Re: [RESEND PATCH v5 1/5] mfd: RK808: Add RK818 support

[Lee Jones]

On Thu, 09 Jun 2016, Wadim Egorov wrote:

> 
> 
> On 09.06.2016 13:12, Lee Jones wrote:
> > On Thu, 09 Jun 2016, Wadim Egorov wrote:
> >> On 08.06.2016 16:17, Lee Jones wrote:
> >>> On Thu, 02 Jun 2016, Wadim Egorov wrote:
> >>>
> >>>> The RK818 chip is a power management IC for multimedia and handheld
> >>> "Power Management IC (PMIC)"
> >>>
> >>>> devices. It contains the following components:
> >>>>
> >>>> - Regulators
> >>>> - RTC
> >>>> - Clkout
> >>> Clocking
> >>>
> >>>> - battery support
> >>> Battery support
> >>>
> >>>> Both chips RK808 and RK818 are using a similar register map.
> >>> "Both RK808 ad RK818 chips"
> >>>
> >>>> So we can reuse the RTC and Clkout functionality.
> >>> Swap '.' for ','.
> >>>
> >>>> Signed-off-by: Wadim Egorov <w.egorov@xxxxxxxxx>
> >>>> ---
> >>>>  drivers/mfd/Kconfig       |   4 +-
> >>>>  drivers/mfd/rk808.c       | 231 ++++++++++++++++++++++++++++++++++++++--------
> >>>>  include/linux/mfd/rk808.h | 162 ++++++++++++++++++++++++++++++--
> >>>>  3 files changed, 350 insertions(+), 47 deletions(-)
> > [...]
> >
> >>>> -struct rk808_reg_data {
> >>>> -	int addr;
> >>>> -	int mask;
> >>>> -	int value;
> >>>> -};
> >>> Why are you moving this to the header?
> >> It is now part of the rk808 struct.
> > Why?
> >
> For me it seemed reasonable. My idea was to have only one chip variant
> check.
> 
> I can move the rk808_reg_data struct back to the c file, check for the
> chip variant within the probe function and initialize the proper
> reg_data variable.
> This would also apply to the mfd_cell variables.

'struct rk808' should only contain information you wish to share with
your subordinate/child devices.  Any data only used in the parent/MFD
driver should be declared, initialised and used only the latter.  No
need to export.

-- 
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org â Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog