1PROC(5) Linux Programmer's Manual PROC(5)
2
6 proc - process information pseudo-filesystem
7
9 The proc filesystem is a pseudo-filesystem which provides an interface
10 to kernel data structures. It is commonly mounted at /proc. Typi‐
11 cally, it is mounted automatically by the system, but it can also be
12 mounted manually using a command such as:
13 mount -t proc proc /proc
15 Most of the files in the proc filesystem are read-only, but some files
17 are writable, allowing kernel variables to be changed.
18 Mount options
20 The proc filesystem supports the following mount options:
21 hidepid=n (since Linux 3.3)
23 This option controls who can access the information in
24 /proc/[pid] directories. The argument, n, is one of the follow‐
25 ing values:
26 0 Everybody may access all /proc/[pid] directories. This is
28 the traditional behavior, and the default if this mount
29 option is not specified.
30 1 Users may not access files and subdirectories inside any
32 /proc/[pid] directories but their own (the /proc/[pid]
33 directories themselves remain visible). Sensitive files
34 such as /proc/[pid]/cmdline and /proc/[pid]/status are now
35 protected against other users. This makes it impossible to
36 learn whether any user is running a specific program (so
37 long as the program doesn't otherwise reveal itself by its
38 behavior).
39 2 As for mode 1, but in addition the /proc/[pid] directories
41 belonging to other users become invisible. This means that
42 /proc/[pid] entries can no longer be used to discover the
43 PIDs on the system. This doesn't hide the fact that a
44 process with a specific PID value exists (it can be learned
45 by other means, for example, by "kill -0 $PID"), but it
46 hides a process's UID and GID, which could otherwise be
47 learned by employing stat(2) on a /proc/[pid] directory.
48 This greatly complicates an attacker's task of gathering
49 information about running processes (e.g., discovering
50 whether some daemon is running with elevated privileges,
51 whether another user is running some sensitive program,
52 whether other users are running any program at all, and so
53 on).
54 gid=gid (since Linux 3.3)
56 Specifies the ID of a group whose members are authorized to
57 learn process information otherwise prohibited by hidepid (i.e.,
58 users in this group behave as though /proc was mounted with
59 hidepid=0). This group should be used instead of approaches
60 such as putting nonroot users into the sudoers(5) file.
61 Overview
63 Underneath /proc, there are the following general groups of files and
64 subdirectories:
65 /proc/[pid] subdirectories
67 Each one of these subdirectories contains files and subdirecto‐
68 ries exposing information about the process with the correspond‐
69 ing process ID.
70 Underneath each of the /proc/[pid] directories, a task subdirec‐
72 tory contains subdirectories of the form task/[tid], which con‐
73 tain corresponding information about each of the threads in the
74 process, where tid is the kernel thread ID of the thread.
75 The /proc/[pid] subdirectories are visible when iterating
77 through /proc with getdents(2) (and thus are visible when one
78 uses ls(1) to view the contents of /proc).
79 /proc/[tid] subdirectories
81 Each one of these subdirectories contains files and subdirecto‐
82 ries exposing information about the thread with the correspond‐
83 ing thread ID. The contents of these directories are the same
84 as the corresponding /proc/[pid]/task/[tid] directories.
85 The /proc/[tid] subdirectories are not visible when iterating
87 through /proc with getdents(2) (and thus are not visible when
88 one uses ls(1) to view the contents of /proc).
89 /proc/self
91 When a process accesses this magic symbolic link, it resolves to
92 the process's own /proc/[pid] directory.
93 /proc/thread-self
95 When a thread accesses this magic symbolic link, it resolves to
96 the process's own /proc/self/task/[tid] directory.
97 /proc/[a-z]*
99 Various other files and subdirectories under /proc expose sys‐
100 tem-wide information.
101 All of the above are described in more detail below.
103 Files and directories
105 The following list provides details of many of the files and directo‐
106 ries under the /proc hierarchy.
107 /proc/[pid]
109 There is a numerical subdirectory for each running process; the
110 subdirectory is named by the process ID. Each /proc/[pid] sub‐
111 directory contains the pseudo-files and directories described
112 below.
113 The files inside each /proc/[pid] directory are normally owned
115 by the effective user and effective group ID of the process.
116 However, as a security measure, the ownership is made root:root
117 if the process's "dumpable" attribute is set to a value other
118 than 1.
119 Before Linux 4.11, root:root meant the "global" root user ID and
121 group ID (i.e., UID 0 and GID 0 in the initial user namespace).
122 Since Linux 4.11, if the process is in a noninitial user names‐
123 pace that has a valid mapping for user (group) ID 0 inside the
124 namespace, then the user (group) ownership of the files under
125 /proc/[pid] is instead made the same as the root user (group) ID
126 of the namespace. This means that inside a container, things
127 work as expected for the container "root" user.
128 The process's "dumpable" attribute may change for the following
130 reasons:
131 * The attribute was explicitly set via the prctl(2)
133 PR_SET_DUMPABLE operation.
134 * The attribute was reset to the value in the file
136 /proc/sys/fs/suid_dumpable (described below), for the reasons
137 described in prctl(2).
138 Resetting the "dumpable" attribute to 1 reverts the ownership of
140 the /proc/[pid]/* files to the process's effective UID and GID.
141 Note, however, that if the effective UID or GID is subsequently
142 modified, then the "dumpable" attribute may be reset, as
143 described in prctl(2). Therefore, it may be desirable to reset
144 the "dumpable" attribute after making any desired changes to the
145 process's effective UID or GID.
146 /proc/[pid]/attr
148 The files in this directory provide an API for security modules.
149 The contents of this directory are files that can be read and
150 written in order to set security-related attributes. This
151 directory was added to support SELinux, but the intention was
152 that the API be general enough to support other security mod‐
153 ules. For the purpose of explanation, examples of how SELinux
154 uses these files are provided below.
155 This directory is present only if the kernel was configured with
157 CONFIG_SECURITY.
158 /proc/[pid]/attr/current (since Linux 2.6.0)
160 The contents of this file represent the current security
161 attributes of the process.
162 In SELinux, this file is used to get the security context of a
164 process. Prior to Linux 2.6.11, this file could not be used to
165 set the security context (a write was always denied), since
166 SELinux limited process security transitions to execve(2) (see
167 the description of /proc/[pid]/attr/exec, below). Since Linux
168 2.6.11, SELinux lifted this restriction and began supporting
169 "set" operations via writes to this node if authorized by pol‐
170 icy, although use of this operation is only suitable for appli‐
171 cations that are trusted to maintain any desired separation
172 between the old and new security contexts.
173 Prior to Linux 2.6.28, SELinux did not allow threads within a
175 multithreaded process to set their security context via this
176 node as it would yield an inconsistency among the security con‐
177 texts of the threads sharing the same memory space. Since Linux
178 2.6.28, SELinux lifted this restriction and began supporting
179 "set" operations for threads within a multithreaded process if
180 the new security context is bounded by the old security context,
181 where the bounded relation is defined in policy and guarantees
182 that the new security context has a subset of the permissions of
183 the old security context.
184 Other security modules may choose to support "set" operations
186 via writes to this node.
187 /proc/[pid]/attr/exec (since Linux 2.6.0)
189 This file represents the attributes to assign to the process
190 upon a subsequent execve(2).
191 In SELinux, this is needed to support role/domain transitions,
193 and execve(2) is the preferred point to make such transitions
194 because it offers better control over the initialization of the
195 process in the new security label and the inheritance of state.
196 In SELinux, this attribute is reset on execve(2) so that the new
197 program reverts to the default behavior for any execve(2) calls
198 that it may make. In SELinux, a process can set only its own
199 /proc/[pid]/attr/exec attribute.
200 /proc/[pid]/attr/fscreate (since Linux 2.6.0)
202 This file represents the attributes to assign to files created
203 by subsequent calls to open(2), mkdir(2), symlink(2), and
204 mknod(2)
205 SELinux employs this file to support creation of a file (using
207 the aforementioned system calls) in a secure state, so that
208 there is no risk of inappropriate access being obtained between
209 the time of creation and the time that attributes are set. In
210 SELinux, this attribute is reset on execve(2), so that the new
211 program reverts to the default behavior for any file creation
212 calls it may make, but the attribute will persist across multi‐
213 ple file creation calls within a program unless it is explicitly
214 reset. In SELinux, a process can set only its own
215 /proc/[pid]/attr/fscreate attribute.
216 /proc/[pid]/attr/keycreate (since Linux 2.6.18)
218 If a process writes a security context into this file, all sub‐
219 sequently created keys (add_key(2)) will be labeled with this
220 context. For further information, see the kernel source file
221 Documentation/security/keys/core.rst (or file Documenta‐
222 tion/security/keys.txt on Linux between 3.0 and 4.13, or Docu‐
223 mentation/keys.txt before Linux 3.0).
224 /proc/[pid]/attr/prev (since Linux 2.6.0)
226 This file contains the security context of the process before
227 the last execve(2); that is, the previous value of
228 /proc/[pid]/attr/current.
229 /proc/[pid]/attr/socketcreate (since Linux 2.6.18)
231 If a process writes a security context into this file, all sub‐
232 sequently created sockets will be labeled with this context.
233 /proc/[pid]/autogroup (since Linux 2.6.38)
235 See sched(7).
236 /proc/[pid]/auxv (since 2.6.0)
238 This contains the contents of the ELF interpreter information
239 passed to the process at exec time. The format is one unsigned
240 long ID plus one unsigned long value for each entry. The last
241 entry contains two zeros. See also getauxval(3).
242 Permission to access this file is governed by a ptrace access
244 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
245 /proc/[pid]/cgroup (since Linux 2.6.24)
247 See cgroups(7).
248 /proc/[pid]/clear_refs (since Linux 2.6.22)
250 This is a write-only file, writable only by owner of the
252 process.
253 The following values may be written to the file:
255 1 (since Linux 2.6.22)
257 Reset the PG_Referenced and ACCESSED/YOUNG bits for all
258 the pages associated with the process. (Before kernel
259 2.6.32, writing any nonzero value to this file had this
260 effect.)
261 2 (since Linux 2.6.32)
263 Reset the PG_Referenced and ACCESSED/YOUNG bits for all
264 anonymous pages associated with the process.
265 3 (since Linux 2.6.32)
267 Reset the PG_Referenced and ACCESSED/YOUNG bits for all
268 file-mapped pages associated with the process.
269 Clearing the PG_Referenced and ACCESSED/YOUNG bits provides a
271 method to measure approximately how much memory a process is
272 using. One first inspects the values in the "Referenced" fields
273 for the VMAs shown in /proc/[pid]/smaps to get an idea of the
274 memory footprint of the process. One then clears the PG_Refer‐
275 enced and ACCESSED/YOUNG bits and, after some measured time
276 interval, once again inspects the values in the "Referenced"
277 fields to get an idea of the change in memory footprint of the
278 process during the measured interval. If one is interested only
279 in inspecting the selected mapping types, then the value 2 or 3
280 can be used instead of 1.
281 Further values can be written to affect different properties:
283 4 (since Linux 3.11)
285 Clear the soft-dirty bit for all the pages associated
286 with the process. This is used (in conjunction with
287 /proc/[pid]/pagemap) by the check-point restore system to
288 discover which pages of a process have been dirtied since
289 the file /proc/[pid]/clear_refs was written to.
290 5 (since Linux 4.0)
292 Reset the peak resident set size ("high water mark") to
293 the process's current resident set size value.
294 Writing any value to /proc/[pid]/clear_refs other than those
296 listed above has no effect.
297 The /proc/[pid]/clear_refs file is present only if the CON‐
299 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
300 /proc/[pid]/cmdline
302 This read-only file holds the complete command line for the
303 process, unless the process is a zombie. In the latter case,
304 there is nothing in this file: that is, a read on this file will
305 return 0 characters. The command-line arguments appear in this
306 file as a set of strings separated by null bytes ('\0'), with a
307 further null byte after the last string.
308 If, after an execve(2), the process modifies its argv strings,
310 those changes will show up here. This is not the same thing as
311 modifying the argv array.
312 Furthermore, a process may change the memory location that this
314 file refers via prctl(2) operations such as PR_SET_MM_ARG_START.
315 Think of this file as the command line that the process wants
317 you to see.
318 /proc/[pid]/comm (since Linux 2.6.33)
320 This file exposes the process's comm value—that is, the command
321 name associated with the process. Different threads in the same
322 process may have different comm values, accessible via
323 /proc/[pid]/task/[tid]/comm. A thread may modify its comm
324 value, or that of any of other thread in the same thread group
325 (see the discussion of CLONE_THREAD in clone(2)), by writing to
326 the file /proc/self/task/[tid]/comm. Strings longer than
327 TASK_COMM_LEN [22m(16) characters (including the terminating null
328 byte) are silently truncated.
329 This file provides a superset of the prctl(2) PR_SET_NAME and
331 PR_GET_NAME operations, and is employed by pthread_setname_np(3)
332 when used to rename threads other than the caller. The value in
333 this file is used for the %e specifier in /proc/sys/ker‐
334 nel/core_pattern; see proc(5).
335 /proc/[pid]/coredump_filter (since Linux 2.6.23)
337 See core(5).
338 /proc/[pid]/cpuset (since Linux 2.6.12)
340 See cpuset(7).
341 /proc/[pid]/cwd
343 This is a symbolic link to the current working directory of the
344 process. To find out the current working directory of process
345 20, for instance, you can do this:
346 $ cd /proc/20/cwd; /bin/pwd
348 Note that the pwd command is often a shell built-in, and might
350 not work properly. In bash(1), you may use pwd -P.
351 In a multithreaded process, the contents of this symbolic link
353 are not available if the main thread has already terminated
354 (typically by calling pthread_exit(3)).
355 Permission to dereference or read (readlink(2)) this symbolic
357 link is governed by a ptrace access mode
358 PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
359 /proc/[pid]/environ
361 This file contains the initial environment that was set when the
362 currently executing program was started via execve(2). The
363 entries are separated by null bytes ('\0'), and there may be a
364 null byte at the end. Thus, to print out the environment of
365 process 1, you would do:
366 $ cat /proc/1/environ | tr '\000' '\n'
368 If, after an execve(2), the process modifies its environment
370 (e.g., by calling functions such as putenv(3) or modifying the
371 environ(7) variable directly), this file will not reflect those
372 changes.
373 Furthermore, a process may change the memory location that this
375 file refers via prctl(2) operations such as PR_SET_MM_ENV_START.
376 Permission to access this file is governed by a ptrace access
378 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
379 /proc/[pid]/exe
381 Under Linux 2.2 and later, this file is a symbolic link contain‐
382 ing the actual pathname of the executed command. This symbolic
383 link can be dereferenced normally; attempting to open it will
384 open the executable. You can even type /proc/[pid]/exe to run
385 another copy of the same executable that is being run by process
386 [pid]. If the pathname has been unlinked, the symbolic link
387 will contain the string '(deleted)' appended to the original
388 pathname. In a multithreaded process, the contents of this sym‐
389 bolic link are not available if the main thread has already ter‐
390 minated (typically by calling pthread_exit(3)).
391 Permission to dereference or read (readlink(2)) this symbolic
393 link is governed by a ptrace access mode
394 PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
395 Under Linux 2.0 and earlier, /proc/[pid]/exe is a pointer to the
397 binary which was executed, and appears as a symbolic link. A
398 readlink(2) call on this file under Linux 2.0 returns a string
399 in the format:
400 [device]:inode
402 For example, [0301]:1502 would be inode 1502 on device major 03
404 (IDE, MFM, etc. drives) minor 01 (first partition on the first
405 drive).
406 find(1) with the -inum option can be used to locate the file.
408 /proc/[pid]/fd/
410 This is a subdirectory containing one entry for each file which
411 the process has open, named by its file descriptor, and which is
412 a symbolic link to the actual file. Thus, 0 is standard input,
413 1 standard output, 2 standard error, and so on.
414 For file descriptors for pipes and sockets, the entries will be
416 symbolic links whose content is the file type with the inode. A
417 readlink(2) call on this file returns a string in the format:
418 type:[inode]
420 For example, socket:[2248868] will be a socket and its inode is
422 2248868. For sockets, that inode can be used to find more
423 information in one of the files under /proc/net/.
424 For file descriptors that have no corresponding inode (e.g.,
426 file descriptors produced by bpf(2), epoll_create(2),
427 eventfd(2), inotify_init(2), perf_event_open(2), signalfd(2),
428 timerfd_create(2), and userfaultfd(2)), the entry will be a sym‐
429 bolic link with contents of the form
430 anon_inode:<file-type>
432 In many cases (but not all), the file-type is surrounded by
434 square brackets.
435 For example, an epoll file descriptor will have a symbolic link
437 whose content is the string anon_inode:[eventpoll].
438 In a multithreaded process, the contents of this directory are
440 not available if the main thread has already terminated (typi‐
441 cally by calling pthread_exit(3)).
442 Programs that take a filename as a command-line argument, but
444 don't take input from standard input if no argument is supplied,
445 and programs that write to a file named as a command-line argu‐
446 ment, but don't send their output to standard output if no argu‐
447 ment is supplied, can nevertheless be made to use standard input
448 or standard output by using /proc/[pid]/fd files as command-line
449 arguments. For example, assuming that -i is the flag designat‐
450 ing an input file and -o is the flag designating an output file:
451 $ foobar -i /proc/self/fd/0 -o /proc/self/fd/1 ...
453 and you have a working filter.
455 /proc/self/fd/N is approximately the same as /dev/fd/N in some
457 UNIX and UNIX-like systems. Most Linux MAKEDEV scripts symboli‐
458 cally link /dev/fd to /proc/self/fd, in fact.
459 Most systems provide symbolic links /dev/stdin, /dev/stdout, and
461 /dev/stderr, which respectively link to the files 0, 1, and 2 in
462 /proc/self/fd. Thus the example command above could be written
463 as:
464 $ foobar -i /dev/stdin -o /dev/stdout ...
466 Permission to dereference or read (readlink(2)) the symbolic
468 links in this directory is governed by a ptrace access mode
469 PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
470 Note that for file descriptors referring to inodes (pipes and
472 sockets, see above), those inodes still have permission bits and
473 ownership information distinct from those of the /proc/[pid]/fd
474 entry, and that the owner may differ from the user and group IDs
475 of the process. An unprivileged process may lack permissions to
476 open them, as in this example:
477 $ echo test | sudo -u nobody cat
479 test
480 $ echo test | sudo -u nobody cat /proc/self/fd/0
481 cat: /proc/self/fd/0: Permission denied
482 File descriptor 0 refers to the pipe created by the shell and
484 owned by that shell's user, which is not nobody, so cat does not
485 have permission to create a new file descriptor to read from
486 that inode, even though it can still read from its existing file
487 descriptor 0.
488 /proc/[pid]/fdinfo/ (since Linux 2.6.22)
490 This is a subdirectory containing one entry for each file which
491 the process has open, named by its file descriptor. The files
492 in this directory are readable only by the owner of the process.
493 The contents of each file can be read to obtain information
494 about the corresponding file descriptor. The content depends on
495 the type of file referred to by the corresponding file descrip‐
496 tor.
497 For regular files and directories, we see something like:
499 $ cat /proc/12015/fdinfo/4
501 pos: 1000
502 flags: 01002002
503 mnt_id: 21
504 The fields are as follows:
506 pos This is a decimal number showing the file offset.
508 flags This is an octal number that displays the file access
510 mode and file status flags (see open(2)). If the close-
511 on-exec file descriptor flag is set, then flags will also
512 include the value O_CLOEXEC.
513 Before Linux 3.1, this field incorrectly displayed the
515 setting of O_CLOEXEC at the time the file was opened,
516 rather than the current setting of the close-on-exec
517 flag.
518 mnt_id This field, present since Linux 3.15, is the ID of the
520 mount point containing this file. See the description of
521 /proc/[pid]/mountinfo.
522 For eventfd file descriptors (see eventfd(2)), we see (since
524 Linux 3.8) the following fields:
525 pos: 0
527 flags: 02
528 mnt_id: 10
529 eventfd-count: 40
530 eventfd-count is the current value of the eventfd counter, in
532 hexadecimal.
533 For epoll file descriptors (see epoll(7)), we see (since Linux
535 3.8) the following fields:
536 pos: 0
538 flags: 02
539 mnt_id: 10
540 tfd: 9 events: 19 data: 74253d2500000009
541 tfd: 7 events: 19 data: 74253d2500000007
542 Each of the lines beginning tfd describes one of the file
544 descriptors being monitored via the epoll file descriptor (see
545 epoll_ctl(2) for some details). The tfd field is the number of
546 the file descriptor. The events field is a hexadecimal mask of
547 the events being monitored for this file descriptor. The data
548 field is the data value associated with this file descriptor.
549 For signalfd file descriptors (see signalfd(2)), we see (since
551 Linux 3.8) the following fields:
552 pos: 0
554 flags: 02
555 mnt_id: 10
556 sigmask: 0000000000000006
557 sigmask is the hexadecimal mask of signals that are accepted via
559 this signalfd file descriptor. (In this example, bits 2 and 3
560 are set, corresponding to the signals SIGINT and SIGQUIT; see
561 signal(7).)
562 For inotify file descriptors (see inotify(7)), we see (since
564 Linux 3.8) the following fields:
565 pos: 0
567 flags: 00
568 mnt_id: 11
569 inotify wd:2 ino:7ef82a sdev:800001 mask:800afff ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:2af87e00220ffd73
570 inotify wd:1 ino:192627 sdev:800001 mask:800afff ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:27261900802dfd73
571 Each of the lines beginning with "inotify" displays information
573 about one file or directory that is being monitored. The fields
574 in this line are as follows:
575 wd A watch descriptor number (in decimal).
577 ino The inode number of the target file (in hexadecimal).
579 sdev The ID of the device where the target file resides (in
581 hexadecimal).
582 mask The mask of events being monitored for the target file
584 (in hexadecimal).
585 If the kernel was built with exportfs support, the path to the
587 target file is exposed as a file handle, via three hexadecimal
588 fields: fhandle-bytes, fhandle-type, and f_handle.
589 For fanotify file descriptors (see fanotify(7)), we see (since
591 Linux 3.8) the following fields:
592 pos: 0
594 flags: 02
595 mnt_id: 11
596 fanotify flags:0 event-flags:88002
597 fanotify ino:19264f sdev:800001 mflags:0 mask:1 ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:4f261900a82dfd73
598 The fourth line displays information defined when the fanotify
600 group was created via fanotify_init(2):
601 flags The flags argument given to fanotify_init(2) (expressed
603 in hexadecimal).
604 event-flags
606 The event_f_flags argument given to fanotify_init(2)
607 (expressed in hexadecimal).
608 Each additional line shown in the file contains information
610 about one of the marks in the fanotify group. Most of these
611 fields are as for inotify, except:
612 mflags The flags associated with the mark (expressed in hexadec‐
614 imal).
615 mask The events mask for this mark (expressed in hexadecimal).
617 ignored_mask
619 The mask of events that are ignored for this mark
620 (expressed in hexadecimal).
621 For details on these fields, see fanotify_mark(2).
623 For timerfd file descriptors (see timerfd(2)), we see (since
625 Linux 3.17) the following fields:
626 pos: 0
628 flags: 02004002
629 mnt_id: 13
630 clockid: 0
631 ticks: 0
632 settime flags: 03
633 it_value: (7695568592, 640020877)
634 it_interval: (0, 0)
635 clockid
637 This is the numeric value of the clock ID (corresponding
638 to one of the CLOCK_* constants defined via <time.h>)
639 that is used to mark the progress of the timer (in this
640 example, 0 is CLOCK_REALTIME).
641 ticks This is the number of timer expirations that have
643 occurred, (i.e., the value that read(2) on it would
644 return).
645 settime flags
647 This field lists the flags with which the timerfd was
648 last armed (see timerfd_settime(2)), in octal (in this
649 example, both TFD_TIMER_ABSTIME and TFD_TIMER_CAN‐
650 CEL_ON_SET are set).
651 it_value
653 This field contains the amount of time until the timer
654 will next expire, expressed in seconds and nanoseconds.
655 This is always expressed as a relative value, regardless
656 of whether the timer was created using the
657 TFD_TIMER_ABSTIME flag.
658 it_interval
660 This field contains the interval of the timer, in seconds
661 and nanoseconds. (The it_value and it_interval fields
662 contain the values that timerfd_gettime(2) on this file
663 descriptor would return.)
664 /proc/[pid]/gid_map (since Linux 3.5)
666 See user_namespaces(7).
667 /proc/[pid]/io (since kernel 2.6.20)
669 This file contains I/O statistics for the process, for example:
670 # cat /proc/3828/io
672 rchar: 323934931
673 wchar: 323929600
674 syscr: 632687
675 syscw: 632675
676 read_bytes: 0
677 write_bytes: 323932160
678 cancelled_write_bytes: 0
679 The fields are as follows:
681 rchar: characters read
683 The number of bytes which this task has caused to be read
684 from storage. This is simply the sum of bytes which this
685 process passed to read(2) and similar system calls. It
686 includes things such as terminal I/O and is unaffected by
687 whether or not actual physical disk I/O was required (the
688 read might have been satisfied from pagecache).
689 wchar: characters written
691 The number of bytes which this task has caused, or shall
692 cause to be written to disk. Similar caveats apply here
693 as with rchar.
694 syscr: read syscalls
696 Attempt to count the number of read I/O operations—that
697 is, system calls such as read(2) and pread(2).
698 syscw: write syscalls
700 Attempt to count the number of write I/O operations—that
701 is, system calls such as write(2) and pwrite(2).
702 read_bytes: bytes read
704 Attempt to count the number of bytes which this process
705 really did cause to be fetched from the storage layer.
706 This is accurate for block-backed filesystems.
707 write_bytes: bytes written
709 Attempt to count the number of bytes which this process
710 caused to be sent to the storage layer.
711 cancelled_write_bytes:
713 The big inaccuracy here is truncate. If a process writes
714 1 MB to a file and then deletes the file, it will in fact
715 perform no writeout. But it will have been accounted as
716 having caused 1 MB of write. In other words: this field
717 represents the number of bytes which this process caused
718 to not happen, by truncating pagecache. A task can cause
719 "negative" I/O too. If this task truncates some dirty
720 pagecache, some I/O which another task has been accounted
721 for (in its write_bytes) will not be happening.
722 Note: In the current implementation, things are a bit racy on
724 32-bit systems: if process A reads process B's /proc/[pid]/io
725 while process B is updating one of these 64-bit counters,
726 process A could see an intermediate result.
727 Permission to access this file is governed by a ptrace access
729 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
730 /proc/[pid]/limits (since Linux 2.6.24)
732 This file displays the soft limit, hard limit, and units of mea‐
733 surement for each of the process's resource limits (see getr‐
734 limit(2)). Up to and including Linux 2.6.35, this file is pro‐
735 tected to allow reading only by the real UID of the process.
736 Since Linux 2.6.36, this file is readable by all users on the
737 system.
738 /proc/[pid]/map_files/ (since kernel 3.3)
740 This subdirectory contains entries corresponding to memory-
741 mapped files (see mmap(2)). Entries are named by memory region
742 start and end address pair (expressed as hexadecimal numbers),
743 and are symbolic links to the mapped files themselves. Here is
744 an example, with the output wrapped and reformatted to fit on an
745 80-column display:
746 # ls -l /proc/self/map_files/
748 lr--------. 1 root root 64 Apr 16 21:31
749 3252e00000-3252e20000 -> /usr/lib64/ld-2.15.so
750 ...
751 Although these entries are present for memory regions that were
753 mapped with the MAP_FILE flag, the way anonymous shared memory
754 (regions created with the MAP_ANON | MAP_SHARED flags) is imple‐
755 mented in Linux means that such regions also appear on this
756 directory. Here is an example where the target file is the
757 deleted /dev/zero one:
758 lrw-------. 1 root root 64 Apr 16 21:33
760 7fc075d2f000-7fc075e6f000 -> /dev/zero (deleted)
761 Permission to access this file is governed by a ptrace access
763 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
764 Until kernel version 4.3, this directory appeared only if the
766 CONFIG_CHECKPOINT_RESTORE kernel configuration option was
767 enabled. Additionally, in those kernel versions, privilege
768 (CAP_SYS_ADMIN) was required to view the contents of this direc‐
769 tory.
770 /proc/[pid]/maps
772 A file containing the currently mapped memory regions and their
773 access permissions. See mmap(2) for some further information
774 about memory mappings.
775 Permission to access this file is governed by a ptrace access
777 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
778 The format of the file is:
780 address perms offset dev inode pathname
782 00400000-00452000 r-xp 00000000 08:02 173521 /usr/bin/dbus-daemon
783 00651000-00652000 r--p 00051000 08:02 173521 /usr/bin/dbus-daemon
784 00652000-00655000 rw-p 00052000 08:02 173521 /usr/bin/dbus-daemon
785 00e03000-00e24000 rw-p 00000000 00:00 0 [heap]
786 00e24000-011f7000 rw-p 00000000 00:00 0 [heap]
787 ...
788 35b1800000-35b1820000 r-xp 00000000 08:02 135522 /usr/lib64/ld-2.15.so
789 35b1a1f000-35b1a20000 r--p 0001f000 08:02 135522 /usr/lib64/ld-2.15.so
790 35b1a20000-35b1a21000 rw-p 00020000 08:02 135522 /usr/lib64/ld-2.15.so
791 35b1a21000-35b1a22000 rw-p 00000000 00:00 0
792 35b1c00000-35b1dac000 r-xp 00000000 08:02 135870 /usr/lib64/libc-2.15.so
793 35b1dac000-35b1fac000 ---p 001ac000 08:02 135870 /usr/lib64/libc-2.15.so
794 35b1fac000-35b1fb0000 r--p 001ac000 08:02 135870 /usr/lib64/libc-2.15.so
795 35b1fb0000-35b1fb2000 rw-p 001b0000 08:02 135870 /usr/lib64/libc-2.15.so
796 ...
797 f2c6ff8c000-7f2c7078c000 rw-p 00000000 00:00 0 [stack:986]
798 ...
799 7fffb2c0d000-7fffb2c2e000 rw-p 00000000 00:00 0 [stack]
800 7fffb2d48000-7fffb2d49000 r-xp 00000000 00:00 0 [vdso]
801 The address field is the address space in the process that the
803 mapping occupies. The perms field is a set of permissions:
804 r = read
806 w = write
807 x = execute
808 s = shared
809 p = private (copy on write)
810 The offset field is the offset into the file/whatever; dev is
812 the device (major:minor); inode is the inode on that device. 0
813 indicates that no inode is associated with the memory region, as
814 would be the case with BSS (uninitialized data).
815 The pathname field will usually be the file that is backing the
817 mapping. For ELF files, you can easily coordinate with the off‐
818 set field by looking at the Offset field in the ELF program
819 headers (readelf -l).
820 There are additional helpful pseudo-paths:
822 [stack]
824 The initial process's (also known as the main thread's)
825 stack.
826 [stack:<tid>] (from Linux 3.4 to 4.4)
828 A thread's stack (where the <tid> is a thread ID). It
829 corresponds to the /proc/[pid]/task/[tid]/ path. This
830 field was removed in Linux 4.5, since providing this
831 information for a process with large numbers of threads
832 is expensive.
833 [vdso] The virtual dynamically linked shared object. See
835 vdso(7).
836 [heap] The process's heap.
838 If the pathname field is blank, this is an anonymous mapping as
840 obtained via mmap(2). There is no easy way to coordinate this
841 back to a process's source, short of running it through gdb(1),
842 strace(1), or similar.
843 pathname is shown unescaped except for newline characters, which
845 are replaced with an octal escape sequence. As a result, it is
846 not possible to determine whether the original pathname con‐
847 tained a newline character or the literal \e012 character
848 sequence.
849 If the mapping is file-backed and the file has been deleted, the
851 string " (deleted)" is appended to the pathname. Note that this
852 is ambiguous too.
853 Under Linux 2.0, there is no field giving pathname.
855 /proc/[pid]/mem
857 This file can be used to access the pages of a process's memory
858 through open(2), read(2), and lseek(2).
859 Permission to access this file is governed by a ptrace access
861 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
862 /proc/[pid]/mountinfo (since Linux 2.6.26)
864 This file contains information about mount points in the
865 process's mount namespace (see mount_namespaces(7)). It sup‐
866 plies various information (e.g., propagation state, root of
867 mount for bind mounts, identifier for each mount and its parent)
868 that is missing from the (older) /proc/[pid]/mounts file, and
869 fixes various other problems with that file (e.g., nonextensi‐
870 bility, failure to distinguish per-mount versus per-superblock
871 options).
872 The file contains lines of the form:
87436 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
876(1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
877 The numbers in parentheses are labels for the descriptions
879 below:
880 (1) mount ID: a unique ID for the mount (may be reused after
882 umount(2)).
883 (2) parent ID: the ID of the parent mount (or of self for the
885 root of this mount namespace's mount tree).
886 If a new mount is stacked on top of a previous existing
888 mount (so that it hides the existing mount) at pathname P,
889 then the parent of the new mount is the previous mount at
890 that location. Thus, when looking at all the mounts
891 stacked at a particular location, the top-most mount is the
892 one that is not the parent of any other mount at the same
893 location. (Note, however, that this top-most mount will be
894 accessible only if the longest path subprefix of P that is
895 a mount point is not itself hidden by a stacked mount.)
896 If the parent mount point lies outside the process's root
898 directory (see chroot(2)), the ID shown here won't have a
899 corresponding record in mountinfo whose mount ID (field 1)
900 matches this parent mount ID (because mount points that lie
901 outside the process's root directory are not shown in
902 mountinfo). As a special case of this point, the process's
903 root mount point may have a parent mount (for the initramfs
904 filesystem) that lies outside the process's root directory,
905 and an entry for that mount point will not appear in
906 mountinfo.
907 (3) major:minor: the value of st_dev for files on this filesys‐
909 tem (see stat(2)).
910 (4) root: the pathname of the directory in the filesystem which
912 forms the root of this mount.
913 (5) mount point: the pathname of the mount point relative to
915 the process's root directory.
916 (6) mount options: per-mount options (see mount(2)).
918 (7) optional fields: zero or more fields of the form
920 "tag[:value]"; see below.
921 (8) separator: the end of the optional fields is marked by a
923 single hyphen.
924 (9) filesystem type: the filesystem type in the form
926 "type[.subtype]".
927 (10) mount source: filesystem-specific information or "none".
929 (11) super options: per-superblock options (see mount(2)).
931 Currently, the possible optional fields are shared, master,
933 propagate_from, and unbindable. See mount_namespaces(7) for a
934 description of these fields. Parsers should ignore all unrecog‐
935 nized optional fields.
936 For more information on mount propagation see: Documenta‐
938 tion/filesystems/sharedsubtree.txt in the Linux kernel source
939 tree.
940 /proc/[pid]/mounts (since Linux 2.4.19)
942 This file lists all the filesystems currently mounted in the
943 process's mount namespace (see mount_namespaces(7)). The format
944 of this file is documented in fstab(5).
945 Since kernel version 2.6.15, this file is pollable: after open‐
947 ing the file for reading, a change in this file (i.e., a
948 filesystem mount or unmount) causes select(2) to mark the file
949 descriptor as having an exceptional condition, and poll(2) and
950 epoll_wait(2) mark the file as having a priority event (POLL‐
951 PRI). (Before Linux 2.6.30, a change in this file was indicated
952 by the file descriptor being marked as readable for select(2),
953 and being marked as having an error condition for poll(2) and
954 epoll_wait(2).)
955 /proc/[pid]/mountstats (since Linux 2.6.17)
957 This file exports information (statistics, configuration infor‐
958 mation) about the mount points in the process's mount namespace
959 (see mount_namespaces(7)). Lines in this file have the form:
960 device /dev/sda7 mounted on /home with fstype ext3 [stats]
962 ( 1 ) ( 2 ) (3 ) ( 4 )
963 The fields in each line are:
965 (1) The name of the mounted device (or "nodevice" if there is
967 no corresponding device).
968 (2) The mount point within the filesystem tree.
970 (3) The filesystem type.
972 (4) Optional statistics and configuration information. Cur‐
974 rently (as at Linux 2.6.26), only NFS filesystems export
975 information via this field.
976 This file is readable only by the owner of the process.
978 /proc/[pid]/net (since Linux 2.6.25)
980 See the description of /proc/net.
981 /proc/[pid]/ns/ (since Linux 3.0)
983 This is a subdirectory containing one entry for each namespace
984 that supports being manipulated by setns(2). For more informa‐
985 tion, see namespaces(7).
986 /proc/[pid]/numa_maps (since Linux 2.6.14)
988 See numa(7).
989 /proc/[pid]/oom_adj (since Linux 2.6.11)
991 This file can be used to adjust the score used to select which
992 process should be killed in an out-of-memory (OOM) situation.
993 The kernel uses this value for a bit-shift operation of the
994 process's oom_score value: valid values are in the range -16 to
995 +15, plus the special value -17, which disables OOM-killing
996 altogether for this process. A positive score increases the
997 likelihood of this process being killed by the OOM-killer; a
998 negative score decreases the likelihood.
999 The default value for this file is 0; a new process inherits its
1001 parent's oom_adj setting. A process must be privileged
1002 (CAP_SYS_RESOURCE) to update this file.
1003 Since Linux 2.6.36, use of this file is deprecated in favor of
1005 /proc/[pid]/oom_score_adj.
1006 /proc/[pid]/oom_score (since Linux 2.6.11)
1008 This file displays the current score that the kernel gives to
1009 this process for the purpose of selecting a process for the OOM-
1010 killer. A higher score means that the process is more likely to
1011 be selected by the OOM-killer. The basis for this score is the
1012 amount of memory used by the process, with increases (+) or
1013 decreases (-) for factors including:
1014 * whether the process is privileged (-).
1016 Before kernel 2.6.36 the following factors were also used in the
1018 calculation of oom_score:
1019 * whether the process creates a lot of children using fork(2)
1021 (+);
1022 * whether the process has been running a long time, or has used
1024 a lot of CPU time (-);
1025 * whether the process has a low nice value (i.e., > 0) (+); and
1027 * whether the process is making direct hardware access (-).
1029 The oom_score also reflects the adjustment specified by the
1031 oom_score_adj or oom_adj setting for the process.
1032 /proc/[pid]/oom_score_adj (since Linux 2.6.36)
1034 This file can be used to adjust the badness heuristic used to
1035 select which process gets killed in out-of-memory conditions.
1036 The badness heuristic assigns a value to each candidate task
1038 ranging from 0 (never kill) to 1000 (always kill) to determine
1039 which process is targeted. The units are roughly a proportion
1040 along that range of allowed memory the process may allocate
1041 from, based on an estimation of its current memory and swap use.
1042 For example, if a task is using all allowed memory, its badness
1043 score will be 1000. If it is using half of its allowed memory,
1044 its score will be 500.
1045 There is an additional factor included in the badness score:
1047 root processes are given 3% extra memory over other tasks.
1048 The amount of "allowed" memory depends on the context in which
1050 the OOM-killer was called. If it is due to the memory assigned
1051 to the allocating task's cpuset being exhausted, the allowed
1052 memory represents the set of mems assigned to that cpuset (see
1053 cpuset(7)). If it is due to a mempolicy's node(s) being
1054 exhausted, the allowed memory represents the set of mempolicy
1055 nodes. If it is due to a memory limit (or swap limit) being
1056 reached, the allowed memory is that configured limit. Finally,
1057 if it is due to the entire system being out of memory, the
1058 allowed memory represents all allocatable resources.
1059 The value of oom_score_adj is added to the badness score before
1061 it is used to determine which task to kill. Acceptable values
1062 range from -1000 (OOM_SCORE_ADJ_MIN) to +1000
1063 (OOM_SCORE_ADJ_MAX). This allows user space to control the
1064 preference for OOM-killing, ranging from always preferring a
1065 certain task or completely disabling it from OOM killing. The
1066 lowest possible value, -1000, is equivalent to disabling OOM-
1067 killing entirely for that task, since it will always report a
1068 badness score of 0.
1069 Consequently, it is very simple for user space to define the
1071 amount of memory to consider for each task. Setting an
1072 oom_score_adj value of +500, for example, is roughly equivalent
1073 to allowing the remainder of tasks sharing the same system,
1074 cpuset, mempolicy, or memory controller resources to use at
1075 least 50% more memory. A value of -500, on the other hand,
1076 would be roughly equivalent to discounting 50% of the task's
1077 allowed memory from being considered as scoring against the
1078 task.
1079 For backward compatibility with previous kernels,
1081 /proc/[pid]/oom_adj can still be used to tune the badness score.
1082 Its value is scaled linearly with oom_score_adj.
1083 Writing to /proc/[pid]/oom_score_adj or /proc/[pid]/oom_adj will
1085 change the other with its scaled value.
1086 The choom(1) program provides a command-line interface for
1088 adjusting the oom_score_adj value of a running process or a
1089 newly executed command.
1090 /proc/[pid]/pagemap (since Linux 2.6.25)
1092 This file shows the mapping of each of the process's virtual
1093 pages into physical page frames or swap area. It contains one
1094 64-bit value for each virtual page, with the bits set as fol‐
1095 lows:
1096 63 If set, the page is present in RAM.
1098 62 If set, the page is in swap space
1100 61 (since Linux 3.5)
1102 The page is a file-mapped page or a shared anonymous
1103 page.
1104 60–57 (since Linux 3.11)
1106 Zero
1107 56 (since Linux 4.2)
1109 The page is exclusively mapped.
1110 55 (since Linux 3.11)
1112 PTE is soft-dirty (see the kernel source file Documenta‐
1113 tion/admin-guide/mm/soft-dirty.rst).
1114 54–0 If the page is present in RAM (bit 63), then these bits
1116 provide the page frame number, which can be used to index
1117 /proc/kpageflags and /proc/kpagecount. If the page is
1118 present in swap (bit 62), then bits 4–0 give the swap
1119 type, and bits 54–5 encode the swap offset.
1120 Before Linux 3.11, bits 60–55 were used to encode the base-2 log
1122 of the page size.
1123 To employ /proc/[pid]/pagemap efficiently, use /proc/[pid]/maps
1125 to determine which areas of memory are actually mapped and seek
1126 to skip over unmapped regions.
1127 The /proc/[pid]/pagemap file is present only if the CON‐
1129 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
1130 Permission to access this file is governed by a ptrace access
1132 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1133 /proc/[pid]/personality (since Linux 2.6.28)
1135 This read-only file exposes the process's execution domain, as
1136 set by personality(2). The value is displayed in hexadecimal
1137 notation.
1138 Permission to access this file is governed by a ptrace access
1140 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1141 /proc/[pid]/root
1143 UNIX and Linux support the idea of a per-process root of the
1144 filesystem, set by the chroot(2) system call. This file is a
1145 symbolic link that points to the process's root directory, and
1146 behaves in the same way as exe, and fd/*.
1147 Note however that this file is not merely a symbolic link. It
1149 provides the same view of the filesystem (including namespaces
1150 and the set of per-process mounts) as the process itself. An
1151 example illustrates this point. In one terminal, we start a
1152 shell in new user and mount namespaces, and in that shell we
1153 create some new mount points:
1154 $ PS1='sh1# ' unshare -Urnm
1156 sh1# mount -t tmpfs tmpfs /etc # Mount empty tmpfs at /etc
1157 sh1# mount --bind /usr /dev # Mount /usr at /dev
1158 sh1# echo $$
1159 27123
1160 In a second terminal window, in the initial mount namespace, we
1162 look at the contents of the corresponding mounts in the initial
1163 and new namespaces:
1164 $ PS1='sh2# ' sudo sh
1166 sh2# ls /etc | wc -l # In initial NS
1167 309
1168 sh2# ls /proc/27123/root/etc | wc -l # /etc in other NS
1169 0 # The empty tmpfs dir
1170 sh2# ls /dev | wc -l # In initial NS
1171 205
1172 sh2# ls /proc/27123/root/dev | wc -l # /dev in other NS
1173 11 # Actually bind
1174 # mounted to /usr
1175 sh2# ls /usr | wc -l # /usr in initial NS
1176 11
1177 In a multithreaded process, the contents of the /proc/[pid]/root
1179 symbolic link are not available if the main thread has already
1180 terminated (typically by calling pthread_exit(3)).
1181 Permission to dereference or read (readlink(2)) this symbolic
1183 link is governed by a ptrace access mode
1184 PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1185 /proc/[pid]/seccomp (Linux 2.6.12 to 2.6.22)
1187 This file can be used to read and change the process's secure
1188 computing (seccomp) mode setting. It contains the value 0 if
1189 the process is not in seccomp mode, and 1 if the process is in
1190 strict seccomp mode (see seccomp(2)). Writing 1 to this file
1191 places the process irreversibly in strict seccomp mode. (Fur‐
1192 ther attempts to write to the file fail with the EPERM error.)
1193 In Linux 2.6.23, this file went away, to be replaced by the
1195 prctl(2) PR_GET_SECCOMP and PR_SET_SECCOMP operations (and later
1196 by seccomp(2) and the Seccomp field in /proc/[pid]/status).
1197 /proc/[pid]/setgroups (since Linux 3.19)
1199 See user_namespaces(7).
1200 /proc/[pid]/smaps (since Linux 2.6.14)
1202 This file shows memory consumption for each of the process's
1203 mappings. (The pmap(1) command displays similar information, in
1204 a form that may be easier for parsing.) For each mapping there
1205 is a series of lines such as the following:
1206 00400000-0048a000 r-xp 00000000 fd:03 960637 /bin/bash
1208 Size: 552 kB
1209 Rss: 460 kB
1210 Pss: 100 kB
1211 Shared_Clean: 452 kB
1212 Shared_Dirty: 0 kB
1213 Private_Clean: 8 kB
1214 Private_Dirty: 0 kB
1215 Referenced: 460 kB
1216 Anonymous: 0 kB
1217 AnonHugePages: 0 kB
1218 ShmemHugePages: 0 kB
1219 ShmemPmdMapped: 0 kB
1220 Swap: 0 kB
1221 KernelPageSize: 4 kB
1222 MMUPageSize: 4 kB
1223 KernelPageSize: 4 kB
1224 MMUPageSize: 4 kB
1225 Locked: 0 kB
1226 ProtectionKey: 0
1227 VmFlags: rd ex mr mw me dw
1228 The first of these lines shows the same information as is dis‐
1230 played for the mapping in /proc/[pid]/maps. The following lines
1231 show the size of the mapping, the amount of the mapping that is
1232 currently resident in RAM ("Rss"), the process's proportional
1233 share of this mapping ("Pss"), the number of clean and dirty
1234 shared pages in the mapping, and the number of clean and dirty
1235 private pages in the mapping. "Referenced" indicates the amount
1236 of memory currently marked as referenced or accessed. "Anony‐
1237 mous" shows the amount of memory that does not belong to any
1238 file. "Swap" shows how much would-be-anonymous memory is also
1239 used, but out on swap.
1240 The "KernelPageSize" line (available since Linux 2.6.29) is the
1242 page size used by the kernel to back the virtual memory area.
1243 This matches the size used by the MMU in the majority of cases.
1244 However, one counter-example occurs on PPC64 kernels whereby a
1245 kernel using 64 kB as a base page size may still use 4 kB pages
1246 for the MMU on older processors. To distinguish the two
1247 attributes, the "MMUPageSize" line (also available since Linux
1248 2.6.29) reports the page size used by the MMU.
1249 The "Locked" indicates whether the mapping is locked in memory
1251 or not.
1252 The "ProtectionKey" line (available since Linux 4.9, on x86
1254 only) contains the memory protection key (see pkeys(7)) associ‐
1255 ated with the virtual memory area. This entry is present only
1256 if the kernel was built with the CONFIG_X86_INTEL_MEMORY_PROTEC‐
1257 TION_KEYS configuration option.
1258 The "VmFlags" line (available since Linux 3.8) represents the
1260 kernel flags associated with the virtual memory area, encoded
1261 using the following two-letter codes:
1262 rd - readable
1264 wr - writable
1265 ex - executable
1266 sh - shared
1267 mr - may read
1268 mw - may write
1269 me - may execute
1270 ms - may share
1271 gd - stack segment grows down
1272 pf - pure PFN range
1273 dw - disabled write to the mapped file
1274 lo - pages are locked in memory
1275 io - memory mapped I/O area
1276 sr - sequential read advise provided
1277 rr - random read advise provided
1278 dc - do not copy area on fork
1279 de - do not expand area on remapping
1280 ac - area is accountable
1281 nr - swap space is not reserved for the area
1282 ht - area uses huge tlb pages
1283 sf - perform synchronous page faults (since Linux 4.15)
1284 nl - non-linear mapping (removed in Linux 4.0)
1285 ar - architecture specific flag
1286 wf - wipe on fork (since Linux 4.14)
1287 dd - do not include area into core dump
1288 sd - soft-dirty flag (since Linux 3.13)
1289 mm - mixed map area
1290 hg - huge page advise flag
1291 nh - no-huge page advise flag
1292 mg - mergeable advise flag
1293 um - userfaultfd missing pages tracking (since Linux 4.3)
1294 uw - userfaultfd wprotect pages tracking (since Linux 4.3)
1295 "ProtectionKey" field contains the memory protection key (see
1297 pkeys(7)) associated with the virtual memory area. Present only
1298 if the kernel was built with the CONFIG_X86_INTEL_MEMORY_PROTEC‐
1299 TION_KEYS configuration option. (since Linux 4.6)
1300 The /proc/[pid]/smaps file is present only if the CON‐
1302 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
1303 /proc/[pid]/stack (since Linux 2.6.29)
1305 This file provides a symbolic trace of the function calls in
1306 this process's kernel stack. This file is provided only if the
1307 kernel was built with the CONFIG_STACKTRACE configuration
1308 option.
1309 Permission to access this file is governed by a ptrace access
1311 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1312 /proc/[pid]/stat
1314 Status information about the process. This is used by ps(1).
1315 It is defined in the kernel source file fs/proc/array.c.
1316 The fields, in order, with their proper scanf(3) format speci‐
1318 fiers, are listed below. Whether or not certain of these fields
1319 display valid information is governed by a ptrace access mode
1320 PTRACE_MODE_READ_FSCREDS | PTRACE_MODE_NOAUDIT check (refer to
1321 ptrace(2)). If the check denies access, then the field value is
1322 displayed as 0. The affected fields are indicated with the
1323 marking [PT].
1324 (1) pid %d
1326 The process ID.
1327 (2) comm %s
1329 The filename of the executable, in parentheses. This
1330 is visible whether or not the executable is swapped
1331 out.
1332 (3) state %c
1334 One of the following characters, indicating process
1335 state:
1336 R Running
1338 S Sleeping in an interruptible wait
1340 D Waiting in uninterruptible disk sleep
1342 Z Zombie
1344 T Stopped (on a signal) or (before Linux 2.6.33)
1346 trace stopped
1347 t Tracing stop (Linux 2.6.33 onward)
1349 W Paging (only before Linux 2.6.0)
1351 X Dead (from Linux 2.6.0 onward)
1353 x Dead (Linux 2.6.33 to 3.13 only)
1355 K Wakekill (Linux 2.6.33 to 3.13 only)
1357 W Waking (Linux 2.6.33 to 3.13 only)
1359 P Parked (Linux 3.9 to 3.13 only)
1361 (4) ppid %d
1363 The PID of the parent of this process.
1364 (5) pgrp %d
1366 The process group ID of the process.
1367 (6) session %d
1369 The session ID of the process.
1370 (7) tty_nr %d
1372 The controlling terminal of the process. (The minor
1373 device number is contained in the combination of bits
1374 31 to 20 and 7 to 0; the major device number is in
1375 bits 15 to 8.)
1376 (8) tpgid %d
1378 The ID of the foreground process group of the control‐
1379 ling terminal of the process.
1380 (9) flags %u
1382 The kernel flags word of the process. For bit mean‐
1383 ings, see the PF_* defines in the Linux kernel source
1384 file include/linux/sched.h. Details depend on the
1385 kernel version.
1386 The format for this field was %lu before Linux 2.6.
1388 (10) minflt %lu
1390 The number of minor faults the process has made which
1391 have not required loading a memory page from disk.
1392 (11) cminflt %lu
1394 The number of minor faults that the process's waited-
1395 for children have made.
1396 (12) majflt %lu
1398 The number of major faults the process has made which
1399 have required loading a memory page from disk.
1400 (13) cmajflt %lu
1402 The number of major faults that the process's waited-
1403 for children have made.
1404 (14) utime %lu
1406 Amount of time that this process has been scheduled in
1407 user mode, measured in clock ticks (divide by
1408 sysconf(_SC_CLK_TCK)). This includes guest time,
1409 guest_time (time spent running a virtual CPU, see
1410 below), so that applications that are not aware of the
1411 guest time field do not lose that time from their cal‐
1412 culations.
1413 (15) stime %lu
1415 Amount of time that this process has been scheduled in
1416 kernel mode, measured in clock ticks (divide by
1417 sysconf(_SC_CLK_TCK)).
1418 (16) cutime %ld
1420 Amount of time that this process's waited-for children
1421 have been scheduled in user mode, measured in clock
1422 ticks (divide by sysconf(_SC_CLK_TCK)). (See also
1423 times(2).) This includes guest time, cguest_time
1424 (time spent running a virtual CPU, see below).
1425 (17) cstime %ld
1427 Amount of time that this process's waited-for children
1428 have been scheduled in kernel mode, measured in clock
1429 ticks (divide by sysconf(_SC_CLK_TCK)).
1430 (18) priority %ld
1432 (Explanation for Linux 2.6) For processes running a
1433 real-time scheduling policy (policy below; see
1434 sched_setscheduler(2)), this is the negated scheduling
1435 priority, minus one; that is, a number in the range -2
1436 to -100, corresponding to real-time priorities 1 to
1437 99. For processes running under a non-real-time
1438 scheduling policy, this is the raw nice value (setpri‐
1439 ority(2)) as represented in the kernel. The kernel
1440 stores nice values as numbers in the range 0 (high) to
1441 39 (low), corresponding to the user-visible nice range
1442 of -20 to 19.
1443 Before Linux 2.6, this was a scaled value based on the
1445 scheduler weighting given to this process.
1446 (19) nice %ld
1448 The nice value (see setpriority(2)), a value in the
1449 range 19 (low priority) to -20 (high priority).
1450 (20) num_threads %ld
1452 Number of threads in this process (since Linux 2.6).
1453 Before kernel 2.6, this field was hard coded to 0 as a
1454 placeholder for an earlier removed field.
1455 (21) itrealvalue %ld
1457 The time in jiffies before the next SIGALRM is sent to
1458 the process due to an interval timer. Since kernel
1459 2.6.17, this field is no longer maintained, and is
1460 hard coded as 0.
1461 (22) starttime %llu
1463 The time the process started after system boot. In
1464 kernels before Linux 2.6, this value was expressed in
1465 jiffies. Since Linux 2.6, the value is expressed in
1466 clock ticks (divide by sysconf(_SC_CLK_TCK)).
1467 The format for this field was %lu before Linux 2.6.
1469 (23) vsize %lu
1471 Virtual memory size in bytes.
1472 (24) rss %ld
1474 Resident Set Size: number of pages the process has in
1475 real memory. This is just the pages which count
1476 toward text, data, or stack space. This does not
1477 include pages which have not been demand-loaded in, or
1478 which are swapped out.
1479 (25) rsslim %lu
1481 Current soft limit in bytes on the rss of the process;
1482 see the description of RLIMIT_RSS in getrlimit(2).
1483 (26) startcode %lu [PT]
1485 The address above which program text can run.
1486 (27) endcode %lu [PT]
1488 The address below which program text can run.
1489 (28) startstack %lu [PT]
1491 The address of the start (i.e., bottom) of the stack.
1492 (29) kstkesp %lu [PT]
1494 The current value of ESP (stack pointer), as found in
1495 the kernel stack page for the process.
1496 (30) kstkeip %lu [PT]
1498 The current EIP (instruction pointer).
1499 (31) signal %lu
1501 The bitmap of pending signals, displayed as a decimal
1502 number. Obsolete, because it does not provide infor‐
1503 mation on real-time signals; use /proc/[pid]/status
1504 instead.
1505 (32) blocked %lu
1507 The bitmap of blocked signals, displayed as a decimal
1508 number. Obsolete, because it does not provide infor‐
1509 mation on real-time signals; use /proc/[pid]/status
1510 instead.
1511 (33) sigignore %lu
1513 The bitmap of ignored signals, displayed as a decimal
1514 number. Obsolete, because it does not provide infor‐
1515 mation on real-time signals; use /proc/[pid]/status
1516 instead.
1517 (34) sigcatch %lu
1519 The bitmap of caught signals, displayed as a decimal
1520 number. Obsolete, because it does not provide infor‐
1521 mation on real-time signals; use /proc/[pid]/status
1522 instead.
1523 (35) wchan %lu [PT]
1525 This is the "channel" in which the process is waiting.
1526 It is the address of a location in the kernel where
1527 the process is sleeping. The corresponding symbolic
1528 name can be found in /proc/[pid]/wchan.
1529 (36) nswap %lu
1531 Number of pages swapped (not maintained).
1532 (37) cnswap %lu
1534 Cumulative nswap for child processes (not maintained).
1535 (38) exit_signal %d (since Linux 2.1.22)
1537 Signal to be sent to parent when we die.
1538 (39) processor %d (since Linux 2.2.8)
1540 CPU number last executed on.
1541 (40) rt_priority %u (since Linux 2.5.19)
1543 Real-time scheduling priority, a number in the range 1
1544 to 99 for processes scheduled under a real-time pol‐
1545 icy, or 0, for non-real-time processes (see
1546 sched_setscheduler(2)).
1547 (41) policy %u (since Linux 2.5.19)
1549 Scheduling policy (see sched_setscheduler(2)). Decode
1550 using the SCHED_* constants in linux/sched.h.
1551 The format for this field was %lu before Linux 2.6.22.
1553 (42) delayacct_blkio_ticks %llu (since Linux 2.6.18)
1555 Aggregated block I/O delays, measured in clock ticks
1556 (centiseconds).
1557 (43) guest_time %lu (since Linux 2.6.24)
1559 Guest time of the process (time spent running a vir‐
1560 tual CPU for a guest operating system), measured in
1561 clock ticks (divide by sysconf(_SC_CLK_TCK)).
1562 (44) cguest_time %ld (since Linux 2.6.24)
1564 Guest time of the process's children, measured in
1565 clock ticks (divide by sysconf(_SC_CLK_TCK)).
1566 (45) start_data %lu (since Linux 3.3) [PT]
1568 Address above which program initialized and uninitial‐
1569 ized (BSS) data are placed.
1570 (46) end_data %lu (since Linux 3.3) [PT]
1572 Address below which program initialized and uninitial‐
1573 ized (BSS) data are placed.
1574 (47) start_brk %lu (since Linux 3.3) [PT]
1576 Address above which program heap can be expanded with
1577 brk(2).
1578 (48) arg_start %lu (since Linux 3.5) [PT]
1580 Address above which program command-line arguments
1581 (argv) are placed.
1582 (49) arg_end %lu (since Linux 3.5) [PT]
1584 Address below program command-line arguments (argv)
1585 are placed.
1586 (50) env_start %lu (since Linux 3.5) [PT]
1588 Address above which program environment is placed.
1589 (51) env_end %lu (since Linux 3.5) [PT]
1591 Address below which program environment is placed.
1592 (52) exit_code %d (since Linux 3.5) [PT]
1594 The thread's exit status in the form reported by wait‐
1595 pid(2).
1596 /proc/[pid]/statm
1598 Provides information about memory usage, measured in pages. The
1599 columns are:
1600 size (1) total program size
1602 (same as VmSize in /proc/[pid]/status)
1603 resident (2) resident set size
1604 (same as VmRSS in /proc/[pid]/status)
1605 shared (3) number of resident shared pages (i.e., backed by a file)
1606 (same as RssFile+RssShmem in /proc/[pid]/status)
1607 text (4) text (code)
1608 lib (5) library (unused since Linux 2.6; always 0)
1609 data (6) data + stack
1610 dt (7) dirty pages (unused since Linux 2.6; always 0)
1611 /proc/[pid]/status
1613 Provides much of the information in /proc/[pid]/stat and
1614 /proc/[pid]/statm in a format that's easier for humans to parse.
1615 Here's an example:
1616 $ cat /proc/$$/status
1618 Name: bash
1619 Umask: 0022
1620 State: S (sleeping)
1621 Tgid: 17248
1622 Ngid: 0
1623 Pid: 17248
1624 PPid: 17200
1625 TracerPid: 0
1626 Uid: 1000 1000 1000 1000
1627 Gid: 100 100 100 100
1628 FDSize: 256
1629 Groups: 16 33 100
1630 NStgid: 17248
1631 NSpid: 17248
1632 NSpgid: 17248
1633 NSsid: 17200
1634 VmPeak: 131168 kB
1635 VmSize: 131168 kB
1636 VmLck: 0 kB
1637 VmPin: 0 kB
1638 VmHWM: 13484 kB
1639 VmRSS: 13484 kB
1640 RssAnon: 10264 kB
1641 RssFile: 3220 kB
1642 RssShmem: 0 kB
1643 VmData: 10332 kB
1644 VmStk: 136 kB
1645 VmExe: 992 kB
1646 VmLib: 2104 kB
1647 VmPTE: 76 kB
1648 VmPMD: 12 kB
1649 VmSwap: 0 kB
1650 HugetlbPages: 0 kB # 4.4
1651 CoreDumping: 0 # 4.15
1652 Threads: 1
1653 SigQ: 0/3067
1654 SigPnd: 0000000000000000
1655 ShdPnd: 0000000000000000
1656 SigBlk: 0000000000010000
1657 SigIgn: 0000000000384004
1658 SigCgt: 000000004b813efb
1659 CapInh: 0000000000000000
1660 CapPrm: 0000000000000000
1661 CapEff: 0000000000000000
1662 CapBnd: ffffffffffffffff
1663 CapAmb: 0000000000000000
1664 NoNewPrivs: 0
1665 Seccomp: 0
1666 Speculation_Store_Bypass: vulnerable
1667 Cpus_allowed: 00000001
1668 Cpus_allowed_list: 0
1669 Mems_allowed: 1
1670 Mems_allowed_list: 0
1671 voluntary_ctxt_switches: 150
1672 nonvoluntary_ctxt_switches: 545
1673 The fields are as follows:
1675 * Name: Command run by this process.
1677 * Umask: Process umask, expressed in octal with a leading zero;
1679 see umask(2). (Since Linux 4.7.)
1680 * State: Current state of the process. One of "R (running)", "S
1682 (sleeping)", "D (disk sleep)", "T (stopped)", "t (tracing
1683 stop)", "Z (zombie)", or "X (dead)".
1684 * Tgid: Thread group ID (i.e., Process ID).
1686 * Ngid: NUMA group ID (0 if none; since Linux 3.13).
1688 * Pid: Thread ID (see gettid(2)).
1690 * PPid: PID of parent process.
1692 * TracerPid: PID of process tracing this process (0 if not being
1694 traced).
1695 * Uid, Gid: Real, effective, saved set, and filesystem UIDs
1697 (GIDs).
1698 * FDSize: Number of file descriptor slots currently allocated.
1700 * Groups: Supplementary group list.
1702 * NStgid: Thread group ID (i.e., PID) in each of the PID names‐
1704 paces of which [pid] is a member. The leftmost entry shows
1705 the value with respect to the PID namespace of the process
1706 that mounted this procfs (or the root namespace if mounted by
1707 the kernel), followed by the value in successively nested
1708 inner namespaces. (Since Linux 4.1.)
1709 * NSpid: Thread ID in each of the PID namespaces of which [pid]
1711 is a member. The fields are ordered as for NStgid. (Since
1712 Linux 4.1.)
1713 * NSpgid: Process group ID in each of the PID namespaces of
1715 which [pid] is a member. The fields are ordered as for NSt‐
1716 gid. (Since Linux 4.1.)
1717 * NSsid: descendant namespace session ID hierarchy Session ID in
1719 each of the PID namespaces of which [pid] is a member. The
1720 fields are ordered as for NStgid. (Since Linux 4.1.)
1721 * VmPeak: Peak virtual memory size.
1723 * VmSize: Virtual memory size.
1725 * VmLck: Locked memory size (see mlock(2)).
1727 * VmPin: Pinned memory size (since Linux 3.2). These are pages
1729 that can't be moved because something needs to directly access
1730 physical memory.
1731 * VmHWM: Peak resident set size ("high water mark").
1733 * VmRSS: Resident set size. Note that the value here is the sum
1735 of RssAnon, RssFile, and RssShmem.
1736 * RssAnon: Size of resident anonymous memory. (since Linux
1738 4.5).
1739 * RssFile: Size of resident file mappings. (since Linux 4.5).
1741 * RssShmem: Size of resident shared memory (includes System V
1743 shared memory, mappings from tmpfs(5), and shared anonymous
1744 mappings). (since Linux 4.5).
1745 * VmData, VmStk, VmExe: Size of data, stack, and text segments.
1747 * VmLib: Shared library code size.
1749 * VmPTE: Page table entries size (since Linux 2.6.10).
1751 * VmPMD: Size of second-level page tables (added in Linux 4.0;
1753 removed in Linux 4.15).
1754 * VmSwap: Swapped-out virtual memory size by anonymous private
1756 pages; shmem swap usage is not included (since Linux 2.6.34).
1757 * HugetlbPages: Size of hugetlb memory portions (since Linux
1759 4.4).
1760 * CoreDumping: Contains the value 1 if the process is currently
1762 dumping core, and 0 if it is not (since Linux 4.15). This
1763 information can be used by a monitoring process to avoid
1764 killing a process that is currently dumping core, which could
1765 result in a corrupted core dump file.
1766 * Threads: Number of threads in process containing this thread.
1768 * SigQ: This field contains two slash-separated numbers that
1770 relate to queued signals for the real user ID of this process.
1771 The first of these is the number of currently queued signals
1772 for this real user ID, and the second is the resource limit on
1773 the number of queued signals for this process (see the
1774 description of RLIMIT_SIGPENDING in getrlimit(2)).
1775 * SigPnd, ShdPnd: Mask (expressed in hexadecimal) of signals
1777 pending for thread and for process as a whole (see pthreads(7)
1778 and signal(7)).
1779 * SigBlk, SigIgn, SigCgt: Masks (expressed in hexadecimal) indi‐
1781 cating signals being blocked, ignored, and caught (see sig‐
1782 nal(7)).
1783 * CapInh, CapPrm, CapEff: Masks (expressed in hexadecimal) of
1785 capabilities enabled in inheritable, permitted, and effective
1786 sets (see capabilities(7)).
1787 * CapBnd: Capability bounding set, expressed in hexadecimal
1789 (since Linux 2.6.26, see capabilities(7)).
1790 * CapAmb: Ambient capability set, expressed in hexadecimal
1792 (since Linux 4.3, see capabilities(7)).
1793 * NoNewPrivs: Value of the no_new_privs bit (since Linux 4.10,
1795 see prctl(2)).
1796 * Seccomp: Seccomp mode of the process (since Linux 3.8, see
1798 seccomp(2)). 0 means SECCOMP_MODE_DISABLED; 1 means SEC‐
1799 COMP_MODE_STRICT; 2 means SECCOMP_MODE_FILTER. This field is
1800 provided only if the kernel was built with the CONFIG_SECCOMP
1801 kernel configuration option enabled.
1802 * Speculation_Store_Bypass: Speculation flaw mitigation state
1804 (since Linux 4.17, see prctl(2)).
1805 * Cpus_allowed: Hexadecimal mask of CPUs on which this process
1807 may run (since Linux 2.6.24, see cpuset(7)).
1808 * Cpus_allowed_list: Same as previous, but in "list format"
1810 (since Linux 2.6.26, see cpuset(7)).
1811 * Mems_allowed: Mask of memory nodes allowed to this process
1813 (since Linux 2.6.24, see cpuset(7)).
1814 * Mems_allowed_list: Same as previous, but in "list format"
1816 (since Linux 2.6.26, see cpuset(7)).
1817 * voluntary_ctxt_switches, nonvoluntary_ctxt_switches: Number of
1819 voluntary and involuntary context switches (since Linux
1820 2.6.23).
1821 /proc/[pid]/syscall (since Linux 2.6.27)
1823 This file exposes the system call number and argument registers
1824 for the system call currently being executed by the process,
1825 followed by the values of the stack pointer and program counter
1826 registers. The values of all six argument registers are
1827 exposed, although most system calls use fewer registers.
1828 If the process is blocked, but not in a system call, then the
1830 file displays -1 in place of the system call number, followed by
1831 just the values of the stack pointer and program counter. If
1832 process is not blocked, then the file contains just the string
1833 "running".
1834 This file is present only if the kernel was configured with CON‐
1836 FIG_HAVE_ARCH_TRACEHOOK.
1837 Permission to access this file is governed by a ptrace access
1839 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1840 /proc/[pid]/task (since Linux 2.6.0)
1842 This is a directory that contains one subdirectory for each
1843 thread in the process. The name of each subdirectory is the
1844 numerical thread ID ([tid]) of the thread (see gettid(2)).
1845 Within each of these subdirectories, there is a set of files
1847 with the same names and contents as under the /proc/[pid] direc‐
1848 tories. For attributes that are shared by all threads, the con‐
1849 tents for each of the files under the task/[tid] subdirectories
1850 will be the same as in the corresponding file in the parent
1851 /proc/[pid] directory (e.g., in a multithreaded process, all of
1852 the task/[tid]/cwd files will have the same value as the
1853 /proc/[pid]/cwd file in the parent directory, since all of the
1854 threads in a process share a working directory). For attributes
1855 that are distinct for each thread, the corresponding files under
1856 task/[tid] may have different values (e.g., various fields in
1857 each of the task/[tid]/status files may be different for each
1858 thread), or they might not exist in /proc/[pid] at all.
1859 In a multithreaded process, the contents of the /proc/[pid]/task
1861 directory are not available if the main thread has already ter‐
1862 minated (typically by calling pthread_exit(3)).
1863 /proc/[pid]/task/[tid]/children (since Linux 3.5)
1865 A space-separated list of child tasks of this task. Each child
1866 task is represented by its TID.
1867 This option is intended for use by the checkpoint-restore (CRIU)
1869 system, and reliably provides a list of children only if all of
1870 the child processes are stopped or frozen. It does not work
1871 properly if children of the target task exit while the file is
1872 being read! Exiting children may cause non-exiting children to
1873 be omitted from the list. This makes this interface even more
1874 unreliable than classic PID-based approaches if the inspected
1875 task and its children aren't frozen, and most code should proba‐
1876 bly not use this interface.
1877 Until Linux 4.2, the presence of this file was governed by the
1879 CONFIG_CHECKPOINT_RESTORE kernel configuration option. Since
1880 Linux 4.2, it is governed by the CONFIG_PROC_CHILDREN option.
1881 /proc/[pid]/timers (since Linux 3.10)
1883 A list of the POSIX timers for this process. Each timer is
1884 listed with a line that starts with the string "ID:". For exam‐
1885 ple:
1886 ID: 1
1888 signal: 60/00007fff86e452a8
1889 notify: signal/pid.2634
1890 ClockID: 0
1891 ID: 0
1892 signal: 60/00007fff86e452a8
1893 notify: signal/pid.2634
1894 ClockID: 1
1895 The lines shown for each timer have the following meanings:
1897 ID The ID for this timer. This is not the same as the timer
1899 ID returned by timer_create(2); rather, it is the same
1900 kernel-internal ID that is available via the si_timerid
1901 field of the siginfo_t structure (see sigaction(2)).
1902 signal This is the signal number that this timer uses to deliver
1904 notifications followed by a slash, and then the
1905 sigev_value value supplied to the signal handler. Valid
1906 only for timers that notify via a signal.
1907 notify The part before the slash specifies the mechanism that
1909 this timer uses to deliver notifications, and is one of
1910 "thread", "signal", or "none". Immediately following the
1911 slash is either the string "tid" for timers with
1912 SIGEV_THREAD_ID notification, or "pid" for timers that
1913 notify by other mechanisms. Following the "." is the PID
1914 of the process (or the kernel thread ID of the thread)
1915 that will be delivered a signal if the timer delivers
1916 notifications via a signal.
1917 ClockID
1919 This field identifies the clock that the timer uses for
1920 measuring time. For most clocks, this is a number that
1921 matches one of the user-space CLOCK_* constants exposed
1922 via <time.h>. CLOCK_PROCESS_CPUTIME_ID timers display
1923 with a value of -6 in this field.
1924 CLOCK_THREAD_CPUTIME_ID timers display with a value of -2
1925 in this field.
1926 This file is available only when the kernel was configured with
1928 CONFIG_CHECKPOINT_RESTORE.
1929 /proc/[pid]/timerslack_ns (since Linux 4.6)
1931 This file exposes the process's "current" timer slack value,
1932 expressed in nanoseconds. The file is writable, allowing the
1933 process's timer slack value to be changed. Writing 0 to this
1934 file resets the "current" timer slack to the "default" timer
1935 slack value. For further details, see the discussion of
1936 PR_SET_TIMERSLACK in prctl(2).
1937 Initially, permission to access this file was governed by a
1939 ptrace access mode PTRACE_MODE_ATTACH_FSCREDS check (see
1940 ptrace(2)). However, this was subsequently deemed too strict a
1941 requirement (and had the side effect that requiring a process to
1942 have the CAP_SYS_PTRACE capability would also allow it to view
1943 and change any process's memory). Therefore, since Linux 4.9,
1944 only the (weaker) CAP_SYS_NICE capability is required to access
1945 this file.
1946 /proc/[pid]/uid_map, /proc/[pid]/gid_map (since Linux 3.5)
1948 See user_namespaces(7).
1949 /proc/[pid]/wchan (since Linux 2.6.0)
1951 The symbolic name corresponding to the location in the kernel
1952 where the process is sleeping.
1953 Permission to access this file is governed by a ptrace access
1955 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1956 /proc/[tid]
1958 There is a numerical subdirectory for each running thread that
1959 is not a thread group leader (i.e., a thread whose thread ID is
1960 not the same as its process ID); the subdirectory is named by
1961 the thread ID. Each one of these subdirectories contains files
1962 and subdirectories exposing information about the thread with
1963 the thread ID tid. The contents of these directories are the
1964 same as the corresponding /proc/[pid]/task/[tid] directories.
1965 The /proc/[tid] subdirectories are not visible when iterating
1967 through /proc with getdents(2) (and thus are not visible when
1968 one uses ls(1) to view the contents of /proc). However, the
1969 pathnames of these directories are visible to (i.e., usable as
1970 arguments in) system calls that operate on pathnames.
1971 /proc/apm
1973 Advanced power management version and battery information when
1974 CONFIG_APM is defined at kernel compilation time.
1975 /proc/buddyinfo
1977 This file contains information which is used for diagnosing mem‐
1978 ory fragmentation issues. Each line starts with the identifica‐
1979 tion of the node and the name of the zone which together iden‐
1980 tify a memory region. This is then followed by the count of
1981 available chunks of a certain order in which these zones are
1982 split. The size in bytes of a certain order is given by the
1983 formula:
1984 (2^order) * PAGE_SIZE
1986 The binary buddy allocator algorithm inside the kernel will
1988 split one chunk into two chunks of a smaller order (thus with
1989 half the size) or combine two contiguous chunks into one larger
1990 chunk of a higher order (thus with double the size) to satisfy
1991 allocation requests and to counter memory fragmentation. The
1992 order matches the column number, when starting to count at zero.
1993 For example on an x86-64 system:
1995 Node 0, zone DMA 1 1 1 0 2 1 1 0 1 1 3
1997 Node 0, zone DMA32 65 47 4 81 52 28 13 10 5 1 404
1998 Node 0, zone Normal 216 55 189 101 84 38 37 27 5 3 587
1999 In this example, there is one node containing three zones and
2001 there are 11 different chunk sizes. If the page size is 4 kilo‐
2002 bytes, then the first zone called DMA (on x86 the first 16
2003 megabyte of memory) has 1 chunk of 4 kilobytes (order 0) avail‐
2004 able and has 3 chunks of 4 megabytes (order 10) available.
2005 If the memory is heavily fragmented, the counters for higher
2007 order chunks will be zero and allocation of large contiguous
2008 areas will fail.
2009 Further information about the zones can be found in /proc/zone‐
2011 info.
2012 /proc/bus
2014 Contains subdirectories for installed busses.
2015 /proc/bus/pccard
2017 Subdirectory for PCMCIA devices when CONFIG_PCMCIA is set at
2018 kernel compilation time.
2019 /proc/bus/pccard/drivers
2021 /proc/bus/pci
2023 Contains various bus subdirectories and pseudo-files containing
2024 information about PCI busses, installed devices, and device
2025 drivers. Some of these files are not ASCII.
2026 /proc/bus/pci/devices
2028 Information about PCI devices. They may be accessed through
2029 lspci(8) and setpci(8).
2030 /proc/cgroups (since Linux 2.6.24)
2032 See cgroups(7).
2033 /proc/cmdline
2035 Arguments passed to the Linux kernel at boot time. Often done
2036 via a boot manager such as lilo(8) or grub(8).
2037 /proc/config.gz (since Linux 2.6)
2039 This file exposes the configuration options that were used to
2040 build the currently running kernel, in the same format as they
2041 would be shown in the .config file that resulted when configur‐
2042 ing the kernel (using make xconfig, make config, or similar).
2043 The file contents are compressed; view or search them using
2044 zcat(1) and zgrep(1). As long as no changes have been made to
2045 the following file, the contents of /proc/config.gz are the same
2046 as those provided by:
2047 cat /lib/modules/$(uname -r)/build/.config
2049 /proc/config.gz is provided only if the kernel is configured
2051 with CONFIG_IKCONFIG_PROC.
2052 /proc/crypto
2054 A list of the ciphers provided by the kernel crypto API. For
2055 details, see the kernel Linux Kernel Crypto API documentation
2056 available under the kernel source directory Documenta‐
2057 tion/crypto/ (or Documentation/DocBook before 4.10; the documen‐
2058 tation can be built using a command such as make htmldocs in the
2059 root directory of the kernel source tree).
2060 /proc/cpuinfo
2062 This is a collection of CPU and system architecture dependent
2063 items, for each supported architecture a different list. Two
2064 common entries are processor which gives CPU number and
2065 bogomips; a system constant that is calculated during kernel
2066 initialization. SMP machines have information for each CPU.
2067 The lscpu(1) command gathers its information from this file.
2068 /proc/devices
2070 Text listing of major numbers and device groups. This can be
2071 used by MAKEDEV scripts for consistency with the kernel.
2072 /proc/diskstats (since Linux 2.5.69)
2074 This file contains disk I/O statistics for each disk device.
2075 See the Linux kernel source file Documentation/iostats.txt for
2076 further information.
2077 /proc/dma
2079 This is a list of the registered ISA DMA (direct memory access)
2080 channels in use.
2081 /proc/driver
2083 Empty subdirectory.
2084 /proc/execdomains
2086 List of the execution domains (ABI personalities).
2087 /proc/fb
2089 Frame buffer information when CONFIG_FB is defined during kernel
2090 compilation.
2091 /proc/filesystems
2093 A text listing of the filesystems which are supported by the
2094 kernel, namely filesystems which were compiled into the kernel
2095 or whose kernel modules are currently loaded. (See also
2096 filesystems(5).) If a filesystem is marked with "nodev", this
2097 means that it does not require a block device to be mounted
2098 (e.g., virtual filesystem, network filesystem).
2099 Incidentally, this file may be used by mount(8) when no filesys‐
2101 tem is specified and it didn't manage to determine the filesys‐
2102 tem type. Then filesystems contained in this file are tried
2103 (excepted those that are marked with "nodev").
2104 /proc/fs
2106 Contains subdirectories that in turn contain files with informa‐
2107 tion about (certain) mounted filesystems.
2108 /proc/ide
2110 This directory exists on systems with the IDE bus. There are
2111 directories for each IDE channel and attached device. Files
2112 include:
2113 cache buffer size in KB
2115 capacity number of sectors
2116 driver driver version
2117 geometry physical and logical geometry
2118 identify in hexadecimal
2119 media media type
2120 model manufacturer's model number
2121 settings drive settings
2122 smart_thresholds IDE disk management thresholds (in hex)
2123 smart_values IDE disk management values (in hex)
2124 The hdparm(8) utility provides access to this information in a
2126 friendly format.
2127 /proc/interrupts
2129 This is used to record the number of interrupts per CPU per IO
2130 device. Since Linux 2.6.24, for the i386 and x86-64 architec‐
2131 tures, at least, this also includes interrupts internal to the
2132 system (that is, not associated with a device as such), such as
2133 NMI (nonmaskable interrupt), LOC (local timer interrupt), and
2134 for SMP systems, TLB (TLB flush interrupt), RES (rescheduling
2135 interrupt), CAL (remote function call interrupt), and possibly
2136 others. Very easy to read formatting, done in ASCII.
2137 /proc/iomem
2139 I/O memory map in Linux 2.4.
2140 /proc/ioports
2142 This is a list of currently registered Input-Output port regions
2143 that are in use.
2144 /proc/kallsyms (since Linux 2.5.71)
2146 This holds the kernel exported symbol definitions used by the
2147 modules(X) tools to dynamically link and bind loadable modules.
2148 In Linux 2.5.47 and earlier, a similar file with slightly dif‐
2149 ferent syntax was named ksyms.
2150 /proc/kcore
2152 This file represents the physical memory of the system and is
2153 stored in the ELF core file format. With this pseudo-file, and
2154 an unstripped kernel (/usr/src/linux/vmlinux) binary, GDB can be
2155 used to examine the current state of any kernel data structures.
2156 The total length of the file is the size of physical memory
2158 (RAM) plus 4 KiB.
2159 /proc/keys (since Linux 2.6.10)
2161 See keyrings(7).
2162 /proc/key-users (since Linux 2.6.10)
2164 See keyrings(7).
2165 /proc/kmsg
2167 This file can be used instead of the syslog(2) system call to
2168 read kernel messages. A process must have superuser privileges
2169 to read this file, and only one process should read this file.
2170 This file should not be read if a syslog process is running
2171 which uses the syslog(2) system call facility to log kernel mes‐
2172 sages.
2173 Information in this file is retrieved with the dmesg(1) program.
2175 /proc/kpagecgroup (since Linux 4.3)
2177 This file contains a 64-bit inode number of the memory cgroup
2178 each page is charged to, indexed by page frame number (see the
2179 discussion of /proc/[pid]/pagemap).
2180 The /proc/kpagecgroup file is present only if the CONFIG_MEMCG
2182 kernel configuration option is enabled.
2183 /proc/kpagecount (since Linux 2.6.25)
2185 This file contains a 64-bit count of the number of times each
2186 physical page frame is mapped, indexed by page frame number (see
2187 the discussion of /proc/[pid]/pagemap).
2188 The /proc/kpagecount file is present only if the CON‐
2190 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
2191 /proc/kpageflags (since Linux 2.6.25)
2193 This file contains 64-bit masks corresponding to each physical
2194 page frame; it is indexed by page frame number (see the discus‐
2195 sion of /proc/[pid]/pagemap). The bits are as follows:
2196 0 - KPF_LOCKED
2198 1 - KPF_ERROR
2199 2 - KPF_REFERENCED
2200 3 - KPF_UPTODATE
2201 4 - KPF_DIRTY
2202 5 - KPF_LRU
2203 6 - KPF_ACTIVE
2204 7 - KPF_SLAB
2205 8 - KPF_WRITEBACK
2206 9 - KPF_RECLAIM
2207 10 - KPF_BUDDY
2208 11 - KPF_MMAP (since Linux 2.6.31)
2209 12 - KPF_ANON (since Linux 2.6.31)
2210 13 - KPF_SWAPCACHE (since Linux 2.6.31)
2211 14 - KPF_SWAPBACKED (since Linux 2.6.31)
2212 15 - KPF_COMPOUND_HEAD (since Linux 2.6.31)
2213 16 - KPF_COMPOUND_TAIL (since Linux 2.6.31)
2214 17 - KPF_HUGE (since Linux 2.6.31)
2215 18 - KPF_UNEVICTABLE (since Linux 2.6.31)
2216 19 - KPF_HWPOISON (since Linux 2.6.31)
2217 20 - KPF_NOPAGE (since Linux 2.6.31)
2218 21 - KPF_KSM (since Linux 2.6.32)
2219 22 - KPF_THP (since Linux 3.4)
2220 23 - KPF_BALLOON (since Linux 3.18)
2221 24 - KPF_ZERO_PAGE (since Linux 4.0)
2222 25 - KPF_IDLE (since Linux 4.3)
2223 For further details on the meanings of these bits, see the ker‐
2225 nel source file Documentation/admin-guide/mm/pagemap.rst.
2226 Before kernel 2.6.29, KPF_WRITEBACK, KPF_RECLAIM, KPF_BUDDY, and
2227 KPF_LOCKED did not report correctly.
2228 The /proc/kpageflags file is present only if the CON‐
2230 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
2231 /proc/ksyms (Linux 1.1.23–2.5.47)
2233 See /proc/kallsyms.
2234 /proc/loadavg
2236 The first three fields in this file are load average figures
2237 giving the number of jobs in the run queue (state R) or waiting
2238 for disk I/O (state D) averaged over 1, 5, and 15 minutes. They
2239 are the same as the load average numbers given by uptime(1) and
2240 other programs. The fourth field consists of two numbers sepa‐
2241 rated by a slash (/). The first of these is the number of cur‐
2242 rently runnable kernel scheduling entities (processes, threads).
2243 The value after the slash is the number of kernel scheduling
2244 entities that currently exist on the system. The fifth field is
2245 the PID of the process that was most recently created on the
2246 system.
2247 /proc/locks
2249 This file shows current file locks (flock(2) and fcntl(2)) and
2250 leases (fcntl(2)).
2251 An example of the content shown in this file is the following:
2253 1: POSIX ADVISORY READ 5433 08:01:7864448 128 128
2255 2: FLOCK ADVISORY WRITE 2001 08:01:7864554 0 EOF
2256 3: FLOCK ADVISORY WRITE 1568 00:2f:32388 0 EOF
2257 4: POSIX ADVISORY WRITE 699 00:16:28457 0 EOF
2258 5: POSIX ADVISORY WRITE 764 00:16:21448 0 0
2259 6: POSIX ADVISORY READ 3548 08:01:7867240 1 1
2260 7: POSIX ADVISORY READ 3548 08:01:7865567 1826 2335
2261 8: OFDLCK ADVISORY WRITE -1 08:01:8713209 128 191
2262 The fields shown in each line are as follows:
2264 (1) The ordinal position of the lock in the list.
2266 (2) The lock type. Values that may appear here include:
2268 FLOCK This is a BSD file lock created using flock(2).
2270 OFDLCK This is an open file description (OFD) lock created
2272 using fcntl(2).
2273 POSIX This is a POSIX byte-range lock created using
2275 fcntl(2).
2276 (3) Among the strings that can appear here are the following:
2278 ADVISORY
2280 This is an advisory lock.
2281 MANDATORY
2283 This is a mandatory lock.
2284 (4) The type of lock. Values that can appear here are:
2286 READ This is a POSIX or OFD read lock, or a BSD shared
2288 lock.
2289 WRITE This is a POSIX or OFD write lock, or a BSD exclusive
2291 lock.
2292 (5) The PID of the process that owns the lock.
2294 Because OFD locks are not owned by a single process (since
2296 multiple processes may have file descriptors that refer to
2297 the same open file description), the value -1 is displayed
2298 in this field for OFD locks. (Before kernel 4.14, a bug
2299 meant that the PID of the process that initially acquired
2300 the lock was displayed instead of the value -1.)
2301 (6) Three colon-separated subfields that identify the major and
2303 minor device ID of the device containing the filesystem
2304 where the locked file resides, followed by the inode number
2305 of the locked file.
2306 (7) The byte offset of the first byte of the lock. For BSD
2308 locks, this value is always 0.
2309 (8) The byte offset of the last byte of the lock. EOF in this
2311 field means that the lock extends to the end of the file.
2312 For BSD locks, the value shown is always EOF.
2313 Since Linux 4.9, the list of locks shown in /proc/locks is fil‐
2315 tered to show just the locks for the processes in the PID names‐
2316 pace (see pid_namespaces(7)) for which the /proc filesystem was
2317 mounted. (In the initial PID namespace, there is no filtering
2318 of the records shown in this file.)
2319 The lslocks(8) command provides a bit more information about
2321 each lock.
2322 /proc/malloc (only up to and including Linux 2.2)
2324 This file is present only if CONFIG_DEBUG_MALLOC was defined
2325 during compilation.
2326 /proc/meminfo
2328 This file reports statistics about memory usage on the system.
2329 It is used by free(1) to report the amount of free and used mem‐
2330 ory (both physical and swap) on the system as well as the shared
2331 memory and buffers used by the kernel. Each line of the file
2332 consists of a parameter name, followed by a colon, the value of
2333 the parameter, and an option unit of measurement (e.g., "kB").
2334 The list below describes the parameter names and the format
2335 specifier required to read the field value. Except as noted
2336 below, all of the fields have been present since at least Linux
2337 2.6.0. Some fields are displayed only if the kernel was config‐
2338 ured with various options; those dependencies are noted in the
2339 list.
2340 MemTotal %lu
2342 Total usable RAM (i.e., physical RAM minus a few reserved
2343 bits and the kernel binary code).
2344 MemFree %lu
2346 The sum of LowFree+HighFree.
2347 MemAvailable %lu (since Linux 3.14)
2349 An estimate of how much memory is available for starting
2350 new applications, without swapping.
2351 Buffers %lu
2353 Relatively temporary storage for raw disk blocks that
2354 shouldn't get tremendously large (20 MB or so).
2355 Cached %lu
2357 In-memory cache for files read from the disk (the page
2358 cache). Doesn't include SwapCached.
2359 SwapCached %lu
2361 Memory that once was swapped out, is swapped back in but
2362 still also is in the swap file. (If memory pressure is
2363 high, these pages don't need to be swapped out again
2364 because they are already in the swap file. This saves
2365 I/O.)
2366 Active %lu
2368 Memory that has been used more recently and usually not
2369 reclaimed unless absolutely necessary.
2370 Inactive %lu
2372 Memory which has been less recently used. It is more
2373 eligible to be reclaimed for other purposes.
2374 Active(anon) %lu (since Linux 2.6.28)
2376 [To be documented.]
2377 Inactive(anon) %lu (since Linux 2.6.28)
2379 [To be documented.]
2380 Active(file) %lu (since Linux 2.6.28)
2382 [To be documented.]
2383 Inactive(file) %lu (since Linux 2.6.28)
2385 [To be documented.]
2386 Unevictable %lu (since Linux 2.6.28)
2388 (From Linux 2.6.28 to 2.6.30, CONFIG_UNEVICTABLE_LRU was
2389 required.) [To be documented.]
2390 Mlocked %lu (since Linux 2.6.28)
2392 (From Linux 2.6.28 to 2.6.30, CONFIG_UNEVICTABLE_LRU was
2393 required.) [To be documented.]
2394 HighTotal %lu
2396 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2397 Total amount of highmem. Highmem is all memory above
2398 ~860 MB of physical memory. Highmem areas are for use by
2399 user-space programs, or for the page cache. The kernel
2400 must use tricks to access this memory, making it slower
2401 to access than lowmem.
2402 HighFree %lu
2404 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2405 Amount of free highmem.
2406 LowTotal %lu
2408 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2409 Total amount of lowmem. Lowmem is memory which can be
2410 used for everything that highmem can be used for, but it
2411 is also available for the kernel's use for its own data
2412 structures. Among many other things, it is where every‐
2413 thing from Slab is allocated. Bad things happen when
2414 you're out of lowmem.
2415 LowFree %lu
2417 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2418 Amount of free lowmem.
2419 MmapCopy %lu (since Linux 2.6.29)
2421 (CONFIG_MMU is required.) [To be documented.]
2422 SwapTotal %lu
2424 Total amount of swap space available.
2425 SwapFree %lu
2427 Amount of swap space that is currently unused.
2428 Dirty %lu
2430 Memory which is waiting to get written back to the disk.
2431 Writeback %lu
2433 Memory which is actively being written back to the disk.
2434 AnonPages %lu (since Linux 2.6.18)
2436 Non-file backed pages mapped into user-space page tables.
2437 Mapped %lu
2439 Files which have been mapped into memory (with mmap(2)),
2440 such as libraries.
2441 Shmem %lu (since Linux 2.6.32)
2443 Amount of memory consumed in tmpfs(5) filesystems.
2444 KReclaimable %lu (since Linux 4.20)
2446 Kernel allocations that the kernel will attempt to
2447 reclaim under memory pressure. Includes SReclaimable
2448 (below), and other direct allocations with a shrinker.
2449 Slab %lu
2451 In-kernel data structures cache. (See slabinfo(5).)
2452 SReclaimable %lu (since Linux 2.6.19)
2454 Part of Slab, that might be reclaimed, such as caches.
2455 SUnreclaim %lu (since Linux 2.6.19)
2457 Part of Slab, that cannot be reclaimed on memory pres‐
2458 sure.
2459 KernelStack %lu (since Linux 2.6.32)
2461 Amount of memory allocated to kernel stacks.
2462 PageTables %lu (since Linux 2.6.18)
2464 Amount of memory dedicated to the lowest level of page
2465 tables.
2466 Quicklists %lu (since Linux 2.6.27)
2468 (CONFIG_QUICKLIST is required.) [To be documented.]
2469 NFS_Unstable %lu (since Linux 2.6.18)
2471 NFS pages sent to the server, but not yet committed to
2472 stable storage.
2473 Bounce %lu (since Linux 2.6.18)
2475 Memory used for block device "bounce buffers".
2476 WritebackTmp %lu (since Linux 2.6.26)
2478 Memory used by FUSE for temporary writeback buffers.
2479 CommitLimit %lu (since Linux 2.6.10)
2481 This is the total amount of memory currently available to
2482 be allocated on the system, expressed in kilobytes. This
2483 limit is adhered to only if strict overcommit accounting
2484 is enabled (mode 2 in /proc/sys/vm/overcommit_memory).
2485 The limit is calculated according to the formula
2486 described under /proc/sys/vm/overcommit_memory. For fur‐
2487 ther details, see the kernel source file Documenta‐
2488 tion/vm/overcommit-accounting.rst.
2489 Committed_AS %lu
2491 The amount of memory presently allocated on the system.
2492 The committed memory is a sum of all of the memory which
2493 has been allocated by processes, even if it has not been
2494 "used" by them as of yet. A process which allocates 1 GB
2495 of memory (using malloc(3) or similar), but touches only
2496 300 MB of that memory will show up as using only 300 MB
2497 of memory even if it has the address space allocated for
2498 the entire 1 GB.
2499 This 1 GB is memory which has been "committed" to by the
2501 VM and can be used at any time by the allocating applica‐
2502 tion. With strict overcommit enabled on the system (mode
2503 2 in /proc/sys/vm/overcommit_memory), allocations which
2504 would exceed the CommitLimit will not be permitted. This
2505 is useful if one needs to guarantee that processes will
2506 not fail due to lack of memory once that memory has been
2507 successfully allocated.
2508 VmallocTotal %lu
2510 Total size of vmalloc memory area.
2511 VmallocUsed %lu
2513 Amount of vmalloc area which is used. Since Linux 4.4,
2514 this field is no longer calculated, and is hard coded as
2515 0. See /proc/vmallocinfo.
2516 VmallocChunk %lu
2518 Largest contiguous block of vmalloc area which is free.
2519 Since Linux 4.4, this field is no longer calculated and
2520 is hard coded as 0. See /proc/vmallocinfo.
2521 HardwareCorrupted %lu (since Linux 2.6.32)
2523 (CONFIG_MEMORY_FAILURE is required.) [To be documented.]
2524 LazyFree %lu (since Linux 4.12)
2526 Shows the amount of memory marked by madvise(2)
2527 MADV_FREE.
2528 AnonHugePages %lu (since Linux 2.6.38)
2530 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Non-file
2531 backed huge pages mapped into user-space page tables.
2532 ShmemHugePages %lu (since Linux 4.8)
2534 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Memory used
2535 by shared memory (shmem) and tmpfs(5) allocated with huge
2536 pages.
2537 ShmemPmdMapped %lu (since Linux 4.8)
2539 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Shared memory
2540 mapped into user space with huge pages.
2541 CmaTotal %lu (since Linux 3.1)
2543 Total CMA (Contiguous Memory Allocator) pages. (CON‐
2544 FIG_CMA is required.)
2545 CmaFree %lu (since Linux 3.1)
2547 Free CMA (Contiguous Memory Allocator) pages. (CON‐
2548 FIG_CMA is required.)
2549 HugePages_Total %lu
2551 (CONFIG_HUGETLB_PAGE is required.) The size of the pool
2552 of huge pages.
2553 HugePages_Free %lu
2555 (CONFIG_HUGETLB_PAGE is required.) The number of huge
2556 pages in the pool that are not yet allocated.
2557 HugePages_Rsvd %lu (since Linux 2.6.17)
2559 (CONFIG_HUGETLB_PAGE is required.) This is the number of
2560 huge pages for which a commitment to allocate from the
2561 pool has been made, but no allocation has yet been made.
2562 These reserved huge pages guarantee that an application
2563 will be able to allocate a huge page from the pool of
2564 huge pages at fault time.
2565 HugePages_Surp %lu (since Linux 2.6.24)
2567 (CONFIG_HUGETLB_PAGE is required.) This is the number of
2568 huge pages in the pool above the value in
2569 /proc/sys/vm/nr_hugepages. The maximum number of surplus
2570 huge pages is controlled by /proc/sys/vm/nr_overcom‐
2571 mit_hugepages.
2572 Hugepagesize %lu
2574 (CONFIG_HUGETLB_PAGE is required.) The size of huge
2575 pages.
2576 DirectMap4k %lu (since Linux 2.6.27)
2578 Number of bytes of RAM linearly mapped by kernel in 4 kB
2579 pages. (x86.)
2580 DirectMap4M %lu (since Linux 2.6.27)
2582 Number of bytes of RAM linearly mapped by kernel in 4 MB
2583 pages. (x86 with CONFIG_X86_64 or CONFIG_X86_PAE
2584 enabled.)
2585 DirectMap2M %lu (since Linux 2.6.27)
2587 Number of bytes of RAM linearly mapped by kernel in 2 MB
2588 pages. (x86 with neither CONFIG_X86_64 nor CON‐
2589 FIG_X86_PAE enabled.)
2590 DirectMap1G %lu (since Linux 2.6.27)
2592 (x86 with CONFIG_X86_64 and CONFIG_X86_DIRECT_GBPAGES
2593 enabled.)
2594 /proc/modules
2596 A text list of the modules that have been loaded by the system.
2597 See also lsmod(8).
2598 /proc/mounts
2600 Before kernel 2.4.19, this file was a list of all the filesys‐
2601 tems currently mounted on the system. With the introduction of
2602 per-process mount namespaces in Linux 2.4.19 (see mount_names‐
2603 paces(7)), this file became a link to /proc/self/mounts, which
2604 lists the mount points of the process's own mount namespace.
2605 The format of this file is documented in fstab(5).
2606 /proc/mtrr
2608 Memory Type Range Registers. See the Linux kernel source file
2609 Documentation/x86/mtrr.txt (or Documentation/mtrr.txt before
2610 Linux 2.6.28) for details.
2611 /proc/net
2613 This directory contains various files and subdirectories con‐
2614 taining information about the networking layer. The files con‐
2615 tain ASCII structures and are, therefore, readable with cat(1).
2616 However, the standard netstat(8) suite provides much cleaner
2617 access to these files.
2618 With the advent of network namespaces, various information
2620 relating to the network stack is virtualized (see network_names‐
2621 paces(7)). Thus, since Linux 2.6.25, /proc/net is a symbolic
2622 link to the directory /proc/self/net, which contains the same
2623 files and directories as listed below. However, these files and
2624 directories now expose information for the network namespace of
2625 which the process is a member.
2626 /proc/net/arp
2628 This holds an ASCII readable dump of the kernel ARP table used
2629 for address resolutions. It will show both dynamically learned
2630 and preprogrammed ARP entries. The format is:
2631 IP address HW type Flags HW address Mask Device
2633 192.168.0.50 0x1 0x2 00:50:BF:25:68:F3 * eth0
2634 192.168.0.250 0x1 0xc 00:00:00:00:00:00 * eth0
2635 Here "IP address" is the IPv4 address of the machine and the "HW
2637 type" is the hardware type of the address from RFC 826. The
2638 flags are the internal flags of the ARP structure (as defined in
2639 /usr/include/linux/if_arp.h) and the "HW address" is the data
2640 link layer mapping for that IP address if it is known.
2641 /proc/net/dev
2643 The dev pseudo-file contains network device status information.
2644 This gives the number of received and sent packets, the number
2645 of errors and collisions and other basic statistics. These are
2646 used by the ifconfig(8) program to report device status. The
2647 format is:
2648 Inter-| Receive | Transmit
2650 face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
2651 lo: 2776770 11307 0 0 0 0 0 0 2776770 11307 0 0 0 0 0 0
2652 eth0: 1215645 2751 0 0 0 0 0 0 1782404 4324 0 0 0 427 0 0
2653 ppp0: 1622270 5552 1 0 0 0 0 0 354130 5669 0 0 0 0 0 0
2654 tap0: 7714 81 0 0 0 0 0 0 7714 81 0 0 0 0 0 0
2655 /proc/net/dev_mcast
2657 Defined in /usr/src/linux/net/core/dev_mcast.c:
2658 indx interface_name dmi_u dmi_g dmi_address
2660 2 eth0 1 0 01005e000001
2661 3 eth1 1 0 01005e000001
2662 4 eth2 1 0 01005e000001
2663 /proc/net/igmp
2665 Internet Group Management Protocol. Defined in
2666 /usr/src/linux/net/core/igmp.c.
2667 /proc/net/rarp
2669 This file uses the same format as the arp file and contains the
2670 current reverse mapping database used to provide rarp(8) reverse
2671 address lookup services. If RARP is not configured into the
2672 kernel, this file will not be present.
2673 /proc/net/raw
2675 Holds a dump of the RAW socket table. Much of the information
2676 is not of use apart from debugging. The "sl" value is the ker‐
2677 nel hash slot for the socket, the "local_address" is the local
2678 address and protocol number pair. "St" is the internal status
2679 of the socket. The "tx_queue" and "rx_queue" are the outgoing
2680 and incoming data queue in terms of kernel memory usage. The
2681 "tr", "tm->when", and "rexmits" fields are not used by RAW. The
2682 "uid" field holds the effective UID of the creator of the
2683 socket.
2684 /proc/net/snmp
2686 This file holds the ASCII data needed for the IP, ICMP, TCP, and
2687 UDP management information bases for an SNMP agent.
2688 /proc/net/tcp
2690 Holds a dump of the TCP socket table. Much of the information
2691 is not of use apart from debugging. The "sl" value is the ker‐
2692 nel hash slot for the socket, the "local_address" is the local
2693 address and port number pair. The "rem_address" is the remote
2694 address and port number pair (if connected). "St" is the inter‐
2695 nal status of the socket. The "tx_queue" and "rx_queue" are the
2696 outgoing and incoming data queue in terms of kernel memory
2697 usage. The "tr", "tm->when", and "rexmits" fields hold internal
2698 information of the kernel socket state and are useful only for
2699 debugging. The "uid" field holds the effective UID of the cre‐
2700 ator of the socket.
2701 /proc/net/udp
2703 Holds a dump of the UDP socket table. Much of the information
2704 is not of use apart from debugging. The "sl" value is the ker‐
2705 nel hash slot for the socket, the "local_address" is the local
2706 address and port number pair. The "rem_address" is the remote
2707 address and port number pair (if connected). "St" is the inter‐
2708 nal status of the socket. The "tx_queue" and "rx_queue" are the
2709 outgoing and incoming data queue in terms of kernel memory
2710 usage. The "tr", "tm->when", and "rexmits" fields are not used
2711 by UDP. The "uid" field holds the effective UID of the creator
2712 of the socket. The format is:
2713 sl local_address rem_address st tx_queue rx_queue tr rexmits tm->when uid
2715 1: 01642C89:0201 0C642C89:03FF 01 00000000:00000001 01:000071BA 00000000 0
2716 1: 00000000:0801 00000000:0000 0A 00000000:00000000 00:00000000 6F000100 0
2717 1: 00000000:0201 00000000:0000 0A 00000000:00000000 00:00000000 00000000 0
2718 /proc/net/unix
2720 Lists the UNIX domain sockets present within the system and
2721 their status. The format is:
2722 Num RefCount Protocol Flags Type St Inode Path
2724 0: 00000002 00000000 00000000 0001 03 42
2725 1: 00000001 00000000 00010000 0001 01 1948 /dev/printer
2726 The fields are as follows:
2728 Num: the kernel table slot number.
2730 RefCount: the number of users of the socket.
2732 Protocol: currently always 0.
2734 Flags: the internal kernel flags holding the status of the
2736 socket.
2737 Type: the socket type. For SOCK_STREAM sockets, this is
2739 0001; for SOCK_DGRAM sockets, it is 0002; and for
2740 SOCK_SEQPACKET sockets, it is 0005.
2741 St: the internal state of the socket.
2743 Inode: the inode number of the socket.
2745 Path: the bound pathname (if any) of the socket. Sockets in
2747 the abstract namespace are included in the list, and
2748 are shown with a Path that commences with the charac‐
2749 ter '@'.
2750 /proc/net/netfilter/nfnetlink_queue
2752 This file contains information about netfilter user-space queue‐
2753 ing, if used. Each line represents a queue. Queues that have
2754 not been subscribed to by user space are not shown.
2755 1 4207 0 2 65535 0 0 0 1
2757 (1) (2) (3)(4) (5) (6) (7) (8)
2758 The fields in each line are:
2760 (1) The ID of the queue. This matches what is specified in the
2762 --queue-num or --queue-balance options to the iptables(8)
2763 NFQUEUE target. See iptables-extensions(8) for more infor‐
2764 mation.
2765 (2) The netlink port ID subscribed to the queue.
2767 (3) The number of packets currently queued and waiting to be
2769 processed by the application.
2770 (4) The copy mode of the queue. It is either 1 (metadata only)
2772 or 2 (also copy payload data to user space).
2773 (5) Copy range; that is, how many bytes of packet payload
2775 should be copied to user space at most.
2776 (6) queue dropped. Number of packets that had to be dropped by
2778 the kernel because too many packets are already waiting for
2779 user space to send back the mandatory accept/drop verdicts.
2780 (7) queue user dropped. Number of packets that were dropped
2782 within the netlink subsystem. Such drops usually happen
2783 when the corresponding socket buffer is full; that is, user
2784 space is not able to read messages fast enough.
2785 (8) sequence number. Every queued packet is associated with a
2787 (32-bit) monotonically-increasing sequence number. This
2788 shows the ID of the most recent packet queued.
2789 The last number exists only for compatibility reasons and is
2791 always 1.
2792 /proc/partitions
2794 Contains the major and minor numbers of each partition as well
2795 as the number of 1024-byte blocks and the partition name.
2796 /proc/pci
2798 This is a listing of all PCI devices found during kernel ini‐
2799 tialization and their configuration.
2800 This file has been deprecated in favor of a new /proc interface
2802 for PCI (/proc/bus/pci). It became optional in Linux 2.2
2803 (available with CONFIG_PCI_OLD_PROC set at kernel compilation).
2804 It became once more nonoptionally enabled in Linux 2.4. Next,
2805 it was deprecated in Linux 2.6 (still available with CON‐
2806 FIG_PCI_LEGACY_PROC set), and finally removed altogether since
2807 Linux 2.6.17.
2808 /proc/profile (since Linux 2.4)
2810 This file is present only if the kernel was booted with the pro‐
2811 file=1 command-line option. It exposes kernel profiling infor‐
2812 mation in a binary format for use by readprofile(1). Writing
2813 (e.g., an empty string) to this file resets the profiling coun‐
2814 ters; on some architectures, writing a binary integer "profiling
2815 multiplier" of size sizeof(int) sets the profiling interrupt
2816 frequency.
2817 /proc/scsi
2819 A directory with the scsi mid-level pseudo-file and various SCSI
2820 low-level driver directories, which contain a file for each SCSI
2821 host in this system, all of which give the status of some part
2822 of the SCSI IO subsystem. These files contain ASCII structures
2823 and are, therefore, readable with cat(1).
2824 You can also write to some of the files to reconfigure the sub‐
2826 system or switch certain features on or off.
2827 /proc/scsi/scsi
2829 This is a listing of all SCSI devices known to the kernel. The
2830 listing is similar to the one seen during bootup. scsi cur‐
2831 rently supports only the add-single-device command which allows
2832 root to add a hotplugged device to the list of known devices.
2833 The command
2835 echo 'scsi add-single-device 1 0 5 0' > /proc/scsi/scsi
2837 will cause host scsi1 to scan on SCSI channel 0 for a device on
2839 ID 5 LUN 0. If there is already a device known on this address
2840 or the address is invalid, an error will be returned.
2841 /proc/scsi/[drivername]
2843 [drivername] can currently be NCR53c7xx, aha152x, aha1542,
2844 aha1740, aic7xxx, buslogic, eata_dma, eata_pio, fdomain, in2000,
2845 pas16, qlogic, scsi_debug, seagate, t128, u15-24f, ultrastore,
2846 or wd7000. These directories show up for all drivers that reg‐
2847 istered at least one SCSI HBA. Every directory contains one
2848 file per registered host. Every host-file is named after the
2849 number the host was assigned during initialization.
2850 Reading these files will usually show driver and host configura‐
2852 tion, statistics, and so on.
2853 Writing to these files allows different things on different
2855 hosts. For example, with the latency and nolatency commands,
2856 root can switch on and off command latency measurement code in
2857 the eata_dma driver. With the lockup and unlock commands, root
2858 can control bus lockups simulated by the scsi_debug driver.
2859 /proc/self
2861 This directory refers to the process accessing the /proc
2862 filesystem, and is identical to the /proc directory named by the
2863 process ID of the same process.
2864 /proc/slabinfo
2866 Information about kernel caches. See slabinfo(5) for details.
2867 /proc/stat
2869 kernel/system statistics. Varies with architecture. Common
2870 entries include:
2871 cpu 10132153 290696 3084719 46828483 16683 0 25195 0 175628 0
2873 cpu0 1393280 32966 572056 13343292 6130 0 17875 0 23933 0
2874 The amount of time, measured in units of USER_HZ
2875 (1/100ths of a second on most architectures, use
2876 sysconf(_SC_CLK_TCK) to obtain the right value), that the
2877 system ("cpu" line) or the specific CPU ("cpuN" line)
2878 spent in various states:
2879 user (1) Time spent in user mode.
2881 nice (2) Time spent in user mode with low priority
2883 (nice).
2884 system (3) Time spent in system mode.
2886 idle (4) Time spent in the idle task. This value
2888 should be USER_HZ times the second entry in the
2889 /proc/uptime pseudo-file.
2890 iowait (since Linux 2.5.41)
2892 (5) Time waiting for I/O to complete. This value
2893 is not reliable, for the following reasons:
2894 1. The CPU will not wait for I/O to complete;
2896 iowait is the time that a task is waiting for
2897 I/O to complete. When a CPU goes into idle
2898 state for outstanding task I/O, another task
2899 will be scheduled on this CPU.
2900 2. On a multi-core CPU, the task waiting for I/O
2902 to complete is not running on any CPU, so the
2903 iowait of each CPU is difficult to calculate.
2904 3. The value in this field may decrease in certain
2906 conditions.
2907 irq (since Linux 2.6.0)
2909 (6) Time servicing interrupts.
2910 softirq (since Linux 2.6.0)
2912 (7) Time servicing softirqs.
2913 steal (since Linux 2.6.11)
2915 (8) Stolen time, which is the time spent in other
2916 operating systems when running in a virtualized
2917 environment
2918 guest (since Linux 2.6.24)
2920 (9) Time spent running a virtual CPU for guest
2921 operating systems under the control of the Linux
2922 kernel.
2923 guest_nice (since Linux 2.6.33)
2925 (10) Time spent running a niced guest (virtual CPU
2926 for guest operating systems under the control of
2927 the Linux kernel).
2928 page 5741 1808
2930 The number of pages the system paged in and the number
2931 that were paged out (from disk).
2932 swap 1 0
2934 The number of swap pages that have been brought in and
2935 out.
2936 intr 1462898
2938 This line shows counts of interrupts serviced since boot
2939 time, for each of the possible system interrupts. The
2940 first column is the total of all interrupts serviced
2941 including unnumbered architecture specific interrupts;
2942 each subsequent column is the total for that particular
2943 numbered interrupt. Unnumbered interrupts are not shown,
2944 only summed into the total.
2945 disk_io: (2,0):(31,30,5764,1,2) (3,0):...
2947 (major,disk_idx):(noinfo, read_io_ops, blks_read,
2948 write_io_ops, blks_written)
2949 (Linux 2.4 only)
2950 ctxt 115315
2952 The number of context switches that the system underwent.
2953 btime 769041601
2955 boot time, in seconds since the Epoch, 1970-01-01
2956 00:00:00 +0000 (UTC).
2957 processes 86031
2959 Number of forks since boot.
2960 procs_running 6
2962 Number of processes in runnable state. (Linux 2.5.45
2963 onward.)
2964 procs_blocked 2
2966 Number of processes blocked waiting for I/O to complete.
2967 (Linux 2.5.45 onward.)
2968 softirq 229245889 94 60001584 13619 5175704 2471304 28 51212741
2970 59130143 0 51240672
2971 This line shows the number of softirq for all CPUs. The
2972 first column is the total of all softirqs and each subse‐
2973 quent column is the total for particular softirq. (Linux
2974 2.6.31 onward.)
2975 /proc/swaps
2977 Swap areas in use. See also swapon(8).
2978 /proc/sys
2980 This directory (present since 1.3.57) contains a number of files
2981 and subdirectories corresponding to kernel variables. These
2982 variables can be read and sometimes modified using the /proc
2983 filesystem, and the (deprecated) sysctl(2) system call.
2984 String values may be terminated by either '\0' or '\n'.
2986 Integer and long values may be written either in decimal or in
2988 hexadecimal notation (e.g. 0x3FFF). When writing multiple inte‐
2989 ger or long values, these may be separated by any of the follow‐
2990 ing whitespace characters: ' ', '\t', or '\n'. Using other sep‐
2991 arators leads to the error EINVAL.
2992 /proc/sys/abi (since Linux 2.4.10)
2994 This directory may contain files with application binary infor‐
2995 mation. See the Linux kernel source file Documenta‐
2996 tion/sysctl/abi.txt for more information.
2997 /proc/sys/debug
2999 This directory may be empty.
3000 /proc/sys/dev
3002 This directory contains device-specific information (e.g.,
3003 dev/cdrom/info). On some systems, it may be empty.
3004 /proc/sys/fs
3006 This directory contains the files and subdirectories for kernel
3007 variables related to filesystems.
3008 /proc/sys/fs/aio-max-nr and /proc/sys/fs/aio-nr (since Linux 2.6.4)
3010 aio-nr is the running total of the number of events specified by
3011 io_setup(2) calls for all currently active AIO contexts. If
3012 aio-nr reaches aio-max-nr, then io_setup(2) will fail with the
3013 error EAGAIN. Raising aio-max-nr does not result in the preal‐
3014 location or resizing of any kernel data structures.
3015 /proc/sys/fs/binfmt_misc
3017 Documentation for files in this directory can be found in the
3018 Linux kernel source in the file Documenta‐
3019 tion/admin-guide/binfmt-misc.rst (or in Documenta‐
3020 tion/binfmt_misc.txt on older kernels).
3021 /proc/sys/fs/dentry-state (since Linux 2.2)
3023 This file contains information about the status of the directory
3024 cache (dcache). The file contains six numbers, nr_dentry,
3025 nr_unused, age_limit (age in seconds), want_pages (pages
3026 requested by system) and two dummy values.
3027 * nr_dentry is the number of allocated dentries (dcache
3029 entries). This field is unused in Linux 2.2.
3030 * nr_unused is the number of unused dentries.
3032 * age_limit is the age in seconds after which dcache entries can
3034 be reclaimed when memory is short.
3035 * want_pages is nonzero when the kernel has called
3037 shrink_dcache_pages() and the dcache isn't pruned yet.
3038 /proc/sys/fs/dir-notify-enable
3040 This file can be used to disable or enable the dnotify interface
3041 described in fcntl(2) on a system-wide basis. A value of 0 in
3042 this file disables the interface, and a value of 1 enables it.
3043 /proc/sys/fs/dquot-max
3045 This file shows the maximum number of cached disk quota entries.
3046 On some (2.4) systems, it is not present. If the number of free
3047 cached disk quota entries is very low and you have some awesome
3048 number of simultaneous system users, you might want to raise the
3049 limit.
3050 /proc/sys/fs/dquot-nr
3052 This file shows the number of allocated disk quota entries and
3053 the number of free disk quota entries.
3054 /proc/sys/fs/epoll (since Linux 2.6.28)
3056 This directory contains the file max_user_watches, which can be
3057 used to limit the amount of kernel memory consumed by the epoll
3058 interface. For further details, see epoll(7).
3059 /proc/sys/fs/file-max
3061 This file defines a system-wide limit on the number of open
3062 files for all processes. System calls that fail when encounter‐
3063 ing this limit fail with the error ENFILE. (See also setr‐
3064 limit(2), which can be used by a process to set the per-process
3065 limit, RLIMIT_NOFILE, on the number of files it may open.) If
3066 you get lots of error messages in the kernel log about running
3067 out of file handles (look for "VFS: file-max limit <number>
3068 reached"), try increasing this value:
3069 echo 100000 > /proc/sys/fs/file-max
3071 Privileged processes (CAP_SYS_ADMIN) can override the file-max
3073 limit.
3074 /proc/sys/fs/file-nr
3076 This (read-only) file contains three numbers: the number of
3077 allocated file handles (i.e., the number of files presently
3078 opened); the number of free file handles; and the maximum number
3079 of file handles (i.e., the same value as /proc/sys/fs/file-max).
3080 If the number of allocated file handles is close to the maximum,
3081 you should consider increasing the maximum. Before Linux 2.6,
3082 the kernel allocated file handles dynamically, but it didn't
3083 free them again. Instead the free file handles were kept in a
3084 list for reallocation; the "free file handles" value indicates
3085 the size of that list. A large number of free file handles
3086 indicates that there was a past peak in the usage of open file
3087 handles. Since Linux 2.6, the kernel does deallocate freed file
3088 handles, and the "free file handles" value is always zero.
3089 /proc/sys/fs/inode-max (only present until Linux 2.2)
3091 This file contains the maximum number of in-memory inodes. This
3092 value should be 3–4 times larger than the value in file-max,
3093 since stdin, stdout and network sockets also need an inode to
3094 handle them. When you regularly run out of inodes, you need to
3095 increase this value.
3096 Starting with Linux 2.4, there is no longer a static limit on
3098 the number of inodes, and this file is removed.
3099 /proc/sys/fs/inode-nr
3101 This file contains the first two values from inode-state.
3102 /proc/sys/fs/inode-state
3104 This file contains seven numbers: nr_inodes, nr_free_inodes,
3105 preshrink, and four dummy values (always zero).
3106 nr_inodes is the number of inodes the system has allocated.
3108 nr_free_inodes represents the number of free inodes.
3109 preshrink is nonzero when the nr_inodes > inode-max and the sys‐
3111 tem needs to prune the inode list instead of allocating more;
3112 since Linux 2.4, this field is a dummy value (always zero).
3113 /proc/sys/fs/inotify (since Linux 2.6.13)
3115 This directory contains files max_queued_events,
3116 max_user_instances, and max_user_watches, that can be used to
3117 limit the amount of kernel memory consumed by the inotify inter‐
3118 face. For further details, see inotify(7).
3119 /proc/sys/fs/lease-break-time
3121 This file specifies the grace period that the kernel grants to a
3122 process holding a file lease (fcntl(2)) after it has sent a sig‐
3123 nal to that process notifying it that another process is waiting
3124 to open the file. If the lease holder does not remove or down‐
3125 grade the lease within this grace period, the kernel forcibly
3126 breaks the lease.
3127 /proc/sys/fs/leases-enable
3129 This file can be used to enable or disable file leases
3130 (fcntl(2)) on a system-wide basis. If this file contains the
3131 value 0, leases are disabled. A nonzero value enables leases.
3132 /proc/sys/fs/mount-max (since Linux 4.9)
3134 The value in this file specifies the maximum number of mounts
3135 that may exist in a mount namespace. The default value in this
3136 file is 100,000.
3137 /proc/sys/fs/mqueue (since Linux 2.6.6)
3139 This directory contains files msg_max, msgsize_max, and
3140 queues_max, controlling the resources used by POSIX message
3141 queues. See mq_overview(7) for details.
3142 /proc/sys/fs/nr_open (since Linux 2.6.25)
3144 This file imposes a ceiling on the value to which the
3145 RLIMIT_NOFILE resource limit can be raised (see getrlimit(2)).
3146 This ceiling is enforced for both unprivileged and privileged
3147 process. The default value in this file is 1048576. (Before
3148 Linux 2.6.25, the ceiling for RLIMIT_NOFILE was hard-coded to
3149 the same value.)
3150 /proc/sys/fs/overflowgid and /proc/sys/fs/overflowuid
3152 These files allow you to change the value of the fixed UID and
3153 GID. The default is 65534. Some filesystems support only
3154 16-bit UIDs and GIDs, although in Linux UIDs and GIDs are 32
3155 bits. When one of these filesystems is mounted with writes
3156 enabled, any UID or GID that would exceed 65535 is translated to
3157 the overflow value before being written to disk.
3158 /proc/sys/fs/pipe-max-size (since Linux 2.6.35)
3160 See pipe(7).
3161 /proc/sys/fs/pipe-user-pages-hard (since Linux 4.5)
3163 See pipe(7).
3164 /proc/sys/fs/pipe-user-pages-soft (since Linux 4.5)
3166 See pipe(7).
3167 /proc/sys/fs/protected_fifos (since Linux 4.19)
3169 The value in this file is/can be set to one of the following:
3170 0 Writing to FIFOs is unrestricted.
3172 1 Don't allow O_CREAT open(2) on FIFOs that the caller doesn't
3174 own in world-writable sticky directories, unless the FIFO is
3175 owned by the owner of the directory.
3176 2 As for the value 1, but the restriction also applies to
3178 group-writable sticky directories.
3179 The intent of the above protections is to avoid unintentional
3181 writes to an attacker-controlled FIFO when a program expected to
3182 create a regular file.
3183 /proc/sys/fs/protected_hardlinks (since Linux 3.6)
3185 When the value in this file is 0, no restrictions are placed on
3186 the creation of hard links (i.e., this is the historical behav‐
3187 ior before Linux 3.6). When the value in this file is 1, a hard
3188 link can be created to a target file only if one of the follow‐
3189 ing conditions is true:
3190 * The calling process has the CAP_FOWNER capability in its user
3192 namespace and the file UID has a mapping in the namespace.
3193 * The filesystem UID of the process creating the link matches
3195 the owner (UID) of the target file (as described in creden‐
3196 tials(7), a process's filesystem UID is normally the same as
3197 its effective UID).
3198 * All of the following conditions are true:
3200 · the target is a regular file;
3202 · the target file does not have its set-user-ID mode bit
3204 enabled;
3205 · the target file does not have both its set-group-ID and
3207 group-executable mode bits enabled; and
3208 · the caller has permission to read and write the target
3210 file (either via the file's permissions mask or because
3211 it has suitable capabilities).
3212 The default value in this file is 0. Setting the value to 1
3214 prevents a longstanding class of security issues caused by hard-
3215 link-based time-of-check, time-of-use races, most commonly seen
3216 in world-writable directories such as /tmp. The common method
3217 of exploiting this flaw is to cross privilege boundaries when
3218 following a given hard link (i.e., a root process follows a hard
3219 link created by another user). Additionally, on systems without
3220 separated partitions, this stops unauthorized users from "pin‐
3221 ning" vulnerable set-user-ID and set-group-ID files against
3222 being upgraded by the administrator, or linking to special
3223 files.
3224 /proc/sys/fs/protected_regular (since Linux 4.19)
3226 The value in this file is/can be set to one of the following:
3227 0 Writing to regular files is unrestricted.
3229 1 Don't allow O_CREAT open(2) on regular files that the caller
3231 doesn't own in world-writable sticky directories, unless the
3232 regular file is owned by the owner of the directory.
3233 2 As for the value 1, but the restriction also applies to
3235 group-writable sticky directories.
3236 The intent of the above protections is similar to pro‐
3238 tected_fifos, but allows an application to avoid writes to an
3239 attacker-controlled regular file, where the application expected
3240 to create one.
3241 /proc/sys/fs/protected_symlinks (since Linux 3.6)
3243 When the value in this file is 0, no restrictions are placed on
3244 following symbolic links (i.e., this is the historical behavior
3245 before Linux 3.6). When the value in this file is 1, symbolic
3246 links are followed only in the following circumstances:
3247 * the filesystem UID of the process following the link matches
3249 the owner (UID) of the symbolic link (as described in creden‐
3250 tials(7), a process's filesystem UID is normally the same as
3251 its effective UID);
3252 * the link is not in a sticky world-writable directory; or
3254 * the symbolic link and its parent directory have the same
3256 owner (UID)
3257 A system call that fails to follow a symbolic link because of
3259 the above restrictions returns the error EACCES in errno.
3260 The default value in this file is 0. Setting the value to 1
3262 avoids a longstanding class of security issues based on time-of-
3263 check, time-of-use races when accessing symbolic links.
3264 /proc/sys/fs/suid_dumpable (since Linux 2.6.13)
3266 The value in this file is assigned to a process's "dumpable"
3267 flag in the circumstances described in prctl(2). In effect, the
3268 value in this file determines whether core dump files are pro‐
3269 duced for set-user-ID or otherwise protected/tainted binaries.
3270 The "dumpable" setting also affects the ownership of files in a
3271 process's /proc/[pid] directory, as described above.
3272 Three different integer values can be specified:
3274 0 (default)
3276 This provides the traditional (pre-Linux 2.6.13) behav‐
3277 ior. A core dump will not be produced for a process
3278 which has changed credentials (by calling seteuid(2),
3279 setgid(2), or similar, or by executing a set-user-ID or
3280 set-group-ID program) or whose binary does not have read
3281 permission enabled.
3282 1 ("debug")
3284 All processes dump core when possible. (Reasons why a
3285 process might nevertheless not dump core are described in
3286 core(5).) The core dump is owned by the filesystem user
3287 ID of the dumping process and no security is applied.
3288 This is intended for system debugging situations only:
3289 this mode is insecure because it allows unprivileged
3290 users to examine the memory contents of privileged pro‐
3291 cesses.
3292 2 ("suidsafe")
3294 Any binary which normally would not be dumped (see "0"
3295 above) is dumped readable by root only. This allows the
3296 user to remove the core dump file but not to read it.
3297 For security reasons core dumps in this mode will not
3298 overwrite one another or other files. This mode is
3299 appropriate when administrators are attempting to debug
3300 problems in a normal environment.
3301 Additionally, since Linux 3.6, /proc/sys/kernel/core_pat‐
3303 tern must either be an absolute pathname or a pipe com‐
3304 mand, as detailed in core(5). Warnings will be written
3305 to the kernel log if core_pattern does not follow these
3306 rules, and no core dump will be produced.
3307 For details of the effect of a process's "dumpable" setting on
3309 ptrace access mode checking, see ptrace(2).
3310 /proc/sys/fs/super-max
3312 This file controls the maximum number of superblocks, and thus
3313 the maximum number of mounted filesystems the kernel can have.
3314 You need increase only super-max if you need to mount more
3315 filesystems than the current value in super-max allows you to.
3316 /proc/sys/fs/super-nr
3318 This file contains the number of filesystems currently mounted.
3319 /proc/sys/kernel
3321 This directory contains files controlling a range of kernel
3322 parameters, as described below.
3323 /proc/sys/kernel/acct
3325 This file contains three numbers: highwater, lowwater, and fre‐
3326 quency. If BSD-style process accounting is enabled, these val‐
3327 ues control its behavior. If free space on filesystem where the
3328 log lives goes below lowwater percent, accounting suspends. If
3329 free space gets above highwater percent, accounting resumes.
3330 frequency determines how often the kernel checks the amount of
3331 free space (value is in seconds). Default values are 4, 2 and
3332 30. That is, suspend accounting if 2% or less space is free;
3333 resume it if 4% or more space is free; consider information
3334 about amount of free space valid for 30 seconds.
3335 /proc/sys/kernel/auto_msgmni (Linux 2.6.27 to 3.18)
3337 From Linux 2.6.27 to 3.18, this file was used to control recom‐
3338 puting of the value in /proc/sys/kernel/msgmni upon the addition
3339 or removal of memory or upon IPC namespace creation/removal.
3340 Echoing "1" into this file enabled msgmni automatic recomputing
3341 (and triggered a recomputation of msgmni based on the current
3342 amount of available memory and number of IPC namespaces). Echo‐
3343 ing "0" disabled automatic recomputing. (Automatic recomputing
3344 was also disabled if a value was explicitly assigned to
3345 /proc/sys/kernel/msgmni.) The default value in auto_msgmni was
3346 1.
3347 Since Linux 3.19, the content of this file has no effect
3349 (because msgmni defaults to near the maximum value possible),
3350 and reads from this file always return the value "0".
3351 /proc/sys/kernel/cap_last_cap (since Linux 3.2)
3353 See capabilities(7).
3354 /proc/sys/kernel/cap-bound (from Linux 2.2 to 2.6.24)
3356 This file holds the value of the kernel capability bounding set
3357 (expressed as a signed decimal number). This set is ANDed
3358 against the capabilities permitted to a process during
3359 execve(2). Starting with Linux 2.6.25, the system-wide capabil‐
3360 ity bounding set disappeared, and was replaced by a per-thread
3361 bounding set; see capabilities(7).
3362 /proc/sys/kernel/core_pattern
3364 See core(5).
3365 /proc/sys/kernel/core_pipe_limit
3367 See core(5).
3368 /proc/sys/kernel/core_uses_pid
3370 See core(5).
3371 /proc/sys/kernel/ctrl-alt-del
3373 This file controls the handling of Ctrl-Alt-Del from the key‐
3374 board. When the value in this file is 0, Ctrl-Alt-Del is
3375 trapped and sent to the init(1) program to handle a graceful
3376 restart. When the value is greater than zero, Linux's reaction
3377 to a Vulcan Nerve Pinch (tm) will be an immediate reboot, with‐
3378 out even syncing its dirty buffers. Note: when a program (like
3379 dosemu) has the keyboard in "raw" mode, the ctrl-alt-del is
3380 intercepted by the program before it ever reaches the kernel tty
3381 layer, and it's up to the program to decide what to do with it.
3382 /proc/sys/kernel/dmesg_restrict (since Linux 2.6.37)
3384 The value in this file determines who can see kernel syslog con‐
3385 tents. A value of 0 in this file imposes no restrictions. If
3386 the value is 1, only privileged users can read the kernel sys‐
3387 log. (See syslog(2) for more details.) Since Linux 3.4, only
3388 users with the CAP_SYS_ADMIN capability may change the value in
3389 this file.
3390 /proc/sys/kernel/domainname and /proc/sys/kernel/hostname
3392 can be used to set the NIS/YP domainname and the hostname of
3393 your box in exactly the same way as the commands domainname(1)
3394 and hostname(1), that is:
3395 # echo 'darkstar' > /proc/sys/kernel/hostname
3397 # echo 'mydomain' > /proc/sys/kernel/domainname
3398 has the same effect as
3400 # hostname 'darkstar'
3402 # domainname 'mydomain'
3403 Note, however, that the classic darkstar.frop.org has the host‐
3405 name "darkstar" and DNS (Internet Domain Name Server) domainname
3406 "frop.org", not to be confused with the NIS (Network Information
3407 Service) or YP (Yellow Pages) domainname. These two domain
3408 names are in general different. For a detailed discussion see
3409 the hostname(1) man page.
3410 /proc/sys/kernel/hotplug
3412 This file contains the pathname for the hotplug policy agent.
3413 The default value in this file is /sbin/hotplug.
3414 /proc/sys/kernel/htab-reclaim (before Linux 2.4.9.2)
3416 (PowerPC only) If this file is set to a nonzero value, the Pow‐
3417 erPC htab (see kernel file Documentation/powerpc/ppc_htab.txt)
3418 is pruned each time the system hits the idle loop.
3419 /proc/sys/kernel/keys/*
3421 This directory contains various files that define parameters and
3422 limits for the key-management facility. These files are
3423 described in keyrings(7).
3424 /proc/sys/kernel/kptr_restrict (since Linux 2.6.38)
3426 The value in this file determines whether kernel addresses are
3427 exposed via /proc files and other interfaces. A value of 0 in
3428 this file imposes no restrictions. If the value is 1, kernel
3429 pointers printed using the %pK format specifier will be replaced
3430 with zeros unless the user has the CAP_SYSLOG capability. If
3431 the value is 2, kernel pointers printed using the %pK format
3432 specifier will be replaced with zeros regardless of the user's
3433 capabilities. The initial default value for this file was 1,
3434 but the default was changed to 0 in Linux 2.6.39. Since Linux
3435 3.4, only users with the CAP_SYS_ADMIN capability can change the
3436 value in this file.
3437 /proc/sys/kernel/l2cr
3439 (PowerPC only) This file contains a flag that controls the L2
3440 cache of G3 processor boards. If 0, the cache is disabled.
3441 Enabled if nonzero.
3442 /proc/sys/kernel/modprobe
3444 This file contains the pathname for the kernel module loader.
3445 The default value is /sbin/modprobe. The file is present only
3446 if the kernel is built with the CONFIG_MODULES (CONFIG_KMOD in
3447 Linux 2.6.26 and earlier) option enabled. It is described by
3448 the Linux kernel source file Documentation/kmod.txt (present
3449 only in kernel 2.4 and earlier).
3450 /proc/sys/kernel/modules_disabled (since Linux 2.6.31)
3452 A toggle value indicating if modules are allowed to be loaded in
3453 an otherwise modular kernel. This toggle defaults to off (0),
3454 but can be set true (1). Once true, modules can be neither
3455 loaded nor unloaded, and the toggle cannot be set back to false.
3456 The file is present only if the kernel is built with the CON‐
3457 FIG_MODULES option enabled.
3458 /proc/sys/kernel/msgmax (since Linux 2.2)
3460 This file defines a system-wide limit specifying the maximum
3461 number of bytes in a single message written on a System V mes‐
3462 sage queue.
3463 /proc/sys/kernel/msgmni (since Linux 2.4)
3465 This file defines the system-wide limit on the number of message
3466 queue identifiers. See also /proc/sys/kernel/auto_msgmni.
3467 /proc/sys/kernel/msgmnb (since Linux 2.2)
3469 This file defines a system-wide parameter used to initialize the
3470 msg_qbytes setting for subsequently created message queues. The
3471 msg_qbytes setting specifies the maximum number of bytes that
3472 may be written to the message queue.
3473 /proc/sys/kernel/ngroups_max (since Linux 2.6.4)
3475 This is a read-only file that displays the upper limit on the
3476 number of a process's group memberships.
3477 /proc/sys/kernel/ns_last_pid (since Linux 3.3)
3479 See pid_namespaces(7).
3480 /proc/sys/kernel/ostype and /proc/sys/kernel/osrelease
3482 These files give substrings of /proc/version.
3483 /proc/sys/kernel/overflowgid and /proc/sys/kernel/overflowuid
3485 These files duplicate the files /proc/sys/fs/overflowgid and
3486 /proc/sys/fs/overflowuid.
3487 /proc/sys/kernel/panic
3489 This file gives read/write access to the kernel variable
3490 panic_timeout. If this is zero, the kernel will loop on a
3491 panic; if nonzero, it indicates that the kernel should autore‐
3492 boot after this number of seconds. When you use the software
3493 watchdog device driver, the recommended setting is 60.
3494 /proc/sys/kernel/panic_on_oops (since Linux 2.5.68)
3496 This file controls the kernel's behavior when an oops or BUG is
3497 encountered. If this file contains 0, then the system tries to
3498 continue operation. If it contains 1, then the system delays a
3499 few seconds (to give klogd time to record the oops output) and
3500 then panics. If the /proc/sys/kernel/panic file is also
3501 nonzero, then the machine will be rebooted.
3502 /proc/sys/kernel/pid_max (since Linux 2.5.34)
3504 This file specifies the value at which PIDs wrap around (i.e.,
3505 the value in this file is one greater than the maximum PID).
3506 PIDs greater than this value are not allocated; thus, the value
3507 in this file also acts as a system-wide limit on the total num‐
3508 ber of processes and threads. The default value for this file,
3509 32768, results in the same range of PIDs as on earlier kernels.
3510 On 32-bit platforms, 32768 is the maximum value for pid_max. On
3511 64-bit systems, pid_max can be set to any value up to 2^22
3512 (PID_MAX_LIMIT, approximately 4 million).
3513 /proc/sys/kernel/powersave-nap (PowerPC only)
3515 This file contains a flag. If set, Linux-PPC will use the "nap"
3516 mode of powersaving, otherwise the "doze" mode will be used.
3517 /proc/sys/kernel/printk
3519 See syslog(2).
3520 /proc/sys/kernel/pty (since Linux 2.6.4)
3522 This directory contains two files relating to the number of UNIX
3523 98 pseudoterminals (see pts(4)) on the system.
3524 /proc/sys/kernel/pty/max
3526 This file defines the maximum number of pseudoterminals.
3527 /proc/sys/kernel/pty/nr
3529 This read-only file indicates how many pseudoterminals are cur‐
3530 rently in use.
3531 /proc/sys/kernel/random
3533 This directory contains various parameters controlling the oper‐
3534 ation of the file /dev/random. See random(4) for further infor‐
3535 mation.
3536 /proc/sys/kernel/random/uuid (since Linux 2.4)
3538 Each read from this read-only file returns a randomly generated
3539 128-bit UUID, as a string in the standard UUID format.
3540 /proc/sys/kernel/randomize_va_space (since Linux 2.6.12)
3542 Select the address space layout randomization (ASLR) policy for
3543 the system (on architectures that support ASLR). Three values
3544 are supported for this file:
3545 0 Turn ASLR off. This is the default for architectures that
3547 don't support ASLR, and when the kernel is booted with the
3548 norandmaps parameter.
3549 1 Make the addresses of mmap(2) allocations, the stack, and the
3551 VDSO page randomized. Among other things, this means that
3552 shared libraries will be loaded at randomized addresses. The
3553 text segment of PIE-linked binaries will also be loaded at a
3554 randomized address. This value is the default if the kernel
3555 was configured with CONFIG_COMPAT_BRK.
3556 2 (Since Linux 2.6.25) Also support heap randomization. This
3558 value is the default if the kernel was not configured with
3559 CONFIG_COMPAT_BRK.
3560 /proc/sys/kernel/real-root-dev
3562 This file is documented in the Linux kernel source file Documen‐
3563 tation/admin-guide/initrd.rst (or Documentation/initrd.txt
3564 before Linux 4.10).
3565 /proc/sys/kernel/reboot-cmd (Sparc only)
3567 This file seems to be a way to give an argument to the SPARC
3568 ROM/Flash boot loader. Maybe to tell it what to do after
3569 rebooting?
3570 /proc/sys/kernel/rtsig-max
3572 (Only in kernels up to and including 2.6.7; see setrlimit(2))
3573 This file can be used to tune the maximum number of POSIX real-
3574 time (queued) signals that can be outstanding in the system.
3575 /proc/sys/kernel/rtsig-nr
3577 (Only in kernels up to and including 2.6.7.) This file shows
3578 the number of POSIX real-time signals currently queued.
3579 /proc/[pid]/sched_autogroup_enabled (since Linux 2.6.38)
3581 See sched(7).
3582 /proc/sys/kernel/sched_child_runs_first (since Linux 2.6.23)
3584 If this file contains the value zero, then, after a fork(2), the
3585 parent is first scheduled on the CPU. If the file contains a
3586 nonzero value, then the child is scheduled first on the CPU.
3587 (Of course, on a multiprocessor system, the parent and the child
3588 might both immediately be scheduled on a CPU.)
3589 /proc/sys/kernel/sched_rr_timeslice_ms (since Linux 3.9)
3591 See sched_rr_get_interval(2).
3592 /proc/sys/kernel/sched_rt_period_us (since Linux 2.6.25)
3594 See sched(7).
3595 /proc/sys/kernel/sched_rt_runtime_us (since Linux 2.6.25)
3597 See sched(7).
3598 /proc/sys/kernel/seccomp (since Linux 4.14)
3600 This directory provides additional seccomp information and con‐
3601 figuration. See seccomp(2) for further details.
3602 /proc/sys/kernel/sem (since Linux 2.4)
3604 This file contains 4 numbers defining limits for System V IPC
3605 semaphores. These fields are, in order:
3606 SEMMSL The maximum semaphores per semaphore set.
3608 SEMMNS A system-wide limit on the number of semaphores in all
3610 semaphore sets.
3611 SEMOPM The maximum number of operations that may be specified
3613 in a semop(2) call.
3614 SEMMNI A system-wide limit on the maximum number of semaphore
3616 identifiers.
3617 /proc/sys/kernel/sg-big-buff
3619 This file shows the size of the generic SCSI device (sg) buffer.
3620 You can't tune it just yet, but you could change it at compile
3621 time by editing include/scsi/sg.h and changing the value of
3622 SG_BIG_BUFF. However, there shouldn't be any reason to change
3623 this value.
3624 /proc/sys/kernel/shm_rmid_forced (since Linux 3.1)
3626 If this file is set to 1, all System V shared memory segments
3627 will be marked for destruction as soon as the number of attached
3628 processes falls to zero; in other words, it is no longer possi‐
3629 ble to create shared memory segments that exist independently of
3630 any attached process.
3631 The effect is as though a shmctl(2) IPC_RMID is performed on all
3633 existing segments as well as all segments created in the future
3634 (until this file is reset to 0). Note that existing segments
3635 that are attached to no process will be immediately destroyed
3636 when this file is set to 1. Setting this option will also
3637 destroy segments that were created, but never attached, upon
3638 termination of the process that created the segment with
3639 shmget(2).
3640 Setting this file to 1 provides a way of ensuring that all Sys‐
3642 tem V shared memory segments are counted against the resource
3643 usage and resource limits (see the description of RLIMIT_AS in
3644 getrlimit(2)) of at least one process.
3645 Because setting this file to 1 produces behavior that is non‐
3647 standard and could also break existing applications, the default
3648 value in this file is 0. Set this file to 1 only if you have a
3649 good understanding of the semantics of the applications using
3650 System V shared memory on your system.
3651 /proc/sys/kernel/shmall (since Linux 2.2)
3653 This file contains the system-wide limit on the total number of
3654 pages of System V shared memory.
3655 /proc/sys/kernel/shmmax (since Linux 2.2)
3657 This file can be used to query and set the run-time limit on the
3658 maximum (System V IPC) shared memory segment size that can be
3659 created. Shared memory segments up to 1 GB are now supported in
3660 the kernel. This value defaults to SHMMAX.
3661 /proc/sys/kernel/shmmni (since Linux 2.4)
3663 This file specifies the system-wide maximum number of System V
3664 shared memory segments that can be created.
3665 /proc/sys/kernel/sysctl_writes_strict (since Linux 3.16)
3667 The value in this file determines how the file offset affects
3668 the behavior of updating entries in files under /proc/sys. The
3669 file has three possible values:
3670 -1 This provides legacy handling, with no printk warnings.
3672 Each write(2) must fully contain the value to be written,
3673 and multiple writes on the same file descriptor will over‐
3674 write the entire value, regardless of the file position.
3675 0 (default) This provides the same behavior as for -1, but
3677 printk warnings are written for processes that perform
3678 writes when the file offset is not 0.
3679 1 Respect the file offset when writing strings into /proc/sys
3681 files. Multiple writes will append to the value buffer.
3682 Anything written beyond the maximum length of the value buf‐
3683 fer will be ignored. Writes to numeric /proc/sys entries
3684 must always be at file offset 0 and the value must be fully
3685 contained in the buffer provided to write(2).
3686 /proc/sys/kernel/sysrq
3688 This file controls the functions allowed to be invoked by the
3689 SysRq key. By default, the file contains 1 meaning that every
3690 possible SysRq request is allowed (in older kernel versions,
3691 SysRq was disabled by default, and you were required to specifi‐
3692 cally enable it at run-time, but this is not the case any more).
3693 Possible values in this file are:
3694 0 Disable sysrq completely
3696 1 Enable all functions of sysrq
3698 > 1 Bit mask of allowed sysrq functions, as follows:
3700 2 Enable control of console logging level
3701 4 Enable control of keyboard (SAK, unraw)
3702 8 Enable debugging dumps of processes etc.
3703 16 Enable sync command
3704 32 Enable remount read-only
3705 64 Enable signaling of processes (term, kill, oom-kill)
3706 128 Allow reboot/poweroff
3707 256 Allow nicing of all real-time tasks
3708 This file is present only if the CONFIG_MAGIC_SYSRQ kernel con‐
3710 figuration option is enabled. For further details see the Linux
3711 kernel source file Documentation/admin-guide/sysrq.rst (or Docu‐
3712 mentation/sysrq.txt before Linux 4.10).
3713 /proc/sys/kernel/version
3715 This file contains a string such as:
3716 #5 Wed Feb 25 21:49:24 MET 1998
3718 The "#5" means that this is the fifth kernel built from this
3720 source base and the date following it indicates the time the
3721 kernel was built.
3722 /proc/sys/kernel/threads-max (since Linux 2.3.11)
3724 This file specifies the system-wide limit on the number of
3725 threads (tasks) that can be created on the system.
3726 Since Linux 4.1, the value that can be written to threads-max is
3728 bounded. The minimum value that can be written is 20. The max‐
3729 imum value that can be written is given by the constant
3730 FUTEX_TID_MASK [22m(0x3fffffff). If a value outside of this range
3731 is written to threads-max, the error EINVAL occurs.
3732 The value written is checked against the available RAM pages.
3734 If the thread structures would occupy too much (more than 1/8th)
3735 of the available RAM pages, threads-max is reduced accordingly.
3736 /proc/sys/kernel/yama/ptrace_scope (since Linux 3.5)
3738 See ptrace(2).
3739 /proc/sys/kernel/zero-paged (PowerPC only)
3741 This file contains a flag. When enabled (nonzero), Linux-PPC
3742 will pre-zero pages in the idle loop, possibly speeding up
3743 get_free_pages.
3744 /proc/sys/net
3746 This directory contains networking stuff. Explanations for some
3747 of the files under this directory can be found in tcp(7) and
3748 ip(7).
3749 /proc/sys/net/core/bpf_jit_enable
3751 See bpf(2).
3752 /proc/sys/net/core/somaxconn
3754 This file defines a ceiling value for the backlog argument of
3755 listen(2); see the listen(2) manual page for details.
3756 /proc/sys/proc
3758 This directory may be empty.
3759 /proc/sys/sunrpc
3761 This directory supports Sun remote procedure call for network
3762 filesystem (NFS). On some systems, it is not present.
3763 /proc/sys/user (since Linux 4.9)
3765 See namespaces(7).
3766 /proc/sys/vm
3768 This directory contains files for memory management tuning, buf‐
3769 fer and cache management.
3770 /proc/sys/vm/admin_reserve_kbytes (since Linux 3.10)
3772 This file defines the amount of free memory (in KiB) on the sys‐
3773 tem that should be reserved for users with the capability
3774 CAP_SYS_ADMIN.
3775 The default value in this file is the minimum of [3% of free
3777 pages, 8MiB] expressed as KiB. The default is intended to pro‐
3778 vide enough for the superuser to log in and kill a process, if
3779 necessary, under the default overcommit 'guess' mode (i.e., 0 in
3780 /proc/sys/vm/overcommit_memory).
3781 Systems running in "overcommit never" mode (i.e., 2 in
3783 /proc/sys/vm/overcommit_memory) should increase the value in
3784 this file to account for the full virtual memory size of the
3785 programs used to recover (e.g., login(1) ssh(1), and top(1))
3786 Otherwise, the superuser may not be able to log in to recover
3787 the system. For example, on x86-64 a suitable value is 131072
3788 (128MiB reserved).
3789 Changing the value in this file takes effect whenever an appli‐
3791 cation requests memory.
3792 /proc/sys/vm/compact_memory (since Linux 2.6.35)
3794 When 1 is written to this file, all zones are compacted such
3795 that free memory is available in contiguous blocks where possi‐
3796 ble. The effect of this action can be seen by examining
3797 /proc/buddyinfo.
3798 Present only if the kernel was configured with CONFIG_COM‐
3800 PACTION.
3801 /proc/sys/vm/drop_caches (since Linux 2.6.16)
3803 Writing to this file causes the kernel to drop clean caches,
3804 dentries, and inodes from memory, causing that memory to become
3805 free. This can be useful for memory management testing and per‐
3806 forming reproducible filesystem benchmarks. Because writing to
3807 this file causes the benefits of caching to be lost, it can
3808 degrade overall system performance.
3809 To free pagecache, use:
3811 echo 1 > /proc/sys/vm/drop_caches
3813 To free dentries and inodes, use:
3815 echo 2 > /proc/sys/vm/drop_caches
3817 To free pagecache, dentries and inodes, use:
3819 echo 3 > /proc/sys/vm/drop_caches
3821 Because writing to this file is a nondestructive operation and
3823 dirty objects are not freeable, the user should run sync(1)
3824 first.
3825 /proc/sys/vm/legacy_va_layout (since Linux 2.6.9)
3827 If nonzero, this disables the new 32-bit memory-mapping layout;
3828 the kernel will use the legacy (2.4) layout for all processes.
3829 /proc/sys/vm/memory_failure_early_kill (since Linux 2.6.32)
3831 Control how to kill processes when an uncorrected memory error
3832 (typically a 2-bit error in a memory module) that cannot be han‐
3833 dled by the kernel is detected in the background by hardware.
3834 In some cases (like the page still having a valid copy on disk),
3835 the kernel will handle the failure transparently without affect‐
3836 ing any applications. But if there is no other up-to-date copy
3837 of the data, it will kill processes to prevent any data corrup‐
3838 tions from propagating.
3839 The file has one of the following values:
3841 1: Kill all processes that have the corrupted-and-not-reload‐
3843 able page mapped as soon as the corruption is detected.
3844 Note that this is not supported for a few types of pages,
3845 such as kernel internally allocated data or the swap cache,
3846 but works for the majority of user pages.
3847 0: Unmap the corrupted page from all processes and kill a
3849 process only if it tries to access the page.
3850 The kill is performed using a SIGBUS signal with si_code set to
3852 BUS_MCEERR_AO. Processes can handle this if they want to; see
3853 sigaction(2) for more details.
3854 This feature is active only on architectures/platforms with
3856 advanced machine check handling and depends on the hardware
3857 capabilities.
3858 Applications can override the memory_failure_early_kill setting
3860 individually with the prctl(2) PR_MCE_KILL operation.
3861 Present only if the kernel was configured with CONFIG_MEM‐
3863 ORY_FAILURE.
3864 /proc/sys/vm/memory_failure_recovery (since Linux 2.6.32)
3866 Enable memory failure recovery (when supported by the platform).
3867 1: Attempt recovery.
3869 0: Always panic on a memory failure.
3871 Present only if the kernel was configured with CONFIG_MEM‐
3873 ORY_FAILURE.
3874 /proc/sys/vm/oom_dump_tasks (since Linux 2.6.25)
3876 Enables a system-wide task dump (excluding kernel threads) to be
3877 produced when the kernel performs an OOM-killing. The dump
3878 includes the following information for each task (thread,
3879 process): thread ID, real user ID, thread group ID (process ID),
3880 virtual memory size, resident set size, the CPU that the task is
3881 scheduled on, oom_adj score (see the description of
3882 /proc/[pid]/oom_adj), and command name. This is helpful to
3883 determine why the OOM-killer was invoked and to identify the
3884 rogue task that caused it.
3885 If this contains the value zero, this information is suppressed.
3887 On very large systems with thousands of tasks, it may not be
3888 feasible to dump the memory state information for each one.
3889 Such systems should not be forced to incur a performance penalty
3890 in OOM situations when the information may not be desired.
3891 If this is set to nonzero, this information is shown whenever
3893 the OOM-killer actually kills a memory-hogging task.
3894 The default value is 0.
3896 /proc/sys/vm/oom_kill_allocating_task (since Linux 2.6.24)
3898 This enables or disables killing the OOM-triggering task in out-
3899 of-memory situations.
3900 If this is set to zero, the OOM-killer will scan through the
3902 entire tasklist and select a task based on heuristics to kill.
3903 This normally selects a rogue memory-hogging task that frees up
3904 a large amount of memory when killed.
3905 If this is set to nonzero, the OOM-killer simply kills the task
3907 that triggered the out-of-memory condition. This avoids a pos‐
3908 sibly expensive tasklist scan.
3909 If /proc/sys/vm/panic_on_oom is nonzero, it takes precedence
3911 over whatever value is used in /proc/sys/vm/oom_kill_allocat‐
3912 ing_task.
3913 The default value is 0.
3915 /proc/sys/vm/overcommit_kbytes (since Linux 3.14)
3917 This writable file provides an alternative to /proc/sys/vm/over‐
3918 commit_ratio for controlling the CommitLimit when
3919 /proc/sys/vm/overcommit_memory has the value 2. It allows the
3920 amount of memory overcommitting to be specified as an absolute
3921 value (in kB), rather than as a percentage, as is done with
3922 overcommit_ratio. This allows for finer-grained control of Com‐
3923 mitLimit on systems with extremely large memory sizes.
3924 Only one of overcommit_kbytes or overcommit_ratio can have an
3926 effect: if overcommit_kbytes has a nonzero value, then it is
3927 used to calculate CommitLimit, otherwise overcommit_ratio is
3928 used. Writing a value to either of these files causes the value
3929 in the other file to be set to zero.
3930 /proc/sys/vm/overcommit_memory
3932 This file contains the kernel virtual memory accounting mode.
3933 Values are:
3934 0: heuristic overcommit (this is the default)
3936 1: always overcommit, never check
3937 2: always check, never overcommit
3938 In mode 0, calls of mmap(2) with MAP_NORESERVE are not checked,
3940 and the default check is very weak, leading to the risk of get‐
3941 ting a process "OOM-killed".
3942 In mode 1, the kernel pretends there is always enough memory,
3944 until memory actually runs out. One use case for this mode is
3945 scientific computing applications that employ large sparse
3946 arrays. In Linux kernel versions before 2.6.0, any nonzero
3947 value implies mode 1.
3948 In mode 2 (available since Linux 2.6), the total virtual address
3950 space that can be allocated (CommitLimit in /proc/meminfo) is
3951 calculated as
3952 CommitLimit = (total_RAM - total_huge_TLB) *
3954 overcommit_ratio / 100 + total_swap
3955 where:
3957 * total_RAM is the total amount of RAM on the system;
3959 * total_huge_TLB is the amount of memory set aside for
3961 huge pages;
3962 * overcommit_ratio is the value in /proc/sys/vm/overcom‐
3964 mit_ratio; and
3965 * total_swap is the amount of swap space.
3967 For example, on a system with 16 GB of physical RAM, 16 GB of
3969 swap, no space dedicated to huge pages, and an overcommit_ratio
3970 of 50, this formula yields a CommitLimit of 24 GB.
3971 Since Linux 3.14, if the value in /proc/sys/vm/overcommit_kbytes
3973 is nonzero, then CommitLimit is instead calculated as:
3974 CommitLimit = overcommit_kbytes + total_swap
3976 See also the description of /proc/sys/vm/admin_reserve_kbytes
3978 and /proc/sys/vm/user_reserve_kbytes.
3979 /proc/sys/vm/overcommit_ratio (since Linux 2.6.0)
3981 This writable file defines a percentage by which memory can be
3982 overcommitted. The default value in the file is 50. See the
3983 description of /proc/sys/vm/overcommit_memory.
3984 /proc/sys/vm/panic_on_oom (since Linux 2.6.18)
3986 This enables or disables a kernel panic in an out-of-memory sit‐
3987 uation.
3988 If this file is set to the value 0, the kernel's OOM-killer will
3990 kill some rogue process. Usually, the OOM-killer is able to
3991 kill a rogue process and the system will survive.
3992 If this file is set to the value 1, then the kernel normally
3994 panics when out-of-memory happens. However, if a process limits
3995 allocations to certain nodes using memory policies (mbind(2)
3996 MPOL_BIND) or cpusets (cpuset(7)) and those nodes reach memory
3997 exhaustion status, one process may be killed by the OOM-killer.
3998 No panic occurs in this case: because other nodes' memory may be
3999 free, this means the system as a whole may not have reached an
4000 out-of-memory situation yet.
4001 If this file is set to the value 2, the kernel always panics
4003 when an out-of-memory condition occurs.
4004 The default value is 0. 1 and 2 are for failover of clustering.
4006 Select either according to your policy of failover.
4007 /proc/sys/vm/swappiness
4009 The value in this file controls how aggressively the kernel will
4010 swap memory pages. Higher values increase aggressiveness, lower
4011 values decrease aggressiveness. The default value is 60.
4012 /proc/sys/vm/user_reserve_kbytes (since Linux 3.10)
4014 Specifies an amount of memory (in KiB) to reserve for user pro‐
4015 cesses. This is intended to prevent a user from starting a sin‐
4016 gle memory hogging process, such that they cannot recover (kill
4017 the hog). The value in this file has an effect only when
4018 /proc/sys/vm/overcommit_memory is set to 2 ("overcommit never"
4019 mode). In this case, the system reserves an amount of memory
4020 that is the minimum of [3% of current process size,
4021 user_reserve_kbytes].
4022 The default value in this file is the minimum of [3% of free
4024 pages, 128MiB] expressed as KiB.
4025 If the value in this file is set to zero, then a user will be
4027 allowed to allocate all free memory with a single process (minus
4028 the amount reserved by /proc/sys/vm/admin_reserve_kbytes). Any
4029 subsequent attempts to execute a command will result in "fork:
4030 Cannot allocate memory".
4031 Changing the value in this file takes effect whenever an appli‐
4033 cation requests memory.
4034 /proc/sys/vm/unprivileged_userfaultfd (since Linux 5.2)
4036 This (writable) file exposes a flag that controls whether
4037 unprivileged processes are allowed to employ userfaultfd(2). If
4038 this file has the value 1, then unprivileged processes may use
4039 userfaultfd(2). If this file has the value 0, then only pro‐
4040 cesses that have the CAP_SYS_PTRACE capability may employ user‐
4041 faultfd(2). The default value in this file is 1.
4042 /proc/sysrq-trigger (since Linux 2.4.21)
4044 Writing a character to this file triggers the same SysRq func‐
4045 tion as typing ALT-SysRq-<character> (see the description of
4046 /proc/sys/kernel/sysrq). This file is normally writable only by
4047 root. For further details see the Linux kernel source file Doc‐
4048 umentation/admin-guide/sysrq.rst (or Documentation/sysrq.txt
4049 before Linux 4.10).
4050 /proc/sysvipc
4052 Subdirectory containing the pseudo-files msg, sem and shm.
4053 These files list the System V Interprocess Communication (IPC)
4054 objects (respectively: message queues, semaphores, and shared
4055 memory) that currently exist on the system, providing similar
4056 information to that available via ipcs(1). These files have
4057 headers and are formatted (one IPC object per line) for easy
4058 understanding. sysvipc(7) provides further background on the
4059 information shown by these files.
4060 /proc/thread-self (since Linux 3.17)
4062 This directory refers to the thread accessing the /proc filesys‐
4063 tem, and is identical to the /proc/self/task/[tid] directory
4064 named by the process thread ID ([tid]) of the same thread.
4065 /proc/timer_list (since Linux 2.6.21)
4067 This read-only file exposes a list of all currently pending
4068 (high-resolution) timers, all clock-event sources, and their
4069 parameters in a human-readable form.
4070 /proc/timer_stats (from Linux 2.6.21 until Linux 4.10)
4072 This is a debugging facility to make timer (ab)use in a Linux
4073 system visible to kernel and user-space developers. It can be
4074 used by kernel and user-space developers to verify that their
4075 code does not make undue use of timers. The goal is to avoid
4076 unnecessary wakeups, thereby optimizing power consumption.
4077 If enabled in the kernel (CONFIG_TIMER_STATS), but not used, it
4079 has almost zero run-time overhead and a relatively small data-
4080 structure overhead. Even if collection is enabled at run time,
4081 overhead is low: all the locking is per-CPU and lookup is
4082 hashed.
4083 The /proc/timer_stats file is used both to control sampling
4085 facility and to read out the sampled information.
4086 The timer_stats functionality is inactive on bootup. A sampling
4088 period can be started using the following command:
4089 # echo 1 > /proc/timer_stats
4091 The following command stops a sampling period:
4093 # echo 0 > /proc/timer_stats
4095 The statistics can be retrieved by:
4097 $ cat /proc/timer_stats
4099 While sampling is enabled, each readout from /proc/timer_stats
4101 will see newly updated statistics. Once sampling is disabled,
4102 the sampled information is kept until a new sample period is
4103 started. This allows multiple readouts.
4104 Sample output from /proc/timer_stats:
4106 $ cat /proc/timer_stats
4108 Timer Stats Version: v0.3
4109 Sample period: 1.764 s
4110 Collection: active
4111 255, 0 swapper/3 hrtimer_start_range_ns (tick_sched_timer)
4112 71, 0 swapper/1 hrtimer_start_range_ns (tick_sched_timer)
4113 58, 0 swapper/0 hrtimer_start_range_ns (tick_sched_timer)
4114 4, 1694 gnome-shell mod_delayed_work_on (delayed_work_timer_fn)
4115 17, 7 rcu_sched rcu_gp_kthread (process_timeout)
4116 ...
4117 1, 4911 kworker/u16:0 mod_delayed_work_on (delayed_work_timer_fn)
4118 1D, 2522 kworker/0:0 queue_delayed_work_on (delayed_work_timer_fn)
4119 1029 total events, 583.333 events/sec
4120 The output columns are:
4122 * a count of the number of events, optionally (since Linux
4124 2.6.23) followed by the letter 'D' if this is a deferrable
4125 timer;
4126 * the PID of the process that initialized the timer;
4128 * the name of the process that initialized the timer;
4130 * the function where the timer was initialized; and
4132 * (in parentheses) the callback function that is associated
4134 with the timer.
4135 During the Linux 4.11 development cycle, this file was removed
4137 because of security concerns, as it exposes information across
4138 namespaces. Furthermore, it is possible to obtain the same
4139 information via in-kernel tracing facilities such as ftrace.
4140 /proc/tty
4142 Subdirectory containing the pseudo-files and subdirectories for
4143 tty drivers and line disciplines.
4144 /proc/uptime
4146 This file contains two numbers (values in seconds): the uptime
4147 of the system (including time spent in suspend) and the amount
4148 of time spent in the idle process.
4149 /proc/version
4151 This string identifies the kernel version that is currently run‐
4152 ning. It includes the contents of /proc/sys/kernel/ostype,
4153 /proc/sys/kernel/osrelease and /proc/sys/kernel/version. For
4154 example:
4155 Linux version 1.0.9 (quinlan@phaze) #1 Sat May 14 01:51:54 EDT 1994
4157 /proc/vmstat (since Linux 2.6.0)
4159 This file displays various virtual memory statistics. Each line
4160 of this file contains a single name-value pair, delimited by
4161 white space. Some lines are present only if the kernel was con‐
4162 figured with suitable options. (In some cases, the options
4163 required for particular files have changed across kernel ver‐
4164 sions, so they are not listed here. Details can be found by
4165 consulting the kernel source code.) The following fields may be
4166 present:
4167 nr_free_pages (since Linux 2.6.31)
4169 nr_alloc_batch (since Linux 3.12)
4171 nr_inactive_anon (since Linux 2.6.28)
4173 nr_active_anon (since Linux 2.6.28)
4175 nr_inactive_file (since Linux 2.6.28)
4177 nr_active_file (since Linux 2.6.28)
4179 nr_unevictable (since Linux 2.6.28)
4181 nr_mlock (since Linux 2.6.28)
4183 nr_anon_pages (since Linux 2.6.18)
4185 nr_mapped (since Linux 2.6.0)
4187 nr_file_pages (since Linux 2.6.18)
4189 nr_dirty (since Linux 2.6.0)
4191 nr_writeback (since Linux 2.6.0)
4193 nr_slab_reclaimable (since Linux 2.6.19)
4195 nr_slab_unreclaimable (since Linux 2.6.19)
4197 nr_page_table_pages (since Linux 2.6.0)
4199 nr_kernel_stack (since Linux 2.6.32)
4201 Amount of memory allocated to kernel stacks.
4202 nr_unstable (since Linux 2.6.0)
4204 nr_bounce (since Linux 2.6.12)
4206 nr_vmscan_write (since Linux 2.6.19)
4208 nr_vmscan_immediate_reclaim (since Linux 3.2)
4210 nr_writeback_temp (since Linux 2.6.26)
4212 nr_isolated_anon (since Linux 2.6.32)
4214 nr_isolated_file (since Linux 2.6.32)
4216 nr_shmem (since Linux 2.6.32)
4218 Pages used by shmem and tmpfs(5).
4219 nr_dirtied (since Linux 2.6.37)
4221 nr_written (since Linux 2.6.37)
4223 nr_pages_scanned (since Linux 3.17)
4225 numa_hit (since Linux 2.6.18)
4227 numa_miss (since Linux 2.6.18)
4229 numa_foreign (since Linux 2.6.18)
4231 numa_interleave (since Linux 2.6.18)
4233 numa_local (since Linux 2.6.18)
4235 numa_other (since Linux 2.6.18)
4237 workingset_refault (since Linux 3.15)
4239 workingset_activate (since Linux 3.15)
4241 workingset_nodereclaim (since Linux 3.15)
4243 nr_anon_transparent_hugepages (since Linux 2.6.38)
4245 nr_free_cma (since Linux 3.7)
4247 Number of free CMA (Contiguous Memory Allocator) pages.
4248 nr_dirty_threshold (since Linux 2.6.37)
4250 nr_dirty_background_threshold (since Linux 2.6.37)
4252 pgpgin (since Linux 2.6.0)
4254 pgpgout (since Linux 2.6.0)
4256 pswpin (since Linux 2.6.0)
4258 pswpout (since Linux 2.6.0)
4260 pgalloc_dma (since Linux 2.6.5)
4262 pgalloc_dma32 (since Linux 2.6.16)
4264 pgalloc_normal (since Linux 2.6.5)
4266 pgalloc_high (since Linux 2.6.5)
4268 pgalloc_movable (since Linux 2.6.23)
4270 pgfree (since Linux 2.6.0)
4272 pgactivate (since Linux 2.6.0)
4274 pgdeactivate (since Linux 2.6.0)
4276 pgfault (since Linux 2.6.0)
4278 pgmajfault (since Linux 2.6.0)
4280 pgrefill_dma (since Linux 2.6.5)
4282 pgrefill_dma32 (since Linux 2.6.16)
4284 pgrefill_normal (since Linux 2.6.5)
4286 pgrefill_high (since Linux 2.6.5)
4288 pgrefill_movable (since Linux 2.6.23)
4290 pgsteal_kswapd_dma (since Linux 3.4)
4292 pgsteal_kswapd_dma32 (since Linux 3.4)
4294 pgsteal_kswapd_normal (since Linux 3.4)
4296 pgsteal_kswapd_high (since Linux 3.4)
4298 pgsteal_kswapd_movable (since Linux 3.4)
4300 pgsteal_direct_dma
4302 pgsteal_direct_dma32 (since Linux 3.4)
4304 pgsteal_direct_normal (since Linux 3.4)
4306 pgsteal_direct_high (since Linux 3.4)
4308 pgsteal_direct_movable (since Linux 2.6.23)
4310 pgscan_kswapd_dma
4312 pgscan_kswapd_dma32 (since Linux 2.6.16)
4314 pgscan_kswapd_normal (since Linux 2.6.5)
4316 pgscan_kswapd_high
4318 pgscan_kswapd_movable (since Linux 2.6.23)
4320 pgscan_direct_dma
4322 pgscan_direct_dma32 (since Linux 2.6.16)
4324 pgscan_direct_normal
4326 pgscan_direct_high
4328 pgscan_direct_movable (since Linux 2.6.23)
4330 pgscan_direct_throttle (since Linux 3.6)
4332 zone_reclaim_failed (since linux 2.6.31)
4334 pginodesteal (since linux 2.6.0)
4336 slabs_scanned (since linux 2.6.5)
4338 kswapd_inodesteal (since linux 2.6.0)
4340 kswapd_low_wmark_hit_quickly (since 2.6.33)
4342 kswapd_high_wmark_hit_quickly (since 2.6.33)
4344 pageoutrun (since Linux 2.6.0)
4346 allocstall (since Linux 2.6.0)
4348 pgrotated (since Linux 2.6.0)
4350 drop_pagecache (since Linux 3.15)
4352 drop_slab (since Linux 3.15)
4354 numa_pte_updates (since Linux 3.8)
4356 numa_huge_pte_updates (since Linux 3.13)
4358 numa_hint_faults (since Linux 3.8)
4360 numa_hint_faults_local (since Linux 3.8)
4362 numa_pages_migrated (since Linux 3.8)
4364 pgmigrate_success (since Linux 3.8)
4366 pgmigrate_fail (since Linux 3.8)
4368 compact_migrate_scanned (since Linux 3.8)
4370 compact_free_scanned (since Linux 3.8)
4372 compact_isolated (since Linux 3.8)
4374 compact_stall (since Linux 2.6.35)
4376 See the kernel source file Documenta‐
4377 tion/admin-guide/mm/transhuge.rst.
4378 compact_fail (since Linux 2.6.35)
4380 See the kernel source file Documenta‐
4381 tion/admin-guide/mm/transhuge.rst.
4382 compact_success (since Linux 2.6.35)
4384 See the kernel source file Documenta‐
4385 tion/admin-guide/mm/transhuge.rst.
4386 htlb_buddy_alloc_success (since Linux 2.6.26)
4388 htlb_buddy_alloc_fail (since Linux 2.6.26)
4390 unevictable_pgs_culled (since Linux 2.6.28)
4392 unevictable_pgs_scanned (since Linux 2.6.28)
4394 unevictable_pgs_rescued (since Linux 2.6.28)
4396 unevictable_pgs_mlocked (since Linux 2.6.28)
4398 unevictable_pgs_munlocked (since Linux 2.6.28)
4400 unevictable_pgs_cleared (since Linux 2.6.28)
4402 unevictable_pgs_stranded (since Linux 2.6.28)
4404 thp_fault_alloc (since Linux 2.6.39)
4406 See the kernel source file Documenta‐
4407 tion/admin-guide/mm/transhuge.rst.
4408 thp_fault_fallback (since Linux 2.6.39)
4410 See the kernel source file Documenta‐
4411 tion/admin-guide/mm/transhuge.rst.
4412 thp_collapse_alloc (since Linux 2.6.39)
4414 See the kernel source file Documenta‐
4415 tion/admin-guide/mm/transhuge.rst.
4416 thp_collapse_alloc_failed (since Linux 2.6.39)
4418 See the kernel source file Documenta‐
4419 tion/admin-guide/mm/transhuge.rst.
4420 thp_split (since Linux 2.6.39)
4422 See the kernel source file Documenta‐
4423 tion/admin-guide/mm/transhuge.rst.
4424 thp_zero_page_alloc (since Linux 3.8)
4426 See the kernel source file Documenta‐
4427 tion/admin-guide/mm/transhuge.rst.
4428 thp_zero_page_alloc_failed (since Linux 3.8)
4430 See the kernel source file Documenta‐
4431 tion/admin-guide/mm/transhuge.rst.
4432 balloon_inflate (since Linux 3.18)
4434 balloon_deflate (since Linux 3.18)
4436 balloon_migrate (since Linux 3.18)
4438 nr_tlb_remote_flush (since Linux 3.12)
4440 nr_tlb_remote_flush_received (since Linux 3.12)
4442 nr_tlb_local_flush_all (since Linux 3.12)
4444 nr_tlb_local_flush_one (since Linux 3.12)
4446 vmacache_find_calls (since Linux 3.16)
4448 vmacache_find_hits (since Linux 3.16)
4450 vmacache_full_flushes (since Linux 3.19)
4452 /proc/zoneinfo (since Linux 2.6.13)
4454 This file displays information about memory zones. This is use‐
4455 ful for analyzing virtual memory behavior.
4456
4458 Many files contain strings (e.g., the environment and command line)
4459 that are in the internal format, with subfields terminated by null
4460 bytes ('\0'). When inspecting such files, you may find that the
4461 results are more readable if you use a command of the following form to
4462 display them:
4463 $ cat file | tr '\000' '\n'
4465 This manual page is incomplete, possibly inaccurate, and is the kind of
4467 thing that needs to be updated very often.
4468
4470 cat(1), dmesg(1), find(1), free(1), htop(1), init(1), ps(1), pstree(1),
4471 tr(1), uptime(1), chroot(2), mmap(2), readlink(2), syslog(2),
4472 slabinfo(5), sysfs(5), hier(7), namespaces(7), time(7), arp(8),
4473 hdparm(8), ifconfig(8), lsmod(8), lspci(8), mount(8), netstat(8),
4474 procinfo(8), route(8), sysctl(8)
4475 The Linux kernel source files: Documentation/filesystems/proc.txt, Doc‐
4477 umentation/sysctl/fs.txt, Documentation/sysctl/kernel.txt, Documenta‐
4478 tion/sysctl/net.txt, and Documentation/sysctl/vm.txt.
4479
4481 This page is part of release 5.07 of the Linux man-pages project. A
4482 description of the project, information about reporting bugs, and the
4483 latest version of this page, can be found at
4484 https://www.kernel.org/doc/man-pages/.
4485Linux 2020-04-11 PROC(5)