1 .. SPDX-License-Identifier: GPL-2.0 1 .. SPDX-License-Identifier: GPL-2.0
2 .. include:: <isonum.txt> 2 .. include:: <isonum.txt>
3 3
4 ============================================== 4 ===============================================
5 ``intel_pstate`` CPU Performance Scaling Drive 5 ``intel_pstate`` CPU Performance Scaling Driver
6 ============================================== 6 ===============================================
7 7
8 :Copyright: |copy| 2017 Intel Corporation 8 :Copyright: |copy| 2017 Intel Corporation
9 9
10 :Author: Rafael J. Wysocki <rafael.j.wysocki@in 10 :Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
11 11
12 12
13 General Information 13 General Information
14 =================== 14 ===================
15 15
16 ``intel_pstate`` is a part of the 16 ``intel_pstate`` is a part of the
17 :doc:`CPU performance scaling subsystem <cpufr 17 :doc:`CPU performance scaling subsystem <cpufreq>` in the Linux kernel
18 (``CPUFreq``). It is a scaling driver for the 18 (``CPUFreq``). It is a scaling driver for the Sandy Bridge and later
19 generations of Intel processors. Note, howeve 19 generations of Intel processors. Note, however, that some of those processors
20 may not be supported. [To understand ``intel_ 20 may not be supported. [To understand ``intel_pstate`` it is necessary to know
21 how ``CPUFreq`` works in general, so this is t !! 21 how ``CPUFreq`` works in general, so this is the time to read :doc:`cpufreq` if
22 Documentation/admin-guide/pm/cpufreq.rst if yo !! 22 you have not done that yet.]
23 23
24 For the processors supported by ``intel_pstate 24 For the processors supported by ``intel_pstate``, the P-state concept is broader
25 than just an operating frequency or an operati 25 than just an operating frequency or an operating performance point (see the
26 LinuxCon Europe 2015 presentation by Kristen A 26 LinuxCon Europe 2015 presentation by Kristen Accardi [1]_ for more
27 information about that). For this reason, the 27 information about that). For this reason, the representation of P-states used
28 by ``intel_pstate`` internally follows the har 28 by ``intel_pstate`` internally follows the hardware specification (for details
29 refer to Intel Software Developer’s Manual [ 29 refer to Intel Software Developer’s Manual [2]_). However, the ``CPUFreq`` core
30 uses frequencies for identifying operating per 30 uses frequencies for identifying operating performance points of CPUs and
31 frequencies are involved in the user space int 31 frequencies are involved in the user space interface exposed by it, so
32 ``intel_pstate`` maps its internal representat 32 ``intel_pstate`` maps its internal representation of P-states to frequencies too
33 (fortunately, that mapping is unambiguous). A 33 (fortunately, that mapping is unambiguous). At the same time, it would not be
34 practical for ``intel_pstate`` to supply the ` 34 practical for ``intel_pstate`` to supply the ``CPUFreq`` core with a table of
35 available frequencies due to the possible size 35 available frequencies due to the possible size of it, so the driver does not do
36 that. Some functionality of the core is limit 36 that. Some functionality of the core is limited by that.
37 37
38 Since the hardware P-state selection interface 38 Since the hardware P-state selection interface used by ``intel_pstate`` is
39 available at the logical CPU level, the driver 39 available at the logical CPU level, the driver always works with individual
40 CPUs. Consequently, if ``intel_pstate`` is in 40 CPUs. Consequently, if ``intel_pstate`` is in use, every ``CPUFreq`` policy
41 object corresponds to one logical CPU and ``CP 41 object corresponds to one logical CPU and ``CPUFreq`` policies are effectively
42 equivalent to CPUs. In particular, this means 42 equivalent to CPUs. In particular, this means that they become "inactive" every
43 time the corresponding CPU is taken offline an 43 time the corresponding CPU is taken offline and need to be re-initialized when
44 it goes back online. 44 it goes back online.
45 45
46 ``intel_pstate`` is not modular, so it cannot 46 ``intel_pstate`` is not modular, so it cannot be unloaded, which means that the
47 only way to pass early-configuration-time para 47 only way to pass early-configuration-time parameters to it is via the kernel
48 command line. However, its configuration can 48 command line. However, its configuration can be adjusted via ``sysfs`` to a
49 great extent. In some configurations it even 49 great extent. In some configurations it even is possible to unregister it via
50 ``sysfs`` which allows another ``CPUFreq`` sca 50 ``sysfs`` which allows another ``CPUFreq`` scaling driver to be loaded and
51 registered (see `below <status_attr_>`_). 51 registered (see `below <status_attr_>`_).
52 52
53 53
54 Operation Modes 54 Operation Modes
55 =============== 55 ===============
56 56
57 ``intel_pstate`` can operate in two different !! 57 ``intel_pstate`` can operate in three different modes: in the active mode with
58 active mode, it uses its own internal performa !! 58 or without hardware-managed P-states support and in the passive mode. Which of
59 allows the hardware to do performance scaling !! 59 them will be in effect depends on what kernel command line options are used and
60 mode it responds to requests made by a generic !! 60 on the capabilities of the processor.
61 a certain performance scaling algorithm. Whic <<
62 depends on what kernel command line options ar <<
63 the processor. <<
64 61
65 Active Mode 62 Active Mode
66 ----------- 63 -----------
67 64
68 This is the default operation mode of ``intel_ !! 65 This is the default operation mode of ``intel_pstate``. If it works in this
69 hardware-managed P-states (HWP) support. If i !! 66 mode, the ``scaling_driver`` policy attribute in ``sysfs`` for all ``CPUFreq``
70 ``scaling_driver`` policy attribute in ``sysfs !! 67 policies contains the string "intel_pstate".
71 contains the string "intel_pstate". <<
72 68
73 In this mode the driver bypasses the scaling g 69 In this mode the driver bypasses the scaling governors layer of ``CPUFreq`` and
74 provides its own scaling algorithms for P-stat 70 provides its own scaling algorithms for P-state selection. Those algorithms
75 can be applied to ``CPUFreq`` policies in the 71 can be applied to ``CPUFreq`` policies in the same way as generic scaling
76 governors (that is, through the ``scaling_gove 72 governors (that is, through the ``scaling_governor`` policy attribute in
77 ``sysfs``). [Note that different P-state sele 73 ``sysfs``). [Note that different P-state selection algorithms may be chosen for
78 different policies, but that is not recommende 74 different policies, but that is not recommended.]
79 75
80 They are not generic scaling governors, but th 76 They are not generic scaling governors, but their names are the same as the
81 names of some of those governors. Moreover, c 77 names of some of those governors. Moreover, confusingly enough, they generally
82 do not work in the same way as the generic gov 78 do not work in the same way as the generic governors they share the names with.
83 For example, the ``powersave`` P-state selecti 79 For example, the ``powersave`` P-state selection algorithm provided by
84 ``intel_pstate`` is not a counterpart of the g 80 ``intel_pstate`` is not a counterpart of the generic ``powersave`` governor
85 (roughly, it corresponds to the ``schedutil`` 81 (roughly, it corresponds to the ``schedutil`` and ``ondemand`` governors).
86 82
87 There are two P-state selection algorithms pro 83 There are two P-state selection algorithms provided by ``intel_pstate`` in the
88 active mode: ``powersave`` and ``performance`` 84 active mode: ``powersave`` and ``performance``. The way they both operate
89 depends on whether or not the hardware-managed 85 depends on whether or not the hardware-managed P-states (HWP) feature has been
90 enabled in the processor and possibly on the p 86 enabled in the processor and possibly on the processor model.
91 87
92 Which of the P-state selection algorithms is u 88 Which of the P-state selection algorithms is used by default depends on the
93 :c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMA 89 :c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE` kernel configuration option.
94 Namely, if that option is set, the ``performan 90 Namely, if that option is set, the ``performance`` algorithm will be used by
95 default, and the other one will be used by def 91 default, and the other one will be used by default if it is not set.
96 92
97 Active Mode With HWP 93 Active Mode With HWP
98 ~~~~~~~~~~~~~~~~~~~~ 94 ~~~~~~~~~~~~~~~~~~~~
99 95
100 If the processor supports the HWP feature, it 96 If the processor supports the HWP feature, it will be enabled during the
101 processor initialization and cannot be disable 97 processor initialization and cannot be disabled after that. It is possible
102 to avoid enabling it by passing the ``intel_ps 98 to avoid enabling it by passing the ``intel_pstate=no_hwp`` argument to the
103 kernel in the command line. 99 kernel in the command line.
104 100
105 If the HWP feature has been enabled, ``intel_p 101 If the HWP feature has been enabled, ``intel_pstate`` relies on the processor to
106 select P-states by itself, but still it can gi 102 select P-states by itself, but still it can give hints to the processor's
107 internal P-state selection logic. What those 103 internal P-state selection logic. What those hints are depends on which P-state
108 selection algorithm has been applied to the gi 104 selection algorithm has been applied to the given policy (or to the CPU it
109 corresponds to). 105 corresponds to).
110 106
111 Even though the P-state selection is carried o 107 Even though the P-state selection is carried out by the processor automatically,
112 ``intel_pstate`` registers utilization update 108 ``intel_pstate`` registers utilization update callbacks with the CPU scheduler
113 in this mode. However, they are not used for 109 in this mode. However, they are not used for running a P-state selection
114 algorithm, but for periodic updates of the cur 110 algorithm, but for periodic updates of the current CPU frequency information to
115 be made available from the ``scaling_cur_freq` 111 be made available from the ``scaling_cur_freq`` policy attribute in ``sysfs``.
116 112
117 HWP + ``performance`` 113 HWP + ``performance``
118 ..................... 114 .....................
119 115
120 In this configuration ``intel_pstate`` will wr 116 In this configuration ``intel_pstate`` will write 0 to the processor's
121 Energy-Performance Preference (EPP) knob (if s 117 Energy-Performance Preference (EPP) knob (if supported) or its
122 Energy-Performance Bias (EPB) knob (otherwise) 118 Energy-Performance Bias (EPB) knob (otherwise), which means that the processor's
123 internal P-state selection logic is expected t 119 internal P-state selection logic is expected to focus entirely on performance.
124 120
125 This will override the EPP/EPB setting coming 121 This will override the EPP/EPB setting coming from the ``sysfs`` interface
126 (see `Energy vs Performance Hints`_ below). M !! 122 (see `Energy vs Performance Hints`_ below).
127 the EPP/EPB to a value different from 0 ("perf <<
128 configuration will be rejected. <<
129 123
130 Also, in this configuration the range of P-sta 124 Also, in this configuration the range of P-states available to the processor's
131 internal P-state selection logic is always res 125 internal P-state selection logic is always restricted to the upper boundary
132 (that is, the maximum P-state that the driver 126 (that is, the maximum P-state that the driver is allowed to use).
133 127
134 HWP + ``powersave`` 128 HWP + ``powersave``
135 ................... 129 ...................
136 130
137 In this configuration ``intel_pstate`` will se 131 In this configuration ``intel_pstate`` will set the processor's
138 Energy-Performance Preference (EPP) knob (if s 132 Energy-Performance Preference (EPP) knob (if supported) or its
139 Energy-Performance Bias (EPB) knob (otherwise) 133 Energy-Performance Bias (EPB) knob (otherwise) to whatever value it was
140 previously set to via ``sysfs`` (or whatever d 134 previously set to via ``sysfs`` (or whatever default value it was
141 set to by the platform firmware). This usuall 135 set to by the platform firmware). This usually causes the processor's
142 internal P-state selection logic to be less pe 136 internal P-state selection logic to be less performance-focused.
143 137
144 Active Mode Without HWP 138 Active Mode Without HWP
145 ~~~~~~~~~~~~~~~~~~~~~~~ 139 ~~~~~~~~~~~~~~~~~~~~~~~
146 140
147 This operation mode is optional for processors !! 141 This is the default operation mode for processors that do not support the HWP
148 feature or when the ``intel_pstate=no_hwp`` ar !! 142 feature. It also is used by default with the ``intel_pstate=no_hwp`` argument
149 the command line. The active mode is used in !! 143 in the kernel command line. However, in this mode ``intel_pstate`` may refuse
150 ``intel_pstate=active`` argument is passed to !! 144 to work with the given processor if it does not recognize it. [Note that
151 In this mode ``intel_pstate`` may refuse to wo !! 145 ``intel_pstate`` will never refuse to work with any processor with the HWP
152 recognized by it. [Note that ``intel_pstate`` !! 146 feature enabled.]
153 any processor with the HWP feature enabled.] <<
154 147
155 In this mode ``intel_pstate`` registers utiliz 148 In this mode ``intel_pstate`` registers utilization update callbacks with the
156 CPU scheduler in order to run a P-state select 149 CPU scheduler in order to run a P-state selection algorithm, either
157 ``powersave`` or ``performance``, depending on 150 ``powersave`` or ``performance``, depending on the ``scaling_governor`` policy
158 setting in ``sysfs``. The current CPU frequen 151 setting in ``sysfs``. The current CPU frequency information to be made
159 available from the ``scaling_cur_freq`` policy 152 available from the ``scaling_cur_freq`` policy attribute in ``sysfs`` is
160 periodically updated by those utilization upda 153 periodically updated by those utilization update callbacks too.
161 154
162 ``performance`` 155 ``performance``
163 ............... 156 ...............
164 157
165 Without HWP, this P-state selection algorithm 158 Without HWP, this P-state selection algorithm is always the same regardless of
166 the processor model and platform configuration 159 the processor model and platform configuration.
167 160
168 It selects the maximum P-state it is allowed t 161 It selects the maximum P-state it is allowed to use, subject to limits set via
169 ``sysfs``, every time the driver configuration 162 ``sysfs``, every time the driver configuration for the given CPU is updated
170 (e.g. via ``sysfs``). 163 (e.g. via ``sysfs``).
171 164
172 This is the default P-state selection algorith 165 This is the default P-state selection algorithm if the
173 :c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMA 166 :c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE` kernel configuration option
174 is set. 167 is set.
175 168
176 ``powersave`` 169 ``powersave``
177 ............. 170 .............
178 171
179 Without HWP, this P-state selection algorithm 172 Without HWP, this P-state selection algorithm is similar to the algorithm
180 implemented by the generic ``schedutil`` scali 173 implemented by the generic ``schedutil`` scaling governor except that the
181 utilization metric used by it is based on numb 174 utilization metric used by it is based on numbers coming from feedback
182 registers of the CPU. It generally selects P- 175 registers of the CPU. It generally selects P-states proportional to the
183 current CPU utilization. 176 current CPU utilization.
184 177
185 This algorithm is run by the driver's utilizat 178 This algorithm is run by the driver's utilization update callback for the
186 given CPU when it is invoked by the CPU schedu 179 given CPU when it is invoked by the CPU scheduler, but not more often than
187 every 10 ms. Like in the ``performance`` case 180 every 10 ms. Like in the ``performance`` case, the hardware configuration
188 is not touched if the new P-state turns out to 181 is not touched if the new P-state turns out to be the same as the current
189 one. 182 one.
190 183
191 This is the default P-state selection algorith 184 This is the default P-state selection algorithm if the
192 :c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMA 185 :c:macro:`CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE` kernel configuration option
193 is not set. 186 is not set.
194 187
195 Passive Mode 188 Passive Mode
196 ------------ 189 ------------
197 190
198 This is the default operation mode of ``intel_ !! 191 This mode is used if the ``intel_pstate=passive`` argument is passed to the
199 hardware-managed P-states (HWP) support. It i !! 192 kernel in the command line (it implies the ``intel_pstate=no_hwp`` setting too).
200 ``intel_pstate=passive`` argument is passed to !! 193 Like in the active mode without HWP support, in this mode ``intel_pstate`` may
201 regardless of whether or not the given process !! 194 refuse to work with the given processor if it does not recognize it.
202 ``intel_pstate=no_hwp`` setting causes the dri <<
203 if it is not combined with ``intel_pstate=acti <<
204 without HWP support, in this mode ``intel_psta <<
205 processors that are not recognized by it if HW <<
206 through the kernel command line. <<
207 195
208 If the driver works in this mode, the ``scalin 196 If the driver works in this mode, the ``scaling_driver`` policy attribute in
209 ``sysfs`` for all ``CPUFreq`` policies contain 197 ``sysfs`` for all ``CPUFreq`` policies contains the string "intel_cpufreq".
210 Then, the driver behaves like a regular ``CPUF 198 Then, the driver behaves like a regular ``CPUFreq`` scaling driver. That is,
211 it is invoked by generic scaling governors whe 199 it is invoked by generic scaling governors when necessary to talk to the
212 hardware in order to change the P-state of a C 200 hardware in order to change the P-state of a CPU (in particular, the
213 ``schedutil`` governor can invoke it directly 201 ``schedutil`` governor can invoke it directly from scheduler context).
214 202
215 While in this mode, ``intel_pstate`` can be us 203 While in this mode, ``intel_pstate`` can be used with all of the (generic)
216 scaling governors listed by the ``scaling_avai 204 scaling governors listed by the ``scaling_available_governors`` policy attribute
217 in ``sysfs`` (and the P-state selection algori 205 in ``sysfs`` (and the P-state selection algorithms described above are not
218 used). Then, it is responsible for the config 206 used). Then, it is responsible for the configuration of policy objects
219 corresponding to CPUs and provides the ``CPUFr 207 corresponding to CPUs and provides the ``CPUFreq`` core (and the scaling
220 governors attached to the policy objects) with 208 governors attached to the policy objects) with accurate information on the
221 maximum and minimum operating frequencies supp 209 maximum and minimum operating frequencies supported by the hardware (including
222 the so-called "turbo" frequency ranges). In o 210 the so-called "turbo" frequency ranges). In other words, in the passive mode
223 the entire range of available P-states is expo 211 the entire range of available P-states is exposed by ``intel_pstate`` to the
224 ``CPUFreq`` core. However, in this mode the d 212 ``CPUFreq`` core. However, in this mode the driver does not register
225 utilization update callbacks with the CPU sche 213 utilization update callbacks with the CPU scheduler and the ``scaling_cur_freq``
226 information comes from the ``CPUFreq`` core (a 214 information comes from the ``CPUFreq`` core (and is the last frequency selected
227 by the current scaling governor for the given 215 by the current scaling governor for the given policy).
228 216
229 217
230 .. _turbo: 218 .. _turbo:
231 219
232 Turbo P-states Support 220 Turbo P-states Support
233 ====================== 221 ======================
234 222
235 In the majority of cases, the entire range of 223 In the majority of cases, the entire range of P-states available to
236 ``intel_pstate`` can be divided into two sub-r 224 ``intel_pstate`` can be divided into two sub-ranges that correspond to
237 different types of processor behavior, above a 225 different types of processor behavior, above and below a boundary that
238 will be referred to as the "turbo threshold" i 226 will be referred to as the "turbo threshold" in what follows.
239 227
240 The P-states above the turbo threshold are ref 228 The P-states above the turbo threshold are referred to as "turbo P-states" and
241 the whole sub-range of P-states they belong to 229 the whole sub-range of P-states they belong to is referred to as the "turbo
242 range". These names are related to the Turbo 230 range". These names are related to the Turbo Boost technology allowing a
243 multicore processor to opportunistically incre 231 multicore processor to opportunistically increase the P-state of one or more
244 cores if there is enough power to do that and 232 cores if there is enough power to do that and if that is not going to cause the
245 thermal envelope of the processor package to b 233 thermal envelope of the processor package to be exceeded.
246 234
247 Specifically, if software sets the P-state of 235 Specifically, if software sets the P-state of a CPU core within the turbo range
248 (that is, above the turbo threshold), the proc 236 (that is, above the turbo threshold), the processor is permitted to take over
249 performance scaling control for that core and 237 performance scaling control for that core and put it into turbo P-states of its
250 choice going forward. However, that permissio 238 choice going forward. However, that permission is interpreted differently by
251 different processor generations. Namely, the 239 different processor generations. Namely, the Sandy Bridge generation of
252 processors will never use any P-states above t 240 processors will never use any P-states above the last one set by software for
253 the given core, even if it is within the turbo 241 the given core, even if it is within the turbo range, whereas all of the later
254 processor generations will take it as a licens 242 processor generations will take it as a license to use any P-states from the
255 turbo range, even above the one set by softwar 243 turbo range, even above the one set by software. In other words, on those
256 processors setting any P-state from the turbo 244 processors setting any P-state from the turbo range will enable the processor
257 to put the given core into all turbo P-states 245 to put the given core into all turbo P-states up to and including the maximum
258 supported one as it sees fit. 246 supported one as it sees fit.
259 247
260 One important property of turbo P-states is th 248 One important property of turbo P-states is that they are not sustainable. More
261 precisely, there is no guarantee that any CPUs 249 precisely, there is no guarantee that any CPUs will be able to stay in any of
262 those states indefinitely, because the power d 250 those states indefinitely, because the power distribution within the processor
263 package may change over time or the thermal e 251 package may change over time or the thermal envelope it was designed for might
264 be exceeded if a turbo P-state was used for to 252 be exceeded if a turbo P-state was used for too long.
265 253
266 In turn, the P-states below the turbo threshol 254 In turn, the P-states below the turbo threshold generally are sustainable. In
267 fact, if one of them is set by software, the p 255 fact, if one of them is set by software, the processor is not expected to change
268 it to a lower one unless in a thermal stress o 256 it to a lower one unless in a thermal stress or a power limit violation
269 situation (a higher P-state may still be used 257 situation (a higher P-state may still be used if it is set for another CPU in
270 the same package at the same time, for example 258 the same package at the same time, for example).
271 259
272 Some processors allow multiple cores to be in 260 Some processors allow multiple cores to be in turbo P-states at the same time,
273 but the maximum P-state that can be set for th 261 but the maximum P-state that can be set for them generally depends on the number
274 of cores running concurrently. The maximum tu 262 of cores running concurrently. The maximum turbo P-state that can be set for 3
275 cores at the same time usually is lower than t 263 cores at the same time usually is lower than the analogous maximum P-state for
276 2 cores, which in turn usually is lower than t 264 2 cores, which in turn usually is lower than the maximum turbo P-state that can
277 be set for 1 core. The one-core maximum turbo 265 be set for 1 core. The one-core maximum turbo P-state is thus the maximum
278 supported one overall. 266 supported one overall.
279 267
280 The maximum supported turbo P-state, the turbo 268 The maximum supported turbo P-state, the turbo threshold (the maximum supported
281 non-turbo P-state) and the minimum supported P 269 non-turbo P-state) and the minimum supported P-state are specific to the
282 processor model and can be determined by readi 270 processor model and can be determined by reading the processor's model-specific
283 registers (MSRs). Moreover, some processors s 271 registers (MSRs). Moreover, some processors support the Configurable TDP
284 (Thermal Design Power) feature and, when that 272 (Thermal Design Power) feature and, when that feature is enabled, the turbo
285 threshold effectively becomes a configurable v 273 threshold effectively becomes a configurable value that can be set by the
286 platform firmware. 274 platform firmware.
287 275
288 Unlike ``_PSS`` objects in the ACPI tables, `` 276 Unlike ``_PSS`` objects in the ACPI tables, ``intel_pstate`` always exposes
289 the entire range of available P-states, includ 277 the entire range of available P-states, including the whole turbo range, to the
290 ``CPUFreq`` core and (in the passive mode) to 278 ``CPUFreq`` core and (in the passive mode) to generic scaling governors. This
291 generally causes turbo P-states to be set more 279 generally causes turbo P-states to be set more often when ``intel_pstate`` is
292 used relative to ACPI-based CPU performance sc 280 used relative to ACPI-based CPU performance scaling (see `below <acpi-cpufreq_>`_
293 for more information). 281 for more information).
294 282
295 Moreover, since ``intel_pstate`` always knows 283 Moreover, since ``intel_pstate`` always knows what the real turbo threshold is
296 (even if the Configurable TDP feature is enabl 284 (even if the Configurable TDP feature is enabled in the processor), its
297 ``no_turbo`` attribute in ``sysfs`` (described 285 ``no_turbo`` attribute in ``sysfs`` (described `below <no_turbo_attr_>`_) should
298 work as expected in all cases (that is, if set 286 work as expected in all cases (that is, if set to disable turbo P-states, it
299 always should prevent ``intel_pstate`` from us 287 always should prevent ``intel_pstate`` from using them).
300 288
301 289
302 Processor Support 290 Processor Support
303 ================= 291 =================
304 292
305 To handle a given processor ``intel_pstate`` r 293 To handle a given processor ``intel_pstate`` requires a number of different
306 pieces of information on it to be known, inclu 294 pieces of information on it to be known, including:
307 295
308 * The minimum supported P-state. 296 * The minimum supported P-state.
309 297
310 * The maximum supported `non-turbo P-state <t 298 * The maximum supported `non-turbo P-state <turbo_>`_.
311 299
312 * Whether or not turbo P-states are supported 300 * Whether or not turbo P-states are supported at all.
313 301
314 * The maximum supported `one-core turbo P-sta 302 * The maximum supported `one-core turbo P-state <turbo_>`_ (if turbo P-states
315 are supported). 303 are supported).
316 304
317 * The scaling formula to translate the driver 305 * The scaling formula to translate the driver's internal representation
318 of P-states into frequencies and the other 306 of P-states into frequencies and the other way around.
319 307
320 Generally, ways to obtain that information are 308 Generally, ways to obtain that information are specific to the processor model
321 or family. Although it often is possible to o 309 or family. Although it often is possible to obtain all of it from the processor
322 itself (using model-specific registers), there 310 itself (using model-specific registers), there are cases in which hardware
323 manuals need to be consulted to get to it too. 311 manuals need to be consulted to get to it too.
324 312
325 For this reason, there is a list of supported 313 For this reason, there is a list of supported processors in ``intel_pstate`` and
326 the driver initialization will fail if the det 314 the driver initialization will fail if the detected processor is not in that
327 list, unless it supports the HWP feature. [Th !! 315 list, unless it supports the `HWP feature <Active Mode_>`_. [The interface to
328 information listed above is the same for all o !! 316 obtain all of the information listed above is the same for all of the processors
329 HWP feature, which is why ``intel_pstate`` wor !! 317 supporting the HWP feature, which is why they all are supported by
>> 318 ``intel_pstate``.]
330 319
331 320
332 User Space Interface in ``sysfs`` 321 User Space Interface in ``sysfs``
333 ================================= 322 =================================
334 323
335 Global Attributes 324 Global Attributes
336 ----------------- 325 -----------------
337 326
338 ``intel_pstate`` exposes several global attrib 327 ``intel_pstate`` exposes several global attributes (files) in ``sysfs`` to
339 control its functionality at the system level. 328 control its functionality at the system level. They are located in the
340 ``/sys/devices/system/cpu/intel_pstate/`` dire 329 ``/sys/devices/system/cpu/intel_pstate/`` directory and affect all CPUs.
341 330
342 Some of them are not present if the ``intel_ps 331 Some of them are not present if the ``intel_pstate=per_cpu_perf_limits``
343 argument is passed to the kernel in the comman 332 argument is passed to the kernel in the command line.
344 333
345 ``max_perf_pct`` 334 ``max_perf_pct``
346 Maximum P-state the driver is allowed 335 Maximum P-state the driver is allowed to set in percent of the
347 maximum supported performance level (t 336 maximum supported performance level (the highest supported `turbo
348 P-state <turbo_>`_). 337 P-state <turbo_>`_).
349 338
350 This attribute will not be exposed if 339 This attribute will not be exposed if the
351 ``intel_pstate=per_cpu_perf_limits`` a 340 ``intel_pstate=per_cpu_perf_limits`` argument is present in the kernel
352 command line. 341 command line.
353 342
354 ``min_perf_pct`` 343 ``min_perf_pct``
355 Minimum P-state the driver is allowed 344 Minimum P-state the driver is allowed to set in percent of the
356 maximum supported performance level (t 345 maximum supported performance level (the highest supported `turbo
357 P-state <turbo_>`_). 346 P-state <turbo_>`_).
358 347
359 This attribute will not be exposed if 348 This attribute will not be exposed if the
360 ``intel_pstate=per_cpu_perf_limits`` a 349 ``intel_pstate=per_cpu_perf_limits`` argument is present in the kernel
361 command line. 350 command line.
362 351
363 ``num_pstates`` 352 ``num_pstates``
364 Number of P-states supported by the pr 353 Number of P-states supported by the processor (between 0 and 255
365 inclusive) including both turbo and no 354 inclusive) including both turbo and non-turbo P-states (see
366 `Turbo P-states Support`_). 355 `Turbo P-states Support`_).
367 356
368 This attribute is present only if the <<
369 for all of the CPUs in the system. <<
370 <<
371 The value of this attribute is not aff 357 The value of this attribute is not affected by the ``no_turbo``
372 setting described `below <no_turbo_att 358 setting described `below <no_turbo_attr_>`_.
373 359
374 This attribute is read-only. 360 This attribute is read-only.
375 361
376 ``turbo_pct`` 362 ``turbo_pct``
377 Ratio of the `turbo range <turbo_>`_ s 363 Ratio of the `turbo range <turbo_>`_ size to the size of the entire
378 range of supported P-states, in percen 364 range of supported P-states, in percent.
379 365
380 This attribute is present only if the <<
381 for all of the CPUs in the system. <<
382 <<
383 This attribute is read-only. 366 This attribute is read-only.
384 367
385 .. _no_turbo_attr: 368 .. _no_turbo_attr:
386 369
387 ``no_turbo`` 370 ``no_turbo``
388 If set (equal to 1), the driver is not 371 If set (equal to 1), the driver is not allowed to set any turbo P-states
389 (see `Turbo P-states Support`_). If u !! 372 (see `Turbo P-states Support`_). If unset (equalt to 0, which is the
390 default), turbo P-states can be set by 373 default), turbo P-states can be set by the driver.
391 [Note that ``intel_pstate`` does not s 374 [Note that ``intel_pstate`` does not support the general ``boost``
392 attribute (supported by some other sca 375 attribute (supported by some other scaling drivers) which is replaced
393 by this one.] 376 by this one.]
394 377
395 This attribute does not affect the max !! 378 This attrubute does not affect the maximum supported frequency value
396 supplied to the ``CPUFreq`` core and e 379 supplied to the ``CPUFreq`` core and exposed via the policy interface,
397 but it affects the maximum possible va 380 but it affects the maximum possible value of per-policy P-state limits
398 (see `Interpretation of Policy Attribu 381 (see `Interpretation of Policy Attributes`_ below for details).
399 382
400 ``hwp_dynamic_boost`` 383 ``hwp_dynamic_boost``
401 This attribute is only present if ``in 384 This attribute is only present if ``intel_pstate`` works in the
402 `active mode with the HWP feature enab 385 `active mode with the HWP feature enabled <Active Mode With HWP_>`_ in
403 the processor. If set (equal to 1), i 386 the processor. If set (equal to 1), it causes the minimum P-state limit
404 to be increased dynamically for a shor 387 to be increased dynamically for a short time whenever a task previously
405 waiting on I/O is selected to run on a 388 waiting on I/O is selected to run on a given logical CPU (the purpose
406 of this mechanism is to improve perfor 389 of this mechanism is to improve performance).
407 390
408 This setting has no effect on logical 391 This setting has no effect on logical CPUs whose minimum P-state limit
409 is directly set to the highest non-tur 392 is directly set to the highest non-turbo P-state or above it.
410 393
411 .. _status_attr: 394 .. _status_attr:
412 395
413 ``status`` 396 ``status``
414 Operation mode of the driver: "active" 397 Operation mode of the driver: "active", "passive" or "off".
415 398
416 "active" 399 "active"
417 The driver is functional and i 400 The driver is functional and in the `active mode
418 <Active Mode_>`_. 401 <Active Mode_>`_.
419 402
420 "passive" 403 "passive"
421 The driver is functional and i 404 The driver is functional and in the `passive mode
422 <Passive Mode_>`_. 405 <Passive Mode_>`_.
423 406
424 "off" 407 "off"
425 The driver is not functional ( 408 The driver is not functional (it is not registered as a scaling
426 driver with the ``CPUFreq`` co 409 driver with the ``CPUFreq`` core).
427 410
428 This attribute can be written to in or 411 This attribute can be written to in order to change the driver's
429 operation mode or to unregister it. T 412 operation mode or to unregister it. The string written to it must be
430 one of the possible values of it and, 413 one of the possible values of it and, if successful, the write will
431 cause the driver to switch over to the 414 cause the driver to switch over to the operation mode represented by
432 that string - or to be unregistered in 415 that string - or to be unregistered in the "off" case. [Actually,
433 switching over from the active mode to 416 switching over from the active mode to the passive mode or the other
434 way around causes the driver to be unr 417 way around causes the driver to be unregistered and registered again
435 with a different set of callbacks, so 418 with a different set of callbacks, so all of its settings (the global
436 as well as the per-policy ones) are th 419 as well as the per-policy ones) are then reset to their default
437 values, possibly depending on the targ 420 values, possibly depending on the target operation mode.]
438 421
439 ``energy_efficiency`` !! 422 That only is supported in some configurations, though (for example, if
440 This attribute is only present on plat !! 423 the `HWP feature is enabled in the processor <Active Mode With HWP_>`_,
441 Lake or Coffee Lake desktop CPU model. !! 424 the operation mode of the driver cannot be changed), and if it is not
442 optimizations are disabled on these CP !! 425 supported in the current configuration, writes to this attribute will
443 Enabling energy-efficiency optimizatio !! 426 fail with an appropriate error.
444 frequency with or without the HWP feat <<
445 optimizations are done only in the tur <<
446 they are done in the entire available <<
447 attribute to "1" enables the energy-ef <<
448 to "0" disables them. <<
449 427
450 Interpretation of Policy Attributes 428 Interpretation of Policy Attributes
451 ----------------------------------- 429 -----------------------------------
452 430
453 The interpretation of some ``CPUFreq`` policy 431 The interpretation of some ``CPUFreq`` policy attributes described in
454 Documentation/admin-guide/pm/cpufreq.rst is sp !! 432 :doc:`cpufreq` is special with ``intel_pstate`` as the current scaling driver
455 as the current scaling driver and it generally !! 433 and it generally depends on the driver's `operation mode <Operation Modes_>`_.
456 `operation mode <Operation Modes_>`_. <<
457 434
458 First of all, the values of the ``cpuinfo_max_ 435 First of all, the values of the ``cpuinfo_max_freq``, ``cpuinfo_min_freq`` and
459 ``scaling_cur_freq`` attributes are produced b 436 ``scaling_cur_freq`` attributes are produced by applying a processor-specific
460 multiplier to the internal P-state representat 437 multiplier to the internal P-state representation used by ``intel_pstate``.
461 Also, the values of the ``scaling_max_freq`` a 438 Also, the values of the ``scaling_max_freq`` and ``scaling_min_freq``
462 attributes are capped by the frequency corresp 439 attributes are capped by the frequency corresponding to the maximum P-state that
463 the driver is allowed to set. 440 the driver is allowed to set.
464 441
465 If the ``no_turbo`` `global attribute <no_turb 442 If the ``no_turbo`` `global attribute <no_turbo_attr_>`_ is set, the driver is
466 not allowed to use turbo P-states, so the maxi 443 not allowed to use turbo P-states, so the maximum value of ``scaling_max_freq``
467 and ``scaling_min_freq`` is limited to the max 444 and ``scaling_min_freq`` is limited to the maximum non-turbo P-state frequency.
468 Accordingly, setting ``no_turbo`` causes ``sca 445 Accordingly, setting ``no_turbo`` causes ``scaling_max_freq`` and
469 ``scaling_min_freq`` to go down to that value 446 ``scaling_min_freq`` to go down to that value if they were above it before.
470 However, the old values of ``scaling_max_freq` 447 However, the old values of ``scaling_max_freq`` and ``scaling_min_freq`` will be
471 restored after unsetting ``no_turbo``, unless 448 restored after unsetting ``no_turbo``, unless these attributes have been written
472 to after ``no_turbo`` was set. 449 to after ``no_turbo`` was set.
473 450
474 If ``no_turbo`` is not set, the maximum possib 451 If ``no_turbo`` is not set, the maximum possible value of ``scaling_max_freq``
475 and ``scaling_min_freq`` corresponds to the ma 452 and ``scaling_min_freq`` corresponds to the maximum supported turbo P-state,
476 which also is the value of ``cpuinfo_max_freq` 453 which also is the value of ``cpuinfo_max_freq`` in either case.
477 454
478 Next, the following policy attributes have spe 455 Next, the following policy attributes have special meaning if
479 ``intel_pstate`` works in the `active mode <Ac 456 ``intel_pstate`` works in the `active mode <Active Mode_>`_:
480 457
481 ``scaling_available_governors`` 458 ``scaling_available_governors``
482 List of P-state selection algorithms p 459 List of P-state selection algorithms provided by ``intel_pstate``.
483 460
484 ``scaling_governor`` 461 ``scaling_governor``
485 P-state selection algorithm provided b 462 P-state selection algorithm provided by ``intel_pstate`` currently in
486 use with the given policy. 463 use with the given policy.
487 464
488 ``scaling_cur_freq`` 465 ``scaling_cur_freq``
489 Frequency of the average P-state of th 466 Frequency of the average P-state of the CPU represented by the given
490 policy for the time interval between t 467 policy for the time interval between the last two invocations of the
491 driver's utilization update callback b 468 driver's utilization update callback by the CPU scheduler for that CPU.
492 469
493 One more policy attribute is present if the HW !! 470 One more policy attribute is present if the `HWP feature is enabled in the
494 processor: !! 471 processor <Active Mode With HWP_>`_:
495 472
496 ``base_frequency`` 473 ``base_frequency``
497 Shows the base frequency of the CPU. A 474 Shows the base frequency of the CPU. Any frequency above this will be
498 in the turbo frequency range. 475 in the turbo frequency range.
499 476
500 The meaning of these attributes in the `passiv 477 The meaning of these attributes in the `passive mode <Passive Mode_>`_ is the
501 same as for other scaling drivers. 478 same as for other scaling drivers.
502 479
503 Additionally, the value of the ``scaling_drive 480 Additionally, the value of the ``scaling_driver`` attribute for ``intel_pstate``
504 depends on the operation mode of the driver. 481 depends on the operation mode of the driver. Namely, it is either
505 "intel_pstate" (in the `active mode <Active Mo 482 "intel_pstate" (in the `active mode <Active Mode_>`_) or "intel_cpufreq" (in the
506 `passive mode <Passive Mode_>`_). 483 `passive mode <Passive Mode_>`_).
507 484
508 Coordination of P-State Limits 485 Coordination of P-State Limits
509 ------------------------------ 486 ------------------------------
510 487
511 ``intel_pstate`` allows P-state limits to be s 488 ``intel_pstate`` allows P-state limits to be set in two ways: with the help of
512 the ``max_perf_pct`` and ``min_perf_pct`` `glo 489 the ``max_perf_pct`` and ``min_perf_pct`` `global attributes
513 <Global Attributes_>`_ or via the ``scaling_ma 490 <Global Attributes_>`_ or via the ``scaling_max_freq`` and ``scaling_min_freq``
514 ``CPUFreq`` policy attributes. The coordinati 491 ``CPUFreq`` policy attributes. The coordination between those limits is based
515 on the following rules, regardless of the curr 492 on the following rules, regardless of the current operation mode of the driver:
516 493
517 1. All CPUs are affected by the global limits 494 1. All CPUs are affected by the global limits (that is, none of them can be
518 requested to run faster than the global ma 495 requested to run faster than the global maximum and none of them can be
519 requested to run slower than the global mi 496 requested to run slower than the global minimum).
520 497
521 2. Each individual CPU is affected by its own 498 2. Each individual CPU is affected by its own per-policy limits (that is, it
522 cannot be requested to run faster than its 499 cannot be requested to run faster than its own per-policy maximum and it
523 cannot be requested to run slower than its 500 cannot be requested to run slower than its own per-policy minimum). The
524 effective performance depends on whether t 501 effective performance depends on whether the platform supports per core
525 P-states, hyper-threading is enabled and o 502 P-states, hyper-threading is enabled and on current performance requests
526 from other CPUs. When platform doesn't sup 503 from other CPUs. When platform doesn't support per core P-states, the
527 effective performance can be more than the 504 effective performance can be more than the policy limits set on a CPU, if
528 other CPUs are requesting higher performan 505 other CPUs are requesting higher performance at that moment. Even with per
529 core P-states support, when hyper-threadin 506 core P-states support, when hyper-threading is enabled, if the sibling CPU
530 is requesting higher performance, the othe 507 is requesting higher performance, the other siblings will get higher
531 performance than their policy limits. 508 performance than their policy limits.
532 509
533 3. The global and per-policy limits can be se 510 3. The global and per-policy limits can be set independently.
534 511
535 In the `active mode with the HWP feature enabl !! 512 If the `HWP feature is enabled in the processor <Active Mode With HWP_>`_, the
536 resulting effective values are written into ha !! 513 resulting effective values are written into its registers whenever the limits
537 limits change in order to request its internal !! 514 change in order to request its internal P-state selection logic to always set
538 set P-states within these limits. Otherwise, !! 515 P-states within these limits. Otherwise, the limits are taken into account by
539 by scaling governors (in the `passive mode <Pa !! 516 scaling governors (in the `passive mode <Passive Mode_>`_) and by the driver
540 every time before setting a new P-state for a 517 every time before setting a new P-state for a CPU.
541 518
542 Additionally, if the ``intel_pstate=per_cpu_pe 519 Additionally, if the ``intel_pstate=per_cpu_perf_limits`` command line argument
543 is passed to the kernel, ``max_perf_pct`` and 520 is passed to the kernel, ``max_perf_pct`` and ``min_perf_pct`` are not exposed
544 at all and the only way to set the limits is b 521 at all and the only way to set the limits is by using the policy attributes.
545 522
546 523
547 Energy vs Performance Hints 524 Energy vs Performance Hints
548 --------------------------- 525 ---------------------------
549 526
550 If the hardware-managed P-states (HWP) is enab !! 527 If ``intel_pstate`` works in the `active mode with the HWP feature enabled
551 attributes, intended to allow user space to he !! 528 <Active Mode With HWP_>`_ in the processor, additional attributes are present
552 processor's internal P-state selection logic b !! 529 in every ``CPUFreq`` policy directory in ``sysfs``. They are intended to allow
553 energy-efficiency, or somewhere between the tw !! 530 user space to help ``intel_pstate`` to adjust the processor's internal P-state
554 ``CPUFreq`` policy directory in ``sysfs``. Th !! 531 selection logic by focusing it on performance or on energy-efficiency, or
>> 532 somewhere between the two extremes:
555 533
556 ``energy_performance_preference`` 534 ``energy_performance_preference``
557 Current value of the energy vs perform 535 Current value of the energy vs performance hint for the given policy
558 (or the CPU represented by it). 536 (or the CPU represented by it).
559 537
560 The hint can be changed by writing to 538 The hint can be changed by writing to this attribute.
561 539
562 ``energy_performance_available_preferences`` 540 ``energy_performance_available_preferences``
563 List of strings that can be written to 541 List of strings that can be written to the
564 ``energy_performance_preference`` attr 542 ``energy_performance_preference`` attribute.
565 543
566 They represent different energy vs per 544 They represent different energy vs performance hints and should be
567 self-explanatory, except that ``defaul 545 self-explanatory, except that ``default`` represents whatever hint
568 value was set by the platform firmware 546 value was set by the platform firmware.
569 547
570 Strings written to the ``energy_performance_pr 548 Strings written to the ``energy_performance_preference`` attribute are
571 internally translated to integer values writte 549 internally translated to integer values written to the processor's
572 Energy-Performance Preference (EPP) knob (if s 550 Energy-Performance Preference (EPP) knob (if supported) or its
573 Energy-Performance Bias (EPB) knob. It is also !! 551 Energy-Performance Bias (EPB) knob.
574 integer value between 0 to 255, if the EPP fea <<
575 feature is not present, writing integer value <<
576 supported. In this case, user can use the <<
577 "/sys/devices/system/cpu/cpu*/power/energy_per <<
578 552
579 [Note that tasks may by migrated from one CPU 553 [Note that tasks may by migrated from one CPU to another by the scheduler's
580 load-balancing algorithm and if different ener 554 load-balancing algorithm and if different energy vs performance hints are
581 set for those CPUs, that may lead to undesirab 555 set for those CPUs, that may lead to undesirable outcomes. To avoid such
582 issues it is better to set the same energy vs 556 issues it is better to set the same energy vs performance hint for all CPUs
583 or to pin every task potentially sensitive to 557 or to pin every task potentially sensitive to them to a specific CPU.]
584 558
585 .. _acpi-cpufreq: 559 .. _acpi-cpufreq:
586 560
587 ``intel_pstate`` vs ``acpi-cpufreq`` 561 ``intel_pstate`` vs ``acpi-cpufreq``
588 ==================================== 562 ====================================
589 563
590 On the majority of systems supported by ``inte 564 On the majority of systems supported by ``intel_pstate``, the ACPI tables
591 provided by the platform firmware contain ``_P 565 provided by the platform firmware contain ``_PSS`` objects returning information
592 that can be used for CPU performance scaling ( 566 that can be used for CPU performance scaling (refer to the ACPI specification
593 [3]_ for details on the ``_PSS`` objects and t 567 [3]_ for details on the ``_PSS`` objects and the format of the information
594 returned by them). 568 returned by them).
595 569
596 The information returned by the ACPI ``_PSS`` 570 The information returned by the ACPI ``_PSS`` objects is used by the
597 ``acpi-cpufreq`` scaling driver. On systems s 571 ``acpi-cpufreq`` scaling driver. On systems supported by ``intel_pstate``
598 the ``acpi-cpufreq`` driver uses the same hard 572 the ``acpi-cpufreq`` driver uses the same hardware CPU performance scaling
599 interface, but the set of P-states it can use 573 interface, but the set of P-states it can use is limited by the ``_PSS``
600 output. 574 output.
601 575
602 On those systems each ``_PSS`` object returns 576 On those systems each ``_PSS`` object returns a list of P-states supported by
603 the corresponding CPU which basically is a sub 577 the corresponding CPU which basically is a subset of the P-states range that can
604 be used by ``intel_pstate`` on the same system 578 be used by ``intel_pstate`` on the same system, with one exception: the whole
605 `turbo range <turbo_>`_ is represented by one 579 `turbo range <turbo_>`_ is represented by one item in it (the topmost one). By
606 convention, the frequency returned by ``_PSS`` 580 convention, the frequency returned by ``_PSS`` for that item is greater by 1 MHz
607 than the frequency of the highest non-turbo P- 581 than the frequency of the highest non-turbo P-state listed by it, but the
608 corresponding P-state representation (followin 582 corresponding P-state representation (following the hardware specification)
609 returned for it matches the maximum supported 583 returned for it matches the maximum supported turbo P-state (or is the
610 special value 255 meaning essentially "go as h 584 special value 255 meaning essentially "go as high as you can get").
611 585
612 The list of P-states returned by ``_PSS`` is r 586 The list of P-states returned by ``_PSS`` is reflected by the table of
613 available frequencies supplied by ``acpi-cpufr 587 available frequencies supplied by ``acpi-cpufreq`` to the ``CPUFreq`` core and
614 scaling governors and the minimum and maximum 588 scaling governors and the minimum and maximum supported frequencies reported by
615 it come from that list as well. In particular 589 it come from that list as well. In particular, given the special representation
616 of the turbo range described above, this means 590 of the turbo range described above, this means that the maximum supported
617 frequency reported by ``acpi-cpufreq`` is high 591 frequency reported by ``acpi-cpufreq`` is higher by 1 MHz than the frequency
618 of the highest supported non-turbo P-state lis 592 of the highest supported non-turbo P-state listed by ``_PSS`` which, of course,
619 affects decisions made by the scaling governor 593 affects decisions made by the scaling governors, except for ``powersave`` and
620 ``performance``. 594 ``performance``.
621 595
622 For example, if a given governor attempts to s 596 For example, if a given governor attempts to select a frequency proportional to
623 estimated CPU load and maps the load of 100% t 597 estimated CPU load and maps the load of 100% to the maximum supported frequency
624 (possibly multiplied by a constant), then it w 598 (possibly multiplied by a constant), then it will tend to choose P-states below
625 the turbo threshold if ``acpi-cpufreq`` is use 599 the turbo threshold if ``acpi-cpufreq`` is used as the scaling driver, because
626 in that case the turbo range corresponds to a 600 in that case the turbo range corresponds to a small fraction of the frequency
627 band it can use (1 MHz vs 1 GHz or more). In 601 band it can use (1 MHz vs 1 GHz or more). In consequence, it will only go to
628 the turbo range for the highest loads and the 602 the turbo range for the highest loads and the other loads above 50% that might
629 benefit from running at turbo frequencies will 603 benefit from running at turbo frequencies will be given non-turbo P-states
630 instead. 604 instead.
631 605
632 One more issue related to that may appear on s 606 One more issue related to that may appear on systems supporting the
633 `Configurable TDP feature <turbo_>`_ allowing 607 `Configurable TDP feature <turbo_>`_ allowing the platform firmware to set the
634 turbo threshold. Namely, if that is not coord 608 turbo threshold. Namely, if that is not coordinated with the lists of P-states
635 returned by ``_PSS`` properly, there may be mo 609 returned by ``_PSS`` properly, there may be more than one item corresponding to
636 a turbo P-state in those lists and there may b 610 a turbo P-state in those lists and there may be a problem with avoiding the
637 turbo range (if desirable or necessary). Usua 611 turbo range (if desirable or necessary). Usually, to avoid using turbo
638 P-states overall, ``acpi-cpufreq`` simply avoi 612 P-states overall, ``acpi-cpufreq`` simply avoids using the topmost state listed
639 by ``_PSS``, but that is not sufficient when t 613 by ``_PSS``, but that is not sufficient when there are other turbo P-states in
640 the list returned by it. 614 the list returned by it.
641 615
642 Apart from the above, ``acpi-cpufreq`` works l 616 Apart from the above, ``acpi-cpufreq`` works like ``intel_pstate`` in the
643 `passive mode <Passive Mode_>`_, except that t 617 `passive mode <Passive Mode_>`_, except that the number of P-states it can set
644 is limited to the ones listed by the ACPI ``_P 618 is limited to the ones listed by the ACPI ``_PSS`` objects.
645 619
646 620
647 Kernel Command Line Options for ``intel_pstate 621 Kernel Command Line Options for ``intel_pstate``
648 ============================================== 622 ================================================
649 623
650 Several kernel command line options can be use 624 Several kernel command line options can be used to pass early-configuration-time
651 parameters to ``intel_pstate`` in order to enf 625 parameters to ``intel_pstate`` in order to enforce specific behavior of it. All
652 of them have to be prepended with the ``intel_ 626 of them have to be prepended with the ``intel_pstate=`` prefix.
653 627
654 ``disable`` 628 ``disable``
655 Do not register ``intel_pstate`` as th 629 Do not register ``intel_pstate`` as the scaling driver even if the
656 processor is supported by it. 630 processor is supported by it.
657 631
658 ``active`` <<
659 Register ``intel_pstate`` in the `acti <<
660 with. <<
661 <<
662 ``passive`` 632 ``passive``
663 Register ``intel_pstate`` in the `pass 633 Register ``intel_pstate`` in the `passive mode <Passive Mode_>`_ to
664 start with. 634 start with.
665 635
>> 636 This option implies the ``no_hwp`` one described below.
>> 637
666 ``force`` 638 ``force``
667 Register ``intel_pstate`` as the scali 639 Register ``intel_pstate`` as the scaling driver instead of
668 ``acpi-cpufreq`` even if the latter is 640 ``acpi-cpufreq`` even if the latter is preferred on the given system.
669 641
670 This may prevent some platform feature 642 This may prevent some platform features (such as thermal controls and
671 power capping) that rely on the availa 643 power capping) that rely on the availability of ACPI P-states
672 information from functioning as expect 644 information from functioning as expected, so it should be used with
673 caution. 645 caution.
674 646
675 This option does not work with process 647 This option does not work with processors that are not supported by
676 ``intel_pstate`` and on platforms wher 648 ``intel_pstate`` and on platforms where the ``pcc-cpufreq`` scaling
677 driver is used instead of ``acpi-cpufr 649 driver is used instead of ``acpi-cpufreq``.
678 650
679 ``no_hwp`` 651 ``no_hwp``
680 Do not enable the hardware-managed P-s !! 652 Do not enable the `hardware-managed P-states (HWP) feature
681 supported by the processor. !! 653 <Active Mode With HWP_>`_ even if it is supported by the processor.
682 654
683 ``hwp_only`` 655 ``hwp_only``
684 Register ``intel_pstate`` as the scali 656 Register ``intel_pstate`` as the scaling driver only if the
685 hardware-managed P-states (HWP) featur !! 657 `hardware-managed P-states (HWP) feature <Active Mode With HWP_>`_ is
>> 658 supported by the processor.
686 659
687 ``support_acpi_ppc`` 660 ``support_acpi_ppc``
688 Take ACPI ``_PPC`` performance limits 661 Take ACPI ``_PPC`` performance limits into account.
689 662
690 If the preferred power management prof 663 If the preferred power management profile in the FADT (Fixed ACPI
691 Description Table) is set to "Enterpri 664 Description Table) is set to "Enterprise Server" or "Performance
692 Server", the ACPI ``_PPC`` limits are 665 Server", the ACPI ``_PPC`` limits are taken into account by default
693 and this option has no effect. 666 and this option has no effect.
694 667
695 ``per_cpu_perf_limits`` 668 ``per_cpu_perf_limits``
696 Use per-logical-CPU P-State limits (se 669 Use per-logical-CPU P-State limits (see `Coordination of P-state
697 Limits`_ for details). 670 Limits`_ for details).
698 671
699 672
700 Diagnostics and Tuning 673 Diagnostics and Tuning
701 ====================== 674 ======================
702 675
703 Trace Events 676 Trace Events
704 ------------ 677 ------------
705 678
706 There are two static trace events that can be 679 There are two static trace events that can be used for ``intel_pstate``
707 diagnostics. One of them is the ``cpu_frequen 680 diagnostics. One of them is the ``cpu_frequency`` trace event generally used
708 by ``CPUFreq``, and the other one is the ``pst 681 by ``CPUFreq``, and the other one is the ``pstate_sample`` trace event specific
709 to ``intel_pstate``. Both of them are trigger 682 to ``intel_pstate``. Both of them are triggered by ``intel_pstate`` only if
710 it works in the `active mode <Active Mode_>`_. 683 it works in the `active mode <Active Mode_>`_.
711 684
712 The following sequence of shell commands can b 685 The following sequence of shell commands can be used to enable them and see
713 their output (if the kernel is generally confi 686 their output (if the kernel is generally configured to support event tracing)::
714 687
715 # cd /sys/kernel/tracing/ !! 688 # cd /sys/kernel/debug/tracing/
716 # echo 1 > events/power/pstate_sample/enable 689 # echo 1 > events/power/pstate_sample/enable
717 # echo 1 > events/power/cpu_frequency/enable 690 # echo 1 > events/power/cpu_frequency/enable
718 # cat trace 691 # cat trace
719 gnome-terminal--4510 [001] ..s. 1177.680733 692 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
720 cat-5235 [002] ..s. 1177.681723: cpu_freque 693 cat-5235 [002] ..s. 1177.681723: cpu_frequency: state=2900000 cpu_id=2
721 694
722 If ``intel_pstate`` works in the `passive mode 695 If ``intel_pstate`` works in the `passive mode <Passive Mode_>`_, the
723 ``cpu_frequency`` trace event will be triggere 696 ``cpu_frequency`` trace event will be triggered either by the ``schedutil``
724 scaling governor (for the policies it is attac 697 scaling governor (for the policies it is attached to), or by the ``CPUFreq``
725 core (for the policies with other scaling gove 698 core (for the policies with other scaling governors).
726 699
727 ``ftrace`` 700 ``ftrace``
728 ---------- 701 ----------
729 702
730 The ``ftrace`` interface can be used for low-l 703 The ``ftrace`` interface can be used for low-level diagnostics of
731 ``intel_pstate``. For example, to check how o 704 ``intel_pstate``. For example, to check how often the function to set a
732 P-state is called, the ``ftrace`` filter can b !! 705 P-state is called, the ``ftrace`` filter can be set to to
733 :c:func:`intel_pstate_set_pstate`:: 706 :c:func:`intel_pstate_set_pstate`::
734 707
735 # cd /sys/kernel/tracing/ !! 708 # cd /sys/kernel/debug/tracing/
736 # cat available_filter_functions | grep -i ps 709 # cat available_filter_functions | grep -i pstate
737 intel_pstate_set_pstate 710 intel_pstate_set_pstate
738 intel_pstate_cpu_init 711 intel_pstate_cpu_init
739 ... 712 ...
740 # echo intel_pstate_set_pstate > set_ftrace_f 713 # echo intel_pstate_set_pstate > set_ftrace_filter
741 # echo function > current_tracer 714 # echo function > current_tracer
742 # cat trace | head -15 715 # cat trace | head -15
743 # tracer: function 716 # tracer: function
744 # 717 #
745 # entries-in-buffer/entries-written: 80/80 718 # entries-in-buffer/entries-written: 80/80 #P:4
746 # 719 #
747 # _-----=> irqs- 720 # _-----=> irqs-off
748 # / _----=> need- 721 # / _----=> need-resched
749 # | / _---=> hardi 722 # | / _---=> hardirq/softirq
750 # || / _--=> preem 723 # || / _--=> preempt-depth
751 # ||| / delay 724 # ||| / delay
752 # TASK-PID CPU# |||| TIMESTAM 725 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
753 # | | | |||| | 726 # | | | |||| | |
754 Xorg-3129 [000] ..s. 2537.64484 727 Xorg-3129 [000] ..s. 2537.644844: intel_pstate_set_pstate <-intel_pstate_timer_func
755 gnome-terminal--4510 [002] ..s. 2537.64984 728 gnome-terminal--4510 [002] ..s. 2537.649844: intel_pstate_set_pstate <-intel_pstate_timer_func
756 gnome-shell-3409 [001] ..s. 2537.65085 729 gnome-shell-3409 [001] ..s. 2537.650850: intel_pstate_set_pstate <-intel_pstate_timer_func
757 <idle>-0 [000] ..s. 2537.65484 730 <idle>-0 [000] ..s. 2537.654843: intel_pstate_set_pstate <-intel_pstate_timer_func
758 731
759 732
760 References 733 References
761 ========== 734 ==========
762 735
763 .. [1] Kristen Accardi, *Balancing Power and P 736 .. [1] Kristen Accardi, *Balancing Power and Performance in the Linux Kernel*,
764 https://events.static.linuxfound.org/si 737 https://events.static.linuxfound.org/sites/events/files/slides/LinuxConEurope_2015.pdf
765 738
766 .. [2] *Intel® 64 and IA-32 Architectures Sof 739 .. [2] *Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 3: System Programming Guide*,
767 https://www.intel.com/content/www/us/en 740 https://www.intel.com/content/www/us/en/architecture-and-technology/64-ia-32-architectures-software-developer-system-programming-manual-325384.html
768 741
769 .. [3] *Advanced Configuration and Power Inter 742 .. [3] *Advanced Configuration and Power Interface Specification*,
770 https://uefi.org/sites/default/files/re 743 https://uefi.org/sites/default/files/resources/ACPI_6_3_final_Jan30.pdf
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.