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 Date:           September 2007
 112 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 113 Description:    Discover cpuidle policy and mechanism
 114
 115                 Various CPUs today support multiple idle levels that are
 116                 differentiated by varying exit latencies and power
 117                 consumption during idle.
 118
 119                 Idle policy (governor) is differentiated from idle mechanism
 120                 (driver)
 121
 122                 current_driver: displays current idle mechanism
 123
 124                 current_governor_ro: displays current idle policy
 125
 126                 See files in Documentation/cpuidle/ for more information.
 127
 128
 129 What:           /sys/devices/system/cpu/cpu#/cpufreq/*
 130 Date:           pre-git history
 131 Contact:        linux-pm@vger.kernel.org
 132 Description:    Discover and change clock speed of CPUs
 133
 134                 Clock scaling allows you to change the clock speed of the
 135                 CPUs on the fly. This is a nice method to save battery
 136                 power, because the lower the clock speed, the less power
 137                 the CPU consumes.
 138
 139                 There are many knobs to tweak in this directory.
 140
 141                 See files in Documentation/cpu-freq/ for more information.
 142
 143                 In particular, read Documentation/cpu-freq/user-guide.txt
 144                 to learn how to control the knobs.
 145
 146
 147 What:           /sys/devices/system/cpu/cpu#/cpufreq/freqdomain_cpus
 148 Date:           June 2013
 149 Contact:        linux-pm@vger.kernel.org
 150 Description:    Discover CPUs in the same CPU frequency coordination domain
 151
 152                 freqdomain_cpus is the list of CPUs (online+offline) that share
 153                 the same clock/freq domain (possibly at the hardware level).
 154                 That information may be hidden from the cpufreq core and the
 155                 value of related_cpus may be different from freqdomain_cpus. This
 156                 attribute is useful for user space DVFS controllers to get better
 157                 power/performance results for platforms using acpi-cpufreq.
 158
 159                 This file is only present if the acpi-cpufreq driver is in use.
 160
 161
 162 What:           /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
 163 Date:           August 2008
 164 KernelVersion:  2.6.27
 165 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 166 Description:    Disable L3 cache indices
 167
 168                 These files exist in every CPU's cache/index3 directory. Each
 169                 cache_disable_{0,1} file corresponds to one disable slot which
 170                 can be used to disable a cache index. Reading from these files
 171                 on a processor with this functionality will return the currently
 172                 disabled index for that node. There is one L3 structure per
 173                 node, or per internal node on MCM machines. Writing a valid
 174                 index to one of these files will cause the specificed cache
 175                 index to be disabled.
 176
 177                 All AMD processors with L3 caches provide this functionality.
 178                 For details, see BKDGs at
 179                 http://developer.amd.com/documentation/guides/Pages/default.aspx
 180
 181
 182 What:           /sys/devices/system/cpu/cpufreq/boost
 183 Date:           August 2012
 184 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 185 Description:    Processor frequency boosting control
 186
 187                 This switch controls the boost setting for the whole system.
 188                 Boosting allows the CPU and the firmware to run at a frequency
 189                 beyound it's nominal limit.
 190                 More details can be found in
 191                 Documentation/admin-guide/pm/cpufreq.rst
 192
 193
 194 What:           /sys/devices/system/cpu/cpu#/crash_notes
 195                 /sys/devices/system/cpu/cpu#/crash_notes_size
 196 Date:           April 2013
 197 Contact:        kexec@lists.infradead.org
 198 Description:    address and size of the percpu note.
 199
 200                 crash_notes: the physical address of the memory that holds the
 201                 note of cpu#.
 202
 203                 crash_notes_size: size of the note of cpu#.
 204
 205
 206 What:           /sys/devices/system/cpu/intel_pstate/max_perf_pct
 207                 /sys/devices/system/cpu/intel_pstate/min_perf_pct
 208                 /sys/devices/system/cpu/intel_pstate/no_turbo
 209 Date:           February 2013
 210 Contact:        linux-pm@vger.kernel.org
 211 Description:    Parameters for the Intel P-state driver
 212
 213                 Logic for selecting the current P-state in Intel
 214                 Sandybridge+ processors. The three knobs control
 215                 limits for the P-state that will be requested by the
 216                 driver.
 217
 218                 max_perf_pct: limits the maximum P state that will be requested by
 219                 the driver stated as a percentage of the available performance.
 220
 221                 min_perf_pct: limits the minimum P state that will be requested by
 222                 the driver stated as a percentage of the available performance.
 223
 224                 no_turbo: limits the driver to selecting P states below the turbo
 225                 frequency range.
 226
 227                 More details can be found in
 228                 Documentation/admin-guide/pm/intel_pstate.rst
 229
 230 What:           /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
 231 Date:           July 2014(documented, existed before August 2008)
 232 Contact:        Sudeep Holla <sudeep.holla@arm.com>
 233                 Linux kernel mailing list <linux-kernel@vger.kernel.org>
 234 Description:    Parameters for the CPU cache attributes
 235
 236                 allocation_policy:
 237                         - WriteAllocate: allocate a memory location to a cache line
 238                                          on a cache miss because of a write
 239                         - ReadAllocate: allocate a memory location to a cache line
 240                                         on a cache miss because of a read
 241                         - ReadWriteAllocate: both writeallocate and readallocate
 242
 243                 attributes: LEGACY used only on IA64 and is same as write_policy
 244
 245                 coherency_line_size: the minimum amount of data in bytes that gets
 246                                      transferred from memory to cache
 247
 248                 level: the cache hierarchy in the multi-level cache configuration
 249
 250                 number_of_sets: total number of sets in the cache, a set is a
 251                                 collection of cache lines with the same cache index
 252
 253                 physical_line_partition: number of physical cache line per cache tag
 254
 255                 shared_cpu_list: the list of logical cpus sharing the cache
 256
 257                 shared_cpu_map: logical cpu mask containing the list of cpus sharing
 258                                 the cache
 259
 260                 size: the total cache size in kB
 261
 262                 type:
 263                         - Instruction: cache that only holds instructions
 264                         - Data: cache that only caches data
 265                         - Unified: cache that holds both data and instructions
 266
 267                 ways_of_associativity: degree of freedom in placing a particular block
 268                                         of memory in the cache
 269
 270                 write_policy:
 271                         - WriteThrough: data is written to both the cache line
 272                                         and to the block in the lower-level memory
 273                         - WriteBack: data is written only to the cache line and
 274                                      the modified cache line is written to main
 275                                      memory only when it is replaced
 276
 277
 278 What:           /sys/devices/system/cpu/cpu*/cache/index*/id
 279 Date:           September 2016
 280 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 281 Description:    Cache id
 282
 283                 The id provides a unique number for a specific instance of
 284                 a cache of a particular type. E.g. there may be a level
 285                 3 unified cache on each socket in a server and we may
 286                 assign them ids 0, 1, 2, ...
 287
 288                 Note that id value can be non-contiguous. E.g. level 1
 289                 caches typically exist per core, but there may not be a
 290                 power of two cores on a socket, so these caches may be
 291                 numbered 0, 1, 2, 3, 4, 5, 8, 9, 10, ...
 292
 293 What:           /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats
 294                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
 295                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat
 296                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
 297                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
 298                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
 299                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
 300                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
 301                 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset
 302 Date:           March 2016
 303 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 304                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 305 Description:    POWERNV CPUFreq driver's frequency throttle stats directory and
 306                 attributes
 307
 308                 'cpuX/cpufreq/throttle_stats' directory contains the CPU frequency
 309                 throttle stat attributes for the chip. The throttle stats of a cpu
 310                 is common across all the cpus belonging to a chip. Below are the
 311                 throttle attributes exported in the 'throttle_stats' directory:
 312
 313                 - turbo_stat : This file gives the total number of times the max
 314                 frequency is throttled to lower frequency in turbo (at and above
 315                 nominal frequency) range of frequencies.
 316
 317                 - sub_turbo_stat : This file gives the total number of times the
 318                 max frequency is throttled to lower frequency in sub-turbo(below
 319                 nominal frequency) range of frequencies.
 320
 321                 - unthrottle : This file gives the total number of times the max
 322                 frequency is unthrottled after being throttled.
 323
 324                 - powercap : This file gives the total number of times the max
 325                 frequency is throttled due to 'Power Capping'.
 326
 327                 - overtemp : This file gives the total number of times the max
 328                 frequency is throttled due to 'CPU Over Temperature'.
 329
 330                 - supply_fault : This file gives the total number of times the
 331                 max frequency is throttled due to 'Power Supply Failure'.
 332
 333                 - overcurrent : This file gives the total number of times the
 334                 max frequency is throttled due to 'Overcurrent'.
 335
 336                 - occ_reset : This file gives the total number of times the max
 337                 frequency is throttled due to 'OCC Reset'.
 338
 339                 The sysfs attributes representing different throttle reasons like
 340                 powercap, overtemp, supply_fault, overcurrent and occ_reset map to
 341                 the reasons provided by OCC firmware for throttling the frequency.
 342
 343 What:           /sys/devices/system/cpu/cpufreq/policyX/throttle_stats
 344                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat
 345                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat
 346                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle
 347                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap
 348                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp
 349                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault
 350                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent
 351                 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset
 352 Date:           March 2016
 353 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 354                 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
 355 Description:    POWERNV CPUFreq driver's frequency throttle stats directory and
 356                 attributes
 357
 358                 'policyX/throttle_stats' directory and all the attributes are same as
 359                 the /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory and
 360                 attributes which give the frequency throttle information of the chip.
 361
 362 What:           /sys/devices/system/cpu/cpuX/regs/
 363                 /sys/devices/system/cpu/cpuX/regs/identification/
 364                 /sys/devices/system/cpu/cpuX/regs/identification/midr_el1
 365                 /sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
 366 Date:           June 2016
 367 Contact:        Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
 368 Description:    AArch64 CPU registers
 369                 'identification' directory exposes the CPU ID registers for
 370                  identifying model and revision of the CPU.
 371
 372 What:           /sys/devices/system/cpu/cpu#/cpu_capacity
 373 Date:           December 2016
 374 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 375 Description:    information about CPUs heterogeneity.
 376
 377                 cpu_capacity: capacity of cpu#.
 378
 379 What:           /sys/devices/system/cpu/vulnerabilities
 380                 /sys/devices/system/cpu/vulnerabilities/meltdown
 381                 /sys/devices/system/cpu/vulnerabilities/spectre_v1
 382                 /sys/devices/system/cpu/vulnerabilities/spectre_v2
 383 Date:           January 2018
 384 Contact:        Linux kernel mailing list <linux-kernel@vger.kernel.org>
 385 Description:    Information about CPU vulnerabilities
 386
 387                 The files are named after the code names of CPU
 388                 vulnerabilities. The output of those files reflects the
 389                 state of the CPUs in the system. Possible output values:
 390
 391                 "Not affected"    CPU is not affected by the vulnerability
 392                 "Vulnerable"      CPU is affected and no mitigation in effect
 393                 "Mitigation: $M"  CPU is affected and mitigation $M is in effect