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 | | Lowest 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_lowest_nonlinear_freq 243 /sys/devices/system/cpu/cpufreq/policy0/amd_pstate_max_freq 244 245 246``amd_pstate_highest_perf / amd_pstate_max_freq`` 247 248Maximum CPPC performance and CPU frequency that the driver is allowed to 249set, in percent of the maximum supported CPPC performance level (the highest 250performance supported in `AMD CPPC Performance Capability <perf_cap_>`_). 251In some ASICs, the highest CPPC performance is not the one in the ``_CPC`` 252table, so we need to expose it to sysfs. If boost is not active, but 253still supported, this maximum frequency will be larger than the one in 254``cpuinfo``. 255This attribute is read-only. 256 257``amd_pstate_lowest_nonlinear_freq`` 258 259The lowest non-linear CPPC CPU frequency that the driver is allowed to set, 260in percent of the maximum supported CPPC performance level. (Please see the 261lowest non-linear performance in `AMD CPPC Performance Capability 262<perf_cap_>`_.) 263This attribute is read-only. 264 265``energy_performance_available_preferences`` 266 267A list of all the supported EPP preferences that could be used for 268``energy_performance_preference`` on this system. 269These profiles represent different hints that are provided 270to the low-level firmware about the user's desired energy vs efficiency 271tradeoff. ``default`` represents the epp value is set by platform 272firmware. This attribute is read-only. 273 274``energy_performance_preference`` 275 276The current energy performance preference can be read from this attribute. 277and user can change current preference according to energy or performance needs 278Please get all support profiles list from 279``energy_performance_available_preferences`` attribute, all the profiles are 280integer values defined between 0 to 255 when EPP feature is enabled by platform 281firmware, if EPP feature is disabled, driver will ignore the written value 282This attribute is read-write. 283 284``boost`` 285The `boost` sysfs attribute provides control over the CPU core 286performance boost, allowing users to manage the maximum frequency limitation 287of the CPU. This attribute can be used to enable or disable the boost feature 288on individual CPUs. 289 290When the boost feature is enabled, the CPU can dynamically increase its frequency 291beyond the base frequency, providing enhanced performance for demanding workloads. 292On the other hand, disabling the boost feature restricts the CPU to operate at the 293base frequency, which may be desirable in certain scenarios to prioritize power 294efficiency or manage temperature. 295 296To manipulate the `boost` attribute, users can write a value of `0` to disable the 297boost or `1` to enable it, for the respective CPU using the sysfs path 298`/sys/devices/system/cpu/cpuX/cpufreq/boost`, where `X` represents the CPU number. 299 300Other performance and frequency values can be read back from 301``/sys/devices/system/cpu/cpuX/acpi_cppc/``, see :ref:`cppc_sysfs`. 302 303 304``amd-pstate`` vs ``acpi-cpufreq`` 305====================================== 306 307On the majority of AMD platforms supported by ``acpi-cpufreq``, the ACPI tables 308provided by the platform firmware are used for CPU performance scaling, but 309only provide 3 P-states on AMD processors. 310However, on modern AMD APU and CPU series, hardware provides the Collaborative 311Processor Performance Control according to the ACPI protocol and customizes this 312for AMD platforms. That is, fine-grained and continuous frequency ranges 313instead of the legacy hardware P-states. ``amd-pstate`` is the kernel 314module which supports the new AMD P-States mechanism on most of the future AMD 315platforms. The AMD P-States mechanism is the more performance and energy 316efficiency frequency management method on AMD processors. 317 318 319``amd-pstate`` Driver Operation Modes 320====================================== 321 322``amd_pstate`` CPPC has 3 operation modes: autonomous (active) mode, 323non-autonomous (passive) mode and guided autonomous (guided) mode. 324Active/passive/guided mode can be chosen by different kernel parameters. 325 326- In autonomous mode, platform ignores the desired performance level request 327 and takes into account only the values set to the minimum, maximum and energy 328 performance preference registers. 329- In non-autonomous mode, platform gets desired performance level 330 from OS directly through Desired Performance Register. 331- In guided-autonomous mode, platform sets operating performance level 332 autonomously according to the current workload and within the limits set by 333 OS through min and max performance registers. 334 335Active Mode 336------------ 337 338``amd_pstate=active`` 339 340This is the low-level firmware control mode which is implemented by ``amd_pstate_epp`` 341driver with ``amd_pstate=active`` passed to the kernel in the command line. 342In this mode, ``amd_pstate_epp`` driver provides a hint to the hardware if software 343wants to bias toward performance (0x0) or energy efficiency (0xff) to the CPPC firmware. 344then CPPC power algorithm will calculate the runtime workload and adjust the realtime 345cores frequency according to the power supply and thermal, core voltage and some other 346hardware conditions. 347 348Passive Mode 349------------ 350 351``amd_pstate=passive`` 352 353It will be enabled if the ``amd_pstate=passive`` is passed to the kernel in the command line. 354In this mode, ``amd_pstate`` driver software specifies a desired QoS target in the CPPC 355performance scale as a relative number. This can be expressed as percentage of nominal 356performance (infrastructure max). Below the nominal sustained performance level, 357desired performance expresses the average performance level of the processor subject 358to the Performance Reduction Tolerance register. Above the nominal performance level, 359processor must provide at least nominal performance requested and go higher if current 360operating conditions allow. 361 362Guided Mode 363----------- 364 365``amd_pstate=guided`` 366 367If ``amd_pstate=guided`` is passed to kernel command line option then this mode 368is activated. In this mode, driver requests minimum and maximum performance 369level and the platform autonomously selects a performance level in this range 370and appropriate to the current workload. 371 372``amd-pstate`` Preferred Core 373================================= 374 375The core frequency is subjected to the process variation in semiconductors. 376Not all cores are able to reach the maximum frequency respecting the 377infrastructure limits. Consequently, AMD has redefined the concept of 378maximum frequency of a part. This means that a fraction of cores can reach 379maximum frequency. To find the best process scheduling policy for a given 380scenario, OS needs to know the core ordering informed by the platform through 381highest performance capability register of the CPPC interface. 382 383``amd-pstate`` preferred core enables the scheduler to prefer scheduling on 384cores that can achieve a higher frequency with lower voltage. The preferred 385core rankings can dynamically change based on the workload, platform conditions, 386thermals and ageing. 387 388The priority metric will be initialized by the ``amd-pstate`` driver. The ``amd-pstate`` 389driver will also determine whether or not ``amd-pstate`` preferred core is 390supported by the platform. 391 392``amd-pstate`` driver will provide an initial core ordering when the system boots. 393The platform uses the CPPC interfaces to communicate the core ranking to the 394operating system and scheduler to make sure that OS is choosing the cores 395with highest performance firstly for scheduling the process. When ``amd-pstate`` 396driver receives a message with the highest performance change, it will 397update the core ranking and set the cpu's priority. 398 399``amd-pstate`` Preferred Core Switch 400===================================== 401Kernel Parameters 402----------------- 403 404``amd-pstate`` peferred core`` has two states: enable and disable. 405Enable/disable states can be chosen by different kernel parameters. 406Default enable ``amd-pstate`` preferred core. 407 408``amd_prefcore=disable`` 409 410For systems that support ``amd-pstate`` preferred core, the core rankings will 411always be advertised by the platform. But OS can choose to ignore that via the 412kernel parameter ``amd_prefcore=disable``. 413 414User Space Interface in ``sysfs`` - General 415=========================================== 416 417Global Attributes 418----------------- 419 420``amd-pstate`` exposes several global attributes (files) in ``sysfs`` to 421control its functionality at the system level. They are located in the 422``/sys/devices/system/cpu/amd_pstate/`` directory and affect all CPUs. 423 424``status`` 425 Operation mode of the driver: "active", "passive", "guided" or "disable". 426 427 "active" 428 The driver is functional and in the ``active mode`` 429 430 "passive" 431 The driver is functional and in the ``passive mode`` 432 433 "guided" 434 The driver is functional and in the ``guided mode`` 435 436 "disable" 437 The driver is unregistered and not functional now. 438 439 This attribute can be written to in order to change the driver's 440 operation mode or to unregister it. The string written to it must be 441 one of the possible values of it and, if successful, writing one of 442 these values to the sysfs file will cause the driver to switch over 443 to the operation mode represented by that string - or to be 444 unregistered in the "disable" case. 445 446``prefcore`` 447 Preferred core state of the driver: "enabled" or "disabled". 448 449 "enabled" 450 Enable the ``amd-pstate`` preferred core. 451 452 "disabled" 453 Disable the ``amd-pstate`` preferred core 454 455 456 This attribute is read-only to check the state of preferred core set 457 by the kernel parameter. 458 459``cpupower`` tool support for ``amd-pstate`` 460=============================================== 461 462``amd-pstate`` is supported by the ``cpupower`` tool, which can be used to dump 463frequency information. Development is in progress to support more and more 464operations for the new ``amd-pstate`` module with this tool. :: 465 466 root@hr-test1:/home/ray# cpupower frequency-info 467 analyzing CPU 0: 468 driver: amd-pstate 469 CPUs which run at the same hardware frequency: 0 470 CPUs which need to have their frequency coordinated by software: 0 471 maximum transition latency: 131 us 472 hardware limits: 400 MHz - 4.68 GHz 473 available cpufreq governors: ondemand conservative powersave userspace performance schedutil 474 current policy: frequency should be within 400 MHz and 4.68 GHz. 475 The governor "schedutil" may decide which speed to use 476 within this range. 477 current CPU frequency: Unable to call hardware 478 current CPU frequency: 4.02 GHz (asserted by call to kernel) 479 boost state support: 480 Supported: yes 481 Active: yes 482 AMD PSTATE Highest Performance: 166. Maximum Frequency: 4.68 GHz. 483 AMD PSTATE Nominal Performance: 117. Nominal Frequency: 3.30 GHz. 484 AMD PSTATE Lowest Non-linear Performance: 39. Lowest Non-linear Frequency: 1.10 GHz. 485 AMD PSTATE Lowest Performance: 15. Lowest Frequency: 400 MHz. 486 487 488Diagnostics and Tuning 489======================= 490 491Trace Events 492-------------- 493 494There are two static trace events that can be used for ``amd-pstate`` 495diagnostics. One of them is the ``cpu_frequency`` trace event generally used 496by ``CPUFreq``, and the other one is the ``amd_pstate_perf`` trace event 497specific to ``amd-pstate``. The following sequence of shell commands can 498be used to enable them and see their output (if the kernel is 499configured to support event tracing). :: 500 501 root@hr-test1:/home/ray# cd /sys/kernel/tracing/ 502 root@hr-test1:/sys/kernel/tracing# echo 1 > events/amd_cpu/enable 503 root@hr-test1:/sys/kernel/tracing# cat trace 504 # tracer: nop 505 # 506 # entries-in-buffer/entries-written: 47827/42233061 #P:2 507 # 508 # _-----=> irqs-off 509 # / _----=> need-resched 510 # | / _---=> hardirq/softirq 511 # || / _--=> preempt-depth 512 # ||| / delay 513 # TASK-PID CPU# |||| TIMESTAMP FUNCTION 514 # | | | |||| | | 515 <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 516 <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 517 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 518 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 519 <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 520 <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 521 <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 522 523The ``cpu_frequency`` trace event will be triggered either by the ``schedutil`` scaling 524governor (for the policies it is attached to), or by the ``CPUFreq`` core (for the 525policies with other scaling governors). 526 527 528Tracer Tool 529------------- 530 531``amd_pstate_tracer.py`` can record and parse ``amd-pstate`` trace log, then 532generate performance plots. This utility can be used to debug and tune the 533performance of ``amd-pstate`` driver. The tracer tool needs to import intel 534pstate tracer. 535 536Tracer tool located in ``linux/tools/power/x86/amd_pstate_tracer``. It can be 537used in two ways. If trace file is available, then directly parse the file 538with command :: 539 540 ./amd_pstate_trace.py [-c cpus] -t <trace_file> -n <test_name> 541 542Or generate trace file with root privilege, then parse and plot with command :: 543 544 sudo ./amd_pstate_trace.py [-c cpus] -n <test_name> -i <interval> [-m kbytes] 545 546The test result can be found in ``results/test_name``. Following is the example 547about part of the output. :: 548 549 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 550 CPU_005 712 116384 39 49 166 0.7565 9645075 2214891 38431470 25.1 11.646 469 2.496 kworker/5:0-40 551 CPU_006 712 116408 39 49 166 0.6769 8950227 1839034 37192089 24.06 11.272 470 2.496 kworker/6:0-1264 552 553Unit Tests for amd-pstate 554------------------------- 555 556``amd-pstate-ut`` is a test module for testing the ``amd-pstate`` driver. 557 558 * It can help all users to verify their processor support (SBIOS/Firmware or Hardware). 559 560 * Kernel can have a basic function test to avoid the kernel regression during the update. 561 562 * We can introduce more functional or performance tests to align the result together, it will benefit power and performance scale optimization. 563 5641. Test case descriptions 565 566 1). Basic tests 567 568 Test prerequisite and basic functions for the ``amd-pstate`` driver. 569 570 +---------+--------------------------------+------------------------------------------------------------------------------------+ 571 | Index | Functions | Description | 572 +=========+================================+====================================================================================+ 573 | 1 | amd_pstate_ut_acpi_cpc_valid || Check whether the _CPC object is present in SBIOS. | 574 | | || | 575 | | || The detail refer to `Processor Support <processor_support_>`_. | 576 +---------+--------------------------------+------------------------------------------------------------------------------------+ 577 | 2 | amd_pstate_ut_check_enabled || Check whether AMD P-State is enabled. | 578 | | || | 579 | | || AMD P-States and ACPI hardware P-States always can be supported in one processor. | 580 | | | But AMD P-States has the higher priority and if it is enabled with | 581 | | | :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_set_enable``, it will respond to the | 582 | | | request from AMD P-States. | 583 +---------+--------------------------------+------------------------------------------------------------------------------------+ 584 | 3 | amd_pstate_ut_check_perf || Check if the each performance values are reasonable. | 585 | | || highest_perf >= nominal_perf > lowest_nonlinear_perf > lowest_perf > 0. | 586 +---------+--------------------------------+------------------------------------------------------------------------------------+ 587 | 4 | amd_pstate_ut_check_freq || Check if the each frequency values and max freq when set support boost mode | 588 | | | are reasonable. | 589 | | || max_freq >= nominal_freq > lowest_nonlinear_freq > min_freq > 0 | 590 | | || If boost is not active but supported, this maximum frequency will be larger than | 591 | | | the one in ``cpuinfo``. | 592 +---------+--------------------------------+------------------------------------------------------------------------------------+ 593 594 2). Tbench test 595 596 Test and monitor the cpu changes when running tbench benchmark under the specified governor. 597 These changes include desire performance, frequency, load, performance, energy etc. 598 The specified governor is ondemand or schedutil. 599 Tbench can also be tested on the ``acpi-cpufreq`` kernel driver for comparison. 600 601 3). Gitsource test 602 603 Test and monitor the cpu changes when running gitsource benchmark under the specified governor. 604 These changes include desire performance, frequency, load, time, energy etc. 605 The specified governor is ondemand or schedutil. 606 Gitsource can also be tested on the ``acpi-cpufreq`` kernel driver for comparison. 607 608#. How to execute the tests 609 610 We use test module in the kselftest frameworks to implement it. 611 We create ``amd-pstate-ut`` module and tie it into kselftest.(for 612 details refer to Linux Kernel Selftests [4]_). 613 614 1). Build 615 616 + open the :c:macro:`CONFIG_X86_AMD_PSTATE` configuration option. 617 + set the :c:macro:`CONFIG_X86_AMD_PSTATE_UT` configuration option to M. 618 + make project 619 + make selftest :: 620 621 $ cd linux 622 $ make -C tools/testing/selftests 623 624 + make perf :: 625 626 $ cd tools/perf/ 627 $ make 628 629 630 2). Installation & Steps :: 631 632 $ make -C tools/testing/selftests install INSTALL_PATH=~/kselftest 633 $ cp tools/perf/perf /usr/bin/perf 634 $ sudo ./kselftest/run_kselftest.sh -c amd-pstate 635 636 3). Specified test case :: 637 638 $ cd ~/kselftest/amd-pstate 639 $ sudo ./run.sh -t basic 640 $ sudo ./run.sh -t tbench 641 $ sudo ./run.sh -t tbench -m acpi-cpufreq 642 $ sudo ./run.sh -t gitsource 643 $ sudo ./run.sh -t gitsource -m acpi-cpufreq 644 $ ./run.sh --help 645 ./run.sh: illegal option -- - 646 Usage: ./run.sh [OPTION...] 647 [-h <help>] 648 [-o <output-file-for-dump>] 649 [-c <all: All testing, 650 basic: Basic testing, 651 tbench: Tbench testing, 652 gitsource: Gitsource testing.>] 653 [-t <tbench time limit>] 654 [-p <tbench process number>] 655 [-l <loop times for tbench>] 656 [-i <amd tracer interval>] 657 [-m <comparative test: acpi-cpufreq>] 658 659 660 4). Results 661 662 + basic 663 664 When you finish test, you will get the following log info :: 665 666 $ dmesg | grep "amd_pstate_ut" | tee log.txt 667 [12977.570663] amd_pstate_ut: 1 amd_pstate_ut_acpi_cpc_valid success! 668 [12977.570673] amd_pstate_ut: 2 amd_pstate_ut_check_enabled success! 669 [12977.571207] amd_pstate_ut: 3 amd_pstate_ut_check_perf success! 670 [12977.571212] amd_pstate_ut: 4 amd_pstate_ut_check_freq success! 671 672 + tbench 673 674 When you finish test, you will get selftest.tbench.csv and png images. 675 The selftest.tbench.csv file contains the raw data and the drop of the comparative test. 676 The png images shows the performance, energy and performan per watt of each test. 677 Open selftest.tbench.csv : 678 679 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 680 + Governor | Round | Des-perf | Freq | Load | Performance | Energy | Performance Per Watt | 681 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 682 + Unit | | | GHz | | MB/s | J | MB/J | 683 +=================================================+==============+==========+=========+==========+=============+=========+======================+ 684 + amd-pstate-ondemand | 1 | | | | 2504.05 | 1563.67 | 158.5378 | 685 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 686 + amd-pstate-ondemand | 2 | | | | 2243.64 | 1430.32 | 155.2941 | 687 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 688 + amd-pstate-ondemand | 3 | | | | 2183.88 | 1401.32 | 154.2860 | 689 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 690 + amd-pstate-ondemand | Average | | | | 2310.52 | 1465.1 | 156.1268 | 691 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 692 + amd-pstate-schedutil | 1 | 165.329 | 1.62257 | 99.798 | 2136.54 | 1395.26 | 151.5971 | 693 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 694 + amd-pstate-schedutil | 2 | 166 | 1.49761 | 99.9993 | 2100.56 | 1380.5 | 150.6377 | 695 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 696 + amd-pstate-schedutil | 3 | 166 | 1.47806 | 99.9993 | 2084.12 | 1375.76 | 149.9737 | 697 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 698 + amd-pstate-schedutil | Average | 165.776 | 1.53275 | 99.9322 | 2107.07 | 1383.84 | 150.7399 | 699 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 700 + acpi-cpufreq-ondemand | 1 | | | | 2529.9 | 1564.4 | 160.0997 | 701 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 702 + acpi-cpufreq-ondemand | 2 | | | | 2249.76 | 1432.97 | 155.4297 | 703 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 704 + acpi-cpufreq-ondemand | 3 | | | | 2181.46 | 1406.88 | 153.5060 | 705 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 706 + acpi-cpufreq-ondemand | Average | | | | 2320.37 | 1468.08 | 156.4741 | 707 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 708 + acpi-cpufreq-schedutil | 1 | | | | 2137.64 | 1385.24 | 152.7723 | 709 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 710 + acpi-cpufreq-schedutil | 2 | | | | 2107.05 | 1372.23 | 152.0138 | 711 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 712 + acpi-cpufreq-schedutil | 3 | | | | 2085.86 | 1365.35 | 151.2433 | 713 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 714 + acpi-cpufreq-schedutil | Average | | | | 2110.18 | 1374.27 | 152.0136 | 715 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 716 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | -9.0584 | -6.3899 | -2.8506 | 717 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 718 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | | | | 8.8053 | -5.5463 | -3.4503 | 719 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 720 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | -0.4245 | -0.2029 | -0.2219 | 721 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 722 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | -0.1473 | 0.6963 | -0.8378 | 723 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 724 725 + gitsource 726 727 When you finish test, you will get selftest.gitsource.csv and png images. 728 The selftest.gitsource.csv file contains the raw data and the drop of the comparative test. 729 The png images shows the performance, energy and performan per watt of each test. 730 Open selftest.gitsource.csv : 731 732 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 733 + Governor | Round | Des-perf | Freq | Load | Time | Energy | Performance Per Watt | 734 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 735 + Unit | | | GHz | | s | J | 1/J | 736 +=================================================+==============+==========+==========+==========+=============+=========+======================+ 737 + amd-pstate-ondemand | 1 | 50.119 | 2.10509 | 23.3076 | 475.69 | 865.78 | 0.001155027 | 738 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 739 + amd-pstate-ondemand | 2 | 94.8006 | 1.98771 | 56.6533 | 467.1 | 839.67 | 0.001190944 | 740 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 741 + amd-pstate-ondemand | 3 | 76.6091 | 2.53251 | 43.7791 | 467.69 | 855.85 | 0.001168429 | 742 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 743 + amd-pstate-ondemand | Average | 73.8429 | 2.20844 | 41.2467 | 470.16 | 853.767 | 0.001171279 | 744 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 745 + amd-pstate-schedutil | 1 | 165.919 | 1.62319 | 98.3868 | 464.17 | 866.8 | 0.001153668 | 746 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 747 + amd-pstate-schedutil | 2 | 165.97 | 1.31309 | 99.5712 | 480.15 | 880.4 | 0.001135847 | 748 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 749 + amd-pstate-schedutil | 3 | 165.973 | 1.28448 | 99.9252 | 481.79 | 867.02 | 0.001153375 | 750 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 751 + amd-pstate-schedutil | Average | 165.954 | 1.40692 | 99.2944 | 475.37 | 871.407 | 0.001147569 | 752 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 753 + acpi-cpufreq-ondemand | 1 | | | | 2379.62 | 742.96 | 0.001345967 | 754 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 755 + acpi-cpufreq-ondemand | 2 | | | | 441.74 | 817.49 | 0.001223256 | 756 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 757 + acpi-cpufreq-ondemand | 3 | | | | 455.48 | 820.01 | 0.001219497 | 758 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 759 + acpi-cpufreq-ondemand | Average | | | | 425.613 | 793.487 | 0.001260260 | 760 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 761 + acpi-cpufreq-schedutil | 1 | | | | 459.69 | 838.54 | 0.001192548 | 762 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 763 + acpi-cpufreq-schedutil | 2 | | | | 466.55 | 830.89 | 0.001203528 | 764 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 765 + acpi-cpufreq-schedutil | 3 | | | | 470.38 | 837.32 | 0.001194286 | 766 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 767 + acpi-cpufreq-schedutil | Average | | | | 465.54 | 835.583 | 0.001196769 | 768 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 769 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | 9.3810 | 5.3051 | -5.0379 | 770 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 771 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | 124.7392 | -36.2934 | 140.7329 | 1.1081 | 2.0661 | -2.0242 | 772 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 773 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | 10.4665 | 7.5968 | -7.0605 | 774 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 775 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | 2.1115 | 4.2873 | -4.1110 | 776 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 777 778Reference 779=========== 780 781.. [1] AMD64 Architecture Programmer's Manual Volume 2: System Programming, 782 https://www.amd.com/system/files/TechDocs/24593.pdf 783 784.. [2] Advanced Configuration and Power Interface Specification, 785 https://uefi.org/sites/default/files/resources/ACPI_Spec_6_4_Jan22.pdf 786 787.. [3] Processor Programming Reference (PPR) for AMD Family 19h Model 51h, Revision A1 Processors 788 https://www.amd.com/system/files/TechDocs/56569-A1-PUB.zip 789 790.. [4] Linux Kernel Selftests, 791 https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html 792