[Letux-kernel] X1600 upstreaming efforts
H. Nikolaus Schaller
hns at goldelico.com
Fri Mar 28 08:06:29 CET 2025
Hi,
> Am 27.03.2025 um 22:15 schrieb Andreas Kemnade <andreas at kemnade.info>:
>
> Am Wed, 26 Mar 2025 18:15:17 +0100
> schrieb "H. Nikolaus Schaller" <hns at goldelico.com>:
>
>> Hi,
>> I hope with 6.14.0 rolled out, I now have a little time to work on more x1600 upstreaming.
>>
>> I have started with the x1600 USB PHY
>>
>> There were some glitches in formatting of old and new macros, but otherwise it is fine.
>>
>> But checkpatch complained about a missing bindings.yaml.
>>
>> The is simple to add ingenic,x1600-phy except that we now must allow/require two register values.
>>
>> So we need to make the maxItems value depend on the compatible enum.
>> This is beyond my knowledge of the YAML programming language.
>>
>> So is there a good description how we can make this conditional?
>>
> maybe:
>
> soc/mediatek/mediatek,pwrap.yaml
>
> It is about minItems depending on compatible, but that should be also
> interesting.
So something like
diff --git a/Documentation/devicetree/bindings/phy/ingenic,phy-usb.yaml b/Documentation/devicetree/bindings/phy/ingenic,phy-usb.yaml
index 30b42008db063..cc6aa69b24e48 100644
--- a/Documentation/devicetree/bindings/phy/ingenic,phy-usb.yaml
+++ b/Documentation/devicetree/bindings/phy/ingenic,phy-usb.yaml
@@ -20,11 +20,15 @@ properties:
- ingenic,jz4775-phy
- ingenic,jz4780-phy
- ingenic,x1000-phy
+ - ingenic,x1600-phy
- ingenic,x1830-phy
- ingenic,x2000-phy
reg:
maxItems: 1
+ items:
+ - description: CGU register set (@0x10000000)
+ - description: x1600 extended register set (@0x10078000)
clocks:
maxItems: 1
@@ -42,6 +46,18 @@ required:
- vcc-supply
- '#phy-cells'
+allOf:
+ - if:
+ properties:
+ compatible:
+ contains:
+ const: ingenic,x1600-phy
+ then:
+ properties:
+ reg:
+ minItems: 2
+ maxItems: 2
+
additionalProperties: false
What I am not sure is if it is allowed to have two "- description" for maxItems: 1
or if the additional description should go into the "then:" section.
Looks as if I have to remove dust from the dtbscheck things to get some insights on that.
And the yaml-mantainers will complain anyways :)
BR and thanks,
Nikolaus
More information about the Letux-kernel
mailing list