AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-16 09:50:42 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity

docs/vm: hugetlbpage.txt: convert to ReST format

Browse Source 
Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
This commit is contained in:
Mike Rapoport 2018-03-21 21:22:23 +02:00 committed by Jonathan Corbet
parent aa9f34e5da
commit 148723f711
1 changed files with 136 additions and 101 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

237
Documentation/vm/hugetlbpage.txt
View File

@ -1,3 +1,11 @@
.. _hugetlbpage:
=============
HugeTLB Pages
=============
Overview
========
The intent of this file is to give a brief summary of hugetlbpage support in The intent of this file is to give a brief summary of hugetlbpage support in
the Linux kernel. This support is built on top of multiple page size support the Linux kernel. This support is built on top of multiple page size support
@ -18,53 +26,59 @@ First the Linux kernel needs to be built with the CONFIG_HUGETLBFS
automatically when CONFIG_HUGETLBFS is selected) configuration automatically when CONFIG_HUGETLBFS is selected) configuration
options. options.
The /proc/meminfo file provides information about the total number of The ``/proc/meminfo`` file provides information about the total number of
persistent hugetlb pages in the kernel's huge page pool. It also displays persistent hugetlb pages in the kernel's huge page pool. It also displays
default huge page size and information about the number of free, reserved default huge page size and information about the number of free, reserved
and surplus huge pages in the pool of huge pages of default size. and surplus huge pages in the pool of huge pages of default size.
The huge page size is needed for generating the proper alignment and The huge page size is needed for generating the proper alignment and
size of the arguments to system calls that map huge page regions. size of the arguments to system calls that map huge page regions.
The output of "cat /proc/meminfo" will include lines like: The output of ``cat /proc/meminfo`` will include lines like::
..... HugePages_Total: uuu
HugePages_Total: uuu HugePages_Free: vvv
HugePages_Free: vvv HugePages_Rsvd: www
HugePages_Rsvd: www HugePages_Surp: xxx
HugePages_Surp: xxx Hugepagesize: yyy kB
Hugepagesize: yyy kB Hugetlb: zzz kB
Hugetlb: zzz kB
where: where:
HugePages_Total is the size of the pool of huge pages.
HugePages_Free is the number of huge pages in the pool that are not yet HugePages_Total
allocated. is the size of the pool of huge pages.
HugePages_Rsvd is short for "reserved," and is the number of huge pages for HugePages_Free
which a commitment to allocate from the pool has been made, is the number of huge pages in the pool that are not yet
but no allocation has yet been made. Reserved huge pages allocated.
guarantee that an application will be able to allocate a HugePages_Rsvd
huge page from the pool of huge pages at fault time. is short for "reserved," and is the number of huge pages for
HugePages_Surp is short for "surplus," and is the number of huge pages in which a commitment to allocate from the pool has been made,
the pool above the value in /proc/sys/vm/nr_hugepages. The but no allocation has yet been made. Reserved huge pages
maximum number of surplus huge pages is controlled by guarantee that an application will be able to allocate a
/proc/sys/vm/nr_overcommit_hugepages. huge page from the pool of huge pages at fault time.
Hugepagesize is the default hugepage size (in Kb). HugePages_Surp
Hugetlb is the total amount of memory (in kB), consumed by huge is short for "surplus," and is the number of huge pages in
pages of all sizes. the pool above the value in ``/proc/sys/vm/nr_hugepages``. The
If huge pages of different sizes are in use, this number maximum number of surplus huge pages is controlled by
will exceed HugePages_Total * Hugepagesize. To get more ``/proc/sys/vm/nr_overcommit_hugepages``.
detailed information, please, refer to Hugepagesize
/sys/kernel/mm/hugepages (described below). is the default hugepage size (in Kb).
Hugetlb
is the total amount of memory (in kB), consumed by huge
pages of all sizes.
If huge pages of different sizes are in use, this number
will exceed HugePages_Total \* Hugepagesize. To get more
detailed information, please, refer to
``/sys/kernel/mm/hugepages`` (described below).
/proc/filesystems should also show a filesystem of type "hugetlbfs" configured ``/proc/filesystems`` should also show a filesystem of type "hugetlbfs"
in the kernel. configured in the kernel.
/proc/sys/vm/nr_hugepages indicates the current number of "persistent" huge ``/proc/sys/vm/nr_hugepages`` indicates the current number of "persistent" huge
pages in the kernel's huge page pool. "Persistent" huge pages will be pages in the kernel's huge page pool. "Persistent" huge pages will be
returned to the huge page pool when freed by a task. A user with root returned to the huge page pool when freed by a task. A user with root
privileges can dynamically allocate more or free some persistent huge pages privileges can dynamically allocate more or free some persistent huge pages
by increasing or decreasing the value of 'nr_hugepages'. by increasing or decreasing the value of ``nr_hugepages``.
Pages that are used as huge pages are reserved inside the kernel and cannot Pages that are used as huge pages are reserved inside the kernel and cannot
be used for other purposes. Huge pages cannot be swapped out under be used for other purposes. Huge pages cannot be swapped out under
@ -86,10 +100,10 @@ with a huge page size selection parameter "hugepagesz=<size>". <size> must
be specified in bytes with optional scale suffix [kKmMgG]. The default huge be specified in bytes with optional scale suffix [kKmMgG]. The default huge
page size may be selected with the "default_hugepagesz=<size>" boot parameter. page size may be selected with the "default_hugepagesz=<size>" boot parameter.
When multiple huge page sizes are supported, /proc/sys/vm/nr_hugepages When multiple huge page sizes are supported, ``/proc/sys/vm/nr_hugepages``
indicates the current number of pre-allocated huge pages of the default size. indicates the current number of pre-allocated huge pages of the default size.
Thus, one can use the following command to dynamically allocate/deallocate Thus, one can use the following command to dynamically allocate/deallocate
default sized persistent huge pages: default sized persistent huge pages::
echo 20 > /proc/sys/vm/nr_hugepages echo 20 > /proc/sys/vm/nr_hugepages
@ -98,7 +112,7 @@ huge page pool to 20, allocating or freeing huge pages, as required.
On a NUMA platform, the kernel will attempt to distribute the huge page pool On a NUMA platform, the kernel will attempt to distribute the huge page pool
over all the set of allowed nodes specified by the NUMA memory policy of the over all the set of allowed nodes specified by the NUMA memory policy of the
task that modifies nr_hugepages. The default for the allowed nodes--when the task that modifies ``nr_hugepages``. The default for the allowed nodes--when the
task has default memory policy--is all on-line nodes with memory. Allowed task has default memory policy--is all on-line nodes with memory. Allowed
nodes with insufficient available, contiguous memory for a huge page will be nodes with insufficient available, contiguous memory for a huge page will be
silently skipped when allocating persistent huge pages. See the discussion silently skipped when allocating persistent huge pages. See the discussion
@ -117,51 +131,52 @@ init files. This will enable the kernel to allocate huge pages early in
the boot process when the possibility of getting physical contiguous pages the boot process when the possibility of getting physical contiguous pages
is still very high. Administrators can verify the number of huge pages is still very high. Administrators can verify the number of huge pages
actually allocated by checking the sysctl or meminfo. To check the per node actually allocated by checking the sysctl or meminfo. To check the per node
distribution of huge pages in a NUMA system, use: distribution of huge pages in a NUMA system, use::
cat /sys/devices/system/node/node*/meminfo | fgrep Huge cat /sys/devices/system/node/node*/meminfo | fgrep Huge
/proc/sys/vm/nr_overcommit_hugepages specifies how large the pool of ``/proc/sys/vm/nr_overcommit_hugepages`` specifies how large the pool of
huge pages can grow, if more huge pages than /proc/sys/vm/nr_hugepages are huge pages can grow, if more huge pages than ``/proc/sys/vm/nr_hugepages`` are
requested by applications. Writing any non-zero value into this file requested by applications. Writing any non-zero value into this file
indicates that the hugetlb subsystem is allowed to try to obtain that indicates that the hugetlb subsystem is allowed to try to obtain that
number of "surplus" huge pages from the kernel's normal page pool, when the number of "surplus" huge pages from the kernel's normal page pool, when the
persistent huge page pool is exhausted. As these surplus huge pages become persistent huge page pool is exhausted. As these surplus huge pages become
unused, they are freed back to the kernel's normal page pool. unused, they are freed back to the kernel's normal page pool.
When increasing the huge page pool size via nr_hugepages, any existing surplus When increasing the huge page pool size via ``nr_hugepages``, any existing
pages will first be promoted to persistent huge pages. Then, additional surplus pages will first be promoted to persistent huge pages. Then, additional
huge pages will be allocated, if necessary and if possible, to fulfill huge pages will be allocated, if necessary and if possible, to fulfill
the new persistent huge page pool size. the new persistent huge page pool size.
The administrator may shrink the pool of persistent huge pages for The administrator may shrink the pool of persistent huge pages for
the default huge page size by setting the nr_hugepages sysctl to a the default huge page size by setting the ``nr_hugepages`` sysctl to a
smaller value. The kernel will attempt to balance the freeing of huge pages smaller value. The kernel will attempt to balance the freeing of huge pages
across all nodes in the memory policy of the task modifying nr_hugepages. across all nodes in the memory policy of the task modifying ``nr_hugepages``.
Any free huge pages on the selected nodes will be freed back to the kernel's Any free huge pages on the selected nodes will be freed back to the kernel's
normal page pool. normal page pool.
Caveat: Shrinking the persistent huge page pool via nr_hugepages such that Caveat: Shrinking the persistent huge page pool via ``nr_hugepages`` such that
it becomes less than the number of huge pages in use will convert the balance it becomes less than the number of huge pages in use will convert the balance
of the in-use huge pages to surplus huge pages. This will occur even if of the in-use huge pages to surplus huge pages. This will occur even if
the number of surplus pages it would exceed the overcommit value. As long as the number of surplus pages it would exceed the overcommit value. As long as
this condition holds--that is, until nr_hugepages+nr_overcommit_hugepages is this condition holds--that is, until ``nr_hugepages+nr_overcommit_hugepages`` is
increased sufficiently, or the surplus huge pages go out of use and are freed-- increased sufficiently, or the surplus huge pages go out of use and are freed--
no more surplus huge pages will be allowed to be allocated. no more surplus huge pages will be allowed to be allocated.
With support for multiple huge page pools at run-time available, much of With support for multiple huge page pools at run-time available, much of
the huge page userspace interface in /proc/sys/vm has been duplicated in sysfs. the huge page userspace interface in ``/proc/sys/vm`` has been duplicated in
The /proc interfaces discussed above have been retained for backwards sysfs.
compatibility. The root huge page control directory in sysfs is: The ``/proc`` interfaces discussed above have been retained for backwards
compatibility. The root huge page control directory in sysfs is::
/sys/kernel/mm/hugepages /sys/kernel/mm/hugepages
For each huge page size supported by the running kernel, a subdirectory For each huge page size supported by the running kernel, a subdirectory
will exist, of the form: will exist, of the form::
hugepages-${size}kB hugepages-${size}kB
Inside each of these directories, the same set of files will exist: Inside each of these directories, the same set of files will exist::
nr_hugepages nr_hugepages
nr_hugepages_mempolicy nr_hugepages_mempolicy
@ -176,33 +191,33 @@ which function as described above for the default huge page-sized case.
Interaction of Task Memory Policy with Huge Page Allocation/Freeing Interaction of Task Memory Policy with Huge Page Allocation/Freeing
=================================================================== ===================================================================
Whether huge pages are allocated and freed via the /proc interface or Whether huge pages are allocated and freed via the ``/proc`` interface or
the /sysfs interface using the nr_hugepages_mempolicy attribute, the NUMA the ``/sysfs`` interface using the ``nr_hugepages_mempolicy`` attribute, the
nodes from which huge pages are allocated or freed are controlled by the NUMA nodes from which huge pages are allocated or freed are controlled by the
NUMA memory policy of the task that modifies the nr_hugepages_mempolicy NUMA memory policy of the task that modifies the ``nr_hugepages_mempolicy``
sysctl or attribute. When the nr_hugepages attribute is used, mempolicy sysctl or attribute. When the ``nr_hugepages`` attribute is used, mempolicy
is ignored. is ignored.
The recommended method to allocate or free huge pages to/from the kernel The recommended method to allocate or free huge pages to/from the kernel
huge page pool, using the nr_hugepages example above, is: huge page pool, using the ``nr_hugepages`` example above, is::
numactl --interleave <node-list> echo 20 \ numactl --interleave <node-list> echo 20 \
>/proc/sys/vm/nr_hugepages_mempolicy >/proc/sys/vm/nr_hugepages_mempolicy
or, more succinctly: or, more succinctly::
numactl -m <node-list> echo 20 >/proc/sys/vm/nr_hugepages_mempolicy numactl -m <node-list> echo 20 >/proc/sys/vm/nr_hugepages_mempolicy
This will allocate or free abs(20 - nr_hugepages) to or from the nodes This will allocate or free ``abs(20 - nr_hugepages)`` to or from the nodes
specified in <node-list>, depending on whether number of persistent huge pages specified in <node-list>, depending on whether number of persistent huge pages
is initially less than or greater than 20, respectively. No huge pages will be is initially less than or greater than 20, respectively. No huge pages will be
allocated nor freed on any node not included in the specified <node-list>. allocated nor freed on any node not included in the specified <node-list>.
When adjusting the persistent hugepage count via nr_hugepages_mempolicy, any When adjusting the persistent hugepage count via ``nr_hugepages_mempolicy``, any
memory policy mode--bind, preferred, local or interleave--may be used. The memory policy mode--bind, preferred, local or interleave--may be used. The
resulting effect on persistent huge page allocation is as follows: resulting effect on persistent huge page allocation is as follows:
1) Regardless of mempolicy mode [see Documentation/vm/numa_memory_policy.txt], #. Regardless of mempolicy mode [see Documentation/vm/numa_memory_policy.txt],
persistent huge pages will be distributed across the node or nodes persistent huge pages will be distributed across the node or nodes
specified in the mempolicy as if "interleave" had been specified. specified in the mempolicy as if "interleave" had been specified.
However, if a node in the policy does not contain sufficient contiguous However, if a node in the policy does not contain sufficient contiguous
@ -212,7 +227,7 @@ resulting effect on persistent huge page allocation is as follows:
possibly, allocation of persistent huge pages on nodes not allowed by possibly, allocation of persistent huge pages on nodes not allowed by
the task's memory policy. the task's memory policy.
2) One or more nodes may be specified with the bind or interleave policy. #. One or more nodes may be specified with the bind or interleave policy.
If more than one node is specified with the preferred policy, only the If more than one node is specified with the preferred policy, only the
lowest numeric id will be used. Local policy will select the node where lowest numeric id will be used. Local policy will select the node where
the task is running at the time the nodes_allowed mask is constructed. the task is running at the time the nodes_allowed mask is constructed.
@ -222,20 +237,20 @@ resulting effect on persistent huge page allocation is as follows:
indeterminate. Thus, local policy is not very useful for this purpose. indeterminate. Thus, local policy is not very useful for this purpose.
Any of the other mempolicy modes may be used to specify a single node. Any of the other mempolicy modes may be used to specify a single node.
3) The nodes allowed mask will be derived from any non-default task mempolicy, #. The nodes allowed mask will be derived from any non-default task mempolicy,
whether this policy was set explicitly by the task itself or one of its whether this policy was set explicitly by the task itself or one of its
ancestors, such as numactl. This means that if the task is invoked from a ancestors, such as numactl. This means that if the task is invoked from a
shell with non-default policy, that policy will be used. One can specify a shell with non-default policy, that policy will be used. One can specify a
node list of "all" with numactl --interleave or --membind [-m] to achieve node list of "all" with numactl --interleave or --membind [-m] to achieve
interleaving over all nodes in the system or cpuset. interleaving over all nodes in the system or cpuset.
4) Any task mempolicy specified--e.g., using numactl--will be constrained by #. Any task mempolicy specified--e.g., using numactl--will be constrained by
the resource limits of any cpuset in which the task runs. Thus, there will the resource limits of any cpuset in which the task runs. Thus, there will
be no way for a task with non-default policy running in a cpuset with a be no way for a task with non-default policy running in a cpuset with a
subset of the system nodes to allocate huge pages outside the cpuset subset of the system nodes to allocate huge pages outside the cpuset
without first moving to a cpuset that contains all of the desired nodes. without first moving to a cpuset that contains all of the desired nodes.
5) Boot-time huge page allocation attempts to distribute the requested number #. Boot-time huge page allocation attempts to distribute the requested number
of huge pages over all on-lines nodes with memory. of huge pages over all on-lines nodes with memory.
Per Node Hugepages Attributes Per Node Hugepages Attributes
@ -243,22 +258,22 @@ Per Node Hugepages Attributes
A subset of the contents of the root huge page control directory in sysfs, A subset of the contents of the root huge page control directory in sysfs,
described above, will be replicated under each the system device of each described above, will be replicated under each the system device of each
NUMA node with memory in: NUMA node with memory in::
/sys/devices/system/node/node[0-9]*/hugepages/ /sys/devices/system/node/node[0-9]*/hugepages/
Under this directory, the subdirectory for each supported huge page size Under this directory, the subdirectory for each supported huge page size
contains the following attribute files: contains the following attribute files::
nr_hugepages nr_hugepages
free_hugepages free_hugepages
surplus_hugepages surplus_hugepages
The free_' and surplus_' attribute files are read-only. They return the number The free\_' and surplus\_' attribute files are read-only. They return the number
of free and surplus [overcommitted] huge pages, respectively, on the parent of free and surplus [overcommitted] huge pages, respectively, on the parent
node. node.
The nr_hugepages attribute returns the total number of huge pages on the The ``nr_hugepages`` attribute returns the total number of huge pages on the
specified node. When this attribute is written, the number of persistent huge specified node. When this attribute is written, the number of persistent huge
pages on the parent node will be adjusted to the specified value, if sufficient pages on the parent node will be adjusted to the specified value, if sufficient
resources exist, regardless of the task's mempolicy or cpuset constraints. resources exist, regardless of the task's mempolicy or cpuset constraints.
@ -273,37 +288,51 @@ Using Huge Pages
If the user applications are going to request huge pages using mmap system If the user applications are going to request huge pages using mmap system
call, then it is required that system administrator mount a file system of call, then it is required that system administrator mount a file system of
type hugetlbfs: type hugetlbfs::
mount -t hugetlbfs \ mount -t hugetlbfs \
-o uid=<value>,gid=<value>,mode=<value>,pagesize=<value>,size=<value>,\ -o uid=<value>,gid=<value>,mode=<value>,pagesize=<value>,size=<value>,\
min_size=<value>,nr_inodes=<value> none /mnt/huge min_size=<value>,nr_inodes=<value> none /mnt/huge
This command mounts a (pseudo) filesystem of type hugetlbfs on the directory This command mounts a (pseudo) filesystem of type hugetlbfs on the directory
/mnt/huge. Any files created on /mnt/huge uses huge pages. The uid and gid ``/mnt/huge``. Any files created on ``/mnt/huge`` uses huge pages.
options sets the owner and group of the root of the file system. By default
the uid and gid of the current process are taken. The mode option sets the The ``uid`` and ``gid`` options sets the owner and group of the root of the
mode of root of file system to value & 01777. This value is given in octal. file system. By default the ``uid`` and ``gid`` of the current process
By default the value 0755 is picked. If the platform supports multiple huge are taken.
page sizes, the pagesize option can be used to specify the huge page size and
associated pool. pagesize is specified in bytes. If pagesize is not specified The ``mode`` option sets the mode of root of file system to value & 01777.
the platform's default huge page size and associated pool will be used. The This value is given in octal. By default the value 0755 is picked.
size option sets the maximum value of memory (huge pages) allowed for that
filesystem (/mnt/huge). The size option can be specified in bytes, or as a If the platform supports multiple huge page sizes, the ``pagesize`` option can
percentage of the specified huge page pool (nr_hugepages). The size is be used to specify the huge page size and associated pool. ``pagesize``
rounded down to HPAGE_SIZE boundary. The min_size option sets the minimum is specified in bytes. If ``pagesize`` is not specified the platform's
value of memory (huge pages) allowed for the filesystem. min_size can be default huge page size and associated pool will be used.
specified in the same way as size, either bytes or a percentage of the
huge page pool. At mount time, the number of huge pages specified by The ``size`` option sets the maximum value of memory (huge pages) allowed
min_size are reserved for use by the filesystem. If there are not enough for that filesystem (``/mnt/huge``). The ``size`` option can be specified
free huge pages available, the mount will fail. As huge pages are allocated in bytes, or as a percentage of the specified huge page pool (``nr_hugepages``).
to the filesystem and freed, the reserve count is adjusted so that the sum The size is rounded down to HPAGE_SIZE boundary.
of allocated and reserved huge pages is always at least min_size. The option
nr_inodes sets the maximum number of inodes that /mnt/huge can use. If the The ``min_size`` option sets the minimum value of memory (huge pages) allowed
size, min_size or nr_inodes option is not provided on command line then for the filesystem. ``min_size`` can be specified in the same way as ``size``,
no limits are set. For pagesize, size, min_size and nr_inodes options, you either bytes or a percentage of the huge page pool.
can use [G|g]/[M|m]/[K|k] to represent giga/mega/kilo. For example, size=2K At mount time, the number of huge pages specified by ``min_size`` are reserved
has the same meaning as size=2048. for use by the filesystem.
If there are not enough free huge pages available, the mount will fail.
As huge pages are allocated to the filesystem and freed, the reserve count
is adjusted so that the sum of allocated and reserved huge pages is always
at least ``min_size``.
The option ``nr_inodes`` sets the maximum number of inodes that ``/mnt/huge``
can use.
If the ``size``, ``min_size`` or ``nr_inodes`` option is not provided on
command line then no limits are set.
For ``pagesize``, ``size``, ``min_size`` and ``nr_inodes`` options, you can
use [G|g]/[M|m]/[K|k] to represent giga/mega/kilo.
For example, size=2K has the same meaning as size=2048.
While read system calls are supported on files that reside on hugetlb While read system calls are supported on files that reside on hugetlb
file systems, write system calls are not. file systems, write system calls are not.
@ -313,12 +342,12 @@ used to change the file attributes on hugetlbfs.
Also, it is important to note that no such mount command is required if Also, it is important to note that no such mount command is required if
applications are going to use only shmat/shmget system calls or mmap with applications are going to use only shmat/shmget system calls or mmap with
MAP_HUGETLB. For an example of how to use mmap with MAP_HUGETLB see map_hugetlb MAP_HUGETLB. For an example of how to use mmap with MAP_HUGETLB see
below. :ref:`map_hugetlb <map_hugetlb>` below.
Users who wish to use hugetlb memory via shared memory segment should be a Users who wish to use hugetlb memory via shared memory segment should be a
member of a supplementary group and system admin needs to configure that gid member of a supplementary group and system admin needs to configure that gid
into /proc/sys/vm/hugetlb_shm_group. It is possible for same or different into ``/proc/sys/vm/hugetlb_shm_group``. It is possible for same or different
applications to use any combination of mmaps and shm* calls, though the mount of applications to use any combination of mmaps and shm* calls, though the mount of
filesystem will be required for using mmap calls without MAP_HUGETLB. filesystem will be required for using mmap calls without MAP_HUGETLB.
@ -332,15 +361,21 @@ a hugetlb page and the length is smaller than the hugepage size.
Examples Examples
======== ========
1) map_hugetlb: see tools/testing/selftests/vm/map_hugetlb.c .. _map_hugetlb:
2) hugepage-shm: see tools/testing/selftests/vm/hugepage-shm.c ``map_hugetlb``
see tools/testing/selftests/vm/map_hugetlb.c
3) hugepage-mmap: see tools/testing/selftests/vm/hugepage-mmap.c ``hugepage-shm``
see tools/testing/selftests/vm/hugepage-shm.c
4) The libhugetlbfs (https://github.com/libhugetlbfs/libhugetlbfs) library ``hugepage-mmap``
provides a wide range of userspace tools to help with huge page usability, see tools/testing/selftests/vm/hugepage-mmap.c
environment setup, and control.
The `libhugetlbfs`_ library provides a wide range of userspace tools
to help with huge page usability, environment setup, and control.
.. _libhugetlbfs: https://github.com/libhugetlbfs/libhugetlbfs
Kernel development regression testing Kernel development regression testing
===================================== =====================================