253 lines · plain
1.. SPDX-License-Identifier: GPL-2.02 3=====4Tmpfs5=====6 7Tmpfs is a file system which keeps all of its files in virtual memory.8 9 10Everything in tmpfs is temporary in the sense that no files will be11created on your hard drive. If you unmount a tmpfs instance,12everything stored therein is lost.13 14tmpfs puts everything into the kernel internal caches and grows and15shrinks to accommodate the files it contains and is able to swap16unneeded pages out to swap space, if swap was enabled for the tmpfs17mount. tmpfs also supports THP.18 19tmpfs extends ramfs with a few userspace configurable options listed and20explained further below, some of which can be reconfigured dynamically on the21fly using a remount ('mount -o remount ...') of the filesystem. A tmpfs22filesystem can be resized but it cannot be resized to a size below its current23usage. tmpfs also supports POSIX ACLs, and extended attributes for the24trusted.*, security.* and user.* namespaces. ramfs does not use swap and you25cannot modify any parameter for a ramfs filesystem. The size limit of a ramfs26filesystem is how much memory you have available, and so care must be taken if27used so to not run out of memory.28 29An alternative to tmpfs and ramfs is to use brd to create RAM disks30(/dev/ram*), which allows you to simulate a block device disk in physical RAM.31To write data you would just then need to create an regular filesystem on top32this ramdisk. As with ramfs, brd ramdisks cannot swap. brd ramdisks are also33configured in size at initialization and you cannot dynamically resize them.34Contrary to brd ramdisks, tmpfs has its own filesystem, it does not rely on the35block layer at all.36 37Since tmpfs lives completely in the page cache and optionally on swap,38all tmpfs pages will be shown as "Shmem" in /proc/meminfo and "Shared" in39free(1). Notice that these counters also include shared memory40(shmem, see ipcs(1)). The most reliable way to get the count is41using df(1) and du(1).42 43tmpfs has the following uses:44 451) There is always a kernel internal mount which you will not see at46 all. This is used for shared anonymous mappings and SYSV shared47 memory.48 49 This mount does not depend on CONFIG_TMPFS. If CONFIG_TMPFS is not50 set, the user visible part of tmpfs is not built. But the internal51 mechanisms are always present.52 532) glibc 2.2 and above expects tmpfs to be mounted at /dev/shm for54 POSIX shared memory (shm_open, shm_unlink). Adding the following55 line to /etc/fstab should take care of this::56 57 tmpfs /dev/shm tmpfs defaults 0 058 59 Remember to create the directory that you intend to mount tmpfs on60 if necessary.61 62 This mount is _not_ needed for SYSV shared memory. The internal63 mount is used for that. (In the 2.3 kernel versions it was64 necessary to mount the predecessor of tmpfs (shm fs) to use SYSV65 shared memory.)66 673) Some people (including me) find it very convenient to mount it68 e.g. on /tmp and /var/tmp and have a big swap partition. And now69 loop mounts of tmpfs files do work, so mkinitrd shipped by most70 distributions should succeed with a tmpfs /tmp.71 724) And probably a lot more I do not know about :-)73 74 75tmpfs has three mount options for sizing:76 77========= ============================================================78size The limit of allocated bytes for this tmpfs instance. The79 default is half of your physical RAM without swap. If you80 oversize your tmpfs instances the machine will deadlock81 since the OOM handler will not be able to free that memory.82nr_blocks The same as size, but in blocks of PAGE_SIZE.83nr_inodes The maximum number of inodes for this instance. The default84 is half of the number of your physical RAM pages, or (on a85 machine with highmem) the number of lowmem RAM pages,86 whichever is the lower.87========= ============================================================88 89These parameters accept a suffix k, m or g for kilo, mega and giga and90can be changed on remount. The size parameter also accepts a suffix %91to limit this tmpfs instance to that percentage of your physical RAM:92the default, when neither size nor nr_blocks is specified, is size=50%93 94If nr_blocks=0 (or size=0), blocks will not be limited in that instance;95if nr_inodes=0, inodes will not be limited. It is generally unwise to96mount with such options, since it allows any user with write access to97use up all the memory on the machine; but enhances the scalability of98that instance in a system with many CPUs making intensive use of it.99 100If nr_inodes is not 0, that limited space for inodes is also used up by101extended attributes: "df -i"'s IUsed and IUse% increase, IFree decreases.102 103tmpfs blocks may be swapped out, when there is a shortage of memory.104tmpfs has a mount option to disable its use of swap:105 106====== ===========================================================107noswap Disables swap. Remounts must respect the original settings.108 By default swap is enabled.109====== ===========================================================110 111tmpfs also supports Transparent Huge Pages which requires a kernel112configured with CONFIG_TRANSPARENT_HUGEPAGE and with huge supported for113your system (has_transparent_hugepage(), which is architecture specific).114The mount options for this are:115 116================ ==============================================================117huge=never Do not allocate huge pages. This is the default.118huge=always Attempt to allocate huge page every time a new page is needed.119huge=within_size Only allocate huge page if it will be fully within i_size.120 Also respect madvise(2) hints.121huge=advise Only allocate huge page if requested with madvise(2).122================ ==============================================================123 124See also Documentation/admin-guide/mm/transhuge.rst, which describes the125sysfs file /sys/kernel/mm/transparent_hugepage/shmem_enabled: which can126be used to deny huge pages on all tmpfs mounts in an emergency, or to127force huge pages on all tmpfs mounts for testing.128 129tmpfs also supports quota with the following mount options130 131======================== =================================================132quota User and group quota accounting and enforcement133 is enabled on the mount. Tmpfs is using hidden134 system quota files that are initialized on mount.135usrquota User quota accounting and enforcement is enabled136 on the mount.137grpquota Group quota accounting and enforcement is enabled138 on the mount.139usrquota_block_hardlimit Set global user quota block hard limit.140usrquota_inode_hardlimit Set global user quota inode hard limit.141grpquota_block_hardlimit Set global group quota block hard limit.142grpquota_inode_hardlimit Set global group quota inode hard limit.143======================== =================================================144 145None of the quota related mount options can be set or changed on remount.146 147Quota limit parameters accept a suffix k, m or g for kilo, mega and giga148and can't be changed on remount. Default global quota limits are taking149effect for any and all user/group/project except root the first time the150quota entry for user/group/project id is being accessed - typically the151first time an inode with a particular id ownership is being created after152the mount. In other words, instead of the limits being initialized to zero,153they are initialized with the particular value provided with these mount154options. The limits can be changed for any user/group id at any time as they155normally can be.156 157Note that tmpfs quotas do not support user namespaces so no uid/gid158translation is done if quotas are enabled inside user namespaces.159 160tmpfs has a mount option to set the NUMA memory allocation policy for161all files in that instance (if CONFIG_NUMA is enabled) - which can be162adjusted on the fly via 'mount -o remount ...'163 164======================== ==============================================165mpol=default use the process allocation policy166 (see set_mempolicy(2))167mpol=prefer:Node prefers to allocate memory from the given Node168mpol=bind:NodeList allocates memory only from nodes in NodeList169mpol=interleave prefers to allocate from each node in turn170mpol=interleave:NodeList allocates from each node of NodeList in turn171mpol=local prefers to allocate memory from the local node172======================== ==============================================173 174NodeList format is a comma-separated list of decimal numbers and ranges,175a range being two hyphen-separated decimal numbers, the smallest and176largest node numbers in the range. For example, mpol=bind:0-3,5,7,9-15177 178A memory policy with a valid NodeList will be saved, as specified, for179use at file creation time. When a task allocates a file in the file180system, the mount option memory policy will be applied with a NodeList,181if any, modified by the calling task's cpuset constraints182[See Documentation/admin-guide/cgroup-v1/cpusets.rst] and any optional flags,183listed below. If the resulting NodeLists is the empty set, the effective184memory policy for the file will revert to "default" policy.185 186NUMA memory allocation policies have optional flags that can be used in187conjunction with their modes. These optional flags can be specified188when tmpfs is mounted by appending them to the mode before the NodeList.189See Documentation/admin-guide/mm/numa_memory_policy.rst for a list of190all available memory allocation policy mode flags and their effect on191memory policy.192 193::194 195 =static is equivalent to MPOL_F_STATIC_NODES196 =relative is equivalent to MPOL_F_RELATIVE_NODES197 198For example, mpol=bind=static:NodeList, is the equivalent of an199allocation policy of MPOL_BIND | MPOL_F_STATIC_NODES.200 201Note that trying to mount a tmpfs with an mpol option will fail if the202running kernel does not support NUMA; and will fail if its nodelist203specifies a node which is not online. If your system relies on that204tmpfs being mounted, but from time to time runs a kernel built without205NUMA capability (perhaps a safe recovery kernel), or with fewer nodes206online, then it is advisable to omit the mpol option from automatic207mount options. It can be added later, when the tmpfs is already mounted208on MountPoint, by 'mount -o remount,mpol=Policy:NodeList MountPoint'.209 210 211To specify the initial root directory you can use the following mount212options:213 214==== ==================================215mode The permissions as an octal number216uid The user id217gid The group id218==== ==================================219 220These options do not have any effect on remount. You can change these221parameters with chmod(1), chown(1) and chgrp(1) on a mounted filesystem.222 223 224tmpfs has a mount option to select whether it will wrap at 32- or 64-bit inode225numbers:226 227======= ========================228inode64 Use 64-bit inode numbers229inode32 Use 32-bit inode numbers230======= ========================231 232On a 32-bit kernel, inode32 is implicit, and inode64 is refused at mount time.233On a 64-bit kernel, CONFIG_TMPFS_INODE64 sets the default. inode64 avoids the234possibility of multiple files with the same inode number on a single device;235but risks glibc failing with EOVERFLOW once 33-bit inode numbers are reached -236if a long-lived tmpfs is accessed by 32-bit applications so ancient that237opening a file larger than 2GiB fails with EINVAL.238 239 240So 'mount -t tmpfs -o size=10G,nr_inodes=10k,mode=700 tmpfs /mytmpfs'241will give you tmpfs instance on /mytmpfs which can allocate 10GB242RAM/SWAP in 10240 inodes and it is only accessible by root.243 244 245:Author:246 Christoph Rohland <cr@sap.com>, 1.12.01247:Updated:248 Hugh Dickins, 4 June 2007249:Updated:250 KOSAKI Motohiro, 16 Mar 2010251:Updated:252 Chris Down, 13 July 2020253