Prev: Badness at fs/sysfs/symlink.c:82 during qeth initalization
Next: perf pmu interface changes -v3
From: Krzysztof Halasa on 9 Jul 2010 06:30
Hello Ben,

I'm trying to use your solo6x10 driver with a card based on 6010 chip.
The driver seems to detect the card correctly, but when I open
/dev/video0 and then close it, I got a kernel panic. I traced the
problem to invalid pointers in q->bufs[i]->queue in a call to
videobuf_queue_cancel().

The kernel is current Linus' tip (.35-rc4+) + your "staging solo6x10".
The same seems to happen with 2.6.34 + the driver included "manually".
x86-64. The card is Commell's MP-6010.

It goes like this:

# modprobe solo6x10 video_type=1

solo6x10: module is from the staging directory, the quality is unknown,
you have been warned.
solo6010 0000:08:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
solo6010 0000:08:00.0: Enabled 2 i2c adapters
solo6010 0000:08:00.0: Initialized 1 tw28xx chip: tw2815[1]
solo6010 0000:08:00.0: Display as /dev/video0 with 4 inputs (1 extended)
solo6010 0000:08:00.0: Encoders as /dev/video1-4
solo6010 0000:08:00.0: Alsa sound card as Softlogic0

# cat /dev/video0 > /dev/null ^C

- videobuf_queue_cancel() called
- about to call list_del(&q->bufs[i]->queue) with:
i = 0
&q->bufs[i]->queue is a valid pointer = 0xffff88033cf97c00
q->bufs[i]->queue.prev = 0xdead000000200200
q->bufs[i]->queue.next = 0xdead000000100100

The backtrace is:
videobuf_queue_cancel()
__videobuf_read_stop()
videobuf_stop()
solo_v4l2_release()
v4l2_release()
fput()
filp_close()
....

Any ideas?
--
Krzysztof Halasa
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Krzysztof Halasa on 17 Jul 2010 05:10
Krzysztof Halasa <khc(a)pm.waw.pl> writes:

> I'm trying to use your solo6x10 driver with a card based on 6010 chip.
> The driver seems to detect the card correctly, but when I open
> /dev/video0 and then close it, I got a kernel panic. I traced the
> problem to invalid pointers in q->bufs[i]->queue in a call to
> videobuf_queue_cancel().

BTW I have fixed it and will work on this driver a bit.
Will post the results soon.
--
Krzysztof Halasa
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: Badness at fs/sysfs/symlink.c:82 during qeth initalization
Next: perf pmu interface changes -v3