random freezes after resuming from sleep

Bug #21646 reported by Francesco Accattapà
6
Affects Status Importance Assigned to Milestone
linux-source-2.6.15 (Ubuntu)
Fix Released
Medium
Ben Collins

Bug Description

The system randomly crashes after resuming on an updated breezy-preview
installation. Neither the mouse nor the keyboard are responsive. It sometimes
freezes after a few seconds, sometimes after a few minutes; however, the problem
seems to manifest only after the opening of an application. The machine is an
iBook G4.

Here is the /proc/cpuinfo file:

----------------------------------------------------------
processor : 0
cpu : 7447A, altivec supported
clock : 1199MHz
revision : 0.2 (pvr 8003 0102)
bogomips : 1196.03
machine : PowerBook6,5
motherboard : PowerBook6,5 MacRISC3 Power Macintosh
detected as : 287 (iBook G4)
pmac flags : 0000001b
L2 cache : 512K unified
memory : 768MB
pmac-generation : NewWorld
----------------------------------------------------------

Attached to this report are the xorg.conf file and the output of "lspci -vv".

Kernel version: linux-image-2.6.12-8-powerpc 2.6.12-8.12
pbbuttonsd version: 0.6.6-3ubuntu3

Please let me know if I can provide more information.

Revision history for this message
Francesco Accattapà (callipeo) wrote :

Created an attachment (id=3792)
lspci -vv output

Revision history for this message
Francesco Accattapà (callipeo) wrote :

Created an attachment (id=3793)
xorg.conf file

Revision history for this message
Ben Collins (ben-collins) wrote :

Is it possible to reproduce this from the console, to see if there are any errors printed by the kernel prior
to the freeze?

Revision history for this message
Francesco Accattapà (callipeo) wrote :

I've tried to sleep/resume several times from the console, and the computer had
never gone frozen. However I should add that while yesterday the problem
appeared systematically (one resume->one freeze), today I tried to sleep/resume
inside gnome and the computer did not freeze. I tried only once, but it is
significant compared to the situation of the last days IMHO. Could it be a local
problem? I have a spare partition available: could a further check on another
fresh installation (a pure 5.10-preview system, without updates or changes to
the default configuration) be helpful?

Revision history for this message
Francesco Accattapà (callipeo) wrote :

Today the problem appeared two times, after ~30 minutes from the resuming. I
will try to not put the laptop to sleep for a couple of days, to check if the
problem is actually sleep-related.

Revision history for this message
Francesco Accattapà (callipeo) wrote :

I did not put the laptop to sleep since my last comment, and it never froze.
Could it be helpful if I would manage to get the SysRq Keys working?

Revision history for this message
Ben Collins (ben-collins) wrote :

(In reply to comment #6)
> I did not put the laptop to sleep since my last comment, and it never froze.
> Could it be helpful if I would manage to get the SysRq Keys working?

Very helpful. Atleast we'll know where the kernel is when the freeze occurs.

Revision history for this message
Francesco Accattapà (callipeo) wrote :

I have applied the patch founded at
http://patchwork.ozlabs.org/linuxppc//patch?id=968, so Fn+ESC acts as ALT+F13. I
can confirm that the key combination works, but I could not use it to obtain a
trace. I have followed the instruction at
https://wiki.ubuntu.com//DebuggingSystemCrash, but the laptop only freezes
inside a graphical environment and there is nothing significant in
/var/log/kern.log. BTW [sysrq]+s followed by [sysrq]+u followed by [sysrq]+b
does not work either. I'll try again to reproduce the problem from the console.
If there is a better way to use the SysRq keys on an iBook than the patch I've
applied, please let me know.

Revision history for this message
Arne Caspari (arne-datafloater) wrote :

I had this problem also with my iBook ( it got fixed with recent kernels ). In
my case it was related to CPU frequency scaling and it never occurred when I let
the CPU always run on a fixed frequency.

Maybe you want to try that also.

Have you verified you really boot into a recent kernel ( with uname -a )?

Revision history for this message
Francesco Accattapà (callipeo) wrote :

I just tried with the kernel 2.6.12-9.18 and powernowd disabled: it is the same.
Total freeze after ~30 seconds.

It's a bit frustrating since I have the time and the will to help debugging, but
I can't find a way to do it.

Revision history for this message
Ben Collins (ben-collins) wrote :

I'm not sure what to tell you. This seems to be an isolated event. Without
debugging information, I don't have much to go on.

If you have the will and time, you could try compling your own stock 2.6.13
kernel and see if the problem still occurs with that (use /boot/config-2.6.12-9*
as a starting point for your own .config).

Revision history for this message
Francesco Accattapà (callipeo) wrote :

(In reply to comment #11)
> I'm not sure what to tell you. This seems to be an isolated event. Without
> debugging information, I don't have much to go on.

I fully understand.

> If you have the will and time, you could try compling your own stock 2.6.13
> kernel and see if the problem still occurs with that

Will try.

Revision history for this message
Francesco Accattapà (callipeo) wrote :

Just tried with kernel 2.6.13.2: nothing changes.

Revision history for this message
Ben Collins (ben-collins) wrote :

If possible, please upgrade to Dapper's 2.6.15-7 kernel. If you do not want to
upgrade to Dapper, then you can also wait for the Dapper Flight 2 CD's, which
are due out within the next few days.

Let me know if this bug still exists with this kernel.

Revision history for this message
Francesco Accattapà (callipeo) wrote :

I've tried today with 2.6.15-7 and it seems to work well.

Revision history for this message
Ben Collins (ben-collins) wrote :

Thanks

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.