Tmpfs¶

Tmpfs is a file system which keeps all of its files in virtual memory.

Everything in tmpfs is temporary in the sense that no files will becreated on your hard drive. If you unmount a tmpfs instance,everything stored therein is lost.

tmpfs puts everything into the kernel internal caches and grows andshrinks to accommodate the files it contains and is able to swapunneeded pages out to swap space, if swap was enabled for the tmpfsmount. tmpfs also supports THP.

tmpfs extends ramfs with a few userspace configurable options listed andexplained further below, some of which can be reconfigured dynamically on thefly using a remount (‘mount -o remount ...’) of the filesystem. A tmpfsfilesystem can be resized but it cannot be resized to a size below its currentusage. tmpfs also supports POSIX ACLs, and extended attributes for thetrusted.*, security.* and user.* namespaces. ramfs does not use swap and youcannot modify any parameter for a ramfs filesystem. The size limit of a ramfsfilesystem is how much memory you have available, and so care must be taken ifused so to not run out of memory.

An alternative to tmpfs and ramfs is to use brd to create RAM disks(/dev/ram*), which allows you to simulate a block device disk in physical RAM.To write data you would just then need to create an regular filesystem on topthis ramdisk. As with ramfs, brd ramdisks cannot swap. brd ramdisks are alsoconfigured in size at initialization and you cannot dynamically resize them.Contrary to brd ramdisks, tmpfs has its own filesystem, it does not rely on theblock layer at all.

Since tmpfs lives completely in the page cache and optionally on swap,all tmpfs pages will be shown as “Shmem” in /proc/meminfo and “Shared” infree(1). Notice that these counters also include shared memory(shmem, see ipcs(1)). The most reliable way to get the count isusing df(1) and du(1).

tmpfs has the following uses:

1. There is always a kernel internal mount which you will not see atall. This is used for shared anonymous mappings and SYSV sharedmemory.

   This mount does not depend on CONFIG_TMPFS. If CONFIG_TMPFS is notset, the user visible part of tmpfs is not built. But the internalmechanisms are always present.

2. glibc 2.2 and above expects tmpfs to be mounted at /dev/shm forPOSIX shared memory (shm_open, shm_unlink). Adding the followingline to /etc/fstab should take care of this:

   tmpfs   /dev/shm        tmpfs   defaults        0 0

   Remember to create the directory that you intend to mount tmpfs onif necessary.

   This mount is _not_ needed for SYSV shared memory. The internalmount is used for that. (In the 2.3 kernel versions it wasnecessary to mount the predecessor of tmpfs (shm fs) to use SYSVshared memory.)

3. Some people (including me) find it very convenient to mount ite.g. on /tmp and /var/tmp and have a big swap partition. And nowloop mounts of tmpfs files do work, so mkinitrd shipped by mostdistributions should succeed with a tmpfs /tmp.

4. And probably a lot more I do not know about :-)

tmpfs has three mount options for sizing:

These parameters accept a suffix k, m or g for kilo, mega and giga andcan be changed on remount. The size parameter also accepts a suffix %to limit this tmpfs instance to that percentage of your physical RAM:the default, when neither size nor nr_blocks is specified, is size=50%

If nr_blocks=0 (or size=0), blocks will not be limited in that instance;if nr_inodes=0, inodes will not be limited. It is generally unwise tomount with such options, since it allows any user with write access touse up all the memory on the machine; but enhances the scalability ofthat instance in a system with many CPUs making intensive use of it.

If nr_inodes is not 0, that limited space for inodes is also used up byextended attributes: “df -i“‘s IUsed and IUse% increase, IFree decreases.

tmpfs blocks may be swapped out, when there is a shortage of memory.tmpfs has a mount option to disable its use of swap:

tmpfs also supports Transparent Huge Pages which requires a kernelconfigured with CONFIG_TRANSPARENT_HUGEPAGE and with huge supported foryour system (has_transparent_hugepage(), which is architecture specific).The mount options for this are:

See alsoTransparent Hugepage Support, which describes thesysfs file /sys/kernel/mm/transparent_hugepage/shmem_enabled: which canbe used to deny huge pages on all tmpfs mounts in an emergency, or toforce huge pages on all tmpfs mounts for testing.

tmpfs also supports quota with the following mount options

None of the quota related mount options can be set or changed on remount.

Quota limit parameters accept a suffix k, m or g for kilo, mega and gigaand can’t be changed on remount. Default global quota limits are takingeffect for any and all user/group/project except root the first time thequota entry for user/group/project id is being accessed - typically thefirst time an inode with a particular id ownership is being created afterthe mount. In other words, instead of the limits being initialized to zero,they are initialized with the particular value provided with these mountoptions. The limits can be changed for any user/group id at any time as theynormally can be.

Note that tmpfs quotas do not support user namespaces so no uid/gidtranslation is done if quotas are enabled inside user namespaces.

tmpfs has a mount option to set the NUMA memory allocation policy forall files in that instance (if CONFIG_NUMA is enabled) - which can beadjusted on the fly via ‘mount -o remount ...’

NodeList format is a comma-separated list of decimal numbers and ranges,a range being two hyphen-separated decimal numbers, the smallest andlargest node numbers in the range. For example, mpol=bind:0-3,5,7,9-15

A memory policy with a valid NodeList will be saved, as specified, foruse at file creation time. When a task allocates a file in the filesystem, the mount option memory policy will be applied with a NodeList,if any, modified by the calling task’s cpuset constraints[SeeCPUSETS] and any optional flags,listed below. If the resulting NodeLists is the empty set, the effectivememory policy for the file will revert to “default” policy.

NUMA memory allocation policies have optional flags that can be used inconjunction with their modes. These optional flags can be specifiedwhen tmpfs is mounted by appending them to the mode before the NodeList.SeeNUMA Memory Policy for a list ofall available memory allocation policy mode flags and their effect onmemory policy.

=static         is equivalent to        MPOL_F_STATIC_NODES=relative       is equivalent to        MPOL_F_RELATIVE_NODES

For example, mpol=bind=static:NodeList, is the equivalent of anallocation policy of MPOL_BIND | MPOL_F_STATIC_NODES.

Note that trying to mount a tmpfs with an mpol option will fail if therunning kernel does not support NUMA; and will fail if its nodelistspecifies a node which is not online. If your system relies on thattmpfs being mounted, but from time to time runs a kernel built withoutNUMA capability (perhaps a safe recovery kernel), or with fewer nodesonline, then it is advisable to omit the mpol option from automaticmount options. It can be added later, when the tmpfs is already mountedon MountPoint, by ‘mount -o remount,mpol=Policy:NodeList MountPoint’.

To specify the initial root directory you can use the following mountoptions:

These options do not have any effect on remount. You can change theseparameters with chmod(1), chown(1) and chgrp(1) on a mounted filesystem.

tmpfs has a mount option to select whether it will wrap at 32- or 64-bit inodenumbers:

On a 32-bit kernel, inode32 is implicit, and inode64 is refused at mount time.On a 64-bit kernel, CONFIG_TMPFS_INODE64 sets the default. inode64 avoids thepossibility of multiple files with the same inode number on a single device;but risks glibc failing with EOVERFLOW once 33-bit inode numbers are reached -if a long-lived tmpfs is accessed by 32-bit applications so ancient thatopening a file larger than 2GiB fails with EINVAL.

So ‘mount -t tmpfs -o size=10G,nr_inodes=10k,mode=700 tmpfs /mytmpfs’will give you tmpfs instance on /mytmpfs which can allocate 10GBRAM/SWAP in 10240 inodes and it is only accessible by root.

tmpfs has the following mounting options for case-insensitive lookup support:

This option doesn’t render the entire filesystem case-insensitive. One needs tostill set the casefold flag per directory, by flipping the +F attribute in anempty directory. Nevertheless, new directories will inherit the attribute. Themountpoint itself cannot be made case-insensitive.

Example:

$ mount -t tmpfs -o casefold=utf8-12.1.0,strict_encoding fs_name /mytmpfs$ mount -t tmpfs -o casefold fs_name /mytmpfs

Author:

Christoph Rohland <cr@sap.com>, 1.12.01

Updated:

Hugh Dickins, 4 June 2007

Updated:

KOSAKI Motohiro, 16 Mar 2010

Updated:

Chris Down, 13 July 2020

Updated:

André Almeida, 23 Aug 2024