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