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 284Other performance and frequency values can be read back from 285``/sys/devices/system/cpu/cpuX/acpi_cppc/``, see :ref:`cppc_sysfs`. 286 287 288``amd-pstate`` vs ``acpi-cpufreq`` 289====================================== 290 291On the majority of AMD platforms supported by ``acpi-cpufreq``, the ACPI tables 292provided by the platform firmware are used for CPU performance scaling, but 293only provide 3 P-states on AMD processors. 294However, on modern AMD APU and CPU series, hardware provides the Collaborative 295Processor Performance Control according to the ACPI protocol and customizes this 296for AMD platforms. That is, fine-grained and continuous frequency ranges 297instead of the legacy hardware P-states. ``amd-pstate`` is the kernel 298module which supports the new AMD P-States mechanism on most of the future AMD 299platforms. The AMD P-States mechanism is the more performance and energy 300efficiency frequency management method on AMD processors. 301 302 303``amd-pstate`` Driver Operation Modes 304====================================== 305 306``amd_pstate`` CPPC has 3 operation modes: autonomous (active) mode, 307non-autonomous (passive) mode and guided autonomous (guided) mode. 308Active/passive/guided mode can be chosen by different kernel parameters. 309 310- In autonomous mode, platform ignores the desired performance level request 311 and takes into account only the values set to the minimum, maximum and energy 312 performance preference registers. 313- In non-autonomous mode, platform gets desired performance level 314 from OS directly through Desired Performance Register. 315- In guided-autonomous mode, platform sets operating performance level 316 autonomously according to the current workload and within the limits set by 317 OS through min and max performance registers. 318 319Active Mode 320------------ 321 322``amd_pstate=active`` 323 324This is the low-level firmware control mode which is implemented by ``amd_pstate_epp`` 325driver with ``amd_pstate=active`` passed to the kernel in the command line. 326In this mode, ``amd_pstate_epp`` driver provides a hint to the hardware if software 327wants to bias toward performance (0x0) or energy efficiency (0xff) to the CPPC firmware. 328then CPPC power algorithm will calculate the runtime workload and adjust the realtime 329cores frequency according to the power supply and thermal, core voltage and some other 330hardware conditions. 331 332Passive Mode 333------------ 334 335``amd_pstate=passive`` 336 337It will be enabled if the ``amd_pstate=passive`` is passed to the kernel in the command line. 338In this mode, ``amd_pstate`` driver software specifies a desired QoS target in the CPPC 339performance scale as a relative number. This can be expressed as percentage of nominal 340performance (infrastructure max). Below the nominal sustained performance level, 341desired performance expresses the average performance level of the processor subject 342to the Performance Reduction Tolerance register. Above the nominal performance level, 343processor must provide at least nominal performance requested and go higher if current 344operating conditions allow. 345 346Guided Mode 347----------- 348 349``amd_pstate=guided`` 350 351If ``amd_pstate=guided`` is passed to kernel command line option then this mode 352is activated. In this mode, driver requests minimum and maximum performance 353level and the platform autonomously selects a performance level in this range 354and appropriate to the current workload. 355 356``amd-pstate`` Preferred Core 357================================= 358 359The core frequency is subjected to the process variation in semiconductors. 360Not all cores are able to reach the maximum frequency respecting the 361infrastructure limits. Consequently, AMD has redefined the concept of 362maximum frequency of a part. This means that a fraction of cores can reach 363maximum frequency. To find the best process scheduling policy for a given 364scenario, OS needs to know the core ordering informed by the platform through 365highest performance capability register of the CPPC interface. 366 367``amd-pstate`` preferred core enables the scheduler to prefer scheduling on 368cores that can achieve a higher frequency with lower voltage. The preferred 369core rankings can dynamically change based on the workload, platform conditions, 370thermals and ageing. 371 372The priority metric will be initialized by the ``amd-pstate`` driver. The ``amd-pstate`` 373driver will also determine whether or not ``amd-pstate`` preferred core is 374supported by the platform. 375 376``amd-pstate`` driver will provide an initial core ordering when the system boots. 377The platform uses the CPPC interfaces to communicate the core ranking to the 378operating system and scheduler to make sure that OS is choosing the cores 379with highest performance firstly for scheduling the process. When ``amd-pstate`` 380driver receives a message with the highest performance change, it will 381update the core ranking and set the cpu's priority. 382 383``amd-pstate`` Preferred Core Switch 384===================================== 385Kernel Parameters 386----------------- 387 388``amd-pstate`` peferred core`` has two states: enable and disable. 389Enable/disable states can be chosen by different kernel parameters. 390Default enable ``amd-pstate`` preferred core. 391 392``amd_prefcore=disable`` 393 394For systems that support ``amd-pstate`` preferred core, the core rankings will 395always be advertised by the platform. But OS can choose to ignore that via the 396kernel parameter ``amd_prefcore=disable``. 397 398User Space Interface in ``sysfs`` - General 399=========================================== 400 401Global Attributes 402----------------- 403 404``amd-pstate`` exposes several global attributes (files) in ``sysfs`` to 405control its functionality at the system level. They are located in the 406``/sys/devices/system/cpu/amd_pstate/`` directory and affect all CPUs. 407 408``status`` 409 Operation mode of the driver: "active", "passive" or "disable". 410 411 "active" 412 The driver is functional and in the ``active mode`` 413 414 "passive" 415 The driver is functional and in the ``passive mode`` 416 417 "guided" 418 The driver is functional and in the ``guided mode`` 419 420 "disable" 421 The driver is unregistered and not functional now. 422 423 This attribute can be written to in order to change the driver's 424 operation mode or to unregister it. The string written to it must be 425 one of the possible values of it and, if successful, writing one of 426 these values to the sysfs file will cause the driver to switch over 427 to the operation mode represented by that string - or to be 428 unregistered in the "disable" case. 429 430``prefcore`` 431 Preferred core state of the driver: "enabled" or "disabled". 432 433 "enabled" 434 Enable the ``amd-pstate`` preferred core. 435 436 "disabled" 437 Disable the ``amd-pstate`` preferred core 438 439 440 This attribute is read-only to check the state of preferred core set 441 by the kernel parameter. 442 443``cpupower`` tool support for ``amd-pstate`` 444=============================================== 445 446``amd-pstate`` is supported by the ``cpupower`` tool, which can be used to dump 447frequency information. Development is in progress to support more and more 448operations for the new ``amd-pstate`` module with this tool. :: 449 450 root@hr-test1:/home/ray# cpupower frequency-info 451 analyzing CPU 0: 452 driver: amd-pstate 453 CPUs which run at the same hardware frequency: 0 454 CPUs which need to have their frequency coordinated by software: 0 455 maximum transition latency: 131 us 456 hardware limits: 400 MHz - 4.68 GHz 457 available cpufreq governors: ondemand conservative powersave userspace performance schedutil 458 current policy: frequency should be within 400 MHz and 4.68 GHz. 459 The governor "schedutil" may decide which speed to use 460 within this range. 461 current CPU frequency: Unable to call hardware 462 current CPU frequency: 4.02 GHz (asserted by call to kernel) 463 boost state support: 464 Supported: yes 465 Active: yes 466 AMD PSTATE Highest Performance: 166. Maximum Frequency: 4.68 GHz. 467 AMD PSTATE Nominal Performance: 117. Nominal Frequency: 3.30 GHz. 468 AMD PSTATE Lowest Non-linear Performance: 39. Lowest Non-linear Frequency: 1.10 GHz. 469 AMD PSTATE Lowest Performance: 15. Lowest Frequency: 400 MHz. 470 471 472Diagnostics and Tuning 473======================= 474 475Trace Events 476-------------- 477 478There are two static trace events that can be used for ``amd-pstate`` 479diagnostics. One of them is the ``cpu_frequency`` trace event generally used 480by ``CPUFreq``, and the other one is the ``amd_pstate_perf`` trace event 481specific to ``amd-pstate``. The following sequence of shell commands can 482be used to enable them and see their output (if the kernel is 483configured to support event tracing). :: 484 485 root@hr-test1:/home/ray# cd /sys/kernel/tracing/ 486 root@hr-test1:/sys/kernel/tracing# echo 1 > events/amd_cpu/enable 487 root@hr-test1:/sys/kernel/tracing# cat trace 488 # tracer: nop 489 # 490 # entries-in-buffer/entries-written: 47827/42233061 #P:2 491 # 492 # _-----=> irqs-off 493 # / _----=> need-resched 494 # | / _---=> hardirq/softirq 495 # || / _--=> preempt-depth 496 # ||| / delay 497 # TASK-PID CPU# |||| TIMESTAMP FUNCTION 498 # | | | |||| | | 499 <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 500 <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 501 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 502 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 503 <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 504 <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 505 <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 506 507The ``cpu_frequency`` trace event will be triggered either by the ``schedutil`` scaling 508governor (for the policies it is attached to), or by the ``CPUFreq`` core (for the 509policies with other scaling governors). 510 511 512Tracer Tool 513------------- 514 515``amd_pstate_tracer.py`` can record and parse ``amd-pstate`` trace log, then 516generate performance plots. This utility can be used to debug and tune the 517performance of ``amd-pstate`` driver. The tracer tool needs to import intel 518pstate tracer. 519 520Tracer tool located in ``linux/tools/power/x86/amd_pstate_tracer``. It can be 521used in two ways. If trace file is available, then directly parse the file 522with command :: 523 524 ./amd_pstate_trace.py [-c cpus] -t <trace_file> -n <test_name> 525 526Or generate trace file with root privilege, then parse and plot with command :: 527 528 sudo ./amd_pstate_trace.py [-c cpus] -n <test_name> -i <interval> [-m kbytes] 529 530The test result can be found in ``results/test_name``. Following is the example 531about part of the output. :: 532 533 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 534 CPU_005 712 116384 39 49 166 0.7565 9645075 2214891 38431470 25.1 11.646 469 2.496 kworker/5:0-40 535 CPU_006 712 116408 39 49 166 0.6769 8950227 1839034 37192089 24.06 11.272 470 2.496 kworker/6:0-1264 536 537Unit Tests for amd-pstate 538------------------------- 539 540``amd-pstate-ut`` is a test module for testing the ``amd-pstate`` driver. 541 542 * It can help all users to verify their processor support (SBIOS/Firmware or Hardware). 543 544 * Kernel can have a basic function test to avoid the kernel regression during the update. 545 546 * We can introduce more functional or performance tests to align the result together, it will benefit power and performance scale optimization. 547 5481. Test case descriptions 549 550 1). Basic tests 551 552 Test prerequisite and basic functions for the ``amd-pstate`` driver. 553 554 +---------+--------------------------------+------------------------------------------------------------------------------------+ 555 | Index | Functions | Description | 556 +=========+================================+====================================================================================+ 557 | 1 | amd_pstate_ut_acpi_cpc_valid || Check whether the _CPC object is present in SBIOS. | 558 | | || | 559 | | || The detail refer to `Processor Support <processor_support_>`_. | 560 +---------+--------------------------------+------------------------------------------------------------------------------------+ 561 | 2 | amd_pstate_ut_check_enabled || Check whether AMD P-State is enabled. | 562 | | || | 563 | | || AMD P-States and ACPI hardware P-States always can be supported in one processor. | 564 | | | But AMD P-States has the higher priority and if it is enabled with | 565 | | | :c:macro:`MSR_AMD_CPPC_ENABLE` or ``cppc_set_enable``, it will respond to the | 566 | | | request from AMD P-States. | 567 +---------+--------------------------------+------------------------------------------------------------------------------------+ 568 | 3 | amd_pstate_ut_check_perf || Check if the each performance values are reasonable. | 569 | | || highest_perf >= nominal_perf > lowest_nonlinear_perf > lowest_perf > 0. | 570 +---------+--------------------------------+------------------------------------------------------------------------------------+ 571 | 4 | amd_pstate_ut_check_freq || Check if the each frequency values and max freq when set support boost mode | 572 | | | are reasonable. | 573 | | || max_freq >= nominal_freq > lowest_nonlinear_freq > min_freq > 0 | 574 | | || If boost is not active but supported, this maximum frequency will be larger than | 575 | | | the one in ``cpuinfo``. | 576 +---------+--------------------------------+------------------------------------------------------------------------------------+ 577 578 2). Tbench test 579 580 Test and monitor the cpu changes when running tbench benchmark under the specified governor. 581 These changes include desire performance, frequency, load, performance, energy etc. 582 The specified governor is ondemand or schedutil. 583 Tbench can also be tested on the ``acpi-cpufreq`` kernel driver for comparison. 584 585 3). Gitsource test 586 587 Test and monitor the cpu changes when running gitsource benchmark under the specified governor. 588 These changes include desire performance, frequency, load, time, energy etc. 589 The specified governor is ondemand or schedutil. 590 Gitsource can also be tested on the ``acpi-cpufreq`` kernel driver for comparison. 591 592#. How to execute the tests 593 594 We use test module in the kselftest frameworks to implement it. 595 We create ``amd-pstate-ut`` module and tie it into kselftest.(for 596 details refer to Linux Kernel Selftests [4]_). 597 598 1). Build 599 600 + open the :c:macro:`CONFIG_X86_AMD_PSTATE` configuration option. 601 + set the :c:macro:`CONFIG_X86_AMD_PSTATE_UT` configuration option to M. 602 + make project 603 + make selftest :: 604 605 $ cd linux 606 $ make -C tools/testing/selftests 607 608 + make perf :: 609 610 $ cd tools/perf/ 611 $ make 612 613 614 2). Installation & Steps :: 615 616 $ make -C tools/testing/selftests install INSTALL_PATH=~/kselftest 617 $ cp tools/perf/perf /usr/bin/perf 618 $ sudo ./kselftest/run_kselftest.sh -c amd-pstate 619 620 3). Specified test case :: 621 622 $ cd ~/kselftest/amd-pstate 623 $ sudo ./run.sh -t basic 624 $ sudo ./run.sh -t tbench 625 $ sudo ./run.sh -t tbench -m acpi-cpufreq 626 $ sudo ./run.sh -t gitsource 627 $ sudo ./run.sh -t gitsource -m acpi-cpufreq 628 $ ./run.sh --help 629 ./run.sh: illegal option -- - 630 Usage: ./run.sh [OPTION...] 631 [-h <help>] 632 [-o <output-file-for-dump>] 633 [-c <all: All testing, 634 basic: Basic testing, 635 tbench: Tbench testing, 636 gitsource: Gitsource testing.>] 637 [-t <tbench time limit>] 638 [-p <tbench process number>] 639 [-l <loop times for tbench>] 640 [-i <amd tracer interval>] 641 [-m <comparative test: acpi-cpufreq>] 642 643 644 4). Results 645 646 + basic 647 648 When you finish test, you will get the following log info :: 649 650 $ dmesg | grep "amd_pstate_ut" | tee log.txt 651 [12977.570663] amd_pstate_ut: 1 amd_pstate_ut_acpi_cpc_valid success! 652 [12977.570673] amd_pstate_ut: 2 amd_pstate_ut_check_enabled success! 653 [12977.571207] amd_pstate_ut: 3 amd_pstate_ut_check_perf success! 654 [12977.571212] amd_pstate_ut: 4 amd_pstate_ut_check_freq success! 655 656 + tbench 657 658 When you finish test, you will get selftest.tbench.csv and png images. 659 The selftest.tbench.csv file contains the raw data and the drop of the comparative test. 660 The png images shows the performance, energy and performan per watt of each test. 661 Open selftest.tbench.csv : 662 663 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 664 + Governor | Round | Des-perf | Freq | Load | Performance | Energy | Performance Per Watt | 665 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 666 + Unit | | | GHz | | MB/s | J | MB/J | 667 +=================================================+==============+==========+=========+==========+=============+=========+======================+ 668 + amd-pstate-ondemand | 1 | | | | 2504.05 | 1563.67 | 158.5378 | 669 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 670 + amd-pstate-ondemand | 2 | | | | 2243.64 | 1430.32 | 155.2941 | 671 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 672 + amd-pstate-ondemand | 3 | | | | 2183.88 | 1401.32 | 154.2860 | 673 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 674 + amd-pstate-ondemand | Average | | | | 2310.52 | 1465.1 | 156.1268 | 675 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 676 + amd-pstate-schedutil | 1 | 165.329 | 1.62257 | 99.798 | 2136.54 | 1395.26 | 151.5971 | 677 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 678 + amd-pstate-schedutil | 2 | 166 | 1.49761 | 99.9993 | 2100.56 | 1380.5 | 150.6377 | 679 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 680 + amd-pstate-schedutil | 3 | 166 | 1.47806 | 99.9993 | 2084.12 | 1375.76 | 149.9737 | 681 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 682 + amd-pstate-schedutil | Average | 165.776 | 1.53275 | 99.9322 | 2107.07 | 1383.84 | 150.7399 | 683 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 684 + acpi-cpufreq-ondemand | 1 | | | | 2529.9 | 1564.4 | 160.0997 | 685 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 686 + acpi-cpufreq-ondemand | 2 | | | | 2249.76 | 1432.97 | 155.4297 | 687 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 688 + acpi-cpufreq-ondemand | 3 | | | | 2181.46 | 1406.88 | 153.5060 | 689 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 690 + acpi-cpufreq-ondemand | Average | | | | 2320.37 | 1468.08 | 156.4741 | 691 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 692 + acpi-cpufreq-schedutil | 1 | | | | 2137.64 | 1385.24 | 152.7723 | 693 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 694 + acpi-cpufreq-schedutil | 2 | | | | 2107.05 | 1372.23 | 152.0138 | 695 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 696 + acpi-cpufreq-schedutil | 3 | | | | 2085.86 | 1365.35 | 151.2433 | 697 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 698 + acpi-cpufreq-schedutil | Average | | | | 2110.18 | 1374.27 | 152.0136 | 699 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 700 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | -9.0584 | -6.3899 | -2.8506 | 701 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 702 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | | | | 8.8053 | -5.5463 | -3.4503 | 703 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 704 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | -0.4245 | -0.2029 | -0.2219 | 705 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 706 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | -0.1473 | 0.6963 | -0.8378 | 707 +-------------------------------------------------+--------------+----------+---------+----------+-------------+---------+----------------------+ 708 709 + gitsource 710 711 When you finish test, you will get selftest.gitsource.csv and png images. 712 The selftest.gitsource.csv file contains the raw data and the drop of the comparative test. 713 The png images shows the performance, energy and performan per watt of each test. 714 Open selftest.gitsource.csv : 715 716 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 717 + Governor | Round | Des-perf | Freq | Load | Time | Energy | Performance Per Watt | 718 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 719 + Unit | | | GHz | | s | J | 1/J | 720 +=================================================+==============+==========+==========+==========+=============+=========+======================+ 721 + amd-pstate-ondemand | 1 | 50.119 | 2.10509 | 23.3076 | 475.69 | 865.78 | 0.001155027 | 722 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 723 + amd-pstate-ondemand | 2 | 94.8006 | 1.98771 | 56.6533 | 467.1 | 839.67 | 0.001190944 | 724 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 725 + amd-pstate-ondemand | 3 | 76.6091 | 2.53251 | 43.7791 | 467.69 | 855.85 | 0.001168429 | 726 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 727 + amd-pstate-ondemand | Average | 73.8429 | 2.20844 | 41.2467 | 470.16 | 853.767 | 0.001171279 | 728 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 729 + amd-pstate-schedutil | 1 | 165.919 | 1.62319 | 98.3868 | 464.17 | 866.8 | 0.001153668 | 730 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 731 + amd-pstate-schedutil | 2 | 165.97 | 1.31309 | 99.5712 | 480.15 | 880.4 | 0.001135847 | 732 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 733 + amd-pstate-schedutil | 3 | 165.973 | 1.28448 | 99.9252 | 481.79 | 867.02 | 0.001153375 | 734 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 735 + amd-pstate-schedutil | Average | 165.954 | 1.40692 | 99.2944 | 475.37 | 871.407 | 0.001147569 | 736 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 737 + acpi-cpufreq-ondemand | 1 | | | | 2379.62 | 742.96 | 0.001345967 | 738 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 739 + acpi-cpufreq-ondemand | 2 | | | | 441.74 | 817.49 | 0.001223256 | 740 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 741 + acpi-cpufreq-ondemand | 3 | | | | 455.48 | 820.01 | 0.001219497 | 742 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 743 + acpi-cpufreq-ondemand | Average | | | | 425.613 | 793.487 | 0.001260260 | 744 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 745 + acpi-cpufreq-schedutil | 1 | | | | 459.69 | 838.54 | 0.001192548 | 746 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 747 + acpi-cpufreq-schedutil | 2 | | | | 466.55 | 830.89 | 0.001203528 | 748 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 749 + acpi-cpufreq-schedutil | 3 | | | | 470.38 | 837.32 | 0.001194286 | 750 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 751 + acpi-cpufreq-schedutil | Average | | | | 465.54 | 835.583 | 0.001196769 | 752 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 753 + acpi-cpufreq-ondemand VS acpi-cpufreq-schedutil | Comprison(%) | | | | 9.3810 | 5.3051 | -5.0379 | 754 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 755 + amd-pstate-ondemand VS amd-pstate-schedutil | Comprison(%) | 124.7392 | -36.2934 | 140.7329 | 1.1081 | 2.0661 | -2.0242 | 756 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 757 + acpi-cpufreq-ondemand VS amd-pstate-ondemand | Comprison(%) | | | | 10.4665 | 7.5968 | -7.0605 | 758 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 759 + acpi-cpufreq-schedutil VS amd-pstate-schedutil | Comprison(%) | | | | 2.1115 | 4.2873 | -4.1110 | 760 +-------------------------------------------------+--------------+----------+----------+----------+-------------+---------+----------------------+ 761 762Reference 763=========== 764 765.. [1] AMD64 Architecture Programmer's Manual Volume 2: System Programming, 766 https://www.amd.com/system/files/TechDocs/24593.pdf 767 768.. [2] Advanced Configuration and Power Interface Specification, 769 https://uefi.org/sites/default/files/resources/ACPI_Spec_6_4_Jan22.pdf 770 771.. [3] Processor Programming Reference (PPR) for AMD Family 19h Model 51h, Revision A1 Processors 772 https://www.amd.com/system/files/TechDocs/56569-A1-PUB.zip 773 774.. [4] Linux Kernel Selftests, 775 https://www.kernel.org/doc/html/latest/dev-tools/kselftest.html 776