1 # SOME DESCRIPTIVE TITLE
2 # Copyright (C) YEAR Free Software Foundation, Inc.
3 # This file is distributed under the same license as the PACKAGE package.
4 # FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.
9 "Project-Id-Version: PACKAGE VERSION\n"
10 "POT-Creation-Date: 2013-03-22 01:05+0900\n"
11 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
12 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
13 "Language-Team: LANGUAGE <LL@li.org>\n"
16 "Content-Type: text/plain; charset=UTF-8\n"
17 "Content-Transfer-Encoding: 8bit\n"
20 #: build/C/man2/alloc_hugepages.2:25
22 msgid "ALLOC_HUGEPAGES"
26 #: build/C/man2/alloc_hugepages.2:25
32 #: build/C/man2/alloc_hugepages.2:25 build/C/man2/cacheflush.2:25 build/C/man2/fallocate.2:11 build/C/man2/madvise.2:35 build/C/man3/mallinfo.3:26 build/C/man3/malloc_stats.3:26 build/C/man3/malloc_trim.3:26 build/C/man3/mallopt.3:26 build/C/man2/mlock.2:26 build/C/man2/mmap.2:40 build/C/man2/mmap2.2:29 build/C/man2/mprotect.2:38 build/C/man2/mremap.2:30 build/C/man2/msync.2:25 build/C/man2/posix_fadvise.2:28 build/C/man2/readahead.2:28 build/C/man2/remap_file_pages.2:28 build/C/man3/shm_open.3:26 build/C/man7/shm_overview.7:27 build/C/man2/shmctl.2:45 build/C/man2/shmget.2:38 build/C/man2/shmop.2:41 build/C/man2/subpage_prot.2:30 build/C/man2/sync_file_range.2:30
38 #: build/C/man2/alloc_hugepages.2:25 build/C/man3/alloca.3:43 build/C/man2/cacheflush.2:25 build/C/man2/fallocate.2:11 build/C/man2/madvise.2:35 build/C/man3/mallinfo.3:26 build/C/man3/malloc_get_state.3:25 build/C/man3/malloc_hook.3:10 build/C/man3/malloc_info.3:25 build/C/man3/malloc_stats.3:26 build/C/man3/malloc_trim.3:26 build/C/man3/malloc_usable_size.3:25 build/C/man3/mallopt.3:26 build/C/man3/mcheck.3:25 build/C/man2/mlock.2:26 build/C/man2/mmap.2:40 build/C/man2/mmap2.2:29 build/C/man2/mprotect.2:38 build/C/man2/mremap.2:30 build/C/man2/msync.2:25 build/C/man3/mtrace.3:25 build/C/man2/posix_fadvise.2:28 build/C/man3/posix_fallocate.3:25 build/C/man3/posix_memalign.3:29 build/C/man2/readahead.2:28 build/C/man2/remap_file_pages.2:28 build/C/man3/shm_open.3:26 build/C/man7/shm_overview.7:27 build/C/man2/shmctl.2:45 build/C/man2/shmget.2:38 build/C/man2/shmop.2:41 build/C/man2/subpage_prot.2:30 build/C/man2/sync_file_range.2:30
40 msgid "Linux Programmer's Manual"
44 #: build/C/man2/alloc_hugepages.2:26 build/C/man3/alloca.3:44 build/C/man2/cacheflush.2:26 build/C/man2/fallocate.2:12 build/C/man2/madvise.2:36 build/C/man3/mallinfo.3:27 build/C/man3/malloc_get_state.3:26 build/C/man3/malloc_hook.3:11 build/C/man3/malloc_info.3:26 build/C/man3/malloc_stats.3:27 build/C/man3/malloc_trim.3:27 build/C/man3/malloc_usable_size.3:26 build/C/man3/mallopt.3:27 build/C/man3/mcheck.3:26 build/C/man2/mlock.2:27 build/C/man2/mmap.2:41 build/C/man2/mmap2.2:30 build/C/man2/mprotect.2:39 build/C/man2/mremap.2:31 build/C/man2/msync.2:26 build/C/man3/mtrace.3:26 build/C/man2/posix_fadvise.2:29 build/C/man3/posix_fallocate.3:26 build/C/man3/posix_memalign.3:30 build/C/man2/readahead.2:29 build/C/man2/remap_file_pages.2:29 build/C/man3/shm_open.3:27 build/C/man7/shm_overview.7:28 build/C/man2/shmctl.2:46 build/C/man2/shmget.2:39 build/C/man2/shmop.2:42 build/C/man2/subpage_prot.2:31 build/C/man2/sync_file_range.2:31
50 #: build/C/man2/alloc_hugepages.2:28
51 msgid "alloc_hugepages, free_hugepages - allocate or free huge pages"
55 #: build/C/man2/alloc_hugepages.2:28 build/C/man3/alloca.3:46 build/C/man2/cacheflush.2:28 build/C/man2/fallocate.2:14 build/C/man2/madvise.2:38 build/C/man3/mallinfo.3:29 build/C/man3/malloc_get_state.3:28 build/C/man3/malloc_hook.3:15 build/C/man3/malloc_info.3:28 build/C/man3/malloc_stats.3:29 build/C/man3/malloc_trim.3:29 build/C/man3/malloc_usable_size.3:28 build/C/man3/mallopt.3:29 build/C/man3/mcheck.3:28 build/C/man2/mlock.2:29 build/C/man2/mmap.2:43 build/C/man2/mmap2.2:32 build/C/man2/mprotect.2:41 build/C/man2/mremap.2:33 build/C/man2/msync.2:28 build/C/man3/mtrace.3:28 build/C/man2/posix_fadvise.2:31 build/C/man3/posix_fallocate.3:28 build/C/man3/posix_memalign.3:32 build/C/man2/readahead.2:31 build/C/man2/remap_file_pages.2:31 build/C/man3/shm_open.3:29 build/C/man2/shmctl.2:48 build/C/man2/shmget.2:41 build/C/man2/shmop.2:44 build/C/man2/subpage_prot.2:33 build/C/man2/sync_file_range.2:33
60 #. asmlinkage unsigned long sys_alloc_hugepages(int key, unsigned long addr,
61 #. unsigned long len, int prot, int flag);
63 #: build/C/man2/alloc_hugepages.2:34
66 "B<void *alloc_hugepages(int >I<key>B<, void *>I<addr>B<, size_t "
68 "B< int >I<prot>B<, int >I<flag>B<);>\n"
71 #. asmlinkage int sys_free_hugepages(unsigned long addr);
73 #: build/C/man2/alloc_hugepages.2:37
75 msgid "B<int free_hugepages(void *>I<addr>B<);>\n"
79 #: build/C/man2/alloc_hugepages.2:38 build/C/man3/alloca.3:50 build/C/man2/cacheflush.2:34 build/C/man2/fallocate.2:22 build/C/man2/madvise.2:50 build/C/man3/mallinfo.3:33 build/C/man3/malloc_get_state.3:36 build/C/man3/malloc_hook.3:33 build/C/man3/malloc_info.3:34 build/C/man3/malloc_stats.3:33 build/C/man3/malloc_trim.3:33 build/C/man3/malloc_usable_size.3:32 build/C/man3/mallopt.3:33 build/C/man3/mcheck.3:40 build/C/man2/mlock.2:39 build/C/man2/mmap.2:54 build/C/man2/mmap2.2:39 build/C/man2/mprotect.2:47 build/C/man2/mremap.2:42 build/C/man2/msync.2:32 build/C/man3/mtrace.3:34 build/C/man2/posix_fadvise.2:50 build/C/man3/posix_fallocate.3:46 build/C/man3/posix_memalign.3:84 build/C/man2/readahead.2:38 build/C/man2/remap_file_pages.2:39 build/C/man3/shm_open.3:41 build/C/man7/shm_overview.7:30 build/C/man2/shmctl.2:56 build/C/man2/shmget.2:49 build/C/man2/shmop.2:53 build/C/man2/subpage_prot.2:41 build/C/man2/sync_file_range.2:41
85 #: build/C/man2/alloc_hugepages.2:49
87 "The system calls B<alloc_hugepages>() and B<free_hugepages>() were "
88 "introduced in Linux 2.5.36 and removed again in 2.5.54. They existed only "
89 "on i386 and ia64 (when built with B<CONFIG_HUGETLB_PAGE>). In Linux 2.4.20 "
90 "the syscall numbers exist, but the calls fail with the error B<ENOSYS>."
94 #: build/C/man2/alloc_hugepages.2:57
96 "On i386 the memory management hardware knows about ordinary pages (4 KiB) "
97 "and huge pages (2 or 4 MiB). Similarly ia64 knows about huge pages of "
98 "several sizes. These system calls serve to map huge pages into the "
99 "process's memory or to free them again. Huge pages are locked into memory, "
100 "and are not swapped."
104 #: build/C/man2/alloc_hugepages.2:66
106 "The I<key> argument is an identifier. When zero the pages are private, and "
107 "not inherited by children. When positive the pages are shared with other "
108 "applications using the same I<key>, and inherited by child processes."
112 #: build/C/man2/alloc_hugepages.2:81
114 "The I<addr> argument of B<free_hugepages>() tells which page is being "
115 "freed: it was the return value of a call to B<alloc_hugepages>(). (The "
116 "memory is first actually freed when all users have released it.) The "
117 "I<addr> argument of B<alloc_hugepages>() is a hint, that the kernel may or "
118 "may not follow. Addresses must be properly aligned."
122 #: build/C/man2/alloc_hugepages.2:86
124 "The I<len> argument is the length of the required segment. It must be a "
125 "multiple of the huge page size."
129 #: build/C/man2/alloc_hugepages.2:94
131 "The I<prot> argument specifies the memory protection of the segment. It is "
132 "one of B<PROT_READ>, B<PROT_WRITE>, B<PROT_EXEC>."
136 #: build/C/man2/alloc_hugepages.2:109
138 "The I<flag> argument is ignored, unless I<key> is positive. In that case, "
139 "if I<flag> is B<IPC_CREAT>, then a new huge page segment is created when "
140 "none with the given key existed. If this flag is not set, then B<ENOENT> is "
141 "returned when no segment with the given key exists."
145 #: build/C/man2/alloc_hugepages.2:109 build/C/man3/alloca.3:60 build/C/man2/cacheflush.2:53 build/C/man2/fallocate.2:112 build/C/man2/madvise.2:268 build/C/man3/malloc_get_state.3:61 build/C/man3/malloc_info.3:48 build/C/man3/malloc_trim.3:51 build/C/man3/malloc_usable_size.3:40 build/C/man3/mallopt.3:379 build/C/man3/mcheck.3:129 build/C/man2/mlock.2:116 build/C/man2/mmap.2:375 build/C/man2/mmap2.2:54 build/C/man2/mprotect.2:78 build/C/man2/mremap.2:127 build/C/man2/msync.2:68 build/C/man2/posix_fadvise.2:87 build/C/man3/posix_fallocate.3:66 build/C/man3/posix_memalign.3:141 build/C/man2/readahead.2:65 build/C/man2/remap_file_pages.2:124 build/C/man3/shm_open.3:169 build/C/man2/shmctl.2:274 build/C/man2/shmget.2:180 build/C/man2/shmop.2:171 build/C/man2/subpage_prot.2:66 build/C/man2/sync_file_range.2:141
151 #: build/C/man2/alloc_hugepages.2:118
153 "On success, B<alloc_hugepages>() returns the allocated virtual address, and "
154 "B<free_hugepages>() returns zero. On error, -1 is returned, and I<errno> "
155 "is set appropriately."
159 #: build/C/man2/alloc_hugepages.2:118 build/C/man2/cacheflush.2:59 build/C/man2/fallocate.2:115 build/C/man2/madvise.2:275 build/C/man3/malloc_info.3:53 build/C/man3/malloc_trim.3:56 build/C/man3/mallopt.3:384 build/C/man2/mlock.2:122 build/C/man2/mmap.2:392 build/C/man2/mmap2.2:61 build/C/man2/mprotect.2:85 build/C/man2/mremap.2:135 build/C/man2/msync.2:73 build/C/man2/posix_fadvise.2:90 build/C/man3/posix_fallocate.3:72 build/C/man3/posix_memalign.3:155 build/C/man2/readahead.2:71 build/C/man2/remap_file_pages.2:131 build/C/man3/shm_open.3:178 build/C/man2/shmctl.2:296 build/C/man2/shmget.2:184 build/C/man2/shmop.2:185 build/C/man2/subpage_prot.2:71 build/C/man2/sync_file_range.2:147
165 #: build/C/man2/alloc_hugepages.2:119 build/C/man2/fallocate.2:154
171 #: build/C/man2/alloc_hugepages.2:122
172 msgid "The system call is not supported on this kernel."
176 #: build/C/man2/alloc_hugepages.2:122
182 #: build/C/man2/alloc_hugepages.2:126
184 "I</proc/sys/vm/nr_hugepages> Number of configured hugetlb pages. This can "
185 "be read and written."
189 #: build/C/man2/alloc_hugepages.2:130
191 "I</proc/meminfo> Gives info on the number of configured hugetlb pages and on "
192 "their size in the three variables HugePages_Total, HugePages_Free, "
197 #: build/C/man2/alloc_hugepages.2:130 build/C/man3/alloca.3:65 build/C/man2/cacheflush.2:75 build/C/man2/fallocate.2:195 build/C/man2/madvise.2:325 build/C/man3/mallinfo.3:110 build/C/man3/malloc_get_state.3:86 build/C/man3/malloc_hook.3:80 build/C/man3/malloc_info.3:61 build/C/man3/malloc_stats.3:55 build/C/man3/malloc_trim.3:60 build/C/man3/malloc_usable_size.3:48 build/C/man3/mallopt.3:392 build/C/man3/mcheck.3:145 build/C/man2/mlock.2:186 build/C/man2/mmap.2:497 build/C/man2/mmap2.2:77 build/C/man2/mprotect.2:113 build/C/man2/mremap.2:185 build/C/man2/msync.2:96 build/C/man3/mtrace.3:78 build/C/man2/posix_fadvise.2:112 build/C/man3/posix_fallocate.3:103 build/C/man3/posix_memalign.3:180 build/C/man2/readahead.2:87 build/C/man2/remap_file_pages.2:153 build/C/man3/shm_open.3:249 build/C/man7/shm_overview.7:101 build/C/man2/shmctl.2:368 build/C/man2/shmget.2:231 build/C/man2/shmop.2:231 build/C/man2/subpage_prot.2:96 build/C/man2/sync_file_range.2:180
199 msgid "CONFORMING TO"
203 #: build/C/man2/alloc_hugepages.2:133
205 "These calls are specific to Linux on Intel processors, and should not be "
206 "used in programs intended to be portable."
210 #: build/C/man2/alloc_hugepages.2:133 build/C/man3/alloca.3:73 build/C/man2/madvise.2:346 build/C/man3/malloc_get_state.3:88 build/C/man3/malloc_hook.3:82 build/C/man3/malloc_info.3:63 build/C/man3/malloc_stats.3:57 build/C/man3/malloc_trim.3:62 build/C/man3/malloc_usable_size.3:50 build/C/man3/mcheck.3:147 build/C/man2/mlock.2:213 build/C/man2/mmap.2:515 build/C/man2/mmap2.2:79 build/C/man2/mprotect.2:122 build/C/man2/mremap.2:191 build/C/man3/mtrace.3:80 build/C/man2/posix_fadvise.2:121 build/C/man3/posix_memalign.3:219 build/C/man3/shm_open.3:255 build/C/man7/shm_overview.7:103 build/C/man2/shmctl.2:373 build/C/man2/shmget.2:237 build/C/man2/shmop.2:250 build/C/man2/subpage_prot.2:98 build/C/man2/sync_file_range.2:183
216 #: build/C/man2/alloc_hugepages.2:141
218 "These system calls are gone; they existed only in Linux 2.5.36 through to "
219 "2.5.54. Now the hugetlbfs file system can be used instead. Memory backed "
220 "by huge pages (if the CPU supports them) is obtained by using B<mmap>(2) to "
221 "map files in this virtual file system."
225 #: build/C/man2/alloc_hugepages.2:145
227 "The maximal number of huge pages can be specified using the B<hugepages=> "
232 #: build/C/man2/alloc_hugepages.2:150 build/C/man3/alloca.3:155 build/C/man2/cacheflush.2:89 build/C/man2/fallocate.2:203 build/C/man2/madvise.2:379 build/C/man3/mallinfo.3:279 build/C/man3/malloc_get_state.3:114 build/C/man3/malloc_hook.3:142 build/C/man3/malloc_info.3:260 build/C/man3/malloc_stats.3:67 build/C/man3/malloc_trim.3:82 build/C/man3/malloc_usable_size.3:64 build/C/man3/mallopt.3:580 build/C/man3/mcheck.3:208 build/C/man2/mlock.2:336 build/C/man2/mmap.2:730 build/C/man2/mmap2.2:98 build/C/man2/mprotect.2:229 build/C/man2/mremap.2:214 build/C/man2/msync.2:122 build/C/man3/mtrace.3:172 build/C/man2/posix_fadvise.2:189 build/C/man3/posix_fallocate.3:133 build/C/man3/posix_memalign.3:275 build/C/man2/readahead.2:98 build/C/man2/remap_file_pages.2:164 build/C/man3/shm_open.3:280 build/C/man7/shm_overview.7:127 build/C/man2/shmctl.2:425 build/C/man2/shmget.2:313 build/C/man2/shmop.2:295 build/C/man2/subpage_prot.2:134 build/C/man2/sync_file_range.2:222
238 #: build/C/man2/alloc_hugepages.2:157 build/C/man3/alloca.3:162 build/C/man2/cacheflush.2:96 build/C/man2/fallocate.2:210 build/C/man2/madvise.2:386 build/C/man3/mallinfo.3:286 build/C/man3/malloc_get_state.3:121 build/C/man3/malloc_hook.3:149 build/C/man3/malloc_info.3:267 build/C/man3/malloc_stats.3:74 build/C/man3/malloc_trim.3:89 build/C/man3/malloc_usable_size.3:71 build/C/man3/mallopt.3:587 build/C/man3/mcheck.3:215 build/C/man2/mlock.2:343 build/C/man2/mmap.2:737 build/C/man2/mmap2.2:105 build/C/man2/mprotect.2:236 build/C/man2/mremap.2:221 build/C/man2/msync.2:129 build/C/man3/mtrace.3:179 build/C/man2/posix_fadvise.2:196 build/C/man3/posix_fallocate.3:140 build/C/man3/posix_memalign.3:282 build/C/man2/readahead.2:105 build/C/man2/remap_file_pages.2:171 build/C/man3/shm_open.3:287 build/C/man7/shm_overview.7:134 build/C/man2/shmctl.2:432 build/C/man2/shmget.2:320 build/C/man2/shmop.2:302 build/C/man2/subpage_prot.2:141 build/C/man2/sync_file_range.2:229
240 "This page is part of release 3.50 of the Linux I<man-pages> project. A "
241 "description of the project, and information about reporting bugs, can be "
242 "found at http://www.kernel.org/doc/man-pages/."
246 #: build/C/man3/alloca.3:43
252 #: build/C/man3/alloca.3:43
258 #: build/C/man3/alloca.3:43 build/C/man3/malloc_get_state.3:25 build/C/man3/malloc_hook.3:10 build/C/man3/malloc_info.3:25 build/C/man3/malloc_usable_size.3:25 build/C/man3/mcheck.3:25 build/C/man3/mtrace.3:25 build/C/man3/posix_fallocate.3:25 build/C/man3/posix_memalign.3:29
264 #: build/C/man3/alloca.3:46
265 msgid "alloca - allocate memory that is automatically freed"
269 #: build/C/man3/alloca.3:48
270 msgid "B<#include E<lt>alloca.hE<gt>>"
274 #: build/C/man3/alloca.3:50
275 msgid "B<void *alloca(size_t >I<size>B<);>"
279 #: build/C/man3/alloca.3:60
281 "The B<alloca>() function allocates I<size> bytes of space in the stack "
282 "frame of the caller. This temporary space is automatically freed when the "
283 "function that called B<alloca>() returns to its caller."
287 #: build/C/man3/alloca.3:65
289 "The B<alloca>() function returns a pointer to the beginning of the "
290 "allocated space. If the allocation causes stack overflow, program behavior "
295 #: build/C/man3/alloca.3:67
296 msgid "This function is not in POSIX.1-2001."
300 #: build/C/man3/alloca.3:73
302 "There is evidence that the B<alloca>() function appeared in 32V, PWB, "
303 "PWB.2, 3BSD, and 4BSD. There is a man page for it in 4.3BSD. Linux uses "
308 #: build/C/man3/alloca.3:88
310 "The B<alloca>() function is machine- and compiler-dependent. For certain "
311 "applications, its use can improve efficiency compared to the use of "
312 "B<malloc>(3) plus B<free>(3). In certain cases, it can also simplify "
313 "memory deallocation in applications that use B<longjmp>(3) or "
314 "B<siglongjmp>(3). Otherwise, its use is discouraged."
318 #: build/C/man3/alloca.3:97
320 "Because the space allocated by B<alloca>() is allocated within the stack "
321 "frame, that space is automatically freed if the function return is jumped "
322 "over by a call to B<longjmp>(3) or B<siglongjmp>(3)."
326 #: build/C/man3/alloca.3:102
327 msgid "Do not attempt to B<free>(3) space allocated by B<alloca>()!"
331 #: build/C/man3/alloca.3:102
333 msgid "Notes on the GNU version"
337 #: build/C/man3/alloca.3:124
339 "Normally, B<gcc>(1) translates calls to B<alloca>() with inlined code. "
340 "This is not done when either the I<-ansi>, I<-std=c89>, I<-std=c99>, or the "
341 "I<-fno-builtin> option is given (and the header I<E<lt>alloca.hE<gt>> is not "
342 "included). But beware! By default the glibc version of "
343 "I<E<lt>stdlib.hE<gt>> includes I<E<lt>alloca.hE<gt>> and that contains the "
348 #: build/C/man3/alloca.3:127
350 msgid " #define alloca(size) __builtin_alloca (size)\n"
354 #: build/C/man3/alloca.3:130
355 msgid "with messy consequences if one has a private version of this function."
359 #: build/C/man3/alloca.3:134
361 "The fact that the code is inlined means that it is impossible to take the "
362 "address of this function, or to change its behavior by linking with a "
367 #: build/C/man3/alloca.3:138
369 "The inlined code often consists of a single instruction adjusting the stack "
370 "pointer, and does not check for stack overflow. Thus, there is no NULL "
375 #: build/C/man3/alloca.3:138 build/C/man2/cacheflush.2:82 build/C/man3/mallinfo.3:114 build/C/man3/mallopt.3:406 build/C/man2/mlock.2:304 build/C/man2/mmap.2:587 build/C/man3/mtrace.3:102 build/C/man2/posix_fadvise.2:178 build/C/man2/shmget.2:299
381 #: build/C/man3/alloca.3:143
383 "There is no error indication if the stack frame cannot be extended. "
384 "(However, after a failed allocation, the program is likely to receive a "
385 "B<SIGSEGV> signal if it attempts to access the unallocated space.)"
389 #: build/C/man3/alloca.3:151
391 "On many systems B<alloca>() cannot be used inside the list of arguments of "
392 "a function call, because the stack space reserved by B<alloca>() would "
393 "appear on the stack in the middle of the space for the function arguments."
397 #: build/C/man3/alloca.3:151 build/C/man2/fallocate.2:198 build/C/man2/madvise.2:371 build/C/man3/mallinfo.3:270 build/C/man3/malloc_get_state.3:111 build/C/man3/malloc_hook.3:137 build/C/man3/malloc_info.3:254 build/C/man3/malloc_stats.3:61 build/C/man3/malloc_trim.3:78 build/C/man3/malloc_usable_size.3:62 build/C/man3/mallopt.3:566 build/C/man3/mcheck.3:204 build/C/man2/mlock.2:329 build/C/man2/mmap.2:712 build/C/man2/mmap2.2:92 build/C/man2/mprotect.2:226 build/C/man2/mremap.2:199 build/C/man2/msync.2:118 build/C/man3/mtrace.3:167 build/C/man2/posix_fadvise.2:183 build/C/man3/posix_fallocate.3:128 build/C/man3/posix_memalign.3:270 build/C/man2/readahead.2:92 build/C/man2/remap_file_pages.2:157 build/C/man3/shm_open.3:269 build/C/man7/shm_overview.7:114 build/C/man2/shmctl.2:418 build/C/man2/shmget.2:305 build/C/man2/shmop.2:287 build/C/man2/subpage_prot.2:128 build/C/man2/sync_file_range.2:217
403 #: build/C/man3/alloca.3:155
404 msgid "B<brk>(2), B<longjmp>(3), B<malloc>(3)"
408 #: build/C/man2/cacheflush.2:25
414 #: build/C/man2/cacheflush.2:25
420 #: build/C/man2/cacheflush.2:28
421 msgid "cacheflush - flush contents of instruction and/or data cache"
425 #: build/C/man2/cacheflush.2:31
427 msgid "B<#include E<lt>asm/cachectl.hE<gt>>\n"
431 #: build/C/man2/cacheflush.2:33
433 msgid "B<int cacheflush(char *>I<addr>B<, int >I<nbytes>B<, int >I<cache>B<);>\n"
437 #: build/C/man2/cacheflush.2:43
439 "B<cacheflush>() flushes the contents of the indicated cache(s) for the user "
440 "addresses in the range I<addr> to I<(addr+nbytes-1)>. I<cache> may be one "
445 #: build/C/man2/cacheflush.2:43
451 #: build/C/man2/cacheflush.2:46
452 msgid "Flush the instruction cache."
456 #: build/C/man2/cacheflush.2:46
462 #: build/C/man2/cacheflush.2:49
463 msgid "Write back to memory and invalidate the affected valid cache lines."
467 #: build/C/man2/cacheflush.2:49
473 #: build/C/man2/cacheflush.2:53
474 msgid "Same as B<(ICACHE|DCACHE)>."
478 #: build/C/man2/cacheflush.2:59
480 "B<cacheflush>() returns 0 on success or -1 on error. If errors are "
481 "detected, I<errno> will indicate the error."
485 #: build/C/man2/cacheflush.2:60 build/C/man2/mmap2.2:62 build/C/man2/mremap.2:142 build/C/man2/shmctl.2:305 build/C/man2/subpage_prot.2:72
491 #: build/C/man2/cacheflush.2:67
493 "Some or all of the address range I<addr> to I<(addr+nbytes-1)> is not "
498 #: build/C/man2/cacheflush.2:67 build/C/man2/fallocate.2:127 build/C/man2/madvise.2:282 build/C/man3/malloc_info.3:54 build/C/man2/mlock.2:157 build/C/man2/mlock.2:164 build/C/man2/mlock.2:176 build/C/man2/mmap.2:423 build/C/man2/mmap.2:431 build/C/man2/mmap.2:436 build/C/man2/mmap2.2:65 build/C/man2/mprotect.2:95 build/C/man2/mremap.2:151 build/C/man2/msync.2:80 build/C/man2/posix_fadvise.2:94 build/C/man3/posix_fallocate.3:81 build/C/man3/posix_memalign.3:156 build/C/man2/readahead.2:76 build/C/man2/remap_file_pages.2:132 build/C/man2/remap_file_pages.2:139 build/C/man3/shm_open.3:211 build/C/man2/shmctl.2:319 build/C/man2/shmget.2:198 build/C/man2/shmop.2:200 build/C/man2/shmop.2:223 build/C/man2/subpage_prot.2:77 build/C/man2/sync_file_range.2:152
504 #: build/C/man2/cacheflush.2:75
505 msgid "I<cache> is not one of B<ICACHE>, B<DCACHE>, or B<BCACHE>."
508 #. FIXME This system call was only on MIPS back in 1.2 days, but
509 #. by now it is on a number of other architectures (but not i386).
510 #. Investigate the details and update this page.
511 #. Irix 6.5 appears to have a cacheflush() syscall -- mtk
513 #: build/C/man2/cacheflush.2:82
515 "This Linux-specific system call is only available on MIPS-based systems. It "
516 "should not be used in programs intended to be portable."
520 #: build/C/man2/cacheflush.2:89
522 "The current implementation ignores the I<addr> and I<nbytes> arguments. "
523 "Therefore, the whole cache is always flushed."
527 #: build/C/man2/fallocate.2:11
533 #: build/C/man2/fallocate.2:11 build/C/man3/posix_fallocate.3:25 build/C/man2/shmop.2:41
539 #: build/C/man2/fallocate.2:14
540 msgid "fallocate - manipulate file space"
544 #: build/C/man2/fallocate.2:18 build/C/man2/readahead.2:35
547 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
548 "B<#include E<lt>fcntl.hE<gt>>\n"
552 #: build/C/man2/fallocate.2:21
555 "B<int fallocate(int >I<fd>B<, int >I<mode>B<, off_t >I<offset>B<, off_t "
560 #: build/C/man2/fallocate.2:27
562 "This is a nonportable, Linux-specific system call. For the portable, "
563 "POSIX.1-specified method of ensuring that space is allocated for a file, see "
564 "B<posix_fallocate>(3)."
568 #: build/C/man2/fallocate.2:37
570 "B<fallocate>() allows the caller to directly manipulate the allocated disk "
571 "space for the file referred to by I<fd> for the byte range starting at "
572 "I<offset> and continuing for I<len> bytes."
576 #: build/C/man2/fallocate.2:42
578 "The I<mode> argument determines the operation to be performed on the given "
579 "range. Details of the supported operations are given in the subsections "
584 #: build/C/man2/fallocate.2:42
586 msgid "Allocating disk space"
590 #: build/C/man2/fallocate.2:61
592 "The default operation (i.e., I<mode> is zero) of B<fallocate>() allocates "
593 "and initializes to zero the disk space within the range specified by "
594 "I<offset> and I<len>. The file size (as reported by B<stat>(2)) will be "
595 "changed if I<offset>+I<len> is greater than the file size. This default "
596 "behavior closely resembles the behavior of the B<posix_fallocate>(3) "
597 "library function, and is intended as a method of optimally implementing that "
602 #: build/C/man2/fallocate.2:67
604 "After a successful call, subsequent writes into the range specified by "
605 "I<offset> and I<len> are guaranteed not to fail because of lack of disk "
610 #: build/C/man2/fallocate.2:78
612 "If the B<FALLOC_FL_KEEP_SIZE> flag is specified in I<mode>, the behavior of "
613 "the call is similar, but the file size will not be changed even if "
614 "I<offset>+I<len> is greater than the file size. Preallocating zeroed blocks "
615 "beyond the end of the file in this manner is useful for optimizing append "
620 #: build/C/man2/fallocate.2:82
622 "Because allocation is done in block size chunks, B<fallocate>() may "
623 "allocate a larger range of disk space than was specified."
627 #: build/C/man2/fallocate.2:82
629 msgid "Deallocating file space"
633 #: build/C/man2/fallocate.2:97
635 "Specifying the B<FALLOC_FL_PUNCH_HOLE> flag (available since Linux 2.6.38) "
636 "in I<mode> deallocates space (i.e., creates a hole) in the byte range "
637 "starting at I<offset> and continuing for I<len> bytes. Within the specified "
638 "range, partial file system blocks are zeroed, and whole file system blocks "
639 "are removed from the file. After a successful call, subsequent reads from "
640 "this range will return zeroes."
644 #: build/C/man2/fallocate.2:108
646 "The B<FALLOC_FL_PUNCH_HOLE> flag must be ORed with B<FALLOC_FL_KEEP_SIZE> in "
647 "I<mode>; in other words, even when punching off the end of the file, the "
648 "file size (as reported by B<stat>(2)) does not change."
652 #: build/C/man2/fallocate.2:112
654 "Not all file systems support B<FALLOC_FL_PUNCH_HOLE>; if a file system "
655 "doesn't support the operation, an error is returned."
659 #: build/C/man2/fallocate.2:115
660 msgid "B<fallocate>() returns zero on success, and -1 on failure."
664 #: build/C/man2/fallocate.2:116 build/C/man2/madvise.2:279 build/C/man2/mmap.2:417 build/C/man2/posix_fadvise.2:91 build/C/man3/posix_fallocate.3:73 build/C/man2/readahead.2:72 build/C/man2/sync_file_range.2:148
670 #: build/C/man2/fallocate.2:120 build/C/man3/posix_fallocate.3:77
671 msgid "I<fd> is not a valid file descriptor, or is not opened for writing."
675 #: build/C/man2/fallocate.2:120 build/C/man3/posix_fallocate.3:77
681 #: build/C/man2/fallocate.2:124
682 msgid "I<offset>+I<len> exceeds the maximum file size."
686 #: build/C/man2/fallocate.2:124
692 #: build/C/man2/fallocate.2:127
693 msgid "A signal was caught during execution."
697 #: build/C/man2/fallocate.2:139 build/C/man3/posix_fallocate.3:87
698 msgid "I<offset> was less than 0, or I<len> was less than or equal to 0."
702 #: build/C/man2/fallocate.2:139 build/C/man2/madvise.2:310 build/C/man2/sync_file_range.2:160
708 #: build/C/man2/fallocate.2:142
709 msgid "An I/O error occurred while reading from or writing to a file system."
713 #: build/C/man2/fallocate.2:142 build/C/man2/mmap.2:452 build/C/man3/posix_fallocate.3:87
719 #: build/C/man2/fallocate.2:149
721 "I<fd> does not refer to a regular file or a directory. (If I<fd> is a pipe "
722 "or FIFO, a different error results.)"
726 #: build/C/man2/fallocate.2:149 build/C/man3/posix_fallocate.3:91 build/C/man2/shmget.2:216 build/C/man2/sync_file_range.2:166
732 #: build/C/man2/fallocate.2:154 build/C/man3/posix_fallocate.3:96
734 "There is not enough space left on the device containing the file referred to "
739 #: build/C/man2/fallocate.2:158
740 msgid "This kernel does not implement B<fallocate>()."
744 #: build/C/man2/fallocate.2:158
746 msgid "B<EOPNOTSUPP>"
750 #: build/C/man2/fallocate.2:167
752 "The file system containing the file referred to by I<fd> does not support "
753 "this operation; or the I<mode> is not supported by the file system "
754 "containing the file referred to by I<fd>."
758 #: build/C/man2/fallocate.2:167 build/C/man2/mlock.2:143 build/C/man2/mlock.2:182 build/C/man2/mmap.2:460 build/C/man2/shmctl.2:344 build/C/man2/shmget.2:224
764 #: build/C/man2/fallocate.2:183
766 "The file referred to by I<fd> is marked immutable (see B<chattr>(1)). Or: "
767 "I<mode> specifies B<FALLOC_FL_PUNCH_HOLE> and the file referred to by I<fd> "
768 "is marked append-only (see B<chattr>(1))."
772 #: build/C/man2/fallocate.2:183 build/C/man2/posix_fadvise.2:97 build/C/man3/posix_fallocate.3:96 build/C/man2/sync_file_range.2:169
778 #: build/C/man2/fallocate.2:187
779 msgid "I<fd> refers to a pipe or FIFO."
783 #: build/C/man2/fallocate.2:187 build/C/man3/malloc_info.3:58 build/C/man3/mcheck.3:134 build/C/man2/mmap2.2:74 build/C/man2/posix_fadvise.2:104 build/C/man3/posix_fallocate.3:100 build/C/man3/posix_memalign.3:165 build/C/man2/readahead.2:82 build/C/man2/remap_file_pages.2:148 build/C/man3/shm_open.3:247 build/C/man2/subpage_prot.2:90 build/C/man2/sync_file_range.2:177
788 #. See http://sourceware.org/bugzilla/show_bug.cgi?id=14964
790 #: build/C/man2/fallocate.2:195
792 "B<fallocate>() is available on Linux since kernel 2.6.23. Support is "
793 "provided by glibc since version 2.10. The B<FALLOC_FL_*> flags are defined "
794 "in glibc headers only since version 2.18."
798 #: build/C/man2/fallocate.2:198
799 msgid "B<fallocate>() is Linux-specific."
803 #: build/C/man2/fallocate.2:203
804 msgid "B<fallocate>(1), B<ftruncate>(2), B<posix_fadvise>(3), B<posix_fallocate>(3)"
808 #: build/C/man2/madvise.2:35
814 #: build/C/man2/madvise.2:35 build/C/man3/malloc_info.3:25
820 #: build/C/man2/madvise.2:38
821 msgid "madvise - give advice about use of memory"
825 #: build/C/man2/madvise.2:40 build/C/man2/msync.2:30 build/C/man3/shm_open.3:31
826 msgid "B<#include E<lt>sys/mman.hE<gt>>"
830 #: build/C/man2/madvise.2:42
831 msgid "B<int madvise(void *>I<addr>B<, size_t >I<length>B<, int >I<advice>B<);>"
835 #: build/C/man2/madvise.2:46 build/C/man2/posix_fadvise.2:43 build/C/man3/posix_fallocate.3:39 build/C/man3/posix_memalign.3:49
836 msgid "Feature Test Macro Requirements for glibc (see B<feature_test_macros>(7)):"
840 #: build/C/man2/madvise.2:50
841 msgid "B<madvise>(): _BSD_SOURCE"
845 #: build/C/man2/madvise.2:68
847 "The B<madvise>() system call advises the kernel about how to handle paging "
848 "input/output in the address range beginning at address I<addr> and with size "
849 "I<length> bytes. It allows an application to tell the kernel how it expects "
850 "to use some mapped or shared memory areas, so that the kernel can choose "
851 "appropriate read-ahead and caching techniques. This call does not influence "
852 "the semantics of the application (except in the case of B<MADV_DONTNEED>), "
853 "but may influence its performance. The kernel is free to ignore the advice."
857 #: build/C/man2/madvise.2:72
858 msgid "The advice is indicated in the I<advice> argument which can be"
862 #: build/C/man2/madvise.2:72
864 msgid "B<MADV_NORMAL>"
868 #: build/C/man2/madvise.2:76
869 msgid "No special treatment. This is the default."
873 #: build/C/man2/madvise.2:76
875 msgid "B<MADV_RANDOM>"
879 #: build/C/man2/madvise.2:80
881 "Expect page references in random order. (Hence, read ahead may be less "
882 "useful than normally.)"
886 #: build/C/man2/madvise.2:80
888 msgid "B<MADV_SEQUENTIAL>"
892 #: build/C/man2/madvise.2:85
894 "Expect page references in sequential order. (Hence, pages in the given "
895 "range can be aggressively read ahead, and may be freed soon after they are "
900 #: build/C/man2/madvise.2:85
902 msgid "B<MADV_WILLNEED>"
906 #: build/C/man2/madvise.2:89
908 "Expect access in the near future. (Hence, it might be a good idea to read "
913 #: build/C/man2/madvise.2:89
915 msgid "B<MADV_DONTNEED>"
919 #: build/C/man2/madvise.2:100
921 "Do not expect access in the near future. (For the time being, the "
922 "application is finished with the given range, so the kernel can free "
923 "resources associated with it.) Subsequent accesses of pages in this range "
924 "will succeed, but will result either in reloading of the memory contents "
925 "from the underlying mapped file (see B<mmap>(2)) or zero-fill-on-demand "
926 "pages for mappings without an underlying file."
930 #: build/C/man2/madvise.2:100
932 msgid "B<MADV_REMOVE> (Since Linux 2.6.16)"
936 #. Databases want to use this feature to drop a section of their
937 #. bufferpool (shared memory segments) - without writing back to
938 #. disk/swap space. This feature is also useful for supporting
939 #. hot-plug memory on UML.
941 #: build/C/man2/madvise.2:113
943 "Free up a given range of pages and its associated backing store. Currently, "
944 "only shmfs/tmpfs supports this; other file systems return with the error "
949 #: build/C/man2/madvise.2:113
951 msgid "B<MADV_DONTFORK> (Since Linux 2.6.16)"
954 #. See http://lwn.net/Articles/171941/
955 #. [PATCH] madvise MADV_DONTFORK/MADV_DOFORK
956 #. Currently, copy-on-write may change the physical address of
957 #. a page even if the user requested that the page is pinned in
958 #. memory (either by mlock or by get_user_pages). This happens
959 #. if the process forks meanwhile, and the parent writes to that
960 #. page. As a result, the page is orphaned: in case of
961 #. get_user_pages, the application will never see any data hardware
962 #. DMA's into this page after the COW. In case of mlock'd memory,
963 #. the parent is not getting the realtime/security benefits of mlock.
965 #. In particular, this affects the Infiniband modules which do DMA from
966 #. and into user pages all the time.
968 #. This patch adds madvise options to control whether memory range is
969 #. inherited across fork. Useful e.g. for when hardware is doing DMA
970 #. from/into these pages. Could also be useful to an application
971 #. wanting to speed up its forks by cutting large areas out of
974 #. SEE ALSO: http://lwn.net/Articles/171941/
975 #. "Tweaks to madvise() and posix_fadvise()", 14 Feb 2006
977 #: build/C/man2/madvise.2:144
979 "Do not make the pages in this range available to the child after a "
980 "B<fork>(2). This is useful to prevent copy-on-write semantics from changing "
981 "the physical location of a page(s) if the parent writes to it after a "
982 "B<fork>(2). (Such page relocations cause problems for hardware that DMAs "
987 #: build/C/man2/madvise.2:144
989 msgid "B<MADV_DOFORK> (Since Linux 2.6.16)"
993 #: build/C/man2/madvise.2:150
995 "Undo the effect of B<MADV_DONTFORK>, restoring the default behavior, whereby "
996 "a mapping is inherited across B<fork>(2)."
1000 #: build/C/man2/madvise.2:150
1002 msgid "B<MADV_HWPOISON> (Since Linux 2.6.32)"
1006 #: build/C/man2/madvise.2:162
1008 "Poison a page and handle it like a hardware memory corruption. This "
1009 "operation is only available for privileged (B<CAP_SYS_ADMIN>) processes. "
1010 "This operation may result in the calling process receiving a B<SIGBUS> and "
1011 "the page being unmapped. This feature is intended for testing of memory "
1012 "error-handling code; it is only available if the kernel was configured with "
1013 "B<CONFIG_MEMORY_FAILURE>."
1017 #: build/C/man2/madvise.2:162
1019 msgid "B<MADV_SOFT_OFFLINE> (Since Linux 2.6.33)"
1023 #: build/C/man2/madvise.2:180
1025 "Soft offline the pages in the range specified by I<addr> and I<length>. The "
1026 "memory of each page in the specified range is preserved (i.e., when next "
1027 "accessed, the same content will be visible, but in a new physical page "
1028 "frame), and the original page is offlined (i.e., no longer used, and taken "
1029 "out of normal memory management). The effect of the B<MADV_SOFT_OFFLINE> "
1030 "operation is invisible to (i.e., does not change the semantics of) the "
1031 "calling process. This feature is intended for testing of memory "
1032 "error-handling code; it is only available if the kernel was configured with "
1033 "B<CONFIG_MEMORY_FAILURE>."
1037 #: build/C/man2/madvise.2:180
1039 msgid "B<MADV_MERGEABLE> (since Linux 2.6.32)"
1043 #: build/C/man2/madvise.2:205
1045 "Enable Kernel Samepage Merging (KSM) for the pages in the range specified by "
1046 "I<addr> and I<length>. The kernel regularly scans those areas of user "
1047 "memory that have been marked as mergeable, looking for pages with identical "
1048 "content. These are replaced by a single write-protected page (which is "
1049 "automatically copied if a process later wants to update the content of the "
1050 "page). KSM only merges private anonymous pages (see B<mmap>(2)). The KSM "
1051 "feature is intended for applications that generate many instances of the "
1052 "same data (e.g., virtualization systems such as KVM). It can consume a lot "
1053 "of processing power; use with care. See the Linux kernel source file "
1054 "I<Documentation/vm/ksm.txt> for more details. The B<MADV_MERGEABLE> and "
1055 "B<MADV_UNMERGEABLE> operations are only available if the kernel was "
1056 "configured with B<CONFIG_KSM>."
1060 #: build/C/man2/madvise.2:205
1062 msgid "B<MADV_UNMERGEABLE> (since Linux 2.6.32)"
1066 #: build/C/man2/madvise.2:214
1068 "Undo the effect of an earlier B<MADV_MERGEABLE> operation on the specified "
1069 "address range; KSM unmerges whatever pages it had merged in the address "
1070 "range specified by I<addr> and I<length>."
1074 #: build/C/man2/madvise.2:214
1076 msgid "B<MADV_HUGEPAGE> (since Linux 2.6.38)"
1079 #. http://lwn.net/Articles/358904/
1080 #. https://lwn.net/Articles/423584/
1082 #: build/C/man2/madvise.2:243
1084 "Enables Transparent Huge Pages (THP) for pages in the range specified by "
1085 "I<addr> and I<length>. Currently, Transparent Huge Pages only work with "
1086 "private anonymous pages (see B<mmap>(2)). The kernel will regularly scan "
1087 "the areas marked as huge page candidates to replace them with huge pages. "
1088 "The kernel will also allocate huge pages directly when the region is "
1089 "naturally aligned to the huge page size (see B<posix_memalign>(2)). This "
1090 "feature is primarily aimed at applications that use large mappings of data "
1091 "and access large regions of that memory at a time (e.g., virtualization "
1092 "systems such as QEMU). It can very easily waste memory (e.g., a 2MB mapping "
1093 "that only ever accesses 1 byte will result in 2MB of wired memory instead of "
1094 "one 4KB page). See the Linux kernel source file "
1095 "I<Documentation/vm/transhuge.txt> for more details. The B<MADV_HUGEPAGE> "
1096 "and B<MADV_NOHUGEPAGE> operations are only available if the kernel was "
1097 "configured with B<CONFIG_TRANSPARENT_HUGEPAGE>."
1101 #: build/C/man2/madvise.2:243
1103 msgid "B<MADV_NOHUGEPAGE> (since Linux 2.6.38)"
1107 #: build/C/man2/madvise.2:250
1109 "Ensures that memory in the address range specified by I<addr> and I<length> "
1110 "will not be collapsed into huge pages."
1114 #: build/C/man2/madvise.2:250
1116 msgid "B<MADV_DONTDUMP> (since Linux 3.4)"
1120 #: build/C/man2/madvise.2:264
1122 "Exclude from a core dump those pages in the range specified by I<addr> and "
1123 "I<length>. This is useful in applications that have large areas of memory "
1124 "that are known not to be useful in a core dump. The effect of "
1125 "B<MADV_DONTDUMP> takes precedence over the bit mask that is set via the "
1126 "I</proc/PID/coredump_filter> file (see B<core>(5))."
1130 #: build/C/man2/madvise.2:264
1132 msgid "B<MADV_DODUMP> (since Linux 3.4)"
1136 #: build/C/man2/madvise.2:268
1137 msgid "Undo the effect of an earlier B<MADV_DONTDUMP>."
1141 #: build/C/man2/madvise.2:275
1143 "On success B<madvise>() returns zero. On error, it returns -1 and I<errno> "
1144 "is set appropriately."
1148 #: build/C/man2/madvise.2:276 build/C/man2/mlock.2:154 build/C/man2/mmap.2:413 build/C/man2/mremap.2:136
1154 #: build/C/man2/madvise.2:279
1155 msgid "A kernel resource was temporarily unavailable."
1159 #: build/C/man2/madvise.2:282
1160 msgid "The map exists, but the area maps something that isn't a file."
1164 #: build/C/man2/madvise.2:285
1165 msgid "This error can occur for the following reasons:"
1169 #: build/C/man2/madvise.2:286 build/C/man2/madvise.2:292 build/C/man2/madvise.2:295 build/C/man2/madvise.2:298 build/C/man2/madvise.2:301 build/C/man3/mallopt.3:235 build/C/man3/mallopt.3:241
1177 #: build/C/man2/madvise.2:292
1178 msgid "The value I<len> is negative."
1182 #: build/C/man2/madvise.2:295
1183 msgid "I<addr> is not page-aligned."
1187 #: build/C/man2/madvise.2:298
1188 msgid "I<advice> is not a valid value"
1192 #: build/C/man2/madvise.2:301
1194 "The application is attempting to release locked or shared pages (with "
1195 "B<MADV_DONTNEED>)."
1199 #: build/C/man2/madvise.2:309
1201 "B<MADV_MERGEABLE> or B<MADV_UNMERGEABLE> was specified in I<advice>, but the "
1202 "kernel was not configured with B<CONFIG_KSM>."
1206 #: build/C/man2/madvise.2:316
1208 "(for B<MADV_WILLNEED>) Paging in this area would exceed the process's "
1209 "maximum resident set size."
1213 #: build/C/man2/madvise.2:316 build/C/man2/madvise.2:321 build/C/man2/mlock.2:123 build/C/man2/mlock.2:131 build/C/man2/mlock.2:169 build/C/man2/mmap.2:456 build/C/man2/mprotect.2:100 build/C/man2/mprotect.2:103 build/C/man2/mremap.2:179 build/C/man2/msync.2:93 build/C/man3/posix_memalign.3:162 build/C/man2/shmctl.2:328 build/C/man2/shmget.2:213 build/C/man2/shmop.2:214 build/C/man2/subpage_prot.2:87 build/C/man2/sync_file_range.2:163
1219 #: build/C/man2/madvise.2:321
1220 msgid "(for B<MADV_WILLNEED>) Not enough memory: paging in failed."
1224 #: build/C/man2/madvise.2:325
1226 "Addresses in the specified range are not currently mapped, or are outside "
1227 "the address space of the process."
1230 #. FIXME . Write a posix_madvise(3) page.
1232 #: build/C/man2/madvise.2:337
1234 "POSIX.1b. POSIX.1-2001 describes B<posix_madvise>(3) with constants "
1235 "B<POSIX_MADV_NORMAL>, etc., with a behavior close to that described here. "
1236 "There is a similar B<posix_fadvise>(2) for file access."
1240 #: build/C/man2/madvise.2:346
1242 "B<MADV_REMOVE>, B<MADV_DONTFORK>, B<MADV_DOFORK>, B<MADV_HWPOISON>, "
1243 "B<MADV_MERGEABLE>, and B<MADV_UNMERGEABLE> are Linux-specific."
1247 #: build/C/man2/madvise.2:347 build/C/man2/mlock.2:267 build/C/man2/shmget.2:293
1253 #: build/C/man2/madvise.2:355
1255 "The current Linux implementation (2.4.0) views this system call more as a "
1256 "command than as advice and hence may return an error when it cannot do what "
1257 "it usually would do in response to this advice. (See the ERRORS description "
1258 "above.) This is nonstandard behavior."
1264 #. function first appeared in 4.4BSD.
1266 #: build/C/man2/madvise.2:371
1268 "The Linux implementation requires that the address I<addr> be page-aligned, "
1269 "and allows I<length> to be zero. If there are some parts of the specified "
1270 "address range that are not mapped, the Linux version of B<madvise>() "
1271 "ignores them and applies the call to the rest (but returns B<ENOMEM> from "
1272 "the system call, as it should)."
1276 #: build/C/man2/madvise.2:379
1278 "B<getrlimit>(2), B<mincore>(2), B<mmap>(2), B<mprotect>(2), B<msync>(2), "
1279 "B<munmap>(2), B<core>(5)"
1283 #: build/C/man3/mallinfo.3:26
1289 #: build/C/man3/mallinfo.3:26 build/C/man3/malloc_stats.3:26
1295 #: build/C/man3/mallinfo.3:29
1296 msgid "mallinfo - obtain memory allocation information"
1300 #: build/C/man3/mallinfo.3:31 build/C/man3/malloc_stats.3:31 build/C/man3/malloc_trim.3:31 build/C/man3/malloc_usable_size.3:30 build/C/man3/mallopt.3:31
1301 msgid "B<#include E<lt>malloc.hE<gt>>"
1305 #: build/C/man3/mallinfo.3:33
1306 msgid "B<struct mallinfo mallinfo(void);>"
1310 #: build/C/man3/mallinfo.3:41
1312 "The B<mallinfo>() function returns a copy of a structure containing "
1313 "information about memory allocations performed by B<malloc>(3) and related "
1314 "functions. This structure is defined as follows:"
1318 #: build/C/man3/mallinfo.3:56
1321 "struct mallinfo {\n"
1322 " int arena; /* Non-mmapped space allocated (bytes) */\n"
1323 " int ordblks; /* Number of free chunks */\n"
1324 " int smblks; /* Number of free fastbin blocks */\n"
1325 " int hblks; /* Number of mmapped regions */\n"
1326 " int hblkhd; /* Space allocated in mmapped regions (bytes) */\n"
1327 " int usmblks; /* Maximum total allocated space (bytes) */\n"
1328 " int fsmblks; /* Space in freed fastbin blocks (bytes) */\n"
1329 " int uordblks; /* Total allocated space (bytes) */\n"
1330 " int fordblks; /* Total free space (bytes) */\n"
1331 " int keepcost; /* Top-most, releasable space (bytes) */\n"
1336 #: build/C/man3/mallinfo.3:62
1337 msgid "The fields of the I<mallinfo> structure contain the following information:"
1341 #: build/C/man3/mallinfo.3:62
1347 #: build/C/man3/mallinfo.3:68
1349 "The total amount of memory allocated by means other than B<mmap>(2) (i.e., "
1350 "memory allocated on the heap). This figure includes both in-use blocks and "
1351 "blocks on the free list."
1355 #: build/C/man3/mallinfo.3:68
1361 #: build/C/man3/mallinfo.3:71
1362 msgid "The number of ordinary (i.e., non-fastbin) free blocks."
1366 #: build/C/man3/mallinfo.3:71
1372 #: build/C/man3/mallinfo.3:75
1373 msgid "The number of fastbin free blocks (see B<mallopt>(3))."
1377 #: build/C/man3/mallinfo.3:75
1383 #: build/C/man3/mallinfo.3:83
1385 "The number of blocks currently allocated using B<mmap>(2). (See the "
1386 "discussion of B<M_MMAP_THRESHOLD> in B<mallopt>(3).)"
1390 #: build/C/man3/mallinfo.3:83
1396 #: build/C/man3/mallinfo.3:87
1397 msgid "The number of bytes in blocks currently allocated using B<mmap>(2)."
1401 #: build/C/man3/mallinfo.3:87
1407 #: build/C/man3/mallinfo.3:92
1409 "The \"highwater mark\" for allocated space\\(emthat is, the maximum amount "
1410 "of space that was ever allocated. This field is maintained only in "
1411 "nonthreading environments."
1415 #: build/C/man3/mallinfo.3:92
1421 #: build/C/man3/mallinfo.3:95
1422 msgid "The total number of bytes in fastbin free blocks."
1426 #: build/C/man3/mallinfo.3:95
1432 #: build/C/man3/mallinfo.3:98
1433 msgid "The total number of bytes used by in-use allocations."
1437 #: build/C/man3/mallinfo.3:98
1443 #: build/C/man3/mallinfo.3:101
1444 msgid "The total number of bytes in free blocks."
1448 #: build/C/man3/mallinfo.3:101
1454 #. Available already in glibc 2.0, possibly earlier
1456 #: build/C/man3/mallinfo.3:110
1458 "The total amount of releasable free space at the top of the heap. This is "
1459 "the maximum number of bytes that could ideally (i.e., ignoring page "
1460 "alignment restrictions, and so on) be released by B<malloc_trim>(3)."
1464 #: build/C/man3/mallinfo.3:114
1466 "This function is not specified by POSIX or the C standards. A similar "
1467 "function exists on many System V derivatives, and was specified in the SVID."
1470 #. FIXME http://sourceware.org/bugzilla/show_bug.cgi?id=208
1471 #. See the 24 Aug 2011 mail by Paul Pluzhnikov:
1472 #. "[patch] Fix mallinfo() to accumulate results for all arenas"
1473 #. on libc-alpha@sourceware.org
1475 #: build/C/man3/mallinfo.3:126
1477 "B<Information is returned for only the main memory allocation area.> "
1478 "Allocations in other arenas are excluded. See B<malloc_stats>(3) and "
1479 "B<malloc_info>(3) for alternatives that include information about other "
1484 #: build/C/man3/mallinfo.3:134
1486 "The fields of the I<mallinfo> structure are typed as I<int>. However, "
1487 "because some internal bookkeeping values may be of type I<long>, the "
1488 "reported values may wrap around zero and thus be inaccurate."
1492 #: build/C/man3/mallinfo.3:134 build/C/man3/malloc_hook.3:89 build/C/man3/malloc_info.3:82 build/C/man3/mallopt.3:466 build/C/man3/mcheck.3:159 build/C/man2/mmap.2:630 build/C/man2/mprotect.2:148 build/C/man3/mtrace.3:108
1498 #: build/C/man3/mallinfo.3:140
1500 "The program below employs B<mallinfo>() to retrieve memory allocation "
1501 "statistics before and after allocating and freeing some blocks of memory. "
1502 "The statistics are displayed on standard output."
1506 #: build/C/man3/mallinfo.3:144
1508 "The first two command-line arguments specify the number and size of blocks "
1509 "to be allocated with B<malloc>(3)."
1513 #: build/C/man3/mallinfo.3:158
1515 "The remaining three arguments specify which of the allocated blocks should "
1516 "be freed with B<free>(3). These three arguments are optional, and specify "
1517 "(in order): the step size to be used in the loop that frees blocks (the "
1518 "default is 1, meaning free all blocks in the range); the ordinal position of "
1519 "the first block to be freed (default 0, meaning the first allocated block); "
1520 "and a number one greater than the ordinal position of the last block to be "
1521 "freed (default is one greater than the maximum block number). If these "
1522 "three arguments are omitted, then the defaults cause all allocated blocks to "
1527 #: build/C/man3/mallinfo.3:162
1529 "In the following example run of the program, 1000 allocations of 100 bytes "
1530 "are performed, and then every second allocated block is freed:"
1534 #: build/C/man3/mallinfo.3:177
1537 "$ B<./a.out 1000 100 2>\n"
1538 "============== Before allocating blocks ==============\n"
1539 "Total non-mmapped bytes (arena): 0\n"
1540 "# of free chunks (ordblks): 1\n"
1541 "# of free fastbin blocks (smblks): 0\n"
1542 "# of mapped regions (hblks): 0\n"
1543 "Bytes in mapped regions (hblkhd): 0\n"
1544 "Max. total allocated space (usmblks): 0\n"
1545 "Free bytes held in fastbins (fsmblks): 0\n"
1546 "Total allocated space (uordblks): 0\n"
1547 "Total free space (fordblks): 0\n"
1548 "Topmost releasable block (keepcost): 0\n"
1552 #: build/C/man3/mallinfo.3:189
1555 "============== After allocating blocks ==============\n"
1556 "Total non-mmapped bytes (arena): 135168\n"
1557 "# of free chunks (ordblks): 1\n"
1558 "# of free fastbin blocks (smblks): 0\n"
1559 "# of mapped regions (hblks): 0\n"
1560 "Bytes in mapped regions (hblkhd): 0\n"
1561 "Max. total allocated space (usmblks): 0\n"
1562 "Free bytes held in fastbins (fsmblks): 0\n"
1563 "Total allocated space (uordblks): 104000\n"
1564 "Total free space (fordblks): 31168\n"
1565 "Topmost releasable block (keepcost): 31168\n"
1569 #: build/C/man3/mallinfo.3:201
1572 "============== After freeing blocks ==============\n"
1573 "Total non-mmapped bytes (arena): 135168\n"
1574 "# of free chunks (ordblks): 501\n"
1575 "# of free fastbin blocks (smblks): 0\n"
1576 "# of mapped regions (hblks): 0\n"
1577 "Bytes in mapped regions (hblkhd): 0\n"
1578 "Max. total allocated space (usmblks): 0\n"
1579 "Free bytes held in fastbins (fsmblks): 0\n"
1580 "Total allocated space (uordblks): 52000\n"
1581 "Total free space (fordblks): 83168\n"
1582 "Topmost releasable block (keepcost): 31168\n"
1586 #: build/C/man3/mallinfo.3:203 build/C/man3/malloc_info.3:165 build/C/man3/mallopt.3:532 build/C/man3/mcheck.3:176 build/C/man2/mprotect.2:165
1588 msgid "Program source"
1592 #: build/C/man3/mallinfo.3:208
1595 "#include E<lt>malloc.hE<gt>\n"
1596 "#include \"tlpi_hdr.h\"\n"
1600 #: build/C/man3/mallinfo.3:213
1604 "display_mallinfo(void)\n"
1606 " struct mallinfo mi;\n"
1610 #: build/C/man3/mallinfo.3:215
1612 msgid " mi = mallinfo();\n"
1616 #: build/C/man3/mallinfo.3:227
1619 " printf(\"Total non-mmapped bytes (arena): %d\\en\", mi.arena);\n"
1620 " printf(\"# of free chunks (ordblks): %d\\en\", mi.ordblks);\n"
1621 " printf(\"# of free fastbin blocks (smblks): %d\\en\", mi.smblks);\n"
1622 " printf(\"# of mapped regions (hblks): %d\\en\", mi.hblks);\n"
1623 " printf(\"Bytes in mapped regions (hblkhd): %d\\en\", mi.hblkhd);\n"
1624 " printf(\"Max. total allocated space (usmblks): %d\\en\", mi.usmblks);\n"
1625 " printf(\"Free bytes held in fastbins (fsmblks): %d\\en\", mi.fsmblks);\n"
1626 " printf(\"Total allocated space (uordblks): %d\\en\", "
1628 " printf(\"Total free space (fordblks): %d\\en\", "
1630 " printf(\"Topmost releasable block (keepcost): %d\\en\", "
1636 #: build/C/man3/mallinfo.3:235
1640 "main(int argc, char *argv[])\n"
1642 "#define MAX_ALLOCS 2000000\n"
1643 " char *alloc[MAX_ALLOCS];\n"
1644 " int numBlocks, j, freeBegin, freeEnd, freeStep;\n"
1645 " size_t blockSize;\n"
1649 #: build/C/man3/mallinfo.3:239
1652 " if (argc E<lt> 3 || strcmp(argv[1], \"--help\") == 0)\n"
1653 " usageErr(\"%s num-blocks block-size [free-step [start-free \"\n"
1654 " \"[end-free]]]\\en\", argv[0]);\n"
1658 #: build/C/man3/mallinfo.3:245
1661 " numBlocks = atoi(argv[1]);\n"
1662 " blockSize = atoi(argv[2]);\n"
1663 " freeStep = (argc E<gt> 3) ? atoi(argv[3]) : 1;\n"
1664 " freeBegin = (argc E<gt> 4) ? atoi(argv[4]) : 0;\n"
1665 " freeEnd = (argc E<gt> 5) ? atoi(argv[5]) : numBlocks;\n"
1669 #: build/C/man3/mallinfo.3:248
1672 " printf(\"============== Before allocating blocks "
1673 "==============\\en\");\n"
1674 " display_mallinfo();\n"
1678 #: build/C/man3/mallinfo.3:252
1681 " for (j = 0; j E<lt> numBlocks; j++) {\n"
1682 " if (numBlocks E<gt>= MAX_ALLOCS)\n"
1683 " fatal(\"Too many allocations\");\n"
1687 #: build/C/man3/mallinfo.3:257
1690 " alloc[j] = malloc(blockSize);\n"
1691 " if (alloc[j] == NULL)\n"
1692 " errExit(\"malloc\");\n"
1697 #: build/C/man3/mallinfo.3:260
1700 " printf(\"\\en============== After allocating blocks "
1701 "==============\\en\");\n"
1702 " display_mallinfo();\n"
1706 #: build/C/man3/mallinfo.3:263
1709 " for (j = freeBegin; j E<lt> freeEnd; j += freeStep)\n"
1710 " free(alloc[j]);\n"
1714 #: build/C/man3/mallinfo.3:266
1717 " printf(\"\\en============== After freeing blocks "
1718 "==============\\en\");\n"
1719 " display_mallinfo();\n"
1723 #: build/C/man3/mallinfo.3:269 build/C/man3/malloc_info.3:253 build/C/man3/mallopt.3:565 build/C/man3/mcheck.3:203 build/C/man2/mmap.2:711
1726 " exit(EXIT_SUCCESS);\n"
1731 #: build/C/man3/mallinfo.3:279
1733 "B<mmap>(2), B<malloc>(3), B<malloc_info>(3), B<malloc_stats>(3), "
1734 "B<malloc_trim>(3), B<mallopt>(3)"
1738 #: build/C/man3/malloc_get_state.3:25
1740 msgid "MALLOC_GET_STATE"
1744 #: build/C/man3/malloc_get_state.3:25
1750 #: build/C/man3/malloc_get_state.3:28
1752 "malloc_get_state, malloc_set_state - record and restore state of malloc "
1757 #: build/C/man3/malloc_get_state.3:31 build/C/man3/posix_memalign.3:35
1759 msgid "B<#include E<lt>stdlib.hE<gt>>\n"
1763 #: build/C/man3/malloc_get_state.3:33
1765 msgid "B<void* malloc_get_state(void);>\n"
1769 #: build/C/man3/malloc_get_state.3:35
1771 msgid "B<int malloc_set_state(void *>I<state>B<);>\n"
1775 #: build/C/man3/malloc_get_state.3:53
1777 "The B<malloc_get_state>() function records the current state of all "
1778 "B<malloc>(3) internal bookkeeping variables (but not the actual contents of "
1779 "the heap or the state of B<malloc_hook>(3) functions pointers). The state "
1780 "is recorded in a system-dependent opaque data structure dynamically "
1781 "allocated via B<malloc>(3), and a pointer to that data structure is returned "
1782 "as the function result. (It is the caller's responsibility to B<free>(3) "
1787 #: build/C/man3/malloc_get_state.3:61
1789 "The B<malloc_set_state>() function restores the state of all B<malloc>(3) "
1790 "internal bookkeeping variables to the values recorded in the opaque data "
1791 "structure pointed to by I<state>."
1795 #: build/C/man3/malloc_get_state.3:68
1797 "On success, B<malloc_get_state>() returns a pointer to a newly allocated "
1798 "opaque data structure. On error (for example, memory could not be allocated "
1799 "for the data structure), B<malloc_get_state>() returns NULL."
1802 #. if(ms->magic != MALLOC_STATE_MAGIC) return -1;
1803 #. /* Must fail if the major version is too high. */
1804 #. if((ms->version & ~0xffl) > (MALLOC_STATE_VERSION & ~0xffl)) return -2;
1806 #: build/C/man3/malloc_get_state.3:86
1808 "On success, B<malloc_set_state>() returns 0. If the implementation detects "
1809 "that I<state> does not point to a correctly formed data structure, "
1810 "B<malloc_set_state>() returns -1. If the implementation detects that the "
1811 "version of the data structure referred to by I<state> is a more recent "
1812 "version than this implementation knows about, B<malloc_set_state>() returns "
1817 #: build/C/man3/malloc_get_state.3:88 build/C/man3/malloc_hook.3:82 build/C/man3/mcheck.3:147 build/C/man3/mtrace.3:80
1818 msgid "These functions are GNU extensions."
1822 #: build/C/man3/malloc_get_state.3:94
1824 "These functions are especially useful when using this B<malloc>(3) "
1825 "implementation as part of a shared library, and the heap contents are "
1826 "saved/restored via some other method. This technique is used by the GNU "
1827 "Emacs to implement its \"dumping\" function."
1830 #. i.e., calls __malloc_check_init()
1831 #. i.e., malloc checking is not already in use
1832 #. and the caller requested malloc checking
1834 #: build/C/man3/malloc_get_state.3:111
1836 "Hook function pointers are never saved or restored by these functions, with "
1837 "two exceptions: if malloc checking (see B<mallopt>(3)) was in use when "
1838 "B<malloc_get_state>() was called, then B<malloc_set_state>() resets malloc "
1839 "checking hooks if possible; if malloc checking was not in use in the "
1840 "recorded state, but the caller has requested malloc checking, then the hooks "
1845 #: build/C/man3/malloc_get_state.3:114
1846 msgid "B<malloc>(3), B<mallopt>(3)"
1850 #: build/C/man3/malloc_hook.3:10
1856 #: build/C/man3/malloc_hook.3:10
1862 #: build/C/man3/malloc_hook.3:15
1864 "__malloc_hook, __malloc_initialize_hook, __memalign_hook, __free_hook, "
1865 "__realloc_hook, __after_morecore_hook - malloc debugging variables"
1869 #: build/C/man3/malloc_hook.3:18 build/C/man3/malloc_info.3:31 build/C/man3/posix_memalign.3:41
1871 msgid "B<#include E<lt>malloc.hE<gt>>\n"
1875 #: build/C/man3/malloc_hook.3:20
1877 msgid "B<void *(*__malloc_hook)(size_t >I<size>B<, const void *>I<caller>B<);>\n"
1881 #: build/C/man3/malloc_hook.3:23
1884 "B<void *(*__realloc_hook)(void *>I<ptr>B<, size_t >I<size>B<, const void "
1885 "*>I<caller>B<);>\n"
1889 #: build/C/man3/malloc_hook.3:26
1892 "B<void *(*__memalign_hook)(size_t >I<alignment>B<, size_t >I<size>B<,>\n"
1893 "B< const void *>I<caller>B<);>\n"
1897 #: build/C/man3/malloc_hook.3:28
1899 msgid "B<void (*__free_hook)(void *>I<ptr>B<, const void *>I<caller>B<);>\n"
1903 #: build/C/man3/malloc_hook.3:30
1905 msgid "B<void (*__malloc_initialize_hook)(void);>\n"
1909 #: build/C/man3/malloc_hook.3:32
1911 msgid "B<void (*__after_morecore_hook)(void);>\n"
1915 #: build/C/man3/malloc_hook.3:43
1917 "The GNU C library lets you modify the behavior of B<malloc>(3), "
1918 "B<realloc>(3), and B<free>(3) by specifying appropriate hook functions. "
1919 "You can use these hooks to help you debug programs that use dynamic memory "
1920 "allocation, for example."
1924 #: build/C/man3/malloc_hook.3:50
1926 "The variable B<__malloc_initialize_hook> points at a function that is called "
1927 "once when the malloc implementation is initialized. This is a weak "
1928 "variable, so it can be overridden in the application with a definition like "
1933 #: build/C/man3/malloc_hook.3:53
1935 msgid " void (*__malloc_initialize_hook)(void) = my_init_hook;\n"
1939 #: build/C/man3/malloc_hook.3:58
1940 msgid "Now the function I<my_init_hook>() can do the initialization of all hooks."
1944 #: build/C/man3/malloc_hook.3:74
1946 "The four functions pointed to by B<__malloc_hook>, B<__realloc_hook>, "
1947 "B<__memalign_hook>, B<__free_hook> have a prototype like the functions "
1948 "B<malloc>(3), B<realloc>(3), B<memalign>(3), B<free>(3), respectively, "
1949 "except that they have a final argument I<caller> that gives the address of "
1950 "the caller of B<malloc>(3), etc."
1954 #: build/C/man3/malloc_hook.3:80
1956 "The variable B<__after_morecore_hook> points at a function that is called "
1957 "each time after B<sbrk>(2) was asked for more memory."
1960 #. https://bugzilla.redhat.com/show_bug.cgi?id=450187
1961 #. http://sourceware.org/bugzilla/show_bug.cgi?id=9957
1963 #: build/C/man3/malloc_hook.3:89
1965 "The use of these hook functions is not safe in multithreaded programs, and "
1966 "they are now deprecated. Programmers should instead preempt calls to the "
1967 "relevant functions by defining and exporting functions such as \"malloc\" "
1972 #: build/C/man3/malloc_hook.3:91
1973 msgid "Here is a short example of how to use these variables."
1977 #: build/C/man3/malloc_hook.3:95
1980 "#include E<lt>stdio.hE<gt>\n"
1981 "#include E<lt>malloc.hE<gt>\n"
1985 #: build/C/man3/malloc_hook.3:99
1988 "/* Prototypes for our hooks. */\n"
1989 "static void my_init_hook(void);\n"
1990 "static void *my_malloc_hook(size_t, const void *);\n"
1994 #: build/C/man3/malloc_hook.3:102
1997 "/* Variables to save original hooks. */\n"
1998 "static void *(*old_malloc_hook)(size_t, const void *);\n"
2002 #: build/C/man3/malloc_hook.3:105
2005 "/* Override initializing hook from the C library. */\n"
2006 "void (*__malloc_initialize_hook) (void) = my_init_hook;\n"
2010 #: build/C/man3/malloc_hook.3:112
2014 "my_init_hook(void)\n"
2016 " old_malloc_hook = __malloc_hook;\n"
2017 " __malloc_hook = my_malloc_hook;\n"
2022 #: build/C/man3/malloc_hook.3:117
2026 "my_malloc_hook(size_t size, const void *caller)\n"
2032 #: build/C/man3/malloc_hook.3:120
2035 " /* Restore all old hooks */\n"
2036 " __malloc_hook = old_malloc_hook;\n"
2040 #: build/C/man3/malloc_hook.3:123
2043 " /* Call recursively */\n"
2044 " result = malloc(size);\n"
2048 #: build/C/man3/malloc_hook.3:126
2051 " /* Save underlying hooks */\n"
2052 " old_malloc_hook = __malloc_hook;\n"
2056 #: build/C/man3/malloc_hook.3:130
2059 " /* printf() might call malloc(), so protect it too. */\n"
2060 " printf(\"malloc(%u) called from %p returns %p\\en\",\n"
2061 " (unsigned int) size, caller, result);\n"
2065 #: build/C/man3/malloc_hook.3:133
2068 " /* Restore our own hooks */\n"
2069 " __malloc_hook = my_malloc_hook;\n"
2073 #: build/C/man3/malloc_hook.3:136
2081 #: build/C/man3/malloc_hook.3:142
2082 msgid "B<mallinfo>(3), B<malloc>(3), B<mcheck>(3), B<mtrace>(3)"
2086 #: build/C/man3/malloc_info.3:25
2092 #: build/C/man3/malloc_info.3:28
2093 msgid "malloc_info - export malloc state to a stream"
2097 #: build/C/man3/malloc_info.3:33
2099 msgid "B<int malloc_info(int >I<options>B<, FILE *>I<fp>B<);>\n"
2103 #: build/C/man3/malloc_info.3:44
2105 "The B<malloc_info>() function exports an XML string that describes the "
2106 "current state of the memory-allocation implementation in the caller. The "
2107 "string is printed on the file stream I<fp>. The exported string includes "
2108 "information about all arenas (see B<malloc>(3))."
2112 #: build/C/man3/malloc_info.3:48
2113 msgid "As currently implemented, I<options> must be zero."
2117 #: build/C/man3/malloc_info.3:53
2118 msgid "On success, B<malloc_info>() returns 0; on error, it returns -1."
2122 #: build/C/man3/malloc_info.3:58
2123 msgid "I<options> was nonzero."
2127 #: build/C/man3/malloc_info.3:61
2128 msgid "B<malloc_info>() was added to glibc in version 2.10."
2132 #: build/C/man3/malloc_info.3:63 build/C/man3/malloc_stats.3:57 build/C/man3/malloc_trim.3:62 build/C/man3/malloc_usable_size.3:50
2133 msgid "This function is a GNU extension."
2137 #: build/C/man3/malloc_info.3:69
2139 "The memory-allocation information is provided as an XML string (rather than "
2140 "a C structure) because the information may change over time (according to "
2141 "changes in the underlying implementation). The output XML string includes a "
2146 #: build/C/man3/malloc_info.3:75
2148 "The B<open_memstream>(3) function can be used to send the output of "
2149 "B<malloc_info>() directly into a buffer in memory, rather than to a file."
2153 #: build/C/man3/malloc_info.3:82
2155 "The B<malloc_info>() function is designed to address deficiencies in "
2156 "B<malloc_stats>(3) and B<mallinfo>(3)."
2160 #: build/C/man3/malloc_info.3:93
2162 "The program below takes up to four command-line arguments, of which the "
2163 "first three are mandatory. The first argument specifies the number of "
2164 "threads that the program should create. All of the threads, including the "
2165 "main thread, allocate the number of blocks of memory specified by the second "
2166 "argument. The third argument controls the size of the blocks to be "
2167 "allocated. The main thread creates blocks of this size, the second thread "
2168 "created by the program allocates blocks of twice this size, the third thread "
2169 "allocates blocks of three times this size, and so on."
2173 #: build/C/man3/malloc_info.3:100
2175 "The program calls B<malloc_info>() twice to display the memory-allocation "
2176 "state. The first call takes place before any threads are created or memory "
2177 "allocated. The second call is performed after all threads have allocated "
2182 #: build/C/man3/malloc_info.3:108
2184 "In the following example, the command-line arguments specify the creation of "
2185 "one additional thread, and both the main thread and the additional thread "
2186 "allocate 10000 blocks of memory. After the blocks of memory have been "
2187 "allocated, B<malloc_info>() shows the state of two allocation arenas."
2191 #: build/C/man3/malloc_info.3:133
2194 "$ B<getconf GNU_LIBC_VERSION>\n"
2196 "$ B<./a.out 1 10000 100>\n"
2197 "============ Before allocating blocks ============\n"
2198 "E<lt>malloc version=\"1\"E<gt>\n"
2199 "E<lt>heap nr=\"0\"E<gt>\n"
2201 "E<lt>/sizesE<gt>\n"
2202 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2203 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2204 "E<lt>system type=\"current\" size=\"135168\"/E<gt>\n"
2205 "E<lt>system type=\"max\" size=\"135168\"/E<gt>\n"
2206 "E<lt>aspace type=\"total\" size=\"135168\"/E<gt>\n"
2207 "E<lt>aspace type=\"mprotect\" size=\"135168\"/E<gt>\n"
2209 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2210 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2211 "E<lt>system type=\"current\" size=\"135168\"/E<gt>\n"
2212 "E<lt>system type=\"max\" size=\"135168\"/E<gt>\n"
2213 "E<lt>aspace type=\"total\" size=\"135168\"/E<gt>\n"
2214 "E<lt>aspace type=\"mprotect\" size=\"135168\"/E<gt>\n"
2215 "E<lt>/mallocE<gt>\n"
2219 #: build/C/man3/malloc_info.3:163
2222 "============ After allocating blocks ============\n"
2223 "E<lt>malloc version=\"1\"E<gt>\n"
2224 "E<lt>heap nr=\"0\"E<gt>\n"
2226 "E<lt>/sizesE<gt>\n"
2227 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2228 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2229 "E<lt>system type=\"current\" size=\"1081344\"/E<gt>\n"
2230 "E<lt>system type=\"max\" size=\"1081344\"/E<gt>\n"
2231 "E<lt>aspace type=\"total\" size=\"1081344\"/E<gt>\n"
2232 "E<lt>aspace type=\"mprotect\" size=\"1081344\"/E<gt>\n"
2234 "E<lt>heap nr=\"1\"E<gt>\n"
2236 "E<lt>/sizesE<gt>\n"
2237 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2238 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2239 "E<lt>system type=\"current\" size=\"1032192\"/E<gt>\n"
2240 "E<lt>system type=\"max\" size=\"1032192\"/E<gt>\n"
2241 "E<lt>aspace type=\"total\" size=\"1032192\"/E<gt>\n"
2242 "E<lt>aspace type=\"mprotect\" size=\"1032192\"/E<gt>\n"
2244 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2245 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2246 "E<lt>system type=\"current\" size=\"2113536\"/E<gt>\n"
2247 "E<lt>system type=\"max\" size=\"2113536\"/E<gt>\n"
2248 "E<lt>aspace type=\"total\" size=\"2113536\"/E<gt>\n"
2249 "E<lt>aspace type=\"mprotect\" size=\"2113536\"/E<gt>\n"
2250 "E<lt>/mallocE<gt>\n"
2254 #: build/C/man3/malloc_info.3:173
2257 "#include E<lt>unistd.hE<gt>\n"
2258 "#include E<lt>stdlib.hE<gt>\n"
2259 "#include E<lt>pthread.hE<gt>\n"
2260 "#include E<lt>malloc.hE<gt>\n"
2261 "#include E<lt>errno.hE<gt>\n"
2265 #: build/C/man3/malloc_info.3:176
2268 "static size_t blockSize;\n"
2269 "static int numThreads, numBlocks;\n"
2273 #: build/C/man3/malloc_info.3:179
2276 "#define errExit(msg) do { perror(msg); exit(EXIT_FAILURE); \\e\n"
2281 #: build/C/man3/malloc_info.3:185
2285 "thread_func(void *arg)\n"
2288 " int tn = (int) arg;\n"
2292 #: build/C/man3/malloc_info.3:188
2295 " /* The multiplier \\(aq(2 + tn)\\(aq ensures that each thread "
2297 " the main thread) allocates a different amount of memory */\n"
2301 #: build/C/man3/malloc_info.3:192
2304 " for (j = 0; j E<lt> numBlocks; j++)\n"
2305 " if (malloc(blockSize * (2 + tn)) == NULL)\n"
2306 " errExit(\"malloc-thread\");\n"
2310 #: build/C/man3/malloc_info.3:196
2313 " sleep(100); /* Sleep until main thread terminates */\n"
2319 #: build/C/man3/malloc_info.3:202
2323 "main(int argc, char *argv[])\n"
2325 " int j, tn, sleepTime;\n"
2326 " pthread_t *thr;\n"
2330 #: build/C/man3/malloc_info.3:209
2333 " if (argc E<lt> 4) {\n"
2334 " fprintf(stderr,\n"
2335 " \"%s num-threads num-blocks block-size [sleep-time]\\en\",\n"
2337 " exit(EXIT_FAILURE);\n"
2342 #: build/C/man3/malloc_info.3:214
2345 " numThreads = atoi(argv[1]);\n"
2346 " numBlocks = atoi(argv[2]);\n"
2347 " blockSize = atoi(argv[3]);\n"
2348 " sleepTime = (argc E<gt> 4) ? atoi(argv[4]) : 0;\n"
2352 #: build/C/man3/malloc_info.3:218
2355 " thr = calloc(numThreads, sizeof(pthread_t));\n"
2356 " if (thr == NULL)\n"
2357 " errExit(\"calloc\");\n"
2361 #: build/C/man3/malloc_info.3:221
2364 " printf(\"============ Before allocating blocks ============\\en\");\n"
2365 " malloc_info(0, stdout);\n"
2369 #: build/C/man3/malloc_info.3:223
2371 msgid " /* Create threads that allocate different amounts of memory */\n"
2375 #: build/C/man3/malloc_info.3:229
2378 " for (tn = 0; tn E<lt> numThreads; tn++) {\n"
2379 " errno = pthread_create(&thr[tn], NULL, thread_func,\n"
2381 " if (errno != 0)\n"
2382 " errExit(\"pthread_create\");\n"
2386 #: build/C/man3/malloc_info.3:234
2389 " /* If we add a sleep interval after the start-up of each\n"
2390 " thread, the threads likely won\\(aqt contend for malloc\n"
2391 " mutexes, and therefore additional arenas won\\(aqt be\n"
2392 " allocated (see malloc(3)). */\n"
2396 #: build/C/man3/malloc_info.3:238
2399 " if (sleepTime E<gt> 0)\n"
2400 " sleep(sleepTime);\n"
2405 #: build/C/man3/malloc_info.3:240
2407 msgid " /* The main thread also allocates some memory */\n"
2411 #: build/C/man3/malloc_info.3:244
2414 " for (j = 0; j E<lt> numBlocks; j++)\n"
2415 " if (malloc(blockSize) == NULL)\n"
2416 " errExit(\"malloc\");\n"
2420 #: build/C/man3/malloc_info.3:247
2423 " sleep(2); /* Give all threads a chance to\n"
2424 " complete allocations */\n"
2428 #: build/C/man3/malloc_info.3:250
2431 " printf(\"\\en============ After allocating blocks ============\\en\");\n"
2432 " malloc_info(0, stdout);\n"
2436 #: build/C/man3/malloc_info.3:260
2438 "B<mallinfo>(3), B<malloc>(3), B<malloc_stats>(3), B<mallopt>(3), "
2439 "B<open_memstream>(3)"
2443 #: build/C/man3/malloc_stats.3:26
2445 msgid "MALLOC_STATS"
2449 #: build/C/man3/malloc_stats.3:29
2450 msgid "malloc_stats - print memory allocation statistics"
2454 #: build/C/man3/malloc_stats.3:33
2455 msgid "B<void malloc_stats(void);>"
2459 #. Available already in glibc 2.0, possibly earlier
2461 #: build/C/man3/malloc_stats.3:55
2463 "The B<malloc_stats>() function prints (on standard error) statistics about "
2464 "memory allocated by B<malloc>(3) and related functions. For each arena "
2465 "(allocation area), this function prints the total amount of memory allocated "
2466 "and the total number of bytes consumed by in-use allocations. (These two "
2467 "values correspond to the I<arena> and I<uordblks> fields retrieved by "
2468 "B<mallinfo>(3).) In addition, the function prints the sum of these two "
2469 "statistics for all arenas, and the maximum number of blocks and bytes that "
2470 "were ever simultaneously allocated using B<mmap>(2)."
2474 #: build/C/man3/malloc_stats.3:61
2476 "More detailed information about memory allocations in the main arena can be "
2477 "obtained using B<mallinfo>(3)."
2481 #: build/C/man3/malloc_stats.3:67
2482 msgid "B<mmap>(2), B<mallinfo>(3), B<malloc>(3), B<malloc_info>(3), B<mallopt>(3)"
2486 #: build/C/man3/malloc_trim.3:26
2492 #: build/C/man3/malloc_trim.3:26 build/C/man3/malloc_usable_size.3:25
2498 #: build/C/man3/malloc_trim.3:29
2499 msgid "malloc_trim - release free memory from the top of the heap"
2503 #: build/C/man3/malloc_trim.3:33
2504 msgid "B<void malloc_trim(size_t >I<pad>B<);>"
2508 #: build/C/man3/malloc_trim.3:40
2510 "The B<malloc_trim>() function attempts to release free memory at the top of "
2511 "the heap (by calling B<sbrk>(2) with a suitable argument)."
2515 #: build/C/man3/malloc_trim.3:51
2517 "The I<pad> argument specifies the amount of free space to leave untrimmed at "
2518 "the top of the heap. If this argument is 0, only the minimum amount of "
2519 "memory is maintained at the top of the heap (i.e., one page or less). A "
2520 "nonzero argument can be used to maintain some trailing space at the top of "
2521 "the heap in order to allow future allocations to be made without having to "
2522 "extend the heap with B<sbrk>(2)."
2526 #: build/C/man3/malloc_trim.3:56
2528 "The B<malloc_trim>() function returns 1 if memory was actually released "
2529 "back to the system, or 0 if it was not possible to release any memory."
2533 #. Available already in glibc 2.0, possibly earlier
2535 #: build/C/man3/malloc_trim.3:60
2536 msgid "No errors are defined."
2540 #: build/C/man3/malloc_trim.3:71
2542 "This function is automatically called by B<free>(3) in certain "
2543 "circumstances; see the discussion of B<M_TOP_PAD> and B<M_TRIM_THRESHOLD> in "
2548 #: build/C/man3/malloc_trim.3:74
2550 "This function cannot release free memory located at places other than the "
2554 #. malloc/malloc.c::mTRIm():
2555 #. return result | (av == &main_arena ? sYSTRIm (pad, av) : 0);
2557 #: build/C/man3/malloc_trim.3:78
2558 msgid "This function releases only memory in the main arena."
2562 #: build/C/man3/malloc_trim.3:82
2563 msgid "B<sbrk>(2), B<malloc>(3), B<mallopt>(3)"
2567 #: build/C/man3/malloc_usable_size.3:25
2569 msgid "MALLOC_USABLE_SIZE"
2573 #: build/C/man3/malloc_usable_size.3:28
2574 msgid "malloc_usable_size - obtain size of block of memory allocated from heap"
2578 #: build/C/man3/malloc_usable_size.3:32
2579 msgid "B<size_t malloc_usable_size (void *>I<ptr>B<);>"
2583 #: build/C/man3/malloc_usable_size.3:40
2585 "The B<malloc_usable_size>() function returns the number of usable bytes in "
2586 "the block pointed to by I<ptr>, a pointer to a block of memory allocated by "
2587 "B<malloc>(3) or a related function."
2591 #: build/C/man3/malloc_usable_size.3:48
2593 "B<malloc_usable_size>() returns the number of usable bytes in the block of "
2594 "allocated memory pointed to by I<ptr>. If I<ptr> is NULL, 0 is returned."
2598 #: build/C/man3/malloc_usable_size.3:60
2600 "The value returned by B<malloc_usable_size>() may be greater than the "
2601 "requested size of the allocation because of alignment and minimum size "
2602 "constraints. Although the excess bytes can be overwritten by the "
2603 "application without ill effects, this is not good programming practice: the "
2604 "number of excess bytes in an allocation depends on the underlying "
2609 #: build/C/man3/malloc_usable_size.3:62
2610 msgid "The main use of this function is for debugging and introspection."
2614 #: build/C/man3/malloc_usable_size.3:64
2615 msgid "B<malloc>(3)"
2619 #: build/C/man3/mallopt.3:26
2625 #: build/C/man3/mallopt.3:26
2631 #: build/C/man3/mallopt.3:29
2632 msgid "mallopt - set memory allocation parameters"
2636 #: build/C/man3/mallopt.3:33
2637 msgid "B<int mallopt(int >I<param>B<, int >I<value>B<);>"
2641 #: build/C/man3/mallopt.3:44
2643 "The B<mallopt>() function adjusts parameters that control the behavior of "
2644 "the memory-allocation functions (see B<malloc>(3)). The I<param> argument "
2645 "specifies the parameter to be modified, and I<value> specifies the new value "
2646 "for that parameter."
2650 #: build/C/man3/mallopt.3:47
2651 msgid "The following values can be specified for I<param>:"
2655 #: build/C/man3/mallopt.3:47
2657 msgid "B<M_CHECK_ACTION>"
2661 #: build/C/man3/mallopt.3:53
2663 "Setting this parameter controls how glibc responds when various kinds of "
2664 "programming errors are detected (e.g., freeing the same pointer twice). The "
2665 "3 least significant bits (2, 1, and 0) of the value assigned to this "
2666 "parameter determine the glibc behavior, as follows:"
2670 #: build/C/man3/mallopt.3:54
2676 #: build/C/man3/mallopt.3:64
2678 "If this bit is set, then print a one-line message on I<stderr> that provides "
2679 "details about the error. The message starts with the string \"***\\ glibc "
2680 "detected\\ ***\", followed by the program name, the name of the "
2681 "memory-allocation function in which the error was detected, a brief "
2682 "description of the error, and the memory address where the error was "
2687 #: build/C/man3/mallopt.3:64
2693 #: build/C/man3/mallopt.3:79
2695 "If this bit is set, then, after printing any error message specified by bit "
2696 "0, the program is terminated by calling B<abort>(3). In glibc versions "
2697 "since 2.4, if bit 0 is also set, then, between printing the error message "
2698 "and aborting, the program also prints a stack trace in the manner of "
2699 "B<backtrace>(3), and prints the process's memory mapping in the style of "
2700 "I</proc/[pid]/maps> (see B<proc>(5))."
2704 #: build/C/man3/mallopt.3:79
2706 msgid "Bit 2 (since glibc 2.4)"
2710 #: build/C/man3/mallopt.3:86
2712 "This bit has an effect only if bit 0 is also set. If this bit is set, then "
2713 "the one-line message describing the error is simplified to contain just the "
2714 "name of the function where the error was detected and the brief description "
2719 #: build/C/man3/mallopt.3:91
2720 msgid "The remaining bits in I<value> are ignored."
2724 #: build/C/man3/mallopt.3:95
2726 "Combining the above details, the following numeric values are meaningful for "
2727 "B<M_CHECK_ACTION>:"
2731 #: build/C/man3/mallopt.3:96
2737 #: build/C/man3/mallopt.3:98
2738 msgid "Ignore error conditions; continue execution (with undefined results)."
2742 #: build/C/man3/mallopt.3:98
2748 #: build/C/man3/mallopt.3:100
2749 msgid "Print a detailed error message and continue execution."
2753 #: build/C/man3/mallopt.3:100
2759 #: build/C/man3/mallopt.3:102
2760 msgid "Abort the program."
2764 #: build/C/man3/mallopt.3:102
2770 #: build/C/man3/mallopt.3:105
2772 "Print detailed error message, stack trace, and memory mappings, and abort "
2777 #: build/C/man3/mallopt.3:105
2783 #: build/C/man3/mallopt.3:107
2784 msgid "Print a simple error message and continue execution."
2788 #: build/C/man3/mallopt.3:107
2794 #: build/C/man3/mallopt.3:110
2796 "Print simple error message, stack trace, and memory mappings, and abort the "
2801 #: build/C/man3/mallopt.3:116
2803 "Since glibc 2.3.4, the default value for the B<M_CHECK_ACTION> parameter is "
2804 "3. In glibc version 2.3.3 and earlier, the default value is 1."
2808 #: build/C/man3/mallopt.3:121
2810 "Using a nonzero B<M_CHECK_ACTION> value can be useful because otherwise a "
2811 "crash may happen much later, and the true cause of the problem is then very "
2812 "hard to track down."
2816 #: build/C/man3/mallopt.3:121
2818 msgid "B<M_MMAP_MAX>"
2821 #. The following text adapted from comments in the glibc source:
2823 #: build/C/man3/mallopt.3:131
2825 "This parameter specifies the maximum number of allocation requests that may "
2826 "be simultaneously serviced using B<mmap>(2). This parameter exists because "
2827 "some systems have a limited number of internal tables for use by B<mmap>(2), "
2828 "and using more than a few of them may degrade performance."
2832 #: build/C/man3/mallopt.3:138
2834 "The default value is 65,536, a value which has no special significance and "
2835 "which servers only as a safeguard. Setting this parameter to 0 disables the "
2836 "use of B<mmap>(2) for servicing large allocation requests."
2840 #: build/C/man3/mallopt.3:138
2842 msgid "B<M_MMAP_THRESHOLD>"
2846 #: build/C/man3/mallopt.3:147
2848 "For allocations greater than or equal to the limit specified (in bytes) by "
2849 "B<M_MMAP_THRESHOLD> that can't be satisfied from the free list, the "
2850 "memory-allocation functions employ B<mmap>(2) instead of increasing the "
2851 "program break using B<sbrk>(2)."
2855 #: build/C/man3/mallopt.3:167
2857 "Allocating memory using B<mmap>(2) has the significant advantage that the "
2858 "allocated memory blocks can always be independently released back to the "
2859 "system. (By contrast, the heap can be trimmed only if memory is freed at "
2860 "the top end.) On the other hand, there are some disadvantages to the use of "
2861 "B<mmap>(2): deallocated space is not placed on the free list for reuse by "
2862 "later allocations; memory may be wasted because B<mmap>(2) allocations must "
2863 "be page-aligned; and the kernel must perform the expensive task of zeroing "
2864 "out memory allocated via B<mmap>(2). Balancing these factors leads to a "
2865 "default setting of 128*1024 for the B<M_MMAP_THRESHOLD> parameter."
2869 #: build/C/man3/mallopt.3:174
2871 "The lower limit for this parameter is 0. The upper limit is "
2872 "B<DEFAULT_MMAP_THRESHOLD_MAX>: 512*1024 on 32-bit systems or "
2873 "I<4*1024*1024*sizeof(long)> on 64-bit systems."
2877 #: build/C/man3/mallopt.3:192
2879 "I<Note:> Nowadays, glibc uses a dynamic mmap threshold by default. The "
2880 "initial value of the threshold is 128*1024, but when blocks larger than the "
2881 "current threshold and less than or equal to B<DEFAULT_MMAP_THRESHOLD_MAX> "
2882 "are freed, the threshold is adjusted upwards to the size of the freed "
2883 "block. When dynamic mmap thresholding is in effect, the threshold for "
2884 "trimming the heap is also dynamically adjusted to be twice the dynamic mmap "
2885 "threshold. Dynamic adjustment of the mmap threshold is disabled if any of "
2886 "the B<M_TRIM_THRESHOLD>, B<M_TOP_PAD>, B<M_MMAP_THRESHOLD>, or B<M_MMAP_MAX> "
2887 "parameters is set."
2891 #: build/C/man3/mallopt.3:192
2893 msgid "B<M_MXFAST> (since glibc 2.3)"
2896 #. The following text adapted from comments in the glibc sources:
2898 #: build/C/man3/mallopt.3:212
2900 "Set the upper limit for memory allocation requests that are satisfied using "
2901 "\"fastbins\". (The measurement unit for this parameter is bytes.) Fastbins "
2902 "are storage areas that hold deallocated blocks of memory of the same size "
2903 "without merging adjacent free blocks. Subsequent reallocation of blocks of "
2904 "the same size can be handled very quickly by allocating from the fastbin, "
2905 "although memory fragmentation and the overall memory footprint of the "
2906 "program can increase. The default value for this parameter is "
2907 "I<64*sizeof(size_t)/4> (i.e., 64 on 32-bit architectures). The range for "
2908 "this parameter is 0 to I<80*sizeof(size_t)/4>. Setting B<M_MXFAST> to 0 "
2909 "disables the use of fastbins."
2913 #: build/C/man3/mallopt.3:212
2915 msgid "B<M_PERTURB> (since glibc 2.4)"
2919 #: build/C/man3/mallopt.3:227
2921 "If this parameter is set to a nonzero value, then bytes of allocated memory "
2922 "(other than allocations via B<calloc>(3)) are initialized to the complement "
2923 "of the value in the least significant byte of I<value>, and when allocated "
2924 "memory is released using B<free>(3), the freed bytes are set to the least "
2925 "significant byte of I<value>. This can be useful for detecting errors where "
2926 "programs incorrectly rely on allocated memory being initialized to zero, or "
2927 "reuse values in memory that has already been freed."
2931 #: build/C/man3/mallopt.3:227
2933 msgid "B<M_TOP_PAD>"
2937 #: build/C/man3/mallopt.3:234
2939 "This parameter defines the amount of padding to employ when calling "
2940 "B<sbrk>(2) to modify the program break. (The measurement unit for this "
2941 "parameter is bytes.) This parameter has an effect in the following "
2946 #: build/C/man3/mallopt.3:241
2948 "When the program break is increased, then B<M_TOP_PAD> bytes are added to "
2949 "the B<sbrk>(2) request."
2953 #: build/C/man3/mallopt.3:247
2955 "When the heap is trimmed as a consequence of calling B<free>(3) (see the "
2956 "discussion of B<M_TRIM_THRESHOLD>) this much free space is preserved at the "
2961 #: build/C/man3/mallopt.3:251
2963 "In either case, the amount of padding is always rounded to a system page "
2968 #: build/C/man3/mallopt.3:258
2970 "Modifying B<M_TOP_PAD> is a trade-off between increasing the number of "
2971 "system calls (when the parameter is set low) and wasting unused memory at "
2972 "the top of the heap (when the parameter is set high)."
2975 #. DEFAULT_TOP_PAD in glibc source
2977 #: build/C/man3/mallopt.3:261
2978 msgid "The default value for this parameter is 128*1024."
2982 #: build/C/man3/mallopt.3:261
2984 msgid "B<M_TRIM_THRESHOLD>"
2988 #: build/C/man3/mallopt.3:277
2990 "When the amount of contiguous free memory at the top of the heap grows "
2991 "sufficiently large, B<free>(3) employs B<sbrk>(2) to release this memory "
2992 "back to the system. (This can be useful in programs that continue to "
2993 "execute for a long period after freeing a significant amount of memory.) "
2994 "The B<M_TRIM_THRESHOLD> parameter specifies the minimum size (in bytes) that "
2995 "this block of memory must reach before B<sbrk>(2) is used to trim the heap."
2999 #: build/C/man3/mallopt.3:282
3001 "The default value for this parameter is 128*1024. Setting "
3002 "B<M_TRIM_THRESHOLD> to -1 disables trimming completely."
3005 #. FIXME Do the arena parameters need to be documented?
3007 #. .BR M_ARENA_TEST " (since glibc 2.10)"
3009 #. .BR M_ARENA_MAX " (since glibc 2.10)"
3011 #. Environment variables
3012 #. MALLOC_ARENA_MAX_
3013 #. MALLOC_ARENA_TEST_
3015 #. http://udrepper.livejournal.com/20948.html describes some details
3016 #. of the MALLOC_ARENA_* environment variables.
3018 #. These macros aren't enabled in production releases until 2.15?
3019 #. (see glibc malloc/Makefile)
3021 #: build/C/man3/mallopt.3:305
3023 "Modifying B<M_TRIM_THRESHOLD> is a trade-off between increasing the number "
3024 "of system calls (when the parameter is set low) and wasting unused memory "
3025 "at the top of the heap (when the parameter is set high)."
3029 #: build/C/man3/mallopt.3:305
3031 msgid "Environment variables"
3035 #: build/C/man3/mallopt.3:320
3037 "A number of environment variables can be defined to modify some of the same "
3038 "parameters as are controlled by B<mallopt>(). Using these variables has the "
3039 "advantage that the source code of the program need not be changed. To be "
3040 "effective, these variables must be defined before the first call to a "
3041 "memory-allocation function. (If the same parameters are adjusted via "
3042 "B<mallopt>() then the B<mallopt>() settings take precedence.) For "
3043 "security reasons, these variables are ignored in set-user-ID and "
3044 "set-group-ID programs."
3048 #: build/C/man3/mallopt.3:323
3050 "The environment variables are as follows (note the trailing underscore at "
3051 "the end of the name of each variable):"
3055 #: build/C/man3/mallopt.3:323
3057 msgid "B<MALLOC_CHECK_>"
3060 #. On glibc 2.12/x86, a simple malloc()+free() loop is about 70% slower
3061 #. when MALLOC_CHECK_ was set.
3063 #: build/C/man3/mallopt.3:340
3065 "This environment variable controls the same parameter as B<mallopt>() "
3066 "B<M_CHECK_ACTION>. If this variable is set to a nonzero value, then a "
3067 "special implementation of the memory-allocation functions is used. (This is "
3068 "accomplished using the B<malloc_hook>(3) feature.) This implementation "
3069 "performs additional error checking, but is slower than the standard set of "
3070 "memory-allocation functions. (This implementation does not detect all "
3071 "possible errors; memory leaks can still occur.)"
3075 #: build/C/man3/mallopt.3:345
3077 "The value assigned to this environment variable should be a single digit, "
3078 "whose meaning is as described for B<M_CHECK_ACTION>. Any characters beyond "
3079 "the initial digit are ignored."
3083 #: build/C/man3/mallopt.3:354
3085 "For security reasons, the effect of B<MALLOC_CHECK_> is disabled by default "
3086 "for set-user-ID and set-group-ID programs. However, if the file "
3087 "I</etc/suid-debug> exists (the content of the file is irrelevant), then "
3088 "B<MALLOC_CHECK_> also has an effect for set-user-ID and set-group-ID "
3093 #: build/C/man3/mallopt.3:354
3095 msgid "B<MALLOC_MMAP_MAX_>"
3099 #: build/C/man3/mallopt.3:359
3100 msgid "Controls the same parameter as B<mallopt>() B<M_MMAP_MAX>."
3104 #: build/C/man3/mallopt.3:359
3106 msgid "B<MALLOC_MMAP_THRESHOLD_>"
3110 #: build/C/man3/mallopt.3:364
3111 msgid "Controls the same parameter as B<mallopt>() B<M_MMAP_THRESHOLD>."
3115 #: build/C/man3/mallopt.3:364
3117 msgid "B<MALLOC_PERTURB_>"
3121 #: build/C/man3/mallopt.3:369
3122 msgid "Controls the same parameter as B<mallopt>() B<M_PERTURB>."
3126 #: build/C/man3/mallopt.3:369
3128 msgid "B<MALLOC_TRIM_THRESHOLD_>"
3132 #: build/C/man3/mallopt.3:374
3133 msgid "Controls the same parameter as B<mallopt>() B<M_TRIM_THRESHOLD>."
3137 #: build/C/man3/mallopt.3:374
3139 msgid "B<MALLOC_TOP_PAD_>"
3143 #: build/C/man3/mallopt.3:379
3144 msgid "Controls the same parameter as B<mallopt>() B<M_TOP_PAD>."
3148 #: build/C/man3/mallopt.3:384
3149 msgid "On success, B<mallopt>() returns 1. On error, it returns 0."
3153 #. Available already in glibc 2.0, possibly earlier
3155 #: build/C/man3/mallopt.3:392
3156 msgid "On error, I<errno> is I<not> set."
3161 #: build/C/man3/mallopt.3:406
3163 "This function is not specified by POSIX or the C standards. A similar "
3164 "function exists on many System V derivatives, but the range of values for "
3165 "I<param> varies across systems. The SVID defined options B<M_MXFAST>, "
3166 "B<M_NLBLKS>, B<M_GRAIN>, and B<M_KEEP>, but only the first of these is "
3167 "implemented in glibc."
3171 #: build/C/man3/mallopt.3:410
3172 msgid "Specifying an invalid value for I<param> does not generate an error."
3175 #. FIXME This looks buggy:
3176 #. setting the M_MXFAST limit rounds up: (s + SIZE_SZ) & ~MALLOC_ALIGN_MASK)
3177 #. malloc requests are rounded up:
3178 #. (req) + SIZE_SZ + MALLOC_ALIGN_MASK) & ~MALLOC_ALIGN_MASK
3179 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=12129
3181 #: build/C/man3/mallopt.3:418
3183 "A calculation error within the glibc implementation means that a call of the "
3188 #: build/C/man3/mallopt.3:421
3190 msgid " mallopt(M_MXFAST, n)\n"
3193 #. Bins are multiples of 2 * sizeof(size_t) + sizeof(size_t)
3195 #: build/C/man3/mallopt.3:433
3197 "does not result in fastbins being employed for all allocations of size up to "
3198 "I<n>. To ensure desired results, I<n> should be rounded up to the next "
3199 "multiple greater than or equal to I<(2k+1)*sizeof(size_t)>, where I<k> is an "
3203 #. FIXME MALLOC_MMAP_THRESHOLD_ and MALLOC_MMAP_MAX_
3204 #. do have an effect for set-user-ID programs (but not
3205 #. set-group-ID programs).
3206 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=12155
3208 #: build/C/man3/mallopt.3:445
3210 "The B<MALLOC_MMAP_THRESHOLD_> and B<MALLOC_MMAP_MAX_> variables are I<not> "
3211 "ignored in set-group-ID programs."
3214 #. FIXME http://sources.redhat.com/bugzilla/show_bug.cgi?id=12140
3216 #: build/C/man3/mallopt.3:466
3218 "If B<mallopt>() is used to set B<M_PERTURB>, then, as expected, the bytes "
3219 "of allocated memory are initialized to the complement of the byte in "
3220 "I<value>, and when that memory is freed, the bytes of the region are "
3221 "initialized to the byte specified in I<value>. However, there is an "
3222 "off-by-I<sizeof(size_t)> error in the implementation: instead of "
3223 "initializing precisely the block of memory being freed by the call "
3224 "I<free(p)>, the block starting at I<p+sizeof(size_t)> is initialized."
3228 #: build/C/man3/mallopt.3:475
3230 "The program below demonstrates the use of B<M_CHECK_ACTION>. If the program "
3231 "is supplied with an (integer) command-line argument, then that argument is "
3232 "used to set the B<M_CHECK_ACTION> parameter. The program then allocates a "
3233 "block of memory, and frees it twice (an error)."
3237 #: build/C/man3/mallopt.3:479
3239 "The following shell session shows what happens when we run this program "
3240 "under glibc, with the default value for B<M_CHECK_ACTION>:"
3244 #: build/C/man3/mallopt.3:499
3248 "main(): returned from first free() call\n"
3249 "*** glibc detected *** ./a.out: double free or corruption (top): 0x09d30008 "
3251 "======= Backtrace: =========\n"
3252 "/lib/libc.so.6(+0x6c501)[0x523501]\n"
3253 "/lib/libc.so.6(+0x6dd70)[0x524d70]\n"
3254 "/lib/libc.so.6(cfree+0x6d)[0x527e5d]\n"
3255 "\\&./a.out[0x80485db]\n"
3256 "/lib/libc.so.6(__libc_start_main+0xe7)[0x4cdce7]\n"
3257 "\\&./a.out[0x8048471]\n"
3258 "======= Memory map: ========\n"
3259 "001e4000-001fe000 r-xp 00000000 08:06 1083555 /lib/libgcc_s.so.1\n"
3260 "001fe000-001ff000 r--p 00019000 08:06 1083555 /lib/libgcc_s.so.1\n"
3261 "[some lines omitted]\n"
3262 "b7814000-b7817000 rw-p 00000000 00:00 0\n"
3263 "bff53000-bff74000 rw-p 00000000 00:00 0 [stack]\n"
3264 "Aborted (core dumped)\n"
3268 #: build/C/man3/mallopt.3:504
3270 "The following runs show the results when employing other values for "
3271 "B<M_CHECK_ACTION>:"
3275 #: build/C/man3/mallopt.3:517
3278 "$ B<./a.out 1> # Diagnose error and continue\n"
3279 "main(): returned from first free() call\n"
3280 "*** glibc detected *** ./a.out: double free or corruption (top): 0x09cbe008 "
3282 "main(): returned from second free() call\n"
3283 "$ B<./a.out 2> # Abort without error message\n"
3284 "main(): returned from first free() call\n"
3285 "Aborted (core dumped)\n"
3286 "$ B<./a.out 0> # Ignore error and continue\n"
3287 "main(): returned from first free() call\n"
3288 "main(): returned from second free() call\n"
3292 #: build/C/man3/mallopt.3:523
3294 "The next run shows how to set the same parameter using the B<MALLOC_CHECK_> "
3295 "environment variable:"
3299 #: build/C/man3/mallopt.3:530
3302 "$ B<MALLOC_CHECK_=1 ./a.out>\n"
3303 "main(): returned from first free() call\n"
3304 "*** glibc detected *** ./a.out: free(): invalid pointer: 0x092c2008 ***\n"
3305 "main(): returned from second free() call\n"
3309 #: build/C/man3/mallopt.3:538
3312 "#include E<lt>malloc.hE<gt>\n"
3313 "#include E<lt>stdio.hE<gt>\n"
3314 "#include E<lt>stdlib.hE<gt>\n"
3318 #: build/C/man3/mallopt.3:543 build/C/man3/mcheck.3:187
3322 "main(int argc, char *argv[])\n"
3328 #: build/C/man3/mallopt.3:550
3331 " if (argc E<gt> 1) {\n"
3332 " if (mallopt(M_CHECK_ACTION, atoi(argv[1])) != 1) {\n"
3333 " fprintf(stderr, \"mallopt() failed\");\n"
3334 " exit(EXIT_FAILURE);\n"
3340 #: build/C/man3/mallopt.3:556
3343 " p = malloc(1000);\n"
3344 " if (p == NULL) {\n"
3345 " fprintf(stderr, \"malloc() failed\");\n"
3346 " exit(EXIT_FAILURE);\n"
3351 #: build/C/man3/mallopt.3:559
3355 " printf(\"main(): returned from first free() call\\en\");\n"
3359 #: build/C/man3/mallopt.3:562
3363 " printf(\"main(): returned from second free() call\\en\");\n"
3367 #: build/C/man3/mallopt.3:580
3369 "B<mmap>(2), B<sbrk>(2), B<mallinfo>(3), B<malloc>(3), B<malloc_hook>(3), "
3370 "B<malloc_info>(3), B<malloc_stats>(3), B<malloc_trim>(3), B<mcheck>(3), "
3371 "B<mtrace>(3), B<posix_memalign>(3)"
3375 #: build/C/man3/mcheck.3:25
3381 #: build/C/man3/mcheck.3:25 build/C/man3/mtrace.3:25
3387 #: build/C/man3/mcheck.3:28
3389 "mcheck, mcheck_check_all, mcheck_pedantic, mprobe - heap consistency "
3394 #: build/C/man3/mcheck.3:31
3396 msgid "B<#include E<lt>mcheck.hE<gt>>\n"
3400 #: build/C/man3/mcheck.3:33
3402 msgid "B<int mcheck(void (*>I<abortfunc>B<)(enum mcheck_status >I<mstatus>B<));>\n"
3406 #: build/C/man3/mcheck.3:35
3409 "B<int mcheck_pedantic(void (*>I<abortfunc>B<)(enum mcheck_status "
3410 ">I<mstatus>B<));>\n"
3414 #: build/C/man3/mcheck.3:37
3416 msgid "B<void mcheck_check_all(void);>\n"
3420 #: build/C/man3/mcheck.3:39
3422 msgid "B<enum mcheck_status mprobe(void *>I<ptr>B<);>\n"
3426 #: build/C/man3/mcheck.3:51
3428 "The B<mcheck>() function installs a set of debugging hooks for the "
3429 "B<malloc>(3) family of memory-allocation functions. These hooks cause "
3430 "certain consistency checks to be performed on the state of the heap. The "
3431 "checks can detect application errors such as freeing a block of memory more "
3432 "than once or corrupting the bookkeeping data structures that immediately "
3433 "precede a block of allocated memory."
3437 #: build/C/man3/mcheck.3:63
3439 "To be effective, the B<mcheck>() function must be called before the first "
3440 "call to B<malloc>(3) or a related function. In cases where this is "
3441 "difficult to ensure, linking the program with I<-mcheck> inserts an implicit "
3442 "call to B<mcheck>() (with a NULL argument) before the first call to a "
3443 "memory-allocation function."
3447 #: build/C/man3/mcheck.3:71
3449 "The B<mcheck_pedantic>() function is similar to B<mcheck>(), but performs "
3450 "checks on all allocated blocks whenever one of the memory-allocation "
3451 "functions is called. This can be very slow!"
3455 #: build/C/man3/mcheck.3:78
3457 "The B<mcheck_check_all>() function causes an immediate check on all "
3458 "allocated blocks. This call is only effective if B<mcheck>() is called "
3463 #: build/C/man3/mcheck.3:91
3465 "If the system detects an inconsistency in the heap, the caller-supplied "
3466 "function pointed to by I<abortfunc> is invoked with a single argument "
3467 "argument, I<mstatus>, that indicates what type of inconsistency was "
3468 "detected. If I<abortfunc> is NULL, a default function prints an error "
3469 "message on I<stderr> and calls B<abort>(3)."
3473 #: build/C/man3/mcheck.3:103
3475 "The B<mprobe>() function performs a consistency check on the block of "
3476 "allocated memory pointed to by I<ptr>. The B<mcheck>() function should be "
3477 "called beforehand (otherwise B<mprobe>() returns B<MCHECK_DISABLED>)."
3481 #: build/C/man3/mcheck.3:111
3483 "The following list describes the values returned by B<mprobe>() or passed "
3484 "as the I<mstatus> argument when I<abortfunc> is invoked:"
3488 #: build/C/man3/mcheck.3:111
3490 msgid "B<MCHECK_DISABLED> (B<mprobe>() only)"
3494 #: build/C/man3/mcheck.3:116
3496 "B<mcheck>() was not called before the first memory allocation function was "
3497 "called. Consistency checking is not possible."
3501 #: build/C/man3/mcheck.3:116
3503 msgid "B<MCHECK_OK> (B<mprobe>() only)"
3507 #: build/C/man3/mcheck.3:119
3508 msgid "No inconsistency detected."
3512 #: build/C/man3/mcheck.3:119
3514 msgid "B<MCHECK_HEAD>"
3518 #: build/C/man3/mcheck.3:122
3519 msgid "Memory preceding an allocated block was clobbered."
3523 #: build/C/man3/mcheck.3:122
3525 msgid "B<MCHECK_TAIL>"
3529 #: build/C/man3/mcheck.3:125
3530 msgid "Memory following an allocated block was clobbered."
3534 #: build/C/man3/mcheck.3:125
3536 msgid "B<MCHECK_FREE>"
3540 #: build/C/man3/mcheck.3:129
3541 msgid "A block of memory was freed twice."
3545 #: build/C/man3/mcheck.3:134
3546 msgid "B<mcheck>() and B<mcheck_pedantic>() return 0 on success, or -1 on error."
3550 #: build/C/man3/mcheck.3:145
3552 "The B<mcheck_pedantic>() and B<mcheck_check_all>() functions are available "
3553 "since glibc 2.2. The B<mcheck>() and B<mprobe>() functions are present "
3554 "since at least glibc 2.0"
3557 #. But is MALLOC_CHECK_ slower?
3559 #: build/C/man3/mcheck.3:159
3561 "Linking a program with I<-lmcheck> and using the B<MALLOC_CHECK_> "
3562 "environment variable (described in B<mallopt>(3)) cause the same kinds of "
3563 "errors to be detected. But, using B<MALLOC_CHECK_> does not require the "
3564 "application to be relinked."
3568 #: build/C/man3/mcheck.3:165
3570 "The program below calls B<mcheck>() with a NULL argument and then frees the "
3571 "same block of memory twice. The following shell session demonstrates what "
3572 "happens when running the program:"
3576 #: build/C/man3/mcheck.3:170
3584 #: build/C/man3/mcheck.3:174
3587 "About to free a second time\n"
3588 "block freed twice\n"
3589 "Aborted (core dumped)\n"
3593 #: build/C/man3/mcheck.3:182
3596 "#include E<lt>stdlib.hE<gt>\n"
3597 "#include E<lt>stdio.hE<gt>\n"
3598 "#include E<lt>mcheck.hE<gt>\n"
3602 #: build/C/man3/mcheck.3:190
3605 " if (mcheck(NULL) != 0) {\n"
3606 " fprintf(stderr, \"mcheck() failed\\en\");\n"
3610 #: build/C/man3/mcheck.3:193
3613 " exit(EXIT_FAILURE);\n"
3618 #: build/C/man3/mcheck.3:195
3620 msgid " p = malloc(1000);\n"
3624 #: build/C/man3/mcheck.3:200
3627 " fprintf(stderr, \"About to free\\en\");\n"
3629 " fprintf(stderr, \"\\enAbout to free a second time\\en\");\n"
3634 #: build/C/man3/mcheck.3:208
3635 msgid "B<malloc>(3), B<mallopt>(3), B<mtrace>(3)"
3639 #: build/C/man2/mlock.2:26
3645 #: build/C/man2/mlock.2:26
3651 #: build/C/man2/mlock.2:29
3652 msgid "mlock, munlock, mlockall, munlockall - lock and unlock memory"
3656 #: build/C/man2/mlock.2:32 build/C/man2/mmap.2:46 build/C/man2/mmap2.2:35 build/C/man2/mprotect.2:44 build/C/man2/mremap.2:38
3658 msgid "B<#include E<lt>sys/mman.hE<gt>>\n"
3662 #: build/C/man2/mlock.2:35
3665 "B<int mlock(const void *>I<addr>B<, size_t >I<len>B<);>\n"
3666 "B<int munlock(const void *>I<addr>B<, size_t >I<len>B<);>\n"
3670 #: build/C/man2/mlock.2:38
3673 "B<int mlockall(int >I<flags>B<);>\n"
3674 "B<int munlockall(void);>\n"
3678 #: build/C/man2/mlock.2:54
3680 "B<mlock>() and B<mlockall>() respectively lock part or all of the calling "
3681 "process's virtual address space into RAM, preventing that memory from being "
3682 "paged to the swap area. B<munlock>() and B<munlockall>() perform the "
3683 "converse operation, respectively unlocking part or all of the calling "
3684 "process's virtual address space, so that pages in the specified virtual "
3685 "address range may once more to be swapped out if required by the kernel "
3686 "memory manager. Memory locking and unlocking are performed in units of "
3691 #: build/C/man2/mlock.2:54
3693 msgid "mlock() and munlock()"
3697 #: build/C/man2/mlock.2:64
3699 "B<mlock>() locks pages in the address range starting at I<addr> and "
3700 "continuing for I<len> bytes. All pages that contain a part of the specified "
3701 "address range are guaranteed to be resident in RAM when the call returns "
3702 "successfully; the pages are guaranteed to stay in RAM until later unlocked."
3706 #: build/C/man2/mlock.2:73
3708 "B<munlock>() unlocks pages in the address range starting at I<addr> and "
3709 "continuing for I<len> bytes. After this call, all pages that contain a part "
3710 "of the specified memory range can be moved to external swap space again by "
3715 #: build/C/man2/mlock.2:73
3717 msgid "mlockall() and munlockall()"
3721 #: build/C/man2/mlock.2:83
3723 "B<mlockall>() locks all pages mapped into the address space of the calling "
3724 "process. This includes the pages of the code, data and stack segment, as "
3725 "well as shared libraries, user space kernel data, shared memory, and "
3726 "memory-mapped files. All mapped pages are guaranteed to be resident in RAM "
3727 "when the call returns successfully; the pages are guaranteed to stay in RAM "
3728 "until later unlocked."
3732 #: build/C/man2/mlock.2:88
3734 "The I<flags> argument is constructed as the bitwise OR of one or more of the "
3735 "following constants:"
3739 #: build/C/man2/mlock.2:88
3741 msgid "B<MCL_CURRENT>"
3745 #: build/C/man2/mlock.2:92
3747 "Lock all pages which are currently mapped into the address space of the "
3752 #: build/C/man2/mlock.2:92
3754 msgid "B<MCL_FUTURE>"
3758 #: build/C/man2/mlock.2:99
3760 "Lock all pages which will become mapped into the address space of the "
3761 "process in the future. These could be for instance new pages required by a "
3762 "growing heap and stack as well as new memory mapped files or shared memory "
3767 #: build/C/man2/mlock.2:112
3769 "If B<MCL_FUTURE> has been specified, then a later system call (e.g., "
3770 "B<mmap>(2), B<sbrk>(2), B<malloc>(3)), may fail if it would cause the number "
3771 "of locked bytes to exceed the permitted maximum (see below). In the same "
3772 "circumstances, stack growth may likewise fail: the kernel will deny stack "
3773 "expansion and deliver a B<SIGSEGV> signal to the process."
3777 #: build/C/man2/mlock.2:116
3779 "B<munlockall>() unlocks all pages mapped into the address space of the "
3784 #: build/C/man2/mlock.2:122
3786 "On success these system calls return 0. On error, -1 is returned, I<errno> "
3787 "is set appropriately, and no changes are made to any locks in the address "
3788 "space of the process."
3792 #: build/C/man2/mlock.2:131
3794 "(Linux 2.6.9 and later) the caller had a nonzero B<RLIMIT_MEMLOCK> soft "
3795 "resource limit, but tried to lock more memory than the limit permitted. "
3796 "This limit is not enforced if the process is privileged (B<CAP_IPC_LOCK>)."
3799 #. In the case of mlock(), this check is somewhat buggy: it doesn't
3800 #. take into account whether the to-be-locked range overlaps with
3801 #. already locked pages. Thus, suppose we allocate
3802 #. (num_physpages / 4 + 1) of memory, and lock those pages once using
3803 #. mlock(), and then lock the *same* page range a second time.
3804 #. In the case, the second mlock() call will fail, since the check
3805 #. calculates that the process is trying to lock (num_physpages / 2 + 2)
3806 #. pages, which of course is not true. (MTK, Nov 04, kernel 2.4.28)
3808 #: build/C/man2/mlock.2:143
3810 "(Linux 2.4 and earlier) the calling process tried to lock more than half of "
3814 #. SVr4 documents an additional EAGAIN error code.
3816 #: build/C/man2/mlock.2:149
3818 "The caller is not privileged, but needs privilege (B<CAP_IPC_LOCK>) to "
3819 "perform the requested operation."
3823 #: build/C/man2/mlock.2:154
3824 msgid "For B<mlock>() and B<munlock>():"
3828 #: build/C/man2/mlock.2:157
3829 msgid "Some or all of the specified address range could not be locked."
3833 #: build/C/man2/mlock.2:164
3835 "The result of the addition I<start>+I<len> was less than I<start> (e.g., the "
3836 "addition may have resulted in an overflow)."
3840 #: build/C/man2/mlock.2:169
3841 msgid "(Not on Linux) I<addr> was not a multiple of the page size."
3845 #: build/C/man2/mlock.2:173
3847 "Some of the specified address range does not correspond to mapped pages in "
3848 "the address space of the process."
3852 #: build/C/man2/mlock.2:176
3853 msgid "For B<mlockall>():"
3857 #: build/C/man2/mlock.2:179
3858 msgid "Unknown I<flags> were specified."
3862 #: build/C/man2/mlock.2:182
3863 msgid "For B<munlockall>():"
3867 #: build/C/man2/mlock.2:186
3868 msgid "(Linux 2.6.8 and earlier) The caller was not privileged (B<CAP_IPC_LOCK>)."
3872 #: build/C/man2/mlock.2:188
3873 msgid "POSIX.1-2001, SVr4."
3877 #: build/C/man2/mlock.2:188 build/C/man2/mmap.2:501 build/C/man2/msync.2:105
3879 msgid "AVAILABILITY"
3883 #: build/C/man2/mlock.2:200
3885 "On POSIX systems on which B<mlock>() and B<munlock>() are available, "
3886 "B<_POSIX_MEMLOCK_RANGE> is defined in I<E<lt>unistd.hE<gt>> and the number "
3887 "of bytes in a page can be determined from the constant B<PAGESIZE> (if "
3888 "defined) in I<E<lt>limits.hE<gt>> or by calling I<sysconf(_SC_PAGESIZE)>."
3891 #. POSIX.1-2001: It shall be defined to -1 or 0 or 200112L.
3892 #. -1: unavailable, 0: ask using sysconf().
3893 #. glibc defines it to 1.
3895 #: build/C/man2/mlock.2:213
3897 "On POSIX systems on which B<mlockall>() and B<munlockall>() are available, "
3898 "B<_POSIX_MEMLOCK> is defined in I<E<lt>unistd.hE<gt>> to a value greater "
3899 "than 0. (See also B<sysconf>(3).)"
3903 #: build/C/man2/mlock.2:231
3905 "Memory locking has two main applications: real-time algorithms and "
3906 "high-security data processing. Real-time applications require deterministic "
3907 "timing, and, like scheduling, paging is one major cause of unexpected "
3908 "program execution delays. Real-time applications will usually also switch "
3909 "to a real-time scheduler with B<sched_setscheduler>(2). Cryptographic "
3910 "security software often handles critical bytes like passwords or secret keys "
3911 "as data structures. As a result of paging, these secrets could be "
3912 "transferred onto a persistent swap store medium, where they might be "
3913 "accessible to the enemy long after the security software has erased the "
3914 "secrets in RAM and terminated. (But be aware that the suspend mode on "
3915 "laptops and some desktop computers will save a copy of the system's RAM to "
3916 "disk, regardless of memory locks.)"
3920 #: build/C/man2/mlock.2:244
3922 "Real-time processes that are using B<mlockall>() to prevent delays on page "
3923 "faults should reserve enough locked stack pages before entering the "
3924 "time-critical section, so that no page fault can be caused by function "
3925 "calls. This can be achieved by calling a function that allocates a "
3926 "sufficiently large automatic variable (an array) and writes to the memory "
3927 "occupied by this array in order to touch these stack pages. This way, "
3928 "enough pages will be mapped for the stack and can be locked into RAM. The "
3929 "dummy writes ensure that not even copy-on-write page faults can occur in the "
3934 #: build/C/man2/mlock.2:250
3936 "Memory locks are not inherited by a child created via B<fork>(2) and are "
3937 "automatically removed (unlocked) during an B<execve>(2) or when the process "
3942 #: build/C/man2/mlock.2:254
3944 "The memory lock on an address range is automatically removed if the address "
3945 "range is unmapped via B<munmap>(2)."
3949 #: build/C/man2/mlock.2:267
3951 "Memory locks do not stack, that is, pages which have been locked several "
3952 "times by calls to B<mlock>() or B<mlockall>() will be unlocked by a single "
3953 "call to B<munlock>() for the corresponding range or by B<munlockall>(). "
3954 "Pages which are mapped to several locations or by several processes stay "
3955 "locked into RAM as long as they are locked at least at one location or by at "
3956 "least one process."
3960 #: build/C/man2/mlock.2:278
3962 "Under Linux, B<mlock>() and B<munlock>() automatically round I<addr> down "
3963 "to the nearest page boundary. However, POSIX.1-2001 allows an "
3964 "implementation to require that I<addr> is page aligned, so portable "
3965 "applications should ensure this."
3969 #: build/C/man2/mlock.2:291
3971 "The I<VmLck> field of the Linux-specific I</proc/PID/status> file shows how "
3972 "many kilobytes of memory the process with ID I<PID> has locked using "
3973 "B<mlock>(), B<mlockall>(), and B<mmap>(2) B<MAP_LOCKED>."
3977 #: build/C/man2/mlock.2:291
3979 msgid "Limits and permissions"
3983 #: build/C/man2/mlock.2:298
3985 "In Linux 2.6.8 and earlier, a process must be privileged (B<CAP_IPC_LOCK>) "
3986 "in order to lock memory and the B<RLIMIT_MEMLOCK> soft resource limit "
3987 "defines a limit on how much memory the process may lock."
3991 #: build/C/man2/mlock.2:304
3993 "Since Linux 2.6.9, no limits are placed on the amount of memory that a "
3994 "privileged process can lock and the B<RLIMIT_MEMLOCK> soft resource limit "
3995 "instead defines a limit on how much memory an unprivileged process may lock."
3999 #: build/C/man2/mlock.2:312
4001 "In the 2.4 series Linux kernels up to and including 2.4.17, a bug caused the "
4002 "B<mlockall>() B<MCL_FUTURE> flag to be inherited across a B<fork>(2). This "
4003 "was rectified in kernel 2.4.18."
4006 #. See the following LKML thread:
4007 #. http://marc.theaimsgroup.com/?l=linux-kernel&m=113801392825023&w=2
4008 #. "Rationale for RLIMIT_MEMLOCK"
4011 #: build/C/man2/mlock.2:329
4013 "Since kernel 2.6.9, if a privileged process calls I<mlockall(MCL_FUTURE)> "
4014 "and later drops privileges (loses the B<CAP_IPC_LOCK> capability by, for "
4015 "example, setting its effective UID to a nonzero value), then subsequent "
4016 "memory allocations (e.g., B<mmap>(2), B<brk>(2)) will fail if the "
4017 "B<RLIMIT_MEMLOCK> resource limit is encountered."
4021 #: build/C/man2/mlock.2:336
4023 "B<mmap>(2), B<setrlimit>(2), B<shmctl>(2), B<sysconf>(3), B<proc>(5), "
4024 "B<capabilities>(7)"
4028 #: build/C/man2/mmap.2:40
4034 #: build/C/man2/mmap.2:40
4040 #: build/C/man2/mmap.2:43
4041 msgid "mmap, munmap - map or unmap files or devices into memory"
4045 #: build/C/man2/mmap.2:51
4048 "B<void *mmap(void *>I<addr>B<, size_t >I<length>B<, int >I<prot>B<, int "
4050 "B< int >I<fd>B<, off_t >I<offset>B<);>\n"
4051 "B<int munmap(void *>I<addr>B<, size_t >I<length>B<);>\n"
4055 #: build/C/man2/mmap.2:54
4056 msgid "See NOTES for information on feature test macro requirements."
4060 #: build/C/man2/mmap.2:63
4062 "B<mmap>() creates a new mapping in the virtual address space of the calling "
4063 "process. The starting address for the new mapping is specified in I<addr>. "
4064 "The I<length> argument specifies the length of the mapping."
4067 #. Before Linux 2.6.24, the address was rounded up to the next page
4068 #. boundary; since 2.6.24, it is rounded down!
4070 #: build/C/man2/mmap.2:77
4072 "If I<addr> is NULL, then the kernel chooses the address at which to create "
4073 "the mapping; this is the most portable method of creating a new mapping. If "
4074 "I<addr> is not NULL, then the kernel takes it as a hint about where to place "
4075 "the mapping; on Linux, the mapping will be created at a nearby page "
4076 "boundary. The address of the new mapping is returned as the result of the "
4081 #: build/C/man2/mmap.2:89
4083 "The contents of a file mapping (as opposed to an anonymous mapping; see "
4084 "B<MAP_ANONYMOUS> below), are initialized using I<length> bytes starting at "
4085 "offset I<offset> in the file (or other object) referred to by the file "
4086 "descriptor I<fd>. I<offset> must be a multiple of the page size as returned "
4087 "by I<sysconf(_SC_PAGE_SIZE)>."
4091 #: build/C/man2/mmap.2:97
4093 "The I<prot> argument describes the desired memory protection of the mapping "
4094 "(and must not conflict with the open mode of the file). It is either "
4095 "B<PROT_NONE> or the bitwise OR of one or more of the following flags:"
4099 #: build/C/man2/mmap.2:97 build/C/man2/mprotect.2:73
4101 msgid "B<PROT_EXEC>"
4105 #: build/C/man2/mmap.2:100
4106 msgid "Pages may be executed."
4110 #: build/C/man2/mmap.2:100 build/C/man2/mprotect.2:67
4112 msgid "B<PROT_READ>"
4116 #: build/C/man2/mmap.2:103
4117 msgid "Pages may be read."
4121 #: build/C/man2/mmap.2:103 build/C/man2/mprotect.2:70
4123 msgid "B<PROT_WRITE>"
4127 #: build/C/man2/mmap.2:106
4128 msgid "Pages may be written."
4132 #: build/C/man2/mmap.2:106 build/C/man2/mprotect.2:64
4134 msgid "B<PROT_NONE>"
4138 #: build/C/man2/mmap.2:109
4139 msgid "Pages may not be accessed."
4143 #: build/C/man2/mmap.2:118
4145 "The I<flags> argument determines whether updates to the mapping are visible "
4146 "to other processes mapping the same region, and whether updates are carried "
4147 "through to the underlying file. This behavior is determined by including "
4148 "exactly one of the following values in I<flags>:"
4152 #: build/C/man2/mmap.2:118
4154 msgid "B<MAP_SHARED>"
4158 #: build/C/man2/mmap.2:128
4160 "Share this mapping. Updates to the mapping are visible to other processes "
4161 "that map this file, and are carried through to the underlying file. The "
4162 "file may not actually be updated until B<msync>(2) or B<munmap>() is "
4167 #: build/C/man2/mmap.2:128
4169 msgid "B<MAP_PRIVATE>"
4173 #: build/C/man2/mmap.2:137
4175 "Create a private copy-on-write mapping. Updates to the mapping are not "
4176 "visible to other processes mapping the same file, and are not carried "
4177 "through to the underlying file. It is unspecified whether changes made to "
4178 "the file after the B<mmap>() call are visible in the mapped region."
4182 #: build/C/man2/mmap.2:139
4183 msgid "Both of these flags are described in POSIX.1-2001."
4187 #: build/C/man2/mmap.2:142
4188 msgid "In addition, zero or more of the following values can be ORed in I<flags>:"
4192 #: build/C/man2/mmap.2:142
4194 msgid "B<MAP_32BIT> (since Linux 2.4.20, 2.6)"
4197 #. See http://lwn.net/Articles/294642 "Tangled up in threads", 19 Aug 08
4199 #: build/C/man2/mmap.2:158
4201 "Put the mapping into the first 2 Gigabytes of the process address space. "
4202 "This flag is only supported on x86-64, for 64-bit programs. It was added to "
4203 "allow thread stacks to be allocated somewhere in the first 2GB of memory, so "
4204 "as to improve context-switch performance on some early 64-bit processors. "
4205 "Modern x86-64 processors no longer have this performance problem, so use of "
4206 "this flag is not required on those systems. The B<MAP_32BIT> flag is "
4207 "ignored when B<MAP_FIXED> is set."
4211 #: build/C/man2/mmap.2:158
4217 #: build/C/man2/mmap.2:163
4218 msgid "Synonym for B<MAP_ANONYMOUS>. Deprecated."
4222 #: build/C/man2/mmap.2:163
4224 msgid "B<MAP_ANONYMOUS>"
4228 #: build/C/man2/mmap.2:185
4230 "The mapping is not backed by any file; its contents are initialized to "
4231 "zero. The I<fd> and I<offset> arguments are ignored; however, some "
4232 "implementations require I<fd> to be -1 if B<MAP_ANONYMOUS> (or B<MAP_ANON>) "
4233 "is specified, and portable applications should ensure this. The use of "
4234 "B<MAP_ANONYMOUS> in conjunction with B<MAP_SHARED> is only supported on "
4235 "Linux since kernel 2.4."
4239 #: build/C/man2/mmap.2:185
4241 msgid "B<MAP_DENYWRITE>"
4244 #. Introduced in 1.1.36, removed in 1.3.24.
4246 #: build/C/man2/mmap.2:193
4248 "This flag is ignored. (Long ago, it signaled that attempts to write to the "
4249 "underlying file should fail with B<ETXTBUSY>. But this was a source of "
4250 "denial-of-service attacks.)"
4254 #: build/C/man2/mmap.2:193
4256 msgid "B<MAP_EXECUTABLE>"
4259 #. Introduced in 1.1.38, removed in 1.3.24. Flag tested in proc_follow_link.
4260 #. (Long ago, it signaled that the underlying file is an executable.
4261 #. However, that information was not really used anywhere.)
4262 #. Linus talked about DOS related to MAP_EXECUTABLE, but he was thinking of
4265 #: build/C/man2/mmap.2:201
4266 msgid "This flag is ignored."
4270 #: build/C/man2/mmap.2:201
4275 #. On some systems, this was required as the opposite of
4276 #. MAP_ANONYMOUS -- mtk, 1 May 2007
4278 #: build/C/man2/mmap.2:207
4279 msgid "Compatibility flag. Ignored."
4283 #: build/C/man2/mmap.2:207
4285 msgid "B<MAP_FIXED>"
4289 #: build/C/man2/mmap.2:225
4291 "Don't interpret I<addr> as a hint: place the mapping at exactly that "
4292 "address. I<addr> must be a multiple of the page size. If the memory region "
4293 "specified by I<addr> and I<len> overlaps pages of any existing mapping(s), "
4294 "then the overlapped part of the existing mapping(s) will be discarded. If "
4295 "the specified address cannot be used, B<mmap>() will fail. Because "
4296 "requiring a fixed address for a mapping is less portable, the use of this "
4297 "option is discouraged."
4301 #: build/C/man2/mmap.2:225
4303 msgid "B<MAP_GROWSDOWN>"
4307 #: build/C/man2/mmap.2:230
4309 "Used for stacks. Indicates to the kernel virtual memory system that the "
4310 "mapping should extend downward in memory."
4314 #: build/C/man2/mmap.2:230
4316 msgid "B<MAP_HUGETLB> (since Linux 2.6.32)"
4320 #: build/C/man2/mmap.2:236
4322 "Allocate the mapping using \"huge pages.\" See the Linux kernel source file "
4323 "I<Documentation/vm/hugetlbpage.txt> for further information."
4327 #: build/C/man2/mmap.2:236
4329 msgid "B<MAP_LOCKED> (since Linux 2.5.37)"
4332 #. If set, the mapped pages will not be swapped out.
4334 #: build/C/man2/mmap.2:242
4336 "Lock the pages of the mapped region into memory in the manner of "
4337 "B<mlock>(2). This flag is ignored in older kernels."
4341 #: build/C/man2/mmap.2:242
4343 msgid "B<MAP_NONBLOCK> (since Linux 2.5.46)"
4347 #: build/C/man2/mmap.2:257
4349 "Only meaningful in conjunction with B<MAP_POPULATE>. Don't perform "
4350 "read-ahead: only create page tables entries for pages that are already "
4351 "present in RAM. Since Linux 2.6.23, this flag causes B<MAP_POPULATE> to do "
4352 "nothing. One day the combination of B<MAP_POPULATE> and B<MAP_NONBLOCK> may "
4357 #: build/C/man2/mmap.2:257
4359 msgid "B<MAP_NORESERVE>"
4363 #: build/C/man2/mmap.2:272
4365 "Do not reserve swap space for this mapping. When swap space is reserved, "
4366 "one has the guarantee that it is possible to modify the mapping. When swap "
4367 "space is not reserved one might get B<SIGSEGV> upon a write if no physical "
4368 "memory is available. See also the discussion of the file "
4369 "I</proc/sys/vm/overcommit_memory> in B<proc>(5). In kernels before 2.6, "
4370 "this flag only had effect for private writable mappings."
4374 #: build/C/man2/mmap.2:272
4376 msgid "B<MAP_POPULATE> (since Linux 2.5.46)"
4380 #: build/C/man2/mmap.2:279
4382 "Populate (prefault) page tables for a mapping. For a file mapping, this "
4383 "causes read-ahead on the file. Later accesses to the mapping will not be "
4384 "blocked by page faults. B<MAP_POPULATE> is only supported for private "
4385 "mappings since Linux 2.6.23."
4389 #: build/C/man2/mmap.2:279
4391 msgid "B<MAP_STACK> (since Linux 2.6.27)"
4394 #. See http://lwn.net/Articles/294642 "Tangled up in threads", 19 Aug 08
4395 #. commit cd98a04a59e2f94fa64d5bf1e26498d27427d5e7
4396 #. http://thread.gmane.org/gmane.linux.kernel/720412
4397 #. "pthread_create() slow for many threads; also time to revisit 64b
4398 #. context switch optimization?"
4400 #: build/C/man2/mmap.2:292
4402 "Allocate the mapping at an address suitable for a process or thread stack. "
4403 "This flag is currently a no-op, but is used in the glibc threading "
4404 "implementation so that if some architectures require special treatment for "
4405 "stack allocations, support can later be transparently implemented for glibc."
4409 #: build/C/man2/mmap.2:292
4411 msgid "B<MAP_UNINITIALIZED> (since Linux 2.6.33)"
4415 #: build/C/man2/mmap.2:302
4417 "Don't clear anonymous pages. This flag is intended to improve performance "
4418 "on embedded devices. This flag is only honored if the kernel was configured "
4419 "with the B<CONFIG_MMAP_ALLOW_UNINITIALIZED> option. Because of the security "
4420 "implications, that option is normally enabled only on embedded devices "
4421 "(i.e., devices where one has complete control of the contents of user "
4426 #: build/C/man2/mmap.2:310
4428 "Of the above flags, only B<MAP_FIXED> is specified in POSIX.1-2001. "
4429 "However, most systems also support B<MAP_ANONYMOUS> (or its synonym "
4434 #: build/C/man2/mmap.2:317
4436 "Some systems document the additional flags B<MAP_AUTOGROW>, "
4437 "B<MAP_AUTORESRV>, B<MAP_COPY>, and B<MAP_LOCAL>."
4441 #: build/C/man2/mmap.2:323
4443 "Memory mapped by B<mmap>() is preserved across B<fork>(2), with the same "
4448 #: build/C/man2/mmap.2:331
4450 "A file is mapped in multiples of the page size. For a file that is not a "
4451 "multiple of the page size, the remaining memory is zeroed when mapped, and "
4452 "writes to that region are not written out to the file. The effect of "
4453 "changing the size of the underlying file of a mapping on the pages that "
4454 "correspond to added or removed regions of the file is unspecified."
4458 #: build/C/man2/mmap.2:331
4464 #: build/C/man2/mmap.2:341
4466 "The B<munmap>() system call deletes the mappings for the specified address "
4467 "range, and causes further references to addresses within the range to "
4468 "generate invalid memory references. The region is also automatically "
4469 "unmapped when the process is terminated. On the other hand, closing the "
4470 "file descriptor does not unmap the region."
4474 #: build/C/man2/mmap.2:351
4476 "The address I<addr> must be a multiple of the page size. All pages "
4477 "containing a part of the indicated range are unmapped, and subsequent "
4478 "references to these pages will generate B<SIGSEGV>. It is not an error if "
4479 "the indicated range does not contain any mapped pages."
4483 #: build/C/man2/mmap.2:351
4485 msgid "Timestamps changes for file-backed mappings"
4489 #: build/C/man2/mmap.2:358
4491 "For file-backed mappings, the I<st_atime> field for the mapped file may be "
4492 "updated at any time between the B<mmap>() and the corresponding unmapping; "
4493 "the first reference to a mapped page will update the field if it has not "
4498 #: build/C/man2/mmap.2:375
4500 "The I<st_ctime> and I<st_mtime> field for a file mapped with B<PROT_WRITE> "
4501 "and B<MAP_SHARED> will be updated after a write to the mapped region, and "
4502 "before a subsequent B<msync>(2) with the B<MS_SYNC> or B<MS_ASYNC> flag, if "
4507 #: build/C/man2/mmap.2:392
4509 "On success, B<mmap>() returns a pointer to the mapped area. On error, the "
4510 "value B<MAP_FAILED> (that is, I<(void\\ *)\\ -1>) is returned, and I<errno> "
4511 "is set appropriately. On success, B<munmap>() returns 0, on failure -1, "
4512 "and I<errno> is set (probably to B<EINVAL>)."
4516 #: build/C/man2/mmap.2:393 build/C/man2/mprotect.2:86 build/C/man3/shm_open.3:185 build/C/man3/shm_open.3:190 build/C/man2/shmctl.2:297 build/C/man2/shmget.2:188 build/C/man2/shmop.2:191
4522 #: build/C/man2/mmap.2:413
4524 "A file descriptor refers to a non-regular file. Or B<MAP_PRIVATE> was "
4525 "requested, but I<fd> is not open for reading. Or B<MAP_SHARED> was "
4526 "requested and B<PROT_WRITE> is set, but I<fd> is not open in read/write "
4527 "(B<O_RDWR>) mode. Or B<PROT_WRITE> is set, but the file is append-only."
4531 #: build/C/man2/mmap.2:417
4533 "The file has been locked, or too much memory has been locked (see "
4538 #: build/C/man2/mmap.2:423
4539 msgid "I<fd> is not a valid file descriptor (and B<MAP_ANONYMOUS> was not set)."
4543 #: build/C/man2/mmap.2:431
4545 "We don't like I<addr>, I<length>, or I<offset> (e.g., they are too large, or "
4546 "not aligned on a page boundary)."
4550 #: build/C/man2/mmap.2:436
4551 msgid "(since Linux 2.6.12) I<length> was 0."
4555 #: build/C/man2/mmap.2:444
4557 "I<flags> contained neither B<MAP_PRIVATE> or B<MAP_SHARED>, or contained "
4558 "both of these values."
4562 #: build/C/man2/mmap.2:444 build/C/man3/shm_open.3:227 build/C/man2/shmget.2:204
4567 #. [2.6.7] shmem_zero_setup()-->shmem_file_setup()-->get_empty_filp()
4569 #: build/C/man2/mmap.2:452 build/C/man2/shmget.2:208
4570 msgid "The system limit on the total number of open files has been reached."
4574 #: build/C/man2/mmap.2:456
4576 "The underlying file system of the specified file does not support memory "
4581 #: build/C/man2/mmap.2:460
4583 "No memory is available, or the process's maximum number of mappings would "
4584 "have been exceeded."
4587 #. (Since 2.4.25 / 2.6.0.)
4589 #: build/C/man2/mmap.2:469
4591 "The I<prot> argument asks for B<PROT_EXEC> but the mapped area belongs to a "
4592 "file on a file system that was mounted no-exec."
4596 #: build/C/man2/mmap.2:469
4602 #: build/C/man2/mmap.2:475
4604 "B<MAP_DENYWRITE> was set but the object specified by I<fd> is open for "
4609 #: build/C/man2/mmap.2:475 build/C/man2/shmctl.2:339
4611 msgid "B<EOVERFLOW>"
4615 #: build/C/man2/mmap.2:487
4617 "On 32-bit architecture together with the large file extension (i.e., using "
4618 "64-bit I<off_t>): the number of pages used for I<length> plus number of "
4619 "pages used for I<offset> would overflow I<unsigned long> (32 bits)."
4623 #: build/C/man2/mmap.2:489
4624 msgid "Use of a mapped region can result in these signals:"
4628 #: build/C/man2/mmap.2:489
4634 #: build/C/man2/mmap.2:492
4635 msgid "Attempted write into a region mapped as read-only."
4639 #: build/C/man2/mmap.2:492
4645 #: build/C/man2/mmap.2:497
4647 "Attempted access to a portion of the buffer that does not correspond to the "
4648 "file (for example, beyond the end of the file, including the case where "
4649 "another process has truncated the file)."
4652 #. SVr4 documents additional error codes ENXIO and ENODEV.
4653 #. SUSv2 documents additional error codes EMFILE and EOVERFLOW.
4655 #: build/C/man2/mmap.2:501
4656 msgid "SVr4, 4.4BSD, POSIX.1-2001."
4659 #. POSIX.1-2001: It shall be defined to -1 or 0 or 200112L.
4660 #. -1: unavailable, 0: ask using sysconf().
4661 #. glibc defines it to 1.
4663 #: build/C/man2/mmap.2:515
4665 "On POSIX systems on which B<mmap>(), B<msync>(2) and B<munmap>() are "
4666 "available, B<_POSIX_MAPPED_FILES> is defined in I<E<lt>unistd.hE<gt>> to a "
4667 "value greater than 0. (See also B<sysconf>(3).)"
4670 #. Since around glibc 2.1/2.2, depending on the platform.
4672 #: build/C/man2/mmap.2:530
4674 "This page describes the interface provided by the glibc B<mmap>() wrapper "
4675 "function. Originally, this function invoked a system call of the same "
4676 "name. Since kernel 2.4, that system call has been superseded by "
4677 "B<mmap2>(2), and nowadays the glibc B<mmap>() wrapper function invokes "
4678 "B<mmap2>(2) with a suitably adjusted value for I<offset>."
4682 #: build/C/man2/mmap.2:543
4684 "On some hardware architectures (e.g., i386), B<PROT_WRITE> implies "
4685 "B<PROT_READ>. It is architecture dependent whether B<PROT_READ> implies "
4686 "B<PROT_EXEC> or not. Portable programs should always set B<PROT_EXEC> if "
4687 "they intend to execute code in the new mapping."
4691 #: build/C/man2/mmap.2:558
4693 "The portable way to create a mapping is to specify I<addr> as 0 (NULL), and "
4694 "omit B<MAP_FIXED> from I<flags>. In this case, the system chooses the "
4695 "address for the mapping; the address is chosen so as not to conflict with "
4696 "any existing mapping, and will not be 0. If the B<MAP_FIXED> flag is "
4697 "specified, and I<addr> is 0 (NULL), then the mapped address will be 0 "
4702 #: build/C/man2/mmap.2:587
4704 "Certain I<flags> constants are defined only if either B<_BSD_SOURCE> or "
4705 "B<_SVID_SOURCE> is defined. (Requiring B<_GNU_SOURCE> also suffices, and "
4706 "requiring that macro specifically would have been more logical, since these "
4707 "flags are all Linux specific.) The relevant flags are: B<MAP_32BIT>, "
4708 "B<MAP_ANONYMOUS> (and the synonym B<MAP_ANON>), B<MAP_DENYWRITE>, "
4709 "B<MAP_EXECUTABLE>, B<MAP_FILE>, B<MAP_GROWSDOWN>, B<MAP_HUGETLB>, "
4710 "B<MAP_LOCKED>, B<MAP_NONBLOCK>, B<MAP_NORESERVE>, B<MAP_POPULATE>, and "
4715 #: build/C/man2/mmap.2:592
4717 "On Linux there are no guarantees like those suggested above under "
4718 "B<MAP_NORESERVE>. By default, any process can be killed at any moment when "
4719 "the system runs out of memory."
4723 #: build/C/man2/mmap.2:599
4725 "In kernels before 2.6.7, the B<MAP_POPULATE> flag only has effect if I<prot> "
4726 "is specified as B<PROT_NONE>."
4730 #: build/C/man2/mmap.2:614
4732 "SUSv3 specifies that B<mmap>() should fail if I<length> is 0. However, in "
4733 "kernels before 2.6.12, B<mmap>() succeeded in this case: no mapping was "
4734 "created and the call returned I<addr>. Since kernel 2.6.12, B<mmap>() "
4735 "fails with the error B<EINVAL> for this case."
4739 #: build/C/man2/mmap.2:630
4741 "POSIX specifies that the system shall always zero fill any partial page at "
4742 "the end of the object and that system will never write any modification of "
4743 "the object beyond its end. On Linux, when you write data to such partial "
4744 "page after the end of the object, the data stays in the page cache even "
4745 "after the file is closed and unmapped and even though the data is never "
4746 "written to the file itself, subsequent mappings may see the modified "
4747 "content. In some cases, this could be fixed by calling B<msync>(2) before "
4748 "the unmap takes place; however, this doesn't work on tmpfs (for example, "
4749 "when using POSIX shared memory interface documented in B<shm_overview>(7))."
4753 #: build/C/man2/mmap.2:642
4755 "The following program prints part of the file specified in its first "
4756 "command-line argument to standard output. The range of bytes to be printed "
4757 "is specified via offset and length values in the second and third "
4758 "command-line arguments. The program creates a memory mapping of the "
4759 "required pages of the file and then uses B<write>(2) to output the desired "
4764 #: build/C/man2/mmap.2:650
4767 "#include E<lt>sys/mman.hE<gt>\n"
4768 "#include E<lt>sys/stat.hE<gt>\n"
4769 "#include E<lt>fcntl.hE<gt>\n"
4770 "#include E<lt>stdio.hE<gt>\n"
4771 "#include E<lt>stdlib.hE<gt>\n"
4772 "#include E<lt>unistd.hE<gt>\n"
4776 #: build/C/man2/mmap.2:653 build/C/man2/mprotect.2:178
4779 "#define handle_error(msg) \\e\n"
4780 " do { perror(msg); exit(EXIT_FAILURE); } while (0)\n"
4784 #: build/C/man2/mmap.2:663
4788 "main(int argc, char *argv[])\n"
4792 " struct stat sb;\n"
4793 " off_t offset, pa_offset;\n"
4799 #: build/C/man2/mmap.2:668
4802 " if (argc E<lt> 3 || argc E<gt> 4) {\n"
4803 " fprintf(stderr, \"%s file offset [length]\\en\", argv[0]);\n"
4804 " exit(EXIT_FAILURE);\n"
4809 #: build/C/man2/mmap.2:672
4812 " fd = open(argv[1], O_RDONLY);\n"
4814 " handle_error(\"open\");\n"
4818 #: build/C/man2/mmap.2:675
4821 " if (fstat(fd, &sb) == -1) /* To obtain file size */\n"
4822 " handle_error(\"fstat\");\n"
4826 #: build/C/man2/mmap.2:679
4829 " offset = atoi(argv[2]);\n"
4830 " pa_offset = offset & ~(sysconf(_SC_PAGE_SIZE) - 1);\n"
4831 " /* offset for mmap() must be page aligned */\n"
4835 #: build/C/man2/mmap.2:684
4838 " if (offset E<gt>= sb.st_size) {\n"
4839 " fprintf(stderr, \"offset is past end of file\\en\");\n"
4840 " exit(EXIT_FAILURE);\n"
4845 #: build/C/man2/mmap.2:690
4848 " if (argc == 4) {\n"
4849 " length = atoi(argv[3]);\n"
4850 " if (offset + length E<gt> sb.st_size)\n"
4851 " length = sb.st_size - offset;\n"
4852 " /* Can\\(aqt display bytes past end of file */\n"
4856 #: build/C/man2/mmap.2:694
4859 " } else { /* No length arg ==E<gt> display to end of file */\n"
4860 " length = sb.st_size - offset;\n"
4865 #: build/C/man2/mmap.2:699
4868 " addr = mmap(NULL, length + offset - pa_offset, PROT_READ,\n"
4869 " MAP_PRIVATE, fd, pa_offset);\n"
4870 " if (addr == MAP_FAILED)\n"
4871 " handle_error(\"mmap\");\n"
4875 #: build/C/man2/mmap.2:704
4878 " s = write(STDOUT_FILENO, addr + offset - pa_offset, length);\n"
4879 " if (s != length) {\n"
4881 " handle_error(\"write\");\n"
4885 #: build/C/man2/mmap.2:708
4888 " fprintf(stderr, \"partial write\");\n"
4889 " exit(EXIT_FAILURE);\n"
4894 #: build/C/man2/mmap.2:725
4896 "B<getpagesize>(2), B<mincore>(2), B<mlock>(2), B<mmap2>(2), B<mprotect>(2), "
4897 "B<mremap>(2), B<msync>(2), B<remap_file_pages>(2), B<setrlimit>(2), "
4898 "B<shmat>(2), B<shm_open>(3), B<shm_overview>(7)"
4902 #: build/C/man2/mmap.2:730 build/C/man2/msync.2:122
4903 msgid "B.O. Gallmeister, POSIX.4, O'Reilly, pp. 128-129 and 389-391."
4907 #: build/C/man2/mmap2.2:29
4913 #: build/C/man2/mmap2.2:29
4919 #: build/C/man2/mmap2.2:32
4920 msgid "mmap2 - map files or devices into memory"
4924 #: build/C/man2/mmap2.2:38
4927 "B<void *mmap2(void *>I<addr>B<, size_t >I<length>B<, int >I<prot>B<,>\n"
4928 "B< int >I<flags>B<, int >I<fd>B<, off_t >I<pgoffset>B<);>\n"
4932 #: build/C/man2/mmap2.2:43
4934 "This is probably not the system call you are interested; instead, see "
4935 "B<mmap>(2), which describes the glibc wrapper function that invokes this "
4940 #: build/C/man2/mmap2.2:54
4942 "The B<mmap2>() system call provides the same interface as B<mmap>(2), "
4943 "except that the final argument specifies the offset into the file in "
4944 "4096-byte units (instead of bytes, as is done by B<mmap>(2)). This enables "
4945 "applications that use a 32-bit I<off_t> to map large files (up to 2^44 "
4950 #: build/C/man2/mmap2.2:61
4952 "On success, B<mmap2>() returns a pointer to the mapped area. On error -1 "
4953 "is returned and I<errno> is set appropriately."
4957 #: build/C/man2/mmap2.2:65
4958 msgid "Problem with getting the data from user space."
4962 #: build/C/man2/mmap2.2:70
4964 "(Various platforms where the page size is not 4096 bytes.) I<offset * 4096> "
4965 "is not a multiple of the system page size."
4969 #: build/C/man2/mmap2.2:74
4970 msgid "B<mmap2>() can also return any of the errors described in B<mmap>(2)."
4974 #: build/C/man2/mmap2.2:77
4975 msgid "B<mmap2>() is available since Linux 2.3.31."
4979 #: build/C/man2/mmap2.2:79 build/C/man2/subpage_prot.2:98
4980 msgid "This system call is Linux-specific."
4984 #: build/C/man2/mmap2.2:85
4986 "Nowadays, the glibc B<mmap>() wrapper function invokes this system call "
4987 "rather than the B<mmap>(2) system call."
4990 #. ia64 can have page sizes ranging from 4kB to 64kB.
4991 #. On cris, it looks like the unit might also be the page size,
4992 #. which is 8192 bytes. -- mtk, June 2007
4994 #: build/C/man2/mmap2.2:92
4996 "On ia64, the unit for I<offset> is actually the system page size, rather "
5001 #: build/C/man2/mmap2.2:98
5002 msgid "B<getpagesize>(2), B<mmap>(2), B<mremap>(2), B<msync>(2), B<shm_open>(3)"
5006 #: build/C/man2/mprotect.2:38
5012 #: build/C/man2/mprotect.2:38
5018 #: build/C/man2/mprotect.2:41
5019 msgid "mprotect - set protection on a region of memory"
5023 #: build/C/man2/mprotect.2:46
5025 msgid "B<int mprotect(void *>I<addr>B<, size_t >I<len>B<, int >I<prot>B<);>\n"
5029 #: build/C/man2/mprotect.2:54
5031 "B<mprotect>() changes protection for the calling process's memory page(s) "
5032 "containing any part of the address range in the interval [I<addr>,\\ "
5033 "I<addr>+I<len>-1]. I<addr> must be aligned to a page boundary."
5037 #: build/C/man2/mprotect.2:59
5039 "If the calling process tries to access memory in a manner that violates the "
5040 "protection, then the kernel generates a B<SIGSEGV> signal for the process."
5044 #: build/C/man2/mprotect.2:64
5046 "I<prot> is either B<PROT_NONE> or a bitwise-or of the other values in the "
5051 #: build/C/man2/mprotect.2:67
5052 msgid "The memory cannot be accessed at all."
5056 #: build/C/man2/mprotect.2:70
5057 msgid "The memory can be read."
5061 #: build/C/man2/mprotect.2:73
5062 msgid "The memory can be modified."
5066 #. Document PROT_GROWSUP and PROT_GROWSDOWN
5068 #: build/C/man2/mprotect.2:78
5069 msgid "The memory can be executed."
5073 #: build/C/man2/mprotect.2:85
5075 "On success, B<mprotect>() returns zero. On error, -1 is returned, and "
5076 "I<errno> is set appropriately."
5080 #: build/C/man2/mprotect.2:95
5082 "The memory cannot be given the specified access. This can happen, for "
5083 "example, if you B<mmap>(2) a file to which you have read-only access, then "
5084 "ask B<mprotect>() to mark it B<PROT_WRITE>."
5087 #. Or: both PROT_GROWSUP and PROT_GROWSDOWN were specified in 'prot'.
5089 #: build/C/man2/mprotect.2:100
5090 msgid "I<addr> is not a valid pointer, or not a multiple of the system page size."
5094 #: build/C/man2/mprotect.2:103
5095 msgid "Internal kernel structures could not be allocated."
5099 #: build/C/man2/mprotect.2:113
5101 "Addresses in the range [I<addr>, I<addr>+I<len>-1] are invalid for the "
5102 "address space of the process, or specify one or more pages that are not "
5103 "mapped. (Before kernel 2.4.19, the error B<EFAULT> was incorrectly produced "
5107 #. SVr4 defines an additional error
5108 #. code EAGAIN. The SVr4 error conditions don't map neatly onto Linux's.
5110 #: build/C/man2/mprotect.2:122
5112 "SVr4, POSIX.1-2001. POSIX says that the behavior of B<mprotect>() is "
5113 "unspecified if it is applied to a region of memory that was not obtained via "
5118 #: build/C/man2/mprotect.2:129
5120 "On Linux it is always permissible to call B<mprotect>() on any address in a "
5121 "process's address space (except for the kernel vsyscall area). In "
5122 "particular it can be used to change existing code mappings to be writable."
5126 #: build/C/man2/mprotect.2:139
5128 "Whether B<PROT_EXEC> has any effect different from B<PROT_READ> is "
5129 "architecture- and kernel version-dependent. On some hardware architectures "
5130 "(e.g., i386), B<PROT_WRITE> implies B<PROT_READ>."
5134 #: build/C/man2/mprotect.2:148
5136 "POSIX.1-2001 says that an implementation may permit access other than that "
5137 "specified in I<prot>, but at a minimum can only allow write access if "
5138 "B<PROT_WRITE> has been set, and must not allow any access if B<PROT_NONE> "
5143 #: build/C/man2/mprotect.2:154
5145 "The program below allocates four pages of memory, makes the third of these "
5146 "pages read-only, and then executes a loop that walks upward through the "
5147 "allocated region modifying bytes."
5151 #: build/C/man2/mprotect.2:157
5152 msgid "An example of what we might see when running the program is the following:"
5156 #: build/C/man2/mprotect.2:163
5160 "Start of region: 0x804c000\n"
5161 "Got SIGSEGV at address: 0x804e000\n"
5165 #: build/C/man2/mprotect.2:175
5168 "#include E<lt>unistd.hE<gt>\n"
5169 "#include E<lt>signal.hE<gt>\n"
5170 "#include E<lt>stdio.hE<gt>\n"
5171 "#include E<lt>malloc.hE<gt>\n"
5172 "#include E<lt>stdlib.hE<gt>\n"
5173 "#include E<lt>errno.hE<gt>\n"
5174 "#include E<lt>sys/mman.hE<gt>\n"
5178 #: build/C/man2/mprotect.2:180
5180 msgid "char *buffer;\n"
5184 #: build/C/man2/mprotect.2:188
5188 "handler(int sig, siginfo_t *si, void *unused)\n"
5190 " printf(\"Got SIGSEGV at address: 0x%lx\\en\",\n"
5191 " (long) si-E<gt>si_addr);\n"
5192 " exit(EXIT_FAILURE);\n"
5197 #: build/C/man2/mprotect.2:195
5201 "main(int argc, char *argv[])\n"
5205 " struct sigaction sa;\n"
5209 #: build/C/man2/mprotect.2:201
5212 " sa.sa_flags = SA_SIGINFO;\n"
5213 " sigemptyset(&sa.sa_mask);\n"
5214 " sa.sa_sigaction = handler;\n"
5215 " if (sigaction(SIGSEGV, &sa, NULL) == -1)\n"
5216 " handle_error(\"sigaction\");\n"
5220 #: build/C/man2/mprotect.2:205
5223 " pagesize = sysconf(_SC_PAGE_SIZE);\n"
5224 " if (pagesize == -1)\n"
5225 " handle_error(\"sysconf\");\n"
5229 #: build/C/man2/mprotect.2:208
5232 " /* Allocate a buffer aligned on a page boundary;\n"
5233 " initial protection is PROT_READ | PROT_WRITE */\n"
5237 #: build/C/man2/mprotect.2:212
5240 " buffer = memalign(pagesize, 4 * pagesize);\n"
5241 " if (buffer == NULL)\n"
5242 " handle_error(\"memalign\");\n"
5246 #: build/C/man2/mprotect.2:214
5248 msgid " printf(\"Start of region: 0x%lx\\en\", (long) buffer);\n"
5252 #: build/C/man2/mprotect.2:218
5255 " if (mprotect(buffer + pagesize * 2, pagesize,\n"
5256 " PROT_READ) == -1)\n"
5257 " handle_error(\"mprotect\");\n"
5261 #: build/C/man2/mprotect.2:221
5264 " for (p = buffer ; ; )\n"
5265 " *(p++) = \\(aqa\\(aq;\n"
5269 #: build/C/man2/mprotect.2:225
5272 " printf(\"Loop completed\\en\"); /* Should never happen */\n"
5273 " exit(EXIT_SUCCESS);\n"
5278 #: build/C/man2/mprotect.2:229
5279 msgid "B<mmap>(2), B<sysconf>(3)"
5283 #: build/C/man2/mremap.2:30
5289 #: build/C/man2/mremap.2:30
5295 #: build/C/man2/mremap.2:33
5296 msgid "mremap - remap a virtual memory address"
5300 #: build/C/man2/mremap.2:36
5302 msgid "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
5306 #: build/C/man2/mremap.2:41
5309 "B<void *mremap(void *>I<old_address>B<, size_t >I<old_size>B<,>\n"
5310 "B< size_t >I<new_size>B<, int >I<flags>B<, ... /* void "
5311 "*>I<new_address>B< */);>\n"
5315 #: build/C/man2/mremap.2:47
5317 "B<mremap>() expands (or shrinks) an existing memory mapping, potentially "
5318 "moving it at the same time (controlled by the I<flags> argument and the "
5319 "available virtual address space)."
5323 #: build/C/man2/mremap.2:61
5325 "I<old_address> is the old address of the virtual memory block that you want "
5326 "to expand (or shrink). Note that I<old_address> has to be page aligned. "
5327 "I<old_size> is the old size of the virtual memory block. I<new_size> is the "
5328 "requested size of the virtual memory block after the resize. An optional "
5329 "fifth argument, I<new_address>, may be provided; see the description of "
5330 "B<MREMAP_FIXED> below."
5334 #: build/C/man2/mremap.2:73
5336 "In Linux the memory is divided into pages. A user process has (one or) "
5337 "several linear virtual memory segments. Each virtual memory segment has one "
5338 "or more mappings to real memory pages (in the page table). Each virtual "
5339 "memory segment has its own protection (access rights), which may cause a "
5340 "segmentation violation if the memory is accessed incorrectly (e.g., writing "
5341 "to a read-only segment). Accessing virtual memory outside of the segments "
5342 "will also cause a segmentation violation."
5346 #: build/C/man2/mremap.2:81
5348 "B<mremap>() uses the Linux page table scheme. B<mremap>() changes the "
5349 "mapping between virtual addresses and memory pages. This can be used to "
5350 "implement a very efficient B<realloc>(3)."
5354 #: build/C/man2/mremap.2:83
5355 msgid "The I<flags> bit-mask argument may be 0, or include the following flag:"
5359 #: build/C/man2/mremap.2:83
5361 msgid "B<MREMAP_MAYMOVE>"
5365 #: build/C/man2/mremap.2:95
5367 "By default, if there is not sufficient space to expand a mapping at its "
5368 "current location, then B<mremap>() fails. If this flag is specified, then "
5369 "the kernel is permitted to relocate the mapping to a new virtual address, if "
5370 "necessary. If the mapping is relocated, then absolute pointers into the old "
5371 "mapping location become invalid (offsets relative to the starting address of "
5372 "the mapping should be employed)."
5376 #: build/C/man2/mremap.2:95
5378 msgid "B<MREMAP_FIXED> (since Linux 2.3.31)"
5382 #: build/C/man2/mremap.2:117
5384 "This flag serves a similar purpose to the B<MAP_FIXED> flag of B<mmap>(2). "
5385 "If this flag is specified, then B<mremap>() accepts a fifth argument, "
5386 "I<void *new_address>, which specifies a page-aligned address to which the "
5387 "mapping must be moved. Any previous mapping at the address range specified "
5388 "by I<new_address> and I<new_size> is unmapped. If B<MREMAP_FIXED> is "
5389 "specified, then B<MREMAP_MAYMOVE> must also be specified."
5393 #: build/C/man2/mremap.2:127
5395 "If the memory segment specified by I<old_address> and I<old_size> is locked "
5396 "(using B<mlock>(2) or similar), then this lock is maintained when the "
5397 "segment is resized and/or relocated. As a consequence, the amount of memory "
5398 "locked by the process may change."
5402 #: build/C/man2/mremap.2:135
5404 "On success B<mremap>() returns a pointer to the new virtual memory area. "
5405 "On error, the value B<MAP_FAILED> (that is, I<(void\\ *)\\ -1>) is returned, "
5406 "and I<errno> is set appropriately."
5410 #: build/C/man2/mremap.2:142
5412 "The caller tried to expand a memory segment that is locked, but this was not "
5413 "possible without exceeding the B<RLIMIT_MEMLOCK> resource limit."
5417 #: build/C/man2/mremap.2:151
5419 "\"Segmentation fault.\" Some address in the range I<old_address> to "
5420 "I<old_address>+I<old_size> is an invalid virtual memory address for this "
5421 "process. You can also get B<EFAULT> even if there exist mappings that cover "
5422 "the whole address space requested, but those mappings are of different "
5427 #: build/C/man2/mremap.2:179
5429 "An invalid argument was given. Possible causes are: I<old_address> was not "
5430 "page aligned; a value other than B<MREMAP_MAYMOVE> or B<MREMAP_FIXED> was "
5431 "specified in I<flags>; I<new_size> was zero; I<new_size> or I<new_address> "
5432 "was invalid; or the new address range specified by I<new_address> and "
5433 "I<new_size> overlapped the old address range specified by I<old_address> and "
5434 "I<old_size>; or B<MREMAP_FIXED> was specified without also specifying "
5435 "B<MREMAP_MAYMOVE>."
5439 #: build/C/man2/mremap.2:185
5441 "The memory area cannot be expanded at the current virtual address, and the "
5442 "B<MREMAP_MAYMOVE> flag is not set in I<flags>. Or, there is not enough "
5443 "(virtual) memory available."
5446 #. 4.2BSD had a (never actually implemented)
5448 #. call with completely different semantics.
5450 #: build/C/man2/mremap.2:191
5452 "This call is Linux-specific, and should not be used in programs intended to "
5457 #: build/C/man2/mremap.2:199
5459 "Prior to version 2.4, glibc did not expose the definition of "
5460 "B<MREMAP_FIXED>, and the prototype for B<mremap>() did not allow for the "
5461 "I<new_address> argument."
5465 #: build/C/man2/mremap.2:208
5467 "B<brk>(2), B<getpagesize>(2), B<getrlimit>(2), B<mlock>(2), B<mmap>(2), "
5468 "B<sbrk>(2), B<malloc>(3), B<realloc>(3)"
5472 #: build/C/man2/mremap.2:214
5474 "Your favorite text book on operating systems for more information on paged "
5475 "memory (e.g., I<Modern Operating Systems> by Andrew S. Tanenbaum, I<Inside "
5476 "Linux> by Randolf Bentson, I<The Design of the UNIX Operating System> by "
5481 #: build/C/man2/msync.2:25
5487 #: build/C/man2/msync.2:25 build/C/man2/remap_file_pages.2:28
5493 #: build/C/man2/msync.2:28
5494 msgid "msync - synchronize a file with a memory map"
5498 #: build/C/man2/msync.2:32
5499 msgid "B<int msync(void *>I<addr>B<, size_t >I<length>B<, int >I<flags>B<);>"
5503 #: build/C/man2/msync.2:48
5505 "B<msync>() flushes changes made to the in-core copy of a file that was "
5506 "mapped into memory using B<mmap>(2) back to disk. Without use of this call "
5507 "there is no guarantee that changes are written back before B<munmap>(2) is "
5508 "called. To be more precise, the part of the file that corresponds to the "
5509 "memory area starting at I<addr> and having length I<length> is updated."
5513 #: build/C/man2/msync.2:68
5515 "The I<flags> argument may have the bits B<MS_ASYNC>, B<MS_SYNC>, and "
5516 "B<MS_INVALIDATE> set, but not both B<MS_ASYNC> and B<MS_SYNC>. B<MS_ASYNC> "
5517 "specifies that an update be scheduled, but the call returns immediately. "
5518 "B<MS_SYNC> asks for an update and waits for it to complete. "
5519 "B<MS_INVALIDATE> asks to invalidate other mappings of the same file (so that "
5520 "they can be updated with the fresh values just written)."
5524 #: build/C/man2/msync.2:73
5526 "On success, zero is returned. On error, -1 is returned, and I<errno> is set "
5531 #: build/C/man2/msync.2:74
5537 #: build/C/man2/msync.2:80
5539 "B<MS_INVALIDATE> was specified in I<flags>, and a memory lock exists for the "
5540 "specified address range."
5544 #: build/C/man2/msync.2:93
5546 "I<addr> is not a multiple of PAGESIZE; or any bit other than B<MS_ASYNC> | "
5547 "B<MS_INVALIDATE> | B<MS_SYNC> is set in I<flags>; or both B<MS_SYNC> and "
5548 "B<MS_ASYNC> are set in I<flags>."
5552 #: build/C/man2/msync.2:96
5553 msgid "The indicated memory (or part of it) was not mapped."
5557 #: build/C/man2/msync.2:98 build/C/man3/posix_fallocate.3:105 build/C/man3/shm_open.3:251 build/C/man7/shm_overview.7:103
5558 msgid "POSIX.1-2001."
5562 #: build/C/man2/msync.2:105
5564 "This call was introduced in Linux 1.3.21, and then used B<EFAULT> instead of "
5565 "B<ENOMEM>. In Linux 2.4.19 this was changed to the POSIX value B<ENOMEM>."
5568 #. POSIX.1-2001: It shall be defined to -1 or 0 or 200112L.
5569 #. -1: unavailable, 0: ask using sysconf().
5570 #. glibc defines them to 1.
5572 #: build/C/man2/msync.2:118
5574 "On POSIX systems on which B<msync>() is available, both "
5575 "B<_POSIX_MAPPED_FILES> and B<_POSIX_SYNCHRONIZED_IO> are defined in "
5576 "I<E<lt>unistd.hE<gt>> to a value greater than 0. (See also B<sysconf>(3).)"
5580 #: build/C/man2/msync.2:120 build/C/man7/shm_overview.7:46
5586 #: build/C/man3/mtrace.3:25
5592 #: build/C/man3/mtrace.3:28
5593 msgid "mtrace, muntrace - malloc tracing"
5597 #: build/C/man3/mtrace.3:30
5598 msgid "B<#include E<lt>mcheck.hE<gt>>"
5602 #: build/C/man3/mtrace.3:32
5603 msgid "B<void mtrace(void);>"
5607 #: build/C/man3/mtrace.3:34
5608 msgid "B<void muntrace(void);>"
5612 #: build/C/man3/mtrace.3:46
5614 "The B<mtrace>() function installs hook functions for the memory-allocation "
5615 "functions (B<malloc>(3), B<realloc>(3) B<memalign>(3), B<free>(3)). These "
5616 "hook functions record tracing information about memory allocation and "
5617 "deallocation. The tracing information can be used to discover memory leaks "
5618 "and attempts to free nonallocated memory in a program."
5622 #: build/C/man3/mtrace.3:57
5624 "The B<muntrace>() function disables the hook functions installed by "
5625 "B<mtrace>(), so that tracing information is no longer recorded for the "
5626 "memory-allocation functions. If no hook functions were successfully "
5627 "installed by B<mtrace>(), B<muntrace>() does nothing."
5631 #: build/C/man3/mtrace.3:65
5633 "When B<mtrace>() is called, it checks the value of the environment variable "
5634 "B<MALLOC_TRACE>, which should contain the pathname of a file in which the "
5635 "tracing information is to be recorded. If the pathname is successfully "
5636 "opened, it is truncated to zero length."
5640 #: build/C/man3/mtrace.3:78
5642 "If B<MALLOC_TRACE> is not set, or the pathname it specifies is invalid or "
5643 "not writable, then no hook functions are installed, and B<mtrace>() has no "
5644 "effect. In set-user-ID and set-group-ID programs, B<MALLOC_TRACE> is "
5645 "ignored, and B<mtrace>() has no effect."
5649 #: build/C/man3/mtrace.3:86
5651 "In normal usage, B<mtrace>() is called once at the start of execution of a "
5652 "program, and B<muntrace>() is never called."
5656 #: build/C/man3/mtrace.3:96
5658 "The tracing output produced after a call to B<mtrace>() is textual, but not "
5659 "designed to be human readable. The GNU C library provides a Perl script, "
5660 "B<mtrace>(1), that interprets the trace log and produces human-readable "
5661 "output. For best results, the traced program should be compiled with "
5662 "debugging enabled, so that line-number information is recorded in the "
5667 #: build/C/man3/mtrace.3:102
5669 "The tracing performed by B<mtrace>() incurs a performance penalty (if "
5670 "B<MALLOC_TRACE> points to a valid, writable pathname)."
5674 #: build/C/man3/mtrace.3:108
5676 "The line-number information produced by B<mtrace>(1) is not always precise: "
5677 "the line number references may refer to the previous or following "
5678 "(non-blank) line of the source code."
5682 #: build/C/man3/mtrace.3:115
5684 "The shell session below demonstrates the use of the B<mtrace>() function "
5685 "and the B<mtrace>(1) command in a program that has memory leaks at two "
5686 "different locations. The demonstration uses the following program:"
5690 #: build/C/man3/mtrace.3:122
5693 "$ B<cat t_mtrace.c>\n"
5694 "#include E<lt>mcheck.hE<gt>\n"
5695 "#include E<lt>stdlib.hE<gt>\n"
5696 "#include E<lt>stdio.hE<gt>\n"
5700 #: build/C/man3/mtrace.3:127
5704 "main(int argc, char *argv[])\n"
5710 #: build/C/man3/mtrace.3:129
5712 msgid " mtrace();\n"
5716 #: build/C/man3/mtrace.3:132
5719 " for (j = 0; j E<lt> 2; j++)\n"
5720 " malloc(100); /* Never freed--a memory leak */\n"
5724 #: build/C/man3/mtrace.3:136
5727 " calloc(16, 16); /* Never freed--a memory leak */\n"
5728 " exit(EXIT_SUCCESS);\n"
5733 #: build/C/man3/mtrace.3:142
5735 "When we run the program as follows, we see that B<mtrace>() diagnosed "
5736 "memory leaks at two different locations in the program:"
5740 #: build/C/man3/mtrace.3:155
5743 "$ B<cc -g t_mtrace.c -o t_mtrace>\n"
5744 "$ B<export MALLOC_TRACE=/tmp/t>\n"
5746 "$ B<mtrace ./t_mtrace $MALLOC_TRACE>\n"
5747 "Memory not freed:\n"
5748 "-----------------\n"
5749 " Address Size Caller\n"
5750 "0x084c9378 0x64 at /home/cecilia/t_mtrace.c:12\n"
5751 "0x084c93e0 0x64 at /home/cecilia/t_mtrace.c:12\n"
5752 "0x084c9448 0x100 at /home/cecilia/t_mtrace.c:16\n"
5756 #: build/C/man3/mtrace.3:167
5758 "The first two messages about unfreed memory correspond to the two "
5759 "B<malloc>(3) calls inside the I<for> loop. The final message corresponds "
5760 "to the call to B<calloc>(3) (which in turn calls B<malloc>(3))."
5764 #: build/C/man3/mtrace.3:172
5765 msgid "B<mtrace>(1), B<malloc>(3), B<malloc_hook>(3), B<mcheck>(3)"
5769 #: build/C/man2/posix_fadvise.2:28
5771 msgid "POSIX_FADVISE"
5775 #: build/C/man2/posix_fadvise.2:28 build/C/man2/sync_file_range.2:30
5781 #: build/C/man2/posix_fadvise.2:31
5782 msgid "posix_fadvise - predeclare an access pattern for file data"
5786 #: build/C/man2/posix_fadvise.2:34 build/C/man3/posix_fallocate.3:31
5788 msgid "B<#include E<lt>fcntl.hE<gt>>\n"
5792 #: build/C/man2/posix_fadvise.2:37
5795 "B<int posix_fadvise(int >I<fd>B<, off_t >I<offset>B<, off_t >I<len>B<, int "
5800 #: build/C/man2/posix_fadvise.2:46
5801 msgid "B<posix_fadvise>():"
5805 #: build/C/man2/posix_fadvise.2:48 build/C/man3/posix_fallocate.3:44
5806 msgid "_XOPEN_SOURCE\\ E<gt>=\\ 600 || _POSIX_C_SOURCE\\ E<gt>=\\ 200112L"
5810 #: build/C/man2/posix_fadvise.2:56
5812 "Programs can use B<posix_fadvise>() to announce an intention to access file "
5813 "data in a specific pattern in the future, thus allowing the kernel to "
5814 "perform appropriate optimizations."
5818 #: build/C/man2/posix_fadvise.2:63
5820 "The I<advice> applies to a (not necessarily existent) region starting at "
5821 "I<offset> and extending for I<len> bytes (or until the end of the file if "
5822 "I<len> is 0) within the file referred to by I<fd>. The I<advice> is not "
5823 "binding; it merely constitutes an expectation on behalf of the application."
5827 #: build/C/man2/posix_fadvise.2:65
5828 msgid "Permissible values for I<advice> include:"
5832 #: build/C/man2/posix_fadvise.2:65
5834 msgid "B<POSIX_FADV_NORMAL>"
5838 #: build/C/man2/posix_fadvise.2:71
5840 "Indicates that the application has no advice to give about its access "
5841 "pattern for the specified data. If no advice is given for an open file, "
5842 "this is the default assumption."
5846 #: build/C/man2/posix_fadvise.2:71
5848 msgid "B<POSIX_FADV_SEQUENTIAL>"
5852 #: build/C/man2/posix_fadvise.2:75
5854 "The application expects to access the specified data sequentially (with "
5855 "lower offsets read before higher ones)."
5859 #: build/C/man2/posix_fadvise.2:75
5861 msgid "B<POSIX_FADV_RANDOM>"
5865 #: build/C/man2/posix_fadvise.2:78
5866 msgid "The specified data will be accessed in random order."
5870 #: build/C/man2/posix_fadvise.2:78
5872 msgid "B<POSIX_FADV_NOREUSE>"
5876 #: build/C/man2/posix_fadvise.2:81
5877 msgid "The specified data will be accessed only once."
5881 #: build/C/man2/posix_fadvise.2:81
5883 msgid "B<POSIX_FADV_WILLNEED>"
5887 #: build/C/man2/posix_fadvise.2:84
5888 msgid "The specified data will be accessed in the near future."
5892 #: build/C/man2/posix_fadvise.2:84
5894 msgid "B<POSIX_FADV_DONTNEED>"
5898 #: build/C/man2/posix_fadvise.2:87
5899 msgid "The specified data will not be accessed in the near future."
5903 #: build/C/man2/posix_fadvise.2:90
5904 msgid "On success, zero is returned. On error, an error number is returned."
5908 #: build/C/man2/posix_fadvise.2:94
5909 msgid "The I<fd> argument was not a valid file descriptor."
5913 #: build/C/man2/posix_fadvise.2:97
5914 msgid "An invalid value was specified for I<advice>."
5918 #: build/C/man2/posix_fadvise.2:104
5920 "The specified file descriptor refers to a pipe or FIFO. (Linux actually "
5921 "returns B<EINVAL> in this case.)"
5924 #. of fadvise64_64()
5926 #: build/C/man2/posix_fadvise.2:112
5928 "Kernel support first appeared in Linux 2.5.60; the underlying system call is "
5929 "called B<fadvise64>(). Library support has been provided since glibc "
5930 "version 2.2, via the wrapper function B<posix_fadvise>()."
5934 #: build/C/man2/posix_fadvise.2:121
5936 "POSIX.1-2001. Note that the type of the I<len> argument was changed from "
5937 "I<size_t> to I<off_t> in POSIX.1-2003 TC1."
5941 #: build/C/man2/posix_fadvise.2:127
5943 "Under Linux, B<POSIX_FADV_NORMAL> sets the readahead window to the default "
5944 "size for the backing device; B<POSIX_FADV_SEQUENTIAL> doubles this size, and "
5945 "B<POSIX_FADV_RANDOM> disables file readahead entirely. These changes affect "
5946 "the entire file, not just the specified region (but other open file handles "
5947 "to the same file are unaffected)."
5951 #: build/C/man2/posix_fadvise.2:134
5953 "B<POSIX_FADV_WILLNEED> initiates a nonblocking read of the specified region "
5954 "into the page cache. The amount of data read may be decreased by the kernel "
5955 "depending on virtual memory load. (A few megabytes will usually be fully "
5956 "satisfied, and more is rarely useful.)"
5960 #: build/C/man2/posix_fadvise.2:138
5962 "In kernels before 2.6.18, B<POSIX_FADV_NOREUSE> had the same semantics as "
5963 "B<POSIX_FADV_WILLNEED>. This was probably a bug; since kernel 2.6.18, this "
5968 #: build/C/man2/posix_fadvise.2:146
5970 "B<POSIX_FADV_DONTNEED> attempts to free cached pages associated with the "
5971 "specified region. This is useful, for example, while streaming large "
5972 "files. A program may periodically request the kernel to free cached data "
5973 "that has already been used, so that more useful cached pages are not "
5974 "discarded instead."
5978 #: build/C/man2/posix_fadvise.2:154
5980 "Pages that have not yet been written out will be unaffected, so if the "
5981 "application wishes to guarantee that pages will be released, it should call "
5982 "B<fsync>(2) or B<fdatasync>(2) first."
5986 #: build/C/man2/posix_fadvise.2:154
5988 msgid "arm_fadvise()"
5992 #: build/C/man2/posix_fadvise.2:166
5994 "The ARM architecture needs 64-bit arguments to be aligned in a suitable pair "
5995 "of registers. On this architecture, the call signature of "
5996 "B<posix_fadvise>() is flawed, since it forces a register to be wasted as "
5997 "padding between the I<fd> and I<len> arguments. Therefore, since Linux "
5998 "2.6.14, ARM defines a different system call that orders the arguments "
6003 #: build/C/man2/posix_fadvise.2:171
6006 "B<long arm_fadvise64_64(int >I<fd>B<, int >I<advice>B<,>\n"
6007 "B< loff_t >I<offset>B<, loff_t >I<len>B<);>\n"
6010 #. No ARM support in glibc.
6012 #: build/C/man2/posix_fadvise.2:178
6014 "The behavior of this system call is otherwise exactly the same as "
6015 "B<posix_fadvise>(). No library support is provided for this system call in "
6020 #: build/C/man2/posix_fadvise.2:183
6022 "In kernels before 2.6.6, if I<len> was specified as 0, then this was "
6023 "interpreted literally as \"zero bytes\", rather than as meaning \"all bytes "
6024 "through to the end of the file\"."
6027 #. FIXME . Write a posix_fadvise(3) page.
6029 #: build/C/man2/posix_fadvise.2:189
6031 "B<readahead>(2), B<sync_file_range>(2), B<posix_fallocate>(3), "
6032 "B<posix_madvise>(3)"
6036 #: build/C/man3/posix_fallocate.3:25
6038 msgid "POSIX_FALLOCATE"
6042 #: build/C/man3/posix_fallocate.3:28
6043 msgid "posix_fallocate - allocate file space"
6047 #: build/C/man3/posix_fallocate.3:33
6049 msgid "B<int posix_fallocate(int >I<fd>B<, off_t >I<offset>B<, off_t >I<len>B<);>\n"
6053 #: build/C/man3/posix_fallocate.3:42
6054 msgid "B<posix_fallocate>():"
6058 #: build/C/man3/posix_fallocate.3:61
6060 "The function B<posix_fallocate>() ensures that disk space is allocated for "
6061 "the file referred to by the descriptor I<fd> for the bytes in the range "
6062 "starting at I<offset> and continuing for I<len> bytes. After a successful "
6063 "call to B<posix_fallocate>(), subsequent writes to bytes in the specified "
6064 "range are guaranteed not to fail because of lack of disk space."
6068 #: build/C/man3/posix_fallocate.3:66
6070 "If the size of the file is less than I<offset>+I<len>, then the file is "
6071 "increased to this size; otherwise the file size is left unchanged."
6075 #: build/C/man3/posix_fallocate.3:72
6077 "B<posix_fallocate>() returns zero on success, or an error number on "
6078 "failure. Note that I<errno> is not set."
6082 #: build/C/man3/posix_fallocate.3:81
6083 msgid "I<offset+len> exceeds the maximum file size."
6087 #: build/C/man3/posix_fallocate.3:91
6088 msgid "I<fd> does not refer to a regular file."
6092 #: build/C/man3/posix_fallocate.3:100
6093 msgid "I<fd> refers to a pipe."
6097 #: build/C/man3/posix_fallocate.3:103
6098 msgid "B<posix_fallocate>() is available since glibc 2.1.94."
6102 #: build/C/man3/posix_fallocate.3:128
6104 "POSIX.1-2008 says that an implementation I<shall> give the B<EINVAL> error "
6105 "if I<len> was 0, or I<offset> was less than 0. POSIX.1-2001 says that an "
6106 "implementation I<shall> give the B<EINVAL> error if I<len> is less than 0, "
6107 "or I<offset> was less than 0, and I<may> give the error if I<len> equals "
6112 #: build/C/man3/posix_fallocate.3:133
6113 msgid "B<fallocate>(1), B<fallocate>(2), B<lseek>(2), B<posix_fadvise>(2)"
6117 #: build/C/man3/posix_memalign.3:29
6119 msgid "POSIX_MEMALIGN"
6123 #: build/C/man3/posix_memalign.3:29
6129 #: build/C/man3/posix_memalign.3:32
6131 "posix_memalign, aligned_alloc, memalign, valloc, pvalloc - allocate aligned "
6136 #: build/C/man3/posix_memalign.3:39
6139 "B<int posix_memalign(void **>I<memptr>B<, size_t >I<alignment>B<, size_t "
6141 "B<void *aligned_alloc(size_t >I<alignment>B<, size_t >I<size>B<);>\n"
6142 "B<void *valloc(size_t >I<size>B<);>\n"
6146 #: build/C/man3/posix_memalign.3:44
6149 "B<void *memalign(size_t >I<alignment>B<, size_t >I<size>B<);>\n"
6150 "B<void *pvalloc(size_t >I<size>B<);>\n"
6154 #: build/C/man3/posix_memalign.3:54
6156 "B<posix_memalign>(): _POSIX_C_SOURCE\\ E<gt>=\\ 200112L || _XOPEN_SOURCE\\ "
6161 #: build/C/man3/posix_memalign.3:57
6162 msgid "B<aligned_alloc>(): _ISOC11_SOURCE"
6166 #: build/C/man3/posix_memalign.3:59
6167 msgid "B<valloc>():"
6171 #: build/C/man3/posix_memalign.3:62
6173 msgid "Since glibc 2.12:"
6177 #: build/C/man3/posix_memalign.3:69
6181 " (_XOPEN_SOURCE\\ E<gt>=\\ 500 ||\n"
6182 " _XOPEN_SOURCE\\ &&\\ _XOPEN_SOURCE_EXTENDED) &&\n"
6183 " !(_POSIX_C_SOURCE\\ E<gt>=\\ 200112L || _XOPEN_SOURCE\\ E<gt>=\\ 600)\n"
6187 #: build/C/man3/posix_memalign.3:71
6189 msgid "Before glibc 2.12:"
6193 #: build/C/man3/posix_memalign.3:75
6195 "_BSD_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 500 || _XOPEN_SOURCE\\ &&\\ "
6196 "_XOPEN_SOURCE_EXTENDED"
6200 #: build/C/man3/posix_memalign.3:82
6202 "(The (nonstandard) header file I<E<lt>malloc.hE<gt>> also exposes the "
6203 "declaration of B<valloc>(); no feature test macros are required.)"
6208 #: build/C/man3/posix_memalign.3:103
6210 "The function B<posix_memalign>() allocates I<size> bytes and places the "
6211 "address of the allocated memory in I<*memptr>. The address of the allocated "
6212 "memory will be a multiple of I<alignment>, which must be a power of two and "
6213 "a multiple of I<sizeof(void *)>. If I<size> is 0, then B<posix_memalign>() "
6214 "returns either NULL, or a unique pointer value that can later be "
6215 "successfully passed to B<free>(3)."
6218 #. The behavior of memalign() for size==0 is as for posix_memalign()
6219 #. but no standards govern this.
6221 #: build/C/man3/posix_memalign.3:114
6223 "The obsolete function B<memalign>() allocates I<size> bytes and returns a "
6224 "pointer to the allocated memory. The memory address will be a multiple of "
6225 "I<alignment>, which must be a power of two."
6229 #: build/C/man3/posix_memalign.3:123
6231 "The function B<aligned_alloc>() is the same as B<memalign>(), except for "
6232 "the added restriction that I<size> should be a multiple of I<alignment>."
6236 #: build/C/man3/posix_memalign.3:132
6238 "The obsolete function B<valloc>() allocates I<size> bytes and returns a "
6239 "pointer to the allocated memory. The memory address will be a multiple of "
6240 "the page size. It is equivalent to I<memalign(sysconf(_SC_PAGESIZE),size)>."
6244 #: build/C/man3/posix_memalign.3:139
6246 "The obsolete function B<pvalloc>() is similar to B<valloc>(), but rounds "
6247 "the size of the allocation up to the next multiple of the system page size."
6251 #: build/C/man3/posix_memalign.3:141
6252 msgid "For all of these functions, the memory is not zeroed."
6256 #: build/C/man3/posix_memalign.3:148
6258 "B<aligned_alloc>(), B<memalign>(), B<valloc>(), and B<pvalloc>() return a "
6259 "pointer to the allocated memory, or NULL if the request fails."
6263 #: build/C/man3/posix_memalign.3:155
6265 "B<posix_memalign>() returns zero on success, or one of the error values "
6266 "listed in the next section on failure. Note that I<errno> is not set."
6270 #: build/C/man3/posix_memalign.3:162
6272 "The I<alignment> argument was not a power of two, or was not a multiple of "
6273 "I<sizeof(void *)>."
6277 #: build/C/man3/posix_memalign.3:165
6278 msgid "There was insufficient memory to fulfill the allocation request."
6282 #: build/C/man3/posix_memalign.3:172
6284 "The functions B<memalign>(), B<valloc>(), and B<pvalloc>() have been "
6285 "available in all Linux libc libraries."
6289 #: build/C/man3/posix_memalign.3:176
6290 msgid "The function B<aligned_alloc>() was added to glibc in version 2.16."
6294 #: build/C/man3/posix_memalign.3:180
6295 msgid "The function B<posix_memalign>() is available since glibc 2.1.91."
6299 #: build/C/man3/posix_memalign.3:187
6301 "The function B<valloc>() appeared in 3.0BSD. It is documented as being "
6302 "obsolete in 4.3BSD, and as legacy in SUSv2. It does not appear in "
6307 #: build/C/man3/posix_memalign.3:191
6308 msgid "The function B<pvalloc>() is a GNU extension."
6312 #: build/C/man3/posix_memalign.3:195
6313 msgid "The function B<memalign>() appears in SunOS 4.1.3 but not in 4.4BSD."
6317 #: build/C/man3/posix_memalign.3:199
6318 msgid "The function B<posix_memalign>() comes from POSIX.1d."
6322 #: build/C/man3/posix_memalign.3:204
6323 msgid "The function B<aligned_alloc>() is specified in the C11 standard."
6327 #: build/C/man3/posix_memalign.3:204
6333 #: build/C/man3/posix_memalign.3:208
6335 "Everybody agrees that B<posix_memalign>() is declared in "
6336 "I<E<lt>stdlib.hE<gt>>."
6340 #: build/C/man3/posix_memalign.3:212
6342 "On some systems B<memalign>() is declared in I<E<lt>stdlib.hE<gt>> instead "
6343 "of I<E<lt>malloc.hE<gt>>."
6347 #: build/C/man3/posix_memalign.3:219
6349 "According to SUSv2, B<valloc>() is declared in I<E<lt>stdlib.hE<gt>>. "
6350 "Libc4,5 and glibc declare it in I<E<lt>malloc.hE<gt>>, and also in "
6351 "I<E<lt>stdlib.hE<gt>> if suitable feature test macros are defined (see "
6356 #: build/C/man3/posix_memalign.3:228
6358 "On many systems there are alignment restrictions, for example, on buffers "
6359 "used for direct block device I/O. POSIX specifies the "
6360 "I<pathconf(path,_PC_REC_XFER_ALIGN)> call that tells what alignment is "
6361 "needed. Now one can use B<posix_memalign>() to satisfy this requirement."
6365 #: build/C/man3/posix_memalign.3:237
6367 "B<posix_memalign>() verifies that I<alignment> matches the requirements "
6368 "detailed above. B<memalign>() may not check that the I<alignment> argument "
6372 #. Other systems allow passing the result of
6379 #: build/C/man3/posix_memalign.3:265
6381 "POSIX requires that memory obtained from B<posix_memalign>() can be freed "
6382 "using B<free>(3). Some systems provide no way to reclaim memory allocated "
6383 "with B<memalign>() or B<valloc>() (because one can only pass to B<free>(3) "
6384 "a pointer gotten from B<malloc>(3), while, for example, B<memalign>() would "
6385 "call B<malloc>(3) and then align the obtained value). The glibc "
6386 "implementation allows memory obtained from any of these functions to be "
6387 "reclaimed with B<free>(3)."
6391 #: build/C/man3/posix_memalign.3:270
6393 "The glibc B<malloc>(3) always returns 8-byte aligned memory addresses, so "
6394 "these functions are only needed if you require larger alignment values."
6398 #: build/C/man3/posix_memalign.3:275
6399 msgid "B<brk>(2), B<getpagesize>(2), B<free>(3), B<malloc>(3)"
6403 #: build/C/man2/readahead.2:28
6409 #: build/C/man2/readahead.2:28 build/C/man7/shm_overview.7:27
6415 #: build/C/man2/readahead.2:31
6416 msgid "readahead - perform file readahead into page cache"
6420 #: build/C/man2/readahead.2:37
6423 "B<ssize_t readahead(int >I<fd>B<, off64_t >I<offset>B<, size_t "
6428 #: build/C/man2/readahead.2:65
6430 "B<readahead>() populates the page cache with data from a file so that "
6431 "subsequent reads from that file will not block on disk I/O. The I<fd> "
6432 "argument is a file descriptor identifying the file which is to be read. The "
6433 "I<offset> argument specifies the starting point from which data is to be "
6434 "read and I<count> specifies the number of bytes to be read. I/O is "
6435 "performed in whole pages, so that I<offset> is effectively rounded down to a "
6436 "page boundary and bytes are read up to the next page boundary greater than "
6437 "or equal to I<(offset+count)>. B<readahead>() does not read beyond the end "
6438 "of the file. B<readahead>() blocks until the specified data has been "
6439 "read. The current file offset of the open file referred to by I<fd> is left "
6444 #: build/C/man2/readahead.2:71
6446 "On success, B<readahead>() returns 0; on failure, -1 is returned, with "
6447 "I<errno> set to indicate the cause of the error."
6451 #: build/C/man2/readahead.2:76
6452 msgid "I<fd> is not a valid file descriptor or is not open for reading."
6456 #: build/C/man2/readahead.2:82
6457 msgid "I<fd> does not refer to a file type to which B<readahead>() can be applied."
6461 #: build/C/man2/readahead.2:87
6463 "The B<readahead>() system call appeared in Linux 2.4.13; glibc support has "
6464 "been provided since version 2.3."
6468 #: build/C/man2/readahead.2:92
6470 "The B<readahead>() system call is Linux-specific, and its use should be "
6471 "avoided in portable applications."
6475 #: build/C/man2/readahead.2:98
6476 msgid "B<lseek>(2), B<madvise>(2), B<mmap>(2), B<posix_fadvise>(2), B<read>(2)"
6480 #: build/C/man2/remap_file_pages.2:28
6482 msgid "REMAP_FILE_PAGES"
6486 #: build/C/man2/remap_file_pages.2:31
6487 msgid "remap_file_pages - create a nonlinear file mapping"
6491 #: build/C/man2/remap_file_pages.2:35
6494 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
6495 "B<#include E<lt>sys/mman.hE<gt>>\n"
6499 #: build/C/man2/remap_file_pages.2:38
6502 "B<int remap_file_pages(void *>I<addr>B<, size_t >I<size>B<, int "
6504 "B< ssize_t >I<pgoff>B<, int >I<flags>B<);>\n"
6508 #: build/C/man2/remap_file_pages.2:51
6510 "The B<remap_file_pages>() system call is used to create a nonlinear "
6511 "mapping, that is, a mapping in which the pages of the file are mapped into a "
6512 "nonsequential order in memory. The advantage of using B<remap_file_pages>() "
6513 "over using repeated calls to B<mmap>(2) is that the former approach does "
6514 "not require the kernel to create additional VMA (Virtual Memory Area) data "
6519 #: build/C/man2/remap_file_pages.2:53
6520 msgid "To create a nonlinear mapping we perform the following steps:"
6524 #: build/C/man2/remap_file_pages.2:53
6530 #: build/C/man2/remap_file_pages.2:61
6532 "Use B<mmap>(2) to create a mapping (which is initially linear). This "
6533 "mapping must be created with the B<MAP_SHARED> flag."
6537 #: build/C/man2/remap_file_pages.2:61
6543 #: build/C/man2/remap_file_pages.2:69
6545 "Use one or more calls to B<remap_file_pages>() to rearrange the "
6546 "correspondence between the pages of the mapping and the pages of the file. "
6547 "It is possible to map the same page of a file into multiple locations within "
6548 "the mapped region."
6552 #: build/C/man2/remap_file_pages.2:80
6554 "The I<pgoff> and I<size> arguments specify the region of the file that is to "
6555 "be relocated within the mapping: I<pgoff> is a file offset in units of the "
6556 "system page size; I<size> is the length of the region in bytes."
6560 #: build/C/man2/remap_file_pages.2:98
6562 "The I<addr> argument serves two purposes. First, it identifies the mapping "
6563 "whose pages we want to rearrange. Thus, I<addr> must be an address that "
6564 "falls within a region previously mapped by a call to B<mmap>(2). Second, "
6565 "I<addr> specifies the address at which the file pages identified by I<pgoff> "
6566 "and I<size> will be placed."
6569 #. This rounding is weird, and not consistent with the treatment of
6570 #. the analogous arguments for munmap()/mprotect() and for mlock().
6573 #: build/C/man2/remap_file_pages.2:112
6575 "The values specified in I<addr> and I<size> should be multiples of the "
6576 "system page size. If they are not, then the kernel rounds I<both> values "
6577 "I<down> to the nearest multiple of the page size."
6581 #: build/C/man2/remap_file_pages.2:116
6582 msgid "The I<prot> argument must be specified as 0."
6586 #: build/C/man2/remap_file_pages.2:124
6588 "The I<flags> argument has the same meaning as for B<mmap>(2), but all flags "
6589 "other than B<MAP_NONBLOCK> are ignored."
6593 #: build/C/man2/remap_file_pages.2:131
6595 "On success, B<remap_file_pages>() returns 0. On error, -1 is returned, and "
6596 "I<errno> is set appropriately."
6600 #: build/C/man2/remap_file_pages.2:139
6602 "I<addr> does not refer to a valid mapping created with the B<MAP_SHARED> "
6606 #. And possibly others from vma->vm_ops->populate()
6608 #: build/C/man2/remap_file_pages.2:148
6609 msgid "I<addr>, I<size>, I<prot>, or I<pgoff> is invalid."
6613 #: build/C/man2/remap_file_pages.2:153
6615 "The B<remap_file_pages>() system call appeared in Linux 2.5.46; glibc "
6616 "support was added in version 2.3.3."
6620 #: build/C/man2/remap_file_pages.2:157
6621 msgid "The B<remap_file_pages>() system call is Linux-specific."
6625 #: build/C/man2/remap_file_pages.2:164
6627 "B<getpagesize>(2), B<mmap>(2), B<mmap2>(2), B<mprotect>(2), B<mremap>(2), "
6632 #: build/C/man3/shm_open.3:26
6638 #: build/C/man3/shm_open.3:26
6644 #: build/C/man3/shm_open.3:29
6645 msgid "shm_open, shm_unlink - create/open or unlink POSIX shared memory objects"
6649 #: build/C/man3/shm_open.3:33
6650 msgid "B<#include E<lt>sys/stat.hE<gt>> /* For mode constants */"
6654 #: build/C/man3/shm_open.3:35
6655 msgid "B<#include E<lt>fcntl.hE<gt>> /* For O_* constants */"
6659 #: build/C/man3/shm_open.3:37
6660 msgid "B<int shm_open(const char *>I<name>B<, int >I<oflag>B<, mode_t >I<mode>B<);>"
6664 #: build/C/man3/shm_open.3:39
6665 msgid "B<int shm_unlink(const char *>I<name>B<);>"
6669 #: build/C/man3/shm_open.3:41
6670 msgid "Link with I<-lrt>."
6674 #: build/C/man3/shm_open.3:53
6676 "B<shm_open>() creates and opens a new, or opens an existing, POSIX shared "
6677 "memory object. A POSIX shared memory object is in effect a handle which can "
6678 "be used by unrelated processes to B<mmap>(2) the same region of shared "
6679 "memory. The B<shm_unlink>() function performs the converse operation, "
6680 "removing an object previously created by B<shm_open>()."
6683 #. glibc allows the initial slash to be omitted, and makes
6684 #. multiple initial slashes equivalent to a single slash.
6685 #. This differs from the implementation of POSIX message queues.
6686 #. glibc allows subdirectory components in the name, in which
6687 #. case the subdirectory must exist under /dev/shm, and allow the
6688 #. required permissions if a user wants to create a shared memory
6689 #. object in that subdirectory.
6691 #: build/C/man3/shm_open.3:74
6693 "The operation of B<shm_open>() is analogous to that of B<open>(2). I<name> "
6694 "specifies the shared memory object to be created or opened. For portable "
6695 "use, a shared memory object should be identified by a name of the form "
6696 "I</somename>; that is, a null-terminated string of up to B<NAME_MAX> (i.e., "
6697 "255) characters consisting of an initial slash, followed by one or more "
6698 "characters, none of which are slashes."
6702 #: build/C/man3/shm_open.3:81
6704 "I<oflag> is a bit mask created by ORing together exactly one of B<O_RDONLY> "
6705 "or B<O_RDWR> and any of the other flags listed here:"
6709 #: build/C/man3/shm_open.3:81
6715 #: build/C/man3/shm_open.3:87
6717 "Open the object for read access. A shared memory object opened in this way "
6718 "can only be B<mmap>(2)ed for read (B<PROT_READ>) access."
6722 #: build/C/man3/shm_open.3:87
6728 #: build/C/man3/shm_open.3:90
6729 msgid "Open the object for read-write access."
6733 #: build/C/man3/shm_open.3:90
6738 #. In truth it is actually the file system IDs on Linux, but these
6739 #. are nearly always the same as the effective IDs. (MTK, Jul 05)
6741 #: build/C/man3/shm_open.3:110
6743 "Create the shared memory object if it does not exist. The user and group "
6744 "ownership of the object are taken from the corresponding effective IDs of "
6745 "the calling process, and the object's permission bits are set according to "
6746 "the low-order 9 bits of I<mode>, except that those bits set in the process "
6747 "file mode creation mask (see B<umask>(2)) are cleared for the new object. "
6748 "A set of macro constants which can be used to define I<mode> is listed in "
6749 "B<open>(2). (Symbolic definitions of these constants can be obtained by "
6750 "including I<E<lt>sys/stat.hE<gt>>.)"
6754 #: build/C/man3/shm_open.3:116
6756 "A new shared memory object initially has zero length\\(emthe size of the "
6757 "object can be set using B<ftruncate>(2). The newly allocated bytes of a "
6758 "shared memory object are automatically initialized to 0."
6762 #: build/C/man3/shm_open.3:116
6768 #: build/C/man3/shm_open.3:125
6770 "If B<O_CREAT> was also specified, and a shared memory object with the given "
6771 "I<name> already exists, return an error. The check for the existence of the "
6772 "object, and its creation if it does not exist, are performed atomically."
6776 #: build/C/man3/shm_open.3:125
6782 #: build/C/man3/shm_open.3:128
6783 msgid "If the shared memory object already exists, truncate it to zero bytes."
6787 #: build/C/man3/shm_open.3:131
6789 "Definitions of these flag values can be obtained by including "
6790 "I<E<lt>fcntl.hE<gt>>."
6794 #: build/C/man3/shm_open.3:142
6796 "On successful completion B<shm_open>() returns a new file descriptor "
6797 "referring to the shared memory object. This file descriptor is guaranteed "
6798 "to be the lowest-numbered file descriptor not previously opened within the "
6799 "process. The B<FD_CLOEXEC> flag (see B<fcntl>(2)) is set for the file "
6804 #: build/C/man3/shm_open.3:151
6806 "The file descriptor is normally used in subsequent calls to B<ftruncate>(2) "
6807 "(for a newly created object) and B<mmap>(2). After a call to B<mmap>(2) "
6808 "the file descriptor may be closed without affecting the memory mapping."
6812 #: build/C/man3/shm_open.3:169
6814 "The operation of B<shm_unlink>() is analogous to B<unlink>(2): it removes a "
6815 "shared memory object name, and, once all processes have unmapped the object, "
6816 "de-allocates and destroys the contents of the associated memory region. "
6817 "After a successful B<shm_unlink>(), attempts to B<shm_open>() an object "
6818 "with the same I<name> will fail (unless B<O_CREAT> was specified, in which "
6819 "case a new, distinct object is created)."
6823 #: build/C/man3/shm_open.3:178
6825 "On success, B<shm_open>() returns a nonnegative file descriptor. On "
6826 "failure, B<shm_open>() returns -1. B<shm_unlink>() returns 0 on success, "
6831 #: build/C/man3/shm_open.3:185
6833 "On failure, I<errno> is set to indicate the cause of the error. Values "
6834 "which may appear in I<errno> include the following:"
6838 #: build/C/man3/shm_open.3:190
6839 msgid "Permission to B<shm_unlink>() the shared memory object was denied."
6843 #: build/C/man3/shm_open.3:200
6845 "Permission was denied to B<shm_open>() I<name> in the specified I<mode>, or "
6846 "B<O_TRUNC> was specified and the caller does not have write permission on "
6851 #: build/C/man3/shm_open.3:200 build/C/man2/shmget.2:194
6857 #: build/C/man3/shm_open.3:211
6859 "Both B<O_CREAT> and B<O_EXCL> were specified to B<shm_open>() and the "
6860 "shared memory object specified by I<name> already exists."
6864 #: build/C/man3/shm_open.3:218
6865 msgid "The I<name> argument to B<shm_open>() was invalid."
6869 #: build/C/man3/shm_open.3:218
6875 #: build/C/man3/shm_open.3:221
6876 msgid "The process already has the maximum number of files open."
6880 #: build/C/man3/shm_open.3:221
6882 msgid "B<ENAMETOOLONG>"
6886 #: build/C/man3/shm_open.3:227
6887 msgid "The length of I<name> exceeds B<PATH_MAX>."
6891 #: build/C/man3/shm_open.3:231
6892 msgid "The limit on the total number of files open on the system has been reached."
6896 #: build/C/man3/shm_open.3:231 build/C/man3/shm_open.3:240 build/C/man2/shmget.2:208
6902 #: build/C/man3/shm_open.3:240
6904 "An attempt was made to B<shm_open>() a I<name> that did not exist, and "
6905 "B<O_CREAT> was not specified."
6909 #: build/C/man3/shm_open.3:247
6910 msgid "An attempt was to made to B<shm_unlink>() a I<name> that does not exist."
6914 #: build/C/man3/shm_open.3:249
6915 msgid "These functions are provided in glibc 2.2 and later."
6919 #: build/C/man3/shm_open.3:255
6921 "POSIX.1-2001 says that the group ownership of a newly created shared memory "
6922 "object is set to either the calling process's effective group ID or \"a "
6923 "system default group ID\"."
6927 #: build/C/man3/shm_open.3:264
6929 "POSIX leaves the behavior of the combination of B<O_RDONLY> and B<O_TRUNC> "
6930 "unspecified. On Linux, this will successfully truncate an existing shared "
6931 "memory object\\(emthis may not be so on other UNIX systems."
6935 #: build/C/man3/shm_open.3:269
6937 "The POSIX shared memory object implementation on Linux 2.4 makes use of a "
6938 "dedicated file system, which is normally mounted under I</dev/shm>."
6942 #: build/C/man3/shm_open.3:280
6944 "B<close>(2), B<fchmod>(2), B<fchown>(2), B<fcntl>(2), B<fstat>(2), "
6945 "B<ftruncate>(2), B<mmap>(2), B<open>(2), B<umask>(2), B<shm_overview>(7)"
6949 #: build/C/man7/shm_overview.7:27
6951 msgid "SHM_OVERVIEW"
6955 #: build/C/man7/shm_overview.7:30
6956 msgid "shm_overview - overview of POSIX shared memory"
6960 #: build/C/man7/shm_overview.7:33
6962 "The POSIX shared memory API allows processes to communicate information by "
6963 "sharing a region of memory."
6967 #: build/C/man7/shm_overview.7:35
6968 msgid "The interfaces employed in the API are:"
6972 #: build/C/man7/shm_overview.7:35
6974 msgid "B<shm_open>(3)"
6978 #: build/C/man7/shm_overview.7:42
6980 "Create and open a new object, or open an existing object. This is analogous "
6981 "to B<open>(2). The call returns a file descriptor for use by the other "
6982 "interfaces listed below."
6986 #: build/C/man7/shm_overview.7:42
6988 msgid "B<ftruncate>(2)"
6992 #: build/C/man7/shm_overview.7:46
6994 "Set the size of the shared memory object. (A newly created shared memory "
6995 "object has a length of zero.)"
6999 #: build/C/man7/shm_overview.7:50
7001 "Map the shared memory object into the virtual address space of the calling "
7006 #: build/C/man7/shm_overview.7:50
7008 msgid "B<munmap>(2)"
7012 #: build/C/man7/shm_overview.7:54
7014 "Unmap the shared memory object from the virtual address space of the calling "
7019 #: build/C/man7/shm_overview.7:54
7021 msgid "B<shm_unlink>(3)"
7025 #: build/C/man7/shm_overview.7:57
7026 msgid "Remove a shared memory object name."
7030 #: build/C/man7/shm_overview.7:57
7036 #: build/C/man7/shm_overview.7:62
7038 "Close the file descriptor allocated by B<shm_open>(3) when it is no longer "
7043 #: build/C/man7/shm_overview.7:62
7049 #: build/C/man7/shm_overview.7:76
7051 "Obtain a I<stat> structure that describes the shared memory object. Among "
7052 "the information returned by this call are the object's size (I<st_size>), "
7053 "permissions (I<st_mode>), owner (I<st_uid>), and group (I<st_gid>)."
7057 #: build/C/man7/shm_overview.7:76
7059 msgid "B<fchown>(2)"
7063 #: build/C/man7/shm_overview.7:79
7064 msgid "To change the ownership of a shared memory object."
7068 #: build/C/man7/shm_overview.7:79
7070 msgid "B<fchmod>(2)"
7074 #: build/C/man7/shm_overview.7:82
7075 msgid "To change the permissions of a shared memory object."
7079 #: build/C/man7/shm_overview.7:82
7085 #: build/C/man7/shm_overview.7:84
7086 msgid "POSIX shared memory is supported since Linux 2.4 and glibc 2.2."
7090 #: build/C/man7/shm_overview.7:84
7096 #: build/C/man7/shm_overview.7:89
7098 "POSIX shared memory objects have kernel persistence: a shared memory object "
7099 "will exist until the system is shut down, or until all processes have "
7100 "unmapped the object and it has been deleted with B<shm_unlink>(3)"
7104 #: build/C/man7/shm_overview.7:89
7110 #: build/C/man7/shm_overview.7:94
7112 "Programs using the POSIX shared memory API must be compiled with I<cc -lrt> "
7113 "to link against the real-time library, I<librt>."
7117 #: build/C/man7/shm_overview.7:94
7119 msgid "Accessing shared memory objects via the file system"
7123 #: build/C/man7/shm_overview.7:101
7125 "On Linux, shared memory objects are created in a (I<tmpfs>) virtual file "
7126 "system, normally mounted under I</dev/shm>. Since kernel 2.6.19, Linux "
7127 "supports the use of access control lists (ACLs) to control the permissions "
7128 "of objects in the virtual file system."
7132 #: build/C/man7/shm_overview.7:106
7134 "Typically, processes must synchronize their access to a shared memory "
7135 "object, using, for example, POSIX semaphores."
7139 #: build/C/man7/shm_overview.7:114
7141 "System V shared memory (B<shmget>(2), B<shmop>(2), etc.) is an older shared "
7142 "memory API. POSIX shared memory provides a simpler, and better designed "
7143 "interface; on the other hand POSIX shared memory is somewhat less widely "
7144 "available (especially on older systems) than System V shared memory."
7148 #: build/C/man7/shm_overview.7:127
7150 "B<fchmod>(2), B<fchown>(2), B<fstat>(2), B<ftruncate>(2), B<mmap>(2), "
7151 "B<mprotect>(2), B<munmap>(2), B<shmget>(2), B<shmop>(2), B<shm_open>(3), "
7152 "B<shm_unlink>(3), B<sem_overview>(7)"
7156 #: build/C/man2/shmctl.2:45
7162 #: build/C/man2/shmctl.2:45 build/C/man2/shmget.2:38
7168 #: build/C/man2/shmctl.2:48
7169 msgid "shmctl - shared memory control"
7173 #: build/C/man2/shmctl.2:51 build/C/man2/shmget.2:44
7174 msgid "B<#include E<lt>sys/ipc.hE<gt>>"
7178 #: build/C/man2/shmctl.2:53 build/C/man2/shmget.2:46
7179 msgid "B<#include E<lt>sys/shm.hE<gt>>"
7183 #: build/C/man2/shmctl.2:55
7184 msgid "B<int shmctl(int >I<shmid>B<, int >I<cmd>B<, struct shmid_ds *>I<buf>B<);>"
7188 #: build/C/man2/shmctl.2:62
7190 "B<shmctl>() performs the control operation specified by I<cmd> on the "
7191 "shared memory segment whose identifier is given in I<shmid>."
7195 #: build/C/man2/shmctl.2:67
7197 "The I<buf> argument is a pointer to a I<shmid_ds> structure, defined in "
7198 "I<E<lt>sys/shm.hE<gt>> as follows:"
7202 #: build/C/man2/shmctl.2:81
7205 "struct shmid_ds {\n"
7206 " struct ipc_perm shm_perm; /* Ownership and permissions */\n"
7207 " size_t shm_segsz; /* Size of segment (bytes) */\n"
7208 " time_t shm_atime; /* Last attach time */\n"
7209 " time_t shm_dtime; /* Last detach time */\n"
7210 " time_t shm_ctime; /* Last change time */\n"
7211 " pid_t shm_cpid; /* PID of creator */\n"
7212 " pid_t shm_lpid; /* PID of last shmat(2)/shmdt(2) */\n"
7213 " shmatt_t shm_nattch; /* No. of current attaches */\n"
7219 #: build/C/man2/shmctl.2:89
7221 "The I<ipc_perm> structure is defined as follows (the highlighted fields are "
7222 "settable using B<IPC_SET>):"
7226 #: build/C/man2/shmctl.2:102
7229 "struct ipc_perm {\n"
7230 " key_t __key; /* Key supplied to shmget(2) */\n"
7231 " uid_t B<uid>; /* Effective UID of owner */\n"
7232 " gid_t B<gid>; /* Effective GID of owner */\n"
7233 " uid_t cuid; /* Effective UID of creator */\n"
7234 " gid_t cgid; /* Effective GID of creator */\n"
7235 " unsigned short B<mode>; /* B<Permissions> + SHM_DEST and\n"
7236 " SHM_LOCKED flags */\n"
7237 " unsigned short __seq; /* Sequence number */\n"
7242 #: build/C/man2/shmctl.2:108
7243 msgid "Valid values for I<cmd> are:"
7247 #: build/C/man2/shmctl.2:109
7253 #: build/C/man2/shmctl.2:118
7255 "Copy information from the kernel data structure associated with I<shmid> "
7256 "into the I<shmid_ds> structure pointed to by I<buf>. The caller must have "
7257 "read permission on the shared memory segment."
7261 #: build/C/man2/shmctl.2:118
7267 #: build/C/man2/shmctl.2:136
7269 "Write the values of some members of the I<shmid_ds> structure pointed to by "
7270 "I<buf> to the kernel data structure associated with this shared memory "
7271 "segment, updating also its I<shm_ctime> member. The following fields can be "
7272 "changed: I<shm_perm.uid>, I<shm_perm.gid>, and (the least significant 9 bits "
7273 "of) I<shm_perm.mode>. The effective UID of the calling process must match "
7274 "the owner (I<shm_perm.uid>) or creator (I<shm_perm.cuid>) of the shared "
7275 "memory segment, or the caller must be privileged."
7279 #: build/C/man2/shmctl.2:136
7285 #: build/C/man2/shmctl.2:153
7287 "Mark the segment to be destroyed. The segment will only actually be "
7288 "destroyed after the last process detaches it (i.e., when the I<shm_nattch> "
7289 "member of the associated structure I<shmid_ds> is zero). The caller must be "
7290 "the owner or creator, or be privileged. If a segment has been marked for "
7291 "destruction, then the (nonstandard) B<SHM_DEST> flag of the "
7292 "I<shm_perm.mode> field in the associated data structure retrieved by "
7293 "B<IPC_STAT> will be set."
7297 #: build/C/man2/shmctl.2:156
7299 "The caller I<must> ensure that a segment is eventually destroyed; otherwise "
7300 "its pages that were faulted in will remain in memory or swap."
7304 #: build/C/man2/shmctl.2:156
7306 msgid "B<IPC_INFO> (Linux-specific)"
7310 #: build/C/man2/shmctl.2:169
7312 "Returns information about system-wide shared memory limits and parameters in "
7313 "the structure pointed to by I<buf>. This structure is of type I<shminfo> "
7314 "(thus, a cast is required), defined in I<E<lt>sys/shm.hE<gt>> if the "
7315 "B<_GNU_SOURCE> feature test macro is defined:"
7319 #: build/C/man2/shmctl.2:183
7322 "struct shminfo {\n"
7323 " unsigned long shmmax; /* Maximum segment size */\n"
7324 " unsigned long shmmin; /* Minimum segment size;\n"
7326 " unsigned long shmmni; /* Maximum number of segments */\n"
7327 " unsigned long shmseg; /* Maximum number of segments\n"
7328 " that a process can attach;\n"
7329 " unused within kernel */\n"
7330 " unsigned long shmall; /* Maximum number of pages of\n"
7331 " shared memory, system-wide */\n"
7336 #: build/C/man2/shmctl.2:196
7338 "The I<shmmni>, I<shmmax>, and I<shmall> settings can be changed via I</proc> "
7339 "files of the same name; see B<proc>(5) for details."
7343 #: build/C/man2/shmctl.2:196
7345 msgid "B<SHM_INFO> (Linux-specific)"
7349 #: build/C/man2/shmctl.2:207
7351 "Returns a I<shm_info> structure whose fields contain information about "
7352 "system resources consumed by shared memory. This structure is defined in "
7353 "I<E<lt>sys/shm.hE<gt>> if the B<_GNU_SOURCE> feature test macro is defined:"
7357 #: build/C/man2/shmctl.2:224
7360 "struct shm_info {\n"
7361 " int used_ids; /* # of currently existing\n"
7363 " unsigned long shm_tot; /* Total number of shared\n"
7364 " memory pages */\n"
7365 " unsigned long shm_rss; /* # of resident shared\n"
7366 " memory pages */\n"
7367 " unsigned long shm_swp; /* # of swapped shared\n"
7368 " memory pages */\n"
7369 " unsigned long swap_attempts;\n"
7370 " /* Unused since Linux 2.4 */\n"
7371 " unsigned long swap_successes;\n"
7372 " /* Unused since Linux 2.4 */\n"
7377 #: build/C/man2/shmctl.2:226
7379 msgid "B<SHM_STAT> (Linux-specific)"
7383 #: build/C/man2/shmctl.2:237
7385 "Returns a I<shmid_ds> structure as for B<IPC_STAT>. However, the I<shmid> "
7386 "argument is not a segment identifier, but instead an index into the kernel's "
7387 "internal array that maintains information about all shared memory segments "
7392 #: build/C/man2/shmctl.2:240
7394 "The caller can prevent or allow swapping of a shared memory segment with the "
7395 "following I<cmd> values:"
7399 #: build/C/man2/shmctl.2:241
7401 msgid "B<SHM_LOCK> (Linux-specific)"
7405 #: build/C/man2/shmctl.2:253
7407 "Prevent swapping of the shared memory segment. The caller must fault in any "
7408 "pages that are required to be present after locking is enabled. If a "
7409 "segment has been locked, then the (nonstandard) B<SHM_LOCKED> flag of the "
7410 "I<shm_perm.mode> field in the associated data structure retrieved by "
7411 "B<IPC_STAT> will be set."
7415 #: build/C/man2/shmctl.2:253
7417 msgid "B<SHM_UNLOCK> (Linux-specific)"
7421 #: build/C/man2/shmctl.2:256
7422 msgid "Unlock the segment, allowing it to be swapped out."
7425 #. There was some weirdness in 2.6.9: SHM_LOCK and SHM_UNLOCK could
7426 #. be applied to a segment, regardless of ownership of the segment.
7427 #. This was a botch-up in the move to RLIMIT_MEMLOCK, and was fixed
7428 #. in 2.6.10. MTK, May 2005
7430 #: build/C/man2/shmctl.2:274
7432 "In kernels before 2.6.10, only a privileged process could employ B<SHM_LOCK> "
7433 "and B<SHM_UNLOCK>. Since kernel 2.6.10, an unprivileged process can employ "
7434 "these operations if its effective UID matches the owner or creator UID of "
7435 "the segment, and (for B<SHM_LOCK>) the amount of memory to be locked falls "
7436 "within the B<RLIMIT_MEMLOCK> resource limit (see B<setrlimit>(2))."
7440 #: build/C/man2/shmctl.2:292
7442 "A successful B<IPC_INFO> or B<SHM_INFO> operation returns the index of the "
7443 "highest used entry in the kernel's internal array recording information "
7444 "about all shared memory segments. (This information can be used with "
7445 "repeated B<SHM_STAT> operations to obtain information about all shared "
7446 "memory segments on the system.) A successful B<SHM_STAT> operation returns "
7447 "the identifier of the shared memory segment whose index was given in "
7448 "I<shmid>. Other operations return 0 on success."
7452 #: build/C/man2/shmctl.2:296
7453 msgid "On error, -1 is returned, and I<errno> is set appropriately."
7457 #: build/C/man2/shmctl.2:305
7459 "B<IPC_STAT> or B<SHM_STAT> is requested and I<shm_perm.mode> does not allow "
7460 "read access for I<shmid>, and the calling process does not have the "
7461 "B<CAP_IPC_OWNER> capability."
7465 #: build/C/man2/shmctl.2:316
7467 "The argument I<cmd> has value B<IPC_SET> or B<IPC_STAT> but the address "
7468 "pointed to by I<buf> isn't accessible."
7472 #: build/C/man2/shmctl.2:316 build/C/man2/shmop.2:197
7478 #: build/C/man2/shmctl.2:319 build/C/man2/shmop.2:200
7479 msgid "I<shmid> points to a removed identifier."
7483 #: build/C/man2/shmctl.2:328
7485 "I<shmid> is not a valid identifier, or I<cmd> is not a valid command. Or: "
7486 "for a B<SHM_STAT> operation, the index value specified in I<shmid> referred "
7487 "to an array slot that is currently unused."
7491 #: build/C/man2/shmctl.2:339
7493 "(In kernels since 2.6.9), B<SHM_LOCK> was specified and the size of the "
7494 "to-be-locked segment would mean that the total bytes in locked shared memory "
7495 "segments would exceed the limit for the real user ID of the calling "
7496 "process. This limit is defined by the B<RLIMIT_MEMLOCK> soft resource limit "
7497 "(see B<setrlimit>(2))."
7501 #: build/C/man2/shmctl.2:344
7503 "B<IPC_STAT> is attempted, and the GID or UID value is too large to be stored "
7504 "in the structure pointed to by I<buf>."
7508 #: build/C/man2/shmctl.2:356
7510 "B<IPC_SET> or B<IPC_RMID> is attempted, and the effective user ID of the "
7511 "calling process is not that of the creator (found in I<shm_perm.cuid>), or "
7512 "the owner (found in I<shm_perm.uid>), and the process was not privileged "
7513 "(Linux: did not have the B<CAP_SYS_ADMIN> capability)."
7517 #: build/C/man2/shmctl.2:368
7519 "Or (in kernels before 2.6.9), B<SHM_LOCK> or B<SHM_UNLOCK> was specified, "
7520 "but the process was not privileged (Linux: did not have the B<CAP_IPC_LOCK> "
7521 "capability). (Since Linux 2.6.9, this error can also occur if the "
7522 "B<RLIMIT_MEMLOCK> is 0 and the caller is not privileged.)"
7525 #. SVr4 documents an additional error condition EMFILE.
7527 #: build/C/man2/shmctl.2:373 build/C/man2/shmget.2:234 build/C/man2/shmop.2:234
7528 msgid "SVr4, POSIX.1-2001."
7531 #. Like Linux, the FreeBSD man pages still document
7532 #. the inclusion of these header files.
7534 #: build/C/man2/shmctl.2:386 build/C/man2/shmget.2:250
7536 "The inclusion of I<E<lt>sys/types.hE<gt>> and I<E<lt>sys/ipc.hE<gt>> isn't "
7537 "required on Linux or by any version of POSIX. However, some old "
7538 "implementations required the inclusion of these header files, and the SVID "
7539 "also documented their inclusion. Applications intended to be portable to "
7540 "such old systems may need to include these header files."
7544 #: build/C/man2/shmctl.2:397
7546 "The B<IPC_INFO>, B<SHM_STAT> and B<SHM_INFO> operations are used by the "
7547 "B<ipcs>(1) program to provide information on allocated resources. In the "
7548 "future these may modified or moved to a /proc file system interface."
7552 #: build/C/man2/shmctl.2:405
7554 "Linux permits a process to attach (B<shmat>(2)) a shared memory segment "
7555 "that has already been marked for deletion using I<shmctl(IPC_RMID)>. This "
7556 "feature is not available on other UNIX implementations; portable "
7557 "applications should avoid relying on it."
7561 #: build/C/man2/shmctl.2:418
7563 "Various fields in a I<struct shmid_ds> were typed as I<short> under Linux "
7564 "2.2 and have become I<long> under Linux 2.4. To take advantage of this, a "
7565 "recompilation under glibc-2.1.91 or later should suffice. (The kernel "
7566 "distinguishes old and new calls by an B<IPC_64> flag in I<cmd>.)"
7570 #: build/C/man2/shmctl.2:425
7572 "B<mlock>(2), B<setrlimit>(2), B<shmget>(2), B<shmop>(2), B<capabilities>(7), "
7577 #: build/C/man2/shmget.2:38
7583 #: build/C/man2/shmget.2:41
7584 msgid "shmget - allocates a shared memory segment"
7588 #: build/C/man2/shmget.2:48
7589 msgid "B<int shmget(key_t >I<key>B<, size_t >I<size>B<, int >I<shmflg>B<);>"
7593 #: build/C/man2/shmget.2:72
7595 "B<shmget>() returns the identifier of the shared memory segment associated "
7596 "with the value of the argument I<key>. A new shared memory segment, with "
7597 "size equal to the value of I<size> rounded up to a multiple of B<PAGE_SIZE>, "
7598 "is created if I<key> has the value B<IPC_PRIVATE> or I<key> isn't "
7599 "B<IPC_PRIVATE>, no shared memory segment corresponding to I<key> exists, and "
7600 "B<IPC_CREAT> is specified in I<shmflg>."
7604 #: build/C/man2/shmget.2:91
7606 "If I<shmflg> specifies both B<IPC_CREAT> and B<IPC_EXCL> and a shared memory "
7607 "segment already exists for I<key>, then B<shmget>() fails with I<errno> set "
7608 "to B<EEXIST>. (This is analogous to the effect of the combination B<O_CREAT "
7609 "| O_EXCL> for B<open>(2).)"
7613 #: build/C/man2/shmget.2:95
7614 msgid "The value I<shmflg> is composed of:"
7618 #: build/C/man2/shmget.2:95
7620 msgid "B<IPC_CREAT>"
7624 #: build/C/man2/shmget.2:102
7626 "to create a new segment. If this flag is not used, then B<shmget>() will "
7627 "find the segment associated with I<key> and check to see if the user has "
7628 "permission to access the segment."
7632 #: build/C/man2/shmget.2:102
7638 #: build/C/man2/shmget.2:105
7639 msgid "used with B<IPC_CREAT> to ensure failure if the segment already exists."
7643 #: build/C/man2/shmget.2:105
7645 msgid "I<mode_flags>"
7649 #: build/C/man2/shmget.2:115
7651 "(least significant 9 bits) specifying the permissions granted to the owner, "
7652 "group, and world. These bits have the same format, and the same meaning, as "
7653 "the I<mode> argument of B<open>(2). Presently, the execute permissions are "
7654 "not used by the system."
7658 #: build/C/man2/shmget.2:115
7660 msgid "B<SHM_HUGETLB> (since Linux 2.6)"
7664 #: build/C/man2/shmget.2:121
7666 "Allocate the segment using \"huge pages.\" See the Linux kernel source file "
7667 "I<Documentation/vm/hugetlbpage.txt> for further information."
7671 #: build/C/man2/shmget.2:121
7673 msgid "B<SHM_NORESERVE> (since Linux 2.6.15)"
7676 #. As at 2.6.17-rc2, this flag has no effect if SHM_HUGETLB was also
7679 #: build/C/man2/shmget.2:140
7681 "This flag serves the same purpose as the B<mmap>(2) B<MAP_NORESERVE> flag. "
7682 "Do not reserve swap space for this segment. When swap space is reserved, "
7683 "one has the guarantee that it is possible to modify the segment. When swap "
7684 "space is not reserved one might get B<SIGSEGV> upon a write if no physical "
7685 "memory is available. See also the discussion of the file "
7686 "I</proc/sys/vm/overcommit_memory> in B<proc>(5)."
7690 #: build/C/man2/shmget.2:148
7692 "When a new shared memory segment is created, its contents are initialized to "
7693 "zero values, and its associated data structure, I<shmid_ds> (see "
7694 "B<shmctl>(2)), is initialized as follows:"
7698 #: build/C/man2/shmget.2:153
7700 "I<shm_perm.cuid> and I<shm_perm.uid> are set to the effective user ID of the "
7705 #: build/C/man2/shmget.2:158
7707 "I<shm_perm.cgid> and I<shm_perm.gid> are set to the effective group ID of "
7708 "the calling process."
7712 #: build/C/man2/shmget.2:163
7714 "The least significant 9 bits of I<shm_perm.mode> are set to the least "
7715 "significant 9 bit of I<shmflg>."
7719 #: build/C/man2/shmget.2:167
7720 msgid "I<shm_segsz> is set to the value of I<size>."
7724 #: build/C/man2/shmget.2:174
7725 msgid "I<shm_lpid>, I<shm_nattch>, I<shm_atime> and I<shm_dtime> are set to 0."
7729 #: build/C/man2/shmget.2:177
7730 msgid "I<shm_ctime> is set to the current time."
7734 #: build/C/man2/shmget.2:180
7736 "If the shared memory segment already exists, the permissions are verified, "
7737 "and a check is made to see if it is marked for destruction."
7741 #: build/C/man2/shmget.2:184
7742 msgid "A valid segment identifier, I<shmid>, is returned on success, -1 on error."
7746 #: build/C/man2/shmget.2:188
7747 msgid "On failure, I<errno> is set to one of the following:"
7751 #: build/C/man2/shmget.2:194
7753 "The user does not have permission to access the shared memory segment, and "
7754 "does not have the B<CAP_IPC_OWNER> capability."
7758 #: build/C/man2/shmget.2:198
7759 msgid "B<IPC_CREAT | IPC_EXCL> was specified and the segment exists."
7763 #: build/C/man2/shmget.2:204
7765 "A new segment was to be created and I<size> E<lt> B<SHMMIN> or I<size> E<gt> "
7766 "B<SHMMAX>, or no new segment was to be created, a segment with given key "
7767 "existed, but I<size> is greater than the size of that segment."
7771 #: build/C/man2/shmget.2:213
7772 msgid "No segment exists for the given I<key>, and B<IPC_CREAT> was not specified."
7776 #: build/C/man2/shmget.2:216
7777 msgid "No memory could be allocated for segment overhead."
7781 #: build/C/man2/shmget.2:224
7783 "All possible shared memory IDs have been taken (B<SHMMNI>), or allocating a "
7784 "segment of the requested I<size> would cause the system to exceed the "
7785 "system-wide limit on shared memory (B<SHMALL>)."
7789 #: build/C/man2/shmget.2:231
7791 "The B<SHM_HUGETLB> flag was specified, but the caller was not privileged "
7792 "(did not have the B<CAP_IPC_LOCK> capability)."
7796 #: build/C/man2/shmget.2:237
7797 msgid "B<SHM_HUGETLB> is a nonportable Linux extension."
7801 #: build/C/man2/shmget.2:260
7803 "B<IPC_PRIVATE> isn't a flag field but a I<key_t> type. If this special "
7804 "value is used for I<key>, the system call ignores everything but the least "
7805 "significant 9 bits of I<shmflg> and creates a new shared memory segment (on "
7810 #: build/C/man2/shmget.2:264
7812 "The following limits on shared memory segment resources affect the "
7817 #: build/C/man2/shmget.2:264
7823 #: build/C/man2/shmget.2:269
7825 "System wide maximum of shared memory pages (on Linux, this limit can be read "
7826 "and modified via I</proc/sys/kernel/shmall>)."
7830 #: build/C/man2/shmget.2:269
7836 #: build/C/man2/shmget.2:274
7838 "Maximum size in bytes for a shared memory segment: policy dependent (on "
7839 "Linux, this limit can be read and modified via I</proc/sys/kernel/shmmax>)."
7843 #: build/C/man2/shmget.2:274
7849 #: build/C/man2/shmget.2:280
7851 "Minimum size in bytes for a shared memory segment: implementation dependent "
7852 "(currently 1 byte, though B<PAGE_SIZE> is the effective minimum size)."
7856 #: build/C/man2/shmget.2:280
7861 #. Kernels between 2.4.x and 2.6.8 had an off-by-one error that meant
7862 #. that we could create one more segment than SHMMNI -- MTK
7863 #. This /proc file is not available in Linux 2.2 and earlier -- MTK
7865 #: build/C/man2/shmget.2:289
7867 "System wide maximum number of shared memory segments: implementation "
7868 "dependent (currently 4096, was 128 before Linux 2.3.99; on Linux, this limit "
7869 "can be read and modified via I</proc/sys/kernel/shmmni>)."
7873 #: build/C/man2/shmget.2:293
7875 "The implementation has no specific limits for the per-process maximum number "
7876 "of shared memory segments (B<SHMSEG>)."
7880 #: build/C/man2/shmget.2:299
7882 "Until version 2.3.30 Linux would return B<EIDRM> for a B<shmget>() on a "
7883 "shared memory segment scheduled for deletion."
7887 #: build/C/man2/shmget.2:305
7889 "The name choice B<IPC_PRIVATE> was perhaps unfortunate, B<IPC_NEW> would "
7890 "more clearly show its function."
7894 #: build/C/man2/shmget.2:313
7896 "B<shmat>(2), B<shmctl>(2), B<shmdt>(2), B<ftok>(3), B<capabilities>(7), "
7897 "B<shm_overview>(7), B<svipc>(7)"
7901 #: build/C/man2/shmop.2:41
7907 #: build/C/man2/shmop.2:44
7908 msgid "shmat, shmdt - shared memory operations"
7912 #: build/C/man2/shmop.2:48
7915 "B<#include E<lt>sys/types.hE<gt>>\n"
7916 "B<#include E<lt>sys/shm.hE<gt>>\n"
7920 #: build/C/man2/shmop.2:50
7923 "B<void *shmat(int >I<shmid>B<, const void *>I<shmaddr>B<, int "
7928 #: build/C/man2/shmop.2:52
7930 msgid "B<int shmdt(const void *>I<shmaddr>B<);>\n"
7934 #: build/C/man2/shmop.2:61
7936 "B<shmat>() attaches the shared memory segment identified by I<shmid> to the "
7937 "address space of the calling process. The attaching address is specified by "
7938 "I<shmaddr> with one of the following criteria:"
7942 #: build/C/man2/shmop.2:67
7944 "If I<shmaddr> is NULL, the system chooses a suitable (unused) address at "
7945 "which to attach the segment."
7949 #: build/C/man2/shmop.2:82
7951 "If I<shmaddr> isn't NULL and B<SHM_RND> is specified in I<shmflg>, the "
7952 "attach occurs at the address equal to I<shmaddr> rounded down to the nearest "
7953 "multiple of B<SHMLBA>. Otherwise I<shmaddr> must be a page-aligned address "
7954 "at which the attach occurs."
7958 #: build/C/man2/shmop.2:92
7960 "If B<SHM_RDONLY> is specified in I<shmflg>, the segment is attached for "
7961 "reading and the process must have read permission for the segment. "
7962 "Otherwise the segment is attached for read and write and the process must "
7963 "have read and write permission for the segment. There is no notion of a "
7964 "write-only shared memory segment."
7968 #: build/C/man2/shmop.2:107
7970 "The (Linux-specific) B<SHM_REMAP> flag may be specified in I<shmflg> to "
7971 "indicate that the mapping of the segment should replace any existing mapping "
7972 "in the range starting at I<shmaddr> and continuing for the size of the "
7973 "segment. (Normally an B<EINVAL> error would result if a mapping already "
7974 "exists in this address range.) In this case, I<shmaddr> must not be NULL."
7978 #: build/C/man2/shmop.2:114
7980 "The B<brk>(2) value of the calling process is not altered by the attach. "
7981 "The segment will automatically be detached at process exit. The same "
7982 "segment may be attached as a read and as a read-write one, and more than "
7983 "once, in the process's address space."
7987 #: build/C/man2/shmop.2:122
7989 "A successful B<shmat>() call updates the members of the I<shmid_ds> "
7990 "structure (see B<shmctl>(2)) associated with the shared memory segment as "
7995 #: build/C/man2/shmop.2:125
7996 msgid "I<shm_atime> is set to the current time."
8000 #: build/C/man2/shmop.2:128 build/C/man2/shmop.2:154
8001 msgid "I<shm_lpid> is set to the process-ID of the calling process."
8005 #: build/C/man2/shmop.2:131
8006 msgid "I<shm_nattch> is incremented by one."
8010 #: build/C/man2/shmop.2:142
8012 "B<shmdt>() detaches the shared memory segment located at the address "
8013 "specified by I<shmaddr> from the address space of the calling process. The "
8014 "to-be-detached segment must be currently attached with I<shmaddr> equal to "
8015 "the value returned by the attaching B<shmat>() call."
8019 #: build/C/man2/shmop.2:148
8021 "On a successful B<shmdt>() call the system updates the members of the "
8022 "I<shmid_ds> structure associated with the shared memory segment as follows:"
8026 #: build/C/man2/shmop.2:151
8027 msgid "I<shm_dtime> is set to the current time."
8031 #: build/C/man2/shmop.2:159
8033 "I<shm_nattch> is decremented by one. If it becomes 0 and the segment is "
8034 "marked for deletion, the segment is deleted."
8038 #: build/C/man2/shmop.2:163
8039 msgid "After a B<fork>(2) the child inherits the attached shared memory segments."
8043 #: build/C/man2/shmop.2:167
8045 "After an B<execve>(2) all attached shared memory segments are detached from "
8050 #: build/C/man2/shmop.2:171
8052 "Upon B<_exit>(2) all attached shared memory segments are detached from the "
8057 #: build/C/man2/shmop.2:179
8059 "On success B<shmat>() returns the address of the attached shared memory "
8060 "segment; on error I<(void\\ *)\\ -1> is returned, and I<errno> is set to "
8061 "indicate the cause of the error."
8065 #: build/C/man2/shmop.2:185
8067 "On success B<shmdt>() returns 0; on error -1 is returned, and I<errno> is "
8068 "set to indicate the cause of the error."
8072 #: build/C/man2/shmop.2:191
8073 msgid "When B<shmat>() fails, I<errno> is set to one of the following:"
8077 #: build/C/man2/shmop.2:197
8079 "The calling process does not have the required permissions for the requested "
8080 "attach type, and does not have the B<CAP_IPC_OWNER> capability."
8084 #: build/C/man2/shmop.2:214
8086 "Invalid I<shmid> value, unaligned (i.e., not page-aligned and B<SHM_RND> was "
8087 "not specified) or invalid I<shmaddr> value, or can't attach segment at "
8088 "I<shmaddr>, or B<SHM_REMAP> was specified and I<shmaddr> was NULL."
8092 #: build/C/man2/shmop.2:217
8093 msgid "Could not allocate memory for the descriptor or for the page tables."
8097 #: build/C/man2/shmop.2:223
8098 msgid "When B<shmdt>() fails, I<errno> is set as follows:"
8101 #. The following since 2.6.17-rc1:
8103 #: build/C/man2/shmop.2:231
8105 "There is no shared memory segment attached at I<shmaddr>; or, I<shmaddr> is "
8106 "not aligned on a page boundary."
8110 #: build/C/man2/shmop.2:250
8112 "In SVID 3 (or perhaps earlier) the type of the I<shmaddr> argument was "
8113 "changed from I<char *> into I<const void *>, and the returned type of "
8114 "B<shmat>() from I<char *> into I<void *>. (Linux libc4 and libc5 have the "
8115 "I<char *> prototypes; glibc2 has I<void *>.)"
8119 #: build/C/man2/shmop.2:262
8121 "Using B<shmat>() with I<shmaddr> equal to NULL is the preferred, portable "
8122 "way of attaching a shared memory segment. Be aware that the shared memory "
8123 "segment attached in this way may be attached at different addresses in "
8124 "different processes. Therefore, any pointers maintained within the shared "
8125 "memory must be made relative (typically to the starting address of the "
8126 "segment), rather than absolute."
8130 #: build/C/man2/shmop.2:267
8132 "On Linux, it is possible to attach a shared memory segment even if it is "
8133 "already marked to be deleted. However, POSIX.1-2001 does not specify this "
8134 "behavior and many other implementations do not support it."
8138 #: build/C/man2/shmop.2:270
8139 msgid "The following system parameter affects B<shmat>():"
8142 #. FIXME A good explanation of the rationale for the existence
8143 #. of SHMLBA would be useful here
8145 #: build/C/man2/shmop.2:270
8150 #. FIXME That last sentence isn't true for all Linux
8151 #. architectures (i.e., SHMLBA != PAGE_SIZE for some architectures)
8154 #: build/C/man2/shmop.2:283
8156 "Segment low boundary address multiple. Must be page aligned. For the "
8157 "current implementation the B<SHMLBA> value is B<PAGE_SIZE>."
8161 #: build/C/man2/shmop.2:287
8163 "The implementation places no intrinsic limit on the per-process maximum "
8164 "number of shared memory segments (B<SHMSEG>)."
8168 #: build/C/man2/shmop.2:295
8170 "B<brk>(2), B<mmap>(2), B<shmctl>(2), B<shmget>(2), B<capabilities>(7), "
8171 "B<shm_overview>(7), B<svipc>(7)"
8175 #: build/C/man2/subpage_prot.2:30
8177 msgid "SUBPAGE_PROT"
8181 #: build/C/man2/subpage_prot.2:30
8187 #: build/C/man2/subpage_prot.2:33
8188 msgid "subpage_prot - define a subpage protection for an address range"
8192 #: build/C/man2/subpage_prot.2:37
8195 "B<long subpage_prot(unsigned long >I<addr>B<, unsigned long >I<len>B<,>\n"
8196 "B< uint32_t *>I<map>B<);>\n"
8200 #: build/C/man2/subpage_prot.2:41
8201 msgid "I<Note>: There is no glibc wrapper for this system call; see NOTES."
8205 #: build/C/man2/subpage_prot.2:47
8207 "The PowerPC-specific B<subpage_prot>() system call provides the facility to "
8208 "control the access permissions on individual 4kB subpages on systems "
8209 "configured with a page size of 64kB."
8213 #: build/C/man2/subpage_prot.2:54
8215 "The protection map is applied to the memory pages in the region starting at "
8216 "I<addr> and continuing for I<len> bytes. Both of these arguments must be "
8217 "aligned to a 64-kB boundary."
8221 #: build/C/man2/subpage_prot.2:66
8223 "The protection map is specified in the buffer pointed to by I<map>. The map "
8224 "has 2 bits per 4kB subpage; thus each 32-bit word specifies the protections "
8225 "of 16 4kB subpages inside a 64kB page (so, the number of 32-bit words "
8226 "pointed to by I<map> should equate to the number of 64-kB pages specified by "
8227 "I<len>). Each 2-bit field in the protection map is either 0 to allow any "
8228 "access, 1 to prevent writes, or 2 or 3 to prevent all accesses."
8232 #: build/C/man2/subpage_prot.2:71
8234 "On success, B<subpage_prot>() returns 0. Otherwise, one of the error codes "
8235 "specified below is returned."
8239 #: build/C/man2/subpage_prot.2:77
8240 msgid "The buffer referred to by I<map> is not accessible."
8244 #: build/C/man2/subpage_prot.2:87
8246 "The I<addr> or I<len> arguments are incorrect. Both of these arguments must "
8247 "be aligned to a multiple of the system page size, and they must not refer to "
8248 "a region outside of the address space of the process or to a region that "
8249 "consists of huge pages."
8253 #: build/C/man2/subpage_prot.2:90 build/C/man2/sync_file_range.2:166
8254 msgid "Out of memory."
8258 #: build/C/man2/subpage_prot.2:96
8260 "This system call is provided on the PowerPC architecture since Linux "
8261 "2.6.25. The system call is provided only if the kernel is configured with "
8262 "B<CONFIG_PPC_64K_PAGES>. No library support is provided."
8266 #: build/C/man2/subpage_prot.2:101
8268 "Glibc does not provide a wrapper for this system call; call it using "
8273 #: build/C/man2/subpage_prot.2:106
8275 "Normal page protections (at the 64-kB page level) also apply; the subpage "
8276 "protection mechanism is an additional constraint, so putting 0 in a 2-bit "
8277 "field won't allow writes to a page that is otherwise write-protected."
8281 #: build/C/man2/subpage_prot.2:106
8286 #. In the initial implementation, it was the case that:
8287 #. In fact the whole process is switched to use 4k hardware pages when the
8288 #. subpage_prot system call is used, but this could be improved in future
8289 #. to switch only the affected segments.
8290 #. But Paul Mackerass says (Oct 2010): I'm pretty sure we now only switch
8291 #. the affected segment, not the whole process.
8293 #: build/C/man2/subpage_prot.2:128
8295 "This system call is provided to assist writing emulators that operate using "
8296 "64-kB pages on PowerPC systems. When emulating systems such as x86, which "
8297 "uses a smaller page size, the emulator can no longer use the "
8298 "memory-management unit (MMU) and normal system calls for controlling page "
8299 "protections. (The emulator could emulate the MMU by checking and possibly "
8300 "remapping the address for each memory access in software, but that is slow.) "
8301 "The idea is that the emulator supplies an array of protection masks to apply "
8302 "to a specified range of virtual addresses. These masks are applied at the "
8303 "level where hardware page-table entries (PTEs) are inserted into the "
8304 "hardware page table based on the Linux PTEs, so the Linux PTEs are not "
8305 "affected. Implicit in this is that the regions of the address space that "
8306 "are protected are switched to use 4-kB hardware pages rather than 64-kB "
8307 "hardware pages (on machines with hardware 64-kB page support)."
8311 #: build/C/man2/subpage_prot.2:131
8312 msgid "B<mprotect>(2), B<syscall>(2)"
8316 #: build/C/man2/subpage_prot.2:134
8317 msgid "I<Documentation/vm/hugetlbpage.txt> in the Linux kernel source tree"
8321 #: build/C/man2/sync_file_range.2:30
8323 msgid "SYNC_FILE_RANGE"
8327 #: build/C/man2/sync_file_range.2:33
8328 msgid "sync_file_range - sync a file segment with disk"
8332 #: build/C/man2/sync_file_range.2:37
8335 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
8336 "B<#include E<lt>fcntl.hE<gt>>\n"
8340 #: build/C/man2/sync_file_range.2:40
8343 "B<int sync_file_range(int >I<fd>B<, off64_t >I<offset>B<, off64_t "
8345 "B< unsigned int >I<flags>B<);>\n"
8349 #: build/C/man2/sync_file_range.2:47
8351 "B<sync_file_range>() permits fine control when synchronizing the open file "
8352 "referred to by the file descriptor I<fd> with disk."
8356 #: build/C/man2/sync_file_range.2:61
8358 "I<offset> is the starting byte of the file range to be synchronized. "
8359 "I<nbytes> specifies the length of the range to be synchronized, in bytes; if "
8360 "I<nbytes> is zero, then all bytes from I<offset> through to the end of file "
8361 "are synchronized. Synchronization is in units of the system page size: "
8362 "I<offset> is rounded down to a page boundary; I<(offset+nbytes-1)> is "
8363 "rounded up to a page boundary."
8367 #: build/C/man2/sync_file_range.2:65
8368 msgid "The I<flags> bit-mask argument can include any of the following values:"
8372 #: build/C/man2/sync_file_range.2:65
8374 msgid "B<SYNC_FILE_RANGE_WAIT_BEFORE>"
8378 #: build/C/man2/sync_file_range.2:70
8380 "Wait upon write-out of all pages in the specified range that have already "
8381 "been submitted to the device driver for write-out before performing any "
8386 #: build/C/man2/sync_file_range.2:70 build/C/man2/sync_file_range.2:120
8388 msgid "B<SYNC_FILE_RANGE_WRITE>"
8392 #: build/C/man2/sync_file_range.2:76
8394 "Initiate write-out of all dirty pages in the specified range which are not "
8395 "presently submitted write-out. Note that even this may block if you attempt "
8396 "to write more than request queue size."
8400 #: build/C/man2/sync_file_range.2:76
8402 msgid "B<SYNC_FILE_RANGE_WAIT_AFTER>"
8406 #: build/C/man2/sync_file_range.2:80
8407 msgid "Wait upon write-out of all pages in the range after performing any write."
8411 #: build/C/man2/sync_file_range.2:84
8412 msgid "Specifying I<flags> as 0 is permitted, as a no-op."
8416 #: build/C/man2/sync_file_range.2:84
8422 #: build/C/man2/sync_file_range.2:100
8424 "This system call is extremely dangerous and should not be used in portable "
8425 "programs. None of these operations writes out the file's metadata. "
8426 "Therefore, unless the application is strictly performing overwrites of "
8427 "already-instantiated disk blocks, there are no guarantees that the data will "
8428 "be available after a crash. There is no user interface to know if a write "
8429 "is purely an overwrite. On file systems using copy-on-write semantics "
8430 "(e.g., I<btrfs>) an overwrite of existing allocated blocks is impossible. "
8431 "When writing into preallocated space, many file systems also require calls "
8432 "into the block allocator, which this system call does not sync out to disk. "
8433 "This system call does not flush disk write caches and thus does not provide "
8434 "any data integrity on systems with volatile disk write caches."
8438 #: build/C/man2/sync_file_range.2:100
8440 msgid "Some details"
8444 #: build/C/man2/sync_file_range.2:108
8446 "B<SYNC_FILE_RANGE_WAIT_BEFORE> and B<SYNC_FILE_RANGE_WAIT_AFTER> will detect "
8447 "any I/O errors or B<ENOSPC> conditions and will return these to the caller."
8451 #: build/C/man2/sync_file_range.2:112
8452 msgid "Useful combinations of the I<flags> bits are:"
8456 #: build/C/man2/sync_file_range.2:112
8458 msgid "B<SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE>"
8462 #: build/C/man2/sync_file_range.2:120
8464 "Ensures that all pages in the specified range which were dirty when "
8465 "B<sync_file_range>() was called are placed under write-out. This is a "
8466 "start-write-for-data-integrity operation."
8470 #: build/C/man2/sync_file_range.2:127
8472 "Start write-out of all dirty pages in the specified range which are not "
8473 "presently under write-out. This is an asynchronous flush-to-disk "
8474 "operation. This is not suitable for data integrity operations."
8478 #: build/C/man2/sync_file_range.2:127
8480 msgid "B<SYNC_FILE_RANGE_WAIT_BEFORE> (or B<SYNC_FILE_RANGE_WAIT_AFTER>)"
8484 #: build/C/man2/sync_file_range.2:134
8486 "Wait for completion of write-out of all pages in the specified range. This "
8487 "can be used after an earlier B<SYNC_FILE_RANGE_WAIT_BEFORE | "
8488 "SYNC_FILE_RANGE_WRITE> operation to wait for completion of that operation, "
8489 "and obtain its result."
8493 #: build/C/man2/sync_file_range.2:134
8496 "B<SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE | "
8497 "SYNC_FILE_RANGE_WAIT_AFTER>"
8501 #: build/C/man2/sync_file_range.2:141
8503 "This is a write-for-data-integrity operation that will ensure that all pages "
8504 "in the specified range which were dirty when B<sync_file_range>() was "
8505 "called are committed to disk."
8509 #: build/C/man2/sync_file_range.2:147
8511 "On success, B<sync_file_range>() returns 0; on failure -1 is returned and "
8512 "I<errno> is set to indicate the error."
8516 #: build/C/man2/sync_file_range.2:152
8517 msgid "I<fd> is not a valid file descriptor."
8521 #: build/C/man2/sync_file_range.2:160
8522 msgid "I<flags> specifies an invalid bit; or I<offset> or I<nbytes> is invalid."
8526 #: build/C/man2/sync_file_range.2:163
8531 #: build/C/man2/sync_file_range.2:169
8532 msgid "Out of disk space."
8535 #. FIXME . (bug?) Actually, how can 'fd' refer to a symbolic link (S_ISLNK)?
8536 #. (In user space at least) it isn't possible to obtain a file descriptor
8537 #. for a symbolic link.
8539 #: build/C/man2/sync_file_range.2:177
8541 "I<fd> refers to something other than a regular file, a block device, a "
8542 "directory, or a symbolic link."
8546 #: build/C/man2/sync_file_range.2:180
8547 msgid "B<sync_file_range>() appeared on Linux in kernel 2.6.17."
8551 #: build/C/man2/sync_file_range.2:183
8553 "This system call is Linux-specific, and should be avoided in portable "
8557 #. See kernel commit edd5cd4a9424f22b0fa08bef5e299d41befd5622
8559 #: build/C/man2/sync_file_range.2:196
8561 "Some architectures (e.g., PowerPC, ARM) need 64-bit arguments to be aligned "
8562 "in a suitable pair of registers. On such architectures, the call signature "
8563 "of B<sync_file_range>() is flawed, since it forces a register to be wasted "
8564 "as padding between the I<fd> and I<offset> arguments. Therefore, these "
8565 "architectures define a different system call that orders the arguments "
8570 #: build/C/man2/sync_file_range.2:201
8573 "B<int sync_file_range2(int >I<fd>B<, unsigned int >I<flags>B<,>\n"
8574 "B< off64_t >I<offset>B<, off64_t >I<nbytes>B<);>\n"
8578 #: build/C/man2/sync_file_range.2:206
8580 "The behavior of this system call is otherwise exactly the same as "
8581 "B<sync_file_range>()."
8585 #: build/C/man2/sync_file_range.2:217
8587 "A system call with this signature first appeared on the ARM architecture in "
8588 "Linux 2.6.20, with the name B<arm_sync_file_range>(). It was renamed in "
8589 "Linux 2.6.22, when the analogous system call was added for PowerPC. On "
8590 "architectures where glibc support is provided, glibc transparently wraps "
8591 "B<sync_file_range2>() under the name B<sync_file_range>()."
8595 #: build/C/man2/sync_file_range.2:222
8596 msgid "B<fdatasync>(2), B<fsync>(2), B<msync>(2), B<sync>(2)"