xref: /freebsd/lib/libc/gen/sysctl.3 (revision 2546665afcaf0d53dc2c7058fee96354b3680f5a)
1.\" Copyright (c) 1993
2.\"	The Regents of the University of California.  All rights reserved.
3.\"
4.\" Redistribution and use in source and binary forms, with or without
5.\" modification, are permitted provided that the following conditions
6.\" are met:
7.\" 1. Redistributions of source code must retain the above copyright
8.\"    notice, this list of conditions and the following disclaimer.
9.\" 2. Redistributions in binary form must reproduce the above copyright
10.\"    notice, this list of conditions and the following disclaimer in the
11.\"    documentation and/or other materials provided with the distribution.
12.\" 3. All advertising materials mentioning features or use of this software
13.\"    must display the following acknowledgement:
14.\"	This product includes software developed by the University of
15.\"	California, Berkeley and its contributors.
16.\" 4. Neither the name of the University nor the names of its contributors
17.\"    may be used to endorse or promote products derived from this software
18.\"    without specific prior written permission.
19.\"
20.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30.\" SUCH DAMAGE.
31.\"
32.\"	@(#)sysctl.3	8.4 (Berkeley) 5/9/95
33.\" $FreeBSD$
34.\"
35.Dd January 23, 2001
36.Dt SYSCTL 3
37.Os
38.Sh NAME
39.Nm sysctl ,
40.Nm sysctlbyname ,
41.Nm sysctlnametomib
42.Nd get or set system information
43.Sh LIBRARY
44.Lb libc
45.Sh SYNOPSIS
46.In sys/types.h
47.In sys/sysctl.h
48.Ft int
49.Fn sysctl "int *name" "u_int namelen" "void *oldp" "size_t *oldlenp" "void *newp" "size_t newlen"
50.Ft int
51.Fn sysctlbyname "const char *name" "void *oldp" "size_t *oldlenp" "void *newp" "size_t newlen"
52.Ft int
53.Fn sysctlnametomib "const char *name" "int *mibp" "size_t *sizep"
54.Sh DESCRIPTION
55The
56.Fn sysctl
57function retrieves system information and allows processes with
58appropriate privileges to set system information.
59The information available from
60.Fn sysctl
61consists of integers, strings, and tables.
62Information may be retrieved and set from the command interface
63using the
64.Xr sysctl 8
65utility.
66.Pp
67Unless explicitly noted below,
68.Fn sysctl
69returns a consistent snapshot of the data requested.
70Consistency is obtained by locking the destination
71buffer into memory so that the data may be copied out without blocking.
72Calls to
73.Fn sysctl
74are serialized to avoid deadlock.
75.Pp
76The state is described using a ``Management Information Base'' (MIB)
77style name, listed in
78.Fa name ,
79which is a
80.Fa namelen
81length array of integers.
82.Pp
83The
84.Fn sysctlbyname
85function accepts an ASCII representation of the name and internally
86looks up the integer name vector.
87Apart from that, it behaves the same
88as the standard
89.Fn sysctl
90function.
91.Pp
92The information is copied into the buffer specified by
93.Fa oldp .
94The size of the buffer is given by the location specified by
95.Fa oldlenp
96before the call,
97and that location gives the amount of data copied after a successful call
98and after a call that returns with the error code
99.Er ENOMEM .
100If the amount of data available is greater
101than the size of the buffer supplied,
102the call supplies as much data as fits in the buffer provided
103and returns with the error code
104.Er ENOMEM .
105If the old value is not desired,
106.Fa oldp
107and
108.Fa oldlenp
109should be set to NULL.
110.Pp
111The size of the available data can be determined by calling
112.Fn sysctl
113with the
114.Dv NULL
115argument for
116.Fa oldp .
117The size of the available data will be returned in the location pointed to by
118.Fa oldlenp .
119For some operations, the amount of space may change often.
120For these operations,
121the system attempts to round up so that the returned size is
122large enough for a call to return the data shortly thereafter.
123.Pp
124To set a new value,
125.Fa newp
126is set to point to a buffer of length
127.Fa newlen
128from which the requested value is to be taken.
129If a new value is not to be set,
130.Fa newp
131should be set to NULL and
132.Fa newlen
133set to 0.
134.Pp
135The
136.Fn sysctlnametomib
137function accepts an ASCII representation of the name,
138looks up the integer name vector,
139and returns the numeric representation in the mib array pointed to by
140.Fa mibp .
141The number of elements in the mib array is given by the location specified by
142.Fa sizep
143before the call,
144and that location gives the number of entries copied after a successful call.
145The resulting
146.Fa mib
147and
148.Fa size
149may be used in subsequent
150.Fn sysctl
151calls to get the data associated with the requested ASCII name.
152This interface is intended for use by applications that want to
153repeatedly request the same variable (the
154.Fn sysctl
155function runs in about a third the time as the same request made via the
156.Fn sysctlbyname
157function).
158The
159.Fn sysctlnametomib
160function is also useful for fetching mib prefixes and then adding
161a final component.
162For example, to fetch process information
163for processes with pid's less than 100:
164.Pp
165.Bd -literal -offset indent -compact
166int i, mib[4];
167size_t len;
168struct kinfo_proc kp;
169
170/* Fill out the first three components of the mib */
171len = 4;
172sysctlnametomib("kern.proc.pid", mib, &len);
173
174/* Fetch and print entries for pid's < 100 */
175for (i = 0; i < 100; i++) {
176	mib[3] = i;
177	len = sizeof(kp);
178	if (sysctl(mib, 4, &kp, &len, NULL, 0) == -1)
179		perror("sysctl");
180	else if (len > 0)
181		printkproc(&kp);
182}
183.Ed
184.Pp
185The top level names are defined with a CTL_ prefix in
186.In sys/sysctl.h ,
187and are as follows.
188The next and subsequent levels down are found in the include files
189listed here, and described in separate sections below.
190.Pp
191.Bl -column CTLXMACHDEPXXX "Next level namesXXXXXX" -offset indent
192.It Sy "Name	Next level names	Description"
193.It "CTL_DEBUG	sys/sysctl.h	Debugging"
194.It "CTL_VFS	sys/mount.h	File system"
195.It "CTL_HW	sys/sysctl.h	Generic CPU, I/O"
196.It "CTL_KERN	sys/sysctl.h	High kernel limits"
197.It "CTL_MACHDEP	sys/sysctl.h	Machine dependent"
198.It "CTL_NET	sys/socket.h	Networking"
199.It "CTL_USER	sys/sysctl.h	User-level"
200.It "CTL_VM	vm/vm_param.h	Virtual memory"
201.El
202.Pp
203For example, the following retrieves the maximum number of processes allowed
204in the system:
205.Pp
206.Bd -literal -offset indent -compact
207int mib[2], maxproc;
208size_t len;
209
210mib[0] = CTL_KERN;
211mib[1] = KERN_MAXPROC;
212len = sizeof(maxproc);
213sysctl(mib, 2, &maxproc, &len, NULL, 0);
214.Ed
215.Pp
216To retrieve the standard search path for the system utilities:
217.Pp
218.Bd -literal -offset indent -compact
219int mib[2];
220size_t len;
221char *p;
222
223mib[0] = CTL_USER;
224mib[1] = USER_CS_PATH;
225sysctl(mib, 2, NULL, &len, NULL, 0);
226p = malloc(len);
227sysctl(mib, 2, p, &len, NULL, 0);
228.Ed
229.Ss CTL_DEBUG
230The debugging variables vary from system to system.
231A debugging variable may be added or deleted without need to recompile
232.Fn sysctl
233to know about it.
234Each time it runs,
235.Fn sysctl
236gets the list of debugging variables from the kernel and
237displays their current values.
238The system defines twenty
239.Pq Vt "struct ctldebug"
240variables named
241.Va debug0
242through
243.Va debug19 .
244They are declared as separate variables so that they can be
245individually initialized at the location of their associated variable.
246The loader prevents multiple use of the same variable by issuing errors
247if a variable is initialized in more than one place.
248For example, to export the variable
249.Va dospecialcheck
250as a debugging variable, the following declaration would be used:
251.Pp
252.Bd -literal -offset indent -compact
253int dospecialcheck = 1;
254struct ctldebug debug5 = { "dospecialcheck", &dospecialcheck };
255.Ed
256.Ss CTL_VFS
257A distinguished second level name, VFS_GENERIC,
258is used to get general information about all file systems.
259One of its third level identifiers is VFS_MAXTYPENUM
260that gives the highest valid file system type number.
261Its other third level identifier is VFS_CONF that
262returns configuration information about the file system
263type given as a fourth level identifier (see
264.Xr getvfsbyname 3
265as an example of its use).
266The remaining second level identifiers are the
267file system type number returned by a
268.Xr statfs 2
269call or from VFS_CONF.
270The third level identifiers available for each file system
271are given in the header file that defines the mount
272argument structure for that file system.
273.Ss CTL_HW
274The string and integer information available for the CTL_HW level
275is detailed below.
276The changeable column shows whether a process with appropriate
277privilege may change the value.
278.Bl -column "Second level nameXXXXXX" integerXXX -offset indent
279.It Sy "Second level name	Type	Changeable"
280.It "HW_MACHINE	string	no"
281.It "HW_MODEL	string	no"
282.It "HW_NCPU	integer	no"
283.It "HW_BYTEORDER	integer	no"
284.It "HW_PHYSMEM	integer	no"
285.It "HW_USERMEM	integer	no"
286.It "HW_PAGESIZE	integer	no"
287.It "HW_FLOATINGPOINT	integer	no"
288.It "HW_MACHINE_ARCH	string	no"
289.\".It "HW_DISKNAMES	integer	no"
290.\".It "HW_DISKSTATS	integer	no"
291.El
292.Pp
293.Bl -tag -width 6n
294.It Li HW_MACHINE
295The machine class.
296.It Li HW_MODEL
297The machine model
298.It Li HW_NCPU
299The number of cpus.
300.It Li HW_BYTEORDER
301The byteorder (4,321, or 1,234).
302.It Li HW_PHYSMEM
303The bytes of physical memory.
304.It Li HW_USERMEM
305The bytes of non-kernel memory.
306.It Li HW_PAGESIZE
307The software page size.
308.It Li HW_FLOATINGPOINT
309Nonzero if the floating point support is in hardware.
310.It Li HW_MACHINE_ARCH
311The machine dependent architecture type.
312.\".It Fa HW_DISKNAMES
313.\".It Fa HW_DISKSTATS
314.El
315.Ss CTL_KERN
316The string and integer information available for the CTL_KERN level
317is detailed below.
318The changeable column shows whether a process with appropriate
319privilege may change the value.
320The types of data currently available are process information,
321system vnodes, the open file entries, routing table entries,
322virtual memory statistics, load average history, and clock rate
323information.
324.Bl -column "KERNXMAXFILESPERPROCXXX" "struct clockrateXXX" -offset indent
325.It Sy "Second level name	Type	Changeable"
326.It "KERN_ARGMAX	integer	no"
327.It "KERN_BOOTFILE	string	yes"
328.It "KERN_BOOTTIME	struct timeval	no"
329.It "KERN_CLOCKRATE	struct clockinfo	no"
330.It "KERN_FILE	struct file	no"
331.It "KERN_HOSTID	integer	yes"
332.It "KERN_HOSTNAME	string	yes"
333.It "KERN_JOB_CONTROL	integer	no"
334.It "KERN_MAXFILES	integer	yes"
335.It "KERN_MAXFILESPERPROC	integer	yes"
336.It "KERN_MAXPROC	integer	no"
337.It "KERN_MAXPROCPERUID	integer	yes"
338.It "KERN_MAXVNODES	integer	yes"
339.It "KERN_NGROUPS	integer	no"
340.It "KERN_NISDOMAINNAME	string	yes"
341.It "KERN_OSRELDATE	integer	no"
342.It "KERN_OSRELEASE	string	no"
343.It "KERN_OSREV	integer	no"
344.It "KERN_OSTYPE	string	no"
345.It "KERN_POSIX1	integer	no"
346.It "KERN_PROC	struct proc	no"
347.It "KERN_PROF	node	not applicable"
348.It "KERN_QUANTUM	integer	yes"
349.It "KERN_SAVED_IDS	integer	no"
350.It "KERN_SECURELVL	integer	raise only"
351.It "KERN_UPDATEINTERVAL	integer	no"
352.It "KERN_VERSION	string	no"
353.It "KERN_VNODE	struct vnode	no"
354.El
355.Pp
356.Bl -tag -width 6n
357.It Li KERN_ARGMAX
358The maximum bytes of argument to
359.Xr execve 2 .
360.It Li KERN_BOOTFILE
361The full pathname of the file from which the kernel was loaded.
362.It Li KERN_BOOTTIME
363A
364.Va struct timeval
365structure is returned.
366This structure contains the time that the system was booted.
367.It Li KERN_CLOCKRATE
368A
369.Va struct clockinfo
370structure is returned.
371This structure contains the clock, statistics clock and profiling clock
372frequencies, the number of micro-seconds per hz tick and the skew rate.
373.It Li KERN_FILE
374Return the entire file table.
375The returned data consists of a single
376.Va struct filehead
377followed by an array of
378.Va struct file ,
379whose size depends on the current number of such objects in the system.
380.It Li KERN_HOSTID
381Get or set the host id.
382.It Li KERN_HOSTNAME
383Get or set the hostname.
384.It Li KERN_JOB_CONTROL
385Return 1 if job control is available on this system, otherwise 0.
386.It Li KERN_MAXFILES
387The maximum number of files that may be open in the system.
388.It Li KERN_MAXFILESPERPROC
389The maximum number of files that may be open for a single process.
390This limit only applies to processes with an effective uid of nonzero
391at the time of the open request.
392Files that have already been opened are not affected if the limit
393or the effective uid is changed.
394.It Li KERN_MAXPROC
395The maximum number of concurrent processes the system will allow.
396.It Li KERN_MAXPROCPERUID
397The maximum number of concurrent processes the system will allow
398for a single effective uid.
399This limit only applies to processes with an effective uid of nonzero
400at the time of a fork request.
401Processes that have already been started are not affected if the limit
402is changed.
403.It Li KERN_MAXVNODES
404The maximum number of vnodes available on the system.
405.It Li KERN_NGROUPS
406The maximum number of supplemental groups.
407.It Li KERN_NISDOMAINNAME
408The name of the current YP/NIS domain.
409.It Li KERN_OSRELDATE
410The kernel release version in the format
411.Ar M Ns Ar mm Ns Ar R Ns Ar xx ,
412where
413.Ar M
414is the major version,
415.Ar mm
416is the two digit minor version,
417.Ar R
418is 0 if release branch, otherwise 1,
419and
420.Ar xx
421is updated when the available APIs change.
422.Pp
423The userland release version is available from
424.In osreldate.h ;
425parse this file if you need to get the release version of
426the currently installed userland.
427.It Li KERN_OSRELEASE
428The system release string.
429.It Li KERN_OSREV
430The system revision string.
431.It Li KERN_OSTYPE
432The system type string.
433.It Li KERN_POSIX1
434The version of
435.St -p1003.1
436with which the system
437attempts to comply.
438.It Li KERN_PROC
439Return the entire process table, or a subset of it.
440An array of pairs of
441.Va struct proc
442followed by corresponding
443.Va struct eproc
444structures is returned,
445whose size depends on the current number of such objects in the system.
446The third and fourth level names are as follows:
447.Bl -column "Third level nameXXXXXX" "Fourth level is:XXXXXX" -offset indent
448.It "Third level name	Fourth level is:"
449.It "KERN_PROC_ALL	None"
450.It "KERN_PROC_PID	A process ID"
451.It "KERN_PROC_PGRP	A process group"
452.It "KERN_PROC_TTY	A tty device"
453.It "KERN_PROC_UID	A user ID"
454.It "KERN_PROC_RUID	A real user ID"
455.El
456.Pp
457If the third level name is KERN_PROC_ARGS then the command line argument
458array is returned in a flattened form, i.e., zero-terminated arguments
459follow each other.
460The total size of array is returned.
461It is also possible for a process to set its own process title this way.
462.Bl -column "Third level nameXXXXXX" "Fourth level is:XXXXXX" -offset indent
463.It Sy "Third level name	Fourth level is:"
464.It "KERN_PROC_ARGS	A process ID"
465.El
466.It Li KERN_PROF
467Return profiling information about the kernel.
468If the kernel is not compiled for profiling,
469attempts to retrieve any of the KERN_PROF values will
470fail with
471.Er ENOENT .
472The third level names for the string and integer profiling information
473is detailed below.
474The changeable column shows whether a process with appropriate
475privilege may change the value.
476.Bl -column "GPROFXGMONPARAMXXX" "struct gmonparamXXX" -offset indent
477.It Sy "Third level name	Type	Changeable"
478.It "GPROF_STATE	integer	yes"
479.It "GPROF_COUNT	u_short[\|]	yes"
480.It "GPROF_FROMS	u_short[\|]	yes"
481.It "GPROF_TOS	struct tostruct	yes"
482.It "GPROF_GMONPARAM	struct gmonparam	no"
483.El
484.Pp
485The variables are as follows:
486.Bl -tag -width 6n
487.It Li GPROF_STATE
488Returns GMON_PROF_ON or GMON_PROF_OFF to show that profiling
489is running or stopped.
490.It Li GPROF_COUNT
491Array of statistical program counter counts.
492.It Li GPROF_FROMS
493Array indexed by program counter of call-from points.
494.It Li GPROF_TOS
495Array of
496.Va struct tostruct
497describing destination of calls and their counts.
498.It Li GPROF_GMONPARAM
499Structure giving the sizes of the above arrays.
500.El
501.It Li KERN_QUANTUM
502The maximum period of time, in microseconds, for which a process is allowed
503to run without being preempted if other processes are in the run queue.
504.It Li KERN_SAVED_IDS
505Returns 1 if saved set-group and saved set-user ID is available.
506.It Li KERN_SECURELVL
507The system security level.
508This level may be raised by processes with appropriate privilege.
509It may not be lowered.
510.It Li KERN_VERSION
511The system version string.
512.It Li KERN_VNODE
513Return the entire vnode table.
514Note, the vnode table is not necessarily a consistent snapshot of
515the system.
516The returned data consists of an array whose size depends on the
517current number of such objects in the system.
518Each element of the array contains the kernel address of a vnode
519.Va struct vnode *
520followed by the vnode itself
521.Va struct vnode .
522.El
523.Ss CTL_MACHDEP
524The set of variables defined is architecture dependent.
525The following variables are defined for the i386 architecture.
526.Bl -column "CONSOLE_DEVICEXXX" "struct bootinfoXXX" -offset indent
527.It Sy "Second level name	Type	Changeable"
528.It Li "CPU_CONSDEV	dev_t	no"
529.It Li "CPU_ADJKERNTZ	int	yes"
530.It Li "CPU_DISRTCSET	int	yes"
531.It Li "CPU_BOOTINFO	struct bootinfo	no"
532.It Li "CPU_WALLCLOCK	int	yes"
533.El
534.Ss CTL_NET
535The string and integer information available for the CTL_NET level
536is detailed below.
537The changeable column shows whether a process with appropriate
538privilege may change the value.
539.Bl -column "Second level nameXXXXXX" "routing messagesXXX" -offset indent
540.It Sy "Second level name	Type	Changeable"
541.It "PF_ROUTE	routing messages	no"
542.It "PF_INET	IPv4 values	yes"
543.It "PF_INET6	IPv6 values	yes"
544.El
545.Pp
546.Bl -tag -width 6n
547.It Li PF_ROUTE
548Return the entire routing table or a subset of it.
549The data is returned as a sequence of routing messages (see
550.Xr route 4
551for the header file, format and meaning).
552The length of each message is contained in the message header.
553.Pp
554The third level name is a protocol number, which is currently always 0.
555The fourth level name is an address family, which may be set to 0 to
556select all address families.
557The fifth and sixth level names are as follows:
558.Bl -column "Fifth level nameXXXXXX" "Sixth level is:XXX" -offset indent
559.It Sy "Fifth level name	Sixth level is:"
560.It "NET_RT_FLAGS	rtflags"
561.It "NET_RT_DUMP	None"
562.It "NET_RT_IFLIST	0 or if_index"
563.It "NET_RT_IFMALIST	0 or if_index"
564.El
565.Pp
566The
567.Dv NET_RT_IFMALIST
568name returns information about multicast group memberships on all interfaces
569if 0 is specified, or for the interface specified by
570.Va if_index .
571.It Li PF_INET
572Get or set various global information about the IPv4
573(Internet Protocol version 4).
574The third level name is the protocol.
575The fourth level name is the variable name.
576The currently defined protocols and names are:
577.Bl -column ProtocolXX VariableXX TypeXX ChangeableXX
578.It Sy "Protocol	Variable	Type	Changeable"
579.It "icmp	bmcastecho	integer	yes"
580.It "icmp	maskrepl	integer	yes"
581.It "ip	forwarding	integer	yes"
582.It "ip	redirect	integer	yes"
583.It "ip	ttl	integer	yes"
584.It "udp	checksum	integer	yes"
585.El
586.Pp
587The variables are as follows:
588.Bl -tag -width 6n
589.It Li icmp.bmcastecho
590Returns 1 if an ICMP echo request to a broadcast or multicast address is
591to be answered.
592.It Li icmp.maskrepl
593Returns 1 if ICMP network mask requests are to be answered.
594.It Li ip.forwarding
595Returns 1 when IP forwarding is enabled for the host,
596meaning that the host is acting as a router.
597.It Li ip.redirect
598Returns 1 when ICMP redirects may be sent by the host.
599This option is ignored unless the host is routing IP packets,
600and should normally be enabled on all systems.
601.It Li ip.ttl
602The maximum time-to-live (hop count) value for an IP packet sourced by
603the system.
604This value applies to normal transport protocols, not to ICMP.
605.It Li udp.checksum
606Returns 1 when UDP checksums are being computed and checked.
607Disabling UDP checksums is strongly discouraged.
608.Pp
609For variables net.inet.*.ipsec, please refer to
610.Xr ipsec 4 .
611.El
612.It Li PF_INET6
613Get or set various global information about the IPv6
614(Internet Protocol version 6).
615The third level name is the protocol.
616The fourth level name is the variable name.
617.Pp
618For variables net.inet6.* please refer to
619.Xr inet6 4 .
620For variables net.inet6.*.ipsec6, please refer to
621.Xr ipsec 4 .
622.El
623.Ss CTL_USER
624The string and integer information available for the CTL_USER level
625is detailed below.
626The changeable column shows whether a process with appropriate
627privilege may change the value.
628.Bl -column "USER_COLL_WEIGHTS_MAXXXX" "integerXXX" -offset indent
629.It Sy "Second level name	Type	Changeable"
630.It "USER_BC_BASE_MAX	integer	no"
631.It "USER_BC_DIM_MAX	integer	no"
632.It "USER_BC_SCALE_MAX	integer	no"
633.It "USER_BC_STRING_MAX	integer	no"
634.It "USER_COLL_WEIGHTS_MAX	integer	no"
635.It "USER_CS_PATH	string	no"
636.It "USER_EXPR_NEST_MAX	integer	no"
637.It "USER_LINE_MAX	integer	no"
638.It "USER_POSIX2_CHAR_TERM	integer	no"
639.It "USER_POSIX2_C_BIND	integer	no"
640.It "USER_POSIX2_C_DEV	integer	no"
641.It "USER_POSIX2_FORT_DEV	integer	no"
642.It "USER_POSIX2_FORT_RUN	integer	no"
643.It "USER_POSIX2_LOCALEDEF	integer	no"
644.It "USER_POSIX2_SW_DEV	integer	no"
645.It "USER_POSIX2_UPE	integer	no"
646.It "USER_POSIX2_VERSION	integer	no"
647.It "USER_RE_DUP_MAX	integer	no"
648.It "USER_STREAM_MAX	integer	no"
649.It "USER_TZNAME_MAX	integer	no"
650.El
651.Bl -tag -width 6n
652.Pp
653.It Li USER_BC_BASE_MAX
654The maximum ibase/obase values in the
655.Xr bc 1
656utility.
657.It Li USER_BC_DIM_MAX
658The maximum array size in the
659.Xr bc 1
660utility.
661.It Li USER_BC_SCALE_MAX
662The maximum scale value in the
663.Xr bc 1
664utility.
665.It Li USER_BC_STRING_MAX
666The maximum string length in the
667.Xr bc 1
668utility.
669.It Li USER_COLL_WEIGHTS_MAX
670The maximum number of weights that can be assigned to any entry of
671the LC_COLLATE order keyword in the locale definition file.
672.It Li USER_CS_PATH
673Return a value for the
674.Ev PATH
675environment variable that finds all the standard utilities.
676.It Li USER_EXPR_NEST_MAX
677The maximum number of expressions that can be nested within
678parenthesis by the
679.Xr expr 1
680utility.
681.It Li USER_LINE_MAX
682The maximum length in bytes of a text-processing utility's input
683line.
684.It Li USER_POSIX2_CHAR_TERM
685Return 1 if the system supports at least one terminal type capable of
686all operations described in
687.St -p1003.2 ,
688otherwise 0.
689.It Li USER_POSIX2_C_BIND
690Return 1 if the system's C-language development facilities support the
691C-Language Bindings Option, otherwise 0.
692.It Li USER_POSIX2_C_DEV
693Return 1 if the system supports the C-Language Development Utilities Option,
694otherwise 0.
695.It Li USER_POSIX2_FORT_DEV
696Return 1 if the system supports the FORTRAN Development Utilities Option,
697otherwise 0.
698.It Li USER_POSIX2_FORT_RUN
699Return 1 if the system supports the FORTRAN Runtime Utilities Option,
700otherwise 0.
701.It Li USER_POSIX2_LOCALEDEF
702Return 1 if the system supports the creation of locales, otherwise 0.
703.It Li USER_POSIX2_SW_DEV
704Return 1 if the system supports the Software Development Utilities Option,
705otherwise 0.
706.It Li USER_POSIX2_UPE
707Return 1 if the system supports the User Portability Utilities Option,
708otherwise 0.
709.It Li USER_POSIX2_VERSION
710The version of
711.St -p1003.2
712with which the system attempts to comply.
713.It Li USER_RE_DUP_MAX
714The maximum number of repeated occurrences of a regular expression
715permitted when using interval notation.
716.It Li USER_STREAM_MAX
717The minimum maximum number of streams that a process may have open
718at any one time.
719.It Li USER_TZNAME_MAX
720The minimum maximum number of types supported for the name of a
721timezone.
722.El
723.Ss CTL_VM
724The string and integer information available for the CTL_VM level
725is detailed below.
726The changeable column shows whether a process with appropriate
727privilege may change the value.
728.Bl -column "Second level nameXXXXXX" "struct loadavgXXX" -offset indent
729.It Sy "Second level name	Type	Changeable"
730.It "VM_LOADAVG	struct loadavg	no"
731.It "VM_METER	struct vmtotal	no"
732.It "VM_PAGEOUT_ALGORITHM	integer	yes"
733.It "VM_SWAPPING_ENABLED	integer	maybe"
734.It "VM_V_CACHE_MAX	integer	yes"
735.It "VM_V_CACHE_MIN	integer	yes"
736.It "VM_V_FREE_MIN	integer	yes"
737.It "VM_V_FREE_RESERVED	integer	yes"
738.It "VM_V_FREE_TARGET	integer	yes"
739.It "VM_V_INACTIVE_TARGET	integer	yes"
740.It "VM_V_PAGEOUT_FREE_MIN	integer	yes"
741.El
742.Pp
743.Bl -tag -width 6n
744.It Li VM_LOADAVG
745Return the load average history.
746The returned data consists of a
747.Va struct loadavg .
748.It Li VM_METER
749Return the system wide virtual memory statistics.
750The returned data consists of a
751.Va struct vmtotal .
752.It Li VM_PAGEOUT_ALGORITHM
7530 if the statistics-based page management algorithm is in use
754or 1 if the near-LRU algorithm is in use.
755.It Li VM_SWAPPING_ENABLED
7561 if process swapping is enabled or 0 if disabled.
757This variable is
758permanently set to 0 if the kernel was built with swapping disabled.
759.It Li VM_V_CACHE_MAX
760Maximum desired size of the cache queue.
761.It Li VM_V_CACHE_MIN
762Minimum desired size of the cache queue.
763If the cache queue size
764falls very far below this value, the pageout daemon is awakened.
765.It Li VM_V_FREE_MIN
766Minimum amount of memory (cache memory plus free memory)
767required to be available before a process waiting on memory will be
768awakened.
769.It Li VM_V_FREE_RESERVED
770Processes will awaken the pageout daemon and wait for memory if the
771number of free and cached pages drops below this value.
772.It Li VM_V_FREE_TARGET
773The total amount of free memory (including cache memory) that the
774pageout daemon tries to maintain.
775.It Li VM_V_INACTIVE_TARGET
776The desired number of inactive pages that the pageout daemon should
777achieve when it runs.
778Inactive pages can be quickly inserted into
779process address space when needed.
780.It Li VM_V_PAGEOUT_FREE_MIN
781If the amount of free and cache memory falls below this value, the
782pageout daemon will enter "memory conserving mode" to avoid deadlock.
783.El
784.Sh RETURN VALUES
785.Rv -std
786.Sh ERRORS
787The following errors may be reported:
788.Bl -tag -width Er
789.It Bq Er EFAULT
790The buffer
791.Fa name ,
792.Fa oldp ,
793.Fa newp ,
794or length pointer
795.Fa oldlenp
796contains an invalid address.
797.It Bq Er EINVAL
798The
799.Fa name
800array is less than two or greater than CTL_MAXNAME.
801.It Bq Er EINVAL
802A non-null
803.Fa newp
804is given and its specified length in
805.Fa newlen
806is too large or too small.
807.It Bq Er ENOMEM
808The length pointed to by
809.Fa oldlenp
810is too short to hold the requested value.
811.It Bq Er ENOMEM
812The smaller of either the length pointed to by
813.Fa oldlenp
814or the estimated size of the returned data exceeds the
815system limit on locked memory.
816.It Bq Er ENOMEM
817Locking the buffer
818.Fa oldp ,
819or a portion of the buffer if the estimated size of the data
820to be returned is smaller,
821would cause the process to exceed its per-process locked memory limit.
822.It Bq Er ENOTDIR
823The
824.Fa name
825array specifies an intermediate rather than terminal name.
826.It Bq Er EISDIR
827The
828.Fa name
829array specifies a terminal name, but the actual name is not terminal.
830.It Bq Er ENOENT
831The
832.Fa name
833array specifies a value that is unknown.
834.It Bq Er EPERM
835An attempt is made to set a read-only value.
836.It Bq Er EPERM
837A process without appropriate privilege attempts to set a value.
838.El
839.Sh FILES
840.Bl -tag -width <netinet/icmpXvar.h> -compact
841.It In sys/sysctl.h
842definitions for top level identifiers, second level kernel and hardware
843identifiers, and user level identifiers
844.It In sys/socket.h
845definitions for second level network identifiers
846.It In sys/gmon.h
847definitions for third level profiling identifiers
848.It In vm/vm_param.h
849definitions for second level virtual memory identifiers
850.It In netinet/in.h
851definitions for third level IPv4/IPv6 identifiers and
852fourth level IPv4/v6 identifiers
853.It In netinet/icmp_var.h
854definitions for fourth level ICMP identifiers
855.It In netinet/icmp6.h
856definitions for fourth level ICMPv6 identifiers
857.It In netinet/udp_var.h
858definitions for fourth level UDP identifiers
859.El
860.Sh SEE ALSO
861.Xr sysconf 3 ,
862.Xr sysctl 8
863.Sh HISTORY
864The
865.Fn sysctl
866function first appeared in
867.Bx 4.4 .
868