X server crashes due to nvidia-driver segmentation faults
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
nvidia-common (Ubuntu) |
Confirmed
|
Undecided
|
Unassigned |
Bug Description
1)
Description: Ubuntu precise (development branch)
Release: 12.04
2)
LC_ALL=C apt-cache policy nvidia-common
nvidia-common:
Installed: 1:0.2.39
Candidate: 1:0.2.39
Version table:
*** 1:0.2.39 0
500 http://
100 /var/lib/
3)
I expected standard behaviour like opening the program by clicking on its icon in the launcher bar
4)
Instead I got a system freeze
My X is freezing and often only a hard reboot helps unfreezing the system. The situation is so bad that from 3 boots 2 are reboots due to freezes. The system usually freezes when doing any 3d-related stuff, when using multimedia applications like flash, or when clicking on the unity-shell (or whatever the name of the programmbar on the left is). The system runs comparable stable (but also freezes often but also is able to unfreeze), when not using nvidia drivers. The same situation was present at ubuntu 11.10.
I have several suggestions grounded on facts but maybe this is a software bug:
a) The laptop has damaged (3d-related) hardware (but neither windows sisoft sandra, nor memtestX86, nor two ther vram and gpu utilities report any problems).
b) There are some bad drivers (Since I was doing cuda programming using the ppa: "http://
Anyway, in case of a X freeze the x logfiles displays many of those lines:
[ 952.821] [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
[ 952.821]
Backtrace:
[ 952.822] 0: /usr/bin/X (xorg_backtrace
[ 952.822] 1: /usr/bin/X (mieqEnqueue+0x263) [0x7fe6b40a39d3]
[ 952.822] 2: /usr/bin/X (0x7fe6b3f3b000
[ 952.822] 3: /usr/lib/
[ 952.822] 4: /usr/bin/X (0x7fe6b3f3b000
[ 952.822] 5: /usr/bin/X (0x7fe6b3f3b000
[ 952.822] 6: /lib/x86_
[ 952.822] 7: /usr/lib/
[ 952.822] 8: /usr/lib/
[ 952.822] 9: /usr/lib/
[ 952.822] 10: /usr/lib/
[ 952.822] 11: /usr/lib/
[ 952.822] 12: /usr/lib/
[ 952.822] 13: /usr/lib/
[ 952.822] 14: /usr/lib/
[ 952.822] 15: /usr/lib/
[ 952.822] 16: /usr/lib/
[ 952.822] 17: /usr/lib/
[ 952.822] 18: /usr/lib/
[ 952.822] 19: /usr/lib/
[ 952.822] 20: /usr/bin/X (BlockHandler+0x4a) [0x7fe6b3f8d4ca]
[ 952.822] 21: /usr/bin/X (WaitForSomethi
[ 952.822] 22: /usr/bin/X (0x7fe6b3f3b000
[ 952.822] 23: /usr/bin/X (0x7fe6b3f3b000
[ 952.822] 24: /lib/x86_
[ 952.822] 25: /usr/bin/X (0x7fe6b3f3b000
[ 952.822] [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
[ 952.822] [mi] mieq is *NOT* the cause. It is a victim.
[ 953.253] [mi] EQ overflow continuing. 100 events have been dropped.
[ 953.253]
Backtrace:
[ 953.253] 0: /usr/bin/X (xorg_backtrace
[ 953.253] 1: /usr/bin/X (0x7fe6b3f3b000
[ 953.253] 2: /usr/lib/
[ 953.253] 3: /usr/bin/X (0x7fe6b3f3b000
[ 953.254] 4: /usr/bin/X (0x7fe6b3f3b000
[ 953.254] 5: /lib/x86_
[ 953.254] 6: /usr/lib/
[ 953.254] 7: /usr/lib/
[ 953.254] 8: /usr/lib/
[ 953.254] 9: /usr/lib/
[ 953.254] 10: /usr/lib/
[ 953.254] 11: /usr/lib/
[ 953.254] 12: /usr/lib/
[ 953.254] 13: /usr/lib/
[ 953.254] 14: /usr/lib/
[ 953.254] 15: /usr/lib/
[ 953.254] 16: /usr/lib/
[ 953.254] 17: /usr/lib/
[ 953.254] 18: /usr/lib/
[ 953.254] 19: /usr/bin/X (BlockHandler+0x4a) [0x7fe6b3f8d4ca]
[ 953.254] 20: /usr/bin/X (WaitForSomethi
[ 953.254] 21: /usr/bin/X (0x7fe6b3f3b000
[ 953.254] 22: /usr/bin/X (0x7fe6b3f3b000
[ 953.254] 23: /lib/x86_
[ 953.254] 24: /usr/bin/X (0x7fe6b3f3b000
[ 989.285] [mi] EQ overflow continuing. 200 events have been dropped.
[ 989.285]
Backtrace:
... and so on.
If you need any more information, please let me know.
ProblemType: Bug
DistroRelease: Ubuntu 12.04
Package: nvidia-common 1:0.2.39
ProcVersionSign
Uname: Linux 3.2.0-17-generic x86_64
NonfreeKernelMo
ApportVersion: 1.94-0ubuntu1
Architecture: amd64
Date: Sun Mar 4 22:47:02 2012
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Release amd64 (20111012)
ProcEnviron:
LANGUAGE=de:en
TERM=xterm
PATH=(custom, user)
LANG=de_DE.UTF-8
SHELL=/bin/bash
SourcePackage: nvidia-common
UpgradeStatus: Upgraded to precise on 2012-03-03 (1 days ago)
Status changed to 'Confirmed' because the bug affects multiple users.