xref: /netbsd-src/share/man/man4/options.4 (revision 8ecbf5f02b752fcb7debe1a8fab1dc82602bc760)
1.\"	$NetBSD: options.4,v 1.516 2020/08/28 16:07:49 fcambus Exp $
2.\"
3.\" Copyright (c) 1996
4.\" 	Perry E. Metzger.  All rights reserved.
5.\"
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in the
13.\"    documentation and/or other materials provided with the distribution.
14.\" 3. All advertising materials mentioning features or use of this software
15.\"    must display the following acknowledgment:
16.\"	This product includes software developed for the NetBSD Project
17.\"	by Perry E. Metzger.
18.\" 4. The name of the author may not be used to endorse or promote products
19.\"    derived from this software without specific prior written permission.
20.\"
21.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
22.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
23.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
24.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
25.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
26.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
27.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
28.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
29.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
30.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31.\"
32.\"
33.Dd August 25, 2020
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 makeoptions Ar name=value
127Defines a
128.Xr make 1
129macro
130.Ar name
131with the value
132.Ar value
133in the kernel Makefile.
134.It Sy maxusers Ar integer
135Set the maxusers variable in the kernel.
136.It Sy no Ar keyword Ar name Op Ar arguments Op ...
137For the
138.Xr config 1
139.Ar keywords
140file-system, makeoptions, options, and pseudo-device,
141.Ar no
142removes the file-system, makeoption, options, or pseudo-device,
143.Ar name .
144This is useful when a kernel configuration file includes another which
145has undesired options.
146.Pp
147For example, a local configuration file that wanted the kitchen sink, but
148not COMPAT_09 or bridging, might be:
149.Bd -literal -offset indent
150include "arch/i386/conf/GENERIC"
151no options COMPAT_09
152no pseudo-device bridge
153.Ed
154.It Sy options Ar option_name Op , Ar option_name=value Op ...
155Specifies (or sets) the option, or comma-separated list of options,
156.Ar option_name .
157Some options expect to be assigned a value, which may be an integer,
158a double-quoted word, a bare word, or an empty string
159.Pq Qq .
160Note that those are eventually handled by the C compiler, so the rules
161of that language apply.
162.Pp
163.Em Note :
164Options that are not defined by device definition files are passed to
165the compile process as
166.Fl D
167flags to the C compiler.
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_80
402Enable binary compatibility with
403.Nx 8.0 .
404.It Cd options COMPAT_90
405Enable binary compatibility with
406.Nx 9.0 .
407.It Cd options COMPAT_BSDPTY
408This option is currently on by default and enables the pty multiplexer
409.Xr ptm 4
410and
411.Xr ptmx 4
412to find and use ptys named
413.Pa /dev/ptyXX
414(master) and
415.Pa /dev/ttyXX
416(slave).
417Eventually this option will become optional as ptyfs based pseudo-ttys become
418the default, see
419.Xr mount_ptyfs 8 .
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_NOMID
450Enable compatibility with
451.Xr a.out 5
452executables that lack a machine ID.
453This includes
454.Nx 0.8 Ns 's
455ZMAGIC format, and 386BSD and BSDI's
456QMAGIC, NMAGIC, and OMAGIC
457.Xr a.out 5
458formats.
459.It Cd options COMPAT_NETBSD32
460On those architectures that support it, this enables binary
461compatibility with 32-bit applications built for the same architecture.
462This is currently limited to the amd64 and sparc64 ports, and only
463applicable for 64-bit kernels.
464.It Cd options COMPAT_AOUT_M68K
465On m68k architectures which have switched to ELF,
466this enables binary compatibility with
467.Nx Ns /m68k
468.Xr a.out 5
469executables on
470.Nx Ns /m68k
471ELF kernels.
472This handles alignment incompatibility of m68k ABI between
473a.out and ELF which causes the structure padding differences.
474Currently only some system calls which use
475.Va struct stat
476are adjusted and some binaries which use
477.Xr sysctl 3
478to retrieve network details would not work properly.
479.It Cd options EMUL_NATIVEROOT=string
480Just like emulated binaries first try looking up files in
481an emulation root (e.g.
482.Pa /emul/linux )
483before looking them up in real root, this option causes native
484binaries to first look up files in an "emulation" directory too.
485This can be useful to test an amd64 kernel on top of an i386 system
486before full migration: by unpacking the amd64 distribution in e.g.
487.Pa /emul/netbsd64
488and specifying that location as
489.Cd EMUL_NATIVEROOT ,
490native amd64 binaries can be run while the root file system remains
491populated with i386 binaries.
492Beware of
493.Pa /dev
494incompatibilities between i386 and amd64 if you do this.
495.It Cd options EXEC_ELF_NOTELESS
496Run unidentified ELF binaries as
497.Nx
498binaries.
499This might be needed for very old
500.Nx
501ELF binaries on some archs.
502These old binaries didn't contain an appropriate
503.Li .note.netbsd.ident
504section, and thus can't be identified by the kernel as
505.Nx
506binaries otherwise.
507Beware - if this option is on, the kernel would run
508.Em any
509unknown ELF binaries as if they were
510.Nx
511binaries.
512.El
513.Ss Debugging Options
514.Bl -ohang
515.It Cd options DDB
516Compiles in a kernel debugger for diagnosing kernel problems.
517See
518.Xr ddb 4
519for details.
520.Em NOTE :
521not available on all architectures.
522.It Cd options DDB_FROMCONSOLE=integer
523If set to non-zero, DDB may be entered by sending a break on a serial
524console or by a special key sequence on a graphics console.
525A value of "0" ignores console breaks or key sequences.
526If not explicitly specified, the default value is "1".
527Note that this sets the value of the
528.Em ddb.fromconsole
529.Xr sysctl 3
530variable which may be changed at run time \(em see
531.Xr sysctl 8
532for details.
533.It Cd options DDB_HISTORY_SIZE=integer
534If this is non-zero, enable history editing in the kernel debugger
535and set the size of the history to this value.
536.It Cd options DDB_ONPANIC
537The default if not specified is
538.Dq 1
539- just enter into DDB.
540If set to
541.Dq 0
542the kernel will attempt to print out a stack trace
543and reboot the system.
544If set to
545.Dq -1
546then neither a stack trace is printed or DDB entered -
547it is as if DDB were not compiled into the kernel.
548Note that this sets the value of the
549.Em ddb.onpanic
550.Xr sysctl 3
551variable which may be changed at run time \(em see
552.Xr sysctl 8
553for details.
554.It Cd options DDB_COMMANDONENTER=string
555This option specify commands which will be executed on each entry to DDB.
556This sets the default value of the
557.Em ddb.commandonenter
558.Xr sysctl 3
559variable which may be changed at run time.
560.It Cd options DDB_BREAK_CHAR=integer
561This option overrides using break to enter the kernel debugger
562on the serial console.
563The value given is the ASCII value to be used instead.
564This is currently only supported by the com driver.
565.It Cd options DDB_VERBOSE_HELP
566This option adds more verbose descriptions to the
567.Em help
568command.
569.It Cd options DDB_PANICSTACKFRAMES=integer
570Number of stack frames to display on panic.
571Useful to avoid scrolling away the interesting frames on a glass tty.
572Default value is
573.Dv 65535
574(all frames), useful value around
575.Dv 10 .
576.It Cd options KGDB
577Compiles in a remote kernel debugger stub for diagnosing kernel problems
578using the
579.Dq remote target
580feature of gdb.
581See
582.Xr gdb 1
583for details.
584.Em NOTE :
585not available on all architectures.
586.It Cd options KGDB_DEV
587Device number
588.Po
589as a
590.Dv dev_t
591.Pc
592of kgdb device.
593.It Cd options KGDB_DEVADDR
594Memory address of kgdb device.
595.It Cd options KGDB_DEVMODE
596Permissions of kgdb device.
597.It Cd options KGDB_DEVNAME
598Device name of kgdb device.
599.It Cd options KGDB_DEVRATE
600Baud rate of kgdb device.
601.It Cd makeoptions DEBUG="-g"
602The
603.Fl g
604flag causes
605.Pa netbsd.gdb
606to be built in addition to
607.Pa netbsd .
608.Pa netbsd.gdb
609is useful for debugging kernel crash dumps with gdb.
610See
611.Xr gdb 1
612for details.
613.It Cd options DEBUG
614Turns on miscellaneous kernel debugging.
615Since options are turned into preprocessor defines (see above),
616.Em options DEBUG
617is equivalent to doing a
618.Em #define DEBUG
619throughout the kernel.
620Much of the kernel has
621.Em #ifdef DEBUG
622conditionalized debugging code.
623Note that many parts of the kernel (typically device drivers) include their own
624.Em #ifdef XXX_DEBUG
625conditionals instead.
626This option also turns on certain other options,
627which may decrease system performance.
628Systems with this option are not suitable for regular use, and are
629intended only for debugging or looking for bugs.
630.It Cd options DIAGNOSTIC
631Adds code to the kernel that does internal consistency checks.
632This code will cause the kernel to panic if corruption of internal data
633structures is detected.
634Historically, the performance degradation is sufficiently small that
635it is reasonable for systems with
636.Em options DIAGNOSTIC
637to be in production use, with the real consideration not being
638performance but instead a preference for more panics versus continued
639operation with undetected problems.
640.It Cd options LOCKDEBUG
641Adds code to the kernel to detect incorrect use of locking primitives
642(mutex, rwlock).
643This code will cause the kernel to check for dead lock conditions.
644It will also check for memory being freed to not contain initialised
645lock primitives.
646Functions for use in
647.Xr ddb 4
648to check lock chains etc. are also enabled.
649These checks are very expensive and can decrease performance on
650multi-processor machines by a factor of three.
651.It Cd options KDTRACE_HOOKS
652Adds hooks for the DTrace tracing facility, which allows users to
653analyze many aspects of system and application behavior.
654See
655.Xr dtrace 1
656for details.
657.It Cd options KSTACK_CHECK_MAGIC
658Check kernel stack usage and panic if stack overflow is detected.
659This check is performance sensitive because it scans stack on each context
660switch.
661.It Cd options KTRACE
662Add hooks for the system call tracing facility, which allows users to
663watch the system call invocation behavior of processes.
664See
665.Xr ktrace 1
666for details.
667.It Cd options MSGBUFSIZE=integer
668This option sets the size of the kernel message buffer.
669This buffer holds the kernel output of
670.Fn printf
671when not (yet) read by
672.Xr syslogd 8 .
673This is particularly useful when the system has crashed and you wish to lookup
674the kernel output from just before the crash.
675Also, since the autoconfig output becomes more and more verbose,
676it sometimes happens that the message buffer overflows before
677.Xr syslogd 8
678was able to read it.
679Note that not all systems are capable of obtaining a variable sized message
680buffer.
681There are also some systems on which memory contents are not preserved
682across reboots.
683.It Cd options KERNHIST
684Enables the kernel history logs, which create in-memory traces of
685various kernel activities.
686These logs can be displayed by using
687.Cm show kernhist
688from DDB.
689See the kernel source file
690.Pa sys/kern/kern_history.c
691and the
692.Xr kernhist 9
693manual for details.
694.It Cd options KERNHIST_PRINT
695Prints the kernel history logs on the system console as entries are added.
696Note that the output is
697.Em extremely
698voluminous, so this option is really only useful for debugging
699the very earliest parts of kernel initialization.
700.It Cd options UVMHIST
701Like
702.Em KERNHIST ,
703it enables the UVM history logs.
704These logs can be displayed by using
705.Cm show kernhist
706from DDB.
707See the kernel source file
708.Pa sys/uvm/uvm_stat.c
709for details.
710.It Cd options UVMHIST_PRINT
711Like
712.Em UVMHIST ,
713it prints the UVM history logs on the system console as entries are added.
714Note that the output is
715.Em extremely
716voluminous, so this option is really only useful for debugging
717the very earliest parts of kernel initialization.
718.It Cd options UVMHIST_MAPHIST_SIZE
719Set the size of the
720.Dq maphist
721kernel history.
722The default is 100.
723This option depends upon the
724.Em UVMHIST
725option.
726.It Cd options UVMHIST_PDHIST_SIZE
727Set the size of the
728.Dq pdhist
729kernel history.
730The default is 100.
731This option depends upon the
732.Em UVMHIST
733option.
734.It Cd options BIOHIST
735Like
736.Em KERNHIST ,
737it enables the BIO history logs.
738These logs can be displayed by using
739.Cm show kernhist
740from DDB, and can help in debugging problems with Buffered I/O operations.
741See the kernel source file
742.Pa sys/kern/vfs_vio.c
743for details.
744.It Cd options BIOHIST_PRINT
745Like
746.Em BIOHIST ,
747it prints the BIO history logs on the system console as entries are added.
748Note that the output is
749.Em extremely
750voluminous, so this option is really only useful for debugging
751the very earliest parts of kernel initialization.
752.It Cd options BIOHIST_SIZE
753Set the size of the
754.Dq biohist
755kernel history.
756The default is 500.
757This option depends upon the
758.Em BIOHIST
759option.
760.El
761.Ss File Systems
762.Bl -ohang
763.It Cd file-system FFS
764Includes code implementing the Berkeley Fast File System
765.Po Em FFS Pc .
766Most machines need this if they are not running diskless.
767.It Cd file-system EXT2FS
768Includes code implementing the Second Extended File System
769.Po Em ext2 Pc ,
770revision 0 and revision 1 with the
771.Em filetype ,
772.Em sparse_super
773and
774.Em large_file
775options.
776This is the most commonly used file system on the Linux operating system,
777and is provided here for compatibility.
778Some of the specific features of
779.Em ext2
780like the "behavior on errors" are not implemented.
781See
782.Xr mount_ext2fs 8
783for details.
784.It Cd file-system LFS
785.Bq Em EXPERIMENTAL
786Include the Log-structured File System
787.Po Em LFS Pc .
788See
789.Xr mount_lfs 8
790and
791.Xr newfs_lfs 8
792for details.
793.It Cd file-system MFS
794Include the Memory File System
795.Po Em MFS Pc .
796This file system stores files in swappable memory, and produces
797notable performance improvements when it is used as the file store
798for
799.Pa /tmp
800and similar file systems.
801See
802.Xr mount_mfs 8
803for details.
804.It Cd file-system NFS
805Include the client side of the Network File System (NFS)
806remote file sharing protocol.
807Although the bulk of the code implementing NFS
808is kernel based, several user level daemons are needed for it to work.
809See
810.Xr mount_nfs 8
811for details.
812.It Cd file-system CD9660
813Includes code for the ISO 9660 + Rock Ridge file system, which is
814the standard file system on many CD-ROM discs.
815Useful primarily if you have a CD-ROM drive.
816See
817.Xr mount_cd9660 8
818for details.
819.It Cd file-system MSDOSFS
820Includes the MS-DOS FAT file system, which is reportedly still used
821by unfortunate people who have not heard about
822.Nx .
823Also implements the Windows 95 extensions to the same, which permit
824the use of longer, mixed case file names.
825See
826.Xr mount_msdos 8
827and
828.Xr fsck_msdos 8
829for details.
830.It Cd file-system NTFS
831.Bq Em EXPERIMENTAL
832Includes code for the Microsoft Windows NT file system.
833See
834.Xr mount_ntfs 8
835for details.
836.It Cd file-system FDESC
837Includes code for a file system, conventionally mounted on
838.Pa /dev/fd ,
839which permits access to the per-process file descriptor space via
840special files in the file system.
841See
842.Xr mount_fdesc 8
843for details.
844Note that this facility is redundant, and thus unneeded on most
845.Nx
846systems, since the
847.Xr fd 4
848pseudo-device driver already provides identical functionality.
849On most
850.Nx
851systems, instances of
852.Xr fd 4
853are mknoded under
854.Pa /dev/fd/
855and on
856.Pa /dev/stdin ,
857.Pa /dev/stdout ,
858and
859.Pa /dev/stderr .
860.It Cd file-system KERNFS
861Includes code which permits the mounting of a special file system
862(normally mounted on
863.Pa /kern )
864in which files representing various kernel variables and parameters
865may be found.
866See
867.Xr mount_kernfs 8
868for details.
869.It Cd file-system NULLFS
870Includes code for a loopback file system.
871This permits portions of the file hierarchy to be re-mounted in other places.
872The code really exists to provide an example of a stackable file system layer.
873See
874.Xr mount_null 8
875for details.
876.It Cd file-system OVERLAY
877Includes code for a file system filter.
878This permits the overlay file system to intercept all access to an underlying
879file system.
880This file system is intended to serve as an example of a stacking file
881system which has a need to interpose itself between an underlying file
882system and all other access.
883See
884.Xr mount_overlay 8
885for details.
886.It Cd file-system PROCFS
887Includes code for a special file system (conventionally mounted on
888.Pa /proc )
889in which the process space becomes visible in the file system.
890Among
891other things, the memory spaces of processes running on the system are
892visible as files, and signals may be sent to processes by writing to
893.Pa ctl
894files in the procfs namespace.
895See
896.Xr mount_procfs 8
897for details.
898.It Cd file-system UDF
899.Bq Em EXPERIMENTAL
900Includes code for the UDF file system commonly found on CD and DVD
901media but also on USB sticks.
902Currently supports read and write access up to UDF 2.01 and somewhat limited
903write support for UDF 2.50.
904It is marked experimental since there is no
905.Xr fsck_udf 8 .
906See
907.Xr mount_udf 8
908for details.
909.It Cd file-system UMAPFS
910Includes a loopback file system in which user and group IDs may be
911remapped \(em this can be useful when mounting alien file systems with
912different UIDs and GIDs than the local system.
913See
914.Xr mount_umap 8
915for details.
916.It Cd file-system UNION
917.Bq Em EXPERIMENTAL
918Includes code for the union file system, which permits directories to
919be mounted on top of each other in such a way that both file systems
920remain visible \(em this permits tricks like allowing writing (and the
921deleting of files) on a read-only file system like a CD-ROM by
922mounting a local writable file system on top of the read-only file
923system.
924See
925.Xr mount_union 8
926for details.
927.It Cd file-system CODA
928.Bq Em EXPERIMENTAL
929Includes code for the Coda file system.
930Coda is a distributed file system like NFS and AFS.
931It is freely available, like NFS, but it functions much like AFS in being a
932.Dq stateful
933file system.
934Both Coda and AFS cache files on your local machine to improve performance.
935Then Coda goes a step further than AFS by letting you access the cached
936files when there is no available network, viz. disconnected laptops and
937network outages.
938In Coda, both the client and server are outside the kernel which makes
939them easier to experiment with.
940Coda is available for several UNIX and non-UNIX platforms.
941See
942.Lk http://www.coda.cs.cmu.edu
943for more details.
944.Em NOTE :
945You also need to enable the pseudo-device, vcoda, for the Coda
946file system to work.
947.It Cd file-system PTYFS
948Includes code for a special file system (normally mounted on
949.Pa /dev/pts )
950in which pseudo-terminal slave devices become visible in the file system.
951See
952.Xr mount_ptyfs 8
953for details.
954.It Cd file-system TMPFS
955Includes code for the efficient memory file system, normally used over
956.Pa /tmp .
957See
958.Xr mount_tmpfs 8
959for details.
960.It Cd file-system PUFFS
961Includes kernel support for the pass-to-userspace framework file system.
962It can be used to implement file system functionality in userspace.
963See
964.Xr puffs 3
965for more details.
966This enables for example sshfs:
967.Xr mount_psshfs 8 .
968.El
969.Ss File System Options
970.Bl -ohang
971.It Cd options DISKLABEL_EI
972Enable
973.Dq Endian-Independent
974.Xr disklabel 5
975support.
976This allows a system to recognize a disklabel written in the other byte order.
977For writing, when a label already exists, its byte order is preserved.
978Otherwise, a new label is written in the native byte order.
979To specify the byte order explicitly, the
980.Fl F
981option of
982.Xr disklabel 8
983should be used with the
984.Fl B
985option in order to avoid using
986.Xr ioctl 2 ,
987which results in the default behavior explained above.
988At the moment this option is restricted to the following ports:
989amd64, bebox, emips, epoc32, evbarm, i386, ibmnws, landisk, mvmeppc, prep,
990.\" riscv,
991rs6000, sandpoint,
992.\" usermode,
993xen, and zaurus;
994also to machines of the
995evbmips and evbppc ports that support
996Master Boot Record (MBR).
997.It Cd options MAGICLINKS
998Enables the expansion of special strings
999.Po
1000beginning with
1001.Dq @
1002.Pc
1003when traversing symbolic links.
1004See
1005.Xr symlink 7
1006for a list of supported strings.
1007Note that this option only controls the enabling of this feature
1008by the kernel at boot-up.
1009This feature can still be manipulated with the
1010.Xr sysctl 8
1011command regardless of the setting of this option.
1012.It Cd options NFSSERVER
1013Include the server side of the
1014.Em NFS
1015(Network File System) remote file sharing protocol.
1016Although the bulk of the code implementing
1017.Em NFS
1018is kernel based, several user level daemons are needed for it to
1019work.
1020See
1021.Xr mountd 8
1022and
1023.Xr nfsd 8
1024for details.
1025.It Cd options NVNODE=integer
1026This option sets the size of the cache used by the name-to-inode translation
1027routines, (a.k.a. the
1028.Fn namei
1029cache, though called by many other names in the kernel source).
1030By default, this cache has
1031.Dv ( NPROC + NTEXT + 100 )
1032entries (NPROC set as 20 + 16 * MAXUSERS and NTEXT as 80 + NPROC / 8).
1033A reasonable way to derive a value of
1034.Dv NVNODE ,
1035should you notice a large number of namei cache misses with a tool such as
1036.Xr systat 1 ,
1037is to examine your system's current computed value with
1038.Xr sysctl 8 ,
1039(which calls this parameter "kern.maxvnodes") and to increase this value
1040until either the namei cache hit rate improves or it is determined that
1041your system does not benefit substantially from an increase in the size of
1042the namei cache.
1043.It Cd options NAMECACHE_ENTER_REVERSE
1044Causes the namei cache to always enter a reverse mapping (vnode -> name)
1045as well as a normal one.
1046Normally, this is already done for directory vnodes, to speed up the getcwd
1047operation.
1048This option will cause longer hash chains in the reverse cache, and thus
1049slow down getcwd somewhat.
1050However, it does make vnode -> path translations possible in some cases.
1051For now, only useful if strict
1052.Pa /proc/#/maps
1053emulation for Linux binaries is required.
1054.El
1055.Ss Options for FFS/UFS File Systems
1056.Bl -ohang
1057.It Cd options APPLE_UFS
1058Enable support for UFS file systems created on Mac OS X.
1059.It Cd options FFS_EI
1060Enable
1061.Dq Endian-Independent
1062FFS support.
1063This allows a system to mount an FFS file system created for another
1064architecture, at a small performance cost for all FFS file systems.
1065See also
1066.Xr newfs 8 ,
1067.Xr fsck_ffs 8 ,
1068.Xr dumpfs 8
1069for file system byte order status and manipulation.
1070.It Cd options FFS_NO_SNAPSHOT
1071Disable support for the creation of file system internal snapshot
1072of FFS file systems.
1073Maybe useful for install media kernels, small memory systems and
1074embedded systems which don't require the snapshot support.
1075.It Cd options QUOTA
1076Enables kernel support for traditional quotas in FFS.
1077Traditional quotas store the quota information in external files and
1078require
1079.Xr quotacheck 8
1080and
1081.Xr quotaon 8
1082at boot time.
1083Traditional quotas are limited to 32-bit sizes and are at this point
1084considered a legacy feature.
1085.It Cd options QUOTA2
1086Enables kernel support for in-volume quotas in FFS.
1087The quota information is file system metadata maintained by
1088.Xr fsck 8
1089and/or WAPBL journaling.
1090MFS volumes can also use
1091.Dv QUOTA2
1092quotas; see
1093.Xr mount_mfs 8
1094for more information.
1095.It Cd options UFS_DIRHASH
1096Increase lookup performance by maintaining in-core hash tables
1097for large directories.
1098.It Cd options UFS_EXTATTR
1099Enable extended attribute support for UFS1 file systems.
1100.It Cd options WAPBL
1101Enable
1102.Dq Write Ahead Physical Block Logging file system journaling .
1103This provides rapid file system consistency checking after a system outage.
1104It also provides better general use performance over regular FFS.
1105See also
1106.Xr wapbl 4 .
1107.El
1108.Ss Options for the LFS File System
1109.Bl -ohang
1110.\" .It Cd options LFS_KERNEL_RFW
1111.\" There is no documentation for this. (XXX)
1112.\" .It Cd options LFS_QUOTA
1113.\" Enable traditional quota support for LFS.
1114.\" (It is not clear if this works.)
1115.\" .It Cd options LFS_QUOTA2
1116.\" Enable modernized 64-bit quota support for LFS.
1117.\" (This does not work yet.)
1118.It Cd options LFS_EI
1119Enable
1120.Dq Endian-Independent
1121LFS support.
1122This allows (at a small performance cost) mounting an LFS file system
1123created for another architecture.
1124.\" .It Cd options LFS_EXTATTR
1125.\" Enable extended attribute support for LFS.
1126.\" (It is not clear if this works.)
1127.It Cd options LFS_DIRHASH
1128Increase lookup performance by maintaining in-core hash tables
1129for large directories.
1130.El
1131.Ss Options for the ext2fs File System
1132.Bl -ohang
1133.It Cd options EXT2FS_SYSTEM_FLAGS
1134This option changes the behavior of the APPEND and IMMUTABLE flags
1135for a file on an
1136.Em ext2
1137file system.
1138Without this option, the superuser or owner of the file can
1139set and clear them.
1140With this option, only the superuser can set them, and
1141they can't be cleared if the securelevel is greater than 0.
1142See also
1143.Xr chflags 1
1144and
1145.Xr secmodel_securelevel 9 .
1146.El
1147.Ss Options for the NFS File System
1148.Bl -ohang
1149.It Cd options NFS_BOOT_BOOTP
1150Enable use of the BOOTP protocol (RFCs 951 and 1048) to get configuration
1151information if NFS is used to mount the root file system.
1152See
1153.Xr diskless 8
1154for details.
1155.It Cd options NFS_BOOT_BOOTSTATIC
1156Enable use of static values defined as
1157.Dq NFS_BOOTSTATIC_MYIP ,
1158.Dq NFS_BOOTSTATIC_GWIP ,
1159.Dq NFS_BOOTSTATIC_SERVADDR ,
1160and
1161.Dq NFS_BOOTSTATIC_SERVER
1162in kernel options to get configuration information
1163if NFS is used to mount the root file system.
1164.It Cd options NFS_BOOT_DHCP
1165Same as
1166.Dq NFS_BOOT_BOOTP ,
1167but use the DHCP extensions to the
1168BOOTP protocol (RFC 1541).
1169.It Cd options NFS_BOOT_BOOTP_REQFILE
1170Specifies the string sent in the bp_file field of the BOOTP/DHCP
1171request packet.
1172.It Cd options NFS_BOOT_BOOTPARAM
1173Enable use of the BOOTPARAM protocol, consisting of RARP and
1174BOOTPARAM RPC, to get configuration information if NFS
1175is used to mount the root file system.
1176See
1177.Xr diskless 8
1178for details.
1179.It Cd options NFS_BOOT_RWSIZE=value
1180Set the initial NFS read and write sizes for diskless-boot requests.
1181The normal default is 8Kbytes.
1182This option provides a way to lower the value (e.g., to 1024 bytes)
1183as a workaround for buggy network interface cards or boot PROMs.
1184Once booted, the read and write request sizes can be increased by
1185remounting the file system.
1186See
1187.Xr mount_nfs 8
1188for details.
1189.It Cd options NFS_V2_ONLY
1190Reduce the size of the NFS client code by omitting code that's only required
1191for NFSv3 and NQNFS support, leaving only that code required to use NFSv2
1192servers.
1193.It Cd options NFS_BOOT_TCP
1194Use NFS over TCP instead of the default UDP, for mounting root.
1195.El
1196.Ss Buffer queue strategy options
1197The following options enable alternative buffer queue strategies.
1198.Bl -ohang
1199.It Cd options BUFQ_READPRIO
1200Enable alternate buffer queue strategy for disk I/O.
1201In the default strategy, outstanding disk requests are ordered by
1202sector number and sent to the disk, regardless of whether the
1203operation is a read or write; this option gives priority to issuing
1204read requests over write requests.
1205Although requests may therefore be issued out of sector-order, causing
1206more seeks and thus lower overall throughput, interactive system
1207responsiveness under heavy disk I/O load may be improved, as processes
1208blocking on disk reads are serviced sooner (file writes typically
1209don't cause applications to block).
1210The performance effect varies greatly depending on the hardware, drive
1211firmware, file system configuration, workload, and desired performance
1212trade-off.
1213Systems using drive write-cache (most modern IDE disks, by default)
1214are unlikely to benefit and may well suffer; such disks acknowledge
1215writes very quickly, and optimize them internally according to
1216physical layout.
1217Giving these disks as many requests to work with as possible (the
1218standard strategy) will typically produce the best results, especially
1219if the drive has a large cache; the drive will silently complete
1220writes from cache as it seeks for reads.
1221Disks that support a large number of concurrent tagged requests (SCSI
1222disks and many hardware RAID controllers) expose this internal
1223scheduling with tagged responses, and don't block for reads; such
1224disks may not see a noticeable difference with either strategy.
1225However, if IDE disks are run with write-cache disabled for safety,
1226writes are not acknowledged until actually completed, and only one
1227request can be outstanding; a large number of small writes in one
1228locality can keep the disk busy, starving reads elsewhere on the disk.
1229Such systems are likely to see the most benefit from this option.
1230Finally, the performance interaction of this option with ffs soft
1231dependencies can be subtle, as that mechanism can drastically alter
1232the workload for file system metadata writes.
1233.It Cd options BUFQ_PRIOCSCAN
1234Enable another buffer queue strategy for disk I/O, per-priority cyclical scan.
1235.It Cd options NEW_BUFQ_STRATEGY
1236Synonym of
1237.Em BUFQ_READPRIO .
1238.El
1239.Ss Miscellaneous Options
1240.Bl -ohang
1241.It Cd options CPU_UCODE
1242Support cpu microcode loading via
1243.Xr cpuctl 8 .
1244.It Cd options MEMORY_DISK_DYNAMIC
1245This option makes the
1246.Xr md 4
1247RAM disk size dynamically sized.
1248It is incompatible with
1249.Xr mdsetimage 8 .
1250.It Cd options MEMORY_DISK_HOOKS
1251This option allows for some machine dependent functions to be called when
1252the
1253.Xr md 4
1254RAM disk driver is configured.
1255This can result in automatically loading a RAM disk from floppy on
1256open (among other things).
1257.It Cd options MEMORY_DISK_IS_ROOT
1258Forces the
1259.Xr md 4
1260RAM disk to be the root device.
1261This can only be overridden when
1262the kernel is booted in the 'ask-for-root' mode.
1263.It Cd options MEMORY_DISK_ROOT_SIZE=integer
1264Allocates the given number of 512 byte blocks as memory for the
1265.Xr md 4
1266RAM disk, to be populated with
1267.Xr mdsetimage 8 .
1268.It Cd options MEMORY_DISK_SERVER=0
1269Do not include the interface to a userland memory disk server process.
1270Per default, this option is set to 1, including the support code.
1271Useful for install media kernels.
1272.It Cd options MEMORY_DISK_RBFLAGS=value
1273This option sets the
1274.Xr reboot 2
1275flags used when booting with a memory disk as root file system.
1276Possible values include
1277.Dv RB_AUTOBOOT
1278(boot in the usual fashion - default value), and
1279.Dv RB_SINGLE
1280(boot in single-user mode).
1281.It Cd options MODULAR
1282Enables the framework for kernel modules (see
1283.Xr module 7 ) .
1284.It Cd options MODULAR_DEFAULT_AUTOLOAD
1285Enables the autoloading of kernel modules by default.
1286This sets the default value of the
1287.Em kern.module.autoload
1288.Xr sysctl 3
1289variable which may be changed at run time.
1290.It Cd options VND_COMPRESSION
1291Enables the
1292.Xr vnd 4
1293driver to also handle compressed images.
1294See
1295.Xr vndcompress 1 ,
1296.Xr vnd 4
1297and
1298.Xr vnconfig 8
1299for more information.
1300.It Cd options SPLDEBUG
1301Help the kernel programmer find bugs related to the interrupt priority
1302level.
1303When
1304.Fn spllower
1305or
1306.Fn splraise
1307changes the current CPU's interrupt priority level to or from
1308.Dv IPL_HIGH ,
1309record a backtrace.
1310Read
1311.Xr i386/return_address 9
1312for caveats about collecting backtraces.
1313This feature is experimental, and it is only available on i386.
1314See
1315.Pa sys/kern/subr_spldebug.c .
1316.It Cd options TFTPROOT
1317Download the root memory disk through TFTP at root mount time.
1318This enables the use of a root RAM disk without requiring it to be
1319embedded in the kernel using
1320.Xr mdsetimage 8 .
1321The RAM disk name is obtained using DHCP's filename parameter.
1322This option requires
1323.Em MEMORY_DISK_HOOKS
1324and
1325.Em MEMORY_DISK_DYNAMIC .
1326It is incompatible with
1327.Em MEMORY_DISK_ROOT_SIZE .
1328.It Cd options HZ=integer
1329On ports that support it, set the system clock frequency (see
1330.Xr hz 9 )
1331to the supplied value.
1332Handle with care.
1333.It Cd options NTP
1334Turns on in-kernel precision timekeeping support used by software
1335implementing
1336.Em NTP
1337(Network Time Protocol, RFC 1305).
1338The
1339.Em NTP
1340option adds an in-kernel Phase-Locked Loop (PLL) for normal
1341.Em NTP
1342operation, and a Frequency-Locked Loop (FLL) for intermittently-connected
1343operation.
1344.Xr ntpd 8
1345will employ a user-level PLL when kernel support is unavailable,
1346but the in-kernel version has lower latency and more precision, and
1347so typically keeps much better time.
1348.Pp
1349The interface to the kernel
1350.Em NTP
1351support is provided by the
1352.Xr ntp_adjtime 2
1353and
1354.Xr ntp_gettime 2
1355system calls, which are intended for use by
1356.Xr ntpd 8
1357and are enabled by the option.
1358On systems with sub-microsecond resolution timers, or where (HZ/100000)
1359is not an integer, the
1360.Em NTP
1361option also enables extended-precision arithmetic to keep track of
1362fractional clock ticks at NTP time-format precision.
1363.It Cd options PPS_SYNC
1364This option enables a kernel serial line discipline for receiving time
1365phase signals from an external reference clock such as a radio clock.
1366.Po
1367The
1368.Em NTP
1369option (which see) must be on if the
1370.Em PPS_SYNC
1371option is used
1372.Pc .
1373Some reference clocks generate a Pulse Per Second (PPS) signal in
1374phase with their time source.
1375The
1376.Em PPS
1377line discipline receives this signal on either the data leads
1378or the DCD control lead of a serial port.
1379.Pp
1380.Em NTP
1381uses the PPS signal to discipline the local clock oscillator to a high
1382degree of precision (typically less than 50 microseconds in time and
13830.1 ppm in accuracy).
1384.Em PPS
1385can also generate a serial output pulse when the system receives a PPS
1386interrupt.
1387This can be used to measure the system interrupt latency and thus calibrate
1388.Em NTP
1389to account for it.
1390Using
1391.Em PPS
1392usually requires a gadget box
1393to convert from TTL to RS-232 signal levels.
1394The gadget box and PPS are described in more detail in the HTML documentation
1395for
1396.Xr ntpd 8
1397in
1398.Pa /usr/share/doc/html/ntp .
1399.Pp
1400.Nx
1401currently supports this option in
1402.Xr com 4
1403and
1404.Xr zsc 4 .
1405.It Cd options SETUIDSCRIPTS
1406Allows scripts with the setuid bit set to execute as the effective
1407user rather than the real user, just like binary executables.
1408.Pp
1409.Em NOTE :
1410Using this option will also enable
1411.Em options FDSCRIPTS
1412.It Cd options FDSCRIPTS
1413Allows execution of scripts with the execute bit set, but not the
1414read bit, by opening the file and passing the file descriptor to
1415the shell, rather than the filename.
1416.Pp
1417.Em NOTE :
1418Execute only (non-readable) scripts will have
1419.Va argv[0]
1420set to
1421.Pa /dev/fd/* .
1422What this option allows as far as security is
1423concerned, is the ability to safely ensure that the correct script
1424is run by the interpreter, as it is passed as an already open file.
1425.It Cd options RTC_OFFSET=integer
1426The kernel (and typically the hardware battery backed-up clock on
1427those machines that have one) keeps time in
1428.Em UTC
1429(Universal Coordinated Time, once known as
1430.Em GMT ,
1431or Greenwich Mean Time)
1432and not in the time of the local time zone.
1433The
1434.Em RTC_OFFSET
1435option is used on some ports (such as the i386) to tell the kernel
1436that the hardware clock is offset from
1437.Em UTC
1438by the specified number of minutes.
1439This is typically used when a machine boots several operating
1440systems and one of them wants the hardware clock to run in the
1441local time zone and not in
1442.Em UTC ,
1443e.g.
1444.Em RTC_OFFSET=300
1445means
1446the hardware clock is set to US Eastern Time (300 minutes behind
1447.Em UTC ) ,
1448and not
1449.Em UTC .
1450(Note:
1451.Em RTC_OFFSET
1452is used to initialize a kernel variable named
1453.Va rtc_offset
1454which is the source actually used to determine the clock offset, and
1455which may be accessed via the kern.rtc_offset sysctl variable.
1456See
1457.Xr sysctl 8
1458and
1459.Xr sysctl 3
1460for details.
1461Since the kernel clock is initialized from the hardware clock very
1462early in the boot process, it is not possible to meaningfully change
1463.Va rtc_offset
1464in system initialization scripts.
1465Changing this value currently may only be done at kernel compile
1466time or by patching the kernel and rebooting).
1467.Pp
1468.Em NOTE :
1469Unfortunately, in many cases where the hardware clock
1470is kept in local time, it is adjusted for Daylight Savings
1471Time; this means that attempting to use
1472.Em RTC_OFFSET
1473to let
1474.Nx
1475coexist with such an operating system, like Windows,
1476would necessitate changing
1477.Em RTC_OFFSET
1478twice a year.
1479As such, this solution is imperfect.
1480.It Cd options MAXUPRC=integer
1481Sets the soft
1482.Dv RLIMIT_NPROC
1483resource limit, which specifies the maximum number of simultaneous
1484processes a user is permitted to run, for process 0;
1485this value is inherited by its child processes.
1486It defaults to
1487.Em CHILD_MAX ,
1488which is currently defined to be 160.
1489Setting
1490.Em MAXUPRC
1491to a value less than
1492.Em CHILD_MAX
1493is not permitted, as this would result in a violation of the semantics of
1494.St -p1003.1-90 .
1495.It Cd options NOFILE=integer
1496Sets the soft
1497.Dv RLIMIT_NOFILE
1498resource limit, which specifies the maximum number of open
1499file descriptors for each process;
1500this value is inherited by its child processes.
1501It defaults to
1502.Em OPEN_MAX ,
1503which is currently defined to be 128.
1504.It Cd options MAXFILES=integer
1505Sets the default value of the
1506.Em kern.maxfiles
1507sysctl variable, which indicates the maximum number of files that may
1508be open in the system.
1509.It Cd options DEFCORENAME=string
1510Sets the default value of the
1511.Em kern.defcorename
1512sysctl variable, otherwise it is set to
1513.Nm %n.core .
1514See
1515.Xr sysctl 8
1516and
1517.Xr sysctl 3
1518for details.
1519.It Cd options RASOPS_CLIPPING
1520Enables clipping within the
1521.Nm rasops
1522raster-console output system.
1523.Em NOTE :
1524only available on architectures that use
1525.Nm rasops
1526for console output.
1527.It Cd options RASOPS_SMALL
1528Removes optimized character writing code from the
1529.Nm rasops
1530raster-console output system.
1531.Em NOTE :
1532only available on architectures that use
1533.Nm rasops
1534for console output.
1535.It Cd options INCLUDE_CONFIG_FILE
1536Embeds the kernel config file used to define the kernel in the kernel
1537binary itself.
1538The embedded data also includes any files directly included by the config
1539file itself, e.g.
1540.Pa GENERIC.local
1541or
1542.Pa std.$MACHINE .
1543The embedded config file can be extracted from the resulting kernel with
1544.Xr config 1
1545.Fl x ,
1546or by the following command:
1547.Bd -literal -offset indent
1548strings netbsd | sed -n 's/^_CFG_//p' | unvis
1549.Ed
1550.It Cd options INCLUDE_JUST_CONFIG
1551Similar to the above option, but includes just the actual config file,
1552not any included files.
1553.It Cd options PIPE_SOCKETPAIR
1554Use slower, but smaller socketpair(2)-based pipe implementation instead
1555of default faster, but bigger one.
1556Primarily useful for installation kernels.
1557.It Cd options USERCONF
1558Compiles in the in-kernel device configuration manager.
1559See
1560.Xr userconf 4
1561for details.
1562.It Cd options SCDEBUG_DEFAULT
1563Used with the
1564.Cd options SYSCALL_DEBUG
1565described below to choose which types of events are displayed.
1566.Pp
1567.Bl -tag -width "SCDEBUG_KERNHIST" -compact -offset indent
1568.It Dv SCDEBUG_CALLS
1569Show system call entry points.
1570.It Dv SCDEBUG_RETURNS
1571Show system call exit points.
1572.It Dv SCDEBUG_ALL
1573Show all system call requests, including unimplemented calls.
1574.It Dv SCDEBUG_SHOWARGS
1575Show the arguments provided.
1576.It Dv SCDEBUG_KERNHIST
1577Store a restricted form of the system call debug in a kernel history
1578instead of printing it to the console.
1579This option relies upon
1580.Cd options KERNHIST .
1581.El
1582.Pp
1583The default value is
1584.Dv (SCDEBUG_CALLS|SCDEBUG_RETURNS|SCDEBUG_SHOWARGS) .
1585.It Cd options SYSCALL_DEBUG
1586Useful for debugging system call issues, usually in early single user bringup.
1587By default, writes entries to the system console for most system call events.
1588Can be configured with the
1589.Cd options SCDEBUG_DEFAULT
1590option to to use the
1591.Cd options KERNHIST
1592facility instead.
1593.It Cd options SYSCALL_STATS
1594Count the number of times each system call number is called.
1595The values can be read through the sysctl interface and displayed using
1596.Xr systat 1 .
1597.Em NOTE :
1598not yet available on all architectures.
1599.It Cd options SYSCALL_TIMES
1600Count the time spent (using
1601.Fn cpu_counter32 )
1602in each system call.
1603.Em NOTE :
1604Using this option will also enable
1605.Cd options SYSCALL_STATS .
1606.It Cd options SYSCALL_TIMES_HASCOUNTER
1607Force use of
1608.Fn cpu_counter32
1609even if
1610.Fn cpu_hascounter
1611reports false.
1612Useful for systems where the cycle counter doesn't run at a constant rate
1613(e.g. Soekris boxes).
1614.It Cd options XSERVER_DDB
1615A supplement to XSERVER that adds support for entering
1616.Xr ddb 4
1617while in X11.
1618.It Cd options FILEASSOC
1619Support for
1620.Xr fileassoc 9 .
1621Required for
1622.Cd options PAX_SEGVGUARD
1623and
1624.Cd pseudo-device veriexec .
1625.It Cd options FILEASSOC_NHOOKS=integer
1626Number of storage slots per file for
1627.Xr fileassoc 9 .
1628Default is 4.
1629.El
1630.Ss Networking Options
1631.Bl -ohang
1632.It Cd options GATEWAY
1633Enables
1634.Em IPFORWARDING
1635and (on most ports) increases the size of
1636.Em NMBCLUSTERS .
1637In general,
1638.Em GATEWAY
1639is used to indicate that a system should act as a router, and
1640.Em IPFORWARDING
1641is not invoked directly.
1642(Note that
1643.Em GATEWAY
1644has no impact on protocols other than IP).
1645.Em GATEWAY
1646option also compiles IPv4 and IPv6 fast forwarding code into the kernel.
1647.It Cd options IPFORWARDING=value
1648If
1649.Em value
1650is 1 this enables IP routing behavior.
1651If
1652.Em value
1653is 0 (the default), it disables it.
1654The
1655.Em GATEWAY
1656option sets this to 1 automatically.
1657With this option enabled, the machine will forward IP datagrams destined
1658for other machines between its interfaces.
1659Note that even without this option, the kernel will
1660still forward some packets (such as source routed packets) \(em removing
1661.Em GATEWAY
1662and
1663.Em IPFORWARDING
1664is insufficient to stop all routing through a bastion host on a
1665firewall \(em source routing is controlled independently.
1666Note that IP forwarding may be turned on and off independently of the
1667setting of the
1668.Em IPFORWARDING
1669option through the use of the
1670.Em net.inet.ip.forwarding
1671sysctl variable.
1672If
1673.Em net.inet.ip.forwarding
1674is 1, IP forwarding is on.
1675See
1676.Xr sysctl 8
1677and
1678.Xr sysctl 3
1679for details.
1680.It Cd options IFA_STATS
1681Tells the kernel to maintain per-address statistics on bytes sent
1682and received over (currently) Internet and AppleTalk addresses.
1683.\"This can be a fairly expensive operation, so you probably want to
1684.\"keep this disabled.
1685The option is not recommended as it degrades system stability.
1686.It Cd options IFQ_MAXLEN=value
1687Increases the allowed size of the network interface packet queues.
1688The default queue size is 50 packets, and you do not normally need
1689to increase it.
1690.It Cd options IPSELSRC
1691Includes support for source-address selection policies.
1692See
1693.Xr in_getifa 9 .
1694.It Cd options MROUTING
1695Includes support for IP multicast routers.
1696You certainly want
1697.Em INET
1698with this.
1699Multicast routing is controlled by the
1700.Xr mrouted 8
1701daemon.
1702See also option
1703.Cd PIM .
1704.It Cd options PIM
1705Includes support for Protocol Independent Multicast (PIM) routing.
1706You need
1707.Em MROUTING
1708and
1709.Em INET
1710with this.
1711Software using this can be found e.g. in
1712.Pa pkgsrc/net/xorp .
1713.It Cd options INET
1714Includes support for the TCP/IP protocol stack.
1715You almost certainly want this.
1716See
1717.Xr inet 4
1718for details.
1719.It Cd options INET6
1720Includes support for the IPv6 protocol stack.
1721See
1722.Xr inet6 4
1723for details.
1724Unlike
1725.Em INET ,
1726.Em INET6
1727enables multicast routing code as well.
1728This option requires
1729.Em INET
1730at this moment, but it should not.
1731.It Cd options ND6_DEBUG
1732The option sets the default value of net.inet6.icmp6.nd6_debug to 1,
1733for debugging IPv6 neighbor discovery protocol handling.
1734See
1735.Xr sysctl 3
1736for details.
1737.It Cd options IPSEC
1738Includes support for the IPsec protocol, using the implementation derived from
1739.Ox ,
1740relying on
1741.Xr opencrypto 9
1742to carry out cryptographic operations.
1743See
1744.Xr ipsec 4
1745for details.
1746.It Cd options IPSEC_DEBUG
1747Enables debugging code in IPsec stack.
1748See
1749.Xr ipsec 4
1750for details.
1751The
1752.Cd IPSEC
1753option includes support for IPsec Network Address Translator traversal
1754(NAT-T), as described in RFCs 3947 and 3948.
1755This feature might be patent-encumbered in some countries.
1756.It Cd options ALTQ
1757Enabled ALTQ (Alternate Queueing).
1758For simple rate-limiting, use
1759.Xr tbrconfig 8
1760to set up the interface transmission rate.
1761To use queueing disciplines, their appropriate kernel options should also
1762be defined (documented below).
1763Queueing disciplines are managed by
1764.Xr altqd 8 .
1765See
1766.Xr altq 9
1767for details.
1768.It Cd options ALTQ_HFSC
1769Include support for ALTQ-implemented HFSC (Hierarchical Fair Service Curve)
1770module.
1771HFSC supports both link-sharing and guaranteed real-time services.
1772HFSC employs a service curve based QoS model, and its unique feature
1773is an ability to decouple delay and bandwidth allocation.
1774Requires
1775.Em ALTQ_RED
1776to use the RED queueing discipline on HFSC classes, or
1777.Em ALTQ_RIO
1778to use the RIO queueing discipline on HFSC classes.
1779This option assumes
1780.Em ALTQ .
1781.It Cd options ALTQ_PRIQ
1782Include support for ALTQ-implemented PRIQ (Priority Queueing).
1783PRIQ implements a simple priority-based queueing discipline.
1784A higher priority class is always served first.
1785Requires
1786.Em ALTQ_RED
1787to use the RED queueing discipline on HFSC classes, or
1788.Em ALTQ_RIO
1789to use the RIO queueing discipline on HFSC classes.
1790This option assumes
1791.Em ALTQ .
1792.It Cd options ALTQ_WFQ
1793Include support for ALTQ-implemented WFQ (Weighted Fair Queueing).
1794WFQ implements a weighted-round robin scheduler for a set of queues.
1795A weight can be assigned to each queue to give a different proportion
1796of the link capacity.
1797A hash function is used to map a flow to one of a set of queues.
1798This option assumes
1799.Em ALTQ .
1800.It Cd options ALTQ_FIFOQ
1801Include support for ALTQ-implemented FIFO queueing.
1802FIFOQ is a simple drop-tail FIFO (First In, First Out) queueing discipline.
1803This option assumes
1804.Em ALTQ .
1805.It Cd options ALTQ_RIO
1806Include support for ALTQ-implemented RIO (RED with In/Out).
1807The original RIO has 2 sets of RED parameters; one for in-profile
1808packets and the other for out-of-profile packets.
1809At the ingress of the network, profile meters tag packets as IN or
1810OUT based on contracted profiles for customers.
1811Inside the network, IN packets receive preferential treatment by
1812the RIO dropper.
1813ALTQ/RIO has 3 drop precedence levels defined for the Assured Forwarding
1814PHB of DiffServ (RFC 2597).
1815This option assumes
1816.Em ALTQ .
1817.It Cd options ALTQ_BLUE
1818Include support for ALTQ-implemented Blue buffer management.
1819Blue is another active buffer management mechanism.
1820This option assumes
1821.Em ALTQ .
1822.It Cd options ALTQ_FLOWVALVE
1823Include support for ALTQ-implemented Flowvalve.
1824Flowvalve is a simple implementation of a RED penalty box that identifies
1825and punishes misbehaving flows.
1826This option requires
1827.Em ALTQ_RED
1828and assumes
1829.Em ALTQ .
1830.It Cd options ALTQ_CDNR
1831Include support for ALTQ-implemented CDNR (diffserv traffic conditioner)
1832packet marking/manipulation.
1833Traffic conditioners are components to meter, mark, or drop incoming
1834packets according to some rules.
1835As opposed to queueing disciplines, traffic conditioners handle incoming
1836packets at an input interface.
1837This option assumes
1838.Em ALTQ .
1839.It Cd options ALTQ_NOPCC
1840Disables use of processor cycle counter to measure time in ALTQ.
1841This option should be defined for a non-Pentium i386 CPU which does not
1842have TSC, SMP (per-CPU counters are not in sync), or power management
1843which affects processor cycle counter.
1844This option assumes
1845.Em ALTQ .
1846.It Cd options ALTQ_IPSEC
1847Include support for IPsec in IPv4 ALTQ.
1848This option assumes
1849.Em ALTQ .
1850.It Cd options ALTQ_JOBS
1851Include support for ALTQ-implemented JoBS (Joint Buffer Management
1852and Scheduling).
1853This option assumes
1854.Em ALTQ .
1855.It Cd options ALTQ_AFMAP
1856Include support for an undocumented ALTQ feature that is used to map an IP
1857flow to an ATM VC (Virtual Circuit).
1858This option assumes
1859.Em ALTQ .
1860.It Cd options ALTQ_LOCALQ
1861Include support for ALTQ-implemented local queues.
1862Its practical use is undefined.
1863Assumes
1864.Em ALTQ .
1865.It Cd options SUBNETSARELOCAL
1866Sets default value for net.inet.ip.subnetsarelocal variable, which
1867controls whether non-directly-connected subnets of connected networks
1868are considered "local" for purposes of choosing the MSS for a TCP
1869connection.
1870This is mostly present for historic reasons and completely irrelevant if
1871you enable Path MTU discovery.
1872.It Cd options HOSTZEROBROADCAST
1873Sets default value for net.inet.ip.hostzerobroadcast variable, which
1874controls whether the zeroth host address of each connected subnet is
1875also considered a broadcast address.
1876Default value is "1", for compatibility with old systems; if this is
1877set to zero on all hosts on a subnet, you should be able to fit an extra
1878host per subnet on the
1879".0" address.
1880.It Cd options MCLSHIFT=value
1881This option is the base-2 logarithm of the size of mbuf clusters.
1882The
1883.Bx
1884networking stack keeps network packets in a linked
1885list, or chain, of kernel buffer objects called mbufs.
1886The system provides larger mbuf clusters as an optimization for
1887large packets, instead of using long chains for large packets.
1888The mbuf cluster size,
1889or
1890.Em MCLBYTES ,
1891must be a power of two, and is computed as two raised to the power
1892.Em MCLSHIFT .
1893On systems with Ethernet network adapters,
1894.Em MCLSHIFT
1895is often set to 11, giving 2048-byte mbuf clusters, large enough to
1896hold a 1500-byte Ethernet frame in a single cluster.
1897Systems with network interfaces supporting larger frame sizes like
1898ATM, FDDI, or HIPPI may perform better with
1899.Em MCLSHIFT
1900set to 12 or 13, giving mbuf cluster sizes of 4096 and 8192 bytes,
1901respectively.
1902.It Cd options NETATALK
1903Include support for the AppleTalk protocol stack.
1904The kernel provides provision for the
1905.Em Datagram Delivery Protocol
1906(DDP), providing SOCK_DGRAM support and AppleTalk routing.
1907This stack is used by the
1908.Em NETATALK
1909package, which adds support for AppleTalk server services via user
1910libraries and applications.
1911.It Cd options BLUETOOTH
1912Include support for the Bluetooth protocol stack.
1913See
1914.Xr bluetooth 4
1915for details.
1916.It Cd options IPNOPRIVPORTS
1917Normally, only root can bind a socket descriptor to a so-called
1918.Dq privileged
1919TCP port, that is, a port number in the range 0-1023.
1920This option eliminates those checks from the kernel.
1921This can be useful if there is a desire to allow daemons without
1922privileges to bind those ports, e.g., on firewalls.
1923The security tradeoffs in doing this are subtle.
1924This option should only be used by experts.
1925.It Cd options TCP_DEBUG
1926Record the last
1927.Em TCP_NDEBUG
1928TCP packets with SO_DEBUG set, and decode to the console if
1929.Em tcpconsdebug
1930is set.
1931.It Cd options TCP_NDEBUG
1932Number of packets to record for
1933.Em TCP_DEBUG .
1934Defaults to 100.
1935.It Cd options TCP_SENDSPACE=value
1936.It Cd options TCP_RECVSPACE=value
1937These options set the max TCP window size to other sizes than the default.
1938The TCP window sizes can be altered via
1939.Xr sysctl 8
1940as well.
1941.It Cd options TCP_INIT_WIN=value
1942This option sets the initial TCP window size for non-local connections,
1943which is used when the transmission starts.
1944The default size is 1, but if the machine should act more aggressively,
1945the initial size can be set to some other value.
1946The initial TCP window size can be set via
1947.Xr sysctl 8
1948as well.
1949.It Cd options TCP_SIGNATURE
1950Enable MD5 TCP signatures (RFC 2385) to protect BGP sessions.
1951.It Cd options IPFILTER_LOG
1952This option, in conjunction with
1953.Em pseudo-device ipfilter ,
1954enables logging of IP packets using IP-Filter.
1955.It Cd options IPFILTER_LOOKUP
1956This option enables the
1957IP-Filter
1958.Xr ippool 8
1959functionality to be enabled.
1960.It Cd options IPFILTER_COMPAT
1961This option enables older IP-Filter binaries to work.
1962.It Cd options IPFILTER_DEFAULT_BLOCK
1963This option sets the default policy of IP-Filter.
1964If it is set, IP-Filter will block packets by default.
1965.It Cd options MBUFTRACE
1966This option can help track down mbuf leaks.
1967When enabled, mbufs are tagged with the devices and protocols using them.
1968This can significantly decrease network performance, particularly
1969on MP systems.
1970This additional information can be viewed with
1971.Xr netstat 1 :
1972.Dl Ic netstat Fl mssv
1973Not all devices or protocols support this option.
1974.El
1975.Ss Sysctl Related Options
1976.Bl -ohang
1977.It Cd options SYSCTL_DISALLOW_CREATE
1978Disallows the creation or deletion of nodes from the sysctl tree, as
1979well as the assigning of descriptions to nodes that lack them, by any
1980process.
1981These operations are still available to kernel sub-systems, including
1982loadable kernel modules.
1983.It Cd options SYSCTL_DISALLOW_KWRITE
1984Prevents processes from adding nodes to the sysctl tree that make
1985existing kernel memory areas writable.
1986Sections of kernel memory can still be read and new nodes that own
1987their own data may still be writable.
1988.It Cd options SYSCTL_DEBUG_SETUP
1989Causes the SYSCTL_SETUP routines to print a brief message when they
1990are invoked.
1991This is merely meant as an aid in determining the order in which
1992sections of the tree are created.
1993.It Cd options SYSCTL_DEBUG_CREATE
1994Prints a message each time
1995.Fn sysctl_create ,
1996the function that adds nodes to the tree, is called.
1997.It Cd options SYSCTL_INCLUDE_DESCR
1998Causes the kernel to include short, human readable descriptions for
1999nodes in the sysctl tree.
2000The descriptions can be retrieved programmatically (see
2001.Xr sysctl 3 ) ,
2002or by the sysctl binary itself (see
2003.Xr sysctl 8 ) .
2004The descriptions are meant to give an indication of the purpose and/or
2005effects of a given node's value, not replace the documentation for the
2006given subsystem as a whole.
2007.El
2008.Ss System V IPC Options
2009.Bl -ohang
2010.It Cd options SYSVMSG
2011Includes support for
2012.At V
2013style message queues.
2014See
2015.Xr msgctl 2 ,
2016.Xr msgget 2 ,
2017.Xr msgrcv 2 ,
2018.Xr msgsnd 2 .
2019.It Cd options SYSVSEM
2020Includes support for
2021.At V
2022style semaphores.
2023See
2024.Xr semctl 2 ,
2025.Xr semget 2 ,
2026.Xr semop 2 .
2027.It Cd options SEMMNI=value
2028Sets the number of
2029.At V
2030style semaphore identifiers.
2031The GENERIC config file for your port will have the default.
2032.It Cd options SEMMNS=value
2033Sets the number of
2034.At V
2035style semaphores in the system.
2036The GENERIC config file for your port will have the default.
2037.It Cd options SEMUME=value
2038Sets the maximum number of undo entries per process for
2039.At V
2040style semaphores.
2041The GENERIC config file for your port will have the default.
2042.It Cd options SEMMNU=value
2043Sets the number of undo structures in the system for
2044.At V
2045style semaphores.
2046The GENERIC config file for your port will have the default.
2047.It Cd options SYSVSHM
2048Includes support for
2049.At V
2050style shared memory.
2051See
2052.Xr shmat 2 ,
2053.Xr shmctl 2 ,
2054.Xr shmdt 2 ,
2055.Xr shmget 2 .
2056.It Cd options SHMMAXPGS=value
2057Sets the maximum number of
2058.At V
2059style shared memory pages that are available through the
2060.Xr shmget 2
2061system call.
2062Default value is 1024 on most ports.
2063See
2064.Pa /usr/include/machine/vmparam.h
2065for the default.
2066.El
2067.Ss VM Related Options
2068.Bl -ohang
2069.It Cd options NMBCLUSTERS=value
2070The number of mbuf clusters the kernel supports.
2071Mbuf clusters are MCLBYTES in size (usually 2k).
2072This is used to compute the size of the kernel VM map
2073.Em mb_map ,
2074which maps mbuf clusters.
2075Default on most ports is 1024 (2048 with
2076.Dq options GATEWAY
2077).
2078See
2079.Pa /usr/include/machine/param.h
2080for exact default information.
2081Increase this value if you get
2082.Dq mclpool limit reached
2083messages.
2084.It Cd options NKMEMPAGES=value
2085.It Cd options NKMEMPAGES_MIN=value
2086.It Cd options NKMEMPAGES_MAX=value
2087Size of kernel VM map
2088.Em kmem_map ,
2089in PAGE_SIZE-sized chunks (the VM page size; this value may be read
2090from the
2091.Xr sysctl 8
2092variable
2093.Em hw.pagesize
2094).
2095This VM map is used to map the kernel malloc arena.
2096The kernel attempts to auto-size this map based on the amount of
2097physical memory in the system.
2098Platform-specific code may place bounds on this computed size,
2099which may be viewed with the
2100.Xr sysctl 8
2101variable
2102.Em vm.nkmempages .
2103See
2104.Pa /usr/include/machine/param.h
2105for the default upper and lower bounds.
2106The related options
2107.Sq NKMEMPAGES_MIN
2108and
2109.Sq NKMEMPAGES_MAX
2110allow the bounds to be overridden in the kernel configuration file.
2111These options are provided in the event the computed value is
2112insufficient resulting in an
2113.Dq out of space in kmem_map
2114panic.
2115.It Cd options SB_MAX=value
2116Sets the max size in bytes that a socket buffer is allowed to occupy.
2117The default is 256k, but sometimes it needs to be increased, for example
2118when using large TCP windows.
2119This option can be changed via
2120.Xr sysctl 8
2121as well.
2122.It Cd options SOMAXKVA=value
2123Sets the maximum size of kernel virtual memory that the socket buffers
2124are allowed to use.
2125The default is 16MB, but in situations where for example large TCP
2126windows are used this value must also be increased.
2127This option can be changed via
2128.Xr sysctl 8
2129as well.
2130.It Cd options BUFCACHE=value
2131Size of the buffer cache as a percentage of total available RAM.
2132Ignored if BUFPAGES is also specified.
2133.It Cd options NBUF=value
2134Sets the number of buffer headers available, i.e., the number of
2135open files that may have a buffer cache entry.
2136Each buffer header
2137requires MAXBSIZE (machine dependent, but usually 65536) bytes.
2138The default value is machine dependent, but is usually equal to the
2139value of BUFPAGES.
2140.It Cd options BUFPAGES=value
2141These options set the number of pages available for the buffer cache.
2142Their default value is a machine dependent value, often calculated as
2143between 5% and 10% of total available RAM.
2144.It Cd options MAXTSIZ=bytes
2145Sets the maximum size limit of a process' text segment.
2146See
2147.Pa /usr/include/machine/vmparam.h
2148for the port-specific default.
2149.It Cd options DFLDSIZ=bytes
2150Sets the default size limit of a process' data segment, the value that
2151will be returned as the soft limit for
2152.Dv RLIMIT_DATA
2153(as returned by
2154.Xr getrlimit 2 ) .
2155See
2156.Pa /usr/include/machine/vmparam.h
2157for the port-specific default.
2158.It Cd options MAXDSIZ=bytes
2159Sets the maximum size limit of a process' data segment, the value that
2160will be returned as the hard limit for
2161.Dv RLIMIT_DATA
2162(as returned by
2163.Xr getrlimit 2 ) .
2164See
2165.Pa /usr/include/machine/vmparam.h
2166for the port-specific default.
2167.It Cd options DFLSSIZ=bytes
2168Sets the default size limit of a process' stack segment, the value that
2169will be returned as the soft limit for
2170.Dv RLIMIT_STACK
2171(as returned by
2172.Xr getrlimit 2 ) .
2173See
2174.Pa /usr/include/machine/vmparam.h
2175for the port-specific default.
2176.It Cd options MAXSSIZ=bytes
2177Sets the maximum size limit of a process' stack segment, the value that
2178will be returned as the hard limit for
2179.Dv RLIMIT_STACK
2180(as returned by
2181.Xr getrlimit 2 ) .
2182See
2183.Pa /usr/include/machine/vmparam.h
2184for the port-specific default.
2185.It Cd options DUMP_ON_PANIC=integer
2186Defaults to one.
2187If set to zero, the kernel will not dump to the dump device when
2188it panics, though dumps can still be forced via
2189.Xr ddb 4
2190with the
2191.Dq sync
2192command.
2193Note that this sets the value of the
2194.Em kern.dump_on_panic
2195.Xr sysctl 3
2196variable which may be changed at run time \(em see
2197.Xr sysctl 8
2198for details.
2199.It Cd options VMSWAP
2200Enable paging device/file support.
2201This option is on by default.
2202.It Cd options PDPOLICY_CLOCKPRO
2203Use CLOCK-Pro, an alternative page replace policy.
2204.El
2205.Ss Security Options
2206.Bl -ohang
2207.It Cd options INSECURE
2208Initializes the kernel security level with \-1 instead of 0.
2209This means that the system always starts in secure level \-1 mode, even when
2210running multiuser, unless the securelevel variable is set to value > \-1 in
2211.Pa /etc/rc.conf .
2212In this case the kernel security level will be raised to that value when the
2213.Pa /etc/rc.d/securelevel
2214script is run during system startup.
2215See the manual page for
2216.Xr init 8
2217for details on the implications of this.
2218The kernel secure level may manipulated by the superuser by altering the
2219.Em kern.securelevel
2220.Xr sysctl 3
2221variable (the secure level may only be lowered by a call from process ID 1,
2222i.e.,
2223.Xr init 8 ) .
2224See also
2225.Xr secmodel_securelevel 9 ,
2226.Xr sysctl 8
2227and
2228.Xr sysctl 3 .
2229.It Cd options VERIFIED_EXEC_FP_SHA256
2230Enables support for SHA256 hashes in Veriexec.
2231.It Cd options VERIFIED_EXEC_FP_SHA384
2232Enables support for SHA384 hashes in Veriexec.
2233.It Cd options VERIFIED_EXEC_FP_SHA512
2234Enables support for SHA512 hashes in Veriexec.
2235.It Cd options PAX_MPROTECT=value
2236Enables PaX MPROTECT,
2237.Xr mprotect 2
2238restrictions from the PaX project.
2239.Pp
2240The
2241.Ar value
2242is the default value for the
2243.Em global
2244knob, see
2245.Xr sysctl 3 .
2246If 0, PaX MPROTECT will be enabled only if explicitly set on programs
2247using
2248.Xr paxctl 8 .
2249If 1, PaX MPROTECT will be enabled for all programs.
2250Programs can be exempted using
2251.Xr paxctl 8 .
2252.Pp
2253See
2254.Xr security 7
2255for more details.
2256.It Cd options PAX_SEGVGUARD=value
2257Enables PaX Segvguard.
2258Requires
2259.Cd options FILEASSOC .
2260.Pp
2261The
2262.Ar value
2263is the default value for the
2264.Em global
2265knob, see
2266.Xr sysctl 3 .
2267If 0, PaX Segvguard will be enabled only if explicitly set on programs
2268using
2269.Xr paxctl 8 .
2270If 1, PaX Segvguard will be enabled to all programs, and exemption can
2271be done using
2272.Xr paxctl 8 .
2273.Pp
2274See
2275.Xr security 7
2276for more details.
2277.It Cd options PAX_ASLR=value
2278Enables PaX ASLR.
2279.Pp
2280The
2281.Ar value
2282is the default value for the
2283.Em global
2284knob, see
2285.Xr sysctl 3 .
2286If 0, PaX ASLR will be enabled only if explicitly set on programs
2287using
2288.Xr paxctl 8 .
2289If 1, PaX ASLR will be enabled to all programs, and exemption can
2290be done using
2291.Xr paxctl 8 .
2292.Pp
2293See
2294.Xr security 7
2295for more details.
2296.It Cd options USER_VA0_DISABLE_DEFAULT=value
2297Sets the initial value of the flag which controls whether user programs
2298can map virtual address 0.
2299The flag can be changed at runtime by
2300.Xr sysctl 3 .
2301.It Cd options KASAN
2302Enables Kernel Address Sanitizer.
2303.Em NOTE :
2304not available on all architectures.
2305.It Cd options KASLR
2306Enables Kernel ASLR.
2307This randomizes the location of the kernel image in memory.
2308.Em NOTE :
2309not available on all architectures.
2310.It Cd options SVS
2311Enables Separate Virtual Space.
2312On architectures that are designed to function with a shared address
2313space, this option explicitly isolates the kernel and user spaces.
2314.Em NOTE :
2315not available on all architectures.
2316.El
2317.Ss amiga-specific Options
2318.Bl -ohang
2319.It Cd options BB060STUPIDROM
2320When the bootloader (which passes AmigaOS ROM information) claims
2321we have a 68060 CPU without FPU, go look into the Processor
2322Configuration Register (PCR) to find out.
2323You need this with Amiga ROMs up to (at least) V40.xxx (OS3.1),
2324when you boot via the bootblocks and don't have a DraCo.
2325.It Cd options IOBZCLOCK=frequency
2326The IOBlix boards come with two different serial master clocks: older ones
2327use 24 MHz, newer ones use 22.1184 MHz.
2328The driver normally assumes the latter.
2329If your board uses 24 MHz, you can recompile your kernel with
2330options IOBZCLOCK=24000000
2331or patch the kernel variable iobzclock to the same value.
2332.It Cd options LIMITMEM=value
2333If there, limit the part of the first memory bank used by
2334.Nx
2335to value megabytes.
2336Default is unlimited.
2337.It Cd options P5PPC68KBOARD
2338Add special support for Phase5 mixed 68k+PPC boards.
2339Currently, this only affects rebooting from
2340.Nx
2341and is only needed on 68040+PPC, not on
234268060+PPC; without this, affected machines will hang after
2343.Nx
2344has shut
2345down and will only restart after a keyboard reset or a power cycle.
2346.El
2347.Ss atari-specific Options
2348.Bl -ohang
2349.It Cd options DISKLABEL_AHDI
2350Include support for AHDI (native Atari) disklabels.
2351.It Cd options DISKLABEL_NBDA
2352Include support for
2353.Nx Ns /atari
2354labels.
2355If you don't set this option, it will be set automatically.
2356.Nx Ns /atari
2357will not work without it.
2358.It Cd options FALCON_SCSI
2359Include support for the 5380-SCSI configuration as found on the Falcon.
2360.It Cd options RELOC_KERNEL
2361If set, the kernel will relocate itself to TT-RAM, if possible.
2362This will give you a slightly faster system.
2363.Em Beware
2364that on some TT030 systems,
2365the system will frequently dump with MMU-faults with this option enabled.
2366.It Cd options SERCONSOLE
2367Allow the modem1-port to act as the system-console.
2368A carrier should be active on modem1 during system boot to active
2369the console functionality.
2370.It Cd options TT_SCSI
2371Include support for the 5380-SCSI configuration as found on the TT030
2372and Hades.
2373.El
2374.Ss i386-specific Options
2375.Bl -ohang
2376.It Cd options CPURESET_DELAY=value
2377Specifies the time (in millisecond) to wait before doing a hardware reset
2378in the last phase of a reboot.
2379This gives the user a chance to see error messages from the shutdown
2380operations (like NFS unmounts, buffer cache flush, etc ...).
2381Setting this to 0 will disable the delay.
2382Default is 2 seconds.
2383.It Cd options USER_LDT
2384Include i386-specific system calls for modifying the local descriptor table,
2385used by Windows emulators.
2386.It Cd options PAE
2387Enable PAE (Physical Address Extension) mode.
2388PAE permits up to 36 bits physical addressing (64GB of physical
2389memory), and turns physical addresses to 64 bits entities in the
2390memory management subsystem.
2391Userland virtual address space remains at 32 bits (4GB).
2392PAE mode is required to enable the NX/XD (No-eXecute/eXecute Disable)
2393bit for pages, which allows marking certain ones as not being executable.
2394Any attempt to execute code from such a page will raise an exception.
2395.It Cd options REALBASEMEM=integer
2396Overrides the base memory size passed in from the boot block.
2397(Value given in kilobytes.)
2398Use this option only if the boot block reports the size incorrectly.
2399(Note that some BIOSes put the extended BIOS
2400data area at the top of base memory, and therefore report a smaller
2401base memory size to prevent programs overwriting it.
2402This is correct behavior, and you should not use the
2403.Em REALBASEMEM
2404option to access this memory).
2405.It Cd options SPECTRE_V2_GCC_MITIGATION=1
2406Enable GCC-specific Spectre variant 2 mitigations.
2407For 32-bit kernels this means these options:
2408.Bd -literal -offset indent
2409-mindirect-branch=thunk -mindirect-branch-register
2410.Ed
2411.Pp
2412For 64-bit kernels this means these options:
2413.Bd -literal -offset indent
2414-mindirect-branch=thunk-inline -mindirect-branch-register
2415.Ed
2416.It Cd options REALEXTMEM=integer
2417Overrides the extended memory size passed in from the boot block.
2418(Value given in kilobytes.
2419Extended memory does not include the first megabyte.)
2420Use this option only if the boot block reports the size incorrectly.
2421.It Cd options CYRIX_CACHE_WORKS
2422Relevant only to the Cyrix 486DLC CPU.
2423This option is used to turn on the cache in hold-flush mode.
2424It is not turned on by default because it is known to have problems in
2425certain motherboard implementations.
2426.It Cd options CYRIX_CACHE_REALLY_WORKS
2427Relevant only to the Cyrix 486DLC CPU.
2428This option is used to turn on the cache in write-back mode.
2429It is not turned on by default because it is known to have problems in
2430certain motherboard implementations.
2431In order for this option to take effect, option
2432.Em CYRIX_CACHE_WORKS
2433must also be specified.
2434.It Cd options PCIBIOS
2435Enable support for initializing the PCI bus using information from
2436the BIOS.
2437See
2438.Xr pcibios 4
2439for details.
2440.It Cd options MTRR
2441Include support for accessing MTRR registers from user-space.
2442See
2443.Xr i386_get_mtrr 2 .
2444.It Cd options BEEP_ONHALT
2445Make the system speaker emit several beeps when it is completely safe to
2446power down the computer after a
2447.Xr halt 8
2448command.
2449Requires
2450.Xr sysbeep 4
2451support.
2452.It Cd options BEEP_ONHALT_COUNT=times
2453Number of times to beep the speaker when
2454.Cd options BEEP_ONHALT
2455is enabled.
2456Defaults to 3.
2457.It Cd options BEEP_ONHALT_PITCH=hz
2458The tone frequency used when
2459.Cd options BEEP_ONHALT
2460option, in hertz.
2461Defaults to 1500.
2462.It Cd options BEEP_ONHALT_PERIOD=msecs
2463The duration of each beep when
2464.Cd options BEEP_ONHALT
2465is enabled, in milliseconds.
2466Defaults to 250.
2467.It Cd options MULTIBOOT
2468Makes the kernel Multiboot-compliant, allowing it to be booted through
2469a Multiboot-compliant boot manager such as GRUB.
2470See
2471.Xr multiboot 8
2472for more information.
2473.It Cd options SPLASHSCREEN
2474Display a splash screen during boot.
2475.El
2476.Ss isa-specific Options
2477Options specific to
2478.Xr isa 4
2479busses.
2480.Bl -ohang
2481.It Cd options PCIC_ISA_ALLOC_IOBASE=address, PCIC_ISA_ALLOC_IOSIZE=size
2482Control the section of IO bus space used for PCMCIA bus space mapping.
2483Ideally the probed defaults are satisfactory, however in practice
2484that is not always the case.
2485See
2486.Xr pcmcia 4
2487for details.
2488.It Cd options PCIC_ISA_INTR_ALLOC_MASK=mask
2489Controls the allowable interrupts that may be used for PCMCIA
2490devices.
2491This mask is a logical-or of power-of-2s of allowable interrupts:
2492.Bd -literal -offset 04n
2493.Em "IRQ Val      IRQ Val      IRQ Val       IRQ Val"
2494 0  0x0001    4  0x0010    8  0x0100    12  0x1000
2495 1  0x0002    5  0x0020    9  0x0200    13  0x2000
2496 2  0x0004    6  0x0040   10  0x0400    14  0x4000
2497 3  0x0008    7  0x0080   11  0x0800    15  0x8000
2498.Ed
2499.It Cd options PCKBC_CNATTACH_SELFTEST
2500Perform a self test of the keyboard controller before attaching it as a
2501console.
2502This might be necessary on machines where we boot on cold iron, and
2503pckbc refuses to talk until we request a self test.
2504Currently only the netwinder port uses it.
2505.It Cd options PCKBD_CNATTACH_MAY_FAIL
2506If this option is set the PS/2 keyboard will not be used as the console
2507if it cannot be found during boot.
2508This allows other keyboards, like USB, to be the console keyboard.
2509.It Cd options PCKBD_LAYOUT=layout
2510Sets the default keyboard layout, see
2511.Xr pckbd 4 .
2512.El
2513.Ss m68k-specific Options
2514.Bl -ohang
2515.It Cd options FPU_EMULATE
2516Include support for MC68881/MC68882 emulator.
2517.It Cd options FPSP
2518Include support for 68040 floating point.
2519.It Cd options M68020,M68030,M68040,M68060
2520Include support for a specific CPU,
2521at least one (the one you are using) should be specified.
2522.It Cd options M060SP
2523Include software support for 68060.
2524This provides emulation of unimplemented
2525integer instructions as well as emulation of unimplemented floating point
2526instructions and data types and software support for floating point traps.
2527.El
2528.Ss powerpc-specific Options (OEA Only)
2529.Bl -ohang
2530.It Cd options PMAP_MEMLIMIT=value
2531Limit the amount of memory seen by the kernel to
2532.Ar value
2533bytes.
2534.It Cd options PTEGCOUNT=value
2535Specify the size of the page table as
2536.Ar value
2537PTE groups.
2538Normally, one PTEG is allocated per physical page frame.
2539.El
2540.Ss sparc-specific Options
2541.Bl -ohang
2542.It Cd options AUDIO_DEBUG
2543Enable simple event debugging of the logging of the
2544.Xr audio 4
2545device.
2546.It Cd options BLINK
2547Enable blinking of LED.
2548Blink rate is full cycle every N seconds for
2549N < then current load average.
2550See
2551.Xr getloadavg 3 .
2552.\" .It Cd options COLORFONT_CACHE
2553.\" What does this do?
2554.It Cd options COUNT_SW_LEFTOVERS
2555Count how many times the sw SCSI device has left 3, 2, 1 and 0 in the
2556sw_3_leftover, sw_2_leftover, sw_1_leftover, and sw_0_leftover
2557variables accessible from
2558.Xr ddb 4 .
2559See
2560.Xr sw 4 .
2561.It Cd options DEBUG_ALIGN
2562Adds debugging messages calls when user-requested alignment fault
2563handling happens.
2564.It Cd options DEBUG_EMUL
2565Adds debugging messages calls for emulated floating point and
2566alignment fixing operations.
2567.It Cd options EXTREME_DEBUG
2568Adds debugging functions callable from
2569.Xr ddb 4 .
2570The debug_pagetables, test_region and print_fe_map
2571functions print information about page tables for the SUN4M
2572platforms only.
2573.It Cd options EXTREME_EXTREME_DEBUG
2574Adds extra info to
2575.Em options EXTREME_DEBUG .
2576.It Cd options FPU_CONTEXT
2577Make
2578.Em options COMPAT_SVR4
2579getcontext and setcontext include floating point registers.
2580.It Cd options MAGMA_DEBUG
2581Adds debugging messages to the
2582.Xr magma 4
2583device.
2584.It Cd options RASTERCONS_FULLSCREEN
2585Use the entire screen for the console.
2586.It Cd options RASTERCONS_SMALLFONT
2587Use the Fixed font on the console, instead of the normal font.
2588.It Cd options SUN4
2589Support sun4 class machines.
2590.It Cd options SUN4C
2591Support sun4c class machines.
2592.It Cd options SUN4M
2593Support sun4m class machines.
2594.It Cd options SUN4_MMU3L
2595.\" XXX ???
2596Enable support for sun4 3-level MMU machines.
2597.It Cd options V9
2598Enable SPARC V9 assembler in
2599.Xr ddb 4 .
2600.El
2601.Ss sparc64-specific Options
2602.Bl -ohang
2603.It Cd options AUDIO_DEBUG
2604Enable simple event debugging of the logging of the
2605.Xr audio 4
2606device.
2607.It Cd options BLINK
2608Enable blinking of LED.
2609Blink rate is full cycle every N seconds for
2610N < then current load average.
2611See
2612.Xr getloadavg 3 .
2613.El
2614.Ss x68k-specific Options
2615.Bl -ohang
2616.It Cd options EXTENDED_MEMORY
2617Include support for extended memory, e.g., TS-6BE16 and 060turbo on-board.
2618.It Cd options JUPITER
2619Include support for Jupiter-X MPU accelerator
2620.It Cd options ZSCONSOLE,ZSCN_SPEED=value
2621Use the built-in serial port as the system-console.
2622Speed is specified in bps, defaults to 9600.
2623.It Cd options ITE_KERNEL_ATTR=value
2624Set the kernel message attribute for ITE.
2625Value, an integer, is a logical or of the following values:
2626.Bl -tag -width 4n -compact -offset indent
2627.It 1
2628color inversed
2629.It 2
2630underlined
2631.It 4
2632bolded
2633.El
2634.El
2635.Ss x86-specific Options
2636.Bl -ohang
2637.It Cd options NO_PCI_MSI_MSIX
2638Disable support for MSI/MSIX in the kernel.
2639See
2640.Xr pci_msi 9
2641for details of MSI/MSIX support
2642.It Cd options NO_PREEMPTION
2643Disables
2644.Xr kpreempt 9
2645support in the kernel.
2646.El
2647.\" The following requests should be uncommented and used where appropriate.
2648.\" .Sh FILES
2649.\" .Sh EXAMPLES
2650.Sh SEE ALSO
2651.Xr config 1 ,
2652.Xr gcc 1 ,
2653.Xr gdb 1 ,
2654.Xr ktrace 1 ,
2655.Xr quota 1 ,
2656.Xr vndcompress 1 ,
2657.Xr gettimeofday 2 ,
2658.Xr i386_get_mtrr 2 ,
2659.Xr i386_iopl 2 ,
2660.Xr msgctl 2 ,
2661.Xr msgget 2 ,
2662.Xr msgrcv 2 ,
2663.Xr msgsnd 2 ,
2664.Xr ntp_adjtime 2 ,
2665.Xr ntp_gettime 2 ,
2666.Xr reboot 2 ,
2667.Xr semctl 2 ,
2668.Xr semget 2 ,
2669.Xr semop 2 ,
2670.Xr shmat 2 ,
2671.Xr shmctl 2 ,
2672.Xr shmdt 2 ,
2673.Xr shmget 2 ,
2674.Xr sysctl 3 ,
2675.Xr apm 4 ,
2676.Xr ddb 4 ,
2677.Xr inet 4 ,
2678.Xr md 4 ,
2679.Xr pcibios 4 ,
2680.Xr pcmcia 4 ,
2681.Xr ppp 4 ,
2682.Xr userconf 4 ,
2683.Xr vnd 4 ,
2684.Xr wscons 4 ,
2685.Xr config 5 ,
2686.Xr edquota 8 ,
2687.Xr init 8 ,
2688.Xr mdsetimage 8 ,
2689.Xr mount_cd9660 8 ,
2690.Xr mount_fdesc 8 ,
2691.Xr mount_kernfs 8 ,
2692.Xr mount_lfs 8 ,
2693.Xr mount_mfs 8 ,
2694.Xr mount_msdos 8 ,
2695.Xr mount_nfs 8 ,
2696.Xr mount_ntfs 8 ,
2697.Xr mount_null 8 ,
2698.Xr mount_portal 8 ,
2699.Xr mount_procfs 8 ,
2700.Xr mount_udf 8 ,
2701.Xr mount_umap 8 ,
2702.Xr mount_union 8 ,
2703.Xr mrouted 8 ,
2704.Xr newfs_lfs 8 ,
2705.Xr ntpd 8 ,
2706.Xr quotaon 8 ,
2707.Xr rpc.rquotad 8 ,
2708.Xr sysctl 8 ,
2709.Xr in_getifa 9 ,
2710.Xr kernhist 9
2711.Sh HISTORY
2712The
2713.Nm
2714man page first appeared in
2715.Nx 1.3 .
2716