1 .\" Copyright (c) 1991, 1993
2 .\" The Regents of the University of California. All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
12 .\" 3. All advertising materials mentioning features or use of this software
13 .\" must display the following acknowledgement:
14 .\" This product includes software developed by the University of
15 .\" California, Berkeley and its contributors.
16 .\" 4. Neither the name of the University nor the names of its contributors
17 .\" may be used to endorse or promote products derived from this software
18 .\" without specific prior written permission.
20 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 .\" @(#)mmap.2 8.4 (Berkeley) 5/11/95
40 .Nd map files or devices into memory
42 .Fd #include <sys/types.h>
43 .Fd #include <sys/mman.h>
45 .Fn mmap "void * addr" "size_t len" "int prot" "int flags" "int fd" "off_t offset"
49 function causes the pages starting at
51 and continuing for at most
53 bytes to be mapped from the object described by
55 starting at byte offset
59 is not a multiple of the pagesize, the mapped region may extend past the
61 Any such extension beyond the end of the mapped object will be zero-filled.
65 is non-zero, it is used as a hint to the system.
66 (As a convenience to the system, the actual address of the region may differ
67 from the address supplied.)
70 is zero, an address will be selected by the system.
71 The actual starting address of the region is returned.
74 deletes any previous mapping in the allocated address range.
76 The protections (region accessibility) are specified in the
82 .Bl -tag -width MAP_FIXEDX
84 Pages may be executed.
93 parameter specifies the type of the mapped object, mapping options and
94 whether modifications made to the mapped copy of the page are private
95 to the process or are to be shared with other references.
96 Sharing, mapping type and options are specified in the
100 the following values:
102 .Bl -tag -width MAP_FIXEDX
104 Map anonymous memory not associated with any specific file.
105 The file descriptor used for creating
110 parameter is ignored.
112 .\"Mapped from a regular file or character-special device memory.
114 Do not permit the system to select a different address than the one
116 If the specified address cannot be used,
119 If MAP_FIXED is specified,
121 must be a multiple of the pagesize.
122 Use of this option is discouraged.
123 .It Dv MAP_HASSEMAPHORE
124 Notify the kernel that the region may contain semaphores and that special
125 handling may be necessary.
127 Permit regions to be inherited across
131 Modifications are private.
133 Modifications are shared.
135 This option is only available if your system has been compiled with
136 VM_STACK defined when compiling the kernel. This is the default for
137 i386 only. Consider adding -DVM_STACK to COPTFLAGS in your /etc/make.conf
138 to enable this option for other architechures. MAP_STACK implies
145 must include at least PROT_READ and PROT_WRITE. This option creates
146 a memory region that grows to at most
148 bytes in size, starting from the stack top and growing down. The
149 stack top is the starting address returned by the call, plus
151 bytes. The bottom of the stack at maximum growth is the starting
152 address returned by the call.
154 Causes data dirtied via this VM map to be flushed to physical media
155 only when necessary (usually by the pager) rather then gratuitously.
156 Typically this prevents the update daemons from flushing pages dirtied
157 through such maps and thus allows efficient sharing of memory across
158 unassociated processes using a file-backed shared memory map. Without
159 this option any VM pages you dirty may be flushed to disk every so often
160 (every 30-60 seconds usually) which can create performance problems if you
161 do not need that to occur (such as when you are using shared file-backed
162 mmap regions for IPC purposes). Note that VM/filesystem coherency is
163 maintained whether you use MAP_NOSYNC or not. This option is not portable
164 across UNIX platforms (yet), though some may implement the same behavior
169 function will flush all dirty data and metadata associated with a file,
170 including dirty NOSYNC VM data, to physical media. The
174 system call generally do not flush dirty NOSYNC VM data.
177 system call is obsolete since
179 implements a coherent filesystem buffer cache. However, it may be
180 used to associate dirty VM pages with filesystem buffers and thus cause
181 them to be flushed to physical media sooner rather then later.
186 function does not unmap pages, see
188 for further information.
190 The current design does not allow a process to specify the location of
192 In the future we may define an additional mapping type,
195 the file descriptor argument specifies a file or device to which swapping
198 Upon successful completion,
200 returns a pointer to the mapped region.
201 Otherwise, a value of MAP_FAILED is returned and
203 is set to indicate the error.
211 was specified as part of the
215 was not open for reading.
220 were specified as part of the
226 was not open for writing.
229 is not a valid open file descriptor.
232 was specified and the
234 parameter was not page aligned, or part of the desired address space
235 resides out of the valid address space for a user process.
241 was specified and the
243 parameter was not -1.
246 has not been specified and
248 did not reference a regular or character special file.
251 was not page-aligned. (See BUGS below.)
254 was specified and the
256 parameter wasn't available.
258 was specified and insufficient memory was available.
270 is limited to 2GB. Mmapping slightly more than 2GB doesn't work, but
271 it is possible to map a window of size (filesize % 2GB) for file sizes
272 of slightly less than 2G, 4GB, 6GB and 8GB.
274 The limit is imposed for a variety of reasons. Most of them have to do
277 not wanting to use 64 bit offsets in the VM system due to
278 the extreme performance penalty. So
280 uses 32bit page indexes and
283 a maximum of 8TB filesizes. It's actually bugs in
284 the filesystem code that causes the limit to be further restricted to
285 1TB (loss of precision when doing blockno calculations).
287 Another reason for the 2GB limit is that filesystem metadata can
288 reside at negative offsets.
290 We currently can only deal with page aligned file offsets.