1 .\" Copyright (c) 2002 by Michael Kerrisk <mtk.manpages@gmail.com>
3 .\" %%%LICENSE_START(VERBATIM)
4 .\" Permission is granted to make and distribute verbatim copies of this
5 .\" manual provided the copyright notice and this permission notice are
6 .\" preserved on all copies.
8 .\" Permission is granted to copy and distribute modified versions of this
9 .\" manual under the conditions for verbatim copying, provided that the
10 .\" entire resulting derived work is distributed under the terms of a
11 .\" permission notice identical to this one.
13 .\" Since the Linux kernel and libraries are constantly changing, this
14 .\" manual page may be incorrect or out-of-date. The author(s) assume no
15 .\" responsibility for errors or omissions, or for damages resulting from
16 .\" the use of the information contained herein. The author(s) may not
17 .\" have taken the same level of care in the production of this manual,
18 .\" which is licensed free of charge, as they might when working
21 .\" Formatted or processed versions of this manual, if unaccompanied by
22 .\" the source, must acknowledge the copyright and authors of this work.
25 .\" 6 Aug 2002 - Initial Creation
26 .\" Modified 2003-05-23, Michael Kerrisk, <mtk.manpages@gmail.com>
27 .\" Modified 2004-05-27, Michael Kerrisk, <mtk.manpages@gmail.com>
28 .\" 2004-12-08, mtk Added O_NOATIME for CAP_FOWNER
29 .\" 2005-08-16, mtk, Added CAP_AUDIT_CONTROL and CAP_AUDIT_WRITE
30 .\" 2008-07-15, Serge Hallyn <serue@us.bbm.com>
31 .\" Document file capabilities, per-process capability
32 .\" bounding set, changed semantics for CAP_SETPCAP,
33 .\" and other changes in 2.6.2[45].
34 .\" Add CAP_MAC_ADMIN, CAP_MAC_OVERRIDE, CAP_SETFCAP.
36 .\" Add text describing circumstances in which CAP_SETPCAP
37 .\" (theoretically) permits a thread to change the
38 .\" capability sets of another thread.
39 .\" Add section describing rules for programmatically
40 .\" adjusting thread capability sets.
41 .\" Describe rationale for capability bounding set.
42 .\" Document "securebits" flags.
43 .\" Add text noting that if we set the effective flag for one file
44 .\" capability, then we must also set the effective flag for all
45 .\" other capabilities where the permitted or inheritable bit is set.
46 .\" 2011-09-07, mtk/Serge hallyn: Add CAP_SYSLOG
48 .TH CAPABILITIES 7 2014-09-21 "Linux" "Linux Programmer's Manual"
50 capabilities \- overview of Linux capabilities
52 For the purpose of performing permission checks,
53 traditional UNIX implementations distinguish two categories of processes:
55 processes (whose effective user ID is 0, referred to as superuser or root),
58 processes (whose effective UID is nonzero).
59 Privileged processes bypass all kernel permission checks,
60 while unprivileged processes are subject to full permission
61 checking based on the process's credentials
62 (usually: effective UID, effective GID, and supplementary group list).
64 Starting with kernel 2.2, Linux divides the privileges traditionally
65 associated with superuser into distinct units, known as
67 which can be independently enabled and disabled.
68 Capabilities are a per-thread attribute.
71 The following list shows the capabilities implemented on Linux,
72 and the operations or behaviors that each capability permits:
74 .BR CAP_AUDIT_CONTROL " (since Linux 2.6.11)"
75 Enable and disable kernel auditing; change auditing filter rules;
76 retrieve auditing status and filtering rules.
78 .BR CAP_AUDIT_READ " (since Linux 3.16)"
79 .\" commit a29b694aa1739f9d76538e34ae25524f9c549d59
80 .\" commit 3a101b8de0d39403b2c7e5c23fd0b005668acf48
81 Allow reading the audit log via a multicast netlink socket.
83 .BR CAP_AUDIT_WRITE " (since Linux 2.6.11)"
84 Write records to kernel auditing log.
86 .BR CAP_BLOCK_SUSPEND " (since Linux 3.5)"
87 Employ features that can block system suspend
90 .IR /proc/sys/wake_lock ).
93 Make arbitrary changes to file UIDs and GIDs (see
97 Bypass file read, write, and execute permission checks.
98 (DAC is an abbreviation of "discretionary access control".)
100 .B CAP_DAC_READ_SEARCH
104 Bypass file read permission checks and
105 directory read and execute permission checks;
108 .BR open_by_handle_at (2).
117 Bypass permission checks on operations that normally
118 require the filesystem UID of the process to match the UID of
122 excluding those operations covered by
125 .BR CAP_DAC_READ_SEARCH ;
127 set extended file attributes (see
131 set Access Control Lists (ACLs) on arbitrary files;
133 ignore directory sticky bit on file deletion;
137 for arbitrary files in
145 Don't clear set-user-ID and set-group-ID permission
146 bits when a file is modified;
147 set the set-group-ID bit for a file whose GID does not match
148 the filesystem or any of the supplementary GIDs of the calling process.
151 .\" FIXME . As at Linux 3.2, there are some strange uses of this capability
152 .\" in other places; they probably should be replaced with something else.
160 Bypass permission checks for operations on System V IPC objects.
163 Bypass permission checks for sending signals (see
165 This includes use of the
169 .\" FIXME . CAP_KILL also has an effect for threads + setting child
170 .\" termination signal to other than SIGCHLD: without this
171 .\" capability, the termination signal reverts to SIGCHLD
172 .\" if the child does an exec(). What is the rationale
175 .BR CAP_LEASE " (since Linux 2.4)"
176 Establish leases on arbitrary files (see
179 .B CAP_LINUX_IMMUTABLE
184 .\" These attributes are now available on ext2, ext3, Reiserfs, XFS, JFS
188 .BR CAP_MAC_ADMIN " (since Linux 2.6.25)"
189 Override Mandatory Access Control (MAC).
190 Implemented for the Smack Linux Security Module (LSM).
192 .BR CAP_MAC_OVERRIDE " (since Linux 2.6.25)"
193 Allow MAC configuration or state changes.
194 Implemented for the Smack LSM.
196 .BR CAP_MKNOD " (since Linux 2.4)"
197 Create special files using
201 Perform various network-related operations:
205 interface configuration;
207 administration of IP firewall, masquerading, and accounting;
209 modify routing tables;
211 bind to any address for transparent proxying;
213 set type-of-service (TOS)
215 clear driver statistics;
217 set promiscuous mode;
219 enabling multicasting;
223 to set the following socket options:
227 (for a priority outside the range 0 to 6),
234 .B CAP_NET_BIND_SERVICE
235 Bind a socket to Internet domain privileged ports
236 (port numbers less than 1024).
239 (Unused) Make socket broadcasts, and listen to multicasts.
245 use RAW and PACKET sockets;
247 bind to any address for transparent proxying.
250 .\" Also various IP options and setsockopt(SO_BINDTODEVICE)
253 Make arbitrary manipulations of process GIDs and supplementary GID list;
254 forge GID when passing socket credentials via UNIX domain sockets;
255 write a group ID mapping in a user namespace (see
256 .BR user_namespaces (7)).
258 .BR CAP_SETFCAP " (since Linux 2.6.24)"
259 Set file capabilities.
262 If file capabilities are not supported:
263 grant or remove any capability in the
264 caller's permitted capability set to or from any other process.
267 is not available when the kernel is configured to support
268 file capabilities, since
270 has entirely different semantics for such kernels.)
272 If file capabilities are supported:
273 add any capability from the calling thread's bounding set
274 to its inheritable set;
275 drop capabilities from the bounding set (via
277 .BR PR_CAPBSET_DROP );
283 Make arbitrary manipulations of process UIDs
288 forge UID when passing socket credentials via UNIX domain sockets;
289 write a user ID mapping in a user namespace (see
290 .BR user_namespaces (7)).
291 .\" FIXME CAP_SETUID also an effect in exec(); document this.
297 Perform a range of system administration operations including:
305 .BR setdomainname (2);
309 operations (since Linux 2.6.37,
311 should be used to permit such operations);
322 operations on arbitrary System V IPC objects;
328 perform operations on
332 Extended Attributes (see
336 .BR lookup_dcookie (2);
342 and (before Linux 2.6.25)
344 I/O scheduling classes;
346 forge PID when passing socket credentials via UNIX domain sockets;
349 .IR /proc/sys/fs/file-max ,
350 the system-wide limit on the number of open files,
351 in system calls that open files (e.g.,
359 flags that create new namespaces with
363 (but, since Linux 3.8,
364 creating user namespaces does not require any capability);
367 .BR perf_event_open (2);
382 .BR fanotify_init (2);
399 to insert characters into the input queue of a terminal other than
400 the caller's controlling terminal;
410 perform various privileged block-device
414 perform various privileged filesystem
418 perform administrative operations on many device drivers.
433 Load and unload kernel modules
437 .BR delete_module (2));
438 in kernels before 2.6.25:
439 drop capabilities from the system-wide capability bounding set.
445 Raise process nice value
448 and change the nice value for arbitrary processes;
450 set real-time scheduling policies for calling process,
451 and set scheduling policies and priorities for arbitrary processes
452 .RB ( sched_setscheduler (2),
453 .BR sched_setparam (2),
454 .BR shed_setattr (2));
456 set CPU affinity for arbitrary processes
457 .RB ( sched_setaffinity (2));
459 set I/O scheduling class and priority for arbitrary processes
460 .RB ( ioprio_set (2));
463 .BR migrate_pages (2)
464 to arbitrary processes and allow processes
465 to be migrated to arbitrary nodes;
466 .\" FIXME CAP_SYS_NICE also has the following effect for
467 .\" migrate_pages(2):
468 .\" do_migrate_pages(mm, &old, &new,
469 .\" capable(CAP_SYS_NICE) ? MPOL_MF_MOVE_ALL : MPOL_MF_MOVE);
474 to arbitrary processes;
493 Trace arbitrary processes using
497 .BR get_robust_list (2)
498 to arbitrary processes;
500 transfer data to or from the memory of arbitrary processes using
501 .BR process_vm_readv (2)
503 .BR process_vm_writev (2).
505 inspect processes using
514 Perform I/O port operations
527 open devices for accessing x86 model-specific registers (MSRs, see
531 .IR /proc/sys/vm/mmap_min_addr ;
533 create memory mappings at addresses below the value specified by
534 .IR /proc/sys/vm/mmap_min_addr ;
544 perform various SCSI device commands;
546 perform certain operations on
552 perform a range of device-specific operations on other devices.
560 Use reserved space on ext2 filesystems;
564 calls controlling ext3 journaling;
566 override disk quota limits;
568 increase resource limits (see
575 override maximum number of consoles on console allocation;
577 override maximum number of keymaps;
579 allow more than 64hz interrupts from the real-time clock;
583 limit for a System V message queue above the limit in
584 .I /proc/sys/kernel/msgmnb
591 .I /proc/sys/fs/pipe-size-max
592 limit when setting the capacity of a pipe using the
599 to increase the capacity of a pipe above the limit specified by
600 .IR /proc/sys/fs/pipe-max-size ;
603 .I /proc/sys/fs/mqueue/queues_max
604 limit when creating POSIX message queues (see
605 .BR mq_overview (7));
613 .IR /proc/PID/oom_score_adj
614 to a value lower than the value last set by a process with
615 .BR CAP_SYS_RESOURCE .
621 .RB ( settimeofday (2),
624 set real-time (hardware) clock.
626 .B CAP_SYS_TTY_CONFIG
629 employ various privileged
631 operations on virtual terminals.
633 .BR CAP_SYSLOG " (since Linux 2.6.37)"
642 for information on which operations require privilege.
644 View kernel addresses exposed via
646 and other interfaces when
647 .IR /proc/sys/kernel/kptr_restrict
649 (See the discussion of the
656 .BR CAP_WAKE_ALARM " (since Linux 3.0)"
657 Trigger something that will wake up the system (set
658 .B CLOCK_REALTIME_ALARM
660 .B CLOCK_BOOTTIME_ALARM
663 .SS Past and current implementation
664 A full implementation of capabilities requires that:
666 For all privileged operations,
667 the kernel must check whether the thread has the required
668 capability in its effective set.
670 The kernel must provide system calls allowing a thread's capability sets to
671 be changed and retrieved.
673 The filesystem must support attaching capabilities to an executable file,
674 so that a process gains those capabilities when the file is executed.
676 Before kernel 2.6.24, only the first two of these requirements are met;
677 since kernel 2.6.24, all three requirements are met.
679 .SS Thread capability sets
680 Each thread has three capability sets containing zero or more
681 of the above capabilities:
684 This is a limiting superset for the effective
685 capabilities that the thread may assume.
686 It is also a limiting superset for the capabilities that
687 may be added to the inheritable set by a thread that does not have the
689 capability in its effective set.
691 If a thread drops a capability from its permitted set,
692 it can never reacquire that capability (unless it
694 either a set-user-ID-root program, or
695 a program whose associated file capabilities grant that capability).
698 This is a set of capabilities preserved across an
700 It provides a mechanism for a process to assign capabilities
701 to the permitted set of the new program during an
705 This is the set of capabilities used by the kernel to
706 perform permission checks for the thread.
710 inherits copies of its parent's capability sets.
711 See below for a discussion of the treatment of capabilities during
716 a thread may manipulate its own capability sets (see below).
718 Since Linux 3.2, the file
719 .I /proc/sys/kernel/cap_last_cap
720 .\" commit 73efc0394e148d0e15583e13712637831f926720
721 exposes the numerical value of the highest capability
722 supported by the running kernel;
723 this can be used to determine the highest bit
724 that may be set in a capability set.
726 .SS File capabilities
727 Since kernel 2.6.24, the kernel supports
728 associating capability sets with an executable file using
730 The file capability sets are stored in an extended attribute (see
733 .IR "security.capability" .
734 Writing to this extended attribute requires the
737 The file capability sets,
738 in conjunction with the capability sets of the thread,
739 determine the capabilities of a thread after an
742 The three file capability sets are:
744 .IR Permitted " (formerly known as " forced ):
745 These capabilities are automatically permitted to the thread,
746 regardless of the thread's inheritable capabilities.
748 .IR Inheritable " (formerly known as " allowed ):
749 This set is ANDed with the thread's inheritable set to determine which
750 inheritable capabilities are enabled in the permitted set of
755 This is not a set, but rather just a single bit.
756 If this bit is set, then during an
758 all of the new permitted capabilities for the thread are
759 also raised in the effective set.
760 If this bit is not set, then after an
762 none of the new permitted capabilities is in the new effective set.
764 Enabling the file effective capability bit implies
765 that any file permitted or inheritable capability that causes a
766 thread to acquire the corresponding permitted capability during an
768 (see the transformation rules described below) will also acquire that
769 capability in its effective set.
770 Therefore, when assigning capabilities to a file
772 .BR cap_set_file (3),
774 if we specify the effective flag as being enabled for any capability,
775 then the effective flag must also be specified as enabled
776 for all other capabilities for which the corresponding permitted or
777 inheritable flags is enabled.
779 .SS Transformation of capabilities during execve()
783 the kernel calculates the new capabilities of
784 the process using the following algorithm:
788 P'(permitted) = (P(inheritable) & F(inheritable)) |
789 (F(permitted) & cap_bset)
791 P'(effective) = F(effective) ? P'(permitted) : 0
793 P'(inheritable) = P(inheritable) [i.e., unchanged]
800 denotes the value of a thread capability set before the
803 denotes the value of a capability set after the
806 denotes a file capability set
808 is the value of the capability bounding set (described below).
811 .SS Capabilities and execution of programs by root
812 In order to provide an all-powerful
814 using capability sets, during an
817 If a set-user-ID-root program is being executed,
818 or the real user ID of the process is 0 (root)
819 then the file inheritable and permitted sets are defined to be all ones
820 (i.e., all capabilities enabled).
822 If a set-user-ID-root program is being executed,
823 then the file effective bit is defined to be one (enabled).
825 The upshot of the above rules,
826 combined with the capabilities transformations described above,
827 is that when a process
829 a set-user-ID-root program, or when a process with an effective UID of 0
832 it gains all capabilities in its permitted and effective capability sets,
833 except those masked out by the capability bounding set.
834 .\" If a process with real UID 0, and nonzero effective UID does an
835 .\" exec(), then it gets all capabilities in its
836 .\" permitted set, and no effective capabilities
837 This provides semantics that are the same as those provided by
838 traditional UNIX systems.
839 .SS Capability bounding set
840 The capability bounding set is a security mechanism that can be used
841 to limit the capabilities that can be gained during an
843 The bounding set is used in the following ways:
847 the capability bounding set is ANDed with the file permitted
848 capability set, and the result of this operation is assigned to the
849 thread's permitted capability set.
850 The capability bounding set thus places a limit on the permitted
851 capabilities that may be granted by an executable file.
854 The capability bounding set acts as a limiting superset for
855 the capabilities that a thread can add to its inheritable set using
857 This means that if a capability is not in the bounding set,
858 then a thread can't add this capability to its
859 inheritable set, even if it was in its permitted capabilities,
860 and thereby cannot have this capability preserved in its
861 permitted set when it
863 a file that has the capability in its inheritable set.
865 Note that the bounding set masks the file permitted capabilities,
866 but not the inherited capabilities.
867 If a thread maintains a capability in its inherited set
868 that is not in its bounding set,
869 then it can still gain that capability in its permitted set
870 by executing a file that has the capability in its inherited set.
872 Depending on the kernel version, the capability bounding set is either
873 a system-wide attribute, or a per-process attribute.
875 .B "Capability bounding set prior to Linux 2.6.25"
877 In kernels before 2.6.25, the capability bounding set is a system-wide
878 attribute that affects all threads on the system.
879 The bounding set is accessible via the file
880 .IR /proc/sys/kernel/cap-bound .
881 (Confusingly, this bit mask parameter is expressed as a
882 signed decimal number in
883 .IR /proc/sys/kernel/cap-bound .)
887 process may set capabilities in the capability bounding set;
888 other than that, the superuser (more precisely: programs with the
890 capability) may only clear capabilities from this set.
892 On a standard system the capability bounding set always masks out the
895 To remove this restriction (dangerous!), modify the definition of
898 .I include/linux/capability.h
899 and rebuild the kernel.
901 The system-wide capability bounding set feature was added
902 to Linux starting with kernel version 2.2.11.
905 .B "Capability bounding set from Linux 2.6.25 onward"
907 From Linux 2.6.25, the
908 .I "capability bounding set"
909 is a per-thread attribute.
910 (There is no longer a system-wide capability bounding set.)
912 The bounding set is inherited at
914 from the thread's parent, and is preserved across an
917 A thread may remove capabilities from its capability bounding set using the
920 operation, provided it has the
923 Once a capability has been dropped from the bounding set,
924 it cannot be restored to that set.
925 A thread can determine if a capability is in its bounding set using the
930 Removing capabilities from the bounding set is supported only if file
931 capabilities are compiled into the kernel.
932 In kernels before Linux 2.6.33,
933 file capabilities were an optional feature configurable via the
934 .B CONFIG_SECURITY_FILE_CAPABILITIES
936 Since Linux 2.6.33, the configuration option has been removed
937 and file capabilities are always part of the kernel.
938 When file capabilities are compiled into the kernel, the
940 process (the ancestor of all processes) begins with a full bounding set.
941 If file capabilities are not compiled into the kernel, then
943 begins with a full bounding set minus
945 because this capability has a different meaning when there are
946 no file capabilities.
948 Removing a capability from the bounding set does not remove it
949 from the thread's inherited set.
950 However it does prevent the capability from being added
951 back into the thread's inherited set in the future.
954 .SS Effect of user ID changes on capabilities
955 To preserve the traditional semantics for transitions between
956 0 and nonzero user IDs,
957 the kernel makes the following changes to a thread's capability
958 sets on changes to the thread's real, effective, saved set,
959 and filesystem user IDs (using
964 If one or more of the real, effective or saved set user IDs
965 was previously 0, and as a result of the UID changes all of these IDs
966 have a nonzero value,
967 then all capabilities are cleared from the permitted and effective
970 If the effective user ID is changed from 0 to nonzero,
971 then all capabilities are cleared from the effective set.
973 If the effective user ID is changed from nonzero to 0,
974 then the permitted set is copied to the effective set.
976 If the filesystem user ID is changed from 0 to nonzero (see
978 then the following capabilities are cleared from the effective set:
980 .BR CAP_DAC_OVERRIDE ,
981 .BR CAP_DAC_READ_SEARCH ,
984 .B CAP_LINUX_IMMUTABLE
985 (since Linux 2.6.30),
986 .BR CAP_MAC_OVERRIDE ,
989 (since Linux 2.6.30).
990 If the filesystem UID is changed from nonzero to 0,
991 then any of these capabilities that are enabled in the permitted set
992 are enabled in the effective set.
994 If a thread that has a 0 value for one or more of its user IDs wants
995 to prevent its permitted capability set being cleared when it resets
996 all of its user IDs to nonzero values, it can do so using the
1001 .SS Programmatically adjusting capability sets
1002 A thread can retrieve and change its capability sets using the
1008 .BR cap_get_proc (3)
1010 .BR cap_set_proc (3),
1011 both provided in the
1014 is preferred for this purpose.
1015 The following rules govern changes to the thread capability sets:
1017 If the caller does not have the
1020 the new inheritable set must be a subset of the combination
1021 of the existing inheritable and permitted sets.
1023 (Since Linux 2.6.25)
1024 The new inheritable set must be a subset of the combination of the
1025 existing inheritable set and the capability bounding set.
1027 The new permitted set must be a subset of the existing permitted set
1028 (i.e., it is not possible to acquire permitted capabilities
1029 that the thread does not currently have).
1031 The new effective set must be a subset of the new permitted set.
1032 .SS The securebits flags: establishing a capabilities-only environment
1033 .\" For some background:
1034 .\" see http://lwn.net/Articles/280279/ and
1035 .\" http://article.gmane.org/gmane.linux.kernel.lsm/5476/
1036 Starting with kernel 2.6.26,
1037 and with a kernel in which file capabilities are enabled,
1038 Linux implements a set of per-thread
1040 flags that can be used to disable special handling of capabilities for UID 0
1042 These flags are as follows:
1045 Setting this flag allows a thread that has one or more 0 UIDs to retain
1046 its capabilities when it switches all of its UIDs to a nonzero value.
1047 If this flag is not set,
1048 then such a UID switch causes the thread to lose all capabilities.
1049 This flag is always cleared on an
1051 (This flag provides the same functionality as the older
1056 .B SECBIT_NO_SETUID_FIXUP
1057 Setting this flag stops the kernel from adjusting capability sets when
1058 the threads's effective and filesystem UIDs are switched between
1059 zero and nonzero values.
1061 .IR "Effect of User ID Changes on Capabilities" .)
1064 If this bit is set, then the kernel does not grant capabilities
1065 when a set-user-ID-root program is executed, or when a process with
1066 an effective or real UID of 0 calls
1069 .IR "Capabilities and execution of programs by root" .)
1071 Each of the above "base" flags has a companion "locked" flag.
1072 Setting any of the "locked" flags is irreversible,
1073 and has the effect of preventing further changes to the
1074 corresponding "base" flag.
1075 The locked flags are:
1076 .BR SECBIT_KEEP_CAPS_LOCKED ,
1077 .BR SECBIT_NO_SETUID_FIXUP_LOCKED ,
1079 .BR SECBIT_NOROOT_LOCKED .
1083 flags can be modified and retrieved using the
1085 .B PR_SET_SECUREBITS
1087 .B PR_GET_SECUREBITS
1091 capability is required to modify the flags.
1095 flags are inherited by child processes.
1098 all of the flags are preserved, except
1100 which is always cleared.
1102 An application can use the following call to lock itself,
1103 and all of its descendants,
1104 into an environment where the only way of gaining capabilities
1105 is by executing a program with associated file capabilities:
1109 prctl(PR_SET_SECUREBITS,
1110 SECBIT_KEEP_CAPS_LOCKED |
1111 SECBIT_NO_SETUID_FIXUP |
1112 SECBIT_NO_SETUID_FIXUP_LOCKED |
1114 SECBIT_NOROOT_LOCKED);
1117 .SS Interaction with user namespaces
1118 For a discussion of the interaction of capabilities and user namespaces, see
1119 .BR user_namespaces (7).
1122 No standards govern capabilities, but the Linux capability implementation
1123 is based on the withdrawn POSIX.1e draft standard; see
1124 .UR http://wt.tuxomania.net\:/publications\:/posix.1e/
1127 Since kernel 2.5.27, capabilities are an optional kernel component,
1128 and can be enabled/disabled via the
1129 .B CONFIG_SECURITY_CAPABILITIES
1130 kernel configuration option.
1133 .I /proc/PID/task/TID/status
1134 file can be used to view the capability sets of a thread.
1137 file shows the capability sets of a process's main thread.
1138 Before Linux 3.8, nonexistent capabilities were shown as being
1139 enabled (1) in these sets.
1141 .\" 7b9a7ec565505699f503b4fcf61500dceb36e744
1142 all nonexistent capabilities (above
1144 are shown as disabled (0).
1148 package provides a suite of routines for setting and
1149 getting capabilities that is more comfortable and less likely
1150 to change than the interface provided by
1154 This package also provides the
1161 .UR http://www.kernel.org\:/pub\:/linux\:/libs\:/security\:/linux-privs
1164 Before kernel 2.6.24, and since kernel 2.6.24 if
1165 file capabilities are not enabled, a thread with the
1167 capability can manipulate the capabilities of threads other than itself.
1168 However, this is only theoretically possible,
1169 since no thread ever has
1171 in either of these cases:
1173 In the pre-2.6.25 implementation the system-wide capability bounding set,
1174 .IR /proc/sys/kernel/cap-bound ,
1175 always masks out this capability, and this can not be changed
1176 without modifying the kernel source and rebuilding.
1178 If file capabilities are disabled in the current implementation, then
1180 starts out with this capability removed from its per-process bounding
1181 set, and that bounding set is inherited by all other processes
1182 created on the system.
1189 .BR cap_copy_ext (3),
1190 .BR cap_from_text (3),
1191 .BR cap_get_file (3),
1192 .BR cap_get_proc (3),
1197 .BR credentials (7),
1198 .BR user_namespaces (7),
1203 .I include/linux/capability.h
1204 in the Linux kernel source tree
1206 This page is part of release 3.78 of the Linux
1209 A description of the project,
1210 information about reporting bugs,
1211 and the latest version of this page,
1213 \%http://www.kernel.org/doc/man\-pages/.