1 =================================== 1 ===================================
2 Documentation for /proc/sys/kernel/ 2 Documentation for /proc/sys/kernel/
3 =================================== 3 ===================================
4 4
5 .. See scripts/check-sysctl-docs to keep this 5 .. See scripts/check-sysctl-docs to keep this up to date
6 6
7 7
8 Copyright (c) 1998, 1999, Rik van Riel <riel@n 8 Copyright (c) 1998, 1999, Rik van Riel <riel@nl.linux.org>
9 9
10 Copyright (c) 2009, Shen Feng<shen@cn.fu 10 Copyright (c) 2009, Shen Feng<shen@cn.fujitsu.com>
11 11
12 For general info and legal blurb, please look 12 For general info and legal blurb, please look in
13 Documentation/admin-guide/sysctl/index.rst. 13 Documentation/admin-guide/sysctl/index.rst.
14 14
15 ---------------------------------------------- 15 ------------------------------------------------------------------------------
16 16
17 This file contains documentation for the sysct 17 This file contains documentation for the sysctl files in
18 ``/proc/sys/kernel/``. 18 ``/proc/sys/kernel/``.
19 19
20 The files in this directory can be used to tun 20 The files in this directory can be used to tune and monitor
21 miscellaneous and general things in the operat 21 miscellaneous and general things in the operation of the Linux
22 kernel. Since some of the files *can* be used 22 kernel. Since some of the files *can* be used to screw up your
23 system, it is advisable to read both documenta 23 system, it is advisable to read both documentation and source
24 before actually making adjustments. 24 before actually making adjustments.
25 25
26 Currently, these files might (depending on you 26 Currently, these files might (depending on your configuration)
27 show up in ``/proc/sys/kernel``: 27 show up in ``/proc/sys/kernel``:
28 28
29 .. contents:: :local: 29 .. contents:: :local:
30 30
31 31
32 acct 32 acct
33 ==== 33 ====
34 34
35 :: 35 ::
36 36
37 highwater lowwater frequency 37 highwater lowwater frequency
38 38
39 If BSD-style process accounting is enabled the 39 If BSD-style process accounting is enabled these values control
40 its behaviour. If free space on filesystem whe 40 its behaviour. If free space on filesystem where the log lives
41 goes below ``lowwater``\ % accounting suspends !! 41 goes below ``lowwater``% accounting suspends. If free space gets
42 above ``highwater``\ % accounting resumes. ``f !! 42 above ``highwater``% accounting resumes. ``frequency`` determines
43 how often do we check the amount of free space 43 how often do we check the amount of free space (value is in
44 seconds). Default: 44 seconds). Default:
45 45
46 :: 46 ::
47 47
48 4 2 30 48 4 2 30
49 49
50 That is, suspend accounting if free space drop 50 That is, suspend accounting if free space drops below 2%; resume it
51 if it increases to at least 4%; consider infor 51 if it increases to at least 4%; consider information about amount of
52 free space valid for 30 seconds. 52 free space valid for 30 seconds.
53 53
54 54
55 acpi_video_flags 55 acpi_video_flags
56 ================ 56 ================
57 57
58 See Documentation/power/video.rst. This allows 58 See Documentation/power/video.rst. This allows the video resume mode to be set,
59 in a similar fashion to the ``acpi_sleep`` ker 59 in a similar fashion to the ``acpi_sleep`` kernel parameter, by
60 combining the following values: 60 combining the following values:
61 61
62 = ======= 62 = =======
63 1 s3_bios 63 1 s3_bios
64 2 s3_mode 64 2 s3_mode
65 4 s3_beep 65 4 s3_beep
66 = ======= 66 = =======
67 67
68 arch <<
69 ==== <<
70 <<
71 The machine hardware name, the same output as <<
72 (e.g. ``x86_64`` or ``aarch64``). <<
73 68
74 auto_msgmni 69 auto_msgmni
75 =========== 70 ===========
76 71
77 This variable has no effect and may be removed 72 This variable has no effect and may be removed in future kernel
78 releases. Reading it always returns 0. 73 releases. Reading it always returns 0.
79 Up to Linux 3.17, it enabled/disabled automati 74 Up to Linux 3.17, it enabled/disabled automatic recomputing of
80 `msgmni`_ 75 `msgmni`_
81 upon memory add/remove or upon IPC namespace c 76 upon memory add/remove or upon IPC namespace creation/removal.
82 Echoing "1" into this file enabled msgmni auto 77 Echoing "1" into this file enabled msgmni automatic recomputing.
83 Echoing "0" turned it off. The default value w 78 Echoing "0" turned it off. The default value was 1.
84 79
85 80
86 bootloader_type (x86 only) 81 bootloader_type (x86 only)
87 ========================== 82 ==========================
88 83
89 This gives the bootloader type number as indic 84 This gives the bootloader type number as indicated by the bootloader,
90 shifted left by 4, and OR'd with the low four 85 shifted left by 4, and OR'd with the low four bits of the bootloader
91 version. The reason for this encoding is that 86 version. The reason for this encoding is that this used to match the
92 ``type_of_loader`` field in the kernel header; 87 ``type_of_loader`` field in the kernel header; the encoding is kept for
93 backwards compatibility. That is, if the full 88 backwards compatibility. That is, if the full bootloader type number
94 is 0x15 and the full version number is 0x234, 89 is 0x15 and the full version number is 0x234, this file will contain
95 the value 340 = 0x154. 90 the value 340 = 0x154.
96 91
97 See the ``type_of_loader`` and ``ext_loader_ty 92 See the ``type_of_loader`` and ``ext_loader_type`` fields in
98 Documentation/arch/x86/boot.rst for additional !! 93 Documentation/x86/boot.rst for additional information.
99 94
100 95
101 bootloader_version (x86 only) 96 bootloader_version (x86 only)
102 ============================= 97 =============================
103 98
104 The complete bootloader version number. In th 99 The complete bootloader version number. In the example above, this
105 file will contain the value 564 = 0x234. 100 file will contain the value 564 = 0x234.
106 101
107 See the ``type_of_loader`` and ``ext_loader_ve 102 See the ``type_of_loader`` and ``ext_loader_ver`` fields in
108 Documentation/arch/x86/boot.rst for additional !! 103 Documentation/x86/boot.rst for additional information.
109 104
110 105
111 bpf_stats_enabled 106 bpf_stats_enabled
112 ================= 107 =================
113 108
114 Controls whether the kernel should collect sta 109 Controls whether the kernel should collect statistics on BPF programs
115 (total time spent running, number of times run 110 (total time spent running, number of times run...). Enabling
116 statistics causes a slight reduction in perfor 111 statistics causes a slight reduction in performance on each program
117 run. The statistics can be seen using ``bpftoo 112 run. The statistics can be seen using ``bpftool``.
118 113
119 = =================================== 114 = ===================================
120 0 Don't collect statistics (default). 115 0 Don't collect statistics (default).
121 1 Collect statistics. 116 1 Collect statistics.
122 = =================================== 117 = ===================================
123 118
124 119
125 cad_pid 120 cad_pid
126 ======= 121 =======
127 122
128 This is the pid which will be signalled on reb 123 This is the pid which will be signalled on reboot (notably, by
129 Ctrl-Alt-Delete). Writing a value to this file 124 Ctrl-Alt-Delete). Writing a value to this file which doesn't
130 correspond to a running process will result in 125 correspond to a running process will result in ``-ESRCH``.
131 126
132 See also `ctrl-alt-del`_. 127 See also `ctrl-alt-del`_.
133 128
134 129
135 cap_last_cap 130 cap_last_cap
136 ============ 131 ============
137 132
138 Highest valid capability of the running kernel 133 Highest valid capability of the running kernel. Exports
139 ``CAP_LAST_CAP`` from the kernel. 134 ``CAP_LAST_CAP`` from the kernel.
140 135
141 136
142 .. _core_pattern: <<
143 <<
144 core_pattern 137 core_pattern
145 ============ 138 ============
146 139
147 ``core_pattern`` is used to specify a core dum 140 ``core_pattern`` is used to specify a core dumpfile pattern name.
148 141
149 * max length 127 characters; default value is 142 * max length 127 characters; default value is "core"
150 * ``core_pattern`` is used as a pattern templa 143 * ``core_pattern`` is used as a pattern template for the output
151 filename; certain string patterns (beginning 144 filename; certain string patterns (beginning with '%') are
152 substituted with their actual values. 145 substituted with their actual values.
153 * backward compatibility with ``core_uses_pid` 146 * backward compatibility with ``core_uses_pid``:
154 147
155 If ``core_pattern`` does not include " 148 If ``core_pattern`` does not include "%p" (default does not)
156 and ``core_uses_pid`` is set, then .PI 149 and ``core_uses_pid`` is set, then .PID will be appended to
157 the filename. 150 the filename.
158 151
159 * corename format specifiers 152 * corename format specifiers
160 153
161 ======== ====================== 154 ======== ==========================================
162 %<NUL> '%' is dropped 155 %<NUL> '%' is dropped
163 %% output one '%' 156 %% output one '%'
164 %p pid 157 %p pid
165 %P global pid (init PID n 158 %P global pid (init PID namespace)
166 %i tid 159 %i tid
167 %I global tid (init PID n 160 %I global tid (init PID namespace)
168 %u uid (in initial user n 161 %u uid (in initial user namespace)
169 %g gid (in initial user n 162 %g gid (in initial user namespace)
170 %d dump mode, matches ``P 163 %d dump mode, matches ``PR_SET_DUMPABLE`` and
171 ``/proc/sys/fs/suid_du 164 ``/proc/sys/fs/suid_dumpable``
172 %s signal number 165 %s signal number
173 %t UNIX time of dump 166 %t UNIX time of dump
174 %h hostname 167 %h hostname
175 %e executable filename (m 168 %e executable filename (may be shortened, could be changed by prctl etc)
176 %f executable filename 169 %f executable filename
177 %E executable path 170 %E executable path
178 %c maximum size of core f 171 %c maximum size of core file by resource limit RLIMIT_CORE
179 %C CPU the task ran on <<
180 %<OTHER> both are dropped 172 %<OTHER> both are dropped
181 ======== ====================== 173 ======== ==========================================
182 174
183 * If the first character of the pattern is a ' 175 * If the first character of the pattern is a '|', the kernel will treat
184 the rest of the pattern as a command to run. 176 the rest of the pattern as a command to run. The core dump will be
185 written to the standard input of that progra 177 written to the standard input of that program instead of to a file.
186 178
187 179
188 core_pipe_limit 180 core_pipe_limit
189 =============== 181 ===============
190 182
191 This sysctl is only applicable when `core_patt 183 This sysctl is only applicable when `core_pattern`_ is configured to
192 pipe core files to a user space helper (when t 184 pipe core files to a user space helper (when the first character of
193 ``core_pattern`` is a '|', see above). 185 ``core_pattern`` is a '|', see above).
194 When collecting cores via a pipe to an applica 186 When collecting cores via a pipe to an application, it is occasionally
195 useful for the collecting application to gathe 187 useful for the collecting application to gather data about the
196 crashing process from its ``/proc/pid`` direct 188 crashing process from its ``/proc/pid`` directory.
197 In order to do this safely, the kernel must wa 189 In order to do this safely, the kernel must wait for the collecting
198 process to exit, so as not to remove the crash 190 process to exit, so as not to remove the crashing processes proc files
199 prematurely. 191 prematurely.
200 This in turn creates the possibility that a mi 192 This in turn creates the possibility that a misbehaving userspace
201 collecting process can block the reaping of a 193 collecting process can block the reaping of a crashed process simply
202 by never exiting. 194 by never exiting.
203 This sysctl defends against that. 195 This sysctl defends against that.
204 It defines how many concurrent crashing proces 196 It defines how many concurrent crashing processes may be piped to user
205 space applications in parallel. 197 space applications in parallel.
206 If this value is exceeded, then those crashing 198 If this value is exceeded, then those crashing processes above that
207 value are noted via the kernel log and their c 199 value are noted via the kernel log and their cores are skipped.
208 0 is a special value, indicating that unlimite 200 0 is a special value, indicating that unlimited processes may be
209 captured in parallel, but that no waiting will 201 captured in parallel, but that no waiting will take place (i.e. the
210 collecting process is not guaranteed access to 202 collecting process is not guaranteed access to ``/proc/<crashing
211 pid>/``). 203 pid>/``).
212 This value defaults to 0. 204 This value defaults to 0.
213 205
214 206
215 core_uses_pid 207 core_uses_pid
216 ============= 208 =============
217 209
218 The default coredump filename is "core". By s 210 The default coredump filename is "core". By setting
219 ``core_uses_pid`` to 1, the coredump filename 211 ``core_uses_pid`` to 1, the coredump filename becomes core.PID.
220 If `core_pattern`_ does not include "%p" (defa 212 If `core_pattern`_ does not include "%p" (default does not)
221 and ``core_uses_pid`` is set, then .PID will b 213 and ``core_uses_pid`` is set, then .PID will be appended to
222 the filename. 214 the filename.
223 215
224 216
225 ctrl-alt-del 217 ctrl-alt-del
226 ============ 218 ============
227 219
228 When the value in this file is 0, ctrl-alt-del 220 When the value in this file is 0, ctrl-alt-del is trapped and
229 sent to the ``init(1)`` program to handle a gr 221 sent to the ``init(1)`` program to handle a graceful restart.
230 When, however, the value is > 0, Linux's react 222 When, however, the value is > 0, Linux's reaction to a Vulcan
231 Nerve Pinch (tm) will be an immediate reboot, 223 Nerve Pinch (tm) will be an immediate reboot, without even
232 syncing its dirty buffers. 224 syncing its dirty buffers.
233 225
234 Note: 226 Note:
235 when a program (like dosemu) has the keyboar 227 when a program (like dosemu) has the keyboard in 'raw'
236 mode, the ctrl-alt-del is intercepted by the 228 mode, the ctrl-alt-del is intercepted by the program before it
237 ever reaches the kernel tty layer, and it's 229 ever reaches the kernel tty layer, and it's up to the program
238 to decide what to do with it. 230 to decide what to do with it.
239 231
240 232
241 dmesg_restrict 233 dmesg_restrict
242 ============== 234 ==============
243 235
244 This toggle indicates whether unprivileged use 236 This toggle indicates whether unprivileged users are prevented
245 from using ``dmesg(8)`` to view messages from 237 from using ``dmesg(8)`` to view messages from the kernel's log
246 buffer. 238 buffer.
247 When ``dmesg_restrict`` is set to 0 there are 239 When ``dmesg_restrict`` is set to 0 there are no restrictions.
248 When ``dmesg_restrict`` is set to 1, users mus 240 When ``dmesg_restrict`` is set to 1, users must have
249 ``CAP_SYSLOG`` to use ``dmesg(8)``. 241 ``CAP_SYSLOG`` to use ``dmesg(8)``.
250 242
251 The kernel config option ``CONFIG_SECURITY_DME 243 The kernel config option ``CONFIG_SECURITY_DMESG_RESTRICT`` sets the
252 default value of ``dmesg_restrict``. 244 default value of ``dmesg_restrict``.
253 245
254 246
255 domainname & hostname 247 domainname & hostname
256 ===================== 248 =====================
257 249
258 These files can be used to set the NIS/YP doma 250 These files can be used to set the NIS/YP domainname and the
259 hostname of your box in exactly the same way a 251 hostname of your box in exactly the same way as the commands
260 domainname and hostname, i.e.:: 252 domainname and hostname, i.e.::
261 253
262 # echo "darkstar" > /proc/sys/kernel/h 254 # echo "darkstar" > /proc/sys/kernel/hostname
263 # echo "mydomain" > /proc/sys/kernel/d 255 # echo "mydomain" > /proc/sys/kernel/domainname
264 256
265 has the same effect as:: 257 has the same effect as::
266 258
267 # hostname "darkstar" 259 # hostname "darkstar"
268 # domainname "mydomain" 260 # domainname "mydomain"
269 261
270 Note, however, that the classic darkstar.frop. 262 Note, however, that the classic darkstar.frop.org has the
271 hostname "darkstar" and DNS (Internet Domain N 263 hostname "darkstar" and DNS (Internet Domain Name Server)
272 domainname "frop.org", not to be confused with 264 domainname "frop.org", not to be confused with the NIS (Network
273 Information Service) or YP (Yellow Pages) doma 265 Information Service) or YP (Yellow Pages) domainname. These two
274 domain names are in general different. For a d 266 domain names are in general different. For a detailed discussion
275 see the ``hostname(1)`` man page. 267 see the ``hostname(1)`` man page.
276 268
277 269
278 firmware_config 270 firmware_config
279 =============== 271 ===============
280 272
281 See Documentation/driver-api/firmware/fallback 273 See Documentation/driver-api/firmware/fallback-mechanisms.rst.
282 274
283 The entries in this directory allow the firmwa 275 The entries in this directory allow the firmware loader helper
284 fallback to be controlled: 276 fallback to be controlled:
285 277
286 * ``force_sysfs_fallback``, when set to 1, for 278 * ``force_sysfs_fallback``, when set to 1, forces the use of the
287 fallback; 279 fallback;
288 * ``ignore_sysfs_fallback``, when set to 1, ig 280 * ``ignore_sysfs_fallback``, when set to 1, ignores any fallback.
289 281
290 282
291 ftrace_dump_on_oops 283 ftrace_dump_on_oops
292 =================== 284 ===================
293 285
294 Determines whether ``ftrace_dump()`` should be 286 Determines whether ``ftrace_dump()`` should be called on an oops (or
295 kernel panic). This will output the contents o 287 kernel panic). This will output the contents of the ftrace buffers to
296 the console. This is very useful for capturin 288 the console. This is very useful for capturing traces that lead to
297 crashes and outputting them to a serial consol 289 crashes and outputting them to a serial console.
298 290
299 ======================= ====================== !! 291 = ===================================================
300 0 Disabled (default). !! 292 0 Disabled (default).
301 1 Dump buffers of all CP !! 293 1 Dump buffers of all CPUs.
302 2(orig_cpu) Dump the buffer of the !! 294 2 Dump the buffer of the CPU that triggered the oops.
303 oops. !! 295 = ===================================================
304 <instance> Dump the specific inst <<
305 <instance>=2(orig_cpu) Dump the specific inst <<
306 that triggered the oop <<
307 ======================= ====================== <<
308 <<
309 Multiple instance dump is also supported, and <<
310 by commas. If global buffer also needs to be d <<
311 the dump mode (1/2/orig_cpu) first for global <<
312 <<
313 So for example to dump "foo" and "bar" instanc <<
314 user can:: <<
315 <<
316 echo "foo,bar" > /proc/sys/kernel/ftrace_dum <<
317 <<
318 To dump global buffer and "foo" instance buffe <<
319 CPUs along with the "bar" instance buffer on C <<
320 oops, user can:: <<
321 296
322 echo "1,foo,bar=2" > /proc/sys/kernel/ftrace <<
323 297
324 ftrace_enabled, stack_tracer_enabled 298 ftrace_enabled, stack_tracer_enabled
325 ==================================== 299 ====================================
326 300
327 See Documentation/trace/ftrace.rst. 301 See Documentation/trace/ftrace.rst.
328 302
329 303
330 hardlockup_all_cpu_backtrace 304 hardlockup_all_cpu_backtrace
331 ============================ 305 ============================
332 306
333 This value controls the hard lockup detector b 307 This value controls the hard lockup detector behavior when a hard
334 lockup condition is detected as to whether or 308 lockup condition is detected as to whether or not to gather further
335 debug information. If enabled, arch-specific a 309 debug information. If enabled, arch-specific all-CPU stack dumping
336 will be initiated. 310 will be initiated.
337 311
338 = ============================================ 312 = ============================================
339 0 Do nothing. This is the default behavior. 313 0 Do nothing. This is the default behavior.
340 1 On detection capture more debug information. 314 1 On detection capture more debug information.
341 = ============================================ 315 = ============================================
342 316
343 317
344 hardlockup_panic 318 hardlockup_panic
345 ================ 319 ================
346 320
347 This parameter can be used to control whether 321 This parameter can be used to control whether the kernel panics
348 when a hard lockup is detected. 322 when a hard lockup is detected.
349 323
350 = =========================== 324 = ===========================
351 0 Don't panic on hard lockup. 325 0 Don't panic on hard lockup.
352 1 Panic on hard lockup. 326 1 Panic on hard lockup.
353 = =========================== 327 = ===========================
354 328
355 See Documentation/admin-guide/lockup-watchdogs 329 See Documentation/admin-guide/lockup-watchdogs.rst for more information.
356 This can also be set using the nmi_watchdog ke 330 This can also be set using the nmi_watchdog kernel parameter.
357 331
358 332
359 hotplug 333 hotplug
360 ======= 334 =======
361 335
362 Path for the hotplug policy agent. 336 Path for the hotplug policy agent.
363 Default value is ``CONFIG_UEVENT_HELPER_PATH`` 337 Default value is ``CONFIG_UEVENT_HELPER_PATH``, which in turn defaults
364 to the empty string. 338 to the empty string.
365 339
366 This file only exists when ``CONFIG_UEVENT_HEL 340 This file only exists when ``CONFIG_UEVENT_HELPER`` is enabled. Most
367 modern systems rely exclusively on the netlink 341 modern systems rely exclusively on the netlink-based uevent source and
368 don't need this. 342 don't need this.
369 343
370 344
371 hung_task_all_cpu_backtrace 345 hung_task_all_cpu_backtrace
372 =========================== 346 ===========================
373 347
374 If this option is set, the kernel will send an 348 If this option is set, the kernel will send an NMI to all CPUs to dump
375 their backtraces when a hung task is detected. 349 their backtraces when a hung task is detected. This file shows up if
376 CONFIG_DETECT_HUNG_TASK and CONFIG_SMP are ena 350 CONFIG_DETECT_HUNG_TASK and CONFIG_SMP are enabled.
377 351
378 0: Won't show all CPUs backtraces when a hung 352 0: Won't show all CPUs backtraces when a hung task is detected.
379 This is the default behavior. 353 This is the default behavior.
380 354
381 1: Will non-maskably interrupt all CPUs and du 355 1: Will non-maskably interrupt all CPUs and dump their backtraces when
382 a hung task is detected. 356 a hung task is detected.
383 357
384 358
385 hung_task_panic 359 hung_task_panic
386 =============== 360 ===============
387 361
388 Controls the kernel's behavior when a hung tas 362 Controls the kernel's behavior when a hung task is detected.
389 This file shows up if ``CONFIG_DETECT_HUNG_TAS 363 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
390 364
391 = ============================================ 365 = =================================================
392 0 Continue operation. This is the default beha 366 0 Continue operation. This is the default behavior.
393 1 Panic immediately. 367 1 Panic immediately.
394 = ============================================ 368 = =================================================
395 369
396 370
397 hung_task_check_count 371 hung_task_check_count
398 ===================== 372 =====================
399 373
400 The upper bound on the number of tasks that ar 374 The upper bound on the number of tasks that are checked.
401 This file shows up if ``CONFIG_DETECT_HUNG_TAS 375 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
402 376
403 377
404 hung_task_timeout_secs 378 hung_task_timeout_secs
405 ====================== 379 ======================
406 380
407 When a task in D state did not get scheduled 381 When a task in D state did not get scheduled
408 for more than this value report a warning. 382 for more than this value report a warning.
409 This file shows up if ``CONFIG_DETECT_HUNG_TAS 383 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
410 384
411 0 means infinite timeout, no checking is done. 385 0 means infinite timeout, no checking is done.
412 386
413 Possible values to set are in range {0:``LONG_ 387 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
414 388
415 389
416 hung_task_check_interval_secs 390 hung_task_check_interval_secs
417 ============================= 391 =============================
418 392
419 Hung task check interval. If hung task checkin 393 Hung task check interval. If hung task checking is enabled
420 (see `hung_task_timeout_secs`_), the check is 394 (see `hung_task_timeout_secs`_), the check is done every
421 ``hung_task_check_interval_secs`` seconds. 395 ``hung_task_check_interval_secs`` seconds.
422 This file shows up if ``CONFIG_DETECT_HUNG_TAS 396 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
423 397
424 0 (default) means use ``hung_task_timeout_secs 398 0 (default) means use ``hung_task_timeout_secs`` as checking
425 interval. 399 interval.
426 400
427 Possible values to set are in range {0:``LONG_ 401 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
428 402
429 403
430 hung_task_warnings 404 hung_task_warnings
431 ================== 405 ==================
432 406
433 The maximum number of warnings to report. Duri 407 The maximum number of warnings to report. During a check interval
434 if a hung task is detected, this value is decr 408 if a hung task is detected, this value is decreased by 1.
435 When this value reaches 0, no more warnings wi 409 When this value reaches 0, no more warnings will be reported.
436 This file shows up if ``CONFIG_DETECT_HUNG_TAS 410 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
437 411
438 -1: report an infinite number of warnings. 412 -1: report an infinite number of warnings.
439 413
440 414
441 hyperv_record_panic_msg 415 hyperv_record_panic_msg
442 ======================= 416 =======================
443 417
444 Controls whether the panic kmsg data should be 418 Controls whether the panic kmsg data should be reported to Hyper-V.
445 419
446 = ============================================ 420 = =========================================================
447 0 Do not report panic kmsg data. 421 0 Do not report panic kmsg data.
448 1 Report the panic kmsg data. This is the defa 422 1 Report the panic kmsg data. This is the default behavior.
449 = ============================================ 423 = =========================================================
450 424
451 425
452 ignore-unaligned-usertrap 426 ignore-unaligned-usertrap
453 ========================= 427 =========================
454 428
455 On architectures where unaligned accesses caus 429 On architectures where unaligned accesses cause traps, and where this
456 feature is supported (``CONFIG_SYSCTL_ARCH_UNA 430 feature is supported (``CONFIG_SYSCTL_ARCH_UNALIGN_NO_WARN``;
457 currently, ``arc``, ``parisc`` and ``loongarch !! 431 currently, ``arc`` and ``ia64``), controls whether all unaligned traps
458 unaligned traps are logged. !! 432 are logged.
459 433
460 = ============================================ 434 = =============================================================
461 0 Log all unaligned accesses. 435 0 Log all unaligned accesses.
462 1 Only warn the first time a process traps. Th 436 1 Only warn the first time a process traps. This is the default
463 setting. 437 setting.
464 = ============================================ 438 = =============================================================
465 439
466 See also `unaligned-trap`_. !! 440 See also `unaligned-trap`_ and `unaligned-dump-stack`_. On ``ia64``,
467 !! 441 this allows system administrators to override the
468 io_uring_disabled !! 442 ``IA64_THREAD_UAC_NOPRINT`` ``prctl`` and avoid logs being flooded.
469 ================= <<
470 <<
471 Prevents all processes from creating new io_ur <<
472 shrinks the kernel's attack surface. <<
473 <<
474 = ============================================ <<
475 0 All processes can create io_uring instances <<
476 default setting. <<
477 1 io_uring creation is disabled (io_uring_setu <<
478 -EPERM) for unprivileged processes not in th <<
479 Existing io_uring instances can still be use <<
480 documentation for io_uring_group for more in <<
481 2 io_uring creation is disabled for all proces <<
482 always fails with -EPERM. Existing io_uring <<
483 used. <<
484 = ============================================ <<
485 <<
486 <<
487 io_uring_group <<
488 ============== <<
489 <<
490 When io_uring_disabled is set to 1, a process <<
491 privileged (CAP_SYS_ADMIN) or be in the io_uri <<
492 to create an io_uring instance. If io_uring_g <<
493 default), only processes with the CAP_SYS_ADMI <<
494 io_uring instances. <<
495 443
496 444
497 kexec_load_disabled 445 kexec_load_disabled
498 =================== 446 ===================
499 447
500 A toggle indicating if the syscalls ``kexec_lo !! 448 A toggle indicating if the ``kexec_load`` syscall has been disabled.
501 ``kexec_file_load`` have been disabled. !! 449 This value defaults to 0 (false: ``kexec_load`` enabled), but can be
502 This value defaults to 0 (false: ``kexec_*load !! 450 set to 1 (true: ``kexec_load`` disabled).
503 set to 1 (true: ``kexec_*load`` disabled). <<
504 Once true, kexec can no longer be used, and th 451 Once true, kexec can no longer be used, and the toggle cannot be set
505 back to false. 452 back to false.
506 This allows a kexec image to be loaded before 453 This allows a kexec image to be loaded before disabling the syscall,
507 allowing a system to set up (and later use) an 454 allowing a system to set up (and later use) an image without it being
508 altered. 455 altered.
509 Generally used together with the `modules_disa 456 Generally used together with the `modules_disabled`_ sysctl.
510 457
511 kexec_load_limit_panic <<
512 ====================== <<
513 <<
514 This parameter specifies a limit to the number <<
515 ``kexec_load`` and ``kexec_file_load`` can be <<
516 image. It can only be set with a more restrict <<
517 current one. <<
518 <<
519 == =========================================== <<
520 -1 Unlimited calls to kexec. This is the defau <<
521 N Number of calls left. <<
522 == =========================================== <<
523 <<
524 kexec_load_limit_reboot <<
525 ======================= <<
526 <<
527 Similar functionality as ``kexec_load_limit_pa <<
528 image. <<
529 458
530 kptr_restrict 459 kptr_restrict
531 ============= 460 =============
532 461
533 This toggle indicates whether restrictions are 462 This toggle indicates whether restrictions are placed on
534 exposing kernel addresses via ``/proc`` and ot 463 exposing kernel addresses via ``/proc`` and other interfaces.
535 464
536 When ``kptr_restrict`` is set to 0 (the defaul 465 When ``kptr_restrict`` is set to 0 (the default) the address is hashed
537 before printing. 466 before printing.
538 (This is the equivalent to %p.) 467 (This is the equivalent to %p.)
539 468
540 When ``kptr_restrict`` is set to 1, kernel poi 469 When ``kptr_restrict`` is set to 1, kernel pointers printed using the
541 %pK format specifier will be replaced with 0s 470 %pK format specifier will be replaced with 0s unless the user has
542 ``CAP_SYSLOG`` and effective user and group id 471 ``CAP_SYSLOG`` and effective user and group ids are equal to the real
543 ids. 472 ids.
544 This is because %pK checks are done at read() 473 This is because %pK checks are done at read() time rather than open()
545 time, so if permissions are elevated between t 474 time, so if permissions are elevated between the open() and the read()
546 (e.g via a setuid binary) then %pK will not le 475 (e.g via a setuid binary) then %pK will not leak kernel pointers to
547 unprivileged users. 476 unprivileged users.
548 Note, this is a temporary solution only. 477 Note, this is a temporary solution only.
549 The correct long-term solution is to do the pe 478 The correct long-term solution is to do the permission checks at
550 open() time. 479 open() time.
551 Consider removing world read permissions from 480 Consider removing world read permissions from files that use %pK, and
552 using `dmesg_restrict`_ to protect against use 481 using `dmesg_restrict`_ to protect against uses of %pK in ``dmesg(8)``
553 if leaking kernel pointer values to unprivileg 482 if leaking kernel pointer values to unprivileged users is a concern.
554 483
555 When ``kptr_restrict`` is set to 2, kernel poi 484 When ``kptr_restrict`` is set to 2, kernel pointers printed using
556 %pK will be replaced with 0s regardless of pri 485 %pK will be replaced with 0s regardless of privileges.
557 486
558 487
559 modprobe 488 modprobe
560 ======== 489 ========
561 490
562 The full path to the usermode helper for autol 491 The full path to the usermode helper for autoloading kernel modules,
563 by default ``CONFIG_MODPROBE_PATH``, which in 492 by default ``CONFIG_MODPROBE_PATH``, which in turn defaults to
564 "/sbin/modprobe". This binary is executed whe 493 "/sbin/modprobe". This binary is executed when the kernel requests a
565 module. For example, if userspace passes an u 494 module. For example, if userspace passes an unknown filesystem type
566 to mount(), then the kernel will automatically 495 to mount(), then the kernel will automatically request the
567 corresponding filesystem module by executing t 496 corresponding filesystem module by executing this usermode helper.
568 This usermode helper should insert the needed 497 This usermode helper should insert the needed module into the kernel.
569 498
570 This sysctl only affects module autoloading. 499 This sysctl only affects module autoloading. It has no effect on the
571 ability to explicitly insert modules. 500 ability to explicitly insert modules.
572 501
573 This sysctl can be used to debug module loadin 502 This sysctl can be used to debug module loading requests::
574 503
575 echo '#! /bin/sh' > /tmp/modprobe 504 echo '#! /bin/sh' > /tmp/modprobe
576 echo 'echo "$@" >> /tmp/modprobe.log' >> / 505 echo 'echo "$@" >> /tmp/modprobe.log' >> /tmp/modprobe
577 echo 'exec /sbin/modprobe "$@"' >> /tmp/mo 506 echo 'exec /sbin/modprobe "$@"' >> /tmp/modprobe
578 chmod a+x /tmp/modprobe 507 chmod a+x /tmp/modprobe
579 echo /tmp/modprobe > /proc/sys/kernel/modp 508 echo /tmp/modprobe > /proc/sys/kernel/modprobe
580 509
581 Alternatively, if this sysctl is set to the em 510 Alternatively, if this sysctl is set to the empty string, then module
582 autoloading is completely disabled. The kerne 511 autoloading is completely disabled. The kernel will not try to
583 execute a usermode helper at all, nor will it 512 execute a usermode helper at all, nor will it call the
584 kernel_module_request LSM hook. 513 kernel_module_request LSM hook.
585 514
586 If CONFIG_STATIC_USERMODEHELPER=y is set in th 515 If CONFIG_STATIC_USERMODEHELPER=y is set in the kernel configuration,
587 then the configured static usermode helper ove 516 then the configured static usermode helper overrides this sysctl,
588 except that the empty string is still accepted 517 except that the empty string is still accepted to completely disable
589 module autoloading as described above. 518 module autoloading as described above.
590 519
591 modules_disabled 520 modules_disabled
592 ================ 521 ================
593 522
594 A toggle value indicating if modules are allow 523 A toggle value indicating if modules are allowed to be loaded
595 in an otherwise modular kernel. This toggle d 524 in an otherwise modular kernel. This toggle defaults to off
596 (0), but can be set true (1). Once true, modu 525 (0), but can be set true (1). Once true, modules can be
597 neither loaded nor unloaded, and the toggle ca 526 neither loaded nor unloaded, and the toggle cannot be set back
598 to false. Generally used with the `kexec_load 527 to false. Generally used with the `kexec_load_disabled`_ toggle.
599 528
600 529
601 .. _msgmni: 530 .. _msgmni:
602 531
603 msgmax, msgmnb, and msgmni 532 msgmax, msgmnb, and msgmni
604 ========================== 533 ==========================
605 534
606 ``msgmax`` is the maximum size of an IPC messa 535 ``msgmax`` is the maximum size of an IPC message, in bytes. 8192 by
607 default (``MSGMAX``). 536 default (``MSGMAX``).
608 537
609 ``msgmnb`` is the maximum size of an IPC queue 538 ``msgmnb`` is the maximum size of an IPC queue, in bytes. 16384 by
610 default (``MSGMNB``). 539 default (``MSGMNB``).
611 540
612 ``msgmni`` is the maximum number of IPC queues 541 ``msgmni`` is the maximum number of IPC queues. 32000 by default
613 (``MSGMNI``). 542 (``MSGMNI``).
614 543
615 All of these parameters are set per ipc namesp <<
616 in POSIX message queues is limited by ``RLIMIT <<
617 respected hierarchically in the each user name <<
618 544
619 msg_next_id, sem_next_id, and shm_next_id (Sys 545 msg_next_id, sem_next_id, and shm_next_id (System V IPC)
620 ============================================== 546 ========================================================
621 547
622 These three toggles allows to specify desired 548 These three toggles allows to specify desired id for next allocated IPC
623 object: message, semaphore or shared memory re 549 object: message, semaphore or shared memory respectively.
624 550
625 By default they are equal to -1, which means g 551 By default they are equal to -1, which means generic allocation logic.
626 Possible values to set are in range {0:``INT_M 552 Possible values to set are in range {0:``INT_MAX``}.
627 553
628 Notes: 554 Notes:
629 1) kernel doesn't guarantee, that new object 555 1) kernel doesn't guarantee, that new object will have desired id. So,
630 it's up to userspace, how to handle an ob 556 it's up to userspace, how to handle an object with "wrong" id.
631 2) Toggle with non-default value will be set 557 2) Toggle with non-default value will be set back to -1 by kernel after
632 successful IPC object allocation. If an I 558 successful IPC object allocation. If an IPC object allocation syscall
633 fails, it is undefined if the value remai 559 fails, it is undefined if the value remains unmodified or is reset to -1.
634 560
635 561
636 ngroups_max 562 ngroups_max
637 =========== 563 ===========
638 564
639 Maximum number of supplementary groups, _i.e._ 565 Maximum number of supplementary groups, _i.e._ the maximum size which
640 ``setgroups`` will accept. Exports ``NGROUPS_M 566 ``setgroups`` will accept. Exports ``NGROUPS_MAX`` from the kernel.
641 567
642 568
643 569
644 nmi_watchdog 570 nmi_watchdog
645 ============ 571 ============
646 572
647 This parameter can be used to control the NMI 573 This parameter can be used to control the NMI watchdog
648 (i.e. the hard lockup detector) on x86 systems 574 (i.e. the hard lockup detector) on x86 systems.
649 575
650 = ================================= 576 = =================================
651 0 Disable the hard lockup detector. 577 0 Disable the hard lockup detector.
652 1 Enable the hard lockup detector. 578 1 Enable the hard lockup detector.
653 = ================================= 579 = =================================
654 580
655 The hard lockup detector monitors each CPU for 581 The hard lockup detector monitors each CPU for its ability to respond to
656 timer interrupts. The mechanism utilizes CPU p 582 timer interrupts. The mechanism utilizes CPU performance counter registers
657 that are programmed to generate Non-Maskable I 583 that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
658 while a CPU is busy. Hence, the alternative na 584 while a CPU is busy. Hence, the alternative name 'NMI watchdog'.
659 585
660 The NMI watchdog is disabled by default if the 586 The NMI watchdog is disabled by default if the kernel is running as a guest
661 in a KVM virtual machine. This default can be 587 in a KVM virtual machine. This default can be overridden by adding::
662 588
663 nmi_watchdog=1 589 nmi_watchdog=1
664 590
665 to the guest kernel command line (see 591 to the guest kernel command line (see
666 Documentation/admin-guide/kernel-parameters.rs 592 Documentation/admin-guide/kernel-parameters.rst).
667 593
668 594
669 nmi_wd_lpm_factor (PPC only) <<
670 ============================ <<
671 <<
672 Factor to apply to the NMI watchdog timeout (o <<
673 set to 1). This factor represents the percenta <<
674 ``watchdog_thresh`` when calculating the NMI w <<
675 LPM. The soft lockup timeout is not impacted. <<
676 <<
677 A value of 0 means no change. The default valu <<
678 watchdog is set to 30s (based on ``watchdog_th <<
679 <<
680 <<
681 numa_balancing 595 numa_balancing
682 ============== 596 ==============
683 597
684 Enables/disables and configures automatic page !! 598 Enables/disables automatic page fault based NUMA memory
685 balancing. Memory is moved automatically to n !! 599 balancing. Memory is moved automatically to nodes
686 The value to set can be the result of ORing th !! 600 that access it often.
687 !! 601
688 = ================================= !! 602 Enables/disables automatic NUMA memory balancing. On NUMA machines, there
689 0 NUMA_BALANCING_DISABLED !! 603 is a performance penalty if remote memory is accessed by a CPU. When this
690 1 NUMA_BALANCING_NORMAL !! 604 feature is enabled the kernel samples what task thread is accessing memory
691 2 NUMA_BALANCING_MEMORY_TIERING !! 605 by periodically unmapping pages and later trapping a page fault. At the
692 = ================================= !! 606 time of the page fault, it is determined if the data being accessed should
693 !! 607 be migrated to a local memory node.
694 Or NUMA_BALANCING_NORMAL to optimize page plac <<
695 NUMA nodes to reduce remote accessing. On NUM <<
696 performance penalty if remote memory is access <<
697 feature is enabled the kernel samples what tas <<
698 memory by periodically unmapping pages and lat <<
699 fault. At the time of the page fault, it is de <<
700 being accessed should be migrated to a local m <<
701 608
702 The unmapping of pages and trapping faults inc 609 The unmapping of pages and trapping faults incur additional overhead that
703 ideally is offset by improved memory locality 610 ideally is offset by improved memory locality but there is no universal
704 guarantee. If the target workload is already b 611 guarantee. If the target workload is already bound to NUMA nodes then this
705 feature should be disabled. !! 612 feature should be disabled. Otherwise, if the system overhead from the
>> 613 feature is too high then the rate the kernel samples for NUMA hinting
>> 614 faults may be controlled by the `numa_balancing_scan_period_min_ms,
>> 615 numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms,
>> 616 numa_balancing_scan_size_mb`_, and numa_balancing_settle_count sysctls.
>> 617
>> 618
>> 619 numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb
>> 620 ===============================================================================================================================
>> 621
>> 622
>> 623 Automatic NUMA balancing scans tasks address space and unmaps pages to
>> 624 detect if pages are properly placed or if the data should be migrated to a
>> 625 memory node local to where the task is running. Every "scan delay" the task
>> 626 scans the next "scan size" number of pages in its address space. When the
>> 627 end of the address space is reached the scanner restarts from the beginning.
>> 628
>> 629 In combination, the "scan delay" and "scan size" determine the scan rate.
>> 630 When "scan delay" decreases, the scan rate increases. The scan delay and
>> 631 hence the scan rate of every task is adaptive and depends on historical
>> 632 behaviour. If pages are properly placed then the scan delay increases,
>> 633 otherwise the scan delay decreases. The "scan size" is not adaptive but
>> 634 the higher the "scan size", the higher the scan rate.
>> 635
>> 636 Higher scan rates incur higher system overhead as page faults must be
>> 637 trapped and potentially data must be migrated. However, the higher the scan
>> 638 rate, the more quickly a tasks memory is migrated to a local node if the
>> 639 workload pattern changes and minimises performance impact due to remote
>> 640 memory accesses. These sysctls control the thresholds for scan delays and
>> 641 the number of pages scanned.
>> 642
>> 643 ``numa_balancing_scan_period_min_ms`` is the minimum time in milliseconds to
>> 644 scan a tasks virtual memory. It effectively controls the maximum scanning
>> 645 rate for each task.
>> 646
>> 647 ``numa_balancing_scan_delay_ms`` is the starting "scan delay" used for a task
>> 648 when it initially forks.
>> 649
>> 650 ``numa_balancing_scan_period_max_ms`` is the maximum time in milliseconds to
>> 651 scan a tasks virtual memory. It effectively controls the minimum scanning
>> 652 rate for each task.
706 653
707 Or NUMA_BALANCING_MEMORY_TIERING to optimize p !! 654 ``numa_balancing_scan_size_mb`` is how many megabytes worth of pages are
708 different types of memory (represented as diff !! 655 scanned for a given scan.
709 place the hot pages in the fast memory. This <<
710 unmapping and page fault too. <<
711 <<
712 numa_balancing_promote_rate_limit_MBps <<
713 ====================================== <<
714 <<
715 Too high promotion/demotion throughput between <<
716 may hurt application latency. This can be use <<
717 promotion throughput. The per-node max promot <<
718 will be limited to be no more than the set val <<
719 656
720 A rule of thumb is to set this to less than 1/ <<
721 write bandwidth. <<
722 657
723 oops_all_cpu_backtrace 658 oops_all_cpu_backtrace
724 ====================== 659 ======================
725 660
726 If this option is set, the kernel will send an 661 If this option is set, the kernel will send an NMI to all CPUs to dump
727 their backtraces when an oops event occurs. It 662 their backtraces when an oops event occurs. It should be used as a last
728 resort in case a panic cannot be triggered (to 663 resort in case a panic cannot be triggered (to protect VMs running, for
729 example) or kdump can't be collected. This fil 664 example) or kdump can't be collected. This file shows up if CONFIG_SMP
730 is enabled. 665 is enabled.
731 666
732 0: Won't show all CPUs backtraces when an oops 667 0: Won't show all CPUs backtraces when an oops is detected.
733 This is the default behavior. 668 This is the default behavior.
734 669
735 1: Will non-maskably interrupt all CPUs and du 670 1: Will non-maskably interrupt all CPUs and dump their backtraces when
736 an oops event is detected. 671 an oops event is detected.
737 672
738 673
739 oops_limit <<
740 ========== <<
741 <<
742 Number of kernel oopses after which the kernel <<
743 ``panic_on_oops`` is not set. Setting this to <<
744 the count. Setting this to 1 has the same eff <<
745 ``panic_on_oops=1``. The default value is 1000 <<
746 <<
747 <<
748 osrelease, ostype & version 674 osrelease, ostype & version
749 =========================== 675 ===========================
750 676
751 :: 677 ::
752 678
753 # cat osrelease 679 # cat osrelease
754 2.1.88 680 2.1.88
755 # cat ostype 681 # cat ostype
756 Linux 682 Linux
757 # cat version 683 # cat version
758 #5 Wed Feb 25 21:49:24 MET 1998 684 #5 Wed Feb 25 21:49:24 MET 1998
759 685
760 The files ``osrelease`` and ``ostype`` should 686 The files ``osrelease`` and ``ostype`` should be clear enough.
761 ``version`` 687 ``version``
762 needs a little more clarification however. The 688 needs a little more clarification however. The '#5' means that
763 this is the fifth kernel built from this sourc 689 this is the fifth kernel built from this source base and the
764 date behind it indicates the time the kernel w 690 date behind it indicates the time the kernel was built.
765 The only way to tune these values is to rebuil 691 The only way to tune these values is to rebuild the kernel :-)
766 692
767 693
768 overflowgid & overflowuid 694 overflowgid & overflowuid
769 ========================= 695 =========================
770 696
771 if your architecture did not always support 32 697 if your architecture did not always support 32-bit UIDs (i.e. arm,
772 i386, m68k, sh, and sparc32), a fixed UID and 698 i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to
773 applications that use the old 16-bit UID/GID s 699 applications that use the old 16-bit UID/GID system calls, if the
774 actual UID or GID would exceed 65535. 700 actual UID or GID would exceed 65535.
775 701
776 These sysctls allow you to change the value of 702 These sysctls allow you to change the value of the fixed UID and GID.
777 The default is 65534. 703 The default is 65534.
778 704
779 705
780 panic 706 panic
781 ===== 707 =====
782 708
783 The value in this file determines the behaviou 709 The value in this file determines the behaviour of the kernel on a
784 panic: 710 panic:
785 711
786 * if zero, the kernel will loop forever; 712 * if zero, the kernel will loop forever;
787 * if negative, the kernel will reboot immediat 713 * if negative, the kernel will reboot immediately;
788 * if positive, the kernel will reboot after th 714 * if positive, the kernel will reboot after the corresponding number
789 of seconds. 715 of seconds.
790 716
791 When you use the software watchdog, the recomm 717 When you use the software watchdog, the recommended setting is 60.
792 718
793 719
794 panic_on_io_nmi 720 panic_on_io_nmi
795 =============== 721 ===============
796 722
797 Controls the kernel's behavior when a CPU rece 723 Controls the kernel's behavior when a CPU receives an NMI caused by
798 an IO error. 724 an IO error.
799 725
800 = ============================================ 726 = ==================================================================
801 0 Try to continue operation (default). 727 0 Try to continue operation (default).
802 1 Panic immediately. The IO error triggered an 728 1 Panic immediately. The IO error triggered an NMI. This indicates a
803 serious system condition which could result 729 serious system condition which could result in IO data corruption.
804 Rather than continuing, panicking might be a 730 Rather than continuing, panicking might be a better choice. Some
805 servers issue this sort of NMI when the dump 731 servers issue this sort of NMI when the dump button is pushed,
806 and you can use this option to take a crash 732 and you can use this option to take a crash dump.
807 = ============================================ 733 = ==================================================================
808 734
809 735
810 panic_on_oops 736 panic_on_oops
811 ============= 737 =============
812 738
813 Controls the kernel's behaviour when an oops o 739 Controls the kernel's behaviour when an oops or BUG is encountered.
814 740
815 = ============================================ 741 = ===================================================================
816 0 Try to continue operation. 742 0 Try to continue operation.
817 1 Panic immediately. If the `panic` sysctl is 743 1 Panic immediately. If the `panic` sysctl is also non-zero then the
818 machine will be rebooted. 744 machine will be rebooted.
819 = ============================================ 745 = ===================================================================
820 746
821 747
822 panic_on_stackoverflow 748 panic_on_stackoverflow
823 ====================== 749 ======================
824 750
825 Controls the kernel's behavior when detecting 751 Controls the kernel's behavior when detecting the overflows of
826 kernel, IRQ and exception stacks except a user 752 kernel, IRQ and exception stacks except a user stack.
827 This file shows up if ``CONFIG_DEBUG_STACKOVER 753 This file shows up if ``CONFIG_DEBUG_STACKOVERFLOW`` is enabled.
828 754
829 = ========================== 755 = ==========================
830 0 Try to continue operation. 756 0 Try to continue operation.
831 1 Panic immediately. 757 1 Panic immediately.
832 = ========================== 758 = ==========================
833 759
834 760
835 panic_on_unrecovered_nmi 761 panic_on_unrecovered_nmi
836 ======================== 762 ========================
837 763
838 The default Linux behaviour on an NMI of eithe 764 The default Linux behaviour on an NMI of either memory or unknown is
839 to continue operation. For many environments s 765 to continue operation. For many environments such as scientific
840 computing it is preferable that the box is tak 766 computing it is preferable that the box is taken out and the error
841 dealt with than an uncorrected parity/ECC erro 767 dealt with than an uncorrected parity/ECC error get propagated.
842 768
843 A small number of systems do generate NMIs for 769 A small number of systems do generate NMIs for bizarre random reasons
844 such as power management so the default is off 770 such as power management so the default is off. That sysctl works like
845 the existing panic controls already in that di 771 the existing panic controls already in that directory.
846 772
847 773
848 panic_on_warn 774 panic_on_warn
849 ============= 775 =============
850 776
851 Calls panic() in the WARN() path when set to 1 777 Calls panic() in the WARN() path when set to 1. This is useful to avoid
852 a kernel rebuild when attempting to kdump at t 778 a kernel rebuild when attempting to kdump at the location of a WARN().
853 779
854 = ============================================ 780 = ================================================
855 0 Only WARN(), default behaviour. 781 0 Only WARN(), default behaviour.
856 1 Call panic() after printing out WARN() locat 782 1 Call panic() after printing out WARN() location.
857 = ============================================ 783 = ================================================
858 784
859 785
860 panic_print 786 panic_print
861 =========== 787 ===========
862 788
863 Bitmask for printing system info when panic ha 789 Bitmask for printing system info when panic happens. User can chose
864 combination of the following bits: 790 combination of the following bits:
865 791
866 ===== ======================================= 792 ===== ============================================
867 bit 0 print all tasks info 793 bit 0 print all tasks info
868 bit 1 print system memory info 794 bit 1 print system memory info
869 bit 2 print timer info 795 bit 2 print timer info
870 bit 3 print locks info if ``CONFIG_LOCKDEP`` 796 bit 3 print locks info if ``CONFIG_LOCKDEP`` is on
871 bit 4 print ftrace buffer 797 bit 4 print ftrace buffer
872 bit 5 print all printk messages in buffer 798 bit 5 print all printk messages in buffer
873 bit 6 print all CPUs backtrace (if available <<
874 bit 7 print only tasks in uninterruptible (bl <<
875 ===== ======================================= 799 ===== ============================================
876 800
877 So for example to print tasks and memory info 801 So for example to print tasks and memory info on panic, user can::
878 802
879 echo 3 > /proc/sys/kernel/panic_print 803 echo 3 > /proc/sys/kernel/panic_print
880 804
881 805
882 panic_on_rcu_stall 806 panic_on_rcu_stall
883 ================== 807 ==================
884 808
885 When set to 1, calls panic() after RCU stall d 809 When set to 1, calls panic() after RCU stall detection messages. This
886 is useful to define the root cause of RCU stal 810 is useful to define the root cause of RCU stalls using a vmcore.
887 811
888 = ============================================ 812 = ============================================================
889 0 Do not panic() when RCU stall takes place, d 813 0 Do not panic() when RCU stall takes place, default behavior.
890 1 panic() after printing RCU stall messages. 814 1 panic() after printing RCU stall messages.
891 = ============================================ 815 = ============================================================
892 816
893 max_rcu_stall_to_panic <<
894 ====================== <<
895 <<
896 When ``panic_on_rcu_stall`` is set to 1, this <<
897 number of times that RCU can stall before pani <<
898 <<
899 When ``panic_on_rcu_stall`` is set to 0, this <<
900 817
901 perf_cpu_time_max_percent 818 perf_cpu_time_max_percent
902 ========================= 819 =========================
903 820
904 Hints to the kernel how much CPU time it shoul 821 Hints to the kernel how much CPU time it should be allowed to
905 use to handle perf sampling events. If the pe 822 use to handle perf sampling events. If the perf subsystem
906 is informed that its samples are exceeding thi 823 is informed that its samples are exceeding this limit, it
907 will drop its sampling frequency to attempt to 824 will drop its sampling frequency to attempt to reduce its CPU
908 usage. 825 usage.
909 826
910 Some perf sampling happens in NMIs. If these 827 Some perf sampling happens in NMIs. If these samples
911 unexpectedly take too long to execute, the NMI 828 unexpectedly take too long to execute, the NMIs can become
912 stacked up next to each other so much that not 829 stacked up next to each other so much that nothing else is
913 allowed to execute. 830 allowed to execute.
914 831
915 ===== ======================================== 832 ===== ========================================================
916 0 Disable the mechanism. Do not monitor o 833 0 Disable the mechanism. Do not monitor or correct perf's
917 sampling rate no matter how CPU time it 834 sampling rate no matter how CPU time it takes.
918 835
919 1-100 Attempt to throttle perf's sample rate t 836 1-100 Attempt to throttle perf's sample rate to this
920 percentage of CPU. Note: the kernel cal 837 percentage of CPU. Note: the kernel calculates an
921 "expected" length of each sample event. 838 "expected" length of each sample event. 100 here means
922 100% of that expected length. Even if t 839 100% of that expected length. Even if this is set to
923 100, you may still see sample throttling 840 100, you may still see sample throttling if this
924 length is exceeded. Set to 0 if you tru 841 length is exceeded. Set to 0 if you truly do not care
925 how much CPU is consumed. 842 how much CPU is consumed.
926 ===== ======================================== 843 ===== ========================================================
927 844
928 845
929 perf_event_paranoid 846 perf_event_paranoid
930 =================== 847 ===================
931 848
932 Controls use of the performance events system 849 Controls use of the performance events system by unprivileged
933 users (without CAP_PERFMON). The default valu 850 users (without CAP_PERFMON). The default value is 2.
934 851
935 For backward compatibility reasons access to s 852 For backward compatibility reasons access to system performance
936 monitoring and observability remains open for 853 monitoring and observability remains open for CAP_SYS_ADMIN
937 privileged processes but CAP_SYS_ADMIN usage f 854 privileged processes but CAP_SYS_ADMIN usage for secure system
938 performance monitoring and observability opera 855 performance monitoring and observability operations is discouraged
939 with respect to CAP_PERFMON use cases. 856 with respect to CAP_PERFMON use cases.
940 857
941 === ========================================= 858 === ==================================================================
942 -1 Allow use of (almost) all events by all u 859 -1 Allow use of (almost) all events by all users.
943 860
944 Ignore mlock limit after perf_event_mlock 861 Ignore mlock limit after perf_event_mlock_kb without
945 ``CAP_IPC_LOCK``. 862 ``CAP_IPC_LOCK``.
946 863
947 >=0 Disallow ftrace function tracepoint by us 864 >=0 Disallow ftrace function tracepoint by users without
948 ``CAP_PERFMON``. 865 ``CAP_PERFMON``.
949 866
950 Disallow raw tracepoint access by users w 867 Disallow raw tracepoint access by users without ``CAP_PERFMON``.
951 868
952 >=1 Disallow CPU event access by users withou 869 >=1 Disallow CPU event access by users without ``CAP_PERFMON``.
953 870
954 >=2 Disallow kernel profiling by users withou 871 >=2 Disallow kernel profiling by users without ``CAP_PERFMON``.
955 === ========================================= 872 === ==================================================================
956 873
957 874
958 perf_event_max_stack 875 perf_event_max_stack
959 ==================== 876 ====================
960 877
961 Controls maximum number of stack frames to cop 878 Controls maximum number of stack frames to copy for (``attr.sample_type &
962 PERF_SAMPLE_CALLCHAIN``) configured events, fo 879 PERF_SAMPLE_CALLCHAIN``) configured events, for instance, when using
963 '``perf record -g``' or '``perf trace --call-g 880 '``perf record -g``' or '``perf trace --call-graph fp``'.
964 881
965 This can only be done when no events are in us 882 This can only be done when no events are in use that have callchains
966 enabled, otherwise writing to this file will r 883 enabled, otherwise writing to this file will return ``-EBUSY``.
967 884
968 The default value is 127. 885 The default value is 127.
969 886
970 887
971 perf_event_mlock_kb 888 perf_event_mlock_kb
972 =================== 889 ===================
973 890
974 Control size of per-cpu ring buffer not counte 891 Control size of per-cpu ring buffer not counted against mlock limit.
975 892
976 The default value is 512 + 1 page 893 The default value is 512 + 1 page
977 894
978 895
979 perf_event_max_contexts_per_stack 896 perf_event_max_contexts_per_stack
980 ================================= 897 =================================
981 898
982 Controls maximum number of stack frame context 899 Controls maximum number of stack frame context entries for
983 (``attr.sample_type & PERF_SAMPLE_CALLCHAIN``) 900 (``attr.sample_type & PERF_SAMPLE_CALLCHAIN``) configured events, for
984 instance, when using '``perf record -g``' or ' 901 instance, when using '``perf record -g``' or '``perf trace --call-graph fp``'.
985 902
986 This can only be done when no events are in us 903 This can only be done when no events are in use that have callchains
987 enabled, otherwise writing to this file will r 904 enabled, otherwise writing to this file will return ``-EBUSY``.
988 905
989 The default value is 8. 906 The default value is 8.
990 907
991 908
992 perf_user_access (arm64 and riscv only) <<
993 ======================================= <<
994 <<
995 Controls user space access for reading perf ev <<
996 <<
997 arm64 <<
998 ===== <<
999 <<
1000 The default value is 0 (access disabled). <<
1001 <<
1002 When set to 1, user space can read performanc <<
1003 directly. <<
1004 <<
1005 See Documentation/arch/arm64/perf.rst for mor <<
1006 <<
1007 riscv <<
1008 ===== <<
1009 <<
1010 When set to 0, user space access is disabled. <<
1011 <<
1012 The default value is 1, user space can read p <<
1013 registers through perf, any direct access wit <<
1014 an illegal instruction. <<
1015 <<
1016 When set to 2, which enables legacy mode (use <<
1017 and insret CSRs only). Note that this legacy <<
1018 removed once all user space applications are <<
1019 <<
1020 Note that the time CSR is always directly acc <<
1021 <<
1022 pid_max 909 pid_max
1023 ======= 910 =======
1024 911
1025 PID allocation wrap value. When the kernel's 912 PID allocation wrap value. When the kernel's next PID value
1026 reaches this value, it wraps back to a minimu 913 reaches this value, it wraps back to a minimum PID value.
1027 PIDs of value ``pid_max`` or larger are not a 914 PIDs of value ``pid_max`` or larger are not allocated.
1028 915
1029 916
1030 ns_last_pid 917 ns_last_pid
1031 =========== 918 ===========
1032 919
1033 The last pid allocated in the current (the on 920 The last pid allocated in the current (the one task using this sysctl
1034 lives in) pid namespace. When selecting a pid 921 lives in) pid namespace. When selecting a pid for a next task on fork
1035 kernel tries to allocate a number starting fr 922 kernel tries to allocate a number starting from this one.
1036 923
1037 924
1038 powersave-nap (PPC only) 925 powersave-nap (PPC only)
1039 ======================== 926 ========================
1040 927
1041 If set, Linux-PPC will use the 'nap' mode of 928 If set, Linux-PPC will use the 'nap' mode of powersaving,
1042 otherwise the 'doze' mode will be used. 929 otherwise the 'doze' mode will be used.
1043 930
1044 931
1045 ============================================= 932 ==============================================================
1046 933
1047 printk 934 printk
1048 ====== 935 ======
1049 936
1050 The four values in printk denote: ``console_l 937 The four values in printk denote: ``console_loglevel``,
1051 ``default_message_loglevel``, ``minimum_conso 938 ``default_message_loglevel``, ``minimum_console_loglevel`` and
1052 ``default_console_loglevel`` respectively. 939 ``default_console_loglevel`` respectively.
1053 940
1054 These values influence printk() behavior when 941 These values influence printk() behavior when printing or
1055 logging error messages. See '``man 2 syslog`` 942 logging error messages. See '``man 2 syslog``' for more info on
1056 the different loglevels. 943 the different loglevels.
1057 944
1058 ======================== ==================== 945 ======================== =====================================
1059 console_loglevel messages with a high 946 console_loglevel messages with a higher priority than
1060 this will be printed 947 this will be printed to the console
1061 default_message_loglevel messages without an 948 default_message_loglevel messages without an explicit priority
1062 will be printed with 949 will be printed with this priority
1063 minimum_console_loglevel minimum (highest) va 950 minimum_console_loglevel minimum (highest) value to which
1064 console_loglevel can 951 console_loglevel can be set
1065 default_console_loglevel default value for co 952 default_console_loglevel default value for console_loglevel
1066 ======================== ==================== 953 ======================== =====================================
1067 954
1068 955
1069 printk_delay 956 printk_delay
1070 ============ 957 ============
1071 958
1072 Delay each printk message in ``printk_delay`` 959 Delay each printk message in ``printk_delay`` milliseconds
1073 960
1074 Value from 0 - 10000 is allowed. 961 Value from 0 - 10000 is allowed.
1075 962
1076 963
1077 printk_ratelimit 964 printk_ratelimit
1078 ================ 965 ================
1079 966
1080 Some warning messages are rate limited. ``pri 967 Some warning messages are rate limited. ``printk_ratelimit`` specifies
1081 the minimum length of time between these mess 968 the minimum length of time between these messages (in seconds).
1082 The default value is 5 seconds. 969 The default value is 5 seconds.
1083 970
1084 A value of 0 will disable rate limiting. 971 A value of 0 will disable rate limiting.
1085 972
1086 973
1087 printk_ratelimit_burst 974 printk_ratelimit_burst
1088 ====================== 975 ======================
1089 976
1090 While long term we enforce one message per `p 977 While long term we enforce one message per `printk_ratelimit`_
1091 seconds, we do allow a burst of messages to p 978 seconds, we do allow a burst of messages to pass through.
1092 ``printk_ratelimit_burst`` specifies the numb 979 ``printk_ratelimit_burst`` specifies the number of messages we can
1093 send before ratelimiting kicks in. 980 send before ratelimiting kicks in.
1094 981
1095 The default value is 10 messages. 982 The default value is 10 messages.
1096 983
1097 984
1098 printk_devkmsg 985 printk_devkmsg
1099 ============== 986 ==============
1100 987
1101 Control the logging to ``/dev/kmsg`` from use 988 Control the logging to ``/dev/kmsg`` from userspace:
1102 989
1103 ========= =================================== 990 ========= =============================================
1104 ratelimit default, ratelimited 991 ratelimit default, ratelimited
1105 on unlimited logging to /dev/kmsg from 992 on unlimited logging to /dev/kmsg from userspace
1106 off logging to /dev/kmsg disabled 993 off logging to /dev/kmsg disabled
1107 ========= =================================== 994 ========= =============================================
1108 995
1109 The kernel command line parameter ``printk.de 996 The kernel command line parameter ``printk.devkmsg=`` overrides this and is
1110 a one-time setting until next reboot: once se 997 a one-time setting until next reboot: once set, it cannot be changed by
1111 this sysctl interface anymore. 998 this sysctl interface anymore.
1112 999
1113 ============================================= 1000 ==============================================================
1114 1001
1115 1002
1116 pty 1003 pty
1117 === 1004 ===
1118 1005
1119 See Documentation/filesystems/devpts.rst. 1006 See Documentation/filesystems/devpts.rst.
1120 1007
1121 1008
1122 random 1009 random
1123 ====== 1010 ======
1124 1011
1125 This is a directory, with the following entri 1012 This is a directory, with the following entries:
1126 1013
1127 * ``boot_id``: a UUID generated the first tim 1014 * ``boot_id``: a UUID generated the first time this is retrieved, and
1128 unvarying after that; 1015 unvarying after that;
1129 1016
1130 * ``uuid``: a UUID generated every time this <<
1131 thus be used to generate UUIDs at will); <<
1132 <<
1133 * ``entropy_avail``: the pool's entropy count 1017 * ``entropy_avail``: the pool's entropy count, in bits;
1134 1018
1135 * ``poolsize``: the entropy pool size, in bit 1019 * ``poolsize``: the entropy pool size, in bits;
1136 1020
1137 * ``urandom_min_reseed_secs``: obsolete (used 1021 * ``urandom_min_reseed_secs``: obsolete (used to determine the minimum
1138 number of seconds between urandom pool rese !! 1022 number of seconds between urandom pool reseeding).
1139 writable for compatibility purposes, but wr !! 1023
1140 on any RNG behavior; !! 1024 * ``uuid``: a UUID generated every time this is retrieved (this can
>> 1025 thus be used to generate UUIDs at will);
1141 1026
1142 * ``write_wakeup_threshold``: when the entrop 1027 * ``write_wakeup_threshold``: when the entropy count drops below this
1143 (as a number of bits), processes waiting to 1028 (as a number of bits), processes waiting to write to ``/dev/random``
1144 are woken up. This file is writable for com !! 1029 are woken up.
1145 writing to it has no effect on any RNG beha !! 1030
>> 1031 If ``drivers/char/random.c`` is built with ``ADD_INTERRUPT_BENCH``
>> 1032 defined, these additional entries are present:
>> 1033
>> 1034 * ``add_interrupt_avg_cycles``: the average number of cycles between
>> 1035 interrupts used to feed the pool;
>> 1036
>> 1037 * ``add_interrupt_avg_deviation``: the standard deviation seen on the
>> 1038 number of cycles between interrupts used to feed the pool.
1146 1039
1147 1040
1148 randomize_va_space 1041 randomize_va_space
1149 ================== 1042 ==================
1150 1043
1151 This option can be used to select the type of 1044 This option can be used to select the type of process address
1152 space randomization that is used in the syste 1045 space randomization that is used in the system, for architectures
1153 that support this feature. 1046 that support this feature.
1154 1047
1155 == ========================================= 1048 == ===========================================================================
1156 0 Turn the process address space randomizat 1049 0 Turn the process address space randomization off. This is the
1157 default for architectures that do not sup 1050 default for architectures that do not support this feature anyways,
1158 and kernels that are booted with the "nor 1051 and kernels that are booted with the "norandmaps" parameter.
1159 1052
1160 1 Make the addresses of mmap base, stack an 1053 1 Make the addresses of mmap base, stack and VDSO page randomized.
1161 This, among other things, implies that sh 1054 This, among other things, implies that shared libraries will be
1162 loaded to random addresses. Also for PIE 1055 loaded to random addresses. Also for PIE-linked binaries, the
1163 location of code start is randomized. Th 1056 location of code start is randomized. This is the default if the
1164 ``CONFIG_COMPAT_BRK`` option is enabled. 1057 ``CONFIG_COMPAT_BRK`` option is enabled.
1165 1058
1166 2 Additionally enable heap randomization. 1059 2 Additionally enable heap randomization. This is the default if
1167 ``CONFIG_COMPAT_BRK`` is disabled. 1060 ``CONFIG_COMPAT_BRK`` is disabled.
1168 1061
1169 There are a few legacy applications out t 1062 There are a few legacy applications out there (such as some ancient
1170 versions of libc.so.5 from 1996) that ass 1063 versions of libc.so.5 from 1996) that assume that brk area starts
1171 just after the end of the code+bss. Thes 1064 just after the end of the code+bss. These applications break when
1172 start of the brk area is randomized. The 1065 start of the brk area is randomized. There are however no known
1173 non-legacy applications that would be bro 1066 non-legacy applications that would be broken this way, so for most
1174 systems it is safe to choose full randomi 1067 systems it is safe to choose full randomization.
1175 1068
1176 Systems with ancient and/or broken binari 1069 Systems with ancient and/or broken binaries should be configured
1177 with ``CONFIG_COMPAT_BRK`` enabled, which 1070 with ``CONFIG_COMPAT_BRK`` enabled, which excludes the heap from process
1178 address space randomization. 1071 address space randomization.
1179 == ========================================= 1072 == ===========================================================================
1180 1073
1181 1074
1182 real-root-dev 1075 real-root-dev
1183 ============= 1076 =============
1184 1077
1185 See Documentation/admin-guide/initrd.rst. 1078 See Documentation/admin-guide/initrd.rst.
1186 1079
1187 1080
1188 reboot-cmd (SPARC only) 1081 reboot-cmd (SPARC only)
1189 ======================= 1082 =======================
1190 1083
1191 ??? This seems to be a way to give an argumen 1084 ??? This seems to be a way to give an argument to the Sparc
1192 ROM/Flash boot loader. Maybe to tell it what 1085 ROM/Flash boot loader. Maybe to tell it what to do after
1193 rebooting. ??? 1086 rebooting. ???
1194 1087
1195 1088
1196 sched_energy_aware 1089 sched_energy_aware
1197 ================== 1090 ==================
1198 1091
1199 Enables/disables Energy Aware Scheduling (EAS 1092 Enables/disables Energy Aware Scheduling (EAS). EAS starts
1200 automatically on platforms where it can run ( 1093 automatically on platforms where it can run (that is,
1201 platforms with asymmetric CPU topologies and 1094 platforms with asymmetric CPU topologies and having an Energy
1202 Model available). If your platform happens to 1095 Model available). If your platform happens to meet the
1203 requirements for EAS but you do not want to u 1096 requirements for EAS but you do not want to use it, change
1204 this value to 0. On Non-EAS platforms, write !! 1097 this value to 0.
1205 read doesn't return anything. <<
1206 1098
1207 task_delayacct 1099 task_delayacct
1208 =============== 1100 ===============
1209 1101
1210 Enables/disables task delay accounting (see 1102 Enables/disables task delay accounting (see
1211 Documentation/accounting/delay-accounting.rst 1103 Documentation/accounting/delay-accounting.rst. Enabling this feature incurs
1212 a small amount of overhead in the scheduler b 1104 a small amount of overhead in the scheduler but is useful for debugging
1213 and performance tuning. It is required by som 1105 and performance tuning. It is required by some tools such as iotop.
1214 1106
1215 sched_schedstats 1107 sched_schedstats
1216 ================ 1108 ================
1217 1109
1218 Enables/disables scheduler statistics. Enabli 1110 Enables/disables scheduler statistics. Enabling this feature
1219 incurs a small amount of overhead in the sche 1111 incurs a small amount of overhead in the scheduler but is
1220 useful for debugging and performance tuning. 1112 useful for debugging and performance tuning.
1221 1113
1222 sched_util_clamp_min 1114 sched_util_clamp_min
1223 ==================== 1115 ====================
1224 1116
1225 Max allowed *minimum* utilization. 1117 Max allowed *minimum* utilization.
1226 1118
1227 Default value is 1024, which is the maximum p 1119 Default value is 1024, which is the maximum possible value.
1228 1120
1229 It means that any requested uclamp.min value 1121 It means that any requested uclamp.min value cannot be greater than
1230 sched_util_clamp_min, i.e., it is restricted 1122 sched_util_clamp_min, i.e., it is restricted to the range
1231 [0:sched_util_clamp_min]. 1123 [0:sched_util_clamp_min].
1232 1124
1233 sched_util_clamp_max 1125 sched_util_clamp_max
1234 ==================== 1126 ====================
1235 1127
1236 Max allowed *maximum* utilization. 1128 Max allowed *maximum* utilization.
1237 1129
1238 Default value is 1024, which is the maximum p 1130 Default value is 1024, which is the maximum possible value.
1239 1131
1240 It means that any requested uclamp.max value 1132 It means that any requested uclamp.max value cannot be greater than
1241 sched_util_clamp_max, i.e., it is restricted 1133 sched_util_clamp_max, i.e., it is restricted to the range
1242 [0:sched_util_clamp_max]. 1134 [0:sched_util_clamp_max].
1243 1135
1244 sched_util_clamp_min_rt_default 1136 sched_util_clamp_min_rt_default
1245 =============================== 1137 ===============================
1246 1138
1247 By default Linux is tuned for performance. Wh 1139 By default Linux is tuned for performance. Which means that RT tasks always run
1248 at the highest frequency and most capable (hi 1140 at the highest frequency and most capable (highest capacity) CPU (in
1249 heterogeneous systems). 1141 heterogeneous systems).
1250 1142
1251 Uclamp achieves this by setting the requested 1143 Uclamp achieves this by setting the requested uclamp.min of all RT tasks to
1252 1024 by default, which effectively boosts the 1144 1024 by default, which effectively boosts the tasks to run at the highest
1253 frequency and biases them to run on the bigge 1145 frequency and biases them to run on the biggest CPU.
1254 1146
1255 This knob allows admins to change the default 1147 This knob allows admins to change the default behavior when uclamp is being
1256 used. In battery powered devices particularly 1148 used. In battery powered devices particularly, running at the maximum
1257 capacity and frequency will increase energy c 1149 capacity and frequency will increase energy consumption and shorten the battery
1258 life. 1150 life.
1259 1151
1260 This knob is only effective for RT tasks whic 1152 This knob is only effective for RT tasks which the user hasn't modified their
1261 requested uclamp.min value via sched_setattr( 1153 requested uclamp.min value via sched_setattr() syscall.
1262 1154
1263 This knob will not escape the range constrain 1155 This knob will not escape the range constraint imposed by sched_util_clamp_min
1264 defined above. 1156 defined above.
1265 1157
1266 For example if 1158 For example if
1267 1159
1268 sched_util_clamp_min_rt_default = 800 1160 sched_util_clamp_min_rt_default = 800
1269 sched_util_clamp_min = 600 1161 sched_util_clamp_min = 600
1270 1162
1271 Then the boost will be clamped to 600 because 1163 Then the boost will be clamped to 600 because 800 is outside of the permissible
1272 range of [0:600]. This could happen for insta 1164 range of [0:600]. This could happen for instance if a powersave mode will
1273 restrict all boosts temporarily by modifying 1165 restrict all boosts temporarily by modifying sched_util_clamp_min. As soon as
1274 this restriction is lifted, the requested sch 1166 this restriction is lifted, the requested sched_util_clamp_min_rt_default
1275 will take effect. 1167 will take effect.
1276 1168
1277 seccomp 1169 seccomp
1278 ======= 1170 =======
1279 1171
1280 See Documentation/userspace-api/seccomp_filte 1172 See Documentation/userspace-api/seccomp_filter.rst.
1281 1173
1282 1174
1283 sg-big-buff 1175 sg-big-buff
1284 =========== 1176 ===========
1285 1177
1286 This file shows the size of the generic SCSI 1178 This file shows the size of the generic SCSI (sg) buffer.
1287 You can't tune it just yet, but you could cha 1179 You can't tune it just yet, but you could change it on
1288 compile time by editing ``include/scsi/sg.h`` 1180 compile time by editing ``include/scsi/sg.h`` and changing
1289 the value of ``SG_BIG_BUFF``. 1181 the value of ``SG_BIG_BUFF``.
1290 1182
1291 There shouldn't be any reason to change this 1183 There shouldn't be any reason to change this value. If
1292 you can come up with one, you probably know w 1184 you can come up with one, you probably know what you
1293 are doing anyway :) 1185 are doing anyway :)
1294 1186
1295 1187
1296 shmall 1188 shmall
1297 ====== 1189 ======
1298 1190
1299 This parameter sets the total amount of share !! 1191 This parameter sets the total amount of shared memory pages that
1300 inside ipc namespace. The shared memory pages !! 1192 can be used system wide. Hence, ``shmall`` should always be at least
1301 namespace separately and is not inherited. He !! 1193 ``ceil(shmmax/PAGE_SIZE)``.
1302 least ``ceil(shmmax/PAGE_SIZE)``. <<
1303 1194
1304 If you are not sure what the default ``PAGE_S 1195 If you are not sure what the default ``PAGE_SIZE`` is on your Linux
1305 system, you can run the following command:: 1196 system, you can run the following command::
1306 1197
1307 # getconf PAGE_SIZE 1198 # getconf PAGE_SIZE
1308 1199
1309 To reduce or disable the ability to allocate <<
1310 new ipc namespace, set this parameter to the <<
1311 creation of a new ipc namespace in the curren <<
1312 be used. <<
1313 1200
1314 shmmax 1201 shmmax
1315 ====== 1202 ======
1316 1203
1317 This value can be used to query and set the r 1204 This value can be used to query and set the run time limit
1318 on the maximum shared memory segment size tha 1205 on the maximum shared memory segment size that can be created.
1319 Shared memory segments up to 1Gb are now supp 1206 Shared memory segments up to 1Gb are now supported in the
1320 kernel. This value defaults to ``SHMMAX``. 1207 kernel. This value defaults to ``SHMMAX``.
1321 1208
1322 1209
1323 shmmni 1210 shmmni
1324 ====== 1211 ======
1325 1212
1326 This value determines the maximum number of s 1213 This value determines the maximum number of shared memory segments.
1327 4096 by default (``SHMMNI``). 1214 4096 by default (``SHMMNI``).
1328 1215
1329 1216
1330 shm_rmid_forced 1217 shm_rmid_forced
1331 =============== 1218 ===============
1332 1219
1333 Linux lets you set resource limits, including 1220 Linux lets you set resource limits, including how much memory one
1334 process can consume, via ``setrlimit(2)``. U 1221 process can consume, via ``setrlimit(2)``. Unfortunately, shared memory
1335 segments are allowed to exist without associa 1222 segments are allowed to exist without association with any process, and
1336 thus might not be counted against any resourc 1223 thus might not be counted against any resource limits. If enabled,
1337 shared memory segments are automatically dest 1224 shared memory segments are automatically destroyed when their attach
1338 count becomes zero after a detach or a proces 1225 count becomes zero after a detach or a process termination. It will
1339 also destroy segments that were created, but 1226 also destroy segments that were created, but never attached to, on exit
1340 from the process. The only use left for ``IP 1227 from the process. The only use left for ``IPC_RMID`` is to immediately
1341 destroy an unattached segment. Of course, th 1228 destroy an unattached segment. Of course, this breaks the way things are
1342 defined, so some applications might stop work 1229 defined, so some applications might stop working. Note that this
1343 feature will do you no good unless you also c 1230 feature will do you no good unless you also configure your resource
1344 limits (in particular, ``RLIMIT_AS`` and ``RL 1231 limits (in particular, ``RLIMIT_AS`` and ``RLIMIT_NPROC``). Most systems don't
1345 need this. 1232 need this.
1346 1233
1347 Note that if you change this from 0 to 1, alr 1234 Note that if you change this from 0 to 1, already created segments
1348 without users and with a dead originative pro 1235 without users and with a dead originative process will be destroyed.
1349 1236
1350 1237
1351 sysctl_writes_strict 1238 sysctl_writes_strict
1352 ==================== 1239 ====================
1353 1240
1354 Control how file position affects the behavio 1241 Control how file position affects the behavior of updating sysctl values
1355 via the ``/proc/sys`` interface: 1242 via the ``/proc/sys`` interface:
1356 1243
1357 == ====================================== 1244 == ======================================================================
1358 -1 Legacy per-write sysctl value handling 1245 -1 Legacy per-write sysctl value handling, with no printk warnings.
1359 Each write syscall must fully contain 1246 Each write syscall must fully contain the sysctl value to be
1360 written, and multiple writes on the sa 1247 written, and multiple writes on the same sysctl file descriptor
1361 will rewrite the sysctl value, regardl 1248 will rewrite the sysctl value, regardless of file position.
1362 0 Same behavior as above, but warn about 1249 0 Same behavior as above, but warn about processes that perform writes
1363 to a sysctl file descriptor when the f 1250 to a sysctl file descriptor when the file position is not 0.
1364 1 (default) Respect file position when w 1251 1 (default) Respect file position when writing sysctl strings. Multiple
1365 writes will append to the sysctl value 1252 writes will append to the sysctl value buffer. Anything past the max
1366 length of the sysctl value buffer will 1253 length of the sysctl value buffer will be ignored. Writes to numeric
1367 sysctl entries must always be at file 1254 sysctl entries must always be at file position 0 and the value must
1368 be fully contained in the buffer sent 1255 be fully contained in the buffer sent in the write syscall.
1369 == ====================================== 1256 == ======================================================================
1370 1257
1371 1258
1372 softlockup_all_cpu_backtrace 1259 softlockup_all_cpu_backtrace
1373 ============================ 1260 ============================
1374 1261
1375 This value controls the soft lockup detector 1262 This value controls the soft lockup detector thread's behavior
1376 when a soft lockup condition is detected as t 1263 when a soft lockup condition is detected as to whether or not
1377 to gather further debug information. If enabl 1264 to gather further debug information. If enabled, each cpu will
1378 be issued an NMI and instructed to capture st 1265 be issued an NMI and instructed to capture stack trace.
1379 1266
1380 This feature is only applicable for architect 1267 This feature is only applicable for architectures which support
1381 NMI. 1268 NMI.
1382 1269
1383 = =========================================== 1270 = ============================================
1384 0 Do nothing. This is the default behavior. 1271 0 Do nothing. This is the default behavior.
1385 1 On detection capture more debug information 1272 1 On detection capture more debug information.
1386 = =========================================== 1273 = ============================================
1387 1274
1388 1275
1389 softlockup_panic 1276 softlockup_panic
1390 ================= 1277 =================
1391 1278
1392 This parameter can be used to control whether 1279 This parameter can be used to control whether the kernel panics
1393 when a soft lockup is detected. 1280 when a soft lockup is detected.
1394 1281
1395 = =========================================== 1282 = ============================================
1396 0 Don't panic on soft lockup. 1283 0 Don't panic on soft lockup.
1397 1 Panic on soft lockup. 1284 1 Panic on soft lockup.
1398 = =========================================== 1285 = ============================================
1399 1286
1400 This can also be set using the softlockup_pan 1287 This can also be set using the softlockup_panic kernel parameter.
1401 1288
1402 1289
1403 soft_watchdog 1290 soft_watchdog
1404 ============= 1291 =============
1405 1292
1406 This parameter can be used to control the sof 1293 This parameter can be used to control the soft lockup detector.
1407 1294
1408 = ================================= 1295 = =================================
1409 0 Disable the soft lockup detector. 1296 0 Disable the soft lockup detector.
1410 1 Enable the soft lockup detector. 1297 1 Enable the soft lockup detector.
1411 = ================================= 1298 = =================================
1412 1299
1413 The soft lockup detector monitors CPUs for th 1300 The soft lockup detector monitors CPUs for threads that are hogging the CPUs
1414 without rescheduling voluntarily, and thus pr 1301 without rescheduling voluntarily, and thus prevent the 'migration/N' threads
1415 from running, causing the watchdog work fail 1302 from running, causing the watchdog work fail to execute. The mechanism depends
1416 on the CPUs ability to respond to timer inter 1303 on the CPUs ability to respond to timer interrupts which are needed for the
1417 watchdog work to be queued by the watchdog ti 1304 watchdog work to be queued by the watchdog timer function, otherwise the NMI
1418 watchdog — if enabled — can detect a hard 1305 watchdog — if enabled — can detect a hard lockup condition.
1419 1306
1420 1307
1421 split_lock_mitigate (x86 only) <<
1422 ============================== <<
1423 <<
1424 On x86, each "split lock" imposes a system-wi <<
1425 systems, large numbers of split locks from un <<
1426 denials of service to well-behaved and potent <<
1427 <<
1428 The kernel mitigates these bad users by detec <<
1429 penalties: forcing them to wait and only allo <<
1430 locks at a time. <<
1431 <<
1432 These mitigations can make those bad applicat <<
1433 split_lock_mitigate=0 may restore some applic <<
1434 increase system exposure to denial of service <<
1435 <<
1436 = =========================================== <<
1437 0 Disable the mitigation mode - just warns th <<
1438 and exposes the system to denials of servic <<
1439 1 Enable the mitigation mode (this is the def <<
1440 lockers with intentional performance degrad <<
1441 = =========================================== <<
1442 <<
1443 <<
1444 stack_erasing 1308 stack_erasing
1445 ============= 1309 =============
1446 1310
1447 This parameter can be used to control kernel 1311 This parameter can be used to control kernel stack erasing at the end
1448 of syscalls for kernels built with ``CONFIG_G 1312 of syscalls for kernels built with ``CONFIG_GCC_PLUGIN_STACKLEAK``.
1449 1313
1450 That erasing reduces the information which ke 1314 That erasing reduces the information which kernel stack leak bugs
1451 can reveal and blocks some uninitialized stac 1315 can reveal and blocks some uninitialized stack variable attacks.
1452 The tradeoff is the performance impact: on a 1316 The tradeoff is the performance impact: on a single CPU system kernel
1453 compilation sees a 1% slowdown, other systems 1317 compilation sees a 1% slowdown, other systems and workloads may vary.
1454 1318
1455 = =========================================== 1319 = ====================================================================
1456 0 Kernel stack erasing is disabled, STACKLEAK 1320 0 Kernel stack erasing is disabled, STACKLEAK_METRICS are not updated.
1457 1 Kernel stack erasing is enabled (default), 1321 1 Kernel stack erasing is enabled (default), it is performed before
1458 returning to the userspace at the end of sy 1322 returning to the userspace at the end of syscalls.
1459 = =========================================== 1323 = ====================================================================
1460 1324
1461 1325
1462 stop-a (SPARC only) 1326 stop-a (SPARC only)
1463 =================== 1327 ===================
1464 1328
1465 Controls Stop-A: 1329 Controls Stop-A:
1466 1330
1467 = ==================================== 1331 = ====================================
1468 0 Stop-A has no effect. 1332 0 Stop-A has no effect.
1469 1 Stop-A breaks to the PROM (default). 1333 1 Stop-A breaks to the PROM (default).
1470 = ==================================== 1334 = ====================================
1471 1335
1472 Stop-A is always enabled on a panic, so that 1336 Stop-A is always enabled on a panic, so that the user can return to
1473 the boot PROM. 1337 the boot PROM.
1474 1338
1475 1339
1476 sysrq 1340 sysrq
1477 ===== 1341 =====
1478 1342
1479 See Documentation/admin-guide/sysrq.rst. 1343 See Documentation/admin-guide/sysrq.rst.
1480 1344
1481 1345
1482 tainted 1346 tainted
1483 ======= 1347 =======
1484 1348
1485 Non-zero if the kernel has been tainted. Nume 1349 Non-zero if the kernel has been tainted. Numeric values, which can be
1486 ORed together. The letters are seen in "Taint 1350 ORed together. The letters are seen in "Tainted" line of Oops reports.
1487 1351
1488 ====== ===== ============================== 1352 ====== ===== ==============================================================
1489 1 `(P)` proprietary module was loaded 1353 1 `(P)` proprietary module was loaded
1490 2 `(F)` module was force loaded 1354 2 `(F)` module was force loaded
1491 4 `(S)` kernel running on an out of sp 1355 4 `(S)` kernel running on an out of specification system
1492 8 `(R)` module was force unloaded 1356 8 `(R)` module was force unloaded
1493 16 `(M)` processor reported a Machine C 1357 16 `(M)` processor reported a Machine Check Exception (MCE)
1494 32 `(B)` bad page referenced or some un 1358 32 `(B)` bad page referenced or some unexpected page flags
1495 64 `(U)` taint requested by userspace a 1359 64 `(U)` taint requested by userspace application
1496 128 `(D)` kernel died recently, i.e. the 1360 128 `(D)` kernel died recently, i.e. there was an OOPS or BUG
1497 256 `(A)` an ACPI table was overridden b 1361 256 `(A)` an ACPI table was overridden by user
1498 512 `(W)` kernel issued warning 1362 512 `(W)` kernel issued warning
1499 1024 `(C)` staging driver was loaded 1363 1024 `(C)` staging driver was loaded
1500 2048 `(I)` workaround for bug in platform 1364 2048 `(I)` workaround for bug in platform firmware applied
1501 4096 `(O)` externally-built ("out-of-tree 1365 4096 `(O)` externally-built ("out-of-tree") module was loaded
1502 8192 `(E)` unsigned module was loaded 1366 8192 `(E)` unsigned module was loaded
1503 16384 `(L)` soft lockup occurred 1367 16384 `(L)` soft lockup occurred
1504 32768 `(K)` kernel has been live patched 1368 32768 `(K)` kernel has been live patched
1505 65536 `(X)` Auxiliary taint, defined and u 1369 65536 `(X)` Auxiliary taint, defined and used by for distros
1506 131072 `(T)` The kernel was built with the 1370 131072 `(T)` The kernel was built with the struct randomization plugin
1507 ====== ===== ============================== 1371 ====== ===== ==============================================================
1508 1372
1509 See Documentation/admin-guide/tainted-kernels 1373 See Documentation/admin-guide/tainted-kernels.rst for more information.
1510 1374
1511 Note: 1375 Note:
1512 writes to this sysctl interface will fail w 1376 writes to this sysctl interface will fail with ``EINVAL`` if the kernel is
1513 booted with the command line option ``panic 1377 booted with the command line option ``panic_on_taint=<bitmask>,nousertaint``
1514 and any of the ORed together values being w 1378 and any of the ORed together values being written to ``tainted`` match with
1515 the bitmask declared on panic_on_taint. 1379 the bitmask declared on panic_on_taint.
1516 See Documentation/admin-guide/kernel-parame 1380 See Documentation/admin-guide/kernel-parameters.rst for more details on
1517 that particular kernel command line option 1381 that particular kernel command line option and its optional
1518 ``nousertaint`` switch. 1382 ``nousertaint`` switch.
1519 1383
1520 threads-max 1384 threads-max
1521 =========== 1385 ===========
1522 1386
1523 This value controls the maximum number of thr 1387 This value controls the maximum number of threads that can be created
1524 using ``fork()``. 1388 using ``fork()``.
1525 1389
1526 During initialization the kernel sets this va 1390 During initialization the kernel sets this value such that even if the
1527 maximum number of threads is created, the thr 1391 maximum number of threads is created, the thread structures occupy only
1528 a part (1/8th) of the available RAM pages. 1392 a part (1/8th) of the available RAM pages.
1529 1393
1530 The minimum value that can be written to ``th 1394 The minimum value that can be written to ``threads-max`` is 1.
1531 1395
1532 The maximum value that can be written to ``th 1396 The maximum value that can be written to ``threads-max`` is given by the
1533 constant ``FUTEX_TID_MASK`` (0x3fffffff). 1397 constant ``FUTEX_TID_MASK`` (0x3fffffff).
1534 1398
1535 If a value outside of this range is written t 1399 If a value outside of this range is written to ``threads-max`` an
1536 ``EINVAL`` error occurs. 1400 ``EINVAL`` error occurs.
1537 1401
1538 1402
1539 traceoff_on_warning 1403 traceoff_on_warning
1540 =================== 1404 ===================
1541 1405
1542 When set, disables tracing (see Documentation 1406 When set, disables tracing (see Documentation/trace/ftrace.rst) when a
1543 ``WARN()`` is hit. 1407 ``WARN()`` is hit.
1544 1408
1545 1409
1546 tracepoint_printk 1410 tracepoint_printk
1547 ================= 1411 =================
1548 1412
1549 When tracepoints are sent to printk() (enable 1413 When tracepoints are sent to printk() (enabled by the ``tp_printk``
1550 boot parameter), this entry provides runtime 1414 boot parameter), this entry provides runtime control::
1551 1415
1552 echo 0 > /proc/sys/kernel/tracepoint_prin 1416 echo 0 > /proc/sys/kernel/tracepoint_printk
1553 1417
1554 will stop tracepoints from being sent to prin 1418 will stop tracepoints from being sent to printk(), and::
1555 1419
1556 echo 1 > /proc/sys/kernel/tracepoint_prin 1420 echo 1 > /proc/sys/kernel/tracepoint_printk
1557 1421
1558 will send them to printk() again. 1422 will send them to printk() again.
1559 1423
1560 This only works if the kernel was booted with 1424 This only works if the kernel was booted with ``tp_printk`` enabled.
1561 1425
1562 See Documentation/admin-guide/kernel-paramete 1426 See Documentation/admin-guide/kernel-parameters.rst and
1563 Documentation/trace/boottime-trace.rst. 1427 Documentation/trace/boottime-trace.rst.
1564 1428
1565 1429
>> 1430 .. _unaligned-dump-stack:
>> 1431
>> 1432 unaligned-dump-stack (ia64)
>> 1433 ===========================
>> 1434
>> 1435 When logging unaligned accesses, controls whether the stack is
>> 1436 dumped.
>> 1437
>> 1438 = ===================================================
>> 1439 0 Do not dump the stack. This is the default setting.
>> 1440 1 Dump the stack.
>> 1441 = ===================================================
>> 1442
>> 1443 See also `ignore-unaligned-usertrap`_.
>> 1444
>> 1445
1566 unaligned-trap 1446 unaligned-trap
1567 ============== 1447 ==============
1568 1448
1569 On architectures where unaligned accesses cau 1449 On architectures where unaligned accesses cause traps, and where this
1570 feature is supported (``CONFIG_SYSCTL_ARCH_UN 1450 feature is supported (``CONFIG_SYSCTL_ARCH_UNALIGN_ALLOW``; currently,
1571 ``arc``, ``parisc`` and ``loongarch``), contr !! 1451 ``arc`` and ``parisc``), controls whether unaligned traps are caught
1572 are caught and emulated (instead of failing). !! 1452 and emulated (instead of failing).
1573 1453
1574 = =========================================== 1454 = ========================================================
1575 0 Do not emulate unaligned accesses. 1455 0 Do not emulate unaligned accesses.
1576 1 Emulate unaligned accesses. This is the def 1456 1 Emulate unaligned accesses. This is the default setting.
1577 = =========================================== 1457 = ========================================================
1578 1458
1579 See also `ignore-unaligned-usertrap`_. 1459 See also `ignore-unaligned-usertrap`_.
1580 1460
1581 1461
1582 unknown_nmi_panic 1462 unknown_nmi_panic
1583 ================= 1463 =================
1584 1464
1585 The value in this file affects behavior of ha 1465 The value in this file affects behavior of handling NMI. When the
1586 value is non-zero, unknown NMI is trapped and 1466 value is non-zero, unknown NMI is trapped and then panic occurs. At
1587 that time, kernel debugging information is di 1467 that time, kernel debugging information is displayed on console.
1588 1468
1589 NMI switch that most IA32 servers have fires 1469 NMI switch that most IA32 servers have fires unknown NMI up, for
1590 example. If a system hangs up, try pressing 1470 example. If a system hangs up, try pressing the NMI switch.
1591 1471
1592 1472
1593 unprivileged_bpf_disabled 1473 unprivileged_bpf_disabled
1594 ========================= 1474 =========================
1595 1475
1596 Writing 1 to this entry will disable unprivil 1476 Writing 1 to this entry will disable unprivileged calls to ``bpf()``;
1597 once disabled, calling ``bpf()`` without ``CA 1477 once disabled, calling ``bpf()`` without ``CAP_SYS_ADMIN`` or ``CAP_BPF``
1598 will return ``-EPERM``. Once set to 1, this c 1478 will return ``-EPERM``. Once set to 1, this can't be cleared from the
1599 running kernel anymore. 1479 running kernel anymore.
1600 1480
1601 Writing 2 to this entry will also disable unp 1481 Writing 2 to this entry will also disable unprivileged calls to ``bpf()``,
1602 however, an admin can still change this setti 1482 however, an admin can still change this setting later on, if needed, by
1603 writing 0 or 1 to this entry. 1483 writing 0 or 1 to this entry.
1604 1484
1605 If ``BPF_UNPRIV_DEFAULT_OFF`` is enabled in t 1485 If ``BPF_UNPRIV_DEFAULT_OFF`` is enabled in the kernel config, then this
1606 entry will default to 2 instead of 0. 1486 entry will default to 2 instead of 0.
1607 1487
1608 = =========================================== 1488 = =============================================================
1609 0 Unprivileged calls to ``bpf()`` are enabled 1489 0 Unprivileged calls to ``bpf()`` are enabled
1610 1 Unprivileged calls to ``bpf()`` are disable 1490 1 Unprivileged calls to ``bpf()`` are disabled without recovery
1611 2 Unprivileged calls to ``bpf()`` are disable 1491 2 Unprivileged calls to ``bpf()`` are disabled
1612 = =========================================== 1492 = =============================================================
1613 <<
1614 <<
1615 warn_limit <<
1616 ========== <<
1617 <<
1618 Number of kernel warnings after which the ker <<
1619 ``panic_on_warn`` is not set. Setting this to <<
1620 the warning count. Setting this to 1 has the <<
1621 ``panic_on_warn=1``. The default value is 0. <<
1622 <<
1623 1493
1624 watchdog 1494 watchdog
1625 ======== 1495 ========
1626 1496
1627 This parameter can be used to disable or enab 1497 This parameter can be used to disable or enable the soft lockup detector
1628 *and* the NMI watchdog (i.e. the hard lockup 1498 *and* the NMI watchdog (i.e. the hard lockup detector) at the same time.
1629 1499
1630 = ============================== 1500 = ==============================
1631 0 Disable both lockup detectors. 1501 0 Disable both lockup detectors.
1632 1 Enable both lockup detectors. 1502 1 Enable both lockup detectors.
1633 = ============================== 1503 = ==============================
1634 1504
1635 The soft lockup detector and the NMI watchdog 1505 The soft lockup detector and the NMI watchdog can also be disabled or
1636 enabled individually, using the ``soft_watchd 1506 enabled individually, using the ``soft_watchdog`` and ``nmi_watchdog``
1637 parameters. 1507 parameters.
1638 If the ``watchdog`` parameter is read, for ex 1508 If the ``watchdog`` parameter is read, for example by executing::
1639 1509
1640 cat /proc/sys/kernel/watchdog 1510 cat /proc/sys/kernel/watchdog
1641 1511
1642 the output of this command (0 or 1) shows the 1512 the output of this command (0 or 1) shows the logical OR of
1643 ``soft_watchdog`` and ``nmi_watchdog``. 1513 ``soft_watchdog`` and ``nmi_watchdog``.
1644 1514
1645 1515
1646 watchdog_cpumask 1516 watchdog_cpumask
1647 ================ 1517 ================
1648 1518
1649 This value can be used to control on which cp 1519 This value can be used to control on which cpus the watchdog may run.
1650 The default cpumask is all possible cores, bu 1520 The default cpumask is all possible cores, but if ``NO_HZ_FULL`` is
1651 enabled in the kernel config, and cores are s 1521 enabled in the kernel config, and cores are specified with the
1652 ``nohz_full=`` boot argument, those cores are 1522 ``nohz_full=`` boot argument, those cores are excluded by default.
1653 Offline cores can be included in this mask, a 1523 Offline cores can be included in this mask, and if the core is later
1654 brought online, the watchdog will be started 1524 brought online, the watchdog will be started based on the mask value.
1655 1525
1656 Typically this value would only be touched in 1526 Typically this value would only be touched in the ``nohz_full`` case
1657 to re-enable cores that by default were not r 1527 to re-enable cores that by default were not running the watchdog,
1658 if a kernel lockup was suspected on those cor 1528 if a kernel lockup was suspected on those cores.
1659 1529
1660 The argument value is the standard cpulist fo 1530 The argument value is the standard cpulist format for cpumasks,
1661 so for example to enable the watchdog on core 1531 so for example to enable the watchdog on cores 0, 2, 3, and 4 you
1662 might say:: 1532 might say::
1663 1533
1664 echo 0,2-4 > /proc/sys/kernel/watchdog_cpum 1534 echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask
1665 1535
1666 1536
1667 watchdog_thresh 1537 watchdog_thresh
1668 =============== 1538 ===============
1669 1539
1670 This value can be used to control the frequen 1540 This value can be used to control the frequency of hrtimer and NMI
1671 events and the soft and hard lockup threshold 1541 events and the soft and hard lockup thresholds. The default threshold
1672 is 10 seconds. 1542 is 10 seconds.
1673 1543
1674 The softlockup threshold is (``2 * watchdog_t 1544 The softlockup threshold is (``2 * watchdog_thresh``). Setting this
1675 tunable to zero will disable lockup detection 1545 tunable to zero will disable lockup detection altogether.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.