Bug#739179: Still happens with 340 series driver

[Tim Gokcen]

Seems to still happen with the 340 series of drivers. I just had two 
crashes in a row; the first one was a freeze while running, and the 
second happened before X even finished booting when I restarted. 
Relevant packages installed:

ii glx-alternative-nvidia 0.5.1~bpo70+1 amd64
ii libegl1-nvidia:amd64 340.65-2~bpo70+1 amd64
ii libgl1-nvidia-glx:amd64 340.65-2~bpo70+1 amd64
ii libgl1-nvidia-glx:i386 340.65-2~bpo70+1 i386
ii libgl1-nvidia-glx-i386 340.65-2~bpo70+1 i386
ii libgles1-nvidia:amd64 340.65-2~bpo70+1 amd64
ii libgles2-nvidia:amd64 340.65-2~bpo70+1 amd64
ii libnvidia-eglcore:amd64 340.65-2~bpo70+1 amd64
ii libnvidia-ml1:amd64 340.65-2~bpo70+1 amd64
ii linux-headers-3.16.0-0.bpo.4-all 3.16.7-ckt2-1~bpo70+1 amd64
ii linux-headers-3.16.0-0.bpo.4-all-amd64 3.16.7-ckt2-1~bpo70+1 amd64
ii linux-headers-3.16.0-0.bpo.4-amd64 3.16.7-ckt2-1~bpo70+1 amd64
ii linux-headers-3.16.0-0.bpo.4-common 3.16.7-ckt2-1~bpo70+1 amd64
ii linux-image-3.16.0-0.bpo.4-amd64 3.16.7-ckt2-1~bpo70+1 amd64
ii linux-image-amd64 3.16+63~bpo70+1 amd64
ii nvidia-alternative 340.65-2~bpo70+1 amd64
ii nvidia-driver 340.65-2~bpo70+1 amd64
ii nvidia-driver-bin 340.65-2~bpo70+1 amd64
ii nvidia-glx 340.65-2~bpo70+1 amd64
ii nvidia-installer-cleanup 20131102+1~bpo70+1 amd64
ii nvidia-kernel-common 20131102+1~bpo70+1 amd64
ii nvidia-kernel-dkms 340.65-2~bpo70+1 amd64
ii nvidia-modprobe 334.16-1~bpo70+1 amd64
ii nvidia-settings 319.72-1~bpo70+1 amd64
ii nvidia-support 20131102+1~bpo70+1 amd64
ii nvidia-vdpau-driver:amd64 340.65-2~bpo70+1 amd64
ii nvidia-xconfig 340.46-1~bpo70+1 amd64
ii xserver-xorg-video-nvidia 340.65-2~bpo70+1 amd64

The boot record in /var/log/messages contains an interesting warning 
from the driver right before the crash:

Feb 11 10:24:36 tg2014 kernel: [   50.289566] NVRM: Your system is not 
currently configured to drive a VGA console
Feb 11 10:24:36 tg2014 kernel: [   50.289568] NVRM: on the primary VGA 
device. The NVIDIA Linux graphics driver
Feb 11 10:24:36 tg2014 kernel: [   50.289568] NVRM: requires the use of 
a text-mode VGA console. Use of other console
Feb 11 10:24:36 tg2014 kernel: [   50.289569] NVRM: drivers including, 
but not limited to, vesafb, may result in
Feb 11 10:24:36 tg2014 kernel: [   50.289570] NVRM: corruption and 
stability problems, and is not supported.
Feb 11 10:24:36 tg2014 kernel: [   50.529177] NVRM: GPU at 0000:01:00.0 
has fallen off the bus.
Feb 11 10:24:36 tg2014 kernel: [   50.529197] NVRM: GPU at 0000:01:00.0 
has fallen off the bus.

 From my own research I understand that the VGA/vesafb warning is 
because this machine is booting in EFI mode where there is no 'raw' VGA 
device available, only an EFI framebuffer device.

There is some older discussion on the Nvidia forums at

https://devtalk.nvidia.com/default/topic/537302/linux/reproducible-nvrm-gpu-at-0000-01-00-0-has-fallen-off-the-bus-both-screens-black-xorg-at-100-/1/

The final post in that thread (from 2013/11/13) asserts that the issue 
may be due to powermizer settings.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-nvidia-devel/attachments/20150211/ef425d90/attachment.html>