cpufreq: Use per-policy frequency QoS

Replace the CPU device PM QoS used for the management of min and max
frequency constraints in cpufreq (and its users) with per-policy
frequency QoS to avoid problems with cpufreq policies covering
more then one CPU.

Namely, a cpufreq driver is registered with the subsys interface
which calls cpufreq_add_dev() for each CPU, starting from CPU0, so
currently the PM QoS notifiers are added to the first CPU in the
policy (i.e. CPU0 in the majority of cases).

In turn, when the cpufreq driver is unregistered, the subsys interface
doing that calls cpufreq_remove_dev() for each CPU, starting from CPU0,
and the PM QoS notifiers are only removed when cpufreq_remove_dev() is
called for the last CPU in the policy, say CPUx, which as a rule is
not CPU0 if the policy covers more than one CPU.  Then, the PM QoS
notifiers cannot be removed, because CPUx does not have them, and
they are still there in the device PM QoS notifiers list of CPU0,
which prevents new PM QoS notifiers from being registered for CPU0
on the next attempt to register the cpufreq driver.

The same issue occurs when the first CPU in the policy goes offline
before unregistering the driver.

After this change it does not matter which CPU is the policy CPU at
the driver registration time and whether or not it is online all the
time, because the frequency QoS is per policy and not per CPU.

Fixes: 67d874c3b2c6 ("cpufreq: Register notifiers with the PM QoS framework")
Reported-by: Dmitry Osipenko <digetx@gmail.com>
Tested-by: Dmitry Osipenko <digetx@gmail.com>
Reported-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Sudeep Holla <sudeep.holla@arm.com>
Diagnosed-by: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lore.kernel.org/linux-pm/5ad2624194baa2f53acc1f1e627eb7684c577a19.1562210705.git.viresh.kumar@linaro.org/T/#md2d89e95906b8c91c15f582146173dce2e86e99f
Link: https://lore.kernel.org/linux-pm/20191017094612.6tbkwoq4harsjcqv@vireshk-i7/T/#m30d48cc23b9a80467fbaa16e30f90b3828a5a29b
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>

3 files changed, 48 insertions, 56 deletions

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index bffc11b87247..8478ff6f3045 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -720,7 +720,7 @@ static ssize_t store_##file_name					\
 	if (ret != 1)							\
 		return -EINVAL;						\
 									\
-	ret = dev_pm_qos_update_request(policy->object##_freq_req, val);\
+	ret = freq_qos_update_request(policy->object##_freq_req, val);\
 	return ret >= 0 ? count : ret;					\
 }
 
@@ -1202,19 +1202,21 @@ static struct cpufreq_policy *cpufreq_policy_alloc(unsigned int cpu)
 		goto err_free_real_cpus;
 	}
 
+	freq_constraints_init(&policy->constraints);
+
 	policy->nb_min.notifier_call = cpufreq_notifier_min;
 	policy->nb_max.notifier_call = cpufreq_notifier_max;
 
-	ret = dev_pm_qos_add_notifier(dev, &policy->nb_min,
-				      DEV_PM_QOS_MIN_FREQUENCY);
+	ret = freq_qos_add_notifier(&policy->constraints, FREQ_QOS_MIN,
+				    &policy->nb_min);
 	if (ret) {
 		dev_err(dev, "Failed to register MIN QoS notifier: %d (%*pbl)\n",
 			ret, cpumask_pr_args(policy->cpus));
 		goto err_kobj_remove;
 	}
 
-	ret = dev_pm_qos_add_notifier(dev, &policy->nb_max,
-				      DEV_PM_QOS_MAX_FREQUENCY);
+	ret = freq_qos_add_notifier(&policy->constraints, FREQ_QOS_MAX,
+				    &policy->nb_max);
 	if (ret) {
 		dev_err(dev, "Failed to register MAX QoS notifier: %d (%*pbl)\n",
 			ret, cpumask_pr_args(policy->cpus));
@@ -1232,8 +1234,8 @@ static struct cpufreq_policy *cpufreq_policy_alloc(unsigned int cpu)
 	return policy;
 
 err_min_qos_notifier:
-	dev_pm_qos_remove_notifier(dev, &policy->nb_min,
-				   DEV_PM_QOS_MIN_FREQUENCY);
+	freq_qos_remove_notifier(&policy->constraints, FREQ_QOS_MIN,
+				 &policy->nb_min);
 err_kobj_remove:
 	cpufreq_policy_put_kobj(policy);
 err_free_real_cpus:
@@ -1250,7 +1252,6 @@ err_free_policy:
 
 static void cpufreq_policy_free(struct cpufreq_policy *policy)
 {
-	struct device *dev = get_cpu_device(policy->cpu);
 	unsigned long flags;
 	int cpu;
 
@@ -1262,10 +1263,10 @@ static void cpufreq_policy_free(struct cpufreq_policy *policy)
 		per_cpu(cpufreq_cpu_data, cpu) = NULL;
 	write_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
-	dev_pm_qos_remove_notifier(dev, &policy->nb_max,
-				   DEV_PM_QOS_MAX_FREQUENCY);
-	dev_pm_qos_remove_notifier(dev, &policy->nb_min,
-				   DEV_PM_QOS_MIN_FREQUENCY);
+	freq_qos_remove_notifier(&policy->constraints, FREQ_QOS_MAX,
+				 &policy->nb_max);
+	freq_qos_remove_notifier(&policy->constraints, FREQ_QOS_MIN,
+				 &policy->nb_min);
 
 	if (policy->max_freq_req) {
 		/*
@@ -1274,10 +1275,10 @@ static void cpufreq_policy_free(struct cpufreq_policy *policy)
 		 */
 		blocking_notifier_call_chain(&cpufreq_policy_notifier_list,
 					     CPUFREQ_REMOVE_POLICY, policy);
-		dev_pm_qos_remove_request(policy->max_freq_req);
+		freq_qos_remove_request(policy->max_freq_req);
 	}
 
-	dev_pm_qos_remove_request(policy->min_freq_req);
+	freq_qos_remove_request(policy->min_freq_req);
 	kfree(policy->min_freq_req);
 
 	cpufreq_policy_put_kobj(policy);
@@ -1357,8 +1358,6 @@ static int cpufreq_online(unsigned int cpu)
 	cpumask_and(policy->cpus, policy->cpus, cpu_online_mask);
 
 	if (new_policy) {
-		struct device *dev = get_cpu_device(cpu);
-
 		for_each_cpu(j, policy->related_cpus) {
 			per_cpu(cpufreq_cpu_data, j) = policy;
 			add_cpu_dev_symlink(policy, j);
@@ -1369,36 +1368,31 @@ static int cpufreq_online(unsigned int cpu)
 		if (!policy->min_freq_req)
 			goto out_destroy_policy;
 
-		ret = dev_pm_qos_add_request(dev, policy->min_freq_req,
-					     DEV_PM_QOS_MIN_FREQUENCY,
-					     policy->min);
+		ret = freq_qos_add_request(&policy->constraints,
+					   policy->min_freq_req, FREQ_QOS_MIN,
+					   policy->min);
 		if (ret < 0) {
 			/*
-			 * So we don't call dev_pm_qos_remove_request() for an
+			 * So we don't call freq_qos_remove_request() for an
 			 * uninitialized request.
 			 */
 			kfree(policy->min_freq_req);
 			policy->min_freq_req = NULL;
-
-			dev_err(dev, "Failed to add min-freq constraint (%d)\n",
-				ret);
 			goto out_destroy_policy;
 		}
 
 		/*
 		 * This must be initialized right here to avoid calling
-		 * dev_pm_qos_remove_request() on uninitialized request in case
+		 * freq_qos_remove_request() on uninitialized request in case
 		 * of errors.
 		 */
 		policy->max_freq_req = policy->min_freq_req + 1;
 
-		ret = dev_pm_qos_add_request(dev, policy->max_freq_req,
-					     DEV_PM_QOS_MAX_FREQUENCY,
-					     policy->max);
+		ret = freq_qos_add_request(&policy->constraints,
+					   policy->max_freq_req, FREQ_QOS_MAX,
+					   policy->max);
 		if (ret < 0) {
 			policy->max_freq_req = NULL;
-			dev_err(dev, "Failed to add max-freq constraint (%d)\n",
-				ret);
 			goto out_destroy_policy;
 		}
 
@@ -2374,7 +2368,6 @@ int cpufreq_set_policy(struct cpufreq_policy *policy,
 		       struct cpufreq_policy *new_policy)
 {
 	struct cpufreq_governor *old_gov;
-	struct device *cpu_dev = get_cpu_device(policy->cpu);
 	int ret;
 
 	pr_debug("setting new policy for CPU %u: %u - %u kHz\n",
@@ -2386,8 +2379,8 @@ int cpufreq_set_policy(struct cpufreq_policy *policy,
 	 * PM QoS framework collects all the requests from users and provide us
 	 * the final aggregated value here.
 	 */
-	new_policy->min = dev_pm_qos_read_value(cpu_dev, DEV_PM_QOS_MIN_FREQUENCY);
-	new_policy->max = dev_pm_qos_read_value(cpu_dev, DEV_PM_QOS_MAX_FREQUENCY);
+	new_policy->min = freq_qos_read_value(&policy->constraints, FREQ_QOS_MIN);
+	new_policy->max = freq_qos_read_value(&policy->constraints, FREQ_QOS_MAX);
 
 	/* verify the cpu speed can be set within this limit */
 	ret = cpufreq_driver->verify(new_policy);
@@ -2518,7 +2511,7 @@ static int cpufreq_boost_set_sw(int state)
 			break;
 		}
 
-		ret = dev_pm_qos_update_request(policy->max_freq_req, policy->max);
+		ret = freq_qos_update_request(policy->max_freq_req, policy->max);
 		if (ret < 0)
 			break;
 	}
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index 9f02de9a1b47..53a51c169451 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -1088,10 +1088,10 @@ static ssize_t store_no_turbo(struct kobject *a, struct kobj_attribute *b,
 
 static struct cpufreq_driver intel_pstate;
 
-static void update_qos_request(enum dev_pm_qos_req_type type)
+static void update_qos_request(enum freq_qos_req_type type)
 {
 	int max_state, turbo_max, freq, i, perf_pct;
-	struct dev_pm_qos_request *req;
+	struct freq_qos_request *req;
 	struct cpufreq_policy *policy;
 
 	for_each_possible_cpu(i) {
@@ -1112,7 +1112,7 @@ static void update_qos_request(enum dev_pm_qos_req_type type)
 		else
 			turbo_max = cpu->pstate.turbo_pstate;
 
-		if (type == DEV_PM_QOS_MIN_FREQUENCY) {
+		if (type == FREQ_QOS_MIN) {
 			perf_pct = global.min_perf_pct;
 		} else {
 			req++;
@@ -1122,7 +1122,7 @@ static void update_qos_request(enum dev_pm_qos_req_type type)
 		freq = DIV_ROUND_UP(turbo_max * perf_pct, 100);
 		freq *= cpu->pstate.scaling;
 
-		if (dev_pm_qos_update_request(req, freq) < 0)
+		if (freq_qos_update_request(req, freq) < 0)
 			pr_warn("Failed to update freq constraint: CPU%d\n", i);
 	}
 }
@@ -1153,7 +1153,7 @@ static ssize_t store_max_perf_pct(struct kobject *a, struct kobj_attribute *b,
 	if (intel_pstate_driver == &intel_pstate)
 		intel_pstate_update_policies();
 	else
-		update_qos_request(DEV_PM_QOS_MAX_FREQUENCY);
+		update_qos_request(FREQ_QOS_MAX);
 
 	mutex_unlock(&intel_pstate_driver_lock);
 
@@ -1187,7 +1187,7 @@ static ssize_t store_min_perf_pct(struct kobject *a, struct kobj_attribute *b,
 	if (intel_pstate_driver == &intel_pstate)
 		intel_pstate_update_policies();
 	else
-		update_qos_request(DEV_PM_QOS_MIN_FREQUENCY);
+		update_qos_request(FREQ_QOS_MIN);
 
 	mutex_unlock(&intel_pstate_driver_lock);
 
@@ -2381,7 +2381,7 @@ static unsigned int intel_cpufreq_fast_switch(struct cpufreq_policy *policy,
 static int intel_cpufreq_cpu_init(struct cpufreq_policy *policy)
 {
 	int max_state, turbo_max, min_freq, max_freq, ret;
-	struct dev_pm_qos_request *req;
+	struct freq_qos_request *req;
 	struct cpudata *cpu;
 	struct device *dev;
 
@@ -2416,15 +2416,15 @@ static int intel_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	max_freq = DIV_ROUND_UP(turbo_max * global.max_perf_pct, 100);
 	max_freq *= cpu->pstate.scaling;
 
-	ret = dev_pm_qos_add_request(dev, req, DEV_PM_QOS_MIN_FREQUENCY,
-				     min_freq);
+	ret = freq_qos_add_request(&policy->constraints, req, FREQ_QOS_MIN,
+				   min_freq);
 	if (ret < 0) {
 		dev_err(dev, "Failed to add min-freq constraint (%d)\n", ret);
 		goto free_req;
 	}
 
-	ret = dev_pm_qos_add_request(dev, req + 1, DEV_PM_QOS_MAX_FREQUENCY,
-				     max_freq);
+	ret = freq_qos_add_request(&policy->constraints, req + 1, FREQ_QOS_MAX,
+				   max_freq);
 	if (ret < 0) {
 		dev_err(dev, "Failed to add max-freq constraint (%d)\n", ret);
 		goto remove_min_req;
@@ -2435,7 +2435,7 @@ static int intel_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	return 0;
 
 remove_min_req:
-	dev_pm_qos_remove_request(req);
+	freq_qos_remove_request(req);
 free_req:
 	kfree(req);
 pstate_exit:
@@ -2446,12 +2446,12 @@ pstate_exit:
 
 static int intel_cpufreq_cpu_exit(struct cpufreq_policy *policy)
 {
-	struct dev_pm_qos_request *req;
+	struct freq_qos_request *req;
 
 	req = policy->driver_data;
 
-	dev_pm_qos_remove_request(req + 1);
-	dev_pm_qos_remove_request(req);
+	freq_qos_remove_request(req + 1);
+	freq_qos_remove_request(req);
 	kfree(req);
 
 	return intel_pstate_cpu_exit(policy);
diff --git a/drivers/cpufreq/ppc_cbe_cpufreq_pmi.c b/drivers/cpufreq/ppc_cbe_cpufreq_pmi.c
index bc9dd30395c4..037fe23bc6ed 100644
--- a/drivers/cpufreq/ppc_cbe_cpufreq_pmi.c
+++ b/drivers/cpufreq/ppc_cbe_cpufreq_pmi.c
@@ -65,7 +65,7 @@ EXPORT_SYMBOL_GPL(cbe_cpufreq_set_pmode_pmi);
 static void cbe_cpufreq_handle_pmi(pmi_message_t pmi_msg)
 {
 	struct cpufreq_policy *policy;
-	struct dev_pm_qos_request *req;
+	struct freq_qos_request *req;
 	u8 node, slow_mode;
 	int cpu, ret;
 
@@ -86,7 +86,7 @@ static void cbe_cpufreq_handle_pmi(pmi_message_t pmi_msg)
 
 	req = policy->driver_data;
 
-	ret = dev_pm_qos_update_request(req,
+	ret = freq_qos_update_request(req,
 			policy->freq_table[slow_mode].frequency);
 	if (ret < 0)
 		pr_warn("Failed to update freq constraint: %d\n", ret);
@@ -103,7 +103,7 @@ static struct pmi_handler cbe_pmi_handler = {
 
 void cbe_cpufreq_pmi_policy_init(struct cpufreq_policy *policy)
 {
-	struct dev_pm_qos_request *req;
+	struct freq_qos_request *req;
 	int ret;
 
 	if (!cbe_cpufreq_has_pmi)
@@ -113,9 +113,8 @@ void cbe_cpufreq_pmi_policy_init(struct cpufreq_policy *policy)
 	if (!req)
 		return;
 
-	ret = dev_pm_qos_add_request(get_cpu_device(policy->cpu), req,
-				     DEV_PM_QOS_MAX_FREQUENCY,
-				     policy->freq_table[0].frequency);
+	ret = freq_qos_add_request(&policy->constraints, req, FREQ_QOS_MAX,
+				   policy->freq_table[0].frequency);
 	if (ret < 0) {
 		pr_err("Failed to add freq constraint (%d)\n", ret);
 		kfree(req);
@@ -128,10 +127,10 @@ EXPORT_SYMBOL_GPL(cbe_cpufreq_pmi_policy_init);
 
 void cbe_cpufreq_pmi_policy_exit(struct cpufreq_policy *policy)
 {
-	struct dev_pm_qos_request *req = policy->driver_data;
+	struct freq_qos_request *req = policy->driver_data;
 
 	if (cbe_cpufreq_has_pmi) {
-		dev_pm_qos_remove_request(req);
+		freq_qos_remove_request(req);
 		kfree(req);
 	}
 }