1.\"- 2.\" Copyright (c) 2001 Dag-Erling Coïdan Smørgrav 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 zdestroy , 36.Nm zalloc , 37.Nm zfree 38.Nd zone allocator 39.Sh SYNOPSIS 40.In sys/param.h 41.In sys/queue.h 42.In vm/vm_zone.h 43.Ft void 44.Fn zbootinit "vm_zone_t z" "char *name" "int size" "void *item" "int nitems" 45.Ft int 46.Fn zinitna "vm_zone_t z" "struct vm_object *obj" "char *name" "int size" "int nentries" "int flags" "int zalloc" 47.Ft vm_zone_t 48.Fn zinit "char *name" "int size" "int nentries" "int flags" "int zalloc" 49.Ft void 50.Fn zdestroy "vm_zone_t z" 51.Ft void * 52.Fn zalloc "vm_zone_t z" 53.Ft void 54.Fn zfree "vm_zone_t z" "void *item" 55.Sh DESCRIPTION 56The zone allocator is deprecated. 57Use 58.In sys/objcache.h 59for new developments. 60.Pp 61The zone allocator provides an efficient interface for managing 62dynamically-sized collections of items of similar size. 63The zone allocator can work with preallocated zones as well as with 64runtime-allocated ones, and is therefore available much earlier in the 65boot process than other memory management routines. 66.Pp 67A zone is an extensible collection of items of identical size. 68The zone allocator keeps track of which items are in use and which 69are not, and provides functions for allocating items from the zone and 70for releasing them back (which makes them available for later use). 71.Pp 72The zone allocator stores state information inside the items proper 73while they are not allocated, 74so structures that will be managed by the zone allocator 75and wish to use the type stable property of zones by leaving some fields 76pre-filled between allocations, must reserve 77two pointers at the very beginning for internal use by the zone 78allocator, as follows: 79.Bd -literal 80struct my_item { 81 struct my_item *z_rsvd1; 82 struct my_item *z_rsvd2; 83 /* rest of structure */ 84}; 85.Ed 86.Pp 87Alternatively they should assume those entries corrupted 88after each allocation. 89After the first allocation of an item, 90it will have been cleared to zeroes, however subsequent allocations 91will retain the contents as of the last free, with the exception of the 92fields mentioned above. 93.Pp 94Zones are created in one of two fashions, depending how far along the 95boot process is. 96.Pp 97If the VM system is fully initialized, a dynamically allocated zone can 98be created using 99.Fn zinit . 100The 101.Fa name 102argument should be a pointer to a short, descriptive name for the 103zone; it is used for statistics and debugging purposes. 104The 105.Fa size 106and 107.Fa nentries 108are the size of the items held by the zone and the initial size (in 109items) of the zone, respectively. 110The 111.Fa flags 112argument should have the 113.Dv ZONE_INTERRUPT 114bit set if there is a chance that items may be allocated from the zone in 115interrupt context; note that in this case, the zone will never grow 116larger than 117.Fa nentries 118items. 119The 120.Fa flags 121argument should have the 122.Dv ZONE_DESTROYABLE 123bit set if the zone is to be destroyed with 124.Fn zdestroy . 125The final argument, 126.Fa zalloc , 127indicates the number of VM pages by which the zone should grow every 128time it fills up. 129.Pp 130If the VM system is not yet fully initialized, the zone allocator 131cannot dynamically allocate VM pages from which to dole out items, so 132the caller needs to provide a static pool of items. 133In this case, the initialization is done in two stages: first, 134.Fn zbootinit 135is called before first use of the zone; later, when the VM system is 136up, the initialization of the zone is completed by calling 137.Fn zinitna . 138.Pp 139The first argument to 140.Fn zbootinit 141is a pointer to a static 142.Vt "struct vm_zone" 143to initialize. 144The second and third are the name of the zone and the size of the 145items it will hold. 146The fourth argument is a pointer to a static array of items from which 147the zone allocator will draw until the zone is fully initialized. 148The 149.Fa nitems 150argument is the number of items in the array. 151.Pp 152The arguments to 153.Fa zinitna 154are the same as for 155.Fa zinit , 156with the addition of a pointer to the zone to initialize, and a 157pointer to a 158.Vt "struct vm_object" 159from which to allocate pages in the 160.Dv ZONE_INTERRUPT 161case. 162.Pp 163To release all the memory allocated for a zone, call 164.Fn zdestroy . 165Only zones created with 166.Fn zinit 167and with the 168.Dv ZONE_DESTROYABLE 169flag can be destroyed. 170.Pp 171To allocate an item from a zone, simply call 172.Fn zalloc 173with a pointer to that zone; it will return a pointer to an item, or 174.Dv NULL 175in the rare case where all items in the zone are in use and the 176allocator is unable to grow the zone. 177.Pp 178Items are released back to the zone from which they were allocated by 179calling 180.Fn zfree 181with a pointer to the zone and a pointer to the item. 182.Pp 183The zone functions are not SMP-safe. 184The MP lock must be held while calling them. 185.Sh RETURN VALUES 186The 187.Fn zinitna 188function returns 1 on success and 0 on failure; the only failure case 189is inability to preallocate address space for an interrupt-safe zone. 190.Pp 191The 192.Fn zinit 193function returns a pointer to a fully initialized 194.Vt "struct vm_zone" , 195or 196.Dv NULL 197if it was unable to 198.Fn kmalloc 199a 200.Vt "struct vm_zone" 201or the 202.Dv ZONE_INTERRUPT 203flag was specified and 204.Fn zinitna 205failed to preallocate address space. 206.Pp 207The 208.Fn zalloc 209function returns a pointer to an item, or 210.Dv NULL 211if the zone ran out of unused items and the allocator was unable to 212enlarge it. 213.Sh SEE ALSO 214.Xr memory 9 215.Sh HISTORY 216The zone allocator first appeared in 217.Fx 3.0 . 218.Sh AUTHORS 219.An -nosplit 220The zone allocator was written by 221.An John S. Dyson . 222.Pp 223This manual page was written by 224.An Dag-Erling Co\(:idan Sm\(/orgrav Aq Mt des@FreeBSD.org . 225