1MMAP(3P) POSIX Programmer's Manual MMAP(3P)
2
6 This manual page is part of the POSIX Programmer's Manual. The Linux
7 implementation of this interface may differ (consult the corresponding
8 Linux manual page for details of Linux behavior), or the interface may
9 not be implemented on Linux.
10
13 mmap — map pages of memory
14
16 #include <sys/mman.h>
17 void *mmap(void *addr, size_t len, int prot, int flags,
19 int fildes, off_t off);
20
22 The mmap() function shall establish a mapping between an address space
23 of a process and a memory object.
24 The mmap() function shall be supported for the following memory
26 objects:
27 * Regular files
29 * Shared memory objects
31 * Typed memory objects
33 Support for any other type of file is unspecified.
35 The format of the call is as follows:
37 pa=mmap(addr, len, prot, flags, fildes, off);
39 The mmap() function shall establish a mapping between the address space
41 of the process at an address pa for len bytes to the memory object rep‐
42 resented by the file descriptor fildes at offset off for len bytes. The
43 value of pa is an implementation-defined function of the parameter addr
44 and the values of flags, further described below. A successful mmap()
45 call shall return pa as its result. The address range starting at pa
46 and continuing for len bytes shall be legitimate for the possible (not
47 necessarily current) address space of the process. The range of bytes
48 starting at off and continuing for len bytes shall be legitimate for
49 the possible (not necessarily current) offsets in the memory object
50 represented by fildes.
51 If fildes represents a typed memory object opened with either the
53 POSIX_TYPED_MEM_ALLOCATE flag or the POSIX_TYPED_MEM_ALLOCATE_CONTIG
54 flag, the memory object to be mapped shall be that portion of the typed
55 memory object allocated by the implementation as specified below. In
56 this case, if off is non-zero, the behavior of mmap() is undefined. If
57 fildes refers to a valid typed memory object that is not accessible
58 from the calling process, mmap() shall fail.
59 The mapping established by mmap() shall replace any previous mappings
61 for those whole pages containing any part of the address space of the
62 process starting at pa and continuing for len bytes.
63 If the size of the mapped file changes after the call to mmap() as a
65 result of some other operation on the mapped file, the effect of refer‐
66 ences to portions of the mapped region that correspond to added or
67 removed portions of the file is unspecified.
68 If len is zero, mmap() shall fail and no mapping shall be established.
70 The parameter prot determines whether read, write, execute, or some
72 combination of accesses are permitted to the data being mapped. The
73 prot shall be either PROT_NONE or the bitwise-inclusive OR of one or
74 more of the other flags in the following table, defined in the
75 <sys/mman.h> header.
76 ┌──────────────────┬──────────────────────────┐
78 │Symbolic Constant │ Description │
79 ├──────────────────┼──────────────────────────┤
80 │PROT_READ │ Data can be read. │
81 │PROT_WRITE │ Data can be written. │
82 │PROT_EXEC │ Data can be executed. │
83 │PROT_NONE │ Data cannot be accessed. │
84 └──────────────────┴──────────────────────────┘
85 If an implementation cannot support the combination of access types
86 specified by prot, the call to mmap() shall fail.
87 An implementation may permit accesses other than those specified by
89 prot; however, the implementation shall not permit a write to succeed
90 where PROT_WRITE has not been set and shall not permit any access where
91 PROT_NONE alone has been set. The implementation shall support at least
92 the following values of prot: PROT_NONE, PROT_READ, PROT_WRITE, and the
93 bitwise-inclusive OR of PROT_READ and PROT_WRITE. The file descriptor
94 fildes shall have been opened with read permission, regardless of the
95 protection options specified. If PROT_WRITE is specified, the applica‐
96 tion shall ensure that it has opened the file descriptor fildes with
97 write permission unless MAP_PRIVATE is specified in the flags parameter
98 as described below.
99 The parameter flags provides other information about the handling of
101 the mapped data. The value of flags is the bitwise-inclusive OR of
102 these options, defined in <sys/mman.h>:
103 ┌──────────────────┬─────────────────────────┐
105 │Symbolic Constant │ Description │
106 ├──────────────────┼─────────────────────────┤
107 │MAP_SHARED │ Changes are shared. │
108 │MAP_PRIVATE │ Changes are private. │
109 │MAP_FIXED │ Interpret addr exactly. │
110 └──────────────────┴─────────────────────────┘
111 It is implementation-defined whether MAP_FIXED shall be supported.
112 MAP_FIXED shall be supported on XSI-conformant systems.
113 MAP_SHARED and MAP_PRIVATE describe the disposition of write references
115 to the memory object. If MAP_SHARED is specified, write references
116 shall change the underlying object. If MAP_PRIVATE is specified, modi‐
117 fications to the mapped data by the calling process shall be visible
118 only to the calling process and shall not change the underlying object.
119 It is unspecified whether modifications to the underlying object done
120 after the MAP_PRIVATE mapping is established are visible through the
121 MAP_PRIVATE mapping. Either MAP_SHARED or MAP_PRIVATE can be specified,
122 but not both. The mapping type is retained across fork().
123 The state of synchronization objects such as mutexes, semaphores, bar‐
125 riers, and conditional variables placed in shared memory mapped with
126 MAP_SHARED becomes undefined when the last region in any process con‐
127 taining the synchronization object is unmapped.
128 When fildes represents a typed memory object opened with either the
130 POSIX_TYPED_MEM_ALLOCATE flag or the POSIX_TYPED_MEM_ALLOCATE_CONTIG
131 flag, mmap() shall, if there are enough resources available, map len
132 bytes allocated from the corresponding typed memory object which were
133 not previously allocated to any process in any processor that may
134 access that typed memory object. If there are not enough resources
135 available, the function shall fail. If fildes represents a typed memory
136 object opened with the POSIX_TYPED_MEM_ALLOCATE_CONTIG flag, these
137 allocated bytes shall be contiguous within the typed memory object. If
138 fildes represents a typed memory object opened with the
139 POSIX_TYPED_MEM_ALLOCATE flag, these allocated bytes may be composed of
140 non-contiguous fragments within the typed memory object. If fildes rep‐
141 resents a typed memory object opened with neither the
142 POSIX_TYPED_MEM_ALLOCATE_CONTIG flag nor the POSIX_TYPED_MEM_ALLOCATE
143 flag, len bytes starting at offset off within the typed memory object
144 are mapped, exactly as when mapping a file or shared memory object. In
145 this case, if two processes map an area of typed memory using the same
146 off and len values and using file descriptors that refer to the same
147 memory pool (either from the same port or from a different port), both
148 processes shall map the same region of storage.
149 When MAP_FIXED is set in the flags argument, the implementation is
151 informed that the value of pa shall be addr, exactly. If MAP_FIXED is
152 set, mmap() may return MAP_FAILED and set errno to [EINVAL]. If a
153 MAP_FIXED request is successful, the mapping established by mmap()
154 replaces any previous mappings for the pages in the range [pa,pa+len)
155 of the process.
156 When MAP_FIXED is not set, the implementation uses addr in an implemen‐
158 tation-defined manner to arrive at pa. The pa so chosen shall be an
159 area of the address space that the implementation deems suitable for a
160 mapping of len bytes to the file. All implementations interpret an addr
161 value of 0 as granting the implementation complete freedom in selecting
162 pa, subject to constraints described below. A non-zero value of addr is
163 taken to be a suggestion of a process address near which the mapping
164 should be placed. When the implementation selects a value for pa, it
165 never places a mapping at address 0, nor does it replace any extant
166 mapping.
167 If MAP_FIXED is specified and addr is non-zero, it shall have the same
169 remainder as the off parameter, modulo the page size as returned by
170 sysconf() when passed _SC_PAGESIZE or _SC_PAGE_SIZE. The implementation
171 may require that off is a multiple of the page size. If MAP_FIXED is
172 specified, the implementation may require that addr is a multiple of
173 the page size. The system performs mapping operations over whole pages.
174 Thus, while the parameter len need not meet a size or alignment con‐
175 straint, the system shall include, in any mapping operation, any par‐
176 tial page specified by the address range starting at pa and continuing
177 for len bytes.
178 The system shall always zero-fill any partial page at the end of an
180 object. Further, the system shall never write out any modified portions
181 of the last page of an object which are beyond its end. References
182 within the address range starting at pa and continuing for len bytes to
183 whole pages following the end of an object shall result in delivery of
184 a SIGBUS signal.
185 An implementation may generate SIGBUS signals when a reference would
187 cause an error in the mapped object, such as out-of-space condition.
188 The mmap() function shall add an extra reference to the file associated
190 with the file descriptor fildes which is not removed by a subsequent
191 close() on that file descriptor. This reference shall be removed when
192 there are no more mappings to the file.
193 The last data access timestamp of the mapped file may be marked for
195 update at any time between the mmap() call and the corresponding mun‐
196 map() call. The initial read or write reference to a mapped region
197 shall cause the file's last data access timestamp to be marked for
198 update if it has not already been marked for update.
199 The last data modification and last file status change timestamps of a
201 file that is mapped with MAP_SHARED and PROT_WRITE shall be marked for
202 update at some point in the interval between a write reference to the
203 mapped region and the next call to msync() with MS_ASYNC or MS_SYNC for
204 that portion of the file by any process. If there is no such call and
205 if the underlying file is modified as a result of a write reference,
206 then these timestamps shall be marked for update at some time after the
207 write reference.
208 There may be implementation-defined limits on the number of memory
210 regions that can be mapped (per process or per system).
211 If such a limit is imposed, whether the number of memory regions that
213 can be mapped by a process is decreased by the use of shmat() is imple‐
214 mentation-defined.
215 If mmap() fails for reasons other than [EBADF], [EINVAL], or [ENOTSUP],
217 some of the mappings in the address range starting at addr and continu‐
218 ing for len bytes may have been unmapped.
219
221 Upon successful completion, the mmap() function shall return the
222 address at which the mapping was placed (pa); otherwise, it shall
223 return a value of MAP_FAILED and set errno to indicate the error. The
224 symbol MAP_FAILED is defined in the <sys/mman.h> header. No successful
225 return from mmap() shall return the value MAP_FAILED.
226
228 The mmap() function shall fail if:
229 EACCES The fildes argument is not open for read, regardless of the pro‐
231 tection specified, or fildes is not open for write and
232 PROT_WRITE was specified for a MAP_SHARED type mapping.
233 EAGAIN The mapping could not be locked in memory, if required by mlock‐
235 all(), due to a lack of resources.
236 EBADF The fildes argument is not a valid open file descriptor.
238 EINVAL The value of len is zero.
240 EINVAL The value of flags is invalid (neither MAP_PRIVATE nor
242 MAP_SHARED is set).
243 EMFILE The number of mapped regions would exceed an implementation-
245 defined limit (per process or per system).
246 ENODEV The fildes argument refers to a file whose type is not supported
248 by mmap().
249 ENOMEM MAP_FIXED was specified, and the range [addr,addr+len) exceeds
251 that allowed for the address space of a process; or, if
252 MAP_FIXED was not specified and there is insufficient room in
253 the address space to effect the mapping.
254 ENOMEM The mapping could not be locked in memory, if required by mlock‐
256 all(), because it would require more space than the system is
257 able to supply.
258 ENOMEM Not enough unallocated memory resources remain in the typed mem‐
260 ory object designated by fildes to allocate len bytes.
261 ENOTSUP
263 MAP_FIXED or MAP_PRIVATE was specified in the flags argument and
264 the implementation does not support this functionality.
265 The implementation does not support the combination of
267 accesses requested in the prot argument.
268 ENXIO Addresses in the range [off,off+len) are invalid for the object
270 specified by fildes.
271 ENXIO MAP_FIXED was specified in flags and the combination of addr,
273 len, and off is invalid for the object specified by fildes.
274 ENXIO The fildes argument refers to a typed memory object that is not
276 accessible from the calling process.
277 EOVERFLOW
279 The file is a regular file and the value of off plus len exceeds
280 the offset maximum established in the open file description
281 associated with fildes.
282 The mmap() function may fail if:
284 EINVAL The addr argument (if MAP_FIXED was specified) or off is not a
286 multiple of the page size as returned by sysconf(), or is con‐
287 sidered invalid by the implementation.
288 The following sections are informative.
290
292 None.
293
295 Use of mmap() may reduce the amount of memory available to other memory
296 allocation functions.
297 Use of MAP_FIXED may result in unspecified behavior in further use of
299 malloc() and shmat(). The use of MAP_FIXED is discouraged, as it may
300 prevent an implementation from making the most effective use of
301 resources. Most implementations require that off and addr are multiples
302 of the page size as returned by sysconf().
303 The application must ensure correct synchronization when using mmap()
305 in conjunction with any other file access method, such as read() and
306 write(), standard input/output, and shmat().
307 The mmap() function allows access to resources via address space manip‐
309 ulations, instead of read()/write(). Once a file is mapped, all a
310 process has to do to access it is use the data at the address to which
311 the file was mapped. So, using pseudo-code to illustrate the way in
312 which an existing program might be changed to use mmap(), the follow‐
313 ing:
314 fildes = open(...)
316 lseek(fildes, some_offset)
317 read(fildes, buf, len)
318 /* Use data in buf. */
319 becomes:
321 fildes = open(...)
323 address = mmap(0, len, PROT_READ, MAP_PRIVATE, fildes, some_offset)
324 /* Use data at address. */
325
327 After considering several other alternatives, it was decided to adopt
328 the mmap() definition found in SVR4 for mapping memory objects into
329 process address spaces. The SVR4 definition is minimal, in that it
330 describes only what has been built, and what appears to be necessary
331 for a general and portable mapping facility.
332 Note that while mmap() was first designed for mapping files, it is
334 actually a general-purpose mapping facility. It can be used to map any
335 appropriate object, such as memory, files, devices, and so on, into the
336 address space of a process.
337 When a mapping is established, it is possible that the implementation
339 may need to map more than is requested into the address space of the
340 process because of hardware requirements. An application, however, can‐
341 not count on this behavior. Implementations that do not use a paged
342 architecture may simply allocate a common memory region and return the
343 address of it; such implementations probably do not allocate any more
344 than is necessary. References past the end of the requested area are
345 unspecified.
346 If an application requests a mapping that would overlay existing map‐
348 pings in the process, it might be desirable that an implementation
349 detect this and inform the application. However, the default, portable
350 (not MAP_FIXED) operation does not overlay existing mappings. On the
351 other hand, if the program specifies a fixed address mapping (which
352 requires some implementation knowledge to determine a suitable address,
353 if the function is supported at all), then the program is presumed to
354 be successfully managing its own address space and should be trusted
355 when it asks to map over existing data structures. Furthermore, it is
356 also desirable to make as few system calls as possible, and it might be
357 considered onerous to require an munmap() before an mmap() to the same
358 address range. This volume of POSIX.1‐2008 specifies that the new map‐
359 pings replace any existing mappings, following existing practice in
360 this regard.
361 It is not expected that all hardware implementations are able to sup‐
363 port all combinations of permissions at all addresses. Implementations
364 are required to disallow write access to mappings without write permis‐
365 sion and to disallow access to mappings without any access permission.
366 Other than these restrictions, implementations may allow access types
367 other than those requested by the application. For example, if the
368 application requests only PROT_WRITE, the implementation may also allow
369 read access. A call to mmap() fails if the implementation cannot sup‐
370 port allowing all the access requested by the application. For example,
371 some implementations cannot support a request for both write access and
372 execute access simultaneously. All implementations must support
373 requests for no access, read access, write access, and both read and
374 write access. Strictly conforming code must only rely on the required
375 checks. These restrictions allow for portability across a wide range of
376 hardware.
377 The MAP_FIXED address treatment is likely to fail for non-page-aligned
379 values and for certain architecture-dependent address ranges. Conform‐
380 ing implementations cannot count on being able to choose address values
381 for MAP_FIXED without utilizing non-portable, implementation-defined
382 knowledge. Nonetheless, MAP_FIXED is provided as a standard interface
383 conforming to existing practice for utilizing such knowledge when it is
384 available.
385 Similarly, in order to allow implementations that do not support vir‐
387 tual addresses, support for directly specifying any mapping addresses
388 via MAP_FIXED is not required and thus a conforming application may not
389 count on it.
390 The MAP_PRIVATE function can be implemented efficiently when memory
392 protection hardware is available. When such hardware is not available,
393 implementations can implement such ``mappings'' by simply making a real
394 copy of the relevant data into process private memory, though this
395 tends to behave similarly to read().
396 The function has been defined to allow for many different models of
398 using shared memory. However, all uses are not equally portable across
399 all machine architectures. In particular, the mmap() function allows
400 the system as well as the application to specify the address at which
401 to map a specific region of a memory object. The most portable way to
402 use the function is always to let the system choose the address, speci‐
403 fying NULL as the value for the argument addr and not to specify
404 MAP_FIXED.
405 If it is intended that a particular region of a memory object be mapped
407 at the same address in a group of processes (on machines where this is
408 even possible), then MAP_FIXED can be used to pass in the desired map‐
409 ping address. The system can still be used to choose the desired
410 address if the first such mapping is made without specifying MAP_FIXED,
411 and then the resulting mapping address can be passed to subsequent pro‐
412 cesses for them to pass in via MAP_FIXED. The availability of a spe‐
413 cific address range cannot be guaranteed, in general.
414 The mmap() function can be used to map a region of memory that is
416 larger than the current size of the object. Memory access within the
417 mapping but beyond the current end of the underlying objects may result
418 in SIGBUS signals being sent to the process. The reason for this is
419 that the size of the object can be manipulated by other processes and
420 can change at any moment. The implementation should tell the applica‐
421 tion that a memory reference is outside the object where this can be
422 detected; otherwise, written data may be lost and read data may not
423 reflect actual data in the object.
424 Note that references beyond the end of the object do not extend the
426 object as the new end cannot be determined precisely by most virtual
427 memory hardware. Instead, the size can be directly manipulated by
428 ftruncate().
429 Process memory locking does apply to shared memory regions, and the
431 MEMLOCK_FUTURE argument to mlockall() can be relied upon to cause new
432 shared memory regions to be automatically locked.
433 Existing implementations of mmap() return the value −1 when unsuccess‐
435 ful. Since the casting of this value to type void * cannot be guaran‐
436 teed by the ISO C standard to be distinct from a successful value, this
437 volume of POSIX.1‐2008 defines the symbol MAP_FAILED, which a conform‐
438 ing implementation does not return as the result of a successful call.
439
441 None.
442
444 exec, fcntl(), fork(), lockf(), msync(), munmap(), mprotect(),
445 posix_typed_mem_open(), shmat(), sysconf()
446 The Base Definitions volume of POSIX.1‐2008, <sys_mman.h>
448
450 Portions of this text are reprinted and reproduced in electronic form
451 from IEEE Std 1003.1, 2013 Edition, Standard for Information Technology
452 -- Portable Operating System Interface (POSIX), The Open Group Base
453 Specifications Issue 7, Copyright (C) 2013 by the Institute of Electri‐
454 cal and Electronics Engineers, Inc and The Open Group. (This is
455 POSIX.1-2008 with the 2013 Technical Corrigendum 1 applied.) In the
456 event of any discrepancy between this version and the original IEEE and
457 The Open Group Standard, the original IEEE and The Open Group Standard
458 is the referee document. The original Standard can be obtained online
459 at http://www.unix.org/online.html .
460 Any typographical or formatting errors that appear in this page are
462 most likely to have been introduced during the conversion of the source
463 files to man page format. To report such errors, see https://www.ker‐
464 nel.org/doc/man-pages/reporting_bugs.html .
465IEEE/The Open Group 2013 MMAP(3P)