Thermel zone failure message on boot (NanoPi NEO2)

[SamuraiJapon]

I am getting a thermal zone failure message during boot, which has always happened with kernels after 6.6, at least as far as I know.

Is this error actually harmful?

 

HardWare:NanoPi NEO2

Armbian Version:

 Packages:     Debian stable (bookworm)  v25.2.3 for NanoPi Neo 2 running Armbian Linux 6.12.20-current-sunxi64
 Support:      DIY (community maintained)

 Kernel:         Linux nanopineo2 6.12.20-current-sunxi64 #1 SMP Sat Mar 22 19:54:28 UTC 2025 aarch64 GNU/Linux

 

dmesg output

[    7.018086] core: _opp_supported_by_regulators: OPP minuV: 1320000 maxuV: 1320000, not supported by regulator
[    7.018125] cpu cpu0: _opp_add: OPP not supported by regulators (1104000000)
[    7.018245] core: _opp_supported_by_regulators: OPP minuV: 1320000 maxuV: 1320000, not supported by regulator
[    7.018258] cpu cpu0: _opp_add: OPP not supported by regulators (1200000000)
[    7.018413] core: _opp_supported_by_regulators: OPP minuV: 1340000 maxuV: 1340000, not supported by regulator
[    7.018426] cpu cpu0: _opp_add: OPP not supported by regulators (1296000000)
[    7.018552] core: _opp_supported_by_regulators: OPP minuV: 1400000 maxuV: 1400000, not supported by regulator
[    7.018564] cpu cpu0: _opp_add: OPP not supported by regulators (1368000000)
[    7.019699] thermal thermal_zone0: binding cdev cpufreq-cpu0 to trip 3 failed: -22
[    7.019785] thermal thermal_zone0: binding cdev cpufreq-cpu0 to trip 4 failed: -22

dmesg.txt

Edited May 6 by SamuraiJapon

[laibsch]

On 5/6/2025 at 6:53 PM, SamuraiJapon said:

Is this error actually harmful?

 

 

You should of course monitor the temperatures now to see if the board becomes abnormally hot (use your hand or the programs sensors or btop, for example).  That being said, a quick web search suggests to me that you are simply seeing a software problem AKA bug.  I suggest you open a bug ticket on the Armbian bug tracker and patiently wait for the board maintainer to have a look into the issue.

 

https://www.armbian.com/nanopi-neo-2/ lists https://github.com/spendist as the maintainer for your board.  Maybe you can send an e-mail?

 

sources:

• https://forum.armbian.com/topic/9501-opzero-h5-after-update-41917-the-cpu-temperature-is-wrong/
• https://community.nxp.com/t5/i-MX-Processors/IMX8MP-not-able-to-read-Thermal-zone0-temperature/m-p/1655665
• https://community.nxp.com/t5/QorIQ/Temperature-is-not-available-at-sys-class-thermal-thermal-zoneX/m-p/1343009

[SamuraiJapon]

laibsch, thanks for your answer.
I tried several kernels from the 6.x.y series on Armbian, but the same error occurred in all of them.
I think this error did not occur with the 4.14 series kernel downloaded from the board manufacturer's official website.
However, when I built and started the 4.14.111+ kernel downloaded from the board manufacturer's github, the following error occurred.

 

dmesg (Friendlycore linux kernel 4.14.111+)
[ 2.243976] thermal thermal_zone0: failed to read out thermal zone (-16)

 

It seems that there has been an error in the thermal driver (dts) of the allwinner H5 SoC for quite some time.
I have already uninstalled Armbian and cannot take logs, so I would like to submit a bug report once I install Armbian again and can take logs.

Edited 6 hours ago by SamuraiJapon