Documentation/ABI/testing/sysfs-devices-system-cpu

   1 What:           /sys/devices/system/cpu/
   2 Date:           pre-git history
   3 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
   4 Description:
   5                 A collection of both global and individual CPU attributes
   6
   7                 Individual CPU attributes are contained in subdirectories
   8                 named by the kernel's logical CPU number, e.g.:
   9
  10                 /sys/devices/system/cpu/cpu#/
  11
  12 What:           /sys/devices/system/cpu/kernel_max
  13                 /sys/devices/system/cpu/offline
  14                 /sys/devices/system/cpu/online
  15                 /sys/devices/system/cpu/possible
  16                 /sys/devices/system/cpu/present
  17 Date:           December 2008
  18 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
  19 Description:    CPU topology files that describe kernel limits related to
  20                 hotplug. Briefly:
  21
  22                 kernel_max: the maximum cpu index allowed by the kernel
  23                 configuration.
  24
  25                 offline: cpus that are not online because they have been
  26                 HOTPLUGGED off or exceed the limit of cpus allowed by the
  27                 kernel configuration (kernel_max above).
  28
  29                 online: cpus that are online and being scheduled.
  30
  31                 possible: cpus that have been allocated resources and can be
  32                 brought online if they are present.
  33
  34                 present: cpus that have been identified as being present in
  35                 the system.
  36
  37                 See Documentation/admin-guide/cputopology.rst for more information.
  38
  39
  40 What:           /sys/devices/system/cpu/probe
  41                 /sys/devices/system/cpu/release
  42 Date:           November 2009
  43 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
  44 Description:    Dynamic addition and removal of CPU's.  This is not hotplug
  45                 removal, this is meant complete removal/addition of the CPU
  46                 from the system.
  47
  48                 probe: writes to this file will dynamically add a CPU to the
  49                 system.  Information written to the file to add CPU's is
  50                 architecture specific.
  51
  52                 release: writes to this file dynamically remove a CPU from
  53                 the system.  Information writtento the file to remove CPU's
  54                 is architecture specific.
  55
  56 What:           /sys/devices/system/cpu/cpu#/node
  57 Date:           October 2009
  58 Contact:        Linux memory management mailing list <linux-mm@kvack.org>
  59 Description:    Discover NUMA node a CPU belongs to
  60
  61                 When CONFIG_NUMA is enabled, a symbolic link that points
  62                 to the corresponding NUMA node directory.
  63
  64                 For example, the following symlink is created for cpu42
  65                 in NUMA node 2:
  66
  67                 /sys/devices/system/cpu/cpu42/node2 -> ../../node/node2
  68
  69
  70 What:           /sys/devices/system/cpu/cpu#/topology/core_id
  71                 /sys/devices/system/cpu/cpu#/topology/core_siblings
  72                 /sys/devices/system/cpu/cpu#/topology/core_siblings_list
  73                 /sys/devices/system/cpu/cpu#/topology/physical_package_id
  74                 /sys/devices/system/cpu/cpu#/topology/thread_siblings
  75                 /sys/devices/system/cpu/cpu#/topology/thread_siblings_list
  76 Date:           December 2008
  77 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
  78 Description:    CPU topology files that describe a logical CPU's relationship
  79                 to other cores and threads in the same physical package.
  80
  81                 One cpu# directory is created per logical CPU in the system,
  82                 e.g. /sys/devices/system/cpu/cpu42/.
  83
  84                 Briefly, the files above are:
  85
  86                 core_id: the CPU core ID of cpu#. Typically it is the
  87                 hardware platform's identifier (rather than the kernel's).
  88                 The actual value is architecture and platform dependent.
  89
  90                 core_siblings: internal kernel map of cpu#'s hardware threads
  91                 within the same physical_package_id.
  92
  93                 core_siblings_list: human-readable list of the logical CPU
  94                 numbers within the same physical_package_id as cpu#.
  95
  96                 physical_package_id: physical package id of cpu#. Typically
  97                 corresponds to a physical socket number, but the actual value
  98                 is architecture and platform dependent.
  99
 100                 thread_siblings: internel kernel map of cpu#'s hardware
 101                 threads within the same core as cpu#
 102
 103                 thread_siblings_list: human-readable list of cpu#'s hardware
 104                 threads within the same core as cpu#
 105
 106                 See Documentation/admin-guide/cputopology.rst for more information.
 107
 108
 109 What:           /sys/devices/system/cpu/cpuidle/available_governors
 110                 /sys/devices/system/cpu/cpuidle/current_driver
 111                 /sys/devices/system/cpu/cpuidle/current_governor
 112                 /sys/devices/system/cpu/cpuidle/current_governer_ro
 113 Date:           September 2007
 114 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 115 Description:    Discover cpuidle policy and mechanism
 116
 117                 Various CPUs today support multiple idle levels that are
 118                 differentiated by varying exit latencies and power
 119                 consumption during idle.
 120
 121                 Idle policy (governor) is differentiated from idle mechanism
 122                 (driver).
 123
 124                 available_governors: (RO) displays a space separated list of
 125                 available governors.
 126
 127                 current_driver: (RO) displays current idle mechanism.
 128
 129                 current_governor: (RW) displays current idle policy. Users can
 130                 switch the governor at runtime by writing to this file.
 131
 132                 current_governor_ro: (RO) displays current idle policy.
 133
 134                 See Documentation/admin-guide/pm/cpuidle.rst and
 135                 Documentation/driver-api/pm/cpuidle.rst for more information.
 136
 137
 138 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/name
 139                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/latency
 140                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/power
 141                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/time
 142                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/usage
 143                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/above
 144                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/below
 145 Date:           September 2007
 146 KernelVersion:  v2.6.24
 147 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 148 Description:
 149                 The directory /sys/devices/system/cpu/cpuX/cpuidle contains per
 150                 logical CPU specific cpuidle information for each online cpu X.
 151                 The processor idle states which are available for use have the
 152                 following attributes:
 153
 154                 name: (RO) Name of the idle state (string).
 155
 156                 latency: (RO) The latency to exit out of this idle state (in
 157                 microseconds).
 158
 159                 power: (RO) The power consumed while in this idle state (in
 160                 milliwatts).
 161
 162                 time: (RO) The total time spent in this idle state (in microseconds).
 163
 164                 usage: (RO) Number of times this state was entered (a count).
 165
 166                 above: (RO) Number of times this state was entered, but the
 167                        observed CPU idle duration was too short for it (a count).
 168
 169                 below: (RO) Number of times this state was entered, but the
 170                        observed CPU idle duration was too long for it (a count).
 171
 172 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/desc
 173 Date:           February 2008
 174 KernelVersion:  v2.6.25
 175 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 176 Description:
 177                 (RO) A small description about the idle state (string).
 178
 179
 180 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/disable
 181 Date:           March 2012
 182 KernelVersion:  v3.10
 183 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 184 Description:
 185                 (RW) Option to disable this idle state (bool). The behavior and
 186                 the effect of the disable variable depends on the implementation
 187                 of a particular governor. In the ladder governor, for example,
 188                 it is not coherent, i.e. if one is disabling a light state, then
 189                 all deeper states are disabled as well, but the disable variable
 190                 does not reflect it. Likewise, if one enables a deep state but a
 191                 lighter state still is disabled, then this has no effect.
 192
 193 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/default_status
 194 Date:           December 2019
 195 KernelVersion:  v5.6
 196 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 197 Description:
 198                 (RO) The default status of this state, "enabled" or "disabled".
 199
 200 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/residency
 201 Date:           March 2014
 202 KernelVersion:  v3.15
 203 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 204 Description:
 205                 (RO) Display the target residency i.e. the minimum amount of
 206                 time (in microseconds) this cpu should spend in this idle state
 207                 to make the transition worth the effort.
 208
 209 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/
 210 Date:           March 2018
 211 KernelVersion:  v4.17
 212 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 213 Description:
 214                 Idle state usage statistics related to suspend-to-idle.
 215
 216                 This attribute group is only present for states that can be
 217                 used in suspend-to-idle with suspended timekeeping.
 218
 219 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/time
 220 Date:           March 2018
 221 KernelVersion:  v4.17
 222 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 223 Description:
 224                 Total time spent by the CPU in suspend-to-idle (with scheduler
 225                 tick suspended) after requesting this state.
 226
 227 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/usage
 228 Date:           March 2018
 229 KernelVersion:  v4.17
 230 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 231 Description:
 232                 Total number of times this state has been requested by the CPU
 233                 while entering suspend-to-idle.
 234
 235 What:           /sys/devices/system/cpu/cpu#/cpufreq/*
 236 Date:           pre-git history
 237 Contact:        linux-pm@vger.kernel.org
 238 Description:    Discover and change clock speed of CPUs
 239
 240                 Clock scaling allows you to change the clock speed of the
 241                 CPUs on the fly. This is a nice method to save battery
 242                 power, because the lower the clock speed, the less power
 243                 the CPU consumes.
 244
 245                 There are many knobs to tweak in this directory.
 246
 247                 See files in Documentation/cpu-freq/ for more information.
 248
 249
 250 What:           /sys/devices/system/cpu/cpu#/cpufreq/freqdomain_cpus
 251 Date:           June 2013
 252 Contact:        linux-pm@vger.kernel.org
 253 Description:    Discover CPUs in the same CPU frequency coordination domain
 254
 255                 freqdomain_cpus is the list of CPUs (online+offline) that share
 256                 the same clock/freq domain (possibly at the hardware level).
 257                 That information may be hidden from the cpufreq core and the
 258                 value of related_cpus may be different from freqdomain_cpus. This
 259                 attribute is useful for user space DVFS controllers to get better
 260                 power/performance results for platforms using acpi-cpufreq.
 261
 262                 This file is only present if the acpi-cpufreq driver is in use.
 263
 264
 265 What:           /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
 266 Date:           August 2008
 267 KernelVersion:  2.6.27
 268 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 269 Description:    Disable L3 cache indices
 270
 271                 These files exist in every CPU's cache/index3 directory. Each
 272                 cache_disable_{0,1} file corresponds to one disable slot which
 273                 can be used to disable a cache index. Reading from these files
 274                 on a processor with this functionality will return the currently
 275                 disabled index for that node. There is one L3 structure per
 276                 node, or per internal node on MCM machines. Writing a valid
 277                 index to one of these files will cause the specificed cache
 278                 index to be disabled.
 279
 280                 All AMD processors with L3 caches provide this functionality.
 281                 For details, see BKDGs at
 282                 http://developer.amd.com/documentation/guides/Pages/default.aspx
 283
 284
 285 What:           /sys/devices/system/cpu/cpufreq/boost
 286 Date:           August 2012
 287 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 288 Description:    Processor frequency boosting control
 289
 290                 This switch controls the boost setting for the whole system.
 291                 Boosting allows the CPU and the firmware to run at a frequency
 292                 beyound it's nominal limit.
 293                 More details can be found in
 294                 Documentation/admin-guide/pm/cpufreq.rst
 295
 296
 297 What:           /sys/devices/system/cpu/cpu#/crash_notes
 298                 /sys/devices/system/cpu/cpu#/crash_notes_size
 299 Date:           April 2013
 300 Contact:        kexec@lists.infradead.org
 301 Description:    address and size of the percpu note.
 302
 303                 crash_notes: the physical address of the memory that holds the
 304                 note of cpu#.
 305
 306                 crash_notes_size: size of the note of cpu#.
 307
 308
 309 What:           /sys/devices/system/cpu/intel_pstate/max_perf_pct
 310                 /sys/devices/system/cpu/intel_pstate/min_perf_pct
 311                 /sys/devices/system/cpu/intel_pstate/no_turbo
 312 Date:           February 2013
 313 Contact:        linux-pm@vger.kernel.org
 314 Description:    Parameters for the Intel P-state driver
 315
 316                 Logic for selecting the current P-state in Intel
 317                 Sandybridge+ processors. The three knobs control
 318                 limits for the P-state that will be requested by the
 319                 driver.
 320
 321                 max_perf_pct: limits the maximum P state that will be requested by
 322                 the driver stated as a percentage of the available performance.
 323
 324                 min_perf_pct: limits the minimum P state that will be requested by
 325                 the driver stated as a percentage of the available performance.
 326
 327                 no_turbo: limits the driver to selecting P states below the turbo
 328                 frequency range.
 329
 330                 More details can be found in
 331                 Documentation/admin-guide/pm/intel_pstate.rst
 332
 333 What:           /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
 334 Date:           July 2014(documented, existed before August 2008)
 335 Contact:        Sudeep Holla <sudeep.holla@arm.com>
 336                 Linux kernel mailing list <linux-kernel@vger.kernel.org>
 337 Description:    Parameters for the CPU cache attributes
 338
 339                 allocation_policy:
 340                         - WriteAllocate: allocate a memory location to a cache line
 341                                          on a cache miss because of a write
 342                         - ReadAllocate: allocate a memory location to a cache line
 343                                         on a cache miss because of a read
 344                         - ReadWriteAllocate: both writeallocate and readallocate
 345
 346                 attributes: LEGACY used only on IA64 and is same as write_policy
 347
 348                 coherency_line_size: the minimum amount of data in bytes that gets
 349                                      transferred from memory to cache
 350
 351                 level: the cache hierarchy in the multi-level cache configuration
 352
 353                 number_of_sets: total number of sets in the cache, a set is a
 354                                 collection of cache lines with the same cache index
 355
 356                 physical_line_partition: number of physical cache line per cache tag
 357
 358                 shared_cpu_list: the list of logical cpus sharing the cache
 359
 360                 shared_cpu_map: logical cpu mask containing the list of cpus sharing
 361                                 the cache
 362
 363                 size: the total cache size in kB
 364
 365                 type:
 366                         - Instruction: cache that only holds instructions
 367                         - Data: cache that only caches data
 368                         - Unified: cache that holds both data and instructions
 369
 370                 ways_of_associativity: degree of freedom in placing a particular block
 371                                         of memory in the cache
 372
 373                 write_policy:
 374                         - WriteThrough: data is written to both the cache line
 375                                         and to the block in the lower-level memory
 376                         - WriteBack: data is written only to the cache line and
 377                                      the modified cache line is written to main
 378                                      memory only when it is replaced
 379
 380
 381 What:           /sys/devices/system/cpu/cpu*/cache/index*/id
 382 Date:           September 2016
 383 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 384 Description:    Cache id
 385
 386                 The id provides a unique number for a specific instance of
 387                 a cache of a particular type. E.g. there may be a level
 388                 3 unified cache on each socket in a server and we may
 389                 assign them ids 0, 1, 2, ...
 390
 391                 Note that id value can be non-contiguous. E.g. level 1
 392                 caches typically exist per core, but there may not be a
 393                 power of two cores on a socket, so these caches may be
 394                 numbered 0, 1, 2, 3, 4, 5, 8, 9, 10, ...
 395
 396 What:           /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats
 397                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
 398                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat
 399                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
 400                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
 401                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
 402                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
 403                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
 404                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset
 405 Date:           March 2016
 406 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 407                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 408 Description:    POWERNV CPUFreq driver's frequency throttle stats directory and
 409                 attributes
 410
 411                 'cpuX/cpufreq/throttle_stats' directory contains the CPU frequency
 412                 throttle stat attributes for the chip. The throttle stats of a cpu
 413                 is common across all the cpus belonging to a chip. Below are the
 414                 throttle attributes exported in the 'throttle_stats' directory:
 415
 416                 - turbo_stat : This file gives the total number of times the max
 417                 frequency is throttled to lower frequency in turbo (at and above
 418                 nominal frequency) range of frequencies.
 419
 420                 - sub_turbo_stat : This file gives the total number of times the
 421                 max frequency is throttled to lower frequency in sub-turbo(below
 422                 nominal frequency) range of frequencies.
 423
 424                 - unthrottle : This file gives the total number of times the max
 425                 frequency is unthrottled after being throttled.
 426
 427                 - powercap : This file gives the total number of times the max
 428                 frequency is throttled due to 'Power Capping'.
 429
 430                 - overtemp : This file gives the total number of times the max
 431                 frequency is throttled due to 'CPU Over Temperature'.
 432
 433                 - supply_fault : This file gives the total number of times the
 434                 max frequency is throttled due to 'Power Supply Failure'.
 435
 436                 - overcurrent : This file gives the total number of times the
 437                 max frequency is throttled due to 'Overcurrent'.
 438
 439                 - occ_reset : This file gives the total number of times the max
 440                 frequency is throttled due to 'OCC Reset'.
 441
 442                 The sysfs attributes representing different throttle reasons like
 443                 powercap, overtemp, supply_fault, overcurrent and occ_reset map to
 444                 the reasons provided by OCC firmware for throttling the frequency.
 445
 446 What:           /sys/devices/system/cpu/cpufreq/policyX/throttle_stats
 447                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat
 448                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat
 449                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle
 450                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap
 451                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp
 452                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault
 453                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent
 454                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset
 455 Date:           March 2016
 456 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 457                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 458 Description:    POWERNV CPUFreq driver's frequency throttle stats directory and
 459                 attributes
 460
 461                 'policyX/throttle_stats' directory and all the attributes are same as
 462                 the /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory and
 463                 attributes which give the frequency throttle information of the chip.
 464
 465 What:           /sys/devices/system/cpu/cpuX/regs/
 466                 /sys/devices/system/cpu/cpuX/regs/identification/
 467                 /sys/devices/system/cpu/cpuX/regs/identification/midr_el1
 468                 /sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
 469 Date:           June 2016
 470 Contact:        Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
 471 Description:    AArch64 CPU registers
 472                 'identification' directory exposes the CPU ID registers for
 473                  identifying model and revision of the CPU.
 474
 475 What:           /sys/devices/system/cpu/cpu#/cpu_capacity
 476 Date:           December 2016
 477 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 478 Description:    information about CPUs heterogeneity.
 479
 480                 cpu_capacity: capacity of cpu#.
 481
 482 What:           /sys/devices/system/cpu/vulnerabilities
 483                 /sys/devices/system/cpu/vulnerabilities/meltdown
 484                 /sys/devices/system/cpu/vulnerabilities/spectre_v1
 485                 /sys/devices/system/cpu/vulnerabilities/spectre_v2
 486                 /sys/devices/system/cpu/vulnerabilities/spec_store_bypass
 487                 /sys/devices/system/cpu/vulnerabilities/l1tf
 488                 /sys/devices/system/cpu/vulnerabilities/mds
 489                 /sys/devices/system/cpu/vulnerabilities/srbds
 490                 /sys/devices/system/cpu/vulnerabilities/tsx_async_abort
 491                 /sys/devices/system/cpu/vulnerabilities/itlb_multihit
 492 Date:           January 2018
 493 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 494 Description:    Information about CPU vulnerabilities
 495
 496                 The files are named after the code names of CPU
 497                 vulnerabilities. The output of those files reflects the
 498                 state of the CPUs in the system. Possible output values:
 499
 500                 "Not affected"    CPU is not affected by the vulnerability
 501                 "Vulnerable"      CPU is affected and no mitigation in effect
 502                 "Mitigation: $M"  CPU is affected and mitigation $M is in effect
 503
 504                 See also: Documentation/admin-guide/hw-vuln/index.rst
 505
 506 What:           /sys/devices/system/cpu/smt
 507                 /sys/devices/system/cpu/smt/active
 508                 /sys/devices/system/cpu/smt/control
 509 Date:           June 2018
 510 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 511 Description:    Control Symetric Multi Threading (SMT)
 512
 513                 active:  Tells whether SMT is active (enabled and siblings online)
 514
 515                 control: Read/write interface to control SMT. Possible
 516                          values:
 517
 518                          "on"             SMT is enabled
 519                          "off"            SMT is disabled
 520                          "forceoff"       SMT is force disabled. Cannot be changed.
 521                          "notsupported"   SMT is not supported by the CPU
 522                          "notimplemented" SMT runtime toggling is not
 523                                           implemented for the architecture
 524
 525                          If control status is "forceoff" or "notsupported" writes
 526                          are rejected.
 527
 528 What:           /sys/devices/system/cpu/cpu#/power/energy_perf_bias
 529 Date:           March 2019
 530 Contact:        linux-pm@vger.kernel.org
 531 Description:    Intel Energy and Performance Bias Hint (EPB)
 532
 533                 EPB for the given CPU in a sliding scale 0 - 15, where a value
 534                 of 0 corresponds to a hint preference for highest performance
 535                 and a value of 15 corresponds to the maximum energy savings.
 536
 537                 In order to change the EPB value for the CPU, write either
 538                 a number in the 0 - 15 sliding scale above, or one of the
 539                 strings: "performance", "balance-performance", "normal",
 540                 "balance-power", "power" (that represent values reflected by
 541                 their meaning), to this attribute.
 542
 543                 This attribute is present for all online CPUs supporting the
 544                 Intel EPB feature.
 545
 546 What:           /sys/devices/system/cpu/umwait_control
 547                 /sys/devices/system/cpu/umwait_control/enable_c02
 548                 /sys/devices/system/cpu/umwait_control/max_time
 549 Date:           May 2019
 550 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 551 Description:    Umwait control
 552
 553                 enable_c02: Read/write interface to control umwait C0.2 state
 554                         Read returns C0.2 state status:
 555                                 0: C0.2 is disabled
 556                                 1: C0.2 is enabled
 557
 558                         Write 'y' or '1'  or 'on' to enable C0.2 state.
 559                         Write 'n' or '0'  or 'off' to disable C0.2 state.
 560
 561                         The interface is case insensitive.
 562
 563                 max_time: Read/write interface to control umwait maximum time
 564                           in TSC-quanta that the CPU can reside in either C0.1
 565                           or C0.2 state. The time is an unsigned 32-bit number.
 566                           Note that a value of zero means there is no limit.
 567                           Low order two bits must be zero.
 568
 569 What:           /sys/devices/system/cpu/svm
 570 Date:           August 2019
 571 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 572                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 573 Description:    Secure Virtual Machine
 574
 575                 If 1, it means the system is using the Protected Execution
 576                 Facility in POWER9 and newer processors. i.e., it is a Secure
 577                 Virtual Machine.
 578
 579 What:           /sys/devices/system/cpu/cpuX/purr
 580 Date:           Apr 2005
 581 Contact:        Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 582 Description:    PURR ticks for this CPU since the system boot.
 583
 584                 The Processor Utilization Resources Register (PURR) is
 585                 a 64-bit counter which provides an estimate of the
 586                 resources used by the CPU thread. The contents of this
 587                 register increases monotonically. This sysfs interface
 588                 exposes the number of PURR ticks for cpuX.
 589
 590 What:           /sys/devices/system/cpu/cpuX/spurr
 591 Date:           Dec 2006
 592 Contact:        Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 593 Description:    SPURR ticks for this CPU since the system boot.
 594
 595                 The Scaled Processor Utilization Resources Register
 596                 (SPURR) is a 64-bit counter that provides a frequency
 597                 invariant estimate of the resources used by the CPU
 598                 thread. The contents of this register increases
 599                 monotonically. This sysfs interface exposes the number
 600                 of SPURR ticks for cpuX.
 601
 602 What:           /sys/devices/system/cpu/cpuX/idle_purr
 603 Date:           Apr 2020
 604 Contact:        Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 605 Description:    PURR ticks for cpuX when it was idle.
 606
 607                 This sysfs interface exposes the number of PURR ticks
 608                 for cpuX when it was idle.
 609
 610 What:           /sys/devices/system/cpu/cpuX/idle_spurr
 611 Date:           Apr 2020
 612 Contact:        Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 613 Description:    SPURR ticks for cpuX when it was idle.
 614
 615                 This sysfs interface exposes the number of SPURR ticks
 616                 for cpuX when it was idle.