[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9fc72d19-1827-7681-4eb7-ba7c67f325fa@kernel.org>
Date: Fri, 25 Mar 2022 08:36:57 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Johan Jonker <jbx6244@...il.com>, heiko@...ech.de
Cc: robh+dt@...nel.org, krzk+dt@...nel.org, mturquette@...libre.com,
sboyd@...nel.org, linux-clk@...r.kernel.org,
devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-rockchip@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1] dt-bindings: clock: convert rockchip,rk3188-cru.txt to
YAML
On 24/03/2022 20:51, Johan Jonker wrote:
> Hi Heiko, Krzysztof,
>
> Question for the Rockchip clock maintainer:
> What clock should be used here and other SoCs with several clock parents
> in the tree?
>
> The clock.yaml produces a lot off notifications like:
>
> /arch/arm/boot/dts/rk3036-evb.dtb: clock-controller@...00000: 'clocks'
> is a dependency of 'assigned-clocks'
>
> Same for power-controller:
>
> /arch/arm/boot/dts/rk3288-evb-rk808.dtb: power-controller: 'clocks' is a
> dependency of 'assigned-clocks'
>
> Just followed rk3399.yaml that got approved recent.
> The current CRU nodes are in use for a long time this way.
> Converting a simple .txt comes with a lot of extra work each time. :\
> Please advise.
Same for me, same for me...
I replied to this part to Stephen's email. Here I'll reply to other parts:
(...)
>>> +
>>> +properties:
>>> + compatible:
>>> + enum:
>>> + - rockchip,rk3066a-cru
>>> + - rockchip,rk3188-cru
>>> + - rockchip,rk3188a-cru
>>> +
>>> + reg:
>>> + maxItems: 1
>>> +
>>> + "#clock-cells":
>>> + const: 1
>>> +
>>> + "#reset-cells":
>>> + const: 1
>>> +
>>> + clocks:
>>> + minItems: 1
>>
>
>> You need maxItems. Would be also nice to describe what should be here as
>> input
>
> Just followed rockchip,rk3399-cru.yaml that got rob+dt's ack.
>
> Is this dependency only for simple clock consumers suitable for complex
> clock trees with pll's, mux, etc? Shouldn't that be relaxed a bit?
minItems means that your clock controller can take any number of other
clock inputs. This is not true, for sure. Your device probably has fixed
number of external clocks (usually provided on the board if this is main
SoC clock controller).
>
>>
>>
>>> +
>>> + assigned-clock-rates:
>>> + minItems: 1
>>> + maxItems: 64
>>> +
>>> + assigned-clocks:
>>> + minItems: 1
>>> + maxItems: 64
>>
>> Both assigned-xxx should not be necessary. Did you try validate the dtbs
>> without these?
>
> That's how it's currently done for Rockchip cru's.
I understand, but if you remove assigned-xxx from schema what happens?
>
>
> ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- make dtbs_check
> DT_SCHEMA_FILES=Documentation/devicetree/bindings/clock/rockchip,rk3188-cru.yaml
There is no such file in my tree...
>
>>
>> I guess you added "clocks" above because of these, so you need to
>> correct the DTS because asigned-xxx depend on clocks property.
>
> See comment above.
>
>>
>>> +
>>> + rockchip,grf:
>>> + $ref: /schemas/types.yaml#/definitions/phandle
>>> + description:
>>> + Phandle to the syscon managing the "general register files" (GRF),
>>> + if missing pll rates are not changeable, due to the missing pll lock status.
>>> +
>>> +required:
>>> + - compatible
>>> + - reg
>>> + - "#clock-cells"
>>> + - "#reset-cells"
>>> +
>>> +additionalProperties: false
>>> +
>>> +examples:
>>> + - |
>>> + cru: cru@...00000 {
>>
>> Node name: clock-controller
>
> Which schema?
Devicetree specification (see chapter for node naming). Nodes should
have generic name. There is also list of example names to use.
Best regards,
Krzysztof
Powered by blists - more mailing lists