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: <20191127084253.16356-1-geert+renesas@glider.be> <20191127084253.16356-3-geert+renesas@glider.be>
In-Reply-To: <20191127084253.16356-3-geert+renesas@glider.be>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Thu, 12 Dec 2019 14:20:45 +0100
Message-ID: <CACRpkdYyY0eGipdK6ixZxLtdJ5px=U2mOa79VZb00NEEAEL=6g@mail.gmail.com>
Subject: Re: [PATCH v3 2/7] gpiolib: Add support for gpiochipN-based table lookup
To:     Geert Uytterhoeven <geert+renesas@glider.be>
Cc:     Bartosz Golaszewski <bgolaszewski@baylibre.com>,
        Jonathan Corbet <corbet@lwn.net>,
        Rob Herring <robh+dt@kernel.org>,
        Mark Rutland <mark.rutland@arm.com>,
        Harish Jenny K N <harish_kandiga@mentor.com>,
        Eugeniu Rosca <erosca@de.adit-jv.com>,
        Alexander Graf <graf@amazon.com>,
        Peter Maydell <peter.maydell@linaro.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Phil Reid <preid@electromag.com.au>,
        Marc Zyngier <marc.zyngier@arm.com>,
        Christoffer Dall <christoffer.dall@arm.com>,
        Magnus Damm <magnus.damm@gmail.com>,
        "open list:GPIO SUBSYSTEM" <linux-gpio@vger.kernel.org>,
        Linux Doc Mailing List <linux-doc@vger.kernel.org>,
        "open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS" 
        <devicetree@vger.kernel.org>,
        Linux-Renesas <linux-renesas-soc@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        QEMU Developers <qemu-devel@nongnu.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Geert!

On Wed, Nov 27, 2019 at 9:43 AM Geert Uytterhoeven
<geert+renesas@glider.be> wrote:

> Currently GPIO controllers can only be referred to by label in GPIO
> lookup tables.
>
> Add support for looking them up by "gpiochipN" name, with "N" either the
> corresponding GPIO device's ID number, or the GPIO controller's first
> GPIO number.
>
> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>

What the commit message is missing is a rationale, why is this needed?

> If this is rejected, the GPIO Aggregator documentation must be updated.
>
> The second variant is currently used by the legacy sysfs interface only,
> so perhaps the chip->base check should be dropped?

Anything improving the sysfs is actively discouraged by me.
If it is just about staying compatible it is another thing.

> +static int gpiochip_match_id(struct gpio_chip *chip, void *data)
> +{
> +       int id = (uintptr_t)data;
> +
> +       return id == chip->base || id == chip->gpiodev->id;
> +}
>  static struct gpio_chip *find_chip_by_name(const char *name)
>  {
> -       return gpiochip_find((void *)name, gpiochip_match_name);
> +       struct gpio_chip *chip;
> +       int id;
> +
> +       chip = gpiochip_find((void *)name, gpiochip_match_name);
> +       if (chip)
> +               return chip;
> +
> +       if (!str_has_prefix(name, GPIOCHIP_NAME))
> +               return NULL;
> +
> +       if (kstrtoint(name + strlen(GPIOCHIP_NAME), 10, &id))
> +               return NULL;
> +
> +       return gpiochip_find((void *)(uintptr_t)id, gpiochip_match_id);

Isn't it easier to just  augment the existing match function to
check like this:

static int gpiochip_match_name(struct gpio_chip *chip, void *data)
{
        const char *name = data;

        if (!strcmp(chip->label, name))
               return 0;
        return !strcmp(dev_name(&chip->gpiodev->dev), name);
}

We should I guess also add some kerneldoc to say we first
match on the label and second on dev_name().

Yours,
Linus Walleij