Commit | Line | Data |
---|---|---|
1da177e4 LT |
1 | ============================ |
2 | KERNEL KEY RETENTION SERVICE | |
3 | ============================ | |
4 | ||
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 other kernel services. | |
8 | ||
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. | |
12 | ||
13 | The key service can be configured on by enabling: | |
14 | ||
15 | "Security options"/"Enable access key retention support" (CONFIG_KEYS) | |
16 | ||
17 | This document has the following sections: | |
18 | ||
19 | - Key overview | |
20 | - Key service overview | |
21 | - Key access permissions | |
22 | - New procfs files | |
23 | - Userspace system call interface | |
24 | - Kernel services | |
76d8aeab | 25 | - Notes on accessing payload contents |
1da177e4 LT |
26 | - Defining a key type |
27 | - Request-key callback service | |
28 | - Key access filesystem | |
29 | ||
30 | ||
31 | ============ | |
32 | KEY OVERVIEW | |
33 | ============ | |
34 | ||
35 | In this context, keys represent units of cryptographic data, authentication | |
36 | tokens, keyrings, etc.. These are represented in the kernel by struct key. | |
37 | ||
38 | Each key has a number of attributes: | |
39 | ||
40 | - A serial number. | |
41 | - A type. | |
42 | - A description (for matching a key in a search). | |
43 | - Access control information. | |
44 | - An expiry time. | |
45 | - A payload. | |
46 | - State. | |
47 | ||
48 | ||
76d8aeab DH |
49 | (*) Each key is issued a serial number of type key_serial_t that is unique for |
50 | the lifetime of that key. All serial numbers are positive non-zero 32-bit | |
51 | integers. | |
1da177e4 LT |
52 | |
53 | Userspace programs can use a key's serial numbers as a way to gain access | |
54 | to it, subject to permission checking. | |
55 | ||
56 | (*) Each key is of a defined "type". Types must be registered inside the | |
76d8aeab DH |
57 | kernel by a kernel service (such as a filesystem) before keys of that type |
58 | can be added or used. Userspace programs cannot define new types directly. | |
1da177e4 | 59 | |
76d8aeab DH |
60 | Key types are represented in the kernel by struct key_type. This defines a |
61 | number of operations that can be performed on a key of that type. | |
1da177e4 LT |
62 | |
63 | Should a type be removed from the system, all the keys of that type will | |
64 | be invalidated. | |
65 | ||
66 | (*) Each key has a description. This should be a printable string. The key | |
76d8aeab DH |
67 | type provides an operation to perform a match between the description on a |
68 | key and a criterion string. | |
1da177e4 LT |
69 | |
70 | (*) Each key has an owner user ID, a group ID and a permissions mask. These | |
71 | are used to control what a process may do to a key from userspace, and | |
72 | whether a kernel service will be able to find the key. | |
73 | ||
74 | (*) Each key can be set to expire at a specific time by the key type's | |
75 | instantiation function. Keys can also be immortal. | |
76 | ||
76d8aeab DH |
77 | (*) Each key can have a payload. This is a quantity of data that represent the |
78 | actual "key". In the case of a keyring, this is a list of keys to which | |
79 | the keyring links; in the case of a user-defined key, it's an arbitrary | |
80 | blob of data. | |
1da177e4 LT |
81 | |
82 | Having a payload is not required; and the payload can, in fact, just be a | |
83 | value stored in the struct key itself. | |
84 | ||
85 | When a key is instantiated, the key type's instantiation function is | |
86 | called with a blob of data, and that then creates the key's payload in | |
87 | some way. | |
88 | ||
89 | Similarly, when userspace wants to read back the contents of the key, if | |
90 | permitted, another key type operation will be called to convert the key's | |
91 | attached payload back into a blob of data. | |
92 | ||
93 | (*) Each key can be in one of a number of basic states: | |
94 | ||
76d8aeab DH |
95 | (*) Uninstantiated. The key exists, but does not have any data attached. |
96 | Keys being requested from userspace will be in this state. | |
1da177e4 LT |
97 | |
98 | (*) Instantiated. This is the normal state. The key is fully formed, and | |
99 | has data attached. | |
100 | ||
101 | (*) Negative. This is a relatively short-lived state. The key acts as a | |
102 | note saying that a previous call out to userspace failed, and acts as | |
103 | a throttle on key lookups. A negative key can be updated to a normal | |
104 | state. | |
105 | ||
106 | (*) Expired. Keys can have lifetimes set. If their lifetime is exceeded, | |
107 | they traverse to this state. An expired key can be updated back to a | |
108 | normal state. | |
109 | ||
110 | (*) Revoked. A key is put in this state by userspace action. It can't be | |
111 | found or operated upon (apart from by unlinking it). | |
112 | ||
113 | (*) Dead. The key's type was unregistered, and so the key is now useless. | |
114 | ||
115 | ||
116 | ==================== | |
117 | KEY SERVICE OVERVIEW | |
118 | ==================== | |
119 | ||
120 | The key service provides a number of features besides keys: | |
121 | ||
122 | (*) The key service defines two special key types: | |
123 | ||
124 | (+) "keyring" | |
125 | ||
126 | Keyrings are special keys that contain a list of other keys. Keyring | |
127 | lists can be modified using various system calls. Keyrings should not | |
128 | be given a payload when created. | |
129 | ||
130 | (+) "user" | |
131 | ||
132 | A key of this type has a description and a payload that are arbitrary | |
133 | blobs of data. These can be created, updated and read by userspace, | |
134 | and aren't intended for use by kernel services. | |
135 | ||
136 | (*) Each process subscribes to three keyrings: a thread-specific keyring, a | |
137 | process-specific keyring, and a session-specific keyring. | |
138 | ||
139 | The thread-specific keyring is discarded from the child when any sort of | |
140 | clone, fork, vfork or execve occurs. A new keyring is created only when | |
141 | required. | |
142 | ||
76d8aeab DH |
143 | The process-specific keyring is replaced with an empty one in the child on |
144 | clone, fork, vfork unless CLONE_THREAD is supplied, in which case it is | |
145 | shared. execve also discards the process's process keyring and creates a | |
146 | new one. | |
1da177e4 LT |
147 | |
148 | The session-specific keyring is persistent across clone, fork, vfork and | |
149 | execve, even when the latter executes a set-UID or set-GID binary. A | |
150 | process can, however, replace its current session keyring with a new one | |
151 | by using PR_JOIN_SESSION_KEYRING. It is permitted to request an anonymous | |
152 | new one, or to attempt to create or join one of a specific name. | |
153 | ||
154 | The ownership of the thread keyring changes when the real UID and GID of | |
155 | the thread changes. | |
156 | ||
157 | (*) Each user ID resident in the system holds two special keyrings: a user | |
158 | specific keyring and a default user session keyring. The default session | |
159 | keyring is initialised with a link to the user-specific keyring. | |
160 | ||
161 | When a process changes its real UID, if it used to have no session key, it | |
162 | will be subscribed to the default session key for the new UID. | |
163 | ||
164 | If a process attempts to access its session key when it doesn't have one, | |
165 | it will be subscribed to the default for its current UID. | |
166 | ||
167 | (*) Each user has two quotas against which the keys they own are tracked. One | |
168 | limits the total number of keys and keyrings, the other limits the total | |
169 | amount of description and payload space that can be consumed. | |
170 | ||
171 | The user can view information on this and other statistics through procfs | |
172 | files. | |
173 | ||
174 | Process-specific and thread-specific keyrings are not counted towards a | |
175 | user's quota. | |
176 | ||
177 | If a system call that modifies a key or keyring in some way would put the | |
178 | user over quota, the operation is refused and error EDQUOT is returned. | |
179 | ||
76d8aeab DH |
180 | (*) There's a system call interface by which userspace programs can create and |
181 | manipulate keys and keyrings. | |
1da177e4 | 182 | |
76d8aeab DH |
183 | (*) There's a kernel interface by which services can register types and search |
184 | for keys. | |
1da177e4 LT |
185 | |
186 | (*) There's a way for the a search done from the kernel to call back to | |
187 | userspace to request a key that can't be found in a process's keyrings. | |
188 | ||
189 | (*) An optional filesystem is available through which the key database can be | |
190 | viewed and manipulated. | |
191 | ||
192 | ||
193 | ====================== | |
194 | KEY ACCESS PERMISSIONS | |
195 | ====================== | |
196 | ||
76d8aeab DH |
197 | Keys have an owner user ID, a group access ID, and a permissions mask. The mask |
198 | has up to eight bits each for user, group and other access. Only five of each | |
199 | set of eight bits are defined. These permissions granted are: | |
1da177e4 LT |
200 | |
201 | (*) View | |
202 | ||
203 | This permits a key or keyring's attributes to be viewed - including key | |
204 | type and description. | |
205 | ||
206 | (*) Read | |
207 | ||
208 | This permits a key's payload to be viewed or a keyring's list of linked | |
209 | keys. | |
210 | ||
211 | (*) Write | |
212 | ||
76d8aeab DH |
213 | This permits a key's payload to be instantiated or updated, or it allows a |
214 | link to be added to or removed from a keyring. | |
1da177e4 LT |
215 | |
216 | (*) Search | |
217 | ||
218 | This permits keyrings to be searched and keys to be found. Searches can | |
219 | only recurse into nested keyrings that have search permission set. | |
220 | ||
221 | (*) Link | |
222 | ||
223 | This permits a key or keyring to be linked to. To create a link from a | |
224 | keyring to a key, a process must have Write permission on the keyring and | |
225 | Link permission on the key. | |
226 | ||
227 | For changing the ownership, group ID or permissions mask, being the owner of | |
228 | the key or having the sysadmin capability is sufficient. | |
229 | ||
230 | ||
231 | ================ | |
232 | NEW PROCFS FILES | |
233 | ================ | |
234 | ||
235 | Two files have been added to procfs by which an administrator can find out | |
236 | about the status of the key service: | |
237 | ||
238 | (*) /proc/keys | |
239 | ||
240 | This lists all the keys on the system, giving information about their | |
76d8aeab DH |
241 | type, description and permissions. The payload of the key is not available |
242 | this way: | |
1da177e4 LT |
243 | |
244 | SERIAL FLAGS USAGE EXPY PERM UID GID TYPE DESCRIPTION: SUMMARY | |
245 | 00000001 I----- 39 perm 1f0000 0 0 keyring _uid_ses.0: 1/4 | |
246 | 00000002 I----- 2 perm 1f0000 0 0 keyring _uid.0: empty | |
247 | 00000007 I----- 1 perm 1f0000 0 0 keyring _pid.1: empty | |
248 | 0000018d I----- 1 perm 1f0000 0 0 keyring _pid.412: empty | |
249 | 000004d2 I--Q-- 1 perm 1f0000 32 -1 keyring _uid.32: 1/4 | |
250 | 000004d3 I--Q-- 3 perm 1f0000 32 -1 keyring _uid_ses.32: empty | |
251 | 00000892 I--QU- 1 perm 1f0000 0 0 user metal:copper: 0 | |
252 | 00000893 I--Q-N 1 35s 1f0000 0 0 user metal:silver: 0 | |
253 | 00000894 I--Q-- 1 10h 1f0000 0 0 user metal:gold: 0 | |
254 | ||
255 | The flags are: | |
256 | ||
257 | I Instantiated | |
258 | R Revoked | |
259 | D Dead | |
260 | Q Contributes to user's quota | |
261 | U Under contruction by callback to userspace | |
262 | N Negative key | |
263 | ||
264 | This file must be enabled at kernel configuration time as it allows anyone | |
265 | to list the keys database. | |
266 | ||
267 | (*) /proc/key-users | |
268 | ||
269 | This file lists the tracking data for each user that has at least one key | |
270 | on the system. Such data includes quota information and statistics: | |
271 | ||
272 | [root@andromeda root]# cat /proc/key-users | |
273 | 0: 46 45/45 1/100 13/10000 | |
274 | 29: 2 2/2 2/100 40/10000 | |
275 | 32: 2 2/2 2/100 40/10000 | |
276 | 38: 2 2/2 2/100 40/10000 | |
277 | ||
278 | The format of each line is | |
279 | <UID>: User ID to which this applies | |
280 | <usage> Structure refcount | |
281 | <inst>/<keys> Total number of keys and number instantiated | |
282 | <keys>/<max> Key count quota | |
283 | <bytes>/<max> Key size quota | |
284 | ||
285 | ||
286 | =============================== | |
287 | USERSPACE SYSTEM CALL INTERFACE | |
288 | =============================== | |
289 | ||
290 | Userspace can manipulate keys directly through three new syscalls: add_key, | |
291 | request_key and keyctl. The latter provides a number of functions for | |
292 | manipulating keys. | |
293 | ||
294 | When referring to a key directly, userspace programs should use the key's | |
295 | serial number (a positive 32-bit integer). However, there are some special | |
296 | values available for referring to special keys and keyrings that relate to the | |
297 | process making the call: | |
298 | ||
299 | CONSTANT VALUE KEY REFERENCED | |
300 | ============================== ====== =========================== | |
301 | KEY_SPEC_THREAD_KEYRING -1 thread-specific keyring | |
302 | KEY_SPEC_PROCESS_KEYRING -2 process-specific keyring | |
303 | KEY_SPEC_SESSION_KEYRING -3 session-specific keyring | |
304 | KEY_SPEC_USER_KEYRING -4 UID-specific keyring | |
305 | KEY_SPEC_USER_SESSION_KEYRING -5 UID-session keyring | |
306 | KEY_SPEC_GROUP_KEYRING -6 GID-specific keyring | |
307 | ||
308 | ||
309 | The main syscalls are: | |
310 | ||
311 | (*) Create a new key of given type, description and payload and add it to the | |
312 | nominated keyring: | |
313 | ||
314 | key_serial_t add_key(const char *type, const char *desc, | |
315 | const void *payload, size_t plen, | |
316 | key_serial_t keyring); | |
317 | ||
318 | If a key of the same type and description as that proposed already exists | |
319 | in the keyring, this will try to update it with the given payload, or it | |
320 | will return error EEXIST if that function is not supported by the key | |
76d8aeab DH |
321 | type. The process must also have permission to write to the key to be able |
322 | to update it. The new key will have all user permissions granted and no | |
323 | group or third party permissions. | |
1da177e4 | 324 | |
76d8aeab DH |
325 | Otherwise, this will attempt to create a new key of the specified type and |
326 | description, and to instantiate it with the supplied payload and attach it | |
327 | to the keyring. In this case, an error will be generated if the process | |
328 | does not have permission to write to the keyring. | |
1da177e4 LT |
329 | |
330 | The payload is optional, and the pointer can be NULL if not required by | |
331 | the type. The payload is plen in size, and plen can be zero for an empty | |
332 | payload. | |
333 | ||
76d8aeab DH |
334 | A new keyring can be generated by setting type "keyring", the keyring name |
335 | as the description (or NULL) and setting the payload to NULL. | |
1da177e4 LT |
336 | |
337 | User defined keys can be created by specifying type "user". It is | |
338 | recommended that a user defined key's description by prefixed with a type | |
339 | ID and a colon, such as "krb5tgt:" for a Kerberos 5 ticket granting | |
340 | ticket. | |
341 | ||
342 | Any other type must have been registered with the kernel in advance by a | |
343 | kernel service such as a filesystem. | |
344 | ||
345 | The ID of the new or updated key is returned if successful. | |
346 | ||
347 | ||
348 | (*) Search the process's keyrings for a key, potentially calling out to | |
349 | userspace to create it. | |
350 | ||
351 | key_serial_t request_key(const char *type, const char *description, | |
352 | const char *callout_info, | |
353 | key_serial_t dest_keyring); | |
354 | ||
355 | This function searches all the process's keyrings in the order thread, | |
356 | process, session for a matching key. This works very much like | |
357 | KEYCTL_SEARCH, including the optional attachment of the discovered key to | |
358 | a keyring. | |
359 | ||
360 | If a key cannot be found, and if callout_info is not NULL, then | |
361 | /sbin/request-key will be invoked in an attempt to obtain a key. The | |
362 | callout_info string will be passed as an argument to the program. | |
363 | ||
364 | ||
365 | The keyctl syscall functions are: | |
366 | ||
367 | (*) Map a special key ID to a real key ID for this process: | |
368 | ||
369 | key_serial_t keyctl(KEYCTL_GET_KEYRING_ID, key_serial_t id, | |
370 | int create); | |
371 | ||
76d8aeab DH |
372 | The special key specified by "id" is looked up (with the key being created |
373 | if necessary) and the ID of the key or keyring thus found is returned if | |
374 | it exists. | |
1da177e4 LT |
375 | |
376 | If the key does not yet exist, the key will be created if "create" is | |
377 | non-zero; and the error ENOKEY will be returned if "create" is zero. | |
378 | ||
379 | ||
380 | (*) Replace the session keyring this process subscribes to with a new one: | |
381 | ||
382 | key_serial_t keyctl(KEYCTL_JOIN_SESSION_KEYRING, const char *name); | |
383 | ||
384 | If name is NULL, an anonymous keyring is created attached to the process | |
385 | as its session keyring, displacing the old session keyring. | |
386 | ||
387 | If name is not NULL, if a keyring of that name exists, the process | |
388 | attempts to attach it as the session keyring, returning an error if that | |
389 | is not permitted; otherwise a new keyring of that name is created and | |
390 | attached as the session keyring. | |
391 | ||
392 | To attach to a named keyring, the keyring must have search permission for | |
393 | the process's ownership. | |
394 | ||
395 | The ID of the new session keyring is returned if successful. | |
396 | ||
397 | ||
398 | (*) Update the specified key: | |
399 | ||
400 | long keyctl(KEYCTL_UPDATE, key_serial_t key, const void *payload, | |
401 | size_t plen); | |
402 | ||
403 | This will try to update the specified key with the given payload, or it | |
404 | will return error EOPNOTSUPP if that function is not supported by the key | |
76d8aeab DH |
405 | type. The process must also have permission to write to the key to be able |
406 | to update it. | |
1da177e4 LT |
407 | |
408 | The payload is of length plen, and may be absent or empty as for | |
409 | add_key(). | |
410 | ||
411 | ||
412 | (*) Revoke a key: | |
413 | ||
414 | long keyctl(KEYCTL_REVOKE, key_serial_t key); | |
415 | ||
416 | This makes a key unavailable for further operations. Further attempts to | |
417 | use the key will be met with error EKEYREVOKED, and the key will no longer | |
418 | be findable. | |
419 | ||
420 | ||
421 | (*) Change the ownership of a key: | |
422 | ||
423 | long keyctl(KEYCTL_CHOWN, key_serial_t key, uid_t uid, gid_t gid); | |
424 | ||
76d8aeab DH |
425 | This function permits a key's owner and group ID to be changed. Either one |
426 | of uid or gid can be set to -1 to suppress that change. | |
1da177e4 LT |
427 | |
428 | Only the superuser can change a key's owner to something other than the | |
429 | key's current owner. Similarly, only the superuser can change a key's | |
430 | group ID to something other than the calling process's group ID or one of | |
431 | its group list members. | |
432 | ||
433 | ||
434 | (*) Change the permissions mask on a key: | |
435 | ||
436 | long keyctl(KEYCTL_SETPERM, key_serial_t key, key_perm_t perm); | |
437 | ||
438 | This function permits the owner of a key or the superuser to change the | |
439 | permissions mask on a key. | |
440 | ||
441 | Only bits the available bits are permitted; if any other bits are set, | |
442 | error EINVAL will be returned. | |
443 | ||
444 | ||
445 | (*) Describe a key: | |
446 | ||
447 | long keyctl(KEYCTL_DESCRIBE, key_serial_t key, char *buffer, | |
448 | size_t buflen); | |
449 | ||
450 | This function returns a summary of the key's attributes (but not its | |
451 | payload data) as a string in the buffer provided. | |
452 | ||
453 | Unless there's an error, it always returns the amount of data it could | |
454 | produce, even if that's too big for the buffer, but it won't copy more | |
455 | than requested to userspace. If the buffer pointer is NULL then no copy | |
456 | will take place. | |
457 | ||
458 | A process must have view permission on the key for this function to be | |
459 | successful. | |
460 | ||
461 | If successful, a string is placed in the buffer in the following format: | |
462 | ||
463 | <type>;<uid>;<gid>;<perm>;<description> | |
464 | ||
465 | Where type and description are strings, uid and gid are decimal, and perm | |
466 | is hexadecimal. A NUL character is included at the end of the string if | |
467 | the buffer is sufficiently big. | |
468 | ||
469 | This can be parsed with | |
470 | ||
471 | sscanf(buffer, "%[^;];%d;%d;%o;%s", type, &uid, &gid, &mode, desc); | |
472 | ||
473 | ||
474 | (*) Clear out a keyring: | |
475 | ||
476 | long keyctl(KEYCTL_CLEAR, key_serial_t keyring); | |
477 | ||
478 | This function clears the list of keys attached to a keyring. The calling | |
479 | process must have write permission on the keyring, and it must be a | |
480 | keyring (or else error ENOTDIR will result). | |
481 | ||
482 | ||
483 | (*) Link a key into a keyring: | |
484 | ||
485 | long keyctl(KEYCTL_LINK, key_serial_t keyring, key_serial_t key); | |
486 | ||
76d8aeab DH |
487 | This function creates a link from the keyring to the key. The process must |
488 | have write permission on the keyring and must have link permission on the | |
489 | key. | |
1da177e4 | 490 | |
76d8aeab DH |
491 | Should the keyring not be a keyring, error ENOTDIR will result; and if the |
492 | keyring is full, error ENFILE will result. | |
1da177e4 LT |
493 | |
494 | The link procedure checks the nesting of the keyrings, returning ELOOP if | |
495 | it appears to deep or EDEADLK if the link would introduce a cycle. | |
496 | ||
497 | ||
498 | (*) Unlink a key or keyring from another keyring: | |
499 | ||
500 | long keyctl(KEYCTL_UNLINK, key_serial_t keyring, key_serial_t key); | |
501 | ||
502 | This function looks through the keyring for the first link to the | |
503 | specified key, and removes it if found. Subsequent links to that key are | |
504 | ignored. The process must have write permission on the keyring. | |
505 | ||
76d8aeab DH |
506 | If the keyring is not a keyring, error ENOTDIR will result; and if the key |
507 | is not present, error ENOENT will be the result. | |
1da177e4 LT |
508 | |
509 | ||
510 | (*) Search a keyring tree for a key: | |
511 | ||
512 | key_serial_t keyctl(KEYCTL_SEARCH, key_serial_t keyring, | |
513 | const char *type, const char *description, | |
514 | key_serial_t dest_keyring); | |
515 | ||
76d8aeab DH |
516 | This searches the keyring tree headed by the specified keyring until a key |
517 | is found that matches the type and description criteria. Each keyring is | |
518 | checked for keys before recursion into its children occurs. | |
1da177e4 LT |
519 | |
520 | The process must have search permission on the top level keyring, or else | |
521 | error EACCES will result. Only keyrings that the process has search | |
522 | permission on will be recursed into, and only keys and keyrings for which | |
523 | a process has search permission can be matched. If the specified keyring | |
524 | is not a keyring, ENOTDIR will result. | |
525 | ||
526 | If the search succeeds, the function will attempt to link the found key | |
527 | into the destination keyring if one is supplied (non-zero ID). All the | |
528 | constraints applicable to KEYCTL_LINK apply in this case too. | |
529 | ||
530 | Error ENOKEY, EKEYREVOKED or EKEYEXPIRED will be returned if the search | |
531 | fails. On success, the resulting key ID will be returned. | |
532 | ||
533 | ||
534 | (*) Read the payload data from a key: | |
535 | ||
536 | key_serial_t keyctl(KEYCTL_READ, key_serial_t keyring, char *buffer, | |
537 | size_t buflen); | |
538 | ||
539 | This function attempts to read the payload data from the specified key | |
540 | into the buffer. The process must have read permission on the key to | |
541 | succeed. | |
542 | ||
543 | The returned data will be processed for presentation by the key type. For | |
544 | instance, a keyring will return an array of key_serial_t entries | |
545 | representing the IDs of all the keys to which it is subscribed. The user | |
546 | defined key type will return its data as is. If a key type does not | |
547 | implement this function, error EOPNOTSUPP will result. | |
548 | ||
549 | As much of the data as can be fitted into the buffer will be copied to | |
550 | userspace if the buffer pointer is not NULL. | |
551 | ||
76d8aeab DH |
552 | On a successful return, the function will always return the amount of data |
553 | available rather than the amount copied. | |
1da177e4 LT |
554 | |
555 | ||
556 | (*) Instantiate a partially constructed key. | |
557 | ||
558 | key_serial_t keyctl(KEYCTL_INSTANTIATE, key_serial_t key, | |
559 | const void *payload, size_t plen, | |
560 | key_serial_t keyring); | |
561 | ||
562 | If the kernel calls back to userspace to complete the instantiation of a | |
563 | key, userspace should use this call to supply data for the key before the | |
564 | invoked process returns, or else the key will be marked negative | |
565 | automatically. | |
566 | ||
567 | The process must have write access on the key to be able to instantiate | |
568 | it, and the key must be uninstantiated. | |
569 | ||
570 | If a keyring is specified (non-zero), the key will also be linked into | |
76d8aeab DH |
571 | that keyring, however all the constraints applying in KEYCTL_LINK apply in |
572 | this case too. | |
1da177e4 LT |
573 | |
574 | The payload and plen arguments describe the payload data as for add_key(). | |
575 | ||
576 | ||
577 | (*) Negatively instantiate a partially constructed key. | |
578 | ||
579 | key_serial_t keyctl(KEYCTL_NEGATE, key_serial_t key, | |
580 | unsigned timeout, key_serial_t keyring); | |
581 | ||
582 | If the kernel calls back to userspace to complete the instantiation of a | |
583 | key, userspace should use this call mark the key as negative before the | |
584 | invoked process returns if it is unable to fulfil the request. | |
585 | ||
586 | The process must have write access on the key to be able to instantiate | |
587 | it, and the key must be uninstantiated. | |
588 | ||
589 | If a keyring is specified (non-zero), the key will also be linked into | |
76d8aeab DH |
590 | that keyring, however all the constraints applying in KEYCTL_LINK apply in |
591 | this case too. | |
1da177e4 LT |
592 | |
593 | ||
3e30148c DH |
594 | (*) Set the default request-key destination keyring. |
595 | ||
596 | long keyctl(KEYCTL_SET_REQKEY_KEYRING, int reqkey_defl); | |
597 | ||
598 | This sets the default keyring to which implicitly requested keys will be | |
599 | attached for this thread. reqkey_defl should be one of these constants: | |
600 | ||
601 | CONSTANT VALUE NEW DEFAULT KEYRING | |
602 | ====================================== ====== ======================= | |
603 | KEY_REQKEY_DEFL_NO_CHANGE -1 No change | |
604 | KEY_REQKEY_DEFL_DEFAULT 0 Default[1] | |
605 | KEY_REQKEY_DEFL_THREAD_KEYRING 1 Thread keyring | |
606 | KEY_REQKEY_DEFL_PROCESS_KEYRING 2 Process keyring | |
607 | KEY_REQKEY_DEFL_SESSION_KEYRING 3 Session keyring | |
608 | KEY_REQKEY_DEFL_USER_KEYRING 4 User keyring | |
609 | KEY_REQKEY_DEFL_USER_SESSION_KEYRING 5 User session keyring | |
610 | KEY_REQKEY_DEFL_GROUP_KEYRING 6 Group keyring | |
611 | ||
612 | The old default will be returned if successful and error EINVAL will be | |
613 | returned if reqkey_defl is not one of the above values. | |
614 | ||
615 | The default keyring can be overridden by the keyring indicated to the | |
616 | request_key() system call. | |
617 | ||
618 | Note that this setting is inherited across fork/exec. | |
619 | ||
620 | [1] The default default is: the thread keyring if there is one, otherwise | |
621 | the process keyring if there is one, otherwise the session keyring if | |
622 | there is one, otherwise the user default session keyring. | |
623 | ||
624 | ||
1da177e4 LT |
625 | =============== |
626 | KERNEL SERVICES | |
627 | =============== | |
628 | ||
629 | The kernel services for key managment are fairly simple to deal with. They can | |
630 | be broken down into two areas: keys and key types. | |
631 | ||
632 | Dealing with keys is fairly straightforward. Firstly, the kernel service | |
633 | registers its type, then it searches for a key of that type. It should retain | |
634 | the key as long as it has need of it, and then it should release it. For a | |
76d8aeab DH |
635 | filesystem or device file, a search would probably be performed during the open |
636 | call, and the key released upon close. How to deal with conflicting keys due to | |
637 | two different users opening the same file is left to the filesystem author to | |
638 | solve. | |
639 | ||
640 | When accessing a key's payload contents, certain precautions must be taken to | |
641 | prevent access vs modification races. See the section "Notes on accessing | |
642 | payload contents" for more information. | |
1da177e4 LT |
643 | |
644 | (*) To search for a key, call: | |
645 | ||
646 | struct key *request_key(const struct key_type *type, | |
647 | const char *description, | |
648 | const char *callout_string); | |
649 | ||
650 | This is used to request a key or keyring with a description that matches | |
651 | the description specified according to the key type's match function. This | |
652 | permits approximate matching to occur. If callout_string is not NULL, then | |
653 | /sbin/request-key will be invoked in an attempt to obtain the key from | |
654 | userspace. In that case, callout_string will be passed as an argument to | |
655 | the program. | |
656 | ||
657 | Should the function fail error ENOKEY, EKEYEXPIRED or EKEYREVOKED will be | |
658 | returned. | |
659 | ||
3e30148c DH |
660 | If successful, the key will have been attached to the default keyring for |
661 | implicitly obtained request-key keys, as set by KEYCTL_SET_REQKEY_KEYRING. | |
662 | ||
1da177e4 LT |
663 | |
664 | (*) When it is no longer required, the key should be released using: | |
665 | ||
666 | void key_put(struct key *key); | |
667 | ||
668 | This can be called from interrupt context. If CONFIG_KEYS is not set then | |
669 | the argument will not be parsed. | |
670 | ||
671 | ||
672 | (*) Extra references can be made to a key by calling the following function: | |
673 | ||
674 | struct key *key_get(struct key *key); | |
675 | ||
676 | These need to be disposed of by calling key_put() when they've been | |
677 | finished with. The key pointer passed in will be returned. If the pointer | |
678 | is NULL or CONFIG_KEYS is not set then the key will not be dereferenced and | |
679 | no increment will take place. | |
680 | ||
681 | ||
682 | (*) A key's serial number can be obtained by calling: | |
683 | ||
684 | key_serial_t key_serial(struct key *key); | |
685 | ||
686 | If key is NULL or if CONFIG_KEYS is not set then 0 will be returned (in the | |
687 | latter case without parsing the argument). | |
688 | ||
689 | ||
690 | (*) If a keyring was found in the search, this can be further searched by: | |
691 | ||
692 | struct key *keyring_search(struct key *keyring, | |
693 | const struct key_type *type, | |
694 | const char *description) | |
695 | ||
696 | This searches the keyring tree specified for a matching key. Error ENOKEY | |
697 | is returned upon failure. If successful, the returned key will need to be | |
698 | released. | |
699 | ||
700 | ||
701 | (*) To check the validity of a key, this function can be called: | |
702 | ||
703 | int validate_key(struct key *key); | |
704 | ||
705 | This checks that the key in question hasn't expired or and hasn't been | |
706 | revoked. Should the key be invalid, error EKEYEXPIRED or EKEYREVOKED will | |
707 | be returned. If the key is NULL or if CONFIG_KEYS is not set then 0 will be | |
708 | returned (in the latter case without parsing the argument). | |
709 | ||
710 | ||
711 | (*) To register a key type, the following function should be called: | |
712 | ||
713 | int register_key_type(struct key_type *type); | |
714 | ||
715 | This will return error EEXIST if a type of the same name is already | |
716 | present. | |
717 | ||
718 | ||
719 | (*) To unregister a key type, call: | |
720 | ||
721 | void unregister_key_type(struct key_type *type); | |
722 | ||
723 | ||
76d8aeab DH |
724 | =================================== |
725 | NOTES ON ACCESSING PAYLOAD CONTENTS | |
726 | =================================== | |
727 | ||
728 | The simplest payload is just a number in key->payload.value. In this case, | |
729 | there's no need to indulge in RCU or locking when accessing the payload. | |
730 | ||
731 | More complex payload contents must be allocated and a pointer to them set in | |
732 | key->payload.data. One of the following ways must be selected to access the | |
733 | data: | |
734 | ||
735 | (1) Unmodifyable key type. | |
736 | ||
737 | If the key type does not have a modify method, then the key's payload can | |
738 | be accessed without any form of locking, provided that it's known to be | |
739 | instantiated (uninstantiated keys cannot be "found"). | |
740 | ||
741 | (2) The key's semaphore. | |
742 | ||
743 | The semaphore could be used to govern access to the payload and to control | |
744 | the payload pointer. It must be write-locked for modifications and would | |
745 | have to be read-locked for general access. The disadvantage of doing this | |
746 | is that the accessor may be required to sleep. | |
747 | ||
748 | (3) RCU. | |
749 | ||
750 | RCU must be used when the semaphore isn't already held; if the semaphore | |
751 | is held then the contents can't change under you unexpectedly as the | |
752 | semaphore must still be used to serialise modifications to the key. The | |
753 | key management code takes care of this for the key type. | |
754 | ||
755 | However, this means using: | |
756 | ||
757 | rcu_read_lock() ... rcu_dereference() ... rcu_read_unlock() | |
758 | ||
759 | to read the pointer, and: | |
760 | ||
761 | rcu_dereference() ... rcu_assign_pointer() ... call_rcu() | |
762 | ||
763 | to set the pointer and dispose of the old contents after a grace period. | |
764 | Note that only the key type should ever modify a key's payload. | |
765 | ||
766 | Furthermore, an RCU controlled payload must hold a struct rcu_head for the | |
767 | use of call_rcu() and, if the payload is of variable size, the length of | |
768 | the payload. key->datalen cannot be relied upon to be consistent with the | |
769 | payload just dereferenced if the key's semaphore is not held. | |
770 | ||
771 | ||
1da177e4 LT |
772 | =================== |
773 | DEFINING A KEY TYPE | |
774 | =================== | |
775 | ||
776 | A kernel service may want to define its own key type. For instance, an AFS | |
777 | filesystem might want to define a Kerberos 5 ticket key type. To do this, it | |
778 | author fills in a struct key_type and registers it with the system. | |
779 | ||
780 | The structure has a number of fields, some of which are mandatory: | |
781 | ||
782 | (*) const char *name | |
783 | ||
784 | The name of the key type. This is used to translate a key type name | |
785 | supplied by userspace into a pointer to the structure. | |
786 | ||
787 | ||
788 | (*) size_t def_datalen | |
789 | ||
790 | This is optional - it supplies the default payload data length as | |
791 | contributed to the quota. If the key type's payload is always or almost | |
792 | always the same size, then this is a more efficient way to do things. | |
793 | ||
794 | The data length (and quota) on a particular key can always be changed | |
795 | during instantiation or update by calling: | |
796 | ||
797 | int key_payload_reserve(struct key *key, size_t datalen); | |
798 | ||
76d8aeab DH |
799 | With the revised data length. Error EDQUOT will be returned if this is not |
800 | viable. | |
1da177e4 LT |
801 | |
802 | ||
803 | (*) int (*instantiate)(struct key *key, const void *data, size_t datalen); | |
804 | ||
805 | This method is called to attach a payload to a key during construction. | |
76d8aeab DH |
806 | The payload attached need not bear any relation to the data passed to this |
807 | function. | |
1da177e4 LT |
808 | |
809 | If the amount of data attached to the key differs from the size in | |
810 | keytype->def_datalen, then key_payload_reserve() should be called. | |
811 | ||
812 | This method does not have to lock the key in order to attach a payload. | |
813 | The fact that KEY_FLAG_INSTANTIATED is not set in key->flags prevents | |
814 | anything else from gaining access to the key. | |
815 | ||
76d8aeab | 816 | It is safe to sleep in this method. |
1da177e4 LT |
817 | |
818 | ||
819 | (*) int (*duplicate)(struct key *key, const struct key *source); | |
820 | ||
821 | If this type of key can be duplicated, then this method should be | |
76d8aeab DH |
822 | provided. It is called to copy the payload attached to the source into the |
823 | new key. The data length on the new key will have been updated and the | |
824 | quota adjusted already. | |
1da177e4 LT |
825 | |
826 | This method will be called with the source key's semaphore read-locked to | |
76d8aeab DH |
827 | prevent its payload from being changed, thus RCU constraints need not be |
828 | applied to the source key. | |
829 | ||
830 | This method does not have to lock the destination key in order to attach a | |
831 | payload. The fact that KEY_FLAG_INSTANTIATED is not set in key->flags | |
832 | prevents anything else from gaining access to the key. | |
833 | ||
834 | It is safe to sleep in this method. | |
1da177e4 LT |
835 | |
836 | ||
837 | (*) int (*update)(struct key *key, const void *data, size_t datalen); | |
838 | ||
76d8aeab DH |
839 | If this type of key can be updated, then this method should be provided. |
840 | It is called to update a key's payload from the blob of data provided. | |
1da177e4 LT |
841 | |
842 | key_payload_reserve() should be called if the data length might change | |
76d8aeab DH |
843 | before any changes are actually made. Note that if this succeeds, the type |
844 | is committed to changing the key because it's already been altered, so all | |
845 | memory allocation must be done first. | |
846 | ||
847 | The key will have its semaphore write-locked before this method is called, | |
848 | but this only deters other writers; any changes to the key's payload must | |
849 | be made under RCU conditions, and call_rcu() must be used to dispose of | |
850 | the old payload. | |
1da177e4 | 851 | |
76d8aeab DH |
852 | key_payload_reserve() should be called before the changes are made, but |
853 | after all allocations and other potentially failing function calls are | |
854 | made. | |
1da177e4 | 855 | |
76d8aeab | 856 | It is safe to sleep in this method. |
1da177e4 LT |
857 | |
858 | ||
859 | (*) int (*match)(const struct key *key, const void *desc); | |
860 | ||
861 | This method is called to match a key against a description. It should | |
862 | return non-zero if the two match, zero if they don't. | |
863 | ||
864 | This method should not need to lock the key in any way. The type and | |
865 | description can be considered invariant, and the payload should not be | |
866 | accessed (the key may not yet be instantiated). | |
867 | ||
868 | It is not safe to sleep in this method; the caller may hold spinlocks. | |
869 | ||
870 | ||
871 | (*) void (*destroy)(struct key *key); | |
872 | ||
76d8aeab DH |
873 | This method is optional. It is called to discard the payload data on a key |
874 | when it is being destroyed. | |
1da177e4 | 875 | |
76d8aeab DH |
876 | This method does not need to lock the key to access the payload; it can |
877 | consider the key as being inaccessible at this time. Note that the key's | |
878 | type may have been changed before this function is called. | |
1da177e4 LT |
879 | |
880 | It is not safe to sleep in this method; the caller may hold spinlocks. | |
881 | ||
882 | ||
883 | (*) void (*describe)(const struct key *key, struct seq_file *p); | |
884 | ||
885 | This method is optional. It is called during /proc/keys reading to | |
886 | summarise a key's description and payload in text form. | |
887 | ||
76d8aeab DH |
888 | This method will be called with the RCU read lock held. rcu_dereference() |
889 | should be used to read the payload pointer if the payload is to be | |
890 | accessed. key->datalen cannot be trusted to stay consistent with the | |
891 | contents of the payload. | |
892 | ||
893 | The description will not change, though the key's state may. | |
894 | ||
895 | It is not safe to sleep in this method; the RCU read lock is held by the | |
896 | caller. | |
1da177e4 LT |
897 | |
898 | ||
899 | (*) long (*read)(const struct key *key, char __user *buffer, size_t buflen); | |
900 | ||
901 | This method is optional. It is called by KEYCTL_READ to translate the | |
76d8aeab DH |
902 | key's payload into something a blob of data for userspace to deal with. |
903 | Ideally, the blob should be in the same format as that passed in to the | |
904 | instantiate and update methods. | |
1da177e4 LT |
905 | |
906 | If successful, the blob size that could be produced should be returned | |
907 | rather than the size copied. | |
908 | ||
76d8aeab DH |
909 | This method will be called with the key's semaphore read-locked. This will |
910 | prevent the key's payload changing. It is not necessary to use RCU locking | |
911 | when accessing the key's payload. It is safe to sleep in this method, such | |
912 | as might happen when the userspace buffer is accessed. | |
1da177e4 LT |
913 | |
914 | ||
915 | ============================ | |
916 | REQUEST-KEY CALLBACK SERVICE | |
917 | ============================ | |
918 | ||
919 | To create a new key, the kernel will attempt to execute the following command | |
920 | line: | |
921 | ||
922 | /sbin/request-key create <key> <uid> <gid> \ | |
923 | <threadring> <processring> <sessionring> <callout_info> | |
924 | ||
925 | <key> is the key being constructed, and the three keyrings are the process | |
926 | keyrings from the process that caused the search to be issued. These are | |
927 | included for two reasons: | |
928 | ||
929 | (1) There may be an authentication token in one of the keyrings that is | |
930 | required to obtain the key, eg: a Kerberos Ticket-Granting Ticket. | |
931 | ||
932 | (2) The new key should probably be cached in one of these rings. | |
933 | ||
934 | This program should set it UID and GID to those specified before attempting to | |
935 | access any more keys. It may then look around for a user specific process to | |
936 | hand the request off to (perhaps a path held in placed in another key by, for | |
937 | example, the KDE desktop manager). | |
938 | ||
939 | The program (or whatever it calls) should finish construction of the key by | |
940 | calling KEYCTL_INSTANTIATE, which also permits it to cache the key in one of | |
941 | the keyrings (probably the session ring) before returning. Alternatively, the | |
942 | key can be marked as negative with KEYCTL_NEGATE; this also permits the key to | |
943 | be cached in one of the keyrings. | |
944 | ||
945 | If it returns with the key remaining in the unconstructed state, the key will | |
946 | be marked as being negative, it will be added to the session keyring, and an | |
947 | error will be returned to the key requestor. | |
948 | ||
76d8aeab DH |
949 | Supplementary information may be provided from whoever or whatever invoked this |
950 | service. This will be passed as the <callout_info> parameter. If no such | |
1da177e4 LT |
951 | information was made available, then "-" will be passed as this parameter |
952 | instead. | |
953 | ||
954 | ||
955 | Similarly, the kernel may attempt to update an expired or a soon to expire key | |
956 | by executing: | |
957 | ||
958 | /sbin/request-key update <key> <uid> <gid> \ | |
959 | <threadring> <processring> <sessionring> | |
960 | ||
961 | In this case, the program isn't required to actually attach the key to a ring; | |
962 | the rings are provided for reference. |