1 ===================================
2 Documentation for /proc/sys/kernel/
3 ===================================
5 .. See scripts/check-sysctl-docs to keep this up to date
8 Copyright (c) 1998, 1999, Rik van Riel <riel@nl.linux.org>
10 Copyright (c) 2009, Shen Feng<shen@cn.fujitsu.com>
12 For general info and legal blurb, please look in :doc:`index`.
14 ------------------------------------------------------------------------------
16 This file contains documentation for the sysctl files in
17 ``/proc/sys/kernel/`` and is valid for Linux kernel version 2.2.
19 The files in this directory can be used to tune and monitor
20 miscellaneous and general things in the operation of the Linux
21 kernel. Since some of the files *can* be used to screw up your
22 system, it is advisable to read both documentation and source
23 before actually making adjustments.
25 Currently, these files might (depending on your configuration)
26 show up in ``/proc/sys/kernel``:
36 highwater lowwater frequency
38 If BSD-style process accounting is enabled these values control
39 its behaviour. If free space on filesystem where the log lives
40 goes below ``lowwater``% accounting suspends. If free space gets
41 above ``highwater``% accounting resumes. ``frequency`` determines
42 how often do we check the amount of free space (value is in
49 That is, suspend accounting if free space drops below 2%; resume it
50 if it increases to at least 4%; consider information about amount of
51 free space valid for 30 seconds.
57 See :doc:`/power/video`. This allows the video resume mode to be set,
58 in a similar fashion to the ``acpi_sleep`` kernel parameter, by
59 combining the following values:
71 This variable has no effect and may be removed in future kernel
72 releases. Reading it always returns 0.
73 Up to Linux 3.17, it enabled/disabled automatic recomputing of
75 upon memory add/remove or upon IPC namespace creation/removal.
76 Echoing "1" into this file enabled msgmni automatic recomputing.
77 Echoing "0" turned it off. The default value was 1.
80 bootloader_type (x86 only)
81 ==========================
83 This gives the bootloader type number as indicated by the bootloader,
84 shifted left by 4, and OR'd with the low four bits of the bootloader
85 version. The reason for this encoding is that this used to match the
86 ``type_of_loader`` field in the kernel header; the encoding is kept for
87 backwards compatibility. That is, if the full bootloader type number
88 is 0x15 and the full version number is 0x234, this file will contain
89 the value 340 = 0x154.
91 See the ``type_of_loader`` and ``ext_loader_type`` fields in
92 :doc:`/x86/boot` for additional information.
95 bootloader_version (x86 only)
96 =============================
98 The complete bootloader version number. In the example above, this
99 file will contain the value 564 = 0x234.
101 See the ``type_of_loader`` and ``ext_loader_ver`` fields in
102 :doc:`/x86/boot` for additional information.
108 Highest valid capability of the running kernel. Exports
109 ``CAP_LAST_CAP`` from the kernel.
115 ``core_pattern`` is used to specify a core dumpfile pattern name.
117 * max length 127 characters; default value is "core"
118 * ``core_pattern`` is used as a pattern template for the output
119 filename; certain string patterns (beginning with '%') are
120 substituted with their actual values.
121 * backward compatibility with ``core_uses_pid``:
123 If ``core_pattern`` does not include "%p" (default does not)
124 and ``core_uses_pid`` is set, then .PID will be appended to
127 * corename format specifiers
129 ======== ==========================================
130 %<NUL> '%' is dropped
133 %P global pid (init PID namespace)
135 %I global tid (init PID namespace)
136 %u uid (in initial user namespace)
137 %g gid (in initial user namespace)
138 %d dump mode, matches ``PR_SET_DUMPABLE`` and
139 ``/proc/sys/fs/suid_dumpable``
143 %e executable filename (may be shortened)
145 %c maximum size of core file by resource limit RLIMIT_CORE
146 %<OTHER> both are dropped
147 ======== ==========================================
149 * If the first character of the pattern is a '|', the kernel will treat
150 the rest of the pattern as a command to run. The core dump will be
151 written to the standard input of that program instead of to a file.
157 This sysctl is only applicable when `core_pattern`_ is configured to
158 pipe core files to a user space helper (when the first character of
159 ``core_pattern`` is a '|', see above).
160 When collecting cores via a pipe to an application, it is occasionally
161 useful for the collecting application to gather data about the
162 crashing process from its ``/proc/pid`` directory.
163 In order to do this safely, the kernel must wait for the collecting
164 process to exit, so as not to remove the crashing processes proc files
166 This in turn creates the possibility that a misbehaving userspace
167 collecting process can block the reaping of a crashed process simply
169 This sysctl defends against that.
170 It defines how many concurrent crashing processes may be piped to user
171 space applications in parallel.
172 If this value is exceeded, then those crashing processes above that
173 value are noted via the kernel log and their cores are skipped.
174 0 is a special value, indicating that unlimited processes may be
175 captured in parallel, but that no waiting will take place (i.e. the
176 collecting process is not guaranteed access to ``/proc/<crashing
178 This value defaults to 0.
184 The default coredump filename is "core". By setting
185 ``core_uses_pid`` to 1, the coredump filename becomes core.PID.
186 If `core_pattern`_ does not include "%p" (default does not)
187 and ``core_uses_pid`` is set, then .PID will be appended to
194 When the value in this file is 0, ctrl-alt-del is trapped and
195 sent to the ``init(1)`` program to handle a graceful restart.
196 When, however, the value is > 0, Linux's reaction to a Vulcan
197 Nerve Pinch (tm) will be an immediate reboot, without even
198 syncing its dirty buffers.
201 when a program (like dosemu) has the keyboard in 'raw'
202 mode, the ctrl-alt-del is intercepted by the program before it
203 ever reaches the kernel tty layer, and it's up to the program
204 to decide what to do with it.
210 This toggle indicates whether unprivileged users are prevented
211 from using ``dmesg(8)`` to view messages from the kernel's log
213 When ``dmesg_restrict`` is set to 0 there are no restrictions.
214 When ``dmesg_restrict`` is set set to 1, users must have
215 ``CAP_SYSLOG`` to use ``dmesg(8)``.
217 The kernel config option ``CONFIG_SECURITY_DMESG_RESTRICT`` sets the
218 default value of ``dmesg_restrict``.
221 domainname & hostname
222 =====================
224 These files can be used to set the NIS/YP domainname and the
225 hostname of your box in exactly the same way as the commands
226 domainname and hostname, i.e.::
228 # echo "darkstar" > /proc/sys/kernel/hostname
229 # echo "mydomain" > /proc/sys/kernel/domainname
231 has the same effect as::
233 # hostname "darkstar"
234 # domainname "mydomain"
236 Note, however, that the classic darkstar.frop.org has the
237 hostname "darkstar" and DNS (Internet Domain Name Server)
238 domainname "frop.org", not to be confused with the NIS (Network
239 Information Service) or YP (Yellow Pages) domainname. These two
240 domain names are in general different. For a detailed discussion
241 see the ``hostname(1)`` man page.
244 hardlockup_all_cpu_backtrace
245 ============================
247 This value controls the hard lockup detector behavior when a hard
248 lockup condition is detected as to whether or not to gather further
249 debug information. If enabled, arch-specific all-CPU stack dumping
252 = ============================================
253 0 Do nothing. This is the default behavior.
254 1 On detection capture more debug information.
255 = ============================================
261 This parameter can be used to control whether the kernel panics
262 when a hard lockup is detected.
264 = ===========================
265 0 Don't panic on hard lockup.
266 1 Panic on hard lockup.
267 = ===========================
269 See :doc:`/admin-guide/lockup-watchdogs` for more information.
270 This can also be set using the nmi_watchdog kernel parameter.
276 Path for the hotplug policy agent.
277 Default value is "``/sbin/hotplug``".
283 Controls the kernel's behavior when a hung task is detected.
284 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
286 = =================================================
287 0 Continue operation. This is the default behavior.
289 = =================================================
292 hung_task_check_count
293 =====================
295 The upper bound on the number of tasks that are checked.
296 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
299 hung_task_timeout_secs
300 ======================
302 When a task in D state did not get scheduled
303 for more than this value report a warning.
304 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
306 0 means infinite timeout, no checking is done.
308 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
311 hung_task_check_interval_secs
312 =============================
314 Hung task check interval. If hung task checking is enabled
315 (see `hung_task_timeout_secs`_), the check is done every
316 ``hung_task_check_interval_secs`` seconds.
317 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
319 0 (default) means use ``hung_task_timeout_secs`` as checking
322 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
328 The maximum number of warnings to report. During a check interval
329 if a hung task is detected, this value is decreased by 1.
330 When this value reaches 0, no more warnings will be reported.
331 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
333 -1: report an infinite number of warnings.
336 hyperv_record_panic_msg
337 =======================
339 Controls whether the panic kmsg data should be reported to Hyper-V.
341 = =========================================================
342 0 Do not report panic kmsg data.
343 1 Report the panic kmsg data. This is the default behavior.
344 = =========================================================
350 A toggle indicating if the ``kexec_load`` syscall has been disabled.
351 This value defaults to 0 (false: ``kexec_load`` enabled), but can be
352 set to 1 (true: ``kexec_load`` disabled).
353 Once true, kexec can no longer be used, and the toggle cannot be set
355 This allows a kexec image to be loaded before disabling the syscall,
356 allowing a system to set up (and later use) an image without it being
358 Generally used together with the `modules_disabled`_ sysctl.
364 This toggle indicates whether restrictions are placed on
365 exposing kernel addresses via ``/proc`` and other interfaces.
367 When ``kptr_restrict`` is set to 0 (the default) the address is hashed
369 (This is the equivalent to %p.)
371 When ``kptr_restrict`` is set to 1, kernel pointers printed using the
372 %pK format specifier will be replaced with 0s unless the user has
373 ``CAP_SYSLOG`` and effective user and group ids are equal to the real
375 This is because %pK checks are done at read() time rather than open()
376 time, so if permissions are elevated between the open() and the read()
377 (e.g via a setuid binary) then %pK will not leak kernel pointers to
379 Note, this is a temporary solution only.
380 The correct long-term solution is to do the permission checks at
382 Consider removing world read permissions from files that use %pK, and
383 using `dmesg_restrict`_ to protect against uses of %pK in ``dmesg(8)``
384 if leaking kernel pointer values to unprivileged users is a concern.
386 When ``kptr_restrict`` is set to 2, kernel pointers printed using
387 %pK will be replaced with 0s regardless of privileges.
393 This gives the full path of the modprobe command which the kernel will
394 use to load modules. This can be used to debug module loading
397 echo '#! /bin/sh' > /tmp/modprobe
398 echo 'echo "$@" >> /tmp/modprobe.log' >> /tmp/modprobe
399 echo 'exec /sbin/modprobe "$@"' >> /tmp/modprobe
400 chmod a+x /tmp/modprobe
401 echo /tmp/modprobe > /proc/sys/kernel/modprobe
403 This only applies when the *kernel* is requesting that the module be
404 loaded; it won't have any effect if the module is being loaded
405 explicitly using ``modprobe`` from userspace.
411 A toggle value indicating if modules are allowed to be loaded
412 in an otherwise modular kernel. This toggle defaults to off
413 (0), but can be set true (1). Once true, modules can be
414 neither loaded nor unloaded, and the toggle cannot be set back
415 to false. Generally used with the `kexec_load_disabled`_ toggle.
420 msgmax, msgmnb, and msgmni
421 ==========================
423 ``msgmax`` is the maximum size of an IPC message, in bytes. 8192 by
424 default (``MSGMAX``).
426 ``msgmnb`` is the maximum size of an IPC queue, in bytes. 16384 by
427 default (``MSGMNB``).
429 ``msgmni`` is the maximum number of IPC queues. 32000 by default
433 msg_next_id, sem_next_id, and shm_next_id (System V IPC)
434 ========================================================
436 These three toggles allows to specify desired id for next allocated IPC
437 object: message, semaphore or shared memory respectively.
439 By default they are equal to -1, which means generic allocation logic.
440 Possible values to set are in range {0:``INT_MAX``}.
443 1) kernel doesn't guarantee, that new object will have desired id. So,
444 it's up to userspace, how to handle an object with "wrong" id.
445 2) Toggle with non-default value will be set back to -1 by kernel after
446 successful IPC object allocation. If an IPC object allocation syscall
447 fails, it is undefined if the value remains unmodified or is reset to -1.
452 The path to the usermode helper for autoloading kernel modules, by
453 default "/sbin/modprobe". This binary is executed when the kernel
454 requests a module. For example, if userspace passes an unknown
455 filesystem type to mount(), then the kernel will automatically request
456 the corresponding filesystem module by executing this usermode helper.
457 This usermode helper should insert the needed module into the kernel.
459 This sysctl only affects module autoloading. It has no effect on the
460 ability to explicitly insert modules.
462 If this sysctl is set to the empty string, then module autoloading is
463 completely disabled. The kernel will not try to execute a usermode
464 helper at all, nor will it call the kernel_module_request LSM hook.
466 If CONFIG_STATIC_USERMODEHELPER=y is set in the kernel configuration,
467 then the configured static usermode helper overrides this sysctl,
468 except that the empty string is still accepted to completely disable
469 module autoloading as described above.
474 This parameter can be used to control the NMI watchdog
475 (i.e. the hard lockup detector) on x86 systems.
477 = =================================
478 0 Disable the hard lockup detector.
479 1 Enable the hard lockup detector.
480 = =================================
482 The hard lockup detector monitors each CPU for its ability to respond to
483 timer interrupts. The mechanism utilizes CPU performance counter registers
484 that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
485 while a CPU is busy. Hence, the alternative name 'NMI watchdog'.
487 The NMI watchdog is disabled by default if the kernel is running as a guest
488 in a KVM virtual machine. This default can be overridden by adding::
492 to the guest kernel command line (see :doc:`/admin-guide/kernel-parameters`).
498 Enables/disables automatic page fault based NUMA memory
499 balancing. Memory is moved automatically to nodes
500 that access it often.
502 Enables/disables automatic NUMA memory balancing. On NUMA machines, there
503 is a performance penalty if remote memory is accessed by a CPU. When this
504 feature is enabled the kernel samples what task thread is accessing memory
505 by periodically unmapping pages and later trapping a page fault. At the
506 time of the page fault, it is determined if the data being accessed should
507 be migrated to a local memory node.
509 The unmapping of pages and trapping faults incur additional overhead that
510 ideally is offset by improved memory locality but there is no universal
511 guarantee. If the target workload is already bound to NUMA nodes then this
512 feature should be disabled. Otherwise, if the system overhead from the
513 feature is too high then the rate the kernel samples for NUMA hinting
514 faults may be controlled by the `numa_balancing_scan_period_min_ms,
515 numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms,
516 numa_balancing_scan_size_mb`_, and numa_balancing_settle_count sysctls.
519 numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb
520 ===============================================================================================================================
523 Automatic NUMA balancing scans tasks address space and unmaps pages to
524 detect if pages are properly placed or if the data should be migrated to a
525 memory node local to where the task is running. Every "scan delay" the task
526 scans the next "scan size" number of pages in its address space. When the
527 end of the address space is reached the scanner restarts from the beginning.
529 In combination, the "scan delay" and "scan size" determine the scan rate.
530 When "scan delay" decreases, the scan rate increases. The scan delay and
531 hence the scan rate of every task is adaptive and depends on historical
532 behaviour. If pages are properly placed then the scan delay increases,
533 otherwise the scan delay decreases. The "scan size" is not adaptive but
534 the higher the "scan size", the higher the scan rate.
536 Higher scan rates incur higher system overhead as page faults must be
537 trapped and potentially data must be migrated. However, the higher the scan
538 rate, the more quickly a tasks memory is migrated to a local node if the
539 workload pattern changes and minimises performance impact due to remote
540 memory accesses. These sysctls control the thresholds for scan delays and
541 the number of pages scanned.
543 ``numa_balancing_scan_period_min_ms`` is the minimum time in milliseconds to
544 scan a tasks virtual memory. It effectively controls the maximum scanning
547 ``numa_balancing_scan_delay_ms`` is the starting "scan delay" used for a task
548 when it initially forks.
550 ``numa_balancing_scan_period_max_ms`` is the maximum time in milliseconds to
551 scan a tasks virtual memory. It effectively controls the minimum scanning
554 ``numa_balancing_scan_size_mb`` is how many megabytes worth of pages are
555 scanned for a given scan.
558 osrelease, ostype & version
559 ===========================
568 #5 Wed Feb 25 21:49:24 MET 1998
570 The files ``osrelease`` and ``ostype`` should be clear enough.
572 needs a little more clarification however. The '#5' means that
573 this is the fifth kernel built from this source base and the
574 date behind it indicates the time the kernel was built.
575 The only way to tune these values is to rebuild the kernel :-)
578 overflowgid & overflowuid
579 =========================
581 if your architecture did not always support 32-bit UIDs (i.e. arm,
582 i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to
583 applications that use the old 16-bit UID/GID system calls, if the
584 actual UID or GID would exceed 65535.
586 These sysctls allow you to change the value of the fixed UID and GID.
587 The default is 65534.
593 The value in this file determines the behaviour of the kernel on a
596 * if zero, the kernel will loop forever;
597 * if negative, the kernel will reboot immediately;
598 * if positive, the kernel will reboot after the corresponding number
601 When you use the software watchdog, the recommended setting is 60.
607 Controls the kernel's behavior when a CPU receives an NMI caused by
610 = ==================================================================
611 0 Try to continue operation (default).
612 1 Panic immediately. The IO error triggered an NMI. This indicates a
613 serious system condition which could result in IO data corruption.
614 Rather than continuing, panicking might be a better choice. Some
615 servers issue this sort of NMI when the dump button is pushed,
616 and you can use this option to take a crash dump.
617 = ==================================================================
623 Controls the kernel's behaviour when an oops or BUG is encountered.
625 = ===================================================================
626 0 Try to continue operation.
627 1 Panic immediately. If the `panic` sysctl is also non-zero then the
628 machine will be rebooted.
629 = ===================================================================
632 panic_on_stackoverflow
633 ======================
635 Controls the kernel's behavior when detecting the overflows of
636 kernel, IRQ and exception stacks except a user stack.
637 This file shows up if ``CONFIG_DEBUG_STACKOVERFLOW`` is enabled.
639 = ==========================
640 0 Try to continue operation.
642 = ==========================
645 panic_on_unrecovered_nmi
646 ========================
648 The default Linux behaviour on an NMI of either memory or unknown is
649 to continue operation. For many environments such as scientific
650 computing it is preferable that the box is taken out and the error
651 dealt with than an uncorrected parity/ECC error get propagated.
653 A small number of systems do generate NMIs for bizarre random reasons
654 such as power management so the default is off. That sysctl works like
655 the existing panic controls already in that directory.
661 Calls panic() in the WARN() path when set to 1. This is useful to avoid
662 a kernel rebuild when attempting to kdump at the location of a WARN().
664 = ================================================
665 0 Only WARN(), default behaviour.
666 1 Call panic() after printing out WARN() location.
667 = ================================================
673 Bitmask for printing system info when panic happens. User can chose
674 combination of the following bits:
676 ===== ============================================
677 bit 0 print all tasks info
678 bit 1 print system memory info
679 bit 2 print timer info
680 bit 3 print locks info if ``CONFIG_LOCKDEP`` is on
681 bit 4 print ftrace buffer
682 ===== ============================================
684 So for example to print tasks and memory info on panic, user can::
686 echo 3 > /proc/sys/kernel/panic_print
692 When set to 1, calls panic() after RCU stall detection messages. This
693 is useful to define the root cause of RCU stalls using a vmcore.
695 = ============================================================
696 0 Do not panic() when RCU stall takes place, default behavior.
697 1 panic() after printing RCU stall messages.
698 = ============================================================
701 perf_cpu_time_max_percent
702 =========================
704 Hints to the kernel how much CPU time it should be allowed to
705 use to handle perf sampling events. If the perf subsystem
706 is informed that its samples are exceeding this limit, it
707 will drop its sampling frequency to attempt to reduce its CPU
710 Some perf sampling happens in NMIs. If these samples
711 unexpectedly take too long to execute, the NMIs can become
712 stacked up next to each other so much that nothing else is
715 ===== ========================================================
716 0 Disable the mechanism. Do not monitor or correct perf's
717 sampling rate no matter how CPU time it takes.
719 1-100 Attempt to throttle perf's sample rate to this
720 percentage of CPU. Note: the kernel calculates an
721 "expected" length of each sample event. 100 here means
722 100% of that expected length. Even if this is set to
723 100, you may still see sample throttling if this
724 length is exceeded. Set to 0 if you truly do not care
725 how much CPU is consumed.
726 ===== ========================================================
732 Controls use of the performance events system by unprivileged
733 users (without CAP_SYS_ADMIN). The default value is 2.
735 === ==================================================================
736 -1 Allow use of (almost) all events by all users.
738 Ignore mlock limit after perf_event_mlock_kb without
741 >=0 Disallow ftrace function tracepoint by users without
744 Disallow raw tracepoint access by users without ``CAP_SYS_ADMIN``.
746 >=1 Disallow CPU event access by users without ``CAP_SYS_ADMIN``.
748 >=2 Disallow kernel profiling by users without ``CAP_SYS_ADMIN``.
749 === ==================================================================
755 Controls maximum number of stack frames to copy for (``attr.sample_type &
756 PERF_SAMPLE_CALLCHAIN``) configured events, for instance, when using
757 '``perf record -g``' or '``perf trace --call-graph fp``'.
759 This can only be done when no events are in use that have callchains
760 enabled, otherwise writing to this file will return ``-EBUSY``.
762 The default value is 127.
768 Control size of per-cpu ring buffer not counted agains mlock limit.
770 The default value is 512 + 1 page
773 perf_event_max_contexts_per_stack
774 =================================
776 Controls maximum number of stack frame context entries for
777 (``attr.sample_type & PERF_SAMPLE_CALLCHAIN``) configured events, for
778 instance, when using '``perf record -g``' or '``perf trace --call-graph fp``'.
780 This can only be done when no events are in use that have callchains
781 enabled, otherwise writing to this file will return ``-EBUSY``.
783 The default value is 8.
789 PID allocation wrap value. When the kernel's next PID value
790 reaches this value, it wraps back to a minimum PID value.
791 PIDs of value ``pid_max`` or larger are not allocated.
797 The last pid allocated in the current (the one task using this sysctl
798 lives in) pid namespace. When selecting a pid for a next task on fork
799 kernel tries to allocate a number starting from this one.
802 powersave-nap (PPC only)
803 ========================
805 If set, Linux-PPC will use the 'nap' mode of powersaving,
806 otherwise the 'doze' mode will be used.
809 ==============================================================
814 The four values in printk denote: ``console_loglevel``,
815 ``default_message_loglevel``, ``minimum_console_loglevel`` and
816 ``default_console_loglevel`` respectively.
818 These values influence printk() behavior when printing or
819 logging error messages. See '``man 2 syslog``' for more info on
820 the different loglevels.
822 ======================== =====================================
823 console_loglevel messages with a higher priority than
824 this will be printed to the console
825 default_message_loglevel messages without an explicit priority
826 will be printed with this priority
827 minimum_console_loglevel minimum (highest) value to which
828 console_loglevel can be set
829 default_console_loglevel default value for console_loglevel
830 ======================== =====================================
836 Delay each printk message in ``printk_delay`` milliseconds
838 Value from 0 - 10000 is allowed.
844 Some warning messages are rate limited. ``printk_ratelimit`` specifies
845 the minimum length of time between these messages (in seconds).
846 The default value is 5 seconds.
848 A value of 0 will disable rate limiting.
851 printk_ratelimit_burst
852 ======================
854 While long term we enforce one message per `printk_ratelimit`_
855 seconds, we do allow a burst of messages to pass through.
856 ``printk_ratelimit_burst`` specifies the number of messages we can
857 send before ratelimiting kicks in.
859 The default value is 10 messages.
865 Control the logging to ``/dev/kmsg`` from userspace:
867 ========= =============================================
868 ratelimit default, ratelimited
869 on unlimited logging to /dev/kmsg from userspace
870 off logging to /dev/kmsg disabled
871 ========= =============================================
873 The kernel command line parameter ``printk.devkmsg=`` overrides this and is
874 a one-time setting until next reboot: once set, it cannot be changed by
875 this sysctl interface anymore.
877 ==============================================================
883 See Documentation/filesystems/devpts.txt.
889 This option can be used to select the type of process address
890 space randomization that is used in the system, for architectures
891 that support this feature.
893 == ===========================================================================
894 0 Turn the process address space randomization off. This is the
895 default for architectures that do not support this feature anyways,
896 and kernels that are booted with the "norandmaps" parameter.
898 1 Make the addresses of mmap base, stack and VDSO page randomized.
899 This, among other things, implies that shared libraries will be
900 loaded to random addresses. Also for PIE-linked binaries, the
901 location of code start is randomized. This is the default if the
902 ``CONFIG_COMPAT_BRK`` option is enabled.
904 2 Additionally enable heap randomization. This is the default if
905 ``CONFIG_COMPAT_BRK`` is disabled.
907 There are a few legacy applications out there (such as some ancient
908 versions of libc.so.5 from 1996) that assume that brk area starts
909 just after the end of the code+bss. These applications break when
910 start of the brk area is randomized. There are however no known
911 non-legacy applications that would be broken this way, so for most
912 systems it is safe to choose full randomization.
914 Systems with ancient and/or broken binaries should be configured
915 with ``CONFIG_COMPAT_BRK`` enabled, which excludes the heap from process
916 address space randomization.
917 == ===========================================================================
923 See :doc:`/admin-guide/initrd`.
926 reboot-cmd (SPARC only)
927 =======================
929 ??? This seems to be a way to give an argument to the Sparc
930 ROM/Flash boot loader. Maybe to tell it what to do after
937 Enables/disables Energy Aware Scheduling (EAS). EAS starts
938 automatically on platforms where it can run (that is,
939 platforms with asymmetric CPU topologies and having an Energy
940 Model available). If your platform happens to meet the
941 requirements for EAS but you do not want to use it, change
948 Enables/disables scheduler statistics. Enabling this feature
949 incurs a small amount of overhead in the scheduler but is
950 useful for debugging and performance tuning.
956 See :doc:`/userspace-api/seccomp_filter`.
962 This file shows the size of the generic SCSI (sg) buffer.
963 You can't tune it just yet, but you could change it on
964 compile time by editing ``include/scsi/sg.h`` and changing
965 the value of ``SG_BIG_BUFF``.
967 There shouldn't be any reason to change this value. If
968 you can come up with one, you probably know what you
975 This parameter sets the total amount of shared memory pages that
976 can be used system wide. Hence, ``shmall`` should always be at least
977 ``ceil(shmmax/PAGE_SIZE)``.
979 If you are not sure what the default ``PAGE_SIZE`` is on your Linux
980 system, you can run the following command::
988 This value can be used to query and set the run time limit
989 on the maximum shared memory segment size that can be created.
990 Shared memory segments up to 1Gb are now supported in the
991 kernel. This value defaults to ``SHMMAX``.
997 This value determines the maximum number of shared memory segments.
998 4096 by default (``SHMMNI``).
1004 Linux lets you set resource limits, including how much memory one
1005 process can consume, via ``setrlimit(2)``. Unfortunately, shared memory
1006 segments are allowed to exist without association with any process, and
1007 thus might not be counted against any resource limits. If enabled,
1008 shared memory segments are automatically destroyed when their attach
1009 count becomes zero after a detach or a process termination. It will
1010 also destroy segments that were created, but never attached to, on exit
1011 from the process. The only use left for ``IPC_RMID`` is to immediately
1012 destroy an unattached segment. Of course, this breaks the way things are
1013 defined, so some applications might stop working. Note that this
1014 feature will do you no good unless you also configure your resource
1015 limits (in particular, ``RLIMIT_AS`` and ``RLIMIT_NPROC``). Most systems don't
1018 Note that if you change this from 0 to 1, already created segments
1019 without users and with a dead originative process will be destroyed.
1022 sysctl_writes_strict
1023 ====================
1025 Control how file position affects the behavior of updating sysctl values
1026 via the ``/proc/sys`` interface:
1028 == ======================================================================
1029 -1 Legacy per-write sysctl value handling, with no printk warnings.
1030 Each write syscall must fully contain the sysctl value to be
1031 written, and multiple writes on the same sysctl file descriptor
1032 will rewrite the sysctl value, regardless of file position.
1033 0 Same behavior as above, but warn about processes that perform writes
1034 to a sysctl file descriptor when the file position is not 0.
1035 1 (default) Respect file position when writing sysctl strings. Multiple
1036 writes will append to the sysctl value buffer. Anything past the max
1037 length of the sysctl value buffer will be ignored. Writes to numeric
1038 sysctl entries must always be at file position 0 and the value must
1039 be fully contained in the buffer sent in the write syscall.
1040 == ======================================================================
1043 softlockup_all_cpu_backtrace
1044 ============================
1046 This value controls the soft lockup detector thread's behavior
1047 when a soft lockup condition is detected as to whether or not
1048 to gather further debug information. If enabled, each cpu will
1049 be issued an NMI and instructed to capture stack trace.
1051 This feature is only applicable for architectures which support
1054 = ============================================
1055 0 Do nothing. This is the default behavior.
1056 1 On detection capture more debug information.
1057 = ============================================
1063 This parameter can be used to control whether the kernel panics
1064 when a soft lockup is detected.
1066 = ============================================
1067 0 Don't panic on soft lockup.
1068 1 Panic on soft lockup.
1069 = ============================================
1071 This can also be set using the softlockup_panic kernel parameter.
1077 This parameter can be used to control the soft lockup detector.
1079 = =================================
1080 0 Disable the soft lockup detector.
1081 1 Enable the soft lockup detector.
1082 = =================================
1084 The soft lockup detector monitors CPUs for threads that are hogging the CPUs
1085 without rescheduling voluntarily, and thus prevent the 'watchdog/N' threads
1086 from running. The mechanism depends on the CPUs ability to respond to timer
1087 interrupts which are needed for the 'watchdog/N' threads to be woken up by
1088 the watchdog timer function, otherwise the NMI watchdog — if enabled — can
1089 detect a hard lockup condition.
1095 This parameter can be used to control kernel stack erasing at the end
1096 of syscalls for kernels built with ``CONFIG_GCC_PLUGIN_STACKLEAK``.
1098 That erasing reduces the information which kernel stack leak bugs
1099 can reveal and blocks some uninitialized stack variable attacks.
1100 The tradeoff is the performance impact: on a single CPU system kernel
1101 compilation sees a 1% slowdown, other systems and workloads may vary.
1103 = ====================================================================
1104 0 Kernel stack erasing is disabled, STACKLEAK_METRICS are not updated.
1105 1 Kernel stack erasing is enabled (default), it is performed before
1106 returning to the userspace at the end of syscalls.
1107 = ====================================================================
1115 = ====================================
1116 0 Stop-A has no effect.
1117 1 Stop-A breaks to the PROM (default).
1118 = ====================================
1120 Stop-A is always enabled on a panic, so that the user can return to
1127 See :doc:`/admin-guide/sysrq`.
1133 Non-zero if the kernel has been tainted. Numeric values, which can be
1134 ORed together. The letters are seen in "Tainted" line of Oops reports.
1136 ====== ===== ==============================================================
1137 1 `(P)` proprietary module was loaded
1138 2 `(F)` module was force loaded
1139 4 `(S)` SMP kernel oops on an officially SMP incapable processor
1140 8 `(R)` module was force unloaded
1141 16 `(M)` processor reported a Machine Check Exception (MCE)
1142 32 `(B)` bad page referenced or some unexpected page flags
1143 64 `(U)` taint requested by userspace application
1144 128 `(D)` kernel died recently, i.e. there was an OOPS or BUG
1145 256 `(A)` an ACPI table was overridden by user
1146 512 `(W)` kernel issued warning
1147 1024 `(C)` staging driver was loaded
1148 2048 `(I)` workaround for bug in platform firmware applied
1149 4096 `(O)` externally-built ("out-of-tree") module was loaded
1150 8192 `(E)` unsigned module was loaded
1151 16384 `(L)` soft lockup occurred
1152 32768 `(K)` kernel has been live patched
1153 65536 `(X)` Auxiliary taint, defined and used by for distros
1154 131072 `(T)` The kernel was built with the struct randomization plugin
1155 ====== ===== ==============================================================
1157 See :doc:`/admin-guide/tainted-kernels` for more information.
1163 This value controls the maximum number of threads that can be created
1166 During initialization the kernel sets this value such that even if the
1167 maximum number of threads is created, the thread structures occupy only
1168 a part (1/8th) of the available RAM pages.
1170 The minimum value that can be written to ``threads-max`` is 1.
1172 The maximum value that can be written to ``threads-max`` is given by the
1173 constant ``FUTEX_TID_MASK`` (0x3fffffff).
1175 If a value outside of this range is written to ``threads-max`` an
1176 ``EINVAL`` error occurs.
1182 The value in this file affects behavior of handling NMI. When the
1183 value is non-zero, unknown NMI is trapped and then panic occurs. At
1184 that time, kernel debugging information is displayed on console.
1186 NMI switch that most IA32 servers have fires unknown NMI up, for
1187 example. If a system hangs up, try pressing the NMI switch.
1193 This parameter can be used to disable or enable the soft lockup detector
1194 *and* the NMI watchdog (i.e. the hard lockup detector) at the same time.
1196 = ==============================
1197 0 Disable both lockup detectors.
1198 1 Enable both lockup detectors.
1199 = ==============================
1201 The soft lockup detector and the NMI watchdog can also be disabled or
1202 enabled individually, using the ``soft_watchdog`` and ``nmi_watchdog``
1204 If the ``watchdog`` parameter is read, for example by executing::
1206 cat /proc/sys/kernel/watchdog
1208 the output of this command (0 or 1) shows the logical OR of
1209 ``soft_watchdog`` and ``nmi_watchdog``.
1215 This value can be used to control on which cpus the watchdog may run.
1216 The default cpumask is all possible cores, but if ``NO_HZ_FULL`` is
1217 enabled in the kernel config, and cores are specified with the
1218 ``nohz_full=`` boot argument, those cores are excluded by default.
1219 Offline cores can be included in this mask, and if the core is later
1220 brought online, the watchdog will be started based on the mask value.
1222 Typically this value would only be touched in the ``nohz_full`` case
1223 to re-enable cores that by default were not running the watchdog,
1224 if a kernel lockup was suspected on those cores.
1226 The argument value is the standard cpulist format for cpumasks,
1227 so for example to enable the watchdog on cores 0, 2, 3, and 4 you
1230 echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask
1236 This value can be used to control the frequency of hrtimer and NMI
1237 events and the soft and hard lockup thresholds. The default threshold
1240 The softlockup threshold is (``2 * watchdog_thresh``). Setting this
1241 tunable to zero will disable lockup detection altogether.