Re: fwnode_for_each_child_node() and OF backend discrepancy

From: Michael Walle <michael@walle.cc>
To: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>,
	Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	linux-acpi@vger.kernel.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: fwnode_for_each_child_node() and OF backend discrepancy
Date: Tue, 28 Jun 2022 13:36:44 +0200	[thread overview]
Message-ID: <f17d3ecfecf4491dd15b1fa092205f3f@walle.cc> (raw)
In-Reply-To: <Yrrhs3D++V79/4Jk@smile.fi.intel.com>

Am 2022-06-28 13:10, schrieb Andy Shevchenko:
> On Mon, Jun 27, 2022 at 02:49:51PM +0200, Michael Walle wrote:
>> Hi,
>> 
>> I tired to iterate over all child nodes, regardless if they are 
>> available
>> or not. Now there is that handy fwnode_for_each_child_node() (and the
>> fwnode_for_each_available_child_node()). The only thing is the OF 
>> backend
>> already skips disabled nodes [1], making fwnode_for_each_child_node() 
>> and
>> fwnode_for_each_available_child_node() behave the same with the OF 
>> backend.
>> 
>> Doesn't seem to be noticed by anyone for now. I'm not sure how to fix 
>> that
>> one. fwnode_for_each_child_node() and also 
>> fwnode_get_next_child_node() are
>> used by a handful of drivers. I've looked at some, but couldn't decide
>> whether they really want to iterate over all child nodes or just the 
>> enabled
>> ones.
>> 
>> Any thoughts?
> 
> It was discussed at least twice this year (in regard to some new IIO 
> drivers)
> and Rob told that iterating over disabled (not available) nodes in OF 
> kinda
> legacy/design mistake. That's why device_for_each_child_node() goes 
> only
> over available nodes only.

Mh, but then the fwnode_for_each_child_node() is very misleading, esp.
with the presence of fwnode_for_each_available_child_node().

> So, why do you need to iterate over disabled ones?

I was trying to fix the lan966x driver [1] which doesn't work if there
are disabled nodes in between. My steps would have been:
  (1) change fwnode_for_each_available_child_node() to
      fwnode_for_each_child_node(), maybe with a fixes tag, as it's
      easy to backport
  (2) introduce new compatibles and deduce the number of ports
      according to the compatible string and not by counting
      the child nodes.
  (3) keep the old behavior for the legacy compatible and mark it
      as deprecated in the binding
  (4) move the device tree over to the new compatible string

-michael

[1] 
https://elixir.bootlin.com/linux/v5.19-rc4/source/drivers/net/ethernet/microchip/lan966x/lan966x_main.c#L1029