Aw: Re: [PATCH 4/4] phy: Realtek Otto Serdes: add devicetree documentation

[Krzysztof Kozlowski]

On 05/10/2024 12:28, Markus Stockhausen wrote:
>> ...
>>> + The primary serdes register memory location. Other SerDes control and
>>> + management registers are distributed all over the I/O memory space and
>>> + identified by the driver automatically.
>>> +
>>> + controlled-ports:
>>> + description: |
>>> + A bit mask defining the ports that are actively controlled by the driver. In
>>> + case a bit is not set the driver will only process read operations on the
>>> + SerDes. If not set the driver will run all ports in read only mode.
>>
>> You have never tested it.
> 
> Maybe I left the wrong impression. I'm currently very hard trying to get this driver
> working on 4 different SoCs. The only environment I can do this porperly is an OpenWrt
> toolchain and a lot of patches. With other bits in place I can even start a Zyxel
> GS1920 (RTL839x) through serial console, setup the SerDes and activate the attached
> Realtek PHYs.
> 
> https://github.com/openwrt/openwrt/pull/16577
> https://github.com/openwrt/openwrt/pull/16123

I am talking about binding. You never tested the binding.

Please run `make dt_binding_check` (see
Documentation/devicetree/bindings/writing-schema.rst for instructions).
Maybe you need to update your dtschema and yamllint.

> 
> An older DT from last weekend:
> 
> https://github.com/openwrt/openwrt/pull/16123/commits/716b01afd35fcc708c0204ca237d8541a8000ffa
> 
>>> +
>>> + "#phy-cells":
>>> + const: 4
>>> + description: |
>>> + The first number defines the SerDes to use. The second number a linked
>>> + SerDes. E.g. if a octa 1G PHY is attached to two QSGMII SerDes. The third
>>> + number is the first switch port this SerDes is working for, the fourth number
>>> + is the last switch port the SerDes is working for.
>>> +
>>> + cmd-setup:
>>> + description: |
>>> + A field of 16 bit values that contain a patch/command sequence to run on the
>>> + SerDes registers during driver setup.
>>
>> None of these fields match DT. Drop all of driver related stuff.
> 
> Hm, what would be the best way to run magic vendor patch sequences that setup
> registers and do not pollute driver?

These belong to driver. What do you mean "pollute"? The driver is the
place for these, not DT!

Best regards,
Krzysztof