Plasma 6.2: Black screen on wakeup

Hi all,

I’ve been getting black screens of death on my laptop waking up recently. This has only started happening in the past few days, and was not the case with 6.1 (did happen rarely on 6.0, before it was patched). This is on the wayland session.

Everything works fine, but (inconsistently) on wakeup the screen will be black, with no cursor visible either. Switching to a TTY is fine, and I can login as normal. Running startplasma-wayland will register my login plasma sound effect, but then the TTY will just hang until a CTRL-C is sent. However, I can actually still log into hyprland by running Hyprland, so something is still functioning behind everything.

Running killall startplasma-wayland;startplasma-wayland does indeed successfully work and bring me back to SDDM, but I’d rather not lose all my progress/current work in open apps.

Tried looking for a new issue on bugzilla, could only see pretty old bugs open. I don’t think this is my issue though, as it’s happened twice in the last 2 days after updates, and hadn’t for months and months before that.

If no-one’s reported or encountered this yet, what sort of logs could I take that would be most useful for a developer? Not familiar enough with the depths of logging to know where or what’s useful.

Edit: Not sure if this is relevant, but TTY1 had a blinking text cursor, TTY2, 3, 5, 6 were fine, and TTY4 was non-functional. After killall startplasma-wayland, TTY 4 had my plasma session open though.

SPECS:

Platform:

OS: Arch Linux, latest
Plasma Version: 6.2.1
Frameworks Version: 6.7.0
Qt Version: 6.8.0
Kernel Version: 6.11.3-arch1-1 (64-bit)
Graphics Platform: Wayland

Hardware

Laptop Model: Lenovo Thinkpad T480s
Graphics Processor: Mesa Intel UHD Graphics 620
CPU: intil i5-8250
RAM: 16Gb

since plasma 6.2 and continuing in 6.2.1
BLACK SCREEN

Still have to CTL-ALT # to terminal, CTL-ALT# to open plasma desktop

same issue on a CachyOS partition
Lenovo Legion T5 26ARA8
AMD Ryzen 7 7700
Radeon RX 6600
samsung s32ag55 monitor

(successful upgrade on totally different system)

possibly related to ddcutil, powerdevil and plasma-desktop ??

Working system with Manjaro Stable. Need I back away from the bleeding edge???

My workaround is to press Ctrl+Alt+F3 and then Ctrl+Alt+F2.
Depending on the distro it can be other function keys.

It happens to me after resume of hibernate or suspend, but not always.

Operating System: CachyOS Linux
KDE Plasma Version: 6.2.1
KDE Frameworks Version: 6.7.0
Qt Version: 6.8.0
Kernel Version: 6.11.3-3-cachyos (64-bit)
Graphics Platform: X11
Processors: 16 × AMD Ryzen 7 5800H with Radeon Graphics
Memory: 28.3 GiB of RAM
Graphics Processor: AMD Radeon Graphics
Manufacturer: Trigkey
Product Name: S5

I suspect that ddcutil, powerdevil and plasma-desktop are involved.

Your monitor brand and model should also be considered part of the issue

Also experiencing this after the update.

Fedora 40, Plasma 6.2.0, KDE Frameworks 6.7.0, Qt 6.7.2, Wayland.

I locked the session, the lock screen displayed successfully, I left the workplace. Returned to the computer not waking up to basic key presses.

Ctrl-Alt-Fx works, but Plasma session, which is on F2 for me, can’t be seen.

I was able to get into the running session by carefully entering my password. After that it’s even weirder: the docked external screens display windows (not with the capital w, thankfully, sad haha) and allow normal interactions. The main laptop screen stays pitch black.

UPDATE: at least using the external screens I was able to save all my work and do a graceful reboot. Now all screens are working again.

Same issue, in my case switching to another tty and back to plasma does not work consistently.

Happening on a Thinkpad T14s Gen 2 (Intel iRIS Xe graphics) on Wayland session

have to CTL-ALT # to terminal, CTL-ALT# to open plasma desktop

Same issue. Arch Linux, since 6.2.0 (ongoing 6.2.2): Desktop-PC (AMD APU): trying to wake up from Sleep results in a complete system freeze (black screen, cursor visible). Neither the Wayland session nor the X session work. Cannot switch to another TTY either as the system becomes completely unresponsive. All I can do is to forcibly shutdown. Also tried with a fresh Home folder, same thing. Only thing that helps is going back to Plasma < 6.2 (e.g. 6.1.5 works fine).

I have a HDMI 32" monitor: Peak PMO S322-IQSC

Same here on Fedora Kinoite 41 (but not each rpm-ostree version; IIRC, 2024-10-20 or 2024-10-21 had this constantly).

System’s dead, no tty switching possible, no ctrl-alt-backspace, nothing working, except power off.

Note that rebasing to Fedora Silverblue 41 (Gnome) does not show this issue.

No really useful log entries in journal (journalctl -S12:00; opened laptop lid on 2024-10-26 at 12:22):

Oct 26 12:22:21 atomic.fritz.box kernel: Freezing user space processes completed (elapsed 0.001 seconds)
Oct 26 12:22:21 atomic.fritz.box kernel: OOM killer disabled.
Oct 26 12:22:21 atomic.fritz.box kernel: Freezing remaining freezable tasks
Oct 26 12:22:21 atomic.fritz.box kernel: Freezing remaining freezable tasks completed (elapsed 0.001 seconds)
Oct 26 12:22:21 atomic.fritz.box kernel: printk: Suspending console(s) (use no_console_suspend to debug)
Oct 26 12:22:21 atomic.fritz.box kernel: PM: suspend devices took 0.198 seconds
Oct 26 12:22:21 atomic.fritz.box kernel: ACPI: EC: interrupt blocked
Oct 26 12:22:21 atomic.fritz.box kernel: ACPI: PM: Preparing to enter system sleep state S3

Did anyone file a bug report?

Edit: system is reachable from network (was a firewall issue). I can ssh into it, but still no signs of things gone wrong.

1 Like

I have filed a bug report.
My view is that this affects only a small number of people with specific hardware.
I have narrowed down the bug containing package to these four:
4 extra/plasma-workspace 6.1.5-1 → 6.2.2-1
3 extra/plasma5support 6.1.5-1 → 6.2.2-1
2 extra/polkit-kde-agent 6.1.5-1 → 6.2.2-1
1 extra/powerdevil 6.1.5-1 → 6.2.2-1

What’s the issue number or a link to the ticket? Searching for black screen, suspend, wake up did not show your report.

1 Like

sorry, lost track of the issue number

black screen after momentary plasma desktop appearance

now I have narrowed down the fault to be either:
2 extra/polkit-kde-agent 6.1.5-1 → 6.2.2-1
1 extra/powerdevil 6.1.5-1 → 6.2.2-1

do I need those packages?

Are you able to find the bug report in bugs.kde.org logging with your account? I would like to add informations.

OS: Fedora 40
Plasma Version: 6.2.2
Frameworks Version: 6.7.0
Qt version: 6.7.2
Kernel version: 6.11.4-201.fc40.x86_64 (64 bit)
Graphics platform: Wayland

Model: ThinkBook 15 G3 ACL
Graphics processor: AMD Radeon Graphics
CPU: MD Ryzen 7 5700U with Radeon Graphics
RAM: 16 Gb

A temporary workaround recommendation to those affected: when leaving, instead of allowing the locked screen to go black after a timeout, just put the whole machine to sleep manually. When you’re back, it wakes up normally, and you don’t need to apply any log out / log in and tty shenanigans.

1 Like

I am also having this issue. I uninstalled powerdevil and rebooted, and the problem persists. It’s not powerdevil.

I had this happen to me to, and with Kate open with a file… Thankfully the Kate swap wasn’t deleted so I was able to recover my changes. This needs to be patched ASAP.

I’ve bound my power button to Sleep like @anton_k suggested in case this happens to me again. Usually I have it bound to showing the logout screen.

Can also confirm this happens on Arch on the Framework 16. I also noticed it happening on someone else’s laptop on Fedora 40 on the Lenovo Yoga 7i, but I’m not sure if this is still happening there.

1 Like

This does also occur on my Sofa PC which is connected to a 1920x1080 screen via hdmi to a receiver.
Sometimes when I switch from vt back to graphics screen the screen is briefly visible and then goes dark again. After some repetitions I can get it to stay.
This also might happen if I switch away from a working graphics screen.

System Information:

Operating System: EndeavourOS
KDE Plasma Version: 6.2.1
KDE Frameworks Version: 6.7.0
Qt Version: 6.8.0
Kernel Version: 6.11.5-2-cachyos (64-bit)
Graphics Platform: Wayland
Processors: 12 × Intel® Xeon® W-2133 CPU @ 3.60GHz
Memory: 125,4 GiB of RAM
Graphics Processor: NVIDIA TITAN Xp/PCIe/SSE2
Manufacturer: LENOVO
Product Name: 30BFS75R02
System Version: ThinkStation P520

It may not be a Plasma problem. I used the workaround in this bug id 2321268 at bugzilla dot redhat dot com. I’m not allowed to link it.

sudo systemctl edit --stdin --drop-in=50-keep-warm.conf systemd-suspend.service <<EOF
[Service]
Environment=SYSTEMD_SLEEP_FREEZE_USER_SESSIONS=0
EOF

The output was:

Successfully installed edited file '/etc/systemd/system/systemd-suspend.service.d/50-keep-warm.conf'

I saved that so I can revert this change later. I did a systemctl daemon-reload. This completely resolved the issue for me.