2 * Copyright (c) 2000-2014 Apple Inc. All rights reserved.
4 * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
6 * This file contains Original Code and/or Modifications of Original Code
7 * as defined in and that are subject to the Apple Public Source License
8 * Version 2.0 (the 'License'). You may not use this file except in
9 * compliance with the License. The rights granted to you under the License
10 * may not be used to create, or enable the creation or redistribution of,
11 * unlawful or unlicensed copies of an Apple operating system, or to
12 * circumvent, violate, or enable the circumvention or violation of, any
13 * terms of an Apple operating system software license agreement.
15 * Please obtain a copy of the License at
16 * http://www.opensource.apple.com/apsl/ and read it before using this file.
18 * The Original Code and all software distributed under the License are
19 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
20 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
21 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
22 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
23 * Please see the License for the specific language governing rights and
24 * limitations under the License.
26 * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
28 /* Copyright (c) 1995 NeXT Computer, Inc. All Rights Reserved */
30 * Copyright (c) 1989, 1991, 1993
31 * The Regents of the University of California. All rights reserved.
33 * Redistribution and use in source and binary forms, with or without
34 * modification, are permitted provided that the following conditions
36 * 1. Redistributions of source code must retain the above copyright
37 * notice, this list of conditions and the following disclaimer.
38 * 2. Redistributions in binary form must reproduce the above copyright
39 * notice, this list of conditions and the following disclaimer in the
40 * documentation and/or other materials provided with the distribution.
41 * 3. All advertising materials mentioning features or use of this software
42 * must display the following acknowledgement:
43 * This product includes software developed by the University of
44 * California, Berkeley and its contributors.
45 * 4. Neither the name of the University nor the names of its contributors
46 * may be used to endorse or promote products derived from this software
47 * without specific prior written permission.
49 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
50 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
51 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
52 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
53 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
54 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
55 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
56 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
57 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
58 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
61 * @(#)mount.h 8.21 (Berkeley) 5/20/95
64 * NOTICE: This file was modified by McAfee Research in 2004 to introduce
65 * support for mandatory and extensible security protections. This notice
66 * is included in support of clause 2.2 (b) of the Apple Public License,
70 #ifndef _SYS_MOUNT_INTERNAL_H_
71 #define _SYS_MOUNT_INTERNAL_H_
73 #include <sys/appleapiopts.h>
75 #include <sys/ucred.h>
77 #include <sys/kernel_types.h>
78 #include <sys/namei.h>
80 #include <sys/queue.h>
82 #include <net/radix.h>
83 #include <sys/socket.h> /* XXX for AF_MAX */
84 #include <sys/vfs_context.h> /* XXX for AF_MAX */
85 #include <sys/mount.h>
86 #include <sys/cdefs.h>
87 #include <sys/sysctl.h>
91 #if defined(__i386__) || defined(__x86_64__)
92 typedef uint64_t pending_io_t
;
93 #define INCR_PENDING_IO(a, b) OSAddAtomic64((int64_t)(a), (int64_t *)&(b));
95 typedef uint32_t pending_io_t
;
96 #define INCR_PENDING_IO(a, b) OSAddAtomic((int32_t)(a), (int32_t *)&(b));
101 * Structure per mounted file system. Each mounted file system has an
102 * array of operations and an instance record. The file systems are
103 * put on a doubly linked list.
105 TAILQ_HEAD(vnodelst
, vnode
);
108 TAILQ_ENTRY(mount
) mnt_list
; /* mount list */
109 int32_t mnt_count
; /* reference on the mount */
110 lck_mtx_t mnt_mlock
; /* mutex that protects mount point */
111 struct vfsops
*mnt_op
; /* operations on fs */
112 struct vfstable
*mnt_vtable
; /* configuration info */
113 struct vnode
*mnt_vnodecovered
; /* vnode we mounted on */
114 struct vnodelst mnt_vnodelist
; /* list of vnodes this mount */
115 struct vnodelst mnt_workerqueue
; /* list of vnodes this mount */
116 struct vnodelst mnt_newvnodes
; /* list of vnodes this mount */
117 uint32_t mnt_flag
; /* flags */
118 uint32_t mnt_kern_flag
; /* kernel only flags */
119 uint32_t mnt_compound_ops
; /* Available compound operations */
120 uint32_t mnt_lflag
; /* mount life cycle flags */
121 uint32_t mnt_maxsymlinklen
; /* max size of short symlink */
122 struct vfsstatfs mnt_vfsstat
; /* cache of filesystem stats */
123 qaddr_t mnt_data
; /* private data */
124 /* Cached values of the IO constraints for the device */
125 uint32_t mnt_maxreadcnt
; /* Max. byte count for read */
126 uint32_t mnt_maxwritecnt
; /* Max. byte count for write */
127 uint32_t mnt_segreadcnt
; /* Max. segment count for read */
128 uint32_t mnt_segwritecnt
; /* Max. segment count for write */
129 uint32_t mnt_maxsegreadsize
; /* Max. segment read size */
130 uint32_t mnt_maxsegwritesize
; /* Max. segment write size */
131 uint32_t mnt_alignmentmask
; /* Mask of bits that aren't addressable via DMA */
132 uint32_t mnt_devblocksize
; /* the underlying device block size */
133 uint32_t mnt_ioqueue_depth
; /* the maxiumum number of commands a device can accept */
134 uint32_t mnt_ioscale
; /* scale the various throttles/limits imposed on the amount of I/O in flight */
135 uint32_t mnt_ioflags
; /* flags for underlying device */
136 uint32_t mnt_minsaturationbytecount
; /* if non-zero, mininum amount of writes (in bytes) needed to max out throughput */
137 pending_io_t mnt_pending_write_size
__attribute__((aligned(sizeof(pending_io_t
)))); /* byte count of pending writes */
138 pending_io_t mnt_pending_read_size
__attribute__((aligned(sizeof(pending_io_t
)))); /* byte count of pending reads */
139 struct timeval mnt_last_write_issued_timestamp
;
140 struct timeval mnt_last_write_completed_timestamp
;
141 int64_t mnt_max_swappin_available
;
143 lck_rw_t mnt_rwlock
; /* mutex readwrite lock */
144 lck_mtx_t mnt_renamelock
; /* mutex that serializes renames that change shape of tree */
145 vnode_t mnt_devvp
; /* the device mounted on for local file systems */
146 uint32_t mnt_devbsdunit
; /* the BSD unit number of the device */
147 uint64_t mnt_throttle_mask
; /* the throttle mask of what devices will be affected by I/O from this mnt */
148 void *mnt_throttle_info
; /* used by the throttle code */
149 int32_t mnt_crossref
; /* refernces to cover lookups crossing into mp */
150 int32_t mnt_iterref
; /* refernces to cover iterations; drained makes it -ve */
152 int32_t mnt_numtriggers
; /* num of trigger vnodes for this mount */
153 vfs_trigger_callback_t
*mnt_triggercallback
;
154 void *mnt_triggerdata
;
156 /* XXX 3762912 hack to support HFS filesystem 'owner' */
160 struct label
*mnt_mntlabel
; /* MAC mount label */
161 struct label
*mnt_fslabel
; /* MAC default fs label */
164 * cache the rootvp of the last mount point
165 * in the chain in the mount struct pointed
166 * to by the vnode sitting in '/'
167 * this cache is used to shortcircuit the
168 * mount chain traversal and allows us
169 * to traverse to the true underlying rootvp
170 * in 1 easy step inside of 'cache_lookup_path'
172 * make sure to validate against the cached vid
173 * in case the rootvp gets stolen away since
174 * we don't take an explicit long term reference
175 * on it when we mount it
177 vnode_t mnt_realrootvp
;
178 uint32_t mnt_realrootvp_vid
;
180 * bumped each time a mount or unmount
181 * occurs... its used to invalidate
182 * 'mnt_realrootvp' from the cache
184 uint32_t mnt_generation
;
186 * if 'MNTK_AUTH_CACHE_TIMEOUT' is
187 * set, then 'mnt_authcache_ttl' is
188 * the time-to-live for the per-vnode authentication cache
189 * on this mount... if zero, no cache is maintained...
190 * if 'MNTK_AUTH_CACHE_TIMEOUT' isn't set, its the
191 * time-to-live for the cached lookup right for
192 * volumes marked 'MNTK_AUTH_OPAQUE'.
194 int mnt_authcache_ttl
;
195 char fstypename_override
[MFSTYPENAMELEN
];
197 uint32_t mnt_iobufinuse
;
201 * default number of seconds to keep cached lookup
202 * rights valid on mounts marked MNTK_AUTH_OPAQUE
204 #define CACHED_LOOKUP_RIGHT_TTL 2
209 #define MNT_IOFLAGS_FUA_SUPPORTED 0x00000001
210 #define MNT_IOFLAGS_UNMAP_SUPPORTED 0x00000002
211 #define MNT_IOFLAGS_IOSCHED_SUPPORTED 0x00000004
212 #define MNT_IOFLAGS_CSUNMAP_SUPPORTED 0x00000008
213 #define MNT_IOFLAGS_SWAPPIN_SUPPORTED 0x00000010
214 #define MNT_IOFLAGS_FUSION_DRIVE 0x00000020
217 * ioqueue depth for devices that don't report one
219 #define MNT_DEFAULT_IOQUEUE_DEPTH 32
221 /* mount point to which dead vps point to */
222 extern struct mount
* dead_mountp
;
225 * Internal filesystem control flags stored in mnt_kern_flag.
227 * MNTK_UNMOUNT locks the mount entry so that name lookup cannot proceed
228 * past the mount point. This keeps the subtree stable during mounts
231 * Note: We are counting down on new bit assignments. This is
232 * because the bits here were broken out from the high bits
233 * of the mount flags.
235 #define MNTK_SWAP_MOUNT 0x00000100 /* we are swapping to this mount */
236 #define MNTK_DENY_READDIREXT 0x00000200 /* Deny Extended-style readdir's for this volume */
237 #define MNTK_PERMIT_UNMOUNT 0x00000400 /* Allow (non-forced) unmounts by UIDs other than the one that mounted the volume */
239 #define MNTK_TYPENAME_OVERRIDE 0x00000800 /* override the fstypename for statfs() */
240 #endif /* NFSCLIENT */
241 #define MNTK_KERNEL_MOUNT 0x00001000 /* mount came from kernel side */
242 #ifdef CONFIG_IMGSRC_ACCESS
243 #define MNTK_HAS_MOVED 0x00002000
244 #define MNTK_BACKS_ROOT 0x00004000
245 #endif /* CONFIG_IMGSRC_ACCESS */
246 #define MNTK_AUTH_CACHE_TTL 0x00008000 /* rights cache has TTL - TTL of 0 disables cache */
247 #define MNTK_PATH_FROM_ID 0x00010000 /* mounted file system supports id-to-path lookups */
248 #define MNTK_UNMOUNT_PREFLIGHT 0x00020000 /* mounted file system wants preflight check during unmount */
249 #define MNTK_NAMED_STREAMS 0x00040000 /* mounted file system supports Named Streams VNOPs */
250 #define MNTK_EXTENDED_ATTRS 0x00080000 /* mounted file system supports Extended Attributes VNOPs */
251 #define MNTK_LOCK_LOCAL 0x00100000 /* advisory locking is done above the VFS itself */
252 #define MNTK_VIRTUALDEV 0x00200000 /* mounted on a virtual device i.e. a disk image */
253 #define MNTK_ROOTDEV 0x00400000 /* this filesystem resides on the same device as the root */
254 #define MNTK_SSD 0x00800000 /* underlying device is of the solid state variety */
255 #define MNTK_UNMOUNT 0x01000000 /* unmount in progress */
256 #define MNTK_MWAIT 0x02000000 /* waiting for unmount to finish */
257 #define MNTK_WANTRDWR 0x04000000 /* upgrade to read/write requested */
259 #define MNT_REVEND 0x08000000 /* Reverse endian FS */
260 #endif /* REV_ENDIAN_FS */
261 #define MNTK_DIR_HARDLINKS 0x10000000 /* mounted file system supports directory hard links */
262 #define MNTK_AUTH_OPAQUE 0x20000000 /* authorisation decisions are not made locally */
263 #define MNTK_AUTH_OPAQUE_ACCESS 0x40000000 /* VNOP_ACCESS is reliable for remote auth */
264 #define MNTK_EXTENDED_SECURITY 0x80000000 /* extended security supported */
266 #define MNT_LNOTRESP 0x00000001 /* mount not responding */
267 #define MNT_LUNMOUNT 0x00000002 /* mount in unmount */
268 #define MNT_LFORCE 0x00000004 /* mount in forced unmount */
269 #define MNT_LDRAIN 0x00000008 /* mount in drain */
270 #define MNT_LITER 0x00000010 /* mount in iteration */
271 #define MNT_LNEWVN 0x00000020 /* mount has new vnodes created */
272 #define MNT_LWAIT 0x00000040 /* wait for unmount op */
273 #define MNT_LITERWAIT 0x00000080 /* mount in iteration */
274 #define MNT_LDEAD 0x00000100 /* mount already unmounted*/
275 #define MNT_LNOSUB 0x00000200 /* submount - no recursion */
279 * Generic file handle
281 #define NFS_MAX_FH_SIZE NFSV4_MAX_FH_SIZE
282 #define NFSV4_MAX_FH_SIZE 128
283 #define NFSV3_MAX_FH_SIZE 64
284 #define NFSV2_MAX_FH_SIZE 32
286 int fh_len
; /* length of file handle */
287 unsigned char fh_data
[NFS_MAX_FH_SIZE
]; /* file handle value */
289 typedef struct fhandle fhandle_t
;
294 * Filesystem configuration information. One of these exists for each
295 * type of filesystem supported by the kernel. These are searched at
296 * mount time to identify the requested filesystem.
299 struct vfsops
*vfc_vfsops
; /* filesystem operations vector */
300 char vfc_name
[MFSNAMELEN
]; /* filesystem type name */
301 int vfc_typenum
; /* historic filesystem type number */
302 int vfc_refcount
; /* number mounted of this type */
303 int vfc_flags
; /* permanent flags */
304 int (*vfc_mountroot
)(mount_t
, vnode_t
, vfs_context_t
); /* if != NULL, routine to mount root */
305 struct vfstable
*vfc_next
; /* next in list */
306 int32_t vfc_reserved1
;
307 int32_t vfc_reserved2
;
308 int vfc_vfsflags
; /* for optional types */
309 void * vfc_descptr
; /* desc table allocated address */
310 int vfc_descsize
; /* size allocated for desc table */
311 struct sysctl_oid
*vfc_sysctl
; /* dynamically registered sysctl node */
315 #define VFC_VFSLOCALARGS 0x002
316 #define VFC_VFSGENERICARGS 0x004
317 #define VFC_VFSNATIVEXATTR 0x010
318 #define VFC_VFSCANMOUNTROOT 0x020
319 #define VFC_VFSPREFLIGHT 0x040
320 #define VFC_VFSREADDIR_EXTENDED 0x080
321 #define VFC_VFS64BITREADY 0x100
322 #define VFC_VFSNOMACLABEL 0x1000
323 #define VFC_VFSVNOP_PAGEINV2 0x2000
324 #define VFC_VFSVNOP_PAGEOUTV2 0x4000
325 #define VFC_VFSVNOP_NOUPDATEID_RENAME 0x8000
326 #define VFC_VFSVNOP_SECLUDE_RENAME 0x10000
328 extern int maxvfstypenum
; /* highest defined filesystem type */
329 extern struct vfstable
*vfsconf
; /* head of list of filesystem types */
330 extern const int maxvfsslots
; /* Maximum statically allocated slots available to be used */
331 extern int numused_vfsslots
; /* number of statically allocated slots already used */
332 extern int numregistered_fses
; /* number of total registered filesystems */
334 /* the following two are xnu private */
335 struct vfstable
* vfstable_add(struct vfstable
*);
336 int vfstable_del(struct vfstable
*);
339 struct vfsmount_args
{
346 void * mnt_fsdata
; /* FS specific */
353 * LP64 *user* version of statfs structure.
354 * NOTE - must be kept in sync with struct statfs in mount.h
356 struct user64_statfs
{
357 short f_otype
; /* TEMPORARY SHADOW COPY OF f_type */
358 short f_oflags
; /* TEMPORARY SHADOW COPY OF f_flags */
359 user64_long_t f_bsize
; /* fundamental file system block size */
360 user64_long_t f_iosize
; /* optimal transfer block size */
361 user64_long_t f_blocks
; /* total data blocks in file system */
362 user64_long_t f_bfree
; /* free blocks in fs */
363 user64_long_t f_bavail
; /* free blocks avail to non-superuser */
364 user64_long_t f_files
; /* total file nodes in file system */
365 user64_long_t f_ffree
; /* free file nodes in fs */
366 fsid_t f_fsid
; /* file system id */
367 uid_t f_owner
; /* user that mounted the filesystem */
368 short f_reserved1
; /* spare for later */
369 short f_type
; /* type of filesystem */
370 user64_long_t f_flags
; /* copy of mount exported flags */
371 user64_long_t f_reserved2
[2]; /* reserved for future use */
372 char f_fstypename
[MFSNAMELEN
]; /* fs type name */
373 char f_mntonname
[MNAMELEN
]; /* directory on which mounted */
374 char f_mntfromname
[MNAMELEN
];/* mounted filesystem */
375 char f_reserved3
; /* For alignment */
376 user64_long_t f_reserved4
[4]; /* For future use */
380 * ILP32 *user* version of statfs structure.
381 * NOTE - must be kept in sync with struct statfs in mount.h
383 struct user32_statfs
{
384 short f_otype
; /* TEMPORARY SHADOW COPY OF f_type */
385 short f_oflags
; /* TEMPORARY SHADOW COPY OF f_flags */
386 user32_long_t f_bsize
; /* fundamental file system block size */
387 user32_long_t f_iosize
; /* optimal transfer block size */
388 user32_long_t f_blocks
; /* total data blocks in file system */
389 user32_long_t f_bfree
; /* free blocks in fs */
390 user32_long_t f_bavail
; /* free blocks avail to non-superuser */
391 user32_long_t f_files
; /* total file nodes in file system */
392 user32_long_t f_ffree
; /* free file nodes in fs */
393 fsid_t f_fsid
; /* file system id */
394 uid_t f_owner
; /* user that mounted the filesystem */
395 short f_reserved1
; /* spare for later */
396 short f_type
; /* type of filesystem */
397 user32_long_t f_flags
; /* copy of mount exported flags */
398 user32_long_t f_reserved2
[2]; /* reserved for future use */
399 char f_fstypename
[MFSNAMELEN
]; /* fs type name */
400 char f_mntonname
[MNAMELEN
]; /* directory on which mounted */
401 char f_mntfromname
[MNAMELEN
];/* mounted filesystem */
402 char f_reserved3
; /* For alignment */
403 user32_long_t f_reserved4
[4]; /* For future use */
407 * throttle I/Os are affected only by normal I/Os happening on the same spindle. Currently we use a 64-bit integer to
408 * represent what devices are affected, so we can handle at most 64 different spindles. Since
409 * throttled I/O is usually useful in non-server environment only, this number is enough in most cases.
411 #define LOWPRI_MAX_NUM_DEV 64
415 extern uint32_t mount_generation
;
416 extern TAILQ_HEAD(mntlist
, mount
) mountlist
;
417 void mount_list_lock(void);
418 void mount_list_unlock(void);
419 void mount_lock_init(mount_t
);
420 void mount_lock_destroy(mount_t
);
421 void mount_lock(mount_t
);
422 void mount_lock_spin(mount_t
);
423 void mount_unlock(mount_t
);
424 void mount_lock_renames(mount_t
);
425 void mount_unlock_renames(mount_t
);
426 void mount_ref(mount_t
, int);
427 void mount_drop(mount_t
, int);
428 int mount_refdrain(mount_t
);
430 /* vfs_rootmountalloc should be kept as a private api */
431 errno_t
vfs_rootmountalloc(const char *, const char *, mount_t
*mpp
);
433 int vfs_mountroot(void);
434 void vfs_unmountall(void);
435 int safedounmount(struct mount
*, int, vfs_context_t
);
436 int dounmount(struct mount
*, int, int, vfs_context_t
);
437 void dounmount_submounts(struct mount
*, int, vfs_context_t
);
439 /* xnu internal api */
440 void mount_dropcrossref(mount_t
, vnode_t
, int);
441 mount_t
mount_lookupby_volfsid(int, int);
442 mount_t
mount_list_lookupby_fsid(fsid_t
*, int, int);
443 int mount_list_add(mount_t
);
444 void mount_list_remove(mount_t
);
445 int mount_iterref(mount_t
, int);
446 int mount_isdrained(mount_t
, int);
447 void mount_iterdrop(mount_t
);
448 void mount_iterdrain(mount_t
);
449 void mount_iterreset(mount_t
);
451 /* Private NFS spi */
452 #define KERNEL_MOUNT_NOAUTH 0x01 /* Don't check the UID of the directory we are mounting on */
453 #define KERNEL_MOUNT_PERMIT_UNMOUNT 0x02 /* Allow (non-forced) unmounts by users other the one who mounted the volume */
454 /* used by snapshot mounting SPI */
455 #define KERNEL_MOUNT_SNAPSHOT 0x04 /* Mounting a snapshot */
456 #if NFSCLIENT || DEVFS || ROUTEFS
458 * NOTE: kernel_mount() does not force MNT_NOSUID, MNT_NOEXEC, or MNT_NODEC for non-privileged
459 * mounting credentials, as the mount(2) system call does.
461 int kernel_mount(char *, vnode_t
, vnode_t
, const char *, void *, size_t, int, uint32_t, vfs_context_t
);
462 boolean_t
vfs_iskernelmount(mount_t
);
465 /* Throttled I/O API. KPI/SPI is in systm.h. */
467 int throttle_get_io_policy(struct uthread
**ut
);
468 int throttle_get_passive_io_policy(struct uthread
**ut
);
469 void *throttle_info_update_by_mount(mount_t mp
);
470 void rethrottle_thread(uthread_t ut
);
473 /* throttled I/O helper function */
474 /* convert the lowest bit to a device index */
475 extern int num_trailing_0(uint64_t n
);
478 extern lck_mtx_t
* sync_mtx_lck
;
480 extern int sync_timeout
;
484 #endif /* !_SYS_MOUNT_INTERNAL_H_ */