You are not logged in.

#1 2021-04-20 21:04:33

Higlav
Member
Registered: 2016-12-09
Posts: 39

GPU falls off the bus, causing an unresponsive, frozen system

Dear community,

Original system configuration:
Kernel: 5.11.15 in the beginning. Now 5.10.31-1
GPU: Nvidia GeForce GTX 550 Ti (using nvidia-390xx-dkms and nvidia-390xx-utils)
nvidia-smi:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 390.141                Driver Version: 390.141                   |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 550 Ti  Off  | 00000000:01:00.0 N/A |                  N/A |
| 30%   47C    P0    N/A /  N/A |    372MiB /   961MiB |     N/A      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0                    Not Supported                                       |
+-----------------------------------------------------------------------------+

packages:

nvidia-390xx-dkms 390.141-3
nvidia-390xx-utils 390.141-3
linux-lts 5.10.31-1

Since two or three weeks I get random freezes. The screen freezes and no inputs get processed. I rarely get a second or third frame after a couple of minutes and I have to hard-reset the machine. When looking at journalctl, I see that the GPU seems to have fallen off the bus:

NVRM: GPU at PCI:0000:01:00: GPU-a610858e-de2a-ac10-bc36-f21c3d89bc37
NVRM: Xid (PCI:0000:01:00): 79, GPU has fallen off the bus.
NVRM: GPU at 0000:01:00.0 has fallen off the bus.
NVRM: A GPU crash dump has been created. If possible, please run
NVRM: nvidia-bug-report.sh as root to collect this data before
NVRM: the NVIDIA kernel module is unloaded.

On a side note: When I boot the machine, shortly before getty finishes and drops me into my login manager, it wastes one and a half minutes  on

a job is running for /dev/dri/card0

I would love to get rid of this one as well because it decreases my speed when resolving this main issue. Unfortunately I didn't find anything

What I looked up:
Here the OP just replaced the GPU to "resolve" the issue. My GPU does not overheat.
Here there is talk about kernel parameters, but those did not work for neither the OP nor me.

What I tried:

  • Other GPU (same model)

  • Other kernel (linux-zen and linux-lts instead of linux)

  • Reinstalling the packages

Here is the log of one session which ended in me first pushing the power button, then hard-resetting

Any help is greatly appreciated as I really need this computer to run reliably.

Offline

#2 2021-04-24 09:15:36

Higlav
Member
Registered: 2016-12-09
Posts: 39

Re: GPU falls off the bus, causing an unresponsive, frozen system

I reinstalled my entire system last night but I still get GPU-caused freezes, although so far just once and the journal shows something different this time:

Apr 24 10:18:01 sun kernel: NVRM: GPU at PCI:0000:01:00: GPU-a610858e-de2a-ac10-bc36-f21c3d89bc37
Apr 24 10:18:01 sun kernel: NVRM: Xid (PCI:0000:01:00): 16, Head 00000000 Count 0000d16b
Apr 24 10:18:05 sun /usr/lib/gdm-x-session[792]: (EE) NVIDIA(0): The NVIDIA X driver has encountered an error; attempting to
Apr 24 10:18:05 sun /usr/lib/gdm-x-session[792]: (EE) NVIDIA(0):     recover...
Apr 24 10:18:05 sun kernel: NVRM: Xid (PCI:0000:01:00): 8, Channel 00000002
Apr 24 10:18:19 sun kernel: NVRM: Xid (PCI:0000:01:00): 8, Channel 00000001
Apr 24 10:18:27 sun kernel: NVRM: Xid (PCI:0000:01:00): 8, Channel 00000001
Apr 24 10:18:36 sun kernel: NVRM: Xid (PCI:0000:01:00): 8, Channel 00000001
Apr 24 10:18:44 sun kernel: NVRM: Xid (PCI:0000:01:00): 8, Channel 00000001
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) NVIDIA(0): Error recovery was successful.
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (EE) event3  - Microsoft Microsoft® 2.4GHz Transceiver v8.0: client bug: event processing lagging behind by 24226ms, your system is too slow
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event4  - Microsoft Microsoft® 2.4GHz Transceiver v8.0 Mouse: SYN_DROPPED event - some input events have been lost.
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "29"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event1  - Power Button: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "32"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event0  - Power Button: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "33"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event3  - Microsoft Microsoft® 2.4GHz Transceiver v8.0: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "34"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event4  - Microsoft Microsoft® 2.4GHz Transceiver v8.0 Mouse: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "35"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "36"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "37"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event7  - Microsoft Microsoft® 2.4GHz Transceiver v8.0 System Control: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "38"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event21 - UVC Camera (046d:0990): device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "39"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event8  - Eee PC WMI hotkeys: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "35"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event5  - Microsoft Microsoft® 2.4GHz Transceiver v8.0 Consumer Control: device removed
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (**) Option "fd" "36"
Apr 24 10:18:45 sun /usr/lib/gdm-x-session[792]: (II) event6  - Microsoft Microsoft® 2.4GHz Transceiver v8.0 Consumer Control: device removed
Apr 24 10:18:48 sun kernel: nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing.
Apr 24 10:18:50 sun kernel: nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000857d:0:0
Apr 24 10:19:11 sun tracker-extract-3[2639]: Corrupt JPEG data: 12 extraneous bytes before marker 0xe1

My current package versions are:

nvidia-390xx-dkms 390.143-3
nvidia-390xx-utils 390.143-3
linux 5.11.16.arch1-1

I would really appreciate any help or guidance. I've never had this problem until around a month ago, when it slowly started to appear and I fear that my motherboard might be faulty and needs to be replaced, which would hurt me immensely as that is one of the few good components my computer has (because, you know, the GPU is definitely not it).

Offline

Board footer

Powered by FluxBB