pwc driver bugs

DISCUSSION Udo van den Heuvel 3 participants 4 messages May 24, 2012

Share & tools

Udo van den Heuvel

author

2012-05-24 15:03 UTC #0 lore raw

Hello,

I discovered some issues in the pwc driver for certain USB webcams. I opened some bugs[1].
I do think both issues might be related and the crash issue is quite serious.

Please have a look, I can assist by testing any patches with fixes. (for ~3.3.7)

Kind regards,
Udo

[1]: https://bugzilla.kernel.org/show_bug.cgi?id=43291

     https://bugzilla.kernel.org/show_bug.cgi?id=43290

My last locking rework for pwc mistakenly assumed that videbuf2 does its own locking, but it does not! This patch fixes the missing locking by moving over the the video_device lock, and introducing a separate lock for the videobuf2_queue.

Signed-off-by Hans de Goede <[email protected]>

Signed-off-by Mauro Carvalho Chehab <[email protected]>

Udo van den Heuvel

author

2012-05-24 16:38 UTC #2 lore raw

Quoted 3 lines · Alan Cox

commit ceede9fa8939e40ad0ddb4ad1355f45c6f1d3478
looks rather relevant here.

The commit does not apply cleanly to 3.3.7.
What are the dependencies for this patch to apply, so I can try to test this code?

Kind regards,
Udo

Hans de Goede

2012-05-24 19:10 UTC #3 lore raw

In reply to Alan Cox

Alan,

Thanks for adding me to the CC on this. Udo, I've
responded to your bugs in bugzilla.

Regards,

Hans

Quoted 30 lines · Alan Cox

Quoted 9 lines · On Thu, 24 May 2012 17:03:49 +0200 Udo van den Heuvel<[email protected]> wrote:
Hello,
I discovered some issues in the pwc driver for certain USB webcams. I opened some bugs[1].
I do think both issues might be related and the crash issue is quite serious.
Please have a look, I can assist by testing any patches with fixes. (for ~3.3.7)
commit ceede9fa8939e40ad0ddb4ad1355f45c6f1d3478
looks rather relevant here.
commit ceede9fa8939e40ad0ddb4ad1355f45c6f1d3478
Author: Hans de Goede<[email protected]>
Date:   Wed May 9 04:43:12 2012 -0300
[media] pwc: Fix locking
My last locking rework for pwc mistakenly assumed that videbuf2 does its own locking, but it does not! This patch fixes the missing locking by moving over the the video_device lock, and introducing a separate lock for the videobuf2_queue.
Signed-off-by Hans de Goede<[email protected]>
Signed-off-by Mauro Carvalho Chehab<[email protected]>