From mboxrd@z Thu Jan 1 00:00:00 1970 From: Viresh Kumar Subject: Re: [PATCH v5 3/7] ARM: dts: Exynos542x/5800: add CPU OPP properties Date: Fri, 11 Dec 2015 09:02:53 +0530 Message-ID: <20151211033253.GN3612@ubuntu> References: <1449766729-435-1-git-send-email-b.zolnierkie@samsung.com> <1449766729-435-4-git-send-email-b.zolnierkie@samsung.com> <20151211031646.GL3612@ubuntu> <566A4231.9050608@osg.samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mail-pf0-f181.google.com ([209.85.192.181]:34138 "EHLO mail-pf0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751352AbbLKDc5 (ORCPT ); Thu, 10 Dec 2015 22:32:57 -0500 Received: by pfbg73 with SMTP id g73so59776805pfb.1 for ; Thu, 10 Dec 2015 19:32:57 -0800 (PST) Content-Disposition: inline In-Reply-To: <566A4231.9050608@osg.samsung.com> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Javier Martinez Canillas Cc: Bartlomiej Zolnierkiewicz , Thomas Abraham , Sylwester Nawrocki , Mike Turquette , Kukjin Kim , Kukjin Kim , Krzysztof Kozlowski , Ben Gamari , Tomasz Figa , Lukasz Majewski , Heiko Stuebner , Chanwoo Choi , Kevin Hilman , Tobias Jakobi , Anand Moon , linux-samsung-soc@vger.kernel.org, linux-pm@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, Doug Anderson , Andreas Faerber On 11-12-15, 00:25, Javier Martinez Canillas wrote: > The problem is that the big and LITTLE cores have different ordering per SoCs: > > - Exynos5420 and Exynos5800: cpu0-3 (Cortex-A15) and cpu4-7 (Coretx-A7) > - Exynos5422: cpu0-3 (Cortex-A7) and cpu4-7 (Cortex-A15) > > So the OPP tables are set in this DTSI file, to prevent the OPP tables > in the Exynos5422 to be inverted for the cluster 0 and 1. Oh dude, that's really *ugly*. :) Reusing files/definitions is fine to the point where things are readable. But you have screwed it up so very badly. Over that, why can't you keep cpu0-3 as A7 and 4-7 as a15 for all the cases? The only worrying thing for you should be that CPU0 within the kenrel should be controllable, right? i.e. you want a A15 to boot 5800 and A7 to boot 5422. If yes, than you could have kept the CPUs in 5422 as: 0-3: A7 4-7: A15 and in 5420 as: 4-7: A15 0-3: A7 Wouldnt' that work ? -- viresh