Re: [PATCH v3 0/2] Add support for LT3074 low voltage linear regulator

[Guenter Roeck <linux@roeck-us.net>]

On 4/22/25 03:09, Encarnacion, Cedric justine wrote:
>> From: Guenter Roeck <groeck7@gmail.com> On Behalf Of Guenter Roeck
>> Sent: Monday, April 21, 2025 9:23 PM
>> To: Encarnacion, Cedric justine <Cedricjustine.Encarnacion@analog.com>; Rob
>> Herring <robh@kernel.org>; Krzysztof Kozlowski <krzk+dt@kernel.org>; Conor
>> Dooley <conor+dt@kernel.org>; Jean Delvare <jdelvare@suse.com>; Jonathan
>> Corbet <corbet@lwn.net>; Delphine CC Chiu <Delphine_CC_Chiu@Wiwynn.com>
>> Cc: devicetree@vger.kernel.org; linux-kernel@vger.kernel.org; linux-
>> hwmon@vger.kernel.org; linux-doc@vger.kernel.org; linux-i2c@vger.kernel.org
>> Subject: Re: [PATCH v3 0/2] Add support for LT3074 low voltage linear regulator
>>
>> [External]
>>
>> On 4/21/25 05:18, Cedric Encarnacion wrote:
>>> Introduce hardware monitoring and regulator support for LT3074. The
>>> component is an ultrafast, ultralow noise 3A, 5.5V dropout linear
>>> regulator with a PMBus serial interface that allows telemetry for
>>> input/output voltage, output current, and die temperature. It has a
>>> single channel and requires a bias voltage which can be monitored via
>>> manufacturer-specific registers.
>>>
>>> Signed-off-by: Cedric Encarnacion
>>> <cedricjustine.encarnacion@analog.com>
>>> ---
>>> Changes in v3:
>>> -
>>>    * Non-nested dt-binding regulator support.
>>>    * PMBUS_REGULATOR_ONE("vout") ->
>> PMBUS_REGULATOR_ONE("regulator").
>>>      New macro needs node-name to be matched.
>>>
>>
>> Does that mean we can (and maybe should) drop the name parameter of that
>> macro ?
>>
>> This isn't affecting this patch. If needed I'll do that separately.
>> Just asking.
> 
> This is used by regulator_desc->of_match to match a node name in dt. By
> dropping the name, does this mean the of_match won't be assigned, or
> of_match="some-default-regulator-name"? I did some tests, and it appears
> the regulators dt properties can only be parsed when the name parameter
> exists, and name parameter == node-name. I have tried simply removing the
> of_match assignment from the regulator_desc macro, but it does not work in
> my case.
> 

No, I was wondering if the name parameter always needs to be "regulator".
I did not (want to) suggest that the name should be dropped from the structure,
only if it always needs to be "regulator". If it always needs to be "regulator",
it could be hard-coded and would not have to be a macro parameter.

Sorry for the confusion. Just forget that I asked.

Guenter