blob: f9e1828cbf30d489bbaa39abb5a8f935e5aa63b3 [file] [log] [blame]
#include <config.h>
#include <malloc.h>
#include <string.h>
#include <mem_malloc.h>
#include <stdio.h>
#include <module.h>
A version of malloc/free/realloc written by Doug Lea and released to the
public domain. Send questions/comments/complaints/performance data
* VERSION 2.6.6 Sun Mar 5 19:10:03 2000 Doug Lea (dl at gee)
Note: There may be an updated version of this malloc obtainable at
Check before installing!
* Why use this malloc?
This is not the fastest, most space-conserving, most portable, or
most tunable malloc ever written. However it is among the fastest
while also being among the most space-conserving, portable and tunable.
Consistent balance across these factors results in a good general-purpose
allocator. For a high-level description, see
* Synopsis of public routines
(Much fuller descriptions are contained in the program documentation below.)
malloc(size_t n);
Return a pointer to a newly allocated chunk of at least n bytes, or null
if no space is available.
free(Void_t* p);
Release the chunk of memory pointed to by p, or no effect if p is null.
realloc(Void_t* p, size_t n);
Return a pointer to a chunk of size n that contains the same data
as does chunk p up to the minimum of (n, p's size) bytes, or null
if no space is available. The returned pointer may or may not be
the same as p. If p is null, equivalent to malloc. Unless the
#define REALLOC_ZERO_BYTES_FREES below is set, realloc with a
size argument of zero (re)allocates a minimum-sized chunk.
memalign(size_t alignment, size_t n);
Return a pointer to a newly allocated chunk of n bytes, aligned
in accord with the alignment argument, which must be a power of
valloc(size_t n);
Equivalent to memalign(pagesize, n), where pagesize is the page
size of the system (or as near to this as can be figured out from
all the includes/defines below.)
pvalloc(size_t n);
Equivalent to valloc(minimum-page-that-holds(n)), that is,
round up n to nearest pagesize.
calloc(size_t unit, size_t quantity);
Returns a pointer to quantity * unit bytes, with all locations
set to zero.
cfree(Void_t* p);
Equivalent to free(p).
malloc_trim(size_t pad);
Release all but pad bytes of freed top-most memory back
to the system. Return 1 if successful, else 0.
malloc_usable_size(Void_t* p);
Report the number usable allocated bytes associated with allocated
chunk p. This may or may not report more bytes than were requested,
due to alignment and minimum size constraints.
Prints brief summary statistics on stderr.
Returns (by copy) a struct containing various summary statistics.
mallopt(int parameter_number, int parameter_value)
Changes one of the tunable parameters described below. Returns
1 if successful in changing the parameter, else 0.
* Vital statistics:
Alignment: 8-byte
8 byte alignment is currently hardwired into the design. This
seems to suffice for all current machines and C compilers.
Assumed pointer representation: 4 or 8 bytes
Code for 8-byte pointers is untested by me but has worked
reliably by Wolfram Gloger, who contributed most of the
changes supporting this.
Assumed size_t representation: 4 or 8 bytes
Note that size_t is allowed to be 4 bytes even if pointers are 8.
Minimum overhead per allocated chunk: 4 or 8 bytes
Each malloced chunk has a hidden overhead of 4 bytes holding size
and status information.
Minimum allocated size: 4-byte ptrs: 16 bytes (including 4 overhead)
8-byte ptrs: 24/32 bytes (including, 4/8 overhead)
When a chunk is freed, 12 (for 4byte ptrs) or 20 (for 8 byte
ptrs but 4 byte size) or 24 (for 8/8) additional bytes are
needed; 4 (8) for a trailing size field
and 8 (16) bytes for free list pointers. Thus, the minimum
allocatable size is 16/24/32 bytes.
Even a request for zero bytes (i.e., malloc(0)) returns a
pointer to something of the minimum allocatable size.
Maximum allocated size: 4-byte size_t: 2^31 - 8 bytes
8-byte size_t: 2^63 - 16 bytes
It is assumed that (possibly signed) size_t bit values suffice to
represent chunk sizes. `Possibly signed' is due to the fact
that `size_t' may be defined on a system as either a signed or
an unsigned type. To be conservative, values that would appear
as negative numbers are avoided.
Requests for sizes with a negative sign bit when the request
size is treaded as a long will return null.
Maximum overhead wastage per allocated chunk: normally 15 bytes
Alignnment demands, plus the minimum allocatable size restriction
make the normal worst-case wastage 15 bytes (i.e., up to 15
more bytes will be allocated than were requested in malloc), with
two exceptions:
1. Because requests for zero bytes allocate non-zero space,
the worst case wastage for a request of zero bytes is 24 bytes.
2. For requests >= mmap_threshold that are serviced via
mmap(), the worst case wastage is 8 bytes plus the remainder
from a system page (the minimal mmap unit); typically 4096 bytes.
* Limitations
Here are some features that are NOT currently supported
* No user-definable hooks for callbacks and the like.
* No automated mechanism for fully checking that all accesses
to malloced memory stay within their bounds.
* No support for compaction.
* Synopsis of compile-time options:
People have reported using previous versions of this malloc on all
versions of Unix, sometimes by tweaking some of the defines
below. It has been tested most extensively on Solaris and
Linux. It is also reported to work on WIN32 platforms.
People have also reported adapting this malloc for use in
stand-alone embedded systems.
The implementation is in straight, hand-tuned ANSI C. Among other
consequences, it uses a lot of macros. Because of this, to be at
all usable, this code should be compiled using an optimizing compiler
(for example gcc -O2) that can simplify expressions and control
__STD_C (default: derived from C compiler defines)
Nonzero if using ANSI-standard C compiler, a C++ compiler, or
a C compiler sufficiently close to ANSI to get away with it.
DEBUG (default: NOT defined)
Define to enable debugging. Adds fairly extensive assertion-based
checking to help track down memory errors, but noticeably slows down
Define this if you think that realloc(p, 0) should be equivalent
to free(p). Otherwise, since malloc returns a unique pointer for
malloc(0), so does realloc(p, 0).
HAVE_MEMCPY (default: defined)
Define if you are not otherwise using ANSI STD C, but still
have memcpy and memset in your C library and want to use them.
Otherwise, simple internal versions are supplied.
USE_MEMCPY (default: 1 if HAVE_MEMCPY is defined, 0 otherwise)
Define as 1 if you want the C library versions of memset and
memcpy called in realloc and calloc (otherwise macro versions are used).
At least on some platforms, the simple macro versions usually
outperform libc versions.
HAVE_MMAP (default: defined as 1)
Define to non-zero to optionally make malloc() use mmap() to
allocate very large blocks.
HAVE_MREMAP (default: defined as 0 unless Linux libc set)
Define to non-zero to optionally make realloc() use mremap() to
reallocate very large blocks.
malloc_getpagesize (default: derived from system #includes)
Either a constant or routine call returning the system page size.
HAVE_USR_INCLUDE_MALLOC_H (default: NOT defined)
Optionally define if you are on a system with a /usr/include/malloc.h
that declares struct mallinfo. It is not at all necessary to
define this even if you do, but will ensure consistency.
INTERNAL_SIZE_T (default: size_t)
Define to a 32-bit type (probably `unsigned int') if you are on a
64-bit machine, yet do not want or need to allow malloc requests of
greater than 2^31 to be handled. This saves space, especially for
very small chunks.
INTERNAL_LINUX_C_LIB (default: NOT defined)
Defined only when compiled as part of Linux libc.
Also note that there is some odd internal name-mangling via defines
(for example, internally, `malloc' is named `mALLOc') needed
when compiling in this case. These look funny but don't otherwise
affect anything.
WIN32 (default: undefined)
Define this on MS win (95, nt) platforms to compile in sbrk emulation.
LACKS_UNISTD_H (default: undefined if not WIN32)
Define this if your system does not have a <unistd.h>.
LACKS_SYS_PARAM_H (default: undefined if not WIN32)
Define this if your system does not have a <sys/param.h>.
MORECORE (default: sbrk)
The name of the routine to call to obtain more memory from the system.
NULL (default: -1)
The value returned upon failure of MORECORE.
True (1) if the routine mapped to MORECORE zeroes out memory (which
holds for sbrk).
Default values of tunable parameters (described in detail below)
controlling interaction with host system routines (sbrk, mmap, etc).
These values may also be changed dynamically via mallopt(). The
preset defaults are those that give best performance for typical
USE_DL_PREFIX (default: undefined)
Prefix all public routines with the string 'dl'. Useful to
quickly avoid procedure declaration conflicts and linker symbol
conflicts with existing memory allocation routines.
#define DEFAULT_TRIM_THRESHOLD (128 * 1024)
M_TRIM_THRESHOLD is the maximum amount of unused top-most memory
to keep before releasing via malloc_trim in free().
Automatic trimming is mainly useful in long-lived programs.
Because trimming via sbrk can be slow on some systems, and can
sometimes be wasteful (in cases where programs immediately
afterward allocate more large chunks) the value should be high
enough so that your overall system performance would improve by
The trim threshold and the mmap control parameters (see below)
can be traded off with one another. Trimming and mmapping are
two different ways of releasing unused memory back to the
system. Between these two, it is often possible to keep
system-level demands of a long-lived program down to a bare
minimum. For example, in one test suite of sessions measuring
the XF86 X server on Linux, using a trim threshold of 128K and a
mmap threshold of 192K led to near-minimal long term resource
If you are using this malloc in a long-lived program, it should
pay to experiment with these values. As a rough guide, you
might set to a value close to the average size of a process
(program) running on your system. Releasing this much memory
would allow such a process to run in memory. Generally, it's
worth it to tune for trimming rather tham memory mapping when a
program undergoes phases where several large chunks are
allocated and released in ways that can reuse each other's
storage, perhaps mixed with phases where there are no such
chunks at all. And in well-behaved long-lived programs,
controlling release of large blocks via trimming versus mapping
is usually faster.
However, in most programs, these parameters serve mainly as
protection against the system-level effects of carrying around
massive amounts of unneeded memory. Since frequent calls to
sbrk, mmap, and munmap otherwise degrade performance, the default
parameters are set to relatively high values that serve only as
The default trim value is high enough to cause trimming only in
fairly extreme (by current memory consumption standards) cases.
It must be greater than page size to have any useful effect. To
disable trimming completely, you can set to (unsigned long)(-1);
#define DEFAULT_TOP_PAD (0)
M_TOP_PAD is the amount of extra `padding' space to allocate or
retain whenever sbrk is called. It is used in two ways internally:
* When sbrk is called to extend the top of the arena to satisfy
a new malloc request, this much padding is added to the sbrk
* When malloc_trim is called automatically from free(),
it is used as the `pad' argument.
In both cases, the actual amount of padding is rounded
so that the end of the arena is always a system page boundary.
The main reason for using padding is to avoid calling sbrk so
often. Having even a small pad greatly reduces the likelihood
that nearly every malloc request during program start-up (or
after trimming) will invoke sbrk, which needlessly wastes
Automatic rounding-up to page-size units is normally sufficient
to avoid measurable overhead, so the default is 0. However, in
systems where sbrk is relatively slow, it can pay to increase
this value, at the expense of carrying around more memory than
the program needs.
#define DEFAULT_MMAP_THRESHOLD (128 * 1024)
M_MMAP_THRESHOLD is the request size threshold for using mmap()
to service a request. Requests of at least this size that cannot
be allocated using already-existing space will be serviced via mmap.
(If enough normal freed space already exists it is used instead.)
Using mmap segregates relatively large chunks of memory so that
they can be individually obtained and released from the host
system. A request serviced through mmap is never reused by any
other request (at least not directly; the system may just so
happen to remap successive requests to the same locations).
Segregating space in this way has the benefit that mmapped space
can ALWAYS be individually released back to the system, which
helps keep the system level memory demands of a long-lived
program low. Mapped memory can never become `locked' between
other chunks, as can happen with normally allocated chunks, which
menas that even trimming via malloc_trim would not release them.
However, it has the disadvantages that:
1. The space cannot be reclaimed, consolidated, and then
used to service later requests, as happens with normal chunks.
2. It can lead to more wastage because of mmap page alignment
3. It causes malloc performance to be more dependent on host
system memory management support routines which may vary in
implementation quality and may impose arbitrary
limitations. Generally, servicing a request via normal
malloc steps is faster than going through a system's mmap.
All together, these considerations should lead you to use mmap
only for relatively large requests.
#define DEFAULT_MMAP_MAX (0)
M_MMAP_MAX is the maximum number of requests to simultaneously
service using mmap. This parameter exists because:
1. Some systems have a limited number of internal tables for
use by mmap.
2. In most systems, overreliance on mmap can degrade overall
3. If a program allocates many large regions, it is probably
better off using normal sbrk-based allocation routines that
can reclaim and reallocate normal heap memory. Using a
small value allows transition into this mode after the
first few allocations.
Setting to 0 disables all use of mmap. If HAVE_MMAP is not set,
the default value is 0, and attempts to set it to non-zero values
in mallopt will fail.
INTERNAL_SIZE_T is the word-size used for internal bookkeeping
of chunk sizes. On a 64-bit machine, you can reduce malloc
overhead by defining INTERNAL_SIZE_T to be a 32 bit `unsigned int'
at the expense of not being able to handle requests greater than
2^31. This limitation is hardly ever a concern; you are encouraged
to set this. However, the default version is the same as size_t.
#define INTERNAL_SIZE_T size_t
REALLOC_ZERO_BYTES_FREES should be set if a call to
realloc with zero bytes should be the same as a call to free.
Some people think it should. Otherwise, since this malloc
returns a unique pointer for malloc(0), so does realloc(p, 0).
Define HAVE_MMAP to optionally make malloc() use mmap() to
allocate very large blocks. These will be returned to the
operating system immediately after a free().
#define HAVE_MMAP 0 /* Not available for barebox */
Define HAVE_MREMAP to make realloc() use mremap() to re-allocate
large blocks. This is currently only possible on Linux with
kernel versions newer than 1.3.77.
#undef HAVE_MREMAP /* Not available for barebox */
This version of malloc supports the standard SVID/XPG mallinfo
routine that returns a struct containing the same kind of
information you can get from malloc_stats. It should work on
any SVID/XPG compliant system that has a /usr/include/malloc.h
defining struct mallinfo. (If you'd like to install such a thing
yourself, cut out the preliminary declarations as described above
and below and save them in a malloc.h file. But there's no
compelling reason to bother to do this.)
The main declaration needed is the mallinfo struct that is returned
(by-copy) by mallinfo(). The SVID/XPG malloinfo struct contains a
bunch of fields, most of which are not even meaningful in this
version of malloc. Some of these fields are are instead filled by
mallinfo() with other numbers that might possibly be of interest.
HAVE_USR_INCLUDE_MALLOC_H should be set if you have a
/usr/include/malloc.h file that includes a declaration of struct
mallinfo. If so, it is included; else an SVID2/XPG2 compliant
version is declared below. These must be precisely the same for
mallinfo() to work.
/* SVID2/XPG mallinfo structure */
struct mallinfo
int arena; /* total space allocated from system */
int ordblks; /* number of non-inuse chunks */
int smblks; /* unused -- always zero */
int hblks; /* number of mmapped regions */
int hblkhd; /* total space in mmapped regions */
int usmblks; /* unused -- always zero */
int fsmblks; /* unused -- always zero */
int uordblks; /* total allocated space */
int fordblks; /* total non-inuse space */
int keepcost; /* top-most, releasable (via malloc_trim) space */
/* SVID2/XPG mallopt options */
#define M_MXFAST 1 /* UNUSED in this malloc */
#define M_NLBLKS 2 /* UNUSED in this malloc */
#define M_GRAIN 3 /* UNUSED in this malloc */
#define M_KEEP 4 /* UNUSED in this malloc */
/* mallopt options that actually do something */
#define M_TOP_PAD -2
#define M_MMAP_MAX -4
Access to system page size. To the extent possible, this malloc
manages memory from the system in page-size units.
The following mechanics for getpagesize were adapted from
bsd/gnu getpagesize.h
#define malloc_getpagesize 4096
Type declarations
struct malloc_chunk
INTERNAL_SIZE_T prev_size; /* Size of previous chunk (if free). */
INTERNAL_SIZE_T size; /* Size in bytes, including overhead. */
struct malloc_chunk *fd; /* double links -- used only if free. */
struct malloc_chunk *bk;
typedef struct malloc_chunk *mchunkptr;
malloc_chunk details:
(The following includes lightly edited explanations by Colin Plumb.)
Chunks of memory are maintained using a `boundary tag' method as
described in e.g., Knuth or Standish. (See the paper by Paul
Wilson for a
survey of such techniques.) Sizes of free chunks are stored both
in the front of each chunk and at the end. This makes
consolidating fragmented chunks into bigger chunks very fast. The
size fields also hold bits representing whether chunks are free or
in use.
An allocated chunk looks like this:
chunk-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Size of previous chunk, if allocated | |
| Size of chunk, in bytes |P|
mem-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| User data starts here... .
. .
. (malloc_usable_space() bytes) .
. |
nextchunk-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Size of chunk |
Where "chunk" is the front of the chunk for the purpose of most of
the malloc code, but "mem" is the pointer that is returned to the
user. "Nextchunk" is the beginning of the next contiguous chunk.
Chunks always begin on even word boundries, so the mem portion
(which is returned to the user) is also on an even word boundary, and
thus double-word aligned.
Free chunks are stored in circular doubly-linked lists, and look like this:
chunk-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Size of previous chunk |
`head:' | Size of chunk, in bytes |P|
mem-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Forward pointer to next chunk in list |
| Back pointer to previous chunk in list |
| Unused space (may be 0 bytes long) .
. .
. |
nextchunk-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
`foot:' | Size of chunk, in bytes |
The P (PREV_INUSE) bit, stored in the unused low-order bit of the
chunk size (which is always a multiple of two words), is an in-use
bit for the *previous* chunk. If that bit is *clear*, then the
word before the current chunk size contains the previous chunk
size, and can be used to find the front of the previous chunk.
(The very first chunk allocated always has this bit set,
preventing access to non-existent (or non-owned) memory.)
Note that the `foot' of the current chunk is actually represented
as the prev_size of the NEXT chunk. (This makes it easier to
deal with alignments etc).
The two exceptions to all this are
1. The special chunk `top', which doesn't bother using the
trailing size field since there is no
next contiguous chunk that would have to index off it. (After
initialization, `top' is forced to always exist. If it would
become less than MINSIZE bytes long, it is replenished via
2. Chunks allocated via mmap, which have the second-lowest-order
bit (IS_MMAPPED) set in their size fields. Because they are
never merged or traversed from any other chunk, they have no
foot size or inuse information.
Available chunks are kept in any of several places (all declared below):
* `av': An array of chunks serving as bin headers for consolidated
chunks. Each bin is doubly linked. The bins are approximately
proportionally (log) spaced. There are a lot of these bins
(128). This may look excessive, but works very well in
practice. All procedures maintain the invariant that no
consolidated chunk physically borders another one. Chunks in
bins are kept in size order, with ties going to the
approximately least recently used chunk.
The chunks in each bin are maintained in decreasing sorted order by
size. This is irrelevant for the small bins, which all contain
the same-sized chunks, but facilitates best-fit allocation for
larger chunks. (These lists are just sequential. Keeping them in
order almost never requires enough traversal to warrant using
fancier ordered data structures.) Chunks of the same size are
linked with the most recently freed at the front, and allocations
are taken from the back. This results in LRU or FIFO allocation
order, which tends to give each chunk an equal opportunity to be
consolidated with adjacent freed chunks, resulting in larger free
chunks and less fragmentation.
* `top': The top-most available chunk (i.e., the one bordering the
end of available memory) is treated specially. It is never
included in any bin, is used only if no other chunk is
available, and is released back to the system if it is very
large (see M_TRIM_THRESHOLD).
* `last_remainder': A bin holding only the remainder of the
most recently split (non-top) chunk. This bin is checked
before other non-fitting chunks, so as to provide better
locality for runs of sequentially allocated chunks.
* Implicitly, through the host system's memory mapping tables.
If supported, requests greater than a threshold are usually
serviced via calls to mmap, and then later released via munmap.
/* sizes, alignments */
#define SIZE_SZ (sizeof(INTERNAL_SIZE_T))
#define MINSIZE (sizeof(struct malloc_chunk))
/* conversion from malloc headers to user pointers, and back */
#define chunk2mem(p) ((void*)((char*)(p) + 2*SIZE_SZ))
#define mem2chunk(mem) ((mchunkptr)((char*)(mem) - 2*SIZE_SZ))
/* pad request bytes into a usable size */
#define request2size(req) \
(((long)((req) + (SIZE_SZ + MALLOC_ALIGN_MASK)) < \
/* Check if m has acceptable alignment */
#define aligned_OK(m) (((unsigned long)((m)) & (MALLOC_ALIGN_MASK)) == 0)
Physical chunk operations
/* size field is or'ed with PREV_INUSE when previous adjacent chunk in use */
#define PREV_INUSE 0x1
/* size field is or'ed with IS_MMAPPED if the chunk was obtained with mmap() */
#define IS_MMAPPED 0x2
/* Bits to mask off when extracting size */
/* Ptr to next physical malloc_chunk. */
#define next_chunk(p) ((mchunkptr)( ((char*)(p)) + ((p)->size & ~PREV_INUSE) ))
/* Ptr to previous physical malloc_chunk */
#define prev_chunk(p)\
((mchunkptr)( ((char*)(p)) - ((p)->prev_size) ))
/* Treat space at ptr + offset as a chunk */
#define chunk_at_offset(p, s) ((mchunkptr)(((char*)(p)) + (s)))
Dealing with use bits
/* extract p's inuse bit */
#define inuse(p)\
((((mchunkptr)(((char*)(p))+((p)->size & ~PREV_INUSE)))->size) & PREV_INUSE)
/* extract inuse bit of previous chunk */
#define prev_inuse(p) ((p)->size & PREV_INUSE)
/* check for mmap()'ed chunk */
#define chunk_is_mmapped(p) ((p)->size & IS_MMAPPED)
/* set/clear chunk as in use without otherwise disturbing */
#define set_inuse(p)\
((mchunkptr)(((char*)(p)) + ((p)->size & ~PREV_INUSE)))->size |= PREV_INUSE
#define clear_inuse(p)\
((mchunkptr)(((char*)(p)) + ((p)->size & ~PREV_INUSE)))->size &= ~(PREV_INUSE)
/* check/set/clear inuse bits in known places */
#define inuse_bit_at_offset(p, s)\
(((mchunkptr)(((char*)(p)) + (s)))->size & PREV_INUSE)
#define set_inuse_bit_at_offset(p, s)\
(((mchunkptr)(((char*)(p)) + (s)))->size |= PREV_INUSE)
#define clear_inuse_bit_at_offset(p, s)\
(((mchunkptr)(((char*)(p)) + (s)))->size &= ~(PREV_INUSE))
Dealing with size fields
/* Get size, ignoring use bits */
#define chunksize(p) ((p)->size & ~(SIZE_BITS))
/* Set size at head, without disturbing its use bit */
#define set_head_size(p, s) ((p)->size = (((p)->size & PREV_INUSE) | (s)))
/* Set size/use ignoring previous bits in header */
#define set_head(p, s) ((p)->size = (s))
/* Set size at footer (only when chunk is not in use) */
#define set_foot(p, s) (((mchunkptr)((char*)(p) + (s)))->prev_size = (s))
The bins, `av_' are an array of pairs of pointers serving as the
heads of (initially empty) doubly-linked lists of chunks, laid out
in a way so that each pair can be treated as if it were in a
malloc_chunk. (This way, the fd/bk offsets for linking bin heads
and chunks are the same).
Bins for sizes < 512 bytes contain chunks of all the same size, spaced
8 bytes apart. Larger bins are approximately logarithmically
spaced. (See the table below.) The `av_' array is never mentioned
directly in the code, but instead via bin access macros.
Bin layout:
64 bins of size 8
32 bins of size 64
16 bins of size 512
8 bins of size 4096
4 bins of size 32768
2 bins of size 262144
1 bin of size what's left
There is actually a little bit of slop in the numbers in bin_index
for the sake of speed. This makes no difference elsewhere.
The special chunks `top' and `last_remainder' get their own bins,
(this is implemented via yet more trickery with the av_ array),
although `top' is never properly linked to its bin since it is
always handled specially.
#define NAV 128 /* number of bins */
typedef struct malloc_chunk *mbinptr;
/* access macros */
#define bin_at(i) ((mbinptr)((char*)&(av_[2*(i) + 2]) - 2*SIZE_SZ))
#define next_bin(b) ((mbinptr)((char*)(b) + 2 * sizeof(mbinptr)))
#define prev_bin(b) ((mbinptr)((char*)(b) - 2 * sizeof(mbinptr)))
The first 2 bins are never indexed. The corresponding av_ cells are instead
used for bookkeeping. This is not to save space, but to simplify
indexing, maintain locality, and avoid some initialization tests.
#define top (bin_at(0)->fd) /* The topmost chunk */
#define last_remainder (bin_at(1)) /* remainder from last split */
Because top initially points to its own bin with initial
zero size, thus forcing extension on the first malloc request,
we avoid having any special code in malloc to check whether
it even exists yet. But we still need to in malloc_extend_top.
#define initial_top ((mchunkptr)(bin_at(0)))
/* Helper macro to initialize bins */
#define IAV(i) bin_at(i), bin_at(i)
static mbinptr av_[NAV * 2 + 2] = {
IAV (0), IAV (1), IAV (2), IAV (3), IAV (4), IAV (5), IAV (6), IAV (7),
IAV (8), IAV (9), IAV (10), IAV (11), IAV (12), IAV (13), IAV (14),
IAV (15),
IAV (16), IAV (17), IAV (18), IAV (19), IAV (20), IAV (21), IAV (22),
IAV (23),
IAV (24), IAV (25), IAV (26), IAV (27), IAV (28), IAV (29), IAV (30),
IAV (31),
IAV (32), IAV (33), IAV (34), IAV (35), IAV (36), IAV (37), IAV (38),
IAV (39),
IAV (40), IAV (41), IAV (42), IAV (43), IAV (44), IAV (45), IAV (46),
IAV (47),
IAV (48), IAV (49), IAV (50), IAV (51), IAV (52), IAV (53), IAV (54),
IAV (55),
IAV (56), IAV (57), IAV (58), IAV (59), IAV (60), IAV (61), IAV (62),
IAV (63),
IAV (64), IAV (65), IAV (66), IAV (67), IAV (68), IAV (69), IAV (70),
IAV (71),
IAV (72), IAV (73), IAV (74), IAV (75), IAV (76), IAV (77), IAV (78),
IAV (79),
IAV (80), IAV (81), IAV (82), IAV (83), IAV (84), IAV (85), IAV (86),
IAV (87),
IAV (88), IAV (89), IAV (90), IAV (91), IAV (92), IAV (93), IAV (94),
IAV (95),
IAV (96), IAV (97), IAV (98), IAV (99), IAV (100), IAV (101), IAV (102),
IAV (103),
IAV (104), IAV (105), IAV (106), IAV (107), IAV (108), IAV (109),
IAV (110), IAV (111),
IAV (112), IAV (113), IAV (114), IAV (115), IAV (116), IAV (117),
IAV (118), IAV (119),
IAV (120), IAV (121), IAV (122), IAV (123), IAV (124), IAV (125),
IAV (126), IAV (127)
/* field-extraction macros */
#define first(b) ((b)->fd)
#define last(b) ((b)->bk)
Indexing into bins
#define bin_index(sz) \
(((((unsigned long)(sz)) >> 9) == 0) ? (((unsigned long)(sz)) >> 3): \
((((unsigned long)(sz)) >> 9) <= 4) ? 56 + (((unsigned long)(sz)) >> 6): \
((((unsigned long)(sz)) >> 9) <= 20) ? 91 + (((unsigned long)(sz)) >> 9): \
((((unsigned long)(sz)) >> 9) <= 84) ? 110 + (((unsigned long)(sz)) >> 12): \
((((unsigned long)(sz)) >> 9) <= 340) ? 119 + (((unsigned long)(sz)) >> 15): \
((((unsigned long)(sz)) >> 9) <= 1364) ? 124 + (((unsigned long)(sz)) >> 18): \
bins for chunks < 512 are all spaced 8 bytes apart, and hold
identically sized chunks. This is exploited in malloc.
#define MAX_SMALLBIN 63
#define smallbin_index(sz) (((unsigned long)(sz)) >> 3)
Requests are `small' if both the corresponding and the next bin are small
#define is_small_request(nb) (nb < MAX_SMALLBIN_SIZE - SMALLBIN_WIDTH)
To help compensate for the large number of bins, a one-level index
structure is used for bin-by-bin searching. `binblocks' is a
one-word bitvector recording whether groups of BINBLOCKWIDTH bins
have any (possibly) non-empty bins, so they can be skipped over
all at once during during traversals. The bits are NOT always
cleared as soon as all bins in a block are empty, but instead only
when all are noticed to be empty during traversal in malloc.
#define BINBLOCKWIDTH 4 /* bins per block */
#define binblocks (bin_at(0)->size) /* bitvector of nonempty blocks */
/* bin<->block macros */
#define idx2binblock(ix) ((unsigned)1 << (ix / BINBLOCKWIDTH))
#define mark_binblock(ii) (binblocks |= idx2binblock(ii))
#define clear_binblock(ii) (binblocks &= ~(idx2binblock(ii)))
/* ----------------------------------------------------------------------- */
/* Other static bookkeeping data */
/* variables holding tunable values */
#ifndef __BAREBOX__
static unsigned long trim_threshold = DEFAULT_TRIM_THRESHOLD;
static unsigned int n_mmaps_max = DEFAULT_MMAP_MAX;
static unsigned long mmap_threshold = DEFAULT_MMAP_THRESHOLD;
static unsigned long top_pad = DEFAULT_TOP_PAD;
/* The first value returned from sbrk */
static char *sbrk_base = (char*)(-1);
/* The maximum memory obtained from system via sbrk */
static unsigned long max_sbrked_mem;
/* The maximum via either sbrk or mmap */
static unsigned long max_total_mem;
/* internal working copy of mallinfo */
static struct mallinfo current_mallinfo;
/* The total memory obtained from system via sbrk */
#define sbrked_mem (current_mallinfo.arena)
/* Tracking mmaps */
static unsigned long mmapped_mem;
Macro-based internal utilities
Linking chunks in bin lists.
Call these only with variables, not arbitrary expressions, as arguments.
Place chunk p of size s in its bin, in size order,
putting it ahead of others of same size.
#define frontlink(P, S, IDX, BK, FD) \
{ \
{ \
IDX = smallbin_index(S); \
mark_binblock(IDX); \
BK = bin_at(IDX); \
FD = BK->fd; \
P->bk = BK; \
P->fd = FD; \
FD->bk = BK->fd = P; \
} \
else \
{ \
IDX = bin_index(S); \
BK = bin_at(IDX); \
FD = BK->fd; \
if (FD == BK) mark_binblock(IDX); \
else \
{ \
while (FD != BK && S < chunksize(FD)) FD = FD->fd; \
BK = FD->bk; \
} \
P->bk = BK; \
P->fd = FD; \
FD->bk = BK->fd = P; \
} \
/* take a chunk off a list */
#define unlink(P, BK, FD) \
{ \
BK = P->bk; \
FD = P->fd; \
FD->bk = BK; \
BK->fd = FD; \
} \
/* Place p as the last remainder */
#define link_last_remainder(P) \
{ \
last_remainder->fd = last_remainder->bk = P; \
P->fd = P->bk = last_remainder; \
/* Clear the last_remainder bin */
#define clear_last_remainder \
(last_remainder->fd = last_remainder->bk = last_remainder)
/* Routines dealing with mmap(). */
Extend the top-most chunk by obtaining memory from system.
Main interface to sbrk (but see also malloc_trim).
static void malloc_extend_top(INTERNAL_SIZE_T nb)
char *brk; /* return value from sbrk */
INTERNAL_SIZE_T front_misalign; /* unusable bytes at front of sbrked space */
INTERNAL_SIZE_T correction; /* bytes for 2nd sbrk call */
char *new_brk; /* return of 2nd sbrk call */
INTERNAL_SIZE_T top_size; /* new size of top chunk */
mchunkptr old_top = top; /* Record state of old top */
INTERNAL_SIZE_T old_top_size = chunksize(old_top);
char *old_end = (char *) (chunk_at_offset(old_top, old_top_size));
/* Pad request with top_pad plus minimal overhead */
INTERNAL_SIZE_T sbrk_size = nb + top_pad + MINSIZE;
unsigned long pagesz = malloc_getpagesize;
/* If not the first time through, round to preserve page boundary */
/* Otherwise, we need to correct to a page size below anyway. */
/* (We also correct below if an intervening foreign sbrk call.) */
if (sbrk_base != (char*)(-1))
sbrk_size = (sbrk_size + (pagesz - 1)) & ~(pagesz - 1);
brk = (char*)(sbrk(sbrk_size));
/* Fail if sbrk failed or if a foreign sbrk call killed our space */
if (brk == (char*)(NULL) || (brk < old_end && old_top != initial_top))
sbrked_mem += sbrk_size;
if (brk == old_end) { /* can just add bytes to current top */
top_size = sbrk_size + old_top_size;
set_head (top, top_size | PREV_INUSE);
} else {
if (sbrk_base == (char*)(-1)) /* First time through. Record base */
sbrk_base = brk;
else /* Someone else called sbrk(). Count those bytes as sbrked_mem. */
sbrked_mem += brk - (char*)old_end;
/* Guarantee alignment of first new chunk made from this space */
front_misalign =
(unsigned long) chunk2mem(brk) & MALLOC_ALIGN_MASK;
if (front_misalign > 0) {
correction = (MALLOC_ALIGNMENT) - front_misalign;
brk += correction;
} else
correction = 0;
/* Guarantee the next brk will be at a page boundary */
correction += ((((unsigned long) (brk + sbrk_size)) +
(pagesz - 1)) & ~(pagesz - 1)) -
((unsigned long) (brk + sbrk_size));
/* Allocate correction */
new_brk = (char*) (sbrk(correction));
if (new_brk == (char*)(NULL))
sbrked_mem += correction;
top = (mchunkptr) brk;
top_size = new_brk - brk + correction;
set_head (top, top_size | PREV_INUSE);
if (old_top != initial_top) {
/* There must have been an intervening foreign sbrk call. */
/* A double fencepost is necessary to prevent consolidation */
/* If not enough space to do this, then user did something very wrong */
if (old_top_size < MINSIZE) {
set_head (top, PREV_INUSE); /* will force null return from malloc */
/* Also keep size a multiple of MALLOC_ALIGNMENT */
old_top_size = (old_top_size -
set_head_size (old_top, old_top_size);
chunk_at_offset (old_top, old_top_size)->size =
chunk_at_offset (old_top, old_top_size + SIZE_SZ)->size =
/* If possible, release the rest. */
if (old_top_size >= MINSIZE)
free(chunk2mem (old_top));
if ((unsigned long) sbrked_mem > (unsigned long) max_sbrked_mem)
max_sbrked_mem = sbrked_mem;
if ((unsigned long) (mmapped_mem + sbrked_mem) > (unsigned long) max_total_mem)
max_total_mem = mmapped_mem + sbrked_mem;
/* Main public routines */
Malloc Algorthim:
The requested size is first converted into a usable form, `nb'.
This currently means to add 4 bytes overhead plus possibly more to
obtain 8-byte alignment and/or to obtain a size of at least
MINSIZE (currently 16 bytes), the smallest allocatable size.
(All fits are considered `exact' if they are within MINSIZE bytes.)
From there, the first successful of the following steps is taken:
1. The bin corresponding to the request size is scanned, and if
a chunk of exactly the right size is found, it is taken.
2. The most recently remaindered chunk is used if it is big
enough. This is a form of (roving) first fit, used only in
the absence of exact fits. Runs of consecutive requests use
the remainder of the chunk used for the previous such request
whenever possible. This limited use of a first-fit style
allocation strategy tends to give contiguous chunks
coextensive lifetimes, which improves locality and can reduce
fragmentation in the long run.
3. Other bins are scanned in increasing size order, using a
chunk big enough to fulfill the request, and splitting off
any remainder. This search is strictly by best-fit; i.e.,
the smallest (with ties going to approximately the least
recently used) chunk that fits is selected.
4. If large enough, the chunk bordering the end of memory
(`top') is split off. (This use of `top' is in accord with
the best-fit search rule. In effect, `top' is treated as
larger (and thus less well fitting) than any other available
chunk since it can be extended to be as large as necessary
(up to system limitations).
5. If the request size meets the mmap threshold and the
system supports mmap, and there are few enough currently
allocated mmapped regions, and a call to mmap succeeds,
the request is allocated via direct memory mapping.
6. Otherwise, the top of memory is extended by
obtaining more space from the system (normally using sbrk,
but definable to anything else via the MORECORE macro).
Memory is gathered from the system (in system page-sized
units) in a way that allows chunks obtained across different
sbrk calls to be consolidated, but does not require
contiguous memory. Thus, it should be safe to intersperse
mallocs with other sbrk calls.
All allocations are made from the the `lowest' part of any found
chunk. (The implementation invariant is that prev_inuse is
always true of any allocated chunk; i.e., that each allocated
chunk borders either a previously allocated and still in-use chunk,
or the base of its memory arena.)
void *malloc(size_t bytes)
mchunkptr victim; /* inspected/selected chunk */
INTERNAL_SIZE_T victim_size; /* its size */
int idx; /* index for bin traversal */
mbinptr bin; /* associated bin */
mchunkptr remainder; /* remainder from a split */
long remainder_size; /* its size */
int remainder_index; /* its bin index */
unsigned long block; /* block traverser bit */
int startidx; /* first bin of a traversed block */
mchunkptr fwd; /* misc temp for linking */
mchunkptr bck; /* misc temp for linking */
mbinptr q; /* misc temp */
if ((long) bytes < 0)
return NULL;
nb = request2size(bytes); /* padded request size; */
/* Check for exact match in a bin */
if (is_small_request(nb)) { /* Faster version for small requests */
idx = smallbin_index(nb);
/* No traversal or size check necessary for small bins. */
q = bin_at(idx);
victim = last(q);
/* Also scan the next one, since it would have a remainder < MINSIZE */
if (victim == q) {
q = next_bin(q);
victim = last(q);
if (victim != q) {
victim_size = chunksize(victim);
unlink(victim, bck, fwd);
set_inuse_bit_at_offset(victim, victim_size);
return chunk2mem(victim);
idx += 2; /* Set for bin scan below. We've already scanned 2 bins. */
} else {
idx = bin_index(nb);
bin = bin_at(idx);
for (victim = last(bin); victim != bin; victim = victim->bk) {
victim_size = chunksize(victim);
remainder_size = victim_size - nb;
if (remainder_size >= (long)MINSIZE) { /* too big */
--idx; /* adjust to rescan below after checking last remainder */
else if (remainder_size >= 0) { /* exact fit */
unlink(victim, bck, fwd);
set_inuse_bit_at_offset(victim, victim_size);
return chunk2mem(victim);
/* Try to use the last split-off remainder */
if ((victim = last_remainder->fd) != last_remainder) {
victim_size = chunksize(victim);
remainder_size = victim_size - nb;
if (remainder_size >= (long)MINSIZE) { /* re-split */
remainder = chunk_at_offset(victim, nb);
set_head(victim, nb | PREV_INUSE);
set_head(remainder, remainder_size | PREV_INUSE);
set_foot(remainder, remainder_size);
return chunk2mem(victim);
if (remainder_size >= 0) { /* exhaust */
set_inuse_bit_at_offset(victim, victim_size);
return chunk2mem(victim);
/* Else place in bin */
frontlink(victim, victim_size, remainder_index, bck, fwd);
If there are any possibly nonempty big-enough blocks,
search for best fitting chunk by scanning bins in blockwidth units.
if ((block = idx2binblock (idx)) <= binblocks) {
/* Get to the first marked block */
if ((block & binblocks) == 0) {
/* force to an even block boundary */
block <<= 1;
while ((block & binblocks) == 0) {
block <<= 1;
/* For each possibly nonempty block ... */
for (;;) {
startidx = idx; /* (track incomplete blocks) */
q = bin = bin_at(idx);
/* For each bin in this block ... */
do {
/* Find and use first big enough chunk ... */
for (victim = last(bin); victim != bin;
victim = victim->bk) {
victim_size = chunksize(victim);
remainder_size = victim_size - nb;
if (remainder_size >= (long)MINSIZE) { /* split */
remainder =
chunk_at_offset (victim,
unlink(victim, bck, fwd);
remainder_size |
return chunk2mem(victim);
} else if (remainder_size >= 0) { /* take */
unlink(victim, bck, fwd);
return chunk2mem(victim);
bin = next_bin (bin);
} while ((++idx & (BINBLOCKWIDTH - 1)) != 0);
/* Clear out the block bit. */
do { /* Possibly backtrack to try to clear a partial block */
if ((startidx & (BINBLOCKWIDTH - 1)) == 0) {
binblocks &= ~block;
q = prev_bin(q);
} while (first(q) == q);
/* Get to the next possibly nonempty block */
if ((block <<= 1) <= binblocks && (block != 0)) {
while ((block & binblocks) == 0) {
block <<= 1;
} else
/* Try to use top chunk */
/* Require that there be a remainder, ensuring top always exists */
if ((remainder_size = chunksize (top) - nb) < (long) MINSIZE) {
/* Try to extend */
if ((remainder_size = chunksize(top) - nb) < (long) MINSIZE)
return NULL; /* propagate failure */
victim = top;
set_head(victim, nb | PREV_INUSE);
top = chunk_at_offset(victim, nb);
set_head(top, remainder_size | PREV_INUSE);
return chunk2mem(victim);
free() algorithm :
1. free(0) has no effect.
2. If the chunk was allocated via mmap, it is release via munmap().
3. If a returned chunk borders the current high end of memory,
it is consolidated into the top, and if the total unused
topmost memory exceeds the trim threshold, malloc_trim is
4. Other chunks are consolidated as they arrive, and
placed in corresponding bins. (This includes the case of
consolidating with the current `last_remainder').
void free(void *mem)
mchunkptr p; /* chunk corresponding to mem */
INTERNAL_SIZE_T hd; /* its head field */
INTERNAL_SIZE_T sz; /* its size */
int idx; /* its bin index */
mchunkptr next; /* next contiguous chunk */
INTERNAL_SIZE_T nextsz; /* its size */
INTERNAL_SIZE_T prevsz; /* size of previous contiguous chunk */
mchunkptr bck; /* misc temp for linking */
mchunkptr fwd; /* misc temp for linking */
int islr; /* track whether merging with last_remainder */
if (!mem) /* free(0) has no effect */
p = mem2chunk(mem);
hd = p->size;
sz = hd & ~PREV_INUSE;
next = chunk_at_offset(p, sz);
nextsz = chunksize(next);
if (next == top) { /* merge with top */
sz += nextsz;
if (!(hd & PREV_INUSE)) { /* consolidate backward */
prevsz = p->prev_size;
p = chunk_at_offset(p, -((long) prevsz));
sz += prevsz;
unlink (p, bck, fwd);
set_head(p, sz | PREV_INUSE);
top = p;
if ((unsigned long) (sz) >= (unsigned long)trim_threshold)
set_head(next, nextsz); /* clear inuse bit */
islr = 0;
if (!(hd & PREV_INUSE)) { /* consolidate backward */
prevsz = p->prev_size;
p = chunk_at_offset(p, -((long) prevsz));
sz += prevsz;
if (p->fd == last_remainder) /* keep as last_remainder */
islr = 1;
unlink(p, bck, fwd);
if (!(inuse_bit_at_offset(next, nextsz))) { /* consolidate forward */
sz += nextsz;
if (!islr && next->fd == last_remainder) { /* re-insert last_remainder */
islr = 1;
} else
unlink(next, bck, fwd);
set_head(p, sz | PREV_INUSE);
set_foot(p, sz);
if (!islr)
frontlink(p, sz, idx, bck, fwd);
Realloc algorithm:
Chunks that were obtained via mmap cannot be extended or shrunk
unless HAVE_MREMAP is defined, in which case mremap is used.
Otherwise, if their reallocation is for additional space, they are
copied. If for less, they are just left alone.
Otherwise, if the reallocation is for additional space, and the
chunk can be extended, it is, else a malloc-copy-free sequence is
taken. There are several different ways that a chunk could be
extended. All are tried:
* Extending forward into following adjacent free chunk.
* Shifting backwards, joining preceding adjacent space
* Both shifting backwards and extending forward.
* Extending into newly sbrked space
Unless the #define REALLOC_ZERO_BYTES_FREES is set, realloc with a
size argument of zero (re)allocates a minimum-sized chunk.
If the reallocation is for less space, and the new request is for
a `small' (<512 bytes) size, then the newly unused space is lopped
off and freed.
The old unix realloc convention of allowing the last-free'd chunk
to be used as an argument to realloc is no longer supported.
I don't know of any programs still relying on this feature,
and allowing it would also allow too many other incorrect
usages of realloc to be sensible.
void *realloc(void *oldmem, size_t bytes)
INTERNAL_SIZE_T nb; /* padded request size */
mchunkptr oldp; /* chunk corresponding to oldmem */
INTERNAL_SIZE_T oldsize; /* its size */
mchunkptr newp; /* chunk to return */
INTERNAL_SIZE_T newsize; /* its size */
void *newmem; /* corresponding user mem */
mchunkptr next; /* next contiguous chunk after oldp */
INTERNAL_SIZE_T nextsize; /* its size */
mchunkptr prev; /* previous contiguous chunk before oldp */
INTERNAL_SIZE_T prevsize; /* its size */
mchunkptr remainder; /* holds split off extra space from newp */
INTERNAL_SIZE_T remainder_size; /* its size */
mchunkptr bck; /* misc temp for linking */
mchunkptr fwd; /* misc temp for linking */
if (bytes == 0) {
return NULL;
if ((long)bytes < 0)
return NULL;
/* realloc of null is supposed to be same as malloc */
if (!oldmem)
return malloc(bytes);
newp = oldp = mem2chunk(oldmem);
newsize = oldsize = chunksize(oldp);
nb = request2size(bytes);
if ((long)(oldsize) < (long)(nb)) {
/* Try expanding forward */
next = chunk_at_offset(oldp, oldsize);
if (next == top || !inuse(next)) {
nextsize = chunksize(next);
/* Forward into top only if a remainder */
if (next == top) {
if ((long)(nextsize + newsize) >=
(long)(nb + MINSIZE)) {
newsize += nextsize;
top = chunk_at_offset(oldp, nb);
set_head (top,
(newsize - nb) | PREV_INUSE);
set_head_size(oldp, nb);
return chunk2mem(oldp);
/* Forward into next chunk */
else if (((long) (nextsize + newsize) >= (long) (nb))) {
unlink(next, bck, fwd);
newsize += nextsize;
goto split;
} else {
next = NULL;
nextsize = 0;
/* Try shifting backwards. */
if (!prev_inuse(oldp)) {
prev = prev_chunk(oldp);
prevsize = chunksize(prev);
/* try forward + backward first to save a later consolidation */
if (next) {
/* into top */
if (next == top) {
if ((long)
(nextsize + prevsize + newsize) >=
(long)(nb + MINSIZE)) {
unlink (prev, bck, fwd);
newp = prev;
newsize += prevsize + nextsize;
newmem = chunk2mem(newp);
memcpy(newmem, oldmem,
oldsize - SIZE_SZ);
top = chunk_at_offset(newp, nb);
(newsize -
nb) | PREV_INUSE);
set_head_size(newp, nb);
return newmem;
/* into next chunk */
else if (((long)(nextsize + prevsize + newsize)
>= (long)(nb))) {
unlink(next, bck, fwd);
unlink(prev, bck, fwd);
newp = prev;
newsize += nextsize + prevsize;
newmem = chunk2mem(newp);
memcpy(newmem, oldmem,
oldsize - SIZE_SZ);
goto split;
/* backward only */
if (prev && (long)(prevsize + newsize) >= (long)nb) {
unlink(prev, bck, fwd);
newp = prev;
newsize += prevsize;
newmem = chunk2mem(newp);
memcpy(newmem, oldmem, oldsize - SIZE_SZ);
goto split;
/* Must allocate */
newmem = malloc(bytes);
if (!newmem) /* propagate failure */
return NULL;
/* Avoid copy if newp is next chunk after oldp. */
/* (This can only happen when new chunk is sbrk'ed.) */
if ((newp = mem2chunk(newmem)) == next_chunk(oldp)) {
newsize += chunksize(newp);
newp = oldp;
goto split;
/* Otherwise copy, free, and exit */
memcpy(newmem, oldmem, oldsize - SIZE_SZ);
return newmem;
split: /* split off extra room in old or expanded chunk */
if (newsize - nb >= MINSIZE) { /* split off remainder */
remainder = chunk_at_offset(newp, nb);
remainder_size = newsize - nb;
set_head_size(newp, nb);
set_head(remainder, remainder_size | PREV_INUSE);
set_inuse_bit_at_offset(remainder, remainder_size);
free (chunk2mem(remainder)); /* let free() deal with it */
} else {
set_head_size(newp, newsize);
set_inuse_bit_at_offset(newp, newsize);
return chunk2mem(newp);
memalign algorithm:
memalign requests more than enough space from malloc, finds a spot
within that chunk that meets the alignment request, and then
possibly frees the leading and trailing space.
The alignment argument must be a power of two. This property is not
checked by memalign, so misuse may result in random runtime errors.
8-byte alignment is guaranteed by normal malloc calls, so don't
bother calling memalign with an argument of 8 or less.
Overreliance on memalign is a sure way to fragment space.
void *memalign(size_t alignment, size_t bytes)
INTERNAL_SIZE_T nb; /* padded request size */
char *m; /* memory returned by malloc call */
mchunkptr p; /* corresponding chunk */
char *brk; /* alignment point within p */
mchunkptr newp; /* chunk to return */
INTERNAL_SIZE_T newsize; /* its size */
INTERNAL_SIZE_T leadsize; /* leading space befor alignment point */
mchunkptr remainder; /* spare room at end to split off */
long remainder_size; /* its size */
if ((long) bytes < 0)
return NULL;
/* If need less alignment than we give anyway, just relay to malloc */
if (alignment <= MALLOC_ALIGNMENT)
return malloc(bytes);
/* Otherwise, ensure that it is at least a minimum chunk size */
if (alignment < MINSIZE)
alignment = MINSIZE;
/* Call malloc with worst case padding to hit alignment. */
nb = request2size(bytes);
m = (char*)(malloc (nb + alignment + MINSIZE));
if (!m)
return NULL; /* propagate failure */
p = mem2chunk(m);
if ((((unsigned long)(m)) % alignment) == 0) { /* aligned */
} else { /* misaligned */
Find an aligned spot inside chunk.
Since we need to give back leading space in a chunk of at
least MINSIZE, if the first calculation places us at
a spot with less than MINSIZE leader, we can move to the
next aligned spot -- we've allocated enough total room so that
this is always possible.
brk = (char*) mem2chunk(((unsigned long) (m + alignment - 1)) &
-((signed) alignment));
if ((long)(brk - (char*)(p)) < MINSIZE)
brk = brk + alignment;
newp = (mchunkptr)brk;
leadsize = brk - (char*)(p);
newsize = chunksize(p) - leadsize;
/* give back leader, use the rest */
set_head(newp, newsize | PREV_INUSE);
set_inuse_bit_at_offset(newp, newsize);
set_head_size(p, leadsize);
p = newp;
/* Also give back spare room at the end */
remainder_size = chunksize(p) - nb;
if (remainder_size >= (long)MINSIZE) {
remainder = chunk_at_offset(p, nb);
set_head(remainder, remainder_size | PREV_INUSE);
set_head_size(p, nb);
free (chunk2mem(remainder));
return chunk2mem(p);
#if 0
* valloc just invokes memalign with alignment argument equal
* to the page size of the system (or as near to this as can
* be figured out from all the includes/defines above.)
void *valloc(size_t bytes)
return memalign(malloc_getpagesize, bytes);
* pvalloc just invokes valloc for the nearest pagesize
* that will accommodate request
void *pvalloc (size_t bytes)
size_t pagesize = malloc_getpagesize;
return memalign(pagesize, (bytes + pagesize - 1) & ~(pagesize - 1));
* calloc calls malloc, then zeroes out the allocated chunk.
void *calloc(size_t n, size_t elem_size)
mchunkptr p;
INTERNAL_SIZE_T sz = n * elem_size;
/* check if expand_top called, in which case don't need to clear */
mchunkptr oldtop = top;
INTERNAL_SIZE_T oldtopsize = chunksize(top);
void *mem = malloc(sz);
if ((long)n < 0)
return NULL;
if (!mem)
return NULL;
else {
p = mem2chunk(mem);
/* Two optional cases in which clearing not necessary */
csz = chunksize(p);
if (p == oldtop && csz > oldtopsize) {
/* clear only the bytes from non-freshly-sbrked memory */
csz = oldtopsize;
memset(mem, 0, csz - SIZE_SZ);
return mem;
* cfree just calls free. It is needed/defined on some systems
* that pair it with calloc, presumably for odd historical reasons.
#if !defined(INTERNAL_LINUX_C_LIB) || !defined(__ELF__)
void cfree(void *mem)
Malloc_trim gives memory back to the system (via negative
arguments to sbrk) if there is unused memory at the `high' end of
the malloc pool. You can call this after freeing large blocks of
memory to potentially reduce the system-level memory requirements
of a program. However, it cannot guarantee to reduce memory. Under
some allocation patterns, some large free blocks of memory will be
locked between two used chunks, so they cannot be given back to
the system.
The `pad' argument to malloc_trim represents the amount of free
trailing space to leave untrimmed. If this argument is zero,
only the minimum amount of memory to maintain internal data
structures will be left (one page or less). Non-zero arguments
can be supplied to maintain enough trailing space to service
future expected allocations without having to re-obtain memory
from the system.
Malloc_trim returns 1 if it actually released any memory, else 0.
int malloc_trim(size_t pad)
long top_size; /* Amount of top-most memory */
long extra; /* Amount to release */
char *current_brk; /* address returned by pre-check sbrk call */
char *new_brk; /* address returned by negative sbrk call */
unsigned long pagesz = malloc_getpagesize;
top_size = chunksize(top);
extra = ((top_size - pad - MINSIZE + (pagesz - 1)) / pagesz -
1) * pagesz;
if (extra < (long)pagesz) /* Not enough memory to release */
return 0;
else {
/* Test to make sure no one else called sbrk */
current_brk = (char*)(sbrk(0));
if (current_brk != (char*)(top) + top_size)
return 0; /* Apparently we don't own memory; must fail */
else {
new_brk = (char *) (sbrk(-extra));
if (new_brk == (char*)(NULL)) { /* sbrk failed? */
/* Try to figure out what we have */
current_brk = (char*)(sbrk (0));
top_size = current_brk - (char*) top;
if (top_size >= (long)MINSIZE) { /* if not, we are very very dead! */
sbrked_mem = current_brk - sbrk_base;
set_head(top, top_size | PREV_INUSE);
return 0;
else {
/* Success. Adjust top accordingly. */
set_head(top, (top_size - extra) | PREV_INUSE);
sbrked_mem -= extra;
return 1;
* malloc_usable_size:
* This routine tells you how many bytes you can actually use in an
* allocated chunk, which may be more than you requested (although
* often not). You can use this many bytes without worrying about
* overwriting other allocated objects. Not a particularly great
* programming practice, but still sometimes useful.
size_t malloc_usable_size(void *mem)
mchunkptr p;
if (!mem)
return 0;
else {
p = mem2chunk(mem);
if (!chunk_is_mmapped(p)) {
if (!inuse(p))
return 0;
return chunksize(p) - SIZE_SZ;
return chunksize(p) - 2 * SIZE_SZ;
/* Utility to update current_mallinfo for malloc_stats and mallinfo() */
static void malloc_update_mallinfo(void)
int i;
mbinptr b;
mchunkptr p;
#ifdef DEBUG
mchunkptr q;
INTERNAL_SIZE_T avail = chunksize(top);
int navail = ((long)(avail) >= (long)MINSIZE) ? 1 : 0;
for (i = 1; i < NAV; ++i) {
b = bin_at (i);
for (p = last(b); p != b; p = p->bk) {
#ifdef DEBUG
for (q = next_chunk(p);
q < top && inuse(q)
&& (long) (chunksize(q)) >= (long)MINSIZE;
q = next_chunk(q))
avail += chunksize(p);
current_mallinfo.ordblks = navail;
current_mallinfo.uordblks = sbrked_mem - avail;
current_mallinfo.fordblks = avail;
current_mallinfo.hblks = n_mmaps;
current_mallinfo.hblkhd = mmapped_mem;
current_mallinfo.keepcost = chunksize(top);
Prints on the amount of space obtain from the system (both
via sbrk and mmap), the maximum amount (which may be more than
current if malloc_trim and/or munmap got called), the maximum
number of simultaneous mmap regions used, and the current number
of bytes allocated via malloc (or realloc, etc) but not yet
freed. (Note that this is the number of bytes allocated, not the
number requested. It will be larger than the number requested
because of alignment and bookkeeping overhead.)
* mallinfo returns a copy of updated current mallinfo.
void malloc_stats(void)
printf("max system bytes = %10u\n", (unsigned int)(max_total_mem));
printf("system bytes = %10u\n",
(unsigned int)(sbrked_mem + mmapped_mem));
printf("in use bytes = %10u\n",
(unsigned int)(current_mallinfo.uordblks + mmapped_mem));
fprintf(stderr, "max mmap regions = %10u\n",
(unsigned int) max_n_mmaps);
mallopt is the general SVID/XPG interface to tunable parameters.
The format is to provide a (parameter-number, parameter-value) pair.
mallopt then sets the corresponding parameter to the argument
value if it can (i.e., so long as the value is meaningful),
and returns 1 if successful else 0.
See descriptions of tunable parameters above.
#ifndef __BAREBOX__
int mallopt(int param_number, int value)
switch (param_number) {
trim_threshold = value;
return 1;
case M_TOP_PAD:
top_pad = value;
return 1;
mmap_threshold = value;
return 1;
case M_MMAP_MAX:
if (value != 0)
return 0;
n_mmaps_max = value;
return 1;
return 0;
V2.6.6 Sun Dec 5 07:42:19 1999 Doug Lea (dl at gee)
* return null for negative arguments
* Added Several WIN32 cleanups from Martin C. Fong <>
* Add 'LACKS_SYS_PARAM_H' for those systems without 'sys/param.h'
(e.g. WIN32 platforms)
* Cleanup up header file inclusion for WIN32 platforms
* Cleanup code to avoid Microsoft Visual C++ compiler complaints
* Add 'USE_DL_PREFIX' to quickly allow co-existence with existing
memory allocation routines
* Set 'malloc_getpagesize' for WIN32 platforms (needs more work)
* Use 'assert' rather than 'ASSERT' in WIN32 code to conform to
usage of 'assert' in non-WIN32 code
* Improve WIN32 'sbrk()' emulation's 'findRegion()' routine to
avoid infinite loop
* Always call 'fREe()' rather than 'free()'
V2.6.5 Wed Jun 17 15:57:31 1998 Doug Lea (dl at gee)
* Fixed ordering problem with boundary-stamping
V2.6.3 Sun May 19 08:17:58 1996 Doug Lea (dl at gee)
* Added pvalloc, as recommended by H.J. Liu
* Added 64bit pointer support mainly from Wolfram Gloger
* Added anonymously donated WIN32 sbrk emulation
* Malloc, calloc, getpagesize: add optimizations from Raymond Nijssen
* malloc_extend_top: fix mask error that caused wastage after
foreign sbrks
* Add linux mremap support code from HJ Liu
V2.6.2 Tue Dec 5 06:52:55 1995 Doug Lea (dl at gee)
* Integrated most documentation with the code.
* Add support for mmap, with help from
Wolfram Gloger (
* Use last_remainder in more cases.
* Pack bins using idea from
* Use ordered bins instead of best-fit threshhold
* Eliminate block-local decls to simplify tracing and debugging.
* Support another case of realloc via move into top
* Fix error occuring when initial sbrk_base not word-aligned.
* Rely on page size for units instead of SBRK_UNIT to
avoid surprises about sbrk alignment conventions.
* Add mallinfo, mallopt. Thanks to Raymond Nijssen
( for the suggestion.
* Add `pad' argument to malloc_trim and top_pad mallopt parameter.
* More precautions for cases where other routines call sbrk,
courtesy of Wolfram Gloger (
* Added macros etc., allowing use in linux libc from
H.J. Lu (
* Inverted this history list
V2.6.1 Sat Dec 2 14:10:57 1995 Doug Lea (dl at gee)
* Re-tuned and fixed to behave more nicely with V2.6.0 changes.
* Removed all preallocation code since under current scheme
the work required to undo bad preallocations exceeds
the work saved in good cases for most test programs.
* No longer use return list or unconsolidated bins since
no scheme using them consistently outperforms those that don't
given above changes.
* Use best fit for very large chunks to prevent some worst-cases.
* Added some support for debugging
V2.6.0 Sat Nov 4 07:05:23 1995 Doug Lea (dl at gee)
* Removed footers when chunks are in use. Thanks to
Paul Wilson ( for the suggestion.
V2.5.4 Wed Nov 1 07:54:51 1995 Doug Lea (dl at gee)
* Added malloc_trim, with help from Wolfram Gloger
V2.5.3 Tue Apr 26 10:16:01 1994 Doug Lea (dl at g)
V2.5.2 Tue Apr 5 16:20:40 1994 Doug Lea (dl at g)
* realloc: try to expand in both directions
* malloc: swap order of clean-bin strategy;
* realloc: only conditionally expand backwards
* Try not to scavenge used bins
* Use bin counts as a guide to preallocation
* Occasionally bin return list chunks in first scan
* Add a few optimizations from
V2.5.1 Sat Aug 14 15:40:43 1993 Doug Lea (dl at g)
* faster bin computation & slightly different binning
* merged all consolidations to one part of malloc proper
(eliminating old malloc_find_space & malloc_clean_bin)
* Scan 2 returns chunks (not just 1)
* Propagate failure in realloc if malloc returns 0
* Add stuff to allow compilation on non-ANSI compilers
V2.5 Sat Aug 7 07:41:59 1993 Doug Lea (dl at
* removed potential for odd address access in prev_chunk
* removed dependency on getpagesize.h
* misc cosmetics and a bit more internal documentation
* anticosmetics: mangled names in macros to evade debugger strangeness
* tested on sparc, hp-700, dec-mips, rs6000
with gcc & native cc (hp, dec only) allowing
Detlefs & Zorn comparison study (in SIGPLAN Notices.)
Trial version Fri Aug 28 13:14:29 1992 Doug Lea (dl at
* Based loosely on libg++-1.2X malloc. (It retains some of the overall
structure of old version, but most details differ.)