Re: [PATCH 08/10] ARM: dts: NSP: Add a default compatible for switch node

[Florian Fainelli]

On 11/10/20 2:48 PM, Vladimir Oltean wrote:
> On Tue, Nov 10, 2020 at 02:40:43PM -0800, Florian Fainelli wrote:
>> On 11/10/20 2:37 PM, Vladimir Oltean wrote:
>>> On Mon, Nov 09, 2020 at 07:31:11PM -0800, Florian Fainelli wrote:
>>>> Provide a default compatible string which is based on the 58522 SRAB
>>>> compatible, this allows us to have sane defaults and silences the
>>>> following warnings:
>>>>
>>>>  arch/arm/boot/dts/bcm958522er.dt.yaml:
>>>>     ethernet-switch@36000: compatible: 'oneOf' conditional failed,
>>>> one
>>>>     must be fixed:
>>>>             ['brcm,bcm5301x-srab'] is too short
>>>>             'brcm,bcm5325' was expected
>>>>             'brcm,bcm53115' was expected
>>>>             'brcm,bcm53125' was expected
>>>>             'brcm,bcm53128' was expected
>>>>             'brcm,bcm5365' was expected
>>>>             'brcm,bcm5395' was expected
>>>>             'brcm,bcm5389' was expected
>>>>             'brcm,bcm5397' was expected
>>>>             'brcm,bcm5398' was expected
>>>>             'brcm,bcm11360-srab' was expected
>>>>             'brcm,bcm5301x-srab' is not one of ['brcm,bcm53010-srab',
>>>>     'brcm,bcm53011-srab', 'brcm,bcm53012-srab', 'brcm,bcm53018-srab',
>>>>     'brcm,bcm53019-srab']
>>>>             'brcm,bcm5301x-srab' is not one of ['brcm,bcm11404-srab',
>>>>     'brcm,bcm11407-srab', 'brcm,bcm11409-srab', 'brcm,bcm58310-srab',
>>>>     'brcm,bcm58311-srab', 'brcm,bcm58313-srab']
>>>>             'brcm,bcm5301x-srab' is not one of ['brcm,bcm58522-srab',
>>>>     'brcm,bcm58523-srab', 'brcm,bcm58525-srab', 'brcm,bcm58622-srab',
>>>>     'brcm,bcm58623-srab', 'brcm,bcm58625-srab', 'brcm,bcm88312-srab']
>>>>             'brcm,bcm5301x-srab' is not one of ['brcm,bcm3384-switch',
>>>>     'brcm,bcm6328-switch', 'brcm,bcm6368-switch']
>>>>             From schema:
>>>>     Documentation/devicetree/bindings/net/dsa/b53.yaml
>>>>
>>>> Signed-off-by: Florian Fainelli <f.fainelli@xxxxxxxxx>
>>>> ---
>>>>  arch/arm/boot/dts/bcm-nsp.dtsi | 2 +-
>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/arch/arm/boot/dts/bcm-nsp.dtsi b/arch/arm/boot/dts/bcm-nsp.dtsi
>>>> index 09fd7e55c069..8453865d1439 100644
>>>> --- a/arch/arm/boot/dts/bcm-nsp.dtsi
>>>> +++ b/arch/arm/boot/dts/bcm-nsp.dtsi
>>>> @@ -386,7 +386,7 @@ ccbtimer1: timer@35000 {
>>>>  		};
>>>>
>>>>  		srab: ethernet-switch@36000 {
>>>> -			compatible = "brcm,nsp-srab";
>>>> +			compatible = "brcm,bcm58522-srab", "brcm,nsp-srab";
>>>>  			reg = <0x36000 0x1000>,
>>>>  			      <0x3f308 0x8>,
>>>>  			      <0x3f410 0xc>;
>>>> --
>>>> 2.25.1
>>>>
>>>
>>> I am not getting this.
>>> The line:
>>> #include "bcm-nsp.dtsi"
>>>
>>> can be found in:
>>>
>>> arch/arm/boot/dts/bcm988312hr.dts
>>> arch/arm/boot/dts/bcm958625hr.dts
>>> arch/arm/boot/dts/bcm958622hr.dts
>>> arch/arm/boot/dts/bcm958625k.dts
>>> arch/arm/boot/dts/bcm958522er.dts
>>> arch/arm/boot/dts/bcm958525er.dts
>>> arch/arm/boot/dts/bcm958623hr.dts
>>> arch/arm/boot/dts/bcm958525xmc.dts
>>>
>>>
>>> The pattern for the other DTS files that include this seems to be to
>>> overwrite the compatible locally in bcm958522er.dts, like this:
>>>
>>> &srab {
>>> 	compatible = "brcm,bcm58522-srab", "brcm,nsp-srab";
>>> };
>>>
>>> Is there a reason why you are choosing to put an SoC specific compatible
>>> in the common bcm-nsp.dtsi?
>>
>> It is necessary to silence the warnings provided in the commit message
>> even when the srab node is disabled, since the dt_binding_check rule
>> will check all of the nodes matching the pattern. If there is a better
>> way to do this, I would gladly do it differently.
>> -- 
>> Florian
> 
> I am still not getting it. The exact 3 lines from above will not change
> the "status" property from "disabled" to "okay", so I don't understand
> why it matters whether it's enabled or not. The dt_binding_check error
> isn't in the DTSI, it's in bcm958522er.dts. All that needs to be done is
> that the bcm958522er.dts needs to override the compatible from the DTSI
> and only the compatible, I believe. With no occurrence of an incomplete
> list of compatibles in any final DTS, the dt_binding_check should not
> complain about that single occurrence in the DTSI as far as I know (and
> I did not test this).

There is not a switch being enabled in
arch/arm/boot/dts/bcm958522er.dts, so sure, I could add the 3 lines you
quote above and that would silence the warning, but that does not scale
at all across DTS files including bcm5301x.dtsi for instance, it sort of
does for those including bcm-nsp.dtsi.
-- 
Florian