ialloc.c 33.5 KB
Newer Older
1
/*
2
 *  linux/fs/ext4/ialloc.c
3 4 5 6 7 8 9 10 11 12 13 14 15 16
 *
 * Copyright (C) 1992, 1993, 1994, 1995
 * Remy Card (card@masi.ibp.fr)
 * Laboratoire MASI - Institut Blaise Pascal
 * Universite Pierre et Marie Curie (Paris VI)
 *
 *  BSD ufs-inspired inode and directory allocation by
 *  Stephen Tweedie (sct@redhat.com), 1993
 *  Big-endian to little-endian byte-swapping/bitmaps by
 *        David S. Miller (davem@caip.rutgers.edu), 1995
 */

#include <linux/time.h>
#include <linux/fs.h>
17
#include <linux/jbd2.h>
18 19 20 21 22 23
#include <linux/stat.h>
#include <linux/string.h>
#include <linux/quotaops.h>
#include <linux/buffer_head.h>
#include <linux/random.h>
#include <linux/bitops.h>
24
#include <linux/blkdev.h>
25
#include <asm/byteorder.h>
26 27
#include "ext4.h"
#include "ext4_jbd2.h"
28 29
#include "xattr.h"
#include "acl.h"
A
Andreas Dilger 已提交
30
#include "group.h"
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

/*
 * ialloc.c contains the inodes allocation and deallocation routines
 */

/*
 * The free inodes are managed by bitmaps.  A file system contains several
 * blocks groups.  Each group contains 1 bitmap block for blocks, 1 bitmap
 * block for inodes, N blocks for the inode table and data blocks.
 *
 * The file system contains group descriptors which are located after the
 * super block.  Each descriptor contains the number of the bitmap block and
 * the free blocks count in the block.
 */

A
Andreas Dilger 已提交
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
/*
 * To avoid calling the atomic setbit hundreds or thousands of times, we only
 * need to use it within a single byte (to ensure we get endianness right).
 * We can use memset for the rest of the bitmap as there are no other users.
 */
void mark_bitmap_end(int start_bit, int end_bit, char *bitmap)
{
	int i;

	if (start_bit >= end_bit)
		return;

	ext4_debug("mark end bits +%d through +%d used\n", start_bit, end_bit);
	for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++)
		ext4_set_bit(i, bitmap);
	if (i < end_bit)
		memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3);
}

/* Initializes an uninitialized inode bitmap */
66 67
unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh,
				ext4_group_t block_group,
A
Andreas Dilger 已提交
68 69 70 71 72 73 74 75 76
				struct ext4_group_desc *gdp)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);

	J_ASSERT_BH(bh, buffer_locked(bh));

	/* If checksum is bad mark all blocks and inodes use to prevent
	 * allocation, essentially implementing a per-group read-only flag. */
	if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) {
77
		ext4_error(sb, __func__, "Checksum bad for group %u",
A
Andreas Dilger 已提交
78
			   block_group);
79 80 81
		ext4_free_blks_set(sb, gdp, 0);
		ext4_free_inodes_set(sb, gdp, 0);
		ext4_itable_unused_set(sb, gdp, 0);
A
Andreas Dilger 已提交
82 83 84 85 86
		memset(bh->b_data, 0xff, sb->s_blocksize);
		return 0;
	}

	memset(bh->b_data, 0, (EXT4_INODES_PER_GROUP(sb) + 7) / 8);
87
	mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), sb->s_blocksize * 8,
A
Andreas Dilger 已提交
88 89 90 91
			bh->b_data);

	return EXT4_INODES_PER_GROUP(sb);
}
92 93 94 95 96 97 98 99

/*
 * Read the inode allocation bitmap for a given block_group, reading
 * into the specified slot in the superblock's bitmap cache.
 *
 * Return buffer_head of bitmap on success or NULL.
 */
static struct buffer_head *
100
ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
101
{
102
	struct ext4_group_desc *desc;
103
	struct buffer_head *bh = NULL;
104
	ext4_fsblk_t bitmap_blk;
105

106
	desc = ext4_get_group_desc(sb, block_group, NULL);
107
	if (!desc)
108 109 110 111 112 113
		return NULL;
	bitmap_blk = ext4_inode_bitmap(sb, desc);
	bh = sb_getblk(sb, bitmap_blk);
	if (unlikely(!bh)) {
		ext4_error(sb, __func__,
			    "Cannot read inode bitmap - "
114
			    "block_group = %u, inode_bitmap = %llu",
115 116 117
			    block_group, bitmap_blk);
		return NULL;
	}
118
	if (bitmap_uptodate(bh))
119 120
		return bh;

121
	lock_buffer(bh);
122 123 124 125
	if (bitmap_uptodate(bh)) {
		unlock_buffer(bh);
		return bh;
	}
126
	spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
A
Andreas Dilger 已提交
127
	if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
128
		ext4_init_inode_bitmap(sb, bh, block_group, desc);
129
		set_bitmap_uptodate(bh);
130
		set_buffer_uptodate(bh);
131
		spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
A
Aneesh Kumar K.V 已提交
132
		unlock_buffer(bh);
133
		return bh;
A
Andreas Dilger 已提交
134
	}
135
	spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
	if (buffer_uptodate(bh)) {
		/*
		 * if not uninit if bh is uptodate,
		 * bitmap is also uptodate
		 */
		set_bitmap_uptodate(bh);
		unlock_buffer(bh);
		return bh;
	}
	/*
	 * submit the buffer_head for read. We can
	 * safely mark the bitmap as uptodate now.
	 * We do it here so the bitmap uptodate bit
	 * get set with buffer lock held.
	 */
	set_bitmap_uptodate(bh);
152 153 154
	if (bh_submit_read(bh) < 0) {
		put_bh(bh);
		ext4_error(sb, __func__,
155
			    "Cannot read inode bitmap - "
156
			    "block_group = %u, inode_bitmap = %llu",
157 158 159
			    block_group, bitmap_blk);
		return NULL;
	}
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
	return bh;
}

/*
 * NOTE! When we get the inode, we're the only people
 * that have access to it, and as such there are no
 * race conditions we have to worry about. The inode
 * is not on the hash-lists, and it cannot be reached
 * through the filesystem because the directory entry
 * has been deleted earlier.
 *
 * HOWEVER: we must make sure that we get no aliases,
 * which means that we have to call "clear_inode()"
 * _before_ we mark the inode not in use in the inode
 * bitmaps. Otherwise a newly created file might use
 * the same inode number (not actually the same pointer
 * though), and then we'd have two inodes sharing the
 * same inode number and space on the harddisk.
 */
179
void ext4_free_inode(handle_t *handle, struct inode *inode)
180
{
181
	struct super_block *sb = inode->i_sb;
182 183 184 185
	int is_directory;
	unsigned long ino;
	struct buffer_head *bitmap_bh = NULL;
	struct buffer_head *bh2;
186
	ext4_group_t block_group;
187
	unsigned long bit;
188 189
	struct ext4_group_desc *gdp;
	struct ext4_super_block *es;
190
	struct ext4_sb_info *sbi;
191
	int fatal = 0, err, count, cleared;
192 193

	if (atomic_read(&inode->i_count) > 1) {
194 195
		printk(KERN_ERR "ext4_free_inode: inode has count=%d\n",
		       atomic_read(&inode->i_count));
196 197 198
		return;
	}
	if (inode->i_nlink) {
199 200
		printk(KERN_ERR "ext4_free_inode: inode has nlink=%d\n",
		       inode->i_nlink);
201 202 203
		return;
	}
	if (!sb) {
204 205
		printk(KERN_ERR "ext4_free_inode: inode on "
		       "nonexistent device\n");
206 207
		return;
	}
208
	sbi = EXT4_SB(sb);
209 210

	ino = inode->i_ino;
211
	ext4_debug("freeing inode %lu\n", ino);
212 213 214 215 216
	trace_mark(ext4_free_inode,
		   "dev %s ino %lu mode %d uid %lu gid %lu bocks %llu",
		   sb->s_id, inode->i_ino, inode->i_mode,
		   (unsigned long) inode->i_uid, (unsigned long) inode->i_gid,
		   (unsigned long long) inode->i_blocks);
217 218 219 220 221

	/*
	 * Note: we must free any quota before locking the superblock,
	 * as writing the quota to disk may need the lock as well.
	 */
222
	vfs_dq_init(inode);
223
	ext4_xattr_delete_inode(handle, inode);
224 225
	vfs_dq_free_inode(inode);
	vfs_dq_drop(inode);
226 227 228 229

	is_directory = S_ISDIR(inode->i_mode);

	/* Do this BEFORE marking the inode not in use or returning an error */
230
	clear_inode(inode);
231

232 233
	es = EXT4_SB(sb)->s_es;
	if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
234 235
		ext4_error(sb, "ext4_free_inode",
			   "reserved or nonexistent inode %lu", ino);
236 237
		goto error_return;
	}
238 239
	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
	bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
240
	bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
241 242 243 244
	if (!bitmap_bh)
		goto error_return;

	BUFFER_TRACE(bitmap_bh, "get_write_access");
245
	fatal = ext4_journal_get_write_access(handle, bitmap_bh);
246 247 248 249
	if (fatal)
		goto error_return;

	/* Ok, now we can actually update the inode bitmaps.. */
250 251 252 253
	spin_lock(sb_bgl_lock(sbi, block_group));
	cleared = ext4_clear_bit(bit, bitmap_bh->b_data);
	spin_unlock(sb_bgl_lock(sbi, block_group));
	if (!cleared)
254 255
		ext4_error(sb, "ext4_free_inode",
			   "bit already cleared for inode %lu", ino);
256
	else {
257
		gdp = ext4_get_group_desc(sb, block_group, &bh2);
258 259

		BUFFER_TRACE(bh2, "get_write_access");
260
		fatal = ext4_journal_get_write_access(handle, bh2);
261 262 263 264
		if (fatal) goto error_return;

		if (gdp) {
			spin_lock(sb_bgl_lock(sbi, block_group));
265 266 267 268 269 270
			count = ext4_free_inodes_count(sb, gdp) + 1;
			ext4_free_inodes_set(sb, gdp, count);
			if (is_directory) {
				count = ext4_used_dirs_count(sb, gdp) - 1;
				ext4_used_dirs_set(sb, gdp, count);
			}
A
Andreas Dilger 已提交
271 272
			gdp->bg_checksum = ext4_group_desc_csum(sbi,
							block_group, gdp);
273 274 275 276 277
			spin_unlock(sb_bgl_lock(sbi, block_group));
			percpu_counter_inc(&sbi->s_freeinodes_counter);
			if (is_directory)
				percpu_counter_dec(&sbi->s_dirs_counter);

278
			if (sbi->s_log_groups_per_flex) {
279 280 281 282
				ext4_group_t f;

				f = ext4_flex_group(sbi, block_group);
				atomic_inc(&sbi->s_flex_groups[f].free_inodes);
283
			}
284
		}
285 286
		BUFFER_TRACE(bh2, "call ext4_handle_dirty_metadata");
		err = ext4_handle_dirty_metadata(handle, NULL, bh2);
287 288
		if (!fatal) fatal = err;
	}
289 290
	BUFFER_TRACE(bitmap_bh, "call ext4_handle_dirty_metadata");
	err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
291 292 293 294 295
	if (!fatal)
		fatal = err;
	sb->s_dirt = 1;
error_return:
	brelse(bitmap_bh);
296
	ext4_std_error(sb, fatal);
297 298 299 300 301 302 303 304 305 306 307 308
}

/*
 * There are two policies for allocating an inode.  If the new inode is
 * a directory, then a forward search is made for a block group with both
 * free space and a low directory-to-inode ratio; if that fails, then of
 * the groups with above-average free space, that group with the fewest
 * directories already is chosen.
 *
 * For other inodes, search forward from the parent directory\'s block
 * group to find a free inode.
 */
309 310
static int find_group_dir(struct super_block *sb, struct inode *parent,
				ext4_group_t *best_group)
311
{
312
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
313
	unsigned int freei, avefreei;
314
	struct ext4_group_desc *desc, *best_desc = NULL;
315 316
	ext4_group_t group;
	int ret = -1;
317

318
	freei = percpu_counter_read_positive(&EXT4_SB(sb)->s_freeinodes_counter);
319 320 321
	avefreei = freei / ngroups;

	for (group = 0; group < ngroups; group++) {
322
		desc = ext4_get_group_desc(sb, group, NULL);
323
		if (!desc || !ext4_free_inodes_count(sb, desc))
324
			continue;
325
		if (ext4_free_inodes_count(sb, desc) < avefreei)
326 327
			continue;
		if (!best_desc ||
328 329
		    (ext4_free_blks_count(sb, desc) >
		     ext4_free_blks_count(sb, best_desc))) {
330
			*best_group = group;
331
			best_desc = desc;
332
			ret = 0;
333 334
		}
	}
335
	return ret;
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
#define free_block_ratio 10

static int find_group_flex(struct super_block *sb, struct inode *parent,
			   ext4_group_t *best_group)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_group_desc *desc;
	struct buffer_head *bh;
	struct flex_groups *flex_group = sbi->s_flex_groups;
	ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
	ext4_group_t parent_fbg_group = ext4_flex_group(sbi, parent_group);
	ext4_group_t ngroups = sbi->s_groups_count;
	int flex_size = ext4_flex_bg_size(sbi);
	ext4_group_t best_flex = parent_fbg_group;
	int blocks_per_flex = sbi->s_blocks_per_group * flex_size;
	int flexbg_free_blocks;
	int flex_freeb_ratio;
	ext4_group_t n_fbg_groups;
	ext4_group_t i;

	n_fbg_groups = (sbi->s_groups_count + flex_size - 1) >>
		sbi->s_log_groups_per_flex;

find_close_to_parent:
362
	flexbg_free_blocks = atomic_read(&flex_group[best_flex].free_blocks);
363
	flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex;
364
	if (atomic_read(&flex_group[best_flex].free_inodes) &&
365 366 367 368 369 370 371 372 373 374 375 376
	    flex_freeb_ratio > free_block_ratio)
		goto found_flexbg;

	if (best_flex && best_flex == parent_fbg_group) {
		best_flex--;
		goto find_close_to_parent;
	}

	for (i = 0; i < n_fbg_groups; i++) {
		if (i == parent_fbg_group || i == parent_fbg_group - 1)
			continue;

377
		flexbg_free_blocks = atomic_read(&flex_group[i].free_blocks);
378 379 380
		flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex;

		if (flex_freeb_ratio > free_block_ratio &&
381
		    (atomic_read(&flex_group[i].free_inodes))) {
382 383 384 385
			best_flex = i;
			goto found_flexbg;
		}

386 387 388 389
		if ((atomic_read(&flex_group[best_flex].free_inodes) == 0) ||
		    ((atomic_read(&flex_group[i].free_blocks) >
		      atomic_read(&flex_group[best_flex].free_blocks)) &&
		     atomic_read(&flex_group[i].free_inodes)))
390 391 392
			best_flex = i;
	}

393 394
	if (!atomic_read(&flex_group[best_flex].free_inodes) ||
	    !atomic_read(&flex_group[best_flex].free_blocks))
395 396 397 398 399 400
		return -1;

found_flexbg:
	for (i = best_flex * flex_size; i < ngroups &&
		     i < (best_flex + 1) * flex_size; i++) {
		desc = ext4_get_group_desc(sb, i, &bh);
401
		if (ext4_free_inodes_count(sb, desc)) {
402 403 404 405 406 407 408 409 410 411
			*best_group = i;
			goto out;
		}
	}

	return -1;
out:
	return 0;
}

412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
struct orlov_stats {
	__u32 free_inodes;
	__u32 free_blocks;
	__u32 used_dirs;
};

/*
 * Helper function for Orlov's allocator; returns critical information
 * for a particular block group or flex_bg.  If flex_size is 1, then g
 * is a block group number; otherwise it is flex_bg number.
 */
void get_orlov_stats(struct super_block *sb, ext4_group_t g,
		       int flex_size, struct orlov_stats *stats)
{
	struct ext4_group_desc *desc;
	ext4_group_t		ngroups = EXT4_SB(sb)->s_groups_count;
	int			i;

	stats->free_inodes = 0;
	stats->free_blocks = 0;
	stats->used_dirs = 0;

	g *= flex_size;

	for (i = 0; i < flex_size; i++) {
		if (g >= ngroups)
			break;
		desc = ext4_get_group_desc(sb, g++, NULL);
		if (!desc)
			continue;

		stats->free_inodes += ext4_free_inodes_count(sb, desc);
		stats->free_blocks += ext4_free_blks_count(sb, desc);
		stats->used_dirs += ext4_used_dirs_count(sb, desc);
	}
}

449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
/*
 * Orlov's allocator for directories.
 *
 * We always try to spread first-level directories.
 *
 * If there are blockgroups with both free inodes and free blocks counts
 * not worse than average we return one with smallest directory count.
 * Otherwise we simply return a random group.
 *
 * For the rest rules look so:
 *
 * It's OK to put directory into a group unless
 * it has too many directories already (max_dirs) or
 * it has too few free inodes left (min_inodes) or
 * it has too few free blocks left (min_blocks) or
464
 * Parent's group is preferred, if it doesn't satisfy these
465 466 467 468 469
 * conditions we search cyclically through the rest. If none
 * of the groups look good we just look for a group with more
 * free inodes than average (starting at parent's group).
 */

470
static int find_group_orlov(struct super_block *sb, struct inode *parent,
471
			    ext4_group_t *group, int mode)
472
{
473
	ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
474
	struct ext4_sb_info *sbi = EXT4_SB(sb);
475
	ext4_group_t ngroups = sbi->s_groups_count;
476
	int inodes_per_group = EXT4_INODES_PER_GROUP(sb);
477
	unsigned int freei, avefreei;
478
	ext4_fsblk_t freeb, avefreeb;
479
	unsigned int ndirs;
480
	int max_dirs, min_inodes;
481
	ext4_grpblk_t min_blocks;
482
	ext4_group_t i, grp, g;
483
	struct ext4_group_desc *desc;
484 485 486 487 488 489 490 491
	struct orlov_stats stats;
	int flex_size = ext4_flex_bg_size(sbi);

	if (flex_size > 1) {
		ngroups = (ngroups + flex_size - 1) >>
			sbi->s_log_groups_per_flex;
		parent_group >>= sbi->s_log_groups_per_flex;
	}
492 493 494 495

	freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter);
	avefreei = freei / ngroups;
	freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
496
	avefreeb = freeb;
A
Andrew Morton 已提交
497
	do_div(avefreeb, ngroups);
498 499
	ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter);

500 501 502
	if (S_ISDIR(mode) &&
	    ((parent == sb->s_root->d_inode) ||
	     (EXT4_I(parent)->i_flags & EXT4_TOPDIR_FL))) {
503
		int best_ndir = inodes_per_group;
504
		int ret = -1;
505

506 507
		get_random_bytes(&grp, sizeof(grp));
		parent_group = (unsigned)grp % ngroups;
508
		for (i = 0; i < ngroups; i++) {
509 510 511
			g = (parent_group + i) % ngroups;
			get_orlov_stats(sb, g, flex_size, &stats);
			if (!stats.free_inodes)
512
				continue;
513
			if (stats.used_dirs >= best_ndir)
514
				continue;
515
			if (stats.free_inodes < avefreei)
516
				continue;
517
			if (stats.free_blocks < avefreeb)
518
				continue;
519
			grp = g;
520
			ret = 0;
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546
			best_ndir = stats.used_dirs;
		}
		if (ret)
			goto fallback;
	found_flex_bg:
		if (flex_size == 1) {
			*group = grp;
			return 0;
		}

		/*
		 * We pack inodes at the beginning of the flexgroup's
		 * inode tables.  Block allocation decisions will do
		 * something similar, although regular files will
		 * start at 2nd block group of the flexgroup.  See
		 * ext4_ext_find_goal() and ext4_find_near().
		 */
		grp *= flex_size;
		for (i = 0; i < flex_size; i++) {
			if (grp+i >= sbi->s_groups_count)
				break;
			desc = ext4_get_group_desc(sb, grp+i, NULL);
			if (desc && ext4_free_inodes_count(sb, desc)) {
				*group = grp+i;
				return 0;
			}
547 548 549 550 551
		}
		goto fallback;
	}

	max_dirs = ndirs / ngroups + inodes_per_group / 16;
552 553 554 555 556 557 558 559 560 561 562 563 564 565
	min_inodes = avefreei - inodes_per_group*flex_size / 4;
	if (min_inodes < 1)
		min_inodes = 1;
	min_blocks = avefreeb - EXT4_BLOCKS_PER_GROUP(sb)*flex_size / 4;

	/*
	 * Start looking in the flex group where we last allocated an
	 * inode for this parent directory
	 */
	if (EXT4_I(parent)->i_last_alloc_group != ~0) {
		parent_group = EXT4_I(parent)->i_last_alloc_group;
		if (flex_size > 1)
			parent_group >>= sbi->s_log_groups_per_flex;
	}
566 567

	for (i = 0; i < ngroups; i++) {
568 569 570
		grp = (parent_group + i) % ngroups;
		get_orlov_stats(sb, grp, flex_size, &stats);
		if (stats.used_dirs >= max_dirs)
571
			continue;
572
		if (stats.free_inodes < min_inodes)
573
			continue;
574
		if (stats.free_blocks < min_blocks)
575
			continue;
576
		goto found_flex_bg;
577 578 579
	}

fallback:
580 581 582
	ngroups = sbi->s_groups_count;
	avefreei = freei / ngroups;
	parent_group = EXT4_I(parent)->i_block_group;
583
	for (i = 0; i < ngroups; i++) {
584 585
		grp = (parent_group + i) % ngroups;
		desc = ext4_get_group_desc(sb, grp, NULL);
586
		if (desc && ext4_free_inodes_count(sb, desc) &&
587 588
		    ext4_free_inodes_count(sb, desc) >= avefreei) {
			*group = grp;
589
			return 0;
590
		}
591 592 593 594 595 596 597 598 599 600 601 602 603 604
	}

	if (avefreei) {
		/*
		 * The free-inodes counter is approximate, and for really small
		 * filesystems the above test can fail to find any blockgroups
		 */
		avefreei = 0;
		goto fallback;
	}

	return -1;
}

605
static int find_group_other(struct super_block *sb, struct inode *parent,
606
			    ext4_group_t *group, int mode)
607
{
608 609
	ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
610
	struct ext4_group_desc *desc;
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650
	ext4_group_t i, last;
	int flex_size = ext4_flex_bg_size(EXT4_SB(sb));

	/*
	 * Try to place the inode is the same flex group as its
	 * parent.  If we can't find space, use the Orlov algorithm to
	 * find another flex group, and store that information in the
	 * parent directory's inode information so that use that flex
	 * group for future allocations.
	 */
	if (flex_size > 1) {
		int retry = 0;

	try_again:
		parent_group &= ~(flex_size-1);
		last = parent_group + flex_size;
		if (last > ngroups)
			last = ngroups;
		for  (i = parent_group; i < last; i++) {
			desc = ext4_get_group_desc(sb, i, NULL);
			if (desc && ext4_free_inodes_count(sb, desc)) {
				*group = i;
				return 0;
			}
		}
		if (!retry && EXT4_I(parent)->i_last_alloc_group != ~0) {
			retry = 1;
			parent_group = EXT4_I(parent)->i_last_alloc_group;
			goto try_again;
		}
		/*
		 * If this didn't work, use the Orlov search algorithm
		 * to find a new flex group; we pass in the mode to
		 * avoid the topdir algorithms.
		 */
		*group = parent_group + flex_size;
		if (*group > ngroups)
			*group = 0;
		return find_group_orlov(sb, parent, group, mode);
	}
651 652 653 654

	/*
	 * Try to place the inode in its parent directory
	 */
655 656
	*group = parent_group;
	desc = ext4_get_group_desc(sb, *group, NULL);
657 658
	if (desc && ext4_free_inodes_count(sb, desc) &&
			ext4_free_blks_count(sb, desc))
659
		return 0;
660 661 662 663 664 665 666 667 668 669

	/*
	 * We're going to place this inode in a different blockgroup from its
	 * parent.  We want to cause files in a common directory to all land in
	 * the same blockgroup.  But we want files which are in a different
	 * directory which shares a blockgroup with our parent to land in a
	 * different blockgroup.
	 *
	 * So add our directory's i_ino into the starting point for the hash.
	 */
670
	*group = (*group + parent->i_ino) % ngroups;
671 672 673 674 675 676

	/*
	 * Use a quadratic hash to find a group with a free inode and some free
	 * blocks.
	 */
	for (i = 1; i < ngroups; i <<= 1) {
677 678 679 680
		*group += i;
		if (*group >= ngroups)
			*group -= ngroups;
		desc = ext4_get_group_desc(sb, *group, NULL);
681 682
		if (desc && ext4_free_inodes_count(sb, desc) &&
				ext4_free_blks_count(sb, desc))
683
			return 0;
684 685 686 687 688 689
	}

	/*
	 * That failed: try linear search for a free inode, even if that group
	 * has no free blocks.
	 */
690
	*group = parent_group;
691
	for (i = 0; i < ngroups; i++) {
692 693 694
		if (++*group >= ngroups)
			*group = 0;
		desc = ext4_get_group_desc(sb, *group, NULL);
695
		if (desc && ext4_free_inodes_count(sb, desc))
696
			return 0;
697 698 699 700 701
	}

	return -1;
}

702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774
/*
 * claim the inode from the inode bitmap. If the group
 * is uninit we need to take the groups's sb_bgl_lock
 * and clear the uninit flag. The inode bitmap update
 * and group desc uninit flag clear should be done
 * after holding sb_bgl_lock so that ext4_read_inode_bitmap
 * doesn't race with the ext4_claim_inode
 */
static int ext4_claim_inode(struct super_block *sb,
			struct buffer_head *inode_bitmap_bh,
			unsigned long ino, ext4_group_t group, int mode)
{
	int free = 0, retval = 0, count;
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_group_desc *gdp = ext4_get_group_desc(sb, group, NULL);

	spin_lock(sb_bgl_lock(sbi, group));
	if (ext4_set_bit(ino, inode_bitmap_bh->b_data)) {
		/* not a free inode */
		retval = 1;
		goto err_ret;
	}
	ino++;
	if ((group == 0 && ino < EXT4_FIRST_INO(sb)) ||
			ino > EXT4_INODES_PER_GROUP(sb)) {
		spin_unlock(sb_bgl_lock(sbi, group));
		ext4_error(sb, __func__,
			   "reserved inode or inode > inodes count - "
			   "block_group = %u, inode=%lu", group,
			   ino + group * EXT4_INODES_PER_GROUP(sb));
		return 1;
	}
	/* If we didn't allocate from within the initialized part of the inode
	 * table then we need to initialize up to this inode. */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) {

		if (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
			gdp->bg_flags &= cpu_to_le16(~EXT4_BG_INODE_UNINIT);
			/* When marking the block group with
			 * ~EXT4_BG_INODE_UNINIT we don't want to depend
			 * on the value of bg_itable_unused even though
			 * mke2fs could have initialized the same for us.
			 * Instead we calculated the value below
			 */

			free = 0;
		} else {
			free = EXT4_INODES_PER_GROUP(sb) -
				ext4_itable_unused_count(sb, gdp);
		}

		/*
		 * Check the relative inode number against the last used
		 * relative inode number in this group. if it is greater
		 * we need to  update the bg_itable_unused count
		 *
		 */
		if (ino > free)
			ext4_itable_unused_set(sb, gdp,
					(EXT4_INODES_PER_GROUP(sb) - ino));
	}
	count = ext4_free_inodes_count(sb, gdp) - 1;
	ext4_free_inodes_set(sb, gdp, count);
	if (S_ISDIR(mode)) {
		count = ext4_used_dirs_count(sb, gdp) + 1;
		ext4_used_dirs_set(sb, gdp, count);
	}
	gdp->bg_checksum = ext4_group_desc_csum(sbi, group, gdp);
err_ret:
	spin_unlock(sb_bgl_lock(sbi, group));
	return retval;
}

775 776 777 778 779 780 781 782 783 784
/*
 * There are two policies for allocating an inode.  If the new inode is
 * a directory, then a forward search is made for a block group with both
 * free space and a low directory-to-inode ratio; if that fails, then of
 * the groups with above-average free space, that group with the fewest
 * directories already is chosen.
 *
 * For other inodes, search forward from the parent directory's block
 * group to find a free inode.
 */
785
struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode)
786 787
{
	struct super_block *sb;
A
Aneesh Kumar K.V 已提交
788 789
	struct buffer_head *inode_bitmap_bh = NULL;
	struct buffer_head *group_desc_bh;
790
	ext4_group_t group = 0;
791
	unsigned long ino = 0;
792 793 794
	struct inode *inode;
	struct ext4_group_desc *gdp = NULL;
	struct ext4_super_block *es;
795 796
	struct ext4_inode_info *ei;
	struct ext4_sb_info *sbi;
797
	int ret2, err = 0;
798
	struct inode *ret;
799 800
	ext4_group_t i;
	int free = 0;
801
	static int once = 1;
802
	ext4_group_t flex_group;
803 804 805 806 807 808

	/* Cannot create files in a deleted directory */
	if (!dir || !dir->i_nlink)
		return ERR_PTR(-EPERM);

	sb = dir->i_sb;
809 810
	trace_mark(ext4_request_inode, "dev %s dir %lu mode %d", sb->s_id,
		   dir->i_ino, mode);
811 812 813
	inode = new_inode(sb);
	if (!inode)
		return ERR_PTR(-ENOMEM);
814
	ei = EXT4_I(inode);
815

816
	sbi = EXT4_SB(sb);
817
	es = sbi->s_es;
818

819
	if (sbi->s_log_groups_per_flex && test_opt(sb, OLDALLOC)) {
820
		ret2 = find_group_flex(sb, dir, &group);
821
		if (ret2 == -1) {
822
			ret2 = find_group_other(sb, dir, &group, mode);
823 824
			if (ret2 == 0 && once)
				once = 0;
825 826 827 828
				printk(KERN_NOTICE "ext4: find_group_flex "
				       "failed, fallback succeeded dir %lu\n",
				       dir->i_ino);
		}
829 830 831
		goto got_group;
	}

832
	if (S_ISDIR(mode)) {
833
		if (test_opt(sb, OLDALLOC))
834
			ret2 = find_group_dir(sb, dir, &group);
835
		else
836
			ret2 = find_group_orlov(sb, dir, &group, mode);
837
	} else
838
		ret2 = find_group_other(sb, dir, &group, mode);
839

840
got_group:
841
	EXT4_I(dir)->i_last_alloc_group = group;
842
	err = -ENOSPC;
843
	if (ret2 == -1)
844 845 846 847 848
		goto out;

	for (i = 0; i < sbi->s_groups_count; i++) {
		err = -EIO;

A
Aneesh Kumar K.V 已提交
849
		gdp = ext4_get_group_desc(sb, group, &group_desc_bh);
850 851 852
		if (!gdp)
			goto fail;

A
Aneesh Kumar K.V 已提交
853 854 855
		brelse(inode_bitmap_bh);
		inode_bitmap_bh = ext4_read_inode_bitmap(sb, group);
		if (!inode_bitmap_bh)
856 857 858 859 860
			goto fail;

		ino = 0;

repeat_in_this_group:
861
		ino = ext4_find_next_zero_bit((unsigned long *)
A
Aneesh Kumar K.V 已提交
862 863 864
					      inode_bitmap_bh->b_data,
					      EXT4_INODES_PER_GROUP(sb), ino);

865
		if (ino < EXT4_INODES_PER_GROUP(sb)) {
866

A
Aneesh Kumar K.V 已提交
867 868 869
			BUFFER_TRACE(inode_bitmap_bh, "get_write_access");
			err = ext4_journal_get_write_access(handle,
							    inode_bitmap_bh);
870 871 872
			if (err)
				goto fail;

873 874 875 876 877 878 879
			BUFFER_TRACE(group_desc_bh, "get_write_access");
			err = ext4_journal_get_write_access(handle,
								group_desc_bh);
			if (err)
				goto fail;
			if (!ext4_claim_inode(sb, inode_bitmap_bh,
						ino, group, mode)) {
880
				/* we won it */
A
Aneesh Kumar K.V 已提交
881
				BUFFER_TRACE(inode_bitmap_bh,
882 883
					"call ext4_handle_dirty_metadata");
				err = ext4_handle_dirty_metadata(handle,
A
Aneesh Kumar K.V 已提交
884 885
								 inode,
							inode_bitmap_bh);
886 887
				if (err)
					goto fail;
888 889
				/* zero bit is inode number 1*/
				ino++;
890 891 892
				goto got;
			}
			/* we lost it */
A
Aneesh Kumar K.V 已提交
893
			ext4_handle_release_buffer(handle, inode_bitmap_bh);
894
			ext4_handle_release_buffer(handle, group_desc_bh);
895

896
			if (++ino < EXT4_INODES_PER_GROUP(sb))
897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913
				goto repeat_in_this_group;
		}

		/*
		 * This case is possible in concurrent environment.  It is very
		 * rare.  We cannot repeat the find_group_xxx() call because
		 * that will simply return the same blockgroup, because the
		 * group descriptor metadata has not yet been updated.
		 * So we just go onto the next blockgroup.
		 */
		if (++group == sbi->s_groups_count)
			group = 0;
	}
	err = -ENOSPC;
	goto out;

got:
A
Andreas Dilger 已提交
914 915 916
	/* We may have to initialize the block bitmap if it isn't already */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM) &&
	    gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
A
Aneesh Kumar K.V 已提交
917
		struct buffer_head *block_bitmap_bh;
A
Andreas Dilger 已提交
918

A
Aneesh Kumar K.V 已提交
919 920 921
		block_bitmap_bh = ext4_read_block_bitmap(sb, group);
		BUFFER_TRACE(block_bitmap_bh, "get block bitmap access");
		err = ext4_journal_get_write_access(handle, block_bitmap_bh);
A
Andreas Dilger 已提交
922
		if (err) {
A
Aneesh Kumar K.V 已提交
923
			brelse(block_bitmap_bh);
A
Andreas Dilger 已提交
924 925 926 927 928 929 930 931
			goto fail;
		}

		free = 0;
		spin_lock(sb_bgl_lock(sbi, group));
		/* recheck and clear flag under lock if we still need to */
		if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
			free = ext4_free_blocks_after_init(sb, group, gdp);
A
Aneesh Kumar K.V 已提交
932
			gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
933
			ext4_free_blks_set(sb, gdp, free);
934 935
			gdp->bg_checksum = ext4_group_desc_csum(sbi, group,
								gdp);
A
Andreas Dilger 已提交
936 937 938 939 940
		}
		spin_unlock(sb_bgl_lock(sbi, group));

		/* Don't need to dirty bitmap block if we didn't change it */
		if (free) {
A
Aneesh Kumar K.V 已提交
941
			BUFFER_TRACE(block_bitmap_bh, "dirty block bitmap");
942
			err = ext4_handle_dirty_metadata(handle,
A
Aneesh Kumar K.V 已提交
943
							NULL, block_bitmap_bh);
A
Andreas Dilger 已提交
944 945
		}

A
Aneesh Kumar K.V 已提交
946
		brelse(block_bitmap_bh);
A
Andreas Dilger 已提交
947 948 949
		if (err)
			goto fail;
	}
A
Aneesh Kumar K.V 已提交
950 951
	BUFFER_TRACE(group_desc_bh, "call ext4_handle_dirty_metadata");
	err = ext4_handle_dirty_metadata(handle, NULL, group_desc_bh);
952 953
	if (err)
		goto fail;
954 955 956 957 958 959

	percpu_counter_dec(&sbi->s_freeinodes_counter);
	if (S_ISDIR(mode))
		percpu_counter_inc(&sbi->s_dirs_counter);
	sb->s_dirt = 1;

960 961
	if (sbi->s_log_groups_per_flex) {
		flex_group = ext4_flex_group(sbi, group);
962
		atomic_dec(&sbi->s_flex_groups[flex_group].free_inodes);
963 964
	}

965
	inode->i_uid = current_fsuid();
966
	if (test_opt(sb, GRPID))
967 968 969 970 971 972
		inode->i_gid = dir->i_gid;
	else if (dir->i_mode & S_ISGID) {
		inode->i_gid = dir->i_gid;
		if (S_ISDIR(mode))
			mode |= S_ISGID;
	} else
973
		inode->i_gid = current_fsgid();
974 975
	inode->i_mode = mode;

A
Andreas Dilger 已提交
976
	inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);
977 978
	/* This is the optimal IO size (for stat), not the fs block size */
	inode->i_blocks = 0;
K
Kalpak Shah 已提交
979 980
	inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime =
						       ext4_current_time(inode);
981 982 983 984 985

	memset(ei->i_data, 0, sizeof(ei->i_data));
	ei->i_dir_start_lookup = 0;
	ei->i_disksize = 0;

986
	/*
987 988 989
	 * Don't inherit extent flag from directory, amongst others. We set
	 * extent flag on newly created directory and file only if -o extent
	 * mount option is specified
990
	 */
991 992
	ei->i_flags =
		ext4_mask_flags(mode, EXT4_I(dir)->i_flags & EXT4_FL_INHERITED);
993 994 995
	ei->i_file_acl = 0;
	ei->i_dtime = 0;
	ei->i_block_group = group;
996
	ei->i_last_alloc_group = ~0;
997

998
	ext4_set_inode_flags(inode);
999
	if (IS_DIRSYNC(inode))
1000
		ext4_handle_sync(handle);
A
Al Viro 已提交
1001 1002 1003 1004
	if (insert_inode_locked(inode) < 0) {
		err = -EINVAL;
		goto fail_drop;
	}
1005 1006 1007 1008
	spin_lock(&sbi->s_next_gen_lock);
	inode->i_generation = sbi->s_next_generation++;
	spin_unlock(&sbi->s_next_gen_lock);

1009
	ei->i_state = EXT4_STATE_NEW;
K
Kalpak Shah 已提交
1010 1011

	ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize;
1012 1013

	ret = inode;
1014
	if (vfs_dq_alloc_inode(inode)) {
1015 1016 1017 1018
		err = -EDQUOT;
		goto fail_drop;
	}

1019
	err = ext4_init_acl(handle, inode, dir);
1020 1021 1022
	if (err)
		goto fail_free_drop;

1023
	err = ext4_init_security(handle, inode, dir);
1024 1025 1026
	if (err)
		goto fail_free_drop;

1027
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) {
1028
		/* set extent flag only for directory, file and normal symlink*/
1029
		if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) {
1030 1031 1032
			EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL;
			ext4_ext_tree_init(handle, inode);
		}
A
Alex Tomas 已提交
1033
	}
1034

1035 1036 1037 1038 1039 1040
	err = ext4_mark_inode_dirty(handle, inode);
	if (err) {
		ext4_std_error(sb, err);
		goto fail_free_drop;
	}

1041
	ext4_debug("allocating inode %lu\n", inode->i_ino);
1042 1043
	trace_mark(ext4_allocate_inode, "dev %s ino %lu dir %lu mode %d",
		   sb->s_id, inode->i_ino, dir->i_ino, mode);
1044 1045
	goto really_out;
fail:
1046
	ext4_std_error(sb, err);
1047 1048 1049 1050
out:
	iput(inode);
	ret = ERR_PTR(err);
really_out:
A
Aneesh Kumar K.V 已提交
1051
	brelse(inode_bitmap_bh);
1052 1053 1054
	return ret;

fail_free_drop:
1055
	vfs_dq_free_inode(inode);
1056 1057

fail_drop:
1058
	vfs_dq_drop(inode);
1059 1060
	inode->i_flags |= S_NOQUOTA;
	inode->i_nlink = 0;
A
Al Viro 已提交
1061
	unlock_new_inode(inode);
1062
	iput(inode);
A
Aneesh Kumar K.V 已提交
1063
	brelse(inode_bitmap_bh);
1064 1065 1066 1067
	return ERR_PTR(err);
}

/* Verify that we are loading a valid orphan from disk */
1068
struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
1069
{
1070
	unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count);
1071
	ext4_group_t block_group;
1072
	int bit;
1073
	struct buffer_head *bitmap_bh;
1074
	struct inode *inode = NULL;
1075
	long err = -EIO;
1076 1077 1078

	/* Error cases - e2fsck has already cleaned up for us */
	if (ino > max_ino) {
1079
		ext4_warning(sb, __func__,
1080
			     "bad orphan ino %lu!  e2fsck was run?", ino);
1081
		goto error;
1082 1083
	}

1084 1085
	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
	bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
1086
	bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
1087
	if (!bitmap_bh) {
1088
		ext4_warning(sb, __func__,
1089
			     "inode bitmap error for orphan %lu", ino);
1090
		goto error;
1091 1092 1093 1094 1095 1096
	}

	/* Having the inode bit set should be a 100% indicator that this
	 * is a valid orphan (no e2fsck run on fs).  Orphans also include
	 * inodes that were being truncated, so we can't check i_nlink==0.
	 */
1097 1098 1099 1100 1101 1102 1103
	if (!ext4_test_bit(bit, bitmap_bh->b_data))
		goto bad_orphan;

	inode = ext4_iget(sb, ino);
	if (IS_ERR(inode))
		goto iget_failed;

1104 1105 1106 1107 1108 1109 1110 1111
	/*
	 * If the orphans has i_nlinks > 0 then it should be able to be
	 * truncated, otherwise it won't be removed from the orphan list
	 * during processing and an infinite loop will result.
	 */
	if (inode->i_nlink && !ext4_can_truncate(inode))
		goto bad_orphan;

1112 1113 1114 1115 1116 1117 1118 1119 1120
	if (NEXT_ORPHAN(inode) > max_ino)
		goto bad_orphan;
	brelse(bitmap_bh);
	return inode;

iget_failed:
	err = PTR_ERR(inode);
	inode = NULL;
bad_orphan:
1121
	ext4_warning(sb, __func__,
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
		     "bad orphan inode %lu!  e2fsck was run?", ino);
	printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n",
	       bit, (unsigned long long)bitmap_bh->b_blocknr,
	       ext4_test_bit(bit, bitmap_bh->b_data));
	printk(KERN_NOTICE "inode=%p\n", inode);
	if (inode) {
		printk(KERN_NOTICE "is_bad_inode(inode)=%d\n",
		       is_bad_inode(inode));
		printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
		       NEXT_ORPHAN(inode));
		printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
1133
		printk(KERN_NOTICE "i_nlink=%u\n", inode->i_nlink);
1134
		/* Avoid freeing blocks if we got a bad deleted inode */
1135
		if (inode->i_nlink == 0)
1136 1137 1138 1139
			inode->i_blocks = 0;
		iput(inode);
	}
	brelse(bitmap_bh);
1140 1141
error:
	return ERR_PTR(err);
1142 1143
}

1144
unsigned long ext4_count_free_inodes(struct super_block *sb)
1145 1146
{
	unsigned long desc_count;
1147
	struct ext4_group_desc *gdp;
1148
	ext4_group_t i;
1149 1150
#ifdef EXT4FS_DEBUG
	struct ext4_super_block *es;
1151 1152 1153
	unsigned long bitmap_count, x;
	struct buffer_head *bitmap_bh = NULL;

1154
	es = EXT4_SB(sb)->s_es;
1155 1156 1157
	desc_count = 0;
	bitmap_count = 0;
	gdp = NULL;
1158
	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
1159
		gdp = ext4_get_group_desc(sb, i, NULL);
1160 1161
		if (!gdp)
			continue;
1162
		desc_count += ext4_free_inodes_count(sb, gdp);
1163
		brelse(bitmap_bh);
1164
		bitmap_bh = ext4_read_inode_bitmap(sb, i);
1165 1166 1167
		if (!bitmap_bh)
			continue;

1168
		x = ext4_count_free(bitmap_bh, EXT4_INODES_PER_GROUP(sb) / 8);
E
Eric Sandeen 已提交
1169
		printk(KERN_DEBUG "group %lu: stored = %d, counted = %lu\n",
1170
			i, ext4_free_inodes_count(sb, gdp), x);
1171 1172 1173
		bitmap_count += x;
	}
	brelse(bitmap_bh);
1174 1175 1176
	printk(KERN_DEBUG "ext4_count_free_inodes: "
	       "stored = %u, computed = %lu, %lu\n",
	       le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
1177 1178 1179
	return desc_count;
#else
	desc_count = 0;
1180
	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
1181
		gdp = ext4_get_group_desc(sb, i, NULL);
1182 1183
		if (!gdp)
			continue;
1184
		desc_count += ext4_free_inodes_count(sb, gdp);
1185 1186 1187 1188 1189 1190 1191
		cond_resched();
	}
	return desc_count;
#endif
}

/* Called at mount-time, super-block is locked */
1192
unsigned long ext4_count_dirs(struct super_block * sb)
1193 1194
{
	unsigned long count = 0;
1195
	ext4_group_t i;
1196

1197
	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
1198
		struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);
1199 1200
		if (!gdp)
			continue;
1201
		count += ext4_used_dirs_count(sb, gdp);
1202 1203 1204
	}
	return count;
}