Re: [PATCH] dt-bindings: memory: convert Samsung Exynos DMC to dtschema

From: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
To: Rob Herring <robh+dt@kernel.org>
Cc: Lukasz Luba <lukasz.luba@arm.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	devicetree@vger.kernel.org,
	linux-arm-kernel <linux-arm-kernel@lists.infradead.org>,
	linux-samsung-soc <linux-samsung-soc@vger.kernel.org>,
	"open list:THERMAL" <linux-pm@vger.kernel.org>
Subject: Re: [PATCH] dt-bindings: memory: convert Samsung Exynos DMC to dtschema
Date: Fri, 20 Aug 2021 15:53:38 +0200	[thread overview]
Message-ID: <d3cf3485-6996-5cb9-6a8a-9b868f86bbea@canonical.com> (raw)
In-Reply-To: <CAL_Jsq+iG1t=yxsyp2whd_L4p=9udrSvZMiy-y75ppcLYK9acA@mail.gmail.com>

On 20/08/2021 15:31, Rob Herring wrote:
> On Wed, Aug 18, 2021 at 9:43 AM Krzysztof Kozlowski
> <krzysztof.kozlowski@canonical.com> wrote:
>>
>> On 18/08/2021 16:31, Rob Herring wrote:
>>> On Mon, Aug 16, 2021 at 3:32 AM Krzysztof Kozlowski
>>> <krzysztof.kozlowski@canonical.com> wrote:
>>>>
>>>> On 16/08/2021 09:53, Lukasz Luba wrote:
>>>>> Hi Krzysztof,
>>>>>
>>>>> On 8/13/21 1:54 PM, Krzysztof Kozlowski wrote:
>>>>>> Convert Samsung Exynos5422 SoC frequency and voltage scaling for
>>>>>> Dynamic Memory Controller to DT schema format using json-schema.
>>>>>>
>>>>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
>>>>>> ---
>>>>>>   .../memory-controllers/exynos5422-dmc.txt     |  84 -----------
>>>>>>   .../samsung,exynos5422-dmc.yaml               | 137 ++++++++++++++++++
>>>>>>   MAINTAINERS                                   |   2 +-
>>>>>
>>>>> I'm not an expert in this DT scripts and why it complains. Maybe it
>>>>> complains because the "samsung,exynos-ppmu" is defined in the .txt
>>>>> file... (?)
>>>>> Although, in general looks OK.
>>>>>
>>>>> Acked-by: Lukasz Luba <lukasz.luba@arm.com>
>>>>
>>>> I think the warning (triggered by DT_CHECKER_FLAGS=-m) can be ignored
>>>> because it complains about compatible in example which is not present in
>>>> the bindings. Usually it means someone wrote example not matching the
>>>> bindings (e.g. a typo in compatible) but here it is on purpose.
>>>
>>> Ultimately, it will mean the binding is undocumented (or a typo). But
>>> right now, it means the binding is undocumented with a schema. It's
>>> off by default because there's still about 80 warnings. It's turned on
>>> for the bot so we don't add more. So please don't ignore it.
>>>
>>
>> By "Ignore" I meant here that it is a false positive, so it can be
>> ignored. The warning is about compatible "samsung,exynos-ppmu" used in
>> the example. However this exynos-ppmu is not part of this bindings and
>> is documented elsewhere:
>> Documentation/devicetree/bindings/devfreq/event/exynos-ppmu.txt
> 
> Yes, I understood all that. Let me be clear. I don't want examples
> which don't have a schema. So drop the node or convert
> exynos-ppmu.txt.
> 
> First, when we do add a schema, then we likely have to go fix the
> examples. It happens a lot and is not caught with folks using
> DT_SCHEMA_FILES. Second, I don't like reporting errors that users
> should have gotten by default. We can't turn this check on by default
> until the 80 or so cases we already have in the binding examples are
> fixed.

Understood, I'll send both of these bindings.

Best regards,
Krzysztof