From: Simon Horman Subject: Re: [PATCH v2 0/2] Renesas R8A77970 CPG/MSSR clock support Date: Wed, 13 Sep 2017 09:28:29 +0200 Message-ID: <20170913072829.GB6756@verge.net.au> References: <20170908213418.823028011@cogentembedded.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-clk-owner@vger.kernel.org To: Geert Uytterhoeven Cc: Sergei Shtylyov , Michael Turquette , Stephen Boyd , Rob Herring , Mark Rutland , Geert Uytterhoeven , linux-clk , "devicetree@vger.kernel.org" , Linux-Renesas List-Id: devicetree@vger.kernel.org On Mon, Sep 11, 2017 at 04:38:43PM +0200, Geert Uytterhoeven wrote: > Hi Sergei, > > On Fri, Sep 8, 2017 at 11:34 PM, Sergei Shtylyov > wrote: > > Here's the set of 2 patches against the 'clk-next' branch of CLK group's > > 'linux.git' repo. > > Thank you, looks good! > > > As the DT patches in the R8A77970/Eagle board support series > > depend on the patch #1 of this series, it might make sense to merge this patch > > to the 'renesas.git' repo as well... > > > > [1/2] dt-bindings: clock: add R8A77970 CPG core clock definitions > > The modern way to avoid this dependency is to use hardcoded numbers in > the initial submission of the DTS, cfr. > https://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git/commit/?h=arm64-dt-for-v4.15&id=d917e0b24811eadeba419ba7318b967ee15933b3 > > After the definitions have hit upstream, you can replace the hardcoded > numbers by symbols, cfr. > https://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers.git/commit/?h=topic/r8a77995-integration&id=d3f6987db28a4657bf3304eeeb8e2cd8dd510ad0 > https://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers.git/commit/?h=topic/r8a77995-integration&id=81426223583d5575767c77684be28c40034785bd > > So I don't think Simon will merge the core clock definitions patch, unless > you plan to submit a full DTS describing the whole SoC in the initial > submissiong. Thanks, I would prefer if the DT patches followed the scheme described by Geert such that they do not depend on any part of this series. I expect that will allow the smoothest path for acceptance of those patches.