From mboxrd@z Thu Jan 1 00:00:00 1970 From: Simon Horman Subject: Re: [PATCH 3/3] arm: dts: r7s72100: Add peripherals nodes Date: Wed, 11 Jan 2017 11:33:17 +0100 Message-ID: <20170111103316.GD12030@verge.net.au> References: <1483990318-26927-1-git-send-email-jacopo+renesas@jmondi.org> <1483990318-26927-4-git-send-email-jacopo+renesas@jmondi.org> <2351875.0S8UmoVGa8@avalon> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from kirsty.vergenet.net ([202.4.237.240]:55320 "EHLO kirsty.vergenet.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1763424AbdAKKda (ORCPT ); Wed, 11 Jan 2017 05:33:30 -0500 Content-Disposition: inline In-Reply-To: <2351875.0S8UmoVGa8@avalon> Sender: linux-gpio-owner@vger.kernel.org List-Id: linux-gpio@vger.kernel.org To: Laurent Pinchart Cc: Geert Uytterhoeven , Jacopo Mondi , Magnus Damm , Geert Uytterhoeven , Chris Brandt , Linus Walleij , Linux-Renesas , "linux-gpio@vger.kernel.org" On Tue, Jan 10, 2017 at 09:58:19PM +0200, Laurent Pinchart wrote: > Hi Geert, > > On Tuesday 10 Jan 2017 16:07:01 Geert Uytterhoeven wrote: > > On Mon, Jan 9, 2017 at 8:31 PM, Jacopo Mondi wrote: > > > From: Magnus Damm > > > > > > This is a squash of several commits, adding peripherals groups > > > configuration to r7s72100 device tree, and enabling some of them on > > > Genmai evaluation board > > > > > > Signed-off-by: Jacopo Mondi > > > > Thanks for the rework! > > > > > arch/arm/boot/dts/r7s72100-genmai.dts | 51 ++++++++++++ > > > arch/arm/boot/dts/r7s72100.dtsi | 151 +++++++++++++++++++++++++++++ > > > > This path should be split in multiple parts: > > - Add the pfc node to r7s72100.dtsi, > > - Add the gpio nodes to r7s72100.dtsi, > > - 4 patches for r7s72100-genmai.dts, adding support for LEDs, SCIF, > > Ethernet, and SPI. > > I can agree about the .dtsi/.dts split, but isn't this going a bit overboard ? I would like the split so that different patches touch different files to be made. I am willing to be flexible regarding adding more than one IP block in a single patch if the patches would otherwise be very small and unlikely to lead to breakage. >>From my PoV a key motivation for splitting things up is to make it easier to selectively revert or backport individual features. I personally don't have much cause to do either on a fine-grained basis of late. So I'm happy to consider being more flexible with regards to patch granularity.