From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751410AbcEJJRW (ORCPT ); Tue, 10 May 2016 05:17:22 -0400 Received: from mga02.intel.com ([134.134.136.20]:18988 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751036AbcEJJRU (ORCPT ); Tue, 10 May 2016 05:17:20 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,604,1455004800"; d="scan'208";a="972382033" From: yu.c.chen@intel.com To: linux-pm@vger.kernel.org Cc: Srinivas Pandruvada , "Rafael J. Wysocki" , Len Brown , Viresh Kumar , Zhang Rui , linux-kernel@vger.kernel.org, Chen Yu Subject: [PATCH][RFC] cpufreq: intel_pstate: Avoid time-costly synchronization if schedutil is not enabled Date: Tue, 10 May 2016 17:24:12 +0800 Message-Id: <1462872252-16152-1-git-send-email-yu.c.chen@intel.com> X-Mailer: git-send-email 2.7.4 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Chen Yu Commit bb6ab52f2bef ("intel_pstate: Do not set utilization update hook too early") moved the assignment of utilization update hook from intel_pstate_init_cpu to intel_pstate_set_policy(), however after moving, we need to leverage synchronize_sched() to deal with synchronization, which is time-costly because it needs to wait for all the CPUs to go through a grace period. Actuall we don't need to synchronize_sched if schedutil has not been enabled, which also can address the problem in our case, thus bypass the synchronize_sched during bootup. Fixes: bb6ab52f2bef "intel_pstate: Do not set utilization update hook too early" Link: https://bugzilla.kernel.org/show_bug.cgi?id=116371 Tested-by: Tian Ye Signed-off-by: Chen Yu --- drivers/cpufreq/intel_pstate.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index b230eba..9811e4e 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -162,6 +162,7 @@ struct _pid { * struct cpudata - Per CPU instance data storage * @cpu: CPU number for this instance data * @update_util: CPUFreq utility callback information + * @update_util_set: CPUFreq utility callback is set * @pstate: Stores P state limits for this CPU * @vid: Stores VID limits for this CPU * @pid: Stores PID parameters for this CPU @@ -179,6 +180,7 @@ struct cpudata { int cpu; struct update_util_data update_util; + bool update_util_set; struct pstate_data pstate; struct vid_data vid; @@ -1287,11 +1289,18 @@ static void intel_pstate_set_update_util_hook(unsigned int cpu_num) /* Prevent intel_pstate_update_util() from using stale data. */ cpu->sample.time = 0; cpufreq_set_update_util_data(cpu_num, &cpu->update_util); + cpu->update_util_set = true; } static void intel_pstate_clear_update_util_hook(unsigned int cpu) { + struct cpudata *cpu_data = all_cpu_data[cpu]; + + if (!cpu_data->update_util_set) + return; + cpufreq_set_update_util_data(cpu, NULL); + cpu_data->update_util_set = false; synchronize_sched(); } -- 2.7.4