1MBIND(2) Linux Programmer's Manual MBIND(2)
2
6 mbind - set memory policy for a memory range
7
9 #include <numaif.h>
10 long mbind(void *addr, unsigned long len, int mode,
12 const unsigned long *nodemask, unsigned long maxnode,
13 unsigned flags);
14 Link with -lnuma.
16
18 mbind() sets the NUMA memory policy, which consists of a policy mode
19 and zero or more nodes, for the memory range starting with addr and
20 continuing for len bytes. The memory policy defines from which node
21 memory is allocated.
22 If the memory range specified by the addr and len arguments includes an
24 "anonymous" region of memory—that is a region of memory created using
25 the mmap(2) system call with the MAP_ANONYMOUS—or a memory-mapped file,
26 mapped using the mmap(2) system call with the MAP_PRIVATE flag, pages
27 will be allocated only according to the specified policy when the
28 application writes (stores) to the page. For anonymous regions, an
29 initial read access will use a shared page in the kernel containing all
30 zeros. For a file mapped with MAP_PRIVATE, an initial read access will
31 allocate pages according to the memory policy of the thread that causes
32 the page to be allocated. This may not be the thread that called
33 mbind().
34 The specified policy will be ignored for any MAP_SHARED mappings in the
36 specified memory range. Rather the pages will be allocated according
37 to the memory policy of the thread that caused the page to be allo‐
38 cated. Again, this may not be the thread that called mbind().
39 If the specified memory range includes a shared memory region created
41 using the shmget(2) system call and attached using the shmat(2) system
42 call, pages allocated for the anonymous or shared memory region will be
43 allocated according to the policy specified, regardless of which
44 process attached to the shared memory segment causes the allocation.
45 If, however, the shared memory region was created with the SHM_HUGETLB
46 flag, the huge pages will be allocated according to the policy speci‐
47 fied only if the page allocation is caused by the process that calls
48 mbind() for that region.
49 By default, mbind() has an effect only for new allocations; if the
51 pages inside the range have been already touched before setting the
52 policy, then the policy has no effect. This default behavior may be
53 overridden by the MPOL_MF_MOVE and MPOL_MF_MOVE_ALL flags described
54 below.
55 The mode argument must specify one of MPOL_DEFAULT, MPOL_BIND,
57 MPOL_INTERLEAVE, MPOL_PREFERRED, or MPOL_LOCAL (which are described in
58 detail below). All policy modes except MPOL_DEFAULT require the caller
59 to specify the node or nodes to which the mode applies, via the node‐
60 mask argument.
61 The mode argument may also include an optional mode flag. The sup‐
63 ported mode flags are:
64 MPOL_F_STATIC_NODES (since Linux-2.6.26)
66 A nonempty nodemask specifies physical node IDs. Linux does not
67 remap the nodemask when the thread moves to a different cpuset
68 context, nor when the set of nodes allowed by the thread's cur‐
69 rent cpuset context changes.
70 MPOL_F_RELATIVE_NODES (since Linux-2.6.26)
72 A nonempty nodemask specifies node IDs that are relative to the
73 set of node IDs allowed by the thread's current cpuset.
74 nodemask points to a bit mask of nodes containing up to maxnode bits.
76 The bit mask size is rounded to the next multiple of sizeof(unsigned
77 long), but the kernel will use bits only up to maxnode. A NULL value
78 of nodemask or a maxnode value of zero specifies the empty set of
79 nodes. If the value of maxnode is zero, the nodemask argument is
80 ignored. Where a nodemask is required, it must contain at least one
81 node that is on-line, allowed by the thread's current cpuset context
82 (unless the MPOL_F_STATIC_NODES mode flag is specified), and contains
83 memory.
84 The mode argument must include one of the following values:
86 MPOL_DEFAULT
88 This mode requests that any nondefault policy be removed,
89 restoring default behavior. When applied to a range of memory
90 via mbind(), this means to use the thread memory policy, which
91 may have been set with set_mempolicy(2). If the mode of the
92 thread memory policy is also MPOL_DEFAULT, the system-wide
93 default policy will be used. The system-wide default policy
94 allocates pages on the node of the CPU that triggers the alloca‐
95 tion. For MPOL_DEFAULT, the nodemask and maxnode arguments must
96 be specify the empty set of nodes.
97 MPOL_BIND
99 This mode specifies a strict policy that restricts memory allo‐
100 cation to the nodes specified in nodemask. If nodemask speci‐
101 fies more than one node, page allocations will come from the
102 node with sufficient free memory that is closest to the node
103 where the allocation takes place. Pages will not be allocated
104 from any node not specified in the IR nodemask . (Before Linux
105 2.6.26, page allocations came from the node with the lowest
106 numeric node ID first, until that node contained no free memory.
107 Allocations then came from the node with the next highest node
108 ID specified in nodemask and so forth, until none of the speci‐
109 fied nodes contained free memory.)
110 MPOL_INTERLEAVE
112 This mode specifies that page allocations be interleaved across
113 the set of nodes specified in nodemask. This optimizes for
114 bandwidth instead of latency by spreading out pages and memory
115 accesses to those pages across multiple nodes. To be effective
116 the memory area should be fairly large, at least 1 MB or bigger
117 with a fairly uniform access pattern. Accesses to a single page
118 of the area will still be limited to the memory bandwidth of a
119 single node.
120 MPOL_PREFERRED
122 This mode sets the preferred node for allocation. The kernel
123 will try to allocate pages from this node first and fall back to
124 other nodes if the preferred nodes is low on free memory. If
125 nodemask specifies more than one node ID, the first node in the
126 mask will be selected as the preferred node. If the nodemask
127 and maxnode arguments specify the empty set, then the memory is
128 allocated on the node of the CPU that triggered the allocation.
129 MPOL_LOCAL (since Linux 3.8)
131 This mode specifies "local allocation"; the memory is allocated
132 on the node of the CPU that triggered the allocation (the "local
133 node"). The nodemask and maxnode arguments must specify the
134 empty set. If the "local node" is low on free memory, the ker‐
135 nel will try to allocate memory from other nodes. The kernel
136 will allocate memory from the "local node" whenever memory for
137 this node is available. If the "local node" is not allowed by
138 the thread's current cpuset context, the kernel will try to
139 allocate memory from other nodes. The kernel will allocate mem‐
140 ory from the "local node" whenever it becomes allowed by the
141 thread's current cpuset context. By contrast, MPOL_DEFAULT
142 reverts to the memory policy of the thread (which may be set via
143 set_mempolicy(2)); that policy may be something other than
144 "local allocation".
145 If MPOL_MF_STRICT is passed in flags and mode is not MPOL_DEFAULT, then
147 the call fails with the error EIO if the existing pages in the memory
148 range don't follow the policy.
149 If MPOL_MF_MOVE is specified in flags, then the kernel will attempt to
151 move all the existing pages in the memory range so that they follow the
152 policy. Pages that are shared with other processes will not be moved.
153 If MPOL_MF_STRICT is also specified, then the call fails with the error
154 EIO if some pages could not be moved.
155 If MPOL_MF_MOVE_ALL is passed in flags, then the kernel will attempt to
157 move all existing pages in the memory range regardless of whether other
158 processes use the pages. The calling thread must be privileged
159 (CAP_SYS_NICE) to use this flag. If MPOL_MF_STRICT is also specified,
160 then the call fails with the error EIO if some pages could not be
161 moved.
162
164 On success, mbind() returns 0; on error, -1 is returned and errno is
165 set to indicate the error.
166
168 EFAULT Part or all of the memory range specified by nodemask and maxn‐
169 ode points outside your accessible address space. Or, there was
170 an unmapped hole in the specified memory range specified by addr
171 and len.
172 EINVAL An invalid value was specified for flags or mode; or addr + len
174 was less than addr; or addr is not a multiple of the system page
175 size. Or, mode is MPOL_DEFAULT and nodemask specified a
176 nonempty set; or mode is MPOL_BIND or MPOL_INTERLEAVE and node‐
177 mask is empty. Or, maxnode exceeds a kernel-imposed limit. Or,
178 nodemask specifies one or more node IDs that are greater than
179 the maximum supported node ID. Or, none of the node IDs speci‐
180 fied by nodemask are on-line and allowed by the thread's current
181 cpuset context, or none of the specified nodes contain memory.
182 Or, the mode argument specified both MPOL_F_STATIC_NODES and
183 MPOL_F_RELATIVE_NODES.
184 EIO MPOL_MF_STRICT was specified and an existing page was already on
186 a node that does not follow the policy; or MPOL_MF_MOVE or
187 MPOL_MF_MOVE_ALL was specified and the kernel was unable to move
188 all existing pages in the range.
189 ENOMEM Insufficient kernel memory was available.
191 EPERM The flags argument included the MPOL_MF_MOVE_ALL flag and the
193 caller does not have the CAP_SYS_NICE privilege.
194
196 The mbind() system call was added to the Linux kernel in version 2.6.7.
197
199 This system call is Linux-specific.
200
202 For information on library support, see numa(7).
203 NUMA policy is not supported on a memory-mapped file range that was
205 mapped with the MAP_SHARED flag.
206 The MPOL_DEFAULT mode can have different effects for mbind() and
208 set_mempolicy(2). When MPOL_DEFAULT is specified for set_mempolicy(2),
209 the thread's memory policy reverts to the system default policy or
210 local allocation. When MPOL_DEFAULT is specified for a range of memory
211 using mbind(), any pages subsequently allocated for that range will use
212 the thread's memory policy, as set by set_mempolicy(2). This effec‐
213 tively removes the explicit policy from the specified range, "falling
214 back" to a possibly nondefault policy. To select explicit "local allo‐
215 cation" for a memory range, specify a mode of MPOL_LOCAL or MPOL_PRE‐
216 FERRED with an empty set of nodes. This method will work for set_mem‐
217 policy(2), as well.
218 Support for huge page policy was added with 2.6.16. For interleave
220 policy to be effective on huge page mappings the policied memory needs
221 to be tens of megabytes or larger.
222 MPOL_MF_STRICT is ignored on huge page mappings.
224 MPOL_MF_MOVE and MPOL_MF_MOVE_ALL are available only on Linux 2.6.16
226 and later.
227
229 get_mempolicy(2), getcpu(2), mmap(2), set_mempolicy(2), shmat(2),
230 shmget(2), numa(3), cpuset(7), numa(7), numactl(8)
231
233 This page is part of release 4.16 of the Linux man-pages project. A
234 description of the project, information about reporting bugs, and the
235 latest version of this page, can be found at
236 https://www.kernel.org/doc/man-pages/.
237Linux 2017-09-15 MBIND(2)