Message-ID: <939ffe2e-9b03-528a-3d27-e9eac7a04ded@linaro.org>
Date:   Wed, 22 Feb 2023 09:40:08 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.8.0
Subject: Re: [PATCH v1 1/4] dt-bindings: mfd: Add DT bindings for TI TPS6594
 PMIC
Content-Language: en-US
To:     Julien Panis <jpanis@baylibre.com>, lee@kernel.org,
        robh+dt@kernel.org, krzysztof.kozlowski+dt@linaro.org,
        corbet@lwn.net
Cc:     hdegoede@redhat.com, eric.auger@redhat.com, jgg@ziepe.ca,
        razor@blackwall.org, suma.hegde@amd.com,
        stephen@networkplumber.org, arnd@arndb.de,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-doc@vger.kernel.org, eblanc@baylibre.com,
        jneanne@baylibre.com
References: <20230216114410.183489-1-jpanis@baylibre.com>
 <20230216114410.183489-2-jpanis@baylibre.com>
 <c50503f0-dce1-a3b0-2973-8a22b5ef8bc2@linaro.org>
 <4e64838c-b727-923b-b6d5-413a0681977c@baylibre.com>
 <8de5a3bb-a0b1-8d69-cf61-0c33f42c56f6@linaro.org>
 <85183c04-40e3-fd97-c4ca-06795fe99e40@baylibre.com>
 <ce5f8e9c-0e05-3391-1393-25ea8086f10c@linaro.org>
 <633753f7-2b8a-15bb-ba55-1c5a6f2eb3f1@baylibre.com>
From:   Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
In-Reply-To: <633753f7-2b8a-15bb-ba55-1c5a6f2eb3f1@baylibre.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 21/02/2023 16:18, Julien Panis wrote:
>>>> It looks the property should be only in the drivers, not in the DT.
>>> I will remove 'ti,use-crc;' property from the DT. This will be only in
>>> the driver.
>>> Do you also consider that a property such as 'ti,is-secondary-pmic;'
>>> would not be acceptable either ? From driver point of view, this
>>> primary/secondary role on SPMI bus is a 'built-in' property of the
>>> PMIC (CRC must be enabled only via primary PMIC but using the
>>> primary PMIC does not imply that CRC is necessarily used).
>> Depends, I am not sure. Are the PMICs in some kind of hierarchical
>> topology? Like one is parent of another? If not (so both are
>> parallel/equal children of SPMI bus), then some property to indicate
>> which one is the main PMIC makes sense.
> 
> There is no hierarchical topology.
> So, I will consider identifying in DT which one is the main PMIC.

Yes. Such property would be also better than the "use-crc" as it
describes the hardware, not desired Linux driver behavior.


Best regards,
Krzysztof