1 .. SPDX-License-Identifier: GPL-2.0
2
3 =====
4 Tmpfs
5 =====
6
7 Tmpfs is a file system which keeps all of its
8
9
10 Everything in tmpfs is temporary in the sense
11 created on your hard drive. If you unmount a t
12 everything stored therein is lost.
13
14 tmpfs puts everything into the kernel internal
15 shrinks to accommodate the files it contains a
16 unneeded pages out to swap space, if swap was
17 mount. tmpfs also supports THP.
18
19 tmpfs extends ramfs with a few userspace confi
20 explained further below, some of which can be
21 fly using a remount ('mount -o remount ...') o
22 filesystem can be resized but it cannot be res
23 usage. tmpfs also supports POSIX ACLs, and ext
24 trusted.*, security.* and user.* namespaces. r
25 cannot modify any parameter for a ramfs filesy
26 filesystem is how much memory you have availab
27 used so to not run out of memory.
28
29 An alternative to tmpfs and ramfs is to use br
30 (/dev/ram*), which allows you to simulate a bl
31 To write data you would just then need to crea
32 this ramdisk. As with ramfs, brd ramdisks cann
33 configured in size at initialization and you c
34 Contrary to brd ramdisks, tmpfs has its own fi
35 block layer at all.
36
37 Since tmpfs lives completely in the page cache
38 all tmpfs pages will be shown as "Shmem" in /p
39 free(1). Notice that these counters also inclu
40 (shmem, see ipcs(1)). The most reliable way to
41 using df(1) and du(1).
42
43 tmpfs has the following uses:
44
45 1) There is always a kernel internal mount whi
46 all. This is used for shared anonymous mapp
47 memory.
48
49 This mount does not depend on CONFIG_TMPFS.
50 set, the user visible part of tmpfs is not
51 mechanisms are always present.
52
53 2) glibc 2.2 and above expects tmpfs to be mou
54 POSIX shared memory (shm_open, shm_unlink).
55 line to /etc/fstab should take care of this
56
57 tmpfs /dev/shm tmpfs defaul
58
59 Remember to create the directory that you i
60 if necessary.
61
62 This mount is _not_ needed for SYSV shared
63 mount is used for that. (In the 2.3 kernel
64 necessary to mount the predecessor of tmpfs
65 shared memory.)
66
67 3) Some people (including me) find it very con
68 e.g. on /tmp and /var/tmp and have a big sw
69 loop mounts of tmpfs files do work, so mkin
70 distributions should succeed with a tmpfs /
71
72 4) And probably a lot more I do not know about
73
74
75 tmpfs has three mount options for sizing:
76
77 ========= ===================================
78 size The limit of allocated bytes for th
79 default is half of your physical RA
80 oversize your tmpfs instances the m
81 since the OOM handler will not be a
82 nr_blocks The same as size, but in blocks of
83 nr_inodes The maximum number of inodes for th
84 is half of the number of your physi
85 machine with highmem) the number of
86 whichever is the lower.
87 ========= ===================================
88
89 These parameters accept a suffix k, m or g for
90 can be changed on remount. The size parameter
91 to limit this tmpfs instance to that percentag
92 the default, when neither size nor nr_blocks i
93
94 If nr_blocks=0 (or size=0), blocks will not be
95 if nr_inodes=0, inodes will not be limited. I
96 mount with such options, since it allows any u
97 use up all the memory on the machine; but enha
98 that instance in a system with many CPUs makin
99
100 If nr_inodes is not 0, that limited space for
101 extended attributes: "df -i"'s IUsed and IUse%
102
103 tmpfs blocks may be swapped out, when there is
104 tmpfs has a mount option to disable its use of
105
106 ====== ======================================
107 noswap Disables swap. Remounts must respect t
108 By default swap is enabled.
109 ====== ======================================
110
111 tmpfs also supports Transparent Huge Pages whi
112 configured with CONFIG_TRANSPARENT_HUGEPAGE an
113 your system (has_transparent_hugepage(), which
114 The mount options for this are:
115
116 ================ =============================
117 huge=never Do not allocate huge pages.
118 huge=always Attempt to allocate huge page
119 huge=within_size Only allocate huge page if it
120 Also respect madvise(2) hints
121 huge=advise Only allocate huge page if re
122 ================ =============================
123
124 See also Documentation/admin-guide/mm/transhug
125 sysfs file /sys/kernel/mm/transparent_hugepage
126 be used to deny huge pages on all tmpfs mounts
127 force huge pages on all tmpfs mounts for testi
128
129 tmpfs also supports quota with the following m
130
131 ======================== =====================
132 quota User and group quota
133 is enabled on the mou
134 system quota files th
135 usrquota User quota accounting
136 on the mount.
137 grpquota Group quota accountin
138 on the mount.
139 usrquota_block_hardlimit Set global user quota
140 usrquota_inode_hardlimit Set global user quota
141 grpquota_block_hardlimit Set global group quot
142 grpquota_inode_hardlimit Set global group quot
143 ======================== =====================
144
145 None of the quota related mount options can be
146
147 Quota limit parameters accept a suffix k, m or
148 and can't be changed on remount. Default globa
149 effect for any and all user/group/project exce
150 quota entry for user/group/project id is being
151 first time an inode with a particular id owner
152 the mount. In other words, instead of the limi
153 they are initialized with the particular value
154 options. The limits can be changed for any use
155 normally can be.
156
157 Note that tmpfs quotas do not support user nam
158 translation is done if quotas are enabled insi
159
160 tmpfs has a mount option to set the NUMA memor
161 all files in that instance (if CONFIG_NUMA is
162 adjusted on the fly via 'mount -o remount ...'
163
164 ======================== =====================
165 mpol=default use the process alloc
166 (see set_mempolicy(2)
167 mpol=prefer:Node prefers to allocate m
168 mpol=bind:NodeList allocates memory only
169 mpol=interleave prefers to allocate f
170 mpol=interleave:NodeList allocates from each n
171 mpol=local prefers to allocate m
172 ======================== =====================
173
174 NodeList format is a comma-separated list of d
175 a range being two hyphen-separated decimal num
176 largest node numbers in the range. For exampl
177
178 A memory policy with a valid NodeList will be
179 use at file creation time. When a task alloca
180 system, the mount option memory policy will be
181 if any, modified by the calling task's cpuset
182 [See Documentation/admin-guide/cgroup-v1/cpuse
183 listed below. If the resulting NodeLists is t
184 memory policy for the file will revert to "def
185
186 NUMA memory allocation policies have optional
187 conjunction with their modes. These optional
188 when tmpfs is mounted by appending them to the
189 See Documentation/admin-guide/mm/numa_memory_p
190 all available memory allocation policy mode fl
191 memory policy.
192
193 ::
194
195 =static is equivalent to
196 =relative is equivalent to
197
198 For example, mpol=bind=static:NodeList, is the
199 allocation policy of MPOL_BIND | MPOL_F_STATIC
200
201 Note that trying to mount a tmpfs with an mpol
202 running kernel does not support NUMA; and will
203 specifies a node which is not online. If your
204 tmpfs being mounted, but from time to time run
205 NUMA capability (perhaps a safe recovery kerne
206 online, then it is advisable to omit the mpol
207 mount options. It can be added later, when th
208 on MountPoint, by 'mount -o remount,mpol=Polic
209
210
211 To specify the initial root directory you can
212 options:
213
214 ==== ==================================
215 mode The permissions as an octal number
216 uid The user id
217 gid The group id
218 ==== ==================================
219
220 These options do not have any effect on remoun
221 parameters with chmod(1), chown(1) and chgrp(1
222
223
224 tmpfs has a mount option to select whether it
225 numbers:
226
227 ======= ========================
228 inode64 Use 64-bit inode numbers
229 inode32 Use 32-bit inode numbers
230 ======= ========================
231
232 On a 32-bit kernel, inode32 is implicit, and i
233 On a 64-bit kernel, CONFIG_TMPFS_INODE64 sets
234 possibility of multiple files with the same in
235 but risks glibc failing with EOVERFLOW once 33
236 if a long-lived tmpfs is accessed by 32-bit ap
237 opening a file larger than 2GiB fails with EIN
238
239
240 So 'mount -t tmpfs -o size=10G,nr_inodes=10k,m
241 will give you tmpfs instance on /mytmpfs which
242 RAM/SWAP in 10240 inodes and it is only access
243
244
245 :Author:
246 Christoph Rohland <cr@sap.com>, 1.12.01
247 :Updated:
248 Hugh Dickins, 4 June 2007
249 :Updated:
250 KOSAKI Motohiro, 16 Mar 2010
251 :Updated:
252 Chris Down, 13 July 2020
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.