Re: [PATCH 08/14] arm64: dts: meson-gxl: fix internal phy compatible

2019-08-21 Thread Neil Armstrong
On 20/08/2019 22:38, Martin Blumenstingl wrote:
> adding Jerome
> 
> On Wed, Aug 14, 2019 at 4:31 PM Neil Armstrong  
> wrote:
>>
>> This fixes the following DT schemas check errors:
>> meson-gxl-s805x-libretech-ac.dt.yaml: ethernet-phy@8: compatible: 
>> ['ethernet-phy-id0181.4400', 'ethernet-phy-ieee802.3-c22'] is not valid 
>> under any of the given schemas
>>
>> Signed-off-by: Neil Armstrong 
>> ---
>>  arch/arm64/boot/dts/amlogic/meson-gxl.dtsi | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi 
>> b/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi
>> index ee1ecdbcc958..43eb158bee24 100644
>> --- a/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi
>> +++ b/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi
>> @@ -709,7 +709,7 @@
>> #size-cells = <0>;
>>
>> internal_phy: ethernet-phy@8 {
>> -   compatible = "ethernet-phy-id0181.4400", 
>> "ethernet-phy-ieee802.3-c22";
>> +   compatible = "ethernet-phy-id0181.4400";
> on G12A there was a specific reason (iirc it was because the PHY ID
> can be any arbitrary value programmed into some register) why we added
> it with a compatible string
> Jerome, do we have the same situation on GXL/GXM as well?

Yes the ID in encoded in the register, but we use the very basic mmio mux, but 
we should
switch to the same g12a-mdio-mux at some point..

=> infinite TODO list !

Neil

> 
> if not I prefer to drop the compatible string because it's probably
> from a time where the PHY dt-bindings stated "add the PHY ID
> compatible string if you know it" while the actual suggestion was
> "only add it if reading the ID doesn't work for some reason"
> 
> 
> Martin
> 



Re: [PATCH 08/14] arm64: dts: meson-gxl: fix internal phy compatible

2019-08-20 Thread Martin Blumenstingl
adding Jerome

On Wed, Aug 14, 2019 at 4:31 PM Neil Armstrong  wrote:
>
> This fixes the following DT schemas check errors:
> meson-gxl-s805x-libretech-ac.dt.yaml: ethernet-phy@8: compatible: 
> ['ethernet-phy-id0181.4400', 'ethernet-phy-ieee802.3-c22'] is not valid under 
> any of the given schemas
>
> Signed-off-by: Neil Armstrong 
> ---
>  arch/arm64/boot/dts/amlogic/meson-gxl.dtsi | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi 
> b/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi
> index ee1ecdbcc958..43eb158bee24 100644
> --- a/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi
> +++ b/arch/arm64/boot/dts/amlogic/meson-gxl.dtsi
> @@ -709,7 +709,7 @@
> #size-cells = <0>;
>
> internal_phy: ethernet-phy@8 {
> -   compatible = "ethernet-phy-id0181.4400", 
> "ethernet-phy-ieee802.3-c22";
> +   compatible = "ethernet-phy-id0181.4400";
on G12A there was a specific reason (iirc it was because the PHY ID
can be any arbitrary value programmed into some register) why we added
it with a compatible string
Jerome, do we have the same situation on GXL/GXM as well?

if not I prefer to drop the compatible string because it's probably
from a time where the PHY dt-bindings stated "add the PHY ID
compatible string if you know it" while the actual suggestion was
"only add it if reading the ID doesn't work for some reason"


Martin