Re: [PATCH 2/2] gpio: mt7621: Make the irqchip immutable

[Sergio Paracuellos <sergio.paracuellos@xxxxxxxxx>]

On Tue, Sep 13, 2022 at 4:43 PM Bartosz Golaszewski <brgl@xxxxxxxx> wrote:
>
> On Fri, Sep 9, 2022 at 11:56 AM Sergio Paracuellos
> <sergio.paracuellos@xxxxxxxxx> wrote:
> >
> > Commit 6c846d026d49 ("gpio: Don't fiddle with irqchips marked as
> > immutable") added a warning to indicate if the gpiolib is altering the
> > internals of irqchips.  Following this change the following warnings
> > are now observed for the mt7621 driver:
> >
> > gpio gpiochip0: (1e000600.gpio-bank0): not an immutable chip, please consider fixing it!
> > gpio gpiochip1: (1e000600.gpio-bank1): not an immutable chip, please consider fixing it!
> > gpio gpiochip2: (1e000600.gpio-bank2): not an immutable chip, please consider fixing it!
> >
> > Fix this by making the irqchip in the mt7621 driver immutable.
> >
> > Tested-by: Arınç ÜNAL <arinc.unal@xxxxxxxxxx>
> > Signed-off-by: Sergio Paracuellos <sergio.paracuellos@xxxxxxxxx>
> > ---
> >  drivers/gpio/gpio-mt7621.c | 21 +++++++++++++++------
> >  1 file changed, 15 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/gpio/gpio-mt7621.c b/drivers/gpio/gpio-mt7621.c
> > index 05891dd3f96e..729cc8f72343 100644
> > --- a/drivers/gpio/gpio-mt7621.c
> > +++ b/drivers/gpio/gpio-mt7621.c
> > @@ -111,6 +111,8 @@ mediatek_gpio_irq_unmask(struct irq_data *d)
> >         unsigned long flags;
> >         u32 rise, fall, high, low;
> >
> > +       gpiochip_enable_irq(gc, d->hwirq);
> > +
> >         spin_lock_irqsave(&rg->lock, flags);
> >         rise = mtk_gpio_r32(rg, GPIO_REG_REDGE);
> >         fall = mtk_gpio_r32(rg, GPIO_REG_FEDGE);
> > @@ -142,6 +144,8 @@ mediatek_gpio_irq_mask(struct irq_data *d)
> >         mtk_gpio_w32(rg, GPIO_REG_HLVL, high & ~BIT(pin));
> >         mtk_gpio_w32(rg, GPIO_REG_LLVL, low & ~BIT(pin));
> >         spin_unlock_irqrestore(&rg->lock, flags);
> > +
> > +       gpiochip_disable_irq(gc, d->hwirq);
> >  }
> >
> >  static int
> > @@ -203,6 +207,16 @@ mediatek_gpio_xlate(struct gpio_chip *chip,
> >         return gpio % MTK_BANK_WIDTH;
> >  }
> >
> > +static const struct irq_chip mt7621_irq_chip = {
> > +       .name           = "mt7621-gpio",
> > +       .irq_mask_ack   = mediatek_gpio_irq_mask,
> > +       .irq_mask       = mediatek_gpio_irq_mask,
> > +       .irq_unmask     = mediatek_gpio_irq_unmask,
> > +       .irq_set_type   = mediatek_gpio_irq_type,
> > +       .flags          = IRQCHIP_IMMUTABLE,
> > +       GPIOCHIP_IRQ_RESOURCE_HELPERS,
> > +};
> > +
> >  static int
> >  mediatek_gpio_bank_probe(struct device *dev,
> >                          struct fwnode_handle *fwnode, int bank)
> > @@ -239,11 +253,6 @@ mediatek_gpio_bank_probe(struct device *dev,
> >                 return -ENOMEM;
> >
> >         rg->chip.offset = bank * MTK_BANK_WIDTH;
> > -       rg->irq_chip.name = dev_name(dev);
> > -       rg->irq_chip.irq_unmask = mediatek_gpio_irq_unmask;
> > -       rg->irq_chip.irq_mask = mediatek_gpio_irq_mask;
> > -       rg->irq_chip.irq_mask_ack = mediatek_gpio_irq_mask;
> > -       rg->irq_chip.irq_set_type = mediatek_gpio_irq_type;
> >
> >         if (mtk->gpio_irq) {
> >                 struct gpio_irq_chip *girq;
> > @@ -263,7 +272,7 @@ mediatek_gpio_bank_probe(struct device *dev,
> >                 }
> >
> >                 girq = &rg->chip.irq;
> > -               girq->chip = &rg->irq_chip;
> > +               gpio_irq_chip_set_chip(girq, &mt7621_irq_chip);
> >                 /* This will let us handle the parent IRQ in the driver */
> >                 girq->parent_handler = NULL;
> >                 girq->num_parents = 0;
> > --
> > 2.25.1
> >
>
> This should go into fixes but depends on the other patch which is
> material for the next release cycle. Can you resend this separately?

Sure, I will!

Thanks,
    Sergio Paracuellos

>
> Bart