Commit | Line | Data |
---|---|---|
1da177e4 LT |
1 | Documentation for /proc/sys/kernel/* kernel version 2.2.10 |
2 | (c) 1998, 1999, Rik van Riel <riel@nl.linux.org> | |
760df93e | 3 | (c) 2009, Shen Feng<shen@cn.fujitsu.com> |
1da177e4 LT |
4 | |
5 | For general info and legal blurb, please look in README. | |
6 | ||
7 | ============================================================== | |
8 | ||
9 | This file contains documentation for the sysctl files in | |
10 | /proc/sys/kernel/ and is valid for Linux kernel version 2.2. | |
11 | ||
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. | |
17 | ||
18 | Currently, these files might (depending on your configuration) | |
19 | show up in /proc/sys/kernel: | |
807094c0 | 20 | |
1da177e4 | 21 | - acct |
807094c0 BP |
22 | - acpi_video_flags |
23 | - auto_msgmni | |
d75757ab PA |
24 | - bootloader_type [ X86 only ] |
25 | - bootloader_version [ X86 only ] | |
c114728a | 26 | - callhome [ S390 only ] |
73efc039 | 27 | - cap_last_cap |
1da177e4 | 28 | - core_pattern |
a293980c | 29 | - core_pipe_limit |
1da177e4 LT |
30 | - core_uses_pid |
31 | - ctrl-alt-del | |
eaf06b24 | 32 | - dmesg_restrict |
1da177e4 LT |
33 | - domainname |
34 | - hostname | |
35 | - hotplug | |
270750db AT |
36 | - hung_task_panic |
37 | - hung_task_check_count | |
38 | - hung_task_timeout_secs | |
39 | - hung_task_warnings | |
7984754b | 40 | - kexec_load_disabled |
455cd5ab | 41 | - kptr_restrict |
0741f4d2 | 42 | - kstack_depth_to_print [ X86 only ] |
1da177e4 | 43 | - l2cr [ PPC only ] |
ac76cff2 | 44 | - modprobe ==> Documentation/debugging-modules.txt |
3d43321b | 45 | - modules_disabled |
03f59566 | 46 | - msg_next_id [ sysv ipc ] |
1da177e4 LT |
47 | - msgmax |
48 | - msgmnb | |
49 | - msgmni | |
760df93e | 50 | - nmi_watchdog |
1da177e4 LT |
51 | - osrelease |
52 | - ostype | |
53 | - overflowgid | |
54 | - overflowuid | |
55 | - panic | |
807094c0 BP |
56 | - panic_on_oops |
57 | - panic_on_unrecovered_nmi | |
55af7796 | 58 | - panic_on_stackoverflow |
1da177e4 LT |
59 | - pid_max |
60 | - powersave-nap [ PPC only ] | |
61 | - printk | |
807094c0 BP |
62 | - printk_delay |
63 | - printk_ratelimit | |
64 | - printk_ratelimit_burst | |
1ec7fd50 | 65 | - randomize_va_space |
1da177e4 LT |
66 | - real-root-dev ==> Documentation/initrd.txt |
67 | - reboot-cmd [ SPARC only ] | |
68 | - rtsig-max | |
69 | - rtsig-nr | |
70 | - sem | |
03f59566 | 71 | - sem_next_id [ sysv ipc ] |
1da177e4 | 72 | - sg-big-buff [ generic SCSI device (sg) ] |
03f59566 | 73 | - shm_next_id [ sysv ipc ] |
b34a6b1d | 74 | - shm_rmid_forced |
1da177e4 LT |
75 | - shmall |
76 | - shmmax [ sysv ipc ] | |
77 | - shmmni | |
78 | - stop-a [ SPARC only ] | |
79 | - sysrq ==> Documentation/sysrq.txt | |
80 | - tainted | |
81 | - threads-max | |
760df93e | 82 | - unknown_nmi_panic |
08825c90 | 83 | - watchdog_thresh |
1da177e4 LT |
84 | - version |
85 | ||
86 | ============================================================== | |
87 | ||
88 | acct: | |
89 | ||
90 | highwater lowwater frequency | |
91 | ||
92 | If BSD-style process accounting is enabled these values control | |
93 | its behaviour. If free space on filesystem where the log lives | |
94 | goes below <lowwater>% accounting suspends. If free space gets | |
95 | above <highwater>% accounting resumes. <Frequency> determines | |
96 | how often do we check the amount of free space (value is in | |
97 | seconds). Default: | |
98 | 4 2 30 | |
99 | That is, suspend accounting if there left <= 2% free; resume it | |
100 | if we got >=4%; consider information about amount of free space | |
101 | valid for 30 seconds. | |
102 | ||
807094c0 BP |
103 | ============================================================== |
104 | ||
105 | acpi_video_flags: | |
106 | ||
107 | flags | |
108 | ||
109 | See Doc*/kernel/power/video.txt, it allows mode of video boot to be | |
110 | set during run time. | |
111 | ||
112 | ============================================================== | |
113 | ||
114 | auto_msgmni: | |
115 | ||
116 | Enables/Disables automatic recomputing of msgmni upon memory add/remove | |
117 | or upon ipc namespace creation/removal (see the msgmni description | |
118 | above). Echoing "1" into this file enables msgmni automatic recomputing. | |
119 | Echoing "0" turns it off. auto_msgmni default value is 1. | |
120 | ||
121 | ||
1da177e4 LT |
122 | ============================================================== |
123 | ||
d75757ab PA |
124 | bootloader_type: |
125 | ||
126 | x86 bootloader identification | |
127 | ||
128 | This gives the bootloader type number as indicated by the bootloader, | |
129 | shifted left by 4, and OR'd with the low four bits of the bootloader | |
130 | version. The reason for this encoding is that this used to match the | |
131 | type_of_loader field in the kernel header; the encoding is kept for | |
132 | backwards compatibility. That is, if the full bootloader type number | |
133 | is 0x15 and the full version number is 0x234, this file will contain | |
134 | the value 340 = 0x154. | |
135 | ||
136 | See the type_of_loader and ext_loader_type fields in | |
137 | Documentation/x86/boot.txt for additional information. | |
138 | ||
139 | ============================================================== | |
140 | ||
141 | bootloader_version: | |
142 | ||
143 | x86 bootloader version | |
144 | ||
145 | The complete bootloader version number. In the example above, this | |
146 | file will contain the value 564 = 0x234. | |
147 | ||
148 | See the type_of_loader and ext_loader_ver fields in | |
149 | Documentation/x86/boot.txt for additional information. | |
150 | ||
151 | ============================================================== | |
152 | ||
c114728a HJP |
153 | callhome: |
154 | ||
155 | Controls the kernel's callhome behavior in case of a kernel panic. | |
156 | ||
157 | The s390 hardware allows an operating system to send a notification | |
158 | to a service organization (callhome) in case of an operating system panic. | |
159 | ||
160 | When the value in this file is 0 (which is the default behavior) | |
161 | nothing happens in case of a kernel panic. If this value is set to "1" | |
162 | the complete kernel oops message is send to the IBM customer service | |
163 | organization in case the mainframe the Linux operating system is running | |
164 | on has a service contract with IBM. | |
165 | ||
166 | ============================================================== | |
167 | ||
73efc039 DB |
168 | cap_last_cap |
169 | ||
170 | Highest valid capability of the running kernel. Exports | |
171 | CAP_LAST_CAP from the kernel. | |
172 | ||
173 | ============================================================== | |
174 | ||
1da177e4 LT |
175 | core_pattern: |
176 | ||
177 | core_pattern is used to specify a core dumpfile pattern name. | |
cd081041 | 178 | . max length 128 characters; default value is "core" |
1da177e4 LT |
179 | . core_pattern is used as a pattern template for the output filename; |
180 | certain string patterns (beginning with '%') are substituted with | |
181 | their actual values. | |
182 | . backward compatibility with core_uses_pid: | |
183 | If core_pattern does not include "%p" (default does not) | |
184 | and core_uses_pid is set, then .PID will be appended to | |
185 | the filename. | |
186 | . corename format specifiers: | |
187 | %<NUL> '%' is dropped | |
188 | %% output one '%' | |
189 | %p pid | |
65aafb1e | 190 | %P global pid (init PID namespace) |
1da177e4 LT |
191 | %u uid |
192 | %g gid | |
12a2b4b2 ON |
193 | %d dump mode, matches PR_SET_DUMPABLE and |
194 | /proc/sys/fs/suid_dumpable | |
1da177e4 LT |
195 | %s signal number |
196 | %t UNIX time of dump | |
197 | %h hostname | |
57cc083a JS |
198 | %e executable filename (may be shortened) |
199 | %E executable path | |
1da177e4 | 200 | %<OTHER> both are dropped |
cd081041 MU |
201 | . If the first character of the pattern is a '|', the kernel will treat |
202 | the rest of the pattern as a command to run. The core dump will be | |
203 | written to the standard input of that program instead of to a file. | |
1da177e4 LT |
204 | |
205 | ============================================================== | |
206 | ||
a293980c NH |
207 | core_pipe_limit: |
208 | ||
807094c0 BP |
209 | This sysctl is only applicable when core_pattern is configured to pipe |
210 | core files to a user space helper (when the first character of | |
211 | core_pattern is a '|', see above). When collecting cores via a pipe | |
212 | to an application, it is occasionally useful for the collecting | |
213 | application to gather data about the crashing process from its | |
214 | /proc/pid directory. In order to do this safely, the kernel must wait | |
215 | for the collecting process to exit, so as not to remove the crashing | |
216 | processes proc files prematurely. This in turn creates the | |
217 | possibility that a misbehaving userspace collecting process can block | |
218 | the reaping of a crashed process simply by never exiting. This sysctl | |
219 | defends against that. It defines how many concurrent crashing | |
220 | processes may be piped to user space applications in parallel. If | |
221 | this value is exceeded, then those crashing processes above that value | |
222 | are noted via the kernel log and their cores are skipped. 0 is a | |
223 | special value, indicating that unlimited processes may be captured in | |
224 | parallel, but that no waiting will take place (i.e. the collecting | |
225 | process is not guaranteed access to /proc/<crashing pid>/). This | |
226 | value defaults to 0. | |
a293980c NH |
227 | |
228 | ============================================================== | |
229 | ||
1da177e4 LT |
230 | core_uses_pid: |
231 | ||
232 | The default coredump filename is "core". By setting | |
233 | core_uses_pid to 1, the coredump filename becomes core.PID. | |
234 | If core_pattern does not include "%p" (default does not) | |
235 | and core_uses_pid is set, then .PID will be appended to | |
236 | the filename. | |
237 | ||
238 | ============================================================== | |
239 | ||
240 | ctrl-alt-del: | |
241 | ||
242 | When the value in this file is 0, ctrl-alt-del is trapped and | |
243 | sent to the init(1) program to handle a graceful restart. | |
244 | When, however, the value is > 0, Linux's reaction to a Vulcan | |
245 | Nerve Pinch (tm) will be an immediate reboot, without even | |
246 | syncing its dirty buffers. | |
247 | ||
248 | Note: when a program (like dosemu) has the keyboard in 'raw' | |
249 | mode, the ctrl-alt-del is intercepted by the program before it | |
250 | ever reaches the kernel tty layer, and it's up to the program | |
251 | to decide what to do with it. | |
252 | ||
253 | ============================================================== | |
254 | ||
eaf06b24 DR |
255 | dmesg_restrict: |
256 | ||
807094c0 BP |
257 | This toggle indicates whether unprivileged users are prevented |
258 | from using dmesg(8) to view messages from the kernel's log buffer. | |
259 | When dmesg_restrict is set to (0) there are no restrictions. When | |
38ef4c2e | 260 | dmesg_restrict is set set to (1), users must have CAP_SYSLOG to use |
eaf06b24 DR |
261 | dmesg(8). |
262 | ||
807094c0 BP |
263 | The kernel config option CONFIG_SECURITY_DMESG_RESTRICT sets the |
264 | default value of dmesg_restrict. | |
eaf06b24 DR |
265 | |
266 | ============================================================== | |
267 | ||
1da177e4 LT |
268 | domainname & hostname: |
269 | ||
270 | These files can be used to set the NIS/YP domainname and the | |
271 | hostname of your box in exactly the same way as the commands | |
272 | domainname and hostname, i.e.: | |
273 | # echo "darkstar" > /proc/sys/kernel/hostname | |
274 | # echo "mydomain" > /proc/sys/kernel/domainname | |
275 | has the same effect as | |
276 | # hostname "darkstar" | |
277 | # domainname "mydomain" | |
278 | ||
279 | Note, however, that the classic darkstar.frop.org has the | |
280 | hostname "darkstar" and DNS (Internet Domain Name Server) | |
281 | domainname "frop.org", not to be confused with the NIS (Network | |
282 | Information Service) or YP (Yellow Pages) domainname. These two | |
283 | domain names are in general different. For a detailed discussion | |
284 | see the hostname(1) man page. | |
285 | ||
286 | ============================================================== | |
287 | ||
288 | hotplug: | |
289 | ||
290 | Path for the hotplug policy agent. | |
291 | Default value is "/sbin/hotplug". | |
292 | ||
293 | ============================================================== | |
294 | ||
270750db AT |
295 | hung_task_panic: |
296 | ||
297 | Controls the kernel's behavior when a hung task is detected. | |
298 | This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. | |
299 | ||
300 | 0: continue operation. This is the default behavior. | |
301 | ||
302 | 1: panic immediately. | |
303 | ||
304 | ============================================================== | |
305 | ||
306 | hung_task_check_count: | |
307 | ||
308 | The upper bound on the number of tasks that are checked. | |
309 | This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. | |
310 | ||
311 | ============================================================== | |
312 | ||
313 | hung_task_timeout_secs: | |
314 | ||
315 | Check interval. When a task in D state did not get scheduled | |
316 | for more than this value report a warning. | |
317 | This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. | |
318 | ||
319 | 0: means infinite timeout - no checking done. | |
320 | ||
321 | ============================================================== | |
322 | ||
323 | hung_task_warning: | |
324 | ||
325 | The maximum number of warnings to report. During a check interval | |
326 | When this value is reached, no more the warnings will be reported. | |
327 | This file shows up if CONFIG_DETECT_HUNG_TASK is enabled. | |
328 | ||
329 | -1: report an infinite number of warnings. | |
330 | ||
331 | ============================================================== | |
332 | ||
7984754b KC |
333 | kexec_load_disabled: |
334 | ||
335 | A toggle indicating if the kexec_load syscall has been disabled. This | |
336 | value defaults to 0 (false: kexec_load enabled), but can be set to 1 | |
337 | (true: kexec_load disabled). Once true, kexec can no longer be used, and | |
338 | the toggle cannot be set back to false. This allows a kexec image to be | |
339 | loaded before disabling the syscall, allowing a system to set up (and | |
340 | later use) an image without it being altered. Generally used together | |
341 | with the "modules_disabled" sysctl. | |
342 | ||
343 | ============================================================== | |
344 | ||
455cd5ab DR |
345 | kptr_restrict: |
346 | ||
347 | This toggle indicates whether restrictions are placed on | |
312b4e22 RM |
348 | exposing kernel addresses via /proc and other interfaces. |
349 | ||
350 | When kptr_restrict is set to (0), the default, there are no restrictions. | |
351 | ||
352 | When kptr_restrict is set to (1), kernel pointers printed using the %pK | |
353 | format specifier will be replaced with 0's unless the user has CAP_SYSLOG | |
354 | and effective user and group ids are equal to the real ids. This is | |
355 | because %pK checks are done at read() time rather than open() time, so | |
356 | if permissions are elevated between the open() and the read() (e.g via | |
357 | a setuid binary) then %pK will not leak kernel pointers to unprivileged | |
358 | users. Note, this is a temporary solution only. The correct long-term | |
359 | solution is to do the permission checks at open() time. Consider removing | |
360 | world read permissions from files that use %pK, and using dmesg_restrict | |
361 | to protect against uses of %pK in dmesg(8) if leaking kernel pointer | |
362 | values to unprivileged users is a concern. | |
363 | ||
364 | When kptr_restrict is set to (2), kernel pointers printed using | |
365 | %pK will be replaced with 0's regardless of privileges. | |
455cd5ab DR |
366 | |
367 | ============================================================== | |
368 | ||
0741f4d2 CE |
369 | kstack_depth_to_print: (X86 only) |
370 | ||
371 | Controls the number of words to print when dumping the raw | |
372 | kernel stack. | |
373 | ||
374 | ============================================================== | |
375 | ||
807094c0 BP |
376 | l2cr: (PPC only) |
377 | ||
378 | This flag controls the L2 cache of G3 processor boards. If | |
379 | 0, the cache is disabled. Enabled if nonzero. | |
380 | ||
381 | ============================================================== | |
382 | ||
3d43321b KC |
383 | modules_disabled: |
384 | ||
385 | A toggle value indicating if modules are allowed to be loaded | |
386 | in an otherwise modular kernel. This toggle defaults to off | |
387 | (0), but can be set true (1). Once true, modules can be | |
388 | neither loaded nor unloaded, and the toggle cannot be set back | |
7984754b | 389 | to false. Generally used with the "kexec_load_disabled" toggle. |
3d43321b KC |
390 | |
391 | ============================================================== | |
392 | ||
03f59566 SK |
393 | msg_next_id, sem_next_id, and shm_next_id: |
394 | ||
395 | These three toggles allows to specify desired id for next allocated IPC | |
396 | object: message, semaphore or shared memory respectively. | |
397 | ||
398 | By default they are equal to -1, which means generic allocation logic. | |
399 | Possible values to set are in range {0..INT_MAX}. | |
400 | ||
401 | Notes: | |
402 | 1) kernel doesn't guarantee, that new object will have desired id. So, | |
403 | it's up to userspace, how to handle an object with "wrong" id. | |
404 | 2) Toggle with non-default value will be set back to -1 by kernel after | |
405 | successful IPC object allocation. | |
406 | ||
407 | ============================================================== | |
408 | ||
807094c0 BP |
409 | nmi_watchdog: |
410 | ||
411 | Enables/Disables the NMI watchdog on x86 systems. When the value is | |
412 | non-zero the NMI watchdog is enabled and will continuously test all | |
413 | online cpus to determine whether or not they are still functioning | |
414 | properly. Currently, passing "nmi_watchdog=" parameter at boot time is | |
415 | required for this function to work. | |
416 | ||
417 | If LAPIC NMI watchdog method is in use (nmi_watchdog=2 kernel | |
418 | parameter), the NMI watchdog shares registers with oprofile. By | |
419 | disabling the NMI watchdog, oprofile may have more registers to | |
420 | utilize. | |
421 | ||
422 | ============================================================== | |
423 | ||
10fc05d0 MG |
424 | numa_balancing |
425 | ||
426 | Enables/disables automatic page fault based NUMA memory | |
427 | balancing. Memory is moved automatically to nodes | |
428 | that access it often. | |
429 | ||
430 | Enables/disables automatic NUMA memory balancing. On NUMA machines, there | |
431 | is a performance penalty if remote memory is accessed by a CPU. When this | |
432 | feature is enabled the kernel samples what task thread is accessing memory | |
433 | by periodically unmapping pages and later trapping a page fault. At the | |
434 | time of the page fault, it is determined if the data being accessed should | |
435 | be migrated to a local memory node. | |
436 | ||
437 | The unmapping of pages and trapping faults incur additional overhead that | |
438 | ideally is offset by improved memory locality but there is no universal | |
439 | guarantee. If the target workload is already bound to NUMA nodes then this | |
440 | feature should be disabled. Otherwise, if the system overhead from the | |
441 | feature is too high then the rate the kernel samples for NUMA hinting | |
442 | faults may be controlled by the numa_balancing_scan_period_min_ms, | |
930aa174 | 443 | numa_balancing_scan_delay_ms, numa_balancing_scan_period_max_ms, |
de1c9ce6 RR |
444 | numa_balancing_scan_size_mb, numa_balancing_settle_count sysctls and |
445 | numa_balancing_migrate_deferred. | |
10fc05d0 MG |
446 | |
447 | ============================================================== | |
448 | ||
449 | numa_balancing_scan_period_min_ms, numa_balancing_scan_delay_ms, | |
930aa174 | 450 | numa_balancing_scan_period_max_ms, numa_balancing_scan_size_mb |
10fc05d0 MG |
451 | |
452 | Automatic NUMA balancing scans tasks address space and unmaps pages to | |
453 | detect if pages are properly placed or if the data should be migrated to a | |
454 | memory node local to where the task is running. Every "scan delay" the task | |
455 | scans the next "scan size" number of pages in its address space. When the | |
456 | end of the address space is reached the scanner restarts from the beginning. | |
457 | ||
458 | In combination, the "scan delay" and "scan size" determine the scan rate. | |
459 | When "scan delay" decreases, the scan rate increases. The scan delay and | |
460 | hence the scan rate of every task is adaptive and depends on historical | |
461 | behaviour. If pages are properly placed then the scan delay increases, | |
462 | otherwise the scan delay decreases. The "scan size" is not adaptive but | |
463 | the higher the "scan size", the higher the scan rate. | |
464 | ||
465 | Higher scan rates incur higher system overhead as page faults must be | |
466 | trapped and potentially data must be migrated. However, the higher the scan | |
467 | rate, the more quickly a tasks memory is migrated to a local node if the | |
468 | workload pattern changes and minimises performance impact due to remote | |
469 | memory accesses. These sysctls control the thresholds for scan delays and | |
470 | the number of pages scanned. | |
471 | ||
598f0ec0 MG |
472 | numa_balancing_scan_period_min_ms is the minimum time in milliseconds to |
473 | scan a tasks virtual memory. It effectively controls the maximum scanning | |
474 | rate for each task. | |
10fc05d0 MG |
475 | |
476 | numa_balancing_scan_delay_ms is the starting "scan delay" used for a task | |
477 | when it initially forks. | |
478 | ||
598f0ec0 MG |
479 | numa_balancing_scan_period_max_ms is the maximum time in milliseconds to |
480 | scan a tasks virtual memory. It effectively controls the minimum scanning | |
481 | rate for each task. | |
10fc05d0 MG |
482 | |
483 | numa_balancing_scan_size_mb is how many megabytes worth of pages are | |
484 | scanned for a given scan. | |
485 | ||
de1c9ce6 RR |
486 | numa_balancing_migrate_deferred is how many page migrations get skipped |
487 | unconditionally, after a page migration is skipped because a page is shared | |
488 | with other tasks. This reduces page migration overhead, and determines | |
489 | how much stronger the "move task near its memory" policy scheduler becomes, | |
490 | versus the "move memory near its task" memory management policy, for workloads | |
491 | with shared memory. | |
492 | ||
10fc05d0 MG |
493 | ============================================================== |
494 | ||
1da177e4 LT |
495 | osrelease, ostype & version: |
496 | ||
497 | # cat osrelease | |
498 | 2.1.88 | |
499 | # cat ostype | |
500 | Linux | |
501 | # cat version | |
502 | #5 Wed Feb 25 21:49:24 MET 1998 | |
503 | ||
504 | The files osrelease and ostype should be clear enough. Version | |
505 | needs a little more clarification however. The '#5' means that | |
506 | this is the fifth kernel built from this source base and the | |
507 | date behind it indicates the time the kernel was built. | |
508 | The only way to tune these values is to rebuild the kernel :-) | |
509 | ||
510 | ============================================================== | |
511 | ||
512 | overflowgid & overflowuid: | |
513 | ||
807094c0 BP |
514 | if your architecture did not always support 32-bit UIDs (i.e. arm, |
515 | i386, m68k, sh, and sparc32), a fixed UID and GID will be returned to | |
516 | applications that use the old 16-bit UID/GID system calls, if the | |
517 | actual UID or GID would exceed 65535. | |
1da177e4 LT |
518 | |
519 | These sysctls allow you to change the value of the fixed UID and GID. | |
520 | The default is 65534. | |
521 | ||
522 | ============================================================== | |
523 | ||
524 | panic: | |
525 | ||
807094c0 BP |
526 | The value in this file represents the number of seconds the kernel |
527 | waits before rebooting on a panic. When you use the software watchdog, | |
528 | the recommended setting is 60. | |
529 | ||
530 | ============================================================== | |
531 | ||
532 | panic_on_unrecovered_nmi: | |
533 | ||
534 | The default Linux behaviour on an NMI of either memory or unknown is | |
535 | to continue operation. For many environments such as scientific | |
536 | computing it is preferable that the box is taken out and the error | |
537 | dealt with than an uncorrected parity/ECC error get propagated. | |
538 | ||
539 | A small number of systems do generate NMI's for bizarre random reasons | |
540 | such as power management so the default is off. That sysctl works like | |
541 | the existing panic controls already in that directory. | |
1da177e4 LT |
542 | |
543 | ============================================================== | |
544 | ||
545 | panic_on_oops: | |
546 | ||
547 | Controls the kernel's behaviour when an oops or BUG is encountered. | |
548 | ||
549 | 0: try to continue operation | |
550 | ||
a982ac06 | 551 | 1: panic immediately. If the `panic' sysctl is also non-zero then the |
8b23d04d | 552 | machine will be rebooted. |
1da177e4 LT |
553 | |
554 | ============================================================== | |
555 | ||
55af7796 MH |
556 | panic_on_stackoverflow: |
557 | ||
558 | Controls the kernel's behavior when detecting the overflows of | |
559 | kernel, IRQ and exception stacks except a user stack. | |
560 | This file shows up if CONFIG_DEBUG_STACKOVERFLOW is enabled. | |
561 | ||
562 | 0: try to continue operation. | |
563 | ||
564 | 1: panic immediately. | |
565 | ||
566 | ============================================================== | |
567 | ||
14c63f17 DH |
568 | perf_cpu_time_max_percent: |
569 | ||
570 | Hints to the kernel how much CPU time it should be allowed to | |
571 | use to handle perf sampling events. If the perf subsystem | |
572 | is informed that its samples are exceeding this limit, it | |
573 | will drop its sampling frequency to attempt to reduce its CPU | |
574 | usage. | |
575 | ||
576 | Some perf sampling happens in NMIs. If these samples | |
577 | unexpectedly take too long to execute, the NMIs can become | |
578 | stacked up next to each other so much that nothing else is | |
579 | allowed to execute. | |
580 | ||
581 | 0: disable the mechanism. Do not monitor or correct perf's | |
582 | sampling rate no matter how CPU time it takes. | |
583 | ||
584 | 1-100: attempt to throttle perf's sample rate to this | |
585 | percentage of CPU. Note: the kernel calculates an | |
586 | "expected" length of each sample event. 100 here means | |
587 | 100% of that expected length. Even if this is set to | |
588 | 100, you may still see sample throttling if this | |
589 | length is exceeded. Set to 0 if you truly do not care | |
590 | how much CPU is consumed. | |
591 | ||
592 | ============================================================== | |
593 | ||
55af7796 | 594 | |
1da177e4 LT |
595 | pid_max: |
596 | ||
beb7dd86 | 597 | PID allocation wrap value. When the kernel's next PID value |
1da177e4 LT |
598 | reaches this value, it wraps back to a minimum PID value. |
599 | PIDs of value pid_max or larger are not allocated. | |
600 | ||
601 | ============================================================== | |
602 | ||
b8f566b0 PE |
603 | ns_last_pid: |
604 | ||
605 | The last pid allocated in the current (the one task using this sysctl | |
606 | lives in) pid namespace. When selecting a pid for a next task on fork | |
607 | kernel tries to allocate a number starting from this one. | |
608 | ||
609 | ============================================================== | |
610 | ||
1da177e4 LT |
611 | powersave-nap: (PPC only) |
612 | ||
613 | If set, Linux-PPC will use the 'nap' mode of powersaving, | |
614 | otherwise the 'doze' mode will be used. | |
615 | ||
616 | ============================================================== | |
617 | ||
618 | printk: | |
619 | ||
620 | The four values in printk denote: console_loglevel, | |
621 | default_message_loglevel, minimum_console_loglevel and | |
622 | default_console_loglevel respectively. | |
623 | ||
624 | These values influence printk() behavior when printing or | |
625 | logging error messages. See 'man 2 syslog' for more info on | |
626 | the different loglevels. | |
627 | ||
628 | - console_loglevel: messages with a higher priority than | |
629 | this will be printed to the console | |
87889e15 | 630 | - default_message_loglevel: messages without an explicit priority |
1da177e4 LT |
631 | will be printed with this priority |
632 | - minimum_console_loglevel: minimum (highest) value to which | |
633 | console_loglevel can be set | |
634 | - default_console_loglevel: default value for console_loglevel | |
635 | ||
636 | ============================================================== | |
637 | ||
807094c0 BP |
638 | printk_delay: |
639 | ||
640 | Delay each printk message in printk_delay milliseconds | |
641 | ||
642 | Value from 0 - 10000 is allowed. | |
643 | ||
644 | ============================================================== | |
645 | ||
1da177e4 LT |
646 | printk_ratelimit: |
647 | ||
648 | Some warning messages are rate limited. printk_ratelimit specifies | |
649 | the minimum length of time between these messages (in jiffies), by | |
650 | default we allow one every 5 seconds. | |
651 | ||
652 | A value of 0 will disable rate limiting. | |
653 | ||
654 | ============================================================== | |
655 | ||
656 | printk_ratelimit_burst: | |
657 | ||
658 | While long term we enforce one message per printk_ratelimit | |
659 | seconds, we do allow a burst of messages to pass through. | |
660 | printk_ratelimit_burst specifies the number of messages we can | |
661 | send before ratelimiting kicks in. | |
662 | ||
663 | ============================================================== | |
664 | ||
807094c0 | 665 | randomize_va_space: |
1ec7fd50 JK |
666 | |
667 | This option can be used to select the type of process address | |
668 | space randomization that is used in the system, for architectures | |
669 | that support this feature. | |
670 | ||
b7f5ab6f HS |
671 | 0 - Turn the process address space randomization off. This is the |
672 | default for architectures that do not support this feature anyways, | |
673 | and kernels that are booted with the "norandmaps" parameter. | |
1ec7fd50 JK |
674 | |
675 | 1 - Make the addresses of mmap base, stack and VDSO page randomized. | |
676 | This, among other things, implies that shared libraries will be | |
b7f5ab6f HS |
677 | loaded to random addresses. Also for PIE-linked binaries, the |
678 | location of code start is randomized. This is the default if the | |
679 | CONFIG_COMPAT_BRK option is enabled. | |
1ec7fd50 | 680 | |
b7f5ab6f HS |
681 | 2 - Additionally enable heap randomization. This is the default if |
682 | CONFIG_COMPAT_BRK is disabled. | |
683 | ||
684 | There are a few legacy applications out there (such as some ancient | |
1ec7fd50 | 685 | versions of libc.so.5 from 1996) that assume that brk area starts |
b7f5ab6f HS |
686 | just after the end of the code+bss. These applications break when |
687 | start of the brk area is randomized. There are however no known | |
1ec7fd50 | 688 | non-legacy applications that would be broken this way, so for most |
b7f5ab6f HS |
689 | systems it is safe to choose full randomization. |
690 | ||
691 | Systems with ancient and/or broken binaries should be configured | |
692 | with CONFIG_COMPAT_BRK enabled, which excludes the heap from process | |
693 | address space randomization. | |
1ec7fd50 JK |
694 | |
695 | ============================================================== | |
696 | ||
1da177e4 LT |
697 | reboot-cmd: (Sparc only) |
698 | ||
699 | ??? This seems to be a way to give an argument to the Sparc | |
700 | ROM/Flash boot loader. Maybe to tell it what to do after | |
701 | rebooting. ??? | |
702 | ||
703 | ============================================================== | |
704 | ||
705 | rtsig-max & rtsig-nr: | |
706 | ||
707 | The file rtsig-max can be used to tune the maximum number | |
708 | of POSIX realtime (queued) signals that can be outstanding | |
709 | in the system. | |
710 | ||
711 | rtsig-nr shows the number of RT signals currently queued. | |
712 | ||
713 | ============================================================== | |
714 | ||
715 | sg-big-buff: | |
716 | ||
717 | This file shows the size of the generic SCSI (sg) buffer. | |
718 | You can't tune it just yet, but you could change it on | |
719 | compile time by editing include/scsi/sg.h and changing | |
720 | the value of SG_BIG_BUFF. | |
721 | ||
722 | There shouldn't be any reason to change this value. If | |
723 | you can come up with one, you probably know what you | |
724 | are doing anyway :) | |
725 | ||
726 | ============================================================== | |
727 | ||
358e419f CALP |
728 | shmall: |
729 | ||
730 | This parameter sets the total amount of shared memory pages that | |
731 | can be used system wide. Hence, SHMALL should always be at least | |
732 | ceil(shmmax/PAGE_SIZE). | |
733 | ||
734 | If you are not sure what the default PAGE_SIZE is on your Linux | |
735 | system, you can run the following command: | |
736 | ||
737 | # getconf PAGE_SIZE | |
738 | ||
739 | ============================================================== | |
740 | ||
807094c0 | 741 | shmmax: |
1da177e4 LT |
742 | |
743 | This value can be used to query and set the run time limit | |
744 | on the maximum shared memory segment size that can be created. | |
807094c0 | 745 | Shared memory segments up to 1Gb are now supported in the |
1da177e4 LT |
746 | kernel. This value defaults to SHMMAX. |
747 | ||
748 | ============================================================== | |
749 | ||
b34a6b1d VK |
750 | shm_rmid_forced: |
751 | ||
752 | Linux lets you set resource limits, including how much memory one | |
753 | process can consume, via setrlimit(2). Unfortunately, shared memory | |
754 | segments are allowed to exist without association with any process, and | |
755 | thus might not be counted against any resource limits. If enabled, | |
756 | shared memory segments are automatically destroyed when their attach | |
757 | count becomes zero after a detach or a process termination. It will | |
758 | also destroy segments that were created, but never attached to, on exit | |
759 | from the process. The only use left for IPC_RMID is to immediately | |
760 | destroy an unattached segment. Of course, this breaks the way things are | |
761 | defined, so some applications might stop working. Note that this | |
762 | feature will do you no good unless you also configure your resource | |
763 | limits (in particular, RLIMIT_AS and RLIMIT_NPROC). Most systems don't | |
764 | need this. | |
765 | ||
766 | Note that if you change this from 0 to 1, already created segments | |
767 | without users and with a dead originative process will be destroyed. | |
768 | ||
769 | ============================================================== | |
770 | ||
807094c0 | 771 | tainted: |
1da177e4 LT |
772 | |
773 | Non-zero if the kernel has been tainted. Numeric values, which | |
774 | can be ORed together: | |
775 | ||
bb20698d GKH |
776 | 1 - A module with a non-GPL license has been loaded, this |
777 | includes modules with no license. | |
778 | Set by modutils >= 2.4.9 and module-init-tools. | |
779 | 2 - A module was force loaded by insmod -f. | |
780 | Set by modutils >= 2.4.9 and module-init-tools. | |
781 | 4 - Unsafe SMP processors: SMP with CPUs not designed for SMP. | |
782 | 8 - A module was forcibly unloaded from the system by rmmod -f. | |
783 | 16 - A hardware machine check error occurred on the system. | |
784 | 32 - A bad page was discovered on the system. | |
785 | 64 - The user has asked that the system be marked "tainted". This | |
786 | could be because they are running software that directly modifies | |
787 | the hardware, or for other reasons. | |
788 | 128 - The system has died. | |
789 | 256 - The ACPI DSDT has been overridden with one supplied by the user | |
790 | instead of using the one provided by the hardware. | |
791 | 512 - A kernel warning has occurred. | |
792 | 1024 - A module from drivers/staging was loaded. | |
f5fe184b LF |
793 | 2048 - The system is working around a severe firmware bug. |
794 | 4096 - An out-of-tree module has been loaded. | |
66cc69e3 MD |
795 | 8192 - An unsigned module has been loaded in a kernel supporting module |
796 | signature. | |
1da177e4 | 797 | |
760df93e SF |
798 | ============================================================== |
799 | ||
760df93e SF |
800 | unknown_nmi_panic: |
801 | ||
807094c0 BP |
802 | The value in this file affects behavior of handling NMI. When the |
803 | value is non-zero, unknown NMI is trapped and then panic occurs. At | |
804 | that time, kernel debugging information is displayed on console. | |
760df93e | 805 | |
807094c0 BP |
806 | NMI switch that most IA32 servers have fires unknown NMI up, for |
807 | example. If a system hangs up, try pressing the NMI switch. | |
08825c90 LZ |
808 | |
809 | ============================================================== | |
810 | ||
811 | watchdog_thresh: | |
812 | ||
813 | This value can be used to control the frequency of hrtimer and NMI | |
814 | events and the soft and hard lockup thresholds. The default threshold | |
815 | is 10 seconds. | |
816 | ||
817 | The softlockup threshold is (2 * watchdog_thresh). Setting this | |
818 | tunable to zero will disable lockup detection altogether. | |
819 | ||
820 | ============================================================== |