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: 2013-09-28 04:06+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.3:142 build/C/man2/readahead.2:65 build/C/man2/remap_file_pages.2:124 build/C/man3/shm_open.3:171 build/C/man2/shmctl.2:279 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.2:119 build/C/man2/madvise.2:275 build/C/man3/malloc_info.3:55 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.3:157 build/C/man2/readahead.2:71 build/C/man2/remap_file_pages.2:131 build/C/man3/shm_open.3:180 build/C/man2/shmctl.2:301 build/C/man2/shmget.2:185 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.2:158
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.2:199 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.3:63 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.3:182 build/C/man2/readahead.2:87 build/C/man2/remap_file_pages.2:153 build/C/man3/shm_open.3:251 build/C/man7/shm_overview.7:101 build/C/man2/shmctl.2:373 build/C/man2/shmget.2:232 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.3:65 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.3:221 build/C/man2/readahead.2:92 build/C/man3/shm_open.3:257 build/C/man7/shm_overview.7:103 build/C/man2/shmctl.2:378 build/C/man2/shmget.2:238 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 filesystem can be used instead. Memory backed by "
220 "huge pages (if the CPU supports them) is obtained by using B<mmap>(2) to "
221 "map files in this virtual filesystem."
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.3:157 build/C/man2/cacheflush.2:89 build/C/man2/fallocate.2:207 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.3:262 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.2:737 build/C/man2/mmap2.2:98 build/C/man2/mprotect.2:229 build/C/man2/mremap.2:214 build/C/man2/msync.2:124 build/C/man3/mtrace.3:172 build/C/man2/posix_fadvise.2:197 build/C/man3/posix_fallocate.3:133 build/C/man3/posix_memalign.3:277 build/C/man2/readahead.2:103 build/C/man2/remap_file_pages.2:164 build/C/man3/shm_open.3:282 build/C/man7/shm_overview.7:127 build/C/man2/shmctl.2:431 build/C/man2/shmget.2:314 build/C/man2/shmop.2:295 build/C/man2/subpage_prot.2:134 build/C/man2/sync_file_range.2:226
238 #: build/C/man2/alloc_hugepages.2:157 build/C/man3/alloca.3:164 build/C/man2/cacheflush.2:96 build/C/man2/fallocate.2:214 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.3:269 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.2:744 build/C/man2/mmap2.2:105 build/C/man2/mprotect.2:236 build/C/man2/mremap.2:221 build/C/man2/msync.2:131 build/C/man3/mtrace.3:179 build/C/man2/posix_fadvise.2:204 build/C/man3/posix_fallocate.3:140 build/C/man3/posix_memalign.3:284 build/C/man2/readahead.2:110 build/C/man2/remap_file_pages.2:171 build/C/man3/shm_open.3:289 build/C/man7/shm_overview.7:134 build/C/man2/shmctl.2:438 build/C/man2/shmget.2:321 build/C/man2/shmop.2:302 build/C/man2/subpage_prot.2:141 build/C/man2/sync_file_range.2:233
240 "This page is part of release 3.54 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<-std=c11> option is given B<and> the header I<E<lt>alloca.hE<gt>> is not "
342 "included. Otherwise (without an -ansi or -std=c* option) the glibc version "
343 "of I<E<lt>stdlib.hE<gt>> includes I<E<lt>alloca.hE<gt>> and that contains "
348 #: build/C/man3/alloca.3:129
352 " #define alloca(size) __builtin_alloca (size)\n"
357 #: build/C/man3/alloca.3:132
358 msgid "with messy consequences if one has a private version of this function."
362 #: build/C/man3/alloca.3:136
364 "The fact that the code is inlined means that it is impossible to take the "
365 "address of this function, or to change its behavior by linking with a "
370 #: build/C/man3/alloca.3:140
372 "The inlined code often consists of a single instruction adjusting the stack "
373 "pointer, and does not check for stack overflow. Thus, there is no NULL "
378 #: build/C/man3/alloca.3:140 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.2:186 build/C/man2/shmget.2:300
384 #: build/C/man3/alloca.3:145
386 "There is no error indication if the stack frame cannot be extended. "
387 "(However, after a failed allocation, the program is likely to receive a "
388 "B<SIGSEGV> signal if it attempts to access the unallocated space.)"
392 #: build/C/man3/alloca.3:153
394 "On many systems B<alloca>() cannot be used inside the list of arguments of "
395 "a function call, because the stack space reserved by B<alloca>() would "
396 "appear on the stack in the middle of the space for the function arguments."
400 #: build/C/man3/alloca.3:153 build/C/man2/fallocate.2:202 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.3:256 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.2:120 build/C/man3/mtrace.3:167 build/C/man2/posix_fadvise.2:191 build/C/man3/posix_fallocate.3:128 build/C/man3/posix_memalign.3:272 build/C/man2/readahead.2:97 build/C/man2/remap_file_pages.2:157 build/C/man3/shm_open.3:271 build/C/man7/shm_overview.7:114 build/C/man2/shmctl.2:424 build/C/man2/shmget.2:306 build/C/man2/shmop.2:287 build/C/man2/subpage_prot.2:128 build/C/man2/sync_file_range.2:221
406 #: build/C/man3/alloca.3:157
407 msgid "B<brk>(2), B<longjmp>(3), B<malloc>(3)"
411 #: build/C/man2/cacheflush.2:25
417 #: build/C/man2/cacheflush.2:25
423 #: build/C/man2/cacheflush.2:28
424 msgid "cacheflush - flush contents of instruction and/or data cache"
428 #: build/C/man2/cacheflush.2:31
430 msgid "B<#include E<lt>asm/cachectl.hE<gt>>\n"
434 #: build/C/man2/cacheflush.2:33
436 msgid "B<int cacheflush(char *>I<addr>B<, int >I<nbytes>B<, int >I<cache>B<);>\n"
440 #: build/C/man2/cacheflush.2:43
442 "B<cacheflush>() flushes the contents of the indicated cache(s) for the user "
443 "addresses in the range I<addr> to I<(addr+nbytes-1)>. I<cache> may be one "
448 #: build/C/man2/cacheflush.2:43
454 #: build/C/man2/cacheflush.2:46
455 msgid "Flush the instruction cache."
459 #: build/C/man2/cacheflush.2:46
465 #: build/C/man2/cacheflush.2:49
466 msgid "Write back to memory and invalidate the affected valid cache lines."
470 #: build/C/man2/cacheflush.2:49
476 #: build/C/man2/cacheflush.2:53
477 msgid "Same as B<(ICACHE|DCACHE)>."
481 #: build/C/man2/cacheflush.2:59
483 "B<cacheflush>() returns 0 on success or -1 on error. If errors are "
484 "detected, I<errno> will indicate the error."
488 #: build/C/man2/cacheflush.2:60 build/C/man2/mmap2.2:62 build/C/man2/mremap.2:142 build/C/man2/shmctl.2:310 build/C/man2/subpage_prot.2:72
494 #: build/C/man2/cacheflush.2:67
496 "Some or all of the address range I<addr> to I<(addr+nbytes-1)> is not "
501 #: build/C/man2/cacheflush.2:67 build/C/man2/fallocate.2:131 build/C/man2/madvise.2:282 build/C/man3/malloc_info.3:56 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.3:158 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.3:213 build/C/man2/shmctl.2:324 build/C/man2/shmget.2:199 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
507 #: build/C/man2/cacheflush.2:75
508 msgid "I<cache> is not one of B<ICACHE>, B<DCACHE>, or B<BCACHE>."
511 #. FIXME This system call was only on MIPS back in 1.2 days, but
512 #. by now it is on a number of other architectures (but not i386).
513 #. Investigate the details and update this page.
514 #. Irix 6.5 appears to have a cacheflush() syscall -- mtk
516 #: build/C/man2/cacheflush.2:82
518 "This Linux-specific system call is available only on MIPS-based systems. It "
519 "should not be used in programs intended to be portable."
523 #: build/C/man2/cacheflush.2:89
525 "The current implementation ignores the I<addr> and I<nbytes> arguments. "
526 "Therefore, the whole cache is always flushed."
530 #: build/C/man2/fallocate.2:11
536 #: build/C/man2/fallocate.2:11
542 #: build/C/man2/fallocate.2:14
543 msgid "fallocate - manipulate file space"
547 #: build/C/man2/fallocate.2:18 build/C/man2/readahead.2:35
550 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
551 "B<#include E<lt>fcntl.hE<gt>>\n"
555 #: build/C/man2/fallocate.2:21
558 "B<int fallocate(int >I<fd>B<, int >I<mode>B<, off_t >I<offset>B<, off_t "
563 #: build/C/man2/fallocate.2:27
565 "This is a nonportable, Linux-specific system call. For the portable, "
566 "POSIX.1-specified method of ensuring that space is allocated for a file, see "
567 "B<posix_fallocate>(3)."
571 #: build/C/man2/fallocate.2:37
573 "B<fallocate>() allows the caller to directly manipulate the allocated disk "
574 "space for the file referred to by I<fd> for the byte range starting at "
575 "I<offset> and continuing for I<len> bytes."
579 #: build/C/man2/fallocate.2:42
581 "The I<mode> argument determines the operation to be performed on the given "
582 "range. Details of the supported operations are given in the subsections "
587 #: build/C/man2/fallocate.2:42
589 msgid "Allocating disk space"
593 #: build/C/man2/fallocate.2:61
595 "The default operation (i.e., I<mode> is zero) of B<fallocate>() allocates "
596 "and initializes to zero the disk space within the range specified by "
597 "I<offset> and I<len>. The file size (as reported by B<stat>(2)) will be "
598 "changed if I<offset>+I<len> is greater than the file size. This default "
599 "behavior closely resembles the behavior of the B<posix_fallocate>(3) "
600 "library function, and is intended as a method of optimally implementing that "
605 #: build/C/man2/fallocate.2:67
607 "After a successful call, subsequent writes into the range specified by "
608 "I<offset> and I<len> are guaranteed not to fail because of lack of disk "
613 #: build/C/man2/fallocate.2:78
615 "If the B<FALLOC_FL_KEEP_SIZE> flag is specified in I<mode>, the behavior of "
616 "the call is similar, but the file size will not be changed even if "
617 "I<offset>+I<len> is greater than the file size. Preallocating zeroed blocks "
618 "beyond the end of the file in this manner is useful for optimizing append "
623 #: build/C/man2/fallocate.2:82
625 "Because allocation is done in block size chunks, B<fallocate>() may "
626 "allocate a larger range of disk space than was specified."
630 #: build/C/man2/fallocate.2:82
632 msgid "Deallocating file space"
636 #: build/C/man2/fallocate.2:97
638 "Specifying the B<FALLOC_FL_PUNCH_HOLE> flag (available since Linux 2.6.38) "
639 "in I<mode> deallocates space (i.e., creates a hole) in the byte range "
640 "starting at I<offset> and continuing for I<len> bytes. Within the specified "
641 "range, partial filesystem blocks are zeroed, and whole filesystem blocks are "
642 "removed from the file. After a successful call, subsequent reads from this "
643 "range will return zeroes."
647 #: build/C/man2/fallocate.2:108
649 "The B<FALLOC_FL_PUNCH_HOLE> flag must be ORed with B<FALLOC_FL_KEEP_SIZE> in "
650 "I<mode>; in other words, even when punching off the end of the file, the "
651 "file size (as reported by B<stat>(2)) does not change."
655 #: build/C/man2/fallocate.2:112
657 "Not all filesystems support B<FALLOC_FL_PUNCH_HOLE>; if a filesystem doesn't "
658 "support the operation, an error is returned."
662 #: build/C/man2/fallocate.2:119
664 "On success, B<fallocate>() returns zero. On error, -1 is returned and "
665 "I<errno> is set to indicate the error."
669 #: build/C/man2/fallocate.2:120 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
675 #: build/C/man2/fallocate.2:124 build/C/man3/posix_fallocate.3:77
676 msgid "I<fd> is not a valid file descriptor, or is not opened for writing."
680 #: build/C/man2/fallocate.2:124 build/C/man3/posix_fallocate.3:77
686 #: build/C/man2/fallocate.2:128
687 msgid "I<offset>+I<len> exceeds the maximum file size."
691 #: build/C/man2/fallocate.2:128
697 #: build/C/man2/fallocate.2:131
698 msgid "A signal was caught during execution."
702 #: build/C/man2/fallocate.2:143 build/C/man3/posix_fallocate.3:87
703 msgid "I<offset> was less than 0, or I<len> was less than or equal to 0."
707 #: build/C/man2/fallocate.2:143 build/C/man2/madvise.2:310 build/C/man2/sync_file_range.2:160
713 #: build/C/man2/fallocate.2:146
714 msgid "An I/O error occurred while reading from or writing to a filesystem."
718 #: build/C/man2/fallocate.2:146 build/C/man2/mmap.2:452 build/C/man3/posix_fallocate.3:87
724 #: build/C/man2/fallocate.2:153
726 "I<fd> does not refer to a regular file or a directory. (If I<fd> is a pipe "
727 "or FIFO, a different error results.)"
731 #: build/C/man2/fallocate.2:153 build/C/man3/posix_fallocate.3:91 build/C/man2/shmget.2:217 build/C/man2/sync_file_range.2:166
737 #: build/C/man2/fallocate.2:158 build/C/man3/posix_fallocate.3:96
739 "There is not enough space left on the device containing the file referred to "
744 #: build/C/man2/fallocate.2:162
745 msgid "This kernel does not implement B<fallocate>()."
749 #: build/C/man2/fallocate.2:162
751 msgid "B<EOPNOTSUPP>"
755 #: build/C/man2/fallocate.2:171
757 "The filesystem containing the file referred to by I<fd> does not support "
758 "this operation; or the I<mode> is not supported by the filesystem containing "
759 "the file referred to by I<fd>."
763 #: build/C/man2/fallocate.2:171 build/C/man2/mlock.2:143 build/C/man2/mlock.2:182 build/C/man2/mmap.2:460 build/C/man2/shmctl.2:349 build/C/man2/shmget.2:225
769 #: build/C/man2/fallocate.2:187
771 "The file referred to by I<fd> is marked immutable (see B<chattr>(1)). Or: "
772 "I<mode> specifies B<FALLOC_FL_PUNCH_HOLE> and the file referred to by I<fd> "
773 "is marked append-only (see B<chattr>(1))."
777 #: build/C/man2/fallocate.2:187 build/C/man2/posix_fadvise.2:97 build/C/man3/posix_fallocate.3:96 build/C/man2/sync_file_range.2:169
783 #: build/C/man2/fallocate.2:191
784 msgid "I<fd> refers to a pipe or FIFO."
788 #: build/C/man2/fallocate.2:191 build/C/man3/malloc_info.3:60 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.3:167 build/C/man2/readahead.2:82 build/C/man2/remap_file_pages.2:148 build/C/man3/shm_open.3:249 build/C/man2/subpage_prot.2:90 build/C/man2/sync_file_range.2:177
793 #. See http://sourceware.org/bugzilla/show_bug.cgi?id=14964
795 #: build/C/man2/fallocate.2:199
797 "B<fallocate>() is available on Linux since kernel 2.6.23. Support is "
798 "provided by glibc since version 2.10. The B<FALLOC_FL_*> flags are defined "
799 "in glibc headers only since version 2.18."
803 #: build/C/man2/fallocate.2:202
804 msgid "B<fallocate>() is Linux-specific."
808 #: build/C/man2/fallocate.2:207
809 msgid "B<fallocate>(1), B<ftruncate>(2), B<posix_fadvise>(3), B<posix_fallocate>(3)"
813 #: build/C/man2/madvise.2:35
819 #: build/C/man2/madvise.2:35
825 #: build/C/man2/madvise.2:38
826 msgid "madvise - give advice about use of memory"
830 #: build/C/man2/madvise.2:40 build/C/man2/msync.2:30 build/C/man3/shm_open.3:31
831 msgid "B<#include E<lt>sys/mman.hE<gt>>"
835 #: build/C/man2/madvise.2:42
836 msgid "B<int madvise(void *>I<addr>B<, size_t >I<length>B<, int >I<advice>B<);>"
840 #: 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
841 msgid "Feature Test Macro Requirements for glibc (see B<feature_test_macros>(7)):"
845 #: build/C/man2/madvise.2:50
846 msgid "B<madvise>(): _BSD_SOURCE"
850 #: build/C/man2/madvise.2:68
852 "The B<madvise>() system call advises the kernel about how to handle paging "
853 "input/output in the address range beginning at address I<addr> and with size "
854 "I<length> bytes. It allows an application to tell the kernel how it expects "
855 "to use some mapped or shared memory areas, so that the kernel can choose "
856 "appropriate read-ahead and caching techniques. This call does not influence "
857 "the semantics of the application (except in the case of B<MADV_DONTNEED>), "
858 "but may influence its performance. The kernel is free to ignore the advice."
862 #: build/C/man2/madvise.2:72
863 msgid "The advice is indicated in the I<advice> argument which can be"
867 #: build/C/man2/madvise.2:72
869 msgid "B<MADV_NORMAL>"
873 #: build/C/man2/madvise.2:76
874 msgid "No special treatment. This is the default."
878 #: build/C/man2/madvise.2:76
880 msgid "B<MADV_RANDOM>"
884 #: build/C/man2/madvise.2:80
886 "Expect page references in random order. (Hence, read ahead may be less "
887 "useful than normally.)"
891 #: build/C/man2/madvise.2:80
893 msgid "B<MADV_SEQUENTIAL>"
897 #: build/C/man2/madvise.2:85
899 "Expect page references in sequential order. (Hence, pages in the given "
900 "range can be aggressively read ahead, and may be freed soon after they are "
905 #: build/C/man2/madvise.2:85
907 msgid "B<MADV_WILLNEED>"
911 #: build/C/man2/madvise.2:89
913 "Expect access in the near future. (Hence, it might be a good idea to read "
918 #: build/C/man2/madvise.2:89
920 msgid "B<MADV_DONTNEED>"
924 #: build/C/man2/madvise.2:100
926 "Do not expect access in the near future. (For the time being, the "
927 "application is finished with the given range, so the kernel can free "
928 "resources associated with it.) Subsequent accesses of pages in this range "
929 "will succeed, but will result either in reloading of the memory contents "
930 "from the underlying mapped file (see B<mmap>(2)) or zero-fill-on-demand "
931 "pages for mappings without an underlying file."
935 #: build/C/man2/madvise.2:100
937 msgid "B<MADV_REMOVE> (Since Linux 2.6.16)"
941 #. Databases want to use this feature to drop a section of their
942 #. bufferpool (shared memory segments) - without writing back to
943 #. disk/swap space. This feature is also useful for supporting
944 #. hot-plug memory on UML.
946 #: build/C/man2/madvise.2:113
948 "Free up a given range of pages and its associated backing store. Currently, "
949 "only shmfs/tmpfs supports this; other filesystems return with the error "
954 #: build/C/man2/madvise.2:113
956 msgid "B<MADV_DONTFORK> (Since Linux 2.6.16)"
959 #. See http://lwn.net/Articles/171941/
960 #. [PATCH] madvise MADV_DONTFORK/MADV_DOFORK
961 #. Currently, copy-on-write may change the physical address of
962 #. a page even if the user requested that the page is pinned in
963 #. memory (either by mlock or by get_user_pages). This happens
964 #. if the process forks meanwhile, and the parent writes to that
965 #. page. As a result, the page is orphaned: in case of
966 #. get_user_pages, the application will never see any data hardware
967 #. DMA's into this page after the COW. In case of mlock'd memory,
968 #. the parent is not getting the realtime/security benefits of mlock.
970 #. In particular, this affects the Infiniband modules which do DMA from
971 #. and into user pages all the time.
973 #. This patch adds madvise options to control whether memory range is
974 #. inherited across fork. Useful e.g. for when hardware is doing DMA
975 #. from/into these pages. Could also be useful to an application
976 #. wanting to speed up its forks by cutting large areas out of
979 #. SEE ALSO: http://lwn.net/Articles/171941/
980 #. "Tweaks to madvise() and posix_fadvise()", 14 Feb 2006
982 #: build/C/man2/madvise.2:144
984 "Do not make the pages in this range available to the child after a "
985 "B<fork>(2). This is useful to prevent copy-on-write semantics from changing "
986 "the physical location of a page(s) if the parent writes to it after a "
987 "B<fork>(2). (Such page relocations cause problems for hardware that DMAs "
992 #: build/C/man2/madvise.2:144
994 msgid "B<MADV_DOFORK> (Since Linux 2.6.16)"
998 #: build/C/man2/madvise.2:150
1000 "Undo the effect of B<MADV_DONTFORK>, restoring the default behavior, whereby "
1001 "a mapping is inherited across B<fork>(2)."
1005 #: build/C/man2/madvise.2:150
1007 msgid "B<MADV_HWPOISON> (Since Linux 2.6.32)"
1011 #: build/C/man2/madvise.2:162
1013 "Poison a page and handle it like a hardware memory corruption. This "
1014 "operation is available only for privileged (B<CAP_SYS_ADMIN>) processes. "
1015 "This operation may result in the calling process receiving a B<SIGBUS> and "
1016 "the page being unmapped. This feature is intended for testing of memory "
1017 "error-handling code; it is available only if the kernel was configured with "
1018 "B<CONFIG_MEMORY_FAILURE>."
1022 #: build/C/man2/madvise.2:162
1024 msgid "B<MADV_SOFT_OFFLINE> (Since Linux 2.6.33)"
1028 #: build/C/man2/madvise.2:180
1030 "Soft offline the pages in the range specified by I<addr> and I<length>. The "
1031 "memory of each page in the specified range is preserved (i.e., when next "
1032 "accessed, the same content will be visible, but in a new physical page "
1033 "frame), and the original page is offlined (i.e., no longer used, and taken "
1034 "out of normal memory management). The effect of the B<MADV_SOFT_OFFLINE> "
1035 "operation is invisible to (i.e., does not change the semantics of) the "
1036 "calling process. This feature is intended for testing of memory "
1037 "error-handling code; it is available only if the kernel was configured with "
1038 "B<CONFIG_MEMORY_FAILURE>."
1042 #: build/C/man2/madvise.2:180
1044 msgid "B<MADV_MERGEABLE> (since Linux 2.6.32)"
1048 #: build/C/man2/madvise.2:205
1050 "Enable Kernel Samepage Merging (KSM) for the pages in the range specified by "
1051 "I<addr> and I<length>. The kernel regularly scans those areas of user "
1052 "memory that have been marked as mergeable, looking for pages with identical "
1053 "content. These are replaced by a single write-protected page (which is "
1054 "automatically copied if a process later wants to update the content of the "
1055 "page). KSM merges only private anonymous pages (see B<mmap>(2)). The KSM "
1056 "feature is intended for applications that generate many instances of the "
1057 "same data (e.g., virtualization systems such as KVM). It can consume a lot "
1058 "of processing power; use with care. See the Linux kernel source file "
1059 "I<Documentation/vm/ksm.txt> for more details. The B<MADV_MERGEABLE> and "
1060 "B<MADV_UNMERGEABLE> operations are available only if the kernel was "
1061 "configured with B<CONFIG_KSM>."
1065 #: build/C/man2/madvise.2:205
1067 msgid "B<MADV_UNMERGEABLE> (since Linux 2.6.32)"
1071 #: build/C/man2/madvise.2:214
1073 "Undo the effect of an earlier B<MADV_MERGEABLE> operation on the specified "
1074 "address range; KSM unmerges whatever pages it had merged in the address "
1075 "range specified by I<addr> and I<length>."
1079 #: build/C/man2/madvise.2:214
1081 msgid "B<MADV_HUGEPAGE> (since Linux 2.6.38)"
1084 #. http://lwn.net/Articles/358904/
1085 #. https://lwn.net/Articles/423584/
1087 #: build/C/man2/madvise.2:243
1089 "Enables Transparent Huge Pages (THP) for pages in the range specified by "
1090 "I<addr> and I<length>. Currently, Transparent Huge Pages work only with "
1091 "private anonymous pages (see B<mmap>(2)). The kernel will regularly scan "
1092 "the areas marked as huge page candidates to replace them with huge pages. "
1093 "The kernel will also allocate huge pages directly when the region is "
1094 "naturally aligned to the huge page size (see B<posix_memalign>(2)). This "
1095 "feature is primarily aimed at applications that use large mappings of data "
1096 "and access large regions of that memory at a time (e.g., virtualization "
1097 "systems such as QEMU). It can very easily waste memory (e.g., a 2MB mapping "
1098 "that only ever accesses 1 byte will result in 2MB of wired memory instead of "
1099 "one 4KB page). See the Linux kernel source file "
1100 "I<Documentation/vm/transhuge.txt> for more details. The B<MADV_HUGEPAGE> "
1101 "and B<MADV_NOHUGEPAGE> operations are available only if the kernel was "
1102 "configured with B<CONFIG_TRANSPARENT_HUGEPAGE>."
1106 #: build/C/man2/madvise.2:243
1108 msgid "B<MADV_NOHUGEPAGE> (since Linux 2.6.38)"
1112 #: build/C/man2/madvise.2:250
1114 "Ensures that memory in the address range specified by I<addr> and I<length> "
1115 "will not be collapsed into huge pages."
1119 #: build/C/man2/madvise.2:250
1121 msgid "B<MADV_DONTDUMP> (since Linux 3.4)"
1125 #: build/C/man2/madvise.2:264
1127 "Exclude from a core dump those pages in the range specified by I<addr> and "
1128 "I<length>. This is useful in applications that have large areas of memory "
1129 "that are known not to be useful in a core dump. The effect of "
1130 "B<MADV_DONTDUMP> takes precedence over the bit mask that is set via the "
1131 "I</proc/PID/coredump_filter> file (see B<core>(5))."
1135 #: build/C/man2/madvise.2:264
1137 msgid "B<MADV_DODUMP> (since Linux 3.4)"
1141 #: build/C/man2/madvise.2:268
1142 msgid "Undo the effect of an earlier B<MADV_DONTDUMP>."
1146 #: build/C/man2/madvise.2:275
1148 "On success B<madvise>() returns zero. On error, it returns -1 and I<errno> "
1149 "is set appropriately."
1153 #: build/C/man2/madvise.2:276 build/C/man2/mlock.2:154 build/C/man2/mmap.2:413 build/C/man2/mremap.2:136
1159 #: build/C/man2/madvise.2:279
1160 msgid "A kernel resource was temporarily unavailable."
1164 #: build/C/man2/madvise.2:282
1165 msgid "The map exists, but the area maps something that isn't a file."
1169 #: build/C/man2/madvise.2:285
1170 msgid "This error can occur for the following reasons:"
1174 #: 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
1182 #: build/C/man2/madvise.2:292
1183 msgid "The value I<len> is negative."
1187 #: build/C/man2/madvise.2:295
1188 msgid "I<addr> is not page-aligned."
1192 #: build/C/man2/madvise.2:298
1193 msgid "I<advice> is not a valid value"
1197 #: build/C/man2/madvise.2:301
1199 "The application is attempting to release locked or shared pages (with "
1200 "B<MADV_DONTNEED>)."
1204 #: build/C/man2/madvise.2:309
1206 "B<MADV_MERGEABLE> or B<MADV_UNMERGEABLE> was specified in I<advice>, but the "
1207 "kernel was not configured with B<CONFIG_KSM>."
1211 #: build/C/man2/madvise.2:316
1213 "(for B<MADV_WILLNEED>) Paging in this area would exceed the process's "
1214 "maximum resident set size."
1218 #: 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.3:164 build/C/man2/shmctl.2:333 build/C/man2/shmget.2:214 build/C/man2/shmop.2:214 build/C/man2/subpage_prot.2:87 build/C/man2/sync_file_range.2:163
1224 #: build/C/man2/madvise.2:321
1225 msgid "(for B<MADV_WILLNEED>) Not enough memory: paging in failed."
1229 #: build/C/man2/madvise.2:325
1231 "Addresses in the specified range are not currently mapped, or are outside "
1232 "the address space of the process."
1235 #. FIXME . Write a posix_madvise(3) page.
1237 #: build/C/man2/madvise.2:337
1239 "POSIX.1b. POSIX.1-2001 describes B<posix_madvise>(3) with constants "
1240 "B<POSIX_MADV_NORMAL>, etc., with a behavior close to that described here. "
1241 "There is a similar B<posix_fadvise>(2) for file access."
1245 #: build/C/man2/madvise.2:346
1247 "B<MADV_REMOVE>, B<MADV_DONTFORK>, B<MADV_DOFORK>, B<MADV_HWPOISON>, "
1248 "B<MADV_MERGEABLE>, and B<MADV_UNMERGEABLE> are Linux-specific."
1252 #: build/C/man2/madvise.2:347 build/C/man2/mlock.2:267 build/C/man2/shmget.2:294
1258 #: build/C/man2/madvise.2:355
1260 "The current Linux implementation (2.4.0) views this system call more as a "
1261 "command than as advice and hence may return an error when it cannot do what "
1262 "it usually would do in response to this advice. (See the ERRORS description "
1263 "above.) This is nonstandard behavior."
1269 #. function first appeared in 4.4BSD.
1271 #: build/C/man2/madvise.2:371
1273 "The Linux implementation requires that the address I<addr> be page-aligned, "
1274 "and allows I<length> to be zero. If there are some parts of the specified "
1275 "address range that are not mapped, the Linux version of B<madvise>() "
1276 "ignores them and applies the call to the rest (but returns B<ENOMEM> from "
1277 "the system call, as it should)."
1281 #: build/C/man2/madvise.2:379
1283 "B<getrlimit>(2), B<mincore>(2), B<mmap>(2), B<mprotect>(2), B<msync>(2), "
1284 "B<munmap>(2), B<core>(5)"
1288 #: build/C/man3/mallinfo.3:26
1294 #: build/C/man3/mallinfo.3:26 build/C/man3/malloc_stats.3:26
1300 #: build/C/man3/mallinfo.3:29
1301 msgid "mallinfo - obtain memory allocation information"
1305 #: 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
1306 msgid "B<#include E<lt>malloc.hE<gt>>"
1310 #: build/C/man3/mallinfo.3:33
1311 msgid "B<struct mallinfo mallinfo(void);>"
1315 #: build/C/man3/mallinfo.3:41
1317 "The B<mallinfo>() function returns a copy of a structure containing "
1318 "information about memory allocations performed by B<malloc>(3) and related "
1319 "functions. This structure is defined as follows:"
1323 #: build/C/man3/mallinfo.3:56
1326 "struct mallinfo {\n"
1327 " int arena; /* Non-mmapped space allocated (bytes) */\n"
1328 " int ordblks; /* Number of free chunks */\n"
1329 " int smblks; /* Number of free fastbin blocks */\n"
1330 " int hblks; /* Number of mmapped regions */\n"
1331 " int hblkhd; /* Space allocated in mmapped regions (bytes) */\n"
1332 " int usmblks; /* Maximum total allocated space (bytes) */\n"
1333 " int fsmblks; /* Space in freed fastbin blocks (bytes) */\n"
1334 " int uordblks; /* Total allocated space (bytes) */\n"
1335 " int fordblks; /* Total free space (bytes) */\n"
1336 " int keepcost; /* Top-most, releasable space (bytes) */\n"
1341 #: build/C/man3/mallinfo.3:62
1342 msgid "The fields of the I<mallinfo> structure contain the following information:"
1346 #: build/C/man3/mallinfo.3:62
1352 #: build/C/man3/mallinfo.3:68
1354 "The total amount of memory allocated by means other than B<mmap>(2) (i.e., "
1355 "memory allocated on the heap). This figure includes both in-use blocks and "
1356 "blocks on the free list."
1360 #: build/C/man3/mallinfo.3:68
1366 #: build/C/man3/mallinfo.3:71
1367 msgid "The number of ordinary (i.e., non-fastbin) free blocks."
1371 #: build/C/man3/mallinfo.3:71
1377 #: build/C/man3/mallinfo.3:75
1378 msgid "The number of fastbin free blocks (see B<mallopt>(3))."
1382 #: build/C/man3/mallinfo.3:75
1388 #: build/C/man3/mallinfo.3:83
1390 "The number of blocks currently allocated using B<mmap>(2). (See the "
1391 "discussion of B<M_MMAP_THRESHOLD> in B<mallopt>(3).)"
1395 #: build/C/man3/mallinfo.3:83
1401 #: build/C/man3/mallinfo.3:87
1402 msgid "The number of bytes in blocks currently allocated using B<mmap>(2)."
1406 #: build/C/man3/mallinfo.3:87
1412 #: build/C/man3/mallinfo.3:92
1414 "The \"highwater mark\" for allocated space\\(emthat is, the maximum amount "
1415 "of space that was ever allocated. This field is maintained only in "
1416 "nonthreading environments."
1420 #: build/C/man3/mallinfo.3:92
1426 #: build/C/man3/mallinfo.3:95
1427 msgid "The total number of bytes in fastbin free blocks."
1431 #: build/C/man3/mallinfo.3:95
1437 #: build/C/man3/mallinfo.3:98
1438 msgid "The total number of bytes used by in-use allocations."
1442 #: build/C/man3/mallinfo.3:98
1448 #: build/C/man3/mallinfo.3:101
1449 msgid "The total number of bytes in free blocks."
1453 #: build/C/man3/mallinfo.3:101
1459 #. Available already in glibc 2.0, possibly earlier
1461 #: build/C/man3/mallinfo.3:110
1463 "The total amount of releasable free space at the top of the heap. This is "
1464 "the maximum number of bytes that could ideally (i.e., ignoring page "
1465 "alignment restrictions, and so on) be released by B<malloc_trim>(3)."
1469 #: build/C/man3/mallinfo.3:114
1471 "This function is not specified by POSIX or the C standards. A similar "
1472 "function exists on many System V derivatives, and was specified in the SVID."
1475 #. FIXME http://sourceware.org/bugzilla/show_bug.cgi?id=208
1476 #. See the 24 Aug 2011 mail by Paul Pluzhnikov:
1477 #. "[patch] Fix mallinfo() to accumulate results for all arenas"
1478 #. on libc-alpha@sourceware.org
1480 #: build/C/man3/mallinfo.3:126
1482 "B<Information is returned for only the main memory allocation area.> "
1483 "Allocations in other arenas are excluded. See B<malloc_stats>(3) and "
1484 "B<malloc_info>(3) for alternatives that include information about other "
1489 #: build/C/man3/mallinfo.3:134
1491 "The fields of the I<mallinfo> structure are typed as I<int>. However, "
1492 "because some internal bookkeeping values may be of type I<long>, the "
1493 "reported values may wrap around zero and thus be inaccurate."
1497 #: build/C/man3/mallinfo.3:134 build/C/man3/malloc_hook.3:89 build/C/man3/malloc_info.3:84 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
1503 #: build/C/man3/mallinfo.3:140
1505 "The program below employs B<mallinfo>() to retrieve memory allocation "
1506 "statistics before and after allocating and freeing some blocks of memory. "
1507 "The statistics are displayed on standard output."
1511 #: build/C/man3/mallinfo.3:144
1513 "The first two command-line arguments specify the number and size of blocks "
1514 "to be allocated with B<malloc>(3)."
1518 #: build/C/man3/mallinfo.3:158
1520 "The remaining three arguments specify which of the allocated blocks should "
1521 "be freed with B<free>(3). These three arguments are optional, and specify "
1522 "(in order): the step size to be used in the loop that frees blocks (the "
1523 "default is 1, meaning free all blocks in the range); the ordinal position of "
1524 "the first block to be freed (default 0, meaning the first allocated block); "
1525 "and a number one greater than the ordinal position of the last block to be "
1526 "freed (default is one greater than the maximum block number). If these "
1527 "three arguments are omitted, then the defaults cause all allocated blocks to "
1532 #: build/C/man3/mallinfo.3:162
1534 "In the following example run of the program, 1000 allocations of 100 bytes "
1535 "are performed, and then every second allocated block is freed:"
1539 #: build/C/man3/mallinfo.3:177
1542 "$ B<./a.out 1000 100 2>\n"
1543 "============== Before allocating blocks ==============\n"
1544 "Total non-mmapped bytes (arena): 0\n"
1545 "# of free chunks (ordblks): 1\n"
1546 "# of free fastbin blocks (smblks): 0\n"
1547 "# of mapped regions (hblks): 0\n"
1548 "Bytes in mapped regions (hblkhd): 0\n"
1549 "Max. total allocated space (usmblks): 0\n"
1550 "Free bytes held in fastbins (fsmblks): 0\n"
1551 "Total allocated space (uordblks): 0\n"
1552 "Total free space (fordblks): 0\n"
1553 "Topmost releasable block (keepcost): 0\n"
1557 #: build/C/man3/mallinfo.3:189
1560 "============== After allocating blocks ==============\n"
1561 "Total non-mmapped bytes (arena): 135168\n"
1562 "# of free chunks (ordblks): 1\n"
1563 "# of free fastbin blocks (smblks): 0\n"
1564 "# of mapped regions (hblks): 0\n"
1565 "Bytes in mapped regions (hblkhd): 0\n"
1566 "Max. total allocated space (usmblks): 0\n"
1567 "Free bytes held in fastbins (fsmblks): 0\n"
1568 "Total allocated space (uordblks): 104000\n"
1569 "Total free space (fordblks): 31168\n"
1570 "Topmost releasable block (keepcost): 31168\n"
1574 #: build/C/man3/mallinfo.3:201
1577 "============== After freeing blocks ==============\n"
1578 "Total non-mmapped bytes (arena): 135168\n"
1579 "# of free chunks (ordblks): 501\n"
1580 "# of free fastbin blocks (smblks): 0\n"
1581 "# of mapped regions (hblks): 0\n"
1582 "Bytes in mapped regions (hblkhd): 0\n"
1583 "Max. total allocated space (usmblks): 0\n"
1584 "Free bytes held in fastbins (fsmblks): 0\n"
1585 "Total allocated space (uordblks): 52000\n"
1586 "Total free space (fordblks): 83168\n"
1587 "Topmost releasable block (keepcost): 31168\n"
1591 #: build/C/man3/mallinfo.3:203 build/C/man3/malloc_info.3:167 build/C/man3/mallopt.3:532 build/C/man3/mcheck.3:176 build/C/man2/mprotect.2:165
1593 msgid "Program source"
1597 #: build/C/man3/mallinfo.3:208
1600 "#include E<lt>malloc.hE<gt>\n"
1601 "#include \"tlpi_hdr.h\"\n"
1605 #: build/C/man3/mallinfo.3:213
1609 "display_mallinfo(void)\n"
1611 " struct mallinfo mi;\n"
1615 #: build/C/man3/mallinfo.3:215
1617 msgid " mi = mallinfo();\n"
1621 #: build/C/man3/mallinfo.3:227
1624 " printf(\"Total non-mmapped bytes (arena): %d\\en\", mi.arena);\n"
1625 " printf(\"# of free chunks (ordblks): %d\\en\", mi.ordblks);\n"
1626 " printf(\"# of free fastbin blocks (smblks): %d\\en\", mi.smblks);\n"
1627 " printf(\"# of mapped regions (hblks): %d\\en\", mi.hblks);\n"
1628 " printf(\"Bytes in mapped regions (hblkhd): %d\\en\", mi.hblkhd);\n"
1629 " printf(\"Max. total allocated space (usmblks): %d\\en\", mi.usmblks);\n"
1630 " printf(\"Free bytes held in fastbins (fsmblks): %d\\en\", mi.fsmblks);\n"
1631 " printf(\"Total allocated space (uordblks): %d\\en\", "
1633 " printf(\"Total free space (fordblks): %d\\en\", "
1635 " printf(\"Topmost releasable block (keepcost): %d\\en\", "
1641 #: build/C/man3/mallinfo.3:235
1645 "main(int argc, char *argv[])\n"
1647 "#define MAX_ALLOCS 2000000\n"
1648 " char *alloc[MAX_ALLOCS];\n"
1649 " int numBlocks, j, freeBegin, freeEnd, freeStep;\n"
1650 " size_t blockSize;\n"
1654 #: build/C/man3/mallinfo.3:239
1657 " if (argc E<lt> 3 || strcmp(argv[1], \"--help\") == 0)\n"
1658 " usageErr(\"%s num-blocks block-size [free-step [start-free \"\n"
1659 " \"[end-free]]]\\en\", argv[0]);\n"
1663 #: build/C/man3/mallinfo.3:245
1666 " numBlocks = atoi(argv[1]);\n"
1667 " blockSize = atoi(argv[2]);\n"
1668 " freeStep = (argc E<gt> 3) ? atoi(argv[3]) : 1;\n"
1669 " freeBegin = (argc E<gt> 4) ? atoi(argv[4]) : 0;\n"
1670 " freeEnd = (argc E<gt> 5) ? atoi(argv[5]) : numBlocks;\n"
1674 #: build/C/man3/mallinfo.3:248
1677 " printf(\"============== Before allocating blocks "
1678 "==============\\en\");\n"
1679 " display_mallinfo();\n"
1683 #: build/C/man3/mallinfo.3:252
1686 " for (j = 0; j E<lt> numBlocks; j++) {\n"
1687 " if (numBlocks E<gt>= MAX_ALLOCS)\n"
1688 " fatal(\"Too many allocations\");\n"
1692 #: build/C/man3/mallinfo.3:257
1695 " alloc[j] = malloc(blockSize);\n"
1696 " if (alloc[j] == NULL)\n"
1697 " errExit(\"malloc\");\n"
1702 #: build/C/man3/mallinfo.3:260
1705 " printf(\"\\en============== After allocating blocks "
1706 "==============\\en\");\n"
1707 " display_mallinfo();\n"
1711 #: build/C/man3/mallinfo.3:263
1714 " for (j = freeBegin; j E<lt> freeEnd; j += freeStep)\n"
1715 " free(alloc[j]);\n"
1719 #: build/C/man3/mallinfo.3:266
1722 " printf(\"\\en============== After freeing blocks "
1723 "==============\\en\");\n"
1724 " display_mallinfo();\n"
1728 #: build/C/man3/mallinfo.3:269 build/C/man3/malloc_info.3:255 build/C/man3/mallopt.3:565 build/C/man3/mcheck.3:203 build/C/man2/mmap.2:711
1731 " exit(EXIT_SUCCESS);\n"
1736 #: build/C/man3/mallinfo.3:279
1738 "B<mmap>(2), B<malloc>(3), B<malloc_info>(3), B<malloc_stats>(3), "
1739 "B<malloc_trim>(3), B<mallopt>(3)"
1743 #: build/C/man3/malloc_get_state.3:25
1745 msgid "MALLOC_GET_STATE"
1749 #: build/C/man3/malloc_get_state.3:25
1755 #: build/C/man3/malloc_get_state.3:28
1757 "malloc_get_state, malloc_set_state - record and restore state of malloc "
1762 #: build/C/man3/malloc_get_state.3:31 build/C/man3/posix_memalign.3:35
1764 msgid "B<#include E<lt>stdlib.hE<gt>>\n"
1768 #: build/C/man3/malloc_get_state.3:33
1770 msgid "B<void* malloc_get_state(void);>\n"
1774 #: build/C/man3/malloc_get_state.3:35
1776 msgid "B<int malloc_set_state(void *>I<state>B<);>\n"
1780 #: build/C/man3/malloc_get_state.3:53
1782 "The B<malloc_get_state>() function records the current state of all "
1783 "B<malloc>(3) internal bookkeeping variables (but not the actual contents of "
1784 "the heap or the state of B<malloc_hook>(3) functions pointers). The state "
1785 "is recorded in a system-dependent opaque data structure dynamically "
1786 "allocated via B<malloc>(3), and a pointer to that data structure is returned "
1787 "as the function result. (It is the caller's responsibility to B<free>(3) "
1792 #: build/C/man3/malloc_get_state.3:61
1794 "The B<malloc_set_state>() function restores the state of all B<malloc>(3) "
1795 "internal bookkeeping variables to the values recorded in the opaque data "
1796 "structure pointed to by I<state>."
1800 #: build/C/man3/malloc_get_state.3:68
1802 "On success, B<malloc_get_state>() returns a pointer to a newly allocated "
1803 "opaque data structure. On error (for example, memory could not be allocated "
1804 "for the data structure), B<malloc_get_state>() returns NULL."
1807 #. if(ms->magic != MALLOC_STATE_MAGIC) return -1;
1808 #. /* Must fail if the major version is too high. */
1809 #. if((ms->version & ~0xffl) > (MALLOC_STATE_VERSION & ~0xffl)) return -2;
1811 #: build/C/man3/malloc_get_state.3:86
1813 "On success, B<malloc_set_state>() returns 0. If the implementation detects "
1814 "that I<state> does not point to a correctly formed data structure, "
1815 "B<malloc_set_state>() returns -1. If the implementation detects that the "
1816 "version of the data structure referred to by I<state> is a more recent "
1817 "version than this implementation knows about, B<malloc_set_state>() returns "
1822 #: 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
1823 msgid "These functions are GNU extensions."
1827 #: build/C/man3/malloc_get_state.3:94
1829 "These functions are especially useful when using this B<malloc>(3) "
1830 "implementation as part of a shared library, and the heap contents are "
1831 "saved/restored via some other method. This technique is used by the GNU "
1832 "Emacs to implement its \"dumping\" function."
1835 #. i.e., calls __malloc_check_init()
1836 #. i.e., malloc checking is not already in use
1837 #. and the caller requested malloc checking
1839 #: build/C/man3/malloc_get_state.3:111
1841 "Hook function pointers are never saved or restored by these functions, with "
1842 "two exceptions: if malloc checking (see B<mallopt>(3)) was in use when "
1843 "B<malloc_get_state>() was called, then B<malloc_set_state>() resets malloc "
1844 "checking hooks if possible; if malloc checking was not in use in the "
1845 "recorded state, but the caller has requested malloc checking, then the hooks "
1850 #: build/C/man3/malloc_get_state.3:114
1851 msgid "B<malloc>(3), B<mallopt>(3)"
1855 #: build/C/man3/malloc_hook.3:10
1861 #: build/C/man3/malloc_hook.3:10
1867 #: build/C/man3/malloc_hook.3:15
1869 "__malloc_hook, __malloc_initialize_hook, __memalign_hook, __free_hook, "
1870 "__realloc_hook, __after_morecore_hook - malloc debugging variables"
1874 #: build/C/man3/malloc_hook.3:18 build/C/man3/malloc_info.3:31 build/C/man3/posix_memalign.3:41
1876 msgid "B<#include E<lt>malloc.hE<gt>>\n"
1880 #: build/C/man3/malloc_hook.3:20
1882 msgid "B<void *(*__malloc_hook)(size_t >I<size>B<, const void *>I<caller>B<);>\n"
1886 #: build/C/man3/malloc_hook.3:23
1889 "B<void *(*__realloc_hook)(void *>I<ptr>B<, size_t >I<size>B<, const void "
1890 "*>I<caller>B<);>\n"
1894 #: build/C/man3/malloc_hook.3:26
1897 "B<void *(*__memalign_hook)(size_t >I<alignment>B<, size_t >I<size>B<,>\n"
1898 "B< const void *>I<caller>B<);>\n"
1902 #: build/C/man3/malloc_hook.3:28
1904 msgid "B<void (*__free_hook)(void *>I<ptr>B<, const void *>I<caller>B<);>\n"
1908 #: build/C/man3/malloc_hook.3:30
1910 msgid "B<void (*__malloc_initialize_hook)(void);>\n"
1914 #: build/C/man3/malloc_hook.3:32
1916 msgid "B<void (*__after_morecore_hook)(void);>\n"
1920 #: build/C/man3/malloc_hook.3:43
1922 "The GNU C library lets you modify the behavior of B<malloc>(3), "
1923 "B<realloc>(3), and B<free>(3) by specifying appropriate hook functions. "
1924 "You can use these hooks to help you debug programs that use dynamic memory "
1925 "allocation, for example."
1929 #: build/C/man3/malloc_hook.3:50
1931 "The variable B<__malloc_initialize_hook> points at a function that is called "
1932 "once when the malloc implementation is initialized. This is a weak "
1933 "variable, so it can be overridden in the application with a definition like "
1938 #: build/C/man3/malloc_hook.3:53
1940 msgid " void (*__malloc_initialize_hook)(void) = my_init_hook;\n"
1944 #: build/C/man3/malloc_hook.3:58
1945 msgid "Now the function I<my_init_hook>() can do the initialization of all hooks."
1949 #: build/C/man3/malloc_hook.3:74
1951 "The four functions pointed to by B<__malloc_hook>, B<__realloc_hook>, "
1952 "B<__memalign_hook>, B<__free_hook> have a prototype like the functions "
1953 "B<malloc>(3), B<realloc>(3), B<memalign>(3), B<free>(3), respectively, "
1954 "except that they have a final argument I<caller> that gives the address of "
1955 "the caller of B<malloc>(3), etc."
1959 #: build/C/man3/malloc_hook.3:80
1961 "The variable B<__after_morecore_hook> points at a function that is called "
1962 "each time after B<sbrk>(2) was asked for more memory."
1965 #. https://bugzilla.redhat.com/show_bug.cgi?id=450187
1966 #. http://sourceware.org/bugzilla/show_bug.cgi?id=9957
1968 #: build/C/man3/malloc_hook.3:89
1970 "The use of these hook functions is not safe in multithreaded programs, and "
1971 "they are now deprecated. Programmers should instead preempt calls to the "
1972 "relevant functions by defining and exporting functions such as \"malloc\" "
1977 #: build/C/man3/malloc_hook.3:91
1978 msgid "Here is a short example of how to use these variables."
1982 #: build/C/man3/malloc_hook.3:95
1985 "#include E<lt>stdio.hE<gt>\n"
1986 "#include E<lt>malloc.hE<gt>\n"
1990 #: build/C/man3/malloc_hook.3:99
1993 "/* Prototypes for our hooks. */\n"
1994 "static void my_init_hook(void);\n"
1995 "static void *my_malloc_hook(size_t, const void *);\n"
1999 #: build/C/man3/malloc_hook.3:102
2002 "/* Variables to save original hooks. */\n"
2003 "static void *(*old_malloc_hook)(size_t, const void *);\n"
2007 #: build/C/man3/malloc_hook.3:105
2010 "/* Override initializing hook from the C library. */\n"
2011 "void (*__malloc_initialize_hook) (void) = my_init_hook;\n"
2015 #: build/C/man3/malloc_hook.3:112
2019 "my_init_hook(void)\n"
2021 " old_malloc_hook = __malloc_hook;\n"
2022 " __malloc_hook = my_malloc_hook;\n"
2027 #: build/C/man3/malloc_hook.3:117
2031 "my_malloc_hook(size_t size, const void *caller)\n"
2037 #: build/C/man3/malloc_hook.3:120
2040 " /* Restore all old hooks */\n"
2041 " __malloc_hook = old_malloc_hook;\n"
2045 #: build/C/man3/malloc_hook.3:123
2048 " /* Call recursively */\n"
2049 " result = malloc(size);\n"
2053 #: build/C/man3/malloc_hook.3:126
2056 " /* Save underlying hooks */\n"
2057 " old_malloc_hook = __malloc_hook;\n"
2061 #: build/C/man3/malloc_hook.3:130
2064 " /* printf() might call malloc(), so protect it too. */\n"
2065 " printf(\"malloc(%u) called from %p returns %p\\en\",\n"
2066 " (unsigned int) size, caller, result);\n"
2070 #: build/C/man3/malloc_hook.3:133
2073 " /* Restore our own hooks */\n"
2074 " __malloc_hook = my_malloc_hook;\n"
2078 #: build/C/man3/malloc_hook.3:136
2086 #: build/C/man3/malloc_hook.3:142
2087 msgid "B<mallinfo>(3), B<malloc>(3), B<mcheck>(3), B<mtrace>(3)"
2091 #: build/C/man3/malloc_info.3:25
2097 #: build/C/man3/malloc_info.3:25 build/C/man2/shmget.2:38
2103 #: build/C/man3/malloc_info.3:28
2104 msgid "malloc_info - export malloc state to a stream"
2108 #: build/C/man3/malloc_info.3:33
2110 msgid "B<int malloc_info(int >I<options>B<, FILE *>I<fp>B<);>\n"
2114 #: build/C/man3/malloc_info.3:44
2116 "The B<malloc_info>() function exports an XML string that describes the "
2117 "current state of the memory-allocation implementation in the caller. The "
2118 "string is printed on the file stream I<fp>. The exported string includes "
2119 "information about all arenas (see B<malloc>(3))."
2123 #: build/C/man3/malloc_info.3:48
2124 msgid "As currently implemented, I<options> must be zero."
2128 #: build/C/man3/malloc_info.3:55
2130 "On success, B<malloc_info>() returns 0; on error, it returns -1, with "
2131 "I<errno> set to indicate the cause."
2135 #: build/C/man3/malloc_info.3:60
2136 msgid "I<options> was nonzero."
2140 #: build/C/man3/malloc_info.3:63
2141 msgid "B<malloc_info>() was added to glibc in version 2.10."
2145 #: build/C/man3/malloc_info.3:65 build/C/man3/malloc_stats.3:57 build/C/man3/malloc_trim.3:62 build/C/man3/malloc_usable_size.3:50
2146 msgid "This function is a GNU extension."
2150 #: build/C/man3/malloc_info.3:71
2152 "The memory-allocation information is provided as an XML string (rather than "
2153 "a C structure) because the information may change over time (according to "
2154 "changes in the underlying implementation). The output XML string includes a "
2159 #: build/C/man3/malloc_info.3:77
2161 "The B<open_memstream>(3) function can be used to send the output of "
2162 "B<malloc_info>() directly into a buffer in memory, rather than to a file."
2166 #: build/C/man3/malloc_info.3:84
2168 "The B<malloc_info>() function is designed to address deficiencies in "
2169 "B<malloc_stats>(3) and B<mallinfo>(3)."
2173 #: build/C/man3/malloc_info.3:95
2175 "The program below takes up to four command-line arguments, of which the "
2176 "first three are mandatory. The first argument specifies the number of "
2177 "threads that the program should create. All of the threads, including the "
2178 "main thread, allocate the number of blocks of memory specified by the second "
2179 "argument. The third argument controls the size of the blocks to be "
2180 "allocated. The main thread creates blocks of this size, the second thread "
2181 "created by the program allocates blocks of twice this size, the third thread "
2182 "allocates blocks of three times this size, and so on."
2186 #: build/C/man3/malloc_info.3:102
2188 "The program calls B<malloc_info>() twice to display the memory-allocation "
2189 "state. The first call takes place before any threads are created or memory "
2190 "allocated. The second call is performed after all threads have allocated "
2195 #: build/C/man3/malloc_info.3:110
2197 "In the following example, the command-line arguments specify the creation of "
2198 "one additional thread, and both the main thread and the additional thread "
2199 "allocate 10000 blocks of memory. After the blocks of memory have been "
2200 "allocated, B<malloc_info>() shows the state of two allocation arenas."
2204 #: build/C/man3/malloc_info.3:135
2207 "$ B<getconf GNU_LIBC_VERSION>\n"
2209 "$ B<./a.out 1 10000 100>\n"
2210 "============ Before allocating blocks ============\n"
2211 "E<lt>malloc version=\"1\"E<gt>\n"
2212 "E<lt>heap nr=\"0\"E<gt>\n"
2214 "E<lt>/sizesE<gt>\n"
2215 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2216 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2217 "E<lt>system type=\"current\" size=\"135168\"/E<gt>\n"
2218 "E<lt>system type=\"max\" size=\"135168\"/E<gt>\n"
2219 "E<lt>aspace type=\"total\" size=\"135168\"/E<gt>\n"
2220 "E<lt>aspace type=\"mprotect\" size=\"135168\"/E<gt>\n"
2222 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2223 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2224 "E<lt>system type=\"current\" size=\"135168\"/E<gt>\n"
2225 "E<lt>system type=\"max\" size=\"135168\"/E<gt>\n"
2226 "E<lt>aspace type=\"total\" size=\"135168\"/E<gt>\n"
2227 "E<lt>aspace type=\"mprotect\" size=\"135168\"/E<gt>\n"
2228 "E<lt>/mallocE<gt>\n"
2232 #: build/C/man3/malloc_info.3:165
2235 "============ After allocating blocks ============\n"
2236 "E<lt>malloc version=\"1\"E<gt>\n"
2237 "E<lt>heap nr=\"0\"E<gt>\n"
2239 "E<lt>/sizesE<gt>\n"
2240 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2241 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2242 "E<lt>system type=\"current\" size=\"1081344\"/E<gt>\n"
2243 "E<lt>system type=\"max\" size=\"1081344\"/E<gt>\n"
2244 "E<lt>aspace type=\"total\" size=\"1081344\"/E<gt>\n"
2245 "E<lt>aspace type=\"mprotect\" size=\"1081344\"/E<gt>\n"
2247 "E<lt>heap nr=\"1\"E<gt>\n"
2249 "E<lt>/sizesE<gt>\n"
2250 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2251 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2252 "E<lt>system type=\"current\" size=\"1032192\"/E<gt>\n"
2253 "E<lt>system type=\"max\" size=\"1032192\"/E<gt>\n"
2254 "E<lt>aspace type=\"total\" size=\"1032192\"/E<gt>\n"
2255 "E<lt>aspace type=\"mprotect\" size=\"1032192\"/E<gt>\n"
2257 "E<lt>total type=\"fast\" count=\"0\" size=\"0\"/E<gt>\n"
2258 "E<lt>total type=\"rest\" count=\"0\" size=\"0\"/E<gt>\n"
2259 "E<lt>system type=\"current\" size=\"2113536\"/E<gt>\n"
2260 "E<lt>system type=\"max\" size=\"2113536\"/E<gt>\n"
2261 "E<lt>aspace type=\"total\" size=\"2113536\"/E<gt>\n"
2262 "E<lt>aspace type=\"mprotect\" size=\"2113536\"/E<gt>\n"
2263 "E<lt>/mallocE<gt>\n"
2267 #: build/C/man3/malloc_info.3:175
2270 "#include E<lt>unistd.hE<gt>\n"
2271 "#include E<lt>stdlib.hE<gt>\n"
2272 "#include E<lt>pthread.hE<gt>\n"
2273 "#include E<lt>malloc.hE<gt>\n"
2274 "#include E<lt>errno.hE<gt>\n"
2278 #: build/C/man3/malloc_info.3:178
2281 "static size_t blockSize;\n"
2282 "static int numThreads, numBlocks;\n"
2286 #: build/C/man3/malloc_info.3:181
2289 "#define errExit(msg) do { perror(msg); exit(EXIT_FAILURE); \\e\n"
2294 #: build/C/man3/malloc_info.3:187
2298 "thread_func(void *arg)\n"
2301 " int tn = (int) arg;\n"
2305 #: build/C/man3/malloc_info.3:190
2308 " /* The multiplier \\(aq(2 + tn)\\(aq ensures that each thread "
2310 " the main thread) allocates a different amount of memory */\n"
2314 #: build/C/man3/malloc_info.3:194
2317 " for (j = 0; j E<lt> numBlocks; j++)\n"
2318 " if (malloc(blockSize * (2 + tn)) == NULL)\n"
2319 " errExit(\"malloc-thread\");\n"
2323 #: build/C/man3/malloc_info.3:198
2326 " sleep(100); /* Sleep until main thread terminates */\n"
2332 #: build/C/man3/malloc_info.3:204
2336 "main(int argc, char *argv[])\n"
2338 " int j, tn, sleepTime;\n"
2339 " pthread_t *thr;\n"
2343 #: build/C/man3/malloc_info.3:211
2346 " if (argc E<lt> 4) {\n"
2347 " fprintf(stderr,\n"
2348 " \"%s num-threads num-blocks block-size [sleep-time]\\en\",\n"
2350 " exit(EXIT_FAILURE);\n"
2355 #: build/C/man3/malloc_info.3:216
2358 " numThreads = atoi(argv[1]);\n"
2359 " numBlocks = atoi(argv[2]);\n"
2360 " blockSize = atoi(argv[3]);\n"
2361 " sleepTime = (argc E<gt> 4) ? atoi(argv[4]) : 0;\n"
2365 #: build/C/man3/malloc_info.3:220
2368 " thr = calloc(numThreads, sizeof(pthread_t));\n"
2369 " if (thr == NULL)\n"
2370 " errExit(\"calloc\");\n"
2374 #: build/C/man3/malloc_info.3:223
2377 " printf(\"============ Before allocating blocks ============\\en\");\n"
2378 " malloc_info(0, stdout);\n"
2382 #: build/C/man3/malloc_info.3:225
2384 msgid " /* Create threads that allocate different amounts of memory */\n"
2388 #: build/C/man3/malloc_info.3:231
2391 " for (tn = 0; tn E<lt> numThreads; tn++) {\n"
2392 " errno = pthread_create(&thr[tn], NULL, thread_func,\n"
2394 " if (errno != 0)\n"
2395 " errExit(\"pthread_create\");\n"
2399 #: build/C/man3/malloc_info.3:236
2402 " /* If we add a sleep interval after the start-up of each\n"
2403 " thread, the threads likely won\\(aqt contend for malloc\n"
2404 " mutexes, and therefore additional arenas won\\(aqt be\n"
2405 " allocated (see malloc(3)). */\n"
2409 #: build/C/man3/malloc_info.3:240
2412 " if (sleepTime E<gt> 0)\n"
2413 " sleep(sleepTime);\n"
2418 #: build/C/man3/malloc_info.3:242
2420 msgid " /* The main thread also allocates some memory */\n"
2424 #: build/C/man3/malloc_info.3:246
2427 " for (j = 0; j E<lt> numBlocks; j++)\n"
2428 " if (malloc(blockSize) == NULL)\n"
2429 " errExit(\"malloc\");\n"
2433 #: build/C/man3/malloc_info.3:249
2436 " sleep(2); /* Give all threads a chance to\n"
2437 " complete allocations */\n"
2441 #: build/C/man3/malloc_info.3:252
2444 " printf(\"\\en============ After allocating blocks ============\\en\");\n"
2445 " malloc_info(0, stdout);\n"
2449 #: build/C/man3/malloc_info.3:262
2451 "B<mallinfo>(3), B<malloc>(3), B<malloc_stats>(3), B<mallopt>(3), "
2452 "B<open_memstream>(3)"
2456 #: build/C/man3/malloc_stats.3:26
2458 msgid "MALLOC_STATS"
2462 #: build/C/man3/malloc_stats.3:29
2463 msgid "malloc_stats - print memory allocation statistics"
2467 #: build/C/man3/malloc_stats.3:33
2468 msgid "B<void malloc_stats(void);>"
2472 #. Available already in glibc 2.0, possibly earlier
2474 #: build/C/man3/malloc_stats.3:55
2476 "The B<malloc_stats>() function prints (on standard error) statistics about "
2477 "memory allocated by B<malloc>(3) and related functions. For each arena "
2478 "(allocation area), this function prints the total amount of memory allocated "
2479 "and the total number of bytes consumed by in-use allocations. (These two "
2480 "values correspond to the I<arena> and I<uordblks> fields retrieved by "
2481 "B<mallinfo>(3).) In addition, the function prints the sum of these two "
2482 "statistics for all arenas, and the maximum number of blocks and bytes that "
2483 "were ever simultaneously allocated using B<mmap>(2)."
2487 #: build/C/man3/malloc_stats.3:61
2489 "More detailed information about memory allocations in the main arena can be "
2490 "obtained using B<mallinfo>(3)."
2494 #: build/C/man3/malloc_stats.3:67
2495 msgid "B<mmap>(2), B<mallinfo>(3), B<malloc>(3), B<malloc_info>(3), B<mallopt>(3)"
2499 #: build/C/man3/malloc_trim.3:26
2505 #: build/C/man3/malloc_trim.3:26 build/C/man3/malloc_usable_size.3:25
2511 #: build/C/man3/malloc_trim.3:29
2512 msgid "malloc_trim - release free memory from the top of the heap"
2516 #: build/C/man3/malloc_trim.3:33
2517 msgid "B<void malloc_trim(size_t >I<pad>B<);>"
2521 #: build/C/man3/malloc_trim.3:40
2523 "The B<malloc_trim>() function attempts to release free memory at the top of "
2524 "the heap (by calling B<sbrk>(2) with a suitable argument)."
2528 #: build/C/man3/malloc_trim.3:51
2530 "The I<pad> argument specifies the amount of free space to leave untrimmed at "
2531 "the top of the heap. If this argument is 0, only the minimum amount of "
2532 "memory is maintained at the top of the heap (i.e., one page or less). A "
2533 "nonzero argument can be used to maintain some trailing space at the top of "
2534 "the heap in order to allow future allocations to be made without having to "
2535 "extend the heap with B<sbrk>(2)."
2539 #: build/C/man3/malloc_trim.3:56
2541 "The B<malloc_trim>() function returns 1 if memory was actually released "
2542 "back to the system, or 0 if it was not possible to release any memory."
2546 #. Available already in glibc 2.0, possibly earlier
2548 #: build/C/man3/malloc_trim.3:60
2549 msgid "No errors are defined."
2553 #: build/C/man3/malloc_trim.3:71
2555 "This function is automatically called by B<free>(3) in certain "
2556 "circumstances; see the discussion of B<M_TOP_PAD> and B<M_TRIM_THRESHOLD> in "
2561 #: build/C/man3/malloc_trim.3:74
2563 "This function cannot release free memory located at places other than the "
2567 #. malloc/malloc.c::mTRIm():
2568 #. return result | (av == &main_arena ? sYSTRIm (pad, av) : 0);
2570 #: build/C/man3/malloc_trim.3:78
2571 msgid "This function releases only memory in the main arena."
2575 #: build/C/man3/malloc_trim.3:82
2576 msgid "B<sbrk>(2), B<malloc>(3), B<mallopt>(3)"
2580 #: build/C/man3/malloc_usable_size.3:25
2582 msgid "MALLOC_USABLE_SIZE"
2586 #: build/C/man3/malloc_usable_size.3:28
2587 msgid "malloc_usable_size - obtain size of block of memory allocated from heap"
2591 #: build/C/man3/malloc_usable_size.3:32
2592 msgid "B<size_t malloc_usable_size (void *>I<ptr>B<);>"
2596 #: build/C/man3/malloc_usable_size.3:40
2598 "The B<malloc_usable_size>() function returns the number of usable bytes in "
2599 "the block pointed to by I<ptr>, a pointer to a block of memory allocated by "
2600 "B<malloc>(3) or a related function."
2604 #: build/C/man3/malloc_usable_size.3:48
2606 "B<malloc_usable_size>() returns the number of usable bytes in the block of "
2607 "allocated memory pointed to by I<ptr>. If I<ptr> is NULL, 0 is returned."
2611 #: build/C/man3/malloc_usable_size.3:60
2613 "The value returned by B<malloc_usable_size>() may be greater than the "
2614 "requested size of the allocation because of alignment and minimum size "
2615 "constraints. Although the excess bytes can be overwritten by the "
2616 "application without ill effects, this is not good programming practice: the "
2617 "number of excess bytes in an allocation depends on the underlying "
2622 #: build/C/man3/malloc_usable_size.3:62
2623 msgid "The main use of this function is for debugging and introspection."
2627 #: build/C/man3/malloc_usable_size.3:64
2628 msgid "B<malloc>(3)"
2632 #: build/C/man3/mallopt.3:26
2638 #: build/C/man3/mallopt.3:26
2644 #: build/C/man3/mallopt.3:29
2645 msgid "mallopt - set memory allocation parameters"
2649 #: build/C/man3/mallopt.3:33
2650 msgid "B<int mallopt(int >I<param>B<, int >I<value>B<);>"
2654 #: build/C/man3/mallopt.3:44
2656 "The B<mallopt>() function adjusts parameters that control the behavior of "
2657 "the memory-allocation functions (see B<malloc>(3)). The I<param> argument "
2658 "specifies the parameter to be modified, and I<value> specifies the new value "
2659 "for that parameter."
2663 #: build/C/man3/mallopt.3:47
2664 msgid "The following values can be specified for I<param>:"
2668 #: build/C/man3/mallopt.3:47
2670 msgid "B<M_CHECK_ACTION>"
2674 #: build/C/man3/mallopt.3:53
2676 "Setting this parameter controls how glibc responds when various kinds of "
2677 "programming errors are detected (e.g., freeing the same pointer twice). The "
2678 "3 least significant bits (2, 1, and 0) of the value assigned to this "
2679 "parameter determine the glibc behavior, as follows:"
2683 #: build/C/man3/mallopt.3:54
2689 #: build/C/man3/mallopt.3:64
2691 "If this bit is set, then print a one-line message on I<stderr> that provides "
2692 "details about the error. The message starts with the string \"***\\ glibc "
2693 "detected\\ ***\", followed by the program name, the name of the "
2694 "memory-allocation function in which the error was detected, a brief "
2695 "description of the error, and the memory address where the error was "
2700 #: build/C/man3/mallopt.3:64
2706 #: build/C/man3/mallopt.3:79
2708 "If this bit is set, then, after printing any error message specified by bit "
2709 "0, the program is terminated by calling B<abort>(3). In glibc versions "
2710 "since 2.4, if bit 0 is also set, then, between printing the error message "
2711 "and aborting, the program also prints a stack trace in the manner of "
2712 "B<backtrace>(3), and prints the process's memory mapping in the style of "
2713 "I</proc/[pid]/maps> (see B<proc>(5))."
2717 #: build/C/man3/mallopt.3:79
2719 msgid "Bit 2 (since glibc 2.4)"
2723 #: build/C/man3/mallopt.3:86
2725 "This bit has an effect only if bit 0 is also set. If this bit is set, then "
2726 "the one-line message describing the error is simplified to contain just the "
2727 "name of the function where the error was detected and the brief description "
2732 #: build/C/man3/mallopt.3:91
2733 msgid "The remaining bits in I<value> are ignored."
2737 #: build/C/man3/mallopt.3:95
2739 "Combining the above details, the following numeric values are meaningful for "
2740 "B<M_CHECK_ACTION>:"
2744 #: build/C/man3/mallopt.3:96
2750 #: build/C/man3/mallopt.3:98
2751 msgid "Ignore error conditions; continue execution (with undefined results)."
2755 #: build/C/man3/mallopt.3:98
2761 #: build/C/man3/mallopt.3:100
2762 msgid "Print a detailed error message and continue execution."
2766 #: build/C/man3/mallopt.3:100
2772 #: build/C/man3/mallopt.3:102
2773 msgid "Abort the program."
2777 #: build/C/man3/mallopt.3:102
2783 #: build/C/man3/mallopt.3:105
2785 "Print detailed error message, stack trace, and memory mappings, and abort "
2790 #: build/C/man3/mallopt.3:105
2796 #: build/C/man3/mallopt.3:107
2797 msgid "Print a simple error message and continue execution."
2801 #: build/C/man3/mallopt.3:107
2807 #: build/C/man3/mallopt.3:110
2809 "Print simple error message, stack trace, and memory mappings, and abort the "
2814 #: build/C/man3/mallopt.3:116
2816 "Since glibc 2.3.4, the default value for the B<M_CHECK_ACTION> parameter is "
2817 "3. In glibc version 2.3.3 and earlier, the default value is 1."
2821 #: build/C/man3/mallopt.3:121
2823 "Using a nonzero B<M_CHECK_ACTION> value can be useful because otherwise a "
2824 "crash may happen much later, and the true cause of the problem is then very "
2825 "hard to track down."
2829 #: build/C/man3/mallopt.3:121
2831 msgid "B<M_MMAP_MAX>"
2834 #. The following text adapted from comments in the glibc source:
2836 #: build/C/man3/mallopt.3:131
2838 "This parameter specifies the maximum number of allocation requests that may "
2839 "be simultaneously serviced using B<mmap>(2). This parameter exists because "
2840 "some systems have a limited number of internal tables for use by B<mmap>(2), "
2841 "and using more than a few of them may degrade performance."
2845 #: build/C/man3/mallopt.3:138
2847 "The default value is 65,536, a value which has no special significance and "
2848 "which servers only as a safeguard. Setting this parameter to 0 disables the "
2849 "use of B<mmap>(2) for servicing large allocation requests."
2853 #: build/C/man3/mallopt.3:138
2855 msgid "B<M_MMAP_THRESHOLD>"
2859 #: build/C/man3/mallopt.3:147
2861 "For allocations greater than or equal to the limit specified (in bytes) by "
2862 "B<M_MMAP_THRESHOLD> that can't be satisfied from the free list, the "
2863 "memory-allocation functions employ B<mmap>(2) instead of increasing the "
2864 "program break using B<sbrk>(2)."
2868 #: build/C/man3/mallopt.3:167
2870 "Allocating memory using B<mmap>(2) has the significant advantage that the "
2871 "allocated memory blocks can always be independently released back to the "
2872 "system. (By contrast, the heap can be trimmed only if memory is freed at "
2873 "the top end.) On the other hand, there are some disadvantages to the use of "
2874 "B<mmap>(2): deallocated space is not placed on the free list for reuse by "
2875 "later allocations; memory may be wasted because B<mmap>(2) allocations must "
2876 "be page-aligned; and the kernel must perform the expensive task of zeroing "
2877 "out memory allocated via B<mmap>(2). Balancing these factors leads to a "
2878 "default setting of 128*1024 for the B<M_MMAP_THRESHOLD> parameter."
2882 #: build/C/man3/mallopt.3:174
2884 "The lower limit for this parameter is 0. The upper limit is "
2885 "B<DEFAULT_MMAP_THRESHOLD_MAX>: 512*1024 on 32-bit systems or "
2886 "I<4*1024*1024*sizeof(long)> on 64-bit systems."
2890 #: build/C/man3/mallopt.3:192
2892 "I<Note:> Nowadays, glibc uses a dynamic mmap threshold by default. The "
2893 "initial value of the threshold is 128*1024, but when blocks larger than the "
2894 "current threshold and less than or equal to B<DEFAULT_MMAP_THRESHOLD_MAX> "
2895 "are freed, the threshold is adjusted upwards to the size of the freed "
2896 "block. When dynamic mmap thresholding is in effect, the threshold for "
2897 "trimming the heap is also dynamically adjusted to be twice the dynamic mmap "
2898 "threshold. Dynamic adjustment of the mmap threshold is disabled if any of "
2899 "the B<M_TRIM_THRESHOLD>, B<M_TOP_PAD>, B<M_MMAP_THRESHOLD>, or B<M_MMAP_MAX> "
2900 "parameters is set."
2904 #: build/C/man3/mallopt.3:192
2906 msgid "B<M_MXFAST> (since glibc 2.3)"
2909 #. The following text adapted from comments in the glibc sources:
2911 #: build/C/man3/mallopt.3:212
2913 "Set the upper limit for memory allocation requests that are satisfied using "
2914 "\"fastbins\". (The measurement unit for this parameter is bytes.) Fastbins "
2915 "are storage areas that hold deallocated blocks of memory of the same size "
2916 "without merging adjacent free blocks. Subsequent reallocation of blocks of "
2917 "the same size can be handled very quickly by allocating from the fastbin, "
2918 "although memory fragmentation and the overall memory footprint of the "
2919 "program can increase. The default value for this parameter is "
2920 "I<64*sizeof(size_t)/4> (i.e., 64 on 32-bit architectures). The range for "
2921 "this parameter is 0 to I<80*sizeof(size_t)/4>. Setting B<M_MXFAST> to 0 "
2922 "disables the use of fastbins."
2926 #: build/C/man3/mallopt.3:212
2928 msgid "B<M_PERTURB> (since glibc 2.4)"
2932 #: build/C/man3/mallopt.3:227
2934 "If this parameter is set to a nonzero value, then bytes of allocated memory "
2935 "(other than allocations via B<calloc>(3)) are initialized to the complement "
2936 "of the value in the least significant byte of I<value>, and when allocated "
2937 "memory is released using B<free>(3), the freed bytes are set to the least "
2938 "significant byte of I<value>. This can be useful for detecting errors where "
2939 "programs incorrectly rely on allocated memory being initialized to zero, or "
2940 "reuse values in memory that has already been freed."
2944 #: build/C/man3/mallopt.3:227
2946 msgid "B<M_TOP_PAD>"
2950 #: build/C/man3/mallopt.3:234
2952 "This parameter defines the amount of padding to employ when calling "
2953 "B<sbrk>(2) to modify the program break. (The measurement unit for this "
2954 "parameter is bytes.) This parameter has an effect in the following "
2959 #: build/C/man3/mallopt.3:241
2961 "When the program break is increased, then B<M_TOP_PAD> bytes are added to "
2962 "the B<sbrk>(2) request."
2966 #: build/C/man3/mallopt.3:247
2968 "When the heap is trimmed as a consequence of calling B<free>(3) (see the "
2969 "discussion of B<M_TRIM_THRESHOLD>) this much free space is preserved at the "
2974 #: build/C/man3/mallopt.3:251
2976 "In either case, the amount of padding is always rounded to a system page "
2981 #: build/C/man3/mallopt.3:258
2983 "Modifying B<M_TOP_PAD> is a trade-off between increasing the number of "
2984 "system calls (when the parameter is set low) and wasting unused memory at "
2985 "the top of the heap (when the parameter is set high)."
2988 #. DEFAULT_TOP_PAD in glibc source
2990 #: build/C/man3/mallopt.3:261
2991 msgid "The default value for this parameter is 128*1024."
2995 #: build/C/man3/mallopt.3:261
2997 msgid "B<M_TRIM_THRESHOLD>"
3001 #: build/C/man3/mallopt.3:277
3003 "When the amount of contiguous free memory at the top of the heap grows "
3004 "sufficiently large, B<free>(3) employs B<sbrk>(2) to release this memory "
3005 "back to the system. (This can be useful in programs that continue to "
3006 "execute for a long period after freeing a significant amount of memory.) "
3007 "The B<M_TRIM_THRESHOLD> parameter specifies the minimum size (in bytes) that "
3008 "this block of memory must reach before B<sbrk>(2) is used to trim the heap."
3012 #: build/C/man3/mallopt.3:282
3014 "The default value for this parameter is 128*1024. Setting "
3015 "B<M_TRIM_THRESHOLD> to -1 disables trimming completely."
3018 #. FIXME Do the arena parameters need to be documented?
3020 #. .BR M_ARENA_TEST " (since glibc 2.10)"
3022 #. .BR M_ARENA_MAX " (since glibc 2.10)"
3024 #. Environment variables
3025 #. MALLOC_ARENA_MAX_
3026 #. MALLOC_ARENA_TEST_
3028 #. http://udrepper.livejournal.com/20948.html describes some details
3029 #. of the MALLOC_ARENA_* environment variables.
3031 #. These macros aren't enabled in production releases until 2.15?
3032 #. (see glibc malloc/Makefile)
3034 #: build/C/man3/mallopt.3:305
3036 "Modifying B<M_TRIM_THRESHOLD> is a trade-off between increasing the number "
3037 "of system calls (when the parameter is set low) and wasting unused memory "
3038 "at the top of the heap (when the parameter is set high)."
3042 #: build/C/man3/mallopt.3:305
3044 msgid "Environment variables"
3048 #: build/C/man3/mallopt.3:320
3050 "A number of environment variables can be defined to modify some of the same "
3051 "parameters as are controlled by B<mallopt>(). Using these variables has the "
3052 "advantage that the source code of the program need not be changed. To be "
3053 "effective, these variables must be defined before the first call to a "
3054 "memory-allocation function. (If the same parameters are adjusted via "
3055 "B<mallopt>() then the B<mallopt>() settings take precedence.) For "
3056 "security reasons, these variables are ignored in set-user-ID and "
3057 "set-group-ID programs."
3061 #: build/C/man3/mallopt.3:323
3063 "The environment variables are as follows (note the trailing underscore at "
3064 "the end of the name of each variable):"
3068 #: build/C/man3/mallopt.3:323
3070 msgid "B<MALLOC_CHECK_>"
3073 #. On glibc 2.12/x86, a simple malloc()+free() loop is about 70% slower
3074 #. when MALLOC_CHECK_ was set.
3076 #: build/C/man3/mallopt.3:340
3078 "This environment variable controls the same parameter as B<mallopt>() "
3079 "B<M_CHECK_ACTION>. If this variable is set to a nonzero value, then a "
3080 "special implementation of the memory-allocation functions is used. (This is "
3081 "accomplished using the B<malloc_hook>(3) feature.) This implementation "
3082 "performs additional error checking, but is slower than the standard set of "
3083 "memory-allocation functions. (This implementation does not detect all "
3084 "possible errors; memory leaks can still occur.)"
3088 #: build/C/man3/mallopt.3:345
3090 "The value assigned to this environment variable should be a single digit, "
3091 "whose meaning is as described for B<M_CHECK_ACTION>. Any characters beyond "
3092 "the initial digit are ignored."
3096 #: build/C/man3/mallopt.3:354
3098 "For security reasons, the effect of B<MALLOC_CHECK_> is disabled by default "
3099 "for set-user-ID and set-group-ID programs. However, if the file "
3100 "I</etc/suid-debug> exists (the content of the file is irrelevant), then "
3101 "B<MALLOC_CHECK_> also has an effect for set-user-ID and set-group-ID "
3106 #: build/C/man3/mallopt.3:354
3108 msgid "B<MALLOC_MMAP_MAX_>"
3112 #: build/C/man3/mallopt.3:359
3113 msgid "Controls the same parameter as B<mallopt>() B<M_MMAP_MAX>."
3117 #: build/C/man3/mallopt.3:359
3119 msgid "B<MALLOC_MMAP_THRESHOLD_>"
3123 #: build/C/man3/mallopt.3:364
3124 msgid "Controls the same parameter as B<mallopt>() B<M_MMAP_THRESHOLD>."
3128 #: build/C/man3/mallopt.3:364
3130 msgid "B<MALLOC_PERTURB_>"
3134 #: build/C/man3/mallopt.3:369
3135 msgid "Controls the same parameter as B<mallopt>() B<M_PERTURB>."
3139 #: build/C/man3/mallopt.3:369
3141 msgid "B<MALLOC_TRIM_THRESHOLD_>"
3145 #: build/C/man3/mallopt.3:374
3146 msgid "Controls the same parameter as B<mallopt>() B<M_TRIM_THRESHOLD>."
3150 #: build/C/man3/mallopt.3:374
3152 msgid "B<MALLOC_TOP_PAD_>"
3156 #: build/C/man3/mallopt.3:379
3157 msgid "Controls the same parameter as B<mallopt>() B<M_TOP_PAD>."
3161 #: build/C/man3/mallopt.3:384
3162 msgid "On success, B<mallopt>() returns 1. On error, it returns 0."
3166 #. Available already in glibc 2.0, possibly earlier
3168 #: build/C/man3/mallopt.3:392
3169 msgid "On error, I<errno> is I<not> set."
3174 #: build/C/man3/mallopt.3:406
3176 "This function is not specified by POSIX or the C standards. A similar "
3177 "function exists on many System V derivatives, but the range of values for "
3178 "I<param> varies across systems. The SVID defined options B<M_MXFAST>, "
3179 "B<M_NLBLKS>, B<M_GRAIN>, and B<M_KEEP>, but only the first of these is "
3180 "implemented in glibc."
3184 #: build/C/man3/mallopt.3:410
3185 msgid "Specifying an invalid value for I<param> does not generate an error."
3188 #. FIXME This looks buggy:
3189 #. setting the M_MXFAST limit rounds up: (s + SIZE_SZ) & ~MALLOC_ALIGN_MASK)
3190 #. malloc requests are rounded up:
3191 #. (req) + SIZE_SZ + MALLOC_ALIGN_MASK) & ~MALLOC_ALIGN_MASK
3192 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=12129
3194 #: build/C/man3/mallopt.3:418
3196 "A calculation error within the glibc implementation means that a call of the "
3201 #: build/C/man3/mallopt.3:421
3203 msgid " mallopt(M_MXFAST, n)\n"
3206 #. Bins are multiples of 2 * sizeof(size_t) + sizeof(size_t)
3208 #: build/C/man3/mallopt.3:433
3210 "does not result in fastbins being employed for all allocations of size up to "
3211 "I<n>. To ensure desired results, I<n> should be rounded up to the next "
3212 "multiple greater than or equal to I<(2k+1)*sizeof(size_t)>, where I<k> is an "
3216 #. FIXME MALLOC_MMAP_THRESHOLD_ and MALLOC_MMAP_MAX_
3217 #. do have an effect for set-user-ID programs (but not
3218 #. set-group-ID programs).
3219 #. http://sources.redhat.com/bugzilla/show_bug.cgi?id=12155
3221 #: build/C/man3/mallopt.3:445
3223 "The B<MALLOC_MMAP_THRESHOLD_> and B<MALLOC_MMAP_MAX_> variables are I<not> "
3224 "ignored in set-group-ID programs."
3227 #. FIXME http://sources.redhat.com/bugzilla/show_bug.cgi?id=12140
3229 #: build/C/man3/mallopt.3:466
3231 "If B<mallopt>() is used to set B<M_PERTURB>, then, as expected, the bytes "
3232 "of allocated memory are initialized to the complement of the byte in "
3233 "I<value>, and when that memory is freed, the bytes of the region are "
3234 "initialized to the byte specified in I<value>. However, there is an "
3235 "off-by-I<sizeof(size_t)> error in the implementation: instead of "
3236 "initializing precisely the block of memory being freed by the call "
3237 "I<free(p)>, the block starting at I<p+sizeof(size_t)> is initialized."
3241 #: build/C/man3/mallopt.3:475
3243 "The program below demonstrates the use of B<M_CHECK_ACTION>. If the program "
3244 "is supplied with an (integer) command-line argument, then that argument is "
3245 "used to set the B<M_CHECK_ACTION> parameter. The program then allocates a "
3246 "block of memory, and frees it twice (an error)."
3250 #: build/C/man3/mallopt.3:479
3252 "The following shell session shows what happens when we run this program "
3253 "under glibc, with the default value for B<M_CHECK_ACTION>:"
3257 #: build/C/man3/mallopt.3:499
3261 "main(): returned from first free() call\n"
3262 "*** glibc detected *** ./a.out: double free or corruption (top): 0x09d30008 "
3264 "======= Backtrace: =========\n"
3265 "/lib/libc.so.6(+0x6c501)[0x523501]\n"
3266 "/lib/libc.so.6(+0x6dd70)[0x524d70]\n"
3267 "/lib/libc.so.6(cfree+0x6d)[0x527e5d]\n"
3268 "\\&./a.out[0x80485db]\n"
3269 "/lib/libc.so.6(__libc_start_main+0xe7)[0x4cdce7]\n"
3270 "\\&./a.out[0x8048471]\n"
3271 "======= Memory map: ========\n"
3272 "001e4000-001fe000 r-xp 00000000 08:06 1083555 /lib/libgcc_s.so.1\n"
3273 "001fe000-001ff000 r--p 00019000 08:06 1083555 /lib/libgcc_s.so.1\n"
3274 "[some lines omitted]\n"
3275 "b7814000-b7817000 rw-p 00000000 00:00 0\n"
3276 "bff53000-bff74000 rw-p 00000000 00:00 0 [stack]\n"
3277 "Aborted (core dumped)\n"
3281 #: build/C/man3/mallopt.3:504
3283 "The following runs show the results when employing other values for "
3284 "B<M_CHECK_ACTION>:"
3288 #: build/C/man3/mallopt.3:517
3291 "$ B<./a.out 1> # Diagnose error and continue\n"
3292 "main(): returned from first free() call\n"
3293 "*** glibc detected *** ./a.out: double free or corruption (top): 0x09cbe008 "
3295 "main(): returned from second free() call\n"
3296 "$ B<./a.out 2> # Abort without error message\n"
3297 "main(): returned from first free() call\n"
3298 "Aborted (core dumped)\n"
3299 "$ B<./a.out 0> # Ignore error and continue\n"
3300 "main(): returned from first free() call\n"
3301 "main(): returned from second free() call\n"
3305 #: build/C/man3/mallopt.3:523
3307 "The next run shows how to set the same parameter using the B<MALLOC_CHECK_> "
3308 "environment variable:"
3312 #: build/C/man3/mallopt.3:530
3315 "$ B<MALLOC_CHECK_=1 ./a.out>\n"
3316 "main(): returned from first free() call\n"
3317 "*** glibc detected *** ./a.out: free(): invalid pointer: 0x092c2008 ***\n"
3318 "main(): returned from second free() call\n"
3322 #: build/C/man3/mallopt.3:538
3325 "#include E<lt>malloc.hE<gt>\n"
3326 "#include E<lt>stdio.hE<gt>\n"
3327 "#include E<lt>stdlib.hE<gt>\n"
3331 #: build/C/man3/mallopt.3:543 build/C/man3/mcheck.3:187
3335 "main(int argc, char *argv[])\n"
3341 #: build/C/man3/mallopt.3:550
3344 " if (argc E<gt> 1) {\n"
3345 " if (mallopt(M_CHECK_ACTION, atoi(argv[1])) != 1) {\n"
3346 " fprintf(stderr, \"mallopt() failed\");\n"
3347 " exit(EXIT_FAILURE);\n"
3353 #: build/C/man3/mallopt.3:556
3356 " p = malloc(1000);\n"
3357 " if (p == NULL) {\n"
3358 " fprintf(stderr, \"malloc() failed\");\n"
3359 " exit(EXIT_FAILURE);\n"
3364 #: build/C/man3/mallopt.3:559
3368 " printf(\"main(): returned from first free() call\\en\");\n"
3372 #: build/C/man3/mallopt.3:562
3376 " printf(\"main(): returned from second free() call\\en\");\n"
3380 #: build/C/man3/mallopt.3:580
3382 "B<mmap>(2), B<sbrk>(2), B<mallinfo>(3), B<malloc>(3), B<malloc_hook>(3), "
3383 "B<malloc_info>(3), B<malloc_stats>(3), B<malloc_trim>(3), B<mcheck>(3), "
3384 "B<mtrace>(3), B<posix_memalign>(3)"
3388 #: build/C/man3/mcheck.3:25
3394 #: build/C/man3/mcheck.3:25 build/C/man3/mtrace.3:25
3400 #: build/C/man3/mcheck.3:28
3402 "mcheck, mcheck_check_all, mcheck_pedantic, mprobe - heap consistency "
3407 #: build/C/man3/mcheck.3:31
3409 msgid "B<#include E<lt>mcheck.hE<gt>>\n"
3413 #: build/C/man3/mcheck.3:33
3415 msgid "B<int mcheck(void (*>I<abortfunc>B<)(enum mcheck_status >I<mstatus>B<));>\n"
3419 #: build/C/man3/mcheck.3:35
3422 "B<int mcheck_pedantic(void (*>I<abortfunc>B<)(enum mcheck_status "
3423 ">I<mstatus>B<));>\n"
3427 #: build/C/man3/mcheck.3:37
3429 msgid "B<void mcheck_check_all(void);>\n"
3433 #: build/C/man3/mcheck.3:39
3435 msgid "B<enum mcheck_status mprobe(void *>I<ptr>B<);>\n"
3439 #: build/C/man3/mcheck.3:51
3441 "The B<mcheck>() function installs a set of debugging hooks for the "
3442 "B<malloc>(3) family of memory-allocation functions. These hooks cause "
3443 "certain consistency checks to be performed on the state of the heap. The "
3444 "checks can detect application errors such as freeing a block of memory more "
3445 "than once or corrupting the bookkeeping data structures that immediately "
3446 "precede a block of allocated memory."
3450 #: build/C/man3/mcheck.3:63
3452 "To be effective, the B<mcheck>() function must be called before the first "
3453 "call to B<malloc>(3) or a related function. In cases where this is "
3454 "difficult to ensure, linking the program with I<-mcheck> inserts an implicit "
3455 "call to B<mcheck>() (with a NULL argument) before the first call to a "
3456 "memory-allocation function."
3460 #: build/C/man3/mcheck.3:71
3462 "The B<mcheck_pedantic>() function is similar to B<mcheck>(), but performs "
3463 "checks on all allocated blocks whenever one of the memory-allocation "
3464 "functions is called. This can be very slow!"
3468 #: build/C/man3/mcheck.3:78
3470 "The B<mcheck_check_all>() function causes an immediate check on all "
3471 "allocated blocks. This call is effective only if B<mcheck>() is called "
3476 #: build/C/man3/mcheck.3:91
3478 "If the system detects an inconsistency in the heap, the caller-supplied "
3479 "function pointed to by I<abortfunc> is invoked with a single argument "
3480 "argument, I<mstatus>, that indicates what type of inconsistency was "
3481 "detected. If I<abortfunc> is NULL, a default function prints an error "
3482 "message on I<stderr> and calls B<abort>(3)."
3486 #: build/C/man3/mcheck.3:103
3488 "The B<mprobe>() function performs a consistency check on the block of "
3489 "allocated memory pointed to by I<ptr>. The B<mcheck>() function should be "
3490 "called beforehand (otherwise B<mprobe>() returns B<MCHECK_DISABLED>)."
3494 #: build/C/man3/mcheck.3:111
3496 "The following list describes the values returned by B<mprobe>() or passed "
3497 "as the I<mstatus> argument when I<abortfunc> is invoked:"
3501 #: build/C/man3/mcheck.3:111
3503 msgid "B<MCHECK_DISABLED> (B<mprobe>() only)"
3507 #: build/C/man3/mcheck.3:116
3509 "B<mcheck>() was not called before the first memory allocation function was "
3510 "called. Consistency checking is not possible."
3514 #: build/C/man3/mcheck.3:116
3516 msgid "B<MCHECK_OK> (B<mprobe>() only)"
3520 #: build/C/man3/mcheck.3:119
3521 msgid "No inconsistency detected."
3525 #: build/C/man3/mcheck.3:119
3527 msgid "B<MCHECK_HEAD>"
3531 #: build/C/man3/mcheck.3:122
3532 msgid "Memory preceding an allocated block was clobbered."
3536 #: build/C/man3/mcheck.3:122
3538 msgid "B<MCHECK_TAIL>"
3542 #: build/C/man3/mcheck.3:125
3543 msgid "Memory following an allocated block was clobbered."
3547 #: build/C/man3/mcheck.3:125
3549 msgid "B<MCHECK_FREE>"
3553 #: build/C/man3/mcheck.3:129
3554 msgid "A block of memory was freed twice."
3558 #: build/C/man3/mcheck.3:134
3559 msgid "B<mcheck>() and B<mcheck_pedantic>() return 0 on success, or -1 on error."
3563 #: build/C/man3/mcheck.3:145
3565 "The B<mcheck_pedantic>() and B<mcheck_check_all>() functions are available "
3566 "since glibc 2.2. The B<mcheck>() and B<mprobe>() functions are present "
3567 "since at least glibc 2.0"
3570 #. But is MALLOC_CHECK_ slower?
3572 #: build/C/man3/mcheck.3:159
3574 "Linking a program with I<-lmcheck> and using the B<MALLOC_CHECK_> "
3575 "environment variable (described in B<mallopt>(3)) cause the same kinds of "
3576 "errors to be detected. But, using B<MALLOC_CHECK_> does not require the "
3577 "application to be relinked."
3581 #: build/C/man3/mcheck.3:165
3583 "The program below calls B<mcheck>() with a NULL argument and then frees the "
3584 "same block of memory twice. The following shell session demonstrates what "
3585 "happens when running the program:"
3589 #: build/C/man3/mcheck.3:170
3597 #: build/C/man3/mcheck.3:174
3600 "About to free a second time\n"
3601 "block freed twice\n"
3602 "Aborted (core dumped)\n"
3606 #: build/C/man3/mcheck.3:182
3609 "#include E<lt>stdlib.hE<gt>\n"
3610 "#include E<lt>stdio.hE<gt>\n"
3611 "#include E<lt>mcheck.hE<gt>\n"
3615 #: build/C/man3/mcheck.3:190
3618 " if (mcheck(NULL) != 0) {\n"
3619 " fprintf(stderr, \"mcheck() failed\\en\");\n"
3623 #: build/C/man3/mcheck.3:193
3626 " exit(EXIT_FAILURE);\n"
3631 #: build/C/man3/mcheck.3:195
3633 msgid " p = malloc(1000);\n"
3637 #: build/C/man3/mcheck.3:200
3640 " fprintf(stderr, \"About to free\\en\");\n"
3642 " fprintf(stderr, \"\\enAbout to free a second time\\en\");\n"
3647 #: build/C/man3/mcheck.3:208
3648 msgid "B<malloc>(3), B<mallopt>(3), B<mtrace>(3)"
3652 #: build/C/man2/mlock.2:26
3658 #: build/C/man2/mlock.2:26
3664 #: build/C/man2/mlock.2:29
3665 msgid "mlock, munlock, mlockall, munlockall - lock and unlock memory"
3669 #: 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
3671 msgid "B<#include E<lt>sys/mman.hE<gt>>\n"
3675 #: build/C/man2/mlock.2:35
3678 "B<int mlock(const void *>I<addr>B<, size_t >I<len>B<);>\n"
3679 "B<int munlock(const void *>I<addr>B<, size_t >I<len>B<);>\n"
3683 #: build/C/man2/mlock.2:38
3686 "B<int mlockall(int >I<flags>B<);>\n"
3687 "B<int munlockall(void);>\n"
3691 #: build/C/man2/mlock.2:54
3693 "B<mlock>() and B<mlockall>() respectively lock part or all of the calling "
3694 "process's virtual address space into RAM, preventing that memory from being "
3695 "paged to the swap area. B<munlock>() and B<munlockall>() perform the "
3696 "converse operation, respectively unlocking part or all of the calling "
3697 "process's virtual address space, so that pages in the specified virtual "
3698 "address range may once more to be swapped out if required by the kernel "
3699 "memory manager. Memory locking and unlocking are performed in units of "
3704 #: build/C/man2/mlock.2:54
3706 msgid "mlock() and munlock()"
3710 #: build/C/man2/mlock.2:64
3712 "B<mlock>() locks pages in the address range starting at I<addr> and "
3713 "continuing for I<len> bytes. All pages that contain a part of the specified "
3714 "address range are guaranteed to be resident in RAM when the call returns "
3715 "successfully; the pages are guaranteed to stay in RAM until later unlocked."
3719 #: build/C/man2/mlock.2:73
3721 "B<munlock>() unlocks pages in the address range starting at I<addr> and "
3722 "continuing for I<len> bytes. After this call, all pages that contain a part "
3723 "of the specified memory range can be moved to external swap space again by "
3728 #: build/C/man2/mlock.2:73
3730 msgid "mlockall() and munlockall()"
3734 #: build/C/man2/mlock.2:83
3736 "B<mlockall>() locks all pages mapped into the address space of the calling "
3737 "process. This includes the pages of the code, data and stack segment, as "
3738 "well as shared libraries, user space kernel data, shared memory, and "
3739 "memory-mapped files. All mapped pages are guaranteed to be resident in RAM "
3740 "when the call returns successfully; the pages are guaranteed to stay in RAM "
3741 "until later unlocked."
3745 #: build/C/man2/mlock.2:88
3747 "The I<flags> argument is constructed as the bitwise OR of one or more of the "
3748 "following constants:"
3752 #: build/C/man2/mlock.2:88
3754 msgid "B<MCL_CURRENT>"
3758 #: build/C/man2/mlock.2:92
3760 "Lock all pages which are currently mapped into the address space of the "
3765 #: build/C/man2/mlock.2:92
3767 msgid "B<MCL_FUTURE>"
3771 #: build/C/man2/mlock.2:99
3773 "Lock all pages which will become mapped into the address space of the "
3774 "process in the future. These could be for instance new pages required by a "
3775 "growing heap and stack as well as new memory mapped files or shared memory "
3780 #: build/C/man2/mlock.2:112
3782 "If B<MCL_FUTURE> has been specified, then a later system call (e.g., "
3783 "B<mmap>(2), B<sbrk>(2), B<malloc>(3)), may fail if it would cause the number "
3784 "of locked bytes to exceed the permitted maximum (see below). In the same "
3785 "circumstances, stack growth may likewise fail: the kernel will deny stack "
3786 "expansion and deliver a B<SIGSEGV> signal to the process."
3790 #: build/C/man2/mlock.2:116
3792 "B<munlockall>() unlocks all pages mapped into the address space of the "
3797 #: build/C/man2/mlock.2:122
3799 "On success these system calls return 0. On error, -1 is returned, I<errno> "
3800 "is set appropriately, and no changes are made to any locks in the address "
3801 "space of the process."
3805 #: build/C/man2/mlock.2:131
3807 "(Linux 2.6.9 and later) the caller had a nonzero B<RLIMIT_MEMLOCK> soft "
3808 "resource limit, but tried to lock more memory than the limit permitted. "
3809 "This limit is not enforced if the process is privileged (B<CAP_IPC_LOCK>)."
3812 #. In the case of mlock(), this check is somewhat buggy: it doesn't
3813 #. take into account whether the to-be-locked range overlaps with
3814 #. already locked pages. Thus, suppose we allocate
3815 #. (num_physpages / 4 + 1) of memory, and lock those pages once using
3816 #. mlock(), and then lock the *same* page range a second time.
3817 #. In the case, the second mlock() call will fail, since the check
3818 #. calculates that the process is trying to lock (num_physpages / 2 + 2)
3819 #. pages, which of course is not true. (MTK, Nov 04, kernel 2.4.28)
3821 #: build/C/man2/mlock.2:143
3823 "(Linux 2.4 and earlier) the calling process tried to lock more than half of "
3827 #. SVr4 documents an additional EAGAIN error code.
3829 #: build/C/man2/mlock.2:149
3831 "The caller is not privileged, but needs privilege (B<CAP_IPC_LOCK>) to "
3832 "perform the requested operation."
3836 #: build/C/man2/mlock.2:154
3837 msgid "For B<mlock>() and B<munlock>():"
3841 #: build/C/man2/mlock.2:157
3842 msgid "Some or all of the specified address range could not be locked."
3846 #: build/C/man2/mlock.2:164
3848 "The result of the addition I<start>+I<len> was less than I<start> (e.g., the "
3849 "addition may have resulted in an overflow)."
3853 #: build/C/man2/mlock.2:169
3854 msgid "(Not on Linux) I<addr> was not a multiple of the page size."
3858 #: build/C/man2/mlock.2:173
3860 "Some of the specified address range does not correspond to mapped pages in "
3861 "the address space of the process."
3865 #: build/C/man2/mlock.2:176
3866 msgid "For B<mlockall>():"
3870 #: build/C/man2/mlock.2:179
3871 msgid "Unknown I<flags> were specified."
3875 #: build/C/man2/mlock.2:182
3876 msgid "For B<munlockall>():"
3880 #: build/C/man2/mlock.2:186
3881 msgid "(Linux 2.6.8 and earlier) The caller was not privileged (B<CAP_IPC_LOCK>)."
3885 #: build/C/man2/mlock.2:188
3886 msgid "POSIX.1-2001, SVr4."
3890 #: build/C/man2/mlock.2:188 build/C/man2/mmap.2:501 build/C/man2/msync.2:105
3892 msgid "AVAILABILITY"
3896 #: build/C/man2/mlock.2:200
3898 "On POSIX systems on which B<mlock>() and B<munlock>() are available, "
3899 "B<_POSIX_MEMLOCK_RANGE> is defined in I<E<lt>unistd.hE<gt>> and the number "
3900 "of bytes in a page can be determined from the constant B<PAGESIZE> (if "
3901 "defined) in I<E<lt>limits.hE<gt>> or by calling I<sysconf(_SC_PAGESIZE)>."
3904 #. POSIX.1-2001: It shall be defined to -1 or 0 or 200112L.
3905 #. -1: unavailable, 0: ask using sysconf().
3906 #. glibc defines it to 1.
3908 #: build/C/man2/mlock.2:213
3910 "On POSIX systems on which B<mlockall>() and B<munlockall>() are available, "
3911 "B<_POSIX_MEMLOCK> is defined in I<E<lt>unistd.hE<gt>> to a value greater "
3912 "than 0. (See also B<sysconf>(3).)"
3916 #: build/C/man2/mlock.2:231
3918 "Memory locking has two main applications: real-time algorithms and "
3919 "high-security data processing. Real-time applications require deterministic "
3920 "timing, and, like scheduling, paging is one major cause of unexpected "
3921 "program execution delays. Real-time applications will usually also switch "
3922 "to a real-time scheduler with B<sched_setscheduler>(2). Cryptographic "
3923 "security software often handles critical bytes like passwords or secret keys "
3924 "as data structures. As a result of paging, these secrets could be "
3925 "transferred onto a persistent swap store medium, where they might be "
3926 "accessible to the enemy long after the security software has erased the "
3927 "secrets in RAM and terminated. (But be aware that the suspend mode on "
3928 "laptops and some desktop computers will save a copy of the system's RAM to "
3929 "disk, regardless of memory locks.)"
3933 #: build/C/man2/mlock.2:244
3935 "Real-time processes that are using B<mlockall>() to prevent delays on page "
3936 "faults should reserve enough locked stack pages before entering the "
3937 "time-critical section, so that no page fault can be caused by function "
3938 "calls. This can be achieved by calling a function that allocates a "
3939 "sufficiently large automatic variable (an array) and writes to the memory "
3940 "occupied by this array in order to touch these stack pages. This way, "
3941 "enough pages will be mapped for the stack and can be locked into RAM. The "
3942 "dummy writes ensure that not even copy-on-write page faults can occur in the "
3947 #: build/C/man2/mlock.2:250
3949 "Memory locks are not inherited by a child created via B<fork>(2) and are "
3950 "automatically removed (unlocked) during an B<execve>(2) or when the process "
3955 #: build/C/man2/mlock.2:254
3957 "The memory lock on an address range is automatically removed if the address "
3958 "range is unmapped via B<munmap>(2)."
3962 #: build/C/man2/mlock.2:267
3964 "Memory locks do not stack, that is, pages which have been locked several "
3965 "times by calls to B<mlock>() or B<mlockall>() will be unlocked by a single "
3966 "call to B<munlock>() for the corresponding range or by B<munlockall>(). "
3967 "Pages which are mapped to several locations or by several processes stay "
3968 "locked into RAM as long as they are locked at least at one location or by at "
3969 "least one process."
3973 #: build/C/man2/mlock.2:278
3975 "Under Linux, B<mlock>() and B<munlock>() automatically round I<addr> down "
3976 "to the nearest page boundary. However, POSIX.1-2001 allows an "
3977 "implementation to require that I<addr> is page aligned, so portable "
3978 "applications should ensure this."
3982 #: build/C/man2/mlock.2:291
3984 "The I<VmLck> field of the Linux-specific I</proc/PID/status> file shows how "
3985 "many kilobytes of memory the process with ID I<PID> has locked using "
3986 "B<mlock>(), B<mlockall>(), and B<mmap>(2) B<MAP_LOCKED>."
3990 #: build/C/man2/mlock.2:291
3992 msgid "Limits and permissions"
3996 #: build/C/man2/mlock.2:298
3998 "In Linux 2.6.8 and earlier, a process must be privileged (B<CAP_IPC_LOCK>) "
3999 "in order to lock memory and the B<RLIMIT_MEMLOCK> soft resource limit "
4000 "defines a limit on how much memory the process may lock."
4004 #: build/C/man2/mlock.2:304
4006 "Since Linux 2.6.9, no limits are placed on the amount of memory that a "
4007 "privileged process can lock and the B<RLIMIT_MEMLOCK> soft resource limit "
4008 "instead defines a limit on how much memory an unprivileged process may lock."
4012 #: build/C/man2/mlock.2:312
4014 "In the 2.4 series Linux kernels up to and including 2.4.17, a bug caused the "
4015 "B<mlockall>() B<MCL_FUTURE> flag to be inherited across a B<fork>(2). This "
4016 "was rectified in kernel 2.4.18."
4019 #. See the following LKML thread:
4020 #. http://marc.theaimsgroup.com/?l=linux-kernel&m=113801392825023&w=2
4021 #. "Rationale for RLIMIT_MEMLOCK"
4024 #: build/C/man2/mlock.2:329
4026 "Since kernel 2.6.9, if a privileged process calls I<mlockall(MCL_FUTURE)> "
4027 "and later drops privileges (loses the B<CAP_IPC_LOCK> capability by, for "
4028 "example, setting its effective UID to a nonzero value), then subsequent "
4029 "memory allocations (e.g., B<mmap>(2), B<brk>(2)) will fail if the "
4030 "B<RLIMIT_MEMLOCK> resource limit is encountered."
4034 #: build/C/man2/mlock.2:336
4036 "B<mmap>(2), B<setrlimit>(2), B<shmctl>(2), B<sysconf>(3), B<proc>(5), "
4037 "B<capabilities>(7)"
4041 #: build/C/man2/mmap.2:40
4047 #: build/C/man2/mmap.2:40
4053 #: build/C/man2/mmap.2:43
4054 msgid "mmap, munmap - map or unmap files or devices into memory"
4058 #: build/C/man2/mmap.2:51
4061 "B<void *mmap(void *>I<addr>B<, size_t >I<length>B<, int >I<prot>B<, int "
4063 "B< int >I<fd>B<, off_t >I<offset>B<);>\n"
4064 "B<int munmap(void *>I<addr>B<, size_t >I<length>B<);>\n"
4068 #: build/C/man2/mmap.2:54
4069 msgid "See NOTES for information on feature test macro requirements."
4073 #: build/C/man2/mmap.2:63
4075 "B<mmap>() creates a new mapping in the virtual address space of the calling "
4076 "process. The starting address for the new mapping is specified in I<addr>. "
4077 "The I<length> argument specifies the length of the mapping."
4080 #. Before Linux 2.6.24, the address was rounded up to the next page
4081 #. boundary; since 2.6.24, it is rounded down!
4083 #: build/C/man2/mmap.2:77
4085 "If I<addr> is NULL, then the kernel chooses the address at which to create "
4086 "the mapping; this is the most portable method of creating a new mapping. If "
4087 "I<addr> is not NULL, then the kernel takes it as a hint about where to place "
4088 "the mapping; on Linux, the mapping will be created at a nearby page "
4089 "boundary. The address of the new mapping is returned as the result of the "
4094 #: build/C/man2/mmap.2:89
4096 "The contents of a file mapping (as opposed to an anonymous mapping; see "
4097 "B<MAP_ANONYMOUS> below), are initialized using I<length> bytes starting at "
4098 "offset I<offset> in the file (or other object) referred to by the file "
4099 "descriptor I<fd>. I<offset> must be a multiple of the page size as returned "
4100 "by I<sysconf(_SC_PAGE_SIZE)>."
4104 #: build/C/man2/mmap.2:97
4106 "The I<prot> argument describes the desired memory protection of the mapping "
4107 "(and must not conflict with the open mode of the file). It is either "
4108 "B<PROT_NONE> or the bitwise OR of one or more of the following flags:"
4112 #: build/C/man2/mmap.2:97 build/C/man2/mprotect.2:73
4114 msgid "B<PROT_EXEC>"
4118 #: build/C/man2/mmap.2:100
4119 msgid "Pages may be executed."
4123 #: build/C/man2/mmap.2:100 build/C/man2/mprotect.2:67
4125 msgid "B<PROT_READ>"
4129 #: build/C/man2/mmap.2:103
4130 msgid "Pages may be read."
4134 #: build/C/man2/mmap.2:103 build/C/man2/mprotect.2:70
4136 msgid "B<PROT_WRITE>"
4140 #: build/C/man2/mmap.2:106
4141 msgid "Pages may be written."
4145 #: build/C/man2/mmap.2:106 build/C/man2/mprotect.2:64
4147 msgid "B<PROT_NONE>"
4151 #: build/C/man2/mmap.2:109
4152 msgid "Pages may not be accessed."
4156 #: build/C/man2/mmap.2:118
4158 "The I<flags> argument determines whether updates to the mapping are visible "
4159 "to other processes mapping the same region, and whether updates are carried "
4160 "through to the underlying file. This behavior is determined by including "
4161 "exactly one of the following values in I<flags>:"
4165 #: build/C/man2/mmap.2:118
4167 msgid "B<MAP_SHARED>"
4171 #: build/C/man2/mmap.2:128
4173 "Share this mapping. Updates to the mapping are visible to other processes "
4174 "that map this file, and are carried through to the underlying file. The "
4175 "file may not actually be updated until B<msync>(2) or B<munmap>() is "
4180 #: build/C/man2/mmap.2:128
4182 msgid "B<MAP_PRIVATE>"
4186 #: build/C/man2/mmap.2:137
4188 "Create a private copy-on-write mapping. Updates to the mapping are not "
4189 "visible to other processes mapping the same file, and are not carried "
4190 "through to the underlying file. It is unspecified whether changes made to "
4191 "the file after the B<mmap>() call are visible in the mapped region."
4195 #: build/C/man2/mmap.2:139
4196 msgid "Both of these flags are described in POSIX.1-2001."
4200 #: build/C/man2/mmap.2:142
4201 msgid "In addition, zero or more of the following values can be ORed in I<flags>:"
4205 #: build/C/man2/mmap.2:142
4207 msgid "B<MAP_32BIT> (since Linux 2.4.20, 2.6)"
4210 #. See http://lwn.net/Articles/294642 "Tangled up in threads", 19 Aug 08
4212 #: build/C/man2/mmap.2:158
4214 "Put the mapping into the first 2 Gigabytes of the process address space. "
4215 "This flag is supported only on x86-64, for 64-bit programs. It was added to "
4216 "allow thread stacks to be allocated somewhere in the first 2GB of memory, so "
4217 "as to improve context-switch performance on some early 64-bit processors. "
4218 "Modern x86-64 processors no longer have this performance problem, so use of "
4219 "this flag is not required on those systems. The B<MAP_32BIT> flag is "
4220 "ignored when B<MAP_FIXED> is set."
4224 #: build/C/man2/mmap.2:158
4230 #: build/C/man2/mmap.2:163
4231 msgid "Synonym for B<MAP_ANONYMOUS>. Deprecated."
4235 #: build/C/man2/mmap.2:163
4237 msgid "B<MAP_ANONYMOUS>"
4241 #: build/C/man2/mmap.2:185
4243 "The mapping is not backed by any file; its contents are initialized to "
4244 "zero. The I<fd> and I<offset> arguments are ignored; however, some "
4245 "implementations require I<fd> to be -1 if B<MAP_ANONYMOUS> (or B<MAP_ANON>) "
4246 "is specified, and portable applications should ensure this. The use of "
4247 "B<MAP_ANONYMOUS> in conjunction with B<MAP_SHARED> is supported on Linux "
4248 "only since kernel 2.4."
4252 #: build/C/man2/mmap.2:185
4254 msgid "B<MAP_DENYWRITE>"
4257 #. Introduced in 1.1.36, removed in 1.3.24.
4259 #: build/C/man2/mmap.2:193
4261 "This flag is ignored. (Long ago, it signaled that attempts to write to the "
4262 "underlying file should fail with B<ETXTBUSY>. But this was a source of "
4263 "denial-of-service attacks.)"
4267 #: build/C/man2/mmap.2:193
4269 msgid "B<MAP_EXECUTABLE>"
4272 #. Introduced in 1.1.38, removed in 1.3.24. Flag tested in proc_follow_link.
4273 #. (Long ago, it signaled that the underlying file is an executable.
4274 #. However, that information was not really used anywhere.)
4275 #. Linus talked about DOS related to MAP_EXECUTABLE, but he was thinking of
4278 #: build/C/man2/mmap.2:201
4279 msgid "This flag is ignored."
4283 #: build/C/man2/mmap.2:201
4288 #. On some systems, this was required as the opposite of
4289 #. MAP_ANONYMOUS -- mtk, 1 May 2007
4291 #: build/C/man2/mmap.2:207
4292 msgid "Compatibility flag. Ignored."
4296 #: build/C/man2/mmap.2:207
4298 msgid "B<MAP_FIXED>"
4302 #: build/C/man2/mmap.2:225
4304 "Don't interpret I<addr> as a hint: place the mapping at exactly that "
4305 "address. I<addr> must be a multiple of the page size. If the memory region "
4306 "specified by I<addr> and I<len> overlaps pages of any existing mapping(s), "
4307 "then the overlapped part of the existing mapping(s) will be discarded. If "
4308 "the specified address cannot be used, B<mmap>() will fail. Because "
4309 "requiring a fixed address for a mapping is less portable, the use of this "
4310 "option is discouraged."
4314 #: build/C/man2/mmap.2:225
4316 msgid "B<MAP_GROWSDOWN>"
4320 #: build/C/man2/mmap.2:230
4322 "Used for stacks. Indicates to the kernel virtual memory system that the "
4323 "mapping should extend downward in memory."
4327 #: build/C/man2/mmap.2:230
4329 msgid "B<MAP_HUGETLB> (since Linux 2.6.32)"
4333 #: build/C/man2/mmap.2:236
4335 "Allocate the mapping using \"huge pages.\" See the Linux kernel source file "
4336 "I<Documentation/vm/hugetlbpage.txt> for further information."
4340 #: build/C/man2/mmap.2:236
4342 msgid "B<MAP_LOCKED> (since Linux 2.5.37)"
4345 #. If set, the mapped pages will not be swapped out.
4347 #: build/C/man2/mmap.2:242
4349 "Lock the pages of the mapped region into memory in the manner of "
4350 "B<mlock>(2). This flag is ignored in older kernels."
4354 #: build/C/man2/mmap.2:242
4356 msgid "B<MAP_NONBLOCK> (since Linux 2.5.46)"
4360 #: build/C/man2/mmap.2:257
4362 "Only meaningful in conjunction with B<MAP_POPULATE>. Don't perform "
4363 "read-ahead: create page tables entries only for pages that are already "
4364 "present in RAM. Since Linux 2.6.23, this flag causes B<MAP_POPULATE> to do "
4365 "nothing. One day the combination of B<MAP_POPULATE> and B<MAP_NONBLOCK> may "
4370 #: build/C/man2/mmap.2:257
4372 msgid "B<MAP_NORESERVE>"
4376 #: build/C/man2/mmap.2:272
4378 "Do not reserve swap space for this mapping. When swap space is reserved, "
4379 "one has the guarantee that it is possible to modify the mapping. When swap "
4380 "space is not reserved one might get B<SIGSEGV> upon a write if no physical "
4381 "memory is available. See also the discussion of the file "
4382 "I</proc/sys/vm/overcommit_memory> in B<proc>(5). In kernels before 2.6, "
4383 "this flag had effect only for private writable mappings."
4387 #: build/C/man2/mmap.2:272
4389 msgid "B<MAP_POPULATE> (since Linux 2.5.46)"
4393 #: build/C/man2/mmap.2:279
4395 "Populate (prefault) page tables for a mapping. For a file mapping, this "
4396 "causes read-ahead on the file. Later accesses to the mapping will not be "
4397 "blocked by page faults. B<MAP_POPULATE> is supported for private mappings "
4398 "only since Linux 2.6.23."
4402 #: build/C/man2/mmap.2:279
4404 msgid "B<MAP_STACK> (since Linux 2.6.27)"
4407 #. See http://lwn.net/Articles/294642 "Tangled up in threads", 19 Aug 08
4408 #. commit cd98a04a59e2f94fa64d5bf1e26498d27427d5e7
4409 #. http://thread.gmane.org/gmane.linux.kernel/720412
4410 #. "pthread_create() slow for many threads; also time to revisit 64b
4411 #. context switch optimization?"
4413 #: build/C/man2/mmap.2:292
4415 "Allocate the mapping at an address suitable for a process or thread stack. "
4416 "This flag is currently a no-op, but is used in the glibc threading "
4417 "implementation so that if some architectures require special treatment for "
4418 "stack allocations, support can later be transparently implemented for glibc."
4422 #: build/C/man2/mmap.2:292
4424 msgid "B<MAP_UNINITIALIZED> (since Linux 2.6.33)"
4428 #: build/C/man2/mmap.2:302
4430 "Don't clear anonymous pages. This flag is intended to improve performance "
4431 "on embedded devices. This flag is honored only if the kernel was configured "
4432 "with the B<CONFIG_MMAP_ALLOW_UNINITIALIZED> option. Because of the security "
4433 "implications, that option is normally enabled only on embedded devices "
4434 "(i.e., devices where one has complete control of the contents of user "
4439 #: build/C/man2/mmap.2:310
4441 "Of the above flags, only B<MAP_FIXED> is specified in POSIX.1-2001. "
4442 "However, most systems also support B<MAP_ANONYMOUS> (or its synonym "
4447 #: build/C/man2/mmap.2:317
4449 "Some systems document the additional flags B<MAP_AUTOGROW>, "
4450 "B<MAP_AUTORESRV>, B<MAP_COPY>, and B<MAP_LOCAL>."
4454 #: build/C/man2/mmap.2:323
4456 "Memory mapped by B<mmap>() is preserved across B<fork>(2), with the same "
4461 #: build/C/man2/mmap.2:331
4463 "A file is mapped in multiples of the page size. For a file that is not a "
4464 "multiple of the page size, the remaining memory is zeroed when mapped, and "
4465 "writes to that region are not written out to the file. The effect of "
4466 "changing the size of the underlying file of a mapping on the pages that "
4467 "correspond to added or removed regions of the file is unspecified."
4471 #: build/C/man2/mmap.2:331
4477 #: build/C/man2/mmap.2:341
4479 "The B<munmap>() system call deletes the mappings for the specified address "
4480 "range, and causes further references to addresses within the range to "
4481 "generate invalid memory references. The region is also automatically "
4482 "unmapped when the process is terminated. On the other hand, closing the "
4483 "file descriptor does not unmap the region."
4487 #: build/C/man2/mmap.2:351
4489 "The address I<addr> must be a multiple of the page size. All pages "
4490 "containing a part of the indicated range are unmapped, and subsequent "
4491 "references to these pages will generate B<SIGSEGV>. It is not an error if "
4492 "the indicated range does not contain any mapped pages."
4496 #: build/C/man2/mmap.2:351
4498 msgid "Timestamps changes for file-backed mappings"
4502 #: build/C/man2/mmap.2:358
4504 "For file-backed mappings, the I<st_atime> field for the mapped file may be "
4505 "updated at any time between the B<mmap>() and the corresponding unmapping; "
4506 "the first reference to a mapped page will update the field if it has not "
4511 #: build/C/man2/mmap.2:375
4513 "The I<st_ctime> and I<st_mtime> field for a file mapped with B<PROT_WRITE> "
4514 "and B<MAP_SHARED> will be updated after a write to the mapped region, and "
4515 "before a subsequent B<msync>(2) with the B<MS_SYNC> or B<MS_ASYNC> flag, if "
4520 #: build/C/man2/mmap.2:392
4522 "On success, B<mmap>() returns a pointer to the mapped area. On error, the "
4523 "value B<MAP_FAILED> (that is, I<(void\\ *)\\ -1>) is returned, and I<errno> "
4524 "is set appropriately. On success, B<munmap>() returns 0, on failure -1, "
4525 "and I<errno> is set (probably to B<EINVAL>)."
4529 #: build/C/man2/mmap.2:393 build/C/man2/mprotect.2:86 build/C/man3/shm_open.3:187 build/C/man3/shm_open.3:192 build/C/man2/shmctl.2:302 build/C/man2/shmget.2:189 build/C/man2/shmop.2:191
4535 #: build/C/man2/mmap.2:413
4537 "A file descriptor refers to a non-regular file. Or B<MAP_PRIVATE> was "
4538 "requested, but I<fd> is not open for reading. Or B<MAP_SHARED> was "
4539 "requested and B<PROT_WRITE> is set, but I<fd> is not open in read/write "
4540 "(B<O_RDWR>) mode. Or B<PROT_WRITE> is set, but the file is append-only."
4544 #: build/C/man2/mmap.2:417
4546 "The file has been locked, or too much memory has been locked (see "
4551 #: build/C/man2/mmap.2:423
4552 msgid "I<fd> is not a valid file descriptor (and B<MAP_ANONYMOUS> was not set)."
4556 #: build/C/man2/mmap.2:431
4558 "We don't like I<addr>, I<length>, or I<offset> (e.g., they are too large, or "
4559 "not aligned on a page boundary)."
4563 #: build/C/man2/mmap.2:436
4564 msgid "(since Linux 2.6.12) I<length> was 0."
4568 #: build/C/man2/mmap.2:444
4570 "I<flags> contained neither B<MAP_PRIVATE> or B<MAP_SHARED>, or contained "
4571 "both of these values."
4575 #: build/C/man2/mmap.2:444 build/C/man3/shm_open.3:229 build/C/man2/shmget.2:205
4580 #. [2.6.7] shmem_zero_setup()-->shmem_file_setup()-->get_empty_filp()
4582 #: build/C/man2/mmap.2:452 build/C/man2/shmget.2:209
4583 msgid "The system limit on the total number of open files has been reached."
4587 #: build/C/man2/mmap.2:456
4589 "The underlying filesystem of the specified file does not support memory "
4594 #: build/C/man2/mmap.2:460
4596 "No memory is available, or the process's maximum number of mappings would "
4597 "have been exceeded."
4600 #. (Since 2.4.25 / 2.6.0.)
4602 #: build/C/man2/mmap.2:469
4604 "The I<prot> argument asks for B<PROT_EXEC> but the mapped area belongs to a "
4605 "file on a filesystem that was mounted no-exec."
4609 #: build/C/man2/mmap.2:469
4615 #: build/C/man2/mmap.2:475
4617 "B<MAP_DENYWRITE> was set but the object specified by I<fd> is open for "
4622 #: build/C/man2/mmap.2:475 build/C/man2/shmctl.2:344
4624 msgid "B<EOVERFLOW>"
4628 #: build/C/man2/mmap.2:487
4630 "On 32-bit architecture together with the large file extension (i.e., using "
4631 "64-bit I<off_t>): the number of pages used for I<length> plus number of "
4632 "pages used for I<offset> would overflow I<unsigned long> (32 bits)."
4636 #: build/C/man2/mmap.2:489
4637 msgid "Use of a mapped region can result in these signals:"
4641 #: build/C/man2/mmap.2:489
4647 #: build/C/man2/mmap.2:492
4648 msgid "Attempted write into a region mapped as read-only."
4652 #: build/C/man2/mmap.2:492
4658 #: build/C/man2/mmap.2:497
4660 "Attempted access to a portion of the buffer that does not correspond to the "
4661 "file (for example, beyond the end of the file, including the case where "
4662 "another process has truncated the file)."
4665 #. SVr4 documents additional error codes ENXIO and ENODEV.
4666 #. SUSv2 documents additional error codes EMFILE and EOVERFLOW.
4668 #: build/C/man2/mmap.2:501
4669 msgid "SVr4, 4.4BSD, POSIX.1-2001."
4672 #. POSIX.1-2001: It shall be defined to -1 or 0 or 200112L.
4673 #. -1: unavailable, 0: ask using sysconf().
4674 #. glibc defines it to 1.
4676 #: build/C/man2/mmap.2:515
4678 "On POSIX systems on which B<mmap>(), B<msync>(2) and B<munmap>() are "
4679 "available, B<_POSIX_MAPPED_FILES> is defined in I<E<lt>unistd.hE<gt>> to a "
4680 "value greater than 0. (See also B<sysconf>(3).)"
4683 #. Since around glibc 2.1/2.2, depending on the platform.
4685 #: build/C/man2/mmap.2:530
4687 "This page describes the interface provided by the glibc B<mmap>() wrapper "
4688 "function. Originally, this function invoked a system call of the same "
4689 "name. Since kernel 2.4, that system call has been superseded by "
4690 "B<mmap2>(2), and nowadays the glibc B<mmap>() wrapper function invokes "
4691 "B<mmap2>(2) with a suitably adjusted value for I<offset>."
4695 #: build/C/man2/mmap.2:543
4697 "On some hardware architectures (e.g., i386), B<PROT_WRITE> implies "
4698 "B<PROT_READ>. It is architecture dependent whether B<PROT_READ> implies "
4699 "B<PROT_EXEC> or not. Portable programs should always set B<PROT_EXEC> if "
4700 "they intend to execute code in the new mapping."
4704 #: build/C/man2/mmap.2:558
4706 "The portable way to create a mapping is to specify I<addr> as 0 (NULL), and "
4707 "omit B<MAP_FIXED> from I<flags>. In this case, the system chooses the "
4708 "address for the mapping; the address is chosen so as not to conflict with "
4709 "any existing mapping, and will not be 0. If the B<MAP_FIXED> flag is "
4710 "specified, and I<addr> is 0 (NULL), then the mapped address will be 0 "
4715 #: build/C/man2/mmap.2:587
4717 "Certain I<flags> constants are defined only if either B<_BSD_SOURCE> or "
4718 "B<_SVID_SOURCE> is defined. (Requiring B<_GNU_SOURCE> also suffices, and "
4719 "requiring that macro specifically would have been more logical, since these "
4720 "flags are all Linux specific.) The relevant flags are: B<MAP_32BIT>, "
4721 "B<MAP_ANONYMOUS> (and the synonym B<MAP_ANON>), B<MAP_DENYWRITE>, "
4722 "B<MAP_EXECUTABLE>, B<MAP_FILE>, B<MAP_GROWSDOWN>, B<MAP_HUGETLB>, "
4723 "B<MAP_LOCKED>, B<MAP_NONBLOCK>, B<MAP_NORESERVE>, B<MAP_POPULATE>, and "
4728 #: build/C/man2/mmap.2:592
4730 "On Linux there are no guarantees like those suggested above under "
4731 "B<MAP_NORESERVE>. By default, any process can be killed at any moment when "
4732 "the system runs out of memory."
4736 #: build/C/man2/mmap.2:599
4738 "In kernels before 2.6.7, the B<MAP_POPULATE> flag has effect only if I<prot> "
4739 "is specified as B<PROT_NONE>."
4743 #: build/C/man2/mmap.2:614
4745 "SUSv3 specifies that B<mmap>() should fail if I<length> is 0. However, in "
4746 "kernels before 2.6.12, B<mmap>() succeeded in this case: no mapping was "
4747 "created and the call returned I<addr>. Since kernel 2.6.12, B<mmap>() "
4748 "fails with the error B<EINVAL> for this case."
4752 #: build/C/man2/mmap.2:630
4754 "POSIX specifies that the system shall always zero fill any partial page at "
4755 "the end of the object and that system will never write any modification of "
4756 "the object beyond its end. On Linux, when you write data to such partial "
4757 "page after the end of the object, the data stays in the page cache even "
4758 "after the file is closed and unmapped and even though the data is never "
4759 "written to the file itself, subsequent mappings may see the modified "
4760 "content. In some cases, this could be fixed by calling B<msync>(2) before "
4761 "the unmap takes place; however, this doesn't work on tmpfs (for example, "
4762 "when using POSIX shared memory interface documented in B<shm_overview>(7))."
4766 #: build/C/man2/mmap.2:642
4768 "The following program prints part of the file specified in its first "
4769 "command-line argument to standard output. The range of bytes to be printed "
4770 "is specified via offset and length values in the second and third "
4771 "command-line arguments. The program creates a memory mapping of the "
4772 "required pages of the file and then uses B<write>(2) to output the desired "
4777 #: build/C/man2/mmap.2:650
4780 "#include E<lt>sys/mman.hE<gt>\n"
4781 "#include E<lt>sys/stat.hE<gt>\n"
4782 "#include E<lt>fcntl.hE<gt>\n"
4783 "#include E<lt>stdio.hE<gt>\n"
4784 "#include E<lt>stdlib.hE<gt>\n"
4785 "#include E<lt>unistd.hE<gt>\n"
4789 #: build/C/man2/mmap.2:653 build/C/man2/mprotect.2:178
4792 "#define handle_error(msg) \\e\n"
4793 " do { perror(msg); exit(EXIT_FAILURE); } while (0)\n"
4797 #: build/C/man2/mmap.2:663
4801 "main(int argc, char *argv[])\n"
4805 " struct stat sb;\n"
4806 " off_t offset, pa_offset;\n"
4812 #: build/C/man2/mmap.2:668
4815 " if (argc E<lt> 3 || argc E<gt> 4) {\n"
4816 " fprintf(stderr, \"%s file offset [length]\\en\", argv[0]);\n"
4817 " exit(EXIT_FAILURE);\n"
4822 #: build/C/man2/mmap.2:672
4825 " fd = open(argv[1], O_RDONLY);\n"
4827 " handle_error(\"open\");\n"
4831 #: build/C/man2/mmap.2:675
4834 " if (fstat(fd, &sb) == -1) /* To obtain file size */\n"
4835 " handle_error(\"fstat\");\n"
4839 #: build/C/man2/mmap.2:679
4842 " offset = atoi(argv[2]);\n"
4843 " pa_offset = offset & ~(sysconf(_SC_PAGE_SIZE) - 1);\n"
4844 " /* offset for mmap() must be page aligned */\n"
4848 #: build/C/man2/mmap.2:684
4851 " if (offset E<gt>= sb.st_size) {\n"
4852 " fprintf(stderr, \"offset is past end of file\\en\");\n"
4853 " exit(EXIT_FAILURE);\n"
4858 #: build/C/man2/mmap.2:690
4861 " if (argc == 4) {\n"
4862 " length = atoi(argv[3]);\n"
4863 " if (offset + length E<gt> sb.st_size)\n"
4864 " length = sb.st_size - offset;\n"
4865 " /* Can\\(aqt display bytes past end of file */\n"
4869 #: build/C/man2/mmap.2:694
4872 " } else { /* No length arg ==E<gt> display to end of file */\n"
4873 " length = sb.st_size - offset;\n"
4878 #: build/C/man2/mmap.2:699
4881 " addr = mmap(NULL, length + offset - pa_offset, PROT_READ,\n"
4882 " MAP_PRIVATE, fd, pa_offset);\n"
4883 " if (addr == MAP_FAILED)\n"
4884 " handle_error(\"mmap\");\n"
4888 #: build/C/man2/mmap.2:704
4891 " s = write(STDOUT_FILENO, addr + offset - pa_offset, length);\n"
4892 " if (s != length) {\n"
4894 " handle_error(\"write\");\n"
4898 #: build/C/man2/mmap.2:708
4901 " fprintf(stderr, \"partial write\");\n"
4902 " exit(EXIT_FAILURE);\n"
4907 #: build/C/man2/mmap.2:725
4909 "B<getpagesize>(2), B<mincore>(2), B<mlock>(2), B<mmap2>(2), B<mprotect>(2), "
4910 "B<mremap>(2), B<msync>(2), B<remap_file_pages>(2), B<setrlimit>(2), "
4911 "B<shmat>(2), B<shm_open>(3), B<shm_overview>(7)"
4915 #: build/C/man2/mmap.2:732
4917 "The descriptions of the following files in B<proc>(5): I</proc/[pid]/maps>, "
4918 "I</proc/[pid]/map_files>, and I</proc/[pid]/smaps>."
4922 #: build/C/man2/mmap.2:737 build/C/man2/msync.2:124
4923 msgid "B.O. Gallmeister, POSIX.4, O'Reilly, pp. 128-129 and 389-391."
4927 #: build/C/man2/mmap2.2:29
4933 #: build/C/man2/mmap2.2:29
4939 #: build/C/man2/mmap2.2:32
4940 msgid "mmap2 - map files or devices into memory"
4944 #: build/C/man2/mmap2.2:38
4947 "B<void *mmap2(void *>I<addr>B<, size_t >I<length>B<, int >I<prot>B<,>\n"
4948 "B< int >I<flags>B<, int >I<fd>B<, off_t >I<pgoffset>B<);>\n"
4952 #: build/C/man2/mmap2.2:43
4954 "This is probably not the system call that you are interested in; instead, "
4955 "see B<mmap>(2), which describes the glibc wrapper function that invokes this "
4960 #: build/C/man2/mmap2.2:54
4962 "The B<mmap2>() system call provides the same interface as B<mmap>(2), "
4963 "except that the final argument specifies the offset into the file in "
4964 "4096-byte units (instead of bytes, as is done by B<mmap>(2)). This enables "
4965 "applications that use a 32-bit I<off_t> to map large files (up to 2^44 "
4970 #: build/C/man2/mmap2.2:61
4972 "On success, B<mmap2>() returns a pointer to the mapped area. On error -1 "
4973 "is returned and I<errno> is set appropriately."
4977 #: build/C/man2/mmap2.2:65
4978 msgid "Problem with getting the data from user space."
4982 #: build/C/man2/mmap2.2:70
4984 "(Various platforms where the page size is not 4096 bytes.) I<offset\\ *\\ "
4985 "4096> is not a multiple of the system page size."
4989 #: build/C/man2/mmap2.2:74
4990 msgid "B<mmap2>() can also return any of the errors described in B<mmap>(2)."
4994 #: build/C/man2/mmap2.2:77
4995 msgid "B<mmap2>() is available since Linux 2.3.31."
4999 #: build/C/man2/mmap2.2:79 build/C/man2/subpage_prot.2:98
5000 msgid "This system call is Linux-specific."
5004 #: build/C/man2/mmap2.2:85
5006 "Nowadays, the glibc B<mmap>() wrapper function invokes this system call "
5007 "rather than the B<mmap>(2) system call."
5010 #. ia64 can have page sizes ranging from 4kB to 64kB.
5011 #. On cris, it looks like the unit might also be the page size,
5012 #. which is 8192 bytes. -- mtk, June 2007
5014 #: build/C/man2/mmap2.2:92
5016 "On ia64, the unit for I<offset> is actually the system page size, rather "
5021 #: build/C/man2/mmap2.2:98
5022 msgid "B<getpagesize>(2), B<mmap>(2), B<mremap>(2), B<msync>(2), B<shm_open>(3)"
5026 #: build/C/man2/mprotect.2:38
5032 #: build/C/man2/mprotect.2:38
5038 #: build/C/man2/mprotect.2:41
5039 msgid "mprotect - set protection on a region of memory"
5043 #: build/C/man2/mprotect.2:46
5045 msgid "B<int mprotect(void *>I<addr>B<, size_t >I<len>B<, int >I<prot>B<);>\n"
5049 #: build/C/man2/mprotect.2:54
5051 "B<mprotect>() changes protection for the calling process's memory page(s) "
5052 "containing any part of the address range in the interval [I<addr>,\\ "
5053 "I<addr>+I<len>-1]. I<addr> must be aligned to a page boundary."
5057 #: build/C/man2/mprotect.2:59
5059 "If the calling process tries to access memory in a manner that violates the "
5060 "protection, then the kernel generates a B<SIGSEGV> signal for the process."
5064 #: build/C/man2/mprotect.2:64
5066 "I<prot> is either B<PROT_NONE> or a bitwise-or of the other values in the "
5071 #: build/C/man2/mprotect.2:67
5072 msgid "The memory cannot be accessed at all."
5076 #: build/C/man2/mprotect.2:70
5077 msgid "The memory can be read."
5081 #: build/C/man2/mprotect.2:73
5082 msgid "The memory can be modified."
5086 #. Document PROT_GROWSUP and PROT_GROWSDOWN
5088 #: build/C/man2/mprotect.2:78
5089 msgid "The memory can be executed."
5093 #: build/C/man2/mprotect.2:85
5095 "On success, B<mprotect>() returns zero. On error, -1 is returned, and "
5096 "I<errno> is set appropriately."
5100 #: build/C/man2/mprotect.2:95
5102 "The memory cannot be given the specified access. This can happen, for "
5103 "example, if you B<mmap>(2) a file to which you have read-only access, then "
5104 "ask B<mprotect>() to mark it B<PROT_WRITE>."
5107 #. Or: both PROT_GROWSUP and PROT_GROWSDOWN were specified in 'prot'.
5109 #: build/C/man2/mprotect.2:100
5110 msgid "I<addr> is not a valid pointer, or not a multiple of the system page size."
5114 #: build/C/man2/mprotect.2:103
5115 msgid "Internal kernel structures could not be allocated."
5119 #: build/C/man2/mprotect.2:113
5121 "Addresses in the range [I<addr>, I<addr>+I<len>-1] are invalid for the "
5122 "address space of the process, or specify one or more pages that are not "
5123 "mapped. (Before kernel 2.4.19, the error B<EFAULT> was incorrectly produced "
5127 #. SVr4 defines an additional error
5128 #. code EAGAIN. The SVr4 error conditions don't map neatly onto Linux's.
5130 #: build/C/man2/mprotect.2:122
5132 "SVr4, POSIX.1-2001. POSIX says that the behavior of B<mprotect>() is "
5133 "unspecified if it is applied to a region of memory that was not obtained via "
5138 #: build/C/man2/mprotect.2:129
5140 "On Linux it is always permissible to call B<mprotect>() on any address in a "
5141 "process's address space (except for the kernel vsyscall area). In "
5142 "particular it can be used to change existing code mappings to be writable."
5146 #: build/C/man2/mprotect.2:139
5148 "Whether B<PROT_EXEC> has any effect different from B<PROT_READ> is "
5149 "architecture- and kernel version-dependent. On some hardware architectures "
5150 "(e.g., i386), B<PROT_WRITE> implies B<PROT_READ>."
5154 #: build/C/man2/mprotect.2:148
5156 "POSIX.1-2001 says that an implementation may permit access other than that "
5157 "specified in I<prot>, but at a minimum can allow write access only if "
5158 "B<PROT_WRITE> has been set, and must not allow any access if B<PROT_NONE> "
5163 #: build/C/man2/mprotect.2:154
5165 "The program below allocates four pages of memory, makes the third of these "
5166 "pages read-only, and then executes a loop that walks upward through the "
5167 "allocated region modifying bytes."
5171 #: build/C/man2/mprotect.2:157
5172 msgid "An example of what we might see when running the program is the following:"
5176 #: build/C/man2/mprotect.2:163
5180 "Start of region: 0x804c000\n"
5181 "Got SIGSEGV at address: 0x804e000\n"
5185 #: build/C/man2/mprotect.2:175
5188 "#include E<lt>unistd.hE<gt>\n"
5189 "#include E<lt>signal.hE<gt>\n"
5190 "#include E<lt>stdio.hE<gt>\n"
5191 "#include E<lt>malloc.hE<gt>\n"
5192 "#include E<lt>stdlib.hE<gt>\n"
5193 "#include E<lt>errno.hE<gt>\n"
5194 "#include E<lt>sys/mman.hE<gt>\n"
5198 #: build/C/man2/mprotect.2:180
5200 msgid "char *buffer;\n"
5204 #: build/C/man2/mprotect.2:188
5208 "handler(int sig, siginfo_t *si, void *unused)\n"
5210 " printf(\"Got SIGSEGV at address: 0x%lx\\en\",\n"
5211 " (long) si-E<gt>si_addr);\n"
5212 " exit(EXIT_FAILURE);\n"
5217 #: build/C/man2/mprotect.2:195
5221 "main(int argc, char *argv[])\n"
5225 " struct sigaction sa;\n"
5229 #: build/C/man2/mprotect.2:201
5232 " sa.sa_flags = SA_SIGINFO;\n"
5233 " sigemptyset(&sa.sa_mask);\n"
5234 " sa.sa_sigaction = handler;\n"
5235 " if (sigaction(SIGSEGV, &sa, NULL) == -1)\n"
5236 " handle_error(\"sigaction\");\n"
5240 #: build/C/man2/mprotect.2:205
5243 " pagesize = sysconf(_SC_PAGE_SIZE);\n"
5244 " if (pagesize == -1)\n"
5245 " handle_error(\"sysconf\");\n"
5249 #: build/C/man2/mprotect.2:208
5252 " /* Allocate a buffer aligned on a page boundary;\n"
5253 " initial protection is PROT_READ | PROT_WRITE */\n"
5257 #: build/C/man2/mprotect.2:212
5260 " buffer = memalign(pagesize, 4 * pagesize);\n"
5261 " if (buffer == NULL)\n"
5262 " handle_error(\"memalign\");\n"
5266 #: build/C/man2/mprotect.2:214
5268 msgid " printf(\"Start of region: 0x%lx\\en\", (long) buffer);\n"
5272 #: build/C/man2/mprotect.2:218
5275 " if (mprotect(buffer + pagesize * 2, pagesize,\n"
5276 " PROT_READ) == -1)\n"
5277 " handle_error(\"mprotect\");\n"
5281 #: build/C/man2/mprotect.2:221
5284 " for (p = buffer ; ; )\n"
5285 " *(p++) = \\(aqa\\(aq;\n"
5289 #: build/C/man2/mprotect.2:225
5292 " printf(\"Loop completed\\en\"); /* Should never happen */\n"
5293 " exit(EXIT_SUCCESS);\n"
5298 #: build/C/man2/mprotect.2:229
5299 msgid "B<mmap>(2), B<sysconf>(3)"
5303 #: build/C/man2/mremap.2:30
5309 #: build/C/man2/mremap.2:30
5315 #: build/C/man2/mremap.2:33
5316 msgid "mremap - remap a virtual memory address"
5320 #: build/C/man2/mremap.2:36
5322 msgid "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
5326 #: build/C/man2/mremap.2:41
5329 "B<void *mremap(void *>I<old_address>B<, size_t >I<old_size>B<,>\n"
5330 "B< size_t >I<new_size>B<, int >I<flags>B<, ... /* void "
5331 "*>I<new_address>B< */);>\n"
5335 #: build/C/man2/mremap.2:47
5337 "B<mremap>() expands (or shrinks) an existing memory mapping, potentially "
5338 "moving it at the same time (controlled by the I<flags> argument and the "
5339 "available virtual address space)."
5343 #: build/C/man2/mremap.2:61
5345 "I<old_address> is the old address of the virtual memory block that you want "
5346 "to expand (or shrink). Note that I<old_address> has to be page aligned. "
5347 "I<old_size> is the old size of the virtual memory block. I<new_size> is the "
5348 "requested size of the virtual memory block after the resize. An optional "
5349 "fifth argument, I<new_address>, may be provided; see the description of "
5350 "B<MREMAP_FIXED> below."
5354 #: build/C/man2/mremap.2:73
5356 "In Linux the memory is divided into pages. A user process has (one or) "
5357 "several linear virtual memory segments. Each virtual memory segment has one "
5358 "or more mappings to real memory pages (in the page table). Each virtual "
5359 "memory segment has its own protection (access rights), which may cause a "
5360 "segmentation violation if the memory is accessed incorrectly (e.g., writing "
5361 "to a read-only segment). Accessing virtual memory outside of the segments "
5362 "will also cause a segmentation violation."
5366 #: build/C/man2/mremap.2:81
5368 "B<mremap>() uses the Linux page table scheme. B<mremap>() changes the "
5369 "mapping between virtual addresses and memory pages. This can be used to "
5370 "implement a very efficient B<realloc>(3)."
5374 #: build/C/man2/mremap.2:83
5375 msgid "The I<flags> bit-mask argument may be 0, or include the following flag:"
5379 #: build/C/man2/mremap.2:83
5381 msgid "B<MREMAP_MAYMOVE>"
5385 #: build/C/man2/mremap.2:95
5387 "By default, if there is not sufficient space to expand a mapping at its "
5388 "current location, then B<mremap>() fails. If this flag is specified, then "
5389 "the kernel is permitted to relocate the mapping to a new virtual address, if "
5390 "necessary. If the mapping is relocated, then absolute pointers into the old "
5391 "mapping location become invalid (offsets relative to the starting address of "
5392 "the mapping should be employed)."
5396 #: build/C/man2/mremap.2:95
5398 msgid "B<MREMAP_FIXED> (since Linux 2.3.31)"
5402 #: build/C/man2/mremap.2:117
5404 "This flag serves a similar purpose to the B<MAP_FIXED> flag of B<mmap>(2). "
5405 "If this flag is specified, then B<mremap>() accepts a fifth argument, "
5406 "I<void\\ *new_address>, which specifies a page-aligned address to which the "
5407 "mapping must be moved. Any previous mapping at the address range specified "
5408 "by I<new_address> and I<new_size> is unmapped. If B<MREMAP_FIXED> is "
5409 "specified, then B<MREMAP_MAYMOVE> must also be specified."
5413 #: build/C/man2/mremap.2:127
5415 "If the memory segment specified by I<old_address> and I<old_size> is locked "
5416 "(using B<mlock>(2) or similar), then this lock is maintained when the "
5417 "segment is resized and/or relocated. As a consequence, the amount of memory "
5418 "locked by the process may change."
5422 #: build/C/man2/mremap.2:135
5424 "On success B<mremap>() returns a pointer to the new virtual memory area. "
5425 "On error, the value B<MAP_FAILED> (that is, I<(void\\ *)\\ -1>) is returned, "
5426 "and I<errno> is set appropriately."
5430 #: build/C/man2/mremap.2:142
5432 "The caller tried to expand a memory segment that is locked, but this was not "
5433 "possible without exceeding the B<RLIMIT_MEMLOCK> resource limit."
5437 #: build/C/man2/mremap.2:151
5439 "\"Segmentation fault.\" Some address in the range I<old_address> to "
5440 "I<old_address>+I<old_size> is an invalid virtual memory address for this "
5441 "process. You can also get B<EFAULT> even if there exist mappings that cover "
5442 "the whole address space requested, but those mappings are of different "
5447 #: build/C/man2/mremap.2:179
5449 "An invalid argument was given. Possible causes are: I<old_address> was not "
5450 "page aligned; a value other than B<MREMAP_MAYMOVE> or B<MREMAP_FIXED> was "
5451 "specified in I<flags>; I<new_size> was zero; I<new_size> or I<new_address> "
5452 "was invalid; or the new address range specified by I<new_address> and "
5453 "I<new_size> overlapped the old address range specified by I<old_address> and "
5454 "I<old_size>; or B<MREMAP_FIXED> was specified without also specifying "
5455 "B<MREMAP_MAYMOVE>."
5459 #: build/C/man2/mremap.2:185
5461 "The memory area cannot be expanded at the current virtual address, and the "
5462 "B<MREMAP_MAYMOVE> flag is not set in I<flags>. Or, there is not enough "
5463 "(virtual) memory available."
5466 #. 4.2BSD had a (never actually implemented)
5468 #. call with completely different semantics.
5470 #: build/C/man2/mremap.2:191
5472 "This call is Linux-specific, and should not be used in programs intended to "
5477 #: build/C/man2/mremap.2:199
5479 "Prior to version 2.4, glibc did not expose the definition of "
5480 "B<MREMAP_FIXED>, and the prototype for B<mremap>() did not allow for the "
5481 "I<new_address> argument."
5485 #: build/C/man2/mremap.2:208
5487 "B<brk>(2), B<getpagesize>(2), B<getrlimit>(2), B<mlock>(2), B<mmap>(2), "
5488 "B<sbrk>(2), B<malloc>(3), B<realloc>(3)"
5492 #: build/C/man2/mremap.2:214
5494 "Your favorite text book on operating systems for more information on paged "
5495 "memory (e.g., I<Modern Operating Systems> by Andrew S. Tanenbaum, I<Inside "
5496 "Linux> by Randolf Bentson, I<The Design of the UNIX Operating System> by "
5501 #: build/C/man2/msync.2:25
5507 #: build/C/man2/msync.2:25 build/C/man2/remap_file_pages.2:28
5513 #: build/C/man2/msync.2:28
5514 msgid "msync - synchronize a file with a memory map"
5518 #: build/C/man2/msync.2:32
5519 msgid "B<int msync(void *>I<addr>B<, size_t >I<length>B<, int >I<flags>B<);>"
5523 #: build/C/man2/msync.2:48
5525 "B<msync>() flushes changes made to the in-core copy of a file that was "
5526 "mapped into memory using B<mmap>(2) back to disk. Without use of this call "
5527 "there is no guarantee that changes are written back before B<munmap>(2) is "
5528 "called. To be more precise, the part of the file that corresponds to the "
5529 "memory area starting at I<addr> and having length I<length> is updated."
5533 #: build/C/man2/msync.2:68
5535 "The I<flags> argument may have the bits B<MS_ASYNC>, B<MS_SYNC>, and "
5536 "B<MS_INVALIDATE> set, but not both B<MS_ASYNC> and B<MS_SYNC>. B<MS_ASYNC> "
5537 "specifies that an update be scheduled, but the call returns immediately. "
5538 "B<MS_SYNC> asks for an update and waits for it to complete. "
5539 "B<MS_INVALIDATE> asks to invalidate other mappings of the same file (so that "
5540 "they can be updated with the fresh values just written)."
5544 #: build/C/man2/msync.2:73
5546 "On success, zero is returned. On error, -1 is returned, and I<errno> is set "
5551 #: build/C/man2/msync.2:74
5557 #: build/C/man2/msync.2:80
5559 "B<MS_INVALIDATE> was specified in I<flags>, and a memory lock exists for the "
5560 "specified address range."
5564 #: build/C/man2/msync.2:93
5566 "I<addr> is not a multiple of PAGESIZE; or any bit other than B<MS_ASYNC> | "
5567 "B<MS_INVALIDATE> | B<MS_SYNC> is set in I<flags>; or both B<MS_SYNC> and "
5568 "B<MS_ASYNC> are set in I<flags>."
5572 #: build/C/man2/msync.2:96
5573 msgid "The indicated memory (or part of it) was not mapped."
5577 #: build/C/man2/msync.2:98 build/C/man3/posix_fallocate.3:105 build/C/man3/shm_open.3:253 build/C/man7/shm_overview.7:103
5578 msgid "POSIX.1-2001."
5582 #: build/C/man2/msync.2:105
5584 "This call was introduced in Linux 1.3.21, and then used B<EFAULT> instead of "
5585 "B<ENOMEM>. In Linux 2.4.19 this was changed to the POSIX value B<ENOMEM>."
5588 #. POSIX.1-2001: It shall be defined to -1 or 0 or 200112L.
5589 #. -1: unavailable, 0: ask using sysconf().
5590 #. glibc defines them to 1.
5592 #: build/C/man2/msync.2:120
5594 "On POSIX systems on which B<msync>() is available, both "
5595 "B<_POSIX_MAPPED_FILES> and B<_POSIX_SYNCHRONIZED_IO> are defined in "
5596 "I<E<lt>unistd.hE<gt>> to a value greater than 0. (See also B<sysconf>(3).)"
5600 #: build/C/man2/msync.2:122 build/C/man7/shm_overview.7:46
5606 #: build/C/man3/mtrace.3:25
5612 #: build/C/man3/mtrace.3:28
5613 msgid "mtrace, muntrace - malloc tracing"
5617 #: build/C/man3/mtrace.3:30
5618 msgid "B<#include E<lt>mcheck.hE<gt>>"
5622 #: build/C/man3/mtrace.3:32
5623 msgid "B<void mtrace(void);>"
5627 #: build/C/man3/mtrace.3:34
5628 msgid "B<void muntrace(void);>"
5632 #: build/C/man3/mtrace.3:46
5634 "The B<mtrace>() function installs hook functions for the memory-allocation "
5635 "functions (B<malloc>(3), B<realloc>(3) B<memalign>(3), B<free>(3)). These "
5636 "hook functions record tracing information about memory allocation and "
5637 "deallocation. The tracing information can be used to discover memory leaks "
5638 "and attempts to free nonallocated memory in a program."
5642 #: build/C/man3/mtrace.3:57
5644 "The B<muntrace>() function disables the hook functions installed by "
5645 "B<mtrace>(), so that tracing information is no longer recorded for the "
5646 "memory-allocation functions. If no hook functions were successfully "
5647 "installed by B<mtrace>(), B<muntrace>() does nothing."
5651 #: build/C/man3/mtrace.3:65
5653 "When B<mtrace>() is called, it checks the value of the environment variable "
5654 "B<MALLOC_TRACE>, which should contain the pathname of a file in which the "
5655 "tracing information is to be recorded. If the pathname is successfully "
5656 "opened, it is truncated to zero length."
5660 #: build/C/man3/mtrace.3:78
5662 "If B<MALLOC_TRACE> is not set, or the pathname it specifies is invalid or "
5663 "not writable, then no hook functions are installed, and B<mtrace>() has no "
5664 "effect. In set-user-ID and set-group-ID programs, B<MALLOC_TRACE> is "
5665 "ignored, and B<mtrace>() has no effect."
5669 #: build/C/man3/mtrace.3:86
5671 "In normal usage, B<mtrace>() is called once at the start of execution of a "
5672 "program, and B<muntrace>() is never called."
5676 #: build/C/man3/mtrace.3:96
5678 "The tracing output produced after a call to B<mtrace>() is textual, but not "
5679 "designed to be human readable. The GNU C library provides a Perl script, "
5680 "B<mtrace>(1), that interprets the trace log and produces human-readable "
5681 "output. For best results, the traced program should be compiled with "
5682 "debugging enabled, so that line-number information is recorded in the "
5687 #: build/C/man3/mtrace.3:102
5689 "The tracing performed by B<mtrace>() incurs a performance penalty (if "
5690 "B<MALLOC_TRACE> points to a valid, writable pathname)."
5694 #: build/C/man3/mtrace.3:108
5696 "The line-number information produced by B<mtrace>(1) is not always precise: "
5697 "the line number references may refer to the previous or following "
5698 "(non-blank) line of the source code."
5702 #: build/C/man3/mtrace.3:115
5704 "The shell session below demonstrates the use of the B<mtrace>() function "
5705 "and the B<mtrace>(1) command in a program that has memory leaks at two "
5706 "different locations. The demonstration uses the following program:"
5710 #: build/C/man3/mtrace.3:122
5713 "$ B<cat t_mtrace.c>\n"
5714 "#include E<lt>mcheck.hE<gt>\n"
5715 "#include E<lt>stdlib.hE<gt>\n"
5716 "#include E<lt>stdio.hE<gt>\n"
5720 #: build/C/man3/mtrace.3:127
5724 "main(int argc, char *argv[])\n"
5730 #: build/C/man3/mtrace.3:129
5732 msgid " mtrace();\n"
5736 #: build/C/man3/mtrace.3:132
5739 " for (j = 0; j E<lt> 2; j++)\n"
5740 " malloc(100); /* Never freed--a memory leak */\n"
5744 #: build/C/man3/mtrace.3:136
5747 " calloc(16, 16); /* Never freed--a memory leak */\n"
5748 " exit(EXIT_SUCCESS);\n"
5753 #: build/C/man3/mtrace.3:142
5755 "When we run the program as follows, we see that B<mtrace>() diagnosed "
5756 "memory leaks at two different locations in the program:"
5760 #: build/C/man3/mtrace.3:155
5763 "$ B<cc -g t_mtrace.c -o t_mtrace>\n"
5764 "$ B<export MALLOC_TRACE=/tmp/t>\n"
5766 "$ B<mtrace ./t_mtrace $MALLOC_TRACE>\n"
5767 "Memory not freed:\n"
5768 "-----------------\n"
5769 " Address Size Caller\n"
5770 "0x084c9378 0x64 at /home/cecilia/t_mtrace.c:12\n"
5771 "0x084c93e0 0x64 at /home/cecilia/t_mtrace.c:12\n"
5772 "0x084c9448 0x100 at /home/cecilia/t_mtrace.c:16\n"
5776 #: build/C/man3/mtrace.3:167
5778 "The first two messages about unfreed memory correspond to the two "
5779 "B<malloc>(3) calls inside the I<for> loop. The final message corresponds "
5780 "to the call to B<calloc>(3) (which in turn calls B<malloc>(3))."
5784 #: build/C/man3/mtrace.3:172
5785 msgid "B<mtrace>(1), B<malloc>(3), B<malloc_hook>(3), B<mcheck>(3)"
5789 #: build/C/man2/posix_fadvise.2:28
5791 msgid "POSIX_FADVISE"
5795 #: build/C/man2/posix_fadvise.2:28 build/C/man2/readahead.2:28 build/C/man2/sync_file_range.2:30
5801 #: build/C/man2/posix_fadvise.2:31
5802 msgid "posix_fadvise - predeclare an access pattern for file data"
5806 #: build/C/man2/posix_fadvise.2:34 build/C/man3/posix_fallocate.3:31
5808 msgid "B<#include E<lt>fcntl.hE<gt>>\n"
5812 #: build/C/man2/posix_fadvise.2:37
5815 "B<int posix_fadvise(int >I<fd>B<, off_t >I<offset>B<, off_t >I<len>B<, int "
5820 #: build/C/man2/posix_fadvise.2:46
5821 msgid "B<posix_fadvise>():"
5825 #: build/C/man2/posix_fadvise.2:48 build/C/man3/posix_fallocate.3:44
5826 msgid "_XOPEN_SOURCE\\ E<gt>=\\ 600 || _POSIX_C_SOURCE\\ E<gt>=\\ 200112L"
5830 #: build/C/man2/posix_fadvise.2:56
5832 "Programs can use B<posix_fadvise>() to announce an intention to access file "
5833 "data in a specific pattern in the future, thus allowing the kernel to "
5834 "perform appropriate optimizations."
5838 #: build/C/man2/posix_fadvise.2:63
5840 "The I<advice> applies to a (not necessarily existent) region starting at "
5841 "I<offset> and extending for I<len> bytes (or until the end of the file if "
5842 "I<len> is 0) within the file referred to by I<fd>. The I<advice> is not "
5843 "binding; it merely constitutes an expectation on behalf of the application."
5847 #: build/C/man2/posix_fadvise.2:65
5848 msgid "Permissible values for I<advice> include:"
5852 #: build/C/man2/posix_fadvise.2:65
5854 msgid "B<POSIX_FADV_NORMAL>"
5858 #: build/C/man2/posix_fadvise.2:71
5860 "Indicates that the application has no advice to give about its access "
5861 "pattern for the specified data. If no advice is given for an open file, "
5862 "this is the default assumption."
5866 #: build/C/man2/posix_fadvise.2:71
5868 msgid "B<POSIX_FADV_SEQUENTIAL>"
5872 #: build/C/man2/posix_fadvise.2:75
5874 "The application expects to access the specified data sequentially (with "
5875 "lower offsets read before higher ones)."
5879 #: build/C/man2/posix_fadvise.2:75
5881 msgid "B<POSIX_FADV_RANDOM>"
5885 #: build/C/man2/posix_fadvise.2:78
5886 msgid "The specified data will be accessed in random order."
5890 #: build/C/man2/posix_fadvise.2:78
5892 msgid "B<POSIX_FADV_NOREUSE>"
5896 #: build/C/man2/posix_fadvise.2:81
5897 msgid "The specified data will be accessed only once."
5901 #: build/C/man2/posix_fadvise.2:81
5903 msgid "B<POSIX_FADV_WILLNEED>"
5907 #: build/C/man2/posix_fadvise.2:84
5908 msgid "The specified data will be accessed in the near future."
5912 #: build/C/man2/posix_fadvise.2:84
5914 msgid "B<POSIX_FADV_DONTNEED>"
5918 #: build/C/man2/posix_fadvise.2:87
5919 msgid "The specified data will not be accessed in the near future."
5923 #: build/C/man2/posix_fadvise.2:90
5924 msgid "On success, zero is returned. On error, an error number is returned."
5928 #: build/C/man2/posix_fadvise.2:94
5929 msgid "The I<fd> argument was not a valid file descriptor."
5933 #: build/C/man2/posix_fadvise.2:97
5934 msgid "An invalid value was specified for I<advice>."
5938 #: build/C/man2/posix_fadvise.2:104
5940 "The specified file descriptor refers to a pipe or FIFO. (Linux actually "
5941 "returns B<EINVAL> in this case.)"
5944 #. of fadvise64_64()
5946 #: build/C/man2/posix_fadvise.2:112
5948 "Kernel support first appeared in Linux 2.5.60; the underlying system call is "
5949 "called B<fadvise64>(). Library support has been provided since glibc "
5950 "version 2.2, via the wrapper function B<posix_fadvise>()."
5954 #: build/C/man2/posix_fadvise.2:121
5956 "POSIX.1-2001. Note that the type of the I<len> argument was changed from "
5957 "I<size_t> to I<off_t> in POSIX.1-2003 TC1."
5961 #: build/C/man2/posix_fadvise.2:127
5963 "Under Linux, B<POSIX_FADV_NORMAL> sets the readahead window to the default "
5964 "size for the backing device; B<POSIX_FADV_SEQUENTIAL> doubles this size, and "
5965 "B<POSIX_FADV_RANDOM> disables file readahead entirely. These changes affect "
5966 "the entire file, not just the specified region (but other open file handles "
5967 "to the same file are unaffected)."
5971 #: build/C/man2/posix_fadvise.2:134
5973 "B<POSIX_FADV_WILLNEED> initiates a nonblocking read of the specified region "
5974 "into the page cache. The amount of data read may be decreased by the kernel "
5975 "depending on virtual memory load. (A few megabytes will usually be fully "
5976 "satisfied, and more is rarely useful.)"
5980 #: build/C/man2/posix_fadvise.2:138
5982 "In kernels before 2.6.18, B<POSIX_FADV_NOREUSE> had the same semantics as "
5983 "B<POSIX_FADV_WILLNEED>. This was probably a bug; since kernel 2.6.18, this "
5988 #: build/C/man2/posix_fadvise.2:146
5990 "B<POSIX_FADV_DONTNEED> attempts to free cached pages associated with the "
5991 "specified region. This is useful, for example, while streaming large "
5992 "files. A program may periodically request the kernel to free cached data "
5993 "that has already been used, so that more useful cached pages are not "
5994 "discarded instead."
5998 #: build/C/man2/posix_fadvise.2:154
6000 "Pages that have not yet been written out will be unaffected, so if the "
6001 "application wishes to guarantee that pages will be released, it should call "
6002 "B<fsync>(2) or B<fdatasync>(2) first."
6006 #: build/C/man2/posix_fadvise.2:154
6008 msgid "Architecture-specific variants"
6012 #: build/C/man2/posix_fadvise.2:171
6014 "Some architectures require 64-bit arguments to be aligned in a suitable pair "
6015 "of registers (see B<syscall>(2) for further detail). On such "
6016 "architectures, the call signature of B<posix_fadvise>() shown in the "
6017 "SYNOPSIS would force a register to be wasted as padding between the I<fd> "
6018 "and I<len> arguments. Therefore, these architectures define a version of "
6019 "the system call that orders the arguments suitably, but otherwise is "
6020 "otherwise exactly the same as B<posix_fadvise>()."
6024 #: build/C/man2/posix_fadvise.2:173
6025 msgid "For example, since Linux 2.6.14, ARM has the following system call:"
6029 #: build/C/man2/posix_fadvise.2:178
6032 "B<long arm_fadvise64_64(int >I<fd>B<, int >I<advice>B<,>\n"
6033 "B< loff_t >I<offset>B<, loff_t >I<len>B<);>\n"
6037 #: build/C/man2/posix_fadvise.2:186
6039 "These architecture-specific details are generally hidden from applications "
6040 "by the glibc B<posix_fadvise>() wrapper function, which invokes the "
6041 "appropriate architecture-specific system call."
6045 #: build/C/man2/posix_fadvise.2:191
6047 "In kernels before 2.6.6, if I<len> was specified as 0, then this was "
6048 "interpreted literally as \"zero bytes\", rather than as meaning \"all bytes "
6049 "through to the end of the file\"."
6052 #. FIXME . Write a posix_fadvise(3) page.
6054 #: build/C/man2/posix_fadvise.2:197
6056 "B<readahead>(2), B<sync_file_range>(2), B<posix_fallocate>(3), "
6057 "B<posix_madvise>(3)"
6061 #: build/C/man3/posix_fallocate.3:25
6063 msgid "POSIX_FALLOCATE"
6067 #: build/C/man3/posix_fallocate.3:25 build/C/man2/shmop.2:41
6073 #: build/C/man3/posix_fallocate.3:28
6074 msgid "posix_fallocate - allocate file space"
6078 #: build/C/man3/posix_fallocate.3:33
6080 msgid "B<int posix_fallocate(int >I<fd>B<, off_t >I<offset>B<, off_t >I<len>B<);>\n"
6084 #: build/C/man3/posix_fallocate.3:42
6085 msgid "B<posix_fallocate>():"
6089 #: build/C/man3/posix_fallocate.3:61
6091 "The function B<posix_fallocate>() ensures that disk space is allocated for "
6092 "the file referred to by the descriptor I<fd> for the bytes in the range "
6093 "starting at I<offset> and continuing for I<len> bytes. After a successful "
6094 "call to B<posix_fallocate>(), subsequent writes to bytes in the specified "
6095 "range are guaranteed not to fail because of lack of disk space."
6099 #: build/C/man3/posix_fallocate.3:66
6101 "If the size of the file is less than I<offset>+I<len>, then the file is "
6102 "increased to this size; otherwise the file size is left unchanged."
6106 #: build/C/man3/posix_fallocate.3:72
6108 "B<posix_fallocate>() returns zero on success, or an error number on "
6109 "failure. Note that I<errno> is not set."
6113 #: build/C/man3/posix_fallocate.3:81
6114 msgid "I<offset+len> exceeds the maximum file size."
6118 #: build/C/man3/posix_fallocate.3:91
6119 msgid "I<fd> does not refer to a regular file."
6123 #: build/C/man3/posix_fallocate.3:100
6124 msgid "I<fd> refers to a pipe."
6128 #: build/C/man3/posix_fallocate.3:103
6129 msgid "B<posix_fallocate>() is available since glibc 2.1.94."
6133 #: build/C/man3/posix_fallocate.3:128
6135 "POSIX.1-2008 says that an implementation I<shall> give the B<EINVAL> error "
6136 "if I<len> was 0, or I<offset> was less than 0. POSIX.1-2001 says that an "
6137 "implementation I<shall> give the B<EINVAL> error if I<len> is less than 0, "
6138 "or I<offset> was less than 0, and I<may> give the error if I<len> equals "
6143 #: build/C/man3/posix_fallocate.3:133
6144 msgid "B<fallocate>(1), B<fallocate>(2), B<lseek>(2), B<posix_fadvise>(2)"
6148 #: build/C/man3/posix_memalign.3:29
6150 msgid "POSIX_MEMALIGN"
6154 #: build/C/man3/posix_memalign.3:29
6160 #: build/C/man3/posix_memalign.3:32
6162 "posix_memalign, aligned_alloc, memalign, valloc, pvalloc - allocate aligned "
6167 #: build/C/man3/posix_memalign.3:39
6170 "B<int posix_memalign(void **>I<memptr>B<, size_t >I<alignment>B<, size_t "
6172 "B<void *aligned_alloc(size_t >I<alignment>B<, size_t >I<size>B<);>\n"
6173 "B<void *valloc(size_t >I<size>B<);>\n"
6177 #: build/C/man3/posix_memalign.3:44
6180 "B<void *memalign(size_t >I<alignment>B<, size_t >I<size>B<);>\n"
6181 "B<void *pvalloc(size_t >I<size>B<);>\n"
6185 #: build/C/man3/posix_memalign.3:54
6187 "B<posix_memalign>(): _POSIX_C_SOURCE\\ E<gt>=\\ 200112L || _XOPEN_SOURCE\\ "
6192 #: build/C/man3/posix_memalign.3:57
6193 msgid "B<aligned_alloc>(): _ISOC11_SOURCE"
6197 #: build/C/man3/posix_memalign.3:59
6198 msgid "B<valloc>():"
6202 #: build/C/man3/posix_memalign.3:62
6204 msgid "Since glibc 2.12:"
6208 #: build/C/man3/posix_memalign.3:69
6212 " (_XOPEN_SOURCE\\ E<gt>=\\ 500 ||\n"
6213 " _XOPEN_SOURCE\\ &&\\ _XOPEN_SOURCE_EXTENDED) &&\n"
6214 " !(_POSIX_C_SOURCE\\ E<gt>=\\ 200112L || _XOPEN_SOURCE\\ E<gt>=\\ 600)\n"
6218 #: build/C/man3/posix_memalign.3:71
6220 msgid "Before glibc 2.12:"
6224 #: build/C/man3/posix_memalign.3:75
6226 "_BSD_SOURCE || _XOPEN_SOURCE\\ E<gt>=\\ 500 || _XOPEN_SOURCE\\ &&\\ "
6227 "_XOPEN_SOURCE_EXTENDED"
6231 #: build/C/man3/posix_memalign.3:82
6233 "(The (nonstandard) header file I<E<lt>malloc.hE<gt>> also exposes the "
6234 "declaration of B<valloc>(); no feature test macros are required.)"
6239 #: build/C/man3/posix_memalign.3:104
6241 "The function B<posix_memalign>() allocates I<size> bytes and places the "
6242 "address of the allocated memory in I<*memptr>. The address of the allocated "
6243 "memory will be a multiple of I<alignment>, which must be a power of two and "
6244 "a multiple of I<sizeof(void\\ *)>. If I<size> is 0, then the value placed "
6245 "in I<*memptr> is either NULL, or a unique pointer value that can later be "
6246 "successfully passed to B<free>(3)."
6249 #. The behavior of memalign() for size==0 is as for posix_memalign()
6250 #. but no standards govern this.
6252 #: build/C/man3/posix_memalign.3:115
6254 "The obsolete function B<memalign>() allocates I<size> bytes and returns a "
6255 "pointer to the allocated memory. The memory address will be a multiple of "
6256 "I<alignment>, which must be a power of two."
6260 #: build/C/man3/posix_memalign.3:124
6262 "The function B<aligned_alloc>() is the same as B<memalign>(), except for "
6263 "the added restriction that I<size> should be a multiple of I<alignment>."
6267 #: build/C/man3/posix_memalign.3:133
6269 "The obsolete function B<valloc>() allocates I<size> bytes and returns a "
6270 "pointer to the allocated memory. The memory address will be a multiple of "
6271 "the page size. It is equivalent to I<memalign(sysconf(_SC_PAGESIZE),size)>."
6275 #: build/C/man3/posix_memalign.3:140
6277 "The obsolete function B<pvalloc>() is similar to B<valloc>(), but rounds "
6278 "the size of the allocation up to the next multiple of the system page size."
6282 #: build/C/man3/posix_memalign.3:142
6283 msgid "For all of these functions, the memory is not zeroed."
6287 #: build/C/man3/posix_memalign.3:149
6289 "B<aligned_alloc>(), B<memalign>(), B<valloc>(), and B<pvalloc>() return a "
6290 "pointer to the allocated memory, or NULL if the request fails."
6294 #: build/C/man3/posix_memalign.3:157
6296 "B<posix_memalign>() returns zero on success, or one of the error values "
6297 "listed in the next section on failure. The value of I<errno> is "
6298 "indeterminate after a call to B<posix_memalign>()."
6302 #: build/C/man3/posix_memalign.3:164
6304 "The I<alignment> argument was not a power of two, or was not a multiple of "
6305 "I<sizeof(void\\ *)>."
6309 #: build/C/man3/posix_memalign.3:167
6310 msgid "There was insufficient memory to fulfill the allocation request."
6314 #: build/C/man3/posix_memalign.3:174
6316 "The functions B<memalign>(), B<valloc>(), and B<pvalloc>() have been "
6317 "available in all Linux libc libraries."
6321 #: build/C/man3/posix_memalign.3:178
6322 msgid "The function B<aligned_alloc>() was added to glibc in version 2.16."
6326 #: build/C/man3/posix_memalign.3:182
6327 msgid "The function B<posix_memalign>() is available since glibc 2.1.91."
6331 #: build/C/man3/posix_memalign.3:189
6333 "The function B<valloc>() appeared in 3.0BSD. It is documented as being "
6334 "obsolete in 4.3BSD, and as legacy in SUSv2. It does not appear in "
6339 #: build/C/man3/posix_memalign.3:193
6340 msgid "The function B<pvalloc>() is a GNU extension."
6344 #: build/C/man3/posix_memalign.3:197
6345 msgid "The function B<memalign>() appears in SunOS 4.1.3 but not in 4.4BSD."
6349 #: build/C/man3/posix_memalign.3:201
6350 msgid "The function B<posix_memalign>() comes from POSIX.1d."
6354 #: build/C/man3/posix_memalign.3:206
6355 msgid "The function B<aligned_alloc>() is specified in the C11 standard."
6359 #: build/C/man3/posix_memalign.3:206
6365 #: build/C/man3/posix_memalign.3:210
6367 "Everybody agrees that B<posix_memalign>() is declared in "
6368 "I<E<lt>stdlib.hE<gt>>."
6372 #: build/C/man3/posix_memalign.3:214
6374 "On some systems B<memalign>() is declared in I<E<lt>stdlib.hE<gt>> instead "
6375 "of I<E<lt>malloc.hE<gt>>."
6379 #: build/C/man3/posix_memalign.3:221
6381 "According to SUSv2, B<valloc>() is declared in I<E<lt>stdlib.hE<gt>>. "
6382 "Libc4,5 and glibc declare it in I<E<lt>malloc.hE<gt>>, and also in "
6383 "I<E<lt>stdlib.hE<gt>> if suitable feature test macros are defined (see "
6388 #: build/C/man3/posix_memalign.3:230
6390 "On many systems there are alignment restrictions, for example, on buffers "
6391 "used for direct block device I/O. POSIX specifies the "
6392 "I<pathconf(path,_PC_REC_XFER_ALIGN)> call that tells what alignment is "
6393 "needed. Now one can use B<posix_memalign>() to satisfy this requirement."
6397 #: build/C/man3/posix_memalign.3:239
6399 "B<posix_memalign>() verifies that I<alignment> matches the requirements "
6400 "detailed above. B<memalign>() may not check that the I<alignment> argument "
6404 #. Other systems allow passing the result of
6411 #: build/C/man3/posix_memalign.3:267
6413 "POSIX requires that memory obtained from B<posix_memalign>() can be freed "
6414 "using B<free>(3). Some systems provide no way to reclaim memory allocated "
6415 "with B<memalign>() or B<valloc>() (because one can pass to B<free>(3) "
6416 "only a pointer obtained from B<malloc>(3), while, for example, B<memalign>() "
6417 "would call B<malloc>(3) and then align the obtained value). The glibc "
6418 "implementation allows memory obtained from any of these functions to be "
6419 "reclaimed with B<free>(3)."
6423 #: build/C/man3/posix_memalign.3:272
6425 "The glibc B<malloc>(3) always returns 8-byte aligned memory addresses, so "
6426 "these functions are needed only if you require larger alignment values."
6430 #: build/C/man3/posix_memalign.3:277
6431 msgid "B<brk>(2), B<getpagesize>(2), B<free>(3), B<malloc>(3)"
6435 #: build/C/man2/readahead.2:28
6441 #: build/C/man2/readahead.2:31
6442 msgid "readahead - perform file readahead into page cache"
6446 #: build/C/man2/readahead.2:37
6449 "B<ssize_t readahead(int >I<fd>B<, off64_t >I<offset>B<, size_t "
6454 #: build/C/man2/readahead.2:65
6456 "B<readahead>() populates the page cache with data from a file so that "
6457 "subsequent reads from that file will not block on disk I/O. The I<fd> "
6458 "argument is a file descriptor identifying the file which is to be read. The "
6459 "I<offset> argument specifies the starting point from which data is to be "
6460 "read and I<count> specifies the number of bytes to be read. I/O is "
6461 "performed in whole pages, so that I<offset> is effectively rounded down to a "
6462 "page boundary and bytes are read up to the next page boundary greater than "
6463 "or equal to I<(offset+count)>. B<readahead>() does not read beyond the end "
6464 "of the file. B<readahead>() blocks until the specified data has been "
6465 "read. The current file offset of the open file referred to by I<fd> is left "
6470 #: build/C/man2/readahead.2:71
6472 "On success, B<readahead>() returns 0; on failure, -1 is returned, with "
6473 "I<errno> set to indicate the cause of the error."
6477 #: build/C/man2/readahead.2:76
6478 msgid "I<fd> is not a valid file descriptor or is not open for reading."
6482 #: build/C/man2/readahead.2:82
6483 msgid "I<fd> does not refer to a file type to which B<readahead>() can be applied."
6487 #: build/C/man2/readahead.2:87
6489 "The B<readahead>() system call appeared in Linux 2.4.13; glibc support has "
6490 "been provided since version 2.3."
6494 #: build/C/man2/readahead.2:92
6496 "The B<readahead>() system call is Linux-specific, and its use should be "
6497 "avoided in portable applications."
6501 #: build/C/man2/readahead.2:97
6503 "On some 32-bit architectures, the calling signature for this system call "
6504 "differs, for the reasons described in B<syscall>(2)."
6508 #: build/C/man2/readahead.2:103
6509 msgid "B<lseek>(2), B<madvise>(2), B<mmap>(2), B<posix_fadvise>(2), B<read>(2)"
6513 #: build/C/man2/remap_file_pages.2:28
6515 msgid "REMAP_FILE_PAGES"
6519 #: build/C/man2/remap_file_pages.2:31
6520 msgid "remap_file_pages - create a nonlinear file mapping"
6524 #: build/C/man2/remap_file_pages.2:35
6527 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
6528 "B<#include E<lt>sys/mman.hE<gt>>\n"
6532 #: build/C/man2/remap_file_pages.2:38
6535 "B<int remap_file_pages(void *>I<addr>B<, size_t >I<size>B<, int "
6537 "B< ssize_t >I<pgoff>B<, int >I<flags>B<);>\n"
6541 #: build/C/man2/remap_file_pages.2:51
6543 "The B<remap_file_pages>() system call is used to create a nonlinear "
6544 "mapping, that is, a mapping in which the pages of the file are mapped into a "
6545 "nonsequential order in memory. The advantage of using B<remap_file_pages>() "
6546 "over using repeated calls to B<mmap>(2) is that the former approach does "
6547 "not require the kernel to create additional VMA (Virtual Memory Area) data "
6552 #: build/C/man2/remap_file_pages.2:53
6553 msgid "To create a nonlinear mapping we perform the following steps:"
6557 #: build/C/man2/remap_file_pages.2:53
6563 #: build/C/man2/remap_file_pages.2:61
6565 "Use B<mmap>(2) to create a mapping (which is initially linear). This "
6566 "mapping must be created with the B<MAP_SHARED> flag."
6570 #: build/C/man2/remap_file_pages.2:61
6576 #: build/C/man2/remap_file_pages.2:69
6578 "Use one or more calls to B<remap_file_pages>() to rearrange the "
6579 "correspondence between the pages of the mapping and the pages of the file. "
6580 "It is possible to map the same page of a file into multiple locations within "
6581 "the mapped region."
6585 #: build/C/man2/remap_file_pages.2:80
6587 "The I<pgoff> and I<size> arguments specify the region of the file that is to "
6588 "be relocated within the mapping: I<pgoff> is a file offset in units of the "
6589 "system page size; I<size> is the length of the region in bytes."
6593 #: build/C/man2/remap_file_pages.2:98
6595 "The I<addr> argument serves two purposes. First, it identifies the mapping "
6596 "whose pages we want to rearrange. Thus, I<addr> must be an address that "
6597 "falls within a region previously mapped by a call to B<mmap>(2). Second, "
6598 "I<addr> specifies the address at which the file pages identified by I<pgoff> "
6599 "and I<size> will be placed."
6602 #. This rounding is weird, and not consistent with the treatment of
6603 #. the analogous arguments for munmap()/mprotect() and for mlock().
6606 #: build/C/man2/remap_file_pages.2:112
6608 "The values specified in I<addr> and I<size> should be multiples of the "
6609 "system page size. If they are not, then the kernel rounds I<both> values "
6610 "I<down> to the nearest multiple of the page size."
6614 #: build/C/man2/remap_file_pages.2:116
6615 msgid "The I<prot> argument must be specified as 0."
6619 #: build/C/man2/remap_file_pages.2:124
6621 "The I<flags> argument has the same meaning as for B<mmap>(2), but all flags "
6622 "other than B<MAP_NONBLOCK> are ignored."
6626 #: build/C/man2/remap_file_pages.2:131
6628 "On success, B<remap_file_pages>() returns 0. On error, -1 is returned, and "
6629 "I<errno> is set appropriately."
6633 #: build/C/man2/remap_file_pages.2:139
6635 "I<addr> does not refer to a valid mapping created with the B<MAP_SHARED> "
6639 #. And possibly others from vma->vm_ops->populate()
6641 #: build/C/man2/remap_file_pages.2:148
6642 msgid "I<addr>, I<size>, I<prot>, or I<pgoff> is invalid."
6646 #: build/C/man2/remap_file_pages.2:153
6648 "The B<remap_file_pages>() system call appeared in Linux 2.5.46; glibc "
6649 "support was added in version 2.3.3."
6653 #: build/C/man2/remap_file_pages.2:157
6654 msgid "The B<remap_file_pages>() system call is Linux-specific."
6658 #: build/C/man2/remap_file_pages.2:164
6660 "B<getpagesize>(2), B<mmap>(2), B<mmap2>(2), B<mprotect>(2), B<mremap>(2), "
6665 #: build/C/man3/shm_open.3:26
6671 #: build/C/man3/shm_open.3:26
6677 #: build/C/man3/shm_open.3:29
6678 msgid "shm_open, shm_unlink - create/open or unlink POSIX shared memory objects"
6682 #: build/C/man3/shm_open.3:33
6683 msgid "B<#include E<lt>sys/stat.hE<gt>> /* For mode constants */"
6687 #: build/C/man3/shm_open.3:35
6688 msgid "B<#include E<lt>fcntl.hE<gt>> /* For O_* constants */"
6692 #: build/C/man3/shm_open.3:37
6693 msgid "B<int shm_open(const char *>I<name>B<, int >I<oflag>B<, mode_t >I<mode>B<);>"
6697 #: build/C/man3/shm_open.3:39
6698 msgid "B<int shm_unlink(const char *>I<name>B<);>"
6702 #: build/C/man3/shm_open.3:41
6703 msgid "Link with I<-lrt>."
6707 #: build/C/man3/shm_open.3:53
6709 "B<shm_open>() creates and opens a new, or opens an existing, POSIX shared "
6710 "memory object. A POSIX shared memory object is in effect a handle which can "
6711 "be used by unrelated processes to B<mmap>(2) the same region of shared "
6712 "memory. The B<shm_unlink>() function performs the converse operation, "
6713 "removing an object previously created by B<shm_open>()."
6716 #. glibc allows the initial slash to be omitted, and makes
6717 #. multiple initial slashes equivalent to a single slash.
6718 #. This differs from the implementation of POSIX message queues.
6719 #. glibc allows subdirectory components in the name, in which
6720 #. case the subdirectory must exist under /dev/shm, and allow the
6721 #. required permissions if a user wants to create a shared memory
6722 #. object in that subdirectory.
6724 #: build/C/man3/shm_open.3:74
6726 "The operation of B<shm_open>() is analogous to that of B<open>(2). I<name> "
6727 "specifies the shared memory object to be created or opened. For portable "
6728 "use, a shared memory object should be identified by a name of the form "
6729 "I</somename>; that is, a null-terminated string of up to B<NAME_MAX> (i.e., "
6730 "255) characters consisting of an initial slash, followed by one or more "
6731 "characters, none of which are slashes."
6735 #: build/C/man3/shm_open.3:81
6737 "I<oflag> is a bit mask created by ORing together exactly one of B<O_RDONLY> "
6738 "or B<O_RDWR> and any of the other flags listed here:"
6742 #: build/C/man3/shm_open.3:81
6748 #: build/C/man3/shm_open.3:89
6750 "Open the object for read access. A shared memory object opened in this way "
6751 "can be B<mmap>(2)ed only for read (B<PROT_READ>) access."
6755 #: build/C/man3/shm_open.3:89
6761 #: build/C/man3/shm_open.3:92
6762 msgid "Open the object for read-write access."
6766 #: build/C/man3/shm_open.3:92
6771 #. In truth it is actually the filesystem IDs on Linux, but these
6772 #. are nearly always the same as the effective IDs. (MTK, Jul 05)
6774 #: build/C/man3/shm_open.3:112
6776 "Create the shared memory object if it does not exist. The user and group "
6777 "ownership of the object are taken from the corresponding effective IDs of "
6778 "the calling process, and the object's permission bits are set according to "
6779 "the low-order 9 bits of I<mode>, except that those bits set in the process "
6780 "file mode creation mask (see B<umask>(2)) are cleared for the new object. "
6781 "A set of macro constants which can be used to define I<mode> is listed in "
6782 "B<open>(2). (Symbolic definitions of these constants can be obtained by "
6783 "including I<E<lt>sys/stat.hE<gt>>.)"
6787 #: build/C/man3/shm_open.3:118
6789 "A new shared memory object initially has zero length\\(emthe size of the "
6790 "object can be set using B<ftruncate>(2). The newly allocated bytes of a "
6791 "shared memory object are automatically initialized to 0."
6795 #: build/C/man3/shm_open.3:118
6801 #: build/C/man3/shm_open.3:127
6803 "If B<O_CREAT> was also specified, and a shared memory object with the given "
6804 "I<name> already exists, return an error. The check for the existence of the "
6805 "object, and its creation if it does not exist, are performed atomically."
6809 #: build/C/man3/shm_open.3:127
6815 #: build/C/man3/shm_open.3:130
6816 msgid "If the shared memory object already exists, truncate it to zero bytes."
6820 #: build/C/man3/shm_open.3:133
6822 "Definitions of these flag values can be obtained by including "
6823 "I<E<lt>fcntl.hE<gt>>."
6827 #: build/C/man3/shm_open.3:144
6829 "On successful completion B<shm_open>() returns a new file descriptor "
6830 "referring to the shared memory object. This file descriptor is guaranteed "
6831 "to be the lowest-numbered file descriptor not previously opened within the "
6832 "process. The B<FD_CLOEXEC> flag (see B<fcntl>(2)) is set for the file "
6837 #: build/C/man3/shm_open.3:153
6839 "The file descriptor is normally used in subsequent calls to B<ftruncate>(2) "
6840 "(for a newly created object) and B<mmap>(2). After a call to B<mmap>(2) "
6841 "the file descriptor may be closed without affecting the memory mapping."
6845 #: build/C/man3/shm_open.3:171
6847 "The operation of B<shm_unlink>() is analogous to B<unlink>(2): it removes a "
6848 "shared memory object name, and, once all processes have unmapped the object, "
6849 "de-allocates and destroys the contents of the associated memory region. "
6850 "After a successful B<shm_unlink>(), attempts to B<shm_open>() an object "
6851 "with the same I<name> will fail (unless B<O_CREAT> was specified, in which "
6852 "case a new, distinct object is created)."
6856 #: build/C/man3/shm_open.3:180
6858 "On success, B<shm_open>() returns a nonnegative file descriptor. On "
6859 "failure, B<shm_open>() returns -1. B<shm_unlink>() returns 0 on success, "
6864 #: build/C/man3/shm_open.3:187
6866 "On failure, I<errno> is set to indicate the cause of the error. Values "
6867 "which may appear in I<errno> include the following:"
6871 #: build/C/man3/shm_open.3:192
6872 msgid "Permission to B<shm_unlink>() the shared memory object was denied."
6876 #: build/C/man3/shm_open.3:202
6878 "Permission was denied to B<shm_open>() I<name> in the specified I<mode>, or "
6879 "B<O_TRUNC> was specified and the caller does not have write permission on "
6884 #: build/C/man3/shm_open.3:202 build/C/man2/shmget.2:195
6890 #: build/C/man3/shm_open.3:213
6892 "Both B<O_CREAT> and B<O_EXCL> were specified to B<shm_open>() and the "
6893 "shared memory object specified by I<name> already exists."
6897 #: build/C/man3/shm_open.3:220
6898 msgid "The I<name> argument to B<shm_open>() was invalid."
6902 #: build/C/man3/shm_open.3:220
6908 #: build/C/man3/shm_open.3:223
6909 msgid "The process already has the maximum number of files open."
6913 #: build/C/man3/shm_open.3:223
6915 msgid "B<ENAMETOOLONG>"
6919 #: build/C/man3/shm_open.3:229
6920 msgid "The length of I<name> exceeds B<PATH_MAX>."
6924 #: build/C/man3/shm_open.3:233
6925 msgid "The limit on the total number of files open on the system has been reached."
6929 #: build/C/man3/shm_open.3:233 build/C/man3/shm_open.3:242 build/C/man2/shmget.2:209
6935 #: build/C/man3/shm_open.3:242
6937 "An attempt was made to B<shm_open>() a I<name> that did not exist, and "
6938 "B<O_CREAT> was not specified."
6942 #: build/C/man3/shm_open.3:249
6943 msgid "An attempt was to made to B<shm_unlink>() a I<name> that does not exist."
6947 #: build/C/man3/shm_open.3:251
6948 msgid "These functions are provided in glibc 2.2 and later."
6952 #: build/C/man3/shm_open.3:257
6954 "POSIX.1-2001 says that the group ownership of a newly created shared memory "
6955 "object is set to either the calling process's effective group ID or \"a "
6956 "system default group ID\"."
6960 #: build/C/man3/shm_open.3:266
6962 "POSIX leaves the behavior of the combination of B<O_RDONLY> and B<O_TRUNC> "
6963 "unspecified. On Linux, this will successfully truncate an existing shared "
6964 "memory object\\(emthis may not be so on other UNIX systems."
6968 #: build/C/man3/shm_open.3:271
6970 "The POSIX shared memory object implementation on Linux 2.4 makes use of a "
6971 "dedicated filesystem, which is normally mounted under I</dev/shm>."
6975 #: build/C/man3/shm_open.3:282
6977 "B<close>(2), B<fchmod>(2), B<fchown>(2), B<fcntl>(2), B<fstat>(2), "
6978 "B<ftruncate>(2), B<mmap>(2), B<open>(2), B<umask>(2), B<shm_overview>(7)"
6982 #: build/C/man7/shm_overview.7:27
6984 msgid "SHM_OVERVIEW"
6988 #: build/C/man7/shm_overview.7:27
6994 #: build/C/man7/shm_overview.7:30
6995 msgid "shm_overview - overview of POSIX shared memory"
6999 #: build/C/man7/shm_overview.7:33
7001 "The POSIX shared memory API allows processes to communicate information by "
7002 "sharing a region of memory."
7006 #: build/C/man7/shm_overview.7:35
7007 msgid "The interfaces employed in the API are:"
7011 #: build/C/man7/shm_overview.7:35
7013 msgid "B<shm_open>(3)"
7017 #: build/C/man7/shm_overview.7:42
7019 "Create and open a new object, or open an existing object. This is analogous "
7020 "to B<open>(2). The call returns a file descriptor for use by the other "
7021 "interfaces listed below."
7025 #: build/C/man7/shm_overview.7:42
7027 msgid "B<ftruncate>(2)"
7031 #: build/C/man7/shm_overview.7:46
7033 "Set the size of the shared memory object. (A newly created shared memory "
7034 "object has a length of zero.)"
7038 #: build/C/man7/shm_overview.7:50
7040 "Map the shared memory object into the virtual address space of the calling "
7045 #: build/C/man7/shm_overview.7:50
7047 msgid "B<munmap>(2)"
7051 #: build/C/man7/shm_overview.7:54
7053 "Unmap the shared memory object from the virtual address space of the calling "
7058 #: build/C/man7/shm_overview.7:54
7060 msgid "B<shm_unlink>(3)"
7064 #: build/C/man7/shm_overview.7:57
7065 msgid "Remove a shared memory object name."
7069 #: build/C/man7/shm_overview.7:57
7075 #: build/C/man7/shm_overview.7:62
7077 "Close the file descriptor allocated by B<shm_open>(3) when it is no longer "
7082 #: build/C/man7/shm_overview.7:62
7088 #: build/C/man7/shm_overview.7:76
7090 "Obtain a I<stat> structure that describes the shared memory object. Among "
7091 "the information returned by this call are the object's size (I<st_size>), "
7092 "permissions (I<st_mode>), owner (I<st_uid>), and group (I<st_gid>)."
7096 #: build/C/man7/shm_overview.7:76
7098 msgid "B<fchown>(2)"
7102 #: build/C/man7/shm_overview.7:79
7103 msgid "To change the ownership of a shared memory object."
7107 #: build/C/man7/shm_overview.7:79
7109 msgid "B<fchmod>(2)"
7113 #: build/C/man7/shm_overview.7:82
7114 msgid "To change the permissions of a shared memory object."
7118 #: build/C/man7/shm_overview.7:82
7124 #: build/C/man7/shm_overview.7:84
7125 msgid "POSIX shared memory is supported since Linux 2.4 and glibc 2.2."
7129 #: build/C/man7/shm_overview.7:84
7135 #: build/C/man7/shm_overview.7:89
7137 "POSIX shared memory objects have kernel persistence: a shared memory object "
7138 "will exist until the system is shut down, or until all processes have "
7139 "unmapped the object and it has been deleted with B<shm_unlink>(3)"
7143 #: build/C/man7/shm_overview.7:89
7149 #: build/C/man7/shm_overview.7:94
7151 "Programs using the POSIX shared memory API must be compiled with I<cc -lrt> "
7152 "to link against the real-time library, I<librt>."
7156 #: build/C/man7/shm_overview.7:94
7158 msgid "Accessing shared memory objects via the filesystem"
7162 #: build/C/man7/shm_overview.7:101
7164 "On Linux, shared memory objects are created in a (I<tmpfs>) virtual "
7165 "filesystem, normally mounted under I</dev/shm>. Since kernel 2.6.19, Linux "
7166 "supports the use of access control lists (ACLs) to control the permissions "
7167 "of objects in the virtual filesystem."
7171 #: build/C/man7/shm_overview.7:106
7173 "Typically, processes must synchronize their access to a shared memory "
7174 "object, using, for example, POSIX semaphores."
7178 #: build/C/man7/shm_overview.7:114
7180 "System V shared memory (B<shmget>(2), B<shmop>(2), etc.) is an older shared "
7181 "memory API. POSIX shared memory provides a simpler, and better designed "
7182 "interface; on the other hand POSIX shared memory is somewhat less widely "
7183 "available (especially on older systems) than System V shared memory."
7187 #: build/C/man7/shm_overview.7:127
7189 "B<fchmod>(2), B<fchown>(2), B<fstat>(2), B<ftruncate>(2), B<mmap>(2), "
7190 "B<mprotect>(2), B<munmap>(2), B<shmget>(2), B<shmop>(2), B<shm_open>(3), "
7191 "B<shm_unlink>(3), B<sem_overview>(7)"
7195 #: build/C/man2/shmctl.2:45
7201 #: build/C/man2/shmctl.2:45
7207 #: build/C/man2/shmctl.2:48
7208 msgid "shmctl - System V shared memory control"
7212 #: build/C/man2/shmctl.2:51 build/C/man2/shmget.2:44
7213 msgid "B<#include E<lt>sys/ipc.hE<gt>>"
7217 #: build/C/man2/shmctl.2:53 build/C/man2/shmget.2:46
7218 msgid "B<#include E<lt>sys/shm.hE<gt>>"
7222 #: build/C/man2/shmctl.2:55
7223 msgid "B<int shmctl(int >I<shmid>B<, int >I<cmd>B<, struct shmid_ds *>I<buf>B<);>"
7227 #: build/C/man2/shmctl.2:62
7229 "B<shmctl>() performs the control operation specified by I<cmd> on the "
7230 "System V shared memory segment whose identifier is given in I<shmid>."
7234 #: build/C/man2/shmctl.2:67
7236 "The I<buf> argument is a pointer to a I<shmid_ds> structure, defined in "
7237 "I<E<lt>sys/shm.hE<gt>> as follows:"
7241 #: build/C/man2/shmctl.2:81
7244 "struct shmid_ds {\n"
7245 " struct ipc_perm shm_perm; /* Ownership and permissions */\n"
7246 " size_t shm_segsz; /* Size of segment (bytes) */\n"
7247 " time_t shm_atime; /* Last attach time */\n"
7248 " time_t shm_dtime; /* Last detach time */\n"
7249 " time_t shm_ctime; /* Last change time */\n"
7250 " pid_t shm_cpid; /* PID of creator */\n"
7251 " pid_t shm_lpid; /* PID of last shmat(2)/shmdt(2) */\n"
7252 " shmatt_t shm_nattch; /* No. of current attaches */\n"
7258 #: build/C/man2/shmctl.2:89
7260 "The I<ipc_perm> structure is defined as follows (the highlighted fields are "
7261 "settable using B<IPC_SET>):"
7265 #: build/C/man2/shmctl.2:102
7268 "struct ipc_perm {\n"
7269 " key_t __key; /* Key supplied to shmget(2) */\n"
7270 " uid_t B<uid>; /* Effective UID of owner */\n"
7271 " gid_t B<gid>; /* Effective GID of owner */\n"
7272 " uid_t cuid; /* Effective UID of creator */\n"
7273 " gid_t cgid; /* Effective GID of creator */\n"
7274 " unsigned short B<mode>; /* B<Permissions> + SHM_DEST and\n"
7275 " SHM_LOCKED flags */\n"
7276 " unsigned short __seq; /* Sequence number */\n"
7281 #: build/C/man2/shmctl.2:108
7282 msgid "Valid values for I<cmd> are:"
7286 #: build/C/man2/shmctl.2:109
7292 #: build/C/man2/shmctl.2:118
7294 "Copy information from the kernel data structure associated with I<shmid> "
7295 "into the I<shmid_ds> structure pointed to by I<buf>. The caller must have "
7296 "read permission on the shared memory segment."
7300 #: build/C/man2/shmctl.2:118
7306 #: build/C/man2/shmctl.2:136
7308 "Write the values of some members of the I<shmid_ds> structure pointed to by "
7309 "I<buf> to the kernel data structure associated with this shared memory "
7310 "segment, updating also its I<shm_ctime> member. The following fields can be "
7311 "changed: I<shm_perm.uid>, I<shm_perm.gid>, and (the least significant 9 bits "
7312 "of) I<shm_perm.mode>. The effective UID of the calling process must match "
7313 "the owner (I<shm_perm.uid>) or creator (I<shm_perm.cuid>) of the shared "
7314 "memory segment, or the caller must be privileged."
7318 #: build/C/man2/shmctl.2:136
7324 #: build/C/man2/shmctl.2:153
7326 "Mark the segment to be destroyed. The segment will only actually be "
7327 "destroyed after the last process detaches it (i.e., when the I<shm_nattch> "
7328 "member of the associated structure I<shmid_ds> is zero). The caller must be "
7329 "the owner or creator, or be privileged. If a segment has been marked for "
7330 "destruction, then the (nonstandard) B<SHM_DEST> flag of the "
7331 "I<shm_perm.mode> field in the associated data structure retrieved by "
7332 "B<IPC_STAT> will be set."
7336 #: build/C/man2/shmctl.2:156
7338 "The caller I<must> ensure that a segment is eventually destroyed; otherwise "
7339 "its pages that were faulted in will remain in memory or swap."
7343 #: build/C/man2/shmctl.2:161
7345 "See also the description of I</proc/sys/kernel/shm_rmid_forced> in "
7350 #: build/C/man2/shmctl.2:161
7352 msgid "B<IPC_INFO> (Linux-specific)"
7356 #: build/C/man2/shmctl.2:174
7358 "Returns information about system-wide shared memory limits and parameters in "
7359 "the structure pointed to by I<buf>. This structure is of type I<shminfo> "
7360 "(thus, a cast is required), defined in I<E<lt>sys/shm.hE<gt>> if the "
7361 "B<_GNU_SOURCE> feature test macro is defined:"
7365 #: build/C/man2/shmctl.2:188
7368 "struct shminfo {\n"
7369 " unsigned long shmmax; /* Maximum segment size */\n"
7370 " unsigned long shmmin; /* Minimum segment size;\n"
7372 " unsigned long shmmni; /* Maximum number of segments */\n"
7373 " unsigned long shmseg; /* Maximum number of segments\n"
7374 " that a process can attach;\n"
7375 " unused within kernel */\n"
7376 " unsigned long shmall; /* Maximum number of pages of\n"
7377 " shared memory, system-wide */\n"
7382 #: build/C/man2/shmctl.2:201
7384 "The I<shmmni>, I<shmmax>, and I<shmall> settings can be changed via I</proc> "
7385 "files of the same name; see B<proc>(5) for details."
7389 #: build/C/man2/shmctl.2:201
7391 msgid "B<SHM_INFO> (Linux-specific)"
7395 #: build/C/man2/shmctl.2:212
7397 "Returns a I<shm_info> structure whose fields contain information about "
7398 "system resources consumed by shared memory. This structure is defined in "
7399 "I<E<lt>sys/shm.hE<gt>> if the B<_GNU_SOURCE> feature test macro is defined:"
7403 #: build/C/man2/shmctl.2:229
7406 "struct shm_info {\n"
7407 " int used_ids; /* # of currently existing\n"
7409 " unsigned long shm_tot; /* Total number of shared\n"
7410 " memory pages */\n"
7411 " unsigned long shm_rss; /* # of resident shared\n"
7412 " memory pages */\n"
7413 " unsigned long shm_swp; /* # of swapped shared\n"
7414 " memory pages */\n"
7415 " unsigned long swap_attempts;\n"
7416 " /* Unused since Linux 2.4 */\n"
7417 " unsigned long swap_successes;\n"
7418 " /* Unused since Linux 2.4 */\n"
7423 #: build/C/man2/shmctl.2:231
7425 msgid "B<SHM_STAT> (Linux-specific)"
7429 #: build/C/man2/shmctl.2:242
7431 "Returns a I<shmid_ds> structure as for B<IPC_STAT>. However, the I<shmid> "
7432 "argument is not a segment identifier, but instead an index into the kernel's "
7433 "internal array that maintains information about all shared memory segments "
7438 #: build/C/man2/shmctl.2:245
7440 "The caller can prevent or allow swapping of a shared memory segment with the "
7441 "following I<cmd> values:"
7445 #: build/C/man2/shmctl.2:246
7447 msgid "B<SHM_LOCK> (Linux-specific)"
7451 #: build/C/man2/shmctl.2:258
7453 "Prevent swapping of the shared memory segment. The caller must fault in any "
7454 "pages that are required to be present after locking is enabled. If a "
7455 "segment has been locked, then the (nonstandard) B<SHM_LOCKED> flag of the "
7456 "I<shm_perm.mode> field in the associated data structure retrieved by "
7457 "B<IPC_STAT> will be set."
7461 #: build/C/man2/shmctl.2:258
7463 msgid "B<SHM_UNLOCK> (Linux-specific)"
7467 #: build/C/man2/shmctl.2:261
7468 msgid "Unlock the segment, allowing it to be swapped out."
7471 #. There was some weirdness in 2.6.9: SHM_LOCK and SHM_UNLOCK could
7472 #. be applied to a segment, regardless of ownership of the segment.
7473 #. This was a botch-up in the move to RLIMIT_MEMLOCK, and was fixed
7474 #. in 2.6.10. MTK, May 2005
7476 #: build/C/man2/shmctl.2:279
7478 "In kernels before 2.6.10, only a privileged process could employ B<SHM_LOCK> "
7479 "and B<SHM_UNLOCK>. Since kernel 2.6.10, an unprivileged process can employ "
7480 "these operations if its effective UID matches the owner or creator UID of "
7481 "the segment, and (for B<SHM_LOCK>) the amount of memory to be locked falls "
7482 "within the B<RLIMIT_MEMLOCK> resource limit (see B<setrlimit>(2))."
7486 #: build/C/man2/shmctl.2:297
7488 "A successful B<IPC_INFO> or B<SHM_INFO> operation returns the index of the "
7489 "highest used entry in the kernel's internal array recording information "
7490 "about all shared memory segments. (This information can be used with "
7491 "repeated B<SHM_STAT> operations to obtain information about all shared "
7492 "memory segments on the system.) A successful B<SHM_STAT> operation returns "
7493 "the identifier of the shared memory segment whose index was given in "
7494 "I<shmid>. Other operations return 0 on success."
7498 #: build/C/man2/shmctl.2:301
7499 msgid "On error, -1 is returned, and I<errno> is set appropriately."
7503 #: build/C/man2/shmctl.2:310
7505 "B<IPC_STAT> or B<SHM_STAT> is requested and I<shm_perm.mode> does not allow "
7506 "read access for I<shmid>, and the calling process does not have the "
7507 "B<CAP_IPC_OWNER> capability."
7511 #: build/C/man2/shmctl.2:321
7513 "The argument I<cmd> has value B<IPC_SET> or B<IPC_STAT> but the address "
7514 "pointed to by I<buf> isn't accessible."
7518 #: build/C/man2/shmctl.2:321 build/C/man2/shmop.2:197
7524 #: build/C/man2/shmctl.2:324 build/C/man2/shmop.2:200
7525 msgid "I<shmid> points to a removed identifier."
7529 #: build/C/man2/shmctl.2:333
7531 "I<shmid> is not a valid identifier, or I<cmd> is not a valid command. Or: "
7532 "for a B<SHM_STAT> operation, the index value specified in I<shmid> referred "
7533 "to an array slot that is currently unused."
7537 #: build/C/man2/shmctl.2:344
7539 "(In kernels since 2.6.9), B<SHM_LOCK> was specified and the size of the "
7540 "to-be-locked segment would mean that the total bytes in locked shared memory "
7541 "segments would exceed the limit for the real user ID of the calling "
7542 "process. This limit is defined by the B<RLIMIT_MEMLOCK> soft resource limit "
7543 "(see B<setrlimit>(2))."
7547 #: build/C/man2/shmctl.2:349
7549 "B<IPC_STAT> is attempted, and the GID or UID value is too large to be stored "
7550 "in the structure pointed to by I<buf>."
7554 #: build/C/man2/shmctl.2:361
7556 "B<IPC_SET> or B<IPC_RMID> is attempted, and the effective user ID of the "
7557 "calling process is not that of the creator (found in I<shm_perm.cuid>), or "
7558 "the owner (found in I<shm_perm.uid>), and the process was not privileged "
7559 "(Linux: did not have the B<CAP_SYS_ADMIN> capability)."
7563 #: build/C/man2/shmctl.2:373
7565 "Or (in kernels before 2.6.9), B<SHM_LOCK> or B<SHM_UNLOCK> was specified, "
7566 "but the process was not privileged (Linux: did not have the B<CAP_IPC_LOCK> "
7567 "capability). (Since Linux 2.6.9, this error can also occur if the "
7568 "B<RLIMIT_MEMLOCK> is 0 and the caller is not privileged.)"
7571 #. SVr4 documents an additional error condition EMFILE.
7573 #: build/C/man2/shmctl.2:378 build/C/man2/shmget.2:235 build/C/man2/shmop.2:234
7574 msgid "SVr4, POSIX.1-2001."
7577 #. Like Linux, the FreeBSD man pages still document
7578 #. the inclusion of these header files.
7580 #: build/C/man2/shmctl.2:391 build/C/man2/shmget.2:251
7582 "The inclusion of I<E<lt>sys/types.hE<gt>> and I<E<lt>sys/ipc.hE<gt>> isn't "
7583 "required on Linux or by any version of POSIX. However, some old "
7584 "implementations required the inclusion of these header files, and the SVID "
7585 "also documented their inclusion. Applications intended to be portable to "
7586 "such old systems may need to include these header files."
7590 #: build/C/man2/shmctl.2:403
7592 "The B<IPC_INFO>, B<SHM_STAT> and B<SHM_INFO> operations are used by the "
7593 "B<ipcs>(1) program to provide information on allocated resources. In the "
7594 "future these may modified or moved to a I</proc> filesystem interface."
7598 #: build/C/man2/shmctl.2:411
7600 "Linux permits a process to attach (B<shmat>(2)) a shared memory segment "
7601 "that has already been marked for deletion using I<shmctl(IPC_RMID)>. This "
7602 "feature is not available on other UNIX implementations; portable "
7603 "applications should avoid relying on it."
7607 #: build/C/man2/shmctl.2:424
7609 "Various fields in a I<struct shmid_ds> were typed as I<short> under Linux "
7610 "2.2 and have become I<long> under Linux 2.4. To take advantage of this, a "
7611 "recompilation under glibc-2.1.91 or later should suffice. (The kernel "
7612 "distinguishes old and new calls by an B<IPC_64> flag in I<cmd>.)"
7616 #: build/C/man2/shmctl.2:431
7618 "B<mlock>(2), B<setrlimit>(2), B<shmget>(2), B<shmop>(2), B<capabilities>(7), "
7623 #: build/C/man2/shmget.2:38
7629 #: build/C/man2/shmget.2:41
7630 msgid "shmget - allocates a System V shared memory segment"
7634 #: build/C/man2/shmget.2:48
7635 msgid "B<int shmget(key_t >I<key>B<, size_t >I<size>B<, int >I<shmflg>B<);>"
7639 #: build/C/man2/shmget.2:72
7641 "B<shmget>() returns the identifier of the System V shared memory segment "
7642 "associated with the value of the argument I<key>. A new shared memory "
7643 "segment, with size equal to the value of I<size> rounded up to a multiple of "
7644 "B<PAGE_SIZE>, is created if I<key> has the value B<IPC_PRIVATE> or I<key> "
7645 "isn't B<IPC_PRIVATE>, no shared memory segment corresponding to I<key> "
7646 "exists, and B<IPC_CREAT> is specified in I<shmflg>."
7650 #: build/C/man2/shmget.2:91
7652 "If I<shmflg> specifies both B<IPC_CREAT> and B<IPC_EXCL> and a shared memory "
7653 "segment already exists for I<key>, then B<shmget>() fails with I<errno> set "
7654 "to B<EEXIST>. (This is analogous to the effect of the combination B<O_CREAT "
7655 "| O_EXCL> for B<open>(2).)"
7659 #: build/C/man2/shmget.2:95
7660 msgid "The value I<shmflg> is composed of:"
7664 #: build/C/man2/shmget.2:95
7666 msgid "B<IPC_CREAT>"
7670 #: build/C/man2/shmget.2:102
7672 "to create a new segment. If this flag is not used, then B<shmget>() will "
7673 "find the segment associated with I<key> and check to see if the user has "
7674 "permission to access the segment."
7678 #: build/C/man2/shmget.2:102
7684 #: build/C/man2/shmget.2:105
7685 msgid "used with B<IPC_CREAT> to ensure failure if the segment already exists."
7689 #: build/C/man2/shmget.2:105
7691 msgid "I<mode_flags>"
7695 #: build/C/man2/shmget.2:115
7697 "(least significant 9 bits) specifying the permissions granted to the owner, "
7698 "group, and world. These bits have the same format, and the same meaning, as "
7699 "the I<mode> argument of B<open>(2). Presently, the execute permissions are "
7700 "not used by the system."
7704 #: build/C/man2/shmget.2:115
7706 msgid "B<SHM_HUGETLB> (since Linux 2.6)"
7710 #: build/C/man2/shmget.2:121
7712 "Allocate the segment using \"huge pages.\" See the Linux kernel source file "
7713 "I<Documentation/vm/hugetlbpage.txt> for further information."
7717 #: build/C/man2/shmget.2:121
7719 msgid "B<SHM_NORESERVE> (since Linux 2.6.15)"
7722 #. As at 2.6.17-rc2, this flag has no effect if SHM_HUGETLB was also
7725 #: build/C/man2/shmget.2:140
7727 "This flag serves the same purpose as the B<mmap>(2) B<MAP_NORESERVE> flag. "
7728 "Do not reserve swap space for this segment. When swap space is reserved, "
7729 "one has the guarantee that it is possible to modify the segment. When swap "
7730 "space is not reserved one might get B<SIGSEGV> upon a write if no physical "
7731 "memory is available. See also the discussion of the file "
7732 "I</proc/sys/vm/overcommit_memory> in B<proc>(5)."
7736 #: build/C/man2/shmget.2:148
7738 "When a new shared memory segment is created, its contents are initialized to "
7739 "zero values, and its associated data structure, I<shmid_ds> (see "
7740 "B<shmctl>(2)), is initialized as follows:"
7744 #: build/C/man2/shmget.2:153
7746 "I<shm_perm.cuid> and I<shm_perm.uid> are set to the effective user ID of the "
7751 #: build/C/man2/shmget.2:158
7753 "I<shm_perm.cgid> and I<shm_perm.gid> are set to the effective group ID of "
7754 "the calling process."
7758 #: build/C/man2/shmget.2:163
7760 "The least significant 9 bits of I<shm_perm.mode> are set to the least "
7761 "significant 9 bit of I<shmflg>."
7765 #: build/C/man2/shmget.2:167
7766 msgid "I<shm_segsz> is set to the value of I<size>."
7770 #: build/C/man2/shmget.2:174
7771 msgid "I<shm_lpid>, I<shm_nattch>, I<shm_atime> and I<shm_dtime> are set to 0."
7775 #: build/C/man2/shmget.2:177
7776 msgid "I<shm_ctime> is set to the current time."
7780 #: build/C/man2/shmget.2:180
7782 "If the shared memory segment already exists, the permissions are verified, "
7783 "and a check is made to see if it is marked for destruction."
7787 #: build/C/man2/shmget.2:185
7789 "On success, a valid shared memory identifier is returned. On errir, -1 is "
7790 "returned, and I<errno> is set to indicate the error."
7794 #: build/C/man2/shmget.2:189
7795 msgid "On failure, I<errno> is set to one of the following:"
7799 #: build/C/man2/shmget.2:195
7801 "The user does not have permission to access the shared memory segment, and "
7802 "does not have the B<CAP_IPC_OWNER> capability."
7806 #: build/C/man2/shmget.2:199
7807 msgid "B<IPC_CREAT | IPC_EXCL> was specified and the segment exists."
7811 #: build/C/man2/shmget.2:205
7813 "A new segment was to be created and I<size> E<lt> B<SHMMIN> or I<size> E<gt> "
7814 "B<SHMMAX>, or no new segment was to be created, a segment with given key "
7815 "existed, but I<size> is greater than the size of that segment."
7819 #: build/C/man2/shmget.2:214
7820 msgid "No segment exists for the given I<key>, and B<IPC_CREAT> was not specified."
7824 #: build/C/man2/shmget.2:217
7825 msgid "No memory could be allocated for segment overhead."
7829 #: build/C/man2/shmget.2:225
7831 "All possible shared memory IDs have been taken (B<SHMMNI>), or allocating a "
7832 "segment of the requested I<size> would cause the system to exceed the "
7833 "system-wide limit on shared memory (B<SHMALL>)."
7837 #: build/C/man2/shmget.2:232
7839 "The B<SHM_HUGETLB> flag was specified, but the caller was not privileged "
7840 "(did not have the B<CAP_IPC_LOCK> capability)."
7844 #: build/C/man2/shmget.2:238
7845 msgid "B<SHM_HUGETLB> is a nonportable Linux extension."
7849 #: build/C/man2/shmget.2:261
7851 "B<IPC_PRIVATE> isn't a flag field but a I<key_t> type. If this special "
7852 "value is used for I<key>, the system call ignores everything but the least "
7853 "significant 9 bits of I<shmflg> and creates a new shared memory segment (on "
7858 #: build/C/man2/shmget.2:265
7860 "The following limits on shared memory segment resources affect the "
7865 #: build/C/man2/shmget.2:265
7871 #: build/C/man2/shmget.2:270
7873 "System wide maximum of shared memory pages (on Linux, this limit can be read "
7874 "and modified via I</proc/sys/kernel/shmall>)."
7878 #: build/C/man2/shmget.2:270
7884 #: build/C/man2/shmget.2:275
7886 "Maximum size in bytes for a shared memory segment: policy dependent (on "
7887 "Linux, this limit can be read and modified via I</proc/sys/kernel/shmmax>)."
7891 #: build/C/man2/shmget.2:275
7897 #: build/C/man2/shmget.2:281
7899 "Minimum size in bytes for a shared memory segment: implementation dependent "
7900 "(currently 1 byte, though B<PAGE_SIZE> is the effective minimum size)."
7904 #: build/C/man2/shmget.2:281
7909 #. Kernels between 2.4.x and 2.6.8 had an off-by-one error that meant
7910 #. that we could create one more segment than SHMMNI -- MTK
7911 #. This /proc file is not available in Linux 2.2 and earlier -- MTK
7913 #: build/C/man2/shmget.2:290
7915 "System wide maximum number of shared memory segments: implementation "
7916 "dependent (currently 4096, was 128 before Linux 2.3.99; on Linux, this limit "
7917 "can be read and modified via I</proc/sys/kernel/shmmni>)."
7921 #: build/C/man2/shmget.2:294
7923 "The implementation has no specific limits for the per-process maximum number "
7924 "of shared memory segments (B<SHMSEG>)."
7928 #: build/C/man2/shmget.2:300
7930 "Until version 2.3.30 Linux would return B<EIDRM> for a B<shmget>() on a "
7931 "shared memory segment scheduled for deletion."
7935 #: build/C/man2/shmget.2:306
7937 "The name choice B<IPC_PRIVATE> was perhaps unfortunate, B<IPC_NEW> would "
7938 "more clearly show its function."
7942 #: build/C/man2/shmget.2:314
7944 "B<shmat>(2), B<shmctl>(2), B<shmdt>(2), B<ftok>(3), B<capabilities>(7), "
7945 "B<shm_overview>(7), B<svipc>(7)"
7949 #: build/C/man2/shmop.2:41
7955 #: build/C/man2/shmop.2:44
7956 msgid "shmat, shmdt - System V shared memory operations"
7960 #: build/C/man2/shmop.2:48
7963 "B<#include E<lt>sys/types.hE<gt>>\n"
7964 "B<#include E<lt>sys/shm.hE<gt>>\n"
7968 #: build/C/man2/shmop.2:50
7971 "B<void *shmat(int >I<shmid>B<, const void *>I<shmaddr>B<, int "
7976 #: build/C/man2/shmop.2:52
7978 msgid "B<int shmdt(const void *>I<shmaddr>B<);>\n"
7982 #: build/C/man2/shmop.2:61
7984 "B<shmat>() attaches the System V shared memory segment identified by "
7985 "I<shmid> to the address space of the calling process. The attaching address "
7986 "is specified by I<shmaddr> with one of the following criteria:"
7990 #: build/C/man2/shmop.2:67
7992 "If I<shmaddr> is NULL, the system chooses a suitable (unused) address at "
7993 "which to attach the segment."
7997 #: build/C/man2/shmop.2:82
7999 "If I<shmaddr> isn't NULL and B<SHM_RND> is specified in I<shmflg>, the "
8000 "attach occurs at the address equal to I<shmaddr> rounded down to the nearest "
8001 "multiple of B<SHMLBA>. Otherwise I<shmaddr> must be a page-aligned address "
8002 "at which the attach occurs."
8006 #: build/C/man2/shmop.2:92
8008 "If B<SHM_RDONLY> is specified in I<shmflg>, the segment is attached for "
8009 "reading and the process must have read permission for the segment. "
8010 "Otherwise the segment is attached for read and write and the process must "
8011 "have read and write permission for the segment. There is no notion of a "
8012 "write-only shared memory segment."
8016 #: build/C/man2/shmop.2:107
8018 "The (Linux-specific) B<SHM_REMAP> flag may be specified in I<shmflg> to "
8019 "indicate that the mapping of the segment should replace any existing mapping "
8020 "in the range starting at I<shmaddr> and continuing for the size of the "
8021 "segment. (Normally an B<EINVAL> error would result if a mapping already "
8022 "exists in this address range.) In this case, I<shmaddr> must not be NULL."
8026 #: build/C/man2/shmop.2:114
8028 "The B<brk>(2) value of the calling process is not altered by the attach. "
8029 "The segment will automatically be detached at process exit. The same "
8030 "segment may be attached as a read and as a read-write one, and more than "
8031 "once, in the process's address space."
8035 #: build/C/man2/shmop.2:122
8037 "A successful B<shmat>() call updates the members of the I<shmid_ds> "
8038 "structure (see B<shmctl>(2)) associated with the shared memory segment as "
8043 #: build/C/man2/shmop.2:125
8044 msgid "I<shm_atime> is set to the current time."
8048 #: build/C/man2/shmop.2:128 build/C/man2/shmop.2:154
8049 msgid "I<shm_lpid> is set to the process-ID of the calling process."
8053 #: build/C/man2/shmop.2:131
8054 msgid "I<shm_nattch> is incremented by one."
8058 #: build/C/man2/shmop.2:142
8060 "B<shmdt>() detaches the shared memory segment located at the address "
8061 "specified by I<shmaddr> from the address space of the calling process. The "
8062 "to-be-detached segment must be currently attached with I<shmaddr> equal to "
8063 "the value returned by the attaching B<shmat>() call."
8067 #: build/C/man2/shmop.2:148
8069 "On a successful B<shmdt>() call the system updates the members of the "
8070 "I<shmid_ds> structure associated with the shared memory segment as follows:"
8074 #: build/C/man2/shmop.2:151
8075 msgid "I<shm_dtime> is set to the current time."
8079 #: build/C/man2/shmop.2:159
8081 "I<shm_nattch> is decremented by one. If it becomes 0 and the segment is "
8082 "marked for deletion, the segment is deleted."
8086 #: build/C/man2/shmop.2:163
8087 msgid "After a B<fork>(2) the child inherits the attached shared memory segments."
8091 #: build/C/man2/shmop.2:167
8093 "After an B<execve>(2) all attached shared memory segments are detached from "
8098 #: build/C/man2/shmop.2:171
8100 "Upon B<_exit>(2) all attached shared memory segments are detached from the "
8105 #: build/C/man2/shmop.2:179
8107 "On success B<shmat>() returns the address of the attached shared memory "
8108 "segment; on error I<(void\\ *)\\ -1> is returned, and I<errno> is set to "
8109 "indicate the cause of the error."
8113 #: build/C/man2/shmop.2:185
8115 "On success B<shmdt>() returns 0; on error -1 is returned, and I<errno> is "
8116 "set to indicate the cause of the error."
8120 #: build/C/man2/shmop.2:191
8121 msgid "When B<shmat>() fails, I<errno> is set to one of the following:"
8125 #: build/C/man2/shmop.2:197
8127 "The calling process does not have the required permissions for the requested "
8128 "attach type, and does not have the B<CAP_IPC_OWNER> capability."
8132 #: build/C/man2/shmop.2:214
8134 "Invalid I<shmid> value, unaligned (i.e., not page-aligned and B<SHM_RND> was "
8135 "not specified) or invalid I<shmaddr> value, or can't attach segment at "
8136 "I<shmaddr>, or B<SHM_REMAP> was specified and I<shmaddr> was NULL."
8140 #: build/C/man2/shmop.2:217
8141 msgid "Could not allocate memory for the descriptor or for the page tables."
8145 #: build/C/man2/shmop.2:223
8146 msgid "When B<shmdt>() fails, I<errno> is set as follows:"
8149 #. The following since 2.6.17-rc1:
8151 #: build/C/man2/shmop.2:231
8153 "There is no shared memory segment attached at I<shmaddr>; or, I<shmaddr> is "
8154 "not aligned on a page boundary."
8158 #: build/C/man2/shmop.2:250
8160 "In SVID 3 (or perhaps earlier) the type of the I<shmaddr> argument was "
8161 "changed from I<char\\ *> into I<const void\\ *>, and the returned type of "
8162 "B<shmat>() from I<char\\ *> into I<void\\ *>. (Linux libc4 and libc5 have "
8163 "the I<char\\ *> prototypes; glibc2 has I<void\\ *>.)"
8167 #: build/C/man2/shmop.2:262
8169 "Using B<shmat>() with I<shmaddr> equal to NULL is the preferred, portable "
8170 "way of attaching a shared memory segment. Be aware that the shared memory "
8171 "segment attached in this way may be attached at different addresses in "
8172 "different processes. Therefore, any pointers maintained within the shared "
8173 "memory must be made relative (typically to the starting address of the "
8174 "segment), rather than absolute."
8178 #: build/C/man2/shmop.2:267
8180 "On Linux, it is possible to attach a shared memory segment even if it is "
8181 "already marked to be deleted. However, POSIX.1-2001 does not specify this "
8182 "behavior and many other implementations do not support it."
8186 #: build/C/man2/shmop.2:270
8187 msgid "The following system parameter affects B<shmat>():"
8190 #. FIXME A good explanation of the rationale for the existence
8191 #. of SHMLBA would be useful here
8193 #: build/C/man2/shmop.2:270
8198 #. FIXME That last sentence isn't true for all Linux
8199 #. architectures (i.e., SHMLBA != PAGE_SIZE for some architectures)
8202 #: build/C/man2/shmop.2:283
8204 "Segment low boundary address multiple. Must be page aligned. For the "
8205 "current implementation the B<SHMLBA> value is B<PAGE_SIZE>."
8209 #: build/C/man2/shmop.2:287
8211 "The implementation places no intrinsic limit on the per-process maximum "
8212 "number of shared memory segments (B<SHMSEG>)."
8216 #: build/C/man2/shmop.2:295
8218 "B<brk>(2), B<mmap>(2), B<shmctl>(2), B<shmget>(2), B<capabilities>(7), "
8219 "B<shm_overview>(7), B<svipc>(7)"
8223 #: build/C/man2/subpage_prot.2:30
8225 msgid "SUBPAGE_PROT"
8229 #: build/C/man2/subpage_prot.2:30
8235 #: build/C/man2/subpage_prot.2:33
8236 msgid "subpage_prot - define a subpage protection for an address range"
8240 #: build/C/man2/subpage_prot.2:37
8243 "B<long subpage_prot(unsigned long >I<addr>B<, unsigned long >I<len>B<,>\n"
8244 "B< uint32_t *>I<map>B<);>\n"
8248 #: build/C/man2/subpage_prot.2:41
8249 msgid "I<Note>: There is no glibc wrapper for this system call; see NOTES."
8253 #: build/C/man2/subpage_prot.2:47
8255 "The PowerPC-specific B<subpage_prot>() system call provides the facility to "
8256 "control the access permissions on individual 4kB subpages on systems "
8257 "configured with a page size of 64kB."
8261 #: build/C/man2/subpage_prot.2:54
8263 "The protection map is applied to the memory pages in the region starting at "
8264 "I<addr> and continuing for I<len> bytes. Both of these arguments must be "
8265 "aligned to a 64-kB boundary."
8269 #: build/C/man2/subpage_prot.2:66
8271 "The protection map is specified in the buffer pointed to by I<map>. The map "
8272 "has 2 bits per 4kB subpage; thus each 32-bit word specifies the protections "
8273 "of 16 4kB subpages inside a 64kB page (so, the number of 32-bit words "
8274 "pointed to by I<map> should equate to the number of 64-kB pages specified by "
8275 "I<len>). Each 2-bit field in the protection map is either 0 to allow any "
8276 "access, 1 to prevent writes, or 2 or 3 to prevent all accesses."
8280 #: build/C/man2/subpage_prot.2:71
8282 "On success, B<subpage_prot>() returns 0. Otherwise, one of the error codes "
8283 "specified below is returned."
8287 #: build/C/man2/subpage_prot.2:77
8288 msgid "The buffer referred to by I<map> is not accessible."
8292 #: build/C/man2/subpage_prot.2:87
8294 "The I<addr> or I<len> arguments are incorrect. Both of these arguments must "
8295 "be aligned to a multiple of the system page size, and they must not refer to "
8296 "a region outside of the address space of the process or to a region that "
8297 "consists of huge pages."
8301 #: build/C/man2/subpage_prot.2:90 build/C/man2/sync_file_range.2:166
8302 msgid "Out of memory."
8306 #: build/C/man2/subpage_prot.2:96
8308 "This system call is provided on the PowerPC architecture since Linux "
8309 "2.6.25. The system call is provided only if the kernel is configured with "
8310 "B<CONFIG_PPC_64K_PAGES>. No library support is provided."
8314 #: build/C/man2/subpage_prot.2:101
8316 "Glibc does not provide a wrapper for this system call; call it using "
8321 #: build/C/man2/subpage_prot.2:106
8323 "Normal page protections (at the 64-kB page level) also apply; the subpage "
8324 "protection mechanism is an additional constraint, so putting 0 in a 2-bit "
8325 "field won't allow writes to a page that is otherwise write-protected."
8329 #: build/C/man2/subpage_prot.2:106
8334 #. In the initial implementation, it was the case that:
8335 #. In fact the whole process is switched to use 4k hardware pages when the
8336 #. subpage_prot system call is used, but this could be improved in future
8337 #. to switch only the affected segments.
8338 #. But Paul Mackerass says (Oct 2010): I'm pretty sure we now only switch
8339 #. the affected segment, not the whole process.
8341 #: build/C/man2/subpage_prot.2:128
8343 "This system call is provided to assist writing emulators that operate using "
8344 "64-kB pages on PowerPC systems. When emulating systems such as x86, which "
8345 "uses a smaller page size, the emulator can no longer use the "
8346 "memory-management unit (MMU) and normal system calls for controlling page "
8347 "protections. (The emulator could emulate the MMU by checking and possibly "
8348 "remapping the address for each memory access in software, but that is slow.) "
8349 "The idea is that the emulator supplies an array of protection masks to apply "
8350 "to a specified range of virtual addresses. These masks are applied at the "
8351 "level where hardware page-table entries (PTEs) are inserted into the "
8352 "hardware page table based on the Linux PTEs, so the Linux PTEs are not "
8353 "affected. Implicit in this is that the regions of the address space that "
8354 "are protected are switched to use 4-kB hardware pages rather than 64-kB "
8355 "hardware pages (on machines with hardware 64-kB page support)."
8359 #: build/C/man2/subpage_prot.2:131
8360 msgid "B<mprotect>(2), B<syscall>(2)"
8364 #: build/C/man2/subpage_prot.2:134
8365 msgid "I<Documentation/vm/hugetlbpage.txt> in the Linux kernel source tree"
8369 #: build/C/man2/sync_file_range.2:30
8371 msgid "SYNC_FILE_RANGE"
8375 #: build/C/man2/sync_file_range.2:33
8376 msgid "sync_file_range - sync a file segment with disk"
8380 #: build/C/man2/sync_file_range.2:37
8383 "B<#define _GNU_SOURCE> /* See feature_test_macros(7) */\n"
8384 "B<#include E<lt>fcntl.hE<gt>>\n"
8388 #: build/C/man2/sync_file_range.2:40
8391 "B<int sync_file_range(int >I<fd>B<, off64_t >I<offset>B<, off64_t "
8393 "B< unsigned int >I<flags>B<);>\n"
8397 #: build/C/man2/sync_file_range.2:47
8399 "B<sync_file_range>() permits fine control when synchronizing the open file "
8400 "referred to by the file descriptor I<fd> with disk."
8404 #: build/C/man2/sync_file_range.2:61
8406 "I<offset> is the starting byte of the file range to be synchronized. "
8407 "I<nbytes> specifies the length of the range to be synchronized, in bytes; if "
8408 "I<nbytes> is zero, then all bytes from I<offset> through to the end of file "
8409 "are synchronized. Synchronization is in units of the system page size: "
8410 "I<offset> is rounded down to a page boundary; I<(offset+nbytes-1)> is "
8411 "rounded up to a page boundary."
8415 #: build/C/man2/sync_file_range.2:65
8416 msgid "The I<flags> bit-mask argument can include any of the following values:"
8420 #: build/C/man2/sync_file_range.2:65
8422 msgid "B<SYNC_FILE_RANGE_WAIT_BEFORE>"
8426 #: build/C/man2/sync_file_range.2:70
8428 "Wait upon write-out of all pages in the specified range that have already "
8429 "been submitted to the device driver for write-out before performing any "
8434 #: build/C/man2/sync_file_range.2:70 build/C/man2/sync_file_range.2:120
8436 msgid "B<SYNC_FILE_RANGE_WRITE>"
8440 #: build/C/man2/sync_file_range.2:76
8442 "Initiate write-out of all dirty pages in the specified range which are not "
8443 "presently submitted write-out. Note that even this may block if you attempt "
8444 "to write more than request queue size."
8448 #: build/C/man2/sync_file_range.2:76
8450 msgid "B<SYNC_FILE_RANGE_WAIT_AFTER>"
8454 #: build/C/man2/sync_file_range.2:80
8455 msgid "Wait upon write-out of all pages in the range after performing any write."
8459 #: build/C/man2/sync_file_range.2:84
8460 msgid "Specifying I<flags> as 0 is permitted, as a no-op."
8464 #: build/C/man2/sync_file_range.2:84
8470 #: build/C/man2/sync_file_range.2:100
8472 "This system call is extremely dangerous and should not be used in portable "
8473 "programs. None of these operations writes out the file's metadata. "
8474 "Therefore, unless the application is strictly performing overwrites of "
8475 "already-instantiated disk blocks, there are no guarantees that the data will "
8476 "be available after a crash. There is no user interface to know if a write "
8477 "is purely an overwrite. On filesystems using copy-on-write semantics (e.g., "
8478 "I<btrfs>) an overwrite of existing allocated blocks is impossible. When "
8479 "writing into preallocated space, many filesystems also require calls into "
8480 "the block allocator, which this system call does not sync out to disk. This "
8481 "system call does not flush disk write caches and thus does not provide any "
8482 "data integrity on systems with volatile disk write caches."
8486 #: build/C/man2/sync_file_range.2:100
8488 msgid "Some details"
8492 #: build/C/man2/sync_file_range.2:108
8494 "B<SYNC_FILE_RANGE_WAIT_BEFORE> and B<SYNC_FILE_RANGE_WAIT_AFTER> will detect "
8495 "any I/O errors or B<ENOSPC> conditions and will return these to the caller."
8499 #: build/C/man2/sync_file_range.2:112
8500 msgid "Useful combinations of the I<flags> bits are:"
8504 #: build/C/man2/sync_file_range.2:112
8506 msgid "B<SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE>"
8510 #: build/C/man2/sync_file_range.2:120
8512 "Ensures that all pages in the specified range which were dirty when "
8513 "B<sync_file_range>() was called are placed under write-out. This is a "
8514 "start-write-for-data-integrity operation."
8518 #: build/C/man2/sync_file_range.2:127
8520 "Start write-out of all dirty pages in the specified range which are not "
8521 "presently under write-out. This is an asynchronous flush-to-disk "
8522 "operation. This is not suitable for data integrity operations."
8526 #: build/C/man2/sync_file_range.2:127
8528 msgid "B<SYNC_FILE_RANGE_WAIT_BEFORE> (or B<SYNC_FILE_RANGE_WAIT_AFTER>)"
8532 #: build/C/man2/sync_file_range.2:134
8534 "Wait for completion of write-out of all pages in the specified range. This "
8535 "can be used after an earlier B<SYNC_FILE_RANGE_WAIT_BEFORE | "
8536 "SYNC_FILE_RANGE_WRITE> operation to wait for completion of that operation, "
8537 "and obtain its result."
8541 #: build/C/man2/sync_file_range.2:134
8544 "B<SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE | "
8545 "SYNC_FILE_RANGE_WAIT_AFTER>"
8549 #: build/C/man2/sync_file_range.2:141
8551 "This is a write-for-data-integrity operation that will ensure that all pages "
8552 "in the specified range which were dirty when B<sync_file_range>() was "
8553 "called are committed to disk."
8557 #: build/C/man2/sync_file_range.2:147
8559 "On success, B<sync_file_range>() returns 0; on failure -1 is returned and "
8560 "I<errno> is set to indicate the error."
8564 #: build/C/man2/sync_file_range.2:152
8565 msgid "I<fd> is not a valid file descriptor."
8569 #: build/C/man2/sync_file_range.2:160
8570 msgid "I<flags> specifies an invalid bit; or I<offset> or I<nbytes> is invalid."
8574 #: build/C/man2/sync_file_range.2:163
8579 #: build/C/man2/sync_file_range.2:169
8580 msgid "Out of disk space."
8583 #. FIXME . (bug?) Actually, how can 'fd' refer to a symbolic link (S_ISLNK)?
8584 #. (In user space at least) it isn't possible to obtain a file descriptor
8585 #. for a symbolic link.
8587 #: build/C/man2/sync_file_range.2:177
8589 "I<fd> refers to something other than a regular file, a block device, a "
8590 "directory, or a symbolic link."
8594 #: build/C/man2/sync_file_range.2:180
8595 msgid "B<sync_file_range>() appeared on Linux in kernel 2.6.17."
8599 #: build/C/man2/sync_file_range.2:183
8601 "This system call is Linux-specific, and should be avoided in portable "
8605 #. See kernel commit edd5cd4a9424f22b0fa08bef5e299d41befd5622
8607 #: build/C/man2/sync_file_range.2:200
8609 "Some architectures (e.g., PowerPC, ARM) need 64-bit arguments to be aligned "
8610 "in a suitable pair of registers. On such architectures, the call signature "
8611 "of B<sync_file_range>() shown in the SYNOPSIS would force a register to be "
8612 "wasted as padding between the I<fd> and I<offset> arguments. (See "
8613 "B<syscall>(2) for details.) Therefore, these architectures define a "
8614 "different system call that orders the arguments suitably:"
8618 #: build/C/man2/sync_file_range.2:205
8621 "B<int sync_file_range2(int >I<fd>B<, unsigned int >I<flags>B<,>\n"
8622 "B< off64_t >I<offset>B<, off64_t >I<nbytes>B<);>\n"
8626 #: build/C/man2/sync_file_range.2:210
8628 "The behavior of this system call is otherwise exactly the same as "
8629 "B<sync_file_range>()."
8633 #: build/C/man2/sync_file_range.2:221
8635 "A system call with this signature first appeared on the ARM architecture in "
8636 "Linux 2.6.20, with the name B<arm_sync_file_range>(). It was renamed in "
8637 "Linux 2.6.22, when the analogous system call was added for PowerPC. On "
8638 "architectures where glibc support is provided, glibc transparently wraps "
8639 "B<sync_file_range2>() under the name B<sync_file_range>()."
8643 #: build/C/man2/sync_file_range.2:226
8644 msgid "B<fdatasync>(2), B<fsync>(2), B<msync>(2), B<sync>(2)"