Should regulator core support parsing OF based fwnode? (was: Re: [PATCH v8 2/5] leds: Add of_led_get() and led_put())

From: Jacek Anaszewski <jacek.anaszewski@gmail.com>
To: Mark Brown <broonie@kernel.org>
Cc: Jean-Jacques Hiblot <jjhiblot@ti.com>,
	Sebastian Reichel <sebastian.reichel@collabora.com>,
	pavel@ucw.cz, robh+dt@kernel.org, mark.rutland@arm.com,
	lee.jones@linaro.org, daniel.thompson@linaro.org,
	linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
	tomi.valkeinen@ti.com, dmurphy@ti.com,
	linux-leds@vger.kernel.org, Liam Girdwood <lgirdwood@gmail.com>
Subject: Should regulator core support parsing OF based fwnode? (was: Re: [PATCH v8 2/5] leds: Add of_led_get() and led_put())
Date: Thu, 3 Oct 2019 22:27:26 +0200	[thread overview]
Message-ID: <a9f668f9-ad26-4e18-178a-8403b8b3b1db@gmail.com> (raw)
In-Reply-To: <20191003194140.GE6090@sirena.co.uk>

On 10/3/19 9:41 PM, Mark Brown wrote:
> On Thu, Oct 03, 2019 at 09:21:06PM +0200, Jacek Anaszewski wrote:
>> On 10/3/19 8:35 PM, Mark Brown wrote:
>>> On Thu, Oct 03, 2019 at 07:43:17PM +0200, Jacek Anaszewski wrote:
>>>> On 10/3/19 2:47 PM, Jean-Jacques Hiblot wrote:
>>>>> On 03/10/2019 12:42, Sebastian Reichel wrote:
>>>>>> On Thu, Oct 03, 2019 at 10:28:09AM +0200, Jean-Jacques Hiblot wrote:
> 
>>> This mail has nothing relevant in the subject line and pages of quotes
>>> before the question for me, it's kind of lucky I noticed it....
> 
>> Isn't it all about creating proper filters?
> 
> My point there is that there's nothing obvious in the mail that suggests
> it should get past filters - just being CCed on a mail isn't super
> reliable, people often get pulled in due to things like checkpatch or
> someone copying a CC list from an earlier patch series where there were
> things were relevant.

OK, updated the subject.

>>>> I wonder if it wouldn't make sense to add support for fwnode
>>>> parsing to regulator core. Or maybe it is either somehow supported
>>>> or not supported on purpose?
> 
>>> Anything attempting to use the regulator DT bindings in ACPI has very
>>> serious problems, ACPI has its own power model which isn't compatible
>>> with that used in DT.
> 
>> We have a means for checking if fwnode refers to of_node:
> 
>> is_of_node(const struct fwnode_handle *fwnode)
> 
>> Couldn't it be employed for OF case?
> 
> Why would we want to do that?  We'd continue to support only DT systems,
> just with code that's less obviously DT only and would need to put
> checks in.  I'm not seeing an upside here.

For instance few weeks ago we had a patch [0] in the LED core switching
from using struct device's of_node property to fwnode for conveying
device property data. And this transition to fwnode property API can be
observed as a frequent pattern across subsystems.

Recently there is an ongoing effort aiming to add generic support for
handling regulators in the LED core [1], but it turns out to require
bringing back initialization of of_node property for
devm_regulator_get_optional() to work properly.

Support for OF related fwnodes in regulator core could help reducing
this noise.

[0]
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/drivers/leds/led-class.c?id=fd81d7e946c6bdb86dbf0bd88fee3e1a545e7979
[1]
https://lore.kernel.org/linux-leds/20190923102059.17818-4-jjhiblot@ti.com/

-- 
Best regards,
Jacek Anaszewski