Arch Linux

mistrob

Member

Registered: 2024-09-17

Posts: 6

KDE-NVidia refuses to lock / sleep monitors

Around 2-3 weeks ago I started to have problems with KDE not putting my monitors to sleep and/or locking my machine after 5 minutes.
Sometimes, it will remain unlocked with the screens on, sometimes it will lock and the screens stay on, rarely will it lock and the screens go to sleep.

This was working fine up to I believe when 6.2.1 Plasma was released.
My main monitor is an OLED, so leaving it on can damage it. 

Anyone started to have this issue recently? 

I am currently fully patched up and running Plasma 6.2.2

V1del

Forum Moderator

Registered: 2012-10-16

Posts: 23,875

Re: KDE-NVidia refuses to lock / sleep monitors

Is powerdevil running when it doesn't work? what output do you get from

systemctl --user status plasma-powerdevil
systemd-inhibit --list

mistrob

Member

Registered: 2024-09-17

Posts: 6

Re: KDE-NVidia refuses to lock / sleep monitors

Is powerdevil running when it doesn't work? what output do you get from

systemctl --user status plasma-powerdevil
systemd-inhibit --list

systemd-inhibit --list
● plasma-powerdevil.service - Powerdevil
     Loaded: loaded (/usr/lib/systemd/user/plasma-powerdevil.service; static)
     Active: active (running) since Tue 2024-10-22 21:41:33 EDT; 3 days ago
Invocation: 50dec8a
   Main PID: 4096 (org_kde_powerde)
      Tasks: 11 (limit: 38249)
     Memory: 11.6M (peak: 25.8M swap: 13.4M swap peak: 14.6M zswap: 1.8M)
        CPU: 37.235s
     CGroup: /user.slice/user-1000.slice/user@1000.service/background.slice/plasma-powerdevil.service
             └─4096 /usr/lib/org_kde_powerdevil

Oct 26 09:52:02 x org_kde_powerdevil[4096]: busno=8. Monitor apparently returns -EIO for unsupported features. This cannot be relied on.
Oct 26 09:52:02 x org_kde_powerdevil[4096]: busno=8. Monitor apparently returns -EIO for unsupported features. This cannot be relied on.
Oct 26 09:52:02 x org_kde_powerdevil[4096]: busno=7, Feature 0x41 should not exist but ddc_get_nontable_vcp_value() succeeds, returning mh=0xff ml=0xff s>
Oct 26 09:52:03 x org_kde_powerdevil[4096]: busno=7, Feature 0x00 should not exist but ddc_get_nontable_vcp_value() succeeds, returning mh=0xff ml=0xff s>
Oct 26 09:52:03 x org_kde_powerdevil[4096]: busno=7, All features that should not exist detected. Monitor does not indicate unsupported
Oct 26 09:52:03 x org_kde_powerdevil[4096]: Watch thread started
Oct 26 09:52:03 x org_kde_powerdevil[4096]: Display redetection finished.
Oct 26 09:52:04 x org_kde_powerdevil[4096]: dh=Display_Handle[i2c-7: fd=52], Replacing adjusted sleep multiplier 0.00 with 1.00 for SE_POST_WRITE or SE_P>
Oct 26 09:52:04 x org_kde_powerdevil[4096]: dh=Display_Handle[i2c-6: fd=52], Replacing adjusted sleep multiplier 0.00 with 1.00 for SE_POST_WRITE or SE_P>
Oct 26 09:52:04 x org_kde_powerdevil[4096]: dh=Display_Handle[i2c-8: fd=52], Replacing adjusted sleep multiplier 0.00 with 1.00 for SE_POST_WRITE or SE_P>

Prior to recent updates (about 2-3 weeks ago) it worked fine.  I did have a problem around the same time where all my monitors were set to 7% brightness, I had to adjust the slider to 70% under power "change screen brightness" to 70% to get it to stop.

Before I figured out the exact problem, I was using ddcutil to manually set the brightness on each monitor when I unlock and I got an error unsupported as well but it did change the brightness and was able to toggle hdr and wide color gamut no problem despite the same -EIO unsupported features messages.

Last edited by mistrob (2024-10-26 14:01:58)