xref: /original-bsd/sbin/newfs/newfs.8 (revision 95ecee29)
1.\" Copyright (c) 1983, 1987, 1991, 1993
2.\"	The Regents of the University of California.  All rights reserved.
3.\"
4.\" %sccs.include.redist.man%
5.\"
6.\"     @(#)newfs.8	8.2 (Berkeley) 12/11/93
7.\"
8.Dd
9.Dt NEWFS 8
10.Os BSD 4.2
11.Sh NAME
12.Nm newfs ,
13.Nm mfs
14.Nd construct a new file system
15.Sh SYNOPSIS
16.Nm newfs
17.Op Fl N
18.Op Ar newfs-options
19.Ar special
20.Nm mount_mfs
21.Op Fl F Ar mount_flags
22.Op Ar newfs-options
23.Ar special node
24.Sh DESCRIPTION
25.Nm Newfs
26replaces the more obtuse
27.Xr mkfs 8
28program.
29Before running
30.Nm newfs
31or
32.Nm mount_mfs ,
33the disk must be labeled using
34.Xr disklabel 8 .
35.Nm Newfs
36builds a file system on the specified special device
37basing its defaults on the information in the disk label.
38Typically the defaults are reasonable, however
39.Nm newfs
40has numerous options to allow the defaults to be selectively overridden.
41.Pp
42.Nm Mount_mfs
43is used to build a file system in virtual memory and then mount it
44on a specified node.
45.Nm Mount_mfs
46exits and the contents of the file system are lost
47when the file system is unmounted.
48If
49.Nm mount_mfs
50is sent a signal while running,
51for example during system shutdown,
52it will attempt to unmount its
53corresponding file system.
54The parameters to
55.Nm mount_mfs
56are the same as those to
57.Nm newfs .
58The special file is only used to read the disk label which provides
59a set of configuration parameters for the memory based file system.
60The special file is typically that of the primary swap area,
61since that is where the file system will be backed up when
62free memory gets low and the memory supporting
63the file system has to be paged.
64.Pp
65The following options define the general layout policies.
66.Bl -tag -width Fl
67.It Fl N
68Causes the file system parameters to be printed out
69without really creating the file system.
70.It Fl O
71Creates a 4.3BSD format filesystem.
72This options is primarily used to build root filesystems
73that can be understood by older boot ROMs.
74.It Fl b Ar block-size
75The block size of the file system in bytes.
76.It Fl f Ar frag-size
77The fragment size of the file system in bytes.
78.It Fl m Ar free space \&%
79The percentage of space reserved from normal users; the minimum
80free space threshold.  The default value used is 10%.
81See
82.Xr tunefs 8
83for more details on how to set this option.
84.It Fl o Ar optimization\ preference
85.Pq ``space'' or ``time''
86The file system can either be instructed to try to minimize the time spent
87allocating blocks, or to try to minimize the space fragmentation on the disk.
88If the value of minfree (see above) is less than 10%,
89the default is to optimize for space;
90if the value of minfree is greater than or equal to 10%,
91the default is to optimize for time.
92See
93.Xr tunefs 8
94for more details on how to set this option.
95.It Fl a Ar maxcontig
96This specifies the maximum number of contiguous blocks that will
97be laid out before forcing a rotational delay (see
98.Fl d
99below).
100The default value is one.
101See
102.Xr tunefs 8
103for more details on how to set this option.
104.It Fl d Ar rotdelay
105This specifies the expected time (in milliseconds)
106to service a transfer completion
107interrupt and initiate a new transfer on the same disk.
108The default is 4 milliseconds.
109See
110.Xr tunefs 8
111for more details on how to set this option.
112.It Fl e Ar maxbpg
113This indicates the maximum number of blocks any single file can
114allocate out of a cylinder group before it is forced to begin
115allocating blocks from another cylinder group.
116The default is about one quarter of the total blocks in a cylinder group.
117See
118.Xr tunefs 8
119for more details on how to set this option.
120.It Fl i Ar number of bytes per inode
121This specifies the density of inodes in the file system.
122The default is to create an inode for each 2048 bytes of data space.
123If fewer inodes are desired, a larger number should be used;
124to create more inodes a smaller number should be given.
125.It Fl c Ar #cylinders/group
126The number of cylinders per cylinder group in a file system.
127The default value used is 16.
128.It Fl s Ar size
129The size of the file system in sectors.
130.Pp
131The following options override the standard sizes for the disk geometry.
132Their default values are taken from the disk label.
133Changing these defaults is useful only when using
134.Nm newfs
135to build a file system whose raw image will eventually be used
136on a different type of disk than the one on which it is initially
137created (for example on a write-once disk).
138Note that changing any of these values from their
139defaults will make it impossible for
140.Xr fsck
141to find the alternate superblocks if the standard superblock is lost.
142.It Fl r Ar revolutions/minute
143The speed of the disk in revolutions per minute.
144.It Fl S Ar sector-size
145The size of a sector in bytes (almost never anything but 512).
146.It Fl u Ar sectors/track
147The number of sectors per track available for data
148allocation by the file system.  This does not
149include sectors reserved at the end of each track
150for bad block replacement (see
151.Fl p
152below).
153.It Fl t Ar #tracks/cylinder
154The number of tracks/cylinder available for data
155allocation by the file system.
156.It Fl p Ar spare sectors per track
157Spare sectors (bad sector replacements) are physical sectors
158that occupy space at the end of each track.
159They are not counted as part of the sectors/track
160.Pq Fl u
161since they are not available to the file system for data allocation.
162.It Fl x Ar spare sectors per cylinder
163Spare sectors (bad sector replacements) are physical sectors
164that occupy space at the end of the last track in the cylinder.
165They are deducted from the sectors/track
166.Pq Fl u
167of the last track of each cylinder
168since they are not available to the file system for data allocation.
169.It Fl l Ar hardware sector interleave
170Used to describe perturbations in the media format to
171compensate for a slow controller.
172Interleave is physical sector interleave on each track,
173specified as the denominator of the ratio:
174.Dl sectors read/sectors passed over
175Thus an interleave of 1/1 implies contiguous layout, while 1/2
176implies logical sector 0 is separated by one sector from logical
177sector 1.
178.It Fl k Ar sector \&0 skew , per track
179Used to describe perturbations in the media format to
180compensate for a slow controller.
181Track skew is the offset of sector 0 on track N
182relative to sector 0 on track N-1 on the same cylinder.
183.Pp
184.El
185The following option applies only to
186.Nm mount_mfs .
187.Bl -tag -width Fl
188.It Fl F Ar mount flags
189Used to pass in a decimal numeric value to be passed
190as mount flags when running as a memory based file system.
191This option is primarily intended for use when
192.Nm mount_mfs
193is started by the
194.Xr mount 8
195command.
196.El
197.Sh SEE ALSO
198.Xr disktab 5 ,
199.Xr fs 5 ,
200.Xr dumpfs 8 ,
201.Xr disklabel 8 ,
202.Xr diskpart 8 ,
203.Xr fsck 8 ,
204.Xr format 8 ,
205.Xr tunefs 8
206.Rs
207.%A M. McKusick
208.%A W. Joy
209.%A S. Leffler
210.%A R. Fabry
211.%T A Fast File System for UNIX ,
212.%J ACM Transactions on Computer Systems 2
213.%V 3
214.%P pp 181-197
215.%D August 1984
216.%O (reprinted in the BSD System Manager's Manual)
217.Re
218.Sh HISTORY
219The
220.Nm
221command appeared in
222.Bx 4.2 .
223