Re: [PATCH v4 6/7] riscv: dts: starfive: jh7110: Add ethernet device node

[yanhong wang]

On 2023/2/3 15:09, Krzysztof Kozlowski wrote:
> On 03/02/2023 04:14, yanhong wang wrote:
>> 
>> 
>> On 2023/1/18 23:51, Krzysztof Kozlowski wrote:
>>> On 18/01/2023 07:17, Yanhong Wang wrote:
>>>> Add JH7110 ethernet device node to support gmac driver for the JH7110
>>>> RISC-V SoC.
>>>>
>>>> Signed-off-by: Yanhong Wang <yanhong.wang@xxxxxxxxxxxxxxxx>
>>>> ---
>>>>  arch/riscv/boot/dts/starfive/jh7110.dtsi | 93 ++++++++++++++++++++++++
>>>>  1 file changed, 93 insertions(+)
>>>>
>>>> diff --git a/arch/riscv/boot/dts/starfive/jh7110.dtsi b/arch/riscv/boot/dts/starfive/jh7110.dtsi
>>>> index c22e8f1d2640..c6de6e3b1a25 100644
>>>> --- a/arch/riscv/boot/dts/starfive/jh7110.dtsi
>>>> +++ b/arch/riscv/boot/dts/starfive/jh7110.dtsi
>>>> @@ -433,5 +433,98 @@
>>>>  			reg-shift = <2>;
>>>>  			status = "disabled";
>>>>  		};
>>>> +
>>>> +		stmmac_axi_setup: stmmac-axi-config {
>>>
>>> Why your bindings example is different?
>>>
>> 
>> There are two gmacs on the StarFive VF2 board, and the two
>> gmacs use the same configuration on axi, so the 
>> stmmac_axi_setup is independent, which is different
>> from the bindings example.
>> 
>> 
>>> Were the bindings tested? Ahh, no they were not... Can you send only
>>> tested patches?
>>>
>>> Was this tested?
>>>
>> Yes, the bindings have been tested on the StarFive VF2 board and work normally.
> 
> Then please tell me how did you test the bindings on the board? How is
> it even possible and how the board is related to bindings? As you could
> easily see from Rob's reply they fail, so I have doubts that they were
> tested. If you still claim they were - please paste the output from
> testing command.
> 

Sorry, I didn't check all the bindings, only the modified ones, the command 
used is as follows: 
"make dt_binding_check DT_SCHEMA_FILES=Documentation/devicetree/bindings/net/snps,dwmac.yaml"
"make dt_binding_check DT_SCHEMA_FILES=Documentation/devicetree/bindings/net/starfive,jh7110-dwmac.yaml"

> 
> Best regards,
> Krzysztof
>