xref: /freebsd/sbin/mount_nullfs/mount_nullfs.8 (revision 1b6c76a2fe091c74f08427e6c870851025a9cf67)
1.\"
2.\" Copyright (c) 1992, 1993, 1994
3.\"	The Regents of the University of California.  All rights reserved.
4.\"
5.\" This code is derived from software donated to Berkeley by
6.\" John Heidemann of the UCLA Ficus project.
7.\"
8.\"
9.\" Redistribution and use in source and binary forms, with or without
10.\" modification, are permitted provided that the following conditions
11.\" are met:
12.\" 1. Redistributions of source code must retain the above copyright
13.\"    notice, this list of conditions and the following disclaimer.
14.\" 2. Redistributions in binary form must reproduce the above copyright
15.\"    notice, this list of conditions and the following disclaimer in the
16.\"    documentation and/or other materials provided with the distribution.
17.\" 3. All advertising materials mentioning features or use of this software
18.\"    must display the following acknowledgement:
19.\"	This product includes software developed by the University of
20.\"	California, Berkeley and its contributors.
21.\" 4. Neither the name of the University nor the names of its contributors
22.\"    may be used to endorse or promote products derived from this software
23.\"    without specific prior written permission.
24.\"
25.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
26.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
27.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
28.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
29.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
30.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
31.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
32.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
33.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
34.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
35.\" SUCH DAMAGE.
36.\"
37.\"     @(#)mount_null.8	8.6 (Berkeley) 5/1/95
38.\" $FreeBSD$
39.\"
40.Dd May 1, 1995
41.Dt MOUNT_NULLFS 8
42.Os
43.Sh NAME
44.Nm mount_nullfs
45.Nd "mount a loopback filesystem sub-tree; demonstrate the use of a null file system layer"
46.Sh SYNOPSIS
47.Nm
48.Op Fl o Ar options
49.Ar target
50.Ar mount-point
51.Sh DESCRIPTION
52The
53.Nm
54command creates a
55null layer, duplicating a sub-tree of the file system
56name space under another part of the global file system namespace.
57This allows existing files and directories to be accessed
58using a different pathname.
59.Pp
60The primary differences between a virtual copy of the filesystem
61and a symbolic link are that
62.Xr getcwd 3
63functions correctly in the virtual copy, and that other filesystems
64may be mounted on the virtual copy without affecting the original.
65A different device number for the virtual copy is returned by
66.Xr stat 2 ,
67but in other respects it is indistinguishable from the original.
68.Pp
69The
70.Nm
71filesystem differs from a traditional
72loopback file system in two respects: it is implemented using
73a stackable layers techniques, and it's
74.Do null-node Dc Ns s
75stack above
76all lower-layer vnodes, not just over directory vnodes.
77.Pp
78The options are as follows:
79.Bl -tag -width indent
80.It Fl o
81Options are specified with a
82.Fl o
83flag followed by a comma separated string of options.
84See the
85.Xr mount 8
86man page for possible options and their meanings.
87.El
88.Pp
89The null layer has two purposes.
90First, it serves as a demonstration of layering by providing a layer
91which does nothing.
92(It actually does everything the loopback file system does,
93which is slightly more than nothing.)
94Second, the null layer can serve as a prototype layer.
95Since it provides all necessary layer framework,
96new file system layers can be created very easily by starting
97with a null layer.
98.Pp
99The remainder of this man page examines the null layer as a basis
100for constructing new layers.
101.\"
102.\"
103.Sh INSTANTIATING NEW NULL LAYERS
104New null layers are created with
105.Nm .
106.Nm Mount_nullfs
107takes two arguments, the pathname
108of the lower vfs (target-pn) and the pathname where the null
109layer will appear in the namespace (mount-point-pn).  After
110the null layer is put into place, the contents
111of target-pn subtree will be aliased under mount-point-pn.
112.\"
113.\"
114.Sh OPERATION OF A NULL LAYER
115The null layer is the minimum file system layer,
116simply bypassing all possible operations to the lower layer
117for processing there.  The majority of its activity centers
118on the bypass routine, through which nearly all vnode operations
119pass.
120.Pp
121The bypass routine accepts arbitrary vnode operations for
122handling by the lower layer.  It begins by examining vnode
123operation arguments and replacing any null-nodes by their
124lower-layer equivalents.  It then invokes the operation
125on the lower layer.  Finally, it replaces the null-nodes
126in the arguments and, if a vnode is returned by the operation,
127stacks a null-node on top of the returned vnode.
128.Pp
129Although bypass handles most operations,
130.Em vop_getattr ,
131.Em vop_inactive ,
132.Em vop_reclaim ,
133and
134.Em vop_print
135are not bypassed.
136.Em Vop_getattr
137must change the fsid being returned.
138.Em Vop_inactive
139and
140.Em vop_reclaim
141are not bypassed so that
142they can handle freeing null-layer specific data.
143.Em Vop_print
144is not bypassed to avoid excessive debugging
145information.
146.\"
147.\"
148.Sh INSTANTIATING VNODE STACKS
149Mounting associates the null layer with a lower layer,
150in effect stacking two VFSes.  Vnode stacks are instead
151created on demand as files are accessed.
152.Pp
153The initial mount creates a single vnode stack for the
154root of the new null layer.  All other vnode stacks
155are created as a result of vnode operations on
156this or other null vnode stacks.
157.Pp
158New vnode stacks come into existence as a result of
159an operation which returns a vnode.
160The bypass routine stacks a null-node above the new
161vnode before returning it to the caller.
162.Pp
163For example, imagine mounting a null layer with
164.Bd -literal -offset indent
165mount_nullfs /usr/include /dev/layer/null
166.Ed
167Changing directory to
168.Pa /dev/layer/null
169will assign
170the root null-node (which was created when the null layer was mounted).
171Now consider opening
172.Pa sys .
173A vop_lookup would be
174done on the root null-node.  This operation would bypass through
175to the lower layer which would return a vnode representing
176the UFS
177.Pa sys .
178Null_bypass then builds a null-node
179aliasing the UFS
180.Pa sys
181and returns this to the caller.
182Later operations on the null-node
183.Pa sys
184will repeat this
185process when constructing other vnode stacks.
186.\"
187.\"
188.Sh CREATING OTHER FILE SYSTEM LAYERS
189One of the easiest ways to construct new file system layers is to make
190a copy of the null layer, rename all files and variables, and
191then begin modifying the copy.
192.Xr Sed 1
193can be used to easily rename
194all variables.
195.Pp
196The umap layer is an example of a layer descended from the
197null layer.
198.\"
199.\"
200.Sh INVOKING OPERATIONS ON LOWER LAYERS
201There are two techniques to invoke operations on a lower layer
202when the operation cannot be completely bypassed.  Each method
203is appropriate in different situations.  In both cases,
204it is the responsibility of the aliasing layer to make
205the operation arguments "correct" for the lower layer
206by mapping a vnode argument to the lower layer.
207.Pp
208The first approach is to call the aliasing layer's bypass routine.
209This method is most suitable when you wish to invoke the operation
210currently being handled on the lower layer.
211It has the advantage that
212the bypass routine already must do argument mapping.
213An example of this is
214.Em null_getattrs
215in the null layer.
216.Pp
217A second approach is to directly invoke vnode operations on
218the lower layer with the
219.Em VOP_OPERATIONNAME
220interface.
221The advantage of this method is that it is easy to invoke
222arbitrary operations on the lower layer.  The disadvantage
223is that vnode arguments must be manually mapped.
224.\"
225.\"
226.Sh SEE ALSO
227.Xr mount 8
228.Pp
229UCLA Technical Report CSD-910056,
230.Em "Stackable Layers: an Architecture for File System Development" .
231.Sh BUGS
232THIS FILESYSTEM TYPE IS NOT YET FULLY SUPPORTED (READ: IT DOESN'T WORK)
233AND USING IT MAY, IN FACT, DESTROY DATA ON YOUR SYSTEM.  USE AT YOUR
234OWN RISK.  BEWARE OF DOG.  SLIPPERY WHEN WET.
235.Pp
236This code also needs an owner in order to be less dangerous - serious
237hackers can apply by sending mail to hackers@freebsd.org and announcing
238their intent to take it over.
239.Sh HISTORY
240The
241.Nm
242utility first appeared in
243.Bx 4.4 .
244