>
>> Are cheese & camstream the appropriate v4l2 frontends for testing
>> with. Is there something more low level I could test with?
>>
>> I using Fedora 10, my kernel version is 2.6.27.9-159.fc10.i686.
>>
>> Thanks in advance.
>>
>> -Paul
>>
>> usb 1-2: new high speed USB device using ehci_hcd and address 20
>> usb 1-2: configuration #1 chosen from 1 choice
>> em28xx new video device (eb1a:2860): interface 0, class 255
>> em28xx Doesn't have usb audio class
>> em28xx #0: Alternate settings: 8
>> em28xx #0: Alternate setting 0, max size= 0
>> em28xx #0: Alternate setting 1, max size= 0
>> em28xx #0: Alternate setting 2, max size= 1448
>> em28xx #0: Alternate setting 3, max size= 2048
>> em28xx #0: Alternate setting 4, max size= 2304
>> em28xx #0: Alternate setting 5, max size= 2580
>> em28xx #0: Alternate setting 6, max size= 2892
>> em28xx #0: Alternate setting 7, max size= 3072
>> em28xx #0: chip ID is em2860
>> saa7115' 1-0025: saa7113 found (1f7113d0e100000) @ 0x4a (em28xx #0)
>> em28xx #0: found i2c device @ 0x4a [saa7113h]
>> em28xx #0: Your board has no unique USB ID.
>> em28xx #0: A hint were successfully done, based on i2c devicelist
>> hash. em28xx #0: This method is not 100% failproof.
>> em28xx #0: If the board were missdetected, please email this log to:
>> em28xx #0:    V4L Mailing List  <video4linux-list@redhat.com>
>> em28xx #0: Board detected as PointNix Intra-Oral Camera
>> em28xx #0: Registering snapshot button...
>> input: em28xx snapshot button as
>> /devices/pci0000:00/0000:00:1d.7/usb1/1-2/input/input16
>> em28xx #0: V4L2 device registered as /dev/video0 and /dev/vbi0
>> em28xx-audio.c: probing for em28x1 non standard usbaudio
>> em28xx-audio.c: Copyright (C) 2006 Markus Rechberger
>> em28xx #0: Found PointNix Intra-Oral Camera
>> usb 1-2: New USB device found, idVendor=eb1a, idProduct=2860
>> usb 1-2: New USB device strings: Mfr=0, Product=0, SerialNumber=0
>>
>> --
>> video4linux-list mailing list
>> Unsubscribe
>> mailto:video4linux-list-request@redhat.com?subject=unsubscribe
>> https://www.redhat.com/mailman/listinfo/video4linux-list
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 2:28 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> Markus & Douglas,
>
> Thanks for your help. Do I need more than just the kernel headers
> installed? Both the full v4l-dvb and em28xx-new start to compile, but
> generate errors an error before finishing.
>
> thanks,
> Paul

Paul,

If you pastebin the output of attempting to compile v4l-dvb, I will
see if it's something obvious.

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Devin,

Here it is,

CC [M]  /home/raid5/kernels/v4l-dvb/v4l/bttv-input.o
In file included from /home/raid5/kernels/v4l-dvb/v4l/bttvp.h:36,
                 from /home/raid5/kernels/v4l-dvb/v4l/bttv-input.c:28:
include/linux/pci.h:1126: error: expected declaration specifiers or
'...' before '(' token
include/linux/pci.h:1126: error: expected declaration specifiers or
'...' before '(' token
include/linux/pci.h:1126: error: static declaration of
'ioremap_nocache' follows non-static declaration
include/asm/io_64.h:176: error: previous declaration of
'ioremap_nocache' was here
include/linux/pci.h: In function 'ioremap_nocache':
include/linux/pci.h:1127: error: number of arguments doesn't match prototype
include/asm/io_64.h:176: error: prototype declaration
include/linux/pci.h:1131: error: 'pdev' undeclared (first use in this function)
include/linux/pci.h:1131: error: (Each undeclared identifier is
reported only once
include/linux/pci.h:1131: error: for each function it appears in.)
include/linux/pci.h:1131: error: 'bar' undeclared (first use in this function)
make[3]: *** [/home/raid5/kernels/v4l-dvb/v4l/bttv-input.o] Error 1
make[2]: *** [_module_/home/raid5/kernels/v4l-dvb/v4l] Error 2
make[2]: Leaving directory `/usr/src/kernels/2.6.27.9-73.fc9.x86_64'
make[1]: *** [default] Error 2
make[1]: Leaving directory `/home/raid5/kernels/v4l-dvb/v4l'
make: *** [all] Error 2

thanks,
Paul

On Wed, Dec 31, 2008 at 12:33 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 2:28 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> Markus & Douglas,
>>
>> Thanks for your help. Do I need more than just the kernel headers
>> installed? Both the full v4l-dvb and em28xx-new start to compile, but
>> generate errors an error before finishing.
>>
>> thanks,
>> Paul
>
> Paul,
>
> If you pastebin the output of attempting to compile v4l-dvb, I will
> see if it's something obvious.
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 2:34 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> Devin,
>
> Here it is,
>
> CC [M]  /home/raid5/kernels/v4l-dvb/v4l/bttv-input.o
> In file included from /home/raid5/kernels/v4l-dvb/v4l/bttvp.h:36,
>                 from /home/raid5/kernels/v4l-dvb/v4l/bttv-input.c:28:
> include/linux/pci.h:1126: error: expected declaration specifiers or

Ok, you're running 2.6.27.9 I presume.

Please update to the latest version of v4l-dvb, as that issue should
have been fixed yesterday.

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

OK, I can compile now, but when I go to modprobe em28xx I get this error.

em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600

thanks,
Paul

On Wed, Dec 31, 2008 at 12:42 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 2:39 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> Devin,
>>
>> I just ran the "hg clone http://linuxtv.org/hg/v4l-dvb" command a few
>> minutes ago. Is there a different way to get the latest version?
>>
>> thanks,
>> Paul
>
> Hmmm...  No, that should have done it.
>
> As a workaround you can just edit v4l/compat.h and look
>
> To work around the issue, edit v4l/compat.h as follows.  Find the block:
>
> #if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 28)
> #define snd_BUG_ON(cond) ?? ?? ?? ??WARN((cond), "BUG? (%s)\n",
> __stringify(cond))
> #define pci_ioremap_bar(pci, a) ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? \
> ?? ?? ?? ?? ioremap_nocache(pci_resource_start(pci, a), ?? ??\
> ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? ?? pci_resource_len(pci, a))
> #endif
>
> and change the KERNEL_VERSION(2, 6, 28) to KERNEL_VERSION(2, 6, 27)
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>
> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>
> thanks,
> Paul

Do you have some mix of files from the em28xx-new and the v4l-dvb
codebase?  Or did you jam a reference in the include path to
em28xx-new's include directory?

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

The v4l-dvb directory is clean aside from the KERNEL_VERSION(2, 6, 27) change.

thanks,
Paul

On Wed, Dec 31, 2008 at 1:06 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>>
>> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>>
>> thanks,
>> Paul
>
> Do you have some mix of files from the em28xx-new and the v4l-dvb
> codebase?  Or did you jam a reference in the include path to
> em28xx-new's include directory?
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

OK, after a restart the driver loads properly. It's still not working,
but I have to look at a couple of things.

thanks,
Paul

On Wed, Dec 31, 2008 at 1:09 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> The v4l-dvb directory is clean aside from the KERNEL_VERSION(2, 6, 27) change.
>
> thanks,
> Paul
>
> On Wed, Dec 31, 2008 at 1:06 PM, Devin Heitmueller
> <devin.heitmueller@gmail.com> wrote:
>> On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>>>
>>> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>>>
>>> thanks,
>>> Paul
>>
>> Do you have some mix of files from the em28xx-new and the v4l-dvb
>> codebase?  Or did you jam a reference in the include path to
>> em28xx-new's include directory?
>>
>> Devin
>>
>> --
>> Devin J. Heitmueller
>> http://www.devinheitmueller.com
>> AIM: devinheitmueller
>>
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

When I start camstream with the new driver this is what I get. This is
a slightly different setup that I first described. It a x86_64 with
Fedora 9.

CCamWindow::CCamWindow()
CWebCamViewer::CWebCamViewer(0xfbf8c0, 0x0)
CVideoDevice::Init()
Using mmap(), VMBuf.size = 6651904
CVideoDevice::Init(): mmap() failed (22). Falling back to non-mmap()ed mode.
Allocating own buffer.
Trying to find video options for PointNix Intra-Oral Camera:/dev/video0
Creating new video options
<!DOCTYPE Configuration>
<config>
 <defaults/>
 <videodevices>
  <videoconfiguration name="PointNix Intra-Oral Camera" >
   <basename>snapshot</basename>
   <textfont>system</textfont>
   <textcolor>#ffff00</textcolor>
   <timeinimage>false</timeinimage>
   <fileformat>PNG</fileformat>
   <savetodisk>true</savetodisk>
   <ftptoserver>false</ftptoserver>
   <saveoption>1</saveoption>
   <maxsequence>1000</maxsequence>
   <sequence>0</sequence>
   <ftpserver></ftpserver>
   <ftppath></ftppath>
   <ftpuser></ftpuser>
   <ftppass></ftppass>
   <ftppassive>false</ftppassive>
   <ftpunique>true</ftpunique>
  </videoconfiguration>
 </videodevices>
</config>

CSnapshotSettingsDlg::CSnapshotSettingsDlg(...)
QFont::setRawName(): Invalid XLFD: "system"
CVideoSettingsDlg::SizeChanged(720x576)
CVideoSettingsDlg::FramerateChanged(10)
CCamPanel::SetSize(720x576)
CCamPanel::SetImageSize(720x576)
CCamPanel::SetVisibleSize(720x576)
CCamPanel::SetSize(720x576)
CCamPanel::SetImageSize(720x576)
CCamPanel::SetVisibleSize(720x576)
RecalcTotalViewSize: resize viewport(720x576)
EnableRGB: +
CVideoDevice::SetPalette picked palette 8 [yuyv]
CVideoDevice::CreateImagesRGB()
 allocating space for RGB
CVideoDevice::StartCapture() go!
Segmentation fault

On Wed, Dec 31, 2008 at 1:20 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> OK, after a restart the driver loads properly. It's still not working,
> but I have to look at a couple of things.
>
> thanks,
> Paul
>
> On Wed, Dec 31, 2008 at 1:09 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> The v4l-dvb directory is clean aside from the KERNEL_VERSION(2, 6, 27) change.
>>
>> thanks,
>> Paul
>>
>> On Wed, Dec 31, 2008 at 1:06 PM, Devin Heitmueller
>> <devin.heitmueller@gmail.com> wrote:
>>> On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>>>>
>>>> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>>>>
>>>> thanks,
>>>> Paul
>>>
>>> Do you have some mix of files from the em28xx-new and the v4l-dvb
>>> codebase?  Or did you jam a reference in the include path to
>>> em28xx-new's include directory?
>>>
>>> Devin
>>>
>>> --
>>> Devin J. Heitmueller
>>> http://www.devinheitmueller.com
>>> AIM: devinheitmueller
>>>
>>
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 3:58 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> When I start camstream with the new driver this is what I get. This is
> a slightly different setup that I first described. It a x86_64 with
> Fedora 9.
>
> CCamWindow::CCamWindow()
> CWebCamViewer::CWebCamViewer(0xfbf8c0, 0x0)
> CVideoDevice::Init()
> Using mmap(), VMBuf.size = 6651904
> CVideoDevice::Init(): mmap() failed (22). Falling back to non-mmap()ed mode.
> Allocating own buffer.
> Trying to find video options for PointNix Intra-Oral Camera:/dev/video0
> Creating new video options
> <!DOCTYPE Configuration>
> <config>
>  <defaults/>
>  <videodevices>
>  <videoconfiguration name="PointNix Intra-Oral Camera" >
>   <basename>snapshot</basename>
>   <textfont>system</textfont>
>   <textcolor>#ffff00</textcolor>
>   <timeinimage>false</timeinimage>
>   <fileformat>PNG</fileformat>
>   <savetodisk>true</savetodisk>
>   <ftptoserver>false</ftptoserver>
>   <saveoption>1</saveoption>
>   <maxsequence>1000</maxsequence>
>   <sequence>0</sequence>
>   <ftpserver></ftpserver>
>   <ftppath></ftppath>
>   <ftpuser></ftpuser>
>   <ftppass></ftppass>
>   <ftppassive>false</ftppassive>
>   <ftpunique>true</ftpunique>
>  </videoconfiguration>
>  </videodevices>
> </config>
>
> CSnapshotSettingsDlg::CSnapshotSettingsDlg(...)
> QFont::setRawName(): Invalid XLFD: "system"
> CVideoSettingsDlg::SizeChanged(720x576)
> CVideoSettingsDlg::FramerateChanged(10)
> CCamPanel::SetSize(720x576)
> CCamPanel::SetImageSize(720x576)
> CCamPanel::SetVisibleSize(720x576)
> CCamPanel::SetSize(720x576)
> CCamPanel::SetImageSize(720x576)
> CCamPanel::SetVisibleSize(720x576)
> RecalcTotalViewSize: resize viewport(720x576)
> EnableRGB: +
> CVideoDevice::SetPalette picked palette 8 [yuyv]
> CVideoDevice::CreateImagesRGB()
>  allocating space for RGB
> CVideoDevice::StartCapture() go!
> Segmentation fault
>
> On Wed, Dec 31, 2008 at 1:20 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> OK, after a restart the driver loads properly. It's still not working,
>> but I have to look at a couple of things.
>>
>> thanks,
>> Paul
>>
>> On Wed, Dec 31, 2008 at 1:09 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> The v4l-dvb directory is clean aside from the KERNEL_VERSION(2, 6, 27) change.
>>>
>>> thanks,
>>> Paul
>>>
>>> On Wed, Dec 31, 2008 at 1:06 PM, Devin Heitmueller
>>> <devin.heitmueller@gmail.com> wrote:
>>>> On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>>> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>>>>>
>>>>> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>>>>>
>>>>> thanks,
>>>>> Paul
>>>>
>>>> Do you have some mix of files from the em28xx-new and the v4l-dvb
>>>> codebase?  Or did you jam a reference in the include path to
>>>> em28xx-new's include directory?
>>>>
>>>> Devin
>>>>
>>>> --
>>>> Devin J. Heitmueller
>>>> http://www.devinheitmueller.com
>>>> AIM: devinheitmueller
>>>>
>>>
>>
>
> --
> video4linux-list mailing list
> Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/video4linux-list
>

Seems like the code that attempts to guess what hardware you have if
it uses the default USB ID resulted in it finding a device with the
same profile as the PointNix camera.

Was there anything in the dmesg output when the segfault occurred?

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Well I've got it to sort of work. If I use the skype and look at the
video preview then it seems to work. However if I close skype and
reopen it, it no longer works. I have to un-plug and plug the device
to get it to work again. And no other program works yet.

Here is the status of the other cam programs:
camstream                 -> seg fault
gstreamer-properties  -> seg fault
cheese                      -> uses test input
fswebcam                  -> black screen with green stripe

thanks,
Paul

On Wed, Dec 31, 2008 at 1:58 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> When I start camstream with the new driver this is what I get. This is
> a slightly different setup that I first described. It a x86_64 with
> Fedora 9.
>
> CCamWindow::CCamWindow()
> CWebCamViewer::CWebCamViewer(0xfbf8c0, 0x0)
> CVideoDevice::Init()
> Using mmap(), VMBuf.size = 6651904
> CVideoDevice::Init(): mmap() failed (22). Falling back to non-mmap()ed mode.
> Allocating own buffer.
> Trying to find video options for PointNix Intra-Oral Camera:/dev/video0
> Creating new video options
> <!DOCTYPE Configuration>
> <config>
>  <defaults/>
>  <videodevices>
>  <videoconfiguration name="PointNix Intra-Oral Camera" >
>   <basename>snapshot</basename>
>   <textfont>system</textfont>
>   <textcolor>#ffff00</textcolor>
>   <timeinimage>false</timeinimage>
>   <fileformat>PNG</fileformat>
>   <savetodisk>true</savetodisk>
>   <ftptoserver>false</ftptoserver>
>   <saveoption>1</saveoption>
>   <maxsequence>1000</maxsequence>
>   <sequence>0</sequence>
>   <ftpserver></ftpserver>
>   <ftppath></ftppath>
>   <ftpuser></ftpuser>
>   <ftppass></ftppass>
>   <ftppassive>false</ftppassive>
>   <ftpunique>true</ftpunique>
>  </videoconfiguration>
>  </videodevices>
> </config>
>
> CSnapshotSettingsDlg::CSnapshotSettingsDlg(...)
> QFont::setRawName(): Invalid XLFD: "system"
> CVideoSettingsDlg::SizeChanged(720x576)
> CVideoSettingsDlg::FramerateChanged(10)
> CCamPanel::SetSize(720x576)
> CCamPanel::SetImageSize(720x576)
> CCamPanel::SetVisibleSize(720x576)
> CCamPanel::SetSize(720x576)
> CCamPanel::SetImageSize(720x576)
> CCamPanel::SetVisibleSize(720x576)
> RecalcTotalViewSize: resize viewport(720x576)
> EnableRGB: +
> CVideoDevice::SetPalette picked palette 8 [yuyv]
> CVideoDevice::CreateImagesRGB()
>  allocating space for RGB
> CVideoDevice::StartCapture() go!
> Segmentation fault
>
> On Wed, Dec 31, 2008 at 1:20 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> OK, after a restart the driver loads properly. It's still not working,
>> but I have to look at a couple of things.
>>
>> thanks,
>> Paul
>>
>> On Wed, Dec 31, 2008 at 1:09 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> The v4l-dvb directory is clean aside from the KERNEL_VERSION(2, 6, 27) change.
>>>
>>> thanks,
>>> Paul
>>>
>>> On Wed, Dec 31, 2008 at 1:06 PM, Devin Heitmueller
>>> <devin.heitmueller@gmail.com> wrote:
>>>> On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>>> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>>>>>
>>>>> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>>>>>
>>>>> thanks,
>>>>> Paul
>>>>
>>>> Do you have some mix of files from the em28xx-new and the v4l-dvb
>>>> codebase?  Or did you jam a reference in the include path to
>>>> em28xx-new's include directory?
>>>>
>>>> Devin
>>>>
>>>> --
>>>> Devin J. Heitmueller
>>>> http://www.devinheitmueller.com
>>>> AIM: devinheitmueller
>>>>
>>>
>>
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Devin,

Thanks for your response. There isn't much in dmesg, this is all I get
camstream[5910]: segfault at 0 ip 000000000043ea76 sp 00007fff7724f908
error 6 in camstream[400000+5a000]

Yeah, it says "Your board has no unique USB ID" and I don't see
pointnix intra-oral camera anywhere else. How do I figure out what
configuration "DVD Maker USB 2.0" should use? And how to I tell em28xx
to use that configuration?

thanks,
Paul

On Wed, Dec 31, 2008 at 2:23 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 3:58 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> When I start camstream with the new driver this is what I get. This is
>> a slightly different setup that I first described. It a x86_64 with
>> Fedora 9.
>>
>> CCamWindow::CCamWindow()
>> CWebCamViewer::CWebCamViewer(0xfbf8c0, 0x0)
>> CVideoDevice::Init()
>> Using mmap(), VMBuf.size = 6651904
>> CVideoDevice::Init(): mmap() failed (22). Falling back to non-mmap()ed mode.
>> Allocating own buffer.
>> Trying to find video options for PointNix Intra-Oral Camera:/dev/video0
>> Creating new video options
>> <!DOCTYPE Configuration>
>> <config>
>>  <defaults/>
>>  <videodevices>
>>  <videoconfiguration name="PointNix Intra-Oral Camera" >
>>   <basename>snapshot</basename>
>>   <textfont>system</textfont>
>>   <textcolor>#ffff00</textcolor>
>>   <timeinimage>false</timeinimage>
>>   <fileformat>PNG</fileformat>
>>   <savetodisk>true</savetodisk>
>>   <ftptoserver>false</ftptoserver>
>>   <saveoption>1</saveoption>
>>   <maxsequence>1000</maxsequence>
>>   <sequence>0</sequence>
>>   <ftpserver></ftpserver>
>>   <ftppath></ftppath>
>>   <ftpuser></ftpuser>
>>   <ftppass></ftppass>
>>   <ftppassive>false</ftppassive>
>>   <ftpunique>true</ftpunique>
>>  </videoconfiguration>
>>  </videodevices>
>> </config>
>>
>> CSnapshotSettingsDlg::CSnapshotSettingsDlg(...)
>> QFont::setRawName(): Invalid XLFD: "system"
>> CVideoSettingsDlg::SizeChanged(720x576)
>> CVideoSettingsDlg::FramerateChanged(10)
>> CCamPanel::SetSize(720x576)
>> CCamPanel::SetImageSize(720x576)
>> CCamPanel::SetVisibleSize(720x576)
>> CCamPanel::SetSize(720x576)
>> CCamPanel::SetImageSize(720x576)
>> CCamPanel::SetVisibleSize(720x576)
>> RecalcTotalViewSize: resize viewport(720x576)
>> EnableRGB: +
>> CVideoDevice::SetPalette picked palette 8 [yuyv]
>> CVideoDevice::CreateImagesRGB()
>>  allocating space for RGB
>> CVideoDevice::StartCapture() go!
>> Segmentation fault
>>
>> On Wed, Dec 31, 2008 at 1:20 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> OK, after a restart the driver loads properly. It's still not working,
>>> but I have to look at a couple of things.
>>>
>>> thanks,
>>> Paul
>>>
>>> On Wed, Dec 31, 2008 at 1:09 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>> The v4l-dvb directory is clean aside from the KERNEL_VERSION(2, 6, 27) change.
>>>>
>>>> thanks,
>>>> Paul
>>>>
>>>> On Wed, Dec 31, 2008 at 1:06 PM, Devin Heitmueller
>>>> <devin.heitmueller@gmail.com> wrote:
>>>>> On Wed, Dec 31, 2008 at 3:03 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>>>> OK, I can compile now, but when I go to modprobe em28xx I get this error.
>>>>>>
>>>>>> em28xx: Unknown symbol ir_codes_ati_tv_wonder_hd_600
>>>>>>
>>>>>> thanks,
>>>>>> Paul
>>>>>
>>>>> Do you have some mix of files from the em28xx-new and the v4l-dvb
>>>>> codebase?  Or did you jam a reference in the include path to
>>>>> em28xx-new's include directory?
>>>>>
>>>>> Devin
>>>>>
>>>>> --
>>>>> Devin J. Heitmueller
>>>>> http://www.devinheitmueller.com
>>>>> AIM: devinheitmueller
>>>>>
>>>>
>>>
>>
>> --
>> video4linux-list mailing list
>> Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
>> https://www.redhat.com/mailman/listinfo/video4linux-list
>>
>
> Seems like the code that attempts to guess what hardware you have if
> it uses the default USB ID resulted in it finding a device with the
> same profile as the PointNix camera.
>
> Was there anything in the dmesg output when the segfault occurred?
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 4:30 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> Devin,
>
> Thanks for your response. There isn't much in dmesg, this is all I get
> camstream[5910]: segfault at 0 ip 000000000043ea76 sp 00007fff7724f908
> error 6 in camstream[400000+5a000]
>
> Yeah, it says "Your board has no unique USB ID" and I don't see
> pointnix intra-oral camera anywhere else. How do I figure out what
> configuration "DVD Maker USB 2.0" should use? And how to I tell em28xx
> to use that configuration?

When the device doesn't have it's own USB ID (using Empia's USB ID
instead), we use a heuristic based on either a hash of the eeprom or a
hash of the list of i2c devices.  It's not surprising that the i2c
hash would match in your case, since both devices only have a single
saa7113 and no tuner.

Could you please send the dmesg output, now that you have the latest
v4l-dvb code compiled, which should include the eeprom hash?  From
this, we can build a profile for your device that includes the proper
saa7113 configuration (in particular the different inputs your device
has).

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Here it is,

em28xx: New device @ 480 Mbps (eb1a:2860, interface 0, class 0)
em28xx #0: Identified as Unknown EM2750/28xx video grabber (card=1)
em28xx #0: chip ID is em2860
saa7115' 0-0025: saa7113 found (1f7113d0e100000) @ 0x4a (em28xx #0)
em28xx #0: board has no eeprom
em28xx #0: found i2c device @ 0x4a [saa7113h]
em28xx #0: Your board has no unique USB ID.
em28xx #0: A hint were successfully done, based on i2c devicelist hash.
em28xx #0: This method is not 100% failproof.
em28xx #0: If the board were missdetected, please email this log to:
em28xx #0:      V4L Mailing List  <video4linux-list@redhat.com>
em28xx #0: Board detected as PointNix Intra-Oral Camera
em28xx #0: Registering snapshot button...
input: em28xx snapshot button as
/devices/pci0000:00/0000:00:02.1/usb1/1-1/1-1.4/input/input15
em28xx #0: Config register raw data: 0x00
em28xx #0: No AC97 audio processor
em28xx #0: v4l2 driver version 0.1.1
em28xx #0: V4L2 device registered as /dev/video0 and /dev/vbi0
em28xx-audio.c: probing for em28x1 non standard usbaudio
em28xx-audio.c: Copyright (C) 2006 Markus Rechberger
usb 1-1.4: New USB device found, idVendor=eb1a, idProduct=2860
usb 1-1.4: New USB device strings: Mfr=0, Product=0, SerialNumber=0

thanks,
Paul

On Wed, Dec 31, 2008 at 3:20 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 4:30 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> Devin,
>>
>> Thanks for your response. There isn't much in dmesg, this is all I get
>> camstream[5910]: segfault at 0 ip 000000000043ea76 sp 00007fff7724f908
>> error 6 in camstream[400000+5a000]
>>
>> Yeah, it says "Your board has no unique USB ID" and I don't see
>> pointnix intra-oral camera anywhere else. How do I figure out what
>> configuration "DVD Maker USB 2.0" should use? And how to I tell em28xx
>> to use that configuration?
>
> When the device doesn't have it's own USB ID (using Empia's USB ID
> instead), we use a heuristic based on either a hash of the eeprom or a
> hash of the list of i2c devices.  It's not surprising that the i2c
> hash would match in your case, since both devices only have a single
> saa7113 and no tuner.
>
> Could you please send the dmesg output, now that you have the latest
> v4l-dvb code compiled, which should include the eeprom hash?  From
> this, we can build a profile for your device that includes the proper
> saa7113 configuration (in particular the different inputs your device
> has).
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 5:24 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> Here it is,
>
> em28xx: New device @ 480 Mbps (eb1a:2860, interface 0, class 0)
> em28xx #0: Identified as Unknown EM2750/28xx video grabber (card=1)
> em28xx #0: chip ID is em2860
> saa7115' 0-0025: saa7113 found (1f7113d0e100000) @ 0x4a (em28xx #0)
> em28xx #0: board has no eeprom
> em28xx #0: found i2c device @ 0x4a [saa7113h]
> em28xx #0: Your board has no unique USB ID.
> em28xx #0: A hint were successfully done, based on i2c devicelist hash.
> em28xx #0: This method is not 100% failproof.
> em28xx #0: If the board were missdetected, please email this log to:
> em28xx #0:      V4L Mailing List  <video4linux-list@redhat.com>
> em28xx #0: Board detected as PointNix Intra-Oral Camera
> em28xx #0: Registering snapshot button...
> input: em28xx snapshot button as
> /devices/pci0000:00/0000:00:02.1/usb1/1-1/1-1.4/input/input15
> em28xx #0: Config register raw data: 0x00
> em28xx #0: No AC97 audio processor
> em28xx #0: v4l2 driver version 0.1.1
> em28xx #0: V4L2 device registered as /dev/video0 and /dev/vbi0
> em28xx-audio.c: probing for em28x1 non standard usbaudio
> em28xx-audio.c: Copyright (C) 2006 Markus Rechberger
> usb 1-1.4: New USB device found, idVendor=eb1a, idProduct=2860
> usb 1-1.4: New USB device strings: Mfr=0, Product=0, SerialNumber=0

Yeah, that's going to be a tough one - they're using the default Empia
USB ID, they have no eeprom at all to use in a hash, and it does
indeed have an identical i2c device layout to the Pointnix camera.

I'm not sure how we're going to be able to tell the two apart, which
really is a problem since the Pointnix camera has different inputs on
the saa7113 than this device.

Mauro, any suggestions?

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

How was it working with Skype? Can we hard code the settings for testing?

thanks,
Paul

On Wed, Dec 31, 2008 at 3:35 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 5:24 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> Here it is,
>>
>> em28xx: New device @ 480 Mbps (eb1a:2860, interface 0, class 0)
>> em28xx #0: Identified as Unknown EM2750/28xx video grabber (card=1)
>> em28xx #0: chip ID is em2860
>> saa7115' 0-0025: saa7113 found (1f7113d0e100000) @ 0x4a (em28xx #0)
>> em28xx #0: board has no eeprom
>> em28xx #0: found i2c device @ 0x4a [saa7113h]
>> em28xx #0: Your board has no unique USB ID.
>> em28xx #0: A hint were successfully done, based on i2c devicelist hash.
>> em28xx #0: This method is not 100% failproof.
>> em28xx #0: If the board were missdetected, please email this log to:
>> em28xx #0:      V4L Mailing List  <video4linux-list@redhat.com>
>> em28xx #0: Board detected as PointNix Intra-Oral Camera
>> em28xx #0: Registering snapshot button...
>> input: em28xx snapshot button as
>> /devices/pci0000:00/0000:00:02.1/usb1/1-1/1-1.4/input/input15
>> em28xx #0: Config register raw data: 0x00
>> em28xx #0: No AC97 audio processor
>> em28xx #0: v4l2 driver version 0.1.1
>> em28xx #0: V4L2 device registered as /dev/video0 and /dev/vbi0
>> em28xx-audio.c: probing for em28x1 non standard usbaudio
>> em28xx-audio.c: Copyright (C) 2006 Markus Rechberger
>> usb 1-1.4: New USB device found, idVendor=eb1a, idProduct=2860
>> usb 1-1.4: New USB device strings: Mfr=0, Product=0, SerialNumber=0
>
> Yeah, that's going to be a tough one - they're using the default Empia
> USB ID, they have no eeprom at all to use in a hash, and it does
> indeed have an identical i2c device layout to the Pointnix camera.
>
> I'm not sure how we're going to be able to tell the two apart, which
> really is a problem since the Pointnix camera has different inputs on
> the saa7113 than this device.
>
> Mauro, any suggestions?
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> How was it working with Skype? Can we hard code the settings for testing?
>
> thanks,
> Paul

Well, regardless of the device profile issue, it's not clear why it is
segfaulting.  In theory, it should work but you should only get a
single input available instead of being able to pick between composite
and s-video.

I think with this device you should be able to set a modprobe option
for "debug=1" which might provide a bit more insight.

Try this.  Unplug the device, run "make unload", and then "modprobe
em28xx".  Then plug the device in.  You should start seeing more info
in the dmesg output.  Then start the app that segfaults and report the
dmesg contents.

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Thanks,

I have to run now, but I'll try this tomorrow.

On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> How was it working with Skype? Can we hard code the settings for testing?
>>
>> thanks,
>> Paul
>
> Well, regardless of the device profile issue, it's not clear why it is
> segfaulting.  In theory, it should work but you should only get a
> single input available instead of being able to pick between composite
> and s-video.
>
> I think with this device you should be able to set a modprobe option
> for "debug=1" which might provide a bit more insight.
>
> Try this.  Unplug the device, run "make unload", and then "modprobe
> em28xx".  Then plug the device in.  You should start seeing more info
> in the dmesg output.  Then start the app that segfaults and report the
> dmesg contents.
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

I get "em28xx: Unknown parameter `debug'" when I try "modprobe em28xx debug=1".

thanks,
Paul

On Wed, Dec 31, 2008 at 4:04 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> Thanks,
>
> I have to run now, but I'll try this tomorrow.
>
> On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
> <devin.heitmueller@gmail.com> wrote:
>> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> How was it working with Skype? Can we hard code the settings for testing?
>>>
>>> thanks,
>>> Paul
>>
>> Well, regardless of the device profile issue, it's not clear why it is
>> segfaulting.  In theory, it should work but you should only get a
>> single input available instead of being able to pick between composite
>> and s-video.
>>
>> I think with this device you should be able to set a modprobe option
>> for "debug=1" which might provide a bit more insight.
>>
>> Try this.  Unplug the device, run "make unload", and then "modprobe
>> em28xx".  Then plug the device in.  You should start seeing more info
>> in the dmesg output.  Then start the app that segfaults and report the
>> dmesg contents.
>>
>> Devin
>>
>> --
>> Devin J. Heitmueller
>> http://www.devinheitmueller.com
>> AIM: devinheitmueller
>>
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

Any thoughts on the debug parameter? Do I set the debug level a different way?

thanks,
Paul

On Thu, Jan 1, 2009 at 9:45 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
> I get "em28xx: Unknown parameter `debug'" when I try "modprobe em28xx debug=1".
>
> thanks,
> Paul
>
> On Wed, Dec 31, 2008 at 4:04 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> Thanks,
>>
>> I have to run now, but I'll try this tomorrow.
>>
>> On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
>> <devin.heitmueller@gmail.com> wrote:
>>> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>> How was it working with Skype? Can we hard code the settings for testing?
>>>>
>>>> thanks,
>>>> Paul
>>>
>>> Well, regardless of the device profile issue, it's not clear why it is
>>> segfaulting.  In theory, it should work but you should only get a
>>> single input available instead of being able to pick between composite
>>> and s-video.
>>>
>>> I think with this device you should be able to set a modprobe option
>>> for "debug=1" which might provide a bit more insight.
>>>
>>> Try this.  Unplug the device, run "make unload", and then "modprobe
>>> em28xx".  Then plug the device in.  You should start seeing more info
>>> in the dmesg output.  Then start the app that segfaults and report the
>>> dmesg contents.
>>>
>>> Devin
>>>
>>> --
>>> Devin J. Heitmueller
>>> http://www.devinheitmueller.com
>>> AIM: devinheitmueller
>>>
>>
>

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[CityK]

Paul Thomas wrote:
> Any thoughts on the debug parameter? Do I set the debug level a different way?
>
> thanks,
> Paul
>
> On Thu, Jan 1, 2009 at 9:45 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
>   
>> I get "em28xx: Unknown parameter `debug'" when I try "modprobe em28xx debug=1".
>>
>> thanks,
>> Paul

Use /sbin/modinfo to see relevant parameters for a module.  You'll
likely want core_debug

Also please stop top posting (rather, post below the text that you are
quoting).  Thanks

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
<devin.heitmueller@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> How was it working with Skype? Can we hard code the settings for testing?
>>
>> thanks,
>> Paul
>
> Well, regardless of the device profile issue, it's not clear why it is
> segfaulting.  In theory, it should work but you should only get a
> single input available instead of being able to pick between composite
> and s-video.
>
> I think with this device you should be able to set a modprobe option
> for "debug=1" which might provide a bit more insight.
>
> Try this.  Unplug the device, run "make unload", and then "modprobe
> em28xx".  Then plug the device in.  You should start seeing more info
> in the dmesg output.  Then start the app that segfaults and report the
> dmesg contents.
>
> Devin
>
> --
> Devin J. Heitmueller
> http://www.devinheitmueller.com
> AIM: devinheitmueller
>

Thanks CityK. I was able to enable core_debug=1. Here is what I have
in dmesg after I camstream seg faults:

Em28xx: Initialized (Em28xx Audio Extension) extension
em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
em28xx #0 em28xx_init_isoc :em28xx: called em28xx_prepare_isoc
em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
camstream[18436]: segfault at 0 ip 000000000043ea76 sp
00007fff1ca5d7a8 error 6 in camstream[400000+5a000]
em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)

thanks,
Paul

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

On Fri, Jan 2, 2009 at 9:50 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
> On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
> <devin.heitmueller@gmail.com> wrote:
>> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> How was it working with Skype? Can we hard code the settings for testing?
>>>
>>> thanks,
>>> Paul
>>
>> Well, regardless of the device profile issue, it's not clear why it is
>> segfaulting.  In theory, it should work but you should only get a
>> single input available instead of being able to pick between composite
>> and s-video.
>>
>> I think with this device you should be able to set a modprobe option
>> for "debug=1" which might provide a bit more insight.
>>
>> Try this.  Unplug the device, run "make unload", and then "modprobe
>> em28xx".  Then plug the device in.  You should start seeing more info
>> in the dmesg output.  Then start the app that segfaults and report the
>> dmesg contents.
>>
>> Devin
>>
>> --
>> Devin J. Heitmueller
>> http://www.devinheitmueller.com
>> AIM: devinheitmueller
>>
>
> Thanks CityK. I was able to enable core_debug=1. Here is what I have
> in dmesg after I camstream seg faults:
>
> Em28xx: Initialized (Em28xx Audio Extension) extension
> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
> em28xx #0 em28xx_init_isoc :em28xx: called em28xx_prepare_isoc
> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> camstream[18436]: segfault at 0 ip 000000000043ea76 sp
> 00007fff1ca5d7a8 error 6 in camstream[400000+5a000]
> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>
> thanks,
> Paul
>

Here is an update on the programs working with em28xx

Working programs
Skype
UCview

Here is the status of the other cam programs that aren't working yet:
camstream                 -> seg fault
gstreamer-properties  -> seg fault
cheese                      -> uses test input
fswebcam                  -> black screen with green stripe

I hope this helps

thanks,
Paul

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

On Sat, Jan 3, 2009 at 10:22 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
> On Fri, Jan 2, 2009 at 9:50 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
>> <devin.heitmueller@gmail.com> wrote:
>>> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>> How was it working with Skype? Can we hard code the settings for testing?
>>>>
>>>> thanks,
>>>> Paul
>>>
>>> Well, regardless of the device profile issue, it's not clear why it is
>>> segfaulting.  In theory, it should work but you should only get a
>>> single input available instead of being able to pick between composite
>>> and s-video.
>>>
>>> I think with this device you should be able to set a modprobe option
>>> for "debug=1" which might provide a bit more insight.
>>>
>>> Try this.  Unplug the device, run "make unload", and then "modprobe
>>> em28xx".  Then plug the device in.  You should start seeing more info
>>> in the dmesg output.  Then start the app that segfaults and report the
>>> dmesg contents.
>>>
>>> Devin
>>>
>>> --
>>> Devin J. Heitmueller
>>> http://www.devinheitmueller.com
>>> AIM: devinheitmueller
>>>
>>
>> Thanks CityK. I was able to enable core_debug=1. Here is what I have
>> in dmesg after I camstream seg faults:
>>
>> Em28xx: Initialized (Em28xx Audio Extension) extension
>> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
>> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
>> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
>> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
>> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
>> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
>> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
>> em28xx #0 em28xx_init_isoc :em28xx: called em28xx_prepare_isoc
>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>> camstream[18436]: segfault at 0 ip 000000000043ea76 sp
>> 00007fff1ca5d7a8 error 6 in camstream[400000+5a000]
>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>>
>> thanks,
>> Paul
>>
>
> Here is an update on the programs working with em28xx
>
> Working programs
> Skype
> UCview
>
> Here is the status of the other cam programs that aren't working yet:
> camstream                 -> seg fault
> gstreamer-properties  -> seg fault
> cheese                      -> uses test input
> fswebcam                  -> black screen with green stripe
>
> I hope this helps
>
> thanks,
> Paul
>

OK, I was able to make some more progress. I can use fswebcam on my
x86_64 box if I comment out the input selection section. Now I'm
trying to get fswebcam to work on my embedded arm board
(http://opencircuits.com/Linuxstamp), but when I run it I get a kernel
oops. I've included the oops below. I'm now using this
(git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git)
git tree.

Any ideas?

thanks,
Paul

linuxstamp:/home/user/fswebcam-20070108# ./fswebcam -v test.jpeg
--- Opening /dev/video0...
Trying source module v4l2...
/dev/video0 opened.
src_v4l2_get_capability,81: /dev/video0 information:
src_v4l2_get_capability,82: cap.driver: "em28xx"
src_v4l2_get_capability,83: cap.card: "PointNix Intra-Oral Camera"
src_v4l2_get_capability,84: cap.bus_info: "1-1.7"
src_v4l2_get_capability,85: cap.capabilities=0x05020041
src_v4l2_get_capability,86: - VIDEO_CAPTURE
src_v4l2_get_capability,93: - AUDIO
src_v4l2_get_capability,95: - READWRITE
src_v4l2_get_capability,97: - STREAMING
src_v4l2_set_pix_format,530: Device offers the following V4L2 pixel formats:
src_v4l2_set_pix_format,542: 0: YUYV (16bpp YUY2, 4:2:2, packed)
Using palette YUYV
src_v4l2_set_mmap,648: mmap information:
src_v4l2_set_mmap,649: frames=4
src_v4l2_set_mmap,694: 0 length=221184
src_v4l2_set_mmap,694: 1 length=221184
src_v4l2_set_mmap,694: 2 length=221184
src_v4l2_set_mmap,694: 3 length=221184
kernel BUG at arch/arm/mm/dma-mapping.c:495!
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = c13c4000
[00000000] *pgd=21311031, *pte=00000000, *ppte=00000000
Internal error: Oops: 817 [#1] PREEMPT
Modules linked in: saa7115 v4l2_common em28xx videodev v4l1_compat
videobuf_vmalloc videobuf_core ir_common tveeprom
CPU: 0    Not tainted  (2.6.28-15476-g7d3b56b #2)
PC is at __bug+0x20/0x2c
LR is at vprintk+0x2f0/0x34c
pc : [<c00265c0>]    lr : [<c003b078>]    psr: 60000013
sp : c1ff1ba4  ip : c03944fc  fp : c1ff1bb0
r10: 00000028  r9 : c1d78628  r8 : c1d1f000
r7 : 00000020  r6 : 00000000  r5 : c1d49800  r4 : ffc09000
r3 : 00000000  r2 : c1ff0000  r1 : 00000002  r0 : 00000030
Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
Control: c000717f  Table: 213c4000  DAC: 00000015
Process fswebcam (pid: 2383, stack limit = 0xc1ff0260)
Stack: (0xc1ff1ba4 to 0xc1ff2000)
1ba0:          c1ff1bc0 c1ff1bb4 c0027504 c00265b0 c1ff1c84 c1ff1bc4
c01e2a40
1bc0: c0027470 c1ff1bf4 c1ff1bd4 c0053810 c003360c 00000000 c037001c
c1c713c0
1be0: 00000017 c1ff0000 c1ff1c08 c1ff1c08 c1ff1bfc c02bd9b4 c0033704
c1ff1c34
1c00: c1ff1c0c c02bb9a8 c003360c ffffdd0e c1ff0000 00000002 c1ff1c38
c0376d64
1c20: ffffdd0e c1ff0000 c1ff1c70 c1ff1c38 c02bc0b4 c0043c5c 00000000
00200200
1c40: ffffdd0e c00443ec c1c713c0 c0398b80 00000001 00000001 c1d78000
00000001
1c60: c1dd9400 00000020 00000000 00008200 c1d78628 00000028 c1ff1cac
c1ff1c88
1c80: c01e4380 c01e2904 c1ff1c94 bf036528 c004464c c1dbd3c0 c1e31400
c1d78000
1ca0: c1ff1cd8 c1ff1cb0 bf036ed0 c01e40ec c1d78000 c1d7cbe0 00000000
c1d379cc
1cc0: bf0150a0 bf038098 c1ff1e2c c1ff1cfc c1ff1cdc bf031904 bf036c68
bf031000
1ce0: ffffffea c1d7cbe0 c1d379cc c1ff1e2c c1ff1d20 c1ff1d00 bf011ff8
bf031850
1d00: c1d379c0 c1ff1e2c c044560f c044560f 00000000 c1ff1d38 c1ff1d24
bf032b98
1d20: bf011d44 bf032b68 000223d0 c1ff1e18 c1ff1d3c bf0244c0 bf032b78 00000001
1d40: 00000000 60000013 c0398a4c c03989e0 60000013 c1d49000 00000000 c1d490d8
1d60: 00000001 00000001 00000000 c1d379c0 c1e0be00 c044560f c1e50ce0 c1ff1d88
1d80: c003624c c003360c 00000000 c1d49000 00000000 c0398a70 00000001 00000009
1da0: c1ff1dc4 c1ff1db0 c0191088 c0036204 c1ff0000 00000000 c1ff1dd4 c1ff1dc8
1dc0: c01a7720 c019103c c1ff1df0 c1ff1df0 c1ff1ddc c003f048 c003360c c1ff0000
1de0: 00000102 c1ff1e20 c1ff1df4 c003f150 00000000 000223d0 fffffff2 00000000
1e00: 00000000 c044560f 00000000 c1ff1ed4 c1ff1e1c bf026b68 bf023128 00000003
1e20: 00000000 c1ff1e2c c1e50ce0 00000000 00000001 00000000 00000000 00000000
1e40: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001
1e60: 00000000 00000000 00000000 00000000 c1ff1e7c c01a76b4 c01aac20 c1ff1ea0
1e80: c1ff1e8c c01a76f0 c003360c c1ff1eb0 c1ff1e9c c004f4ac c003360c 00000004
1ea0: c1d49000 c1ff1f10 c0385478 c044560f 000223d0 c1e50ce0 c044560f c0022fe4
1ec0: c1ff0000 00018d94 c1ff1ef0 c1ff1ed8 bf0222e4 bf0269c4 c1e50ce0 ffffffe7
1ee0: 000223d0 c1ff1f10 c1ff1ef4 c00993dc bf0222ac c1e50ce0 c1f235a0 00000003
1f00: 000223d0 c1ff1f84 c1ff1f14 c0099d18 c009936c 00000000 c1ff1f34 c1ff1f28
1f20: c01a7720 c019103c c1ff1f50 c1ff1f50 c1ff1f3c c003f048 c003360c c1ff0000
1f40: 00000102 c1ff1f80 c1ff1f54 c003f150 c003efc4 00000000 00000001 c1e50ce0
1f60: fffffff7 c044560f 00000036 c0022fe4 c1ff0000 00018d94 c1ff1fa4 c1ff1f88
1f80: c0099d80 c0099858 00000000 00000000 00000004 00022280 00000000 c1ff1fa8
1fa0: c0022e40 c0099d50 00000000 00000004 00000003 c044560f 000223d0 00022414
1fc0: 00000000 00000004 00022280 000223d0 00000001 0000000c 00018d94 000220a0
1fe0: 00021540 be81e728 00013e68 40129424 80000010 00000003 00000000 00000000
Backtrace:
[<c00265a0>] (__bug+0x0/0x2c) from [<c0027504>] (dma_cache_maint+0xa4/0xb8)
[<c0027460>] (dma_cache_maint+0x0/0xb8) from [<c01e2a40>]
(usb_hcd_submit_urb+0x14c/0x890)
[<c01e28f4>] (usb_hcd_submit_urb+0x0/0x890) from [<c01e4380>]
(usb_submit_urb+0x2a4/0x2ac)
[<c01e40dc>] (usb_submit_urb+0x0/0x2ac) from [<bf036ed0>]
(em28xx_init_isoc+0x278/0x2c4 [em28xx])
 r6:c1d78000 r5:c1e31400 r4:c1dbd3c0
[<bf036c58>] (em28xx_init_isoc+0x0/0x2c4 [em28xx]) from [<bf031904>]
(buffer_prepare+0xc4/0xf4 [em28xx])
[<bf031840>] (buffer_prepare+0x0/0xf4 [em28xx]) from [<bf011ff8>]
(videobuf_qbuf+0x2c4/0x400 [videobuf_core])
 r7:c1ff1e2c r6:c1d379cc r5:c1d7cbe0 r4:ffffffea
[<bf011d34>] (videobuf_qbuf+0x0/0x400 [videobuf_core]) from
[<bf032b98>] (vidioc_qbuf+0x30/0x3c [em28xx])
 r8:00000000 r7:c044560f r6:c044560f r5:c1ff1e2c r4:c1d379c0
[<bf032b68>] (vidioc_qbuf+0x0/0x3c [em28xx]) from [<bf0244c0>]
(__video_do_ioctl+0x13a8/0x389c [videodev])
 r5:000223d0 r4:bf032b68
[<bf023118>] (__video_do_ioctl+0x0/0x389c [videodev]) from
[<bf026b68>] (video_ioctl2+0x1b4/0x2a4 [videodev])
[<bf0269b4>] (video_ioctl2+0x0/0x2a4 [videodev]) from [<bf0222e4>]
(v4l2_ioctl+0x48/0x54 [videodev])
[<bf02229c>] (v4l2_ioctl+0x0/0x54 [videodev]) from [<c00993dc>]
(vfs_ioctl+0x80/0x94)
 r6:000223d0 r5:ffffffe7 r4:c1e50ce0
[<c009935c>] (vfs_ioctl+0x0/0x94) from [<c0099d18>] (do_vfs_ioctl+0x4d0/0x4f8)
 r7:000223d0 r6:00000003 r5:c1f235a0 r4:c1e50ce0
[<c0099848>] (do_vfs_ioctl+0x0/0x4f8) from [<c0099d80>] (sys_ioctl+0x40/0x5c)
[<c0099d40>] (sys_ioctl+0x0/0x5c) from [<c0022e40>] (ret_fast_syscall+0x0/0x2c)
 r6:00022280 r5:00000004 r4:00000000
Code: e1a01000 e59f000c eb0052c5 e3a03000 (e5833000)
---[ end trace 68fb0b4324720f98 ]---

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

On Mon, Jan 5, 2009 at 5:33 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> On Sat, Jan 3, 2009 at 10:22 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
>> On Fri, Jan 2, 2009 at 9:50 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>> On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
>>> <devin.heitmueller@gmail.com> wrote:
>>>> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
>>>>> How was it working with Skype? Can we hard code the settings for testing?
>>>>>
>>>>> thanks,
>>>>> Paul
>>>>
>>>> Well, regardless of the device profile issue, it's not clear why it is
>>>> segfaulting.  In theory, it should work but you should only get a
>>>> single input available instead of being able to pick between composite
>>>> and s-video.
>>>>
>>>> I think with this device you should be able to set a modprobe option
>>>> for "debug=1" which might provide a bit more insight.
>>>>
>>>> Try this.  Unplug the device, run "make unload", and then "modprobe
>>>> em28xx".  Then plug the device in.  You should start seeing more info
>>>> in the dmesg output.  Then start the app that segfaults and report the
>>>> dmesg contents.
>>>>
>>>> Devin
>>>>
>>>> --
>>>> Devin J. Heitmueller
>>>> http://www.devinheitmueller.com
>>>> AIM: devinheitmueller
>>>>
>>>
>>> Thanks CityK. I was able to enable core_debug=1. Here is what I have
>>> in dmesg after I camstream seg faults:
>>>
>>> Em28xx: Initialized (Em28xx Audio Extension) extension
>>> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
>>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>>> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
>>> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
>>> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
>>> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
>>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>>> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
>>> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
>>> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
>>> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
>>> em28xx #0 em28xx_init_isoc :em28xx: called em28xx_prepare_isoc
>>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>>> camstream[18436]: segfault at 0 ip 000000000043ea76 sp
>>> 00007fff1ca5d7a8 error 6 in camstream[400000+5a000]
>>> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
>>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>>> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
>>>
>>> thanks,
>>> Paul
>>>
>>
>> Here is an update on the programs working with em28xx
>>
>> Working programs
>> Skype
>> UCview
>>
>> Here is the status of the other cam programs that aren't working yet:
>> camstream                 -> seg fault
>> gstreamer-properties  -> seg fault
>> cheese                      -> uses test input
>> fswebcam                  -> black screen with green stripe
>>
>> I hope this helps
>>
>> thanks,
>> Paul
>>
>
> OK, I was able to make some more progress. I can use fswebcam on my
> x86_64 box if I comment out the input selection section. Now I'm
> trying to get fswebcam to work on my embedded arm board
> (http://opencircuits.com/Linuxstamp), but when I run it I get a kernel
> oops. I've included the oops below. I'm now using this
> (git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git)
> git tree.
>
> Any ideas?
>
> thanks,
> Paul
>
> linuxstamp:/home/user/fswebcam-20070108# ./fswebcam -v test.jpeg
> --- Opening /dev/video0...
> Trying source module v4l2...
> /dev/video0 opened.
> src_v4l2_get_capability,81: /dev/video0 information:
> src_v4l2_get_capability,82: cap.driver: "em28xx"
> src_v4l2_get_capability,83: cap.card: "PointNix Intra-Oral Camera"
> src_v4l2_get_capability,84: cap.bus_info: "1-1.7"
> src_v4l2_get_capability,85: cap.capabilities=0x05020041
> src_v4l2_get_capability,86: - VIDEO_CAPTURE
> src_v4l2_get_capability,93: - AUDIO
> src_v4l2_get_capability,95: - READWRITE
> src_v4l2_get_capability,97: - STREAMING
> src_v4l2_set_pix_format,530: Device offers the following V4L2 pixel formats:
> src_v4l2_set_pix_format,542: 0: YUYV (16bpp YUY2, 4:2:2, packed)
> Using palette YUYV
> src_v4l2_set_mmap,648: mmap information:
> src_v4l2_set_mmap,649: frames=4
> src_v4l2_set_mmap,694: 0 length=221184
> src_v4l2_set_mmap,694: 1 length=221184
> src_v4l2_set_mmap,694: 2 length=221184
> src_v4l2_set_mmap,694: 3 length=221184
> kernel BUG at arch/arm/mm/dma-mapping.c:495!
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
> pgd = c13c4000
> [00000000] *pgd=21311031, *pte=00000000, *ppte=00000000
> Internal error: Oops: 817 [#1] PREEMPT
> Modules linked in: saa7115 v4l2_common em28xx videodev v4l1_compat
> videobuf_vmalloc videobuf_core ir_common tveeprom
> CPU: 0    Not tainted  (2.6.28-15476-g7d3b56b #2)
> PC is at __bug+0x20/0x2c
> LR is at vprintk+0x2f0/0x34c
> pc : [<c00265c0>]    lr : [<c003b078>]    psr: 60000013
> sp : c1ff1ba4  ip : c03944fc  fp : c1ff1bb0
> r10: 00000028  r9 : c1d78628  r8 : c1d1f000
> r7 : 00000020  r6 : 00000000  r5 : c1d49800  r4 : ffc09000
> r3 : 00000000  r2 : c1ff0000  r1 : 00000002  r0 : 00000030
> Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
> Control: c000717f  Table: 213c4000  DAC: 00000015
> Process fswebcam (pid: 2383, stack limit = 0xc1ff0260)
> Stack: (0xc1ff1ba4 to 0xc1ff2000)
> 1ba0:          c1ff1bc0 c1ff1bb4 c0027504 c00265b0 c1ff1c84 c1ff1bc4
> c01e2a40
> 1bc0: c0027470 c1ff1bf4 c1ff1bd4 c0053810 c003360c 00000000 c037001c
> c1c713c0
> 1be0: 00000017 c1ff0000 c1ff1c08 c1ff1c08 c1ff1bfc c02bd9b4 c0033704
> c1ff1c34
> 1c00: c1ff1c0c c02bb9a8 c003360c ffffdd0e c1ff0000 00000002 c1ff1c38
> c0376d64
> 1c20: ffffdd0e c1ff0000 c1ff1c70 c1ff1c38 c02bc0b4 c0043c5c 00000000
> 00200200
> 1c40: ffffdd0e c00443ec c1c713c0 c0398b80 00000001 00000001 c1d78000
> 00000001
> 1c60: c1dd9400 00000020 00000000 00008200 c1d78628 00000028 c1ff1cac
> c1ff1c88
> 1c80: c01e4380 c01e2904 c1ff1c94 bf036528 c004464c c1dbd3c0 c1e31400
> c1d78000
> 1ca0: c1ff1cd8 c1ff1cb0 bf036ed0 c01e40ec c1d78000 c1d7cbe0 00000000
> c1d379cc
> 1cc0: bf0150a0 bf038098 c1ff1e2c c1ff1cfc c1ff1cdc bf031904 bf036c68
> bf031000
> 1ce0: ffffffea c1d7cbe0 c1d379cc c1ff1e2c c1ff1d20 c1ff1d00 bf011ff8
> bf031850
> 1d00: c1d379c0 c1ff1e2c c044560f c044560f 00000000 c1ff1d38 c1ff1d24
> bf032b98
> 1d20: bf011d44 bf032b68 000223d0 c1ff1e18 c1ff1d3c bf0244c0 bf032b78 00000001
> 1d40: 00000000 60000013 c0398a4c c03989e0 60000013 c1d49000 00000000 c1d490d8
> 1d60: 00000001 00000001 00000000 c1d379c0 c1e0be00 c044560f c1e50ce0 c1ff1d88
> 1d80: c003624c c003360c 00000000 c1d49000 00000000 c0398a70 00000001 00000009
> 1da0: c1ff1dc4 c1ff1db0 c0191088 c0036204 c1ff0000 00000000 c1ff1dd4 c1ff1dc8
> 1dc0: c01a7720 c019103c c1ff1df0 c1ff1df0 c1ff1ddc c003f048 c003360c c1ff0000
> 1de0: 00000102 c1ff1e20 c1ff1df4 c003f150 00000000 000223d0 fffffff2 00000000
> 1e00: 00000000 c044560f 00000000 c1ff1ed4 c1ff1e1c bf026b68 bf023128 00000003
> 1e20: 00000000 c1ff1e2c c1e50ce0 00000000 00000001 00000000 00000000 00000000
> 1e40: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001
> 1e60: 00000000 00000000 00000000 00000000 c1ff1e7c c01a76b4 c01aac20 c1ff1ea0
> 1e80: c1ff1e8c c01a76f0 c003360c c1ff1eb0 c1ff1e9c c004f4ac c003360c 00000004
> 1ea0: c1d49000 c1ff1f10 c0385478 c044560f 000223d0 c1e50ce0 c044560f c0022fe4
> 1ec0: c1ff0000 00018d94 c1ff1ef0 c1ff1ed8 bf0222e4 bf0269c4 c1e50ce0 ffffffe7
> 1ee0: 000223d0 c1ff1f10 c1ff1ef4 c00993dc bf0222ac c1e50ce0 c1f235a0 00000003
> 1f00: 000223d0 c1ff1f84 c1ff1f14 c0099d18 c009936c 00000000 c1ff1f34 c1ff1f28
> 1f20: c01a7720 c019103c c1ff1f50 c1ff1f50 c1ff1f3c c003f048 c003360c c1ff0000
> 1f40: 00000102 c1ff1f80 c1ff1f54 c003f150 c003efc4 00000000 00000001 c1e50ce0
> 1f60: fffffff7 c044560f 00000036 c0022fe4 c1ff0000 00018d94 c1ff1fa4 c1ff1f88
> 1f80: c0099d80 c0099858 00000000 00000000 00000004 00022280 00000000 c1ff1fa8
> 1fa0: c0022e40 c0099d50 00000000 00000004 00000003 c044560f 000223d0 00022414
> 1fc0: 00000000 00000004 00022280 000223d0 00000001 0000000c 00018d94 000220a0
> 1fe0: 00021540 be81e728 00013e68 40129424 80000010 00000003 00000000 00000000
> Backtrace:
> [<c00265a0>] (__bug+0x0/0x2c) from [<c0027504>] (dma_cache_maint+0xa4/0xb8)
> [<c0027460>] (dma_cache_maint+0x0/0xb8) from [<c01e2a40>]
> (usb_hcd_submit_urb+0x14c/0x890)
> [<c01e28f4>] (usb_hcd_submit_urb+0x0/0x890) from [<c01e4380>]
> (usb_submit_urb+0x2a4/0x2ac)
> [<c01e40dc>] (usb_submit_urb+0x0/0x2ac) from [<bf036ed0>]
> (em28xx_init_isoc+0x278/0x2c4 [em28xx])
>  r6:c1d78000 r5:c1e31400 r4:c1dbd3c0
> [<bf036c58>] (em28xx_init_isoc+0x0/0x2c4 [em28xx]) from [<bf031904>]
> (buffer_prepare+0xc4/0xf4 [em28xx])
> [<bf031840>] (buffer_prepare+0x0/0xf4 [em28xx]) from [<bf011ff8>]
> (videobuf_qbuf+0x2c4/0x400 [videobuf_core])
>  r7:c1ff1e2c r6:c1d379cc r5:c1d7cbe0 r4:ffffffea
> [<bf011d34>] (videobuf_qbuf+0x0/0x400 [videobuf_core]) from
> [<bf032b98>] (vidioc_qbuf+0x30/0x3c [em28xx])
>  r8:00000000 r7:c044560f r6:c044560f r5:c1ff1e2c r4:c1d379c0
> [<bf032b68>] (vidioc_qbuf+0x0/0x3c [em28xx]) from [<bf0244c0>]
> (__video_do_ioctl+0x13a8/0x389c [videodev])
>  r5:000223d0 r4:bf032b68
> [<bf023118>] (__video_do_ioctl+0x0/0x389c [videodev]) from
> [<bf026b68>] (video_ioctl2+0x1b4/0x2a4 [videodev])
> [<bf0269b4>] (video_ioctl2+0x0/0x2a4 [videodev]) from [<bf0222e4>]
> (v4l2_ioctl+0x48/0x54 [videodev])
> [<bf02229c>] (v4l2_ioctl+0x0/0x54 [videodev]) from [<c00993dc>]
> (vfs_ioctl+0x80/0x94)
>  r6:000223d0 r5:ffffffe7 r4:c1e50ce0
> [<c009935c>] (vfs_ioctl+0x0/0x94) from [<c0099d18>] (do_vfs_ioctl+0x4d0/0x4f8)
>  r7:000223d0 r6:00000003 r5:c1f235a0 r4:c1e50ce0
> [<c0099848>] (do_vfs_ioctl+0x0/0x4f8) from [<c0099d80>] (sys_ioctl+0x40/0x5c)
> [<c0099d40>] (sys_ioctl+0x0/0x5c) from [<c0022e40>] (ret_fast_syscall+0x0/0x2c)
>  r6:00022280 r5:00000004 r4:00000000
> Code: e1a01000 e59f000c eb0052c5 e3a03000 (e5833000)
> ---[ end trace 68fb0b4324720f98 ]---
>

I thought a kernel oops was a big deal? Anyway, if the em28 device
isn't well supported. What would be a well supported device that has
an component input and preferably has been tested on ARM.

thanks,
Paul

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

On Wed, Jan 7, 2009 at 12:27 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> I thought a kernel oops was a big deal? Anyway, if the em28 device
> isn't well supported. What would be a well supported device that has
> an component input and preferably has been tested on ARM.
>
> thanks,
> Paul

Hello Paul,

I don't think anyone is saying that a kernel oops isn't a big deal.
The reality though is that this is a project worked on by a very small
number of *volunteers* and there is no assertion that your issue will
be dealt with as quickly as you like.

This issue is compounded by the fact that as far as we know it only
occurs on a single piece of hardware that nobody ever claimed was
supported and no developer has in their possession to debug the issue
with.  In cases like this the issue it may only take two or three
hours to debug if someone like myself has the device in question, but
I'm not going to spend my own money to go out and buy one.

By all means, if you want to debug the issue, we would be happy to
accept patches.  Or if you want to mail me the hardware so I could
take a look and see if I can debug the issue, that is something else
we can discuss.  But operating under the assumption that you are
somehow entitled to have a developer donate his time to address your
problem is perhaps not the best way to get help.

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Devin Heitmueller]

A quick look at the code does show something interesting:

There are a number of cases where we dereference the result of the
"INPUT" macro as follows without checking the number of inputs
defined:

route.input = INPUT(index)->vmux;

and here is the macro definition:

#define INPUT(nr) (&em28xx_boards[dev->model].input[nr])

It may be the case that a NULL pointer deference would occur if there
was only one input defined (as is the case for the PointNix camera).

As a test, you might want to copy the other two inputs for the
PointNix device profile from some other device, and see if you still
hits an oops during input selection.

Devin

-- 
Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Mauro Carvalho Chehab]

On Wed, 7 Jan 2009 13:24:10 -0500
"Devin Heitmueller" <devin.heitmueller@gmail.com> wrote:

> A quick look at the code does show something interesting:
> 
> There are a number of cases where we dereference the result of the
> "INPUT" macro as follows without checking the number of inputs
> defined:
> 
> route.input = INPUT(index)->vmux;
> 
> and here is the macro definition:
> 
> #define INPUT(nr) (&em28xx_boards[dev->model].input[nr])
> 
> It may be the case that a NULL pointer deference would occur if there
> was only one input defined (as is the case for the PointNix camera).
> 
> As a test, you might want to copy the other two inputs for the
> PointNix device profile from some other device, and see if you still
> hits an oops during input selection.

I've reviewed the input stuff at em28xx driver, to avoid accessing input
entries that aren't defined (so, filled with zeros).

Cheers,
Mauro

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Andy Walls]

On Mon, 2009-01-05 at 17:33 -0700, Paul Thomas wrote:
> On Sat, Jan 3, 2009 at 10:22 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
> > On Fri, Jan 2, 2009 at 9:50 AM, Paul Thomas <pthomas8589@gmail.com> wrote:
> >> On Wed, Dec 31, 2008 at 3:52 PM, Devin Heitmueller
> >> <devin.heitmueller@gmail.com> wrote:
> >>> On Wed, Dec 31, 2008 at 5:44 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> >>>> How was it working with Skype? Can we hard code the settings for testing?
> >>>>
> >>>> thanks,
> >>>> Paul
> >>>
> >>> Well, regardless of the device profile issue, it's not clear why it is
> >>> segfaulting.  In theory, it should work but you should only get a
> >>> single input available instead of being able to pick between composite
> >>> and s-video.
> >>>
> >>> I think with this device you should be able to set a modprobe option
> >>> for "debug=1" which might provide a bit more insight.
> >>>
> >>> Try this.  Unplug the device, run "make unload", and then "modprobe
> >>> em28xx".  Then plug the device in.  You should start seeing more info
> >>> in the dmesg output.  Then start the app that segfaults and report the
> >>> dmesg contents.
> >>>
> >>> Devin
> >>>
> >>> --
> >>> Devin J. Heitmueller
> >>> http://www.devinheitmueller.com
> >>> AIM: devinheitmueller
> >>>
> >>
> >> Thanks CityK. I was able to enable core_debug=1. Here is what I have
> >> in dmesg after I camstream seg faults:
> >>
> >> Em28xx: Initialized (Em28xx Audio Extension) extension
> >> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
> >> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> >> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> >> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
> >> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
> >> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
> >> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
> >> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> >> em28xx #0 em28xx_set_alternate :minimum isoc packet size: 2888 (alt=6)
> >> em28xx #0 em28xx_set_alternate :setting alternate 6 with wMaxPacketSize=2892
> >> em28xx #0 em28xx_accumulator_set :em28xx Scale: (1,1)-(179,71)
> >> em28xx #0 em28xx_capture_area_set :em28xx Area Set: (180,72)
> >> em28xx #0 em28xx_init_isoc :em28xx: called em28xx_prepare_isoc
> >> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> >> camstream[18436]: segfault at 0 ip 000000000043ea76 sp
> >> 00007fff1ca5d7a8 error 6 in camstream[400000+5a000]
> >> em28xx #0 em28xx_uninit_isoc :em28xx: called em28xx_uninit_isoc
> >> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> >> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> >> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> >> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> >> em28xx #0 em28xx_irq_callback :urb resubmit failed (error=-1)
> >>
> >> thanks,
> >> Paul
> >>
> >
> > Here is an update on the programs working with em28xx
> >
> > Working programs
> > Skype
> > UCview
> >
> > Here is the status of the other cam programs that aren't working yet:
> > camstream                 -> seg fault
> > gstreamer-properties  -> seg fault
> > cheese                      -> uses test input
> > fswebcam                  -> black screen with green stripe
> >
> > I hope this helps
> >
> > thanks,
> > Paul
> >
> 
> OK, I was able to make some more progress. I can use fswebcam on my
> x86_64 box if I comment out the input selection section. Now I'm
> trying to get fswebcam to work on my embedded arm board
> (http://opencircuits.com/Linuxstamp), but when I run it I get a kernel
> oops. I've included the oops below. I'm now using this
> (git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git)
> git tree.
> 
> Any ideas?
> 
> thanks,
> Paul






> linuxstamp:/home/user/fswebcam-20070108# ./fswebcam -v test.jpeg
> --- Opening /dev/video0...
> Trying source module v4l2...
> /dev/video0 opened.
> src_v4l2_get_capability,81: /dev/video0 information:
> src_v4l2_get_capability,82: cap.driver: "em28xx"
> src_v4l2_get_capability,83: cap.card: "PointNix Intra-Oral Camera"
> src_v4l2_get_capability,84: cap.bus_info: "1-1.7"
> src_v4l2_get_capability,85: cap.capabilities=0x05020041
> src_v4l2_get_capability,86: - VIDEO_CAPTURE
> src_v4l2_get_capability,93: - AUDIO
> src_v4l2_get_capability,95: - READWRITE
> src_v4l2_get_capability,97: - STREAMING
> src_v4l2_set_pix_format,530: Device offers the following V4L2 pixel formats:
> src_v4l2_set_pix_format,542: 0: YUYV (16bpp YUY2, 4:2:2, packed)
> Using palette YUYV
> src_v4l2_set_mmap,648: mmap information:
> src_v4l2_set_mmap,649: frames=4
> src_v4l2_set_mmap,694: 0 length=221184
> src_v4l2_set_mmap,694: 1 length=221184
> src_v4l2_set_mmap,694: 2 length=221184
> src_v4l2_set_mmap,694: 3 length=221184
> kernel BUG at arch/arm/mm/dma-mapping.c:495!
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
> pgd = c13c4000
> [00000000] *pgd=21311031, *pte=00000000, *ppte=00000000
> Internal error: Oops: 817 [#1] PREEMPT
> Modules linked in: saa7115 v4l2_common em28xx videodev v4l1_compat
> videobuf_vmalloc videobuf_core ir_common tveeprom
> CPU: 0    Not tainted  (2.6.28-15476-g7d3b56b #2)
> PC is at __bug+0x20/0x2c
> LR is at vprintk+0x2f0/0x34c
> pc : [<c00265c0>]    lr : [<c003b078>]    psr: 60000013
> sp : c1ff1ba4  ip : c03944fc  fp : c1ff1bb0
> r10: 00000028  r9 : c1d78628  r8 : c1d1f000
> r7 : 00000020  r6 : 00000000  r5 : c1d49800  r4 : ffc09000
> r3 : 00000000  r2 : c1ff0000  r1 : 00000002  r0 : 00000030
> Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
> Control: c000717f  Table: 213c4000  DAC: 00000015
> Process fswebcam (pid: 2383, stack limit = 0xc1ff0260)
> Stack: (0xc1ff1ba4 to 0xc1ff2000)
> 1ba0:          c1ff1bc0 c1ff1bb4 c0027504 c00265b0 c1ff1c84 c1ff1bc4
> c01e2a40
> 1bc0: c0027470 c1ff1bf4 c1ff1bd4 c0053810 c003360c 00000000 c037001c
> c1c713c0
> 1be0: 00000017 c1ff0000 c1ff1c08 c1ff1c08 c1ff1bfc c02bd9b4 c0033704
> c1ff1c34
> 1c00: c1ff1c0c c02bb9a8 c003360c ffffdd0e c1ff0000 00000002 c1ff1c38
> c0376d64
> 1c20: ffffdd0e c1ff0000 c1ff1c70 c1ff1c38 c02bc0b4 c0043c5c 00000000
> 00200200
> 1c40: ffffdd0e c00443ec c1c713c0 c0398b80 00000001 00000001 c1d78000
> 00000001
> 1c60: c1dd9400 00000020 00000000 00008200 c1d78628 00000028 c1ff1cac
> c1ff1c88
> 1c80: c01e4380 c01e2904 c1ff1c94 bf036528 c004464c c1dbd3c0 c1e31400
> c1d78000
> 1ca0: c1ff1cd8 c1ff1cb0 bf036ed0 c01e40ec c1d78000 c1d7cbe0 00000000
> c1d379cc
> 1cc0: bf0150a0 bf038098 c1ff1e2c c1ff1cfc c1ff1cdc bf031904 bf036c68
> bf031000
> 1ce0: ffffffea c1d7cbe0 c1d379cc c1ff1e2c c1ff1d20 c1ff1d00 bf011ff8
> bf031850
> 1d00: c1d379c0 c1ff1e2c c044560f c044560f 00000000 c1ff1d38 c1ff1d24
> bf032b98
> 1d20: bf011d44 bf032b68 000223d0 c1ff1e18 c1ff1d3c bf0244c0 bf032b78 00000001
> 1d40: 00000000 60000013 c0398a4c c03989e0 60000013 c1d49000 00000000 c1d490d8
> 1d60: 00000001 00000001 00000000 c1d379c0 c1e0be00 c044560f c1e50ce0 c1ff1d88
> 1d80: c003624c c003360c 00000000 c1d49000 00000000 c0398a70 00000001 00000009
> 1da0: c1ff1dc4 c1ff1db0 c0191088 c0036204 c1ff0000 00000000 c1ff1dd4 c1ff1dc8
> 1dc0: c01a7720 c019103c c1ff1df0 c1ff1df0 c1ff1ddc c003f048 c003360c c1ff0000
> 1de0: 00000102 c1ff1e20 c1ff1df4 c003f150 00000000 000223d0 fffffff2 00000000
> 1e00: 00000000 c044560f 00000000 c1ff1ed4 c1ff1e1c bf026b68 bf023128 00000003
> 1e20: 00000000 c1ff1e2c c1e50ce0 00000000 00000001 00000000 00000000 00000000
> 1e40: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001
> 1e60: 00000000 00000000 00000000 00000000 c1ff1e7c c01a76b4 c01aac20 c1ff1ea0
> 1e80: c1ff1e8c c01a76f0 c003360c c1ff1eb0 c1ff1e9c c004f4ac c003360c 00000004
> 1ea0: c1d49000 c1ff1f10 c0385478 c044560f 000223d0 c1e50ce0 c044560f c0022fe4
> 1ec0: c1ff0000 00018d94 c1ff1ef0 c1ff1ed8 bf0222e4 bf0269c4 c1e50ce0 ffffffe7
> 1ee0: 000223d0 c1ff1f10 c1ff1ef4 c00993dc bf0222ac c1e50ce0 c1f235a0 00000003
> 1f00: 000223d0 c1ff1f84 c1ff1f14 c0099d18 c009936c 00000000 c1ff1f34 c1ff1f28
> 1f20: c01a7720 c019103c c1ff1f50 c1ff1f50 c1ff1f3c c003f048 c003360c c1ff0000
> 1f40: 00000102 c1ff1f80 c1ff1f54 c003f150 c003efc4 00000000 00000001 c1e50ce0
> 1f60: fffffff7 c044560f 00000036 c0022fe4 c1ff0000 00018d94 c1ff1fa4 c1ff1f88
> 1f80: c0099d80 c0099858 00000000 00000000 00000004 00022280 00000000 c1ff1fa8
> 1fa0: c0022e40 c0099d50 00000000 00000004 00000003 c044560f 000223d0 00022414
> 1fc0: 00000000 00000004 00022280 000223d0 00000001 0000000c 00018d94 000220a0
> 1fe0: 00021540 be81e728 00013e68 40129424 80000010 00000003 00000000 00000000
> Backtrace:
> [<c00265a0>] (__bug+0x0/0x2c) from [<c0027504>] (dma_cache_maint+0xa4/0xb8)
> [<c0027460>] (dma_cache_maint+0x0/0xb8) from [<c01e2a40>]
> (usb_hcd_submit_urb+0x14c/0x890)
> [<c01e28f4>] (usb_hcd_submit_urb+0x0/0x890) from [<c01e4380>]
> (usb_submit_urb+0x2a4/0x2ac)
> [<c01e40dc>] (usb_submit_urb+0x0/0x2ac) from [<bf036ed0>]
> (em28xx_init_isoc+0x278/0x2c4 [em28xx])
>  r6:c1d78000 r5:c1e31400 r4:c1dbd3c0
> [<bf036c58>] (em28xx_init_isoc+0x0/0x2c4 [em28xx]) from [<bf031904>]
> (buffer_prepare+0xc4/0xf4 [em28xx])
> [<bf031840>] (buffer_prepare+0x0/0xf4 [em28xx]) from [<bf011ff8>]
> (videobuf_qbuf+0x2c4/0x400 [videobuf_core])
>  r7:c1ff1e2c r6:c1d379cc r5:c1d7cbe0 r4:ffffffea
> [<bf011d34>] (videobuf_qbuf+0x0/0x400 [videobuf_core]) from
> [<bf032b98>] (vidioc_qbuf+0x30/0x3c [em28xx])
>  r8:00000000 r7:c044560f r6:c044560f r5:c1ff1e2c r4:c1d379c0
> [<bf032b68>] (vidioc_qbuf+0x0/0x3c [em28xx]) from [<bf0244c0>]
> (__video_do_ioctl+0x13a8/0x389c [videodev])
>  r5:000223d0 r4:bf032b68
> [<bf023118>] (__video_do_ioctl+0x0/0x389c [videodev]) from
> [<bf026b68>] (video_ioctl2+0x1b4/0x2a4 [videodev])
> [<bf0269b4>] (video_ioctl2+0x0/0x2a4 [videodev]) from [<bf0222e4>]
> (v4l2_ioctl+0x48/0x54 [videodev])
> [<bf02229c>] (v4l2_ioctl+0x0/0x54 [videodev]) from [<c00993dc>]
> (vfs_ioctl+0x80/0x94)
>  r6:000223d0 r5:ffffffe7 r4:c1e50ce0
> [<c009935c>] (vfs_ioctl+0x0/0x94) from [<c0099d18>] (do_vfs_ioctl+0x4d0/0x4f8)
>  r7:000223d0 r6:00000003 r5:c1f235a0 r4:c1e50ce0
> [<c0099848>] (do_vfs_ioctl+0x0/0x4f8) from [<c0099d80>] (sys_ioctl+0x40/0x5c)
> [<c0099d40>] (sys_ioctl+0x0/0x5c) from [<c0022e40>] (ret_fast_syscall+0x0/0x2c)
>  r6:00022280 r5:00000004 r4:00000000
> Code: e1a01000 e59f000c eb0052c5 e3a03000 (e5833000)
> ---[ end trace 68fb0b4324720f98 ]---


Well, I don't have kernel 2.6.28 code lying around, but it looks like a
BUG() that was being emitted Oops'ed instead of completing.

The ARMv5te disassmbly of the code bytes show this:

   0:   e1a01000        mov     r1, r0
   4:   e59f000c        ldr     r0, [pc, #12]   ; 0x18
   8:   eb0052c5        bl      0x14b24
   c:   e3a03000        mov     r3, #0  ; 0x0
  10:   e5833000        str     r3, [r3]       <=== Ooops here

r3 is 0/NULL in the Oops dump, but that just means there was a problem
with the BUG() call (I think).  That doesn't help much.

Without the 2.6.28 ARM kernel source at hand, I can't say much more
aside from the em82xx_init_isoc() is the first v4l-dvb driver function
encountered going through the backtrace and that it looks like the chain
of events was spurred by the memory mapping activity:

> ...
>src_v4l2_set_mmap,694: 3 length=221184


Regards,
Andy



--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

On Wed, Jan 7, 2009 at 6:50 PM, Mauro Carvalho Chehab
<mchehab@infradead.org> wrote:
> On Wed, 7 Jan 2009 13:24:10 -0500
> "Devin Heitmueller" <devin.heitmueller@gmail.com> wrote:
>
>> A quick look at the code does show something interesting:
>>
>> There are a number of cases where we dereference the result of the
>> "INPUT" macro as follows without checking the number of inputs
>> defined:
>>
>> route.input = INPUT(index)->vmux;
>>
>> and here is the macro definition:
>>
>> #define INPUT(nr) (&em28xx_boards[dev->model].input[nr])
>>
>> It may be the case that a NULL pointer deference would occur if there
>> was only one input defined (as is the case for the PointNix camera).
>>
>> As a test, you might want to copy the other two inputs for the
>> PointNix device profile from some other device, and see if you still
>> hits an oops during input selection.
>
> I've reviewed the input stuff at em28xx driver, to avoid accessing input
> entries that aren't defined (so, filled with zeros).
>
> Cheers,
> Mauro
>

So, I'm finally able to get the source to compile again. I'm now using
a gcc 4.3.2 cross-compiler instead of a gcc 3.4.5. The three things
that make it work nicely are to use the "make release DIR=" command,
add "ARCH=arm CROSS_COMPILE=arm-unknown-gnu-" to the v4l-dvb make
command and finally run "make install" from the embedded side.

Anyway, I still get the oops with the latest tree. Also did some more
tests on my x86_64 box it looks like I have to run ucview before
fswebcam will work.

Is there any way this is being caused by improper ioctl calls from user-space?

The other thing that is odd is that there seems to be a need for some
physical memory. I have 512MB of swap space, but unless I have > 6MB
of physical memory I get a "Cannot allocate memory" error.

thanks,
Paul

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

Re: em28xx issues

[Paul Thomas]

On Mon, Jan 12, 2009 at 6:17 PM, Paul Thomas <pthomas8589@gmail.com> wrote:
> On Wed, Jan 7, 2009 at 6:50 PM, Mauro Carvalho Chehab
> <mchehab@infradead.org> wrote:
>> On Wed, 7 Jan 2009 13:24:10 -0500
>> "Devin Heitmueller" <devin.heitmueller@gmail.com> wrote:
>>
>>> A quick look at the code does show something interesting:
>>>
>>> There are a number of cases where we dereference the result of the
>>> "INPUT" macro as follows without checking the number of inputs
>>> defined:
>>>
>>> route.input = INPUT(index)->vmux;
>>>
>>> and here is the macro definition:
>>>
>>> #define INPUT(nr) (&em28xx_boards[dev->model].input[nr])
>>>
>>> It may be the case that a NULL pointer deference would occur if there
>>> was only one input defined (as is the case for the PointNix camera).
>>>
>>> As a test, you might want to copy the other two inputs for the
>>> PointNix device profile from some other device, and see if you still
>>> hits an oops during input selection.
>>
>> I've reviewed the input stuff at em28xx driver, to avoid accessing input
>> entries that aren't defined (so, filled with zeros).
>>
>> Cheers,
>> Mauro
>>
>
> So, I'm finally able to get the source to compile again. I'm now using
> a gcc 4.3.2 cross-compiler instead of a gcc 3.4.5. The three things
> that make it work nicely are to use the "make release DIR=" command,
> add "ARCH=arm CROSS_COMPILE=arm-unknown-gnu-" to the v4l-dvb make
> command and finally run "make install" from the embedded side.
>
> Anyway, I still get the oops with the latest tree. Also did some more
> tests on my x86_64 box it looks like I have to run ucview before
> fswebcam will work.
>
> Is there any way this is being caused by improper ioctl calls from user-space?
>
> The other thing that is odd is that there seems to be a need for some
> physical memory. I have 512MB of swap space, but unless I have > 6MB
> of physical memory I get a "Cannot allocate memory" error.
>
> thanks,
> Paul
>

I think some of this could be seen from the backtrace I posted before,
but I can clarify a little. I've been putting printks in the kernel to
follow the error.

The calling user space function from fswebcam is ioctl(s->fd,
VIDIOC_QBUF, &s->buf)

In v4l2-ioctl.c in the VIDIOC_QBUF case it gets to the
"ops->vidioc_qbuf(file, fh, p)" call

In em28xx-video.c in function vidioc_qbuf the return line "return
(videobuf_qbuf(&fh->vb_vidq, b));" calls videobuf_qbuf

In videobuf-core.c in function videobuf_qbuf it gets to "retval =
q->ops->buf_prepare(q, buf, field);"

in em28xx-video.c in function buffer_prepare it gets to "rc =
em28xx_init_isoc(dev, EM28XX_NUM_PACKETS,
				      EM28XX_NUM_BUFS, dev->max_pkt_size,
				      em28xx_isoc_copy);"

in em28xx-core.c in function em28xx_init_isoc it gets to "rc =
usb_submit_urb(dev->isoc_ctl.urb[i], GFP_ATOMIC);"

The last to functions I didn't trace because they were called to much,
but the backtrace says they are usb_hcd_submit_urb and finally
dma_cache_maint.

I'll try and add some more debugging in to see if I can see anything
else. Also I got a usbvision board that does the exact same thing, but
I haven't followed the call chain on it. The virtual driver works fine
with fswebcam.

I've seen a lot of the OMAP patches running around, but has anyone
else have good success using arm?

thanks,
Paul

--
video4linux-list mailing list
Unsubscribe mailto:video4linux-list-request@redhat.com?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list