Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
MIME-Version: 1.0
References: <20220419071503.1596423-1-patrick.rudolph@9elements.com> <20220419071503.1596423-3-patrick.rudolph@9elements.com>
In-Reply-To: <20220419071503.1596423-3-patrick.rudolph@9elements.com>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Thu, 21 Apr 2022 22:53:02 +0200
Message-ID: <CACRpkdY8Xm0L-NY3nzySnDiP105BvonE-LJkSH=vmeMNFAPuFg@mail.gmail.com>
Subject: Re: [PATCH 2/2] pinctrl: Add Cypress cy8c95x0 support
To:     Patrick Rudolph <patrick.rudolph@9elements.com>
Cc:     Naresh Solanki <naresh.solanki@9elements.com>,
        linux-kernel@vger.kernel.org, linux-gpio@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi Patrick, Naresh,

thanks for your patch! This is a VERY good looking driver so I don't
think we will need a lot of iterations before it is finished. (Famous
last words, but it looks very good.)

On Tue, Apr 19, 2022 at 9:17 AM Patrick Rudolph
<patrick.rudolph@9elements.com> wrote:

> +static const struct i2c_device_id cy8c95x0_id[] = {
> +       { "cy8c9520", 20, },
> +       { "cy8c9540", 40, },
> +       { "cy8c9560", 60, },
> +       { }

So here is the ngpios for each expander.

> +       chip->tpin = chip->driver_data & CY8C95X0_GPIO_MASK;
> +       ngpios = -1;
> +       if (client->dev.of_node)
> +               of_property_read_u32(client->dev.of_node, "ngpios", &ngpios);
> +
> +       if (ngpios < 0 || ngpios > (chip->driver_data & CY8C95X0_GPIO_MASK))
> +               ngpios = chip->driver_data & CY8C95X0_GPIO_MASK;

So why should this be configurable from the device tree at all?

I understand if not all gpios are used, but we usually just expose all
GPIOs on an expander.

Yours,
Linus Walleij