On 13/07/2023 21:30, Andrew Lunn wrote:
> On Thu, Jul 13, 2023 at 08:53:34PM +0200, Krzysztof Kozlowski wrote:
>> On 13/07/2023 17:59, Andrew Lunn wrote:
>>>>> +examples:
>>>>> + - |
>>>>> + mdio {
>>>>> + #address-cells = <1>;
>>>>> + #size-cells = <0>;
>>>>> + phy: ethernet-phy@0 {
>>>>> + reg = <0>;
>>>>> + };
>>>>
>>>> Drop this node, quite obvious.
>>>
>>> Dumb question. Isn't it needed since it is referenced by phy-handle =
>>> <&phy> below. Without it, the fragment is not valid DT and so the
>>> checking tools will fail?
>>
>> No, because the example is compiled with silencing missing phandles.
>
> Ah, thanks.
>
> This is a rather odd device, there is no other like it in mainline, so
> i think having that PHY is useful, even if you think it is obvious
> what is going on here.
For almost all other devices in other subsystems we do not provides such
nodes. The "ethernet-phy" node should be obvious because nothing else is
expected to be in "phy-handle". However, if you find it useful, then I
am also fine with it.
Best regards,
Krzysztof
