Bug#820571: nvidia-driver: shutdown hangs after installing nvidia-drivers

Sun Apr 10 13:32:12 UTC 2016

On Sat, 2016-04-09 at 21:46 -0500, Ignacio Vargas wrote:
> Package: nvidia-driver
> Version: 352.79-5
> Severity: important
> 
> Dear Maintainer,

Hello Ignacio,

> Hopefully this will be a decent enough report - it's my first time doing this.
> 
> I did a fresh install of Debian on my commputer recently, loaded on an SSD.
> Everything ran without a hitch - boot times and shutdown times were lightning
> fast. However, I was still using the Nouveau/Gallium drivers for my NVIDIA GTX
> 760 GPU. I decided to install the propietary drivers to be able to take full
> advantage of my hardware. The install went fine, and every program that I tried
> ran well, as expected.
> 
> However, one problem popped up right after the install - shutdowns and reboots
> now take upwards of a minute, hanging after the screen has turned off. Before,
> the shutdown was completed in less than five seconds. The issue also seems to
> affect 'Suspend'. So I have reason to believe there is something wrong with the
> shutdown scripts related to nvidia-drivers. I can't give much more info on the
> subject as I am not very experienced. I tried looking at the shutdown logs but
> couldn't make out anything of them.
> 
> If you require more information I'd be happy to provide it to you

> 
> dmesg:
> [    0.153932] vgaarb: device added: PCI:0000:00:02.0,decodes=io+mem,owns=mem,locks=none
> [    0.153935] vgaarb: setting as boot device: PCI:0000:01:00.0
> [    0.153936] vgaarb: device added: PCI:0000:01:00.0,decodes=io+mem,owns=io+mem,locks=none
> [    0.153937] vgaarb: loaded
> [    0.153938] vgaarb: bridge control possible 0000:01:00.0
> [    0.153938] vgaarb: no bridge control possible 0000:00:02.0
> [    0.469750] Linux agpgart interface v0.103
> [    1.989177] nvidia: module license 'NVIDIA' taints kernel.
> [    1.994886] vgaarb: device changed decodes: PCI:0000:01:00.0,olddecodes=io+mem,decodes=none:owns=io+mem
> [    1.995081] [drm] Initialized nvidia-drm 0.0.0 20150116 for 0000:01:00.0 on minor 0
> [    1.995084] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  352.79  Wed Jan 13 16:17:53 PST 2016
> [    2.127303] snd_hda_intel 0000:01:00.1: Handle vga_switcheroo audio client
> [    2.153178] [drm] Replacing VGA console driver
> [    2.633807] input: HDA NVidia HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card2/input25
> [    2.669518] input: HDA NVidia HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card2/input26
> [    2.669608] input: HDA NVidia HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card2/input27
> [    2.669654] input: HDA NVidia HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card2/input29
> [    3.784607] NVRM: Your system is not currently configured to drive a VGA console
> [    3.784610] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
> [    3.784611] NVRM: requires the use of a text-mode VGA console. Use of other console
> [    3.784612] NVRM: drivers including, but not limited to, vesafb, may result in
> [    3.784612] NVRM: corruption and stability problems, and is not supported.
> [   68.938456]  <EOI>  [<ffffffffa089c9a0>] ? rm_shutdown_gvi_device+0x1f0/0x2f0 [nvidia]
> [   68.938490]  [<ffffffffa03e276a>] ? os_io_write_dword+0xa/0x10 [nvidia]
> [   68.938550]  [<ffffffffa087e410>] ? _nv016241rm+0x8570/0xbd60 [nvidia]
> [   68.938608]  [<ffffffffa0875145>] ? _nv000823rm+0x85/0xb0 [nvidia]
> [   68.938667]  [<ffffffffa089ce94>] ? _nv012224rm+0x164/0x540 [nvidia]
> [   68.938726]  [<ffffffffa088a7ec>] ? _nv012593rm+0x7c/0x170 [nvidia]
> [   68.938785]  [<ffffffffa088eaf2>] ? _nv000718rm+0x2d2/0x360 [nvidia]
> [   68.938844]  [<ffffffffa088edfd>] ? _nv000642rm+0x27d/0x530 [nvidia]
> [   68.938903]  [<ffffffffa0891f27>] ? _nv014263rm+0x87/0xc0 [nvidia]
> [   68.938961]  [<ffffffffa0887ae6>] ? _nv000696rm+0x4b6/0x830 [nvidia]
> [   68.939020]  [<ffffffffa0891e13>] ? rm_ioctl+0x73/0x100 [nvidia]
> [   68.939051]  [<ffffffffa03d8a01>] ? nvidia_ioctl+0x201/0x430 [nvidia]
> [   68.939081]  [<ffffffffa03d736a>] ? nvidia_frontend_ioctl+0x2a/0x60 [nvidia]
> [   68.939111]  [<ffffffffa03d73b9>] ? nvidia_frontend_compat_ioctl+0x19/0x20 [nvidia]
> [  103.854667]  <EOI>  [<ffffffffa089c9a0>] ? rm_shutdown_gvi_device+0x1f0/0x2f0 [nvidia]
> [  103.854697]  [<ffffffffa03e276a>] ? os_io_write_dword+0xa/0x10 [nvidia]
> [  103.854753]  [<ffffffffa087e410>] ? _nv016241rm+0x8570/0xbd60 [nvidia]
> [  103.854808]  [<ffffffffa0875145>] ? _nv000823rm+0x85/0xb0 [nvidia]
> [  103.854863]  [<ffffffffa089ce94>] ? _nv012224rm+0x164/0x540 [nvidia]
> [  103.854919]  [<ffffffffa088a7ec>] ? _nv012593rm+0x7c/0x170 [nvidia]
> [  103.854974]  [<ffffffffa088eaf2>] ? _nv000718rm+0x2d2/0x360 [nvidia]
> [  103.855029]  [<ffffffffa088edfd>] ? _nv000642rm+0x27d/0x530 [nvidia]
> [  103.855084]  [<ffffffffa088dba7>] ? _nv000710rm+0x1c7/0x210 [nvidia]
> [  103.855139]  [<ffffffffa0891bd0>] ? rm_power_management+0x110/0x150 [nvidia]
> [  103.855168]  [<ffffffffa03d7b9e>] ? nv_power_management+0x16e/0x200 [nvidia]

This looks identical to a problem I had a while back, and backtrace is
very, very similar. I'm fairly sure it's nothing to do with our
packaging or Debian in general, but it looks like an upstream problem,
so I reported it [1].

Do TTYs work for you? That was another symptom, and I ended up fixing
the first problem when I found a solution for the TTYs.

Turns out I needed to configure grub to use VBE.

AFter adding this to my /etc/default/grub:

GRUB_GFXMODE=1920x1080x32
GRUB_GFXPAYLOAD_LINUX="keep"
GRUB_VIDEO_BACKEND="vbe"
GRUB_TERMINAL_OUTPUT="gfxterm"
GRUB_FONT_PATH="/boot/grub/fonts/unicode.pf2"

And running "sudo update-grub" and rebooting, both problems went away
from the next boot.

Could you please give this a shot?

Kind regards,
Luca Boccassi

[1] https://devtalk.nvidia.com/default/topic/852598/34x-35x-freeze-at-reboot-shutdown-rcu_sched-self-detected-stall-detected/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part
URL: <http://lists.alioth.debian.org/pipermail/pkg-nvidia-devel/attachments/20160410/47bd97cb/attachment.sig>