ocfs2.h 15.5 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
/* -*- mode: c; c-basic-offset: 8; -*-
 * vim: noexpandtab sw=8 ts=8 sts=0:
 *
 * ocfs2.h
 *
 * Defines macros and structures used in OCFS2
 *
 * Copyright (C) 2002, 2004 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License as published by the Free Software Foundation; either
 * version 2 of the License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will 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.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#ifndef OCFS2_H
#define OCFS2_H

#include <linux/spinlock.h>
#include <linux/sched.h>
#include <linux/wait.h>
#include <linux/list.h>
#include <linux/rbtree.h>
#include <linux/workqueue.h>
#include <linux/kref.h>
36
#include <linux/mutex.h>
37
#include <linux/jbd.h>
38

39 40
/* For union ocfs2_dlm_lksb */
#include "stackglue.h"
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99

#include "ocfs2_fs.h"
#include "ocfs2_lockid.h"

/* Most user visible OCFS2 inodes will have very few pieces of
 * metadata, but larger files (including bitmaps, etc) must be taken
 * into account when designing an access scheme. We allow a small
 * amount of inlined blocks to be stored on an array and grow the
 * structure into a rb tree when necessary. */
#define OCFS2_INODE_MAX_CACHE_ARRAY 2

struct ocfs2_caching_info {
	unsigned int		ci_num_cached;
	union {
		sector_t	ci_array[OCFS2_INODE_MAX_CACHE_ARRAY];
		struct rb_root	ci_tree;
	} ci_cache;
};

/* this limits us to 256 nodes
 * if we need more, we can do a kmalloc for the map */
#define OCFS2_NODE_MAP_MAX_NODES    256
struct ocfs2_node_map {
	u16 num_nodes;
	unsigned long map[BITS_TO_LONGS(OCFS2_NODE_MAP_MAX_NODES)];
};

enum ocfs2_ast_action {
	OCFS2_AST_INVALID = 0,
	OCFS2_AST_ATTACH,
	OCFS2_AST_CONVERT,
	OCFS2_AST_DOWNCONVERT,
};

/* actions for an unlockast function to take. */
enum ocfs2_unlock_action {
	OCFS2_UNLOCK_INVALID = 0,
	OCFS2_UNLOCK_CANCEL_CONVERT,
	OCFS2_UNLOCK_DROP_LOCK,
};

/* ocfs2_lock_res->l_flags flags. */
#define OCFS2_LOCK_ATTACHED      (0x00000001) /* have we initialized
					       * the lvb */
#define OCFS2_LOCK_BUSY          (0x00000002) /* we are currently in
					       * dlm_lock */
#define OCFS2_LOCK_BLOCKED       (0x00000004) /* blocked waiting to
					       * downconvert*/
#define OCFS2_LOCK_LOCAL         (0x00000008) /* newly created inode */
#define OCFS2_LOCK_NEEDS_REFRESH (0x00000010)
#define OCFS2_LOCK_REFRESHING    (0x00000020)
#define OCFS2_LOCK_INITIALIZED   (0x00000040) /* track initialization
					       * for shutdown paths */
#define OCFS2_LOCK_FREEING       (0x00000080) /* help dlmglue track
					       * when to skip queueing
					       * a lock because it's
					       * about to be
					       * dropped. */
#define OCFS2_LOCK_QUEUED        (0x00000100) /* queued for downconvert */
100
#define OCFS2_LOCK_NOCACHE       (0x00000200) /* don't use a holder count */
101 102 103
#define OCFS2_LOCK_PENDING       (0x00000400) /* This lockres is pending a
						 call to dlm_lock.  Only
						 exists with BUSY set. */
104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122

struct ocfs2_lock_res_ops;

typedef void (*ocfs2_lock_callback)(int status, unsigned long data);

struct ocfs2_lock_res {
	void                    *l_priv;
	struct ocfs2_lock_res_ops *l_ops;
	spinlock_t               l_lock;

	struct list_head         l_blocked_list;
	struct list_head         l_mask_waiters;

	enum ocfs2_lock_type     l_type;
	unsigned long		 l_flags;
	char                     l_name[OCFS2_LOCK_ID_MAX_LEN];
	int                      l_level;
	unsigned int             l_ro_holders;
	unsigned int             l_ex_holders;
123
	union ocfs2_dlm_lksb     l_lksb;
124 125 126 127 128 129

	/* used from AST/BAST funcs. */
	enum ocfs2_ast_action    l_action;
	enum ocfs2_unlock_action l_unlock_action;
	int                      l_requested;
	int                      l_blocking;
130
	unsigned int             l_pending_gen;
131 132 133 134

	wait_queue_head_t        l_event;

	struct list_head         l_debug_list;
135 136 137 138 139 140 141 142 143 144 145 146

#ifdef CONFIG_OCFS2_FS_STATS
	unsigned long long	 l_lock_num_prmode; 	   /* PR acquires */
	unsigned long long 	 l_lock_num_exmode; 	   /* EX acquires */
	unsigned int		 l_lock_num_prmode_failed; /* Failed PR gets */
	unsigned int		 l_lock_num_exmode_failed; /* Failed EX gets */
	unsigned long long	 l_lock_total_prmode; 	   /* Tot wait for PR */
	unsigned long long	 l_lock_total_exmode; 	   /* Tot wait for EX */
	unsigned int		 l_lock_max_prmode; 	   /* Max wait for PR */
	unsigned int		 l_lock_max_exmode; 	   /* Max wait for EX */
	unsigned int		 l_lock_refresh;	   /* Disk refreshes */
#endif
147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
};

struct ocfs2_dlm_debug {
	struct kref d_refcnt;
	struct dentry *d_locking_state;
	struct list_head d_lockres_tracking;
};

enum ocfs2_vol_state
{
	VOLUME_INIT = 0,
	VOLUME_MOUNTED,
	VOLUME_DISMOUNTED,
	VOLUME_DISABLED
};

struct ocfs2_alloc_stats
{
	atomic_t moves;
	atomic_t local_data;
	atomic_t bitmap_data;
	atomic_t bg_allocs;
	atomic_t bg_extends;
};

enum ocfs2_local_alloc_state
{
	OCFS2_LA_UNUSED = 0,
	OCFS2_LA_ENABLED,
	OCFS2_LA_DISABLED
};

enum ocfs2_mount_options
{
	OCFS2_MOUNT_HB_LOCAL   = 1 << 0, /* Heartbeat started in local mode */
	OCFS2_MOUNT_BARRIER = 1 << 1,	/* Use block barriers */
	OCFS2_MOUNT_NOINTR  = 1 << 2,   /* Don't catch signals */
	OCFS2_MOUNT_ERRORS_PANIC = 1 << 3, /* Panic on errors */
	OCFS2_MOUNT_DATA_WRITEBACK = 1 << 4, /* No data ordering */
186
	OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */
187 188 189 190 191
};

#define OCFS2_OSB_SOFT_RO	0x0001
#define OCFS2_OSB_HARD_RO	0x0002
#define OCFS2_OSB_ERROR_FS	0x0004
T
Tiger Yang 已提交
192
#define OCFS2_DEFAULT_ATIME_QUANTUM	60
193 194

struct ocfs2_journal;
195
struct ocfs2_slot_info;
196
struct ocfs2_recovery_map;
197 198 199 200 201 202 203 204 205 206
struct ocfs2_super
{
	struct task_struct *commit_task;
	struct super_block *sb;
	struct inode *root_inode;
	struct inode *sys_root_inode;
	struct inode *system_inodes[NUM_SYSTEM_INODES];

	struct ocfs2_slot_info *slot_info;

207 208
	u32 *slot_recovery_generations;

209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
	spinlock_t node_map_lock;

	u64 root_blkno;
	u64 system_dir_blkno;
	u64 bitmap_blkno;
	u32 bitmap_cpg;
	u8 *uuid;
	char *uuid_str;
	u8 *vol_label;
	u64 first_cluster_group_blkno;
	u32 fs_generation;

	u32 s_feature_compat;
	u32 s_feature_incompat;
	u32 s_feature_ro_compat;

225 226 227
	/* Protects s_next_generation, osb_flags and s_inode_steal_slot.
	 * Could protect more on osb as it's very short lived.
	 */
228 229 230
	spinlock_t osb_lock;
	u32 s_next_generation;
	unsigned long osb_flags;
231 232
	s16 s_inode_steal_slot;
	atomic_t s_num_inodes_stolen;
233 234

	unsigned long s_mount_opt;
T
Tiger Yang 已提交
235
	unsigned int s_atime_quantum;
236

237
	unsigned int max_slots;
238
	unsigned int node_num;
239 240
	int slot_num;
	int preferred_slot;
241 242 243 244 245
	int s_sectsize_bits;
	int s_clustersize;
	int s_clustersize_bits;

	atomic_t vol_state;
246
	struct mutex recovery_lock;
247
	struct ocfs2_recovery_map *recovery_map;
248 249 250 251 252
	struct task_struct *recovery_thread_task;
	int disable_recovery;
	wait_queue_head_t checkpoint_event;
	atomic_t needs_checkpoint;
	struct ocfs2_journal *journal;
253
	unsigned long osb_commit_interval;
254

255
	int local_alloc_size;
256 257
	enum ocfs2_local_alloc_state local_alloc_state;
	struct buffer_head *local_alloc_bh;
M
Mark Fasheh 已提交
258
	u64 la_last_gd;
259 260 261 262 263 264 265 266 267

	/* Next two fields are for local node slot recovery during
	 * mount. */
	int dirty;
	struct ocfs2_dinode *local_alloc_copy;

	struct ocfs2_alloc_stats alloc_stats;
	char dev_str[20];		/* "major,minor" of the device */

268
	char osb_cluster_stack[OCFS2_STACK_LABEL_LEN + 1];
269
	struct ocfs2_cluster_connection *cconn;
270 271 272 273 274 275 276 277
	struct ocfs2_lock_res osb_super_lockres;
	struct ocfs2_lock_res osb_rename_lockres;
	struct ocfs2_dlm_debug *osb_dlm_debug;

	struct dentry *osb_debug_root;

	wait_queue_head_t recovery_event;

M
Mark Fasheh 已提交
278 279 280 281 282
	spinlock_t dc_task_lock;
	struct task_struct *dc_task;
	wait_queue_head_t dc_event;
	unsigned long dc_wake_sequence;
	unsigned long dc_work_sequence;
283

284 285 286 287 288 289
	/*
	 * Any thread can add locks to the list, but the downconvert
	 * thread is the only one allowed to remove locks. Any change
	 * to this rule requires updating
	 * ocfs2_downconvert_thread_do_work().
	 */
290 291 292 293 294 295 296 297
	struct list_head blocked_lock_list;
	unsigned long blocked_lock_count;

	wait_queue_head_t		osb_mount_event;

	/* Truncate log info */
	struct inode			*osb_tl_inode;
	struct buffer_head		*osb_tl_bh;
D
David Howells 已提交
298
	struct delayed_work		osb_truncate_log_wq;
299 300 301 302

	struct ocfs2_node_map		osb_recovering_orphan_dirs;
	unsigned int			*osb_orphan_wipes;
	wait_queue_head_t		osb_wipe_event;
303 304 305 306 307 308 309 310 311 312 313 314 315
};

#define OCFS2_SB(sb)	    ((struct ocfs2_super *)(sb)->s_fs_info)

static inline int ocfs2_should_order_data(struct inode *inode)
{
	if (!S_ISREG(inode->i_mode))
		return 0;
	if (OCFS2_SB(inode->i_sb)->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK)
		return 0;
	return 1;
}

M
Mark Fasheh 已提交
316 317 318 319 320 321 322
static inline int ocfs2_sparse_alloc(struct ocfs2_super *osb)
{
	if (osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_SPARSE_ALLOC)
		return 1;
	return 0;
}

323 324 325 326 327 328 329 330 331 332 333 334 335
static inline int ocfs2_writes_unwritten_extents(struct ocfs2_super *osb)
{
	/*
	 * Support for sparse files is a pre-requisite
	 */
	if (!ocfs2_sparse_alloc(osb))
		return 0;

	if (osb->s_feature_ro_compat & OCFS2_FEATURE_RO_COMPAT_UNWRITTEN)
		return 1;
	return 0;
}

336 337 338 339 340 341 342
static inline int ocfs2_supports_inline_data(struct ocfs2_super *osb)
{
	if (osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_INLINE_DATA)
		return 1;
	return 0;
}

343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
/* set / clear functions because cluster events can make these happen
 * in parallel so we want the transitions to be atomic. this also
 * means that any future flags osb_flags must be protected by spinlock
 * too! */
static inline void ocfs2_set_osb_flag(struct ocfs2_super *osb,
				      unsigned long flag)
{
	spin_lock(&osb->osb_lock);
	osb->osb_flags |= flag;
	spin_unlock(&osb->osb_lock);
}

static inline void ocfs2_set_ro_flag(struct ocfs2_super *osb,
				     int hard)
{
	spin_lock(&osb->osb_lock);
	osb->osb_flags &= ~(OCFS2_OSB_SOFT_RO|OCFS2_OSB_HARD_RO);
	if (hard)
		osb->osb_flags |= OCFS2_OSB_HARD_RO;
	else
		osb->osb_flags |= OCFS2_OSB_SOFT_RO;
	spin_unlock(&osb->osb_lock);
}

static inline int ocfs2_is_hard_readonly(struct ocfs2_super *osb)
{
	int ret;

	spin_lock(&osb->osb_lock);
	ret = osb->osb_flags & OCFS2_OSB_HARD_RO;
	spin_unlock(&osb->osb_lock);

	return ret;
}

static inline int ocfs2_is_soft_readonly(struct ocfs2_super *osb)
{
	int ret;

	spin_lock(&osb->osb_lock);
	ret = osb->osb_flags & OCFS2_OSB_SOFT_RO;
	spin_unlock(&osb->osb_lock);

	return ret;
}

389 390 391 392 393 394
static inline int ocfs2_userspace_stack(struct ocfs2_super *osb)
{
	return (osb->s_feature_incompat &
		OCFS2_FEATURE_INCOMPAT_USERSPACE_STACK);
}

S
Sunil Mushran 已提交
395 396 397 398 399
static inline int ocfs2_mount_local(struct ocfs2_super *osb)
{
	return (osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT);
}

J
Joel Becker 已提交
400 401 402 403 404 405 406
static inline int ocfs2_uses_extended_slot_map(struct ocfs2_super *osb)
{
	return (osb->s_feature_incompat &
		OCFS2_FEATURE_INCOMPAT_EXTENDED_SLOT_MAP);
}


407 408 409 410 411 412
#define OCFS2_IS_VALID_DINODE(ptr)					\
	(!strcmp((ptr)->i_signature, OCFS2_INODE_SIGNATURE))

#define OCFS2_RO_ON_INVALID_DINODE(__sb, __di)	do {			\
	typeof(__di) ____di = (__di);					\
	ocfs2_error((__sb), 						\
413
		"Dinode # %llu has bad signature %.*s",			\
414
		(unsigned long long)le64_to_cpu((____di)->i_blkno), 7, 	\
415
		(____di)->i_signature);					\
416
} while (0)
417 418 419 420 421 422 423

#define OCFS2_IS_VALID_EXTENT_BLOCK(ptr)				\
	(!strcmp((ptr)->h_signature, OCFS2_EXTENT_BLOCK_SIGNATURE))

#define OCFS2_RO_ON_INVALID_EXTENT_BLOCK(__sb, __eb)	do {		\
	typeof(__eb) ____eb = (__eb);					\
	ocfs2_error((__sb), 						\
424
		"Extent Block # %llu has bad signature %.*s",		\
425
		(unsigned long long)le64_to_cpu((____eb)->h_blkno), 7,	\
426
		(____eb)->h_signature);					\
427
} while (0)
428 429 430 431 432 433 434

#define OCFS2_IS_VALID_GROUP_DESC(ptr)					\
	(!strcmp((ptr)->bg_signature, OCFS2_GROUP_DESC_SIGNATURE))

#define OCFS2_RO_ON_INVALID_GROUP_DESC(__sb, __gd)	do {		\
	typeof(__gd) ____gd = (__gd);					\
		ocfs2_error((__sb),					\
435
		"Group Descriptor # %llu has bad signature %.*s",	\
436
		(unsigned long long)le64_to_cpu((____gd)->bg_blkno), 7, \
437
		(____gd)->bg_signature);				\
438
} while (0)
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513

static inline unsigned long ino_from_blkno(struct super_block *sb,
					   u64 blkno)
{
	return (unsigned long)(blkno & (u64)ULONG_MAX);
}

static inline u64 ocfs2_clusters_to_blocks(struct super_block *sb,
					   u32 clusters)
{
	int c_to_b_bits = OCFS2_SB(sb)->s_clustersize_bits -
		sb->s_blocksize_bits;

	return (u64)clusters << c_to_b_bits;
}

static inline u32 ocfs2_blocks_to_clusters(struct super_block *sb,
					   u64 blocks)
{
	int b_to_c_bits = OCFS2_SB(sb)->s_clustersize_bits -
		sb->s_blocksize_bits;

	return (u32)(blocks >> b_to_c_bits);
}

static inline unsigned int ocfs2_clusters_for_bytes(struct super_block *sb,
						    u64 bytes)
{
	int cl_bits = OCFS2_SB(sb)->s_clustersize_bits;
	unsigned int clusters;

	bytes += OCFS2_SB(sb)->s_clustersize - 1;
	/* OCFS2 just cannot have enough clusters to overflow this */
	clusters = (unsigned int)(bytes >> cl_bits);

	return clusters;
}

static inline u64 ocfs2_blocks_for_bytes(struct super_block *sb,
					 u64 bytes)
{
	bytes += sb->s_blocksize - 1;
	return bytes >> sb->s_blocksize_bits;
}

static inline u64 ocfs2_clusters_to_bytes(struct super_block *sb,
					  u32 clusters)
{
	return (u64)clusters << OCFS2_SB(sb)->s_clustersize_bits;
}

static inline u64 ocfs2_align_bytes_to_clusters(struct super_block *sb,
						u64 bytes)
{
	int cl_bits = OCFS2_SB(sb)->s_clustersize_bits;
	unsigned int clusters;

	clusters = ocfs2_clusters_for_bytes(sb, bytes);
	return (u64)clusters << cl_bits;
}

static inline u64 ocfs2_align_bytes_to_blocks(struct super_block *sb,
					      u64 bytes)
{
	u64 blocks;

        blocks = ocfs2_blocks_for_bytes(sb, bytes);
	return blocks << sb->s_blocksize_bits;
}

static inline unsigned long ocfs2_align_bytes_to_sectors(u64 bytes)
{
	return (unsigned long)((bytes + 511) >> 9);
}

514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
static inline unsigned int ocfs2_page_index_to_clusters(struct super_block *sb,
							unsigned long pg_index)
{
	u32 clusters = pg_index;
	unsigned int cbits = OCFS2_SB(sb)->s_clustersize_bits;

	if (unlikely(PAGE_CACHE_SHIFT > cbits))
		clusters = pg_index << (PAGE_CACHE_SHIFT - cbits);
	else if (PAGE_CACHE_SHIFT < cbits)
		clusters = pg_index >> (cbits - PAGE_CACHE_SHIFT);

	return clusters;
}

/*
 * Find the 1st page index which covers the given clusters.
 */
531
static inline pgoff_t ocfs2_align_clusters_to_page_index(struct super_block *sb,
532 533 534
							u32 clusters)
{
	unsigned int cbits = OCFS2_SB(sb)->s_clustersize_bits;
535
        pgoff_t index = clusters;
536 537

	if (PAGE_CACHE_SHIFT > cbits) {
538
		index = (pgoff_t)clusters >> (PAGE_CACHE_SHIFT - cbits);
539
	} else if (PAGE_CACHE_SHIFT < cbits) {
540
		index = (pgoff_t)clusters << (cbits - PAGE_CACHE_SHIFT);
541 542 543 544 545
	}

	return index;
}

546 547 548 549 550 551 552 553 554 555 556
static inline unsigned int ocfs2_pages_per_cluster(struct super_block *sb)
{
	unsigned int cbits = OCFS2_SB(sb)->s_clustersize_bits;
	unsigned int pages_per_cluster = 1;

	if (PAGE_CACHE_SHIFT < cbits)
		pages_per_cluster = 1 << (cbits - PAGE_CACHE_SHIFT);

	return pages_per_cluster;
}

557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583
static inline void ocfs2_init_inode_steal_slot(struct ocfs2_super *osb)
{
	spin_lock(&osb->osb_lock);
	osb->s_inode_steal_slot = OCFS2_INVALID_SLOT;
	spin_unlock(&osb->osb_lock);
	atomic_set(&osb->s_num_inodes_stolen, 0);
}

static inline void ocfs2_set_inode_steal_slot(struct ocfs2_super *osb,
					      s16 slot)
{
	spin_lock(&osb->osb_lock);
	osb->s_inode_steal_slot = slot;
	spin_unlock(&osb->osb_lock);
}

static inline s16 ocfs2_get_inode_steal_slot(struct ocfs2_super *osb)
{
	s16 slot;

	spin_lock(&osb->osb_lock);
	slot = osb->s_inode_steal_slot;
	spin_unlock(&osb->osb_lock);

	return slot;
}

584 585 586 587 588 589
#define ocfs2_set_bit ext2_set_bit
#define ocfs2_clear_bit ext2_clear_bit
#define ocfs2_test_bit ext2_test_bit
#define ocfs2_find_next_zero_bit ext2_find_next_zero_bit
#endif  /* OCFS2_H */