linux_dsm_epyc7002/drivers/thermal/fair_share.c

121 lines
3.3 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* fair_share.c - A simple weight based Thermal governor
*
* Copyright (C) 2012 Intel Corp
* Copyright (C) 2012 Durgadoss R <durgadoss.r@intel.com>
*
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*
* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*/
#include <linux/thermal.h>
#include <trace/events/thermal.h>
#include "thermal_core.h"
/**
* get_trip_level: - obtains the current trip level for a zone
* @tz: thermal zone device
*/
static int get_trip_level(struct thermal_zone_device *tz)
{
int count = 0;
thermal: consistently use int for temperatures The thermal code uses int, long and unsigned long for temperatures in different places. Using an unsigned type limits the thermal framework to positive temperatures without need. Also several drivers currently will report temperatures near UINT_MAX for temperatures below 0°C. This will probably immediately shut the machine down due to overtemperature if started below 0°C. 'long' is 64bit on several architectures. This is not needed since INT_MAX °mC is above the melting point of all known materials. Consistently use a plain 'int' for temperatures throughout the thermal code and the drivers. This only changes the places in the drivers where the temperature is passed around as pointer, when drivers internally use another type this is not changed. Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be> Reviewed-by: Jean Delvare <jdelvare@suse.de> Reviewed-by: Lukasz Majewski <l.majewski@samsung.com> Reviewed-by: Darren Hart <dvhart@linux.intel.com> Reviewed-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Peter Feuerer <peter@piie.net> Cc: Punit Agrawal <punit.agrawal@arm.com> Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: Jean Delvare <jdelvare@suse.de> Cc: Peter Feuerer <peter@piie.net> Cc: Heiko Stuebner <heiko@sntech.de> Cc: Lukasz Majewski <l.majewski@samsung.com> Cc: Stephen Warren <swarren@wwwdotorg.org> Cc: Thierry Reding <thierry.reding@gmail.com> Cc: linux-acpi@vger.kernel.org Cc: platform-driver-x86@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-omap@vger.kernel.org Cc: linux-samsung-soc@vger.kernel.org Cc: Guenter Roeck <linux@roeck-us.net> Cc: Rafael J. Wysocki <rjw@rjwysocki.net> Cc: Maxime Ripard <maxime.ripard@free-electrons.com> Cc: Darren Hart <dvhart@infradead.org> Cc: lm-sensors@lm-sensors.org Signed-off-by: Zhang Rui <rui.zhang@intel.com>
2015-07-24 13:12:54 +07:00
int trip_temp;
enum thermal_trip_type trip_type;
if (tz->trips == 0 || !tz->ops->get_trip_temp)
return 0;
for (count = 0; count < tz->trips; count++) {
tz->ops->get_trip_temp(tz, count, &trip_temp);
if (tz->temperature < trip_temp)
break;
}
/*
* count > 0 only if temperature is greater than first trip
* point, in which case, trip_point = count - 1
*/
if (count > 0) {
tz->ops->get_trip_type(tz, count - 1, &trip_type);
trace_thermal_zone_trip(tz, count - 1, trip_type);
}
return count;
}
static long get_target_state(struct thermal_zone_device *tz,
thermal: fair_share: generalize the weight concept The fair share governor has the concept of weights, which is the influence of each cooling device in a thermal zone. The current implementation forces the weights of all cooling devices in a thermal zone to add up to a 100. This complicates setups, as you need to know in advance how many cooling devices you are going to have. If you bind a new cooling device, you have to modify all the other cooling devices weights, which is error prone. Furthermore, you can't specify a "default" weight for platforms since that default value depends on the number of cooling devices in the platform. This patch generalizes the concept of weight by allowing any number to be a "weight". Weights are now relative to each other. Platforms that don't specify weights get the same default value for all their cooling devices, so all their cdevs are considered to be equally influential. It's important to note that previous users of the weights don't need to alter the code: percentages continue to work as they used to. This patch just removes the constraint of all the weights in a thermal zone having to add up to a 100. If they do, you get the same behavior as before. If they don't, fair share now works for that platform. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Durgadoss R <durgadoss.r@intel.com> Acked-by: Durgadoss R <durgadoss.r@intel.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
2015-02-18 23:04:25 +07:00
struct thermal_cooling_device *cdev, int percentage, int level)
{
unsigned long max_state;
cdev->ops->get_max_state(cdev, &max_state);
thermal: fair_share: generalize the weight concept The fair share governor has the concept of weights, which is the influence of each cooling device in a thermal zone. The current implementation forces the weights of all cooling devices in a thermal zone to add up to a 100. This complicates setups, as you need to know in advance how many cooling devices you are going to have. If you bind a new cooling device, you have to modify all the other cooling devices weights, which is error prone. Furthermore, you can't specify a "default" weight for platforms since that default value depends on the number of cooling devices in the platform. This patch generalizes the concept of weight by allowing any number to be a "weight". Weights are now relative to each other. Platforms that don't specify weights get the same default value for all their cooling devices, so all their cdevs are considered to be equally influential. It's important to note that previous users of the weights don't need to alter the code: percentages continue to work as they used to. This patch just removes the constraint of all the weights in a thermal zone having to add up to a 100. If they do, you get the same behavior as before. If they don't, fair share now works for that platform. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Durgadoss R <durgadoss.r@intel.com> Acked-by: Durgadoss R <durgadoss.r@intel.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
2015-02-18 23:04:25 +07:00
return (long)(percentage * level * max_state) / (100 * tz->trips);
}
/**
* fair_share_throttle - throttles devices associated with the given zone
* @tz: thermal_zone_device
* @trip: trip point index
*
* Throttling Logic: This uses three parameters to calculate the new
* throttle state of the cooling devices associated with the given zone.
*
* Parameters used for Throttling:
* P1. max_state: Maximum throttle state exposed by the cooling device.
thermal: fair_share: generalize the weight concept The fair share governor has the concept of weights, which is the influence of each cooling device in a thermal zone. The current implementation forces the weights of all cooling devices in a thermal zone to add up to a 100. This complicates setups, as you need to know in advance how many cooling devices you are going to have. If you bind a new cooling device, you have to modify all the other cooling devices weights, which is error prone. Furthermore, you can't specify a "default" weight for platforms since that default value depends on the number of cooling devices in the platform. This patch generalizes the concept of weight by allowing any number to be a "weight". Weights are now relative to each other. Platforms that don't specify weights get the same default value for all their cooling devices, so all their cdevs are considered to be equally influential. It's important to note that previous users of the weights don't need to alter the code: percentages continue to work as they used to. This patch just removes the constraint of all the weights in a thermal zone having to add up to a 100. If they do, you get the same behavior as before. If they don't, fair share now works for that platform. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Durgadoss R <durgadoss.r@intel.com> Acked-by: Durgadoss R <durgadoss.r@intel.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
2015-02-18 23:04:25 +07:00
* P2. percentage[i]/100:
* How 'effective' the 'i'th device is, in cooling the given zone.
* P3. cur_trip_level/max_no_of_trips:
* This describes the extent to which the devices should be throttled.
* We do not want to throttle too much when we trip a lower temperature,
* whereas the throttling is at full swing if we trip critical levels.
* (Heavily assumes the trip points are in ascending order)
* new_state of cooling device = P3 * P2 * P1
*/
static int fair_share_throttle(struct thermal_zone_device *tz, int trip)
{
struct thermal_instance *instance;
thermal: fair_share: generalize the weight concept The fair share governor has the concept of weights, which is the influence of each cooling device in a thermal zone. The current implementation forces the weights of all cooling devices in a thermal zone to add up to a 100. This complicates setups, as you need to know in advance how many cooling devices you are going to have. If you bind a new cooling device, you have to modify all the other cooling devices weights, which is error prone. Furthermore, you can't specify a "default" weight for platforms since that default value depends on the number of cooling devices in the platform. This patch generalizes the concept of weight by allowing any number to be a "weight". Weights are now relative to each other. Platforms that don't specify weights get the same default value for all their cooling devices, so all their cdevs are considered to be equally influential. It's important to note that previous users of the weights don't need to alter the code: percentages continue to work as they used to. This patch just removes the constraint of all the weights in a thermal zone having to add up to a 100. If they do, you get the same behavior as before. If they don't, fair share now works for that platform. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Durgadoss R <durgadoss.r@intel.com> Acked-by: Durgadoss R <durgadoss.r@intel.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
2015-02-18 23:04:25 +07:00
int total_weight = 0;
int total_instance = 0;
int cur_trip_level = get_trip_level(tz);
list_for_each_entry(instance, &tz->thermal_instances, tz_node) {
thermal: fair_share: generalize the weight concept The fair share governor has the concept of weights, which is the influence of each cooling device in a thermal zone. The current implementation forces the weights of all cooling devices in a thermal zone to add up to a 100. This complicates setups, as you need to know in advance how many cooling devices you are going to have. If you bind a new cooling device, you have to modify all the other cooling devices weights, which is error prone. Furthermore, you can't specify a "default" weight for platforms since that default value depends on the number of cooling devices in the platform. This patch generalizes the concept of weight by allowing any number to be a "weight". Weights are now relative to each other. Platforms that don't specify weights get the same default value for all their cooling devices, so all their cdevs are considered to be equally influential. It's important to note that previous users of the weights don't need to alter the code: percentages continue to work as they used to. This patch just removes the constraint of all the weights in a thermal zone having to add up to a 100. If they do, you get the same behavior as before. If they don't, fair share now works for that platform. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Durgadoss R <durgadoss.r@intel.com> Acked-by: Durgadoss R <durgadoss.r@intel.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
2015-02-18 23:04:25 +07:00
if (instance->trip != trip)
continue;
total_weight += instance->weight;
total_instance++;
}
list_for_each_entry(instance, &tz->thermal_instances, tz_node) {
int percentage;
struct thermal_cooling_device *cdev = instance->cdev;
if (instance->trip != trip)
continue;
thermal: fair_share: generalize the weight concept The fair share governor has the concept of weights, which is the influence of each cooling device in a thermal zone. The current implementation forces the weights of all cooling devices in a thermal zone to add up to a 100. This complicates setups, as you need to know in advance how many cooling devices you are going to have. If you bind a new cooling device, you have to modify all the other cooling devices weights, which is error prone. Furthermore, you can't specify a "default" weight for platforms since that default value depends on the number of cooling devices in the platform. This patch generalizes the concept of weight by allowing any number to be a "weight". Weights are now relative to each other. Platforms that don't specify weights get the same default value for all their cooling devices, so all their cdevs are considered to be equally influential. It's important to note that previous users of the weights don't need to alter the code: percentages continue to work as they used to. This patch just removes the constraint of all the weights in a thermal zone having to add up to a 100. If they do, you get the same behavior as before. If they don't, fair share now works for that platform. Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Cc: Durgadoss R <durgadoss.r@intel.com> Acked-by: Durgadoss R <durgadoss.r@intel.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
2015-02-18 23:04:25 +07:00
if (!total_weight)
percentage = 100 / total_instance;
else
percentage = (instance->weight * 100) / total_weight;
instance->target = get_target_state(tz, cdev, percentage,
cur_trip_level);
thermal: fix race condition when updating cooling device When multiple thermal zones are bound to the same cooling device, multiple kernel threads may want to update the cooling device state by calling thermal_cdev_update(). Having cdev not protected by a mutex can lead to a race condition. Consider the following situation with two kernel threads k1 and k2: Thread k1 Thread k2 || || call thermal_cdev_update() || ... || set_cur_state(cdev, target); call power_actor_set_power() || ... || instance->target = state; || cdev->updated = false; || || cdev->updated = true; || // completes execution call thermal_cdev_update() || // cdev->updated == true || return; || \/ time k2 has already looped through the thermal instances looking for the deepest cooling device state and is preempted right before setting cdev->updated to true. Now, k1 runs, modifies the thermal instance state and sets cdev->updated to false. Then, k1 is preempted and k2 continues the execution by setting cdev->updated to true, therefore preventing k1 from performing the update. Notice that this is not an issue if k2 looks at the instance->target modified by k1 "after" it is assigned by k1. In fact, in this case the update will happen anyway and k1 can safely return immediately from thermal_cdev_update(). This may lead to a situation where a thermal governor never updates the cooling device. For example, this is the case for the step_wise governor: when calling the function thermal_zone_trip_update(), the governor may always get a new state equal to the old one (which, however, wasn't notified to the cooling device) and will therefore skip the update. CC: Zhang Rui <rui.zhang@intel.com> CC: Eduardo Valentin <edubezval@gmail.com> CC: Peter Feuerer <peter@piie.net> Reported-by: Toby Huang <toby.huang@arm.com> Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Zhang Rui <rui.zhang@intel.com>
2016-06-02 21:25:31 +07:00
mutex_lock(&instance->cdev->lock);
instance->cdev->updated = false;
thermal: fix race condition when updating cooling device When multiple thermal zones are bound to the same cooling device, multiple kernel threads may want to update the cooling device state by calling thermal_cdev_update(). Having cdev not protected by a mutex can lead to a race condition. Consider the following situation with two kernel threads k1 and k2: Thread k1 Thread k2 || || call thermal_cdev_update() || ... || set_cur_state(cdev, target); call power_actor_set_power() || ... || instance->target = state; || cdev->updated = false; || || cdev->updated = true; || // completes execution call thermal_cdev_update() || // cdev->updated == true || return; || \/ time k2 has already looped through the thermal instances looking for the deepest cooling device state and is preempted right before setting cdev->updated to true. Now, k1 runs, modifies the thermal instance state and sets cdev->updated to false. Then, k1 is preempted and k2 continues the execution by setting cdev->updated to true, therefore preventing k1 from performing the update. Notice that this is not an issue if k2 looks at the instance->target modified by k1 "after" it is assigned by k1. In fact, in this case the update will happen anyway and k1 can safely return immediately from thermal_cdev_update(). This may lead to a situation where a thermal governor never updates the cooling device. For example, this is the case for the step_wise governor: when calling the function thermal_zone_trip_update(), the governor may always get a new state equal to the old one (which, however, wasn't notified to the cooling device) and will therefore skip the update. CC: Zhang Rui <rui.zhang@intel.com> CC: Eduardo Valentin <edubezval@gmail.com> CC: Peter Feuerer <peter@piie.net> Reported-by: Toby Huang <toby.huang@arm.com> Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Zhang Rui <rui.zhang@intel.com>
2016-06-02 21:25:31 +07:00
mutex_unlock(&instance->cdev->lock);
thermal_cdev_update(cdev);
}
return 0;
}
static struct thermal_governor thermal_gov_fair_share = {
.name = "fair_share",
.throttle = fair_share_throttle,
};
THERMAL_GOVERNOR_DECLARE(thermal_gov_fair_share);