Message-ID: <2fe41e89-9a26-e7ba-6ef6-2c9262bda43d@linaro.org>
Date:   Thu, 22 Jun 2023 10:13:43 +0200
From:   Neil Armstrong <neil.armstrong@...aro.org>
To:     tanure@...ux.com, Conor Dooley <conor.dooley@...rochip.com>
Cc:     Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
        Conor Dooley <conor@...nel.org>,
        Rob Herring <robh+dt@...nel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Conor Dooley <conor+dt@...nel.org>,
        Jerome Brunet <jbrunet@...libre.com>,
        Kevin Hilman <khilman@...libre.com>, Nick <nick@...das.com>,
        Artem <art@...das.com>, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-amlogic@...ts.infradead.org,
        linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v3 2/3] dt-bindings: serial: amlogic,meson-uart: Add
 compatible string for T7

On 22/06/2023 09:36, Lucas Tanure wrote:
> On Thu, Jun 22, 2023 at 8:12 AM Conor Dooley <conor.dooley@...rochip.com> wrote:
>>
>> On Thu, Jun 22, 2023 at 07:43:31AM +0100, Lucas Tanure wrote:
>>> On Thu, Jun 22, 2023 at 7:05 AM Krzysztof Kozlowski
>>> <krzysztof.kozlowski@...aro.org> wrote:
>>>> On 22/06/2023 07:32, Lucas Tanure wrote:
>>>>> On Wed, Jun 21, 2023 at 7:12 PM Conor Dooley <conor@...nel.org> wrote:
>>>>>> On Wed, Jun 21, 2023 at 03:53:04PM +0200, Krzysztof Kozlowski wrote:
>>>>>>> On 21/06/2023 15:32, Lucas Tanure wrote:
>>>>>>>> Amlogic T7 SoCs uses the same UART controller as S4 SoCs and G12A.
>>>>>>>> There is no need for an extra compatible line in the driver, but
>>>>>>>> add T7 compatible line for documentation.
>>>>>>>>
>>>>>>>> Signed-off-by: Lucas Tanure <tanure@...ux.com>
>>>>>>>> ---
>>>>>>>>   .../devicetree/bindings/serial/amlogic,meson-uart.yaml          | 2 ++
>>>>>>>>   1 file changed, 2 insertions(+)
>>>>>>>>
>>>>>>>> diff --git a/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml b/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
>>>>>>>> index 01ec45b3b406..860ab58d87b0 100644
>>>>>>>> --- a/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
>>>>>>>> +++ b/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
>>>>>>>> @@ -33,6 +33,7 @@ properties:
>>>>>>>>                 - amlogic,meson8b-uart
>>>>>>>>                 - amlogic,meson-gx-uart
>>>>>>>>                 - amlogic,meson-s4-uart
>>>>>>>> +              - amlogic,meson-t7-uart
>>>>>>>>             - const: amlogic,meson-ao-uart
>>>>>>>>         - description: Always-on power domain UART controller on G12A SoCs
>>>>>>>>           items:
>>>>>>>> @@ -46,6 +47,7 @@ properties:
>>>>>>>>             - amlogic,meson8b-uart
>>>>>>>>             - amlogic,meson-gx-uart
>>>>>>>>             - amlogic,meson-s4-uart
>>>>>>>> +          - amlogic,meson-t7-uart
>>>>>>>
>>>>>>> It does not look like you tested the DTS against bindings. Please run
>>>>>>> `make dtbs_check` (see
>>>>>>> Documentation/devicetree/bindings/writing-schema.rst or
>>>>>>> https://www.linaro.org/blog/tips-and-tricks-for-validating-devicetree-sources-with-the-devicetree-schema/
>>>>>>> for instructions).
>>>>>>
>>>>>> Check back on the previous version, I should've posted an untested
>>>>>> version of what you need to add.
>>>>> I saw that, but adding a S4 doesn't make sense to me. And you didn't
>>>>> show the entire change, so I can't understand what you want there.
>>>>
>>>> For sure you need something which does not trigger errors. If you claim
>>>> adding S4 as fallback does not make sense, then why did you use it?
>>>> Sending a code which is clearly incorrect does not make sense.
>>>>
>>> Sorry, I think we are talking about different things. It does not make
>>> sense to me to add an S4 line in the documentation when it is already
>>> there. So I could not understand or make sense of the patch Conor sent
>>> in reply to my V2.
>>
>> That is just how it works. You need to spell out exactly which
>> combinations are permitted. The current entry for s4 says that s4 is
>> only permitted in isolation.
>> Since you are adding "amlogic,meson-t7-uart", "amlogic,meson-s4-uart"
>> you need to explicitly allow that combination. You'll notice if you look
>> at the file that the gx uart appears more than once.
>>
>> Given the g12a was the most recently added compatible, it might make
>> sense to follow the pattern that it had set, given the thing your
>> original patch copied the match data from was the g12a. That change to
>> the dt-binding would look like:
>> diff --git a/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml b/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
>> index 01ec45b3b406..eae11e87b88a 100644
>> --- a/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
>> +++ b/Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml
>> @@ -50,6 +50,13 @@ properties:
>>           items:
>>             - const: amlogic,meson-g12a-uart
>>             - const: amlogic,meson-gx-uart
>> +      - description:
>> +          Everything-Else power domain UART controller on G12A compatible SoCs
>> +        items:
>> +          - enum:
>> +              - amlogic,meson-t7-uart
>> +          - const: amlogic,meson-g12a-uart
>> +          - const: amlogic,meson-gx-uart
>>
>>     reg:
>>       maxItems: 1
>>
>> /I/ don't really care whether you do that, or do the s4 version of it,
>> but following the most recent pattern might make more sense. When I
>> suggested s4, it was because I only looked at the driver patch rather
>> than the code itself.
>>
>>> Krzysztof, I will check again with dtbs_check and re-send.
>>
>> Cheers,
>> Conor.
> I am struggling to understand this. Everything I try fails the check.

I just applied Conor's change on top of v6.4-rc1 and ran:
make dt_binding_check DT_SCHEMA_FILES=Documentation/devicetree/bindings/serial/amlogic,meson-uart.yaml

and the check was successful.

Neil

Powered by blists - more mailing lists