1 files changed, 184 insertions, 107 deletions

diff --git a/drivers/cpufreq/cpufreq_governor.c b/drivers/cpufreq/cpufreq_governor.c
index 5a76086ff09b..443442df113b 100644
--- a/drivers/cpufreq/cpufreq_governor.c
+++ b/drivers/cpufreq/cpufreq_governor.c
@@ -22,12 +22,29 @@
 #include <linux/export.h>
 #include <linux/kernel_stat.h>
 #include <linux/mutex.h>
+#include <linux/slab.h>
 #include <linux/tick.h>
 #include <linux/types.h>
 #include <linux/workqueue.h>
 
 #include "cpufreq_governor.h"
 
+static struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy)
+{
+	if (have_governor_per_policy())
+		return &policy->kobj;
+	else
+		return cpufreq_global_kobject;
+}
+
+static struct attribute_group *get_sysfs_attr(struct dbs_data *dbs_data)
+{
+	if (have_governor_per_policy())
+		return dbs_data->cdata->attr_group_gov_pol;
+	else
+		return dbs_data->cdata->attr_group_gov_sys;
+}
+
 static inline u64 get_cpu_idle_time_jiffy(unsigned int cpu, u64 *wall)
 {
 	u64 idle_time;
@@ -50,13 +67,13 @@ static inline u64 get_cpu_idle_time_jiffy(unsigned int cpu, u64 *wall)
 	return cputime_to_usecs(idle_time);
 }
 
-u64 get_cpu_idle_time(unsigned int cpu, u64 *wall)
+u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy)
 {
-	u64 idle_time = get_cpu_idle_time_us(cpu, NULL);
+	u64 idle_time = get_cpu_idle_time_us(cpu, io_busy ? wall : NULL);
 
 	if (idle_time == -1ULL)
 		return get_cpu_idle_time_jiffy(cpu, wall);
-	else
+	else if (!io_busy)
 		idle_time += get_cpu_iowait_time_us(cpu, wall);
 
 	return idle_time;
@@ -65,7 +82,7 @@ EXPORT_SYMBOL_GPL(get_cpu_idle_time);
 
 void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 {
-	struct cpu_dbs_common_info *cdbs = dbs_data->get_cpu_cdbs(cpu);
+	struct cpu_dbs_common_info *cdbs = dbs_data->cdata->get_cpu_cdbs(cpu);
 	struct od_dbs_tuners *od_tuners = dbs_data->tuners;
 	struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
 	struct cpufreq_policy *policy;
@@ -73,7 +90,7 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 	unsigned int ignore_nice;
 	unsigned int j;
 
-	if (dbs_data->governor == GOV_ONDEMAND)
+	if (dbs_data->cdata->governor == GOV_ONDEMAND)
 		ignore_nice = od_tuners->ignore_nice;
 	else
 		ignore_nice = cs_tuners->ignore_nice;
@@ -83,13 +100,22 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 	/* Get Absolute Load (in terms of freq for ondemand gov) */
 	for_each_cpu(j, policy->cpus) {
 		struct cpu_dbs_common_info *j_cdbs;
-		u64 cur_wall_time, cur_idle_time, cur_iowait_time;
-		unsigned int idle_time, wall_time, iowait_time;
+		u64 cur_wall_time, cur_idle_time;
+		unsigned int idle_time, wall_time;
 		unsigned int load;
+		int io_busy = 0;
 
-		j_cdbs = dbs_data->get_cpu_cdbs(j);
+		j_cdbs = dbs_data->cdata->get_cpu_cdbs(j);
 
-		cur_idle_time = get_cpu_idle_time(j, &cur_wall_time);
+		/*
+		 * For the purpose of ondemand, waiting for disk IO is
+		 * an indication that you're performance critical, and
+		 * not that the system is actually idle. So do not add
+		 * the iowait time to the cpu idle time.
+		 */
+		if (dbs_data->cdata->governor == GOV_ONDEMAND)
+			io_busy = od_tuners->io_is_busy;
+		cur_idle_time = get_cpu_idle_time(j, &cur_wall_time, io_busy);
 
 		wall_time = (unsigned int)
 			(cur_wall_time - j_cdbs->prev_cpu_wall);
@@ -117,35 +143,12 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 			idle_time += jiffies_to_usecs(cur_nice_jiffies);
 		}
 
-		if (dbs_data->governor == GOV_ONDEMAND) {
-			struct od_cpu_dbs_info_s *od_j_dbs_info =
-				dbs_data->get_cpu_dbs_info_s(cpu);
-
-			cur_iowait_time = get_cpu_iowait_time_us(j,
-					&cur_wall_time);
-			if (cur_iowait_time == -1ULL)
-				cur_iowait_time = 0;
-
-			iowait_time = (unsigned int) (cur_iowait_time -
-					od_j_dbs_info->prev_cpu_iowait);
-			od_j_dbs_info->prev_cpu_iowait = cur_iowait_time;
-
-			/*
-			 * For the purpose of ondemand, waiting for disk IO is
-			 * an indication that you're performance critical, and
-			 * not that the system is actually idle. So subtract the
-			 * iowait time from the cpu idle time.
-			 */
-			if (od_tuners->io_is_busy && idle_time >= iowait_time)
-				idle_time -= iowait_time;
-		}
-
 		if (unlikely(!wall_time || wall_time < idle_time))
 			continue;
 
 		load = 100 * (wall_time - idle_time) / wall_time;
 
-		if (dbs_data->governor == GOV_ONDEMAND) {
+		if (dbs_data->cdata->governor == GOV_ONDEMAND) {
 			int freq_avg = __cpufreq_driver_getavg(policy, j);
 			if (freq_avg <= 0)
 				freq_avg = policy->cur;
@@ -157,24 +160,42 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu)
 			max_load = load;
 	}
 
-	dbs_data->gov_check_cpu(cpu, max_load);
+	dbs_data->cdata->gov_check_cpu(cpu, max_load);
 }
 EXPORT_SYMBOL_GPL(dbs_check_cpu);
 
-static inline void dbs_timer_init(struct dbs_data *dbs_data, int cpu,
-				  unsigned int sampling_rate)
+static inline void __gov_queue_work(int cpu, struct dbs_data *dbs_data,
+		unsigned int delay)
 {
-	int delay = delay_for_sampling_rate(sampling_rate);
-	struct cpu_dbs_common_info *cdbs = dbs_data->get_cpu_cdbs(cpu);
+	struct cpu_dbs_common_info *cdbs = dbs_data->cdata->get_cpu_cdbs(cpu);
 
-	schedule_delayed_work_on(cpu, &cdbs->work, delay);
+	mod_delayed_work_on(cpu, system_wq, &cdbs->work, delay);
 }
 
-static inline void dbs_timer_exit(struct dbs_data *dbs_data, int cpu)
+void gov_queue_work(struct dbs_data *dbs_data, struct cpufreq_policy *policy,
+		unsigned int delay, bool all_cpus)
 {
-	struct cpu_dbs_common_info *cdbs = dbs_data->get_cpu_cdbs(cpu);
+	int i;
 
-	cancel_delayed_work_sync(&cdbs->work);
+	if (!all_cpus) {
+		__gov_queue_work(smp_processor_id(), dbs_data, delay);
+	} else {
+		for_each_cpu(i, policy->cpus)
+			__gov_queue_work(i, dbs_data, delay);
+	}
+}
+EXPORT_SYMBOL_GPL(gov_queue_work);
+
+static inline void gov_cancel_work(struct dbs_data *dbs_data,
+		struct cpufreq_policy *policy)
+{
+	struct cpu_dbs_common_info *cdbs;
+	int i;
+
+	for_each_cpu(i, policy->cpus) {
+		cdbs = dbs_data->cdata->get_cpu_cdbs(i);
+		cancel_delayed_work_sync(&cdbs->work);
+	}
 }
 
 /* Will return if we need to evaluate cpu load again or not */
@@ -196,31 +217,130 @@ bool need_load_eval(struct cpu_dbs_common_info *cdbs,
 }
 EXPORT_SYMBOL_GPL(need_load_eval);
 
-int cpufreq_governor_dbs(struct dbs_data *dbs_data,
-		struct cpufreq_policy *policy, unsigned int event)
+static void set_sampling_rate(struct dbs_data *dbs_data,
+		unsigned int sampling_rate)
 {
+	if (dbs_data->cdata->governor == GOV_CONSERVATIVE) {
+		struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
+		cs_tuners->sampling_rate = sampling_rate;
+	} else {
+		struct od_dbs_tuners *od_tuners = dbs_data->tuners;
+		od_tuners->sampling_rate = sampling_rate;
+	}
+}
+
+int cpufreq_governor_dbs(struct cpufreq_policy *policy,
+		struct common_dbs_data *cdata, unsigned int event)
+{
+	struct dbs_data *dbs_data;
 	struct od_cpu_dbs_info_s *od_dbs_info = NULL;
 	struct cs_cpu_dbs_info_s *cs_dbs_info = NULL;
-	struct cs_ops *cs_ops = NULL;
 	struct od_ops *od_ops = NULL;
-	struct od_dbs_tuners *od_tuners = dbs_data->tuners;
-	struct cs_dbs_tuners *cs_tuners = dbs_data->tuners;
+	struct od_dbs_tuners *od_tuners = NULL;
+	struct cs_dbs_tuners *cs_tuners = NULL;
 	struct cpu_dbs_common_info *cpu_cdbs;
-	unsigned int *sampling_rate, latency, ignore_nice, j, cpu = policy->cpu;
+	unsigned int sampling_rate, latency, ignore_nice, j, cpu = policy->cpu;
+	int io_busy = 0;
 	int rc;
 
-	cpu_cdbs = dbs_data->get_cpu_cdbs(cpu);
+	if (have_governor_per_policy())
+		dbs_data = policy->governor_data;
+	else
+		dbs_data = cdata->gdbs_data;
+
+	WARN_ON(!dbs_data && (event != CPUFREQ_GOV_POLICY_INIT));
+
+	switch (event) {
+	case CPUFREQ_GOV_POLICY_INIT:
+		if (have_governor_per_policy()) {
+			WARN_ON(dbs_data);
+		} else if (dbs_data) {
+			policy->governor_data = dbs_data;
+			return 0;
+		}
+
+		dbs_data = kzalloc(sizeof(*dbs_data), GFP_KERNEL);
+		if (!dbs_data) {
+			pr_err("%s: POLICY_INIT: kzalloc failed\n", __func__);
+			return -ENOMEM;
+		}
+
+		dbs_data->cdata = cdata;
+		rc = cdata->init(dbs_data);
+		if (rc) {
+			pr_err("%s: POLICY_INIT: init() failed\n", __func__);
+			kfree(dbs_data);
+			return rc;
+		}
+
+		rc = sysfs_create_group(get_governor_parent_kobj(policy),
+				get_sysfs_attr(dbs_data));
+		if (rc) {
+			cdata->exit(dbs_data);
+			kfree(dbs_data);
+			return rc;
+		}
+
+		policy->governor_data = dbs_data;
 
-	if (dbs_data->governor == GOV_CONSERVATIVE) {
-		cs_dbs_info = dbs_data->get_cpu_dbs_info_s(cpu);
-		sampling_rate = &cs_tuners->sampling_rate;
+		/* policy latency is in nS. Convert it to uS first */
+		latency = policy->cpuinfo.transition_latency / 1000;
+		if (latency == 0)
+			latency = 1;
+
+		/* Bring kernel and HW constraints together */
+		dbs_data->min_sampling_rate = max(dbs_data->min_sampling_rate,
+				MIN_LATENCY_MULTIPLIER * latency);
+		set_sampling_rate(dbs_data, max(dbs_data->min_sampling_rate,
+					latency * LATENCY_MULTIPLIER));
+
+		if (dbs_data->cdata->governor == GOV_CONSERVATIVE) {
+			struct cs_ops *cs_ops = dbs_data->cdata->gov_ops;
+
+			cpufreq_register_notifier(cs_ops->notifier_block,
+					CPUFREQ_TRANSITION_NOTIFIER);
+		}
+
+		if (!have_governor_per_policy())
+			cdata->gdbs_data = dbs_data;
+
+		return 0;
+	case CPUFREQ_GOV_POLICY_EXIT:
+		if ((policy->governor->initialized == 1) ||
+				have_governor_per_policy()) {
+			sysfs_remove_group(get_governor_parent_kobj(policy),
+					get_sysfs_attr(dbs_data));
+
+			if (dbs_data->cdata->governor == GOV_CONSERVATIVE) {
+				struct cs_ops *cs_ops = dbs_data->cdata->gov_ops;
+
+				cpufreq_unregister_notifier(cs_ops->notifier_block,
+						CPUFREQ_TRANSITION_NOTIFIER);
+			}
+
+			cdata->exit(dbs_data);
+			kfree(dbs_data);
+			cdata->gdbs_data = NULL;
+		}
+
+		policy->governor_data = NULL;
+		return 0;
+	}
+
+	cpu_cdbs = dbs_data->cdata->get_cpu_cdbs(cpu);
+
+	if (dbs_data->cdata->governor == GOV_CONSERVATIVE) {
+		cs_tuners = dbs_data->tuners;
+		cs_dbs_info = dbs_data->cdata->get_cpu_dbs_info_s(cpu);
+		sampling_rate = cs_tuners->sampling_rate;
 		ignore_nice = cs_tuners->ignore_nice;
-		cs_ops = dbs_data->gov_ops;
 	} else {
-		od_dbs_info = dbs_data->get_cpu_dbs_info_s(cpu);
-		sampling_rate = &od_tuners->sampling_rate;
+		od_tuners = dbs_data->tuners;
+		od_dbs_info = dbs_data->cdata->get_cpu_dbs_info_s(cpu);
+		sampling_rate = od_tuners->sampling_rate;
 		ignore_nice = od_tuners->ignore_nice;
-		od_ops = dbs_data->gov_ops;
+		od_ops = dbs_data->cdata->gov_ops;
+		io_busy = od_tuners->io_is_busy;
 	}
 
 	switch (event) {
@@ -232,96 +352,53 @@ int cpufreq_governor_dbs(struct dbs_data *dbs_data,
 
 		for_each_cpu(j, policy->cpus) {
 			struct cpu_dbs_common_info *j_cdbs =
-				dbs_data->get_cpu_cdbs(j);
+				dbs_data->cdata->get_cpu_cdbs(j);
 
 			j_cdbs->cpu = j;
 			j_cdbs->cur_policy = policy;
 			j_cdbs->prev_cpu_idle = get_cpu_idle_time(j,
-					&j_cdbs->prev_cpu_wall);
+					       &j_cdbs->prev_cpu_wall, io_busy);
 			if (ignore_nice)
 				j_cdbs->prev_cpu_nice =
 					kcpustat_cpu(j).cpustat[CPUTIME_NICE];
 
 			mutex_init(&j_cdbs->timer_mutex);
 			INIT_DEFERRABLE_WORK(&j_cdbs->work,
-					     dbs_data->gov_dbs_timer);
-		}
-
-		if (!policy->governor->initialized) {
-			rc = sysfs_create_group(cpufreq_global_kobject,
-					dbs_data->attr_group);
-			if (rc) {
-				mutex_unlock(&dbs_data->mutex);
-				return rc;
-			}
+					     dbs_data->cdata->gov_dbs_timer);
 		}
 
 		/*
 		 * conservative does not implement micro like ondemand
 		 * governor, thus we are bound to jiffes/HZ
 		 */
-		if (dbs_data->governor == GOV_CONSERVATIVE) {
+		if (dbs_data->cdata->governor == GOV_CONSERVATIVE) {
 			cs_dbs_info->down_skip = 0;
 			cs_dbs_info->enable = 1;
 			cs_dbs_info->requested_freq = policy->cur;
-
-			if (!policy->governor->initialized) {
-				cpufreq_register_notifier(cs_ops->notifier_block,
-						CPUFREQ_TRANSITION_NOTIFIER);
-
-				dbs_data->min_sampling_rate =
-					MIN_SAMPLING_RATE_RATIO *
-					jiffies_to_usecs(10);
-			}
 		} else {
 			od_dbs_info->rate_mult = 1;
 			od_dbs_info->sample_type = OD_NORMAL_SAMPLE;
 			od_ops->powersave_bias_init_cpu(cpu);
-
-			if (!policy->governor->initialized)
-				od_tuners->io_is_busy = od_ops->io_busy();
 		}
 
-		if (policy->governor->initialized)
-			goto unlock;
-
-		/* policy latency is in nS. Convert it to uS first */
-		latency = policy->cpuinfo.transition_latency / 1000;
-		if (latency == 0)
-			latency = 1;
-
-		/* Bring kernel and HW constraints together */
-		dbs_data->min_sampling_rate = max(dbs_data->min_sampling_rate,
-				MIN_LATENCY_MULTIPLIER * latency);
-		*sampling_rate = max(dbs_data->min_sampling_rate, latency *
-				LATENCY_MULTIPLIER);
-unlock:
 		mutex_unlock(&dbs_data->mutex);
 
 		/* Initiate timer time stamp */
 		cpu_cdbs->time_stamp = ktime_get();
 
-		for_each_cpu(j, policy->cpus)
-			dbs_timer_init(dbs_data, j, *sampling_rate);
+		gov_queue_work(dbs_data, policy,
+				delay_for_sampling_rate(sampling_rate), true);
 		break;
 
 	case CPUFREQ_GOV_STOP:
-		if (dbs_data->governor == GOV_CONSERVATIVE)
+		if (dbs_data->cdata->governor == GOV_CONSERVATIVE)
 			cs_dbs_info->enable = 0;
 
-		for_each_cpu(j, policy->cpus)
-			dbs_timer_exit(dbs_data, j);
+		gov_cancel_work(dbs_data, policy);
 
 		mutex_lock(&dbs_data->mutex);
 		mutex_destroy(&cpu_cdbs->timer_mutex);
 
-		if (policy->governor->initialized == 1) {
-			sysfs_remove_group(cpufreq_global_kobject,
-					dbs_data->attr_group);
-			if (dbs_data->governor == GOV_CONSERVATIVE)
-				cpufreq_unregister_notifier(cs_ops->notifier_block,
-						CPUFREQ_TRANSITION_NOTIFIER);
-		}
 		mutex_unlock(&dbs_data->mutex);
 
 		break;