RE: [PATCH mfd v2] mfd: Add Renesas Synchronization Management Unit (SMU) support

[Min Li <min.li.xe@renesas.com>]

> 
> > >
> > > > +static struct mfd_cell rsmu_cm_devs[] = {
> > > > +	[RSMU_PHC] = {
> > > > +		.name = "idtcm-phc",
> > >
> > > Can't you have a nicer name?
> > >
> > Hi Lee
> >
> > I wonder which part of the name that you don't like? PHC stands from PTP
> Hardware Clock.
> > I was following the name convention like tps65912-regulator.
> > Do you accept "8a3400-phc"?
> 
> I think you're trying to put too much information into the device name.
> 
> Currently it's:
> 
>   idt <company name> cm <platform> - phc <exact device type>
> 
> Where usually we have, taking your example:
> 
>   tps65912 <chip> - regulator <subsystem>
> 
> So assuming the PTP HW Clock is just a clock it should be:
> 
>   8a3400 <chip> - clock <subsystem>
> 
> It's difficult to say without seeing the associated child device(s).
> When do you propose to upstream those?  Maybe they should be part of
> this initial set.  I think that would help a lot.
> 

Hi Lee

The PHC driver already existed in the current tree as drivers/ptp/ptp_clockmatrix.c and
ptp_idt82p33.c

Right now, they act as i2c driver. I plan to change them as a normal platform device driver
after this MFD change kicked in

That is why I would prefer the name "phc" instead of "clock" since the driver is not a normal clk driver
under drivers/clk but a ptp clock driver. And down the road, we will have our real clock driver and I wanna
reserve the name "clock" for the real clock driver.

Thanks

Min