Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Thu, 22 Sep 2022 11:19:08 +0100
From:   Lee Jones <lee.jones@linaro.org>
To:     Chunyan Zhang <zhang.lyra@gmail.com>
Cc:     Mark Brown <broonie@kernel.org>,
        Liam Girdwood <lgirdwood@gmail.com>,
        Rob Herring <robh+dt@kernel.org>, devicetree@vger.kernel.org,
        Baolin Wang <baolin.wang7@gmail.com>,
        Orson Zhai <orsonzhai@gmail.com>,
        Chunyan Zhang <chunyan.zhang@unisoc.com>,
        LKML <linux-kernel@vger.kernel.org>, Lee Jones <lee@kernel.org>
Subject: Re: [PATCH v2 2/2] dt-bindings: regulator: Add bindings for Unisoc's
 SC2730 regulator
Message-ID: <Yyw2nAAjN6NxmS09@google.com>
References: <20211008031953.339461-1-zhang.lyra@gmail.com>
 <20211008031953.339461-3-zhang.lyra@gmail.com>
 <YY5wPh0rwRvFjSRG@sirena.org.uk>
 <CAAfSe-uA3iowafC25zRqoTSaub1PbOzUvQgukLm=szEge_abvw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAAfSe-uA3iowafC25zRqoTSaub1PbOzUvQgukLm=szEge_abvw@mail.gmail.com>
Precedence: bulk

On Thu, 22 Sep 2022, Chunyan Zhang wrote:

> Hi Mark,
> 
> Sorry for the late response.
> [1] is the v1 on which we had some discussion. I hope that can help
> recall the issue below.
> 
> On Fri, 12 Nov 2021 at 21:46, Mark Brown <broonie@kernel.org> wrote:
> >
> > On Fri, Oct 08, 2021 at 11:19:53AM +0800, Chunyan Zhang wrote:
> >
> > > +properties:
> > > +  compatible:
> > > +    const: sprd,sc2730-regulator
> >
> > I still don't understand why this MFD subfunction for a specific device
> > is a separate binding with a separate compatible string, the issues I
> > mentioned previously with this just encoding current Linux internals
> > into the DT rather than describing the device still apply.
> 
> I understand your point. But like I described previously [1], if we
> still use the current solution (i.e. use devm_of_platform_populate()
> to register MFD subdevices), a compatible string is required. I'm open
> to switching to other solutions, do you have some suggestions?

Many IPs encompassing multiple functions are described that way in
DT.  I don't have the details for *this* device to hand, so my
comments here aren't specific to this use-case, but describing each
function individually does describe the H/W accurately, which is all
DT calls for.

Can you imagine describing an SoC, which can be considered as a huge
MFD, with only a single node?

Does the regulator functionality have it's own bank of registers?

-- 
DEPRECATED: Please use lee@kernel.org