Re: [PATCH v3 2/8] ARM: dts: hisilicon: add missing compatibles to CRG node

[Yang Xiwen]

On 2/23/2024 2:18 AM, Krzysztof Kozlowski wrote:
> On 22/02/2024 19:13, Yang Xiwen wrote:
> > On 2/23/2024 2:08 AM, Krzysztof Kozlowski wrote:
> > > On 21/02/2024 17:41, Yang Xiwen via B4 Relay wrote:
> > > > From: Yang Xiwen <forbidden405@xxxxxxxxxxx>
> > > >
> > > > Add "syscon" and "simple-mfd" compatibles to CRG node due to recent
> > > > binding changes.
> > > Why? You claimed you added them in the bindings because DTS has them. In
> > > DTS you claim reason is: binding has them.
> > >
> > > That's confusing.
> >
> > Because the old txt based binding claimed there should not be a "syscon"
> > and "simple-mfd".
> >
> >
> > But it exists in hi3798cv200.dtsi. And i think it does no harm to be
> > there. So should i do it in two commits?
> hi3798cv200 is not hi3519, is it? You are adding simple-mfd to one SoC
> because some other has it? I don't see reason to do that. Er, why?


I think it's the careless HiSilicon people who simply forgot to add it. 
CRG core on these SoCs are very similar. They only provided a bunch of 
clocks and resets. Some register offsets in them are even the same 
across SoCs. So I'll say all CRG devices are "syscon" and "simple-mfd".


In fact, i do have TRM for Hi3519 so i can prove what i said is true.


> Best regards,
> Krzysztof

--
Regards,
Yang Xiwen