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 point 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 mount points in the
865 process's mount namespace (see mount_namespaces(7)). It sup‐
866 plies various information (e.g., propagation state, root of
867 mount for bind mounts, identifier for each mount and its parent)
868 that is missing from the (older) /proc/[pid]/mounts file, and
869 fixes various other problems with that file (e.g., nonextensi‐
870 bility, failure to distinguish per-mount versus per-superblock
871 options).
872 The file contains lines of the form:
874 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
876 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
877 The numbers in parentheses are labels for the descriptions be‐
879 low:
880 (1) mount ID: a unique ID for the mount (may be reused after
882 umount(2)).
883 (2) parent ID: the ID of the parent mount (or of self for the
885 root of this mount namespace's mount tree).
886 If a new mount is stacked on top of a previous existing
888 mount (so that it hides the existing mount) at pathname P,
889 then the parent of the new mount is the previous mount at
890 that location. Thus, when looking at all the mounts
891 stacked at a particular location, the top-most mount is the
892 one that is not the parent of any other mount at the same
893 location. (Note, however, that this top-most mount will be
894 accessible only if the longest path subprefix of P that is
895 a mount point is not itself hidden by a stacked mount.)
896 If the parent mount point lies outside the process's root
898 directory (see chroot(2)), the ID shown here won't have a
899 corresponding record in mountinfo whose mount ID (field 1)
900 matches this parent mount ID (because mount points that lie
901 outside the process's root directory are not shown in
902 mountinfo). As a special case of this point, the process's
903 root mount point may have a parent mount (for the initramfs
904 filesystem) that lies outside the process's root directory,
905 and an entry for that mount point will not appear in
906 mountinfo.
907 (3) major:minor: the value of st_dev for files on this filesys‐
909 tem (see stat(2)).
910 (4) root: the pathname of the directory in the filesystem which
912 forms the root of this mount.
913 (5) mount point: the pathname of the mount point relative to
915 the process's root directory.
916 (6) mount options: per-mount options (see mount(2)).
918 (7) optional fields: zero or more fields of the form
920 "tag[:value]"; see below.
921 (8) separator: the end of the optional fields is marked by a
923 single hyphen.
924 (9) filesystem type: the filesystem type in the form
926 "type[.subtype]".
927 (10) mount source: filesystem-specific information or "none".
929 (11) super options: per-superblock options (see mount(2)).
931 Currently, the possible optional fields are shared, master,
933 propagate_from, and unbindable. See mount_namespaces(7) for a
934 description of these fields. Parsers should ignore all unrecog‐
935 nized optional fields.
936 For more information on mount propagation see: Documenta‐
938 tion/filesystems/sharedsubtree.txt in the Linux kernel source
939 tree.
940 /proc/[pid]/mounts (since Linux 2.4.19)
942 This file lists all the filesystems currently mounted in the
943 process's mount namespace (see mount_namespaces(7)). The format
944 of this file is documented in fstab(5).
945 Since kernel version 2.6.15, this file is pollable: after open‐
947 ing the file for reading, a change in this file (i.e., a
948 filesystem mount or unmount) causes select(2) to mark the file
949 descriptor as having an exceptional condition, and poll(2) and
950 epoll_wait(2) mark the file as having a priority event (POLL‐
951 PRI). (Before Linux 2.6.30, a change in this file was indicated
952 by the file descriptor being marked as readable for select(2),
953 and being marked as having an error condition for poll(2) and
954 epoll_wait(2).)
955 /proc/[pid]/mountstats (since Linux 2.6.17)
957 This file exports information (statistics, configuration infor‐
958 mation) about the mount points in the process's mount namespace
959 (see mount_namespaces(7)). Lines in this file have the form:
960 device /dev/sda7 mounted on /home with fstype ext3 [stats]
962 ( 1 ) ( 2 ) (3 ) ( 4 )
963 The fields in each line are:
965 (1) The name of the mounted device (or "nodevice" if there is
967 no corresponding device).
968 (2) The mount point within the filesystem tree.
970 (3) The filesystem type.
972 (4) Optional statistics and configuration information. Cur‐
974 rently (as at Linux 2.6.26), only NFS filesystems export
975 information via this field.
976 This file is readable only by the owner of the process.
978 /proc/[pid]/net (since Linux 2.6.25)
980 See the description of /proc/net.
981 /proc/[pid]/ns/ (since Linux 3.0)
983 This is a subdirectory containing one entry for each namespace
984 that supports being manipulated by setns(2). For more informa‐
985 tion, see namespaces(7).
986 /proc/[pid]/numa_maps (since Linux 2.6.14)
988 See numa(7).
989 /proc/[pid]/oom_adj (since Linux 2.6.11)
991 This file can be used to adjust the score used to select which
992 process should be killed in an out-of-memory (OOM) situation.
993 The kernel uses this value for a bit-shift operation of the
994 process's oom_score value: valid values are in the range -16 to
995 +15, plus the special value -17, which disables OOM-killing al‐
996 together for this process. A positive score increases the like‐
997 lihood of this process being killed by the OOM-killer; a nega‐
998 tive score decreases the likelihood.
999 The default value for this file is 0; a new process inherits its
1001 parent's oom_adj setting. A process must be privileged
1002 (CAP_SYS_RESOURCE) to update this file.
1003 Since Linux 2.6.36, use of this file is deprecated in favor of
1005 /proc/[pid]/oom_score_adj.
1006 /proc/[pid]/oom_score (since Linux 2.6.11)
1008 This file displays the current score that the kernel gives to
1009 this process for the purpose of selecting a process for the OOM-
1010 killer. A higher score means that the process is more likely to
1011 be selected by the OOM-killer. The basis for this score is the
1012 amount of memory used by the process, with increases (+) or de‐
1013 creases (-) for factors including:
1014 * whether the process is privileged (-).
1016 Before kernel 2.6.36 the following factors were also used in the
1018 calculation of oom_score:
1019 * whether the process creates a lot of children using fork(2)
1021 (+);
1022 * whether the process has been running a long time, or has used
1024 a lot of CPU time (-);
1025 * whether the process has a low nice value (i.e., > 0) (+); and
1027 * whether the process is making direct hardware access (-).
1029 The oom_score also reflects the adjustment specified by the
1031 oom_score_adj or oom_adj setting for the process.
1032 /proc/[pid]/oom_score_adj (since Linux 2.6.36)
1034 This file can be used to adjust the badness heuristic used to
1035 select which process gets killed in out-of-memory conditions.
1036 The badness heuristic assigns a value to each candidate task
1038 ranging from 0 (never kill) to 1000 (always kill) to determine
1039 which process is targeted. The units are roughly a proportion
1040 along that range of allowed memory the process may allocate
1041 from, based on an estimation of its current memory and swap use.
1042 For example, if a task is using all allowed memory, its badness
1043 score will be 1000. If it is using half of its allowed memory,
1044 its score will be 500.
1045 There is an additional factor included in the badness score:
1047 root processes are given 3% extra memory over other tasks.
1048 The amount of "allowed" memory depends on the context in which
1050 the OOM-killer was called. If it is due to the memory assigned
1051 to the allocating task's cpuset being exhausted, the allowed
1052 memory represents the set of mems assigned to that cpuset (see
1053 cpuset(7)). If it is due to a mempolicy's node(s) being ex‐
1054 hausted, the allowed memory represents the set of mempolicy
1055 nodes. If it is due to a memory limit (or swap limit) being
1056 reached, the allowed memory is that configured limit. Finally,
1057 if it is due to the entire system being out of memory, the al‐
1058 lowed memory represents all allocatable resources.
1059 The value of oom_score_adj is added to the badness score before
1061 it is used to determine which task to kill. Acceptable values
1062 range from -1000 (OOM_SCORE_ADJ_MIN) to +1000
1063 (OOM_SCORE_ADJ_MAX). This allows user space to control the
1064 preference for OOM-killing, ranging from always preferring a
1065 certain task or completely disabling it from OOM killing. The
1066 lowest possible value, -1000, is equivalent to disabling OOM-
1067 killing entirely for that task, since it will always report a
1068 badness score of 0.
1069 Consequently, it is very simple for user space to define the
1071 amount of memory to consider for each task. Setting an
1072 oom_score_adj value of +500, for example, is roughly equivalent
1073 to allowing the remainder of tasks sharing the same system,
1074 cpuset, mempolicy, or memory controller resources to use at
1075 least 50% more memory. A value of -500, on the other hand,
1076 would be roughly equivalent to discounting 50% of the task's al‐
1077 lowed memory from being considered as scoring against the task.
1078 For backward compatibility with previous kernels,
1080 /proc/[pid]/oom_adj can still be used to tune the badness score.
1081 Its value is scaled linearly with oom_score_adj.
1082 Writing to /proc/[pid]/oom_score_adj or /proc/[pid]/oom_adj will
1084 change the other with its scaled value.
1085 The choom(1) program provides a command-line interface for ad‐
1087 justing the oom_score_adj value of a running process or a newly
1088 executed command.
1089 /proc/[pid]/pagemap (since Linux 2.6.25)
1091 This file shows the mapping of each of the process's virtual
1092 pages into physical page frames or swap area. It contains one
1093 64-bit value for each virtual page, with the bits set as fol‐
1094 lows:
1095 63 If set, the page is present in RAM.
1097 62 If set, the page is in swap space
1099 61 (since Linux 3.5)
1101 The page is a file-mapped page or a shared anonymous
1102 page.
1103 60–57 (since Linux 3.11)
1105 Zero
1106 56 (since Linux 4.2)
1108 The page is exclusively mapped.
1109 55 (since Linux 3.11)
1111 PTE is soft-dirty (see the kernel source file Documenta‐
1112 tion/admin-guide/mm/soft-dirty.rst).
1113 54–0 If the page is present in RAM (bit 63), then these bits
1115 provide the page frame number, which can be used to index
1116 /proc/kpageflags and /proc/kpagecount. If the page is
1117 present in swap (bit 62), then bits 4–0 give the swap
1118 type, and bits 54–5 encode the swap offset.
1119 Before Linux 3.11, bits 60–55 were used to encode the base-2 log
1121 of the page size.
1122 To employ /proc/[pid]/pagemap efficiently, use /proc/[pid]/maps
1124 to determine which areas of memory are actually mapped and seek
1125 to skip over unmapped regions.
1126 The /proc/[pid]/pagemap file is present only if the CON‐
1128 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
1129 Permission to access this file is governed by a ptrace access
1131 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1132 /proc/[pid]/personality (since Linux 2.6.28)
1134 This read-only file exposes the process's execution domain, as
1135 set by personality(2). The value is displayed in hexadecimal
1136 notation.
1137 Permission to access this file is governed by a ptrace access
1139 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1140 /proc/[pid]/root
1142 UNIX and Linux support the idea of a per-process root of the
1143 filesystem, set by the chroot(2) system call. This file is a
1144 symbolic link that points to the process's root directory, and
1145 behaves in the same way as exe, and fd/*.
1146 Note however that this file is not merely a symbolic link. It
1148 provides the same view of the filesystem (including namespaces
1149 and the set of per-process mounts) as the process itself. An
1150 example illustrates this point. In one terminal, we start a
1151 shell in new user and mount namespaces, and in that shell we
1152 create some new mount points:
1153 $ PS1='sh1# ' unshare -Urnm
1155 sh1# mount -t tmpfs tmpfs /etc # Mount empty tmpfs at /etc
1156 sh1# mount --bind /usr /dev # Mount /usr at /dev
1157 sh1# echo $$
1158 27123
1159 In a second terminal window, in the initial mount namespace, we
1161 look at the contents of the corresponding mounts in the initial
1162 and new namespaces:
1163 $ PS1='sh2# ' sudo sh
1165 sh2# ls /etc | wc -l # In initial NS
1166 309
1167 sh2# ls /proc/27123/root/etc | wc -l # /etc in other NS
1168 0 # The empty tmpfs dir
1169 sh2# ls /dev | wc -l # In initial NS
1170 205
1171 sh2# ls /proc/27123/root/dev | wc -l # /dev in other NS
1172 11 # Actually bind
1173 # mounted to /usr
1174 sh2# ls /usr | wc -l # /usr in initial NS
1175 11
1176 In a multithreaded process, the contents of the /proc/[pid]/root
1178 symbolic link are not available if the main thread has already
1179 terminated (typically by calling pthread_exit(3)).
1180 Permission to dereference or read (readlink(2)) this symbolic
1182 link is governed by a ptrace access mode PTRACE_MODE_READ_FS‐
1183 CREDS check; see ptrace(2).
1184 /proc/[pid]/seccomp (Linux 2.6.12 to 2.6.22)
1186 This file can be used to read and change the process's secure
1187 computing (seccomp) mode setting. It contains the value 0 if
1188 the process is not in seccomp mode, and 1 if the process is in
1189 strict seccomp mode (see seccomp(2)). Writing 1 to this file
1190 places the process irreversibly in strict seccomp mode. (Fur‐
1191 ther attempts to write to the file fail with the EPERM error.)
1192 In Linux 2.6.23, this file went away, to be replaced by the
1194 prctl(2) PR_GET_SECCOMP and PR_SET_SECCOMP operations (and later
1195 by seccomp(2) and the Seccomp field in /proc/[pid]/status).
1196 /proc/[pid]/setgroups (since Linux 3.19)
1198 See user_namespaces(7).
1199 /proc/[pid]/smaps (since Linux 2.6.14)
1201 This file shows memory consumption for each of the process's
1202 mappings. (The pmap(1) command displays similar information, in
1203 a form that may be easier for parsing.) For each mapping there
1204 is a series of lines such as the following:
1205 00400000-0048a000 r-xp 00000000 fd:03 960637 /bin/bash
1207 Size: 552 kB
1208 Rss: 460 kB
1209 Pss: 100 kB
1210 Shared_Clean: 452 kB
1211 Shared_Dirty: 0 kB
1212 Private_Clean: 8 kB
1213 Private_Dirty: 0 kB
1214 Referenced: 460 kB
1215 Anonymous: 0 kB
1216 AnonHugePages: 0 kB
1217 ShmemHugePages: 0 kB
1218 ShmemPmdMapped: 0 kB
1219 Swap: 0 kB
1220 KernelPageSize: 4 kB
1221 MMUPageSize: 4 kB
1222 KernelPageSize: 4 kB
1223 MMUPageSize: 4 kB
1224 Locked: 0 kB
1225 ProtectionKey: 0
1226 VmFlags: rd ex mr mw me dw
1227 The first of these lines shows the same information as is dis‐
1229 played for the mapping in /proc/[pid]/maps. The following lines
1230 show the size of the mapping, the amount of the mapping that is
1231 currently resident in RAM ("Rss"), the process's proportional
1232 share of this mapping ("Pss"), the number of clean and dirty
1233 shared pages in the mapping, and the number of clean and dirty
1234 private pages in the mapping. "Referenced" indicates the amount
1235 of memory currently marked as referenced or accessed. "Anony‐
1236 mous" shows the amount of memory that does not belong to any
1237 file. "Swap" shows how much would-be-anonymous memory is also
1238 used, but out on swap.
1239 The "KernelPageSize" line (available since Linux 2.6.29) is the
1241 page size used by the kernel to back the virtual memory area.
1242 This matches the size used by the MMU in the majority of cases.
1243 However, one counter-example occurs on PPC64 kernels whereby a
1244 kernel using 64 kB as a base page size may still use 4 kB pages
1245 for the MMU on older processors. To distinguish the two at‐
1246 tributes, the "MMUPageSize" line (also available since Linux
1247 2.6.29) reports the page size used by the MMU.
1248 The "Locked" indicates whether the mapping is locked in memory
1250 or not.
1251 The "ProtectionKey" line (available since Linux 4.9, on x86
1253 only) contains the memory protection key (see pkeys(7)) associ‐
1254 ated with the virtual memory area. This entry is present only
1255 if the kernel was built with the CONFIG_X86_INTEL_MEMORY_PROTEC‐
1256 TION_KEYS configuration option (since Linux 4.6).
1257 The "VmFlags" line (available since Linux 3.8) represents the
1259 kernel flags associated with the virtual memory area, encoded
1260 using the following two-letter codes:
1261 rd - readable
1263 wr - writable
1264 ex - executable
1265 sh - shared
1266 mr - may read
1267 mw - may write
1268 me - may execute
1269 ms - may share
1270 gd - stack segment grows down
1271 pf - pure PFN range
1272 dw - disabled write to the mapped file
1273 lo - pages are locked in memory
1274 io - memory mapped I/O area
1275 sr - sequential read advise provided
1276 rr - random read advise provided
1277 dc - do not copy area on fork
1278 de - do not expand area on remapping
1279 ac - area is accountable
1280 nr - swap space is not reserved for the area
1281 ht - area uses huge tlb pages
1282 sf - perform synchronous page faults (since Linux 4.15)
1283 nl - non-linear mapping (removed in Linux 4.0)
1284 ar - architecture specific flag
1285 wf - wipe on fork (since Linux 4.14)
1286 dd - do not include area into core dump
1287 sd - soft-dirty flag (since Linux 3.13)
1288 mm - mixed map area
1289 hg - huge page advise flag
1290 nh - no-huge page advise flag
1291 mg - mergeable advise flag
1292 um - userfaultfd missing pages tracking (since Linux 4.3)
1293 uw - userfaultfd wprotect pages tracking (since Linux 4.3)
1294 The /proc/[pid]/smaps file is present only if the CON‐
1296 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
1297 /proc/[pid]/stack (since Linux 2.6.29)
1299 This file provides a symbolic trace of the function calls in
1300 this process's kernel stack. This file is provided only if the
1301 kernel was built with the CONFIG_STACKTRACE configuration op‐
1302 tion.
1303 Permission to access this file is governed by a ptrace access
1305 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1306 /proc/[pid]/stat
1308 Status information about the process. This is used by ps(1).
1309 It is defined in the kernel source file fs/proc/array.c.
1310 The fields, in order, with their proper scanf(3) format speci‐
1312 fiers, are listed below. Whether or not certain of these fields
1313 display valid information is governed by a ptrace access mode
1314 PTRACE_MODE_READ_FSCREDS | PTRACE_MODE_NOAUDIT check (refer to
1315 ptrace(2)). If the check denies access, then the field value is
1316 displayed as 0. The affected fields are indicated with the
1317 marking [PT].
1318 (1) pid %d
1320 The process ID.
1321 (2) comm %s
1323 The filename of the executable, in parentheses. Strings
1324 longer than TASK_COMM_LEN (16) characters (including the
1325 terminating null byte) are silently truncated. This is
1326 visible whether or not the executable is swapped out.
1327 (3) state %c
1329 One of the following characters, indicating process
1330 state:
1331 R Running
1333 S Sleeping in an interruptible wait
1335 D Waiting in uninterruptible disk sleep
1337 Z Zombie
1339 T Stopped (on a signal) or (before Linux 2.6.33) trace
1341 stopped
1342 t Tracing stop (Linux 2.6.33 onward)
1344 W Paging (only before Linux 2.6.0)
1346 X Dead (from Linux 2.6.0 onward)
1348 x Dead (Linux 2.6.33 to 3.13 only)
1350 K Wakekill (Linux 2.6.33 to 3.13 only)
1352 W Waking (Linux 2.6.33 to 3.13 only)
1354 P Parked (Linux 3.9 to 3.13 only)
1356 (4) ppid %d
1358 The PID of the parent of this process.
1359 (5) pgrp %d
1361 The process group ID of the process.
1362 (6) session %d
1364 The session ID of the process.
1365 (7) tty_nr %d
1367 The controlling terminal of the process. (The minor de‐
1368 vice number is contained in the combination of bits 31 to
1369 20 and 7 to 0; the major device number is in bits 15 to
1370 8.)
1371 (8) tpgid %d
1373 The ID of the foreground process group of the controlling
1374 terminal of the process.
1375 (9) flags %u
1377 The kernel flags word of the process. For bit meanings,
1378 see the PF_* defines in the Linux kernel source file in‐
1379 clude/linux/sched.h. Details depend on the kernel ver‐
1380 sion.
1381 The format for this field was %lu before Linux 2.6.
1383 (10) minflt %lu
1385 The number of minor faults the process has made which
1386 have not required loading a memory page from disk.
1387 (11) cminflt %lu
1389 The number of minor faults that the process's waited-for
1390 children have made.
1391 (12) majflt %lu
1393 The number of major faults the process has made which
1394 have required loading a memory page from disk.
1395 (13) cmajflt %lu
1397 The number of major faults that the process's waited-for
1398 children have made.
1399 (14) utime %lu
1401 Amount of time that this process has been scheduled in
1402 user mode, measured in clock ticks (divide by
1403 sysconf(_SC_CLK_TCK)). This includes guest time,
1404 guest_time (time spent running a virtual CPU, see below),
1405 so that applications that are not aware of the guest time
1406 field do not lose that time from their calculations.
1407 (15) stime %lu
1409 Amount of time that this process has been scheduled in
1410 kernel mode, measured in clock ticks (divide by
1411 sysconf(_SC_CLK_TCK)).
1412 (16) cutime %ld
1414 Amount of time that this process's waited-for children
1415 have been scheduled in user mode, measured in clock ticks
1416 (divide by sysconf(_SC_CLK_TCK)). (See also times(2).)
1417 This includes guest time, cguest_time (time spent running
1418 a virtual CPU, see below).
1419 (17) cstime %ld
1421 Amount of time that this process's waited-for children
1422 have been scheduled in kernel mode, measured in clock
1423 ticks (divide by sysconf(_SC_CLK_TCK)).
1424 (18) priority %ld
1426 (Explanation for Linux 2.6) For processes running a real-
1427 time scheduling policy (policy below; see sched_setsched‐
1428 uler(2)), this is the negated scheduling priority, minus
1429 one; that is, a number in the range -2 to -100, corre‐
1430 sponding to real-time priorities 1 to 99. For processes
1431 running under a non-real-time scheduling policy, this is
1432 the raw nice value (setpriority(2)) as represented in the
1433 kernel. The kernel stores nice values as numbers in the
1434 range 0 (high) to 39 (low), corresponding to the user-
1435 visible nice range of -20 to 19.
1436 Before Linux 2.6, this was a scaled value based on the
1438 scheduler weighting given to this process.
1439 (19) nice %ld
1441 The nice value (see setpriority(2)), a value in the range
1442 19 (low priority) to -20 (high priority).
1443 (20) num_threads %ld
1445 Number of threads in this process (since Linux 2.6). Be‐
1446 fore kernel 2.6, this field was hard coded to 0 as a
1447 placeholder for an earlier removed field.
1448 (21) itrealvalue %ld
1450 The time in jiffies before the next SIGALRM is sent to
1451 the process due to an interval timer. Since kernel
1452 2.6.17, this field is no longer maintained, and is hard
1453 coded as 0.
1454 (22) starttime %llu
1456 The time the process started after system boot. In ker‐
1457 nels before Linux 2.6, this value was expressed in
1458 jiffies. Since Linux 2.6, the value is expressed in
1459 clock ticks (divide by sysconf(_SC_CLK_TCK)).
1460 The format for this field was %lu before Linux 2.6.
1462 (23) vsize %lu
1464 Virtual memory size in bytes.
1465 (24) rss %ld
1467 Resident Set Size: number of pages the process has in
1468 real memory. This is just the pages which count toward
1469 text, data, or stack space. This does not include pages
1470 which have not been demand-loaded in, or which are
1471 swapped out. This value is inaccurate; see
1472 /proc/[pid]/statm below.
1473 (25) rsslim %lu
1475 Current soft limit in bytes on the rss of the process;
1476 see the description of RLIMIT_RSS in getrlimit(2).
1477 (26) startcode %lu [PT]
1479 The address above which program text can run.
1480 (27) endcode %lu [PT]
1482 The address below which program text can run.
1483 (28) startstack %lu [PT]
1485 The address of the start (i.e., bottom) of the stack.
1486 (29) kstkesp %lu [PT]
1488 The current value of ESP (stack pointer), as found in the
1489 kernel stack page for the process.
1490 (30) kstkeip %lu [PT]
1492 The current EIP (instruction pointer).
1493 (31) signal %lu
1495 The bitmap of pending signals, displayed as a decimal
1496 number. Obsolete, because it does not provide informa‐
1497 tion on real-time signals; use /proc/[pid]/status in‐
1498 stead.
1499 (32) blocked %lu
1501 The bitmap of blocked signals, displayed as a decimal
1502 number. Obsolete, because it does not provide informa‐
1503 tion on real-time signals; use /proc/[pid]/status in‐
1504 stead.
1505 (33) sigignore %lu
1507 The bitmap of ignored signals, displayed as a decimal
1508 number. Obsolete, because it does not provide informa‐
1509 tion on real-time signals; use /proc/[pid]/status in‐
1510 stead.
1511 (34) sigcatch %lu
1513 The bitmap of caught signals, displayed as a decimal num‐
1514 ber. Obsolete, because it does not provide information
1515 on real-time signals; use /proc/[pid]/status instead.
1516 (35) wchan %lu [PT]
1518 This is the "channel" in which the process is waiting.
1519 It is the address of a location in the kernel where the
1520 process is sleeping. The corresponding symbolic name can
1521 be found in /proc/[pid]/wchan.
1522 (36) nswap %lu
1524 Number of pages swapped (not maintained).
1525 (37) cnswap %lu
1527 Cumulative nswap for child processes (not maintained).
1528 (38) exit_signal %d (since Linux 2.1.22)
1530 Signal to be sent to parent when we die.
1531 (39) processor %d (since Linux 2.2.8)
1533 CPU number last executed on.
1534 (40) rt_priority %u (since Linux 2.5.19)
1536 Real-time scheduling priority, a number in the range 1 to
1537 99 for processes scheduled under a real-time policy, or
1538 0, for non-real-time processes (see sched_setsched‐
1539 uler(2)).
1540 (41) policy %u (since Linux 2.5.19)
1542 Scheduling policy (see sched_setscheduler(2)). Decode
1543 using the SCHED_* constants in linux/sched.h.
1544 The format for this field was %lu before Linux 2.6.22.
1546 (42) delayacct_blkio_ticks %llu (since Linux 2.6.18)
1548 Aggregated block I/O delays, measured in clock ticks
1549 (centiseconds).
1550 (43) guest_time %lu (since Linux 2.6.24)
1552 Guest time of the process (time spent running a virtual
1553 CPU for a guest operating system), measured in clock
1554 ticks (divide by sysconf(_SC_CLK_TCK)).
1555 (44) cguest_time %ld (since Linux 2.6.24)
1557 Guest time of the process's children, measured in clock
1558 ticks (divide by sysconf(_SC_CLK_TCK)).
1559 (45) start_data %lu (since Linux 3.3) [PT]
1561 Address above which program initialized and uninitialized
1562 (BSS) data are placed.
1563 (46) end_data %lu (since Linux 3.3) [PT]
1565 Address below which program initialized and uninitialized
1566 (BSS) data are placed.
1567 (47) start_brk %lu (since Linux 3.3) [PT]
1569 Address above which program heap can be expanded with
1570 brk(2).
1571 (48) arg_start %lu (since Linux 3.5) [PT]
1573 Address above which program command-line arguments (argv)
1574 are placed.
1575 (49) arg_end %lu (since Linux 3.5) [PT]
1577 Address below program command-line arguments (argv) are
1578 placed.
1579 (50) env_start %lu (since Linux 3.5) [PT]
1581 Address above which program environment is placed.
1582 (51) env_end %lu (since Linux 3.5) [PT]
1584 Address below which program environment is placed.
1585 (52) exit_code %d (since Linux 3.5) [PT]
1587 The thread's exit status in the form reported by wait‐
1588 pid(2).
1589 /proc/[pid]/statm
1591 Provides information about memory usage, measured in pages. The
1592 columns are:
1593 size (1) total program size
1595 (same as VmSize in /proc/[pid]/status)
1596 resident (2) resident set size
1597 (inaccurate; same as VmRSS in /proc/[pid]/status)
1598 shared (3) number of resident shared pages
1599 (i.e., backed by a file)
1600 (inaccurate; same as RssFile+RssShmem in
1601 /proc/[pid]/status)
1602 text (4) text (code)
1603 lib (5) library (unused since Linux 2.6; always 0)
1604 data (6) data + stack
1605 dt (7) dirty pages (unused since Linux 2.6; always 0)
1606 Some of these values are inaccurate because of a kernel-internal
1608 scalability optimization. If accurate values are required, use
1609 /proc/[pid]/smaps or /proc/[pid]/smaps_rollup instead, which are
1610 much slower but provide accurate, detailed information.
1611 /proc/[pid]/status
1613 Provides much of the information in /proc/[pid]/stat and
1614 /proc/[pid]/statm in a format that's easier for humans to parse.
1615 Here's an example:
1616 $ cat /proc/$$/status
1618 Name: bash
1619 Umask: 0022
1620 State: S (sleeping)
1621 Tgid: 17248
1622 Ngid: 0
1623 Pid: 17248
1624 PPid: 17200
1625 TracerPid: 0
1626 Uid: 1000 1000 1000 1000
1627 Gid: 100 100 100 100
1628 FDSize: 256
1629 Groups: 16 33 100
1630 NStgid: 17248
1631 NSpid: 17248
1632 NSpgid: 17248
1633 NSsid: 17200
1634 VmPeak: 131168 kB
1635 VmSize: 131168 kB
1636 VmLck: 0 kB
1637 VmPin: 0 kB
1638 VmHWM: 13484 kB
1639 VmRSS: 13484 kB
1640 RssAnon: 10264 kB
1641 RssFile: 3220 kB
1642 RssShmem: 0 kB
1643 VmData: 10332 kB
1644 VmStk: 136 kB
1645 VmExe: 992 kB
1646 VmLib: 2104 kB
1647 VmPTE: 76 kB
1648 VmPMD: 12 kB
1649 VmSwap: 0 kB
1650 HugetlbPages: 0 kB # 4.4
1651 CoreDumping: 0 # 4.15
1652 Threads: 1
1653 SigQ: 0/3067
1654 SigPnd: 0000000000000000
1655 ShdPnd: 0000000000000000
1656 SigBlk: 0000000000010000
1657 SigIgn: 0000000000384004
1658 SigCgt: 000000004b813efb
1659 CapInh: 0000000000000000
1660 CapPrm: 0000000000000000
1661 CapEff: 0000000000000000
1662 CapBnd: ffffffffffffffff
1663 CapAmb: 0000000000000000
1664 NoNewPrivs: 0
1665 Seccomp: 0
1666 Speculation_Store_Bypass: vulnerable
1667 Cpus_allowed: 00000001
1668 Cpus_allowed_list: 0
1669 Mems_allowed: 1
1670 Mems_allowed_list: 0
1671 voluntary_ctxt_switches: 150
1672 nonvoluntary_ctxt_switches: 545
1673 The fields are as follows:
1675 Name Command run by this process. Strings longer than
1677 TASK_COMM_LEN [22m(16) characters (including the terminating
1678 null byte) are silently truncated.
1679 Umask Process umask, expressed in octal with a leading zero;
1681 see umask(2). (Since Linux 4.7.)
1682 State Current state of the process. One of "R (running)", "S
1684 (sleeping)", "D (disk sleep)", "T (stopped)", "t (tracing
1685 stop)", "Z (zombie)", or "X (dead)".
1686 Tgid Thread group ID (i.e., Process ID).
1688 Ngid NUMA group ID (0 if none; since Linux 3.13).
1690 Pid Thread ID (see gettid(2)).
1692 PPid PID of parent process.
1694 TracerPid
1696 PID of process tracing this process (0 if not being
1697 traced).
1698 Uid, Gid
1700 Real, effective, saved set, and filesystem UIDs (GIDs).
1701 FDSize Number of file descriptor slots currently allocated.
1703 Groups Supplementary group list.
1705 NStgid Thread group ID (i.e., PID) in each of the PID namespaces
1707 of which [pid] is a member. The leftmost entry shows the
1708 value with respect to the PID namespace of the process
1709 that mounted this procfs (or the root namespace if
1710 mounted by the kernel), followed by the value in succes‐
1711 sively nested inner namespaces. (Since Linux 4.1.)
1712 NSpid Thread ID in each of the PID namespaces of which [pid] is
1714 a member. The fields are ordered as for NStgid. (Since
1715 Linux 4.1.)
1716 NSpgid Process group ID in each of the PID namespaces of which
1718 [pid] is a member. The fields are ordered as for NStgid.
1719 (Since Linux 4.1.)
1720 NSsid descendant namespace session ID hierarchy Session ID in
1722 each of the PID namespaces of which [pid] is a member.
1723 The fields are ordered as for NStgid. (Since Linux 4.1.)
1724 VmPeak Peak virtual memory size.
1726 VmSize Virtual memory size.
1728 VmLck Locked memory size (see mlock(2)).
1730 VmPin Pinned memory size (since Linux 3.2). These are pages
1732 that can't be moved because something needs to directly
1733 access physical memory.
1734 VmHWM Peak resident set size ("high water mark"). This value
1736 is inaccurate; see /proc/[pid]/statm above.
1737 VmRSS Resident set size. Note that the value here is the sum
1739 of RssAnon, RssFile, and RssShmem. This value is inaccu‐
1740 rate; see /proc/[pid]/statm above.
1741 RssAnon
1743 Size of resident anonymous memory. (since Linux 4.5).
1744 This value is inaccurate; see /proc/[pid]/statm above.
1745 RssFile
1747 Size of resident file mappings. (since Linux 4.5). This
1748 value is inaccurate; see /proc/[pid]/statm above.
1749 RssShmem
1751 Size of resident shared memory (includes System V shared
1752 memory, mappings from tmpfs(5), and shared anonymous map‐
1753 pings). (since Linux 4.5).
1754 VmData, VmStk, VmExe
1756 Size of data, stack, and text segments. This value is
1757 inaccurate; see /proc/[pid]/statm above.
1758 VmLib Shared library code size.
1760 VmPTE Page table entries size (since Linux 2.6.10).
1762 VmPMD Size of second-level page tables (added in Linux 4.0; re‐
1764 moved in Linux 4.15).
1765 VmSwap Swapped-out virtual memory size by anonymous private
1767 pages; shmem swap usage is not included (since Linux
1768 2.6.34). This value is inaccurate; see /proc/[pid]/statm
1769 above.
1770 HugetlbPages
1772 Size of hugetlb memory portions (since Linux 4.4).
1773 CoreDumping
1775 Contains the value 1 if the process is currently dumping
1776 core, and 0 if it is not (since Linux 4.15). This infor‐
1777 mation can be used by a monitoring process to avoid
1778 killing a process that is currently dumping core, which
1779 could result in a corrupted core dump file.
1780 Threads
1782 Number of threads in process containing this thread.
1783 SigQ This field contains two slash-separated numbers that re‐
1785 late to queued signals for the real user ID of this
1786 process. The first of these is the number of currently
1787 queued signals for this real user ID, and the second is
1788 the resource limit on the number of queued signals for
1789 this process (see the description of RLIMIT_SIGPENDING in
1790 getrlimit(2)).
1791 SigPnd, ShdPnd
1793 Mask (expressed in hexadecimal) of signals pending for
1794 thread and for process as a whole (see pthreads(7) and
1795 signal(7)).
1796 SigBlk, SigIgn, SigCgt
1798 Masks (expressed in hexadecimal) indicating signals being
1799 blocked, ignored, and caught (see signal(7)).
1800 CapInh, CapPrm, CapEff
1802 Masks (expressed in hexadecimal) of capabilities enabled
1803 in inheritable, permitted, and effective sets (see capa‐
1804 bilities(7)).
1805 CapBnd Capability bounding set, expressed in hexadecimal (since
1807 Linux 2.6.26, see capabilities(7)).
1808 CapAmb Ambient capability set, expressed in hexadecimal (since
1810 Linux 4.3, see capabilities(7)).
1811 NoNewPrivs
1813 Value of the no_new_privs bit (since Linux 4.10, see
1814 prctl(2)).
1815 Seccomp
1817 Seccomp mode of the process (since Linux 3.8, see sec‐
1818 comp(2)). 0 means SECCOMP_MODE_DISABLED; 1 means SEC‐
1819 COMP_MODE_STRICT; 2 means SECCOMP_MODE_FILTER. This
1820 field is provided only if the kernel was built with the
1821 CONFIG_SECCOMP kernel configuration option enabled.
1822 Speculation_Store_Bypass
1824 Speculation flaw mitigation state (since Linux 4.17, see
1825 prctl(2)).
1826 Cpus_allowed
1828 Hexadecimal mask of CPUs on which this process may run
1829 (since Linux 2.6.24, see cpuset(7)).
1830 Cpus_allowed_list
1832 Same as previous, but in "list format" (since Linux
1833 2.6.26, see cpuset(7)).
1834 Mems_allowed
1836 Mask of memory nodes allowed to this process (since Linux
1837 2.6.24, see cpuset(7)).
1838 Mems_allowed_list
1840 Same as previous, but in "list format" (since Linux
1841 2.6.26, see cpuset(7)).
1842 voluntary_ctxt_switches, nonvoluntary_ctxt_switches
1844 Number of voluntary and involuntary context switches
1845 (since Linux 2.6.23).
1846 /proc/[pid]/syscall (since Linux 2.6.27)
1848 This file exposes the system call number and argument registers
1849 for the system call currently being executed by the process,
1850 followed by the values of the stack pointer and program counter
1851 registers. The values of all six argument registers are ex‐
1852 posed, although most system calls use fewer registers.
1853 If the process is blocked, but not in a system call, then the
1855 file displays -1 in place of the system call number, followed by
1856 just the values of the stack pointer and program counter. If
1857 process is not blocked, then the file contains just the string
1858 "running".
1859 This file is present only if the kernel was configured with CON‐
1861 FIG_HAVE_ARCH_TRACEHOOK.
1862 Permission to access this file is governed by a ptrace access
1864 mode PTRACE_MODE_ATTACH_FSCREDS check; see ptrace(2).
1865 /proc/[pid]/task (since Linux 2.6.0)
1867 This is a directory that contains one subdirectory for each
1868 thread in the process. The name of each subdirectory is the nu‐
1869 merical thread ID ([tid]) of the thread (see gettid(2)).
1870 Within each of these subdirectories, there is a set of files
1872 with the same names and contents as under the /proc/[pid] direc‐
1873 tories. For attributes that are shared by all threads, the con‐
1874 tents for each of the files under the task/[tid] subdirectories
1875 will be the same as in the corresponding file in the parent
1876 /proc/[pid] directory (e.g., in a multithreaded process, all of
1877 the task/[tid]/cwd files will have the same value as the
1878 /proc/[pid]/cwd file in the parent directory, since all of the
1879 threads in a process share a working directory). For attributes
1880 that are distinct for each thread, the corresponding files under
1881 task/[tid] may have different values (e.g., various fields in
1882 each of the task/[tid]/status files may be different for each
1883 thread), or they might not exist in /proc/[pid] at all.
1884 In a multithreaded process, the contents of the /proc/[pid]/task
1886 directory are not available if the main thread has already ter‐
1887 minated (typically by calling pthread_exit(3)).
1888 /proc/[pid]/task/[tid]/children (since Linux 3.5)
1890 A space-separated list of child tasks of this task. Each child
1891 task is represented by its TID.
1892 This option is intended for use by the checkpoint-restore (CRIU)
1894 system, and reliably provides a list of children only if all of
1895 the child processes are stopped or frozen. It does not work
1896 properly if children of the target task exit while the file is
1897 being read! Exiting children may cause non-exiting children to
1898 be omitted from the list. This makes this interface even more
1899 unreliable than classic PID-based approaches if the inspected
1900 task and its children aren't frozen, and most code should proba‐
1901 bly not use this interface.
1902 Until Linux 4.2, the presence of this file was governed by the
1904 CONFIG_CHECKPOINT_RESTORE kernel configuration option. Since
1905 Linux 4.2, it is governed by the CONFIG_PROC_CHILDREN option.
1906 /proc/[pid]/timers (since Linux 3.10)
1908 A list of the POSIX timers for this process. Each timer is
1909 listed with a line that starts with the string "ID:". For exam‐
1910 ple:
1911 ID: 1
1913 signal: 60/00007fff86e452a8
1914 notify: signal/pid.2634
1915 ClockID: 0
1916 ID: 0
1917 signal: 60/00007fff86e452a8
1918 notify: signal/pid.2634
1919 ClockID: 1
1920 The lines shown for each timer have the following meanings:
1922 ID The ID for this timer. This is not the same as the timer
1924 ID returned by timer_create(2); rather, it is the same
1925 kernel-internal ID that is available via the si_timerid
1926 field of the siginfo_t structure (see sigaction(2)).
1927 signal This is the signal number that this timer uses to deliver
1929 notifications followed by a slash, and then the
1930 sigev_value value supplied to the signal handler. Valid
1931 only for timers that notify via a signal.
1932 notify The part before the slash specifies the mechanism that
1934 this timer uses to deliver notifications, and is one of
1935 "thread", "signal", or "none". Immediately following the
1936 slash is either the string "tid" for timers with
1937 SIGEV_THREAD_ID notification, or "pid" for timers that
1938 notify by other mechanisms. Following the "." is the PID
1939 of the process (or the kernel thread ID of the thread)
1940 that will be delivered a signal if the timer delivers no‐
1941 tifications via a signal.
1942 ClockID
1944 This field identifies the clock that the timer uses for
1945 measuring time. For most clocks, this is a number that
1946 matches one of the user-space CLOCK_* constants exposed
1947 via <time.h>. CLOCK_PROCESS_CPUTIME_ID timers display
1948 with a value of -6 in this field.
1949 CLOCK_THREAD_CPUTIME_ID timers display with a value of -2
1950 in this field.
1951 This file is available only when the kernel was configured with
1953 CONFIG_CHECKPOINT_RESTORE.
1954 /proc/[pid]/timerslack_ns (since Linux 4.6)
1956 This file exposes the process's "current" timer slack value, ex‐
1957 pressed in nanoseconds. The file is writable, allowing the
1958 process's timer slack value to be changed. Writing 0 to this
1959 file resets the "current" timer slack to the "default" timer
1960 slack value. For further details, see the discussion of
1961 PR_SET_TIMERSLACK in prctl(2).
1962 Initially, permission to access this file was governed by a
1964 ptrace access mode PTRACE_MODE_ATTACH_FSCREDS check (see
1965 ptrace(2)). However, this was subsequently deemed too strict a
1966 requirement (and had the side effect that requiring a process to
1967 have the CAP_SYS_PTRACE capability would also allow it to view
1968 and change any process's memory). Therefore, since Linux 4.9,
1969 only the (weaker) CAP_SYS_NICE capability is required to access
1970 this file.
1971 /proc/[pid]/uid_map, /proc/[pid]/gid_map (since Linux 3.5)
1973 See user_namespaces(7).
1974 /proc/[pid]/wchan (since Linux 2.6.0)
1976 The symbolic name corresponding to the location in the kernel
1977 where the process is sleeping.
1978 Permission to access this file is governed by a ptrace access
1980 mode PTRACE_MODE_READ_FSCREDS check; see ptrace(2).
1981 /proc/[tid]
1983 There is a numerical subdirectory for each running thread that
1984 is not a thread group leader (i.e., a thread whose thread ID is
1985 not the same as its process ID); the subdirectory is named by
1986 the thread ID. Each one of these subdirectories contains files
1987 and subdirectories exposing information about the thread with
1988 the thread ID tid. The contents of these directories are the
1989 same as the corresponding /proc/[pid]/task/[tid] directories.
1990 The /proc/[tid] subdirectories are not visible when iterating
1992 through /proc with getdents(2) (and thus are not visible when
1993 one uses ls(1) to view the contents of /proc). However, the
1994 pathnames of these directories are visible to (i.e., usable as
1995 arguments in) system calls that operate on pathnames.
1996 /proc/apm
1998 Advanced power management version and battery information when
1999 CONFIG_APM is defined at kernel compilation time.
2000 /proc/buddyinfo
2002 This file contains information which is used for diagnosing mem‐
2003 ory fragmentation issues. Each line starts with the identifica‐
2004 tion of the node and the name of the zone which together iden‐
2005 tify a memory region. This is then followed by the count of
2006 available chunks of a certain order in which these zones are
2007 split. The size in bytes of a certain order is given by the
2008 formula:
2009 (2^order) * PAGE_SIZE
2011 The binary buddy allocator algorithm inside the kernel will
2013 split one chunk into two chunks of a smaller order (thus with
2014 half the size) or combine two contiguous chunks into one larger
2015 chunk of a higher order (thus with double the size) to satisfy
2016 allocation requests and to counter memory fragmentation. The
2017 order matches the column number, when starting to count at zero.
2018 For example on an x86-64 system:
2020 Node 0, zone DMA 1 1 1 0 2 1 1 0 1 1 3
2021 Node 0, zone DMA32 65 47 4 81 52 28 13 10 5 1 404
2022 Node 0, zone Normal 216 55 189 101 84 38 37 27 5 3 587
2023 In this example, there is one node containing three zones and
2025 there are 11 different chunk sizes. If the page size is 4 kilo‐
2026 bytes, then the first zone called DMA (on x86 the first 16
2027 megabyte of memory) has 1 chunk of 4 kilobytes (order 0) avail‐
2028 able and has 3 chunks of 4 megabytes (order 10) available.
2029 If the memory is heavily fragmented, the counters for higher or‐
2031 der chunks will be zero and allocation of large contiguous areas
2032 will fail.
2033 Further information about the zones can be found in /proc/zone‐
2035 info.
2036 /proc/bus
2038 Contains subdirectories for installed busses.
2039 /proc/bus/pccard
2041 Subdirectory for PCMCIA devices when CONFIG_PCMCIA is set at
2042 kernel compilation time.
2043 /proc/bus/pccard/drivers
2045 /proc/bus/pci
2047 Contains various bus subdirectories and pseudo-files containing
2048 information about PCI busses, installed devices, and device
2049 drivers. Some of these files are not ASCII.
2050 /proc/bus/pci/devices
2052 Information about PCI devices. They may be accessed through
2053 lspci(8) and setpci(8).
2054 /proc/cgroups (since Linux 2.6.24)
2056 See cgroups(7).
2057 /proc/cmdline
2059 Arguments passed to the Linux kernel at boot time. Often done
2060 via a boot manager such as lilo(8) or grub(8).
2061 /proc/config.gz (since Linux 2.6)
2063 This file exposes the configuration options that were used to
2064 build the currently running kernel, in the same format as they
2065 would be shown in the .config file that resulted when configur‐
2066 ing the kernel (using make xconfig, make config, or similar).
2067 The file contents are compressed; view or search them using
2068 zcat(1) and zgrep(1). As long as no changes have been made to
2069 the following file, the contents of /proc/config.gz are the same
2070 as those provided by:
2071 cat /lib/modules/$(uname -r)/build/.config
2073 /proc/config.gz is provided only if the kernel is configured
2075 with CONFIG_IKCONFIG_PROC.
2076 /proc/crypto
2078 A list of the ciphers provided by the kernel crypto API. For
2079 details, see the kernel Linux Kernel Crypto API documentation
2080 available under the kernel source directory Documenta‐
2081 tion/crypto/ (or Documentation/DocBook before 4.10; the documen‐
2082 tation can be built using a command such as make htmldocs in the
2083 root directory of the kernel source tree).
2084 /proc/cpuinfo
2086 This is a collection of CPU and system architecture dependent
2087 items, for each supported architecture a different list. Two
2088 common entries are processor which gives CPU number and bo‐
2089 gomips; a system constant that is calculated during kernel ini‐
2090 tialization. SMP machines have information for each CPU. The
2091 lscpu(1) command gathers its information from this file.
2092 /proc/devices
2094 Text listing of major numbers and device groups. This can be
2095 used by MAKEDEV scripts for consistency with the kernel.
2096 /proc/diskstats (since Linux 2.5.69)
2098 This file contains disk I/O statistics for each disk device.
2099 See the Linux kernel source file Documentation/iostats.txt for
2100 further information.
2101 /proc/dma
2103 This is a list of the registered ISA DMA (direct memory access)
2104 channels in use.
2105 /proc/driver
2107 Empty subdirectory.
2108 /proc/execdomains
2110 List of the execution domains (ABI personalities).
2111 /proc/fb
2113 Frame buffer information when CONFIG_FB is defined during kernel
2114 compilation.
2115 /proc/filesystems
2117 A text listing of the filesystems which are supported by the
2118 kernel, namely filesystems which were compiled into the kernel
2119 or whose kernel modules are currently loaded. (See also
2120 filesystems(5).) If a filesystem is marked with "nodev", this
2121 means that it does not require a block device to be mounted
2122 (e.g., virtual filesystem, network filesystem).
2123 Incidentally, this file may be used by mount(8) when no filesys‐
2125 tem is specified and it didn't manage to determine the filesys‐
2126 tem type. Then filesystems contained in this file are tried
2127 (excepted those that are marked with "nodev").
2128 /proc/fs
2130 Contains subdirectories that in turn contain files with informa‐
2131 tion about (certain) mounted filesystems.
2132 /proc/ide
2134 This directory exists on systems with the IDE bus. There are
2135 directories for each IDE channel and attached device. Files in‐
2136 clude:
2137 cache buffer size in KB
2139 capacity number of sectors
2140 driver driver version
2141 geometry physical and logical geometry
2142 identify in hexadecimal
2143 media media type
2144 model manufacturer's model number
2145 settings drive settings
2146 smart_thresholds IDE disk management thresholds (in hex)
2147 smart_values IDE disk management values (in hex)
2148 The hdparm(8) utility provides access to this information in a
2150 friendly format.
2151 /proc/interrupts
2153 This is used to record the number of interrupts per CPU per IO
2154 device. Since Linux 2.6.24, for the i386 and x86-64 architec‐
2155 tures, at least, this also includes interrupts internal to the
2156 system (that is, not associated with a device as such), such as
2157 NMI (nonmaskable interrupt), LOC (local timer interrupt), and
2158 for SMP systems, TLB (TLB flush interrupt), RES (rescheduling
2159 interrupt), CAL (remote function call interrupt), and possibly
2160 others. Very easy to read formatting, done in ASCII.
2161 /proc/iomem
2163 I/O memory map in Linux 2.4.
2164 /proc/ioports
2166 This is a list of currently registered Input-Output port regions
2167 that are in use.
2168 /proc/kallsyms (since Linux 2.5.71)
2170 This holds the kernel exported symbol definitions used by the
2171 modules(X) tools to dynamically link and bind loadable modules.
2172 In Linux 2.5.47 and earlier, a similar file with slightly dif‐
2173 ferent syntax was named ksyms.
2174 /proc/kcore
2176 This file represents the physical memory of the system and is
2177 stored in the ELF core file format. With this pseudo-file, and
2178 an unstripped kernel (/usr/src/linux/vmlinux) binary, GDB can be
2179 used to examine the current state of any kernel data structures.
2180 The total length of the file is the size of physical memory
2182 (RAM) plus 4 KiB.
2183 /proc/keys (since Linux 2.6.10)
2185 See keyrings(7).
2186 /proc/key-users (since Linux 2.6.10)
2188 See keyrings(7).
2189 /proc/kmsg
2191 This file can be used instead of the syslog(2) system call to
2192 read kernel messages. A process must have superuser privileges
2193 to read this file, and only one process should read this file.
2194 This file should not be read if a syslog process is running
2195 which uses the syslog(2) system call facility to log kernel mes‐
2196 sages.
2197 Information in this file is retrieved with the dmesg(1) program.
2199 /proc/kpagecgroup (since Linux 4.3)
2201 This file contains a 64-bit inode number of the memory cgroup
2202 each page is charged to, indexed by page frame number (see the
2203 discussion of /proc/[pid]/pagemap).
2204 The /proc/kpagecgroup file is present only if the CONFIG_MEMCG
2206 kernel configuration option is enabled.
2207 /proc/kpagecount (since Linux 2.6.25)
2209 This file contains a 64-bit count of the number of times each
2210 physical page frame is mapped, indexed by page frame number (see
2211 the discussion of /proc/[pid]/pagemap).
2212 The /proc/kpagecount file is present only if the CON‐
2214 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
2215 /proc/kpageflags (since Linux 2.6.25)
2217 This file contains 64-bit masks corresponding to each physical
2218 page frame; it is indexed by page frame number (see the discus‐
2219 sion of /proc/[pid]/pagemap). The bits are as follows:
2220 0 - KPF_LOCKED
2222 1 - KPF_ERROR
2223 2 - KPF_REFERENCED
2224 3 - KPF_UPTODATE
2225 4 - KPF_DIRTY
2226 5 - KPF_LRU
2227 6 - KPF_ACTIVE
2228 7 - KPF_SLAB
2229 8 - KPF_WRITEBACK
2230 9 - KPF_RECLAIM
2231 10 - KPF_BUDDY
2232 11 - KPF_MMAP (since Linux 2.6.31)
2233 12 - KPF_ANON (since Linux 2.6.31)
2234 13 - KPF_SWAPCACHE (since Linux 2.6.31)
2235 14 - KPF_SWAPBACKED (since Linux 2.6.31)
2236 15 - KPF_COMPOUND_HEAD (since Linux 2.6.31)
2237 16 - KPF_COMPOUND_TAIL (since Linux 2.6.31)
2238 17 - KPF_HUGE (since Linux 2.6.31)
2239 18 - KPF_UNEVICTABLE (since Linux 2.6.31)
2240 19 - KPF_HWPOISON (since Linux 2.6.31)
2241 20 - KPF_NOPAGE (since Linux 2.6.31)
2242 21 - KPF_KSM (since Linux 2.6.32)
2243 22 - KPF_THP (since Linux 3.4)
2244 23 - KPF_BALLOON (since Linux 3.18)
2245 24 - KPF_ZERO_PAGE (since Linux 4.0)
2246 25 - KPF_IDLE (since Linux 4.3)
2247 For further details on the meanings of these bits, see the ker‐
2249 nel source file Documentation/admin-guide/mm/pagemap.rst. Be‐
2250 fore kernel 2.6.29, KPF_WRITEBACK, KPF_RECLAIM, KPF_BUDDY, and
2251 KPF_LOCKED did not report correctly.
2252 The /proc/kpageflags file is present only if the CON‐
2254 FIG_PROC_PAGE_MONITOR kernel configuration option is enabled.
2255 /proc/ksyms (Linux 1.1.23–2.5.47)
2257 See /proc/kallsyms.
2258 /proc/loadavg
2260 The first three fields in this file are load average figures
2261 giving the number of jobs in the run queue (state R) or waiting
2262 for disk I/O (state D) averaged over 1, 5, and 15 minutes. They
2263 are the same as the load average numbers given by uptime(1) and
2264 other programs. The fourth field consists of two numbers sepa‐
2265 rated by a slash (/). The first of these is the number of cur‐
2266 rently runnable kernel scheduling entities (processes, threads).
2267 The value after the slash is the number of kernel scheduling en‐
2268 tities that currently exist on the system. The fifth field is
2269 the PID of the process that was most recently created on the
2270 system.
2271 /proc/locks
2273 This file shows current file locks (flock(2) and fcntl(2)) and
2274 leases (fcntl(2)).
2275 An example of the content shown in this file is the following:
2277 1: POSIX ADVISORY READ 5433 08:01:7864448 128 128
2279 2: FLOCK ADVISORY WRITE 2001 08:01:7864554 0 EOF
2280 3: FLOCK ADVISORY WRITE 1568 00:2f:32388 0 EOF
2281 4: POSIX ADVISORY WRITE 699 00:16:28457 0 EOF
2282 5: POSIX ADVISORY WRITE 764 00:16:21448 0 0
2283 6: POSIX ADVISORY READ 3548 08:01:7867240 1 1
2284 7: POSIX ADVISORY READ 3548 08:01:7865567 1826 2335
2285 8: OFDLCK ADVISORY WRITE -1 08:01:8713209 128 191
2286 The fields shown in each line are as follows:
2288 (1) The ordinal position of the lock in the list.
2290 (2) The lock type. Values that may appear here include:
2292 FLOCK This is a BSD file lock created using flock(2).
2294 OFDLCK This is an open file description (OFD) lock created
2296 using fcntl(2).
2297 POSIX This is a POSIX byte-range lock created using fc‐
2299 ntl(2).
2300 (3) Among the strings that can appear here are the following:
2302 ADVISORY
2304 This is an advisory lock.
2305 MANDATORY
2307 This is a mandatory lock.
2308 (4) The type of lock. Values that can appear here are:
2310 READ This is a POSIX or OFD read lock, or a BSD shared
2312 lock.
2313 WRITE This is a POSIX or OFD write lock, or a BSD exclusive
2315 lock.
2316 (5) The PID of the process that owns the lock.
2318 Because OFD locks are not owned by a single process (since
2320 multiple processes may have file descriptors that refer to
2321 the same open file description), the value -1 is displayed
2322 in this field for OFD locks. (Before kernel 4.14, a bug
2323 meant that the PID of the process that initially acquired
2324 the lock was displayed instead of the value -1.)
2325 (6) Three colon-separated subfields that identify the major and
2327 minor device ID of the device containing the filesystem
2328 where the locked file resides, followed by the inode number
2329 of the locked file.
2330 (7) The byte offset of the first byte of the lock. For BSD
2332 locks, this value is always 0.
2333 (8) The byte offset of the last byte of the lock. EOF in this
2335 field means that the lock extends to the end of the file.
2336 For BSD locks, the value shown is always EOF.
2337 Since Linux 4.9, the list of locks shown in /proc/locks is fil‐
2339 tered to show just the locks for the processes in the PID name‐
2340 space (see pid_namespaces(7)) for which the /proc filesystem was
2341 mounted. (In the initial PID namespace, there is no filtering
2342 of the records shown in this file.)
2343 The lslocks(8) command provides a bit more information about
2345 each lock.
2346 /proc/malloc (only up to and including Linux 2.2)
2348 This file is present only if CONFIG_DEBUG_MALLOC was defined
2349 during compilation.
2350 /proc/meminfo
2352 This file reports statistics about memory usage on the system.
2353 It is used by free(1) to report the amount of free and used mem‐
2354 ory (both physical and swap) on the system as well as the shared
2355 memory and buffers used by the kernel. Each line of the file
2356 consists of a parameter name, followed by a colon, the value of
2357 the parameter, and an option unit of measurement (e.g., "kB").
2358 The list below describes the parameter names and the format
2359 specifier required to read the field value. Except as noted be‐
2360 low, all of the fields have been present since at least Linux
2361 2.6.0. Some fields are displayed only if the kernel was config‐
2362 ured with various options; those dependencies are noted in the
2363 list.
2364 MemTotal %lu
2366 Total usable RAM (i.e., physical RAM minus a few reserved
2367 bits and the kernel binary code).
2368 MemFree %lu
2370 The sum of LowFree+HighFree.
2371 MemAvailable %lu (since Linux 3.14)
2373 An estimate of how much memory is available for starting
2374 new applications, without swapping.
2375 Buffers %lu
2377 Relatively temporary storage for raw disk blocks that
2378 shouldn't get tremendously large (20 MB or so).
2379 Cached %lu
2381 In-memory cache for files read from the disk (the page
2382 cache). Doesn't include SwapCached.
2383 SwapCached %lu
2385 Memory that once was swapped out, is swapped back in but
2386 still also is in the swap file. (If memory pressure is
2387 high, these pages don't need to be swapped out again be‐
2388 cause they are already in the swap file. This saves
2389 I/O.)
2390 Active %lu
2392 Memory that has been used more recently and usually not
2393 reclaimed unless absolutely necessary.
2394 Inactive %lu
2396 Memory which has been less recently used. It is more el‐
2397 igible to be reclaimed for other purposes.
2398 Active(anon) %lu (since Linux 2.6.28)
2400 [To be documented.]
2401 Inactive(anon) %lu (since Linux 2.6.28)
2403 [To be documented.]
2404 Active(file) %lu (since Linux 2.6.28)
2406 [To be documented.]
2407 Inactive(file) %lu (since Linux 2.6.28)
2409 [To be documented.]
2410 Unevictable %lu (since Linux 2.6.28)
2412 (From Linux 2.6.28 to 2.6.30, CONFIG_UNEVICTABLE_LRU was
2413 required.) [To be documented.]
2414 Mlocked %lu (since Linux 2.6.28)
2416 (From Linux 2.6.28 to 2.6.30, CONFIG_UNEVICTABLE_LRU was
2417 required.) [To be documented.]
2418 HighTotal %lu
2420 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2421 Total amount of highmem. Highmem is all memory above
2422 ~860 MB of physical memory. Highmem areas are for use by
2423 user-space programs, or for the page cache. The kernel
2424 must use tricks to access this memory, making it slower
2425 to access than lowmem.
2426 HighFree %lu
2428 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2429 Amount of free highmem.
2430 LowTotal %lu
2432 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2433 Total amount of lowmem. Lowmem is memory which can be
2434 used for everything that highmem can be used for, but it
2435 is also available for the kernel's use for its own data
2436 structures. Among many other things, it is where every‐
2437 thing from Slab is allocated. Bad things happen when
2438 you're out of lowmem.
2439 LowFree %lu
2441 (Starting with Linux 2.6.19, CONFIG_HIGHMEM is required.)
2442 Amount of free lowmem.
2443 MmapCopy %lu (since Linux 2.6.29)
2445 (CONFIG_MMU is required.) [To be documented.]
2446 SwapTotal %lu
2448 Total amount of swap space available.
2449 SwapFree %lu
2451 Amount of swap space that is currently unused.
2452 Dirty %lu
2454 Memory which is waiting to get written back to the disk.
2455 Writeback %lu
2457 Memory which is actively being written back to the disk.
2458 AnonPages %lu (since Linux 2.6.18)
2460 Non-file backed pages mapped into user-space page tables.
2461 Mapped %lu
2463 Files which have been mapped into memory (with mmap(2)),
2464 such as libraries.
2465 Shmem %lu (since Linux 2.6.32)
2467 Amount of memory consumed in tmpfs(5) filesystems.
2468 KReclaimable %lu (since Linux 4.20)
2470 Kernel allocations that the kernel will attempt to re‐
2471 claim under memory pressure. Includes SReclaimable (be‐
2472 low), and other direct allocations with a shrinker.
2473 Slab %lu
2475 In-kernel data structures cache. (See slabinfo(5).)
2476 SReclaimable %lu (since Linux 2.6.19)
2478 Part of Slab, that might be reclaimed, such as caches.
2479 SUnreclaim %lu (since Linux 2.6.19)
2481 Part of Slab, that cannot be reclaimed on memory pres‐
2482 sure.
2483 KernelStack %lu (since Linux 2.6.32)
2485 Amount of memory allocated to kernel stacks.
2486 PageTables %lu (since Linux 2.6.18)
2488 Amount of memory dedicated to the lowest level of page
2489 tables.
2490 Quicklists %lu (since Linux 2.6.27)
2492 (CONFIG_QUICKLIST is required.) [To be documented.]
2493 NFS_Unstable %lu (since Linux 2.6.18)
2495 NFS pages sent to the server, but not yet committed to
2496 stable storage.
2497 Bounce %lu (since Linux 2.6.18)
2499 Memory used for block device "bounce buffers".
2500 WritebackTmp %lu (since Linux 2.6.26)
2502 Memory used by FUSE for temporary writeback buffers.
2503 CommitLimit %lu (since Linux 2.6.10)
2505 This is the total amount of memory currently available to
2506 be allocated on the system, expressed in kilobytes. This
2507 limit is adhered to only if strict overcommit accounting
2508 is enabled (mode 2 in /proc/sys/vm/overcommit_memory).
2509 The limit is calculated according to the formula de‐
2510 scribed under /proc/sys/vm/overcommit_memory. For fur‐
2511 ther details, see the kernel source file Documenta‐
2512 tion/vm/overcommit-accounting.rst.
2513 Committed_AS %lu
2515 The amount of memory presently allocated on the system.
2516 The committed memory is a sum of all of the memory which
2517 has been allocated by processes, even if it has not been
2518 "used" by them as of yet. A process which allocates 1 GB
2519 of memory (using malloc(3) or similar), but touches only
2520 300 MB of that memory will show up as using only 300 MB
2521 of memory even if it has the address space allocated for
2522 the entire 1 GB.
2523 This 1 GB is memory which has been "committed" to by the
2525 VM and can be used at any time by the allocating applica‐
2526 tion. With strict overcommit enabled on the system (mode
2527 2 in /proc/sys/vm/overcommit_memory), allocations which
2528 would exceed the CommitLimit will not be permitted. This
2529 is useful if one needs to guarantee that processes will
2530 not fail due to lack of memory once that memory has been
2531 successfully allocated.
2532 VmallocTotal %lu
2534 Total size of vmalloc memory area.
2535 VmallocUsed %lu
2537 Amount of vmalloc area which is used. Since Linux 4.4,
2538 this field is no longer calculated, and is hard coded as
2539 0. See /proc/vmallocinfo.
2540 VmallocChunk %lu
2542 Largest contiguous block of vmalloc area which is free.
2543 Since Linux 4.4, this field is no longer calculated and
2544 is hard coded as 0. See /proc/vmallocinfo.
2545 HardwareCorrupted %lu (since Linux 2.6.32)
2547 (CONFIG_MEMORY_FAILURE is required.) [To be documented.]
2548 LazyFree %lu (since Linux 4.12)
2550 Shows the amount of memory marked by madvise(2)
2551 MADV_FREE.
2552 AnonHugePages %lu (since Linux 2.6.38)
2554 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Non-file
2555 backed huge pages mapped into user-space page tables.
2556 ShmemHugePages %lu (since Linux 4.8)
2558 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Memory used
2559 by shared memory (shmem) and tmpfs(5) allocated with huge
2560 pages.
2561 ShmemPmdMapped %lu (since Linux 4.8)
2563 (CONFIG_TRANSPARENT_HUGEPAGE is required.) Shared memory
2564 mapped into user space with huge pages.
2565 CmaTotal %lu (since Linux 3.1)
2567 Total CMA (Contiguous Memory Allocator) pages. (CON‐
2568 FIG_CMA is required.)
2569 CmaFree %lu (since Linux 3.1)
2571 Free CMA (Contiguous Memory Allocator) pages. (CON‐
2572 FIG_CMA is required.)
2573 HugePages_Total %lu
2575 (CONFIG_HUGETLB_PAGE is required.) The size of the pool
2576 of huge pages.
2577 HugePages_Free %lu
2579 (CONFIG_HUGETLB_PAGE is required.) The number of huge
2580 pages in the pool that are not yet allocated.
2581 HugePages_Rsvd %lu (since Linux 2.6.17)
2583 (CONFIG_HUGETLB_PAGE is required.) This is the number of
2584 huge pages for which a commitment to allocate from the
2585 pool has been made, but no allocation has yet been made.
2586 These reserved huge pages guarantee that an application
2587 will be able to allocate a huge page from the pool of
2588 huge pages at fault time.
2589 HugePages_Surp %lu (since Linux 2.6.24)
2591 (CONFIG_HUGETLB_PAGE is required.) This is the number of
2592 huge pages in the pool above the value in
2593 /proc/sys/vm/nr_hugepages. The maximum number of surplus
2594 huge pages is controlled by /proc/sys/vm/nr_overcom‐
2595 mit_hugepages.
2596 Hugepagesize %lu
2598 (CONFIG_HUGETLB_PAGE is required.) The size of huge
2599 pages.
2600 DirectMap4k %lu (since Linux 2.6.27)
2602 Number of bytes of RAM linearly mapped by kernel in 4 kB
2603 pages. (x86.)
2604 DirectMap4M %lu (since Linux 2.6.27)
2606 Number of bytes of RAM linearly mapped by kernel in 4 MB
2607 pages. (x86 with CONFIG_X86_64 or CONFIG_X86_PAE en‐
2608 abled.)
2609 DirectMap2M %lu (since Linux 2.6.27)
2611 Number of bytes of RAM linearly mapped by kernel in 2 MB
2612 pages. (x86 with neither CONFIG_X86_64 nor CON‐
2613 FIG_X86_PAE enabled.)
2614 DirectMap1G %lu (since Linux 2.6.27)
2616 (x86 with CONFIG_X86_64 and CONFIG_X86_DIRECT_GBPAGES en‐
2617 abled.)
2618 /proc/modules
2620 A text list of the modules that have been loaded by the system.
2621 See also lsmod(8).
2622 /proc/mounts
2624 Before kernel 2.4.19, this file was a list of all the filesys‐
2625 tems currently mounted on the system. With the introduction of
2626 per-process mount namespaces in Linux 2.4.19 (see mount_name‐
2627 spaces(7)), this file became a link to /proc/self/mounts, which
2628 lists the mount points of the process's own mount namespace.
2629 The format of this file is documented in fstab(5).
2630 /proc/mtrr
2632 Memory Type Range Registers. See the Linux kernel source file
2633 Documentation/x86/mtrr.txt (or Documentation/mtrr.txt before
2634 Linux 2.6.28) for details.
2635 /proc/net
2637 This directory contains various files and subdirectories con‐
2638 taining information about the networking layer. The files con‐
2639 tain ASCII structures and are, therefore, readable with cat(1).
2640 However, the standard netstat(8) suite provides much cleaner ac‐
2641 cess to these files.
2642 With the advent of network namespaces, various information re‐
2644 lating to the network stack is virtualized (see network_name‐
2645 spaces(7)). Thus, since Linux 2.6.25, /proc/net is a symbolic
2646 link to the directory /proc/self/net, which contains the same
2647 files and directories as listed below. However, these files and
2648 directories now expose information for the network namespace of
2649 which the process is a member.
2650 /proc/net/arp
2652 This holds an ASCII readable dump of the kernel ARP table used
2653 for address resolutions. It will show both dynamically learned
2654 and preprogrammed ARP entries. The format is:
2655 IP address HW type Flags HW address Mask Device
2657 192.168.0.50 0x1 0x2 00:50:BF:25:68:F3 * eth0
2658 192.168.0.250 0x1 0xc 00:00:00:00:00:00 * eth0
2659 Here "IP address" is the IPv4 address of the machine and the "HW
2661 type" is the hardware type of the address from RFC 826. The
2662 flags are the internal flags of the ARP structure (as defined in
2663 /usr/include/linux/if_arp.h) and the "HW address" is the data
2664 link layer mapping for that IP address if it is known.
2665 /proc/net/dev
2667 The dev pseudo-file contains network device status information.
2668 This gives the number of received and sent packets, the number
2669 of errors and collisions and other basic statistics. These are
2670 used by the ifconfig(8) program to report device status. The
2671 format is:
2672 Inter-| Receive | Transmit
2674 face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
2675 lo: 2776770 11307 0 0 0 0 0 0 2776770 11307 0 0 0 0 0 0
2676 eth0: 1215645 2751 0 0 0 0 0 0 1782404 4324 0 0 0 427 0 0
2677 ppp0: 1622270 5552 1 0 0 0 0 0 354130 5669 0 0 0 0 0 0
2678 tap0: 7714 81 0 0 0 0 0 0 7714 81 0 0 0 0 0 0
2679 /proc/net/dev_mcast
2681 Defined in /usr/src/linux/net/core/dev_mcast.c:
2682 indx interface_name dmi_u dmi_g dmi_address
2684 2 eth0 1 0 01005e000001
2685 3 eth1 1 0 01005e000001
2686 4 eth2 1 0 01005e000001
2687 /proc/net/igmp
2689 Internet Group Management Protocol. Defined in
2690 /usr/src/linux/net/core/igmp.c.
2691 /proc/net/rarp
2693 This file uses the same format as the arp file and contains the
2694 current reverse mapping database used to provide rarp(8) reverse
2695 address lookup services. If RARP is not configured into the
2696 kernel, this file will not be present.
2697 /proc/net/raw
2699 Holds a dump of the RAW socket table. Much of the information
2700 is not of use apart from debugging. The "sl" value is the ker‐
2701 nel hash slot for the socket, the "local_address" is the local
2702 address and protocol number pair. "St" is the internal status
2703 of the socket. The "tx_queue" and "rx_queue" are the outgoing
2704 and incoming data queue in terms of kernel memory usage. The
2705 "tr", "tm->when", and "rexmits" fields are not used by RAW. The
2706 "uid" field holds the effective UID of the creator of the
2707 socket.
2708 /proc/net/snmp
2710 This file holds the ASCII data needed for the IP, ICMP, TCP, and
2711 UDP management information bases for an SNMP agent.
2712 /proc/net/tcp
2714 Holds a dump of the TCP socket table. Much of the information
2715 is not of use apart from debugging. The "sl" value is the ker‐
2716 nel hash slot for the socket, the "local_address" is the local
2717 address and port number pair. The "rem_address" is the remote
2718 address and port number pair (if connected). "St" is the inter‐
2719 nal status of the socket. The "tx_queue" and "rx_queue" are the
2720 outgoing and incoming data queue in terms of kernel memory us‐
2721 age. The "tr", "tm->when", and "rexmits" fields hold internal
2722 information of the kernel socket state and are useful only for
2723 debugging. The "uid" field holds the effective UID of the cre‐
2724 ator of the socket.
2725 /proc/net/udp
2727 Holds a dump of the UDP socket table. Much of the information
2728 is not of use apart from debugging. The "sl" value is the ker‐
2729 nel hash slot for the socket, the "local_address" is the local
2730 address and port number pair. The "rem_address" is the remote
2731 address and port number pair (if connected). "St" is the inter‐
2732 nal status of the socket. The "tx_queue" and "rx_queue" are the
2733 outgoing and incoming data queue in terms of kernel memory us‐
2734 age. The "tr", "tm->when", and "rexmits" fields are not used by
2735 UDP. The "uid" field holds the effective UID of the creator of
2736 the socket. The format is:
2737 sl local_address rem_address st tx_queue rx_queue tr rexmits tm->when uid
2739 1: 01642C89:0201 0C642C89:03FF 01 00000000:00000001 01:000071BA 00000000 0
2740 1: 00000000:0801 00000000:0000 0A 00000000:00000000 00:00000000 6F000100 0
2741 1: 00000000:0201 00000000:0000 0A 00000000:00000000 00:00000000 00000000 0
2742 /proc/net/unix
2744 Lists the UNIX domain sockets present within the system and
2745 their status. The format is:
2746 Num RefCount Protocol Flags Type St Inode Path
2748 0: 00000002 00000000 00000000 0001 03 42
2749 1: 00000001 00000000 00010000 0001 01 1948 /dev/printer
2750 The fields are as follows:
2752 Num: the kernel table slot number.
2754 RefCount: the number of users of the socket.
2756 Protocol: currently always 0.
2758 Flags: the internal kernel flags holding the status of the
2760 socket.
2761 Type: the socket type. For SOCK_STREAM sockets, this is
2763 0001; for SOCK_DGRAM sockets, it is 0002; and for
2764 SOCK_SEQPACKET sockets, it is 0005.
2765 St: the internal state of the socket.
2767 Inode: the inode number of the socket.
2769 Path: the bound pathname (if any) of the socket. Sockets in
2771 the abstract namespace are included in the list, and
2772 are shown with a Path that commences with the charac‐
2773 ter '@'.
2774 /proc/net/netfilter/nfnetlink_queue
2776 This file contains information about netfilter user-space queue‐
2777 ing, if used. Each line represents a queue. Queues that have
2778 not been subscribed to by user space are not shown.
2779 1 4207 0 2 65535 0 0 0 1
2781 (1) (2) (3)(4) (5) (6) (7) (8)
2782 The fields in each line are:
2784 (1) The ID of the queue. This matches what is specified in the
2786 --queue-num or --queue-balance options to the iptables(8)
2787 NFQUEUE target. See iptables-extensions(8) for more infor‐
2788 mation.
2789 (2) The netlink port ID subscribed to the queue.
2791 (3) The number of packets currently queued and waiting to be
2793 processed by the application.
2794 (4) The copy mode of the queue. It is either 1 (metadata only)
2796 or 2 (also copy payload data to user space).
2797 (5) Copy range; that is, how many bytes of packet payload
2799 should be copied to user space at most.
2800 (6) queue dropped. Number of packets that had to be dropped by
2802 the kernel because too many packets are already waiting for
2803 user space to send back the mandatory accept/drop verdicts.
2804 (7) queue user dropped. Number of packets that were dropped
2806 within the netlink subsystem. Such drops usually happen
2807 when the corresponding socket buffer is full; that is, user
2808 space is not able to read messages fast enough.
2809 (8) sequence number. Every queued packet is associated with a
2811 (32-bit) monotonically increasing sequence number. This
2812 shows the ID of the most recent packet queued.
2813 The last number exists only for compatibility reasons and is al‐
2815 ways 1.
2816 /proc/partitions
2818 Contains the major and minor numbers of each partition as well
2819 as the number of 1024-byte blocks and the partition name.
2820 /proc/pci
2822 This is a listing of all PCI devices found during kernel ini‐
2823 tialization and their configuration.
2824 This file has been deprecated in favor of a new /proc interface
2826 for PCI (/proc/bus/pci). It became optional in Linux 2.2
2827 (available with CONFIG_PCI_OLD_PROC set at kernel compilation).
2828 It became once more nonoptionally enabled in Linux 2.4. Next,
2829 it was deprecated in Linux 2.6 (still available with CON‐
2830 FIG_PCI_LEGACY_PROC set), and finally removed altogether since
2831 Linux 2.6.17.
2832 /proc/profile (since Linux 2.4)
2834 This file is present only if the kernel was booted with the pro‐
2835 file=1 command-line option. It exposes kernel profiling infor‐
2836 mation in a binary format for use by readprofile(1). Writing
2837 (e.g., an empty string) to this file resets the profiling coun‐
2838 ters; on some architectures, writing a binary integer "profiling
2839 multiplier" of size sizeof(int) sets the profiling interrupt
2840 frequency.
2841 /proc/scsi
2843 A directory with the scsi mid-level pseudo-file and various SCSI
2844 low-level driver directories, which contain a file for each SCSI
2845 host in this system, all of which give the status of some part
2846 of the SCSI IO subsystem. These files contain ASCII structures
2847 and are, therefore, readable with cat(1).
2848 You can also write to some of the files to reconfigure the sub‐
2850 system or switch certain features on or off.
2851 /proc/scsi/scsi
2853 This is a listing of all SCSI devices known to the kernel. The
2854 listing is similar to the one seen during bootup. scsi cur‐
2855 rently supports only the add-single-device command which allows
2856 root to add a hotplugged device to the list of known devices.
2857 The command
2859 echo 'scsi add-single-device 1 0 5 0' > /proc/scsi/scsi
2861 will cause host scsi1 to scan on SCSI channel 0 for a device on
2863 ID 5 LUN 0. If there is already a device known on this address
2864 or the address is invalid, an error will be returned.
2865 /proc/scsi/[drivername]
2867 [drivername] can currently be NCR53c7xx, aha152x, aha1542,
2868 aha1740, aic7xxx, buslogic, eata_dma, eata_pio, fdomain, in2000,
2869 pas16, qlogic, scsi_debug, seagate, t128, u15-24f, ultrastore,
2870 or wd7000. These directories show up for all drivers that reg‐
2871 istered at least one SCSI HBA. Every directory contains one
2872 file per registered host. Every host-file is named after the
2873 number the host was assigned during initialization.
2874 Reading these files will usually show driver and host configura‐
2876 tion, statistics, and so on.
2877 Writing to these files allows different things on different
2879 hosts. For example, with the latency and nolatency commands,
2880 root can switch on and off command latency measurement code in
2881 the eata_dma driver. With the lockup and unlock commands, root
2882 can control bus lockups simulated by the scsi_debug driver.
2883 /proc/self
2885 This directory refers to the process accessing the /proc
2886 filesystem, and is identical to the /proc directory named by the
2887 process ID of the same process.
2888 /proc/slabinfo
2890 Information about kernel caches. See slabinfo(5) for details.
2891 /proc/stat
2893 kernel/system statistics. Varies with architecture. Common en‐
2894 tries include:
2895 cpu 10132153 290696 3084719 46828483 16683 0 25195 0 175628 0
2897 cpu0 1393280 32966 572056 13343292 6130 0 17875 0 23933 0
2898 The amount of time, measured in units of USER_HZ
2899 (1/100ths of a second on most architectures, use
2900 sysconf(_SC_CLK_TCK) to obtain the right value), that the
2901 system ("cpu" line) or the specific CPU ("cpuN" line)
2902 spent in various states:
2903 user (1) Time spent in user mode.
2905 nice (2) Time spent in user mode with low priority
2907 (nice).
2908 system (3) Time spent in system mode.
2910 idle (4) Time spent in the idle task. This value
2912 should be USER_HZ times the second entry in the
2913 /proc/uptime pseudo-file.
2914 iowait (since Linux 2.5.41)
2916 (5) Time waiting for I/O to complete. This value
2917 is not reliable, for the following reasons:
2918 1. The CPU will not wait for I/O to complete;
2920 iowait is the time that a task is waiting for
2921 I/O to complete. When a CPU goes into idle
2922 state for outstanding task I/O, another task
2923 will be scheduled on this CPU.
2924 2. On a multi-core CPU, the task waiting for I/O
2926 to complete is not running on any CPU, so the
2927 iowait of each CPU is difficult to calculate.
2928 3. The value in this field may decrease in certain
2930 conditions.
2931 irq (since Linux 2.6.0)
2933 (6) Time servicing interrupts.
2934 softirq (since Linux 2.6.0)
2936 (7) Time servicing softirqs.
2937 steal (since Linux 2.6.11)
2939 (8) Stolen time, which is the time spent in other
2940 operating systems when running in a virtualized
2941 environment
2942 guest (since Linux 2.6.24)
2944 (9) Time spent running a virtual CPU for guest op‐
2945 erating systems under the control of the Linux
2946 kernel.
2947 guest_nice (since Linux 2.6.33)
2949 (10) Time spent running a niced guest (virtual CPU
2950 for guest operating systems under the control of
2951 the Linux kernel).
2952 page 5741 1808
2954 The number of pages the system paged in and the number
2955 that were paged out (from disk).
2956 swap 1 0
2958 The number of swap pages that have been brought in and
2959 out.
2960 intr 1462898
2962 This line shows counts of interrupts serviced since boot
2963 time, for each of the possible system interrupts. The
2964 first column is the total of all interrupts serviced in‐
2965 cluding unnumbered architecture specific interrupts; each
2966 subsequent column is the total for that particular num‐
2967 bered interrupt. Unnumbered interrupts are not shown,
2968 only summed into the total.
2969 disk_io: (2,0):(31,30,5764,1,2) (3,0):...
2971 (major,disk_idx):(noinfo, read_io_ops, blks_read,
2972 write_io_ops, blks_written)
2973 (Linux 2.4 only)
2974 ctxt 115315
2976 The number of context switches that the system underwent.
2977 btime 769041601
2979 boot time, in seconds since the Epoch, 1970-01-01
2980 00:00:00 +0000 (UTC).
2981 processes 86031
2983 Number of forks since boot.
2984 procs_running 6
2986 Number of processes in runnable state. (Linux 2.5.45 on‐
2987 ward.)
2988 procs_blocked 2
2990 Number of processes blocked waiting for I/O to complete.
2991 (Linux 2.5.45 onward.)
2992 softirq 229245889 94 60001584 13619 5175704 2471304 28 51212741
2994 59130143 0 51240672
2995 This line shows the number of softirq for all CPUs. The
2996 first column is the total of all softirqs and each subse‐
2997 quent column is the total for particular softirq. (Linux
2998 2.6.31 onward.)
2999 /proc/swaps
3001 Swap areas in use. See also swapon(8).
3002 /proc/sys
3004 This directory (present since 1.3.57) contains a number of files
3005 and subdirectories corresponding to kernel variables. These
3006 variables can be read and in some cases modified using the /proc
3007 filesystem, and the (deprecated) sysctl(2) system call.
3008 String values may be terminated by either '\0' or '\n'.
3010 Integer and long values may be written either in decimal or in
3012 hexadecimal notation (e.g., 0x3FFF). When writing multiple in‐
3013 teger or long values, these may be separated by any of the fol‐
3014 lowing whitespace characters: ' ', '\t', or '\n'. Using other
3015 separators leads to the error EINVAL.
3016 /proc/sys/abi (since Linux 2.4.10)
3018 This directory may contain files with application binary infor‐
3019 mation. See the Linux kernel source file Documenta‐
3020 tion/sysctl/abi.txt for more information.
3021 /proc/sys/debug
3023 This directory may be empty.
3024 /proc/sys/dev
3026 This directory contains device-specific information (e.g.,
3027 dev/cdrom/info). On some systems, it may be empty.
3028 /proc/sys/fs
3030 This directory contains the files and subdirectories for kernel
3031 variables related to filesystems.
3032 /proc/sys/fs/aio-max-nr and /proc/sys/fs/aio-nr (since Linux 2.6.4)
3034 aio-nr is the running total of the number of events specified by
3035 io_setup(2) calls for all currently active AIO contexts. If
3036 aio-nr reaches aio-max-nr, then io_setup(2) will fail with the
3037 error EAGAIN. Raising aio-max-nr does not result in the preal‐
3038 location or resizing of any kernel data structures.
3039 /proc/sys/fs/binfmt_misc
3041 Documentation for files in this directory can be found in the
3042 Linux kernel source in the file Documentation/ad‐
3043 min-guide/binfmt-misc.rst (or in Documentation/binfmt_misc.txt
3044 on older kernels).
3045 /proc/sys/fs/dentry-state (since Linux 2.2)
3047 This file contains information about the status of the directory
3048 cache (dcache). The file contains six numbers, nr_dentry,
3049 nr_unused, age_limit (age in seconds), want_pages (pages re‐
3050 quested by system) and two dummy values.
3051 * nr_dentry is the number of allocated dentries (dcache en‐
3053 tries). This field is unused in Linux 2.2.
3054 * nr_unused is the number of unused dentries.
3056 * age_limit is the age in seconds after which dcache entries can
3058 be reclaimed when memory is short.
3059 * want_pages is nonzero when the kernel has called
3061 shrink_dcache_pages() and the dcache isn't pruned yet.
3062 /proc/sys/fs/dir-notify-enable
3064 This file can be used to disable or enable the dnotify interface
3065 described in fcntl(2) on a system-wide basis. A value of 0 in
3066 this file disables the interface, and a value of 1 enables it.
3067 /proc/sys/fs/dquot-max
3069 This file shows the maximum number of cached disk quota entries.
3070 On some (2.4) systems, it is not present. If the number of free
3071 cached disk quota entries is very low and you have some awesome
3072 number of simultaneous system users, you might want to raise the
3073 limit.
3074 /proc/sys/fs/dquot-nr
3076 This file shows the number of allocated disk quota entries and
3077 the number of free disk quota entries.
3078 /proc/sys/fs/epoll (since Linux 2.6.28)
3080 This directory contains the file max_user_watches, which can be
3081 used to limit the amount of kernel memory consumed by the epoll
3082 interface. For further details, see epoll(7).
3083 /proc/sys/fs/file-max
3085 This file defines a system-wide limit on the number of open
3086 files for all processes. System calls that fail when encounter‐
3087 ing this limit fail with the error ENFILE. (See also setr‐
3088 limit(2), which can be used by a process to set the per-process
3089 limit, RLIMIT_NOFILE, on the number of files it may open.) If
3090 you get lots of error messages in the kernel log about running
3091 out of file handles (open file descriptions) (look for "VFS:
3092 file-max limit <number> reached"), try increasing this value:
3093 echo 100000 > /proc/sys/fs/file-max
3095 Privileged processes (CAP_SYS_ADMIN) can override the file-max
3097 limit.
3098 /proc/sys/fs/file-nr
3100 This (read-only) file contains three numbers: the number of al‐
3101 located file handles (i.e., the number of open file descrip‐
3102 tions; see open(2)); the number of free file handles; and the
3103 maximum number of file handles (i.e., the same value as
3104 /proc/sys/fs/file-max). If the number of allocated file handles
3105 is close to the maximum, you should consider increasing the max‐
3106 imum. Before Linux 2.6, the kernel allocated file handles dy‐
3107 namically, but it didn't free them again. Instead the free file
3108 handles were kept in a list for reallocation; the "free file
3109 handles" value indicates the size of that list. A large number
3110 of free file handles indicates that there was a past peak in the
3111 usage of open file handles. Since Linux 2.6, the kernel does
3112 deallocate freed file handles, and the "free file handles" value
3113 is always zero.
3114 /proc/sys/fs/inode-max (only present until Linux 2.2)
3116 This file contains the maximum number of in-memory inodes. This
3117 value should be 3–4 times larger than the value in file-max,
3118 since stdin, stdout and network sockets also need an inode to
3119 handle them. When you regularly run out of inodes, you need to
3120 increase this value.
3121 Starting with Linux 2.4, there is no longer a static limit on
3123 the number of inodes, and this file is removed.
3124 /proc/sys/fs/inode-nr
3126 This file contains the first two values from inode-state.
3127 /proc/sys/fs/inode-state
3129 This file contains seven numbers: nr_inodes, nr_free_inodes,
3130 preshrink, and four dummy values (always zero).
3131 nr_inodes is the number of inodes the system has allocated.
3133 nr_free_inodes represents the number of free inodes.
3134 preshrink is nonzero when the nr_inodes > inode-max and the sys‐
3136 tem needs to prune the inode list instead of allocating more;
3137 since Linux 2.4, this field is a dummy value (always zero).
3138 /proc/sys/fs/inotify (since Linux 2.6.13)
3140 This directory contains files max_queued_events, max_user_in‐
3141 stances, and max_user_watches, that can be used to limit the
3142 amount of kernel memory consumed by the inotify interface. For
3143 further details, see inotify(7).
3144 /proc/sys/fs/lease-break-time
3146 This file specifies the grace period that the kernel grants to a
3147 process holding a file lease (fcntl(2)) after it has sent a sig‐
3148 nal to that process notifying it that another process is waiting
3149 to open the file. If the lease holder does not remove or down‐
3150 grade the lease within this grace period, the kernel forcibly
3151 breaks the lease.
3152 /proc/sys/fs/leases-enable
3154 This file can be used to enable or disable file leases (fc‐
3155 ntl(2)) on a system-wide basis. If this file contains the value
3156 0, leases are disabled. A nonzero value enables leases.
3157 /proc/sys/fs/mount-max (since Linux 4.9)
3159 The value in this file specifies the maximum number of mounts
3160 that may exist in a mount namespace. The default value in this
3161 file is 100,000.
3162 /proc/sys/fs/mqueue (since Linux 2.6.6)
3164 This directory contains files msg_max, msgsize_max, and
3165 queues_max, controlling the resources used by POSIX message
3166 queues. See mq_overview(7) for details.
3167 /proc/sys/fs/nr_open (since Linux 2.6.25)
3169 This file imposes a ceiling on the value to which the
3170 RLIMIT_NOFILE resource limit can be raised (see getrlimit(2)).
3171 This ceiling is enforced for both unprivileged and privileged
3172 process. The default value in this file is 1048576. (Before
3173 Linux 2.6.25, the ceiling for RLIMIT_NOFILE was hard-coded to
3174 the same value.)
3175 /proc/sys/fs/overflowgid and /proc/sys/fs/overflowuid
3177 These files allow you to change the value of the fixed UID and
3178 GID. The default is 65534. Some filesystems support only
3179 16-bit UIDs and GIDs, although in Linux UIDs and GIDs are 32
3180 bits. When one of these filesystems is mounted with writes en‐
3181 abled, any UID or GID that would exceed 65535 is translated to
3182 the overflow value before being written to disk.
3183 /proc/sys/fs/pipe-max-size (since Linux 2.6.35)
3185 See pipe(7).
3186 /proc/sys/fs/pipe-user-pages-hard (since Linux 4.5)
3188 See pipe(7).
3189 /proc/sys/fs/pipe-user-pages-soft (since Linux 4.5)
3191 See pipe(7).
3192 /proc/sys/fs/protected_fifos (since Linux 4.19)
3194 The value in this file is/can be set to one of the following:
3195 0 Writing to FIFOs is unrestricted.
3197 1 Don't allow O_CREAT open(2) on FIFOs that the caller doesn't
3199 own in world-writable sticky directories, unless the FIFO is
3200 owned by the owner of the directory.
3201 2 As for the value 1, but the restriction also applies to
3203 group-writable sticky directories.
3204 The intent of the above protections is to avoid unintentional
3206 writes to an attacker-controlled FIFO when a program expected to
3207 create a regular file.
3208 /proc/sys/fs/protected_hardlinks (since Linux 3.6)
3210 When the value in this file is 0, no restrictions are placed on
3211 the creation of hard links (i.e., this is the historical behav‐
3212 ior before Linux 3.6). When the value in this file is 1, a hard
3213 link can be created to a target file only if one of the follow‐
3214 ing conditions is true:
3215 * The calling process has the CAP_FOWNER capability in its user
3217 namespace and the file UID has a mapping in the namespace.
3218 * The filesystem UID of the process creating the link matches
3220 the owner (UID) of the target file (as described in creden‐
3221 tials(7), a process's filesystem UID is normally the same as
3222 its effective UID).
3223 * All of the following conditions are true:
3225 • the target is a regular file;
3227 • the target file does not have its set-user-ID mode bit
3229 enabled;
3230 • the target file does not have both its set-group-ID and
3232 group-executable mode bits enabled; and
3233 • the caller has permission to read and write the target
3235 file (either via the file's permissions mask or because
3236 it has suitable capabilities).
3237 The default value in this file is 0. Setting the value to 1
3239 prevents a longstanding class of security issues caused by hard-
3240 link-based time-of-check, time-of-use races, most commonly seen
3241 in world-writable directories such as /tmp. The common method
3242 of exploiting this flaw is to cross privilege boundaries when
3243 following a given hard link (i.e., a root process follows a hard
3244 link created by another user). Additionally, on systems without
3245 separated partitions, this stops unauthorized users from "pin‐
3246 ning" vulnerable set-user-ID and set-group-ID files against be‐
3247 ing upgraded by the administrator, or linking to special files.
3248 /proc/sys/fs/protected_regular (since Linux 4.19)
3250 The value in this file is/can be set to one of the following:
3251 0 Writing to regular files is unrestricted.
3253 1 Don't allow O_CREAT open(2) on regular files that the caller
3255 doesn't own in world-writable sticky directories, unless the
3256 regular file is owned by the owner of the directory.
3257 2 As for the value 1, but the restriction also applies to
3259 group-writable sticky directories.
3260 The intent of the above protections is similar to protected_fi‐
3262 fos, but allows an application to avoid writes to an attacker-
3263 controlled regular file, where the application expected to cre‐
3264 ate one.
3265 /proc/sys/fs/protected_symlinks (since Linux 3.6)
3267 When the value in this file is 0, no restrictions are placed on
3268 following symbolic links (i.e., this is the historical behavior
3269 before Linux 3.6). When the value in this file is 1, symbolic
3270 links are followed only in the following circumstances:
3271 * the filesystem UID of the process following the link matches
3273 the owner (UID) of the symbolic link (as described in creden‐
3274 tials(7), a process's filesystem UID is normally the same as
3275 its effective UID);
3276 * the link is not in a sticky world-writable directory; or
3278 * the symbolic link and its parent directory have the same
3280 owner (UID)
3281 A system call that fails to follow a symbolic link because of
3283 the above restrictions returns the error EACCES in errno.
3284 The default value in this file is 0. Setting the value to 1
3286 avoids a longstanding class of security issues based on time-of-
3287 check, time-of-use races when accessing symbolic links.
3288 /proc/sys/fs/suid_dumpable (since Linux 2.6.13)
3290 The value in this file is assigned to a process's "dumpable"
3291 flag in the circumstances described in prctl(2). In effect, the
3292 value in this file determines whether core dump files are pro‐
3293 duced for set-user-ID or otherwise protected/tainted binaries.
3294 The "dumpable" setting also affects the ownership of files in a
3295 process's /proc/[pid] directory, as described above.
3296 Three different integer values can be specified:
3298 0 (default)
3300 This provides the traditional (pre-Linux 2.6.13) behav‐
3301 ior. A core dump will not be produced for a process
3302 which has changed credentials (by calling seteuid(2),
3303 setgid(2), or similar, or by executing a set-user-ID or
3304 set-group-ID program) or whose binary does not have read
3305 permission enabled.
3306 1 ("debug")
3308 All processes dump core when possible. (Reasons why a
3309 process might nevertheless not dump core are described in
3310 core(5).) The core dump is owned by the filesystem user
3311 ID of the dumping process and no security is applied.
3312 This is intended for system debugging situations only:
3313 this mode is insecure because it allows unprivileged
3314 users to examine the memory contents of privileged pro‐
3315 cesses.
3316 2 ("suidsafe")
3318 Any binary which normally would not be dumped (see "0"
3319 above) is dumped readable by root only. This allows the
3320 user to remove the core dump file but not to read it.
3321 For security reasons core dumps in this mode will not
3322 overwrite one another or other files. This mode is ap‐
3323 propriate when administrators are attempting to debug
3324 problems in a normal environment.
3325 Additionally, since Linux 3.6, /proc/sys/kernel/core_pat‐
3327 tern must either be an absolute pathname or a pipe com‐
3328 mand, as detailed in core(5). Warnings will be written
3329 to the kernel log if core_pattern does not follow these
3330 rules, and no core dump will be produced.
3331 For details of the effect of a process's "dumpable" setting on
3333 ptrace access mode checking, see ptrace(2).
3334 /proc/sys/fs/super-max
3336 This file controls the maximum number of superblocks, and thus
3337 the maximum number of mounted filesystems the kernel can have.
3338 You need increase only super-max if you need to mount more
3339 filesystems than the current value in super-max allows you to.
3340 /proc/sys/fs/super-nr
3342 This file contains the number of filesystems currently mounted.
3343 /proc/sys/kernel
3345 This directory contains files controlling a range of kernel pa‐
3346 rameters, as described below.
3347 /proc/sys/kernel/acct
3349 This file contains three numbers: highwater, lowwater, and fre‐
3350 quency. If BSD-style process accounting is enabled, these val‐
3351 ues control its behavior. If free space on filesystem where the
3352 log lives goes below lowwater percent, accounting suspends. If
3353 free space gets above highwater percent, accounting resumes.
3354 frequency determines how often the kernel checks the amount of
3355 free space (value is in seconds). Default values are 4, 2 and
3356 30. That is, suspend accounting if 2% or less space is free;
3357 resume it if 4% or more space is free; consider information
3358 about amount of free space valid for 30 seconds.
3359 /proc/sys/kernel/auto_msgmni (Linux 2.6.27 to 3.18)
3361 From Linux 2.6.27 to 3.18, this file was used to control recom‐
3362 puting of the value in /proc/sys/kernel/msgmni upon the addition
3363 or removal of memory or upon IPC namespace creation/removal.
3364 Echoing "1" into this file enabled msgmni automatic recomputing
3365 (and triggered a recomputation of msgmni based on the current
3366 amount of available memory and number of IPC namespaces). Echo‐
3367 ing "0" disabled automatic recomputing. (Automatic recomputing
3368 was also disabled if a value was explicitly assigned to
3369 /proc/sys/kernel/msgmni.) The default value in auto_msgmni was
3370 1.
3371 Since Linux 3.19, the content of this file has no effect (be‐
3373 cause msgmni defaults to near the maximum value possible), and
3374 reads from this file always return the value "0".
3375 /proc/sys/kernel/cap_last_cap (since Linux 3.2)
3377 See capabilities(7).
3378 /proc/sys/kernel/cap-bound (from Linux 2.2 to 2.6.24)
3380 This file holds the value of the kernel capability bounding set
3381 (expressed as a signed decimal number). This set is ANDed
3382 against the capabilities permitted to a process during ex‐
3383 ecve(2). Starting with Linux 2.6.25, the system-wide capability
3384 bounding set disappeared, and was replaced by a per-thread
3385 bounding set; see capabilities(7).
3386 /proc/sys/kernel/core_pattern
3388 See core(5).
3389 /proc/sys/kernel/core_pipe_limit
3391 See core(5).
3392 /proc/sys/kernel/core_uses_pid
3394 See core(5).
3395 /proc/sys/kernel/ctrl-alt-del
3397 This file controls the handling of Ctrl-Alt-Del from the key‐
3398 board. When the value in this file is 0, Ctrl-Alt-Del is
3399 trapped and sent to the init(1) program to handle a graceful
3400 restart. When the value is greater than zero, Linux's reaction
3401 to a Vulcan Nerve Pinch (tm) will be an immediate reboot, with‐
3402 out even syncing its dirty buffers. Note: when a program (like
3403 dosemu) has the keyboard in "raw" mode, the ctrl-alt-del is in‐
3404 tercepted by the program before it ever reaches the kernel tty
3405 layer, and it's up to the program to decide what to do with it.
3406 /proc/sys/kernel/dmesg_restrict (since Linux 2.6.37)
3408 The value in this file determines who can see kernel syslog con‐
3409 tents. A value of 0 in this file imposes no restrictions. If
3410 the value is 1, only privileged users can read the kernel sys‐
3411 log. (See syslog(2) for more details.) Since Linux 3.4, only
3412 users with the CAP_SYS_ADMIN capability may change the value in
3413 this file.
3414 /proc/sys/kernel/domainname and /proc/sys/kernel/hostname
3416 can be used to set the NIS/YP domainname and the hostname of
3417 your box in exactly the same way as the commands domainname(1)
3418 and hostname(1), that is:
3419 # echo 'darkstar' > /proc/sys/kernel/hostname
3421 # echo 'mydomain' > /proc/sys/kernel/domainname
3422 has the same effect as
3424 # hostname 'darkstar'
3426 # domainname 'mydomain'
3427 Note, however, that the classic darkstar.frop.org has the host‐
3429 name "darkstar" and DNS (Internet Domain Name Server) domainname
3430 "frop.org", not to be confused with the NIS (Network Information
3431 Service) or YP (Yellow Pages) domainname. These two domain
3432 names are in general different. For a detailed discussion see
3433 the hostname(1) man page.
3434 /proc/sys/kernel/hotplug
3436 This file contains the pathname for the hotplug policy agent.
3437 The default value in this file is /sbin/hotplug.
3438 /proc/sys/kernel/htab-reclaim (before Linux 2.4.9.2)
3440 (PowerPC only) If this file is set to a nonzero value, the Pow‐
3441 erPC htab (see kernel file Documentation/powerpc/ppc_htab.txt)
3442 is pruned each time the system hits the idle loop.
3443 /proc/sys/kernel/keys/*
3445 This directory contains various files that define parameters and
3446 limits for the key-management facility. These files are de‐
3447 scribed in keyrings(7).
3448 /proc/sys/kernel/kptr_restrict (since Linux 2.6.38)
3450 The value in this file determines whether kernel addresses are
3451 exposed via /proc files and other interfaces. A value of 0 in
3452 this file imposes no restrictions. If the value is 1, kernel
3453 pointers printed using the %pK format specifier will be replaced
3454 with zeros unless the user has the CAP_SYSLOG capability. If
3455 the value is 2, kernel pointers printed using the %pK format
3456 specifier will be replaced with zeros regardless of the user's
3457 capabilities. The initial default value for this file was 1,
3458 but the default was changed to 0 in Linux 2.6.39. Since Linux
3459 3.4, only users with the CAP_SYS_ADMIN capability can change the
3460 value in this file.
3461 /proc/sys/kernel/l2cr
3463 (PowerPC only) This file contains a flag that controls the L2
3464 cache of G3 processor boards. If 0, the cache is disabled. En‐
3465 abled if nonzero.
3466 /proc/sys/kernel/modprobe
3468 This file contains the pathname for the kernel module loader.
3469 The default value is /sbin/modprobe. The file is present only
3470 if the kernel is built with the CONFIG_MODULES (CONFIG_KMOD in
3471 Linux 2.6.26 and earlier) option enabled. It is described by
3472 the Linux kernel source file Documentation/kmod.txt (present
3473 only in kernel 2.4 and earlier).
3474 /proc/sys/kernel/modules_disabled (since Linux 2.6.31)
3476 A toggle value indicating if modules are allowed to be loaded in
3477 an otherwise modular kernel. This toggle defaults to off (0),
3478 but can be set true (1). Once true, modules can be neither
3479 loaded nor unloaded, and the toggle cannot be set back to false.
3480 The file is present only if the kernel is built with the CON‐
3481 FIG_MODULES option enabled.
3482 /proc/sys/kernel/msgmax (since Linux 2.2)
3484 This file defines a system-wide limit specifying the maximum
3485 number of bytes in a single message written on a System V mes‐
3486 sage queue.
3487 /proc/sys/kernel/msgmni (since Linux 2.4)
3489 This file defines the system-wide limit on the number of message
3490 queue identifiers. See also /proc/sys/kernel/auto_msgmni.
3491 /proc/sys/kernel/msgmnb (since Linux 2.2)
3493 This file defines a system-wide parameter used to initialize the
3494 msg_qbytes setting for subsequently created message queues. The
3495 msg_qbytes setting specifies the maximum number of bytes that
3496 may be written to the message queue.
3497 /proc/sys/kernel/ngroups_max (since Linux 2.6.4)
3499 This is a read-only file that displays the upper limit on the
3500 number of a process's group memberships.
3501 /proc/sys/kernel/ns_last_pid (since Linux 3.3)
3503 See pid_namespaces(7).
3504 /proc/sys/kernel/ostype and /proc/sys/kernel/osrelease
3506 These files give substrings of /proc/version.
3507 /proc/sys/kernel/overflowgid and /proc/sys/kernel/overflowuid
3509 These files duplicate the files /proc/sys/fs/overflowgid and
3510 /proc/sys/fs/overflowuid.
3511 /proc/sys/kernel/panic
3513 This file gives read/write access to the kernel variable
3514 panic_timeout. If this is zero, the kernel will loop on a
3515 panic; if nonzero, it indicates that the kernel should autore‐
3516 boot after this number of seconds. When you use the software
3517 watchdog device driver, the recommended setting is 60.
3518 /proc/sys/kernel/panic_on_oops (since Linux 2.5.68)
3520 This file controls the kernel's behavior when an oops or BUG is
3521 encountered. If this file contains 0, then the system tries to
3522 continue operation. If it contains 1, then the system delays a
3523 few seconds (to give klogd time to record the oops output) and
3524 then panics. If the /proc/sys/kernel/panic file is also non‐
3525 zero, then the machine will be rebooted.
3526 /proc/sys/kernel/pid_max (since Linux 2.5.34)
3528 This file specifies the value at which PIDs wrap around (i.e.,
3529 the value in this file is one greater than the maximum PID).
3530 PIDs greater than this value are not allocated; thus, the value
3531 in this file also acts as a system-wide limit on the total num‐
3532 ber of processes and threads. The default value for this file,
3533 32768, results in the same range of PIDs as on earlier kernels.
3534 On 32-bit platforms, 32768 is the maximum value for pid_max. On
3535 64-bit systems, pid_max can be set to any value up to 2^22
3536 (PID_MAX_LIMIT, approximately 4 million).
3537 /proc/sys/kernel/powersave-nap (PowerPC only)
3539 This file contains a flag. If set, Linux-PPC will use the "nap"
3540 mode of powersaving, otherwise the "doze" mode will be used.
3541 /proc/sys/kernel/printk
3543 See syslog(2).
3544 /proc/sys/kernel/pty (since Linux 2.6.4)
3546 This directory contains two files relating to the number of UNIX
3547 98 pseudoterminals (see pts(4)) on the system.
3548 /proc/sys/kernel/pty/max
3550 This file defines the maximum number of pseudoterminals.
3551 /proc/sys/kernel/pty/nr
3553 This read-only file indicates how many pseudoterminals are cur‐
3554 rently in use.
3555 /proc/sys/kernel/random
3557 This directory contains various parameters controlling the oper‐
3558 ation of the file /dev/random. See random(4) for further infor‐
3559 mation.
3560 /proc/sys/kernel/random/uuid (since Linux 2.4)
3562 Each read from this read-only file returns a randomly generated
3563 128-bit UUID, as a string in the standard UUID format.
3564 /proc/sys/kernel/randomize_va_space (since Linux 2.6.12)
3566 Select the address space layout randomization (ASLR) policy for
3567 the system (on architectures that support ASLR). Three values
3568 are supported for this file:
3569 0 Turn ASLR off. This is the default for architectures that
3571 don't support ASLR, and when the kernel is booted with the
3572 norandmaps parameter.
3573 1 Make the addresses of mmap(2) allocations, the stack, and the
3575 VDSO page randomized. Among other things, this means that
3576 shared libraries will be loaded at randomized addresses. The
3577 text segment of PIE-linked binaries will also be loaded at a
3578 randomized address. This value is the default if the kernel
3579 was configured with CONFIG_COMPAT_BRK.
3580 2 (Since Linux 2.6.25) Also support heap randomization. This
3582 value is the default if the kernel was not configured with
3583 CONFIG_COMPAT_BRK.
3584 /proc/sys/kernel/real-root-dev
3586 This file is documented in the Linux kernel source file Documen‐
3587 tation/admin-guide/initrd.rst (or Documentation/initrd.txt be‐
3588 fore Linux 4.10).
3589 /proc/sys/kernel/reboot-cmd (Sparc only)
3591 This file seems to be a way to give an argument to the SPARC
3592 ROM/Flash boot loader. Maybe to tell it what to do after re‐
3593 booting?
3594 /proc/sys/kernel/rtsig-max
3596 (Only in kernels up to and including 2.6.7; see setrlimit(2))
3597 This file can be used to tune the maximum number of POSIX real-
3598 time (queued) signals that can be outstanding in the system.
3599 /proc/sys/kernel/rtsig-nr
3601 (Only in kernels up to and including 2.6.7.) This file shows
3602 the number of POSIX real-time signals currently queued.
3603 /proc/[pid]/sched_autogroup_enabled (since Linux 2.6.38)
3605 See sched(7).
3606 /proc/sys/kernel/sched_child_runs_first (since Linux 2.6.23)
3608 If this file contains the value zero, then, after a fork(2), the
3609 parent is first scheduled on the CPU. If the file contains a
3610 nonzero value, then the child is scheduled first on the CPU.
3611 (Of course, on a multiprocessor system, the parent and the child
3612 might both immediately be scheduled on a CPU.)
3613 /proc/sys/kernel/sched_rr_timeslice_ms (since Linux 3.9)
3615 See sched_rr_get_interval(2).
3616 /proc/sys/kernel/sched_rt_period_us (since Linux 2.6.25)
3618 See sched(7).
3619 /proc/sys/kernel/sched_rt_runtime_us (since Linux 2.6.25)
3621 See sched(7).
3622 /proc/sys/kernel/seccomp (since Linux 4.14)
3624 This directory provides additional seccomp information and con‐
3625 figuration. See seccomp(2) for further details.
3626 /proc/sys/kernel/sem (since Linux 2.4)
3628 This file contains 4 numbers defining limits for System V IPC
3629 semaphores. These fields are, in order:
3630 SEMMSL The maximum semaphores per semaphore set.
3632 SEMMNS A system-wide limit on the number of semaphores in all
3634 semaphore sets.
3635 SEMOPM The maximum number of operations that may be specified
3637 in a semop(2) call.
3638 SEMMNI A system-wide limit on the maximum number of semaphore
3640 identifiers.
3641 /proc/sys/kernel/sg-big-buff
3643 This file shows the size of the generic SCSI device (sg) buffer.
3644 You can't tune it just yet, but you could change it at compile
3645 time by editing include/scsi/sg.h and changing the value of
3646 SG_BIG_BUFF. However, there shouldn't be any reason to change
3647 this value.
3648 /proc/sys/kernel/shm_rmid_forced (since Linux 3.1)
3650 If this file is set to 1, all System V shared memory segments
3651 will be marked for destruction as soon as the number of attached
3652 processes falls to zero; in other words, it is no longer possi‐
3653 ble to create shared memory segments that exist independently of
3654 any attached process.
3655 The effect is as though a shmctl(2) IPC_RMID is performed on all
3657 existing segments as well as all segments created in the future
3658 (until this file is reset to 0). Note that existing segments
3659 that are attached to no process will be immediately destroyed
3660 when this file is set to 1. Setting this option will also de‐
3661 stroy segments that were created, but never attached, upon ter‐
3662 mination of the process that created the segment with shmget(2).
3663 Setting this file to 1 provides a way of ensuring that all Sys‐
3665 tem V shared memory segments are counted against the resource
3666 usage and resource limits (see the description of RLIMIT_AS in
3667 getrlimit(2)) of at least one process.
3668 Because setting this file to 1 produces behavior that is non‐
3670 standard and could also break existing applications, the default
3671 value in this file is 0. Set this file to 1 only if you have a
3672 good understanding of the semantics of the applications using
3673 System V shared memory on your system.
3674 /proc/sys/kernel/shmall (since Linux 2.2)
3676 This file contains the system-wide limit on the total number of
3677 pages of System V shared memory.
3678 /proc/sys/kernel/shmmax (since Linux 2.2)
3680 This file can be used to query and set the run-time limit on the
3681 maximum (System V IPC) shared memory segment size that can be
3682 created. Shared memory segments up to 1 GB are now supported in
3683 the kernel. This value defaults to SHMMAX.
3684 /proc/sys/kernel/shmmni (since Linux 2.4)
3686 This file specifies the system-wide maximum number of System V
3687 shared memory segments that can be created.
3688 /proc/sys/kernel/sysctl_writes_strict (since Linux 3.16)
3690 The value in this file determines how the file offset affects
3691 the behavior of updating entries in files under /proc/sys. The
3692 file has three possible values:
3693 -1 This provides legacy handling, with no printk warnings.
3695 Each write(2) must fully contain the value to be written,
3696 and multiple writes on the same file descriptor will over‐
3697 write the entire value, regardless of the file position.
3698 0 (default) This provides the same behavior as for -1, but
3700 printk warnings are written for processes that perform
3701 writes when the file offset is not 0.
3702 1 Respect the file offset when writing strings into /proc/sys
3704 files. Multiple writes will append to the value buffer.
3705 Anything written beyond the maximum length of the value buf‐
3706 fer will be ignored. Writes to numeric /proc/sys entries
3707 must always be at file offset 0 and the value must be fully
3708 contained in the buffer provided to write(2).
3709 /proc/sys/kernel/sysrq
3711 This file controls the functions allowed to be invoked by the
3712 SysRq key. By default, the file contains 1 meaning that every
3713 possible SysRq request is allowed (in older kernel versions,
3714 SysRq was disabled by default, and you were required to specifi‐
3715 cally enable it at run-time, but this is not the case any more).
3716 Possible values in this file are:
3717 0 Disable sysrq completely
3719 1 Enable all functions of sysrq
3721 > 1 Bit mask of allowed sysrq functions, as follows:
3723 2 Enable control of console logging level
3724 4 Enable control of keyboard (SAK, unraw)
3725 8 Enable debugging dumps of processes etc.
3726 16 Enable sync command
3727 32 Enable remount read-only
3728 64 Enable signaling of processes (term, kill, oom-kill)
3729 128 Allow reboot/poweroff
3730 256 Allow nicing of all real-time tasks
3731 This file is present only if the CONFIG_MAGIC_SYSRQ kernel con‐
3733 figuration option is enabled. For further details see the Linux
3734 kernel source file Documentation/admin-guide/sysrq.rst (or Docu‐
3735 mentation/sysrq.txt before Linux 4.10).
3736 /proc/sys/kernel/version
3738 This file contains a string such as:
3739 #5 Wed Feb 25 21:49:24 MET 1998
3741 The "#5" means that this is the fifth kernel built from this
3743 source base and the date following it indicates the time the
3744 kernel was built.
3745 /proc/sys/kernel/threads-max (since Linux 2.3.11)
3747 This file specifies the system-wide limit on the number of
3748 threads (tasks) that can be created on the system.
3749 Since Linux 4.1, the value that can be written to threads-max is
3751 bounded. The minimum value that can be written is 20. The max‐
3752 imum value that can be written is given by the constant FU‐
3753 TEX_TID_MASK [22m(0x3fffffff). If a value outside of this range is
3754 written to threads-max, the error EINVAL occurs.
3755 The value written is checked against the available RAM pages.
3757 If the thread structures would occupy too much (more than 1/8th)
3758 of the available RAM pages, threads-max is reduced accordingly.
3759 /proc/sys/kernel/yama/ptrace_scope (since Linux 3.5)
3761 See ptrace(2).
3762 /proc/sys/kernel/zero-paged (PowerPC only)
3764 This file contains a flag. When enabled (nonzero), Linux-PPC
3765 will pre-zero pages in the idle loop, possibly speeding up
3766 get_free_pages.
3767 /proc/sys/net
3769 This directory contains networking stuff. Explanations for some
3770 of the files under this directory can be found in tcp(7) and
3771 ip(7).
3772 /proc/sys/net/core/bpf_jit_enable
3774 See bpf(2).
3775 /proc/sys/net/core/somaxconn
3777 This file defines a ceiling value for the backlog argument of
3778 listen(2); see the listen(2) manual page for details.
3779 /proc/sys/proc
3781 This directory may be empty.
3782 /proc/sys/sunrpc
3784 This directory supports Sun remote procedure call for network
3785 filesystem (NFS). On some systems, it is not present.
3786 /proc/sys/user (since Linux 4.9)
3788 See namespaces(7).
3789 /proc/sys/vm
3791 This directory contains files for memory management tuning, buf‐
3792 fer and cache management.
3793 /proc/sys/vm/admin_reserve_kbytes (since Linux 3.10)
3795 This file defines the amount of free memory (in KiB) on the sys‐
3796 tem that should be reserved for users with the capability
3797 CAP_SYS_ADMIN.
3798 The default value in this file is the minimum of [3% of free
3800 pages, 8MiB] expressed as KiB. The default is intended to pro‐
3801 vide enough for the superuser to log in and kill a process, if
3802 necessary, under the default overcommit 'guess' mode (i.e., 0 in
3803 /proc/sys/vm/overcommit_memory).
3804 Systems running in "overcommit never" mode (i.e., 2 in
3806 /proc/sys/vm/overcommit_memory) should increase the value in
3807 this file to account for the full virtual memory size of the
3808 programs used to recover (e.g., login(1) ssh(1), and top(1))
3809 Otherwise, the superuser may not be able to log in to recover
3810 the system. For example, on x86-64 a suitable value is 131072
3811 (128MiB reserved).
3812 Changing the value in this file takes effect whenever an appli‐
3814 cation requests memory.
3815 /proc/sys/vm/compact_memory (since Linux 2.6.35)
3817 When 1 is written to this file, all zones are compacted such
3818 that free memory is available in contiguous blocks where possi‐
3819 ble. The effect of this action can be seen by examining
3820 /proc/buddyinfo.
3821 Present only if the kernel was configured with CONFIG_COM‐
3823 PACTION.
3824 /proc/sys/vm/drop_caches (since Linux 2.6.16)
3826 Writing to this file causes the kernel to drop clean caches,
3827 dentries, and inodes from memory, causing that memory to become
3828 free. This can be useful for memory management testing and per‐
3829 forming reproducible filesystem benchmarks. Because writing to
3830 this file causes the benefits of caching to be lost, it can de‐
3831 grade overall system performance.
3832 To free pagecache, use:
3834 echo 1 > /proc/sys/vm/drop_caches
3836 To free dentries and inodes, use:
3838 echo 2 > /proc/sys/vm/drop_caches
3840 To free pagecache, dentries and inodes, use:
3842 echo 3 > /proc/sys/vm/drop_caches
3844 Because writing to this file is a nondestructive operation and
3846 dirty objects are not freeable, the user should run sync(1)
3847 first.
3848 /proc/sys/vm/legacy_va_layout (since Linux 2.6.9)
3850 If nonzero, this disables the new 32-bit memory-mapping layout;
3851 the kernel will use the legacy (2.4) layout for all processes.
3852 /proc/sys/vm/memory_failure_early_kill (since Linux 2.6.32)
3854 Control how to kill processes when an uncorrected memory error
3855 (typically a 2-bit error in a memory module) that cannot be han‐
3856 dled by the kernel is detected in the background by hardware.
3857 In some cases (like the page still having a valid copy on disk),
3858 the kernel will handle the failure transparently without affect‐
3859 ing any applications. But if there is no other up-to-date copy
3860 of the data, it will kill processes to prevent any data corrup‐
3861 tions from propagating.
3862 The file has one of the following values:
3864 1: Kill all processes that have the corrupted-and-not-reload‐
3866 able page mapped as soon as the corruption is detected.
3867 Note that this is not supported for a few types of pages,
3868 such as kernel internally allocated data or the swap cache,
3869 but works for the majority of user pages.
3870 0: Unmap the corrupted page from all processes and kill a
3872 process only if it tries to access the page.
3873 The kill is performed using a SIGBUS signal with si_code set to
3875 BUS_MCEERR_AO. Processes can handle this if they want to; see
3876 sigaction(2) for more details.
3877 This feature is active only on architectures/platforms with ad‐
3879 vanced machine check handling and depends on the hardware capa‐
3880 bilities.
3881 Applications can override the memory_failure_early_kill setting
3883 individually with the prctl(2) PR_MCE_KILL operation.
3884 Present only if the kernel was configured with CONFIG_MEM‐
3886 ORY_FAILURE.
3887 /proc/sys/vm/memory_failure_recovery (since Linux 2.6.32)
3889 Enable memory failure recovery (when supported by the platform).
3890 1: Attempt recovery.
3892 0: Always panic on a memory failure.
3894 Present only if the kernel was configured with CONFIG_MEM‐
3896 ORY_FAILURE.
3897 /proc/sys/vm/oom_dump_tasks (since Linux 2.6.25)
3899 Enables a system-wide task dump (excluding kernel threads) to be
3900 produced when the kernel performs an OOM-killing. The dump in‐
3901 cludes the following information for each task (thread,
3902 process): thread ID, real user ID, thread group ID (process ID),
3903 virtual memory size, resident set size, the CPU that the task is
3904 scheduled on, oom_adj score (see the description of
3905 /proc/[pid]/oom_adj), and command name. This is helpful to de‐
3906 termine why the OOM-killer was invoked and to identify the rogue
3907 task that caused it.
3908 If this contains the value zero, this information is suppressed.
3910 On very large systems with thousands of tasks, it may not be
3911 feasible to dump the memory state information for each one.
3912 Such systems should not be forced to incur a performance penalty
3913 in OOM situations when the information may not be desired.
3914 If this is set to nonzero, this information is shown whenever
3916 the OOM-killer actually kills a memory-hogging task.
3917 The default value is 0.
3919 /proc/sys/vm/oom_kill_allocating_task (since Linux 2.6.24)
3921 This enables or disables killing the OOM-triggering task in out-
3922 of-memory situations.
3923 If this is set to zero, the OOM-killer will scan through the en‐
3925 tire tasklist and select a task based on heuristics to kill.
3926 This normally selects a rogue memory-hogging task that frees up
3927 a large amount of memory when killed.
3928 If this is set to nonzero, the OOM-killer simply kills the task
3930 that triggered the out-of-memory condition. This avoids a pos‐
3931 sibly expensive tasklist scan.
3932 If /proc/sys/vm/panic_on_oom is nonzero, it takes precedence
3934 over whatever value is used in /proc/sys/vm/oom_kill_allocat‐
3935 ing_task.
3936 The default value is 0.
3938 /proc/sys/vm/overcommit_kbytes (since Linux 3.14)
3940 This writable file provides an alternative to /proc/sys/vm/over‐
3941 commit_ratio for controlling the CommitLimit when
3942 /proc/sys/vm/overcommit_memory has the value 2. It allows the
3943 amount of memory overcommitting to be specified as an absolute
3944 value (in kB), rather than as a percentage, as is done with
3945 overcommit_ratio. This allows for finer-grained control of Com‐
3946 mitLimit on systems with extremely large memory sizes.
3947 Only one of overcommit_kbytes or overcommit_ratio can have an
3949 effect: if overcommit_kbytes has a nonzero value, then it is
3950 used to calculate CommitLimit, otherwise overcommit_ratio is
3951 used. Writing a value to either of these files causes the value
3952 in the other file to be set to zero.
3953 /proc/sys/vm/overcommit_memory
3955 This file contains the kernel virtual memory accounting mode.
3956 Values are:
3957 0: heuristic overcommit (this is the default)
3959 1: always overcommit, never check
3960 2: always check, never overcommit
3961 In mode 0, calls of mmap(2) with MAP_NORESERVE are not checked,
3963 and the default check is very weak, leading to the risk of get‐
3964 ting a process "OOM-killed".
3965 In mode 1, the kernel pretends there is always enough memory,
3967 until memory actually runs out. One use case for this mode is
3968 scientific computing applications that employ large sparse ar‐
3969 rays. In Linux kernel versions before 2.6.0, any nonzero value
3970 implies mode 1.
3971 In mode 2 (available since Linux 2.6), the total virtual address
3973 space that can be allocated (CommitLimit in /proc/meminfo) is
3974 calculated as
3975 CommitLimit = (total_RAM - total_huge_TLB) *
3977 overcommit_ratio / 100 + total_swap
3978 where:
3980 * total_RAM is the total amount of RAM on the system;
3982 * total_huge_TLB is the amount of memory set aside for
3984 huge pages;
3985 * overcommit_ratio is the value in /proc/sys/vm/overcom‐
3987 mit_ratio; and
3988 * total_swap is the amount of swap space.
3990 For example, on a system with 16 GB of physical RAM, 16 GB of
3992 swap, no space dedicated to huge pages, and an overcommit_ratio
3993 of 50, this formula yields a CommitLimit of 24 GB.
3994 Since Linux 3.14, if the value in /proc/sys/vm/overcommit_kbytes
3996 is nonzero, then CommitLimit is instead calculated as:
3997 CommitLimit = overcommit_kbytes + total_swap
3999 See also the description of /proc/sys/vm/admin_reserve_kbytes
4001 and /proc/sys/vm/user_reserve_kbytes.
4002 /proc/sys/vm/overcommit_ratio (since Linux 2.6.0)
4004 This writable file defines a percentage by which memory can be
4005 overcommitted. The default value in the file is 50. See the
4006 description of /proc/sys/vm/overcommit_memory.
4007 /proc/sys/vm/panic_on_oom (since Linux 2.6.18)
4009 This enables or disables a kernel panic in an out-of-memory sit‐
4010 uation.
4011 If this file is set to the value 0, the kernel's OOM-killer will
4013 kill some rogue process. Usually, the OOM-killer is able to
4014 kill a rogue process and the system will survive.
4015 If this file is set to the value 1, then the kernel normally
4017 panics when out-of-memory happens. However, if a process limits
4018 allocations to certain nodes using memory policies (mbind(2)
4019 MPOL_BIND) or cpusets (cpuset(7)) and those nodes reach memory
4020 exhaustion status, one process may be killed by the OOM-killer.
4021 No panic occurs in this case: because other nodes' memory may be
4022 free, this means the system as a whole may not have reached an
4023 out-of-memory situation yet.
4024 If this file is set to the value 2, the kernel always panics
4026 when an out-of-memory condition occurs.
4027 The default value is 0. 1 and 2 are for failover of clustering.
4029 Select either according to your policy of failover.
4030 /proc/sys/vm/swappiness
4032 The value in this file controls how aggressively the kernel will
4033 swap memory pages. Higher values increase aggressiveness, lower
4034 values decrease aggressiveness. The default value is 60.
4035 /proc/sys/vm/user_reserve_kbytes (since Linux 3.10)
4037 Specifies an amount of memory (in KiB) to reserve for user pro‐
4038 cesses. This is intended to prevent a user from starting a sin‐
4039 gle memory hogging process, such that they cannot recover (kill
4040 the hog). The value in this file has an effect only when
4041 /proc/sys/vm/overcommit_memory is set to 2 ("overcommit never"
4042 mode). In this case, the system reserves an amount of memory
4043 that is the minimum of [3% of current process size, user_re‐
4044 serve_kbytes].
4045 The default value in this file is the minimum of [3% of free
4047 pages, 128MiB] expressed as KiB.
4048 If the value in this file is set to zero, then a user will be
4050 allowed to allocate all free memory with a single process (minus
4051 the amount reserved by /proc/sys/vm/admin_reserve_kbytes). Any
4052 subsequent attempts to execute a command will result in "fork:
4053 Cannot allocate memory".
4054 Changing the value in this file takes effect whenever an appli‐
4056 cation requests memory.
4057 /proc/sys/vm/unprivileged_userfaultfd (since Linux 5.2)
4059 This (writable) file exposes a flag that controls whether un‐
4060 privileged processes are allowed to employ userfaultfd(2). If
4061 this file has the value 1, then unprivileged processes may use
4062 userfaultfd(2). If this file has the value 0, then only pro‐
4063 cesses that have the CAP_SYS_PTRACE capability may employ user‐
4064 faultfd(2). The default value in this file is 1.
4065 /proc/sysrq-trigger (since Linux 2.4.21)
4067 Writing a character to this file triggers the same SysRq func‐
4068 tion as typing ALT-SysRq-<character> (see the description of
4069 /proc/sys/kernel/sysrq). This file is normally writable only by
4070 root. For further details see the Linux kernel source file Doc‐
4071 umentation/admin-guide/sysrq.rst (or Documentation/sysrq.txt be‐
4072 fore Linux 4.10).
4073 /proc/sysvipc
4075 Subdirectory containing the pseudo-files msg, sem and shm.
4076 These files list the System V Interprocess Communication (IPC)
4077 objects (respectively: message queues, semaphores, and shared
4078 memory) that currently exist on the system, providing similar
4079 information to that available via ipcs(1). These files have
4080 headers and are formatted (one IPC object per line) for easy un‐
4081 derstanding. sysvipc(7) provides further background on the in‐
4082 formation shown by these files.
4083 /proc/thread-self (since Linux 3.17)
4085 This directory refers to the thread accessing the /proc filesys‐
4086 tem, and is identical to the /proc/self/task/[tid] directory
4087 named by the process thread ID ([tid]) of the same thread.
4088 /proc/timer_list (since Linux 2.6.21)
4090 This read-only file exposes a list of all currently pending
4091 (high-resolution) timers, all clock-event sources, and their pa‐
4092 rameters in a human-readable form.
4093 /proc/timer_stats (from Linux 2.6.21 until Linux 4.10)
4095 This is a debugging facility to make timer (ab)use in a Linux
4096 system visible to kernel and user-space developers. It can be
4097 used by kernel and user-space developers to verify that their
4098 code does not make undue use of timers. The goal is to avoid
4099 unnecessary wakeups, thereby optimizing power consumption.
4100 If enabled in the kernel (CONFIG_TIMER_STATS), but not used, it
4102 has almost zero run-time overhead and a relatively small data-
4103 structure overhead. Even if collection is enabled at run time,
4104 overhead is low: all the locking is per-CPU and lookup is
4105 hashed.
4106 The /proc/timer_stats file is used both to control sampling fa‐
4108 cility and to read out the sampled information.
4109 The timer_stats functionality is inactive on bootup. A sampling
4111 period can be started using the following command:
4112 # echo 1 > /proc/timer_stats
4114 The following command stops a sampling period:
4116 # echo 0 > /proc/timer_stats
4118 The statistics can be retrieved by:
4120 $ cat /proc/timer_stats
4122 While sampling is enabled, each readout from /proc/timer_stats
4124 will see newly updated statistics. Once sampling is disabled,
4125 the sampled information is kept until a new sample period is
4126 started. This allows multiple readouts.
4127 Sample output from /proc/timer_stats:
4129 $ cat /proc/timer_stats
4131 Timer Stats Version: v0.3
4132 Sample period: 1.764 s
4133 Collection: active
4134 255, 0 swapper/3 hrtimer_start_range_ns (tick_sched_timer)
4135 71, 0 swapper/1 hrtimer_start_range_ns (tick_sched_timer)
4136 58, 0 swapper/0 hrtimer_start_range_ns (tick_sched_timer)
4137 4, 1694 gnome-shell mod_delayed_work_on (delayed_work_timer_fn)
4138 17, 7 rcu_sched rcu_gp_kthread (process_timeout)
4139 ...
4140 1, 4911 kworker/u16:0 mod_delayed_work_on (delayed_work_timer_fn)
4141 1D, 2522 kworker/0:0 queue_delayed_work_on (delayed_work_timer_fn)
4142 1029 total events, 583.333 events/sec
4143 The output columns are:
4145 * a count of the number of events, optionally (since Linux
4147 2.6.23) followed by the letter 'D' if this is a deferrable
4148 timer;
4149 * the PID of the process that initialized the timer;
4151 * the name of the process that initialized the timer;
4153 * the function where the timer was initialized; and
4155 * (in parentheses) the callback function that is associated
4157 with the timer.
4158 During the Linux 4.11 development cycle, this file was removed
4160 because of security concerns, as it exposes information across
4161 namespaces. Furthermore, it is possible to obtain the same in‐
4162 formation via in-kernel tracing facilities such as ftrace.
4163 /proc/tty
4165 Subdirectory containing the pseudo-files and subdirectories for
4166 tty drivers and line disciplines.
4167 /proc/uptime
4169 This file contains two numbers (values in seconds): the uptime
4170 of the system (including time spent in suspend) and the amount
4171 of time spent in the idle process.
4172 /proc/version
4174 This string identifies the kernel version that is currently run‐
4175 ning. It includes the contents of /proc/sys/kernel/ostype,
4176 /proc/sys/kernel/osrelease, and /proc/sys/kernel/version. For
4177 example:
4178 Linux version 1.0.9 (quinlan@phaze) #1 Sat May 14 01:51:54 EDT 1994
4180 /proc/vmstat (since Linux 2.6.0)
4182 This file displays various virtual memory statistics. Each line
4183 of this file contains a single name-value pair, delimited by
4184 white space. Some lines are present only if the kernel was con‐
4185 figured with suitable options. (In some cases, the options re‐
4186 quired for particular files have changed across kernel versions,
4187 so they are not listed here. Details can be found by consulting
4188 the kernel source code.) The following fields may be present:
4189 nr_free_pages (since Linux 2.6.31)
4191 nr_alloc_batch (since Linux 3.12)
4193 nr_inactive_anon (since Linux 2.6.28)
4195 nr_active_anon (since Linux 2.6.28)
4197 nr_inactive_file (since Linux 2.6.28)
4199 nr_active_file (since Linux 2.6.28)
4201 nr_unevictable (since Linux 2.6.28)
4203 nr_mlock (since Linux 2.6.28)
4205 nr_anon_pages (since Linux 2.6.18)
4207 nr_mapped (since Linux 2.6.0)
4209 nr_file_pages (since Linux 2.6.18)
4211 nr_dirty (since Linux 2.6.0)
4213 nr_writeback (since Linux 2.6.0)
4215 nr_slab_reclaimable (since Linux 2.6.19)
4217 nr_slab_unreclaimable (since Linux 2.6.19)
4219 nr_page_table_pages (since Linux 2.6.0)
4221 nr_kernel_stack (since Linux 2.6.32)
4223 Amount of memory allocated to kernel stacks.
4224 nr_unstable (since Linux 2.6.0)
4226 nr_bounce (since Linux 2.6.12)
4228 nr_vmscan_write (since Linux 2.6.19)
4230 nr_vmscan_immediate_reclaim (since Linux 3.2)
4232 nr_writeback_temp (since Linux 2.6.26)
4234 nr_isolated_anon (since Linux 2.6.32)
4236 nr_isolated_file (since Linux 2.6.32)
4238 nr_shmem (since Linux 2.6.32)
4240 Pages used by shmem and tmpfs(5).
4241 nr_dirtied (since Linux 2.6.37)
4243 nr_written (since Linux 2.6.37)
4245 nr_pages_scanned (since Linux 3.17)
4247 numa_hit (since Linux 2.6.18)
4249 numa_miss (since Linux 2.6.18)
4251 numa_foreign (since Linux 2.6.18)
4253 numa_interleave (since Linux 2.6.18)
4255 numa_local (since Linux 2.6.18)
4257 numa_other (since Linux 2.6.18)
4259 workingset_refault (since Linux 3.15)
4261 workingset_activate (since Linux 3.15)
4263 workingset_nodereclaim (since Linux 3.15)
4265 nr_anon_transparent_hugepages (since Linux 2.6.38)
4267 nr_free_cma (since Linux 3.7)
4269 Number of free CMA (Contiguous Memory Allocator) pages.
4270 nr_dirty_threshold (since Linux 2.6.37)
4272 nr_dirty_background_threshold (since Linux 2.6.37)
4274 pgpgin (since Linux 2.6.0)
4276 pgpgout (since Linux 2.6.0)
4278 pswpin (since Linux 2.6.0)
4280 pswpout (since Linux 2.6.0)
4282 pgalloc_dma (since Linux 2.6.5)
4284 pgalloc_dma32 (since Linux 2.6.16)
4286 pgalloc_normal (since Linux 2.6.5)
4288 pgalloc_high (since Linux 2.6.5)
4290 pgalloc_movable (since Linux 2.6.23)
4292 pgfree (since Linux 2.6.0)
4294 pgactivate (since Linux 2.6.0)
4296 pgdeactivate (since Linux 2.6.0)
4298 pgfault (since Linux 2.6.0)
4300 pgmajfault (since Linux 2.6.0)
4302 pgrefill_dma (since Linux 2.6.5)
4304 pgrefill_dma32 (since Linux 2.6.16)
4306 pgrefill_normal (since Linux 2.6.5)
4308 pgrefill_high (since Linux 2.6.5)
4310 pgrefill_movable (since Linux 2.6.23)
4312 pgsteal_kswapd_dma (since Linux 3.4)
4314 pgsteal_kswapd_dma32 (since Linux 3.4)
4316 pgsteal_kswapd_normal (since Linux 3.4)
4318 pgsteal_kswapd_high (since Linux 3.4)
4320 pgsteal_kswapd_movable (since Linux 3.4)
4322 pgsteal_direct_dma
4324 pgsteal_direct_dma32 (since Linux 3.4)
4326 pgsteal_direct_normal (since Linux 3.4)
4328 pgsteal_direct_high (since Linux 3.4)
4330 pgsteal_direct_movable (since Linux 2.6.23)
4332 pgscan_kswapd_dma
4334 pgscan_kswapd_dma32 (since Linux 2.6.16)
4336 pgscan_kswapd_normal (since Linux 2.6.5)
4338 pgscan_kswapd_high
4340 pgscan_kswapd_movable (since Linux 2.6.23)
4342 pgscan_direct_dma
4344 pgscan_direct_dma32 (since Linux 2.6.16)
4346 pgscan_direct_normal
4348 pgscan_direct_high
4350 pgscan_direct_movable (since Linux 2.6.23)
4352 pgscan_direct_throttle (since Linux 3.6)
4354 zone_reclaim_failed (since linux 2.6.31)
4356 pginodesteal (since linux 2.6.0)
4358 slabs_scanned (since linux 2.6.5)
4360 kswapd_inodesteal (since linux 2.6.0)
4362 kswapd_low_wmark_hit_quickly (since 2.6.33)
4364 kswapd_high_wmark_hit_quickly (since 2.6.33)
4366 pageoutrun (since Linux 2.6.0)
4368 allocstall (since Linux 2.6.0)
4370 pgrotated (since Linux 2.6.0)
4372 drop_pagecache (since Linux 3.15)
4374 drop_slab (since Linux 3.15)
4376 numa_pte_updates (since Linux 3.8)
4378 numa_huge_pte_updates (since Linux 3.13)
4380 numa_hint_faults (since Linux 3.8)
4382 numa_hint_faults_local (since Linux 3.8)
4384 numa_pages_migrated (since Linux 3.8)
4386 pgmigrate_success (since Linux 3.8)
4388 pgmigrate_fail (since Linux 3.8)
4390 compact_migrate_scanned (since Linux 3.8)
4392 compact_free_scanned (since Linux 3.8)
4394 compact_isolated (since Linux 3.8)
4396 compact_stall (since Linux 2.6.35)
4398 See the kernel source file Documentation/ad‐
4399 min-guide/mm/transhuge.rst.
4400 compact_fail (since Linux 2.6.35)
4402 See the kernel source file Documentation/ad‐
4403 min-guide/mm/transhuge.rst.
4404 compact_success (since Linux 2.6.35)
4406 See the kernel source file Documentation/ad‐
4407 min-guide/mm/transhuge.rst.
4408 htlb_buddy_alloc_success (since Linux 2.6.26)
4410 htlb_buddy_alloc_fail (since Linux 2.6.26)
4412 unevictable_pgs_culled (since Linux 2.6.28)
4414 unevictable_pgs_scanned (since Linux 2.6.28)
4416 unevictable_pgs_rescued (since Linux 2.6.28)
4418 unevictable_pgs_mlocked (since Linux 2.6.28)
4420 unevictable_pgs_munlocked (since Linux 2.6.28)
4422 unevictable_pgs_cleared (since Linux 2.6.28)
4424 unevictable_pgs_stranded (since Linux 2.6.28)
4426 thp_fault_alloc (since Linux 2.6.39)
4428 See the kernel source file Documentation/ad‐
4429 min-guide/mm/transhuge.rst.
4430 thp_fault_fallback (since Linux 2.6.39)
4432 See the kernel source file Documentation/ad‐
4433 min-guide/mm/transhuge.rst.
4434 thp_collapse_alloc (since Linux 2.6.39)
4436 See the kernel source file Documentation/ad‐
4437 min-guide/mm/transhuge.rst.
4438 thp_collapse_alloc_failed (since Linux 2.6.39)
4440 See the kernel source file Documentation/ad‐
4441 min-guide/mm/transhuge.rst.
4442 thp_split (since Linux 2.6.39)
4444 See the kernel source file Documentation/ad‐
4445 min-guide/mm/transhuge.rst.
4446 thp_zero_page_alloc (since Linux 3.8)
4448 See the kernel source file Documentation/ad‐
4449 min-guide/mm/transhuge.rst.
4450 thp_zero_page_alloc_failed (since Linux 3.8)
4452 See the kernel source file Documentation/ad‐
4453 min-guide/mm/transhuge.rst.
4454 balloon_inflate (since Linux 3.18)
4456 balloon_deflate (since Linux 3.18)
4458 balloon_migrate (since Linux 3.18)
4460 nr_tlb_remote_flush (since Linux 3.12)
4462 nr_tlb_remote_flush_received (since Linux 3.12)
4464 nr_tlb_local_flush_all (since Linux 3.12)
4466 nr_tlb_local_flush_one (since Linux 3.12)
4468 vmacache_find_calls (since Linux 3.16)
4470 vmacache_find_hits (since Linux 3.16)
4472 vmacache_full_flushes (since Linux 3.19)
4474 /proc/zoneinfo (since Linux 2.6.13)
4476 This file displays information about memory zones. This is use‐
4477 ful for analyzing virtual memory behavior.
4478
4480 Many files contain strings (e.g., the environment and command line)
4481 that are in the internal format, with subfields terminated by null
4482 bytes ('\0'). When inspecting such files, you may find that the re‐
4483 sults are more readable if you use a command of the following form to
4484 display them:
4485 $ cat file | tr '\000' '\n'
4487 This manual page is incomplete, possibly inaccurate, and is the kind of
4489 thing that needs to be updated very often.
4490
4492 cat(1), dmesg(1), find(1), free(1), htop(1), init(1), ps(1), pstree(1),
4493 tr(1), uptime(1), chroot(2), mmap(2), readlink(2), syslog(2),
4494 slabinfo(5), sysfs(5), hier(7), namespaces(7), time(7), arp(8), hd‐
4495 parm(8), ifconfig(8), lsmod(8), lspci(8), mount(8), netstat(8),
4496 procinfo(8), route(8), sysctl(8)
4497 The Linux kernel source files: Documentation/filesystems/proc.txt, Doc‐
4499 umentation/sysctl/fs.txt, Documentation/sysctl/kernel.txt, Documenta‐
4500 tion/sysctl/net.txt, and Documentation/sysctl/vm.txt.
4501
4503 This page is part of release 5.10 of the Linux man-pages project. A
4504 description of the project, information about reporting bugs, and the
4505 latest version of this page, can be found at
4506 https://www.kernel.org/doc/man-pages/.
4507Linux 2020-12-21 PROC(5)