Bug #1813423 “System freezing or flickering with i915 error [drm...” : Bugs : xorg-server package : Ubuntu

Revision history for this message

Llord Llama (llamalord) wrote on 2019-01-26:

#1

CurrentDmesg.txt Edit (65.7 KiB, text/plain; charset="utf-8")
Dependencies.txt Edit (6.9 KiB, text/plain; charset="utf-8")
DpkgLog.txt Edit (1.5 MiB, text/plain; charset="utf-8")
Lspci.txt Edit (10.5 KiB, text/plain; charset="utf-8")
Lsusb.txt Edit (624 bytes, text/plain; charset="utf-8")
ProcCpuinfo.txt Edit (3.9 KiB, text/plain; charset="utf-8")
ProcCpuinfoMinimal.txt Edit (999 bytes, text/plain; charset="utf-8")
ProcEnviron.txt Edit (119 bytes, text/plain; charset="utf-8")
ProcInterrupts.txt Edit (2.8 KiB, text/plain; charset="utf-8")
ProcModules.txt Edit (5.2 KiB, text/plain; charset="utf-8")
UdevDb.txt Edit (152.3 KiB, text/plain; charset="utf-8")
XorgLog.txt Edit (34.4 KiB, text/plain; charset="utf-8")
XorgLogOld.txt Edit (35.0 KiB, text/plain; charset="utf-8")
Xrandr.txt Edit (16.1 KiB, text/plain; charset="utf-8")
xdpyinfo.txt Edit (18.7 KiB, text/plain; charset="utf-8")

Revision history for this message

Llord Llama (llamalord) wrote on 2019-01-26:

#2

I have records of the last 5 kern.og / syslog / Xorg.log files from when it hung. When it does hang, the last thing visible on the screen stays there stationary, the cursor doesn't move and no keyboard inputs eg. alt+crtl+f2 / crtl+sysrq+RSEINUB, do nothing.

Revision history for this message

Llord Llama (llamalord) wrote on 2019-01-27:

#3

kern.log.26-01-2019_09:20:13 Edit (239.7 KiB, text/plain)

EG, the last kern.log:

Sebastien Bacher (seb128) on 2019-01-28

affects:

xorg (Ubuntu) → xserver-xorg-video-modesetting (Ubuntu)

Revision history for this message

Ubuntu Kernel Bot (ubuntu-kernel-bot) wrote on 2019-01-29: Status changed to Confirmed

#4

This change was made by a bot.

Changed in linux (Ubuntu):
status:	New → Confirmed

Revision history for this message

Llord Llama (llamalord) wrote on 2019-02-01: Re: Frequent system freezing with i915 error "*ERROR* Atomic update failure on pipe A"

#5

I understand that every user would like every bug they file to be considered "Important".

In this case though we have around 23 people who have confirmed this bug affects them on the three separate bug reports. It causes hanging and forced hard shutdowns at unpredictable times as often as every 15 minutes or so. This means there is a high probability of user data loss and even hardware damage.

I respectfully submit that it's reasonable to expect the importance of this to be set to High.

Revision history for this message

Launchpad Janitor (janitor) wrote on 2019-02-01:

#6

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

Changed in xserver-xorg-video-modesetting (Ubuntu):
status:	New → Confirmed

Revision history for this message

Timo Aaltonen (tjaalton) wrote on 2019-02-01:

#7

there is no separate modesetting driver anymore, it's folded in the server

that said, the bug is most likely in the kernel

affects:

xserver-xorg-video-modesetting (Ubuntu) → xorg-server (Ubuntu)

Revision history for this message

Llord Llama (llamalord) wrote on 2019-02-04:

#8

Would it help to post more of the kern.logs from when it's hung?

Revision history for this message

Daniel van Vugt (vanvugt) wrote on 2019-03-21:

#9

We could probably fold bug 1806242 and bug 1658632 in with this one.

tags:

added: disco

Revision history for this message

Daniel van Vugt (vanvugt) wrote on 2019-03-22:

#10

Browsing the Xorg modesetting source code it appears that might be to blame too, since it seems to try to use atomic mode setting.

Revision history for this message

Federico_dv (quaidy) wrote on 2019-03-26:

#11

Same problem here, brand new install 18.10 in a new laptop, i cant pass the login, freezes all and the logs keeps raising

Revision history for this message

Daniel van Vugt (vanvugt) wrote on 2019-03-27:

#12

In theory logging into "Ubuntu on Wayland" should avoid the problem.

Revision history for this message

Avi (ikes73) wrote on 2019-03-27:

#13

Having same issue, freeze without any changes on screen, on Ubuntu 18.10, Dell Precision 5530.

It only started recently after I encrypted my home directory and swap but I don't know if that's related.

Revision history for this message

You-Sheng Yang (vicamo) wrote on 2019-03-28:

#14

https://bbs.archlinux.org/viewtopic.php?id=233230 is this related? It seems it was fixed by "i915.enable_rc6=0" kernel parameter. Could you have a try?

Revision history for this message

Edson T. Marques (edsontmarques) wrote on 2019-03-28: Re: [Bug 1813423] Re: Frequent system freezing with i915 error "*ERROR* Atomic update failure on pipe A"

#15

Download full text (5.9 KiB)

>
> Hello!
>
> I will explain what I did, I have little ability, so i made some research
> to get knowledge on how I could perform the configuration of this parameter.
>
> I found a location with the instructions I followed:
>
> # sudo cat /sys/module/i915/parameters/enable_psr
>
> This command returned -1;
> # sudo nano /etc/modprobe.d/i915.conf
>
> In this file I put:
> options i915 enable_psr = 0;
>

> I saved the file;
>
> performed:
> # sudo update-initramfs -u
>
> rebooted the computer;
>
> performed:
> # sudo cat /sys/module/i915/parameters/enable_psr
>
> This command returned 0;
>
> Then I connected the HDMI cable from my LG monitor and started testing.
>

> I noticed a specific behavior. I had already set up the gnome-shell dock
> to be displayed on all displays ... So I used the mouse pointer to fire the
> quick-launch shortcuts of applications I have recorded in the gnome-shell
> dock, but I made those shots from LG monitor.
>

> Every time I ran an application, there was a delay and a slight freezing
> of the mouse pointer while the application was loading up to a fraction of
> a second after the application was fully loaded. I continued to load the
> various applications, always using Dock shortcuts, and the same problem
> happened with several of these applications. This is a remarkable detail
> because I do not remember similar behavior in linux before... In addition,
> we notice a certain slowness and some truncation in the movement of the
> mouse pointer (only when it is on the LG display).
>
> So I realized that in this system reboot, my WiFi network was not
> activated (this is another uncomfortable instability that has been
> happening, that is, these faults in the load of the wifi or the wired
> network when we have to reboot the system). When I realized that the
> network was not working I completely turned off the notebook and turned it
> on again. Note that the LG monitor was connected all the time ... Then the
> network started up again and I was able to do the load tests again ...
> Again firing the applications through the gnome shell dock shortcuts using
> the LG monitor. This time, however, at the very first shot of the
> gnome-calculator, the complete interface and system freeze has already
> happened, not even the caps-lock button LED works ... Neither Ctrl Alt
> BkSpace ... Neither Ctrl Alt Delete ... or Ctrl Alt F "n" ... nothing. I
> had to turn off the notebook on the power button.
>
> Again I turned off the HDMI connection and re-used only the built-in
> monitor of the notebook.
>
> Very well ... To the point above I was ready to respond to the previous
> email with the alleged result of the proposed test. Then, reading the
> propose again, I realized that the proposed test was about a kernel
> parameter called "enable_rc6" and not "enable_psr" as it appears in the
> Arch Linux forum. Therefore, I tried to repeat the previous test using
> "enable_rc6". However, the "sudo cat
> /sys/module/i915/parameters/enable_rc6" command did not acknowledge the
> existence of this parameter. Nor after I change the
> "/etc/modprobe.d/i915.conf", run the "sudo update-initramfs -u" and restart
> the notebook. Even so I w...

>
> Hello!
>
> I will explain what I did, I have little ability, so i made some research
> to get knowledge on how I could perform the configuration of this parameter.
>
> I found a location with the instructions I followed:
>
> # sudo cat /sys/module/i915/parameters/enable_psr
>
> This command returned -1;
> # sudo nano /etc/modprobe.d/i915.conf
>
> In this file I put:
> options i915 enable_psr = 0;
>

> I saved the file;
>
> performed:
> # sudo update-initramfs -u
>
> rebooted the computer;
>
> performed:
> # sudo cat /sys/module/i915/parameters/enable_psr
>
> This command returned 0;
>
> Then I connected the HDMI cable from my LG monitor and started testing.
>

> I noticed a specific behavior. I had already set up the gnome-shell dock
> to be displayed on all displays ... So I used the mouse pointer to fire the
> quick-launch shortcuts of applications I have recorded in the gnome-shell
> dock, but I made those shots from LG monitor.
>

> Every time I ran an application, there was a delay and a slight freezing
> of the mouse pointer while the application was loading up to a fraction of
> a second after the application was fully loaded. I continued to load the
> various applications, always using Dock shortcuts, and the same problem
> happened with several of these applications. This is a remarkable detail
> because I do not remember similar behavior in linux before... In addition,
> we notice a certain slowness and some truncation in the movement of the
> mouse pointer (only when it is on the LG display).
>
> So I realized that in this system reboot, my WiFi network was not
> activated (this is another uncomfortable instability that has been
> happening, that is, these faults in the load of the wifi or the wired
> network when we have to reboot the system). When I realized that the
> network was not working I completely turned off the notebook and turned it
> on again. Note that the LG monitor was connected all the time ... Then the
> network started up again and I was able to do the load tests again ...
> Again firing the applications through the gnome shell dock shortcuts using
> the LG monitor. This time, however, at the very first shot of the
> gnome-calculator, the complete interface and system freeze has already
> happened, not even the caps-lock button LED works ... Neither Ctrl Alt
> BkSpace ... Neither Ctrl Alt Delete ... or Ctrl Alt F "n" ... nothing. I
> had to turn off the notebook on the power button.
>
> Again I turned off the HDMI connection and re-used only the built-in
> monitor of the notebook.
>
> Very well ... To the point above I was ready to respond to the previous
> email with the alleged result of the proposed test. Then, reading the
> propose again, I realized that the proposed test was about a kernel
> parameter called "enable_rc6" and not "enable_psr" as it appears in the
> Arch Linux forum. Therefore, I tried to repeat the previous test using
> "enable_rc6". However, the "sudo cat
> /sys/module/i915/parameters/enable_rc6" command did not acknowledge the
> existence of this parameter. Nor after I change the
> "/etc/modprobe.d/i915.conf", run the "sudo update-initramfs -u" and restart
> the notebook. Even so I went ahead with the test, reconnected the HDMI,
> turned it off and turned the notebook on again and started firing the apps
> like before. It was not long before a new freeze happened, more or less as
> before, during application loading. I could only point out that the freeze
> time after the mouse strokes is not greater than the application load time,
> it is actually little lower. These small freezes are happening and
> repeating themselves every time an application is triggered until a time
> when a complete system freeze occurs... and noticed also the same "lazy"
> behavior in the mouse pointer movement on LG display.
>
> After that, in another search, there was the guideline to change kermel
> load parameters, not like I did, but directly editing "/etc/default/grub"
> by changing the line "GRUB_CMDLINE_LINUX_DEFAULT="quiet splash" and adding
> "i915.enable_psr = 0" inside the "quiet splash" quotes. Well ... First I
> deleted "/etc/modprobe.d/i915.conf", so I did all the tests again using
> "i915.enable_rc6 = 0" in "/etc/default/grub". However, after making this
> change, and before making the tests, I ran a "sudo update-grub" commands
> and then "sudo update-initramfs -u", I rebooted the notebook with the
> second display connected... Again, the system was frozen in the same way as
> before.
>
> Then I restarted the computer, edited "/etc/default/grub" again and
> changed the parameter to "i915.enable_psr = 0", I ran "update-grub" and
> "update-initramfs -u". I turned off the notebook, turned it on again,
> tested the "sudo cat/sys/module/i915/parameters/enable_psr" by looking at
> its value changed to "0" ... and restarted application load tests, as
> before using the gnome-shell dock on the secondary display ... I did these
> load tests for 5 to 10 minutes and realized only the keeping of annoying
> behavior of the slight freezing of the mouse cursor during the loading of
> each application and lazy behavior again (notice that this does not happen
> when I am only using the built-in display) but, at least so far, I have not
> yet obtained the complete freezing of the system...
>
> ... That was before lunch...
>
> All the previous tests were implemented during the last 60 minutes before
> lunch ... So I left my workstation with the notebook turned on and went to
> lunch.
>
> After lunchtime I was back to work and was putting the finishing touches
> on this story ... and then the same freeze again. I was using gEdit to
> finalize the translation of the text from my language into English ... So I
> moved the mouse, the mouse pointer went from the LG display to the built-in
> display and back to the LG where I edited the document, then the complete
> freezing of the system happened again.
>

> So if I have not done something wrong, I believe that the problem of
> system freezing can not be solved in this way.
>

Daniel van Vugt (vanvugt) on 2019-07-22

tags:

removed: cosmic

Daniel van Vugt (vanvugt) on 2019-07-30

summary:	- Frequent system freezing with i915 error "ERROR Atomic update failure - on pipe A" + System freezing or flickering with i915 error "ERROR Atomic update + failure on pipe A"
summary:	- System freezing or flickering with i915 error "ERROR Atomic update - failure on pipe A" + System freezing or flickering with i915 error [drm:intel_pipe_update_end + [i915]] ERROR Atomic update failure on pipe A
tags:	added: bionic

Revision history for this message

aaron (nelaaro) wrote on 2019-09-05:

#16

https://www.dedoimedo.com/computers/intel-microcode-atomic-update.html

TLDR

`
Apparently, hardware-based panel refresh supports automatic entry and exit from self-refresh. It is also an atomic operation, which means the processor will wait for it to complete until it finishes. Instinctively, we can assume there is some kind of an issue here, because if it takes to long for the screen to refresh, it may appear frozen.

Perhaps the driver misses buffers, or something else in the driver blocks the self-refresh functionality. This is not a guaranteed reason for the problem, but it a decent rule of thumb for what we are facing. And now, we can, with some confidence, attempt to apply a workaround that may help mitigate the freeze issues until a complete solution is provided with a driver update as part of normal distro update cycle. Hopefully.
`

Work arounds

Install Intel's proprietary firmware (Intel microcode)

Under /etc/modprobe.d/, you will need to create a configuration file i915.conf, if it does not already exist. Then, inside this file, you will need to add the following line:

options i915 enable_psr=0

Revision history for this message

gst (g-starck) wrote on 2019-12-09:

#17

I am running with :

options i915 enable_psr=0

in my /etc/modprobe.d/i915.conf ( + options i915 enable_guc=2 ).

but still getting the error messages in dmesg:

$ dmesg | tail
[ 4461.136192] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81211 end=81212) time 3659 us, min 1073, max 1079, scanline start 990, end 112
[ 4462.253219] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81278 end=81279) time 3613 us, min 1073, max 1079, scanline start 1008, end 126
[ 4462.350517] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81284 end=81285) time 3645 us, min 1073, max 1079, scanline start 836, end 1082
[ 4462.588911] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=81293 end=81294) time 3695 us, min 1073, max 1079, scanline start 833, end 1082
[ 4462.712018] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe C (start=81608 end=81609) time 3635 us, min 1073, max 1079, scanline start 876, end 4
[ 4464.044487] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe C (start=81688 end=81689) time 3672 us, min 1073, max 1079, scanline start 1055, end 185
[ 4465.185017] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81454 end=81455) time 3546 us, min 1073, max 1079, scanline start 922, end 36
[ 4465.368990] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81465 end=81466) time 3642 us, min 1073, max 1079, scanline start 949, end 69
[ 4467.289516] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=81575 end=81576) time 3673 us, min 1073, max 1079, scanline start 871, end 1119
[ 4467.636122] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81601 end=81602) time 3663 us, min 1073, max 1079, scanline start 979, end 101
✔

I am waiting for the freeze to come sooner or later..

I am running with :

options i915 enable_psr=0

in my /etc/modprobe.d/i915.conf  ( + options i915 enable_guc=2 ).

but still getting the error messages in dmesg:

$ dmesg | tail
[ 4461.136192] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81211 end=81212) time 3659 us, min 1073, max 1079, scanline start 990, end 112
[ 4462.253219] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81278 end=81279) time 3613 us, min 1073, max 1079, scanline start 1008, end 126
[ 4462.350517] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81284 end=81285) time 3645 us, min 1073, max 1079, scanline start 836, end 1082
[ 4462.588911] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=81293 end=81294) time 3695 us, min 1073, max 1079, scanline start 833, end 1082
[ 4462.712018] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe C (start=81608 end=81609) time 3635 us, min 1073, max 1079, scanline start 876, end 4
[ 4464.044487] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe C (start=81688 end=81689) time 3672 us, min 1073, max 1079, scanline start 1055, end 185
[ 4465.185017] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81454 end=81455) time 3546 us, min 1073, max 1079, scanline start 922, end 36
[ 4465.368990] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81465 end=81466) time 3642 us, min 1073, max 1079, scanline start 949, end 69
[ 4467.289516] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A (start=81575 end=81576) time 3673 us, min 1073, max 1079, scanline start 871, end 1119
[ 4467.636122] [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe B (start=81601 end=81602) time 3663 us, min 1073, max 1079, scanline start 979, end 101
✔

I am waiting for the freeze to come sooner or later..

Revision history for this message

Carcalac (carcalacus) wrote on 2020-04-03:

#18

i have the same issue on ubuntu 19.10

cat /etc/modprobe.d/i915.conf

options i915 enable_psr=0
options i915 enable_guc=2

5.3.0-45-generic

followed by one of the eight cores going crazy ~100% usage and i have to reboot because the system becomes slow..

Affects		Status	Importance	Assigned to	Milestone
	linux (Ubuntu)	Confirmed	Undecided	Unassigned
	xorg-server (Ubuntu)	Confirmed	Undecided	Unassigned

Ubuntu
xorg-server package

System freezing or flickering with i915 error [drm:intel_pipe_update_end [i915]] ERROR Atomic update failure on pipe A

Bug Description

Other bug subscribers

Bug attachments

Remote bug watches

Ubuntuxorg-server package

System freezing or flickering with i915 error [drm:intel_pipe_update_end [i915]] *ERROR* Atomic update failure on pipe A

Bug Description

Other bug subscribers

Bug attachments

Remote bug watches

Ubuntu
xorg-server package

System freezing or flickering with i915 error [drm:intel_pipe_update_end [i915]] ERROR Atomic update failure on pipe A