Screen brightness not restoring after closing laptop lid

Desktop Environments Gnome
1

Unsure if this issue is specifically GNOME related or kernel related but since about a month ago after one of my system updates my Tuxedo laptop suddenly has this weird brightness behaviour.
I close the lid at any % of screen brightness, open it up a few seconds later and bam it’s 100% brightness (even though the value stayed the same) and it only flips back to the intended brightness once I either move the brightness slider or use the brightness keys on my keyboard.

I didn’t have this back in the 6.16.x kernels and before GNOME 49. So either something in 6.17 changed the way the brightness gets restored or in GNOME 49.

Does anybody here have a solution or the same experience?

3 Likes
2

I have been having the same issue exactly as you have described it. I have ignored it despite being an annoyance in the hope that some update to some component in GNOME or something else would resolve it. So far nothing and I have no idea where to look.

2 Likes
3

Thanks for the fast reply, couldn’t find any recent posts about it so thought I might be the only one with the issue.
I’m mostly ignoring it but it is a bummer to get blasted with full brightness late in the evening though :stuck_out_tongue:

Same here, tried to look at the values of the brightness files in /sys/class/backlight/amdgpu_bl1/ but couldn’t really find something meaningful to fix it. Seems to me this is more of a systemd service issue that is supposed to restore the brightness when the lid gets reopened.

2 Likes
4

I have:

cat /var/lib/systemd/backlight/pci-0000:04:00.0:backlight:amdgpu_bl0
24524


cat /var/lib/systemd/backlight/pci-0000:04:00.0:backlight:amdgpu_bl1
17643

but I don’t know how to interpret these values and if they are relevant to this issue.

5

My Intel system has these values:

/sys/class/backlight/intel_backlight/brightness : 20
/sys/class/backlight/intel_backlight/max_brightness : 400

so the current brightness is 5% of the maximum.

Hope this can be of some help.

6

I had this a while ago, I created a resume hook:

Point 5, to run a script to reset the brightness, with something like brightnessctl.

3 Likes
7

It doesn’t unfortunately, because my brightness values are this before closing the lid:

/sys/class/backlight/amdgpu_bl1/brightness : 624
/sys/class/backlight/amdgpu_bl1/max_brightness : 62451
/sys/class/backlight/amdgpu_bl1/actual_brightness : 3084

And after closing the lid and opening it up after 10 seconds:

/sys/class/backlight/amdgpu_bl1/brightness : 624
/sys/class/backlight/amdgpu_bl1/max_brightness : 62451
/sys/class/backlight/amdgpu_bl1/actual_brightness : 63924

So this seems even buggier than expected, the value goes beyond the max brightness for this laptop screen.

@xircon ‘s solution seems like a good workaround for now but this seems like the bug should be mentioned somewhere

8

Thanks for the feedback.
It might be a kernel or GNOME issue (or something else…), really hard to tell. But I believe (or guess) this has already been recognized by some devs, so likely it will be fixed in some future update.

To direct any report to the correct place, could you explore this a bit more to narrow down the reason, e.g. by trying on a different desktop like KDE?

1 Like
9

I’ll try to find another m.2 I have laying around somewhere and load it up with a fresh Endeavour KDE install :slight_smile:

Will report back when it’s done and if there’s any change there.

1 Like
10

I have Arch-Plasma in my multiboot in the same machine. I could test it there. I think however that it has to do something with some component in GNOME and not the kernel.

Had it been the kernel or something else, we would have seen more reports about this issue, I guess.

11

I see the following in journalctl. Can anyone figure out something from it related to the issue described in this topic?

Oct 22 13:48:33 archbox kernel:    acpi_os_write_port+0x13/0x70
Oct 22 13:48:33 archbox kernel:    acpi_ex_system_io_space_handler+0x78/0x120
Oct 22 13:48:33 archbox kernel:    acpi_ev_address_space_dispatch+0x180/0x4a0
Oct 22 13:48:33 archbox kernel:    acpi_ex_access_region+0x284/0x510
Oct 22 13:48:33 archbox kernel:    acpi_ex_field_datum_io+0xa1/0x4f0
Oct 22 13:48:33 archbox kernel:    acpi_ex_write_with_update_rule+0x10d/0x240
Oct 22 13:48:33 archbox kernel:    acpi_ex_insert_into_field+0x29f/0x350
Oct 22 13:48:33 archbox kernel:    acpi_ex_write_data_to_field+0xe6/0x3d0
Oct 22 13:48:33 archbox kernel:    acpi_ex_store_object_to_node+0x1b6/0x3a0
Oct 22 13:48:33 archbox kernel:    acpi_ex_store+0x221/0x490
Oct 22 13:48:33 archbox kernel:    acpi_ex_opcode_1A_1T_1R+0x107/0x670
Oct 22 13:48:33 archbox kernel:    acpi_ds_exec_end_op+0x200/0x880
Oct 22 13:48:33 archbox kernel:    acpi_ps_parse_loop+0x12e/0xa30
Oct 22 13:48:33 archbox kernel:    acpi_ps_parse_aml+0xbd/0x5d0
Oct 22 13:48:33 archbox kernel:    acpi_ps_execute_method+0x171/0x3e0
Oct 22 13:48:33 archbox kernel:    acpi_ns_evaluate+0x197/0x5c0
Oct 22 13:48:33 archbox kernel:    acpi_evaluate_object+0x1ce/0x450
Oct 22 13:48:33 archbox kernel:    __acpi_power_on+0x23/0x100
Oct 22 13:48:33 archbox kernel:    acpi_power_on_list+0x64/0xd0
Oct 22 13:48:33 archbox kernel:    acpi_power_transition+0x7f/0xf0
Oct 22 13:48:33 archbox kernel:    acpi_device_set_power+0x2e6/0x4a0
Oct 22 13:48:33 archbox kernel:    acpi_pci_set_power_state+0x79/0x1e0
Oct 22 13:48:33 archbox kernel:    pci_pm_power_up_and_verify_state+0x2c/0x1c0
Oct 22 13:48:33 archbox kernel:    pci_pm_resume_noirq+0xec/0x1c0
Oct 22 13:48:33 archbox kernel:    dpm_run_callback+0x41/0x1e0
Oct 22 13:48:33 archbox kernel:    async_resume_noirq+0x17c/0x580
Oct 22 13:48:33 archbox kernel:    async_run_entry_fn+0x36/0x140
Oct 22 13:48:33 archbox kernel:    process_one_work+0x193/0x350
Oct 22 13:48:33 archbox kernel:    worker_thread+0x254/0x3a0
Oct 22 13:48:33 archbox kernel:    kthread+0xfc/0x240
Oct 22 13:48:33 archbox kernel:    ret_from_fork+0x1c5/0x1f0
Oct 22 13:48:33 archbox kernel:    ret_from_fork_asm+0x1a/0x30

$ journalctl -p err..alert | grep -i resume

Oct 16 15:50:11 archbox kernel:    pci_pm_resume_noirq+0xec/0x1c0
Oct 16 15:50:11 archbox kernel:    async_resume_noirq+0xf5/0x560
Oct 16 19:12:42 archbox kernel:    pci_pm_resume_noirq+0xec/0x1c0
Oct 16 19:12:42 archbox kernel:    async_resume_noirq+0xf5/0x560
Oct 17 09:36:00 archbox kernel:    pci_pm_resume_noirq+0xec/0x1c0
Oct 17 09:36:00 archbox kernel:    async_resume_noirq+0xf5/0x560
Oct 21 13:05:36 archbox kernel:    pci_pm_resume_noirq+0xec/0x1c0
Oct 21 13:05:36 archbox kernel:    async_resume_noirq+0x17c/0x580
Oct 22 13:48:33 archbox kernel:    pci_pm_resume_noirq+0xec/0x1c0
Oct 22 13:48:33 archbox kernel:    async_resume_noirq+0x17c/0x580
12

@fhireman

Are you using amdgpu too?

Though this doesn’t happen on reboot on my end could it be related to this?

https://wiki.archlinux.org/title/Backlight#Backlight_is_always_at_full_brightness_after_a_reboot_with_amdgpu_driver

:thinking:

13

yeah I have a Ryzen 8845HS with 780M iGPU so using amdgpu.
this seems to be the exact bug that I’m having, thanks so much for sharing!

1 Like
14

I seem to be having this identical issue but on my Debian system.

Some info:

Distributor ID: Debian
Description: Debian GNU/Linux 13 (trixie)
Release: 13
Codename: trixie

Kernel 6.17.3

GNOME Shell 48.4

Enabled extensions:
blur-my-shell@aunetx
uxplay-control@xxanqw
caffeine@patapon.info
icloud-slop@astr64
solaar-extension@sidevesh
quick-settings-tweaks@qwreey
openbar@neuromorph
just-perfection-desktop@just-perfection
top-bar-organizer@julian.gse.jsts.xyz
BringOutSubmenuOfPowerOffLogoutButton@pratap.fastmail.fm
quick-settings-avatar@d-go
wiggle@mechtifs
color-picker@tuberry
auto-accent-colour@Wartybix
smile-extension@mijorus.it
trayIconsReloaded@selfmade.pl
lockkeys@vaina.lt
logomenu@aryan_k
gnome-ui-tune@itstime.tech

15

I have three laptops (HP 845 G10; Thinkpad t14 and x270) with same installation (EndeavourOS and Gnome latest). Two of them AMD (with amdgpu) and one old Intel Laptop (x270).

This happens only on my AMD laptops. And only with newer kernels. LTS is bugless.

So conclusion is that this an AMD GPU only problem with newer kernels or a combination kernel+Gnome. Do we have somebody with Plasma + AMD here to confirm it’s working or not?

16

Checked back today. Some of the latest updated seems to have fixed it. I just resumed the laptop and internal screen brightness was how it should be.

I don’t know what solved the issue as there were updates to kernel, andgpu, firmware, HP laptop firmware, Gnome.

17

What kernel and gnome versions are you running?

You can check via gnome-shell –version, and for the kernel you can use uname -rv

18

❯ gnome-shell --version
GNOME Shell 49.1
❯ uname -rv
6.17.7-arch1-1 #1 SMP PREEMPT_DYNAMIC Sun, 02 Nov 2025 17:27:22 +0000

core/linux-firmware-amdgpu 20251021-1 (25.2 MiB 26.0 MiB) (Installiert)
Firmware files for Linux - Firmware for AMD Radeon GPUs

core/amd-ucode 20251021-1 (128.9 KiB 356.8 KiB) (Installiert)
Microcode update image for AMD CPUs

Works now on both laptops. HP and Thinkpad. So HP firmware update today was not responsible. I assume it’s something in the kernel that has been wrong for some time.

19

I can confirm that indeed the kernel is the issue.

I’ve tried multiple kernels from debian release and some custom compiled kernels.

All of the debian release kernels worked flawlessly and with no issues, ranging from 6.12.4 to the latest.

20

Issue seems to be fixed for me since about a week or so?

GNOME Shell 49.1-2
6.17.7-arch1-1 #1 SMP PREEMPT_DYNAMIC

Unsure which one caused the issue because both gnome-shell and the kernel got updated on 2025-11-04.