Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20221220090248.1134214-1-haibo.chen@nxp.com> <20221220090248.1134214-2-haibo.chen@nxp.com>
 <CACRpkdaPQODoiG+ARAiT-6X2u6n7_aCO7Ywj4z=76A65iUhTpA@mail.gmail.com> <DB7PR04MB4010B71F54175CD7C73C3FC490C69@DB7PR04MB4010.eurprd04.prod.outlook.com>
In-Reply-To: <DB7PR04MB4010B71F54175CD7C73C3FC490C69@DB7PR04MB4010.eurprd04.prod.outlook.com>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Tue, 17 Jan 2023 15:12:56 +0100
Message-ID: <CACRpkdZDbsNsaQ+9sM2BEiKXJhJuGU7jk2M07L5TRZ6H1u2VFg@mail.gmail.com>
Subject: Re: [PATCH 2/2] gpio: vf610: Support building gpio-vf610 driver as module
To:     Bough Chen <haibo.chen@nxp.com>
Cc:     "brgl@bgdev.pl" <brgl@bgdev.pl>,
        "stefan@agner.ch" <stefan@agner.ch>,
        "linux-gpio@vger.kernel.org" <linux-gpio@vger.kernel.org>,
        dl-linux-imx <linux-imx@nxp.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Jan 17, 2023 at 3:01 PM Bough Chen <haibo.chen@nxp.com> wrote:
> > On Tue, Dec 20, 2022 at 10:02 AM <haibo.chen@nxp.com> wrote:
> >
> > > From: Haibo Chen <haibo.chen@nxp.com>
> > >
> > > To support Android GKI, need to build this driver as module.
> > > So change the config as tristate type, and add module license.
> > >
> > > Signed-off-by: Jindong Yue <jindong.yue@nxp.com>
> > > Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
> >
> > OK...
> >
> > >  builtin_platform_driver(vf610_gpio_driver);
> >
> > Then you want something like module_platform_driver() here.
>
> Hi Linus,
>
> Here, though need to build this gpio driver as module due to the Android GKI, but we do not really need to
> rmmod/unbind this driver, so do not need to implement the _exit or remove callback.
> In this case, still need to change to module_platform_driver()?
> If so, after use the module_platform_driver(), is it okay to not implement the __exit /remove callback?

Yes, our project is about maintaining and improving the Linux kernel, GKI
is not a project we are running. If they need stuff from Linux, it happens on
our terms, and that is that modularization should be self-motivated.

Check this article:
https://lwn.net/Articles/872209/

Yours,
Linus Walleij