Frequent Freeze/Crash of Xorg with drivers 310.19 with GTS 250 on 3.2.0-4-amd64

Morning,

I got frequent Freeze/Crash of Xorg with drivers 310.19 with GTS 250 on 3.2.0-4-amd64 (current debian testing kernel)

It happen quickly when i use any 3D function (compositing of the gnome shell desktop, launching a game via wine), but if i let the pc running all the night after reboot, it can run all the night without crashing.

I can see in Xorg.0.conf an error and backtrace :

[ 19206.280] (WW) NVIDIA(0): WAIT (2, 6, 0x8000, 0x0000ac2c, 0x0000b2b0)
[ 19212.525] [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
[ 19212.525]
[ 19212.525] Backtrace:
[ 19212.550] 0: /usr/bin/Xorg (xorg_backtrace+0x36) [0x7f4de4d0ec66]
[ 19212.550] 1: /usr/bin/Xorg (mieqEnqueue+0x26b) [0x7f4de4ceff0b]
[ 19212.550] 2: /usr/bin/Xorg (0x7f4de4b90000+0x66892) [0x7f4de4bf6892]
[ 19212.550] 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7f4ddcd56000+0x55d8) [0x7f4ddcd5b5d8]
[ 19212.550] 4: /usr/bin/Xorg (0x7f4de4b90000+0x8d8d7) [0x7f4de4c1d8d7]
[ 19212.550] 5: /usr/bin/Xorg (0x7f4de4b90000+0xb1ba8) [0x7f4de4c41ba8]
[ 19212.550] 6: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7f4de3eb8000+0xf030) [0x7f4de3ec7030]
[ 19212.550] 7: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x7bd0b) [0x7f4ddd694d0b]
[ 19212.550] 8: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x7d254) [0x7f4ddd696254]
[ 19212.550] 9: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0xf27eb) [0x7f4ddd70b7eb]
[ 19212.550] 10: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0xb6d82) [0x7f4ddd6cfd82]
[ 19212.550] 11: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x4f902c) [0x7f4dddb1202c]
[ 19212.550] 12: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x4cc54a) [0x7f4dddae554a]
[ 19212.550] 13: /usr/bin/Xorg (BlockHandler+0x4a) [0x7f4de4be6bda]
[ 19212.550] 14: /usr/bin/Xorg (WaitForSomething+0x114) [0x7f4de4d0c024]
[ 19212.550] 15: /usr/bin/Xorg (0x7f4de4b90000+0x52b81) [0x7f4de4be2b81]
[ 19212.550] 16: /usr/bin/Xorg (0x7f4de4b90000+0x41ec5) [0x7f4de4bd1ec5]
[ 19212.550] 17: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xfd) [0x7f4de2be2ead]
[ 19212.551] 18: /usr/bin/Xorg (0x7f4de4b90000+0x4219d) [0x7f4de4bd219d]
[ 19212.551]
[ 19212.551] [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
[ 19212.551] [mi] mieq is NOT the cause. It is a victim.
[ 19213.061] [mi] EQ overflow continuing. 100 events have been dropped.
[ 19213.061]
[ 19213.061] Backtrace:
[ 19213.061] 0: /usr/bin/Xorg (xorg_backtrace+0x36) [0x7f4de4d0ec66]
[ 19213.061] 1: /usr/bin/Xorg (0x7f4de4b90000+0x66892) [0x7f4de4bf6892]
[ 19213.061] 2: /usr/lib/xorg/modules/input/evdev_drv.so (0x7f4ddcd56000+0x55d8) [0x7f4ddcd5b5d8]
[ 19213.062] 3: /usr/bin/Xorg (0x7f4de4b90000+0x8d8d7) [0x7f4de4c1d8d7]
[ 19213.062] 4: /usr/bin/Xorg (0x7f4de4b90000+0xb1ba8) [0x7f4de4c41ba8]
[ 19213.062] 5: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7f4de3eb8000+0xf030) [0x7f4de3ec7030]
[ 19213.062] 6: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x7bd0b) [0x7f4ddd694d0b]
[ 19213.062] 7: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x7d254) [0x7f4ddd696254]
[ 19213.062] 8: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0xf27eb) [0x7f4ddd70b7eb]
[ 19213.062] 9: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0xb6d82) [0x7f4ddd6cfd82]
[ 19213.062] 10: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x4f902c) [0x7f4dddb1202c]
[ 19213.062] 11: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f4ddd619000+0x4cc54a) [0x7f4dddae554a]
[ 19213.062] 12: /usr/bin/Xorg (BlockHandler+0x4a) [0x7f4de4be6bda]
[ 19213.062] 13: /usr/bin/Xorg (WaitForSomething+0x114) [0x7f4de4d0c024]
[ 19213.062] 14: /usr/bin/Xorg (0x7f4de4b90000+0x52b81) [0x7f4de4be2b81]
[ 19213.062] 15: /usr/bin/Xorg (0x7f4de4b90000+0x41ec5) [0x7f4de4bd1ec5]
[ 19213.062] 16: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xfd) [0x7f4de2be2ead]
[ 19213.062] 17: /usr/bin/Xorg (0x7f4de4b90000+0x4219d) [0x7f4de4bd219d]
[ 19213.062]
[ 19213.280] (WW) NVIDIA(0): WAIT (1, 6, 0x8000, 0x0000ac2c, 0x0000b2b0)
[ 19213.280] [mi] Increasing EQ size to 512 to prevent dropped events.
[ 19213.280] [mi] EQ processing has resumed after 151 dropped events.
[ 19213.280] [mi] This may be caused my a misbehaving driver monopolizing the server’s resources.
[ 19220.283] (WW) NVIDIA(0): WAIT (2, 6, 0x8000, 0x0000ac2c, 0x0000fd1c)
[ 19227.283] (WW) NVIDIA(0): WAIT (1, 6, 0x8000, 0x0000ac2c, 0x0000fd1c)
[ 19237.284] (EE) NVIDIA(GPU-0): WAIT: (E, 0, 0x827d)

In dmesg i have the following message too :
[19193.871515] NVRM: Xid (0000:01:00): 8, Channel 00000003
[19195.869021] NVRM: os_schedule: Attempted to yield the CPU while in atomic or interrupt context

Bug report archive is available here : http://khertan.net/datas/downloads/Others/nvidia-bug-report.log.gz.

I ll go back to 295.x.x driver for the moment which didn’t generate such problem.

(it could be also a config error, in this case thanks for any advice)

I comfirm this Freeze/Crash with GTS 250 (ASUS ENGTS250)

My /var/log/Xorg.0.log: http://pastebin.com/PyZp51qX

Ubuntu 12.10, KDE

nvidia-current 310.19-0ubuntu1~xedgers~quantal1

kernel 3.7.0-4-generic

with GTS 8800 640M this system is OK.

This is not unique to GTS 250. GTX 650, similar driver locking up problem, though this happens after running desktop for a while.

[ 44666.062] (WW) NVIDIA(1): WAIT (2, 8, 0x8000, 0x0000533c, 0x0000562c)
[ 44670.404] (WW) NVIDIA(1): WAIT (0, 8, 0x8000, 0x0000562c, 0x0000562c)
[ 44678.849] [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
[ 44678.849]
[ 44678.849] Backtrace:
[ 44678.881] 0: /usr/bin/X (xorg_backtrace+0x36) [0x7ffc6b0e0c66]
[ 44678.881] 1: /usr/bin/X (mieqEnqueue+0x26b) [0x7ffc6b0c1f0b]
[ 44678.881] 2: /usr/bin/X (0x7ffc6af62000+0x66892) [0x7ffc6afc8892]
[ 44678.881] 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7ffc62e3f000+0x55d8) [0x7ffc62e445d8]
[ 44678.882] 4: /usr/bin/X (0x7ffc6af62000+0x8d8d7) [0x7ffc6afef8d7]
[ 44678.882] 5: /usr/bin/X (0x7ffc6af62000+0xb1ba8) [0x7ffc6b013ba8]
[ 44678.882] 6: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7ffc6a28a000+0xf030) [0x7ffc6a299030]
[ 44678.882] 7: (vdso) (0x7fff327ed000+0x60c) [0x7fff327ed60c]
[ 44678.882] 8: (vdso) (__vdso_gettimeofday+0x125) [0x7fff327ed9f5]
[ 44678.882] 9: /lib/x86_64-linux-gnu/libc.so.6 (__gettimeofday+0x1a) [0x7ffc6902ff8a]
[ 44678.882] 10: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7ffc639c4000+0xee86e) [0x7ffc63ab286e]
[ 44678.882] 11: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7ffc639c4000+0x7d0ce) [0x7ffc63a410ce]
[ 44678.882] 12: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7ffc639c4000+0xf83ba) [0x7ffc63abc3ba]
[ 44678.882] 13: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7ffc639c4000+0x75710) [0x7ffc63a39710]
[ 44678.882] 14: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7ffc639c4000+0xa5a86) [0x7ffc63a69a86]
[ 44678.882] 15: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7ffc639c4000+0xa5e88) [0x7ffc63a69e88]
[ 44678.882] 16: /usr/bin/X (0x7ffc6af62000+0x96eae) [0x7ffc6aff8eae]
[ 44678.882] 17: /usr/bin/X (0x7ffc6af62000+0xd6dfb) [0x7ffc6b038dfb]
[ 44678.882] 18: /usr/bin/X (miUninstallColormap+0x96) [0x7ffc6b0bedb6]
[ 44678.882] 19: /usr/bin/X (0x7ffc6af62000+0x8b169) [0x7ffc6afed169]
[ 44678.882] 20: /usr/bin/X (0x7ffc6af62000+0x50860) [0x7ffc6afb2860]
[ 44678.882] 21: /usr/bin/X (0x7ffc6af62000+0x52e31) [0x7ffc6afb4e31]
[ 44678.882] 22: /usr/bin/X (0x7ffc6af62000+0x41ec5) [0x7ffc6afa3ec5]
[ 44678.882] 23: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xfd) [0x7ffc68fb1ead]
[ 44678.882] 24: /usr/bin/X (0x7ffc6af62000+0x4219d) [0x7ffc6afa419d]
[ 44678.882]
[ 44678.882] [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
[ 44678.882] [mi] mieq is NOT the cause. It is a victim.
[ 44679.807] (WW) NVIDIA(0): WAIT (2, 8, 0x8000, 0x0000807c, 0x0000a14c)
[ 44686.807] (WW) NVIDIA(0): WAIT (1, 8, 0x8000, 0x0000807c, 0x0000a14c)
[ 44686.807] [mi] Increasing EQ size to 512 to prevent dropped events.
[ 44686.807] [mi] EQ processing has resumed after 79 dropped events.
[ 44686.807] [mi] This may be caused my a misbehaving driver monopolizing the server’s resources.
[ 44689.809] (WW) NVIDIA(0): WAIT (2, 8, 0x8000, 0x0000807c, 0x0000a48c)
[ 44696.809] (WW) NVIDIA(0): WAIT (1, 8, 0x8000, 0x0000807c, 0x0000a48c)
[ 44699.810] (WW) NVIDIA(1): WAIT (2, 8, 0x8000, 0x00002ab4, 0x00002e7c)
[ 44706.810] (WW) NVIDIA(1): WAIT (1, 8, 0x8000, 0x00002ab4, 0x00002e7c)
[ 44709.812] (WW) NVIDIA(0): WAIT (2, 8, 0x8000, 0x0000807c, 0x0000a970)
[ 44716.812] (WW) NVIDIA(0): WAIT (1, 8, 0x8000, 0x0000807c, 0x0000a970)
[ 44719.813] (WW) NVIDIA(1): WAIT (2, 8, 0x8000, 0x00002ab4, 0x00003244)
[ 44726.813] (WW) NVIDIA(1): WAIT (1, 8, 0x8000, 0x00002ab4, 0x00003244)
[ 44729.814] (WW) NVIDIA(0): WAIT (2, 8, 0x8000, 0x0000807c, 0x0000af90)
[ 44736.814] (WW) NVIDIA(0): WAIT (1, 8, 0x8000, 0x0000807c, 0x0000af90)
[ 44739.816] (WW) NVIDIA(1): WAIT (2, 8, 0x8000, 0x00002ab4, 0x0000360c)
[ 44746.816] (WW) NVIDIA(1): WAIT (1, 8, 0x8000, 0x00002ab4, 0x0000360c)
[ 44749.818] (WW) NVIDIA(1): WAIT (2-S, 17, 0xdeadbeef, 0x00002ab4, 0x00009384)
[ 44756.818] (WW) NVIDIA(1): WAIT (1-S, 17, 0xdeadbeef, 0x00002ab4, 0x00009384)
[ 44759.819] (WW) NVIDIA(1): WAIT (2-S, 17, 0xdeadbeef, 0x00002ab4, 0x00009384)
[ 44766.819] (WW) NVIDIA(1): WAIT (1-S, 17, 0xdeadbeef, 0x00002ab4, 0x00009384)
[ 44769.820] (WW) NVIDIA(1): WAIT (2, 8, 0x8000, 0x00002ab4, 0x00009e98)
[ 44776.820] (WW) NVIDIA(1): WAIT (1, 8, 0x8000, 0x00002ab4, 0x00009e98)
[ 44779.821] (WW) NVIDIA(1): WAIT (2, 8, 0x8000, 0x00002ab4, 0x0000c3b8)
[ 44786.821] (WW) NVIDIA(1): WAIT (1, 8, 0x8000, 0x00002ab4, 0x0000c3b8)

This is with Debian X and custom Linux (3.6.7)

[ 116.975] X Protocol Version 11, Revision 0
[ 116.975] Build Operating System: Linux 3.2.0-4-amd64 x86_64 Debian
[ 116.975] Current Operating System: Linux mira 3.6.7+ #1 SMP Sat Nov 17 21:04:12 CST 2012 x86_64
[ 116.975] Kernel command line: BOOT_IMAGE=/vmlinuz-3.6.7+ root=/dev/mapper/raid1-root ro quiet
[ 116.975] Build Date: 30 September 2012 10:53:24AM
[ 116.976] xorg-server 2:1.12.4-1 (Julien Cristau )

Also a problem for 313.09 on GTS 250

nvidia-bug-report.log.gz: http://dl.dropbox.com/u/15965032/nvidia-bug-report.log.gz

The problem is still not resolved. Do you need any additional information that would help fix the driver?

Xorg.0.log: http://pastebin.com/YKazsZWm

Kubuntu 12.10

Driver 313.09-0ubuntu1~xedgers~quantal2 from ppa:xorg-edgers/ppa

$ uname -a
Linux turbid-pc 3.7.0-7-generic #15-Ubuntu SMP Sat Dec 15 16:34:25 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux
$
$ lspci | grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation G92 [GeForce GTS 250] (rev a2)

Its would be good if you all provide reproduction steps .

  1. kernel loads

  2. X11 is loaded and lightdm

  3. I enter the username and password

  4. loaded plasma-desktop

  5. after 5-10 seconds I got Freeze, no response from the keyboard (including Ctr-Alt-Fx, Ctrl-Alt-Backspace). Images are not damaged, but no movement in the picture there. Just mouse over reacts and moves in jumps.

  6. Reboot the system power.

It happened to me about once a week (I don’t turn off the pc) completely at random. Usually while playing games but also just on desktop. I had to login via SSH from tablet and restart X. Mouse worked fine though. The only “solution” I’ve found is installing 32 bit system (also switched from Debian to Ubuntu). Maybe it’s x86_64-specific bug. It was worse before 304.64 (can’t say for sure) and started around 295.93 when Team Fortress 2 on Wine freezed X after ~20-40 minutes of gameplay. On 304.64 or so it was fixed but once a week freeze happened anyway. Now I’m on Ubuntu 12.04 x86_32 with Awesome WM and everything is perfectly stable. As I see by Turbid’s post this freeze happens on Ubuntu x86_64 as well so it’s not a distribution issue.

Any progress on bug fixing?

Also a problem for 313.26 xorg-edgers ppa on Kubuntu 13.04 with GTS 250

$ uname -a
Linux turbid-pc 3.8.0-13-generic #23-Ubuntu SMP Mon Mar 18 18:09:44 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
$ lspci -v
...
...

01:00.0 VGA compatible controller: NVIDIA Corporation G92 [GeForce GTS 250] (rev a2) (prog-if 00 [VGA controller])
	Subsystem: ASUSTeK Computer Inc. Device 8339
	Flags: fast devsel, IRQ 16
	Memory at fd000000 (32-bit, non-prefetchable) 
	Memory at d0000000 (64-bit, prefetchable) 
	Memory at fa000000 (64-bit, non-prefetchable) 
	I/O ports at ec00 
	[virtual] Expansion ROM at fc000000 [disabled] 
	Capabilities: <access denied>
	Kernel driver in use: nvidia

/var/log/Xorg.0.log: http://pastebin.com/aJqHgVnu

nvidia-bug-report.log.gz http://dl.dropbox.com/u/15965032/nvidia-bug-report.log.gz

Any progress on bug fixing?

Hi Khertan,

I believe Franster is correct. This problem is occurring on other hardware.
I’m seeing it on Quadro FX 2800M as well.

https://devtalk.nvidia.com/default/topic/537302/linux/both-screens-black-xorg-at-100-cpu-overflow-errors-in-xorg-0-log-nvidia-driver-310-14-quadro-fx/

Other forum topics are also seeing the same.

https://devtalk.nvidia.com/default/topic/535519/linux/x-hangs-using-100-cpu-wait-and-mieq-overflowing-errors-in-logs/

and

https://devtalk.nvidia.com/default/topic/534892/linux/x-freeze-with-eq-overflows/

Please let me know if you find a resolution to this issue…

Cryptor

I can confirm the same bug. I’m running Kubuntu 12.10
OpenGL renderer string: GeForce GTX 560/PCIe/SSE2
OpenGL version string: 4.3.0 NVIDIA 319.12

I was getting the same error on the 313 drivers and hoped switching to 319 would solve the problem. I’ve also tried disabling my second monitor and compositing to see if either of those were causing the issues.

I have written a GL program that seems to be pretty good at causing the problem. Unfortunately I don’t know what aspect of it is doing so. If there are any suggestions on what I can do to get more information about what may be causing the issue let me know.

Any chance you could make your code available? I have a similar issue to this (Cryptor mentioned my thread, which is here) and I’d like to see if your program triggers the fault on my system.

Unfortunately I can’t. This bug seems very much like a bug I ran into about year ago when we were trying to move our render farm over to Linux. The machines we ran on were capable of running 4 renders at once without breaking a sweat but eventually X would lock up and you’d get log files as above. We ended up concluding that there was some timing lock up between nvidia and X.

Now I’m getting it on my desktop running an entirely new piece of software. Usually the application runs fine though in this case we’re using compute shaders to do quite a bit of work on the GPU. We have 2 different threads sending gl calls to the GPU, but usually a fault there causes the application to crash, not X to hang.

I can try and generate a trace of the gl calls for you to play back.

Kubuntu 13.04, nvidia 319 - the problem is not solved

there is also the issue in Win7 with 310 drivers: https://forums.geforce.com/default/topic/532102/gts-250-1gb-freezing/

Nvidia, you’ll fix it? :)

A little bit more of an update. I’ve managed to narrow down the code to where we upload quite some large chunks of data to the card. It would appear then that the problem is caused either by saturating the bus to the card, or by using all the memory on the card. To test out this theory I swapped my 560 TI for a 570 and everything worked beautifully. Swap back to the 560 and it hangs. So at this point I’m thinking that when the card runs out of memory the driver gets itself stuck in a lock of some sort and somehow this ends up holding up X.

One other thing I noticed. At one point during X hanging I decided to leave it for awhile and see if it recovered. As the rest of the application appears to continue running fine eventually it got to a point where it refreshed everything (i.e. cleared memory and reloaded). For a few seconds after this I could move the mouse and interact fine but then it hung again.

How much memory do your cards have? My 560 Ti has 1Gb. I haven’t noticed GPU memory usage on my system being particularly high when the lockups occur, it’s been between 150Mb and 300Mb. I don’t have another 5xx series card to test with but I had no problems with a 3.7 kernel, 304.64 drivers and a 5500FX card.

I’ve also had it recover very briefly once or twice. The Xorg log file usually mentions the event queue has resumed processing after dropping 200 or so events but it starts backing up again a few seconds later.

Same issue with Nvidia Driver 304.88 & GeForce 8400M G.

DebianVersion 7.0 (wheezy) (64-Bit)
Kernel Linux 3.2.0-4-amd64
GNOME 3.4.2

I have used the same system with Ubuntu 10.04 LTS over 3 years with no problems. The Driver version was 195.xx .

I hope NVIDIA will fix the problem very soon.

can you check this bug with 319.23?

I have checked it with the newest available Version in the Debian Backports:

(II) NVIDIA GLX Module 319.17 Thu Apr 25 21:41:50 PDT 2013

[ 68.168] () NVIDIA(0): Using HorizSync/VertRefresh ranges from the EDID for display
[ 68.168] (
) NVIDIA(0): device Nvidia Default Flat Panel (DFP-0) (Using EDID
[ 68.168] (**) NVIDIA(0): frequencies has been enabled on all display devices.)
[ 290.369] (WW) NVIDIA(0): WAIT (2, 4, 0x8000, 0x00009714, 0x0000a6d8)
[ 292.618] [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
[ 292.618]
[ 292.618] Backtrace:
[ 292.659] 0: /usr/bin/Xorg (xorg_backtrace+0x36) [0x7f6dec9d8d06]
[ 292.659] 1: /usr/bin/Xorg (mieqEnqueue+0x26b) [0x7f6dec9b9fab]
[ 292.659] 2: /usr/bin/Xorg (0x7f6dec85a000+0x668f2) [0x7f6dec8c08f2]
[ 292.659] 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7f6de44ae000+0x55d8) [0x7f6de44b35d8]
[ 292.659] 4: /usr/bin/Xorg (0x7f6dec85a000+0x8d937) [0x7f6dec8e7937]
[ 292.659] 5: /usr/bin/Xorg (0x7f6dec85a000+0xb1c08) [0x7f6dec90bc08]
[ 292.659] 6: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7f6debb82000+0xf030) [0x7f6debb91030]
[ 292.659] 7: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f6de519c000+0x631db) [0x7f6de51ff1db]
[ 292.659] 8: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f6de519c000+0x63ba1) [0x7f6de51ffba1]
[ 292.659] 9: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f6de519c000+0xd5fd6) [0x7f6de5271fd6]
[ 292.659] 10: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0x7f6de519c000+0x485318) [0x7f6de5621318]
[ 292.659] 11: /usr/bin/Xorg (0x7f6dec85a000+0x16ef93) [0x7f6dec9c8f93]
[ 292.659] 12: /usr/bin/Xorg (0x7f6dec85a000+0xd7080) [0x7f6dec931080]
[ 292.659] 13: /usr/bin/Xorg (0x7f6dec85a000+0x109238) [0x7f6dec963238]
[ 292.659] 14: /usr/bin/Xorg (0x7f6dec85a000+0x52e61) [0x7f6dec8ace61]
[ 292.659] 15: /usr/bin/Xorg (0x7f6dec85a000+0x41ec5) [0x7f6dec89bec5]
[ 292.659] 16: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xfd) [0x7f6dea8a9ead]
[ 292.659] 17: /usr/bin/Xorg (0x7f6dec85a000+0x4219d) [0x7f6dec89c19d]

In version 304.88 only the X11 crashes and restarts, in 319.17 the whole Linux crashes !