Loading...
1What: /sys/devices/system/cpu/
2Date: pre-git history
3Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
4Description:
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/cpuX/
11
12What: /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
17Date: December 2008
18Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
19Description: 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
40What: /sys/devices/system/cpu/probe
41 /sys/devices/system/cpu/release
42Date: November 2009
43Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
44Description: 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 written to the file to remove CPU's
54 is architecture specific.
55
56What: /sys/devices/system/cpu/cpuX/node
57Date: October 2009
58Contact: Linux memory management mailing list <linux-mm@kvack.org>
59Description: 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
70What: /sys/devices/system/cpu/cpuX/topology/core_siblings
71 /sys/devices/system/cpu/cpuX/topology/core_siblings_list
72 /sys/devices/system/cpu/cpuX/topology/physical_package_id
73 /sys/devices/system/cpu/cpuX/topology/thread_siblings
74 /sys/devices/system/cpu/cpuX/topology/thread_siblings_list
75 /sys/devices/system/cpu/cpuX/topology/ppin
76Date: December 2008
77Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
78Description: CPU topology files that describe a logical CPU's relationship
79 to other cores and threads in the same physical package.
80
81 One cpuX 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_siblings: internal kernel map of cpuX's hardware threads
87 within the same physical_package_id.
88
89 core_siblings_list: human-readable list of the logical CPU
90 numbers within the same physical_package_id as cpuX.
91
92 physical_package_id: physical package id of cpuX. Typically
93 corresponds to a physical socket number, but the actual value
94 is architecture and platform dependent.
95
96 thread_siblings: internal kernel map of cpuX's hardware
97 threads within the same core as cpuX
98
99 thread_siblings_list: human-readable list of cpuX's hardware
100 threads within the same core as cpuX
101
102 ppin: human-readable Protected Processor Identification
103 Number of the socket the cpu# belongs to. There should be
104 one per physical_package_id. File is readable only to
105 admin.
106
107 See Documentation/admin-guide/cputopology.rst for more information.
108
109
110What: /sys/devices/system/cpu/cpuidle/available_governors
111 /sys/devices/system/cpu/cpuidle/current_driver
112 /sys/devices/system/cpu/cpuidle/current_governor
113 /sys/devices/system/cpu/cpuidle/current_governer_ro
114Date: September 2007
115Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
116Description: Discover cpuidle policy and mechanism
117
118 Various CPUs today support multiple idle levels that are
119 differentiated by varying exit latencies and power
120 consumption during idle.
121
122 Idle policy (governor) is differentiated from idle mechanism
123 (driver).
124
125 available_governors: (RO) displays a space separated list of
126 available governors.
127
128 current_driver: (RO) displays current idle mechanism.
129
130 current_governor: (RW) displays current idle policy. Users can
131 switch the governor at runtime by writing to this file.
132
133 current_governor_ro: (RO) displays current idle policy.
134
135 See Documentation/admin-guide/pm/cpuidle.rst and
136 Documentation/driver-api/pm/cpuidle.rst for more information.
137
138
139What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/name
140 /sys/devices/system/cpu/cpuX/cpuidle/stateN/latency
141 /sys/devices/system/cpu/cpuX/cpuidle/stateN/power
142 /sys/devices/system/cpu/cpuX/cpuidle/stateN/time
143 /sys/devices/system/cpu/cpuX/cpuidle/stateN/usage
144 /sys/devices/system/cpu/cpuX/cpuidle/stateN/above
145 /sys/devices/system/cpu/cpuX/cpuidle/stateN/below
146Date: September 2007
147KernelVersion: v2.6.24
148Contact: Linux power management list <linux-pm@vger.kernel.org>
149Description:
150 The directory /sys/devices/system/cpu/cpuX/cpuidle contains per
151 logical CPU specific cpuidle information for each online cpu X.
152 The processor idle states which are available for use have the
153 following attributes:
154
155 ======== ==== =================================================
156 name: (RO) Name of the idle state (string).
157
158 latency: (RO) The latency to exit out of this idle state (in
159 microseconds).
160
161 power: (RO) The power consumed while in this idle state (in
162 milliwatts).
163
164 time: (RO) The total time spent in this idle state
165 (in microseconds).
166
167 usage: (RO) Number of times this state was entered (a count).
168
169 above: (RO) Number of times this state was entered, but the
170 observed CPU idle duration was too short for it
171 (a count).
172
173 below: (RO) Number of times this state was entered, but the
174 observed CPU idle duration was too long for it
175 (a count).
176 ======== ==== =================================================
177
178What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/desc
179Date: February 2008
180KernelVersion: v2.6.25
181Contact: Linux power management list <linux-pm@vger.kernel.org>
182Description:
183 (RO) A small description about the idle state (string).
184
185
186What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/disable
187Date: March 2012
188KernelVersion: v3.10
189Contact: Linux power management list <linux-pm@vger.kernel.org>
190Description:
191 (RW) Option to disable this idle state (bool). The behavior and
192 the effect of the disable variable depends on the implementation
193 of a particular governor. In the ladder governor, for example,
194 it is not coherent, i.e. if one is disabling a light state, then
195 all deeper states are disabled as well, but the disable variable
196 does not reflect it. Likewise, if one enables a deep state but a
197 lighter state still is disabled, then this has no effect.
198
199What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/default_status
200Date: December 2019
201KernelVersion: v5.6
202Contact: Linux power management list <linux-pm@vger.kernel.org>
203Description:
204 (RO) The default status of this state, "enabled" or "disabled".
205
206What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/residency
207Date: March 2014
208KernelVersion: v3.15
209Contact: Linux power management list <linux-pm@vger.kernel.org>
210Description:
211 (RO) Display the target residency i.e. the minimum amount of
212 time (in microseconds) this cpu should spend in this idle state
213 to make the transition worth the effort.
214
215What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/
216Date: March 2018
217KernelVersion: v4.17
218Contact: Linux power management list <linux-pm@vger.kernel.org>
219Description:
220 Idle state usage statistics related to suspend-to-idle.
221
222 This attribute group is only present for states that can be
223 used in suspend-to-idle with suspended timekeeping.
224
225What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/time
226Date: March 2018
227KernelVersion: v4.17
228Contact: Linux power management list <linux-pm@vger.kernel.org>
229Description:
230 Total time spent by the CPU in suspend-to-idle (with scheduler
231 tick suspended) after requesting this state.
232
233What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/usage
234Date: March 2018
235KernelVersion: v4.17
236Contact: Linux power management list <linux-pm@vger.kernel.org>
237Description:
238 Total number of times this state has been requested by the CPU
239 while entering suspend-to-idle.
240
241What: /sys/devices/system/cpu/cpuX/cpufreq/*
242Date: pre-git history
243Contact: linux-pm@vger.kernel.org
244Description: Discover and change clock speed of CPUs
245
246 Clock scaling allows you to change the clock speed of the
247 CPUs on the fly. This is a nice method to save battery
248 power, because the lower the clock speed, the less power
249 the CPU consumes.
250
251 There are many knobs to tweak in this directory.
252
253 See files in Documentation/cpu-freq/ for more information.
254
255
256What: /sys/devices/system/cpu/cpuX/cpufreq/freqdomain_cpus
257Date: June 2013
258Contact: linux-pm@vger.kernel.org
259Description: Discover CPUs in the same CPU frequency coordination domain
260
261 freqdomain_cpus is the list of CPUs (online+offline) that share
262 the same clock/freq domain (possibly at the hardware level).
263 That information may be hidden from the cpufreq core and the
264 value of related_cpus may be different from freqdomain_cpus. This
265 attribute is useful for user space DVFS controllers to get better
266 power/performance results for platforms using acpi-cpufreq.
267
268 This file is only present if the acpi-cpufreq or the cppc-cpufreq
269 drivers are in use.
270
271
272What: /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
273Date: August 2008
274KernelVersion: 2.6.27
275Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
276Description: Disable L3 cache indices
277
278 These files exist in every CPU's cache/index3 directory. Each
279 cache_disable_{0,1} file corresponds to one disable slot which
280 can be used to disable a cache index. Reading from these files
281 on a processor with this functionality will return the currently
282 disabled index for that node. There is one L3 structure per
283 node, or per internal node on MCM machines. Writing a valid
284 index to one of these files will cause the specified cache
285 index to be disabled.
286
287 All AMD processors with L3 caches provide this functionality.
288 For details, see BKDGs at
289 https://www.amd.com/en/support/tech-docs?keyword=bios+kernel
290
291
292What: /sys/devices/system/cpu/cpufreq/boost
293Date: August 2012
294Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
295Description: Processor frequency boosting control
296
297 This switch controls the boost setting for the whole system.
298 Boosting allows the CPU and the firmware to run at a frequency
299 beyond its nominal limit.
300
301 More details can be found in
302 Documentation/admin-guide/pm/cpufreq.rst
303
304
305What: /sys/devices/system/cpu/cpuX/crash_notes
306 /sys/devices/system/cpu/cpuX/crash_notes_size
307Date: April 2013
308Contact: kexec@lists.infradead.org
309Description: address and size of the percpu note.
310
311 crash_notes: the physical address of the memory that holds the
312 note of cpuX.
313
314 crash_notes_size: size of the note of cpuX.
315
316
317What: /sys/devices/system/cpu/intel_pstate/max_perf_pct
318 /sys/devices/system/cpu/intel_pstate/min_perf_pct
319 /sys/devices/system/cpu/intel_pstate/no_turbo
320Date: February 2013
321Contact: linux-pm@vger.kernel.org
322Description: Parameters for the Intel P-state driver
323
324 Logic for selecting the current P-state in Intel
325 Sandybridge+ processors. The three knobs control
326 limits for the P-state that will be requested by the
327 driver.
328
329 max_perf_pct: limits the maximum P state that will be requested by
330 the driver stated as a percentage of the available performance.
331
332 min_perf_pct: limits the minimum P state that will be requested by
333 the driver stated as a percentage of the available performance.
334
335 no_turbo: limits the driver to selecting P states below the turbo
336 frequency range.
337
338 More details can be found in
339 Documentation/admin-guide/pm/intel_pstate.rst
340
341What: /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
342Date: July 2014(documented, existed before August 2008)
343Contact: Sudeep Holla <sudeep.holla@arm.com>
344 Linux kernel mailing list <linux-kernel@vger.kernel.org>
345Description: Parameters for the CPU cache attributes
346
347 allocation_policy:
348 - WriteAllocate:
349 allocate a memory location to a cache line
350 on a cache miss because of a write
351 - ReadAllocate:
352 allocate a memory location to a cache line
353 on a cache miss because of a read
354 - ReadWriteAllocate:
355 both writeallocate and readallocate
356
357 coherency_line_size:
358 the minimum amount of data in bytes that gets
359 transferred from memory to cache
360
361 level:
362 the cache hierarchy in the multi-level cache configuration
363
364 number_of_sets:
365 total number of sets in the cache, a set is a
366 collection of cache lines with the same cache index
367
368 physical_line_partition:
369 number of physical cache line per cache tag
370
371 shared_cpu_list:
372 the list of logical cpus sharing the cache
373
374 shared_cpu_map:
375 logical cpu mask containing the list of cpus sharing
376 the cache
377
378 size:
379 the total cache size in kB
380
381 type:
382 - Instruction: cache that only holds instructions
383 - Data: cache that only caches data
384 - Unified: cache that holds both data and instructions
385
386 ways_of_associativity:
387 degree of freedom in placing a particular block
388 of memory in the cache
389
390 write_policy:
391 - WriteThrough:
392 data is written to both the cache line
393 and to the block in the lower-level memory
394 - WriteBack:
395 data is written only to the cache line and
396 the modified cache line is written to main
397 memory only when it is replaced
398
399
400What: /sys/devices/system/cpu/cpu*/cache/index*/id
401Date: September 2016
402Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
403Description: Cache id
404
405 The id provides a unique number for a specific instance of
406 a cache of a particular type. E.g. there may be a level
407 3 unified cache on each socket in a server and we may
408 assign them ids 0, 1, 2, ...
409
410 Note that id value can be non-contiguous. E.g. level 1
411 caches typically exist per core, but there may not be a
412 power of two cores on a socket, so these caches may be
413 numbered 0, 1, 2, 3, 4, 5, 8, 9, 10, ...
414
415What: /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats
416 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
417 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat
418 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
419 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
420 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
421 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
422 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
423 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset
424Date: March 2016
425Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
426 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
427Description: POWERNV CPUFreq driver's frequency throttle stats directory and
428 attributes
429
430 'cpuX/cpufreq/throttle_stats' directory contains the CPU frequency
431 throttle stat attributes for the chip. The throttle stats of a cpu
432 is common across all the cpus belonging to a chip. Below are the
433 throttle attributes exported in the 'throttle_stats' directory:
434
435 - turbo_stat : This file gives the total number of times the max
436 frequency is throttled to lower frequency in turbo (at and above
437 nominal frequency) range of frequencies.
438
439 - sub_turbo_stat : This file gives the total number of times the
440 max frequency is throttled to lower frequency in sub-turbo(below
441 nominal frequency) range of frequencies.
442
443 - unthrottle : This file gives the total number of times the max
444 frequency is unthrottled after being throttled.
445
446 - powercap : This file gives the total number of times the max
447 frequency is throttled due to 'Power Capping'.
448
449 - overtemp : This file gives the total number of times the max
450 frequency is throttled due to 'CPU Over Temperature'.
451
452 - supply_fault : This file gives the total number of times the
453 max frequency is throttled due to 'Power Supply Failure'.
454
455 - overcurrent : This file gives the total number of times the
456 max frequency is throttled due to 'Overcurrent'.
457
458 - occ_reset : This file gives the total number of times the max
459 frequency is throttled due to 'OCC Reset'.
460
461 The sysfs attributes representing different throttle reasons like
462 powercap, overtemp, supply_fault, overcurrent and occ_reset map to
463 the reasons provided by OCC firmware for throttling the frequency.
464
465What: /sys/devices/system/cpu/cpufreq/policyX/throttle_stats
466 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat
467 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat
468 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle
469 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap
470 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp
471 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault
472 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent
473 /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset
474Date: March 2016
475Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
476 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
477Description: POWERNV CPUFreq driver's frequency throttle stats directory and
478 attributes
479
480 'policyX/throttle_stats' directory and all the attributes are same as
481 the /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory and
482 attributes which give the frequency throttle information of the chip.
483
484What: /sys/devices/system/cpu/cpuX/regs/
485 /sys/devices/system/cpu/cpuX/regs/identification/
486 /sys/devices/system/cpu/cpuX/regs/identification/midr_el1
487 /sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
488 /sys/devices/system/cpu/cpuX/regs/identification/smidr_el1
489Date: June 2016
490Contact: Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
491Description: AArch64 CPU registers
492
493 'identification' directory exposes the CPU ID registers for
494 identifying model and revision of the CPU and SMCU.
495
496What: /sys/devices/system/cpu/aarch32_el0
497Date: May 2021
498Contact: Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
499Description: Identifies the subset of CPUs in the system that can execute
500 AArch32 (32-bit ARM) applications. If present, the same format as
501 /sys/devices/system/cpu/{offline,online,possible,present} is used.
502 If absent, then all or none of the CPUs can execute AArch32
503 applications and execve() will behave accordingly.
504
505What: /sys/devices/system/cpu/cpuX/cpu_capacity
506Date: December 2016
507Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
508Description: information about CPUs heterogeneity.
509
510 cpu_capacity: capacity of cpuX.
511
512What: /sys/devices/system/cpu/vulnerabilities
513 /sys/devices/system/cpu/vulnerabilities/gather_data_sampling
514 /sys/devices/system/cpu/vulnerabilities/itlb_multihit
515 /sys/devices/system/cpu/vulnerabilities/l1tf
516 /sys/devices/system/cpu/vulnerabilities/mds
517 /sys/devices/system/cpu/vulnerabilities/meltdown
518 /sys/devices/system/cpu/vulnerabilities/mmio_stale_data
519 /sys/devices/system/cpu/vulnerabilities/reg_file_data_sampling
520 /sys/devices/system/cpu/vulnerabilities/retbleed
521 /sys/devices/system/cpu/vulnerabilities/spec_store_bypass
522 /sys/devices/system/cpu/vulnerabilities/spectre_v1
523 /sys/devices/system/cpu/vulnerabilities/spectre_v2
524 /sys/devices/system/cpu/vulnerabilities/srbds
525 /sys/devices/system/cpu/vulnerabilities/tsx_async_abort
526Date: January 2018
527Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
528Description: Information about CPU vulnerabilities
529
530 The files are named after the code names of CPU
531 vulnerabilities. The output of those files reflects the
532 state of the CPUs in the system. Possible output values:
533
534 ================ ==============================================
535 "Not affected" CPU is not affected by the vulnerability
536 "Vulnerable" CPU is affected and no mitigation in effect
537 "Mitigation: $M" CPU is affected and mitigation $M is in effect
538 ================ ==============================================
539
540 See also: Documentation/admin-guide/hw-vuln/index.rst
541
542What: /sys/devices/system/cpu/smt
543 /sys/devices/system/cpu/smt/active
544 /sys/devices/system/cpu/smt/control
545Date: June 2018
546Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
547Description: Control Symmetric Multi Threading (SMT)
548
549 active: Tells whether SMT is active (enabled and siblings online)
550
551 control: Read/write interface to control SMT. Possible
552 values:
553
554 ================ =========================================
555 "on" SMT is enabled
556 "off" SMT is disabled
557 "<N>" SMT is enabled with N threads per core.
558 "forceoff" SMT is force disabled. Cannot be changed.
559 "notsupported" SMT is not supported by the CPU
560 "notimplemented" SMT runtime toggling is not
561 implemented for the architecture
562 ================ =========================================
563
564 If control status is "forceoff" or "notsupported" writes
565 are rejected.
566
567What: /sys/devices/system/cpu/cpuX/power/energy_perf_bias
568Date: March 2019
569Contact: linux-pm@vger.kernel.org
570Description: Intel Energy and Performance Bias Hint (EPB)
571
572 EPB for the given CPU in a sliding scale 0 - 15, where a value
573 of 0 corresponds to a hint preference for highest performance
574 and a value of 15 corresponds to the maximum energy savings.
575
576 In order to change the EPB value for the CPU, write either
577 a number in the 0 - 15 sliding scale above, or one of the
578 strings: "performance", "balance-performance", "normal",
579 "balance-power", "power" (that represent values reflected by
580 their meaning), to this attribute.
581
582 This attribute is present for all online CPUs supporting the
583 Intel EPB feature.
584
585What: /sys/devices/system/cpu/umwait_control
586 /sys/devices/system/cpu/umwait_control/enable_c02
587 /sys/devices/system/cpu/umwait_control/max_time
588Date: May 2019
589Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
590Description: Umwait control
591
592 enable_c02: Read/write interface to control umwait C0.2 state
593 Read returns C0.2 state status:
594 0: C0.2 is disabled
595 1: C0.2 is enabled
596
597 Write 'y' or '1' or 'on' to enable C0.2 state.
598 Write 'n' or '0' or 'off' to disable C0.2 state.
599
600 The interface is case insensitive.
601
602 max_time: Read/write interface to control umwait maximum time
603 in TSC-quanta that the CPU can reside in either C0.1
604 or C0.2 state. The time is an unsigned 32-bit number.
605 Note that a value of zero means there is no limit.
606 Low order two bits must be zero.
607
608What: /sys/devices/system/cpu/svm
609Date: August 2019
610Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
611 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
612Description: Secure Virtual Machine
613
614 If 1, it means the system is using the Protected Execution
615 Facility in POWER9 and newer processors. i.e., it is a Secure
616 Virtual Machine.
617
618What: /sys/devices/system/cpu/cpuX/purr
619Date: Apr 2005
620Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
621Description: PURR ticks for this CPU since the system boot.
622
623 The Processor Utilization Resources Register (PURR) is
624 a 64-bit counter which provides an estimate of the
625 resources used by the CPU thread. The contents of this
626 register increases monotonically. This sysfs interface
627 exposes the number of PURR ticks for cpuX.
628
629What: /sys/devices/system/cpu/cpuX/spurr
630Date: Dec 2006
631Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
632Description: SPURR ticks for this CPU since the system boot.
633
634 The Scaled Processor Utilization Resources Register
635 (SPURR) is a 64-bit counter that provides a frequency
636 invariant estimate of the resources used by the CPU
637 thread. The contents of this register increases
638 monotonically. This sysfs interface exposes the number
639 of SPURR ticks for cpuX.
640
641What: /sys/devices/system/cpu/cpuX/idle_purr
642Date: Apr 2020
643Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
644Description: PURR ticks for cpuX when it was idle.
645
646 This sysfs interface exposes the number of PURR ticks
647 for cpuX when it was idle.
648
649What: /sys/devices/system/cpu/cpuX/idle_spurr
650Date: Apr 2020
651Contact: Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
652Description: SPURR ticks for cpuX when it was idle.
653
654 This sysfs interface exposes the number of SPURR ticks
655 for cpuX when it was idle.
656
657What: /sys/devices/system/cpu/cpuX/mte_tcf_preferred
658Date: July 2021
659Contact: Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
660Description: Preferred MTE tag checking mode
661
662 When a user program specifies more than one MTE tag checking
663 mode, this sysfs node is used to specify which mode should
664 be preferred when scheduling a task on that CPU. Possible
665 values:
666
667 ================ ==============================================
668 "sync" Prefer synchronous mode
669 "asymm" Prefer asymmetric mode
670 "async" Prefer asynchronous mode
671 ================ ==============================================
672
673 See also: Documentation/arch/arm64/memory-tagging-extension.rst
674
675What: /sys/devices/system/cpu/nohz_full
676Date: Apr 2015
677Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
678Description:
679 (RO) the list of CPUs that are in nohz_full mode.
680 These CPUs are set by boot parameter "nohz_full=".
681
682What: /sys/devices/system/cpu/isolated
683Date: Apr 2015
684Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
685Description:
686 (RO) the list of CPUs that are isolated and don't
687 participate in load balancing. These CPUs are set by
688 boot parameter "isolcpus=".
689
690What: /sys/devices/system/cpu/crash_hotplug
691Date: Aug 2023
692Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
693Description:
694 (RO) indicates whether or not the kernel directly supports
695 modifying the crash elfcorehdr for CPU hot un/plug and/or
696 on/offline changes.
1What: /sys/devices/system/cpu/
2Date: pre-git history
3Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
4Description:
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
12What: /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
17Date: December 2008
18Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
19Description: 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
40What: /sys/devices/system/cpu/probe
41 /sys/devices/system/cpu/release
42Date: November 2009
43Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
44Description: 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
56What: /sys/devices/system/cpu/cpu#/node
57Date: October 2009
58Contact: Linux memory management mailing list <linux-mm@kvack.org>
59Description: 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
70What: /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
76Date: December 2008
77Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
78Description: 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
109What: /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
113Date: September 2007
114Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
115Description: 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
143What: /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
148Date: September 2007
149KernelVersion: v2.6.24
150Contact: Linux power management list <linux-pm@vger.kernel.org>
151Description:
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
170What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/desc
171Date: February 2008
172KernelVersion: v2.6.25
173Contact: Linux power management list <linux-pm@vger.kernel.org>
174Description:
175 (RO) A small description about the idle state (string).
176
177
178What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/disable
179Date: March 2012
180KernelVersion: v3.10
181Contact: Linux power management list <linux-pm@vger.kernel.org>
182Description:
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
192What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/residency
193Date: March 2014
194KernelVersion: v3.15
195Contact: Linux power management list <linux-pm@vger.kernel.org>
196Description:
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
201What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/
202Date: March 2018
203KernelVersion: v4.17
204Contact: Linux power management list <linux-pm@vger.kernel.org>
205Description:
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
211What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/time
212Date: March 2018
213KernelVersion: v4.17
214Contact: Linux power management list <linux-pm@vger.kernel.org>
215Description:
216 Total time spent by the CPU in suspend-to-idle (with scheduler
217 tick suspended) after requesting this state.
218
219What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/s2idle/usage
220Date: March 2018
221KernelVersion: v4.17
222Contact: Linux power management list <linux-pm@vger.kernel.org>
223Description:
224 Total number of times this state has been requested by the CPU
225 while entering suspend-to-idle.
226
227What: /sys/devices/system/cpu/cpu#/cpufreq/*
228Date: pre-git history
229Contact: linux-pm@vger.kernel.org
230Description: 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
245What: /sys/devices/system/cpu/cpu#/cpufreq/freqdomain_cpus
246Date: June 2013
247Contact: linux-pm@vger.kernel.org
248Description: 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
260What: /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
261Date: August 2008
262KernelVersion: 2.6.27
263Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
264Description: 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
280What: /sys/devices/system/cpu/cpufreq/boost
281Date: August 2012
282Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
283Description: 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
292What: /sys/devices/system/cpu/cpu#/crash_notes
293 /sys/devices/system/cpu/cpu#/crash_notes_size
294Date: April 2013
295Contact: kexec@lists.infradead.org
296Description: 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
304What: /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
307Date: February 2013
308Contact: linux-pm@vger.kernel.org
309Description: 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
328What: /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
329Date: July 2014(documented, existed before August 2008)
330Contact: Sudeep Holla <sudeep.holla@arm.com>
331 Linux kernel mailing list <linux-kernel@vger.kernel.org>
332Description: 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
376What: /sys/devices/system/cpu/cpu*/cache/index*/id
377Date: September 2016
378Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
379Description: 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
391What: /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
400Date: March 2016
401Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
402 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
403Description: 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
441What: /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
450Date: March 2016
451Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
452 Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
453Description: 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
460What: /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
464Date: June 2016
465Contact: Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
466Description: AArch64 CPU registers
467 'identification' directory exposes the CPU ID registers for
468 identifying model and revision of the CPU.
469
470What: /sys/devices/system/cpu/cpu#/cpu_capacity
471Date: December 2016
472Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
473Description: information about CPUs heterogeneity.
474
475 cpu_capacity: capacity of cpu#.
476
477What: /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
482Date: January 2018
483Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
484Description: 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