Re: [PATCH v1 3/3] ACPI / PMIC: Add TI PMIC TPS68470 operation region driver

From: Sakari Ailus
Date: Wed Jun 07 2017 - 16:11:00 EST


Hi Andy,

On Wed, Jun 07, 2017 at 04:40:13PM +0300, Andy Shevchenko wrote:
> On Wed, Jun 7, 2017 at 3:15 PM, Sakari Ailus <sakari.ailus@xxxxxx> wrote:
> > On Tue, Jun 06, 2017 at 05:23:56PM +0300, Andy Shevchenko wrote:
> >> Follow the pattern, please, I suppose
> >> ti_pmic_tps68470.c
> >
> > This pattern is weird. "ti" in front of the file name is redundant, and in
> > very few places the vendor prefix is used anyway. Especially when the chip
> > has a proper name --- as this one does.
> >
> > I assume for the Intel PMICs it could be there for a couple of reasons which
> > are
> >
> > 1) lack of a clearly unique chip ID and
> >
> > 2) the use of common frameworklet for Intel PMICs.
> >
> > There are also no other PMIC chips supported currently.
> >
> > The pmic_tps68470 naming is in line with the GPIO driver (apart from the
> > dash / underscore difference).
>
> Since
>
> % git ls-files *pmic*
>
> returns somewhat interesting results, I would even go further and use
>
> tps68470.c here
>
> and
>
> s/ti_pmic/tps6840/g
>
> inside the file.
>
> Would it work for you?

This is still a different driver from the tps68470 driver which is an MFD
driver. For clarity, I'd keep pmic as part of the name (and I'd use
tps68470_pmic_ prefix for internal symbols, too).

As PMICs are typically linked to the kernel (vs. being modules), there's no
issue with the module name. I would suppose few if any PMICs will be
compiled as modules in general.

It's not a big deal though. I'm fine either way.

--
Regards,

Sakari Ailus
e-mail: sakari.ailus@xxxxxx XMPP: sailus@xxxxxxxxxxxxxx