[i915] Ubuntu display get shifted after some time.

[Ivan Krasicenko]

It is continuation of this bug:
https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/1986583
where advised fix was applied.

Summay: After ubuntu is turned on, after some time screen gets shifted to side. (and starts to overlap)

I setup webcamera to record screen, and video below shows what it looked like.
after fix were applied.
www.ms.mff.cuni.cz/~krasicei/capture-0231.bug.crop.slow.mp4
(video was cropped, and slowed down 2x)

before fix was applied shift looked somehow like this:
www.ms.mff.cuni.cz/~krasicei/bug17-50-18-8-2022.c.mp4
(video is also cropped, and slowed down 2x)

During shift no one was present on computer.

Is there any idea how to enable some more detailed log or something? I probably can pinpoint exact moment in time when shift happened.
i tried to read log journalctl, but I found nothing interesting.

ProblemType: Bug
DistroRelease: Ubuntu 22.04
Package: xorg 1:7.7+23ubuntu2
ProcVersionSignature: Ubuntu 5.15.0-46.49-generic 5.15.39
Uname: Linux 5.15.0-46-generic x86_64
ApportVersion: 2.20.11-0ubuntu82.1
Architecture: amd64
BootLog:

CasperMD5CheckResult: unknown
Date: Fri Aug 26 02:31:10 2022
DistUpgraded: 2022-08-02 14:36:41,704 DEBUG Running PostInstallScript: '/usr/lib/ubuntu-advantage/upgrade_lts_contract.py'
DistroCodename: jammy
DistroVariant: ubuntu
ExtraDebuggingInterest: Yes, including running git bisection searches
GraphicsCard:
 Intel Corporation Skylake GT2 [HD Graphics 520] [8086:1916] (rev 07) (prog-if 00 [VGA controller])
   Subsystem: Intel Corporation Skylake GT2 [HD Graphics 520] [8086:2015]
InstallationDate: Installed on 2022-02-17 (189 days ago)
InstallationMedia: Ubuntu 20.04.3 LTS "Focal Fossa" - Release amd64 (20210819)
Lsusb:
 Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
 Bus 001 Device 003: ID 222a:0001 ILI Technology Corp. Multi-Touch Screen
 Bus 001 Device 002: ID 1a2c:2d23 China Resource Semico Co., Ltd Keyboard
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
MachineType: Default string Default string
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-5.15.0-46-generic root=UUID=582fb3cc-8ef6-460b-b9ae-a16f81c604a7 ro quiet splash vt.handoff=7
SourcePackage: xorg
UpgradeStatus: Upgraded to jammy on 2022-08-02 (23 days ago)
dmi.bios.date: 10/07/2021
dmi.bios.release: 5.12
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: GSKU0504.V54
dmi.board.asset.tag: Default string
dmi.board.name: SKYBAY
dmi.board.vendor: Default string
dmi.board.version: Default string
dmi.chassis.asset.tag: Default string
dmi.chassis.type: 3
dmi.chassis.vendor: Default string
dmi.chassis.version: Default string
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrGSKU0504.V54:bd10/07/2021:br5.12:svnDefaultstring:pnDefaultstring:pvrDefaultstring:rvnDefaultstring:rnSKYBAY:rvrDefaultstring:cvnDefaultstring:ct3:cvrDefaultstring:skuDefaultstring:
dmi.product.family: Default string
dmi.product.name: Default string
dmi.product.sku: Default string
dmi.product.version: Default string
dmi.sys.vendor: Default string
version.compiz: compiz N/A
version.libdrm2: libdrm2 2.4.110-1ubuntu1
version.libgl1-mesa-dri: libgl1-mesa-dri 22.0.5-0ubuntu0.1
version.libgl1-mesa-glx: libgl1-mesa-glx N/A
version.xserver-xorg-core: xserver-xorg-core 2:21.1.3-2ubuntu2.1
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev N/A
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:19.1.0-2ubuntu1
version.xserver-xorg-video-intel: xserver-xorg-video-intel N/A
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:1.0.17-2build1

[Daniel van Vugt (vanvugt)]

Thanks for the bug report.

Please run these commands:

  journalctl -b0 > journal.txt
  xrandr --verbose > xrandr.txt

and attach the resulting text files here.

[Ivan Krasicenko (teofil321)]

I restarted pc, and this error occured again today at 1AM. i.e. 26.8.2022 01:18:34 +- 1second.
logs should be in attachment.

[Daniel van Vugt (vanvugt)]

Thanks. It sounds like the problem is now happening in Wayland sessions. Since essentially the same bug was first reported in Xorg that would make this a kernel bug.

Please try adding this line to /etc/environment:

  MUTTER_DEBUG_ENABLE_ATOMIC_KMS=0

and then reboot. It will change the way the screen is rendered in Wayland sessions.

[Ivan Krasicenko (teofil321)]

I had a really bad luck making RDP work under wayland (but I need some remote desktop management), so I want to ask, is there some alike adjustment for Xorg? I would like make it work under Xorg, than wayland. Since bug is presented in both cases, i hope it is possible.

[Daniel van Vugt (vanvugt)]

Yes I suspect RDP under Wayland is still buggy -- it's very new and hardly anyone knows about it or uses it that I can tell.

Since this sounds like a kernel bug you don't really need to stick to Wayland. You can go back to using Xorg. Just make sure your log contains the correct driver: "modeset(0)" and not "intel(0)"

As for other ways to possibly solve the shifting problem, I think the next step is to try the latest upstream graphics kernel:

  https://kernel.ubuntu.com/~kernel-ppa/mainline/drm-tip/2022-08-27/amd64/

But beware that test kernels require Secure Boot is disabled in your BIOS.

[Ivan Krasicenko (teofil321)]

Hi,
I tried new kernel (by link above) with xorg, and pc runs around week and yet no shifting occurred.
I also tried on another pc (same hw configuration) (under wayland) use command: MUTTER_DEBUG_ENABLE_ATOMIC_KMS=0
and shifting occurred.

I think I will give it another week, to see if the shifting occurs on xorg.
Question: drm-tip isn't actual kernel version, so if i want to try older but more tested kernel, I have to pick one of older releases? Or this kernel is just fine to use?

[Daniel van Vugt (vanvugt)]

drm-tip is a test kernel used to tell if a fix already exists upstream, and it sounds like it does.

But the fix you need might already exist in a mainline version, so try:

  https://kernel.ubuntu.com/~kernel-ppa/mainline/v5.19.5/amd64/

[Ivan Krasicenko (teofil321)]

Well, looks like bug occured again.
In drm-tip, and also it occured in version v5.19.5

I realized, that I have also computer with se same(i believe) graphic card, and i do not remember for problem to occur on this pc even once.

so I looked for difference in these two computers.

I tried these commands to find difference:
glxinfo -B
lscpu
lshw

And (from my point of view) interesting difference is:

glxinfo:
Pc where error does not occures:
    Version: 22.0.1
    Video memory: 3072MB

pc where error occures:
    Version: 22.0.5
    Video memory: 15928MB

(note, that pc where error does not occures has 8GB ram, and the one where error occures has total 16GB ram.) ie, probably video memory can consume entire ram?
But I did not remember setting that, so these have to be default settings, or do not know.)

lshw:

pc ok: lshw
          product: Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz
          vendor: Intel Corp.
          physical id: 1
          bus info: cpu@0
          version: 6.78.3
          size: 2800MHz
          width: 64 bits

pc nok: lshw
          product: Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz
          vendor: Intel Corp.
          physical id: 1
          bus info: cpu@0
          version: 6.78.3
          size: 2700MHz
          capacity: 2800MHz
          width: 64 bits

there is different "size" value, and pc that does not shows screen shifting misses "capacity".

lscpu differs in these two rows:
pc ok: lscpu

Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; IBRS, IBPB conditional, RSB filling

pc nok: lscpu

Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS Not affected

Main difference is in seccomp, and PBRSB-eIRBS Not affected... have no idea if it is something interesting.

anyway, full list of response commands is in attachment.

I may try to downgrade from version 22.0.5 to 22.0.1 but i am not sure what exactly to downgrade.

P.s.: I just remembered, that collegue did in past some ram upgrade (from 4 to 8) so there is little chance, that pc that is ok, was originally installed with only 4gb ram. (but cannot confirm that)

[Daniel van Vugt (vanvugt)]

Can you try different computers on the same monitor? Just in case it is the monitor.

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This change was made by a bot.

[Ivan Krasicenko (teofil321)]

I cant try different monitors, because it is not monitor, it is display, that is embedded into computer. (basically computer is "all in one"- motherboard, display, case, sound etc)
But to test possibility of faulty hardware we tried run windows on the same computer, for longer time, and no error occurred.

[Ivan Krasicenko (teofil321)]

Anyway,
can you pinpoint me library that probably contains bug, maybe it will not be hard to fix it.
Or at least what to do to downgrade driver from Version: 22.0.5 to 22.0.1?
Or maybe some article describing how this entire graphic thing works.

[Daniel van Vugt (vanvugt)]

If this isn't a hardware fault in the screen itself then it seems like a bug in the kernel graphics driver for Intel GPUs, which you can find here:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/i915