xref: /netbsd-src/share/man/man4/options.4 (revision 796c32c94f6e154afc9de0f63da35c91bb739b45)
1.\"	$NetBSD: options.4,v 1.475 2017/09/14 08:52:41 wiz 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 September 13, 2017
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
176Note that compatibility options for older
177.Nx
178releases includes support for newer releases as well.
179This means that typically only one of these is necessary, with the
180.Cd COMPAT_09
181option enabling all
182.Nx
183compatibility.
184This does not include the
185.Cd COMPAT_43
186or
187.Cd COMPAT_44
188options.
189.Bl -ohang
190.It Cd options COMPAT_09
191Enable binary compatibility with
192.Nx 0.9 .
193This enables support for
19416-bit user, group, and process IDs (following revisions support
19532-bit identifiers).
196It also allows the use of the deprecated
197.Xr getdomainname 3 ,
198.Xr setdomainname 3 ,
199and
200.Xr uname 3
201syscalls.
202This option also allows using numeric file system identifiers rather
203than strings.
204Post
205.Nx 0.9
206versions use string identifiers.
207.It Cd options COMPAT_10
208Enable binary compatibility with
209.Nx 1.0 .
210This option allows the use of the file system name of
211.Dq ufs
212as an alias for
213.Dq ffs .
214The name
215.Dq ffs
216should be used post 1.0 in
217.Pa /etc/fstab
218and other files.
219It also adds old syscalls for the
220.At V
221shared memory interface.
222This was changed post 1.0 to work on 64-bit architectures.
223This option also enables
224.Dq sgtty
225compatibility, without which programs using the old interface produce
226an
227.Dq inappropriate ioctl
228error, and
229.Pa /dev/io
230only works when this option is set in the kernel,
231see
232.Xr io 4
233on ports that support it.
234.It Cd options COMPAT_11
235Enable binary compatibility with
236.Nx 1.1 .
237This allows binaries running on the i386 port to gain direct access to
238the io ports by opening
239.Pa /dev/io
240read/write.
241This functionality was replaced by
242.Xr i386_iopl 2
243post 1.1.
244On the Atari port, the location of the disk label was moved after 1.1.
245When the
246.Em COMPAT_11
247option is set, the kernel will read (pre) 1.1 style disk labels as a
248last resort.
249When a disk label is re-written, the old style label will be replaced
250with a post 1.1 style label.
251This also enables the
252.Em EXEC_ELF_NOTELESS
253option.
254.It Cd options COMPAT_12
255Enable binary compatibility with
256.Nx 1.2 .
257This allows the use of old syscalls for
258.Fn reboot
259and
260.Fn swapon .
261The syscall numbers were changed post 1.2 to add functionality to the
262.Xr reboot 2
263syscall, and the new
264.Xr swapctl 2
265interface was introduced.
266This also enables the
267.Em EXEC_ELF_NOTELESS
268option.
269.It Cd options COMPAT_13
270Enable binary compatibility with
271.Nx 1.3 .
272This allows the use of old syscalls for
273.Fn sigaltstack ,
274and also enables the old
275.Xr swapctl 2
276command
277.Dv SWAP_STATS
278(now called
279.Dv SWAP_OSTATS ) ,
280which does not include the
281.Fa se_path
282member of
283.Va struct swapent .
284.It Cd options COMPAT_14
285Enable binary compatibility with
286.Nx 1.4 .
287This allows some old
288.Xr ioctl 2
289on
290.Xr wscons 4
291to be performed, and allows the
292.Dv NFSSVC_BIOD
293mode of the
294.Xr nfssvc 2
295system call to be used for compatibility with the deprecated nfsiod program.
296.It Cd options COMPAT_15
297Enable binary compatibility with
298.Nx 1.5 .
299Since there were no API changes from
300.Nx 1.5
301and
302.Nx 1.6 ,
303this option does nothing.
304.It Cd options COMPAT_16
305Enable binary compatibility with
306.Nx 1.6 .
307This allows the use of old signal trampoline code which has been deprecated
308with the addition of
309.Xr siginfo 2 .
310.It Cd options COMPAT_20
311Enable binary compatibility with
312.Nx 2.0 .
313This allows the use of old syscalls for
314.Fn statfs ,
315.Fn fstatfs ,
316.Fn getfsstat
317and
318.Fn fhstatfs ,
319which have been deprecated with the addition of the
320.Xr statvfs 2 ,
321.Xr fstatvfs 2 ,
322.Xr getvfsstat 2
323and
324.Xr fhstatvfs 2
325system calls.
326.It Cd options COMPAT_30
327Enable binary compatibility with
328.Nx 3.0 .
329See
330.Xr compat_30 8
331for details about the changes made after the
332.Nx 3.0
333release.
334.It Cd options COMPAT_40
335Enable binary compatibility with
336.Nx 4.0 .
337This allows the use of old
338.Xr ptrace 2
339calls for the SH3 platform.
340It also enables the old
341.Xr mount 2
342system call that did not include the data length parameter.
343The power_event_t structure's pev_switch is filled in.
344.It Cd options COMPAT_43
345Enables compatibility with
346.Bx 4.3 .
347This adds an old syscall for
348.Xr lseek 2 .
349It also adds the ioctls for
350.Dv TIOCGETP
351and
352.Dv TIOCSETP .
353The return values for
354.Xr getpid 2 ,
355.Xr getgid 2 ,
356and
357.Xr getuid 2
358syscalls are modified as well, to return the parent's PID and
359UID as well as the current process's.
360It also enables the deprecated
361.Dv NTTYDISC
362terminal line discipline.
363It also provides backwards compatibility with
364.Dq old
365SIOC[GS]IF{ADDR,DSTADDR,BRDADDR,NETMASK} interface ioctls, including
366binary compatibility with code written before the introduction of the
367sa_len field in sockaddrs.
368It also enables
369support for some older pre
370.Bx 4.4
371socket calls.
372.It Cd options COMPAT_50
373Enable binary compatibility with
374.Nx 5.0 .
375This enables support for the old
376.Ft time_t
377and
378.Ft dev_t
379types as 32 bit, and all the associated kernel interface changes.
380It also enables old
381.Xr gpio 4
382and
383.Xr rnd 4
384interfaces.
385.It Cd options COMPAT_60
386Enable binary compatibility with
387.Nx 6.0 .
388This provides old
389.Xr ccd 4
390interfaces, enables support for old
391.Xr cpuctl 8
392microcode interfaces, and support for the old
393.Ft ptmget
394structure.
395.It Cd options COMPAT_70
396Enable binary compatibility with
397.Nx 7.0 .
398This provides support for old
399.Xr route 4
400interfaces.
401.It Cd options COMPAT_BSDPTY
402This option is currently on by default and enables the pty multiplexer
403.Xr ptm 4
404and
405.Xr ptmx 4
406to find and use ptys named
407.Pa /dev/ptyXX
408(master) and
409.Pa /dev/ttyXX
410(slave).
411Eventually this option will become optional as ptyfs based pseudo-ttys become
412the default, see
413.Xr mount_ptyfs 8 .
414.It Cd options COMPAT_SVR4
415On those architectures that support it, this enables binary
416compatibility with
417.At V.4
418applications built for the same architecture.
419This currently includes the m68k and sparc ports.
420.It Cd options COMPAT_LINUX
421On those architectures that support it, this enables binary
422compatibility with Linux ELF and
423.Xr a.out 5
424applications built for the same architecture.
425This currently includes the alpha, arm, i386, m68k, mips, powerpc and
426x86_64 ports.
427.It Cd options COMPAT_LINUX32
428On those 64 bit architectures that support it, this enables binary
429compatibility with 32 bit Linux binaries.
430For now this is limited to running i386 ELF Linux binaries on amd64.
431.It Cd options COMPAT_SUNOS
432On those architectures that support it, this enables binary
433compatibility with SunOS 4.1 applications built for the same architecture.
434This currently includes the sparc, sparc64 and most or all m68k ports.
435Note that the sparc64 requires the
436.Em COMPAT_NETBSD32
437option for 64-bit kernels, in addition to this option.
438.It Cd options COMPAT_ULTRIX
439On those architectures that support it, this enables binary
440compatibility with ULTRIX applications built for the same architecture.
441This currently is limited to the pmax.
442The functionality of this option is unknown.
443.It Cd options COMPAT_FREEBSD
444On those architectures that support it, this enables binary
445compatibility with
446.Fx
447applications built for the same architecture.
448At the moment this is limited to the i386 port.
449.It Cd options COMPAT_IBCS2
450On those architectures that support it, this enables binary
451compatibility with iBCS2 or SVR3 applications built for the same architecture.
452This is currently limited to the i386 and vax ports.
453.It Cd options COMPAT_OSF1
454On those architectures that support it, this enables binary
455compatibility with Digital
456.Ux
457(formerly OSF/1)
458applications built for the same architecture.
459This is currently limited to the alpha port.
460.It Cd options COMPAT_NOMID
461Enable compatibility with
462.Xr a.out 5
463executables that lack a machine ID.
464This includes
465.Nx 0.8 Ns 's
466ZMAGIC format, and 386BSD and BSDI's
467QMAGIC, NMAGIC, and OMAGIC
468.Xr a.out 5
469formats.
470.It Cd options COMPAT_NETBSD32
471On those architectures that support it, this enables binary
472compatibility with 32-bit applications built for the same architecture.
473This is currently limited to the amd64 and sparc64 ports, and only
474applicable for 64-bit kernels.
475.It Cd options COMPAT_SVR4_32
476On those architectures that support it, this enables binary
477compatibility with 32-bit SVR4 applications built for the same architecture.
478This is currently limited to the sparc64 port, and only applicable for
47964-bit kernels.
480.It Cd options COMPAT_AOUT_M68K
481On m68k architectures which have switched to ELF,
482this enables binary compatibility with
483.Nx Ns /m68k
484.Xr a.out 5
485executables on
486.Nx Ns /m68k
487ELF kernels.
488This handles alignment incompatibility of m68k ABI between
489a.out and ELF which causes the structure padding differences.
490Currently only some system calls which use
491.Va struct stat
492are adjusted and some binaries which use
493.Xr sysctl 3
494to retrieve network details would not work properly.
495.It Cd options EMUL_NATIVEROOT=string
496Just like emulated binaries first try looking up files in
497an emulation root (e.g.
498.Pa /emul/linux )
499before looking them up in real root, this option causes native
500binaries to first look up files in an "emulation" directory too.
501This can be useful to test an amd64 kernel on top of an i386 system
502before full migration: by unpacking the amd64 distribution in e.g.
503.Pa /emul/netbsd64
504and specifying that location as
505.Cd EMUL_NATIVEROOT ,
506native amd64 binaries can be run while the root file system remains
507populated with i386 binaries.
508Beware of
509.Pa /dev
510incompatibilities between i386 and amd64 if you do this.
511.It Cd options EXEC_ELF_NOTELESS
512Run unidentified ELF binaries as
513.Nx
514binaries.
515This might be needed for very old
516.Nx
517ELF binaries on some archs.
518These old binaries didn't contain an appropriate
519.Li .note.netbsd.ident
520section, and thus can't be identified by the kernel as
521.Nx
522binaries otherwise.
523Beware - if this option is on, the kernel would run
524.Em any
525unknown ELF binaries as if they were
526.Nx
527binaries.
528.El
529.Ss Debugging Options
530.Bl -ohang
531.It Cd options DDB
532Compiles in a kernel debugger for diagnosing kernel problems.
533See
534.Xr ddb 4
535for details.
536.Em NOTE :
537not available on all architectures.
538.It Cd options DDB_FROMCONSOLE=integer
539If set to non-zero, DDB may be entered by sending a break on a serial
540console or by a special key sequence on a graphics console.
541A value of "0" ignores console breaks or key sequences.
542If not explicitly specified, the default value is "1".
543Note that this sets the value of the
544.Em ddb.fromconsole
545.Xr sysctl 3
546variable which may be changed at run time -- see
547.Xr sysctl 8
548for details.
549.It Cd options DDB_HISTORY_SIZE=integer
550If this is non-zero, enable history editing in the kernel debugger
551and set the size of the history to this value.
552.It Cd options DDB_ONPANIC
553The default if not specified is
554.Dq 1
555- just enter into DDB.
556If set to
557.Dq 2
558the kernel will
559attempt to print out a stack trace before entering into DDB.
560If set to
561.Dq 0
562the kernel will attempt to print out a stack trace
563and reboot the system.
564If set to
565.Dq -1
566then neither a stack trace is printed or DDB entered -
567it is as if DDB were not compiled into the kernel.
568Note that this sets the value of the
569.Em ddb.onpanic
570.Xr sysctl 3
571variable which may be changed at run time -- see
572.Xr sysctl 8
573for details.
574.It Cd options DDB_COMMANDONENTER=string
575This option specify commands which will be executed on each entry to DDB.
576This sets the default value of the
577.Em ddb.commandonenter
578.Xr sysctl 3
579variable which may be changed at run time.
580.It Cd options DDB_BREAK_CHAR=integer
581This option overrides using break to enter the kernel debugger
582on the serial console.
583The value given is the ASCII value to be used instead.
584This is currently only supported by the com driver.
585.It Cd options DDB_VERBOSE_HELP
586This option adds more verbose descriptions to the
587.Em help
588command.
589.It Cd options KGDB
590Compiles in a remote kernel debugger stub for diagnosing kernel problems
591using the
592.Dq remote target
593feature of gdb.
594See
595.Xr gdb 1
596for details.
597.Em NOTE :
598not available on all architectures.
599.It Cd options KGDB_DEV
600Device number
601.Po
602as a
603.Dv dev_t
604.Pc
605of kgdb device.
606.It Cd options KGDB_DEVADDR
607Memory address of kgdb device.
608.It Cd options KGDB_DEVMODE
609Permissions of kgdb device.
610.It Cd options KGDB_DEVNAME
611Device name of kgdb device.
612.It Cd options KGDB_DEVRATE
613Baud rate of kgdb device.
614.It Cd makeoptions DEBUG="-g"
615The
616.Fl g
617flag causes
618.Pa netbsd.gdb
619to be built in addition to
620.Pa netbsd .
621.Pa netbsd.gdb
622is useful for debugging kernel crash dumps with gdb.
623See
624.Xr gdb 1
625for details.
626This also turns on
627.Em options DEBUG
628(which see).
629.It Cd options DEBUG
630Turns on miscellaneous kernel debugging.
631Since options are turned into preprocessor defines (see above),
632.Em options DEBUG
633is equivalent to doing a
634.Em #define DEBUG
635throughout the kernel.
636Much of the kernel has
637.Em #ifdef DEBUG
638conditionalized debugging code.
639Note that many parts of the kernel (typically device drivers) include their own
640.Em #ifdef XXX_DEBUG
641conditionals instead.
642This option also turns on certain other options,
643which may decrease system performance.
644Systems with this option are not suitable for regular use, and are
645intended only for debugging or looking for bugs.
646.It Cd options DIAGNOSTIC
647Adds code to the kernel that does internal consistency checks.
648This code will cause the kernel to panic if corruption of internal data
649structures is detected.
650Historically, the performance degradation is sufficiently small that
651it is reasonable for systems with
652.Em options DIAGNOSTIC
653to be in production use, with the real consideration not being
654performance but instead a preference for more panics versus continued
655operation with undetected problems.
656.It Cd options LOCKDEBUG
657Adds code to the kernel to detect incorrect use of locking primitives
658(mutex, rwlock).
659This code will cause the kernel to check for dead lock conditions.
660It will also check for memory being freed to not contain initialised
661lock primitives.
662Functions for use in
663.Xr ddb 4
664to check lock chains etc. are also enabled.
665These checks are very expensive and can decrease performance on
666multi-processor machines by a factor of three.
667.It Cd options KDTRACE_HOOKS
668Adds hooks for the DTrace tracing facility, which allows users to
669analyze many aspects of system and application behavior.
670See
671.Xr dtrace 1
672for details.
673.It Cd options KSTACK_CHECK_MAGIC
674Check kernel stack usage and panic if stack overflow is detected.
675This check is performance sensitive because it scans stack on each context
676switch.
677.It Cd options KTRACE
678Add hooks for the system call tracing facility, which allows users to
679watch the system call invocation behavior of processes.
680See
681.Xr ktrace 1
682for details.
683.It Cd options MSGBUFSIZE=integer
684This option sets the size of the kernel message buffer.
685This buffer holds the kernel output of
686.Fn printf
687when not (yet) read by
688.Xr syslogd 8 .
689This is particularly useful when the system has crashed and you wish to lookup
690the kernel output from just before the crash.
691Also, since the autoconfig output becomes more and more verbose,
692it sometimes happens that the message buffer overflows before
693.Xr syslogd 8
694was able to read it.
695Note that not all systems are capable of obtaining a variable sized message
696buffer.
697There are also some systems on which memory contents are not preserved
698across reboots.
699.It Cd options KERNHIST
700Enables the kernel history logs, which create in-memory traces of
701various kernel activities.
702These logs can be displayed by using
703.Cm show kernhist
704from DDB.
705See the kernel source file
706.Pa sys/kern/kern_history.c
707and the
708.Xr kernhist 9
709manual for details.
710.It Cd options KERNHIST_PRINT
711Prints the kernel history logs on the system console as entries are added.
712Note that the output is
713.Em extremely
714voluminous, so this option is really only useful for debugging
715the very earliest parts of kernel initialization.
716.It Cd options UVMHIST
717Like
718.Em KERNHIST ,
719it enables the UVM history logs.
720These logs can be displayed by using
721.Cm show kernhist
722from DDB.
723See the kernel source file
724.Pa sys/uvm/uvm_stat.c
725for details.
726.It Cd options UVMHIST_PRINT
727Like
728.Em UVMHIST ,
729it prints the UVM history logs on the system console as entries are added.
730Note that the output is
731.Em extremely
732voluminous, so this option is really only useful for debugging
733the very earliest parts of kernel initialization.
734.It Cd options UVMHIST_MAPHIST_SIZE
735Set the size of the
736.Dq maphist
737kernel history.
738The default is 100.
739This option depends upon the
740.Em UVMHIST
741option.
742.It Cd options UVMHIST_PDHIST_SIZE
743Set the size of the
744.Dq pdhist
745kernel history.
746The default is 100.
747This option depends upon the
748.Em UVMHIST
749option.
750.It Cd options BIOHIST
751Like
752.Em KERNHIST ,
753it enables the BIO history logs.
754These logs can be displayed by using
755.Cm show kernhist
756from DDB, and can help in debugging problems with Buffered I/O operations.
757See the kernel source file
758.Pa sys/kern/vfs_vio.c
759for details.
760.It Cd options BIOHIST_PRINT
761Like
762.Em BIOHIST ,
763it prints the BIO history logs on the system console as entries are added.
764Note that the output is
765.Em extremely
766voluminous, so this option is really only useful for debugging
767the very earliest parts of kernel initialization.
768.It Cd options BIOHIST_SIZE
769Set the size of the
770.Dq biohist
771kernel history.
772The default is 500.
773This option depends upon the
774.Em BIOHIST
775option.
776.El
777.Ss File Systems
778.Bl -ohang
779.It Cd file-system FFS
780Includes code implementing the Berkeley Fast File System
781.Po Em FFS Pc .
782Most machines need this if they are not running diskless.
783.It Cd file-system EXT2FS
784Includes code implementing the Second Extended File System
785.Po Em ext2 Pc ,
786revision 0 and revision 1 with the
787.Em filetype ,
788.Em sparse_super
789and
790.Em large_file
791options.
792This is the most commonly used file system on the Linux operating system,
793and is provided here for compatibility.
794Some of the specific features of
795.Em ext2
796like the "behavior on errors" are not implemented.
797See
798.Xr mount_ext2fs 8
799for details.
800.It Cd file-system LFS
801.Bq Em EXPERIMENTAL
802Include the Log-structured File System
803.Po Em LFS Pc .
804See
805.Xr mount_lfs 8
806and
807.Xr newfs_lfs 8
808for details.
809.It Cd file-system MFS
810Include the Memory File System
811.Po Em MFS Pc .
812This file system stores files in swappable memory, and produces
813notable performance improvements when it is used as the file store
814for
815.Pa /tmp
816and similar file systems.
817See
818.Xr mount_mfs 8
819for details.
820.It Cd file-system NFS
821Include the client side of the Network File System (NFS)
822remote file sharing protocol.
823Although the bulk of the code implementing NFS
824is kernel based, several user level daemons are needed for it to work.
825See
826.Xr mount_nfs 8
827for details.
828.It Cd file-system CD9660
829Includes code for the ISO 9660 + Rock Ridge file system, which is
830the standard file system on many CD-ROM discs.
831Useful primarily if you have a CD-ROM drive.
832See
833.Xr mount_cd9660 8
834for details.
835.It Cd file-system MSDOSFS
836Includes the MS-DOS FAT file system, which is reportedly still used
837by unfortunate people who have not heard about
838.Nx .
839Also implements the Windows 95 extensions to the same, which permit
840the use of longer, mixed case file names.
841See
842.Xr mount_msdos 8
843and
844.Xr fsck_msdos 8
845for details.
846.It Cd file-system NTFS
847.Bq Em EXPERIMENTAL
848Includes code for the Microsoft Windows NT file system.
849See
850.Xr mount_ntfs 8
851for details.
852.It Cd file-system FDESC
853Includes code for a file system, conventionally mounted on
854.Pa /dev/fd ,
855which permits access to the per-process file descriptor space via
856special files in the file system.
857See
858.Xr mount_fdesc 8
859for details.
860Note that this facility is redundant, and thus unneeded on most
861.Nx
862systems, since the
863.Xr fd 4
864pseudo-device driver already provides identical functionality.
865On most
866.Nx
867systems, instances of
868.Xr fd 4
869are mknoded under
870.Pa /dev/fd/
871and on
872.Pa /dev/stdin ,
873.Pa /dev/stdout ,
874and
875.Pa /dev/stderr .
876.It Cd file-system KERNFS
877Includes code which permits the mounting of a special file system
878(normally mounted on
879.Pa /kern )
880in which files representing various kernel variables and parameters
881may be found.
882See
883.Xr mount_kernfs 8
884for details.
885.It Cd file-system NULLFS
886Includes code for a loopback file system.
887This permits portions of the file hierarchy to be re-mounted in other places.
888The code really exists to provide an example of a stackable file system layer.
889See
890.Xr mount_null 8
891for details.
892.It Cd file-system OVERLAY
893Includes code for a file system filter.
894This permits the overlay file system to intercept all access to an underlying
895file system.
896This file system is intended to serve as an example of a stacking file
897system which has a need to interpose itself between an underlying file
898system and all other access.
899See
900.Xr mount_overlay 8
901for details.
902.It Cd file-system PROCFS
903Includes code for a special file system (conventionally mounted on
904.Pa /proc )
905in which the process space becomes visible in the file system.
906Among
907other things, the memory spaces of processes running on the system are
908visible as files, and signals may be sent to processes by writing to
909.Pa ctl
910files in the procfs namespace.
911See
912.Xr mount_procfs 8
913for details.
914.It Cd file-system UDF
915.Bq Em EXPERIMENTAL
916Includes code for the UDF file system commonly found on CD and DVD
917media but also on USB sticks.
918Currently supports read and write access upto UDF 2.01 and somewhat limited
919write support for UDF 2.50.
920It is marked experimental since there is no
921.Xr fsck_udf 8 .
922See
923.Xr mount_udf 8
924for details.
925.It Cd file-system UMAPFS
926Includes a loopback file system in which user and group IDs may be
927remapped -- this can be useful when mounting alien file systems with
928different UIDs and GIDs than the local system.
929See
930.Xr mount_umap 8
931for details.
932.It Cd file-system UNION
933.Bq Em EXPERIMENTAL
934Includes code for the union file system, which permits directories to
935be mounted on top of each other in such a way that both file systems
936remain visible -- this permits tricks like allowing writing (and the
937deleting of files) on a read-only file system like a CD-ROM by
938mounting a local writable file system on top of the read-only file
939system.
940See
941.Xr mount_union 8
942for details.
943.It Cd file-system CODA
944.Bq Em EXPERIMENTAL
945Includes code for the Coda file system.
946Coda is a distributed file system like NFS and AFS.
947It is freely available, like NFS, but it functions much like AFS in being a
948.Dq stateful
949file system.
950Both Coda and AFS cache files on your local machine to improve performance.
951Then Coda goes a step further than AFS by letting you access the cached
952files when there is no available network, viz. disconnected laptops and
953network outages.
954In Coda, both the client and server are outside the kernel which makes
955them easier to experiment with.
956Coda is available for several UNIX and non-UNIX platforms.
957See
958.Lk http://www.coda.cs.cmu.edu
959for more details.
960.Em NOTE :
961You also need to enable the pseudo-device, vcoda, for the Coda
962file system to work.
963.It Cd file-system SMBFS
964.Bq Em EXPERIMENTAL
965Includes code for the SMB/CIFS file system.
966See
967.Xr mount_smbfs 8
968for details.
969.Em NOTE :
970You also need to enable the pseudo-device, nsmb, for the SMB
971file system to work.
972.It Cd file-system PTYFS
973.Bq Em EXPERIMENTAL
974Includes code for a special file system (normally mounted on
975.Pa /dev/pts )
976in which pseudo-terminal slave devices become visible in the file system.
977See
978.Xr mount_ptyfs 8
979for details.
980.It Cd file-system TMPFS
981Includes code for the efficient memory file system, normally used over
982.Pa /tmp .
983See
984.Xr mount_tmpfs 8
985for details.
986.It Cd file-system PUFFS
987Includes kernel support for the pass-to-userspace framework file system.
988It can be used to implement file system functionality in userspace.
989See
990.Xr puffs 3
991for more details.
992This enables for example sshfs:
993.Xr mount_psshfs 8 .
994.El
995.Ss File System Options
996.Bl -ohang
997.It Cd options DISKLABEL_EI
998Enable
999.Dq Endian-Independent
1000.Xr disklabel 5
1001support.
1002This allows a system to recognize a disklabel written in the other byte order.
1003For writing, when a label already exists, its byte order is preserved.
1004Otherwise, a new label is written in the native byte order.
1005To specify the byte order explicitly, the
1006.Fl F
1007option of
1008.Xr disklabel 8
1009should be used with the
1010.Fl B
1011option in order to avoid using
1012.Xr ioctl 2 ,
1013which results in the default behavior explained above.
1014At the moment this option is restricted to the following ports:
1015amd64, bebox, emips, epoc32, evbarm, i386, ibmnws, landisk, mvmeppc, prep,
1016.\" riscv,
1017rs6000, sandpoint,
1018.\" usermode,
1019xen, and zaurus;
1020also to machines of the
1021.\" evbarm64,
1022evbmips and evbppc ports that support
1023Master Boot Record (MBR).
1024.It Cd options MAGICLINKS
1025Enables the expansion of special strings
1026.Po
1027beginning with
1028.Dq @
1029.Pc
1030when traversing symbolic links.
1031See
1032.Xr symlink 7
1033for a list of supported strings.
1034Note that this option only controls the enabling of this feature
1035by the kernel at boot-up.
1036This feature can still be manipulated with the
1037.Xr sysctl 8
1038command regardless of the setting of this option.
1039.It Cd options NFSSERVER
1040Include the server side of the
1041.Em NFS
1042(Network File System) remote file sharing protocol.
1043Although the bulk of the code implementing
1044.Em NFS
1045is kernel based, several user level daemons are needed for it to
1046work.
1047See
1048.Xr mountd 8
1049and
1050.Xr nfsd 8
1051for details.
1052.It Cd options NVNODE=integer
1053This option sets the size of the cache used by the name-to-inode translation
1054routines, (a.k.a. the
1055.Fn namei
1056cache, though called by many other names in the kernel source).
1057By default, this cache has
1058.Dv (NPROC + NTEXT + 100)
1059entries (NPROC set as 20 + 16 * MAXUSERS and NTEXT as 80 + NPROC / 8).
1060A reasonable way to derive a value of
1061.Dv NVNODE ,
1062should you notice a large number of namei cache misses with a tool such as
1063.Xr systat 1 ,
1064is to examine your system's current computed value with
1065.Xr sysctl 8 ,
1066(which calls this parameter "kern.maxvnodes") and to increase this value
1067until either the namei cache hit rate improves or it is determined that
1068your system does not benefit substantially from an increase in the size of
1069the namei cache.
1070.It Cd options NAMECACHE_ENTER_REVERSE
1071Causes the namei cache to always enter a reverse mapping (vnode -> name)
1072as well as a normal one.
1073Normally, this is already done for directory vnodes, to speed up the getcwd
1074operation.
1075This option will cause longer hash chains in the reverse cache, and thus
1076slow down getcwd somewhat.
1077However, it does make vnode -> path translations possible in some cases.
1078For now, only useful if strict
1079.Pa /proc/#/maps
1080emulation for Linux binaries is required.
1081.El
1082.Ss Options for FFS/UFS File Systems
1083.Bl -ohang
1084.It Cd options WAPBL
1085Enable
1086.Dq Write Ahead Physical Block Logging file system journaling .
1087This provides rapid file system consistency checking after a system outage.
1088It also provides better general use performance over regular FFS.
1089See also
1090.Xr wapbl 4 .
1091.It Cd options QUOTA
1092Enables kernel support for traditional quotas in FFS.
1093Traditional quotas store the quota information in external files and
1094require
1095.Xr quotacheck 8
1096and
1097.Xr quotaon 8
1098at boot time.
1099Traditional quotas are limited to 32-bit sizes and are at this point
1100considered a legacy feature.
1101.It Cd options QUOTA2
1102Enables kernel support for in-volume quotas in FFS.
1103The quota information is file system metadata maintained by
1104.Xr fsck 8
1105and/or WAPBL journaling.
1106MFS volumes can also use
1107.Dv QUOTA2
1108quotas; see
1109.Xr mount_mfs 8
1110for more information.
1111.It Cd options FFS_EI
1112Enable
1113.Dq Endian-Independent
1114FFS support.
1115This allows a system to mount an FFS file system created for another
1116architecture, at a small performance cost for all FFS file systems.
1117See also
1118.Xr newfs 8 ,
1119.Xr fsck_ffs 8 ,
1120.Xr dumpfs 8
1121for file system byte order status and manipulation.
1122.It Cd options FFS_NO_SNAPSHOT
1123Disable support for the creation of file system internal snapshot
1124of FFS file systems.
1125Maybe useful for install media kernels, small memory systems and
1126embedded systems which don't require the snapshot support.
1127.It Cd options UFS_EXTATTR
1128Enable extended attribute support for UFS1 file systems.
1129.It Cd options UFS_DIRHASH
1130Increase lookup performance by maintaining in-core hash tables
1131for large directories.
1132.El
1133.Ss Options for the LFS File System
1134.Bl -ohang
1135.\" .It Cd options LFS_KERNEL_RFW
1136.\" There is no documentation for this. (XXX)
1137.\" .It Cd options LFS_QUOTA
1138.\" Enable traditional quota support for LFS.
1139.\" (It is not clear if this works.)
1140.\" .It Cd options LFS_QUOTA2
1141.\" Enable modernized 64-bit quota support for LFS.
1142.\" (This does not work yet.)
1143.It Cd options LFS_EI
1144Enable
1145.Dq Endian-Independent
1146LFS support.
1147This allows (at a small performance cost) mounting an LFS file system
1148created for another architecture.
1149.\" .It Cd options LFS_EXTATTR
1150.\" Enable extended attribute support for LFS.
1151.\" (It is not clear if this works.)
1152.It Cd options LFS_DIRHASH
1153Increase lookup performance by maintaining in-core hash tables
1154for large directories.
1155.El
1156.Ss Options for the ext2fs File System
1157.Bl -ohang
1158.It Cd options EXT2FS_SYSTEM_FLAGS
1159This option changes the behavior of the APPEND and IMMUTABLE flags
1160for a file on an
1161.Em ext2
1162file system.
1163Without this option, the superuser or owner of the file can
1164set and clear them.
1165With this option, only the superuser can set them, and
1166they can't be cleared if the securelevel is greater than 0.
1167See also
1168.Xr chflags 1
1169and
1170.Xr secmodel_securelevel 9 .
1171.El
1172.Ss Options for the NFS File System
1173.Bl -ohang
1174.It Cd options NFS_BOOT_BOOTP
1175Enable use of the BOOTP protocol (RFCs 951 and 1048) to get configuration
1176information if NFS is used to mount the root file system.
1177See
1178.Xr diskless 8
1179for details.
1180.It Cd options NFS_BOOT_BOOTSTATIC
1181Enable use of static values defined as
1182.Dq NFS_BOOTSTATIC_MYIP ,
1183.Dq NFS_BOOTSTATIC_GWIP ,
1184.Dq NFS_BOOTSTATIC_SERVADDR ,
1185and
1186.Dq NFS_BOOTSTATIC_SERVER
1187in kernel options to get configuration information
1188if NFS is used to mount the root file system.
1189.It Cd options NFS_BOOT_DHCP
1190Same as
1191.Dq NFS_BOOT_BOOTP ,
1192but use the DHCP extensions to the
1193BOOTP protocol (RFC 1541).
1194.It Cd options NFS_BOOT_BOOTP_REQFILE
1195Specifies the string sent in the bp_file field of the BOOTP/DHCP
1196request packet.
1197.It Cd options NFS_BOOT_BOOTPARAM
1198Enable use of the BOOTPARAM protocol, consisting of RARP and
1199BOOTPARAM RPC, to get configuration information if NFS
1200is used to mount the root file system.
1201See
1202.Xr diskless 8
1203for details.
1204.It Cd options NFS_BOOT_RWSIZE=value
1205Set the initial NFS read and write sizes for diskless-boot requests.
1206The normal default is 8Kbytes.
1207This option provides a way to lower the value (e.g., to 1024 bytes)
1208as a workaround for buggy network interface cards or boot PROMs.
1209Once booted, the read and write request sizes can be increased by
1210remounting the file system.
1211See
1212.Xr mount_nfs 8
1213for details.
1214.It Cd options NFS_V2_ONLY
1215Reduce the size of the NFS client code by omitting code that's only required
1216for NFSv3 and NQNFS support, leaving only that code required to use NFSv2
1217servers.
1218.It Cd options NFS_BOOT_TCP
1219Use NFS over TCP instead of the default UDP, for mounting root.
1220.El
1221.Ss Buffer queue strategy options
1222The following options enable alternative buffer queue strategies.
1223.Bl -ohang
1224.It Cd options BUFQ_READPRIO
1225Enable experimental buffer queue strategy for disk I/O.
1226In the default strategy, outstanding disk requests are ordered by
1227sector number and sent to the disk, regardless of whether the
1228operation is a read or write; this option gives priority to issuing
1229read requests over write requests.
1230Although requests may therefore be issued out of sector-order, causing
1231more seeks and thus lower overall throughput, interactive system
1232responsiveness under heavy disk I/O load may be improved, as processes
1233blocking on disk reads are serviced sooner (file writes typically
1234don't cause applications to block).
1235The performance effect varies greatly depending on the hardware, drive
1236firmware, file system configuration, workload, and desired performance
1237trade-off.
1238Systems using drive write-cache (most modern IDE disks, by default)
1239are unlikely to benefit and may well suffer; such disks acknowledge
1240writes very quickly, and optimize them internally according to
1241physical layout.
1242Giving these disks as many requests to work with as possible (the
1243standard strategy) will typically produce the best results, especially
1244if the drive has a large cache; the drive will silently complete
1245writes from cache as it seeks for reads.
1246Disks that support a large number of concurrent tagged requests (SCSI
1247disks and many hardware RAID controllers) expose this internal
1248scheduling with tagged responses, and don't block for reads; such
1249disks may not see a noticeable difference with either strategy.
1250However, if IDE disks are run with write-cache disabled for safety,
1251writes are not acknowledged until actually completed, and only one
1252request can be outstanding; a large number of small writes in one
1253locality can keep the disk busy, starving reads elsewhere on the disk.
1254Such systems are likely to see the most benefit from this option.
1255Finally, the performance interaction of this option with ffs soft
1256dependencies can be subtle, as that mechanism can drastically alter
1257the workload for file system metadata writes.
1258.It Cd options BUFQ_PRIOCSCAN
1259Enable another buffer queue strategy for disk I/O, per-priority cyclical scan.
1260.It Cd options NEW_BUFQ_STRATEGY
1261Synonym of
1262.Em BUFQ_READPRIO .
1263.El
1264.Ss Miscellaneous Options
1265.Bl -ohang
1266.It Cd options CPU_UCODE
1267Support cpu microcode loading via
1268.Xr cpuctl 8 .
1269.It Cd options MEMORY_DISK_DYNAMIC
1270This option makes the
1271.Xr md 4
1272RAM disk size dynamically sized.
1273It is incompatible with
1274.Xr mdsetimage 8 .
1275.It Cd options MEMORY_DISK_HOOKS
1276This option allows for some machine dependent functions to be called when
1277the
1278.Xr md 4
1279RAM disk driver is configured.
1280This can result in automatically loading a RAM disk from floppy on
1281open (among other things).
1282.It Cd options MEMORY_DISK_IS_ROOT
1283Forces the
1284.Xr md 4
1285RAM disk to be the root device.
1286This can only be overridden when
1287the kernel is booted in the 'ask-for-root' mode.
1288.It Cd options MEMORY_DISK_ROOT_SIZE=integer
1289Allocates the given number of 512 byte blocks as memory for the
1290.Xr md 4
1291RAM disk, to be populated with
1292.Xr mdsetimage 8 .
1293.It Cd options MEMORY_DISK_SERVER=0
1294Do not include the interface to a userland memory disk server process.
1295Per default, this option is set to 1, including the support code.
1296Useful for install media kernels.
1297.It Cd options MEMORY_DISK_RBFLAGS=value
1298This option sets the
1299.Xr reboot 2
1300flags used when booting with a memory disk as root file system.
1301Possible values include
1302.Dv RB_AUTOBOOT
1303(boot in the usual fashion - default value), and
1304.Dv RB_SINGLE
1305(boot in single-user mode).
1306.It Cd options MODULAR
1307Enables the framework for kernel modules (see
1308.Xr module 7 ) .
1309.It Cd options MODULAR_DEFAULT_AUTOLOAD
1310Enables the autoloading of kernel modules by default.
1311This sets the default value of the
1312.Em kern.module.autoload
1313.Xr sysctl 3
1314variable which may be changed at run time.
1315.It Cd options VND_COMPRESSION
1316Enables the
1317.Xr vnd 4
1318driver to also handle compressed images.
1319See
1320.Xr vndcompress 1 ,
1321.Xr vnd 4
1322and
1323.Xr vnconfig 8
1324for more information.
1325.It Cd options SPLDEBUG
1326Help the kernel programmer find bugs related to the interrupt priority
1327level.
1328When
1329.Fn spllower
1330or
1331.Fn splraise
1332changes the current CPU's interrupt priority level to or from
1333.Dv IPL_HIGH ,
1334record a backtrace.
1335Read
1336.Xr i386/return_address 9
1337for caveats about collecting backtraces.
1338This feature is experimental, and it is only available on i386.
1339See
1340.Pa sys/kern/subr_spldebug.c .
1341.It Cd options TFTPROOT
1342Download the root memory disk through TFTP at root mount time.
1343This enables the use of a root RAM disk without requiring it to be
1344embedded in the kernel using
1345.Xr mdsetimage 8 .
1346The RAM disk name is obtained using DHCP's filename parameter.
1347This option requires
1348.Em MEMORY_DISK_HOOKS ,
1349.Em MEMORY_DISK_DYNAMIC ,
1350and
1351.Em MEMORY_DISK_IS_ROOT .
1352It is incompatible with
1353.Em MEMORY_DISK_ROOT_SIZE .
1354.It Cd options HZ=integer
1355On ports that support it, set the system clock frequency (see
1356.Xr hz 9 )
1357to the supplied value.
1358Handle with care.
1359.It Cd options NTP
1360Turns on in-kernel precision timekeeping support used by software
1361implementing
1362.Em NTP
1363(Network Time Protocol, RFC 1305).
1364The
1365.Em NTP
1366option adds an in-kernel Phase-Locked Loop (PLL) for normal
1367.Em NTP
1368operation, and a Frequency-Locked Loop (FLL) for intermittently-connected
1369operation.
1370.Xr ntpd 8
1371will employ a user-level PLL when kernel support is unavailable,
1372but the in-kernel version has lower latency and more precision, and
1373so typically keeps much better time.
1374.Pp
1375The interface to the kernel
1376.Em NTP
1377support is provided by the
1378.Xr ntp_adjtime 2
1379and
1380.Xr ntp_gettime 2
1381system calls, which are intended for use by
1382.Xr ntpd 8
1383and are enabled by the option.
1384On systems with sub-microsecond resolution timers, or where (HZ/100000)
1385is not an integer, the
1386.Em NTP
1387option also enables extended-precision arithmetic to keep track of
1388fractional clock ticks at NTP time-format precision.
1389.It Cd options PPS_SYNC
1390This option enables a kernel serial line discipline for receiving time
1391phase signals from an external reference clock such as a radio clock.
1392.Po
1393The
1394.Em NTP
1395option (which see) must be on if the
1396.Em PPS_SYNC
1397option is used
1398.Pc .
1399Some reference clocks generate a Pulse Per Second (PPS) signal in
1400phase with their time source.
1401The
1402.Em PPS
1403line discipline receives this signal on either the data leads
1404or the DCD control lead of a serial port.
1405.Pp
1406.Em NTP
1407uses the PPS signal to discipline the local clock oscillator to a high
1408degree of precision (typically less than 50 microseconds in time and
14090.1 ppm in accuracy).
1410.Em PPS
1411can also generate a serial output pulse when the system receives a PPS
1412interrupt.
1413This can be used to measure the system interrupt latency and thus calibrate
1414.Em NTP
1415to account for it.
1416Using
1417.Em PPS
1418usually requires a gadget box
1419to convert from TTL to RS-232 signal levels.
1420The gadget box and PPS are described in more detail in the HTML documentation
1421for
1422.Xr ntpd 8
1423in
1424.Pa /usr/share/doc/html/ntp .
1425.Pp
1426.Nx
1427currently supports this option in
1428.Xr com 4
1429and
1430.Xr zsc 4 .
1431.It Cd options SETUIDSCRIPTS
1432Allows scripts with the setuid bit set to execute as the effective
1433user rather than the real user, just like binary executables.
1434.Pp
1435.Em NOTE :
1436Using this option will also enable
1437.Em options FDSCRIPTS
1438.It Cd options FDSCRIPTS
1439Allows execution of scripts with the execute bit set, but not the
1440read bit, by opening the file and passing the file descriptor to
1441the shell, rather than the filename.
1442.Pp
1443.Em NOTE :
1444Execute only (non-readable) scripts will have
1445.Va argv[0]
1446set to
1447.Pa /dev/fd/* .
1448What this option allows as far as security is
1449concerned, is the ability to safely ensure that the correct script
1450is run by the interpreter, as it is passed as an already open file.
1451.It Cd options RTC_OFFSET=integer
1452The kernel (and typically the hardware battery backed-up clock on
1453those machines that have one) keeps time in
1454.Em UTC
1455(Universal Coordinated Time, once known as
1456.Em GMT ,
1457or Greenwich Mean Time)
1458and not in the time of the local time zone.
1459The
1460.Em RTC_OFFSET
1461option is used on some ports (such as the i386) to tell the kernel
1462that the hardware clock is offset from
1463.Em UTC
1464by the specified number of minutes.
1465This is typically used when a machine boots several operating
1466systems and one of them wants the hardware clock to run in the
1467local time zone and not in
1468.Em UTC ,
1469e.g.
1470.Em RTC_OFFSET=300
1471means
1472the hardware clock is set to US Eastern Time (300 minutes behind
1473.Em UTC ) ,
1474and not
1475.Em UTC .
1476(Note:
1477.Em RTC_OFFSET
1478is used to initialize a kernel variable named
1479.Va rtc_offset
1480which is the source actually used to determine the clock offset, and
1481which may be accessed via the kern.rtc_offset sysctl variable.
1482See
1483.Xr sysctl 8
1484and
1485.Xr sysctl 3
1486for details.
1487Since the kernel clock is initialized from the hardware clock very
1488early in the boot process, it is not possible to meaningfully change
1489.Va rtc_offset
1490in system initialization scripts.
1491Changing this value currently may only be done at kernel compile
1492time or by patching the kernel and rebooting).
1493.Pp
1494.Em NOTE :
1495Unfortunately, in many cases where the hardware clock
1496is kept in local time, it is adjusted for Daylight Savings
1497Time; this means that attempting to use
1498.Em RTC_OFFSET
1499to let
1500.Nx
1501coexist with such an operating system, like Windows,
1502would necessitate changing
1503.Em RTC_OFFSET
1504twice a year.
1505As such, this solution is imperfect.
1506.It Cd options MAXUPRC=integer
1507Sets the soft
1508.Dv RLIMIT_NPROC
1509resource limit, which specifies the maximum number of simultaneous
1510processes a user is permitted to run, for process 0;
1511this value is inherited by its child processes.
1512It defaults to
1513.Em CHILD_MAX ,
1514which is currently defined to be 160.
1515Setting
1516.Em MAXUPRC
1517to a value less than
1518.Em CHILD_MAX
1519is not permitted, as this would result in a violation of the semantics of
1520.St -p1003.1-90 .
1521.It Cd options NOFILE=integer
1522Sets the soft
1523.Dv RLIMIT_NOFILE
1524resource limit, which specifies the maximum number of open
1525file descriptors for each process;
1526this value is inherited by its child processes.
1527It defaults to
1528.Em OPEN_MAX ,
1529which is currently defined to be 128.
1530.It Cd options MAXFILES=integer
1531Sets the default value of the
1532.Em kern.maxfiles
1533sysctl variable, which indicates the maximum number of files that may
1534be open in the system.
1535.It Cd options DEFCORENAME=string
1536Sets the default value of the
1537.Em kern.defcorename
1538sysctl variable, otherwise it is set to
1539.Nm %n.core .
1540See
1541.Xr sysctl 8
1542and
1543.Xr sysctl 3
1544for details.
1545.It Cd options RASOPS_CLIPPING
1546Enables clipping within the
1547.Nm rasops
1548raster-console output system.
1549.Em NOTE :
1550only available on architectures that use
1551.Nm rasops
1552for console output.
1553.It Cd options RASOPS_SMALL
1554Removes optimized character writing code from the
1555.Nm rasops
1556raster-console output system.
1557.Em NOTE :
1558only available on architectures that use
1559.Nm rasops
1560for console output.
1561.It Cd options INCLUDE_CONFIG_FILE
1562Embeds the kernel config file used to define the kernel in the kernel
1563binary itself.
1564The embedded data also includes any files directly included by the config
1565file itself, e.g.
1566.Pa GENERIC.local
1567or
1568.Pa std.$MACHINE .
1569The embedded config file can be extracted from the resulting kernel with
1570.Xr config 1
1571.Fl x ,
1572or by the following command:
1573.Bd -literal -offset indent
1574strings netbsd | sed -n 's/^_CFG_//p' | unvis
1575.Ed
1576.It Cd options INCLUDE_JUST_CONFIG
1577Similar to the above option, but includes just the actual config file,
1578not any included files.
1579.It Cd options PIPE_SOCKETPAIR
1580Use slower, but smaller socketpair(2)-based pipe implementation instead
1581of default faster, but bigger one.
1582Primarily useful for installation kernels.
1583.It Cd options USERCONF
1584Compiles in the in-kernel device configuration manager.
1585See
1586.Xr userconf 4
1587for details.
1588.It Cd options PERFCTRS
1589Compiles in kernel support for CPU performance-monitoring counters.
1590See
1591.Xr pmc 1
1592for details.
1593.Em NOTE :
1594not available on all architectures.
1595.It Cd options SCDEBUG_DEFAULT
1596Used with the
1597.Cd options SYSCALL_DEBUG
1598described below to choose which types of events are displayed.
1599.Pp
1600.Bl -tag -width "SCDEBUG_KERNHIST" -compact -offset indent
1601.It Dv SCDEBUG_CALLS
1602Show system call entry points.
1603.It Dv SCDEBUG_RETURNS
1604Show system call exit points.
1605.It Dv SCDEBUG_ALL
1606Show all system call requestes, including unimplemented calls.
1607.It Dv SCDEBUG_SHOWARGS
1608Show the arguments provided.
1609.It Dv SCDEBUG_KERNHIST
1610Store a restricted form of the system call debug in a kernel history
1611instead of printing it to the console.
1612This option relies upon
1613.Cd options KERNHIST .
1614.El
1615.Pp
1616The default value is
1617.Dv (SCDEBUG_CALLS|SCDEBUG_RETURNS|SCDEBUG_SHOWARGS) .
1618.It Cd options SYSCALL_DEBUG
1619Useful for debugging system call issues, usually in early single user bringup.
1620By default, writes entries to the system console for most system call events.
1621Can be configured with the
1622.Cd options SCDEBUG_DEFAULT
1623option to to use the
1624.Cd options KERNHIST
1625facility instead.
1626.It Cd options SYSCALL_STATS
1627Count the number of times each system call number is called.
1628The values can be read through the sysctl interface and displayed using
1629.Xr systat 1 .
1630.Em NOTE :
1631not yet available on all architectures.
1632.It Cd options SYSCALL_TIMES
1633Count the time spent (using
1634.Fn cpu_counter32 )
1635in each system call.
1636.Em NOTE :
1637Using this option will also enable
1638.Cd options SYSCALL_STATS .
1639.It Cd options SYSCALL_TIMES_HASCOUNTER
1640Force use of
1641.Fn cpu_counter32
1642even if
1643.Fn cpu_hascounter
1644reports false.
1645Useful for systems where the cycle counter doesn't run at a constant rate
1646(e.g. Soekris boxes).
1647.It Cd options XSERVER_DDB
1648A supplement to XSERVER that adds support for entering
1649.Xr ddb 4
1650while in X11.
1651.It Cd options FILEASSOC
1652Support for
1653.Xr fileassoc 9 .
1654Required for
1655.Cd options PAX_SEGVGUARD
1656and
1657.Cd pseudo-device veriexec .
1658.It Cd options FILEASSOC_NHOOKS=integer
1659Number of storage slots per file for
1660.Xr fileassoc 9 .
1661Default is 4.
1662.El
1663.Ss Networking Options
1664.Bl -ohang
1665.It Cd options GATEWAY
1666Enables
1667.Em IPFORWARDING
1668(which see)
1669and (on most ports) increases the size of
1670.Em NMBCLUSTERS
1671(which see).
1672In general,
1673.Em GATEWAY
1674is used to indicate that a system should act as a router, and
1675.Em IPFORWARDING
1676is not invoked directly.
1677(Note that
1678.Em GATEWAY
1679has no impact on protocols other than IP, such as CLNP).
1680.Em GATEWAY
1681option also compiles IPv4 and IPv6 fast forwarding code into the kernel.
1682.It Cd options ICMPPRINTFS
1683The
1684.Em ICMPPRINTFS
1685option will enable debugging information to be printed about
1686the
1687.Xr icmp 4
1688protocol.
1689.It Cd options IPFORWARDING=value
1690If
1691.Em value
1692is 1 this enables IP routing behavior.
1693If
1694.Em value
1695is 0 (the default), it disables it.
1696The
1697.Em GATEWAY
1698option sets this to 1 automatically.
1699With this option enabled, the machine will forward IP datagrams destined
1700for other machines between its interfaces.
1701Note that even without this option, the kernel will
1702still forward some packets (such as source routed packets) -- removing
1703.Em GATEWAY
1704and
1705.Em IPFORWARDING
1706is insufficient to stop all routing through a bastion host on a
1707firewall -- source routing is controlled independently.
1708To turn off source routing, use
1709.Em options IPFORWSRCRT=0
1710(which see).
1711Note that IP forwarding may be turned on and off independently of the
1712setting of the
1713.Em IPFORWARDING
1714option through the use of the
1715.Em net.inet.ip.forwarding
1716sysctl variable.
1717If
1718.Em net.inet.ip.forwarding
1719is 1, IP forwarding is on.
1720See
1721.Xr sysctl 8
1722and
1723.Xr sysctl 3
1724for details.
1725.It Cd options IPFORWSRCRT=value
1726If
1727.Em value
1728is set to zero, source routing of IP datagrams is turned off.
1729If
1730.Em value
1731is set to one (the default) or the option is absent, source routed IP
1732datagrams are forwarded by the machine.
1733Note that source routing of IP packets may be turned on and off
1734independently of the setting of the
1735.Em IPFORWSRCRT
1736option through the use of the
1737.Em net.inet.ip.forwsrcrt
1738sysctl variable.
1739If
1740.Em net.inet.ip.forwsrcrt
1741is 1, forwarding of source routed IP datagrams is on.
1742See
1743.Xr sysctl 8
1744and
1745.Xr sysctl 3
1746for details.
1747.It Cd options IFA_STATS
1748Tells the kernel to maintain per-address statistics on bytes sent
1749and received over (currently) Internet and AppleTalk addresses.
1750.\"This can be a fairly expensive operation, so you probably want to
1751.\"keep this disabled.
1752The option is not recommended as it degrades system stability.
1753.It Cd options IFQ_MAXLEN=value
1754Increases the allowed size of the network interface packet queues.
1755The default queue size is 50 packets, and you do not normally need
1756to increase it.
1757.It Cd options IPSELSRC
1758Includes support for source-address selection policies.
1759See
1760.Xr in_getifa 9 .
1761.It Cd options MROUTING
1762Includes support for IP multicast routers.
1763You certainly want
1764.Em INET
1765with this.
1766Multicast routing is controlled by the
1767.Xr mrouted 8
1768daemon.
1769See also option
1770.Cd PIM .
1771.It Cd options PIM
1772Includes support for Protocol Independent Multicast (PIM) routing.
1773You need
1774.Em MROUTING
1775and
1776.Em INET
1777with this.
1778Software using this can be found e.g. in
1779.Pa pkgsrc/net/xorp .
1780.It Cd options INET
1781Includes support for the TCP/IP protocol stack.
1782You almost certainly want this.
1783See
1784.Xr inet 4
1785for details.
1786.It Cd options INET6
1787Includes support for the IPv6 protocol stack.
1788See
1789.Xr inet6 4
1790for details.
1791Unlike
1792.Em INET ,
1793.Em INET6
1794enables multicast routing code as well.
1795This option requires
1796.Em INET
1797at this moment, but it should not.
1798.It Cd options ND6_DEBUG
1799The option sets the default value of net.inet6.icmp6.nd6_debug to 1,
1800for debugging IPv6 neighbor discovery protocol handling.
1801See
1802.Xr sysctl 3
1803for details.
1804.It Cd options IPSEC
1805Includes support for the IPsec protocol, using the implementation derived from
1806.Ox ,
1807relying on
1808.Xr opencrypto 9
1809to carry out cryptographic operations.
1810See
1811.Xr fast_ipsec 4
1812for details.
1813.It Cd options IPSEC_DEBUG
1814Enables debugging code in IPsec stack.
1815See
1816.Xr ipsec 4
1817for details.
1818The
1819.Cd IPSEC
1820option includes support for IPsec Network Address Translator traversal
1821(NAT-T), as described in RFCs 3947 and 3948.
1822This feature might be patent-encumbered in some countries.
1823.It Cd options ALTQ
1824Enabled ALTQ (Alternate Queueing).
1825For simple rate-limiting, use
1826.Xr tbrconfig 8
1827to set up the interface transmission rate.
1828To use queueing disciplines, their appropriate kernel options should also
1829be defined (documented below).
1830Queueing disciplines are managed by
1831.Xr altqd 8 .
1832See
1833.Xr altq 9
1834for details.
1835.It Cd options ALTQ_HFSC
1836Include support for ALTQ-implemented HFSC (Hierarchical Fair Service Curve)
1837module.
1838HFSC supports both link-sharing and guaranteed real-time services.
1839HFSC employs a service curve based QoS model, and its unique feature
1840is an ability to decouple delay and bandwidth allocation.
1841Requires
1842.Em ALTQ_RED
1843to use the RED queueing discipline on HFSC classes, or
1844.Em ALTQ_RIO
1845to use the RIO queueing discipline on HFSC classes.
1846This option assumes
1847.Em ALTQ .
1848.It Cd options ALTQ_PRIQ
1849Include support for ALTQ-implemented PRIQ (Priority Queueing).
1850PRIQ implements a simple priority-based queueing discipline.
1851A higher priority class is always served first.
1852Requires
1853.Em ALTQ_RED
1854to use the RED queueing discipline on HFSC classes, or
1855.Em ALTQ_RIO
1856to use the RIO queueing discipline on HFSC classes.
1857This option assumes
1858.Em ALTQ .
1859.It Cd options ALTQ_WFQ
1860Include support for ALTQ-implemented WFQ (Weighted Fair Queueing).
1861WFQ implements a weighted-round robin scheduler for a set of queues.
1862A weight can be assigned to each queue to give a different proportion
1863of the link capacity.
1864A hash function is used to map a flow to one of a set of queues.
1865This option assumes
1866.Em ALTQ .
1867.It Cd options ALTQ_FIFOQ
1868Include support for ALTQ-implemented FIFO queueing.
1869FIFOQ is a simple drop-tail FIFO (First In, First Out) queueing discipline.
1870This option assumes
1871.Em ALTQ .
1872.It Cd options ALTQ_RIO
1873Include support for ALTQ-implemented RIO (RED with In/Out).
1874The original RIO has 2 sets of RED parameters; one for in-profile
1875packets and the other for out-of-profile packets.
1876At the ingress of the network, profile meters tag packets as IN or
1877OUT based on contracted profiles for customers.
1878Inside the network, IN packets receive preferential treatment by
1879the RIO dropper.
1880ALTQ/RIO has 3 drop precedence levels defined for the Assured Forwarding
1881PHB of DiffServ (RFC 2597).
1882This option assumes
1883.Em ALTQ .
1884.It Cd options ALTQ_BLUE
1885Include support for ALTQ-implemented Blue buffer management.
1886Blue is another active buffer management mechanism.
1887This option assumes
1888.Em ALTQ .
1889.It Cd options ALTQ_FLOWVALVE
1890Include support for ALTQ-implemented Flowvalve.
1891Flowvalve is a simple implementation of a RED penalty box that identifies
1892and punishes misbehaving flows.
1893This option requires
1894.Em ALTQ_RED
1895and assumes
1896.Em ALTQ .
1897.It Cd options ALTQ_CDNR
1898Include support for ALTQ-implemented CDNR (diffserv traffic conditioner)
1899packet marking/manipulation.
1900Traffic conditioners are components to meter, mark, or drop incoming
1901packets according to some rules.
1902As opposed to queueing disciplines, traffic conditioners handle incoming
1903packets at an input interface.
1904This option assumes
1905.Em ALTQ .
1906.It Cd options ALTQ_NOPCC
1907Disables use of processor cycle counter to measure time in ALTQ.
1908This option should be defined for a non-Pentium i386 CPU which does not
1909have TSC, SMP (per-CPU counters are not in sync), or power management
1910which affects processor cycle counter.
1911This option assumes
1912.Em ALTQ .
1913.It Cd options ALTQ_IPSEC
1914Include support for IPsec in IPv4 ALTQ.
1915This option assumes
1916.Em ALTQ .
1917.It Cd options ALTQ_JOBS
1918Include support for ALTQ-implemented JoBS (Joint Buffer Management
1919and Scheduling).
1920This option assumes
1921.Em ALTQ .
1922.It Cd options ALTQ_AFMAP
1923Include support for an undocumented ALTQ feature that is used to map an IP
1924flow to an ATM VC (Virtual Circuit).
1925This option assumes
1926.Em ALTQ .
1927.It Cd options ALTQ_LOCALQ
1928Include support for ALTQ-implemented local queues.
1929Its practical use is undefined.
1930Assumes
1931.Em ALTQ .
1932.It Cd options SUBNETSARELOCAL
1933Sets default value for net.inet.ip.subnetsarelocal variable, which
1934controls whether non-directly-connected subnets of connected networks
1935are considered "local" for purposes of choosing the MSS for a TCP
1936connection.
1937This is mostly present for historic reasons and completely irrelevant if
1938you enable Path MTU discovery.
1939.It Cd options HOSTZEROBROADCAST
1940Sets default value for net.inet.ip.hostzerobroadcast variable, which
1941controls whether the zeroth host address of each connected subnet is
1942also considered a broadcast address.
1943Default value is "1", for compatibility with old systems; if this is
1944set to zero on all hosts on a subnet, you should be able to fit an extra
1945host per subnet on the
1946".0" address.
1947.It Cd options MCLSHIFT=value
1948This option is the base-2 logarithm of the size of mbuf clusters.
1949The
1950.Bx
1951networking stack keeps network packets in a linked
1952list, or chain, of kernel buffer objects called mbufs.
1953The system provides larger mbuf clusters as an optimization for
1954large packets, instead of using long chains for large packets.
1955The mbuf cluster size,
1956or
1957.Em MCLBYTES ,
1958must be a power of two, and is computed as two raised to the power
1959.Em MCLSHIFT .
1960On systems with Ethernet network adapters,
1961.Em MCLSHIFT
1962is often set to 11, giving 2048-byte mbuf clusters, large enough to
1963hold a 1500-byte Ethernet frame in a single cluster.
1964Systems with network interfaces supporting larger frame sizes like
1965ATM, FDDI, or HIPPI may perform better with
1966.Em MCLSHIFT
1967set to 12 or 13, giving mbuf cluster sizes of 4096 and 8192 bytes,
1968respectively.
1969.It Cd options NETATALK
1970Include support for the AppleTalk protocol stack.
1971The kernel provides provision for the
1972.Em Datagram Delivery Protocol
1973(DDP), providing SOCK_DGRAM support and AppleTalk routing.
1974This stack is used by the
1975.Em NETATALK
1976package, which adds support for AppleTalk server services via user
1977libraries and applications.
1978.It Cd options BLUETOOTH
1979Include support for the Bluetooth protocol stack.
1980See
1981.Xr bluetooth 4
1982for details.
1983.It Cd options IPNOPRIVPORTS
1984Normally, only root can bind a socket descriptor to a so-called
1985.Dq privileged
1986TCP port, that is, a port number in the range 0-1023.
1987This option eliminates those checks from the kernel.
1988This can be useful if there is a desire to allow daemons without
1989privileges to bind those ports, e.g., on firewalls.
1990The security tradeoffs in doing this are subtle.
1991This option should only be used by experts.
1992.It Cd options TCP_DEBUG
1993Record the last
1994.Em TCP_NDEBUG
1995TCP packets with SO_DEBUG set, and decode to the console if
1996.Em tcpconsdebug
1997is set.
1998.It Cd options TCP_NDEBUG
1999Number of packets to record for
2000.Em TCP_DEBUG .
2001Defaults to 100.
2002.It Cd options TCP_SENDSPACE=value
2003.It Cd options TCP_RECVSPACE=value
2004These options set the max TCP window size to other sizes than the default.
2005The TCP window sizes can be altered via
2006.Xr sysctl 8
2007as well.
2008.It Cd options TCP_INIT_WIN=value
2009This option sets the initial TCP window size for non-local connections,
2010which is used when the transmission starts.
2011The default size is 1, but if the machine should act more aggressively,
2012the initial size can be set to some other value.
2013The initial TCP window size can be set via
2014.Xr sysctl 8
2015as well.
2016.It Cd options TCP_SIGNATURE
2017Enable MD5 TCP signatures (RFC 2385) to protect BGP sessions.
2018.It Cd options IPFILTER_LOG
2019This option, in conjunction with
2020.Em pseudo-device ipfilter ,
2021enables logging of IP packets using IP-Filter.
2022.It Cd options IPFILTER_LOOKUP
2023This option enables the
2024IP-Filter
2025.Xr ippool 8
2026functionality to be enabled.
2027.It Cd options IPFILTER_COMPAT
2028This option enables older IP-Filter binaries to work.
2029.It Cd options IPFILTER_DEFAULT_BLOCK
2030This option sets the default policy of IP-Filter.
2031If it is set, IP-Filter will block packets by default.
2032.It Cd options BRIDGE_IPF
2033This option causes
2034.Em bridge
2035devices to use the IP and/or IPv6 filtering hooks, forming
2036a link-layer filter that uses protocol-layer rules.
2037This option assumes the presence of
2038.Em pseudo-device ipfilter .
2039.It Cd options MBUFTRACE
2040This option can help track down mbuf leaks.
2041When enabled, mbufs are tagged with the devices and protocols using them,
2042which slightly decreases network performance.
2043This additional information can be viewed with
2044.Xr netstat 1 :
2045.Dl Ic netstat Fl mssv
2046Not all devices or protocols support this option.
2047.El
2048.Ss Sysctl Related Options
2049.Bl -ohang
2050.It Cd options SYSCTL_DISALLOW_CREATE
2051Disallows the creation or deletion of nodes from the sysctl tree, as
2052well as the assigning of descriptions to nodes that lack them, by any
2053process.
2054These operations are still available to kernel sub-systems, including
2055loadable kernel modules.
2056.It Cd options SYSCTL_DISALLOW_KWRITE
2057Prevents processes from adding nodes to the sysctl tree that make
2058existing kernel memory areas writable.
2059Sections of kernel memory can still be read and new nodes that own
2060their own data may still be writable.
2061.It Cd options SYSCTL_DEBUG_SETUP
2062Causes the SYSCTL_SETUP routines to print a brief message when they
2063are invoked.
2064This is merely meant as an aid in determining the order in which
2065sections of the tree are created.
2066.It Cd options SYSCTL_DEBUG_CREATE
2067Prints a message each time
2068.Fn sysctl_create ,
2069the function that adds nodes to the tree, is called.
2070.It Cd options SYSCTL_INCLUDE_DESCR
2071Causes the kernel to include short, human readable descriptions for
2072nodes in the sysctl tree.
2073The descriptions can be retrieved programmatically (see
2074.Xr sysctl 3 ) ,
2075or by the sysctl binary itself (see
2076.Xr sysctl 8 ) .
2077The descriptions are meant to give an indication of the purpose and/or
2078effects of a given node's value, not replace the documentation for the
2079given subsystem as a whole.
2080.El
2081.Ss System V IPC Options
2082.Bl -ohang
2083.It Cd options SYSVMSG
2084Includes support for
2085.At V
2086style message queues.
2087See
2088.Xr msgctl 2 ,
2089.Xr msgget 2 ,
2090.Xr msgrcv 2 ,
2091.Xr msgsnd 2 .
2092.It Cd options SYSVSEM
2093Includes support for
2094.At V
2095style semaphores.
2096See
2097.Xr semctl 2 ,
2098.Xr semget 2 ,
2099.Xr semop 2 .
2100.It Cd options SEMMNI=value
2101Sets the number of
2102.At V
2103style semaphore identifiers.
2104The GENERIC config file for your port will have the default.
2105.It Cd options SEMMNS=value
2106Sets the number of
2107.At V
2108style semaphores in the system.
2109The GENERIC config file for your port will have the default.
2110.It Cd options SEMUME=value
2111Sets the maximum number of undo entries per process for
2112.At V
2113style semaphores.
2114The GENERIC config file for your port will have the default.
2115.It Cd options SEMMNU=value
2116Sets the number of undo structures in the system for
2117.At V
2118style semaphores.
2119The GENERIC config file for your port will have the default.
2120.It Cd options SYSVSHM
2121Includes support for
2122.At V
2123style shared memory.
2124See
2125.Xr shmat 2 ,
2126.Xr shmctl 2 ,
2127.Xr shmdt 2 ,
2128.Xr shmget 2 .
2129.It Cd options SHMMAXPGS=value
2130Sets the maximum number of
2131.At V
2132style shared memory pages that are available through the
2133.Xr shmget 2
2134system call.
2135Default value is 1024 on most ports.
2136See
2137.Pa /usr/include/machine/vmparam.h
2138for the default.
2139.El
2140.Ss VM Related Options
2141.Bl -ohang
2142.It Cd options NMBCLUSTERS=value
2143The number of mbuf clusters the kernel supports.
2144Mbuf clusters are MCLBYTES in size (usually 2k).
2145This is used to compute the size of the kernel VM map
2146.Em mb_map ,
2147which maps mbuf clusters.
2148Default on most ports is 1024 (2048 with
2149.Dq options GATEWAY
2150).
2151See
2152.Pa /usr/include/machine/param.h
2153for exact default information.
2154Increase this value if you get
2155.Dq mclpool limit reached
2156messages.
2157.It Cd options NKMEMPAGES=value
2158.It Cd options NKMEMPAGES_MIN=value
2159.It Cd options NKMEMPAGES_MAX=value
2160Size of kernel VM map
2161.Em kmem_map ,
2162in PAGE_SIZE-sized chunks (the VM page size; this value may be read
2163from the
2164.Xr sysctl 8
2165variable
2166.Em hw.pagesize
2167).
2168This VM map is used to map the kernel malloc arena.
2169The kernel attempts to auto-size this map based on the amount of
2170physical memory in the system.
2171Platform-specific code may place bounds on this computed size,
2172which may be viewed with the
2173.Xr sysctl 8
2174variable
2175.Em vm.nkmempages .
2176See
2177.Pa /usr/include/machine/param.h
2178for the default upper and lower bounds.
2179The related options
2180.Sq NKMEMPAGES_MIN
2181and
2182.Sq NKMEMPAGES_MAX
2183allow the bounds to be overridden in the kernel configuration file.
2184These options are provided in the event the computed value is
2185insufficient resulting in an
2186.Dq out of space in kmem_map
2187panic.
2188.It Cd options SB_MAX=value
2189Sets the max size in bytes that a socket buffer is allowed to occupy.
2190The default is 256k, but sometimes it needs to be increased, for example
2191when using large TCP windows.
2192This option can be changed via
2193.Xr sysctl 8
2194as well.
2195.It Cd options SOMAXKVA=value
2196Sets the maximum size of kernel virtual memory that the socket buffers
2197are allowed to use.
2198The default is 16MB, but in situations where for example large TCP
2199windows are used this value must also be increased.
2200This option can be changed via
2201.Xr sysctl 8
2202as well.
2203.It Cd options BUFCACHE=value
2204Size of the buffer cache as a percentage of total available RAM.
2205Ignored if BUFPAGES is also specified.
2206.It Cd options NBUF=value
2207Sets the number of buffer headers available, i.e., the number of
2208open files that may have a buffer cache entry.
2209Each buffer header
2210requires MAXBSIZE (machine dependent, but usually 65536) bytes.
2211The default value is machine dependent, but is usually equal to the
2212value of BUFPAGES.
2213.It Cd options BUFPAGES=value
2214These options set the number of pages available for the buffer cache.
2215Their default value is a machine dependent value, often calculated as
2216between 5% and 10% of total available RAM.
2217.It Cd options MAXTSIZ=bytes
2218Sets the maximum size limit of a process' text segment.
2219See
2220.Pa /usr/include/machine/vmparam.h
2221for the port-specific default.
2222.It Cd options DFLDSIZ=bytes
2223Sets the default size limit of a process' data segment, the value that
2224will be returned as the soft limit for
2225.Dv RLIMIT_DATA
2226(as returned by
2227.Xr getrlimit 2 ) .
2228See
2229.Pa /usr/include/machine/vmparam.h
2230for the port-specific default.
2231.It Cd options MAXDSIZ=bytes
2232Sets the maximum size limit of a process' data segment, the value that
2233will be returned as the hard limit for
2234.Dv RLIMIT_DATA
2235(as returned by
2236.Xr getrlimit 2 ) .
2237See
2238.Pa /usr/include/machine/vmparam.h
2239for the port-specific default.
2240.It Cd options DFLSSIZ=bytes
2241Sets the default size limit of a process' stack segment, the value that
2242will be returned as the soft limit for
2243.Dv RLIMIT_STACK
2244(as returned by
2245.Xr getrlimit 2 ) .
2246See
2247.Pa /usr/include/machine/vmparam.h
2248for the port-specific default.
2249.It Cd options MAXSSIZ=bytes
2250Sets the maximum size limit of a process' stack segment, the value that
2251will be returned as the hard limit for
2252.Dv RLIMIT_STACK
2253(as returned by
2254.Xr getrlimit 2 ) .
2255See
2256.Pa /usr/include/machine/vmparam.h
2257for the port-specific default.
2258.It Cd options DUMP_ON_PANIC=integer
2259Defaults to one.
2260If set to zero, the kernel will not dump to the dump device when
2261it panics, though dumps can still be forced via
2262.Xr ddb 4
2263with the
2264.Dq sync
2265command.
2266Note that this sets the value of the
2267.Em kern.dump_on_panic
2268.Xr sysctl 3
2269variable which may be changed at run time -- see
2270.Xr sysctl 8
2271for details.
2272.It Cd options USE_TOPDOWN_VM
2273User space memory allocations (as made by
2274.Xr mmap 2 )
2275will be arranged in a
2276.Dq top down
2277fashion instead of the traditional
2278.Dq upwards from MAXDSIZ \&+ vm_daddr
2279method.
2280This includes the placement of
2281.Xr ld.so 1 .
2282Arranging memory in this manner allows either (or both of) the heap or
2283.Xr mmap 2
2284allocated space to grow larger than traditionally possible.
2285This option is not available on all ports, but is instead expected to be
2286offered on a port-by-port basis, after which some ports will commit to
2287using it by default.
2288See the files
2289.Pa /usr/include/uvm/uvm_param.h
2290for some implementation details, and
2291.Pa /usr/include/machine/vmparam.h
2292for port specific details including availability.
2293.It Cd options VMSWAP
2294Enable paging device/file support.
2295This option is on by default.
2296.It Cd options PDPOLICY_CLOCKPRO
2297Use CLOCK-Pro, an alternative page replace policy.
2298.El
2299.Ss Security Options
2300.Bl -ohang
2301.It Cd options INSECURE
2302Initializes the kernel security level with \-1 instead of 0.
2303This means that the system always starts in secure level \-1 mode, even when
2304running multiuser, unless the securelevel variable is set to value > \-1 in
2305.Pa /etc/rc.conf .
2306In this case the kernel security level will be raised to that value when the
2307.Pa /etc/rc.d/securelevel
2308script is run during system startup.
2309See the manual page for
2310.Xr init 8
2311for details on the implications of this.
2312The kernel secure level may manipulated by the superuser by altering the
2313.Em kern.securelevel
2314.Xr sysctl 3
2315variable (the secure level may only be lowered by a call from process ID 1,
2316i.e.,
2317.Xr init 8 ) .
2318See also
2319.Xr secmodel_securelevel 9 ,
2320.Xr sysctl 8
2321and
2322.Xr sysctl 3 .
2323.It Cd options VERIFIED_EXEC_FP_SHA256
2324Enables support for SHA256 hashes in Veriexec.
2325.It Cd options VERIFIED_EXEC_FP_SHA384
2326Enables support for SHA384 hashes in Veriexec.
2327.It Cd options VERIFIED_EXEC_FP_SHA512
2328Enables support for SHA512 hashes in Veriexec.
2329.It Cd options PAX_MPROTECT=value
2330Enables PaX MPROTECT,
2331.Xr mprotect 2
2332restrictions from the PaX project.
2333.Pp
2334The
2335.Ar value
2336is the default value for the
2337.Em global
2338knob, see
2339.Xr sysctl 3 .
2340If 0, PaX MPROTECT will be enabled only if explicitly set on programs
2341using
2342.Xr paxctl 8 .
2343If 1, PaX MPROTECT will be enabled for all programs.
2344Programs can be exempted using
2345.Xr paxctl 8 .
2346.Pp
2347See
2348.Xr security 7
2349for more details.
2350.It Cd options PAX_SEGVGUARD=value
2351Enables PaX Segvguard.
2352Requires
2353.Cd options FILEASSOC .
2354.Pp
2355The
2356.Ar value
2357is the default value for the
2358.Em global
2359knob, see
2360.Xr sysctl 3 .
2361If 0, PaX Segvguard will be enabled only if explicitly set on programs
2362using
2363.Xr paxctl 8 .
2364If 1, PaX Segvguard will be enabled to all programs, and exemption can
2365be done using
2366.Xr paxctl 8 .
2367.Pp
2368See
2369.Xr security 7
2370for more details.
2371.It Cd options PAX_ASLR=value
2372Enables PaX ASLR.
2373.Pp
2374The
2375.Ar value
2376is the default value for the
2377.Em global
2378knob, see
2379.Xr sysctl 3 .
2380If 0, PaX ASLR will be enabled only if explicitly set on programs
2381using
2382.Xr paxctl 8 .
2383If 1, PaX ASLR will be enabled to all programs, and exemption can
2384be done using
2385.Xr paxctl 8 .
2386.Pp
2387See
2388.Xr security 7
2389for more details.
2390.It Cd options USER_VA0_DISABLE_DEFAULT=value
2391Sets the initial value of the flag which controls whether user programs
2392can map virtual address 0.
2393The flag can be changed at runtime by
2394.Xr sysctl 3 .
2395.El
2396.Ss amiga-specific Options
2397.Bl -ohang
2398.It Cd options BB060STUPIDROM
2399When the bootloader (which passes AmigaOS ROM information) claims
2400we have a 68060 CPU without FPU, go look into the Processor
2401Configuration Register (PCR) to find out.
2402You need this with Amiga ROMs up to (at least) V40.xxx (OS3.1),
2403when you boot via the bootblocks and don't have a DraCo.
2404.It Cd options IOBZCLOCK=frequency
2405The IOBlix boards come with two different serial master clocks: older ones
2406use 24 MHz, newer ones use 22.1184 MHz.
2407The driver normally assumes the latter.
2408If your board uses 24 MHz, you can recompile your kernel with
2409options IOBZCLOCK=24000000
2410or patch the kernel variable iobzclock to the same value.
2411.It Cd options LIMITMEM=value
2412If there, limit the part of the first memory bank used by
2413.Nx
2414to value megabytes.
2415Default is unlimited.
2416.It Cd options P5PPC68KBOARD
2417Add special support for Phase5 mixed 68k+PPC boards.
2418Currently, this only affects rebooting from
2419.Nx
2420and is only needed on 68040+PPC, not on
242168060+PPC; without this, affected machines will hang after
2422.Nx
2423has shut
2424down and will only restart after a keyboard reset or a power cycle.
2425.El
2426.Ss atari-specific Options
2427.Bl -ohang
2428.It Cd options DISKLABEL_AHDI
2429Include support for AHDI (native Atari) disklabels.
2430.It Cd options DISKLABEL_NBDA
2431Include support for
2432.Nx Ns /atari
2433labels.
2434If you don't set this option, it will be set automatically.
2435.Nx Ns /atari
2436will not work without it.
2437.It Cd options FALCON_SCSI
2438Include support for the 5380-SCSI configuration as found on the Falcon.
2439.It Cd options RELOC_KERNEL
2440If set, the kernel will relocate itself to TT-RAM, if possible.
2441This will give you a slightly faster system.
2442.Em Beware
2443that on some TT030 systems,
2444the system will frequently dump with MMU-faults with this option enabled.
2445.It Cd options SERCONSOLE
2446Allow the modem1-port to act as the system-console.
2447A carrier should be active on modem1 during system boot to active
2448the console functionality.
2449.It Cd options TT_SCSI
2450Include support for the 5380-SCSI configuration as found on the TT030
2451and Hades.
2452.El
2453.Ss i386-specific Options
2454.Bl -ohang
2455.It Cd options CPURESET_DELAY=value
2456Specifies the time (in millisecond) to wait before doing a hardware reset
2457in the last phase of a reboot.
2458This gives the user a chance to see error messages from the shutdown
2459operations (like NFS unmounts, buffer cache flush, etc ...).
2460Setting this to 0 will disable the delay.
2461Default is 2 seconds.
2462.It Cd options USER_LDT
2463Include i386-specific system calls for modifying the local descriptor table,
2464used by Windows emulators.
2465.It Cd options PAE
2466Enable PAE (Physical Address Extension) mode.
2467PAE permits up to 36 bits physical addressing (64GB of physical
2468memory), and turns physical addresses to 64 bits entities in the
2469memory management subsystem.
2470Userland virtual address space remains at 32 bits (4GB).
2471PAE mode is required to enable the NX/XD (No-eXecute/eXecute Disable)
2472bit for pages, which allows marking certain ones as not being executable.
2473Any attempt to execute code from such a page will raise an exception.
2474.It Cd options REALBASEMEM=integer
2475Overrides the base memory size passed in from the boot block.
2476(Value given in kilobytes.)
2477Use this option only if the boot block reports the size incorrectly.
2478(Note that some BIOSes put the extended BIOS
2479data area at the top of base memory, and therefore report a smaller
2480base memory size to prevent programs overwriting it.
2481This is correct behavior, and you should not use the
2482.Em REALBASEMEM
2483option to access this memory).
2484.It Cd options REALEXTMEM=integer
2485Overrides the extended memory size passed in from the boot block.
2486(Value given in kilobytes.
2487Extended memory does not include the first megabyte.)
2488Use this option only if the boot block reports the size incorrectly.
2489.It Cd options CYRIX_CACHE_WORKS
2490Relevant only to the Cyrix 486DLC CPU.
2491This option is used to turn on the cache in hold-flush mode.
2492It is not turned on by default because it is known to have problems in
2493certain motherboard implementations.
2494.It Cd options CYRIX_CACHE_REALLY_WORKS
2495Relevant only to the Cyrix 486DLC CPU.
2496This option is used to turn on the cache in write-back mode.
2497It is not turned on by default because it is known to have problems in
2498certain motherboard implementations.
2499In order for this option to take effect, option
2500.Em CYRIX_CACHE_WORKS
2501must also be specified.
2502.It Cd options PCIBIOS
2503Enable support for initializing the PCI bus using information from
2504the BIOS.
2505See
2506.Xr pcibios 4
2507for details.
2508.It Cd options MTRR
2509Include support for accessing MTRR registers from user-space.
2510See
2511.Xr i386_get_mtrr 2 .
2512.It Cd options BEEP_ONHALT
2513Make the system speaker emit several beeps when it is completely safe to
2514power down the computer after a
2515.Xr halt 8
2516command.
2517Requires
2518.Xr sysbeep 4
2519support.
2520.It Cd options BEEP_ONHALT_COUNT=times
2521Number of times to beep the speaker when
2522.Cd options BEEP_ONHALT
2523is enabled.
2524Defaults to 3.
2525.It Cd options BEEP_ONHALT_PITCH=hz
2526The tone frequency used when
2527.Cd options BEEP_ONHALT
2528option, in hertz.
2529Defaults to 1500.
2530.It Cd options BEEP_ONHALT_PERIOD=msecs
2531The duration of each beep when
2532.Cd options BEEP_ONHALT
2533is enabled, in milliseconds.
2534Defaults to 250.
2535.It Cd options MULTIBOOT
2536Makes the kernel Multiboot-compliant, allowing it to be booted through
2537a Multiboot-compliant boot manager such as GRUB.
2538See
2539.Xr multiboot 8
2540for more information.
2541.It Cd options SPLASHSCREEN
2542Display a splash screen during boot.
2543.It Cd options SPLASHSCREEN_PROGRESS
2544Display a progress bar at the splash screen during boot.
2545This option requires
2546.Em SPLASHSCREEN .
2547.El
2548.Ss isa-specific Options
2549Options specific to
2550.Xr isa 4
2551busses.
2552.Bl -ohang
2553.It Cd options PCIC_ISA_ALLOC_IOBASE=address, PCIC_ISA_ALLOC_IOSIZE=size
2554Control the section of IO bus space used for PCMCIA bus space mapping.
2555Ideally the probed defaults are satisfactory, however in practice
2556that is not always the case.
2557See
2558.Xr pcmcia 4
2559for details.
2560.It Cd options PCIC_ISA_INTR_ALLOC_MASK=mask
2561Controls the allowable interrupts that may be used for PCMCIA
2562devices.
2563This mask is a logical-or of power-of-2s of allowable interrupts:
2564.Bd -literal -offset 04n
2565.Em "IRQ Val      IRQ Val      IRQ Val       IRQ Val"
2566 0  0x0001    4  0x0010    8  0x0100    12  0x1000
2567 1  0x0002    5  0x0020    9  0x0200    13  0x2000
2568 2  0x0004    6  0x0040   10  0x0400    14  0x4000
2569 3  0x0008    7  0x0080   11  0x0800    15  0x8000
2570.Ed
2571.It Cd options PCKBC_CNATTACH_SELFTEST
2572Perform a self test of the keyboard controller before attaching it as a
2573console.
2574This might be necessary on machines where we boot on cold iron, and
2575pckbc refuses to talk until we request a self test.
2576Currently only the netwinder port uses it.
2577.It Cd options PCKBD_CNATTACH_MAY_FAIL
2578If this option is set the PS/2 keyboard will not be used as the console
2579if it cannot be found during boot.
2580This allows other keyboards, like USB, to be the console keyboard.
2581.It Cd options PCKBD_LAYOUT=layout
2582Sets the default keyboard layout, see
2583.Xr pckbd 4 .
2584.El
2585.Ss m68k-specific Options
2586.Bl -ohang
2587.It Cd options FPU_EMULATE
2588Include support for MC68881/MC68882 emulator.
2589.It Cd options FPSP
2590Include support for 68040 floating point.
2591.It Cd options M68020,M68030,M68040,M68060
2592Include support for a specific CPU,
2593at least one (the one you are using) should be specified.
2594.It Cd options M060SP
2595Include software support for 68060.
2596This provides emulation of unimplemented
2597integer instructions as well as emulation of unimplemented floating point
2598instructions and data types and software support for floating point traps.
2599.El
2600.Ss powerpc-specific Options (OEA Only)
2601.Bl -ohang
2602.It Cd options PMAP_MEMLIMIT=value
2603Limit the amount of memory seen by the kernel to
2604.Ar value
2605bytes.
2606.It Cd options PTEGCOUNT=value
2607Specify the size of the page table as
2608.Ar value
2609PTE groups.
2610Normally, one PTEG is allocated per physical page frame.
2611.El
2612.Ss sparc-specific Options
2613.Bl -ohang
2614.It Cd options AUDIO_DEBUG
2615Enable simple event debugging of the logging of the
2616.Xr audio 4
2617device.
2618.It Cd options BLINK
2619Enable blinking of LED.
2620Blink rate is full cycle every N seconds for
2621N < then current load average.
2622See
2623.Xr getloadavg 3 .
2624.\" .It Cd options COLORFONT_CACHE
2625.\" What does this do?
2626.It Cd options COUNT_SW_LEFTOVERS
2627Count how many times the sw SCSI device has left 3, 2, 1 and 0 in the
2628sw_3_leftover, sw_2_leftover, sw_1_leftover, and sw_0_leftover
2629variables accessible from
2630.Xr ddb 4 .
2631See
2632.Xr sw 4 .
2633.It Cd options DEBUG_ALIGN
2634Adds debugging messages calls when user-requested alignment fault
2635handling happens.
2636.It Cd options DEBUG_EMUL
2637Adds debugging messages calls for emulated floating point and
2638alignment fixing operations.
2639.It Cd options DEBUG_SVR4
2640Prints registers messages calls for emulated SVR4 getcontext and
2641setcontext operations.
2642See
2643.Em options COMPAT_SVR4 .
2644.It Cd options EXTREME_DEBUG
2645Adds debugging functions callable from
2646.Xr ddb 4 .
2647The debug_pagetables, test_region and print_fe_map
2648functions print information about page tables for the SUN4M
2649platforms only.
2650.It Cd options EXTREME_EXTREME_DEBUG
2651Adds extra info to
2652.Em options EXTREME_DEBUG .
2653.It Cd options FPU_CONTEXT
2654Make
2655.Em options COMPAT_SVR4
2656getcontext and setcontext include floating point registers.
2657.It Cd options MAGMA_DEBUG
2658Adds debugging messages to the
2659.Xr magma 4
2660device.
2661.It Cd options RASTERCONS_FULLSCREEN
2662Use the entire screen for the console.
2663.It Cd options RASTERCONS_SMALLFONT
2664Use the Fixed font on the console, instead of the normal font.
2665.It Cd options SUN4
2666Support sun4 class machines.
2667.It Cd options SUN4C
2668Support sun4c class machines.
2669.It Cd options SUN4M
2670Support sun4m class machines.
2671.It Cd options SUN4_MMU3L
2672.\" XXX ???
2673Enable support for sun4 3-level MMU machines.
2674.It Cd options V9
2675Enable SPARC V9 assembler in
2676.Xr ddb 4 .
2677.El
2678.Ss sparc64-specific Options
2679.Bl -ohang
2680.It Cd options AUDIO_DEBUG
2681Enable simple event debugging of the logging of the
2682.Xr audio 4
2683device.
2684.It Cd options BLINK
2685Enable blinking of LED.
2686Blink rate is full cycle every N seconds for
2687N < then current load average.
2688See
2689.Xr getloadavg 3 .
2690.El
2691.Ss x68k-specific Options
2692.Bl -ohang
2693.It Cd options EXTENDED_MEMORY
2694Include support for extended memory, e.g., TS-6BE16 and 060turbo on-board.
2695.It Cd options JUPITER
2696Include support for Jupiter-X MPU accelerator
2697.It Cd options ZSCONSOLE,ZSCN_SPEED=value
2698Use the built-in serial port as the system-console.
2699Speed is specified in bps, defaults to 9600.
2700.It Cd options ITE_KERNEL_ATTR=value
2701Set the kernel message attribute for ITE.
2702Value, an integer, is a logical or of the following values:
2703.Bl -tag -width 4n -compact -offset indent
2704.It 1
2705color inversed
2706.It 2
2707underlined
2708.It 4
2709bolded
2710.El
2711.El
2712.\" The following requests should be uncommented and used where appropriate.
2713.\" .Sh FILES
2714.\" .Sh EXAMPLES
2715.Sh SEE ALSO
2716.Xr config 1 ,
2717.Xr gdb 1 ,
2718.Xr ktrace 1 ,
2719.Xr pmc 1 ,
2720.Xr quota 1 ,
2721.Xr vndcompress 1 ,
2722.Xr gettimeofday 2 ,
2723.Xr i386_get_mtrr 2 ,
2724.Xr i386_iopl 2 ,
2725.Xr msgctl 2 ,
2726.Xr msgget 2 ,
2727.Xr msgrcv 2 ,
2728.Xr msgsnd 2 ,
2729.Xr ntp_adjtime 2 ,
2730.Xr ntp_gettime 2 ,
2731.Xr reboot 2 ,
2732.Xr semctl 2 ,
2733.Xr semget 2 ,
2734.Xr semop 2 ,
2735.Xr shmat 2 ,
2736.Xr shmctl 2 ,
2737.Xr shmdt 2 ,
2738.Xr shmget 2 ,
2739.Xr sysctl 3 ,
2740.Xr apm 4 ,
2741.Xr ddb 4 ,
2742.Xr inet 4 ,
2743.Xr md 4 ,
2744.Xr pcibios 4 ,
2745.Xr pcmcia 4 ,
2746.Xr ppp 4 ,
2747.Xr userconf 4 ,
2748.Xr vnd 4 ,
2749.Xr wscons 4 ,
2750.Xr config 5 ,
2751.Xr edquota 8 ,
2752.Xr init 8 ,
2753.Xr mdsetimage 8 ,
2754.Xr mount_cd9660 8 ,
2755.Xr mount_fdesc 8 ,
2756.Xr mount_kernfs 8 ,
2757.Xr mount_lfs 8 ,
2758.Xr mount_mfs 8 ,
2759.Xr mount_msdos 8 ,
2760.Xr mount_nfs 8 ,
2761.Xr mount_ntfs 8 ,
2762.Xr mount_null 8 ,
2763.Xr mount_portal 8 ,
2764.Xr mount_procfs 8 ,
2765.Xr mount_udf 8 ,
2766.Xr mount_umap 8 ,
2767.Xr mount_union 8 ,
2768.Xr mrouted 8 ,
2769.Xr newfs_lfs 8 ,
2770.Xr ntpd 8 ,
2771.Xr quotaon 8 ,
2772.Xr rpc.rquotad 8 ,
2773.Xr sysctl 8 ,
2774.Xr in_getifa 9 ,
2775.Xr kernhist 9
2776.Sh HISTORY
2777The
2778.Nm
2779man page first appeared in
2780.Nx 1.3 .
2781.Sh BUGS
2782The
2783.Em EON
2784option should be a pseudo-device, and is also very fragile.
2785