1.\" $NetBSD: options.4,v 1.168 2002/04/07 04:20:17 lukem 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 November 20, 2001 34.Os 35.Dt OPTIONS 4 36.Sh NAME 37.Nm options 38.Nd Miscellaneous kernel configuration options 39.Sh SYNOPSIS 40.Cd options ... 41.Sh DESCRIPTION 42This manual page describes a number of miscellaneous kernel 43configuration options that may be specified in a kernel config 44file. 45See 46.Xr config 8 47for information on how to configure and build kernels. 48.Em Note : 49Options are passed to the compile process as -D flags to the C 50compiler. 51.Ss Compatibility Options 52.Bl -ohang 53.It Cd options COMPAT_09 54Enable binary compatibility with 55.Nx 0.9 . 56This enables support for 5716-bit user, group, and process ids (following revisions support 5832-bit identifiers), 59It also allows the use of the deprecated 60.Xr getdomainname 3 , 61.Xr setdomainname 3 , 62and 63.Xr uname 3 64syscalls. 65This option also allows using numeric filesystem identifiers rather 66than strings. 67Post 68.Nx 0.9 69versions use string identifiers. 70.It Cd options COMPAT_10 71Enable binary compatibility with 72.Nx 1.0 . 73This option allows the use of the filesystem name of 74.Dq ufs 75as an alias for 76.Dq ffs . 77The name 78.Dq ffs 79should be used post 1.0 in 80.Pa /etc/fstab 81and other files. 82It also adds old syscalls for the 83.At V 84shared memory interface. 85This was changed post 1.0 to work on 64-bit architectures. 86This option also enables 87.Dq sgtty 88compatibility, without which programs using the old interface produce 89an 90.Dq inappropriate ioctl 91error, and 92.Pa /dev/io 93only works when this option is set in the kernel, 94see 95.Xr io 4 96on ports that support it. 97.It Cd options COMPAT_11 98Enable binary compatibility with 99.Nx 1.1 . 100This allows binaries running on the i386 port to gain direct access to 101the io ports by opening 102.Pa /dev/io 103read/write. 104This functionality was replaced by 105.Xr i386_iopl 2 106post 1.1. 107On the 108.Tn Atari 109port, the location of the disk label was moved after 1.1. 110When the 111.Em COMPAT_11 112option is set, the kernel will read (pre) 1.1 style disk labels as a 113last resort. 114When a disklabel is re-written, the old style label will be replaced 115with a post 1.1 style label. 116This also enables 117.Em EXEC_ELF_NOTELESS 118option. 119.It Cd options COMPAT_12 120Enable binary compatibility with 121.Nx 1.2 . 122This allows the use of old syscalls for 123.Fn reboot 124and 125.Fn swapon . 126The syscall numbers were changed post 1.2 to add functionality to the 127.Xr reboot 2 128syscall, and the new 129.Xr swapctl 2 130interface was introduced. 131This also enables 132.Em EXEC_ELF_NOTELESS 133option. 134.It Cd options COMPAT_13 135Enable binary compatibility with 136.Nx 1.3 . 137This allows the use of old syscalls for 138.Fn sigaltstack , 139and also enables the old 140.Xr swapctl 2 141command 142.Dv SWAP_STATS 143(now called 144.Dv SWAP_OSTATS ) , 145which does not include the 146.Fa se_path 147member of 148.Va struct swapent . 149.It Cd options COMPAT_14 150Enable binary compatibility with 151.Nx 1.4 . 152This allows some old 153.Xr ioctl 2 154on 155.Xr wscons 4 156to be performed, and allows the 157.Dv NFSSVC_BIOD 158mode of the 159.Xr nfssvc 2 160system call to be used for compatibility with the deprecated nfsiod program. 161.It Cd options COMPAT_43 162Enables compatibility with 163.Bx 4.3 . 164This adds an old syscall for 165.Xr lseek 2 . 166It also adds the ioctls for 167.Dv TIOCGETP 168and 169.Dv TIOCSETP . 170The return values for 171.Xr getpid 2 , 172.Xr getgid 2 , 173and 174.Xr getuid 2 175syscalls are modified as well, to return the parent's pid and 176uid as well as the current process's. 177It also enables the deprecated 178.Dv NTTYDISC 179terminal line discipline. 180It also provides backwards compatibility with 181.Dq old 182SIOC[GS]IF{ADDR,DSTADDR,BRDADDR,NETMASK} interface ioctls, including 183binary compatibility with code written before the introduction of the 184sa_len field in sockaddrs. 185It also enables 186support for some older pre 187.Bx 4.4 188socket calls. 189.It Cd options COMPAT_SVR4 190On those architectures that support it, this enables binary 191compatibility with 192.At V.4 193applications built for the same architecture. 194This currently includes the i386, m68k and sparc port. 195.It Cd options COMPAT_LINUX 196On those architectures that support it, this enables binary 197compatibility with Linux ELF and 198.Xr a.out 5 199applications built for the same architecture. 200This currently includes the alpha, i386, powerpc, and m68k 201ports. There is also a minimal support for running Linux 202binaries on Mips ports, but it is experimental at 203the moment. 204.It Cd options COMPAT_SUNOS 205On those architectures that support it, this enables binary 206compatibility with 207.Tn SunOS 4.1 208applications built for the same architecture. 209This currently includes the sparc, sparc64 and most or all m68k ports. 210Note that the sparc64 requires the 211.Em COMPAT_NETBSD32 212option for 64-bit kernels, in addition to this option. 213.It Cd options COMPAT_ULTRIX 214On those architectures that support it, this enables binary 215compatibility with 216.Tn Ultrix 217applications built for the same architecture. 218This currently is limited to the pmax. 219The functionality of this option is unknown. 220.It Cd options COMPAT_FREEBSD 221On those architectures that support it, this enables binary 222compatibility with 223.Fx 224applications built for the same architecture. 225At the moment this is limited to the i386 port. 226.It Cd options COMPAT_HPUX 227On those architectures that support it, this enables binary 228compatibility with 229.Tn HP/UX 230applications built for the same architecture. 231This is limited to the hp300 port, and has some known bugs. 232A limited set of programs do work. 233.It Cd options COMPAT_IBCS2 234On those architectures that support it, this enables binary 235compatibility with iBCS2 or SVR3 applications built for the same architecture. 236This is currently limited to the i386 and vax ports. 237.It Cd options COMPAT_IRIX 238On those architectures that support it, this enables binary 239compatibility with IRIX o32 binaries built for the same architecture. 240This feature is experimental, and it is currently limited to 241the sgimips port of 242.Nx . 243.It Cd options COMPAT_OSF1 244On those architectures that support it, this enables binary 245compatibility with 246.Tn Digital 247.Ux 248.Po 249formerly 250.Tn OSF/1 251.Pc 252applications built for the same architecture. 253This is currently limited to the alpha port. 254.It Cd options COMPAT_NOMID 255Enable compatibility with 256.Xr a.out 5 257executables that lack a machine ID. 258This includes 259.Nx 0.8 Ns 's 260ZMAGIC format, and 386BSD and BSDI's 261QMAGIC, NMAGIC, and OMAGIC 262.Xr a.out 5 263formats. 264.It Cd options COMPAT_NETBSD32 265On those architectures that support it, this enables binary 266compatibility with 32-bit applications built for the same architecture. 267This is currently limited to the sparc64 port, and only applicable for 26864-bit kernels. 269.It Cd options COMPAT_SVR4_32 270On those architectures that support it, this enables binary 271compatibility with 32-bit SVR4 applications built for the same architecture. 272This is currently limited to the sparc64 port, and only applicable for 27364-bit kernels. 274.It Cd options COMPAT_AOUT_M68K 275On m68k architectures which have switched to ELF, 276this enable binary compatibility with 277.Nx Ns Tn /m68k 278.Xr a.out 5 279executables on 280.Nx Ns Tn /m68k 281ELF kernel. 282This handles alignment incompatibility of m68k ABI between 283a.out and ELF which causes the structure padding differences. 284Currently only some system calls which use 285.Va struct stat 286are adjusted and some binaries which use 287.Xr sysctl 3 288to retrieve network details would not work properly. 289.It Cd options EXEC_ELF_NOTELESS 290Run unidentified ELF binaries as 291.Nx 292binaries. 293This might be needed for very old 294.Nx 295ELF binaries on some archs. 296These old binaries didn't contain an appropriate 297.Li .note.netbsd.ident 298section, and thus can't be identified by the kernel as 299.Nx 300binaries otherwise. 301Beware - if this option is on, the kernel would run 302.Em any 303unknown ELF binaries as if they were 304.Nx 305binaries. 306.El 307.Ss Debugging Options 308.Bl -ohang 309.It Cd options DDB 310Compiles in a kernel debugger for diagnosing kernel problems. 311See 312.Xr ddb 4 313for details. 314.Em NOTE : 315not available on all architectures. 316.It Cd options DDB_FROMCONSOLE=integer 317If set to non-zero, DDB may be entered by sending a break on a serial 318console or by a special key sequence on a graphics console. 319A value of "0" ignores console breaks or key sequences, 320It not explicitly specified, the default value is "1". 321Note that this sets the value of the 322.Em ddb.fromconsole 323.Xr sysctl 3 324variable which may be changed at run time -- see 325.Xr sysctl 8 326for details. 327.It Cd options DDB_HISTORY_SIZE=integer 328If this is non-zero, enable history editing in the kernel debugger 329and set the size of the history to this value. 330.It Cd options DDB_ONPANIC 331If set to non-zero, DDB will be entered upon kernel panic. 332The default if not specified is "1". 333Note that this sets the value of the 334.Em ddb.onpanic 335.Xr sysctl 3 336variable which may be changed at run time -- see 337.Xr sysctl 8 338for details. 339.It Cd options DDB_BREAK_CHAR=integer 340This option overrides the using break to enter the kernel debugger 341on the serial console. 342The value given will is the ascii value to be used instead. 343This is currently only supported by the com driver. 344.It Cd options KGDB 345Compiles in a remote kernel debugger stub for diagnosing kernel problems 346using the 347.Dq remote target 348feature of gdb. 349See 350.Xr gdb 1 351for details. 352.Em NOTE : 353not available on all architectures. 354.It Cd options KGDB_DEV 355Device number 356.Po 357as a 358.Dv dev_t 359.Pc 360of kgdb device. 361.It Cd options KGDB_DEVADDR 362Memory address of kgdb device. 363.It Cd options KGDB_DEVMODE 364Permissions of kgdb device. 365.It Cd options KGDB_DEVNAME 366Device name of kgdb device. 367.It Cd options KGDB_DEVRATE 368Baud rate of kgdb device. 369.It Cd makeoptions DEBUG="-g" 370The -g flag causes 371.Pa netbsd.gdb 372to be built in addition to 373.Pa netbsd . 374.Pa netbsd.gdb 375is useful for debugging kernel crash dumps with gdb. 376The command 377.Dl gdb -k 378invokes gdb in kernel debugger mode. 379See 380.Xr gdb 1 381for details. 382This also turns on 383.Em options DEBUG 384(which see). 385.It Cd options DEBUG 386Turns on miscellaneous kernel debugging. 387Since options are turned into preprocessor defines (see above), 388.Em options DEBUG 389is equivalent to doing a 390.Em #define DEBUG 391throughout the kernel. 392Much of the kernel has 393.Em #ifdef DEBUG 394conditionalized debugging code. 395Note that many parts of the kernel (typically device drivers) include their own 396.Em #ifdef XXX_DEBUG 397conditionals instead. 398This option also turns on certain other options, 399which may decrease system performance. 400.It Cd options DIAGNOSTIC 401Adds code to the kernel that does internal consistency checks. 402This code will cause the kernel to panic if corruption of internal data 403structures is detected. These checks can decrease performance up to 15%. 404.It Cd options KTRACE 405Add hooks for the system call tracing facility, which allows users to 406watch the system call invocation behavior of processes. 407See 408.Xr ktrace 1 409for details. 410.It Cd options MSGBUFSIZE=integer 411This option sets the size of the kernel message buffer. 412This buffer holds the kernel output of 413.Fn printf 414when not (yet) read by 415.Xr syslogd 8 . 416This is particularly useful when the system has crashed and you wish to lookup 417the kernel output from just before the crash. 418Also, since the autoconfig output becomes more and more verbose, 419it sometimes happens that the message buffer overflows before 420.Xr syslogd 8 421was able to read it. 422Note that not all systems are capable of obtaining a variable sized message 423buffer. 424There are also some systems on which memory contents are not preserved 425across reboots. 426.It Cd options MALLOCLOG 427Enables an event log for 428.Xr malloc 9 . 429Useful for tracking down 430.Dq Data modified on freelist 431and 432.Dq multiple free 433problems. 434.It Cd options MALLOCLOGSIZE=integer 435Defines the number of entries in the malloc log. 436Default is 100000 entries. 437.It Cd options UVMHIST 438Enables the UVM history logs, which create in-memory traces of 439various UVM activities. These logs can be displayed be calling 440.Fn uvmhist_dump 441or 442.Fn uvm_hist 443with appropriate arguments from DDB. 444See the kernel source file sys/uvm/uvm_stat.c for details. 445.It Cd options UVMHIST_PRINT 446Prints the UVM history logs on the system console as entries are added. 447Note that the output is 448.Em extremely 449voluminous, so this option is really only useful for debugging 450the very earliest parts of kernel initialization. 451.El 452.Ss File Systems 453.Bl -ohang 454.It Cd file-system FFS 455Includes code implementing the Berkeley Fast File System 456.Em ( FFS ) . 457Most machines need this if they are not running diskless. 458.It Cd file-system EXT2FS 459Includes code implementing the Second Extended File System 460.Em ( EXT2FS ) 461, revision 0 and revision 1 with the 462.Em filetype 463and 464.Em sparse_super 465options. This is the most commonly used file system on the Linux operating 466system, and is provided here for compatibility. 467Some of the specific features of 468.Em EXT2FS 469like the "behavior on errors" are not implemented. 470This file system can't be used with UID or GID greater than 65535. 471See 472.Xr mount_ext2fs 8 473for details. 474.It Cd file-system LFS 475.Em [EXPERIMENTAL] 476Include the Log-structured File System 477.Em ( LFS ) . 478See 479.Xr mount_lfs 8 480and 481.Xr newfs_lfs 8 482for details. 483.It Cd file-system MFS 484Include the Memory File System 485.Em ( MFS ) . 486This file system stores files in swappable memory, and produces 487notable performance improvements when it is used as the file store 488for 489.Pa /tmp 490and similar file systems. 491See 492.Xr mount_mfs 8 493for details. 494.It Cd file-system NFS 495Include the client side of the Network File System 496.Pq Tn NFS 497remote file sharing protocol. 498Although the bulk of the code implementing 499.Tn NFS 500is kernel based, several user level daemons are needed for it to work. 501See 502.Xr mount_nfs 8 503for details. 504.It Cd file-system CD9660 505Includes code for the 506.Tn ISO 5079660 + Rock Ridge file system, which is the standard file system on many 508.Tn CD-ROM 509discs. 510Useful primarily if you have a 511.Tn CD-ROM 512drive. 513See 514.Xr mount_cd9660 8 515for details. 516.It Cd file-system MSDOSFS 517Includes the 518.Tn MS-DOS 519FAT file system, which is reportedly still used 520by unfortunate people who have not heard about 521.Nx . 522Also implements the 523.Tn Windows 95 524extensions to the same, which permit the use of longer, mixed case 525file names. 526See 527.Xr mount_msdos 8 528and 529.Xr fsck_msdos 8 530for details. 531.It Cd file-system NTFS 532.Em [EXPERIMENTAL] 533Includes code for the 534.Tn Microsoft Windows NT 535file system. 536See 537.Xr mount_ntfs 8 538for details. 539.It Cd file-system FDESC 540Includes code for a file system, conventionally mounted on 541.Pa /dev/fd , 542which permits access to the per-process file descriptor space via 543special files in the file system. 544See 545.Xr mount_fdesc 8 546for details. 547Note that this facility is redundant, and thus unneeded on most 548.Nx 549systems, since the 550.Xr fd 4 551pseudodevice driver already provides identical functionality. 552On most 553.Nx 554systems, instances of 555.Xr fd 4 556are mknoded under 557.Pa /dev/fd/ 558and on 559.Pa /dev/stdin , 560.Pa /dev/stdout , 561and 562.Pa /dev/stderr . 563.It Cd file-system KERNFS 564Includes code which permits the mounting of a special file system 565(normally mounted on 566.Pa /kern ) 567in which files representing various kernel variables and parameters 568may be found. 569See 570.Xr mount_kernfs 8 571for details. 572.It Cd file-system NULLFS 573Includes code for a loopback file system. 574This permits portions of the file hierarchy to be re-mounted in other places. 575The code really exists to provide an example of a stackable file system layer. 576See 577.Xr mount_null 8 578for details. 579.It Cd file-system OVERLAY 580Includes code for a file system filter. 581This permits the overlay file system to intercept all access to an underlying 582file system. This file system is intended to serve as an example of 583a stacking file system which has a need to interpose itself between an 584underlying file system and all other access. 585See 586.Xr mount_overlay 8 587for details. 588.It Cd file-system PORTAL 589.Em [EXPERIMENTAL] 590Includes the portal filesystem. 591This permits interesting tricks like opening 592.Tn TCP 593sockets by opening files in the file system. 594The portal file system is conventionally mounted on 595.Pa /p 596and is partially implemented by a special daemon. 597See 598.Xr mount_portal 8 599for details. 600.It Cd file-system PROCFS 601Includes code for a special file system (conventionally mounted on 602.Pa /proc ) 603in which the process space becomes visible in the file system. 604Among 605other things, the memory spaces of processes running on the system are 606visible as files, and signals may be sent to processes by writing to 607.Pa ctl 608files in the procfs namespace. 609See 610.Xr mount_procfs 8 611for details. 612.It Cd file-system UMAPFS 613Includes a loopback file system in which user and group ids may be 614remapped -- this can be useful when mounting alien file systems with 615different uids and gids than the local system. 616See 617.Xr mount_umap 8 618for details. 619.It Cd file-system UNION 620.Em [EXPERIMENTAL] 621Includes code for the union file system, which permits directories to 622be mounted on top of each other in such a way that both file systems 623remain visible -- this permits tricks like allowing writing (and the 624deleting of files) on a read-only file system like a 625.Tn CD-ROM 626by mounting a local writable file system on top of the read-only file system. 627See 628.Xr mount_union 8 629for details. 630.It Cd file-system CODA 631.Em [EXPERIMENTAL] 632Includes code for the Coda file system. 633Coda is a distributed file system like NFS and AFS. It is 634freely available, like NFS, but it functions much like AFS in being a 635.Dq stateful 636file system. Both Coda and AFS cache files on your local 637machine to improve performance. Then Coda goes a step further than AFS 638by letting you access the cached files when there is no available 639network, viz. disconnected laptops and network outages. In Coda, both 640the client and server are outside the kernel which makes them easier 641to experiment with. Coda is available for several UNIX and non-UNIX 642platforms. 643See http://www.coda.cs.cmu.edu for more details. 644NOTE: You also need to enable the pseudo-device, vcoda, for the Coda 645filesystem to work. 646.El 647.Ss File System Options 648.Bl -ohang 649.It Cd options NFSSERVER 650Include the server side of the 651.Em NFS 652(Network File System) remote file sharing protocol. 653Although the bulk of the code implementing 654.Em NFS 655is kernel based, several user level daemons are needed for it to 656work. 657See 658.Xr mountd 8 659and 660.Xr nfsd 8 661for details. 662.It Cd options QUOTA 663Enables kernel support for file system quotas. 664See 665.Xr quotaon 8 , 666.Xr edquota 8 , 667and 668.Xr quota 1 669for details. 670Note that quotas only work on 671.Dq ffs 672file systems, although 673.Xr rpc.rquotad 8 674permits them to be accessed over 675.Em NFS . 676.It Cd options FFS_EI 677Enable ``Endian-Independent'' FFS support. 678This allows a system to mount an FFS filesystem created for another 679architecture, at a small performance cost for all FFS filesystems. 680See also 681.Xr newfs 8 , 682.Xr fsck_ffs 8 , 683.Xr dumpfs 8 684for filesystem byte order status and manipulation. 685.It Cd options NVNODE=integer 686This option sets the size of the cache used by the name-to-inode translation 687routines, (a.k.a. the 688.Fn namei 689cache, though called by many other names in the kernel source). 690By default, this cache has 691.Dv NPROC 692(set as 20 + 16 * MAXUSERS) * (80 + NPROC / 8) entries. 693A reasonable way to derive a value of 694.Dv NVNODE , 695should you notice a large number of namei cache misses with a tool such as 696.Xr systat 1 , 697is to examine your system's current computed value with 698.Xr sysctl 8 , 699(which calls this parameter "kern.maxvnodes") and to increase this value 700until either the namei cache hit rate improves or it is determined that 701your system does not benefit substantially from an increase in the size of 702the namei cache. 703.It Cd options NAMECACHE_ENTER_REVERSE 704Causes the namei cache to always enter a reverse mapping (vnode -\*[Gt] name) 705as well as a normal one. Normally, this is already done for directory 706vnodes, to speed up the getcwd operation. This option will cause 707longer hash chains in the reverse cache, and thus slow down 708getcwd somewhat. However, it does make vnode -\*[Gt] path translations 709possible in some cases. For now, only useful if strict /proc/#/maps 710emulation for Linux binaries is required. 711.It Cd options EXT2FS_SYSTEM_FLAGS 712This option changes the behavior of the APPEND and IMMUTABLE flags 713for a file on an 714.Em EXT2FS 715filesystem. 716Without this option, the superuser or owner of the file can 717set and clear them. 718With this option, only the superuser can set them, and 719they can't be cleared if the securelevel is greater than 0. 720See also 721.Xr chflags 1 . 722.It Cd options NFS_BOOT_BOOTP 723Enable use of the BOOTP protocol (RFC 951, 1048) to get configuration 724information if NFS is used to mount the root file system. 725See 726.Xr diskless 8 727for details. 728.It Cd options NFS_BOOT_DHCP 729Same as 730.Dq NFS_BOOT_BOOTP , 731but use the DHCP extensions to the 732BOOTP protocol (RFC 1541). 733.It Cd options NFS_BOOT_BOOTP_REQFILE 734Specifies the string sent in the bp_file field of the BOOTP / DHCP 735request packet. 736.It Cd options NFS_BOOT_BOOTPARAM 737Enable use of the BOOTPARAM protocol, consisting of RARP and 738BOOTPARAM RPC, to get configuration information if NFS 739is used to mount the root file system. 740See 741.Xr diskless 8 742for details. 743.It Cd options NFS_BOOT_RWSIZE=value 744Set the initial NFS read and write sizes for diskless-boot requests. 745The normal default is 8Kbytes. This option provides a way to lower 746the value (e.g., to 1024 bytes) as a workaround for buggy network 747interface cards or boot proms. Once booted, the read and write request 748sizes can be increased by remounting the filesystem. See 749.Xr mount_nfs 8 750for details. 751.It Cd options NFS_V2_ONLY 752Reduce the size of the NFS client code by omitting code that's only required 753for NFSv3 and NQNFS support, leaving only that code required to use NFSv2 754servers. 755.El 756.Ss Miscellaneous Options 757.Bl -ohang 758.It Cd options LKM 759Enable loadable kernel modules. 760See 761.Xr lkm 4 762for details. 763.Em NOTE : 764not available on all architectures. 765.It Cd options INSECURE 766Hardwires the kernel security level at -1. 767This means that the system 768always runs in secure level 0 mode, even when running multiuser. 769See the manual page for 770.Xr init 8 771for details on the implications of this. 772The kernel secure level may manipulated by the superuser by altering the 773.Em kern.securelevel 774.Xr sysctl 3 775variable (the secure level may only be lowered by a call from process ID 1, 776i.e. 777.Xr init 8 ) . 778See also 779.Xr sysctl 8 780and 781.Xr sysctl 3 . 782.It Cd options UCONSOLE 783Normally, only the superuser can execute the 784.Dv TIOCCONS 785.Xr ioctl 2 , 786which redirects console output to a non-console tty. 787See 788.Xr tty 4 789for details. 790This option permits any user to execute the 791.Dv TIOCCONS 792.Xr ioctl 2 . 793This is useful on 794machines such as personal workstations which run 795.Xr X 7 796servers, where one would prefer to permit console output to be 797viewed in a window without requiring a suid root program to do it. 798.It Cd options MEMORY_DISK_HOOKS 799This option allows for some machine dependent functions to be called when 800the 801.Xr md 4 802.Tn RAM 803disk driver is configured. 804This can result in automatically loading a 805.Tn RAM 806disk from floppy on open (among other things). 807.It Cd options MEMORY_DISK_IS_ROOT 808Forces the 809.Xr md 4 810.Tn RAM 811disk to be the root device. 812This can only be overridden when 813the kernel is booted in the 'ask-for-root' mode. 814.It Cd options MEMORY_DISK_ROOT_SIZE=integer 815Allocates the given number of 512 byte blocks as memory for the 816.Xr md 4 817.Tn RAM 818disk, to be populated with 819.Xr mdsetimage 8 . 820.It Cd options VNODE_OP_NOINLINE 821Do not inline the VOP_*() calls in the kernel. 822On i386 GENERIC, this saves 36k of kernel text. Useful 823for install media kernels, small memory systems and embedded systems. 824.It Cd options HZ=integer 825On ports that support it, set the system clock frequency (see 826.Xr hz 9 ) 827to the supplied value. Handle with care. 828.It Cd options NTP 829Turns on in-kernel precision timekeeping support used by software 830implementing 831.Em NTP 832(Network Time Protocol, RFC1305). 833The 834.Em NTP 835option adds an in-kernel Phase-Locked Loop (PLL) for normal 836.Em NTP 837operation, and a Frequency-Locked Loop (FLL) for intermittently-connected 838operation. 839.Xr ntpd 8 840will employ a user-level PLL when kernel support is unavailable, 841but the in-kernel version has lower latency and more precision, and 842so typically keeps much better time. 843The interface to the kernel 844.Em NTP 845support is provided by the 846.Xr ntp_adjtime 2 847and 848.Xr ntp_gettime 2 849system calls, which are intended for use by 850.Xr ntpd 8 851and are enabled by the option. 852On systems with sub-microsecond resolution timers, or where (HZ / 100000) 853is not an integer, the 854.Em NTP 855option also enables extended-precision arithmetic to keep track of 856fractional clock ticks at NTP time-format precision. 857.It Cd options PPS_SYNC 858This option enables a kernel serial line discipline for receiving time 859phase signals from an external reference clock such as a radio clock. 860(The 861.Em NTP 862option (which see) must be on if the 863.Em PPS_SYNC 864option is used.) 865Some reference clocks generate a Pulse Per Second (PPS) signal in 866phase with their time source. 867The 868.Em PPS 869line discipline receives this signal on either the data leads 870or the DCD control lead of a serial port. 871.Em NTP 872uses the PPS signal to discipline the local clock oscillator to a high 873degree of precision (typically less than 50 microseconds in time and 8740.1 ppm in accuracy). 875.Em PPS 876can also generate a serial output pulse when the system receives a PPS 877interrupt. 878This can be used to measure the system interrupt latency and thus calibrate 879.Em NTP 880to account for it. 881Using 882.Em PPS 883usually requires a 884gadget box 885to convert from TTL to RS-232 signal levels. 886The gadget box and PPS are described in more detail in the HTML documentation 887for 888.Xr ntpd 8 889in 890.Pa /usr/share/doc/html/ntp . 891.It Cd options SETUIDSCRIPTS 892Allows scripts with the setuid bit set to execute as the effective 893user rather than the real user, just like binary executables. 894.Pp 895.Em NOTE : 896Using this option will also enable 897.Em options FDSCRIPTS 898.It Cd option FDSCRIPTS 899Allows execution of scripts with the execute bit set, but not the 900read bit, by opening the file and passing the file descriptor to 901the shell, rather than the filename. 902.Pp 903.Em NOTE : 904Execute only (non-readable) scripts will have 905.Va argv[0] 906set to 907.Pa /dev/fd/* . 908What this option allows as far as security is 909concerned, is the ability to safely ensure that the correct script 910is run by the interpreter, as it is passed as an already open file. 911.It Cd options PUCCN 912Enables treating serial ports found on PCI boards 913.Xr puc 4 914as potential console devices. The method for choosing such a console 915device is port dependent. 916.It Cd options RTC_OFFSET=integer 917The kernel (and typically the hardware battery backed-up clock on 918those machines that have one) keeps time in 919.Em UTC 920(Universal Coordinated Time, once known as 921.Em GMT , 922or Greenwich Mean Time) 923and not in the time of the local time zone. 924The 925.Em RTC_OFFSET 926option is used on some ports (such as the i386) to tell the kernel 927that the hardware clock is offset from 928.Em UTC 929by the specified number of minutes. 930This is typically used when a machine boots several operating 931systems and one of them wants the hardware clock to run in the 932local time zone and not in 933.Em UTC , 934e.g. 935.Em RTC_OFFSET=300 936means 937the hardware clock is set to US Eastern Time (300 minutes behind 938.Em UTC ) , 939and not 940.Em UTC . 941(Note: 942.Em RTC_OFFSET 943is used to initialize a kernel variable named 944.Va rtc_offset 945which is the source actually used to determine the clock offset, and 946which may be accessed via the kern.rtc_offset sysctl variable. 947See 948.Xr sysctl 8 949and 950.Xr sysctl 3 951for details. 952Since the kernel clock is initialized from the hardware clock very 953early in the boot process, it is not possible to meaningfully change 954.Va rtc_offset 955in system initialization scripts. 956Changing this value currently may only be done at kernel compile 957time or by patching the kernel and rebooting). 958.Pp 959.Em NOTE : 960Unfortunately, in many cases where the hardware clock 961is kept in local time, it is adjusted for Daylight Savings 962Time; this means that attempting to use 963.Em RTC_OFFSET 964to let 965.Nx 966coexist with such an operating system, like Windows, 967would necessitate changing 968.Em RTC_OFFSET 969twice a year. As such, this solution is imperfect. 970.It Cd options KMEMSTATS 971The kernel memory allocator, 972.Xr malloc 9 , 973will keep statistics on its performance if this option is enabled. 974Unfortunately, this option therefore essentially disables the 975.Fn MALLOC 976and 977.Fn FREE 978forms of the memory allocator, which are used to enhance the performance 979of certain critical sections of code in the kernel. 980This option therefore can lead to a significant decrease in the 981performance of certain code in the kernel if enabled. 982Examples of such code include the 983.Fn namei 984routine, the 985.Xr ccd 4 986driver, 987the 988.Xr ncr 4 989driver, 990and much of the networking code. 991.It Cd options MAXUPRC=integer 992Sets the soft 993.Dv RLIMIT_NPROC 994resource limit, which specifies the maximum number of simultaneous 995processes a user is permitted to run, for process 0; 996this value is inherited by its child processes. 997It defaults to 998.Em CHILD_MAX , 999which is currently defined to be 160. 1000Setting 1001.Em MAXUPRC 1002to a value less than 1003.Em CHILD_MAX 1004is not permitted, as this would result in a violation of the semantics of 1005.St -p1003.1-90 . 1006.It Cd options DEFCORENAME=string 1007Sets the default value of the 1008.Em kern.defcorename 1009sysctl variable, otherwise it is set to 1010.Nm %n.core . 1011See 1012.Xr sysctl 8 1013and 1014.Xr sysctl 3 1015for details. 1016.It Cd options RASOPS_CLIPPING 1017Enables clipping within the 1018.Nm rasops 1019raster-console output system. 1020.Em NOTE : 1021only available on architectures that use 1022.Nm rasops 1023for console output. 1024.It Cd options RASOPS_SMALL 1025Removes optimized character writing code from the 1026.Nm rasops 1027raster-console output system. 1028.Em NOTE : 1029only available on architectures that use 1030.Nm rasops 1031for console output. 1032.It Cd options INCLUDE_CONFIG_FILE 1033Embeds the kernel config file used to define the kernel in the kernel 1034binary itself. The embedded data also includes any files directly 1035included by the config file itself, e.g. 1036.Pa GENERIC.local 1037or 1038.Pa std.$MACHINE . 1039The embedded config file can be extracted from the resulting kernel by 1040the following command: 1041.Bd -literal -offset indent 1042strings netbsd | sed -n 's/^_CFG_//p' | unvis 1043.Ed 1044.It Cd options INCLUDE_JUST_CONFIG 1045Similar to the above option, but includes just the actual config file, 1046not any included files. 1047.It Cd options PIPE_SOCKETPAIR 1048Use slower, but smaller socketpair(2)-based pipe implementation instead 1049of default faster, but bigger one. Primarily useful for installation 1050kernels. 1051.El 1052.Ss Networking Options 1053.Bl -ohang 1054.It Cd options GATEWAY 1055Enables 1056.Em IPFORWARDING 1057(which see) 1058and (on most ports) increases the size of 1059.Em NMBCLUSTERS 1060(which see). 1061In general, 1062.Em GATEWAY 1063is used to indicate that a system should act as a router, and 1064.Em IPFORWARDING 1065is not invoked directly. 1066(Note that 1067.Em GATEWAY 1068has no impact on protocols other than 1069.Tn IP , 1070such as 1071.Tn CLNP 1072or 1073.Tn XNS ) . 1074.It Cd options IPFORWARDING=value 1075If 1076.Em value 1077is 1 this enables IP routing behavior. If 1078.Em value 1079is 0 (the default), it disables it. The 1080.Em GATEWAY 1081option sets this to 1 automatically. 1082With this option enabled, the machine will forward IP datagrams destined 1083for other machines between its interfaces. 1084Note that even without this option, the kernel will 1085still forward some packets (such as source routed packets) -- removing 1086.Em GATEWAY 1087and 1088.Em IPFORWARDING 1089is insufficient to stop all routing through a bastion host on a 1090firewall -- source routing is controlled independently. 1091To turn off source routing, use 1092.Em options IPFORWSRCRT=0 1093(which see). 1094Note that IP forwarding may be turned on and off independently of the 1095setting of the 1096.Em IPFORWARDING 1097option through the use of the 1098.Em net.inet.ip.forwarding 1099sysctl variable. 1100If 1101.Em net.inet.ip.forwarding 1102is 1, IP forwarding is on. 1103See 1104.Xr sysctl 8 1105and 1106.Xr sysctl 3 1107for details. 1108.It Cd options IPFORWSRCRT=value 1109If 1110.Em value 1111is set to zero, source routing of IP datagrams is turned off. 1112If 1113.Em value 1114is set to one (the default) or the option is absent, source routed IP 1115datagrams are forwarded by the machine. 1116Note that source routing of IP packets may be turned on and off 1117independently of the setting of the 1118.Em IPFORWSRCRT 1119option through the use of the 1120.Em net.inet.ip.forwsrcrt 1121sysctl variable. 1122If 1123.Em net.inet.ip.forwsrcrt 1124is 1, forwarding of source routed IP datagrams is on. 1125See 1126.Xr sysctl 8 1127and 1128.Xr sysctl 3 1129for details. 1130.It Cd options IFA_STATS 1131Tells the kernel to maintain per-address statistics on bytes sent 1132and received over (currently) internet and appletalk addresses. 1133.\"This can be a fairly expensive operation, so you probably want to 1134.\"keep this disabled. 1135The option is not recommended as it degrades system stability. 1136.It Cd options MROUTING 1137Includes support for IP multicast routers. 1138You certainly want 1139.Em INET 1140with this. 1141Multicast routing is controlled by the 1142.Xr mrouted 8 1143daemon. 1144.It Cd options INET 1145Includes support for the 1146.Tn TCP/IP 1147protocol stack. 1148You almost certainly want this. 1149See 1150.Xr inet 4 1151for details. 1152This option is currently required. 1153.It Cd options INET6 1154Includes support for the 1155.Tn IPv6 1156protocol stack. 1157See 1158.Xr inet6 4 1159for details. 1160Unlike 1161.Em INET , 1162.Em INET6 1163enables multicast routing code as well. 1164This option requires 1165.Em INET 1166at this moment, but it should not. 1167.It Cd options ND6_DEBUG 1168The option sets the default value of net.inet6.icmp6.nd6_debug to 1, 1169for debugging IPv6 neighbor discovery protocol handling. 1170See 1171.Xr sysctl 3 1172for details. 1173.It Cd options IPSEC 1174Includes support for the 1175.Tn IPsec 1176protocol. 1177See 1178.Xr ipsec 4 1179for details. 1180.Em IPSEC 1181will enable 1182secret key management part, 1183policy management part, 1184.Tn AH 1185and 1186.Tn IPComp . 1187Kernel binary will not be subject to export control in most of countries, 1188even if compiled with 1189.Em IPSEC . 1190For example, it should be okay to export it from within the United States 1191to the outside. 1192.Em INET6 1193and 1194.Em IPSEC 1195are orthogonal so you can get IPv4-only kernel with IPsec support, 1196IPv4/v6 dual support kernel without IPsec, and so forth. 1197This option requires 1198.Em INET 1199at this moment, but it should not. 1200.It Cd options IPSEC_DEBUG 1201Enables debugging code in 1202.Tn IPsec 1203stack. 1204This option assumes 1205.Em IPSEC . 1206.It Cd options IPSEC_ESP 1207Includes support for 1208.Tn IPsec 1209.Tn ESP 1210protocol. 1211See 1212.Xr ipsec 4 1213for details. 1214.Em IPSEC_ESP 1215will enable source code that is subject to export control in some countries 1216.Pq including the United States , 1217and compiled kernel binary will be subject to certain restriction. 1218This option assumes 1219.Em IPSEC . 1220.It Cd options SUBNETSARELOCAL 1221Sets default value for net.inet.ip.subnetsarelocal variable, which 1222controls whether non-directly-connected subnets of connected networks 1223are considered "local" for purposes of choosing the MSS for a TCP 1224connection. This is mostly present for historic reasons and 1225completely irrelevant if you enable Path MTU discovery. 1226.It Cd options HOSTZEROBROADCAST 1227Sets default value for net.inet.ip.hostzerobroadcast variable, which 1228controls whether the zeroth host address of each connected subnet is 1229also considered a broadcast address. Default value is "1", for 1230compatibility with old systems; if this is set to zero on all hosts on 1231a subnet, you should be able to fit an extra host per subnet on the 1232".0" address. 1233.It Cd options MCLSHIFT=value 1234This option is the base-2 logarithm of the size of mbuf clusters. 1235The 1236.Bx 1237networking stack keeps network packets in a linked 1238list, or chain, of kernel buffer objects called mbufs. 1239The system provides larger mbuf clusters as an optimization for 1240large packets, instead of using long chains for large packets. 1241The mbuf cluster size, 1242or 1243.Em MCLBYTES , 1244must be a power of two, and is computed as two raised to the power 1245.Em MCLSHIFT . 1246On systems with Ethernet network adaptors, 1247.Em MCLSHIFT 1248is often set to 11, giving 2048-byte mbuf clusters, large enough to 1249hold a 1500-byte 1250.Tn Ethernet 1251frame in a single cluster. 1252Systems with network interfaces supporting larger frame sizes like 1253.Tn ATM , 1254.Tn FDDI , 1255or 1256.Tn HIPPI 1257may perform better with 1258.Em MCLSHIFT 1259set to 12 or 13, giving mbuf cluster sizes of 4096 and 8192 bytes, 1260respectively. 1261.It Cd options NS 1262Include support for the 1263.Tn Xerox 1264.Tn XNS 1265protocol stack. 1266See 1267.Xr ns 4 1268for details. 1269.It Cd options ISO,TPIP 1270Include support for the ubiquitous 1271.Tn OSI 1272protocol stack. 1273See 1274.Xr iso 4 1275for details. 1276This option assumes 1277.Em INET . 1278.It Cd options EON 1279Include support for tunneling 1280.Tn OSI 1281protocols over 1282.Tn IP . 1283Known to be broken, or at least very fragile, and undocumented. 1284.It Cd options CCITT,LLC,HDLC 1285Include support for the 1286.Tn CCITT 1287(nee 1288.Tn ITU-TSS ) 1289.Tn X.25 1290protocol stack. 1291The state of this code is currently unknown, and probably contains bugs. 1292This option assumes 1293.Em INET . 1294.It Cd options NETATALK 1295Include support for the 1296.Tn AppleTalk 1297protocol stack. 1298The kernel provides provision for the 1299.Em Datagram Delivery Protocol 1300(DDP), providing SOCK_DGRAM support and 1301.Tn AppleTalk 1302routing. 1303This stack is used by the 1304.Em NETATALK 1305package, which adds support for 1306.Tn AppleTalk 1307server services via user libraries and applications. 1308.It Cd options IPNOPRIVPORTS 1309Normally, only root can bind a socket descriptor to a so-called 1310.Dq privileged 1311.Tn TCP 1312port, that is, a port number in the range 0-1023. 1313This option eliminates those checks from the kernel. 1314This can be useful if there is a desire to allow daemons without 1315privileges to bind those ports, e.g. on firewalls. 1316The security tradeoffs in doing this are subtle. 1317This option should only be used by experts. 1318.It Cd options TCP_COMPAT_42 1319.Tn TCP 1320bug compatibility with 1321.Bx 4.2 . 1322In 1323.Bx 4.2 , 1324.Tn TCP 1325sequence numbers were 32-bit signed values. 1326Modern implementations of TCP use unsigned values. 1327This option clamps the initial sequence number to start in 1328the range 2^31 rather than the full unsigned range of 2^32. 1329Also, under 1330.Bx 4.2 , 1331keepalive packets must contain at least one byte or else 1332the remote end would not respond. 1333.It Cd options TCP_DEBUG 1334Record the last 1335.Em TCP_NDEBUG 1336TCP packets with SO_DEBUG set, and decode to the console if 1337.Em tcpconsdebug 1338is set. 1339.It Cd options TCP_NDEBUG 1340Number of packets to record for 1341.Em TCP_DEBUG . 1342Defaults to 100. 1343.It Cd options PFIL_HOOKS 1344This option turns on the packet filter interface hooks. 1345See 1346.Xr pfil 9 1347for details. 1348This option assumes 1349.Em INET . 1350.It Cd options IPFILTER_LOG 1351This option, in conjunction with 1352.Em pseudo-device ipfilter , 1353enables logging of IP packets using ip-filter. 1354.It Cd options IPFILTER_DEFAULT_BLOCK 1355This option sets the default policy of ip-filter. 1356If it is set, ip-filter will block packets by default. 1357.It Cd options PPP_BSDCOMP 1358Enable support for BSD-compress 1359.Pq Sq bsdcomp 1360compression in ppp. 1361.It Cd options PPP_DEFLATE 1362Enable support for deflate 1363compression in ppp. 1364.It Cd options PPP_FILTER 1365This option turns on 1366.Xr pcap 3 1367based filtering for ppp connections. 1368This option is used by 1369.Xr pppd 8 1370which needs to be compiled with 1371.Em PPP_FILTER 1372defined (the current default). 1373.El 1374.Ss System V IPC Options 1375.Bl -ohang 1376.It Cd options SYSVMSG 1377Includes support for 1378.At V 1379style message queues. 1380See 1381.Xr msgctl 2 , 1382.Xr msgget 2 , 1383.Xr msgrcv 2 , 1384.Xr msgsnd 2 . 1385.It Cd options SYSVSEM 1386Includes support for 1387.At V 1388style semaphores. 1389See 1390.Xr semctl 2 , 1391.Xr semget 2 , 1392.Xr semop 2 . 1393.It Cd options SEMMNI=value 1394Sets the number of 1395.At V 1396style semaphore identifiers. The GENERIC config file for your port 1397will have the default. 1398.It Cd options SEMMNS=value 1399Sets the number of 1400.At V 1401style semaphores in the system. The GENERIC config file for your port 1402will have the default. 1403.It Cd options SEMUME=value 1404Sets the maximum number of undo entries per process for 1405.At V 1406style semaphores. The GENERIC config file for your port 1407will have the default. 1408.It Cd options SEMMNU=value 1409Sets the number of undo structures in the system for 1410.At V 1411style semaphores. The GENERIC config file for your port 1412will have the default. 1413.It Cd options SYSVSHM 1414Includes support for 1415.At V 1416style shared memory. 1417See 1418.Xr shmat 2 , 1419.Xr shmctl 2 , 1420.Xr shmdt 2 , 1421.Xr shmget 2 . 1422.It Cd options SHMMAXPGS=value 1423Sets the maximum number of 1424.At V 1425style shared memory pages that are available through the 1426.Xr shmget 2 1427system call. 1428Default value is 1024 on most ports. 1429See 1430.Pa /usr/include/machine/vmparam.h 1431for the default. 1432.El 1433.Ss VM Related Options 1434.Bl -ohang 1435.It Cd options NMBCLUSTERS=value 1436The number of mbuf clusters the kernel supports. Mbuf clusters are 1437MCLBYTES in size (usually 2k). This is used to compute the size of 1438the kernel VM map 1439.Em mb_map , 1440which maps mbuf clusters. 1441Default on most ports is 256 (512 with 1442.Dq options GATEWAY 1443). 1444See 1445.Pa /usr/include/machine/param.h 1446for exact default information. 1447Increase this value if you get 1448.Dq mb_map full 1449messages. 1450.It Cd options NKMEMPAGES=value 1451.It Cd options NKMEMPAGES_MIN=value 1452.It Cd options NKMEMPAGES_MAX=value 1453Size of kernel VM map 1454.Em kmem_map , 1455in PAGE_SIZE-sized chunks (the VM page size; this value may be read 1456from the 1457.Xr sysctl 8 1458variable 1459.Em hw.pagesize 1460). 1461This VM map is used to map the kernel malloc arena. 1462The kernel attempts to auto-size this map based on the amount of 1463physical memory in the system. Platform-specific code may place 1464bounds on this computed size, which may be viewed with the 1465.Xr sysctl 8 1466variable 1467.Em vm.nkmempages . 1468See 1469.Pa /usr/include/machine/param.h 1470for the default upper and lower bounds. 1471The related options 1472.Sq NKMEMPAGES_MIN 1473and 1474.Sq NKMEMPAGES_MAX 1475allow the bounds to be overridden in the kernel configuration file. 1476These options are provided in the event the computed value is 1477insufficient resulting in an 1478.Dq out of space in kmem_map 1479panic. 1480.It Cd options BUFCACHE=value 1481Size of the buffer cache as a percentage of total available 1482.Tn RAM . 1483Ignored if BUFPAGES is also specified. 1484.It Cd options NBUF=value 1485.It Cd options BUFPAGES=value 1486These options set the number of pages available for the buffer cache. 1487Their default value is a machine dependent value, often calculated as 1488between 5% and 10% of total available 1489.Tn RAM . 1490.It Cd options MAXTSIZ=bytes 1491Sets the maximum size limit of a process' text segment. See 1492.Pa /usr/include/machine/vmparam.h 1493for the port-specific default. 1494.It Cd options DFLDSIZ=bytes 1495Sets the default size limit of a process' data segment, the value that 1496will be returned as the soft limit for 1497.Dv RLIMIT_DATA 1498(as returned by 1499.Xr getrlimit 2 ) . 1500See 1501.Pa /usr/include/machine/vmparam.h 1502for the port-specific default. 1503.It Cd options MAXDSIZ=bytes 1504Sets the maximum size limit of a process' data segment, the value that 1505will be returned as the hard limit for 1506.Dv RLIMIT_DATA 1507(as returned by 1508.Xr getrlimit 2 ) . 1509See 1510.Pa /usr/include/machine/vmparam.h 1511for the port-specific default. 1512.It Cd options DFLSSIZ=bytes 1513Sets the default size limit of a process' stack segment, the value that 1514will be returned as the soft limit for 1515.Dv RLIMIT_STACK 1516(as returned by 1517.Xr getrlimit 2 ) . 1518See 1519.Pa /usr/include/machine/vmparam.h 1520for the port-specific default. 1521.It Cd options MAXSSIZ=bytes 1522Sets the maximum size limit of a process' stack segment, the value that 1523will be returned as the hard limit for 1524.Dv RLIMIT_STACK 1525(as returned by 1526.Xr getrlimit 2 ) . 1527See 1528.Pa /usr/include/machine/vmparam.h 1529for the port-specific default. 1530.El 1531.Ss amiga-specific Options 1532.Bl -ohang 1533.It Cd options BB060STUPIDROM 1534When the bootloader (which passes 1535.Tn AmigaOS 1536.Tn ROM 1537information) claims we have a 68060 1538.Tn CPU 1539without 1540.Tn FPU , 1541go look into the Processor Configuration Register (PCR) to find out. 1542You need this with 1543.Tn Amiga 1544.Tn ROM Ns s 1545up to (at least) V40.xxx (OS3.1), 1546when you boot via the bootblocks and don't have a DraCo. 1547.It Cd options IOBZCLOCK=frequency 1548The IOBlix boards come with two different serial master clocks: older ones 1549use 24 MHz, newer ones use 22.1184 MHz. The driver normally assumes the latter. 1550If your board uses 24 MHz, you can recompile your kernel with 1551options IOBZCLOCK=24000000 1552or patch the kernel variable 1553.Tn iobzclock 1554to the same value. 1555.It Cd options LIMITMEM=value 1556If there, limit the part of the first memory bank used by 1557.Nx 1558to value megabytes. 1559Default is unlimited. 1560.It Cd options NKPTADD=addvalue 1561.It Cd options NKPTADDSHIFT=shiftvalue 1562The 1563.Tn CPU 1564specific 1565.Tn MMU 1566table for the kernel is pre-allocated at kernel startup time. 1567Part of it is scaled with 1568.Va maxproc , 1569to have enough room to hold the user program 1570.Tn MMU 1571tables; the second part is a fixed amount for the kernel itself. 1572.Pp 1573The third part accounts for the size of the file buffer cache. 1574Its size is either 1575.Dv NKPTADD 1576pages (if defined) or memory size in bytes divided by two to 1577the power of 1578.Dv NKPTADDSHIFT . 1579The default is undefined 1580.Dv NKPTADD 1581and 1582.Dv NKPTADDSHIFT=24 , 1583allowing for 16 buffers per megabyte of main memory (while 1584a GENERIC kernel allocates about half of that). 1585When you get "can't get KPT page" panics, you should increase 1586.Dv NKPTADD 1587(if defined), or decrease 1588.Dv NKPTADDSHIFT 1589by one. 1590.It Cd options P5PPC68KBOARD 1591Add special support for Phase5 mixed 68k+PPC boards. Currently, this only 1592affects rebooting from 1593.Nx 1594and is only needed on 68040+PPC, not on 159568060+PPC; without this, affected machines will hang after 1596.Nx 1597has shut 1598down and will only restart after a keyboard reset or a power cycle. 1599.El 1600.Ss arm32-specific Options 1601.Bl -ohang 1602.It Cd options FRENCH_KEYBOARD 1603Include translation for French keyboards when using 1604.Xr pccons 4 1605on a Shark. 1606.It Cd options FINNISH_KEYBOARD 1607Include translation for Finnish keyboards when using 1608.Xr pccons 4 1609on a Shark. 1610.It Cd options GERMAN_KEYBOARD 1611Include translation for German keyboards when using 1612.Xr pccons 4 1613on a Shark. 1614.It Cd options NORWEGIAN_KEYBOARD 1615Include translation for French keyboards when using 1616.Xr pccons 4 1617on a Shark. 1618.El 1619.Ss atari-specific Options 1620.Bl -ohang 1621.It Cd options DISKLABEL_AHDI 1622Include support for AHDI (native Atari) disklabels. 1623.It Cd options DISKLABEL_NBDA 1624Include support for 1625.Nx Ns Tn /atari 1626labels. 1627If you don't set this option, it will be set automatically. 1628.Nx Ns Tn /atari 1629will not work without it. 1630.It Cd options FALCON_SCSI 1631Include support for the 5380-SCSI configuration as found on the Falcon. 1632.It Cd options RELOC_KERNEL 1633If set, the kernel will relocate itself to TT-RAM, if possible. 1634This will give you a slightly faster system. 1635.Em Beware 1636that on some TT030 systems, 1637the system will frequently dump with MMU-faults with this option enabled. 1638.It Cd options SERCONSOLE 1639Allow the modem1-port to act as the system-console. 1640A carrier should be active on modem1 during system boot to active 1641the console functionality. 1642.It Cd options TT_SCSI 1643Include support for the 5380-SCSI configuration as found on the TT030 1644and Hades. 1645.El 1646.Ss i386-specific Options 1647.Bl -ohang 1648.It Cd options I386_CPU,I486_CPU,I586_CPU,I686_CPU 1649Include support for a particular class of 1650.Tn CPU 1651.Po 1652.Tn i386 , 1653.Tn i486 , 1654.Tn Pentium , 1655or 1656.Tn Pentium Pro 1657.Pc . 1658If the appropriate class for your 1659.Tn CPU 1660is not configured, the kernel will use the highest class available 1661that will work. 1662In general, using the correct 1663.Tn CPU 1664class will result in the best performance. 1665At least one of these options must be present. 1666.It Cd options CPURESET_DELAY=value 1667specifies the time (in millisecond) to wait before doing a hardware reset 1668in the last phase of a reboot. This gives the user a chance to see error 1669messages from the shutdown operations (like NFS unmounts, buffer cache flush, 1670etc ...). Setting this to 0 will disable the delay. Default is 2 seconds. 1671.It Cd options MATH_EMULATE 1672Include the floating point emulator. 1673This is useful only for 1674.Tn CPU Ns s 1675that lack an 1676internal Floating Point Unit 1677.Pq Tn FPU 1678or co-processor. 1679.It Cd options VM86 1680Include support for virtual 8086 mode, used by 1681.Tn DOS 1682emulators and X servers to run BIOS code, e.g. for some VESA routines. 1683.It Cd options USER_LDT 1684Include i386-specific system calls for modifying the local descriptor table, 1685used by Windows emulators. 1686.It Cd options REALBASEMEM=integer 1687Overrides the base memory size passed in from the boot block. 1688(Value given in kilobytes.) 1689Use this option only if the boot block reports the size incorrectly. 1690(Note that some 1691.Tn BIOS Ns es 1692put the extended 1693.Tn BIOS 1694data area at the top of base memory, and therefore report a smaller 1695base memory size to prevent programs overwriting it. 1696This is correct behavior, and you should not use the 1697.Em REALBASEMEM 1698option to access this memory). 1699.It Cd options REALEXTMEM=integer 1700Overrides the extended memory size passed in from the boot block. 1701(Value given in kilobytes. Extended memory does not include the first megabyte.) 1702Use this option only if the boot block reports the size incorrectly. 1703.It Cd options FRENCH_KBD,FINNISH_KBD,GERMAN_KBD,NORWEGIAN_KBD 1704Select a non-US keyboard layout for the 1705.Em pccons 1706console driver. 1707.It Cd options CYRIX_CACHE_WORKS 1708Relevant only to the Cyrix 486DLC cpu. This option is used to turn on 1709the cache in hold-flush mode. It is not turned on by default because it 1710is known to have problems in certain motherboard implementations. 1711.It Cd options CYRIX_CACHE_REALLY_WORKS 1712Relevant only to the Cyrix 486DLC cpu. This option is used to turn on 1713the cache in write-back mode. It is not turned on by default because it 1714is known to have problems in certain motherboard implementations. In order 1715for this option to take effect, option 1716.Em CYRIX_CACHE_WORKS 1717must also be specified. 1718.It Cd options PCIBIOS 1719Enable support for initializing the 1720.Tn PCI 1721bus using information from the 1722.Tn BIOS . 1723See 1724.Xr pcibios 4 1725for details. 1726.El 1727.Ss isa-specific Options 1728Options specific to 1729.Xr isa 4 1730busses. 1731.Bl -ohang 1732.It Cd options PCIC_ISA_ALLOC_IOBASE=address, PCIC_ISA_ALLOC_IOSIZE=size 1733Control the section of IO bus space used for PCMCIA bus space mapping. 1734Ideally the probed defaults are satisfactory, however in practice 1735that is not always the case. See 1736.Xr pcmcia 4 1737for details. 1738.It Cd options PCIC_ISA_INTR_ALLOC_MASK=mask 1739Controls the allowable interrupts that may be used for 1740.Tn PCMCIA 1741devices. This mask is a logical-or of power-of-2s of allowable 1742interrupts: 1743.Bd -literal -offset 04n 1744.Em "IRQ Val IRQ Val IRQ Val IRQ Val" 1745 0 0x0001 4 0x0010 8 0x0100 12 0x1000 1746 1 0x0002 5 0x0020 9 0x0200 13 0x2000 1747 2 0x0004 6 0x0040 10 0x0400 14 0x4000 1748 3 0x0008 7 0x0080 11 0x0800 15 0x8000 1749.Ed 1750.El 1751.Ss m68k-specific Options 1752.Bl -ohang 1753.It Cd options FPU_EMULATE 1754Include support for MC68881/MC68882 emulator. 1755.It Cd options FPSP 1756Include support for 68040 floating point. 1757.It Cd options M68020,M68030,M68040,M68060 1758Include support for a specific 1759.Tn CPU , 1760at least one (the one you are using) should be specified. 1761.It Cd options M060SP 1762Include software support for 68060. 1763This provides emulation of unimplemented 1764integer instructions as well as emulation of unimplemented floating point 1765instructions and data types and software support for floating point traps. 1766.El 1767.Ss sparc-specific Options 1768.Bl -ohang 1769.It Cd options AUDIO_DEBUG 1770Enable simple event debugging of the logging of the 1771.Xr audio 4 1772device. 1773.It Cd options BLINK 1774Enable blinking of LED. Blink rate is full cycle every N seconds for 1775N \*[Lt] then current load average. See 1776.Xr getloadavg 3 . 1777.\" .It Cd options COLORFONT_CACHE 1778.\" this is totally fucked up.. what does this do? 1779.It Cd options COUNT_SW_LEFTOVERS 1780Count how many times the sw SCSI device has left 3, 2, 1 and 0 in the 1781sw_3_leftover, sw_2_leftover, sw_1_leftover, and sw_0_leftover 1782variables accessible from 1783.Xr ddb 4 . 1784See 1785.Xr sw 4 . 1786.It Cd options DEBUG_ALIGN 1787Adds debugging messages calls when user-requested alignment fault 1788handling happens. 1789.It Cd options DEBUG_EMUL 1790Adds debugging messages calls for emulated floating point and 1791alignment fixing operations. 1792.It Cd options DEBUG_SVR4 1793Prints registers messages calls for emulated SVR4 getcontext and 1794setcontext operations. See 1795.Em options COMPAT_SVR4 . 1796.It Cd options EXTREME_DEBUG 1797Adds debugging functions callable from 1798.Xr ddb 4 . The debug_pagetables, test_region and print_fe_map 1799functions print information about page tables for the SUN4M 1800platforms only. 1801.It Cd options EXTREME_EXTREME_DEBUG 1802Adds extra info to 1803.Em options EXTREME_DEBUG . 1804.It Cd options FPU_CONTEXT 1805Make 1806.Em options COMPAT_SVR4 1807getcontext and setcontext include floating point registers. 1808.It Cd options MAGMA_DEBUG 1809Adds debugging messages to the 1810.Xr magma 4 1811device. 1812.It Cd options RASTERCONS_FULLSCREEN 1813Use the entire screen for the console. 1814.It Cd options RASTERCONS_SMALLFONT 1815Use a the fixed font on the console, instead of the normal font. 1816.It Cd options SUN4 1817Support sun4 class machines. 1818.It Cd options SUN4C 1819Support sun4c class machines. 1820.It Cd options SUN4M 1821Support sun4m class machines. 1822.It Cd options SUN4_MMU3L 1823.\" XXX ??? 1824Enable support for sun4 3-level MMU machines. 1825.It Cd options V9 1826Enable SPARC V9 assembler in 1827.Xr ddb 4 . 1828.El 1829.Ss sparc64-specific Options 1830.Bl -ohang 1831.It Cd options AUDIO_DEBUG 1832Enable simple event debugging of the logging of the 1833.Xr audio 4 1834device. 1835.It Cd options BLINK 1836Enable blinking of LED. Blink rate is full cycle every N seconds for 1837N \*[Lt] then current load average. See 1838.Xr getloadavg 3 . 1839.El 1840.Ss x68k-specific Options 1841.Bl -ohang 1842.It Cd options EXTENDED_MEMORY 1843Include support for extended memory e.g. TS-6BE16 and 060turbo on-board. 1844.It Cd options JUPITER 1845Include support for Jupiter-X MPU accelerator 1846.It Cd options ZSCONSOLE,ZSCN_SPEED=value 1847Use the built-in serial port as the system-console. 1848Speed is specified in bps, defaults to 9600. 1849.It Cd options ITE_KERNEL_ATTR=value 1850Set the kernel message attribute for ITE. 1851Value, an integer, is a logical or of the following values: 1852.Bl -tag -width 4n -compact -offset indent 1853.It 1 1854color inversed 1855.It 2 1856underlined 1857.It 4 1858bolded 1859.El 1860.El 1861.\" The following requests should be uncommented and used where appropriate. 1862.\" .Sh FILES 1863.\" .Sh EXAMPLES 1864.Sh SEE ALSO 1865.Xr gdb 1 , 1866.Xr ktrace 1 , 1867.Xr quota 1 , 1868.Xr gettimeofday 2 , 1869.Xr i386_iopl 2 , 1870.Xr msgctl 2 , 1871.Xr msgget 2 , 1872.Xr msgrcv 2 , 1873.Xr msgsnd 2 , 1874.Xr ntp_adjtime 2 , 1875.Xr ntp_gettime 2 , 1876.Xr semctl 2 , 1877.Xr semget 2 , 1878.Xr semop 2 , 1879.Xr shmat 2 , 1880.Xr shmctl 2 , 1881.Xr shmdt 2 , 1882.Xr shmget 2 , 1883.Xr sysctl 3 , 1884.Xr apm 4 , 1885.Xr ddb 4 , 1886.Xr inet 4 , 1887.Xr iso 4 , 1888.Xr lkm 4 , 1889.Xr md 4 , 1890.Xr ns 4 , 1891.Xr pcibios 4 , 1892.Xr pcmcia 4 , 1893.Xr config 8 , 1894.Xr edquota 8 , 1895.Xr init 8 , 1896.Xr mdsetimage 8 , 1897.Xr mount_cd9660 8 , 1898.Xr mount_fdesc 8 , 1899.Xr mount_kernfs 8 , 1900.Xr mount_lfs 8 , 1901.Xr mount_mfs 8 , 1902.Xr mount_msdos 8 , 1903.Xr mount_nfs 8 , 1904.Xr mount_ntfs 8 , 1905.Xr mount_null 8 , 1906.Xr mount_portal 8 , 1907.Xr mount_procfs 8 , 1908.Xr mount_umap 8 , 1909.Xr mount_union 8 , 1910.Xr mrouted 8 , 1911.Xr newfs_lfs 8 , 1912.Xr ntpd 8 , 1913.Xr quotaon 8 , 1914.Xr rpc.rquotad 8 , 1915.Xr sysctl 8 1916.Sh HISTORY 1917The 1918.Nm 1919man page first appeared in 1920.Nx 1.3 . 1921.Sh BUGS 1922The 1923.Em INET 1924and the 1925.Em VNODEPAGER 1926options should not be required. 1927The 1928.Em EON 1929option should be a pseudo-device, and is also very fragile. 1930