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;
MIME-Version: 1.0
References: <cover.1686063941.git.oleksii_moisieiev@epam.com>
 <43109a0f2f362222fca79e2afd15c46ed9a32977.1686063941.git.oleksii_moisieiev@epam.com>
 <ZIAxLdexyKBnMOmU@surfacebook> <87sf9ihfc7.fsf@epam.com>
In-Reply-To: <87sf9ihfc7.fsf@epam.com>
From:   Andy Shevchenko <andy.shevchenko@gmail.com>
Date:   Thu, 20 Jul 2023 19:11:25 +0300
Message-ID: <CAHp75Vf+H_wnhT=2w=A9M7wFeOkf_m1M1gmL9vd8WHNid7+YBg@mail.gmail.com>
Subject: Re: [PATCH v3 3/4] pinctrl: Implementation of the generic
 scmi-pinctrl driver
To:     Oleksii Moisieiev <Oleksii_Moisieiev@epam.com>
Cc:     "sudeep.holla@arm.com" <sudeep.holla@arm.com>,
        Cristian Marussi <cristian.marussi@arm.com>,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Conor Dooley <conor+dt@kernel.org>,
        Linus Walleij <linus.walleij@linaro.org>,
        "linux-arm-kernel@lists.infradead.org" 
        <linux-arm-kernel@lists.infradead.org>,
        "devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-gpio@vger.kernel.org" <linux-gpio@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Thu, Jul 20, 2023 at 4:40=E2=80=AFPM Oleksii Moisieiev
<Oleksii_Moisieiev@epam.com> wrote:
> andy.shevchenko@gmail.com writes:
> > Tue, Jun 06, 2023 at 04:22:28PM +0000, Oleksii Moisieiev kirjoitti:

...

> >> +    devm_kfree(pmx->dev, pmx->functions[selector].groups);
> >
> > Red Flag. Please, elaborate.
>
> Thank you for the review.
> I did some research regarding this and now I'm confused. Could you
> please explain to me why it's a red flag?
> IIUC devm_alloc/free functions are the calls to the resource-managed
> alloc/free command, which is bound to the device.
> pinctrl-scmi driver does devm_pinctrl_register_and_init which does
> devres_alloc and doesn't open devres_group like
> scmi_alloc_init_protocol_instance (thanks to Cristian detailed
> explanation).
>
> As was mentioned in Documentation/driver-api/driver-model/devres.rst:
>
> ```
> No matter what, all devres entries are released on driver detach.  On
> release, the associated release function is invoked and then the
> devres entry is freed.
> ```

Precisely. So, why do you intervene in this?

> Also there is devm_pinctrl_get call listed in the managed interfaces.
>
> My understanding is that all resources, bound to the particular device
> will be freed on driver detach.
>
> Also I found some examples of using devm_alloc/free like from dt_node_to_=
map
> call in pinctrl-simple.c driver.
>
> I agree that I need to implement .remove callback with proper cleanup,
> but why can't I use devm_* here?

You can use devm_*(), but what's the point if you call release
yourself? That's quite a red flag usually shows a bigger issue
(misunderstanding of the objects lifetimes and their interaction).

> Maybe I've misunderstood your point.

--=20
With Best Regards,
Andy Shevchenko