xref: /netbsd-src/share/man/man4/options.4 (revision 80d9064ac03cbb6a4174695f0d5b237c8766d3d0)
1.\"	$NetBSD: options.4,v 1.442 2014/08/24 07:59:22 jnemeth Exp $
2.\"
3.\" Copyright (c) 1996
4.\" 	Perry E. Metzger.  All rights reserved.
5.\"
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in the
13.\"    documentation and/or other materials provided with the distribution.
14.\" 3. All advertising materials mentioning features or use of this software
15.\"    must display the following acknowledgment:
16.\"	This product includes software developed for the NetBSD Project
17.\"	by Perry E. Metzger.
18.\" 4. The name of the author may not be used to endorse or promote products
19.\"    derived from this software without specific prior written permission.
20.\"
21.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
22.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
23.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
24.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
25.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
26.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
27.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
28.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
29.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
30.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31.\"
32.\"
33.Dd August 23, 2014
34.Dt OPTIONS 4
35.Os
36.Sh NAME
37.Nm options
38.Nd Miscellaneous kernel configuration options
39.Sh SYNOPSIS
40.Cd cinclude ...
41.Cd config ...
42.Cd [no] file-system ...
43.Cd ident ...
44.Cd include ...
45.Cd [no] makeoptions ...
46.Cd maxusers ...
47.Cd [no] options ...
48.Cd [no] pseudo-device ...
49.Sh DESCRIPTION
50This manual page describes a number of miscellaneous kernel
51configuration options that may be specified in a kernel config
52file.
53See
54.Xr config 1
55and
56.Xr config 5
57for information on how to configure and build kernels.
58.Pp
59The
60.Ar no
61form removes a previously specified option.
62.Ss Keywords
63The following keywords are recognized in a kernel configuration file:
64.Bl -ohang
65.It Sy cinclude Qq Ar filename
66Conditionally includes another kernel configuration file whose name is
67.Ar filename ,
68which may be double-quoted and may be an explicit path or relative to
69the kernel source directory.
70Failure to open the named file is ignored.
71.It Sy config Ar exec_name No root on Ar rootdev Oo type Ar fstype Oc \
72Oo dumps on Ar dumpdev Oc
73Defines a configuration whose kernel executable is named
74.Ar exec_name ,
75normally
76.Dq netbsd ,
77with its root file system of type
78.Ar fstype
79on the device
80.Ar rootdev ,
81and optionally specifying the location of kernel core dumps on the device
82.Ar dumpdev .
83.Ar dev
84or
85.Ar dumpdev
86and
87.Ar fstype
88may be specified as
89.Dq \&? ,
90which is a wild card.
91The root
92.Ar fstype
93and
94.Ar dumpdev
95are optional and assumed to be wild carded if they are not specified.
96.It Ar device_instance No at Ar attachment \
97Oo Ar locators value Oo ... Oc Oc Oo flags Ar value Oc
98Define an instance of the device driver
99.Ar device_instance
100that attaches to the bus or device named
101.Ar attachment .
102An
103.Ar attachment
104may require additional information on where the device can be found, such
105as an address, channel, function, offset, and/or slot, referred to as
106.Ar locators ,
107whose
108.Ar value
109often may be a wild card,
110.Dq \&? .
111Some device drivers have one or more
112.Ar flags
113that can be adjusted to affect the way they operate.
114.It Sy file-system Ar fs_name Op , Ar fs_name Op ...
115Include support for the file-system
116.Ar fs_name .
117.It Sy ident Qq Ar string
118Sets the kernel identification string to
119.Ar string .
120.It Sy include Qq Ar filename
121Functions the same as
122.Ar cinclude ,
123except failure to open
124.Ar filename
125produces a fatal error.
126.It Sy options Ar option_name Op , Ar option_name=value Op ...
127Specifies (or sets) the option, or comma-separated list of options,
128.Ar option_name .
129Some options expect to be assigned a value, which may be an integer,
130a double-quoted word, a bare word, or an empty string
131.Pq Qq .
132Note that those are eventually handled by the C compiler, so the rules
133of that language apply.
134.Pp
135.Em Note :
136Options that are not defined by device definition files are passed to
137the compile process as
138.Fl D
139flags to the C compiler.
140.It Sy makeoptions Ar name=value
141Defines a
142.Xr make 1
143macro
144.Ar name
145with the value
146.Ar value
147in the kernel Makefile.
148.It Sy maxusers Ar integer
149Set the maxusers variable in the kernel.
150.It Sy no Ar keyword Ar name Op Ar arguments Op ...
151For the
152.Xr config 1
153.Ar keywords
154file-system, makeoptions, options, and pseudo-device,
155.Ar no
156removes the file-system, makeoption, options, or pseudo-device,
157.Ar name .
158This is useful when a kernel configuration file includes another which
159has undesired options.
160.Pp
161For example, a local configuration file that wanted the kitchen sink, but
162not COMPAT_09 or bridging, might be:
163.Bd -literal -offset indent
164include "arch/i386/conf/GENERIC"
165no options COMPAT_09
166no pseudo-device bridge
167.Ed
168.It Sy pseudo-device Ar name Op Ar N
169Includes support for the pseudo-device
170.Ar name .
171Some pseudo-devices can have multiple or
172.Ar N
173instances.
174.El
175.Ss Compatibility Options
176.Bl -ohang
177.It Cd options COMPAT_09
178Enable binary compatibility with
179.Nx 0.9 .
180This enables support for
18116-bit user, group, and process IDs (following revisions support
18232-bit identifiers).
183It also allows the use of the deprecated
184.Xr getdomainname 3 ,
185.Xr setdomainname 3 ,
186and
187.Xr uname 3
188syscalls.
189This option also allows using numeric file system identifiers rather
190than strings.
191Post
192.Nx 0.9
193versions use string identifiers.
194.It Cd options COMPAT_10
195Enable binary compatibility with
196.Nx 1.0 .
197This option allows the use of the file system name of
198.Dq ufs
199as an alias for
200.Dq ffs .
201The name
202.Dq ffs
203should be used post 1.0 in
204.Pa /etc/fstab
205and other files.
206It also adds old syscalls for the
207.At V
208shared memory interface.
209This was changed post 1.0 to work on 64-bit architectures.
210This option also enables
211.Dq sgtty
212compatibility, without which programs using the old interface produce
213an
214.Dq inappropriate ioctl
215error, and
216.Pa /dev/io
217only works when this option is set in the kernel,
218see
219.Xr io 4
220on ports that support it.
221.It Cd options COMPAT_11
222Enable binary compatibility with
223.Nx 1.1 .
224This allows binaries running on the i386 port to gain direct access to
225the io ports by opening
226.Pa /dev/io
227read/write.
228This functionality was replaced by
229.Xr i386_iopl 2
230post 1.1.
231On the
232.Tn Atari
233port, the location of the disk label was moved after 1.1.
234When the
235.Em COMPAT_11
236option is set, the kernel will read (pre) 1.1 style disk labels as a
237last resort.
238When a disk label is re-written, the old style label will be replaced
239with a post 1.1 style label.
240This also enables the
241.Em EXEC_ELF_NOTELESS
242option.
243.It Cd options COMPAT_12
244Enable binary compatibility with
245.Nx 1.2 .
246This allows the use of old syscalls for
247.Fn reboot
248and
249.Fn swapon .
250The syscall numbers were changed post 1.2 to add functionality to the
251.Xr reboot 2
252syscall, and the new
253.Xr swapctl 2
254interface was introduced.
255This also enables the
256.Em EXEC_ELF_NOTELESS
257option.
258.It Cd options COMPAT_13
259Enable binary compatibility with
260.Nx 1.3 .
261This allows the use of old syscalls for
262.Fn sigaltstack ,
263and also enables the old
264.Xr swapctl 2
265command
266.Dv SWAP_STATS
267(now called
268.Dv SWAP_OSTATS ) ,
269which does not include the
270.Fa se_path
271member of
272.Va struct swapent .
273.It Cd options COMPAT_14
274Enable binary compatibility with
275.Nx 1.4 .
276This allows some old
277.Xr ioctl 2
278on
279.Xr wscons 4
280to be performed, and allows the
281.Dv NFSSVC_BIOD
282mode of the
283.Xr nfssvc 2
284system call to be used for compatibility with the deprecated nfsiod program.
285.It Cd options COMPAT_15
286Enable binary compatibility with
287.Nx 1.5 .
288Since there were no API changes from
289.Nx 1.5
290and
291.Nx 1.6 ,
292this option does nothing.
293.It Cd options COMPAT_16
294Enable binary compatibility with
295.Nx 1.6 .
296This allows the use of old signal trampoline code which has been deprecated
297with the addition of
298.Xr siginfo 2 .
299.It Cd options COMPAT_20
300Enable binary compatibility with
301.Nx 2.0 .
302This allows the use of old syscalls for
303.Fn statfs ,
304.Fn fstatfs ,
305.Fn getfsstat
306and
307.Fn fhstatfs ,
308which have been deprecated with the addition of the
309.Xr statvfs 2 ,
310.Xr fstatvfs 2 ,
311.Xr getvfsstat 2
312and
313.Xr fhstatvfs 2
314system calls.
315.It Cd options COMPAT_30
316Enable binary compatibility with
317.Nx 3.0 .
318See
319.Xr compat_30 8
320for details about the changes made after the
321.Nx 3.0
322release.
323.It Cd options COMPAT_40
324Enable binary compatibility with
325.Nx 4.0 .
326.It Cd options COMPAT_43
327Enables compatibility with
328.Bx 4.3 .
329This adds an old syscall for
330.Xr lseek 2 .
331It also adds the ioctls for
332.Dv TIOCGETP
333and
334.Dv TIOCSETP .
335The return values for
336.Xr getpid 2 ,
337.Xr getgid 2 ,
338and
339.Xr getuid 2
340syscalls are modified as well, to return the parent's PID and
341UID as well as the current process's.
342It also enables the deprecated
343.Dv NTTYDISC
344terminal line discipline.
345It also provides backwards compatibility with
346.Dq old
347SIOC[GS]IF{ADDR,DSTADDR,BRDADDR,NETMASK} interface ioctls, including
348binary compatibility with code written before the introduction of the
349sa_len field in sockaddrs.
350It also enables
351support for some older pre
352.Bx 4.4
353socket calls.
354.It Cd options COMPAT_50
355Enable binary compatibility with
356.Nx 5.0 .
357.It Cd options COMPAT_60
358Enable binary compatibility with
359.Nx 6.0 .
360.It Cd options COMPAT_70
361Enable binary compatibility with
362.Nx 7.0 .
363.It Cd options COMPAT_BSDPTY
364This option is currently on by default and enables the pty multiplexer
365.Xr ptm 4
366and
367.Xr ptmx 4
368to find and use ptys named
369.Pa /dev/ptyXX
370(master) and
371.Pa /dev/ttyXX
372(slave).
373Eventually this option will become optional as ptyfs based pseudo-ttys become
374the default, see
375.Xr mount_ptyfs 8 .
376.It Cd options COMPAT_SVR4
377On those architectures that support it, this enables binary
378compatibility with
379.At V.4
380applications built for the same architecture.
381This currently includes the i386, m68k, and sparc ports.
382.It Cd options COMPAT_LINUX
383On those architectures that support it, this enables binary
384compatibility with Linux ELF and
385.Xr a.out 5
386applications built for the same architecture.
387This currently includes the alpha, arm, i386, m68k, mips, powerpc and
388x86_64 ports.
389.It Cd options COMPAT_LINUX32
390On those 64 bit architectures that support it, this enables binary
391compatibility with 32 bit Linux binaries.
392For now this is limited to running i386 ELF Linux binaries on amd64.
393.It Cd options COMPAT_SUNOS
394On those architectures that support it, this enables binary
395compatibility with
396.Tn SunOS 4.1
397applications built for the same architecture.
398This currently includes the sparc, sparc64 and most or all m68k ports.
399Note that the sparc64 requires the
400.Em COMPAT_NETBSD32
401option for 64-bit kernels, in addition to this option.
402.It Cd options COMPAT_ULTRIX
403On those architectures that support it, this enables binary
404compatibility with
405.Tn ULTRIX
406applications built for the same architecture.
407This currently is limited to the pmax.
408The functionality of this option is unknown.
409.It Cd options COMPAT_FREEBSD
410On those architectures that support it, this enables binary
411compatibility with
412.Fx
413applications built for the same architecture.
414At the moment this is limited to the i386 port.
415.It Cd options COMPAT_IBCS2
416On those architectures that support it, this enables binary
417compatibility with iBCS2 or SVR3 applications built for the same architecture.
418This is currently limited to the i386 and vax ports.
419.It Cd options COMPAT_OSF1
420On those architectures that support it, this enables binary
421compatibility with
422.Tn Digital
423.Ux
424.Po
425formerly
426.Tn OSF/1
427.Pc
428applications built for the same architecture.
429This is currently limited to the alpha port.
430.It Cd options COMPAT_NOMID
431Enable compatibility with
432.Xr a.out 5
433executables that lack a machine ID.
434This includes
435.Nx 0.8 Ns 's
436ZMAGIC format, and 386BSD and BSDI's
437QMAGIC, NMAGIC, and OMAGIC
438.Xr a.out 5
439formats.
440.It Cd options COMPAT_NETBSD32
441On those architectures that support it, this enables binary
442compatibility with 32-bit applications built for the same architecture.
443This is currently limited to the amd64 and sparc64 ports, and only
444applicable for 64-bit kernels.
445.It Cd options COMPAT_SVR4_32
446On those architectures that support it, this enables binary
447compatibility with 32-bit SVR4 applications built for the same architecture.
448This is currently limited to the sparc64 port, and only applicable for
44964-bit kernels.
450.It Cd options COMPAT_AOUT_M68K
451On m68k architectures which have switched to ELF,
452this enables binary compatibility with
453.Nx Ns Tn /m68k
454.Xr a.out 5
455executables on
456.Nx Ns Tn /m68k
457ELF kernels.
458This handles alignment incompatibility of m68k ABI between
459a.out and ELF which causes the structure padding differences.
460Currently only some system calls which use
461.Va struct stat
462are adjusted and some binaries which use
463.Xr sysctl 3
464to retrieve network details would not work properly.
465.It Cd options EMUL_NATIVEROOT=string
466Just like emulated binaries first try looking up files in
467an emulation root (e.g.
468.Pa /emul/linux )
469before looking them up in real root, this option causes native
470binaries to first look up files in an "emulation" directory too.
471This can be useful to test an amd64 kernel on top of an i386 system
472before full migration: by unpacking the amd64 distribution in e.g.
473.Pa /emul/netbsd64
474and specifying that location as
475.Cd EMUL_NATIVEROOT ,
476native amd64 binaries can be run while the root file system remains
477populated with i386 binaries.
478Beware of
479.Pa /dev
480incompatibilities between i386 and amd64 if you do this.
481.It Cd options EXEC_ELF_NOTELESS
482Run unidentified ELF binaries as
483.Nx
484binaries.
485This might be needed for very old
486.Nx
487ELF binaries on some archs.
488These old binaries didn't contain an appropriate
489.Li .note.netbsd.ident
490section, and thus can't be identified by the kernel as
491.Nx
492binaries otherwise.
493Beware - if this option is on, the kernel would run
494.Em any
495unknown ELF binaries as if they were
496.Nx
497binaries.
498.It Cd options P1003_1B_SEMAPHORE
499Includes kernel support for the standard C library
500.Pq libc
501functions that implement semaphores as specified in
502.St -p1003.1-96 .
503.El
504.Ss Debugging Options
505.Bl -ohang
506.It Cd options DDB
507Compiles in a kernel debugger for diagnosing kernel problems.
508See
509.Xr ddb 4
510for details.
511.Em NOTE :
512not available on all architectures.
513.It Cd options DDB_FROMCONSOLE=integer
514If set to non-zero, DDB may be entered by sending a break on a serial
515console or by a special key sequence on a graphics console.
516A value of "0" ignores console breaks or key sequences.
517If not explicitly specified, the default value is "1".
518Note that this sets the value of the
519.Em ddb.fromconsole
520.Xr sysctl 3
521variable which may be changed at run time -- see
522.Xr sysctl 8
523for details.
524.It Cd options DDB_HISTORY_SIZE=integer
525If this is non-zero, enable history editing in the kernel debugger
526and set the size of the history to this value.
527.It Cd options DDB_ONPANIC
528The default if not specified is
529.Dq 1
530- just enter into DDB.
531If set to
532.Dq 2
533the kernel will
534attempt to print out a stack trace before entering into DDB.
535If set to
536.Dq 0
537the kernel will attempt to print out a stack trace
538and reboot the system.
539If set to
540.Dq -1
541then neither a stack trace is printed or DDB entered -
542it is as if DDB were not compiled into the kernel.
543Note that this sets the value of the
544.Em ddb.onpanic
545.Xr sysctl 3
546variable which may be changed at run time -- see
547.Xr sysctl 8
548for details.
549.It Cd options DDB_COMMANDONENTER=string
550This option specify commands which will be executed on each entry to DDB.
551This sets the default value of the
552.Em ddb.commandonenter
553.Xr sysctl 3
554variable which may be changed at run time.
555.It Cd options DDB_BREAK_CHAR=integer
556This option overrides using break to enter the kernel debugger
557on the serial console.
558The value given is the ASCII value to be used instead.
559This is currently only supported by the com driver.
560.It Cd options DDB_VERBOSE_HELP
561This option adds more verbose descriptions to the
562.Em help
563command.
564.It Cd options KGDB
565Compiles in a remote kernel debugger stub for diagnosing kernel problems
566using the
567.Dq remote target
568feature of gdb.
569See
570.Xr gdb 1
571for details.
572.Em NOTE :
573not available on all architectures.
574.It Cd options KGDB_DEV
575Device number
576.Po
577as a
578.Dv dev_t
579.Pc
580of kgdb device.
581.It Cd options KGDB_DEVADDR
582Memory address of kgdb device.
583.It Cd options KGDB_DEVMODE
584Permissions of kgdb device.
585.It Cd options KGDB_DEVNAME
586Device name of kgdb device.
587.It Cd options KGDB_DEVRATE
588Baud rate of kgdb device.
589.It Cd makeoptions DEBUG="-g"
590The
591.Fl g
592flag causes
593.Pa netbsd.gdb
594to be built in addition to
595.Pa netbsd .
596.Pa netbsd.gdb
597is useful for debugging kernel crash dumps with gdb.
598See
599.Xr gdb 1
600for details.
601This also turns on
602.Em options DEBUG
603(which see).
604.It Cd options DEBUG
605Turns on miscellaneous kernel debugging.
606Since options are turned into preprocessor defines (see above),
607.Em options DEBUG
608is equivalent to doing a
609.Em #define DEBUG
610throughout the kernel.
611Much of the kernel has
612.Em #ifdef DEBUG
613conditionalized debugging code.
614Note that many parts of the kernel (typically device drivers) include their own
615.Em #ifdef XXX_DEBUG
616conditionals instead.
617This option also turns on certain other options,
618which may decrease system performance.
619Systems with this option are not suitable for regular use, and are
620intended only for debugging or looking for bugs.
621.It Cd options DIAGNOSTIC
622Adds code to the kernel that does internal consistency checks.
623This code will cause the kernel to panic if corruption of internal data
624structures is detected.
625Historically, the performance degradation is sufficiently small that
626it is reasonable for systems with
627.Em options DIAGNOSTIC
628to be in production use, with the real consideration not being
629performance but instead a preference for more panics versus continued
630operation with undetected problems.
631.It Cd options LOCKDEBUG
632Adds code to the kernel to detect incorrect use of locking primitives
633(mutex, rwlock).
634This code will cause the kernel to check for dead lock conditions.
635It will also check for memory being freed to not contain initialised
636lock primitives.
637Functions for use in
638.Xr ddb 4
639to check lock chains etc. are also enabled.
640These checks are very expensive and can decrease performance on
641multi-processor machines by a factor of three.
642.It Cd options KSTACK_CHECK_MAGIC
643Check kernel stack usage and panic if stack overflow is detected.
644This check is performance sensitive because it scans stack on each context
645switch.
646.It Cd options KTRACE
647Add hooks for the system call tracing facility, which allows users to
648watch the system call invocation behavior of processes.
649See
650.Xr ktrace 1
651for details.
652.It Cd options MSGBUFSIZE=integer
653This option sets the size of the kernel message buffer.
654This buffer holds the kernel output of
655.Fn printf
656when not (yet) read by
657.Xr syslogd 8 .
658This is particularly useful when the system has crashed and you wish to lookup
659the kernel output from just before the crash.
660Also, since the autoconfig output becomes more and more verbose,
661it sometimes happens that the message buffer overflows before
662.Xr syslogd 8
663was able to read it.
664Note that not all systems are capable of obtaining a variable sized message
665buffer.
666There are also some systems on which memory contents are not preserved
667across reboots.
668.It Cd options MALLOCLOG
669Enables an event log for
670.Xr malloc 9 .
671Useful for tracking down
672.Dq Data modified on freelist
673and
674.Dq multiple free
675problems.
676.It Cd options MALLOCLOGSIZE=integer
677Defines the number of entries in the malloc log.
678Default is 100000 entries.
679.It Cd options KERNHIST
680Enables the kernel history logs, which create in-memory traces of
681various kernel activities.
682These logs can be displayed by using
683.Cm show kernhist
684from DDB.
685See the kernel source file
686.Pa sys/kern/kern_history.c
687for details.
688.It Cd options KERNHIST_PRINT
689Prints the kernel history logs on the system console as entries are added.
690Note that the output is
691.Em extremely
692voluminous, so this option is really only useful for debugging
693the very earliest parts of kernel initialization.
694.It Cd options UVMHIST
695Like
696.Em KERNHIST ,
697it enables the UVM history logs.
698These logs can be displayed by using
699.Cm show kernhist
700from DDB.
701See the kernel source file
702.Pa sys/uvm/uvm_stat.c
703for details.
704.It Cd options UVMHIST_PRINT
705Like
706.Em UVMHIST ,
707it prints the UVM history logs on the system console as entries are added.
708Note that the output is
709.Em extremely
710voluminous, so this option is really only useful for debugging
711the very earliest parts of kernel initialization.
712.El
713.Ss File Systems
714.Bl -ohang
715.It Cd file-system FFS
716Includes code implementing the Berkeley Fast File System
717.Po Em FFS Pc .
718Most machines need this if they are not running diskless.
719.It Cd file-system EXT2FS
720Includes code implementing the Second Extended File System
721.Po Em ext2 Pc ,
722revision 0 and revision 1 with the
723.Em filetype ,
724.Em sparse_super
725and
726.Em large_file
727options.
728This is the most commonly used file system on the Linux operating system,
729and is provided here for compatibility.
730Some of the specific features of
731.Em ext2
732like the "behavior on errors" are not implemented.
733See
734.Xr mount_ext2fs 8
735for details.
736.It Cd file-system LFS
737.Bq Em EXPERIMENTAL
738Include the Log-structured File System
739.Po Em LFS Pc .
740See
741.Xr mount_lfs 8
742and
743.Xr newfs_lfs 8
744for details.
745.It Cd file-system MFS
746Include the Memory File System
747.Po Em MFS Pc .
748This file system stores files in swappable memory, and produces
749notable performance improvements when it is used as the file store
750for
751.Pa /tmp
752and similar file systems.
753See
754.Xr mount_mfs 8
755for details.
756.It Cd file-system NFS
757Include the client side of the Network File System
758.Pq Tn NFS
759remote file sharing protocol.
760Although the bulk of the code implementing
761.Tn NFS
762is kernel based, several user level daemons are needed for it to work.
763See
764.Xr mount_nfs 8
765for details.
766.It Cd file-system CD9660
767Includes code for the
768.Tn ISO
7699660 + Rock Ridge file system, which is the standard file system on many
770.Tn CD-ROM
771discs.
772Useful primarily if you have a
773.Tn CD-ROM
774drive.
775See
776.Xr mount_cd9660 8
777for details.
778.It Cd file-system MSDOSFS
779Includes the
780.Tn MS-DOS
781FAT file system, which is reportedly still used
782by unfortunate people who have not heard about
783.Nx .
784Also implements the
785.Tn Windows 95
786extensions to the same, which permit the use of longer, mixed case
787file names.
788See
789.Xr mount_msdos 8
790and
791.Xr fsck_msdos 8
792for details.
793.It Cd file-system NTFS
794.Bq Em EXPERIMENTAL
795Includes code for the
796.Tn Microsoft Windows NT
797file system.
798See
799.Xr mount_ntfs 8
800for details.
801.It Cd file-system FDESC
802Includes code for a file system, conventionally mounted on
803.Pa /dev/fd ,
804which permits access to the per-process file descriptor space via
805special files in the file system.
806See
807.Xr mount_fdesc 8
808for details.
809Note that this facility is redundant, and thus unneeded on most
810.Nx
811systems, since the
812.Xr fd 4
813pseudo-device driver already provides identical functionality.
814On most
815.Nx
816systems, instances of
817.Xr fd 4
818are mknoded under
819.Pa /dev/fd/
820and on
821.Pa /dev/stdin ,
822.Pa /dev/stdout ,
823and
824.Pa /dev/stderr .
825.It Cd file-system KERNFS
826Includes code which permits the mounting of a special file system
827(normally mounted on
828.Pa /kern )
829in which files representing various kernel variables and parameters
830may be found.
831See
832.Xr mount_kernfs 8
833for details.
834.It Cd file-system NULLFS
835Includes code for a loopback file system.
836This permits portions of the file hierarchy to be re-mounted in other places.
837The code really exists to provide an example of a stackable file system layer.
838See
839.Xr mount_null 8
840for details.
841.It Cd file-system OVERLAY
842Includes code for a file system filter.
843This permits the overlay file system to intercept all access to an underlying
844file system.
845This file system is intended to serve as an example of a stacking file
846system which has a need to interpose itself between an underlying file
847system and all other access.
848See
849.Xr mount_overlay 8
850for details.
851.It Cd file-system PROCFS
852Includes code for a special file system (conventionally mounted on
853.Pa /proc )
854in which the process space becomes visible in the file system.
855Among
856other things, the memory spaces of processes running on the system are
857visible as files, and signals may be sent to processes by writing to
858.Pa ctl
859files in the procfs namespace.
860See
861.Xr mount_procfs 8
862for details.
863.It Cd file-system UDF
864.Bq Em EXPERIMENTAL
865Includes code for the UDF file system commonly found on CD and DVD
866media but also on USB sticks.
867Currently supports read and write access upto UDF 2.01 and somewhat limited
868write support for UDF 2.50.
869It is marked experimental since there is no
870.Xr fsck_udf 8 .
871See
872.Xr mount_udf 8
873for details.
874.It Cd file-system UMAPFS
875Includes a loopback file system in which user and group IDs may be
876remapped -- this can be useful when mounting alien file systems with
877different UIDs and GIDs than the local system.
878See
879.Xr mount_umap 8
880for details.
881.It Cd file-system UNION
882.Bq Em EXPERIMENTAL
883Includes code for the union file system, which permits directories to
884be mounted on top of each other in such a way that both file systems
885remain visible -- this permits tricks like allowing writing (and the
886deleting of files) on a read-only file system like a
887.Tn CD-ROM
888by mounting a local writable file system on top of the read-only file system.
889See
890.Xr mount_union 8
891for details.
892.It Cd file-system CODA
893.Bq Em EXPERIMENTAL
894Includes code for the Coda file system.
895Coda is a distributed file system like NFS and AFS.
896It is freely available, like NFS, but it functions much like AFS in being a
897.Dq stateful
898file system.
899Both Coda and AFS cache files on your local machine to improve performance.
900Then Coda goes a step further than AFS by letting you access the cached
901files when there is no available network, viz. disconnected laptops and
902network outages.
903In Coda, both the client and server are outside the kernel which makes
904them easier to experiment with.
905Coda is available for several UNIX and non-UNIX platforms.
906See
907.Lk http://www.coda.cs.cmu.edu
908for more details.
909.Em NOTE :
910You also need to enable the pseudo-device, vcoda, for the Coda
911file system to work.
912.It Cd file-system SMBFS
913.Bq Em EXPERIMENTAL
914Includes code for the SMB/CIFS file system.
915See
916.Xr mount_smbfs 8
917for details.
918.Em NOTE :
919You also need to enable the pseudo-device, nsmb, for the SMB
920file system to work.
921.It Cd file-system PTYFS
922.Bq Em EXPERIMENTAL
923Includes code for a special file system (normally mounted on
924.Pa /dev/pts )
925in which pseudo-terminal slave devices become visible in the file system.
926See
927.Xr mount_ptyfs 8
928for details.
929.It Cd file-system TMPFS
930Includes code for the efficient memory file system, normally used over
931.Pa /tmp .
932See
933.Xr mount_tmpfs 8
934for details.
935.It Cd file-system PUFFS
936Includes kernel support for the pass-to-userspace framework file system.
937It can be used to implement file system functionality in userspace.
938See
939.Xr puffs 3
940for more details.
941This enables for example sshfs:
942.Xr mount_psshfs 8 .
943.El
944.Ss File System Options
945.Bl -ohang
946.It Cd options MAGICLINKS
947Enables the expansion of special strings
948.Po
949beginning with
950.Dq @
951.Pc
952when traversing symbolic links.
953See
954.Xr symlink 7
955for a list of supported strings.
956Note that this option only controls the enabling of this feature
957by the kernel at boot-up.
958This feature can still be manipulated with the
959.Xr sysctl 8
960command regardless of the setting of this option.
961.It Cd options NFSSERVER
962Include the server side of the
963.Em NFS
964(Network File System) remote file sharing protocol.
965Although the bulk of the code implementing
966.Em NFS
967is kernel based, several user level daemons are needed for it to
968work.
969See
970.Xr mountd 8
971and
972.Xr nfsd 8
973for details.
974.It Cd options NVNODE=integer
975This option sets the size of the cache used by the name-to-inode translation
976routines, (a.k.a. the
977.Fn namei
978cache, though called by many other names in the kernel source).
979By default, this cache has
980.Dv NPROC
981(set as 20 + 16 * MAXUSERS) * (80 + NPROC / 8) entries.
982A reasonable way to derive a value of
983.Dv NVNODE ,
984should you notice a large number of namei cache misses with a tool such as
985.Xr systat 1 ,
986is to examine your system's current computed value with
987.Xr sysctl 8 ,
988(which calls this parameter "kern.maxvnodes") and to increase this value
989until either the namei cache hit rate improves or it is determined that
990your system does not benefit substantially from an increase in the size of
991the namei cache.
992.It Cd options NAMECACHE_ENTER_REVERSE
993Causes the namei cache to always enter a reverse mapping (vnode -\*[Gt] name)
994as well as a normal one.
995Normally, this is already done for directory vnodes, to speed up the getcwd
996operation.
997This option will cause longer hash chains in the reverse cache, and thus
998slow down getcwd somewhat.
999However, it does make vnode -\*[Gt] path translations possible in some cases.
1000For now, only useful if strict
1001.Pa /proc/#/maps
1002emulation for Linux binaries is required.
1003.El
1004.Ss Options for FFS/UFS File Systems
1005.Bl -ohang
1006.It Cd options WAPBL
1007Enable
1008.Dq Write Ahead Physical Block Logging file system journaling .
1009This provides rapid file system consistency checking after a system outage.
1010It also provides better general use performance over regular FFS.
1011See also
1012.Xr wapbl 4 .
1013.It Cd options QUOTA
1014Enables kernel support for traditional quotas in FFS.
1015Traditional quotas store the quota information in external files and
1016require
1017.Xr quotacheck 8
1018and
1019.Xr quotaon 8
1020at boot time.
1021Traditional quotas are limited to 32-bit sizes and are at this point
1022considered a legacy feature.
1023.It Cd options QUOTA2
1024Enables kernel support for in-volume quotas in FFS.
1025The quota information is file system metadata maintained by
1026.Xr fsck 8
1027and/or WAPBL journaling.
1028MFS volumes can also use
1029.Dv QUOTA2
1030quotas; see
1031.Xr mount_mfs 8
1032for more information.
1033.It Cd options FFS_EI
1034Enable
1035.Dq Endian-Independent
1036FFS support.
1037This allows a system to mount an FFS file system created for another
1038architecture, at a small performance cost for all FFS file systems.
1039See also
1040.Xr newfs 8 ,
1041.Xr fsck_ffs 8 ,
1042.Xr dumpfs 8
1043for file system byte order status and manipulation.
1044.It Cd options FFS_NO_SNAPSHOT
1045Disable support for the creation of file system internal snapshot
1046of FFS file systems.
1047Maybe useful for install media kernels, small memory systems and
1048embedded systems which don't require the snapshot support.
1049.It Cd options UFS_EXTATTR
1050Enable extended attribute support for UFS1 file systems.
1051.It Cd options UFS_DIRHASH
1052Increase lookup performance by maintaining in-core hash tables
1053for large directories.
1054.El
1055.Ss Options for the LFS File System
1056.Bl -ohang
1057.\" .It Cd options LFS_KERNEL_RFW
1058.\" There is no documentation for this. (XXX)
1059.\" .It Cd options LFS_QUOTA
1060.\" Enable traditional quota support for LFS.
1061.\" (It is not clear if this works.)
1062.\" .It Cd options LFS_QUOTA2
1063.\" Enable modernized 64-bit quota support for LFS.
1064.\" (This does not work yet.)
1065.It Cd options LFS_EI
1066Enable
1067.Dq Endian-Independent
1068LFS support.
1069This allows (at a small performance cost) mounting an LFS file system
1070created for another architecture.
1071.\" .It Cd options LFS_EXTATTR
1072.\" Enable extended attribute support for LFS.
1073.\" (It is not clear if this works.)
1074.It Cd options LFS_DIRHASH
1075Increase lookup performance by maintaining in-core hash tables
1076for large directories.
1077.El
1078.Ss Options for the ext2fs File System
1079.Bl -ohang
1080.It Cd options EXT2FS_SYSTEM_FLAGS
1081This option changes the behavior of the APPEND and IMMUTABLE flags
1082for a file on an
1083.Em ext2
1084file system.
1085Without this option, the superuser or owner of the file can
1086set and clear them.
1087With this option, only the superuser can set them, and
1088they can't be cleared if the securelevel is greater than 0.
1089See also
1090.Xr chflags 1
1091and
1092.Xr secmodel_securelevel 9 .
1093.El
1094.Ss Options for the NFS File System
1095.Bl -ohang
1096.It Cd options NFS_BOOT_BOOTP
1097Enable use of the BOOTP protocol (RFCs 951 and 1048) to get configuration
1098information if NFS is used to mount the root file system.
1099See
1100.Xr diskless 8
1101for details.
1102.It Cd options NFS_BOOT_BOOTSTATIC
1103Enable use of static values defined as
1104.Dq NFS_BOOTSTATIC_MYIP ,
1105.Dq NFS_BOOTSTATIC_GWIP ,
1106.Dq NFS_BOOTSTATIC_SERVADDR ,
1107and
1108.Dq NFS_BOOTSTATIC_SERVER
1109in kernel options to get configuration information
1110if NFS is used to mount the root file system.
1111.It Cd options NFS_BOOT_DHCP
1112Same as
1113.Dq NFS_BOOT_BOOTP ,
1114but use the DHCP extensions to the
1115BOOTP protocol (RFC 1541).
1116.It Cd options NFS_BOOT_BOOTP_REQFILE
1117Specifies the string sent in the bp_file field of the BOOTP/DHCP
1118request packet.
1119.It Cd options NFS_BOOT_BOOTPARAM
1120Enable use of the BOOTPARAM protocol, consisting of RARP and
1121BOOTPARAM RPC, to get configuration information if NFS
1122is used to mount the root file system.
1123See
1124.Xr diskless 8
1125for details.
1126.It Cd options NFS_BOOT_RWSIZE=value
1127Set the initial NFS read and write sizes for diskless-boot requests.
1128The normal default is 8Kbytes.
1129This option provides a way to lower the value (e.g., to 1024 bytes)
1130as a workaround for buggy network interface cards or boot PROMs.
1131Once booted, the read and write request sizes can be increased by
1132remounting the file system.
1133See
1134.Xr mount_nfs 8
1135for details.
1136.It Cd options NFS_V2_ONLY
1137Reduce the size of the NFS client code by omitting code that's only required
1138for NFSv3 and NQNFS support, leaving only that code required to use NFSv2
1139servers.
1140.El
1141.Ss Buffer queue strategy options
1142The following options enable alternative buffer queue strategies.
1143.Bl -ohang
1144.It Cd options BUFQ_READPRIO
1145Enable experimental buffer queue strategy for disk I/O.
1146In the default strategy, outstanding disk requests are ordered by
1147sector number and sent to the disk, regardless of whether the
1148operation is a read or write; this option gives priority to issuing
1149read requests over write requests.
1150Although requests may therefore be issued out of sector-order, causing
1151more seeks and thus lower overall throughput, interactive system
1152responsiveness under heavy disk I/O load may be improved, as processes
1153blocking on disk reads are serviced sooner (file writes typically
1154don't cause applications to block).
1155The performance effect varies greatly depending on the hardware, drive
1156firmware, file system configuration, workload, and desired performance
1157trade-off.
1158Systems using drive write-cache (most modern IDE disks, by default)
1159are unlikely to benefit and may well suffer; such disks acknowledge
1160writes very quickly, and optimize them internally according to
1161physical layout.
1162Giving these disks as many requests to work with as possible (the
1163standard strategy) will typically produce the best results, especially
1164if the drive has a large cache; the drive will silently complete
1165writes from cache as it seeks for reads.
1166Disks that support a large number of concurrent tagged requests (SCSI
1167disks and many hardware RAID controllers) expose this internal
1168scheduling with tagged responses, and don't block for reads; such
1169disks may not see a noticeable difference with either strategy.
1170However, if IDE disks are run with write-cache disabled for safety,
1171writes are not acknowledged until actually completed, and only one
1172request can be outstanding; a large number of small writes in one
1173locality can keep the disk busy, starving reads elsewhere on the disk.
1174Such systems are likely to see the most benefit from this option.
1175Finally, the performance interaction of this option with ffs soft
1176dependencies can be subtle, as that mechanism can drastically alter
1177the workload for file system metadata writes.
1178.It Cd options BUFQ_PRIOCSCAN
1179Enable another buffer queue strategy for disk I/O, per-priority cyclical scan.
1180.It Cd options NEW_BUFQ_STRATEGY
1181Synonym of
1182.Em BUFQ_READPRIO .
1183.El
1184.Ss Miscellaneous Options
1185.Bl -ohang
1186.It Cd options CPU_UCODE
1187Support cpu microcode loading via
1188.Xr cpuctl 8 .
1189.It Cd options MEMORY_DISK_DYNAMIC
1190This option makes the
1191.Xr md 4
1192.Tn RAM
1193disk size dynamically sized.
1194It is incompatible with
1195.Xr mdsetimage 8 .
1196.It Cd options MEMORY_DISK_HOOKS
1197This option allows for some machine dependent functions to be called when
1198the
1199.Xr md 4
1200.Tn RAM
1201disk driver is configured.
1202This can result in automatically loading a
1203.Tn RAM
1204disk from floppy on open (among other things).
1205.It Cd options MEMORY_DISK_IS_ROOT
1206Forces the
1207.Xr md 4
1208.Tn RAM
1209disk to be the root device.
1210This can only be overridden when
1211the kernel is booted in the 'ask-for-root' mode.
1212.It Cd options MEMORY_DISK_ROOT_SIZE=integer
1213Allocates the given number of 512 byte blocks as memory for the
1214.Xr md 4
1215.Tn RAM
1216disk, to be populated with
1217.Xr mdsetimage 8 .
1218.It Cd options MEMORY_DISK_SERVER=0
1219Do not include the interface to a userland memory disk server process.
1220Per default, this option is set to 1, including the support code.
1221Useful for install media kernels.
1222.It Cd options MEMORY_DISK_RBFLAGS=value
1223This option sets the
1224.Xr reboot 2
1225flags used when booting with a memory disk as root file system.
1226Possible values include
1227.Dv RB_AUTOBOOT
1228(boot in the usual fashion - default value), and
1229.Dv RB_SINGLE
1230(boot in single-user mode).
1231.It Cd options MODULAR
1232Enables the framework for kernel modules (see
1233.Xr module 7 ) .
1234.It Cd options MODULAR_DEFAULT_AUTOLOAD
1235Enables the autoloading of kernel modules by default.
1236This sets the default value of the
1237.Em kern.module.autoload
1238.Xr sysctl 3
1239variable which may be changed at run time.
1240.It Cd options VND_COMPRESSION
1241Enables the
1242.Xr vnd 4
1243driver to also handle compressed images.
1244See
1245.Xr vndcompress 1 ,
1246.Xr vnd 4
1247and
1248.Xr vnconfig 8
1249for more information.
1250.It Cd options SPLDEBUG
1251Help the kernel programmer find bugs related to the interrupt priority
1252level.
1253When
1254.Fn spllower
1255or
1256.Fn splraise
1257changes the current CPU's interrupt priority level to or from
1258.Dv IPL_HIGH ,
1259record a backtrace.
1260Read
1261.Xr return_address 9
1262for caveats about collecting backtraces.
1263This feature is experimental, and it is only available on i386.
1264See
1265.Pa sys/kern/subr_spldebug.c .
1266.It Cd options TFTPROOT
1267Download the root memory disk through TFTP at root mount time.
1268This enables the use of a root
1269.Tn RAM
1270disk without requiring it to be embedded in the kernel using
1271.Xr mdsetimage 8 .
1272The
1273.Tn RAM
1274disk name is obtained using DHCP's filename parameter.
1275This option requires
1276.Em MEMORY_DISK_HOOKS ,
1277.Em MEMORY_DISK_DYNAMIC ,
1278and
1279.Em MEMORY_DISK_IS_ROOT .
1280It is incompatible with
1281.Em MEMORY_DISK_ROOT_SIZE .
1282.It Cd options MALLOC_NOINLINE
1283Time critical fixed size memory allocation is performed with
1284.Fn MALLOC
1285and
1286.Fn FREE .
1287Normally these expand to inline code, but with
1288.Em MALLOC_NOINLINE
1289these call the normal
1290.Fn malloc
1291and
1292.Fn free
1293functions.
1294Useful for install media kernels, small memory systems and embedded systems.
1295.It Cd options HZ=integer
1296On ports that support it, set the system clock frequency (see
1297.Xr hz 9 )
1298to the supplied value.
1299Handle with care.
1300.It Cd options NTP
1301Turns on in-kernel precision timekeeping support used by software
1302implementing
1303.Em NTP
1304(Network Time Protocol, RFC 1305).
1305The
1306.Em NTP
1307option adds an in-kernel Phase-Locked Loop (PLL) for normal
1308.Em NTP
1309operation, and a Frequency-Locked Loop (FLL) for intermittently-connected
1310operation.
1311.Xr ntpd 8
1312will employ a user-level PLL when kernel support is unavailable,
1313but the in-kernel version has lower latency and more precision, and
1314so typically keeps much better time.
1315.Pp
1316The interface to the kernel
1317.Em NTP
1318support is provided by the
1319.Xr ntp_adjtime 2
1320and
1321.Xr ntp_gettime 2
1322system calls, which are intended for use by
1323.Xr ntpd 8
1324and are enabled by the option.
1325On systems with sub-microsecond resolution timers, or where (HZ/100000)
1326is not an integer, the
1327.Em NTP
1328option also enables extended-precision arithmetic to keep track of
1329fractional clock ticks at NTP time-format precision.
1330.It Cd options PPS_SYNC
1331This option enables a kernel serial line discipline for receiving time
1332phase signals from an external reference clock such as a radio clock.
1333.Po
1334The
1335.Em NTP
1336option (which see) must be on if the
1337.Em PPS_SYNC
1338option is used
1339.Pc .
1340Some reference clocks generate a Pulse Per Second (PPS) signal in
1341phase with their time source.
1342The
1343.Em PPS
1344line discipline receives this signal on either the data leads
1345or the DCD control lead of a serial port.
1346.Pp
1347.Em NTP
1348uses the PPS signal to discipline the local clock oscillator to a high
1349degree of precision (typically less than 50 microseconds in time and
13500.1 ppm in accuracy).
1351.Em PPS
1352can also generate a serial output pulse when the system receives a PPS
1353interrupt.
1354This can be used to measure the system interrupt latency and thus calibrate
1355.Em NTP
1356to account for it.
1357Using
1358.Em PPS
1359usually requires a gadget box
1360to convert from
1361.Tn TTL
1362to
1363.Tn RS-232
1364signal levels.
1365The gadget box and PPS are described in more detail in the HTML documentation
1366for
1367.Xr ntpd 8
1368in
1369.Pa /usr/share/doc/html/ntp .
1370.Pp
1371.Nx
1372currently supports this option in
1373.Xr com 4
1374and
1375.Xr zsc 4 .
1376.It Cd options SETUIDSCRIPTS
1377Allows scripts with the setuid bit set to execute as the effective
1378user rather than the real user, just like binary executables.
1379.Pp
1380.Em NOTE :
1381Using this option will also enable
1382.Em options FDSCRIPTS
1383.It Cd options FDSCRIPTS
1384Allows execution of scripts with the execute bit set, but not the
1385read bit, by opening the file and passing the file descriptor to
1386the shell, rather than the filename.
1387.Pp
1388.Em NOTE :
1389Execute only (non-readable) scripts will have
1390.Va argv[0]
1391set to
1392.Pa /dev/fd/* .
1393What this option allows as far as security is
1394concerned, is the ability to safely ensure that the correct script
1395is run by the interpreter, as it is passed as an already open file.
1396.It Cd options RTC_OFFSET=integer
1397The kernel (and typically the hardware battery backed-up clock on
1398those machines that have one) keeps time in
1399.Em UTC
1400(Universal Coordinated Time, once known as
1401.Em GMT ,
1402or Greenwich Mean Time)
1403and not in the time of the local time zone.
1404The
1405.Em RTC_OFFSET
1406option is used on some ports (such as the i386) to tell the kernel
1407that the hardware clock is offset from
1408.Em UTC
1409by the specified number of minutes.
1410This is typically used when a machine boots several operating
1411systems and one of them wants the hardware clock to run in the
1412local time zone and not in
1413.Em UTC ,
1414e.g.
1415.Em RTC_OFFSET=300
1416means
1417the hardware clock is set to US Eastern Time (300 minutes behind
1418.Em UTC ) ,
1419and not
1420.Em UTC .
1421(Note:
1422.Em RTC_OFFSET
1423is used to initialize a kernel variable named
1424.Va rtc_offset
1425which is the source actually used to determine the clock offset, and
1426which may be accessed via the kern.rtc_offset sysctl variable.
1427See
1428.Xr sysctl 8
1429and
1430.Xr sysctl 3
1431for details.
1432Since the kernel clock is initialized from the hardware clock very
1433early in the boot process, it is not possible to meaningfully change
1434.Va rtc_offset
1435in system initialization scripts.
1436Changing this value currently may only be done at kernel compile
1437time or by patching the kernel and rebooting).
1438.Pp
1439.Em NOTE :
1440Unfortunately, in many cases where the hardware clock
1441is kept in local time, it is adjusted for Daylight Savings
1442Time; this means that attempting to use
1443.Em RTC_OFFSET
1444to let
1445.Nx
1446coexist with such an operating system, like Windows,
1447would necessitate changing
1448.Em RTC_OFFSET
1449twice a year.
1450As such, this solution is imperfect.
1451.It Cd options KMEMSTATS
1452The kernel memory allocator,
1453.Xr malloc 9 ,
1454will keep statistics on its performance if this option is enabled.
1455Unfortunately, this option therefore essentially disables the
1456.Fn MALLOC
1457and
1458.Fn FREE
1459forms of the memory allocator, which are used to enhance the performance
1460of certain critical sections of code in the kernel.
1461This option therefore can lead to a significant decrease in the
1462performance of certain code in the kernel if enabled.
1463Examples of such code include the
1464.Fn namei
1465routine, the
1466.Xr ccd 4
1467driver,
1468and much of the networking code.
1469.It Cd options MAXUPRC=integer
1470Sets the soft
1471.Dv RLIMIT_NPROC
1472resource limit, which specifies the maximum number of simultaneous
1473processes a user is permitted to run, for process 0;
1474this value is inherited by its child processes.
1475It defaults to
1476.Em CHILD_MAX ,
1477which is currently defined to be 160.
1478Setting
1479.Em MAXUPRC
1480to a value less than
1481.Em CHILD_MAX
1482is not permitted, as this would result in a violation of the semantics of
1483.St -p1003.1-90 .
1484.It Cd options NOFILE=integer
1485Sets the soft
1486.Dv RLIMIT_NOFILE
1487resource limit, which specifies the maximum number of open
1488file descriptors for each process;
1489this value is inherited by its child processes.
1490It defaults to
1491.Em OPEN_MAX ,
1492which is currently defined to be 64.
1493.It Cd options MAXFILES=integer
1494Sets the default value of the
1495.Em kern.maxfiles
1496sysctl variable, which indicates the maximum number of files that may
1497be open in the system.
1498.It Cd options DEFCORENAME=string
1499Sets the default value of the
1500.Em kern.defcorename
1501sysctl variable, otherwise it is set to
1502.Nm %n.core .
1503See
1504.Xr sysctl 8
1505and
1506.Xr sysctl 3
1507for details.
1508.It Cd options RASOPS_CLIPPING
1509Enables clipping within the
1510.Nm rasops
1511raster-console output system.
1512.Em NOTE :
1513only available on architectures that use
1514.Nm rasops
1515for console output.
1516.It Cd options RASOPS_SMALL
1517Removes optimized character writing code from the
1518.Nm rasops
1519raster-console output system.
1520.Em NOTE :
1521only available on architectures that use
1522.Nm rasops
1523for console output.
1524.It Cd options INCLUDE_CONFIG_FILE
1525Embeds the kernel config file used to define the kernel in the kernel
1526binary itself.
1527The embedded data also includes any files directly included by the config
1528file itself, e.g.
1529.Pa GENERIC.local
1530or
1531.Pa std.$MACHINE .
1532The embedded config file can be extracted from the resulting kernel with
1533.Xr config 1
1534.Fl x ,
1535or by the following command:
1536.Bd -literal -offset indent
1537strings netbsd | sed -n 's/^_CFG_//p' | unvis
1538.Ed
1539.It Cd options INCLUDE_JUST_CONFIG
1540Similar to the above option, but includes just the actual config file,
1541not any included files.
1542.It Cd options PIPE_SOCKETPAIR
1543Use slower, but smaller socketpair(2)-based pipe implementation instead
1544of default faster, but bigger one.
1545Primarily useful for installation kernels.
1546.It Cd options USERCONF
1547Compiles in the in-kernel device configuration manager.
1548See
1549.Xr userconf 4
1550for details.
1551.It Cd options PERFCTRS
1552Compiles in kernel support for CPU performance-monitoring counters.
1553See
1554.Xr pmc 1
1555for details.
1556.Em NOTE :
1557not available on all architectures.
1558.It Cd options SYSCALL_STATS
1559Count the number of times each system call number is called.
1560The values can be read through the sysctl interface and displayed using
1561.Xr systat 1 .
1562.Em NOTE :
1563not yet available on all architectures.
1564.It Cd options SYSCALL_TIMES
1565Count the time spent (using
1566.Fn cpu_counter32 )
1567in each system call.
1568.Em NOTE :
1569Using this option will also enable
1570.Cd options SYSCALL_STATS .
1571.It Cd options SYSCALL_TIMES_HASCOUNTER
1572Force use of
1573.Fn cpu_counter32
1574even if
1575.Fn cpu_hascounter
1576reports false.
1577Useful for systems where the cycle counter doesn't run at a constant rate
1578(e.g. Soekris boxes).
1579.It Cd options XSERVER_DDB
1580A supplement to XSERVER that adds support for entering
1581.Xr ddb 4
1582while in
1583.Tn X11 .
1584.It Cd options FILEASSOC
1585Support for
1586.Xr fileassoc 9 .
1587.It Cd options FILEASSOC_NHOOKS=integer
1588Number of storage slots per file for
1589.Xr fileassoc 9 .
1590Default is 4.
1591.El
1592.Ss Networking Options
1593.Bl -ohang
1594.It Cd options GATEWAY
1595Enables
1596.Em IPFORWARDING
1597(which see)
1598and (on most ports) increases the size of
1599.Em NMBCLUSTERS
1600(which see).
1601In general,
1602.Em GATEWAY
1603is used to indicate that a system should act as a router, and
1604.Em IPFORWARDING
1605is not invoked directly.
1606(Note that
1607.Em GATEWAY
1608has no impact on protocols other than
1609.Tn IP ,
1610such as
1611.Tn CLNP ) .
1612.Em GATEWAY
1613option also compiles IPv4 and IPv6 fast forwarding code into the kernel.
1614.It Cd options ICMPPRINTFS
1615The
1616.Em ICMPPRINTFS
1617option will enable debugging information to be printed about
1618the
1619.Xr icmp 4
1620protocol.
1621.It Cd options IPFORWARDING=value
1622If
1623.Em value
1624is 1 this enables IP routing behavior.
1625If
1626.Em value
1627is 0 (the default), it disables it.
1628The
1629.Em GATEWAY
1630option sets this to 1 automatically.
1631With this option enabled, the machine will forward IP datagrams destined
1632for other machines between its interfaces.
1633Note that even without this option, the kernel will
1634still forward some packets (such as source routed packets) -- removing
1635.Em GATEWAY
1636and
1637.Em IPFORWARDING
1638is insufficient to stop all routing through a bastion host on a
1639firewall -- source routing is controlled independently.
1640To turn off source routing, use
1641.Em options IPFORWSRCRT=0
1642(which see).
1643Note that IP forwarding may be turned on and off independently of the
1644setting of the
1645.Em IPFORWARDING
1646option through the use of the
1647.Em net.inet.ip.forwarding
1648sysctl variable.
1649If
1650.Em net.inet.ip.forwarding
1651is 1, IP forwarding is on.
1652See
1653.Xr sysctl 8
1654and
1655.Xr sysctl 3
1656for details.
1657.It Cd options IPFORWSRCRT=value
1658If
1659.Em value
1660is set to zero, source routing of IP datagrams is turned off.
1661If
1662.Em value
1663is set to one (the default) or the option is absent, source routed IP
1664datagrams are forwarded by the machine.
1665Note that source routing of IP packets may be turned on and off
1666independently of the setting of the
1667.Em IPFORWSRCRT
1668option through the use of the
1669.Em net.inet.ip.forwsrcrt
1670sysctl variable.
1671If
1672.Em net.inet.ip.forwsrcrt
1673is 1, forwarding of source routed IP datagrams is on.
1674See
1675.Xr sysctl 8
1676and
1677.Xr sysctl 3
1678for details.
1679.It Cd options IFA_STATS
1680Tells the kernel to maintain per-address statistics on bytes sent
1681and received over (currently) Internet and AppleTalk addresses.
1682.\"This can be a fairly expensive operation, so you probably want to
1683.\"keep this disabled.
1684The option is not recommended as it degrades system stability.
1685.It Cd options IFQ_MAXLEN=value
1686Increases the allowed size of the network interface packet queues.
1687The default queue size is 50 packets, and you do not normally need
1688to increase it.
1689.It Cd options IPSELSRC
1690Includes support for source-address selection policies.
1691See
1692.Xr in_getifa 9 .
1693.It Cd options MROUTING
1694Includes support for IP multicast routers.
1695You certainly want
1696.Em INET
1697with this.
1698Multicast routing is controlled by the
1699.Xr mrouted 8
1700daemon.
1701See also option
1702.Cd PIM .
1703.It Cd options PIM
1704Includes support for Protocol Independent Multicast (PIM) routing.
1705You need
1706.Em MROUTING
1707and
1708.Em INET
1709with this.
1710Software using this can be found e.g. in
1711.Pa pkgsrc/net/xorp .
1712.It Cd options INET
1713Includes support for the
1714.Tn TCP/IP
1715protocol stack.
1716You almost certainly want this.
1717See
1718.Xr inet 4
1719for details.
1720.It Cd options INET6
1721Includes support for the
1722.Tn IPv6
1723protocol stack.
1724See
1725.Xr inet6 4
1726for details.
1727Unlike
1728.Em INET ,
1729.Em INET6
1730enables multicast routing code as well.
1731This option requires
1732.Em INET
1733at this moment, but it should not.
1734.It Cd options ND6_DEBUG
1735The option sets the default value of net.inet6.icmp6.nd6_debug to 1,
1736for debugging IPv6 neighbor discovery protocol handling.
1737See
1738.Xr sysctl 3
1739for details.
1740.It Cd options IPSEC
1741Includes support for the
1742.Tn IPsec
1743protocol, using the implementation derived from
1744.Ox ,
1745relying on
1746.Xr opencrypto 9
1747to carry out cryptographic operations.
1748See
1749.Xr fast_ipsec 4
1750for details.
1751.It Cd options IPSEC_DEBUG
1752Enables debugging code in
1753.Tn IPsec
1754stack.
1755See
1756.Xr ipsec 4
1757for details.
1758The
1759.Cd IPSEC
1760option includes support for
1761.Tn IPsec
1762Network Address Translator traversal (NAT-T), as described in RFCs 3947
1763and 3948.
1764This feature might be patent-encumbered in some countries.
1765.It Cd options ALTQ
1766Enabled ALTQ (Alternate Queueing).
1767For simple rate-limiting, use
1768.Xr tbrconfig 8
1769to set up the interface transmission rate.
1770To use queueing disciplines, their appropriate kernel options should also
1771be defined (documented below).
1772Queueing disciplines are managed by
1773.Xr altqd 8 .
1774See
1775.Xr altq 9
1776for details.
1777.It Cd options ALTQ_HFSC
1778Include support for ALTQ-implemented HFSC (Hierarchical Fair Service Curve)
1779module.
1780HFSC supports both link-sharing and guaranteed real-time services.
1781HFSC employs a service curve based QoS model, and its unique feature
1782is an ability to decouple delay and bandwidth allocation.
1783Requires
1784.Em ALTQ_RED
1785to use the RED queueing discipline on HFSC classes, or
1786.Em ALTQ_RIO
1787to use the RIO queueing discipline on HFSC classes.
1788This option assumes
1789.Em ALTQ .
1790.It Cd options ALTQ_PRIQ
1791Include support for ALTQ-implemented PRIQ (Priority Queueing).
1792PRIQ implements a simple priority-based queueing discipline.
1793A higher priority class is always served first.
1794Requires
1795.Em ALTQ_RED
1796to use the RED queueing discipline on HFSC classes, or
1797.Em ALTQ_RIO
1798to use the RIO queueing discipline on HFSC classes.
1799This option assumes
1800.Em ALTQ .
1801.It Cd options ALTQ_WFQ
1802Include support for ALTQ-implemented WFQ (Weighted Fair Queueing).
1803WFQ implements a weighted-round robin scheduler for a set of queues.
1804A weight can be assigned to each queue to give a different proportion
1805of the link capacity.
1806A hash function is used to map a flow to one of a set of queues.
1807This option assumes
1808.Em ALTQ .
1809.It Cd options ALTQ_FIFOQ
1810Include support for ALTQ-implemented FIFO queueing.
1811FIFOQ is a simple drop-tail FIFO (First In, First Out) queueing discipline.
1812This option assumes
1813.Em ALTQ .
1814.It Cd options ALTQ_RIO
1815Include support for ALTQ-implemented RIO (RED with In/Out).
1816The original RIO has 2 sets of RED parameters; one for in-profile
1817packets and the other for out-of-profile packets.
1818At the ingress of the network, profile meters tag packets as IN or
1819OUT based on contracted profiles for customers.
1820Inside the network, IN packets receive preferential treatment by
1821the RIO dropper.
1822ALTQ/RIO has 3 drop precedence levels defined for the Assured Forwarding
1823PHB of DiffServ (RFC 2597).
1824This option assumes
1825.Em ALTQ .
1826.It Cd options ALTQ_BLUE
1827Include support for ALTQ-implemented Blue buffer management.
1828Blue is another active buffer management mechanism.
1829This option assumes
1830.Em ALTQ .
1831.It Cd options ALTQ_FLOWVALVE
1832Include support for ALTQ-implemented Flowvalve.
1833Flowvalve is a simple implementation of a RED penalty box that identifies
1834and punishes misbehaving flows.
1835This option requires
1836.Em ALTQ_RED
1837and assumes
1838.Em ALTQ .
1839.It Cd options ALTQ_CDNR
1840Include support for ALTQ-implemented CDNR (diffserv traffic conditioner)
1841packet marking/manipulation.
1842Traffic conditioners are components to meter, mark, or drop incoming
1843packets according to some rules.
1844As opposed to queueing disciplines, traffic conditioners handle incoming
1845packets at an input interface.
1846This option assumes
1847.Em ALTQ .
1848.It Cd options ALTQ_NOPCC
1849Disables use of processor cycle counter to measure time in ALTQ.
1850This option should be defined for a non-Pentium i386 CPU which does not
1851have TSC, SMP (per-CPU counters are not in sync), or power management
1852which affects processor cycle counter.
1853This option assumes
1854.Em ALTQ .
1855.It Cd options ALTQ_IPSEC
1856Include support for IPsec in IPv4 ALTQ.
1857This option assumes
1858.Em ALTQ .
1859.It Cd options ALTQ_JOBS
1860Include support for ALTQ-implemented JoBS (Joint Buffer Management
1861and Scheduling).
1862This option assumes
1863.Em ALTQ .
1864.It Cd options ALTQ_AFMAP
1865Include support for an undocumented ALTQ feature that is used to map an IP
1866flow to an ATM VC (Virtual Circuit).
1867This option assumes
1868.Em ALTQ .
1869.It Cd options ALTQ_LOCALQ
1870Include support for ALTQ-implemented local queues.
1871Its practical use is undefined.
1872Assumes
1873.Em ALTQ .
1874.It Cd options SUBNETSARELOCAL
1875Sets default value for net.inet.ip.subnetsarelocal variable, which
1876controls whether non-directly-connected subnets of connected networks
1877are considered "local" for purposes of choosing the MSS for a TCP
1878connection.
1879This is mostly present for historic reasons and completely irrelevant if
1880you enable Path MTU discovery.
1881.It Cd options HOSTZEROBROADCAST
1882Sets default value for net.inet.ip.hostzerobroadcast variable, which
1883controls whether the zeroth host address of each connected subnet is
1884also considered a broadcast address.
1885Default value is "1", for compatibility with old systems; if this is
1886set to zero on all hosts on a subnet, you should be able to fit an extra
1887host per subnet on the
1888".0" address.
1889.It Cd options MCLSHIFT=value
1890This option is the base-2 logarithm of the size of mbuf clusters.
1891The
1892.Bx
1893networking stack keeps network packets in a linked
1894list, or chain, of kernel buffer objects called mbufs.
1895The system provides larger mbuf clusters as an optimization for
1896large packets, instead of using long chains for large packets.
1897The mbuf cluster size,
1898or
1899.Em MCLBYTES ,
1900must be a power of two, and is computed as two raised to the power
1901.Em MCLSHIFT .
1902On systems with Ethernet network adapters,
1903.Em MCLSHIFT
1904is often set to 11, giving 2048-byte mbuf clusters, large enough to
1905hold a 1500-byte
1906.Tn Ethernet
1907frame in a single cluster.
1908Systems with network interfaces supporting larger frame sizes like
1909.Tn ATM ,
1910.Tn FDDI ,
1911or
1912.Tn HIPPI
1913may perform better with
1914.Em MCLSHIFT
1915set to 12 or 13, giving mbuf cluster sizes of 4096 and 8192 bytes,
1916respectively.
1917.It Cd options ISO,TPIP
1918Include support for the ubiquitous
1919.Tn OSI
1920protocol stack.
1921See
1922.Xr iso 4
1923for details.
1924This option assumes
1925.Em INET .
1926.It Cd options EON
1927Include support for tunneling
1928.Tn OSI
1929protocols over
1930.Tn IP .
1931Known to be broken, or at least very fragile, and undocumented.
1932.It Cd options NETATALK
1933Include support for the
1934.Tn AppleTalk
1935protocol stack.
1936The kernel provides provision for the
1937.Em Datagram Delivery Protocol
1938(DDP), providing SOCK_DGRAM support and
1939.Tn AppleTalk
1940routing.
1941This stack is used by the
1942.Em NETATALK
1943package, which adds support for
1944.Tn AppleTalk
1945server services via user libraries and applications.
1946.It Cd options BLUETOOTH
1947Include support for the
1948.Tn Bluetooth
1949protocol stack.
1950See
1951.Xr bluetooth 4
1952for details.
1953.It Cd options IPNOPRIVPORTS
1954Normally, only root can bind a socket descriptor to a so-called
1955.Dq privileged
1956.Tn TCP
1957port, that is, a port number in the range 0-1023.
1958This option eliminates those checks from the kernel.
1959This can be useful if there is a desire to allow daemons without
1960privileges to bind those ports, e.g., on firewalls.
1961The security tradeoffs in doing this are subtle.
1962This option should only be used by experts.
1963.It Cd options TCP_COMPAT_42
1964.Tn TCP
1965bug compatibility with
1966.Bx 4.2 .
1967In
1968.Bx 4.2 ,
1969.Tn TCP
1970sequence numbers were 32-bit signed values.
1971Modern implementations of TCP use unsigned values.
1972This option clamps the initial sequence number to start in
1973the range 2^31 rather than the full unsigned range of 2^32.
1974Also, under
1975.Bx 4.2 ,
1976keepalive packets must contain at least one byte or else
1977the remote end would not respond.
1978.It Cd options TCP_DEBUG
1979Record the last
1980.Em TCP_NDEBUG
1981TCP packets with SO_DEBUG set, and decode to the console if
1982.Em tcpconsdebug
1983is set.
1984.It Cd options TCP_NDEBUG
1985Number of packets to record for
1986.Em TCP_DEBUG .
1987Defaults to 100.
1988.It Cd options TCP_SENDSPACE=value
1989.It Cd options TCP_RECVSPACE=value
1990These options set the max TCP window size to other sizes than the default.
1991The TCP window sizes can be altered via
1992.Xr sysctl 8
1993as well.
1994.It Cd options TCP_INIT_WIN=value
1995This option sets the initial TCP window size for non-local connections,
1996which is used when the transmission starts.
1997The default size is 1, but if the machine should act more aggressively,
1998the initial size can be set to some other value.
1999The initial TCP window size can be set via
2000.Xr sysctl 8
2001as well.
2002.It Cd options IPFILTER_LOG
2003This option, in conjunction with
2004.Em pseudo-device ipfilter ,
2005enables logging of IP packets using IP-Filter.
2006.It Cd options IPFILTER_LOOKUP
2007This option enables the
2008IP-Filter
2009.Xr ippool 8
2010functionality to be enabled.
2011.It Cd options IPFILTER_COMPAT
2012This option enables older IP-Filter binaries to work.
2013.It Cd options IPFILTER_DEFAULT_BLOCK
2014This option sets the default policy of IP-Filter.
2015If it is set, IP-Filter will block packets by default.
2016.It Cd options BRIDGE_IPF
2017This option causes
2018.Em bridge
2019devices to use the IP and/or IPv6 filtering hooks, forming
2020a link-layer filter that uses protocol-layer rules.
2021This option assumes the presence of
2022.Em pseudo-device ipfilter .
2023.It Cd options MBUFTRACE
2024This option can help track down mbuf leaks.
2025When enabled, mbufs are tagged with the devices and protocols using them,
2026which slightly decreases network performance.
2027This additional information can be viewed with
2028.Xr netstat 1 :
2029.Dl Ic netstat Fl mssv
2030Not all devices or protocols support this option.
2031.El
2032.Ss Sysctl Related Options
2033.Bl -ohang
2034.It Cd options SYSCTL_DISALLOW_CREATE
2035Disallows the creation or deletion of nodes from the sysctl tree, as
2036well as the assigning of descriptions to nodes that lack them, by any
2037process.
2038These operations are still available to kernel sub-systems, including
2039loadable kernel modules.
2040.It Cd options SYSCTL_DISALLOW_KWRITE
2041Prevents processes from adding nodes to the sysctl tree that make
2042existing kernel memory areas writable.
2043Sections of kernel memory can still be read and new nodes that own
2044their own data may still be writable.
2045.It Cd options SYSCTL_DEBUG_SETUP
2046Causes the SYSCTL_SETUP routines to print a brief message when they
2047are invoked.
2048This is merely meant as an aid in determining the order in which
2049sections of the tree are created.
2050.It Cd options SYSCTL_DEBUG_CREATE
2051Prints a message each time
2052.Fn sysctl_create ,
2053the function that adds nodes to the tree, is called.
2054.It Cd options SYSCTL_INCLUDE_DESCR
2055Causes the kernel to include short, human readable descriptions for
2056nodes in the sysctl tree.
2057The descriptions can be retrieved programmatically (see
2058.Xr sysctl 3 ) ,
2059or by the sysctl binary itself (see
2060.Xr sysctl 8 ) .
2061The descriptions are meant to give an indication of the purpose and/or
2062effects of a given node's value, not replace the documentation for the
2063given subsystem as a whole.
2064.El
2065.Ss System V IPC Options
2066.Bl -ohang
2067.It Cd options SYSVMSG
2068Includes support for
2069.At V
2070style message queues.
2071See
2072.Xr msgctl 2 ,
2073.Xr msgget 2 ,
2074.Xr msgrcv 2 ,
2075.Xr msgsnd 2 .
2076.It Cd options SYSVSEM
2077Includes support for
2078.At V
2079style semaphores.
2080See
2081.Xr semctl 2 ,
2082.Xr semget 2 ,
2083.Xr semop 2 .
2084.It Cd options SEMMNI=value
2085Sets the number of
2086.At V
2087style semaphore identifiers.
2088The GENERIC config file for your port will have the default.
2089.It Cd options SEMMNS=value
2090Sets the number of
2091.At V
2092style semaphores in the system.
2093The GENERIC config file for your port will have the default.
2094.It Cd options SEMUME=value
2095Sets the maximum number of undo entries per process for
2096.At V
2097style semaphores.
2098The GENERIC config file for your port will have the default.
2099.It Cd options SEMMNU=value
2100Sets the number of undo structures in the system for
2101.At V
2102style semaphores.
2103The GENERIC config file for your port will have the default.
2104.It Cd options SYSVSHM
2105Includes support for
2106.At V
2107style shared memory.
2108See
2109.Xr shmat 2 ,
2110.Xr shmctl 2 ,
2111.Xr shmdt 2 ,
2112.Xr shmget 2 .
2113.It Cd options SHMMAXPGS=value
2114Sets the maximum number of
2115.At V
2116style shared memory pages that are available through the
2117.Xr shmget 2
2118system call.
2119Default value is 1024 on most ports.
2120See
2121.Pa /usr/include/machine/vmparam.h
2122for the default.
2123.El
2124.Ss VM Related Options
2125.Bl -ohang
2126.It Cd options NMBCLUSTERS=value
2127The number of mbuf clusters the kernel supports.
2128Mbuf clusters are MCLBYTES in size (usually 2k).
2129This is used to compute the size of the kernel VM map
2130.Em mb_map ,
2131which maps mbuf clusters.
2132Default on most ports is 1024 (2048 with
2133.Dq options GATEWAY
2134).
2135See
2136.Pa /usr/include/machine/param.h
2137for exact default information.
2138Increase this value if you get
2139.Dq mclpool limit reached
2140messages.
2141.It Cd options NKMEMPAGES=value
2142.It Cd options NKMEMPAGES_MIN=value
2143.It Cd options NKMEMPAGES_MAX=value
2144Size of kernel VM map
2145.Em kmem_map ,
2146in PAGE_SIZE-sized chunks (the VM page size; this value may be read
2147from the
2148.Xr sysctl 8
2149variable
2150.Em hw.pagesize
2151).
2152This VM map is used to map the kernel malloc arena.
2153The kernel attempts to auto-size this map based on the amount of
2154physical memory in the system.
2155Platform-specific code may place bounds on this computed size,
2156which may be viewed with the
2157.Xr sysctl 8
2158variable
2159.Em vm.nkmempages .
2160See
2161.Pa /usr/include/machine/param.h
2162for the default upper and lower bounds.
2163The related options
2164.Sq NKMEMPAGES_MIN
2165and
2166.Sq NKMEMPAGES_MAX
2167allow the bounds to be overridden in the kernel configuration file.
2168These options are provided in the event the computed value is
2169insufficient resulting in an
2170.Dq out of space in kmem_map
2171panic.
2172.It Cd options SB_MAX=value
2173Sets the max size in bytes that a socket buffer is allowed to occupy.
2174The default is 256k, but sometimes it needs to be increased, for example
2175when using large TCP windows.
2176This option can be changed via
2177.Xr sysctl 8
2178as well.
2179.It Cd options SOMAXKVA=value
2180Sets the maximum size of kernel virtual memory that the socket buffers
2181are allowed to use.
2182The default is 16MB, but in situations where for example large TCP
2183windows are used this value must also be increased.
2184This option can be changed via
2185.Xr sysctl 8
2186as well.
2187.It Cd options BUFCACHE=value
2188Size of the buffer cache as a percentage of total available
2189.Tn RAM .
2190Ignored if BUFPAGES is also specified.
2191.It Cd options NBUF=value
2192Sets the number of buffer headers available, i.e., the number of
2193open files that may have a buffer cache entry.
2194Each buffer header
2195requires MAXBSIZE (machine dependent, but usually 65536) bytes.
2196The default value is machine dependent, but is usually equal to the
2197value of BUFPAGES.
2198If an architecture dependent VM_MAX_KERNEL_BUF constant is defined
2199then NBUF may be reduced at run time so that the storage allocated
2200for buffer headers doesn't exceed that limit.
2201.It Cd options BUFPAGES=value
2202These options set the number of pages available for the buffer cache.
2203Their default value is a machine dependent value, often calculated as
2204between 5% and 10% of total available
2205.Tn RAM .
2206.It Cd options MAXTSIZ=bytes
2207Sets the maximum size limit of a process' text segment.
2208See
2209.Pa /usr/include/machine/vmparam.h
2210for the port-specific default.
2211.It Cd options DFLDSIZ=bytes
2212Sets the default size limit of a process' data segment, the value that
2213will be returned as the soft limit for
2214.Dv RLIMIT_DATA
2215(as returned by
2216.Xr getrlimit 2 ) .
2217See
2218.Pa /usr/include/machine/vmparam.h
2219for the port-specific default.
2220.It Cd options MAXDSIZ=bytes
2221Sets the maximum size limit of a process' data segment, the value that
2222will be returned as the hard limit for
2223.Dv RLIMIT_DATA
2224(as returned by
2225.Xr getrlimit 2 ) .
2226See
2227.Pa /usr/include/machine/vmparam.h
2228for the port-specific default.
2229.It Cd options DFLSSIZ=bytes
2230Sets the default size limit of a process' stack segment, the value that
2231will be returned as the soft limit for
2232.Dv RLIMIT_STACK
2233(as returned by
2234.Xr getrlimit 2 ) .
2235See
2236.Pa /usr/include/machine/vmparam.h
2237for the port-specific default.
2238.It Cd options MAXSSIZ=bytes
2239Sets the maximum size limit of a process' stack segment, the value that
2240will be returned as the hard limit for
2241.Dv RLIMIT_STACK
2242(as returned by
2243.Xr getrlimit 2 ) .
2244See
2245.Pa /usr/include/machine/vmparam.h
2246for the port-specific default.
2247.It Cd options DUMP_ON_PANIC=integer
2248Defaults to one.
2249If set to zero, the kernel will not dump to the dump device when
2250it panics, though dumps can still be forced via
2251.Xr ddb 4
2252with the
2253.Dq sync
2254command.
2255Note that this sets the value of the
2256.Em kern.dump_on_panic
2257.Xr sysctl 3
2258variable which may be changed at run time -- see
2259.Xr sysctl 8
2260for details.
2261.It Cd options USE_TOPDOWN_VM
2262User space memory allocations (as made by
2263.Xr mmap 2 )
2264will be arranged in a
2265.Dq top down
2266fashion instead of the traditional
2267.Dq upwards from MAXDSIZ \&+ vm_daddr
2268method.
2269This includes the placement of
2270.Xr ld.so 1 .
2271Arranging memory in this manner allows either (or both of) the heap or
2272.Xr mmap 2
2273allocated space to grow larger than traditionally possible.
2274This option is not available on all ports, but is instead expected to be
2275offered on a port-by-port basis, after which some ports will commit to
2276using it by default.
2277See the files
2278.Pa /usr/include/uvm/uvm_param.h
2279for some implementation details, and
2280.Pa /usr/include/machine/vmparam.h
2281for port specific details including availability.
2282.It Cd options VMSWAP
2283Enable paging device/file support.
2284This option is on by default.
2285.It Cd options PDPOLICY_CLOCKPRO
2286Use CLOCK-Pro, an alternative page replace policy.
2287.El
2288.Ss Security Options
2289.Bl -ohang
2290.It Cd options INSECURE
2291Initializes the kernel security level with \-1 instead of 0.
2292This means that the system always starts in secure level \-1 mode, even when
2293running multiuser, unless the securelevel variable is set to value > \-1 in
2294.Pa /etc/rc.conf .
2295In this case the kernel security level will be raised to that value when the
2296.Pa /etc/rc.d/securelevel
2297script is run during system startup.
2298See the manual page for
2299.Xr init 8
2300for details on the implications of this.
2301The kernel secure level may manipulated by the superuser by altering the
2302.Em kern.securelevel
2303.Xr sysctl 3
2304variable (the secure level may only be lowered by a call from process ID 1,
2305i.e.,
2306.Xr init 8 ) .
2307See also
2308.Xr secmodel_securelevel 9 ,
2309.Xr sysctl 8
2310and
2311.Xr sysctl 3 .
2312.It Cd options VERIFIED_EXEC_FP_MD5
2313Enables support for MD5 hashes in Veriexec.
2314.It Cd options VERIFIED_EXEC_FP_SHA1
2315Enables support for SHA1 hashes in Veriexec.
2316.It Cd options VERIFIED_EXEC_FP_RMD160
2317Enables support for RMD160 hashes in Veriexec.
2318.It Cd options VERIFIED_EXEC_FP_SHA256
2319Enables support for SHA256 hashes in Veriexec.
2320.It Cd options VERIFIED_EXEC_FP_SHA384
2321Enables support for SHA384 hashes in Veriexec.
2322.It Cd options VERIFIED_EXEC_FP_SHA512
2323Enables support for SHA512 hashes in Veriexec.
2324.It Cd options PAX_MPROTECT=value
2325Enables PaX MPROTECT,
2326.Xr mprotect 2
2327restrictions from the PaX project.
2328.Pp
2329The
2330.Ar value
2331is the default value for the
2332.Em global
2333knob, see
2334.Xr sysctl 3 .
2335If 0, PaX MPROTECT will be enabled only if explicitly set on programs
2336using
2337.Xr paxctl 8 .
2338If 1, PaX MPROTECT will be enabled for all programs.
2339Programs can be exempted using
2340.Xr paxctl 8 .
2341.Pp
2342See
2343.Xr security 7
2344for more details.
2345.It Cd options PAX_SEGVGUARD=value
2346Enables PaX Segvguard.
2347.Pp
2348The
2349.Ar value
2350is the default value for the
2351.Em global
2352knob, see
2353.Xr sysctl 3 .
2354If 0, PaX Segvguard will be enabled only if explicitly set on programs
2355using
2356.Xr paxctl 8 .
2357If 1, PaX Segvguard will be enabled to all programs, and exemption can
2358be done using
2359.Xr paxctl 8 .
2360.Pp
2361See
2362.Xr security 7
2363for more details.
2364.It Cd options PAX_ASLR=value
2365Enables PaX ASLR.
2366.Pp
2367The
2368.Ar value
2369is the default value for the
2370.Em global
2371knob, see
2372.Xr sysctl 3 .
2373If 0, PaX ASLR will be enabled only if explicitly set on programs
2374using
2375.Xr paxctl 8 .
2376If 1, PaX ASLR will be enabled to all programs, and exemption can
2377be done using
2378.Xr paxctl 8 .
2379.Pp
2380See
2381.Xr security 7
2382for more details.
2383.It Cd options USER_VA0_DISABLE_DEFAULT=value
2384Sets the initial value of the flag which controls whether user programs
2385can map virtual address 0.
2386The flag can be changed at runtime by
2387.Xr sysctl 3 .
2388.El
2389.Ss amiga-specific Options
2390.Bl -ohang
2391.It Cd options BB060STUPIDROM
2392When the bootloader (which passes
2393.Tn AmigaOS
2394.Tn ROM
2395information) claims we have a 68060
2396.Tn CPU
2397without
2398.Tn FPU ,
2399go look into the Processor Configuration Register (PCR) to find out.
2400You need this with
2401.Tn Amiga
2402.Tn ROM Ns s
2403up to (at least) V40.xxx (OS3.1),
2404when you boot via the bootblocks and don't have a DraCo.
2405.It Cd options IOBZCLOCK=frequency
2406The IOBlix boards come with two different serial master clocks: older ones
2407use 24 MHz, newer ones use 22.1184 MHz.
2408The driver normally assumes the latter.
2409If your board uses 24 MHz, you can recompile your kernel with
2410options IOBZCLOCK=24000000
2411or patch the kernel variable
2412.Tn iobzclock
2413to the same value.
2414.It Cd options LIMITMEM=value
2415If there, limit the part of the first memory bank used by
2416.Nx
2417to value megabytes.
2418Default is unlimited.
2419.It Cd options P5PPC68KBOARD
2420Add special support for Phase5 mixed 68k+PPC boards.
2421Currently, this only affects rebooting from
2422.Nx
2423and is only needed on 68040+PPC, not on
242468060+PPC; without this, affected machines will hang after
2425.Nx
2426has shut
2427down and will only restart after a keyboard reset or a power cycle.
2428.El
2429.Ss atari-specific Options
2430.Bl -ohang
2431.It Cd options DISKLABEL_AHDI
2432Include support for AHDI (native Atari) disklabels.
2433.It Cd options DISKLABEL_NBDA
2434Include support for
2435.Nx Ns Tn /atari
2436labels.
2437If you don't set this option, it will be set automatically.
2438.Nx Ns Tn /atari
2439will not work without it.
2440.It Cd options FALCON_SCSI
2441Include support for the 5380-SCSI configuration as found on the Falcon.
2442.It Cd options RELOC_KERNEL
2443If set, the kernel will relocate itself to TT-RAM, if possible.
2444This will give you a slightly faster system.
2445.Em Beware
2446that on some TT030 systems,
2447the system will frequently dump with MMU-faults with this option enabled.
2448.It Cd options SERCONSOLE
2449Allow the modem1-port to act as the system-console.
2450A carrier should be active on modem1 during system boot to active
2451the console functionality.
2452.It Cd options TT_SCSI
2453Include support for the 5380-SCSI configuration as found on the TT030
2454and Hades.
2455.El
2456.Ss i386-specific Options
2457.Bl -ohang
2458.It Cd options CPURESET_DELAY=value
2459Specifies the time (in millisecond) to wait before doing a hardware reset
2460in the last phase of a reboot.
2461This gives the user a chance to see error messages from the shutdown
2462operations (like NFS unmounts, buffer cache flush, etc ...).
2463Setting this to 0 will disable the delay.
2464Default is 2 seconds.
2465.It Cd options VM86
2466Include support for virtual 8086 mode, used by
2467.Tn DOS
2468emulators and X servers to run BIOS code, e.g., for some VESA routines.
2469.It Cd options USER_LDT
2470Include i386-specific system calls for modifying the local descriptor table,
2471used by Windows emulators.
2472.It Cd options PAE
2473Enable
2474.Tn PAE (Physical Address Extension)
2475mode.
2476.Tn PAE
2477permits up to 36 bits physical addressing (64GB of physical memory), and
2478turns physical addresses to 64 bits entities in the memory management
2479subsystem.
2480Userland virtual address space remains at 32 bits (4GB).
2481.Tn PAE
2482mode is required to enable the
2483.Tn NX/XD (No-eXecute/eXecute Disable)
2484bit for pages, which allows marking certain ones as not being executable.
2485Any attempt to execute code from such a page will raise an exception.
2486.It Cd options REALBASEMEM=integer
2487Overrides the base memory size passed in from the boot block.
2488(Value given in kilobytes.)
2489Use this option only if the boot block reports the size incorrectly.
2490(Note that some
2491.Tn BIOS Ns es
2492put the extended
2493.Tn BIOS
2494data area at the top of base memory, and therefore report a smaller
2495base memory size to prevent programs overwriting it.
2496This is correct behavior, and you should not use the
2497.Em REALBASEMEM
2498option to access this memory).
2499.It Cd options REALEXTMEM=integer
2500Overrides the extended memory size passed in from the boot block.
2501(Value given in kilobytes.
2502Extended memory does not include the first megabyte.)
2503Use this option only if the boot block reports the size incorrectly.
2504.It Cd options CYRIX_CACHE_WORKS
2505Relevant only to the Cyrix 486DLC CPU.
2506This option is used to turn on the cache in hold-flush mode.
2507It is not turned on by default because it is known to have problems in
2508certain motherboard implementations.
2509.It Cd options CYRIX_CACHE_REALLY_WORKS
2510Relevant only to the Cyrix 486DLC CPU.
2511This option is used to turn on the cache in write-back mode.
2512It is not turned on by default because it is known to have problems in
2513certain motherboard implementations.
2514In order for this option to take effect, option
2515.Em CYRIX_CACHE_WORKS
2516must also be specified.
2517.It Cd options PCIBIOS
2518Enable support for initializing the
2519.Tn PCI
2520bus using information from the
2521.Tn BIOS .
2522See
2523.Xr pcibios 4
2524for details.
2525.It Cd options KSTACK_CHECK_DR0
2526Detect kernel stack overflow using DR0 register.
2527This option uses DR0 register exclusively so you can't use DR0 register for
2528other purpose (e.g., hardware breakpoint) if you turn this on.
2529.It Cd options MTRR
2530Include support for accessing MTRR registers from user-space.
2531See
2532.Xr i386_get_mtrr 2 .
2533.It Cd options BEEP_ONHALT
2534Make the system speaker emit several beeps when it is completely safe to
2535power down the computer after a
2536.Xr halt 8
2537command.
2538Requires
2539.Xr sysbeep 4
2540support.
2541.It Cd options BEEP_ONHALT_COUNT=times
2542Number of times to beep the speaker when
2543.Cd options BEEP_ONHALT
2544is enabled.
2545Defaults to 3.
2546.It Cd options BEEP_ONHALT_PITCH=hz
2547The tone frequency used when
2548.Cd options BEEP_ONHALT
2549option, in hertz.
2550Defaults to 1500.
2551.It Cd options BEEP_ONHALT_PERIOD=msecs
2552The duration of each beep when
2553.Cd options BEEP_ONHALT
2554is enabled, in milliseconds.
2555Defaults to 250.
2556.It Cd options MULTIBOOT
2557Makes the kernel Multiboot-compliant, allowing it to be booted through
2558a Multiboot-compliant boot manager such as GRUB.
2559See
2560.Xr multiboot 8
2561for more information.
2562.It Cd options SPLASHSCREEN
2563Display a splash screen during boot.
2564.It Cd options SPLASHSCREEN_PROGRESS
2565Display a progress bar at the splash screen during boot.
2566This option requires
2567.Em SPLASHSCREEN .
2568.El
2569.Ss isa-specific Options
2570Options specific to
2571.Xr isa 4
2572busses.
2573.Bl -ohang
2574.It Cd options PCIC_ISA_ALLOC_IOBASE=address, PCIC_ISA_ALLOC_IOSIZE=size
2575Control the section of IO bus space used for PCMCIA bus space mapping.
2576Ideally the probed defaults are satisfactory, however in practice
2577that is not always the case.
2578See
2579.Xr pcmcia 4
2580for details.
2581.It Cd options PCIC_ISA_INTR_ALLOC_MASK=mask
2582Controls the allowable interrupts that may be used for
2583.Tn PCMCIA
2584devices.
2585This mask is a logical-or of power-of-2s of allowable interrupts:
2586.Bd -literal -offset 04n
2587.Em "IRQ Val      IRQ Val      IRQ Val       IRQ Val"
2588 0  0x0001    4  0x0010    8  0x0100    12  0x1000
2589 1  0x0002    5  0x0020    9  0x0200    13  0x2000
2590 2  0x0004    6  0x0040   10  0x0400    14  0x4000
2591 3  0x0008    7  0x0080   11  0x0800    15  0x8000
2592.Ed
2593.It Cd options PCKBC_CNATTACH_SELFTEST
2594Perform a self test of the keyboard controller before attaching it as a
2595console.
2596This might be necessary on machines where we boot on cold iron, and
2597pckbc refuses to talk until we request a self test.
2598Currently only the netwinder port uses it.
2599.It Cd options PCKBD_CNATTACH_MAY_FAIL
2600If this option is set the PS/2 keyboard will not be used as the console
2601if it cannot be found during boot.
2602This allows other keyboards, like USB, to be the console keyboard.
2603.It Cd options PCKBD_LAYOUT=layout
2604Sets the default keyboard layout, see
2605.Xr pckbd 4 .
2606.El
2607.Ss m68k-specific Options
2608.Bl -ohang
2609.It Cd options FPU_EMULATE
2610Include support for MC68881/MC68882 emulator.
2611.It Cd options FPSP
2612Include support for 68040 floating point.
2613.It Cd options M68020,M68030,M68040,M68060
2614Include support for a specific
2615.Tn CPU ,
2616at least one (the one you are using) should be specified.
2617.It Cd options M060SP
2618Include software support for 68060.
2619This provides emulation of unimplemented
2620integer instructions as well as emulation of unimplemented floating point
2621instructions and data types and software support for floating point traps.
2622.El
2623.Ss powerpc-specific Options (OEA Only)
2624.Bl -ohang
2625.It Cd options PMAP_MEMLIMIT=value
2626Limit the amount of memory seen by the kernel to
2627.Ar value
2628bytes.
2629.It Cd options PTEGCOUNT=value
2630Specify the size of the page table as
2631.Ar value
2632PTE groups.
2633Normally, one PTEG is allocated per physical page frame.
2634.El
2635.Ss sparc-specific Options
2636.Bl -ohang
2637.It Cd options AUDIO_DEBUG
2638Enable simple event debugging of the logging of the
2639.Xr audio 4
2640device.
2641.It Cd options BLINK
2642Enable blinking of LED.
2643Blink rate is full cycle every N seconds for
2644N \*[Lt] then current load average.
2645See
2646.Xr getloadavg 3 .
2647.\" .It Cd options COLORFONT_CACHE
2648.\" What does this do?
2649.It Cd options COUNT_SW_LEFTOVERS
2650Count how many times the sw SCSI device has left 3, 2, 1 and 0 in the
2651sw_3_leftover, sw_2_leftover, sw_1_leftover, and sw_0_leftover
2652variables accessible from
2653.Xr ddb 4 .
2654See
2655.Xr sw 4 .
2656.It Cd options DEBUG_ALIGN
2657Adds debugging messages calls when user-requested alignment fault
2658handling happens.
2659.It Cd options DEBUG_EMUL
2660Adds debugging messages calls for emulated floating point and
2661alignment fixing operations.
2662.It Cd options DEBUG_SVR4
2663Prints registers messages calls for emulated SVR4 getcontext and
2664setcontext operations.
2665See
2666.Em options COMPAT_SVR4 .
2667.It Cd options EXTREME_DEBUG
2668Adds debugging functions callable from
2669.Xr ddb 4 .
2670The debug_pagetables, test_region and print_fe_map
2671functions print information about page tables for the SUN4M
2672platforms only.
2673.It Cd options EXTREME_EXTREME_DEBUG
2674Adds extra info to
2675.Em options EXTREME_DEBUG .
2676.It Cd options FPU_CONTEXT
2677Make
2678.Em options COMPAT_SVR4
2679getcontext and setcontext include floating point registers.
2680.It Cd options MAGMA_DEBUG
2681Adds debugging messages to the
2682.Xr magma 4
2683device.
2684.It Cd options RASTERCONS_FULLSCREEN
2685Use the entire screen for the console.
2686.It Cd options RASTERCONS_SMALLFONT
2687Use the Fixed font on the console, instead of the normal font.
2688.It Cd options SUN4
2689Support sun4 class machines.
2690.It Cd options SUN4C
2691Support sun4c class machines.
2692.It Cd options SUN4M
2693Support sun4m class machines.
2694.It Cd options SUN4_MMU3L
2695.\" XXX ???
2696Enable support for sun4 3-level MMU machines.
2697.It Cd options V9
2698Enable SPARC V9 assembler in
2699.Xr ddb 4 .
2700.El
2701.Ss sparc64-specific Options
2702.Bl -ohang
2703.It Cd options AUDIO_DEBUG
2704Enable simple event debugging of the logging of the
2705.Xr audio 4
2706device.
2707.It Cd options BLINK
2708Enable blinking of LED.
2709Blink rate is full cycle every N seconds for
2710N \*[Lt] then current load average.
2711See
2712.Xr getloadavg 3 .
2713.El
2714.Ss x68k-specific Options
2715.Bl -ohang
2716.It Cd options EXTENDED_MEMORY
2717Include support for extended memory, e.g., TS-6BE16 and 060turbo on-board.
2718.It Cd options JUPITER
2719Include support for Jupiter-X MPU accelerator
2720.It Cd options ZSCONSOLE,ZSCN_SPEED=value
2721Use the built-in serial port as the system-console.
2722Speed is specified in bps, defaults to 9600.
2723.It Cd options ITE_KERNEL_ATTR=value
2724Set the kernel message attribute for ITE.
2725Value, an integer, is a logical or of the following values:
2726.Bl -tag -width 4n -compact -offset indent
2727.It 1
2728color inversed
2729.It 2
2730underlined
2731.It 4
2732bolded
2733.El
2734.El
2735.\" The following requests should be uncommented and used where appropriate.
2736.\" .Sh FILES
2737.\" .Sh EXAMPLES
2738.Sh SEE ALSO
2739.Xr config 1 ,
2740.Xr gdb 1 ,
2741.Xr ktrace 1 ,
2742.Xr pmc 1 ,
2743.Xr quota 1 ,
2744.Xr vndcompress 1 ,
2745.Xr gettimeofday 2 ,
2746.Xr i386_get_mtrr 2 ,
2747.Xr i386_iopl 2 ,
2748.Xr msgctl 2 ,
2749.Xr msgget 2 ,
2750.Xr msgrcv 2 ,
2751.Xr msgsnd 2 ,
2752.Xr ntp_adjtime 2 ,
2753.Xr ntp_gettime 2 ,
2754.Xr reboot 2 ,
2755.Xr semctl 2 ,
2756.Xr semget 2 ,
2757.Xr semop 2 ,
2758.Xr shmat 2 ,
2759.Xr shmctl 2 ,
2760.Xr shmdt 2 ,
2761.Xr shmget 2 ,
2762.Xr sysctl 3 ,
2763.Xr apm 4 ,
2764.Xr ddb 4 ,
2765.Xr inet 4 ,
2766.Xr iso 4 ,
2767.Xr md 4 ,
2768.Xr pcibios 4 ,
2769.Xr pcmcia 4 ,
2770.Xr ppp 4 ,
2771.Xr userconf 4 ,
2772.Xr vnd 4 ,
2773.Xr wscons 4 ,
2774.Xr config 5 ,
2775.Xr edquota 8 ,
2776.Xr init 8 ,
2777.Xr mdsetimage 8 ,
2778.Xr mount_cd9660 8 ,
2779.Xr mount_fdesc 8 ,
2780.Xr mount_kernfs 8 ,
2781.Xr mount_lfs 8 ,
2782.Xr mount_mfs 8 ,
2783.Xr mount_msdos 8 ,
2784.Xr mount_nfs 8 ,
2785.Xr mount_ntfs 8 ,
2786.Xr mount_null 8 ,
2787.Xr mount_portal 8 ,
2788.Xr mount_procfs 8 ,
2789.Xr mount_udf 8 ,
2790.Xr mount_umap 8 ,
2791.Xr mount_union 8 ,
2792.Xr mrouted 8 ,
2793.Xr newfs_lfs 8 ,
2794.Xr ntpd 8 ,
2795.Xr quotaon 8 ,
2796.Xr rpc.rquotad 8 ,
2797.Xr sysctl 8 ,
2798.Xr in_getifa 9
2799.Sh HISTORY
2800The
2801.Nm
2802man page first appeared in
2803.Nx 1.3 .
2804.Sh BUGS
2805The
2806.Em EON
2807option should be a pseudo-device, and is also very fragile.
2808