xref: /dragonfly/share/man/man9/zone.9 (revision 984263bc)
1.\"-
2.\" Copyright (c) 2001 Dag-Erling Co�dan Sm�rgrav
3.\" All rights reserved.
4.\"
5.\" Redistribution and use in source and binary forms, with or without
6.\" modification, are permitted provided that the following conditions
7.\" are met:
8.\" 1. Redistributions of source code must retain the above copyright
9.\"    notice, this list of conditions and the following disclaimer.
10.\" 2. Redistributions in binary form must reproduce the above copyright
11.\"    notice, this list of conditions and the following disclaimer in the
12.\"    documentation and/or other materials provided with the distribution.
13.\"
14.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
24.\" SUCH DAMAGE.
25.\"
26.\" $FreeBSD: src/share/man/man9/zone.9,v 1.9.2.4 2002/05/02 20:01:29 asmodai Exp $
27.\"
28.Dd January 27, 2001
29.Dt ZONE 9
30.Os
31.Sh NAME
32.Nm zbootinit ,
33.Nm zinitna ,
34.Nm zinit ,
35.Nm zalloc ,
36.Nm zfree ,
37.Nm zalloci ,
38.Nm zfreei ,
39.Nd zone allocator
40.Sh SYNOPSIS
41.In sys/param.h
42.In sys/queue.h
43.In vm/vm_zone.h
44.Ft void
45.Fn zbootinit "vm_zone_t z" "char *name" "int size" "void *item" "int nitems"
46.Ft int
47.Fn zinitna "vm_zone_t z" "struct vm_object *obj" "char *name" "int size" "int nentries" "int flags" "int zalloc"
48.Ft vm_zone_t
49.Fn zinit "char *name" "int size" "int nentries" "int flags" "int zalloc"
50.Ft void *
51.Fn zalloc "vm_zone_t z"
52.Ft void
53.Fn zfree "vm_zone_t z" "void *item"
54.Ft void *
55.Fn zalloci "vm_zone_t z"
56.Ft void
57.Fn zfreei "vm_zone_t z" "void *item"
58.Sh DESCRIPTION
59The zone allocator provides an efficient interface for managing
60dynamically-sized collections of items of similar size.
61The zone allocator can work with preallocated zones as well as with
62runtime-allocated ones, and is therefore available much earlier in the
63boot process than other memory management routines.
64.Pp
65A zone is an extensible collection of items of identical size.
66The zone allocator keeps track of which items are in use and which
67are not, and provides functions for allocating items from the zone and
68for releasing them back (which makes them available for later use).
69.Pp
70The zone allocator stores state information inside the items proper
71while they are not allocated,
72so structures that will be managed by the zone allocator
73and wish to use the type stable property of zones by leaving some fields
74pre-filled between allocations, must reserve
75two pointers at the very beginning for internal use by the zone
76allocator, as follows:
77.Bd -literal
78struct my_item {
79        struct my_item  *z_rsvd1;
80        struct my_item  *z_rsvd2;
81        /* rest of structure */
82};
83.Ed
84.Pp
85Alternatively they should assume those entries corrupted
86after each allocation.
87After the first allocation of an item,
88it will have been cleared to zeroes, however subsequent allocations
89will retain the contents as of the last free, with the exception of the
90fields mentioned above.
91.Pp
92Zones are created in one of two fashions, depending how far along the
93boot process is.
94.Pp
95If the VM system is fully initialized, a dynamically allocated zone can
96be created using
97.Fn zinit .
98The
99.Fa name
100argument should be a pointer to a short, descriptive name for the
101zone; it is used for statistics and debugging purposes.
102The
103.Fa size
104and
105.Fa nentries
106are the size of the items held by the zone and the initial size (in
107items) of the zone, respectively.
108The
109.Fa flags
110argument should be set to
111.Dv ZONE_INTERRUPT
112if there is a chance that items may be allocated from the zone in
113interrupt context; note that in this case, the zone will never grow
114larger than
115.Fa nentries
116items.
117In all other cases,
118.Fa flags
119should be set to 0.
120The final argument,
121.Fa zalloc ,
122indicates the number of VM pages by which the zone should grow every
123time it fills up.
124.Pp
125If the VM system is not yet fully initialized, the zone allocator
126cannot dynamically allocate VM pages from which to dole out items, so
127the caller needs to provide a static pool of items.
128In this case, the initialization is done in two stages: first,
129.Fn zbootinit
130is called before first use of the zone; later, when the VM system is
131up, the initialization of the zone is completed by calling
132.Fn zinitna .
133.Pp
134The first argument to
135.Fn zbootinit
136is a pointer to a static
137.Vt "struct vm_zone"
138to initialize.
139The second and third are the name of the zone and the size of the
140items it will hold.
141The fourth argument is a pointer to a static array of items from which
142the zone allocator will draw until the zone is fully initialized.
143The
144.Fa nitems
145argument is the number of items in the array.
146.Pp
147The arguments to
148.Fa zinitna
149are the same as for
150.Fa zinit ,
151with the addition of a pointer to the zone to initialize, and a
152pointer to a
153.Vt "struct vm_object"
154from which to allocate pages in the
155.Dv ZONE_INTERRUPT
156case.
157.Pp
158To allocate an item from a zone, simply call
159.Fn zalloc
160with a pointer to that zone; it will return a pointer to an item, or
161.Dv NULL
162in the rare case where all items in the zone are in use and the
163allocator is unable to grow the zone.
164Note that
165.Fn zalloc
166is not interrupt safe.
167For an interrupt safe allocation use
168.Fn zalloci ,
169which has the same semantics as
170.Fn zalloc .
171.Pp
172Items are released back to the zone from which they were allocated by
173calling
174.Fn zfree
175with a pointer to the zone and a pointer to the item.
176Use the corresponding
177.Fn zfreei
178function if the allocation was done using
179.Fn zalloci .
180.Sh RETURN VALUES
181The
182.Fn zinitna
183function returns 1 on success and 0 on failure; the only failure case
184is inability to preallocate address space for an interrupt-safe zone.
185.Pp
186The
187.Fn zinit
188function returns a pointer to a fully initialized
189.Vt "struct vm_zone" ,
190or
191.Dv NULL
192if it was unable to
193.Fn malloc
194a
195.Vt "struct vm_zone"
196or the
197.Dv ZONE_INTERRUPT
198flag was specified and
199.Fn zinitna
200failed to preallocate address space.
201.Pp
202The
203.Fn zalloc
204function returns a pointer to an item, or
205.Dv NULL
206if the zone ran out of unused items and the allocator was unable to
207enlarge it.
208.Sh SEE ALSO
209.Xr malloc 9
210.Sh HISTORY
211The zone allocator first appeared in
212.Fx 3.0 .
213.Sh AUTHORS
214.An -nosplit
215The zone allocator was written by
216.An John S. Dyson .
217.Pp
218This manual page was written by
219.An Dag-Erling Co\(:idan Sm\(/orgrav Aq des@FreeBSD.org .
220