nVidia driver 361 problem causing lockup at log in

[oshunluvr]

This is my third attempt at a 16.04 install (for various reasons). About 2-3 hours into configuring and playing around, all of a sudden I could no longer log in. The login splash would display and sit forever (or as long as I had the patience for). I have now idea what I installed or changed that might have caused the problem.

I booted back into my old 15.04 install and looked at the log files on 16.04 and found hundreds of lines that looked like this:

(WW) NVIDIA(0): WAIT (2, 8, 0x8000, 0x0000c46c, 0x0000c474)

Sorry I don't have the exact errors, apparently the fix I did also wiped out the old log files - which is odd because in past installs it would just rename them with a "1".

This error had occurred similarly on my previous attempt to use 16.04, but the error messages in xorg.0.log were different - more verbose:

[ 35.656] (WW) NVIDIA(0): WAIT (2, 8, 0x8000, 0x0000c46c, 0x0000c474)(EE) [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
(EE)
(EE) Backtrace:
(EE) 0: /usr/lib/xorg/Xorg (xorg_backtrace+0x4e) [0x55922fff050e]
(EE) 1: /usr/lib/xorg/Xorg (mieqEnqueue+0x253) [0x55922ffd1fd3]
(EE) 2: /usr/lib/xorg/Xorg (QueuePointerEvents+0x52) [0x55922feaa672]
(EE) 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fb1133dd000+0x61f3) [0x7fb1133e31f3]
(EE) 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fb1133dd000+0x6a5d) [0x7fb1133e3a5d]
(EE) 5: /usr/lib/xorg/Xorg (0x55922fe3e000+0x94248) [0x55922fed2248]
(EE) 6: /usr/lib/xorg/Xorg (0x55922fe3e000+0xb9712) [0x55922fef7712]
(EE) 7: /lib/x86_64-linux-gnu/libc.so.6 (0x7fb11960b000+0x354a0) [0x7fb1196404a0]
(EE) 8: (vdso) (__vdso_gettimeofday+0xd9) [0x7ffe70fa9e09]
(EE) 9: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0xad37e) [0x7fb1147d337e]
(EE) 10: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0x5882e) [0x7fb11477e82e]
(EE) 11: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0xb8b8d) [0x7fb1147deb8d]
(EE) 12: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/libglx.so (0x7fb116fd1000+0x80bad6) [0x7fb1177dcad6]
(EE)
(EE) [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
(EE) [mi] mieq is *NOT* the cause. It is a victim.
(EE) [mi] EQ overflow continuing. 100 events have been dropped.
(EE)
(EE) Backtrace:
(EE) 0: /usr/lib/xorg/Xorg (xorg_backtrace+0x4e) [0x55922fff050e]
(EE) 1: /usr/lib/xorg/Xorg (QueuePointerEvents+0x52) [0x55922feaa672]
(EE) 2: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fb1133dd000+0x61f3) [0x7fb1133e31f3]
(EE) 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fb1133dd000+0x6a5d) [0x7fb1133e3a5d]
(EE) 4: /usr/lib/xorg/Xorg (0x55922fe3e000+0x94248) [0x55922fed2248]
(EE) 5: /usr/lib/xorg/Xorg (0x55922fe3e000+0xb9712) [0x55922fef7712]
(EE) 6: /lib/x86_64-linux-gnu/libc.so.6 (0x7fb11960b000+0x354a0) [0x7fb1196404a0]
(EE) 7: (vdso) (0x7ffe70fa9000+0x9a9) [0x7ffe70fa99a9]
(EE) 8: (vdso) (__vdso_gettimeofday+0x12f) [0x7ffe70fa9e5f]
(EE) 9: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0xad37e) [0x7fb1147d337e]
(EE) 10: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0x5882e) [0x7fb11477e82e]
(EE) 11: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0xb8b8d) [0x7fb1147deb8d]
(EE) 12: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/libglx.so (0x7fb116fd1000+0x80bad6) [0x7fb1177dcad6]
(EE)
(EE) [mi] EQ overflow continuing. 200 events have been dropped.
(EE)
(EE) Backtrace:
(EE) 0: /usr/lib/xorg/Xorg (xorg_backtrace+0x4e) [0x55922fff050e]
(EE) 1: /usr/lib/xorg/Xorg (QueuePointerEvents+0x52) [0x55922feaa672]
(EE) 2: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fb1133dd000+0x61f3) [0x7fb1133e31f3]
(EE) 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fb1133dd000+0x6a5d) [0x7fb1133e3a5d]
(EE) 4: /usr/lib/xorg/Xorg (0x55922fe3e000+0x94248) [0x55922fed2248]
(EE) 5: /usr/lib/xorg/Xorg (0x55922fe3e000+0xb9712) [0x55922fef7712]
(EE) 6: /lib/x86_64-linux-gnu/libc.so.6 (0x7fb11960b000+0x354a0) [0x7fb1196404a0]
(EE) 7: (vdso) (0x7ffe70fa9000+0x9ad) [0x7ffe70fa99ad]
(EE) 8: (vdso) (__vdso_gettimeofday+0x12f) [0x7ffe70fa9e5f]
(EE) 9: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0xad37e) [0x7fb1147d337e]
(EE) 10: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0x5882e) [0x7fb11477e82e]
(EE) 11: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/nvidia_drv.so (0x7fb114726000+0xb8b8d) [0x7fb1147deb8d]
(EE) 12: /usr/lib/x86_64-linux-gnu/xorg/extra-modules/libglx.so (0x7fb116fd1000+0x80bad6) [0x7fb1177dcad6]
(EE)
[ 42.656] (WW) NVIDIA(0): WAIT (1, 8, 0x8000, 0x0000c46c, 0x0000c474)
[ 42.656] [mi] Increasing EQ size to 1024 to prevent dropped events.
[ 42.656] [mi] EQ processing has resumed after 251 dropped events.
[ 42.656] [mi] This may be caused by a misbehaving driver monopolizing the server's resources.

Finding no relief on the internet, I logged into the text console and removed nvidia-361 and installed nvidia-340 instead. It appears you can't install -346 or -352 without dragging in -361. Then rebooted. I am now posting from my 16.04 install.

I can only speculate that the issue is related to the nvidia driver and desktop effects and/or the kernel. I changed a few effects settings and some other things but I wasn't keeping track of what I changed so who knows at this point. I'll likely stay with -340 until several updates to the driver and kernel have occurred. Hopefully, the problem won't return.

[Snowhog]

Now that you have a working 16.04, have you installed dkms?

[oshunluvr]

Now that you have a working 16.04, have you installed dkms?

Yes, you have to in order to install the nvidia driver. I also use virtualbox which also requires dkms.

[vinnywright]

I dont know what may be your problem ,,,,,, but I have been using the drivers from @hear for a wile now ,,,,

#graphics drivers (nvidia)
deb http://ppa.launchpad.net/graphics-drivers/ppa/ubuntu/ xenial main

and am on the nvidia-364 right now .

I am having a similer problem on my 14.04 install where the login screen just sits their forever after tying to login ,,,,,but from some logs I think it's a PAM issue .

switching to a TTY and running "startx" gets me in ,,,,,,,,,, it may (running "startx" from a TTY)give you some more info on where it's braking.

however I am also on

deb http://archive.neon.kde.org/stable/ xenial main

So

VINNY

[oshunluvr]

Yeah Vinny, I just added that ppa last night but haven't yet tried the newer driver. I'm travelling this week so I won't have time until I return.

[oshunluvr]

OK, upgraded to the nvidia-367 drivers. It worked for a while, but started the same behavior after a few hours (much longer than before).

I did some research and ran across similar errors over the years. Several of them referenced default memory allocation not being substantial enough for composting and 3d graphics. This prompted a memory that I had used a kernel parameter in the past to supposedly increase performance: "vmalloc=" (virtual memory allocation, I think?). Some of these older posts had stated increasing this was the solution, so I edited default/grab and added "vmalloc=512M" to it. Supposedly the default is only 128M. Possibly my older hardware (circa 2006) doesn't have the ability to increase this without help.

Anyway, since I have done this no lockups so far (about 2 days). Fingers crossed and knock wood!

[oshunluvr]

Well, back to 340. 367 worked until the latest upgrade, then started the same WAIT behavior. Seems like nVidia is slipping...

[oshunluvr]

Now on Neon. Same issue with nvidia drivers occurs.