1 Documentation 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>
5 For general info and legal blurb, please look in README.
7 ==============================================================
9 This file contains documentation for the sysctl files in
10 /proc/sys/kernel/ and is valid for Linux kernel version 2.2.
12 The files in this directory can be used to tune and monitor
13 miscellaneous and general things in the operation of the Linux
14 kernel. Since some of the files _can_ be used to screw up your
15 system, it is advisable to read both documentation and source
16 before actually making adjustments.
18 Currently, these files might (depending on your configuration)
19 show up in /proc/sys/kernel:
24 - bootloader_type [ X86 only ]
25 - bootloader_version [ X86 only ]
26 - callhome [ S390 only ]
36 - hardlockup_all_cpu_backtrace
38 - hung_task_check_count
39 - hung_task_timeout_secs
43 - kstack_depth_to_print [ X86 only ]
45 - modprobe ==> Documentation/debugging-modules.txt
47 - msg_next_id [ sysv ipc ]
58 - panic_on_stackoverflow
59 - panic_on_unrecovered_nmi
62 - powersave-nap [ PPC only ]
66 - printk_ratelimit_burst
68 - real-root-dev ==> Documentation/initrd.txt
69 - reboot-cmd [ SPARC only ]
73 - sem_next_id [ sysv ipc ]
74 - sg-big-buff [ generic SCSI device (sg) ]
75 - shm_next_id [ sysv ipc ]
80 - softlockup_all_cpu_backtrace
82 - stop-a [ SPARC only ]
83 - sysrq ==> Documentation/sysrq.txt
84 - sysctl_writes_strict
92 ==============================================================
96 highwater lowwater frequency
98 If BSD-style process accounting is enabled these values control
99 its behaviour. If free space on filesystem where the log lives
100 goes below <lowwater>% accounting suspends. If free space gets
101 above <highwater>% accounting resumes. <Frequency> determines
102 how often do we check the amount of free space (value is in
105 That is, suspend accounting if there left <= 2% free; resume it
106 if we got >=4%; consider information about amount of free space
107 valid for 30 seconds.
109 ==============================================================
115 See Doc*/kernel/power/video.txt, it allows mode of video boot to be
118 ==============================================================
122 This variable has no effect and may be removed in future kernel
123 releases. Reading it always returns 0.
124 Up to Linux 3.17, it enabled/disabled automatic recomputing of msgmni
125 upon memory add/remove or upon ipc namespace creation/removal.
126 Echoing "1" into this file enabled msgmni automatic recomputing.
127 Echoing "0" turned it off. auto_msgmni default value was 1.
130 ==============================================================
134 x86 bootloader identification
136 This gives the bootloader type number as indicated by the bootloader,
137 shifted left by 4, and OR'd with the low four bits of the bootloader
138 version. The reason for this encoding is that this used to match the
139 type_of_loader field in the kernel header; the encoding is kept for
140 backwards compatibility. That is, if the full bootloader type number
141 is 0x15 and the full version number is 0x234, this file will contain
142 the value 340 = 0x154.
144 See the type_of_loader and ext_loader_type fields in
145 Documentation/x86/boot.txt for additional information.
147 ==============================================================
151 x86 bootloader version
153 The complete bootloader version number. In the example above, this
154 file will contain the value 564 = 0x234.
156 See the type_of_loader and ext_loader_ver fields in
157 Documentation/x86/boot.txt for additional information.
159 ==============================================================
163 Controls the kernel's callhome behavior in case of a kernel panic.
165 The s390 hardware allows an operating system to send a notification
166 to a service organization (callhome) in case of an operating system panic.
168 When the value in this file is 0 (which is the default behavior)
169 nothing happens in case of a kernel panic. If this value is set to "1"
170 the complete kernel oops message is send to the IBM customer service
171 organization in case the mainframe the Linux operating system is running
172 on has a service contract with IBM.
174 ==============================================================
178 Highest valid capability of the running kernel. Exports
179 CAP_LAST_CAP from the kernel.
181 ==============================================================
185 core_pattern is used to specify a core dumpfile pattern name.
186 . max length 128 characters; default value is "core"
187 . core_pattern is used as a pattern template for the output filename;
188 certain string patterns (beginning with '%') are substituted with
190 . backward compatibility with core_uses_pid:
191 If core_pattern does not include "%p" (default does not)
192 and core_uses_pid is set, then .PID will be appended to
194 . corename format specifiers:
195 %<NUL> '%' is dropped
198 %P global pid (init PID namespace)
200 %I global tid (init PID namespace)
201 %u uid (in initial user namespace)
202 %g gid (in initial user namespace)
203 %d dump mode, matches PR_SET_DUMPABLE and
204 /proc/sys/fs/suid_dumpable
208 %e executable filename (may be shortened)
210 %<OTHER> both are dropped
211 . If the first character of the pattern is a '|', the kernel will treat
212 the rest of the pattern as a command to run. The core dump will be
213 written to the standard input of that program instead of to a file.
215 ==============================================================
219 This sysctl is only applicable when core_pattern is configured to pipe
220 core files to a user space helper (when the first character of
221 core_pattern is a '|', see above). When collecting cores via a pipe
222 to an application, it is occasionally useful for the collecting
223 application to gather data about the crashing process from its
224 /proc/pid directory. In order to do this safely, the kernel must wait
225 for the collecting process to exit, so as not to remove the crashing
226 processes proc files prematurely. This in turn creates the
227 possibility that a misbehaving userspace collecting process can block
228 the reaping of a crashed process simply by never exiting. This sysctl
229 defends against that. It defines how many concurrent crashing
230 processes may be piped to user space applications in parallel. If
231 this value is exceeded, then those crashing processes above that value
232 are noted via the kernel log and their cores are skipped. 0 is a
233 special value, indicating that unlimited processes may be captured in
234 parallel, but that no waiting will take place (i.e. the collecting
235 process is not guaranteed access to /proc/<crashing pid>/). This
238 ==============================================================
242 The default coredump filename is "core". By setting
243 core_uses_pid to 1, the coredump filename becomes core.PID.
244 If core_pattern does not include "%p" (default does not)
245 and core_uses_pid is set, then .PID will be appended to
248 ==============================================================
252 When the value in this file is 0, ctrl-alt-del is trapped and
253 sent to the init(1) program to handle a graceful restart.
254 When, however, the value is > 0, Linux's reaction to a Vulcan
255 Nerve Pinch (tm) will be an immediate reboot, without even
256 syncing its dirty buffers.
258 Note: when a program (like dosemu) has the keyboard in 'raw'
259 mode, the ctrl-alt-del is intercepted by the program before it
260 ever reaches the kernel tty layer, and it's up to the program
261 to decide what to do with it.
263 ==============================================================
267 This toggle indicates whether unprivileged users are prevented
268 from using dmesg(8) to view messages from the kernel's log buffer.
269 When dmesg_restrict is set to (0) there are no restrictions. When
270 dmesg_restrict is set set to (1), users must have CAP_SYSLOG to use
273 The kernel config option CONFIG_SECURITY_DMESG_RESTRICT sets the
274 default value of dmesg_restrict.
276 ==============================================================
278 domainname & hostname:
280 These files can be used to set the NIS/YP domainname and the
281 hostname of your box in exactly the same way as the commands
282 domainname and hostname, i.e.:
283 # echo "darkstar" > /proc/sys/kernel/hostname
284 # echo "mydomain" > /proc/sys/kernel/domainname
285 has the same effect as
286 # hostname "darkstar"
287 # domainname "mydomain"
289 Note, however, that the classic darkstar.frop.org has the
290 hostname "darkstar" and DNS (Internet Domain Name Server)
291 domainname "frop.org", not to be confused with the NIS (Network
292 Information Service) or YP (Yellow Pages) domainname. These two
293 domain names are in general different. For a detailed discussion
294 see the hostname(1) man page.
296 ==============================================================
297 hardlockup_all_cpu_backtrace:
299 This value controls the hard lockup detector behavior when a hard
300 lockup condition is detected as to whether or not to gather further
301 debug information. If enabled, arch-specific all-CPU stack dumping
304 0: do nothing. This is the default behavior.
306 1: on detection capture more debug information.
307 ==============================================================
311 Path for the hotplug policy agent.
312 Default value is "/sbin/hotplug".
314 ==============================================================
318 Controls the kernel's behavior when a hung task is detected.
319 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
321 0: continue operation. This is the default behavior.
323 1: panic immediately.
325 ==============================================================
327 hung_task_check_count:
329 The upper bound on the number of tasks that are checked.
330 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
332 ==============================================================
334 hung_task_timeout_secs:
336 Check interval. When a task in D state did not get scheduled
337 for more than this value report a warning.
338 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
340 0: means infinite timeout - no checking done.
341 Possible values to set are in range {0..LONG_MAX/HZ}.
343 ==============================================================
347 The maximum number of warnings to report. During a check interval
348 if a hung task is detected, this value is decreased by 1.
349 When this value reaches 0, no more warnings will be reported.
350 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
352 -1: report an infinite number of warnings.
354 ==============================================================
358 A toggle indicating if the kexec_load syscall has been disabled. This
359 value defaults to 0 (false: kexec_load enabled), but can be set to 1
360 (true: kexec_load disabled). Once true, kexec can no longer be used, and
361 the toggle cannot be set back to false. This allows a kexec image to be
362 loaded before disabling the syscall, allowing a system to set up (and
363 later use) an image without it being altered. Generally used together
364 with the "modules_disabled" sysctl.
366 ==============================================================
370 This toggle indicates whether restrictions are placed on
371 exposing kernel addresses via /proc and other interfaces.
373 When kptr_restrict is set to (0), the default, there are no restrictions.
375 When kptr_restrict is set to (1), kernel pointers printed using the %pK
376 format specifier will be replaced with 0's unless the user has CAP_SYSLOG
377 and effective user and group ids are equal to the real ids. This is
378 because %pK checks are done at read() time rather than open() time, so
379 if permissions are elevated between the open() and the read() (e.g via
380 a setuid binary) then %pK will not leak kernel pointers to unprivileged
381 users. Note, this is a temporary solution only. The correct long-term
382 solution is to do the permission checks at open() time. Consider removing
383 world read permissions from files that use %pK, and using dmesg_restrict
384 to protect against uses of %pK in dmesg(8) if leaking kernel pointer
385 values to unprivileged users is a concern.
387 When kptr_restrict is set to (2), kernel pointers printed using
388 %pK will be replaced with 0's regardless of privileges.
390 ==============================================================
392 kstack_depth_to_print: (X86 only)
394 Controls the number of words to print when dumping the raw
397 ==============================================================
401 This flag controls the L2 cache of G3 processor boards. If
402 0, the cache is disabled. Enabled if nonzero.
404 ==============================================================
408 A toggle value indicating if modules are allowed to be loaded
409 in an otherwise modular kernel. This toggle defaults to off
410 (0), but can be set true (1). Once true, modules can be
411 neither loaded nor unloaded, and the toggle cannot be set back
412 to false. Generally used with the "kexec_load_disabled" toggle.
414 ==============================================================
416 msg_next_id, sem_next_id, and shm_next_id:
418 These three toggles allows to specify desired id for next allocated IPC
419 object: message, semaphore or shared memory respectively.
421 By default they are equal to -1, which means generic allocation logic.
422 Possible values to set are in range {0..INT_MAX}.
425 1) kernel doesn't guarantee, that new object will have desired id. So,
426 it's up to userspace, how to handle an object with "wrong" id.
427 2) Toggle with non-default value will be set back to -1 by kernel after
428 successful IPC object allocation.
430 ==============================================================
434 This parameter can be used to control the NMI watchdog
435 (i.e. the hard lockup detector) on x86 systems.
437 0 - disable the hard lockup detector
438 1 - enable the hard lockup detector
440 The hard lockup detector monitors each CPU for its ability to respond to
441 timer interrupts. The mechanism utilizes CPU performance counter registers
442 that are programmed to generate Non-Maskable Interrupts (NMIs) periodically
443 while a CPU is busy. Hence, the alternative name 'NMI watchdog'.
445 The NMI watchdog is disabled by default if the kernel is running as a guest
446 in a KVM virtual machine. This default can be overridden by adding
450 to the guest kernel command line (see Documentation/kernel-parameters.txt).
452 ==============================================================
456 Enables/disables automatic page fault based NUMA memory
457 balancing. Memory is moved automatically to nodes
458 that access it often.
460 Enables/disables automatic NUMA memory balancing. On NUMA machines, there
461 is a performance penalty if remote memory is accessed by a CPU. When this
462 feature is enabled the kernel samples what task thread is accessing memory
463 by periodically unmapping pages and later trapping a page fault. At the
464 time of the page fault, it is determined if the data being accessed should
465 be migrated to a local memory node.
467 The unmapping of pages and trapping faults incur additional overhead that
468 ideally is offset by improved memory locality but there is no universal
469 guarantee. If the target workload is already bound to NUMA nodes then this
470 feature should be disabled. Otherwise, if the system overhead from the
471 feature is too high then the rate the kernel samples for NUMA hinting
472 faults may be controlled by the numa_balancing_scan_period_min_ms,
473 numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms,
474 numa_balancing_scan_size_mb, and numa_balancing_settle_count sysctls.
476 ==============================================================
478 numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms,
479 numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb
481 Automatic NUMA balancing scans tasks address space and unmaps pages to
482 detect if pages are properly placed or if the data should be migrated to a
483 memory node local to where the task is running. Every "scan delay" the task
484 scans the next "scan size" number of pages in its address space. When the
485 end of the address space is reached the scanner restarts from the beginning.
487 In combination, the "scan delay" and "scan size" determine the scan rate.
488 When "scan delay" decreases, the scan rate increases. The scan delay and
489 hence the scan rate of every task is adaptive and depends on historical
490 behaviour. If pages are properly placed then the scan delay increases,
491 otherwise the scan delay decreases. The "scan size" is not adaptive but
492 the higher the "scan size", the higher the scan rate.
494 Higher scan rates incur higher system overhead as page faults must be
495 trapped and potentially data must be migrated. However, the higher the scan
496 rate, the more quickly a tasks memory is migrated to a local node if the
497 workload pattern changes and minimises performance impact due to remote
498 memory accesses. These sysctls control the thresholds for scan delays and
499 the number of pages scanned.
501 numa_balancing_scan_period_min_ms is the minimum time in milliseconds to
502 scan a tasks virtual memory. It effectively controls the maximum scanning
505 numa_balancing_scan_delay_ms is the starting "scan delay" used for a task
506 when it initially forks.
508 numa_balancing_scan_period_max_ms is the maximum time in milliseconds to
509 scan a tasks virtual memory. It effectively controls the minimum scanning
512 numa_balancing_scan_size_mb is how many megabytes worth of pages are
513 scanned for a given scan.
515 ==============================================================
517 osrelease, ostype & version:
524 #5 Wed Feb 25 21:49:24 MET 1998
526 The files osrelease and ostype should be clear enough. Version
527 needs a little more clarification however. The '#5' means that
528 this is the fifth kernel built from this source base and the
529 date behind it indicates the time the kernel was built.
530 The only way to tune these values is to rebuild the kernel :-)
532 ==============================================================
534 overflowgid & overflowuid:
536 if your architecture did not always support 32-bit UIDs (i.e. arm,
537 i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to
538 applications that use the old 16-bit UID/GID system calls, if the
539 actual UID or GID would exceed 65535.
541 These sysctls allow you to change the value of the fixed UID and GID.
542 The default is 65534.
544 ==============================================================
548 The value in this file represents the number of seconds the kernel
549 waits before rebooting on a panic. When you use the software watchdog,
550 the recommended setting is 60.
552 ==============================================================
556 Controls the kernel's behaviour when an oops or BUG is encountered.
558 0: try to continue operation
560 1: panic immediately. If the `panic' sysctl is also non-zero then the
561 machine will be rebooted.
563 ==============================================================
565 panic_on_stackoverflow:
567 Controls the kernel's behavior when detecting the overflows of
568 kernel, IRQ and exception stacks except a user stack.
569 This file shows up if CONFIG_DEBUG_STACKOVERFLOW is enabled.
571 0: try to continue operation.
573 1: panic immediately.
575 ==============================================================
577 panic_on_unrecovered_nmi:
579 The default Linux behaviour on an NMI of either memory or unknown is
580 to continue operation. For many environments such as scientific
581 computing it is preferable that the box is taken out and the error
582 dealt with than an uncorrected parity/ECC error get propagated.
584 A small number of systems do generate NMI's for bizarre random reasons
585 such as power management so the default is off. That sysctl works like
586 the existing panic controls already in that directory.
588 ==============================================================
592 Calls panic() in the WARN() path when set to 1. This is useful to avoid
593 a kernel rebuild when attempting to kdump at the location of a WARN().
595 0: only WARN(), default behaviour.
597 1: call panic() after printing out WARN() location.
599 ==============================================================
601 perf_cpu_time_max_percent:
603 Hints to the kernel how much CPU time it should be allowed to
604 use to handle perf sampling events. If the perf subsystem
605 is informed that its samples are exceeding this limit, it
606 will drop its sampling frequency to attempt to reduce its CPU
609 Some perf sampling happens in NMIs. If these samples
610 unexpectedly take too long to execute, the NMIs can become
611 stacked up next to each other so much that nothing else is
614 0: disable the mechanism. Do not monitor or correct perf's
615 sampling rate no matter how CPU time it takes.
617 1-100: attempt to throttle perf's sample rate to this
618 percentage of CPU. Note: the kernel calculates an
619 "expected" length of each sample event. 100 here means
620 100% of that expected length. Even if this is set to
621 100, you may still see sample throttling if this
622 length is exceeded. Set to 0 if you truly do not care
623 how much CPU is consumed.
625 ==============================================================
630 PID allocation wrap value. When the kernel's next PID value
631 reaches this value, it wraps back to a minimum PID value.
632 PIDs of value pid_max or larger are not allocated.
634 ==============================================================
638 The last pid allocated in the current (the one task using this sysctl
639 lives in) pid namespace. When selecting a pid for a next task on fork
640 kernel tries to allocate a number starting from this one.
642 ==============================================================
644 powersave-nap: (PPC only)
646 If set, Linux-PPC will use the 'nap' mode of powersaving,
647 otherwise the 'doze' mode will be used.
649 ==============================================================
653 The four values in printk denote: console_loglevel,
654 default_message_loglevel, minimum_console_loglevel and
655 default_console_loglevel respectively.
657 These values influence printk() behavior when printing or
658 logging error messages. See 'man 2 syslog' for more info on
659 the different loglevels.
661 - console_loglevel: messages with a higher priority than
662 this will be printed to the console
663 - default_message_loglevel: messages without an explicit priority
664 will be printed with this priority
665 - minimum_console_loglevel: minimum (highest) value to which
666 console_loglevel can be set
667 - default_console_loglevel: default value for console_loglevel
669 ==============================================================
673 Delay each printk message in printk_delay milliseconds
675 Value from 0 - 10000 is allowed.
677 ==============================================================
681 Some warning messages are rate limited. printk_ratelimit specifies
682 the minimum length of time between these messages (in jiffies), by
683 default we allow one every 5 seconds.
685 A value of 0 will disable rate limiting.
687 ==============================================================
689 printk_ratelimit_burst:
691 While long term we enforce one message per printk_ratelimit
692 seconds, we do allow a burst of messages to pass through.
693 printk_ratelimit_burst specifies the number of messages we can
694 send before ratelimiting kicks in.
696 ==============================================================
700 This option can be used to select the type of process address
701 space randomization that is used in the system, for architectures
702 that support this feature.
704 0 - Turn the process address space randomization off. This is the
705 default for architectures that do not support this feature anyways,
706 and kernels that are booted with the "norandmaps" parameter.
708 1 - Make the addresses of mmap base, stack and VDSO page randomized.
709 This, among other things, implies that shared libraries will be
710 loaded to random addresses. Also for PIE-linked binaries, the
711 location of code start is randomized. This is the default if the
712 CONFIG_COMPAT_BRK option is enabled.
714 2 - Additionally enable heap randomization. This is the default if
715 CONFIG_COMPAT_BRK is disabled.
717 There are a few legacy applications out there (such as some ancient
718 versions of libc.so.5 from 1996) that assume that brk area starts
719 just after the end of the code+bss. These applications break when
720 start of the brk area is randomized. There are however no known
721 non-legacy applications that would be broken this way, so for most
722 systems it is safe to choose full randomization.
724 Systems with ancient and/or broken binaries should be configured
725 with CONFIG_COMPAT_BRK enabled, which excludes the heap from process
726 address space randomization.
728 ==============================================================
730 reboot-cmd: (Sparc only)
732 ??? This seems to be a way to give an argument to the Sparc
733 ROM/Flash boot loader. Maybe to tell it what to do after
736 ==============================================================
738 rtsig-max & rtsig-nr:
740 The file rtsig-max can be used to tune the maximum number
741 of POSIX realtime (queued) signals that can be outstanding
744 rtsig-nr shows the number of RT signals currently queued.
746 ==============================================================
750 This file shows the size of the generic SCSI (sg) buffer.
751 You can't tune it just yet, but you could change it on
752 compile time by editing include/scsi/sg.h and changing
753 the value of SG_BIG_BUFF.
755 There shouldn't be any reason to change this value. If
756 you can come up with one, you probably know what you
759 ==============================================================
763 This parameter sets the total amount of shared memory pages that
764 can be used system wide. Hence, SHMALL should always be at least
765 ceil(shmmax/PAGE_SIZE).
767 If you are not sure what the default PAGE_SIZE is on your Linux
768 system, you can run the following command:
772 ==============================================================
776 This value can be used to query and set the run time limit
777 on the maximum shared memory segment size that can be created.
778 Shared memory segments up to 1Gb are now supported in the
779 kernel. This value defaults to SHMMAX.
781 ==============================================================
785 Linux lets you set resource limits, including how much memory one
786 process can consume, via setrlimit(2). Unfortunately, shared memory
787 segments are allowed to exist without association with any process, and
788 thus might not be counted against any resource limits. If enabled,
789 shared memory segments are automatically destroyed when their attach
790 count becomes zero after a detach or a process termination. It will
791 also destroy segments that were created, but never attached to, on exit
792 from the process. The only use left for IPC_RMID is to immediately
793 destroy an unattached segment. Of course, this breaks the way things are
794 defined, so some applications might stop working. Note that this
795 feature will do you no good unless you also configure your resource
796 limits (in particular, RLIMIT_AS and RLIMIT_NPROC). Most systems don't
799 Note that if you change this from 0 to 1, already created segments
800 without users and with a dead originative process will be destroyed.
802 ==============================================================
804 sysctl_writes_strict:
806 Control how file position affects the behavior of updating sysctl values
807 via the /proc/sys interface:
809 -1 - Legacy per-write sysctl value handling, with no printk warnings.
810 Each write syscall must fully contain the sysctl value to be
811 written, and multiple writes on the same sysctl file descriptor
812 will rewrite the sysctl value, regardless of file position.
813 0 - (default) Same behavior as above, but warn about processes that
814 perform writes to a sysctl file descriptor when the file position
816 1 - Respect file position when writing sysctl strings. Multiple writes
817 will append to the sysctl value buffer. Anything past the max length
818 of the sysctl value buffer will be ignored. Writes to numeric sysctl
819 entries must always be at file position 0 and the value must be
820 fully contained in the buffer sent in the write syscall.
822 ==============================================================
824 softlockup_all_cpu_backtrace:
826 This value controls the soft lockup detector thread's behavior
827 when a soft lockup condition is detected as to whether or not
828 to gather further debug information. If enabled, each cpu will
829 be issued an NMI and instructed to capture stack trace.
831 This feature is only applicable for architectures which support
834 0: do nothing. This is the default behavior.
836 1: on detection capture more debug information.
838 ==============================================================
842 This parameter can be used to control the soft lockup detector.
844 0 - disable the soft lockup detector
845 1 - enable the soft lockup detector
847 The soft lockup detector monitors CPUs for threads that are hogging the CPUs
848 without rescheduling voluntarily, and thus prevent the 'watchdog/N' threads
849 from running. The mechanism depends on the CPUs ability to respond to timer
850 interrupts which are needed for the 'watchdog/N' threads to be woken up by
851 the watchdog timer function, otherwise the NMI watchdog - if enabled - can
852 detect a hard lockup condition.
854 ==============================================================
858 Non-zero if the kernel has been tainted. Numeric values, which
859 can be ORed together:
861 1 - A module with a non-GPL license has been loaded, this
862 includes modules with no license.
863 Set by modutils >= 2.4.9 and module-init-tools.
864 2 - A module was force loaded by insmod -f.
865 Set by modutils >= 2.4.9 and module-init-tools.
866 4 - Unsafe SMP processors: SMP with CPUs not designed for SMP.
867 8 - A module was forcibly unloaded from the system by rmmod -f.
868 16 - A hardware machine check error occurred on the system.
869 32 - A bad page was discovered on the system.
870 64 - The user has asked that the system be marked "tainted". This
871 could be because they are running software that directly modifies
872 the hardware, or for other reasons.
873 128 - The system has died.
874 256 - The ACPI DSDT has been overridden with one supplied by the user
875 instead of using the one provided by the hardware.
876 512 - A kernel warning has occurred.
877 1024 - A module from drivers/staging was loaded.
878 2048 - The system is working around a severe firmware bug.
879 4096 - An out-of-tree module has been loaded.
880 8192 - An unsigned module has been loaded in a kernel supporting module
882 16384 - A soft lockup has previously occurred on the system.
883 32768 - The kernel has been live patched.
885 ==============================================================
889 This value controls the maximum number of threads that can be created
892 During initialization the kernel sets this value such that even if the
893 maximum number of threads is created, the thread structures occupy only
894 a part (1/8th) of the available RAM pages.
896 The minimum value that can be written to threads-max is 20.
897 The maximum value that can be written to threads-max is given by the
898 constant FUTEX_TID_MASK (0x3fffffff).
899 If a value outside of this range is written to threads-max an error
902 The value written is checked against the available RAM pages. If the
903 thread structures would occupy too much (more than 1/8th) of the
904 available RAM pages threads-max is reduced accordingly.
906 ==============================================================
910 The value in this file affects behavior of handling NMI. When the
911 value is non-zero, unknown NMI is trapped and then panic occurs. At
912 that time, kernel debugging information is displayed on console.
914 NMI switch that most IA32 servers have fires unknown NMI up, for
915 example. If a system hangs up, try pressing the NMI switch.
917 ==============================================================
921 This parameter can be used to disable or enable the soft lockup detector
922 _and_ the NMI watchdog (i.e. the hard lockup detector) at the same time.
924 0 - disable both lockup detectors
925 1 - enable both lockup detectors
927 The soft lockup detector and the NMI watchdog can also be disabled or
928 enabled individually, using the soft_watchdog and nmi_watchdog parameters.
929 If the watchdog parameter is read, for example by executing
931 cat /proc/sys/kernel/watchdog
933 the output of this command (0 or 1) shows the logical OR of soft_watchdog
936 ==============================================================
940 This value can be used to control on which cpus the watchdog may run.
941 The default cpumask is all possible cores, but if NO_HZ_FULL is
942 enabled in the kernel config, and cores are specified with the
943 nohz_full= boot argument, those cores are excluded by default.
944 Offline cores can be included in this mask, and if the core is later
945 brought online, the watchdog will be started based on the mask value.
947 Typically this value would only be touched in the nohz_full case
948 to re-enable cores that by default were not running the watchdog,
949 if a kernel lockup was suspected on those cores.
951 The argument value is the standard cpulist format for cpumasks,
952 so for example to enable the watchdog on cores 0, 2, 3, and 4 you
955 echo 0,2-4 > /proc/sys/kernel/watchdog_cpumask
957 ==============================================================
961 This value can be used to control the frequency of hrtimer and NMI
962 events and the soft and hard lockup thresholds. The default threshold
965 The softlockup threshold is (2 * watchdog_thresh). Setting this
966 tunable to zero will disable lockup detection altogether.
968 ==============================================================