[mirror_zfs-debian.git] / include / sys / zfs_znode.h

/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
 */

#ifndef	_SYS_FS_ZFS_ZNODE_H
#define	_SYS_FS_ZFS_ZNODE_H

#ifdef _KERNEL
#include <sys/isa_defs.h>
#include <sys/types32.h>
#include <sys/attr.h>
#include <sys/list.h>
#include <sys/dmu.h>
#include <sys/sa.h>
#include <sys/zfs_vfsops.h>
#include <sys/rrwlock.h>
#include <sys/zfs_sa.h>
#include <sys/zfs_stat.h>
#endif
#include <sys/zfs_acl.h>
#include <sys/zil.h>

#ifdef	__cplusplus
extern "C" {
#endif

/*
 * Additional file level attributes, that are stored
 * in the upper half of zp_flags
 */
#define	ZFS_READONLY		0x0000000100000000ull
#define	ZFS_HIDDEN		0x0000000200000000ull
#define	ZFS_SYSTEM		0x0000000400000000ull
#define	ZFS_ARCHIVE		0x0000000800000000ull
#define	ZFS_IMMUTABLE		0x0000001000000000ull
#define	ZFS_NOUNLINK		0x0000002000000000ull
#define	ZFS_APPENDONLY		0x0000004000000000ull
#define	ZFS_NODUMP		0x0000008000000000ull
#define	ZFS_OPAQUE		0x0000010000000000ull
#define	ZFS_AV_QUARANTINED	0x0000020000000000ull
#define	ZFS_AV_MODIFIED		0x0000040000000000ull
#define	ZFS_REPARSE		0x0000080000000000ull
#define	ZFS_OFFLINE		0x0000100000000000ull
#define	ZFS_SPARSE		0x0000200000000000ull

#define	ZFS_ATTR_SET(zp, attr, value, pflags, tx) \
{ \
	if (value) \
		pflags |= attr; \
	else \
		pflags &= ~attr; \
	VERIFY(0 == sa_update(zp->z_sa_hdl, SA_ZPL_FLAGS(ZTOZSB(zp)), \
	    &pflags, sizeof (pflags), tx)); \
}

/*
 * Define special zfs pflags
 */
#define	ZFS_XATTR		0x1		/* is an extended attribute */
#define	ZFS_INHERIT_ACE		0x2		/* ace has inheritable ACEs */
#define	ZFS_ACL_TRIVIAL		0x4		/* files ACL is trivial */
#define	ZFS_ACL_OBJ_ACE		0x8		/* ACL has CMPLX Object ACE */
#define	ZFS_ACL_PROTECTED	0x10		/* ACL protected */
#define	ZFS_ACL_DEFAULTED	0x20		/* ACL should be defaulted */
#define	ZFS_ACL_AUTO_INHERIT	0x40		/* ACL should be inherited */
#define	ZFS_BONUS_SCANSTAMP	0x80		/* Scanstamp in bonus area */
#define	ZFS_NO_EXECS_DENIED	0x100		/* exec was given to everyone */

#define	SA_ZPL_ATIME(z)		z->z_attr_table[ZPL_ATIME]
#define	SA_ZPL_MTIME(z)		z->z_attr_table[ZPL_MTIME]
#define	SA_ZPL_CTIME(z)		z->z_attr_table[ZPL_CTIME]
#define	SA_ZPL_CRTIME(z)	z->z_attr_table[ZPL_CRTIME]
#define	SA_ZPL_GEN(z)		z->z_attr_table[ZPL_GEN]
#define	SA_ZPL_DACL_ACES(z)	z->z_attr_table[ZPL_DACL_ACES]
#define	SA_ZPL_XATTR(z)		z->z_attr_table[ZPL_XATTR]
#define	SA_ZPL_SYMLINK(z)	z->z_attr_table[ZPL_SYMLINK]
#define	SA_ZPL_RDEV(z)		z->z_attr_table[ZPL_RDEV]
#define	SA_ZPL_SCANSTAMP(z)	z->z_attr_table[ZPL_SCANSTAMP]
#define	SA_ZPL_UID(z)		z->z_attr_table[ZPL_UID]
#define	SA_ZPL_GID(z)		z->z_attr_table[ZPL_GID]
#define	SA_ZPL_PARENT(z)	z->z_attr_table[ZPL_PARENT]
#define	SA_ZPL_LINKS(z)		z->z_attr_table[ZPL_LINKS]
#define	SA_ZPL_MODE(z)		z->z_attr_table[ZPL_MODE]
#define	SA_ZPL_DACL_COUNT(z)	z->z_attr_table[ZPL_DACL_COUNT]
#define	SA_ZPL_FLAGS(z)		z->z_attr_table[ZPL_FLAGS]
#define	SA_ZPL_SIZE(z)		z->z_attr_table[ZPL_SIZE]
#define	SA_ZPL_ZNODE_ACL(z)	z->z_attr_table[ZPL_ZNODE_ACL]
#define	SA_ZPL_PAD(z)		z->z_attr_table[ZPL_PAD]

/*
 * Is ID ephemeral?
 */
#define	IS_EPHEMERAL(x)		(x > MAXUID)

/*
 * Should we use FUIDs?
 */
#define	USE_FUIDS(version, os)	(version >= ZPL_VERSION_FUID && \
    spa_version(dmu_objset_spa(os)) >= SPA_VERSION_FUID)
#define	USE_SA(version, os) (version >= ZPL_VERSION_SA && \
    spa_version(dmu_objset_spa(os)) >= SPA_VERSION_SA)

#define	MASTER_NODE_OBJ	1

/*
 * Special attributes for master node.
 * "userquota@" and "groupquota@" are also valid (from
 * zfs_userquota_prop_prefixes[]).
 */
#define	ZFS_FSID		"FSID"
#define	ZFS_UNLINKED_SET	"DELETE_QUEUE"
#define	ZFS_ROOT_OBJ		"ROOT"
#define	ZPL_VERSION_STR		"VERSION"
#define	ZFS_FUID_TABLES		"FUID"
#define	ZFS_SHARES_DIR		"SHARES"
#define	ZFS_SA_ATTRS		"SA_ATTRS"

#define	ZFS_MAX_BLOCKSIZE	(SPA_MAXBLOCKSIZE)

/* Path component length */
/*
 * The generic fs code uses MAXNAMELEN to represent
 * what the largest component length is.  Unfortunately,
 * this length includes the terminating NULL.  ZFS needs
 * to tell the users via pathconf() and statvfs() what the
 * true maximum length of a component is, excluding the NULL.
 */
#define	ZFS_MAXNAMELEN	(MAXNAMELEN - 1)

/*
 * Convert mode bits (zp_mode) to BSD-style DT_* values for storing in
 * the directory entries.  On Linux systems this value is already
 * defined correctly as part of the /usr/include/dirent.h header file.
 */
#ifndef IFTODT
#define	IFTODT(mode) (((mode) & S_IFMT) >> 12)
#endif

/*
 * The directory entry has the type (currently unused on Solaris) in the
 * top 4 bits, and the object number in the low 48 bits.  The "middle"
 * 12 bits are unused.
 */
#define	ZFS_DIRENT_TYPE(de) BF64_GET(de, 60, 4)
#define	ZFS_DIRENT_OBJ(de) BF64_GET(de, 0, 48)

/*
 * Directory entry locks control access to directory entries.
 * They are used to protect creates, deletes, and renames.
 * Each directory znode has a mutex and a list of locked names.
 */
#ifdef _KERNEL
typedef struct zfs_dirlock {
	char		*dl_name;	/* directory entry being locked */
	uint32_t	dl_sharecnt;	/* 0 if exclusive, > 0 if shared */
	uint8_t		dl_namelock;	/* 1 if z_name_lock is NOT held */
	uint16_t	dl_namesize;	/* set if dl_name was allocated */
	kcondvar_t	dl_cv;		/* wait for entry to be unlocked */
	struct znode	*dl_dzp;	/* directory znode */
	struct zfs_dirlock *dl_next;	/* next in z_dirlocks list */
} zfs_dirlock_t;

typedef struct znode {
	uint64_t	z_id;		/* object ID for this znode */
	kmutex_t	z_lock;		/* znode modification lock */
	krwlock_t	z_parent_lock;	/* parent lock for directories */
	krwlock_t	z_name_lock;	/* "master" lock for dirent locks */
	zfs_dirlock_t	*z_dirlocks;	/* directory entry lock list */
	kmutex_t	z_range_lock;	/* protects changes to z_range_avl */
	avl_tree_t	z_range_avl;	/* avl tree of file range locks */
	uint8_t		z_unlinked;	/* file has been unlinked */
	uint8_t		z_atime_dirty;	/* atime needs to be synced */
	uint8_t		z_zn_prefetch;	/* Prefetch znodes? */
	uint8_t		z_moved;	/* Has this znode been moved? */
	uint_t		z_blksz;	/* block size in bytes */
	uint_t		z_seq;		/* modification sequence number */
	uint64_t	z_mapcnt;	/* number of pages mapped to file */
	uint64_t	z_gen;		/* generation (cached) */
	uint64_t	z_size;		/* file size (cached) */
	uint64_t	z_atime[2];	/* atime (cached) */
	uint64_t	z_links;	/* file links (cached) */
	uint64_t	z_pflags;	/* pflags (cached) */
	uint64_t	z_uid;		/* uid fuid (cached) */
	uint64_t	z_gid;		/* gid fuid (cached) */
	mode_t		z_mode;		/* mode (cached) */
	uint32_t	z_sync_cnt;	/* synchronous open count */
	kmutex_t	z_acl_lock;	/* acl data lock */
	zfs_acl_t	*z_acl_cached;	/* cached acl */
	list_node_t	z_link_node;	/* all znodes in fs link */
	sa_handle_t	*z_sa_hdl;	/* handle to sa data */
	boolean_t	z_is_sa;	/* are we native sa? */
	boolean_t	z_is_zvol;	/* are we used by the zvol */
	boolean_t	z_is_mapped;	/* are we mmap'ed */
	struct inode	z_inode;	/* generic vfs inode */
} znode_t;


/*
 * Range locking rules
 * --------------------
 * 1. When truncating a file (zfs_create, zfs_setattr, zfs_space) the whole
 *    file range needs to be locked as RL_WRITER. Only then can the pages be
 *    freed etc and zp_size reset. zp_size must be set within range lock.
 * 2. For writes and punching holes (zfs_write & zfs_space) just the range
 *    being written or freed needs to be locked as RL_WRITER.
 *    Multiple writes at the end of the file must coordinate zp_size updates
 *    to ensure data isn't lost. A compare and swap loop is currently used
 *    to ensure the file size is at least the offset last written.
 * 3. For reads (zfs_read, zfs_get_data & zfs_putapage) just the range being
 *    read needs to be locked as RL_READER. A check against zp_size can then
 *    be made for reading beyond end of file.
 */

/*
 * Convert between znode pointers and inode pointers
 */
#define	ZTOI(znode)	(&((znode)->z_inode))
#define	ITOZ(inode)	(container_of((inode), znode_t, z_inode))
#define	ZTOZSB(znode)	((zfs_sb_t *)(ZTOI(znode)->i_sb->s_fs_info))
#define	ITOZSB(inode)	((zfs_sb_t *)((inode)->i_sb->s_fs_info))

#define	S_ISDEV(mode)	(S_ISCHR(mode) || S_ISBLK(mode) || S_ISFIFO(mode))

/*
 * ZFS_ENTER() is called on entry to each ZFS inode and vfs operation.
 * ZFS_EXIT() must be called before exitting the vop.
 * ZFS_VERIFY_ZP() verifies the znode is valid.
 */
#define	ZFS_ENTER(zsb) \
	{ \
		rrw_enter(&(zsb)->z_teardown_lock, RW_READER, FTAG); \
		if ((zsb)->z_unmounted) { \
			ZFS_EXIT(zsb); \
			return (EIO); \
		} \
	}

#define	ZFS_EXIT(zsb) \
	{ \
		rrw_exit(&(zsb)->z_teardown_lock, FTAG); \
		tsd_exit(); \
	}

#define	ZFS_VERIFY_ZP(zp) \
	if ((zp)->z_sa_hdl == NULL) { \
		ZFS_EXIT(ZTOZSB(zp)); \
		return (EIO); \
	}

/*
 * Macros for dealing with dmu_buf_hold
 */
#define	ZFS_OBJ_HASH(obj_num)	((obj_num) & (ZFS_OBJ_MTX_SZ - 1))
#define	ZFS_OBJ_MUTEX(zsb, obj_num)	\
	(&(zsb)->z_hold_mtx[ZFS_OBJ_HASH(obj_num)])
#define	ZFS_OBJ_HOLD_ENTER(zsb, obj_num) \
	mutex_enter(ZFS_OBJ_MUTEX((zsb), (obj_num)))
#define	ZFS_OBJ_HOLD_TRYENTER(zsb, obj_num) \
	mutex_tryenter(ZFS_OBJ_MUTEX((zsb), (obj_num)))
#define	ZFS_OBJ_HOLD_EXIT(zsb, obj_num) \
	mutex_exit(ZFS_OBJ_MUTEX((zsb), (obj_num)))
#define	ZFS_OBJ_HOLD_OWNED(zsb, obj_num) \
	mutex_owned(ZFS_OBJ_MUTEX((zsb), (obj_num)))

/*
 * Macros to encode/decode ZFS stored time values from/to struct timespec
 */
#define	ZFS_TIME_ENCODE(tp, stmp)		\
{						\
	(stmp)[0] = (uint64_t)(tp)->tv_sec;	\
	(stmp)[1] = (uint64_t)(tp)->tv_nsec;	\
}

#define	ZFS_TIME_DECODE(tp, stmp)		\
{						\
	(tp)->tv_sec = (time_t)(stmp)[0];		\
	(tp)->tv_nsec = (long)(stmp)[1];		\
}

/*
 * Timestamp defines
 */
#define	ACCESSED		(ATTR_ATIME)
#define	STATE_CHANGED		(ATTR_CTIME)
#define	CONTENT_MODIFIED	(ATTR_MTIME | ATTR_CTIME)

#define	ZFS_ACCESSTIME_STAMP(zsb, zp) \
	if ((zsb)->z_atime && !(zfs_is_readonly(zsb))) \
		zfs_tstamp_update_setup(zp, ACCESSED, NULL, NULL, B_FALSE);

extern int	zfs_init_fs(zfs_sb_t *, znode_t **);
extern void	zfs_set_dataprop(objset_t *);
extern void	zfs_create_fs(objset_t *os, cred_t *cr, nvlist_t *,
    dmu_tx_t *tx);
extern void	zfs_tstamp_update_setup(znode_t *, uint_t, uint64_t [2],
    uint64_t [2], boolean_t);
extern void	zfs_grow_blocksize(znode_t *, uint64_t, dmu_tx_t *);
extern int	zfs_freesp(znode_t *, uint64_t, uint64_t, int, boolean_t);
extern void	zfs_znode_init(void);
extern void	zfs_znode_fini(void);
extern int	zfs_zget(zfs_sb_t *, uint64_t, znode_t **);
extern int	zfs_rezget(znode_t *);
extern void	zfs_zinactive(znode_t *);
extern void	zfs_znode_delete(znode_t *, dmu_tx_t *);
extern void	zfs_remove_op_tables(void);
extern int	zfs_create_op_tables(void);
extern int	zfs_sync(struct super_block *, int, cred_t *);
extern dev_t	zfs_cmpldev(uint64_t);
extern int	zfs_get_zplprop(objset_t *os, zfs_prop_t prop, uint64_t *value);
extern int	zfs_get_stats(objset_t *os, nvlist_t *nv);
extern void	zfs_znode_dmu_fini(znode_t *);
extern int	zfs_inode_alloc(struct super_block *, struct inode **ip);
extern void	zfs_inode_destroy(struct inode *);
extern void	zfs_inode_update(znode_t *);

extern void zfs_log_create(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
    znode_t *dzp, znode_t *zp, char *name, vsecattr_t *, zfs_fuid_info_t *,
    vattr_t *vap);
extern int zfs_log_create_txtype(zil_create_t, vsecattr_t *vsecp,
    vattr_t *vap);
extern void zfs_log_remove(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
    znode_t *dzp, char *name, uint64_t foid);
#define	ZFS_NO_OBJECT	0	/* no object id */
extern void zfs_log_link(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
    znode_t *dzp, znode_t *zp, char *name);
extern void zfs_log_symlink(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
    znode_t *dzp, znode_t *zp, char *name, char *link);
extern void zfs_log_rename(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
    znode_t *sdzp, char *sname, znode_t *tdzp, char *dname, znode_t *szp);
extern void zfs_log_write(zilog_t *zilog, dmu_tx_t *tx, int txtype,
    znode_t *zp, offset_t off, ssize_t len, int ioflag);
extern void zfs_log_truncate(zilog_t *zilog, dmu_tx_t *tx, int txtype,
    znode_t *zp, uint64_t off, uint64_t len);
extern void zfs_log_setattr(zilog_t *zilog, dmu_tx_t *tx, int txtype,
    znode_t *zp, vattr_t *vap, uint_t mask_applied, zfs_fuid_info_t *fuidp);
extern void zfs_log_acl(zilog_t *zilog, dmu_tx_t *tx, znode_t *zp,
    vsecattr_t *vsecp, zfs_fuid_info_t *fuidp);
extern void zfs_xvattr_set(znode_t *zp, xvattr_t *xvap, dmu_tx_t *tx);
extern void zfs_upgrade(zfs_sb_t *zsb, dmu_tx_t *tx);
extern int zfs_create_share_dir(zfs_sb_t *zsb, dmu_tx_t *tx);

#if defined(HAVE_UIO_RW)
extern caddr_t zfs_map_page(page_t *, enum seg_rw);
extern void zfs_unmap_page(page_t *, caddr_t);
#endif /* HAVE_UIO_RW */

extern zil_get_data_t zfs_get_data;
extern zil_replay_func_t *zfs_replay_vector[TX_MAX_TYPE];
extern int zfsfstype;

#endif /* _KERNEL */

extern int zfs_obj_to_path(objset_t *osp, uint64_t obj, char *buf, int len);

#ifdef	__cplusplus
}
#endif

#endif	/* _SYS_FS_ZFS_ZNODE_H */
Commit	Line	Data
34dc7c2f BB	1	/*
	2	* CDDL HEADER START
	3	*
	4	* The contents of this file are subject to the terms of the
	5	* Common Development and Distribution License (the "License").
	6	* You may not use this file except in compliance with the License.
	7	*
	8	* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
	9	* or http://www.opensolaris.org/os/licensing.
	10	* See the License for the specific language governing permissions
	11	* and limitations under the License.
	12	*
	13	* When distributing Covered Code, include this CDDL HEADER in each
	14	* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
	15	* If applicable, add the following below this CDDL HEADER, with the
	16	* fields enclosed by brackets "[]" replaced with your own identifying
	17	* information: Portions Copyright [yyyy] [name of copyright owner]
	18	*
	19	* CDDL HEADER END
	20	*/
	21	/*
572e2857	22	* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
34dc7c2f BB	23	*/
	24
	25	#ifndef _SYS_FS_ZFS_ZNODE_H
	26	#define _SYS_FS_ZFS_ZNODE_H
	27
34dc7c2f BB	28	#ifdef _KERNEL
	29	#include <sys/isa_defs.h>
	30	#include <sys/types32.h>
	31	#include <sys/attr.h>
	32	#include <sys/list.h>
	33	#include <sys/dmu.h>
428870ff	34	#include <sys/sa.h>
34dc7c2f BB	35	#include <sys/zfs_vfsops.h>
34dc7c2f BB	36	#include <sys/rrwlock.h>
428870ff	37	#include <sys/zfs_sa.h>
572e2857	38	#include <sys/zfs_stat.h>
34dc7c2f BB	39	#endif
	40	#include <sys/zfs_acl.h>
	41	#include <sys/zil.h>
	42
	43	#ifdef __cplusplus
	44	extern "C" {
	45	#endif
	46
	47	/*
	48	* Additional file level attributes, that are stored
	49	* in the upper half of zp_flags
	50	*/
c5b3a7bb RC	51	#define ZFS_READONLY 0x0000000100000000ull
	52	#define ZFS_HIDDEN 0x0000000200000000ull
	53	#define ZFS_SYSTEM 0x0000000400000000ull
	54	#define ZFS_ARCHIVE 0x0000000800000000ull
	55	#define ZFS_IMMUTABLE 0x0000001000000000ull
	56	#define ZFS_NOUNLINK 0x0000002000000000ull
	57	#define ZFS_APPENDONLY 0x0000004000000000ull
	58	#define ZFS_NODUMP 0x0000008000000000ull
	59	#define ZFS_OPAQUE 0x0000010000000000ull
5484965a BB	60	#define ZFS_AV_QUARANTINED 0x0000020000000000ull
5484965a BB	61	#define ZFS_AV_MODIFIED 0x0000040000000000ull
c5b3a7bb RC	62	#define ZFS_REPARSE 0x0000080000000000ull
	63	#define ZFS_OFFLINE 0x0000100000000000ull
	64	#define ZFS_SPARSE 0x0000200000000000ull
34dc7c2f	65
428870ff	66	#define ZFS_ATTR_SET(zp, attr, value, pflags, tx) \
34dc7c2f BB	67	{ \
34dc7c2f BB	68	if (value) \
428870ff	69	pflags \|= attr; \
34dc7c2f	70	else \
428870ff	71	pflags &= ~attr; \
5484965a	72	VERIFY(0 == sa_update(zp->z_sa_hdl, SA_ZPL_FLAGS(ZTOZSB(zp)), \
428870ff	73	&pflags, sizeof (pflags), tx)); \
34dc7c2f BB	74	}
	75
	76	/*
	77	* Define special zfs pflags
	78	*/
	79	#define ZFS_XATTR 0x1 /* is an extended attribute */
	80	#define ZFS_INHERIT_ACE 0x2 /* ace has inheritable ACEs */
5484965a BB	81	#define ZFS_ACL_TRIVIAL 0x4 /* files ACL is trivial */
5484965a BB	82	#define ZFS_ACL_OBJ_ACE 0x8 /* ACL has CMPLX Object ACE */
34dc7c2f BB	83	#define ZFS_ACL_PROTECTED 0x10 /* ACL protected */
	84	#define ZFS_ACL_DEFAULTED 0x20 /* ACL should be defaulted */
	85	#define ZFS_ACL_AUTO_INHERIT 0x40 /* ACL should be inherited */
	86	#define ZFS_BONUS_SCANSTAMP 0x80 /* Scanstamp in bonus area */
45d1cae3	87	#define ZFS_NO_EXECS_DENIED 0x100 /* exec was given to everyone */
34dc7c2f	88
428870ff BB	89	#define SA_ZPL_ATIME(z) z->z_attr_table[ZPL_ATIME]
	90	#define SA_ZPL_MTIME(z) z->z_attr_table[ZPL_MTIME]
	91	#define SA_ZPL_CTIME(z) z->z_attr_table[ZPL_CTIME]
	92	#define SA_ZPL_CRTIME(z) z->z_attr_table[ZPL_CRTIME]
	93	#define SA_ZPL_GEN(z) z->z_attr_table[ZPL_GEN]
	94	#define SA_ZPL_DACL_ACES(z) z->z_attr_table[ZPL_DACL_ACES]
	95	#define SA_ZPL_XATTR(z) z->z_attr_table[ZPL_XATTR]
	96	#define SA_ZPL_SYMLINK(z) z->z_attr_table[ZPL_SYMLINK]
	97	#define SA_ZPL_RDEV(z) z->z_attr_table[ZPL_RDEV]
	98	#define SA_ZPL_SCANSTAMP(z) z->z_attr_table[ZPL_SCANSTAMP]
	99	#define SA_ZPL_UID(z) z->z_attr_table[ZPL_UID]
	100	#define SA_ZPL_GID(z) z->z_attr_table[ZPL_GID]
	101	#define SA_ZPL_PARENT(z) z->z_attr_table[ZPL_PARENT]
	102	#define SA_ZPL_LINKS(z) z->z_attr_table[ZPL_LINKS]
	103	#define SA_ZPL_MODE(z) z->z_attr_table[ZPL_MODE]
	104	#define SA_ZPL_DACL_COUNT(z) z->z_attr_table[ZPL_DACL_COUNT]
	105	#define SA_ZPL_FLAGS(z) z->z_attr_table[ZPL_FLAGS]
	106	#define SA_ZPL_SIZE(z) z->z_attr_table[ZPL_SIZE]
	107	#define SA_ZPL_ZNODE_ACL(z) z->z_attr_table[ZPL_ZNODE_ACL]
	108	#define SA_ZPL_PAD(z) z->z_attr_table[ZPL_PAD]
	109
34dc7c2f BB	110	/*
	111	* Is ID ephemeral?
	112	*/
	113	#define IS_EPHEMERAL(x) (x > MAXUID)
	114
	115	/*
	116	* Should we use FUIDs?
	117	*/
428870ff	118	#define USE_FUIDS(version, os) (version >= ZPL_VERSION_FUID && \
34dc7c2f	119	spa_version(dmu_objset_spa(os)) >= SPA_VERSION_FUID)
428870ff BB	120	#define USE_SA(version, os) (version >= ZPL_VERSION_SA && \
428870ff BB	121	spa_version(dmu_objset_spa(os)) >= SPA_VERSION_SA)
34dc7c2f BB	122
	123	#define MASTER_NODE_OBJ 1
	124
	125	/*
	126	* Special attributes for master node.
9babb374 BB	127	* "userquota@" and "groupquota@" are also valid (from
9babb374 BB	128	* zfs_userquota_prop_prefixes[]).
34dc7c2f BB	129	*/
	130	#define ZFS_FSID "FSID"
	131	#define ZFS_UNLINKED_SET "DELETE_QUEUE"
	132	#define ZFS_ROOT_OBJ "ROOT"
	133	#define ZPL_VERSION_STR "VERSION"
	134	#define ZFS_FUID_TABLES "FUID"
9babb374	135	#define ZFS_SHARES_DIR "SHARES"
428870ff	136	#define ZFS_SA_ATTRS "SA_ATTRS"
34dc7c2f BB	137
	138	#define ZFS_MAX_BLOCKSIZE (SPA_MAXBLOCKSIZE)
	139
	140	/* Path component length */
	141	/*
	142	* The generic fs code uses MAXNAMELEN to represent
	143	* what the largest component length is. Unfortunately,
	144	* this length includes the terminating NULL. ZFS needs
	145	* to tell the users via pathconf() and statvfs() what the
	146	* true maximum length of a component is, excluding the NULL.
	147	*/
	148	#define ZFS_MAXNAMELEN (MAXNAMELEN - 1)
	149
	150	/*
	151	* Convert mode bits (zp_mode) to BSD-style DT_* values for storing in
054bc00b BB	152	* the directory entries. On Linux systems this value is already
054bc00b BB	153	* defined correctly as part of the /usr/include/dirent.h header file.
34dc7c2f	154	*/
054bc00b	155	#ifndef IFTODT
34dc7c2f	156	#define IFTODT(mode) (((mode) & S_IFMT) >> 12)
054bc00b	157	#endif
34dc7c2f BB	158
	159	/*
	160	* The directory entry has the type (currently unused on Solaris) in the
	161	* top 4 bits, and the object number in the low 48 bits. The "middle"
	162	* 12 bits are unused.
	163	*/
	164	#define ZFS_DIRENT_TYPE(de) BF64_GET(de, 60, 4)
	165	#define ZFS_DIRENT_OBJ(de) BF64_GET(de, 0, 48)
	166
34dc7c2f BB	167	/*
	168	* Directory entry locks control access to directory entries.
	169	* They are used to protect creates, deletes, and renames.
	170	* Each directory znode has a mutex and a list of locked names.
	171	*/
	172	#ifdef _KERNEL
	173	typedef struct zfs_dirlock {
	174	char dl_name; / directory entry being locked */
	175	uint32_t dl_sharecnt; /* 0 if exclusive, > 0 if shared */
428870ff	176	uint8_t dl_namelock; /* 1 if z_name_lock is NOT held */
34dc7c2f BB	177	uint16_t dl_namesize; /* set if dl_name was allocated */
	178	kcondvar_t dl_cv; /* wait for entry to be unlocked */
	179	struct znode dl_dzp; / directory znode */
	180	struct zfs_dirlock dl_next; / next in z_dirlocks list */
	181	} zfs_dirlock_t;
	182
	183	typedef struct znode {
34dc7c2f BB	184	uint64_t z_id; /* object ID for this znode */
34dc7c2f BB	185	kmutex_t z_lock; /* znode modification lock */
34dc7c2f BB	186	krwlock_t z_parent_lock; /* parent lock for directories */
	187	krwlock_t z_name_lock; /* "master" lock for dirent locks */
	188	zfs_dirlock_t z_dirlocks; / directory entry lock list */
	189	kmutex_t z_range_lock; /* protects changes to z_range_avl */
	190	avl_tree_t z_range_avl; /* avl tree of file range locks */
	191	uint8_t z_unlinked; /* file has been unlinked */
	192	uint8_t z_atime_dirty; /* atime needs to be synced */
	193	uint8_t z_zn_prefetch; /* Prefetch znodes? */
572e2857	194	uint8_t z_moved; /* Has this znode been moved? */
34dc7c2f BB	195	uint_t z_blksz; /* block size in bytes */
	196	uint_t z_seq; /* modification sequence number */
	197	uint64_t z_mapcnt; /* number of pages mapped to file */
428870ff BB	198	uint64_t z_gen; /* generation (cached) */
	199	uint64_t z_size; /* file size (cached) */
	200	uint64_t z_atime[2]; /* atime (cached) */
	201	uint64_t z_links; /* file links (cached) */
	202	uint64_t z_pflags; /* pflags (cached) */
572e2857 BB	203	uint64_t z_uid; /* uid fuid (cached) */
572e2857 BB	204	uint64_t z_gid; /* gid fuid (cached) */
428870ff	205	mode_t z_mode; /* mode (cached) */
34dc7c2f BB	206	uint32_t z_sync_cnt; /* synchronous open count */
34dc7c2f BB	207	kmutex_t z_acl_lock; /* acl data lock */
45d1cae3	208	zfs_acl_t z_acl_cached; / cached acl */
34dc7c2f	209	list_node_t z_link_node; /* all znodes in fs link */
428870ff BB	210	sa_handle_t z_sa_hdl; / handle to sa data */
428870ff BB	211	boolean_t z_is_sa; /* are we native sa? */
3c4988c8	212	boolean_t z_is_zvol; /* are we used by the zvol */
c0d35759	213	boolean_t z_is_mapped; /* are we mmap'ed */
7304b6e5	214	struct inode z_inode; /* generic vfs inode */
34dc7c2f BB	215	} znode_t;
	216
	217
	218	/*
	219	* Range locking rules
	220	* --------------------
	221	* 1. When truncating a file (zfs_create, zfs_setattr, zfs_space) the whole
	222	* file range needs to be locked as RL_WRITER. Only then can the pages be
	223	* freed etc and zp_size reset. zp_size must be set within range lock.
	224	* 2. For writes and punching holes (zfs_write & zfs_space) just the range
	225	* being written or freed needs to be locked as RL_WRITER.
	226	* Multiple writes at the end of the file must coordinate zp_size updates
	227	* to ensure data isn't lost. A compare and swap loop is currently used
	228	* to ensure the file size is at least the offset last written.
	229	* 3. For reads (zfs_read, zfs_get_data & zfs_putapage) just the range being
	230	* read needs to be locked as RL_READER. A check against zp_size can then
	231	* be made for reading beyond end of file.
	232	*/
	233
	234	/*
7304b6e5	235	* Convert between znode pointers and inode pointers
34dc7c2f	236	*/
3558fd73 BB	237	#define ZTOI(znode) (&((znode)->z_inode))
3558fd73 BB	238	#define ITOZ(inode) (container_of((inode), znode_t, z_inode))
2cf7f52b BB	239	#define ZTOZSB(znode) ((zfs_sb_t *)(ZTOI(znode)->i_sb->s_fs_info))
2cf7f52b BB	240	#define ITOZSB(inode) ((zfs_sb_t *)((inode)->i_sb->s_fs_info))
34dc7c2f	241
3558fd73	242	#define S_ISDEV(mode) (S_ISCHR(mode) \|\| S_ISBLK(mode) \|\| S_ISFIFO(mode))
7304b6e5	243
34dc7c2f	244	/*
7304b6e5	245	* ZFS_ENTER() is called on entry to each ZFS inode and vfs operation.
34dc7c2f BB	246	* ZFS_EXIT() must be called before exitting the vop.
	247	* ZFS_VERIFY_ZP() verifies the znode is valid.
	248	*/
3558fd73	249	#define ZFS_ENTER(zsb) \
34dc7c2f	250	{ \
3558fd73 BB	251	rrw_enter(&(zsb)->z_teardown_lock, RW_READER, FTAG); \
	252	if ((zsb)->z_unmounted) { \
	253	ZFS_EXIT(zsb); \
34dc7c2f BB	254	return (EIO); \
	255	} \
	256	}
	257
3558fd73 BB	258	#define ZFS_EXIT(zsb) \
	259	{ \
	260	rrw_exit(&(zsb)->z_teardown_lock, FTAG); \
	261	tsd_exit(); \
	262	}
34dc7c2f BB	263
34dc7c2f BB	264	#define ZFS_VERIFY_ZP(zp) \
428870ff	265	if ((zp)->z_sa_hdl == NULL) { \
3558fd73	266	ZFS_EXIT(ZTOZSB(zp)); \
34dc7c2f	267	return (EIO); \
3558fd73	268	}
34dc7c2f BB	269
	270	/*
	271	* Macros for dealing with dmu_buf_hold
	272	*/
	273	#define ZFS_OBJ_HASH(obj_num) ((obj_num) & (ZFS_OBJ_MTX_SZ - 1))
3558fd73 BB	274	#define ZFS_OBJ_MUTEX(zsb, obj_num) \
	275	(&(zsb)->z_hold_mtx[ZFS_OBJ_HASH(obj_num)])
	276	#define ZFS_OBJ_HOLD_ENTER(zsb, obj_num) \
	277	mutex_enter(ZFS_OBJ_MUTEX((zsb), (obj_num)))
	278	#define ZFS_OBJ_HOLD_TRYENTER(zsb, obj_num) \
	279	mutex_tryenter(ZFS_OBJ_MUTEX((zsb), (obj_num)))
	280	#define ZFS_OBJ_HOLD_EXIT(zsb, obj_num) \
	281	mutex_exit(ZFS_OBJ_MUTEX((zsb), (obj_num)))
d6bd8eaa BB	282	#define ZFS_OBJ_HOLD_OWNED(zsb, obj_num) \
d6bd8eaa BB	283	mutex_owned(ZFS_OBJ_MUTEX((zsb), (obj_num)))
34dc7c2f BB	284
	285	/*
	286	* Macros to encode/decode ZFS stored time values from/to struct timespec
	287	*/
	288	#define ZFS_TIME_ENCODE(tp, stmp) \
	289	{ \
b128c09f	290	(stmp)[0] = (uint64_t)(tp)->tv_sec; \
34dc7c2f BB	291	(stmp)[1] = (uint64_t)(tp)->tv_nsec; \
	292	}
	293
	294	#define ZFS_TIME_DECODE(tp, stmp) \
	295	{ \
	296	(tp)->tv_sec = (time_t)(stmp)[0]; \
	297	(tp)->tv_nsec = (long)(stmp)[1]; \
	298	}
	299
	300	/*
	301	* Timestamp defines
	302	*/
3558fd73 BB	303	#define ACCESSED (ATTR_ATIME)
	304	#define STATE_CHANGED (ATTR_CTIME)
	305	#define CONTENT_MODIFIED (ATTR_MTIME \| ATTR_CTIME)
34dc7c2f	306
3558fd73	307	#define ZFS_ACCESSTIME_STAMP(zsb, zp) \
2cf7f52b	308	if ((zsb)->z_atime && !(zfs_is_readonly(zsb))) \
428870ff	309	zfs_tstamp_update_setup(zp, ACCESSED, NULL, NULL, B_FALSE);
34dc7c2f	310
3558fd73	311	extern int zfs_init_fs(zfs_sb_t , znode_t *);
34dc7c2f BB	312	extern void zfs_set_dataprop(objset_t *);
	313	extern void zfs_create_fs(objset_t os, cred_t cr, nvlist_t *,
	314	dmu_tx_t *tx);
428870ff BB	315	extern void zfs_tstamp_update_setup(znode_t *, uint_t, uint64_t [2],
428870ff BB	316	uint64_t [2], boolean_t);
34dc7c2f BB	317	extern void zfs_grow_blocksize(znode_t , uint64_t, dmu_tx_t );
	318	extern int zfs_freesp(znode_t *, uint64_t, uint64_t, int, boolean_t);
	319	extern void zfs_znode_init(void);
	320	extern void zfs_znode_fini(void);
3558fd73	321	extern int zfs_zget(zfs_sb_t , uint64_t, znode_t *);
34dc7c2f BB	322	extern int zfs_rezget(znode_t *);
	323	extern void zfs_zinactive(znode_t *);
	324	extern void zfs_znode_delete(znode_t , dmu_tx_t );
0bc8fd78 BB	325	extern void zfs_remove_op_tables(void);
0bc8fd78 BB	326	extern int zfs_create_op_tables(void);
03f9ba9d	327	extern int zfs_sync(struct super_block , int, cred_t );
34dc7c2f BB	328	extern dev_t zfs_cmpldev(uint64_t);
34dc7c2f BB	329	extern int zfs_get_zplprop(objset_t os, zfs_prop_t prop, uint64_t value);
34dc7c2f BB	330	extern int zfs_get_stats(objset_t os, nvlist_t nv);
34dc7c2f BB	331	extern void zfs_znode_dmu_fini(znode_t *);
3558fd73 BB	332	extern int zfs_inode_alloc(struct super_block , struct inode *ip);
3558fd73 BB	333	extern void zfs_inode_destroy(struct inode *);
960e08fe	334	extern void zfs_inode_update(znode_t *);
34dc7c2f BB	335
	336	extern void zfs_log_create(zilog_t zilog, dmu_tx_t tx, uint64_t txtype,
	337	znode_t dzp, znode_t zp, char name, vsecattr_t , zfs_fuid_info_t *,
	338	vattr_t *vap);
	339	extern int zfs_log_create_txtype(zil_create_t, vsecattr_t *vsecp,
	340	vattr_t *vap);
	341	extern void zfs_log_remove(zilog_t zilog, dmu_tx_t tx, uint64_t txtype,
572e2857 BB	342	znode_t dzp, char name, uint64_t foid);
572e2857 BB	343	#define ZFS_NO_OBJECT 0 /* no object id */
34dc7c2f BB	344	extern void zfs_log_link(zilog_t zilog, dmu_tx_t tx, uint64_t txtype,
	345	znode_t dzp, znode_t zp, char *name);
	346	extern void zfs_log_symlink(zilog_t zilog, dmu_tx_t tx, uint64_t txtype,
	347	znode_t dzp, znode_t zp, char name, char link);
	348	extern void zfs_log_rename(zilog_t zilog, dmu_tx_t tx, uint64_t txtype,
	349	znode_t sdzp, char sname, znode_t tdzp, char dname, znode_t *szp);
	350	extern void zfs_log_write(zilog_t zilog, dmu_tx_t tx, int txtype,
	351	znode_t *zp, offset_t off, ssize_t len, int ioflag);
	352	extern void zfs_log_truncate(zilog_t zilog, dmu_tx_t tx, int txtype,
	353	znode_t *zp, uint64_t off, uint64_t len);
	354	extern void zfs_log_setattr(zilog_t zilog, dmu_tx_t tx, int txtype,
5484965a	355	znode_t zp, vattr_t vap, uint_t mask_applied, zfs_fuid_info_t *fuidp);
34dc7c2f BB	356	extern void zfs_log_acl(zilog_t zilog, dmu_tx_t tx, znode_t *zp,
34dc7c2f BB	357	vsecattr_t vsecp, zfs_fuid_info_t fuidp);
428870ff	358	extern void zfs_xvattr_set(znode_t zp, xvattr_t xvap, dmu_tx_t *tx);
3558fd73 BB	359	extern void zfs_upgrade(zfs_sb_t zsb, dmu_tx_t tx);
3558fd73 BB	360	extern int zfs_create_share_dir(zfs_sb_t zsb, dmu_tx_t tx);
34dc7c2f	361
60101509	362	#if defined(HAVE_UIO_RW)
b128c09f BB	363	extern caddr_t zfs_map_page(page_t *, enum seg_rw);
b128c09f BB	364	extern void zfs_unmap_page(page_t *, caddr_t);
60101509	365	#endif /* HAVE_UIO_RW */
b128c09f	366
34dc7c2f BB	367	extern zil_get_data_t zfs_get_data;
	368	extern zil_replay_func_t *zfs_replay_vector[TX_MAX_TYPE];
	369	extern int zfsfstype;
	370
	371	#endif /* _KERNEL */
	372
	373	extern int zfs_obj_to_path(objset_t osp, uint64_t obj, char buf, int len);
	374
	375	#ifdef __cplusplus
	376	}
	377	#endif
	378
	379	#endif /* _SYS_FS_ZFS_ZNODE_H */