Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
In-Reply-To: <20180815111318.GY2414@sirena.org.uk>
References: <20180814170617.100087-1-dianders@chromium.org>
 <20180814170617.100087-2-dianders@chromium.org> <bf70f01d-41ac-dcdc-b503-5e0eb1e70329@codeaurora.org>
 <CAD=FV=VNCOyyq5+xvSudTmtBgPm4-s7gj7GRCvuL5vSmA6yDVA@mail.gmail.com>
 <aa2bd1f6-74b4-a63b-d115-88d6e16ed95d@codeaurora.org> <CAD=FV=WBw2SKgPq8Atb+SkziUku8p7N4GWemHYZZvoMSE0bK9g@mail.gmail.com>
 <20180815111318.GY2414@sirena.org.uk>
From:   Doug Anderson <dianders@chromium.org>
Date:   Thu, 16 Aug 2018 13:07:47 -0700
Message-ID: <CAD=FV=VT6b1SmboyuBSGCnm17HTxxvAL8Hec23Dmc6Yfk5jYOg@mail.gmail.com>
Subject: Re: [PATCH 1/4] regulator: core: If consumers don't call
 regulator_set_load() assume max
To:     Mark Brown <broonie@kernel.org>
Cc:     David Collins <collinsd@codeaurora.org>,
        linux-arm-msm <linux-arm-msm@vger.kernel.org>,
        Bjorn Andersson <bjorn.andersson@linaro.org>,
        Stephen Boyd <swboyd@chromium.org>,
        Liam Girdwood <lgirdwood@gmail.com>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi,

On Wed, Aug 15, 2018 at 4:13 AM, Mark Brown <broonie@kernel.org> wrote:
> On Tue, Aug 14, 2018 at 04:56:42PM -0700, Doug Anderson wrote:
>
>> IMO about the best we could hope to do would be to map "mode" from
>> children to parent.  AKA: perhaps you could assume that if a child is
>> in a higher power mode that perhaps a parent should be too?
>
> That's not going to work well - different regulators have wildly
> different abilities to deliver current which is the whole reason why
> modes are so fuzzy and hard to use in the first place.  A high power
> load for a low noise regulator designed to feed analogue circuits might
> not even make it out of the lowest power LDO mode of a DCDC designed to
> supply the main application processors in the system or (more
> relevantly) provide the main step down for a bunch of LDOs.

OK, fair enough.  I'll drop this patch and rebase the later patches in
the series without it since I think they're still useful.

I'll work on either adding more regulator_set_load() calls to clients
or perhaps disabling the "regulator-allow-set-load" for a bunch of
rails.  David: presumably if we have a rail that we never need to be
on-and-in-low-power-mode can just be left in high power mode all the
time?  There should be no advantage of being in low power mode for a
regulator that is off, right?


-Doug