off by one error, maxpins

[Luiz Otavio O Souza]

Yes, please commit it.

And by the way... There is another bug in that code, the gpio driver
automatically adjust itself to the number of GPIO pins available on SoC:

ar71xx_gpiovar.h:#define AR71XX_GPIO_PINS 12
ar71xx_gpiovar.h:#define AR724X_GPIO_PINS 18
ar71xx_gpiovar.h:#define AR91XX_GPIO_PINS 22
ar933xreg.h:#define AR933X_GPIO_COUNT 30
ar934xreg.h:#define AR934X_GPIO_COUNT 23

But the size of the gpio_pins struct is statically defined (and with a
small size for most of the supported SoC):

struct ar71xx_gpio_softc {
        ....
        struct gpio_pin         gpio_pins[AR71XX_GPIO_PINS];
};

I'll write a patch to allocate this structure dynamically based on the SoC
we're running on.

Luiz


On 5 September 2013 20:14, Adrian Chadd <adrian at freebsd.org> wrote:

> Commit ahoy. :)
>
>
> -a
>
>
> On 5 September 2013 15:47, Sean Bruno <sean_bruno at yahoo.com> wrote:
>
> > On Thu, 2013-09-05 at 16:42 -0600, Warner Losh wrote:
> > > Shouldn't npins and/or maxpin just be bumped by 1 instead?
> > >
> > > Warner
> > >
> > > On Sep 5, 2013, at 4:38 PM, Sean Bruno wrote:
> > >
> > > > http://people.freebsd.org/~sbruno/maxpin_plus_one.txt
> > > >
> >
> > It looks like its  *intentionally* assigned as the h/w specific (GPIO
> > COUNT) -1 in ar71xx_gpio_pin_max()
> >
> > Userland assumes that this value is a 0-based thing.
> >
> _______________________________________________
> freebsd-embedded at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-embedded
> To unsubscribe, send any mail to "freebsd-embedded-unsubscribe at freebsd.org
> "
>