xref: /freebsd/share/man/man5/fs.5 (revision 2b743a9e9ddc6736208dc8ca1ce06ce64ad20a19)
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 October 31, 2006
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_unrefs;        /* number of unreferenced inodes */
157	int64_t	 fs_sparecon64[16]; /* old rotation block list head */
158	int64_t	 fs_sblockloc;     /* byte offset of standard superblock */
159	struct	csum_total fs_cstotal;  /* cylinder summary information */
160	ufs_time_t fs_time;        /* last time written */
161	int64_t	 fs_size;          /* number of blocks in fs */
162	int64_t	 fs_dsize;         /* number of data blocks in fs */
163	ufs2_daddr_t fs_csaddr;    /* blk addr of cyl grp summary area */
164	int64_t	 fs_pendingblocks; /* blocks in process of being freed */
165	int32_t	 fs_pendinginodes; /* inodes in process of being freed */
166	int32_t	 fs_snapinum[FSMAXSNAP]; /* list of snapshot inode numbers */
167	int32_t	 fs_avgfilesize;   /* expected average file size */
168	int32_t	 fs_avgfpdir;      /* expected # of files per directory */
169	int32_t	 fs_save_cgsize;   /* save real cg size to use fs_bsize */
170	int32_t	 fs_sparecon32[26]; /* reserved for future constants */
171	int32_t  fs_flags;         /* see FS_ flags below */
172	int32_t	 fs_contigsumsize; /* size of cluster summary array */
173	int32_t	 fs_maxsymlinklen; /* max length of an internal symlink */
174	int32_t	 fs_old_inodefmt;  /* format of on-disk inodes */
175	u_int64_t fs_maxfilesize;  /* maximum representable file size */
176	int64_t	 fs_qbmask;        /* ~fs_bmask for use with 64-bit size */
177	int64_t	 fs_qfmask;        /* ~fs_fmask for use with 64-bit size */
178	int32_t	 fs_state;         /* validate fs_clean field */
179	int32_t	 fs_old_postblformat; /* format of positional layout tables */
180	int32_t	 fs_old_nrpos;     /* number of rotational positions */
181	int32_t	 fs_spare5[2];     /* old fs_postbloff */
182	                           /* old fs_rotbloff */
183	int32_t	 fs_magic;         /* magic number */
184};
185
186/*
187 * Filesystem identification
188 */
189#define	FS_UFS1_MAGIC	0x011954    /* UFS1 fast filesystem magic number */
190#define	FS_UFS2_MAGIC	0x19540119  /* UFS2 fast filesystem magic number */
191#define	FS_OKAY		0x7c269d38  /* superblock checksum */
192#define FS_42INODEFMT	-1      /* 4.2BSD inode format */
193#define FS_44INODEFMT	2       /* 4.4BSD inode format */
194
195/*
196 * Preference for optimization.
197 */
198#define FS_OPTTIME	0	/* minimize allocation time */
199#define FS_OPTSPACE	1	/* minimize disk fragmentation */
200.Ed
201.Pp
202Each disk drive contains some number of file systems.
203A file system consists of a number of cylinder groups.
204Each cylinder group has inodes and data.
205.Pp
206A file system is described by its super-block, which in turn
207describes the cylinder groups.
208The super-block is critical
209data and is replicated in each cylinder group to protect against
210catastrophic loss.
211This is done at file system creation
212time and the critical
213super-block data does not change, so the copies need not be
214referenced further unless disaster strikes.
215.Pp
216Addresses stored in inodes are capable of addressing fragments
217of `blocks'.
218File system blocks of at most size
219.Dv MAXBSIZE
220can
221be optionally broken into 2, 4, or 8 pieces, each of which is
222addressable; these pieces may be
223.Dv DEV_BSIZE ,
224or some multiple of
225a
226.Dv DEV_BSIZE
227unit.
228.Pp
229Large files consist of exclusively large data blocks.
230To avoid
231undue wasted disk space, the last data block of a small file is
232allocated as only as many fragments of a large block as are
233necessary.
234The file system format retains only a single pointer
235to such a fragment, which is a piece of a single large block that
236has been divided.
237The size of such a fragment is determinable from
238information in the inode, using the
239.Fn blksize fs ip lbn
240macro.
241.Pp
242The file system records space availability at the fragment level;
243to determine block availability, aligned fragments are examined.
244.Pp
245The root inode is the root of the file system.
246Inode 0 cannot be used for normal purposes and
247historically bad blocks were linked to inode 1,
248thus the root inode is 2 (inode 1 is no longer used for
249this purpose, however numerous dump tapes make this
250assumption, so we are stuck with it).
251.Pp
252The
253.Fa fs_minfree
254element gives the minimum acceptable percentage of file system
255blocks that may be free.
256If the freelist drops below this level
257only the super-user may continue to allocate blocks.
258The
259.Fa fs_minfree
260element
261may be set to 0 if no reserve of free blocks is deemed necessary,
262however severe performance degradations will be observed if the
263file system is run at greater than 90% full; thus the default
264value of
265.Fa fs_minfree
266is 10%.
267.Pp
268Empirically the best trade-off between block fragmentation and
269overall disk utilization at a loading of 90% comes with a
270fragmentation of 8, thus the default fragment size is an eighth
271of the block size.
272.Pp
273The element
274.Fa fs_optim
275specifies whether the file system should try to minimize the time spent
276allocating blocks, or if it should attempt to minimize the space
277fragmentation on the disk.
278If the value of fs_minfree (see above) is less than 10%,
279then the file system defaults to optimizing for space to avoid
280running out of full sized blocks.
281If the value of minfree is greater than or equal to 10%,
282fragmentation is unlikely to be problematical, and
283the file system defaults to optimizing for time.
284.Pp
285.Em Cylinder group related limits :
286Each cylinder keeps track of the availability of blocks at different
287rotational positions, so that sequential blocks can be laid out
288with minimum rotational latency.
289With the default of 8 distinguished
290rotational positions, the resolution of the
291summary information is 2ms for a typical 3600 rpm drive.
292.Pp
293The element
294.Fa fs_old_rotdelay
295gives the minimum number of milliseconds to initiate
296another disk transfer on the same cylinder.
297It is used in determining the rotationally optimal
298layout for disk blocks within a file;
299the default value for
300.Fa fs_old_rotdelay
301is 2ms.
302.Pp
303Each file system has a statically allocated number of inodes.
304An inode is allocated for each
305.Dv NBPI
306bytes of disk space.
307The inode allocation strategy is extremely conservative.
308.Pp
309.Dv MINBSIZE
310is the smallest allowable block size.
311With a
312.Dv MINBSIZE
313of 4096
314it is possible to create files of size
3152^32 with only two levels of indirection.
316.Dv MINBSIZE
317must be big enough to hold a cylinder group block,
318thus changes to
319.Pq Fa struct cg
320must keep its size within
321.Dv MINBSIZE .
322Note that super-blocks are never more than size
323.Dv SBLOCKSIZE .
324.Pp
325The path name on which the file system is mounted is maintained in
326.Fa fs_fsmnt .
327.Dv MAXMNTLEN
328defines the amount of space allocated in
329the super-block for this name.
330The limit on the amount of summary information per file system
331is defined by
332.Dv MAXCSBUFS .
333For a 4096 byte block size, it is currently parameterized for a
334maximum of two million cylinders.
335.Pp
336Per cylinder group information is summarized in blocks allocated
337from the first cylinder group's data blocks.
338These blocks are read in from
339.Fa fs_csaddr
340(size
341.Fa fs_cssize )
342in addition to the super-block.
343.Pp
344.Sy N.B. :
345.Fn sizeof "struct csum"
346must be a power of two in order for
347the
348.Fn fs_cs
349macro to work.
350.Pp
351The
352.Em "Super-block for a file system" :
353The size of the rotational layout tables
354is limited by the fact that the super-block is of size
355.Dv SBLOCKSIZE .
356The size of these tables is
357.Em inversely
358proportional to the block
359size of the file system.
360The size of the tables is
361increased when sector sizes are not powers of two,
362as this increases the number of cylinders
363included before the rotational pattern repeats
364.Pq Fa fs_cpc .
365The size of the rotational layout
366tables is derived from the number of bytes remaining in
367.Pq Fa struct fs .
368.Pp
369The number of blocks of data per cylinder group
370is limited because cylinder groups are at most one block.
371The inode and free block tables
372must fit into a single block after deducting space for
373the cylinder group structure
374.Pq Fa struct cg .
375.Pp
376The
377.Em Inode :
378The inode is the focus of all file activity in the
379.Ux
380file system.
381There is a unique inode allocated
382for each active file,
383each current directory, each mounted-on file,
384text file, and the root.
385An inode is `named' by its device/i-number pair.
386For further information, see the include file
387.In ufs/ufs/inode.h .
388.Sh HISTORY
389A super-block structure named filsys appeared in
390.At v6 .
391The file system described in this manual appeared
392in
393.Bx 4.2 .
394