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/``.
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 Controls whether the kernel should collect statistics on BPF programs
109 (total time spent running, number of times run...). Enabling
110 statistics causes a slight reduction in performance on each program
111 run. The statistics can be seen using ``bpftool``.
113 = ===================================
114 0 Don't collect statistics (default).
115 1 Collect statistics.
116 = ===================================
122 This is the pid which will be signalled on reboot (notably, by
123 Ctrl-Alt-Delete). Writing a value to this file which doesn't
124 correspond to a running process will result in ``-ESRCH``.
126 See also `ctrl-alt-del`_.
132 Highest valid capability of the running kernel. Exports
133 ``CAP_LAST_CAP`` from the kernel.
139 ``core_pattern`` is used to specify a core dumpfile pattern name.
141 * max length 127 characters; default value is "core"
142 * ``core_pattern`` is used as a pattern template for the output
143 filename; certain string patterns (beginning with '%') are
144 substituted with their actual values.
145 * backward compatibility with ``core_uses_pid``:
147 If ``core_pattern`` does not include "%p" (default does not)
148 and ``core_uses_pid`` is set, then .PID will be appended to
151 * corename format specifiers
153 ======== ==========================================
154 %<NUL> '%' is dropped
157 %P global pid (init PID namespace)
159 %I global tid (init PID namespace)
160 %u uid (in initial user namespace)
161 %g gid (in initial user namespace)
162 %d dump mode, matches ``PR_SET_DUMPABLE`` and
163 ``/proc/sys/fs/suid_dumpable``
167 %e executable filename (may be shortened, could be changed by prctl etc)
168 %f executable filename
170 %c maximum size of core file by resource limit RLIMIT_CORE
171 %<OTHER> both are dropped
172 ======== ==========================================
174 * If the first character of the pattern is a '|', the kernel will treat
175 the rest of the pattern as a command to run. The core dump will be
176 written to the standard input of that program instead of to a file.
182 This sysctl is only applicable when `core_pattern`_ is configured to
183 pipe core files to a user space helper (when the first character of
184 ``core_pattern`` is a '|', see above).
185 When collecting cores via a pipe to an application, it is occasionally
186 useful for the collecting application to gather data about the
187 crashing process from its ``/proc/pid`` directory.
188 In order to do this safely, the kernel must wait for the collecting
189 process to exit, so as not to remove the crashing processes proc files
191 This in turn creates the possibility that a misbehaving userspace
192 collecting process can block the reaping of a crashed process simply
194 This sysctl defends against that.
195 It defines how many concurrent crashing processes may be piped to user
196 space applications in parallel.
197 If this value is exceeded, then those crashing processes above that
198 value are noted via the kernel log and their cores are skipped.
199 0 is a special value, indicating that unlimited processes may be
200 captured in parallel, but that no waiting will take place (i.e. the
201 collecting process is not guaranteed access to ``/proc/<crashing
203 This value defaults to 0.
209 The default coredump filename is "core". By setting
210 ``core_uses_pid`` to 1, the coredump filename becomes core.PID.
211 If `core_pattern`_ does not include "%p" (default does not)
212 and ``core_uses_pid`` is set, then .PID will be appended to
219 When the value in this file is 0, ctrl-alt-del is trapped and
220 sent to the ``init(1)`` program to handle a graceful restart.
221 When, however, the value is > 0, Linux's reaction to a Vulcan
222 Nerve Pinch (tm) will be an immediate reboot, without even
223 syncing its dirty buffers.
226 when a program (like dosemu) has the keyboard in 'raw'
227 mode, the ctrl-alt-del is intercepted by the program before it
228 ever reaches the kernel tty layer, and it's up to the program
229 to decide what to do with it.
235 This toggle indicates whether unprivileged users are prevented
236 from using ``dmesg(8)`` to view messages from the kernel's log
238 When ``dmesg_restrict`` is set to 0 there are no restrictions.
239 When ``dmesg_restrict`` is set to 1, users must have
240 ``CAP_SYSLOG`` to use ``dmesg(8)``.
242 The kernel config option ``CONFIG_SECURITY_DMESG_RESTRICT`` sets the
243 default value of ``dmesg_restrict``.
246 domainname & hostname
247 =====================
249 These files can be used to set the NIS/YP domainname and the
250 hostname of your box in exactly the same way as the commands
251 domainname and hostname, i.e.::
253 # echo "darkstar" > /proc/sys/kernel/hostname
254 # echo "mydomain" > /proc/sys/kernel/domainname
256 has the same effect as::
258 # hostname "darkstar"
259 # domainname "mydomain"
261 Note, however, that the classic darkstar.frop.org has the
262 hostname "darkstar" and DNS (Internet Domain Name Server)
263 domainname "frop.org", not to be confused with the NIS (Network
264 Information Service) or YP (Yellow Pages) domainname. These two
265 domain names are in general different. For a detailed discussion
266 see the ``hostname(1)`` man page.
272 See :doc:`/driver-api/firmware/fallback-mechanisms`.
274 The entries in this directory allow the firmware loader helper
275 fallback to be controlled:
277 * ``force_sysfs_fallback``, when set to 1, forces the use of the
279 * ``ignore_sysfs_fallback``, when set to 1, ignores any fallback.
285 Determines whether ``ftrace_dump()`` should be called on an oops (or
286 kernel panic). This will output the contents of the ftrace buffers to
287 the console. This is very useful for capturing traces that lead to
288 crashes and outputting them to a serial console.
290 = ===================================================
291 0 Disabled (default).
292 1 Dump buffers of all CPUs.
293 2 Dump the buffer of the CPU that triggered the oops.
294 = ===================================================
297 ftrace_enabled, stack_tracer_enabled
298 ====================================
300 See :doc:`/trace/ftrace`.
303 hardlockup_all_cpu_backtrace
304 ============================
306 This value controls the hard lockup detector behavior when a hard
307 lockup condition is detected as to whether or not to gather further
308 debug information. If enabled, arch-specific all-CPU stack dumping
311 = ============================================
312 0 Do nothing. This is the default behavior.
313 1 On detection capture more debug information.
314 = ============================================
320 This parameter can be used to control whether the kernel panics
321 when a hard lockup is detected.
323 = ===========================
324 0 Don't panic on hard lockup.
325 1 Panic on hard lockup.
326 = ===========================
328 See :doc:`/admin-guide/lockup-watchdogs` for more information.
329 This can also be set using the nmi_watchdog kernel parameter.
335 Path for the hotplug policy agent.
336 Default value is "``/sbin/hotplug``".
339 hung_task_all_cpu_backtrace
340 ===========================
342 If this option is set, the kernel will send an NMI to all CPUs to dump
343 their backtraces when a hung task is detected. This file shows up if
344 CONFIG_DETECT_HUNG_TASK and CONFIG_SMP are enabled.
346 0: Won't show all CPUs backtraces when a hung task is detected.
347 This is the default behavior.
349 1: Will non-maskably interrupt all CPUs and dump their backtraces when
350 a hung task is detected.
356 Controls the kernel's behavior when a hung task is detected.
357 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
359 = =================================================
360 0 Continue operation. This is the default behavior.
362 = =================================================
365 hung_task_check_count
366 =====================
368 The upper bound on the number of tasks that are checked.
369 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
372 hung_task_timeout_secs
373 ======================
375 When a task in D state did not get scheduled
376 for more than this value report a warning.
377 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
379 0 means infinite timeout, no checking is done.
381 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
384 hung_task_check_interval_secs
385 =============================
387 Hung task check interval. If hung task checking is enabled
388 (see `hung_task_timeout_secs`_), the check is done every
389 ``hung_task_check_interval_secs`` seconds.
390 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
392 0 (default) means use ``hung_task_timeout_secs`` as checking
395 Possible values to set are in range {0:``LONG_MAX``/``HZ``}.
401 The maximum number of warnings to report. During a check interval
402 if a hung task is detected, this value is decreased by 1.
403 When this value reaches 0, no more warnings will be reported.
404 This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
406 -1: report an infinite number of warnings.
409 hyperv_record_panic_msg
410 =======================
412 Controls whether the panic kmsg data should be reported to Hyper-V.
414 = =========================================================
415 0 Do not report panic kmsg data.
416 1 Report the panic kmsg data. This is the default behavior.
417 = =========================================================
420 ignore-unaligned-usertrap
421 =========================
423 On architectures where unaligned accesses cause traps, and where this
424 feature is supported (``CONFIG_SYSCTL_ARCH_UNALIGN_NO_WARN``;
425 currently, ``arc`` and ``ia64``), controls whether all unaligned traps
428 = =============================================================
429 0 Log all unaligned accesses.
430 1 Only warn the first time a process traps. This is the default
432 = =============================================================
434 See also `unaligned-trap`_ and `unaligned-dump-stack`_. On ``ia64``,
435 this allows system administrators to override the
436 ``IA64_THREAD_UAC_NOPRINT`` ``prctl`` and avoid logs being flooded.
442 A toggle indicating if the ``kexec_load`` syscall has been disabled.
443 This value defaults to 0 (false: ``kexec_load`` enabled), but can be
444 set to 1 (true: ``kexec_load`` disabled).
445 Once true, kexec can no longer be used, and the toggle cannot be set
447 This allows a kexec image to be loaded before disabling the syscall,
448 allowing a system to set up (and later use) an image without it being
450 Generally used together with the `modules_disabled`_ sysctl.
456 This toggle indicates whether restrictions are placed on
457 exposing kernel addresses via ``/proc`` and other interfaces.
459 When ``kptr_restrict`` is set to 0 (the default) the address is hashed
461 (This is the equivalent to %p.)
463 When ``kptr_restrict`` is set to 1, kernel pointers printed using the
464 %pK format specifier will be replaced with 0s unless the user has
465 ``CAP_SYSLOG`` and effective user and group ids are equal to the real
467 This is because %pK checks are done at read() time rather than open()
468 time, so if permissions are elevated between the open() and the read()
469 (e.g via a setuid binary) then %pK will not leak kernel pointers to
471 Note, this is a temporary solution only.
472 The correct long-term solution is to do the permission checks at
474 Consider removing world read permissions from files that use %pK, and
475 using `dmesg_restrict`_ to protect against uses of %pK in ``dmesg(8)``
476 if leaking kernel pointer values to unprivileged users is a concern.
478 When ``kptr_restrict`` is set to 2, kernel pointers printed using
479 %pK will be replaced with 0s regardless of privileges.
485 The full path to the usermode helper for autoloading kernel modules,
486 by default ``CONFIG_MODPROBE_PATH``, which in turn defaults to
487 "/sbin/modprobe". This binary is executed when the kernel requests a
488 module. For example, if userspace passes an unknown filesystem type
489 to mount(), then the kernel will automatically request the
490 corresponding filesystem module by executing this usermode helper.
491 This usermode helper should insert the needed module into the kernel.
493 This sysctl only affects module autoloading. It has no effect on the
494 ability to explicitly insert modules.
496 This sysctl can be used to debug module loading requests::
498 echo '#! /bin/sh' > /tmp/modprobe
499 echo 'echo "$@" >> /tmp/modprobe.log' >> /tmp/modprobe
500 echo 'exec /sbin/modprobe "$@"' >> /tmp/modprobe
501 chmod a+x /tmp/modprobe
502 echo /tmp/modprobe > /proc/sys/kernel/modprobe
504 Alternatively, if this sysctl is set to the empty string, then module
505 autoloading is completely disabled. The kernel will not try to
506 execute a usermode helper at all, nor will it call the
507 kernel_module_request LSM hook.
509 If CONFIG_STATIC_USERMODEHELPER=y is set in the kernel configuration,
510 then the configured static usermode helper overrides this sysctl,
511 except that the empty string is still accepted to completely disable
512 module autoloading as described above.
517 A toggle value indicating if modules are allowed to be loaded
518 in an otherwise modular kernel. This toggle defaults to off
519 (0), but can be set true (1). Once true, modules can be
520 neither loaded nor unloaded, and the toggle cannot be set back
521 to false. Generally used with the `kexec_load_disabled`_ toggle.
526 msgmax, msgmnb, and msgmni
527 ==========================
529 ``msgmax`` is the maximum size of an IPC message, in bytes. 8192 by
530 default (``MSGMAX``).
532 ``msgmnb`` is the maximum size of an IPC queue, in bytes. 16384 by
533 default (``MSGMNB``).
535 ``msgmni`` is the maximum number of IPC queues. 32000 by default
539 msg_next_id, sem_next_id, and shm_next_id (System V IPC)
540 ========================================================
542 These three toggles allows to specify desired id for next allocated IPC
543 object: message, semaphore or shared memory respectively.
545 By default they are equal to -1, which means generic allocation logic.
546 Possible values to set are in range {0:``INT_MAX``}.
549 1) kernel doesn't guarantee, that new object will have desired id. So,
550 it's up to userspace, how to handle an object with "wrong" id.
551 2) Toggle with non-default value will be set back to -1 by kernel after
552 successful IPC object allocation. If an IPC object allocation syscall
553 fails, it is undefined if the value remains unmodified or is reset to -1.
559 Maximum number of supplementary groups, _i.e._ the maximum size which
560 ``setgroups`` will accept. Exports ``NGROUPS_MAX`` from the kernel.
567 This parameter can be used to control the NMI watchdog
568 (i.e. the hard lockup detector) on x86 systems.
570 = =================================
571 0 Disable the hard lockup detector.
572 1 Enable the hard lockup detector.
573 = =================================
575 The hard lockup detector monitors each CPU for its ability to respond to
576 timer interrupts. The mechanism utilizes CPU performance counter registers
577 that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
578 while a CPU is busy. Hence, the alternative name 'NMI watchdog'.
580 The NMI watchdog is disabled by default if the kernel is running as a guest
581 in a KVM virtual machine. This default can be overridden by adding::
585 to the guest kernel command line (see :doc:`/admin-guide/kernel-parameters`).
591 Enables/disables automatic page fault based NUMA memory
592 balancing. Memory is moved automatically to nodes
593 that access it often.
595 Enables/disables automatic NUMA memory balancing. On NUMA machines, there
596 is a performance penalty if remote memory is accessed by a CPU. When this
597 feature is enabled the kernel samples what task thread is accessing memory
598 by periodically unmapping pages and later trapping a page fault. At the
599 time of the page fault, it is determined if the data being accessed should
600 be migrated to a local memory node.
602 The unmapping of pages and trapping faults incur additional overhead that
603 ideally is offset by improved memory locality but there is no universal
604 guarantee. If the target workload is already bound to NUMA nodes then this
605 feature should be disabled. Otherwise, if the system overhead from the
606 feature is too high then the rate the kernel samples for NUMA hinting
607 faults may be controlled by the `numa_balancing_scan_period_min_ms,
608 numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms,
609 numa_balancing_scan_size_mb`_, and numa_balancing_settle_count sysctls.
612 numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb
613 ===============================================================================================================================
616 Automatic NUMA balancing scans tasks address space and unmaps pages to
617 detect if pages are properly placed or if the data should be migrated to a
618 memory node local to where the task is running. Every "scan delay" the task
619 scans the next "scan size" number of pages in its address space. When the
620 end of the address space is reached the scanner restarts from the beginning.
622 In combination, the "scan delay" and "scan size" determine the scan rate.
623 When "scan delay" decreases, the scan rate increases. The scan delay and
624 hence the scan rate of every task is adaptive and depends on historical
625 behaviour. If pages are properly placed then the scan delay increases,
626 otherwise the scan delay decreases. The "scan size" is not adaptive but
627 the higher the "scan size", the higher the scan rate.
629 Higher scan rates incur higher system overhead as page faults must be
630 trapped and potentially data must be migrated. However, the higher the scan
631 rate, the more quickly a tasks memory is migrated to a local node if the
632 workload pattern changes and minimises performance impact due to remote
633 memory accesses. These sysctls control the thresholds for scan delays and
634 the number of pages scanned.
636 ``numa_balancing_scan_period_min_ms`` is the minimum time in milliseconds to
637 scan a tasks virtual memory. It effectively controls the maximum scanning
640 ``numa_balancing_scan_delay_ms`` is the starting "scan delay" used for a task
641 when it initially forks.
643 ``numa_balancing_scan_period_max_ms`` is the maximum time in milliseconds to
644 scan a tasks virtual memory. It effectively controls the minimum scanning
647 ``numa_balancing_scan_size_mb`` is how many megabytes worth of pages are
648 scanned for a given scan.
651 oops_all_cpu_backtrace
652 ======================
654 If this option is set, the kernel will send an NMI to all CPUs to dump
655 their backtraces when an oops event occurs. It should be used as a last
656 resort in case a panic cannot be triggered (to protect VMs running, for
657 example) or kdump can't be collected. This file shows up if CONFIG_SMP
660 0: Won't show all CPUs backtraces when an oops is detected.
661 This is the default behavior.
663 1: Will non-maskably interrupt all CPUs and dump their backtraces when
664 an oops event is detected.
667 osrelease, ostype & version
668 ===========================
677 #5 Wed Feb 25 21:49:24 MET 1998
679 The files ``osrelease`` and ``ostype`` should be clear enough.
681 needs a little more clarification however. The '#5' means that
682 this is the fifth kernel built from this source base and the
683 date behind it indicates the time the kernel was built.
684 The only way to tune these values is to rebuild the kernel :-)
687 overflowgid & overflowuid
688 =========================
690 if your architecture did not always support 32-bit UIDs (i.e. arm,
691 i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to
692 applications that use the old 16-bit UID/GID system calls, if the
693 actual UID or GID would exceed 65535.
695 These sysctls allow you to change the value of the fixed UID and GID.
696 The default is 65534.
702 The value in this file determines the behaviour of the kernel on a
705 * if zero, the kernel will loop forever;
706 * if negative, the kernel will reboot immediately;
707 * if positive, the kernel will reboot after the corresponding number
710 When you use the software watchdog, the recommended setting is 60.
716 Controls the kernel's behavior when a CPU receives an NMI caused by
719 = ==================================================================
720 0 Try to continue operation (default).
721 1 Panic immediately. The IO error triggered an NMI. This indicates a
722 serious system condition which could result in IO data corruption.
723 Rather than continuing, panicking might be a better choice. Some
724 servers issue this sort of NMI when the dump button is pushed,
725 and you can use this option to take a crash dump.
726 = ==================================================================
732 Controls the kernel's behaviour when an oops or BUG is encountered.
734 = ===================================================================
735 0 Try to continue operation.
736 1 Panic immediately. If the `panic` sysctl is also non-zero then the
737 machine will be rebooted.
738 = ===================================================================
741 panic_on_stackoverflow
742 ======================
744 Controls the kernel's behavior when detecting the overflows of
745 kernel, IRQ and exception stacks except a user stack.
746 This file shows up if ``CONFIG_DEBUG_STACKOVERFLOW`` is enabled.
748 = ==========================
749 0 Try to continue operation.
751 = ==========================
754 panic_on_unrecovered_nmi
755 ========================
757 The default Linux behaviour on an NMI of either memory or unknown is
758 to continue operation. For many environments such as scientific
759 computing it is preferable that the box is taken out and the error
760 dealt with than an uncorrected parity/ECC error get propagated.
762 A small number of systems do generate NMIs for bizarre random reasons
763 such as power management so the default is off. That sysctl works like
764 the existing panic controls already in that directory.
770 Calls panic() in the WARN() path when set to 1. This is useful to avoid
771 a kernel rebuild when attempting to kdump at the location of a WARN().
773 = ================================================
774 0 Only WARN(), default behaviour.
775 1 Call panic() after printing out WARN() location.
776 = ================================================
782 Bitmask for printing system info when panic happens. User can chose
783 combination of the following bits:
785 ===== ============================================
786 bit 0 print all tasks info
787 bit 1 print system memory info
788 bit 2 print timer info
789 bit 3 print locks info if ``CONFIG_LOCKDEP`` is on
790 bit 4 print ftrace buffer
791 ===== ============================================
793 So for example to print tasks and memory info on panic, user can::
795 echo 3 > /proc/sys/kernel/panic_print
801 When set to 1, calls panic() after RCU stall detection messages. This
802 is useful to define the root cause of RCU stalls using a vmcore.
804 = ============================================================
805 0 Do not panic() when RCU stall takes place, default behavior.
806 1 panic() after printing RCU stall messages.
807 = ============================================================
810 perf_cpu_time_max_percent
811 =========================
813 Hints to the kernel how much CPU time it should be allowed to
814 use to handle perf sampling events. If the perf subsystem
815 is informed that its samples are exceeding this limit, it
816 will drop its sampling frequency to attempt to reduce its CPU
819 Some perf sampling happens in NMIs. If these samples
820 unexpectedly take too long to execute, the NMIs can become
821 stacked up next to each other so much that nothing else is
824 ===== ========================================================
825 0 Disable the mechanism. Do not monitor or correct perf's
826 sampling rate no matter how CPU time it takes.
828 1-100 Attempt to throttle perf's sample rate to this
829 percentage of CPU. Note: the kernel calculates an
830 "expected" length of each sample event. 100 here means
831 100% of that expected length. Even if this is set to
832 100, you may still see sample throttling if this
833 length is exceeded. Set to 0 if you truly do not care
834 how much CPU is consumed.
835 ===== ========================================================
841 Controls use of the performance events system by unprivileged
842 users (without CAP_PERFMON). The default value is 2.
844 For backward compatibility reasons access to system performance
845 monitoring and observability remains open for CAP_SYS_ADMIN
846 privileged processes but CAP_SYS_ADMIN usage for secure system
847 performance monitoring and observability operations is discouraged
848 with respect to CAP_PERFMON use cases.
850 === ==================================================================
851 -1 Allow use of (almost) all events by all users.
853 Ignore mlock limit after perf_event_mlock_kb without
856 >=0 Disallow ftrace function tracepoint by users without
859 Disallow raw tracepoint access by users without ``CAP_PERFMON``.
861 >=1 Disallow CPU event access by users without ``CAP_PERFMON``.
863 >=2 Disallow kernel profiling by users without ``CAP_PERFMON``.
864 === ==================================================================
870 Controls maximum number of stack frames to copy for (``attr.sample_type &
871 PERF_SAMPLE_CALLCHAIN``) configured events, for instance, when using
872 '``perf record -g``' or '``perf trace --call-graph fp``'.
874 This can only be done when no events are in use that have callchains
875 enabled, otherwise writing to this file will return ``-EBUSY``.
877 The default value is 127.
883 Control size of per-cpu ring buffer not counted against mlock limit.
885 The default value is 512 + 1 page
888 perf_event_max_contexts_per_stack
889 =================================
891 Controls maximum number of stack frame context entries for
892 (``attr.sample_type & PERF_SAMPLE_CALLCHAIN``) configured events, for
893 instance, when using '``perf record -g``' or '``perf trace --call-graph fp``'.
895 This can only be done when no events are in use that have callchains
896 enabled, otherwise writing to this file will return ``-EBUSY``.
898 The default value is 8.
904 PID allocation wrap value. When the kernel's next PID value
905 reaches this value, it wraps back to a minimum PID value.
906 PIDs of value ``pid_max`` or larger are not allocated.
912 The last pid allocated in the current (the one task using this sysctl
913 lives in) pid namespace. When selecting a pid for a next task on fork
914 kernel tries to allocate a number starting from this one.
917 powersave-nap (PPC only)
918 ========================
920 If set, Linux-PPC will use the 'nap' mode of powersaving,
921 otherwise the 'doze' mode will be used.
924 ==============================================================
929 The four values in printk denote: ``console_loglevel``,
930 ``default_message_loglevel``, ``minimum_console_loglevel`` and
931 ``default_console_loglevel`` respectively.
933 These values influence printk() behavior when printing or
934 logging error messages. See '``man 2 syslog``' for more info on
935 the different loglevels.
937 ======================== =====================================
938 console_loglevel messages with a higher priority than
939 this will be printed to the console
940 default_message_loglevel messages without an explicit priority
941 will be printed with this priority
942 minimum_console_loglevel minimum (highest) value to which
943 console_loglevel can be set
944 default_console_loglevel default value for console_loglevel
945 ======================== =====================================
951 Delay each printk message in ``printk_delay`` milliseconds
953 Value from 0 - 10000 is allowed.
959 Some warning messages are rate limited. ``printk_ratelimit`` specifies
960 the minimum length of time between these messages (in seconds).
961 The default value is 5 seconds.
963 A value of 0 will disable rate limiting.
966 printk_ratelimit_burst
967 ======================
969 While long term we enforce one message per `printk_ratelimit`_
970 seconds, we do allow a burst of messages to pass through.
971 ``printk_ratelimit_burst`` specifies the number of messages we can
972 send before ratelimiting kicks in.
974 The default value is 10 messages.
980 Control the logging to ``/dev/kmsg`` from userspace:
982 ========= =============================================
983 ratelimit default, ratelimited
984 on unlimited logging to /dev/kmsg from userspace
985 off logging to /dev/kmsg disabled
986 ========= =============================================
988 The kernel command line parameter ``printk.devkmsg=`` overrides this and is
989 a one-time setting until next reboot: once set, it cannot be changed by
990 this sysctl interface anymore.
992 ==============================================================
998 See Documentation/filesystems/devpts.rst.
1004 This is a directory, with the following entries:
1006 * ``boot_id``: a UUID generated the first time this is retrieved, and
1007 unvarying after that;
1009 * ``entropy_avail``: the pool's entropy count, in bits;
1011 * ``poolsize``: the entropy pool size, in bits;
1013 * ``urandom_min_reseed_secs``: obsolete (used to determine the minimum
1014 number of seconds between urandom pool reseeding).
1016 * ``uuid``: a UUID generated every time this is retrieved (this can
1017 thus be used to generate UUIDs at will);
1019 * ``write_wakeup_threshold``: when the entropy count drops below this
1020 (as a number of bits), processes waiting to write to ``/dev/random``
1023 If ``drivers/char/random.c`` is built with ``ADD_INTERRUPT_BENCH``
1024 defined, these additional entries are present:
1026 * ``add_interrupt_avg_cycles``: the average number of cycles between
1027 interrupts used to feed the pool;
1029 * ``add_interrupt_avg_deviation``: the standard deviation seen on the
1030 number of cycles between interrupts used to feed the pool.
1036 This option can be used to select the type of process address
1037 space randomization that is used in the system, for architectures
1038 that support this feature.
1040 == ===========================================================================
1041 0 Turn the process address space randomization off. This is the
1042 default for architectures that do not support this feature anyways,
1043 and kernels that are booted with the "norandmaps" parameter.
1045 1 Make the addresses of mmap base, stack and VDSO page randomized.
1046 This, among other things, implies that shared libraries will be
1047 loaded to random addresses. Also for PIE-linked binaries, the
1048 location of code start is randomized. This is the default if the
1049 ``CONFIG_COMPAT_BRK`` option is enabled.
1051 2 Additionally enable heap randomization. This is the default if
1052 ``CONFIG_COMPAT_BRK`` is disabled.
1054 There are a few legacy applications out there (such as some ancient
1055 versions of libc.so.5 from 1996) that assume that brk area starts
1056 just after the end of the code+bss. These applications break when
1057 start of the brk area is randomized. There are however no known
1058 non-legacy applications that would be broken this way, so for most
1059 systems it is safe to choose full randomization.
1061 Systems with ancient and/or broken binaries should be configured
1062 with ``CONFIG_COMPAT_BRK`` enabled, which excludes the heap from process
1063 address space randomization.
1064 == ===========================================================================
1070 See :doc:`/admin-guide/initrd`.
1073 reboot-cmd (SPARC only)
1074 =======================
1076 ??? This seems to be a way to give an argument to the Sparc
1077 ROM/Flash boot loader. Maybe to tell it what to do after
1084 Enables/disables Energy Aware Scheduling (EAS). EAS starts
1085 automatically on platforms where it can run (that is,
1086 platforms with asymmetric CPU topologies and having an Energy
1087 Model available). If your platform happens to meet the
1088 requirements for EAS but you do not want to use it, change
1094 Enables/disables task delay accounting (see
1095 :doc:`accounting/delay-accounting.rst`). Enabling this feature incurs
1096 a small amount of overhead in the scheduler but is useful for debugging
1097 and performance tuning. It is required by some tools such as iotop.
1102 Enables/disables scheduler statistics. Enabling this feature
1103 incurs a small amount of overhead in the scheduler but is
1104 useful for debugging and performance tuning.
1106 sched_util_clamp_min
1107 ====================
1109 Max allowed *minimum* utilization.
1111 Default value is 1024, which is the maximum possible value.
1113 It means that any requested uclamp.min value cannot be greater than
1114 sched_util_clamp_min, i.e., it is restricted to the range
1115 [0:sched_util_clamp_min].
1117 sched_util_clamp_max
1118 ====================
1120 Max allowed *maximum* utilization.
1122 Default value is 1024, which is the maximum possible value.
1124 It means that any requested uclamp.max value cannot be greater than
1125 sched_util_clamp_max, i.e., it is restricted to the range
1126 [0:sched_util_clamp_max].
1128 sched_util_clamp_min_rt_default
1129 ===============================
1131 By default Linux is tuned for performance. Which means that RT tasks always run
1132 at the highest frequency and most capable (highest capacity) CPU (in
1133 heterogeneous systems).
1135 Uclamp achieves this by setting the requested uclamp.min of all RT tasks to
1136 1024 by default, which effectively boosts the tasks to run at the highest
1137 frequency and biases them to run on the biggest CPU.
1139 This knob allows admins to change the default behavior when uclamp is being
1140 used. In battery powered devices particularly, running at the maximum
1141 capacity and frequency will increase energy consumption and shorten the battery
1144 This knob is only effective for RT tasks which the user hasn't modified their
1145 requested uclamp.min value via sched_setattr() syscall.
1147 This knob will not escape the range constraint imposed by sched_util_clamp_min
1152 sched_util_clamp_min_rt_default = 800
1153 sched_util_clamp_min = 600
1155 Then the boost will be clamped to 600 because 800 is outside of the permissible
1156 range of [0:600]. This could happen for instance if a powersave mode will
1157 restrict all boosts temporarily by modifying sched_util_clamp_min. As soon as
1158 this restriction is lifted, the requested sched_util_clamp_min_rt_default
1164 See :doc:`/userspace-api/seccomp_filter`.
1170 This file shows the size of the generic SCSI (sg) buffer.
1171 You can't tune it just yet, but you could change it on
1172 compile time by editing ``include/scsi/sg.h`` and changing
1173 the value of ``SG_BIG_BUFF``.
1175 There shouldn't be any reason to change this value. If
1176 you can come up with one, you probably know what you
1183 This parameter sets the total amount of shared memory pages that
1184 can be used system wide. Hence, ``shmall`` should always be at least
1185 ``ceil(shmmax/PAGE_SIZE)``.
1187 If you are not sure what the default ``PAGE_SIZE`` is on your Linux
1188 system, you can run the following command::
1196 This value can be used to query and set the run time limit
1197 on the maximum shared memory segment size that can be created.
1198 Shared memory segments up to 1Gb are now supported in the
1199 kernel. This value defaults to ``SHMMAX``.
1205 This value determines the maximum number of shared memory segments.
1206 4096 by default (``SHMMNI``).
1212 Linux lets you set resource limits, including how much memory one
1213 process can consume, via ``setrlimit(2)``. Unfortunately, shared memory
1214 segments are allowed to exist without association with any process, and
1215 thus might not be counted against any resource limits. If enabled,
1216 shared memory segments are automatically destroyed when their attach
1217 count becomes zero after a detach or a process termination. It will
1218 also destroy segments that were created, but never attached to, on exit
1219 from the process. The only use left for ``IPC_RMID`` is to immediately
1220 destroy an unattached segment. Of course, this breaks the way things are
1221 defined, so some applications might stop working. Note that this
1222 feature will do you no good unless you also configure your resource
1223 limits (in particular, ``RLIMIT_AS`` and ``RLIMIT_NPROC``). Most systems don't
1226 Note that if you change this from 0 to 1, already created segments
1227 without users and with a dead originative process will be destroyed.
1230 sysctl_writes_strict
1231 ====================
1233 Control how file position affects the behavior of updating sysctl values
1234 via the ``/proc/sys`` interface:
1236 == ======================================================================
1237 -1 Legacy per-write sysctl value handling, with no printk warnings.
1238 Each write syscall must fully contain the sysctl value to be
1239 written, and multiple writes on the same sysctl file descriptor
1240 will rewrite the sysctl value, regardless of file position.
1241 0 Same behavior as above, but warn about processes that perform writes
1242 to a sysctl file descriptor when the file position is not 0.
1243 1 (default) Respect file position when writing sysctl strings. Multiple
1244 writes will append to the sysctl value buffer. Anything past the max
1245 length of the sysctl value buffer will be ignored. Writes to numeric
1246 sysctl entries must always be at file position 0 and the value must
1247 be fully contained in the buffer sent in the write syscall.
1248 == ======================================================================
1251 softlockup_all_cpu_backtrace
1252 ============================
1254 This value controls the soft lockup detector thread's behavior
1255 when a soft lockup condition is detected as to whether or not
1256 to gather further debug information. If enabled, each cpu will
1257 be issued an NMI and instructed to capture stack trace.
1259 This feature is only applicable for architectures which support
1262 = ============================================
1263 0 Do nothing. This is the default behavior.
1264 1 On detection capture more debug information.
1265 = ============================================
1271 This parameter can be used to control whether the kernel panics
1272 when a soft lockup is detected.
1274 = ============================================
1275 0 Don't panic on soft lockup.
1276 1 Panic on soft lockup.
1277 = ============================================
1279 This can also be set using the softlockup_panic kernel parameter.
1285 This parameter can be used to control the soft lockup detector.
1287 = =================================
1288 0 Disable the soft lockup detector.
1289 1 Enable the soft lockup detector.
1290 = =================================
1292 The soft lockup detector monitors CPUs for threads that are hogging the CPUs
1293 without rescheduling voluntarily, and thus prevent the 'watchdog/N' threads
1294 from running. The mechanism depends on the CPUs ability to respond to timer
1295 interrupts which are needed for the 'watchdog/N' threads to be woken up by
1296 the watchdog timer function, otherwise the NMI watchdog — if enabled — can
1297 detect a hard lockup condition.
1303 This parameter can be used to control kernel stack erasing at the end
1304 of syscalls for kernels built with ``CONFIG_GCC_PLUGIN_STACKLEAK``.
1306 That erasing reduces the information which kernel stack leak bugs
1307 can reveal and blocks some uninitialized stack variable attacks.
1308 The tradeoff is the performance impact: on a single CPU system kernel
1309 compilation sees a 1% slowdown, other systems and workloads may vary.
1311 = ====================================================================
1312 0 Kernel stack erasing is disabled, STACKLEAK_METRICS are not updated.
1313 1 Kernel stack erasing is enabled (default), it is performed before
1314 returning to the userspace at the end of syscalls.
1315 = ====================================================================
1323 = ====================================
1324 0 Stop-A has no effect.
1325 1 Stop-A breaks to the PROM (default).
1326 = ====================================
1328 Stop-A is always enabled on a panic, so that the user can return to
1335 See :doc:`/admin-guide/sysrq`.
1341 Non-zero if the kernel has been tainted. Numeric values, which can be
1342 ORed together. The letters are seen in "Tainted" line of Oops reports.
1344 ====== ===== ==============================================================
1345 1 `(P)` proprietary module was loaded
1346 2 `(F)` module was force loaded
1347 4 `(S)` kernel running on an out of specification system
1348 8 `(R)` module was force unloaded
1349 16 `(M)` processor reported a Machine Check Exception (MCE)
1350 32 `(B)` bad page referenced or some unexpected page flags
1351 64 `(U)` taint requested by userspace application
1352 128 `(D)` kernel died recently, i.e. there was an OOPS or BUG
1353 256 `(A)` an ACPI table was overridden by user
1354 512 `(W)` kernel issued warning
1355 1024 `(C)` staging driver was loaded
1356 2048 `(I)` workaround for bug in platform firmware applied
1357 4096 `(O)` externally-built ("out-of-tree") module was loaded
1358 8192 `(E)` unsigned module was loaded
1359 16384 `(L)` soft lockup occurred
1360 32768 `(K)` kernel has been live patched
1361 65536 `(X)` Auxiliary taint, defined and used by for distros
1362 131072 `(T)` The kernel was built with the struct randomization plugin
1363 ====== ===== ==============================================================
1365 See :doc:`/admin-guide/tainted-kernels` for more information.
1368 writes to this sysctl interface will fail with ``EINVAL`` if the kernel is
1369 booted with the command line option ``panic_on_taint=<bitmask>,nousertaint``
1370 and any of the ORed together values being written to ``tainted`` match with
1371 the bitmask declared on panic_on_taint.
1372 See :doc:`/admin-guide/kernel-parameters` for more details on that particular
1373 kernel command line option and its optional ``nousertaint`` switch.
1378 This value controls the maximum number of threads that can be created
1381 During initialization the kernel sets this value such that even if the
1382 maximum number of threads is created, the thread structures occupy only
1383 a part (1/8th) of the available RAM pages.
1385 The minimum value that can be written to ``threads-max`` is 1.
1387 The maximum value that can be written to ``threads-max`` is given by the
1388 constant ``FUTEX_TID_MASK`` (0x3fffffff).
1390 If a value outside of this range is written to ``threads-max`` an
1391 ``EINVAL`` error occurs.
1397 When set, disables tracing (see :doc:`/trace/ftrace`) when a
1404 When tracepoints are sent to printk() (enabled by the ``tp_printk``
1405 boot parameter), this entry provides runtime control::
1407 echo 0 > /proc/sys/kernel/tracepoint_printk
1409 will stop tracepoints from being sent to printk(), and::
1411 echo 1 > /proc/sys/kernel/tracepoint_printk
1413 will send them to printk() again.
1415 This only works if the kernel was booted with ``tp_printk`` enabled.
1417 See :doc:`/admin-guide/kernel-parameters` and
1418 :doc:`/trace/boottime-trace`.
1421 .. _unaligned-dump-stack:
1423 unaligned-dump-stack (ia64)
1424 ===========================
1426 When logging unaligned accesses, controls whether the stack is
1429 = ===================================================
1430 0 Do not dump the stack. This is the default setting.
1432 = ===================================================
1434 See also `ignore-unaligned-usertrap`_.
1440 On architectures where unaligned accesses cause traps, and where this
1441 feature is supported (``CONFIG_SYSCTL_ARCH_UNALIGN_ALLOW``; currently,
1442 ``arc`` and ``parisc``), controls whether unaligned traps are caught
1443 and emulated (instead of failing).
1445 = ========================================================
1446 0 Do not emulate unaligned accesses.
1447 1 Emulate unaligned accesses. This is the default setting.
1448 = ========================================================
1450 See also `ignore-unaligned-usertrap`_.
1456 The value in this file affects behavior of handling NMI. When the
1457 value is non-zero, unknown NMI is trapped and then panic occurs. At
1458 that time, kernel debugging information is displayed on console.
1460 NMI switch that most IA32 servers have fires unknown NMI up, for
1461 example. If a system hangs up, try pressing the NMI switch.
1464 unprivileged_bpf_disabled
1465 =========================
1467 Writing 1 to this entry will disable unprivileged calls to ``bpf()``;
1468 once disabled, calling ``bpf()`` without ``CAP_SYS_ADMIN`` or ``CAP_BPF``
1469 will return ``-EPERM``. Once set to 1, this can't be cleared from the
1470 running kernel anymore.
1472 Writing 2 to this entry will also disable unprivileged calls to ``bpf()``,
1473 however, an admin can still change this setting later on, if needed, by
1474 writing 0 or 1 to this entry.
1476 If ``BPF_UNPRIV_DEFAULT_OFF`` is enabled in the kernel config, then this
1477 entry will default to 2 instead of 0.
1479 = =============================================================
1480 0 Unprivileged calls to ``bpf()`` are enabled
1481 1 Unprivileged calls to ``bpf()`` are disabled without recovery
1482 2 Unprivileged calls to ``bpf()`` are disabled
1483 = =============================================================
1488 This parameter can be used to disable or enable the soft lockup detector
1489 *and* the NMI watchdog (i.e. the hard lockup detector) at the same time.
1491 = ==============================
1492 0 Disable both lockup detectors.
1493 1 Enable both lockup detectors.
1494 = ==============================
1496 The soft lockup detector and the NMI watchdog can also be disabled or
1497 enabled individually, using the ``soft_watchdog`` and ``nmi_watchdog``
1499 If the ``watchdog`` parameter is read, for example by executing::
1501 cat /proc/sys/kernel/watchdog
1503 the output of this command (0 or 1) shows the logical OR of
1504 ``soft_watchdog`` and ``nmi_watchdog``.
1510 This value can be used to control on which cpus the watchdog may run.
1511 The default cpumask is all possible cores, but if ``NO_HZ_FULL`` is
1512 enabled in the kernel config, and cores are specified with the
1513 ``nohz_full=`` boot argument, those cores are excluded by default.
1514 Offline cores can be included in this mask, and if the core is later
1515 brought online, the watchdog will be started based on the mask value.
1517 Typically this value would only be touched in the ``nohz_full`` case
1518 to re-enable cores that by default were not running the watchdog,
1519 if a kernel lockup was suspected on those cores.
1521 The argument value is the standard cpulist format for cpumasks,
1522 so for example to enable the watchdog on cores 0, 2, 3, and 4 you
1525 echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask
1531 This value can be used to control the frequency of hrtimer and NMI
1532 events and the soft and hard lockup thresholds. The default threshold
1535 The softlockup threshold is (``2 * watchdog_thresh``). Setting this
1536 tunable to zero will disable lockup detection altogether.