xfs_mount.h 14.7 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
2 3
 * Copyright (c) 2000-2005 Silicon Graphics, Inc.
 * All Rights Reserved.
L
Linus Torvalds 已提交
4
 *
5 6
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
L
Linus Torvalds 已提交
7 8
 * published by the Free Software Foundation.
 *
9 10 11 12
 * This program is distributed in the hope that it would be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
L
Linus Torvalds 已提交
13
 *
14 15 16
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write the Free Software Foundation,
 * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
L
Linus Torvalds 已提交
17 18 19 20
 */
#ifndef __XFS_MOUNT_H__
#define	__XFS_MOUNT_H__

21
struct xlog;
L
Linus Torvalds 已提交
22
struct xfs_inode;
23
struct xfs_mru_cache;
24
struct xfs_nameops;
D
David Chinner 已提交
25
struct xfs_ail;
C
Christoph Hellwig 已提交
26
struct xfs_quotainfo;
27
struct xfs_dir_ops;
28
struct xfs_da_geometry;
C
Christoph Hellwig 已提交
29

30 31 32 33 34 35 36 37 38 39
/* dynamic preallocation free space thresholds, 5% down to 1% */
enum {
	XFS_LOWSP_1_PCNT = 0,
	XFS_LOWSP_2_PCNT,
	XFS_LOWSP_3_PCNT,
	XFS_LOWSP_4_PCNT,
	XFS_LOWSP_5_PCNT,
	XFS_LOWSP_MAX,
};

40 41 42 43 44 45 46
/*
 * Error Configuration
 *
 * Error classes define the subsystem the configuration belongs to.
 * Error numbers define the errors that are configurable.
 */
enum {
47
	XFS_ERR_METADATA,
48 49 50
	XFS_ERR_CLASS_MAX,
};
enum {
51
	XFS_ERR_DEFAULT,
52 53 54 55 56 57 58 59
	XFS_ERR_ERRNO_MAX,
};

struct xfs_error_cfg {
	struct xfs_kobj	kobj;
	int		max_retries;
};

L
Linus Torvalds 已提交
60
typedef struct xfs_mount {
C
Christoph Hellwig 已提交
61
	struct super_block	*m_super;
L
Linus Torvalds 已提交
62
	xfs_tid_t		m_tid;		/* next unused tid for fs */
D
David Chinner 已提交
63
	struct xfs_ail		*m_ail;		/* fs active log item list */
64 65

	struct xfs_sb		m_sb;		/* copy of fs superblock */
E
Eric Sandeen 已提交
66
	spinlock_t		m_sb_lock;	/* sb counter lock */
67
	struct percpu_counter	m_icount;	/* allocated inodes counter */
68
	struct percpu_counter	m_ifree;	/* free inodes counter */
69
	struct percpu_counter	m_fdblocks;	/* free block counter */
70

L
Linus Torvalds 已提交
71 72 73
	struct xfs_buf		*m_sb_bp;	/* buffer for superblock */
	char			*m_fsname;	/* filesystem name */
	int			m_fsname_len;	/* strlen of fs name */
74 75
	char			*m_rtname;	/* realtime device name */
	char			*m_logname;	/* external log device name */
L
Linus Torvalds 已提交
76 77 78
	int			m_bsize;	/* fs logical block size */
	xfs_agnumber_t		m_agfrotor;	/* last ag where space found */
	xfs_agnumber_t		m_agirotor;	/* last ag dir inode alloced */
E
Eric Sandeen 已提交
79
	spinlock_t		m_agirotor_lock;/* .. and lock protecting it */
L
Linus Torvalds 已提交
80 81 82 83 84
	xfs_agnumber_t		m_maxagi;	/* highest inode alloc group */
	uint			m_readio_log;	/* min read size log bytes */
	uint			m_readio_blocks; /* min read size blocks */
	uint			m_writeio_log;	/* min write size log bytes */
	uint			m_writeio_blocks; /* min write size blocks */
85 86
	struct xfs_da_geometry	*m_dir_geo;	/* directory block geometry */
	struct xfs_da_geometry	*m_attr_geo;	/* attribute block geometry */
87
	struct xlog		*m_log;		/* log specific stuff */
L
Linus Torvalds 已提交
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102
	int			m_logbufs;	/* number of log buffers */
	int			m_logbsize;	/* size of each log buffer */
	uint			m_rsumlevels;	/* rt summary levels */
	uint			m_rsumsize;	/* size of rt summary, bytes */
	struct xfs_inode	*m_rbmip;	/* pointer to bitmap inode */
	struct xfs_inode	*m_rsumip;	/* pointer to summary inode */
	struct xfs_inode	*m_rootip;	/* pointer to root directory */
	struct xfs_quotainfo	*m_quotainfo;	/* disk quota information */
	xfs_buftarg_t		*m_ddev_targp;	/* saves taking the address */
	xfs_buftarg_t		*m_logdev_targp;/* ptr to log device */
	xfs_buftarg_t		*m_rtdev_targp;	/* ptr to rt device */
	__uint8_t		m_blkbit_log;	/* blocklog + NBBY */
	__uint8_t		m_blkbb_log;	/* blocklog - BBSHIFT */
	__uint8_t		m_agno_log;	/* log #ag's */
	__uint8_t		m_agino_log;	/* #bits for agino in inum */
103
	uint			m_inode_cluster_size;/* min inode buf size */
L
Linus Torvalds 已提交
104 105 106
	uint			m_blockmask;	/* sb_blocksize-1 */
	uint			m_blockwsize;	/* sb_blocksize in words */
	uint			m_blockwmask;	/* blockwsize-1 */
107 108 109 110 111 112
	uint			m_alloc_mxr[2];	/* max alloc btree records */
	uint			m_alloc_mnr[2];	/* min alloc btree records */
	uint			m_bmap_dmxr[2];	/* max bmap btree records */
	uint			m_bmap_dmnr[2];	/* min bmap btree records */
	uint			m_inobt_mxr[2];	/* max inobt btree records */
	uint			m_inobt_mnr[2];	/* min inobt btree records */
L
Linus Torvalds 已提交
113 114
	uint			m_ag_maxlevels;	/* XFS_AG_MAXLEVELS */
	uint			m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
115
	uint			m_in_maxlevels;	/* max inobt btree levels. */
116 117
	struct radix_tree_root	m_perag_tree;	/* per-ag accounting info */
	spinlock_t		m_perag_lock;	/* lock for m_perag_tree */
118
	struct mutex		m_growlock;	/* growfs mutex */
L
Linus Torvalds 已提交
119 120
	int			m_fixedfsid[2];	/* unchanged for life of FS */
	uint			m_dmevmask;	/* DMI events for this FS */
121
	__uint64_t		m_flags;	/* global mount flags */
L
Linus Torvalds 已提交
122 123
	int			m_ialloc_inos;	/* inodes in inode allocation */
	int			m_ialloc_blks;	/* blocks in inode allocation */
124 125
	int			m_ialloc_min_blks;/* min blocks in sparse inode
						   * allocation */
L
Linus Torvalds 已提交
126 127
	int			m_inoalign_mask;/* mask sb_inoalignmt if used */
	uint			m_qflags;	/* quota status flags */
128
	struct xfs_trans_resv	m_resv;		/* precomputed res values */
L
Linus Torvalds 已提交
129 130 131
	__uint64_t		m_maxicount;	/* maximum inode count */
	__uint64_t		m_resblks;	/* total reserved blocks */
	__uint64_t		m_resblks_avail;/* available reserved blocks */
132
	__uint64_t		m_resblks_save;	/* reserved blks @ remount,ro */
L
Linus Torvalds 已提交
133 134
	int			m_dalign;	/* stripe unit */
	int			m_swidth;	/* stripe width */
135
	int			m_sinoalign;	/* stripe unit inode alignment */
L
Linus Torvalds 已提交
136
	__uint8_t		m_sectbb_log;	/* sectlog - BBSHIFT */
137
	const struct xfs_nameops *m_dirnameops;	/* vector of dir name ops */
138
	const struct xfs_dir_ops *m_dir_inode_ops; /* vector of dir inode ops */
D
Dave Chinner 已提交
139
	const struct xfs_dir_ops *m_nondir_inode_ops; /* !dir inode ops */
L
Linus Torvalds 已提交
140 141
	uint			m_chsize;	/* size of next field */
	atomic_t		m_active_trans;	/* number trans frozen */
142
	struct xfs_mru_cache	*m_filestream;  /* per-mount filestream data */
143
	struct delayed_work	m_reclaim_work;	/* background inode reclaim */
144 145
	struct delayed_work	m_eofblocks_work; /* background eof blocks
						     trimming */
146
	bool			m_update_sb;	/* sb needs update in mount */
147 148
	int64_t			m_low_space[XFS_LOWSP_MAX];
						/* low free space thresholds */
B
Brian Foster 已提交
149
	struct xfs_kobj		m_kobj;
150
	struct xfs_kobj		m_error_kobj;
151
	struct xfs_kobj		m_error_meta_kobj;
152
	struct xfs_error_cfg	m_error_cfg[XFS_ERR_CLASS_MAX][XFS_ERR_ERRNO_MAX];
153
	struct xstats		m_stats;	/* per-fs stats */
154

155
	struct workqueue_struct *m_buf_workqueue;
156 157
	struct workqueue_struct	*m_data_workqueue;
	struct workqueue_struct	*m_unwritten_workqueue;
158
	struct workqueue_struct	*m_cil_workqueue;
D
Dave Chinner 已提交
159 160
	struct workqueue_struct	*m_reclaim_workqueue;
	struct workqueue_struct	*m_log_workqueue;
161
	struct workqueue_struct *m_eofblocks_workqueue;
162 163 164 165 166 167 168 169 170 171 172

	/*
	 * Generation of the filesysyem layout.  This is incremented by each
	 * growfs, and used by the pNFS server to ensure the client updates
	 * its view of the block device once it gets a layout that might
	 * reference the newly added blocks.  Does not need to be persistent
	 * as long as we only allow file system size increments, but if we
	 * ever support shrinks it would have to be persisted in addition
	 * to various other kinds of pain inflicted on the pNFS server.
	 */
	__uint32_t		m_generation;
173 174 175 176 177 178 179 180 181 182 183

#ifdef DEBUG
	/*
	 * DEBUG mode instrumentation to test and/or trigger delayed allocation
	 * block killing in the event of failed writes. When enabled, all
	 * buffered writes are forced to fail. All delalloc blocks in the range
	 * of the write (including pre-existing delalloc blocks!) are tossed as
	 * part of the write failure error handling sequence.
	 */
	bool			m_fail_writes;
#endif
L
Linus Torvalds 已提交
184 185 186 187 188
} xfs_mount_t;

/*
 * Flags for m_flags.
 */
D
David Chinner 已提交
189
#define XFS_MOUNT_WSYNC		(1ULL << 0)	/* for nfs - all metadata ops
L
Linus Torvalds 已提交
190 191
						   must be synchronous except
						   for space allocations */
D
David Chinner 已提交
192
#define XFS_MOUNT_WAS_CLEAN	(1ULL << 3)
193
#define XFS_MOUNT_FS_SHUTDOWN	(1ULL << 4)	/* atomic stop of all filesystem
L
Linus Torvalds 已提交
194 195
						   operations, typically for
						   disk errors in metadata */
196
#define XFS_MOUNT_DISCARD	(1ULL << 5)	/* discard unused blocks */
197
#define XFS_MOUNT_NOALIGN	(1ULL << 7)	/* turn off stripe alignment
L
Linus Torvalds 已提交
198
						   allocations */
199
#define XFS_MOUNT_ATTR2		(1ULL << 8)	/* allow use of attr2 format */
200
#define XFS_MOUNT_GRPID		(1ULL << 9)	/* group-ID assigned from directory */
201 202
#define XFS_MOUNT_NORECOVERY	(1ULL << 10)	/* no recovery - dirty fs */
#define XFS_MOUNT_DFLT_IOSIZE	(1ULL << 12)	/* set default i/o size */
203 204
#define XFS_MOUNT_SMALL_INUMS	(1ULL << 14)	/* user wants 32bit inodes */
#define XFS_MOUNT_32BITINODES	(1ULL << 15)	/* inode32 allocator active */
205 206
#define XFS_MOUNT_NOUUID	(1ULL << 16)	/* ignore uuid during mount */
#define XFS_MOUNT_BARRIER	(1ULL << 17)
207
#define XFS_MOUNT_IKEEP		(1ULL << 18)	/* keep empty inode clusters*/
208
#define XFS_MOUNT_SWALLOC	(1ULL << 19)	/* turn on stripe width
L
Linus Torvalds 已提交
209
						 * allocation */
210
#define XFS_MOUNT_RDONLY	(1ULL << 20)	/* read-only fs */
211 212
#define XFS_MOUNT_DIRSYNC	(1ULL << 21)	/* synchronous directory ops */
#define XFS_MOUNT_COMPAT_IOSIZE	(1ULL << 22)	/* don't report large preferred
213
						 * I/O size in stat() */
214 215
#define XFS_MOUNT_FILESTREAMS	(1ULL << 24)	/* enable the filestreams
						   allocator */
216
#define XFS_MOUNT_NOATTR2	(1ULL << 25)	/* disable use of attr2 format */
217

D
Dave Chinner 已提交
218 219
#define XFS_MOUNT_DAX		(1ULL << 62)	/* TEST ONLY! */

L
Linus Torvalds 已提交
220 221 222 223 224 225 226 227

/*
 * Default minimum read and write sizes.
 */
#define XFS_READIO_LOG_LARGE	16
#define XFS_WRITEIO_LOG_LARGE	16

/*
228 229
 * Max and min values for mount-option defined I/O
 * preallocation sizes.
L
Linus Torvalds 已提交
230
 */
231
#define XFS_MAX_IO_LOG		30	/* 1G */
L
Linus Torvalds 已提交
232 233 234 235 236 237
#define XFS_MIN_IO_LOG		PAGE_SHIFT

/*
 * Synchronous read and write sizes.  This should be
 * better for NFSv2 wsync filesystems.
 */
M
Malcolm Parsons 已提交
238 239
#define	XFS_WSYNC_READIO_LOG	15	/* 32k */
#define	XFS_WSYNC_WRITEIO_LOG	14	/* 16k */
L
Linus Torvalds 已提交
240

241 242
/*
 * Allow large block sizes to be reported to userspace programs if the
243
 * "largeio" mount option is used.
244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264
 *
 * If compatibility mode is specified, simply return the basic unit of caching
 * so that we don't get inefficient read/modify/write I/O from user apps.
 * Otherwise....
 *
 * If the underlying volume is a stripe, then return the stripe width in bytes
 * as the recommended I/O size. It is not a stripe and we've set a default
 * buffered I/O size, return that, otherwise return the compat default.
 */
static inline unsigned long
xfs_preferred_iosize(xfs_mount_t *mp)
{
	if (mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE)
		return PAGE_CACHE_SIZE;
	return (mp->m_swidth ?
		(mp->m_swidth << mp->m_sb.sb_blocklog) :
		((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ?
			(1 << (int)MAX(mp->m_readio_log, mp->m_writeio_log)) :
			PAGE_CACHE_SIZE));
}

D
David Chinner 已提交
265 266
#define XFS_LAST_UNMOUNT_WAS_CLEAN(mp)	\
				((mp)->m_flags & XFS_MOUNT_WAS_CLEAN)
L
Linus Torvalds 已提交
267
#define XFS_FORCED_SHUTDOWN(mp)	((mp)->m_flags & XFS_MOUNT_FS_SHUTDOWN)
268 269
void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname,
		int lnnum);
L
Linus Torvalds 已提交
270
#define xfs_force_shutdown(m,f)	\
271
	xfs_do_force_shutdown(m, f, __FILE__, __LINE__)
L
Linus Torvalds 已提交
272

C
Christoph Hellwig 已提交
273 274 275 276 277 278 279
#define SHUTDOWN_META_IO_ERROR	0x0001	/* write attempt to metadata failed */
#define SHUTDOWN_LOG_IO_ERROR	0x0002	/* write attempt to the log failed */
#define SHUTDOWN_FORCE_UMOUNT	0x0004	/* shutdown from a forced unmount */
#define SHUTDOWN_CORRUPT_INCORE	0x0008	/* corrupt in-memory data structures */
#define SHUTDOWN_REMOTE_REQ	0x0010	/* shutdown came from remote cell */
#define SHUTDOWN_DEVICE_REQ	0x0020	/* failed all paths to the device */

L
Linus Torvalds 已提交
280 281 282
/*
 * Flags for xfs_mountfs
 */
283
#define XFS_MFSI_QUIET		0x40	/* Be silent if mount errors found */
L
Linus Torvalds 已提交
284

285 286
static inline xfs_agnumber_t
xfs_daddr_to_agno(struct xfs_mount *mp, xfs_daddr_t d)
L
Linus Torvalds 已提交
287
{
288 289 290
	xfs_daddr_t ld = XFS_BB_TO_FSBT(mp, d);
	do_div(ld, mp->m_sb.sb_agblocks);
	return (xfs_agnumber_t) ld;
L
Linus Torvalds 已提交
291 292
}

293 294
static inline xfs_agblock_t
xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d)
L
Linus Torvalds 已提交
295
{
296 297
	xfs_daddr_t ld = XFS_BB_TO_FSBT(mp, d);
	return (xfs_agblock_t) do_div(ld, mp->m_sb.sb_agblocks);
L
Linus Torvalds 已提交
298 299
}

300 301 302 303 304 305 306 307 308 309 310 311 312 313
#ifdef DEBUG
static inline bool
xfs_mp_fail_writes(struct xfs_mount *mp)
{
	return mp->m_fail_writes;
}
#else
static inline bool
xfs_mp_fail_writes(struct xfs_mount *mp)
{
	return 0;
}
#endif

314 315
/*
 * Per-ag incore structure, copies of information in agf and agi, to improve the
316
 * performance of allocation group selection.
317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
 */
typedef struct xfs_perag {
	struct xfs_mount *pag_mount;	/* owner filesystem */
	xfs_agnumber_t	pag_agno;	/* AG this structure belongs to */
	atomic_t	pag_ref;	/* perag reference count */
	char		pagf_init;	/* this agf's entry is initialized */
	char		pagi_init;	/* this agi's entry is initialized */
	char		pagf_metadata;	/* the agf is preferred to be metadata */
	char		pagi_inodeok;	/* The agi is ok for inodes */
	__uint8_t	pagf_levels[XFS_BTNUM_AGF];
					/* # of levels in bno & cnt btree */
	__uint32_t	pagf_flcount;	/* count of blocks in freelist */
	xfs_extlen_t	pagf_freeblks;	/* total free blocks */
	xfs_extlen_t	pagf_longest;	/* longest free space */
	__uint32_t	pagf_btreeblks;	/* # of blocks held in AGF btrees */
	xfs_agino_t	pagi_freecount;	/* number of free inodes */
	xfs_agino_t	pagi_count;	/* number of allocated inodes */

	/*
	 * Inode allocation search lookup optimisation.
	 * If the pagino matches, the search for new inodes
	 * doesn't need to search the near ones again straight away
	 */
	xfs_agino_t	pagl_pagino;
	xfs_agino_t	pagl_leftrec;
	xfs_agino_t	pagl_rightrec;
	spinlock_t	pagb_lock;	/* lock for pagb_tree */
	struct rb_root	pagb_tree;	/* ordered tree of busy extents */

	atomic_t        pagf_fstrms;    /* # of filestreams active in this AG */

	spinlock_t	pag_ici_lock;	/* incore inode cache lock */
	struct radix_tree_root pag_ici_root;	/* incore inode cache root */
	int		pag_ici_reclaimable;	/* reclaimable inodes */
	struct mutex	pag_ici_reclaim_lock;	/* serialisation point */
	unsigned long	pag_ici_reclaim_cursor;	/* reclaim restart point */

	/* buffer cache index */
	spinlock_t	pag_buf_lock;	/* lock for pag_buf_tree */
	struct rb_root	pag_buf_tree;	/* ordered tree of active buffers */

	/* for rcu-safe freeing */
	struct rcu_head	rcu_head;
	int		pagb_count;	/* pagb slots in use */
} xfs_perag_t;

363
extern void	xfs_uuid_table_free(void);
364
extern int	xfs_log_sbcount(xfs_mount_t *);
E
Eric Sandeen 已提交
365
extern __uint64_t xfs_default_resblks(xfs_mount_t *mp);
C
Christoph Hellwig 已提交
366
extern int	xfs_mountfs(xfs_mount_t *mp);
367 368
extern int	xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount,
				     xfs_agnumber_t *maxagi);
369
extern void	xfs_unmountfs(xfs_mount_t *);
D
Dave Chinner 已提交
370

371
extern int	xfs_mod_icount(struct xfs_mount *mp, int64_t delta);
372
extern int	xfs_mod_ifree(struct xfs_mount *mp, int64_t delta);
373 374
extern int	xfs_mod_fdblocks(struct xfs_mount *mp, int64_t delta,
				 bool reserved);
D
Dave Chinner 已提交
375 376
extern int	xfs_mod_frextents(struct xfs_mount *mp, int64_t delta);

L
Linus Torvalds 已提交
377
extern struct xfs_buf *xfs_getsb(xfs_mount_t *, int);
378
extern int	xfs_readsb(xfs_mount_t *, int);
L
Linus Torvalds 已提交
379
extern void	xfs_freesb(xfs_mount_t *);
380
extern bool	xfs_fs_writable(struct xfs_mount *mp, int level);
381
extern int	xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t);
L
Linus Torvalds 已提交
382

C
Christoph Hellwig 已提交
383 384
extern int	xfs_dev_is_read_only(struct xfs_mount *, char *);

385 386
extern void	xfs_set_low_space_thresholds(struct xfs_mount *);

387 388 389
int	xfs_zero_extent(struct xfs_inode *ip, xfs_fsblock_t start_fsb,
			xfs_off_t count_fsb);

390 391 392
struct xfs_error_cfg * xfs_error_get_cfg(struct xfs_mount *mp,
		int error_class, int error);

L
Linus Torvalds 已提交
393
#endif	/* __XFS_MOUNT_H__ */