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: <20210516061425.8757-1-aardelean@deviqon.com>
In-Reply-To: <20210516061425.8757-1-aardelean@deviqon.com>
From:   Bartosz Golaszewski <bgolaszewski@baylibre.com>
Date:   Fri, 21 May 2021 14:42:15 +0200
Message-ID: <CAMpxmJVeZ2zASLN-TqasWHDvLbWN1K=4ZFTzw36_KTs6N7S-5g@mail.gmail.com>
Subject: Re: [PATCH] gpio: gpio-stmpe: fully use convert probe to device-managed
To:     Alexandru Ardelean <aardelean@deviqon.com>
Cc:     linux-gpio <linux-gpio@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Linus Walleij <linus.walleij@linaro.org>,
        alexandre.torgue@foss.st.com,
        Maxime Coquelin <mcoquelin.stm32@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sun, May 16, 2021 at 8:14 AM Alexandru Ardelean
<aardelean@deviqon.com> wrote:
>
> The driver doesn't look like it can be built as a kmod, so leaks cannot
> happen via a rmmod mechanism.
> The remove hook was removed via commit 3b52bb960ec6 ("gpio: stmpe: make
> it explicitly non-modular").
>
> The IRQ is registered via devm_request_threaded_irq(), making the driver
> only partially device-managed.
>
> In any case all resources should be made device-managed, mostly as a good
> practice. That way at least the unwinding on error is happening in reverse
> order (as the probe).
>
> This change also removes platform_set_drvdata() since the information is
> never retrieved to be used in the driver.
>
> Signed-off-by: Alexandru Ardelean <aardelean@deviqon.com>

I applied the patch and tweaked the commit message because as Andy
pointed out - this driver can be unbound over sysfs.

Bart