xorg crashes randomly with seg fault

[Stephan]

This is similar to bug 578750, but on my system this particular bug only started to occur recently.

I'm getting frequent crashes of the X server with this stacktrace in the logs:

Backtrace:
0: /usr/bin/X (xorg_backtrace+0x3b) [0x80e938b]
1: /usr/bin/X (0x8048000+0x61c8d) [0x80a9c8d]
2: (vdso) (__kernel_rt_sigreturn+0x0) [0x7c8410]
Segmentation fault at address 0x40080884

This was without debugging symbols, a version after I installed xorg with debugging symbols can be found in Xorg.0.log.old

More info:
I'm using "nv" driver with TwinView (via xorg.conf) and this configuration has been working pretty stable for a while. I can't precisely say, when instability returned, but possibly the upgrade of xserver-xorg-core to 2:1.7.6-2ubuntu7.3 escalated this situation. I did the upgrade three days ago, and while it feels that the situation started earlier than that, during the last (3) days the system crashed several times a day (or even: hour).

Suspecting TwinView to have a finger in the pie I disabled TwinViw but X crashed only a few seconds after restarting.

It can't see a connection to any application, usually I'm in Eclipse (with gtk) when it happens, but that's only where I spend most of the time. The crash might be related to either some mouse activity (never saw it without doing something with the mouse). One crash, e.g., happened about when I clicked on the (KDE) desktop.

Another symptom I've been seeing is not crashing but freezing (no reaction to any input device, including not-working keyboard LEDs) - not sure if this is related. Usually, when the system freezes first the full screen flickers shortly before the freeze - and I believe this symptom equally applies to the freeze and the crash. At other times the flicker occurs and after a second or so that system becomes responsive again (micro freeze, you might say).

Please let me know what further information would help to identify the problem.

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: xserver-xorg-core 2:1.7.6-2ubuntu7.3
ProcVersionSignature: Ubuntu 2.6.32-24.39-generic 2.6.32.15+drm33.5
Uname: Linux 2.6.32-24-generic i686
NonfreeKernelModules: nvidia
Architecture: i386
Date: Thu Aug 19 16:45:28 2010
DkmsStatus:
 nvidia-current, 195.36.24, 2.6.32-23-generic, i686: installed
 nvidia-current, 195.36.24, 2.6.32-24-generic, i686: installed
Lsusb:
 Bus 002 Device 003: ID 045e:00dd Microsoft Corp.
 Bus 002 Device 002: ID 046d:c050 Logitech, Inc. RX 250 Optical Mouse
 Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 001 Device 004: ID 058f:6362 Alcor Micro Corp. Hi-Speed 21-in-1 Flash Card Reader/Writer (Internal/External)
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
MachineType: System manufacturer System Product Name
ProcCmdLine: root=UUID=0b16b3df-690d-4404-afc9-9560c74fdc5a ro quiet splash noapic
ProcEnviron:
 LANGUAGE=en_AU:en
 PATH=(custom, user)
 LANG=en_AU.UTF-8
 SHELL=/bin/bash
SourcePackage: xorg-server
dmi.bios.date: 06/15/2006
dmi.bios.vendor: Phoenix Technologies, LTD
dmi.bios.version: ASUS M2N-SLI DELUXE ACPI BIOS Revision 0304
dmi.board.name: M2N-SLI DELUXE
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: 1.XX
dmi.chassis.asset.tag: 123456789000
dmi.chassis.type: 3
dmi.chassis.vendor: Chassis Manufacture
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnPhoenixTechnologies,LTD:bvrASUSM2N-SLIDELUXEACPIBIOSRevision0304:bd06/15/2006:svnSystemmanufacturer:pnSystemProductName:pvrSystemVersion:rvnASUSTeKComputerINC.:rnM2N-SLIDELUXE:rvr1.XX:cvnChassisManufacture:ct3:cvrChassisVersion:
dmi.product.name: System Product Name
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer
system:
 distro: Ubuntu
 codename: lucid
 architecture: i686
 kernel: 2.6.32-24-generic

[Stephan (stephan-h)]

As I just had another crash (exact same backtrace), I found that it correlates
also to these entries in syslog:

[ 2610.985055] NVRM: Xid (0007:00): 6, PE0001
...

Interestingly, this problem is logged several times without crashing
(might correlate with those micro-freezes I mentioned?),
but eventually the NVRM: Xid precedes the death of X and kdm
(in this case with a 7 second delay, hm)

That machine worked pretty stable until some days ago, but at this point
it is not usable for work: on one hand crashes happen frequently,
but I cannot see it correlate to any specific action that I could avoid.

All this implies that I'm more than willing to provide additional information
for analysing the problem.

[Stephan (stephan-h)]

Update: I gave it another chance after finding bug 526857 comment 5.
I applied that workaround and the machine has been running now for
a few hours without crash nor freeze.

I still had a few
  NVRM: Xid (0007:00): 6, PE0001
which did not correlate to any observable freeze/flicker, then:
Aug 21 18:19:49 luna kernel: [ 5907.594173] NVRM: Xid (0007:00): 3, C 00000001 SC 00000002 M 00000860 Data ffffff78
Aug 21 19:06:59 luna kernel: [ 8736.997129] NVRM: Xid (0007:00): 3, C 00000001 SC 00000002 M 000002ac Data 00000004
Aug 21 19:07:03 luna kernel: [ 8741.289690] NVRM: Xid (0007:00): 3, C 00000001 SC 00000002 M 000002ac Data 00000004
Aug 21 19:07:07 luna kernel: [ 8745.370445] NVRM: Xid (0007:00): 3, C 00000001 SC 00000002 M 000002ac Data 00000004
which caused a flicker and a temporary freeze. The time between
log-entries could well be equal to the time of the freeze.

[Stephan (stephan-h)]

Ouch, machine died again (X crash this time):

Backtrace:
0: /usr/bin/X (xorg_backtrace+0x3b) [0x80e938b]
1: /usr/bin/X (0x8048000+0x61c8d) [0x80a9c8d]
2: (vdso) (__kernel_rt_sigreturn+0x0) [0x461410]
Segmentation fault at address 0x40300884

Caught signal 11 (Segmentation fault). Server aborting

This time nothing with NVRM: Xid.
Unsure if this means that these are two different bugs.
At least now we know that the workaround from bug 526857 comment 5
is no solution for this problem.

[Stephan (stephan-h)]

Some more updates:

- I suspected a broken CD drive to influence the behavior, disconnected the CD and again saw some crashes.

- I tried all 3 performance levels (see bug 526857 comment 5), X still crashes in all these configurations.

- I observed one freeze (not crash) while logging into KDE, no mouse action at that time. (see above hypothesis about a connection between mouse activity and the bug). Because it was a hard freeze (not even sys-req emergency umount or emergency reboot worked), no traces of the problem could be found on the machine.

-----------------------

Given the lack of activity in this report, let me ask whether there is any use in providing more info in this bug?

I intentionally did not try any other versions of any software involved, because the infected machine could be a good testbed for future attempts to narrow down what exactly is broken here.

However, I can't wait forever, since, as mentioned repeatedly, the machine is in this state unusable for any serious work.
I will have to install a working OS on this machine soon.

[Per Wahlström (per-wahlstrom)]

Running Ubuntu 9.10, and after replacing motherboard, CPU and RAM, this started happening to me too.

Old setup, ----> no freezez/crashes:

MB: Asus P5W DH Deluxe
CPU: Core 2 Duo E6600
RAM: 4x1 Gb Corsair whatever
Asus nvidia 8800GT card
Ubuntu 9.10, kernel 2.6.31-14
nvidia driver version: 185.18.36

New setup ---> freezes occur:

MB: Asus P6T SE
CPU: Core i7 930
RAM: 6x2Gb Corsair XMS3 1333MHz
Same Asus nvidia 8800GT card
Same ubuntu installation on same HD

Running glxgears is a sure way to reproduce the system freeze: causes almost instant freeze, every time.

Watching video will cause a freeze after some time, if I have switched desktop with compiz fusion 3D cube desktop before.
Switching CPU to 1 core enables me to run glxgears for some time, but still occasional freeze can happen when for example searching on google and the search-box auto-complete is active.

Syslog after a 1-second 'mini-freeze':
kernel: [11144.963412] NVRM: Xid (0002:00): 6, PE007f
kernel: [11144.965148] NVRM: Xid (0002:00): 7, Ch 0000007f M 00001ffc D ffffffff intr ffffffff
kernel: [11144.966874] NVRM: Xid (0002:00): 26, Ch 0000007f M 00001ffc D ffffffff intr ffffffff
kernel: [11144.968600] NVRM: Xid (0002:00): 4, Ch 00000005 acquireValue 11111111 dmaPut 2004be6c dmaGet 2004be6c

Syslog after a total freeze 25 seconds later:
kernel: [11170.422373] NVRM: Xid (0002:00): 6, PE007f
kernel: [11170.425820] NVRM: Xid (0002:00): 7, Ch 0000007f M 00001ffc D ffffffff intr ffffffff
kernel: [11170.429223] NVRM: Xid (0002:00): 26, Ch 0000007f M 00001ffc D ffffffff intr ffffffff
kernel: [11181.488364] NVRM: Xid (0002:00): 8, Channel 00000005

Tried installing latest driver from nvidia (256.53), glxgears causes instant freeze here too.

To recap, new MB, CPU, RAM made the freezes start occurring.

[Mike Homer (homerhomer)]

I'm having this too. :(
[ 247.613222] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001518 3f800000 0000000d
[ 248.530114] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001518 3f800000 0000000d
[ 249.089444] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001514 00000000 0000000d
[ 251.625693] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001b0c 01007002 00000003
[ 255.682719] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001518 3f800000 0000000d
[ 256.439137] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001420 00000000 00000004
[ 262.402521] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001340 00000006 00000005
[ 278.367232] NVRM: Xid (0002:00): 13, 0005 00000000 00008297 00001b0c 10007010

[dino99 (9d9)]

That version is no more supported; please open a new bug report if the actual archive found version also has the same issue.

[Stephan (stephan-h)]

"invalid" is a funny classification for a bug that made Ubuntu unusable on certain machines at the time of reporting.

After 5 years with no action you may be sure that I'm not using a nvidia card/driver any more, bye.