Re: [PATCH linux-next] clk: sparx5: Review changes

From: Arnd Bergmann <arnd@arndb.de>
To: Stephen Boyd <sboyd@kernel.org>
Cc: Lars Povlsen <lars.povlsen@microchip.com>,
	SoC Team <soc@kernel.org>,
	Steen Hegelund <Steen.Hegelund@microchip.com>,
	Microchip Linux Driver Support <UNGLinuxDriver@microchip.com>,
	linux-clk <linux-clk@vger.kernel.org>,
	Linux ARM <linux-arm-kernel@lists.infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Alexandre Belloni <alexandre.belloni@bootlin.com>
Subject: Re: [PATCH linux-next] clk: sparx5: Review changes
Date: Mon, 27 Jul 2020 22:11:49 +0200	[thread overview]
Message-ID: <CAK8P3a27u7UkjGJ32KsRp02FoF34mKW0gRUsms0+YBkxy+L4Gg@mail.gmail.com> (raw)
In-Reply-To: <159587873833.1360974.11729154337431621644@swboyd.mtv.corp.google.com>

On Mon, Jul 27, 2020 at 9:39 PM Stephen Boyd <sboyd@kernel.org> wrote:
> Quoting Arnd Bergmann (2020-07-27 05:02:56)
> > On Mon, Jul 27, 2020 at 1:22 PM Lars Povlsen <lars.povlsen@microchip.com> wrote:
> > >
> > > This address the review comments for the sparx5 clk driver from Stephen
> > > Boyd <sboyd@kernel.org>:
> > >
> > >  - Remove unused include of of_address.h
> > >  - Remove unused member in s5_hw_clk struct
> > >  - Changed type (to unsigned long) for freq in s5_pll_conf struct
> > >  - Use .parent_data instead of looking up parent name
> > >  - Use devm_of_clk_add_hw_provider
> > >  - Some minor comsmetics
> > >
> > > The patch is intended for linux-next (incremental), as the original
> > > driver was already merged.
> > >
> > > Signed-off-by: Lars Povlsen <lars.povlsen@microchip.com>
> >
> > Hi Lars, thank you for addressing these!
> >
> > Generally speaking, you should avoid having patches that list a
> > number of unrelated things that are done by a single commit.
> >
> > Splitting this up into six patches is probably a little too much,
> > but maybe you can find a better balance. What I'd like to see
> > is to split the purely cosmetic changes from those that might
> > actually change the behavior, and then for each patch that
> > changes something non-obvious, explain why this was done.
> >
>
> Why was the clk driver merged to linux-next outside of the clk tree? Was
> there some sort of dependency?

I merged the entire series of the base platform support along with
a few core drivers. I had asked for the series to be submitted to
soc@kernel.org after all parts had been reviewed, but I missed that
the clk driver was still missing maintainer review (I saw you had
reviewed some patches, but apparently that was just the binding,
not the driver).

I rebased the 'arm/newsoc' branch the other day to fix another mistake,
so if you prefer, I can rebase it again and drop the clk driver or
all the sparx5 patches.

       Arnd