Received-SPF: pass (google.com: domain of linux-kernel+bounces-52222-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: Bartosz Golaszewski <brgl@bgdev.pl>
In-Reply-To: <v5cgkajd7vqz63jsflpx56lvofozh77w4kwomlm5kkwcb36uxs@yqlkx2eduqhk>
Precedence: bulk
MIME-Version: 1.0
References: <20240204-bus_cleanup-gpio-v1-1-82d91b4ee1de@marliere.net>
 <CAMRc=MdSApjK2b66LxLE9XZufvT-+SThuqjdr+yG-hsKEb4sVw@mail.gmail.com> <v5cgkajd7vqz63jsflpx56lvofozh77w4kwomlm5kkwcb36uxs@yqlkx2eduqhk>
Date: Mon, 5 Feb 2024 00:45:57 -0800
Message-ID: <CAMRc=MehgMNZ39wrtbFXFFs6gUN8m8K4NhYWUVNj=ETziKMfpA@mail.gmail.com>
Subject: Re: [PATCH] gpio: gpiolib: make gpio_bus_type const
To: "Ricardo B. Marliere" <ricardo@marliere.net>
Cc: Linus Walleij <linus.walleij@linaro.org>, linux-gpio@vger.kernel.org, 
	linux-kernel@vger.kernel.org, Greg Kroah-Hartman <gregkh@linuxfoundation.org>, 
	Bartosz Golaszewski <brgl@bgdev.pl>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sun, 4 Feb 2024 20:41:52 +0100, "Ricardo B. Marliere"
<ricardo@marliere.net> said:
> Hi Bartosz,
>
> On  4 Feb 19:34, Bartosz Golaszewski wrote:
>> On Sun, Feb 4, 2024 at 5:29=E2=80=AFPM Ricardo B. Marliere <ricardo@marl=
iere.net> wrote:
>> >
>> > Now that the driver core can properly handle constant struct bus_type,
>> > move the gpio_bus_type variable to be a constant structure as well,
>> > placing it into read-only memory which can not be modified at runtime.
>> >
>> > Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>> > Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
>> > Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
>> > ---
>> >  drivers/gpio/gpiolib.c | 2 +-
>> >  1 file changed, 1 insertion(+), 1 deletion(-)
>> >
>> > diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
>> > index d50a786f8176..24d046268a01 100644
>> > --- a/drivers/gpio/gpiolib.c
>> > +++ b/drivers/gpio/gpiolib.c
>> > @@ -63,7 +63,7 @@ static int gpio_bus_match(struct device *dev, struct=
 device_driver *drv)
>> >         return 1;
>> >  }
>> >
>> > -static struct bus_type gpio_bus_type =3D {
>> > +static const struct bus_type gpio_bus_type =3D {
>> >         .name =3D "gpio",
>> >         .match =3D gpio_bus_match,
>> >  };
>> >
>> > ---
>> > base-commit: 3eac8bbed22e940ac1645a884f221bef408f675c
>> > change-id: 20240204-bus_cleanup-gpio-57eea8d32a5a
>> >
>> > Best regards,
>> > --
>> > Ricardo B. Marliere <ricardo@marliere.net>
>> >
>>
>> Ha! Does the same work now for struct device_type?
>
> From a quick grep it seems to be the case, but maybe you should wait for
> Greg for confirmation. Also found this:
> https://lore.kernel.org/all/1305850262-9575-5-git-send-email-gregkh@suse.=
de/
>
> Best regards,
> -	Ricardo.
>
>
>
>>
>> Because I just sent this:
>> https://lore.kernel.org/linux-gpio/20240201162758.50733-1-brgl@bgdev.pl/
>>
>> And maybe I should make it const right away.
>>
>> Bart
>

Turns out it has been like that for years:

commit aed65af1cc2f6fc9ded5a8158f1405a02cf6d2ff
Author: Stephen Hemminger <shemminger@vyatta.com>
Date:   Mon Mar 28 09:12:52 2011 -0700

    drivers: make device_type const

    The device_type structure does not contain data that changes
    during usage and should be const. This allows devices to declare
    the struct const.

    I have patches to change all the subsystems, but need the infra
    structure change first.

    Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
    Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

I queued your patch too.

Bart