Received-SPF: pass (google.com: domain of linux-kernel+bounces-169770-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240505214754.891700-1-andreas@kemnade.info> <20240505214754.891700-4-andreas@kemnade.info>
In-Reply-To: <20240505214754.891700-4-andreas@kemnade.info>
From: Andy Shevchenko <andy.shevchenko@gmail.com>
Date: Mon, 6 May 2024 15:05:52 +0300
Message-ID: <CAHp75Vdnwrxw96prr9hyLdZ2u6t1uNcj6pyxCp52UoVOpatTpg@mail.gmail.com>
Subject: Re: [PATCH v2 3/3] Input: ektf2127 - add ektf2232 support
To: Andreas Kemnade <andreas@kemnade.info>
Cc: dmitry.torokhov@gmail.com, robh@kernel.org, krzk+dt@kernel.org, 
	conor+dt@kernel.org, u.kleine-koenig@pengutronix.de, hdegoede@redhat.com, 
	siebren.vroegindeweij@hotmail.com, linux-input@vger.kernel.org, 
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Mon, May 6, 2024 at 12:48=E2=80=AFAM Andreas Kemnade <andreas@kemnade.in=
fo> wrote:
>
> The chip is similar, but has status bits at different positions,
> so use the correct bits.

..

> @@ -46,6 +47,11 @@ struct ektf2127_ts {
>         struct input_dev *input;
>         struct gpio_desc *power_gpios;
>         struct touchscreen_properties prop;
> +       int status_shift;
> +};
> +
> +struct ektf2127_i2c_chip_data {
> +       int status_shift;
>  };
>
>  static void ektf2127_parse_coordinates(const u8 *buf, unsigned int touch=
_count,

I'm wondering if you are using --histogram diff algo when preparing the pat=
ches.

..

> +       chip_data =3D device_get_match_data(&client->dev);
> +       if (!chip_data)
> +               chip_data =3D (const struct ektf2127_i2c_chip_data *)id->=
driver_data;
> +       if (!chip_data) {
> +               dev_err(&client->dev, "missing chip data\n");
> +               return -EINVAL;

First of all, there is a special combined API for this, please use it
instead. Second, use dev_err_probe().

> +       }

..

>  #ifdef CONFIG_OF

Side note: Ideally we should kill this ugliness. But it can be done later o=
n.


--
With Best Regards,
Andy Shevchenko