Fedora 41 wayland (Plasma) is a burden on my CPU, renders KDE/Plasma sluggish, and sucks my battery empty

Ask Fedora
1

Hi,
I got a new/another laptop. I freshly installed fedora 41. I wanted KDE/Plasma (not gnome) so that’s what I additionally installed and activated (following the instructions how to do that).

All looked fine (like on my previous laptop). But after a very short while, I notices the window system is sluggish, the fan running constantly, in short the machine is under heavy load.

It’s so slow that working is not fluently doable. it gets worse when starting a browser, it seems somehow graphics rendering that brings it to its knees (videos are not possible).

Typing top shows that it’s invariably kwin_wayland that take the undisputed top-spot
on resource consumption (but plasmashell is mentioned at second place; also firefox ranks high if running).

It makes KDE basically unusable (even if I close the lid of the notebook, it seems to consume energy at quite some , instead of hibernating or similar.
I did not do many systematic experiments, but it takes 60 or 70 minutes of work to empty a completely full battery and it’s unpleasant work, because the large latency of the interface, the mouse, the keyboard…)

What can I do?

  • can I deactivate wayland and get some alternative? I don’t want to use gnome
  • shall I go back to fedora 40? on my older laptop (some company, similar line of models =
    HP elitebooks) it works fine…
  • are there alteratives to Gnome/KDE, maybe simpler more lightweight, that support
    fluently and easily configurable virtual desktops?
  • shall I just wait , F41 is jquite new, maybe it’s a know problem and will be fixed at some point?

One thing I don’t want to do (I don’t currenrly have the time), to experiment around of tweak kernel settings or esotheric stuff (at least not like "one could try this, or try that’') unless
it’s clear what the problem is and what it’s solution. So information like “that’s a known problem of wayland, and currenrly unsolved on some platforms” is valuable information as well (in which case I would have to activate a differnet window manager).

Thanks, Martin

2

We do not know the full details of your hardware.

The command inxi -Fzxx will provide the details needed for most of the hardware.

We have seen similar when a laptop contains a GPU without having hardware acceleration in the driver and the CPU is tasked with graphics rendering.

This is only a guess and cannot be confirmed without seeing the results of the above command for better analysis.

3

Which ISO did you use to install Fedora? The lines I copied tell me you did not install the KDE spin, but installed (added) KDE later. Why so complicated, and where did you find instructions to add KDE to your setup? How did you do that?
I have a feeling this way of doing things is either not correct, or something went wrong doing so.
Please tell us more about the why and how of this installation?

4

Hi, thanks a lot.

Indeed I did not install some spin. Why did I do that? Well, not particular
rational decision. Basically “lazyness”: I installed fedora quite a number
of times (and earlier some other distribution). I always log the steps,
like works and what not, and what the stumbling blocks are and how I solved
them; that was helpful in the early years of linuxes, when stuff was not so
mature, nowadays it often works mostly like a breeze. And when I need to
install something the next round, I simply follow the route that worked
last time. And in this case, the workstation-route with KDE (and other
stuff) later, always worked for me.

Where did I find the instructions? I did

   dnf group install fedora

as is listed on KDE - Fedora Project Wiki, which I think I did also in the
previous installation (the one where KDE/Wayland had not performance
touble). This piece I did not write down in detail, just
``installing/activating KDE: easy, as last time’’ :slight_smile:

More about the why and how of the installation? Well, it’s a new laptop (or
maybe almost new? I got it from the tech guys, they had it on stock, and
maybe it was a machine that they got back from some others). It had some
windows installed, which I erased by repartitioning the disk (I just took
the suggested default, without bothering of making any specific partition
decisions), I booted from the ISO-image on USB, that worked without warning
or unusual. I afterwards installed (via DNF) a number ``apps’’ I want, but
nothing out of the ordinary and nothing that I would expect interferes in a
drastic way with the window system.

There is one thing that comes to my mind that may had an affect. When
“migrating” to a new laptop, I have the habit to backup my old /home/user/
and, when the new one is up-and-running, copy it 100% to the fresh system,
under the same user name. That way, “all” stuff works as before
(passphrases, configurations of numerous tools, etc) so I can contninue my
real work without starting to configure endless stuff to make my self ``at
home’’ at a naked computer.

That implies I copied whatever KDE configurations/.caches there might have
been, maybe that interferes. I did not investigate (it just occurs to me
now), but in previous such situations it was never problematic. And
actually, on the new laptop, my main KDE adaptation (the virtual desktops
and their behavior) was not taken over by copying my home, so I configured
that as one of the first things myself (the only tweak of KDE/plasma
desktop that I did, but a must-have for me…)

That’s all I remember, how I did it (quite conventional and smooth, I
think, were it now for the “resource problem” afterwards).

Thanks, Martin

5

Hi, thanks, here is the output for the diagnosis command (I ran it not as
superusers, but I guess the serial number are not needed).

thanks, Martin

[msteffen@fedora ~]$ inxi -Fzxx
System:
  Kernel: 6.11.10-300.fc41.x86_64 arch: x86_64 bits: 64 compiler: gcc
    v: 2.43.1-2.fc41
  Desktop: KDE Plasma v: 6.2.4 tk: Qt v: N/A wm: kwin_wayland dm: 1: GDM
    2: SDDM note: stopped Distro: Fedora Linux 41 (Workstation Edition)
Machine:
  Type: Laptop System: HP product: HP EliteBook 840 G8 Notebook PC v: SBKPF
    serial: <superuser required> Chassis: type: 10 serial: <superuser required>
  Mobo: HP model: 880D v: KBC Version 30.56.00 serial: <superuser required>
    part-nu: 19X35AV UEFI: HP v: T76 Ver. 01.18.00 date: 08/28/2024
Battery:
  ID-1: BAT0 charge: 30.2 Wh (72.4%) condition: 41.7/53.2 Wh (78.3%)
    volts: 11.6 min: 11.6 model: Hewlett-Packard Primary serial: <filter>
    status: discharging
CPU:
  Info: quad core model: 11th Gen Intel Core i5-1135G7 bits: 64 type: MT MCP
    arch: Tiger Lake rev: 1 cache: L1: 320 KiB L2: 5 MiB L3: 8 MiB
  Speed (MHz): avg: 1700 min/max: 400/4200 cores: 1: 1700 2: 1700 3: 1700
    4: 1700 5: 1700 6: 1700 7: 1700 8: 1700 bogomips: 38707
  Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx
Graphics:
  Device-1: Intel TigerLake-LP GT2 [Iris Xe Graphics] vendor: Hewlett-Packard
    driver: N/A arch: Gen-12.1 bus-ID: 00:02.0 chip-ID: 8086:9a49
  Device-2: Quanta HP HD Camera driver: uvcvideo type: USB rev: 2.0
    speed: 480 Mb/s lanes: 1 bus-ID: 3-2:2 chip-ID: 0408:5348
  Display: wayland server: Xwayland v: 24.1.4 compositor: kwin_wayland
    driver: X: loaded: modesetting unloaded: fbdev,vesa dri: iris gpu: N/A
    display-ID: 0
  Monitor-1: Unknown-1 res: 1920x1080 size: N/A
  API: EGL v: 1.5 platforms: device: 0 drv: swrast surfaceless: drv: swrast
    wayland: drv: swrast x11: drv: swrast inactive: gbm
  API: OpenGL v: 4.5 vendor: mesa v: 24.2.8 glx-v: 1.4 direct-render: yes
    renderer: llvmpipe (LLVM 19.1.0 256 bits) device-ID: ffffffff:ffffffff
    display-ID: :0.0
  API: Vulkan v: 1.3.296 surfaces: xcb,xlib,wayland device: 0 type: cpu
    driver: N/A device-ID: 10005:0000
Audio:
  Device-1: Intel Tiger Lake-LP Smart Sound Audio vendor: Hewlett-Packard
    driver: sof-audio-pci-intel-tgl bus-ID: 00:1f.3 chip-ID: 8086:a0c8
  API: ALSA v: k6.11.10-300.fc41.x86_64 status: kernel-api
  Server-1: JACK v: 1.9.22 status: off
  Server-2: PipeWire v: 1.2.7 status: active with: 1: pipewire-pulse
    status: active 2: wireplumber status: active 3: pipewire-alsa type: plugin
Network:
  Device-1: Intel Wi-Fi 6 AX201 driver: iwlwifi v: kernel bus-ID: 00:14.3
    chip-ID: 8086:a0f0
  IF: wlp0s20f3 state: up mac: <filter>
Bluetooth:
  Device-1: Intel AX201 Bluetooth driver: btusb v: 0.8 type: USB rev: 2.0
    speed: 12 Mb/s lanes: 1 bus-ID: 3-10:3 chip-ID: 8087:0026
  Report: btmgmt ID: hci0 rfk-id: 0 state: up address: <filter> bt-v: 5.2
    lmp-v: 11
RAID:
  Hardware-1: Intel Volume Management Device NVMe RAID Controller driver: vmd
    v: 0.6 bus-ID: 00:0e.0 chip-ID: 8086:9a0b
Drives:
  Local Storage: total: 476.94 GiB used: 388.45 GiB (81.4%)
  ID-1: /dev/nvme0n1 vendor: Micron model: MTFDHBA512TDV-1AZ1AABHA
    size: 476.94 GiB speed: 31.6 Gb/s lanes: 4 serial: <filter> temp: 31.9 C
Partition:
  ID-1: / size: 475.35 GiB used: 388.07 GiB (81.6%) fs: btrfs
    dev: /dev/nvme0n1p3
  ID-2: /boot size: 973.4 MiB used: 365.3 MiB (37.5%) fs: ext4
    dev: /dev/nvme0n1p2
  ID-3: /boot/efi size: 598.8 MiB used: 19.3 MiB (3.2%) fs: vfat
    dev: /dev/nvme0n1p1
  ID-4: /home size: 475.35 GiB used: 388.07 GiB (81.6%) fs: btrfs
    dev: /dev/nvme0n1p3
Swap:
  ID-1: swap-1 type: zram size: 8 GiB used: 0 KiB (0.0%) priority: 100
    dev: /dev/zram0
Sensors:
  System Temperatures: cpu: 70.0 C mobo: N/A
  Fan Speeds (rpm): N/A
Info:
  Memory: total: 16 GiB note: est. available: 15.3 GiB used: 3.98 GiB (26.0%)
  Processes: 309 Power: uptime: 38m wakeups: 0 Init: systemd v: 256
    target: graphical (5) default: graphical
  Packages: pm: rpm pkgs: N/A note: see --rpm Compilers: gcc: 14.2.1
    Shell: Bash v: 5.2.32 running-in: emacs inxi: 3.3.36
[msteffen@fedora ~]$
6

The intel GPU should have the i915 driver in use but that does not show as active in the inxi output you posted.
To me that indicates as the reason for the overload on the CPU.

First try rebooting and if the i915 driver still does not load we will need to dig deeper to determine why.

7

thanks. I rebooted (actually I had rebooted earlier already) with shutdown -r now, but the situation is unchanged, and inxy -Fzxx did not show difference wrt to Graphics:

Martin

8

Use sudo dmesg and peruse the output to see what is happening and why the i915 driver is not being loaded. It may be a firmware that does not load or a driver that fails to load even after the firmware does load.

The command lspci -nnk | grep -iA3 vga should show the details of the GPU to assist in troubleshooting.

9

Hi Martin,

Give this a try:
create a file /etc/modprobe.d/Iris.conf with the following:
options i915 modeset=1

Then reboot …

10

thanks, I did the lspci command, and it gave the following

lspci -nnk | grep -iA3 vga
00:02.0 VGA compatible controller [0300]: Intel Corporation TigerLake-LP GT2 [Iris Xe Graphics] [8086:9a49] (rev 01)
	DeviceName: Onboard IGD
	Subsystem: Hewlett-Packard Company Device [103c:880d]
	Kernel modules: i915, xe

Not that I am familiar what lspci actually does, but I see i915 mentioned and no errors/warnings.

I did also dmesg, and in the long output, I could not find i915 (but I was not sure what exactly I should look at). best, Martin

11

Hi Martin,

See if this process is running:
ps -elf | grep tracker-miner-fs

and in top see if it is grabbing a lot of CPU …

If it is, then you might want to have a look at this:

12

Hi, Einer, changing the /etc/modprobe.d/Iris.conf with that line did not change the behavior after a reboot. Martin

13

Hi, noe, the tracker-miner-fs is not the culprit, it’s not even running according to ps. Martin

14

Here is an update: in the meantime the problem disappeared, with no further interventions
from my side or digging deeper. I had and still have real work currently to do under some deadline, so
I had not patience and time to figure something out (or surf the web for further potential ways around as it seemed out of may area of knowledge), I simply
avoided graphics/browsing the net etc to keep the load down, focussed on my work and lived with it.

But I realize now that the problem disappeared. Maybe as consequence of one of the regular “dnf update” actions, though I can’t say which, as I payed no attention to which packages were updated (can be quite a lot). Seems like “wait until the problem disappears” turned out to work.

best, Martin