admin-guide/mm/numa_memory_policy.rst

12 supported platforms with Non-Uniform Memory Access architectures since 2.4.?.
18 (``Documentation/admin-guide/cgroup-v1/cpusets.rst``)
21 programming interface that a NUMA-aware application can take advantage of.  When
30 ------------------------
43 	not to overload the initial boot node with boot-time
47 	this is an optional, per-task policy.  When defined for a
56 	[clone() w/o the CLONE_VM flag] and exec*().  This allows a parent task
63 	In a multi-threaded task, task policies apply only to the thread
95 	  mmap()ed with the MAP_ANONYMOUS flag.  If a VMA policy is
97 	  used the MAP_SHARED flag.  If the file mapping used the
98 	  MAP_PRIVATE flag, the VMA policy will only be applied when
100 	  mapping-- i.e., at Copy-On-Write.
103 	  virtual address space--a.k.a. threads--independent of when
108 	  are NOT inheritable across exec().  Thus, only NUMA-aware
111 	* A task may install a new VMA policy on a sub-range of a
128 	policies--using the mbind() system call specifying a range of
140 	support allocation at fault time--a.k.a lazy allocation--so hugetlbfs
163 -----------------------------
165 A NUMA memory policy consists of a "mode", optional mode flags, and
166 an optional set of nodes.  The mode determines the behavior of the
167 policy, the optional mode flags determine the behavior of the mode,
177 Default Mode--MPOL_DEFAULT
178 	This mode is only used in the memory policy APIs.  Internally,
180 	policy scopes.  Any existing non-default policy will simply be
189 	When specified in one of the memory policy APIs, the Default mode
193 	be non-empty.
196 	This mode specifies that memory must come from the set of
202 	This mode specifies that the allocation should be attempted
208 	Internally, the Preferred policy uses a single node--the
210 	mode flag MPOL_F_LOCAL is set, the preferred_node is ignored
218 	mode.  If an empty nodemask is passed, the policy cannot use
223 	This mode specifies that page allocations be interleaved, on a
225 	This mode also behaves slightly differently, based on the
229 	Interleave mode indexes the set of nodes specified by the
238 	For allocation of page cache pages, Interleave mode indexes
246 	interleaved system default policy works in this mode.
248 NUMA memory policy supports the following optional mode flags:
251 	This flag specifies that the nodemask passed by
255 	Without this flag, any time a mempolicy is rebound because of a
261 	With this flag, if the user-specified nodes overlap with the
267 	mems 1-3 that sets an Interleave policy over the same set.  If
268 	the cpuset's mems change to 3-5, the Interleave will now occur
269 	over nodes 3, 4, and 5.  With this flag, however, since only node
275 	MPOL_F_RELATIVE_NODES flag.  It also cannot be used for
280 	This flag specifies that the nodemask passed
282 	set of allowed nodes.  The kernel stores the user-passed nodemask,
286 	Without this flag (and without MPOL_F_STATIC_NODES), anytime a
292 	1,3,5 may be remapped to 7-9 and then to 1-3 if the set of
295 	With this flag, the remap is done so that the node numbers from
305 	the user's nodemask when the set of allowed nodes is only 0-3),
310 	mems 2-5 that sets an Interleave policy over the same set with
311 	MPOL_F_RELATIVE_NODES.  If the cpuset's mems change to 3-7, the
312 	interleave now occurs over nodes 3,5-7.  If the cpuset's mems
313 	then change to 0,2-3,5, then the interleave occurs over nodes
314 	0,2-3,5.
317 	nodemasks to specify memory policies using this flag should
320 	memory nodes 0 to N-1, where N is the number of memory nodes the
326 	MPOL_F_STATIC_NODES flag.  It also cannot be used for
345 During run-time "usage" of the policy, we attempt to minimize atomic operations
353 2) examination of the policy to determine the policy mode and associated node
386    used for non-shared policies.  For this reason, shared policies are marked
387    as such, and the extra reference is dropped "conditionally"--i.e., only
412    prefix, are defined in <linux/syscalls.h>; the mode and flag
417 	long set_mempolicy(int mode, const unsigned long *nmask,
420 Set's the calling task's "task/process memory policy" to mode
421 specified by the 'mode' argument and the set of nodes defined by
423 'maxnode' ids.  Optional mode flags may be passed by combining the
424 'mode' argument with the flag (for example: MPOL_INTERLEAVE |
432 	long get_mempolicy(int *mode,
445 	long mbind(void *start, unsigned long len, int mode,
449 mbind() installs the policy specified by (mode, nmask, maxnodes) as a
467 The numactl(8) tool is packaged with the run-time version of the library
469 package the headers and compile-time libraries in a separate development