Re: [PATCH V2 2/3] gpio: tegra: Remove the need of keeping device handle for gpio driver

[Stephen Warren]

On 04/19/2016 06:43 AM, Laxman Dewangan wrote:
> On Tuesday 19 April 2016 06:03 PM, Thierry Reding wrote:
> > * PGP Signed by an unknown key
> >
> > On Tue, Apr 19, 2016 at 03:13:39PM +0530, Laxman Dewangan wrote:
> > > Remove the file static device handle variable for keeping device handle
> > > of driver as this is just required for error prints. The required device
> > > handle are available from gpiochip structure.
> > >
> > > Signed-off-by: Laxman Dewangan <ldewangan@xxxxxxxxxx>
> > > ---
> > >   drivers/gpio/gpio-tegra.c | 8 ++++----
> > >   1 file changed, 4 insertions(+), 4 deletions(-)
> > >
> > > diff --git a/drivers/gpio/gpio-tegra.c b/drivers/gpio/gpio-tegra.c
> > > index 1b0c497..de022a9 100644
> > > --- a/drivers/gpio/gpio-tegra.c
> > > +++ b/drivers/gpio/gpio-tegra.c
> > > @@ -80,7 +80,6 @@ struct tegra_gpio_soc_config {
> > >       u32 upper_offset;
> > >   };
> > > -static struct device *dev;
> > >   static struct irq_domain *irq_domain;
> > >   static void __iomem *regs;
> > >   static u32 tegra_gpio_bank_count;
> > > @@ -240,7 +239,8 @@ static int tegra_gpio_irq_set_type(struct
> > > irq_data *d, unsigned int type)
> > >       ret = gpiochip_lock_as_irq(&tegra_gpio_chip, gpio);
> > >       if (ret) {
> > > -        dev_err(dev, "unable to lock Tegra GPIO %d as IRQ\n", gpio);
> > > +        dev_err(tegra_gpio_chip.parent,
> > > +            "unable to lock Tegra GPIO %d as IRQ\n", gpio);
> > Can't we get rid of the tegra_gpio_chip global variable altogether? We
> > set a struct tegra_gpio_bank * as the chip data for each of the
> > interrupts, so if we added a back link to the GPIO chip to each bank
> > we could easily get at the GPIO chip (and its parent device) from the
> > IRQ chip implementation.
>
> We can get rid of all the global variables and can have only one (this
> is because our REGS macros uses the stride and changing then part of
> argument is too much change for single patch) as suggested by Stephen.

The REGS macro can take a parameter that points to the struct 
tegra_gpio_chip, which in turn can point at the struct 
tegra_gpio_soc_config. That doesn't sound like too many parameters/... 
to me. There's little point getting rid of all-but-one global.