Hi Biju,
On Tue, May 9, 2023 at 9:35 AM Biju Das <biju.das.jz@xxxxxxxxxxxxxx> wrote:
> > Subject: Re: [PATCH v2 4/5] mfd: Add Renesas PMIC RAA215300 RTC driver
> > On Tue, May 9, 2023 at 9:06 AM Biju Das <biju.das.jz@xxxxxxxxxxxxxx> wrote:
> > > > Subject: [PATCH v2 4/5] mfd: Add Renesas PMIC RAA215300 RTC driver
> > > >
> > > > Currently, it is not possible to instantiate the i2c client driver
> > > > using MFD cell as it is not a platform driver. Add support for
> > > > Renesas PMIC
> > > > RAA215300 RTC platform driver, so that it can be instantiated by MFD
> > API.
> > > > The rtc device is created by using i2c_new_ancillary_device() and it
> > > > register as the rtc device by the helper function provided by
> > > > rtc-isl2108 driver.
> > >
> > > Not sure this platform driver has to be placed in RTC subsystem rather
> > > than MFD subsystem as PMIC MFD driver, can instantiate it using MFD cell??
> >
> > Can't you just instantiate the i2c ancillary device from the PMIC driver,
> > and drop the new mfd and platform rtc drivers?
>
> Yes, it is possible.
>
> The only reason MFD is used for future expansion like adding support for
>
> 1) battery charger
> Or
> 2) regulator
I'd just start with a skeleton regulator ("PMIC") driver...
> In that case, we can share regmap to sub devices. But these drivers are not currently planned.
>
> Apart from that,
>
> 1) It avoids subsystem dependencies, ie, PMIC driver directly calling rtc driver
> for registering RTC device.
You mean the direct call into raa215300_rtc_probe_helper()?
I think that can be solved by enhancing i2c_new_ancillary_device() to
take a proper device name, instead of using "dummy"?
> 2) I guess, the current split will also give some modular design.
> We have a control to enable or disable the driver.
> ie, Enable the driver PMIC with RTC enabled and
> disable the driver PMIC with RTC not enabled(ie, XIN/XOUT grounded) in the system design
> by controlling the config "PMIC_RAA215300_RTC".
You already have CONFIG_RTC_DRV_ISL1208, and look at
"renesas,rtc-enabled" before activating the RTC?
> These are the reasons for creating MFD driver and platform device.
>
> I may be totally wrong. Please correct me if that is the case.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
