On Mon, Sep 28, 2020 at 4:00 PM Andy Shevchenko
<andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
>
> On Mon, Sep 28, 2020 at 03:13:53PM +0200, Bartosz Golaszewski wrote:
> > On Mon, Sep 28, 2020 at 3:00 PM Andy Shevchenko
> > <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
> > >
> > > On Mon, Sep 28, 2020 at 12:41:53PM +0200, Bartosz Golaszewski wrote:
> > > > From: Bartosz Golaszewski <bgolaszewski@xxxxxxxxxxxx>
> > > >
> > > > While we do check the "chip-name" property in probe(), we never actually
> > > > use it. Let's pass the chip label to the driver using device properties
> > > > as we'll want to allow users to define their own once dynamically
> > > > created chips are supported.
> > > >
> > > > The property is renamed to "chip-label" to not cause any confusion with
> > > > the actual chip name which is of the form: "gpiochipX".
> > > >
> >
> > ^^^ here, see below
> >
> > > > If the "chip-label" property is missing, let's do what most devices in
> > > > drivers/gpio/ do and use dev_name().
> > >
> > > ...
> > >
> > > > + snprintf(chip_label, sizeof(chip_label),
> > > > + "gpio-mockup-%c", i + 'A');
> > > > + properties[prop++] = PROPERTY_ENTRY_STRING("chip-label",
> > > > + chip_label);
> > >
> > > You added new property, now count is up to 4. But at the same time
> > >
> > > #define GPIO_MOCKUP_MAX_PROP 4
> > >
> > > how do you avoid overflow?
> > >
> >
> > I renamed the property, the previous "chip-name" is no longer used. In
> > fact it was never used but was accounted for in GPIO_MOCKUP_MAX_PROP.
>
> Either I'm missing something or...
>
> Current code in linux-next has 3 properties to be possible
>
> PROPERTY_ENTRY_U32("gpio-base", base);
> PROPERTY_ENTRY_U16("nr-gpios", ngpio);
> PROPERTY_ENTRY_BOOL("named-gpio-lines");
>
> You adding here
> PROPERTY_ENTRY_STRING("chip-label", chip_label);
>
> Altogether after this patch is 4 which is maximum, but since array is passed by
> a solely pointer, the terminator is a must.
>
Thanks for explaining my code to me. Yes you're right and I'm not sure
why I missed this. :)
I'll fix this in v3.
Actually this means the code is wrong even before this series - it's
just that we don't use the "chip-name" property.
Bartosz
