Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210308043607.957156-1-hsinyi@chromium.org> <20210308043607.957156-3-hsinyi@chromium.org>
 <20210308171644.GE4656@sirena.org.uk>
In-Reply-To: <20210308171644.GE4656@sirena.org.uk>
From:   Hsin-Yi Wang <hsinyi@chromium.org>
Date:   Tue, 9 Mar 2021 21:34:37 +0800
Message-ID: <CAJMQK-jZnQ1pt435Dz=xAbuU=wJcMJkSFSnTtuARTW_rGwRCFg@mail.gmail.com>
Subject: Re: [PATCH v16 2/2] i2c: core: support bus regulator controlling in adapter
To:     Mark Brown <broonie@kernel.org>
Cc:     Wolfram Sang <wsa@kernel.org>,
        Bartosz Golaszewski <bgolaszewski@baylibre.com>,
        linux-i2c@vger.kernel.org,
        Matthias Brugger <matthias.bgg@gmail.com>,
        lkml <linux-kernel@vger.kernel.org>,
        "moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE" 
        <linux-arm-kernel@lists.infradead.org>,
        "moderated list:ARM/Mediatek SoC support" 
        <linux-mediatek@lists.infradead.org>,
        Bibby Hsieh <bibby.hsieh@mediatek.com>,
        Marek Szyprowski <m.szyprowski@samsung.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Mar 9, 2021 at 1:17 AM Mark Brown <broonie@kernel.org> wrote:
>
> On Mon, Mar 08, 2021 at 12:36:07PM +0800, Hsin-Yi Wang wrote:
>
> > +     adap->bus_regulator = devm_regulator_get(&adap->dev, "bus");
> > +     if (IS_ERR(adap->bus_regulator)) {
> > +             res = PTR_ERR(adap->bus_regulator);
> > +             goto out_reg;
> > +     }
>
> Idiomatically supplies should be named as they are by the chip datasheet
> rather than just a generic name like this, and I'm guessing that systems
> that have supplies like this will often already have something
> requesting the supply (eg, it's quite common for consumer drivers to do
> this) under that name.  I can see this being a useful thing to factor
> out into the core but it seems like it'd be better to have it enabled by
> having the controllers (or devices) pass a supply name (or possibly
> requested regulator) to the core rather than by just hard coding a name
> in the core so bindings look as expected.
>

I'll move the regulator request into device instead of core in the
next version. Thanks.

> I do also wonder if it's better to put the feature on the clients rather
> than the controller, I don't think it makes much difference though.