uma_zcreate, uma_zalloc, uma_zfree, uma_zdestroy, uma_zone_set_max --
zone allocator
#include <sys/param.h>
#include <sys/queue.h>
#include <vm/uma.h>
uma_zone_t
uma_zcreate(char *name, int size, uma_ctor ctor, uma_dtor dtor,
uma_init uminit, uma_fini fini, int align, u_int16_t flags);
void *
uma_zalloc(uma_zone_t zone, int flags);
void
uma_zfree(uma_zone_t zone, void *item);
void
uma_zdestroy(uma_zone_t zone);
void
uma_zone_set_max(uma_zone_t zone, int nitems);
The zone allocator provides an efficient interface for managing dynamically-sized
collections of items of similar size. The zone allocator can
work with preallocated zones as well as with runtime-allocated ones, and
is therefore available much earlier in the boot process than other memory
management routines.
A zone is an extensible collection of items of identical size. The zone
allocator keeps track of which items are in use and which are not, and
provides functions for allocating items from the zone and for releasing
them back (which makes them available for later use).
The zone allocator stores state information inside the items proper while
they are not allocated, so structures that will be managed by the zone
allocator and wish to use the type stable property of zones by leaving
some fields pre-filled between allocations, must reserve two pointers at
the very beginning for internal use by the zone allocator, as follows:
struct my_item {
struct my_item *z_rsvd1;
struct my_item *z_rsvd2;
/* rest of structure */
};
Alternatively they should assume those entries corrupted after each allocation.
After the first allocation of an item, it will have been cleared
to zeroes, however subsequent allocations will retain the contents as of
the last free, with the exception of the fields mentioned above.
The uma_zcreate() function creates a new zone from which items may then
be allocated from. The name argument is a text name of the zone for
debugging and stats; this memory should not be freed until the zone has
been deallocated.
The ctor and dtor arguments are callback functions that are called by the
uma subsystem at the time of the call to uma_zalloc() and uma_zfree()
respectively. Their purpose is to provide hooks for initializing or
destroying things that need to be done at the time of the allocation or
release of a resource. A good usage for the ctor and dtor callbacks
might be to adjust a global count of the number of objects allocated.
The uminit and fini arguments are used to optimize the allocation of
objects from the zone. They are called by the uma subsystem whenever it
needs to allocate or free several items to satisfy requests or memory
pressure. A good use for the uminit and fini callbacks might be to initialize
and destroy mutexes contained within the object. This would
allow one to re-use already initialized mutexes when an object is
returned from the uma subsystem's object cache. They are not called on
each call to uma_zalloc() and uma_zfree() but rather in a batch mode on
several objects.
To allocate an item from a zone, simply call uma_zalloc() with a pointer
to that zone and set the flags argument to selected flags as documented
in malloc(9). It will return a pointer to an item if successful, or NULL
in the rare case where all items in the zone are in use and the allocator
is unable to grow the zone or when M_NOWAIT is specified.
Items are released back to the zone from which they were allocated by
calling uma_zfree() with a pointer to the zone and a pointer to the item.
Created zones, which are empty, can be destroyed using uma_zdestroy(),
freeing all memory that was allocated for the zone. All items allocated
from the zone with uma_zalloc() must have been freed with uma_zfree()
before.
The purpose of uma_zone_set_max() is to limit the maximum amount of memory
that the system can dedicated toward the zone specified by the zone
argument. The nitems argument gives the upper limit of items in the
zone. This limits the total number of items in the zone which includes:
allocated items, free items and free items in the per-cpu caches. On
systems with more than one CPU it may not be possible to allocate the
specified number of items even when there is no shortage of memory,
because all of the remaining free items may be in the caches of the other
CPUs when the limit is hit.
The uma_zalloc() function returns a pointer to an item, or NULL if the
zone ran out of unused items and the allocator was unable to enlarge it.
malloc(9)
The zone allocator first appeared in FreeBSD 3.0. It was radically
changed in FreeBSD 5.0 to function as a slab allocator.
The zone allocator was written by John S. Dyson. The zone allocator was
rewritten in large parts by Jeff Roberson <[email protected]> to function
as a slab allocator.
This manual page was written by Dag-Erling Coidan Smorgrav
<[email protected]>. Changes for UMA by Jeroen Ruigrok van der Werven
<[email protected]>.
FreeBSD 5.2.1 July 21, 2003 FreeBSD 5.2.1 [ Back ] |
