Re: [PATCH 3/3] riscv: dts: spacemit: define a SPI controller node

[Alex Elder <elder@riscstar.com>]

On 9/18/25 9:06 AM, Yixun Lan wrote:
> Hi Alex,
> 
> On 08:51 Thu 18 Sep     , Alex Elder wrote:
>> On 9/18/25 8:32 AM, Yixun Lan wrote:
>>>> +			spi3: spi@d401c000 {
>>>> +				compatible = "spacemit,k1-spi";
>>>> +				reg = <0x0 0xd401c000 0x0 0x30>;
>>>> +				#address-cells = <1>;
>>>> +				#size-cells = <0>;
>>>> +				clocks = <&syscon_apbc CLK_SSP3>,
>>>> +					 <&syscon_apbc CLK_SSP3_BUS>;
>>> ..
>>>> +				clock-names = "core",
>>>> +					      "bus";
>>> can you simply put them together in one line? it's kind of tedious to split..
>>
>> Sure I can do that.  I've seen it both ways.
>>
> right, it's merely a coding style I want to enforce, to make it slightly consistent
> 
>>>> +				resets = <&syscon_apbc RESET_SSP3>;
>>>> +				interrupts-extended = <&plic 55>;
>>> why use interrupts-extended?
>>
>> Because it specifies both the controller and interrupt number
>> explicitly.  Why *not* use interrupts-extended?
>>
> It's just unnecessary, the SPI node will fall back to find parent node's interrupt
> which already specific as &plic, brings no benefits

The benefit it brings is that I don't have to search backward to
see what the interrupt controller is.  I realize it's redundant
but I do prefer interrupts-extended over just interrupts.

					-Alex

> 
>>>> +				spacemit,k1-ssp-id = <3>;
>>>> +				dmas = <&pdma 20>,
>>>> +				       <&pdma 19>;
>