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 op‐
29 tion is not specified.
30 1 Users may not access files and subdirectories inside any
32 /proc/[pid] directories but their own (the /proc/[pid] di‐
33 rectories themselves remain visible). Sensitive files such
34 as /proc/[pid]/cmdline and /proc/[pid]/status are now pro‐
35 tected 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 in‐
49 formation about running processes (e.g., discovering whether
50 some daemon is running with elevated privileges, whether an‐
51 other user is running some sensitive program, whether other
52 users are running any program at all, and so on).
53 gid=gid (since Linux 3.3)
55 Specifies the ID of a group whose members are authorized to
56 learn process information otherwise prohibited by hidepid (i.e.,
57 users in this group behave as though /proc was mounted with
58 hidepid=0). This group should be used instead of approaches
59 such as putting nonroot users into the sudoers(5) file.
60 Overview
62 Underneath /proc, there are the following general groups of files and
63 subdirectories:
64 /proc/[pid] subdirectories
66 Each one of these subdirectories contains files and subdirecto‐
67 ries exposing information about the process with the correspond‐
68 ing process ID.
69 Underneath each of the /proc/[pid] directories, a task subdirec‐
71 tory contains subdirectories of the form task/[tid], which con‐
72 tain corresponding information about each of the threads in the
73 process, where tid is the kernel thread ID of the thread.
74 The /proc/[pid] subdirectories are visible when iterating
76 through /proc with getdents(2) (and thus are visible when one
77 uses ls(1) to view the contents of /proc).
78 /proc/[tid] subdirectories
80 Each one of these subdirectories contains files and subdirecto‐
81 ries exposing information about the thread with the correspond‐
82 ing thread ID. The contents of these directories are the same
83 as the corresponding /proc/[pid]/task/[tid] directories.
84 The /proc/[tid] subdirectories are not visible when iterating
86 through /proc with getdents(2) (and thus are not visible when
87 one uses ls(1) to view the contents of /proc).
88 /proc/self
90 When a process accesses this magic symbolic link, it resolves to
91 the process's own /proc/[pid] directory.
92 /proc/thread-self
94 When a thread accesses this magic symbolic link, it resolves to
95 the process's own /proc/self/task/[tid] directory.
96 /proc/[a-z]*
98 Various other files and subdirectories under /proc expose sys‐
99 tem-wide information.
100 All of the above are described in more detail below.
102 Files and directories
104 The following list provides details of many of the files and directo‐
105 ries under the /proc hierarchy.
106 /proc/[pid]
108 There is a numerical subdirectory for each running process; the
109 subdirectory is named by the process ID. Each /proc/[pid] sub‐
110 directory contains the pseudo-files and directories described
111 below.
112 The files inside each /proc/[pid] directory are normally owned
114 by the effective user and effective group ID of the process.
115 However, as a security measure, the ownership is made root:root
116 if the process's "dumpable" attribute is set to a value other
117 than 1.
118 Before Linux 4.11, root:root meant the "global" root user ID and
120 group ID (i.e., UID 0 and GID 0 in the initial user namespace).
121 Since Linux 4.11, if the process is in a noninitial user name‐
122 space that has a valid mapping for user (group) ID 0 inside the
123 namespace, then the user (group) ownership of the files under
124 /proc/[pid] is instead made the same as the root user (group) ID
125 of the namespace. This means that inside a container, things
126 work as expected for the container "root" user.
127 The process's "dumpable" attribute may change for the following
129 reasons:
130 * The attribute was explicitly set via the prctl(2)
132 PR_SET_DUMPABLE operation.
133 * The attribute was reset to the value in the file
135 /proc/sys/fs/suid_dumpable (described below), for the reasons
136 described in prctl(2).
137 Resetting the "dumpable" attribute to 1 reverts the ownership of
139 the /proc/[pid]/* files to the process's effective UID and GID.
140 Note, however, that if the effective UID or GID is subsequently
141 modified, then the "dumpable" attribute may be reset, as de‐
142 scribed in prctl(2). Therefore, it may be desirable to reset
143 the "dumpable" attribute after making any desired changes to the
144 process's effective UID or GID.
145 /proc/[pid]/attr
147 The files in this directory provide an API for security modules.
148 The contents of this directory are files that can be read and
149 written in order to set security-related attributes. This di‐
150 rectory was added to support SELinux, but the intention was that
151 the API be general enough to support other security modules.
152 For the purpose of explanation, examples of how SELinux uses
153 these files are provided below.
154 This directory is present only if the kernel was configured with
156 CONFIG_SECURITY.
157 /proc/[pid]/attr/current (since Linux 2.6.0)
159 The contents of this file represent the current security at‐
160 tributes of the process.
161 In SELinux, this file is used to get the security context of a
163 process. Prior to Linux 2.6.11, this file could not be used to
164 set the security context (a write was always denied), since
165 SELinux limited process security transitions to execve(2) (see
166 the description of /proc/[pid]/attr/exec, below). Since Linux
167 2.6.11, SELinux lifted this restriction and began supporting
168 "set" operations via writes to this node if authorized by pol‐
169 icy, although use of this operation is only suitable for appli‐
170 cations that are trusted to maintain any desired separation be‐
171 tween the old and new security contexts.
172 Prior to Linux 2.6.28, SELinux did not allow threads within a
174 multithreaded process to set their security context via this
175 node as it would yield an inconsistency among the security con‐
176 texts of the threads sharing the same memory space. Since Linux
177 2.6.28, SELinux lifted this restriction and began supporting
178 "set" operations for threads within a multithreaded process if
179 the new security context is bounded by the old security context,
180 where the bounded relation is defined in policy and guarantees
181 that the new security context has a subset of the permissions of
182 the old security context.
183 Other security modules may choose to support "set" operations
185 via writes to this node.
186 /proc/[pid]/attr/exec (since Linux 2.6.0)
188 This file represents the attributes to assign to the process
189 upon a subsequent execve(2).
190 In SELinux, this is needed to support role/domain transitions,
192 and execve(2) is the preferred point to make such transitions
193 because it offers better control over the initialization of the
194 process in the new security label and the inheritance of state.
195 In SELinux, this attribute is reset on execve(2) so that the new
196 program reverts to the default behavior for any execve(2) calls
197 that it may make. In SELinux, a process can set only its own
198 /proc/[pid]/attr/exec attribute.
199 /proc/[pid]/attr/fscreate (since Linux 2.6.0)
201 This file represents the attributes to assign to files created
202 by subsequent calls to open(2), mkdir(2), symlink(2), and
203 mknod(2)
204 SELinux employs this file to support creation of a file (using
206 the aforementioned system calls) in a secure state, so that
207 there is no risk of inappropriate access being obtained between
208 the time of creation and the time that attributes are set. In
209 SELinux, this attribute is reset on execve(2), so that the new
210 program reverts to the default behavior for any file creation
211 calls it may make, but the attribute will persist across multi‐
212 ple file creation calls within a program unless it is explicitly
213 reset. In SELinux, a process can set only its own
214 /proc/[pid]/attr/fscreate attribute.
215 /proc/[pid]/attr/keycreate (since Linux 2.6.18)
217 If a process writes a security context into this file, all sub‐
218 sequently created keys (add_key(2)) will be labeled with this
219 context. For further information, see the kernel source file
220 Documentation/security/keys/core.rst (or file Documentation/se‐
221 curity/keys.txt on Linux between 3.0 and 4.13, or Documenta‐
222 tion/keys.txt before Linux 3.0).
223 /proc/[pid]/attr/prev (since Linux 2.6.0)
225 This file contains the security context of the process before
226 the last execve(2); that is, the previous value of
227 /proc/[pid]/attr/current.
228 /proc/[pid]/attr/socketcreate (since Linux 2.6.18)
230 If a process writes a security context into this file, all sub‐
231 sequently created sockets will be labeled with this context.
232 /proc/[pid]/autogroup (since Linux 2.6.38)
234 See sched(7).
235 /proc/[pid]/auxv (since 2.6.0)
237 This contains the contents of the ELF interpreter information
238 passed to the process at exec time. The format is one unsigned
239 long ID plus one unsigned long value for each entry. The last
240 entry contains two zeros. See also getauxval(3).
241 Permission to access this file is governed by a ptrace access
243 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
244 /proc/[pid]/cgroup (since Linux 2.6.24)
246 See cgroups(7).
247 /proc/[pid]/clear_refs (since Linux 2.6.22)
249 This is a write-only file, writable only by owner of the
251 process.
252 The following values may be written to the file:
254 1 (since Linux 2.6.22)
256 Reset the PG_Referenced and ACCESSED/YOUNG bits for all
257 the pages associated with the process. (Before kernel
258 2.6.32, writing any nonzero value to this file had this
259 effect.)
260 2 (since Linux 2.6.32)
262 Reset the PG_Referenced and ACCESSED/YOUNG bits for all
263 anonymous pages associated with the process.
264 3 (since Linux 2.6.32)
266 Reset the PG_Referenced and ACCESSED/YOUNG bits for all
267 file-mapped pages associated with the process.
268 Clearing the PG_Referenced and ACCESSED/YOUNG bits provides a
270 method to measure approximately how much memory a process is us‐
271 ing. One first inspects the values in the "Referenced" fields
272 for the VMAs shown in /proc/[pid]/smaps to get an idea of the
273 memory footprint of the process. One then clears the PG_Refer‐
274 enced and ACCESSED/YOUNG bits and, after some measured time in‐
275 terval, once again inspects the values in the "Referenced"
276 fields to get an idea of the change in memory footprint of the
277 process during the measured interval. If one is interested only
278 in inspecting the selected mapping types, then the value 2 or 3
279 can be used instead of 1.
280 Further values can be written to affect different properties:
282 4 (since Linux 3.11)
284 Clear the soft-dirty bit for all the pages associated
285 with the process. This is used (in conjunction with
286 /proc/[pid]/pagemap) by the check-point restore system to
287 discover which pages of a process have been dirtied since
288 the file /proc/[pid]/clear_refs was written to.
289 5 (since Linux 4.0)
291 Reset the peak resident set size ("high water mark") to
292 the process's current resident set size value.
293 Writing any value to /proc/[pid]/clear_refs other than those
295 listed above has no effect.
296 The /proc/[pid]/clear_refs file is present only if the CON‐
298 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
299 /proc/[pid]/cmdline
301 This read-only file holds the complete command line for the
302 process, unless the process is a zombie. In the latter case,
303 there is nothing in this file: that is, a read on this file will
304 return 0 characters. The command-line arguments appear in this
305 file as a set of strings separated by null bytes ('\0'), with a
306 further null byte after the last string.
307 If, after an execve(2), the process modifies its argv strings,
309 those changes will show up here. This is not the same thing as
310 modifying the argv array.
311 Furthermore, a process may change the memory location that this
313 file refers via prctl(2) operations such as PR_SET_MM_ARG_START.
314 Think of this file as the command line that the process wants
316 you to see.
317 /proc/[pid]/comm (since Linux 2.6.33)
319 This file exposes the process's comm value—that is, the command
320 name associated with the process. Different threads in the same
321 process may have different comm values, accessible via
322 /proc/[pid]/task/[tid]/comm. A thread may modify its comm
323 value, or that of any of other thread in the same thread group
324 (see the discussion of CLONE_THREAD in clone(2)), by writing to
325 the file /proc/self/task/[tid]/comm. Strings longer than
326 TASK_COMM_LEN (16) characters (including the terminating null
327 byte) are silently truncated.
328 This file provides a superset of the prctl(2) PR_SET_NAME and
330 PR_GET_NAME operations, and is employed by pthread_setname_np(3)
331 when used to rename threads other than the caller. The value in
332 this file is used for the %e specifier in /proc/sys/ker‐
333 nel/core_pattern; see core(5).
334 /proc/[pid]/coredump_filter (since Linux 2.6.23)
336 See core(5).
337 /proc/[pid]/cpuset (since Linux 2.6.12)
339 See cpuset(7).
340 /proc/[pid]/cwd
342 This is a symbolic link to the current working directory of the
343 process. To find out the current working directory of process
344 20, for instance, you can do this:
345 $ cd /proc/20/cwd; pwd -P
347 In a multithreaded process, the contents of this symbolic link
349 are not available if the main thread has already terminated
350 (typically by calling pthread_exit(3)).
351 Permission to dereference or read (readlink(2)) this symbolic
353 link is governed by a ptrace access mode PTRACE_MODE_READ_FS‐
354 CREDS check; see ptrace(2).
355 /proc/[pid]/environ
357 This file contains the initial environment that was set when the
358 currently executing program was started via execve(2). The en‐
359 tries are separated by null bytes ('\0'), and there may be a
360 null byte at the end. Thus, to print out the environment of
361 process 1, you would do:
362 $ cat /proc/1/environ | tr '\000' '\n'
364 If, after an execve(2), the process modifies its environment
366 (e.g., by calling functions such as putenv(3) or modifying the
367 environ(7) variable directly), this file will not reflect those
368 changes.
369 Furthermore, a process may change the memory location that this
371 file refers via prctl(2) operations such as PR_SET_MM_ENV_START.
372 Permission to access this file is governed by a ptrace access
374 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
375 /proc/[pid]/exe
377 Under Linux 2.2 and later, this file is a symbolic link contain‐
378 ing the actual pathname of the executed command. This symbolic
379 link can be dereferenced normally; attempting to open it will
380 open the executable. You can even type /proc/[pid]/exe to run
381 another copy of the same executable that is being run by process
382 [pid]. If the pathname has been unlinked, the symbolic link
383 will contain the string '(deleted)' appended to the original
384 pathname. In a multithreaded process, the contents of this sym‐
385 bolic link are not available if the main thread has already ter‐
386 minated (typically by calling pthread_exit(3)).
387 Permission to dereference or read (readlink(2)) this symbolic
389 link is governed by a ptrace access mode PTRACE_MODE_READ_FS‐
390 CREDS check; see ptrace(2).
391 Under Linux 2.0 and earlier, /proc/[pid]/exe is a pointer to the
393 binary which was executed, and appears as a symbolic link. A
394 readlink(2) call on this file under Linux 2.0 returns a string
395 in the format:
396 [device]:inode
398 For example, [0301]:1502 would be inode 1502 on device major 03
400 (IDE, MFM, etc. drives) minor 01 (first partition on the first
401 drive).
402 find(1) with the -inum option can be used to locate the file.
404 /proc/[pid]/fd/
406 This is a subdirectory containing one entry for each file which
407 the process has open, named by its file descriptor, and which is
408 a symbolic link to the actual file. Thus, 0 is standard input,
409 1 standard output, 2 standard error, and so on.
410 For file descriptors for pipes and sockets, the entries will be
412 symbolic links whose content is the file type with the inode. A
413 readlink(2) call on this file returns a string in the format:
414 type:[inode]
416 For example, socket:[2248868] will be a socket and its inode is
418 2248868. For sockets, that inode can be used to find more in‐
419 formation in one of the files under /proc/net/.
420 For file descriptors that have no corresponding inode (e.g.,
422 file descriptors produced by bpf(2), epoll_create(2),
423 eventfd(2), inotify_init(2), perf_event_open(2), signalfd(2),
424 timerfd_create(2), and userfaultfd(2)), the entry will be a sym‐
425 bolic link with contents of the form
426 anon_inode:<file-type>
428 In many cases (but not all), the file-type is surrounded by
430 square brackets.
431 For example, an epoll file descriptor will have a symbolic link
433 whose content is the string anon_inode:[eventpoll].
434 In a multithreaded process, the contents of this directory are
436 not available if the main thread has already terminated (typi‐
437 cally by calling pthread_exit(3)).
438 Programs that take a filename as a command-line argument, but
440 don't take input from standard input if no argument is supplied,
441 and programs that write to a file named as a command-line argu‐
442 ment, but don't send their output to standard output if no argu‐
443 ment is supplied, can nevertheless be made to use standard input
444 or standard output by using /proc/[pid]/fd files as command-line
445 arguments. For example, assuming that -i is the flag designat‐
446 ing an input file and -o is the flag designating an output file:
447 $ foobar -i /proc/self/fd/0 -o /proc/self/fd/1 ...
449 and you have a working filter.
451 /proc/self/fd/N is approximately the same as /dev/fd/N in some
453 UNIX and UNIX-like systems. Most Linux MAKEDEV scripts symboli‐
454 cally link /dev/fd to /proc/self/fd, in fact.
455 Most systems provide symbolic links /dev/stdin, /dev/stdout, and
457 /dev/stderr, which respectively link to the files 0, 1, and 2 in
458 /proc/self/fd. Thus the example command above could be written
459 as:
460 $ foobar -i /dev/stdin -o /dev/stdout ...
462 Permission to dereference or read (readlink(2)) the symbolic
464 links in this directory is governed by a ptrace access mode
465 PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
466 Note that for file descriptors referring to inodes (pipes and
468 sockets, see above), those inodes still have permission bits and
469 ownership information distinct from those of the /proc/[pid]/fd
470 entry, and that the owner may differ from the user and group IDs
471 of the process. An unprivileged process may lack permissions to
472 open them, as in this example:
473 $ echo test | sudo -u nobody cat
475 test
476 $ echo test | sudo -u nobody cat /proc/self/fd/0
477 cat: /proc/self/fd/0: Permission denied
478 File descriptor 0 refers to the pipe created by the shell and
480 owned by that shell's user, which is not nobody, so cat does not
481 have permission to create a new file descriptor to read from
482 that inode, even though it can still read from its existing file
483 descriptor 0.
484 /proc/[pid]/fdinfo/ (since Linux 2.6.22)
486 This is a subdirectory containing one entry for each file which
487 the process has open, named by its file descriptor. The files
488 in this directory are readable only by the owner of the process.
489 The contents of each file can be read to obtain information
490 about the corresponding file descriptor. The content depends on
491 the type of file referred to by the corresponding file descrip‐
492 tor.
493 For regular files and directories, we see something like:
495 $ cat /proc/12015/fdinfo/4
497 pos: 1000
498 flags: 01002002
499 mnt_id: 21
500 The fields are as follows:
502 pos This is a decimal number showing the file offset.
504 flags This is an octal number that displays the file access
506 mode and file status flags (see open(2)). If the close-
507 on-exec file descriptor flag is set, then flags will also
508 include the value O_CLOEXEC.
509 Before Linux 3.1, this field incorrectly displayed the
511 setting of O_CLOEXEC at the time the file was opened,
512 rather than the current setting of the close-on-exec
513 flag.
514 mnt_id This field, present since Linux 3.15, is the ID of the
516 mount containing this file. See the description of
517 /proc/[pid]/mountinfo.
518 For eventfd file descriptors (see eventfd(2)), we see (since
520 Linux 3.8) the following fields:
521 pos: 0
523 flags: 02
524 mnt_id: 10
525 eventfd-count: 40
526 eventfd-count is the current value of the eventfd counter, in
528 hexadecimal.
529 For epoll file descriptors (see epoll(7)), we see (since Linux
531 3.8) the following fields:
532 pos: 0
534 flags: 02
535 mnt_id: 10
536 tfd: 9 events: 19 data: 74253d2500000009
537 tfd: 7 events: 19 data: 74253d2500000007
538 Each of the lines beginning tfd describes one of the file de‐
540 scriptors being monitored via the epoll file descriptor (see
541 epoll_ctl(2) for some details). The tfd field is the number of
542 the file descriptor. The events field is a hexadecimal mask of
543 the events being monitored for this file descriptor. The data
544 field is the data value associated with this file descriptor.
545 For signalfd file descriptors (see signalfd(2)), we see (since
547 Linux 3.8) the following fields:
548 pos: 0
550 flags: 02
551 mnt_id: 10
552 sigmask: 0000000000000006
553 sigmask is the hexadecimal mask of signals that are accepted via
555 this signalfd file descriptor. (In this example, bits 2 and 3
556 are set, corresponding to the signals SIGINT and SIGQUIT; see
557 signal(7).)
558 For inotify file descriptors (see inotify(7)), we see (since
560 Linux 3.8) the following fields:
561 pos: 0
563 flags: 00
564 mnt_id: 11
565 inotify wd:2 ino:7ef82a sdev:800001 mask:800afff ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:2af87e00220ffd73
566 inotify wd:1 ino:192627 sdev:800001 mask:800afff ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:27261900802dfd73
567 Each of the lines beginning with "inotify" displays information
569 about one file or directory that is being monitored. The fields
570 in this line are as follows:
571 wd A watch descriptor number (in decimal).
573 ino The inode number of the target file (in hexadecimal).
575 sdev The ID of the device where the target file resides (in
577 hexadecimal).
578 mask The mask of events being monitored for the target file
580 (in hexadecimal).
581 If the kernel was built with exportfs support, the path to the
583 target file is exposed as a file handle, via three hexadecimal
584 fields: fhandle-bytes, fhandle-type, and f_handle.
585 For fanotify file descriptors (see fanotify(7)), we see (since
587 Linux 3.8) the following fields:
588 pos: 0
590 flags: 02
591 mnt_id: 11
592 fanotify flags:0 event-flags:88002
593 fanotify ino:19264f sdev:800001 mflags:0 mask:1 ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:4f261900a82dfd73
594 The fourth line displays information defined when the fanotify
596 group was created via fanotify_init(2):
597 flags The flags argument given to fanotify_init(2) (expressed
599 in hexadecimal).
600 event-flags
602 The event_f_flags argument given to fanotify_init(2) (ex‐
603 pressed in hexadecimal).
604 Each additional line shown in the file contains information
606 about one of the marks in the fanotify group. Most of these
607 fields are as for inotify, except:
608 mflags The flags associated with the mark (expressed in hexadec‐
610 imal).
611 mask The events mask for this mark (expressed in hexadecimal).
613 ignored_mask
615 The mask of events that are ignored for this mark (ex‐
616 pressed in hexadecimal).
617 For details on these fields, see fanotify_mark(2).
619 For timerfd file descriptors (see timerfd(2)), we see (since
621 Linux 3.17) the following fields:
622 pos: 0
624 flags: 02004002
625 mnt_id: 13
626 clockid: 0
627 ticks: 0
628 settime flags: 03
629 it_value: (7695568592, 640020877)
630 it_interval: (0, 0)
631 clockid
633 This is the numeric value of the clock ID (corresponding
634 to one of the CLOCK_* constants defined via <time.h>)
635 that is used to mark the progress of the timer (in this
636 example, 0 is CLOCK_REALTIME).
637 ticks This is the number of timer expirations that have oc‐
639 curred, (i.e., the value that read(2) on it would re‐
640 turn).
641 settime flags
643 This field lists the flags with which the timerfd was
644 last armed (see timerfd_settime(2)), in octal (in this
645 example, both TFD_TIMER_ABSTIME and TFD_TIMER_CAN‐
646 CEL_ON_SET are set).
647 it_value
649 This field contains the amount of time until the timer
650 will next expire, expressed in seconds and nanoseconds.
651 This is always expressed as a relative value, regardless
652 of whether the timer was created using the TFD_TIMER_AB‐
653 STIME flag.
654 it_interval
656 This field contains the interval of the timer, in seconds
657 and nanoseconds. (The it_value and it_interval fields
658 contain the values that timerfd_gettime(2) on this file
659 descriptor would return.)
660 /proc/[pid]/gid_map (since Linux 3.5)
662 See user_namespaces(7).
663 /proc/[pid]/io (since kernel 2.6.20)
665 This file contains I/O statistics for the process, for example:
666 # cat /proc/3828/io
668 rchar: 323934931
669 wchar: 323929600
670 syscr: 632687
671 syscw: 632675
672 read_bytes: 0
673 write_bytes: 323932160
674 cancelled_write_bytes: 0
675 The fields are as follows:
677 rchar: characters read
679 The number of bytes which this task has caused to be read
680 from storage. This is simply the sum of bytes which this
681 process passed to read(2) and similar system calls. It
682 includes things such as terminal I/O and is unaffected by
683 whether or not actual physical disk I/O was required (the
684 read might have been satisfied from pagecache).
685 wchar: characters written
687 The number of bytes which this task has caused, or shall
688 cause to be written to disk. Similar caveats apply here
689 as with rchar.
690 syscr: read syscalls
692 Attempt to count the number of read I/O operations—that
693 is, system calls such as read(2) and pread(2).
694 syscw: write syscalls
696 Attempt to count the number of write I/O operations—that
697 is, system calls such as write(2) and pwrite(2).
698 read_bytes: bytes read
700 Attempt to count the number of bytes which this process
701 really did cause to be fetched from the storage layer.
702 This is accurate for block-backed filesystems.
703 write_bytes: bytes written
705 Attempt to count the number of bytes which this process
706 caused to be sent to the storage layer.
707 cancelled_write_bytes:
709 The big inaccuracy here is truncate. If a process writes
710 1 MB to a file and then deletes the file, it will in fact
711 perform no writeout. But it will have been accounted as
712 having caused 1 MB of write. In other words: this field
713 represents the number of bytes which this process caused
714 to not happen, by truncating pagecache. A task can cause
715 "negative" I/O too. If this task truncates some dirty
716 pagecache, some I/O which another task has been accounted
717 for (in its write_bytes) will not be happening.
718 Note: In the current implementation, things are a bit racy on
720 32-bit systems: if process A reads process B's /proc/[pid]/io
721 while process B is updating one of these 64-bit counters,
722 process A could see an intermediate result.
723 Permission to access this file is governed by a ptrace access
725 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
726 /proc/[pid]/limits (since Linux 2.6.24)
728 This file displays the soft limit, hard limit, and units of mea‐
729 surement for each of the process's resource limits (see getr‐
730 limit(2)). Up to and including Linux 2.6.35, this file is pro‐
731 tected to allow reading only by the real UID of the process.
732 Since Linux 2.6.36, this file is readable by all users on the
733 system.
734 /proc/[pid]/map_files/ (since kernel 3.3)
736 This subdirectory contains entries corresponding to memory-
737 mapped files (see mmap(2)). Entries are named by memory region
738 start and end address pair (expressed as hexadecimal numbers),
739 and are symbolic links to the mapped files themselves. Here is
740 an example, with the output wrapped and reformatted to fit on an
741 80-column display:
742 # ls -l /proc/self/map_files/
744 lr--------. 1 root root 64 Apr 16 21:31
745 3252e00000-3252e20000 -> /usr/lib64/ld-2.15.so
746 ...
747 Although these entries are present for memory regions that were
749 mapped with the MAP_FILE flag, the way anonymous shared memory
750 (regions created with the MAP_ANON | MAP_SHARED flags) is imple‐
751 mented in Linux means that such regions also appear on this di‐
752 rectory. Here is an example where the target file is the
753 deleted /dev/zero one:
754 lrw-------. 1 root root 64 Apr 16 21:33
756 7fc075d2f000-7fc075e6f000 -> /dev/zero (deleted)
757 Permission to access this file is governed by a ptrace access
759 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
760 Until kernel version 4.3, this directory appeared only if the
762 CONFIG_CHECKPOINT_RESTORE kernel configuration option was en‐
763 abled.
764 Capabilities are required to read the contents of the symbolic
766 links in this directory: before Linux 5.9, the reading process
767 requires CAP_SYS_ADMIN in the initial user namespace; since
768 Linux 5.9, the reading process must have either CAP_SYS_ADMIN or
769 CAP_CHECKPOINT_RESTORE in the user namespace where it resides.
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 in‐
831 formation for a process with large numbers of threads is
832 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 \012 character se‐
848 quence.
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 mounts in the process's
865 mount namespace (see mount_namespaces(7)). It supplies various
866 information (e.g., propagation state, root of mount for bind
867 mounts, identifier for each mount and its parent) that is miss‐
868 ing from the (older) /proc/[pid]/mounts file, and fixes various
869 other problems with that file (e.g., nonextensibility, failure
870 to distinguish per-mount versus per-superblock options).
871 The file contains lines of the form:
873 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
875 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
876 The numbers in parentheses are labels for the descriptions be‐
878 low:
879 (1) mount ID: a unique ID for the mount (may be reused after
881 umount(2)).
882 (2) parent ID: the ID of the parent mount (or of self for the
884 root of this mount namespace's mount tree).
885 If a new mount is stacked on top of a previous existing
887 mount (so that it hides the existing mount) at pathname P,
888 then the parent of the new mount is the previous mount at
889 that location. Thus, when looking at all the mounts
890 stacked at a particular location, the top-most mount is the
891 one that is not the parent of any other mount at the same
892 location. (Note, however, that this top-most mount will be
893 accessible only if the longest path subprefix of P that is
894 a mount point is not itself hidden by a stacked mount.)
895 If the parent mount lies outside the process's root direc‐
897 tory (see chroot(2)), the ID shown here won't have a corre‐
898 sponding record in mountinfo whose mount ID (field 1)
899 matches this parent mount ID (because mounts that lie out‐
900 side the process's root directory are not shown in
901 mountinfo). As a special case of this point, the process's
902 root mount may have a parent mount (for the initramfs
903 filesystem) that lies outside the process's root directory,
904 and an entry for that mount will not appear in mountinfo.
905 (3) major:minor: the value of st_dev for files on this filesys‐
907 tem (see stat(2)).
908 (4) root: the pathname of the directory in the filesystem which
910 forms the root of this mount.
911 (5) mount point: the pathname of the mount point relative to
913 the process's root directory.
914 (6) mount options: per-mount options (see mount(2)).
916 (7) optional fields: zero or more fields of the form
918 "tag[:value]"; see below.
919 (8) separator: the end of the optional fields is marked by a
921 single hyphen.
922 (9) filesystem type: the filesystem type in the form
924 "type[.subtype]".
925 (10) mount source: filesystem-specific information or "none".
927 (11) super options: per-superblock options (see mount(2)).
929 Currently, the possible optional fields are shared, master,
931 propagate_from, and unbindable. See mount_namespaces(7) for a
932 description of these fields. Parsers should ignore all unrecog‐
933 nized optional fields.
934 For more information on mount propagation see: Documenta‐
936 tion/filesystems/sharedsubtree.txt in the Linux kernel source
937 tree.
938 /proc/[pid]/mounts (since Linux 2.4.19)
940 This file lists all the filesystems currently mounted in the
941 process's mount namespace (see mount_namespaces(7)). The format
942 of this file is documented in fstab(5).
943 Since kernel version 2.6.15, this file is pollable: after open‐
945 ing the file for reading, a change in this file (i.e., a
946 filesystem mount or unmount) causes select(2) to mark the file
947 descriptor as having an exceptional condition, and poll(2) and
948 epoll_wait(2) mark the file as having a priority event (POLL‐
949 PRI). (Before Linux 2.6.30, a change in this file was indicated
950 by the file descriptor being marked as readable for select(2),
951 and being marked as having an error condition for poll(2) and
952 epoll_wait(2).)
953 /proc/[pid]/mountstats (since Linux 2.6.17)
955 This file exports information (statistics, configuration infor‐
956 mation) about the mounts in the process's mount namespace (see
957 mount_namespaces(7)). Lines in this file have the form:
958 device /dev/sda7 mounted on /home with fstype ext3 [stats]
960 ( 1 ) ( 2 ) (3 ) ( 4 )
961 The fields in each line are:
963 (1) The name of the mounted device (or "nodevice" if there is
965 no corresponding device).
966 (2) The mount point within the filesystem tree.
968 (3) The filesystem type.
970 (4) Optional statistics and configuration information. Cur‐
972 rently (as at Linux 2.6.26), only NFS filesystems export
973 information via this field.
974 This file is readable only by the owner of the process.
976 /proc/[pid]/net (since Linux 2.6.25)
978 See the description of /proc/net.
979 /proc/[pid]/ns/ (since Linux 3.0)
981 This is a subdirectory containing one entry for each namespace
982 that supports being manipulated by setns(2). For more informa‐
983 tion, see namespaces(7).
984 /proc/[pid]/numa_maps (since Linux 2.6.14)
986 See numa(7).
987 /proc/[pid]/oom_adj (since Linux 2.6.11)
989 This file can be used to adjust the score used to select which
990 process should be killed in an out-of-memory (OOM) situation.
991 The kernel uses this value for a bit-shift operation of the
992 process's oom_score value: valid values are in the range -16 to
993 +15, plus the special value -17, which disables OOM-killing al‐
994 together for this process. A positive score increases the like‐
995 lihood of this process being killed by the OOM-killer; a nega‐
996 tive score decreases the likelihood.
997 The default value for this file is 0; a new process inherits its
999 parent's oom_adj setting. A process must be privileged
1000 (CAP_SYS_RESOURCE) to update this file.
1001 Since Linux 2.6.36, use of this file is deprecated in favor of
1003 /proc/[pid]/oom_score_adj.
1004 /proc/[pid]/oom_score (since Linux 2.6.11)
1006 This file displays the current score that the kernel gives to
1007 this process for the purpose of selecting a process for the OOM-
1008 killer. A higher score means that the process is more likely to
1009 be selected by the OOM-killer. The basis for this score is the
1010 amount of memory used by the process, with increases (+) or de‐
1011 creases (-) for factors including:
1012 * whether the process is privileged (-).
1014 Before kernel 2.6.36 the following factors were also used in the
1016 calculation of oom_score:
1017 * whether the process creates a lot of children using fork(2)
1019 (+);
1020 * whether the process has been running a long time, or has used
1022 a lot of CPU time (-);
1023 * whether the process has a low nice value (i.e., > 0) (+); and
1025 * whether the process is making direct hardware access (-).
1027 The oom_score also reflects the adjustment specified by the
1029 oom_score_adj or oom_adj setting for the process.
1030 /proc/[pid]/oom_score_adj (since Linux 2.6.36)
1032 This file can be used to adjust the badness heuristic used to
1033 select which process gets killed in out-of-memory conditions.
1034 The badness heuristic assigns a value to each candidate task
1036 ranging from 0 (never kill) to 1000 (always kill) to determine
1037 which process is targeted. The units are roughly a proportion
1038 along that range of allowed memory the process may allocate
1039 from, based on an estimation of its current memory and swap use.
1040 For example, if a task is using all allowed memory, its badness
1041 score will be 1000. If it is using half of its allowed memory,
1042 its score will be 500.
1043 There is an additional factor included in the badness score:
1045 root processes are given 3% extra memory over other tasks.
1046 The amount of "allowed" memory depends on the context in which
1048 the OOM-killer was called. If it is due to the memory assigned
1049 to the allocating task's cpuset being exhausted, the allowed
1050 memory represents the set of mems assigned to that cpuset (see
1051 cpuset(7)). If it is due to a mempolicy's node(s) being ex‐
1052 hausted, the allowed memory represents the set of mempolicy
1053 nodes. If it is due to a memory limit (or swap limit) being
1054 reached, the allowed memory is that configured limit. Finally,
1055 if it is due to the entire system being out of memory, the al‐
1056 lowed memory represents all allocatable resources.
1057 The value of oom_score_adj is added to the badness score before
1059 it is used to determine which task to kill. Acceptable values
1060 range from -1000 (OOM_SCORE_ADJ_MIN) to +1000
1061 (OOM_SCORE_ADJ_MAX). This allows user space to control the
1062 preference for OOM-killing, ranging from always preferring a
1063 certain task or completely disabling it from OOM-killing. The
1064 lowest possible value, -1000, is equivalent to disabling OOM-
1065 killing entirely for that task, since it will always report a
1066 badness score of 0.
1067 Consequently, it is very simple for user space to define the
1069 amount of memory to consider for each task. Setting an
1070 oom_score_adj value of +500, for example, is roughly equivalent
1071 to allowing the remainder of tasks sharing the same system,
1072 cpuset, mempolicy, or memory controller resources to use at
1073 least 50% more memory. A value of -500, on the other hand,
1074 would be roughly equivalent to discounting 50% of the task's al‐
1075 lowed memory from being considered as scoring against the task.
1076 For backward compatibility with previous kernels,
1078 /proc/[pid]/oom_adj can still be used to tune the badness score.
1079 Its value is scaled linearly with oom_score_adj.
1080 Writing to /proc/[pid]/oom_score_adj or /proc/[pid]/oom_adj will
1082 change the other with its scaled value.
1083 The choom(1) program provides a command-line interface for ad‐
1085 justing the oom_score_adj value of a running process or a newly
1086 executed command.
1087 /proc/[pid]/pagemap (since Linux 2.6.25)
1089 This file shows the mapping of each of the process's virtual
1090 pages into physical page frames or swap area. It contains one
1091 64-bit value for each virtual page, with the bits set as fol‐
1092 lows:
1093 63 If set, the page is present in RAM.
1095 62 If set, the page is in swap space
1097 61 (since Linux 3.5)
1099 The page is a file-mapped page or a shared anonymous
1100 page.
1101 60–57 (since Linux 3.11)
1103 Zero
1104 56 (since Linux 4.2)
1106 The page is exclusively mapped.
1107 55 (since Linux 3.11)
1109 PTE is soft-dirty (see the kernel source file Documenta‐
1110 tion/admin-guide/mm/soft-dirty.rst).
1111 54–0 If the page is present in RAM (bit 63), then these bits
1113 provide the page frame number, which can be used to index
1114 /proc/kpageflags and /proc/kpagecount. If the page is
1115 present in swap (bit 62), then bits 4–0 give the swap
1116 type, and bits 54–5 encode the swap offset.
1117 Before Linux 3.11, bits 60–55 were used to encode the base-2 log
1119 of the page size.
1120 To employ /proc/[pid]/pagemap efficiently, use /proc/[pid]/maps
1122 to determine which areas of memory are actually mapped and seek
1123 to skip over unmapped regions.
1124 The /proc/[pid]/pagemap file is present only if the CON‐
1126 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
1127 Permission to access this file is governed by a ptrace access
1129 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1130 /proc/[pid]/personality (since Linux 2.6.28)
1132 This read-only file exposes the process's execution domain, as
1133 set by personality(2). The value is displayed in hexadecimal
1134 notation.
1135 Permission to access this file is governed by a ptrace access
1137 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1138 /proc/[pid]/root
1140 UNIX and Linux support the idea of a per-process root of the
1141 filesystem, set by the chroot(2) system call. This file is a
1142 symbolic link that points to the process's root directory, and
1143 behaves in the same way as exe, and fd/*.
1144 Note however that this file is not merely a symbolic link. It
1146 provides the same view of the filesystem (including namespaces
1147 and the set of per-process mounts) as the process itself. An
1148 example illustrates this point. In one terminal, we start a
1149 shell in new user and mount namespaces, and in that shell we
1150 create some new mounts:
1151 $ PS1='sh1# ' unshare -Urnm
1153 sh1# mount -t tmpfs tmpfs /etc # Mount empty tmpfs at /etc
1154 sh1# mount --bind /usr /dev # Mount /usr at /dev
1155 sh1# echo $$
1156 27123
1157 In a second terminal window, in the initial mount namespace, we
1159 look at the contents of the corresponding mounts in the initial
1160 and new namespaces:
1161 $ PS1='sh2# ' sudo sh
1163 sh2# ls /etc | wc -l # In initial NS
1164 309
1165 sh2# ls /proc/27123/root/etc | wc -l # /etc in other NS
1166 0 # The empty tmpfs dir
1167 sh2# ls /dev | wc -l # In initial NS
1168 205
1169 sh2# ls /proc/27123/root/dev | wc -l # /dev in other NS
1170 11 # Actually bind
1171 # mounted to /usr
1172 sh2# ls /usr | wc -l # /usr in initial NS
1173 11
1174 In a multithreaded process, the contents of the /proc/[pid]/root
1176 symbolic link are not available if the main thread has already
1177 terminated (typically by calling pthread_exit(3)).
1178 Permission to dereference or read (readlink(2)) this symbolic
1180 link is governed by a ptrace access mode PTRACE_MODE_READ_FS‐
1181 CREDS check; see ptrace(2).
1182 /proc/[pid]/projid_map (since Linux 3.7)
1184 See user_namespaces(7).
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 at‐
1247 tributes, 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 (since Linux 4.6).
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 The /proc/[pid]/smaps file is present only if the CON‐
1297 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
1298 /proc/[pid]/stack (since Linux 2.6.29)
1300 This file provides a symbolic trace of the function calls in
1301 this process's kernel stack. This file is provided only if the
1302 kernel was built with the CONFIG_STACKTRACE configuration op‐
1303 tion.
1304 Permission to access this file is governed by a ptrace access
1306 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1307 /proc/[pid]/stat
1309 Status information about the process. This is used by ps(1).
1310 It is defined in the kernel source file fs/proc/array.c.
1311 The fields, in order, with their proper scanf(3) format speci‐
1313 fiers, are listed below. Whether or not certain of these fields
1314 display valid information is governed by a ptrace access mode
1315 PTRACE_MODE_READ_FSCREDS | PTRACE_MODE_NOAUDIT check (refer to
1316 ptrace(2)). If the check denies access, then the field value is
1317 displayed as 0. The affected fields are indicated with the
1318 marking [PT].
1319 (1) pid %d
1321 The process ID.
1322 (2) comm %s
1324 The filename of the executable, in parentheses. Strings
1325 longer than TASK_COMM_LEN (16) characters (including the
1326 terminating null byte) are silently truncated. This is
1327 visible whether or not the executable is swapped out.
1328 (3) state %c
1330 One of the following characters, indicating process
1331 state:
1332 R Running
1334 S Sleeping in an interruptible wait
1336 D Waiting in uninterruptible disk sleep
1338 Z Zombie
1340 T Stopped (on a signal) or (before Linux 2.6.33) trace
1342 stopped
1343 t Tracing stop (Linux 2.6.33 onward)
1345 W Paging (only before Linux 2.6.0)
1347 X Dead (from Linux 2.6.0 onward)
1349 x Dead (Linux 2.6.33 to 3.13 only)
1351 K Wakekill (Linux 2.6.33 to 3.13 only)
1353 W Waking (Linux 2.6.33 to 3.13 only)
1355 P Parked (Linux 3.9 to 3.13 only)
1357 (4) ppid %d
1359 The PID of the parent of this process.
1360 (5) pgrp %d
1362 The process group ID of the process.
1363 (6) session %d
1365 The session ID of the process.
1366 (7) tty_nr %d
1368 The controlling terminal of the process. (The minor de‐
1369 vice number is contained in the combination of bits 31 to
1370 20 and 7 to 0; the major device number is in bits 15 to
1371 8.)
1372 (8) tpgid %d
1374 The ID of the foreground process group of the controlling
1375 terminal of the process.
1376 (9) flags %u
1378 The kernel flags word of the process. For bit meanings,
1379 see the PF_* defines in the Linux kernel source file in‐
1380 clude/linux/sched.h. Details depend on the kernel ver‐
1381 sion.
1382 The format for this field was %lu before Linux 2.6.
1384 (10) minflt %lu
1386 The number of minor faults the process has made which
1387 have not required loading a memory page from disk.
1388 (11) cminflt %lu
1390 The number of minor faults that the process's waited-for
1391 children have made.
1392 (12) majflt %lu
1394 The number of major faults the process has made which
1395 have required loading a memory page from disk.
1396 (13) cmajflt %lu
1398 The number of major faults that the process's waited-for
1399 children have made.
1400 (14) utime %lu
1402 Amount of time that this process has been scheduled in
1403 user mode, measured in clock ticks (divide by
1404 sysconf(_SC_CLK_TCK)). This includes guest time,
1405 guest_time (time spent running a virtual CPU, see below),
1406 so that applications that are not aware of the guest time
1407 field do not lose that time from their calculations.
1408 (15) stime %lu
1410 Amount of time that this process has been scheduled in
1411 kernel mode, measured in clock ticks (divide by
1412 sysconf(_SC_CLK_TCK)).
1413 (16) cutime %ld
1415 Amount of time that this process's waited-for children
1416 have been scheduled in user mode, measured in clock ticks
1417 (divide by sysconf(_SC_CLK_TCK)). (See also times(2).)
1418 This includes guest time, cguest_time (time spent running
1419 a virtual CPU, see below).
1420 (17) cstime %ld
1422 Amount of time that this process's waited-for children
1423 have been scheduled in kernel mode, measured in clock
1424 ticks (divide by sysconf(_SC_CLK_TCK)).
1425 (18) priority %ld
1427 (Explanation for Linux 2.6) For processes running a real-
1428 time scheduling policy (policy below; see sched_setsched‐
1429 uler(2)), this is the negated scheduling priority, minus
1430 one; that is, a number in the range -2 to -100, corre‐
1431 sponding to real-time priorities 1 to 99. For processes
1432 running under a non-real-time scheduling policy, this is
1433 the raw nice value (setpriority(2)) as represented in the
1434 kernel. The kernel stores nice values as numbers in the
1435 range 0 (high) to 39 (low), corresponding to the user-
1436 visible nice range of -20 to 19.
1437 Before Linux 2.6, this was a scaled value based on the
1439 scheduler weighting given to this process.
1440 (19) nice %ld
1442 The nice value (see setpriority(2)), a value in the range
1443 19 (low priority) to -20 (high priority).
1444 (20) num_threads %ld
1446 Number of threads in this process (since Linux 2.6). Be‐
1447 fore kernel 2.6, this field was hard coded to 0 as a
1448 placeholder for an earlier removed field.
1449 (21) itrealvalue %ld
1451 The time in jiffies before the next SIGALRM is sent to
1452 the process due to an interval timer. Since kernel
1453 2.6.17, this field is no longer maintained, and is hard
1454 coded as 0.
1455 (22) starttime %llu
1457 The time the process started after system boot. In ker‐
1458 nels before Linux 2.6, this value was expressed in
1459 jiffies. Since Linux 2.6, the value is expressed in
1460 clock ticks (divide by sysconf(_SC_CLK_TCK)).
1461 The format for this field was %lu before Linux 2.6.
1463 (23) vsize %lu
1465 Virtual memory size in bytes.
1466 (24) rss %ld
1468 Resident Set Size: number of pages the process has in
1469 real memory. This is just the pages which count toward
1470 text, data, or stack space. This does not include pages
1471 which have not been demand-loaded in, or which are
1472 swapped out. This value is inaccurate; see
1473 /proc/[pid]/statm below.
1474 (25) rsslim %lu
1476 Current soft limit in bytes on the rss of the process;
1477 see the description of RLIMIT_RSS in getrlimit(2).
1478 (26) startcode %lu [PT]
1480 The address above which program text can run.
1481 (27) endcode %lu [PT]
1483 The address below which program text can run.
1484 (28) startstack %lu [PT]
1486 The address of the start (i.e., bottom) of the stack.
1487 (29) kstkesp %lu [PT]
1489 The current value of ESP (stack pointer), as found in the
1490 kernel stack page for the process.
1491 (30) kstkeip %lu [PT]
1493 The current EIP (instruction pointer).
1494 (31) signal %lu
1496 The bitmap of pending signals, displayed as a decimal
1497 number. Obsolete, because it does not provide informa‐
1498 tion on real-time signals; use /proc/[pid]/status in‐
1499 stead.
1500 (32) blocked %lu
1502 The bitmap of blocked signals, displayed as a decimal
1503 number. Obsolete, because it does not provide informa‐
1504 tion on real-time signals; use /proc/[pid]/status in‐
1505 stead.
1506 (33) sigignore %lu
1508 The bitmap of ignored signals, displayed as a decimal
1509 number. Obsolete, because it does not provide informa‐
1510 tion on real-time signals; use /proc/[pid]/status in‐
1511 stead.
1512 (34) sigcatch %lu
1514 The bitmap of caught signals, displayed as a decimal num‐
1515 ber. Obsolete, because it does not provide information
1516 on real-time signals; use /proc/[pid]/status instead.
1517 (35) wchan %lu [PT]
1519 This is the "channel" in which the process is waiting.
1520 It is the address of a location in the kernel where the
1521 process is sleeping. The corresponding symbolic name can
1522 be found in /proc/[pid]/wchan.
1523 (36) nswap %lu
1525 Number of pages swapped (not maintained).
1526 (37) cnswap %lu
1528 Cumulative nswap for child processes (not maintained).
1529 (38) exit_signal %d (since Linux 2.1.22)
1531 Signal to be sent to parent when we die.
1532 (39) processor %d (since Linux 2.2.8)
1534 CPU number last executed on.
1535 (40) rt_priority %u (since Linux 2.5.19)
1537 Real-time scheduling priority, a number in the range 1 to
1538 99 for processes scheduled under a real-time policy, or
1539 0, for non-real-time processes (see sched_setsched‐
1540 uler(2)).
1541 (41) policy %u (since Linux 2.5.19)
1543 Scheduling policy (see sched_setscheduler(2)). Decode
1544 using the SCHED_* constants in linux/sched.h.
1545 The format for this field was %lu before Linux 2.6.22.
1547 (42) delayacct_blkio_ticks %llu (since Linux 2.6.18)
1549 Aggregated block I/O delays, measured in clock ticks
1550 (centiseconds).
1551 (43) guest_time %lu (since Linux 2.6.24)
1553 Guest time of the process (time spent running a virtual
1554 CPU for a guest operating system), measured in clock
1555 ticks (divide by sysconf(_SC_CLK_TCK)).
1556 (44) cguest_time %ld (since Linux 2.6.24)
1558 Guest time of the process's children, measured in clock
1559 ticks (divide by sysconf(_SC_CLK_TCK)).
1560 (45) start_data %lu (since Linux 3.3) [PT]
1562 Address above which program initialized and uninitialized
1563 (BSS) data are placed.
1564 (46) end_data %lu (since Linux 3.3) [PT]
1566 Address below which program initialized and uninitialized
1567 (BSS) data are placed.
1568 (47) start_brk %lu (since Linux 3.3) [PT]
1570 Address above which program heap can be expanded with
1571 brk(2).
1572 (48) arg_start %lu (since Linux 3.5) [PT]
1574 Address above which program command-line arguments (argv)
1575 are placed.
1576 (49) arg_end %lu (since Linux 3.5) [PT]
1578 Address below program command-line arguments (argv) are
1579 placed.
1580 (50) env_start %lu (since Linux 3.5) [PT]
1582 Address above which program environment is placed.
1583 (51) env_end %lu (since Linux 3.5) [PT]
1585 Address below which program environment is placed.
1586 (52) exit_code %d (since Linux 3.5) [PT]
1588 The thread's exit status in the form reported by wait‐
1589 pid(2).
1590 /proc/[pid]/statm
1592 Provides information about memory usage, measured in pages. The
1593 columns are:
1594 size (1) total program size
1596 (same as VmSize in /proc/[pid]/status)
1597 resident (2) resident set size
1598 (inaccurate; same as VmRSS in /proc/[pid]/status)
1599 shared (3) number of resident shared pages
1600 (i.e., backed by a file)
1601 (inaccurate; same as RssFile+RssShmem in
1602 /proc/[pid]/status)
1603 text (4) text (code)
1604 lib (5) library (unused since Linux 2.6; always 0)
1605 data (6) data + stack
1606 dt (7) dirty pages (unused since Linux 2.6; always 0)
1607 Some of these values are inaccurate because of a kernel-internal
1609 scalability optimization. If accurate values are required, use
1610 /proc/[pid]/smaps or /proc/[pid]/smaps_rollup instead, which are
1611 much slower but provide accurate, detailed information.
1612 /proc/[pid]/status
1614 Provides much of the information in /proc/[pid]/stat and
1615 /proc/[pid]/statm in a format that's easier for humans to parse.
1616 Here's an example:
1617 $ cat /proc/$$/status
1619 Name: bash
1620 Umask: 0022
1621 State: S (sleeping)
1622 Tgid: 17248
1623 Ngid: 0
1624 Pid: 17248
1625 PPid: 17200
1626 TracerPid: 0
1627 Uid: 1000 1000 1000 1000
1628 Gid: 100 100 100 100
1629 FDSize: 256
1630 Groups: 16 33 100
1631 NStgid: 17248
1632 NSpid: 17248
1633 NSpgid: 17248
1634 NSsid: 17200
1635 VmPeak: 131168 kB
1636 VmSize: 131168 kB
1637 VmLck: 0 kB
1638 VmPin: 0 kB
1639 VmHWM: 13484 kB
1640 VmRSS: 13484 kB
1641 RssAnon: 10264 kB
1642 RssFile: 3220 kB
1643 RssShmem: 0 kB
1644 VmData: 10332 kB
1645 VmStk: 136 kB
1646 VmExe: 992 kB
1647 VmLib: 2104 kB
1648 VmPTE: 76 kB
1649 VmPMD: 12 kB
1650 VmSwap: 0 kB
1651 HugetlbPages: 0 kB # 4.4
1652 CoreDumping: 0 # 4.15
1653 Threads: 1
1654 SigQ: 0/3067
1655 SigPnd: 0000000000000000
1656 ShdPnd: 0000000000000000
1657 SigBlk: 0000000000010000
1658 SigIgn: 0000000000384004
1659 SigCgt: 000000004b813efb
1660 CapInh: 0000000000000000
1661 CapPrm: 0000000000000000
1662 CapEff: 0000000000000000
1663 CapBnd: ffffffffffffffff
1664 CapAmb: 0000000000000000
1665 NoNewPrivs: 0
1666 Seccomp: 0
1667 Speculation_Store_Bypass: vulnerable
1668 Cpus_allowed: 00000001
1669 Cpus_allowed_list: 0
1670 Mems_allowed: 1
1671 Mems_allowed_list: 0
1672 voluntary_ctxt_switches: 150
1673 nonvoluntary_ctxt_switches: 545
1674 The fields are as follows:
1676 Name Command run by this process. Strings longer than
1678 TASK_COMM_LEN [22m(16) characters (including the terminating
1679 null byte) are silently truncated.
1680 Umask Process umask, expressed in octal with a leading zero;
1682 see umask(2). (Since Linux 4.7.)
1683 State Current state of the process. One of "R (running)", "S
1685 (sleeping)", "D (disk sleep)", "T (stopped)", "t (tracing
1686 stop)", "Z (zombie)", or "X (dead)".
1687 Tgid Thread group ID (i.e., Process ID).
1689 Ngid NUMA group ID (0 if none; since Linux 3.13).
1691 Pid Thread ID (see gettid(2)).
1693 PPid PID of parent process.
1695 TracerPid
1697 PID of process tracing this process (0 if not being
1698 traced).
1699 Uid, Gid
1701 Real, effective, saved set, and filesystem UIDs (GIDs).
1702 FDSize Number of file descriptor slots currently allocated.
1704 Groups Supplementary group list.
1706 NStgid Thread group ID (i.e., PID) in each of the PID namespaces
1708 of which [pid] is a member. The leftmost entry shows the
1709 value with respect to the PID namespace of the process
1710 that mounted this procfs (or the root namespace if
1711 mounted by the kernel), followed by the value in succes‐
1712 sively nested inner namespaces. (Since Linux 4.1.)
1713 NSpid Thread ID in each of the PID namespaces of which [pid] is
1715 a member. The fields are ordered as for NStgid. (Since
1716 Linux 4.1.)
1717 NSpgid Process group ID in each of the PID namespaces of which
1719 [pid] is a member. The fields are ordered as for NStgid.
1720 (Since Linux 4.1.)
1721 NSsid descendant namespace session ID hierarchy Session ID in
1723 each of the PID namespaces of which [pid] is a member.
1724 The fields are ordered as for NStgid. (Since Linux 4.1.)
1725 VmPeak Peak virtual memory size.
1727 VmSize Virtual memory size.
1729 VmLck Locked memory size (see mlock(2)).
1731 VmPin Pinned memory size (since Linux 3.2). These are pages
1733 that can't be moved because something needs to directly
1734 access physical memory.
1735 VmHWM Peak resident set size ("high water mark"). This value
1737 is inaccurate; see /proc/[pid]/statm above.
1738 VmRSS Resident set size. Note that the value here is the sum
1740 of RssAnon, RssFile, and RssShmem. This value is inaccu‐
1741 rate; see /proc/[pid]/statm above.
1742 RssAnon
1744 Size of resident anonymous memory. (since Linux 4.5).
1745 This value is inaccurate; see /proc/[pid]/statm above.
1746 RssFile
1748 Size of resident file mappings. (since Linux 4.5). This
1749 value is inaccurate; see /proc/[pid]/statm above.
1750 RssShmem
1752 Size of resident shared memory (includes System V shared
1753 memory, mappings from tmpfs(5), and shared anonymous map‐
1754 pings). (since Linux 4.5).
1755 VmData, VmStk, VmExe
1757 Size of data, stack, and text segments. This value is
1758 inaccurate; see /proc/[pid]/statm above.
1759 VmLib Shared library code size.
1761 VmPTE Page table entries size (since Linux 2.6.10).
1763 VmPMD Size of second-level page tables (added in Linux 4.0; re‐
1765 moved in Linux 4.15).
1766 VmSwap Swapped-out virtual memory size by anonymous private
1768 pages; shmem swap usage is not included (since Linux
1769 2.6.34). This value is inaccurate; see /proc/[pid]/statm
1770 above.
1771 HugetlbPages
1773 Size of hugetlb memory portions (since Linux 4.4).
1774 CoreDumping
1776 Contains the value 1 if the process is currently dumping
1777 core, and 0 if it is not (since Linux 4.15). This infor‐
1778 mation can be used by a monitoring process to avoid
1779 killing a process that is currently dumping core, which
1780 could result in a corrupted core dump file.
1781 Threads
1783 Number of threads in process containing this thread.
1784 SigQ This field contains two slash-separated numbers that re‐
1786 late to queued signals for the real user ID of this
1787 process. The first of these is the number of currently
1788 queued signals for this real user ID, and the second is
1789 the resource limit on the number of queued signals for
1790 this process (see the description of RLIMIT_SIGPENDING in
1791 getrlimit(2)).
1792 SigPnd, ShdPnd
1794 Mask (expressed in hexadecimal) of signals pending for
1795 thread and for process as a whole (see pthreads(7) and
1796 signal(7)).
1797 SigBlk, SigIgn, SigCgt
1799 Masks (expressed in hexadecimal) indicating signals being
1800 blocked, ignored, and caught (see signal(7)).
1801 CapInh, CapPrm, CapEff
1803 Masks (expressed in hexadecimal) of capabilities enabled
1804 in inheritable, permitted, and effective sets (see capa‐
1805 bilities(7)).
1806 CapBnd Capability bounding set, expressed in hexadecimal (since
1808 Linux 2.6.26, see capabilities(7)).
1809 CapAmb Ambient capability set, expressed in hexadecimal (since
1811 Linux 4.3, see capabilities(7)).
1812 NoNewPrivs
1814 Value of the no_new_privs bit (since Linux 4.10, see
1815 prctl(2)).
1816 Seccomp
1818 Seccomp mode of the process (since Linux 3.8, see sec‐
1819 comp(2)). 0 means SECCOMP_MODE_DISABLED; 1 means SEC‐
1820 COMP_MODE_STRICT; 2 means SECCOMP_MODE_FILTER. This
1821 field is provided only if the kernel was built with the
1822 CONFIG_SECCOMP kernel configuration option enabled.
1823 Speculation_Store_Bypass
1825 Speculation flaw mitigation state (since Linux 4.17, see
1826 prctl(2)).
1827 Cpus_allowed
1829 Hexadecimal mask of CPUs on which this process may run
1830 (since Linux 2.6.24, see cpuset(7)).
1831 Cpus_allowed_list
1833 Same as previous, but in "list format" (since Linux
1834 2.6.26, see cpuset(7)).
1835 Mems_allowed
1837 Mask of memory nodes allowed to this process (since Linux
1838 2.6.24, see cpuset(7)).
1839 Mems_allowed_list
1841 Same as previous, but in "list format" (since Linux
1842 2.6.26, see cpuset(7)).
1843 voluntary_ctxt_switches, nonvoluntary_ctxt_switches
1845 Number of voluntary and involuntary context switches
1846 (since Linux 2.6.23).
1847 /proc/[pid]/syscall (since Linux 2.6.27)
1849 This file exposes the system call number and argument registers
1850 for the system call currently being executed by the process,
1851 followed by the values of the stack pointer and program counter
1852 registers. The values of all six argument registers are ex‐
1853 posed, although most system calls use fewer registers.
1854 If the process is blocked, but not in a system call, then the
1856 file displays -1 in place of the system call number, followed by
1857 just the values of the stack pointer and program counter. If
1858 process is not blocked, then the file contains just the string
1859 "running".
1860 This file is present only if the kernel was configured with CON‐
1862 FIG_HAVE_ARCH_TRACEHOOK.
1863 Permission to access this file is governed by a ptrace access
1865 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1866 /proc/[pid]/task (since Linux 2.6.0)
1868 This is a directory that contains one subdirectory for each
1869 thread in the process. The name of each subdirectory is the nu‐
1870 merical thread ID ([tid]) of the thread (see gettid(2)).
1871 Within each of these subdirectories, there is a set of files
1873 with the same names and contents as under the /proc/[pid] direc‐
1874 tories. For attributes that are shared by all threads, the con‐
1875 tents for each of the files under the task/[tid] subdirectories
1876 will be the same as in the corresponding file in the parent
1877 /proc/[pid] directory (e.g., in a multithreaded process, all of
1878 the task/[tid]/cwd files will have the same value as the
1879 /proc/[pid]/cwd file in the parent directory, since all of the
1880 threads in a process share a working directory). For attributes
1881 that are distinct for each thread, the corresponding files under
1882 task/[tid] may have different values (e.g., various fields in
1883 each of the task/[tid]/status files may be different for each
1884 thread), or they might not exist in /proc/[pid] at all.
1885 In a multithreaded process, the contents of the /proc/[pid]/task
1887 directory are not available if the main thread has already ter‐
1888 minated (typically by calling pthread_exit(3)).
1889 /proc/[pid]/task/[tid]/children (since Linux 3.5)
1891 A space-separated list of child tasks of this task. Each child
1892 task is represented by its TID.
1893 This option is intended for use by the checkpoint-restore (CRIU)
1895 system, and reliably provides a list of children only if all of
1896 the child processes are stopped or frozen. It does not work
1897 properly if children of the target task exit while the file is
1898 being read! Exiting children may cause non-exiting children to
1899 be omitted from the list. This makes this interface even more
1900 unreliable than classic PID-based approaches if the inspected
1901 task and its children aren't frozen, and most code should proba‐
1902 bly not use this interface.
1903 Until Linux 4.2, the presence of this file was governed by the
1905 CONFIG_CHECKPOINT_RESTORE kernel configuration option. Since
1906 Linux 4.2, it is governed by the CONFIG_PROC_CHILDREN option.
1907 /proc/[pid]/timers (since Linux 3.10)
1909 A list of the POSIX timers for this process. Each timer is
1910 listed with a line that starts with the string "ID:". For exam‐
1911 ple:
1912 ID: 1
1914 signal: 60/00007fff86e452a8
1915 notify: signal/pid.2634
1916 ClockID: 0
1917 ID: 0
1918 signal: 60/00007fff86e452a8
1919 notify: signal/pid.2634
1920 ClockID: 1
1921 The lines shown for each timer have the following meanings:
1923 ID The ID for this timer. This is not the same as the timer
1925 ID returned by timer_create(2); rather, it is the same
1926 kernel-internal ID that is available via the si_timerid
1927 field of the siginfo_t structure (see sigaction(2)).
1928 signal This is the signal number that this timer uses to deliver
1930 notifications followed by a slash, and then the
1931 sigev_value value supplied to the signal handler. Valid
1932 only for timers that notify via a signal.
1933 notify The part before the slash specifies the mechanism that
1935 this timer uses to deliver notifications, and is one of
1936 "thread", "signal", or "none". Immediately following the
1937 slash is either the string "tid" for timers with
1938 SIGEV_THREAD_ID notification, or "pid" for timers that
1939 notify by other mechanisms. Following the "." is the PID
1940 of the process (or the kernel thread ID of the thread)
1941 that will be delivered a signal if the timer delivers no‐
1942 tifications via a signal.
1943 ClockID
1945 This field identifies the clock that the timer uses for
1946 measuring time. For most clocks, this is a number that
1947 matches one of the user-space CLOCK_* constants exposed
1948 via <time.h>. CLOCK_PROCESS_CPUTIME_ID timers display
1949 with a value of -6 in this field.
1950 CLOCK_THREAD_CPUTIME_ID timers display with a value of -2
1951 in this field.
1952 This file is available only when the kernel was configured with
1954 CONFIG_CHECKPOINT_RESTORE.
1955 /proc/[pid]/timerslack_ns (since Linux 4.6)
1957 This file exposes the process's "current" timer slack value, ex‐
1958 pressed in nanoseconds. The file is writable, allowing the
1959 process's timer slack value to be changed. Writing 0 to this
1960 file resets the "current" timer slack to the "default" timer
1961 slack value. For further details, see the discussion of
1962 PR_SET_TIMERSLACK in prctl(2).
1963 Initially, permission to access this file was governed by a
1965 ptrace access mode PTRACE_MODE_ATTACH_FSCREDS check (see
1966 ptrace(2)). However, this was subsequently deemed too strict a
1967 requirement (and had the side effect that requiring a process to
1968 have the CAP_SYS_PTRACE capability would also allow it to view
1969 and change any process's memory). Therefore, since Linux 4.9,
1970 only the (weaker) CAP_SYS_NICE capability is required to access
1971 this file.
1972 /proc/[pid]/uid_map (since Linux 3.5)
1974 See user_namespaces(7).
1975 /proc/[pid]/wchan (since Linux 2.6.0)
1977 The symbolic name corresponding to the location in the kernel
1978 where the process is sleeping.
1979 Permission to access this file is governed by a ptrace access
1981 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1982 /proc/[tid]
1984 There is a numerical subdirectory for each running thread that
1985 is not a thread group leader (i.e., a thread whose thread ID is
1986 not the same as its process ID); the subdirectory is named by
1987 the thread ID. Each one of these subdirectories contains files
1988 and subdirectories exposing information about the thread with
1989 the thread ID tid. The contents of these directories are the
1990 same as the corresponding /proc/[pid]/task/[tid] directories.
1991 The /proc/[tid] subdirectories are not visible when iterating
1993 through /proc with getdents(2) (and thus are not visible when
1994 one uses ls(1) to view the contents of /proc). However, the
1995 pathnames of these directories are visible to (i.e., usable as
1996 arguments in) system calls that operate on pathnames.
1997 /proc/apm
1999 Advanced power management version and battery information when
2000 CONFIG_APM is defined at kernel compilation time.
2001 /proc/buddyinfo
2003 This file contains information which is used for diagnosing mem‐
2004 ory fragmentation issues. Each line starts with the identifica‐
2005 tion of the node and the name of the zone which together iden‐
2006 tify a memory region. This is then followed by the count of
2007 available chunks of a certain order in which these zones are
2008 split. The size in bytes of a certain order is given by the
2009 formula:
2010 (2^order) * PAGE_SIZE
2012 The binary buddy allocator algorithm inside the kernel will
2014 split one chunk into two chunks of a smaller order (thus with
2015 half the size) or combine two contiguous chunks into one larger
2016 chunk of a higher order (thus with double the size) to satisfy
2017 allocation requests and to counter memory fragmentation. The
2018 order matches the column number, when starting to count at zero.
2019 For example on an x86-64 system:
2021 Node 0, zone DMA 1 1 1 0 2 1 1 0 1 1 3
2022 Node 0, zone DMA32 65 47 4 81 52 28 13 10 5 1 404
2023 Node 0, zone Normal 216 55 189 101 84 38 37 27 5 3 587
2024 In this example, there is one node containing three zones and
2026 there are 11 different chunk sizes. If the page size is 4 kilo‐
2027 bytes, then the first zone called DMA (on x86 the first 16
2028 megabyte of memory) has 1 chunk of 4 kilobytes (order 0) avail‐
2029 able and has 3 chunks of 4 megabytes (order 10) available.
2030 If the memory is heavily fragmented, the counters for higher or‐
2032 der chunks will be zero and allocation of large contiguous areas
2033 will fail.
2034 Further information about the zones can be found in /proc/zone‐
2036 info.
2037 /proc/bus
2039 Contains subdirectories for installed busses.
2040 /proc/bus/pccard
2042 Subdirectory for PCMCIA devices when CONFIG_PCMCIA is set at
2043 kernel compilation time.
2044 /proc/bus/pccard/drivers
2046 /proc/bus/pci
2048 Contains various bus subdirectories and pseudo-files containing
2049 information about PCI busses, installed devices, and device
2050 drivers. Some of these files are not ASCII.
2051 /proc/bus/pci/devices
2053 Information about PCI devices. They may be accessed through
2054 lspci(8) and setpci(8).
2055 /proc/cgroups (since Linux 2.6.24)
2057 See cgroups(7).
2058 /proc/cmdline
2060 Arguments passed to the Linux kernel at boot time. Often done
2061 via a boot manager such as lilo(8) or grub(8).
2062 /proc/config.gz (since Linux 2.6)
2064 This file exposes the configuration options that were used to
2065 build the currently running kernel, in the same format as they
2066 would be shown in the .config file that resulted when configur‐
2067 ing the kernel (using make xconfig, make config, or similar).
2068 The file contents are compressed; view or search them using
2069 zcat(1) and zgrep(1). As long as no changes have been made to
2070 the following file, the contents of /proc/config.gz are the same
2071 as those provided by:
2072 cat /lib/modules/$(uname -r)/build/.config
2074 /proc/config.gz is provided only if the kernel is configured
2076 with CONFIG_IKCONFIG_PROC.
2077 /proc/crypto
2079 A list of the ciphers provided by the kernel crypto API. For
2080 details, see the kernel Linux Kernel Crypto API documentation
2081 available under the kernel source directory Documenta‐
2082 tion/crypto/ (or Documentation/DocBook before 4.10; the documen‐
2083 tation can be built using a command such as make htmldocs in the
2084 root directory of the kernel source tree).
2085 /proc/cpuinfo
2087 This is a collection of CPU and system architecture dependent
2088 items, for each supported architecture a different list. Two
2089 common entries are processor which gives CPU number and bo‐
2090 gomips; a system constant that is calculated during kernel ini‐
2091 tialization. SMP machines have information for each CPU. The
2092 lscpu(1) command gathers its information from this file.
2093 /proc/devices
2095 Text listing of major numbers and device groups. This can be
2096 used by MAKEDEV scripts for consistency with the kernel.
2097 /proc/diskstats (since Linux 2.5.69)
2099 This file contains disk I/O statistics for each disk device.
2100 See the Linux kernel source file Documentation/iostats.txt for
2101 further information.
2102 /proc/dma
2104 This is a list of the registered ISA DMA (direct memory access)
2105 channels in use.
2106 /proc/driver
2108 Empty subdirectory.
2109 /proc/execdomains
2111 List of the execution domains (ABI personalities).
2112 /proc/fb
2114 Frame buffer information when CONFIG_FB is defined during kernel
2115 compilation.
2116 /proc/filesystems
2118 A text listing of the filesystems which are supported by the
2119 kernel, namely filesystems which were compiled into the kernel
2120 or whose kernel modules are currently loaded. (See also
2121 filesystems(5).) If a filesystem is marked with "nodev", this
2122 means that it does not require a block device to be mounted
2123 (e.g., virtual filesystem, network filesystem).
2124 Incidentally, this file may be used by mount(8) when no filesys‐
2126 tem is specified and it didn't manage to determine the filesys‐
2127 tem type. Then filesystems contained in this file are tried
2128 (excepted those that are marked with "nodev").
2129 /proc/fs
2131 Contains subdirectories that in turn contain files with informa‐
2132 tion about (certain) mounted filesystems.
2133 /proc/ide
2135 This directory exists on systems with the IDE bus. There are
2136 directories for each IDE channel and attached device. Files in‐
2137 clude:
2138 cache buffer size in KB
2140 capacity number of sectors
2141 driver driver version
2142 geometry physical and logical geometry
2143 identify in hexadecimal
2144 media media type
2145 model manufacturer's model number
2146 settings drive settings
2147 smart_thresholds IDE disk management thresholds (in hex)
2148 smart_values IDE disk management values (in hex)
2149 The hdparm(8) utility provides access to this information in a
2151 friendly format.
2152 /proc/interrupts
2154 This is used to record the number of interrupts per CPU per IO
2155 device. Since Linux 2.6.24, for the i386 and x86-64 architec‐
2156 tures, at least, this also includes interrupts internal to the
2157 system (that is, not associated with a device as such), such as
2158 NMI (nonmaskable interrupt), LOC (local timer interrupt), and
2159 for SMP systems, TLB (TLB flush interrupt), RES (rescheduling
2160 interrupt), CAL (remote function call interrupt), and possibly
2161 others. Very easy to read formatting, done in ASCII.
2162 /proc/iomem
2164 I/O memory map in Linux 2.4.
2165 /proc/ioports
2167 This is a list of currently registered Input-Output port regions
2168 that are in use.
2169 /proc/kallsyms (since Linux 2.5.71)
2171 This holds the kernel exported symbol definitions used by the
2172 modules(X) tools to dynamically link and bind loadable modules.
2173 In Linux 2.5.47 and earlier, a similar file with slightly dif‐
2174 ferent syntax was named ksyms.
2175 /proc/kcore
2177 This file represents the physical memory of the system and is
2178 stored in the ELF core file format. With this pseudo-file, and
2179 an unstripped kernel (/usr/src/linux/vmlinux) binary, GDB can be
2180 used to examine the current state of any kernel data structures.
2181 The total length of the file is the size of physical memory
2183 (RAM) plus 4 KiB.
2184 /proc/keys (since Linux 2.6.10)
2186 See keyrings(7).
2187 /proc/key-users (since Linux 2.6.10)
2189 See keyrings(7).
2190 /proc/kmsg
2192 This file can be used instead of the syslog(2) system call to
2193 read kernel messages. A process must have superuser privileges
2194 to read this file, and only one process should read this file.
2195 This file should not be read if a syslog process is running
2196 which uses the syslog(2) system call facility to log kernel mes‐
2197 sages.
2198 Information in this file is retrieved with the dmesg(1) program.
2200 /proc/kpagecgroup (since Linux 4.3)
2202 This file contains a 64-bit inode number of the memory cgroup
2203 each page is charged to, indexed by page frame number (see the
2204 discussion of /proc/[pid]/pagemap).
2205 The /proc/kpagecgroup file is present only if the CONFIG_MEMCG
2207 kernel configuration option is enabled.
2208 /proc/kpagecount (since Linux 2.6.25)
2210 This file contains a 64-bit count of the number of times each
2211 physical page frame is mapped, indexed by page frame number (see
2212 the discussion of /proc/[pid]/pagemap).
2213 The /proc/kpagecount file is present only if the CON‐
2215 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
2216 /proc/kpageflags (since Linux 2.6.25)
2218 This file contains 64-bit masks corresponding to each physical
2219 page frame; it is indexed by page frame number (see the discus‐
2220 sion of /proc/[pid]/pagemap). The bits are as follows:
2221 0 - KPF_LOCKED
2223 1 - KPF_ERROR
2224 2 - KPF_REFERENCED
2225 3 - KPF_UPTODATE
2226 4 - KPF_DIRTY
2227 5 - KPF_LRU
2228 6 - KPF_ACTIVE
2229 7 - KPF_SLAB
2230 8 - KPF_WRITEBACK
2231 9 - KPF_RECLAIM
2232 10 - KPF_BUDDY
2233 11 - KPF_MMAP (since Linux 2.6.31)
2234 12 - KPF_ANON (since Linux 2.6.31)
2235 13 - KPF_SWAPCACHE (since Linux 2.6.31)
2236 14 - KPF_SWAPBACKED (since Linux 2.6.31)
2237 15 - KPF_COMPOUND_HEAD (since Linux 2.6.31)
2238 16 - KPF_COMPOUND_TAIL (since Linux 2.6.31)
2239 17 - KPF_HUGE (since Linux 2.6.31)
2240 18 - KPF_UNEVICTABLE (since Linux 2.6.31)
2241 19 - KPF_HWPOISON (since Linux 2.6.31)
2242 20 - KPF_NOPAGE (since Linux 2.6.31)
2243 21 - KPF_KSM (since Linux 2.6.32)
2244 22 - KPF_THP (since Linux 3.4)
2245 23 - KPF_BALLOON (since Linux 3.18)
2246 24 - KPF_ZERO_PAGE (since Linux 4.0)
2247 25 - KPF_IDLE (since Linux 4.3)
2248 For further details on the meanings of these bits, see the ker‐
2250 nel source file Documentation/admin-guide/mm/pagemap.rst. Be‐
2251 fore kernel 2.6.29, KPF_WRITEBACK, KPF_RECLAIM, KPF_BUDDY, and
2252 KPF_LOCKED did not report correctly.
2253 The /proc/kpageflags file is present only if the CON‐
2255 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
2256 /proc/ksyms (Linux 1.1.23–2.5.47)
2258 See /proc/kallsyms.
2259 /proc/loadavg
2261 The first three fields in this file are load average figures
2262 giving the number of jobs in the run queue (state R) or waiting
2263 for disk I/O (state D) averaged over 1, 5, and 15 minutes. They
2264 are the same as the load average numbers given by uptime(1) and
2265 other programs. The fourth field consists of two numbers sepa‐
2266 rated by a slash (/). The first of these is the number of cur‐
2267 rently runnable kernel scheduling entities (processes, threads).
2268 The value after the slash is the number of kernel scheduling en‐
2269 tities that currently exist on the system. The fifth field is
2270 the PID of the process that was most recently created on the
2271 system.
2272 /proc/locks
2274 This file shows current file locks (flock(2) and fcntl(2)) and
2275 leases (fcntl(2)).
2276 An example of the content shown in this file is the following:
2278 1: POSIX ADVISORY READ 5433 08:01:7864448 128 128
2280 2: FLOCK ADVISORY WRITE 2001 08:01:7864554 0 EOF
2281 3: FLOCK ADVISORY WRITE 1568 00:2f:32388 0 EOF
2282 4: POSIX ADVISORY WRITE 699 00:16:28457 0 EOF
2283 5: POSIX ADVISORY WRITE 764 00:16:21448 0 0
2284 6: POSIX ADVISORY READ 3548 08:01:7867240 1 1
2285 7: POSIX ADVISORY READ 3548 08:01:7865567 1826 2335
2286 8: OFDLCK ADVISORY WRITE -1 08:01:8713209 128 191
2287 The fields shown in each line are as follows:
2289 (1) The ordinal position of the lock in the list.
2291 (2) The lock type. Values that may appear here include:
2293 FLOCK This is a BSD file lock created using flock(2).
2295 OFDLCK This is an open file description (OFD) lock created
2297 using fcntl(2).
2298 POSIX This is a POSIX byte-range lock created using fc‐
2300 ntl(2).
2301 (3) Among the strings that can appear here are the following:
2303 ADVISORY
2305 This is an advisory lock.
2306 MANDATORY
2308 This is a mandatory lock.
2309 (4) The type of lock. Values that can appear here are:
2311 READ This is a POSIX or OFD read lock, or a BSD shared
2313 lock.
2314 WRITE This is a POSIX or OFD write lock, or a BSD exclusive
2316 lock.
2317 (5) The PID of the process that owns the lock.
2319 Because OFD locks are not owned by a single process (since
2321 multiple processes may have file descriptors that refer to
2322 the same open file description), the value -1 is displayed
2323 in this field for OFD locks. (Before kernel 4.14, a bug
2324 meant that the PID of the process that initially acquired
2325 the lock was displayed instead of the value -1.)
2326 (6) Three colon-separated subfields that identify the major and
2328 minor device ID of the device containing the filesystem
2329 where the locked file resides, followed by the inode number
2330 of the locked file.
2331 (7) The byte offset of the first byte of the lock. For BSD
2333 locks, this value is always 0.
2334 (8) The byte offset of the last byte of the lock. EOF in this
2336 field means that the lock extends to the end of the file.
2337 For BSD locks, the value shown is always EOF.
2338 Since Linux 4.9, the list of locks shown in /proc/locks is fil‐
2340 tered to show just the locks for the processes in the PID name‐
2341 space (see pid_namespaces(7)) for which the /proc filesystem was
2342 mounted. (In the initial PID namespace, there is no filtering
2343 of the records shown in this file.)
2344 The lslocks(8) command provides a bit more information about
2346 each lock.
2347 /proc/malloc (only up to and including Linux 2.2)
2349 This file is present only if CONFIG_DEBUG_MALLOC was defined
2350 during compilation.
2351 /proc/meminfo
2353 This file reports statistics about memory usage on the system.
2354 It is used by free(1) to report the amount of free and used mem‐
2355 ory (both physical and swap) on the system as well as the shared
2356 memory and buffers used by the kernel. Each line of the file
2357 consists of a parameter name, followed by a colon, the value of
2358 the parameter, and an option unit of measurement (e.g., "kB").
2359 The list below describes the parameter names and the format
2360 specifier required to read the field value. Except as noted be‐
2361 low, all of the fields have been present since at least Linux
2362 2.6.0. Some fields are displayed only if the kernel was config‐
2363 ured with various options; those dependencies are noted in the
2364 list.
2365 MemTotal %lu
2367 Total usable RAM (i.e., physical RAM minus a few reserved
2368 bits and the kernel binary code).
2369 MemFree %lu
2371 The sum of LowFree+HighFree.
2372 MemAvailable %lu (since Linux 3.14)
2374 An estimate of how much memory is available for starting
2375 new applications, without swapping.
2376 Buffers %lu
2378 Relatively temporary storage for raw disk blocks that
2379 shouldn't get tremendously large (20 MB or so).
2380 Cached %lu
2382 In-memory cache for files read from the disk (the page
2383 cache). Doesn't include SwapCached.
2384 SwapCached %lu
2386 Memory that once was swapped out, is swapped back in but
2387 still also is in the swap file. (If memory pressure is
2388 high, these pages don't need to be swapped out again be‐
2389 cause they are already in the swap file. This saves
2390 I/O.)
2391 Active %lu
2393 Memory that has been used more recently and usually not
2394 reclaimed unless absolutely necessary.
2395 Inactive %lu
2397 Memory which has been less recently used. It is more el‐
2398 igible to be reclaimed for other purposes.
2399 Active(anon) %lu (since Linux 2.6.28)
2401 [To be documented.]
2402 Inactive(anon) %lu (since Linux 2.6.28)
2404 [To be documented.]
2405 Active(file) %lu (since Linux 2.6.28)
2407 [To be documented.]
2408 Inactive(file) %lu (since Linux 2.6.28)
2410 [To be documented.]
2411 Unevictable %lu (since Linux 2.6.28)
2413 (From Linux 2.6.28 to 2.6.30, CONFIG_UNEVICTABLE_LRU was
2414 required.) [To be documented.]
2415 Mlocked %lu (since Linux 2.6.28)
2417 (From Linux 2.6.28 to 2.6.30, CONFIG_UNEVICTABLE_LRU was
2418 required.) [To be documented.]
2419 HighTotal %lu
2421 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2422 Total amount of highmem. Highmem is all memory above
2423 ~860 MB of physical memory. Highmem areas are for use by
2424 user-space programs, or for the page cache. The kernel
2425 must use tricks to access this memory, making it slower
2426 to access than lowmem.
2427 HighFree %lu
2429 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2430 Amount of free highmem.
2431 LowTotal %lu
2433 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2434 Total amount of lowmem. Lowmem is memory which can be
2435 used for everything that highmem can be used for, but it
2436 is also available for the kernel's use for its own data
2437 structures. Among many other things, it is where every‐
2438 thing from Slab is allocated. Bad things happen when
2439 you're out of lowmem.
2440 LowFree %lu
2442 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2443 Amount of free lowmem.
2444 MmapCopy %lu (since Linux 2.6.29)
2446 (CONFIG_MMU is required.) [To be documented.]
2447 SwapTotal %lu
2449 Total amount of swap space available.
2450 SwapFree %lu
2452 Amount of swap space that is currently unused.
2453 Dirty %lu
2455 Memory which is waiting to get written back to the disk.
2456 Writeback %lu
2458 Memory which is actively being written back to the disk.
2459 AnonPages %lu (since Linux 2.6.18)
2461 Non-file backed pages mapped into user-space page tables.
2462 Mapped %lu
2464 Files which have been mapped into memory (with mmap(2)),
2465 such as libraries.
2466 Shmem %lu (since Linux 2.6.32)
2468 Amount of memory consumed in tmpfs(5) filesystems.
2469 KReclaimable %lu (since Linux 4.20)
2471 Kernel allocations that the kernel will attempt to re‐
2472 claim under memory pressure. Includes SReclaimable (be‐
2473 low), and other direct allocations with a shrinker.
2474 Slab %lu
2476 In-kernel data structures cache. (See slabinfo(5).)
2477 SReclaimable %lu (since Linux 2.6.19)
2479 Part of Slab, that might be reclaimed, such as caches.
2480 SUnreclaim %lu (since Linux 2.6.19)
2482 Part of Slab, that cannot be reclaimed on memory pres‐
2483 sure.
2484 KernelStack %lu (since Linux 2.6.32)
2486 Amount of memory allocated to kernel stacks.
2487 PageTables %lu (since Linux 2.6.18)
2489 Amount of memory dedicated to the lowest level of page
2490 tables.
2491 Quicklists %lu (since Linux 2.6.27)
2493 (CONFIG_QUICKLIST is required.) [To be documented.]
2494 NFS_Unstable %lu (since Linux 2.6.18)
2496 NFS pages sent to the server, but not yet committed to
2497 stable storage.
2498 Bounce %lu (since Linux 2.6.18)
2500 Memory used for block device "bounce buffers".
2501 WritebackTmp %lu (since Linux 2.6.26)
2503 Memory used by FUSE for temporary writeback buffers.
2504 CommitLimit %lu (since Linux 2.6.10)
2506 This is the total amount of memory currently available to
2507 be allocated on the system, expressed in kilobytes. This
2508 limit is adhered to only if strict overcommit accounting
2509 is enabled (mode 2 in /proc/sys/vm/overcommit_memory).
2510 The limit is calculated according to the formula de‐
2511 scribed under /proc/sys/vm/overcommit_memory. For fur‐
2512 ther details, see the kernel source file Documenta‐
2513 tion/vm/overcommit-accounting.rst.
2514 Committed_AS %lu
2516 The amount of memory presently allocated on the system.
2517 The committed memory is a sum of all of the memory which
2518 has been allocated by processes, even if it has not been
2519 "used" by them as of yet. A process which allocates 1 GB
2520 of memory (using malloc(3) or similar), but touches only
2521 300 MB of that memory will show up as using only 300 MB
2522 of memory even if it has the address space allocated for
2523 the entire 1 GB.
2524 This 1 GB is memory which has been "committed" to by the
2526 VM and can be used at any time by the allocating applica‐
2527 tion. With strict overcommit enabled on the system (mode
2528 2 in /proc/sys/vm/overcommit_memory), allocations which
2529 would exceed the CommitLimit will not be permitted. This
2530 is useful if one needs to guarantee that processes will
2531 not fail due to lack of memory once that memory has been
2532 successfully allocated.
2533 VmallocTotal %lu
2535 Total size of vmalloc memory area.
2536 VmallocUsed %lu
2538 Amount of vmalloc area which is used. Since Linux 4.4,
2539 this field is no longer calculated, and is hard coded as
2540 0. See /proc/vmallocinfo.
2541 VmallocChunk %lu
2543 Largest contiguous block of vmalloc area which is free.
2544 Since Linux 4.4, this field is no longer calculated and
2545 is hard coded as 0. See /proc/vmallocinfo.
2546 HardwareCorrupted %lu (since Linux 2.6.32)
2548 (CONFIG_MEMORY_FAILURE is required.) [To be documented.]
2549 LazyFree %lu (since Linux 4.12)
2551 Shows the amount of memory marked by madvise(2)
2552 MADV_FREE.
2553 AnonHugePages %lu (since Linux 2.6.38)
2555 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Non-file
2556 backed huge pages mapped into user-space page tables.
2557 ShmemHugePages %lu (since Linux 4.8)
2559 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Memory used
2560 by shared memory (shmem) and tmpfs(5) allocated with huge
2561 pages.
2562 ShmemPmdMapped %lu (since Linux 4.8)
2564 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Shared memory
2565 mapped into user space with huge pages.
2566 CmaTotal %lu (since Linux 3.1)
2568 Total CMA (Contiguous Memory Allocator) pages. (CON‐
2569 FIG_CMA is required.)
2570 CmaFree %lu (since Linux 3.1)
2572 Free CMA (Contiguous Memory Allocator) pages. (CON‐
2573 FIG_CMA is required.)
2574 HugePages_Total %lu
2576 (CONFIG_HUGETLB_PAGE is required.) The size of the pool
2577 of huge pages.
2578 HugePages_Free %lu
2580 (CONFIG_HUGETLB_PAGE is required.) The number of huge
2581 pages in the pool that are not yet allocated.
2582 HugePages_Rsvd %lu (since Linux 2.6.17)
2584 (CONFIG_HUGETLB_PAGE is required.) This is the number of
2585 huge pages for which a commitment to allocate from the
2586 pool has been made, but no allocation has yet been made.
2587 These reserved huge pages guarantee that an application
2588 will be able to allocate a huge page from the pool of
2589 huge pages at fault time.
2590 HugePages_Surp %lu (since Linux 2.6.24)
2592 (CONFIG_HUGETLB_PAGE is required.) This is the number of
2593 huge pages in the pool above the value in
2594 /proc/sys/vm/nr_hugepages. The maximum number of surplus
2595 huge pages is controlled by /proc/sys/vm/nr_overcom‐
2596 mit_hugepages.
2597 Hugepagesize %lu
2599 (CONFIG_HUGETLB_PAGE is required.) The size of huge
2600 pages.
2601 DirectMap4k %lu (since Linux 2.6.27)
2603 Number of bytes of RAM linearly mapped by kernel in 4 kB
2604 pages. (x86.)
2605 DirectMap4M %lu (since Linux 2.6.27)
2607 Number of bytes of RAM linearly mapped by kernel in 4 MB
2608 pages. (x86 with CONFIG_X86_64 or CONFIG_X86_PAE en‐
2609 abled.)
2610 DirectMap2M %lu (since Linux 2.6.27)
2612 Number of bytes of RAM linearly mapped by kernel in 2 MB
2613 pages. (x86 with neither CONFIG_X86_64 nor CON‐
2614 FIG_X86_PAE enabled.)
2615 DirectMap1G %lu (since Linux 2.6.27)
2617 (x86 with CONFIG_X86_64 and CONFIG_X86_DIRECT_GBPAGES en‐
2618 abled.)
2619 /proc/modules
2621 A text list of the modules that have been loaded by the system.
2622 See also lsmod(8).
2623 /proc/mounts
2625 Before kernel 2.4.19, this file was a list of all the filesys‐
2626 tems currently mounted on the system. With the introduction of
2627 per-process mount namespaces in Linux 2.4.19 (see mount_name‐
2628 spaces(7)), this file became a link to /proc/self/mounts, which
2629 lists the mounts of the process's own mount namespace. The for‐
2630 mat of this file is documented in fstab(5).
2631 /proc/mtrr
2633 Memory Type Range Registers. See the Linux kernel source file
2634 Documentation/x86/mtrr.txt (or Documentation/mtrr.txt before
2635 Linux 2.6.28) for details.
2636 /proc/net
2638 This directory contains various files and subdirectories con‐
2639 taining information about the networking layer. The files con‐
2640 tain ASCII structures and are, therefore, readable with cat(1).
2641 However, the standard netstat(8) suite provides much cleaner ac‐
2642 cess to these files.
2643 With the advent of network namespaces, various information re‐
2645 lating to the network stack is virtualized (see network_name‐
2646 spaces(7)). Thus, since Linux 2.6.25, /proc/net is a symbolic
2647 link to the directory /proc/self/net, which contains the same
2648 files and directories as listed below. However, these files and
2649 directories now expose information for the network namespace of
2650 which the process is a member.
2651 /proc/net/arp
2653 This holds an ASCII readable dump of the kernel ARP table used
2654 for address resolutions. It will show both dynamically learned
2655 and preprogrammed ARP entries. The format is:
2656 IP address HW type Flags HW address Mask Device
2658 192.168.0.50 0x1 0x2 00:50:BF:25:68:F3 * eth0
2659 192.168.0.250 0x1 0xc 00:00:00:00:00:00 * eth0
2660 Here "IP address" is the IPv4 address of the machine and the "HW
2662 type" is the hardware type of the address from RFC 826. The
2663 flags are the internal flags of the ARP structure (as defined in
2664 /usr/include/linux/if_arp.h) and the "HW address" is the data
2665 link layer mapping for that IP address if it is known.
2666 /proc/net/dev
2668 The dev pseudo-file contains network device status information.
2669 This gives the number of received and sent packets, the number
2670 of errors and collisions and other basic statistics. These are
2671 used by the ifconfig(8) program to report device status. The
2672 format is:
2673 Inter-| Receive | Transmit
2675 face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
2676 lo: 2776770 11307 0 0 0 0 0 0 2776770 11307 0 0 0 0 0 0
2677 eth0: 1215645 2751 0 0 0 0 0 0 1782404 4324 0 0 0 427 0 0
2678 ppp0: 1622270 5552 1 0 0 0 0 0 354130 5669 0 0 0 0 0 0
2679 tap0: 7714 81 0 0 0 0 0 0 7714 81 0 0 0 0 0 0
2680 /proc/net/dev_mcast
2682 Defined in /usr/src/linux/net/core/dev_mcast.c:
2683 indx interface_name dmi_u dmi_g dmi_address
2685 2 eth0 1 0 01005e000001
2686 3 eth1 1 0 01005e000001
2687 4 eth2 1 0 01005e000001
2688 /proc/net/igmp
2690 Internet Group Management Protocol. Defined in
2691 /usr/src/linux/net/core/igmp.c.
2692 /proc/net/rarp
2694 This file uses the same format as the arp file and contains the
2695 current reverse mapping database used to provide rarp(8) reverse
2696 address lookup services. If RARP is not configured into the
2697 kernel, this file will not be present.
2698 /proc/net/raw
2700 Holds a dump of the RAW socket table. Much of the information
2701 is not of use apart from debugging. The "sl" value is the ker‐
2702 nel hash slot for the socket, the "local_address" is the local
2703 address and protocol number pair. "St" is the internal status
2704 of the socket. The "tx_queue" and "rx_queue" are the outgoing
2705 and incoming data queue in terms of kernel memory usage. The
2706 "tr", "tm->when", and "rexmits" fields are not used by RAW. The
2707 "uid" field holds the effective UID of the creator of the
2708 socket.
2709 /proc/net/snmp
2711 This file holds the ASCII data needed for the IP, ICMP, TCP, and
2712 UDP management information bases for an SNMP agent.
2713 /proc/net/tcp
2715 Holds a dump of the TCP socket table. Much of the information
2716 is not of use apart from debugging. The "sl" value is the ker‐
2717 nel hash slot for the socket, the "local_address" is the local
2718 address and port number pair. The "rem_address" is the remote
2719 address and port number pair (if connected). "St" is the inter‐
2720 nal status of the socket. The "tx_queue" and "rx_queue" are the
2721 outgoing and incoming data queue in terms of kernel memory us‐
2722 age. The "tr", "tm->when", and "rexmits" fields hold internal
2723 information of the kernel socket state and are useful only for
2724 debugging. The "uid" field holds the effective UID of the cre‐
2725 ator of the socket.
2726 /proc/net/udp
2728 Holds a dump of the UDP socket table. Much of the information
2729 is not of use apart from debugging. The "sl" value is the ker‐
2730 nel hash slot for the socket, the "local_address" is the local
2731 address and port number pair. The "rem_address" is the remote
2732 address and port number pair (if connected). "St" is the inter‐
2733 nal status of the socket. The "tx_queue" and "rx_queue" are the
2734 outgoing and incoming data queue in terms of kernel memory us‐
2735 age. The "tr", "tm->when", and "rexmits" fields are not used by
2736 UDP. The "uid" field holds the effective UID of the creator of
2737 the socket. The format is:
2738 sl local_address rem_address st tx_queue rx_queue tr rexmits tm->when uid
2740 1: 01642C89:0201 0C642C89:03FF 01 00000000:00000001 01:000071BA 00000000 0
2741 1: 00000000:0801 00000000:0000 0A 00000000:00000000 00:00000000 6F000100 0
2742 1: 00000000:0201 00000000:0000 0A 00000000:00000000 00:00000000 00000000 0
2743 /proc/net/unix
2745 Lists the UNIX domain sockets present within the system and
2746 their status. The format is:
2747 Num RefCount Protocol Flags Type St Inode Path
2749 0: 00000002 00000000 00000000 0001 03 42
2750 1: 00000001 00000000 00010000 0001 01 1948 /dev/printer
2751 The fields are as follows:
2753 Num: the kernel table slot number.
2755 RefCount: the number of users of the socket.
2757 Protocol: currently always 0.
2759 Flags: the internal kernel flags holding the status of the
2761 socket.
2762 Type: the socket type. For SOCK_STREAM sockets, this is
2764 0001; for SOCK_DGRAM sockets, it is 0002; and for
2765 SOCK_SEQPACKET sockets, it is 0005.
2766 St: the internal state of the socket.
2768 Inode: the inode number of the socket.
2770 Path: the bound pathname (if any) of the socket. Sockets in
2772 the abstract namespace are included in the list, and
2773 are shown with a Path that commences with the charac‐
2774 ter '@'.
2775 /proc/net/netfilter/nfnetlink_queue
2777 This file contains information about netfilter user-space queue‐
2778 ing, if used. Each line represents a queue. Queues that have
2779 not been subscribed to by user space are not shown.
2780 1 4207 0 2 65535 0 0 0 1
2782 (1) (2) (3)(4) (5) (6) (7) (8)
2783 The fields in each line are:
2785 (1) The ID of the queue. This matches what is specified in the
2787 --queue-num or --queue-balance options to the iptables(8)
2788 NFQUEUE target. See iptables-extensions(8) for more infor‐
2789 mation.
2790 (2) The netlink port ID subscribed to the queue.
2792 (3) The number of packets currently queued and waiting to be
2794 processed by the application.
2795 (4) The copy mode of the queue. It is either 1 (metadata only)
2797 or 2 (also copy payload data to user space).
2798 (5) Copy range; that is, how many bytes of packet payload
2800 should be copied to user space at most.
2801 (6) queue dropped. Number of packets that had to be dropped by
2803 the kernel because too many packets are already waiting for
2804 user space to send back the mandatory accept/drop verdicts.
2805 (7) queue user dropped. Number of packets that were dropped
2807 within the netlink subsystem. Such drops usually happen
2808 when the corresponding socket buffer is full; that is, user
2809 space is not able to read messages fast enough.
2810 (8) sequence number. Every queued packet is associated with a
2812 (32-bit) monotonically increasing sequence number. This
2813 shows the ID of the most recent packet queued.
2814 The last number exists only for compatibility reasons and is al‐
2816 ways 1.
2817 /proc/partitions
2819 Contains the major and minor numbers of each partition as well
2820 as the number of 1024-byte blocks and the partition name.
2821 /proc/pci
2823 This is a listing of all PCI devices found during kernel ini‐
2824 tialization and their configuration.
2825 This file has been deprecated in favor of a new /proc interface
2827 for PCI (/proc/bus/pci). It became optional in Linux 2.2
2828 (available with CONFIG_PCI_OLD_PROC set at kernel compilation).
2829 It became once more nonoptionally enabled in Linux 2.4. Next,
2830 it was deprecated in Linux 2.6 (still available with CON‐
2831 FIG_PCI_LEGACY_PROC set), and finally removed altogether since
2832 Linux 2.6.17.
2833 /proc/profile (since Linux 2.4)
2835 This file is present only if the kernel was booted with the pro‐
2836 file=1 command-line option. It exposes kernel profiling infor‐
2837 mation in a binary format for use by readprofile(1). Writing
2838 (e.g., an empty string) to this file resets the profiling coun‐
2839 ters; on some architectures, writing a binary integer "profiling
2840 multiplier" of size sizeof(int) sets the profiling interrupt
2841 frequency.
2842 /proc/scsi
2844 A directory with the scsi mid-level pseudo-file and various SCSI
2845 low-level driver directories, which contain a file for each SCSI
2846 host in this system, all of which give the status of some part
2847 of the SCSI IO subsystem. These files contain ASCII structures
2848 and are, therefore, readable with cat(1).
2849 You can also write to some of the files to reconfigure the sub‐
2851 system or switch certain features on or off.
2852 /proc/scsi/scsi
2854 This is a listing of all SCSI devices known to the kernel. The
2855 listing is similar to the one seen during bootup. scsi cur‐
2856 rently supports only the add-single-device command which allows
2857 root to add a hotplugged device to the list of known devices.
2858 The command
2860 echo 'scsi add-single-device 1 0 5 0' > /proc/scsi/scsi
2862 will cause host scsi1 to scan on SCSI channel 0 for a device on
2864 ID 5 LUN 0. If there is already a device known on this address
2865 or the address is invalid, an error will be returned.
2866 /proc/scsi/[drivername]
2868 [drivername] can currently be NCR53c7xx, aha152x, aha1542,
2869 aha1740, aic7xxx, buslogic, eata_dma, eata_pio, fdomain, in2000,
2870 pas16, qlogic, scsi_debug, seagate, t128, u15-24f, ultrastore,
2871 or wd7000. These directories show up for all drivers that reg‐
2872 istered at least one SCSI HBA. Every directory contains one
2873 file per registered host. Every host-file is named after the
2874 number the host was assigned during initialization.
2875 Reading these files will usually show driver and host configura‐
2877 tion, statistics, and so on.
2878 Writing to these files allows different things on different
2880 hosts. For example, with the latency and nolatency commands,
2881 root can switch on and off command latency measurement code in
2882 the eata_dma driver. With the lockup and unlock commands, root
2883 can control bus lockups simulated by the scsi_debug driver.
2884 /proc/self
2886 This directory refers to the process accessing the /proc
2887 filesystem, and is identical to the /proc directory named by the
2888 process ID of the same process.
2889 /proc/slabinfo
2891 Information about kernel caches. See slabinfo(5) for details.
2892 /proc/stat
2894 kernel/system statistics. Varies with architecture. Common en‐
2895 tries include:
2896 cpu 10132153 290696 3084719 46828483 16683 0 25195 0 175628 0
2898 cpu0 1393280 32966 572056 13343292 6130 0 17875 0 23933 0
2899 The amount of time, measured in units of USER_HZ
2900 (1/100ths of a second on most architectures, use
2901 sysconf(_SC_CLK_TCK) to obtain the right value), that the
2902 system ("cpu" line) or the specific CPU ("cpuN" line)
2903 spent in various states:
2904 user (1) Time spent in user mode.
2906 nice (2) Time spent in user mode with low priority
2908 (nice).
2909 system (3) Time spent in system mode.
2911 idle (4) Time spent in the idle task. This value
2913 should be USER_HZ times the second entry in the
2914 /proc/uptime pseudo-file.
2915 iowait (since Linux 2.5.41)
2917 (5) Time waiting for I/O to complete. This value
2918 is not reliable, for the following reasons:
2919 1. The CPU will not wait for I/O to complete;
2921 iowait is the time that a task is waiting for
2922 I/O to complete. When a CPU goes into idle
2923 state for outstanding task I/O, another task
2924 will be scheduled on this CPU.
2925 2. On a multi-core CPU, the task waiting for I/O
2927 to complete is not running on any CPU, so the
2928 iowait of each CPU is difficult to calculate.
2929 3. The value in this field may decrease in certain
2931 conditions.
2932 irq (since Linux 2.6.0)
2934 (6) Time servicing interrupts.
2935 softirq (since Linux 2.6.0)
2937 (7) Time servicing softirqs.
2938 steal (since Linux 2.6.11)
2940 (8) Stolen time, which is the time spent in other
2941 operating systems when running in a virtualized
2942 environment
2943 guest (since Linux 2.6.24)
2945 (9) Time spent running a virtual CPU for guest op‐
2946 erating systems under the control of the Linux
2947 kernel.
2948 guest_nice (since Linux 2.6.33)
2950 (10) Time spent running a niced guest (virtual CPU
2951 for guest operating systems under the control of
2952 the Linux kernel).
2953 page 5741 1808
2955 The number of pages the system paged in and the number
2956 that were paged out (from disk).
2957 swap 1 0
2959 The number of swap pages that have been brought in and
2960 out.
2961 intr 1462898
2963 This line shows counts of interrupts serviced since boot
2964 time, for each of the possible system interrupts. The
2965 first column is the total of all interrupts serviced in‐
2966 cluding unnumbered architecture specific interrupts; each
2967 subsequent column is the total for that particular num‐
2968 bered interrupt. Unnumbered interrupts are not shown,
2969 only summed into the total.
2970 disk_io: (2,0):(31,30,5764,1,2) (3,0):...
2972 (major,disk_idx):(noinfo, read_io_ops, blks_read,
2973 write_io_ops, blks_written)
2974 (Linux 2.4 only)
2975 ctxt 115315
2977 The number of context switches that the system underwent.
2978 btime 769041601
2980 boot time, in seconds since the Epoch, 1970-01-01
2981 00:00:00 +0000 (UTC).
2982 processes 86031
2984 Number of forks since boot.
2985 procs_running 6
2987 Number of processes in runnable state. (Linux 2.5.45 on‐
2988 ward.)
2989 procs_blocked 2
2991 Number of processes blocked waiting for I/O to complete.
2992 (Linux 2.5.45 onward.)
2993 softirq 229245889 94 60001584 13619 5175704 2471304 28 51212741
2995 59130143 0 51240672
2996 This line shows the number of softirq for all CPUs. The
2997 first column is the total of all softirqs and each subse‐
2998 quent column is the total for particular softirq. (Linux
2999 2.6.31 onward.)
3000 /proc/swaps
3002 Swap areas in use. See also swapon(8).
3003 /proc/sys
3005 This directory (present since 1.3.57) contains a number of files
3006 and subdirectories corresponding to kernel variables. These
3007 variables can be read and in some cases modified using the /proc
3008 filesystem, and the (deprecated) sysctl(2) system call.
3009 String values may be terminated by either '\0' or '\n'.
3011 Integer and long values may be written either in decimal or in
3013 hexadecimal notation (e.g., 0x3FFF). When writing multiple in‐
3014 teger or long values, these may be separated by any of the fol‐
3015 lowing whitespace characters: ' ', '\t', or '\n'. Using other
3016 separators leads to the error EINVAL.
3017 /proc/sys/abi (since Linux 2.4.10)
3019 This directory may contain files with application binary infor‐
3020 mation. See the Linux kernel source file Documenta‐
3021 tion/sysctl/abi.txt for more information.
3022 /proc/sys/debug
3024 This directory may be empty.
3025 /proc/sys/dev
3027 This directory contains device-specific information (e.g.,
3028 dev/cdrom/info). On some systems, it may be empty.
3029 /proc/sys/fs
3031 This directory contains the files and subdirectories for kernel
3032 variables related to filesystems.
3033 /proc/sys/fs/aio-max-nr and /proc/sys/fs/aio-nr (since Linux 2.6.4)
3035 aio-nr is the running total of the number of events specified by
3036 io_setup(2) calls for all currently active AIO contexts. If
3037 aio-nr reaches aio-max-nr, then io_setup(2) will fail with the
3038 error EAGAIN. Raising aio-max-nr does not result in the preal‐
3039 location or resizing of any kernel data structures.
3040 /proc/sys/fs/binfmt_misc
3042 Documentation for files in this directory can be found in the
3043 Linux kernel source in the file Documentation/ad‐
3044 min-guide/binfmt-misc.rst (or in Documentation/binfmt_misc.txt
3045 on older kernels).
3046 /proc/sys/fs/dentry-state (since Linux 2.2)
3048 This file contains information about the status of the directory
3049 cache (dcache). The file contains six numbers, nr_dentry,
3050 nr_unused, age_limit (age in seconds), want_pages (pages re‐
3051 quested by system) and two dummy values.
3052 * nr_dentry is the number of allocated dentries (dcache en‐
3054 tries). This field is unused in Linux 2.2.
3055 * nr_unused is the number of unused dentries.
3057 * age_limit is the age in seconds after which dcache entries can
3059 be reclaimed when memory is short.
3060 * want_pages is nonzero when the kernel has called
3062 shrink_dcache_pages() and the dcache isn't pruned yet.
3063 /proc/sys/fs/dir-notify-enable
3065 This file can be used to disable or enable the dnotify interface
3066 described in fcntl(2) on a system-wide basis. A value of 0 in
3067 this file disables the interface, and a value of 1 enables it.
3068 /proc/sys/fs/dquot-max
3070 This file shows the maximum number of cached disk quota entries.
3071 On some (2.4) systems, it is not present. If the number of free
3072 cached disk quota entries is very low and you have some awesome
3073 number of simultaneous system users, you might want to raise the
3074 limit.
3075 /proc/sys/fs/dquot-nr
3077 This file shows the number of allocated disk quota entries and
3078 the number of free disk quota entries.
3079 /proc/sys/fs/epoll (since Linux 2.6.28)
3081 This directory contains the file max_user_watches, which can be
3082 used to limit the amount of kernel memory consumed by the epoll
3083 interface. For further details, see epoll(7).
3084 /proc/sys/fs/file-max
3086 This file defines a system-wide limit on the number of open
3087 files for all processes. System calls that fail when encounter‐
3088 ing this limit fail with the error ENFILE. (See also setr‐
3089 limit(2), which can be used by a process to set the per-process
3090 limit, RLIMIT_NOFILE, on the number of files it may open.) If
3091 you get lots of error messages in the kernel log about running
3092 out of file handles (open file descriptions) (look for "VFS:
3093 file-max limit <number> reached"), try increasing this value:
3094 echo 100000 > /proc/sys/fs/file-max
3096 Privileged processes (CAP_SYS_ADMIN) can override the file-max
3098 limit.
3099 /proc/sys/fs/file-nr
3101 This (read-only) file contains three numbers: the number of al‐
3102 located file handles (i.e., the number of open file descrip‐
3103 tions; see open(2)); the number of free file handles; and the
3104 maximum number of file handles (i.e., the same value as
3105 /proc/sys/fs/file-max). If the number of allocated file handles
3106 is close to the maximum, you should consider increasing the max‐
3107 imum. Before Linux 2.6, the kernel allocated file handles dy‐
3108 namically, but it didn't free them again. Instead the free file
3109 handles were kept in a list for reallocation; the "free file
3110 handles" value indicates the size of that list. A large number
3111 of free file handles indicates that there was a past peak in the
3112 usage of open file handles. Since Linux 2.6, the kernel does
3113 deallocate freed file handles, and the "free file handles" value
3114 is always zero.
3115 /proc/sys/fs/inode-max (only present until Linux 2.2)
3117 This file contains the maximum number of in-memory inodes. This
3118 value should be 3–4 times larger than the value in file-max,
3119 since stdin, stdout and network sockets also need an inode to
3120 handle them. When you regularly run out of inodes, you need to
3121 increase this value.
3122 Starting with Linux 2.4, there is no longer a static limit on
3124 the number of inodes, and this file is removed.
3125 /proc/sys/fs/inode-nr
3127 This file contains the first two values from inode-state.
3128 /proc/sys/fs/inode-state
3130 This file contains seven numbers: nr_inodes, nr_free_inodes,
3131 preshrink, and four dummy values (always zero).
3132 nr_inodes is the number of inodes the system has allocated.
3134 nr_free_inodes represents the number of free inodes.
3135 preshrink is nonzero when the nr_inodes > inode-max and the sys‐
3137 tem needs to prune the inode list instead of allocating more;
3138 since Linux 2.4, this field is a dummy value (always zero).
3139 /proc/sys/fs/inotify (since Linux 2.6.13)
3141 This directory contains files max_queued_events, max_user_in‐
3142 stances, and max_user_watches, that can be used to limit the
3143 amount of kernel memory consumed by the inotify interface. For
3144 further details, see inotify(7).
3145 /proc/sys/fs/lease-break-time
3147 This file specifies the grace period that the kernel grants to a
3148 process holding a file lease (fcntl(2)) after it has sent a sig‐
3149 nal to that process notifying it that another process is waiting
3150 to open the file. If the lease holder does not remove or down‐
3151 grade the lease within this grace period, the kernel forcibly
3152 breaks the lease.
3153 /proc/sys/fs/leases-enable
3155 This file can be used to enable or disable file leases (fc‐
3156 ntl(2)) on a system-wide basis. If this file contains the value
3157 0, leases are disabled. A nonzero value enables leases.
3158 /proc/sys/fs/mount-max (since Linux 4.9)
3160 The value in this file specifies the maximum number of mounts
3161 that may exist in a mount namespace. The default value in this
3162 file is 100,000.
3163 /proc/sys/fs/mqueue (since Linux 2.6.6)
3165 This directory contains files msg_max, msgsize_max, and
3166 queues_max, controlling the resources used by POSIX message
3167 queues. See mq_overview(7) for details.
3168 /proc/sys/fs/nr_open (since Linux 2.6.25)
3170 This file imposes a ceiling on the value to which the
3171 RLIMIT_NOFILE resource limit can be raised (see getrlimit(2)).
3172 This ceiling is enforced for both unprivileged and privileged
3173 process. The default value in this file is 1048576. (Before
3174 Linux 2.6.25, the ceiling for RLIMIT_NOFILE was hard-coded to
3175 the same value.)
3176 /proc/sys/fs/overflowgid and /proc/sys/fs/overflowuid
3178 These files allow you to change the value of the fixed UID and
3179 GID. The default is 65534. Some filesystems support only
3180 16-bit UIDs and GIDs, although in Linux UIDs and GIDs are 32
3181 bits. When one of these filesystems is mounted with writes en‐
3182 abled, any UID or GID that would exceed 65535 is translated to
3183 the overflow value before being written to disk.
3184 /proc/sys/fs/pipe-max-size (since Linux 2.6.35)
3186 See pipe(7).
3187 /proc/sys/fs/pipe-user-pages-hard (since Linux 4.5)
3189 See pipe(7).
3190 /proc/sys/fs/pipe-user-pages-soft (since Linux 4.5)
3192 See pipe(7).
3193 /proc/sys/fs/protected_fifos (since Linux 4.19)
3195 The value in this file is/can be set to one of the following:
3196 0 Writing to FIFOs is unrestricted.
3198 1 Don't allow O_CREAT open(2) on FIFOs that the caller doesn't
3200 own in world-writable sticky directories, unless the FIFO is
3201 owned by the owner of the directory.
3202 2 As for the value 1, but the restriction also applies to
3204 group-writable sticky directories.
3205 The intent of the above protections is to avoid unintentional
3207 writes to an attacker-controlled FIFO when a program expected to
3208 create a regular file.
3209 /proc/sys/fs/protected_hardlinks (since Linux 3.6)
3211 When the value in this file is 0, no restrictions are placed on
3212 the creation of hard links (i.e., this is the historical behav‐
3213 ior before Linux 3.6). When the value in this file is 1, a hard
3214 link can be created to a target file only if one of the follow‐
3215 ing conditions is true:
3216 * The calling process has the CAP_FOWNER capability in its user
3218 namespace and the file UID has a mapping in the namespace.
3219 * The filesystem UID of the process creating the link matches
3221 the owner (UID) of the target file (as described in creden‐
3222 tials(7), a process's filesystem UID is normally the same as
3223 its effective UID).
3224 * All of the following conditions are true:
3226 • the target is a regular file;
3228 • the target file does not have its set-user-ID mode bit
3230 enabled;
3231 • the target file does not have both its set-group-ID and
3233 group-executable mode bits enabled; and
3234 • the caller has permission to read and write the target
3236 file (either via the file's permissions mask or because
3237 it has suitable capabilities).
3238 The default value in this file is 0. Setting the value to 1
3240 prevents a longstanding class of security issues caused by hard-
3241 link-based time-of-check, time-of-use races, most commonly seen
3242 in world-writable directories such as /tmp. The common method
3243 of exploiting this flaw is to cross privilege boundaries when
3244 following a given hard link (i.e., a root process follows a hard
3245 link created by another user). Additionally, on systems without
3246 separated partitions, this stops unauthorized users from "pin‐
3247 ning" vulnerable set-user-ID and set-group-ID files against be‐
3248 ing upgraded by the administrator, or linking to special files.
3249 /proc/sys/fs/protected_regular (since Linux 4.19)
3251 The value in this file is/can be set to one of the following:
3252 0 Writing to regular files is unrestricted.
3254 1 Don't allow O_CREAT open(2) on regular files that the caller
3256 doesn't own in world-writable sticky directories, unless the
3257 regular file is owned by the owner of the directory.
3258 2 As for the value 1, but the restriction also applies to
3260 group-writable sticky directories.
3261 The intent of the above protections is similar to protected_fi‐
3263 fos, but allows an application to avoid writes to an attacker-
3264 controlled regular file, where the application expected to cre‐
3265 ate one.
3266 /proc/sys/fs/protected_symlinks (since Linux 3.6)
3268 When the value in this file is 0, no restrictions are placed on
3269 following symbolic links (i.e., this is the historical behavior
3270 before Linux 3.6). When the value in this file is 1, symbolic
3271 links are followed only in the following circumstances:
3272 * the filesystem UID of the process following the link matches
3274 the owner (UID) of the symbolic link (as described in creden‐
3275 tials(7), a process's filesystem UID is normally the same as
3276 its effective UID);
3277 * the link is not in a sticky world-writable directory; or
3279 * the symbolic link and its parent directory have the same
3281 owner (UID)
3282 A system call that fails to follow a symbolic link because of
3284 the above restrictions returns the error EACCES in errno.
3285 The default value in this file is 0. Setting the value to 1
3287 avoids a longstanding class of security issues based on time-of-
3288 check, time-of-use races when accessing symbolic links.
3289 /proc/sys/fs/suid_dumpable (since Linux 2.6.13)
3291 The value in this file is assigned to a process's "dumpable"
3292 flag in the circumstances described in prctl(2). In effect, the
3293 value in this file determines whether core dump files are pro‐
3294 duced for set-user-ID or otherwise protected/tainted binaries.
3295 The "dumpable" setting also affects the ownership of files in a
3296 process's /proc/[pid] directory, as described above.
3297 Three different integer values can be specified:
3299 0 (default)
3301 This provides the traditional (pre-Linux 2.6.13) behav‐
3302 ior. A core dump will not be produced for a process
3303 which has changed credentials (by calling seteuid(2),
3304 setgid(2), or similar, or by executing a set-user-ID or
3305 set-group-ID program) or whose binary does not have read
3306 permission enabled.
3307 1 ("debug")
3309 All processes dump core when possible. (Reasons why a
3310 process might nevertheless not dump core are described in
3311 core(5).) The core dump is owned by the filesystem user
3312 ID of the dumping process and no security is applied.
3313 This is intended for system debugging situations only:
3314 this mode is insecure because it allows unprivileged
3315 users to examine the memory contents of privileged pro‐
3316 cesses.
3317 2 ("suidsafe")
3319 Any binary which normally would not be dumped (see "0"
3320 above) is dumped readable by root only. This allows the
3321 user to remove the core dump file but not to read it.
3322 For security reasons core dumps in this mode will not
3323 overwrite one another or other files. This mode is ap‐
3324 propriate when administrators are attempting to debug
3325 problems in a normal environment.
3326 Additionally, since Linux 3.6, /proc/sys/kernel/core_pat‐
3328 tern must either be an absolute pathname or a pipe com‐
3329 mand, as detailed in core(5). Warnings will be written
3330 to the kernel log if core_pattern does not follow these
3331 rules, and no core dump will be produced.
3332 For details of the effect of a process's "dumpable" setting on
3334 ptrace access mode checking, see ptrace(2).
3335 /proc/sys/fs/super-max
3337 This file controls the maximum number of superblocks, and thus
3338 the maximum number of mounted filesystems the kernel can have.
3339 You need increase only super-max if you need to mount more
3340 filesystems than the current value in super-max allows you to.
3341 /proc/sys/fs/super-nr
3343 This file contains the number of filesystems currently mounted.
3344 /proc/sys/kernel
3346 This directory contains files controlling a range of kernel pa‐
3347 rameters, as described below.
3348 /proc/sys/kernel/acct
3350 This file contains three numbers: highwater, lowwater, and fre‐
3351 quency. If BSD-style process accounting is enabled, these val‐
3352 ues control its behavior. If free space on filesystem where the
3353 log lives goes below lowwater percent, accounting suspends. If
3354 free space gets above highwater percent, accounting resumes.
3355 frequency determines how often the kernel checks the amount of
3356 free space (value is in seconds). Default values are 4, 2, and
3357 30. That is, suspend accounting if 2% or less space is free;
3358 resume it if 4% or more space is free; consider information
3359 about amount of free space valid for 30 seconds.
3360 /proc/sys/kernel/auto_msgmni (Linux 2.6.27 to 3.18)
3362 From Linux 2.6.27 to 3.18, this file was used to control recom‐
3363 puting of the value in /proc/sys/kernel/msgmni upon the addition
3364 or removal of memory or upon IPC namespace creation/removal.
3365 Echoing "1" into this file enabled msgmni automatic recomputing
3366 (and triggered a recomputation of msgmni based on the current
3367 amount of available memory and number of IPC namespaces). Echo‐
3368 ing "0" disabled automatic recomputing. (Automatic recomputing
3369 was also disabled if a value was explicitly assigned to
3370 /proc/sys/kernel/msgmni.) The default value in auto_msgmni was
3371 1.
3372 Since Linux 3.19, the content of this file has no effect (be‐
3374 cause msgmni defaults to near the maximum value possible), and
3375 reads from this file always return the value "0".
3376 /proc/sys/kernel/cap_last_cap (since Linux 3.2)
3378 See capabilities(7).
3379 /proc/sys/kernel/cap-bound (from Linux 2.2 to 2.6.24)
3381 This file holds the value of the kernel capability bounding set
3382 (expressed as a signed decimal number). This set is ANDed
3383 against the capabilities permitted to a process during ex‐
3384 ecve(2). Starting with Linux 2.6.25, the system-wide capability
3385 bounding set disappeared, and was replaced by a per-thread
3386 bounding set; see capabilities(7).
3387 /proc/sys/kernel/core_pattern
3389 See core(5).
3390 /proc/sys/kernel/core_pipe_limit
3392 See core(5).
3393 /proc/sys/kernel/core_uses_pid
3395 See core(5).
3396 /proc/sys/kernel/ctrl-alt-del
3398 This file controls the handling of Ctrl-Alt-Del from the key‐
3399 board. When the value in this file is 0, Ctrl-Alt-Del is
3400 trapped and sent to the init(1) program to handle a graceful
3401 restart. When the value is greater than zero, Linux's reaction
3402 to a Vulcan Nerve Pinch (tm) will be an immediate reboot, with‐
3403 out even syncing its dirty buffers. Note: when a program (like
3404 dosemu) has the keyboard in "raw" mode, the ctrl-alt-del is in‐
3405 tercepted by the program before it ever reaches the kernel tty
3406 layer, and it's up to the program to decide what to do with it.
3407 /proc/sys/kernel/dmesg_restrict (since Linux 2.6.37)
3409 The value in this file determines who can see kernel syslog con‐
3410 tents. A value of 0 in this file imposes no restrictions. If
3411 the value is 1, only privileged users can read the kernel sys‐
3412 log. (See syslog(2) for more details.) Since Linux 3.4, only
3413 users with the CAP_SYS_ADMIN capability may change the value in
3414 this file.
3415 /proc/sys/kernel/domainname and /proc/sys/kernel/hostname
3417 can be used to set the NIS/YP domainname and the hostname of
3418 your box in exactly the same way as the commands domainname(1)
3419 and hostname(1), that is:
3420 # echo 'darkstar' > /proc/sys/kernel/hostname
3422 # echo 'mydomain' > /proc/sys/kernel/domainname
3423 has the same effect as
3425 # hostname 'darkstar'
3427 # domainname 'mydomain'
3428 Note, however, that the classic darkstar.frop.org has the host‐
3430 name "darkstar" and DNS (Internet Domain Name Server) domainname
3431 "frop.org", not to be confused with the NIS (Network Information
3432 Service) or YP (Yellow Pages) domainname. These two domain
3433 names are in general different. For a detailed discussion see
3434 the hostname(1) man page.
3435 /proc/sys/kernel/hotplug
3437 This file contains the pathname for the hotplug policy agent.
3438 The default value in this file is /sbin/hotplug.
3439 /proc/sys/kernel/htab-reclaim (before Linux 2.4.9.2)
3441 (PowerPC only) If this file is set to a nonzero value, the Pow‐
3442 erPC htab (see kernel file Documentation/powerpc/ppc_htab.txt)
3443 is pruned each time the system hits the idle loop.
3444 /proc/sys/kernel/keys/*
3446 This directory contains various files that define parameters and
3447 limits for the key-management facility. These files are de‐
3448 scribed in keyrings(7).
3449 /proc/sys/kernel/kptr_restrict (since Linux 2.6.38)
3451 The value in this file determines whether kernel addresses are
3452 exposed via /proc files and other interfaces. A value of 0 in
3453 this file imposes no restrictions. If the value is 1, kernel
3454 pointers printed using the %pK format specifier will be replaced
3455 with zeros unless the user has the CAP_SYSLOG capability. If
3456 the value is 2, kernel pointers printed using the %pK format
3457 specifier will be replaced with zeros regardless of the user's
3458 capabilities. The initial default value for this file was 1,
3459 but the default was changed to 0 in Linux 2.6.39. Since Linux
3460 3.4, only users with the CAP_SYS_ADMIN capability can change the
3461 value in this file.
3462 /proc/sys/kernel/l2cr
3464 (PowerPC only) This file contains a flag that controls the L2
3465 cache of G3 processor boards. If 0, the cache is disabled. En‐
3466 abled if nonzero.
3467 /proc/sys/kernel/modprobe
3469 This file contains the pathname for the kernel module loader.
3470 The default value is /sbin/modprobe. The file is present only
3471 if the kernel is built with the CONFIG_MODULES (CONFIG_KMOD in
3472 Linux 2.6.26 and earlier) option enabled. It is described by
3473 the Linux kernel source file Documentation/kmod.txt (present
3474 only in kernel 2.4 and earlier).
3475 /proc/sys/kernel/modules_disabled (since Linux 2.6.31)
3477 A toggle value indicating if modules are allowed to be loaded in
3478 an otherwise modular kernel. This toggle defaults to off (0),
3479 but can be set true (1). Once true, modules can be neither
3480 loaded nor unloaded, and the toggle cannot be set back to false.
3481 The file is present only if the kernel is built with the CON‐
3482 FIG_MODULES option enabled.
3483 /proc/sys/kernel/msgmax (since Linux 2.2)
3485 This file defines a system-wide limit specifying the maximum
3486 number of bytes in a single message written on a System V mes‐
3487 sage queue.
3488 /proc/sys/kernel/msgmni (since Linux 2.4)
3490 This file defines the system-wide limit on the number of message
3491 queue identifiers. See also /proc/sys/kernel/auto_msgmni.
3492 /proc/sys/kernel/msgmnb (since Linux 2.2)
3494 This file defines a system-wide parameter used to initialize the
3495 msg_qbytes setting for subsequently created message queues. The
3496 msg_qbytes setting specifies the maximum number of bytes that
3497 may be written to the message queue.
3498 /proc/sys/kernel/ngroups_max (since Linux 2.6.4)
3500 This is a read-only file that displays the upper limit on the
3501 number of a process's group memberships.
3502 /proc/sys/kernel/ns_last_pid (since Linux 3.3)
3504 See pid_namespaces(7).
3505 /proc/sys/kernel/ostype and /proc/sys/kernel/osrelease
3507 These files give substrings of /proc/version.
3508 /proc/sys/kernel/overflowgid and /proc/sys/kernel/overflowuid
3510 These files duplicate the files /proc/sys/fs/overflowgid and
3511 /proc/sys/fs/overflowuid.
3512 /proc/sys/kernel/panic
3514 This file gives read/write access to the kernel variable
3515 panic_timeout. If this is zero, the kernel will loop on a
3516 panic; if nonzero, it indicates that the kernel should autore‐
3517 boot after this number of seconds. When you use the software
3518 watchdog device driver, the recommended setting is 60.
3519 /proc/sys/kernel/panic_on_oops (since Linux 2.5.68)
3521 This file controls the kernel's behavior when an oops or BUG is
3522 encountered. If this file contains 0, then the system tries to
3523 continue operation. If it contains 1, then the system delays a
3524 few seconds (to give klogd time to record the oops output) and
3525 then panics. If the /proc/sys/kernel/panic file is also non‐
3526 zero, then the machine will be rebooted.
3527 /proc/sys/kernel/pid_max (since Linux 2.5.34)
3529 This file specifies the value at which PIDs wrap around (i.e.,
3530 the value in this file is one greater than the maximum PID).
3531 PIDs greater than this value are not allocated; thus, the value
3532 in this file also acts as a system-wide limit on the total num‐
3533 ber of processes and threads. The default value for this file,
3534 32768, results in the same range of PIDs as on earlier kernels.
3535 On 32-bit platforms, 32768 is the maximum value for pid_max. On
3536 64-bit systems, pid_max can be set to any value up to 2^22
3537 (PID_MAX_LIMIT, approximately 4 million).
3538 /proc/sys/kernel/powersave-nap (PowerPC only)
3540 This file contains a flag. If set, Linux-PPC will use the "nap"
3541 mode of powersaving, otherwise the "doze" mode will be used.
3542 /proc/sys/kernel/printk
3544 See syslog(2).
3545 /proc/sys/kernel/pty (since Linux 2.6.4)
3547 This directory contains two files relating to the number of UNIX
3548 98 pseudoterminals (see pts(4)) on the system.
3549 /proc/sys/kernel/pty/max
3551 This file defines the maximum number of pseudoterminals.
3552 /proc/sys/kernel/pty/nr
3554 This read-only file indicates how many pseudoterminals are cur‐
3555 rently in use.
3556 /proc/sys/kernel/random
3558 This directory contains various parameters controlling the oper‐
3559 ation of the file /dev/random. See random(4) for further infor‐
3560 mation.
3561 /proc/sys/kernel/random/uuid (since Linux 2.4)
3563 Each read from this read-only file returns a randomly generated
3564 128-bit UUID, as a string in the standard UUID format.
3565 /proc/sys/kernel/randomize_va_space (since Linux 2.6.12)
3567 Select the address space layout randomization (ASLR) policy for
3568 the system (on architectures that support ASLR). Three values
3569 are supported for this file:
3570 0 Turn ASLR off. This is the default for architectures that
3572 don't support ASLR, and when the kernel is booted with the
3573 norandmaps parameter.
3574 1 Make the addresses of mmap(2) allocations, the stack, and the
3576 VDSO page randomized. Among other things, this means that
3577 shared libraries will be loaded at randomized addresses. The
3578 text segment of PIE-linked binaries will also be loaded at a
3579 randomized address. This value is the default if the kernel
3580 was configured with CONFIG_COMPAT_BRK.
3581 2 (Since Linux 2.6.25) Also support heap randomization. This
3583 value is the default if the kernel was not configured with
3584 CONFIG_COMPAT_BRK.
3585 /proc/sys/kernel/real-root-dev
3587 This file is documented in the Linux kernel source file Documen‐
3588 tation/admin-guide/initrd.rst (or Documentation/initrd.txt be‐
3589 fore Linux 4.10).
3590 /proc/sys/kernel/reboot-cmd (Sparc only)
3592 This file seems to be a way to give an argument to the SPARC
3593 ROM/Flash boot loader. Maybe to tell it what to do after re‐
3594 booting?
3595 /proc/sys/kernel/rtsig-max
3597 (Only in kernels up to and including 2.6.7; see setrlimit(2))
3598 This file can be used to tune the maximum number of POSIX real-
3599 time (queued) signals that can be outstanding in the system.
3600 /proc/sys/kernel/rtsig-nr
3602 (Only in kernels up to and including 2.6.7.) This file shows
3603 the number of POSIX real-time signals currently queued.
3604 /proc/[pid]/sched_autogroup_enabled (since Linux 2.6.38)
3606 See sched(7).
3607 /proc/sys/kernel/sched_child_runs_first (since Linux 2.6.23)
3609 If this file contains the value zero, then, after a fork(2), the
3610 parent is first scheduled on the CPU. If the file contains a
3611 nonzero value, then the child is scheduled first on the CPU.
3612 (Of course, on a multiprocessor system, the parent and the child
3613 might both immediately be scheduled on a CPU.)
3614 /proc/sys/kernel/sched_rr_timeslice_ms (since Linux 3.9)
3616 See sched_rr_get_interval(2).
3617 /proc/sys/kernel/sched_rt_period_us (since Linux 2.6.25)
3619 See sched(7).
3620 /proc/sys/kernel/sched_rt_runtime_us (since Linux 2.6.25)
3622 See sched(7).
3623 /proc/sys/kernel/seccomp (since Linux 4.14)
3625 This directory provides additional seccomp information and con‐
3626 figuration. See seccomp(2) for further details.
3627 /proc/sys/kernel/sem (since Linux 2.4)
3629 This file contains 4 numbers defining limits for System V IPC
3630 semaphores. These fields are, in order:
3631 SEMMSL The maximum semaphores per semaphore set.
3633 SEMMNS A system-wide limit on the number of semaphores in all
3635 semaphore sets.
3636 SEMOPM The maximum number of operations that may be specified
3638 in a semop(2) call.
3639 SEMMNI A system-wide limit on the maximum number of semaphore
3641 identifiers.
3642 /proc/sys/kernel/sg-big-buff
3644 This file shows the size of the generic SCSI device (sg) buffer.
3645 You can't tune it just yet, but you could change it at compile
3646 time by editing include/scsi/sg.h and changing the value of
3647 SG_BIG_BUFF. However, there shouldn't be any reason to change
3648 this value.
3649 /proc/sys/kernel/shm_rmid_forced (since Linux 3.1)
3651 If this file is set to 1, all System V shared memory segments
3652 will be marked for destruction as soon as the number of attached
3653 processes falls to zero; in other words, it is no longer possi‐
3654 ble to create shared memory segments that exist independently of
3655 any attached process.
3656 The effect is as though a shmctl(2) IPC_RMID is performed on all
3658 existing segments as well as all segments created in the future
3659 (until this file is reset to 0). Note that existing segments
3660 that are attached to no process will be immediately destroyed
3661 when this file is set to 1. Setting this option will also de‐
3662 stroy segments that were created, but never attached, upon ter‐
3663 mination of the process that created the segment with shmget(2).
3664 Setting this file to 1 provides a way of ensuring that all Sys‐
3666 tem V shared memory segments are counted against the resource
3667 usage and resource limits (see the description of RLIMIT_AS in
3668 getrlimit(2)) of at least one process.
3669 Because setting this file to 1 produces behavior that is non‐
3671 standard and could also break existing applications, the default
3672 value in this file is 0. Set this file to 1 only if you have a
3673 good understanding of the semantics of the applications using
3674 System V shared memory on your system.
3675 /proc/sys/kernel/shmall (since Linux 2.2)
3677 This file contains the system-wide limit on the total number of
3678 pages of System V shared memory.
3679 /proc/sys/kernel/shmmax (since Linux 2.2)
3681 This file can be used to query and set the run-time limit on the
3682 maximum (System V IPC) shared memory segment size that can be
3683 created. Shared memory segments up to 1 GB are now supported in
3684 the kernel. This value defaults to SHMMAX.
3685 /proc/sys/kernel/shmmni (since Linux 2.4)
3687 This file specifies the system-wide maximum number of System V
3688 shared memory segments that can be created.
3689 /proc/sys/kernel/sysctl_writes_strict (since Linux 3.16)
3691 The value in this file determines how the file offset affects
3692 the behavior of updating entries in files under /proc/sys. The
3693 file has three possible values:
3694 -1 This provides legacy handling, with no printk warnings.
3696 Each write(2) must fully contain the value to be written,
3697 and multiple writes on the same file descriptor will over‐
3698 write the entire value, regardless of the file position.
3699 0 (default) This provides the same behavior as for -1, but
3701 printk warnings are written for processes that perform
3702 writes when the file offset is not 0.
3703 1 Respect the file offset when writing strings into /proc/sys
3705 files. Multiple writes will append to the value buffer.
3706 Anything written beyond the maximum length of the value buf‐
3707 fer will be ignored. Writes to numeric /proc/sys entries
3708 must always be at file offset 0 and the value must be fully
3709 contained in the buffer provided to write(2).
3710 /proc/sys/kernel/sysrq
3712 This file controls the functions allowed to be invoked by the
3713 SysRq key. By default, the file contains 1 meaning that every
3714 possible SysRq request is allowed (in older kernel versions,
3715 SysRq was disabled by default, and you were required to specifi‐
3716 cally enable it at run-time, but this is not the case any more).
3717 Possible values in this file are:
3718 0 Disable sysrq completely
3720 1 Enable all functions of sysrq
3722 > 1 Bit mask of allowed sysrq functions, as follows:
3724 2 Enable control of console logging level
3725 4 Enable control of keyboard (SAK, unraw)
3726 8 Enable debugging dumps of processes etc.
3727 16 Enable sync command
3728 32 Enable remount read-only
3729 64 Enable signaling of processes (term, kill, oom-kill)
3730 128 Allow reboot/poweroff
3731 256 Allow nicing of all real-time tasks
3732 This file is present only if the CONFIG_MAGIC_SYSRQ kernel con‐
3734 figuration option is enabled. For further details see the Linux
3735 kernel source file Documentation/admin-guide/sysrq.rst (or Docu‐
3736 mentation/sysrq.txt before Linux 4.10).
3737 /proc/sys/kernel/version
3739 This file contains a string such as:
3740 #5 Wed Feb 25 21:49:24 MET 1998
3742 The "#5" means that this is the fifth kernel built from this
3744 source base and the date following it indicates the time the
3745 kernel was built.
3746 /proc/sys/kernel/threads-max (since Linux 2.3.11)
3748 This file specifies the system-wide limit on the number of
3749 threads (tasks) that can be created on the system.
3750 Since Linux 4.1, the value that can be written to threads-max is
3752 bounded. The minimum value that can be written is 20. The max‐
3753 imum value that can be written is given by the constant FU‐
3754 TEX_TID_MASK [22m(0x3fffffff). If a value outside of this range is
3755 written to threads-max, the error EINVAL occurs.
3756 The value written is checked against the available RAM pages.
3758 If the thread structures would occupy too much (more than 1/8th)
3759 of the available RAM pages, threads-max is reduced accordingly.
3760 /proc/sys/kernel/yama/ptrace_scope (since Linux 3.5)
3762 See ptrace(2).
3763 /proc/sys/kernel/zero-paged (PowerPC only)
3765 This file contains a flag. When enabled (nonzero), Linux-PPC
3766 will pre-zero pages in the idle loop, possibly speeding up
3767 get_free_pages.
3768 /proc/sys/net
3770 This directory contains networking stuff. Explanations for some
3771 of the files under this directory can be found in tcp(7) and
3772 ip(7).
3773 /proc/sys/net/core/bpf_jit_enable
3775 See bpf(2).
3776 /proc/sys/net/core/somaxconn
3778 This file defines a ceiling value for the backlog argument of
3779 listen(2); see the listen(2) manual page for details.
3780 /proc/sys/proc
3782 This directory may be empty.
3783 /proc/sys/sunrpc
3785 This directory supports Sun remote procedure call for network
3786 filesystem (NFS). On some systems, it is not present.
3787 /proc/sys/user (since Linux 4.9)
3789 See namespaces(7).
3790 /proc/sys/vm
3792 This directory contains files for memory management tuning, buf‐
3793 fer, and cache management.
3794 /proc/sys/vm/admin_reserve_kbytes (since Linux 3.10)
3796 This file defines the amount of free memory (in KiB) on the sys‐
3797 tem that should be reserved for users with the capability
3798 CAP_SYS_ADMIN.
3799 The default value in this file is the minimum of [3% of free
3801 pages, 8MiB] expressed as KiB. The default is intended to pro‐
3802 vide enough for the superuser to log in and kill a process, if
3803 necessary, under the default overcommit 'guess' mode (i.e., 0 in
3804 /proc/sys/vm/overcommit_memory).
3805 Systems running in "overcommit never" mode (i.e., 2 in
3807 /proc/sys/vm/overcommit_memory) should increase the value in
3808 this file to account for the full virtual memory size of the
3809 programs used to recover (e.g., login(1) ssh(1), and top(1))
3810 Otherwise, the superuser may not be able to log in to recover
3811 the system. For example, on x86-64 a suitable value is 131072
3812 (128MiB reserved).
3813 Changing the value in this file takes effect whenever an appli‐
3815 cation requests memory.
3816 /proc/sys/vm/compact_memory (since Linux 2.6.35)
3818 When 1 is written to this file, all zones are compacted such
3819 that free memory is available in contiguous blocks where possi‐
3820 ble. The effect of this action can be seen by examining
3821 /proc/buddyinfo.
3822 Present only if the kernel was configured with CONFIG_COM‐
3824 PACTION.
3825 /proc/sys/vm/drop_caches (since Linux 2.6.16)
3827 Writing to this file causes the kernel to drop clean caches,
3828 dentries, and inodes from memory, causing that memory to become
3829 free. This can be useful for memory management testing and per‐
3830 forming reproducible filesystem benchmarks. Because writing to
3831 this file causes the benefits of caching to be lost, it can de‐
3832 grade overall system performance.
3833 To free pagecache, use:
3835 echo 1 > /proc/sys/vm/drop_caches
3837 To free dentries and inodes, use:
3839 echo 2 > /proc/sys/vm/drop_caches
3841 To free pagecache, dentries, and inodes, use:
3843 echo 3 > /proc/sys/vm/drop_caches
3845 Because writing to this file is a nondestructive operation and
3847 dirty objects are not freeable, the user should run sync(1)
3848 first.
3849 /proc/sys/vm/sysctl_hugetlb_shm_group (since Linux 2.6.7)
3851 This writable file contains a group ID that is allowed to allo‐
3852 cate memory using huge pages. If a process has a filesystem
3853 group ID or any supplementary group ID that matches this group
3854 ID, then it can make huge-page allocations without holding the
3855 CAP_IPC_LOCK capability; see memfd_create(2), mmap(2), and
3856 shmget(2).
3857 /proc/sys/vm/legacy_va_layout (since Linux 2.6.9)
3859 If nonzero, this disables the new 32-bit memory-mapping layout;
3860 the kernel will use the legacy (2.4) layout for all processes.
3861 /proc/sys/vm/memory_failure_early_kill (since Linux 2.6.32)
3863 Control how to kill processes when an uncorrected memory error
3864 (typically a 2-bit error in a memory module) that cannot be han‐
3865 dled by the kernel is detected in the background by hardware.
3866 In some cases (like the page still having a valid copy on disk),
3867 the kernel will handle the failure transparently without affect‐
3868 ing any applications. But if there is no other up-to-date copy
3869 of the data, it will kill processes to prevent any data corrup‐
3870 tions from propagating.
3871 The file has one of the following values:
3873 1: Kill all processes that have the corrupted-and-not-reload‐
3875 able page mapped as soon as the corruption is detected.
3876 Note that this is not supported for a few types of pages,
3877 such as kernel internally allocated data or the swap cache,
3878 but works for the majority of user pages.
3879 0: Unmap the corrupted page from all processes and kill a
3881 process only if it tries to access the page.
3882 The kill is performed using a SIGBUS signal with si_code set to
3884 BUS_MCEERR_AO. Processes can handle this if they want to; see
3885 sigaction(2) for more details.
3886 This feature is active only on architectures/platforms with ad‐
3888 vanced machine check handling and depends on the hardware capa‐
3889 bilities.
3890 Applications can override the memory_failure_early_kill setting
3892 individually with the prctl(2) PR_MCE_KILL operation.
3893 Present only if the kernel was configured with CONFIG_MEM‐
3895 ORY_FAILURE.
3896 /proc/sys/vm/memory_failure_recovery (since Linux 2.6.32)
3898 Enable memory failure recovery (when supported by the platform).
3899 1: Attempt recovery.
3901 0: Always panic on a memory failure.
3903 Present only if the kernel was configured with CONFIG_MEM‐
3905 ORY_FAILURE.
3906 /proc/sys/vm/oom_dump_tasks (since Linux 2.6.25)
3908 Enables a system-wide task dump (excluding kernel threads) to be
3909 produced when the kernel performs an OOM-killing. The dump in‐
3910 cludes the following information for each task (thread,
3911 process): thread ID, real user ID, thread group ID (process ID),
3912 virtual memory size, resident set size, the CPU that the task is
3913 scheduled on, oom_adj score (see the description of
3914 /proc/[pid]/oom_adj), and command name. This is helpful to de‐
3915 termine why the OOM-killer was invoked and to identify the rogue
3916 task that caused it.
3917 If this contains the value zero, this information is suppressed.
3919 On very large systems with thousands of tasks, it may not be
3920 feasible to dump the memory state information for each one.
3921 Such systems should not be forced to incur a performance penalty
3922 in OOM situations when the information may not be desired.
3923 If this is set to nonzero, this information is shown whenever
3925 the OOM-killer actually kills a memory-hogging task.
3926 The default value is 0.
3928 /proc/sys/vm/oom_kill_allocating_task (since Linux 2.6.24)
3930 This enables or disables killing the OOM-triggering task in out-
3931 of-memory situations.
3932 If this is set to zero, the OOM-killer will scan through the en‐
3934 tire tasklist and select a task based on heuristics to kill.
3935 This normally selects a rogue memory-hogging task that frees up
3936 a large amount of memory when killed.
3937 If this is set to nonzero, the OOM-killer simply kills the task
3939 that triggered the out-of-memory condition. This avoids a pos‐
3940 sibly expensive tasklist scan.
3941 If /proc/sys/vm/panic_on_oom is nonzero, it takes precedence
3943 over whatever value is used in /proc/sys/vm/oom_kill_allocat‐
3944 ing_task.
3945 The default value is 0.
3947 /proc/sys/vm/overcommit_kbytes (since Linux 3.14)
3949 This writable file provides an alternative to /proc/sys/vm/over‐
3950 commit_ratio for controlling the CommitLimit when
3951 /proc/sys/vm/overcommit_memory has the value 2. It allows the
3952 amount of memory overcommitting to be specified as an absolute
3953 value (in kB), rather than as a percentage, as is done with
3954 overcommit_ratio. This allows for finer-grained control of Com‐
3955 mitLimit on systems with extremely large memory sizes.
3956 Only one of overcommit_kbytes or overcommit_ratio can have an
3958 effect: if overcommit_kbytes has a nonzero value, then it is
3959 used to calculate CommitLimit, otherwise overcommit_ratio is
3960 used. Writing a value to either of these files causes the value
3961 in the other file to be set to zero.
3962 /proc/sys/vm/overcommit_memory
3964 This file contains the kernel virtual memory accounting mode.
3965 Values are:
3966 0: heuristic overcommit (this is the default)
3968 1: always overcommit, never check
3969 2: always check, never overcommit
3970 In mode 0, calls of mmap(2) with MAP_NORESERVE are not checked,
3972 and the default check is very weak, leading to the risk of get‐
3973 ting a process "OOM-killed".
3974 In mode 1, the kernel pretends there is always enough memory,
3976 until memory actually runs out. One use case for this mode is
3977 scientific computing applications that employ large sparse ar‐
3978 rays. In Linux kernel versions before 2.6.0, any nonzero value
3979 implies mode 1.
3980 In mode 2 (available since Linux 2.6), the total virtual address
3982 space that can be allocated (CommitLimit in /proc/meminfo) is
3983 calculated as
3984 CommitLimit = (total_RAM - total_huge_TLB) *
3986 overcommit_ratio / 100 + total_swap
3987 where:
3989 * total_RAM is the total amount of RAM on the system;
3991 * total_huge_TLB is the amount of memory set aside for
3993 huge pages;
3994 * overcommit_ratio is the value in /proc/sys/vm/overcom‐
3996 mit_ratio; and
3997 * total_swap is the amount of swap space.
3999 For example, on a system with 16 GB of physical RAM, 16 GB of
4001 swap, no space dedicated to huge pages, and an overcommit_ratio
4002 of 50, this formula yields a CommitLimit of 24 GB.
4003 Since Linux 3.14, if the value in /proc/sys/vm/overcommit_kbytes
4005 is nonzero, then CommitLimit is instead calculated as:
4006 CommitLimit = overcommit_kbytes + total_swap
4008 See also the description of /proc/sys/vm/admin_reserve_kbytes
4010 and /proc/sys/vm/user_reserve_kbytes.
4011 /proc/sys/vm/overcommit_ratio (since Linux 2.6.0)
4013 This writable file defines a percentage by which memory can be
4014 overcommitted. The default value in the file is 50. See the
4015 description of /proc/sys/vm/overcommit_memory.
4016 /proc/sys/vm/panic_on_oom (since Linux 2.6.18)
4018 This enables or disables a kernel panic in an out-of-memory sit‐
4019 uation.
4020 If this file is set to the value 0, the kernel's OOM-killer will
4022 kill some rogue process. Usually, the OOM-killer is able to
4023 kill a rogue process and the system will survive.
4024 If this file is set to the value 1, then the kernel normally
4026 panics when out-of-memory happens. However, if a process limits
4027 allocations to certain nodes using memory policies (mbind(2)
4028 MPOL_BIND) or cpusets (cpuset(7)) and those nodes reach memory
4029 exhaustion status, one process may be killed by the OOM-killer.
4030 No panic occurs in this case: because other nodes' memory may be
4031 free, this means the system as a whole may not have reached an
4032 out-of-memory situation yet.
4033 If this file is set to the value 2, the kernel always panics
4035 when an out-of-memory condition occurs.
4036 The default value is 0. 1 and 2 are for failover of clustering.
4038 Select either according to your policy of failover.
4039 /proc/sys/vm/swappiness
4041 The value in this file controls how aggressively the kernel will
4042 swap memory pages. Higher values increase aggressiveness, lower
4043 values decrease aggressiveness. The default value is 60.
4044 /proc/sys/vm/user_reserve_kbytes (since Linux 3.10)
4046 Specifies an amount of memory (in KiB) to reserve for user pro‐
4047 cesses. This is intended to prevent a user from starting a sin‐
4048 gle memory hogging process, such that they cannot recover (kill
4049 the hog). The value in this file has an effect only when
4050 /proc/sys/vm/overcommit_memory is set to 2 ("overcommit never"
4051 mode). In this case, the system reserves an amount of memory
4052 that is the minimum of [3% of current process size, user_re‐
4053 serve_kbytes].
4054 The default value in this file is the minimum of [3% of free
4056 pages, 128MiB] expressed as KiB.
4057 If the value in this file is set to zero, then a user will be
4059 allowed to allocate all free memory with a single process (minus
4060 the amount reserved by /proc/sys/vm/admin_reserve_kbytes). Any
4061 subsequent attempts to execute a command will result in "fork:
4062 Cannot allocate memory".
4063 Changing the value in this file takes effect whenever an appli‐
4065 cation requests memory.
4066 /proc/sys/vm/unprivileged_userfaultfd (since Linux 5.2)
4068 This (writable) file exposes a flag that controls whether un‐
4069 privileged processes are allowed to employ userfaultfd(2). If
4070 this file has the value 1, then unprivileged processes may use
4071 userfaultfd(2). If this file has the value 0, then only pro‐
4072 cesses that have the CAP_SYS_PTRACE capability may employ user‐
4073 faultfd(2). The default value in this file is 1.
4074 /proc/sysrq-trigger (since Linux 2.4.21)
4076 Writing a character to this file triggers the same SysRq func‐
4077 tion as typing ALT-SysRq-<character> (see the description of
4078 /proc/sys/kernel/sysrq). This file is normally writable only by
4079 root. For further details see the Linux kernel source file Doc‐
4080 umentation/admin-guide/sysrq.rst (or Documentation/sysrq.txt be‐
4081 fore Linux 4.10).
4082 /proc/sysvipc
4084 Subdirectory containing the pseudo-files msg, sem and shm.
4085 These files list the System V Interprocess Communication (IPC)
4086 objects (respectively: message queues, semaphores, and shared
4087 memory) that currently exist on the system, providing similar
4088 information to that available via ipcs(1). These files have
4089 headers and are formatted (one IPC object per line) for easy un‐
4090 derstanding. sysvipc(7) provides further background on the in‐
4091 formation shown by these files.
4092 /proc/thread-self (since Linux 3.17)
4094 This directory refers to the thread accessing the /proc filesys‐
4095 tem, and is identical to the /proc/self/task/[tid] directory
4096 named by the process thread ID ([tid]) of the same thread.
4097 /proc/timer_list (since Linux 2.6.21)
4099 This read-only file exposes a list of all currently pending
4100 (high-resolution) timers, all clock-event sources, and their pa‐
4101 rameters in a human-readable form.
4102 /proc/timer_stats (from Linux 2.6.21 until Linux 4.10)
4104 This is a debugging facility to make timer (ab)use in a Linux
4105 system visible to kernel and user-space developers. It can be
4106 used by kernel and user-space developers to verify that their
4107 code does not make undue use of timers. The goal is to avoid
4108 unnecessary wakeups, thereby optimizing power consumption.
4109 If enabled in the kernel (CONFIG_TIMER_STATS), but not used, it
4111 has almost zero run-time overhead and a relatively small data-
4112 structure overhead. Even if collection is enabled at run time,
4113 overhead is low: all the locking is per-CPU and lookup is
4114 hashed.
4115 The /proc/timer_stats file is used both to control sampling fa‐
4117 cility and to read out the sampled information.
4118 The timer_stats functionality is inactive on bootup. A sampling
4120 period can be started using the following command:
4121 # echo 1 > /proc/timer_stats
4123 The following command stops a sampling period:
4125 # echo 0 > /proc/timer_stats
4127 The statistics can be retrieved by:
4129 $ cat /proc/timer_stats
4131 While sampling is enabled, each readout from /proc/timer_stats
4133 will see newly updated statistics. Once sampling is disabled,
4134 the sampled information is kept until a new sample period is
4135 started. This allows multiple readouts.
4136 Sample output from /proc/timer_stats:
4138 $ cat /proc/timer_stats
4140 Timer Stats Version: v0.3
4141 Sample period: 1.764 s
4142 Collection: active
4143 255, 0 swapper/3 hrtimer_start_range_ns (tick_sched_timer)
4144 71, 0 swapper/1 hrtimer_start_range_ns (tick_sched_timer)
4145 58, 0 swapper/0 hrtimer_start_range_ns (tick_sched_timer)
4146 4, 1694 gnome-shell mod_delayed_work_on (delayed_work_timer_fn)
4147 17, 7 rcu_sched rcu_gp_kthread (process_timeout)
4148 ...
4149 1, 4911 kworker/u16:0 mod_delayed_work_on (delayed_work_timer_fn)
4150 1D, 2522 kworker/0:0 queue_delayed_work_on (delayed_work_timer_fn)
4151 1029 total events, 583.333 events/sec
4152 The output columns are:
4154 * a count of the number of events, optionally (since Linux
4156 2.6.23) followed by the letter 'D' if this is a deferrable
4157 timer;
4158 * the PID of the process that initialized the timer;
4160 * the name of the process that initialized the timer;
4162 * the function where the timer was initialized; and
4164 * (in parentheses) the callback function that is associated
4166 with the timer.
4167 During the Linux 4.11 development cycle, this file was removed
4169 because of security concerns, as it exposes information across
4170 namespaces. Furthermore, it is possible to obtain the same in‐
4171 formation via in-kernel tracing facilities such as ftrace.
4172 /proc/tty
4174 Subdirectory containing the pseudo-files and subdirectories for
4175 tty drivers and line disciplines.
4176 /proc/uptime
4178 This file contains two numbers (values in seconds): the uptime
4179 of the system (including time spent in suspend) and the amount
4180 of time spent in the idle process.
4181 /proc/version
4183 This string identifies the kernel version that is currently run‐
4184 ning. It includes the contents of /proc/sys/kernel/ostype,
4185 /proc/sys/kernel/osrelease, and /proc/sys/kernel/version. For
4186 example:
4187 Linux version 1.0.9 (quinlan@phaze) #1 Sat May 14 01:51:54 EDT 1994
4189 /proc/vmstat (since Linux 2.6.0)
4191 This file displays various virtual memory statistics. Each line
4192 of this file contains a single name-value pair, delimited by
4193 white space. Some lines are present only if the kernel was con‐
4194 figured with suitable options. (In some cases, the options re‐
4195 quired for particular files have changed across kernel versions,
4196 so they are not listed here. Details can be found by consulting
4197 the kernel source code.) The following fields may be present:
4198 nr_free_pages (since Linux 2.6.31)
4200 nr_alloc_batch (since Linux 3.12)
4202 nr_inactive_anon (since Linux 2.6.28)
4204 nr_active_anon (since Linux 2.6.28)
4206 nr_inactive_file (since Linux 2.6.28)
4208 nr_active_file (since Linux 2.6.28)
4210 nr_unevictable (since Linux 2.6.28)
4212 nr_mlock (since Linux 2.6.28)
4214 nr_anon_pages (since Linux 2.6.18)
4216 nr_mapped (since Linux 2.6.0)
4218 nr_file_pages (since Linux 2.6.18)
4220 nr_dirty (since Linux 2.6.0)
4222 nr_writeback (since Linux 2.6.0)
4224 nr_slab_reclaimable (since Linux 2.6.19)
4226 nr_slab_unreclaimable (since Linux 2.6.19)
4228 nr_page_table_pages (since Linux 2.6.0)
4230 nr_kernel_stack (since Linux 2.6.32)
4232 Amount of memory allocated to kernel stacks.
4233 nr_unstable (since Linux 2.6.0)
4235 nr_bounce (since Linux 2.6.12)
4237 nr_vmscan_write (since Linux 2.6.19)
4239 nr_vmscan_immediate_reclaim (since Linux 3.2)
4241 nr_writeback_temp (since Linux 2.6.26)
4243 nr_isolated_anon (since Linux 2.6.32)
4245 nr_isolated_file (since Linux 2.6.32)
4247 nr_shmem (since Linux 2.6.32)
4249 Pages used by shmem and tmpfs(5).
4250 nr_dirtied (since Linux 2.6.37)
4252 nr_written (since Linux 2.6.37)
4254 nr_pages_scanned (since Linux 3.17)
4256 numa_hit (since Linux 2.6.18)
4258 numa_miss (since Linux 2.6.18)
4260 numa_foreign (since Linux 2.6.18)
4262 numa_interleave (since Linux 2.6.18)
4264 numa_local (since Linux 2.6.18)
4266 numa_other (since Linux 2.6.18)
4268 workingset_refault (since Linux 3.15)
4270 workingset_activate (since Linux 3.15)
4272 workingset_nodereclaim (since Linux 3.15)
4274 nr_anon_transparent_hugepages (since Linux 2.6.38)
4276 nr_free_cma (since Linux 3.7)
4278 Number of free CMA (Contiguous Memory Allocator) pages.
4279 nr_dirty_threshold (since Linux 2.6.37)
4281 nr_dirty_background_threshold (since Linux 2.6.37)
4283 pgpgin (since Linux 2.6.0)
4285 pgpgout (since Linux 2.6.0)
4287 pswpin (since Linux 2.6.0)
4289 pswpout (since Linux 2.6.0)
4291 pgalloc_dma (since Linux 2.6.5)
4293 pgalloc_dma32 (since Linux 2.6.16)
4295 pgalloc_normal (since Linux 2.6.5)
4297 pgalloc_high (since Linux 2.6.5)
4299 pgalloc_movable (since Linux 2.6.23)
4301 pgfree (since Linux 2.6.0)
4303 pgactivate (since Linux 2.6.0)
4305 pgdeactivate (since Linux 2.6.0)
4307 pgfault (since Linux 2.6.0)
4309 pgmajfault (since Linux 2.6.0)
4311 pgrefill_dma (since Linux 2.6.5)
4313 pgrefill_dma32 (since Linux 2.6.16)
4315 pgrefill_normal (since Linux 2.6.5)
4317 pgrefill_high (since Linux 2.6.5)
4319 pgrefill_movable (since Linux 2.6.23)
4321 pgsteal_kswapd_dma (since Linux 3.4)
4323 pgsteal_kswapd_dma32 (since Linux 3.4)
4325 pgsteal_kswapd_normal (since Linux 3.4)
4327 pgsteal_kswapd_high (since Linux 3.4)
4329 pgsteal_kswapd_movable (since Linux 3.4)
4331 pgsteal_direct_dma
4333 pgsteal_direct_dma32 (since Linux 3.4)
4335 pgsteal_direct_normal (since Linux 3.4)
4337 pgsteal_direct_high (since Linux 3.4)
4339 pgsteal_direct_movable (since Linux 2.6.23)
4341 pgscan_kswapd_dma
4343 pgscan_kswapd_dma32 (since Linux 2.6.16)
4345 pgscan_kswapd_normal (since Linux 2.6.5)
4347 pgscan_kswapd_high
4349 pgscan_kswapd_movable (since Linux 2.6.23)
4351 pgscan_direct_dma
4353 pgscan_direct_dma32 (since Linux 2.6.16)
4355 pgscan_direct_normal
4357 pgscan_direct_high
4359 pgscan_direct_movable (since Linux 2.6.23)
4361 pgscan_direct_throttle (since Linux 3.6)
4363 zone_reclaim_failed (since linux 2.6.31)
4365 pginodesteal (since linux 2.6.0)
4367 slabs_scanned (since linux 2.6.5)
4369 kswapd_inodesteal (since linux 2.6.0)
4371 kswapd_low_wmark_hit_quickly (since 2.6.33)
4373 kswapd_high_wmark_hit_quickly (since 2.6.33)
4375 pageoutrun (since Linux 2.6.0)
4377 allocstall (since Linux 2.6.0)
4379 pgrotated (since Linux 2.6.0)
4381 drop_pagecache (since Linux 3.15)
4383 drop_slab (since Linux 3.15)
4385 numa_pte_updates (since Linux 3.8)
4387 numa_huge_pte_updates (since Linux 3.13)
4389 numa_hint_faults (since Linux 3.8)
4391 numa_hint_faults_local (since Linux 3.8)
4393 numa_pages_migrated (since Linux 3.8)
4395 pgmigrate_success (since Linux 3.8)
4397 pgmigrate_fail (since Linux 3.8)
4399 compact_migrate_scanned (since Linux 3.8)
4401 compact_free_scanned (since Linux 3.8)
4403 compact_isolated (since Linux 3.8)
4405 compact_stall (since Linux 2.6.35)
4407 See the kernel source file Documentation/ad‐
4408 min-guide/mm/transhuge.rst.
4409 compact_fail (since Linux 2.6.35)
4411 See the kernel source file Documentation/ad‐
4412 min-guide/mm/transhuge.rst.
4413 compact_success (since Linux 2.6.35)
4415 See the kernel source file Documentation/ad‐
4416 min-guide/mm/transhuge.rst.
4417 htlb_buddy_alloc_success (since Linux 2.6.26)
4419 htlb_buddy_alloc_fail (since Linux 2.6.26)
4421 unevictable_pgs_culled (since Linux 2.6.28)
4423 unevictable_pgs_scanned (since Linux 2.6.28)
4425 unevictable_pgs_rescued (since Linux 2.6.28)
4427 unevictable_pgs_mlocked (since Linux 2.6.28)
4429 unevictable_pgs_munlocked (since Linux 2.6.28)
4431 unevictable_pgs_cleared (since Linux 2.6.28)
4433 unevictable_pgs_stranded (since Linux 2.6.28)
4435 thp_fault_alloc (since Linux 2.6.39)
4437 See the kernel source file Documentation/ad‐
4438 min-guide/mm/transhuge.rst.
4439 thp_fault_fallback (since Linux 2.6.39)
4441 See the kernel source file Documentation/ad‐
4442 min-guide/mm/transhuge.rst.
4443 thp_collapse_alloc (since Linux 2.6.39)
4445 See the kernel source file Documentation/ad‐
4446 min-guide/mm/transhuge.rst.
4447 thp_collapse_alloc_failed (since Linux 2.6.39)
4449 See the kernel source file Documentation/ad‐
4450 min-guide/mm/transhuge.rst.
4451 thp_split (since Linux 2.6.39)
4453 See the kernel source file Documentation/ad‐
4454 min-guide/mm/transhuge.rst.
4455 thp_zero_page_alloc (since Linux 3.8)
4457 See the kernel source file Documentation/ad‐
4458 min-guide/mm/transhuge.rst.
4459 thp_zero_page_alloc_failed (since Linux 3.8)
4461 See the kernel source file Documentation/ad‐
4462 min-guide/mm/transhuge.rst.
4463 balloon_inflate (since Linux 3.18)
4465 balloon_deflate (since Linux 3.18)
4467 balloon_migrate (since Linux 3.18)
4469 nr_tlb_remote_flush (since Linux 3.12)
4471 nr_tlb_remote_flush_received (since Linux 3.12)
4473 nr_tlb_local_flush_all (since Linux 3.12)
4475 nr_tlb_local_flush_one (since Linux 3.12)
4477 vmacache_find_calls (since Linux 3.16)
4479 vmacache_find_hits (since Linux 3.16)
4481 vmacache_full_flushes (since Linux 3.19)
4483 /proc/zoneinfo (since Linux 2.6.13)
4485 This file displays information about memory zones. This is use‐
4486 ful for analyzing virtual memory behavior.
4487
4489 Many files contain strings (e.g., the environment and command line)
4490 that are in the internal format, with subfields terminated by null
4491 bytes ('\0'). When inspecting such files, you may find that the re‐
4492 sults are more readable if you use a command of the following form to
4493 display them:
4494 $ cat file | tr '\000' '\n'
4496 This manual page is incomplete, possibly inaccurate, and is the kind of
4498 thing that needs to be updated very often.
4499
4501 cat(1), dmesg(1), find(1), free(1), htop(1), init(1), ps(1), pstree(1),
4502 tr(1), uptime(1), chroot(2), mmap(2), readlink(2), syslog(2),
4503 slabinfo(5), sysfs(5), hier(7), namespaces(7), time(7), arp(8), hd‐
4504 parm(8), ifconfig(8), lsmod(8), lspci(8), mount(8), netstat(8),
4505 procinfo(8), route(8), sysctl(8)
4506 The Linux kernel source files: Documentation/filesystems/proc.txt, Doc‐
4508 umentation/sysctl/fs.txt, Documentation/sysctl/kernel.txt, Documenta‐
4509 tion/sysctl/net.txt, and Documentation/sysctl/vm.txt.
4510
4512 This page is part of release 5.13 of the Linux man-pages project. A
4513 description of the project, information about reporting bugs, and the
4514 latest version of this page, can be found at
4515 https://www.kernel.org/doc/man-pages/.
4516Linux 2021-08-27 PROC(5)