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: <20201228150052.2633-1-nikita.shubin@maquefel.me>
In-Reply-To: <20201228150052.2633-1-nikita.shubin@maquefel.me>
From:   Bartosz Golaszewski <bgolaszewski@baylibre.com>
Date:   Mon, 4 Jan 2021 15:15:07 +0100
Message-ID: <CAMpxmJVN8bH5KJ1-W76Ac6MOH3fy5Xo_B1ozsWkLVacJhxOHsA@mail.gmail.com>
Subject: Re: [PATCH] gpiolib: warning on gpiochip->to_irq defined
To:     Nikita Shubin <nikita.shubin@maquefel.me>
Cc:     Linus Walleij <linus.walleij@linaro.org>,
        Grygorii Strashko <grygorii.strashko@ti.com>,
        Thierry Reding <treding@nvidia.com>,
        linux-gpio <linux-gpio@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Mon, Dec 28, 2020 at 4:02 PM Nikita Shubin <nikita.shubin@maquefel.me> wrote:
>
> gpiochip->to_irq method is redefined in gpiochip_add_irqchip.
>
> A lot of gpiod driver's still define ->to_irq method, let's give
> a gentle warning that they can no longer rely on it, so they can remove
> it on ocassion.
>
> Fixes: e0d8972898139 ("gpio: Implement tighter IRQ chip integration")
> Signed-off-by: Nikita Shubin <nikita.shubin@maquefel.me>
> ---
>  drivers/gpio/gpiolib.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
> index 5ce0c14c637b..44538d1a771a 100644
> --- a/drivers/gpio/gpiolib.c
> +++ b/drivers/gpio/gpiolib.c
> @@ -1489,6 +1489,9 @@ static int gpiochip_add_irqchip(struct gpio_chip *gc,
>                 type = IRQ_TYPE_NONE;
>         }
>
> +       if (gc->to_irq)
> +               chip_err(gc, "to_irq is redefined in %s and you shouldn't rely on it\n", __func__);
> +
>         gc->to_irq = gpiochip_to_irq;
>         gc->irq.default_type = type;
>         gc->irq.lock_key = lock_key;
> --
> 2.29.2
>

I know Linus suggested using chip_err() here but since this doesn't
cause the function to fail, I'd say this should be chip_warn().

Unless Linus has any objections, please change it.

Bartosz