Re: [PATCH 1/3] gpio: swnode: Add ability to specify native chip selects for SPI

From: Linus Walleij <linus.walleij@linaro.org>
To: Charles Keepax <ckeepax@opensource.cirrus.com>
Cc: broonie@kernel.org, brgl@bgdev.pl, linux-gpio@vger.kernel.org,
	 linux-spi@vger.kernel.org, patches@opensource.cirrus.com
Subject: Re: [PATCH 1/3] gpio: swnode: Add ability to specify native chip selects for SPI
Date: Tue, 9 Apr 2024 09:12:01 +0200	[thread overview]
Message-ID: <CACRpkdZQNrDt35d30xJSRz=03rhs6vOODWorpqMsJ=Lo4huJmQ@mail.gmail.com> (raw)
In-Reply-To: <ZhPvW46kGeOnG++E@ediswmail9.ad.cirrus.com>

On Mon, Apr 8, 2024 at 3:21 PM Charles Keepax
<ckeepax@opensource.cirrus.com> wrote:
> On Thu, Apr 04, 2024 at 10:16:35AM +0200, Linus Walleij wrote:
> > On Tue, Mar 26, 2024 at 3:11 PM Charles Keepax
> > <ckeepax@opensource.cirrus.com> wrote:
> > > +const struct software_node swnode_gpio_undefined = {
> > > +       .name = "gpio-internal-undefined",
> > > +};
> > > +EXPORT_SYMBOL_GPL(swnode_gpio_undefined);
> >
> > This needs a comment in the code telling exactly why this is here.
> > It is also taking up space and code here on systems that have no use
> > for it, so I wonder if it is possible to make this optional.
> >
>
> Happy to add the comment, less sure about how to make it
> optional. I could ifdef it based the SPI config, but whilst that
> is the current user the mechanism feels like it is more generic
> than that and could be used in other bindings as well.

That's a fair point.
Maybe a new bool Kconfig symbol that the SPI drivers or
other potential users can select?

Yours,
Linus Walleij