1 .\" Copyright (C) 1994, 1995 by Daniel Quinlan (quinlan@yggdrasil.com)
2 .\" and Copyright (C) 2002-2008 Michael Kerrisk <mtk.manpages@gmail.com>
3 .\" with networking additions from Alan Cox (A.Cox@swansea.ac.uk)
4 .\" and scsi additions from Michael Neuffer (neuffer@mail.uni-mainz.de)
5 .\" and sysctl additions from Andries Brouwer (aeb@cwi.nl)
6 .\" and System V IPC (as well as various other) additions from
7 .\" Michael Kerrisk <mtk.manpages@gmail.com>
9 .\" This is free documentation; you can redistribute it and/or
10 .\" modify it under the terms of the GNU General Public License as
11 .\" published by the Free Software Foundation; either version 2 of
12 .\" the License, or (at your option) any later version.
14 .\" The GNU General Public License's references to "object code"
15 .\" and "executables" are to be interpreted as the output of any
16 .\" document formatting or typesetting system, including
17 .\" intermediate and printed output.
19 .\" This manual is distributed in the hope that it will be useful,
20 .\" but WITHOUT ANY WARRANTY; without even the implied warranty of
21 .\" MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 .\" GNU General Public License for more details.
24 .\" You should have received a copy of the GNU General Public
25 .\" License along with this manual; if not, write to the Free
26 .\" Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111,
29 .\" Modified 1995-05-17 by faith@cs.unc.edu
30 .\" Minor changes by aeb and Marty Leisner (leisner@sdsp.mc.xerox.com).
31 .\" Modified 1996-04-13, 1996-07-22 by aeb@cwi.nl
32 .\" Modified 2001-12-16 by rwhron@earthlink.net
33 .\" Modified 2002-07-13 by jbelton@shaw.ca
34 .\" Modified 2002-07-22, 2003-05-27, 2004-04-06, 2004-05-25
35 .\" by Michael Kerrisk <mtk.manpages@gmail.com>
36 .\" 2004-11-17, mtk -- updated notes on /proc/loadavg
37 .\" 2004-12-01, mtk, rtsig-max and rtsig-nr went away in 2.6.8
38 .\" 2004-12-14, mtk, updated 'statm', and fixed error in order of list
39 .\" 2005-05-12, mtk, updated 'stat'
40 .\" 2005-07-13, mtk, added /proc/sys/fs/mqueue/*
41 .\" 2005-09-16, mtk, Added /proc/sys/fs/suid_dumpable
42 .\" 2005-09-19, mtk, added /proc/zoneinfo
43 .\" 2005-03-01, mtk, moved /proc/sys/fs/mqueue/* material to mq_overview.7.
44 .\" 2008-06-05, mtk, Added /proc/[pid]/oom_score, /proc/[pid]/oom_adj,
45 .\" /proc/[pid]/limits, /proc/[pid]/mountinfo, /proc/[pid]/mountstats,
46 .\" and /proc/[pid]/fdinfo/*.
47 .\" 2008-06-19, mtk, Documented /proc/[pid]/status.
48 .\" 2008-07-15, mtk, added /proc/config.gz
50 .\" FIXME 2.6.13 seems to have /proc/vmcore implemented
51 .\" in the source code, but there is no option available under
52 .\" 'make xconfig'; eventually this should be fixed, and then info
53 .\" from the patch-2.6.13 and change log could be used to write an
54 .\" entry in this man page.
55 .\" Needs CONFIG_VMCORE
57 .\" FIXME cross check against Documentation/filesystems/proc.txt
58 .\" to see what information could be imported from that file
61 .TH PROC 5 2012-08-18 "Linux" "Linux Programmer's Manual"
63 proc \- process information pseudo-file system
67 file system is a pseudo-file system which is used as an interface to
68 kernel data structures.
69 It is commonly mounted at
71 Most of it is read-only, but some files allow kernel variables to be
74 The following outline gives a quick tour through the
80 There is a numerical subdirectory for each running process; the
81 subdirectory is named by the process ID.
82 Each such subdirectory contains the following
83 pseudo-files and directories.
84 .\" FIXME Describe /proc/[pid]/attr and
85 .\" /proc/[pid]/task/[tid]/attr
86 .\" This is a directory
91 .IR /proc/[pid]/auxv " (since 2.6.0-test7)"
92 This contains the contents of the ELF interpreter information passed
93 to the process at exec time.
94 The format is one \fIunsigned long\fP ID
95 plus one \fIunsigned long\fP value for each entry.
96 The last entry contains two zeros.
98 .IR /proc/[pid]/cgroup " (since Linux 2.6.24)"
99 .\" Info in Documentation/cgroups/cgroups.txt
100 This file describes control groups to which the process/task belongs.
101 For each cgroup hierarchy there is one entry containing
102 colon-separated fields of the form:
106 5:cpuacct,cpu,cpuset:/daemons
110 The colon-separated fields are, from left to right:
115 set of subsystems bound to the hierarchy
117 control group in the hierarchy to which the process belongs
120 This file is only present if the
122 kernel configuration option is enabled.
124 .\" FIXME Describe /proc/[pid]/clear_refs
126 .\" "Clears page referenced bits shown in smaps output"
128 .\" CONFIG_PROC_PAGE_MONITOR
130 .I /proc/[pid]/cmdline
131 This holds the complete command line for the process,
132 unless the process is a zombie.
133 .\" In 2.3.26, this also used to be true if the process was swapped out.
134 In the latter case, there is nothing in this file:
135 that is, a read on this file will return 0 characters.
136 The command-line arguments appear in this file as a set of
137 strings separated by null bytes (\(aq\\0\(aq),
138 with a further null byte after the last string.
140 .IR /proc/[pid]/coredump_filter " (since kernel 2.6.23)"
144 .IR /proc/[pid]/cpuset " (since kernel 2.6.12)"
145 .\" and/proc/[pid]/task/[tid]/cpuset
150 This is a symbolic link to the current working directory of the process.
151 To find out the current working directory of process 20,
152 for instance, you can do this:
156 .RB "$" " cd /proc/20/cwd; /bin/pwd"
162 command is often a shell built-in, and might
169 .\" The following was still true as at kernel 2.6.13
170 In a multithreaded process, the contents of this symbolic link
171 are not available if the main thread has already terminated
172 (typically by calling
173 .BR pthread_exit (3)).
175 .I /proc/[pid]/environ
176 This file contains the environment for the process.
177 The entries are separated by null bytes (\(aq\\0\(aq),
178 and there may be a null byte at the end.
179 Thus, to print out the environment of process 1, you would do:
184 .RB "$" " (cat /proc/1/environ; echo) | tr \(aq\e000\(aq \(aq\en\(aq"
190 Under Linux 2.2 and later, this file is a symbolic link
191 containing the actual pathname of the executed command.
192 This symbolic link can be dereferenced normally; attempting to open
193 it will open the executable.
196 to run another copy of the same executable as is being run by
198 .\" The following was still true as at kernel 2.6.13
199 In a multithreaded process, the contents of this symbolic link
200 are not available if the main thread has already terminated
201 (typically by calling
202 .BR pthread_exit (3)).
204 Under Linux 2.0 and earlier
206 is a pointer to the binary which was executed,
207 and appears as a symbolic link.
210 call on this file under Linux 2.0 returns a string in the format:
214 For example, [0301]:1502 would be inode 1502 on device major 03 (IDE,
215 MFM, etc. drives) minor 01 (first partition on the first drive).
220 option can be used to locate the file.
223 This is a subdirectory containing one entry for each file which the
224 process has open, named by its file descriptor, and which is a
225 symbolic link to the actual file.
226 Thus, 0 is standard input, 1 standard output, 2 standard error, etc.
228 .\" The following was still true as at kernel 2.6.13
229 In a multithreaded process, the contents of this directory
230 are not available if the main thread has already terminated
231 (typically by calling
232 .BR pthread_exit (3)).
234 Programs that will take a filename as a command-line argument,
235 but will not take input from standard input if no argument is supplied,
236 or that write to a file named as a command-line argument,
237 but will not send their output to standard output
238 if no argument is supplied, can nevertheless be made to use
239 standard input or standard out using
241 For example, assuming that
243 is the flag designating an input file and
245 is the flag designating an output file:
249 .RB "$" " foobar \-i /proc/self/fd/0 \-o /proc/self/fd/1 ..."
253 and you have a working filter.
254 .\" The following is not true in my tests (MTK):
255 .\" Note that this will not work for
256 .\" programs that seek on their files, as the files in the fd directory
257 .\" are not seekable.
260 is approximately the same as
262 in some UNIX and UNIX-like systems.
263 Most Linux MAKEDEV scripts symbolically link
269 Most systems provide symbolic links
274 which respectively link to the files
281 Thus the example command above could be written as:
285 .RB "$" " foobar \-i /dev/stdin \-o /dev/stdout ..."
288 .\" FIXME Describe /proc/[pid]/loginuid
289 .\" Added in 2.6.11; updating requires CAP_AUDIT_CONTROL
290 .\" CONFIG_AUDITSYSCALL
292 .IR /proc/[pid]/fdinfo/ " (since kernel 2.6.22)"
293 This is a subdirectory containing one entry for each file which the
294 process has open, named by its file descriptor.
295 The contents of each file can be read to obtain information
296 about the corresponding file descriptor, for example:
300 .RB "$" " cat /proc/12015/fdinfo/4"
308 field is a decimal number showing the current file offset.
311 field is an octal number that displays the
312 file access mode and file status flags (see
315 The files in this directory are readable only by the owner of the process.
316 .\" FIXME document /proc/[pid]/io
318 .\" .IR /proc/[pid]/io " (since kernel 2.6.20)"
320 .IR /proc/[pid]/limits " (since kernel 2.6.24)"
321 This file displays the soft limit, hard limit, and units of measurement
322 for each of the process's resource limits (see
324 The file is protected to only allow reading by the real UID of the process.
327 A file containing the currently mapped memory regions and their access
334 address perms offset dev inode pathname
335 08048000-08056000 r-xp 00000000 03:0c 64593 /usr/sbin/gpm
336 08056000-08058000 rw-p 0000d000 03:0c 64593 /usr/sbin/gpm
337 08058000-0805b000 rwxp 00000000 00:00 0
338 40000000-40013000 r-xp 00000000 03:0c 4165 /lib/ld-2.2.4.so
339 40013000-40015000 rw-p 00012000 03:0c 4165 /lib/ld-2.2.4.so
340 4001f000-40135000 r-xp 00000000 03:0c 45494 /lib/libc-2.2.4.so
341 40135000-4013e000 rw-p 00115000 03:0c 45494 /lib/libc-2.2.4.so
342 4013e000-40142000 rw-p 00000000 00:00 0
343 bffff000-c0000000 rwxp 00000000 00:00 0
347 where "address" is the address space in the process that it occupies,
348 "perms" is a set of permissions:
356 p = private (copy on write)
360 "offset" is the offset into the file/whatever, "dev" is the device
361 (major:minor), and "inode" is the inode on that device.
362 0 indicates that no inode is associated with the memory region,
363 as the case would be with BSS (uninitialized data).
365 Under Linux 2.0 there is no field giving pathname.
368 This file can be used to access the pages of a process's memory through
374 .IR /proc/[pid]/mountinfo " (since Linux 2.6.26)"
375 .\" This info adapted from Documentation/filesystems/proc.txt
376 This file contains information about mount points.
377 It contains lines of the form:
381 36 35 98:0 /mnt1 /mnt2 rw,noatime master:1 - ext3 /dev/root rw,errors=continue
382 (1)(2)(3) (4) (5) (6) (7) (8) (9) (10) (11)
386 The numbers in parentheses are labels for the descriptions below:
390 mount ID: unique identifier of the mount (may be reused after
394 parent ID: ID of parent mount (or of self for the top of the mount tree).
397 major:minor: value of
399 for files on file system (see
403 root: root of the mount within the file system.
406 mount point: mount point relative to the process's root.
409 mount options: per-mount options.
412 optional fields: zero or more fields of the form "tag[:value]".
415 separator: marks the end of the optional fields.
418 file system type: name of file system in the form "type[.subtype]".
421 mount source: file system-specific information or "none".
424 super options: per-super block options.
427 Parsers should ignore all unrecognized optional fields.
428 Currently the possible optional fields are:
432 mount is shared in peer group X
435 mount is slave to peer group X
438 mount is slave and receives propagation from peer group X (*)
444 (*) X is the closest dominant peer group under the process's root.
445 If X is the immediate master of the mount,
446 or if there is no dominant peer group under the same root,
447 then only the "master:X" field is present
448 and not the "propagate_from:X" field.
450 For more information on mount propagation see:
451 .I Documentation/filesystems/sharedsubtree.txt
452 in the Linux kernel source tree.
454 .IR /proc/[pid]/mounts " (since Linux 2.4.19)"
455 This is a list of all the file systems currently mounted in the
456 process's mount namespace.
457 The format of this file is documented in
459 Since kernel version 2.6.15, this file is pollable:
460 after opening the file for reading, a change in this file
461 (i.e., a file system mount or unmount) causes
463 to mark the file descriptor as readable, and
467 mark the file as having an error condition.
469 .IR /proc/[pid]/mountstats " (since Linux 2.6.17)"
470 This file exports information (statistics, configuration information)
471 about the mount points in the process's name space.
472 Lines in this file have the form:
475 device /dev/sda7 mounted on /home with fstype ext3 [statistics]
479 The fields in each line are:
483 The name of the mounted device
484 (or "nodevice" if there is no corresponding device).
487 The mount point within the file system tree.
490 The file system type.
493 Optional statistics and configuration information.
494 Currently (as at Linux 2.6.26), only NFS file systems export
495 information via this field.
498 This file is only readable by the owner of the process.
500 .IR /proc/[pid]/ns/ " (since Linux 3.0)"
501 .\" See commit 6b4e306aa3dc94a0545eb9279475b1ab6209a31f
502 This is a subdirectory containing one entry for each namespace that
503 supports being manipulated by
505 For information about namespaces, see
508 .IR /proc/[pid]/ns/ipc " (since Linux 3.0)"
509 Bind mounting this file (see
511 to somewhere else in the filesystem keeps
512 the IPC namespace of the process specified by
514 alive even if all processes currently in the namespace terminate.
516 Opening this file returns a file handle for the IPC namespace
517 of the process specified by
519 As long as this file descriptor remains open,
520 the IPC namespace will remain alive,
521 even if all processes in the namespace terminate.
522 The file descriptor can be passed to
525 .IR /proc/[pid]/ns/net " (since Linux 3.0)"
526 Bind mounting this file (see
528 to somewhere else in the filesystem keeps
529 the network namespace of the process specified by
531 alive even if all processes in the namespace terminate.
533 Opening this file returns a file handle for the network namespace
534 of the process specified by
536 As long as this file descriptor remains open,
537 the network namespace will remain alive,
538 even if all processes in the namespace terminate.
539 The file descriptor can be passed to
542 .IR /proc/[pid]/ns/uts " (since Linux 3.0)"
543 Bind mounting this file (see
545 to somewhere else in the filesystem keeps
546 the UTS namespace of the process specified by
548 alive even if all processes currently in the namespace terminate.
550 Opening this file returns a file handle for the UTS namespace
551 of the process specified by
553 As long as this file descriptor remains open,
554 the UTS namespace will remain alive,
555 even if all processes in the namespace terminate.
556 The file descriptor can be passed to
559 .IR /proc/[pid]/numa_maps " (since Linux 2.6.14)"
563 .IR /proc/[pid]/oom_adj " (since Linux 2.6.11)"
564 This file can be used to adjust the score used to select which process
565 should be killed in an out-of-memory (OOM) situation.
566 The kernel uses this value for a bit-shift operation of the process's
569 valid values are in the range \-16 to +15,
570 plus the special value \-17,
571 which disables OOM-killing altogether for this process.
572 A positive score increases the likelihood of this
573 process being killed by the OOM-killer;
574 a negative score decreases the likelihood.
575 The default value for this file is 0;
576 a new process inherits its parent's
579 A process must be privileged
580 .RB ( CAP_SYS_RESOURCE )
583 .IR /proc/[pid]/oom_score " (since Linux 2.6.11)"
584 .\" See mm/oom_kill.c::badness() in the 2.6.25 sources
585 This file displays the current score that the kernel gives to
586 this process for the purpose of selecting a process
588 A higher score means that the process is more likely to be
589 selected by the OOM-killer.
590 The basis for this score is the amount of memory used by the process,
591 with increases (+) or decreases (\-) for factors including:
592 .\" See mm/oom_kill.c::badness() in the 2.6.25 sources
595 whether the process creates a lot of children using
599 whether the process has been running a long time,
600 or has used a lot of CPU time (\-);
602 whether the process has a low nice value (i.e., > 0) (+);
604 whether the process is privileged (\-); and
605 .\" More precisely, if it has CAP_SYS_ADMIN or CAP_SYS_RESOURCE
607 whether the process is making direct hardware access (\-).
608 .\" More precisely, if it has CAP_SYS_RAWIO
613 also reflects the bit-shift adjustment specified by the
615 setting for the process.
616 .\" FIXME Describe /proc/[pid]/pagemap
618 .\" CONFIG_PROC_PAGE_MONITOR
621 UNIX and Linux support the idea of a per-process root of the
622 file system, set by the
625 This file is a symbolic link that points to the process's
626 root directory, and behaves as exe, fd/*, etc. do.
628 .\" The following was still true as at kernel 2.6.13
629 In a multithreaded process, the contents of this symbolic link
630 are not available if the main thread has already terminated
631 (typically by calling
632 .BR pthread_exit (3)).
633 .\" FIXME Describe /proc/[pid]/seccomp
635 .\" FIXME Describe /proc/[pid]/sessionid
636 .\" Added in 2.6.25; read-only; only readable by real UID
637 .\" CONFIG_AUDITSYSCALL
638 .\" FIXME Describe /proc/[pid]/sched
640 .\" CONFIG_SCHED_DEBUG, and additional fields if CONFIG_SCHEDSTATS
641 .\" Displays various scheduling parameters
642 .\" This file can be written, to reset stats
643 .\" FIXME Describe /proc/[pid]/schedstats and
644 .\" /proc/[pid]/task/[tid]/schedstats
646 .\" CONFIG_SCHEDSTATS
648 .IR /proc/[pid]/smaps " (since Linux 2.6.14)"
649 .\" CONFIG_PROC_PAGE_MONITOR
650 This file shows memory consumption for each of the process's mappings.
651 For each of mappings there is a series of lines such as the following:
655 08048000-080bc000 r-xp 00000000 03:02 13130 /bin/bash
665 The first of these lines shows the same information as is displayed
667 .IR /proc/[pid]/maps .
668 The remaining lines show the size of the mapping,
669 the amount of the mapping that is currently resident in RAM,
670 the number of clean and dirty shared pages in the mapping,
671 and the number of clean and dirty private pages in the mapping.
673 This file is only present if the
679 Status information about the process.
683 .IR /usr/src/linux/fs/proc/array.c "."
685 The fields, in order, with their proper
687 format specifiers, are:
694 The filename of the executable, in parentheses.
695 This is visible whether or not the executable is swapped out.
698 One character from the string "RSDZTW" where R is running, S is
699 sleeping in an interruptible wait, D is waiting in uninterruptible
700 disk sleep, Z is zombie, T is traced or stopped (on a signal),
704 The PID of the parent.
707 The process group ID of the process.
710 The session ID of the process.
713 The controlling terminal of the process.
714 (The minor device number is contained in the combination of bits
716 the major device number is in bits 15 to 8.)
719 .\" This field and following, up to and including wchan added 0.99.1
720 The ID of the foreground process group of the controlling
721 terminal of the process.
723 \fIflags\fP %u (%lu before Linux 2.6.22)
724 The kernel flags word of the process.
726 see the PF_* defines in the Linux kernel source file
727 .IR include/linux/sched.h .
728 Details depend on the kernel version.
731 The number of minor faults the process has made which have not
732 required loading a memory page from disk.
736 The number of minor faults that the process's
737 waited-for children have made.
740 The number of major faults the process has made which have
741 required loading a memory page from disk.
744 The number of major faults that the process's
745 waited-for children have made.
748 Amount of time that this process has been scheduled in user mode,
749 measured in clock ticks (divide by
750 .IR sysconf(_SC_CLK_TCK) ).
751 This includes guest time, \fIguest_time\fP
752 (time spent running a virtual CPU, see below),
753 so that applications that are not aware of the guest time field
754 do not lose that time from their calculations.
757 Amount of time that this process has been scheduled in kernel mode,
758 measured in clock ticks (divide by
759 .IR sysconf(_SC_CLK_TCK) ).
762 Amount of time that this process's
763 waited-for children have been scheduled in user mode,
764 measured in clock ticks (divide by
765 .IR sysconf(_SC_CLK_TCK) ).
768 This includes guest time, \fIcguest_time\fP
769 (time spent running a virtual CPU, see below).
772 Amount of time that this process's
773 waited-for children have been scheduled in kernel mode,
774 measured in clock ticks (divide by
775 .IR sysconf(_SC_CLK_TCK) ).
778 (Explanation for Linux 2.6)
779 For processes running a real-time scheduling policy
782 .BR sched_setscheduler (2)),
783 this is the negated scheduling priority, minus one;
784 that is, a number in the range \-2 to \-100,
785 corresponding to real-time priorities 1 to 99.
786 For processes running under a non-real-time scheduling policy,
787 this is the raw nice value
788 .RB ( setpriority (2))
789 as represented in the kernel.
790 The kernel stores nice values as numbers
791 in the range 0 (high) to 39 (low),
792 corresponding to the user-visible nice range of \-20 to 19.
794 Before Linux 2.6, this was a scaled value based on
795 the scheduler weighting given to this process.
796 .\" And back in kernel 1.2 days things were different again.
800 .BR setpriority (2)),
801 a value in the range 19 (low priority) to \-20 (high priority).
802 .\" Back in kernel 1.2 days things were different.
805 .\" \fIcounter\fP %ld
806 .\" The current maximum size in jiffies of the process's next timeslice,
807 .\" or what is currently left of its current timeslice, if it is the
808 .\" currently running process.
811 .\" The time in jiffies of the process's next timeout.
812 .\" timeout was removed sometime around 2.1/2.2
813 \fInum_threads\fP %ld
814 Number of threads in this process (since Linux 2.6).
815 Before kernel 2.6, this field was hard coded to 0 as a placeholder
816 for an earlier removed field.
819 \fIitrealvalue\fP %ld
820 The time in jiffies before the next
822 is sent to the process due to an interval timer.
823 Since kernel 2.6.17, this field is no longer maintained,
824 and is hard coded as 0.
826 \fIstarttime\fP %llu (was %lu before Linux 2.6)
827 The time the process started after system boot.
828 In kernels before Linux 2.6, this value was expressed in jiffies.
829 Since Linux 2.6, the value is expressed in clock ticks (divide by
830 .IR sysconf(_SC_CLK_TCK) ).
833 Virtual memory size in bytes.
836 Resident Set Size: number of pages the process has in real memory.
837 This is just the pages which
838 count toward text, data, or stack space.
839 This does not include pages
840 which have not been demand-loaded in, or which are swapped out.
843 Current soft limit in bytes on the rss of the process;
844 see the description of
850 The address above which program text can run.
853 The address below which program text can run.
856 The address of the start (i.e., bottom) of the stack.
859 The current value of ESP (stack pointer), as found in the
860 kernel stack page for the process.
863 The current EIP (instruction pointer).
867 The bitmap of pending signals, displayed as a decimal number.
868 Obsolete, because it does not provide information on real-time signals; use
869 .I /proc/[pid]/status
873 The bitmap of blocked signals, displayed as a decimal number.
874 Obsolete, because it does not provide information on real-time signals; use
875 .I /proc/[pid]/status
879 The bitmap of ignored signals, displayed as a decimal number.
880 Obsolete, because it does not provide information on real-time signals; use
881 .I /proc/[pid]/status
885 The bitmap of caught signals, displayed as a decimal number.
886 Obsolete, because it does not provide information on real-time signals; use
887 .I /proc/[pid]/status
891 This is the "channel" in which the process is waiting.
893 address of a system call, and can be looked up in a namelist if you
895 (If you have an up-to-date
896 .IR /etc/psdatabase ,
898 try \fIps \-l\fP to see the WCHAN field in action.)
901 .\" nswap was added in 2.0
902 Number of pages swapped (not maintained).
905 .\" cnswap was added in 2.0
906 Cumulative \fInswap\fP for child processes (not maintained).
908 \fIexit_signal\fP %d (since Linux 2.1.22)
909 Signal to be sent to parent when we die.
911 \fIprocessor\fP %d (since Linux 2.2.8)
912 CPU number last executed on.
914 \fIrt_priority\fP %u (since Linux 2.5.19; was %lu before Linux 2.6.22)
915 Real-time scheduling priority, a number in the range 1 to 99 for
916 processes scheduled under a real-time policy,
917 or 0, for non-real-time processes (see
918 .BR sched_setscheduler (2)).
921 \fIpolicy\fP %u (since Linux 2.5.19; was %lu before Linux 2.6.22)
922 Scheduling policy (see
923 .BR sched_setscheduler (2)).
924 Decode using the SCHED_* constants in
927 \fIdelayacct_blkio_ticks\fP %llu (since Linux 2.6.18)
928 Aggregated block I/O delays, measured in clock ticks (centiseconds).
930 \fIguest_time\fP %lu (since Linux 2.6.24)
931 Guest time of the process (time spent running a virtual CPU
932 for a guest operating system), measured in clock ticks (divide by
933 .IR sysconf(_SC_CLK_TCK) ).
935 \fIcguest_time\fP %ld (since Linux 2.6.24)
936 Guest time of the process's children, measured in clock ticks (divide by
937 .IR sysconf(_SC_CLK_TCK) ).
941 Provides information about memory usage, measured in pages.
946 size total program size
947 (same as VmSize in \fI/proc/[pid]/status\fP)
948 resident resident set size
949 (same as VmRSS in \fI/proc/[pid]/status\fP)
950 share shared pages (from shared mappings)
952 .\" (not including libs; broken, includes data segment)
953 lib library (unused in Linux 2.6)
955 .\" (including libs; broken, includes library text)
956 dt dirty pages (unused in Linux 2.6)
960 .I /proc/[pid]/status
961 Provides much of the information in
965 in a format that's easier for humans to parse.
970 .RB "$" " cat /proc/$$/status"
977 Uid: 1000 1000 1000 1000
993 SigPnd: 0000000000000000
994 ShdPnd: 0000000000000000
995 SigBlk: 0000000000010000
996 SigIgn: 0000000000384004
997 SigCgt: 000000004b813efb
998 CapInh: 0000000000000000
999 CapPrm: 0000000000000000
1000 CapEff: 0000000000000000
1001 CapBnd: ffffffffffffffff
1002 Cpus_allowed: 00000001
1003 Cpus_allowed_list: 0
1005 Mems_allowed_list: 0
1006 voluntary_ctxt_switches: 150
1007 nonvoluntary_ctxt_switches: 545
1011 The fields are as follows:
1015 Command run by this process.
1018 Current state of the process.
1030 Thread group ID (i.e., Process ID).
1037 PID of parent process.
1040 PID of process tracing this process (0 if not being traced).
1043 Real, effective, saved set, and file system UIDs (GIDs).
1046 Number of file descriptor slots currently allocated.
1049 Supplementary group list.
1052 Peak virtual memory size.
1055 Virtual memory size.
1058 Locked memory size (see
1062 Peak resident set size ("high water mark").
1067 .IR VmData ", " VmStk ", " VmExe :
1068 Size of data, stack, and text segments.
1071 Shared library code size.
1074 Page table entries size (since Linux 2.6.10).
1077 Number of threads in process containing this thread.
1080 This field contains two slash-separated numbers that relate to
1081 queued signals for the real user ID of this process.
1082 The first of these is the number of currently queued
1083 signals for this real user ID, and the second is the
1084 resource limit on the number of queued signals for this process
1085 (see the description of
1086 .BR RLIMIT_SIGPENDING
1090 .IR SigPnd ", " ShdPnd :
1091 Number of signals pending for thread and for process as a whole (see
1096 .IR SigBlk ", " SigIgn ", " SigCgt :
1097 Masks indicating signals being blocked, ignored, and caught (see
1100 .IR CapInh ", " CapPrm ", " CapEff :
1101 Masks of capabilities enabled in inheritable, permitted, and effective sets
1103 .BR capabilities (7)).
1106 Capability Bounding set
1107 (since kernel 2.6.26, see
1108 .BR capabilities (7)).
1111 Mask of CPUs on which this process may run
1112 (since Linux 2.6.24, see
1115 .IR Cpus_allowed_list :
1116 Same as previous, but in "list format"
1117 (since Linux 2.6.26, see
1121 Mask of memory nodes allowed to this process
1122 (since Linux 2.6.24, see
1125 .IR Mems_allowed_list :
1126 Same as previous, but in "list format"
1127 (since Linux 2.6.26, see
1130 .IR voluntary_context_switches ", " nonvoluntary_context_switches :
1131 Number of voluntary and involuntary context switches (since Linux 2.6.23).
1134 .IR /proc/[pid]/task " (since Linux 2.6.0-test6)"
1135 This is a directory that contains one subdirectory
1136 for each thread in the process.
1137 The name of each subdirectory is the numerical thread ID
1141 Within each of these subdirectories, there is a set of
1142 files with the same names and contents as under the
1145 For attributes that are shared by all threads, the contents for
1146 each of the files under the
1148 subdirectories will be the same as in the corresponding
1152 (e.g., in a multithreaded process, all of the
1154 files will have the same value as the
1156 file in the parent directory, since all of the threads in a process
1157 share a working directory).
1158 For attributes that are distinct for each thread,
1159 the corresponding files under
1161 may have different values (e.g., various fields in each of the
1162 .I task/[tid]/status
1163 files may be different for each thread).
1165 .\" The following was still true as at kernel 2.6.13
1166 In a multithreaded process, the contents of the
1168 directory are not available if the main thread has already terminated
1169 (typically by calling
1170 .BR pthread_exit (3)).
1173 Advanced power management version and battery information when
1175 is defined at kernel compilation time.
1178 Contains subdirectories for installed busses.
1181 Subdirectory for PCMCIA devices when
1183 is set at kernel compilation time.
1185 .I /proc/bus/pccard/drivers
1188 Contains various bus subdirectories and pseudo-files containing
1189 information about PCI busses, installed devices, and device
1191 Some of these files are not ASCII.
1193 .I /proc/bus/pci/devices
1194 Information about PCI devices.
1195 They may be accessed through
1201 Arguments passed to the Linux kernel at boot time.
1202 Often done via a boot manager such as
1207 .IR /proc/config.gz " (since Linux 2.6)"
1208 This file exposes the configuration options that were used
1209 to build the currently running kernel,
1210 in the same format as they would be shown in the
1212 file that resulted when configuring the kernel (using
1213 .IR "make xconfig" ,
1216 The file contents are compressed; view or search them using
1220 As long as no changes have been made to the following file,
1223 are the same as those provided by :
1227 cat /lib/modules/$(uname \-r)/build/.config
1232 is only provided if the kernel is configured with
1233 .BR CONFIG_IKCONFIG_PROC .
1236 This is a collection of CPU and system architecture dependent items,
1237 for each supported architecture a different list.
1238 Two common entries are \fIprocessor\fP which gives CPU number and
1239 \fIbogomips\fP; a system constant that is calculated
1240 during kernel initialization.
1241 SMP machines have information for
1245 Text listing of major numbers and device groups.
1246 This can be used by MAKEDEV scripts for consistency with the kernel.
1248 .IR /proc/diskstats " (since Linux 2.5.69)"
1249 This file contains disk I/O statistics for each disk device.
1250 See the Linux kernel source file
1251 .I Documentation/iostats.txt
1252 for further information.
1255 This is a list of the registered \fIISA\fP DMA (direct memory access)
1261 .I /proc/execdomains
1262 List of the execution domains (ABI personalities).
1265 Frame buffer information when
1267 is defined during kernel compilation.
1269 .I /proc/filesystems
1270 A text listing of the file systems which are supported by the kernel,
1271 namely file systems which were compiled into the kernel or whose kernel
1272 modules are currently loaded.
1274 .BR filesystems (5).)
1275 If a file system is marked with "nodev",
1276 this means that it does not require a block device to be mounted
1277 (e.g., virtual file system, network file system).
1279 Incidentally, this file may be used by
1281 when no file system is specified and it didn't manage to determine the
1283 Then file systems contained in this file are tried
1284 (excepted those that are marked with "nodev").
1291 exists on systems with the IDE bus.
1292 There are directories for each IDE channel and attached device.
1297 cache buffer size in KB
1298 capacity number of sectors
1299 driver driver version
1300 geometry physical and logical geometry
1301 identify in hexadecimal
1303 model manufacturer's model number
1304 settings drive settings
1305 smart_thresholds in hexadecimal
1306 smart_values in hexadecimal
1312 utility provides access to this information in a friendly format.
1315 This is used to record the number of interrupts per CPU per IO device.
1317 for the i386 and x86_64 architectures, at least, this also includes
1318 interrupts internal to the system (that is, not associated with a device
1319 as such), such as NMI (nonmaskable interrupt), LOC (local timer interrupt),
1320 and for SMP systems, TLB (TLB flush interrupt), RES (rescheduling
1321 interrupt), CAL (remote function call interrupt), and possibly others.
1322 Very easy to read formatting, done in ASCII.
1325 I/O memory map in Linux 2.4.
1328 This is a list of currently registered Input-Output port regions that
1331 .IR /proc/kallsyms " (since Linux 2.5.71)"
1332 This holds the kernel exported symbol definitions used by the
1334 tools to dynamically link and bind loadable modules.
1335 In Linux 2.5.47 and earlier, a similar file with slightly different syntax
1340 This file represents the physical memory of the system and is stored
1341 in the ELF core file format.
1342 With this pseudo-file, and an unstripped
1344 .RI ( /usr/src/linux/vmlinux )
1345 binary, GDB can be used to
1346 examine the current state of any kernel data structures.
1348 The total length of the file is the size of physical memory (RAM) plus
1352 This file can be used instead of the
1354 system call to read kernel messages.
1355 A process must have superuser
1356 privileges to read this file, and only one process should read this
1358 This file should not be read if a syslog process is running
1361 system call facility to log kernel messages.
1363 Information in this file is retrieved with the
1367 .IR /proc/ksyms " (Linux 1.1.23-2.5.47)"
1369 .IR /proc/kallsyms .
1372 The first three fields in this file are load average figures
1373 giving the number of jobs in the run queue (state R)
1374 or waiting for disk I/O (state D) averaged over 1, 5, and 15 minutes.
1375 They are the same as the load average numbers given by
1378 The fourth field consists of two numbers separated by a slash (/).
1379 The first of these is the number of currently runnable kernel
1380 scheduling entities (processes, threads).
1381 The value after the slash is the number of kernel scheduling entities
1382 that currently exist on the system.
1383 The fifth field is the PID of the process that was most
1384 recently created on the system.
1387 This file shows current file locks
1388 .RB ( flock "(2) and " fcntl (2))
1392 .IR /proc/malloc " (only up to and including Linux 2.2)"
1393 .\" It looks like this only ever did something back in 1.0 days
1394 This file is only present if
1395 .B CONFIG_DEBUG_MALLOC
1396 was defined during compilation.
1399 This file reports statistics about memory usage on the system.
1402 to report the amount of free and used memory (both physical and swap)
1403 on the system as well as the shared memory and buffers used by the
1407 A text list of the modules that have been loaded by the system.
1412 Before kernel 2.4.19, this file was a list
1413 of all the file systems currently mounted on the system.
1414 With the introduction of per-process mount namespaces in
1415 Linux 2.4.19, this file became a link to
1416 .IR /proc/self/mounts ,
1417 which lists the mount points of the process's own mount namespace.
1418 The format of this file is documented in
1422 Memory Type Range Registers.
1423 See the Linux kernel source file
1424 .I Documentation/mtrr.txt
1428 various net pseudo-files, all of which give the status of some part of
1429 the networking layer.
1430 These files contain ASCII structures and are,
1431 therefore, readable with
1433 However, the standard
1435 suite provides much cleaner access to these files.
1438 This holds an ASCII readable dump of the kernel ARP table used for
1439 address resolutions.
1440 It will show both dynamically learned and preprogrammed ARP entries.
1446 IP address HW type Flags HW address Mask Device
1447 192.168.0.50 0x1 0x2 00:50:BF:25:68:F3 * eth0
1448 192.168.0.250 0x1 0xc 00:00:00:00:00:00 * eth0
1453 Here "IP address" is the IPv4 address of the machine and the "HW type"
1454 is the hardware type of the address from RFC\ 826.
1455 The flags are the internal
1456 flags of the ARP structure (as defined in
1457 .IR /usr/include/linux/if_arp.h )
1459 the "HW address" is the data link layer mapping for that IP address if
1463 The dev pseudo-file contains network device status information.
1465 the number of received and sent packets, the number of errors and
1467 and other basic statistics.
1468 These are used by the
1470 program to report device status.
1476 Inter-| Receive | Transmit
1477 face |bytes packets errs drop fifo frame compressed multicast|bytes packets errs drop fifo colls carrier compressed
1478 lo: 2776770 11307 0 0 0 0 0 0 2776770 11307 0 0 0 0 0 0
1479 eth0: 1215645 2751 0 0 0 0 0 0 1782404 4324 0 0 0 427 0 0
1480 ppp0: 1622270 5552 1 0 0 0 0 0 354130 5669 0 0 0 0 0 0
1481 tap0: 7714 81 0 0 0 0 0 0 7714 81 0 0 0 0 0 0
1486 .\" .I /proc/net/ipx
1489 .\" .I /proc/net/ipx_route
1492 .I /proc/net/dev_mcast
1494 .IR /usr/src/linux/net/core/dev_mcast.c :
1497 indx interface_name dmi_u dmi_g dmi_address
1498 2 eth0 1 0 01005e000001
1499 3 eth1 1 0 01005e000001
1500 4 eth2 1 0 01005e000001
1505 Internet Group Management Protocol.
1507 .IR /usr/src/linux/net/core/igmp.c .
1510 This file uses the same format as the
1512 file and contains the current reverse mapping database used to provide
1514 reverse address lookup services.
1515 If RARP is not configured into the
1517 this file will not be present.
1520 Holds a dump of the RAW socket table.
1521 Much of the information is not of
1523 apart from debugging.
1524 The "sl" value is the kernel hash slot for the
1526 the "local_address" is the local address and protocol number pair.
1528 the internal status of the socket.
1529 The "tx_queue" and "rx_queue" are the
1530 outgoing and incoming data queue in terms of kernel memory usage.
1531 The "tr", "tm\->when", and "rexmits" fields are not used by RAW.
1533 field holds the effective UID of the creator of the socket.
1535 .\" .I /proc/net/route
1536 .\" No information, but looks similar to
1540 This file holds the ASCII data needed for the IP, ICMP, TCP, and UDP
1542 information bases for an SNMP agent.
1545 Holds a dump of the TCP socket table.
1546 Much of the information is not
1547 of use apart from debugging.
1548 The "sl" value is the kernel hash slot
1549 for the socket, the "local_address" is the local address and port number pair.
1550 The "rem_address" is the remote address and port number pair
1552 \&"St" is the internal status of the socket.
1553 The "tx_queue" and "rx_queue" are the
1554 outgoing and incoming data queue in terms of kernel memory usage.
1555 The "tr", "tm\->when", and "rexmits" fields hold internal information of
1556 the kernel socket state and are only useful for debugging.
1558 field holds the effective UID of the creator of the socket.
1561 Holds a dump of the UDP socket table.
1562 Much of the information is not of
1563 use apart from debugging.
1564 The "sl" value is the kernel hash slot for the
1565 socket, the "local_address" is the local address and port number pair.
1566 The "rem_address" is the remote address and port number pair
1567 (if connected). "St" is the internal status of the socket.
1568 The "tx_queue" and "rx_queue" are the outgoing and incoming data queue
1569 in terms of kernel memory usage.
1570 The "tr", "tm\->when", and "rexmits" fields
1571 are not used by UDP.
1573 field holds the effective UID of the creator of the socket.
1579 sl local_address rem_address st tx_queue rx_queue tr rexmits tm\->when uid
1580 1: 01642C89:0201 0C642C89:03FF 01 00000000:00000001 01:000071BA 00000000 0
1581 1: 00000000:0801 00000000:0000 0A 00000000:00000000 00:00000000 6F000100 0
1582 1: 00000000:0201 00000000:0000 0A 00000000:00000000 00:00000000 00000000 0
1588 Lists the UNIX domain sockets present within the system and their
1594 Num RefCount Protocol Flags Type St Path
1595 0: 00000002 00000000 00000000 0001 03
1596 1: 00000001 00000000 00010000 0001 01 /dev/printer
1601 Here "Num" is the kernel table slot number, "RefCount" is the number
1602 of users of the socket, "Protocol" is currently always 0, "Flags"
1603 represent the internal kernel flags holding the status of the
1605 Currently, type is always "1" (UNIX domain datagram sockets are
1606 not yet supported in the kernel).
1607 \&"St" is the internal state of the
1608 socket and Path is the bound path (if any) of the socket.
1611 Contains major and minor numbers of each partition as well as number
1612 of blocks and partition name.
1615 This is a listing of all PCI devices found during kernel initialization
1616 and their configuration.
1618 This file has been deprecated in favor of a new
1621 .RI ( /proc/bus/pci ).
1622 It became optional in Linux 2.2 (available with
1623 .B CONFIG_PCI_OLD_PROC
1624 set at kernel compilation).
1625 It became once more nonoptionally enabled in Linux 2.4.
1626 Next, it was deprecated in Linux 2.6 (still available with
1627 .B CONFIG_PCI_LEGACY_PROC
1628 set), and finally removed altogether since Linux 2.6.17.
1629 .\" FIXME /proc/sched_debug
1631 .\" .IR /proc/sched_debug " (since Linux 2.6.23)"
1632 .\" See also /proc/[pid]/sched
1635 A directory with the
1637 mid-level pseudo-file and various SCSI low-level
1639 which contain a file for each SCSI host in this system, all of
1640 which give the status of some part of the SCSI IO subsystem.
1641 These files contain ASCII structures and are, therefore, readable with
1644 You can also write to some of the files to reconfigure the subsystem or
1645 switch certain features on or off.
1648 This is a listing of all SCSI devices known to the kernel.
1649 The listing is similar to the one seen during bootup.
1650 scsi currently supports only the \fIadd-single-device\fP command which
1651 allows root to add a hotplugged device to the list of known devices.
1657 echo \(aqscsi add-single-device 1 0 5 0\(aq > /proc/scsi/scsi
1662 host scsi1 to scan on SCSI channel 0 for a device on ID 5 LUN 0.
1664 is already a device known on this address or the address is invalid, an
1665 error will be returned.
1667 .I /proc/scsi/[drivername]
1668 \fI[drivername]\fP can currently be NCR53c7xx, aha152x, aha1542, aha1740,
1669 aic7xxx, buslogic, eata_dma, eata_pio, fdomain, in2000, pas16, qlogic,
1670 scsi_debug, seagate, t128, u15-24f, ultrastore, or wd7000.
1671 These directories show up for all drivers that registered at least one
1673 Every directory contains one file per registered host.
1674 Every host-file is named after the number the host was assigned during
1677 Reading these files will usually show driver and host configuration,
1680 Writing to these files allows different things on different hosts.
1681 For example, with the \fIlatency\fP and \fInolatency\fP commands,
1682 root can switch on and off command latency measurement code in the
1684 With the \fIlockup\fP and \fIunlock\fP commands,
1685 root can control bus lockups simulated by the scsi_debug driver.
1688 This directory refers to the process accessing the
1691 and is identical to the
1693 directory named by the process ID of the same process.
1696 Information about kernel caches.
1697 Since Linux 2.6.16 this file is only present if the
1699 kernel configuration option is enabled.
1721 kernel/system statistics.
1722 Varies with architecture.
1727 \fIcpu 3357 0 4313 1362393\fP
1728 The amount of time, measured in units of
1729 USER_HZ (1/100ths of a second on most architectures, use
1730 .IR sysconf(_SC_CLK_TCK)
1731 to obtain the right value),
1732 .\" 1024 on Alpha and ia64
1733 that the system spent in user mode,
1734 user mode with low priority (nice), system mode, and the
1735 idle task, respectively.
1736 .\" FIXME Actually, the following info about the /proc/stat 'cpu' field
1737 .\" does not seem to be quite right (at least in 2.6.12)
1738 The last value should be USER_HZ times the
1739 second entry in the uptime pseudo-file.
1741 In Linux 2.6 this line includes three additional columns:
1743 \- time waiting for I/O to complete (since 2.5.41);
1745 \- time servicing interrupts (since 2.6.0-test4);
1747 \- time servicing softirqs (since 2.6.0-test4).
1749 Since Linux 2.6.11, there is an eighth column,
1751 \- stolen time, which is the time spent in other operating systems when
1752 running in a virtualized environment
1754 Since Linux 2.6.24, there is a ninth column,
1756 which is the time spent running a virtual CPU for guest
1757 operating systems under the control of the Linux kernel.
1758 .\" See Changelog entry for 5e84cfde51cf303d368fcb48f22059f37b3872de
1760 \fIpage 5741 1808\fP
1761 The number of pages the system paged in and the number that were paged
1765 The number of swap pages that have been brought in and out.
1767 .\" FIXME The following is not the full picture for the 'intr' of
1768 .\" /proc/stat on 2.6:
1770 This line shows counts of interrupts serviced since boot time,
1771 for each of the possible system interrupts.
1772 The first column is the total of all interrupts serviced;
1773 each subsequent column is the total for a particular interrupt.
1775 \fIdisk_io: (2,0):(31,30,5764,1,2) (3,0):\fP...
1776 (major,disk_idx):(noinfo, read_io_ops, blks_read, write_io_ops, blks_written)
1781 The number of context switches that the system underwent.
1783 \fIbtime 769041601\fP
1784 boot time, in seconds since the Epoch, 1970-01-01 00:00:00 +0000 (UTC).
1786 \fIprocesses 86031\fP
1787 Number of forks since boot.
1789 \fIprocs_running 6\fP
1790 Number of processes in runnable state.
1791 (Linux 2.5.45 onward.)
1793 \fIprocs_blocked 2\fP
1794 Number of processes blocked waiting for I/O to complete.
1795 (Linux 2.5.45 onward.)
1804 This directory (present since 1.3.57) contains a number of files
1805 and subdirectories corresponding to kernel variables.
1806 These variables can be read and sometimes modified using
1807 the \fI/proc\fP file system, and the (deprecated)
1811 .IR /proc/sys/abi " (since Linux 2.4.10)"
1812 This directory may contain files with application binary information.
1813 .\" On some systems, it is not present.
1814 See the Linux kernel source file
1815 .I Documentation/sysctl/abi.txt
1816 for more information.
1819 This directory may be empty.
1822 This directory contains device-specific information (e.g.,
1823 .IR dev/cdrom/info ).
1825 some systems, it may be empty.
1828 This directory contains the files and subdirectories for kernel variables
1829 related to file systems.
1831 .I /proc/sys/fs/binfmt_misc
1832 Documentation for files in this directory can be found
1833 in the Linux kernel sources in
1834 .IR Documentation/binfmt_misc.txt .
1836 .IR /proc/sys/fs/dentry-state " (since Linux 2.2)"
1837 This file contains information about the status of the
1838 directory cache (dcache).
1839 The file contains six numbers,
1840 .IR nr_dentry ", " nr_unused ", " age_limit " (age in seconds), "
1842 (pages requested by system) and two dummy values.
1846 is the number of allocated dentries (dcache entries).
1847 This field is unused in Linux 2.2.
1850 is the number of unused dentries.
1853 .\" looks like this is unused in kernels 2.2 to 2.6
1854 is the age in seconds after which dcache entries
1855 can be reclaimed when memory is short.
1858 .\" looks like this is unused in kernels 2.2 to 2.6
1859 is nonzero when the kernel has called shrink_dcache_pages() and the
1860 dcache isn't pruned yet.
1863 .I /proc/sys/fs/dir-notify-enable
1864 This file can be used to disable or enable the
1866 interface described in
1868 on a system-wide basis.
1869 A value of 0 in this file disables the interface,
1870 and a value of 1 enables it.
1872 .I /proc/sys/fs/dquot-max
1873 This file shows the maximum number of cached disk quota entries.
1874 On some (2.4) systems, it is not present.
1875 If the number of free cached disk quota entries is very low and
1876 you have some awesome number of simultaneous system users,
1877 you might want to raise the limit.
1879 .I /proc/sys/fs/dquot-nr
1880 This file shows the number of allocated disk quota
1881 entries and the number of free disk quota entries.
1883 .IR /proc/sys/fs/epoll " (since Linux 2.6.28)"
1884 This directory contains the file
1885 .IR max_user_watches ,
1886 which can be used to limit the amount of kernel memory consumed by the
1889 For further details, see
1892 .I /proc/sys/fs/file-max
1894 a system-wide limit on the number of open files for all processes.
1897 which can be used by a process to set the per-process limit,
1899 on the number of files it may open.)
1901 of error messages about running out of file handles,
1902 try increasing this value:
1908 echo 100000 > /proc/sys/fs/file-max
1914 imposes an upper limit on the value that may be placed in
1918 .IR /proc/sys/fs/file-max ","
1920 .I /proc/sys/fs/inode-max
1921 to 3-4 times the new
1923 .IR /proc/sys/fs/file-max ","
1924 or you will run out of inodes.
1926 Privileged processes
1927 .RB ( CAP_SYS_ADMIN )
1932 .I /proc/sys/fs/file-nr
1933 This (read-only) file gives the number of files presently opened.
1934 It contains three numbers: the number of allocated file handles;
1935 the number of free file handles; and the maximum number of file handles.
1936 The kernel allocates file handles dynamically, but it
1937 doesn't free them again.
1938 If the number of allocated files is close to the
1939 maximum, you should consider increasing the maximum.
1940 When the number of free file handles is
1941 large, you've encountered a peak in your usage of file
1942 handles and you probably don't need to increase the maximum.
1944 .I /proc/sys/fs/inode-max
1945 This file contains the maximum number of in-memory inodes.
1946 On some (2.4) systems, it may not be present.
1947 This value should be 3-4 times larger
1950 since \fIstdin\fP, \fIstdout\fP
1951 and network sockets also need an inode to handle them.
1952 When you regularly run out of inodes, you need to increase this value.
1954 .I /proc/sys/fs/inode-nr
1955 This file contains the first two values from
1958 .I /proc/sys/fs/inode-state
1960 contains seven numbers:
1962 .IR nr_free_inodes ,
1964 and four dummy values.
1966 is the number of inodes the system has allocated.
1967 This can be slightly more than
1969 because Linux allocates them one page full at a time.
1971 represents the number of free inodes.
1977 and the system needs to prune the inode list instead of allocating more.
1979 .IR /proc/sys/fs/inotify " (since Linux 2.6.13)"
1980 This directory contains files
1981 .IR max_queued_events ", " max_user_instances ", and " max_user_watches ,
1982 that can be used to limit the amount of kernel memory consumed by the
1985 For further details, see
1988 .I /proc/sys/fs/lease-break-time
1989 This file specifies the grace period that the kernel grants to a process
1990 holding a file lease
1992 after it has sent a signal to that process notifying it
1993 that another process is waiting to open the file.
1994 If the lease holder does not remove or downgrade the lease within
1995 this grace period, the kernel forcibly breaks the lease.
1997 .I /proc/sys/fs/leases-enable
1998 This file can be used to enable or disable file leases
2000 on a system-wide basis.
2001 If this file contains the value 0, leases are disabled.
2002 A nonzero value enables leases.
2004 .IR /proc/sys/fs/mqueue " (since Linux 2.6.6)"
2005 This directory contains files
2006 .IR msg_max ", " msgsize_max ", and " queues_max ,
2007 controlling the resources used by POSIX message queues.
2012 .IR /proc/sys/fs/overflowgid " and " /proc/sys/fs/overflowuid
2014 allow you to change the value of the fixed UID and GID.
2015 The default is 65534.
2016 Some file systems only support 16-bit UIDs and GIDs, although in Linux
2017 UIDs and GIDs are 32 bits.
2018 When one of these file systems is mounted
2019 with writes enabled, any UID or GID that would exceed 65535 is translated
2020 to the overflow value before being written to disk.
2022 .IR /proc/sys/fs/pipe-max-size " (since Linux 2.6.35)"
2023 The value in this file defines an upper limit for raising the capacity
2028 This limit applies only to unprivileged processes.
2029 The default value for this file is 1,048,576.
2030 The value assigned to this file may be rounded upward,
2031 to reflect the value actually employed for a convenient implementation.
2032 To determine the rounded-up value,
2033 display the contents of this file after assigning a value to it.
2034 The minimum value that can be assigned to this file is the system page size.
2036 .IR /proc/sys/fs/suid_dumpable " (since Linux 2.6.13)"
2037 .\" The following is based on text from Documentation/sysctl/kernel.txt
2038 The value in this file determines whether core dump files are
2039 produced for set-user-ID or otherwise protected/tainted binaries.
2040 Three different integer values can be specified:
2043 This provides the traditional (pre-Linux 2.6.13) behavior.
2044 A core dump will not be produced for a process which has
2045 changed credentials (by calling
2048 or similar, or by executing a set-user-ID or set-group-ID program)
2049 or whose binary does not have read permission enabled.
2052 All processes dump core when possible.
2053 The core dump is owned by the file system user ID of the dumping process
2054 and no security is applied.
2055 This is intended for system debugging situations only.
2056 Ptrace is unchecked.
2058 \fI2\ ("suidsafe")\fP
2059 Any binary which normally would not be dumped (see "0" above)
2060 is dumped readable by root only.
2061 This allows the user to remove the core dump file but not to read it.
2062 For security reasons core dumps in this mode will not overwrite one
2063 another or other files.
2064 This mode is appropriate when administrators are
2065 attempting to debug problems in a normal environment.
2067 .I /proc/sys/fs/super-max
2069 controls the maximum number of superblocks, and
2070 thus the maximum number of mounted file systems the kernel
2072 You only need to increase
2074 if you need to mount more file systems than the current value in
2078 .I /proc/sys/fs/super-nr
2080 contains the number of file systems currently mounted.
2083 This directory contains files controlling a range of kernel parameters,
2086 .I /proc/sys/kernel/acct
2088 contains three numbers:
2093 If BSD-style process accounting is enabled these values control
2095 If free space on file system where the log lives goes below
2097 percent accounting suspends.
2098 If free space gets above
2100 percent accounting resumes.
2103 how often the kernel checks the amount of free space (value is in
2105 Default values are 4, 2 and 30.
2106 That is, suspend accounting if 2% or less space is free; resume it
2107 if 4% or more space is free; consider information about amount of free space
2108 valid for 30 seconds.
2110 .IR /proc/sys/kernel/cap-bound " (from Linux 2.2 to 2.6.24)"
2111 This file holds the value of the kernel
2112 .I "capability bounding set"
2113 (expressed as a signed decimal number).
2114 This set is ANDed against the capabilities permitted to a process
2117 Starting with Linux 2.6.25,
2118 the system-wide capability bounding set disappeared,
2119 and was replaced by a per-thread bounding set; see
2120 .BR capabilities (7).
2122 .I /proc/sys/kernel/core_pattern
2126 .I /proc/sys/kernel/core_uses_pid
2130 .I /proc/sys/kernel/ctrl-alt-del
2132 controls the handling of Ctrl-Alt-Del from the keyboard.
2133 When the value in this file is 0, Ctrl-Alt-Del is trapped and
2136 program to handle a graceful restart.
2137 When the value is greater than zero, Linux's reaction to a Vulcan
2138 Nerve Pinch (tm) will be an immediate reboot, without even
2139 syncing its dirty buffers.
2140 Note: when a program (like dosemu) has the keyboard in "raw"
2141 mode, the ctrl-alt-del is intercepted by the program before it
2142 ever reaches the kernel tty layer, and it's up to the program
2143 to decide what to do with it.
2145 .I /proc/sys/kernel/hotplug
2147 contains the path for the hotplug policy agent.
2148 The default value in this file is
2151 .IR /proc/sys/kernel/domainname " and " /proc/sys/kernel/hostname
2152 can be used to set the NIS/YP domainname and the
2153 hostname of your box in exactly the same way as the commands
2161 .RB "#" " echo \(aqdarkstar\(aq > /proc/sys/kernel/hostname"
2162 .RB "#" " echo \(aqmydomain\(aq > /proc/sys/kernel/domainname"
2166 has the same effect as
2170 .RB "#" " hostname \(aqdarkstar\(aq"
2171 .RB "#" " domainname \(aqmydomain\(aq"
2175 Note, however, that the classic darkstar.frop.org has the
2176 hostname "darkstar" and DNS (Internet Domain Name Server)
2177 domainname "frop.org", not to be confused with the NIS (Network
2178 Information Service) or YP (Yellow Pages) domainname.
2180 domain names are in general different.
2181 For a detailed discussion
2186 .I /proc/sys/kernel/htab-reclaim
2187 (PowerPC only) If this file is set to a nonzero value,
2190 .IR Documentation/powerpc/ppc_htab.txt )
2192 each time the system hits the idle loop.
2194 .I /proc/sys/kernel/l2cr
2195 (PowerPC only) This file
2196 contains a flag that controls the L2 cache of G3 processor
2198 If 0, the cache is disabled.
2201 .I /proc/sys/kernel/modprobe
2202 This file contains the path for the kernel module loader.
2203 The default value is
2204 .IR /sbin/modprobe .
2205 The file is only present if the kernel is built with the
2208 It is described by the Linux kernel source file
2209 .I Documentation/kmod.txt
2210 (only present in kernel 2.4 and earlier).
2212 .I /proc/sys/kernel/msgmax
2214 a system-wide limit specifying the maximum number of bytes in
2215 a single message written on a System V message queue.
2217 .I /proc/sys/kernel/msgmni
2218 This file defines the system-wide limit on the number of
2219 message queue identifiers.
2220 (This file is only present in Linux 2.4 onward.)
2222 .I /proc/sys/kernel/msgmnb
2223 This file defines a system-wide parameter used to initialize the
2225 setting for subsequently created message queues.
2228 setting specifies the maximum number of bytes that may be written to the
2231 .IR /proc/sys/kernel/ostype " and " /proc/sys/kernel/osrelease
2236 .IR /proc/sys/kernel/overflowgid " and " /proc/sys/kernel/overflowuid
2237 These files duplicate the files
2238 .I /proc/sys/fs/overflowgid
2240 .IR /proc/sys/fs/overflowuid .
2242 .I /proc/sys/kernel/panic
2243 This file gives read/write access to the kernel variable
2245 If this is zero, the kernel will loop on a panic; if nonzero
2246 it indicates that the kernel should autoreboot after this number
2249 software watchdog device driver, the recommended setting is 60.
2251 .IR /proc/sys/kernel/panic_on_oops " (since Linux 2.5.68)"
2252 This file controls the kernel's behavior when an oops
2253 or BUG is encountered.
2254 If this file contains 0, then the system
2255 tries to continue operation.
2256 If it contains 1, then the system
2257 delays a few seconds (to give klogd time to record the oops output)
2260 .I /proc/sys/kernel/panic
2261 file is also nonzero then the machine will be rebooted.
2263 .IR /proc/sys/kernel/pid_max " (since Linux 2.5.34)"
2264 This file specifies the value at which PIDs wrap around
2265 (i.e., the value in this file is one greater than the maximum PID).
2266 The default value for this file, 32768,
2267 results in the same range of PIDs as on earlier kernels.
2268 On 32-bit platforms, 32768 is the maximum value for
2272 can be set to any value up to 2^22
2273 .RB ( PID_MAX_LIMIT ,
2274 approximately 4 million).
2275 .\" Prior to 2.6.10, pid_max could also be raised above 32768 on 32-bit
2276 .\" platforms, but this broke /proc/[pid]
2277 .\" See http://marc.theaimsgroup.com/?l=linux-kernel&m=109513010926152&w=2
2279 .IR /proc/sys/kernel/powersave-nap " (PowerPC only)"
2280 This file contains a flag.
2281 If set, Linux-PPC will use the "nap" mode of
2283 otherwise the "doze" mode will be used.
2285 .I /proc/sys/kernel/printk
2286 The four values in this file are
2287 .IR console_loglevel ,
2288 .IR default_message_loglevel ,
2289 .IR minimum_console_level ,
2291 .IR default_console_loglevel .
2292 These values influence
2294 behavior when printing or logging error messages.
2297 for more info on the different loglevels.
2298 Messages with a higher priority than
2300 will be printed to the console.
2301 Messages without an explicit priority will be printed with priority
2302 .IR default_message_level .
2303 .I minimum_console_loglevel
2304 is the minimum (highest) value to which
2307 .I default_console_loglevel
2308 is the default value for
2309 .IR console_loglevel .
2311 .IR /proc/sys/kernel/pty " (since Linux 2.6.4)"
2312 This directory contains two files relating to the number of UNIX 98
2313 pseudoterminals (see
2317 .I /proc/sys/kernel/pty/max
2318 This file defines the maximum number of pseudoterminals.
2320 .I /proc/sys/kernel/pty/nr
2322 indicates how many pseudoterminals are currently in use.
2324 .I /proc/sys/kernel/random
2326 contains various parameters controlling the operation of the file
2330 for further information.
2332 .I /proc/sys/kernel/real-root-dev
2333 This file is documented in the Linux kernel source file
2334 .IR Documentation/initrd.txt .
2336 .IR /proc/sys/kernel/reboot-cmd " (Sparc only) "
2337 This file seems to be a way to give an argument to the SPARC
2338 ROM/Flash boot loader.
2339 Maybe to tell it what to do after
2342 .I /proc/sys/kernel/rtsig-max
2343 (Only in kernels up to and including 2.6.7; see
2345 This file can be used to tune the maximum number
2346 of POSIX real-time (queued) signals that can be outstanding
2349 .I /proc/sys/kernel/rtsig-nr
2350 (Only in kernels up to and including 2.6.7.)
2351 This file shows the number POSIX real-time signals currently queued.
2353 .IR /proc/sys/kernel/sem " (since Linux 2.4)"
2354 This file contains 4 numbers defining limits for System V IPC semaphores.
2355 These fields are, in order:
2358 The maximum semaphores per semaphore set.
2360 A system-wide limit on the number of semaphores in all semaphore sets.
2362 The maximum number of operations that may be specified in a
2366 A system-wide limit on the maximum number of semaphore identifiers.
2369 .I /proc/sys/kernel/sg-big-buff
2371 shows the size of the generic SCSI device (sg) buffer.
2372 You can't tune it just yet, but you could change it at
2373 compile time by editing
2374 .I include/scsi/sg.h
2378 However, there shouldn't be any reason to change this value.
2380 .I /proc/sys/kernel/shmall
2382 contains the system-wide limit on the total number of pages of
2383 System V shared memory.
2385 .I /proc/sys/kernel/shmmax
2387 can be used to query and set the run-time limit
2388 on the maximum (System V IPC) shared memory segment size that can be
2390 Shared memory segments up to 1GB are now supported in the
2392 This value defaults to
2395 .I /proc/sys/kernel/shmmni
2396 (available in Linux 2.4 and onward)
2398 specifies the system-wide maximum number of System V shared memory
2399 segments that can be created.
2401 .I /proc/sys/kernel/sysrq
2402 This file controls the functions allowed to be invoked by the SysRq key.
2404 the file contains 1 meaning that every possible SysRq request is allowed
2405 (in older kernel versions, SysRq was disabled by default,
2406 and you were required to specifically enable it at run-time,
2407 but this is not the case any more).
2408 Possible values in this file are:
2410 0 - disable sysrq completely
2411 1 - enable all functions of sysrq
2412 >1 - bitmask of allowed sysrq functions, as follows:
2413 2 - enable control of console logging level
2414 4 - enable control of keyboard (SAK, unraw)
2415 8 - enable debugging dumps of processes etc.
2416 16 - enable sync command
2417 32 - enable remount read-only
2418 64 - enable signalling of processes (term, kill, oom-kill)
2419 128 - allow reboot/poweroff
2420 256 - allow nicing of all real-time tasks
2422 This file is only present if the
2423 .B CONFIG_MAGIC_SYSRQ
2424 kernel configuration option is enabled.
2425 For further details see the Linux kernel source file
2426 .IR Documentation/sysrq.txt .
2428 .I /proc/sys/kernel/version
2429 This file contains a string like:
2431 #5 Wed Feb 25 21:49:24 MET 1998
2434 this is the fifth kernel built from this source base and the
2435 date behind it indicates the time the kernel was built.
2437 .IR /proc/sys/kernel/threads-max " (since Linux 2.3.11)"
2438 This file specifies the system-wide limit on the number of
2439 threads (tasks) that can be created on the system.
2441 .IR /proc/sys/kernel/zero-paged " (PowerPC only) "
2444 When enabled (nonzero), Linux-PPC will pre-zero pages in
2445 the idle loop, possibly speeding up get_free_pages.
2448 This directory contains networking stuff.
2449 Explanations for some of the files under this directory can be found in
2454 .I /proc/sys/net/core/somaxconn
2455 This file defines a ceiling value for the
2461 manual page for details.
2464 This directory may be empty.
2467 This directory supports Sun remote procedure call for network file system
2469 On some systems, it is not present.
2472 This directory contains files for memory management tuning, buffer and
2475 .IR /proc/sys/vm/drop_caches " (since Linux 2.6.16)"
2476 Writing to this file causes the kernel to drop clean caches, dentries and
2477 inodes from memory, causing that memory to become free.
2479 To free pagecache, use
2480 .IR "echo 1 > /proc/sys/vm/drop_caches" ;
2481 to free dentries and inodes, use
2482 .IR "echo 2 > /proc/sys/vm/drop_caches" ;
2483 to free pagecache, dentries and inodes, use
2484 .IR "echo 3 > /proc/sys/vm/drop_caches" .
2486 Because this is a nondestructive operation and dirty objects
2487 are not freeable, the
2492 .IR /proc/sys/vm/legacy_va_layout " (since Linux 2.6.9)"
2493 .\" The following is from Documentation/filesystems/proc.txt
2494 If nonzero, this disables the new 32-bit memory-mapping layout;
2495 the kernel will use the legacy (2.4) layout for all processes.
2497 .IR /proc/sys/vm/memory_failure_early_kill " (since Linux 2.6.32)"
2498 .\" The following is based on the text in Documentation/sysctl/vm.txt
2499 Control how to kill processes when an uncorrected memory error
2500 (typically a 2-bit error in a memory module)
2501 that cannot be handled by the kernel
2502 is detected in the background by hardware.
2503 In some cases (like the page still having a valid copy on disk),
2504 the kernel will handle the failure
2505 transparently without affecting any applications.
2506 But if there is no other up-to-date copy of the data,
2507 it will kill processes to prevent any data corruptions from propagating.
2509 The file has one of the following values:
2512 Kill all processes that have the corrupted-and-not-reloadable page mapped
2513 as soon as the corruption is detected.
2514 Note this is not supported for a few types of pages, like kernel internally
2515 allocated data or the swap cache, but works for the majority of user pages.
2517 Only unmap the corrupted page from all processes and only kill a process
2518 who tries to access it.
2521 The kill is performed using a
2527 Processes can handle this if they want to; see
2531 This feature is only active on architectures/platforms with advanced machine
2532 check handling and depends on the hardware capabilities.
2534 Applications can override the
2535 .I memory_failure_early_kill
2536 setting individually with the
2541 Only present if the kernel was configured with
2542 .BR CONFIG_MEMORY_FAILURE .
2544 .IR /proc/sys/vm/memory_failure_recovery " (since Linux 2.6.32)"
2545 .\" The following is based on the text in Documentation/sysctl/vm.txt
2546 Enable memory failure recovery (when supported by the platform)
2551 Always panic on a memory failure.
2554 Only present if the kernel was configured with
2555 .BR CONFIG_MEMORY_FAILURE .
2557 .IR /proc/sys/vm/oom_dump_tasks " (since Linux 2.6.25)"
2558 .\" The following is from Documentation/sysctl/vm.txt
2559 Enables a system-wide task dump (excluding kernel threads) to be
2560 produced when the kernel performs an OOM-killing.
2561 The dump includes the following information
2562 for each task (thread, process):
2563 thread ID, real user ID, thread group ID (process ID),
2564 virtual memory size, resident set size,
2565 the CPU that the task is scheduled on,
2566 oom_adj score (see the description of
2567 .IR /proc/[pid]/oom_adj ),
2569 This is helpful to determine why the OOM-killer was invoked
2570 and to identify the rogue task that caused it.
2572 If this contains the value zero, this information is suppressed.
2573 On very large systems with thousands of tasks,
2574 it may not be feasible to dump the memory state information for each one.
2575 Such systems should not be forced to incur a performance penalty in
2576 OOM situations when the information may not be desired.
2578 If this is set to nonzero, this information is shown whenever the
2579 OOM-killer actually kills a memory-hogging task.
2581 The default value is 0.
2583 .IR /proc/sys/vm/oom_kill_allocating_task " (since Linux 2.6.24)"
2584 .\" The following is from Documentation/sysctl/vm.txt
2585 This enables or disables killing the OOM-triggering task in
2586 out-of-memory situations.
2588 If this is set to zero, the OOM-killer will scan through the entire
2589 tasklist and select a task based on heuristics to kill.
2590 This normally selects a rogue memory-hogging task that
2591 frees up a large amount of memory when killed.
2593 If this is set to nonzero, the OOM-killer simply kills the task that
2594 triggered the out-of-memory condition.
2595 This avoids a possibly expensive tasklist scan.
2598 .I /proc/sys/vm/panic_on_oom
2599 is nonzero, it takes precedence over whatever value is used in
2600 .IR /proc/sys/vm/oom_kill_allocating_task .
2602 The default value is 0.
2604 .I /proc/sys/vm/overcommit_memory
2605 This file contains the kernel virtual memory accounting mode.
2609 0: heuristic overcommit (this is the default)
2611 1: always overcommit, never check
2613 2: always check, never overcommit
2620 are not checked, and the default check is very weak,
2621 leading to the risk of getting a process "OOM-killed".
2622 Under Linux 2.4 any nonzero value implies mode 1.
2623 In mode 2 (available since Linux 2.6), the total virtual address space
2624 on the system is limited to (SS + RAM*(r/100)),
2625 where SS is the size of the swap space, and RAM
2626 is the size of the physical memory, and r is the contents of the file
2627 .IR /proc/sys/vm/overcommit_ratio .
2629 .I /proc/sys/vm/overcommit_ratio
2630 See the description of
2631 .IR /proc/sys/vm/overcommit_memory .
2633 .IR /proc/sys/vm/panic_on_oom " (since Linux 2.6.18)"
2634 .\" The following is adapted from Documentation/sysctl/vm.txt
2635 This enables or disables a kernel panic in
2636 an out-of-memory situation.
2638 If this file is set to the value 0,
2639 the kernel's OOM-killer will kill some rogue process.
2640 Usually, the OOM-killer is able to kill a rogue process and the
2641 system will survive.
2643 If this file is set to the value 1,
2644 then the kernel normally panics when out-of-memory happens.
2645 However, if a process limits allocations to certain nodes
2646 using memory policies
2651 and those nodes reach memory exhaustion status,
2652 one process may be killed by the OOM-killer.
2653 No panic occurs in this case:
2654 because other nodes' memory may be free,
2655 this means the system as a whole may not have reached
2656 an out-of-memory situation yet.
2658 If this file is set to the value 2,
2659 the kernel always panics when an out-of-memory condition occurs.
2661 The default value is 0.
2662 1 and 2 are for failover of clustering.
2663 Select either according to your policy of failover.
2665 .IR /proc/sys/vm/swappiness
2666 .\" The following is from Documentation/sysctl/vm.txt
2667 The value in this file controls how aggressively the kernel will swap
2669 Higher values increase aggressiveness, lower values
2670 decrease aggressiveness.
2671 The default value is 60.
2673 .IR /proc/sysrq-trigger " (since Linux 2.4.21)"
2674 Writing a character to this file triggers the same SysRq function as
2675 typing ALT-SysRq-<character> (see the description of
2676 .IR /proc/sys/kernel/sysrq ).
2677 This file is normally only writable by
2679 For further details see the Linux kernel source file
2680 .IR Documentation/sysrq.txt .
2683 Subdirectory containing the pseudo-files
2684 .IR msg ", " sem " and " shm "."
2685 These files list the System V Interprocess Communication (IPC) objects
2686 (respectively: message queues, semaphores, and shared memory)
2687 that currently exist on the system,
2688 providing similar information to that available via
2690 These files have headers and are formatted (one IPC object per line)
2691 for easy understanding.
2693 provides further background on the information shown by these files.
2696 Subdirectory containing the pseudo-files and subdirectories for
2697 tty drivers and line disciplines.
2700 This file contains two numbers: the uptime of the system (seconds),
2701 and the amount of time spent in idle process (seconds).
2704 This string identifies the kernel version that is currently running.
2705 It includes the contents of
2706 .IR /proc/sys/kernel/ostype ,
2707 .I /proc/sys/kernel/osrelease
2709 .IR /proc/sys/kernel/version .
2714 Linux version 1.0.9 (quinlan@phaze) #1 Sat May 14 01:51:54 EDT 1994
2718 .\" FIXME Document /proc/timer_list
2720 .\" .IR /proc/timer_list " (since Linux 2.6.21)"
2721 .\" See the 2.6.21 Change log
2722 .\" FIXME Document /proc/timer_stats
2724 .\" .IR /proc/timer_stats " (since Linux 2.6.21)"
2725 .\" See the 2.6.21 Change log
2727 .IR /proc/vmstat " (since Linux 2.6)"
2728 This file displays various virtual memory statistics.
2730 .IR /proc/zoneinfo " (since Linux 2.6.13)"
2731 This file display information about memory zones.
2732 This is useful for analyzing virtual memory behavior.
2733 .\" FIXME more should be said about /proc/zoneinfo
2735 Many strings (i.e., the environment and command line) are in
2736 the internal format, with subfields terminated by null bytes (\(aq\\0\(aq),
2738 may find that things are more readable if you use \fIod \-c\fP or \fItr
2739 "\\000" "\\n"\fP to read them.
2740 Alternatively, \fIecho \`cat <file>\`\fP works well.
2742 This manual page is incomplete, possibly inaccurate, and is the kind
2743 of thing that needs to be updated very often.
2744 .\" .SH ACKNOWLEDGEMENTS
2745 .\" The material on /proc/sys/fs and /proc/sys/kernel is closely based on
2746 .\" kernel source documentation files written by Rik van Riel.
2773 The Linux kernel source files:
2774 .IR Documentation/filesystems/proc.txt
2776 .IR Documentation/sysctl/vm.txt .