From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tony Lindgren Subject: Re: [PATCH v2 0/3] ARM: dts: Add operating-points-v2 tables for TI platforms Date: Wed, 14 Sep 2016 09:56:38 -0700 Message-ID: <20160914165635.h4afluenzlfhcyo5@atomide.com> References: <20160901024816.32390-1-d-gerlach@ti.com> <20160913220244.txdiak6ikujvwzjk@atomide.com> <57D97E9A.90400@ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <57D97E9A.90400@ti.com> Sender: linux-pm-owner@vger.kernel.org To: Dave Gerlach Cc: Rob Herring , linux-arm-kernel@lists.infradead.org, linux-omap@vger.kernel.org, devicetree@vger.kernel.org, linux-pm@vger.kernel.org, Viresh Kumar , Mark Rutland , Nishanth Menon List-Id: devicetree@vger.kernel.org * Dave Gerlach [160914 09:45]: > On 09/13/2016 05:02 PM, Tony Lindgren wrote: > > Assuming this will have to wait based on what you mentioned about > > am335x cpufreq regression. I guess you'll instead send few patches > > to revert? > > > > Correct, these should not be merged, I will be sending a fixup series until > this series is ready. OK > I needed to resend the ti-cpufreq driver at [2] to get a response about the > dt binding and other updates Viresh asked for and sent this to update the > merged nodes so they would work with the driver. But the driver has not been > merged yet, so in order to re-enable the previously existing cpufreq support > on those platforms I must revert three patches from v1 [1] of this series > that got merged, which I will send. > > I also found that 7694ca6e1d6f ("cpufreq: omap: Use generic platdev driver") > which went in v4.6-rc1 had already broken the existing cpufreq support on > am335x and dra7xx, so it's been missing for some time. I can send a patch > fixing that as well but I will need to remove it once the ti-cpufreq driver > in [2] below gets merged as that driver must create the cpufreq-dt platform > device instead of cpufreq-dt-platdev. So I will send a series that can fix > cpufreq until ti-cpufreq is ready which will give us better support for more > OPPs on am335x, am437x (which has no implementation today), and dra7xx. OK yeah let's fix the regression & thanks for the update. Regards, Tony