[hardy] oops when running cheese on Thinkpad X300

Bug #200990 reported by Chris Jones
10
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Fix Released
High
Stefan Bader

Bug Description

the Thinkpad X300 has a built in webcam:

Bus 004 Device 002: ID 17ef:4807 ChipsBnk

which is supported by the UVC driver:

[ 23.318039] uvcvideo: Found UVC 1.00 device <unnamed> (17ef:4807)

Running gstreamer-properties and testing the default V4L2 input source displays video from the camera as expected.

Running cheese causes the kernel to oops.
(fully updated Hardy)

Attaching a dump of dmesg, which includes the oops.

Revision history for this message
Chris Jones (cmsj) wrote :
Revision history for this message
Chris Jones (cmsj) wrote :

I just tested again with gstreamer-properties and that is Oopsing the kernel now too, so perhaps this is something that changed in the most recent kernel updates.

Changed in linux:
assignee: nobody → ubuntu-kernel-team
importance: Undecided → High
status: New → Triaged
Revision history for this message
Brian Murray (brian-murray) wrote :

I didn't see this particular device listed at http://linux-uvc.berlios.de/ as supported. However, they say "Other UVC compliant video input devices are very likely to be supported."

Revision history for this message
Chris Jones (cmsj) wrote :

It is indeed not specifically listed and I can't reproduce it working at all. My only suspicion at this point is that I was using 2.6.24-10, but I don't think the uvc driver has changed since then, and that version is gone from the archives.

Revision history for this message
Chris Jones (cmsj) wrote :

This is the output of v4l-info on the camera, up to the point where the tool segfaults because the driver has oopsed in uvc_get_video_ctrl (which seems to be the same place as before).

Revision history for this message
Rob Martin (robmartin) wrote :

Logitech Sphere/Orbit not recognized. Cheese and Ekiga crashing. Similar regression occurred during Gusty beta.

Attached are output files requested.

Revision history for this message
Rob Martin (robmartin) wrote :
Revision history for this message
Rob Martin (robmartin) wrote :
Revision history for this message
Chris Jones (cmsj) wrote :

Rob: the OOPS in your kernel dmesg strongly suggests to me that it's actually an audio driver that's crashing (I guess your camera includes a USB microphone). I suggest you report a new bug against linux-ubuntu-modules or something alsa related.

Revision history for this message
Chris Jones (cmsj) wrote :

I have been talking to upstream about this and Laurent.Pinchart has offered a patch for testing. It seems to fix this issue for me - the camera works in cheese multiple times after booting, and suspending multiple times - I can no longer immediately reproduce the OOPS.

I have asked Laurent (ie upstream) to comment on the potential impact of the patch for other cameras, but even if it's super low risk, it seems to be too late to make Hardy now.

Revision history for this message
Stefan Bader (smb) wrote :

@Rob
There has already been a fix to sco_bt (which may cause this). Should be included in -14.

Revision history for this message
Stefan Bader (smb) wrote :

@Ng, thanks for the pointer. Looking at the patch it absolutely makes sense not to access an array before its start. ;-) So it shouldn't harm.

Changed in linux:
assignee: ubuntu-kernel-team → stefan-bader-canonical
status: Triaged → Fix Committed
Revision history for this message
Chris Jones (cmsj) wrote :

This is fixed in 2.6.24-15. Thanks Stefan!

Changed in linux:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.