蜗窝科技

linux thermal framework(4)_thermal governor

作者：huowj 发布于：2025-4-14 10:45 分类：电源管理子系统

1. 介绍

thermal governor是通过一定算法控制cooling device状态来控温的在这篇文章中，我们使用一个简单的step_wise governor来说明整个过程。

2. thermal governor相关的API以及功能分析

2.1 struct thermal_governor

struct thermal_governor {
	const char *name;
	int (*bind_to_tz)(struct thermal_zone_device *tz);
	void (*unbind_from_tz)(struct thermal_zone_device *tz);
	void (*trip_crossed)(struct thermal_zone_device *tz,
			     const struct thermal_trip *trip,
			     bool crossed_up);
	void (*manage)(struct thermal_zone_device *tz);
	void (*update_tz)(struct thermal_zone_device *tz,
			  enum thermal_notify_event reason);
	struct list_head	governor_list;
};

2.2 thermal_governor的注册

linux在lds中静态定义了一个governor_thermal_table,每个thermal governor会在这个table中添加一个entry，以step_wise为例：

#ifdef CONFIG_THERMAL
#define THERMAL_TABLE(name)						\
	. = ALIGN(8);							\
	BOUNDED_SECTION_POST_LABEL(__##name##_thermal_table,		\
				   __##name##_thermal_table,, _end)
#else
#define THERMAL_TABLE(name)
#endif
static struct thermal_governor thermal_gov_step_wise = {
	.name	= "step_wise",
	.manage	= step_wise_manage,
};
THERMAL_GOVERNOR_DECLARE(thermal_gov_step_wise);

thermal core在初始化的过程中，会遍历governor_thermal_table中所有的entry，即所有的governor，将其加进governor_list中，比较这个governor和DEFAULT_THERMAL_GOVERNOR名字是否想同，相同的话，就将系统默认的governor设置为这个governor，遍历所有的thermal_zone，如果这个governor的名字和thermal zone本身有的governor名字相同，则会设置thermal zone的governor

对governor来说，主要的回调函数就是manage，manage会在thermal zone拥有的monitor每次check thermal zone的温度的时候调用该thermal zone对应的governor->manage函数，我们可以看下step_wise的manage函数是怎么调温的

2.3 step_wise温控算法

首先我们还是用这张图结合stepwise的算法来描述以下过程：

static void step_wise_manage(struct thermal_zone_device *tz)
{
	const struct thermal_trip_desc *td;
	struct thermal_instance *instance;

	lockdep_assert_held(&tz->lock);

	/*
	 * Throttling Logic: Use the trend of the thermal zone to throttle.
	 * If the thermal zone is 'heating up', throttle all of the cooling
	 * devices associated with each trip point by one step. If the zone
	 * is 'cooling down', it brings back the performance of the devices
	 * by one step.
	 */
	for_each_trip_desc(tz, td) {
		const struct thermal_trip *trip = &td->trip;

		if (trip->temperature == THERMAL_TEMP_INVALID ||
		    trip->type == THERMAL_TRIP_CRITICAL ||
		    trip->type == THERMAL_TRIP_HOT)
			continue;

		thermal_zone_trip_update(tz, td, td->threshold);
	}

	for_each_trip_desc(tz, td) {
		list_for_each_entry(instance, &td->thermal_instances, trip_node)
			thermal_cdev_update(instance->cdev);
	}
}

1）首先遍历这个thermal zone所有的trip:for_each_trip_desc,过滤掉其中无效或者档位过高的trip，然后调用thermal_zone_trip_update这个函数来更新trip信息，为选择最佳cooling device state做准备
2）在做好更新后，重新遍历trip，thermal_cdev_update->__thermal_cdev_update->thermal_cdev_set_cur_state使state生效

static void thermal_zone_trip_update(struct thermal_zone_device *tz,
				     const struct thermal_trip_desc *td,
				     int trip_threshold)
{
	const struct thermal_trip *trip = &td->trip;
	enum thermal_trend trend = get_tz_trend(tz, trip);
	int trip_id = thermal_zone_trip_id(tz, trip);
	struct thermal_instance *instance;
	bool throttle = false;
if (tz->temperature >= trip_threshold) {
	throttle = true;
	trace_thermal_zone_trip(tz, trip_id, trip->type);
}

dev_dbg(&tz->device, "Trip%d[type=%d,temp=%d]:trend=%d,throttle=%d\n",
	trip_id, trip->type, trip_threshold, trend, throttle);

list_for_each_entry(instance, &td->thermal_instances, trip_node) {
	int old_target;

	old_target = instance->target;
	instance->target = get_target_state(instance, trend, throttle);

	dev_dbg(&instance->cdev->device, "old_target=%d, target=%ld\n",
		old_target, instance->target);

	if (instance->initialized && old_target == instance->target)
		continue;

	instance->initialized = true;
scoped_guard(cooling_dev, instance->cdev) {
		instance->cdev->updated = false; /* cdev needs update */
	}
}

}

1）trend = get_tz_trend，首先得到温度趋势，上升：THERMAL_TREND_RAISING or 下降：THERMAL_TREND_DROPPING
2）throttle：是否达到trip限制的温度
3）遍历这个trip所有的thermal instances，调用get_target_state获得目标状态，如果目标状态和原有状态不同，就设置cooling device的update参数为false，在后面的函数中更新为target state

get_target_state是获取目标状态的主函数：

/*
 * If the temperature is higher than a trip point,
 *    a. if the trend is THERMAL_TREND_RAISING, use higher cooling
 *       state for this trip point
 *    b. if the trend is THERMAL_TREND_DROPPING, do nothing
 * If the temperature is lower than a trip point,
 *    a. if the trend is THERMAL_TREND_RAISING, do nothing
 *    b. if the trend is THERMAL_TREND_DROPPING, use lower cooling
 *       state for this trip point, if the cooling state already
 *       equals lower limit, deactivate the thermal instance
 */
static unsigned long get_target_state(struct thermal_instance *instance,
				enum thermal_trend trend, bool throttle)
{
	struct thermal_cooling_device *cdev = instance->cdev;
	unsigned long cur_state;
/*
 * We keep this instance the way it is by default.
 * Otherwise, we use the current state of the
 * cdev in use to determine the next_target.
 */
cdev->ops->get_cur_state(cdev, &cur_state);
dev_dbg(&cdev->device, "cur_state=%ld\n", cur_state);

if (!instance->initialized) {
	if (throttle)
		return clamp(cur_state + 1, instance->lower, instance->upper);

	return THERMAL_NO_TARGET;
}

if (throttle) {
if (trend == THERMAL_TREND_RAISING)
return clamp(cur_state + 1, instance->lower, instance->upper);
} else if (trend == THERMAL_TREND_DROPPING) {
if (cur_state <= instance->lower)
return THERMAL_NO_TARGET;
	/*
	 * If 'throttle' is false, no mitigation is necessary, so
	 * request the lower state for this instance.
	 */
	return instance->lower;
}

return instance->target;

}

1）注释将选择target state的逻辑总结得很清楚：如果当前温度高于trip温度，如果趋势是上升，选择更高的cooling 状态，如果趋势是下降，do nothing; 如果当前温度低于trip温度，如果趋势是上升，do nothing,如果趋势是下降，用更低的cooling状态，如果已经是最低的状态了，那么就deactive这个thermal instance

标签: thermal