[GM965] X freezes when logging in from screensaver - fails in intelSwapBuffers()/ intelCopyBuffers()

[Bengt Olsson]

Binary package hint: xserver-xorg-video-intel

Running Hardy updated as of today on a Thinkpad T61 with intel X3100 graphics:

$ dpkg -l | grep xserver-xorg-video-intel
ii xserver-xorg-video-intel 2:2.2.1-1ubuntu13.6 X.Org X server -- Intel i8xx, i9xx display d

I occasionally run into X lock-ups where the desktop freezes, except for mouse. Not possible to Ctrl-Alt-F1 out to terminal. Laptop can be reached from ssh and it is seen that the /usr/bin/X process runs at almost 100% CPU. The Xorg.0.log contains repetitions of

 tossed event which came in late
 mieqEnequeue: out-of-order valuator event; dropping.

Using compiz and have direct rendering enabled. Believe it is the same problem that is reported incomment
https://bugs.launchpad.net/ubuntu/hardy/+source/xserver-xorg-video-intel/+bug/212799/comments/44

and that it has something to do with was is reported in

https://bugs.freedesktop.org/show_bug.cgi?id=14428

Attaching Xorg.0.log.old that was active at the latest lock-up.

#0 0xb7fad433 in __kernel_vsyscall ()
#1 0xb7c731a9 in ioctl () from /lib/tls/i686/cmov/libc.so.6
#2 0xb7a731df in ?? () from /usr/lib/libdrm.so.2
#3 0xb7a73294 in drmCommandWrite () from /usr/lib/libdrm.so.2
#4 0xa761bcd6 in intelWaitIrq () from /usr/lib/dri/i965_dri.so
#5 0xa7615f74 in ?? () from /usr/lib/dri/i965_dri.so
#6 0xa760f3bc in ?? () from /usr/lib/dri/i965_dri.so
#7 0xa760e712 in dri_fence_wait () from /usr/lib/dri/i965_dri.so
#8 0xa7611e0d in intelCopyBuffer () from /usr/lib/dri/i965_dri.so
#9 0xa7614620 in intelSwapBuffers () from /usr/lib/dri/i965_dri.so
#10 0xa760b5a7 in ?? () from /usr/lib/dri/i965_dri.so
#11 0xb7ad8ac4 in ?? () from /usr/lib/xorg/modules/extensions//libglx.so
#12 0xb7acb2ce in ?? () from /usr/lib/xorg/modules/extensions//libglx.so
#13 0xb7acec0a in ?? () from /usr/lib/xorg/modules/extensions//libglx.so
#14 0x0808c89f in Dispatch ()
#15 0x08071d1d in main ()

[lspci]
00:00.0 Host bridge: Intel Corporation Mobile PM965/GM965/GL960 Memory Controller Hub (rev 0c)
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

[Bengt Olsson (bengt-blafs)]

Had this problem again but now with the added information:

"[mi] EQ overflowing. The server is probably stuck in an infinite loop."

repeating in the Xorg.0.log file. Maybe because this happened when the laptop was un-attended during the week-end and had been running this loop for a long time.

This is a very troublesome bug and seems to be experienced by others with different drivers, see

http://bugs.gentoo.org/207106

and

http://bugs.freedesktop.org/show_bug.cgi?id=14633

[Szilárd Páll (sin-pecado)]

I can also confirm this problem. In my case It mainly appears when a video file is opened (using Totem but as far as I remember it happened with vlc as well). The freezing does not occur every time when I launch the video player but it is quite frequent. Like in the former reported case the X freezes except of the mouse movement. Restarting the X does not help.

The lines in the Xorg.0.log produced by the (presumable) bug:

"mieqEnequeue: out-of-order valuator event; dropping.
tossed event which came in late"

OS: Ubuntu Hardy with latest updates.

xserver-xorg-video-intel: 2:2.2.1-1ubuntu13.7

Hardware: HP nx7400, with video chipset Intel i945GM.

Attached the X log.

I hope a solution will be found quickly as it is extremely annoying.

[Tormod Volden (tormodvolden)]

From https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/276518/comments/47:

"The "EQ overflowing" error message is a generic error that occurs after the Xserver has hung for any old reason.
So it is most likely that each reporter here is seeing an xserver hang for a *different* reason, but since this esoteric error message is printed in each case, it is confusing and hard to tell the lockups apart. (See bug #259808)"

"Also please note you're not having an "overflow error". That error
message is just a side effect. You are having an "X lockup", and that
is the issue that has to be troubleshot."

See also https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/276518/comments/49

[Tormod Volden (tormodvolden)]

If possible, try the xserver-xorg-core 1.5.3 in Jaunty, since it will add a backtrace to Xorg.0.log which might help identify the real bug behind the lock-up.

[Szilárd Páll (sin-pecado)]

@Tormod

Thanks for the tip, however I have Hardy and I don't really feel like upgrading, not even to Intrepid. At the same time I have checked the xserver-xorg-core 1.5.3 but in order to just install on Hardy I'd need to upgrade quite a bunch of other packages as well and there is no guarantee that it will work.
Any other possibility to try it out?

As far as I understood it is much easier to mark a report duplicate in case if it turn out that it is. Therefore, just to make things clear(er), do you think I should open a new bug report?

Thanks,
Szilard

[Tormod Volden (tormodvolden)]

Szilárd, without a backtrace it is hard to know if it's the same bug, but file a new bug anyway. I am using 1.5.3 packages in Hardy from the xorg-edgers PPA, but I wouldn't recommend it for beginners. The best would be if you could install Jaunty on a separate partition or a USB drive. You can also use a live CD if you install the openssh-server on it and log in from another computer to read the log after Xorg is locked up.

[Szilárd Páll (sin-pecado)]

As I'm not really experienced, I will try to induce the lockup either with a live CD or if it does not work I will try to free up some space and install Jaunty somewhere.

Thanks again for the tips, I will get back to the topic if I have any news.

[Bryce Harrington (bryce)]

[This is an automated message]

Hi bengt-blafs,

Please attach the output of `lspci -vvnn` too.

[Bryce Harrington (bryce)]

[This is an automatic notice.]

We'd like to forward your bug upstream, however upstream requires
that you first test it against their newer driver code.

To save you the effort of building the driver from source, we've built
packages for the driver and its new dependencies.

So you have a couple options:

 1. Download and test .debs for intrepid, from:
     https://edge.launchpad.net/~intel-gfx-testing/+archive

 -or-

 2. Download and test the Jaunty alpha-2 (or newer) Live CD,
     (which includes a beta of the new xserver 1.6 as well).
     See http://cdimage.ubuntu.com/releases/9.04/ for ISOs

Thanks ahead of time! You can simply reply to this email to report your
findings.

P.S., if you wish to forward your bug upstream yourself, please follow
these directions to do so:
  http://intellinuxgraphics.org/how_to_report_bug.html

[Bengt Olsson (bengt-blafs)]

Ok thanks will do, but its on my work machine so it will take some days since I started X-mas holidays now... Also I disabled compiz since a while back and have had no occurrence of the fault since. I will turn on compiz again but, since the lock-ups where very sporadic even with compiz, it may take some time before I can verify that it works (will use the Intrepid debs).

However, I attach the current lspci output (without compiz enabled)

Happy X-mas!

[Bengt Olsson (bengt-blafs)]

OK I've been running the above intrepid debs since dec 29, and just now I got a look-up again, this time when logging in remote via vnc from home. The T61 at work is up running and locked by the screen-saver. The screen saver is running without any "effect" i.e. it is black. I first typed my passwd wrong and when I retyped the passwd and hit "enter" (or pressed "OK" don't remember which) X locked up. I logged in via ssh and saw the usual symptom: X process running at 100% (on one of the cores).

This time with no output in Xorg.0.log (at least not yet after about 20 minutes), but it is clearly the same problem as before. Nothing in syslog or messages. Compiz is enabled.

I vaguely recall that at least a couple of lock-ups have been associated with logging in from the screen-saver.

Attaching lspci output again.

Is there any other tests I can do while in lock-up state now during the week-end (I have to reboot on monday)?

[Bengt Olsson (bengt-blafs)]

Forget the last lspci attachment, didn't use the right switches. When using the -vvnn switches the output was identical from the previous lspci attachment (no diff).

[Bengt Olsson (bengt-blafs)]

Attaching present Xorg.0.log as well

[Bengt Olsson (bengt-blafs)]

Tried to get a backtrace according to https://wiki.ubuntu.com/Backtrace. I attached to the X process but could not interrupt it with Ctrl-c as decribed in the text, gdb died with Ctrl-c. Instead I skipped step 4 (continue) and did backtrace anyway. Don't know if it is any help but here it is anyway.

[Bryce Harrington (bryce)]

Thanks, please install debug packages for mesa, libdrm, and xserver-xorg-video-intel, and if you see the crash once more, please do a 'backtrace full', which will collect line numbers and symbols.

[Bryce Harrington (bryce)]

Btw, the " tossed event which came in late" and mieqEnequeue stuff are generic warnings that get printed out whenever the Xserver has locked up. Same warning is given for a huge variety of different bugs.

[Setting to incomplete waiting on full backtrace.]

[vouchy (stepabest)]

Hi,

I think this may be related to https://bugzilla.redhat.com/show_bug.cgi?id=465884

Rgs
vouchy

[Steve Langasek (vorlon)]

Bengt, are you still seeing this problem on your system? The bug has been incomplete, waiting on a full backtrace, for two months.

[mcNisse (johan-erlands)]

I have this X lock sometimes when I try to unlock the screensavever on two computers, running kubuntu intrepid with kde4.2. One with an intel card and the other one with an ati card.

[Bengt Olsson (bengt-blafs)]

Steve,

Sorry for not reporting before, but I changed back to Hardy and disabled Compiz in the end of January. This for general stability reasons, Hardy performs more solid on my T61, and I can't risk interruptions on my work computer. I think that vouchy's comment above is relevant to this problem as well as a number of others. My feeling is that it is the combination of the hw, the exa acceleration system, the intel driver that is in the core of the problem, just from scanning related posts. Use of Compiz seems to accelerate the problem. But I don't have any more hard facts, and since I changed back as described above I haven't had any more lockups.

So its fair for you to handle this bug as you see fit, since I can't contribute more.

[Steve Langasek (vorlon)]

Ok, I'm dropping the jaunty target for this bug accordingly.

There is still insufficient information here to link this bug to any other bugs, upstream or in other distributions. We need a full backtrace in order to establish that.

[Bryce Harrington (bryce)]

[Still waiting on backtrace]

[Bryce Harrington (bryce)]

We have a PPA with some new tools for debugging X freezes:

  https://launchpad.net/~ubuntu-x-swat/+archive/x-freeze-test

I know how irritating X freezes can be. They're also typically
quite hard to debug, but the information provided by these new
tools should help upstream figure them out.

You can help by doing the following:

 A. Install the PPA packages on Jaunty and boot kernel 2.6.30-rc2
 B. Reproduce your freeze
 C. ssh into the machine and run the steps to collect the info
 D. Attach the tarball of the results to this bug report

With this information, we'll be able to forward your bug upstream.

(For more information on triaging X freeze bugs, see
 https://wiki.ubuntu.com/X/Troubleshooting/Freeze )

[Carybielenberg (cary-bielenberg)]

Have the same problem with Jaunty Kubuntu 64bit, my machine is a R61 with the same hardware. It seems to go fine for about a hour then totally freeze & not respond to any keyboard or mouse activity.

[Bryce Harrington (bryce)]

@Cary, I suspect you probably have an unrelated freeze bug; several kubuntu users have reported similar findings, and it might be due to the Qt update in Jaunty.

Closing bug as expired since info requested in comment #24 has not been provided. Reopen with requested info if issue can be confirmed as still present in Karmic.

[[e-mail] (tc-progra-soft)]

Finally solved for Ubuntu 10.04 Gnome user
I have tested folowing drivers:
nvidia-current
nv,
nouveau,
nvidia closed:
NVIDIA-Linux-x86-173.14.25-pkg1.run
NVIDIA-Linux-x86-195.36.15-pkg1.run
NVIDIA-Linux-x86-195.36.24-pkg1.run
NVIDIA-Linux-x86-195.36.31-pkg1.run
NVIDIA-Linux-x86-256.53.run
NVIDIA-Linux-x86-260.19.04.run
with all these drivers I experienced minor ( artifacts only, artifacts and X crash ) or major problems ( crash of Xserver before login to gnome ).
I was lucky to get frequent traces in logs - minor crashes starts with the following statement in Xorg.0.log
[mi] EQ overflowing. The server is probably stuck in an infinite loop.
2 Weeks lost, with approx 50 X crashes. Linux are you Windows ME clone?
IT IS NOT ANY DRIVER FAULT!
The problem was solved by downgrading gdm package
from: 2.30.2.is.2.30.0-0ubuntu3
to: gdm 2.30.0-0ubuntu5
The last one is a good one!
BEWARE OF gdm 2.30.2.is.2.30.0-0ubuntu3 !!! Ticking Time bomb!
Best regards,
Tomasz