blob: f11ca7979fa67b5bbf08339cf473d110ecf92f7d [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001Documentation for /proc/sys/kernel/* kernel version 2.2.10
2 (c) 1998, 1999, Rik van Riel <riel@nl.linux.org>
Shen Feng760df932009-04-02 16:57:20 -07003 (c) 2009, Shen Feng<shen@cn.fujitsu.com>
Linus Torvalds1da177e2005-04-16 15:20:36 -07004
5For general info and legal blurb, please look in README.
6
7==============================================================
8
9This file contains documentation for the sysctl files in
10/proc/sys/kernel/ and is valid for Linux kernel version 2.2.
11
12The files in this directory can be used to tune and monitor
13miscellaneous and general things in the operation of the Linux
14kernel. Since some of the files _can_ be used to screw up your
15system, it is advisable to read both documentation and source
16before actually making adjustments.
17
18Currently, these files might (depending on your configuration)
19show up in /proc/sys/kernel:
Pavel Machekc255d842006-02-20 18:27:58 -080020- acpi_video_flags
Linus Torvalds1da177e2005-04-16 15:20:36 -070021- acct
Shen Feng760df932009-04-02 16:57:20 -070022- auto_msgmni
Linus Torvalds1da177e2005-04-16 15:20:36 -070023- core_pattern
24- core_uses_pid
25- ctrl-alt-del
26- dentry-state
27- domainname
28- hostname
29- hotplug
30- java-appletviewer [ binfmt_java, obsolete ]
31- java-interpreter [ binfmt_java, obsolete ]
Chuck Ebbert0741f4d2006-12-07 02:14:11 +010032- kstack_depth_to_print [ X86 only ]
Linus Torvalds1da177e2005-04-16 15:20:36 -070033- l2cr [ PPC only ]
Michael Opdenackerac76cff2008-02-13 15:03:32 -080034- modprobe ==> Documentation/debugging-modules.txt
Linus Torvalds1da177e2005-04-16 15:20:36 -070035- msgmax
36- msgmnb
37- msgmni
Shen Feng760df932009-04-02 16:57:20 -070038- nmi_watchdog
Linus Torvalds1da177e2005-04-16 15:20:36 -070039- osrelease
40- ostype
41- overflowgid
42- overflowuid
43- panic
44- pid_max
45- powersave-nap [ PPC only ]
Shen Feng760df932009-04-02 16:57:20 -070046- panic_on_unrecovered_nmi
Linus Torvalds1da177e2005-04-16 15:20:36 -070047- printk
Jiri Kosina1ec7fd52008-02-09 23:24:08 +010048- randomize_va_space
Linus Torvalds1da177e2005-04-16 15:20:36 -070049- real-root-dev ==> Documentation/initrd.txt
50- reboot-cmd [ SPARC only ]
51- rtsig-max
52- rtsig-nr
53- sem
54- sg-big-buff [ generic SCSI device (sg) ]
55- shmall
56- shmmax [ sysv ipc ]
57- shmmni
58- stop-a [ SPARC only ]
59- sysrq ==> Documentation/sysrq.txt
60- tainted
61- threads-max
Shen Feng760df932009-04-02 16:57:20 -070062- unknown_nmi_panic
Linus Torvalds1da177e2005-04-16 15:20:36 -070063- version
64
65==============================================================
66
Pavel Machekc255d842006-02-20 18:27:58 -080067acpi_video_flags:
68
69flags
70
71See Doc*/kernel/power/video.txt, it allows mode of video boot to be
72set during run time.
73
74==============================================================
75
Linus Torvalds1da177e2005-04-16 15:20:36 -070076acct:
77
78highwater lowwater frequency
79
80If BSD-style process accounting is enabled these values control
81its behaviour. If free space on filesystem where the log lives
82goes below <lowwater>% accounting suspends. If free space gets
83above <highwater>% accounting resumes. <Frequency> determines
84how often do we check the amount of free space (value is in
85seconds). Default:
864 2 30
87That is, suspend accounting if there left <= 2% free; resume it
88if we got >=4%; consider information about amount of free space
89valid for 30 seconds.
90
91==============================================================
92
93core_pattern:
94
95core_pattern is used to specify a core dumpfile pattern name.
Matthias Urlichscd081042006-10-11 01:21:57 -070096. max length 128 characters; default value is "core"
Linus Torvalds1da177e2005-04-16 15:20:36 -070097. core_pattern is used as a pattern template for the output filename;
98 certain string patterns (beginning with '%') are substituted with
99 their actual values.
100. backward compatibility with core_uses_pid:
101 If core_pattern does not include "%p" (default does not)
102 and core_uses_pid is set, then .PID will be appended to
103 the filename.
104. corename format specifiers:
105 %<NUL> '%' is dropped
106 %% output one '%'
107 %p pid
108 %u uid
109 %g gid
110 %s signal number
111 %t UNIX time of dump
112 %h hostname
113 %e executable filename
114 %<OTHER> both are dropped
Matthias Urlichscd081042006-10-11 01:21:57 -0700115. If the first character of the pattern is a '|', the kernel will treat
116 the rest of the pattern as a command to run. The core dump will be
117 written to the standard input of that program instead of to a file.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700118
119==============================================================
120
121core_uses_pid:
122
123The default coredump filename is "core". By setting
124core_uses_pid to 1, the coredump filename becomes core.PID.
125If core_pattern does not include "%p" (default does not)
126and core_uses_pid is set, then .PID will be appended to
127the filename.
128
129==============================================================
130
131ctrl-alt-del:
132
133When the value in this file is 0, ctrl-alt-del is trapped and
134sent to the init(1) program to handle a graceful restart.
135When, however, the value is > 0, Linux's reaction to a Vulcan
136Nerve Pinch (tm) will be an immediate reboot, without even
137syncing its dirty buffers.
138
139Note: when a program (like dosemu) has the keyboard in 'raw'
140mode, the ctrl-alt-del is intercepted by the program before it
141ever reaches the kernel tty layer, and it's up to the program
142to decide what to do with it.
143
144==============================================================
145
146domainname & hostname:
147
148These files can be used to set the NIS/YP domainname and the
149hostname of your box in exactly the same way as the commands
150domainname and hostname, i.e.:
151# echo "darkstar" > /proc/sys/kernel/hostname
152# echo "mydomain" > /proc/sys/kernel/domainname
153has the same effect as
154# hostname "darkstar"
155# domainname "mydomain"
156
157Note, however, that the classic darkstar.frop.org has the
158hostname "darkstar" and DNS (Internet Domain Name Server)
159domainname "frop.org", not to be confused with the NIS (Network
160Information Service) or YP (Yellow Pages) domainname. These two
161domain names are in general different. For a detailed discussion
162see the hostname(1) man page.
163
164==============================================================
165
166hotplug:
167
168Path for the hotplug policy agent.
169Default value is "/sbin/hotplug".
170
171==============================================================
172
173l2cr: (PPC only)
174
175This flag controls the L2 cache of G3 processor boards. If
1760, the cache is disabled. Enabled if nonzero.
177
178==============================================================
179
Chuck Ebbert0741f4d2006-12-07 02:14:11 +0100180kstack_depth_to_print: (X86 only)
181
182Controls the number of words to print when dumping the raw
183kernel stack.
184
185==============================================================
186
Linus Torvalds1da177e2005-04-16 15:20:36 -0700187osrelease, ostype & version:
188
189# cat osrelease
1902.1.88
191# cat ostype
192Linux
193# cat version
194#5 Wed Feb 25 21:49:24 MET 1998
195
196The files osrelease and ostype should be clear enough. Version
197needs a little more clarification however. The '#5' means that
198this is the fifth kernel built from this source base and the
199date behind it indicates the time the kernel was built.
200The only way to tune these values is to rebuild the kernel :-)
201
202==============================================================
203
204overflowgid & overflowuid:
205
206if your architecture did not always support 32-bit UIDs (i.e. arm, i386,
207m68k, sh, and sparc32), a fixed UID and GID will be returned to
208applications that use the old 16-bit UID/GID system calls, if the actual
209UID or GID would exceed 65535.
210
211These sysctls allow you to change the value of the fixed UID and GID.
212The default is 65534.
213
214==============================================================
215
216panic:
217
218The value in this file represents the number of seconds the
219kernel waits before rebooting on a panic. When you use the
220software watchdog, the recommended setting is 60.
221
222==============================================================
223
224panic_on_oops:
225
226Controls the kernel's behaviour when an oops or BUG is encountered.
227
2280: try to continue operation
229
Matt LaPlantea982ac02007-05-09 07:35:06 +02002301: panic immediately. If the `panic' sysctl is also non-zero then the
Maxime Bizon8b23d04d2006-08-05 12:14:32 -0700231 machine will be rebooted.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700232
233==============================================================
234
235pid_max:
236
Robert P. J. Daybeb7dd82007-05-09 07:14:03 +0200237PID allocation wrap value. When the kernel's next PID value
Linus Torvalds1da177e2005-04-16 15:20:36 -0700238reaches this value, it wraps back to a minimum PID value.
239PIDs of value pid_max or larger are not allocated.
240
241==============================================================
242
243powersave-nap: (PPC only)
244
245If set, Linux-PPC will use the 'nap' mode of powersaving,
246otherwise the 'doze' mode will be used.
247
248==============================================================
249
250printk:
251
252The four values in printk denote: console_loglevel,
253default_message_loglevel, minimum_console_loglevel and
254default_console_loglevel respectively.
255
256These values influence printk() behavior when printing or
257logging error messages. See 'man 2 syslog' for more info on
258the different loglevels.
259
260- console_loglevel: messages with a higher priority than
261 this will be printed to the console
262- default_message_level: messages without an explicit priority
263 will be printed with this priority
264- minimum_console_loglevel: minimum (highest) value to which
265 console_loglevel can be set
266- default_console_loglevel: default value for console_loglevel
267
268==============================================================
269
270printk_ratelimit:
271
272Some warning messages are rate limited. printk_ratelimit specifies
273the minimum length of time between these messages (in jiffies), by
274default we allow one every 5 seconds.
275
276A value of 0 will disable rate limiting.
277
278==============================================================
279
280printk_ratelimit_burst:
281
282While long term we enforce one message per printk_ratelimit
283seconds, we do allow a burst of messages to pass through.
284printk_ratelimit_burst specifies the number of messages we can
285send before ratelimiting kicks in.
286
287==============================================================
288
Jiri Kosina1ec7fd52008-02-09 23:24:08 +0100289randomize-va-space:
290
291This option can be used to select the type of process address
292space randomization that is used in the system, for architectures
293that support this feature.
294
2950 - Turn the process address space randomization off by default.
296
2971 - Make the addresses of mmap base, stack and VDSO page randomized.
298 This, among other things, implies that shared libraries will be
299 loaded to random addresses. Also for PIE-linked binaries, the location
300 of code start is randomized.
301
302 With heap randomization, the situation is a little bit more
303 complicated.
304 There a few legacy applications out there (such as some ancient
305 versions of libc.so.5 from 1996) that assume that brk area starts
306 just after the end of the code+bss. These applications break when
307 start of the brk area is randomized. There are however no known
308 non-legacy applications that would be broken this way, so for most
309 systems it is safe to choose full randomization. However there is
310 a CONFIG_COMPAT_BRK option for systems with ancient and/or broken
311 binaries, that makes heap non-randomized, but keeps all other
312 parts of process address space randomized if randomize_va_space
313 sysctl is turned on.
314
315==============================================================
316
Linus Torvalds1da177e2005-04-16 15:20:36 -0700317reboot-cmd: (Sparc only)
318
319??? This seems to be a way to give an argument to the Sparc
320ROM/Flash boot loader. Maybe to tell it what to do after
321rebooting. ???
322
323==============================================================
324
325rtsig-max & rtsig-nr:
326
327The file rtsig-max can be used to tune the maximum number
328of POSIX realtime (queued) signals that can be outstanding
329in the system.
330
331rtsig-nr shows the number of RT signals currently queued.
332
333==============================================================
334
335sg-big-buff:
336
337This file shows the size of the generic SCSI (sg) buffer.
338You can't tune it just yet, but you could change it on
339compile time by editing include/scsi/sg.h and changing
340the value of SG_BIG_BUFF.
341
342There shouldn't be any reason to change this value. If
343you can come up with one, you probably know what you
344are doing anyway :)
345
346==============================================================
347
348shmmax:
349
350This value can be used to query and set the run time limit
351on the maximum shared memory segment size that can be created.
352Shared memory segments up to 1Gb are now supported in the
353kernel. This value defaults to SHMMAX.
354
355==============================================================
356
Ravikiran G Thirumalaic4f3b632007-10-16 23:26:09 -0700357softlockup_thresh:
358
Andrew Mortonb4d19cc2008-09-22 13:57:51 -0700359This value can be used to lower the softlockup tolerance threshold. The
360default threshold is 60 seconds. If a cpu is locked up for 60 seconds,
361the kernel complains. Valid values are 1-60 seconds. Setting this
362tunable to zero will disable the softlockup detection altogether.
Ravikiran G Thirumalaic4f3b632007-10-16 23:26:09 -0700363
364==============================================================
365
Linus Torvalds1da177e2005-04-16 15:20:36 -0700366tainted:
367
368Non-zero if the kernel has been tainted. Numeric values, which
369can be ORed together:
370
Greg Kroah-Hartmanbb206982008-10-17 15:01:07 -0700371 1 - A module with a non-GPL license has been loaded, this
372 includes modules with no license.
373 Set by modutils >= 2.4.9 and module-init-tools.
374 2 - A module was force loaded by insmod -f.
375 Set by modutils >= 2.4.9 and module-init-tools.
376 4 - Unsafe SMP processors: SMP with CPUs not designed for SMP.
377 8 - A module was forcibly unloaded from the system by rmmod -f.
378 16 - A hardware machine check error occurred on the system.
379 32 - A bad page was discovered on the system.
380 64 - The user has asked that the system be marked "tainted". This
381 could be because they are running software that directly modifies
382 the hardware, or for other reasons.
383 128 - The system has died.
384 256 - The ACPI DSDT has been overridden with one supplied by the user
385 instead of using the one provided by the hardware.
386 512 - A kernel warning has occurred.
3871024 - A module from drivers/staging was loaded.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700388
Shen Feng760df932009-04-02 16:57:20 -0700389==============================================================
390
391auto_msgmni:
392
393Enables/Disables automatic recomputing of msgmni upon memory add/remove or
394upon ipc namespace creation/removal (see the msgmni description above).
395Echoing "1" into this file enables msgmni automatic recomputing.
396Echoing "0" turns it off.
397auto_msgmni default value is 1.
398
399==============================================================
400
401nmi_watchdog:
402
403Enables/Disables the NMI watchdog on x86 systems. When the value is non-zero
404the NMI watchdog is enabled and will continuously test all online cpus to
405determine whether or not they are still functioning properly. Currently,
406passing "nmi_watchdog=" parameter at boot time is required for this function
407to work.
408
409If LAPIC NMI watchdog method is in use (nmi_watchdog=2 kernel parameter), the
410NMI watchdog shares registers with oprofile. By disabling the NMI watchdog,
411oprofile may have more registers to utilize.
412
413==============================================================
414
415unknown_nmi_panic:
416
417The value in this file affects behavior of handling NMI. When the value is
418non-zero, unknown NMI is trapped and then panic occurs. At that time, kernel
419debugging information is displayed on console.
420
421NMI switch that most IA32 servers have fires unknown NMI up, for example.
422If a system hangs up, try pressing the NMI switch.
423
424==============================================================
425
426panic_on_unrecovered_nmi:
427
428The default Linux behaviour on an NMI of either memory or unknown is to continue
429operation. For many environments such as scientific computing it is preferable
430that the box is taken out and the error dealt with than an uncorrected
431parity/ECC error get propogated.
432
433A small number of systems do generate NMI's for bizarre random reasons such as
434power management so the default is off. That sysctl works like the existing
435panic controls already in that directory.
436