diff options
| author | Viresh Kumar <viresh.kumar@linaro.org> | 2015-12-03 09:37:54 +0530 |
|---|---|---|
| committer | Rafael J. Wysocki <rafael.j.wysocki@intel.com> | 2015-12-09 22:26:12 +0100 |
| commit | f08f638b9c7f1bf3cb9006d3d26bf568d807ede0 (patch) | |
| tree | 5d2312f8f6f7f662332b53e26b9df77ca18547b1 /drivers/cpufreq | |
| parent | 70f43e5e798c8818d97d8d6a9bd4cd3235af9686 (diff) | |
| download | linux-f08f638b9c7f1bf3cb9006d3d26bf568d807ede0.tar.gz | |
cpufreq: ondemand: update update_sampling_rate() to make it more efficient
Currently update_sampling_rate() runs over each online CPU and
cancels/queues timers on all policy->cpus every time. This should be
done just once for any cpu belonging to a policy.
Create a cpumask and keep on clearing it as and when we process
policies, so that we don't have to traverse through all CPUs of the same
policy.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Diffstat (limited to 'drivers/cpufreq')
| -rw-r--r-- | drivers/cpufreq/cpufreq_ondemand.c | 12 |
1 files changed, 11 insertions, 1 deletions
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index f879012cf849..eae51070c034 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -246,6 +246,7 @@ static void update_sampling_rate(struct dbs_data *dbs_data, unsigned int new_rate) { struct od_dbs_tuners *od_tuners = dbs_data->tuners; + struct cpumask cpumask; int cpu; od_tuners->sampling_rate = new_rate = max(new_rate, @@ -256,7 +257,9 @@ static void update_sampling_rate(struct dbs_data *dbs_data, */ mutex_lock(&od_dbs_cdata.mutex); - for_each_online_cpu(cpu) { + cpumask_copy(&cpumask, cpu_online_mask); + + for_each_cpu(cpu, &cpumask) { struct cpufreq_policy *policy; struct od_cpu_dbs_info_s *dbs_info; struct cpu_dbs_info *cdbs; @@ -276,6 +279,9 @@ static void update_sampling_rate(struct dbs_data *dbs_data, policy = shared->policy; + /* clear all CPUs of this policy */ + cpumask_andnot(&cpumask, &cpumask, policy->cpus); + /* * Update sampling rate for CPUs whose policy is governed by * dbs_data. In case of governor_per_policy, only a single @@ -285,6 +291,10 @@ static void update_sampling_rate(struct dbs_data *dbs_data, if (dbs_data != policy->governor_data) continue; + /* + * Checking this for any CPU should be fine, timers for all of + * them are scheduled together. + */ next_sampling = jiffies + usecs_to_jiffies(new_rate); appointed_at = dbs_info->cdbs.timer.expires; |