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
242What:		/sys/devices/system/cpu/cpu#/cpufreq/freqdomain_cpus
243Date:		June 2013
244Contact:	linux-pm@vger.kernel.org
245Description:	Discover CPUs in the same CPU frequency coordination domain
246
247		freqdomain_cpus is the list of CPUs (online+offline) that share
248		the same clock/freq domain (possibly at the hardware level).
249		That information may be hidden from the cpufreq core and the
250		value of related_cpus may be different from freqdomain_cpus. This
251		attribute is useful for user space DVFS controllers to get better
252		power/performance results for platforms using acpi-cpufreq.
253
254		This file is only present if the acpi-cpufreq driver is in use.
255
256
257What:		/sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
258Date:		August 2008
259KernelVersion:	2.6.27
260Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
261Description:	Disable L3 cache indices
262
263		These files exist in every CPU's cache/index3 directory. Each
264		cache_disable_{0,1} file corresponds to one disable slot which
265		can be used to disable a cache index. Reading from these files
266		on a processor with this functionality will return the currently
267		disabled index for that node. There is one L3 structure per
268		node, or per internal node on MCM machines. Writing a valid
269		index to one of these files will cause the specificed cache
270		index to be disabled.
271
272		All AMD processors with L3 caches provide this functionality.
273		For details, see BKDGs at
274		http://developer.amd.com/documentation/guides/Pages/default.aspx
275
276
277What:		/sys/devices/system/cpu/cpufreq/boost
278Date:		August 2012
279Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
280Description:	Processor frequency boosting control
281
282		This switch controls the boost setting for the whole system.
283		Boosting allows the CPU and the firmware to run at a frequency
284		beyound it's nominal limit.
285		More details can be found in
286		Documentation/admin-guide/pm/cpufreq.rst
287
288
289What:		/sys/devices/system/cpu/cpu#/crash_notes
290		/sys/devices/system/cpu/cpu#/crash_notes_size
291Date:		April 2013
292Contact:	kexec@lists.infradead.org
293Description:	address and size of the percpu note.
294
295		crash_notes: the physical address of the memory that holds the
296		note of cpu#.
297
298		crash_notes_size: size of the note of cpu#.
299
300
301What:		/sys/devices/system/cpu/intel_pstate/max_perf_pct
302		/sys/devices/system/cpu/intel_pstate/min_perf_pct
303		/sys/devices/system/cpu/intel_pstate/no_turbo
304Date:		February 2013
305Contact:	linux-pm@vger.kernel.org
306Description:	Parameters for the Intel P-state driver
307
308		Logic for selecting the current P-state in Intel
309		Sandybridge+ processors. The three knobs control
310		limits for the P-state that will be requested by the
311		driver.
312
313		max_perf_pct: limits the maximum P state that will be requested by
314		the driver stated as a percentage of the available performance.
315
316		min_perf_pct: limits the minimum P state that will be requested by
317		the driver stated as a percentage of the available performance.
318
319		no_turbo: limits the driver to selecting P states below the turbo
320		frequency range.
321
322		More details can be found in
323		Documentation/admin-guide/pm/intel_pstate.rst
324
325What:		/sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
326Date:		July 2014(documented, existed before August 2008)
327Contact:	Sudeep Holla <sudeep.holla@arm.com>
328		Linux kernel mailing list <linux-kernel@vger.kernel.org>
329Description:	Parameters for the CPU cache attributes
330
331		allocation_policy:
332			- WriteAllocate: allocate a memory location to a cache line
333					 on a cache miss because of a write
334			- ReadAllocate: allocate a memory location to a cache line
335					on a cache miss because of a read
336			- ReadWriteAllocate: both writeallocate and readallocate
337
338		attributes: LEGACY used only on IA64 and is same as write_policy
339
340		coherency_line_size: the minimum amount of data in bytes that gets
341				     transferred from memory to cache
342
343		level: the cache hierarchy in the multi-level cache configuration
344
345		number_of_sets: total number of sets in the cache, a set is a
346				collection of cache lines with the same cache index
347
348		physical_line_partition: number of physical cache line per cache tag
349
350		shared_cpu_list: the list of logical cpus sharing the cache
351
352		shared_cpu_map: logical cpu mask containing the list of cpus sharing
353				the cache
354
355		size: the total cache size in kB
356
357		type:
358			- Instruction: cache that only holds instructions
359			- Data: cache that only caches data
360			- Unified: cache that holds both data and instructions
361
362		ways_of_associativity: degree of freedom in placing a particular block
363					of memory in the cache
364
365		write_policy:
366			- WriteThrough: data is written to both the cache line
367					and to the block in the lower-level memory
368			- WriteBack: data is written only to the cache line and
369				     the modified cache line is written to main
370				     memory only when it is replaced
371
372
373What:		/sys/devices/system/cpu/cpu*/cache/index*/id
374Date:		September 2016
375Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
376Description:	Cache id
377
378		The id provides a unique number for a specific instance of
379		a cache of a particular type. E.g. there may be a level
380		3 unified cache on each socket in a server and we may
381		assign them ids 0, 1, 2, ...
382
383		Note that id value can be non-contiguous. E.g. level 1
384		caches typically exist per core, but there may not be a
385		power of two cores on a socket, so these caches may be
386		numbered 0, 1, 2, 3, 4, 5, 8, 9, 10, ...
387
388What:		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats
389		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
390		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat
391		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
392		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
393		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
394		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
395		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
396		/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset
397Date:		March 2016
398Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
399		Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
400Description:	POWERNV CPUFreq driver's frequency throttle stats directory and
401		attributes
402
403		'cpuX/cpufreq/throttle_stats' directory contains the CPU frequency
404		throttle stat attributes for the chip. The throttle stats of a cpu
405		is common across all the cpus belonging to a chip. Below are the
406		throttle attributes exported in the 'throttle_stats' directory:
407
408		- turbo_stat : This file gives the total number of times the max
409		frequency is throttled to lower frequency in turbo (at and above
410		nominal frequency) range of frequencies.
411
412		- sub_turbo_stat : This file gives the total number of times the
413		max frequency is throttled to lower frequency in sub-turbo(below
414		nominal frequency) range of frequencies.
415
416		- unthrottle : This file gives the total number of times the max
417		frequency is unthrottled after being throttled.
418
419		- powercap : This file gives the total number of times the max
420		frequency is throttled due to 'Power Capping'.
421
422		- overtemp : This file gives the total number of times the max
423		frequency is throttled due to 'CPU Over Temperature'.
424
425		- supply_fault : This file gives the total number of times the
426		max frequency is throttled due to 'Power Supply Failure'.
427
428		- overcurrent : This file gives the total number of times the
429		max frequency is throttled due to 'Overcurrent'.
430
431		- occ_reset : This file gives the total number of times the max
432		frequency is throttled due to 'OCC Reset'.
433
434		The sysfs attributes representing different throttle reasons like
435		powercap, overtemp, supply_fault, overcurrent and occ_reset map to
436		the reasons provided by OCC firmware for throttling the frequency.
437
438What:		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats
439		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat
440		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat
441		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle
442		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap
443		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp
444		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault
445		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent
446		/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset
447Date:		March 2016
448Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
449		Linux for PowerPC mailing list <linuxppc-dev@ozlabs.org>
450Description:	POWERNV CPUFreq driver's frequency throttle stats directory and
451		attributes
452
453		'policyX/throttle_stats' directory and all the attributes are same as
454		the /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory and
455		attributes which give the frequency throttle information of the chip.
456
457What:		/sys/devices/system/cpu/cpuX/regs/
458		/sys/devices/system/cpu/cpuX/regs/identification/
459		/sys/devices/system/cpu/cpuX/regs/identification/midr_el1
460		/sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
461Date:		June 2016
462Contact:	Linux ARM Kernel Mailing list <linux-arm-kernel@lists.infradead.org>
463Description:	AArch64 CPU registers
464		'identification' directory exposes the CPU ID registers for
465		 identifying model and revision of the CPU.
466
467What:		/sys/devices/system/cpu/cpu#/cpu_capacity
468Date:		December 2016
469Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
470Description:	information about CPUs heterogeneity.
471
472		cpu_capacity: capacity of cpu#.
473
474What:		/sys/devices/system/cpu/vulnerabilities
475		/sys/devices/system/cpu/vulnerabilities/meltdown
476		/sys/devices/system/cpu/vulnerabilities/spectre_v1
477		/sys/devices/system/cpu/vulnerabilities/spectre_v2
478		/sys/devices/system/cpu/vulnerabilities/spec_store_bypass
479		/sys/devices/system/cpu/vulnerabilities/l1tf
480Date:		January 2018
481Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
482Description:	Information about CPU vulnerabilities
483
484		The files are named after the code names of CPU
485		vulnerabilities. The output of those files reflects the
486		state of the CPUs in the system. Possible output values:
487
488		"Not affected"	  CPU is not affected by the vulnerability
489		"Vulnerable"	  CPU is affected and no mitigation in effect
490		"Mitigation: $M"  CPU is affected and mitigation $M is in effect
491
492		Details about the l1tf file can be found in
493		Documentation/admin-guide/l1tf.rst
494
495What:		/sys/devices/system/cpu/smt
496		/sys/devices/system/cpu/smt/active
497		/sys/devices/system/cpu/smt/control
498Date:		June 2018
499Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>
500Description:	Control Symetric Multi Threading (SMT)
501
502		active:  Tells whether SMT is active (enabled and siblings online)
503
504		control: Read/write interface to control SMT. Possible
505			 values:
506
507			 "on"		SMT is enabled
508			 "off"		SMT is disabled
509			 "forceoff"	SMT is force disabled. Cannot be changed.
510			 "notsupported" SMT is not supported by the CPU
511
512			 If control status is "forceoff" or "notsupported" writes
513			 are rejected.
514