Bug 110117

I don't see anything immediately wrong in the log at the first glance at least so it'd be helpful if you could post your dmesg log from after the problem occurred with drm.debug=4 in your kernel boot parameters.

Please post your xorg log if applicable as well.

Steps I took:
1. added drm.debug=4 to /etc/default/grub
2. sudo update-grub
3. reboot
4. reproduced the issue
5. rebooted using ctrl-sysrq-reisub
6. dumped this dmesg into log.txt

Created attachment 143692 [details]
dmesg after reproducing issue and rebooting

Created attachment 143693 [details]
dmesg where i ssh'd into machine during the screen static

Created attachment 143698 [details]
Xorg Log around the recreation of the problem

Created attachment 143699 [details]
Xorg.old log when the error was reproduced

tested the kernel from here today with the same result: https://kernel.ubuntu.com/~kernel-ppa/mainline/v5.1-rc4/

generic amd64

I have updated to Ubuntu 19.04 and still getting the same result.  Please let me know what steps to take next to get this issue resolved.

I would like to add to this some additional troubleshooting I have performed, in case it is of use.



just a short history: I am unable to suspend with any Linux distribution and be able to resume again. Currently running Ubuntu 19.04.

I am attempting to follow the steps here to trace the problem: https://wiki.ubuntu.com/DebuggingKernelSuspend

This is what I have done so far:

$ sudo apt install pm-utils
$ sudo sh -c "sync && echo 1 > /sys/power/pm_trace && pm-suspend"

I resumed but got my usual screen garbage (looks like static on the screen). I ssh'd in and did a sudo reboot

Right away I ran this:

$ dmesg > dmesg_post_boot.txt

I looked for Magic number in the dmesg per the ubuntu wiki above and found these lines:

[    0.950964]   Magic number: 0:66:1001
[    0.950965]   hash matches drivers/base/power/main.c:1012

Now the instructions are to exclude this module from the next boot and see if I can replicate the problem. How do I exclude main.c or do I even want to? Is it even possible that the bug is here?

Created attachment 144270 [details]
Most Recent dmesg with magic number

I can reproduce on the same machine, but I get rainbow static, not grey.

yes i should mention most of my static is grey but there are colours in there as well

I can confirm this behavior at Ubuntu 18.04.2 LTS with Acer PH517-61 notebook.

The minor addition is that the issue concerns only the built-in display. In two display configuration after wake up the external display seems to be normal.

happened to me too using Opensuse Tumbleweed with kernel 5.1.7 and MESA 19.0.5, I have been try to disable power management in amdgpu via kernel paramater but still happen.
this happen to me while I play some game(s) such Mutant Zero Year and Uboat on the STEAM, while other games seems doesnt trigger this static greyed display such as The Witcher 3
I cant log/reproduce any error messages since I'm using the laptop display (Acer Predator Helios 500 Ryzen 2700 with vega56)
but maybe I can show my recorded video taken from my phone while this display glitch happened.

This issue also occurs on Windows. AMD support says that it will be fixed in the next Windows update; could they possibly contribute that same fix to Linux?

-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/726.