[Unstable Update] 2022-10-13 - Kernels, Plasma 5.26, KDE Frameworks 5.99, Nvidia 520.56.06

Announcements Unstable Updates
,
18

9 posts were split to a new topic: Can anyone else confirm if modules from already uninstalled kernels are leftover?

19

This issue is not Manjaro related, as it also happens on Arch, but in Plasma 5.26, I no longer have highlights in my application menu.
What I mean is, that in 5.25, there would be a highlight of objects the mouse is over, with the accent color.
This is no longer the case with 5.26. It happens on at least 2 of my machines, My x64 laptop and my Pinebook Pro, so it’s not architecture or driver related either.

So it’s either caused by Plasma 5.26 or KDE Frameworks 5.99.

20

Did we test frameworks 5.99 against 5.25 when Plasma wasn’t updated yet? currently it is just guessing …

21

You’re right I just checked and no high-liting on mouseover, but not really necessary.

22

Maybe an issue with Breeze theme? I still use Oxygen and the highlighting still appears.

24

I’m having this (edit: multi-monitor) issue with Plasma 5.26 (same with 5.26.1).

25

got the same thing, for now i can work around it by having an xrandr command set as an alias

26

With 5.26.1 one of annoying bugs I had before is gone.
On logon, the primary display was used to set to my laptop’s monitor and then quickly it shifted to the external one, as it was specified in system settings. But now the external monitor is set as the primary one from the very beginning of the Plasma session. That’s great!

27

@Yochanan @Chrysostomus

manjaro-ranger-settings needs to be updated to replace nerd-fonts-terminus with upstream ttf-terminus-nerd. Unfortunately Manjaro’s Gitlab has currently issues, so I’m unable to open a bug report.

28

:white_check_mark:

What issues? :thinking:

29

The “create new issue” button was gone, instead a message on the top was displayed… Perhaps a temporary issue.

30

Plasma 5.26.2 seems to have fixed a lot of multi-monitor issues and I’m not experiencing the one I had anymore.

31

Has pamac lost integration with snap?

32

Downgrade snapd-glib from 1.63-2 to 1.63-1 for now. As soon as we have GNOME 43 packages, all will be well.

33

:information_source: Kernel 6.1-rc# might break backlight control on old/weird laptops, please test

2 Likes
34

I wrote an email to Hans with three of the text files for his reference, but received a command not found error when running the acpidump report.

I run acpi_backlight=native as a kernel parameter normally, but when I changed native to video, that broke my backlight controls, so let’s see what he thinks.

35 
❯ pacman -F acpidump
community/acpica 20220331-1
    usr/bin/acpidump

:wink:

1 Like
36

setting kernel parameter

acpi_backlight=video

might break backlight controls, but you are missing the objective. he is asking to set it and see whether after setting it you get following results;

$ ls /sys/class/backlight
acpi_video0  acpi_video1  intel_backlight

if after setting it, you get acpi_video0 as a result, then you are NOT in the problem group, if you dont (still only get whatever you were getting before as the sole result) you are, please re-evaluate.

@Yochanan wouldn’t it be nice if we create some sort of forum post asking ppl from the testing branch (atleast) to join in the effort to help with this.

1 Like
37

Thanks @koshikas, Hans said essentially the same thing on the Phoronix forum and his email reply to me. Adding acpi_backlight=video breaks my backlight control but does output:

$ ls /sys/class/backlight
acpi_video0  intel_backlight

so I shouldn’t be part of the problem group, like you said.

Hans de Goede reply to my inquiry:

Note: this is a standard / templated reply because I’m getting quite
a few emails about this.

Thank you very much for taking the time to test things on your laptop!

Looking at the logs which you have provided backlight control should
keep working fine for you in 6.1.

Some of you have reported non-working backlight-control when passing
“acpi_backlight=video” on the kernel commandline; and/or some extra
(ACPI) error messages in the kernel logs.

This is expected behavior. The “acpi_backlight=video” test run is only
intended to verify if your BIOS/UEFI claims to support ACPI video
backlight control, which is used as part of the kernel’s heuristics to
determine if the native (intel_backlight / …) control should be used.

Unfortunately ACPI video backlight control does not work on all laptops
which claim to support it, hence the problems which you may have seen
with “acpi_backlight=video”. After the test you can and should remove
the “acpi_backlight=video” kernel parameter again.

Regards,

Hans

38