bsd/sys/mount_internal.h

   1 /*
   2  * Copyright (c) 2000-2007 Apple Inc. All rights reserved.
   3  *
   4  * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
   5  *
   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.
  14  *
  15  * Please obtain a copy of the License at
  16  * http://www.opensource.apple.com/apsl/ and read it before using this file.
  17  *
  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.
  25  *
  26  * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  27  */
  28 /* Copyright (c) 1995 NeXT Computer, Inc. All Rights Reserved */
  29 /*
  30  * Copyright (c) 1989, 1991, 1993
  31  *      The Regents of the University of California.  All rights reserved.
  32  *
  33  * Redistribution and use in source and binary forms, with or without
  34  * modification, are permitted provided that the following conditions
  35  * are met:
  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.
  48  *
  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
  59  * SUCH DAMAGE.
  60  *
  61  *      @(#)mount.h     8.21 (Berkeley) 5/20/95
  62  */
  63 /*
  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,
  67  * Version 2.0.
  68  */
  69
  70 #ifndef _SYS_MOUNT_INTERNAL_H_
  71 #define _SYS_MOUNT_INTERNAL_H_
  72
  73 #include <sys/appleapiopts.h>
  74 #ifndef KERNEL
  75 #include <sys/ucred.h>
  76 #else
  77 #include <sys/kernel_types.h>
  78 #include <sys/namei.h>
  79 #endif
  80 #include <sys/queue.h>
  81 #include <sys/lock.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
  88 struct label;
  89
  90 /*
  91  * Structure per mounted file system.  Each mounted file system has an
  92  * array of operations and an instance record.  The file systems are
  93  * put on a doubly linked list.
  94  */
  95 TAILQ_HEAD(vnodelst, vnode);
  96
  97 struct mount {
  98         TAILQ_ENTRY(mount) mnt_list;            /* mount list */
  99         int32_t         mnt_count;              /* reference on the mount */
 100         lck_mtx_t       mnt_mlock;              /* mutex that protects mount point */
 101         struct vfsops   *mnt_op;                /* operations on fs */
 102         struct vfstable *mnt_vtable;            /* configuration info */
 103         struct vnode    *mnt_vnodecovered;      /* vnode we mounted on */
 104         struct vnodelst mnt_vnodelist;          /* list of vnodes this mount */
 105         struct vnodelst mnt_workerqueue;                /* list of vnodes this mount */
 106         struct vnodelst mnt_newvnodes;          /* list of vnodes this mount */
 107         int             mnt_flag;               /* flags */
 108         int             mnt_kern_flag;          /* kernel only flags */
 109         int             mnt_lflag;                      /* mount life cycle flags */
 110         int             mnt_maxsymlinklen;      /* max size of short symlink */
 111         struct vfsstatfs        mnt_vfsstat;            /* cache of filesystem stats */
 112         qaddr_t         mnt_data;               /* private data */
 113         /* Cached values of the IO constraints for the device */
 114         u_int32_t       mnt_maxreadcnt;         /* Max. byte count for read */
 115         u_int32_t       mnt_maxwritecnt;        /* Max. byte count for write */
 116         u_int32_t       mnt_segreadcnt;         /* Max. segment count for read */
 117         u_int32_t       mnt_segwritecnt;        /* Max. segment count for write */
 118         u_int32_t       mnt_maxsegreadsize;     /* Max. segment read size  */
 119         u_int32_t       mnt_maxsegwritesize;    /* Max. segment write size */
 120         u_int32_t       mnt_alignmentmask;      /* Mask of bits that aren't addressable via DMA */
 121         u_int32_t       mnt_devblocksize;       /* the underlying device block size */
 122         u_int32_t       mnt_ioflags;            /* flags for  underlying device */
 123         lck_rw_t        mnt_rwlock;             /* mutex readwrite lock */
 124         lck_mtx_t       mnt_renamelock;         /* mutex that serializes renames that change shape of tree */
 125         vnode_t         mnt_devvp;              /* the device mounted on for local file systems */
 126         int32_t         mnt_crossref;           /* refernces to cover lookups  crossing into mp */
 127         int32_t         mnt_iterref;            /* refernces to cover iterations; drained makes it -ve  */
 128
 129         /* XXX 3762912 hack to support HFS filesystem 'owner' */
 130         uid_t           mnt_fsowner;
 131         gid_t           mnt_fsgroup;
 132
 133         struct label    *mnt_mntlabel;          /* MAC mount label */
 134         struct label    *mnt_fslabel;           /* MAC default fs label */
 135
 136         /*
 137          * cache the rootvp of the last mount point
 138          * in the chain in the mount struct pointed
 139          * to by the vnode sitting in '/'
 140          * this cache is used to shortcircuit the
 141          * mount chain traversal and allows us
 142          * to traverse to the true underlying rootvp
 143          * in 1 easy step inside of 'cache_lookup_path'
 144          *
 145          * make sure to validate against the cached vid
 146          * in case the rootvp gets stolen away since
 147          * we don't take an explicit long term reference
 148          * on it when we mount it
 149          */
 150         vnode_t         mnt_realrootvp;
 151         int             mnt_realrootvp_vid;
 152         /*
 153          * bumped each time a mount or unmount
 154          * occurs... its used to invalidate
 155          * 'mnt_realrootvp' from the cache
 156          */
 157         int             mnt_generation;
 158         /*
 159          * if 'MNTK_AUTH_CACHE_TIMEOUT' is
 160          * set, then 'mnt_authcache_ttl' is
 161          * the time-to-live for the per-vnode authentication cache
 162          * on this mount... if zero, no cache is maintained...
 163          * if 'MNTK_AUTH_CACHE_TIMEOUT' isn't set, its the
 164          * time-to-live for the cached lookup right for
 165          * volumes marked 'MNTK_AUTH_OPAQUE'.
 166          */
 167         int             mnt_authcache_ttl;
 168         /*
 169          * The proc structure pointer and process ID form a
 170          * sufficiently unique duple identifying the process
 171          * hosting this mount point. Set by vfs_markdependency()
 172          * and utilized in new_vnode() to avoid reclaiming vnodes
 173          * with this dependency (radar 5192010).
 174          */
 175         pid_t           mnt_dependent_pid;
 176         void            *mnt_dependent_process;
 177
 178         struct timeval  last_normal_IO_timestamp;
 179 };
 180
 181 /*
 182  * default number of seconds to keep cached lookup
 183  * rights valid on mounts marked MNTK_AUTH_OPAQUE
 184  */
 185 #define CACHED_LOOKUP_RIGHT_TTL         2
 186
 187 /*
 188  * ioflags
 189  */
 190 #define MNT_IOFLAGS_FUA_SUPPORTED       0x00000001
 191
 192
 193 /* XXX 3762912 hack to support HFS filesystem 'owner' */
 194 #define vfs_setowner(_mp, _uid, _gid)   do {(_mp)->mnt_fsowner = (_uid); (_mp)->mnt_fsgroup = (_gid); } while (0)
 195
 196
 197 /* mount point to which dead vps point to */
 198 extern struct mount * dead_mountp;
 199
 200 /*
 201  * Internal filesystem control flags stored in mnt_kern_flag.
 202  *
 203  * MNTK_UNMOUNT locks the mount entry so that name lookup cannot proceed
 204  * past the mount point.  This keeps the subtree stable during mounts
 205  * and unmounts.
 206  *
 207  * Note:        We are counting down on new bit assignments.  This is
 208  *              because the bits here were broken out from the high bits
 209  *              of the mount flags.
 210  */
 211 #define MNTK_AUTH_CACHE_TTL     0x00008000      /* rights cache has TTL - TTL of 0 disables cache */
 212 #define MNTK_PATH_FROM_ID       0x00010000      /* mounted file system supports id-to-path lookups */
 213 #define MNTK_UNMOUNT_PREFLIGHT  0x00020000      /* mounted file system wants preflight check during unmount */
 214 #define MNTK_NAMED_STREAMS      0x00040000      /* mounted file system supports Named Streams VNOPs */
 215 #define MNTK_EXTENDED_ATTRS     0x00080000      /* mounted file system supports Extended Attributes VNOPs */
 216 #define MNTK_LOCK_LOCAL         0x00100000      /* advisory locking is done above the VFS itself */
 217 #define MNTK_VIRTUALDEV         0x00200000      /* mounted on a virtual device i.e. a disk image */
 218 #define MNTK_ROOTDEV            0x00400000      /* this filesystem resides on the same device as the root */
 219 #define MNTK_UNMOUNT            0x01000000      /* unmount in progress */
 220 #define MNTK_MWAIT              0x02000000      /* waiting for unmount to finish */
 221 #define MNTK_WANTRDWR           0x04000000      /* upgrade to read/write requested */
 222 #if REV_ENDIAN_FS
 223 #define MNT_REVEND              0x08000000      /* Reverse endian FS */
 224 #endif /* REV_ENDIAN_FS */
 225 #define MNTK_FRCUNMOUNT         0x10000000      /* Forced unmount wanted. */
 226 #define MNTK_AUTH_OPAQUE        0x20000000      /* authorisation decisions are not made locally */
 227 #define MNTK_AUTH_OPAQUE_ACCESS 0x40000000      /* VNOP_ACCESS is reliable for remote auth */
 228 #define MNTK_EXTENDED_SECURITY  0x80000000      /* extended security supported */
 229
 230 #define MNT_LBUSY               0x00000001      /* mount is busy */
 231 #define MNT_LUNMOUNT            0x00000002      /* mount in unmount */
 232 #define MNT_LFORCE              0x00000004      /* mount in forced unmount */
 233 #define MNT_LDRAIN              0x00000008      /* mount in drain */
 234 #define MNT_LITER               0x00000010      /* mount in iteration */
 235 #define MNT_LNEWVN              0x00000020      /* mount has new vnodes created */
 236 #define MNT_LWAIT               0x00000040      /* wait for unmount op */
 237 #define MNT_LITERWAIT           0x00000080      /* mount in iteration */
 238 #define MNT_LDEAD               0x00000100      /* mount already unmounted*/
 239
 240
 241 /*
 242  * Generic file handle
 243  */
 244 #define NFS_MAX_FH_SIZE         NFSV4_MAX_FH_SIZE
 245 #define NFSV4_MAX_FH_SIZE       128
 246 #define NFSV3_MAX_FH_SIZE       64
 247 #define NFSV2_MAX_FH_SIZE       32
 248 struct fhandle {
 249         int             fh_len;                         /* length of file handle */
 250         unsigned char   fh_data[NFS_MAX_FH_SIZE];       /* file handle value */
 251 };
 252 typedef struct fhandle  fhandle_t;
 253
 254
 255
 256 /*
 257  * Filesystem configuration information. One of these exists for each
 258  * type of filesystem supported by the kernel. These are searched at
 259  * mount time to identify the requested filesystem.
 260  */
 261 struct vfstable {
 262 /* THE FOLLOWING SHOULD KEEP THE SAME FOR user compat with sysctl */
 263         struct  vfsops *vfc_vfsops;     /* filesystem operations vector */
 264         char    vfc_name[MFSNAMELEN];   /* filesystem type name */
 265         int     vfc_typenum;            /* historic filesystem type number */
 266         int     vfc_refcount;           /* number mounted of this type */
 267         int     vfc_flags;              /* permanent flags */
 268         int     (*vfc_mountroot)(mount_t, vnode_t, vfs_context_t);      /* if != NULL, routine to mount root */
 269         struct  vfstable *vfc_next;     /* next in list */
 270 /* Till the above we SHOULD KEEP THE SAME FOR user compat with sysctl */
 271         int         vfc_threadsafe;     /* FS is thread & premeption safe */
 272         lck_mtx_t   vfc_lock;           /* for non-threaded file systems */
 273         int             vfc_vfsflags;   /* for optional types */
 274         void *          vfc_descptr;    /* desc table allocated address */
 275         int                     vfc_descsize;   /* size allocated for desc table */
 276         int             vfc_64bitready; /* The file system is ready for 64bit */
 277 };
 278
 279 /* vfc_vfsflags: */
 280 #define VFC_VFSLOCALARGS        0x02
 281 #define VFC_VFSGENERICARGS      0x04
 282 #define VFC_VFSNATIVEXATTR      0x10
 283 #define VFC_VFSDIRLINKS         0x20
 284 #define VFC_VFSPREFLIGHT        0x40
 285 #define VFC_VFSREADDIR_EXTENDED 0x80
 286 #define VFC_VFSNOMACLABEL       0x1000
 287
 288 extern int maxvfsconf;          /* highest defined filesystem type */
 289 extern struct vfstable  *vfsconf;       /* head of list of filesystem types */
 290 extern int maxvfsslots;         /* Maximum slots available to be used */
 291 extern int numused_vfsslots;    /* number of slots already used */
 292
 293 /* the following two are xnu private */
 294 struct vfstable *       vfstable_add(struct     vfstable *);
 295 int     vfstable_del(struct vfstable *);
 296
 297
 298 struct vfsmount_args {
 299         union {
 300                 struct {
 301                         char * mnt_fspec;
 302                         void * mnt_fsdata;
 303                 } mnt_localfs_args;
 304                 struct {
 305                         void *  mnt_fsdata;             /* FS specific */
 306                 } mnt_remotefs_args;
 307         } mountfs_args;
 308 };
 309
 310
 311 /*
 312  * LP64 version of statfs structure.
 313  * NOTE - must be kept in sync with struct statfs in mount.h
 314  */
 315 struct user_statfs {
 316         short           f_otype;                /* TEMPORARY SHADOW COPY OF f_type */
 317         short           f_oflags;               /* TEMPORARY SHADOW COPY OF f_flags */
 318         user_long_t     f_bsize __attribute((aligned(8)));              /* fundamental file system block size */
 319         user_long_t     f_iosize;               /* optimal transfer block size */
 320         user_long_t     f_blocks;               /* total data blocks in file system */
 321         user_long_t     f_bfree;                /* free blocks in fs */
 322         user_long_t     f_bavail;               /* free blocks avail to non-superuser */
 323         user_long_t     f_files;                /* total file nodes in file system */
 324         user_long_t     f_ffree;                /* free file nodes in fs */
 325         fsid_t          f_fsid;                 /* file system id */
 326         uid_t           f_owner;                /* user that mounted the filesystem */
 327         short           f_reserved1;    /* spare for later */
 328         short           f_type;                 /* type of filesystem */
 329     user_long_t f_flags;                /* copy of mount exported flags */
 330         user_long_t f_reserved2[2];     /* reserved for future use */
 331         char            f_fstypename[MFSNAMELEN]; /* fs type name */
 332         char            f_mntonname[MNAMELEN];  /* directory on which mounted */
 333         char            f_mntfromname[MNAMELEN];/* mounted filesystem */
 334 #if COMPAT_GETFSSTAT
 335         char            f_reserved3[0]; /* For alignment */
 336         user_long_t     f_reserved4[0]; /* For future use */
 337 #else
 338         char            f_reserved3;    /* For alignment */
 339         user_long_t     f_reserved4[4] __attribute((aligned(8)));       /* For future use */
 340 #endif
 341 };
 342
 343 __BEGIN_DECLS
 344
 345 extern int mount_generation;
 346 extern TAILQ_HEAD(mntlist, mount) mountlist;
 347 void mount_list_lock(void);
 348 void mount_list_unlock(void);
 349 void mount_lock_init(mount_t);
 350 void mount_lock_destroy(mount_t);
 351 void mount_lock(mount_t);
 352 void mount_unlock(mount_t);
 353 void mount_lock_renames(mount_t);
 354 void mount_unlock_renames(mount_t);
 355 void mount_ref(mount_t, int);
 356 void mount_drop(mount_t, int);
 357 int  mount_refdrain(mount_t);
 358
 359 /* vfs_rootmountalloc should be kept as a private api */
 360 errno_t vfs_rootmountalloc(const char *, const char *, mount_t *mpp);
 361 errno_t vfs_init_io_attributes(vnode_t, mount_t);
 362
 363 int     vfs_mountroot(void);
 364 void    vfs_unmountall(void);
 365 int     safedounmount(struct mount *, int, vfs_context_t);
 366 int     dounmount(struct mount *, int, int, vfs_context_t);
 367
 368 /* xnuy internal api */
 369 void  mount_dropcrossref(mount_t, vnode_t, int);
 370 mount_t mount_lookupby_volfsid(int, int);
 371 mount_t mount_list_lookupby_fsid(fsid_t *, int, int);
 372 void mount_list_add(mount_t);
 373 void mount_list_remove(mount_t);
 374 int  mount_iterref(mount_t, int);
 375 int  mount_isdrained(mount_t, int);
 376 void mount_iterdrop(mount_t);
 377 void mount_iterdrain(mount_t);
 378 void mount_iterreset(mount_t);
 379
 380 __END_DECLS
 381
 382 #endif /* !_SYS_MOUNT_INTERNAL_H_ */