From mboxrd@z Thu Jan 1 00:00:00 1970 From: arnd@arndb.de (Arnd Bergmann) Date: Fri, 21 Nov 2014 13:55:43 +0100 Subject: [GIT PULL] ARM: imx: SoC updates for 3.19 In-Reply-To: References: <20141119070306.GF27759@dragon> <201411201730.04562.arnd@arndb.de> Message-ID: <2257892.t3t7utnvDU@wuerfel> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Friday 21 November 2014 16:37:55 Shawn Guo wrote: > On Fri, Nov 21, 2014 at 12:30 AM, Arnd Bergmann wrote: > > The changes all look good, but could you please send them again based on > > v3.18-rc1? I'm trying to avoid back-merges from upstream into the next > > branches. > > The imx/cleanup branch can simply be based on v3.18-rc1. But I choose > to base imx/soc on a latest -rc to resolve a conflict between commit > 410ba8d4a0da (ARM: imx: clk-vf610: get input clocks from assigned > clocks) and c72c553249bb (ARM: imx: clk-vf610: define PLL's clock > tree). The former one is currently sits on imx/soc branch, and the > latter landed on -rc4 as a fix. I think we can handle it in either > way of the following two. > > 1. Drop commit 410ba8d4a0da (ARM: imx: clk-vf610: get input clocks > from assigned clocks) from imx/soc branch and send it as a fix for > 3.18 right way. > > 2. Keep 410ba8d4a0da on imx/soc and leave the merge conflict to Linus. > > What's your preference? I think the best approach in this case would be to rebase your series on top of 410ba8d4a0da. next/soc already contains -rc3 but not -rc4, so this would solve the backmerge problem nicely. > > > > Also, note that we've started splitting out defconfig changes into > > a separate topic branch. Since you are rebasing anyway, please send > > the defconfig change as a separate patch or pull request and also > > update multi_v7_defconfig in the same way. > > Okay, to keep it simple, I will squashed the defconfig updates into > one patch and send it to you. Ok, sounds good, thanks! Arnd