[Libdlo] Kernel panic with udlfb
Bernie Thompson
bernie at berniethompson.com
Mon Oct 11 10:55:12 PDT 2010
Hi Alexander,
Yes, sorry, this is a bug in older versions of udlfb. If no one is
using the framebuffer when someone reads the use_defio sysfs
attribute, udlfb will dereference a null pointer.
It was fixed about 2 months ago (it's also fixed in the version
alreadygoing into 2.6.37), when use_defio was removed. Control of
defio on/off is no longer a sysfs attribute, rather a kernel module
option.
For a fix, you could upgrade to a later version udlfb, or just null
out the use_defio_show() code in your version, and have it return 0;
Best wishes,
Bernie
On Mon, Oct 11, 2010 at 10:46 AM, Alexander Todorov <atodorov at otb.bg> wrote:
> Hi all,
> I got a kernel panic on 2.6.32 kernel with a slightly older version of udlfb
> when trying to extract some udev info:
>
> # udevadm info -q all -a -p `pwd`
>
> Udevadm info starts with the device specified by the devpath and then
> walks up the chain of parent devices. It prints for every device
> found, all possible attributes in the udev rules key format.
> A rule to match, can be composed by the attributes of the device
> and the attributes from one single parent device.
>
> looking at device
> '/devices/pci0000:00/0000:00:1d.7/usb2/2-2/2-2.1/graphics/fb1':
> KERNEL=="fb1"
> SUBSYSTEM=="graphics"
> DRIVER==""
> ATTR{bits_per_pixel}=="16"
> ATTR{blank}==""
> ATTR{console}==""
> ATTR{cursor}==""
> ATTR{mode}==""
> ATTR{pan}=="0,0"
> ATTR{virtual_size}=="1024,768"
> ATTR{name}=="udlfb"
> ATTR{stride}=="2048"
> ATTR{rotate}=="0"
> ATTR{state}=="0"
> ATTR{metrics_bytes_rendered}=="12369280"
> ATTR{metrics_bytes_identical}=="4423288"
> ATTR{metrics_bytes_sent}=="2551490"
> ATTR{metrics_cpu_kcycles_used}=="145906"
> ATTR{use_defio}=="0"
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:Oops: 0000 [#1] SMP
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:last sysfs file:
> /sys/devices/pci0000:00/0000:00:1d.7/usb2/2-2/2-2.1/graphics/fb1/use_defio
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:Process udevadm (pid: 2363, ti=e9c64000 task=f71a2030
> task.ti=e9c64000)
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:Stack:
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:Call Trace:
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:Code: 83 ec 10 83 e8 08 89 1c 24 8b 5c 24 1c 89 74 24 04 89 ce 89 7c
> 24 08 8b 7c 24 14 89 6c 24 0c 8b 6c 24 18 e8 28 1d 39 c7 83 fd 00 <8b> 88 30
> 03 00 00 7c 12 7e 0b 31 c0 eb 38 8d b4 26 00 00 00 00
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:EIP: [<f930f2bb>] edid_show+0x2b/0xa0 [udlfb] SS:ESP 0068:e9c65f10
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:CR2: 0000000000000330
>
> Message from syslogd at sumu01 at Oct 11 20:44:20 ...
> kernel:Kernel panic - not syncing: Fatal exception
>
>
> --
> Regards,
> Alexander.
> _______________________________________________
> Libdlo mailing list
> Libdlo at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/libdlo
>
More information about the Libdlo
mailing list