2 The intent of this file is to give a brief summary of hugetlbpage support in
3 the Linux kernel. This support is built on top of multiple page size support
4 that is provided by most modern architectures. For example, x86 CPUs normally
5 support 4K and 2M (1G if architecturally supported) page sizes, ia64
6 architecture supports multiple page sizes 4K, 8K, 64K, 256K, 1M, 4M, 16M,
7 256M and ppc64 supports 4K and 16M. A TLB is a cache of virtual-to-physical
8 translations. Typically this is a very scarce resource on processor.
9 Operating systems try to make best use of limited number of TLB resources.
10 This optimization is more critical now as bigger and bigger physical memories
11 (several GBs) are more readily available.
13 Users can use the huge page support in Linux kernel by either using the mmap
14 system call or standard SYSV shared memory system calls (shmget, shmat).
16 First the Linux kernel needs to be built with the CONFIG_HUGETLBFS
17 (present under "File systems") and CONFIG_HUGETLB_PAGE (selected
18 automatically when CONFIG_HUGETLBFS is selected) configuration
21 The /proc/meminfo file provides information about the total number of
22 persistent hugetlb pages in the kernel's huge page pool. It also displays
23 information about the number of free, reserved and surplus huge pages and the
24 default huge page size. The huge page size is needed for generating the
25 proper alignment and size of the arguments to system calls that map huge page
28 The output of "cat /proc/meminfo" will include lines like:
38 HugePages_Total is the size of the pool of huge pages.
39 HugePages_Free is the number of huge pages in the pool that are not yet
41 HugePages_Rsvd is short for "reserved," and is the number of huge pages for
42 which a commitment to allocate from the pool has been made,
43 but no allocation has yet been made. Reserved huge pages
44 guarantee that an application will be able to allocate a
45 huge page from the pool of huge pages at fault time.
46 HugePages_Surp is short for "surplus," and is the number of huge pages in
47 the pool above the value in /proc/sys/vm/nr_hugepages. The
48 maximum number of surplus huge pages is controlled by
49 /proc/sys/vm/nr_overcommit_hugepages.
51 /proc/filesystems should also show a filesystem of type "hugetlbfs" configured
54 /proc/sys/vm/nr_hugepages indicates the current number of "persistent" huge
55 pages in the kernel's huge page pool. "Persistent" huge pages will be
56 returned to the huge page pool when freed by a task. A user with root
57 privileges can dynamically allocate more or free some persistent huge pages
58 by increasing or decreasing the value of 'nr_hugepages'.
60 Pages that are used as huge pages are reserved inside the kernel and cannot
61 be used for other purposes. Huge pages cannot be swapped out under
64 Once a number of huge pages have been pre-allocated to the kernel huge page
65 pool, a user with appropriate privilege can use either the mmap system call
66 or shared memory system calls to use the huge pages. See the discussion of
67 Using Huge Pages, below.
69 The administrator can allocate persistent huge pages on the kernel boot
70 command line by specifying the "hugepages=N" parameter, where 'N' = the
71 number of huge pages requested. This is the most reliable method of
72 allocating huge pages as memory has not yet become fragmented.
74 Some platforms support multiple huge page sizes. To allocate huge pages
75 of a specific size, one must precede the huge pages boot command parameters
76 with a huge page size selection parameter "hugepagesz=<size>". <size> must
77 be specified in bytes with optional scale suffix [kKmMgG]. The default huge
78 page size may be selected with the "default_hugepagesz=<size>" boot parameter.
80 When multiple huge page sizes are supported, /proc/sys/vm/nr_hugepages
81 indicates the current number of pre-allocated huge pages of the default size.
82 Thus, one can use the following command to dynamically allocate/deallocate
83 default sized persistent huge pages:
85 echo 20 > /proc/sys/vm/nr_hugepages
87 This command will try to adjust the number of default sized huge pages in the
88 huge page pool to 20, allocating or freeing huge pages, as required.
90 On a NUMA platform, the kernel will attempt to distribute the huge page pool
91 over all the set of allowed nodes specified by the NUMA memory policy of the
92 task that modifies nr_hugepages. The default for the allowed nodes--when the
93 task has default memory policy--is all on-line nodes with memory. Allowed
94 nodes with insufficient available, contiguous memory for a huge page will be
95 silently skipped when allocating persistent huge pages. See the discussion
96 below of the interaction of task memory policy, cpusets and per node attributes
97 with the allocation and freeing of persistent huge pages.
99 The success or failure of huge page allocation depends on the amount of
100 physically contiguous memory that is present in system at the time of the
101 allocation attempt. If the kernel is unable to allocate huge pages from
102 some nodes in a NUMA system, it will attempt to make up the difference by
103 allocating extra pages on other nodes with sufficient available contiguous
106 System administrators may want to put this command in one of the local rc
107 init files. This will enable the kernel to allocate huge pages early in
108 the boot process when the possibility of getting physical contiguous pages
109 is still very high. Administrators can verify the number of huge pages
110 actually allocated by checking the sysctl or meminfo. To check the per node
111 distribution of huge pages in a NUMA system, use:
113 cat /sys/devices/system/node/node*/meminfo | fgrep Huge
115 /proc/sys/vm/nr_overcommit_hugepages specifies how large the pool of
116 huge pages can grow, if more huge pages than /proc/sys/vm/nr_hugepages are
117 requested by applications. Writing any non-zero value into this file
118 indicates that the hugetlb subsystem is allowed to try to obtain that
119 number of "surplus" huge pages from the kernel's normal page pool, when the
120 persistent huge page pool is exhausted. As these surplus huge pages become
121 unused, they are freed back to the kernel's normal page pool.
123 When increasing the huge page pool size via nr_hugepages, any existing surplus
124 pages will first be promoted to persistent huge pages. Then, additional
125 huge pages will be allocated, if necessary and if possible, to fulfill
126 the new persistent huge page pool size.
128 The administrator may shrink the pool of persistent huge pages for
129 the default huge page size by setting the nr_hugepages sysctl to a
130 smaller value. The kernel will attempt to balance the freeing of huge pages
131 across all nodes in the memory policy of the task modifying nr_hugepages.
132 Any free huge pages on the selected nodes will be freed back to the kernel's
135 Caveat: Shrinking the persistent huge page pool via nr_hugepages such that
136 it becomes less than the number of huge pages in use will convert the balance
137 of the in-use huge pages to surplus huge pages. This will occur even if
138 the number of surplus pages it would exceed the overcommit value. As long as
139 this condition holds--that is, until nr_hugepages+nr_overcommit_hugepages is
140 increased sufficiently, or the surplus huge pages go out of use and are freed--
141 no more surplus huge pages will be allowed to be allocated.
143 With support for multiple huge page pools at run-time available, much of
144 the huge page userspace interface in /proc/sys/vm has been duplicated in sysfs.
145 The /proc interfaces discussed above have been retained for backwards
146 compatibility. The root huge page control directory in sysfs is:
148 /sys/kernel/mm/hugepages
150 For each huge page size supported by the running kernel, a subdirectory
151 will exist, of the form:
155 Inside each of these directories, the same set of files will exist:
158 nr_hugepages_mempolicy
159 nr_overcommit_hugepages
164 which function as described above for the default huge page-sized case.
167 Interaction of Task Memory Policy with Huge Page Allocation/Freeing
168 ===================================================================
170 Whether huge pages are allocated and freed via the /proc interface or
171 the /sysfs interface using the nr_hugepages_mempolicy attribute, the NUMA
172 nodes from which huge pages are allocated or freed are controlled by the
173 NUMA memory policy of the task that modifies the nr_hugepages_mempolicy
174 sysctl or attribute. When the nr_hugepages attribute is used, mempolicy
177 The recommended method to allocate or free huge pages to/from the kernel
178 huge page pool, using the nr_hugepages example above, is:
180 numactl --interleave <node-list> echo 20 \
181 >/proc/sys/vm/nr_hugepages_mempolicy
185 numactl -m <node-list> echo 20 >/proc/sys/vm/nr_hugepages_mempolicy
187 This will allocate or free abs(20 - nr_hugepages) to or from the nodes
188 specified in <node-list>, depending on whether number of persistent huge pages
189 is initially less than or greater than 20, respectively. No huge pages will be
190 allocated nor freed on any node not included in the specified <node-list>.
192 When adjusting the persistent hugepage count via nr_hugepages_mempolicy, any
193 memory policy mode--bind, preferred, local or interleave--may be used. The
194 resulting effect on persistent huge page allocation is as follows:
196 1) Regardless of mempolicy mode [see Documentation/vm/numa_memory_policy.txt],
197 persistent huge pages will be distributed across the node or nodes
198 specified in the mempolicy as if "interleave" had been specified.
199 However, if a node in the policy does not contain sufficient contiguous
200 memory for a huge page, the allocation will not "fallback" to the nearest
201 neighbor node with sufficient contiguous memory. To do this would cause
202 undesirable imbalance in the distribution of the huge page pool, or
203 possibly, allocation of persistent huge pages on nodes not allowed by
204 the task's memory policy.
206 2) One or more nodes may be specified with the bind or interleave policy.
207 If more than one node is specified with the preferred policy, only the
208 lowest numeric id will be used. Local policy will select the node where
209 the task is running at the time the nodes_allowed mask is constructed.
210 For local policy to be deterministic, the task must be bound to a cpu or
211 cpus in a single node. Otherwise, the task could be migrated to some
212 other node at any time after launch and the resulting node will be
213 indeterminate. Thus, local policy is not very useful for this purpose.
214 Any of the other mempolicy modes may be used to specify a single node.
216 3) The nodes allowed mask will be derived from any non-default task mempolicy,
217 whether this policy was set explicitly by the task itself or one of its
218 ancestors, such as numactl. This means that if the task is invoked from a
219 shell with non-default policy, that policy will be used. One can specify a
220 node list of "all" with numactl --interleave or --membind [-m] to achieve
221 interleaving over all nodes in the system or cpuset.
223 4) Any task mempolicy specifed--e.g., using numactl--will be constrained by
224 the resource limits of any cpuset in which the task runs. Thus, there will
225 be no way for a task with non-default policy running in a cpuset with a
226 subset of the system nodes to allocate huge pages outside the cpuset
227 without first moving to a cpuset that contains all of the desired nodes.
229 5) Boot-time huge page allocation attempts to distribute the requested number
230 of huge pages over all on-lines nodes with memory.
232 Per Node Hugepages Attributes
233 =============================
235 A subset of the contents of the root huge page control directory in sysfs,
236 described above, will be replicated under each the system device of each
237 NUMA node with memory in:
239 /sys/devices/system/node/node[0-9]*/hugepages/
241 Under this directory, the subdirectory for each supported huge page size
242 contains the following attribute files:
248 The free_' and surplus_' attribute files are read-only. They return the number
249 of free and surplus [overcommitted] huge pages, respectively, on the parent
252 The nr_hugepages attribute returns the total number of huge pages on the
253 specified node. When this attribute is written, the number of persistent huge
254 pages on the parent node will be adjusted to the specified value, if sufficient
255 resources exist, regardless of the task's mempolicy or cpuset constraints.
257 Note that the number of overcommit and reserve pages remain global quantities,
258 as we don't know until fault time, when the faulting task's mempolicy is
259 applied, from which node the huge page allocation will be attempted.
265 If the user applications are going to request huge pages using mmap system
266 call, then it is required that system administrator mount a file system of
270 -o uid=<value>,gid=<value>,mode=<value>,size=<value>,nr_inodes=<value> \
273 This command mounts a (pseudo) filesystem of type hugetlbfs on the directory
274 /mnt/huge. Any files created on /mnt/huge uses huge pages. The uid and gid
275 options sets the owner and group of the root of the file system. By default
276 the uid and gid of the current process are taken. The mode option sets the
277 mode of root of file system to value & 01777. This value is given in octal.
278 By default the value 0755 is picked. The size option sets the maximum value of
279 memory (huge pages) allowed for that filesystem (/mnt/huge). The size is
280 rounded down to HPAGE_SIZE. The option nr_inodes sets the maximum number of
281 inodes that /mnt/huge can use. If the size or nr_inodes option is not
282 provided on command line then no limits are set. For size and nr_inodes
283 options, you can use [G|g]/[M|m]/[K|k] to represent giga/mega/kilo. For
284 example, size=2K has the same meaning as size=2048.
286 While read system calls are supported on files that reside on hugetlb
287 file systems, write system calls are not.
289 Regular chown, chgrp, and chmod commands (with right permissions) could be
290 used to change the file attributes on hugetlbfs.
292 Also, it is important to note that no such mount command is required if the
293 applications are going to use only shmat/shmget system calls or mmap with
294 MAP_HUGETLB. Users who wish to use hugetlb page via shared memory segment
295 should be a member of a supplementary group and system admin needs to
296 configure that gid into /proc/sys/vm/hugetlb_shm_group. It is possible for
297 same or different applications to use any combination of mmaps and shm*
298 calls, though the mount of filesystem will be required for using mmap calls
299 without MAP_HUGETLB. For an example of how to use mmap with MAP_HUGETLB see
305 1) map_hugetlb: see tools/testing/selftests/vm/map_hugetlb.c
307 2) hugepage-shm: see tools/testing/selftests/vm/hugepage-shm.c
309 3) hugepage-mmap: see tools/testing/selftests/vm/hugepage-mmap.c
311 4) The libhugetlbfs (http://libhugetlbfs.sourceforge.net) library provides a
312 wide range of userspace tools to help with huge page usability, environment
313 setup, and control. Furthermore it provides useful test cases that should be
314 used when modifying code to ensure no regressions are introduced.