[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHp75VeR2LG-qAzQs2048K3pyfRkz7OKioHDNyU8bVAPsgy8ew@mail.gmail.com>
Date: Tue, 20 Feb 2018 14:56:33 +0200
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: David Lechner <david@...hnology.com>
Cc: linux-arm Mailing List <linux-arm-kernel@...ts.infradead.org>,
Linux OMAP Mailing List <linux-omap@...r.kernel.org>,
"open list:GPIO SUBSYSTEM" <linux-gpio@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Tony Lindgren <tony@...mide.com>,
Haojian Zhuang <haojian.zhuang@...aro.org>,
Linus Walleij <linus.walleij@...aro.org>
Subject: Re: [PATCH v2] pinctrl: pinctrl-single: Fix pcs_request_gpio() when
bits_per_mux != 0
On Mon, Feb 19, 2018 at 11:57 PM, David Lechner <david@...hnology.com> wrote:
> This fixes pcs_request_gpio() in the pinctrl-single driver when
> bits_per_mux != 0. It appears this was overlooked when the multiple
> pins per register feature was added.
>
> Fixes: 4e7e8017a80e ("pinctrl: pinctrl-single: enhance to configure multiple pins of different modules")
> Signed-off-by: David Lechner <david@...hnology.com>
> ---
>
> v2 changes:
> - don't wrap Fixes: line in commit message since it is a special machine-
> readable line.
>
> There was some discussion in v1 about using DIV_ROUND_UP(), etc. macros, but
> the consensus was to leave it as-is since it matches existing code and that
> macros can be introduced in another patch.
>
> drivers/pinctrl/pinctrl-single.c | 22 +++++++++++++++++++---
> 1 file changed, 19 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/pinctrl/pinctrl-single.c b/drivers/pinctrl/pinctrl-single.c
> index cec7537..a7c5eb3 100644
> --- a/drivers/pinctrl/pinctrl-single.c
> +++ b/drivers/pinctrl/pinctrl-single.c
> @@ -391,9 +391,25 @@ static int pcs_request_gpio(struct pinctrl_dev *pctldev,
> || pin < frange->offset)
> continue;
> mux_bytes = pcs->width / BITS_PER_BYTE;
> - data = pcs->read(pcs->base + pin * mux_bytes) & ~pcs->fmask;
> - data |= frange->gpiofunc;
> - pcs->write(data, pcs->base + pin * mux_bytes);
> +
> + if (pcs->bits_per_mux) {
> + int byte_num, offset, pin_shift;
> +
> + byte_num = (pcs->bits_per_pin * pin) / BITS_PER_BYTE;
> + offset = (byte_num / mux_bytes) * mux_bytes;
> + pin_shift = pin % (pcs->width / pcs->bits_per_pin) *
> + pcs->bits_per_pin;
> +
> + data = pcs->read(pcs->base + offset);
> + data &= ~(pcs->fmask << pin_shift);
> + data |= frange->gpiofunc << pin_shift;
> + pcs->write(data, pcs->base + offset);
> + } else {
> + data = pcs->read(pcs->base + pin * mux_bytes);
> + data &= ~pcs->fmask;
> + data |= frange->gpiofunc;
> + pcs->write(data, pcs->base + pin * mux_bytes);
Just an idea, you may leave this almost untouched and do calculate
pin_shift and offset in condition, like
if (...) {
pin_shift = ...
offset = ...
} else {
pin_shift = 0;
offset = pin * mux_bytes;
}
data = pcs->read(pcs->base + offset);
data &= ~(pcs->fmask << pin_shift);
data |= frange->gpiofunc << pin_shift;
pcs->write(data, pcs->base + offset);
It's also possible to split to two changes, where first introduces
that variables and their default values (see 'else' branch) and second
one introduces an if branch override.
> + }
> break;
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists