1 ============================
2 KERNEL KEY RETENTION SERVICE
3 ============================
5 This service allows cryptographic keys, authentication tokens, cross-domain
6 user mappings, and similar to be cached in the kernel for the use of
7 filesystems and other kernel services.
9 Keyrings are permitted; these are a special type of key that can hold links to
10 other keys. Processes each have three standard keyring subscriptions that a
11 kernel service can search for relevant keys.
13 The key service can be configured on by enabling:
15 "Security options"/"Enable access key retention support" (CONFIG_KEYS)
17 This document has the following sections:
20 - Key service overview
21 - Key access permissions
24 - Userspace system call interface
26 - Notes on accessing payload contents
28 - Request-key callback service
36 In this context, keys represent units of cryptographic data, authentication
37 tokens, keyrings, etc.. These are represented in the kernel by struct key.
39 Each key has a number of attributes:
43 - A description (for matching a key in a search).
44 - Access control information.
50 (*) Each key is issued a serial number of type key_serial_t that is unique for
51 the lifetime of that key. All serial numbers are positive non-zero 32-bit
54 Userspace programs can use a key's serial numbers as a way to gain access
55 to it, subject to permission checking.
57 (*) Each key is of a defined "type". Types must be registered inside the
58 kernel by a kernel service (such as a filesystem) before keys of that type
59 can be added or used. Userspace programs cannot define new types directly.
61 Key types are represented in the kernel by struct key_type. This defines a
62 number of operations that can be performed on a key of that type.
64 Should a type be removed from the system, all the keys of that type will
67 (*) Each key has a description. This should be a printable string. The key
68 type provides an operation to perform a match between the description on a
69 key and a criterion string.
71 (*) Each key has an owner user ID, a group ID and a permissions mask. These
72 are used to control what a process may do to a key from userspace, and
73 whether a kernel service will be able to find the key.
75 (*) Each key can be set to expire at a specific time by the key type's
76 instantiation function. Keys can also be immortal.
78 (*) Each key can have a payload. This is a quantity of data that represent the
79 actual "key". In the case of a keyring, this is a list of keys to which
80 the keyring links; in the case of a user-defined key, it's an arbitrary
83 Having a payload is not required; and the payload can, in fact, just be a
84 value stored in the struct key itself.
86 When a key is instantiated, the key type's instantiation function is
87 called with a blob of data, and that then creates the key's payload in
90 Similarly, when userspace wants to read back the contents of the key, if
91 permitted, another key type operation will be called to convert the key's
92 attached payload back into a blob of data.
94 (*) Each key can be in one of a number of basic states:
96 (*) Uninstantiated. The key exists, but does not have any data attached.
97 Keys being requested from userspace will be in this state.
99 (*) Instantiated. This is the normal state. The key is fully formed, and
102 (*) Negative. This is a relatively short-lived state. The key acts as a
103 note saying that a previous call out to userspace failed, and acts as
104 a throttle on key lookups. A negative key can be updated to a normal
107 (*) Expired. Keys can have lifetimes set. If their lifetime is exceeded,
108 they traverse to this state. An expired key can be updated back to a
111 (*) Revoked. A key is put in this state by userspace action. It can't be
112 found or operated upon (apart from by unlinking it).
114 (*) Dead. The key's type was unregistered, and so the key is now useless.
116 Keys in the last three states are subject to garbage collection. See the
117 section on "Garbage collection".
124 The key service provides a number of features besides keys:
126 (*) The key service defines three special key types:
130 Keyrings are special keys that contain a list of other keys. Keyring
131 lists can be modified using various system calls. Keyrings should not
132 be given a payload when created.
136 A key of this type has a description and a payload that are arbitrary
137 blobs of data. These can be created, updated and read by userspace,
138 and aren't intended for use by kernel services.
142 Like a "user" key, a "logon" key has a payload that is an arbitrary
143 blob of data. It is intended as a place to store secrets which are
144 accessible to the kernel but not to userspace programs.
146 The description can be arbitrary, but must be prefixed with a non-zero
147 length string that describes the key "subclass". The subclass is
148 separated from the rest of the description by a ':'. "logon" keys can
149 be created and updated from userspace, but the payload is only
150 readable from kernel space.
152 (*) Each process subscribes to three keyrings: a thread-specific keyring, a
153 process-specific keyring, and a session-specific keyring.
155 The thread-specific keyring is discarded from the child when any sort of
156 clone, fork, vfork or execve occurs. A new keyring is created only when
159 The process-specific keyring is replaced with an empty one in the child on
160 clone, fork, vfork unless CLONE_THREAD is supplied, in which case it is
161 shared. execve also discards the process's process keyring and creates a
164 The session-specific keyring is persistent across clone, fork, vfork and
165 execve, even when the latter executes a set-UID or set-GID binary. A
166 process can, however, replace its current session keyring with a new one
167 by using PR_JOIN_SESSION_KEYRING. It is permitted to request an anonymous
168 new one, or to attempt to create or join one of a specific name.
170 The ownership of the thread keyring changes when the real UID and GID of
173 (*) Each user ID resident in the system holds two special keyrings: a user
174 specific keyring and a default user session keyring. The default session
175 keyring is initialised with a link to the user-specific keyring.
177 When a process changes its real UID, if it used to have no session key, it
178 will be subscribed to the default session key for the new UID.
180 If a process attempts to access its session key when it doesn't have one,
181 it will be subscribed to the default for its current UID.
183 (*) Each user has two quotas against which the keys they own are tracked. One
184 limits the total number of keys and keyrings, the other limits the total
185 amount of description and payload space that can be consumed.
187 The user can view information on this and other statistics through procfs
188 files. The root user may also alter the quota limits through sysctl files
189 (see the section "New procfs files").
191 Process-specific and thread-specific keyrings are not counted towards a
194 If a system call that modifies a key or keyring in some way would put the
195 user over quota, the operation is refused and error EDQUOT is returned.
197 (*) There's a system call interface by which userspace programs can create and
198 manipulate keys and keyrings.
200 (*) There's a kernel interface by which services can register types and search
203 (*) There's a way for the a search done from the kernel to call back to
204 userspace to request a key that can't be found in a process's keyrings.
206 (*) An optional filesystem is available through which the key database can be
207 viewed and manipulated.
210 ======================
211 KEY ACCESS PERMISSIONS
212 ======================
214 Keys have an owner user ID, a group access ID, and a permissions mask. The mask
215 has up to eight bits each for possessor, user, group and other access. Only
216 six of each set of eight bits are defined. These permissions granted are:
220 This permits a key or keyring's attributes to be viewed - including key
221 type and description.
225 This permits a key's payload to be viewed or a keyring's list of linked
230 This permits a key's payload to be instantiated or updated, or it allows a
231 link to be added to or removed from a keyring.
235 This permits keyrings to be searched and keys to be found. Searches can
236 only recurse into nested keyrings that have search permission set.
240 This permits a key or keyring to be linked to. To create a link from a
241 keyring to a key, a process must have Write permission on the keyring and
242 Link permission on the key.
246 This permits a key's UID, GID and permissions mask to be changed.
248 For changing the ownership, group ID or permissions mask, being the owner of
249 the key or having the sysadmin capability is sufficient.
256 The security class "key" has been added to SELinux so that mandatory access
257 controls can be applied to keys created within various contexts. This support
258 is preliminary, and is likely to change quite significantly in the near future.
259 Currently, all of the basic permissions explained above are provided in SELinux
260 as well; SELinux is simply invoked after all basic permission checks have been
263 The value of the file /proc/self/attr/keycreate influences the labeling of
264 newly-created keys. If the contents of that file correspond to an SELinux
265 security context, then the key will be assigned that context. Otherwise, the
266 key will be assigned the current context of the task that invoked the key
267 creation request. Tasks must be granted explicit permission to assign a
268 particular context to newly-created keys, using the "create" permission in the
271 The default keyrings associated with users will be labeled with the default
272 context of the user if and only if the login programs have been instrumented to
273 properly initialize keycreate during the login process. Otherwise, they will
274 be labeled with the context of the login program itself.
276 Note, however, that the default keyrings associated with the root user are
277 labeled with the default kernel context, since they are created early in the
278 boot process, before root has a chance to log in.
280 The keyrings associated with new threads are each labeled with the context of
281 their associated thread, and both session and process keyrings are handled
289 Two files have been added to procfs by which an administrator can find out
290 about the status of the key service:
294 This lists the keys that are currently viewable by the task reading the
295 file, giving information about their type, description and permissions.
296 It is not possible to view the payload of the key this way, though some
297 information about it may be given.
299 The only keys included in the list are those that grant View permission to
300 the reading process whether or not it possesses them. Note that LSM
301 security checks are still performed, and may further filter out keys that
302 the current process is not authorised to view.
304 The contents of the file look like this:
306 SERIAL FLAGS USAGE EXPY PERM UID GID TYPE DESCRIPTION: SUMMARY
307 00000001 I----- 39 perm 1f3f0000 0 0 keyring _uid_ses.0: 1/4
308 00000002 I----- 2 perm 1f3f0000 0 0 keyring _uid.0: empty
309 00000007 I----- 1 perm 1f3f0000 0 0 keyring _pid.1: empty
310 0000018d I----- 1 perm 1f3f0000 0 0 keyring _pid.412: empty
311 000004d2 I--Q-- 1 perm 1f3f0000 32 -1 keyring _uid.32: 1/4
312 000004d3 I--Q-- 3 perm 1f3f0000 32 -1 keyring _uid_ses.32: empty
313 00000892 I--QU- 1 perm 1f000000 0 0 user metal:copper: 0
314 00000893 I--Q-N 1 35s 1f3f0000 0 0 user metal:silver: 0
315 00000894 I--Q-- 1 10h 003f0000 0 0 user metal:gold: 0
322 Q Contributes to user's quota
323 U Under construction by callback to userspace
326 This file must be enabled at kernel configuration time as it allows anyone
327 to list the keys database.
331 This file lists the tracking data for each user that has at least one key
332 on the system. Such data includes quota information and statistics:
334 [root@andromeda root]# cat /proc/key-users
335 0: 46 45/45 1/100 13/10000
336 29: 2 2/2 2/100 40/10000
337 32: 2 2/2 2/100 40/10000
338 38: 2 2/2 2/100 40/10000
340 The format of each line is
341 <UID>: User ID to which this applies
342 <usage> Structure refcount
343 <inst>/<keys> Total number of keys and number instantiated
344 <keys>/<max> Key count quota
345 <bytes>/<max> Key size quota
348 Four new sysctl files have been added also for the purpose of controlling the
349 quota limits on keys:
351 (*) /proc/sys/kernel/keys/root_maxkeys
352 /proc/sys/kernel/keys/root_maxbytes
354 These files hold the maximum number of keys that root may have and the
355 maximum total number of bytes of data that root may have stored in those
358 (*) /proc/sys/kernel/keys/maxkeys
359 /proc/sys/kernel/keys/maxbytes
361 These files hold the maximum number of keys that each non-root user may
362 have and the maximum total number of bytes of data that each of those
363 users may have stored in their keys.
365 Root may alter these by writing each new limit as a decimal number string to
366 the appropriate file.
369 ===============================
370 USERSPACE SYSTEM CALL INTERFACE
371 ===============================
373 Userspace can manipulate keys directly through three new syscalls: add_key,
374 request_key and keyctl. The latter provides a number of functions for
377 When referring to a key directly, userspace programs should use the key's
378 serial number (a positive 32-bit integer). However, there are some special
379 values available for referring to special keys and keyrings that relate to the
380 process making the call:
382 CONSTANT VALUE KEY REFERENCED
383 ============================== ====== ===========================
384 KEY_SPEC_THREAD_KEYRING -1 thread-specific keyring
385 KEY_SPEC_PROCESS_KEYRING -2 process-specific keyring
386 KEY_SPEC_SESSION_KEYRING -3 session-specific keyring
387 KEY_SPEC_USER_KEYRING -4 UID-specific keyring
388 KEY_SPEC_USER_SESSION_KEYRING -5 UID-session keyring
389 KEY_SPEC_GROUP_KEYRING -6 GID-specific keyring
390 KEY_SPEC_REQKEY_AUTH_KEY -7 assumed request_key()
394 The main syscalls are:
396 (*) Create a new key of given type, description and payload and add it to the
399 key_serial_t add_key(const char *type, const char *desc,
400 const void *payload, size_t plen,
401 key_serial_t keyring);
403 If a key of the same type and description as that proposed already exists
404 in the keyring, this will try to update it with the given payload, or it
405 will return error EEXIST if that function is not supported by the key
406 type. The process must also have permission to write to the key to be able
407 to update it. The new key will have all user permissions granted and no
408 group or third party permissions.
410 Otherwise, this will attempt to create a new key of the specified type and
411 description, and to instantiate it with the supplied payload and attach it
412 to the keyring. In this case, an error will be generated if the process
413 does not have permission to write to the keyring.
415 If the key type supports it, if the description is NULL or an empty
416 string, the key type will try and generate a description from the content
419 The payload is optional, and the pointer can be NULL if not required by
420 the type. The payload is plen in size, and plen can be zero for an empty
423 A new keyring can be generated by setting type "keyring", the keyring name
424 as the description (or NULL) and setting the payload to NULL.
426 User defined keys can be created by specifying type "user". It is
427 recommended that a user defined key's description by prefixed with a type
428 ID and a colon, such as "krb5tgt:" for a Kerberos 5 ticket granting
431 Any other type must have been registered with the kernel in advance by a
432 kernel service such as a filesystem.
434 The ID of the new or updated key is returned if successful.
437 (*) Search the process's keyrings for a key, potentially calling out to
438 userspace to create it.
440 key_serial_t request_key(const char *type, const char *description,
441 const char *callout_info,
442 key_serial_t dest_keyring);
444 This function searches all the process's keyrings in the order thread,
445 process, session for a matching key. This works very much like
446 KEYCTL_SEARCH, including the optional attachment of the discovered key to
449 If a key cannot be found, and if callout_info is not NULL, then
450 /sbin/request-key will be invoked in an attempt to obtain a key. The
451 callout_info string will be passed as an argument to the program.
453 See also Documentation/security/keys-request-key.txt.
456 The keyctl syscall functions are:
458 (*) Map a special key ID to a real key ID for this process:
460 key_serial_t keyctl(KEYCTL_GET_KEYRING_ID, key_serial_t id,
463 The special key specified by "id" is looked up (with the key being created
464 if necessary) and the ID of the key or keyring thus found is returned if
467 If the key does not yet exist, the key will be created if "create" is
468 non-zero; and the error ENOKEY will be returned if "create" is zero.
471 (*) Replace the session keyring this process subscribes to with a new one:
473 key_serial_t keyctl(KEYCTL_JOIN_SESSION_KEYRING, const char *name);
475 If name is NULL, an anonymous keyring is created attached to the process
476 as its session keyring, displacing the old session keyring.
478 If name is not NULL, if a keyring of that name exists, the process
479 attempts to attach it as the session keyring, returning an error if that
480 is not permitted; otherwise a new keyring of that name is created and
481 attached as the session keyring.
483 To attach to a named keyring, the keyring must have search permission for
484 the process's ownership.
486 The ID of the new session keyring is returned if successful.
489 (*) Update the specified key:
491 long keyctl(KEYCTL_UPDATE, key_serial_t key, const void *payload,
494 This will try to update the specified key with the given payload, or it
495 will return error EOPNOTSUPP if that function is not supported by the key
496 type. The process must also have permission to write to the key to be able
499 The payload is of length plen, and may be absent or empty as for
505 long keyctl(KEYCTL_REVOKE, key_serial_t key);
507 This makes a key unavailable for further operations. Further attempts to
508 use the key will be met with error EKEYREVOKED, and the key will no longer
512 (*) Change the ownership of a key:
514 long keyctl(KEYCTL_CHOWN, key_serial_t key, uid_t uid, gid_t gid);
516 This function permits a key's owner and group ID to be changed. Either one
517 of uid or gid can be set to -1 to suppress that change.
519 Only the superuser can change a key's owner to something other than the
520 key's current owner. Similarly, only the superuser can change a key's
521 group ID to something other than the calling process's group ID or one of
522 its group list members.
525 (*) Change the permissions mask on a key:
527 long keyctl(KEYCTL_SETPERM, key_serial_t key, key_perm_t perm);
529 This function permits the owner of a key or the superuser to change the
530 permissions mask on a key.
532 Only bits the available bits are permitted; if any other bits are set,
533 error EINVAL will be returned.
538 long keyctl(KEYCTL_DESCRIBE, key_serial_t key, char *buffer,
541 This function returns a summary of the key's attributes (but not its
542 payload data) as a string in the buffer provided.
544 Unless there's an error, it always returns the amount of data it could
545 produce, even if that's too big for the buffer, but it won't copy more
546 than requested to userspace. If the buffer pointer is NULL then no copy
549 A process must have view permission on the key for this function to be
552 If successful, a string is placed in the buffer in the following format:
554 <type>;<uid>;<gid>;<perm>;<description>
556 Where type and description are strings, uid and gid are decimal, and perm
557 is hexadecimal. A NUL character is included at the end of the string if
558 the buffer is sufficiently big.
560 This can be parsed with
562 sscanf(buffer, "%[^;];%d;%d;%o;%s", type, &uid, &gid, &mode, desc);
565 (*) Clear out a keyring:
567 long keyctl(KEYCTL_CLEAR, key_serial_t keyring);
569 This function clears the list of keys attached to a keyring. The calling
570 process must have write permission on the keyring, and it must be a
571 keyring (or else error ENOTDIR will result).
573 This function can also be used to clear special kernel keyrings if they
574 are appropriately marked if the user has CAP_SYS_ADMIN capability. The
575 DNS resolver cache keyring is an example of this.
578 (*) Link a key into a keyring:
580 long keyctl(KEYCTL_LINK, key_serial_t keyring, key_serial_t key);
582 This function creates a link from the keyring to the key. The process must
583 have write permission on the keyring and must have link permission on the
586 Should the keyring not be a keyring, error ENOTDIR will result; and if the
587 keyring is full, error ENFILE will result.
589 The link procedure checks the nesting of the keyrings, returning ELOOP if
590 it appears too deep or EDEADLK if the link would introduce a cycle.
592 Any links within the keyring to keys that match the new key in terms of
593 type and description will be discarded from the keyring as the new one is
597 (*) Unlink a key or keyring from another keyring:
599 long keyctl(KEYCTL_UNLINK, key_serial_t keyring, key_serial_t key);
601 This function looks through the keyring for the first link to the
602 specified key, and removes it if found. Subsequent links to that key are
603 ignored. The process must have write permission on the keyring.
605 If the keyring is not a keyring, error ENOTDIR will result; and if the key
606 is not present, error ENOENT will be the result.
609 (*) Search a keyring tree for a key:
611 key_serial_t keyctl(KEYCTL_SEARCH, key_serial_t keyring,
612 const char *type, const char *description,
613 key_serial_t dest_keyring);
615 This searches the keyring tree headed by the specified keyring until a key
616 is found that matches the type and description criteria. Each keyring is
617 checked for keys before recursion into its children occurs.
619 The process must have search permission on the top level keyring, or else
620 error EACCES will result. Only keyrings that the process has search
621 permission on will be recursed into, and only keys and keyrings for which
622 a process has search permission can be matched. If the specified keyring
623 is not a keyring, ENOTDIR will result.
625 If the search succeeds, the function will attempt to link the found key
626 into the destination keyring if one is supplied (non-zero ID). All the
627 constraints applicable to KEYCTL_LINK apply in this case too.
629 Error ENOKEY, EKEYREVOKED or EKEYEXPIRED will be returned if the search
630 fails. On success, the resulting key ID will be returned.
633 (*) Read the payload data from a key:
635 long keyctl(KEYCTL_READ, key_serial_t keyring, char *buffer,
638 This function attempts to read the payload data from the specified key
639 into the buffer. The process must have read permission on the key to
642 The returned data will be processed for presentation by the key type. For
643 instance, a keyring will return an array of key_serial_t entries
644 representing the IDs of all the keys to which it is subscribed. The user
645 defined key type will return its data as is. If a key type does not
646 implement this function, error EOPNOTSUPP will result.
648 As much of the data as can be fitted into the buffer will be copied to
649 userspace if the buffer pointer is not NULL.
651 On a successful return, the function will always return the amount of data
652 available rather than the amount copied.
655 (*) Instantiate a partially constructed key.
657 long keyctl(KEYCTL_INSTANTIATE, key_serial_t key,
658 const void *payload, size_t plen,
659 key_serial_t keyring);
660 long keyctl(KEYCTL_INSTANTIATE_IOV, key_serial_t key,
661 const struct iovec *payload_iov, unsigned ioc,
662 key_serial_t keyring);
664 If the kernel calls back to userspace to complete the instantiation of a
665 key, userspace should use this call to supply data for the key before the
666 invoked process returns, or else the key will be marked negative
669 The process must have write access on the key to be able to instantiate
670 it, and the key must be uninstantiated.
672 If a keyring is specified (non-zero), the key will also be linked into
673 that keyring, however all the constraints applying in KEYCTL_LINK apply in
676 The payload and plen arguments describe the payload data as for add_key().
678 The payload_iov and ioc arguments describe the payload data in an iovec
679 array instead of a single buffer.
682 (*) Negatively instantiate a partially constructed key.
684 long keyctl(KEYCTL_NEGATE, key_serial_t key,
685 unsigned timeout, key_serial_t keyring);
686 long keyctl(KEYCTL_REJECT, key_serial_t key,
687 unsigned timeout, unsigned error, key_serial_t keyring);
689 If the kernel calls back to userspace to complete the instantiation of a
690 key, userspace should use this call mark the key as negative before the
691 invoked process returns if it is unable to fulfill the request.
693 The process must have write access on the key to be able to instantiate
694 it, and the key must be uninstantiated.
696 If a keyring is specified (non-zero), the key will also be linked into
697 that keyring, however all the constraints applying in KEYCTL_LINK apply in
700 If the key is rejected, future searches for it will return the specified
701 error code until the rejected key expires. Negating the key is the same
702 as rejecting the key with ENOKEY as the error code.
705 (*) Set the default request-key destination keyring.
707 long keyctl(KEYCTL_SET_REQKEY_KEYRING, int reqkey_defl);
709 This sets the default keyring to which implicitly requested keys will be
710 attached for this thread. reqkey_defl should be one of these constants:
712 CONSTANT VALUE NEW DEFAULT KEYRING
713 ====================================== ====== =======================
714 KEY_REQKEY_DEFL_NO_CHANGE -1 No change
715 KEY_REQKEY_DEFL_DEFAULT 0 Default[1]
716 KEY_REQKEY_DEFL_THREAD_KEYRING 1 Thread keyring
717 KEY_REQKEY_DEFL_PROCESS_KEYRING 2 Process keyring
718 KEY_REQKEY_DEFL_SESSION_KEYRING 3 Session keyring
719 KEY_REQKEY_DEFL_USER_KEYRING 4 User keyring
720 KEY_REQKEY_DEFL_USER_SESSION_KEYRING 5 User session keyring
721 KEY_REQKEY_DEFL_GROUP_KEYRING 6 Group keyring
723 The old default will be returned if successful and error EINVAL will be
724 returned if reqkey_defl is not one of the above values.
726 The default keyring can be overridden by the keyring indicated to the
727 request_key() system call.
729 Note that this setting is inherited across fork/exec.
731 [1] The default is: the thread keyring if there is one, otherwise
732 the process keyring if there is one, otherwise the session keyring if
733 there is one, otherwise the user default session keyring.
736 (*) Set the timeout on a key.
738 long keyctl(KEYCTL_SET_TIMEOUT, key_serial_t key, unsigned timeout);
740 This sets or clears the timeout on a key. The timeout can be 0 to clear
741 the timeout or a number of seconds to set the expiry time that far into
744 The process must have attribute modification access on a key to set its
745 timeout. Timeouts may not be set with this function on negative, revoked
749 (*) Assume the authority granted to instantiate a key
751 long keyctl(KEYCTL_ASSUME_AUTHORITY, key_serial_t key);
753 This assumes or divests the authority required to instantiate the
754 specified key. Authority can only be assumed if the thread has the
755 authorisation key associated with the specified key in its keyrings
758 Once authority is assumed, searches for keys will also search the
759 requester's keyrings using the requester's security label, UID, GID and
762 If the requested authority is unavailable, error EPERM will be returned,
763 likewise if the authority has been revoked because the target key is
764 already instantiated.
766 If the specified key is 0, then any assumed authority will be divested.
768 The assumed authoritative key is inherited across fork and exec.
771 (*) Get the LSM security context attached to a key.
773 long keyctl(KEYCTL_GET_SECURITY, key_serial_t key, char *buffer,
776 This function returns a string that represents the LSM security context
777 attached to a key in the buffer provided.
779 Unless there's an error, it always returns the amount of data it could
780 produce, even if that's too big for the buffer, but it won't copy more
781 than requested to userspace. If the buffer pointer is NULL then no copy
784 A NUL character is included at the end of the string if the buffer is
785 sufficiently big. This is included in the returned count. If no LSM is
786 in force then an empty string will be returned.
788 A process must have view permission on the key for this function to be
792 (*) Install the calling process's session keyring on its parent.
794 long keyctl(KEYCTL_SESSION_TO_PARENT);
796 This functions attempts to install the calling process's session keyring
797 on to the calling process's parent, replacing the parent's current session
800 The calling process must have the same ownership as its parent, the
801 keyring must have the same ownership as the calling process, the calling
802 process must have LINK permission on the keyring and the active LSM module
803 mustn't deny permission, otherwise error EPERM will be returned.
805 Error ENOMEM will be returned if there was insufficient memory to complete
806 the operation, otherwise 0 will be returned to indicate success.
808 The keyring will be replaced next time the parent process leaves the
809 kernel and resumes executing userspace.
812 (*) Invalidate a key.
814 long keyctl(KEYCTL_INVALIDATE, key_serial_t key);
816 This function marks a key as being invalidated and then wakes up the
817 garbage collector. The garbage collector immediately removes invalidated
818 keys from all keyrings and deletes the key when its reference count
821 Keys that are marked invalidated become invisible to normal key operations
822 immediately, though they are still visible in /proc/keys until deleted
823 (they're marked with an 'i' flag).
825 A process must have search permission on the key for this function to be
833 The kernel services for key management are fairly simple to deal with. They can
834 be broken down into two areas: keys and key types.
836 Dealing with keys is fairly straightforward. Firstly, the kernel service
837 registers its type, then it searches for a key of that type. It should retain
838 the key as long as it has need of it, and then it should release it. For a
839 filesystem or device file, a search would probably be performed during the open
840 call, and the key released upon close. How to deal with conflicting keys due to
841 two different users opening the same file is left to the filesystem author to
844 To access the key manager, the following header must be #included:
848 Specific key types should have a header file under include/keys/ that should be
849 used to access that type. For keys of type "user", for example, that would be:
853 Note that there are two different types of pointers to keys that may be
858 This simply points to the key structure itself. Key structures will be at
859 least four-byte aligned.
863 This is equivalent to a struct key *, but the least significant bit is set
864 if the caller "possesses" the key. By "possession" it is meant that the
865 calling processes has a searchable link to the key from one of its
866 keyrings. There are three functions for dealing with these:
868 key_ref_t make_key_ref(const struct key *key, bool possession);
870 struct key *key_ref_to_ptr(const key_ref_t key_ref);
872 bool is_key_possessed(const key_ref_t key_ref);
874 The first function constructs a key reference from a key pointer and
875 possession information (which must be true or false).
877 The second function retrieves the key pointer from a reference and the
878 third retrieves the possession flag.
880 When accessing a key's payload contents, certain precautions must be taken to
881 prevent access vs modification races. See the section "Notes on accessing
882 payload contents" for more information.
884 (*) To search for a key, call:
886 struct key *request_key(const struct key_type *type,
887 const char *description,
888 const char *callout_info);
890 This is used to request a key or keyring with a description that matches
891 the description specified according to the key type's match function. This
892 permits approximate matching to occur. If callout_string is not NULL, then
893 /sbin/request-key will be invoked in an attempt to obtain the key from
894 userspace. In that case, callout_string will be passed as an argument to
897 Should the function fail error ENOKEY, EKEYEXPIRED or EKEYREVOKED will be
900 If successful, the key will have been attached to the default keyring for
901 implicitly obtained request-key keys, as set by KEYCTL_SET_REQKEY_KEYRING.
903 See also Documentation/security/keys-request-key.txt.
906 (*) To search for a key, passing auxiliary data to the upcaller, call:
908 struct key *request_key_with_auxdata(const struct key_type *type,
909 const char *description,
910 const void *callout_info,
914 This is identical to request_key(), except that the auxiliary data is
915 passed to the key_type->request_key() op if it exists, and the callout_info
916 is a blob of length callout_len, if given (the length may be 0).
919 (*) A key can be requested asynchronously by calling one of:
921 struct key *request_key_async(const struct key_type *type,
922 const char *description,
923 const void *callout_info,
928 struct key *request_key_async_with_auxdata(const struct key_type *type,
929 const char *description,
930 const char *callout_info,
934 which are asynchronous equivalents of request_key() and
935 request_key_with_auxdata() respectively.
937 These two functions return with the key potentially still under
938 construction. To wait for construction completion, the following should be
941 int wait_for_key_construction(struct key *key, bool intr);
943 The function will wait for the key to finish being constructed and then
944 invokes key_validate() to return an appropriate value to indicate the state
945 of the key (0 indicates the key is usable).
947 If intr is true, then the wait can be interrupted by a signal, in which
948 case error ERESTARTSYS will be returned.
951 (*) When it is no longer required, the key should be released using:
953 void key_put(struct key *key);
957 void key_ref_put(key_ref_t key_ref);
959 These can be called from interrupt context. If CONFIG_KEYS is not set then
960 the argument will not be parsed.
963 (*) Extra references can be made to a key by calling one of the following
966 struct key *__key_get(struct key *key);
967 struct key *key_get(struct key *key);
969 Keys so references will need to be disposed of by calling key_put() when
970 they've been finished with. The key pointer passed in will be returned.
972 In the case of key_get(), if the pointer is NULL or CONFIG_KEYS is not set
973 then the key will not be dereferenced and no increment will take place.
976 (*) A key's serial number can be obtained by calling:
978 key_serial_t key_serial(struct key *key);
980 If key is NULL or if CONFIG_KEYS is not set then 0 will be returned (in the
981 latter case without parsing the argument).
984 (*) If a keyring was found in the search, this can be further searched by:
986 key_ref_t keyring_search(key_ref_t keyring_ref,
987 const struct key_type *type,
988 const char *description)
990 This searches the keyring tree specified for a matching key. Error ENOKEY
991 is returned upon failure (use IS_ERR/PTR_ERR to determine). If successful,
992 the returned key will need to be released.
994 The possession attribute from the keyring reference is used to control
995 access through the permissions mask and is propagated to the returned key
996 reference pointer if successful.
999 (*) A keyring can be created by:
1001 struct key *keyring_alloc(const char *description, uid_t uid, gid_t gid,
1002 const struct cred *cred,
1004 unsigned long flags,
1007 This creates a keyring with the given attributes and returns it. If dest
1008 is not NULL, the new keyring will be linked into the keyring to which it
1009 points. No permission checks are made upon the destination keyring.
1011 Error EDQUOT can be returned if the keyring would overload the quota (pass
1012 KEY_ALLOC_NOT_IN_QUOTA in flags if the keyring shouldn't be accounted
1013 towards the user's quota). Error ENOMEM can also be returned.
1016 (*) To check the validity of a key, this function can be called:
1018 int validate_key(struct key *key);
1020 This checks that the key in question hasn't expired or and hasn't been
1021 revoked. Should the key be invalid, error EKEYEXPIRED or EKEYREVOKED will
1022 be returned. If the key is NULL or if CONFIG_KEYS is not set then 0 will be
1023 returned (in the latter case without parsing the argument).
1026 (*) To register a key type, the following function should be called:
1028 int register_key_type(struct key_type *type);
1030 This will return error EEXIST if a type of the same name is already
1034 (*) To unregister a key type, call:
1036 void unregister_key_type(struct key_type *type);
1039 Under some circumstances, it may be desirable to deal with a bundle of keys.
1040 The facility provides access to the keyring type for managing such a bundle:
1042 struct key_type key_type_keyring;
1044 This can be used with a function such as request_key() to find a specific
1045 keyring in a process's keyrings. A keyring thus found can then be searched
1046 with keyring_search(). Note that it is not possible to use request_key() to
1047 search a specific keyring, so using keyrings in this way is of limited utility.
1050 ===================================
1051 NOTES ON ACCESSING PAYLOAD CONTENTS
1052 ===================================
1054 The simplest payload is just a number in key->payload.value. In this case,
1055 there's no need to indulge in RCU or locking when accessing the payload.
1057 More complex payload contents must be allocated and a pointer to them set in
1058 key->payload.data. One of the following ways must be selected to access the
1061 (1) Unmodifiable key type.
1063 If the key type does not have a modify method, then the key's payload can
1064 be accessed without any form of locking, provided that it's known to be
1065 instantiated (uninstantiated keys cannot be "found").
1067 (2) The key's semaphore.
1069 The semaphore could be used to govern access to the payload and to control
1070 the payload pointer. It must be write-locked for modifications and would
1071 have to be read-locked for general access. The disadvantage of doing this
1072 is that the accessor may be required to sleep.
1076 RCU must be used when the semaphore isn't already held; if the semaphore
1077 is held then the contents can't change under you unexpectedly as the
1078 semaphore must still be used to serialise modifications to the key. The
1079 key management code takes care of this for the key type.
1081 However, this means using:
1083 rcu_read_lock() ... rcu_dereference() ... rcu_read_unlock()
1085 to read the pointer, and:
1087 rcu_dereference() ... rcu_assign_pointer() ... call_rcu()
1089 to set the pointer and dispose of the old contents after a grace period.
1090 Note that only the key type should ever modify a key's payload.
1092 Furthermore, an RCU controlled payload must hold a struct rcu_head for the
1093 use of call_rcu() and, if the payload is of variable size, the length of
1094 the payload. key->datalen cannot be relied upon to be consistent with the
1095 payload just dereferenced if the key's semaphore is not held.
1102 A kernel service may want to define its own key type. For instance, an AFS
1103 filesystem might want to define a Kerberos 5 ticket key type. To do this, it
1104 author fills in a key_type struct and registers it with the system.
1106 Source files that implement key types should include the following header file:
1110 The structure has a number of fields, some of which are mandatory:
1112 (*) const char *name
1114 The name of the key type. This is used to translate a key type name
1115 supplied by userspace into a pointer to the structure.
1118 (*) size_t def_datalen
1120 This is optional - it supplies the default payload data length as
1121 contributed to the quota. If the key type's payload is always or almost
1122 always the same size, then this is a more efficient way to do things.
1124 The data length (and quota) on a particular key can always be changed
1125 during instantiation or update by calling:
1127 int key_payload_reserve(struct key *key, size_t datalen);
1129 With the revised data length. Error EDQUOT will be returned if this is not
1133 (*) int (*vet_description)(const char *description);
1135 This optional method is called to vet a key description. If the key type
1136 doesn't approve of the key description, it may return an error, otherwise
1140 (*) int (*preparse)(struct key_preparsed_payload *prep);
1142 This optional method permits the key type to attempt to parse payload
1143 before a key is created (add key) or the key semaphore is taken (update or
1144 instantiate key). The structure pointed to by prep looks like:
1146 struct key_preparsed_payload {
1155 Before calling the method, the caller will fill in data and datalen with
1156 the payload blob parameters; quotalen will be filled in with the default
1157 quota size from the key type and the rest will be cleared.
1159 If a description can be proposed from the payload contents, that should be
1160 attached as a string to the description field. This will be used for the
1161 key description if the caller of add_key() passes NULL or "".
1163 The method can attach anything it likes to type_data[] and payload. These
1164 are merely passed along to the instantiate() or update() operations.
1166 The method should return 0 if success ful or a negative error code
1170 (*) void (*free_preparse)(struct key_preparsed_payload *prep);
1172 This method is only required if the preparse() method is provided,
1173 otherwise it is unused. It cleans up anything attached to the
1174 description, type_data and payload fields of the key_preparsed_payload
1175 struct as filled in by the preparse() method.
1178 (*) int (*instantiate)(struct key *key, struct key_preparsed_payload *prep);
1180 This method is called to attach a payload to a key during construction.
1181 The payload attached need not bear any relation to the data passed to this
1184 The prep->data and prep->datalen fields will define the original payload
1185 blob. If preparse() was supplied then other fields may be filled in also.
1187 If the amount of data attached to the key differs from the size in
1188 keytype->def_datalen, then key_payload_reserve() should be called.
1190 This method does not have to lock the key in order to attach a payload.
1191 The fact that KEY_FLAG_INSTANTIATED is not set in key->flags prevents
1192 anything else from gaining access to the key.
1194 It is safe to sleep in this method.
1197 (*) int (*update)(struct key *key, const void *data, size_t datalen);
1199 If this type of key can be updated, then this method should be provided.
1200 It is called to update a key's payload from the blob of data provided.
1202 The prep->data and prep->datalen fields will define the original payload
1203 blob. If preparse() was supplied then other fields may be filled in also.
1205 key_payload_reserve() should be called if the data length might change
1206 before any changes are actually made. Note that if this succeeds, the type
1207 is committed to changing the key because it's already been altered, so all
1208 memory allocation must be done first.
1210 The key will have its semaphore write-locked before this method is called,
1211 but this only deters other writers; any changes to the key's payload must
1212 be made under RCU conditions, and call_rcu() must be used to dispose of
1215 key_payload_reserve() should be called before the changes are made, but
1216 after all allocations and other potentially failing function calls are
1219 It is safe to sleep in this method.
1222 (*) int (*match)(const struct key *key, const void *desc);
1224 This method is called to match a key against a description. It should
1225 return non-zero if the two match, zero if they don't.
1227 This method should not need to lock the key in any way. The type and
1228 description can be considered invariant, and the payload should not be
1229 accessed (the key may not yet be instantiated).
1231 It is not safe to sleep in this method; the caller may hold spinlocks.
1234 (*) void (*revoke)(struct key *key);
1236 This method is optional. It is called to discard part of the payload
1237 data upon a key being revoked. The caller will have the key semaphore
1240 It is safe to sleep in this method, though care should be taken to avoid
1241 a deadlock against the key semaphore.
1244 (*) void (*destroy)(struct key *key);
1246 This method is optional. It is called to discard the payload data on a key
1247 when it is being destroyed.
1249 This method does not need to lock the key to access the payload; it can
1250 consider the key as being inaccessible at this time. Note that the key's
1251 type may have been changed before this function is called.
1253 It is not safe to sleep in this method; the caller may hold spinlocks.
1256 (*) void (*describe)(const struct key *key, struct seq_file *p);
1258 This method is optional. It is called during /proc/keys reading to
1259 summarise a key's description and payload in text form.
1261 This method will be called with the RCU read lock held. rcu_dereference()
1262 should be used to read the payload pointer if the payload is to be
1263 accessed. key->datalen cannot be trusted to stay consistent with the
1264 contents of the payload.
1266 The description will not change, though the key's state may.
1268 It is not safe to sleep in this method; the RCU read lock is held by the
1272 (*) long (*read)(const struct key *key, char __user *buffer, size_t buflen);
1274 This method is optional. It is called by KEYCTL_READ to translate the
1275 key's payload into something a blob of data for userspace to deal with.
1276 Ideally, the blob should be in the same format as that passed in to the
1277 instantiate and update methods.
1279 If successful, the blob size that could be produced should be returned
1280 rather than the size copied.
1282 This method will be called with the key's semaphore read-locked. This will
1283 prevent the key's payload changing. It is not necessary to use RCU locking
1284 when accessing the key's payload. It is safe to sleep in this method, such
1285 as might happen when the userspace buffer is accessed.
1288 (*) int (*request_key)(struct key_construction *cons, const char *op,
1291 This method is optional. If provided, request_key() and friends will
1292 invoke this function rather than upcalling to /sbin/request-key to operate
1293 upon a key of this type.
1295 The aux parameter is as passed to request_key_async_with_auxdata() and
1296 similar or is NULL otherwise. Also passed are the construction record for
1297 the key to be operated upon and the operation type (currently only
1300 This method is permitted to return before the upcall is complete, but the
1301 following function must be called under all circumstances to complete the
1302 instantiation process, whether or not it succeeds, whether or not there's
1305 void complete_request_key(struct key_construction *cons, int error);
1307 The error parameter should be 0 on success, -ve on error. The
1308 construction record is destroyed by this action and the authorisation key
1309 will be revoked. If an error is indicated, the key under construction
1310 will be negatively instantiated if it wasn't already instantiated.
1312 If this method returns an error, that error will be returned to the
1313 caller of request_key*(). complete_request_key() must be called prior to
1316 The key under construction and the authorisation key can be found in the
1317 key_construction struct pointed to by cons:
1319 (*) struct key *key;
1321 The key under construction.
1323 (*) struct key *authkey;
1325 The authorisation key.
1328 ============================
1329 REQUEST-KEY CALLBACK SERVICE
1330 ============================
1332 To create a new key, the kernel will attempt to execute the following command
1335 /sbin/request-key create <key> <uid> <gid> \
1336 <threadring> <processring> <sessionring> <callout_info>
1338 <key> is the key being constructed, and the three keyrings are the process
1339 keyrings from the process that caused the search to be issued. These are
1340 included for two reasons:
1342 (1) There may be an authentication token in one of the keyrings that is
1343 required to obtain the key, eg: a Kerberos Ticket-Granting Ticket.
1345 (2) The new key should probably be cached in one of these rings.
1347 This program should set it UID and GID to those specified before attempting to
1348 access any more keys. It may then look around for a user specific process to
1349 hand the request off to (perhaps a path held in placed in another key by, for
1350 example, the KDE desktop manager).
1352 The program (or whatever it calls) should finish construction of the key by
1353 calling KEYCTL_INSTANTIATE or KEYCTL_INSTANTIATE_IOV, which also permits it to
1354 cache the key in one of the keyrings (probably the session ring) before
1355 returning. Alternatively, the key can be marked as negative with KEYCTL_NEGATE
1356 or KEYCTL_REJECT; this also permits the key to be cached in one of the
1359 If it returns with the key remaining in the unconstructed state, the key will
1360 be marked as being negative, it will be added to the session keyring, and an
1361 error will be returned to the key requestor.
1363 Supplementary information may be provided from whoever or whatever invoked this
1364 service. This will be passed as the <callout_info> parameter. If no such
1365 information was made available, then "-" will be passed as this parameter
1369 Similarly, the kernel may attempt to update an expired or a soon to expire key
1372 /sbin/request-key update <key> <uid> <gid> \
1373 <threadring> <processring> <sessionring>
1375 In this case, the program isn't required to actually attach the key to a ring;
1376 the rings are provided for reference.
1383 Dead keys (for which the type has been removed) will be automatically unlinked
1384 from those keyrings that point to them and deleted as soon as possible by a
1385 background garbage collector.
1387 Similarly, revoked and expired keys will be garbage collected, but only after a
1388 certain amount of time has passed. This time is set as a number of seconds in:
1390 /proc/sys/kernel/keys/gc_delay