microsoft
/
WSL2-Linux-Kernel
зеркало из https://github.com/microsoft/WSL2-Linux-Kernel.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность

Merge branch 'thermal-intel' 

Merge thermal control changes related to Intel platforms for 6.3-rc1:

 - Rework ACPI helper functions for thermal control to retrieve a trip
   point temperature instead of initializing a trip point objetc (Rafael
   Wysocki).

 - Clean up and improve the int340x thermal driver ((Rafael Wysocki).

 - Simplify and clean up the intel_pch thermal driver ((Rafael Wysocki).

 - Fix the Intel powerclamp thermal driver and make it use the common
   idle injection framework (Srinivas Pandruvada).

 - Add two module parameters, cpumask and max_idle, to the Intel powerclamp
   thermal driver to allow it to affect only a specific subset of CPUs
   instead of all of them (Srinivas Pandruvada).

 - Make the Intel quark_dts thermal driver Use generic trip point
   objects instead of its own trip point representation (Daniel
   Lezcano).

 - Add toctree entry for thermal documents and fix two issues in the
   Intel powerclamp driver documentation (Bagas Sanjaya).

* thermal-intel: (25 commits)
  Documentation: powerclamp: Fix numbered lists formatting
  Documentation: powerclamp: Escape wildcard in cpumask description
  Documentation: admin-guide: Add toctree entry for thermal docs
  thermal: intel: powerclamp: Add two module parameters
  Documentation: admin-guide: Move intel_powerclamp documentation
  thermal: intel: powerclamp: Fix duration module parameter
  thermal: intel: powerclamp: Return last requested state as cur_state
  thermal: intel: quark_dts: Use generic trip points
  thermal: intel: powerclamp: Use powercap idle-inject feature
  powercap: idle_inject: Add update callback
  powercap: idle_inject: Export symbols
  thermal: intel: powerclamp: Fix cur_state for multi package system
  thermal: intel: intel_pch: Drop struct board_info
  thermal: intel: intel_pch: Rename board ID symbols
  thermal: intel: intel_pch: Fold suspend and resume routines into their callers
  thermal: intel: intel_pch: Fold two functions into their callers
  thermal: intel: intel_pch: Eliminate device operations object
  thermal: intel: intel_pch: Rename device operations callbacks
  thermal: intel: intel_pch: Eliminate redundant return pointers
  thermal: intel: intel_pch: Make pch_wpt_add_acpi_psv_trip() return int
  ...
This commit is contained in:
Rafael J. Wysocki 2023-02-15 17:18:08 +01:00
Родитель c3bd6d539f fef1f0be10
Коммит badf1f9050
14 изменённых файлов: 694 добавлений и 641 удалений
Показать статистику Скачать Patch файл Скачать Diff файл Показать все файлы Свернуть все файлы

1
Documentation/admin-guide/index.rst
Просмотреть файл

@ -116,6 +116,7 @@ configure specific aspects of kernel behavior to your liking.
svga svga
syscall-user-dispatch syscall-user-dispatch
sysrq sysrq
thermal/index
thunderbolt thunderbolt
ufs ufs
unicode unicode

8
Documentation/admin-guide/thermal/index.rst Normal file
Просмотреть файл

@ -0,0 +1,8 @@
=================
Thermal Subsystem
=================
.. toctree::
:maxdepth: 1
intel_powerclamp

35
Documentation/driver-api/thermal/intel_powerclamp.rst → Documentation/admin-guide/thermal/intel_powerclamp.rst
Просмотреть файл

@ -26,6 +26,8 @@ By:
- Generic Thermal Layer (sysfs) - Generic Thermal Layer (sysfs)
- Kernel APIs (TBD) - Kernel APIs (TBD)
(*) Module Parameters
INTRODUCTION INTRODUCTION
============ ============
@ -153,13 +155,15 @@ b) determine the amount of compensation needed at each target ratio
Compensation to each target ratio consists of two parts: Compensation to each target ratio consists of two parts:
a) steady state error compensation a) steady state error compensation
This is to offset the error occurring when the system can
enter idle without extra wakeups (such as external interrupts). This is to offset the error occurring when the system can
enter idle without extra wakeups (such as external interrupts).
b) dynamic error compensation b) dynamic error compensation
When an excessive amount of wakeups occurs during idle, an
additional idle ratio can be added to quiet interrupts, by When an excessive amount of wakeups occurs during idle, an
slowing down CPU activities. additional idle ratio can be added to quiet interrupts, by
slowing down CPU activities.
A debugfs file is provided for the user to examine compensation A debugfs file is provided for the user to examine compensation
progress and results, such as on a Westmere system:: progress and results, such as on a Westmere system::
@ -281,6 +285,7 @@ cur_state returns value -1 instead of 0 which is to avoid confusing
100% busy state with the disabled state. 100% busy state with the disabled state.
Example usage: Example usage:
- To inject 25% idle time:: - To inject 25% idle time::
$ sudo sh -c "echo 25 > /sys/class/thermal/cooling_device80/cur_state $ sudo sh -c "echo 25 > /sys/class/thermal/cooling_device80/cur_state
@ -318,3 +323,23 @@ device, a PID based userspace thermal controller can manage to
control CPU temperature effectively, when no other thermal influence control CPU temperature effectively, when no other thermal influence
is added. For example, a UltraBook user can compile the kernel under is added. For example, a UltraBook user can compile the kernel under
certain temperature (below most active trip points). certain temperature (below most active trip points).
Module Parameters
=================
``cpumask`` (RW)
A bit mask of CPUs to inject idle. The format of the bitmask is same as
used in other subsystems like in /proc/irq/\*/smp_affinity. The mask is
comma separated 32 bit groups. Each CPU is one bit. For example for a 256
CPU system the full mask is:
ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff,ffffffff
The rightmost mask is for CPU 0-32.
``max_idle`` (RW)
Maximum injected idle time to the total CPU time ratio in percent range
from 1 to 100. Even if the cooling device max_state is always 100 (100%),
this parameter allows to add a max idle percent limit. The default is 50,
to match the current implementation of powerclamp driver. Also doesn't
allow value more than 75, if the cpumask includes every CPU present in
the system.

1
Documentation/driver-api/thermal/index.rst
Просмотреть файл

@ -14,7 +14,6 @@ Thermal
exynos_thermal exynos_thermal
exynos_thermal_emulation exynos_thermal_emulation
intel_powerclamp
nouveau_thermal nouveau_thermal
x86_pkg_temperature_thermal x86_pkg_temperature_thermal
intel_dptf intel_dptf

1
MAINTAINERS
Просмотреть файл

@ -20707,6 +20707,7 @@ S: Supported
Q: https://patchwork.kernel.org/project/linux-pm/list/ Q: https://patchwork.kernel.org/project/linux-pm/list/
T: git git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git thermal T: git git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git thermal
F: Documentation/ABI/testing/sysfs-class-thermal F: Documentation/ABI/testing/sysfs-class-thermal
F: Documentation/admin-guide/thermal/
F: Documentation/devicetree/bindings/thermal/ F: Documentation/devicetree/bindings/thermal/
F: Documentation/driver-api/thermal/ F: Documentation/driver-api/thermal/
F: drivers/thermal/ F: drivers/thermal/

59
drivers/powercap/idle_inject.c
Просмотреть файл

@ -63,13 +63,29 @@ struct idle_inject_thread {
* @idle_duration_us: duration of CPU idle time to inject * @idle_duration_us: duration of CPU idle time to inject
* @run_duration_us: duration of CPU run time to allow * @run_duration_us: duration of CPU run time to allow
* @latency_us: max allowed latency * @latency_us: max allowed latency
* @update: Optional callback deciding whether or not to skip idle
* injection in the given cycle.
* @cpumask: mask of CPUs affected by idle injection * @cpumask: mask of CPUs affected by idle injection
*
* This structure is used to define per instance idle inject device data. Each
* instance has an idle duration, a run duration and mask of CPUs to inject
* idle.
*
* Actual CPU idle time is injected by calling kernel scheduler interface
* play_idle_precise(). There is one optional callback that can be registered
* by calling idle_inject_register_full():
*
* update() - This callback is invoked just before waking up CPUs to inject
* idle. If it returns false, CPUs are not woken up to inject idle in the given
* cycle. It also allows the caller to readjust the idle and run duration by
* calling idle_inject_set_duration() for the next cycle.
*/ */
struct idle_inject_device { struct idle_inject_device {
struct hrtimer timer; struct hrtimer timer;
unsigned int idle_duration_us; unsigned int idle_duration_us;
unsigned int run_duration_us; unsigned int run_duration_us;
unsigned int latency_us; unsigned int latency_us;
bool (*update)(void);
unsigned long cpumask[]; unsigned long cpumask[];
}; };
@ -111,11 +127,12 @@ static enum hrtimer_restart idle_inject_timer_fn(struct hrtimer *timer)
struct idle_inject_device *ii_dev = struct idle_inject_device *ii_dev =
container_of(timer, struct idle_inject_device, timer); container_of(timer, struct idle_inject_device, timer);
if (!ii_dev->update || (ii_dev->update && ii_dev->update()))
idle_inject_wakeup(ii_dev);
duration_us = READ_ONCE(ii_dev->run_duration_us); duration_us = READ_ONCE(ii_dev->run_duration_us);
duration_us += READ_ONCE(ii_dev->idle_duration_us); duration_us += READ_ONCE(ii_dev->idle_duration_us);
idle_inject_wakeup(ii_dev);
hrtimer_forward_now(timer, ns_to_ktime(duration_us * NSEC_PER_USEC)); hrtimer_forward_now(timer, ns_to_ktime(duration_us * NSEC_PER_USEC));
return HRTIMER_RESTART; return HRTIMER_RESTART;
@ -160,6 +177,7 @@ void idle_inject_set_duration(struct idle_inject_device *ii_dev,
WRITE_ONCE(ii_dev->idle_duration_us, idle_duration_us); WRITE_ONCE(ii_dev->idle_duration_us, idle_duration_us);
} }
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_set_duration, IDLE_INJECT);
/** /**
* idle_inject_get_duration - idle and run duration retrieval helper * idle_inject_get_duration - idle and run duration retrieval helper
@ -174,6 +192,7 @@ void idle_inject_get_duration(struct idle_inject_device *ii_dev,
*run_duration_us = READ_ONCE(ii_dev->run_duration_us); *run_duration_us = READ_ONCE(ii_dev->run_duration_us);
*idle_duration_us = READ_ONCE(ii_dev->idle_duration_us); *idle_duration_us = READ_ONCE(ii_dev->idle_duration_us);
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_get_duration, IDLE_INJECT);
/** /**
* idle_inject_set_latency - set the maximum latency allowed * idle_inject_set_latency - set the maximum latency allowed
@ -185,6 +204,7 @@ void idle_inject_set_latency(struct idle_inject_device *ii_dev,
{ {
WRITE_ONCE(ii_dev->latency_us, latency_us); WRITE_ONCE(ii_dev->latency_us, latency_us);
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_set_latency, IDLE_INJECT);
/** /**
* idle_inject_start - start idle injections * idle_inject_start - start idle injections
@ -216,6 +236,7 @@ int idle_inject_start(struct idle_inject_device *ii_dev)
return 0; return 0;
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_start, IDLE_INJECT);
/** /**
* idle_inject_stop - stops idle injections * idle_inject_stop - stops idle injections
@ -262,6 +283,7 @@ void idle_inject_stop(struct idle_inject_device *ii_dev)
cpu_hotplug_enable(); cpu_hotplug_enable();
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_stop, IDLE_INJECT);
/** /**
* idle_inject_setup - prepare the current task for idle injection * idle_inject_setup - prepare the current task for idle injection
@ -290,17 +312,22 @@ static int idle_inject_should_run(unsigned int cpu)
} }
/** /**
* idle_inject_register - initialize idle injection on a set of CPUs * idle_inject_register_full - initialize idle injection on a set of CPUs
* @cpumask: CPUs to be affected by idle injection * @cpumask: CPUs to be affected by idle injection
* @update: This callback is called just before waking up CPUs to inject
* idle
* *
* This function creates an idle injection control device structure for the * This function creates an idle injection control device structure for the
* given set of CPUs and initializes the timer associated with it. It does not * given set of CPUs and initializes the timer associated with it. This
* start any injection cycles. * function also allows to register update()callback.
* It does not start any injection cycles.
* *
* Return: NULL if memory allocation fails, idle injection control device * Return: NULL if memory allocation fails, idle injection control device
* pointer on success. * pointer on success.
*/ */
struct idle_inject_device *idle_inject_register(struct cpumask *cpumask)
struct idle_inject_device *idle_inject_register_full(struct cpumask *cpumask,
bool (*update)(void))
{ {
struct idle_inject_device *ii_dev; struct idle_inject_device *ii_dev;
int cpu, cpu_rb; int cpu, cpu_rb;
@ -313,6 +340,7 @@ struct idle_inject_device *idle_inject_register(struct cpumask *cpumask)
hrtimer_init(&ii_dev->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); hrtimer_init(&ii_dev->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
ii_dev->timer.function = idle_inject_timer_fn; ii_dev->timer.function = idle_inject_timer_fn;
ii_dev->latency_us = UINT_MAX; ii_dev->latency_us = UINT_MAX;
ii_dev->update = update;
for_each_cpu(cpu, to_cpumask(ii_dev->cpumask)) { for_each_cpu(cpu, to_cpumask(ii_dev->cpumask)) {
@ -337,6 +365,24 @@ out_rollback:
return NULL; return NULL;
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_register_full, IDLE_INJECT);
/**
* idle_inject_register - initialize idle injection on a set of CPUs
* @cpumask: CPUs to be affected by idle injection
*
* This function creates an idle injection control device structure for the
* given set of CPUs and initializes the timer associated with it. It does not
* start any injection cycles.
*
* Return: NULL if memory allocation fails, idle injection control device
* pointer on success.
*/
struct idle_inject_device *idle_inject_register(struct cpumask *cpumask)
{
return idle_inject_register_full(cpumask, NULL);
}
EXPORT_SYMBOL_NS_GPL(idle_inject_register, IDLE_INJECT);
/** /**
* idle_inject_unregister - unregister idle injection control device * idle_inject_unregister - unregister idle injection control device
@ -357,6 +403,7 @@ void idle_inject_unregister(struct idle_inject_device *ii_dev)
kfree(ii_dev); kfree(ii_dev);
} }
EXPORT_SYMBOL_NS_GPL(idle_inject_unregister, IDLE_INJECT);
static struct smp_hotplug_thread idle_inject_threads = { static struct smp_hotplug_thread idle_inject_threads = {
.store = &idle_inject_thread.tsk, .store = &idle_inject_thread.tsk,

3
drivers/thermal/intel/Kconfig
Просмотреть файл

@ -3,6 +3,9 @@ config INTEL_POWERCLAMP
tristate "Intel PowerClamp idle injection driver" tristate "Intel PowerClamp idle injection driver"
depends on X86 depends on X86
depends on CPU_SUP_INTEL depends on CPU_SUP_INTEL
depends on CPU_IDLE
select POWERCAP
select IDLE_INJECT
help help
Enable this to enable Intel PowerClamp idle injection driver. This Enable this to enable Intel PowerClamp idle injection driver. This
enforce idle time which results in more package C-state residency. The enforce idle time which results in more package C-state residency. The

127
drivers/thermal/intel/int340x_thermal/int340x_thermal_zone.c
Просмотреть файл

@ -29,24 +29,27 @@ static int int340x_thermal_get_zone_temp(struct thermal_zone_device *zone,
if (conv_temp < 0) if (conv_temp < 0)
return conv_temp; return conv_temp;
*temp = (unsigned long)conv_temp * 10; *temp = conv_temp * 10;
} else } else {
/* _TMP returns the temperature in tenths of degrees Kelvin */ /* _TMP returns the temperature in tenths of degrees Kelvin */
*temp = deci_kelvin_to_millicelsius(tmp); *temp = deci_kelvin_to_millicelsius(tmp);
}
return 0; return 0;
} }
static int int340x_thermal_set_trip_temp(struct thermal_zone_device *zone, static int int340x_thermal_set_trip_temp(struct thermal_zone_device *zone,
int trip, int temp) int trip, int temp)
{ {
struct int34x_thermal_zone *d = zone->devdata; struct int34x_thermal_zone *d = zone->devdata;
char name[] = {'P', 'A', 'T', '0' + trip, '\0'};
acpi_status status; acpi_status status;
char name[10];
snprintf(name, sizeof(name), "PAT%d", trip); if (trip > 9)
return -EINVAL;
status = acpi_execute_simple_method(d->adev->handle, name, status = acpi_execute_simple_method(d->adev->handle, name,
millicelsius_to_deci_kelvin(temp)); millicelsius_to_deci_kelvin(temp));
if (ACPI_FAILURE(status)) if (ACPI_FAILURE(status))
return -EIO; return -EIO;
@ -70,24 +73,34 @@ static int int340x_thermal_read_trips(struct acpi_device *zone_adev,
{ {
int i, ret; int i, ret;
ret = thermal_acpi_trip_critical(zone_adev, &zone_trips[trip_cnt]); ret = thermal_acpi_critical_trip_temp(zone_adev,
if (!ret) &zone_trips[trip_cnt].temperature);
if (!ret) {
zone_trips[trip_cnt].type = THERMAL_TRIP_CRITICAL;
trip_cnt++; trip_cnt++;
}
ret = thermal_acpi_trip_hot(zone_adev, &zone_trips[trip_cnt]); ret = thermal_acpi_hot_trip_temp(zone_adev,
if (!ret) &zone_trips[trip_cnt].temperature);
if (!ret) {
zone_trips[trip_cnt].type = THERMAL_TRIP_HOT;
trip_cnt++; trip_cnt++;
}
ret = thermal_acpi_trip_passive(zone_adev, &zone_trips[trip_cnt]); ret = thermal_acpi_passive_trip_temp(zone_adev,
if (!ret) &zone_trips[trip_cnt].temperature);
if (!ret) {
zone_trips[trip_cnt].type = THERMAL_TRIP_PASSIVE;
trip_cnt++; trip_cnt++;
}
for (i = 0; i < INT340X_THERMAL_MAX_ACT_TRIP_COUNT; i++) { for (i = 0; i < INT340X_THERMAL_MAX_ACT_TRIP_COUNT; i++) {
ret = thermal_acpi_active_trip_temp(zone_adev, i,
ret = thermal_acpi_trip_active(zone_adev, i, &zone_trips[trip_cnt]); &zone_trips[trip_cnt].temperature);
if (ret) if (ret)
break; break;
zone_trips[trip_cnt].type = THERMAL_TRIP_ACTIVE;
trip_cnt++; trip_cnt++;
} }
@ -102,7 +115,7 @@ static struct thermal_zone_params int340x_thermal_params = {
struct int34x_thermal_zone *int340x_thermal_zone_add(struct acpi_device *adev, struct int34x_thermal_zone *int340x_thermal_zone_add(struct acpi_device *adev,
int (*get_temp) (struct thermal_zone_device *, int *)) int (*get_temp) (struct thermal_zone_device *, int *))
{ {
struct int34x_thermal_zone *int34x_thermal_zone; struct int34x_thermal_zone *int34x_zone;
struct thermal_trip *zone_trips; struct thermal_trip *zone_trips;
unsigned long long trip_cnt = 0; unsigned long long trip_cnt = 0;
unsigned long long hyst; unsigned long long hyst;
@ -110,26 +123,25 @@ struct int34x_thermal_zone *int340x_thermal_zone_add(struct acpi_device *adev,
acpi_status status; acpi_status status;
int i, ret; int i, ret;
int34x_thermal_zone = kzalloc(sizeof(*int34x_thermal_zone), int34x_zone = kzalloc(sizeof(*int34x_zone), GFP_KERNEL);
GFP_KERNEL); if (!int34x_zone)
if (!int34x_thermal_zone)
return ERR_PTR(-ENOMEM); return ERR_PTR(-ENOMEM);
int34x_thermal_zone->adev = adev; int34x_zone->adev = adev;
int34x_thermal_zone->ops = kmemdup(&int340x_thermal_zone_ops, int34x_zone->ops = kmemdup(&int340x_thermal_zone_ops,
sizeof(int340x_thermal_zone_ops), GFP_KERNEL); sizeof(int340x_thermal_zone_ops), GFP_KERNEL);
if (!int34x_thermal_zone->ops) { if (!int34x_zone->ops) {
ret = -ENOMEM; ret = -ENOMEM;
goto err_ops_alloc; goto err_ops_alloc;
} }
if (get_temp) if (get_temp)
int34x_thermal_zone->ops->get_temp = get_temp; int34x_zone->ops->get_temp = get_temp;
status = acpi_evaluate_integer(adev->handle, "PATC", NULL, &trip_cnt); status = acpi_evaluate_integer(adev->handle, "PATC", NULL, &trip_cnt);
if (!ACPI_FAILURE(status)) { if (ACPI_SUCCESS(status)) {
int34x_thermal_zone->aux_trip_nr = trip_cnt; int34x_zone->aux_trip_nr = trip_cnt;
trip_mask = BIT(trip_cnt) - 1; trip_mask = BIT(trip_cnt) - 1;
} }
@ -156,49 +168,47 @@ struct int34x_thermal_zone *int340x_thermal_zone_add(struct acpi_device *adev,
for (i = 0; i < trip_cnt; ++i) for (i = 0; i < trip_cnt; ++i)
zone_trips[i].hysteresis = hyst; zone_trips[i].hysteresis = hyst;
int34x_thermal_zone->trips = zone_trips; int34x_zone->trips = zone_trips;
int34x_thermal_zone->lpat_table = acpi_lpat_get_conversion_table( int34x_zone->lpat_table = acpi_lpat_get_conversion_table(adev->handle);
adev->handle);
int34x_thermal_zone->zone = thermal_zone_device_register_with_trips( int34x_zone->zone = thermal_zone_device_register_with_trips(
acpi_device_bid(adev), acpi_device_bid(adev),
zone_trips, trip_cnt, zone_trips, trip_cnt,
trip_mask, int34x_thermal_zone, trip_mask, int34x_zone,
int34x_thermal_zone->ops, int34x_zone->ops,
&int340x_thermal_params, &int340x_thermal_params,
0, 0); 0, 0);
if (IS_ERR(int34x_thermal_zone->zone)) { if (IS_ERR(int34x_zone->zone)) {
ret = PTR_ERR(int34x_thermal_zone->zone); ret = PTR_ERR(int34x_zone->zone);
goto err_thermal_zone; goto err_thermal_zone;
} }
ret = thermal_zone_device_enable(int34x_thermal_zone->zone); ret = thermal_zone_device_enable(int34x_zone->zone);
if (ret) if (ret)
goto err_enable; goto err_enable;
return int34x_thermal_zone; return int34x_zone;
err_enable: err_enable:
thermal_zone_device_unregister(int34x_thermal_zone->zone); thermal_zone_device_unregister(int34x_zone->zone);
err_thermal_zone: err_thermal_zone:
kfree(int34x_thermal_zone->trips); kfree(int34x_zone->trips);
acpi_lpat_free_conversion_table(int34x_thermal_zone->lpat_table); acpi_lpat_free_conversion_table(int34x_zone->lpat_table);
err_trips_alloc: err_trips_alloc:
kfree(int34x_thermal_zone->ops); kfree(int34x_zone->ops);
err_ops_alloc: err_ops_alloc:
kfree(int34x_thermal_zone); kfree(int34x_zone);
return ERR_PTR(ret); return ERR_PTR(ret);
} }
EXPORT_SYMBOL_GPL(int340x_thermal_zone_add); EXPORT_SYMBOL_GPL(int340x_thermal_zone_add);
void int340x_thermal_zone_remove(struct int34x_thermal_zone void int340x_thermal_zone_remove(struct int34x_thermal_zone *int34x_zone)
*int34x_thermal_zone)
{ {
thermal_zone_device_unregister(int34x_thermal_zone->zone); thermal_zone_device_unregister(int34x_zone->zone);
acpi_lpat_free_conversion_table(int34x_thermal_zone->lpat_table); acpi_lpat_free_conversion_table(int34x_zone->lpat_table);
kfree(int34x_thermal_zone->trips); kfree(int34x_zone->trips);
kfree(int34x_thermal_zone->ops); kfree(int34x_zone->ops);
kfree(int34x_thermal_zone); kfree(int34x_zone);
} }
EXPORT_SYMBOL_GPL(int340x_thermal_zone_remove); EXPORT_SYMBOL_GPL(int340x_thermal_zone_remove);
@ -213,22 +223,21 @@ void int340x_thermal_update_trips(struct int34x_thermal_zone *int34x_zone)
mutex_lock(&int34x_zone->zone->lock); mutex_lock(&int34x_zone->zone->lock);
for (i = int34x_zone->aux_trip_nr; i < trip_cnt; i++) { for (i = int34x_zone->aux_trip_nr; i < trip_cnt; i++) {
struct thermal_trip trip; int temp, err;
int err;
switch (zone_trips[i].type) { switch (zone_trips[i].type) {
case THERMAL_TRIP_CRITICAL: case THERMAL_TRIP_CRITICAL:
err = thermal_acpi_trip_critical(zone_adev, &trip); err = thermal_acpi_critical_trip_temp(zone_adev, &temp);
break; break;
case THERMAL_TRIP_HOT: case THERMAL_TRIP_HOT:
err = thermal_acpi_trip_hot(zone_adev, &trip); err = thermal_acpi_hot_trip_temp(zone_adev, &temp);
break; break;
case THERMAL_TRIP_PASSIVE: case THERMAL_TRIP_PASSIVE:
err = thermal_acpi_trip_passive(zone_adev, &trip); err = thermal_acpi_passive_trip_temp(zone_adev, &temp);
break; break;
case THERMAL_TRIP_ACTIVE: case THERMAL_TRIP_ACTIVE:
err = thermal_acpi_trip_active(zone_adev, act_trip_nr++, err = thermal_acpi_active_trip_temp(zone_adev, act_trip_nr++,
&trip); &temp);
break; break;
default: default:
err = -ENODEV; err = -ENODEV;
@ -238,7 +247,7 @@ void int340x_thermal_update_trips(struct int34x_thermal_zone *int34x_zone)
continue; continue;
} }
zone_trips[i].temperature = trip.temperature; zone_trips[i].temperature = temp;
} }
mutex_unlock(&int34x_zone->zone->lock); mutex_unlock(&int34x_zone->zone->lock);

354
drivers/thermal/intel/intel_pch_thermal.c
Просмотреть файл

@ -82,7 +82,6 @@ static char driver_name[] = "Intel PCH thermal driver";
struct pch_thermal_device { struct pch_thermal_device {
void __iomem *hw_base; void __iomem *hw_base;
const struct pch_dev_ops *ops;
struct pci_dev *pdev; struct pci_dev *pdev;
struct thermal_zone_device *tzd; struct thermal_zone_device *tzd;
struct thermal_trip trips[PCH_MAX_TRIPS]; struct thermal_trip trips[PCH_MAX_TRIPS];
@ -90,42 +89,107 @@ struct pch_thermal_device {
}; };
#ifdef CONFIG_ACPI #ifdef CONFIG_ACPI
/* /*
* On some platforms, there is a companion ACPI device, which adds * On some platforms, there is a companion ACPI device, which adds
* passive trip temperature using _PSV method. There is no specific * passive trip temperature using _PSV method. There is no specific
* passive temperature setting in MMIO interface of this PCI device. * passive temperature setting in MMIO interface of this PCI device.
*/ */
static void pch_wpt_add_acpi_psv_trip(struct pch_thermal_device *ptd, static int pch_wpt_add_acpi_psv_trip(struct pch_thermal_device *ptd, int trip)
int *nr_trips)
{ {
struct acpi_device *adev; struct acpi_device *adev;
int ret; int temp;
adev = ACPI_COMPANION(&ptd->pdev->dev); adev = ACPI_COMPANION(&ptd->pdev->dev);
if (!adev) if (!adev)
return; return 0;
ret = thermal_acpi_trip_passive(adev, &ptd->trips[*nr_trips]); if (thermal_acpi_passive_trip_temp(adev, &temp) || temp <= 0)
if (ret || ptd->trips[*nr_trips].temperature <= 0) return 0;
return;
++(*nr_trips); ptd->trips[trip].type = THERMAL_TRIP_PASSIVE;
ptd->trips[trip].temperature = temp;
return 1;
} }
#else #else
static void pch_wpt_add_acpi_psv_trip(struct pch_thermal_device *ptd, static int pch_wpt_add_acpi_psv_trip(struct pch_thermal_device *ptd, int trip)
int *nr_trips)
{ {
return 0;
} }
#endif #endif
static int pch_wpt_init(struct pch_thermal_device *ptd, int *nr_trips) static int pch_thermal_get_temp(struct thermal_zone_device *tzd, int *temp)
{ {
u8 tsel; struct pch_thermal_device *ptd = tzd->devdata;
u16 trip_temp;
*nr_trips = 0; *temp = GET_WPT_TEMP(WPT_TEMP_TSR & readw(ptd->hw_base + WPT_TEMP));
return 0;
}
static void pch_critical(struct thermal_zone_device *tzd)
{
dev_dbg(&tzd->device, "%s: critical temperature reached\n", tzd->type);
}
static struct thermal_zone_device_ops tzd_ops = {
.get_temp = pch_thermal_get_temp,
.critical = pch_critical,
};
enum pch_board_ids {
PCH_BOARD_HSW = 0,
PCH_BOARD_WPT,
PCH_BOARD_SKL,
PCH_BOARD_CNL,
PCH_BOARD_CML,
PCH_BOARD_LWB,
PCH_BOARD_WBG,
};
static const char *board_names[] = {
[PCH_BOARD_HSW] = "pch_haswell",
[PCH_BOARD_WPT] = "pch_wildcat_point",
[PCH_BOARD_SKL] = "pch_skylake",
[PCH_BOARD_CNL] = "pch_cannonlake",
[PCH_BOARD_CML] = "pch_cometlake",
[PCH_BOARD_LWB] = "pch_lewisburg",
[PCH_BOARD_WBG] = "pch_wellsburg",
};
static int intel_pch_thermal_probe(struct pci_dev *pdev,
const struct pci_device_id *id)
{
enum pch_board_ids board_id = id->driver_data;
struct pch_thermal_device *ptd;
int nr_trips = 0;
u16 trip_temp;
u8 tsel;
int err;
ptd = devm_kzalloc(&pdev->dev, sizeof(*ptd), GFP_KERNEL);
if (!ptd)
return -ENOMEM;
pci_set_drvdata(pdev, ptd);
ptd->pdev = pdev;
err = pci_enable_device(pdev);
if (err) {
dev_err(&pdev->dev, "failed to enable pci device\n");
return err;
}
err = pci_request_regions(pdev, driver_name);
if (err) {
dev_err(&pdev->dev, "failed to request pci region\n");
goto error_disable;
}
ptd->hw_base = pci_ioremap_bar(pdev, 0);
if (!ptd->hw_base) {
err = -ENOMEM;
dev_err(&pdev->dev, "failed to map mem base\n");
goto error_release;
}
/* Check if BIOS has already enabled thermal sensor */ /* Check if BIOS has already enabled thermal sensor */
if (WPT_TSEL_ETS & readb(ptd->hw_base + WPT_TSEL)) { if (WPT_TSEL_ETS & readb(ptd->hw_base + WPT_TSEL)) {
@ -140,50 +204,79 @@ static int pch_wpt_init(struct pch_thermal_device *ptd, int *nr_trips)
*/ */
if (tsel & WPT_TSEL_PLDB) { if (tsel & WPT_TSEL_PLDB) {
dev_err(&ptd->pdev->dev, "Sensor can't be enabled\n"); dev_err(&ptd->pdev->dev, "Sensor can't be enabled\n");
return -ENODEV; err = -ENODEV;
goto error_cleanup;
} }
writeb(tsel|WPT_TSEL_ETS, ptd->hw_base + WPT_TSEL); writeb(tsel|WPT_TSEL_ETS, ptd->hw_base + WPT_TSEL);
if (!(WPT_TSEL_ETS & readb(ptd->hw_base + WPT_TSEL))) { if (!(WPT_TSEL_ETS & readb(ptd->hw_base + WPT_TSEL))) {
dev_err(&ptd->pdev->dev, "Sensor can't be enabled\n"); dev_err(&ptd->pdev->dev, "Sensor can't be enabled\n");
return -ENODEV; err = -ENODEV;
goto error_cleanup;
} }
read_trips: read_trips:
trip_temp = readw(ptd->hw_base + WPT_CTT); trip_temp = readw(ptd->hw_base + WPT_CTT);
trip_temp &= 0x1FF; trip_temp &= 0x1FF;
if (trip_temp) { if (trip_temp) {
ptd->trips[*nr_trips].temperature = GET_WPT_TEMP(trip_temp); ptd->trips[nr_trips].temperature = GET_WPT_TEMP(trip_temp);
ptd->trips[*nr_trips].type = THERMAL_TRIP_CRITICAL; ptd->trips[nr_trips++].type = THERMAL_TRIP_CRITICAL;
++(*nr_trips);
} }
trip_temp = readw(ptd->hw_base + WPT_PHL); trip_temp = readw(ptd->hw_base + WPT_PHL);
trip_temp &= 0x1FF; trip_temp &= 0x1FF;
if (trip_temp) { if (trip_temp) {
ptd->trips[*nr_trips].temperature = GET_WPT_TEMP(trip_temp); ptd->trips[nr_trips].temperature = GET_WPT_TEMP(trip_temp);
ptd->trips[*nr_trips].type = THERMAL_TRIP_HOT; ptd->trips[nr_trips++].type = THERMAL_TRIP_HOT;
++(*nr_trips);
} }
pch_wpt_add_acpi_psv_trip(ptd, nr_trips); nr_trips += pch_wpt_add_acpi_psv_trip(ptd, nr_trips);
ptd->tzd = thermal_zone_device_register_with_trips(board_names[board_id],
ptd->trips, nr_trips,
0, ptd, &tzd_ops,
NULL, 0, 0);
if (IS_ERR(ptd->tzd)) {
dev_err(&pdev->dev, "Failed to register thermal zone %s\n",
board_names[board_id]);
err = PTR_ERR(ptd->tzd);
goto error_cleanup;
}
err = thermal_zone_device_enable(ptd->tzd);
if (err)
goto err_unregister;
return 0; return 0;
err_unregister:
thermal_zone_device_unregister(ptd->tzd);
error_cleanup:
iounmap(ptd->hw_base);
error_release:
pci_release_regions(pdev);
error_disable:
pci_disable_device(pdev);
dev_err(&pdev->dev, "pci device failed to probe\n");
return err;
} }
static int pch_wpt_get_temp(struct pch_thermal_device *ptd, int *temp) static void intel_pch_thermal_remove(struct pci_dev *pdev)
{ {
*temp = GET_WPT_TEMP(WPT_TEMP_TSR & readw(ptd->hw_base + WPT_TEMP)); struct pch_thermal_device *ptd = pci_get_drvdata(pdev);
return 0; thermal_zone_device_unregister(ptd->tzd);
iounmap(ptd->hw_base);
pci_set_drvdata(pdev, NULL);
pci_release_regions(pdev);
pci_disable_device(pdev);
} }
/* Cool the PCH when it's overheat in .suspend_noirq phase */ static int intel_pch_thermal_suspend_noirq(struct device *device)
static int pch_wpt_suspend(struct pch_thermal_device *ptd)
{ {
u8 tsel; struct pch_thermal_device *ptd = dev_get_drvdata(device);
int pch_delay_cnt = 0;
u16 pch_thr_temp, pch_cur_temp; u16 pch_thr_temp, pch_cur_temp;
int pch_delay_cnt = 0;
u8 tsel;
/* Shutdown the thermal sensor if it is not enabled by BIOS */ /* Shutdown the thermal sensor if it is not enabled by BIOS */
if (!ptd->bios_enabled) { if (!ptd->bios_enabled) {
@ -246,8 +339,9 @@ static int pch_wpt_suspend(struct pch_thermal_device *ptd)
return 0; return 0;
} }
static int pch_wpt_resume(struct pch_thermal_device *ptd) static int intel_pch_thermal_resume(struct device *device)
{ {
struct pch_thermal_device *ptd = dev_get_drvdata(device);
u8 tsel; u8 tsel;
if (ptd->bios_enabled) if (ptd->bios_enabled)
@ -260,199 +354,29 @@ static int pch_wpt_resume(struct pch_thermal_device *ptd)
return 0; return 0;
} }
struct pch_dev_ops {
int (*hw_init)(struct pch_thermal_device *ptd, int *nr_trips);
int (*get_temp)(struct pch_thermal_device *ptd, int *temp);
int (*suspend)(struct pch_thermal_device *ptd);
int (*resume)(struct pch_thermal_device *ptd);
};
/* dev ops for Wildcat Point */
static const struct pch_dev_ops pch_dev_ops_wpt = {
.hw_init = pch_wpt_init,
.get_temp = pch_wpt_get_temp,
.suspend = pch_wpt_suspend,
.resume = pch_wpt_resume,
};
static int pch_thermal_get_temp(struct thermal_zone_device *tzd, int *temp)
{
struct pch_thermal_device *ptd = tzd->devdata;
return ptd->ops->get_temp(ptd, temp);
}
static void pch_critical(struct thermal_zone_device *tzd)
{
dev_dbg(&tzd->device, "%s: critical temperature reached\n", tzd->type);
}
static struct thermal_zone_device_ops tzd_ops = {
.get_temp = pch_thermal_get_temp,
.critical = pch_critical,
};
enum board_ids {
board_hsw,
board_wpt,
board_skl,
board_cnl,
board_cml,
board_lwb,
board_wbg,
};
static const struct board_info {
const char *name;
const struct pch_dev_ops *ops;
} board_info[] = {
[board_hsw] = {
.name = "pch_haswell",
.ops = &pch_dev_ops_wpt,
},
[board_wpt] = {
.name = "pch_wildcat_point",
.ops = &pch_dev_ops_wpt,
},
[board_skl] = {
.name = "pch_skylake",
.ops = &pch_dev_ops_wpt,
},
[board_cnl] = {
.name = "pch_cannonlake",
.ops = &pch_dev_ops_wpt,
},
[board_cml] = {
.name = "pch_cometlake",
.ops = &pch_dev_ops_wpt,
},
[board_lwb] = {
.name = "pch_lewisburg",
.ops = &pch_dev_ops_wpt,
},
[board_wbg] = {
.name = "pch_wellsburg",
.ops = &pch_dev_ops_wpt,
},
};
static int intel_pch_thermal_probe(struct pci_dev *pdev,
const struct pci_device_id *id)
{
enum board_ids board_id = id->driver_data;
const struct board_info *bi = &board_info[board_id];
struct pch_thermal_device *ptd;
int err;
int nr_trips;
ptd = devm_kzalloc(&pdev->dev, sizeof(*ptd), GFP_KERNEL);
if (!ptd)
return -ENOMEM;
ptd->ops = bi->ops;
pci_set_drvdata(pdev, ptd);
ptd->pdev = pdev;
err = pci_enable_device(pdev);
if (err) {
dev_err(&pdev->dev, "failed to enable pci device\n");
return err;
}
err = pci_request_regions(pdev, driver_name);
if (err) {
dev_err(&pdev->dev, "failed to request pci region\n");
goto error_disable;
}
ptd->hw_base = pci_ioremap_bar(pdev, 0);
if (!ptd->hw_base) {
err = -ENOMEM;
dev_err(&pdev->dev, "failed to map mem base\n");
goto error_release;
}
err = ptd->ops->hw_init(ptd, &nr_trips);
if (err)
goto error_cleanup;
ptd->tzd = thermal_zone_device_register_with_trips(bi->name, ptd->trips,
nr_trips, 0, ptd,
&tzd_ops, NULL, 0, 0);
if (IS_ERR(ptd->tzd)) {
dev_err(&pdev->dev, "Failed to register thermal zone %s\n",
bi->name);
err = PTR_ERR(ptd->tzd);
goto error_cleanup;
}
err = thermal_zone_device_enable(ptd->tzd);
if (err)
goto err_unregister;
return 0;
err_unregister:
thermal_zone_device_unregister(ptd->tzd);
error_cleanup:
iounmap(ptd->hw_base);
error_release:
pci_release_regions(pdev);
error_disable:
pci_disable_device(pdev);
dev_err(&pdev->dev, "pci device failed to probe\n");
return err;
}
static void intel_pch_thermal_remove(struct pci_dev *pdev)
{
struct pch_thermal_device *ptd = pci_get_drvdata(pdev);
thermal_zone_device_unregister(ptd->tzd);
iounmap(ptd->hw_base);
pci_set_drvdata(pdev, NULL);
pci_release_regions(pdev);
pci_disable_device(pdev);
}
static int intel_pch_thermal_suspend_noirq(struct device *device)
{
struct pch_thermal_device *ptd = dev_get_drvdata(device);
return ptd->ops->suspend(ptd);
}
static int intel_pch_thermal_resume(struct device *device)
{
struct pch_thermal_device *ptd = dev_get_drvdata(device);
return ptd->ops->resume(ptd);
}
static const struct pci_device_id intel_pch_thermal_id[] = { static const struct pci_device_id intel_pch_thermal_id[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_HSW_1), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_HSW_1),
.driver_data = board_hsw, }, .driver_data = PCH_BOARD_HSW, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_HSW_2), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_HSW_2),
.driver_data = board_hsw, }, .driver_data = PCH_BOARD_HSW, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_WPT), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_WPT),
.driver_data = board_wpt, }, .driver_data = PCH_BOARD_WPT, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_SKL), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_SKL),
.driver_data = board_skl, }, .driver_data = PCH_BOARD_SKL, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_SKL_H), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_SKL_H),
.driver_data = board_skl, }, .driver_data = PCH_BOARD_SKL, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CNL), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CNL),
.driver_data = board_cnl, }, .driver_data = PCH_BOARD_CNL, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CNL_H), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CNL_H),
.driver_data = board_cnl, }, .driver_data = PCH_BOARD_CNL, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CNL_LP), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CNL_LP),
.driver_data = board_cnl, }, .driver_data = PCH_BOARD_CNL, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CML_H), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_CML_H),
.driver_data = board_cml, }, .driver_data = PCH_BOARD_CML, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_LWB), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_LWB),
.driver_data = board_lwb, }, .driver_data = PCH_BOARD_LWB, },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_WBG), { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCH_THERMAL_DID_WBG),
.driver_data = board_wbg, }, .driver_data = PCH_BOARD_WBG, },
{ 0, }, { 0, },
}; };
MODULE_DEVICE_TABLE(pci, intel_pch_thermal_id); MODULE_DEVICE_TABLE(pci, intel_pch_thermal_id);

571
drivers/thermal/intel/intel_powerclamp.c
Просмотреть файл

@ -2,7 +2,7 @@
/* /*
* intel_powerclamp.c - package c-state idle injection * intel_powerclamp.c - package c-state idle injection
* *
* Copyright (c) 2012, Intel Corporation. * Copyright (c) 2012-2023, Intel Corporation.
* *
* Authors: * Authors:
* Arjan van de Ven <arjan@linux.intel.com> * Arjan van de Ven <arjan@linux.intel.com>
@ -27,23 +27,17 @@
#include <linux/module.h> #include <linux/module.h>
#include <linux/kernel.h> #include <linux/kernel.h>
#include <linux/delay.h> #include <linux/delay.h>
#include <linux/kthread.h>
#include <linux/cpu.h> #include <linux/cpu.h>
#include <linux/thermal.h> #include <linux/thermal.h>
#include <linux/slab.h>
#include <linux/tick.h>
#include <linux/debugfs.h> #include <linux/debugfs.h>
#include <linux/seq_file.h> #include <linux/seq_file.h>
#include <linux/sched/rt.h> #include <linux/idle_inject.h>
#include <uapi/linux/sched/types.h>
#include <asm/nmi.h>
#include <asm/msr.h> #include <asm/msr.h>
#include <asm/mwait.h> #include <asm/mwait.h>
#include <asm/cpu_device_id.h> #include <asm/cpu_device_id.h>
#include <asm/hardirq.h>
#define MAX_TARGET_RATIO (50U) #define MAX_TARGET_RATIO (100U)
/* For each undisturbed clamping period (no extra wake ups during idle time), /* For each undisturbed clamping period (no extra wake ups during idle time),
* we increment the confidence counter for the given target ratio. * we increment the confidence counter for the given target ratio.
* CONFIDENCE_OK defines the level where runtime calibration results are * CONFIDENCE_OK defines the level where runtime calibration results are
@ -57,37 +51,30 @@
static unsigned int target_mwait; static unsigned int target_mwait;
static struct dentry *debug_dir; static struct dentry *debug_dir;
static bool poll_pkg_cstate_enable;
/* user selected target */ /* Idle ratio observed using package C-state counters */
static unsigned int set_target_ratio;
static unsigned int current_ratio; static unsigned int current_ratio;
/* Skip the idle injection till set to true */
static bool should_skip; static bool should_skip;
static unsigned int control_cpu; /* The cpu assigned to collect stat and update struct powerclamp_data {
* control parameters. default to BSP but BSP
* can be offlined.
*/
static bool clamping;
struct powerclamp_worker_data {
struct kthread_worker *worker;
struct kthread_work balancing_work;
struct kthread_delayed_work idle_injection_work;
unsigned int cpu; unsigned int cpu;
unsigned int count; unsigned int count;
unsigned int guard; unsigned int guard;
unsigned int window_size_now; unsigned int window_size_now;
unsigned int target_ratio; unsigned int target_ratio;
unsigned int duration_jiffies;
bool clamping; bool clamping;
}; };
static struct powerclamp_worker_data __percpu *worker_data; static struct powerclamp_data powerclamp_data;
static struct thermal_cooling_device *cooling_dev;
static unsigned long *cpu_clamping_mask; /* bit map for tracking per cpu
* clamping kthread worker
*/
static struct thermal_cooling_device *cooling_dev;
static DEFINE_MUTEX(powerclamp_lock);
/* This duration is in microseconds */
static unsigned int duration; static unsigned int duration;
static unsigned int pkg_cstate_ratio_cur; static unsigned int pkg_cstate_ratio_cur;
static unsigned int window_size; static unsigned int window_size;
@ -104,24 +91,170 @@ static int duration_set(const char *arg, const struct kernel_param *kp)
pr_err("Out of recommended range %lu, between 6-25ms\n", pr_err("Out of recommended range %lu, between 6-25ms\n",
new_duration); new_duration);
ret = -EINVAL; ret = -EINVAL;
goto exit;
} }
duration = clamp(new_duration, 6ul, 25ul); mutex_lock(&powerclamp_lock);
smp_mb(); duration = clamp(new_duration, 6ul, 25ul) * 1000;
mutex_unlock(&powerclamp_lock);
exit: exit:
return ret; return ret;
} }
static int duration_get(char *buf, const struct kernel_param *kp)
{
int ret;
mutex_lock(&powerclamp_lock);
ret = sysfs_emit(buf, "%d\n", duration / 1000);
mutex_unlock(&powerclamp_lock);
return ret;
}
static const struct kernel_param_ops duration_ops = { static const struct kernel_param_ops duration_ops = {
.set = duration_set, .set = duration_set,
.get = duration_get,
};
module_param_cb(duration, &duration_ops, NULL, 0644);
MODULE_PARM_DESC(duration, "forced idle time for each attempt in msec.");
#define DEFAULT_MAX_IDLE 50
#define MAX_ALL_CPU_IDLE 75
static u8 max_idle = DEFAULT_MAX_IDLE;
static cpumask_var_t idle_injection_cpu_mask;
static int allocate_copy_idle_injection_mask(const struct cpumask *copy_mask)
{
if (cpumask_available(idle_injection_cpu_mask))
goto copy_mask;
/* This mask is allocated only one time and freed during module exit */
if (!alloc_cpumask_var(&idle_injection_cpu_mask, GFP_KERNEL))
return -ENOMEM;
copy_mask:
cpumask_copy(idle_injection_cpu_mask, copy_mask);
return 0;
}
/* Return true if the cpumask and idle percent combination is invalid */
static bool check_invalid(cpumask_var_t mask, u8 idle)
{
if (cpumask_equal(cpu_present_mask, mask) && idle > MAX_ALL_CPU_IDLE)
return true;
return false;
}
static int cpumask_set(const char *arg, const struct kernel_param *kp)
{
cpumask_var_t new_mask;
int ret;
mutex_lock(&powerclamp_lock);
/* Can't set mask when cooling device is in use */
if (powerclamp_data.clamping) {
ret = -EAGAIN;
goto skip_cpumask_set;
}
ret = alloc_cpumask_var(&new_mask, GFP_KERNEL);
if (!ret)
goto skip_cpumask_set;
ret = bitmap_parse(arg, strlen(arg), cpumask_bits(new_mask),
nr_cpumask_bits);
if (ret)
goto free_cpumask_set;
if (cpumask_empty(new_mask) || check_invalid(new_mask, max_idle)) {
ret = -EINVAL;
goto free_cpumask_set;
}
/*
* When module parameters are passed from kernel command line
* during insmod, the module parameter callback is called
* before powerclamp_init(), so we can't assume that some
* cpumask can be allocated and copied before here. Also
* in this case this cpumask is used as the default mask.
*/
ret = allocate_copy_idle_injection_mask(new_mask);
free_cpumask_set:
free_cpumask_var(new_mask);
skip_cpumask_set:
mutex_unlock(&powerclamp_lock);
return ret;
}
static int cpumask_get(char *buf, const struct kernel_param *kp)
{
if (!cpumask_available(idle_injection_cpu_mask))
return -ENODEV;
return bitmap_print_to_pagebuf(false, buf, cpumask_bits(idle_injection_cpu_mask),
nr_cpumask_bits);
}
static const struct kernel_param_ops cpumask_ops = {
.set = cpumask_set,
.get = cpumask_get,
};
module_param_cb(cpumask, &cpumask_ops, NULL, 0644);
MODULE_PARM_DESC(cpumask, "Mask of CPUs to use for idle injection.");
static int max_idle_set(const char *arg, const struct kernel_param *kp)
{
u8 new_max_idle;
int ret = 0;
mutex_lock(&powerclamp_lock);
/* Can't set mask when cooling device is in use */
if (powerclamp_data.clamping) {
ret = -EAGAIN;
goto skip_limit_set;
}
ret = kstrtou8(arg, 10, &new_max_idle);
if (ret)
goto skip_limit_set;
if (new_max_idle > MAX_TARGET_RATIO) {
ret = -EINVAL;
goto skip_limit_set;
}
if (check_invalid(idle_injection_cpu_mask, new_max_idle)) {
ret = -EINVAL;
goto skip_limit_set;
}
max_idle = new_max_idle;
skip_limit_set:
mutex_unlock(&powerclamp_lock);
return ret;
}
static const struct kernel_param_ops max_idle_ops = {
.set = max_idle_set,
.get = param_get_int, .get = param_get_int,
}; };
module_param_cb(max_idle, &max_idle_ops, &max_idle, 0644);
module_param_cb(duration, &duration_ops, &duration, 0644); MODULE_PARM_DESC(max_idle, "maximum injected idle time to the total CPU time ratio in percent range:1-100");
MODULE_PARM_DESC(duration, "forced idle time for each attempt in msec.");
struct powerclamp_calibration_data { struct powerclamp_calibration_data {
unsigned long confidence; /* used for calibration, basically a counter unsigned long confidence; /* used for calibration, basically a counter
@ -261,6 +394,9 @@ static unsigned int get_compensation(int ratio)
{ {
unsigned int comp = 0; unsigned int comp = 0;
if (!poll_pkg_cstate_enable)
return 0;
/* we only use compensation if all adjacent ones are good */ /* we only use compensation if all adjacent ones are good */
if (ratio == 1 && if (ratio == 1 &&
cal_data[ratio].confidence >= CONFIDENCE_OK && cal_data[ratio].confidence >= CONFIDENCE_OK &&
@ -302,7 +438,7 @@ static void adjust_compensation(int target_ratio, unsigned int win)
if (d->confidence >= CONFIDENCE_OK) if (d->confidence >= CONFIDENCE_OK)
return; return;
delta = set_target_ratio - current_ratio; delta = powerclamp_data.target_ratio - current_ratio;
/* filter out bad data */ /* filter out bad data */
if (delta >= 0 && delta <= (1+target_ratio/10)) { if (delta >= 0 && delta <= (1+target_ratio/10)) {
if (d->steady_comp) if (d->steady_comp)
@ -341,82 +477,39 @@ static bool powerclamp_adjust_controls(unsigned int target_ratio,
adjust_compensation(target_ratio, win); adjust_compensation(target_ratio, win);
/* if we are above target+guard, skip */ /* if we are above target+guard, skip */
return set_target_ratio + guard <= current_ratio; return powerclamp_data.target_ratio + guard <= current_ratio;
} }
static void clamp_balancing_func(struct kthread_work *work) /*
* This function calculates runtime from the current target ratio.
* This function gets called under powerclamp_lock.
*/
static unsigned int get_run_time(void)
{ {
struct powerclamp_worker_data *w_data;
int sleeptime;
unsigned long target_jiffies;
unsigned int compensated_ratio; unsigned int compensated_ratio;
int interval; /* jiffies to sleep for each attempt */ unsigned int runtime;
w_data = container_of(work, struct powerclamp_worker_data,
balancing_work);
/* /*
* make sure user selected ratio does not take effect until * make sure user selected ratio does not take effect until
* the next round. adjust target_ratio if user has changed * the next round. adjust target_ratio if user has changed
* target such that we can converge quickly. * target such that we can converge quickly.
*/ */
w_data->target_ratio = READ_ONCE(set_target_ratio); powerclamp_data.guard = 1 + powerclamp_data.target_ratio / 20;
w_data->guard = 1 + w_data->target_ratio / 20; powerclamp_data.window_size_now = window_size;
w_data->window_size_now = window_size;
w_data->duration_jiffies = msecs_to_jiffies(duration);
w_data->count++;
/* /*
* systems may have different ability to enter package level * systems may have different ability to enter package level
* c-states, thus we need to compensate the injected idle ratio * c-states, thus we need to compensate the injected idle ratio
* to achieve the actual target reported by the HW. * to achieve the actual target reported by the HW.
*/ */
compensated_ratio = w_data->target_ratio + compensated_ratio = powerclamp_data.target_ratio +
get_compensation(w_data->target_ratio); get_compensation(powerclamp_data.target_ratio);
if (compensated_ratio <= 0) if (compensated_ratio <= 0)
compensated_ratio = 1; compensated_ratio = 1;
interval = w_data->duration_jiffies * 100 / compensated_ratio;
/* align idle time */ runtime = duration * 100 / compensated_ratio - duration;
target_jiffies = roundup(jiffies, interval);
sleeptime = target_jiffies - jiffies;
if (sleeptime <= 0)
sleeptime = 1;
if (clamping && w_data->clamping && cpu_online(w_data->cpu)) return runtime;
kthread_queue_delayed_work(w_data->worker,
&w_data->idle_injection_work,
sleeptime);
}
static void clamp_idle_injection_func(struct kthread_work *work)
{
struct powerclamp_worker_data *w_data;
w_data = container_of(work, struct powerclamp_worker_data,
idle_injection_work.work);
/*
* only elected controlling cpu can collect stats and update
* control parameters.
*/
if (w_data->cpu == control_cpu &&
!(w_data->count % w_data->window_size_now)) {
should_skip =
powerclamp_adjust_controls(w_data->target_ratio,
w_data->guard,
w_data->window_size_now);
smp_mb();
}
if (should_skip)
goto balance;
play_idle(jiffies_to_usecs(w_data->duration_jiffies));
balance:
if (clamping && w_data->clamping && cpu_online(w_data->cpu))
kthread_queue_work(w_data->worker, &w_data->balancing_work);
} }
/* /*
@ -452,128 +545,131 @@ static void poll_pkg_cstate(struct work_struct *dummy)
msr_last = msr_now; msr_last = msr_now;
tsc_last = tsc_now; tsc_last = tsc_now;
if (true == clamping) mutex_lock(&powerclamp_lock);
if (powerclamp_data.clamping)
schedule_delayed_work(&poll_pkg_cstate_work, HZ); schedule_delayed_work(&poll_pkg_cstate_work, HZ);
mutex_unlock(&powerclamp_lock);
} }
static void start_power_clamp_worker(unsigned long cpu) static struct idle_inject_device *ii_dev;
{
struct powerclamp_worker_data *w_data = per_cpu_ptr(worker_data, cpu);
struct kthread_worker *worker;
worker = kthread_create_worker_on_cpu(cpu, 0, "kidle_inj/%ld", cpu); /*
if (IS_ERR(worker)) * This function is called from idle injection core on timer expiry
* for the run duration. This allows powerclamp to readjust or skip
* injecting idle for this cycle.
*/
static bool idle_inject_update(void)
{
bool update = false;
/* We can't sleep in this callback */
if (!mutex_trylock(&powerclamp_lock))
return true;
if (!(powerclamp_data.count % powerclamp_data.window_size_now)) {
should_skip = powerclamp_adjust_controls(powerclamp_data.target_ratio,
powerclamp_data.guard,
powerclamp_data.window_size_now);
update = true;
}
if (update) {
unsigned int runtime = get_run_time();
idle_inject_set_duration(ii_dev, runtime, duration);
}
powerclamp_data.count++;
mutex_unlock(&powerclamp_lock);
if (should_skip)
return false;
return true;
}
/* This function starts idle injection by calling idle_inject_start() */
static void trigger_idle_injection(void)
{
unsigned int runtime = get_run_time();
idle_inject_set_duration(ii_dev, runtime, duration);
idle_inject_start(ii_dev);
powerclamp_data.clamping = true;
}
/*
* This function is called from start_power_clamp() to register
* CPUS with powercap idle injection register and set default
* idle duration and latency.
*/
static int powerclamp_idle_injection_register(void)
{
poll_pkg_cstate_enable = false;
if (cpumask_equal(cpu_present_mask, idle_injection_cpu_mask)) {
ii_dev = idle_inject_register_full(idle_injection_cpu_mask, idle_inject_update);
if (topology_max_packages() == 1 && topology_max_die_per_package() == 1)
poll_pkg_cstate_enable = true;
} else {
ii_dev = idle_inject_register(idle_injection_cpu_mask);
}
if (!ii_dev) {
pr_err("powerclamp: idle_inject_register failed\n");
return -EAGAIN;
}
idle_inject_set_duration(ii_dev, TICK_USEC, duration);
idle_inject_set_latency(ii_dev, UINT_MAX);
return 0;
}
/*
* This function is called from end_power_clamp() to stop idle injection
* and unregister CPUS from powercap idle injection core.
*/
static void remove_idle_injection(void)
{
if (!powerclamp_data.clamping)
return; return;
w_data->worker = worker; powerclamp_data.clamping = false;
w_data->count = 0; idle_inject_stop(ii_dev);
w_data->cpu = cpu;
w_data->clamping = true;
set_bit(cpu, cpu_clamping_mask);
sched_set_fifo(worker->task);
kthread_init_work(&w_data->balancing_work, clamp_balancing_func);
kthread_init_delayed_work(&w_data->idle_injection_work,
clamp_idle_injection_func);
kthread_queue_work(w_data->worker, &w_data->balancing_work);
}
static void stop_power_clamp_worker(unsigned long cpu)
{
struct powerclamp_worker_data *w_data = per_cpu_ptr(worker_data, cpu);
if (!w_data->worker)
return;
w_data->clamping = false;
/*
* Make sure that all works that get queued after this point see
* the clamping disabled. The counter part is not needed because
* there is an implicit memory barrier when the queued work
* is proceed.
*/
smp_wmb();
kthread_cancel_work_sync(&w_data->balancing_work);
kthread_cancel_delayed_work_sync(&w_data->idle_injection_work);
/*
* The balancing work still might be queued here because
* the handling of the "clapming" variable, cancel, and queue
* operations are not synchronized via a lock. But it is not
* a big deal. The balancing work is fast and destroy kthread
* will wait for it.
*/
clear_bit(w_data->cpu, cpu_clamping_mask);
kthread_destroy_worker(w_data->worker);
w_data->worker = NULL;
} }
/*
* This function is called when user change the cooling device
* state from zero to some other value.
*/
static int start_power_clamp(void) static int start_power_clamp(void)
{ {
unsigned long cpu; int ret;
set_target_ratio = clamp(set_target_ratio, 0U, MAX_TARGET_RATIO - 1); ret = powerclamp_idle_injection_register();
/* prevent cpu hotplug */ if (!ret) {
cpus_read_lock(); trigger_idle_injection();
if (poll_pkg_cstate_enable)
/* prefer BSP */ schedule_delayed_work(&poll_pkg_cstate_work, 0);
control_cpu = cpumask_first(cpu_online_mask);
clamping = true;
schedule_delayed_work(&poll_pkg_cstate_work, 0);
/* start one kthread worker per online cpu */
for_each_online_cpu(cpu) {
start_power_clamp_worker(cpu);
} }
cpus_read_unlock();
return 0; return ret;
} }
/*
* This function is called when user change the cooling device
* state from non zero value zero.
*/
static void end_power_clamp(void) static void end_power_clamp(void)
{ {
int i; if (powerclamp_data.clamping) {
remove_idle_injection();
/* idle_inject_unregister(ii_dev);
* Block requeuing in all the kthread workers. They will flush and
* stop faster.
*/
clamping = false;
for_each_set_bit(i, cpu_clamping_mask, num_possible_cpus()) {
pr_debug("clamping worker for cpu %d alive, destroy\n", i);
stop_power_clamp_worker(i);
} }
} }
static int powerclamp_cpu_online(unsigned int cpu)
{
if (clamping == false)
return 0;
start_power_clamp_worker(cpu);
/* prefer BSP as controlling CPU */
if (cpu == 0) {
control_cpu = 0;
smp_mb();
}
return 0;
}
static int powerclamp_cpu_predown(unsigned int cpu)
{
if (clamping == false)
return 0;
stop_power_clamp_worker(cpu);
if (cpu != control_cpu)
return 0;
control_cpu = cpumask_first(cpu_online_mask);
if (control_cpu == cpu)
control_cpu = cpumask_next(cpu, cpu_online_mask);
smp_mb();
return 0;
}
static int powerclamp_get_max_state(struct thermal_cooling_device *cdev, static int powerclamp_get_max_state(struct thermal_cooling_device *cdev,
unsigned long *state) unsigned long *state)
{ {
@ -585,11 +681,9 @@ static int powerclamp_get_max_state(struct thermal_cooling_device *cdev,
static int powerclamp_get_cur_state(struct thermal_cooling_device *cdev, static int powerclamp_get_cur_state(struct thermal_cooling_device *cdev,
unsigned long *state) unsigned long *state)
{ {
if (true == clamping) mutex_lock(&powerclamp_lock);
*state = pkg_cstate_ratio_cur; *state = powerclamp_data.target_ratio;
else mutex_unlock(&powerclamp_lock);
/* to save power, do not poll idle ratio while not clamping */
*state = -1; /* indicates invalid state */
return 0; return 0;
} }
@ -599,24 +693,32 @@ static int powerclamp_set_cur_state(struct thermal_cooling_device *cdev,
{ {
int ret = 0; int ret = 0;
mutex_lock(&powerclamp_lock);
new_target_ratio = clamp(new_target_ratio, 0UL, new_target_ratio = clamp(new_target_ratio, 0UL,
(unsigned long) (MAX_TARGET_RATIO-1)); (unsigned long) (max_idle - 1));
if (set_target_ratio == 0 && new_target_ratio > 0) { if (!powerclamp_data.target_ratio && new_target_ratio > 0) {
pr_info("Start idle injection to reduce power\n"); pr_info("Start idle injection to reduce power\n");
set_target_ratio = new_target_ratio; powerclamp_data.target_ratio = new_target_ratio;
ret = start_power_clamp(); ret = start_power_clamp();
if (ret)
powerclamp_data.target_ratio = 0;
goto exit_set; goto exit_set;
} else if (set_target_ratio > 0 && new_target_ratio == 0) { } else if (powerclamp_data.target_ratio > 0 && new_target_ratio == 0) {
pr_info("Stop forced idle injection\n"); pr_info("Stop forced idle injection\n");
end_power_clamp(); end_power_clamp();
set_target_ratio = 0; powerclamp_data.target_ratio = 0;
} else /* adjust currently running */ { } else /* adjust currently running */ {
set_target_ratio = new_target_ratio; unsigned int runtime;
/* make new set_target_ratio visible to other cpus */
smp_mb(); powerclamp_data.target_ratio = new_target_ratio;
runtime = get_run_time();
idle_inject_set_duration(ii_dev, runtime, duration);
} }
exit_set: exit_set:
mutex_unlock(&powerclamp_lock);
return ret; return ret;
} }
@ -657,7 +759,6 @@ static int powerclamp_debug_show(struct seq_file *m, void *unused)
{ {
int i = 0; int i = 0;
seq_printf(m, "controlling cpu: %d\n", control_cpu);
seq_printf(m, "pct confidence steady dynamic (compensation)\n"); seq_printf(m, "pct confidence steady dynamic (compensation)\n");
for (i = 0; i < MAX_TARGET_RATIO; i++) { for (i = 0; i < MAX_TARGET_RATIO; i++) {
seq_printf(m, "%d\t%lu\t%lu\t%lu\n", seq_printf(m, "%d\t%lu\t%lu\t%lu\n",
@ -680,75 +781,57 @@ static inline void powerclamp_create_debug_files(void)
&powerclamp_debug_fops); &powerclamp_debug_fops);
} }
static enum cpuhp_state hp_state;
static int __init powerclamp_init(void) static int __init powerclamp_init(void)
{ {
int retval; int retval;
cpu_clamping_mask = bitmap_zalloc(num_possible_cpus(), GFP_KERNEL);
if (!cpu_clamping_mask)
return -ENOMEM;
/* probe cpu features and ids here */ /* probe cpu features and ids here */
retval = powerclamp_probe(); retval = powerclamp_probe();
if (retval) if (retval)
goto exit_free; return retval;
mutex_lock(&powerclamp_lock);
retval = allocate_copy_idle_injection_mask(cpu_present_mask);
mutex_unlock(&powerclamp_lock);
if (retval)
return retval;
/* set default limit, maybe adjusted during runtime based on feedback */ /* set default limit, maybe adjusted during runtime based on feedback */
window_size = 2; window_size = 2;
retval = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
"thermal/intel_powerclamp:online",
powerclamp_cpu_online,
powerclamp_cpu_predown);
if (retval < 0)
goto exit_free;
hp_state = retval;
worker_data = alloc_percpu(struct powerclamp_worker_data);
if (!worker_data) {
retval = -ENOMEM;
goto exit_unregister;
}
cooling_dev = thermal_cooling_device_register("intel_powerclamp", NULL, cooling_dev = thermal_cooling_device_register("intel_powerclamp", NULL,
&powerclamp_cooling_ops); &powerclamp_cooling_ops);
if (IS_ERR(cooling_dev)) { if (IS_ERR(cooling_dev))
retval = -ENODEV; return -ENODEV;
goto exit_free_thread;
}
if (!duration) if (!duration)
duration = jiffies_to_msecs(DEFAULT_DURATION_JIFFIES); duration = jiffies_to_usecs(DEFAULT_DURATION_JIFFIES);
powerclamp_create_debug_files(); powerclamp_create_debug_files();
return 0; return 0;
exit_free_thread:
free_percpu(worker_data);
exit_unregister:
cpuhp_remove_state_nocalls(hp_state);
exit_free:
bitmap_free(cpu_clamping_mask);
return retval;
} }
module_init(powerclamp_init); module_init(powerclamp_init);
static void __exit powerclamp_exit(void) static void __exit powerclamp_exit(void)
{ {
mutex_lock(&powerclamp_lock);
end_power_clamp(); end_power_clamp();
cpuhp_remove_state_nocalls(hp_state); mutex_unlock(&powerclamp_lock);
free_percpu(worker_data);
thermal_cooling_device_unregister(cooling_dev); thermal_cooling_device_unregister(cooling_dev);
bitmap_free(cpu_clamping_mask);
cancel_delayed_work_sync(&poll_pkg_cstate_work); cancel_delayed_work_sync(&poll_pkg_cstate_work);
debugfs_remove_recursive(debug_dir); debugfs_remove_recursive(debug_dir);
if (cpumask_available(idle_injection_cpu_mask))
free_cpumask_var(idle_injection_cpu_mask);
} }
module_exit(powerclamp_exit); module_exit(powerclamp_exit);
MODULE_IMPORT_NS(IDLE_INJECT);
MODULE_LICENSE("GPL"); MODULE_LICENSE("GPL");
MODULE_AUTHOR("Arjan van de Ven <arjan@linux.intel.com>"); MODULE_AUTHOR("Arjan van de Ven <arjan@linux.intel.com>");
MODULE_AUTHOR("Jacob Pan <jacob.jun.pan@linux.intel.com>"); MODULE_AUTHOR("Jacob Pan <jacob.jun.pan@linux.intel.com>");

55
drivers/thermal/intel/intel_quark_dts_thermal.c
Просмотреть файл

@ -84,6 +84,7 @@
#define QRK_DTS_MASK_TP_THRES 0xFF #define QRK_DTS_MASK_TP_THRES 0xFF
#define QRK_DTS_SHIFT_TP 8 #define QRK_DTS_SHIFT_TP 8
#define QRK_DTS_ID_TP_CRITICAL 0 #define QRK_DTS_ID_TP_CRITICAL 0
#define QRK_DTS_ID_TP_HOT 1
#define QRK_DTS_SAFE_TP_THRES 105 #define QRK_DTS_SAFE_TP_THRES 105
/* Thermal Sensor Register Lock */ /* Thermal Sensor Register Lock */
@ -104,6 +105,7 @@ struct soc_sensor_entry {
u32 store_ptps; u32 store_ptps;
u32 store_dts_enable; u32 store_dts_enable;
struct thermal_zone_device *tzone; struct thermal_zone_device *tzone;
struct thermal_trip trips[QRK_MAX_DTS_TRIPS];
}; };
static struct soc_sensor_entry *soc_dts; static struct soc_sensor_entry *soc_dts;
@ -172,9 +174,9 @@ static int soc_dts_disable(struct thermal_zone_device *tzd)
return ret; return ret;
} }
static int _get_trip_temp(int trip, int *temp) static int get_trip_temp(int trip)
{ {
int status; int status, temp;
u32 out; u32 out;
mutex_lock(&dts_update_mutex); mutex_lock(&dts_update_mutex);
@ -183,7 +185,7 @@ static int _get_trip_temp(int trip, int *temp)
mutex_unlock(&dts_update_mutex); mutex_unlock(&dts_update_mutex);
if (status) if (status)
return status; return THERMAL_TEMP_INVALID;
/* /*
* Thermal Sensor Programmable Trip Point Register has 8-bit * Thermal Sensor Programmable Trip Point Register has 8-bit
@ -191,21 +193,10 @@ static int _get_trip_temp(int trip, int *temp)
* thresholds. The threshold value is always offset by its * thresholds. The threshold value is always offset by its
* temperature base (50 degree Celsius). * temperature base (50 degree Celsius).
*/ */
*temp = (out >> (trip * QRK_DTS_SHIFT_TP)) & QRK_DTS_MASK_TP_THRES; temp = (out >> (trip * QRK_DTS_SHIFT_TP)) & QRK_DTS_MASK_TP_THRES;
*temp -= QRK_DTS_TEMP_BASE; temp -= QRK_DTS_TEMP_BASE;
return 0; return temp;
}
static inline int sys_get_trip_temp(struct thermal_zone_device *tzd,
int trip, int *temp)
{
return _get_trip_temp(trip, temp);
}
static inline int sys_get_crit_temp(struct thermal_zone_device *tzd, int *temp)
{
return _get_trip_temp(QRK_DTS_ID_TP_CRITICAL, temp);
} }
static int update_trip_temp(struct soc_sensor_entry *aux_entry, static int update_trip_temp(struct soc_sensor_entry *aux_entry,
@ -262,17 +253,6 @@ static inline int sys_set_trip_temp(struct thermal_zone_device *tzd, int trip,
return update_trip_temp(tzd->devdata, trip, temp); return update_trip_temp(tzd->devdata, trip, temp);
} }
static int sys_get_trip_type(struct thermal_zone_device *thermal,
int trip, enum thermal_trip_type *type)
{
if (trip)
*type = THERMAL_TRIP_HOT;
else
*type = THERMAL_TRIP_CRITICAL;
return 0;
}
static int sys_get_curr_temp(struct thermal_zone_device *tzd, static int sys_get_curr_temp(struct thermal_zone_device *tzd,
int *temp) int *temp)
{ {
@ -315,10 +295,7 @@ static int sys_change_mode(struct thermal_zone_device *tzd,
static struct thermal_zone_device_ops tzone_ops = { static struct thermal_zone_device_ops tzone_ops = {
.get_temp = sys_get_curr_temp, .get_temp = sys_get_curr_temp,
.get_trip_temp = sys_get_trip_temp,
.get_trip_type = sys_get_trip_type,
.set_trip_temp = sys_set_trip_temp, .set_trip_temp = sys_set_trip_temp,
.get_crit_temp = sys_get_crit_temp,
.change_mode = sys_change_mode, .change_mode = sys_change_mode,
}; };
@ -385,10 +362,18 @@ static struct soc_sensor_entry *alloc_soc_dts(void)
goto err_ret; goto err_ret;
} }
aux_entry->tzone = thermal_zone_device_register("quark_dts", aux_entry->trips[QRK_DTS_ID_TP_CRITICAL].temperature = get_trip_temp(QRK_DTS_ID_TP_CRITICAL);
QRK_MAX_DTS_TRIPS, aux_entry->trips[QRK_DTS_ID_TP_CRITICAL].type = THERMAL_TRIP_CRITICAL;
wr_mask,
aux_entry, &tzone_ops, NULL, 0, polling_delay); aux_entry->trips[QRK_DTS_ID_TP_HOT].temperature = get_trip_temp(QRK_DTS_ID_TP_HOT);
aux_entry->trips[QRK_DTS_ID_TP_HOT].type = THERMAL_TRIP_HOT;
aux_entry->tzone = thermal_zone_device_register_with_trips("quark_dts",
aux_entry->trips,
QRK_MAX_DTS_TRIPS,
wr_mask,
aux_entry, &tzone_ops,
NULL, 0, polling_delay);
if (IS_ERR(aux_entry->tzone)) { if (IS_ERR(aux_entry->tzone)) {
err = PTR_ERR(aux_entry->tzone); err = PTR_ERR(aux_entry->tzone);
goto err_ret; goto err_ret;

108
drivers/thermal/thermal_acpi.c
Просмотреть файл

@ -21,42 +21,11 @@
#define TEMP_MIN_DECIK 2180 #define TEMP_MIN_DECIK 2180
#define TEMP_MAX_DECIK 4480 #define TEMP_MAX_DECIK 4480
static int thermal_acpi_trip_init(struct acpi_device *adev, static int thermal_acpi_trip_temp(struct acpi_device *adev, char *obj_name,
enum thermal_trip_type type, int id, int *ret_temp)
struct thermal_trip *trip)
{ {
unsigned long long temp; unsigned long long temp;
acpi_status status; acpi_status status;
char obj_name[5];
switch (type) {
case THERMAL_TRIP_ACTIVE:
if (id < 0 || id > 9)
return -EINVAL;
obj_name[1] = 'A';
obj_name[2] = 'C';
obj_name[3] = '0' + id;
break;
case THERMAL_TRIP_PASSIVE:
obj_name[1] = 'P';
obj_name[2] = 'S';
obj_name[3] = 'V';
break;
case THERMAL_TRIP_HOT:
obj_name[1] = 'H';
obj_name[2] = 'O';
obj_name[3] = 'T';
break;
case THERMAL_TRIP_CRITICAL:
obj_name[1] = 'C';
obj_name[2] = 'R';
obj_name[3] = 'T';
break;
}
obj_name[0] = '_';
obj_name[4] = '\0';
status = acpi_evaluate_integer(adev->handle, obj_name, NULL, &temp); status = acpi_evaluate_integer(adev->handle, obj_name, NULL, &temp);
if (ACPI_FAILURE(status)) { if (ACPI_FAILURE(status)) {
@ -65,87 +34,84 @@ static int thermal_acpi_trip_init(struct acpi_device *adev,
} }
if (temp >= TEMP_MIN_DECIK && temp <= TEMP_MAX_DECIK) { if (temp >= TEMP_MIN_DECIK && temp <= TEMP_MAX_DECIK) {
trip->temperature = deci_kelvin_to_millicelsius(temp); *ret_temp = deci_kelvin_to_millicelsius(temp);
} else { } else {
acpi_handle_debug(adev->handle, "%s result %llu out of range\n", acpi_handle_debug(adev->handle, "%s result %llu out of range\n",
obj_name, temp); obj_name, temp);
trip->temperature = THERMAL_TEMP_INVALID; *ret_temp = THERMAL_TEMP_INVALID;
} }
trip->hysteresis = 0;
trip->type = type;
return 0; return 0;
} }
/** /**
* thermal_acpi_trip_active - Get the specified active trip point * thermal_acpi_active_trip_temp - Retrieve active trip point temperature
* @adev: Thermal zone ACPI device object to get the description from. * @adev: Target thermal zone ACPI device object.
* @id: Active cooling level (0 - 9). * @id: Active cooling level (0 - 9).
* @trip: Trip point structure to be populated on success. * @ret_temp: Address to store the retrieved temperature value on success.
* *
* Evaluate the _ACx object for the thermal zone represented by @adev to obtain * Evaluate the _ACx object for the thermal zone represented by @adev to obtain
* the temperature of the active cooling trip point corresponding to the active * the temperature of the active cooling trip point corresponding to the active
* cooling level given by @id and initialize @trip as an active trip point using * cooling level given by @id.
* that temperature value.
* *
* Return 0 on success or a negative error value on failure. * Return 0 on success or a negative error value on failure.
*/ */
int thermal_acpi_trip_active(struct acpi_device *adev, int id, int thermal_acpi_active_trip_temp(struct acpi_device *adev, int id, int *ret_temp)
struct thermal_trip *trip)
{ {
return thermal_acpi_trip_init(adev, THERMAL_TRIP_ACTIVE, id, trip); char obj_name[] = {'_', 'A', 'C', '0' + id, '\0'};
if (id < 0 || id > 9)
return -EINVAL;
return thermal_acpi_trip_temp(adev, obj_name, ret_temp);
} }
EXPORT_SYMBOL_GPL(thermal_acpi_trip_active); EXPORT_SYMBOL_GPL(thermal_acpi_active_trip_temp);
/** /**
* thermal_acpi_trip_passive - Get the passive trip point * thermal_acpi_passive_trip_temp - Retrieve passive trip point temperature
* @adev: Thermal zone ACPI device object to get the description from. * @adev: Target thermal zone ACPI device object.
* @trip: Trip point structure to be populated on success. * @ret_temp: Address to store the retrieved temperature value on success.
* *
* Evaluate the _PSV object for the thermal zone represented by @adev to obtain * Evaluate the _PSV object for the thermal zone represented by @adev to obtain
* the temperature of the passive cooling trip point and initialize @trip as a * the temperature of the passive cooling trip point.
* passive trip point using that temperature value.
* *
* Return 0 on success or -ENODATA on failure. * Return 0 on success or -ENODATA on failure.
*/ */
int thermal_acpi_trip_passive(struct acpi_device *adev, struct thermal_trip *trip) int thermal_acpi_passive_trip_temp(struct acpi_device *adev, int *ret_temp)
{ {
return thermal_acpi_trip_init(adev, THERMAL_TRIP_PASSIVE, INT_MAX, trip); return thermal_acpi_trip_temp(adev, "_PSV", ret_temp);
} }
EXPORT_SYMBOL_GPL(thermal_acpi_trip_passive); EXPORT_SYMBOL_GPL(thermal_acpi_passive_trip_temp);
/** /**
* thermal_acpi_trip_hot - Get the near critical trip point * thermal_acpi_hot_trip_temp - Retrieve hot trip point temperature
* @adev: the ACPI device to get the description from. * @adev: Target thermal zone ACPI device object.
* @trip: a &struct thermal_trip to be filled if the function succeed. * @ret_temp: Address to store the retrieved temperature value on success.
* *
* Evaluate the _HOT object for the thermal zone represented by @adev to obtain * Evaluate the _HOT object for the thermal zone represented by @adev to obtain
* the temperature of the trip point at which the system is expected to be put * the temperature of the trip point at which the system is expected to be put
* into the S4 sleep state and initialize @trip as a hot trip point using that * into the S4 sleep state.
* temperature value.
* *
* Return 0 on success or -ENODATA on failure. * Return 0 on success or -ENODATA on failure.
*/ */
int thermal_acpi_trip_hot(struct acpi_device *adev, struct thermal_trip *trip) int thermal_acpi_hot_trip_temp(struct acpi_device *adev, int *ret_temp)
{ {
return thermal_acpi_trip_init(adev, THERMAL_TRIP_HOT, INT_MAX, trip); return thermal_acpi_trip_temp(adev, "_HOT", ret_temp);
} }
EXPORT_SYMBOL_GPL(thermal_acpi_trip_hot); EXPORT_SYMBOL_GPL(thermal_acpi_hot_trip_temp);
/** /**
* thermal_acpi_trip_critical - Get the critical trip point * thermal_acpi_critical_trip_temp - Retrieve critical trip point temperature
* @adev: the ACPI device to get the description from. * @adev: Target thermal zone ACPI device object.
* @trip: a &struct thermal_trip to be filled if the function succeed. * @ret_temp: Address to store the retrieved temperature value on success.
* *
* Evaluate the _CRT object for the thermal zone represented by @adev to obtain * Evaluate the _CRT object for the thermal zone represented by @adev to obtain
* the temperature of the critical cooling trip point and initialize @trip as a * the temperature of the critical cooling trip point.
* critical trip point using that temperature value.
* *
* Return 0 on success or -ENODATA on failure. * Return 0 on success or -ENODATA on failure.
*/ */
int thermal_acpi_trip_critical(struct acpi_device *adev, struct thermal_trip *trip) int thermal_acpi_critical_trip_temp(struct acpi_device *adev, int *ret_temp)
{ {
return thermal_acpi_trip_init(adev, THERMAL_TRIP_CRITICAL, INT_MAX, trip); return thermal_acpi_trip_temp(adev, "_CRT", ret_temp);
} }
EXPORT_SYMBOL_GPL(thermal_acpi_trip_critical); EXPORT_SYMBOL_GPL(thermal_acpi_critical_trip_temp);

3
include/linux/idle_inject.h
Просмотреть файл

@ -13,6 +13,9 @@ struct idle_inject_device;
struct idle_inject_device *idle_inject_register(struct cpumask *cpumask); struct idle_inject_device *idle_inject_register(struct cpumask *cpumask);
struct idle_inject_device *idle_inject_register_full(struct cpumask *cpumask,
bool (*update)(void));
void idle_inject_unregister(struct idle_inject_device *ii_dev); void idle_inject_unregister(struct idle_inject_device *ii_dev);
int idle_inject_start(struct idle_inject_device *ii_dev); int idle_inject_start(struct idle_inject_device *ii_dev);

9
include/linux/thermal.h
Просмотреть файл

@ -347,11 +347,10 @@ int thermal_zone_get_num_trips(struct thermal_zone_device *tz);
int thermal_zone_get_crit_temp(struct thermal_zone_device *tz, int *temp); int thermal_zone_get_crit_temp(struct thermal_zone_device *tz, int *temp);
#ifdef CONFIG_THERMAL_ACPI #ifdef CONFIG_THERMAL_ACPI
int thermal_acpi_trip_active(struct acpi_device *adev, int id, int thermal_acpi_active_trip_temp(struct acpi_device *adev, int id, int *ret_temp);
struct thermal_trip *trip); int thermal_acpi_passive_trip_temp(struct acpi_device *adev, int *ret_temp);
int thermal_acpi_trip_passive(struct acpi_device *adev, struct thermal_trip *trip); int thermal_acpi_hot_trip_temp(struct acpi_device *adev, int *ret_temp);
int thermal_acpi_trip_hot(struct acpi_device *adev, struct thermal_trip *trip); int thermal_acpi_critical_trip_temp(struct acpi_device *adev, int *ret_temp);
int thermal_acpi_trip_critical(struct acpi_device *adev, struct thermal_trip *trip);
#endif #endif
#ifdef CONFIG_THERMAL #ifdef CONFIG_THERMAL