Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751942AbdHBDop (ORCPT ); Tue, 1 Aug 2017 23:44:45 -0400 Received: from mail-pf0-f174.google.com ([209.85.192.174]:35643 "EHLO mail-pf0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751779AbdHBDom (ORCPT ); Tue, 1 Aug 2017 23:44:42 -0400 Date: Wed, 2 Aug 2017 09:14:08 +0530 From: Viresh Kumar To: Pavan Kondeti Cc: Rafael Wysocki , Peter Zijlstra , joelaf@google.com, linux-pm@vger.kernel.org, LKML , skannan@codeaurora.org, smuckle.linux@gmail.com, eas-dev@lists.linaro.org Subject: Re: [Eas-dev] [PATCH V5 2/2] cpufreq: Process remote callbacks from any CPU if the platform permits Message-ID: <20170802034408.GB26689@vireshk-i7> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 984 Lines: 27 On 01-08-17, 16:30, Pavan Kondeti wrote: > Currently sugov threads in the schedutil governor are pinned to the > policy CPUs. schedutil can now make use of this new > dvfs_possible_from_any_cpu flag and avoid the pinning, right? Actually yes and it would be something as simple as below. Will send a patch for this if no one reports any problems with this. diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 9deedd5f16a5..6ea12a8c8863 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -471,7 +471,10 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy) } sg_policy->thread = thread; - kthread_bind_mask(thread, policy->related_cpus); + + if (!policy->dvfs_possible_from_any_cpu) + kthread_bind_mask(thread, policy->related_cpus); + init_irq_work(&sg_policy->irq_work, sugov_irq_work); mutex_init(&sg_policy->work_lock); -- viresh