xref: /linux/Documentation/admin-guide/pm/intel_pstate.rst (revision 65b09bfa8aa7ebe087093b591525385efb2d58b0)
1.. SPDX-License-Identifier: GPL-2.0
2.. include:: <isonum.txt>
3
4===============================================
5``intel_pstate`` CPU Performance Scaling Driver
6===============================================
7
8:Copyright: |copy| 2017 Intel Corporation
9
10:Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
11
12
13General Information
14===================
15
16``intel_pstate`` is a part of the
17:doc:`CPU performance scaling subsystem <cpufreq>` in the Linux kernel
18(``CPUFreq``).  It is a scaling driver for the Sandy Bridge and later
19generations of Intel processors.  Note, however, that some of those processors
20may not be supported.  [To understand ``intel_pstate`` it is necessary to know
21how ``CPUFreq`` works in general, so this is the time to read
22Documentation/admin-guide/pm/cpufreq.rst if you have not done that yet.]
23
24For the processors supported by ``intel_pstate``, the P-state concept is broader
25than just an operating frequency or an operating performance point (see the
26LinuxCon Europe 2015 presentation by Kristen Accardi [1]_ for more
27information about that).  For this reason, the representation of P-states used
28by ``intel_pstate`` internally follows the hardware specification (for details
29refer to Intel Software Developer’s Manual [2]_).  However, the ``CPUFreq`` core
30uses frequencies for identifying operating performance points of CPUs and
31frequencies are involved in the user space interface exposed by it, so
32``intel_pstate`` maps its internal representation of P-states to frequencies too
33(fortunately, that mapping is unambiguous).  At the same time, it would not be
34practical for ``intel_pstate`` to supply the ``CPUFreq`` core with a table of
35available frequencies due to the possible size of it, so the driver does not do
36that.  Some functionality of the core is limited by that.
37
38Since the hardware P-state selection interface used by ``intel_pstate`` is
39available at the logical CPU level, the driver always works with individual
40CPUs.  Consequently, if ``intel_pstate`` is in use, every ``CPUFreq`` policy
41object corresponds to one logical CPU and ``CPUFreq`` policies are effectively
42equivalent to CPUs.  In particular, this means that they become "inactive" every
43time the corresponding CPU is taken offline and need to be re-initialized when
44it goes back online.
45
46``intel_pstate`` is not modular, so it cannot be unloaded, which means that the
47only way to pass early-configuration-time parameters to it is via the kernel
48command line.  However, its configuration can be adjusted via ``sysfs`` to a
49great extent.  In some configurations it even is possible to unregister it via
50``sysfs`` which allows another ``CPUFreq`` scaling driver to be loaded and
51registered (see :ref:`below <status_attr>`).
52
53.. _operation_modes:
54
55Operation Modes
56===============
57
58``intel_pstate`` can operate in two different modes, active or passive.  In the
59active mode, it uses its own internal performance scaling governor algorithm or
60allows the hardware to do performance scaling by itself, while in the passive
61mode it responds to requests made by a generic ``CPUFreq`` governor implementing
62a certain performance scaling algorithm.  Which of them will be in effect
63depends on what kernel command line options are used and on the capabilities of
64the processor.
65
66.. _active_mode:
67
68Active Mode
69-----------
70
71This is the default operation mode of ``intel_pstate`` for processors with
72hardware-managed P-states (HWP) support.  If it works in this mode, the
73``scaling_driver`` policy attribute in ``sysfs`` for all ``CPUFreq`` policies
74contains the string "intel_pstate".
75
76In this mode the driver bypasses the scaling governors layer of ``CPUFreq`` and
77provides its own scaling algorithms for P-state selection.  Those algorithms
78can be applied to ``CPUFreq`` policies in the same way as generic scaling
79governors (that is, through the ``scaling_governor`` policy attribute in
80``sysfs``).  [Note that different P-state selection algorithms may be chosen for
81different policies, but that is not recommended.]
82
83They are not generic scaling governors, but their names are the same as the
84names of some of those governors.  Moreover, confusingly enough, they generally
85do not work in the same way as the generic governors they share the names with.
86For example, the ``powersave`` P-state selection algorithm provided by
87``intel_pstate`` is not a counterpart of the generic ``powersave`` governor
88(roughly, it corresponds to the ``schedutil`` and ``ondemand`` governors).
89
90There are two P-state selection algorithms provided by ``intel_pstate`` in the
91active mode: ``powersave`` and ``performance``.  The way they both operate
92depends on whether or not the hardware-managed P-states (HWP) feature has been
93enabled in the processor and possibly on the processor model.
94
95Which of the P-state selection algorithms is used by default depends on the
96:c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE` kernel configuration option.
97Namely, if that option is set, the ``performance`` algorithm will be used by
98default, and the other one will be used by default if it is not set.
99
100.. _active_mode_hwp:
101
102Active Mode With HWP
103~~~~~~~~~~~~~~~~~~~~
104
105If the processor supports the HWP feature, it will be enabled during the
106processor initialization and cannot be disabled after that.  It is possible
107to avoid enabling it by passing the ``intel_pstate=no_hwp`` argument to the
108kernel in the command line.
109
110If the HWP feature has been enabled, ``intel_pstate`` relies on the processor to
111select P-states by itself, but still it can give hints to the processor's
112internal P-state selection logic.  What those hints are depends on which P-state
113selection algorithm has been applied to the given policy (or to the CPU it
114corresponds to).
115
116Even though the P-state selection is carried out by the processor automatically,
117``intel_pstate`` registers utilization update callbacks with the CPU scheduler
118in this mode.  However, they are not used for running a P-state selection
119algorithm, but for periodic updates of the current CPU frequency information to
120be made available from the ``scaling_cur_freq`` policy attribute in ``sysfs``.
121
122HWP + ``performance``
123.....................
124
125In this configuration ``intel_pstate`` will write 0 to the processor's
126Energy-Performance Preference (EPP) knob (if supported) or its
127Energy-Performance Bias (EPB) knob (otherwise), which means that the processor's
128internal P-state selection logic is expected to focus entirely on performance.
129
130This will override the EPP/EPB setting coming from the ``sysfs`` interface
131(see :ref:`energy_performance_hints` below).  Moreover, any attempts to change
132the EPP/EPB to a value different from 0 ("performance") via ``sysfs`` in this
133configuration will be rejected.
134
135Also, in this configuration the range of P-states available to the processor's
136internal P-state selection logic is always restricted to the upper boundary
137(that is, the maximum P-state that the driver is allowed to use).
138
139HWP + ``powersave``
140...................
141
142In this configuration ``intel_pstate`` will set the processor's
143Energy-Performance Preference (EPP) knob (if supported) or its
144Energy-Performance Bias (EPB) knob (otherwise) to whatever value it was
145previously set to via ``sysfs`` (or whatever default value it was
146set to by the platform firmware).  This usually causes the processor's
147internal P-state selection logic to be less performance-focused.
148
149Active Mode Without HWP
150~~~~~~~~~~~~~~~~~~~~~~~
151
152This operation mode is optional for processors that do not support the HWP
153feature or when the ``intel_pstate=no_hwp`` argument is passed to the kernel in
154the command line.  The active mode is used in those cases if the
155``intel_pstate=active`` argument is passed to the kernel in the command line.
156In this mode ``intel_pstate`` may refuse to work with processors that are not
157recognized by it.  [Note that ``intel_pstate`` will never refuse to work with
158any processor with the HWP feature enabled.]
159
160In this mode ``intel_pstate`` registers utilization update callbacks with the
161CPU scheduler in order to run a P-state selection algorithm, either
162``powersave`` or ``performance``, depending on the ``scaling_governor`` policy
163setting in ``sysfs``.  The current CPU frequency information to be made
164available from the ``scaling_cur_freq`` policy attribute in ``sysfs`` is
165periodically updated by those utilization update callbacks too.
166
167``performance``
168...............
169
170Without HWP, this P-state selection algorithm is always the same regardless of
171the processor model and platform configuration.
172
173It selects the maximum P-state it is allowed to use, subject to limits set via
174``sysfs``, every time the driver configuration for the given CPU is updated
175(e.g. via ``sysfs``).
176
177This is the default P-state selection algorithm if the
178:c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE` kernel configuration option
179is set.
180
181``powersave``
182.............
183
184Without HWP, this P-state selection algorithm is similar to the algorithm
185implemented by the generic ``schedutil`` scaling governor except that the
186utilization metric used by it is based on numbers coming from feedback
187registers of the CPU.  It generally selects P-states proportional to the
188current CPU utilization.
189
190This algorithm is run by the driver's utilization update callback for the
191given CPU when it is invoked by the CPU scheduler, but not more often than
192every 10 ms.  Like in the ``performance`` case, the hardware configuration
193is not touched if the new P-state turns out to be the same as the current
194one.
195
196This is the default P-state selection algorithm if the
197:c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE` kernel configuration option
198is not set.
199
200.. _passive_mode:
201
202Passive Mode
203------------
204
205This is the default operation mode of ``intel_pstate`` for processors without
206hardware-managed P-states (HWP) support.  It is always used if the
207``intel_pstate=passive`` argument is passed to the kernel in the command line
208regardless of whether or not the given processor supports HWP.  [Note that the
209``intel_pstate=no_hwp`` setting causes the driver to start in the passive mode
210if it is not combined with ``intel_pstate=active``.]  Like in the active mode
211without HWP support, in this mode ``intel_pstate`` may refuse to work with
212processors that are not recognized by it if HWP is prevented from being enabled
213through the kernel command line.
214
215If the driver works in this mode, the ``scaling_driver`` policy attribute in
216``sysfs`` for all ``CPUFreq`` policies contains the string "intel_cpufreq".
217Then, the driver behaves like a regular ``CPUFreq`` scaling driver.  That is,
218it is invoked by generic scaling governors when necessary to talk to the
219hardware in order to change the P-state of a CPU (in particular, the
220``schedutil`` governor can invoke it directly from scheduler context).
221
222While in this mode, ``intel_pstate`` can be used with all of the (generic)
223scaling governors listed by the ``scaling_available_governors`` policy attribute
224in ``sysfs`` (and the P-state selection algorithms described above are not
225used).  Then, it is responsible for the configuration of policy objects
226corresponding to CPUs and provides the ``CPUFreq`` core (and the scaling
227governors attached to the policy objects) with accurate information on the
228maximum and minimum operating frequencies supported by the hardware (including
229the so-called "turbo" frequency ranges).  In other words, in the passive mode
230the entire range of available P-states is exposed by ``intel_pstate`` to the
231``CPUFreq`` core.  However, in this mode the driver does not register
232utilization update callbacks with the CPU scheduler and the ``scaling_cur_freq``
233information comes from the ``CPUFreq`` core (and is the last frequency selected
234by the current scaling governor for the given policy).
235
236
237.. _turbo:
238
239Turbo P-states Support
240======================
241
242In the majority of cases, the entire range of P-states available to
243``intel_pstate`` can be divided into two sub-ranges that correspond to
244different types of processor behavior, above and below a boundary that
245will be referred to as the "turbo threshold" in what follows.
246
247The P-states above the turbo threshold are referred to as "turbo P-states" and
248the whole sub-range of P-states they belong to is referred to as the "turbo
249range".  These names are related to the Turbo Boost technology allowing a
250multicore processor to opportunistically increase the P-state of one or more
251cores if there is enough power to do that and if that is not going to cause the
252thermal envelope of the processor package to be exceeded.
253
254Specifically, if software sets the P-state of a CPU core within the turbo range
255(that is, above the turbo threshold), the processor is permitted to take over
256performance scaling control for that core and put it into turbo P-states of its
257choice going forward.  However, that permission is interpreted differently by
258different processor generations.  Namely, the Sandy Bridge generation of
259processors will never use any P-states above the last one set by software for
260the given core, even if it is within the turbo range, whereas all of the later
261processor generations will take it as a license to use any P-states from the
262turbo range, even above the one set by software.  In other words, on those
263processors setting any P-state from the turbo range will enable the processor
264to put the given core into all turbo P-states up to and including the maximum
265supported one as it sees fit.
266
267One important property of turbo P-states is that they are not sustainable.  More
268precisely, there is no guarantee that any CPUs will be able to stay in any of
269those states indefinitely, because the power distribution within the processor
270package may change over time  or the thermal envelope it was designed for might
271be exceeded if a turbo P-state was used for too long.
272
273In turn, the P-states below the turbo threshold generally are sustainable.  In
274fact, if one of them is set by software, the processor is not expected to change
275it to a lower one unless in a thermal stress or a power limit violation
276situation (a higher P-state may still be used if it is set for another CPU in
277the same package at the same time, for example).
278
279Some processors allow multiple cores to be in turbo P-states at the same time,
280but the maximum P-state that can be set for them generally depends on the number
281of cores running concurrently.  The maximum turbo P-state that can be set for 3
282cores at the same time usually is lower than the analogous maximum P-state for
2832 cores, which in turn usually is lower than the maximum turbo P-state that can
284be set for 1 core.  The one-core maximum turbo P-state is thus the maximum
285supported one overall.
286
287The maximum supported turbo P-state, the turbo threshold (the maximum supported
288non-turbo P-state) and the minimum supported P-state are specific to the
289processor model and can be determined by reading the processor's model-specific
290registers (MSRs).  Moreover, some processors support the Configurable TDP
291(Thermal Design Power) feature and, when that feature is enabled, the turbo
292threshold effectively becomes a configurable value that can be set by the
293platform firmware.
294
295Unlike ``_PSS`` objects in the ACPI tables, ``intel_pstate`` always exposes
296the entire range of available P-states, including the whole turbo range, to the
297``CPUFreq`` core and (in the passive mode) to generic scaling governors.  This
298generally causes turbo P-states to be set more often when ``intel_pstate`` is
299used relative to ACPI-based CPU performance scaling (see
300:ref:`below <acpi-cpufreq>` for more information).
301
302Moreover, since ``intel_pstate`` always knows what the real turbo threshold is
303(even if the Configurable TDP feature is enabled in the processor), its
304``no_turbo`` attribute in ``sysfs`` (described :ref:`below <no_turbo_attr>`) should
305work as expected in all cases (that is, if set to disable turbo P-states, it
306always should prevent ``intel_pstate`` from using them).
307
308
309Processor Support
310=================
311
312To handle a given processor ``intel_pstate`` requires a number of different
313pieces of information on it to be known, including:
314
315 * The minimum supported P-state.
316
317 * The maximum supported :ref:`non-turbo P-state <turbo>`.
318
319 * Whether or not turbo P-states are supported at all.
320
321 * The maximum supported :ref:`one-core turbo P-state <turbo>` (if turbo
322   P-states are supported).
323
324 * The scaling formula to translate the driver's internal representation
325   of P-states into frequencies and the other way around.
326
327Generally, ways to obtain that information are specific to the processor model
328or family.  Although it often is possible to obtain all of it from the processor
329itself (using model-specific registers), there are cases in which hardware
330manuals need to be consulted to get to it too.
331
332For this reason, there is a list of supported processors in ``intel_pstate`` and
333the driver initialization will fail if the detected processor is not in that
334list, unless it supports the HWP feature.  [The interface to obtain all of the
335information listed above is the same for all of the processors supporting the
336HWP feature, which is why ``intel_pstate`` works with all of them.]
337
338
339Support for Hybrid Processors
340=============================
341
342Some processors supported by ``intel_pstate`` contain two or more types of CPU
343cores differing by the maximum turbo P-state, performance vs power characteristics,
344cache sizes, and possibly other properties.  They are commonly referred to as
345hybrid processors.  To support them, ``intel_pstate`` requires HWP to be enabled
346and it assumes the HWP performance units to be the same for all CPUs in the
347system, so a given HWP performance level always represents approximately the
348same physical performance regardless of the core (CPU) type.
349
350Hybrid Processors with SMT
351--------------------------
352
353On systems where SMT (Simultaneous Multithreading), also referred to as
354HyperThreading (HT) in the context of Intel processors, is enabled on at least
355one core, ``intel_pstate`` assigns performance-based priorities to CPUs.  Namely,
356the priority of a given CPU reflects its highest HWP performance level which
357causes the CPU scheduler to generally prefer more performant CPUs, so the less
358performant CPUs are used when the other ones are fully loaded.  SMT siblings
359(that is, logical CPUs sharing one physical core) are given the same priority.
360The scheduler can pull tasks from lower-priority cores and place them on any
361sibling.  Since the scheduler spreads tasks among physical cores, tasks will be
362placed on the SMT siblings of physical cores only after all physical cores are
363busy.
364
365This approach maximizes performance in the majority of cases, but unfortunately
366it also leads to excessive energy usage in some important scenarios, like video
367playback, which is not generally desirable.  While there is no other viable
368choice with SMT enabled because the effective capacity and utilization of SMT
369siblings are hard to determine, hybrid processors without SMT can be handled in
370more energy-efficient ways.
371
372.. _CAS:
373
374Capacity-Aware Scheduling Support
375---------------------------------
376
377The capacity-aware scheduling (CAS) support in the CPU scheduler is enabled by
378``intel_pstate`` by default on hybrid processors without SMT.  CAS generally
379causes the scheduler to put tasks on a CPU so long as there is a sufficient
380amount of spare capacity on it, and if the utilization of a given task is too
381high for it, the task will need to go somewhere else.
382
383Since CAS takes CPU capacities into account, it does not require CPU
384prioritization and it allows tasks to be distributed more symmetrically among
385the more performant and less performant CPUs.  Once placed on a CPU with enough
386capacity to accommodate it, a task may just continue to run there regardless of
387whether or not the other CPUs are fully loaded, so on average CAS reduces the
388utilization of the more performant CPUs which causes the energy usage to be more
389balanced because the more performant CPUs are generally less energy-efficient
390than the less performant ones.
391
392In order to use CAS, the scheduler needs to know the capacity of each CPU in
393the system and it needs to be able to compute scale-invariant utilization of
394CPUs, so ``intel_pstate`` provides it with the requisite information.
395
396First of all, the capacity of each CPU is represented by the ratio of its highest
397HWP performance level, multiplied by 1024, to the highest HWP performance level
398of the most performant CPU in the system, which works because the HWP performance
399units are the same for all CPUs.  Second, the frequency-invariance computations,
400carried out by the scheduler to always express CPU utilization in the same units
401regardless of the frequency it is currently running at, are adjusted to take the
402CPU capacity into account.  All of this happens when ``intel_pstate`` has
403registered itself with the ``CPUFreq`` core and it has figured out that it is
404running on a hybrid processor without SMT.
405
406Energy-Aware Scheduling Support
407-------------------------------
408
409If ``CONFIG_ENERGY_MODEL`` has been set during kernel configuration and
410``intel_pstate`` runs on a hybrid processor without SMT, in addition to enabling
411:ref:`CAS` it registers an Energy Model for the processor.  This allows the
412Energy-Aware Scheduling (EAS) support to be enabled in the CPU scheduler if
413``schedutil`` is used as the  ``CPUFreq`` governor which requires ``intel_pstate``
414to operate in the :ref:`passive mode <passive_mode>`.
415
416The Energy Model registered by ``intel_pstate`` is artificial (that is, it is
417based on abstract cost values and it does not include any real power numbers)
418and it is relatively simple to avoid unnecessary computations in the scheduler.
419There is a performance domain in it for every CPU in the system and the cost
420values for these performance domains have been chosen so that running a task on
421a less performant (small) CPU appears to be always cheaper than running that
422task on a more performant (big) CPU.  However, for two CPUs of the same type,
423the cost difference depends on their current utilization, and the CPU whose
424current utilization is higher generally appears to be a more expensive
425destination for a given task.  This helps to balance the load among CPUs of the
426same type.
427
428Since EAS works on top of CAS, high-utilization tasks are always migrated to
429CPUs with enough capacity to accommodate them, but thanks to EAS, low-utilization
430tasks tend to be placed on the CPUs that look less expensive to the scheduler.
431Effectively, this causes the less performant and less loaded CPUs to be
432preferred as long as they have enough spare capacity to run the given task
433which generally leads to reduced energy usage.
434
435The Energy Model created by ``intel_pstate`` can be inspected by looking at
436the ``energy_model`` directory in ``debugfs`` (typlically mounted on
437``/sys/kernel/debug/``).
438
439
440User Space Interface in ``sysfs``
441=================================
442
443.. _global_attributes:
444
445Global Attributes
446-----------------
447
448``intel_pstate`` exposes several global attributes (files) in ``sysfs`` to
449control its functionality at the system level.  They are located in the
450``/sys/devices/system/cpu/intel_pstate/`` directory and affect all CPUs.
451
452Some of them are not present if the ``intel_pstate=per_cpu_perf_limits``
453argument is passed to the kernel in the command line.
454
455``max_perf_pct``
456	Maximum P-state the driver is allowed to set in percent of the
457	maximum supported performance level (the highest supported :ref:`turbo
458	P-state <turbo>`).
459
460	This attribute will not be exposed if the
461	``intel_pstate=per_cpu_perf_limits`` argument is present in the kernel
462	command line.
463
464``min_perf_pct``
465	Minimum P-state the driver is allowed to set in percent of the
466	maximum supported performance level (the highest supported :ref:`turbo
467	P-state <turbo>`).
468
469	This attribute will not be exposed if the
470	``intel_pstate=per_cpu_perf_limits`` argument is present in the kernel
471	command line.
472
473``num_pstates``
474	Number of P-states supported by the processor (between 0 and 255
475	inclusive) including both turbo and non-turbo P-states (see
476	:ref:`turbo`).
477
478	This attribute is present only if the value exposed by it is the same
479	for all of the CPUs in the system.
480
481	The value of this attribute is not affected by the ``no_turbo``
482	setting described :ref:`below <no_turbo_attr>`.
483
484	This attribute is read-only.
485
486``turbo_pct``
487	Ratio of the :ref:`turbo range <turbo>` size to the size of the entire
488	range of supported P-states, in percent.
489
490	This attribute is present only if the value exposed by it is the same
491	for all of the CPUs in the system.
492
493	This attribute is read-only.
494
495.. _no_turbo_attr:
496
497``no_turbo``
498	If set (equal to 1), the driver is not allowed to set any turbo P-states
499	(see :ref:`turbo`).  If unset (equal to 0, which is the
500	default), turbo P-states can be set by the driver.
501	[Note that ``intel_pstate`` does not support the general ``boost``
502	attribute (supported by some other scaling drivers) which is replaced
503	by this one.]
504
505	This attribute does not affect the maximum supported frequency value
506	supplied to the ``CPUFreq`` core and exposed via the policy interface,
507	but it affects the maximum possible value of per-policy P-state	limits
508	(see :ref:`policy_attributes_interpretation` below for details).
509
510``hwp_dynamic_boost``
511	This attribute is only present if ``intel_pstate`` works in the
512	:ref:`active mode with the HWP feature enabled <active_mode_hwp>` in
513	the processor.  If set (equal to 1), it causes the minimum P-state limit
514	to be increased dynamically for a short time whenever a task previously
515	waiting on I/O is selected to run on a given logical CPU (the purpose
516	of this mechanism is to improve performance).
517
518	This setting has no effect on logical CPUs whose minimum P-state limit
519	is directly set to the highest non-turbo P-state or above it.
520
521.. _status_attr:
522
523``status``
524	Operation mode of the driver: "active", "passive" or "off".
525
526	"active"
527		The driver is functional and in the :ref:`active mode
528		<active_mode>`.
529
530	"passive"
531		The driver is functional and in the :ref:`passive mode
532		<passive_mode>`.
533
534	"off"
535		The driver is not functional (it is not registered as a scaling
536		driver with the ``CPUFreq`` core).
537
538	This attribute can be written to in order to change the driver's
539	operation mode or to unregister it.  The string written to it must be
540	one of the possible values of it and, if successful, the write will
541	cause the driver to switch over to the operation mode represented by
542	that string - or to be unregistered in the "off" case.  [Actually,
543	switching over from the active mode to the passive mode or the other
544	way around causes the driver to be unregistered and registered again
545	with a different set of callbacks, so all of its settings (the global
546	as well as the per-policy ones) are then reset to their default
547	values, possibly depending on the target operation mode.]
548
549``energy_efficiency``
550	This attribute is only present on platforms with CPUs matching the Kaby
551	Lake or Coffee Lake desktop CPU model. By default, energy-efficiency
552	optimizations are disabled on these CPU models if HWP is enabled.
553	Enabling energy-efficiency optimizations may limit maximum operating
554	frequency with or without the HWP feature.  With HWP enabled, the
555	optimizations are done only in the turbo frequency range.  Without it,
556	they are done in the entire available frequency range.  Setting this
557	attribute to "1" enables the energy-efficiency optimizations and setting
558	to "0" disables them.
559
560.. _policy_attributes_interpretation:
561
562Interpretation of Policy Attributes
563-----------------------------------
564
565The interpretation of some ``CPUFreq`` policy attributes described in
566Documentation/admin-guide/pm/cpufreq.rst is special with ``intel_pstate``
567as the current scaling driver and it generally depends on the driver's
568:ref:`operation mode <operation_modes>`.
569
570First of all, the values of the ``cpuinfo_max_freq``, ``cpuinfo_min_freq`` and
571``scaling_cur_freq`` attributes are produced by applying a processor-specific
572multiplier to the internal P-state representation used by ``intel_pstate``.
573Also, the values of the ``scaling_max_freq`` and ``scaling_min_freq``
574attributes are capped by the frequency corresponding to the maximum P-state that
575the driver is allowed to set.
576
577If the ``no_turbo`` :ref:`global attribute <no_turbo_attr>` is set, the driver
578is not allowed to use turbo P-states, so the maximum value of
579``scaling_max_freq`` and ``scaling_min_freq`` is limited to the maximum
580non-turbo P-state frequency.
581Accordingly, setting ``no_turbo`` causes ``scaling_max_freq`` and
582``scaling_min_freq`` to go down to that value if they were above it before.
583However, the old values of ``scaling_max_freq`` and ``scaling_min_freq`` will be
584restored after unsetting ``no_turbo``, unless these attributes have been written
585to after ``no_turbo`` was set.
586
587If ``no_turbo`` is not set, the maximum possible value of ``scaling_max_freq``
588and ``scaling_min_freq`` corresponds to the maximum supported turbo P-state,
589which also is the value of ``cpuinfo_max_freq`` in either case.
590
591Next, the following policy attributes have special meaning if
592``intel_pstate`` works in the :ref:`active mode <active_mode>`:
593
594``scaling_available_governors``
595	List of P-state selection algorithms provided by ``intel_pstate``.
596
597``scaling_governor``
598	P-state selection algorithm provided by ``intel_pstate`` currently in
599	use with the given policy.
600
601``scaling_cur_freq``
602	Frequency of the average P-state of the CPU represented by the given
603	policy for the time interval between the last two invocations of the
604	driver's utilization update callback by the CPU scheduler for that CPU.
605
606One more policy attribute is present if the HWP feature is enabled in the
607processor:
608
609``base_frequency``
610	Shows the base frequency of the CPU. Any frequency above this will be
611	in the turbo frequency range.
612
613The meaning of these attributes in the :ref:`passive mode <passive_mode>` is the
614same as for other scaling drivers.
615
616Additionally, the value of the ``scaling_driver`` attribute for ``intel_pstate``
617depends on the operation mode of the driver.  Namely, it is either
618"intel_pstate" (in the :ref:`active mode <active_mode>`) or "intel_cpufreq"
619(in the :ref:`passive mode <passive_mode>`).
620
621.. _pstate_limits_coordination:
622
623Coordination of P-State Limits
624------------------------------
625
626``intel_pstate`` allows P-state limits to be set in two ways: with the help of
627the ``max_perf_pct`` and ``min_perf_pct`` :ref:`global attributes
628<global_attributes>` or via the ``scaling_max_freq`` and ``scaling_min_freq``
629``CPUFreq`` policy attributes.  The coordination between those limits is based
630on the following rules, regardless of the current operation mode of the driver:
631
632 1. All CPUs are affected by the global limits (that is, none of them can be
633    requested to run faster than the global maximum and none of them can be
634    requested to run slower than the global minimum).
635
636 2. Each individual CPU is affected by its own per-policy limits (that is, it
637    cannot be requested to run faster than its own per-policy maximum and it
638    cannot be requested to run slower than its own per-policy minimum). The
639    effective performance depends on whether the platform supports per core
640    P-states, hyper-threading is enabled and on current performance requests
641    from other CPUs. When platform doesn't support per core P-states, the
642    effective performance can be more than the policy limits set on a CPU, if
643    other CPUs are requesting higher performance at that moment. Even with per
644    core P-states support, when hyper-threading is enabled, if the sibling CPU
645    is requesting higher performance, the other siblings will get higher
646    performance than their policy limits.
647
648 3. The global and per-policy limits can be set independently.
649
650In the :ref:`active mode with the HWP feature enabled <active_mode_hwp>`, the
651resulting effective values are written into hardware registers whenever the
652limits change in order to request its internal P-state selection logic to always
653set P-states within these limits.  Otherwise, the limits are taken into account
654by scaling governors (in the :ref:`passive mode <passive_mode>`) and by the
655driver every time before setting a new P-state for a CPU.
656
657Additionally, if the ``intel_pstate=per_cpu_perf_limits`` command line argument
658is passed to the kernel, ``max_perf_pct`` and ``min_perf_pct`` are not exposed
659at all and the only way to set the limits is by using the policy attributes.
660
661.. _energy_performance_hints:
662
663Energy vs Performance Hints
664---------------------------
665
666If the hardware-managed P-states (HWP) is enabled in the processor, additional
667attributes, intended to allow user space to help ``intel_pstate`` to adjust the
668processor's internal P-state selection logic by focusing it on performance or on
669energy-efficiency, or somewhere between the two extremes, are present in every
670``CPUFreq`` policy directory in ``sysfs``.  They are :
671
672``energy_performance_preference``
673	Current value of the energy vs performance hint for the given policy
674	(or the CPU represented by it).
675
676	The hint can be changed by writing to this attribute.
677
678``energy_performance_available_preferences``
679	List of strings that can be written to the
680	``energy_performance_preference`` attribute.
681
682	They represent different energy vs performance hints and should be
683	self-explanatory, except that ``default`` represents whatever hint
684	value was set by the platform firmware.
685
686Strings written to the ``energy_performance_preference`` attribute are
687internally translated to integer values written to the processor's
688Energy-Performance Preference (EPP) knob (if supported) or its
689Energy-Performance Bias (EPB) knob. It is also possible to write a positive
690integer value between 0 to 255, if the EPP feature is present. If the EPP
691feature is not present, writing integer value to this attribute is not
692supported. In this case, user can use the
693"/sys/devices/system/cpu/cpu*/power/energy_perf_bias" interface.
694
695[Note that tasks may by migrated from one CPU to another by the scheduler's
696load-balancing algorithm and if different energy vs performance hints are
697set for those CPUs, that may lead to undesirable outcomes.  To avoid such
698issues it is better to set the same energy vs performance hint for all CPUs
699or to pin every task potentially sensitive to them to a specific CPU.]
700
701.. _acpi-cpufreq:
702
703``intel_pstate`` vs ``acpi-cpufreq``
704====================================
705
706On the majority of systems supported by ``intel_pstate``, the ACPI tables
707provided by the platform firmware contain ``_PSS`` objects returning information
708that can be used for CPU performance scaling (refer to the ACPI specification
709[3]_ for details on the ``_PSS`` objects and the format of the information
710returned by them).
711
712The information returned by the ACPI ``_PSS`` objects is used by the
713``acpi-cpufreq`` scaling driver.  On systems supported by ``intel_pstate``
714the ``acpi-cpufreq`` driver uses the same hardware CPU performance scaling
715interface, but the set of P-states it can use is limited by the ``_PSS``
716output.
717
718On those systems each ``_PSS`` object returns a list of P-states supported by
719the corresponding CPU which basically is a subset of the P-states range that can
720be used by ``intel_pstate`` on the same system, with one exception: the whole
721:ref:`turbo range <turbo>` is represented by one item in it (the topmost one).
722By convention, the frequency returned by ``_PSS`` for that item is greater by
7231 MHz than the frequency of the highest non-turbo P-state listed by it, but the
724corresponding P-state representation (following the hardware specification)
725returned for it matches the maximum supported turbo P-state (or is the
726special value 255 meaning essentially "go as high as you can get").
727
728The list of P-states returned by ``_PSS`` is reflected by the table of
729available frequencies supplied by ``acpi-cpufreq`` to the ``CPUFreq`` core and
730scaling governors and the minimum and maximum supported frequencies reported by
731it come from that list as well.  In particular, given the special representation
732of the turbo range described above, this means that the maximum supported
733frequency reported by ``acpi-cpufreq`` is higher by 1 MHz than the frequency
734of the highest supported non-turbo P-state listed by ``_PSS`` which, of course,
735affects decisions made by the scaling governors, except for ``powersave`` and
736``performance``.
737
738For example, if a given governor attempts to select a frequency proportional to
739estimated CPU load and maps the load of 100% to the maximum supported frequency
740(possibly multiplied by a constant), then it will tend to choose P-states below
741the turbo threshold if ``acpi-cpufreq`` is used as the scaling driver, because
742in that case the turbo range corresponds to a small fraction of the frequency
743band it can use (1 MHz vs 1 GHz or more).  In consequence, it will only go to
744the turbo range for the highest loads and the other loads above 50% that might
745benefit from running at turbo frequencies will be given non-turbo P-states
746instead.
747
748One more issue related to that may appear on systems supporting the
749:ref:`Configurable TDP feature <turbo>` allowing the platform firmware to set
750the turbo threshold.  Namely, if that is not coordinated with the lists of
751P-states returned by ``_PSS`` properly, there may be more than one item
752corresponding to a turbo P-state in those lists and there may be a problem with
753avoiding the turbo range (if desirable or necessary).  Usually, to avoid using
754turbo P-states overall, ``acpi-cpufreq`` simply avoids using the topmost state
755listed by ``_PSS``, but that is not sufficient when there are other turbo
756P-states in the list returned by it.
757
758Apart from the above, ``acpi-cpufreq`` works like ``intel_pstate`` in the
759:ref:`passive mode <passive_mode>`, except that the number of P-states it can
760set is limited to the ones listed by the ACPI ``_PSS`` objects.
761
762
763Kernel Command Line Options for ``intel_pstate``
764================================================
765
766Several kernel command line options can be used to pass early-configuration-time
767parameters to ``intel_pstate`` in order to enforce specific behavior of it.  All
768of them have to be prepended with the ``intel_pstate=`` prefix.
769
770``disable``
771	Do not register ``intel_pstate`` as the scaling driver even if the
772	processor is supported by it.
773
774``active``
775	Register ``intel_pstate`` in the :ref:`active mode <active_mode>` to
776        start with.
777
778``passive``
779	Register ``intel_pstate`` in the :ref:`passive mode <passive_mode>` to
780	start with.
781
782``force``
783	Register ``intel_pstate`` as the scaling driver instead of
784	``acpi-cpufreq`` even if the latter is preferred on the given system.
785
786	This may prevent some platform features (such as thermal controls and
787	power capping) that rely on the availability of ACPI P-states
788	information from functioning as expected, so it should be used with
789	caution.
790
791	This option does not work with processors that are not supported by
792	``intel_pstate`` and on platforms where the ``pcc-cpufreq`` scaling
793	driver is used instead of ``acpi-cpufreq``.
794
795``no_hwp``
796	Do not enable the hardware-managed P-states (HWP) feature even if it is
797	supported by the processor.
798
799``hwp_only``
800	Register ``intel_pstate`` as the scaling driver only if the
801	hardware-managed P-states (HWP) feature is supported by the processor.
802
803``support_acpi_ppc``
804	Take ACPI ``_PPC`` performance limits into account.
805
806	If the preferred power management profile in the FADT (Fixed ACPI
807	Description Table) is set to "Enterprise Server" or "Performance
808	Server", the ACPI ``_PPC`` limits are taken into account by default
809	and this option has no effect.
810
811``per_cpu_perf_limits``
812	Use per-logical-CPU P-State limits (see
813        :ref:`pstate_limits_coordination` for details).
814
815``no_cas``
816	Do not enable :ref:`capacity-aware scheduling <CAS>` which is enabled
817        by default on hybrid systems without SMT.
818
819Diagnostics and Tuning
820======================
821
822Trace Events
823------------
824
825There are two static trace events that can be used for ``intel_pstate``
826diagnostics.  One of them is the ``cpu_frequency`` trace event generally used
827by ``CPUFreq``, and the other one is the ``pstate_sample`` trace event specific
828to ``intel_pstate``.  Both of them are triggered by ``intel_pstate`` only if
829it works in the :ref:`active mode <active_mode>`.
830
831The following sequence of shell commands can be used to enable them and see
832their output (if the kernel is generally configured to support event tracing)::
833
834 # cd /sys/kernel/tracing/
835 # echo 1 > events/power/pstate_sample/enable
836 # echo 1 > events/power/cpu_frequency/enable
837 # cat trace
838 gnome-terminal--4510  [001] ..s.  1177.680733: pstate_sample: core_busy=107 scaled=94 from=26 to=26 mperf=1143818 aperf=1230607 tsc=29838618 freq=2474476
839 cat-5235  [002] ..s.  1177.681723: cpu_frequency: state=2900000 cpu_id=2
840
841If ``intel_pstate`` works in the :ref:`passive mode <passive_mode>`, the
842``cpu_frequency`` trace event will be triggered either by the ``schedutil``
843scaling governor (for the policies it is attached to), or by the ``CPUFreq``
844core (for the policies with other scaling governors).
845
846``ftrace``
847----------
848
849The ``ftrace`` interface can be used for low-level diagnostics of
850``intel_pstate``.  For example, to check how often the function to set a
851P-state is called, the ``ftrace`` filter can be set to
852:c:func:`intel_pstate_set_pstate`::
853
854 # cd /sys/kernel/tracing/
855 # cat available_filter_functions | grep -i pstate
856 intel_pstate_set_pstate
857 intel_pstate_cpu_init
858 ...
859 # echo intel_pstate_set_pstate > set_ftrace_filter
860 # echo function > current_tracer
861 # cat trace | head -15
862 # tracer: function
863 #
864 # entries-in-buffer/entries-written: 80/80   #P:4
865 #
866 #                              _-----=> irqs-off
867 #                             / _----=> need-resched
868 #                            | / _---=> hardirq/softirq
869 #                            || / _--=> preempt-depth
870 #                            ||| /     delay
871 #           TASK-PID   CPU#  ||||    TIMESTAMP  FUNCTION
872 #              | |       |   ||||       |         |
873             Xorg-3129  [000] ..s.  2537.644844: intel_pstate_set_pstate <-intel_pstate_timer_func
874  gnome-terminal--4510  [002] ..s.  2537.649844: intel_pstate_set_pstate <-intel_pstate_timer_func
875      gnome-shell-3409  [001] ..s.  2537.650850: intel_pstate_set_pstate <-intel_pstate_timer_func
876           <idle>-0     [000] ..s.  2537.654843: intel_pstate_set_pstate <-intel_pstate_timer_func
877
878
879References
880==========
881
882.. [1] Kristen Accardi, *Balancing Power and Performance in the Linux Kernel*,
883       https://events.static.linuxfound.org/sites/events/files/slides/LinuxConEurope_2015.pdf
884
885.. [2] *Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 3: System Programming Guide*,
886       https://www.intel.com/content/www/us/en/architecture-and-technology/64-ia-32-architectures-software-developer-system-programming-manual-325384.html
887
888.. [3] *Advanced Configuration and Power Interface Specification*,
889       https://uefi.org/sites/default/files/resources/ACPI_6_3_final_Jan30.pdf
890