Kernel errors - Trixie

Creating a bug report/issue

I have searched the existing open and closed issues

Required Information

  • DietPi version | G_DIETPI_VERSION_CORE=9
    G_DIETPI_VERSION_SUB=16
    G_DIETPI_VERSION_RC=3
    G_GITBRANCH=‘master’
    G_GITOWNER=‘MichaIng’
    G_LIVE_PATCH_STATUS[0]=‘applied’
    G_LIVE_PATCH_STATUS[1]=‘applied’

  • Distro version | trixie

  • Kernel version | Linux DietPi 6.12.41+deb13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.12.41-1 (2025-08-12) x86_64 GNU/Linux

  • Architecture | amd64

  • SBC model | Native PC (x86_64)

  • Intel NUC13ANHi7

  • 16GB DDR4 SO-DIMM (2 X 8GB)

  • Samsung SSD 990 EVO 1TB (M.2 – NVMe, PCIe 4.0 X4)

  • Samsung SSD 870 QVO 1TB, SATA

  • Thunderbolt: OWC Express 1M2 - Portable USB4 NVMe SSD

    • Samsung 990 PRO NVMe M.2 SSD

Additional Information (if applicable)

The errors below did show up after upgrade to Trixie (I used the script)

When using:

dmesg -l 0,1,2,3

it returns errors:

[    1.669165] pci 0000:03:00.0: ASPM: Could not configure common clock
[    2.575986] nvme 0000:04:00.0: Unable to change power state from D3cold to D0, device inaccessible`
[    4.755806] i915 0000:00:02.0: [drm] `*`ERROR`*` GT0: GuC firmware i915/adlp_guc_70.bin: fetch failed -ENOENT
[    4.756623] i915 0000:00:02.0: [drm] `*`ERROR`*` GT0: GuC initialization failed -ENOENT
[    4.756626] i915 0000:00:02.0: [drm] `*`ERROR`*` GT0: Enabling uc failed (-5)
[    4.756627] i915 0000:00:02.0: [drm] `*`ERROR`*` GT0: Failed to initialize GPU, declaring it wedged!`[    8.019717] ucsi_acpi USBC000:00: error -ETIMEDOUT: PPM init failed

Thanks

On native pc, the regular Debian kernel is used, means we don’t maintain this kernel. Not sure if we can do something on this.

Ping @MichaIng

Did not have any errors with Bookworm

Yes the Debian kernel changed between Bookworm and Trixie.

When running “dmesg” this are errors:

1.669165] pci 0000:03:00.0: ASPM: Could not configure common clock
2.575986] nvme 0000:04:00.0: Unable to change power state from D3cold to D0, device inaccessible
4.755806] i915 0000:00:02.0: [drm] *ERROR* GT0: GuC firmware i915/adlp_guc_70.bin: fetch failed -ENOENT
4.756623] i915 0000:00:02.0: [drm] *ERROR* GT0: GuC initialization failed -ENOENT
[    4.756626] i915 0000:00:02.0: [drm] *ERROR* GT0: Enabling uc failed (-5)
[    4.756627] i915 0000:00:02.0: [drm] *ERROR* GT0: Failed to initialize GPU, declaring it wedged!
8.019717] ucsi_acpi USBC000:00: error -ETIMEDOUT: PPM init failed

Yes that’s the same as above. As already told we don’t manage kernel for native pc and I’m not sure if we can do anything on this. Therefore I ask the developer as you can see on my reply above.

It seems that the main problem is i915:

[ 3.949719] i915 0000:00:02.0: [drm] ERROR GT0: GuC firmware i915/adlp_guc_70.bin: fetch failed -ENOENT
[ 3.949723] i915 0000:00:02.0: [drm] GT0: GuC firmware(s) can be downloaded from Making sure you're not a bot!

The i915adlp_guc_70.bin can be found here:

Would like to solve this issue, as I am unaware of what consequences it may have. Let’s hope @MichaIng can help.

I think this problem could impact many Roon users with NUC and DietPI.

Thanks

Debian split off some firmware into dedicated packages. Do you use a desktop or other GUI application, or do some video encoding/decoding on that machine? If so, the firmware/driver might enhance something:

apt install firmware-intel-graphics

Else, you can ignore it.

We do not pre-install these, but it is included in the dietpi-software > Display Options > Display Driver selection, along with OpenGL, Vulkan and VAAPI libraries.

2 Likes

@MichaIng - Thanks for your support. Highly appreciated.

No, I don’t use any desktop or other GUI application, or do video encoding/decoding on the Intel NUC13ANHi7 (Intel Iris® Xe Graphics eligible)

If I would install:

would the errors go away?

THX

something you would need to test

Did not fix the errors:

root@DietPi:~# dmesg -l 0,1,2,3
[ 3.917626] i915 0000:00:02.0: [drm] ERROR GT0: GuC firmware i915/adlp_guc_70.bin: fetch failed -ENOENT
[ 3.918245] i915 0000:00:02.0: [drm] ERROR GT0: GuC initialization failed -ENOENT
[ 3.918248] i915 0000:00:02.0: [drm] ERROR GT0: Enabling uc failed (-5)
[ 3.918249] i915 0000:00:02.0: [drm] ERROR GT0: Failed to initialize GPU, declaring it wedged!
[ 7.257216] ucsi_acpi USBC000:00: error -ETIMEDOUT: PPM init failed
root@DietPi:~#

The package however definitely contains the firmware it is looking for:

Can you verify it is installed?

dpkg -l firmware-intel-graphics
ls -l /usr/lib/firmware/i915/adlp_guc_70.bin

If you do not use a GUI/HDMI anyway, but for whatever reason these one time per boot session errors annoy you, it makes probably more sense to blacklist the related kernel module:

echo 'blacklist i915' | sudo tee /etc/modprobe.d/disable-gpu.conf
sudo update-initramfs -u

But it might be still pulled in by other/parent drivers. In case lsmod can be checked to see which modules are loaded, and which ones are dependencies of which other ones. And if you blacklist the whole display stack, of course attaching a HDMI screen won’t work anymore, until you remove this blacklist config. But probably a fallback driver is used instead, that does not require any firmware.

It says installed here:

BUT

root@DietPi:~# dpkg -l firmware-intel-graphics
dpkg-query: no packages found matching firmware-intel-graphics
root@DietPi:~# ls -l /usr/lib/firmware/i915/adlp_guc_70.bin
ls: cannot access ‘/usr/lib/firmware/i915/adlp_guc_70.bin’: No such file or directory
root@DietPi:~#

I did an uninstall and installed again:

root@DietPi:~# dpkg -l firmware-intel-graphics
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name                    Version      Architecture Description
++±=======================-============-============-========================================
ii  firmware-intel-graphics 20250410-2   all          Binary firmware for Intel iGPUs and IPUs

and

root@DietPi:~# ls -l /usr/lib/firmware/i915/adlp_guc_70.bin
-rw-r–r-- 1 root root 337472 Apr 24 18:35 /usr/lib/firmware/i915/adlp_guc_70.bin
root@DietPi:~#

after a reboot:

[    3.862971] i915 0000:00:02.0: [drm] Found ALDERLAKE_P/RPL-P (device ID a7a0) display version 13.00 stepping E0
[    3.863696] i915 0000:00:02.0: [drm] VT-d active for gfx access
[    3.863701] i915 0000:00:02.0: vgaarb: deactivate vga console
[    3.863781] i915 0000:00:02.0: [drm] Using Transparent Hugepages
[    3.864340] i915 0000:00:02.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=io+mem:owns=io+mem
[    3.869257] i915 0000:00:02.0: [drm] Finished loading DMC firmware i915/adlp_dmc.bin (v2.20)
[    3.872697] i915 0000:00:02.0: [drm] GT0: GuC firmware i915/adlp_guc_70.bin version 70.36.0
[    3.872703] i915 0000:00:02.0: [drm] GT0: HuC firmware i915/tgl_huc.bin version 7.9.3
[    3.887649] i915 0000:00:02.0: [drm] GT0: HuC: authenticated for all workloads
[    3.888034] i915 0000:00:02.0: [drm] GT0: GUC: submission enabled
[    3.888035] i915 0000:00:02.0: [drm] GT0: GUC: SLPC enabled
[    3.888482] i915 0000:00:02.0: [drm] GT0: GUC: RC enabled

The driver errors seems to have disappeared now:

root@DietPi:~# dmesg -l 0,1,2,3
[    7.237634] ucsi_acpi USBC000:00: con1: failed to get status
[   12.417263] ucsi_acpi USBC000:00: error -EBUSY: PPM init failed
root@DietPi:~#

It seems to me, when upgrading (Native PC) from Bookworm to Trixie (using the script) it is necessary to uninstall the display driver and than install it again (reboot).

Just my 2 cent - I am not an expert.

Just for clarification,

firmware-intel-graphics is the firmware, the entry in dietpi-config is for the driver. So the firmware was not intsalled, only the driver.

But it seems to work they way I descripted it. Only left with these two errors:

[    1.653412] pci 0000:03:00.0: ASPM: Could not configure common clock
[    8.079166] ucsi_acpi USBC000:00: error -EBUSY: PPM init failed

I did not install anything else.

Did not install:

apt install firmware-intel-graphics

Yes, respectively to install the particular firmware package that is not available on Bookworm yet. The menu shows what has been selected once, and is hence stored as CONFIG_GPU_DRIVER setting in dietpi.txt, instead of checking for the actual DEB packages being installed or not.

I’ll change two things:

  1. When selecting the already installed driver option, it does install the intended packages again, if any missing. So it is not needed to select “none”, have all packages purged and then install them all again.
  2. If the Intel driver was selected, i.e. CONFIG_GPU_DRIVER=intel, the Trixie upgrade script installs the additional firmware automatically.
1 Like

Has no been solved after reinstall of " Locale : [C.UTF-8] ", It seems that this has to be done after the upgrade. I don’t think I had tis error before, but I am not quit sure

Now I have only this error left:

root@DietPi:~# dmesg -l 0,1,2,3
[    6.995846] ucsi_acpi USBC000:00: error -ETIMEDOUT: PPM init failed
root@DietPi:~# dmesg -l 0,1,2,3

but can’t find any information about that error

The driver seems to probe ACPI. Seem to be related to USB-C ports, for mode switches and PD. ACPI on OS-side is provided e.g. by systemd-logind, but I am not sure whether this is what the driver is looking for, or a missing hardware-side node (which somewhat makes more sense to me).

Again, if you do not face any issues with the USB-C mode, that e.g. a WiFi adapter is detected as storage device with some Windows firmware blobs, failing the mode switch, or charging a phone or what with that port is slower than you expect, you can/should just ignore it. systemd-logind means two more services, and the dbus package. acpi-support-base would be an alternative, but systemd-logind is needed in some other cases as well, so publicly I always recommend that one:

systemctl unmask systemd-logind
apt install dbus
systemctl start dbus systemd-logind

THX :slight_smile:

I have an OWC Express 1M2 (Portable USB4 NVMe SSD) attached and it works just fine (use it as total backup (DietPi + music). So I will just ignore the error.

THX for your great support.

1 Like