My Ubuntu 20.04 system has been affected by occasional freezes (~ each 1-2 days) which completely blocks any interaction (not even REISUB works) and requires pressing on the power button to restart the computer. This is a brand new installation on a laptop MSI GF65 Thin 10SER. I have done some research already and this seems to be an old problem (, , ) related to the Intel iGPU.

My dmesg is full of lines such as the ones below and inevitably they cause the system to freeze.

[drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=9338 end=9339) time 321 us, min 1063, max 1079, scanline start 1038, end 1090

$ uname -r

5.4.0-58-generic

$ lspci -k | grep -EA3 'VGA|3D|Display':

00:02.0 VGA compatible controller: Intel Corporation UHD Graphics (rev 05) DeviceName: Onboard - Video Subsystem: Micro-Star International Co., Ltd. [MSI] Device 12aa Kernel driver in use: i915
--
01:00.0 VGA compatible controller: NVIDIA Corporation TU106 [GeForce RTX 2060] (rev a1) Subsystem: Micro-Star International Co., Ltd. [MSI] Device 12aa Kernel driver in use: nvidia Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia

I have tried the following:

• Add the kernel options i915.enable_psr=0, i915.enable_dc=0

• Enable early KMS for the Intel graphics driver (add i915 to /etc/initramfs-tools/modules)

• Enable GuC with i915.enable_guc=2, however it isn't working as $ sudo cat /sys/kernel/debug/dri/0/i915_guc_load_status | grep status: returns status: DISABLED

• Change the X.Org display driver to modeset instead of the xserver-xorg-video-intel

• Update the linux kernel to 5.8

• Add the following to /usr/share/X11/xorg.conf.d/10-intel.conf

   Section "Device" Identifier "Intel Graphics" Driver "intel" Option "NoAccel" "True" Option "DRI" "False" EndSection

One observation is that when I change the Prime Profile to Performance mode (i.e. only the nvidia gpu is used and the intel one isn't) I don't get any of these errors in dmesg. However this isn't a solution for me since this options drains my battery very quickly.

I am a new Ubuntu user and I would really appreciate if someone could give a hand or just provide a workaround so I can have a stable system with no crashes. Let me know if you need any additional information. Thanks in advance!

Update 20/12/2020:

Setting GRUB_CMDLINE_LINUX="nouveau.modeset=0" on /etc/default/grub stopped the dmesg atomic update failure errors. However, now glxinfo|egrep "OpenGL renderer" gives OpenGL renderer string: llvmpipe (LLVM 10.0.0, 256 bits) instead of OpenGL renderer string: Mesa Intel(R) UHD Graphics (CML GT2). This is a problem because now I get extremely high CPU usage from gnome-shell. I don't really know what to take out from this information.

Help would be much appreciated.

In the meanwhile I have restored the grub configuration and done some other changes.$ inxi -G

Graphics: Device-1: Intel UHD Graphics driver: i915 v: kernel Device-2: NVIDIA TU106 [GeForce RTX 2060] driver: nvidia v: 450.80.02 Display: x11 server: X.Org 1.20.8 driver: modesetting,nvidia resolution: 1920x1080~144Hz OpenGL: renderer: Mesa Intel UHD Graphics (CML GT2) v: 4.6 Mesa 20.0.8

Update 21/12/2020:

Ok so I have been running my system with the Nvidia Prime performance profile on (I think this means the iGPU isn't being used, only the dGPU - nvidia - is) and I just got a freeze while listening to Spotify (which happens to also be one of the reasons why my computer has frozen in the past), however this time I didn't register any Atomic update error in this session and journalctl -b -1 -e didn't register anything relevant (at least for me). I will follow the answer to this question How can I tell why 20.04 is crashing? and then I will update the topic. Let me know if there's any output/ information you need to understand better the problem.

Update 23/12/2020:

I performed a memtest from and got 0 errors (results below):

I also did a clean install of Ubuntu, Mint and Manjaro. Every one of them would show the Atomic update error in dmesg. I sent the laptop for the repair service as I think the problem is due to hardware malfunction.

3

1 Answer

So the laptop came out from RMA and they told me there was nothing wrong with it. Unfortunately I had to switch to Windows as I can't afford having an unstable system like I described above. In the future I may try to install a newer version of Ubuntu to check if the problem persists.

Your Answer

Sign up or log in

Sign up using Google Sign up using Facebook Sign up using Email and Password Submit

Post as a guest

Post Your Answer Discard

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy