Re: [PATCH v2 2/5] gpio: syscon: Add gpio-syscon for rockchip

[Linus Walleij]

On Thu, May 24, 2018 at 10:35 AM, Heiko StÃbner <heiko@xxxxxxxxx> wrote:
> Am Donnerstag, 24. Mai 2018, 10:28:44 CEST schrieb Linus Walleij:
>> On Wed, May 23, 2018 at 5:12 PM, Heiko StÃbner <heiko@xxxxxxxxx> wrote:
>> > So the gpio controller should definitly also be a subnode.
>> >
>> > The gpio in question is called "mute", so I'd think the gpio-syscon driver
>> > should just define a "rockchip,rk3328-gpio-mute" compatible and contain
>> > all the register voodoo in the driver itself and not define it in the dt.
>> >
>> > So it should probably look like
>> >
>> > grf: syscon at ff100000 {
>> >
>> >         compatible = "rockchip,rk3328-grf", "syscon", "simple-mfd";
>> >
>> >         [all the other syscon sub-devices]
>> >
>> >         gpio_mute: gpio-mute {
>> >
>> >                 compatible = "rockchip,rk3328-gpio-mute";
>> >                 gpio-controller;
>> >                 #gpio-cells = <2>;
>> >
>> >         };
>>
>> I'm sceptic.
>>
>> That doesn't sound like "general purpose input output" at all.
>>
>> It sounds like special purpose, for a mute button.
>>
>> Does it use IRQ? I would recommend implementing
>> drivers/input/keyboard/syscon-keys.c in the same vein
>> as drivers/leds/leds-syscon.c so you can avoid indirection
>> through GPIO for no good reason at all.
>
> To quote Levin from the other mail:
> --------
> The "mute" pin is a output only GPIO, which is already supported by
> setting flags in the gpio-syscon
>   driver. And yes, this pin has a defined function, but can also be used
> for general purpose operation.
> --------
>
> So to summarize, the documentation calls it "mute", but it is usable as
> a general pin, which is the reason Levin is working on it - because on his
> board this pin is used to switch between two voltages (aka a gpio-regulator)
> for the sdmmc controller [3.3V + 1.8V].

OK then, I was wrong! :)

Go ahead with this, sorry for the fuzz.

Yours,
Linus Walleij