[SOLVED] Graphical issues on 12th Gen i5-1240P with Pop_OS!

Framework Laptop 13 Linux
1

Specs:
OS: Pop!_OS 22.04 LTS x86_64
Host: Laptop (12th Gen Intel Core) A4
Kernel: 6.0.12-76060006-generic
Uptime: 13 mins
Packages: 2123 (dpkg), 22 (flatpak)
Shell: zsh 5.8.1
Resolution: 2256x1504
DE: GNOME 42.3.1
WM: Mutter
WM Theme: Pop
Theme: Orchis-Red-Dark-Compact [GTK2/3]
Icons: candy-icons-master [GTK2/3]
Terminal: tilix
CPU: 12th Gen Intel i5-1240P (16) @ 1.700GHz
GPU: Intel Alder Lake-P
Memory: 2614MiB / 15708MiB

Problem:
Hi all. Starting to get worried about the iGPU as I am seeing more and more graphical issues on screen. Recently the very top of the screen has started to get lines of pixel glitches on occasion. This was first present when I connected a 2K monitor to the HDMI expansion port, which then caused the top of the display to glitch regularly. Below is just one example of some of the glitches I have seen in my time using this device (I haven’t had it a month yet.)

image
image545Ă—346 97.8 KB

Before that, the first issues I were seeing were thumbnails, and pictures on websites showing a corrupted/glitched.

Is anyone else experiencing these issues with their 12th gen?

2

I have not tested Pop in a couple weeks as it’s not a primary distro I have had time to deep dive into. That said, try rebooting into “oldkern” from the systemmd boot menu and see if that behaves the same way. If it behaves poorly there as well, reboot into the systemmd boot menu again and this time choose recovery.

This is a partition used for testing and acts like a LiveUSB. If it is happening there as well, please update us here.

3

Thanks @Matt_Hartley, I will do some testing over the course of tomorrow. I’m not sure what the older kernel will be on the system but I was sure I was on a Pop specific kernel before it switched me to a generic one.

4

Sounds good. My Pop kernels are 6.0.12-76060006-generic (actively on now) and I believe my oldkern is 6.0.6-76060006.

5

So tested with 6.0.6 and still got the glitching along the top of the laptop display. I had other windows open full screen before opening Edge, and the first time I saw it this session was shortly opening Edge. I think this may be a browser related issue. Even though it’s not where the Edge window is.

I then booted to my live USB with Pop, that is also using 6.0.6 kern, and noticed the same glitching at the top of the screen, but also the Pop Install wizard window was having some weird ghosting issues going on too. So, Edge is obviously out of the question on this instance.

Video of the glitching at the top of the screen whilst secondary display is attached: https://photos.app.goo.gl/aP3RoGBBFFqnKWvt5

Glitches like the one in the OP don’t happen too often, but enough to notice. But he’s another example where the thumbnails aren’t being displayed correctly and so have this weird dark blue/red appearance. And then the cover art for the MAISONdes track Aiwae just looks glitched completely.

Screenshot from 2023-01-17 02-25-05
Screenshot from 2023-01-17 02-25-05373Ă—635 76.8 KB

6

I’ve seen this on XPS 13 with Pop OS in the past. In my case, it was middle of the screen and it was fairly rare.

First and foremost:

sudo apt update && sudo apt upgrade -y && flatpak update

We need to make sure you’re fully up to date on Pop 22.04

Can you guess about when this started happening? This is usually a graphics driver glitch (the white lines).

On Pop OS, I’d boot into the recovery partition and view everything there. If you notice this is not happening, it’s likely something that was either goofed in a config or an Intel driver oddity.

Try that and see if the behavior happens there as well.

Because Pop does its own thing, our options are limited here as we generally work with Fedora and Ubuntu on stuff like this. But do try recovery and see if that behavior appears there as well.

From there, I’d test an Ubuntu or Fedora USB key. If it does not happen with those, then this falls into a Pop OS issue.

7

My system is fully up to date on these.

I will check this. I think this is more prevalent when trying to drive a 2k monitor. I don’t have any 2k monitors at work, but a 1080p monitor didn’t produce any artefacts/glitching across the top of the screen with the time I was testing it for.

Also wanted to add the following to confirm GPU driver in use:

hwinfo --display
26: PCI 02.0: 0300 VGA compatible controller (VGA)              
  [Created at pci.386]
  Unique ID: _Znp.1drWockLPdB
  SysFS ID: /devices/pci0000:00/0000:00:02.0
  SysFS BusID: 0000:00:02.0
  Hardware Class: graphics card
  Model: "Intel VGA compatible controller"
  Vendor: pci 0x8086 "Intel Corporation"
  Device: pci 0x4626 
  SubVendor: pci 0xf111 
  SubDevice: pci 0x0002 
  Revision: 0x0c
  Driver: "i915"
  Driver Modules: "i915"
  Memory Range: 0x605c000000-0x605cffffff (rw,non-prefetchable)
  Memory Range: 0x4000000000-0x400fffffff (ro,non-prefetchable)
  I/O Ports: 0x3000-0x303f (rw)
  Memory Range: 0x000c0000-0x000dffff (rw,non-prefetchable,disabled)
  IRQ: 181 (2988444 events)
  Module Alias: "pci:v00008086d00004626sv0000F111sd00000002bc03sc00i00"
  Driver Info #0:
    Driver Status: i915 is active
    Driver Activation Cmd: "modprobe i915"
  Config Status: cfg=new, avail=yes, need=no, active=unknown

Primary display adapter: #26

lspci -v | grep -A8 VGA
00:02.0 VGA compatible controller: Intel Corporation Alder Lake-P Integrated Graphics Controller (rev 0c) (prog-if 00 [VGA controller])
	Subsystem: Device f111:0002
	Flags: bus master, fast devsel, latency 0, IRQ 181, IOMMU group 1
	Memory at 605c000000 (64-bit, non-prefetchable) [size=16M]
	Memory at 4000000000 (64-bit, prefetchable) [size=256M]
	I/O ports at 3000 [size=64]
	Expansion ROM at 000c0000 [virtual] [disabled] [size=128K]
	Capabilities: <access denied>
	Kernel driver in use: i915

I have a live Ubuntu USB ready but again need the 2K monitor to do the testing with.

Edit: is it also worth installing the OEM kernel instead, like the ubuntu install guide suggests?

8

One thing I have noticed, since enabling, and switching to Wayland, is that the glitching at the top of the screen between the LUKS authentication screen, and the loading of the log in screen does not occur. Need to further test Wayland as I don’t think I have connected an external monitor whilst using it. But if I can’t reproduce the glitching under Wayland then I guess that helps narrow it somewhat.

Edit:

image
image468Ă—593 131 KB

This glitch happens more when on Wayland it seems

Edit2: I’ve installed Ubuntu now, but still seeing graphics issues, e.g. this green square is supposed to be album art…
image

9

I have found an article from a reddit post reporting similar issues to mine in respect to the screen flicker at the top of the screen

https://www.reddit.com/r/framework/comments/zsqj4x/screen_flickeringglitching_possible_screen_damage/

So I will trial running my laptop with the following kernelstub “i915.enable_psr=0” set and see how I go. I suspect this impact the already dismal battery life I’m getting out of this setup though.

10

So far over the course of yesterday and this morning, with the kernel parameter applied, I have not seen the return of the flickering at the top of the framework laptop display.

11

Thanks for the update! Please let us know if it returns.

12

Granted, this image was taken on a Fedora install on 12th gen, but using one HDMI cable to HDMI expansion card, one DP cable to DP expansion card, drives these 2k displays with zero issues. Note the resolution and refresh rate to both displays.

3-displays
3-displays1920Ă—1080 101 KB

Yes, the guide is always recommended as it’s constantly being updated for Ubuntu - it won’t help on Pop OS.

Anytime you have an issues that “feels” driver/kernel related, recovery partition test on Pop, Live USB for Ubuntu. It provides the following details almost immediately:

  • Works great on recovery partition? Likely an update or configuration change that goofed something.
  • Works the same on the recovery partition? Likely something it’s not liking about that specific display. Using a dock or input conversion (DVI, etc to DP or HDMI, there’s your issue - docks are compatible in Linux until they’re not. :slight_smile: I never, ever used (or trust) docks to handle display or power. I trust them to push USB devices.

Pop OS, while based on the LTS release, has a number of things that differ.

  • Uses systemd boot, so kernel parameters are set using kernelstub.
  • Uses system76-power (DO NOT USE TLP HERE, you will have a bad time).
  • Tends to lean toward newer kernels out of the box.
1 Like
13

@Matt_Hartley yes, I thought as much re OEM kernel so haven’t installed it under Pop_OS. However, as per the links I included above, kernel 5.19 has been causing issues for devices with intel based displays and match the description of what I was experiencing. The kernel stub has resolved the issue though.

I also note that your external monitors work at 144Hz, but my Samsung display only allows 30Hz unless I reduce the resolution? Could be the display is utter trash though, I did get it for free. But my desktop PC with GTX970 drives the display at 59Hz?

Edit: I should note that only the glitching at the top of the screen is resolved at the moment. Image artifacting still occurs occasionally but my understanding is there are intel GPU improvements and fixes coming in kernel 6.1

14

Oh, just to make sure I’m clear on OEM kernel - that’s for Ubuntu. You should be on Pop OS 6.0.12-76060006-generic kernel, that’s what we want for Pop OS. Although in your edit you mentioned about waiting until 6.1. So either way, no OEM kernel for Pop, just ubuntu.

I’d need to see the output of xrandr. Please run that command and copy/paste the output.
Let’s also get some display information as well. With your monitors attached, please run the following:

Copy the contents of displays.txt in your home directory:
sudo apt update && sudo apt install ddcutil && (sudo ddcutil detect && xrandr) > displays.txt

or

Just copy the entire output of both commands:
sudo apt update && sudo apt install ddcutil && (sudo ddcutil detect && xrandr)

And lastly, this is a direct connection, no adapters or hubs to the monitor?

15

Just now caught this. If it’s only driving this on the desktop with dedicated graphics, then yes, that is very likely a display limitation.