move_pages — move individual pages of a process to another node


#include <numaif.h>
long move_pages( int pid,
  unsigned long count,
  void **pages,
  const int *nodes,
  int *status,
  int flags);
[Note] Note
There is no glibc wrapper for this system call; see NOTES.
[Note] Note

Link with −lnuma.


move_pages() moves the specified pages of the process pid to the memory nodes specified by nodes. The result of the move is reflected in status. The flags indicate constraints on the pages to be moved.

pid is the ID of the process in which pages are to be moved. If pid is 0, then move_pages() moves pages of the calling process.

To move pages in another process requires the following privileges:

  • In kernels up to and including Linux 4.12: the caller must be privileged (CAP_SYS_NICE) or the real or effective user ID of the calling process must match the real or saved-set user ID of the target process.

  • The older rules allowed the caller to discover various virtual address choices made by the kernel that could lead to the defeat of address-space-layout randomization for a process owned by the same UID as the caller, the rules were changed starting with Linux 4.13. Since Linux 4.13, permission is governed by a ptrace access mode PTRACE_MODE_READ_REALCREDS check with respect to the target process; see ptrace(2).

count is the number of pages to move. It defines the size of the three arrays pages, nodes, and status.

pages is an array of pointers to the pages that should be moved. These are pointers that should be aligned to page boundaries. Addresses are specified as seen by the process specified by pid.

nodes is an array of integers that specify the desired location for each page. Each element in the array is a node number. nodes can also be NULL, in which case move_pages() does not move any pages but instead will return the node where each page currently resides, in the status array. Obtaining the status of each page may be necessary to determine pages that need to be moved.

status is an array of integers that return the status of each page. The array contains valid values only if move_pages() did not return an error. Preinitialization of the array to a value which cannot represent a real numa node or valid error of status array could help to identify pages that have been migrated.

flags specify what types of pages to move. MPOL_MF_MOVE means that only pages that are in exclusive use by the process are to be moved. MPOL_MF_MOVE_ALL means that pages shared between multiple processes can also be moved. The process must be privileged (CAP_SYS_NICE) to use MPOL_MF_MOVE_ALL.

Page states in the status array

The following values can be returned in each element of the status array.


Identifies the node on which the page resides.


The page is mapped by multiple processes and can be moved only if MPOL_MF_MOVE_ALL is specified.


The page is currently busy and cannot be moved. Try again later. This occurs if a page is undergoing I/O or another kernel subsystem is holding a reference to the page.


This is a zero page or the memory area is not mapped by the process.


Unable to write back a page. The page has to be written back in order to move it since the page is dirty and the filesystem does not provide a migration function that would allow the move of dirty pages.


A dirty page cannot be moved. The filesystem does not provide a migration function and has no ability to write back pages.


The page is not present.


Unable to allocate memory on target node.


On success move_pages() returns zero. On error, it returns −1, and sets errno to indicate the error. If positive value is returned, it is the number of nonmigrated pages.


Positive value

The number of nonmigrated pages if they were the result of nonfatal reasons (since Linux 4.17).


Too many pages to move. Since Linux 2.6.29, the kernel no longer generates this error.


One of the target nodes is not allowed by the current cpuset.


Parameter array could not be accessed.


Flags other than MPOL_MF_MOVE and MPOL_MF_MOVE_ALL was specified or an attempt was made to migrate pages of a kernel thread.


One of the target nodes is not online.


The caller specified MPOL_MF_MOVE_ALL without sufficient privileges (CAP_SYS_NICE). Or, the caller attempted to move pages of a process belonging to another user but did not have privilege to do so (CAP_SYS_NICE).


Process does not exist.


move_pages() first appeared on Linux in version 2.6.18.


This system call is Linux-specific.


Glibc does not provide a wrapper for this system call. For information on library support, see numa(7).

Use get_mempolicy(2) with the MPOL_F_MEMS_ALLOWED flag to obtain the set of nodes that are allowed by the current cpuset. Note that this information is subject to change at any time by manual or automatic reconfiguration of the cpuset.

Use of this function may result in pages whose location (node) violates the memory policy established for the specified addresses (See mbind(2)) and/or the specified process (See set_mempolicy(2)). That is, memory policy does not constrain the destination nodes used by move_pages().

The <numaif.h> header is not included with glibc, but requires installing libnuma−devel or a similar package.


get_mempolicy(2), mbind(2), set_mempolicy(2), numa(3), numa_maps(5), cpuset(7), numa(7), migratepages(8), numastat(8)


This page is part of release 5.13 of the Linux man-pages project. A description of the project, information about reporting bugs, and the latest version of this page, can be found at−pages/.

  This manpage is Copyright (C) 2006 Silicon Graphics, Inc.
                              Christoph Lameter

Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.

FIXME Should programs normally be using move_pages() directly, or should
they rather be using interfaces in the numactl package?
(e.g., compare with recommendation in mbind(2)).
Does this page need to give advice on this topic?