Re: [RFC 00/19] clk: imx: Switch the imx6 and imx7 to clk_hw based API

From: Stephen Boyd
Date: Tue Apr 23 2019 - 17:50:03 EST


Quoting Stephen Boyd (2019-04-10 15:00:07)
> Quoting Fabio Estevam (2019-03-22 09:24:54)
> > Hi Abel,
> >
> > On Fri, Mar 22, 2019 at 12:37 PM Abel Vesa <abel.vesa@xxxxxxx> wrote:
> > >
> > > This work is part of the effort of switching the entire imx clock drivers
> > > towards clk_hw based API. In order to do that, we have to keep the old
> >
> > Maybe I missed some earlier discussions on this topic, but let me ask
> > a fundamental question first: what is the motivation for this API
> > change?
> >
> > I could not find the reasoning inside the series.
>
> The reasoning should be added to the series, but I would like to see clk
> providers only deal with clk_hw structures and not with clk structures.
> This clearly divides the line between clk consumers and clk providers so
> that we know if clk providers are using the clk API themselves. That's
> why I introduced clk_hw_register() in the first place.
>

FYI, I'm dropping this from my review queue under the assumption it will
be sent again. Also, it would be nice to not have the wrappers at all in
the next series so that everything is clk_hw based instead of clk based.
Otherwise, it looks good to me and I'd like to see this work completed.
Thanks!