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/cputopology.txt 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/cputopology.txt for more information.
 107
 108
 109 What:           /sys/devices/system/cpu/cpuidle/current_driver
 110                 /sys/devices/system/cpu/cpuidle/current_governer_ro
 111                 /sys/devices/system/cpu/cpuidle/available_governors
 112                 /sys/devices/system/cpu/cpuidle/current_governor
 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                 current_driver: (RO) displays current idle mechanism
 125
 126                 current_governor_ro: (RO) displays current idle policy
 127
 128                 With the cpuidle_sysfs_switch boot option enabled (meant for
 129                 developer testing), the following three attributes are visible
 130                 instead:
 131
 132                 current_driver: same as described above
 133
 134                 available_governors: (RO) displays a space separated list of
 135                 available governors
 136
 137                 current_governor: (RW) displays current idle policy. Users can
 138                 switch the governor at runtime by writing to this file.
 139
 140                 See files in Documentation/cpuidle/ for more information.
 141
 142
 143 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/name
 144                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/latency
 145                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/power
 146                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/time
 147                 /sys/devices/system/cpu/cpuX/cpuidle/stateN/usage
 148 Date:           September 2007
 149 KernelVersion:  v2.6.24
 150 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 151 Description:
 152                 The directory /sys/devices/system/cpu/cpuX/cpuidle contains per
 153                 logical CPU specific cpuidle information for each online cpu X.
 154                 The processor idle states which are available for use have the
 155                 following attributes:
 156
 157                 name: (RO) Name of the idle state (string).
 158
 159                 latency: (RO) The latency to exit out of this idle state (in
 160                 microseconds).
 161
 162                 power: (RO) The power consumed while in this idle state (in
 163                 milliwatts).
 164
 165                 time: (RO) The total time spent in this idle state (in microseconds).
 166
 167                 usage: (RO) Number of times this state was entered (a count).
 168
 169
 170 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/desc
 171 Date:           February 2008
 172 KernelVersion:  v2.6.25
 173 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 174 Description:
 175                 (RO) A small description about the idle state (string).
 176
 177
 178 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/disable
 179 Date:           March 2012
 180 KernelVersion:  v3.10
 181 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 182 Description:
 183                 (RW) Option to disable this idle state (bool). The behavior and
 184                 the effect of the disable variable depends on the implementation
 185                 of a particular governor. In the ladder governor, for example,
 186                 it is not coherent, i.e. if one is disabling a light state, then
 187                 all deeper states are disabled as well, but the disable variable
 188                 does not reflect it. Likewise, if one enables a deep state but a
 189                 lighter state still is disabled, then this has no effect.
 190
 191
 192 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/residency
 193 Date:           March 2014
 194 KernelVersion:  v3.15
 195 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 196 Description:
 197                 (RO) Display the target residency i.e. the minimum amount of
 198                 time (in microseconds) this cpu should spend in this idle state
 199                 to make the transition worth the effort.
 200
 201 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/
 202 Date:           March 2018
 203 KernelVersion:  v4.17
 204 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 205 Description:
 206                 Idle state usage statistics related to suspend-to-idle.
 207
 208                 This attribute group is only present for states that can be
 209                 used in suspend-to-idle with suspended timekeeping.
 210
 211 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/time
 212 Date:           March 2018
 213 KernelVersion:  v4.17
 214 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 215 Description:
 216                 Total time spent by the CPU in suspend-to-idle (with scheduler
 217                 tick suspended) after requesting this state.
 218
 219 What:           /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/usage
 220 Date:           March 2018
 221 KernelVersion:  v4.17
 222 Contact:        Linux power management list <linux-pm@vger.kernel.org>
 223 Description:
 224                 Total number of times this state has been requested by the CPU
 225                 while entering suspend-to-idle.
 226
 227 What:           /sys/devices/system/cpu/cpu#/cpufreq/*
 228 Date:           pre-git history
 229 Contact:        linux-pm@vger.kernel.org
 230 Description:    Discover and change clock speed of CPUs
 231
 232                 Clock scaling allows you to change the clock speed of the
 233                 CPUs on the fly. This is a nice method to save battery
 234                 power, because the lower the clock speed, the less power
 235                 the CPU consumes.
 236
 237                 There are many knobs to tweak in this directory.
 238
 239                 See files in Documentation/cpu-freq/ for more information.
 240
 241                 In particular, read Documentation/cpu-freq/user-guide.txt
 242                 to learn how to control the knobs.
 243
 244
 245 What:           /sys/devices/system/cpu/cpu#/cpufreq/freqdomain_cpus
 246 Date:           June 2013
 247 Contact:        linux-pm@vger.kernel.org
 248 Description:    Discover CPUs in the same CPU frequency coordination domain
 249
 250                 freqdomain_cpus is the list of CPUs (online+offline) that share
 251                 the same clock/freq domain (possibly at the hardware level).
 252                 That information may be hidden from the cpufreq core and the
 253                 value of related_cpus may be different from freqdomain_cpus. This
 254                 attribute is useful for user space DVFS controllers to get better
 255                 power/performance results for platforms using acpi-cpufreq.
 256
 257                 This file is only present if the acpi-cpufreq driver is in use.
 258
 259
 260 What:           /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
 261 Date:           August 2008
 262 KernelVersion:  2.6.27
 263 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 264 Description:    Disable L3 cache indices
 265
 266                 These files exist in every CPU's cache/index3 directory. Each
 267                 cache_disable_{0,1} file corresponds to one disable slot which
 268                 can be used to disable a cache index. Reading from these files
 269                 on a processor with this functionality will return the currently
 270                 disabled index for that node. There is one L3 structure per
 271                 node, or per internal node on MCM machines. Writing a valid
 272                 index to one of these files will cause the specificed cache
 273                 index to be disabled.
 274
 275                 All AMD processors with L3 caches provide this functionality.
 276                 For details, see BKDGs at
 277                 http://developer.amd.com/documentation/guides/Pages/default.aspx
 278
 279
 280 What:           /sys/devices/system/cpu/cpufreq/boost
 281 Date:           August 2012
 282 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 283 Description:    Processor frequency boosting control
 284
 285                 This switch controls the boost setting for the whole system.
 286                 Boosting allows the CPU and the firmware to run at a frequency
 287                 beyound it's nominal limit.
 288                 More details can be found in
 289                 Documentation/admin-guide/pm/cpufreq.rst
 290
 291
 292 What:           /sys/devices/system/cpu/cpu#/crash_notes
 293                 /sys/devices/system/cpu/cpu#/crash_notes_size
 294 Date:           April 2013
 295 Contact:        kexec@lists.infradead.org
 296 Description:    address and size of the percpu note.
 297
 298                 crash_notes: the physical address of the memory that holds the
 299                 note of cpu#.
 300
 301                 crash_notes_size: size of the note of cpu#.
 302
 303
 304 What:           /sys/devices/system/cpu/intel_pstate/max_perf_pct
 305                 /sys/devices/system/cpu/intel_pstate/min_perf_pct
 306                 /sys/devices/system/cpu/intel_pstate/no_turbo
 307 Date:           February 2013
 308 Contact:        linux-pm@vger.kernel.org
 309 Description:    Parameters for the Intel P-state driver
 310
 311                 Logic for selecting the current P-state in Intel
 312                 Sandybridge+ processors. The three knobs control
 313                 limits for the P-state that will be requested by the
 314                 driver.
 315
 316                 max_perf_pct: limits the maximum P state that will be requested by
 317                 the driver stated as a percentage of the available performance.
 318
 319                 min_perf_pct: limits the minimum P state that will be requested by
 320                 the driver stated as a percentage of the available performance.
 321
 322                 no_turbo: limits the driver to selecting P states below the turbo
 323                 frequency range.
 324
 325                 More details can be found in
 326                 Documentation/admin-guide/pm/intel_pstate.rst
 327
 328 What:           /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
 329 Date:           July 2014(documented, existed before August 2008)
 330 Contact:        Sudeep Holla <sudeep.holla@arm.com>
 331                 Linux kernel mailing list <linux-kernel@vger.kernel.org>
 332 Description:    Parameters for the CPU cache attributes
 333
 334                 allocation_policy:
 335                         - WriteAllocate: allocate a memory location to a cache line
 336                                          on a cache miss because of a write
 337                         - ReadAllocate: allocate a memory location to a cache line
 338                                         on a cache miss because of a read
 339                         - ReadWriteAllocate: both writeallocate and readallocate
 340
 341                 attributes: LEGACY used only on IA64 and is same as write_policy
 342
 343                 coherency_line_size: the minimum amount of data in bytes that gets
 344                                      transferred from memory to cache
 345
 346                 level: the cache hierarchy in the multi-level cache configuration
 347
 348                 number_of_sets: total number of sets in the cache, a set is a
 349                                 collection of cache lines with the same cache index
 350
 351                 physical_line_partition: number of physical cache line per cache tag
 352
 353                 shared_cpu_list: the list of logical cpus sharing the cache
 354
 355                 shared_cpu_map: logical cpu mask containing the list of cpus sharing
 356                                 the cache
 357
 358                 size: the total cache size in kB
 359
 360                 type:
 361                         - Instruction: cache that only holds instructions
 362                         - Data: cache that only caches data
 363                         - Unified: cache that holds both data and instructions
 364
 365                 ways_of_associativity: degree of freedom in placing a particular block
 366                                         of memory in the cache
 367
 368                 write_policy:
 369                         - WriteThrough: data is written to both the cache line
 370                                         and to the block in the lower-level memory
 371                         - WriteBack: data is written only to the cache line and
 372                                      the modified cache line is written to main
 373                                      memory only when it is replaced
 374
 375
 376 What:           /sys/devices/system/cpu/cpu*/cache/index*/id
 377 Date:           September 2016
 378 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 379 Description:    Cache id
 380
 381                 The id provides a unique number for a specific instance of
 382                 a cache of a particular type. E.g. there may be a level
 383                 3 unified cache on each socket in a server and we may
 384                 assign them ids 0, 1, 2, ...
 385
 386                 Note that id value can be non-contiguous. E.g. level 1
 387                 caches typically exist per core, but there may not be a
 388                 power of two cores on a socket, so these caches may be
 389                 numbered 0, 1, 2, 3, 4, 5, 8, 9, 10, ...
 390
 391 What:           /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats
 392                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
 393                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat
 394                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
 395                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
 396                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
 397                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
 398                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
 399                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset
 400 Date:           March 2016
 401 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 402                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 403 Description:    POWERNV CPUFreq driver's frequency throttle stats directory and
 404                 attributes
 405
 406                 'cpuX/cpufreq/throttle_stats' directory contains the CPU frequency
 407                 throttle stat attributes for the chip. The throttle stats of a cpu
 408                 is common across all the cpus belonging to a chip. Below are the
 409                 throttle attributes exported in the 'throttle_stats' directory:
 410
 411                 - turbo_stat : This file gives the total number of times the max
 412                 frequency is throttled to lower frequency in turbo (at and above
 413                 nominal frequency) range of frequencies.
 414
 415                 - sub_turbo_stat : This file gives the total number of times the
 416                 max frequency is throttled to lower frequency in sub-turbo(below
 417                 nominal frequency) range of frequencies.
 418
 419                 - unthrottle : This file gives the total number of times the max
 420                 frequency is unthrottled after being throttled.
 421
 422                 - powercap : This file gives the total number of times the max
 423                 frequency is throttled due to 'Power Capping'.
 424
 425                 - overtemp : This file gives the total number of times the max
 426                 frequency is throttled due to 'CPU Over Temperature'.
 427
 428                 - supply_fault : This file gives the total number of times the
 429                 max frequency is throttled due to 'Power Supply Failure'.
 430
 431                 - overcurrent : This file gives the total number of times the
 432                 max frequency is throttled due to 'Overcurrent'.
 433
 434                 - occ_reset : This file gives the total number of times the max
 435                 frequency is throttled due to 'OCC Reset'.
 436
 437                 The sysfs attributes representing different throttle reasons like
 438                 powercap, overtemp, supply_fault, overcurrent and occ_reset map to
 439                 the reasons provided by OCC firmware for throttling the frequency.
 440
 441 What:           /sys/devices/system/cpu/cpufreq/policyX/throttle_stats
 442                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat
 443                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat
 444                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle
 445                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap
 446                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp
 447                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault
 448                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent
 449                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset
 450 Date:           March 2016
 451 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 452                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 453 Description:    POWERNV CPUFreq driver's frequency throttle stats directory and
 454                 attributes
 455
 456                 'policyX/throttle_stats' directory and all the attributes are same as
 457                 the /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory and
 458                 attributes which give the frequency throttle information of the chip.
 459
 460 What:           /sys/devices/system/cpu/cpuX/regs/
 461                 /sys/devices/system/cpu/cpuX/regs/identification/
 462                 /sys/devices/system/cpu/cpuX/regs/identification/midr_el1
 463                 /sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
 464 Date:           June 2016
 465 Contact:        Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
 466 Description:    AArch64 CPU registers
 467                 'identification' directory exposes the CPU ID registers for
 468                  identifying model and revision of the CPU.
 469
 470 What:           /sys/devices/system/cpu/cpu#/cpu_capacity
 471 Date:           December 2016
 472 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 473 Description:    information about CPUs heterogeneity.
 474
 475                 cpu_capacity: capacity of cpu#.
 476
 477 What:           /sys/devices/system/cpu/vulnerabilities
 478                 /sys/devices/system/cpu/vulnerabilities/meltdown
 479                 /sys/devices/system/cpu/vulnerabilities/spectre_v1
 480                 /sys/devices/system/cpu/vulnerabilities/spectre_v2
 481                 /sys/devices/system/cpu/vulnerabilities/spec_store_bypass
 482 Date:           January 2018
 483 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 484 Description:    Information about CPU vulnerabilities
 485
 486                 The files are named after the code names of CPU
 487                 vulnerabilities. The output of those files reflects the
 488                 state of the CPUs in the system. Possible output values:
 489
 490                 "Not affected"    CPU is not affected by the vulnerability
 491                 "Vulnerable"      CPU is affected and no mitigation in effect
 492                 "Mitigation: $M"  CPU is affected and mitigation $M is in effect