Re: [PATCH v1 1/3] gpio: defer probe if pinctrl cannot be found

[Rob Herring <robherring2-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]

On Fri, Jul 10, 2015 at 4:29 AM, Tomeu Vizoso
<tomeu.vizoso-ZGY8ohtN/8qB+jHODAdFcQ@public.gmane.org> wrote:
> On 1 July 2015 at 19:36, Rob Herring <robherring2-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>> On Wed, Jul 1, 2015 at 7:45 AM, Tomeu Vizoso <tomeu.vizoso-ZGY8ohtN/8qB+jHODAdFcQ@public.gmane.org> wrote:
>>> When an OF node has a pin range for its GPIOs, return -EPROBE_DEFER if
>>> the pin controller isn't available.
>>>
>>> Otherwise, the GPIO range wouldn't be set at all unless the pin
>>> controller probed always before the GPIO chip.
>>>
>>> With this change, the probe of the GPIO chip will be deferred and will
>>> be retried at a later point, hopefully once the pin controller has been
>>> registered and probed already.
>>
>> This will break cases where the pinctrl driver does not exist, but the
>> DT contains pinctrl bindings. We can have similar problems already
>> with clocks though. However, IMO this problem is a bit different in
>> that pinctrl is more likely entirely optional while clocks are often
>> required. You may do all pin setup in bootloader/firmware on some
>> boards and not others. Of course then why put pinctrl in the DT in
>> that case? They could be present just due to how chip vs. board dts
>> files are structured.
>
> I see. My instinct tells me that it would be better if the gpio-ranges
> property was set in the board dts, but I don't really know what each
> mach does with its DTSs.
>
>> We could address this by simply marking the pin controller node
>> disabled. However, ...
>>
>>> @@ -361,7 +361,7 @@ static void of_gpiochip_add_pin_range(struct gpio_chip *chip)
>>>
>>>                 pctldev = of_pinctrl_get(pinspec.np);
>>>                 if (!pctldev)
>>> -                       break;
>>> +                       return -EPROBE_DEFER;
>>
>> But you cannot distinguish that case here. I think of_pinctrl_get
>> needs to set the error code appropriately.
>
> Why not? I was thinking of just doing this before we call of_pinctrl_get():
>
>         if (!of_device_is_available(pinspec.np))
>             continue;

That is exactly what you need, but that should be of_pinctrl_get's
responsibility to check, not the caller's. IIRC, this is the only user
of of_pinctrl_get, so it should be just as easy to change.

Rob