super.c 61.6 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
/*
 * Copyright 2000 by Hans Reiser, licensing governed by reiserfs/README
 *
 * Trivial changes by Alan Cox to add the LFS fixes
 *
 * Trivial Changes:
 * Rights granted to Hans Reiser to redistribute under other terms providing
 * he accepts all liability including but not limited to patent, fitness
 * for purpose, and direct or indirect claims arising from failure to perform.
 *
 * NO WARRANTY
 */

#include <linux/module.h>
#include <linux/vmalloc.h>
#include <linux/time.h>
#include <asm/uaccess.h>
#include <linux/reiserfs_fs.h>
#include <linux/reiserfs_acl.h>
#include <linux/reiserfs_xattr.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/buffer_head.h>
24
#include <linux/exportfs.h>
L
Linus Torvalds 已提交
25
#include <linux/vfs.h>
26
#include <linux/mnt_namespace.h>
L
Linus Torvalds 已提交
27 28 29 30 31 32 33 34 35 36
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/quotaops.h>

struct file_system_type reiserfs_fs_type;

static const char reiserfs_3_5_magic_string[] = REISERFS_SUPER_MAGIC_STRING;
static const char reiserfs_3_6_magic_string[] = REISER2FS_SUPER_MAGIC_STRING;
static const char reiserfs_jr_magic_string[] = REISER2FS_JR_SUPER_MAGIC_STRING;

37
int is_reiserfs_3_5(struct reiserfs_super_block *rs)
L
Linus Torvalds 已提交
38
{
39 40
	return !strncmp(rs->s_v1.s_magic, reiserfs_3_5_magic_string,
			strlen(reiserfs_3_5_magic_string));
L
Linus Torvalds 已提交
41 42
}

43
int is_reiserfs_3_6(struct reiserfs_super_block *rs)
L
Linus Torvalds 已提交
44
{
45 46
	return !strncmp(rs->s_v1.s_magic, reiserfs_3_6_magic_string,
			strlen(reiserfs_3_6_magic_string));
L
Linus Torvalds 已提交
47 48
}

49
int is_reiserfs_jr(struct reiserfs_super_block *rs)
L
Linus Torvalds 已提交
50
{
51 52
	return !strncmp(rs->s_v1.s_magic, reiserfs_jr_magic_string,
			strlen(reiserfs_jr_magic_string));
L
Linus Torvalds 已提交
53 54
}

55
static int is_any_reiserfs_magic_string(struct reiserfs_super_block *rs)
L
Linus Torvalds 已提交
56
{
57 58
	return (is_reiserfs_3_5(rs) || is_reiserfs_3_6(rs) ||
		is_reiserfs_jr(rs));
L
Linus Torvalds 已提交
59 60
}

61
static int reiserfs_remount(struct super_block *s, int *flags, char *data);
62
static int reiserfs_statfs(struct dentry *dentry, struct kstatfs *buf);
L
Linus Torvalds 已提交
63

64
static int reiserfs_sync_fs(struct super_block *s, int wait)
L
Linus Torvalds 已提交
65
{
66 67 68 69 70 71 72 73 74 75 76 77 78
	if (!(s->s_flags & MS_RDONLY)) {
		struct reiserfs_transaction_handle th;
		reiserfs_write_lock(s);
		if (!journal_begin(&th, s, 1))
			if (!journal_end_sync(&th, s, 1))
				reiserfs_flush_old_commits(s);
		s->s_dirt = 0;	/* Even if it's not true.
				 * We'll loop forever in sync_supers otherwise */
		reiserfs_write_unlock(s);
	} else {
		s->s_dirt = 0;
	}
	return 0;
L
Linus Torvalds 已提交
79 80 81 82
}

static void reiserfs_write_super(struct super_block *s)
{
83
	reiserfs_sync_fs(s, 1);
L
Linus Torvalds 已提交
84 85
}

86
static void reiserfs_write_super_lockfs(struct super_block *s)
L
Linus Torvalds 已提交
87
{
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
	struct reiserfs_transaction_handle th;
	reiserfs_write_lock(s);
	if (!(s->s_flags & MS_RDONLY)) {
		int err = journal_begin(&th, s, 1);
		if (err) {
			reiserfs_block_writes(&th);
		} else {
			reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s),
						     1);
			journal_mark_dirty(&th, s, SB_BUFFER_WITH_SB(s));
			reiserfs_block_writes(&th);
			journal_end_sync(&th, s, 1);
		}
	}
	s->s_dirt = 0;
	reiserfs_write_unlock(s);
L
Linus Torvalds 已提交
104 105
}

106 107 108
static void reiserfs_unlockfs(struct super_block *s)
{
	reiserfs_allow_writes(s);
L
Linus Torvalds 已提交
109 110
}

111
extern const struct in_core_key MAX_IN_CORE_KEY;
L
Linus Torvalds 已提交
112 113 114 115 116 117 118 119

/* this is used to delete "save link" when there are no items of a
   file it points to. It can either happen if unlink is completed but
   "save unlink" removal, or if file has both unlink and truncate
   pending and as unlink completes first (because key of "save link"
   protecting unlink is bigger that a key lf "save link" which
   protects truncate), so there left no items to make truncate
   completion on */
120 121
static int remove_save_link_only(struct super_block *s,
				 struct reiserfs_key *key, int oid_free)
L
Linus Torvalds 已提交
122
{
123 124 125 126 127 128 129 130 131 132 133 134 135 136
	struct reiserfs_transaction_handle th;
	int err;

	/* we are going to do one balancing */
	err = journal_begin(&th, s, JOURNAL_PER_BALANCE_CNT);
	if (err)
		return err;

	reiserfs_delete_solid_item(&th, NULL, key);
	if (oid_free)
		/* removals are protected by direct items */
		reiserfs_release_objectid(&th, le32_to_cpu(key->k_objectid));

	return journal_end(&th, s, JOURNAL_PER_BALANCE_CNT);
L
Linus Torvalds 已提交
137
}
138

L
Linus Torvalds 已提交
139 140 141
#ifdef CONFIG_QUOTA
static int reiserfs_quota_on_mount(struct super_block *, int);
#endif
142

L
Linus Torvalds 已提交
143
/* look for uncompleted unlinks and truncates and complete them */
144
static int finish_unfinished(struct super_block *s)
L
Linus Torvalds 已提交
145
{
146 147
	INITIALIZE_PATH(path);
	struct cpu_key max_cpu_key, obj_key;
148
	struct reiserfs_key save_link_key, last_inode_key;
149 150 151 152 153 154 155 156
	int retval = 0;
	struct item_head *ih;
	struct buffer_head *bh;
	int item_pos;
	char *item;
	int done;
	struct inode *inode;
	int truncate;
L
Linus Torvalds 已提交
157
#ifdef CONFIG_QUOTA
158 159
	int i;
	int ms_active_set;
L
Linus Torvalds 已提交
160
#endif
161 162 163 164 165 166 167

	/* compose key to look for "save" links */
	max_cpu_key.version = KEY_FORMAT_3_5;
	max_cpu_key.on_disk_key.k_dir_id = ~0U;
	max_cpu_key.on_disk_key.k_objectid = ~0U;
	set_cpu_key_k_offset(&max_cpu_key, ~0U);
	max_cpu_key.key_length = 3;
L
Linus Torvalds 已提交
168

169 170
	memset(&last_inode_key, 0, sizeof(last_inode_key));

L
Linus Torvalds 已提交
171
#ifdef CONFIG_QUOTA
172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
	/* Needed for iput() to work correctly and not trash data */
	if (s->s_flags & MS_ACTIVE) {
		ms_active_set = 0;
	} else {
		ms_active_set = 1;
		s->s_flags |= MS_ACTIVE;
	}
	/* Turn on quotas so that they are updated correctly */
	for (i = 0; i < MAXQUOTAS; i++) {
		if (REISERFS_SB(s)->s_qf_names[i]) {
			int ret = reiserfs_quota_on_mount(s, i);
			if (ret < 0)
				reiserfs_warning(s,
						 "reiserfs: cannot turn on journalled quota: error %d",
						 ret);
		}
	}
L
Linus Torvalds 已提交
189
#endif
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282

	done = 0;
	REISERFS_SB(s)->s_is_unlinked_ok = 1;
	while (!retval) {
		retval = search_item(s, &max_cpu_key, &path);
		if (retval != ITEM_NOT_FOUND) {
			reiserfs_warning(s,
					 "vs-2140: finish_unfinished: search_by_key returned %d",
					 retval);
			break;
		}

		bh = get_last_bh(&path);
		item_pos = get_item_pos(&path);
		if (item_pos != B_NR_ITEMS(bh)) {
			reiserfs_warning(s,
					 "vs-2060: finish_unfinished: wrong position found");
			break;
		}
		item_pos--;
		ih = B_N_PITEM_HEAD(bh, item_pos);

		if (le32_to_cpu(ih->ih_key.k_dir_id) != MAX_KEY_OBJECTID)
			/* there are no "save" links anymore */
			break;

		save_link_key = ih->ih_key;
		if (is_indirect_le_ih(ih))
			truncate = 1;
		else
			truncate = 0;

		/* reiserfs_iget needs k_dirid and k_objectid only */
		item = B_I_PITEM(bh, ih);
		obj_key.on_disk_key.k_dir_id = le32_to_cpu(*(__le32 *) item);
		obj_key.on_disk_key.k_objectid =
		    le32_to_cpu(ih->ih_key.k_objectid);
		obj_key.on_disk_key.k_offset = 0;
		obj_key.on_disk_key.k_type = 0;

		pathrelse(&path);

		inode = reiserfs_iget(s, &obj_key);
		if (!inode) {
			/* the unlink almost completed, it just did not manage to remove
			   "save" link and release objectid */
			reiserfs_warning(s,
					 "vs-2180: finish_unfinished: iget failed for %K",
					 &obj_key);
			retval = remove_save_link_only(s, &save_link_key, 1);
			continue;
		}

		if (!truncate && inode->i_nlink) {
			/* file is not unlinked */
			reiserfs_warning(s,
					 "vs-2185: finish_unfinished: file %K is not unlinked",
					 &obj_key);
			retval = remove_save_link_only(s, &save_link_key, 0);
			continue;
		}
		DQUOT_INIT(inode);

		if (truncate && S_ISDIR(inode->i_mode)) {
			/* We got a truncate request for a dir which is impossible.
			   The only imaginable way is to execute unfinished truncate request
			   then boot into old kernel, remove the file and create dir with
			   the same key. */
			reiserfs_warning(s,
					 "green-2101: impossible truncate on a directory %k. Please report",
					 INODE_PKEY(inode));
			retval = remove_save_link_only(s, &save_link_key, 0);
			truncate = 0;
			iput(inode);
			continue;
		}

		if (truncate) {
			REISERFS_I(inode)->i_flags |=
			    i_link_saved_truncate_mask;
			/* not completed truncate found. New size was committed together
			   with "save" link */
			reiserfs_info(s, "Truncating %k to %Ld ..",
				      INODE_PKEY(inode), inode->i_size);
			reiserfs_truncate_file(inode,
					       0
					       /*don't update modification time */
					       );
			retval = remove_save_link(inode, truncate);
		} else {
			REISERFS_I(inode)->i_flags |= i_link_saved_unlink_mask;
			/* not completed unlink (rmdir) found */
			reiserfs_info(s, "Removing %k..", INODE_PKEY(inode));
283 284 285 286 287 288 289 290 291 292 293 294
			if (memcmp(&last_inode_key, INODE_PKEY(inode),
					sizeof(last_inode_key))){
				last_inode_key = *INODE_PKEY(inode);
				/* removal gets completed in iput */
				retval = 0;
			} else {
				reiserfs_warning(s, "Dead loop in "
						"finish_unfinished detected, "
						"just remove save link\n");
				retval = remove_save_link_only(s,
							&save_link_key, 0);
			}
295 296 297 298 299 300 301 302
		}

		iput(inode);
		printk("done\n");
		done++;
	}
	REISERFS_SB(s)->s_is_unlinked_ok = 0;

L
Linus Torvalds 已提交
303
#ifdef CONFIG_QUOTA
304 305 306 307 308 309 310 311
	/* Turn quotas off */
	for (i = 0; i < MAXQUOTAS; i++) {
		if (sb_dqopt(s)->files[i])
			vfs_quota_off_mount(s, i);
	}
	if (ms_active_set)
		/* Restore the flag back */
		s->s_flags &= ~MS_ACTIVE;
L
Linus Torvalds 已提交
312
#endif
313 314 315 316 317
	pathrelse(&path);
	if (done)
		reiserfs_info(s, "There were %d uncompleted unlinks/truncates. "
			      "Completed\n", done);
	return retval;
L
Linus Torvalds 已提交
318
}
319

L
Linus Torvalds 已提交
320 321
/* to protect file being unlinked from getting lost we "safe" link files
   being unlinked. This link will be deleted in the same transaction with last
322
   item of file. mounting the filesystem we scan all these links and remove
L
Linus Torvalds 已提交
323
   files which almost got lost */
324 325
void add_save_link(struct reiserfs_transaction_handle *th,
		   struct inode *inode, int truncate)
L
Linus Torvalds 已提交
326
{
327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
	INITIALIZE_PATH(path);
	int retval;
	struct cpu_key key;
	struct item_head ih;
	__le32 link;

	BUG_ON(!th->t_trans_id);

	/* file can only get one "save link" of each kind */
	RFALSE(truncate &&
	       (REISERFS_I(inode)->i_flags & i_link_saved_truncate_mask),
	       "saved link already exists for truncated inode %lx",
	       (long)inode->i_ino);
	RFALSE(!truncate &&
	       (REISERFS_I(inode)->i_flags & i_link_saved_unlink_mask),
	       "saved link already exists for unlinked inode %lx",
	       (long)inode->i_ino);

	/* setup key of "save" link */
	key.version = KEY_FORMAT_3_5;
	key.on_disk_key.k_dir_id = MAX_KEY_OBJECTID;
	key.on_disk_key.k_objectid = inode->i_ino;
	if (!truncate) {
		/* unlink, rmdir, rename */
		set_cpu_key_k_offset(&key, 1 + inode->i_sb->s_blocksize);
		set_cpu_key_k_type(&key, TYPE_DIRECT);

		/* item head of "safe" link */
		make_le_item_head(&ih, &key, key.version,
				  1 + inode->i_sb->s_blocksize, TYPE_DIRECT,
				  4 /*length */ , 0xffff /*free space */ );
	} else {
		/* truncate */
		if (S_ISDIR(inode->i_mode))
			reiserfs_warning(inode->i_sb,
					 "green-2102: Adding a truncate savelink for a directory %k! Please report",
					 INODE_PKEY(inode));
		set_cpu_key_k_offset(&key, 1);
		set_cpu_key_k_type(&key, TYPE_INDIRECT);

		/* item head of "safe" link */
		make_le_item_head(&ih, &key, key.version, 1, TYPE_INDIRECT,
				  4 /*length */ , 0 /*free space */ );
	}
	key.key_length = 3;

	/* look for its place in the tree */
	retval = search_item(inode->i_sb, &key, &path);
	if (retval != ITEM_NOT_FOUND) {
		if (retval != -ENOSPC)
			reiserfs_warning(inode->i_sb, "vs-2100: add_save_link:"
					 "search_by_key (%K) returned %d", &key,
					 retval);
		pathrelse(&path);
		return;
	}
L
Linus Torvalds 已提交
383

384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
	/* body of "save" link */
	link = INODE_PKEY(inode)->k_dir_id;

	/* put "save" link inot tree, don't charge quota to anyone */
	retval =
	    reiserfs_insert_item(th, &path, &key, &ih, NULL, (char *)&link);
	if (retval) {
		if (retval != -ENOSPC)
			reiserfs_warning(inode->i_sb,
					 "vs-2120: add_save_link: insert_item returned %d",
					 retval);
	} else {
		if (truncate)
			REISERFS_I(inode)->i_flags |=
			    i_link_saved_truncate_mask;
		else
			REISERFS_I(inode)->i_flags |= i_link_saved_unlink_mask;
	}
}
L
Linus Torvalds 已提交
403 404

/* this opens transaction unlike add_save_link */
405
int remove_save_link(struct inode *inode, int truncate)
L
Linus Torvalds 已提交
406
{
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428
	struct reiserfs_transaction_handle th;
	struct reiserfs_key key;
	int err;

	/* we are going to do one balancing only */
	err = journal_begin(&th, inode->i_sb, JOURNAL_PER_BALANCE_CNT);
	if (err)
		return err;

	/* setup key of "save" link */
	key.k_dir_id = cpu_to_le32(MAX_KEY_OBJECTID);
	key.k_objectid = INODE_PKEY(inode)->k_objectid;
	if (!truncate) {
		/* unlink, rmdir, rename */
		set_le_key_k_offset(KEY_FORMAT_3_5, &key,
				    1 + inode->i_sb->s_blocksize);
		set_le_key_k_type(KEY_FORMAT_3_5, &key, TYPE_DIRECT);
	} else {
		/* truncate */
		set_le_key_k_offset(KEY_FORMAT_3_5, &key, 1);
		set_le_key_k_type(KEY_FORMAT_3_5, &key, TYPE_INDIRECT);
	}
L
Linus Torvalds 已提交
429

430 431 432 433 434 435 436 437 438 439 440 441 442 443
	if ((truncate &&
	     (REISERFS_I(inode)->i_flags & i_link_saved_truncate_mask)) ||
	    (!truncate &&
	     (REISERFS_I(inode)->i_flags & i_link_saved_unlink_mask)))
		/* don't take quota bytes from anywhere */
		reiserfs_delete_solid_item(&th, NULL, &key);
	if (!truncate) {
		reiserfs_release_objectid(&th, inode->i_ino);
		REISERFS_I(inode)->i_flags &= ~i_link_saved_unlink_mask;
	} else
		REISERFS_I(inode)->i_flags &= ~i_link_saved_truncate_mask;

	return journal_end(&th, inode->i_sb, JOURNAL_PER_BALANCE_CNT);
}
L
Linus Torvalds 已提交
444

445
static void reiserfs_kill_sb(struct super_block *s)
L
Linus Torvalds 已提交
446
{
447
	if (REISERFS_SB(s)) {
448
#ifdef CONFIG_REISERFS_FS_XATTR
449 450 451 452 453
		if (REISERFS_SB(s)->xattr_root) {
			d_invalidate(REISERFS_SB(s)->xattr_root);
			dput(REISERFS_SB(s)->xattr_root);
			REISERFS_SB(s)->xattr_root = NULL;
		}
454
#endif
455 456 457 458 459
		if (REISERFS_SB(s)->priv_root) {
			d_invalidate(REISERFS_SB(s)->priv_root);
			dput(REISERFS_SB(s)->priv_root);
			REISERFS_SB(s)->priv_root = NULL;
		}
460 461
	}

462 463 464 465 466 467 468
	kill_block_super(s);
}

static void reiserfs_put_super(struct super_block *s)
{
	struct reiserfs_transaction_handle th;
	th.t_trans_id = 0;
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485

	/* change file system state to current state if it was mounted with read-write permissions */
	if (!(s->s_flags & MS_RDONLY)) {
		if (!journal_begin(&th, s, 10)) {
			reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s),
						     1);
			set_sb_umount_state(SB_DISK_SUPER_BLOCK(s),
					    REISERFS_SB(s)->s_mount_state);
			journal_mark_dirty(&th, s, SB_BUFFER_WITH_SB(s));
		}
	}

	/* note, journal_release checks for readonly mount, and can decide not
	 ** to do a journal_end
	 */
	journal_release(&th, s);

486
	reiserfs_free_bitmap_cache(s);
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503

	brelse(SB_BUFFER_WITH_SB(s));

	print_statistics(s);

	if (REISERFS_SB(s)->reserved_blocks != 0) {
		reiserfs_warning(s,
				 "green-2005: reiserfs_put_super: reserved blocks left %d",
				 REISERFS_SB(s)->reserved_blocks);
	}

	reiserfs_proc_info_done(s);

	kfree(s->s_fs_info);
	s->s_fs_info = NULL;

	return;
L
Linus Torvalds 已提交
504 505
}

506
static struct kmem_cache *reiserfs_inode_cachep;
L
Linus Torvalds 已提交
507 508 509 510

static struct inode *reiserfs_alloc_inode(struct super_block *sb)
{
	struct reiserfs_inode_info *ei;
511
	ei = (struct reiserfs_inode_info *)
512
	    kmem_cache_alloc(reiserfs_inode_cachep, GFP_KERNEL);
L
Linus Torvalds 已提交
513 514 515 516 517 518 519 520 521 522
	if (!ei)
		return NULL;
	return &ei->vfs_inode;
}

static void reiserfs_destroy_inode(struct inode *inode)
{
	kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode));
}

523
static void init_once(struct kmem_cache * cachep, void *foo)
L
Linus Torvalds 已提交
524
{
525
	struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo;
L
Linus Torvalds 已提交
526

C
Christoph Lameter 已提交
527 528
	INIT_LIST_HEAD(&ei->i_prealloc_list);
	inode_init_once(&ei->vfs_inode);
529
#ifdef CONFIG_REISERFS_FS_POSIX_ACL
C
Christoph Lameter 已提交
530 531
	ei->i_acl_access = NULL;
	ei->i_acl_default = NULL;
532
#endif
L
Linus Torvalds 已提交
533
}
534

L
Linus Torvalds 已提交
535 536 537
static int init_inodecache(void)
{
	reiserfs_inode_cachep = kmem_cache_create("reiser_inode_cache",
538 539
						  sizeof(struct
							 reiserfs_inode_info),
540 541
						  0, (SLAB_RECLAIM_ACCOUNT|
							SLAB_MEM_SPREAD),
542
						  init_once);
L
Linus Torvalds 已提交
543 544 545 546 547 548 549
	if (reiserfs_inode_cachep == NULL)
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
550
	kmem_cache_destroy(reiserfs_inode_cachep);
L
Linus Torvalds 已提交
551 552 553
}

/* we don't mark inodes dirty, we just log them */
554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577
static void reiserfs_dirty_inode(struct inode *inode)
{
	struct reiserfs_transaction_handle th;

	int err = 0;
	if (inode->i_sb->s_flags & MS_RDONLY) {
		reiserfs_warning(inode->i_sb,
				 "clm-6006: writing inode %lu on readonly FS",
				 inode->i_ino);
		return;
	}
	reiserfs_write_lock(inode->i_sb);

	/* this is really only used for atime updates, so they don't have
	 ** to be included in O_SYNC or fsync
	 */
	err = journal_begin(&th, inode->i_sb, 1);
	if (err) {
		reiserfs_write_unlock(inode->i_sb);
		return;
	}
	reiserfs_update_sd(&th, inode);
	journal_end(&th, inode->i_sb, 1);
	reiserfs_write_unlock(inode->i_sb);
L
Linus Torvalds 已提交
578 579
}

580
#ifdef CONFIG_REISERFS_FS_POSIX_ACL
581
static void reiserfs_clear_inode(struct inode *inode)
L
Linus Torvalds 已提交
582
{
583
	struct posix_acl *acl;
L
Linus Torvalds 已提交
584

585 586 587 588
	acl = REISERFS_I(inode)->i_acl_access;
	if (acl && !IS_ERR(acl))
		posix_acl_release(acl);
	REISERFS_I(inode)->i_acl_access = NULL;
L
Linus Torvalds 已提交
589

590 591 592 593
	acl = REISERFS_I(inode)->i_acl_default;
	if (acl && !IS_ERR(acl))
		posix_acl_release(acl);
	REISERFS_I(inode)->i_acl_default = NULL;
L
Linus Torvalds 已提交
594
}
595 596 597
#else
#define reiserfs_clear_inode NULL
#endif
L
Linus Torvalds 已提交
598 599

#ifdef CONFIG_QUOTA
600 601 602 603
static ssize_t reiserfs_quota_write(struct super_block *, int, const char *,
				    size_t, loff_t);
static ssize_t reiserfs_quota_read(struct super_block *, int, char *, size_t,
				   loff_t);
L
Linus Torvalds 已提交
604 605
#endif

606
static const struct super_operations reiserfs_sops = {
607 608 609 610 611 612 613 614 615 616 617 618 619
	.alloc_inode = reiserfs_alloc_inode,
	.destroy_inode = reiserfs_destroy_inode,
	.write_inode = reiserfs_write_inode,
	.dirty_inode = reiserfs_dirty_inode,
	.delete_inode = reiserfs_delete_inode,
	.clear_inode = reiserfs_clear_inode,
	.put_super = reiserfs_put_super,
	.write_super = reiserfs_write_super,
	.sync_fs = reiserfs_sync_fs,
	.write_super_lockfs = reiserfs_write_super_lockfs,
	.unlockfs = reiserfs_unlockfs,
	.statfs = reiserfs_statfs,
	.remount_fs = reiserfs_remount,
L
Linus Torvalds 已提交
620
#ifdef CONFIG_QUOTA
621 622
	.quota_read = reiserfs_quota_read,
	.quota_write = reiserfs_quota_write,
L
Linus Torvalds 已提交
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637
#endif
};

#ifdef CONFIG_QUOTA
#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group")

static int reiserfs_dquot_initialize(struct inode *, int);
static int reiserfs_dquot_drop(struct inode *);
static int reiserfs_write_dquot(struct dquot *);
static int reiserfs_acquire_dquot(struct dquot *);
static int reiserfs_release_dquot(struct dquot *);
static int reiserfs_mark_dquot_dirty(struct dquot *);
static int reiserfs_write_info(struct super_block *, int);
static int reiserfs_quota_on(struct super_block *, int, int, char *);

638 639 640 641 642 643 644 645 646 647 648 649 650
static struct dquot_operations reiserfs_quota_operations = {
	.initialize = reiserfs_dquot_initialize,
	.drop = reiserfs_dquot_drop,
	.alloc_space = dquot_alloc_space,
	.alloc_inode = dquot_alloc_inode,
	.free_space = dquot_free_space,
	.free_inode = dquot_free_inode,
	.transfer = dquot_transfer,
	.write_dquot = reiserfs_write_dquot,
	.acquire_dquot = reiserfs_acquire_dquot,
	.release_dquot = reiserfs_release_dquot,
	.mark_dirty = reiserfs_mark_dquot_dirty,
	.write_info = reiserfs_write_info,
L
Linus Torvalds 已提交
651 652
};

653 654 655 656 657 658 659 660
static struct quotactl_ops reiserfs_qctl_operations = {
	.quota_on = reiserfs_quota_on,
	.quota_off = vfs_quota_off,
	.quota_sync = vfs_quota_sync,
	.get_info = vfs_get_dqinfo,
	.set_info = vfs_set_dqinfo,
	.get_dqblk = vfs_get_dqblk,
	.set_dqblk = vfs_set_dqblk,
L
Linus Torvalds 已提交
661 662 663 664
};
#endif

static struct export_operations reiserfs_export_ops = {
665
	.encode_fh = reiserfs_encode_fh,
C
Christoph Hellwig 已提交
666 667
	.fh_to_dentry = reiserfs_fh_to_dentry,
	.fh_to_parent = reiserfs_fh_to_parent,
668 669
	.get_parent = reiserfs_get_parent,
};
L
Linus Torvalds 已提交
670 671 672 673

/* this struct is used in reiserfs_getopt () for containing the value for those
   mount options that have values rather than being toggles. */
typedef struct {
674 675 676 677 678 679
	char *value;
	int setmask;		/* bitmask which is to set on mount_options bitmask when this
				   value is found, 0 is no bits are to be changed. */
	int clrmask;		/* bitmask which is to clear on mount_options bitmask when  this
				   value is found, 0 is no bits are to be changed. This is
				   applied BEFORE setmask */
L
Linus Torvalds 已提交
680 681 682 683 684 685 686 687
} arg_desc_t;

/* Set this bit in arg_required to allow empty arguments */
#define REISERFS_OPT_ALLOWEMPTY 31

/* this struct is used in reiserfs_getopt() for describing the set of reiserfs
   mount options */
typedef struct {
688 689 690 691 692 693 694 695
	char *option_name;
	int arg_required;	/* 0 if argument is not required, not 0 otherwise */
	const arg_desc_t *values;	/* list of values accepted by an option */
	int setmask;		/* bitmask which is to set on mount_options bitmask when this
				   value is found, 0 is no bits are to be changed. */
	int clrmask;		/* bitmask which is to clear on mount_options bitmask when  this
				   value is found, 0 is no bits are to be changed. This is
				   applied BEFORE setmask */
L
Linus Torvalds 已提交
696 697 698 699
} opt_desc_t;

/* possible values for -o data= */
static const arg_desc_t logging_mode[] = {
700 701 702 703 704 705
	{"ordered", 1 << REISERFS_DATA_ORDERED,
	 (1 << REISERFS_DATA_LOG | 1 << REISERFS_DATA_WRITEBACK)},
	{"journal", 1 << REISERFS_DATA_LOG,
	 (1 << REISERFS_DATA_ORDERED | 1 << REISERFS_DATA_WRITEBACK)},
	{"writeback", 1 << REISERFS_DATA_WRITEBACK,
	 (1 << REISERFS_DATA_ORDERED | 1 << REISERFS_DATA_LOG)},
706
	{.value = NULL}
L
Linus Torvalds 已提交
707 708 709 710
};

/* possible values for -o barrier= */
static const arg_desc_t barrier_mode[] = {
711 712
	{"none", 1 << REISERFS_BARRIER_NONE, 1 << REISERFS_BARRIER_FLUSH},
	{"flush", 1 << REISERFS_BARRIER_FLUSH, 1 << REISERFS_BARRIER_NONE},
713
	{.value = NULL}
L
Linus Torvalds 已提交
714 715 716 717 718
};

/* possible values for "-o block-allocator=" and bits which are to be set in
   s_mount_opt of reiserfs specific part of in-core super block */
static const arg_desc_t balloc[] = {
719 720 721 722 723 724 725
	{"noborder", 1 << REISERFS_NO_BORDER, 0},
	{"border", 0, 1 << REISERFS_NO_BORDER},
	{"no_unhashed_relocation", 1 << REISERFS_NO_UNHASHED_RELOCATION, 0},
	{"hashed_relocation", 1 << REISERFS_HASHED_RELOCATION, 0},
	{"test4", 1 << REISERFS_TEST4, 0},
	{"notest4", 0, 1 << REISERFS_TEST4},
	{NULL, 0, 0}
L
Linus Torvalds 已提交
726 727 728
};

static const arg_desc_t tails[] = {
729 730 731 732
	{"on", 1 << REISERFS_LARGETAIL, 1 << REISERFS_SMALLTAIL},
	{"off", 0, (1 << REISERFS_LARGETAIL) | (1 << REISERFS_SMALLTAIL)},
	{"small", 1 << REISERFS_SMALLTAIL, 1 << REISERFS_LARGETAIL},
	{NULL, 0, 0}
L
Linus Torvalds 已提交
733 734 735
};

static const arg_desc_t error_actions[] = {
736 737 738 739
	{"panic", 1 << REISERFS_ERROR_PANIC,
	 (1 << REISERFS_ERROR_RO | 1 << REISERFS_ERROR_CONTINUE)},
	{"ro-remount", 1 << REISERFS_ERROR_RO,
	 (1 << REISERFS_ERROR_PANIC | 1 << REISERFS_ERROR_CONTINUE)},
L
Linus Torvalds 已提交
740
#ifdef REISERFS_JOURNAL_ERROR_ALLOWS_NO_LOG
741 742
	{"continue", 1 << REISERFS_ERROR_CONTINUE,
	 (1 << REISERFS_ERROR_PANIC | 1 << REISERFS_ERROR_RO)},
L
Linus Torvalds 已提交
743
#endif
744
	{NULL, 0, 0},
L
Linus Torvalds 已提交
745 746 747 748 749 750 751 752
};

/* proceed only one option from a list *cur - string containing of mount options
   opts - array of options which are accepted
   opt_arg - if option is found and requires an argument and if it is specifed
   in the input - pointer to the argument is stored here
   bit_flags - if option requires to set a certain bit - it is set here
   return -1 if unknown option is found, opt->arg_required otherwise */
753 754
static int reiserfs_getopt(struct super_block *s, char **cur, opt_desc_t * opts,
			   char **opt_arg, unsigned long *bit_flags)
L
Linus Torvalds 已提交
755
{
756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866
	char *p;
	/* foo=bar, 
	   ^   ^  ^
	   |   |  +-- option_end
	   |   +-- arg_start
	   +-- option_start
	 */
	const opt_desc_t *opt;
	const arg_desc_t *arg;

	p = *cur;

	/* assume argument cannot contain commas */
	*cur = strchr(p, ',');
	if (*cur) {
		*(*cur) = '\0';
		(*cur)++;
	}

	if (!strncmp(p, "alloc=", 6)) {
		/* Ugly special case, probably we should redo options parser so that
		   it can understand several arguments for some options, also so that
		   it can fill several bitfields with option values. */
		if (reiserfs_parse_alloc_options(s, p + 6)) {
			return -1;
		} else {
			return 0;
		}
	}

	/* for every option in the list */
	for (opt = opts; opt->option_name; opt++) {
		if (!strncmp(p, opt->option_name, strlen(opt->option_name))) {
			if (bit_flags) {
				if (opt->clrmask ==
				    (1 << REISERFS_UNSUPPORTED_OPT))
					reiserfs_warning(s, "%s not supported.",
							 p);
				else
					*bit_flags &= ~opt->clrmask;
				if (opt->setmask ==
				    (1 << REISERFS_UNSUPPORTED_OPT))
					reiserfs_warning(s, "%s not supported.",
							 p);
				else
					*bit_flags |= opt->setmask;
			}
			break;
		}
	}
	if (!opt->option_name) {
		reiserfs_warning(s, "unknown mount option \"%s\"", p);
		return -1;
	}

	p += strlen(opt->option_name);
	switch (*p) {
	case '=':
		if (!opt->arg_required) {
			reiserfs_warning(s,
					 "the option \"%s\" does not require an argument",
					 opt->option_name);
			return -1;
		}
		break;

	case 0:
		if (opt->arg_required) {
			reiserfs_warning(s,
					 "the option \"%s\" requires an argument",
					 opt->option_name);
			return -1;
		}
		break;
	default:
		reiserfs_warning(s, "head of option \"%s\" is only correct",
				 opt->option_name);
		return -1;
	}

	/* move to the argument, or to next option if argument is not required */
	p++;

	if (opt->arg_required
	    && !(opt->arg_required & (1 << REISERFS_OPT_ALLOWEMPTY))
	    && !strlen(p)) {
		/* this catches "option=," if not allowed */
		reiserfs_warning(s, "empty argument for \"%s\"",
				 opt->option_name);
		return -1;
	}

	if (!opt->values) {
		/* *=NULLopt_arg contains pointer to argument */
		*opt_arg = p;
		return opt->arg_required & ~(1 << REISERFS_OPT_ALLOWEMPTY);
	}

	/* values possible for this option are listed in opt->values */
	for (arg = opt->values; arg->value; arg++) {
		if (!strcmp(p, arg->value)) {
			if (bit_flags) {
				*bit_flags &= ~arg->clrmask;
				*bit_flags |= arg->setmask;
			}
			return opt->arg_required;
		}
	}

	reiserfs_warning(s, "bad value \"%s\" for option \"%s\"", p,
			 opt->option_name);
L
Linus Torvalds 已提交
867 868 869 870
	return -1;
}

/* returns 0 if something is wrong in option string, 1 - otherwise */
871 872 873 874 875 876 877 878
static int reiserfs_parse_options(struct super_block *s, char *options,	/* string given via mount's -o */
				  unsigned long *mount_options,
				  /* after the parsing phase, contains the
				     collection of bitflags defining what
				     mount options were selected. */
				  unsigned long *blocks,	/* strtol-ed from NNN of resize=NNN */
				  char **jdev_name,
				  unsigned int *commit_max_age)
L
Linus Torvalds 已提交
879
{
880 881 882 883 884 885 886 887 888 889 890
	int c;
	char *arg = NULL;
	char *pos;
	opt_desc_t opts[] = {
		/* Compatibility stuff, so that -o notail for old setups still work */
		{"tails",.arg_required = 't',.values = tails},
		{"notail",.clrmask =
		 (1 << REISERFS_LARGETAIL) | (1 << REISERFS_SMALLTAIL)},
		{"conv",.setmask = 1 << REISERFS_CONVERT},
		{"attrs",.setmask = 1 << REISERFS_ATTRS},
		{"noattrs",.clrmask = 1 << REISERFS_ATTRS},
L
Linus Torvalds 已提交
891
#ifdef CONFIG_REISERFS_FS_XATTR
892 893
		{"user_xattr",.setmask = 1 << REISERFS_XATTRS_USER},
		{"nouser_xattr",.clrmask = 1 << REISERFS_XATTRS_USER},
L
Linus Torvalds 已提交
894
#else
895 896
		{"user_xattr",.setmask = 1 << REISERFS_UNSUPPORTED_OPT},
		{"nouser_xattr",.clrmask = 1 << REISERFS_UNSUPPORTED_OPT},
L
Linus Torvalds 已提交
897 898
#endif
#ifdef CONFIG_REISERFS_FS_POSIX_ACL
899 900
		{"acl",.setmask = 1 << REISERFS_POSIXACL},
		{"noacl",.clrmask = 1 << REISERFS_POSIXACL},
L
Linus Torvalds 已提交
901
#else
902 903
		{"acl",.setmask = 1 << REISERFS_UNSUPPORTED_OPT},
		{"noacl",.clrmask = 1 << REISERFS_UNSUPPORTED_OPT},
L
Linus Torvalds 已提交
904
#endif
905
		{.option_name = "nolog"},
906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922
		{"replayonly",.setmask = 1 << REPLAYONLY},
		{"block-allocator",.arg_required = 'a',.values = balloc},
		{"data",.arg_required = 'd',.values = logging_mode},
		{"barrier",.arg_required = 'b',.values = barrier_mode},
		{"resize",.arg_required = 'r',.values = NULL},
		{"jdev",.arg_required = 'j',.values = NULL},
		{"nolargeio",.arg_required = 'w',.values = NULL},
		{"commit",.arg_required = 'c',.values = NULL},
		{"usrquota",.setmask = 1 << REISERFS_QUOTA},
		{"grpquota",.setmask = 1 << REISERFS_QUOTA},
		{"noquota",.clrmask = 1 << REISERFS_QUOTA},
		{"errors",.arg_required = 'e',.values = error_actions},
		{"usrjquota",.arg_required =
		 'u' | (1 << REISERFS_OPT_ALLOWEMPTY),.values = NULL},
		{"grpjquota",.arg_required =
		 'g' | (1 << REISERFS_OPT_ALLOWEMPTY),.values = NULL},
		{"jqfmt",.arg_required = 'f',.values = NULL},
923
		{.option_name = NULL}
924 925 926 927 928 929 930 931 932 933 934 935
	};

	*blocks = 0;
	if (!options || !*options)
		/* use default configuration: create tails, journaling on, no
		   conversion to newest format */
		return 1;

	for (pos = options; pos;) {
		c = reiserfs_getopt(s, &pos, opts, &arg, mount_options);
		if (c == -1)
			/* wrong option is given */
936
			return 0;
L
Linus Torvalds 已提交
937

938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958
		if (c == 'r') {
			char *p;

			p = NULL;
			/* "resize=NNN" or "resize=auto" */

			if (!strcmp(arg, "auto")) {
				/* From JFS code, to auto-get the size. */
				*blocks =
				    s->s_bdev->bd_inode->i_size >> s->
				    s_blocksize_bits;
			} else {
				*blocks = simple_strtoul(arg, &p, 0);
				if (*p != '\0') {
					/* NNN does not look like a number */
					reiserfs_warning(s,
							 "reiserfs_parse_options: bad value %s",
							 arg);
					return 0;
				}
			}
L
Linus Torvalds 已提交
959 960
		}

961 962 963 964 965 966 967 968 969 970 971
		if (c == 'c') {
			char *p = NULL;
			unsigned long val = simple_strtoul(arg, &p, 0);
			/* commit=NNN (time in seconds) */
			if (*p != '\0' || val >= (unsigned int)-1) {
				reiserfs_warning(s,
						 "reiserfs_parse_options: bad value %s",
						 arg);
				return 0;
			}
			*commit_max_age = (unsigned int)val;
L
Linus Torvalds 已提交
972 973
		}

974
		if (c == 'w') {
975 976
			reiserfs_warning(s, "reiserfs: nolargeio option is no longer supported");
			return 0;
L
Linus Torvalds 已提交
977 978
		}

979 980 981 982 983 984 985 986 987 988
		if (c == 'j') {
			if (arg && *arg && jdev_name) {
				if (*jdev_name) {	//Hm, already assigned?
					reiserfs_warning(s,
							 "reiserfs_parse_options: journal device was already  specified to be %s",
							 *jdev_name);
					return 0;
				}
				*jdev_name = arg;
			}
L
Linus Torvalds 已提交
989
		}
990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
#ifdef CONFIG_QUOTA
		if (c == 'u' || c == 'g') {
			int qtype = c == 'u' ? USRQUOTA : GRPQUOTA;

			if (sb_any_quota_enabled(s)) {
				reiserfs_warning(s,
						 "reiserfs_parse_options: cannot change journalled quota options when quota turned on.");
				return 0;
			}
			if (*arg) {	/* Some filename specified? */
				if (REISERFS_SB(s)->s_qf_names[qtype]
				    && strcmp(REISERFS_SB(s)->s_qf_names[qtype],
					      arg)) {
					reiserfs_warning(s,
							 "reiserfs_parse_options: %s quota file already specified.",
							 QTYPE2NAME(qtype));
					return 0;
				}
				if (strchr(arg, '/')) {
					reiserfs_warning(s,
							 "reiserfs_parse_options: quotafile must be on filesystem root.");
					return 0;
				}
				REISERFS_SB(s)->s_qf_names[qtype] =
				    kmalloc(strlen(arg) + 1, GFP_KERNEL);
				if (!REISERFS_SB(s)->s_qf_names[qtype]) {
					reiserfs_warning(s,
							 "reiserfs_parse_options: not enough memory for storing quotafile name.");
					return 0;
				}
				strcpy(REISERFS_SB(s)->s_qf_names[qtype], arg);
				*mount_options |= 1 << REISERFS_QUOTA;
			} else {
1023 1024
				kfree(REISERFS_SB(s)->s_qf_names[qtype]);
				REISERFS_SB(s)->s_qf_names[qtype] = NULL;
1025
			}
L
Linus Torvalds 已提交
1026
		}
1027 1028 1029 1030 1031 1032 1033 1034 1035 1036
		if (c == 'f') {
			if (!strcmp(arg, "vfsold"))
				REISERFS_SB(s)->s_jquota_fmt = QFMT_VFS_OLD;
			else if (!strcmp(arg, "vfsv0"))
				REISERFS_SB(s)->s_jquota_fmt = QFMT_VFS_V0;
			else {
				reiserfs_warning(s,
						 "reiserfs_parse_options: unknown quota format specified.");
				return 0;
			}
L
Linus Torvalds 已提交
1037
		}
1038 1039 1040 1041 1042
#else
		if (c == 'u' || c == 'g' || c == 'f') {
			reiserfs_warning(s,
					 "reiserfs_parse_options: journalled quota options not supported.");
			return 0;
L
Linus Torvalds 已提交
1043
		}
1044 1045 1046 1047 1048 1049 1050 1051 1052
#endif
	}

#ifdef CONFIG_QUOTA
	if (!REISERFS_SB(s)->s_jquota_fmt
	    && (REISERFS_SB(s)->s_qf_names[USRQUOTA]
		|| REISERFS_SB(s)->s_qf_names[GRPQUOTA])) {
		reiserfs_warning(s,
				 "reiserfs_parse_options: journalled quota format not specified.");
L
Linus Torvalds 已提交
1053 1054
		return 0;
	}
1055 1056 1057 1058 1059 1060
	/* This checking is not precise wrt the quota type but for our purposes it is sufficient */
	if (!(*mount_options & (1 << REISERFS_QUOTA))
	    && sb_any_quota_enabled(s)) {
		reiserfs_warning(s,
				 "reiserfs_parse_options: quota options must be present when quota is turned on.");
		return 0;
L
Linus Torvalds 已提交
1061 1062
	}
#endif
1063

1064
	return 1;
L
Linus Torvalds 已提交
1065 1066
}

1067 1068 1069 1070 1071 1072
static void switch_data_mode(struct super_block *s, unsigned long mode)
{
	REISERFS_SB(s)->s_mount_opt &= ~((1 << REISERFS_DATA_LOG) |
					 (1 << REISERFS_DATA_ORDERED) |
					 (1 << REISERFS_DATA_WRITEBACK));
	REISERFS_SB(s)->s_mount_opt |= (1 << mode);
L
Linus Torvalds 已提交
1073 1074 1075 1076
}

static void handle_data_mode(struct super_block *s, unsigned long mount_options)
{
1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092
	if (mount_options & (1 << REISERFS_DATA_LOG)) {
		if (!reiserfs_data_log(s)) {
			switch_data_mode(s, REISERFS_DATA_LOG);
			reiserfs_info(s, "switching to journaled data mode\n");
		}
	} else if (mount_options & (1 << REISERFS_DATA_ORDERED)) {
		if (!reiserfs_data_ordered(s)) {
			switch_data_mode(s, REISERFS_DATA_ORDERED);
			reiserfs_info(s, "switching to ordered data mode\n");
		}
	} else if (mount_options & (1 << REISERFS_DATA_WRITEBACK)) {
		if (!reiserfs_data_writeback(s)) {
			switch_data_mode(s, REISERFS_DATA_WRITEBACK);
			reiserfs_info(s, "switching to writeback data mode\n");
		}
	}
L
Linus Torvalds 已提交
1093 1094
}

1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110
static void handle_barrier_mode(struct super_block *s, unsigned long bits)
{
	int flush = (1 << REISERFS_BARRIER_FLUSH);
	int none = (1 << REISERFS_BARRIER_NONE);
	int all_barrier = flush | none;

	if (bits & all_barrier) {
		REISERFS_SB(s)->s_mount_opt &= ~all_barrier;
		if (bits & flush) {
			REISERFS_SB(s)->s_mount_opt |= flush;
			printk("reiserfs: enabling write barrier flush mode\n");
		} else if (bits & none) {
			REISERFS_SB(s)->s_mount_opt |= none;
			printk("reiserfs: write barriers turned off\n");
		}
	}
L
Linus Torvalds 已提交
1111 1112
}

1113
static void handle_attrs(struct super_block *s)
L
Linus Torvalds 已提交
1114
{
1115
	struct reiserfs_super_block *rs = SB_DISK_SUPER_BLOCK(s);
L
Linus Torvalds 已提交
1116

1117 1118 1119 1120 1121
	if (reiserfs_attrs(s)) {
		if (old_format_only(s)) {
			reiserfs_warning(s,
					 "reiserfs: cannot support attributes on 3.5.x disk format");
			REISERFS_SB(s)->s_mount_opt &= ~(1 << REISERFS_ATTRS);
L
Linus Torvalds 已提交
1122 1123
			return;
		}
1124 1125 1126 1127
		if (!(le32_to_cpu(rs->s_flags) & reiserfs_attrs_cleared)) {
			reiserfs_warning(s,
					 "reiserfs: cannot support attributes until flag is set in super-block");
			REISERFS_SB(s)->s_mount_opt &= ~(1 << REISERFS_ATTRS);
L
Linus Torvalds 已提交
1128 1129 1130 1131
		}
	}
}

1132
static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
L
Linus Torvalds 已提交
1133
{
1134 1135 1136 1137 1138 1139 1140 1141
	struct reiserfs_super_block *rs;
	struct reiserfs_transaction_handle th;
	unsigned long blocks;
	unsigned long mount_options = REISERFS_SB(s)->s_mount_opt;
	unsigned long safe_mask = 0;
	unsigned int commit_max_age = (unsigned int)-1;
	struct reiserfs_journal *journal = SB_JOURNAL(s);
	int err;
L
Linus Torvalds 已提交
1142
#ifdef CONFIG_QUOTA
1143
	int i;
L
Linus Torvalds 已提交
1144 1145
#endif

1146
	rs = SB_DISK_SUPER_BLOCK(s);
L
Linus Torvalds 已提交
1147

1148 1149
	if (!reiserfs_parse_options
	    (s, arg, &mount_options, &blocks, NULL, &commit_max_age)) {
L
Linus Torvalds 已提交
1150
#ifdef CONFIG_QUOTA
1151 1152 1153 1154
		for (i = 0; i < MAXQUOTAS; i++) {
			kfree(REISERFS_SB(s)->s_qf_names[i]);
			REISERFS_SB(s)->s_qf_names[i] = NULL;
		}
L
Linus Torvalds 已提交
1155
#endif
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258
		return -EINVAL;
	}

	handle_attrs(s);

	/* Add options that are safe here */
	safe_mask |= 1 << REISERFS_SMALLTAIL;
	safe_mask |= 1 << REISERFS_LARGETAIL;
	safe_mask |= 1 << REISERFS_NO_BORDER;
	safe_mask |= 1 << REISERFS_NO_UNHASHED_RELOCATION;
	safe_mask |= 1 << REISERFS_HASHED_RELOCATION;
	safe_mask |= 1 << REISERFS_TEST4;
	safe_mask |= 1 << REISERFS_ATTRS;
	safe_mask |= 1 << REISERFS_XATTRS_USER;
	safe_mask |= 1 << REISERFS_POSIXACL;
	safe_mask |= 1 << REISERFS_BARRIER_FLUSH;
	safe_mask |= 1 << REISERFS_BARRIER_NONE;
	safe_mask |= 1 << REISERFS_ERROR_RO;
	safe_mask |= 1 << REISERFS_ERROR_CONTINUE;
	safe_mask |= 1 << REISERFS_ERROR_PANIC;
	safe_mask |= 1 << REISERFS_QUOTA;

	/* Update the bitmask, taking care to keep
	 * the bits we're not allowed to change here */
	REISERFS_SB(s)->s_mount_opt =
	    (REISERFS_SB(s)->
	     s_mount_opt & ~safe_mask) | (mount_options & safe_mask);

	if (commit_max_age != 0 && commit_max_age != (unsigned int)-1) {
		journal->j_max_commit_age = commit_max_age;
		journal->j_max_trans_age = commit_max_age;
	} else if (commit_max_age == 0) {
		/* 0 means restore defaults. */
		journal->j_max_commit_age = journal->j_default_max_commit_age;
		journal->j_max_trans_age = JOURNAL_MAX_TRANS_AGE;
	}

	if (blocks) {
		int rc = reiserfs_resize(s, blocks);
		if (rc != 0)
			return rc;
	}

	if (*mount_flags & MS_RDONLY) {
		reiserfs_xattr_init(s, *mount_flags);
		/* remount read-only */
		if (s->s_flags & MS_RDONLY)
			/* it is read-only already */
			return 0;
		/* try to remount file system with read-only permissions */
		if (sb_umount_state(rs) == REISERFS_VALID_FS
		    || REISERFS_SB(s)->s_mount_state != REISERFS_VALID_FS) {
			return 0;
		}

		err = journal_begin(&th, s, 10);
		if (err)
			return err;

		/* Mounting a rw partition read-only. */
		reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1);
		set_sb_umount_state(rs, REISERFS_SB(s)->s_mount_state);
		journal_mark_dirty(&th, s, SB_BUFFER_WITH_SB(s));
	} else {
		/* remount read-write */
		if (!(s->s_flags & MS_RDONLY)) {
			reiserfs_xattr_init(s, *mount_flags);
			return 0;	/* We are read-write already */
		}

		if (reiserfs_is_journal_aborted(journal))
			return journal->j_errno;

		handle_data_mode(s, mount_options);
		handle_barrier_mode(s, mount_options);
		REISERFS_SB(s)->s_mount_state = sb_umount_state(rs);
		s->s_flags &= ~MS_RDONLY;	/* now it is safe to call journal_begin */
		err = journal_begin(&th, s, 10);
		if (err)
			return err;

		/* Mount a partition which is read-only, read-write */
		reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1);
		REISERFS_SB(s)->s_mount_state = sb_umount_state(rs);
		s->s_flags &= ~MS_RDONLY;
		set_sb_umount_state(rs, REISERFS_ERROR_FS);
		/* mark_buffer_dirty (SB_BUFFER_WITH_SB (s), 1); */
		journal_mark_dirty(&th, s, SB_BUFFER_WITH_SB(s));
		REISERFS_SB(s)->s_mount_state = REISERFS_VALID_FS;
	}
	/* this will force a full flush of all journal lists */
	SB_JOURNAL(s)->j_must_wait = 1;
	err = journal_end(&th, s, 10);
	if (err)
		return err;
	s->s_dirt = 0;

	if (!(*mount_flags & MS_RDONLY)) {
		finish_unfinished(s);
		reiserfs_xattr_init(s, *mount_flags);
	}

	return 0;
L
Linus Torvalds 已提交
1259 1260
}

1261
static int read_super_block(struct super_block *s, int offset)
L
Linus Torvalds 已提交
1262
{
1263 1264 1265
	struct buffer_head *bh;
	struct reiserfs_super_block *rs;
	int fs_blocksize;
L
Linus Torvalds 已提交
1266

1267 1268 1269 1270 1271 1272 1273 1274
	bh = sb_bread(s, offset / s->s_blocksize);
	if (!bh) {
		reiserfs_warning(s, "sh-2006: read_super_block: "
				 "bread failed (dev %s, block %lu, size %lu)",
				 reiserfs_bdevname(s), offset / s->s_blocksize,
				 s->s_blocksize);
		return 1;
	}
L
Linus Torvalds 已提交
1275

1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
	rs = (struct reiserfs_super_block *)bh->b_data;
	if (!is_any_reiserfs_magic_string(rs)) {
		brelse(bh);
		return 1;
	}
	//
	// ok, reiserfs signature (old or new) found in at the given offset
	//    
	fs_blocksize = sb_blocksize(rs);
	brelse(bh);
	sb_set_blocksize(s, fs_blocksize);
L
Linus Torvalds 已提交
1287

1288 1289 1290 1291 1292 1293 1294 1295
	bh = sb_bread(s, offset / s->s_blocksize);
	if (!bh) {
		reiserfs_warning(s, "sh-2007: read_super_block: "
				 "bread failed (dev %s, block %lu, size %lu)\n",
				 reiserfs_bdevname(s), offset / s->s_blocksize,
				 s->s_blocksize);
		return 1;
	}
L
Linus Torvalds 已提交
1296

1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
	rs = (struct reiserfs_super_block *)bh->b_data;
	if (sb_blocksize(rs) != s->s_blocksize) {
		reiserfs_warning(s, "sh-2011: read_super_block: "
				 "can't find a reiserfs filesystem on (dev %s, block %Lu, size %lu)\n",
				 reiserfs_bdevname(s),
				 (unsigned long long)bh->b_blocknr,
				 s->s_blocksize);
		brelse(bh);
		return 1;
	}
L
Linus Torvalds 已提交
1307

1308 1309 1310 1311 1312 1313 1314
	if (rs->s_v1.s_root_block == cpu_to_le32(-1)) {
		brelse(bh);
		reiserfs_warning(s,
				 "Unfinished reiserfsck --rebuild-tree run detected. Please run\n"
				 "reiserfsck --rebuild-tree and wait for a completion. If that fails\n"
				 "get newer reiserfsprogs package");
		return 1;
L
Linus Torvalds 已提交
1315 1316
	}

1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346
	SB_BUFFER_WITH_SB(s) = bh;
	SB_DISK_SUPER_BLOCK(s) = rs;

	if (is_reiserfs_jr(rs)) {
		/* magic is of non-standard journal filesystem, look at s_version to
		   find which format is in use */
		if (sb_version(rs) == REISERFS_VERSION_2)
			reiserfs_warning(s,
					 "read_super_block: found reiserfs format \"3.6\""
					 " with non-standard journal");
		else if (sb_version(rs) == REISERFS_VERSION_1)
			reiserfs_warning(s,
					 "read_super_block: found reiserfs format \"3.5\""
					 " with non-standard journal");
		else {
			reiserfs_warning(s,
					 "sh-2012: read_super_block: found unknown "
					 "format \"%u\" of reiserfs with non-standard magic",
					 sb_version(rs));
			return 1;
		}
	} else
		/* s_version of standard format may contain incorrect information,
		   so we just look at the magic string */
		reiserfs_info(s,
			      "found reiserfs format \"%s\" with standard journal\n",
			      is_reiserfs_3_5(rs) ? "3.5" : "3.6");

	s->s_op = &reiserfs_sops;
	s->s_export_op = &reiserfs_export_ops;
L
Linus Torvalds 已提交
1347
#ifdef CONFIG_QUOTA
1348 1349
	s->s_qcop = &reiserfs_qctl_operations;
	s->dq_op = &reiserfs_quota_operations;
L
Linus Torvalds 已提交
1350 1351
#endif

1352 1353 1354 1355 1356
	/* new format is limited by the 32 bit wide i_blocks field, want to
	 ** be one full block below that.
	 */
	s->s_maxbytes = (512LL << 32) - s->s_blocksize;
	return 0;
L
Linus Torvalds 已提交
1357 1358 1359
}

/* after journal replay, reread all bitmap and super blocks */
1360 1361 1362 1363 1364 1365 1366 1367 1368
static int reread_meta_blocks(struct super_block *s)
{
	ll_rw_block(READ, 1, &(SB_BUFFER_WITH_SB(s)));
	wait_on_buffer(SB_BUFFER_WITH_SB(s));
	if (!buffer_uptodate(SB_BUFFER_WITH_SB(s))) {
		reiserfs_warning(s,
				 "reread_meta_blocks, error reading the super");
		return 1;
	}
L
Linus Torvalds 已提交
1369

1370 1371
	return 0;
}
L
Linus Torvalds 已提交
1372 1373 1374 1375 1376 1377 1378 1379 1380

/////////////////////////////////////////////////////
// hash detection stuff

// if root directory is empty - we set default - Yura's - hash and
// warn about it
// FIXME: we look for only one name in a directory. If tea and yura
// bith have the same value - we ask user to send report to the
// mailing list
1381
static __u32 find_hash_out(struct super_block *s)
L
Linus Torvalds 已提交
1382
{
1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453
	int retval;
	struct inode *inode;
	struct cpu_key key;
	INITIALIZE_PATH(path);
	struct reiserfs_dir_entry de;
	__u32 hash = DEFAULT_HASH;

	inode = s->s_root->d_inode;

	do {			// Some serious "goto"-hater was there ;)
		u32 teahash, r5hash, yurahash;

		make_cpu_key(&key, inode, ~0, TYPE_DIRENTRY, 3);
		retval = search_by_entry_key(s, &key, &path, &de);
		if (retval == IO_ERROR) {
			pathrelse(&path);
			return UNSET_HASH;
		}
		if (retval == NAME_NOT_FOUND)
			de.de_entry_num--;
		set_de_name_and_namelen(&de);
		if (deh_offset(&(de.de_deh[de.de_entry_num])) == DOT_DOT_OFFSET) {
			/* allow override in this case */
			if (reiserfs_rupasov_hash(s)) {
				hash = YURA_HASH;
			}
			reiserfs_warning(s, "FS seems to be empty, autodetect "
					 "is using the default hash");
			break;
		}
		r5hash = GET_HASH_VALUE(r5_hash(de.de_name, de.de_namelen));
		teahash = GET_HASH_VALUE(keyed_hash(de.de_name, de.de_namelen));
		yurahash = GET_HASH_VALUE(yura_hash(de.de_name, de.de_namelen));
		if (((teahash == r5hash)
		     &&
		     (GET_HASH_VALUE(deh_offset(&(de.de_deh[de.de_entry_num])))
		      == r5hash)) || ((teahash == yurahash)
				      && (yurahash ==
					  GET_HASH_VALUE(deh_offset
							 (&
							  (de.
							   de_deh[de.
								  de_entry_num])))))
		    || ((r5hash == yurahash)
			&& (yurahash ==
			    GET_HASH_VALUE(deh_offset
					   (&(de.de_deh[de.de_entry_num])))))) {
			reiserfs_warning(s,
					 "Unable to automatically detect hash function. "
					 "Please mount with -o hash={tea,rupasov,r5}",
					 reiserfs_bdevname(s));
			hash = UNSET_HASH;
			break;
		}
		if (GET_HASH_VALUE(deh_offset(&(de.de_deh[de.de_entry_num]))) ==
		    yurahash)
			hash = YURA_HASH;
		else if (GET_HASH_VALUE
			 (deh_offset(&(de.de_deh[de.de_entry_num]))) == teahash)
			hash = TEA_HASH;
		else if (GET_HASH_VALUE
			 (deh_offset(&(de.de_deh[de.de_entry_num]))) == r5hash)
			hash = R5_HASH;
		else {
			reiserfs_warning(s, "Unrecognised hash function");
			hash = UNSET_HASH;
		}
	} while (0);

	pathrelse(&path);
	return hash;
L
Linus Torvalds 已提交
1454 1455 1456
}

// finds out which hash names are sorted with
1457
static int what_hash(struct super_block *s)
L
Linus Torvalds 已提交
1458
{
1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509
	__u32 code;

	code = sb_hash_function_code(SB_DISK_SUPER_BLOCK(s));

	/* reiserfs_hash_detect() == true if any of the hash mount options
	 ** were used.  We must check them to make sure the user isn't
	 ** using a bad hash value
	 */
	if (code == UNSET_HASH || reiserfs_hash_detect(s))
		code = find_hash_out(s);

	if (code != UNSET_HASH && reiserfs_hash_detect(s)) {
		/* detection has found the hash, and we must check against the 
		 ** mount options 
		 */
		if (reiserfs_rupasov_hash(s) && code != YURA_HASH) {
			reiserfs_warning(s, "Error, %s hash detected, "
					 "unable to force rupasov hash",
					 reiserfs_hashname(code));
			code = UNSET_HASH;
		} else if (reiserfs_tea_hash(s) && code != TEA_HASH) {
			reiserfs_warning(s, "Error, %s hash detected, "
					 "unable to force tea hash",
					 reiserfs_hashname(code));
			code = UNSET_HASH;
		} else if (reiserfs_r5_hash(s) && code != R5_HASH) {
			reiserfs_warning(s, "Error, %s hash detected, "
					 "unable to force r5 hash",
					 reiserfs_hashname(code));
			code = UNSET_HASH;
		}
	} else {
		/* find_hash_out was not called or could not determine the hash */
		if (reiserfs_rupasov_hash(s)) {
			code = YURA_HASH;
		} else if (reiserfs_tea_hash(s)) {
			code = TEA_HASH;
		} else if (reiserfs_r5_hash(s)) {
			code = R5_HASH;
		}
	}

	/* if we are mounted RW, and we have a new valid hash code, update 
	 ** the super
	 */
	if (code != UNSET_HASH &&
	    !(s->s_flags & MS_RDONLY) &&
	    code != sb_hash_function_code(SB_DISK_SUPER_BLOCK(s))) {
		set_sb_hash_function_code(SB_DISK_SUPER_BLOCK(s), code);
	}
	return code;
L
Linus Torvalds 已提交
1510 1511 1512
}

// return pointer to appropriate function
1513
static hashf_t hash_function(struct super_block *s)
L
Linus Torvalds 已提交
1514
{
1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526
	switch (what_hash(s)) {
	case TEA_HASH:
		reiserfs_info(s, "Using tea hash to sort names\n");
		return keyed_hash;
	case YURA_HASH:
		reiserfs_info(s, "Using rupasov hash to sort names\n");
		return yura_hash;
	case R5_HASH:
		reiserfs_info(s, "Using r5 hash to sort names\n");
		return r5_hash;
	}
	return NULL;
L
Linus Torvalds 已提交
1527 1528 1529
}

// this is used to set up correct value for old partitions
1530
static int function2code(hashf_t func)
L
Linus Torvalds 已提交
1531
{
1532 1533 1534 1535 1536 1537
	if (func == keyed_hash)
		return TEA_HASH;
	if (func == yura_hash)
		return YURA_HASH;
	if (func == r5_hash)
		return R5_HASH;
L
Linus Torvalds 已提交
1538

1539
	BUG();			// should never happen
L
Linus Torvalds 已提交
1540

1541
	return 0;
L
Linus Torvalds 已提交
1542 1543 1544 1545 1546 1547
}

#define SWARN(silent, s, ...)			\
	if (!(silent))				\
		reiserfs_warning (s, __VA_ARGS__)

1548
static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
L
Linus Torvalds 已提交
1549
{
1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561
	struct inode *root_inode;
	struct reiserfs_transaction_handle th;
	int old_format = 0;
	unsigned long blocks;
	unsigned int commit_max_age = 0;
	int jinit_done = 0;
	struct reiserfs_iget_args args;
	struct reiserfs_super_block *rs;
	char *jdev_name;
	struct reiserfs_sb_info *sbi;
	int errval = -EINVAL;

1562
	sbi = kzalloc(sizeof(struct reiserfs_sb_info), GFP_KERNEL);
1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575
	if (!sbi) {
		errval = -ENOMEM;
		goto error;
	}
	s->s_fs_info = sbi;
	/* Set default values for options: non-aggressive tails, RO on errors */
	REISERFS_SB(s)->s_mount_opt |= (1 << REISERFS_SMALLTAIL);
	REISERFS_SB(s)->s_mount_opt |= (1 << REISERFS_ERROR_RO);
	/* no preallocation minimum, be smart in
	   reiserfs_file_write instead */
	REISERFS_SB(s)->s_alloc_options.preallocmin = 0;
	/* Preallocate by 16 blocks (17-1) at once */
	REISERFS_SB(s)->s_alloc_options.preallocsize = 17;
1576
#ifdef CONFIG_REISERFS_FS_XATTR
1577 1578
	/* Initialize the rwsem for xattr dir */
	init_rwsem(&REISERFS_SB(s)->xattr_dir_sem);
1579
#endif
1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626
	/* setup default block allocator options */
	reiserfs_init_alloc_options(s);

	jdev_name = NULL;
	if (reiserfs_parse_options
	    (s, (char *)data, &(sbi->s_mount_opt), &blocks, &jdev_name,
	     &commit_max_age) == 0) {
		goto error;
	}

	if (blocks) {
		SWARN(silent, s, "jmacd-7: reiserfs_fill_super: resize option "
		      "for remount only");
		goto error;
	}

	/* try old format (undistributed bitmap, super block in 8-th 1k block of a device) */
	if (!read_super_block(s, REISERFS_OLD_DISK_OFFSET_IN_BYTES))
		old_format = 1;
	/* try new format (64-th 1k block), which can contain reiserfs super block */
	else if (read_super_block(s, REISERFS_DISK_OFFSET_IN_BYTES)) {
		SWARN(silent, s,
		      "sh-2021: reiserfs_fill_super: can not find reiserfs on %s",
		      reiserfs_bdevname(s));
		goto error;
	}

	rs = SB_DISK_SUPER_BLOCK(s);
	/* Let's do basic sanity check to verify that underlying device is not
	   smaller than the filesystem. If the check fails then abort and scream,
	   because bad stuff will happen otherwise. */
	if (s->s_bdev && s->s_bdev->bd_inode
	    && i_size_read(s->s_bdev->bd_inode) <
	    sb_block_count(rs) * sb_blocksize(rs)) {
		SWARN(silent, s,
		      "Filesystem on %s cannot be mounted because it is bigger than the device",
		      reiserfs_bdevname(s));
		SWARN(silent, s,
		      "You may need to run fsck or increase size of your LVM partition");
		SWARN(silent, s,
		      "Or may be you forgot to reboot after fdisk when it told you to");
		goto error;
	}

	sbi->s_mount_state = SB_REISERFS_STATE(s);
	sbi->s_mount_state = REISERFS_VALID_FS;

1627
	if ((errval = reiserfs_init_bitmap_cache(s))) {
1628 1629 1630 1631
		SWARN(silent, s,
		      "jmacd-8: reiserfs_fill_super: unable to read bitmap");
		goto error;
	}
1632
	errval = -EINVAL;
L
Linus Torvalds 已提交
1633
#ifdef CONFIG_REISERFS_CHECK
1634 1635
	SWARN(silent, s, "CONFIG_REISERFS_CHECK is set ON");
	SWARN(silent, s, "- it is slow mode for debugging.");
L
Linus Torvalds 已提交
1636 1637
#endif

1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709
	/* make data=ordered the default */
	if (!reiserfs_data_log(s) && !reiserfs_data_ordered(s) &&
	    !reiserfs_data_writeback(s)) {
		REISERFS_SB(s)->s_mount_opt |= (1 << REISERFS_DATA_ORDERED);
	}

	if (reiserfs_data_log(s)) {
		reiserfs_info(s, "using journaled data mode\n");
	} else if (reiserfs_data_ordered(s)) {
		reiserfs_info(s, "using ordered data mode\n");
	} else {
		reiserfs_info(s, "using writeback data mode\n");
	}
	if (reiserfs_barrier_flush(s)) {
		printk("reiserfs: using flush barriers\n");
	}
	// set_device_ro(s->s_dev, 1) ;
	if (journal_init(s, jdev_name, old_format, commit_max_age)) {
		SWARN(silent, s,
		      "sh-2022: reiserfs_fill_super: unable to initialize journal space");
		goto error;
	} else {
		jinit_done = 1;	/* once this is set, journal_release must be called
				 ** if we error out of the mount
				 */
	}
	if (reread_meta_blocks(s)) {
		SWARN(silent, s,
		      "jmacd-9: reiserfs_fill_super: unable to reread meta blocks after journal init");
		goto error;
	}

	if (replay_only(s))
		goto error;

	if (bdev_read_only(s->s_bdev) && !(s->s_flags & MS_RDONLY)) {
		SWARN(silent, s,
		      "clm-7000: Detected readonly device, marking FS readonly");
		s->s_flags |= MS_RDONLY;
	}
	args.objectid = REISERFS_ROOT_OBJECTID;
	args.dirid = REISERFS_ROOT_PARENT_OBJECTID;
	root_inode =
	    iget5_locked(s, REISERFS_ROOT_OBJECTID, reiserfs_find_actor,
			 reiserfs_init_locked_inode, (void *)(&args));
	if (!root_inode) {
		SWARN(silent, s,
		      "jmacd-10: reiserfs_fill_super: get root inode failed");
		goto error;
	}

	if (root_inode->i_state & I_NEW) {
		reiserfs_read_locked_inode(root_inode, &args);
		unlock_new_inode(root_inode);
	}

	s->s_root = d_alloc_root(root_inode);
	if (!s->s_root) {
		iput(root_inode);
		goto error;
	}
	// define and initialize hash function
	sbi->s_hash_function = hash_function(s);
	if (sbi->s_hash_function == NULL) {
		dput(s->s_root);
		s->s_root = NULL;
		goto error;
	}

	if (is_reiserfs_3_5(rs)
	    || (is_reiserfs_jr(rs) && SB_VERSION(s) == REISERFS_VERSION_1))
		set_bit(REISERFS_3_5, &(sbi->s_properties));
1710 1711
	else if (old_format)
		set_bit(REISERFS_OLD_FORMAT, &(sbi->s_properties));
1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727
	else
		set_bit(REISERFS_3_6, &(sbi->s_properties));

	if (!(s->s_flags & MS_RDONLY)) {

		errval = journal_begin(&th, s, 1);
		if (errval) {
			dput(s->s_root);
			s->s_root = NULL;
			goto error;
		}
		reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1);

		set_sb_umount_state(rs, REISERFS_ERROR_FS);
		set_sb_fs_state(rs, 0);

1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742
		/* Clear out s_bmap_nr if it would wrap. We can handle this
		 * case, but older revisions can't. This will cause the
		 * file system to fail mount on those older implementations,
		 * avoiding corruption. -jeffm */
		if (bmap_would_wrap(reiserfs_bmap_count(s)) &&
		    sb_bmap_nr(rs) != 0) {
			reiserfs_warning(s, "super-2030: This file system "
					"claims to use %u bitmap blocks in "
					"its super block, but requires %u. "
					"Clearing to zero.", sb_bmap_nr(rs),
					reiserfs_bmap_count(s));

			set_sb_bmap_nr(rs, 0);
		}

1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811
		if (old_format_only(s)) {
			/* filesystem of format 3.5 either with standard or non-standard
			   journal */
			if (convert_reiserfs(s)) {
				/* and -o conv is given */
				if (!silent)
					reiserfs_info(s,
						      "converting 3.5 filesystem to the 3.6 format");

				if (is_reiserfs_3_5(rs))
					/* put magic string of 3.6 format. 2.2 will not be able to
					   mount this filesystem anymore */
					memcpy(rs->s_v1.s_magic,
					       reiserfs_3_6_magic_string,
					       sizeof
					       (reiserfs_3_6_magic_string));

				set_sb_version(rs, REISERFS_VERSION_2);
				reiserfs_convert_objectid_map_v1(s);
				set_bit(REISERFS_3_6, &(sbi->s_properties));
				clear_bit(REISERFS_3_5, &(sbi->s_properties));
			} else if (!silent) {
				reiserfs_info(s, "using 3.5.x disk format\n");
			}
		}

		journal_mark_dirty(&th, s, SB_BUFFER_WITH_SB(s));
		errval = journal_end(&th, s, 1);
		if (errval) {
			dput(s->s_root);
			s->s_root = NULL;
			goto error;
		}

		if ((errval = reiserfs_xattr_init(s, s->s_flags))) {
			dput(s->s_root);
			s->s_root = NULL;
			goto error;
		}

		/* look for files which were to be removed in previous session */
		finish_unfinished(s);
	} else {
		if (old_format_only(s) && !silent) {
			reiserfs_info(s, "using 3.5.x disk format\n");
		}

		if ((errval = reiserfs_xattr_init(s, s->s_flags))) {
			dput(s->s_root);
			s->s_root = NULL;
			goto error;
		}
	}
	// mark hash in super block: it could be unset. overwrite should be ok
	set_sb_hash_function_code(rs, function2code(sbi->s_hash_function));

	handle_attrs(s);

	reiserfs_proc_info_init(s);

	init_waitqueue_head(&(sbi->s_wait));
	spin_lock_init(&sbi->bitmap_lock);

	return (0);

      error:
	if (jinit_done) {	/* kill the commit thread, free journal ram */
		journal_release_error(NULL, s);
	}
1812 1813

	reiserfs_free_bitmap_cache(s);
1814 1815
	if (SB_BUFFER_WITH_SB(s))
		brelse(SB_BUFFER_WITH_SB(s));
L
Linus Torvalds 已提交
1816
#ifdef CONFIG_QUOTA
1817 1818 1819 1820 1821 1822
	{
		int j;
		for (j = 0; j < MAXQUOTAS; j++) {
			kfree(sbi->s_qf_names[j]);
			sbi->s_qf_names[j] = NULL;
		}
1823
	}
L
Linus Torvalds 已提交
1824
#endif
1825
	kfree(sbi);
L
Linus Torvalds 已提交
1826

1827 1828
	s->s_fs_info = NULL;
	return errval;
L
Linus Torvalds 已提交
1829 1830
}

1831
static int reiserfs_statfs(struct dentry *dentry, struct kstatfs *buf)
L
Linus Torvalds 已提交
1832
{
1833
	struct reiserfs_super_block *rs = SB_DISK_SUPER_BLOCK(dentry->d_sb);
1834 1835 1836 1837 1838

	buf->f_namelen = (REISERFS_MAX_NAME(s->s_blocksize));
	buf->f_bfree = sb_free_blocks(rs);
	buf->f_bavail = buf->f_bfree;
	buf->f_blocks = sb_block_count(rs) - sb_bmap_nr(rs) - 1;
1839
	buf->f_bsize = dentry->d_sb->s_blocksize;
1840 1841 1842
	/* changed to accommodate gcc folks. */
	buf->f_type = REISERFS_SUPER_MAGIC;
	return 0;
L
Linus Torvalds 已提交
1843 1844 1845 1846 1847
}

#ifdef CONFIG_QUOTA
static int reiserfs_dquot_initialize(struct inode *inode, int type)
{
1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866
	struct reiserfs_transaction_handle th;
	int ret, err;

	/* We may create quota structure so we need to reserve enough blocks */
	reiserfs_write_lock(inode->i_sb);
	ret =
	    journal_begin(&th, inode->i_sb,
			  2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb));
	if (ret)
		goto out;
	ret = dquot_initialize(inode, type);
	err =
	    journal_end(&th, inode->i_sb,
			2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb));
	if (!ret && err)
		ret = err;
      out:
	reiserfs_write_unlock(inode->i_sb);
	return ret;
L
Linus Torvalds 已提交
1867 1868 1869 1870
}

static int reiserfs_dquot_drop(struct inode *inode)
{
1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889
	struct reiserfs_transaction_handle th;
	int ret, err;

	/* We may delete quota structure so we need to reserve enough blocks */
	reiserfs_write_lock(inode->i_sb);
	ret =
	    journal_begin(&th, inode->i_sb,
			  2 * REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb));
	if (ret)
		goto out;
	ret = dquot_drop(inode);
	err =
	    journal_end(&th, inode->i_sb,
			2 * REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb));
	if (!ret && err)
		ret = err;
      out:
	reiserfs_write_unlock(inode->i_sb);
	return ret;
L
Linus Torvalds 已提交
1890 1891 1892 1893
}

static int reiserfs_write_dquot(struct dquot *dquot)
{
1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911
	struct reiserfs_transaction_handle th;
	int ret, err;

	reiserfs_write_lock(dquot->dq_sb);
	ret =
	    journal_begin(&th, dquot->dq_sb,
			  REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
	if (ret)
		goto out;
	ret = dquot_commit(dquot);
	err =
	    journal_end(&th, dquot->dq_sb,
			REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
	if (!ret && err)
		ret = err;
      out:
	reiserfs_write_unlock(dquot->dq_sb);
	return ret;
L
Linus Torvalds 已提交
1912 1913 1914 1915
}

static int reiserfs_acquire_dquot(struct dquot *dquot)
{
1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933
	struct reiserfs_transaction_handle th;
	int ret, err;

	reiserfs_write_lock(dquot->dq_sb);
	ret =
	    journal_begin(&th, dquot->dq_sb,
			  REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb));
	if (ret)
		goto out;
	ret = dquot_acquire(dquot);
	err =
	    journal_end(&th, dquot->dq_sb,
			REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb));
	if (!ret && err)
		ret = err;
      out:
	reiserfs_write_unlock(dquot->dq_sb);
	return ret;
L
Linus Torvalds 已提交
1934 1935 1936 1937
}

static int reiserfs_release_dquot(struct dquot *dquot)
{
1938 1939 1940 1941 1942 1943 1944
	struct reiserfs_transaction_handle th;
	int ret, err;

	reiserfs_write_lock(dquot->dq_sb);
	ret =
	    journal_begin(&th, dquot->dq_sb,
			  REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb));
J
Jan Kara 已提交
1945 1946 1947
	if (ret) {
		/* Release dquot anyway to avoid endless cycle in dqput() */
		dquot_release(dquot);
1948
		goto out;
J
Jan Kara 已提交
1949
	}
1950 1951 1952 1953 1954 1955 1956 1957 1958
	ret = dquot_release(dquot);
	err =
	    journal_end(&th, dquot->dq_sb,
			REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb));
	if (!ret && err)
		ret = err;
      out:
	reiserfs_write_unlock(dquot->dq_sb);
	return ret;
L
Linus Torvalds 已提交
1959 1960 1961 1962
}

static int reiserfs_mark_dquot_dirty(struct dquot *dquot)
{
1963 1964 1965 1966 1967 1968 1969
	/* Are we journalling quotas? */
	if (REISERFS_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] ||
	    REISERFS_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) {
		dquot_mark_dquot_dirty(dquot);
		return reiserfs_write_dquot(dquot);
	} else
		return dquot_mark_dquot_dirty(dquot);
L
Linus Torvalds 已提交
1970 1971 1972 1973
}

static int reiserfs_write_info(struct super_block *sb, int type)
{
1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
	struct reiserfs_transaction_handle th;
	int ret, err;

	/* Data block + inode block */
	reiserfs_write_lock(sb);
	ret = journal_begin(&th, sb, 2);
	if (ret)
		goto out;
	ret = dquot_commit_info(sb, type);
	err = journal_end(&th, sb, 2);
	if (!ret && err)
		ret = err;
      out:
	reiserfs_write_unlock(sb);
	return ret;
L
Linus Torvalds 已提交
1989 1990 1991
}

/*
1992
 * Turn on quotas during mount time - we need to find the quota file and such...
L
Linus Torvalds 已提交
1993 1994 1995
 */
static int reiserfs_quota_on_mount(struct super_block *sb, int type)
{
1996
	return vfs_quota_on_mount(sb, REISERFS_SB(sb)->s_qf_names[type],
1997
				  REISERFS_SB(sb)->s_jquota_fmt, type);
L
Linus Torvalds 已提交
1998 1999 2000 2001 2002
}

/*
 * Standard function to be called on quota_on
 */
2003 2004
static int reiserfs_quota_on(struct super_block *sb, int type, int format_id,
			     char *path)
L
Linus Torvalds 已提交
2005
{
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036
	int err;
	struct nameidata nd;

	if (!(REISERFS_SB(sb)->s_mount_opt & (1 << REISERFS_QUOTA)))
		return -EINVAL;
	err = path_lookup(path, LOOKUP_FOLLOW, &nd);
	if (err)
		return err;
	/* Quotafile not on the same filesystem? */
	if (nd.mnt->mnt_sb != sb) {
		path_release(&nd);
		return -EXDEV;
	}
	/* We must not pack tails for quota files on reiserfs for quota IO to work */
	if (!REISERFS_I(nd.dentry->d_inode)->i_flags & i_nopack_mask) {
		reiserfs_warning(sb,
				 "reiserfs: Quota file must have tail packing disabled.");
		path_release(&nd);
		return -EINVAL;
	}
	/* Not journalling quota? No more tests needed... */
	if (!REISERFS_SB(sb)->s_qf_names[USRQUOTA] &&
	    !REISERFS_SB(sb)->s_qf_names[GRPQUOTA]) {
		path_release(&nd);
		return vfs_quota_on(sb, type, format_id, path);
	}
	/* Quotafile not of fs root? */
	if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode)
		reiserfs_warning(sb,
				 "reiserfs: Quota file not on filesystem root. "
				 "Journalled quota will not work.");
L
Linus Torvalds 已提交
2037
	path_release(&nd);
2038
	return vfs_quota_on(sb, type, format_id, path);
L
Linus Torvalds 已提交
2039 2040 2041 2042 2043 2044 2045 2046 2047
}

/* Read data from quotafile - avoid pagecache and such because we cannot afford
 * acquiring the locks... As quota files are never truncated and quota code
 * itself serializes the operations (and noone else should touch the files)
 * we don't have to be afraid of races */
static ssize_t reiserfs_quota_read(struct super_block *sb, int type, char *data,
				   size_t len, loff_t off)
{
2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085
	struct inode *inode = sb_dqopt(sb)->files[type];
	unsigned long blk = off >> sb->s_blocksize_bits;
	int err = 0, offset = off & (sb->s_blocksize - 1), tocopy;
	size_t toread;
	struct buffer_head tmp_bh, *bh;
	loff_t i_size = i_size_read(inode);

	if (off > i_size)
		return 0;
	if (off + len > i_size)
		len = i_size - off;
	toread = len;
	while (toread > 0) {
		tocopy =
		    sb->s_blocksize - offset <
		    toread ? sb->s_blocksize - offset : toread;
		tmp_bh.b_state = 0;
		/* Quota files are without tails so we can safely use this function */
		reiserfs_write_lock(sb);
		err = reiserfs_get_block(inode, blk, &tmp_bh, 0);
		reiserfs_write_unlock(sb);
		if (err)
			return err;
		if (!buffer_mapped(&tmp_bh))	/* A hole? */
			memset(data, 0, tocopy);
		else {
			bh = sb_bread(sb, tmp_bh.b_blocknr);
			if (!bh)
				return -EIO;
			memcpy(data, bh->b_data + offset, tocopy);
			brelse(bh);
		}
		offset = 0;
		toread -= tocopy;
		data += tocopy;
		blk++;
	}
	return len;
L
Linus Torvalds 已提交
2086 2087 2088 2089 2090 2091 2092
}

/* Write to quotafile (we know the transaction is already started and has
 * enough credits) */
static ssize_t reiserfs_quota_write(struct super_block *sb, int type,
				    const char *data, size_t len, loff_t off)
{
2093 2094 2095 2096 2097 2098 2099
	struct inode *inode = sb_dqopt(sb)->files[type];
	unsigned long blk = off >> sb->s_blocksize_bits;
	int err = 0, offset = off & (sb->s_blocksize - 1), tocopy;
	int journal_quota = REISERFS_SB(sb)->s_qf_names[type] != NULL;
	size_t towrite = len;
	struct buffer_head tmp_bh, *bh;

J
Jan Kara 已提交
2100 2101 2102 2103 2104 2105
	if (!current->journal_info) {
		printk(KERN_WARNING "reiserfs: Quota write (off=%Lu, len=%Lu)"
			" cancelled because transaction is not started.\n",
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}
2106
	mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA);
2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136
	while (towrite > 0) {
		tocopy = sb->s_blocksize - offset < towrite ?
		    sb->s_blocksize - offset : towrite;
		tmp_bh.b_state = 0;
		err = reiserfs_get_block(inode, blk, &tmp_bh, GET_BLOCK_CREATE);
		if (err)
			goto out;
		if (offset || tocopy != sb->s_blocksize)
			bh = sb_bread(sb, tmp_bh.b_blocknr);
		else
			bh = sb_getblk(sb, tmp_bh.b_blocknr);
		if (!bh) {
			err = -EIO;
			goto out;
		}
		lock_buffer(bh);
		memcpy(bh->b_data + offset, data, tocopy);
		flush_dcache_page(bh->b_page);
		set_buffer_uptodate(bh);
		unlock_buffer(bh);
		reiserfs_prepare_for_journal(sb, bh, 1);
		journal_mark_dirty(current->journal_info, sb, bh);
		if (!journal_quota)
			reiserfs_add_ordered_list(inode, bh);
		brelse(bh);
		offset = 0;
		towrite -= tocopy;
		data += tocopy;
		blk++;
	}
J
Jan Kara 已提交
2137
out:
2138 2139 2140 2141 2142 2143 2144
	if (len == towrite)
		return err;
	if (inode->i_size < off + len - towrite)
		i_size_write(inode, off + len - towrite);
	inode->i_version++;
	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	mark_inode_dirty(inode);
2145
	mutex_unlock(&inode->i_mutex);
2146
	return len - towrite;
L
Linus Torvalds 已提交
2147 2148 2149 2150
}

#endif

2151 2152 2153
static int get_super_block(struct file_system_type *fs_type,
			   int flags, const char *dev_name,
			   void *data, struct vfsmount *mnt)
L
Linus Torvalds 已提交
2154
{
2155 2156
	return get_sb_bdev(fs_type, flags, dev_name, data, reiserfs_fill_super,
			   mnt);
L
Linus Torvalds 已提交
2157 2158
}

2159
static int __init init_reiserfs_fs(void)
L
Linus Torvalds 已提交
2160 2161 2162
{
	int ret;

2163
	if ((ret = init_inodecache())) {
L
Linus Torvalds 已提交
2164 2165 2166
		return ret;
	}

2167 2168
	if ((ret = reiserfs_xattr_register_handlers()))
		goto failed_reiserfs_xattr_register_handlers;
L
Linus Torvalds 已提交
2169

2170 2171 2172
	reiserfs_proc_info_global_init();
	reiserfs_proc_register_global("version",
				      reiserfs_global_version_in_proc);
L
Linus Torvalds 已提交
2173

2174
	ret = register_filesystem(&reiserfs_fs_type);
L
Linus Torvalds 已提交
2175 2176 2177 2178 2179

	if (ret == 0) {
		return 0;
	}

2180
	reiserfs_xattr_unregister_handlers();
L
Linus Torvalds 已提交
2181

2182 2183 2184 2185
      failed_reiserfs_xattr_register_handlers:
	reiserfs_proc_unregister_global("version");
	reiserfs_proc_info_global_done();
	destroy_inodecache();
L
Linus Torvalds 已提交
2186 2187 2188 2189

	return ret;
}

2190
static void __exit exit_reiserfs_fs(void)
L
Linus Torvalds 已提交
2191
{
2192 2193 2194 2195 2196
	reiserfs_xattr_unregister_handlers();
	reiserfs_proc_unregister_global("version");
	reiserfs_proc_info_global_done();
	unregister_filesystem(&reiserfs_fs_type);
	destroy_inodecache();
L
Linus Torvalds 已提交
2197 2198 2199
}

struct file_system_type reiserfs_fs_type = {
2200 2201 2202
	.owner = THIS_MODULE,
	.name = "reiserfs",
	.get_sb = get_super_block,
2203
	.kill_sb = reiserfs_kill_sb,
2204
	.fs_flags = FS_REQUIRES_DEV,
L
Linus Torvalds 已提交
2205 2206
};

2207 2208 2209
MODULE_DESCRIPTION("ReiserFS journaled filesystem");
MODULE_AUTHOR("Hans Reiser <reiser@namesys.com>");
MODULE_LICENSE("GPL");
L
Linus Torvalds 已提交
2210

2211 2212
module_init(init_reiserfs_fs);
module_exit(exit_reiserfs_fs);