.\" Copyright (c) 2016 by Michael Kerrisk .\" .\" SPDX-License-Identifier: Linux-man-pages-copyleft .\" .TH tmpfs 5 2023-07-28 "Linux man-pages 6.05.01" .SH NAME tmpfs \- a virtual memory filesystem .SH DESCRIPTION The .B tmpfs facility allows the creation of filesystems whose contents reside in virtual memory. Since the files on such filesystems typically reside in RAM, file access is extremely fast. .PP The filesystem is automatically created when mounting a filesystem with the type .B tmpfs via a command such as the following: .PP .in +4n .EX $ sudo mount \-t tmpfs \-o size=10M tmpfs /mnt/mytmpfs .EE .in .PP A .B tmpfs filesystem has the following properties: .IP \[bu] 3 The filesystem can employ swap space when physical memory pressure demands it. .IP \[bu] The filesystem consumes only as much physical memory and swap space as is required to store the current contents of the filesystem. .IP \[bu] During a remount operation .RI ( "mount\ \-o\ remount" ), the filesystem size can be changed (without losing the existing contents of the filesystem). .PP If a .B tmpfs filesystem is unmounted, its contents are discarded (lost). .\" See mm/shmem.c:shmem_parse_options for options it supports. .SS Mount options The .B tmpfs filesystem supports the following mount options: .TP .BR size "=\fIbytes\fP" Specify an upper limit on the size of the filesystem. The size is given in bytes, and rounded up to entire pages. .IP The size may have a .BR k , .BR m , or .B g suffix for Ki, Mi, Gi (binary kilo (kibi), binary mega (mebi), and binary giga (gibi)). .IP The size may also have a % suffix to limit this instance to a percentage of physical RAM. .IP The default, when neither .B size nor .B nr_blocks is specified, is .IR size=50% . .TP .BR nr_blocks "=\fIblocks\fP" The same as .BR size , but in blocks of .BR PAGE_CACHE_SIZE . .IP Blocks may be specified with .BR k , .BR m , or .B g suffixes like .BR size , but not a % suffix. .TP .BR nr_inodes "=\fIinodes\fP" The maximum number of inodes for this instance. The default is half of the number of your physical RAM pages, or (on a machine with highmem) the number of lowmem RAM pages, whichever is smaller. .IP Inodes may be specified with .BR k , .BR m , or .B g suffixes like .BR size , but not a % suffix. .TP .BR mode "=\fImode\fP" Set initial permissions of the root directory. .TP .BR gid "=\fIgid\fP (since Linux 2.5.7)" .\" Technically this is also in some version of Linux 2.4. .\" commit 099445b489625b80b1d6687c9b6072dbeaca4096 Set the initial group ID of the root directory. .TP .BR uid "=\fIuid\fP (since Linux 2.5.7)" .\" Technically this is also in some version of Linux 2.4. .\" commit 099445b489625b80b1d6687c9b6072dbeaca4096 Set the initial user ID of the root directory. .TP .BR huge "=\fIhuge_option\fR (since Linux 4.7.0)" .\" commit 5a6e75f8110c97e2a5488894d4e922187e6cb343 Set the huge table memory allocation policy for all files in this instance (if .B CONFIG_TRANSPARENT_HUGEPAGE is enabled). .IP The .I huge_option value is one of the following: .RS .TP .B never Do not allocate huge pages. This is the default. .TP .B always Attempt to allocate huge pages every time a new page is needed. .TP .B within_size Only allocate huge page if it will be fully within .IR i_size . Also respect .BR fadvise (2) and .BR madvise (2) hints .TP .B advise Only allocate huge pages if requested with .BR fadvise (2) or .BR madvise (2). .TP .B deny For use in emergencies, to force the huge option off from all mounts. .TP .B force Force the huge option on for all mounts; useful for testing. .RE .TP .BR mpol "=\fImpol_option\fR (since Linux 2.6.15)" .\" commit 7339ff8302fd70aabf5f1ae26e0c4905fa74a495 Set the NUMA memory allocation policy for all files in this instance (if .B CONFIG_NUMA is enabled). .IP The .I mpol_option value is one of the following: .RS .TP .B default Use the process allocation policy (see .BR set_mempolicy (2)). .TP .BR prefer ":\fInode\fP" Preferably allocate memory from the given .IR node . .TP .BR bind ":\fInodelist\fP" Allocate memory only from nodes in .IR nodelist . .TP .B interleave Allocate from each node in turn. .TP .BR interleave ":\fInodelist\fP" Allocate from each node of .I in turn. .TP .B local Preferably allocate memory from the local node. .RE .IP In the above, .I nodelist is a comma-separated list of decimal numbers and ranges that specify NUMA nodes. A range is a pair of hyphen-separated decimal numbers, the smallest and largest node numbers in the range. For example, .IR mpol=bind:0\-3,5,7,9\-15 . .SH VERSIONS The .B tmpfs facility was added in Linux 2.4, as a successor to the older .B ramfs facility, which did not provide limit checking or allow for the use of swap space. .SH NOTES In order for user-space tools and applications to create .B tmpfs filesystems, the kernel must be configured with the .B CONFIG_TMPFS option. .PP The .B tmpfs filesystem supports extended attributes (see .BR xattr (7)), but .I user extended attributes are not permitted. .PP An internal shared memory filesystem is used for System V shared memory .RB ( shmget (2)) and shared anonymous mappings .RB ( mmap (2) with the .B MAP_SHARED and .B MAP_ANONYMOUS flags). This filesystem is available regardless of whether the kernel was configured with the .B CONFIG_TMPFS option. .PP A .B tmpfs filesystem mounted at .I /dev/shm is used for the implementation of POSIX shared memory .RB ( shm_overview (7)) and POSIX semaphores .RB ( sem_overview (7)). .PP The amount of memory consumed by all .B tmpfs filesystems is shown in the .I Shmem field of .I /proc/meminfo and in the .I shared field displayed by .BR free (1). .PP The .B tmpfs facility was formerly called .BR shmfs . .SH SEE ALSO .BR df (1), .BR du (1), .BR memfd_create (2), .BR mmap (2), .BR set_mempolicy (2), .BR shm_open (3), .BR mount (8) .PP The kernel source files .I Documentation/filesystems/tmpfs.txt and .IR Documentation/admin\-guide/mm/transhuge.rst .