Is it safe to switch from Nvidia to an open driver?

raijar · November 12, 2025, 10:04am

How do I set SESSION_TYPE to stay on wayland?

Initially wayland is set on.

When the default is NVIDIA (not intel) and I change the cable from one DP port to another, the screen goes black. I can’t get past this situation except by 1) changing the cable back to the original DP port or 2) turning off the computer by pressing its power button.

Below is journalctl list what happens

localhost:~ # cat failure.txt 
marras 12 11:51:10 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:51:10 localhost.localdomain systemd[4744]: drkonqi-coredump-launcher.socket: Unit needs to be started because active unit sockets.target upholds it, but not starting since we tried this too often recently. Will retry later.
marras 12 11:51:19 localhost.localdomain akonadi_mailmerge_agent[4230]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain kalendarac[4091]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain kactivitymanagerd[3944]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain kded6[3854]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_migration_agent[4231]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_archivemail_agent[4217]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_unifiedmailbox_agent[4234]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain polkit-kde-authentication-agent-1[3948]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_sendlater_agent[4233]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain xdg-desktop-portal-kde[3950]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_newmailnotifier_agent[4232]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_followupreminder_agent[4220]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain chrome[4245]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain konsole[4655]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain keepassxc[4065]: qt.qpa.wayland: Creating a fake screen in order for Qt not to crash
marras 12 11:51:19 localhost.localdomain akonadi_control[4165]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_maildispatcher_agent[4227]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain akonadi_mailfilter_agent[4228]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain konsole[4728]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain plasmashell[3936]: requesting unexisting screen available rect -1
marras 12 11:51:19 localhost.localdomain org_kde_powerdevil[3949]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain plasmashell[3936]: There are no outputs - creating placeholder screen
marras 12 11:51:19 localhost.localdomain systemd[1720]: Started dbus-:1.1-org.kde.KSplash@3.service.
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Removing connected display on bus 14
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Emitting DDCA_Display_Status_Event[169.967:  DDCA_EVENT_DISPLAY_DISCONNECTED, card1-DP-4, dref: DDCA_Display_Ref[3], io_path:/dev/i2c-14, ddc working: false]
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Starting 1 callback threads
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4918] libddcutil callback thread 0x7f1cec01ffa0 started
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Started 1 event callback thread(s)
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4998] Invoking callback function 0x7f1d18b780b0 for event DDCA_Display_Status_Event[169.967:  DDCA_EVENT_DISPLAY_DISCONNECTED, card1-DP-4, dref: DDCA_Display_Ref[3], io_path:/dev/i2c-14, ddc working: false] in this thread [  4998]
marras 12 11:51:20 localhost.localdomain org_kde_powerdevil[3949]: [  4998] Callback function 0x7f1d18b780b0 for event DDCA_Display_Status_Event[169.967:  DDCA_EVENT_DISPLAY_DISCONNECTED, card1-DP-4, dref: DDCA_Display_Ref[3], io_path:/dev/i2c-14, ddc working: false] complete
marras 12 11:51:21 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:52:12 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:52:12 localhost.localdomain systemd[4744]: drkonqi-coredump-launcher.socket: Unit needs to be started because active unit sockets.target upholds it, but not starting since we tried this too often recently. Will retry later.
marras 12 11:52:13 localhost.localdomain kwin_wayland[3772]: 0x501: GL_INVALID_VALUE error generated. Size and/or offset out of range.
marras 12 11:52:13 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Adding connected display with bus 14
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Emitting DDCA_Display_Status_Event[223.395:  DDCA_EVENT_DISPLAY_CONNECTED, card1-DP-4, dref: DDCA_Display_Ref[4], io_path:/dev/i2c-14, ddc working: true]
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Starting 1 callback threads
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  4918] libddcutil callback thread 0x7f1cec0133d0 started
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  4918] Started 1 event callback thread(s)
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  5018] Invoking callback function 0x7f1d18b780b0 for event DDCA_Display_Status_Event[223.395:  DDCA_EVENT_DISPLAY_CONNECTED, card1-DP-4, dref: DDCA_Display_Ref[4], io_path:/dev/i2c-14, ddc working: true] in this thread [  5018]
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  5018] Callback function 0x7f1d18b780b0 for event DDCA_Display_Status_Event[223.395:  DDCA_EVENT_DISPLAY_CONNECTED, card1-DP-4, dref: DDCA_Display_Ref[4], io_path:/dev/i2c-14, ddc working: true] complete
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Quiescing libddcutil API...
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Quiesce libddcutil API complete
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Display redetection starting.
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  4917] recheck thread terminating because watch thread terminated
marras 12 11:52:14 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Watch thread terminated.
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Watching for display connection changes, resolved watch mode = Watch_Mode_Xevent, poll loop interval = 100 millisec
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  3949]                                          extra_stabilization_millisec: 0,  stabilization_poll_millisec: 100
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  3949] libddcutil recheck thread 0x556513480110 started
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  3949] libddcutil watch thread 0x55651345f460 started
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Display redetection finished.
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  3949] Unquiescing libddcutil API...
marras 12 11:52:17 localhost.localdomain org_kde_powerdevil[3949]: [  5019] (dw_recheck_displays_func) Recheck interval: Sleeping for 200 milliseconds
marras 12 11:52:19 localhost.localdomain plasma_waitforname[4996]: WaitForName: Service was not registered within timeout
marras 12 11:52:19 localhost.localdomain systemd[1720]: dbus-:1.1-org.kde.KSplash@3.service: Main process exited, code=exited, status=1/FAILURE
marras 12 11:52:19 localhost.localdomain systemd[1720]: dbus-:1.1-org.kde.KSplash@3.service: Failed with result 'exit-code'.
marras 12 11:52:21 localhost.localdomain plasma_waitforname[4999]: WaitForName: Service was not registered within timeout
marras 12 11:52:21 localhost.localdomain systemd[1720]: dbus-:1.1-org.freedesktop.Notifications@3.service: Main process exited, code=exited, status=1/FAILURE
marras 12 11:52:21 localhost.localdomain systemd[1720]: dbus-:1.1-org.freedesktop.Notifications@3.service: Failed with result 'exit-code'.
marras 12 11:52:21 localhost.localdomain kded6[3854]: Failed to notify "Could not activate remote peer 'org.freedesktop.Notifications': unit failed"
marras 12 11:52:22 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:53:54 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:53:54 localhost.localdomain systemd[4744]: drkonqi-coredump-launcher.socket: Unit needs to be started because active unit sockets.target upholds it, but not starting since we tried this too often recently. Will retry later.
marras 12 11:54:01 localhost.localdomain systemd[4744]: Created slice User Background Tasks Slice.
marras 12 11:54:01 localhost.localdomain systemd[4744]: Starting Cleanup of User's Temporary Files and Directories...
marras 12 11:54:01 localhost.localdomain systemd[4744]: Finished Cleanup of User's Temporary Files and Directories.
marras 12 11:54:05 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:54:15 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
marras 12 11:54:15 localhost.localdomain systemd[4744]: drkonqi-coredump-launcher.socket: Unit needs to be started because active unit sockets.target upholds it, but not starting since we tried this too often recently. Will retry later.
marras 12 11:54:17 localhost.localdomain NetworkManager[1253]: <info>  [1762941257.0300] device (wlo1): set-hw-addr: set MAC address to F6:8B:CA:02:05:19 (scanning)
marras 12 11:54:17 localhost.localdomain NetworkManager[1253]: <info>  [1762941257.0499] device (wlo1): supplicant interface state: inactive -> disconnected
marras 12 11:54:17 localhost.localdomain NetworkManager[1253]: <info>  [1762941257.0499] device (p2p-dev-wlo1): supplicant management interface state: inactive -> disconnected
marras 12 11:54:17 localhost.localdomain NetworkManager[1253]: <info>  [1762941257.0557] device (wlo1): supplicant interface state: disconnected -> inactive
marras 12 11:54:17 localhost.localdomain NetworkManager[1253]: <info>  [1762941257.0557] device (p2p-dev-wlo1): supplicant management interface state: disconnected -> inactive
marras 12 11:54:25 localhost.localdomain systemd[4744]: Socket to launch DrKonqi for a systemd-coredump crash was skipped because of an unmet condition check (ConditionUser=!@system).
localhost:~ #

panorain · November 16, 2025, 8:19pm

Hi, I wonder what switcherooctl list shows now.

-Great Hopes

malcolmlewis · November 16, 2025, 11:08pm

@raijar so it could be switching from Nvidia to Intel ARC, but the desktop environment may not work with it… It would be interesting to see if you switch cables back and boot to multi user with the cables swapped back (to the port that’s not working)

So could just edit boot options and add a 3 at the end of the linux line in grub, or boot to nvidia (or the port that works), run systemctl set-default multi-user.target power off, swap cables and boot. Do you get to a tty (console) login?

raijar · November 17, 2025, 12:28pm

I did as I was told.
And the computer worked as you expected.
The computer’s graphical interface didn’t start. But I was able to open a tty session.
Now the only problem is that X won’t start through either DP port.
How do I eliminate the “systemctl set-default multi-user.target” command?
ps. start up my computer to read only mode.

raijar · November 17, 2025, 12:28pm

I did as I was told.
And the computer worked as you expected.
The computer’s graphical interface didn’t start. But I was able to open a tty session.
Now the only problem is that X won’t start through either DP port.
How do I eliminate the “systemctl set-default multi-user.target” command?
ps. I start up my computer to read only mode.

malcolmlewis · November 17, 2025, 12:34pm

@raijar ok, so log into the tty and run inxi -Gxxz and post the output, you can pipe it to a file for recover later eg inxi -Gxxz > inxi.txt

Then log in as root user and run systemctl isolate graphical.target does the desktop start?

Set the default back from the tty after logging in as root user and running systemctl set-default graphical.target

raijar · November 17, 2025, 1:15pm

The command “systemctl isolate graphical.target” does not start Desktop.
After issuing the command “systemctl set-default graphical.target”, the computer starts normally.

But the inxi listing is empty or otherwise faulty?

My fault! I gave the inxi command incorrectly
I’ll try again.

raijar · November 17, 2025, 1:59pm

Below inxi listing

localhost:~ # cat inxi.txt 
Graphics:
  Device-1: Intel Meteor Lake-P [Intel Arc Graphics] vendor: ASUSTeK driver: i915 v: kernel
    arch: Xe-LPG ports: active: none empty: DP-1,DP-2 bus-ID: 00:02.0 chip-ID: 8086:7d55
  Device-2: NVIDIA AD107M [GeForce RTX 4060 Max-Q / Mobile] vendor: ASUSTeK driver: nvidia
    v: 580.105.08 arch: Lovelace pcie: speed: 2.5 GT/s lanes: 8 ports: active: DP-3
    empty: DP-4,HDMI-A-1 bus-ID: 01:00.0 chip-ID: 10de:28a0
  Display: unspecified server: X.org v: 1.21.1.15 with: Xwayland v: 24.1.8 driver: X:
    loaded: modesetting,nvidia unloaded: vesa alternate: fbdev,intel,nouveau,nv dri: iris
    gpu: nvidia,nvidia-nvswitch tty: 240x67
  Monitor-1: DP-3 model: Asus MX279 res: 1920x1080 dpi: 82 diag: 686mm (27")
  API: EGL v: 1.5 platforms: device: 0 drv: nvidia device: 2 drv: iris device: 3 drv: swrast
    gbm: drv: nvidia surfaceless: drv: nvidia inactive: wayland,x11,device-1
  API: OpenGL v: 4.6.0 compat-v: 4.5 vendor: mesa v: 25.2.7 note: console (EGL sourced)
    renderer: NVIDIA GeForce RTX 4060 Laptop GPU/PCIe/SSE2, Mesa Intel Arc Graphics (MTL), llvmpipe
    (LLVM 21.1.5 256 bits)
  API: Vulkan v: 1.4.328 surfaces: N/A device: 0 type: discrete-gpu driver: nvidia
    device-ID: 10de:28a0 device: 1 type: integrated-gpu driver: mesa intel device-ID: 8086:7d55
    device: 2 type: cpu driver: mesa llvmpipe device-ID: 10005:0000
  Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo de: kscreen-console,kscreen-doctor
    gpu: nvidia-settings,nvidia-smi wl: wayland-info x11: xdpyinfo, xprop, xrandr
localhost:~ #

malcolmlewis · November 17, 2025, 2:18pm

@raijar so this is when booting in multi-user and with the DP connection that normally gives a black screen when booting to graphical?

If so, sure it’s not an actual hardware issue? It’s not actually switching to the intel ARC.

OK, need you to boot to multi-user again and connect each DP port in turn and run;
/sbin/lspci -nnk | grep -EA3 "VGA|Display|3D" and post the output, again you can pipe each one to a file with the >.

The reason for above is to see if the display switches to a different class id on switching DP ports.

Plus you still running X11?

The other option is to look at trying the Xe driver on the ARC gpu… again this can all be done at multi-user, then running the isolate command to see if it switches.

To try switching, as root user you would run;

update-bootloader --add-option i915.force_probe=\!7d55
update-bootloader --add-option xe.force_probe=7d55
update-bootloader

To remove it’s the same except --del-option

Then start graphical via systemctl isolate graphical.target.

Be also interesting for you to look at the logs for errors…

raijar · November 17, 2025, 3:54pm

I’m sorry, but I can’t figure out the problem anymore.
I’m wondering if it’s a hardware or software problem, and how I can figure it out.
I could install another distro or even Windows, but that seems like too much work.
However, the thing is, I haven’t been able to get the Intel driver to work with any trick.

malcolmlewis · November 17, 2025, 4:09pm

@raijar You might have to install Windows… there could be some switching that is done to activate/de-activate hardware… That’s why I suggested the ASUS forum…

raijar · November 17, 2025, 5:04pm

OK.

But I’m definitely not switching to Windows. I had to use Windows 3.11 for work groups a long time ago at work. And I hated it from the start.
Vax/VMS was much better!

OrsoBruno · November 17, 2025, 6:54pm

Referring to another thread of yours I guess that we are reading about an Asus NUC 14 Performance NUC14SRKU7 Intel Core Ultra 7 155H NVIDIA RTX 4060 Mini PC.
My understanding is that it is a NUC made for performance, the Nvidia looks like an essential piece of it and guessing from the layout maybe the DP ports are wired to the Nvidia.
If the above is correct, there is no way to have it working without a (proprietary?) Nvidia driver installed. It looks like a “dual graphics” system, not an “Optimus” hybrid laptop where saving battery power might be an issue, it looks indeed like a piece of a powerhouse.
I cannot tell for sure based on public Asus docs, but if you cannot find a setting in the firmware (AKA BIOS) to disable the Nvidia, just install the Nvidia driver and enjoy. I (and Malcolm and others) have been using the proprietary drivers on Tumbleweed for years without showstoppers, so if you are careful with installation and upgrade I don’t understand your worry.

raijar · November 18, 2025, 12:11pm

Thank you very much to all the experts who advised me!

I had a lot of trouble getting my new computer to even boot up. Then the driver problems started. I lost faith in my new computer, an expensive but hopeless task to fix. With great difficulty I managed to install the NVidia driver and I thought it would be a good idea to get the Intel driver working before I even tried to change to the Nvidia driver to another driver.

Usually there is some bug or error in the system that explains the strange behavior, but I have not understood the operation of my computer with the DP ports at all. However, I have thought that it may indeed be a matter of the computer being designed to work this way (as it works now).

I will still investigate the BIOS, but I am content with my fate. Next time I will stick to AMD.

malcolmlewis · November 18, 2025, 12:44pm

@raijar So have you tried Windows 11 on it, to me there is a hardware problem and I’d be looking at sending it back…

raijar · November 18, 2025, 1:25pm

OK.
I have a pretty big renovation going on in my house, so I’m a bit tired of testing my computer…
But I’m also interested in testing Widows 11. So I’ll test it, but not right now.

raijar · November 18, 2025, 3:24pm

I couldn’t find any parameter that affects the NVidia/Intel driver. Attached is the version of BIOS.

raijar · November 20, 2025, 10:08am

The NVdia graphics driver is physically connected to ports 8 and 9.
The Intel driver is only connected to port 11.
I don’t have a suitable cable for port 11, so I can’t test it, but I think it works. So my computer is working as designed.

system · November 27, 2025, 10:08am

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.