Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <67df9824-898c-192f-7baf-a1f05936f0ab@linaro.org>
Date:   Fri, 1 Apr 2022 16:28:52 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.7.0
Subject: Re: [PATCH v2 1/2] spi: dt-bindings: qcom,spi-geni-qcom: convert to
 dtschema
Content-Language: en-US
To:     Kuldeep Singh <singh.kuldeep87k@gmail.com>
Cc:     Andy Gross <agross@kernel.org>,
        Bjorn Andersson <bjorn.andersson@linaro.org>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzk+dt@kernel.org>,
        Mark Brown <broonie@kernel.org>, linux-arm-msm@vger.kernel.org,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-spi@vger.kernel.org
References: <20220331160248.716234-1-krzysztof.kozlowski@linaro.org>
 <20220331175817.GA91341@9a2d8922b8f1>
 <ddc12aab-3cff-16a1-9ec9-a246240f9521@linaro.org>
 <20220401135224.GA40286@9a2d8922b8f1>
From:   Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
In-Reply-To: <20220401135224.GA40286@9a2d8922b8f1>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 01/04/2022 15:52, Kuldeep Singh wrote:
> On Thu, Mar 31, 2022 at 09:40:08PM +0200, Krzysztof Kozlowski wrote:
>> On 31/03/2022 19:58, Kuldeep Singh wrote:
>>> On Thu, Mar 31, 2022 at 06:02:47PM +0200, Krzysztof Kozlowski wrote:
>>>> Convert the GENI based Qualcomm Universal Peripheral (QUP) Serial
>>>> Peripheral Interface (SPI) bindings to DT Schema.
>>>>
>>>> The original bindings in TXT were not complete, so add during conversion
>>>> properties already used in DTS and/or in the driver: reg-names, dmas,
>>>> interconnects, operating points and power-domains.
>>>>
>>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
> 
> [...]
> 
>>>> +  reg:
>>>> +    maxItems: 1
>>>> +
>>>> +  reg-names:
>>>> +    const: se
>>>
>>> Why reg-names is required?
>>> Reg contain max 1 value, we can skip reg-names like other users.
>>>
>>> Also, "se" is used as clock name and using it again for reg-names?
>>> I think this is wrong and reg-names shouldn't be documented.
>>
>> reg-names are not required. If you ask why they are documented? As I
>> wrote in commit msg - bindings were not fully updated to DTSes being used.
> 
> Hi Krzysztof,
> 
> Power-domains, operating-points etc. makes sense, but reg-names?
> 
> With quick search, sm8150.dtsi is the only user of reg-names. Moreover,
> there is no reference in driver to incorporate it.
> Driver(drivers/spi/spi-geni-qcom.c) looks for "se" as clock name and not
> as reg-name. If there's one, kindly help in pointing out.
> 
> It seems you are adding reg-names because some DT is using it and no
> other reason. And with no reference even in driver, that's surely not a
> way to go ahead as you have been advising the same in past.

I don't mind dropping it.

Best regards,
Krzysztof