ACPI: move thermal trip handling to generic thermal layer

The ACPI code currently carries its own thermal trip handling, meaning that
any other thermal implementation will need to reimplement it. Move the code
to the generic thermal layer.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>

drivers/acpi/thermal.c

@@ -37,7 +37,6 @@
 #include <linux/init.h>
 #include <linux/types.h>
 #include <linux/proc_fs.h>
-#include <linux/timer.h>
 #include <linux/jiffies.h>
 #include <linux/kmod.h>
 #include <linux/seq_file.h>
@@ -190,7 +189,6 @@ struct acpi_thermal {
 	struct acpi_thermal_state state;
 	struct acpi_thermal_trips trips;
 	struct acpi_handle_list devices;
-	struct timer_list timer;
 	struct thermal_zone_device *thermal_zone;
 	int tz_enabled;
 	struct mutex lock;
@@ -290,6 +288,11 @@ static int acpi_thermal_set_polling(struct acpi_thermal *tz, int seconds)
 
 	tz->polling_frequency = seconds * 10;	/* Convert value to deci-seconds */
 
+	tz->thermal_zone->polling_delay = seconds * 1000;
+
+	if (tz->tz_enabled)
+		thermal_zone_device_update(tz->thermal_zone);
+
 	ACPI_DEBUG_PRINT((ACPI_DB_INFO,
 			  "Polling frequency set to %lu seconds\n",
 			  tz->polling_frequency/10));
@@ -569,386 +572,11 @@ static int acpi_thermal_get_trip_points(struct acpi_thermal *tz)
 	return acpi_thermal_trips_update(tz, ACPI_TRIPS_INIT);
 }
 
-static int acpi_thermal_critical(struct acpi_thermal *tz)
-{
-	if (!tz || !tz->trips.critical.flags.valid)
-		return -EINVAL;
-
-	if (tz->temperature >= tz->trips.critical.temperature) {
-		printk(KERN_WARNING PREFIX "Critical trip point\n");
-		tz->trips.critical.flags.enabled = 1;
-	} else if (tz->trips.critical.flags.enabled)
-		tz->trips.critical.flags.enabled = 0;
-
-	acpi_bus_generate_proc_event(tz->device, ACPI_THERMAL_NOTIFY_CRITICAL,
-				tz->trips.critical.flags.enabled);
-	acpi_bus_generate_netlink_event(tz->device->pnp.device_class,
-					  dev_name(&tz->device->dev),
-					  ACPI_THERMAL_NOTIFY_CRITICAL,
-					  tz->trips.critical.flags.enabled);
-
-	/* take no action if nocrt is set */
-	if(!nocrt) {
-		printk(KERN_EMERG
-			"Critical temperature reached (%ld C), shutting down.\n",
-			KELVIN_TO_CELSIUS(tz->temperature));
-		orderly_poweroff(true);
-	}
-
-	return 0;
-}
-
-static int acpi_thermal_hot(struct acpi_thermal *tz)
-{
-	if (!tz || !tz->trips.hot.flags.valid)
-		return -EINVAL;
-
-	if (tz->temperature >= tz->trips.hot.temperature) {
-		printk(KERN_WARNING PREFIX "Hot trip point\n");
-		tz->trips.hot.flags.enabled = 1;
-	} else if (tz->trips.hot.flags.enabled)
-		tz->trips.hot.flags.enabled = 0;
-
-	acpi_bus_generate_proc_event(tz->device, ACPI_THERMAL_NOTIFY_HOT,
-				tz->trips.hot.flags.enabled);
-	acpi_bus_generate_netlink_event(tz->device->pnp.device_class,
-					  dev_name(&tz->device->dev),
-					  ACPI_THERMAL_NOTIFY_HOT,
-					  tz->trips.hot.flags.enabled);
-
-	/* TBD: Call user-mode "sleep(S4)" function if nocrt is cleared */
-
-	return 0;
-}
-
-static void acpi_thermal_passive(struct acpi_thermal *tz)
-{
-	int result = 1;
-	struct acpi_thermal_passive *passive = NULL;
-	int trend = 0;
-	int i = 0;
-
-
-	if (!tz || !tz->trips.passive.flags.valid)
-		return;
-
-	passive = &(tz->trips.passive);
-
-	/*
-	 * Above Trip?
-	 * -----------
-	 * Calculate the thermal trend (using the passive cooling equation)
-	 * and modify the performance limit for all passive cooling devices
-	 * accordingly.  Note that we assume symmetry.
-	 */
-	if (tz->temperature >= passive->temperature) {
-		trend =
-		    (passive->tc1 * (tz->temperature - tz->last_temperature)) +
-		    (passive->tc2 * (tz->temperature - passive->temperature));
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "trend[%d]=(tc1[%lu]*(tmp[%lu]-last[%lu]))+(tc2[%lu]*(tmp[%lu]-psv[%lu]))\n",
-				  trend, passive->tc1, tz->temperature,
-				  tz->last_temperature, passive->tc2,
-				  tz->temperature, passive->temperature));
-		passive->flags.enabled = 1;
-		/* Heating up? */
-		if (trend > 0)
-			for (i = 0; i < passive->devices.count; i++)
-				acpi_processor_set_thermal_limit(passive->
-								 devices.
-								 handles[i],
-								 ACPI_PROCESSOR_LIMIT_INCREMENT);
-		/* Cooling off? */
-		else if (trend < 0) {
-			for (i = 0; i < passive->devices.count; i++)
-				/*
-				 * assume that we are on highest
-				 * freq/lowest thrott and can leave
-				 * passive mode, even in error case
-				 */
-				if (!acpi_processor_set_thermal_limit
-				    (passive->devices.handles[i],
-				     ACPI_PROCESSOR_LIMIT_DECREMENT))
-					result = 0;
-			/*
-			 * Leave cooling mode, even if the temp might
-			 * higher than trip point This is because some
-			 * machines might have long thermal polling
-			 * frequencies (tsp) defined. We will fall back
-			 * into passive mode in next cycle (probably quicker)
-			 */
-			if (result) {
-				passive->flags.enabled = 0;
-				ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-						  "Disabling passive cooling, still above threshold,"
-						  " but we are cooling down\n"));
-			}
-		}
-		return;
-	}
-
-	/*
-	 * Below Trip?
-	 * -----------
-	 * Implement passive cooling hysteresis to slowly increase performance
-	 * and avoid thrashing around the passive trip point.  Note that we
-	 * assume symmetry.
-	 */
-	if (!passive->flags.enabled)
-		return;
-	for (i = 0; i < passive->devices.count; i++)
-		if (!acpi_processor_set_thermal_limit
-		    (passive->devices.handles[i],
-		     ACPI_PROCESSOR_LIMIT_DECREMENT))
-			result = 0;
-	if (result) {
-		passive->flags.enabled = 0;
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Disabling passive cooling (zone is cool)\n"));
-	}
-}
-
-static void acpi_thermal_active(struct acpi_thermal *tz)
-{
-	int result = 0;
-	struct acpi_thermal_active *active = NULL;
-	int i = 0;
-	int j = 0;
-	unsigned long maxtemp = 0;
-
-
-	if (!tz)
-		return;
-
-	for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE; i++) {
-		active = &(tz->trips.active[i]);
-		if (!active || !active->flags.valid)
-			break;
-		if (tz->temperature >= active->temperature) {
-			/*
-			 * Above Threshold?
-			 * ----------------
-			 * If not already enabled, turn ON all cooling devices
-			 * associated with this active threshold.
-			 */
-			if (active->temperature > maxtemp)
-				tz->state.active_index = i;
-			maxtemp = active->temperature;
-			if (active->flags.enabled)
-				continue;
-			for (j = 0; j < active->devices.count; j++) {
-				result =
-				    acpi_bus_set_power(active->devices.
-						       handles[j],
-						       ACPI_STATE_D0);
-				if (result) {
-					printk(KERN_WARNING PREFIX
-						      "Unable to turn cooling device [%p] 'on'\n",
-						      active->devices.
-						      handles[j]);
-					continue;
-				}
-				active->flags.enabled = 1;
-				ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-						  "Cooling device [%p] now 'on'\n",
-						  active->devices.handles[j]));
-			}
-			continue;
-		}
-		if (!active->flags.enabled)
-			continue;
-		/*
-		 * Below Threshold?
-		 * ----------------
-		 * Turn OFF all cooling devices associated with this
-		 * threshold.
-		 */
-		for (j = 0; j < active->devices.count; j++) {
-			result = acpi_bus_set_power(active->devices.handles[j],
-						    ACPI_STATE_D3);
-			if (result) {
-				printk(KERN_WARNING PREFIX
-					      "Unable to turn cooling device [%p] 'off'\n",
-					      active->devices.handles[j]);
-				continue;
-			}
-			active->flags.enabled = 0;
-			ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-					  "Cooling device [%p] now 'off'\n",
-					  active->devices.handles[j]));
-		}
-	}
-}
-
-static void acpi_thermal_check(void *context);
-
-static void acpi_thermal_run(unsigned long data)
-{
-	struct acpi_thermal *tz = (struct acpi_thermal *)data;
-	if (!tz->zombie)
-		acpi_os_execute(OSL_GPE_HANDLER, acpi_thermal_check, (void *)data);
-}
-
-static void acpi_thermal_active_off(void *data)
-{
-	int result = 0;
-	struct acpi_thermal *tz = data;
-	int i = 0;
-	int j = 0;
-	struct acpi_thermal_active *active = NULL;
-
-	if (!tz) {
-		printk(KERN_ERR PREFIX "Invalid (NULL) context\n");
-		return;
-	}
-
-	result = acpi_thermal_get_temperature(tz);
-	if (result)
-		return;
-
-	for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE; i++) {
-		active = &(tz->trips.active[i]);
-		if (!active || !active->flags.valid)
-			break;
-		if (tz->temperature >= active->temperature) {
-			/*
-			 * If the thermal temperature is greater than the
-			 * active threshod, unnecessary to turn off the
-			 * the active cooling device.
-			 */
-			continue;
-		}
-		/*
-		 * Below Threshold?
-		 * ----------------
-		 * Turn OFF all cooling devices associated with this
-		 * threshold.
-		 */
-		for (j = 0; j < active->devices.count; j++)
-			result = acpi_bus_set_power(active->devices.handles[j],
-						    ACPI_STATE_D3);
-	}
-}
-
 static void acpi_thermal_check(void *data)
 {
-	int result = 0;
 	struct acpi_thermal *tz = data;
-	unsigned long sleep_time = 0;
-	unsigned long timeout_jiffies = 0;
-	int i = 0;
-	struct acpi_thermal_state state;
-
-
-	if (!tz) {
-		printk(KERN_ERR PREFIX "Invalid (NULL) context\n");
-		return;
-	}
-
-	/* Check if someone else is already running */
-	if (!mutex_trylock(&tz->lock))
-		return;
-
-	state = tz->state;
-
-	result = acpi_thermal_get_temperature(tz);
-	if (result)
-		goto unlock;
-
-	if (!tz->tz_enabled)
-		goto unlock;
-
-	memset(&tz->state, 0, sizeof(tz->state));
-
-	/*
-	 * Check Trip Points
-	 * -----------------
-	 * Compare the current temperature to the trip point values to see
-	 * if we've entered one of the thermal policy states.  Note that
-	 * this function determines when a state is entered, but the 
-	 * individual policy decides when it is exited (e.g. hysteresis).
-	 */
-	if (tz->trips.critical.flags.valid)
-		state.critical |=
-		    (tz->temperature >= tz->trips.critical.temperature);
-	if (tz->trips.hot.flags.valid)
-		state.hot |= (tz->temperature >= tz->trips.hot.temperature);
-	if (tz->trips.passive.flags.valid)
-		state.passive |=
-		    (tz->temperature >= tz->trips.passive.temperature);
-	for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE; i++)
-		if (tz->trips.active[i].flags.valid)
-			state.active |=
-			    (tz->temperature >=
-			     tz->trips.active[i].temperature);
 
-	/*
-	 * Invoke Policy
-	 * -------------
-	 * Separated from the above check to allow individual policy to 
-	 * determine when to exit a given state.
-	 */
-	if (state.critical)
-		acpi_thermal_critical(tz);
-	if (state.hot)
-		acpi_thermal_hot(tz);
-	if (state.passive)
-		acpi_thermal_passive(tz);
-	if (state.active)
-		acpi_thermal_active(tz);
-
-	/*
-	 * Calculate State
-	 * ---------------
-	 * Again, separated from the above two to allow independent policy
-	 * decisions.
-	 */
-	tz->state.critical = tz->trips.critical.flags.enabled;
-	tz->state.hot = tz->trips.hot.flags.enabled;
-	tz->state.passive = tz->trips.passive.flags.enabled;
-	tz->state.active = 0;
-	for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE; i++)
-		tz->state.active |= tz->trips.active[i].flags.enabled;
-
-	/*
-	 * Calculate Sleep Time
-	 * --------------------
-	 * If we're in the passive state, use _TSP's value.  Otherwise
-	 * use the default polling frequency (e.g. _TZP).  If no polling
-	 * frequency is specified then we'll wait forever (at least until
-	 * a thermal event occurs).  Note that _TSP and _TZD values are
-	 * given in 1/10th seconds (we must covert to milliseconds).
-	 */
-	if (tz->state.passive) {
-		sleep_time = tz->trips.passive.tsp * 100;
-		timeout_jiffies =  jiffies + (HZ * sleep_time) / 1000;
-	} else if (tz->polling_frequency > 0) {
-		sleep_time = tz->polling_frequency * 100;
-		timeout_jiffies =  round_jiffies(jiffies + (HZ * sleep_time) / 1000);
-	}
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "%s: temperature[%lu] sleep[%lu]\n",
-			  tz->name, tz->temperature, sleep_time));
-
-	/*
-	 * Schedule Next Poll
-	 * ------------------
-	 */
-	if (!sleep_time) {
-		if (timer_pending(&(tz->timer)))
-			del_timer(&(tz->timer));
-	} else {
-		if (timer_pending(&(tz->timer)))
-			mod_timer(&(tz->timer), timeout_jiffies);
-		else {
-			tz->timer.data = (unsigned long)tz;
-			tz->timer.function = acpi_thermal_run;
-			tz->timer.expires = timeout_jiffies;
-			add_timer(&(tz->timer));
-		}
-	}
-      unlock:
-	mutex_unlock(&tz->lock);
+	thermal_zone_device_update(tz->thermal_zone);
 }
 
 /* sys I/F for generic thermal sysfs support */
@@ -1122,6 +750,29 @@ static int thermal_get_crit_temp(struct thermal_zone_device *thermal,
 		return -EINVAL;
 }
 
+static int thermal_notify(struct thermal_zone_device *thermal, int trip,
+			   enum thermal_trip_type trip_type)
+{
+	u8 type = 0;
+	struct acpi_thermal *tz = thermal->devdata;
+
+	if (trip_type == THERMAL_TRIP_CRITICAL)
+		type = ACPI_THERMAL_NOTIFY_CRITICAL;
+	else if (trip_type == THERMAL_TRIP_HOT)
+		type = ACPI_THERMAL_NOTIFY_HOT;
+	else
+		return 0;
+
+	acpi_bus_generate_proc_event(tz->device, type, 1);
+	acpi_bus_generate_netlink_event(tz->device->pnp.device_class,
+					tz->device->dev.bus_id, type, 1);
+
+	if (trip_type == THERMAL_TRIP_CRITICAL && nocrt)
+		return 1;
+
+	return 0;
+}
+
 typedef int (*cb)(struct thermal_zone_device *, int,
 		  struct thermal_cooling_device *);
 static int acpi_thermal_cooling_device_cb(struct thermal_zone_device *thermal,
@@ -1214,6 +865,7 @@ static struct thermal_zone_device_ops acpi_thermal_zone_ops = {
 	.get_trip_type = thermal_get_trip_type,
 	.get_trip_temp = thermal_get_trip_temp,
 	.get_crit_temp = thermal_get_crit_temp,
+	.notify = thermal_notify,
 };
 
 static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)
@@ -1234,8 +886,21 @@ static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz)
 
 	for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE &&
 			tz->trips.active[i].flags.valid; i++, trips++);
-	tz->thermal_zone = thermal_zone_device_register("acpitz",
-					trips, tz, &acpi_thermal_zone_ops);
+
+	if (tz->trips.passive.flags.valid)
+		tz->thermal_zone =
+			thermal_zone_device_register("acpitz", trips, tz,
+						     &acpi_thermal_zone_ops,
+						     tz->trips.passive.tc1,
+						     tz->trips.passive.tc2,
+						     tz->trips.passive.tsp*100,
+						     tz->polling_frequency*100);
+	else
+		tz->thermal_zone =
+			thermal_zone_device_register("acpitz", trips, tz,
+						     &acpi_thermal_zone_ops,
+						     0, 0, 0,
+						     tz->polling_frequency);
 	if (IS_ERR(tz->thermal_zone))
 		return -ENODEV;
 
@@ -1467,13 +1132,13 @@ static int acpi_thermal_polling_seq_show(struct seq_file *seq, void *offset)
 	if (!tz)
 		goto end;
 
-	if (!tz->polling_frequency) {
+	if (!tz->thermal_zone->polling_delay) {
 		seq_puts(seq, "<polling disabled>\n");
 		goto end;
 	}
 
-	seq_printf(seq, "polling frequency:       %lu seconds\n",
-		   (tz->polling_frequency / 10));
+	seq_printf(seq, "polling frequency:       %d seconds\n",
+		   (tz->thermal_zone->polling_delay / 1000));
 
       end:
 	return 0;
@@ -1703,12 +1368,6 @@ static int acpi_thermal_add(struct acpi_device *device)
 	if (result)
 		goto unregister_thermal_zone;
 
-	init_timer(&tz->timer);
-
-	acpi_thermal_active_off(tz);
-
-	acpi_thermal_check(tz);
-
 	status = acpi_install_notify_handler(device->handle,
 					     ACPI_DEVICE_NOTIFY,
 					     acpi_thermal_notify, tz);
@@ -1737,36 +1396,15 @@ static int acpi_thermal_remove(struct acpi_device *device, int type)
 	acpi_status status = AE_OK;
 	struct acpi_thermal *tz = NULL;
 
-
 	if (!device || !acpi_driver_data(device))
 		return -EINVAL;
 
 	tz = acpi_driver_data(device);
 
-	/* avoid timer adding new defer task */
-	tz->zombie = 1;
-	/* wait for running timer (on other CPUs) finish */
-	del_timer_sync(&(tz->timer));
-	/* synchronize deferred task */
-	acpi_os_wait_events_complete(NULL);
-	/* deferred task may reinsert timer */
-	del_timer_sync(&(tz->timer));
-
 	status = acpi_remove_notify_handler(device->handle,
 					    ACPI_DEVICE_NOTIFY,
 					    acpi_thermal_notify);
 
-	/* Terminate policy */
-	if (tz->trips.passive.flags.valid && tz->trips.passive.flags.enabled) {
-		tz->trips.passive.flags.enabled = 0;
-		acpi_thermal_passive(tz);
-	}
-	if (tz->trips.active[0].flags.valid
-	    && tz->trips.active[0].flags.enabled) {
-		tz->trips.active[0].flags.enabled = 0;
-		acpi_thermal_active(tz);
-	}
-
 	acpi_thermal_remove_fs(device);
 	acpi_thermal_unregister_thermal_zone(tz);
 	mutex_destroy(&tz->lock);

drivers/thermal/thermal_sys.c

@@ -30,6 +30,7 @@
 #include <linux/idr.h>
 #include <linux/thermal.h>
 #include <linux/spinlock.h>
+#include <linux/reboot.h>
 
 MODULE_AUTHOR("Zhang Rui");
 MODULE_DESCRIPTION("Generic thermal management sysfs support");
@@ -517,6 +518,97 @@ thermal_remove_hwmon_sysfs(struct thermal_zone_device *tz)
 }
 #endif
 
+static void thermal_zone_device_set_polling(struct thermal_zone_device *tz,
+					    int delay)
+{
+	cancel_delayed_work(&(tz->poll_queue));
+
+	if (!delay)
+		return;
+
+	if (delay > 1000)
+		schedule_delayed_work(&(tz->poll_queue),
+				      round_jiffies(msecs_to_jiffies(delay)));
+	else
+		schedule_delayed_work(&(tz->poll_queue),
+				      msecs_to_jiffies(delay));
+}
+
+static void thermal_zone_device_passive(struct thermal_zone_device *tz,
+					int temp, int trip_temp, int trip)
+{
+	int trend = 0;
+	struct thermal_cooling_device_instance *instance;
+	struct thermal_cooling_device *cdev;
+	long state, max_state;
+
+	/*
+	 * Above Trip?
+	 * -----------
+	 * Calculate the thermal trend (using the passive cooling equation)
+	 * and modify the performance limit for all passive cooling devices
+	 * accordingly.  Note that we assume symmetry.
+	 */
+	if (temp >= trip_temp) {
+		tz->passive = true;
+
+		trend = (tz->tc1 * (temp - tz->last_temperature)) +
+			(tz->tc2 * (temp - trip_temp));
+
+		/* Heating up? */
+		if (trend > 0) {
+			list_for_each_entry(instance, &tz->cooling_devices,
+					    node) {
+				if (instance->trip != trip)
+					continue;
+				cdev = instance->cdev;
+				cdev->ops->get_cur_state(cdev, &state);
+				cdev->ops->get_max_state(cdev, &max_state);
+				if (state++ < max_state)
+					cdev->ops->set_cur_state(cdev, state);
+			}
+		} else if (trend < 0) { /* Cooling off? */
+			list_for_each_entry(instance, &tz->cooling_devices,
+					    node) {
+				if (instance->trip != trip)
+					continue;
+				cdev = instance->cdev;
+				cdev->ops->get_cur_state(cdev, &state);
+				cdev->ops->get_max_state(cdev, &max_state);
+				if (state > 0)
+					cdev->ops->set_cur_state(cdev, --state);
+			}
+		}
+		return;
+	}
+
+	/*
+	 * Below Trip?
+	 * -----------
+	 * Implement passive cooling hysteresis to slowly increase performance
+	 * and avoid thrashing around the passive trip point.  Note that we
+	 * assume symmetry.
+	 */
+	list_for_each_entry(instance, &tz->cooling_devices, node) {
+		if (instance->trip != trip)
+			continue;
+		cdev = instance->cdev;
+		cdev->ops->get_cur_state(cdev, &state);
+		cdev->ops->get_max_state(cdev, &max_state);
+		if (state > 0)
+			cdev->ops->set_cur_state(cdev, --state);
+		if (state == 0)
+			tz->passive = false;
+	}
+}
+
+static void thermal_zone_device_check(struct work_struct *work)
+{
+	struct thermal_zone_device *tz = container_of(work, struct
+						      thermal_zone_device,
+						      poll_queue.work);
+	thermal_zone_device_update(tz);
+}
 
 /**
  * thermal_zone_bind_cooling_device - bind a cooling device to a thermal zone
@@ -786,21 +878,102 @@ void thermal_cooling_device_unregister(struct
 
 EXPORT_SYMBOL(thermal_cooling_device_unregister);
 
+/**
+ * thermal_zone_device_update - force an update of a thermal zone's state
+ * @ttz:	the thermal zone to update
+ */
+
+void thermal_zone_device_update(struct thermal_zone_device *tz)
+{
+	int count, ret = 0;
+	long temp, trip_temp;
+	enum thermal_trip_type trip_type;
+	struct thermal_cooling_device_instance *instance;
+	struct thermal_cooling_device *cdev;
+
+	mutex_lock(&tz->lock);
+
+	tz->ops->get_temp(tz, &temp);
+
+	for (count = 0; count < tz->trips; count++) {
+		tz->ops->get_trip_type(tz, count, &trip_type);
+		tz->ops->get_trip_temp(tz, count, &trip_temp);
+
+		switch (trip_type) {
+		case THERMAL_TRIP_CRITICAL:
+			if (temp > trip_temp) {
+				if (tz->ops->notify)
+					ret = tz->ops->notify(tz, count,
+							      trip_type);
+				if (!ret) {
+					printk(KERN_EMERG
+					       "Critical temperature reached (%ld C), shutting down.\n",
+					       temp/1000);
+					orderly_poweroff(true);
+				}
+			}
+			break;
+		case THERMAL_TRIP_HOT:
+			if (temp > trip_temp)
+				if (tz->ops->notify)
+					tz->ops->notify(tz, count, trip_type);
+			break;
+		case THERMAL_TRIP_ACTIVE:
+			list_for_each_entry(instance, &tz->cooling_devices,
+					    node) {
+				if (instance->trip != count)
+					continue;
+
+				cdev = instance->cdev;
+
+				if (temp > trip_temp)
+					cdev->ops->set_cur_state(cdev, 1);
+				else
+					cdev->ops->set_cur_state(cdev, 0);
+			}
+			break;
+		case THERMAL_TRIP_PASSIVE:
+			if (temp > trip_temp || tz->passive)
+				thermal_zone_device_passive(tz, temp,
+							    trip_temp, count);
+			break;
+		}
+	}
+	tz->last_temperature = temp;
+	if (tz->passive)
+		thermal_zone_device_set_polling(tz, tz->passive_delay);
+	else if (tz->polling_delay)
+		thermal_zone_device_set_polling(tz, tz->polling_delay);
+	mutex_unlock(&tz->lock);
+}
+EXPORT_SYMBOL(thermal_zone_device_update);
+
 /**
  * thermal_zone_device_register - register a new thermal zone device
  * @type:	the thermal zone device type
  * @trips:	the number of trip points the thermal zone support
  * @devdata:	private device data
  * @ops:	standard thermal zone device callbacks
+ * @tc1:	thermal coefficient 1 for passive calculations
+ * @tc2:	thermal coefficient 2 for passive calculations
+ * @passive_delay: number of milliseconds to wait between polls when
+ *		   performing passive cooling
+ * @polling_delay: number of milliseconds to wait between polls when checking
+ *		   whether trip points have been crossed (0 for interrupt
+ *		   driven systems)
  *
  * thermal_zone_device_unregister() must be called when the device is no
- * longer needed.
+ * longer needed. The passive cooling formula uses tc1 and tc2 as described in
+ * section 11.1.5.1 of the ACPI specification 3.0.
  */
 struct thermal_zone_device *thermal_zone_device_register(char *type,
 							 int trips,
 							 void *devdata, struct
 							 thermal_zone_device_ops
-							 *ops)
+							 *ops, int tc1, int
+							 tc2,
+							 int passive_delay,
+							 int polling_delay)
 {
 	struct thermal_zone_device *tz;
 	struct thermal_cooling_device *pos;
@@ -834,6 +1007,11 @@ struct thermal_zone_device *thermal_zone_device_register(char *type,
 	tz->device.class = &thermal_class;
 	tz->devdata = devdata;
 	tz->trips = trips;
+	tz->tc1 = tc1;
+	tz->tc2 = tc2;
+	tz->passive_delay = passive_delay;
+	tz->polling_delay = polling_delay;
+
 	dev_set_name(&tz->device, "thermal_zone%d", tz->id);
 	result = device_register(&tz->device);
 	if (result) {
@@ -879,6 +1057,10 @@ struct thermal_zone_device *thermal_zone_device_register(char *type,
 		}
 	mutex_unlock(&thermal_list_lock);
 
+	INIT_DELAYED_WORK(&(tz->poll_queue), thermal_zone_device_check);
+
+	thermal_zone_device_update(tz);
+
 	if (!result)
 		return tz;
 
@@ -918,6 +1100,8 @@ void thermal_zone_device_unregister(struct thermal_zone_device *tz)
 		    tz->ops->unbind(tz, cdev);
 	mutex_unlock(&thermal_list_lock);
 
+	thermal_zone_device_set_polling(tz, 0);
+
 	if (tz->type[0])
 		device_remove_file(&tz->device, &dev_attr_type);
 	device_remove_file(&tz->device, &dev_attr_temp);

include/linux/thermal.h

@@ -27,6 +27,7 @@
 
 #include <linux/idr.h>
 #include <linux/device.h>
+#include <linux/workqueue.h>
 
 struct thermal_zone_device;
 struct thermal_cooling_device;
@@ -58,6 +59,8 @@ struct thermal_zone_device_ops {
 	int (*get_trip_temp) (struct thermal_zone_device *, int,
 			      unsigned long *);
 	int (*get_crit_temp) (struct thermal_zone_device *, unsigned long *);
+	int (*notify) (struct thermal_zone_device *, int,
+		       enum thermal_trip_type);
 };
 
 struct thermal_cooling_device_ops {
@@ -104,11 +107,18 @@ struct thermal_zone_device {
 	struct device device;
 	void *devdata;
 	int trips;
+	int tc1;
+	int tc2;
+	int passive_delay;
+	int polling_delay;
+	int last_temperature;
+	bool passive;
 	struct thermal_zone_device_ops *ops;
 	struct list_head cooling_devices;
 	struct idr idr;
 	struct mutex lock;	/* protect cooling devices list */
 	struct list_head node;
+	struct delayed_work poll_queue;
 #if defined(CONFIG_THERMAL_HWMON)
 	struct list_head hwmon_node;
 	struct thermal_hwmon_device *hwmon;
@@ -120,13 +130,16 @@ struct thermal_zone_device {
 struct thermal_zone_device *thermal_zone_device_register(char *, int, void *,
 							 struct
 							 thermal_zone_device_ops
-							 *);
+							 *, int tc1, int tc2,
+							 int passive_freq,
+							 int polling_freq);
 void thermal_zone_device_unregister(struct thermal_zone_device *);
 
 int thermal_zone_bind_cooling_device(struct thermal_zone_device *, int,
 				     struct thermal_cooling_device *);
 int thermal_zone_unbind_cooling_device(struct thermal_zone_device *, int,
 				       struct thermal_cooling_device *);
+void thermal_zone_device_update(struct thermal_zone_device *);
 struct thermal_cooling_device *thermal_cooling_device_register(char *, void *,
 							       struct
 							       thermal_cooling_device_ops