All of lore.kernel.org
 help / color / mirror / Atom feed
From: Viresh Kumar <viresh.kumar@linaro.org>
To: Rafael Wysocki <rjw@rjwysocki.net>
Cc: linaro-kernel@lists.linaro.org, linux-pm@vger.kernel.org,
	sboyd@codeaurora.org, prarit@redhat.com, skannan@codeaurora.org,
	Viresh Kumar <viresh.kumar@linaro.org>
Subject: [PATCH 17/18] cpufreq: Restart governor as soon as possible
Date: Tue, 27 Jan 2015 14:06:23 +0530	[thread overview]
Message-ID: <529898a0821ded6de14eb31c457d2f441382e9f5.1422346933.git.viresh.kumar@linaro.org> (raw)
In-Reply-To: <cover.1422346933.git.viresh.kumar@linaro.org>
In-Reply-To: <cover.1422346933.git.viresh.kumar@linaro.org>

On cpu hot-unplug, we don't need to wait for POST_DEAD notification to restart
the governor if the policy has atleast one online cpu left. We can restart the
governor right from the DOWN_PREPARE notification instead.

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
 drivers/cpufreq/cpufreq.c | 59 +++++++++++++++++++----------------------------
 1 file changed, 24 insertions(+), 35 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 0e03ae1bc96a..bcc042a6221a 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1338,7 +1338,7 @@ static int __cpufreq_remove_dev_prepare(struct device *dev,
 					struct subsys_interface *sif)
 {
 	unsigned int cpu = dev->id, cpus;
-	int ret;
+	int ret = 0;
 	unsigned long flags;
 	struct cpufreq_policy *policy;
 
@@ -1379,25 +1379,35 @@ static int __cpufreq_remove_dev_prepare(struct device *dev,
 				CPUFREQ_NAME_LEN);
 	}
 
-	if (cpu != policy->cpu)
-		return 0;
-
 	if (cpus > 1) {
-		/* Nominate new CPU */
 		down_write(&policy->rwsem);
-		policy->cpu = cpumask_any_but(policy->cpus, cpu);
+		cpumask_clear_cpu(cpu, policy->cpus);
+
+		/* Nominate new CPU */
+		if (cpu == policy->cpu)
+			policy->cpu = cpumask_any(policy->cpus);
 		up_write(&policy->rwsem);
+
+		if (has_target()) {
+			/* Start governor again */
+			ret = __cpufreq_governor(policy, CPUFREQ_GOV_START);
+			if (!ret)
+				ret = __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS);
+
+			if (ret)
+				pr_err("%s: Failed to start governor\n", __func__);
+		}
 	} else if (cpufreq_driver->stop_cpu) {
 		cpufreq_driver->stop_cpu(policy);
 	}
 
-	return 0;
+	return ret;
 }
 
 static int __cpufreq_remove_dev_finish(struct device *dev,
 				       struct subsys_interface *sif)
 {
-	unsigned int cpu = dev->id, cpus;
+	unsigned int cpu = dev->id;
 	int ret;
 	unsigned long flags;
 	struct cpufreq_policy *policy;
@@ -1406,34 +1416,13 @@ static int __cpufreq_remove_dev_finish(struct device *dev,
 	policy = cpufreq_cpu_get_raw(cpu);
 	read_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
-	if (!policy) {
-		pr_debug("%s: No cpu_data found\n", __func__);
-		return -EINVAL;
-	}
-
-	down_write(&policy->rwsem);
-	cpus = cpumask_weight(policy->cpus);
-
-	if (cpus > 1)
-		cpumask_clear_cpu(cpu, policy->cpus);
-	up_write(&policy->rwsem);
-
-	/* Not the last cpu of policy, start governor again ? */
-	if (cpus > 1) {
-		if (!has_target())
-			return 0;
-
-		ret = __cpufreq_governor(policy, CPUFREQ_GOV_START);
-		if (!ret)
-			ret = __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS);
-
-		if (ret) {
-			pr_err("%s: Failed to start governor\n", __func__);
-			return ret;
-		}
-
+	/*
+	 * If this isn't the last cpu of policy, we will fail to get policy here
+	 * as the cpumask in policy->cpus is already cleared. Hence we don't
+	 * need to proceed here anymore
+	 */
+	if (!policy)
 		return 0;
-	}
 
 	/* If cpu is last user of policy, free policy */
 	if (has_target()) {
-- 
2.3.0.rc0.44.ga94655d


  parent reply	other threads:[~2015-01-27  8:37 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-27  8:36 [PATCH 00/18] cpufreq: don't loose cpufreq history on CPU hotplug Viresh Kumar
2015-01-27  8:36 ` [PATCH 01/18] cpufreq: Drop cpufreq_disabled() check from cpufreq_cpu_{get|put}() Viresh Kumar
2015-02-03 22:17   ` Saravana Kannan
2015-01-27  8:36 ` [PATCH 02/18] cpufreq: Create for_each_policy() Viresh Kumar
2015-02-03 22:22   ` Saravana Kannan
2015-02-04  4:53     ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 03/18] cpufreq: Create for_each_governor() Viresh Kumar
2015-02-03 22:23   ` Saravana Kannan
2015-01-27  8:36 ` [PATCH 04/18] cpufreq: Manage fallback policies in a list Viresh Kumar
2015-02-03  0:41   ` Rafael J. Wysocki
2015-02-03  4:10     ` Viresh Kumar
2015-02-03 15:04       ` Rafael J. Wysocki
2015-02-04  6:18         ` Viresh Kumar
2015-02-03 22:28   ` Saravana Kannan
2015-02-04  6:20     ` Viresh Kumar
2015-02-04 22:28       ` Saravana Kannan
2015-02-04 23:20         ` Rafael J. Wysocki
2015-02-05  1:55           ` Saravana Kannan
2015-02-05 15:11             ` Rafael J. Wysocki
2015-02-05 22:55               ` Saravana Kannan
2015-02-17  8:06               ` Viresh Kumar
2015-02-17 18:15                 ` Rafael J. Wysocki
2015-02-18  4:23                   ` Viresh Kumar
2015-02-18 21:15                     ` Saravana Kannan
2015-02-19  3:24                       ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 05/18] cpufreq: Manage governor usage history with 'policy->last_governor' Viresh Kumar
2015-02-12  3:03   ` Saravana Kannan
2015-02-12  7:44     ` Viresh Kumar
2015-02-12  8:00       ` skannan
2015-02-17  8:02         ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 06/18] cpufreq: Reuse policy list instead of per-cpu variable 'cpufreq_cpu_data' Viresh Kumar
2015-02-12  3:13   ` Saravana Kannan
2015-02-12  7:48     ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 07/18] cpufreq: Drop (now) useless check 'cpu > nr_cpu_ids' Viresh Kumar
2015-02-12  3:15   ` Saravana Kannan
2015-02-12  7:50     ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 08/18] cpufreq: Add doc style comment about cpufreq_cpu_{get|put}() Viresh Kumar
2015-02-12  3:19   ` Saravana Kannan
2015-02-12  7:52     ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 09/18] cpufreq: Mark policy->governor = NULL for fallback policies Viresh Kumar
2015-02-12  3:22   ` Saravana Kannan
2015-02-12  7:56     ` Viresh Kumar
2015-01-27  8:36 ` [PATCH 10/18] cpufreq: Don't allow updating inactive-policies from sysfs Viresh Kumar
2015-02-12  3:24   ` Saravana Kannan
2015-01-27  8:36 ` [PATCH 11/18] cpufreq: Track cpu managing sysfs kobjects separately Viresh Kumar
2015-01-27  8:36 ` [PATCH 12/18] cpufreq: Stop migrating sysfs files on hotplug Viresh Kumar
2015-01-27  8:36 ` [PATCH 13/18] cpufreq: Keep a single path for adding managed CPUs Viresh Kumar
2015-01-27  8:36 ` [PATCH 14/18] cpufreq: Remove cpufreq_update_policy() Viresh Kumar
2015-01-27  8:36 ` [PATCH 15/18] cpufreq: Initialize policy->kobj while allocating policy Viresh Kumar
2015-01-27  8:36 ` [PATCH 16/18] cpufreq: Call cpufreq_policy_put_kobj() from cpufreq_policy_free() Viresh Kumar
2015-01-27  8:36 ` Viresh Kumar [this message]
2015-01-27  8:36 ` [PATCH 18/18] cpufreq: Merge __cpufreq_add_dev() and cpufreq_add_dev() Viresh Kumar
2015-01-27 15:06 ` [PATCH 00/18] cpufreq: don't loose cpufreq history on CPU hotplug Rafael J. Wysocki
2015-01-27 14:59   ` Viresh Kumar
2015-01-28 19:35 ` Saravana Kannan
2015-01-29  1:43   ` Viresh Kumar
2015-02-03  0:30 ` Rafael J. Wysocki

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=529898a0821ded6de14eb31c457d2f441382e9f5.1422346933.git.viresh.kumar@linaro.org \
    --to=viresh.kumar@linaro.org \
    --cc=linaro-kernel@lists.linaro.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=prarit@redhat.com \
    --cc=rjw@rjwysocki.net \
    --cc=sboyd@codeaurora.org \
    --cc=skannan@codeaurora.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.