[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20160525181603.GD8206@sirena.org.uk>
Date: Wed, 25 May 2016 19:16:03 +0100
From: Mark Brown <broonie@...nel.org>
To: Rob Herring <robh@...nel.org>
Cc: Nicolin Chen <nicoleotsuka@...il.com>, brian.austin@...rus.com,
Paul.Handrigan@...rus.com, pawel.moll@....com,
mark.rutland@....com, ijc+devicetree@...lion.org.uk,
galak@...eaurora.org, lgirdwood@...il.com, Tim.Howe@...rus.com,
somduttar@...dia.com, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org, alsa-devel@...a-project.org
Subject: Re: [PATCH v6] ASoC: cs53l30: Add codec driver support for Cirrus
CS53L30
On Wed, May 25, 2016 at 12:28:00PM -0500, Rob Herring wrote:
> On Mon, May 23, 2016 at 04:13:57PM -0700, Nicolin Chen wrote:
> > + - VA-supply, VP-supply : power supplies for the device,
> > + as covered in Documentation/devicetree/bindings/regulator/regulator.txt.
> Also, these are missing from the example yet are required. I'd argue
> they should be optional as you could have designs with no regulator
> control.
The general rule is that regulators should only be omitted if they are
physically absent, we can describe simple fixed supplies with no control
so we should do so (though in practice Linux is much more forgiving and
will provide one if the supply is just plain missing from the DT). This
is a simpler rule for people to follow than making them officially
optional and is more robust in the face of things like implementation
changes that allow supplies to become controlable or drivers that
suddenly form a desire to look at their supply voltages, as well as
allowing other OSs to have simpler implementations if it's followed.
Download attachment "signature.asc" of type "application/pgp-signature" (474 bytes)
Powered by blists - more mailing lists