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 ]
37 - hung_task_check_count
38 - hung_task_timeout_secs
42 - kstack_depth_to_print [ X86 only ]
44 - modprobe ==> Documentation/debugging-modules.txt
46 - msg_next_id [ sysv ipc ]
57 - panic_on_stackoverflow
58 - panic_on_unrecovered_nmi
61 - powersave-nap [ PPC only ]
65 - printk_ratelimit_burst
67 - real-root-dev ==> Documentation/initrd.txt
68 - reboot-cmd [ SPARC only ]
72 - sem_next_id [ sysv ipc ]
73 - sg-big-buff [ generic SCSI device (sg) ]
74 - shm_next_id [ sysv ipc ]
79 - softlockup_all_cpu_backtrace
80 - stop-a [ SPARC only ]
81 - sysrq ==> Documentation/sysrq.txt
82 - sysctl_writes_strict
89 ==============================================================
93 highwater lowwater frequency
95 If BSD-style process accounting is enabled these values control
96 its behaviour. If free space on filesystem where the log lives
97 goes below <lowwater>% accounting suspends. If free space gets
98 above <highwater>% accounting resumes. <Frequency> determines
99 how often do we check the amount of free space (value is in
102 That is, suspend accounting if there left <= 2% free; resume it
103 if we got >=4%; consider information about amount of free space
104 valid for 30 seconds.
106 ==============================================================
112 See Doc*/kernel/power/video.txt, it allows mode of video boot to be
115 ==============================================================
119 This variable has no effect and may be removed in future kernel
120 releases. Reading it always returns 0.
121 Up to Linux 3.17, it enabled/disabled automatic recomputing of msgmni
122 upon memory add/remove or upon ipc namespace creation/removal.
123 Echoing "1" into this file enabled msgmni automatic recomputing.
124 Echoing "0" turned it off. auto_msgmni default value was 1.
127 ==============================================================
131 x86 bootloader identification
133 This gives the bootloader type number as indicated by the bootloader,
134 shifted left by 4, and OR'd with the low four bits of the bootloader
135 version. The reason for this encoding is that this used to match the
136 type_of_loader field in the kernel header; the encoding is kept for
137 backwards compatibility. That is, if the full bootloader type number
138 is 0x15 and the full version number is 0x234, this file will contain
139 the value 340 = 0x154.
141 See the type_of_loader and ext_loader_type fields in
142 Documentation/x86/boot.txt for additional information.
144 ==============================================================
148 x86 bootloader version
150 The complete bootloader version number. In the example above, this
151 file will contain the value 564 = 0x234.
153 See the type_of_loader and ext_loader_ver fields in
154 Documentation/x86/boot.txt for additional information.
156 ==============================================================
160 Controls the kernel's callhome behavior in case of a kernel panic.
162 The s390 hardware allows an operating system to send a notification
163 to a service organization (callhome) in case of an operating system panic.
165 When the value in this file is 0 (which is the default behavior)
166 nothing happens in case of a kernel panic. If this value is set to "1"
167 the complete kernel oops message is send to the IBM customer service
168 organization in case the mainframe the Linux operating system is running
169 on has a service contract with IBM.
171 ==============================================================
175 Highest valid capability of the running kernel. Exports
176 CAP_LAST_CAP from the kernel.
178 ==============================================================
182 core_pattern is used to specify a core dumpfile pattern name.
183 . max length 128 characters; default value is "core"
184 . core_pattern is used as a pattern template for the output filename;
185 certain string patterns (beginning with '%') are substituted with
187 . backward compatibility with core_uses_pid:
188 If core_pattern does not include "%p" (default does not)
189 and core_uses_pid is set, then .PID will be appended to
191 . corename format specifiers:
192 %<NUL> '%' is dropped
195 %P global pid (init PID namespace)
197 %I global tid (init PID namespace)
200 %d dump mode, matches PR_SET_DUMPABLE and
201 /proc/sys/fs/suid_dumpable
205 %e executable filename (may be shortened)
207 %<OTHER> both are dropped
208 . If the first character of the pattern is a '|', the kernel will treat
209 the rest of the pattern as a command to run. The core dump will be
210 written to the standard input of that program instead of to a file.
212 ==============================================================
216 This sysctl is only applicable when core_pattern is configured to pipe
217 core files to a user space helper (when the first character of
218 core_pattern is a '|', see above). When collecting cores via a pipe
219 to an application, it is occasionally useful for the collecting
220 application to gather data about the crashing process from its
221 /proc/pid directory. In order to do this safely, the kernel must wait
222 for the collecting process to exit, so as not to remove the crashing
223 processes proc files prematurely. This in turn creates the
224 possibility that a misbehaving userspace collecting process can block
225 the reaping of a crashed process simply by never exiting. This sysctl
226 defends against that. It defines how many concurrent crashing
227 processes may be piped to user space applications in parallel. If
228 this value is exceeded, then those crashing processes above that value
229 are noted via the kernel log and their cores are skipped. 0 is a
230 special value, indicating that unlimited processes may be captured in
231 parallel, but that no waiting will take place (i.e. the collecting
232 process is not guaranteed access to /proc/<crashing pid>/). This
235 ==============================================================
239 The default coredump filename is "core". By setting
240 core_uses_pid to 1, the coredump filename becomes core.PID.
241 If core_pattern does not include "%p" (default does not)
242 and core_uses_pid is set, then .PID will be appended to
245 ==============================================================
249 When the value in this file is 0, ctrl-alt-del is trapped and
250 sent to the init(1) program to handle a graceful restart.
251 When, however, the value is > 0, Linux's reaction to a Vulcan
252 Nerve Pinch (tm) will be an immediate reboot, without even
253 syncing its dirty buffers.
255 Note: when a program (like dosemu) has the keyboard in 'raw'
256 mode, the ctrl-alt-del is intercepted by the program before it
257 ever reaches the kernel tty layer, and it's up to the program
258 to decide what to do with it.
260 ==============================================================
264 This toggle indicates whether unprivileged users are prevented
265 from using dmesg(8) to view messages from the kernel's log buffer.
266 When dmesg_restrict is set to (0) there are no restrictions. When
267 dmesg_restrict is set set to (1), users must have CAP_SYSLOG to use
270 The kernel config option CONFIG_SECURITY_DMESG_RESTRICT sets the
271 default value of dmesg_restrict.
273 ==============================================================
275 domainname & hostname:
277 These files can be used to set the NIS/YP domainname and the
278 hostname of your box in exactly the same way as the commands
279 domainname and hostname, i.e.:
280 # echo "darkstar" > /proc/sys/kernel/hostname
281 # echo "mydomain" > /proc/sys/kernel/domainname
282 has the same effect as
283 # hostname "darkstar"
284 # domainname "mydomain"
286 Note, however, that the classic darkstar.frop.org has the
287 hostname "darkstar" and DNS (Internet Domain Name Server)
288 domainname "frop.org", not to be confused with the NIS (Network
289 Information Service) or YP (Yellow Pages) domainname. These two
290 domain names are in general different. For a detailed discussion
291 see the hostname(1) man page.
293 ==============================================================
297 Path for the hotplug policy agent.
298 Default value is "/sbin/hotplug".
300 ==============================================================
304 Controls the kernel's behavior when a hung task is detected.
305 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
307 0: continue operation. This is the default behavior.
309 1: panic immediately.
311 ==============================================================
313 hung_task_check_count:
315 The upper bound on the number of tasks that are checked.
316 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
318 ==============================================================
320 hung_task_timeout_secs:
322 Check interval. When a task in D state did not get scheduled
323 for more than this value report a warning.
324 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
326 0: means infinite timeout - no checking done.
327 Possible values to set are in range {0..LONG_MAX/HZ}.
329 ==============================================================
333 The maximum number of warnings to report. During a check interval
334 if a hung task is detected, this value is decreased by 1.
335 When this value reaches 0, no more warnings will be reported.
336 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
338 -1: report an infinite number of warnings.
340 ==============================================================
344 A toggle indicating if the kexec_load syscall has been disabled. This
345 value defaults to 0 (false: kexec_load enabled), but can be set to 1
346 (true: kexec_load disabled). Once true, kexec can no longer be used, and
347 the toggle cannot be set back to false. This allows a kexec image to be
348 loaded before disabling the syscall, allowing a system to set up (and
349 later use) an image without it being altered. Generally used together
350 with the "modules_disabled" sysctl.
352 ==============================================================
356 This toggle indicates whether restrictions are placed on
357 exposing kernel addresses via /proc and other interfaces.
359 When kptr_restrict is set to (0), the default, there are no restrictions.
361 When kptr_restrict is set to (1), kernel pointers printed using the %pK
362 format specifier will be replaced with 0's unless the user has CAP_SYSLOG
363 and effective user and group ids are equal to the real ids. This is
364 because %pK checks are done at read() time rather than open() time, so
365 if permissions are elevated between the open() and the read() (e.g via
366 a setuid binary) then %pK will not leak kernel pointers to unprivileged
367 users. Note, this is a temporary solution only. The correct long-term
368 solution is to do the permission checks at open() time. Consider removing
369 world read permissions from files that use %pK, and using dmesg_restrict
370 to protect against uses of %pK in dmesg(8) if leaking kernel pointer
371 values to unprivileged users is a concern.
373 When kptr_restrict is set to (2), kernel pointers printed using
374 %pK will be replaced with 0's regardless of privileges.
376 ==============================================================
378 kstack_depth_to_print: (X86 only)
380 Controls the number of words to print when dumping the raw
383 ==============================================================
387 This flag controls the L2 cache of G3 processor boards. If
388 0, the cache is disabled. Enabled if nonzero.
390 ==============================================================
394 A toggle value indicating if modules are allowed to be loaded
395 in an otherwise modular kernel. This toggle defaults to off
396 (0), but can be set true (1). Once true, modules can be
397 neither loaded nor unloaded, and the toggle cannot be set back
398 to false. Generally used with the "kexec_load_disabled" toggle.
400 ==============================================================
402 msg_next_id, sem_next_id, and shm_next_id:
404 These three toggles allows to specify desired id for next allocated IPC
405 object: message, semaphore or shared memory respectively.
407 By default they are equal to -1, which means generic allocation logic.
408 Possible values to set are in range {0..INT_MAX}.
411 1) kernel doesn't guarantee, that new object will have desired id. So,
412 it's up to userspace, how to handle an object with "wrong" id.
413 2) Toggle with non-default value will be set back to -1 by kernel after
414 successful IPC object allocation.
416 ==============================================================
420 Enables/Disables the NMI watchdog on x86 systems. When the value is
421 non-zero the NMI watchdog is enabled and will continuously test all
422 online cpus to determine whether or not they are still functioning
423 properly. Currently, passing "nmi_watchdog=" parameter at boot time is
424 required for this function to work.
426 If LAPIC NMI watchdog method is in use (nmi_watchdog=2 kernel
427 parameter), the NMI watchdog shares registers with oprofile. By
428 disabling the NMI watchdog, oprofile may have more registers to
431 ==============================================================
435 Enables/disables automatic page fault based NUMA memory
436 balancing. Memory is moved automatically to nodes
437 that access it often.
439 Enables/disables automatic NUMA memory balancing. On NUMA machines, there
440 is a performance penalty if remote memory is accessed by a CPU. When this
441 feature is enabled the kernel samples what task thread is accessing memory
442 by periodically unmapping pages and later trapping a page fault. At the
443 time of the page fault, it is determined if the data being accessed should
444 be migrated to a local memory node.
446 The unmapping of pages and trapping faults incur additional overhead that
447 ideally is offset by improved memory locality but there is no universal
448 guarantee. If the target workload is already bound to NUMA nodes then this
449 feature should be disabled. Otherwise, if the system overhead from the
450 feature is too high then the rate the kernel samples for NUMA hinting
451 faults may be controlled by the numa_balancing_scan_period_min_ms,
452 numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms,
453 numa_balancing_scan_size_mb, and numa_balancing_settle_count sysctls.
455 ==============================================================
457 numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms,
458 numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb
460 Automatic NUMA balancing scans tasks address space and unmaps pages to
461 detect if pages are properly placed or if the data should be migrated to a
462 memory node local to where the task is running. Every "scan delay" the task
463 scans the next "scan size" number of pages in its address space. When the
464 end of the address space is reached the scanner restarts from the beginning.
466 In combination, the "scan delay" and "scan size" determine the scan rate.
467 When "scan delay" decreases, the scan rate increases. The scan delay and
468 hence the scan rate of every task is adaptive and depends on historical
469 behaviour. If pages are properly placed then the scan delay increases,
470 otherwise the scan delay decreases. The "scan size" is not adaptive but
471 the higher the "scan size", the higher the scan rate.
473 Higher scan rates incur higher system overhead as page faults must be
474 trapped and potentially data must be migrated. However, the higher the scan
475 rate, the more quickly a tasks memory is migrated to a local node if the
476 workload pattern changes and minimises performance impact due to remote
477 memory accesses. These sysctls control the thresholds for scan delays and
478 the number of pages scanned.
480 numa_balancing_scan_period_min_ms is the minimum time in milliseconds to
481 scan a tasks virtual memory. It effectively controls the maximum scanning
484 numa_balancing_scan_delay_ms is the starting "scan delay" used for a task
485 when it initially forks.
487 numa_balancing_scan_period_max_ms is the maximum time in milliseconds to
488 scan a tasks virtual memory. It effectively controls the minimum scanning
491 numa_balancing_scan_size_mb is how many megabytes worth of pages are
492 scanned for a given scan.
494 ==============================================================
496 osrelease, ostype & version:
503 #5 Wed Feb 25 21:49:24 MET 1998
505 The files osrelease and ostype should be clear enough. Version
506 needs a little more clarification however. The '#5' means that
507 this is the fifth kernel built from this source base and the
508 date behind it indicates the time the kernel was built.
509 The only way to tune these values is to rebuild the kernel :-)
511 ==============================================================
513 overflowgid & overflowuid:
515 if your architecture did not always support 32-bit UIDs (i.e. arm,
516 i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to
517 applications that use the old 16-bit UID/GID system calls, if the
518 actual UID or GID would exceed 65535.
520 These sysctls allow you to change the value of the fixed UID and GID.
521 The default is 65534.
523 ==============================================================
527 The value in this file represents the number of seconds the kernel
528 waits before rebooting on a panic. When you use the software watchdog,
529 the recommended setting is 60.
531 ==============================================================
535 Controls the kernel's behaviour when an oops or BUG is encountered.
537 0: try to continue operation
539 1: panic immediately. If the `panic' sysctl is also non-zero then the
540 machine will be rebooted.
542 ==============================================================
544 panic_on_stackoverflow:
546 Controls the kernel's behavior when detecting the overflows of
547 kernel, IRQ and exception stacks except a user stack.
548 This file shows up if CONFIG_DEBUG_STACKOVERFLOW is enabled.
550 0: try to continue operation.
552 1: panic immediately.
554 ==============================================================
556 panic_on_unrecovered_nmi:
558 The default Linux behaviour on an NMI of either memory or unknown is
559 to continue operation. For many environments such as scientific
560 computing it is preferable that the box is taken out and the error
561 dealt with than an uncorrected parity/ECC error get propagated.
563 A small number of systems do generate NMI's for bizarre random reasons
564 such as power management so the default is off. That sysctl works like
565 the existing panic controls already in that directory.
567 ==============================================================
571 Calls panic() in the WARN() path when set to 1. This is useful to avoid
572 a kernel rebuild when attempting to kdump at the location of a WARN().
574 0: only WARN(), default behaviour.
576 1: call panic() after printing out WARN() location.
578 ==============================================================
580 perf_cpu_time_max_percent:
582 Hints to the kernel how much CPU time it should be allowed to
583 use to handle perf sampling events. If the perf subsystem
584 is informed that its samples are exceeding this limit, it
585 will drop its sampling frequency to attempt to reduce its CPU
588 Some perf sampling happens in NMIs. If these samples
589 unexpectedly take too long to execute, the NMIs can become
590 stacked up next to each other so much that nothing else is
593 0: disable the mechanism. Do not monitor or correct perf's
594 sampling rate no matter how CPU time it takes.
596 1-100: attempt to throttle perf's sample rate to this
597 percentage of CPU. Note: the kernel calculates an
598 "expected" length of each sample event. 100 here means
599 100% of that expected length. Even if this is set to
600 100, you may still see sample throttling if this
601 length is exceeded. Set to 0 if you truly do not care
602 how much CPU is consumed.
604 ==============================================================
609 PID allocation wrap value. When the kernel's next PID value
610 reaches this value, it wraps back to a minimum PID value.
611 PIDs of value pid_max or larger are not allocated.
613 ==============================================================
617 The last pid allocated in the current (the one task using this sysctl
618 lives in) pid namespace. When selecting a pid for a next task on fork
619 kernel tries to allocate a number starting from this one.
621 ==============================================================
623 powersave-nap: (PPC only)
625 If set, Linux-PPC will use the 'nap' mode of powersaving,
626 otherwise the 'doze' mode will be used.
628 ==============================================================
632 The four values in printk denote: console_loglevel,
633 default_message_loglevel, minimum_console_loglevel and
634 default_console_loglevel respectively.
636 These values influence printk() behavior when printing or
637 logging error messages. See 'man 2 syslog' for more info on
638 the different loglevels.
640 - console_loglevel: messages with a higher priority than
641 this will be printed to the console
642 - default_message_loglevel: messages without an explicit priority
643 will be printed with this priority
644 - minimum_console_loglevel: minimum (highest) value to which
645 console_loglevel can be set
646 - default_console_loglevel: default value for console_loglevel
648 ==============================================================
652 Delay each printk message in printk_delay milliseconds
654 Value from 0 - 10000 is allowed.
656 ==============================================================
660 Some warning messages are rate limited. printk_ratelimit specifies
661 the minimum length of time between these messages (in jiffies), by
662 default we allow one every 5 seconds.
664 A value of 0 will disable rate limiting.
666 ==============================================================
668 printk_ratelimit_burst:
670 While long term we enforce one message per printk_ratelimit
671 seconds, we do allow a burst of messages to pass through.
672 printk_ratelimit_burst specifies the number of messages we can
673 send before ratelimiting kicks in.
675 ==============================================================
679 This option can be used to select the type of process address
680 space randomization that is used in the system, for architectures
681 that support this feature.
683 0 - Turn the process address space randomization off. This is the
684 default for architectures that do not support this feature anyways,
685 and kernels that are booted with the "norandmaps" parameter.
687 1 - Make the addresses of mmap base, stack and VDSO page randomized.
688 This, among other things, implies that shared libraries will be
689 loaded to random addresses. Also for PIE-linked binaries, the
690 location of code start is randomized. This is the default if the
691 CONFIG_COMPAT_BRK option is enabled.
693 2 - Additionally enable heap randomization. This is the default if
694 CONFIG_COMPAT_BRK is disabled.
696 There are a few legacy applications out there (such as some ancient
697 versions of libc.so.5 from 1996) that assume that brk area starts
698 just after the end of the code+bss. These applications break when
699 start of the brk area is randomized. There are however no known
700 non-legacy applications that would be broken this way, so for most
701 systems it is safe to choose full randomization.
703 Systems with ancient and/or broken binaries should be configured
704 with CONFIG_COMPAT_BRK enabled, which excludes the heap from process
705 address space randomization.
707 ==============================================================
709 reboot-cmd: (Sparc only)
711 ??? This seems to be a way to give an argument to the Sparc
712 ROM/Flash boot loader. Maybe to tell it what to do after
715 ==============================================================
717 rtsig-max & rtsig-nr:
719 The file rtsig-max can be used to tune the maximum number
720 of POSIX realtime (queued) signals that can be outstanding
723 rtsig-nr shows the number of RT signals currently queued.
725 ==============================================================
729 This file shows the size of the generic SCSI (sg) buffer.
730 You can't tune it just yet, but you could change it on
731 compile time by editing include/scsi/sg.h and changing
732 the value of SG_BIG_BUFF.
734 There shouldn't be any reason to change this value. If
735 you can come up with one, you probably know what you
738 ==============================================================
742 This parameter sets the total amount of shared memory pages that
743 can be used system wide. Hence, SHMALL should always be at least
744 ceil(shmmax/PAGE_SIZE).
746 If you are not sure what the default PAGE_SIZE is on your Linux
747 system, you can run the following command:
751 ==============================================================
755 This value can be used to query and set the run time limit
756 on the maximum shared memory segment size that can be created.
757 Shared memory segments up to 1Gb are now supported in the
758 kernel. This value defaults to SHMMAX.
760 ==============================================================
764 Linux lets you set resource limits, including how much memory one
765 process can consume, via setrlimit(2). Unfortunately, shared memory
766 segments are allowed to exist without association with any process, and
767 thus might not be counted against any resource limits. If enabled,
768 shared memory segments are automatically destroyed when their attach
769 count becomes zero after a detach or a process termination. It will
770 also destroy segments that were created, but never attached to, on exit
771 from the process. The only use left for IPC_RMID is to immediately
772 destroy an unattached segment. Of course, this breaks the way things are
773 defined, so some applications might stop working. Note that this
774 feature will do you no good unless you also configure your resource
775 limits (in particular, RLIMIT_AS and RLIMIT_NPROC). Most systems don't
778 Note that if you change this from 0 to 1, already created segments
779 without users and with a dead originative process will be destroyed.
781 ==============================================================
783 sysctl_writes_strict:
785 Control how file position affects the behavior of updating sysctl values
786 via the /proc/sys interface:
788 -1 - Legacy per-write sysctl value handling, with no printk warnings.
789 Each write syscall must fully contain the sysctl value to be
790 written, and multiple writes on the same sysctl file descriptor
791 will rewrite the sysctl value, regardless of file position.
792 0 - (default) Same behavior as above, but warn about processes that
793 perform writes to a sysctl file descriptor when the file position
795 1 - Respect file position when writing sysctl strings. Multiple writes
796 will append to the sysctl value buffer. Anything past the max length
797 of the sysctl value buffer will be ignored. Writes to numeric sysctl
798 entries must always be at file position 0 and the value must be
799 fully contained in the buffer sent in the write syscall.
801 ==============================================================
803 softlockup_all_cpu_backtrace:
805 This value controls the soft lockup detector thread's behavior
806 when a soft lockup condition is detected as to whether or not
807 to gather further debug information. If enabled, each cpu will
808 be issued an NMI and instructed to capture stack trace.
810 This feature is only applicable for architectures which support
813 0: do nothing. This is the default behavior.
815 1: on detection capture more debug information.
817 ==============================================================
821 Non-zero if the kernel has been tainted. Numeric values, which
822 can be ORed together:
824 1 - A module with a non-GPL license has been loaded, this
825 includes modules with no license.
826 Set by modutils >= 2.4.9 and module-init-tools.
827 2 - A module was force loaded by insmod -f.
828 Set by modutils >= 2.4.9 and module-init-tools.
829 4 - Unsafe SMP processors: SMP with CPUs not designed for SMP.
830 8 - A module was forcibly unloaded from the system by rmmod -f.
831 16 - A hardware machine check error occurred on the system.
832 32 - A bad page was discovered on the system.
833 64 - The user has asked that the system be marked "tainted". This
834 could be because they are running software that directly modifies
835 the hardware, or for other reasons.
836 128 - The system has died.
837 256 - The ACPI DSDT has been overridden with one supplied by the user
838 instead of using the one provided by the hardware.
839 512 - A kernel warning has occurred.
840 1024 - A module from drivers/staging was loaded.
841 2048 - The system is working around a severe firmware bug.
842 4096 - An out-of-tree module has been loaded.
843 8192 - An unsigned module has been loaded in a kernel supporting module
845 16384 - A soft lockup has previously occurred on the system.
847 ==============================================================
851 The value in this file affects behavior of handling NMI. When the
852 value is non-zero, unknown NMI is trapped and then panic occurs. At
853 that time, kernel debugging information is displayed on console.
855 NMI switch that most IA32 servers have fires unknown NMI up, for
856 example. If a system hangs up, try pressing the NMI switch.
858 ==============================================================
862 This value can be used to control the frequency of hrtimer and NMI
863 events and the soft and hard lockup thresholds. The default threshold
866 The softlockup threshold is (2 * watchdog_thresh). Setting this
867 tunable to zero will disable lockup detection altogether.
869 ==============================================================