After upgrade to kernel 5.8.0-49/50 with Intel Gen7 (Haswell/Ivy Bridge/Bay Trail) graphics a lot of glitches render screen unusable

[hva]

After upgrading to 5.8.0-49, on a laprtop with intel graphics, graphics was suddenly extremely glitchy.
Icons from the desktop flash through the browser, windows top bar gets distorted, all sort of glitches appear, parts of characters are missing in terminal, scrolling is inconsistent and the whole desktop becomes unusable.
Under wayland the graphics are a bit more stable, but lots of glitches appear anyway, video reproduction in browser is stuttering and wobbly.
Booting with previous kernel 5.8.0-48 seems to fix the issues.
5.8.0-50 is broken as well.

this is the result of 'sudo lshw -c video' on my hp Elitebook 8470:

*-display
       description: VGA compatible controller
       product: 3rd Gen Core processor Graphics Controller
       vendor: Intel Corporation
       physical id: 2
       bus info: pci@0000:00:02.0
       version: 09
       width: 64 bits
       clock: 33MHz
       capabilities: msi pm vga_controller bus_master cap_list rom
       configuration: driver=i915 latency=0
       resources: irq:33 memory:d4000000-d43fffff memory:c0000000-cfffffff ioport:4000(size=64) memory:c0000-dffff

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[MichaelB (mrbou)]

Same here, computer is unusable. Have to use the current kernel 5.4.0-72-generic or 5.8.0-48-generic as workaround.

5.8.0-49 and 5.8.0-50 are broken... (Ubuntu 20.04.2)

*-display
       description: VGA compatible controller
       produit: Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller
       fabricant: Intel Corporation
       identifiant matériel: 2
       information bus: pci@0000:00:02.0
       version: 06
       bits: 64 bits
       horloge: 33MHz
       fonctionnalités: msi pm vga_controller bus_master cap_list rom
       configuration : driver=i915 latency=0
       ressources : irq:32 mémoire:f7800000-f7bfffff mémoire:e0000000-efffffff portE/S:f000(taille=64) mémoire:c0000-dffff

[Hazhir Rahmani (h4906r)]

Same here,
Graphics:
  Device-1: Intel 3rd Gen Core processor Graphics driver: i915 v: kernel
  Display: x11 server: X.Org 1.20.9 driver: i915 resolution: 1366x768~60Hz
  OpenGL: renderer: Mesa DRI Intel HD Graphics 4000 (IVB GT2)
  v: 4.2 Mesa 20.2.6

5.8.0-50-generic

[mahsa (ms1133)]

  *-display
       description: VGA compatible controller
       product: 3rd Gen Core processor Graphics Controller
       vendor: Intel Corporation
       physical id: 2
       bus info: pci@0000:00:02.0
       version: 09
       width: 64 bits
       clock: 33MHz
       capabilities: msi pm vga_controller bus_master cap_list rom
       configuration: driver=i915 latency=0
       resources: irq:31 memory:f0000000-f03fffff memory:e0000000-efffffff ioport:3000(size=64) memory:c0000-dffff

[Terry Rudd (terrykrudd)]

Can you please try the 5.8.0-51 kernel from the -proposed PPA and let us know if this issue resolves with the newer kernel?

[eric-1111 (eric-1111)]

can confirm 5.8.0-51 is working. No glitchy rendering and no "GPU HANG"s in dmesg

[Terry Rudd (terrykrudd)]

Thanks eric-1111. I believe this came in via upstream stable commit for the -49 kernel and has since been reverted. Will watch to see if the same is true for others who've hit this issue.

[hva (francesco-hermanitosverdes)]

sorry, was going to test 5.8.0-51 but can't find the PPA

[MichaelB (mrbou)]

Nothing in focal-proposed cannot find 5.8.0-51 and cannot test.

[Tim Passingham (tim-8aw3u04umo)]

linux 5.8.0-51 not available in groovy-proposed - developer options - (19/4/21 13:11 gmt)

[eric-1111 (eric-1111)]

5.8.0-51 is not in groovy-proposed - developer options
(or focal-proposed I'd assume)

It's available here:
https://launchpad.net/~canonical-kernel-team

[eric-1111 (eric-1111)]

Should be this one:
https://launchpad.net/~canonical-kernel-team/+archive/ubuntu/ppa

[Tim Passingham (tim-8aw3u04umo)]

5.8.0-51 became available last night for me in groovy-proposed. I have installed and will test for a couple of days. OK so far. I'll report again in a day or two.

[Andrea (mariofutire86)]

Same here.
If anybody knows how to stress the system please say.

Yesterday -50 run 12 H no problems.
Today -50 failed after 1 H, and now installed -51.

[samvimes (samvimes)]

I'm using 5.8.0-51 from groovy-proposed now and the glitches have disappeared.

[Pascal Kuppler (kupplepa)]

I used 5.8.0-50 and had the same bug, now I upgraded to 5.8.1-050801 and the bug disappeared

[DiagonalArg (diagonalarg)]

Perhaps this is related? That's my report, and I got rid of it by using the intel_iommu=on kernel parameter, but that introduced another error.

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1925126

[Tim Passingham (tim-8aw3u04umo)]

Using 5.8.0-51 has fixed the problem for me.

Thanks very much for fixing this so promptly.

[Vivien GUEANT (vivienfr)]

Using 5.8.0-51 has fixed the problem for me.

My PC :
- Dell Inc. Inspiron 3847/088DT1, BIOS A11 05/07/2019
- EFI v2.31 by American Megatrends - Secure boot disable
- CPU : Intel(R) Core(TM) i3-4150 CPU @ 3.50GHz
- GPU : Intel HD Graphics 4400
- A single screen connected via HDMI. Resolution: 1920x1080
- RAM : 16 GB DDR3

My Ubuntu 20.10 :
- X11
- GNOME 3.38.2
- French language

[Sergio (hackuss)]

Downgrade to kernel 48 and works.

Please update to 51 release asap.

[heynnema (heynnema)]

The -49 and -50 kernels are at fault. Use -48 for now (or -51 or higher, if you can get it). I'm also checking out a possible fix with the Displays settings for Refresh rate.

[FairMiles (fermilesi)]

Same problem in two ThinkPad E430 (Intel HD Graphics 4000; driver i915) when upgrading to -49 or -50

Additionally, with -50 I can't login (Ubuntu 20.04): cursor appears fine but login screen (e.g., logo, username, password box) does not appear. However I can still open non-graphical consoles (e.g., tty3)

Everything works fine booting into -48-generic

[Torsten Bronger (bronger)]

Is there a plan, roadmap, or experience values for when to expect -51 for focal without using the above mentioned – not recommended – PPA? It’s not that I want to complain; it would be simply helpful for my own planning.

[Daniel van Vugt (vanvugt)]

Thanks for your patience. It appears the kernel team is now aiming for version 5.8.0-52 which is in proposed.

[Diego Garcia (d1360-64rc14)]

Response to FairMiles (fermilesi) #22

I also had this error. I fixed it forcing login screen to load with Xorg (instead of wayland)
These are the steps:
https://linuxconfig.org/how-to-disable-wayland-and-enable-xorg-display-server-on-ubuntu-18-04-bionic-beaver-linux

[MichaelB (mrbou)]

5.8.0-52 in proposed works for me. Thanks!

But why wait more than 2 weeks to resolve it?
Thousands of computers are directly impacted and unusable and not everyone know how they can change his kernel version on boot. Only around me, 4 people are impacted and also myself on 2 computers.

[Kleber Sacilotto de Souza (kleber-souza)]

Hello hva and everyone else affected,

Thank you for reporting the problem and testing the kernel in -proposed. The current kernel SRU schedule is to release the fixed kernel to -updates for general availability next week.

[John Magdy Lotfy (zorono)]

It Affects me too...

GNOME version: 3.38.3
Windowing System: X11
GPU: Intel® HD Graphics 4000 (IVB GT2)
CPU: Intel® Core™ i5-3230M CPU @ 2.60GHz × 4
Kernel: 5.8.0-50-generic

Video: https://streamable.com/7fpimk

$ **lspci -nnk | egrep -i '3d|aphics|display|nouveau|nvidia|radeon|trident|vesa|vga'**:
```
00:02.0 VGA compatible controller [0300]: Intel Corporation 3rd Gen Core processor Graphics Controller [8086:0166] (rev 09)
 Subsystem: Hewlett-Packard Company 3rd Gen Core processor Graphics Controller [103c:17f4]

```

$ **uname -a**:
```
Linux Stack-Linux 5.8.0-50-generic #56-Ubuntu SMP Mon Apr 12 17:18:36 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux
```

$ **sudo Xorg -version**:
```
X.Org X Server 1.20.9
X Protocol Version 11, Revision 0
Build Operating System: Linux 4.15.0-140-generic x86_64 Ubuntu
Current Operating System: Linux Stack-Linux 5.8.0-50-generic #56-Ubuntu SMP Mon Apr 12 17:18:36 UTC 2021 x86_64
Kernel command line: BOOT_IMAGE=/vmlinuz-5.8.0-50-generic root=UUID=b603ef31-0b9c-44dc-b76c-00563dacbf42 ro quiet splash vt.handoff=7
Build Date: 08 April 2021 12:24:24PM
xorg-server 2:1.20.9-2ubuntu1.3 (For technical support please see http://www.ubuntu.com/support)
Current version of pixman: 0.38.4
 Before reporting problems, check http://wiki.x.org
 to make sure that you have the latest version.
```

$ **inxi -F**:
```
System:
  Host: BrownTurbo-Linux Kernel: 5.8.0-50-generic x86_64 bits: 64
  Desktop: N/A Distro: Ubuntu 20.10 (Groovy Gorilla)
Machine:
  Type: Laptop System: Hewlett-Packard product: HP ProBook 4540s v: A1008C11
  serial: <superuser/root required>
  Mobo: Hewlett-Packard model: 17F6 v: KBC Version 58.21
  serial: <superuser/root required> UEFI: Hewlett-Packard v: 68IRR Ver. F.68
  date: 04/11/2019
CPU:
  Info: Dual Core model: Intel Core i5-3230M bits: 64 type: MT MCP
  L2 cache: 3072 KiB
  Speed: 1262 MHz min/max: 1200/3200 MHz Core speeds (MHz): 1: 1586 2: 2643
  3: 1799 4: 2146
Graphics:
  Device-1: Intel 3rd Gen Core processor Graphics driver: i915 v: kernel
  Device-2: Sunplus Innovation type: USB driver: uvcvideo
  Display: x11 server: X.Org 1.20.9 driver: modesetting unloaded: fbdev,vesa
  resolution: 1366x768~60Hz
  OpenGL: renderer: Mesa DRI Intel HD Graphics 4000 (IVB GT2)
  v: 4.2 Mesa 20.2.6
Audio:
  Device-1: Intel 7 Series/C216 Family High Definition Audio
  driver: snd_hda_intel
  Sound Server: ALSA v: k5.8.0-50-generic
Network:
  Device-1: Ralink RT3290 Wireless 802.11n 1T/1R PCIe driver: rt2800pci
  IF: wlo1 state: down mac: 48:5a:b6:68:e4:71
  Device-2: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet
  driver: r8169
  IF: enp5s0 state: up speed: 100 Mbps duplex: full mac: 9c:b6:54:1f:62:ed
  IF-ID-1: docker0 state: down mac: 02:42:b7:9b:e8:b3
Drives:
  Local Storage: total: 465.76 GiB used: 335.35 GiB (72.0%)
  ID-1: /dev/sda vendor: Seagate model: ST500LT012-9WS142 size: 465.76 GiB
Partition:
  ID-1: / size: 91.17 GiB used: 14.04 GiB (15.4%) fs: ext4 dev: /dev/sda1
  ID-2: /boot size: 2.69 GiB used: 155.5 MiB (5.7%) fs: ext4 dev: /dev/sda4
  ID-3: /home size: 81.74 GiB used: 62.49 GiB (76.4%) fs: ext4
  dev: /dev/sda7
  ID-4: /var size: 9.10 GiB used: 7.89 GiB (86.7%) fs: ext4 dev: /dev/sda2
S...

[Daniel van Vugt (vanvugt)]

The fixes appear to be released now so you just need to update your system:

  sudo apt update
  sudo apt full-upgrade

and then reboot.

[Mahmoud Elbarbari (m.elbarbary)]

My problem still persists after upgrading

https://answers.launchpad.net/ubuntu/+question/696971

[Daniel van Vugt (vanvugt)]

If you still experience a problem in kernel 5.8.0-52 or later then please open a new bug by running:

  ubuntu-bug linux

[Daniel van Vugt (vanvugt)]

Also the link in comment #30 is NOT related to this bug. It's the wrong hardware generation. You need to open a new bug.

[Mahmoud Elbarbari (m.elbarbary)]

I have opened a new bug: https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.8/+bug/1928656

[Rudi van Vuuren (tell.tales)]

System updated, been using it for a few hours now and all seems well.
I am so thankful for whoever fixed this.

[Paul Dev (pauldev77)]

I was using linux-image-5.8.0-55-generic:amd64 - this kernel still has the issues.
Reverting to 5.8.0-48 removes the issues.
Changing to the 5.4 kernel (latest) also removes the issues

[Daniel van Vugt (vanvugt)]

If you still experience a problem in kernel 5.8.0-52 or later then please open a new bug by running:

  ubuntu-bug linux

[Daniel van Vugt (vanvugt)]

> seems like this bug has not been patched for the vast majority of us,
> given that this is a critical bug i say it should be reopened

Judging by the above comments, it's 2 people out of 51. Comments #30 and #33 are definitely not this bug because that's the wrong hardware generation. So it would seem the "vast majority" of affected users are no longer reporting issues.

You can use bug 1928656 to continue the conversation or might want to open a new bug of your own.

[Tomas Zemek (tzemek)]

Today regression with kernel linux-image-5.11.0-25-generic 5.11.0-25.27~20.04.1:

Aug 8 22:18:11 kernel: [37254.612789] i915 0000:00:02.0: [drm] GPU HANG: ecode 7:0:00000000
Aug 8 22:18:11 kernel: [37254.612858] i915 0000:00:02.0: [drm] Resetting chip for stopped heartbeat on rcs0
Aug 8 22:18:11 kernel: [37254.714421] i915 0000:00:02.0: [drm] VMSVGA FIFO[8731] context reset due to GPU hang
Aug 8 22:18:14 kernel: [37257.670342] i915 0000:00:02.0: [drm] GPU HANG: ecode 7:0:00000000
Aug 8 22:18:14 kernel: [37257.670382] i915 0000:00:02.0: [drm] Resetting chip for stopped heartbeat on rcs0
Aug 8 22:18:14 kernel: [37257.773493] i915 0000:00:02.0: [drm] kwin_x11[7538] context reset due to GPU hang

System is Asus P8Z68V-Pro Gen3, i7-3770, iGPU HD4000:

00:02.0 VGA compatible controller: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor Graphics Controller (rev 09) (prog-if 00 [VGA controller])
        DeviceName: Onboard IGD
        Subsystem: ASUSTeK Computer Inc. Xeon E3-1200 v2/3rd Gen Core processor Graphics Controller
        Flags: bus master, fast devsel, latency 0, IRQ 61
        Memory at f7400000 (64-bit, non-prefetchable) [size=4M]
        Memory at e0000000 (64-bit, prefetchable) [size=256M]
        I/O ports at f000 [size=64]
        Expansion ROM at 000c0000 [virtual] [disabled] [size=128K]
        Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
        Capabilities: [d0] Power Management version 2
        Capabilities: [a4] PCI Advanced Features
        Kernel driver in use: i915
        Kernel modules: i915

Was the fix of this bug properly ported to 5.11?

[Daniel van Vugt (vanvugt)]

There are countless different ways a GPU HANG can occur so let's assume it's a new problem. Please open a new bug by running:

  ubuntu-bug linux-image-5.11.0-25-generic

[Tomas Zemek (tzemek)]

New bug against 5.11 opened https://bugs.launchpad.net/ubuntu/+source/linux-signed-hwe-5.11/+bug/1939296