[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190508075542.GV14916@sirena.org.uk>
Date: Wed, 8 May 2019 16:55:42 +0900
From: Mark Brown <broonie@...nel.org>
To: Dmitry Osipenko <digetx@...il.com>
Cc: Liam Girdwood <lgirdwood@...il.com>,
Thierry Reding <thierry.reding@...il.com>,
Jonathan Hunter <jonathanh@...dia.com>,
Peter De Schrijver <pdeschrijver@...dia.com>,
linux-tegra@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH v1 1/6] regulator: core: Introduce API for
machine-specific regulators coupling
On Sun, Apr 14, 2019 at 08:59:34PM +0300, Dmitry Osipenko wrote:
> Right now regulator core supports only one type of regulators coupling,
> the "voltage max-spread" which keeps voltages of coupled regulators in a
> given range. A more sophisticated coupling may be required in practice,
> one example is the NVIDIA Tegra SoC's which besides the max-spreading
> have other restrictions that must be adhered. Introduce API that allow
> platforms to provide their own custom coupling algorithms.
This is really concerning since it's jumping straight to open coding the
algorithm in platform specific code which isn't great, especially since
that platform specific code is now going to have to handle all possible
board specific restrictions that might be found on that platform. Why
is it not possible to express the rules that exist in a more general
fashion which can be encoded in drivers? I'm not thrilled about later
patches that export core functionality for platform specific use either.
Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)
Powered by blists - more mailing lists