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
References: <CGME20181011090112eucas1p286d8c1edfc1a2a207d8a11c5ad7eb20e@eucas1p2.samsung.com>
 <20180906122436.25610-1-linus.walleij@linaro.org> <20181011090112eucas1p286d8c1edfc1a2a207d8a11c5ad7eb20e~cglSx9qcr2394623946eucas1p2y@eucas1p2.samsung.com>
In-Reply-To: <20181011090112eucas1p286d8c1edfc1a2a207d8a11c5ad7eb20e~cglSx9qcr2394623946eucas1p2y@eucas1p2.samsung.com>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Thu, 11 Oct 2018 11:29:22 +0200
Message-ID: <CACRpkdbTdmQQTg1D9jbcHGcT4kZermwVA+QEnUJGf+VYqoKRkg@mail.gmail.com>
Subject: Re: [PATCH v7] regulator: fixed: Convert to use GPIO descriptor only
To:     Marek Szyprowski <m.szyprowski@samsung.com>
Cc:     Liam Girdwood <lgirdwood@gmail.com>,
        Mark Brown <broonie@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Janusz Krzysztofik <jmkrzyszt@gmail.com>,
        Alexander Shiyan <shc_work@mail.ru>,
        Haojian Zhuang <haojian.zhuang@gmail.com>,
        Aaro Koskinen <aaro.koskinen@iki.fi>,
        Mike Rapoport <rppt@linux.vnet.ibm.com>,
        Robert Jarzmik <robert.jarzmik@free.fr>,
        Philipp Zabel <philipp.zabel@gmail.com>,
        Daniel Mack <zonque@gmail.com>,
        Marc Zyngier <marc.zyngier@arm.com>,
        jacopo <jacopo@jmondi.org>,
        Geert Uytterhoeven <geert+renesas@glider.be>,
        Russell King <rmk+kernel@armlinux.org.uk>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Oct 11, 2018 at 11:01 AM Marek Szyprowski
<m.szyprowski@samsung.com> wrote:

> I've just noticed that this patch causes regression on Samsung
> Exynos4412-based Trats2 board. Conversion to GPIO descriptor breaks
> operation when regulators used shared GPIO:  sii9234 i2c driver
> is not able to get vcc33mhl regulator (it uses shared GPIO enable
> line with vsil12 regulator).

So I guess this means that this physical GPIO line will enable the
vcc33mhl and the vsil12 regulators at the same time?

> This issue has been already pointed in case of commits:
> 37fa23dbccbd97663acc085bd79246f427e603a1
> d1dae72fab2c377ff463742eefd8ac0f9e99b7b9
> ab4d11e2c2329cf7cb7be31ff22489aae4dee5dc

A big sorry for my ignorance, I guess the information overload
on the mailing list just makes me miss the important points.
I'll try to be better, sadly I constantly fail to keep everything
in mind and constantly break things like this.

> Maybe it would be better to first solve the handling of shared enable
> GPIO in the descriptor-based interface before converting more regulators
> and stepping into this issue again?

I am trying to solve it, but I just don't have systems to reproduce all
kinds of things. It's a bit stressful since this is one of those runtime
things that is hard to test when devising a patch for systems I don't
have.

I am kind of relying on system maintainers to test things.

I was aware of the usecase "several consumers takes the same
GPIO line" (Mark told me several times...) so it was in the back of
my mind, but it's just hard to see when we were gonna run into it.
So it is the fixed regulators, on Samsung boards, I see.

Anyways. Let's try to fix it now then instead of me constantly
hitting this and breaking it. It happens because it is just unintuitive
so I share your frustration.

I don't want to generally make it possible for a gpio line to be
retrieved nonexclusively. We need the semantics to help people
do the right thing.

So I was thinking to introduce
gpiod_get_nonexclusive() to explicitly handle this case for the few
systems that use shared GPIO lines, let me see what I can do.

Yours,
Linus Walleij