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: 2015-01-23 22:25+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/get_mempolicy.2:27
26 #: build/C/man2/get_mempolicy.2:27
32 #: build/C/man2/get_mempolicy.2:27 build/C/man2/getcpu.2:12 build/C/man2/mbind.2:45 build/C/man2/migrate_pages.2:18 build/C/man2/move_pages.2:19 build/C/man7/numa.7:29 build/C/man2/set_mempolicy.2:27
38 #: build/C/man2/get_mempolicy.2:27 build/C/man2/getcpu.2:12 build/C/man2/mbind.2:45 build/C/man2/migrate_pages.2:18 build/C/man2/move_pages.2:19 build/C/man7/numa.7:29 build/C/man2/set_mempolicy.2:27
40 msgid "Linux Programmer's Manual"
44 #: build/C/man2/get_mempolicy.2:28 build/C/man2/getcpu.2:13 build/C/man2/mbind.2:46 build/C/man2/migrate_pages.2:19 build/C/man2/move_pages.2:20 build/C/man7/numa.7:30 build/C/man2/set_mempolicy.2:28
50 #: build/C/man2/get_mempolicy.2:30
51 msgid "get_mempolicy - retrieve NUMA memory policy for a process"
55 #: build/C/man2/get_mempolicy.2:30 build/C/man2/getcpu.2:15 build/C/man2/mbind.2:48 build/C/man2/migrate_pages.2:21 build/C/man2/move_pages.2:22 build/C/man2/set_mempolicy.2:30
61 #: build/C/man2/get_mempolicy.2:32
62 msgid "B<#include E<lt>numaif.hE<gt>>"
66 #: build/C/man2/get_mempolicy.2:37
69 "B<int get_mempolicy(int *>I<mode>B<, unsigned long *>I<nodemask>B<,>\n"
70 "B< unsigned long >I<maxnode>B<, unsigned long "
72 "B< unsigned long >I<flags>B<);>\n"
76 #: build/C/man2/get_mempolicy.2:39 build/C/man2/mbind.2:57 build/C/man2/set_mempolicy.2:38
78 msgid "Link with I<-lnuma>.\n"
82 #: build/C/man2/get_mempolicy.2:40 build/C/man2/getcpu.2:25 build/C/man2/mbind.2:58 build/C/man2/migrate_pages.2:31 build/C/man2/move_pages.2:31 build/C/man7/numa.7:32 build/C/man2/set_mempolicy.2:39
88 #: build/C/man2/get_mempolicy.2:45
90 "B<get_mempolicy>() retrieves the NUMA policy of the calling process or of a "
91 "memory address, depending on the setting of I<flags>."
95 #: build/C/man2/get_mempolicy.2:50 build/C/man2/set_mempolicy.2:54
97 "A NUMA machine has different memory controllers with different distances to "
98 "specific CPUs. The memory policy defines from which node memory is "
99 "allocated for the process."
103 #: build/C/man2/get_mempolicy.2:67
105 "If I<flags> is specified as 0, then information about the calling process's "
106 "default policy (as set by B<set_mempolicy>(2)) is returned. The policy "
107 "returned [I<mode> and I<nodemask>] may be used to restore the process's "
108 "policy to its state at the time of the call to B<get_mempolicy>() using "
109 "B<set_mempolicy>(2)."
113 #: build/C/man2/get_mempolicy.2:89
115 "If I<flags> specifies B<MPOL_F_MEMS_ALLOWED> (available since Linux 2.6.24), "
116 "the I<mode> argument is ignored and the set of nodes [memories] that the "
117 "process is allowed to specify in subsequent calls to B<mbind>(2) or "
118 "B<set_mempolicy>(2) [in the absence of any I<mode flags>] is returned in "
119 "I<nodemask>. It is not permitted to combine B<MPOL_F_MEMS_ALLOWED> with "
120 "either B<MPOL_F_ADDR> or B<MPOL_F_NODE>."
124 #: build/C/man2/get_mempolicy.2:103
126 "If I<flags> specifies B<MPOL_F_ADDR>, then information is returned about the "
127 "policy governing the memory address given in I<addr>. This policy may be "
128 "different from the process's default policy if B<mbind>(2) or one of the "
129 "helper functions described in B<numa>(3) has been used to establish a "
130 "policy for the memory range containing I<addr>."
134 #: build/C/man2/get_mempolicy.2:124
136 "If the I<mode> argument is not NULL, then B<get_mempolicy>() will store the "
137 "policy mode and any optional I<mode flags> of the requested NUMA policy in "
138 "the location pointed to by this argument. If I<nodemask> is not NULL, then "
139 "the nodemask associated with the policy will be stored in the location "
140 "pointed to by this argument. I<maxnode> specifies the number of node IDs "
141 "that can be stored into I<nodemask>\\(emthat is, the maximum node ID plus "
142 "one. The value specified by I<maxnode> is always rounded to a multiple of "
143 "I<sizeof(unsigned long)>."
147 #: build/C/man2/get_mempolicy.2:141
149 "If I<flags> specifies both B<MPOL_F_NODE> and B<MPOL_F_ADDR>, "
150 "B<get_mempolicy>() will return the node ID of the node on which the address "
151 "I<addr> is allocated into the location pointed to by I<mode>. If no page "
152 "has yet been allocated for the specified address, B<get_mempolicy>() will "
153 "allocate a page as if the process had performed a read [load] access to that "
154 "address, and return the ID of the node where that page was allocated."
157 #. Note: code returns next interleave node via 'mode' argument -Lee Schermerhorn
159 #: build/C/man2/get_mempolicy.2:166
161 "If I<flags> specifies B<MPOL_F_NODE>, but not B<MPOL_F_ADDR>, and the "
162 "process's current policy is B<MPOL_INTERLEAVE>, then B<get_mempolicy>() "
163 "will return in the location pointed to by a non-NULL I<mode> argument, the "
164 "node ID of the next node that will be used for interleaving of internal "
165 "kernel pages allocated on behalf of the process. These allocations include "
166 "pages for memory-mapped files in process memory ranges mapped using the "
167 "B<mmap>(2) call with the B<MAP_PRIVATE> flag for read accesses, and in "
168 "memory ranges mapped with the B<MAP_SHARED> flag for all accesses."
172 #: build/C/man2/get_mempolicy.2:168
173 msgid "Other flag values are reserved."
177 #: build/C/man2/get_mempolicy.2:171
178 msgid "For an overview of the possible policies see B<set_mempolicy>(2)."
182 #: build/C/man2/get_mempolicy.2:171 build/C/man2/getcpu.2:61 build/C/man2/mbind.2:309 build/C/man2/migrate_pages.2:88 build/C/man2/move_pages.2:143 build/C/man2/set_mempolicy.2:220
188 #: build/C/man2/get_mempolicy.2:178
190 "On success, B<get_mempolicy>() returns 0; on error, -1 is returned and "
191 "I<errno> is set to indicate the error."
195 #: build/C/man2/get_mempolicy.2:178 build/C/man2/getcpu.2:66 build/C/man2/mbind.2:317 build/C/man2/migrate_pages.2:96 build/C/man2/move_pages.2:152 build/C/man2/set_mempolicy.2:227
201 #: build/C/man2/get_mempolicy.2:179 build/C/man2/getcpu.2:67 build/C/man2/mbind.2:319 build/C/man2/move_pages.2:161 build/C/man2/set_mempolicy.2:228
207 #: build/C/man2/get_mempolicy.2:186 build/C/man2/set_mempolicy.2:235
209 "Part of all of the memory range specified by I<nodemask> and I<maxnode> "
210 "points outside your accessible address space."
214 #: build/C/man2/get_mempolicy.2:186 build/C/man2/mbind.2:330 build/C/man2/move_pages.2:164 build/C/man2/set_mempolicy.2:235
220 #: build/C/man2/get_mempolicy.2:230
222 "The value specified by I<maxnode> is less than the number of node IDs "
223 "supported by the system. Or I<flags> specified values other than "
224 "B<MPOL_F_NODE> or B<MPOL_F_ADDR>; or I<flags> specified B<MPOL_F_ADDR> and "
225 "I<addr> is NULL, or I<flags> did not specify B<MPOL_F_ADDR> and I<addr> is "
226 "not NULL. Or, I<flags> specified B<MPOL_F_NODE> but not B<MPOL_F_ADDR> and "
227 "the current process policy is not B<MPOL_INTERLEAVE>. Or, I<flags> "
228 "specified B<MPOL_F_MEMS_ALLOWED> with either B<MPOL_F_ADDR> or "
229 "B<MPOL_F_NODE>. (And there are other B<EINVAL> cases.)"
233 #: build/C/man2/get_mempolicy.2:230 build/C/man2/getcpu.2:70 build/C/man2/mbind.2:402 build/C/man2/migrate_pages.2:112 build/C/man2/move_pages.2:192 build/C/man2/set_mempolicy.2:275
239 #: build/C/man2/get_mempolicy.2:234
241 "The B<get_mempolicy>() system call was added to the Linux kernel in version "
246 #: build/C/man2/get_mempolicy.2:234 build/C/man2/getcpu.2:73 build/C/man2/mbind.2:406 build/C/man2/migrate_pages.2:116 build/C/man2/move_pages.2:195 build/C/man7/numa.7:148 build/C/man2/set_mempolicy.2:279
248 msgid "CONFORMING TO"
252 #: build/C/man2/get_mempolicy.2:236 build/C/man2/mbind.2:408 build/C/man2/migrate_pages.2:118 build/C/man2/move_pages.2:197 build/C/man2/set_mempolicy.2:281
253 msgid "This system call is Linux-specific."
257 #: build/C/man2/get_mempolicy.2:236 build/C/man2/getcpu.2:76 build/C/man2/mbind.2:408 build/C/man2/migrate_pages.2:118 build/C/man2/move_pages.2:197 build/C/man7/numa.7:150 build/C/man2/set_mempolicy.2:281
263 #: build/C/man2/get_mempolicy.2:239 build/C/man2/mbind.2:411 build/C/man2/migrate_pages.2:121 build/C/man2/move_pages.2:200 build/C/man2/set_mempolicy.2:289
264 msgid "For information on library support, see B<numa>(7)."
268 #: build/C/man2/get_mempolicy.2:239 build/C/man2/getcpu.2:139 build/C/man2/mbind.2:460 build/C/man2/migrate_pages.2:148 build/C/man2/move_pages.2:227 build/C/man7/numa.7:183 build/C/man2/set_mempolicy.2:289
274 #: build/C/man2/get_mempolicy.2:247
276 "B<getcpu>(2), B<mbind>(2), B<mmap>(2), B<set_mempolicy>(2), B<numa>(3), "
277 "B<numa>(7), B<numactl>(8)"
281 #: build/C/man2/get_mempolicy.2:247 build/C/man2/getcpu.2:145 build/C/man2/mbind.2:471 build/C/man2/migrate_pages.2:161 build/C/man2/move_pages.2:237 build/C/man7/numa.7:191 build/C/man2/set_mempolicy.2:298
287 #: build/C/man2/get_mempolicy.2:255 build/C/man2/getcpu.2:153 build/C/man2/mbind.2:479 build/C/man2/migrate_pages.2:169 build/C/man2/move_pages.2:245 build/C/man7/numa.7:199 build/C/man2/set_mempolicy.2:306
289 "This page is part of release 3.78 of the Linux I<man-pages> project. A "
290 "description of the project, information about reporting bugs, and the latest "
291 "version of this page, can be found at "
292 "\\%http://www.kernel.org/doc/man-pages/."
296 #: build/C/man2/getcpu.2:12
302 #: build/C/man2/getcpu.2:12
308 #: build/C/man2/getcpu.2:15
309 msgid "getcpu - determine CPU and NUMA node on which the calling thread is running"
313 #: build/C/man2/getcpu.2:18
315 msgid "B<#include E<lt>linux/getcpu.hE<gt>>\n"
319 #: build/C/man2/getcpu.2:21
322 "B<int getcpu(unsigned *>I<cpu>B<, unsigned *>I<node>B<, struct getcpu_cache "
327 #: build/C/man2/getcpu.2:25
328 msgid "I<Note>: There is no glibc wrapper for this system call; see NOTES."
332 #: build/C/man2/getcpu.2:42
334 "The B<getcpu>() system call identifies the processor and node on which the "
335 "calling thread or process is currently running and writes them into the "
336 "integers pointed to by the I<cpu> and I<node> arguments. The processor is a "
337 "unique small integer identifying a CPU. The node is a unique small "
338 "identifier identifying a NUMA node. When either I<cpu> or I<node> is NULL "
339 "nothing is written to the respective pointer."
343 #: build/C/man2/getcpu.2:46
345 "The third argument to this system call is nowadays unused, and should be "
346 "specified as NULL unless portability to Linux 2.6.23 or earlier is required "
351 #: build/C/man2/getcpu.2:61
353 "The information placed in I<cpu> is guaranteed to be current only at the "
354 "time of the call: unless the CPU affinity has been fixed using "
355 "B<sched_setaffinity>(2), the kernel might change the CPU at any time. "
356 "(Normally this does not happen because the scheduler tries to minimize "
357 "movements between CPUs to keep caches hot, but it is possible.) The caller "
358 "must allow for the possibility that the information returned in I<cpu> and "
359 "I<node> is no longer current by the time the call returns."
363 #: build/C/man2/getcpu.2:66
365 "On success, 0 is returned. On error, -1 is returned, and I<errno> is set "
370 #: build/C/man2/getcpu.2:70
371 msgid "Arguments point outside the calling process's address space."
375 #: build/C/man2/getcpu.2:73
376 msgid "B<getcpu>() was added in kernel 2.6.19 for x86_64 and i386."
380 #: build/C/man2/getcpu.2:76
381 msgid "B<getcpu>() is Linux-specific."
385 #: build/C/man2/getcpu.2:82
387 "Linux makes a best effort to make this call as fast possible. The intention "
388 "of B<getcpu>() is to allow programs to make optimizations with per-CPU data "
389 "or for NUMA optimization."
393 #: build/C/man2/getcpu.2:88
395 "Glibc does not provide a wrapper for this system call; call it using "
396 "B<syscall>(2); or use B<sched_getcpu>(3) instead."
399 #. commit 4307d1e5ada595c87f9a4d16db16ba5edb70dcb1
400 #. Author: Ingo Molnar <mingo@elte.hu>
401 #. Date: Wed Nov 7 18:37:48 2007 +0100
402 #. x86: ignore the sys_getcpu() tcache parameter
404 #. ===== Before kernel 2.6.24: =====
407 #. .IR "struct getcpu_cache"
408 #. that is used as a cache by
410 #. The caller should put the cache into a thread-local variable
411 #. if the process is multithreaded,
412 #. because the cache cannot be shared between different threads.
417 #. will use it to speed up operation.
418 #. The information inside the cache is private to the system call
419 #. and should not be accessed by the user program.
420 #. The information placed in the cache can change between kernel releases.
422 #. When no cache is specified
425 #. but always retrieve the current CPU and node information.
429 #. However, the cached information is updated only once per jiffy (see
431 #. This means that the information could theoretically be out of date,
432 #. although in practice the scheduler's attempt to maintain
433 #. soft CPU affinity means that the information is unlikely to change
434 #. over the course of the caching interval.
436 #: build/C/man2/getcpu.2:139
438 "The I<tcache> argument is unused since Linux 2.6.24. In earlier kernels, if "
439 "this argument was non-NULL, then it specified a pointer to a "
440 "caller-allocated buffer in thread-local storage that was used to provide a "
441 "caching mechanism for B<getcpu>(). Use of the cache could speed B<getcpu>() "
442 "calls, at the cost that there was a very small chance that the returned "
443 "information would be out of date. The caching mechanism was considered to "
444 "cause problems when migrating threads between CPUs, and so the argument is "
449 #: build/C/man2/getcpu.2:145
451 "B<mbind>(2), B<sched_setaffinity>(2), B<set_mempolicy>(2), "
452 "B<sched_getcpu>(3), B<cpuset>(7)"
456 #: build/C/man2/mbind.2:45
462 #: build/C/man2/mbind.2:45
468 #: build/C/man2/mbind.2:48
469 msgid "mbind - set memory policy for a memory range"
473 #: build/C/man2/mbind.2:51 build/C/man2/migrate_pages.2:24 build/C/man2/move_pages.2:25 build/C/man2/set_mempolicy.2:33
475 msgid "B<#include E<lt>numaif.hE<gt>>\n"
479 #: build/C/man2/mbind.2:55
482 "B<long mbind(void *>I<addr>B<, unsigned long >I<len>B<, int >I<mode>B<,>\n"
483 "B< const unsigned long *>I<nodemask>B<, unsigned long "
485 "B< unsigned >I<flags>B<);>\n"
489 #: build/C/man2/mbind.2:68
491 "B<mbind>() sets the NUMA memory policy, which consists of a policy mode and "
492 "zero or more nodes, for the memory range starting with I<addr> and "
493 "continuing for I<len> bytes. The memory policy defines from which node "
494 "memory is allocated."
498 #: build/C/man2/mbind.2:90
500 "If the memory range specified by the I<addr> and I<len> arguments includes "
501 "an \"anonymous\" region of memory\\(emthat is a region of memory created "
502 "using the B<mmap>(2) system call with the B<MAP_ANONYMOUS>\\(emor a "
503 "memory-mapped file, mapped using the B<mmap>(2) system call with the "
504 "B<MAP_PRIVATE> flag, pages will be allocated only according to the specified "
505 "policy when the application writes [stores] to the page. For anonymous "
506 "regions, an initial read access will use a shared page in the kernel "
507 "containing all zeros. For a file mapped with B<MAP_PRIVATE>, an initial "
508 "read access will allocate pages according to the process policy of the "
509 "process that causes the page to be allocated. This may not be the process "
510 "that called B<mbind>()."
514 #: build/C/man2/mbind.2:98
516 "The specified policy will be ignored for any B<MAP_SHARED> mappings in the "
517 "specified memory range. Rather the pages will be allocated according to the "
518 "process policy of the process that caused the page to be allocated. Again, "
519 "this may not be the process that called B<mbind>()."
523 #: build/C/man2/mbind.2:115
525 "If the specified memory range includes a shared memory region created using "
526 "the B<shmget>(2) system call and attached using the B<shmat>(2) system "
527 "call, pages allocated for the anonymous or shared memory region will be "
528 "allocated according to the policy specified, regardless which process "
529 "attached to the shared memory segment causes the allocation. If, however, "
530 "the shared memory region was created with the B<SHM_HUGETLB> flag, the huge "
531 "pages will be allocated according to the policy specified only if the page "
532 "allocation is caused by the process that calls B<mbind>() for that region."
536 #: build/C/man2/mbind.2:126
538 "By default, B<mbind>() has an effect only for new allocations; if the pages "
539 "inside the range have been already touched before setting the policy, then "
540 "the policy has no effect. This default behavior may be overridden by the "
541 "B<MPOL_MF_MOVE> and B<MPOL_MF_MOVE_ALL> flags described below."
545 #: build/C/man2/mbind.2:141
547 "The I<mode> argument must specify one of B<MPOL_DEFAULT>, B<MPOL_BIND>, "
548 "B<MPOL_INTERLEAVE>, or B<MPOL_PREFERRED>. All policy modes except "
549 "B<MPOL_DEFAULT> require the caller to specify via the I<nodemask> argument, "
550 "the node or nodes to which the mode applies."
554 #: build/C/man2/mbind.2:149
556 "The I<mode> argument may also include an optional I<mode flag >. The "
557 "supported I<mode flags> are:"
561 #: build/C/man2/mbind.2:149
563 msgid "B<MPOL_F_STATIC_NODES> (since Linux-2.6.26)"
567 #: build/C/man2/mbind.2:159
569 "A nonempty I<nodemask> specifies physical node ids. Linux does not remap "
570 "the I<nodemask> when the process moves to a different cpuset context, nor "
571 "when the set of nodes allowed by the process's current cpuset context "
576 #: build/C/man2/mbind.2:159
578 msgid "B<MPOL_F_RELATIVE_NODES> (since Linux-2.6.26)"
582 #: build/C/man2/mbind.2:165 build/C/man2/set_mempolicy.2:113
584 "A nonempty I<nodemask> specifies node ids that are relative to the set of "
585 "node ids allowed by the process's current cpuset."
589 #: build/C/man2/mbind.2:193
591 "I<nodemask> points to a bit mask of nodes containing up to I<maxnode> bits. "
592 "The bit mask size is rounded to the next multiple of I<sizeof(unsigned "
593 "long)>, but the kernel will use bits only up to I<maxnode>. A NULL value of "
594 "I<nodemask> or a I<maxnode> value of zero specifies the empty set of nodes. "
595 "If the value of I<maxnode> is zero, the I<nodemask> argument is ignored. "
596 "Where a I<nodemask> is required, it must contain at least one node that is "
597 "on-line, allowed by the process's current cpuset context [unless the "
598 "B<MPOL_F_STATIC_NODES> mode flag is specified], and contains memory."
602 #: build/C/man2/mbind.2:215
604 "The B<MPOL_DEFAULT> mode requests that any nondefault policy be removed, "
605 "restoring default behavior. When applied to a range of memory via "
606 "B<mbind>(), this means to use the process policy, which may have been set "
607 "with B<set_mempolicy>(2). If the mode of the process policy is also "
608 "B<MPOL_DEFAULT>, the system-wide default policy will be used. The "
609 "system-wide default policy allocates pages on the node of the CPU that "
610 "triggers the allocation. For B<MPOL_DEFAULT>, the I<nodemask> and "
611 "I<maxnode> arguments must be specify the empty set of nodes."
615 #: build/C/man2/mbind.2:232
617 "The B<MPOL_BIND> mode specifies a strict policy that restricts memory "
618 "allocation to the nodes specified in I<nodemask>. If I<nodemask> specifies "
619 "more than one node, page allocations will come from the node with the lowest "
620 "numeric node ID first, until that node contains no free memory. Allocations "
621 "will then come from the node with the next highest node ID specified in "
622 "I<nodemask> and so forth, until none of the specified nodes contain free "
623 "memory. Pages will not be allocated from any node not specified in the "
628 #: build/C/man2/mbind.2:245
630 "The B<MPOL_INTERLEAVE> mode specifies that page allocations be interleaved "
631 "across the set of nodes specified in I<nodemask>. This optimizes for "
632 "bandwidth instead of latency by spreading out pages and memory accesses to "
633 "those pages across multiple nodes. To be effective the memory area should "
634 "be fairly large, at least 1MB or bigger with a fairly uniform access "
635 "pattern. Accesses to a single page of the area will still be limited to the "
636 "memory bandwidth of a single node."
640 #: build/C/man2/mbind.2:264
642 "B<MPOL_PREFERRED> sets the preferred node for allocation. The kernel will "
643 "try to allocate pages from this node first and fall back to other nodes if "
644 "the preferred nodes is low on free memory. If I<nodemask> specifies more "
645 "than one node ID, the first node in the mask will be selected as the "
646 "preferred node. If the I<nodemask> and I<maxnode> arguments specify the "
647 "empty set, then the memory is allocated on the node of the CPU that "
648 "triggered the allocation. This is the only way to specify \"local "
649 "allocation\" for a range of memory via B<mbind>()."
652 #. According to the kernel code, the following is not true
653 #. --Lee Schermerhorn
654 #. In 2.6.16 or later the kernel will also try to move pages
655 #. to the requested node with this flag.
657 #: build/C/man2/mbind.2:280
659 "If B<MPOL_MF_STRICT> is passed in I<flags> and I<mode> is not "
660 "B<MPOL_DEFAULT>, then the call will fail with the error B<EIO> if the "
661 "existing pages in the memory range don't follow the policy."
665 #: build/C/man2/mbind.2:293
667 "If B<MPOL_MF_MOVE> is specified in I<flags>, then the kernel will attempt to "
668 "move all the existing pages in the memory range so that they follow the "
669 "policy. Pages that are shared with other processes will not be moved. If "
670 "B<MPOL_MF_STRICT> is also specified, then the call will fail with the error "
671 "B<EIO> if some pages could not be moved."
674 #. ---------------------------------------------------------------
676 #: build/C/man2/mbind.2:309
678 "If B<MPOL_MF_MOVE_ALL> is passed in I<flags>, then the kernel will attempt "
679 "to move all existing pages in the memory range regardless of whether other "
680 "processes use the pages. The calling process must be privileged "
681 "(B<CAP_SYS_NICE>) to use this flag. If B<MPOL_MF_STRICT> is also "
682 "specified, then the call will fail with the error B<EIO> if some pages could "
686 #. ---------------------------------------------------------------
688 #: build/C/man2/mbind.2:317
690 "On success, B<mbind>() returns 0; on error, -1 is returned and I<errno> is "
691 "set to indicate the error."
695 #: build/C/man2/mbind.2:330
697 "Part or all of the memory range specified by I<nodemask> and I<maxnode> "
698 "points outside your accessible address space. Or, there was an unmapped "
699 "hole in the specified memory range specified by I<addr> and I<len>."
702 #. As at 2.6.23, this limit is "a page worth of bits", e.g.,
703 #. 8 * 4096 bits, assuming a 4kB page size.
705 #: build/C/man2/mbind.2:378
707 "An invalid value was specified for I<flags> or I<mode>; or I<addr + len> was "
708 "less than I<addr>; or I<addr> is not a multiple of the system page size. "
709 "Or, I<mode> is B<MPOL_DEFAULT> and I<nodemask> specified a nonempty set; or "
710 "I<mode> is B<MPOL_BIND> or B<MPOL_INTERLEAVE> and I<nodemask> is empty. Or, "
711 "I<maxnode> exceeds a kernel-imposed limit. Or, I<nodemask> specifies one or "
712 "more node IDs that are greater than the maximum supported node ID. Or, none "
713 "of the node IDs specified by I<nodemask> are on-line and allowed by the "
714 "process's current cpuset context, or none of the specified nodes contain "
715 "memory. Or, the I<mode> argument specified both B<MPOL_F_STATIC_NODES> and "
716 "B<MPOL_F_RELATIVE_NODES>."
720 #: build/C/man2/mbind.2:378
726 #: build/C/man2/mbind.2:389
728 "B<MPOL_MF_STRICT> was specified and an existing page was already on a node "
729 "that does not follow the policy; or B<MPOL_MF_MOVE> or B<MPOL_MF_MOVE_ALL> "
730 "was specified and the kernel was unable to move all existing pages in the "
735 #: build/C/man2/mbind.2:389 build/C/man2/set_mempolicy.2:272
741 #: build/C/man2/mbind.2:392 build/C/man2/set_mempolicy.2:275
742 msgid "Insufficient kernel memory was available."
746 #: build/C/man2/mbind.2:392 build/C/man2/migrate_pages.2:97 build/C/man2/move_pages.2:180
751 #. ---------------------------------------------------------------
753 #: build/C/man2/mbind.2:402
755 "The I<flags> argument included the B<MPOL_MF_MOVE_ALL> flag and the caller "
756 "does not have the B<CAP_SYS_NICE> privilege."
760 #: build/C/man2/mbind.2:406
761 msgid "The B<mbind>() system call was added to the Linux kernel in version 2.6.7."
765 #: build/C/man2/mbind.2:416
767 "NUMA policy is not supported on a memory-mapped file range that was mapped "
768 "with the B<MAP_SHARED> flag."
772 #: build/C/man2/mbind.2:448
774 "The B<MPOL_DEFAULT> mode can have different effects for B<mbind>() and "
775 "B<set_mempolicy>(2). When B<MPOL_DEFAULT> is specified for "
776 "B<set_mempolicy>(2), the process's policy reverts to system default policy "
777 "or local allocation. When B<MPOL_DEFAULT> is specified for a range of "
778 "memory using B<mbind>(), any pages subsequently allocated for that range "
779 "will use the process's policy, as set by B<set_mempolicy>(2). This "
780 "effectively removes the explicit policy from the specified range, \"falling "
781 "back\" to a possibly nondefault policy. To select explicit \"local "
782 "allocation\" for a memory range, specify a I<mode> of B<MPOL_PREFERRED> with "
783 "an empty set of nodes. This method will work for B<set_mempolicy>(2), as "
788 #: build/C/man2/mbind.2:452
790 "Support for huge page policy was added with 2.6.16. For interleave policy "
791 "to be effective on huge page mappings the policied memory needs to be tens "
792 "of megabytes or larger."
796 #: build/C/man2/mbind.2:455
797 msgid "B<MPOL_MF_STRICT> is ignored on huge page mappings."
801 #: build/C/man2/mbind.2:460
803 "B<MPOL_MF_MOVE> and B<MPOL_MF_MOVE_ALL> are available only on Linux 2.6.16 "
808 #: build/C/man2/mbind.2:471
810 "B<get_mempolicy>(2), B<getcpu>(2), B<mmap>(2), B<set_mempolicy>(2), "
811 "B<shmat>(2), B<shmget>(2), B<numa>(3), B<cpuset>(7), B<numa>(7), "
816 #: build/C/man2/migrate_pages.2:18
818 msgid "MIGRATE_PAGES"
822 #: build/C/man2/migrate_pages.2:18
828 #: build/C/man2/migrate_pages.2:21
829 msgid "migrate_pages - move all pages in a process to another set of nodes"
833 #: build/C/man2/migrate_pages.2:28
836 "B<long migrate_pages(int >I<pid>B<, unsigned long >I<maxnode,>\n"
837 "B< const unsigned long *>I<old_nodes,>\n"
838 "B< const unsigned long *>I<new_nodes>B<);>\n"
842 #: build/C/man2/migrate_pages.2:31 build/C/man2/move_pages.2:31
843 msgid "Link with I<-lnuma>."
847 #: build/C/man2/migrate_pages.2:47
849 "B<migrate_pages>() attempts to move all pages of the process I<pid> that "
850 "are in memory nodes I<old_nodes> to the memory nodes in I<new_nodes>. Pages "
851 "not located in any node in I<old_nodes> will not be migrated. As far as "
852 "possible, the kernel maintains the relative topology relationship inside "
853 "I<old_nodes> during the migration to I<new_nodes>."
857 #: build/C/man2/migrate_pages.2:69
859 "The I<old_nodes> and I<new_nodes> arguments are pointers to bit masks of "
860 "node numbers, with up to I<maxnode> bits in each mask. These masks are "
861 "maintained as arrays of unsigned I<long> integers (in the last I<long> "
862 "integer, the bits beyond those specified by I<maxnode> are ignored). The "
863 "I<maxnode> argument is the maximum node number in the bit mask plus one "
864 "(this is the same as in B<mbind>(2), but different from B<select>(2))."
868 #: build/C/man2/migrate_pages.2:83
870 "The I<pid> argument is the ID of the process whose pages are to be moved. "
871 "To move pages in another process, the caller must be privileged "
872 "(B<CAP_SYS_NICE>) or the real or effective user ID of the calling process "
873 "must match the real or saved-set user ID of the target process. If I<pid> "
874 "is 0, then B<migrate_pages>() moves pages of the calling process."
878 #: build/C/man2/migrate_pages.2:88
880 "Pages shared with another process will be moved only if the initiating "
881 "process has the B<CAP_SYS_NICE> privilege."
885 #: build/C/man2/migrate_pages.2:96
887 "On success B<migrate_pages>() returns the number of pages that could not be "
888 "moved (i.e., a return of zero means that all pages were successfully "
889 "moved). On error, it returns -1, and sets I<errno> to indicate the error."
893 #: build/C/man2/migrate_pages.2:106
895 "Insufficient privilege (B<CAP_SYS_NICE>) to move pages of the process "
896 "specified by I<pid>, or insufficient privilege (B<CAP_SYS_NICE>) to access "
897 "the specified target nodes."
901 #: build/C/man2/migrate_pages.2:106 build/C/man2/move_pages.2:189
906 #. FIXME Document the other errors that can occur for migrate_pages()
908 #: build/C/man2/migrate_pages.2:112
909 msgid "No process matching I<pid> could be found."
913 #: build/C/man2/migrate_pages.2:116
915 "The B<migrate_pages>() system call first appeared on Linux in version "
920 #: build/C/man2/migrate_pages.2:130
922 "Use B<get_mempolicy>(2) with the B<MPOL_F_MEMS_ALLOWED> flag to obtain the "
923 "set of nodes that are allowed by the calling process's cpuset. Note that "
924 "this information is subject to change at any time by manual or automatic "
925 "reconfiguration of the cpuset."
929 #: build/C/man2/migrate_pages.2:142
931 "Use of B<migrate_pages>() may result in pages whose location (node) "
932 "violates the memory policy established for the specified addresses (see "
933 "B<mbind>(2)) and/or the specified process (see B<set_mempolicy>(2)). That "
934 "is, memory policy does not constrain the destination nodes used by "
935 "B<migrate_pages>()."
939 #: build/C/man2/migrate_pages.2:148 build/C/man2/move_pages.2:227
941 "The I<E<lt>numaif.hE<gt>> header is not included with glibc, but requires "
942 "installing I<libnuma-devel> or a similar package."
946 #: build/C/man2/migrate_pages.2:158 build/C/man2/move_pages.2:237
948 "B<get_mempolicy>(2), B<mbind>(2), B<set_mempolicy>(2), B<numa>(3), "
949 "B<numa_maps>(5), B<cpuset>(7), B<numa>(7), B<migratepages>(8), "
954 #: build/C/man2/migrate_pages.2:161
955 msgid "I<Documentation/vm/page_migration> in the Linux kernel source tree"
959 #: build/C/man2/move_pages.2:19
965 #: build/C/man2/move_pages.2:19
971 #: build/C/man2/move_pages.2:22
972 msgid "move_pages - move individual pages of a process to another node"
976 #: build/C/man2/move_pages.2:28
979 "B<long move_pages(int >I<pid>B<, unsigned long count, void **>I<pages>B<,>\n"
980 "B< const int *>I<nodes>B<, int *>I<status>B<, int "
985 #: build/C/man2/move_pages.2:44
987 "B<move_pages>() moves the specified I<pages> of the process I<pid> to the "
988 "memory nodes specified by I<nodes>. The result of the move is reflected in "
989 "I<status>. The I<flags> indicate constraints on the pages to be moved."
993 #: build/C/man2/move_pages.2:57
995 "I<pid> is the ID of the process in which pages are to be moved. To move "
996 "pages in another process, the caller must be privileged (B<CAP_SYS_NICE>) "
997 "or the real or effective user ID of the calling process must match the real "
998 "or saved-set user ID of the target process. If I<pid> is 0, then "
999 "B<move_pages>() moves pages of the calling process."
1003 #: build/C/man2/move_pages.2:65
1005 "I<count> is the number of pages to move. It defines the size of the three "
1006 "arrays I<pages>, I<nodes>, and I<status>."
1009 #. FIXME Describe the result if pointers in the 'pages' array are
1010 #. not aligned to page boundaries
1012 #: build/C/man2/move_pages.2:73
1014 "I<pages> is an array of pointers to the pages that should be moved. These "
1015 "are pointers that should be aligned to page boundaries. Addresses are "
1016 "specified as seen by the process specified by I<pid>."
1020 #: build/C/man2/move_pages.2:86
1022 "I<nodes> is an array of integers that specify the desired location for each "
1023 "page. Each element in the array is a node number. I<nodes> can also be "
1024 "NULL, in which case B<move_pages>() does not move any pages but instead "
1025 "will return the node where each page currently resides, in the I<status> "
1026 "array. Obtaining the status of each page may be necessary to determine "
1027 "pages that need to be moved."
1031 #: build/C/man2/move_pages.2:92
1033 "I<status> is an array of integers that return the status of each page. The "
1034 "array contains valid values only if B<move_pages>() did not return an "
1039 #: build/C/man2/move_pages.2:104
1041 "I<flags> specify what types of pages to move. B<MPOL_MF_MOVE> means that "
1042 "only pages that are in exclusive use by the process are to be moved. "
1043 "B<MPOL_MF_MOVE_ALL> means that pages shared between multiple processes can "
1044 "also be moved. The process must be privileged (B<CAP_SYS_NICE>) to use "
1045 "B<MPOL_MF_MOVE_ALL>."
1049 #: build/C/man2/move_pages.2:104
1051 msgid "Page states in the status array"
1055 #: build/C/man2/move_pages.2:108
1056 msgid "The following values can be returned in each element of the I<status> array."
1060 #: build/C/man2/move_pages.2:108
1062 msgid "B<0..MAX_NUMNODES>"
1066 #: build/C/man2/move_pages.2:111
1067 msgid "Identifies the node on which the page resides."
1071 #: build/C/man2/move_pages.2:111
1077 #: build/C/man2/move_pages.2:116
1079 "The page is mapped by multiple processes and can be moved only if "
1080 "B<MPOL_MF_MOVE_ALL> is specified."
1084 #: build/C/man2/move_pages.2:116
1090 #: build/C/man2/move_pages.2:122
1092 "The page is currently busy and cannot be moved. Try again later. This "
1093 "occurs if a page is undergoing I/O or another kernel subsystem is holding a "
1094 "reference to the page."
1098 #: build/C/man2/move_pages.2:122
1104 #: build/C/man2/move_pages.2:125
1105 msgid "This is a zero page or the memory area is not mapped by the process."
1109 #: build/C/man2/move_pages.2:125
1115 #: build/C/man2/move_pages.2:132
1117 "Unable to write back a page. The page has to be written back in order to "
1118 "move it since the page is dirty and the filesystem does not provide a "
1119 "migration function that would allow the move of dirty pages."
1123 #: build/C/man2/move_pages.2:132
1129 #: build/C/man2/move_pages.2:137
1131 "A dirty page cannot be moved. The filesystem does not provide a migration "
1132 "function and has no ability to write back pages."
1136 #: build/C/man2/move_pages.2:137
1142 #: build/C/man2/move_pages.2:140
1143 msgid "The page is not present."
1147 #: build/C/man2/move_pages.2:140
1153 #: build/C/man2/move_pages.2:143
1154 msgid "Unable to allocate memory on target node."
1157 #. FIXME . Is the following quite true: does the wrapper in numactl
1158 #. do the right thing?
1160 #: build/C/man2/move_pages.2:152
1162 "On success B<move_pages>() returns zero. On error, it returns -1, and sets "
1163 "I<errno> to indicate the error."
1167 #: build/C/man2/move_pages.2:153
1173 #: build/C/man2/move_pages.2:156
1174 msgid "Too many pages to move."
1178 #: build/C/man2/move_pages.2:156
1183 #. FIXME Clarify "current cpuset" in the description of the EACCESS error.
1184 #. Is that the cpuset of the caller or the target?
1186 #: build/C/man2/move_pages.2:161
1187 msgid "One of the target nodes is not allowed by the current cpuset."
1191 #: build/C/man2/move_pages.2:164
1192 msgid "Parameter array could not be accessed."
1196 #: build/C/man2/move_pages.2:171
1198 "Flags other than B<MPOL_MF_MOVE> and B<MPOL_MF_MOVE_ALL> was specified or an "
1199 "attempt was made to migrate pages of a kernel thread."
1203 #: build/C/man2/move_pages.2:171
1209 #: build/C/man2/move_pages.2:174
1210 msgid "One of the target nodes is not online."
1214 #: build/C/man2/move_pages.2:174
1220 #: build/C/man2/move_pages.2:180
1222 "No pages were found that require moving. All pages are either already on "
1223 "the target node, not present, had an invalid address or could not be moved "
1224 "because they were mapped by multiple processes."
1228 #: build/C/man2/move_pages.2:189
1230 "The caller specified B<MPOL_MF_MOVE_ALL> without sufficient privileges "
1231 "(B<CAP_SYS_NICE>). Or, the caller attempted to move pages of a process "
1232 "belonging to another user but did not have privilege to do so "
1233 "(B<CAP_SYS_NICE>)."
1237 #: build/C/man2/move_pages.2:192
1238 msgid "Process does not exist."
1242 #: build/C/man2/move_pages.2:195
1243 msgid "B<move_pages>() first appeared on Linux in version 2.6.18."
1246 #. FIXME Clarify "current cpuset". Is that the cpuset of the caller
1249 #: build/C/man2/move_pages.2:211
1251 "Use B<get_mempolicy>(2) with the B<MPOL_F_MEMS_ALLOWED> flag to obtain the "
1252 "set of nodes that are allowed by the current cpuset. Note that this "
1253 "information is subject to change at any time by manual or automatic "
1254 "reconfiguration of the cpuset."
1258 #: build/C/man2/move_pages.2:221
1260 "Use of this function may result in pages whose location (node) violates the "
1261 "memory policy established for the specified addresses (See B<mbind>(2)) "
1262 "and/or the specified process (See B<set_mempolicy>(2)). That is, memory "
1263 "policy does not constrain the destination nodes used by B<move_pages>()."
1267 #: build/C/man7/numa.7:29
1273 #: build/C/man7/numa.7:29
1279 #: build/C/man7/numa.7:32
1280 msgid "numa - overview of Non-Uniform Memory Architecture"
1284 #: build/C/man7/numa.7:43
1286 "Non-Uniform Memory Access (NUMA) refers to multiprocessor systems whose "
1287 "memory is divided into multiple memory nodes. The access time of a memory "
1288 "node depends on the relative locations of the accessing CPU and the accessed "
1289 "node. (This contrasts with a symmetric multiprocessor system, where the "
1290 "access time for all of the memory is the same for all CPUs.) Normally, each "
1291 "CPU on a NUMA system has a local memory node whose contents can be accessed "
1292 "faster than the memory in the node local to another CPU or the memory on a "
1293 "bus shared by all CPUs."
1297 #: build/C/man7/numa.7:43
1299 msgid "NUMA system calls"
1303 #: build/C/man7/numa.7:54
1305 "The Linux kernel implements the following NUMA-related system calls: "
1306 "B<get_mempolicy>(2), B<mbind>(2), B<migrate_pages>(2), B<move_pages>(2), and "
1307 "B<set_mempolicy>(2). However, applications should normally use the "
1308 "interface provided by I<libnuma>; see \"Library Support\" below."
1312 #: build/C/man7/numa.7:54
1314 msgid "/proc/[number]/numa_maps (since Linux 2.6.14)"
1317 #. See also Changelog-2.6.14
1319 #: build/C/man7/numa.7:58
1321 "This file displays information about a process's NUMA memory policy and "
1326 #: build/C/man7/numa.7:62
1328 "Each line contains information about a memory range used by the process, "
1329 "displaying\\(emamong other information\\(emthe effective memory policy for "
1330 "that memory range and on which nodes the pages have been allocated."
1334 #: build/C/man7/numa.7:70
1336 "I<numa_maps> is a read-only file. When I</proc/E<lt>pidE<gt>/numa_maps> is "
1337 "read, the kernel will scan the virtual address space of the process and "
1338 "report how memory is used. One line is displayed for each unique memory "
1339 "range of the process."
1343 #: build/C/man7/numa.7:77
1345 "The first field of each line shows the starting address of the memory "
1346 "range. This field allows a correlation with the contents of the "
1347 "I</proc/E<lt>pidE<gt>/maps> file, which contains the end address of the "
1348 "range and other information, such as the access permissions and sharing."
1352 #: build/C/man7/numa.7:85
1354 "The second field shows the memory policy currently in effect for the memory "
1355 "range. Note that the effective policy is not necessarily the policy "
1356 "installed by the process for that memory range. Specifically, if the "
1357 "process installed a \"default\" policy for that range, the effective policy "
1358 "for that range will be the process policy, which may or may not be "
1363 #: build/C/man7/numa.7:88
1365 "The rest of the line contains information about the pages allocated in the "
1366 "memory range, as follows:"
1370 #: build/C/man7/numa.7:88
1372 msgid "I<NE<lt>nodeE<gt>=E<lt>nr_pagesE<gt>>"
1376 #: build/C/man7/numa.7:101
1378 "The number of pages allocated on I<E<lt>nodeE<gt>>. I<E<lt>nr_pagesE<gt>> "
1379 "includes only pages currently mapped by the process. Page migration and "
1380 "memory reclaim may have temporarily unmapped pages associated with this "
1381 "memory range. These pages may show up again only after the process has "
1382 "attempted to reference them. If the memory range represents a shared memory "
1383 "area or file mapping, other processes may currently have additional pages "
1384 "mapped in a corresponding memory range."
1388 #: build/C/man7/numa.7:101
1390 msgid "I<file=E<lt>filenameE<gt>>"
1394 #: build/C/man7/numa.7:107
1396 "The file backing the memory range. If the file is mapped as private, write "
1397 "accesses may have generated COW (Copy-On-Write) pages in this memory range. "
1398 "These pages are displayed as anonymous pages."
1402 #: build/C/man7/numa.7:107
1408 #: build/C/man7/numa.7:110
1409 msgid "Memory range is used for the heap."
1413 #: build/C/man7/numa.7:110
1419 #: build/C/man7/numa.7:113
1420 msgid "Memory range is used for the stack."
1424 #: build/C/man7/numa.7:113
1430 #: build/C/man7/numa.7:117
1432 "Huge memory range. The page counts shown are huge pages and not regular "
1437 #: build/C/man7/numa.7:117
1439 msgid "I<anon=E<lt>pagesE<gt>>"
1443 #: build/C/man7/numa.7:120
1444 msgid "The number of anonymous page in the range."
1448 #: build/C/man7/numa.7:120
1450 msgid "I<dirty=E<lt>pagesE<gt>>"
1454 #: build/C/man7/numa.7:123
1455 msgid "Number of dirty pages."
1459 #: build/C/man7/numa.7:123
1461 msgid "I<mapped=E<lt>pagesE<gt>>"
1465 #: build/C/man7/numa.7:130
1466 msgid "Total number of mapped pages, if different from I<dirty> and I<anon> pages."
1470 #: build/C/man7/numa.7:130
1472 msgid "I<mapmax=E<lt>countE<gt>>"
1476 #: build/C/man7/numa.7:136
1478 "Maximum mapcount (number of processes mapping a single page) encountered "
1479 "during the scan. This may be used as an indicator of the degree of sharing "
1480 "occurring in a given memory range."
1484 #: build/C/man7/numa.7:136
1486 msgid "I<swapcache=E<lt>countE<gt>>"
1490 #: build/C/man7/numa.7:139
1491 msgid "Number of pages that have an associated entry on a swap device."
1495 #: build/C/man7/numa.7:139
1497 msgid "I<active=E<lt>pagesE<gt>>"
1501 #: build/C/man7/numa.7:145
1503 "The number of pages on the active list. This field is shown only if "
1504 "different from the number of pages in this range. This means that some "
1505 "inactive pages exist in the memory range that may be removed from memory by "
1510 #: build/C/man7/numa.7:145
1512 msgid "I<writeback=E<lt>pagesE<gt>>"
1516 #: build/C/man7/numa.7:148
1517 msgid "Number of pages that are currently being written out to disk."
1521 #: build/C/man7/numa.7:150
1522 msgid "No standards govern NUMA interfaces."
1526 #: build/C/man7/numa.7:157
1528 "The Linux NUMA system calls and I</proc> interface are available only if the "
1529 "kernel was configured and built with the B<CONFIG_NUMA> option."
1533 #: build/C/man7/numa.7:157
1535 msgid "Library support"
1539 #: build/C/man7/numa.7:166
1541 "Link with I<-lnuma> to get the system call definitions. I<libnuma> and the "
1542 "required I<E<lt>numaif.hE<gt>> header are available in the I<numactl> "
1547 #: build/C/man7/numa.7:183
1549 "However, applications should not use these system calls directly. Instead, "
1550 "the higher level interface provided by the B<numa>(3) functions in the "
1551 "I<numactl> package is recommended. The I<numactl> package is available at "
1552 "E<.UR ftp://oss.sgi.com\\:/www\\:/projects\\:/libnuma\\:/download/> E<.UE .> "
1553 "The package is also included in some Linux distributions. Some "
1554 "distributions include the development library and header in the separate "
1555 "I<numactl-devel> package."
1559 #: build/C/man7/numa.7:191
1561 "B<get_mempolicy>(2), B<mbind>(2), B<move_pages>(2), B<set_mempolicy>(2), "
1562 "B<numa>(3), B<cpuset>(7), B<numactl>(8)"
1566 #: build/C/man2/set_mempolicy.2:27
1568 msgid "SET_MEMPOLICY"
1572 #: build/C/man2/set_mempolicy.2:27
1578 #: build/C/man2/set_mempolicy.2:30
1580 "set_mempolicy - set default NUMA memory policy for a process and its "
1585 #: build/C/man2/set_mempolicy.2:36
1588 "B<long set_mempolicy(int >I<mode>B<, const unsigned long *>I<nodemask>B<,>\n"
1589 "B< unsigned long >I<maxnode>B<);>\n"
1593 #: build/C/man2/set_mempolicy.2:49
1595 "B<set_mempolicy>() sets the NUMA memory policy of the calling process, "
1596 "which consists of a policy mode and zero or more nodes, to the values "
1597 "specified by the I<mode>, I<nodemask> and I<maxnode> arguments."
1601 #: build/C/man2/set_mempolicy.2:75
1603 "This system call defines the default policy for the process. The process "
1604 "policy governs allocation of pages in the process's address space outside of "
1605 "memory ranges controlled by a more specific policy set by B<mbind>(2). The "
1606 "process default policy also controls allocation of any pages for "
1607 "memory-mapped files mapped using the B<mmap>(2) call with the "
1608 "B<MAP_PRIVATE> flag and that are only read [loaded] from by the process and "
1609 "of memory-mapped files mapped using the B<mmap>(2) call with the "
1610 "B<MAP_SHARED> flag, regardless of the access type. The policy is applied "
1611 "only when a new page is allocated for the process. For anonymous memory "
1612 "this is when the page is first touched by the application."
1616 #: build/C/man2/set_mempolicy.2:89
1618 "The I<mode> argument must specify one of B<MPOL_DEFAULT>, B<MPOL_BIND>, "
1619 "B<MPOL_INTERLEAVE>, or B<MPOL_PREFERRED>. All modes except B<MPOL_DEFAULT> "
1620 "require the caller to specify via the I<nodemask> argument one or more "
1625 #: build/C/man2/set_mempolicy.2:97
1627 "The I<mode> argument may also include an optional I<mode flag>. The "
1628 "supported I<mode flags> are:"
1632 #: build/C/man2/set_mempolicy.2:97
1634 msgid "B<MPOL_F_STATIC_NODES> (since Linux 2.6.26)"
1638 #: build/C/man2/set_mempolicy.2:107
1640 "A nonempty I<nodemask> specifies physical node ids. Linux will not remap "
1641 "the I<nodemask> when the process moves to a different cpuset context, nor "
1642 "when the set of nodes allowed by the process's current cpuset context "
1647 #: build/C/man2/set_mempolicy.2:107
1649 msgid "B<MPOL_F_RELATIVE_NODES> (since Linux 2.6.26)"
1653 #: build/C/man2/set_mempolicy.2:133
1655 "I<nodemask> points to a bit mask of node IDs that contains up to I<maxnode> "
1656 "bits. The bit mask size is rounded to the next multiple of "
1657 "I<sizeof(unsigned long)>, but the kernel will use bits only up to "
1658 "I<maxnode>. A NULL value of I<nodemask> or a I<maxnode> value of zero "
1659 "specifies the empty set of nodes. If the value of I<maxnode> is zero, the "
1660 "I<nodemask> argument is ignored."
1664 #: build/C/man2/set_mempolicy.2:154
1666 "Where a I<nodemask> is required, it must contain at least one node that is "
1667 "on-line, allowed by the process's current cpuset context, [unless the "
1668 "B<MPOL_F_STATIC_NODES> mode flag is specified], and contains memory. If the "
1669 "B<MPOL_F_STATIC_NODES> is set in I<mode> and a required I<nodemask> contains "
1670 "no nodes that are allowed by the process's current cpuset context, the "
1671 "memory policy reverts to I<local allocation>. This effectively overrides "
1672 "the specified policy until the process's cpuset context includes one or more "
1673 "of the nodes specified by I<nodemask>."
1677 #: build/C/man2/set_mempolicy.2:165
1679 "The B<MPOL_DEFAULT> mode specifies that any nondefault process memory policy "
1680 "be removed, so that the memory policy \"falls back\" to the system default "
1681 "policy. The system default policy is \"local allocation\"\\(emthat is, "
1682 "allocate memory on the node of the CPU that triggered the allocation. "
1683 "I<nodemask> must be specified as NULL. If the \"local node\" contains no "
1684 "free memory, the system will attempt to allocate memory from a \"near by\" "
1689 #: build/C/man2/set_mempolicy.2:182
1691 "The B<MPOL_BIND> mode defines a strict policy that restricts memory "
1692 "allocation to the nodes specified in I<nodemask>. If I<nodemask> specifies "
1693 "more than one node, page allocations will come from the node with the lowest "
1694 "numeric node ID first, until that node contains no free memory. Allocations "
1695 "will then come from the node with the next highest node ID specified in "
1696 "I<nodemask> and so forth, until none of the specified nodes contain free "
1697 "memory. Pages will not be allocated from any node not specified in the "
1701 #. NOTE: the following sentence doesn't make sense in the context
1702 #. of set_mempolicy() -- no memory area specified.
1703 #. To be effective the memory area should be fairly large,
1704 #. at least 1MB or bigger.
1706 #: build/C/man2/set_mempolicy.2:196
1708 "B<MPOL_INTERLEAVE> interleaves page allocations across the nodes specified "
1709 "in I<nodemask> in numeric node ID order. This optimizes for bandwidth "
1710 "instead of latency by spreading out pages and memory accesses to those pages "
1711 "across multiple nodes. However, accesses to a single page will still be "
1712 "limited to the memory bandwidth of a single node."
1716 #: build/C/man2/set_mempolicy.2:213
1718 "B<MPOL_PREFERRED> sets the preferred node for allocation. The kernel will "
1719 "try to allocate pages from this node first and fall back to \"near by\" "
1720 "nodes if the preferred node is low on free memory. If I<nodemask> specifies "
1721 "more than one node ID, the first node in the mask will be selected as the "
1722 "preferred node. If the I<nodemask> and I<maxnode> arguments specify the "
1723 "empty set, then the policy specifies \"local allocation\" (like the system "
1724 "default policy discussed above)."
1728 #: build/C/man2/set_mempolicy.2:220
1730 "The process memory policy is preserved across an B<execve>(2), and is "
1731 "inherited by child processes created using B<fork>(2) or B<clone>(2)."
1735 #: build/C/man2/set_mempolicy.2:227
1737 "On success, B<set_mempolicy>() returns 0; on error, -1 is returned and "
1738 "I<errno> is set to indicate the error."
1742 #: build/C/man2/set_mempolicy.2:272
1744 "I<mode> is invalid. Or, I<mode> is B<MPOL_DEFAULT> and I<nodemask> is "
1745 "nonempty, or I<mode> is B<MPOL_BIND> or B<MPOL_INTERLEAVE> and I<nodemask> "
1746 "is empty. Or, I<maxnode> specifies more than a page worth of bits. Or, "
1747 "I<nodemask> specifies one or more node IDs that are greater than the maximum "
1748 "supported node ID. Or, none of the node IDs specified by I<nodemask> are "
1749 "on-line and allowed by the process's current cpuset context, or none of the "
1750 "specified nodes contain memory. Or, the I<mode> argument specified both "
1751 "B<MPOL_F_STATIC_NODES> and B<MPOL_F_RELATIVE_NODES>."
1755 #: build/C/man2/set_mempolicy.2:279
1757 "The B<set_mempolicy>(), system call was added to the Linux kernel in version "
1762 #: build/C/man2/set_mempolicy.2:286
1764 "Process policy is not remembered if the page is swapped out. When such a "
1765 "page is paged back in, it will use the policy of the process or memory range "
1766 "that is in effect at the time the page is allocated."
1770 #: build/C/man2/set_mempolicy.2:298
1772 "B<get_mempolicy>(2), B<getcpu>(2), B<mbind>(2), B<mmap>(2), B<numa>(3), "
1773 "B<cpuset>(7), B<numa>(7), B<numactl>(8)"