1.\" Copyright (c) 1983, 1991, 1993 2.\" The Regents of the University of California. All rights reserved. 3.\" 4.\" Redistribution and use in source and binary forms, with or without 5.\" modification, are permitted provided that the following conditions 6.\" are met: 7.\" 1. Redistributions of source code must retain the above copyright 8.\" notice, this list of conditions and the following disclaimer. 9.\" 2. Redistributions in binary form must reproduce the above copyright 10.\" notice, this list of conditions and the following disclaimer in the 11.\" documentation and/or other materials provided with the distribution. 12.\" 3. All advertising materials mentioning features or use of this software 13.\" must display the following acknowledgement: 14.\" This product includes software developed by the University of 15.\" California, Berkeley and its contributors. 16.\" 4. Neither the name of the University nor the names of its contributors 17.\" may be used to endorse or promote products derived from this software 18.\" without specific prior written permission. 19.\" 20.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 21.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 22.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 23.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 24.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 25.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 26.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 27.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 28.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 29.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 30.\" SUCH DAMAGE. 31.\" 32.\" @(#)fs.5 8.2 (Berkeley) 4/19/94 33.\" $FreeBSD$ 34.\" 35.Dd April 19, 1994 36.Dt FS 5 37.Os 38.Sh NAME 39.Nm fs , 40.Nm inode 41.Nd format of file system volume 42.Sh SYNOPSIS 43.In sys/param.h 44.In ufs/ffs/fs.h 45.Pp 46.In sys/types.h 47.In sys/lock.h 48.In ufs/ufs/quota.h 49.In ufs/ufs/inode.h 50.Sh DESCRIPTION 51The files 52.In fs.h 53and 54.In inode.h 55declare several structures, defined variables and macros 56which are used to create and manage the underlying format of 57file system objects on random access devices (disks). 58.Pp 59The block size and number of blocks which 60comprise a file system are parameters of the file system. 61Sectors beginning at 62.Dv BBLOCK 63and continuing for 64.Dv BBSIZE 65are used 66for a disklabel and for some hardware primary 67and secondary bootstrapping programs. 68.Pp 69The actual file system begins at sector 70.Dv SBLOCK 71with the 72.Em super-block 73that is of size 74.Dv SBLOCKSIZE . 75The following structure describes the super-block and is 76from the file 77.In ufs/ffs/fs.h : 78.Bd -literal 79/* 80 * Super block for an FFS filesystem. 81 */ 82struct fs { 83 int32_t fs_firstfield; /* historic filesystem linked list, */ 84 int32_t fs_unused_1; /* used for incore super blocks */ 85 int32_t fs_sblkno; /* offset of super-block in filesys */ 86 int32_t fs_cblkno; /* offset of cyl-block in filesys */ 87 int32_t fs_iblkno; /* offset of inode-blocks in filesys */ 88 int32_t fs_dblkno; /* offset of first data after cg */ 89 int32_t fs_old_cgoffset; /* cylinder group offset in cylinder */ 90 int32_t fs_old_cgmask; /* used to calc mod fs_ntrak */ 91 int32_t fs_old_time; /* last time written */ 92 int32_t fs_old_size; /* number of blocks in fs */ 93 int32_t fs_old_dsize; /* number of data blocks in fs */ 94 int32_t fs_ncg; /* number of cylinder groups */ 95 int32_t fs_bsize; /* size of basic blocks in fs */ 96 int32_t fs_fsize; /* size of frag blocks in fs */ 97 int32_t fs_frag; /* number of frags in a block in fs */ 98/* these are configuration parameters */ 99 int32_t fs_minfree; /* minimum percentage of free blocks */ 100 int32_t fs_old_rotdelay; /* num of ms for optimal next block */ 101 int32_t fs_old_rps; /* disk revolutions per second */ 102/* these fields can be computed from the others */ 103 int32_t fs_bmask; /* ``blkoff'' calc of blk offsets */ 104 int32_t fs_fmask; /* ``fragoff'' calc of frag offsets */ 105 int32_t fs_bshift; /* ``lblkno'' calc of logical blkno */ 106 int32_t fs_fshift; /* ``numfrags'' calc number of frags */ 107/* these are configuration parameters */ 108 int32_t fs_maxcontig; /* max number of contiguous blks */ 109 int32_t fs_maxbpg; /* max number of blks per cyl group */ 110/* these fields can be computed from the others */ 111 int32_t fs_fragshift; /* block to frag shift */ 112 int32_t fs_fsbtodb; /* fsbtodb and dbtofsb shift constant */ 113 int32_t fs_sbsize; /* actual size of super block */ 114 int32_t fs_spare1[2]; /* old fs_csmask */ 115 /* old fs_csshift */ 116 int32_t fs_nindir; /* value of NINDIR */ 117 int32_t fs_inopb; /* value of INOPB */ 118 int32_t fs_old_nspf; /* value of NSPF */ 119/* yet another configuration parameter */ 120 int32_t fs_optim; /* optimization preference, see below */ 121 int32_t fs_old_npsect; /* # sectors/track including spares */ 122 int32_t fs_old_interleave; /* hardware sector interleave */ 123 int32_t fs_old_trackskew; /* sector 0 skew, per track */ 124 int32_t fs_id[2]; /* unique filesystem id */ 125/* sizes determined by number of cylinder groups and their sizes */ 126 int32_t fs_old_csaddr; /* blk addr of cyl grp summary area */ 127 int32_t fs_cssize; /* size of cyl grp summary area */ 128 int32_t fs_cgsize; /* cylinder group size */ 129 int32_t fs_spare2; /* old fs_ntrak */ 130 int32_t fs_old_nsect; /* sectors per track */ 131 int32_t fs_old_spc; /* sectors per cylinder */ 132 int32_t fs_old_ncyl; /* cylinders in filesystem */ 133 int32_t fs_old_cpg; /* cylinders per group */ 134 int32_t fs_ipg; /* inodes per group */ 135 int32_t fs_fpg; /* blocks per group * fs_frag */ 136/* this data must be re-computed after crashes */ 137 struct csum fs_old_cstotal; /* cylinder summary information */ 138/* these fields are cleared at mount time */ 139 int8_t fs_fmod; /* super block modified flag */ 140 int8_t fs_clean; /* filesystem is clean flag */ 141 int8_t fs_ronly; /* mounted read-only flag */ 142 int8_t fs_old_flags; /* old FS_ flags */ 143 u_char fs_fsmnt[MAXMNTLEN]; /* name mounted on */ 144 u_char fs_volname[MAXVOLLEN]; /* volume name */ 145 u_int64_t fs_swuid; /* system-wide uid */ 146 int32_t fs_pad; /* due to alignment of fs_swuid */ 147/* these fields retain the current block allocation info */ 148 int32_t fs_cgrotor; /* last cg searched */ 149 void *fs_ocsp[NOCSPTRS]; /* padding; was list of fs_cs buffers */ 150 u_int8_t *fs_contigdirs; /* # of contiguously allocated dirs */ 151 struct csum *fs_csp; /* cg summary info buffer for fs_cs */ 152 int32_t *fs_maxcluster; /* max cluster in each cyl group */ 153 u_int *fs_active; /* used by snapshots to track fs */ 154 int32_t fs_old_cpc; /* cyl per cycle in postbl */ 155 int32_t fs_maxbsize; /* maximum blocking factor permitted */ 156 int64_t fs_sparecon64[17]; /* old rotation block list head */ 157 int64_t fs_sblockloc; /* byte offset of standard superblock */ 158 struct csum_total fs_cstotal; /* cylinder summary information */ 159 ufs_time_t fs_time; /* last time written */ 160 int64_t fs_size; /* number of blocks in fs */ 161 int64_t fs_dsize; /* number of data blocks in fs */ 162 ufs2_daddr_t fs_csaddr; /* blk addr of cyl grp summary area */ 163 int64_t fs_pendingblocks; /* blocks in process of being freed */ 164 int32_t fs_pendinginodes; /* inodes in process of being freed */ 165 int32_t fs_snapinum[FSMAXSNAP]; /* list of snapshot inode numbers */ 166 int32_t fs_avgfilesize; /* expected average file size */ 167 int32_t fs_avgfpdir; /* expected # of files per directory */ 168 int32_t fs_save_cgsize; /* save real cg size to use fs_bsize */ 169 int32_t fs_sparecon32[26]; /* reserved for future constants */ 170 int32_t fs_flags; /* see FS_ flags below */ 171 int32_t fs_contigsumsize; /* size of cluster summary array */ 172 int32_t fs_maxsymlinklen; /* max length of an internal symlink */ 173 int32_t fs_old_inodefmt; /* format of on-disk inodes */ 174 u_int64_t fs_maxfilesize; /* maximum representable file size */ 175 int64_t fs_qbmask; /* ~fs_bmask for use with 64-bit size */ 176 int64_t fs_qfmask; /* ~fs_fmask for use with 64-bit size */ 177 int32_t fs_state; /* validate fs_clean field */ 178 int32_t fs_old_postblformat; /* format of positional layout tables */ 179 int32_t fs_old_nrpos; /* number of rotational positions */ 180 int32_t fs_spare5[2]; /* old fs_postbloff */ 181 /* old fs_rotbloff */ 182 int32_t fs_magic; /* magic number */ 183}; 184 185/* 186 * Filesystem identification 187 */ 188#define FS_UFS1_MAGIC 0x011954 /* UFS1 fast filesystem magic number */ 189#define FS_UFS2_MAGIC 0x19540119 /* UFS2 fast filesystem magic number */ 190#define FS_OKAY 0x7c269d38 /* superblock checksum */ 191#define FS_42INODEFMT -1 /* 4.2BSD inode format */ 192#define FS_44INODEFMT 2 /* 4.4BSD inode format */ 193 194/* 195 * Preference for optimization. 196 */ 197#define FS_OPTTIME 0 /* minimize allocation time */ 198#define FS_OPTSPACE 1 /* minimize disk fragmentation */ 199.Ed 200.Pp 201Each disk drive contains some number of file systems. 202A file system consists of a number of cylinder groups. 203Each cylinder group has inodes and data. 204.Pp 205A file system is described by its super-block, which in turn 206describes the cylinder groups. The super-block is critical 207data and is replicated in each cylinder group to protect against 208catastrophic loss. This is done at file system creation 209time and the critical 210super-block data does not change, so the copies need not be 211referenced further unless disaster strikes. 212.Pp 213Addresses stored in inodes are capable of addressing fragments 214of `blocks'. File system blocks of at most size 215.Dv MAXBSIZE 216can 217be optionally broken into 2, 4, or 8 pieces, each of which is 218addressable; these pieces may be 219.Dv DEV_BSIZE , 220or some multiple of 221a 222.Dv DEV_BSIZE 223unit. 224.Pp 225Large files consist of exclusively large data blocks. To avoid 226undue wasted disk space, the last data block of a small file is 227allocated as only as many fragments of a large block as are 228necessary. The file system format retains only a single pointer 229to such a fragment, which is a piece of a single large block that 230has been divided. The size of such a fragment is determinable from 231information in the inode, using the 232.Fn blksize fs ip lbn 233macro. 234.Pp 235The file system records space availability at the fragment level; 236to determine block availability, aligned fragments are examined. 237.Pp 238The root inode is the root of the file system. 239Inode 0 can't be used for normal purposes and 240historically bad blocks were linked to inode 1, 241thus the root inode is 2 (inode 1 is no longer used for 242this purpose, however numerous dump tapes make this 243assumption, so we are stuck with it). 244.Pp 245The 246.Fa fs_minfree 247element gives the minimum acceptable percentage of file system 248blocks that may be free. 249If the freelist drops below this level 250only the super-user may continue to allocate blocks. 251The 252.Fa fs_minfree 253element 254may be set to 0 if no reserve of free blocks is deemed necessary, 255however severe performance degradations will be observed if the 256file system is run at greater than 90% full; thus the default 257value of 258.Fa fs_minfree 259is 10%. 260.Pp 261Empirically the best trade-off between block fragmentation and 262overall disk utilization at a loading of 90% comes with a 263fragmentation of 8, thus the default fragment size is an eighth 264of the block size. 265.Pp 266The element 267.Fa fs_optim 268specifies whether the file system should try to minimize the time spent 269allocating blocks, or if it should attempt to minimize the space 270fragmentation on the disk. 271If the value of fs_minfree (see above) is less than 10%, 272then the file system defaults to optimizing for space to avoid 273running out of full sized blocks. 274If the value of minfree is greater than or equal to 10%, 275fragmentation is unlikely to be problematical, and 276the file system defaults to optimizing for time. 277.Pp 278.Em Cylinder group related limits : 279Each cylinder keeps track of the availability of blocks at different 280rotational positions, so that sequential blocks can be laid out 281with minimum rotational latency. 282With the default of 8 distinguished 283rotational positions, the resolution of the 284summary information is 2ms for a typical 3600 rpm drive. 285.Pp 286The element 287.Fa fs_old_rotdelay 288gives the minimum number of milliseconds to initiate 289another disk transfer on the same cylinder. 290It is used in determining the rotationally optimal 291layout for disk blocks within a file; 292the default value for 293.Fa fs_old_rotdelay 294is 2ms. 295.Pp 296Each file system has a statically allocated number of inodes. 297An inode is allocated for each 298.Dv NBPI 299bytes of disk space. 300The inode allocation strategy is extremely conservative. 301.Pp 302.Dv MINBSIZE 303is the smallest allowable block size. 304With a 305.Dv MINBSIZE 306of 4096 307it is possible to create files of size 3082^32 with only two levels of indirection. 309.Dv MINBSIZE 310must be big enough to hold a cylinder group block, 311thus changes to 312.Pq Fa struct cg 313must keep its size within 314.Dv MINBSIZE . 315Note that super-blocks are never more than size 316.Dv SBLOCKSIZE . 317.Pp 318The path name on which the file system is mounted is maintained in 319.Fa fs_fsmnt . 320.Dv MAXMNTLEN 321defines the amount of space allocated in 322the super-block for this name. 323The limit on the amount of summary information per file system 324is defined by 325.Dv MAXCSBUFS . 326For a 4096 byte block size, it is currently parameterized for a 327maximum of two million cylinders. 328.Pp 329Per cylinder group information is summarized in blocks allocated 330from the first cylinder group's data blocks. 331These blocks are read in from 332.Fa fs_csaddr 333(size 334.Fa fs_cssize ) 335in addition to the super-block. 336.Pp 337.Sy N.B. : 338.Fn sizeof "struct csum" 339must be a power of two in order for 340the 341.Fn fs_cs 342macro to work. 343.Pp 344The 345.Em "Super-block for a file system" : 346The size of the rotational layout tables 347is limited by the fact that the super-block is of size 348.Dv SBLOCKSIZE . 349The size of these tables is 350.Em inversely 351proportional to the block 352size of the file system. 353The size of the tables is 354increased when sector sizes are not powers of two, 355as this increases the number of cylinders 356included before the rotational pattern repeats 357.Pq Fa fs_cpc . 358The size of the rotational layout 359tables is derived from the number of bytes remaining in 360.Pq Fa struct fs . 361.Pp 362The number of blocks of data per cylinder group 363is limited because cylinder groups are at most one block. 364The inode and free block tables 365must fit into a single block after deducting space for 366the cylinder group structure 367.Pq Fa struct cg . 368.Pp 369The 370.Em Inode : 371The inode is the focus of all file activity in the 372.Ux 373file system. 374There is a unique inode allocated 375for each active file, 376each current directory, each mounted-on file, 377text file, and the root. 378An inode is `named' by its device/i-number pair. 379For further information, see the include file 380.In ufs/ufs/inode.h . 381.Sh HISTORY 382A super-block structure named filsys appeared in 383.At v6 . 384The file system described in this manual appeared 385in 386.Bx 4.2 . 387