[PATCH] gpio: rcar: Use new line value setter callbacks

[PATCH] gpio: rcar: Use new line value setter callbacks

[Geert Uytterhoeven]

struct gpio_chip now has callbacks for setting line values that return
integers, so they can indicate failures.  Convert the driver to using
them.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
---
 drivers/gpio/gpio-rcar.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/gpio/gpio-rcar.c b/drivers/gpio/gpio-rcar.c
index 4fc7cad5032a7a80..130aa0cac186d0cb 100644
--- a/drivers/gpio/gpio-rcar.c
+++ b/drivers/gpio/gpio-rcar.c
@@ -359,7 +359,7 @@ static int gpio_rcar_get_multiple(struct gpio_chip *chip, unsigned long *mask,
 	return 0;
 }
 
-static void gpio_rcar_set(struct gpio_chip *chip, unsigned offset, int value)
+static int gpio_rcar_set(struct gpio_chip *chip, unsigned int offset, int value)
 {
 	struct gpio_rcar_priv *p = gpiochip_get_data(chip);
 	unsigned long flags;
@@ -367,10 +367,12 @@ static void gpio_rcar_set(struct gpio_chip *chip, unsigned offset, int value)
 	raw_spin_lock_irqsave(&p->lock, flags);
 	gpio_rcar_modify_bit(p, OUTDT, offset, value);
 	raw_spin_unlock_irqrestore(&p->lock, flags);
+
+	return 0;
 }
 
-static void gpio_rcar_set_multiple(struct gpio_chip *chip, unsigned long *mask,
-				   unsigned long *bits)
+static int gpio_rcar_set_multiple(struct gpio_chip *chip, unsigned long *mask,
+				  unsigned long *bits)
 {
 	struct gpio_rcar_priv *p = gpiochip_get_data(chip);
 	unsigned long flags;
@@ -378,7 +380,7 @@ static void gpio_rcar_set_multiple(struct gpio_chip *chip, unsigned long *mask,
 
 	bankmask = mask[0] & GENMASK(chip->ngpio - 1, 0);
 	if (!bankmask)
-		return;
+		return 0;
 
 	raw_spin_lock_irqsave(&p->lock, flags);
 	val = gpio_rcar_read(p, OUTDT);
@@ -386,6 +388,8 @@ static void gpio_rcar_set_multiple(struct gpio_chip *chip, unsigned long *mask,
 	val |= (bankmask & bits[0]);
 	gpio_rcar_write(p, OUTDT, val);
 	raw_spin_unlock_irqrestore(&p->lock, flags);
+
+	return 0;
 }
 
 static int gpio_rcar_direction_output(struct gpio_chip *chip, unsigned offset,
@@ -537,8 +541,8 @@ static int gpio_rcar_probe(struct platform_device *pdev)
 	gpio_chip->get = gpio_rcar_get;
 	gpio_chip->get_multiple = gpio_rcar_get_multiple;
 	gpio_chip->direction_output = gpio_rcar_direction_output;
-	gpio_chip->set = gpio_rcar_set;
-	gpio_chip->set_multiple = gpio_rcar_set_multiple;
+	gpio_chip->set_rv = gpio_rcar_set;
+	gpio_chip->set_multiple_rv = gpio_rcar_set_multiple;
 	gpio_chip->label = name;
 	gpio_chip->parent = dev;
 	gpio_chip->owner = THIS_MODULE;

Re: [PATCH] gpio: rcar: Use new line value setter callbacks

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 222 bytes --]


>  	bankmask = mask[0] & GENMASK(chip->ngpio - 1, 0);
>  	if (!bankmask)
> -		return;
> +		return 0;

Doesn't that mean that the mask is invalid and we could return an error
here? Or is '!bankmask' an expected use-case?


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH] gpio: rcar: Use new line value setter callbacks

[Geert Uytterhoeven]

Hi Wolfram,

On Fri, 13 Jun 2025 at 13:42, Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> >       bankmask = mask[0] & GENMASK(chip->ngpio - 1, 0);
> >       if (!bankmask)
> > -             return;
> > +             return 0;
>
> Doesn't that mean that the mask is invalid and we could return an error
> here? Or is '!bankmask' an expected use-case?

That is a good question!

I _think_ this really can't happen anymore, as the GPIO core is supposed
to check this against the valid mask? Or isn't it?

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH] gpio: rcar: Use new line value setter callbacks

[Bartosz Golaszewski]

On Fri, Jun 13, 2025 at 2:02 PM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>
> Hi Wolfram,
>
> On Fri, 13 Jun 2025 at 13:42, Wolfram Sang
> <wsa+renesas@sang-engineering.com> wrote:
> > >       bankmask = mask[0] & GENMASK(chip->ngpio - 1, 0);
> > >       if (!bankmask)
> > > -             return;
> > > +             return 0;
> >
> > Doesn't that mean that the mask is invalid and we could return an error
> > here? Or is '!bankmask' an expected use-case?
>
> That is a good question!
>
> I _think_ this really can't happen anymore, as the GPIO core is supposed
> to check this against the valid mask? Or isn't it?
>

Yes but this doesn't seem to have anything to do with the valid_mask?
If it's about the number of GPIOs then that too is verified by GPIO
core.

Bart

Re: [PATCH] gpio: rcar: Use new line value setter callbacks

[Geert Uytterhoeven]

Hi Bartosz,

On Tue, 17 Jun 2025 at 11:06, Bartosz Golaszewski <brgl@bgdev.pl> wrote:
> On Fri, Jun 13, 2025 at 2:02 PM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > On Fri, 13 Jun 2025 at 13:42, Wolfram Sang
> > <wsa+renesas@sang-engineering.com> wrote:
> > > >       bankmask = mask[0] & GENMASK(chip->ngpio - 1, 0);
> > > >       if (!bankmask)
> > > > -             return;
> > > > +             return 0;
> > >
> > > Doesn't that mean that the mask is invalid and we could return an error
> > > here? Or is '!bankmask' an expected use-case?
> >
> > That is a good question!
> >
> > I _think_ this really can't happen anymore, as the GPIO core is supposed
> > to check this against the valid mask? Or isn't it?
>
> Yes but this doesn't seem to have anything to do with the valid_mask?
> If it's about the number of GPIOs then that too is verified by GPIO
> core.

Sure, about the collection of valid GPIO offsets.  So it cannot really
happen, and just bailing out with zero sounds fine to me?

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH] gpio: rcar: Use new line value setter callbacks

[Bartosz Golaszewski]

On Tue, Jun 17, 2025 at 11:12 AM Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
>
> Hi Bartosz,
>
> On Tue, 17 Jun 2025 at 11:06, Bartosz Golaszewski <brgl@bgdev.pl> wrote:
> > On Fri, Jun 13, 2025 at 2:02 PM Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> > > On Fri, 13 Jun 2025 at 13:42, Wolfram Sang
> > > <wsa+renesas@sang-engineering.com> wrote:
> > > > >       bankmask = mask[0] & GENMASK(chip->ngpio - 1, 0);
> > > > >       if (!bankmask)
> > > > > -             return;
> > > > > +             return 0;
> > > >
> > > > Doesn't that mean that the mask is invalid and we could return an error
> > > > here? Or is '!bankmask' an expected use-case?
> > >
> > > That is a good question!
> > >
> > > I _think_ this really can't happen anymore, as the GPIO core is supposed
> > > to check this against the valid mask? Or isn't it?
> >
> > Yes but this doesn't seem to have anything to do with the valid_mask?
> > If it's about the number of GPIOs then that too is verified by GPIO
> > core.
>
> Sure, about the collection of valid GPIO offsets.  So it cannot really
> happen, and just bailing out with zero sounds fine to me?
>

If this cannot happen, then why not drop the check?

Bart