1.. SPDX-License-Identifier: GPL-2.0 2.. include:: <isonum.txt> 3 4=============================================== 5``amd-pstate`` CPU Performance Scaling Driver 6=============================================== 7 8:Copyright: |copy| 2021 Advanced Micro Devices, Inc. 9 10:Author: Huang Rui <ray.huang@amd.com> 11 12 13Introduction 14=================== 15 16``amd-pstate`` is the AMD CPU performance scaling driver that introduces a 17new CPU frequency control mechanism on modern AMD APU and CPU series in 18Linux kernel. The new mechanism is based on Collaborative Processor 19Performance Control (CPPC) which provides finer grain frequency management 20than legacy ACPI hardware P-States. Current AMD CPU/APU platforms are using 21the ACPI P-states driver to manage CPU frequency and clocks with switching 22only in 3 P-states. CPPC replaces the ACPI P-states controls and allows a 23flexible, low-latency interface for the Linux kernel to directly 24communicate the performance hints to hardware. 25 26``amd-pstate`` leverages the Linux kernel governors such as ``schedutil``, 27``ondemand``, etc. to manage the performance hints which are provided by 28CPPC hardware functionality that internally follows the hardware 29specification (for details refer to AMD64 Architecture Programmer's Manual 30Volume 2: System Programming [1]_). Currently, ``amd-pstate`` supports basic 31frequency control function according to kernel governors on some of the 32Zen2 and Zen3 processors, and we will implement more AMD specific functions 33in future after we verify them on the hardware and SBIOS. 34 35 36AMD CPPC Overview 37======================= 38 39Collaborative Processor Performance Control (CPPC) interface enumerates a 40continuous, abstract, and unit-less performance value in a scale that is 41not tied to a specific performance state / frequency. This is an ACPI 42standard [2]_ which software can specify application performance goals and 43hints as a relative target to the infrastructure limits. AMD processors 44provide the low latency register model (MSR) instead of an AML code 45interpreter for performance adjustments. ``amd-pstate`` will initialize a 46``struct cpufreq_driver`` instance, ``amd_pstate_driver``, with the callbacks 47to manage each performance update behavior. :: 48 49 Highest Perf ------>+-----------------------+ +-----------------------+ 50 | | | | 51 | | | | 52 | | Max Perf ---->| | 53 | | | | 54 | | | | 55 Nominal Perf ------>+-----------------------+ +-----------------------+ 56 | | | | 57 | | | | 58 | | | | 59 | | | | 60 | | | | 61 | | | | 62 | | Desired Perf ---->| | 63 | | | | 64 | | | | 65 | | | | 66 | | | | 67 | | | | 68 | | | | 69 | | | | 70 | | | | 71 | | | | 72 Lowest non- | | | | 73 linear perf ------>+-----------------------+ +-----------------------+ 74 | | | | 75 | | Min perf ---->| | 76 | | | | 77 Lowest perf ------>+-----------------------+ +-----------------------+ 78 | | | | 79 | | | | 80 | | | | 81 0 ------>+-----------------------+ +-----------------------+ 82 83 AMD P-States Performance Scale 84 85 86.. _perf_cap: 87 88AMD CPPC Performance Capability 89-------------------------------- 90 91Highest Performance (RO) 92......................... 93 94This is the absolute maximum performance an individual processor may reach, 95assuming ideal conditions. This performance level may not be sustainable 96for long durations and may only be achievable if other platform components 97are in a specific state; for example, it may require other processors to be in 98an idle state. This would be equivalent to the highest frequencies 99supported by the processor. 100 101Nominal (Guaranteed) Performance (RO) 102...................................... 103 104This is the maximum sustained performance level of the processor, assuming 105ideal operating conditions. In the absence of an external constraint (power, 106thermal, etc.), this is the performance level the processor is expected to 107be able to maintain continuously. All cores/processors are expected to be 108able to sustain their nominal performance state simultaneously. 109 110Lowest non-linear Performance (RO) 111................................... 112 113This is the lowest performance level at which nonlinear power savings are 114achieved, for example, due to the combined effects of voltage and frequency 115scaling. Above this threshold, lower performance levels should be generally 116more energy efficient than higher performance levels. This register 117effectively conveys the most efficient performance level to ``amd-pstate``. 118 119Lowest Performance (RO) 120........................ 121 122This is the absolute lowest performance level of the processor. Selecting a 123performance level lower than the lowest nonlinear performance level may 124cause an efficiency penalty but should reduce the instantaneous power 125consumption of the processor. 126 127AMD CPPC Performance Control 128------------------------------ 129 130``amd-pstate`` passes performance goals through these registers. The 131register drives the behavior of the desired performance target. 132 133Minimum requested performance (RW) 134................................... 135 136``amd-pstate`` specifies the minimum allowed performance level. 137 138Maximum requested performance (RW) 139................................... 140 141``amd-pstate`` specifies a limit the maximum performance that is expected 142to be supplied by the hardware. 143 144Desired performance target (RW) 145................................... 146 147``amd-pstate`` specifies a desired target in the CPPC performance scale as 148a relative number. This can be expressed as percentage of nominal 149performance (infrastructure max). Below the nominal sustained performance 150level, desired performance expresses the average performance level of the 151processor subject to hardware. Above the nominal performance level, 152the processor must provide at least nominal performance requested and go higher 153if current operating conditions allow. 154 155Energy Performance Preference (EPP) (RW) 156......................................... 157 158This attribute provides a hint to the hardware if software wants to bias 159toward performance (0x0) or energy efficiency (0xff). 160 161 162Key Governors Support 163======================= 164 165``amd-pstate`` can be used with all the (generic) scaling governors listed 166by the ``scaling_available_governors`` policy attribute in ``sysfs``. Then, 167it is responsible for the configuration of policy objects corresponding to 168CPUs and provides the ``CPUFreq`` core (and the scaling governors attached 169to the policy objects) with accurate information on the maximum and minimum 170operating frequencies supported by the hardware. Users can check the 171``scaling_cur_freq`` information comes from the ``CPUFreq`` core. 172 173``amd-pstate`` mainly supports ``schedutil`` and ``ondemand`` for dynamic 174frequency control. It is to fine tune the processor configuration on 175``amd-pstate`` to the ``schedutil`` with CPU CFS scheduler. ``amd-pstate`` 176registers the adjust_perf callback to implement performance update behavior 177similar to CPPC. It is initialized by ``sugov_start`` and then populates the 178CPU's update_util_data pointer to assign ``sugov_update_single_perf`` as the 179utilization update callback function in the CPU scheduler. The CPU scheduler 180will call ``cpufreq_update_util`` and assigns the target performance according 181to the ``struct sugov_cpu`` that the utilization update belongs to. 182Then, ``amd-pstate`` updates the desired performance according to the CPU 183scheduler assigned. 184 185.. _processor_support: 186 187Processor Support 188======================= 189 190The ``amd-pstate`` initialization will fail if the ``_CPC`` entry in the ACPI 191SBIOS does not exist in the detected processor. It uses ``acpi_cpc_valid`` 192to check the existence of ``_CPC``. All Zen based processors support the legacy 193ACPI hardware P-States function, so when ``amd-pstate`` fails initialization, 194the kernel will fall back to initialize the ``acpi-cpufreq`` driver. 195 196There are two types of hardware implementations for ``amd-pstate``: one is 197`Full MSR Support <perf_cap_>`_ and another is `Shared Memory Support 198<perf_cap_>`_. It can use the :c:macro:`X86_FEATURE_CPPC` feature flag to 199indicate the different types. (For details, refer to the Processor Programming 200Reference (PPR) for AMD Family 19h Model 51h, Revision A1 Processors [3]_.) 201``amd-pstate`` is to register different ``static_call`` instances for different 202hardware implementations. 203 204Currently, some of the Zen2 and Zen3 processors support ``amd-pstate``. In the 205future, it will be supported on more and more AMD processors. 206 207Full MSR Support 208----------------- 209 210Some new Zen3 processors such as Cezanne provide the MSR registers directly 211while the :c:macro:`X86_FEATURE_CPPC` CPU feature flag is set. 212``amd-pstate`` can handle the MSR register to implement the fast switch 213function in ``CPUFreq`` that can reduce the latency of frequency control in 214interrupt context. The functions with a ``pstate_xxx`` prefix represent the 215operations on MSR registers. 216 217Shared Memory Support 218---------------------- 219 220If the :c:macro:`X86_FEATURE_CPPC` CPU feature flag is not set, the 221processor supports the shared memory solution. In this case, ``amd-pstate`` 222uses the ``cppc_acpi`` helper methods to implement the callback functions 223that are defined on ``static_call``. The functions with the ``cppc_xxx`` prefix 224represent the operations of ACPI CPPC helpers for the shared memory solution. 225 226 227AMD P-States and ACPI hardware P-States always can be supported in one 228processor. But AMD P-States has the higher priority and if it is enabled 229with :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_set_enable``, it will respond 230to the request from AMD P-States. 231 232 233User Space Interface in ``sysfs`` - Per-policy control 234====================================================== 235 236``amd-pstate`` exposes several global attributes (files) in ``sysfs`` to 237control its functionality at the system level. They are located in the 238``/sys/devices/system/cpu/cpufreq/policyX/`` directory and affect all CPUs. :: 239 240 root@hr-test1:/home/ray# ls /sys/devices/system/cpu/cpufreq/policy0/*amd* 241 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_highest_perf 242 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_hw_prefcore 243 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_lowest_nonlinear_freq 244 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_max_freq 245 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_floor_freq 246 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_floor_count 247 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_prefcore_ranking 248 249 250``amd_pstate_highest_perf / amd_pstate_max_freq`` 251 252Maximum CPPC performance and CPU frequency that the driver is allowed to 253set, in percent of the maximum supported CPPC performance level (the highest 254performance supported in `AMD CPPC Performance Capability <perf_cap_>`_). 255In some ASICs, the highest CPPC performance is not the one in the ``_CPC`` 256table, so we need to expose it to sysfs. If boost is not active, but 257still supported, this maximum frequency will be larger than the one in 258``cpuinfo``. 259This attribute is read-only. 260 261``amd_pstate_lowest_nonlinear_freq`` 262 263The lowest non-linear CPPC CPU frequency that the driver is allowed to set, 264in percent of the maximum supported CPPC performance level. (Please see the 265lowest non-linear performance in `AMD CPPC Performance Capability 266<perf_cap_>`_.) 267This attribute is read-only. 268 269``amd_pstate_hw_prefcore`` 270 271Whether the platform supports the preferred core feature and it has 272been enabled. This attribute is read-only. This file is only visible 273on platforms which support the preferred core feature. 274 275``amd_pstate_prefcore_ranking`` 276 277The performance ranking of the core. This number doesn't have any unit, but 278larger numbers are preferred at the time of reading. This can change at 279runtime based on platform conditions. This attribute is read-only. This file 280is only visible on platforms which support the preferred core feature. 281 282``amd_pstate_floor_freq`` 283 284The floor frequency associated with each CPU. Userspace can write any 285value between ``cpuinfo_min_freq`` and ``scaling_max_freq`` into this 286file. When the system is under power or thermal constraints, the 287platform firmware will attempt to throttle the CPU frequency to the 288value specified in ``amd_pstate_floor_freq`` before throttling it 289further. This allows userspace to specify different floor frequencies 290to different CPUs. For optimal results, threads of the same core 291should have the same floor frequency value. This file is only visible 292on platforms that support the CPPC Performance Priority feature. 293 294 295``amd_pstate_floor_count`` 296 297The number of distinct Floor Performance levels supported by the 298platform. For example, if this value is 2, then the number of unique 299values obtained from the command ``cat 300/sys/devices/system/cpu/cpufreq/policy*/amd_pstate_floor_freq | 301sort -n | uniq`` should be at most this number for the behavior 302described in ``amd_pstate_floor_freq`` to take effect. A zero value 303implies that the platform supports unlimited floor performance levels. 304This file is only visible on platforms that support the CPPC 305Performance Priority feature. 306 307**Note**: When ``amd_pstate_floor_count`` is non-zero, the frequency to 308which the CPU is throttled under power or thermal constraints is 309undefined when the number of unique values of ``amd_pstate_floor_freq`` 310across all CPUs in the system exceeds ``amd_pstate_floor_count``. 311 312``energy_performance_available_preferences`` 313 314A list of all the supported EPP preferences that could be used for 315``energy_performance_preference`` on this system. 316These profiles represent different hints that are provided 317to the low-level firmware about the user's desired energy vs efficiency 318tradeoff. ``default`` represents the epp value is set by platform 319firmware. ``custom`` designates that integer values 0-255 may be written 320as well. ``dynamic`` designates that the EPP is modified dynamically 321on kernel events. See ``Dynamic energy performance profile`` section below to 322know more about the ``dynamic`` mode. This attribute is read-only. 323 324``energy_performance_preference`` 325 326The current energy performance preference can be read from this attribute. 327and user can change current preference according to energy or performance needs 328Coarse named profiles are available in the attribute 329``energy_performance_available_preferences``. 330Users can also write individual integer values between 0 to 255. 331Lower epp values shift the bias towards improved performance while a higher epp 332value shifts the bias towards power-savings. The exact impact can change from 333one platform to the other. If a valid integer was last written, then a number 334will be returned on future reads. If a valid string was last written then a 335string will be returned on future reads. This attribute is read-write. 336 337``boost`` 338The `boost` sysfs attribute provides control over the CPU core 339performance boost, allowing users to manage the maximum frequency limitation 340of the CPU. This attribute can be used to enable or disable the boost feature 341on individual CPUs. 342 343When the boost feature is enabled, the CPU can dynamically increase its frequency 344beyond the base frequency, providing enhanced performance for demanding workloads. 345On the other hand, disabling the boost feature restricts the CPU to operate at the 346base frequency, which may be desirable in certain scenarios to prioritize power 347efficiency or manage temperature. 348 349To manipulate the `boost` attribute, users can write a value of `0` to disable the 350boost or `1` to enable it, for the respective CPU using the sysfs path 351`/sys/devices/system/cpu/cpuX/cpufreq/boost`, where `X` represents the CPU number. 352 353Other performance and frequency values can be read back from 354``/sys/devices/system/cpu/cpuX/acpi_cppc/``, see :ref:`cppc_sysfs`. 355 356Dynamic energy performance profile 357================================== 358The amd-pstate driver supports dynamically selecting the energy performance 359profile based on the system profile and the current power source in active mode. 360 361The ``dynamic`` mode is listed in 362``/sys/devices/system/cpu/cpuX/cpufreq/energy_performance_available_preferences`` 363when available while running under the ``powersave`` governor. The ``dynamic`` 364mode can be toggled on by writing the same to the sysfs file 365``/sys/devices/system/cpu/cpuX/cpufreq/energy_performance_preference`` when 366available. 367 368When ``energy_performance_preference`` is set to ``dynamic``, the driver will 369select a different energy performance profile when the machine is running on 370battery or AC power. The driver will also register with the platform profile 371handler to receive notifications of user desired power state and react to 372those. 373 374``amd-pstate`` vs ``acpi-cpufreq`` 375====================================== 376 377On the majority of AMD platforms supported by ``acpi-cpufreq``, the ACPI tables 378provided by the platform firmware are used for CPU performance scaling, but 379only provide 3 P-states on AMD processors. 380However, on modern AMD APU and CPU series, hardware provides the Collaborative 381Processor Performance Control according to the ACPI protocol and customizes this 382for AMD platforms. That is, fine-grained and continuous frequency ranges 383instead of the legacy hardware P-states. ``amd-pstate`` is the kernel 384module which supports the new AMD P-States mechanism on most of the future AMD 385platforms. The AMD P-States mechanism is the more performance and energy 386efficiency frequency management method on AMD processors. 387 388 389``amd-pstate`` Driver Operation Modes 390====================================== 391 392``amd_pstate`` CPPC has 3 operation modes: autonomous (active) mode, 393non-autonomous (passive) mode and guided autonomous (guided) mode. 394Active/passive/guided mode can be chosen by different kernel parameters. 395 396- In autonomous mode, platform ignores the desired performance level request 397 and takes into account only the values set to the minimum, maximum and energy 398 performance preference registers. 399- In non-autonomous mode, platform gets desired performance level 400 from OS directly through Desired Performance Register. 401- In guided-autonomous mode, platform sets operating performance level 402 autonomously according to the current workload and within the limits set by 403 OS through min and max performance registers. 404 405Active Mode 406------------ 407 408``amd_pstate=active`` 409 410This is the low-level firmware control mode which is implemented by ``amd_pstate_epp`` 411driver with ``amd_pstate=active`` passed to the kernel in the command line. 412In this mode, ``amd_pstate_epp`` driver provides a hint to the hardware if software 413wants to bias toward performance (0x0) or energy efficiency (0xff) to the CPPC firmware. 414then CPPC power algorithm will calculate the runtime workload and adjust the realtime 415cores frequency according to the power supply and thermal, core voltage and some other 416hardware conditions. 417 418Passive Mode 419------------ 420 421``amd_pstate=passive`` 422 423It will be enabled if the ``amd_pstate=passive`` is passed to the kernel in the command line. 424In this mode, ``amd_pstate`` driver software specifies a desired QoS target in the CPPC 425performance scale as a relative number. This can be expressed as percentage of nominal 426performance (infrastructure max). Below the nominal sustained performance level, 427desired performance expresses the average performance level of the processor subject 428to the Performance Reduction Tolerance register. Above the nominal performance level, 429processor must provide at least nominal performance requested and go higher if current 430operating conditions allow. 431 432Guided Mode 433----------- 434 435``amd_pstate=guided`` 436 437If ``amd_pstate=guided`` is passed to kernel command line option then this mode 438is activated. In this mode, driver requests minimum and maximum performance 439level and the platform autonomously selects a performance level in this range 440and appropriate to the current workload. 441 442``amd-pstate`` Preferred Core 443================================= 444 445The core frequency is subjected to the process variation in semiconductors. 446Not all cores are able to reach the maximum frequency respecting the 447infrastructure limits. Consequently, AMD has redefined the concept of 448maximum frequency of a part. This means that a fraction of cores can reach 449maximum frequency. To find the best process scheduling policy for a given 450scenario, OS needs to know the core ordering informed by the platform through 451highest performance capability register of the CPPC interface. 452 453``amd-pstate`` preferred core enables the scheduler to prefer scheduling on 454cores that can achieve a higher frequency with lower voltage. The preferred 455core rankings can dynamically change based on the workload, platform conditions, 456thermals and ageing. 457 458The priority metric will be initialized by the ``amd-pstate`` driver. The ``amd-pstate`` 459driver will also determine whether or not ``amd-pstate`` preferred core is 460supported by the platform. 461 462``amd-pstate`` driver will provide an initial core ordering when the system boots. 463The platform uses the CPPC interfaces to communicate the core ranking to the 464operating system and scheduler to make sure that OS is choosing the cores 465with highest performance firstly for scheduling the process. When ``amd-pstate`` 466driver receives a message with the highest performance change, it will 467update the core ranking and set the cpu's priority. 468 469``amd-pstate`` Preferred Core Switch 470===================================== 471Kernel Parameters 472----------------- 473 474``amd-pstate`` preferred core has two states: enable and disable. 475Enable/disable states can be chosen by different kernel parameters. 476Default enable ``amd-pstate`` preferred core. 477 478``amd_prefcore=disable`` 479 480For systems that support ``amd-pstate`` preferred core, the core rankings will 481always be advertised by the platform. But OS can choose to ignore that via the 482kernel parameter ``amd_prefcore=disable``. 483 484``amd_dynamic_epp`` 485 486When AMD pstate is in auto mode, dynamic EPP will control whether the kernel 487autonomously changes the EPP mode. The default is disabled. It can be enabled 488with the kernel parameter ``amd_dynamic_epp=enable``. 489 490User Space Interface in ``sysfs`` - General 491=========================================== 492 493Global Attributes 494----------------- 495 496``amd-pstate`` exposes several global attributes (files) in ``sysfs`` to 497control its functionality at the system level. They are located in the 498``/sys/devices/system/cpu/amd_pstate/`` directory and affect all CPUs. 499 500``status`` 501 Operation mode of the driver: "active", "passive", "guided" or "disable". 502 503 "active" 504 The driver is functional and in the ``active mode`` 505 506 "passive" 507 The driver is functional and in the ``passive mode`` 508 509 "guided" 510 The driver is functional and in the ``guided mode`` 511 512 "disable" 513 The driver is unregistered and not functional now. 514 515 This attribute can be written to in order to change the driver's 516 operation mode or to unregister it. The string written to it must be 517 one of the possible values of it and, if successful, writing one of 518 these values to the sysfs file will cause the driver to switch over 519 to the operation mode represented by that string - or to be 520 unregistered in the "disable" case. 521 522``prefcore`` 523 Preferred core state of the driver: "enabled" or "disabled". 524 525 "enabled" 526 Enable the ``amd-pstate`` preferred core. 527 528 "disabled" 529 Disable the ``amd-pstate`` preferred core 530 531 532 This attribute is read-only to check the state of preferred core set 533 by the kernel parameter. 534 535``cpupower`` tool support for ``amd-pstate`` 536=============================================== 537 538``amd-pstate`` is supported by the ``cpupower`` tool, which can be used to dump 539frequency information. Development is in progress to support more and more 540operations for the new ``amd-pstate`` module with this tool. :: 541 542 root@hr-test1:/home/ray# cpupower frequency-info 543 analyzing CPU 0: 544 driver: amd-pstate 545 CPUs which run at the same hardware frequency: 0 546 CPUs which need to have their frequency coordinated by software: 0 547 maximum transition latency: 131 us 548 hardware limits: 400 MHz - 4.68 GHz 549 available cpufreq governors: ondemand conservative powersave userspace performance schedutil 550 current policy: frequency should be within 400 MHz and 4.68 GHz. 551 The governor "schedutil" may decide which speed to use 552 within this range. 553 current CPU frequency: Unable to call hardware 554 current CPU frequency: 4.02 GHz (asserted by call to kernel) 555 boost state support: 556 Supported: yes 557 Active: yes 558 AMD PSTATE Highest Performance: 166. Maximum Frequency: 4.68 GHz. 559 AMD PSTATE Nominal Performance: 117. Nominal Frequency: 3.30 GHz. 560 AMD PSTATE Lowest Non-linear Performance: 39. Lowest Non-linear Frequency: 1.10 GHz. 561 AMD PSTATE Lowest Performance: 15. Lowest Frequency: 400 MHz. 562 563 564Diagnostics and Tuning 565======================= 566 567Trace Events 568-------------- 569 570There are two static trace events that can be used for ``amd-pstate`` 571diagnostics. One of them is the ``cpu_frequency`` trace event generally used 572by ``CPUFreq``, and the other one is the ``amd_pstate_perf`` trace event 573specific to ``amd-pstate``. The following sequence of shell commands can 574be used to enable them and see their output (if the kernel is 575configured to support event tracing). :: 576 577 root@hr-test1:/home/ray# cd /sys/kernel/tracing/ 578 root@hr-test1:/sys/kernel/tracing# echo 1 > events/amd_cpu/enable 579 root@hr-test1:/sys/kernel/tracing# cat trace 580 # tracer: nop 581 # 582 # entries-in-buffer/entries-written: 47827/42233061 #P:2 583 # 584 # _-----=> irqs-off 585 # / _----=> need-resched 586 # | / _---=> hardirq/softirq 587 # || / _--=> preempt-depth 588 # ||| / delay 589 # TASK-PID CPU# |||| TIMESTAMP FUNCTION 590 # | | | |||| | | 591 <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 592 <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 593 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 594 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 595 <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 596 <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 597 <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 598 599The ``cpu_frequency`` trace event will be triggered either by the ``schedutil`` scaling 600governor (for the policies it is attached to), or by the ``CPUFreq`` core (for the 601policies with other scaling governors). 602 603 604Tracer Tool 605------------- 606 607``amd_pstate_tracer.py`` can record and parse ``amd-pstate`` trace log, then 608generate performance plots. This utility can be used to debug and tune the 609performance of ``amd-pstate`` driver. The tracer tool needs to import intel 610pstate tracer. 611 612Tracer tool located in ``linux/tools/power/x86/amd_pstate_tracer``. It can be 613used in two ways. If trace file is available, then directly parse the file 614with command :: 615 616 ./amd_pstate_trace.py [-c cpus] -t <trace_file> -n <test_name> 617 618Or generate trace file with root privilege, then parse and plot with command :: 619 620 sudo ./amd_pstate_trace.py [-c cpus] -n <test_name> -i <interval> [-m kbytes] 621 622The test result can be found in ``results/test_name``. Following is the example 623about part of the output. :: 624 625 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 626 CPU_005 712 116384 39 49 166 0.7565 9645075 2214891 38431470 25.1 11.646 469 2.496 kworker/5:0-40 627 CPU_006 712 116408 39 49 166 0.6769 8950227 1839034 37192089 24.06 11.272 470 2.496 kworker/6:0-1264 628 629Unit Tests for amd-pstate 630------------------------- 631 632``amd-pstate-ut`` is a test module for testing the ``amd-pstate`` driver. 633 634 * It can help all users to verify their processor support (SBIOS/Firmware or Hardware). 635 636 * Kernel can have a basic function test to avoid the kernel regression during the update. 637 638 * We can introduce more functional or performance tests to align the result together, it will benefit power and performance scale optimization. 639 6401. Test case descriptions 641 642 1). Basic tests 643 644 Test prerequisite and basic functions for the ``amd-pstate`` driver. 645 646 +---------+--------------------------------+------------------------------------------------------------------------------------+ 647 | Index | Functions | Description | 648 +=========+================================+====================================================================================+ 649 | 1 | amd_pstate_ut_acpi_cpc_valid || Check whether the _CPC object is present in SBIOS. | 650 | | || | 651 | | || The detail refer to `Processor Support <processor_support_>`_. | 652 +---------+--------------------------------+------------------------------------------------------------------------------------+ 653 | 2 | amd_pstate_ut_check_enabled || Check whether AMD P-State is enabled. | 654 | | || | 655 | | || AMD P-States and ACPI hardware P-States always can be supported in one processor. | 656 | | | But AMD P-States has the higher priority and if it is enabled with | 657 | | | :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_set_enable``, it will respond to the | 658 | | | request from AMD P-States. | 659 +---------+--------------------------------+------------------------------------------------------------------------------------+ 660 | 3 | amd_pstate_ut_check_perf || Check if the each performance values are reasonable. | 661 | | || highest_perf >= nominal_perf > lowest_nonlinear_perf > lowest_perf > 0. | 662 +---------+--------------------------------+------------------------------------------------------------------------------------+ 663 | 4 | amd_pstate_ut_check_freq || Check if the each frequency values and max freq when set support boost mode | 664 | | | are reasonable. | 665 | | || max_freq >= nominal_freq > lowest_nonlinear_freq > min_freq > 0 | 666 | | || If boost is not active but supported, this maximum frequency will be larger than | 667 | | | the one in ``cpuinfo``. | 668 +---------+--------------------------------+------------------------------------------------------------------------------------+ 669 670 2). Tbench test 671 672 Test and monitor the cpu changes when running tbench benchmark under the specified governor. 673 These changes include desire performance, frequency, load, performance, energy etc. 674 The specified governor is ondemand or schedutil. 675 Tbench can also be tested on the ``acpi-cpufreq`` kernel driver for comparison. 676 677 3). Gitsource test 678 679 Test and monitor the cpu changes when running gitsource benchmark under the specified governor. 680 These changes include desire performance, frequency, load, time, energy etc. 681 The specified governor is ondemand or schedutil. 682 Gitsource can also be tested on the ``acpi-cpufreq`` kernel driver for comparison. 683 684#. How to execute the tests 685 686 We use test module in the kselftest frameworks to implement it. 687 We create ``amd-pstate-ut`` module and tie it into kselftest.(for 688 details refer to Linux Kernel Selftests [4]_). 689 690 1). Build 691 692 + open the :c:macro:`CONFIG_X86_AMD_PSTATE` configuration option. 693 + set the :c:macro:`CONFIG_X86_AMD_PSTATE_UT` configuration option to M. 694 + make project 695 + make selftest :: 696 697 $ cd linux 698 $ make -C tools/testing/selftests 699 700 + make perf :: 701 702 $ cd tools/perf/ 703 $ make 704 705 706 2). Installation & Steps :: 707 708 $ make -C tools/testing/selftests install INSTALL_PATH=~/kselftest 709 $ cp tools/perf/perf /usr/bin/perf 710 $ sudo ./kselftest/run_kselftest.sh -c amd-pstate 711 712 3). Specified test case :: 713 714 $ cd ~/kselftest/amd-pstate 715 $ sudo ./run.sh -t basic 716 $ sudo ./run.sh -t tbench 717 $ sudo ./run.sh -t tbench -m acpi-cpufreq 718 $ sudo ./run.sh -t gitsource 719 $ sudo ./run.sh -t gitsource -m acpi-cpufreq 720 $ ./run.sh --help 721 ./run.sh: illegal option -- - 722 Usage: ./run.sh [OPTION...] 723 [-h <help>] 724 [-o <output-file-for-dump>] 725 [-c <all: All testing, 726 basic: Basic testing, 727 tbench: Tbench testing, 728 gitsource: Gitsource testing.>] 729 [-t <tbench time limit>] 730 [-p <tbench process number>] 731 [-l <loop times for tbench>] 732 [-i <amd tracer interval>] 733 [-m <comparative test: acpi-cpufreq>] 734 735 736 4). Results 737 738 + basic 739 740 When you finish test, you will get the following log info :: 741 742 $ dmesg | grep "amd_pstate_ut" | tee log.txt 743 [12977.570663] amd_pstate_ut: 1 amd_pstate_ut_acpi_cpc_valid success! 744 [12977.570673] amd_pstate_ut: 2 amd_pstate_ut_check_enabled success! 745 [12977.571207] amd_pstate_ut: 3 amd_pstate_ut_check_perf success! 746 [12977.571212] amd_pstate_ut: 4 amd_pstate_ut_check_freq success! 747 748 + tbench 749 750 When you finish test, you will get selftest.tbench.csv and png images. 751 The selftest.tbench.csv file contains the raw data and the drop of the comparative test. 752 The png images shows the performance, energy and performan per watt of each test. 753 Open selftest.tbench.csv : 754 755 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 756 + Governor | Round | Des-perf | Freq | Load | Performance | Energy | Performance Per Watt | 757 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 758 + Unit | | | GHz | | MB/s | J | MB/J | 759 +=================================================+==============+==========+=========+==========+=============+=========+======================+ 760 + amd-pstate-ondemand | 1 | | | | 2504.05 | 1563.67 | 158.5378 | 761 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 762 + amd-pstate-ondemand | 2 | | | | 2243.64 | 1430.32 | 155.2941 | 763 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 764 + amd-pstate-ondemand | 3 | | | | 2183.88 | 1401.32 | 154.2860 | 765 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 766 + amd-pstate-ondemand | Average | | | | 2310.52 | 1465.1 | 156.1268 | 767 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 768 + amd-pstate-schedutil | 1 | 165.329 | 1.62257 | 99.798 | 2136.54 | 1395.26 | 151.5971 | 769 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 770 + amd-pstate-schedutil | 2 | 166 | 1.49761 | 99.9993 | 2100.56 | 1380.5 | 150.6377 | 771 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 772 + amd-pstate-schedutil | 3 | 166 | 1.47806 | 99.9993 | 2084.12 | 1375.76 | 149.9737 | 773 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 774 + amd-pstate-schedutil | Average | 165.776 | 1.53275 | 99.9322 | 2107.07 | 1383.84 | 150.7399 | 775 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 776 + acpi-cpufreq-ondemand | 1 | | | | 2529.9 | 1564.4 | 160.0997 | 777 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 778 + acpi-cpufreq-ondemand | 2 | | | | 2249.76 | 1432.97 | 155.4297 | 779 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 780 + acpi-cpufreq-ondemand | 3 | | | | 2181.46 | 1406.88 | 153.5060 | 781 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 782 + acpi-cpufreq-ondemand | Average | | | | 2320.37 | 1468.08 | 156.4741 | 783 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 784 + acpi-cpufreq-schedutil | 1 | | | | 2137.64 | 1385.24 | 152.7723 | 785 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 786 + acpi-cpufreq-schedutil | 2 | | | | 2107.05 | 1372.23 | 152.0138 | 787 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 788 + acpi-cpufreq-schedutil | 3 | | | | 2085.86 | 1365.35 | 151.2433 | 789 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 790 + acpi-cpufreq-schedutil | Average | | | | 2110.18 | 1374.27 | 152.0136 | 791 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 792 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | -9.0584 | -6.3899 | -2.8506 | 793 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 794 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | | | | 8.8053 | -5.5463 | -3.4503 | 795 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 796 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | -0.4245 | -0.2029 | -0.2219 | 797 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 798 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | -0.1473 | 0.6963 | -0.8378 | 799 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 800 801 + gitsource 802 803 When you finish test, you will get selftest.gitsource.csv and png images. 804 The selftest.gitsource.csv file contains the raw data and the drop of the comparative test. 805 The png images shows the performance, energy and performan per watt of each test. 806 Open selftest.gitsource.csv : 807 808 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 809 + Governor | Round | Des-perf | Freq | Load | Time | Energy | Performance Per Watt | 810 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 811 + Unit | | | GHz | | s | J | 1/J | 812 +=================================================+==============+==========+==========+==========+=============+=========+======================+ 813 + amd-pstate-ondemand | 1 | 50.119 | 2.10509 | 23.3076 | 475.69 | 865.78 | 0.001155027 | 814 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 815 + amd-pstate-ondemand | 2 | 94.8006 | 1.98771 | 56.6533 | 467.1 | 839.67 | 0.001190944 | 816 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 817 + amd-pstate-ondemand | 3 | 76.6091 | 2.53251 | 43.7791 | 467.69 | 855.85 | 0.001168429 | 818 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 819 + amd-pstate-ondemand | Average | 73.8429 | 2.20844 | 41.2467 | 470.16 | 853.767 | 0.001171279 | 820 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 821 + amd-pstate-schedutil | 1 | 165.919 | 1.62319 | 98.3868 | 464.17 | 866.8 | 0.001153668 | 822 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 823 + amd-pstate-schedutil | 2 | 165.97 | 1.31309 | 99.5712 | 480.15 | 880.4 | 0.001135847 | 824 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 825 + amd-pstate-schedutil | 3 | 165.973 | 1.28448 | 99.9252 | 481.79 | 867.02 | 0.001153375 | 826 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 827 + amd-pstate-schedutil | Average | 165.954 | 1.40692 | 99.2944 | 475.37 | 871.407 | 0.001147569 | 828 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 829 + acpi-cpufreq-ondemand | 1 | | | | 2379.62 | 742.96 | 0.001345967 | 830 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 831 + acpi-cpufreq-ondemand | 2 | | | | 441.74 | 817.49 | 0.001223256 | 832 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 833 + acpi-cpufreq-ondemand | 3 | | | | 455.48 | 820.01 | 0.001219497 | 834 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 835 + acpi-cpufreq-ondemand | Average | | | | 425.613 | 793.487 | 0.001260260 | 836 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 837 + acpi-cpufreq-schedutil | 1 | | | | 459.69 | 838.54 | 0.001192548 | 838 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 839 + acpi-cpufreq-schedutil | 2 | | | | 466.55 | 830.89 | 0.001203528 | 840 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 841 + acpi-cpufreq-schedutil | 3 | | | | 470.38 | 837.32 | 0.001194286 | 842 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 843 + acpi-cpufreq-schedutil | Average | | | | 465.54 | 835.583 | 0.001196769 | 844 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 845 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | 9.3810 | 5.3051 | -5.0379 | 846 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 847 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | 124.7392 | -36.2934 | 140.7329 | 1.1081 | 2.0661 | -2.0242 | 848 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 849 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | 10.4665 | 7.5968 | -7.0605 | 850 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 851 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | 2.1115 | 4.2873 | -4.1110 | 852 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 853 854Reference 855=========== 856 857.. [1] AMD64 Architecture Programmer's Manual Volume 2: System Programming, 858 https://docs.amd.com/v/u/en-US/24593_3.44_APM_Vol2 859 860.. [2] Advanced Configuration and Power Interface Specification, 861 https://uefi.org/sites/default/files/resources/ACPI_Spec_6_4_Jan22.pdf 862 863.. [3] Processor Programming Reference (PPR) for AMD Family 19h Model 51h, Revision A1 Processors 864 https://docs.amd.com/v/u/en-US/56569-A1-PUB_3.03 865 866.. [4] Linux Kernel Selftests, 867 https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html 868