On Thu, 10 Oct 2024, Christian Marangi wrote:
> On Wed, Oct 09, 2024 at 11:55:50AM +0100, Lee Jones wrote:
> > On Wed, 09 Oct 2024, Lee Jones wrote:
> >
> > > On Wed, 09 Oct 2024, Lorenzo Bianconi wrote:
> > >
> > > > On Oct 02, Lee Jones wrote:
> > > > > On Tue, 01 Oct 2024, Lorenzo Bianconi wrote:
> > > > >
> > > > > > From: Christian Marangi <ansuelsmth@xxxxxxxxx>
> > > > > >
> > > > > > Support for Airoha EN7581 Multi Function Device that
> > > > > > expose PINCTRL functionality and PWM functionality.
> > > > >
> > > > > The device is a jumble of pinctrl registers, some of which can oscillate.
> > > > >
> > > > > This is *still* not an MFD.
> > > > >
> > > > > If you wish to spread this functionality over 2 drivers, use syscon to
> > > > > obtain the registers and simple-mfd to automatically probe the drivers.
> > > >
> > > > Hi Lee,
> > > >
> > > > IIUC you are suggesting two possible approaches here:
> > > >
> > > > 1- have a single driver implementing both pinctrl and pwm functionalities.
> > > > This approach will not let us reuse the code for future devices that
> > > > have just one of them in common, like pwm (but we can live with that).
> > >
> > > If you can have one without the other, then they are separate devices.
> > >
> > > > 2- use a device node like the one below (something similar to [0])
> > > >
> > > > system-controller@1fbf0200 {
> > > > compatible = "syscon", "simple-mfd";
> > > > reg = <0x0 0x1fbf0200 0x0 0xc0>;
> > > >
> > > > interrupt-parent = <&gic>;
> > > > interrupts = <GIC_SPI 26 IRQ_TYPE_LEVEL_HIGH>;
> > > >
> > > > gpio-controller;
> > > > #gpio-cells = <2>;
> > > >
> > > > interrupt-controller;
> > > > #interrupt-cells = <2>;
> > > >
> > > > pio: pinctrl {
> > > > compatible = "airoha,en7581-pinctrl";
> > > >
> > > > [ some pinctrl properties here ]
> > > > };
> > > >
> > > > #pwm-cells = <3>;
> > > >
> > > > pwm {
> > > > compatible = "airoha,en7581-pwm";
> > > > };
> > > > };
> > > >
> > > > Please correct me if I am wrong, but using syscon/simple-mfd as compatible
> > > > string for the 'parent' device, will require to introduce the compatible strings
> > > > even for the child devices in order to probe them, correct?
> > > > If so, as pointed out by Christian, this is something nacked by Rob/Krzysztof/Conor
> > > > (this is the main reason why we introduced a full mfd driver here).
> > > >
> > > > @Rob, Krzysztof, Conor: am I right?
> > >
> > > I don't see why separate functionality shouldn't have separate
> > > compatible strings, even if the registers are together. Register layout
> > > and functionality separation are not related.
> >
> > We've been happy to support both pinctrl and pwm devices before:
> >
> > git grep "\-pinctrl\|\-pwm" -- drivers/mfd
> > git grep "\-pinctrl\|\-pwm" -- arch/*/boot/dts
> >
> > git grep "\-pinctrl" -- arch/*/boot/dts | wc -l
> > 602
> > git grep "\-pwm" -- arch/*/boot/dts | wc -l
> > 856
> >
> > What makes this particular device different to all of the others?
> >
>
> Hi Lee,
>
> this would be the final DTS following the "simple-mfd" pattern.
>
> Can you confirm it's correct?
I can't confirm that it's 100% correct, but it looks okay to me.
> mfd: system-controller@1fbf0200 {
Not sure about the mfd: label though.
What is the device?
> compatible = "syscon", "simple-mfd";
> reg = <0x0 0x1fbf0200 0x0 0xc0>;
>
> interrupt-parent = <&gic>;
> interrupts = <GIC_SPI 26 IRQ_TYPE_LEVEL_HIGH>;
>
> gpio-controller;
> #gpio-cells = <2>;
>
> interrupt-controller;
> #interrupt-cells = <2>;
>
> gpio-ranges = <&mfd 0 13 47>;
>
> #pwm-cells = <3>;
>
> pio: pinctrl {
> compatible = "airoha,en7581-pinctrl";
>
> mdio_pins: mdio-pins {
> mux {
> function = "mdio";
> groups = "mdio";
> };
>
> conf {
> pins = "gpio2";
> output-high;
> };
> };
>
> pcie0_rst_pins: pcie0-rst-pins {
> conf {
> pins = "pcie_reset0";
> drive-open-drain = <1>;
> };
> };
>
> pcie1_rst_pins: pcie1-rst-pins {
> conf {
> pins = "pcie_reset1";
> drive-open-drain = <1>;
> };
> };
> };
>
> pwm {
> compatible = "airoha,en7581-pwm";
> };
> };
>
> --
> Ansuel
--
Lee Jones [李琼斯]
