On 1/16/26 11:14 AM, Krzysztof Kozlowski wrote:
> On 16/01/2026 09:50, Tudor Ambarus wrote:
>>
>> Because of the hybrid approach I'm arguing the ACPM child node does not
>> fully describe the hardware, and it's just a firmware abstraction.
>> So option 2/ would be to have just the TMU IP block described with a
>> phandle to the ACPM IPC:
>>
>> soc: soc@0 {
>>     tmu@100a0000 {
>>         compatible = "google,gs101-tmu-top";
>>         reg = <0x100a0000 0x800>;
>>         clocks = <&cmu_misc CLK_GOUT_MISC_TMU_TOP_PCLK>;
>>         interrupts = <GIC_SPI 769 IRQ_TYPE_LEVEL_HIGH 0>;
>>         
>>         /* The "Firmware Phandle" approach */
>>         samsung,acpm-ipc = <&acpm_ipc>;
>>         
>>         #thermal-sensor-cells = <1>;
> 
> Yes, this one, I think it's the best representation.

I was leaning towards this as well. Thank you! I'll start reworking the
patches.

Cheers,
ta


Reply via email to