Ubuntu 20.04 Freezes after a few minutes

Graphics / Linux Linux
, , ,
1

CPU - AMD EPYC 7272
GPU - RTX 3080 Ti
OS - Ubuntu 20.04

I started with a clean Ubuntu 20.04 build and the system stayed awake for ~12hrs. Then I proceeded to download CUDA 11.6.1, using the deb (local) method, following the instructions at Installation Guide Linux :: CUDA Toolkit Documentation. Upon installing CUDA 11.6.1, I now see the system freezing up after about 10 minutes. If the system is power cycled, it comes back.

Is there a way to determine what is causing the system to freeze?

I attached the nvidia bug report.

Thanks,
Kevin

2

nvidia-bug-report.log.zip (290.4 KB)

3

The Xserver is running on the on-board apeed graphics, the nvidia gpu only used for cuda, is this the intended state?
Please create /etc/X11/xorg.conf

Section "Device"
  Identifier "Aspeed"
  Driver "modesetting"
  BusID "PCI:67:0:0"
EndSection

Furthermore, please correctly set up nvidia-presistenced to start on boot and run

grep nvidia /etc/modprobe.d/* /lib/modprobe.d/*

to find a file containing

options nvidia-drm modeset=1

and change 1 to 0
then run

sudo update-initramfs -u

and reboot.