요약·해설과 원문, 전문 번역을 서로 분리했습니다. API 이름, symbol, source path는 원문 표기를 사용합니다.
1. 요약·해설
원문의 핵심 논리와 kernel programming 관점의 보충 설명입니다. 아래의 전문 번역과는 별도로 작성했습니다.
CPU idle policy and state statistics
sysfs-devices-system-cpu:110-245cpuidle driver·governor를 선택하고 state별 latency, power, time, usage, target residency, disable/default status와 suspend-to-idle 통계를 제공합니다.
CPU frequency and performance policy
sysfs-devices-system-cpu:246-399Clock scaling, shared frequency domain, CPPC autonomous selection/activity window/EPP, system boost와 Intel P-state percentage·turbo limits를 제어합니다.
CPU cache controls and identity
sysfs-devices-system-cpu:331-350, 400-473AMD L3 index disable slots와 cache allocation/write policy, geometry, sharing, type, size 및 non-contiguous instance ID를 설명합니다.
Throttle statistics, registers and capacity
sysfs-devices-system-cpu:364-375, 474-571Per-CPU crash notes, POWERNV throttle reasons, AArch64 identification registers, AArch32-capable CPU mask와 heterogeneous CPU capacity를 expose합니다.
Vulnerability status and SMT control
sysfs-devices-system-cpu:572-631Named CPU vulnerability files는 영향·mitigation state를 보고하고 SMT active/control은 enable, thread count, forced disable와 architecture support 상태를 제어합니다.
Energy hints, UMWAIT and secure virtualization
sysfs-devices-system-cpu:632-694Intel EPB와 UMWAIT C0.2/max-time policy를 제어하고 AMD SEV-SNP VMPL 및 POWER9 Protected Execution Facility 사용 여부를 보고합니다.
PowerPC utilization counters and ARM MTE
sysfs-devices-system-cpu:695-751PowerPC PURR/SPURR와 idle variants는 thread resource ticks를 보고하고 ARM64 MTE는 CPU별 preferred tag-checking mode를 선택합니다.
CPU isolation, crash hotplug and online eligibility
sysfs-devices-system-cpu:752-779Boot parameter로 정한 nohz_full·isolated CPU 목록, kexec segment hotplug update 지원 여부와 online 전환 가능한 CPU 목록을 제공합니다.
2. 영어 원문 전체
번역 기준이 된 Linux v6.18.37 원문입니다. 줄 번호는 이 버전의 파일 좌표입니다.
원문 전체 펼치기
1
What: /sys/devices/system/cpu/
2
Date: pre-git history
3
Contact: Linux kernel mailing list <[email protected]>
4
Description:
5
A collection of both global and individual CPU attributes
7
Individual CPU attributes are contained in subdirectories
8
named by the kernel's logical CPU number, e.g.:
10
/sys/devices/system/cpu/cpuX/
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 <[email protected]>
19
Description: CPU topology files that describe kernel limits related to
20
hotplug. Briefly:
22
kernel_max: the maximum cpu index allowed by the kernel
23
configuration.
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).
29
online: cpus that are online and being scheduled.
31
possible: cpus that have been allocated resources and can be
32
brought online if they are present.
34
present: cpus that have been identified as being present in
35
the system.
37
See Documentation/admin-guide/cputopology.rst for more information.
40
What: /sys/devices/system/cpu/probe
41
/sys/devices/system/cpu/release
42
Date: November 2009
43
Contact: Linux kernel mailing list <[email protected]>
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.
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.
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.
56
What: /sys/devices/system/cpu/cpuX/node
57
Date: October 2009
58
Contact: Linux memory management mailing list <[email protected]>
59
Description: Discover NUMA node a CPU belongs to
61
When CONFIG_NUMA is enabled, a symbolic link that points
62
to the corresponding NUMA node directory.
64
For example, the following symlink is created for cpu42
65
in NUMA node 2:
67
/sys/devices/system/cpu/cpu42/node2 -> ../../node/node2
70
What: /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
76
Date: December 2008
77
Contact: Linux kernel mailing list <[email protected]>
78
Description: CPU topology files that describe a logical CPU's relationship
79
to other cores and threads in the same physical package.
81
One cpuX directory is created per logical CPU in the system,
82
e.g. /sys/devices/system/cpu/cpu42/.
84
Briefly, the files above are:
86
core_siblings: internal kernel map of cpuX's hardware threads
87
within the same physical_package_id.
89
core_siblings_list: human-readable list of the logical CPU
90
numbers within the same physical_package_id as cpuX.
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.
96
thread_siblings: internal kernel map of cpuX's hardware
97
threads within the same core as cpuX
99
thread_siblings_list: human-readable list of cpuX's hardware
100
threads within the same core as cpuX
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.
107
See Documentation/admin-guide/cputopology.rst for more information.
110
What: /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
114
/sys/devices/system/cpu/cpuidle/intel_c1_demotion
115
Date: September 2007
116
Contact: Linux kernel mailing list <[email protected]>
117
Description: Discover cpuidle policy and mechanism
119
Various CPUs today support multiple idle levels that are
120
differentiated by varying exit latencies and power
121
consumption during idle.
123
Idle policy (governor) is differentiated from idle mechanism
124
(driver).
126
available_governors: (RO) displays a space separated list of
127
available governors.
129
current_driver: (RO) displays current idle mechanism.
131
current_governor: (RW) displays current idle policy. Users can
132
switch the governor at runtime by writing to this file.
134
current_governor_ro: (RO) displays current idle policy.
136
intel_c1_demotion: (RW) enables/disables the C1 demotion
137
feature on Intel CPUs.
139
See Documentation/admin-guide/pm/cpuidle.rst,
140
Documentation/admin-guide/pm/intel_idle.rst, and
141
Documentation/driver-api/pm/cpuidle.rst for more information.
144
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/name
145
/sys/devices/system/cpu/cpuX/cpuidle/stateN/latency
146
/sys/devices/system/cpu/cpuX/cpuidle/stateN/power
147
/sys/devices/system/cpu/cpuX/cpuidle/stateN/time
148
/sys/devices/system/cpu/cpuX/cpuidle/stateN/usage
149
/sys/devices/system/cpu/cpuX/cpuidle/stateN/above
150
/sys/devices/system/cpu/cpuX/cpuidle/stateN/below
151
Date: September 2007
152
KernelVersion: v2.6.24
153
Contact: Linux power management list <[email protected]>
154
Description:
155
The directory /sys/devices/system/cpu/cpuX/cpuidle contains per
156
logical CPU specific cpuidle information for each online cpu X.
157
The processor idle states which are available for use have the
158
following attributes:
160
======== ==== =================================================
161
name: (RO) Name of the idle state (string).
163
latency: (RO) The latency to exit out of this idle state (in
164
microseconds).
166
power: (RO) The power consumed while in this idle state (in
167
milliwatts).
169
time: (RO) The total time spent in this idle state
170
(in microseconds).
172
usage: (RO) Number of times this state was entered (a count).
174
above: (RO) Number of times this state was entered, but the
175
observed CPU idle duration was too short for it
176
(a count).
178
below: (RO) Number of times this state was entered, but the
179
observed CPU idle duration was too long for it
180
(a count).
181
======== ==== =================================================
183
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/desc
184
Date: February 2008
185
KernelVersion: v2.6.25
186
Contact: Linux power management list <[email protected]>
187
Description:
188
(RO) A small description about the idle state (string).
191
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/disable
192
Date: March 2012
193
KernelVersion: v3.10
194
Contact: Linux power management list <[email protected]>
195
Description:
196
(RW) Option to disable this idle state (bool). The behavior and
197
the effect of the disable variable depends on the implementation
198
of a particular governor. In the ladder governor, for example,
199
it is not coherent, i.e. if one is disabling a light state, then
200
all deeper states are disabled as well, but the disable variable
201
does not reflect it. Likewise, if one enables a deep state but a
202
lighter state still is disabled, then this has no effect.
204
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/default_status
205
Date: December 2019
206
KernelVersion: v5.6
207
Contact: Linux power management list <[email protected]>
208
Description:
209
(RO) The default status of this state, "enabled" or "disabled".
211
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/residency
212
Date: March 2014
213
KernelVersion: v3.15
214
Contact: Linux power management list <[email protected]>
215
Description:
216
(RO) Display the target residency i.e. the minimum amount of
217
time (in microseconds) this cpu should spend in this idle state
218
to make the transition worth the effort.
220
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/
221
Date: March 2018
222
KernelVersion: v4.17
223
Contact: Linux power management list <[email protected]>
224
Description:
225
Idle state usage statistics related to suspend-to-idle.
227
This attribute group is only present for states that can be
228
used in suspend-to-idle with suspended timekeeping.
230
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/time
231
Date: March 2018
232
KernelVersion: v4.17
233
Contact: Linux power management list <[email protected]>
234
Description:
235
Total time spent by the CPU in suspend-to-idle (with scheduler
236
tick suspended) after requesting this state.
238
What: /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/usage
239
Date: March 2018
240
KernelVersion: v4.17
241
Contact: Linux power management list <[email protected]>
242
Description:
243
Total number of times this state has been requested by the CPU
244
while entering suspend-to-idle.
246
What: /sys/devices/system/cpu/cpuX/cpufreq/*
247
Date: pre-git history
248
Contact: [email protected]
249
Description: Discover and change clock speed of CPUs
251
Clock scaling allows you to change the clock speed of the
252
CPUs on the fly. This is a nice method to save battery
253
power, because the lower the clock speed, the less power
254
the CPU consumes.
256
There are many knobs to tweak in this directory.
258
See files in Documentation/cpu-freq/ for more information.
261
What: /sys/devices/system/cpu/cpuX/cpufreq/freqdomain_cpus
262
Date: June 2013
263
Contact: [email protected]
264
Description: Discover CPUs in the same CPU frequency coordination domain
266
freqdomain_cpus is the list of CPUs (online+offline) that share
267
the same clock/freq domain (possibly at the hardware level).
268
That information may be hidden from the cpufreq core and the
269
value of related_cpus may be different from freqdomain_cpus. This
270
attribute is useful for user space DVFS controllers to get better
271
power/performance results for platforms using acpi-cpufreq.
273
This file is only present if the acpi-cpufreq or the cppc-cpufreq
274
drivers are in use.
276
What: /sys/devices/system/cpu/cpuX/cpufreq/auto_select
277
Date: May 2025
278
Contact: [email protected]
279
Description: Autonomous selection enable
281
Read/write interface to control autonomous selection enable
282
Read returns autonomous selection status:
283
0: autonomous selection is disabled
284
1: autonomous selection is enabled
286
Write 'y' or '1' or 'on' to enable autonomous selection.
287
Write 'n' or '0' or 'off' to disable autonomous selection.
289
This file is only present if the cppc-cpufreq driver is in use.
291
What: /sys/devices/system/cpu/cpuX/cpufreq/auto_act_window
292
Date: May 2025
293
Contact: [email protected]
294
Description: Autonomous activity window
296
This file indicates a moving utilization sensitivity window to
297
the platform's autonomous selection policy.
299
Read/write an integer represents autonomous activity window (in
300
microseconds) from/to this file. The max value to write is
301
1270000000 but the max significand is 127. This means that if 128
302
is written to this file, 127 will be stored. If the value is
303
greater than 130, only the first two digits will be saved as
304
significand.
306
Writing a zero value to this file enable the platform to
307
determine an appropriate Activity Window depending on the workload.
309
Writing to this file only has meaning when Autonomous Selection is
310
enabled.
312
This file is only present if the cppc-cpufreq driver is in use.
314
What: /sys/devices/system/cpu/cpuX/cpufreq/energy_performance_preference_val
315
Date: May 2025
316
Contact: [email protected]
317
Description: Energy performance preference
319
Read/write an 8-bit integer from/to this file. This file
320
represents a range of values from 0 (performance preference) to
321
0xFF (energy efficiency preference) that influences the rate of
322
performance increase/decrease and the result of the hardware's
323
energy efficiency and performance optimization policies.
325
Writing to this file only has meaning when Autonomous Selection is
326
enabled.
328
This file is only present if the cppc-cpufreq driver is in use.
331
What: /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1}
332
Date: August 2008
333
KernelVersion: 2.6.27
334
Contact: Linux kernel mailing list <[email protected]>
335
Description: Disable L3 cache indices
337
These files exist in every CPU's cache/index3 directory. Each
338
cache_disable_{0,1} file corresponds to one disable slot which
339
can be used to disable a cache index. Reading from these files
340
on a processor with this functionality will return the currently
341
disabled index for that node. There is one L3 structure per
342
node, or per internal node on MCM machines. Writing a valid
343
index to one of these files will cause the specified cache
344
index to be disabled.
346
All AMD processors with L3 caches provide this functionality.
347
For details, see BKDGs at
348
https://www.amd.com/en/support/tech-docs?keyword=bios+kernel
351
What: /sys/devices/system/cpu/cpufreq/boost
352
Date: August 2012
353
Contact: Linux kernel mailing list <[email protected]>
354
Description: Processor frequency boosting control
356
This switch controls the boost setting for the whole system.
357
Boosting allows the CPU and the firmware to run at a frequency
358
beyond its nominal limit.
360
More details can be found in
361
Documentation/admin-guide/pm/cpufreq.rst
364
What: /sys/devices/system/cpu/cpuX/crash_notes
365
/sys/devices/system/cpu/cpuX/crash_notes_size
366
Date: April 2013
367
Contact: [email protected]
368
Description: address and size of the percpu note.
370
crash_notes: the physical address of the memory that holds the
371
note of cpuX.
373
crash_notes_size: size of the note of cpuX.
376
What: /sys/devices/system/cpu/intel_pstate/max_perf_pct
377
/sys/devices/system/cpu/intel_pstate/min_perf_pct
378
/sys/devices/system/cpu/intel_pstate/no_turbo
379
Date: February 2013
380
Contact: [email protected]
381
Description: Parameters for the Intel P-state driver
383
Logic for selecting the current P-state in Intel
384
Sandybridge+ processors. The three knobs control
385
limits for the P-state that will be requested by the
386
driver.
388
max_perf_pct: limits the maximum P state that will be requested by
389
the driver stated as a percentage of the available performance.
391
min_perf_pct: limits the minimum P state that will be requested by
392
the driver stated as a percentage of the available performance.
394
no_turbo: limits the driver to selecting P states below the turbo
395
frequency range.
397
More details can be found in
398
Documentation/admin-guide/pm/intel_pstate.rst
400
What: /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below>
401
Date: July 2014(documented, existed before August 2008)
402
Contact: Sudeep Holla <[email protected]>
403
Linux kernel mailing list <[email protected]>
404
Description: Parameters for the CPU cache attributes
406
allocation_policy:
407
- WriteAllocate:
408
allocate a memory location to a cache line
409
on a cache miss because of a write
410
- ReadAllocate:
411
allocate a memory location to a cache line
412
on a cache miss because of a read
413
- ReadWriteAllocate:
414
both writeallocate and readallocate
416
coherency_line_size:
417
the minimum amount of data in bytes that gets
418
transferred from memory to cache
420
level:
421
the cache hierarchy in the multi-level cache configuration
423
number_of_sets:
424
total number of sets in the cache, a set is a
425
collection of cache lines with the same cache index
427
physical_line_partition:
428
number of physical cache line per cache tag
430
shared_cpu_list:
431
the list of logical cpus sharing the cache
433
shared_cpu_map:
434
logical cpu mask containing the list of cpus sharing
435
the cache
437
size:
438
the total cache size in kB
440
type:
441
- Instruction: cache that only holds instructions
442
- Data: cache that only caches data
443
- Unified: cache that holds both data and instructions
445
ways_of_associativity:
446
degree of freedom in placing a particular block
447
of memory in the cache
449
write_policy:
450
- WriteThrough:
451
data is written to both the cache line
452
and to the block in the lower-level memory
453
- WriteBack:
454
data is written only to the cache line and
455
the modified cache line is written to main
456
memory only when it is replaced
459
What: /sys/devices/system/cpu/cpu*/cache/index*/id
460
Date: September 2016
461
Contact: Linux kernel mailing list <[email protected]>
462
Description: Cache id
464
The id provides a unique number for a specific instance of
465
a cache of a particular type. E.g. there may be a level
466
3 unified cache on each socket in a server and we may
467
assign them ids 0, 1, 2, ...
469
Note that id value can be non-contiguous. E.g. level 1
470
caches typically exist per core, but there may not be a
471
power of two cores on a socket, so these caches may be
472
numbered 0, 1, 2, 3, 4, 5, 8, 9, 10, ...
474
What: /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats
475
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
476
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat
477
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
478
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
479
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
480
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
481
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
482
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset
483
Date: March 2016
484
Contact: Linux kernel mailing list <[email protected]>
485
Linux for PowerPC mailing list <[email protected]>
486
Description: POWERNV CPUFreq driver's frequency throttle stats directory and
487
attributes
489
'cpuX/cpufreq/throttle_stats' directory contains the CPU frequency
490
throttle stat attributes for the chip. The throttle stats of a cpu
491
is common across all the cpus belonging to a chip. Below are the
492
throttle attributes exported in the 'throttle_stats' directory:
494
- turbo_stat : This file gives the total number of times the max
495
frequency is throttled to lower frequency in turbo (at and above
496
nominal frequency) range of frequencies.
498
- sub_turbo_stat : This file gives the total number of times the
499
max frequency is throttled to lower frequency in sub-turbo(below
500
nominal frequency) range of frequencies.
502
- unthrottle : This file gives the total number of times the max
503
frequency is unthrottled after being throttled.
505
- powercap : This file gives the total number of times the max
506
frequency is throttled due to 'Power Capping'.
508
- overtemp : This file gives the total number of times the max
509
frequency is throttled due to 'CPU Over Temperature'.
511
- supply_fault : This file gives the total number of times the
512
max frequency is throttled due to 'Power Supply Failure'.
514
- overcurrent : This file gives the total number of times the
515
max frequency is throttled due to 'Overcurrent'.
517
- occ_reset : This file gives the total number of times the max
518
frequency is throttled due to 'OCC Reset'.
520
The sysfs attributes representing different throttle reasons like
521
powercap, overtemp, supply_fault, overcurrent and occ_reset map to
522
the reasons provided by OCC firmware for throttling the frequency.
524
What: /sys/devices/system/cpu/cpufreq/policyX/throttle_stats
525
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat
526
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat
527
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle
528
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap
529
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp
530
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault
531
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent
532
/sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset
533
Date: March 2016
534
Contact: Linux kernel mailing list <[email protected]>
535
Linux for PowerPC mailing list <[email protected]>
536
Description: POWERNV CPUFreq driver's frequency throttle stats directory and
537
attributes
539
'policyX/throttle_stats' directory and all the attributes are same as
540
the /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory and
541
attributes which give the frequency throttle information of the chip.
543
What: /sys/devices/system/cpu/cpuX/regs/
544
/sys/devices/system/cpu/cpuX/regs/identification/
545
/sys/devices/system/cpu/cpuX/regs/identification/midr_el1
546
/sys/devices/system/cpu/cpuX/regs/identification/revidr_el1
547
/sys/devices/system/cpu/cpuX/regs/identification/aidr_el1
548
/sys/devices/system/cpu/cpuX/regs/identification/smidr_el1
549
Date: June 2016
550
Contact: Linux ARM Kernel Mailing list <[email protected]>
551
Description: AArch64 CPU registers
553
'identification' directory exposes the CPU ID registers for
554
identifying model and revision of the CPU and SMCU.
556
What: /sys/devices/system/cpu/aarch32_el0
557
Date: May 2021
558
Contact: Linux ARM Kernel Mailing list <[email protected]>
559
Description: Identifies the subset of CPUs in the system that can execute
560
AArch32 (32-bit ARM) applications. If present, the same format as
561
/sys/devices/system/cpu/{offline,online,possible,present} is used.
562
If absent, then all or none of the CPUs can execute AArch32
563
applications and execve() will behave accordingly.
565
What: /sys/devices/system/cpu/cpuX/cpu_capacity
566
Date: December 2016
567
Contact: Linux kernel mailing list <[email protected]>
568
Description: information about CPUs heterogeneity.
570
cpu_capacity: capacity of cpuX.
572
What: /sys/devices/system/cpu/vulnerabilities
573
/sys/devices/system/cpu/vulnerabilities/gather_data_sampling
574
/sys/devices/system/cpu/vulnerabilities/indirect_target_selection
575
/sys/devices/system/cpu/vulnerabilities/itlb_multihit
576
/sys/devices/system/cpu/vulnerabilities/l1tf
577
/sys/devices/system/cpu/vulnerabilities/mds
578
/sys/devices/system/cpu/vulnerabilities/meltdown
579
/sys/devices/system/cpu/vulnerabilities/mmio_stale_data
580
/sys/devices/system/cpu/vulnerabilities/old_microcode
581
/sys/devices/system/cpu/vulnerabilities/reg_file_data_sampling
582
/sys/devices/system/cpu/vulnerabilities/retbleed
583
/sys/devices/system/cpu/vulnerabilities/spec_store_bypass
584
/sys/devices/system/cpu/vulnerabilities/spectre_v1
585
/sys/devices/system/cpu/vulnerabilities/spectre_v2
586
/sys/devices/system/cpu/vulnerabilities/srbds
587
/sys/devices/system/cpu/vulnerabilities/tsa
588
/sys/devices/system/cpu/vulnerabilities/tsx_async_abort
589
/sys/devices/system/cpu/vulnerabilities/vmscape
590
Date: January 2018
591
Contact: Linux kernel mailing list <[email protected]>
592
Description: Information about CPU vulnerabilities
594
The files are named after the code names of CPU
595
vulnerabilities. The output of those files reflects the
596
state of the CPUs in the system. Possible output values:
598
================ ==============================================
599
"Not affected" CPU is not affected by the vulnerability
600
"Vulnerable" CPU is affected and no mitigation in effect
601
"Mitigation: $M" CPU is affected and mitigation $M is in effect
602
================ ==============================================
604
See also: Documentation/admin-guide/hw-vuln/index.rst
606
What: /sys/devices/system/cpu/smt
607
/sys/devices/system/cpu/smt/active
608
/sys/devices/system/cpu/smt/control
609
Date: June 2018
610
Contact: Linux kernel mailing list <[email protected]>
611
Description: Control Symmetric Multi Threading (SMT)
613
active: Tells whether SMT is active (enabled and siblings online)
615
control: Read/write interface to control SMT. Possible
616
values:
618
================ =========================================
619
"on" SMT is enabled
620
"off" SMT is disabled
621
"<N>" SMT is enabled with N threads per core.
622
"forceoff" SMT is force disabled. Cannot be changed.
623
"notsupported" SMT is not supported by the CPU
624
"notimplemented" SMT runtime toggling is not
625
implemented for the architecture
626
================ =========================================
628
If control status is "forceoff" or "notsupported" writes
629
are rejected. Note that enabling SMT on PowerPC skips
630
offline cores.
632
What: /sys/devices/system/cpu/cpuX/power/energy_perf_bias
633
Date: March 2019
634
Contact: [email protected]
635
Description: Intel Energy and Performance Bias Hint (EPB)
637
EPB for the given CPU in a sliding scale 0 - 15, where a value
638
of 0 corresponds to a hint preference for highest performance
639
and a value of 15 corresponds to the maximum energy savings.
641
In order to change the EPB value for the CPU, write either
642
a number in the 0 - 15 sliding scale above, or one of the
643
strings: "performance", "balance-performance", "normal",
644
"balance-power", "power" (that represent values reflected by
645
their meaning), to this attribute.
647
This attribute is present for all online CPUs supporting the
648
Intel EPB feature.
650
What: /sys/devices/system/cpu/umwait_control
651
/sys/devices/system/cpu/umwait_control/enable_c02
652
/sys/devices/system/cpu/umwait_control/max_time
653
Date: May 2019
654
Contact: Linux kernel mailing list <[email protected]>
655
Description: Umwait control
657
enable_c02: Read/write interface to control umwait C0.2 state
658
Read returns C0.2 state status:
659
0: C0.2 is disabled
660
1: C0.2 is enabled
662
Write 'y' or '1' or 'on' to enable C0.2 state.
663
Write 'n' or '0' or 'off' to disable C0.2 state.
665
The interface is case insensitive.
667
max_time: Read/write interface to control umwait maximum time
668
in TSC-quanta that the CPU can reside in either C0.1
669
or C0.2 state. The time is an unsigned 32-bit number.
670
Note that a value of zero means there is no limit.
671
Low order two bits must be zero.
673
What: /sys/devices/system/cpu/sev
674
/sys/devices/system/cpu/sev/vmpl
675
Date: May 2024
676
Contact: Linux kernel mailing list <[email protected]>
677
Description: Secure Encrypted Virtualization (SEV) information
679
This directory is only present when running as an SEV-SNP guest.
681
vmpl: Reports the Virtual Machine Privilege Level (VMPL) at which
682
the SEV-SNP guest is running.
685
What: /sys/devices/system/cpu/svm
686
Date: August 2019
687
Contact: Linux kernel mailing list <[email protected]>
688
Linux for PowerPC mailing list <[email protected]>
689
Description: Secure Virtual Machine
691
If 1, it means the system is using the Protected Execution
692
Facility in POWER9 and newer processors. i.e., it is a Secure
693
Virtual Machine.
695
What: /sys/devices/system/cpu/cpuX/purr
696
Date: Apr 2005
697
Contact: Linux for PowerPC mailing list <[email protected]>
698
Description: PURR ticks for this CPU since the system boot.
700
The Processor Utilization Resources Register (PURR) is
701
a 64-bit counter which provides an estimate of the
702
resources used by the CPU thread. The contents of this
703
register increases monotonically. This sysfs interface
704
exposes the number of PURR ticks for cpuX.
706
What: /sys/devices/system/cpu/cpuX/spurr
707
Date: Dec 2006
708
Contact: Linux for PowerPC mailing list <[email protected]>
709
Description: SPURR ticks for this CPU since the system boot.
711
The Scaled Processor Utilization Resources Register
712
(SPURR) is a 64-bit counter that provides a frequency
713
invariant estimate of the resources used by the CPU
714
thread. The contents of this register increases
715
monotonically. This sysfs interface exposes the number
716
of SPURR ticks for cpuX.
718
What: /sys/devices/system/cpu/cpuX/idle_purr
719
Date: Apr 2020
720
Contact: Linux for PowerPC mailing list <[email protected]>
721
Description: PURR ticks for cpuX when it was idle.
723
This sysfs interface exposes the number of PURR ticks
724
for cpuX when it was idle.
726
What: /sys/devices/system/cpu/cpuX/idle_spurr
727
Date: Apr 2020
728
Contact: Linux for PowerPC mailing list <[email protected]>
729
Description: SPURR ticks for cpuX when it was idle.
731
This sysfs interface exposes the number of SPURR ticks
732
for cpuX when it was idle.
734
What: /sys/devices/system/cpu/cpuX/mte_tcf_preferred
735
Date: July 2021
736
Contact: Linux ARM Kernel Mailing list <[email protected]>
737
Description: Preferred MTE tag checking mode
739
When a user program specifies more than one MTE tag checking
740
mode, this sysfs node is used to specify which mode should
741
be preferred when scheduling a task on that CPU. Possible
742
values:
744
================ ==============================================
745
"sync" Prefer synchronous mode
746
"asymm" Prefer asymmetric mode
747
"async" Prefer asynchronous mode
748
================ ==============================================
750
See also: Documentation/arch/arm64/memory-tagging-extension.rst
752
What: /sys/devices/system/cpu/nohz_full
753
Date: Apr 2015
754
Contact: Linux kernel mailing list <[email protected]>
755
Description:
756
(RO) the list of CPUs that are in nohz_full mode.
757
These CPUs are set by boot parameter "nohz_full=".
759
What: /sys/devices/system/cpu/isolated
760
Date: Apr 2015
761
Contact: Linux kernel mailing list <[email protected]>
762
Description:
763
(RO) the list of CPUs that are isolated and don't
764
participate in load balancing. These CPUs are set by
765
boot parameter "isolcpus=".
767
What: /sys/devices/system/cpu/crash_hotplug
768
Date: Aug 2023
769
Contact: Linux kernel mailing list <[email protected]>
770
Description:
771
(RO) indicates whether or not the kernel updates relevant kexec
772
segments on memory hot un/plug and/or on/offline events, avoiding the
773
need to reload kdump kernel.
775
What: /sys/devices/system/cpu/enabled
776
Date: Nov 2022
777
Contact: Linux kernel mailing list <[email protected]>
778
Description:
779
(RO) the list of CPUs that can be brought online.
3. 한국어 전문 번역
영어 원문의 문단 순서와 의미를 유지한 전체 번역입니다. 코드, 함수명, symbol과 URL은 원문 표기를 유지합니다.
Global and per-CPU attribute hierarchy
1-11| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/ |
| Date | pre-git history |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Global CPU attribute와 개별 CPU attribute를 함께 모아 둔 directory입니다. 개별 CPU attribute는 kernel logical CPU number로 이름 붙인 subdirectory에 들어 있습니다. 예를 들면 /sys/devices/system/cpu/cpuX/입니다. |
/sys/devices/system/cpu/→Global CPU attributes→cpuX logical CPU directory→Per-CPU topology, idle, frequency and architecture data
Global controls와 logical-CPU-specific attributes가 갈라지는 구조입니다.
CPU hotplug topology masks
12-39| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/kernel_max; /sys/devices/system/cpu/offline; /sys/devices/system/cpu/online; /sys/devices/system/cpu/possible; /sys/devices/system/cpu/present |
| Date | 2008년 12월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Hotplug와 관련된 kernel limit를 설명하는 CPU topology files입니다. kernel_max는 kernel configuration이 허용하는 maximum CPU index입니다. offline은 hotplug로 off되었거나 kernel_max limit를 초과해 online이 아닌 CPU들입니다. online은 online 상태이며 scheduling 중인 CPU들입니다. possible은 resource가 할당되어 있고 present하면 online으로 전환할 수 있는 CPU들입니다. present는 system에 존재한다고 식별된 CPU들입니다. 자세한 내용은 Documentation/admin-guide/cputopology.rst를 참조하십시오. |
FileMeaning
kernel_maxMaximum CPU index allowed by configuration
offlinePresent or configured CPUs not currently online
onlineOnline CPUs participating in scheduling
possibleCPUs with resources that may be brought online
presentCPUs identified as physically/logically present
Hotplug lifecycle과 configuration limit 관점의 CPU sets입니다.
Dynamic CPU addition and complete removal
40-55| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/probe; /sys/devices/system/cpu/release |
| Date | 2009년 11월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | CPU를 동적으로 추가하거나 제거합니다. 이는 hotplug removal이 아니라 system에서 CPU를 완전히 제거하거나 추가하기 위한 interface입니다. probe에 architecture-specific 정보를 write하면 CPU가 system에 동적으로 추가됩니다. release에 architecture-specific 정보를 write하면 CPU가 system에서 동적으로 제거됩니다. |
CPU NUMA-node link
56-69| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/node |
| Date | 2009년 10월 |
| Contact | Linux memory management mailing list <[email protected]> |
| Description | CPU가 속한 NUMA node를 찾는 interface입니다. CONFIG_NUMA가 enable되어 있으면 해당 NUMA node directory를 가리키는 symbolic link가 생성됩니다. 예를 들어 NUMA node 2에 속한 cpu42에는 /sys/devices/system/cpu/cpu42/node2 -> ../../node/node2 link가 만들어집니다. |
/sys/devices/system/cpu/cpu42/node2→Symbolic link→../../node/node2→NUMA node 2
Per-CPU directory의 symlink가 canonical node directory를 가리킵니다.
Package, core and thread topology
70-109| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/topology/core_siblings; /sys/devices/system/cpu/cpuX/topology/core_siblings_list; /sys/devices/system/cpu/cpuX/topology/physical_package_id; /sys/devices/system/cpu/cpuX/topology/thread_siblings; /sys/devices/system/cpu/cpuX/topology/thread_siblings_list; /sys/devices/system/cpu/cpuX/topology/ppin |
| Date | 2008년 12월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Logical CPU와 같은 physical package 안의 다른 core 및 thread 사이 관계를 설명하는 topology files입니다. System의 logical CPU마다 cpuX directory가 하나씩 생성됩니다. core_siblings는 cpuX와 physical_package_id가 같은 hardware thread의 internal kernel map이고, core_siblings_list는 같은 package의 logical CPU number를 사람이 읽을 수 있게 나열합니다. physical_package_id는 cpuX의 physical package ID이며 보통 physical socket number에 대응하지만 실제 값은 architecture와 platform에 따라 달라집니다. thread_siblings는 cpuX와 같은 core의 hardware thread를 나타내는 internal kernel map이고, thread_siblings_list는 이를 사람이 읽을 수 있게 나열합니다. ppin은 cpuX가 속한 socket의 Protected Processor Identification Number를 사람이 읽을 수 있게 표시합니다. physical_package_id마다 하나가 있어야 하며 admin만 읽을 수 있습니다. 자세한 내용은 Documentation/admin-guide/cputopology.rst를 참조하십시오. |
AttributeScope / representation
core_siblingsKernel map of threads in same package
core_siblings_listReadable logical CPUs in same package
physical_package_idArchitecture/platform-specific package ID
thread_siblingsKernel map of threads in same core
thread_siblings_listReadable threads in same core
ppinAdmin-only socket Protected Processor ID
Package와 core sharing을 bitmap, list 및 identifier로 표현합니다.
CPU idle driver and governor policy
110-143| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuidle/available_governors; /sys/devices/system/cpu/cpuidle/current_driver; /sys/devices/system/cpu/cpuidle/current_governor; /sys/devices/system/cpu/cpuidle/current_governer_ro; /sys/devices/system/cpu/cpuidle/intel_c1_demotion |
| Date | 2007년 9월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Cpuidle policy와 mechanism을 확인합니다. 오늘날 여러 CPU는 exit latency와 idle power consumption이 다른 복수의 idle level을 지원합니다. Idle policy(governor)는 idle mechanism(driver)과 구분됩니다. available_governors는 사용 가능한 governor를 space-separated list로 표시하는 RO file입니다. current_driver는 현재 idle mechanism을 표시하는 RO file입니다. current_governor는 현재 idle policy를 표시하고 write로 runtime governor를 바꿀 수 있는 RW file입니다. 원문의 What path는 current_governer_ro로 표기되어 있으나 설명에서는 current_governor_ro라 하며, 현재 idle policy를 표시하는 RO file입니다. intel_c1_demotion은 Intel CPU의 C1 demotion 기능을 enable/disable하는 RW file입니다. 자세한 내용은 Documentation/admin-guide/pm/cpuidle.rst, Documentation/admin-guide/pm/intel_idle.rst, Documentation/driver-api/pm/cpuidle.rst를 참조하십시오. |
AttributeModePurpose
available_governorsROAvailable idle policies
current_driverROCurrent idle mechanism
current_governorRWCurrent/selectable idle policy
current_governer_roROCurrent idle policy
intel_c1_demotionRWIntel C1 demotion control
Idle mechanism, policy와 Intel-specific control의 access mode입니다.
Per-state CPU idle statistics
144-182| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/name; /sys/devices/system/cpu/cpuX/cpuidle/stateN/latency; /sys/devices/system/cpu/cpuX/cpuidle/stateN/power; /sys/devices/system/cpu/cpuX/cpuidle/stateN/time; /sys/devices/system/cpu/cpuX/cpuidle/stateN/usage; /sys/devices/system/cpu/cpuX/cpuidle/stateN/above; /sys/devices/system/cpu/cpuX/cpuidle/stateN/below |
| Date | 2007년 9월 |
| KernelVersion | v2.6.24 |
| Contact | Linux power management list <[email protected]> |
| Description | /sys/devices/system/cpu/cpuX/cpuidle directory에는 online logical CPU X마다 CPU별 cpuidle 정보가 들어 있습니다. 사용 가능한 processor idle state는 다음 attribute를 가집니다. name은 idle-state name string, latency는 해당 state에서 빠져나오는 latency(microseconds), power는 해당 state의 power consumption(milliwatts), time은 해당 state에서 보낸 total time(microseconds), usage는 state 진입 횟수입니다. above는 이 state에 진입했지만 관측된 CPU idle duration이 너무 짧았던 횟수이고, below는 진입했지만 관측된 idle duration이 너무 길었던 횟수입니다. 모두 read-only입니다. |
AttributeModeMeaning / unit
nameROIdle-state name string
latencyROExit latency, microseconds
powerROPower consumption, milliwatts
timeROTotal residency time, microseconds
usageROEntry count
aboveROEntries where observed idle was too short
belowROEntries where observed idle was too long
원문의 ASCII 표를 access mode, value와 unit이 드러나는 구조화 표로 옮겼습니다.
Idle-state description
183-190| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/desc |
| Date | 2008년 2월 |
| KernelVersion | v2.6.25 |
| Contact | Linux power management list <[email protected]> |
| Description | (RO) Idle state에 관한 짧은 description string입니다. |
Idle-state disable control
191-203| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/disable |
| Date | 2012년 3월 |
| KernelVersion | v3.10 |
| Contact | Linux power management list <[email protected]> |
| Description | (RW) 해당 idle state를 disable하는 boolean option입니다. disable variable의 behavior와 effect는 governor 구현에 따라 달라집니다. 예를 들어 ladder governor에서는 coherent하지 않습니다. Light state 하나를 disable하면 더 deep한 state도 모두 disable되지만 각 disable variable에는 이 사실이 반영되지 않습니다. 마찬가지로 lighter state가 여전히 disabled인 상태에서 deep state를 enable해도 효과가 없습니다. |
Disable a light idle state→All deeper states become unavailable→Their disable files may not reflect that→Re-enabling only a deeper state has no effect while lighter state stays disabled
Per-state variable과 실제 reachable idle-state chain이 다를 수 있습니다.
Default idle-state status
204-210| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/default_status |
| Date | 2019년 12월 |
| KernelVersion | v5.6 |
| Contact | Linux power management list <[email protected]> |
| Description | (RO) 해당 state의 default status인 "enabled" 또는 "disabled"를 표시합니다. |
Idle-state target residency
211-219| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/residency |
| Date | 2014년 3월 |
| KernelVersion | v3.15 |
| Contact | Linux power management list <[email protected]> |
| Description | (RO) Target residency를 표시합니다. 즉 이 idle state로 transition할 가치가 있으려면 CPU가 해당 state에서 보내야 하는 minimum time을 microseconds 단위로 나타냅니다. |
Suspend-to-idle state statistics group
220-229| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/ |
| Date | 2018년 3월 |
| KernelVersion | v4.17 |
| Contact | Linux power management list <[email protected]> |
| Description | Suspend-to-idle과 관련된 idle-state usage statistics입니다. Timekeeping이 suspended된 suspend-to-idle에서 사용할 수 있는 state에만 이 attribute group이 존재합니다. |
Idle state supports suspend-to-idle→Timekeeping is suspended→CPU requests state during s2idle→s2idle/time and s2idle/usage are updated
지원 state에서 scheduler tick이 멈춘 s2idle request만 별도로 집계합니다.
Suspend-to-idle total time
230-237| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/time |
| Date | 2018년 3월 |
| KernelVersion | v4.17 |
| Contact | Linux power management list <[email protected]> |
| Description | 이 state를 request한 뒤 CPU가 scheduler tick이 suspended된 suspend-to-idle에서 보낸 total time입니다. |
Suspend-to-idle request count
238-245| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpuidle/state<N>/s2idle/usage |
| Date | 2018년 3월 |
| KernelVersion | v4.17 |
| Contact | Linux power management list <[email protected]> |
| Description | CPU가 suspend-to-idle에 진입하면서 이 state를 request한 total count입니다. |
CPU clock-frequency controls
246-260| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpufreq/* |
| Date | pre-git history |
| Contact | [email protected] |
| Description | CPU clock speed를 확인하고 변경합니다. Clock scaling을 사용하면 CPU clock speed를 실행 중에 변경할 수 있습니다. Clock speed가 낮을수록 CPU power consumption도 줄어들기 때문에 battery power를 절약하는 좋은 방법입니다. 이 directory에는 조정할 수 있는 많은 knob가 있습니다. 자세한 내용은 Documentation/cpu-freq/ 아래 file을 참조하십시오. |
CPU frequency coordination domain
261-275| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpufreq/freqdomain_cpus |
| Date | 2013년 6월 |
| Contact | [email protected] |
| Description | 같은 CPU frequency coordination domain의 CPU를 확인합니다. freqdomain_cpus는 같은 clock/frequency domain을 공유하는 online 및 offline CPU 목록이며, hardware level에서 공유할 수도 있습니다. 이 정보는 cpufreq core에 보이지 않을 수 있으므로 related_cpus 값이 freqdomain_cpus와 다를 수 있습니다. 이 attribute는 acpi-cpufreq를 사용하는 platform의 userspace DVFS controller가 더 나은 power/performance 결과를 얻는 데 유용합니다. acpi-cpufreq 또는 cppc-cpufreq driver를 사용할 때만 file이 존재합니다. |
AttributeMeaning
related_cpusCPUs related as visible to cpufreq core
freqdomain_cpusOnline + offline CPUs sharing clock/frequency domain
Core-visible policy relation과 hardware coordination domain이 다를 수 있습니다.
CPPC autonomous selection enable
276-290| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpufreq/auto_select |
| Date | 2025년 5월 |
| Contact | [email protected] |
| Description | Autonomous selection enable을 제어하는 read/write interface입니다. Read 결과 0은 autonomous selection disabled, 1은 enabled를 뜻합니다. 'y', '1', 'on'을 write하면 enable하고, 'n', '0', 'off'를 write하면 disable합니다. cppc-cpufreq driver를 사용할 때만 이 file이 존재합니다. |
ActionDisabledEnabled
Read01
Write aliasesn, 0, offy, 1, on
Read status와 accepted write aliases입니다.
CPPC autonomous activity window
291-313| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpufreq/auto_act_window |
| Date | 2025년 5월 |
| Contact | [email protected] |
| Description | Platform의 autonomous-selection policy에 moving utilization sensitivity window를 알려 줍니다. 이 file에서 microseconds 단위 autonomous activity window를 integer로 read/write합니다. Write 가능한 maximum value는 1270000000이지만 maximum significand는 127입니다. 따라서 128을 write하면 127이 저장됩니다. 값이 130보다 크면 첫 두 digit만 significand로 저장됩니다. 0을 write하면 workload에 따라 적절한 Activity Window를 platform이 결정할 수 있습니다. 이 file에 write하는 동작은 Autonomous Selection이 enabled일 때만 의미가 있습니다. cppc-cpufreq driver를 사용할 때만 file이 존재합니다. |
InputResult
0Platform selects an appropriate window
128127 stored because maximum significand is 127
> 130Only first two digits saved as significand
Maximum write1270000000
Microsecond input의 special handling을 원문 규칙대로 정리했습니다.
CPPC energy-performance preference value
314-330| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpufreq/energy_performance_preference_val |
| Date | 2025년 5월 |
| Contact | [email protected] |
| Description | 8-bit integer를 read/write하는 file입니다. 0(performance preference)부터 0xFF(energy-efficiency preference)까지의 값을 나타내며 performance 증가·감소 속도와 hardware의 energy-efficiency/performance optimization policy 결과에 영향을 줍니다. 이 file에 write하는 동작은 Autonomous Selection이 enabled일 때만 의미가 있습니다. cppc-cpufreq driver를 사용할 때만 file이 존재합니다. |
ValuePreference
0Performance
0xFFEnergy efficiency
8-bit range의 양 끝이 나타내는 optimization preference입니다.
AMD L3 cache-index disable slots
331-350| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1} |
| Date | 2008년 8월 |
| KernelVersion | 2.6.27 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | L3 cache index를 disable합니다. 이 file은 모든 CPU의 cache/index3 directory에 존재합니다. cache_disable_{0,1} file 각각은 cache index 하나를 disable하는 데 쓸 수 있는 disable slot 하나에 대응합니다. 이 기능이 있는 processor에서 file을 읽으면 해당 node에서 현재 disabled인 index를 반환합니다. Node마다 L3 structure가 하나 있으며 MCM machine에서는 internal node마다 하나입니다. Valid index를 write하면 지정한 cache index가 disabled됩니다. L3 cache가 있는 모든 AMD processor가 이 기능을 제공합니다. 자세한 내용은 https://www.amd.com/en/support/tech-docs?keyword=bios+kernel 의 BKDG를 참조하십시오. |
Select cache_disable_0 or cache_disable_1→Write a valid L3 cache index→Node or MCM internal-node L3 structure→Specified cache index disabled
각 slot은 node-local L3 index 하나를 선택해 disable합니다.
System-wide processor frequency boost
351-363| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpufreq/boost |
| Date | 2012년 8월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Processor frequency boosting을 제어합니다. 이 switch는 system 전체의 boost setting을 제어합니다. Boosting을 사용하면 CPU와 firmware가 nominal limit를 넘는 frequency에서 실행될 수 있습니다. 자세한 내용은 Documentation/admin-guide/pm/cpufreq.rst를 참조하십시오. |
Per-CPU crash-note address and size
364-375| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/crash_notes; /sys/devices/system/cpu/cpuX/crash_notes_size |
| Date | 2013년 4월 |
| Contact | [email protected] |
| Description | Per-CPU note의 address와 size를 제공합니다. crash_notes는 cpuX의 note를 보관하는 memory의 physical address이고, crash_notes_size는 cpuX note의 size입니다. |
Intel P-state performance limits
376-399| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/intel_pstate/max_perf_pct; /sys/devices/system/cpu/intel_pstate/min_perf_pct; /sys/devices/system/cpu/intel_pstate/no_turbo |
| Date | 2013년 2월 |
| Contact | [email protected] |
| Description | Intel P-state driver parameter입니다. Intel Sandybridge 이상 processor에서 current P-state를 선택하는 logic이며 세 knob가 driver가 request할 P-state limit를 제어합니다. max_perf_pct는 available performance의 percentage로 maximum requested P-state를 제한합니다. min_perf_pct는 available performance의 percentage로 minimum requested P-state를 제한합니다. no_turbo는 driver가 turbo-frequency range 아래의 P-state만 선택하도록 제한합니다. 자세한 내용은 Documentation/admin-guide/pm/intel_pstate.rst를 참조하십시오. |
AttributeLimit
max_perf_pctMaximum P-state as available-performance percentage
min_perf_pctMinimum P-state as available-performance percentage
no_turboRestrict selection below turbo range
Requested performance range와 turbo eligibility를 제한합니다.
CPU cache policy, geometry and sharing
400-458| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpu*/cache/index*/<set_of_attributes_mentioned_below> |
| Date | 2014년 7월(documented, 2008년 8월 이전부터 존재) |
| Contact | Sudeep Holla <[email protected]>; Linux kernel mailing list <[email protected]> |
| Description | CPU cache attribute parameter입니다. allocation_policy의 WriteAllocate는 write로 인한 cache miss 때 memory location을 cache line에 allocate하고, ReadAllocate는 read miss 때 allocate하며, ReadWriteAllocate는 두 방식을 모두 사용합니다. coherency_line_size는 memory에서 cache로 전송되는 minimum data amount를 bytes로 나타냅니다. level은 multi-level cache configuration의 cache hierarchy level입니다. number_of_sets는 cache의 total set 수이며 set은 같은 cache index를 가진 cache line collection입니다. physical_line_partition은 cache tag 하나당 physical cache line 수입니다. shared_cpu_list는 cache를 공유하는 logical CPU 목록이고 shared_cpu_map은 그 CPU 목록을 담은 logical CPU mask입니다. size는 total cache size(kB)입니다. type은 instruction만 담는 Instruction, data만 cache하는 Data, data와 instruction을 모두 담는 Unified 중 하나입니다. ways_of_associativity는 특정 memory block을 cache에 배치할 수 있는 자유도입니다. write_policy의 WriteThrough는 cache line과 lower-level memory block 모두에 data를 write하고, WriteBack은 cache line에만 write한 뒤 modified line이 교체될 때 main memory에 write합니다. |
AttributeMeaning
allocation_policyWriteAllocate, ReadAllocate or ReadWriteAllocate
coherency_line_sizeMinimum transfer size from memory, bytes
levelCache hierarchy level
number_of_setsTotal sets
physical_line_partitionPhysical lines per cache tag
shared_cpu_list / mapReadable list / logical CPU mask sharing cache
sizeTotal cache size in kB
typeInstruction, Data or Unified
ways_of_associativityPlacement freedom for a memory block
write_policyWriteThrough or WriteBack
Cache policy, geometry, sharing과 capacity attributes를 구조화했습니다.
Cache-instance identifier
459-473| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpu*/cache/index*/id |
| Date | 2016년 9월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | 특정 type의 cache instance 하나에 unique number를 부여합니다. 예를 들어 server의 socket마다 level-3 unified cache가 있다면 ID 0, 1, 2 등을 할당할 수 있습니다. ID 값은 contiguous하지 않을 수 있습니다. Level-1 cache는 보통 core마다 있지만 socket의 core 수가 2의 거듭제곱이 아닐 수 있으므로 0, 1, 2, 3, 4, 5, 8, 9, 10 등으로 번호가 매겨질 수 있습니다. |
POWERNV per-chip throttle statistics
474-523| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub_turbo_stat; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent; /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_reset |
| Date | 2016년 3월 |
| Contact | Linux kernel mailing list <[email protected]>; Linux for PowerPC mailing list <[email protected]> |
| Description | POWERNV CPUFreq driver의 frequency-throttle statistics directory와 attributes입니다. cpuX/cpufreq/throttle_stats에는 chip의 CPU-frequency throttle statistics가 들어 있으며 한 CPU의 statistics는 같은 chip의 모든 CPU에 공통입니다. turbo_stat은 turbo range, 즉 nominal frequency 이상에서 maximum frequency가 lower frequency로 throttle된 total count입니다. sub_turbo_stat은 sub-turbo range, 즉 nominal frequency 아래에서 maximum frequency가 lower frequency로 throttle된 total count입니다. unthrottle은 throttle 뒤 maximum frequency가 unthrottle된 total count입니다. powercap은 Power Capping 때문에, overtemp는 CPU Over Temperature 때문에, supply_fault는 Power Supply Failure 때문에, overcurrent는 Overcurrent 때문에, occ_reset은 OCC Reset 때문에 maximum frequency가 throttle된 total count입니다. powercap, overtemp, supply_fault, overcurrent, occ_reset attributes는 OCC firmware가 제공하는 frequency-throttling reason에 대응합니다. |
AttributeCounted event
turbo_statThrottle within at/above nominal turbo range
sub_turbo_statThrottle below nominal frequency
unthrottleMaximum frequency restored after throttle
powercapPower-capping throttle
overtempCPU over-temperature throttle
supply_faultPower-supply-failure throttle
overcurrentOvercurrent throttle
occ_resetOCC-reset throttle
Frequency range transition과 OCC firmware reason별 counters입니다.
Policy-view POWERNV throttle statistics
524-542| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpufreq/policyX/throttle_stats; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/turbo_stat; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/sub_turbo_stat; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/unthrottle; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/powercap; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overtemp; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/supply_fault; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/overcurrent; /sys/devices/system/cpu/cpufreq/policyX/throttle_stats/occ_reset |
| Date | 2016년 3월 |
| Contact | Linux kernel mailing list <[email protected]>; Linux for PowerPC mailing list <[email protected]> |
| Description | POWERNV CPUFreq driver의 frequency-throttle statistics directory와 attributes입니다. policyX/throttle_stats directory 및 모든 attribute는 /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory와 동일하며 chip의 frequency-throttle 정보를 제공합니다. |
AArch64 CPU identification registers
543-555| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/regs/; /sys/devices/system/cpu/cpuX/regs/identification/; /sys/devices/system/cpu/cpuX/regs/identification/midr_el1; /sys/devices/system/cpu/cpuX/regs/identification/revidr_el1; /sys/devices/system/cpu/cpuX/regs/identification/aidr_el1; /sys/devices/system/cpu/cpuX/regs/identification/smidr_el1 |
| Date | 2016년 6월 |
| Contact | Linux ARM Kernel Mailing list <[email protected]> |
| Description | AArch64 CPU registers입니다. identification directory는 CPU와 SMCU의 model 및 revision을 식별하기 위한 CPU ID registers를 expose합니다. |
FileDirectory
midr_el1cpuX/regs/identification/
revidr_el1cpuX/regs/identification/
aidr_el1cpuX/regs/identification/
smidr_el1cpuX/regs/identification/
Per-CPU identification directory에서 expose하는 registers입니다.
AArch32 userspace-capable CPU mask
556-564| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/aarch32_el0 |
| Date | 2021년 5월 |
| Contact | Linux ARM Kernel Mailing list <[email protected]> |
| Description | System에서 AArch32(32-bit ARM) application을 실행할 수 있는 CPU subset을 식별합니다. File이 존재하면 /sys/devices/system/cpu/{offline,online,possible,present}와 같은 형식을 사용합니다. File이 없으면 모든 CPU 또는 어떤 CPU도 AArch32 application을 실행할 수 없으며 execve()가 그에 맞게 동작합니다. |
Heterogeneous CPU capacity
565-571| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/cpu_capacity |
| Date | 2016년 12월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | CPU heterogeneity에 관한 정보입니다. cpu_capacity는 cpuX의 capacity를 나타냅니다. |
CPU vulnerability status files
572-605| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/vulnerabilities; /sys/devices/system/cpu/vulnerabilities/gather_data_sampling; /sys/devices/system/cpu/vulnerabilities/indirect_target_selection; /sys/devices/system/cpu/vulnerabilities/itlb_multihit; /sys/devices/system/cpu/vulnerabilities/l1tf; /sys/devices/system/cpu/vulnerabilities/mds; /sys/devices/system/cpu/vulnerabilities/meltdown; /sys/devices/system/cpu/vulnerabilities/mmio_stale_data; /sys/devices/system/cpu/vulnerabilities/old_microcode; /sys/devices/system/cpu/vulnerabilities/reg_file_data_sampling; /sys/devices/system/cpu/vulnerabilities/retbleed; /sys/devices/system/cpu/vulnerabilities/spec_store_bypass; /sys/devices/system/cpu/vulnerabilities/spectre_v1; /sys/devices/system/cpu/vulnerabilities/spectre_v2; /sys/devices/system/cpu/vulnerabilities/srbds; /sys/devices/system/cpu/vulnerabilities/tsa; /sys/devices/system/cpu/vulnerabilities/tsx_async_abort; /sys/devices/system/cpu/vulnerabilities/vmscape |
| Date | 2018년 1월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | CPU vulnerability 정보를 제공합니다. File은 CPU vulnerability code name으로 이름 붙이며 output은 system CPU의 상태를 반영합니다. "Not affected"는 CPU가 해당 vulnerability의 영향을 받지 않음을, "Vulnerable"은 영향을 받으며 mitigation이 적용되지 않았음을, "Mitigation: $M"은 영향을 받지만 mitigation $M이 적용 중임을 뜻합니다. 자세한 내용은 Documentation/admin-guide/hw-vuln/index.rst도 참조하십시오. |
OutputMeaning
Not affectedCPU is not affected
VulnerableAffected with no mitigation in effect
Mitigation: $MAffected with mitigation $M in effect
원문의 ASCII status table을 구조화했습니다.
Symmetric Multi-Threading control
606-631| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/smt; /sys/devices/system/cpu/smt/active; /sys/devices/system/cpu/smt/control |
| Date | 2018년 6월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Symmetric Multi Threading(SMT)을 제어합니다. active는 SMT가 active한지, 즉 enabled이며 sibling이 online인지 알려 줍니다. control은 SMT를 제어하는 read/write interface입니다. "on"은 SMT enabled, "off"는 disabled, "<N>"은 core당 N thread로 SMT enabled를 뜻합니다. "forceoff"는 변경할 수 없는 forced-disabled 상태, "notsupported"는 CPU가 SMT를 지원하지 않음, "notimplemented"는 architecture가 runtime SMT toggle을 구현하지 않았음을 뜻합니다. control status가 "forceoff" 또는 "notsupported"이면 write가 거부됩니다. PowerPC에서 SMT를 enable할 때는 offline core를 건너뜁니다. |
ValueMeaning / mutability
onSMT enabled
offSMT disabled
<N>Enabled with N threads per core
forceoffForced disabled; cannot be changed
notsupportedCPU does not support SMT; writes rejected
notimplementedArchitecture lacks runtime toggling
원문의 ASCII control-value table과 write restrictions입니다.
Intel Energy and Performance Bias Hint
632-649| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/power/energy_perf_bias |
| Date | 2019년 3월 |
| Contact | [email protected] |
| Description | 해당 CPU의 Intel Energy and Performance Bias Hint(EPB)입니다. 0부터 15까지의 sliding scale에서 0은 highest-performance preference, 15는 maximum energy savings에 해당합니다. CPU의 EPB를 바꾸려면 0-15 숫자 또는 의미를 반영하는 "performance", "balance-performance", "normal", "balance-power", "power" string 중 하나를 write합니다. Intel EPB 기능을 지원하는 모든 online CPU에 이 attribute가 존재합니다. |
InputPreference
0 / performanceHighest performance
balance-performanceBalanced toward performance
normalNeutral
balance-powerBalanced toward power saving
15 / powerMaximum energy savings
Numeric endpoint와 accepted semantic aliases입니다.
UMWAIT C0.2 and maximum-time controls
650-672| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/umwait_control; /sys/devices/system/cpu/umwait_control/enable_c02; /sys/devices/system/cpu/umwait_control/max_time |
| Date | 2019년 5월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | UMWAIT를 제어합니다. enable_c02는 UMWAIT C0.2 state의 read/write control입니다. Read 결과 0은 C0.2 disabled, 1은 enabled를 뜻합니다. 'y', '1', 'on'을 write하면 C0.2를 enable하고 'n', '0', 'off'를 write하면 disable하며 interface는 case-insensitive입니다. max_time은 CPU가 C0.1 또는 C0.2 state에 머무를 수 있는 UMWAIT maximum time을 TSC-quanta 단위로 제어하는 read/write interface입니다. Time은 unsigned 32-bit number이고 0은 limit가 없음을 뜻하며 low-order 2 bits는 0이어야 합니다. |
AttributeValues / constraints
enable_c02 read0 disabled, 1 enabled
enable_c02 writey/1/on or n/0/off, case-insensitive
max_timeUnsigned 32-bit TSC-quanta; 0 means unlimited
max_time alignmentLow-order two bits must be zero
C0.2 enable aliases와 maximum-time constraints입니다.
SEV-SNP guest privilege level
673-684| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/sev; /sys/devices/system/cpu/sev/vmpl |
| Date | 2024년 5월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | Secure Encrypted Virtualization(SEV) 정보입니다. SEV-SNP guest로 실행 중일 때만 이 directory가 존재합니다. vmpl은 SEV-SNP guest가 실행 중인 Virtual Machine Privilege Level(VMPL)을 보고합니다. |
POWER secure virtual-machine state
685-694| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/svm |
| Date | 2019년 8월 |
| Contact | Linux kernel mailing list <[email protected]>; Linux for PowerPC mailing list <[email protected]> |
| Description | Secure Virtual Machine 상태입니다. 값이 1이면 system이 POWER9 이상 processor의 Protected Execution Facility를 사용 중임을 뜻하며, 즉 Secure Virtual Machine입니다. |
Processor Utilization Resources Register ticks
695-705| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/purr |
| Date | 2005년 4월 |
| Contact | Linux for PowerPC mailing list <[email protected]> |
| Description | System boot 이후 이 CPU의 PURR ticks를 제공합니다. Processor Utilization Resources Register(PURR)는 CPU thread가 사용한 resource를 추정하는 64-bit counter입니다. Register 내용은 monotonically increase합니다. 이 sysfs interface는 cpuX의 PURR tick 수를 expose합니다. |
AttributeReported ticks
purrMonotonic processor-utilization estimate since boot
spurrFrequency-invariant utilization estimate since boot
idle_purrPURR ticks accumulated while idle
idle_spurrSPURR ticks accumulated while idle
전체 resource estimate와 idle-only tick views입니다.
Scaled Processor Utilization Resources ticks
706-717| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/spurr |
| Date | 2006년 12월 |
| Contact | Linux for PowerPC mailing list <[email protected]> |
| Description | System boot 이후 이 CPU의 SPURR ticks를 제공합니다. Scaled Processor Utilization Resources Register(SPURR)는 CPU thread가 사용한 resource를 frequency-invariant하게 추정하는 64-bit counter입니다. Register 내용은 monotonically increase합니다. 이 sysfs interface는 cpuX의 SPURR tick 수를 expose합니다. |
Idle PURR ticks
718-725| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/idle_purr |
| Date | 2020년 4월 |
| Contact | Linux for PowerPC mailing list <[email protected]> |
| Description | cpuX가 idle 상태였을 때의 PURR ticks입니다. 이 sysfs interface는 cpuX가 idle일 때 누적한 PURR tick 수를 expose합니다. |
Idle SPURR ticks
726-733| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/idle_spurr |
| Date | 2020년 4월 |
| Contact | Linux for PowerPC mailing list <[email protected]> |
| Description | cpuX가 idle 상태였을 때의 SPURR ticks입니다. 이 sysfs interface는 cpuX가 idle일 때 누적한 SPURR tick 수를 expose합니다. |
Preferred ARM MTE tag-checking mode
734-751| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/cpuX/mte_tcf_preferred |
| Date | 2021년 7월 |
| Contact | Linux ARM Kernel Mailing list <[email protected]> |
| Description | Preferred MTE tag-checking mode입니다. User program이 둘 이상의 MTE tag-checking mode를 지정하면 이 sysfs node가 해당 CPU에 task를 schedule할 때 우선할 mode를 지정합니다. "sync"는 synchronous mode, "asymm"은 asymmetric mode, "async"는 asynchronous mode를 우선합니다. Documentation/arch/arm64/memory-tagging-extension.rst도 참조하십시오. |
ValuePreferred mode
syncSynchronous
asymmAsymmetric
asyncAsynchronous
원문의 ASCII value table을 구조화했습니다.
Full-dynticks CPU list
752-758| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/nohz_full |
| Date | 2015년 4월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | (RO) nohz_full mode에 있는 CPU 목록입니다. 이 CPU들은 boot parameter "nohz_full="로 설정됩니다. |
AttributeBoot parameterEffect
nohz_fullnohz_full=CPUs in full-dynticks mode
isolatedisolcpus=CPUs excluded from load balancing
Tick suppression과 scheduler load-balancing isolation을 구분합니다.
Load-balancing-isolated CPU list
759-766| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/isolated |
| Date | 2015년 4월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | (RO) Isolated되어 load balancing에 참여하지 않는 CPU 목록입니다. 이 CPU들은 boot parameter "isolcpus="로 설정됩니다. |
Kexec crash-segment hotplug updating
767-774| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/crash_hotplug |
| Date | 2023년 8월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | (RO) Memory hot-unplug/plug 및 CPU offline/online event에서 kernel이 관련 kexec segment를 update하는지 나타냅니다. 이 기능을 사용하면 kdump kernel을 reload할 필요가 없습니다. |
Memory hotplug or CPU online/offline event→Kernel updates relevant kexec segments→Existing kdump kernel remains usable→No kdump-kernel reload required
지원 kernel은 topology 변화에 맞춰 kexec segment를 갱신합니다.
CPUs eligible to be brought online
775-779| 항목 | 한국어 전문 번역 |
|---|---|
| What | /sys/devices/system/cpu/enabled |
| Date | 2022년 11월 |
| Contact | Linux kernel mailing list <[email protected]> |
| Description | (RO) Online으로 전환할 수 있는 CPU 목록입니다. |
CPU hierarchy, hotplug and topology
sysfs-devices-system-cpu:1-109Global CPU directory와 logical CPU instances, online/offline/possible/present masks, architecture-specific probe/release, NUMA node 및 package/core/thread topology를 제공합니다.