X-Git-Url: https://git.saurik.com/apple/xnu.git/blobdiff_plain/0c530ab8987f0ae6a1a3d9284f40182b88852816..b0d623f7f2ae71ed96e60569f61f9a9a27016e80:/bsd/sys/mount_internal.h diff --git a/bsd/sys/mount_internal.h b/bsd/sys/mount_internal.h index ca490eb6e..6cca245d5 100644 --- a/bsd/sys/mount_internal.h +++ b/bsd/sys/mount_internal.h @@ -1,23 +1,29 @@ /* - * Copyright (c) 2000-2005 Apple Computer, Inc. All rights reserved. + * Copyright (c) 2000-2007 Apple Inc. All rights reserved. * - * @APPLE_LICENSE_HEADER_START@ + * @APPLE_OSREFERENCE_LICENSE_HEADER_START@ * - * The contents of this file constitute Original Code as defined in and - * are subject to the Apple Public Source License Version 1.1 (the - * "License"). You may not use this file except in compliance with the - * License. Please obtain a copy of the License at - * http://www.apple.com/publicsource and read it before using this file. + * This file contains Original Code and/or Modifications of Original Code + * as defined in and that are subject to the Apple Public Source License + * Version 2.0 (the 'License'). You may not use this file except in + * compliance with the License. The rights granted to you under the License + * may not be used to create, or enable the creation or redistribution of, + * unlawful or unlicensed copies of an Apple operating system, or to + * circumvent, violate, or enable the circumvention or violation of, any + * terms of an Apple operating system software license agreement. * - * This Original Code and all software distributed under the License are - * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER + * Please obtain a copy of the License at + * http://www.opensource.apple.com/apsl/ and read it before using this file. + * + * The Original Code and all software distributed under the License are + * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT. Please see the - * License for the specific language governing rights and limitations - * under the License. + * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT. + * Please see the License for the specific language governing rights and + * limitations under the License. * - * @APPLE_LICENSE_HEADER_END@ + * @APPLE_OSREFERENCE_LICENSE_HEADER_END@ */ /* Copyright (c) 1995 NeXT Computer, Inc. All Rights Reserved */ /* @@ -54,6 +60,12 @@ * * @(#)mount.h 8.21 (Berkeley) 5/20/95 */ +/* + * NOTICE: This file was modified by McAfee Research in 2004 to introduce + * support for mandatory and extensible security protections. This notice + * is included in support of clause 2.2 (b) of the Apple Public License, + * Version 2.0. + */ #ifndef _SYS_MOUNT_INTERNAL_H_ #define _SYS_MOUNT_INTERNAL_H_ @@ -73,6 +85,17 @@ #include #include +struct label; + +#if defined(__i386__) || defined(__x86_64__) +typedef uint64_t pending_io_t; +#define INCR_PENDING_IO(a, b) OSAddAtomic64((int64_t)(a), (int64_t *)&(b)); +#else +typedef uint32_t pending_io_t; +#define INCR_PENDING_IO(a, b) OSAddAtomic((int32_t)(a), (int32_t *)&(b)); +#endif + + /* * Structure per mounted file system. Each mounted file system has an * array of operations and an instance record. The file systems are @@ -90,30 +113,101 @@ struct mount { struct vnodelst mnt_vnodelist; /* list of vnodes this mount */ struct vnodelst mnt_workerqueue; /* list of vnodes this mount */ struct vnodelst mnt_newvnodes; /* list of vnodes this mount */ - int mnt_flag; /* flags */ - int mnt_kern_flag; /* kernel only flags */ - int mnt_lflag; /* mount life cycle flags */ - int mnt_maxsymlinklen; /* max size of short symlink */ + uint32_t mnt_flag; /* flags */ + uint32_t mnt_kern_flag; /* kernel only flags */ + uint32_t mnt_lflag; /* mount life cycle flags */ + uint32_t mnt_maxsymlinklen; /* max size of short symlink */ struct vfsstatfs mnt_vfsstat; /* cache of filesystem stats */ qaddr_t mnt_data; /* private data */ /* Cached values of the IO constraints for the device */ - u_int32_t mnt_maxreadcnt; /* Max. byte count for read */ - u_int32_t mnt_maxwritecnt; /* Max. byte count for write */ - u_int32_t mnt_segreadcnt; /* Max. segment count for read */ - u_int32_t mnt_segwritecnt; /* Max. segment count for write */ - u_int32_t mnt_maxsegreadsize; /* Max. segment read size */ - u_int32_t mnt_maxsegwritesize; /* Max. segment write size */ - u_int32_t mnt_devblocksize; /* the underlying device block size */ + uint32_t mnt_maxreadcnt; /* Max. byte count for read */ + uint32_t mnt_maxwritecnt; /* Max. byte count for write */ + uint32_t mnt_segreadcnt; /* Max. segment count for read */ + uint32_t mnt_segwritecnt; /* Max. segment count for write */ + uint32_t mnt_maxsegreadsize; /* Max. segment read size */ + uint32_t mnt_maxsegwritesize; /* Max. segment write size */ + uint32_t mnt_alignmentmask; /* Mask of bits that aren't addressable via DMA */ + uint32_t mnt_devblocksize; /* the underlying device block size */ + uint32_t mnt_ioqueue_depth; /* the maxiumum number of commands a device can accept */ + uint32_t mnt_ioscale; /* scale the various throttles/limits imposed on the amount of I/O in flight */ + uint32_t mnt_ioflags; /* flags for underlying device */ + pending_io_t mnt_pending_write_size; /* byte count of pending writes */ + pending_io_t mnt_pending_read_size; /* byte count of pending reads */ + lck_rw_t mnt_rwlock; /* mutex readwrite lock */ - lck_mtx_t mnt_renamelock; /* mutex that serializes renames that change shape of tree */ + lck_mtx_t mnt_renamelock; /* mutex that serializes renames that change shape of tree */ vnode_t mnt_devvp; /* the device mounted on for local file systems */ + uint32_t mnt_devbsdunit; /* the BSD unit number of the device */ + void *mnt_throttle_info; /* used by the throttle code */ int32_t mnt_crossref; /* refernces to cover lookups crossing into mp */ int32_t mnt_iterref; /* refernces to cover iterations; drained makes it -ve */ /* XXX 3762912 hack to support HFS filesystem 'owner' */ uid_t mnt_fsowner; gid_t mnt_fsgroup; + + struct label *mnt_mntlabel; /* MAC mount label */ + struct label *mnt_fslabel; /* MAC default fs label */ + + /* + * cache the rootvp of the last mount point + * in the chain in the mount struct pointed + * to by the vnode sitting in '/' + * this cache is used to shortcircuit the + * mount chain traversal and allows us + * to traverse to the true underlying rootvp + * in 1 easy step inside of 'cache_lookup_path' + * + * make sure to validate against the cached vid + * in case the rootvp gets stolen away since + * we don't take an explicit long term reference + * on it when we mount it + */ + vnode_t mnt_realrootvp; + uint32_t mnt_realrootvp_vid; + /* + * bumped each time a mount or unmount + * occurs... its used to invalidate + * 'mnt_realrootvp' from the cache + */ + uint32_t mnt_generation; + /* + * if 'MNTK_AUTH_CACHE_TIMEOUT' is + * set, then 'mnt_authcache_ttl' is + * the time-to-live for the per-vnode authentication cache + * on this mount... if zero, no cache is maintained... + * if 'MNTK_AUTH_CACHE_TIMEOUT' isn't set, its the + * time-to-live for the cached lookup right for + * volumes marked 'MNTK_AUTH_OPAQUE'. + */ + int mnt_authcache_ttl; + /* + * The proc structure pointer and process ID form a + * sufficiently unique duple identifying the process + * hosting this mount point. Set by vfs_markdependency() + * and utilized in new_vnode() to avoid reclaiming vnodes + * with this dependency (radar 5192010). + */ + pid_t mnt_dependent_pid; + void *mnt_dependent_process; }; + +/* + * default number of seconds to keep cached lookup + * rights valid on mounts marked MNTK_AUTH_OPAQUE + */ +#define CACHED_LOOKUP_RIGHT_TTL 2 + +/* + * ioflags + */ +#define MNT_IOFLAGS_FUA_SUPPORTED 0x00000001 + +/* + * ioqueue depth for devices that don't report one + */ +#define MNT_DEFAULT_IOQUEUE_DEPTH 32 + /* XXX 3762912 hack to support HFS filesystem 'owner' */ #define vfs_setowner(_mp, _uid, _gid) do {(_mp)->mnt_fsowner = (_uid); (_mp)->mnt_fsgroup = (_gid); } while (0) @@ -133,6 +227,11 @@ extern struct mount * dead_mountp; * because the bits here were broken out from the high bits * of the mount flags. */ +#define MNTK_AUTH_CACHE_TTL 0x00008000 /* rights cache has TTL - TTL of 0 disables cache */ +#define MNTK_PATH_FROM_ID 0x00010000 /* mounted file system supports id-to-path lookups */ +#define MNTK_UNMOUNT_PREFLIGHT 0x00020000 /* mounted file system wants preflight check during unmount */ +#define MNTK_NAMED_STREAMS 0x00040000 /* mounted file system supports Named Streams VNOPs */ +#define MNTK_EXTENDED_ATTRS 0x00080000 /* mounted file system supports Extended Attributes VNOPs */ #define MNTK_LOCK_LOCAL 0x00100000 /* advisory locking is done above the VFS itself */ #define MNTK_VIRTUALDEV 0x00200000 /* mounted on a virtual device i.e. a disk image */ #define MNTK_ROOTDEV 0x00400000 /* this filesystem resides on the same device as the root */ @@ -161,7 +260,9 @@ extern struct mount * dead_mountp; /* * Generic file handle */ -#define NFS_MAX_FH_SIZE 64 +#define NFS_MAX_FH_SIZE NFSV4_MAX_FH_SIZE +#define NFSV4_MAX_FH_SIZE 128 +#define NFSV3_MAX_FH_SIZE 64 #define NFSV2_MAX_FH_SIZE 32 struct fhandle { int fh_len; /* length of file handle */ @@ -177,7 +278,6 @@ typedef struct fhandle fhandle_t; * mount time to identify the requested filesystem. */ struct vfstable { -/* THE FOLLOWING SHOULD KEEP THE SAME FOR user compat with sysctl */ struct vfsops *vfc_vfsops; /* filesystem operations vector */ char vfc_name[MFSNAMELEN]; /* filesystem type name */ int vfc_typenum; /* historic filesystem type number */ @@ -185,18 +285,27 @@ struct vfstable { int vfc_flags; /* permanent flags */ int (*vfc_mountroot)(mount_t, vnode_t, vfs_context_t); /* if != NULL, routine to mount root */ struct vfstable *vfc_next; /* next in list */ -/* Till the above we SHOULD KEEP THE SAME FOR user compat with sysctl */ - int vfc_threadsafe; /* FS is thread & premeption safe */ - lck_mtx_t vfc_lock; /* for non-threaded file systems */ + int32_t vfc_reserved1; + int32_t vfc_reserved2; int vfc_vfsflags; /* for optional types */ void * vfc_descptr; /* desc table allocated address */ int vfc_descsize; /* size allocated for desc table */ - int vfc_64bitready; /* The file system is ready for 64bit */ }; -#define VFC_VFSLOCALARGS 0x02 -#define VFC_VFSGENERICARGS 0x04 -#define VFC_VFSNATIVEXATTR 0x10 +/* vfc_vfsflags: */ +#define VFC_VFSLOCALARGS 0x002 +#define VFC_VFSGENERICARGS 0x004 +#define VFC_VFSNATIVEXATTR 0x010 +#define VFC_VFSDIRLINKS 0x020 +#define VFC_VFSPREFLIGHT 0x040 +#define VFC_VFSREADDIR_EXTENDED 0x080 +#define VFC_VFS64BITREADY 0x100 +#ifndef __LP64__ +#define VFC_VFSTHREADSAFE 0x200 +#endif /* __LP64__ */ +#define VFC_VFSNOMACLABEL 0x1000 +#define VFC_VFSVNOP_PAGEINV2 0x2000 +#define VFC_VFSVNOP_PAGEOUTV2 0x4000 extern int maxvfsconf; /* highest defined filesystem type */ @@ -223,71 +332,110 @@ struct vfsmount_args { /* - * LP64 version of statfs structure. + * LP64 *user* version of statfs structure. * NOTE - must be kept in sync with struct statfs in mount.h */ -struct user_statfs { +struct user64_statfs { short f_otype; /* TEMPORARY SHADOW COPY OF f_type */ short f_oflags; /* TEMPORARY SHADOW COPY OF f_flags */ - user_long_t f_bsize __attribute((aligned(8))); /* fundamental file system block size */ - user_long_t f_iosize; /* optimal transfer block size */ - user_long_t f_blocks; /* total data blocks in file system */ - user_long_t f_bfree; /* free blocks in fs */ - user_long_t f_bavail; /* free blocks avail to non-superuser */ - user_long_t f_files; /* total file nodes in file system */ - user_long_t f_ffree; /* free file nodes in fs */ + user64_long_t f_bsize; /* fundamental file system block size */ + user64_long_t f_iosize; /* optimal transfer block size */ + user64_long_t f_blocks; /* total data blocks in file system */ + user64_long_t f_bfree; /* free blocks in fs */ + user64_long_t f_bavail; /* free blocks avail to non-superuser */ + user64_long_t f_files; /* total file nodes in file system */ + user64_long_t f_ffree; /* free file nodes in fs */ fsid_t f_fsid; /* file system id */ uid_t f_owner; /* user that mounted the filesystem */ short f_reserved1; /* spare for later */ short f_type; /* type of filesystem */ - user_long_t f_flags; /* copy of mount exported flags */ - user_long_t f_reserved2[2]; /* reserved for future use */ + user64_long_t f_flags; /* copy of mount exported flags */ + user64_long_t f_reserved2[2]; /* reserved for future use */ char f_fstypename[MFSNAMELEN]; /* fs type name */ char f_mntonname[MNAMELEN]; /* directory on which mounted */ char f_mntfromname[MNAMELEN];/* mounted filesystem */ -#if COMPAT_GETFSSTAT - char f_reserved3[0]; /* For alignment */ - user_long_t f_reserved4[0]; /* For future use */ -#else char f_reserved3; /* For alignment */ - user_long_t f_reserved4[4] __attribute((aligned(8))); /* For future use */ -#endif + user64_long_t f_reserved4[4]; /* For future use */ }; +/* + * ILP32 *user* version of statfs structure. + * NOTE - must be kept in sync with struct statfs in mount.h + */ +struct user32_statfs { + short f_otype; /* TEMPORARY SHADOW COPY OF f_type */ + short f_oflags; /* TEMPORARY SHADOW COPY OF f_flags */ + user32_long_t f_bsize; /* fundamental file system block size */ + user32_long_t f_iosize; /* optimal transfer block size */ + user32_long_t f_blocks; /* total data blocks in file system */ + user32_long_t f_bfree; /* free blocks in fs */ + user32_long_t f_bavail; /* free blocks avail to non-superuser */ + user32_long_t f_files; /* total file nodes in file system */ + user32_long_t f_ffree; /* free file nodes in fs */ + fsid_t f_fsid; /* file system id */ + uid_t f_owner; /* user that mounted the filesystem */ + short f_reserved1; /* spare for later */ + short f_type; /* type of filesystem */ + user32_long_t f_flags; /* copy of mount exported flags */ + user32_long_t f_reserved2[2]; /* reserved for future use */ + char f_fstypename[MFSNAMELEN]; /* fs type name */ + char f_mntonname[MNAMELEN]; /* directory on which mounted */ + char f_mntfromname[MNAMELEN];/* mounted filesystem */ + char f_reserved3; /* For alignment */ + user32_long_t f_reserved4[4]; /* For future use */ +}; + +/* + * throttle I/Os are affected only by normal I/Os happening on the same bsd device node. For example, disk1s3 and + * disk1s5 are the same device node, while disk1s3 and disk2 are not (although disk2 might be a mounted disk image file + * and the disk image file resides on a partition in disk1). The following constant defines the maximum number of + * different bsd device nodes the algorithm can consider, and larger numbers are rounded by this maximum. Since + * throttled I/O is usually useful in non-server environment only, a small number 16 is enough in most cases + */ +#define LOWPRI_MAX_NUM_DEV 16 + __BEGIN_DECLS +extern uint32_t mount_generation; extern TAILQ_HEAD(mntlist, mount) mountlist; void mount_list_lock(void); void mount_list_unlock(void); void mount_lock_init(mount_t); void mount_lock_destroy(mount_t); void mount_lock(mount_t); +void mount_lock_spin(mount_t); void mount_unlock(mount_t); void mount_lock_renames(mount_t); void mount_unlock_renames(mount_t); void mount_ref(mount_t, int); void mount_drop(mount_t, int); +int mount_refdrain(mount_t); /* vfs_rootmountalloc should be kept as a private api */ errno_t vfs_rootmountalloc(const char *, const char *, mount_t *mpp); -errno_t vfs_init_io_attributes(vnode_t, mount_t); int vfs_mountroot(void); void vfs_unmountall(void); -int safedounmount(struct mount *, int, struct proc *); -int dounmount(struct mount *, int, int *, int, struct proc *); +int safedounmount(struct mount *, int, vfs_context_t); +int dounmount(struct mount *, int, int, vfs_context_t); /* xnuy internal api */ void mount_dropcrossref(mount_t, vnode_t, int); -int validfsnode(mount_t); mount_t mount_lookupby_volfsid(int, int); mount_t mount_list_lookupby_fsid(fsid_t *, int, int); +int mount_list_add(mount_t); +void mount_list_remove(mount_t); int mount_iterref(mount_t, int); int mount_isdrained(mount_t, int); void mount_iterdrop(mount_t); void mount_iterdrain(mount_t); void mount_iterreset(mount_t); +/* throttled I/O api */ +int throttle_get_io_policy(struct uthread **ut); +extern void throttle_lowpri_io(boolean_t ok_to_sleep); +int throttle_io_will_be_throttled(int lowpri_window_msecs, mount_t mp); + __END_DECLS #endif /* !_SYS_MOUNT_INTERNAL_H_ */