Re: [PATCH 1/3] dt-bindings: soc: renesas: add MFIS binding documentation

[Krzysztof Kozlowski <krzk@kernel.org>]

On 31/03/2026 09:40, Wolfram Sang wrote:
> 
>>> In case you mean this as unanswered questions to v1: This describes the
>>> device specific second mbox cell. Like Tegra does it here (even with
>>
>> I asked Linux ABI. Device specific numbers is not Linux ABI, because
>> Linux is not device.
> 
> I have no idea what you want here. I see it like this: DT is OS

DT, but we talk about bindings. Bindings is not DT. Bindings bind the SW
and DTS.

> agnostic. So, the flags are to describe the hardware. The Linux driver

Well, then no. Bindings headers do not describe the hardware.


> then does the proper things according to these flags. Other OS driver
> might do other things.
> 
>>> shifts instead of plain numbers):
>>>
>>> include/dt-bindings/mailbox/tegra186-hsp.h
> 
> And I still have no idea why my header is different than the Tegra one.
> 

No clue what is there. I am not going even to investigate, because this
is argument of "I found a bug somewhere, so I can implement same bug".

I responded to this *MULTIPLE* times already on the lists. I explained
the rationale, I even provided old messages from Arnd and other people.

Best regards,
Krzysztof