PM / Runtime: Fix pm_runtime_suspended()

There are some situations (e.g. in __pm_generic_call()), where
pm_runtime_suspended() is used to decide whether or not to execute
a device's (system) ->suspend() callback.  The callback is not
executed if pm_runtime_suspended() returns true, but it does so
for devices that don't even support runtime PM, because the
power.disable_depth device field is ignored by it.  This leads to
problems (i.e. devices are not suspened when they should), so rework
pm_runtime_suspended() so that it returns false if the device's
power.disable_depth field is different from zero.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: stable@kernel.org

Documentation/power/runtime_pm.txt

@@ -379,8 +379,8 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
       zero)
 
   bool pm_runtime_suspended(struct device *dev);
-    - return true if the device's runtime PM status is 'suspended', or false
-      otherwise
+    - return true if the device's runtime PM status is 'suspended' and its
+      'power.disable_depth' field is equal to zero, or false otherwise
 
   void pm_runtime_allow(struct device *dev);
     - set the power.runtime_auto flag for the device and decrease its usage

include/linux/pm_runtime.h

@@ -77,7 +77,8 @@ static inline void device_set_run_wake(struct device *dev, bool enable)
 
 static inline bool pm_runtime_suspended(struct device *dev)
 {
-	return dev->power.runtime_status == RPM_SUSPENDED;
+	return dev->power.runtime_status == RPM_SUSPENDED
+		&& !dev->power.disable_depth;
 }
 
 static inline void pm_runtime_mark_last_busy(struct device *dev)