[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMRc=Mfk9yBSkMQj_egtfNQ8=A+mvyqqy+_KM690GCV-Sts0Lg@mail.gmail.com>
Date: Wed, 3 Sep 2025 18:12:20 +0200
From: Bartosz Golaszewski <brgl@...ev.pl>
To: Andy Shevchenko <andriy.shevchenko@...el.com>
Cc: Linus Walleij <linus.walleij@...aro.org>, Yinbo Zhu <zhuyinbo@...ngson.cn>,
Hoan Tran <hoan@...amperecomputing.com>, Manivannan Sadhasivam <mani@...nel.org>,
Yang Shen <shenyang39@...wei.com>, linux-gpio@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-unisoc@...ts.infradead.org,
Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Subject: Re: [PATCH RESEND 05/14] gpio: ts4800: use generic device properties
On Wed, Sep 3, 2025 at 5:41 PM Andy Shevchenko
<andriy.shevchenko@...el.com> wrote:
>
> On Mon, Aug 25, 2025 at 11:48:46AM +0200, Bartosz Golaszewski wrote:
> >
> > Avoid pulling in linux/of.h by using the generic device properties.
>
> ...
>
> > - retval = of_property_read_u32(node, "ngpios", &ngpios);
> > + retval = device_property_read_u32(dev, "ngpios", &ngpios);
> > if (retval == -EINVAL)
> > ngpios = DEFAULT_PIN_NUMBER;
> > else if (retval)
>
> Don't we have a method in GPIOLIB that does this (can be called explicitly by the drivers)?
>
Sure but we don't have the same behavior in GPIO core, we don't know
what DEFAULT_PIN_NUMBER is. I can't test this functionally so don't
want to change semantics.
Bart
Powered by blists - more mailing lists