Skip to contentSkip to conversation

cpufreq: Locking-related changes in cpufreq_offline() and cpufreq_remove_dev()

PATCH In review Rafael J. Wysocki 2 participants 7 messages May 11, 2022 – May 12, 2022 CPU frequency scaling framework

lore
Share & tools
lore
#0 lore raw

cover cpufreq: Locking-related changes in cpufreq_offline() and cpufreq_remove_dev()

Hi,

This series is based on the observation that the policy rwsem is used (or rather not used) inconsistently in cpufreq_remove_dev() (in summary, it does things without holding the policy rwsem that are done under that rwsem elsewhere).

The first two patches are preparatory (but patch [1/3] is a good enough improvement by itself IMO) and patch [3/3] makes the essential change.

Please refer to the patch changelogs for details.

Thanks!

#1 lore raw

PATCH 1/3 cpufreq: Reorganize checks in cpufreq_offline()

From: Rafael J. Wysocki <[email protected]>

Notice that cpufreq_offline() only needs to check policy_is_inactive() once and rearrange the code in there to make that happen.

No expected functional impact.

---
 drivers/cpufreq/cpufreq.c |   24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)
drivers/cpufreq/cpufreq.c +12 -12
Index: linux-pm/drivers/cpufreq/cpufreq.c
===================================================================
@@ -1590,24 +1590,18 @@ static int cpufreq_offline(unsigned int
 	}
 
 	down_write(&policy->rwsem);
+
 	if (has_target())
 		cpufreq_stop_governor(policy);
 
 	cpumask_clear_cpu(cpu, policy->cpus);
 
-	if (policy_is_inactive(policy)) {
-		if (has_target())
-			strncpy(policy->last_governor, policy->governor->name,
-				CPUFREQ_NAME_LEN);
-		else
-			policy->last_policy = policy->policy;
-	} else if (cpu == policy->cpu) {
-		/* Nominate new CPU */
-		policy->cpu = cpumask_any(policy->cpus);
-	}
-
-	/* Start governor again for active policy */
 	if (!policy_is_inactive(policy)) {
+		/* Nominate a new CPU if necessary. */
+		if (cpu == policy->cpu)
+			policy->cpu = cpumask_any(policy->cpus);
+
+		/* Start the governor again for the active policy. */
 		if (has_target()) {
 			ret = cpufreq_start_governor(policy);
 			if (ret)
@@ -1617,6 +1611,12 @@ static int cpufreq_offline(unsigned int
 		goto unlock;
 	}
 
+	if (has_target())
+		strncpy(policy->last_governor, policy->governor->name,
+			CPUFREQ_NAME_LEN);
+	else
+		policy->last_policy = policy->policy;
+
 	if (cpufreq_thermal_control_enabled(cpufreq_driver)) {
 		cpufreq_cooling_unregister(policy->cdev);
 		policy->cdev = NULL;
maintainer Acked-by on v1 1/3
#2 lore raw

Re: [PATCH v1 1/3] cpufreq: Reorganize checks in cpufreq_offline()

Quoted 11 lines · Rafael J. Wysocki

From: Rafael J. Wysocki <[email protected]>

Notice that cpufreq_offline() only needs to check policy_is_inactive() once and rearrange the code in there to make that happen.

No expected functional impact.

---
 drivers/cpufreq/cpufreq.c |   24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)
Signature · 1 line
viresh
#3 lore raw

PATCH 2/3 cpufreq: Split cpufreq_offline()

In reply to Rafael J. Wysocki

From: Rafael J. Wysocki <[email protected]>

Split the "core" part running under the policy rwsem out of cpufreq_offline() to allow the locking in cpufreq_remove_dev() to be rearranged more easily.

As a side-effect this eliminates the unlock label that's not needed any more.

No expected functional impact.

---
 drivers/cpufreq/cpufreq.c |   33 +++++++++++++++++++--------------
 1 file changed, 19 insertions(+), 14 deletions(-)
drivers/cpufreq/cpufreq.c +19 -14
Index: linux-pm/drivers/cpufreq/cpufreq.c
===================================================================
@@ -1576,21 +1576,10 @@ static int cpufreq_add_dev(struct device
 	return 0;
 }
 
-static int cpufreq_offline(unsigned int cpu)
+static void __cpufreq_offline(unsigned int cpu, struct cpufreq_policy *policy)
 {
-	struct cpufreq_policy *policy;
 	int ret;
 
-	pr_debug("%s: unregistering CPU %u\n", __func__, cpu);
-
-	policy = cpufreq_cpu_get_raw(cpu);
-	if (!policy) {
-		pr_debug("%s: No cpu_data found\n", __func__);
-		return 0;
-	}
-
-	down_write(&policy->rwsem);
-
 	if (has_target())
 		cpufreq_stop_governor(policy);
 
@@ -1608,7 +1597,7 @@ static int cpufreq_offline(unsigned int
 				pr_err("%s: Failed to start governor\n", __func__);
 		}
 
-		goto unlock;
+		return;
 	}
 
 	if (has_target())
@@ -1635,8 +1624,24 @@ static int cpufreq_offline(unsigned int
 		cpufreq_driver->exit(policy);
 		policy->freq_table = NULL;
 	}
+}
+
+static int cpufreq_offline(unsigned int cpu)
+{
+	struct cpufreq_policy *policy;
+
+	pr_debug("%s: unregistering CPU %u\n", __func__, cpu);
+
+	policy = cpufreq_cpu_get_raw(cpu);
+	if (!policy) {
+		pr_debug("%s: No cpu_data found\n", __func__);
+		return 0;
+	}
+
+	down_write(&policy->rwsem);
+
+	__cpufreq_offline(cpu, policy);
 
-unlock:
 	up_write(&policy->rwsem);
 	return 0;
 }
maintainer Acked-by on v1 2/3
#4 lore raw

Re: [PATCH v1 2/3] cpufreq: Split cpufreq_offline()

Quoted 9 lines · Rafael J. Wysocki

From: Rafael J. Wysocki <[email protected]>

Split the "core" part running under the policy rwsem out of cpufreq_offline() to allow the locking in cpufreq_remove_dev() to be rearranged more easily.

As a side-effect this eliminates the unlock label that's not needed any more.

Quoted 6 lines

No expected functional impact.

---
 drivers/cpufreq/cpufreq.c |   33 +++++++++++++++++++--------------
 1 file changed, 19 insertions(+), 14 deletions(-)
Signature · 1 line
viresh
#5 lore raw

PATCH 3/3 cpufreq: Rearrange locking in cpufreq_remove_dev()

In reply to Rafael J. Wysocki

From: Rafael J. Wysocki <[email protected]>

Currently, cpufreq_remove_dev() invokes the ->exit() driver callback without holding the policy rwsem which is inconsistent with what happens if ->exit() is invoked directly from cpufreq_offline().

It also manipulates the real_cpus mask and removes the CPU device symlink without holding the policy rwsem, but cpufreq_offline() holds the rwsem around the modifications thereof.

For consistency, modify cpufreq_remove_dev() to hold the policy rwsem until the ->exit() callback has been called (or it has been determined that it is not necessary to call it).

---
 drivers/cpufreq/cpufreq.c |   21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
drivers/cpufreq/cpufreq.c +14 -7
Index: linux-pm/drivers/cpufreq/cpufreq.c
===================================================================
@@ -1659,19 +1659,26 @@ static void cpufreq_remove_dev(struct de
 	if (!policy)
 		return;
 
+	down_write(&policy->rwsem);
+
 	if (cpu_online(cpu))
-		cpufreq_offline(cpu);
+		__cpufreq_offline(cpu, policy);
 
 	cpumask_clear_cpu(cpu, policy->real_cpus);
 	remove_cpu_dev_symlink(policy, dev);
 
-	if (cpumask_empty(policy->real_cpus)) {
-		/* We did light-weight exit earlier, do full tear down now */
-		if (cpufreq_driver->offline)
-			cpufreq_driver->exit(policy);
-
-		cpufreq_policy_free(policy);
+	if (!cpumask_empty(policy->real_cpus)) {
+		up_write(&policy->rwsem);
+		return;
 	}
+
+	/* We did light-weight exit earlier, do full tear down now */
+	if (cpufreq_driver->offline)
+		cpufreq_driver->exit(policy);
+
+	up_write(&policy->rwsem);
+
+	cpufreq_policy_free(policy);
 }
 
 /**
maintainer Acked-by on v1 3/3
#6 lore raw

Re: [PATCH v1 3/3] cpufreq: Rearrange locking in cpufreq_remove_dev()

Quoted 10 lines · Rafael J. Wysocki

From: Rafael J. Wysocki <[email protected]>

Currently, cpufreq_remove_dev() invokes the ->exit() driver callback without holding the policy rwsem which is inconsistent with what happens if ->exit() is invoked directly from cpufreq_offline().

It also manipulates the real_cpus mask and removes the CPU device symlink without holding the policy rwsem, but cpufreq_offline() holds the rwsem around the modifications thereof.

Quoted 8 lines

For consistency, modify cpufreq_remove_dev() to hold the policy rwsem until the ->exit() callback has been called (or it has been determined that it is not necessary to call it).

---
 drivers/cpufreq/cpufreq.c |   21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
Signature · 1 line
viresh