Loading...
Note: File does not exist in v6.13.7.
1Documentation for /proc/sys/kernel/* kernel version 2.2.10
2 (c) 1998, 1999, Rik van Riel <riel@nl.linux.org>
3 (c) 2009, Shen Feng<shen@cn.fujitsu.com>
4
5For general info and legal blurb, please look in README.
6
7==============================================================
8
9This file contains documentation for the sysctl files in
10/proc/sys/kernel/ and is valid for Linux kernel version 2.2.
11
12The files in this directory can be used to tune and monitor
13miscellaneous and general things in the operation of the Linux
14kernel. Since some of the files _can_ be used to screw up your
15system, it is advisable to read both documentation and source
16before actually making adjustments.
17
18Currently, these files might (depending on your configuration)
19show up in /proc/sys/kernel:
20
21- acct
22- acpi_video_flags
23- auto_msgmni
24- bootloader_type [ X86 only ]
25- bootloader_version [ X86 only ]
26- callhome [ S390 only ]
27- cap_last_cap
28- core_pattern
29- core_pipe_limit
30- core_uses_pid
31- ctrl-alt-del
32- dmesg_restrict
33- domainname
34- hostname
35- hotplug
36- hardlockup_all_cpu_backtrace
37- hardlockup_panic
38- hung_task_panic
39- hung_task_check_count
40- hung_task_timeout_secs
41- hung_task_warnings
42- kexec_load_disabled
43- kptr_restrict
44- l2cr [ PPC only ]
45- modprobe ==> Documentation/debugging-modules.txt
46- modules_disabled
47- msg_next_id [ sysv ipc ]
48- msgmax
49- msgmnb
50- msgmni
51- nmi_watchdog
52- osrelease
53- ostype
54- overflowgid
55- overflowuid
56- panic
57- panic_on_oops
58- panic_on_stackoverflow
59- panic_on_unrecovered_nmi
60- panic_on_warn
61- panic_on_rcu_stall
62- perf_cpu_time_max_percent
63- perf_event_paranoid
64- perf_event_max_stack
65- perf_event_mlock_kb
66- perf_event_max_contexts_per_stack
67- pid_max
68- powersave-nap [ PPC only ]
69- printk
70- printk_delay
71- printk_ratelimit
72- printk_ratelimit_burst
73- pty ==> Documentation/filesystems/devpts.txt
74- randomize_va_space
75- real-root-dev ==> Documentation/admin-guide/initrd.rst
76- reboot-cmd [ SPARC only ]
77- rtsig-max
78- rtsig-nr
79- seccomp/ ==> Documentation/userspace-api/seccomp_filter.rst
80- sem
81- sem_next_id [ sysv ipc ]
82- sg-big-buff [ generic SCSI device (sg) ]
83- shm_next_id [ sysv ipc ]
84- shm_rmid_forced
85- shmall
86- shmmax [ sysv ipc ]
87- shmmni
88- softlockup_all_cpu_backtrace
89- soft_watchdog
90- stop-a [ SPARC only ]
91- sysrq ==> Documentation/admin-guide/sysrq.rst
92- sysctl_writes_strict
93- tainted
94- threads-max
95- unknown_nmi_panic
96- watchdog
97- watchdog_thresh
98- version
99
100==============================================================
101
102acct:
103
104highwater lowwater frequency
105
106If BSD-style process accounting is enabled these values control
107its behaviour. If free space on filesystem where the log lives
108goes below <lowwater>% accounting suspends. If free space gets
109above <highwater>% accounting resumes. <Frequency> determines
110how often do we check the amount of free space (value is in
111seconds). Default:
1124 2 30
113That is, suspend accounting if there left <= 2% free; resume it
114if we got >=4%; consider information about amount of free space
115valid for 30 seconds.
116
117==============================================================
118
119acpi_video_flags:
120
121flags
122
123See Doc*/kernel/power/video.txt, it allows mode of video boot to be
124set during run time.
125
126==============================================================
127
128auto_msgmni:
129
130This variable has no effect and may be removed in future kernel
131releases. Reading it always returns 0.
132Up to Linux 3.17, it enabled/disabled automatic recomputing of msgmni
133upon memory add/remove or upon ipc namespace creation/removal.
134Echoing "1" into this file enabled msgmni automatic recomputing.
135Echoing "0" turned it off. auto_msgmni default value was 1.
136
137
138==============================================================
139
140bootloader_type:
141
142x86 bootloader identification
143
144This gives the bootloader type number as indicated by the bootloader,
145shifted left by 4, and OR'd with the low four bits of the bootloader
146version. The reason for this encoding is that this used to match the
147type_of_loader field in the kernel header; the encoding is kept for
148backwards compatibility. That is, if the full bootloader type number
149is 0x15 and the full version number is 0x234, this file will contain
150the value 340 = 0x154.
151
152See the type_of_loader and ext_loader_type fields in
153Documentation/x86/boot.txt for additional information.
154
155==============================================================
156
157bootloader_version:
158
159x86 bootloader version
160
161The complete bootloader version number. In the example above, this
162file will contain the value 564 = 0x234.
163
164See the type_of_loader and ext_loader_ver fields in
165Documentation/x86/boot.txt for additional information.
166
167==============================================================
168
169callhome:
170
171Controls the kernel's callhome behavior in case of a kernel panic.
172
173The s390 hardware allows an operating system to send a notification
174to a service organization (callhome) in case of an operating system panic.
175
176When the value in this file is 0 (which is the default behavior)
177nothing happens in case of a kernel panic. If this value is set to "1"
178the complete kernel oops message is send to the IBM customer service
179organization in case the mainframe the Linux operating system is running
180on has a service contract with IBM.
181
182==============================================================
183
184cap_last_cap
185
186Highest valid capability of the running kernel. Exports
187CAP_LAST_CAP from the kernel.
188
189==============================================================
190
191core_pattern:
192
193core_pattern is used to specify a core dumpfile pattern name.
194. max length 128 characters; default value is "core"
195. core_pattern is used as a pattern template for the output filename;
196 certain string patterns (beginning with '%') are substituted with
197 their actual values.
198. backward compatibility with core_uses_pid:
199 If core_pattern does not include "%p" (default does not)
200 and core_uses_pid is set, then .PID will be appended to
201 the filename.
202. corename format specifiers:
203 %<NUL> '%' is dropped
204 %% output one '%'
205 %p pid
206 %P global pid (init PID namespace)
207 %i tid
208 %I global tid (init PID namespace)
209 %u uid (in initial user namespace)
210 %g gid (in initial user namespace)
211 %d dump mode, matches PR_SET_DUMPABLE and
212 /proc/sys/fs/suid_dumpable
213 %s signal number
214 %t UNIX time of dump
215 %h hostname
216 %e executable filename (may be shortened)
217 %E executable path
218 %<OTHER> both are dropped
219. If the first character of the pattern is a '|', the kernel will treat
220 the rest of the pattern as a command to run. The core dump will be
221 written to the standard input of that program instead of to a file.
222
223==============================================================
224
225core_pipe_limit:
226
227This sysctl is only applicable when core_pattern is configured to pipe
228core files to a user space helper (when the first character of
229core_pattern is a '|', see above). When collecting cores via a pipe
230to an application, it is occasionally useful for the collecting
231application to gather data about the crashing process from its
232/proc/pid directory. In order to do this safely, the kernel must wait
233for the collecting process to exit, so as not to remove the crashing
234processes proc files prematurely. This in turn creates the
235possibility that a misbehaving userspace collecting process can block
236the reaping of a crashed process simply by never exiting. This sysctl
237defends against that. It defines how many concurrent crashing
238processes may be piped to user space applications in parallel. If
239this value is exceeded, then those crashing processes above that value
240are noted via the kernel log and their cores are skipped. 0 is a
241special value, indicating that unlimited processes may be captured in
242parallel, but that no waiting will take place (i.e. the collecting
243process is not guaranteed access to /proc/<crashing pid>/). This
244value defaults to 0.
245
246==============================================================
247
248core_uses_pid:
249
250The default coredump filename is "core". By setting
251core_uses_pid to 1, the coredump filename becomes core.PID.
252If core_pattern does not include "%p" (default does not)
253and core_uses_pid is set, then .PID will be appended to
254the filename.
255
256==============================================================
257
258ctrl-alt-del:
259
260When the value in this file is 0, ctrl-alt-del is trapped and
261sent to the init(1) program to handle a graceful restart.
262When, however, the value is > 0, Linux's reaction to a Vulcan
263Nerve Pinch (tm) will be an immediate reboot, without even
264syncing its dirty buffers.
265
266Note: when a program (like dosemu) has the keyboard in 'raw'
267mode, the ctrl-alt-del is intercepted by the program before it
268ever reaches the kernel tty layer, and it's up to the program
269to decide what to do with it.
270
271==============================================================
272
273dmesg_restrict:
274
275This toggle indicates whether unprivileged users are prevented
276from using dmesg(8) to view messages from the kernel's log buffer.
277When dmesg_restrict is set to (0) there are no restrictions. When
278dmesg_restrict is set set to (1), users must have CAP_SYSLOG to use
279dmesg(8).
280
281The kernel config option CONFIG_SECURITY_DMESG_RESTRICT sets the
282default value of dmesg_restrict.
283
284==============================================================
285
286domainname & hostname:
287
288These files can be used to set the NIS/YP domainname and the
289hostname of your box in exactly the same way as the commands
290domainname and hostname, i.e.:
291# echo "darkstar" > /proc/sys/kernel/hostname
292# echo "mydomain" > /proc/sys/kernel/domainname
293has the same effect as
294# hostname "darkstar"
295# domainname "mydomain"
296
297Note, however, that the classic darkstar.frop.org has the
298hostname "darkstar" and DNS (Internet Domain Name Server)
299domainname "frop.org", not to be confused with the NIS (Network
300Information Service) or YP (Yellow Pages) domainname. These two
301domain names are in general different. For a detailed discussion
302see the hostname(1) man page.
303
304==============================================================
305hardlockup_all_cpu_backtrace:
306
307This value controls the hard lockup detector behavior when a hard
308lockup condition is detected as to whether or not to gather further
309debug information. If enabled, arch-specific all-CPU stack dumping
310will be initiated.
311
3120: do nothing. This is the default behavior.
313
3141: on detection capture more debug information.
315==============================================================
316
317hardlockup_panic:
318
319This parameter can be used to control whether the kernel panics
320when a hard lockup is detected.
321
322 0 - don't panic on hard lockup
323 1 - panic on hard lockup
324
325See Documentation/lockup-watchdogs.txt for more information. This can
326also be set using the nmi_watchdog kernel parameter.
327
328==============================================================
329
330hotplug:
331
332Path for the hotplug policy agent.
333Default value is "/sbin/hotplug".
334
335==============================================================
336
337hung_task_panic:
338
339Controls the kernel's behavior when a hung task is detected.
340This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
341
3420: continue operation. This is the default behavior.
343
3441: panic immediately.
345
346==============================================================
347
348hung_task_check_count:
349
350The upper bound on the number of tasks that are checked.
351This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
352
353==============================================================
354
355hung_task_timeout_secs:
356
357Check interval. When a task in D state did not get scheduled
358for more than this value report a warning.
359This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
360
3610: means infinite timeout - no checking done.
362Possible values to set are in range {0..LONG_MAX/HZ}.
363
364==============================================================
365
366hung_task_warnings:
367
368The maximum number of warnings to report. During a check interval
369if a hung task is detected, this value is decreased by 1.
370When this value reaches 0, no more warnings will be reported.
371This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
372
373-1: report an infinite number of warnings.
374
375==============================================================
376
377kexec_load_disabled:
378
379A toggle indicating if the kexec_load syscall has been disabled. This
380value defaults to 0 (false: kexec_load enabled), but can be set to 1
381(true: kexec_load disabled). Once true, kexec can no longer be used, and
382the toggle cannot be set back to false. This allows a kexec image to be
383loaded before disabling the syscall, allowing a system to set up (and
384later use) an image without it being altered. Generally used together
385with the "modules_disabled" sysctl.
386
387==============================================================
388
389kptr_restrict:
390
391This toggle indicates whether restrictions are placed on
392exposing kernel addresses via /proc and other interfaces.
393
394When kptr_restrict is set to 0 (the default) the address is hashed before
395printing. (This is the equivalent to %p.)
396
397When kptr_restrict is set to (1), kernel pointers printed using the %pK
398format specifier will be replaced with 0's unless the user has CAP_SYSLOG
399and effective user and group ids are equal to the real ids. This is
400because %pK checks are done at read() time rather than open() time, so
401if permissions are elevated between the open() and the read() (e.g via
402a setuid binary) then %pK will not leak kernel pointers to unprivileged
403users. Note, this is a temporary solution only. The correct long-term
404solution is to do the permission checks at open() time. Consider removing
405world read permissions from files that use %pK, and using dmesg_restrict
406to protect against uses of %pK in dmesg(8) if leaking kernel pointer
407values to unprivileged users is a concern.
408
409When kptr_restrict is set to (2), kernel pointers printed using
410%pK will be replaced with 0's regardless of privileges.
411
412==============================================================
413
414l2cr: (PPC only)
415
416This flag controls the L2 cache of G3 processor boards. If
4170, the cache is disabled. Enabled if nonzero.
418
419==============================================================
420
421modules_disabled:
422
423A toggle value indicating if modules are allowed to be loaded
424in an otherwise modular kernel. This toggle defaults to off
425(0), but can be set true (1). Once true, modules can be
426neither loaded nor unloaded, and the toggle cannot be set back
427to false. Generally used with the "kexec_load_disabled" toggle.
428
429==============================================================
430
431msg_next_id, sem_next_id, and shm_next_id:
432
433These three toggles allows to specify desired id for next allocated IPC
434object: message, semaphore or shared memory respectively.
435
436By default they are equal to -1, which means generic allocation logic.
437Possible values to set are in range {0..INT_MAX}.
438
439Notes:
4401) kernel doesn't guarantee, that new object will have desired id. So,
441it's up to userspace, how to handle an object with "wrong" id.
4422) Toggle with non-default value will be set back to -1 by kernel after
443successful IPC object allocation.
444
445==============================================================
446
447nmi_watchdog:
448
449This parameter can be used to control the NMI watchdog
450(i.e. the hard lockup detector) on x86 systems.
451
452 0 - disable the hard lockup detector
453 1 - enable the hard lockup detector
454
455The hard lockup detector monitors each CPU for its ability to respond to
456timer interrupts. The mechanism utilizes CPU performance counter registers
457that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
458while a CPU is busy. Hence, the alternative name 'NMI watchdog'.
459
460The NMI watchdog is disabled by default if the kernel is running as a guest
461in a KVM virtual machine. This default can be overridden by adding
462
463 nmi_watchdog=1
464
465to the guest kernel command line (see Documentation/admin-guide/kernel-parameters.rst).
466
467==============================================================
468
469numa_balancing
470
471Enables/disables automatic page fault based NUMA memory
472balancing. Memory is moved automatically to nodes
473that access it often.
474
475Enables/disables automatic NUMA memory balancing. On NUMA machines, there
476is a performance penalty if remote memory is accessed by a CPU. When this
477feature is enabled the kernel samples what task thread is accessing memory
478by periodically unmapping pages and later trapping a page fault. At the
479time of the page fault, it is determined if the data being accessed should
480be migrated to a local memory node.
481
482The unmapping of pages and trapping faults incur additional overhead that
483ideally is offset by improved memory locality but there is no universal
484guarantee. If the target workload is already bound to NUMA nodes then this
485feature should be disabled. Otherwise, if the system overhead from the
486feature is too high then the rate the kernel samples for NUMA hinting
487faults may be controlled by the numa_balancing_scan_period_min_ms,
488numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms,
489numa_balancing_scan_size_mb, and numa_balancing_settle_count sysctls.
490
491==============================================================
492
493numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms,
494numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb
495
496Automatic NUMA balancing scans tasks address space and unmaps pages to
497detect if pages are properly placed or if the data should be migrated to a
498memory node local to where the task is running. Every "scan delay" the task
499scans the next "scan size" number of pages in its address space. When the
500end of the address space is reached the scanner restarts from the beginning.
501
502In combination, the "scan delay" and "scan size" determine the scan rate.
503When "scan delay" decreases, the scan rate increases. The scan delay and
504hence the scan rate of every task is adaptive and depends on historical
505behaviour. If pages are properly placed then the scan delay increases,
506otherwise the scan delay decreases. The "scan size" is not adaptive but
507the higher the "scan size", the higher the scan rate.
508
509Higher scan rates incur higher system overhead as page faults must be
510trapped and potentially data must be migrated. However, the higher the scan
511rate, the more quickly a tasks memory is migrated to a local node if the
512workload pattern changes and minimises performance impact due to remote
513memory accesses. These sysctls control the thresholds for scan delays and
514the number of pages scanned.
515
516numa_balancing_scan_period_min_ms is the minimum time in milliseconds to
517scan a tasks virtual memory. It effectively controls the maximum scanning
518rate for each task.
519
520numa_balancing_scan_delay_ms is the starting "scan delay" used for a task
521when it initially forks.
522
523numa_balancing_scan_period_max_ms is the maximum time in milliseconds to
524scan a tasks virtual memory. It effectively controls the minimum scanning
525rate for each task.
526
527numa_balancing_scan_size_mb is how many megabytes worth of pages are
528scanned for a given scan.
529
530==============================================================
531
532osrelease, ostype & version:
533
534# cat osrelease
5352.1.88
536# cat ostype
537Linux
538# cat version
539#5 Wed Feb 25 21:49:24 MET 1998
540
541The files osrelease and ostype should be clear enough. Version
542needs a little more clarification however. The '#5' means that
543this is the fifth kernel built from this source base and the
544date behind it indicates the time the kernel was built.
545The only way to tune these values is to rebuild the kernel :-)
546
547==============================================================
548
549overflowgid & overflowuid:
550
551if your architecture did not always support 32-bit UIDs (i.e. arm,
552i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to
553applications that use the old 16-bit UID/GID system calls, if the
554actual UID or GID would exceed 65535.
555
556These sysctls allow you to change the value of the fixed UID and GID.
557The default is 65534.
558
559==============================================================
560
561panic:
562
563The value in this file represents the number of seconds the kernel
564waits before rebooting on a panic. When you use the software watchdog,
565the recommended setting is 60.
566
567==============================================================
568
569panic_on_io_nmi:
570
571Controls the kernel's behavior when a CPU receives an NMI caused by
572an IO error.
573
5740: try to continue operation (default)
575
5761: panic immediately. The IO error triggered an NMI. This indicates a
577 serious system condition which could result in IO data corruption.
578 Rather than continuing, panicking might be a better choice. Some
579 servers issue this sort of NMI when the dump button is pushed,
580 and you can use this option to take a crash dump.
581
582==============================================================
583
584panic_on_oops:
585
586Controls the kernel's behaviour when an oops or BUG is encountered.
587
5880: try to continue operation
589
5901: panic immediately. If the `panic' sysctl is also non-zero then the
591 machine will be rebooted.
592
593==============================================================
594
595panic_on_stackoverflow:
596
597Controls the kernel's behavior when detecting the overflows of
598kernel, IRQ and exception stacks except a user stack.
599This file shows up if CONFIG_DEBUG_STACKOVERFLOW is enabled.
600
6010: try to continue operation.
602
6031: panic immediately.
604
605==============================================================
606
607panic_on_unrecovered_nmi:
608
609The default Linux behaviour on an NMI of either memory or unknown is
610to continue operation. For many environments such as scientific
611computing it is preferable that the box is taken out and the error
612dealt with than an uncorrected parity/ECC error get propagated.
613
614A small number of systems do generate NMI's for bizarre random reasons
615such as power management so the default is off. That sysctl works like
616the existing panic controls already in that directory.
617
618==============================================================
619
620panic_on_warn:
621
622Calls panic() in the WARN() path when set to 1. This is useful to avoid
623a kernel rebuild when attempting to kdump at the location of a WARN().
624
6250: only WARN(), default behaviour.
626
6271: call panic() after printing out WARN() location.
628
629==============================================================
630
631panic_on_rcu_stall:
632
633When set to 1, calls panic() after RCU stall detection messages. This
634is useful to define the root cause of RCU stalls using a vmcore.
635
6360: do not panic() when RCU stall takes place, default behavior.
637
6381: panic() after printing RCU stall messages.
639
640==============================================================
641
642perf_cpu_time_max_percent:
643
644Hints to the kernel how much CPU time it should be allowed to
645use to handle perf sampling events. If the perf subsystem
646is informed that its samples are exceeding this limit, it
647will drop its sampling frequency to attempt to reduce its CPU
648usage.
649
650Some perf sampling happens in NMIs. If these samples
651unexpectedly take too long to execute, the NMIs can become
652stacked up next to each other so much that nothing else is
653allowed to execute.
654
6550: disable the mechanism. Do not monitor or correct perf's
656 sampling rate no matter how CPU time it takes.
657
6581-100: attempt to throttle perf's sample rate to this
659 percentage of CPU. Note: the kernel calculates an
660 "expected" length of each sample event. 100 here means
661 100% of that expected length. Even if this is set to
662 100, you may still see sample throttling if this
663 length is exceeded. Set to 0 if you truly do not care
664 how much CPU is consumed.
665
666==============================================================
667
668perf_event_paranoid:
669
670Controls use of the performance events system by unprivileged
671users (without CAP_SYS_ADMIN). The default value is 2.
672
673 -1: Allow use of (almost) all events by all users
674 Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK
675>=0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN
676 Disallow raw tracepoint access by users without CAP_SYS_ADMIN
677>=1: Disallow CPU event access by users without CAP_SYS_ADMIN
678>=2: Disallow kernel profiling by users without CAP_SYS_ADMIN
679
680==============================================================
681
682perf_event_max_stack:
683
684Controls maximum number of stack frames to copy for (attr.sample_type &
685PERF_SAMPLE_CALLCHAIN) configured events, for instance, when using
686'perf record -g' or 'perf trace --call-graph fp'.
687
688This can only be done when no events are in use that have callchains
689enabled, otherwise writing to this file will return -EBUSY.
690
691The default value is 127.
692
693==============================================================
694
695perf_event_mlock_kb:
696
697Control size of per-cpu ring buffer not counted agains mlock limit.
698
699The default value is 512 + 1 page
700
701==============================================================
702
703perf_event_max_contexts_per_stack:
704
705Controls maximum number of stack frame context entries for
706(attr.sample_type & PERF_SAMPLE_CALLCHAIN) configured events, for
707instance, when using 'perf record -g' or 'perf trace --call-graph fp'.
708
709This can only be done when no events are in use that have callchains
710enabled, otherwise writing to this file will return -EBUSY.
711
712The default value is 8.
713
714==============================================================
715
716pid_max:
717
718PID allocation wrap value. When the kernel's next PID value
719reaches this value, it wraps back to a minimum PID value.
720PIDs of value pid_max or larger are not allocated.
721
722==============================================================
723
724ns_last_pid:
725
726The last pid allocated in the current (the one task using this sysctl
727lives in) pid namespace. When selecting a pid for a next task on fork
728kernel tries to allocate a number starting from this one.
729
730==============================================================
731
732powersave-nap: (PPC only)
733
734If set, Linux-PPC will use the 'nap' mode of powersaving,
735otherwise the 'doze' mode will be used.
736
737==============================================================
738
739printk:
740
741The four values in printk denote: console_loglevel,
742default_message_loglevel, minimum_console_loglevel and
743default_console_loglevel respectively.
744
745These values influence printk() behavior when printing or
746logging error messages. See 'man 2 syslog' for more info on
747the different loglevels.
748
749- console_loglevel: messages with a higher priority than
750 this will be printed to the console
751- default_message_loglevel: messages without an explicit priority
752 will be printed with this priority
753- minimum_console_loglevel: minimum (highest) value to which
754 console_loglevel can be set
755- default_console_loglevel: default value for console_loglevel
756
757==============================================================
758
759printk_delay:
760
761Delay each printk message in printk_delay milliseconds
762
763Value from 0 - 10000 is allowed.
764
765==============================================================
766
767printk_ratelimit:
768
769Some warning messages are rate limited. printk_ratelimit specifies
770the minimum length of time between these messages (in jiffies), by
771default we allow one every 5 seconds.
772
773A value of 0 will disable rate limiting.
774
775==============================================================
776
777printk_ratelimit_burst:
778
779While long term we enforce one message per printk_ratelimit
780seconds, we do allow a burst of messages to pass through.
781printk_ratelimit_burst specifies the number of messages we can
782send before ratelimiting kicks in.
783
784==============================================================
785
786printk_devkmsg:
787
788Control the logging to /dev/kmsg from userspace:
789
790ratelimit: default, ratelimited
791on: unlimited logging to /dev/kmsg from userspace
792off: logging to /dev/kmsg disabled
793
794The kernel command line parameter printk.devkmsg= overrides this and is
795a one-time setting until next reboot: once set, it cannot be changed by
796this sysctl interface anymore.
797
798==============================================================
799
800randomize_va_space:
801
802This option can be used to select the type of process address
803space randomization that is used in the system, for architectures
804that support this feature.
805
8060 - Turn the process address space randomization off. This is the
807 default for architectures that do not support this feature anyways,
808 and kernels that are booted with the "norandmaps" parameter.
809
8101 - Make the addresses of mmap base, stack and VDSO page randomized.
811 This, among other things, implies that shared libraries will be
812 loaded to random addresses. Also for PIE-linked binaries, the
813 location of code start is randomized. This is the default if the
814 CONFIG_COMPAT_BRK option is enabled.
815
8162 - Additionally enable heap randomization. This is the default if
817 CONFIG_COMPAT_BRK is disabled.
818
819 There are a few legacy applications out there (such as some ancient
820 versions of libc.so.5 from 1996) that assume that brk area starts
821 just after the end of the code+bss. These applications break when
822 start of the brk area is randomized. There are however no known
823 non-legacy applications that would be broken this way, so for most
824 systems it is safe to choose full randomization.
825
826 Systems with ancient and/or broken binaries should be configured
827 with CONFIG_COMPAT_BRK enabled, which excludes the heap from process
828 address space randomization.
829
830==============================================================
831
832reboot-cmd: (Sparc only)
833
834??? This seems to be a way to give an argument to the Sparc
835ROM/Flash boot loader. Maybe to tell it what to do after
836rebooting. ???
837
838==============================================================
839
840rtsig-max & rtsig-nr:
841
842The file rtsig-max can be used to tune the maximum number
843of POSIX realtime (queued) signals that can be outstanding
844in the system.
845
846rtsig-nr shows the number of RT signals currently queued.
847
848==============================================================
849
850sched_schedstats:
851
852Enables/disables scheduler statistics. Enabling this feature
853incurs a small amount of overhead in the scheduler but is
854useful for debugging and performance tuning.
855
856==============================================================
857
858sg-big-buff:
859
860This file shows the size of the generic SCSI (sg) buffer.
861You can't tune it just yet, but you could change it on
862compile time by editing include/scsi/sg.h and changing
863the value of SG_BIG_BUFF.
864
865There shouldn't be any reason to change this value. If
866you can come up with one, you probably know what you
867are doing anyway :)
868
869==============================================================
870
871shmall:
872
873This parameter sets the total amount of shared memory pages that
874can be used system wide. Hence, SHMALL should always be at least
875ceil(shmmax/PAGE_SIZE).
876
877If you are not sure what the default PAGE_SIZE is on your Linux
878system, you can run the following command:
879
880# getconf PAGE_SIZE
881
882==============================================================
883
884shmmax:
885
886This value can be used to query and set the run time limit
887on the maximum shared memory segment size that can be created.
888Shared memory segments up to 1Gb are now supported in the
889kernel. This value defaults to SHMMAX.
890
891==============================================================
892
893shm_rmid_forced:
894
895Linux lets you set resource limits, including how much memory one
896process can consume, via setrlimit(2). Unfortunately, shared memory
897segments are allowed to exist without association with any process, and
898thus might not be counted against any resource limits. If enabled,
899shared memory segments are automatically destroyed when their attach
900count becomes zero after a detach or a process termination. It will
901also destroy segments that were created, but never attached to, on exit
902from the process. The only use left for IPC_RMID is to immediately
903destroy an unattached segment. Of course, this breaks the way things are
904defined, so some applications might stop working. Note that this
905feature will do you no good unless you also configure your resource
906limits (in particular, RLIMIT_AS and RLIMIT_NPROC). Most systems don't
907need this.
908
909Note that if you change this from 0 to 1, already created segments
910without users and with a dead originative process will be destroyed.
911
912==============================================================
913
914sysctl_writes_strict:
915
916Control how file position affects the behavior of updating sysctl values
917via the /proc/sys interface:
918
919 -1 - Legacy per-write sysctl value handling, with no printk warnings.
920 Each write syscall must fully contain the sysctl value to be
921 written, and multiple writes on the same sysctl file descriptor
922 will rewrite the sysctl value, regardless of file position.
923 0 - Same behavior as above, but warn about processes that perform writes
924 to a sysctl file descriptor when the file position is not 0.
925 1 - (default) Respect file position when writing sysctl strings. Multiple
926 writes will append to the sysctl value buffer. Anything past the max
927 length of the sysctl value buffer will be ignored. Writes to numeric
928 sysctl entries must always be at file position 0 and the value must
929 be fully contained in the buffer sent in the write syscall.
930
931==============================================================
932
933softlockup_all_cpu_backtrace:
934
935This value controls the soft lockup detector thread's behavior
936when a soft lockup condition is detected as to whether or not
937to gather further debug information. If enabled, each cpu will
938be issued an NMI and instructed to capture stack trace.
939
940This feature is only applicable for architectures which support
941NMI.
942
9430: do nothing. This is the default behavior.
944
9451: on detection capture more debug information.
946
947==============================================================
948
949soft_watchdog
950
951This parameter can be used to control the soft lockup detector.
952
953 0 - disable the soft lockup detector
954 1 - enable the soft lockup detector
955
956The soft lockup detector monitors CPUs for threads that are hogging the CPUs
957without rescheduling voluntarily, and thus prevent the 'watchdog/N' threads
958from running. The mechanism depends on the CPUs ability to respond to timer
959interrupts which are needed for the 'watchdog/N' threads to be woken up by
960the watchdog timer function, otherwise the NMI watchdog - if enabled - can
961detect a hard lockup condition.
962
963==============================================================
964
965tainted:
966
967Non-zero if the kernel has been tainted. Numeric values, which can be
968ORed together. The letters are seen in "Tainted" line of Oops reports.
969
970 1 (P): A module with a non-GPL license has been loaded, this
971 includes modules with no license.
972 Set by modutils >= 2.4.9 and module-init-tools.
973 2 (F): A module was force loaded by insmod -f.
974 Set by modutils >= 2.4.9 and module-init-tools.
975 4 (S): Unsafe SMP processors: SMP with CPUs not designed for SMP.
976 8 (R): A module was forcibly unloaded from the system by rmmod -f.
977 16 (M): A hardware machine check error occurred on the system.
978 32 (B): A bad page was discovered on the system.
979 64 (U): The user has asked that the system be marked "tainted". This
980 could be because they are running software that directly modifies
981 the hardware, or for other reasons.
982 128 (D): The system has died.
983 256 (A): The ACPI DSDT has been overridden with one supplied by the user
984 instead of using the one provided by the hardware.
985 512 (W): A kernel warning has occurred.
986 1024 (C): A module from drivers/staging was loaded.
987 2048 (I): The system is working around a severe firmware bug.
988 4096 (O): An out-of-tree module has been loaded.
989 8192 (E): An unsigned module has been loaded in a kernel supporting module
990 signature.
991 16384 (L): A soft lockup has previously occurred on the system.
992 32768 (K): The kernel has been live patched.
993 65536 (X): Auxiliary taint, defined and used by for distros.
994131072 (T): The kernel was built with the struct randomization plugin.
995
996==============================================================
997
998threads-max
999
1000This value controls the maximum number of threads that can be created
1001using fork().
1002
1003During initialization the kernel sets this value such that even if the
1004maximum number of threads is created, the thread structures occupy only
1005a part (1/8th) of the available RAM pages.
1006
1007The minimum value that can be written to threads-max is 20.
1008The maximum value that can be written to threads-max is given by the
1009constant FUTEX_TID_MASK (0x3fffffff).
1010If a value outside of this range is written to threads-max an error
1011EINVAL occurs.
1012
1013The value written is checked against the available RAM pages. If the
1014thread structures would occupy too much (more than 1/8th) of the
1015available RAM pages threads-max is reduced accordingly.
1016
1017==============================================================
1018
1019unknown_nmi_panic:
1020
1021The value in this file affects behavior of handling NMI. When the
1022value is non-zero, unknown NMI is trapped and then panic occurs. At
1023that time, kernel debugging information is displayed on console.
1024
1025NMI switch that most IA32 servers have fires unknown NMI up, for
1026example. If a system hangs up, try pressing the NMI switch.
1027
1028==============================================================
1029
1030watchdog:
1031
1032This parameter can be used to disable or enable the soft lockup detector
1033_and_ the NMI watchdog (i.e. the hard lockup detector) at the same time.
1034
1035 0 - disable both lockup detectors
1036 1 - enable both lockup detectors
1037
1038The soft lockup detector and the NMI watchdog can also be disabled or
1039enabled individually, using the soft_watchdog and nmi_watchdog parameters.
1040If the watchdog parameter is read, for example by executing
1041
1042 cat /proc/sys/kernel/watchdog
1043
1044the output of this command (0 or 1) shows the logical OR of soft_watchdog
1045and nmi_watchdog.
1046
1047==============================================================
1048
1049watchdog_cpumask:
1050
1051This value can be used to control on which cpus the watchdog may run.
1052The default cpumask is all possible cores, but if NO_HZ_FULL is
1053enabled in the kernel config, and cores are specified with the
1054nohz_full= boot argument, those cores are excluded by default.
1055Offline cores can be included in this mask, and if the core is later
1056brought online, the watchdog will be started based on the mask value.
1057
1058Typically this value would only be touched in the nohz_full case
1059to re-enable cores that by default were not running the watchdog,
1060if a kernel lockup was suspected on those cores.
1061
1062The argument value is the standard cpulist format for cpumasks,
1063so for example to enable the watchdog on cores 0, 2, 3, and 4 you
1064might say:
1065
1066 echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask
1067
1068==============================================================
1069
1070watchdog_thresh:
1071
1072This value can be used to control the frequency of hrtimer and NMI
1073events and the soft and hard lockup thresholds. The default threshold
1074is 10 seconds.
1075
1076The softlockup threshold is (2 * watchdog_thresh). Setting this
1077tunable to zero will disable lockup detection altogether.
1078
1079==============================================================