man2/madvise.2

   1 .\" Hey Emacs! This file is -*- nroff -*- source.
   2 .\"
   3 .\" Copyright (C) 2001 David Gómez <davidge@jazzfree.com>
   4 .\"
   5 .\" Permission is granted to make and distribute verbatim copies of this
   6 .\" manual provided the copyright notice and this permission notice are
   7 .\" preserved on all copies.
   8 .\"
   9 .\" Permission is granted to copy and distribute modified versions of this
  10 .\" manual under the conditions for verbatim copying, provided that the
  11 .\" entire resulting derived work is distributed under the terms of a
  12 .\" permission notice identical to this one.
  13 .\"
  14 .\" Since the Linux kernel and libraries are constantly changing, this
  15 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  16 .\" responsibility for errors or omissions, or for damages resulting from
  17 .\" the use of the information contained herein.  The author(s) may not
  18 .\" have taken the same level of care in the production of this manual,
  19 .\" which is licensed free of charge, as they might when working
  20 .\" professionally.
  21 .\"
  22 .\" Formatted or processed versions of this manual, if unaccompanied by
  23 .\" the source, must acknowledge the copyright and authors of this work.
  24 .\"
  25 .\" Based on comments from mm/filemap.c. Last modified on 10-06-2001
  26 .\" Modified, 25 Feb 2002, Michael Kerrisk, <mtk.manpages@gmail.com>
  27 .\"     Added notes on MADV_DONTNEED
  28 .\"
  29 .TH MADVISE 2 2007-07-26 "Linux" "Linux Programmer's Manual"
  30 .SH NAME
  31 madvise \- give advice about use of memory
  32 .SH SYNOPSIS
  33 .br
  34 .B #include <sys/mman.h>
  35 .sp
  36 .BI "int madvise(void *" start ", size_t " length ", int " advice );
  37 .sp
  38 .in -4n
  39 Feature Test Macro Requirements for glibc (see
  40 .BR feature_test_macros (7)):
  41 .in
  42 .sp
  43 .BR madvise ():
  44 _BSD_SOURCE
  45 .SH DESCRIPTION
  46 The
  47 .BR madvise ()
  48 system call advises the kernel about how to handle paging input/output in
  49 the address range beginning at address
  50 .I start
  51 and with size
  52 .I length
  53 bytes.
  54 It allows an application to tell the kernel how it expects to use
  55 some mapped or shared memory areas, so that the kernel can choose
  56 appropriate read-ahead and caching techniques.
  57 This call does not influence the semantics of the application
  58 (except in the case of
  59 .BR MADV_DONTNEED ),
  60 but
  61 may influence its performance.
  62 The kernel is free to ignore the advice.
  63 .LP
  64 The advice is indicated in the
  65 .I advice
  66 parameter which can be
  67 .TP
  68 .B MADV_NORMAL
  69 No special treatment.
  70 This is the default.
  71 .TP
  72 .B MADV_RANDOM
  73 Expect page references in random order.
  74 (Hence, read ahead may be less useful than normally.)
  75 .TP
  76 .B MADV_SEQUENTIAL
  77 Expect page references in sequential order.
  78 (Hence, pages in the given range can be aggressively read ahead,
  79 and may be freed soon after they are accessed.)
  80 .TP
  81 .B MADV_WILLNEED
  82 Expect access in the near future.
  83 (Hence, it might be a good idea to read some pages ahead.)
  84 .TP
  85 .B MADV_DONTNEED
  86 Do not expect access in the near future.
  87 (For the time being, the application is finished with the given range,
  88 so the kernel can free resources associated with it.)
  89 Subsequent accesses of pages in this range will succeed, but will result
  90 either in re-loading of the memory contents from the underlying mapped file
  91 (see
  92 .BR mmap (2))
  93 or zero-fill-on-demand pages for mappings
  94 without an underlying file.
  95 .TP
  96 .BR MADV_REMOVE " (Since Linux 2.6.16)"
  97 Free up a given range of pages
  98 and its associated backing store.
  99 Currently,
 100 .\" 2.6.18-rc5
 101 only shmfs/tmpfs supports this; other filesystems return with the
 102 error
 103 .BR ENOSYS .
 104 .\" Databases want to use this feature to drop a section of their
 105 .\" bufferpool (shared memory segments) - without writing back to
 106 .\" disk/swap space.  This feature is also useful for supporting
 107 .\" hot-plug memory on UML.
 108 .TP
 109 .BR MADV_DONTFORK " (Since Linux 2.6.16)"
 110 .\" See http://lwn.net/Articles/171941/
 111 Do not make the pages in this range available to the child after a
 112 .BR fork (2).
 113 This is useful to prevent copy-on-write semantics from changing
 114 the physical location of a page(s) if the parent writes to it after a
 115 .BR fork (2).
 116 (Such page relocations cause problems for hardware that
 117 DMAs into the page(s).)
 118 .\" [PATCH] madvise MADV_DONTFORK/MADV_DOFORK
 119 .\" Currently, copy-on-write may change the physical address of
 120 .\" a page even if the user requested that the page is pinned in
 121 .\" memory (either by mlock or by get_user_pages).  This happens
 122 .\" if the process forks meanwhile, and the parent writes to that
 123 .\" page.  As a result, the page is orphaned: in case of
 124 .\" get_user_pages, the application will never see any data hardware
 125 .\" DMA's into this page after the COW.  In case of mlock'd memory,
 126 .\" the parent is not getting the realtime/security benefits of mlock.
 127 .\"
 128 .\" In particular, this affects the Infiniband modules which do DMA from
 129 .\" and into user pages all the time.
 130 .\"
 131 .\" This patch adds madvise options to control whether memory range is
 132 .\" inherited across fork. Useful e.g. for when hardware is doing DMA
 133 .\" from/into these pages.  Could also be useful to an application
 134 .\" wanting to speed up its forks by cutting large areas out of
 135 .\" consideration.
 136 .TP
 137 .BR MADV_DOFORK " (Since Linux 2.6.16)"
 138 Undo the effect of
 139 .BR MADV_DONTFORK ,
 140 restoring the default behavior, whereby a mapping is inherited across
 141 .BR fork (2).
 142 .SH "RETURN VALUE"
 143 On success
 144 .BR madvise ()
 145 returns zero.
 146 On error, it returns \-1 and
 147 .I errno
 148 is set appropriately.
 149 .SH ERRORS
 150 .TP
 151 .B EAGAIN
 152 A kernel resource was temporarily unavailable.
 153 .TP
 154 .B EBADF
 155 The map exists, but the area maps something that isn't a file.
 156 .TP
 157 .B EINVAL
 158 The value
 159 .I len
 160 is negative,
 161 .\" .I len
 162 .\" is zero,
 163 .I start
 164 is not page-aligned,
 165 .I advice
 166 is not a valid value, or the application is attempting
 167 to release locked or shared pages (with
 168 .BR MADV_DONTNEED ).
 169 .TP
 170 .B EIO
 171 (for
 172 .BR MADV_WILLNEED )
 173 Paging in this area would exceed the process's
 174 maximum resident set size.
 175 .TP
 176 .B ENOMEM
 177 (for
 178 .BR MADV_WILLNEED )
 179 Not enough memory: paging in failed.
 180 .TP
 181 .B ENOMEM
 182 Addresses in the specified range are not currently
 183 mapped, or are outside the address space of the process.
 184 .SH "CONFORMING TO"
 185 POSIX.1b.
 186 POSIX.1-2001 describes
 187 .BR posix_madvise (3)
 188 with constants
 189 .BR POSIX_MADV_NORMAL ,
 190 etc.,
 191 with a behavior close to that described here.
 192 There is a similar
 193 .BR posix_fadvise (2)
 194 for file access.
 195
 196 .BR MADV_REMOVE ,
 197 .BR MADV_DONTFORK ,
 198 and
 199 .B MADV_DOFORK
 200 are Linux-specific.
 201 .SH NOTES
 202 .SS "Linux Notes"
 203 .LP
 204 The current Linux implementation (2.4.0) views this system call
 205 more as a command than as advice and hence may return an error
 206 when it cannot do what it usually would do in response to this
 207 advice.
 208 (See the ERRORS description above.)
 209 This is nonstandard behavior.
 210 .LP
 211 The Linux implementation requires that the address
 212 .I start
 213 be page-aligned, and allows
 214 .I length
 215 to be zero.
 216 If there are some parts of the specified address range
 217 that are not mapped, the Linux version of
 218 .BR madvise ()
 219 ignores them and applies the call to the rest (but returns
 220 .B ENOMEM
 221 from the system call, as it should).
 222 .\" .SH HISTORY
 223 .\" The
 224 .\" .BR madvise ()
 225 .\" function first appeared in 4.4BSD.
 226 .SH "SEE ALSO"
 227 .BR getrlimit (2),
 228 .BR mincore (2),
 229 .BR mmap (2),
 230 .BR mprotect (2),
 231 .BR msync (2),
 232 .BR munmap (2)