1 /* $NetBSD: vfs_init.c,v 1.48 2015/05/06 15:57:08 hannken Exp $ */
2
3 /*-
4 * Copyright (c) 1998, 2000, 2008 The NetBSD Foundation, Inc.
5 * All rights reserved.
6 *
7 * This code is derived from software contributed to The NetBSD Foundation
8 * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
9 * NASA Ames Research Center.
10 *
11 * Redistribution and use in source and binary forms, with or without
12 * modification, are permitted provided that the following conditions
13 * are met:
14 * 1. Redistributions of source code must retain the above copyright
15 * notice, this list of conditions and the following disclaimer.
16 * 2. Redistributions in binary form must reproduce the above copyright
17 * notice, this list of conditions and the following disclaimer in the
18 * documentation and/or other materials provided with the distribution.
19 *
20 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
21 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
22 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
23 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
24 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
25 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
26 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
27 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
28 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
29 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
30 * POSSIBILITY OF SUCH DAMAGE.
31 */
32
33 /*
34 * Copyright (c) 1989, 1993
35 * The Regents of the University of California. All rights reserved.
36 *
37 * This code is derived from software contributed
38 * to Berkeley by John Heidemann of the UCLA Ficus project.
39 *
40 * Source: * @(#)i405_init.c 2.10 92/04/27 UCLA Ficus project
41 *
42 * Redistribution and use in source and binary forms, with or without
43 * modification, are permitted provided that the following conditions
44 * are met:
45 * 1. Redistributions of source code must retain the above copyright
46 * notice, this list of conditions and the following disclaimer.
47 * 2. Redistributions in binary form must reproduce the above copyright
48 * notice, this list of conditions and the following disclaimer in the
49 * documentation and/or other materials provided with the distribution.
50 * 3. Neither the name of the University nor the names of its contributors
51 * may be used to endorse or promote products derived from this software
52 * without specific prior written permission.
53 *
54 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
55 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
56 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
57 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
58 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
59 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
60 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
61 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
62 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
63 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
64 * SUCH DAMAGE.
65 *
66 * @(#)vfs_init.c 8.5 (Berkeley) 5/11/95
67 */
68
69 #include <sys/cdefs.h>
70 __KERNEL_RCSID(0, "$NetBSD: vfs_init.c,v 1.48 2015/05/06 15:57:08 hannken Exp $");
71
72 #include <sys/param.h>
73 #include <sys/mount.h>
74 #include <sys/time.h>
75 #include <sys/vnode.h>
76 #include <sys/stat.h>
77 #include <sys/namei.h>
78 #include <sys/ucred.h>
79 #include <sys/buf.h>
80 #include <sys/errno.h>
81 #include <sys/kmem.h>
82 #include <sys/systm.h>
83 #include <sys/module.h>
84 #include <sys/dirhash.h>
85 #include <sys/sysctl.h>
86 #include <sys/kauth.h>
87
88 /*
89 * Sigh, such primitive tools are these...
90 */
91 #if 0
92 #define DODEBUG(A) A
93 #else
94 #define DODEBUG(A)
95 #endif
96
97 /*
98 * The global list of vnode operations.
99 */
100 extern const struct vnodeop_desc * const vfs_op_descs[];
101
102 /*
103 * These vnodeopv_descs are listed here because they are not
104 * associated with any particular file system, and thus cannot
105 * be initialized by vfs_attach().
106 */
107 extern const struct vnodeopv_desc dead_vnodeop_opv_desc;
108 extern const struct vnodeopv_desc fifo_vnodeop_opv_desc;
109 extern const struct vnodeopv_desc spec_vnodeop_opv_desc;
110
111 const struct vnodeopv_desc * const vfs_special_vnodeopv_descs[] = {
112 &dead_vnodeop_opv_desc,
113 &fifo_vnodeop_opv_desc,
114 &spec_vnodeop_opv_desc,
115 NULL,
116 };
117
118 struct vfs_list_head vfs_list = /* vfs list */
119 LIST_HEAD_INITIALIZER(vfs_list);
120
121 static kauth_listener_t mount_listener;
122
123 /*
124 * This code doesn't work if the defn is **vnodop_defns with cc.
125 * The problem is because of the compiler sometimes putting in an
126 * extra level of indirection for arrays. It's an interesting
127 * "feature" of C.
128 */
129 typedef int (*PFI)(void *);
130
131 /*
132 * A miscellaneous routine.
133 * A generic "default" routine that just returns an error.
134 */
135 /*ARGSUSED*/
136 int
vn_default_error(void * v)137 vn_default_error(void *v)
138 {
139
140 return (EOPNOTSUPP);
141 }
142
143 static struct sysctllog *vfs_sysctllog;
144
145 /*
146 * Top level filesystem related information gathering.
147 */
148 static void
sysctl_vfs_setup(void)149 sysctl_vfs_setup(void)
150 {
151 extern int vfs_magiclinks;
152
153 sysctl_createv(&vfs_sysctllog, 0, NULL, NULL,
154 CTLFLAG_PERMANENT,
155 CTLTYPE_NODE, "generic",
156 SYSCTL_DESCR("Non-specific vfs related information"),
157 NULL, 0, NULL, 0,
158 CTL_VFS, VFS_GENERIC, CTL_EOL);
159 sysctl_createv(&vfs_sysctllog, 0, NULL, NULL,
160 CTLFLAG_PERMANENT,
161 CTLTYPE_STRING, "fstypes",
162 SYSCTL_DESCR("List of file systems present"),
163 sysctl_vfs_generic_fstypes, 0, NULL, 0,
164 CTL_VFS, VFS_GENERIC, CTL_CREATE, CTL_EOL);
165 sysctl_createv(&vfs_sysctllog, 0, NULL, NULL,
166 CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
167 CTLTYPE_INT, "magiclinks",
168 SYSCTL_DESCR("Whether \"magic\" symlinks are expanded"),
169 NULL, 0, &vfs_magiclinks, 0,
170 CTL_VFS, VFS_GENERIC, VFS_MAGICLINKS, CTL_EOL);
171 }
172
173
174 /*
175 * vfs_init.c
176 *
177 * Allocate and fill in operations vectors.
178 *
179 * An undocumented feature of this approach to defining operations is that
180 * there can be multiple entries in vfs_opv_descs for the same operations
181 * vector. This allows third parties to extend the set of operations
182 * supported by another layer in a binary compatibile way. For example,
183 * assume that NFS needed to be modified to support Ficus. NFS has an entry
184 * (probably nfs_vnopdeop_decls) declaring all the operations NFS supports by
185 * default. Ficus could add another entry (ficus_nfs_vnodeop_decl_entensions)
186 * listing those new operations Ficus adds to NFS, all without modifying the
187 * NFS code. (Of couse, the OTW NFS protocol still needs to be munged, but
188 * that is a(whole)nother story.) This is a feature.
189 */
190
191 /*
192 * Init the vector, if it needs it.
193 * Also handle backwards compatibility.
194 */
195 static void
vfs_opv_init_explicit(const struct vnodeopv_desc * vfs_opv_desc)196 vfs_opv_init_explicit(const struct vnodeopv_desc *vfs_opv_desc)
197 {
198 int (**opv_desc_vector)(void *);
199 const struct vnodeopv_entry_desc *opve_descp;
200
201 opv_desc_vector = *(vfs_opv_desc->opv_desc_vector_p);
202
203 for (opve_descp = vfs_opv_desc->opv_desc_ops;
204 opve_descp->opve_op;
205 opve_descp++) {
206 /*
207 * Sanity check: is this operation listed
208 * in the list of operations? We check this
209 * by seeing if its offset is zero. Since
210 * the default routine should always be listed
211 * first, it should be the only one with a zero
212 * offset. Any other operation with a zero
213 * offset is probably not listed in
214 * vfs_op_descs, and so is probably an error.
215 *
216 * A panic here means the layer programmer
217 * has committed the all-too common bug
218 * of adding a new operation to the layer's
219 * list of vnode operations but
220 * not adding the operation to the system-wide
221 * list of supported operations.
222 */
223 if (opve_descp->opve_op->vdesc_offset == 0 &&
224 opve_descp->opve_op->vdesc_offset != VOFFSET(vop_default)) {
225 printf("operation %s not listed in %s.\n",
226 opve_descp->opve_op->vdesc_name, "vfs_op_descs");
227 panic ("vfs_opv_init: bad operation");
228 }
229
230 /*
231 * Fill in this entry.
232 */
233 opv_desc_vector[opve_descp->opve_op->vdesc_offset] =
234 opve_descp->opve_impl;
235 }
236 }
237
238 static void
vfs_opv_init_default(const struct vnodeopv_desc * vfs_opv_desc)239 vfs_opv_init_default(const struct vnodeopv_desc *vfs_opv_desc)
240 {
241 int j;
242 int (**opv_desc_vector)(void *);
243
244 opv_desc_vector = *(vfs_opv_desc->opv_desc_vector_p);
245
246 /*
247 * Force every operations vector to have a default routine.
248 */
249 if (opv_desc_vector[VOFFSET(vop_default)] == NULL)
250 panic("vfs_opv_init: operation vector without default routine.");
251
252 for (j = 0; j < VNODE_OPS_COUNT; j++)
253 if (opv_desc_vector[j] == NULL)
254 opv_desc_vector[j] =
255 opv_desc_vector[VOFFSET(vop_default)];
256 }
257
258 void
vfs_opv_init(const struct vnodeopv_desc * const * vopvdpp)259 vfs_opv_init(const struct vnodeopv_desc * const *vopvdpp)
260 {
261 int (**opv_desc_vector)(void *);
262 int i;
263
264 /*
265 * Allocate the vectors.
266 */
267 for (i = 0; vopvdpp[i] != NULL; i++) {
268 opv_desc_vector =
269 kmem_alloc(VNODE_OPS_COUNT * sizeof(PFI), KM_SLEEP);
270 memset(opv_desc_vector, 0, VNODE_OPS_COUNT * sizeof(PFI));
271 *(vopvdpp[i]->opv_desc_vector_p) = opv_desc_vector;
272 DODEBUG(printf("vector at %p allocated\n",
273 opv_desc_vector_p));
274 }
275
276 /*
277 * ...and fill them in.
278 */
279 for (i = 0; vopvdpp[i] != NULL; i++)
280 vfs_opv_init_explicit(vopvdpp[i]);
281
282 /*
283 * Finally, go back and replace unfilled routines
284 * with their default.
285 */
286 for (i = 0; vopvdpp[i] != NULL; i++)
287 vfs_opv_init_default(vopvdpp[i]);
288 }
289
290 void
vfs_opv_free(const struct vnodeopv_desc * const * vopvdpp)291 vfs_opv_free(const struct vnodeopv_desc * const *vopvdpp)
292 {
293 int i;
294
295 /*
296 * Free the vectors allocated in vfs_opv_init().
297 */
298 for (i = 0; vopvdpp[i] != NULL; i++) {
299 kmem_free(*(vopvdpp[i]->opv_desc_vector_p),
300 VNODE_OPS_COUNT * sizeof(PFI));
301 *(vopvdpp[i]->opv_desc_vector_p) = NULL;
302 }
303 }
304
305 #ifdef DEBUG
306 static void
vfs_op_check(void)307 vfs_op_check(void)
308 {
309 int i;
310
311 DODEBUG(printf("Vnode_interface_init.\n"));
312
313 /*
314 * Check offset of each op.
315 */
316 for (i = 0; vfs_op_descs[i]; i++) {
317 if (vfs_op_descs[i]->vdesc_offset != i)
318 panic("vfs_op_check: vfs_op_desc[] offset mismatch");
319 }
320
321 if (i != VNODE_OPS_COUNT) {
322 panic("vfs_op_check: vnode ops count mismatch (%d != %d)",
323 i, VNODE_OPS_COUNT);
324 }
325
326 DODEBUG(printf ("vfs_opv_numops=%d\n", VNODE_OPS_COUNT));
327 }
328 #endif /* DEBUG */
329
330 /*
331 * Common routine to check if an unprivileged mount is allowed.
332 *
333 * We export just this part (i.e., without the access control) so that if a
334 * secmodel wants to implement finer grained user mounts it can do so without
335 * copying too much code. More elaborate policies (i.e., specific users allowed
336 * to also create devices and/or introduce set-id binaries, or export
337 * file-systems) will require a different implementation.
338 *
339 * This routine is intended to be called from listener context, and as such
340 * does not take credentials as an argument.
341 */
342 int
usermount_common_policy(struct mount * mp,u_long flags)343 usermount_common_policy(struct mount *mp, u_long flags)
344 {
345
346 /* No exporting if unprivileged. */
347 if (flags & MNT_EXPORTED)
348 return EPERM;
349
350 /* Must have 'nosuid' and 'nodev'. */
351 if ((flags & MNT_NODEV) == 0 || (flags & MNT_NOSUID) == 0)
352 return EPERM;
353
354 /* Retain 'noexec'. */
355 if ((mp->mnt_flag & MNT_NOEXEC) && (flags & MNT_NOEXEC) == 0)
356 return EPERM;
357
358 return 0;
359 }
360
361 static int
mount_listener_cb(kauth_cred_t cred,kauth_action_t action,void * cookie,void * arg0,void * arg1,void * arg2,void * arg3)362 mount_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie,
363 void *arg0, void *arg1, void *arg2, void *arg3)
364 {
365 int result;
366 enum kauth_system_req req;
367
368 result = KAUTH_RESULT_DEFER;
369 req = (enum kauth_system_req)(uintptr_t)arg0;
370
371 if (action != KAUTH_SYSTEM_MOUNT)
372 return result;
373
374 if (req == KAUTH_REQ_SYSTEM_MOUNT_GET)
375 result = KAUTH_RESULT_ALLOW;
376 else if (req == KAUTH_REQ_SYSTEM_MOUNT_DEVICE) {
377 vnode_t *devvp = arg2;
378 mode_t access_mode = (mode_t)(unsigned long)arg3;
379 int error;
380
381 error = VOP_ACCESS(devvp, access_mode, cred);
382 if (!error)
383 result = KAUTH_RESULT_ALLOW;
384 }
385
386 return result;
387 }
388
389 /*
390 * Initialize the vnode structures and initialize each file system type.
391 */
392 void
vfsinit(void)393 vfsinit(void)
394 {
395
396 /*
397 * Attach sysctl nodes
398 */
399 sysctl_vfs_setup();
400
401 /*
402 * Initialize the namei pathname buffer pool and cache.
403 */
404 pnbuf_cache = pool_cache_init(MAXPATHLEN, 0, 0, 0, "pnbufpl",
405 NULL, IPL_NONE, NULL, NULL, NULL);
406 KASSERT(pnbuf_cache != NULL);
407
408 /*
409 * Initialize the vnode table
410 */
411 vntblinit();
412
413 /*
414 * Initialize the vnode name cache
415 */
416 nchinit();
417
418 #ifdef DEBUG
419 /*
420 * Check the list of vnode operations.
421 */
422 vfs_op_check();
423 #endif
424
425 /*
426 * Initialize the special vnode operations.
427 */
428 vfs_opv_init(vfs_special_vnodeopv_descs);
429
430 /*
431 * Initialise generic dirhash.
432 */
433 dirhash_init();
434
435 /*
436 * Initialise VFS hooks.
437 */
438 vfs_hooks_init();
439
440 mount_listener = kauth_listen_scope(KAUTH_SCOPE_SYSTEM,
441 mount_listener_cb, NULL);
442
443 /*
444 * Establish each file system which was statically
445 * included in the kernel.
446 */
447 module_init_class(MODULE_CLASS_VFS);
448 }
449
450 /*
451 * Drop a reference to a file system type.
452 */
453 void
vfs_delref(struct vfsops * vfs)454 vfs_delref(struct vfsops *vfs)
455 {
456
457 mutex_enter(&vfs_list_lock);
458 vfs->vfs_refcount--;
459 mutex_exit(&vfs_list_lock);
460 }
461
462 /*
463 * Establish a file system and initialize it.
464 */
465 int
vfs_attach(struct vfsops * vfs)466 vfs_attach(struct vfsops *vfs)
467 {
468 struct vfsops *v;
469 int error = 0;
470
471 mutex_enter(&vfs_list_lock);
472
473 /*
474 * Make sure this file system doesn't already exist.
475 */
476 LIST_FOREACH(v, &vfs_list, vfs_list) {
477 if (strcmp(vfs->vfs_name, v->vfs_name) == 0) {
478 error = EEXIST;
479 goto out;
480 }
481 }
482
483 /*
484 * Initialize the vnode operations for this file system.
485 */
486 vfs_opv_init(vfs->vfs_opv_descs);
487
488 /*
489 * Now initialize the file system itself.
490 */
491 (*vfs->vfs_init)();
492
493 /*
494 * ...and link it into the kernel's list.
495 */
496 LIST_INSERT_HEAD(&vfs_list, vfs, vfs_list);
497
498 /*
499 * Sanity: make sure the reference count is 0.
500 */
501 vfs->vfs_refcount = 0;
502 out:
503 mutex_exit(&vfs_list_lock);
504 return (error);
505 }
506
507 /*
508 * Remove a file system from the kernel.
509 */
510 int
vfs_detach(struct vfsops * vfs)511 vfs_detach(struct vfsops *vfs)
512 {
513 struct vfsops *v;
514 int error = 0;
515
516 mutex_enter(&vfs_list_lock);
517
518 /*
519 * Make sure no one is using the filesystem.
520 */
521 if (vfs->vfs_refcount != 0) {
522 error = EBUSY;
523 goto out;
524 }
525
526 /*
527 * ...and remove it from the kernel's list.
528 */
529 LIST_FOREACH(v, &vfs_list, vfs_list) {
530 if (v == vfs) {
531 LIST_REMOVE(v, vfs_list);
532 break;
533 }
534 }
535
536 if (v == NULL) {
537 error = ESRCH;
538 goto out;
539 }
540
541 /*
542 * Now run the file system-specific cleanups.
543 */
544 (*vfs->vfs_done)();
545
546 /*
547 * Free the vnode operations vector.
548 */
549 vfs_opv_free(vfs->vfs_opv_descs);
550 out:
551 mutex_exit(&vfs_list_lock);
552 return (error);
553 }
554
555 void
vfs_reinit(void)556 vfs_reinit(void)
557 {
558 struct vfsops *vfs;
559
560 mutex_enter(&vfs_list_lock);
561 LIST_FOREACH(vfs, &vfs_list, vfs_list) {
562 if (vfs->vfs_reinit) {
563 vfs->vfs_refcount++;
564 mutex_exit(&vfs_list_lock);
565 (*vfs->vfs_reinit)();
566 mutex_enter(&vfs_list_lock);
567 vfs->vfs_refcount--;
568 }
569 }
570 mutex_exit(&vfs_list_lock);
571 }
572