]>
Commit | Line | Data |
---|---|---|
7e7cd458 MCC |
1 | .. SPDX-License-Identifier: GPL-2.0 |
2 | ||
3 | ===== | |
4 | Tmpfs | |
5 | ===== | |
6 | ||
1da177e4 LT |
7 | Tmpfs is a file system which keeps all files in virtual memory. |
8 | ||
9 | ||
10 | Everything in tmpfs is temporary in the sense that no files will be | |
11 | created on your hard drive. If you unmount a tmpfs instance, | |
12 | everything stored therein is lost. | |
13 | ||
14 | tmpfs puts everything into the kernel internal caches and grows and | |
15 | shrinks to accommodate the files it contains and is able to swap | |
16 | unneeded pages out to swap space. It has maximum size limits which can | |
17 | be adjusted on the fly via 'mount -o remount ...' | |
18 | ||
19 | If you compare it to ramfs (which was the template to create tmpfs) | |
20 | you gain swapping and limit checking. Another similar thing is the RAM | |
21 | disk (/dev/ram*), which simulates a fixed size hard disk in physical | |
22 | RAM, where you have to create an ordinary filesystem on top. Ramdisks | |
7e7cd458 | 23 | cannot swap and you do not have the possibility to resize them. |
1da177e4 LT |
24 | |
25 | Since tmpfs lives completely in the page cache and on swap, all tmpfs | |
0bc126d4 RF |
26 | pages will be shown as "Shmem" in /proc/meminfo and "Shared" in |
27 | free(1). Notice that these counters also include shared memory | |
28 | (shmem, see ipcs(1)). The most reliable way to get the count is | |
29 | using df(1) and du(1). | |
1da177e4 LT |
30 | |
31 | tmpfs has the following uses: | |
32 | ||
33 | 1) There is always a kernel internal mount which you will not see at | |
34 | all. This is used for shared anonymous mappings and SYSV shared | |
7e7cd458 | 35 | memory. |
1da177e4 LT |
36 | |
37 | This mount does not depend on CONFIG_TMPFS. If CONFIG_TMPFS is not | |
38 | set, the user visible part of tmpfs is not build. But the internal | |
39 | mechanisms are always present. | |
40 | ||
41 | 2) glibc 2.2 and above expects tmpfs to be mounted at /dev/shm for | |
42 | POSIX shared memory (shm_open, shm_unlink). Adding the following | |
7e7cd458 | 43 | line to /etc/fstab should take care of this:: |
1da177e4 LT |
44 | |
45 | tmpfs /dev/shm tmpfs defaults 0 0 | |
46 | ||
47 | Remember to create the directory that you intend to mount tmpfs on | |
bf6ee0ae | 48 | if necessary. |
1da177e4 LT |
49 | |
50 | This mount is _not_ needed for SYSV shared memory. The internal | |
51 | mount is used for that. (In the 2.3 kernel versions it was | |
52 | necessary to mount the predecessor of tmpfs (shm fs) to use SYSV | |
53 | shared memory) | |
54 | ||
55 | 3) Some people (including me) find it very convenient to mount it | |
56 | e.g. on /tmp and /var/tmp and have a big swap partition. And now | |
57 | loop mounts of tmpfs files do work, so mkinitrd shipped by most | |
58 | distributions should succeed with a tmpfs /tmp. | |
59 | ||
60 | 4) And probably a lot more I do not know about :-) | |
61 | ||
62 | ||
63 | tmpfs has three mount options for sizing: | |
64 | ||
7e7cd458 MCC |
65 | ========= ============================================================ |
66 | size The limit of allocated bytes for this tmpfs instance. The | |
1da177e4 LT |
67 | default is half of your physical RAM without swap. If you |
68 | oversize your tmpfs instances the machine will deadlock | |
69 | since the OOM handler will not be able to free that memory. | |
7e7cd458 MCC |
70 | nr_blocks The same as size, but in blocks of PAGE_SIZE. |
71 | nr_inodes The maximum number of inodes for this instance. The default | |
1da177e4 | 72 | is half of the number of your physical RAM pages, or (on a |
670e9f34 | 73 | machine with highmem) the number of lowmem RAM pages, |
1da177e4 | 74 | whichever is the lower. |
7e7cd458 | 75 | ========= ============================================================ |
1da177e4 LT |
76 | |
77 | These parameters accept a suffix k, m or g for kilo, mega and giga and | |
78 | can be changed on remount. The size parameter also accepts a suffix % | |
79 | to limit this tmpfs instance to that percentage of your physical RAM: | |
80 | the default, when neither size nor nr_blocks is specified, is size=50% | |
81 | ||
0edd73b3 HD |
82 | If nr_blocks=0 (or size=0), blocks will not be limited in that instance; |
83 | if nr_inodes=0, inodes will not be limited. It is generally unwise to | |
1da177e4 LT |
84 | mount with such options, since it allows any user with write access to |
85 | use up all the memory on the machine; but enhances the scalability of | |
86 | that instance in a system with many cpus making intensive use of it. | |
87 | ||
88 | ||
7339ff83 | 89 | tmpfs has a mount option to set the NUMA memory allocation policy for |
b00dc3ad HD |
90 | all files in that instance (if CONFIG_NUMA is enabled) - which can be |
91 | adjusted on the fly via 'mount -o remount ...' | |
7339ff83 | 92 | |
7e7cd458 | 93 | ======================== ============================================== |
55741696 KM |
94 | mpol=default use the process allocation policy |
95 | (see set_mempolicy(2)) | |
b00dc3ad HD |
96 | mpol=prefer:Node prefers to allocate memory from the given Node |
97 | mpol=bind:NodeList allocates memory only from nodes in NodeList | |
98 | mpol=interleave prefers to allocate from each node in turn | |
99 | mpol=interleave:NodeList allocates from each node of NodeList in turn | |
55741696 | 100 | mpol=local prefers to allocate memory from the local node |
7e7cd458 | 101 | ======================== ============================================== |
b00dc3ad HD |
102 | |
103 | NodeList format is a comma-separated list of decimal numbers and ranges, | |
104 | a range being two hyphen-separated decimal numbers, the smallest and | |
105 | largest node numbers in the range. For example, mpol=bind:0-3,5,7,9-15 | |
7339ff83 | 106 | |
971ada0f LS |
107 | A memory policy with a valid NodeList will be saved, as specified, for |
108 | use at file creation time. When a task allocates a file in the file | |
109 | system, the mount option memory policy will be applied with a NodeList, | |
110 | if any, modified by the calling task's cpuset constraints | |
7e7cd458 MCC |
111 | [See Documentation/admin-guide/cgroup-v1/cpusets.rst] and any optional flags, |
112 | listed below. If the resulting NodeLists is the empty set, the effective | |
113 | memory policy for the file will revert to "default" policy. | |
971ada0f | 114 | |
65d66fc0 DR |
115 | NUMA memory allocation policies have optional flags that can be used in |
116 | conjunction with their modes. These optional flags can be specified | |
117 | when tmpfs is mounted by appending them to the mode before the NodeList. | |
3ecf53e4 MR |
118 | See Documentation/admin-guide/mm/numa_memory_policy.rst for a list of |
119 | all available memory allocation policy mode flags and their effect on | |
120 | memory policy. | |
65d66fc0 | 121 | |
7e7cd458 MCC |
122 | :: |
123 | ||
65d66fc0 DR |
124 | =static is equivalent to MPOL_F_STATIC_NODES |
125 | =relative is equivalent to MPOL_F_RELATIVE_NODES | |
126 | ||
127 | For example, mpol=bind=static:NodeList, is the equivalent of an | |
128 | allocation policy of MPOL_BIND | MPOL_F_STATIC_NODES. | |
129 | ||
ad329b15 HD |
130 | Note that trying to mount a tmpfs with an mpol option will fail if the |
131 | running kernel does not support NUMA; and will fail if its nodelist | |
a210906c HD |
132 | specifies a node which is not online. If your system relies on that |
133 | tmpfs being mounted, but from time to time runs a kernel built without | |
134 | NUMA capability (perhaps a safe recovery kernel), or with fewer nodes | |
135 | online, then it is advisable to omit the mpol option from automatic | |
ad329b15 HD |
136 | mount options. It can be added later, when the tmpfs is already mounted |
137 | on MountPoint, by 'mount -o remount,mpol=Policy:NodeList MountPoint'. | |
138 | ||
7339ff83 | 139 | |
1da177e4 LT |
140 | To specify the initial root directory you can use the following mount |
141 | options: | |
142 | ||
7e7cd458 MCC |
143 | ==== ================================== |
144 | mode The permissions as an octal number | |
145 | uid The user id | |
146 | gid The group id | |
147 | ==== ================================== | |
1da177e4 LT |
148 | |
149 | These options do not have any effect on remount. You can change these | |
150 | parameters with chmod(1), chown(1) and chgrp(1) on a mounted filesystem. | |
151 | ||
152 | ||
153 | So 'mount -t tmpfs -o size=10G,nr_inodes=10k,mode=700 tmpfs /mytmpfs' | |
154 | will give you tmpfs instance on /mytmpfs which can allocate 10GB | |
155 | RAM/SWAP in 10240 inodes and it is only accessible by root. | |
156 | ||
157 | ||
7e7cd458 | 158 | :Author: |
1da177e4 | 159 | Christoph Rohland <cr@sap.com>, 1.12.01 |
7e7cd458 | 160 | :Updated: |
98f32602 | 161 | Hugh Dickins, 4 June 2007 |
7e7cd458 | 162 | :Updated: |
55741696 | 163 | KOSAKI Motohiro, 16 Mar 2010 |