Re: [PATCH v2 03/10] gpio: raspberrypi-exp: Release firmware handle on unbind
From: Bartosz Golaszewski
Date: Mon Oct 26 2020 - 10:45:35 EST
On Mon, Oct 26, 2020 at 3:42 PM Nicolas Saenz Julienne
<nsaenzjulienne@xxxxxxx> wrote:
>
> On Mon, 2020-10-26 at 15:40 +0100, Bartosz Golaszewski wrote:
> > On Thu, Oct 22, 2020 at 5:59 PM Nicolas Saenz Julienne
> > <nsaenzjulienne@xxxxxxx> wrote:
> > > Upon unbinding the device make sure we release RPi's firmware interface.
> > >
> > > Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@xxxxxxx>
> > > ---
> > > drivers/gpio/gpio-raspberrypi-exp.c | 14 +++++++++++++-
> > > 1 file changed, 13 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/gpio/gpio-raspberrypi-exp.c b/drivers/gpio/gpio-raspberrypi-exp.c
> > > index bb100e0124e6..c008336e1131 100644
> > > --- a/drivers/gpio/gpio-raspberrypi-exp.c
> > > +++ b/drivers/gpio/gpio-raspberrypi-exp.c
> > > @@ -231,8 +231,19 @@ static int rpi_exp_gpio_probe(struct platform_device *pdev)
> > > rpi_gpio->gc.get = rpi_exp_gpio_get;
> > > rpi_gpio->gc.set = rpi_exp_gpio_set;
> > > rpi_gpio->gc.can_sleep = true;
> > > + platform_set_drvdata(pdev, rpi_gpio);
> > >
> > > - return devm_gpiochip_add_data(dev, &rpi_gpio->gc, rpi_gpio);
> > > + return gpiochip_add_data(&rpi_gpio->gc, rpi_gpio);
> > > +}
> > > +
> > > +static int rpi_exp_gpio_remove(struct platform_device *pdev)
> > > +{
> > > + struct rpi_exp_gpio *rpi_gpio = platform_get_drvdata(pdev);
> > > +
> > > + gpiochip_remove(&rpi_gpio->gc);
> > > + rpi_firmware_put(rpi_gpio->fw);
> > > +
> > > + return 0;
> > > }
> > >
> > > static const struct of_device_id rpi_exp_gpio_ids[] = {
> > > @@ -247,6 +258,7 @@ static struct platform_driver rpi_exp_gpio_driver = {
> > > .of_match_table = of_match_ptr(rpi_exp_gpio_ids),
> > > },
> > > .probe = rpi_exp_gpio_probe,
> > > + .remove = rpi_exp_gpio_remove,
> > > };
> > > module_platform_driver(rpi_exp_gpio_driver);
> > >
> > > --
> > > 2.28.0
> > >
> >
> > Why not introduce devm_rpi_firmware_get()? That would allow you to
> > keep the driver elegant without re-adding remove().
>
> I like the idea, I'll look into it.
>
> Thanks,
> Nicolas
>
If you can't do it for some reason, then even using devm_add_action() is fine.
Bartosz