linux-gpio.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andy Shevchenko <andy.shevchenko@gmail.com>
To: Aidan MacDonald <aidanmacdonald.0x0@gmail.com>
Cc: Michael Walle <michael@walle.cc>,
	Linus Walleij <linus.walleij@linaro.org>,
	Bartosz Golaszewski <brgl@bgdev.pl>,
	"open list:GPIO SUBSYSTEM" <linux-gpio@vger.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] gpio: regmap: Support registers with more than one bit per GPIO
Date: Sun, 3 Jul 2022 16:09:26 +0200	[thread overview]
Message-ID: <CAHp75VeosViEcLZ3LF92=_bpSmyGthh2j9V4xYtHTt0Z-fn3-Q@mail.gmail.com> (raw)
In-Reply-To: <20220703111057.23246-2-aidanmacdonald.0x0@gmail.com>

On Sun, Jul 3, 2022 at 1:11 PM Aidan MacDonald
<aidanmacdonald.0x0@gmail.com> wrote:
>
> Some devices use a multi-bit register field to change the GPIO
> input/output direction. Add the ->reg_field_xlate() callback to
> support such devices in gpio-regmap.
>
> ->reg_field_xlate() builds on ->reg_mask_xlate() by allowing the
> driver to return a mask and values to describe a register field.
> gpio-regmap will use the mask to isolate the field and compare or
> update it using the values to implement GPIO level and direction
> get and set ops.

Thanks for the proposal. My comments below.

...

> +static void
> +gpio_regmap_simple_field_xlate(struct gpio_regmap *gpio,
> +                              unsigned int base, unsigned int offset,
> +                              unsigned int *reg, unsigned int *mask,
> +                              unsigned int *values)
> +{
> +       gpio->reg_mask_xlate(gpio, base, offset, reg, mask);
> +       values[0] = 0;
> +       values[1] = *mask;

This is a fragile and less compile-check prone approach. If you know
the amount of values, make a specific data type for that, or pass the
length of the output buffer..

> +}

...

> +       unsigned int values[2];

> +       return (val & mask) == values[1];

> +       unsigned int values[2];

How will the callee know that it's only 2 available?


> +       regmap_update_bits(gpio->regmap, reg, mask, values[!!val]);

If we have special meaning of the values, perhaps it needs to follow
an enum of some definitions, so everybody will understand how indices
are mapped to the actual data in the array.

> +       unsigned int values[2];

> +       regmap_update_bits(gpio->regmap, reg, mask, values[1]);

> +       unsigned int values[2];

> +       if ((val & mask) == values[invert])

How do you guarantee this won't overflow? (see above comment about
indices mapping)

> +       unsigned int values[2];

As per above comments.

-- 
With Best Regards,
Andy Shevchenko

  reply	other threads:[~2022-07-03 14:10 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-03 11:10 [PATCH 0/3] gpio-regmap support for register fields and other hooks Aidan MacDonald
2022-07-03 11:10 ` [PATCH 1/3] gpio: regmap: Support registers with more than one bit per GPIO Aidan MacDonald
2022-07-03 14:09   ` Andy Shevchenko [this message]
2022-07-04 16:03     ` Aidan MacDonald
2022-07-04 12:28   ` Michael Walle
2022-07-04 16:01     ` Aidan MacDonald
2022-07-04 19:46       ` Michael Walle
2022-07-06 20:46         ` Aidan MacDonald
2022-07-07  7:44           ` Michael Walle
2022-07-07 14:58             ` Aidan MacDonald
2022-07-03 11:10 ` [PATCH 2/3] gpio: regmap: Support combined GPIO and pin control drivers Aidan MacDonald
2022-07-03 14:14   ` Andy Shevchenko
2022-07-04 15:31     ` Aidan MacDonald
2022-07-03 11:10 ` [PATCH 3/3] gpio: regmap: Support a custom ->to_irq() hook Aidan MacDonald
2022-07-03 14:24   ` Andy Shevchenko
2022-07-04 16:38     ` Aidan MacDonald
2022-07-04 23:05   ` Linus Walleij
2022-07-05 11:09     ` Aidan MacDonald
2022-07-06 11:45       ` Andy Shevchenko
2022-07-06 20:53         ` Aidan MacDonald
2022-07-06 12:02       ` Linus Walleij
2022-07-06 13:50         ` Aidan MacDonald
2022-07-11 11:48           ` Linus Walleij

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAHp75VeosViEcLZ3LF92=_bpSmyGthh2j9V4xYtHTt0Z-fn3-Q@mail.gmail.com' \
    --to=andy.shevchenko@gmail.com \
    --cc=aidanmacdonald.0x0@gmail.com \
    --cc=brgl@bgdev.pl \
    --cc=linus.walleij@linaro.org \
    --cc=linux-gpio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=michael@walle.cc \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).