thermal: Fix potential deadlock in cpu_cooling

cooling_list_lock is covering not just cpufreq_dev_count, but also the
calls to cpufreq_register_notifier() and cpufreq_unregister_notifier().

Since cooling_list_lock is also used within cpufreq_thermal_notifier(),
lockdep reports a potential deadlock. Fix it by testing the condition
under cooling_list_lock and dropping the lock before calling
cpufreq_register_notifier(). And variable cpufreq_dev_count is removed
at the same time, because it's no longer needed after the fix.

Fixes: ae606089621e ("thermal: convert cpu_cooling to use an IDA")
Reported-and-Tested-by: Russell King <linux@armlinux.org.uk>
Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com
Signed-off-by: Zhang Rui <rui.zhang@intel.com>

diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
index 91048eeca28b2dc3d81f234ed60bbfdd796f8277..11b5ea685518f89f3812583f4c1487fa5d5ccf6f 100644 (file)
--- a/drivers/thermal/cpu_cooling.c
+++ b/drivers/thermal/cpu_cooling.c
@@ -107,8 +107,6 @@ struct cpufreq_cooling_device {
 };
 static DEFINE_IDA(cpufreq_ida);
 
-static unsigned int cpufreq_dev_count;
-
 static DEFINE_MUTEX(cooling_list_lock);
 static LIST_HEAD(cpufreq_dev_list);
 
@@ -771,6 +769,7 @@ __cpufreq_cooling_register(struct device_node *np,
        unsigned int freq, i, num_cpus;
        int ret;
        struct thermal_cooling_device_ops *cooling_ops;
+       bool first;
 
        if (!alloc_cpumask_var(&temp_mask, GFP_KERNEL))
                return ERR_PTR(-ENOMEM);
@@ -874,13 +873,14 @@ __cpufreq_cooling_register(struct device_node *np,
        cpufreq_dev->cool_dev = cool_dev;
 
        mutex_lock(&cooling_list_lock);
+       /* Register the notifier for first cpufreq cooling device */
+       first = list_empty(&cpufreq_dev_list);
        list_add(&cpufreq_dev->node, &cpufreq_dev_list);
+       mutex_unlock(&cooling_list_lock);
 
-       /* Register the notifier for first cpufreq cooling device */
-       if (!cpufreq_dev_count++)
+       if (first)
                cpufreq_register_notifier(&thermal_cpufreq_notifier_block,
                                          CPUFREQ_POLICY_NOTIFIER);
-       mutex_unlock(&cooling_list_lock);
 
        goto put_policy;
 
@@ -1021,6 +1021,7 @@ EXPORT_SYMBOL(of_cpufreq_power_cooling_register);
 void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
 {
        struct cpufreq_cooling_device *cpufreq_dev;
+       bool last;
 
        if (!cdev)
                return;
@@ -1028,14 +1029,15 @@ void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
        cpufreq_dev = cdev->devdata;
 
        mutex_lock(&cooling_list_lock);
+       list_del(&cpufreq_dev->node);
        /* Unregister the notifier for the last cpufreq cooling device */
-       if (!--cpufreq_dev_count)
+       last = list_empty(&cpufreq_dev_list);
+       mutex_unlock(&cooling_list_lock);
+
+       if (last)
                cpufreq_unregister_notifier(&thermal_cpufreq_notifier_block,
                                            CPUFREQ_POLICY_NOTIFIER);
 
-       list_del(&cpufreq_dev->node);
-       mutex_unlock(&cooling_list_lock);
-
        thermal_cooling_device_unregister(cpufreq_dev->cool_dev);
        ida_simple_remove(&cpufreq_ida, cpufreq_dev->id);
        kfree(cpufreq_dev->dyn_power_table);