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 ``amd-pstate`` CPU Performance Scaling Driver 5 ``amd-pstate`` CPU Performance Scaling Driver
6 ============================================== 6 ===============================================
7 7
8 :Copyright: |copy| 2021 Advanced Micro Devices 8 :Copyright: |copy| 2021 Advanced Micro Devices, Inc.
9 9
10 :Author: Huang Rui <ray.huang@amd.com> 10 :Author: Huang Rui <ray.huang@amd.com>
11 11
12 12
13 Introduction 13 Introduction
14 =================== 14 ===================
15 15
16 ``amd-pstate`` is the AMD CPU performance scal 16 ``amd-pstate`` is the AMD CPU performance scaling driver that introduces a
17 new CPU frequency control mechanism on modern 17 new CPU frequency control mechanism on modern AMD APU and CPU series in
18 Linux kernel. The new mechanism is based on Co 18 Linux kernel. The new mechanism is based on Collaborative Processor
19 Performance Control (CPPC) which provides fine 19 Performance Control (CPPC) which provides finer grain frequency management
20 than legacy ACPI hardware P-States. Current AM 20 than legacy ACPI hardware P-States. Current AMD CPU/APU platforms are using
21 the ACPI P-states driver to manage CPU frequen 21 the ACPI P-states driver to manage CPU frequency and clocks with switching
22 only in 3 P-states. CPPC replaces the ACPI P-s 22 only in 3 P-states. CPPC replaces the ACPI P-states controls and allows a
23 flexible, low-latency interface for the Linux 23 flexible, low-latency interface for the Linux kernel to directly
24 communicate the performance hints to hardware. 24 communicate the performance hints to hardware.
25 25
26 ``amd-pstate`` leverages the Linux kernel gove 26 ``amd-pstate`` leverages the Linux kernel governors such as ``schedutil``,
27 ``ondemand``, etc. to manage the performance h 27 ``ondemand``, etc. to manage the performance hints which are provided by
28 CPPC hardware functionality that internally fo 28 CPPC hardware functionality that internally follows the hardware
29 specification (for details refer to AMD64 Arch 29 specification (for details refer to AMD64 Architecture Programmer's Manual
30 Volume 2: System Programming [1]_). Currently, 30 Volume 2: System Programming [1]_). Currently, ``amd-pstate`` supports basic
31 frequency control function according to kernel 31 frequency control function according to kernel governors on some of the
32 Zen2 and Zen3 processors, and we will implemen 32 Zen2 and Zen3 processors, and we will implement more AMD specific functions
33 in future after we verify them on the hardware 33 in future after we verify them on the hardware and SBIOS.
34 34
35 35
36 AMD CPPC Overview 36 AMD CPPC Overview
37 ======================= 37 =======================
38 38
39 Collaborative Processor Performance Control (C 39 Collaborative Processor Performance Control (CPPC) interface enumerates a
40 continuous, abstract, and unit-less performanc 40 continuous, abstract, and unit-less performance value in a scale that is
41 not tied to a specific performance state / fre 41 not tied to a specific performance state / frequency. This is an ACPI
42 standard [2]_ which software can specify appli 42 standard [2]_ which software can specify application performance goals and
43 hints as a relative target to the infrastructu 43 hints as a relative target to the infrastructure limits. AMD processors
44 provide the low latency register model (MSR) i 44 provide the low latency register model (MSR) instead of an AML code
45 interpreter for performance adjustments. ``amd 45 interpreter for performance adjustments. ``amd-pstate`` will initialize a
46 ``struct cpufreq_driver`` instance, ``amd_psta 46 ``struct cpufreq_driver`` instance, ``amd_pstate_driver``, with the callbacks
47 to manage each performance update behavior. :: 47 to manage each performance update behavior. ::
48 48
49 Highest Perf ------>+-----------------------+ 49 Highest Perf ------>+-----------------------+ +-----------------------+
50 | | 50 | | | |
51 | | 51 | | | |
52 | | 52 | | Max Perf ---->| |
53 | | 53 | | | |
54 | | 54 | | | |
55 Nominal Perf ------>+-----------------------+ 55 Nominal Perf ------>+-----------------------+ +-----------------------+
56 | | 56 | | | |
57 | | 57 | | | |
58 | | 58 | | | |
59 | | 59 | | | |
60 | | 60 | | | |
61 | | 61 | | | |
62 | | 62 | | Desired Perf ---->| |
63 | | 63 | | | |
64 | | 64 | | | |
65 | | 65 | | | |
66 | | 66 | | | |
67 | | 67 | | | |
68 | | 68 | | | |
69 | | 69 | | | |
70 | | 70 | | | |
71 | | 71 | | | |
72 Lowest non- | | 72 Lowest non- | | | |
73 linear perf ------>+-----------------------+ 73 linear perf ------>+-----------------------+ +-----------------------+
74 | | 74 | | | |
75 | | 75 | | Lowest perf ---->| |
76 | | 76 | | | |
77 Lowest perf ------>+-----------------------+ 77 Lowest perf ------>+-----------------------+ +-----------------------+
78 | | 78 | | | |
79 | | 79 | | | |
80 | | 80 | | | |
81 0 ------>+-----------------------+ 81 0 ------>+-----------------------+ +-----------------------+
82 82
83 AMD P-Sta 83 AMD P-States Performance Scale
84 84
85 85
86 .. _perf_cap: 86 .. _perf_cap:
87 87
88 AMD CPPC Performance Capability 88 AMD CPPC Performance Capability
89 -------------------------------- 89 --------------------------------
90 90
91 Highest Performance (RO) 91 Highest Performance (RO)
92 ......................... 92 .........................
93 93
94 This is the absolute maximum performance an in 94 This is the absolute maximum performance an individual processor may reach,
95 assuming ideal conditions. This performance le 95 assuming ideal conditions. This performance level may not be sustainable
96 for long durations and may only be achievable 96 for long durations and may only be achievable if other platform components
97 are in a specific state; for example, it may r 97 are in a specific state; for example, it may require other processors to be in
98 an idle state. This would be equivalent to the 98 an idle state. This would be equivalent to the highest frequencies
99 supported by the processor. 99 supported by the processor.
100 100
101 Nominal (Guaranteed) Performance (RO) 101 Nominal (Guaranteed) Performance (RO)
102 ...................................... 102 ......................................
103 103
104 This is the maximum sustained performance leve 104 This is the maximum sustained performance level of the processor, assuming
105 ideal operating conditions. In the absence of 105 ideal operating conditions. In the absence of an external constraint (power,
106 thermal, etc.), this is the performance level 106 thermal, etc.), this is the performance level the processor is expected to
107 be able to maintain continuously. All cores/pr 107 be able to maintain continuously. All cores/processors are expected to be
108 able to sustain their nominal performance stat 108 able to sustain their nominal performance state simultaneously.
109 109
110 Lowest non-linear Performance (RO) 110 Lowest non-linear Performance (RO)
111 ................................... 111 ...................................
112 112
113 This is the lowest performance level at which 113 This is the lowest performance level at which nonlinear power savings are
114 achieved, for example, due to the combined eff 114 achieved, for example, due to the combined effects of voltage and frequency
115 scaling. Above this threshold, lower performan 115 scaling. Above this threshold, lower performance levels should be generally
116 more energy efficient than higher performance 116 more energy efficient than higher performance levels. This register
117 effectively conveys the most efficient perform 117 effectively conveys the most efficient performance level to ``amd-pstate``.
118 118
119 Lowest Performance (RO) 119 Lowest Performance (RO)
120 ........................ 120 ........................
121 121
122 This is the absolute lowest performance level 122 This is the absolute lowest performance level of the processor. Selecting a
123 performance level lower than the lowest nonlin 123 performance level lower than the lowest nonlinear performance level may
124 cause an efficiency penalty but should reduce 124 cause an efficiency penalty but should reduce the instantaneous power
125 consumption of the processor. 125 consumption of the processor.
126 126
127 AMD CPPC Performance Control 127 AMD CPPC Performance Control
128 ------------------------------ 128 ------------------------------
129 129
130 ``amd-pstate`` passes performance goals throug 130 ``amd-pstate`` passes performance goals through these registers. The
131 register drives the behavior of the desired pe 131 register drives the behavior of the desired performance target.
132 132
133 Minimum requested performance (RW) 133 Minimum requested performance (RW)
134 ................................... 134 ...................................
135 135
136 ``amd-pstate`` specifies the minimum allowed p 136 ``amd-pstate`` specifies the minimum allowed performance level.
137 137
138 Maximum requested performance (RW) 138 Maximum requested performance (RW)
139 ................................... 139 ...................................
140 140
141 ``amd-pstate`` specifies a limit the maximum p 141 ``amd-pstate`` specifies a limit the maximum performance that is expected
142 to be supplied by the hardware. 142 to be supplied by the hardware.
143 143
144 Desired performance target (RW) 144 Desired performance target (RW)
145 ................................... 145 ...................................
146 146
147 ``amd-pstate`` specifies a desired target in t 147 ``amd-pstate`` specifies a desired target in the CPPC performance scale as
148 a relative number. This can be expressed as pe 148 a relative number. This can be expressed as percentage of nominal
149 performance (infrastructure max). Below the no 149 performance (infrastructure max). Below the nominal sustained performance
150 level, desired performance expresses the avera 150 level, desired performance expresses the average performance level of the
151 processor subject to hardware. Above the nomin 151 processor subject to hardware. Above the nominal performance level,
152 the processor must provide at least nominal pe 152 the processor must provide at least nominal performance requested and go higher
153 if current operating conditions allow. 153 if current operating conditions allow.
154 154
155 Energy Performance Preference (EPP) (RW) 155 Energy Performance Preference (EPP) (RW)
156 ......................................... 156 .........................................
157 157
158 This attribute provides a hint to the hardware 158 This attribute provides a hint to the hardware if software wants to bias
159 toward performance (0x0) or energy efficiency 159 toward performance (0x0) or energy efficiency (0xff).
160 160
161 161
162 Key Governors Support 162 Key Governors Support
163 ======================= 163 =======================
164 164
165 ``amd-pstate`` can be used with all the (gener 165 ``amd-pstate`` can be used with all the (generic) scaling governors listed
166 by the ``scaling_available_governors`` policy 166 by the ``scaling_available_governors`` policy attribute in ``sysfs``. Then,
167 it is responsible for the configuration of pol 167 it is responsible for the configuration of policy objects corresponding to
168 CPUs and provides the ``CPUFreq`` core (and th 168 CPUs and provides the ``CPUFreq`` core (and the scaling governors attached
169 to the policy objects) with accurate informati 169 to the policy objects) with accurate information on the maximum and minimum
170 operating frequencies supported by the hardwar 170 operating frequencies supported by the hardware. Users can check the
171 ``scaling_cur_freq`` information comes from th 171 ``scaling_cur_freq`` information comes from the ``CPUFreq`` core.
172 172
173 ``amd-pstate`` mainly supports ``schedutil`` a 173 ``amd-pstate`` mainly supports ``schedutil`` and ``ondemand`` for dynamic
174 frequency control. It is to fine tune the proc 174 frequency control. It is to fine tune the processor configuration on
175 ``amd-pstate`` to the ``schedutil`` with CPU C 175 ``amd-pstate`` to the ``schedutil`` with CPU CFS scheduler. ``amd-pstate``
176 registers the adjust_perf callback to implemen 176 registers the adjust_perf callback to implement performance update behavior
177 similar to CPPC. It is initialized by ``sugov_ 177 similar to CPPC. It is initialized by ``sugov_start`` and then populates the
178 CPU's update_util_data pointer to assign ``sug 178 CPU's update_util_data pointer to assign ``sugov_update_single_perf`` as the
179 utilization update callback function in the CP 179 utilization update callback function in the CPU scheduler. The CPU scheduler
180 will call ``cpufreq_update_util`` and assigns 180 will call ``cpufreq_update_util`` and assigns the target performance according
181 to the ``struct sugov_cpu`` that the utilizati 181 to the ``struct sugov_cpu`` that the utilization update belongs to.
182 Then, ``amd-pstate`` updates the desired perfo 182 Then, ``amd-pstate`` updates the desired performance according to the CPU
183 scheduler assigned. 183 scheduler assigned.
184 184
185 .. _processor_support: 185 .. _processor_support:
186 186
187 Processor Support 187 Processor Support
188 ======================= 188 =======================
189 189
190 The ``amd-pstate`` initialization will fail if 190 The ``amd-pstate`` initialization will fail if the ``_CPC`` entry in the ACPI
191 SBIOS does not exist in the detected processor 191 SBIOS does not exist in the detected processor. It uses ``acpi_cpc_valid``
192 to check the existence of ``_CPC``. All Zen ba 192 to check the existence of ``_CPC``. All Zen based processors support the legacy
193 ACPI hardware P-States function, so when ``amd 193 ACPI hardware P-States function, so when ``amd-pstate`` fails initialization,
194 the kernel will fall back to initialize the `` 194 the kernel will fall back to initialize the ``acpi-cpufreq`` driver.
195 195
196 There are two types of hardware implementation 196 There are two types of hardware implementations for ``amd-pstate``: one is
197 `Full MSR Support <perf_cap_>`_ and another is 197 `Full MSR Support <perf_cap_>`_ and another is `Shared Memory Support
198 <perf_cap_>`_. It can use the :c:macro:`X86_FE 198 <perf_cap_>`_. It can use the :c:macro:`X86_FEATURE_CPPC` feature flag to
199 indicate the different types. (For details, re 199 indicate the different types. (For details, refer to the Processor Programming
200 Reference (PPR) for AMD Family 19h Model 51h, 200 Reference (PPR) for AMD Family 19h Model 51h, Revision A1 Processors [3]_.)
201 ``amd-pstate`` is to register different ``stat 201 ``amd-pstate`` is to register different ``static_call`` instances for different
202 hardware implementations. 202 hardware implementations.
203 203
204 Currently, some of the Zen2 and Zen3 processor 204 Currently, some of the Zen2 and Zen3 processors support ``amd-pstate``. In the
205 future, it will be supported on more and more 205 future, it will be supported on more and more AMD processors.
206 206
207 Full MSR Support 207 Full MSR Support
208 ----------------- 208 -----------------
209 209
210 Some new Zen3 processors such as Cezanne provi 210 Some new Zen3 processors such as Cezanne provide the MSR registers directly
211 while the :c:macro:`X86_FEATURE_CPPC` CPU feat 211 while the :c:macro:`X86_FEATURE_CPPC` CPU feature flag is set.
212 ``amd-pstate`` can handle the MSR register to 212 ``amd-pstate`` can handle the MSR register to implement the fast switch
213 function in ``CPUFreq`` that can reduce the la 213 function in ``CPUFreq`` that can reduce the latency of frequency control in
214 interrupt context. The functions with a ``psta 214 interrupt context. The functions with a ``pstate_xxx`` prefix represent the
215 operations on MSR registers. 215 operations on MSR registers.
216 216
217 Shared Memory Support 217 Shared Memory Support
218 ---------------------- 218 ----------------------
219 219
220 If the :c:macro:`X86_FEATURE_CPPC` CPU feature 220 If the :c:macro:`X86_FEATURE_CPPC` CPU feature flag is not set, the
221 processor supports the shared memory solution. 221 processor supports the shared memory solution. In this case, ``amd-pstate``
222 uses the ``cppc_acpi`` helper methods to imple 222 uses the ``cppc_acpi`` helper methods to implement the callback functions
223 that are defined on ``static_call``. The funct 223 that are defined on ``static_call``. The functions with the ``cppc_xxx`` prefix
224 represent the operations of ACPI CPPC helpers 224 represent the operations of ACPI CPPC helpers for the shared memory solution.
225 225
226 226
227 AMD P-States and ACPI hardware P-States always 227 AMD P-States and ACPI hardware P-States always can be supported in one
228 processor. But AMD P-States has the higher pri 228 processor. But AMD P-States has the higher priority and if it is enabled
229 with :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_ 229 with :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_set_enable``, it will respond
230 to the request from AMD P-States. 230 to the request from AMD P-States.
231 231
232 232
233 User Space Interface in ``sysfs`` - Per-policy 233 User Space Interface in ``sysfs`` - Per-policy control
234 ============================================== 234 ======================================================
235 235
236 ``amd-pstate`` exposes several global attribut 236 ``amd-pstate`` exposes several global attributes (files) in ``sysfs`` to
237 control its functionality at the system level. 237 control its functionality at the system level. They are located in the
238 ``/sys/devices/system/cpu/cpufreq/policyX/`` d 238 ``/sys/devices/system/cpu/cpufreq/policyX/`` directory and affect all CPUs. ::
239 239
240 root@hr-test1:/home/ray# ls /sys/devices/syst 240 root@hr-test1:/home/ray# ls /sys/devices/system/cpu/cpufreq/policy0/*amd*
241 /sys/devices/system/cpu/cpufreq/policy0/amd_p 241 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_highest_perf
242 /sys/devices/system/cpu/cpufreq/policy0/amd_p 242 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_lowest_nonlinear_freq
243 /sys/devices/system/cpu/cpufreq/policy0/amd_p 243 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_max_freq
244 244
245 245
246 ``amd_pstate_highest_perf / amd_pstate_max_fre 246 ``amd_pstate_highest_perf / amd_pstate_max_freq``
247 247
248 Maximum CPPC performance and CPU frequency tha 248 Maximum CPPC performance and CPU frequency that the driver is allowed to
249 set, in percent of the maximum supported CPPC 249 set, in percent of the maximum supported CPPC performance level (the highest
250 performance supported in `AMD CPPC Performance 250 performance supported in `AMD CPPC Performance Capability <perf_cap_>`_).
251 In some ASICs, the highest CPPC performance is 251 In some ASICs, the highest CPPC performance is not the one in the ``_CPC``
252 table, so we need to expose it to sysfs. If bo 252 table, so we need to expose it to sysfs. If boost is not active, but
253 still supported, this maximum frequency will b 253 still supported, this maximum frequency will be larger than the one in
254 ``cpuinfo``. On systems that support preferred 254 ``cpuinfo``. On systems that support preferred core, the driver will have
255 different values for some cores than others an 255 different values for some cores than others and this will reflect the values
256 advertised by the platform at bootup. 256 advertised by the platform at bootup.
257 This attribute is read-only. 257 This attribute is read-only.
258 258
259 ``amd_pstate_lowest_nonlinear_freq`` 259 ``amd_pstate_lowest_nonlinear_freq``
260 260
261 The lowest non-linear CPPC CPU frequency that 261 The lowest non-linear CPPC CPU frequency that the driver is allowed to set,
262 in percent of the maximum supported CPPC perfo 262 in percent of the maximum supported CPPC performance level. (Please see the
263 lowest non-linear performance in `AMD CPPC Per 263 lowest non-linear performance in `AMD CPPC Performance Capability
264 <perf_cap_>`_.) 264 <perf_cap_>`_.)
265 This attribute is read-only. 265 This attribute is read-only.
266 266
267 ``amd_pstate_hw_prefcore`` 267 ``amd_pstate_hw_prefcore``
268 268
269 Whether the platform supports the preferred co 269 Whether the platform supports the preferred core feature and it has been
270 enabled. This attribute is read-only. 270 enabled. This attribute is read-only.
271 271
272 ``amd_pstate_prefcore_ranking`` 272 ``amd_pstate_prefcore_ranking``
273 273
274 The performance ranking of the core. This numb 274 The performance ranking of the core. This number doesn't have any unit, but
275 larger numbers are preferred at the time of re 275 larger numbers are preferred at the time of reading. This can change at
276 runtime based on platform conditions. This att 276 runtime based on platform conditions. This attribute is read-only.
277 277
278 ``energy_performance_available_preferences`` 278 ``energy_performance_available_preferences``
279 279
280 A list of all the supported EPP preferences th 280 A list of all the supported EPP preferences that could be used for
281 ``energy_performance_preference`` on this syst 281 ``energy_performance_preference`` on this system.
282 These profiles represent different hints that 282 These profiles represent different hints that are provided
283 to the low-level firmware about the user's des 283 to the low-level firmware about the user's desired energy vs efficiency
284 tradeoff. ``default`` represents the epp valu 284 tradeoff. ``default`` represents the epp value is set by platform
285 firmware. This attribute is read-only. 285 firmware. This attribute is read-only.
286 286
287 ``energy_performance_preference`` 287 ``energy_performance_preference``
288 288
289 The current energy performance preference can 289 The current energy performance preference can be read from this attribute.
290 and user can change current preference accordi 290 and user can change current preference according to energy or performance needs
291 Please get all support profiles list from 291 Please get all support profiles list from
292 ``energy_performance_available_preferences`` a 292 ``energy_performance_available_preferences`` attribute, all the profiles are
293 integer values defined between 0 to 255 when E 293 integer values defined between 0 to 255 when EPP feature is enabled by platform
294 firmware, if EPP feature is disabled, driver w 294 firmware, if EPP feature is disabled, driver will ignore the written value
295 This attribute is read-write. 295 This attribute is read-write.
296 296
297 ``boost`` 297 ``boost``
298 The `boost` sysfs attribute provides control o 298 The `boost` sysfs attribute provides control over the CPU core
299 performance boost, allowing users to manage th 299 performance boost, allowing users to manage the maximum frequency limitation
300 of the CPU. This attribute can be used to enab 300 of the CPU. This attribute can be used to enable or disable the boost feature
301 on individual CPUs. 301 on individual CPUs.
302 302
303 When the boost feature is enabled, the CPU can 303 When the boost feature is enabled, the CPU can dynamically increase its frequency
304 beyond the base frequency, providing enhanced 304 beyond the base frequency, providing enhanced performance for demanding workloads.
305 On the other hand, disabling the boost feature 305 On the other hand, disabling the boost feature restricts the CPU to operate at the
306 base frequency, which may be desirable in cert 306 base frequency, which may be desirable in certain scenarios to prioritize power
307 efficiency or manage temperature. 307 efficiency or manage temperature.
308 308
309 To manipulate the `boost` attribute, users can 309 To manipulate the `boost` attribute, users can write a value of `0` to disable the
310 boost or `1` to enable it, for the respective 310 boost or `1` to enable it, for the respective CPU using the sysfs path
311 `/sys/devices/system/cpu/cpuX/cpufreq/boost`, 311 `/sys/devices/system/cpu/cpuX/cpufreq/boost`, where `X` represents the CPU number.
312 312
313 Other performance and frequency values can be 313 Other performance and frequency values can be read back from
314 ``/sys/devices/system/cpu/cpuX/acpi_cppc/``, s 314 ``/sys/devices/system/cpu/cpuX/acpi_cppc/``, see :ref:`cppc_sysfs`.
315 315
316 316
317 ``amd-pstate`` vs ``acpi-cpufreq`` 317 ``amd-pstate`` vs ``acpi-cpufreq``
318 ====================================== 318 ======================================
319 319
320 On the majority of AMD platforms supported by 320 On the majority of AMD platforms supported by ``acpi-cpufreq``, the ACPI tables
321 provided by the platform firmware are used for 321 provided by the platform firmware are used for CPU performance scaling, but
322 only provide 3 P-states on AMD processors. 322 only provide 3 P-states on AMD processors.
323 However, on modern AMD APU and CPU series, har 323 However, on modern AMD APU and CPU series, hardware provides the Collaborative
324 Processor Performance Control according to the 324 Processor Performance Control according to the ACPI protocol and customizes this
325 for AMD platforms. That is, fine-grained and c 325 for AMD platforms. That is, fine-grained and continuous frequency ranges
326 instead of the legacy hardware P-states. ``amd 326 instead of the legacy hardware P-states. ``amd-pstate`` is the kernel
327 module which supports the new AMD P-States mec 327 module which supports the new AMD P-States mechanism on most of the future AMD
328 platforms. The AMD P-States mechanism is the m 328 platforms. The AMD P-States mechanism is the more performance and energy
329 efficiency frequency management method on AMD 329 efficiency frequency management method on AMD processors.
330 330
331 331
332 ``amd-pstate`` Driver Operation Modes 332 ``amd-pstate`` Driver Operation Modes
333 ====================================== 333 ======================================
334 334
335 ``amd_pstate`` CPPC has 3 operation modes: aut 335 ``amd_pstate`` CPPC has 3 operation modes: autonomous (active) mode,
336 non-autonomous (passive) mode and guided auton 336 non-autonomous (passive) mode and guided autonomous (guided) mode.
337 Active/passive/guided mode can be chosen by di 337 Active/passive/guided mode can be chosen by different kernel parameters.
338 338
339 - In autonomous mode, platform ignores the des 339 - In autonomous mode, platform ignores the desired performance level request
340 and takes into account only the values set t 340 and takes into account only the values set to the minimum, maximum and energy
341 performance preference registers. 341 performance preference registers.
342 - In non-autonomous mode, platform gets desire 342 - In non-autonomous mode, platform gets desired performance level
343 from OS directly through Desired Performance 343 from OS directly through Desired Performance Register.
344 - In guided-autonomous mode, platform sets ope 344 - In guided-autonomous mode, platform sets operating performance level
345 autonomously according to the current worklo 345 autonomously according to the current workload and within the limits set by
346 OS through min and max performance registers 346 OS through min and max performance registers.
347 347
348 Active Mode 348 Active Mode
349 ------------ 349 ------------
350 350
351 ``amd_pstate=active`` 351 ``amd_pstate=active``
352 352
353 This is the low-level firmware control mode wh 353 This is the low-level firmware control mode which is implemented by ``amd_pstate_epp``
354 driver with ``amd_pstate=active`` passed to th 354 driver with ``amd_pstate=active`` passed to the kernel in the command line.
355 In this mode, ``amd_pstate_epp`` driver provid 355 In this mode, ``amd_pstate_epp`` driver provides a hint to the hardware if software
356 wants to bias toward performance (0x0) or ener 356 wants to bias toward performance (0x0) or energy efficiency (0xff) to the CPPC firmware.
357 then CPPC power algorithm will calculate the r 357 then CPPC power algorithm will calculate the runtime workload and adjust the realtime
358 cores frequency according to the power supply 358 cores frequency according to the power supply and thermal, core voltage and some other
359 hardware conditions. 359 hardware conditions.
360 360
361 Passive Mode 361 Passive Mode
362 ------------ 362 ------------
363 363
364 ``amd_pstate=passive`` 364 ``amd_pstate=passive``
365 365
366 It will be enabled if the ``amd_pstate=passive 366 It will be enabled if the ``amd_pstate=passive`` is passed to the kernel in the command line.
367 In this mode, ``amd_pstate`` driver software s 367 In this mode, ``amd_pstate`` driver software specifies a desired QoS target in the CPPC
368 performance scale as a relative number. This c 368 performance scale as a relative number. This can be expressed as percentage of nominal
369 performance (infrastructure max). Below the no 369 performance (infrastructure max). Below the nominal sustained performance level,
370 desired performance expresses the average perf 370 desired performance expresses the average performance level of the processor subject
371 to the Performance Reduction Tolerance registe 371 to the Performance Reduction Tolerance register. Above the nominal performance level,
372 processor must provide at least nominal perfor 372 processor must provide at least nominal performance requested and go higher if current
373 operating conditions allow. 373 operating conditions allow.
374 374
375 Guided Mode 375 Guided Mode
376 ----------- 376 -----------
377 377
378 ``amd_pstate=guided`` 378 ``amd_pstate=guided``
379 379
380 If ``amd_pstate=guided`` is passed to kernel c 380 If ``amd_pstate=guided`` is passed to kernel command line option then this mode
381 is activated. In this mode, driver requests m 381 is activated. In this mode, driver requests minimum and maximum performance
382 level and the platform autonomously selects a 382 level and the platform autonomously selects a performance level in this range
383 and appropriate to the current workload. 383 and appropriate to the current workload.
384 384
385 ``amd-pstate`` Preferred Core 385 ``amd-pstate`` Preferred Core
386 ================================= 386 =================================
387 387
388 The core frequency is subjected to the process 388 The core frequency is subjected to the process variation in semiconductors.
389 Not all cores are able to reach the maximum fr 389 Not all cores are able to reach the maximum frequency respecting the
390 infrastructure limits. Consequently, AMD has r 390 infrastructure limits. Consequently, AMD has redefined the concept of
391 maximum frequency of a part. This means that a 391 maximum frequency of a part. This means that a fraction of cores can reach
392 maximum frequency. To find the best process sc 392 maximum frequency. To find the best process scheduling policy for a given
393 scenario, OS needs to know the core ordering i 393 scenario, OS needs to know the core ordering informed by the platform through
394 highest performance capability register of the 394 highest performance capability register of the CPPC interface.
395 395
396 ``amd-pstate`` preferred core enables the sche 396 ``amd-pstate`` preferred core enables the scheduler to prefer scheduling on
397 cores that can achieve a higher frequency with 397 cores that can achieve a higher frequency with lower voltage. The preferred
398 core rankings can dynamically change based on 398 core rankings can dynamically change based on the workload, platform conditions,
399 thermals and ageing. 399 thermals and ageing.
400 400
401 The priority metric will be initialized by the 401 The priority metric will be initialized by the ``amd-pstate`` driver. The ``amd-pstate``
402 driver will also determine whether or not ``am 402 driver will also determine whether or not ``amd-pstate`` preferred core is
403 supported by the platform. 403 supported by the platform.
404 404
405 ``amd-pstate`` driver will provide an initial 405 ``amd-pstate`` driver will provide an initial core ordering when the system boots.
406 The platform uses the CPPC interfaces to commu 406 The platform uses the CPPC interfaces to communicate the core ranking to the
407 operating system and scheduler to make sure th 407 operating system and scheduler to make sure that OS is choosing the cores
408 with highest performance firstly for schedulin 408 with highest performance firstly for scheduling the process. When ``amd-pstate``
409 driver receives a message with the highest per 409 driver receives a message with the highest performance change, it will
410 update the core ranking and set the cpu's prio 410 update the core ranking and set the cpu's priority.
411 411
412 ``amd-pstate`` Preferred Core Switch 412 ``amd-pstate`` Preferred Core Switch
413 ===================================== 413 =====================================
414 Kernel Parameters 414 Kernel Parameters
415 ----------------- 415 -----------------
416 416
417 ``amd-pstate`` peferred core`` has two states: 417 ``amd-pstate`` peferred core`` has two states: enable and disable.
418 Enable/disable states can be chosen by differe 418 Enable/disable states can be chosen by different kernel parameters.
419 Default enable ``amd-pstate`` preferred core. 419 Default enable ``amd-pstate`` preferred core.
420 420
421 ``amd_prefcore=disable`` 421 ``amd_prefcore=disable``
422 422
423 For systems that support ``amd-pstate`` prefer 423 For systems that support ``amd-pstate`` preferred core, the core rankings will
424 always be advertised by the platform. But OS c 424 always be advertised by the platform. But OS can choose to ignore that via the
425 kernel parameter ``amd_prefcore=disable``. 425 kernel parameter ``amd_prefcore=disable``.
426 426
427 User Space Interface in ``sysfs`` - General 427 User Space Interface in ``sysfs`` - General
428 =========================================== 428 ===========================================
429 429
430 Global Attributes 430 Global Attributes
431 ----------------- 431 -----------------
432 432
433 ``amd-pstate`` exposes several global attribut 433 ``amd-pstate`` exposes several global attributes (files) in ``sysfs`` to
434 control its functionality at the system level. 434 control its functionality at the system level. They are located in the
435 ``/sys/devices/system/cpu/amd_pstate/`` direct 435 ``/sys/devices/system/cpu/amd_pstate/`` directory and affect all CPUs.
436 436
437 ``status`` 437 ``status``
438 Operation mode of the driver: "active" 438 Operation mode of the driver: "active", "passive", "guided" or "disable".
439 439
440 "active" 440 "active"
441 The driver is functional and i 441 The driver is functional and in the ``active mode``
442 442
443 "passive" 443 "passive"
444 The driver is functional and i 444 The driver is functional and in the ``passive mode``
445 445
446 "guided" 446 "guided"
447 The driver is functional and i 447 The driver is functional and in the ``guided mode``
448 448
449 "disable" 449 "disable"
450 The driver is unregistered and 450 The driver is unregistered and not functional now.
451 451
452 This attribute can be written to in or 452 This attribute can be written to in order to change the driver's
453 operation mode or to unregister it. T 453 operation mode or to unregister it. The string written to it must be
454 one of the possible values of it and, 454 one of the possible values of it and, if successful, writing one of
455 these values to the sysfs file will ca 455 these values to the sysfs file will cause the driver to switch over
456 to the operation mode represented by t 456 to the operation mode represented by that string - or to be
457 unregistered in the "disable" case. 457 unregistered in the "disable" case.
458 458
459 ``prefcore`` 459 ``prefcore``
460 Preferred core state of the driver: "e 460 Preferred core state of the driver: "enabled" or "disabled".
461 461
462 "enabled" 462 "enabled"
463 Enable the ``amd-pstate`` pref 463 Enable the ``amd-pstate`` preferred core.
464 464
465 "disabled" 465 "disabled"
466 Disable the ``amd-pstate`` pre 466 Disable the ``amd-pstate`` preferred core
467 467
468 468
469 This attribute is read-only to check t 469 This attribute is read-only to check the state of preferred core set
470 by the kernel parameter. 470 by the kernel parameter.
471 471
472 ``cpupower`` tool support for ``amd-pstate`` 472 ``cpupower`` tool support for ``amd-pstate``
473 ============================================== 473 ===============================================
474 474
475 ``amd-pstate`` is supported by the ``cpupower` 475 ``amd-pstate`` is supported by the ``cpupower`` tool, which can be used to dump
476 frequency information. Development is in progr 476 frequency information. Development is in progress to support more and more
477 operations for the new ``amd-pstate`` module w 477 operations for the new ``amd-pstate`` module with this tool. ::
478 478
479 root@hr-test1:/home/ray# cpupower frequency-i 479 root@hr-test1:/home/ray# cpupower frequency-info
480 analyzing CPU 0: 480 analyzing CPU 0:
481 driver: amd-pstate 481 driver: amd-pstate
482 CPUs which run at the same hardware frequen 482 CPUs which run at the same hardware frequency: 0
483 CPUs which need to have their frequency coo 483 CPUs which need to have their frequency coordinated by software: 0
484 maximum transition latency: 131 us 484 maximum transition latency: 131 us
485 hardware limits: 400 MHz - 4.68 GHz 485 hardware limits: 400 MHz - 4.68 GHz
486 available cpufreq governors: ondemand conse 486 available cpufreq governors: ondemand conservative powersave userspace performance schedutil
487 current policy: frequency should be within 487 current policy: frequency should be within 400 MHz and 4.68 GHz.
488 The governor "schedutil" ma 488 The governor "schedutil" may decide which speed to use
489 within this range. 489 within this range.
490 current CPU frequency: Unable to call hardw 490 current CPU frequency: Unable to call hardware
491 current CPU frequency: 4.02 GHz (asserted b 491 current CPU frequency: 4.02 GHz (asserted by call to kernel)
492 boost state support: 492 boost state support:
493 Supported: yes 493 Supported: yes
494 Active: yes 494 Active: yes
495 AMD PSTATE Highest Performance: 166. Maxi 495 AMD PSTATE Highest Performance: 166. Maximum Frequency: 4.68 GHz.
496 AMD PSTATE Nominal Performance: 117. Nomi 496 AMD PSTATE Nominal Performance: 117. Nominal Frequency: 3.30 GHz.
497 AMD PSTATE Lowest Non-linear Performance: 497 AMD PSTATE Lowest Non-linear Performance: 39. Lowest Non-linear Frequency: 1.10 GHz.
498 AMD PSTATE Lowest Performance: 15. Lowest 498 AMD PSTATE Lowest Performance: 15. Lowest Frequency: 400 MHz.
499 499
500 500
501 Diagnostics and Tuning 501 Diagnostics and Tuning
502 ======================= 502 =======================
503 503
504 Trace Events 504 Trace Events
505 -------------- 505 --------------
506 506
507 There are two static trace events that can be 507 There are two static trace events that can be used for ``amd-pstate``
508 diagnostics. One of them is the ``cpu_frequenc 508 diagnostics. One of them is the ``cpu_frequency`` trace event generally used
509 by ``CPUFreq``, and the other one is the ``amd 509 by ``CPUFreq``, and the other one is the ``amd_pstate_perf`` trace event
510 specific to ``amd-pstate``. The following seq 510 specific to ``amd-pstate``. The following sequence of shell commands can
511 be used to enable them and see their output (i 511 be used to enable them and see their output (if the kernel is
512 configured to support event tracing). :: 512 configured to support event tracing). ::
513 513
514 root@hr-test1:/home/ray# cd /sys/kernel/traci 514 root@hr-test1:/home/ray# cd /sys/kernel/tracing/
515 root@hr-test1:/sys/kernel/tracing# echo 1 > e 515 root@hr-test1:/sys/kernel/tracing# echo 1 > events/amd_cpu/enable
516 root@hr-test1:/sys/kernel/tracing# cat trace 516 root@hr-test1:/sys/kernel/tracing# cat trace
517 # tracer: nop 517 # tracer: nop
518 # 518 #
519 # entries-in-buffer/entries-written: 47827/42 519 # entries-in-buffer/entries-written: 47827/42233061 #P:2
520 # 520 #
521 # _-----=> irq 521 # _-----=> irqs-off
522 # / _----=> nee 522 # / _----=> need-resched
523 # | / _---=> har 523 # | / _---=> hardirq/softirq
524 # || / _--=> pre 524 # || / _--=> preempt-depth
525 # ||| / dela 525 # ||| / delay
526 # TASK-PID CPU# |||| TIMESTA 526 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
527 # | | | |||| | 527 # | | | |||| | |
528 <idle>-0 [015] dN... 4995.979 528 <idle>-0 [015] dN... 4995.979886: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=15 changed=false fast_switch=true
529 <idle>-0 [007] d.h.. 4995.979 529 <idle>-0 [007] d.h.. 4995.979893: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=7 changed=false fast_switch=true
530 cat-2161 [000] d.... 4995.980 530 cat-2161 [000] d.... 4995.980841: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=0 changed=false fast_switch=true
531 sshd-2125 [004] d.s.. 4995.980 531 sshd-2125 [004] d.s.. 4995.980968: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=4 changed=false fast_switch=true
532 <idle>-0 [007] d.s.. 4995.980 532 <idle>-0 [007] d.s.. 4995.980968: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=7 changed=false fast_switch=true
533 <idle>-0 [003] d.s.. 4995.980 533 <idle>-0 [003] d.s.. 4995.980971: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=3 changed=false fast_switch=true
534 <idle>-0 [011] d.s.. 4995.980 534 <idle>-0 [011] d.s.. 4995.980996: amd_pstate_perf: amd_min_perf=85 amd_des_perf=85 amd_max_perf=166 cpu_id=11 changed=false fast_switch=true
535 535
536 The ``cpu_frequency`` trace event will be trig 536 The ``cpu_frequency`` trace event will be triggered either by the ``schedutil`` scaling
537 governor (for the policies it is attached to), 537 governor (for the policies it is attached to), or by the ``CPUFreq`` core (for the
538 policies with other scaling governors). 538 policies with other scaling governors).
539 539
540 540
541 Tracer Tool 541 Tracer Tool
542 ------------- 542 -------------
543 543
544 ``amd_pstate_tracer.py`` can record and parse 544 ``amd_pstate_tracer.py`` can record and parse ``amd-pstate`` trace log, then
545 generate performance plots. This utility can b 545 generate performance plots. This utility can be used to debug and tune the
546 performance of ``amd-pstate`` driver. The trac 546 performance of ``amd-pstate`` driver. The tracer tool needs to import intel
547 pstate tracer. 547 pstate tracer.
548 548
549 Tracer tool located in ``linux/tools/power/x86 549 Tracer tool located in ``linux/tools/power/x86/amd_pstate_tracer``. It can be
550 used in two ways. If trace file is available, 550 used in two ways. If trace file is available, then directly parse the file
551 with command :: 551 with command ::
552 552
553 ./amd_pstate_trace.py [-c cpus] -t <trace_fil 553 ./amd_pstate_trace.py [-c cpus] -t <trace_file> -n <test_name>
554 554
555 Or generate trace file with root privilege, th 555 Or generate trace file with root privilege, then parse and plot with command ::
556 556
557 sudo ./amd_pstate_trace.py [-c cpus] -n <test 557 sudo ./amd_pstate_trace.py [-c cpus] -n <test_name> -i <interval> [-m kbytes]
558 558
559 The test result can be found in ``results/test 559 The test result can be found in ``results/test_name``. Following is the example
560 about part of the output. :: 560 about part of the output. ::
561 561
562 common_cpu common_secs common_usecs min_pe 562 common_cpu common_secs common_usecs min_perf des_perf max_perf freq mperf apef tsc load duration_ms sample_num elapsed_time common_comm
563 CPU_005 712 116384 39 563 CPU_005 712 116384 39 49 166 0.7565 9645075 2214891 38431470 25.1 11.646 469 2.496 kworker/5:0-40
564 CPU_006 712 116408 39 564 CPU_006 712 116408 39 49 166 0.6769 8950227 1839034 37192089 24.06 11.272 470 2.496 kworker/6:0-1264
565 565
566 Unit Tests for amd-pstate 566 Unit Tests for amd-pstate
567 ------------------------- 567 -------------------------
568 568
569 ``amd-pstate-ut`` is a test module for testing 569 ``amd-pstate-ut`` is a test module for testing the ``amd-pstate`` driver.
570 570
571 * It can help all users to verify their proce 571 * It can help all users to verify their processor support (SBIOS/Firmware or Hardware).
572 572
573 * Kernel can have a basic function test to av 573 * Kernel can have a basic function test to avoid the kernel regression during the update.
574 574
575 * We can introduce more functional or perform 575 * We can introduce more functional or performance tests to align the result together, it will benefit power and performance scale optimization.
576 576
577 1. Test case descriptions 577 1. Test case descriptions
578 578
579 1). Basic tests 579 1). Basic tests
580 580
581 Test prerequisite and basic functions 581 Test prerequisite and basic functions for the ``amd-pstate`` driver.
582 582
583 +---------+--------------------------- 583 +---------+--------------------------------+------------------------------------------------------------------------------------+
584 | Index | Functions 584 | Index | Functions | Description |
585 +=========+=========================== 585 +=========+================================+====================================================================================+
586 | 1 | amd_pstate_ut_acpi_cpc_val 586 | 1 | amd_pstate_ut_acpi_cpc_valid || Check whether the _CPC object is present in SBIOS. |
587 | | 587 | | || |
588 | | 588 | | || The detail refer to `Processor Support <processor_support_>`_. |
589 +---------+--------------------------- 589 +---------+--------------------------------+------------------------------------------------------------------------------------+
590 | 2 | amd_pstate_ut_check_enable 590 | 2 | amd_pstate_ut_check_enabled || Check whether AMD P-State is enabled. |
591 | | 591 | | || |
592 | | 592 | | || AMD P-States and ACPI hardware P-States always can be supported in one processor. |
593 | | 593 | | | But AMD P-States has the higher priority and if it is enabled with |
594 | | 594 | | | :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_set_enable``, it will respond to the |
595 | | 595 | | | request from AMD P-States. |
596 +---------+--------------------------- 596 +---------+--------------------------------+------------------------------------------------------------------------------------+
597 | 3 | amd_pstate_ut_check_perf 597 | 3 | amd_pstate_ut_check_perf || Check if the each performance values are reasonable. |
598 | | 598 | | || highest_perf >= nominal_perf > lowest_nonlinear_perf > lowest_perf > 0. |
599 +---------+--------------------------- 599 +---------+--------------------------------+------------------------------------------------------------------------------------+
600 | 4 | amd_pstate_ut_check_freq 600 | 4 | amd_pstate_ut_check_freq || Check if the each frequency values and max freq when set support boost mode |
601 | | 601 | | | are reasonable. |
602 | | 602 | | || max_freq >= nominal_freq > lowest_nonlinear_freq > min_freq > 0 |
603 | | 603 | | || If boost is not active but supported, this maximum frequency will be larger than |
604 | | 604 | | | the one in ``cpuinfo``. |
605 +---------+--------------------------- 605 +---------+--------------------------------+------------------------------------------------------------------------------------+
606 606
607 2). Tbench test 607 2). Tbench test
608 608
609 Test and monitor the cpu changes when 609 Test and monitor the cpu changes when running tbench benchmark under the specified governor.
610 These changes include desire performan 610 These changes include desire performance, frequency, load, performance, energy etc.
611 The specified governor is ondemand or 611 The specified governor is ondemand or schedutil.
612 Tbench can also be tested on the ``acp 612 Tbench can also be tested on the ``acpi-cpufreq`` kernel driver for comparison.
613 613
614 3). Gitsource test 614 3). Gitsource test
615 615
616 Test and monitor the cpu changes when 616 Test and monitor the cpu changes when running gitsource benchmark under the specified governor.
617 These changes include desire performan 617 These changes include desire performance, frequency, load, time, energy etc.
618 The specified governor is ondemand or 618 The specified governor is ondemand or schedutil.
619 Gitsource can also be tested on the `` 619 Gitsource can also be tested on the ``acpi-cpufreq`` kernel driver for comparison.
620 620
621 #. How to execute the tests 621 #. How to execute the tests
622 622
623 We use test module in the kselftest framewo 623 We use test module in the kselftest frameworks to implement it.
624 We create ``amd-pstate-ut`` module and tie 624 We create ``amd-pstate-ut`` module and tie it into kselftest.(for
625 details refer to Linux Kernel Selftests [4] 625 details refer to Linux Kernel Selftests [4]_).
626 626
627 1). Build 627 1). Build
628 628
629 + open the :c:macro:`CONFIG_X86_AMD_PS 629 + open the :c:macro:`CONFIG_X86_AMD_PSTATE` configuration option.
630 + set the :c:macro:`CONFIG_X86_AMD_PST 630 + set the :c:macro:`CONFIG_X86_AMD_PSTATE_UT` configuration option to M.
631 + make project 631 + make project
632 + make selftest :: 632 + make selftest ::
633 633
634 $ cd linux 634 $ cd linux
635 $ make -C tools/testing/selftests 635 $ make -C tools/testing/selftests
636 636
637 + make perf :: 637 + make perf ::
638 638
639 $ cd tools/perf/ 639 $ cd tools/perf/
640 $ make 640 $ make
641 641
642 642
643 2). Installation & Steps :: 643 2). Installation & Steps ::
644 644
645 $ make -C tools/testing/selftests inst 645 $ make -C tools/testing/selftests install INSTALL_PATH=~/kselftest
646 $ cp tools/perf/perf /usr/bin/perf 646 $ cp tools/perf/perf /usr/bin/perf
647 $ sudo ./kselftest/run_kselftest.sh -c 647 $ sudo ./kselftest/run_kselftest.sh -c amd-pstate
648 648
649 3). Specified test case :: 649 3). Specified test case ::
650 650
651 $ cd ~/kselftest/amd-pstate 651 $ cd ~/kselftest/amd-pstate
652 $ sudo ./run.sh -t basic 652 $ sudo ./run.sh -t basic
653 $ sudo ./run.sh -t tbench 653 $ sudo ./run.sh -t tbench
654 $ sudo ./run.sh -t tbench -m acpi-cpuf 654 $ sudo ./run.sh -t tbench -m acpi-cpufreq
655 $ sudo ./run.sh -t gitsource 655 $ sudo ./run.sh -t gitsource
656 $ sudo ./run.sh -t gitsource -m acpi-c 656 $ sudo ./run.sh -t gitsource -m acpi-cpufreq
657 $ ./run.sh --help 657 $ ./run.sh --help
658 ./run.sh: illegal option -- - 658 ./run.sh: illegal option -- -
659 Usage: ./run.sh [OPTION...] 659 Usage: ./run.sh [OPTION...]
660 [-h <help>] 660 [-h <help>]
661 [-o <output-file-for-dump>] 661 [-o <output-file-for-dump>]
662 [-c <all: All testing, 662 [-c <all: All testing,
663 basic: Basic testing, 663 basic: Basic testing,
664 tbench: Tbench testing, 664 tbench: Tbench testing,
665 gitsource: Gitsource test 665 gitsource: Gitsource testing.>]
666 [-t <tbench time limit>] 666 [-t <tbench time limit>]
667 [-p <tbench process number>] 667 [-p <tbench process number>]
668 [-l <loop times for tbench>] 668 [-l <loop times for tbench>]
669 [-i <amd tracer interval>] 669 [-i <amd tracer interval>]
670 [-m <comparative test: acpi-cp 670 [-m <comparative test: acpi-cpufreq>]
671 671
672 672
673 4). Results 673 4). Results
674 674
675 + basic 675 + basic
676 676
677 When you finish test, you will get th 677 When you finish test, you will get the following log info ::
678 678
679 $ dmesg | grep "amd_pstate_ut" | tee 679 $ dmesg | grep "amd_pstate_ut" | tee log.txt
680 [12977.570663] amd_pstate_ut: 1 a 680 [12977.570663] amd_pstate_ut: 1 amd_pstate_ut_acpi_cpc_valid success!
681 [12977.570673] amd_pstate_ut: 2 a 681 [12977.570673] amd_pstate_ut: 2 amd_pstate_ut_check_enabled success!
682 [12977.571207] amd_pstate_ut: 3 a 682 [12977.571207] amd_pstate_ut: 3 amd_pstate_ut_check_perf success!
683 [12977.571212] amd_pstate_ut: 4 a 683 [12977.571212] amd_pstate_ut: 4 amd_pstate_ut_check_freq success!
684 684
685 + tbench 685 + tbench
686 686
687 When you finish test, you will get se 687 When you finish test, you will get selftest.tbench.csv and png images.
688 The selftest.tbench.csv file contains 688 The selftest.tbench.csv file contains the raw data and the drop of the comparative test.
689 The png images shows the performance, 689 The png images shows the performance, energy and performan per watt of each test.
690 Open selftest.tbench.csv : 690 Open selftest.tbench.csv :
691 691
692 +------------------------------------ 692 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
693 + Governor 693 + Governor | Round | Des-perf | Freq | Load | Performance | Energy | Performance Per Watt |
694 +------------------------------------ 694 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
695 + Unit 695 + Unit | | | GHz | | MB/s | J | MB/J |
696 +==================================== 696 +=================================================+==============+==========+=========+==========+=============+=========+======================+
697 + amd-pstate-ondemand 697 + amd-pstate-ondemand | 1 | | | | 2504.05 | 1563.67 | 158.5378 |
698 +------------------------------------ 698 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
699 + amd-pstate-ondemand 699 + amd-pstate-ondemand | 2 | | | | 2243.64 | 1430.32 | 155.2941 |
700 +------------------------------------ 700 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
701 + amd-pstate-ondemand 701 + amd-pstate-ondemand | 3 | | | | 2183.88 | 1401.32 | 154.2860 |
702 +------------------------------------ 702 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
703 + amd-pstate-ondemand 703 + amd-pstate-ondemand | Average | | | | 2310.52 | 1465.1 | 156.1268 |
704 +------------------------------------ 704 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
705 + amd-pstate-schedutil 705 + amd-pstate-schedutil | 1 | 165.329 | 1.62257 | 99.798 | 2136.54 | 1395.26 | 151.5971 |
706 +------------------------------------ 706 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
707 + amd-pstate-schedutil 707 + amd-pstate-schedutil | 2 | 166 | 1.49761 | 99.9993 | 2100.56 | 1380.5 | 150.6377 |
708 +------------------------------------ 708 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
709 + amd-pstate-schedutil 709 + amd-pstate-schedutil | 3 | 166 | 1.47806 | 99.9993 | 2084.12 | 1375.76 | 149.9737 |
710 +------------------------------------ 710 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
711 + amd-pstate-schedutil 711 + amd-pstate-schedutil | Average | 165.776 | 1.53275 | 99.9322 | 2107.07 | 1383.84 | 150.7399 |
712 +------------------------------------ 712 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
713 + acpi-cpufreq-ondemand 713 + acpi-cpufreq-ondemand | 1 | | | | 2529.9 | 1564.4 | 160.0997 |
714 +------------------------------------ 714 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
715 + acpi-cpufreq-ondemand 715 + acpi-cpufreq-ondemand | 2 | | | | 2249.76 | 1432.97 | 155.4297 |
716 +------------------------------------ 716 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
717 + acpi-cpufreq-ondemand 717 + acpi-cpufreq-ondemand | 3 | | | | 2181.46 | 1406.88 | 153.5060 |
718 +------------------------------------ 718 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
719 + acpi-cpufreq-ondemand 719 + acpi-cpufreq-ondemand | Average | | | | 2320.37 | 1468.08 | 156.4741 |
720 +------------------------------------ 720 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
721 + acpi-cpufreq-schedutil 721 + acpi-cpufreq-schedutil | 1 | | | | 2137.64 | 1385.24 | 152.7723 |
722 +------------------------------------ 722 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
723 + acpi-cpufreq-schedutil 723 + acpi-cpufreq-schedutil | 2 | | | | 2107.05 | 1372.23 | 152.0138 |
724 +------------------------------------ 724 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
725 + acpi-cpufreq-schedutil 725 + acpi-cpufreq-schedutil | 3 | | | | 2085.86 | 1365.35 | 151.2433 |
726 +------------------------------------ 726 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
727 + acpi-cpufreq-schedutil 727 + acpi-cpufreq-schedutil | Average | | | | 2110.18 | 1374.27 | 152.0136 |
728 +------------------------------------ 728 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
729 + acpi-cpufreq-ondemand VS acpi-cpufr 729 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | -9.0584 | -6.3899 | -2.8506 |
730 +------------------------------------ 730 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
731 + amd-pstate-ondemand VS amd-pstate-s 731 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | | | | 8.8053 | -5.5463 | -3.4503 |
732 +------------------------------------ 732 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
733 + acpi-cpufreq-ondemand VS amd-pstate 733 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | -0.4245 | -0.2029 | -0.2219 |
734 +------------------------------------ 734 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
735 + acpi-cpufreq-schedutil VS amd-pstat 735 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | -0.1473 | 0.6963 | -0.8378 |
736 +------------------------------------ 736 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+
737 737
738 + gitsource 738 + gitsource
739 739
740 When you finish test, you will get se 740 When you finish test, you will get selftest.gitsource.csv and png images.
741 The selftest.gitsource.csv file conta 741 The selftest.gitsource.csv file contains the raw data and the drop of the comparative test.
742 The png images shows the performance, 742 The png images shows the performance, energy and performan per watt of each test.
743 Open selftest.gitsource.csv : 743 Open selftest.gitsource.csv :
744 744
745 +------------------------------------ 745 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
746 + Governor 746 + Governor | Round | Des-perf | Freq | Load | Time | Energy | Performance Per Watt |
747 +------------------------------------ 747 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
748 + Unit 748 + Unit | | | GHz | | s | J | 1/J |
749 +==================================== 749 +=================================================+==============+==========+==========+==========+=============+=========+======================+
750 + amd-pstate-ondemand 750 + amd-pstate-ondemand | 1 | 50.119 | 2.10509 | 23.3076 | 475.69 | 865.78 | 0.001155027 |
751 +------------------------------------ 751 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
752 + amd-pstate-ondemand 752 + amd-pstate-ondemand | 2 | 94.8006 | 1.98771 | 56.6533 | 467.1 | 839.67 | 0.001190944 |
753 +------------------------------------ 753 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
754 + amd-pstate-ondemand 754 + amd-pstate-ondemand | 3 | 76.6091 | 2.53251 | 43.7791 | 467.69 | 855.85 | 0.001168429 |
755 +------------------------------------ 755 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
756 + amd-pstate-ondemand 756 + amd-pstate-ondemand | Average | 73.8429 | 2.20844 | 41.2467 | 470.16 | 853.767 | 0.001171279 |
757 +------------------------------------ 757 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
758 + amd-pstate-schedutil 758 + amd-pstate-schedutil | 1 | 165.919 | 1.62319 | 98.3868 | 464.17 | 866.8 | 0.001153668 |
759 +------------------------------------ 759 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
760 + amd-pstate-schedutil 760 + amd-pstate-schedutil | 2 | 165.97 | 1.31309 | 99.5712 | 480.15 | 880.4 | 0.001135847 |
761 +------------------------------------ 761 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
762 + amd-pstate-schedutil 762 + amd-pstate-schedutil | 3 | 165.973 | 1.28448 | 99.9252 | 481.79 | 867.02 | 0.001153375 |
763 +------------------------------------ 763 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
764 + amd-pstate-schedutil 764 + amd-pstate-schedutil | Average | 165.954 | 1.40692 | 99.2944 | 475.37 | 871.407 | 0.001147569 |
765 +------------------------------------ 765 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
766 + acpi-cpufreq-ondemand 766 + acpi-cpufreq-ondemand | 1 | | | | 2379.62 | 742.96 | 0.001345967 |
767 +------------------------------------ 767 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
768 + acpi-cpufreq-ondemand 768 + acpi-cpufreq-ondemand | 2 | | | | 441.74 | 817.49 | 0.001223256 |
769 +------------------------------------ 769 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
770 + acpi-cpufreq-ondemand 770 + acpi-cpufreq-ondemand | 3 | | | | 455.48 | 820.01 | 0.001219497 |
771 +------------------------------------ 771 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
772 + acpi-cpufreq-ondemand 772 + acpi-cpufreq-ondemand | Average | | | | 425.613 | 793.487 | 0.001260260 |
773 +------------------------------------ 773 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
774 + acpi-cpufreq-schedutil 774 + acpi-cpufreq-schedutil | 1 | | | | 459.69 | 838.54 | 0.001192548 |
775 +------------------------------------ 775 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
776 + acpi-cpufreq-schedutil 776 + acpi-cpufreq-schedutil | 2 | | | | 466.55 | 830.89 | 0.001203528 |
777 +------------------------------------ 777 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
778 + acpi-cpufreq-schedutil 778 + acpi-cpufreq-schedutil | 3 | | | | 470.38 | 837.32 | 0.001194286 |
779 +------------------------------------ 779 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
780 + acpi-cpufreq-schedutil 780 + acpi-cpufreq-schedutil | Average | | | | 465.54 | 835.583 | 0.001196769 |
781 +------------------------------------ 781 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
782 + acpi-cpufreq-ondemand VS acpi-cpufr 782 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | 9.3810 | 5.3051 | -5.0379 |
783 +------------------------------------ 783 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
784 + amd-pstate-ondemand VS amd-pstate-s 784 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | 124.7392 | -36.2934 | 140.7329 | 1.1081 | 2.0661 | -2.0242 |
785 +------------------------------------ 785 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
786 + acpi-cpufreq-ondemand VS amd-pstate 786 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | 10.4665 | 7.5968 | -7.0605 |
787 +------------------------------------ 787 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
788 + acpi-cpufreq-schedutil VS amd-pstat 788 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | 2.1115 | 4.2873 | -4.1110 |
789 +------------------------------------ 789 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+
790 790
791 Reference 791 Reference
792 =========== 792 ===========
793 793
794 .. [1] AMD64 Architecture Programmer's Manual 794 .. [1] AMD64 Architecture Programmer's Manual Volume 2: System Programming,
795 https://www.amd.com/system/files/TechDo 795 https://www.amd.com/system/files/TechDocs/24593.pdf
796 796
797 .. [2] Advanced Configuration and Power Interf 797 .. [2] Advanced Configuration and Power Interface Specification,
798 https://uefi.org/sites/default/files/re 798 https://uefi.org/sites/default/files/resources/ACPI_Spec_6_4_Jan22.pdf
799 799
800 .. [3] Processor Programming Reference (PPR) f 800 .. [3] Processor Programming Reference (PPR) for AMD Family 19h Model 51h, Revision A1 Processors
801 https://www.amd.com/system/files/TechDo 801 https://www.amd.com/system/files/TechDocs/56569-A1-PUB.zip
802 802
803 .. [4] Linux Kernel Selftests, 803 .. [4] Linux Kernel Selftests,
804 https://www.kernel.org/doc/html/latest/ 804 https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.