[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#572134: xserver-xorg-video-intel: i915 945GM gpu crashed after resume



Feeling adventurous, I decided to upgrade to xserver-xorg-video-intel
from experimental (2:2.10.902-1) and kernel
linux-image-2.6.32-4-686. This seems to work fine for me. After a
couple of suspend/resume tests, it kept working.

So for me, this bug can be closed.

Just for a bit more info: pm-suspend worked fine, and Fn-F3 also, but
closing the lid resulted in the problem earlier.

Thanks,

Mark

On Wed, Mar 3, 2010 at 8:44 PM, Mark Borst <mark@borst.org> wrote:
> Hi Cyril
>
> Thanks for having a look at my problem. I have installed debug packages.
>
> I've been trying to reliably reproduce the problem, and it comes down to
> this:
>
> 1. dmesg > dmesg_before; echo mem > /sys/power/state; dmesg > dmesg_after
> works. No problem whatsoever. This is from
> http://intellinuxgraphics.org/suspend-resume.html
>
> 2. pm-suspend works fine. No problem, not after multiple times. I don't get
> the gnome lock-screen thingie.
>
> 3. close laptop lid. This gives the problem. The first time it's fine, the
> second time it gives the problem. Always. I don't know what exactly is
> happening and why this is different. But after closing the lid, I get the
> gnome lock-screen passwordprompt. It seems gdm is trying to restart X, and
> that's failing the second time.
>
> A little problem is that I can't ssh into the box when it's 'broken' as
> network-manager apparantly doesn't work when there is no nm-applet to talk
> to, like when X isn't there.
>
> So there seems to be a dependency on the way the laptop is suspended, but I
> don't know how to determine that properly. Do you have any more suggestions?
>
> Thanks,
>
> Mark
>
> On Tue, Mar 2, 2010 at 12:20 AM, Cyril Brulebois <kibi@debian.org> wrote:
>>
>> Hi Mark,
>>
>> Mark Borst <mark@borst.org> (01/03/2010):
>> > After a resume from suspend, the gpu (945GM) often crashes. Earlier
>> > I got an X screen back, with mouse moving, but no other actions
>> > possible. Then I could still switch to VT1 and reboot the box. I
>> > hoped it would be fixed with newer kernels, so I eagerly upgraded
>> > recently, up till now when I'm running 2.6.33.
>>
>> thanks for trying this. :)
>>
>> > However, the problem hasn't stopped.
>>
>> While this is sad, that's good to know.
>>
>> If you have a few more minutes, it'd be very nice if you could report
>> this bug on upstream's bugzilla following instructions at [1], Cc'ing
>> me, so that I can mark this bug as forwarded there.
>>
>>  1. http://intellinuxgraphics.org/how_to_report_bug.html
>>
>> What could be of some help to make sure you're getting almost all
>> messages: ssh from another box, install the various libdrm*dbg,
>> *-intel-dbg, -core-dbg, and attach X inside gdb. This might help you
>> get a backtrace if X is segfaulting and if no other traces of that are
>> kept.
>>
>> G'luck.
>>
>> Mraw,
>> KiBi.
>>
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.10 (GNU/Linux)
>>
>> iEYEARECAAYFAkuMS6QACgkQeGfVPHR5Nd3hcwCgnAFi2hityKPC9IsNUsQsJHpm
>> Ps4AoMk/fAiizt1VGf7e+i4z4hAPeSKJ
>> =8RC/
>> -----END PGP SIGNATURE-----
>>
>
>



Reply to: