[PATCH v3 1/2] arm64: dts: meson-g12b-odroid-n2: Enable RTC controller node

[Anand Moon]

Hi All,

On Wed, 26 Aug 2020 at 12:29, Anand Moon <linux.amoon at gmail.com> wrote:
>
> hi Jerome
>
> On Tue, 25 Aug 2020 at 20:00, Jerome Brunet <jbrunet at baylibre.com> wrote:
> >
> >
> > On Tue 25 Aug 2020 at 11:01, Anand Moon <linux.amoon at gmail.com> wrote:
> >
> > > Hi Jerome
> > >
> > > On Mon, 24 Aug 2020 at 20:00, Jerome Brunet <jbrunet at baylibre.com> wrote:
> > >>
> > >>
> > >> On Mon 24 Aug 2020 at 15:50, Neil Armstrong <narmstrong at baylibre.com> wrote:
> > >>
> > >> > On 24/08/2020 15:41, Anand Moon wrote:
> > >> >> hi All,
> > >> >>
> > >> >> On Fri, 21 Aug 2020 at 10:13, Anand Moon <linux.amoon at gmail.com> wrote:
> > >> >>>
> > >> >>> Hi Kevin,
> > >> >>>
> > >> >>> Thanks for your review comments.
> > >> >>>
> > >> >>> On Fri, 21 Aug 2020 at 01:03, Kevin Hilman <khilman at baylibre.com> wrote:
> > >> >>>>
> > >> >>>> Anand Moon <linux.amoon at gmail.com> writes:
> > >> >>>>
> > >> >>>>> Enable RTC PCF8563 node on Odroid-N2 SBC, In order to
> > >> >>>>> support the RTC wakealarm feature for suspend and resume.
> > >> >>>>> Also assign an alias to the pcf8563 to rtc0 and meson-vrtc to rtc1
> > >> >>>>> timer device to prevent it being assigned to /dev/rtc0
> > >> >>>>> which disto userspace tools assume is a clock device.
> > >> >>>>>
> > >> >>>>> Cc: Neil Armstrong <narmstrong at baylibre.com>
> > >> >>>>> Cc: Kevin Hilman <khilman at baylibre.com>
> > >> >>>>> Suggested-by: Christian Hewitt <christianshewitt at gmail.com>
> > >> >>>>> Signed-off-by: Anand Moon <linux.amoon at gmail.com>
> > >> >>>>> ---
> > >> >>>>> Changes V3
> > >> >>>>> --Drop the INI GPIOAO.BIT7 pinctrl.
> > >> >>>>
> > >> >>>> Why did you drop this GPIO?  Isn't this the GPIO that the RTC uses to
> > >> >>>> wakeup the system?  If so, this should be included as part of this
> > >> >>>> patch.
> > >> >>>>
> > >> >>>> It probably still works because the bootloader configures this GPIO as
> > >> >>>> input, but the kernel should not rely on the booloader for that, so
> > >> >>>> please include as part of this patch.
> > >> >>>>
> > >> >>>
> > >> >>> Ok I will figure out the correct pinctrl need for this settings.
> > >> >>> looking into the Odroid N2 schematics.
> > >> >>>
> > >> >>
> > >> >> I am trying to map the RTC INT pinctrl, ie RTC INT GPIOAO.BIT7.
> > >> >
> > >> >
> > >> > Simply add:
> > >> >
> > >> > interrupt-parent = <&gpio_intc>;
> > >> > interrupts = <7 IRQ_TYPE_LEVEL_LOW>;
> > >> >
> > >> > to reflect the interrupt connection.
> > >
> > > I have tried this setting, but it is not working.
> > >
> > > [alarm at archl-on2e ~]$ dmesg| grep rtc
> > > [    5.378002] meson-vrtc ff8000a8.rtc: registered as rtc1
> > > [    5.942307] rtc-pcf8563 0-0051: pcf8563_write_block_data: err=-110
> > > addr=0e, data=03
> > > [    5.942316] rtc-pcf8563 0-0051: pcf8563_probe: write error
> > > [    5.945983] rtc-pcf8563: probe of 0-0051 failed with error -5
> > >
> >
> > -110 is timeout ... either you i2c bus is broken and you device is not
> > at 0x51. In both case, it has nothing to do with the interrupt configuration
> >
> I have check the I2C bus on the device for rtc and it return correctly.
>
> $ sudo i2cdetect -l
> i2c-1   i2c             DesignWare HDMI                         I2C adapter
> i2c-0   i2c             Meson I2C adapter                       I2C adapter
> $
> $ sudo i2cdetect -y -r 0
>      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
> 00:          -- -- -- -- -- -- -- -- -- -- -- -- --
> 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 50: -- UU -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 70: -- -- -- -- -- -- -- --
>
> > >
> > >> >
> > >> > No need to setup pinctrl here since the GPIO input is always connected
> > >> > to the gpio irq generator whatever pinctrl mode is set.
> > >>
> > >> It is actually better to setup pinctrl. Yes the irq controller can work
> > >> whatever the pin setup but if an output function is active it can mess with
> > >> what the irq controller gets.
> > >>
> > >> Think about applying/removing bias if necessary too.
> > >>
> > >
> > > Ok, I am trying to add a new pinctrl configuration for
> > > TSIN_A_DIN0 //  TDMB_FS // TDMB_SLV_FS
> > > But it's still not working at my end.
> >
> > Either you are quite confused when it comes to pinctrl or I am.
>
> Basically I dont want to keep repeating silly mistakes.
>
> > TSIN and TDM have nothing to do with an i2c RTC.
>
> Yes I understand this clearly, that's why I have dropped the
> RTC INT gpio pinctrl settings.Without this setting RTC driver
> works fine and their is no issue rtcwakeup during suspend
> and resume in my testing.
>

Just want to clear my confusion on RTC INT gpio setting is not needed.
I did not find any other example to support this changes.
So I have enable the debug logs on rtc-pcf8563.c with this current
patch at my end.

[0] https://pastebin.com/F2UA665H

-Best Regards
Anand