You are not logged in.

Pages: 1

#1 2025-12-31 09:20:35

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Hello,

I recently purchased a Legion Pro 5 16 with the following two GPUs:
- Ryzen 9 8940HX iGPU
- RTX 5070 dGPU

I am using the 6.17.11-hardened1-1-hardened kernel, KDE Plasma 6, and Wayland.
I have configured PRIME to use both GPUs.

I have two problems:
- The system seems to use only the iGPU by default, but the dGPU never seems to turn off
- When I adjust the screen brightness, the backlight does not change

Here are somes logs:

Last boot logs:

❯ journalctl -p 3 -xb

déc. 31 09:03:08 archlinux kernel: ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PCI0.PB2], AE_NOT_FOUND (20250404/dswload2-162)
déc. 31 09:03:08 archlinux kernel: ACPI Error: AE_NOT_FOUND, During name lookup/catalog (20250404/psobject-220)
déc. 31 09:03:08 archlinux kernel: 
déc. 31 09:03:10 archlinux kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:10 archlinux kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:23 ltp-arch-nayrosk bootctl[1381]:  Mount point '/boot' which backs the random seed file is world accessible, which is a security hole! 
déc. 31 09:03:23 ltp-arch-nayrosk bootctl[1381]: Random seed file '/boot/loader/random-seed' is world accessible, which is a security hole!
déc. 31 09:03:25 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:25 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:26 ltp-arch-nayrosk bluetoothd[1411]: Failed to set mode: Failed (0x03)
déc. 31 09:03:35 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:35 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:36 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:36 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:03:42 ltp-arch-nayrosk akonadiserver[9141]: Failed to detect mysqld version!
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: database server stopped unexpectedly
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: Database process exited unexpectedly during initial connection!
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: executable: "/usr/bin/mariadb"
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: arguments: QList("--defaults-file=/home/nayrosk/.local/share/akonadi/mysql.conf", "--datadir=/home/nayrosk/.local/share/akonadi/db_data/", "--socket=/run/user/1000/akonadi/mysql.socket", "--pid-fil>
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: stdout: ""
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: stderr: "/usr/bin/mariadb: unknown variable 'datadir=/home/nayrosk/.local/share/akonadi/db_data/'\n/usr/bin/mariadb: unknown variable 'pid-file=/run/user/1000/akonadi/mysql.pid'\n"
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: exit code: 7
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: process error: "Unknown error"
déc. 31 09:03:47 ltp-arch-nayrosk akonadiserver[9141]: See "/home/nayrosk/.local/share/akonadi/db_data/mysql.err" for more details
déc. 31 09:03:51 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:03:52 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:03:52 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:20:50 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 09:45:32 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:45:32 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:45:32 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:45:59 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:45:59 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:45:59 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 09:51:41 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
déc. 31 10:05:26 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 10:05:26 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 10:06:45 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"
déc. 31 10:06:45 ltp-arch-nayrosk kwin_wayland[8415]: Invalid framebuffer status:  "GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT"

My packages:

❯ yay -Qs nvidia

local/egl-gbm 1.1.2.1-1
    The GBM EGL external platform library
local/egl-wayland 4:1.1.21-1
    EGLStream-based Wayland external platform
local/egl-wayland2 1.0.0.rc.r53.gf4a2d32-1
    EGLStream-based Wayland external platform (2)
local/egl-x11 1.0.4-1
    NVIDIA XLib and XCB EGL Platform Library
local/lib32-nvidia-utils 590.48.01-1
    NVIDIA drivers utilities (32-bit)
local/libvdpau 1.5-3
    Nvidia VDPAU library
local/libxnvctrl 590.48.01-1
    NVIDIA NV-CONTROL X extension
local/linux-firmware-nvidia 20251125-2
    Firmware files for Linux - Firmware for NVIDIA GPUs and SoCs
local/nvidia-open-dkms 590.48.01-1
    NVIDIA open kernel modules - module sources
local/nvidia-prime 1.0-5
    NVIDIA Prime Render Offload configuration and utilities
local/nvidia-prime-rtd3pm 1.1-1
    Configure your discrete NVIDIA GPU to power down when not in use.
local/nvidia-settings 590.48.01-1
    Tool for configuring the NVIDIA graphics driver
local/nvidia-utils 590.48.01-1
    NVIDIA drivers utilities
local/nvtop 3.2.0-1
    GPUs process monitoring for AMD, Intel and NVIDIA

❯ yay -Qs amd

local/amd-ucode 20251125-2
    Microcode update image for AMD CPUs
local/distorm 3.5.2.b-4
    Powerful disassembler library for x86/AMD64
local/lib32-vulkan-radeon 1:25.3.2-1
    Open-source Vulkan driver for AMD GPUs - 32-bit
local/libteam 1.32-3
    Library for controlling team network device
local/linux-firmware-amdgpu 20251125-2
    Firmware files for Linux - Firmware for AMD Radeon GPUs
local/nvtop 3.2.0-1
    GPUs process monitoring for AMD, Intel and NVIDIA
local/vulkan-radeon 1:25.3.2-1
    Open-source Vulkan driver for AMD GPUs
local/xf86-video-amdgpu 25.0.0-1 (xorg-drivers)
    X.org amdgpu video driver

In my mkinitcpio.conf:

MODULES=(btrfs amdgpu radeon nvidia nvidia_modeset nvidia_uvm nvidia_drm)
HOOKS=(base systemd autodetect microcode modconf keyboard sd-vconsole block plymouth sd-encrypt filesystems fsck)

In my kernel's options line (I use systemd-boot):

options rd.luks.name=1732ae68-3bb0-4703-aa43-eeb554fbb863=cryptroot root=/dev/mapper/cryptroot rootflags=subvol=@ splash quiet rw amdgpu.dcdebugmask=0x10 nvidia-drm.modeset=1 nvidia-drm.fbdev=1 fsck.mode=force fsck.repair=yes acpi_backlight=video

P.S: I tried video|vendor|native| for acpi_backlight parameter

Nvidia services:

❯ systemctl list-units --type=service --state=running | grep -i nvidia

  nvidia-persistenced.service   loaded active running NVIDIA Persistence Daemon
  nvidia-powerd.service         loaded active running nvidia-powerd service

❯ systemctl list-unit-files --type=service | grep -i nvidia

nvidia-hibernate.service                     enabled         disabled
nvidia-persistenced.service                  enabled         disabled
nvidia-powerd.service                        enabled         disabled
nvidia-resume.service                        enabled         disabled
nvidia-suspend-then-hibernate.service        enabled         disabled
nvidia-suspend.service                       enabled         disabled

My Wayland configuration:

❯ cat /etc/sddm.conf.d/10-display.conf
[General]
DisplayServer=wayland
GreeterEnvironment=QT_WAYLAND_SHELL_INTEGRATION=layer-shell

[Wayland]
CompositorCommand=kwin_wayland --drm --no-lockscreen --no-global-shortcuts --locale1

I installed and enabled switcheroo-control too.

Nvidia related logs:

❯ cat /proc/driver/nvidia/gpus/0000:01:00.0/power
Runtime D3 status:          Enabled (fine-grained)
Tegra iGPU Rail-Gating:     Disabled
Video Memory:               Off

GPU Hardware Support:
 Video Memory Self Refresh: Supported
 Video Memory Off:          Supported

S0ix Power Management:
 Platform Support:          Not Supported
 Status:                    Disabled

Notebook Dynamic Boost:     Supported

❯ cat /sys/bus/pci/devices/0000:00:01.0/power/control
on

❯ sudo fuser -v /dev/nvidia*
                     USER        PID ACCESS COMMAND
/dev/nvidia0:        root       1413 F.... nvidia-powerd
                     nvidia-persistenced   1417 F.... nvidia-persiste
/dev/nvidiactl:      root       1413 F.... nvidia-powerd
                     nvidia-persistenced   1417 F.... nvidia-persiste
                     nayrosk    9231 F.... electron
/dev/nvidia-modeset: nvidia-persistenced   1417 F.... nvidia-persiste

❯ sudo nvidia-smi
*Wed Dec 31 09:27:45 2025       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 590.48.01              Driver Version: 590.48.01      CUDA Version: 13.1     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 5070 ...    On  |   00000000:01:00.0 Off |                  N/A |
| N/A   46C    P5             12W /   95W |       2MiB /   8151MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+

Can you help me solve this problem? I would like the system to adapt the GPU according to the load, and of course a functionnal backlight switch.

Thank you

[UPDATE]

In order for RTD3 to work, the following parameter had to be manually added to the bootloader's options:

NVreg_DynamicPowerManagement=0x03

Last edited by virgilggx (2026-01-09 17:17:23)

Offline

#2 2026-01-01 15:14:13

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Runtime D3 status: Enabled (fine-grained)
Video Memory: Off

it's kind of idle like it will not fully power down if smtg keeps /dev/nvidia* open/active, like  'nvidia-powerd','nvidia-persistenced' if you want fully turn it off then you should stop these services which will break dynamic boost and yeah it's not recommended. this part is working as designed on wayland.
and the second part

amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data

also you have problematic parameters 

amdgpu.dcdebugmask=0x10
acpi_backlight=video

both breaks backlight controls on modern AMD laptops , also it is not recommended to force NVIDIA modules into the initramfs on a hybrid system.
edit: not both, but not recommended.

Last edited by 5hridhyan (2026-01-03 16:26:52)

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#3 2026-01-02 16:20:08

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Thanks for your reply.
I removed amdgpu.dcdebugmask=0x10  and acpi_backlight=video, but the backlight control still doesn't work!

Offline

#4 2026-01-02 17:23:13

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

what is the output of

ls /sys/class/backlight

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#5 2026-01-02 17:27:09

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

5hridhyan wrote:

what is the output of

ls /sys/class/backlight
❯ ls /sys/class/backlight
amdgpu_bl1  nvidia_0

Offline

#6 2026-01-02 17:51:37

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

"amdgpu_bl1" is the actual panel backlight device, and the "nvidia_0" is exposed by the nvidia driver which does not control internal panel  like it should be ignored and in your case KDE might be picking "nvidia_0" as default. try creating udev rule to ignore the NVIDIA backlight by
"sudo nano /etc/udev/rules.d/90-backlight.rules" and adding 

SUBSYSTEM=="backlight", KERNEL=="nvidia_0", OPTIONS+="ignore_device"

then reload udev and reboot and also clear PowerDevil’s cache if needed.

Last edited by 5hridhyan (2026-01-02 18:03:30)

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#7 2026-01-02 22:55:20

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

5hridhyan wrote:

"amdgpu_bl1" is the actual panel backlight device, and the "nvidia_0" is exposed by the nvidia driver which does not control internal panel  like it should be ignored and in your case KDE might be picking "nvidia_0" as default. try creating udev rule to ignore the NVIDIA backlight by
"sudo nano /etc/udev/rules.d/90-backlight.rules" and adding 

SUBSYSTEM=="backlight", KERNEL=="nvidia_0", OPTIONS+="ignore_device"

then reload udev and reboot and also clear PowerDevil’s cache if needed.

I tested it (but I can't find the PowerDevil cache), I reloaded udev and restarted, but nothing changed, it doesn't work. There are still two GPUs in backlight.

Offline

#8 2026-01-03 03:50:31

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

sorry for the confusion earlier, I was wrong about PowerDevil like I forgot in KDE 6 most of the backlight states is runtime only.
I forgot to ask

5hridhyan wrote:

also it is not recommended to force NVIDIA modules into the initramfs on a hybrid system.

Did you remove those modules from mkinitcpio.conf and rebuild the initramfs?
and try

 echo 50 | sudo tee /sys/class/backlight/amdgpu_bl1/brightness

if the brightness changes then it's a problem of KDE Wayland userspace thing.

Last edited by 5hridhyan (2026-01-03 11:39:27)

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#9 2026-01-03 12:30:44

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

5hridhyan wrote:

sorry for the confusion earlier, I was wrong about PowerDevil like I forgot in KDE 6 most of the backlight states is runtime only.
I forgot to ask

5hridhyan wrote:

also it is not recommended to force NVIDIA modules into the initramfs on a hybrid system.

Did you remove those modules from mkinitcpio.conf and rebuild the initramfs?
and try

 echo 50 | sudo tee /sys/class/backlight/amdgpu_bl1/brightness

if the brightness changes then it's a problem of KDE Wayland userspace thing.


After removing the initramfs modules, it seems to work, but stability is affected. Now SDDM segfaults on launch and I have to force a reboot because nothing responds (once in a while).
In any case, if the dGPU cannot be disabled under Wayland, then I might as well use it exclusively, so I think I will modify the configuration to use only the dGPU.

Offline

#10 2026-01-03 12:42:41

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

virgilggx wrote:

After removing the initramfs modules, it seems to work, but stability is affected. Now SDDM segfaults on launch and I have to force a reboot because nothing responds (once in a while).
In any case, if the dGPU cannot be disabled under Wayland, then I might as well use it exclusively, so I think I will modify the configuration to use only the dGPU.

"it seems to work" ok that confirms it was an early-KMS/module-ordering issue, "but stability is affected" exactly how like only SDDM? or anything else??

Last edited by 5hridhyan (2026-01-03 13:09:00)

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#11 2026-01-03 12:51:24

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

5hridhyan wrote:
virgilggx wrote:

After removing the initramfs modules, it seems to work, but stability is affected. Now SDDM segfaults on launch and I have to force a reboot because nothing responds (once in a while).
In any case, if the dGPU cannot be disabled under Wayland, then I might as well use it exclusively, so I think I will modify the configuration to use only the dGPU.

"it seems to work" ok that confirms it was an early-KMS/module-ordering issue, "but stability is affected" exactly how like only SDDM? or anything else??

SDDM starts up in low brightness mode, then when I log in, it crashes (segfault in the logs) and either the desktop appears and works, or it freezes completely (requiring a forced reboot).

I reverted all the configurations to use only the Nvidia GPU, and that doesn't work either; nvtop still shows the AMD GPU as the computing source.

I followed the documentation: NVIDIA_Optimus#Use_NVIDIA_graphics_only for SDDM but nothing change...

Offline

#12 2026-01-03 12:56:23

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Reverted the changes, right? You added the NVIDIA modules back, and SDDM is working as before, or is it still broken?
edit: as said likey due to KMS issue try "nvidia-drm.modeset=1" as parameters rebuild initramfs and also try sddm with x11.

Last edited by 5hridhyan (2026-01-11 18:03:29)

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#13 2026-01-03 13:14:23

5hridhyan
Member
From: 127.0.0.1
Registered: 2025-12-25
Posts: 73

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

virgilggx wrote:

In any case, if the dGPU cannot be disabled under Wayland, then I might as well use it exclusively, so I think I will modify the configuration to use only the dGPU.

it's a hybrid system where the iGPU owns the internel panel and many more so if you go with configuring to use dGPU only you might end up with other problems and

nvtop still shows the AMD GPU as the computing source.

yes that is a expected behavior.

Last edited by 5hridhyan (2026-01-03 13:16:01)

We don’t finish life. We just learn to love the unfinished lines.  — Grok

Offline

#14 2026-01-03 14:17:08

seth
Member
From: Don't DM me only for attention
Registered: 2012-09-03
Posts: 72,065

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

On the nvidia situation, please illustrate the status quo:
Please post your complete system journal for the boot: 

sudo journalctl -b | curl -F 'file=@-' 0x0.st

And also 

nvidia-smi
glxinfo -B

On the backlight situation: has this been resolved?
If not:

brightnessctl -l

will list the backlight devices, 

brightnessctl -d <device> <value>

will allow you to control a specific device, likely "amdgpu_bl1"

If you want to use the nvidia GPU exclusively, the only sane approach would be to do that in the BIOS/UEFI to rewire the output.

Fwwi, amdgpu.dcdebugmask=0x10 disables PSR - you don't want to do that, but I don't think it would have systematic impact on the backlight control (it just avoids the AMD GPU from having to frequently wake up just in order to kick the panel into the butt to redraw the same output)

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

Offline

#15 2026-01-03 17:51:04

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

seth wrote:

On the nvidia situation, please illustrate the status quo:
Please post your complete system journal for the boot: 

sudo journalctl -b | curl -F 'file=@-' 0x0.st

And also 

nvidia-smi
glxinfo -B

On the backlight situation: has this been resolved?
If not:

brightnessctl -l

will list the backlight devices, 

brightnessctl -d <device> <value>

will allow you to control a specific device, likely "amdgpu_bl1"

If you want to use the nvidia GPU exclusively, the only sane approach would be to do that in the BIOS/UEFI to rewire the output.

Fwwi, amdgpu.dcdebugmask=0x10 disables PSR - you don't want to do that, but I don't think it would have systematic impact on the backlight control (it just avoids the AMD GPU from having to frequently wake up just in order to kick the panel into the butt to redraw the same output)

In mkinitcpio :

MODULES=(btrfs usbhid xhci_hcd amdgpu radeon)
HOOKS=(base systemd autodetect microcode modconf keyboard sd-vconsole block plymouth sd-encrypt filesystems fsck)

In bootloader's entry :

options rd.luks.name=1732ae68-3bb0-4703-aa43-eeb554fbb863=cryptroot root=/dev/mapper/cryptroot rootflags=subvol=@ splash quiet rw nvidia-drm.modeset=1 nvidia-drm.fbdev=1 fsck.mode=force fsck.repair=yes

According to the wiki, I added this to the SDDM settings (in /usr/share/sddm/scripts/Xsetup :

#!/bin/sh
# Xsetup - run as root before the login dialog appears
xrandr --setprovideroutputsource modesetting NVIDIA-0
xrandr --auto

I uninstalled nvidia-prime nvidia-prime-rtd3pm switcheroo-control and restarted.

And no, the backlight problem has not been resolved.

According to brightnessctl, it is the Nvidia device that controls the backlights.
But when I change the brightness, the values do not change:

Device 'nvidia_0' of class 'backlight':
	Current brightness: 100 (100%)
	Max brightness: 100
Device 'amdgpu_bl1' of class 'backlight':
	Current brightness: 65535 (100%)
	Max brightness: 65535

Ideally, I would like to have the following setup:
- Stay on Wayland
- The GPU is chosen according to the load
- If the integrated GPU is used, then the dedicated GPU is turned off (for battery life)

But if it's not possible I would like to use only dGPU

https://0x0.st/Pooq.txt

❯ nvidia-smi
Sat Jan  3 18:37:07 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 590.48.01              Driver Version: 590.48.01      CUDA Version: 13.1     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 5070 ...    On  |   00000000:01:00.0 Off |                  N/A |
| N/A   47C    P2             11W /   78W |       2MiB /   8151MiB |     11%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+
❯ glxinfo -B
name of display: :0
display: :0  screen: 0
direct rendering: Yes
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: AMD (0x1002)
    Device: AMD Radeon 610M (radeonsi, raphael_mendocino, LLVM 21.1.6, DRM 3.64, 6.17.13-hardened1-1-hardened) (0x164e)
    Version: 25.3.2
    Accelerated: yes
    Video memory: 512MB
    Unified memory: no
    Preferred profile: core (0x1)
    Max core profile version: 4.6
    Max compat profile version: 4.6
    Max GLES1 profile version: 1.1
    Max GLES[23] profile version: 3.2
Memory info (GL_ATI_meminfo):
    VBO free memory - total: 14 MB, largest block: 14 MB
    VBO free aux. memory - total: 14475 MB, largest block: 14475 MB
    Texture free memory - total: 14 MB, largest block: 14 MB
    Texture free aux. memory - total: 14475 MB, largest block: 14475 MB
    Renderbuffer free memory - total: 14 MB, largest block: 14 MB
    Renderbuffer free aux. memory - total: 14475 MB, largest block: 14475 MB
Memory info (GL_NVX_gpu_memory_info):
    Dedicated video memory: 512 MB
    Total available memory: 16196 MB
    Currently available dedicated video memory: 14 MB
OpenGL vendor string: AMD
OpenGL renderer string: AMD Radeon 610M (radeonsi, raphael_mendocino, LLVM 21.1.6, DRM 3.64, 6.17.13-hardened1-1-hardened)
OpenGL core profile version string: 4.6 (Core Profile) Mesa 25.3.2-arch1.1
OpenGL core profile shading language version string: 4.60
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile

OpenGL version string: 4.6 (Compatibility Profile) Mesa 25.3.2-arch1.1
OpenGL shading language version string: 4.60
OpenGL context flags: (none)
OpenGL profile mask: compatibility profile

OpenGL ES profile version string: OpenGL ES 3.2 Mesa 25.3.2-arch1.1
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20

Offline

#16 2026-01-03 20:23:25

seth
Member
From: Don't DM me only for attention
Registered: 2012-09-03
Posts: 72,065

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

MODULES=(btrfs usbhid xhci_hcd amdgpu radeon)

Remove the radeon module from that list, it's for pre-gcn GPUs

Fwwi, "nvidia-drm.modeset=1 nvidia-drm.fbdev=1" are the default values.

I added this to the SDDM settings

This is usually no longer necessary, certainly not if the output is attached to the GPU the server runs on.

I uninstalled nvidia-prime nvidia-prime-rtd3pm switcheroo-control and restarted.

Restore nvidia-prime and nvidia-prime-rtd3pm ?

According to brightnessctl, it is the Nvidia device that controls the backlights.

No. No idea where you want to see that, but that's false.

janv. 03 13:40:56 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes

There's no output attached to the nvidia GPU,
According to the nvidia-smi output, nothing is running on that GPU and your glxinfo also reports the AMD GPU.
Therefore

janv. 03 13:40:52 ltp-arch-nayrosk kernel: nvidia_bl: loading out-of-tree module taints kernel.
janv. 03 13:40:52 ltp-arch-nayrosk kernel: nvidia_bl: module verification failed: signature and/or required key missing - tainting kernel
janv. 03 13:40:52 ltp-arch-nayrosk kernel: nvidia_bl: No supported Nvidia graphics adapter found

remove that!
(The module is deprecated bitrot anyway)

brightnessctl -d amdgpu_bl1 s 50%

nb. the "%", because

janv. 03 13:40:53 ltp-arch-nayrosk systemd-backlight[1211]: amdgpu_bl1: Saved brightness 50 is too low; increasing to 3276.

Ideally, I would like to have the following setup:
- The GPU is chosen according to the load

That's not a thing, you can chose what processes to run on the GPU.

- Stay on Wayland
- If the integrated GPU is used, then the dedicated GPU is turned off (for battery life)

That's more plausible.

Leaving aside 

janv. 03 13:40:57 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: GStreamer; Unable to play - "http://a1.v2.phobos.apple.com.edgesuite.net/us/r1000/000/Features/atv/AutumnResources/videos/b6-1.mov"
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: pam_kwallet5(sddm:auth): pam_kwallet5: pam_sm_authenticate
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: [PAM] returning.
janv. 03 13:41:09 ltp-arch-nayrosk sddm[1544]: Authentication for user  "nayrosk"  successful
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: pam_kwallet5(sddm:setcred): pam_kwallet5: pam_sm_setcred
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: pam_unix(sddm:session): session opened for user nayrosk(uid=1000) by nayrosk(uid=0)
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Message received from daemon: LoginSucceeded
janv. 03 13:41:09 ltp-arch-nayrosk kernel: QSGRenderThread[1678]: segfault at 18 ip 000069e71951cc22 sp 000069e6abffe8c0 error 6 in libEGL_mesa.so.0.0.0[24c22,69e7194fe000+2b000] likely on CPU 7 (core 7, socket 0)
janv. 03 13:41:09 ltp-arch-nayrosk kernel: Code: c0 be 01 00 00 00 4c 89 f7 31 c0 52 31 d2 41 57 ff 15 52 be 02 00 58 48 8b 83 58 01 00 00 5a 48 8b 53 38 48 8b bb 78 01 00 00 <48> 89 50 18 48 c7 83 60 01 00 00 00 00 00 00 ff 15 99 c1 02 00 83
janv. 03 13:41:09 ltp-arch-nayrosk systemd-coredump[10759]: Process 1578 (sddm-greeter) of user 964 terminated abnormally with signal 11/SEGV, processing...
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Created slice Slice /system/drkonqi-coredump-processor.
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Created slice Slice /system/systemd-coredump.
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Started Process Core Dump (PID 10759/UID 0).
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Started Pass systemd-coredump journal entries to relevant user for potential DrKonqi handling.
janv. 03 13:41:09 ltp-arch-nayrosk systemd-logind[1438]: New session '2' of user 'nayrosk' with class 'user' and type 'wayland'.

there's some problem w/ the sddm greeter (running on the experimental wayland backend) but you're ultimately logging into a kwin_wayland session.

Since no process is currently attached to the nvidia GPU there's not actual reason for it to not power down, however

janv. 03 13:42:17 ltp-arch-nayrosk sudo[12750]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi
janv. 03 13:42:17 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 13:42:44 ltp-arch-nayrosk sudo[13595]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi
janv. 03 13:42:44 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 13:43:09 ltp-arch-nayrosk sudo[13818]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi
janv. 03 13:43:09 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 13:52:05 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 15:46:17 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 16:58:43 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 18:31:22 ltp-arch-nayrosk sudo[27451]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi

nvidia-smi itself will by necessity wake the GPU and so will fancy desktop widgetlets that poll it's power draw, temperature or fan speed etc.
You can only poll /sys/bus/pci/devices/0000:00:01.0/power/control to figure it's power state (and it will typically take some seconds before it automatically powers down)
For clarification, the "nvidia-smi" output was drawn from the KDE session?

janv. 03 13:40:55 ltp-arch-nayrosk systemd[1]: Started nvidia-powerd service.
janv. 03 13:40:55 ltp-arch-nayrosk nvidia-powerd[1436]: nvidia-powerd version:2.0 (build 1)
janv. 03 13:40:56 ltp-arch-nayrosk nvidia-powerd[1436]: DBus Connection is established

I'd for the time being also disable https://wiki.archlinux.org/title/CPU_fr … dia-powerd

One final obstacle could be 

janv. 03 13:40:37 archlinux kernel: smpboot: CPU0: AMD Ryzen 9 8940HX with Radeon Graphics (family: 0x19, model: 0x61, stepping: 0x2)

because the features is advertised for intel CPUs…
In that case your only option might be https://wiki.archlinux.org/title/Bumble … management - we'll see what's left after cleaning up the rest.

For a principle functionality test, please only boot the multi-user.target (2nd link below, console only, no SDDM nor KDE nor nothing else) and check whether the nvidia GPU enters RTD3 after adressing all of the above.

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

Offline

#17 2026-01-04 10:04:18

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

seth wrote:

MODULES=(btrfs usbhid xhci_hcd amdgpu radeon)

Remove the radeon module from that list, it's for pre-gcn GPUs

Fwwi, "nvidia-drm.modeset=1 nvidia-drm.fbdev=1" are the default values.

I added this to the SDDM settings

This is usually no longer necessary, certainly not if the output is attached to the GPU the server runs on.

I uninstalled nvidia-prime nvidia-prime-rtd3pm switcheroo-control and restarted.

Restore nvidia-prime and nvidia-prime-rtd3pm ?

According to brightnessctl, it is the Nvidia device that controls the backlights.

No. No idea where you want to see that, but that's false.

janv. 03 13:40:56 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes

There's no output attached to the nvidia GPU,
According to the nvidia-smi output, nothing is running on that GPU and your glxinfo also reports the AMD GPU.
Therefore

janv. 03 13:40:52 ltp-arch-nayrosk kernel: nvidia_bl: loading out-of-tree module taints kernel.
janv. 03 13:40:52 ltp-arch-nayrosk kernel: nvidia_bl: module verification failed: signature and/or required key missing - tainting kernel
janv. 03 13:40:52 ltp-arch-nayrosk kernel: nvidia_bl: No supported Nvidia graphics adapter found

remove that!
(The module is deprecated bitrot anyway)

brightnessctl -d amdgpu_bl1 s 50%

nb. the "%", because

janv. 03 13:40:53 ltp-arch-nayrosk systemd-backlight[1211]: amdgpu_bl1: Saved brightness 50 is too low; increasing to 3276.

Ideally, I would like to have the following setup:
- The GPU is chosen according to the load

That's not a thing, you can chose what processes to run on the GPU.

- Stay on Wayland
- If the integrated GPU is used, then the dedicated GPU is turned off (for battery life)

That's more plausible.

Leaving aside 

janv. 03 13:40:57 ltp-arch-nayrosk kernel: amdgpu 0000:05:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Error: "Internal data stream error."
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: GStreamer; Unable to play - "http://a1.v2.phobos.apple.com.edgesuite.net/us/r1000/000/Features/atv/AutumnResources/videos/b6-1.mov"
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: pam_kwallet5(sddm:auth): pam_kwallet5: pam_sm_authenticate
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: [PAM] returning.
janv. 03 13:41:09 ltp-arch-nayrosk sddm[1544]: Authentication for user  "nayrosk"  successful
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: pam_kwallet5(sddm:setcred): pam_kwallet5: pam_sm_setcred
janv. 03 13:41:09 ltp-arch-nayrosk sddm-helper[10679]: pam_unix(sddm:session): session opened for user nayrosk(uid=1000) by nayrosk(uid=0)
janv. 03 13:41:09 ltp-arch-nayrosk sddm-greeter[1578]: Message received from daemon: LoginSucceeded
janv. 03 13:41:09 ltp-arch-nayrosk kernel: QSGRenderThread[1678]: segfault at 18 ip 000069e71951cc22 sp 000069e6abffe8c0 error 6 in libEGL_mesa.so.0.0.0[24c22,69e7194fe000+2b000] likely on CPU 7 (core 7, socket 0)
janv. 03 13:41:09 ltp-arch-nayrosk kernel: Code: c0 be 01 00 00 00 4c 89 f7 31 c0 52 31 d2 41 57 ff 15 52 be 02 00 58 48 8b 83 58 01 00 00 5a 48 8b 53 38 48 8b bb 78 01 00 00 <48> 89 50 18 48 c7 83 60 01 00 00 00 00 00 00 ff 15 99 c1 02 00 83
janv. 03 13:41:09 ltp-arch-nayrosk systemd-coredump[10759]: Process 1578 (sddm-greeter) of user 964 terminated abnormally with signal 11/SEGV, processing...
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Created slice Slice /system/drkonqi-coredump-processor.
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Created slice Slice /system/systemd-coredump.
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Started Process Core Dump (PID 10759/UID 0).
janv. 03 13:41:09 ltp-arch-nayrosk systemd[1]: Started Pass systemd-coredump journal entries to relevant user for potential DrKonqi handling.
janv. 03 13:41:09 ltp-arch-nayrosk systemd-logind[1438]: New session '2' of user 'nayrosk' with class 'user' and type 'wayland'.

there's some problem w/ the sddm greeter (running on the experimental wayland backend) but you're ultimately logging into a kwin_wayland session.

Since no process is currently attached to the nvidia GPU there's not actual reason for it to not power down, however

janv. 03 13:42:17 ltp-arch-nayrosk sudo[12750]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi
janv. 03 13:42:17 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 13:42:44 ltp-arch-nayrosk sudo[13595]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi
janv. 03 13:42:44 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 13:43:09 ltp-arch-nayrosk sudo[13818]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi
janv. 03 13:43:09 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 13:52:05 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 15:46:17 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 16:58:43 ltp-arch-nayrosk kernel: nvidia 0000:01:00.0: Enabling HDA controller
janv. 03 18:31:22 ltp-arch-nayrosk sudo[27451]:  nayrosk : TTY=pts/1 ; PWD=/home/nayrosk ; USER=root ; COMMAND=/usr/bin/nvidia-smi

nvidia-smi itself will by necessity wake the GPU and so will fancy desktop widgetlets that poll it's power draw, temperature or fan speed etc.
You can only poll /sys/bus/pci/devices/0000:00:01.0/power/control to figure it's power state (and it will typically take some seconds before it automatically powers down)
For clarification, the "nvidia-smi" output was drawn from the KDE session?

janv. 03 13:40:55 ltp-arch-nayrosk systemd[1]: Started nvidia-powerd service.
janv. 03 13:40:55 ltp-arch-nayrosk nvidia-powerd[1436]: nvidia-powerd version:2.0 (build 1)
janv. 03 13:40:56 ltp-arch-nayrosk nvidia-powerd[1436]: DBus Connection is established

I'd for the time being also disable https://wiki.archlinux.org/title/CPU_fr … dia-powerd

One final obstacle could be 

janv. 03 13:40:37 archlinux kernel: smpboot: CPU0: AMD Ryzen 9 8940HX with Radeon Graphics (family: 0x19, model: 0x61, stepping: 0x2)

because the features is advertised for intel CPUs…
In that case your only option might be https://wiki.archlinux.org/title/Bumble … management - we'll see what's left after cleaning up the rest.

For a principle functionality test, please only boot the multi-user.target (2nd link below, console only, no SDDM nor KDE nor nothing else) and check whether the nvidia GPU enters RTD3 after adressing all of the above.

Thanks for the explanations.
I followed all the steps you sent me.

I can confirm that even in console mode, the dGPU does not shut down (/sys/bus/pci/devices/0000:00:01.0/power/control returns on).
Also, brightnessctl -d amdgpu_bl1 s 50% does not change anything.

I am attaching my latest boot log: https://0x0.st/Poao.txt

Offline

#18 2026-01-04 12:53:36

seth
Member
From: Don't DM me only for attention
Registered: 2012-09-03
Posts: 72,065

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Please avoid bloating the thread with completely pointless full quotes.

brightnessctl -d amdgpu_bl1 s 50% does not change anything.

Incl. the set value, ie.  "brightnessctl i amdgpu_bl1" does not report 50%?

I can confirm that even in console mode, the dGPU does not shut down (/sys/bus/pci/devices/0000:00:01.0/power/control returns on).

For clarification: https://wiki.archlinux.org/title/PRIME# … Management

Edit: google lied to me!
You have a Turing chip which in this combination simply might not be eligible for RTD3, but:
1. You have added /etc/udev/rules.d/80-nvidia-pm.rules ? What exactly does it look like? Did you ensure the rules are applied?
2. NVreg_DynamicPowerManagement is set to 0x02 ? (In doubt add "nvidia.NVreg_DynamicPowerManagement=0x02" to the https://wiki.archlinux.org/title/Kernel_parameters )
3. Because of "Some users with pre-Ampere card and broken D3 support on newer drivers reported a workaround to disable the GSP firmware with NVreg_EnableGpuFirmware=0" have you tested https://aur.archlinux.org/packages?O=0&K=580xx and https://wiki.archlinux.org/title/NVIDIA … P_firmware (this is not an option w/ nvidia-open! It *needs* the GSP)

10de:2d58 is a blackwell chip, you *have* to use nvidia-open and GSP and RTD3 should work by default!
sad

Last edited by seth (2026-01-04 13:18:24)

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

Offline

#19 2026-01-05 18:01:58

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

seth wrote:

Incl. the set value, ie.  "brightnessctl i amdgpu_bl1" does not report 50%?.

The command works fine, the percentage changes, but the display does not change at all.
The backlights do not change when I adjust the brightness or when I type the command.

seth wrote:

10de:2d58 is a blackwell chip, you *have* to use nvidia-open and GSP and RTD3 should work by default!

I don't understand, am I supposed to use the normal version instead of the DKMS version?

Offline

#20 2026-01-05 22:55:09

seth
Member
From: Don't DM me only for attention
Registered: 2012-09-03
Posts: 72,065

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

I don't understand, am I supposed to use the normal version instead of the DKMS version?

No, I was just stressing that because of my googles! previous mistake

The backlights do not change when I adjust the brightness or when I type the command.

AMD has been fumbling around w/ the backlight algorithm a lot, try to add "amdgpu.dcdebugmask=0x40000" to the https://wiki.archlinux.org/title/Kernel_parameters

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

Offline

#21 2026-01-09 06:24:24

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

seth wrote:

IAMD has been fumbling around w/ the backlight algorithm a lot, try to add "amdgpu.dcdebugmask=0x40000" to the https://wiki.archlinux.org/title/Kernel_parameters

I tried, but nothing change: https://0x0.st/PHku.txt

Offline

#22 2026-01-09 07:59:51

seth
Member
From: Don't DM me only for attention
Registered: 2012-09-03
Posts: 72,065

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Nothing change implies you still get to change the values in sysfs but the panel doesn't respond?

tail /sys/class/backlight/amdgpu_bl*/*

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc

Offline

#23 2026-01-09 08:02:00

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

seth wrote:

Nothing change implies you still get to change the values in sysfs but the panel doesn't respond?

Yes sad

Offline

#24 2026-01-09 11:29:22

Gatoroto
Member
Registered: 2024-07-31
Posts: 3

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

Hello, I have Lenovo Legion 5 Pro 16 as well, but with MD Ryzen 5 5600H CPU and NVIDIA GeForce RTX 3060 Mobile GPU. I was able to got brightness working in KDE+wayland with the following kernel parameters:

amdgpu.backlight=0 acpi_backlight=native

Hope this will help you.

Offline

#25 Yesterday 11:25:15

virgilggx
Member
Registered: 2024-05-12
Posts: 22

Re: Unable to disable the dGPU / brighness issue on Lenovo Pro 5 16

I edited my first post to confirm that I found a solution to disable the NVIDIA card when it is not in use.

Gatoroto wrote:

I was able to got brightness working in KDE+wayland with the following kernel parameters:

amdgpu.backlight=0 acpi_backlight=native

Hope this will help you.

As for the backlights, it still doesn't work, despite these settings.

Offline

Pages: 1

Board footer

Powered by FluxBB