log.c 19.6 KB
Newer Older
D
David Teigland 已提交
1 2
/*
 * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
3
 * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
D
David Teigland 已提交
4 5 6
 *
 * This copyrighted material is made available to anyone wishing to use,
 * modify, copy, or redistribute it subject to the terms and conditions
7
 * of the GNU General Public License version 2.
D
David Teigland 已提交
8 9 10 11 12 13 14
 */

#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
15
#include <linux/gfs2_ondisk.h>
16
#include <linux/crc32.h>
17
#include <linux/lm_interface.h>
18
#include <linux/delay.h>
D
David Teigland 已提交
19 20

#include "gfs2.h"
21
#include "incore.h"
D
David Teigland 已提交
22 23 24 25 26
#include "bmap.h"
#include "glock.h"
#include "log.h"
#include "lops.h"
#include "meta_io.h"
27
#include "util.h"
28
#include "dir.h"
D
David Teigland 已提交
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

#define PULL 1

/**
 * gfs2_struct2blk - compute stuff
 * @sdp: the filesystem
 * @nstruct: the number of structures
 * @ssize: the size of the structures
 *
 * Compute the number of log descriptor blocks needed to hold a certain number
 * of structures of a certain size.
 *
 * Returns: the number of blocks needed (minimum is always 1)
 */

unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
			     unsigned int ssize)
{
	unsigned int blks;
	unsigned int first, second;

	blks = 1;
S
Steven Whitehouse 已提交
51
	first = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / ssize;
D
David Teigland 已提交
52 53

	if (nstruct > first) {
54 55
		second = (sdp->sd_sb.sb_bsize -
			  sizeof(struct gfs2_meta_header)) / ssize;
56
		blks += DIV_ROUND_UP(nstruct - first, second);
D
David Teigland 已提交
57 58 59 60 61
	}

	return blks;
}

62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
/**
 * gfs2_remove_from_ail - Remove an entry from the ail lists, updating counters
 * @mapping: The associated mapping (maybe NULL)
 * @bd: The gfs2_bufdata to remove
 *
 * The log lock _must_ be held when calling this function
 *
 */

void gfs2_remove_from_ail(struct address_space *mapping, struct gfs2_bufdata *bd)
{
	bd->bd_ail = NULL;
	list_del(&bd->bd_ail_st_list);
	list_del(&bd->bd_ail_gl_list);
	atomic_dec(&bd->bd_gl->gl_ail_count);
	if (mapping)
		gfs2_meta_cache_flush(GFS2_I(mapping->host));
	brelse(bd->bd_bh);
}

82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165
/**
 * gfs2_ail1_start_one - Start I/O on a part of the AIL
 * @sdp: the filesystem
 * @tr: the part of the AIL
 *
 */

static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
{
	struct gfs2_bufdata *bd, *s;
	struct buffer_head *bh;
	int retry;

	BUG_ON(!spin_is_locked(&sdp->sd_log_lock));

	do {
		retry = 0;

		list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list,
						 bd_ail_st_list) {
			bh = bd->bd_bh;

			gfs2_assert(sdp, bd->bd_ail == ai);

			if (!buffer_busy(bh)) {
				if (!buffer_uptodate(bh)) {
					gfs2_log_unlock(sdp);
					gfs2_io_error_bh(sdp, bh);
					gfs2_log_lock(sdp);
				}
				list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
				continue;
			}

			if (!buffer_dirty(bh))
				continue;

			list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list);

			gfs2_log_unlock(sdp);
			wait_on_buffer(bh);
			ll_rw_block(WRITE, 1, &bh);
			gfs2_log_lock(sdp);

			retry = 1;
			break;
		}
	} while (retry);
}

/**
 * gfs2_ail1_empty_one - Check whether or not a trans in the AIL has been synced
 * @sdp: the filesystem
 * @ai: the AIL entry
 *
 */

static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int flags)
{
	struct gfs2_bufdata *bd, *s;
	struct buffer_head *bh;

	list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list,
					 bd_ail_st_list) {
		bh = bd->bd_bh;

		gfs2_assert(sdp, bd->bd_ail == ai);

		if (buffer_busy(bh)) {
			if (flags & DIO_ALL)
				continue;
			else
				break;
		}

		if (!buffer_uptodate(bh))
			gfs2_io_error_bh(sdp, bh);

		list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
	}

	return list_empty(&ai->ai_ail1_list);
}

166
static void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags)
D
David Teigland 已提交
167
{
168
	struct list_head *head;
169
	u64 sync_gen;
170 171 172
	struct list_head *first;
	struct gfs2_ail *first_ai, *ai, *tmp;
	int done = 0;
D
David Teigland 已提交
173 174

	gfs2_log_lock(sdp);
175
	head = &sdp->sd_ail1_list;
D
David Teigland 已提交
176 177 178 179 180 181 182 183 184
	if (list_empty(head)) {
		gfs2_log_unlock(sdp);
		return;
	}
	sync_gen = sdp->sd_ail_sync_gen++;

	first = head->prev;
	first_ai = list_entry(first, struct gfs2_ail, ai_list);
	first_ai->ai_sync_gen = sync_gen;
185
	gfs2_ail1_start_one(sdp, first_ai); /* This may drop log lock */
D
David Teigland 已提交
186 187 188 189

	if (flags & DIO_ALL)
		first = NULL;

190
	while(!done) {
191 192
		if (first && (head->prev != first ||
			      gfs2_ail1_empty_one(sdp, first_ai, 0)))
D
David Teigland 已提交
193 194
			break;

195 196
		done = 1;
		list_for_each_entry_safe_reverse(ai, tmp, head, ai_list) {
D
David Teigland 已提交
197 198 199
			if (ai->ai_sync_gen >= sync_gen)
				continue;
			ai->ai_sync_gen = sync_gen;
200 201
			gfs2_ail1_start_one(sdp, ai); /* This may drop log lock */
			done = 0;
D
David Teigland 已提交
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
			break;
		}
	}

	gfs2_log_unlock(sdp);
}

int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags)
{
	struct gfs2_ail *ai, *s;
	int ret;

	gfs2_log_lock(sdp);

	list_for_each_entry_safe_reverse(ai, s, &sdp->sd_ail1_list, ai_list) {
		if (gfs2_ail1_empty_one(sdp, ai, flags))
			list_move(&ai->ai_list, &sdp->sd_ail2_list);
		else if (!(flags & DIO_ALL))
			break;
	}

	ret = list_empty(&sdp->sd_ail1_list);

	gfs2_log_unlock(sdp);

	return ret;
}

230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246

/**
 * gfs2_ail2_empty_one - Check whether or not a trans in the AIL has been synced
 * @sdp: the filesystem
 * @ai: the AIL entry
 *
 */

static void gfs2_ail2_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
{
	struct list_head *head = &ai->ai_ail2_list;
	struct gfs2_bufdata *bd;

	while (!list_empty(head)) {
		bd = list_entry(head->prev, struct gfs2_bufdata,
				bd_ail_st_list);
		gfs2_assert(sdp, bd->bd_ail == ai);
247
		gfs2_remove_from_ail(bd->bd_bh->b_page->mapping, bd);
248 249 250
	}
}

D
David Teigland 已提交
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
static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail)
{
	struct gfs2_ail *ai, *safe;
	unsigned int old_tail = sdp->sd_log_tail;
	int wrap = (new_tail < old_tail);
	int a, b, rm;

	gfs2_log_lock(sdp);

	list_for_each_entry_safe(ai, safe, &sdp->sd_ail2_list, ai_list) {
		a = (old_tail <= ai->ai_first);
		b = (ai->ai_first < new_tail);
		rm = (wrap) ? (a || b) : (a && b);
		if (!rm)
			continue;

		gfs2_ail2_empty_one(sdp, ai);
		list_del(&ai->ai_list);
		gfs2_assert_warn(sdp, list_empty(&ai->ai_ail1_list));
		gfs2_assert_warn(sdp, list_empty(&ai->ai_ail2_list));
		kfree(ai);
	}

	gfs2_log_unlock(sdp);
}

/**
 * gfs2_log_reserve - Make a log reservation
 * @sdp: The GFS2 superblock
 * @blks: The number of blocks to reserve
 *
282
 * Note that we never give out the last few blocks of the journal. Thats
283
 * due to the fact that there is a small number of header blocks
284 285 286 287
 * associated with each log flush. The exact number can't be known until
 * flush time, so we ensure that we have just enough free blocks at all
 * times to avoid running out during a log flush.
 *
D
David Teigland 已提交
288 289 290 291 292 293
 * Returns: errno
 */

int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks)
{
	unsigned int try = 0;
294
	unsigned reserved_blks = 6 * (4096 / sdp->sd_vfs->s_blocksize);
D
David Teigland 已提交
295 296 297 298 299

	if (gfs2_assert_warn(sdp, blks) ||
	    gfs2_assert_warn(sdp, blks <= sdp->sd_jdesc->jd_blocks))
		return -EINVAL;

300
	mutex_lock(&sdp->sd_log_reserve_mutex);
301
	gfs2_log_lock(sdp);
302
	while(sdp->sd_log_blks_free <= (blks + reserved_blks)) {
D
David Teigland 已提交
303 304
		gfs2_log_unlock(sdp);
		gfs2_ail1_empty(sdp, 0);
305
		gfs2_log_flush(sdp, NULL);
D
David Teigland 已提交
306 307 308

		if (try++)
			gfs2_ail1_start(sdp, 0);
309
		gfs2_log_lock(sdp);
D
David Teigland 已提交
310
	}
311 312 313 314 315
	sdp->sd_log_blks_free -= blks;
	gfs2_log_unlock(sdp);
	mutex_unlock(&sdp->sd_log_reserve_mutex);

	down_read(&sdp->sd_log_flush_lock);
D
David Teigland 已提交
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334

	return 0;
}

/**
 * gfs2_log_release - Release a given number of log blocks
 * @sdp: The GFS2 superblock
 * @blks: The number of blocks
 *
 */

void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks)
{

	gfs2_log_lock(sdp);
	sdp->sd_log_blks_free += blks;
	gfs2_assert_withdraw(sdp,
			     sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks);
	gfs2_log_unlock(sdp);
335
	up_read(&sdp->sd_log_flush_lock);
D
David Teigland 已提交
336 337
}

338
static u64 log_bmap(struct gfs2_sbd *sdp, unsigned int lbn)
D
David Teigland 已提交
339
{
340
	struct inode *inode = sdp->sd_jdesc->jd_inode;
D
David Teigland 已提交
341
	int error;
342
	struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 };
D
David Teigland 已提交
343

344 345
	bh_map.b_size = 1 << inode->i_blkbits;
	error = gfs2_block_map(inode, lbn, 0, &bh_map);
346
	if (error || !bh_map.b_blocknr)
347 348
		printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error,
		       (unsigned long long)bh_map.b_blocknr, lbn);
349
	gfs2_assert_withdraw(sdp, !error && bh_map.b_blocknr);
D
David Teigland 已提交
350

351
	return bh_map.b_blocknr;
D
David Teigland 已提交
352 353 354 355 356 357 358 359 360 361 362 363 364 365
}

/**
 * log_distance - Compute distance between two journal blocks
 * @sdp: The GFS2 superblock
 * @newer: The most recent journal block of the pair
 * @older: The older journal block of the pair
 *
 *   Compute the distance (in the journal direction) between two
 *   blocks in the journal
 *
 * Returns: the distance in blocks
 */

S
Steven Whitehouse 已提交
366
static inline unsigned int log_distance(struct gfs2_sbd *sdp, unsigned int newer,
D
David Teigland 已提交
367 368 369 370 371 372 373 374 375 376 377
					unsigned int older)
{
	int dist;

	dist = newer - older;
	if (dist < 0)
		dist += sdp->sd_jdesc->jd_blocks;

	return dist;
}

378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
/**
 * calc_reserved - Calculate the number of blocks to reserve when
 *                 refunding a transaction's unused buffers.
 * @sdp: The GFS2 superblock
 *
 * This is complex.  We need to reserve room for all our currently used
 * metadata buffers (e.g. normal file I/O rewriting file time stamps) and 
 * all our journaled data buffers for journaled files (e.g. files in the 
 * meta_fs like rindex, or files for which chattr +j was done.)
 * If we don't reserve enough space, gfs2_log_refund and gfs2_log_flush
 * will count it as free space (sd_log_blks_free) and corruption will follow.
 *
 * We can have metadata bufs and jdata bufs in the same journal.  So each
 * type gets its own log header, for which we need to reserve a block.
 * In fact, each type has the potential for needing more than one header 
 * in cases where we have more buffers than will fit on a journal page.
 * Metadata journal entries take up half the space of journaled buffer entries.
 * Thus, metadata entries have buf_limit (502) and journaled buffers have
 * databuf_limit (251) before they cause a wrap around.
 *
 * Also, we need to reserve blocks for revoke journal entries and one for an
 * overall header for the lot.
 *
 * Returns: the number of blocks reserved
 */
static unsigned int calc_reserved(struct gfs2_sbd *sdp)
{
	unsigned int reserved = 0;
	unsigned int mbuf_limit, metabufhdrs_needed;
	unsigned int dbuf_limit, databufhdrs_needed;
	unsigned int revokes = 0;

	mbuf_limit = buf_limit(sdp);
	metabufhdrs_needed = (sdp->sd_log_commited_buf +
			      (mbuf_limit - 1)) / mbuf_limit;
	dbuf_limit = databuf_limit(sdp);
	databufhdrs_needed = (sdp->sd_log_commited_databuf +
			      (dbuf_limit - 1)) / dbuf_limit;

	if (sdp->sd_log_commited_revoke)
		revokes = gfs2_struct2blk(sdp, sdp->sd_log_commited_revoke,
					  sizeof(u64));

	reserved = sdp->sd_log_commited_buf + metabufhdrs_needed +
		sdp->sd_log_commited_databuf + databufhdrs_needed +
		revokes;
	/* One for the overall header */
	if (reserved)
		reserved++;
	return reserved;
}

D
David Teigland 已提交
430 431 432 433 434 435 436
static unsigned int current_tail(struct gfs2_sbd *sdp)
{
	struct gfs2_ail *ai;
	unsigned int tail;

	gfs2_log_lock(sdp);

S
Steven Whitehouse 已提交
437
	if (list_empty(&sdp->sd_ail1_list)) {
D
David Teigland 已提交
438
		tail = sdp->sd_log_head;
S
Steven Whitehouse 已提交
439 440
	} else {
		ai = list_entry(sdp->sd_ail1_list.prev, struct gfs2_ail, ai_list);
D
David Teigland 已提交
441 442 443 444 445 446 447 448 449 450 451
		tail = ai->ai_first;
	}

	gfs2_log_unlock(sdp);

	return tail;
}

static inline void log_incr_head(struct gfs2_sbd *sdp)
{
	if (sdp->sd_log_flush_head == sdp->sd_log_tail)
S
Steven Whitehouse 已提交
452
		gfs2_assert_withdraw(sdp, sdp->sd_log_flush_head == sdp->sd_log_head);
D
David Teigland 已提交
453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468

	if (++sdp->sd_log_flush_head == sdp->sd_jdesc->jd_blocks) {
		sdp->sd_log_flush_head = 0;
		sdp->sd_log_flush_wrapped = 1;
	}
}

/**
 * gfs2_log_get_buf - Get and initialize a buffer to use for log control data
 * @sdp: The GFS2 superblock
 *
 * Returns: the buffer_head
 */

struct buffer_head *gfs2_log_get_buf(struct gfs2_sbd *sdp)
{
469
	u64 blkno = log_bmap(sdp, sdp->sd_log_flush_head);
D
David Teigland 已提交
470 471 472
	struct gfs2_log_buf *lb;
	struct buffer_head *bh;

473
	lb = kzalloc(sizeof(struct gfs2_log_buf), GFP_NOFS | __GFP_NOFAIL);
D
David Teigland 已提交
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498
	list_add(&lb->lb_list, &sdp->sd_log_flush_list);

	bh = lb->lb_bh = sb_getblk(sdp->sd_vfs, blkno);
	lock_buffer(bh);
	memset(bh->b_data, 0, bh->b_size);
	set_buffer_uptodate(bh);
	clear_buffer_dirty(bh);
	unlock_buffer(bh);

	log_incr_head(sdp);

	return bh;
}

/**
 * gfs2_log_fake_buf - Build a fake buffer head to write metadata buffer to log
 * @sdp: the filesystem
 * @data: the data the buffer_head should point to
 *
 * Returns: the log buffer descriptor
 */

struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp,
				      struct buffer_head *real)
{
499
	u64 blkno = log_bmap(sdp, sdp->sd_log_flush_head);
D
David Teigland 已提交
500 501 502
	struct gfs2_log_buf *lb;
	struct buffer_head *bh;

503
	lb = kzalloc(sizeof(struct gfs2_log_buf), GFP_NOFS | __GFP_NOFAIL);
D
David Teigland 已提交
504 505 506 507 508 509
	list_add(&lb->lb_list, &sdp->sd_log_flush_list);
	lb->lb_real = real;

	bh = lb->lb_bh = alloc_buffer_head(GFP_NOFS | __GFP_NOFAIL);
	atomic_set(&bh->b_count, 1);
	bh->b_state = (1 << BH_Mapped) | (1 << BH_Uptodate);
510
	set_bh_page(bh, real->b_page, bh_offset(real));
D
David Teigland 已提交
511 512 513 514 515 516 517 518 519
	bh->b_blocknr = blkno;
	bh->b_size = sdp->sd_sb.sb_bsize;
	bh->b_bdev = sdp->sd_vfs->s_bdev;

	log_incr_head(sdp);

	return bh;
}

520
static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail)
D
David Teigland 已提交
521 522 523 524 525 526
{
	unsigned int dist = log_distance(sdp, new_tail, sdp->sd_log_tail);

	ail2_empty(sdp, new_tail);

	gfs2_log_lock(sdp);
527
	sdp->sd_log_blks_free += dist;
S
Steven Whitehouse 已提交
528
	gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks);
D
David Teigland 已提交
529 530 531 532 533 534 535 536 537 538 539 540
	gfs2_log_unlock(sdp);

	sdp->sd_log_tail = new_tail;
}

/**
 * log_write_header - Get and initialize a journal header buffer
 * @sdp: The GFS2 superblock
 *
 * Returns: the initialized log buffer descriptor
 */

541
static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull)
D
David Teigland 已提交
542
{
543
	u64 blkno = log_bmap(sdp, sdp->sd_log_flush_head);
D
David Teigland 已提交
544 545 546
	struct buffer_head *bh;
	struct gfs2_log_header *lh;
	unsigned int tail;
547
	u32 hash;
D
David Teigland 已提交
548 549 550 551 552 553 554 555 556 557 558 559 560 561

	bh = sb_getblk(sdp->sd_vfs, blkno);
	lock_buffer(bh);
	memset(bh->b_data, 0, bh->b_size);
	set_buffer_uptodate(bh);
	clear_buffer_dirty(bh);
	unlock_buffer(bh);

	gfs2_ail1_empty(sdp, 0);
	tail = current_tail(sdp);

	lh = (struct gfs2_log_header *)bh->b_data;
	memset(lh, 0, sizeof(struct gfs2_log_header));
	lh->lh_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
562 563
	lh->lh_header.mh_type = cpu_to_be32(GFS2_METATYPE_LH);
	lh->lh_header.mh_format = cpu_to_be32(GFS2_FORMAT_LH);
564 565 566 567
	lh->lh_sequence = cpu_to_be64(sdp->sd_log_sequence++);
	lh->lh_flags = cpu_to_be32(flags);
	lh->lh_tail = cpu_to_be32(tail);
	lh->lh_blkno = cpu_to_be32(sdp->sd_log_flush_head);
D
David Teigland 已提交
568 569 570 571 572 573 574 575 576
	hash = gfs2_disk_hash(bh->b_data, sizeof(struct gfs2_log_header));
	lh->lh_hash = cpu_to_be32(hash);

	set_buffer_dirty(bh);
	if (sync_dirty_buffer(bh))
		gfs2_io_error_bh(sdp, bh);
	brelse(bh);

	if (sdp->sd_log_tail != tail)
577
		log_pull_tail(sdp, tail);
D
David Teigland 已提交
578 579 580 581 582 583 584 585 586 587 588 589
	else
		gfs2_assert_withdraw(sdp, !pull);

	sdp->sd_log_idle = (tail == sdp->sd_log_flush_head);
	log_incr_head(sdp);
}

static void log_flush_commit(struct gfs2_sbd *sdp)
{
	struct list_head *head = &sdp->sd_log_flush_list;
	struct gfs2_log_buf *lb;
	struct buffer_head *bh;
590
	int flushcount = 0;
D
David Teigland 已提交
591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606

	while (!list_empty(head)) {
		lb = list_entry(head->next, struct gfs2_log_buf, lb_list);
		list_del(&lb->lb_list);
		bh = lb->lb_bh;

		wait_on_buffer(bh);
		if (!buffer_uptodate(bh))
			gfs2_io_error_bh(sdp, bh);
		if (lb->lb_real) {
			while (atomic_read(&bh->b_count) != 1)  /* Grrrr... */
				schedule();
			free_buffer_head(bh);
		} else
			brelse(bh);
		kfree(lb);
607
		flushcount++;
D
David Teigland 已提交
608 609
	}

610 611 612 613 614 615 616 617 618 619 620
	/* If nothing was journaled, the header is unplanned and unwanted. */
	if (flushcount) {
		log_write_header(sdp, 0, 0);
	} else {
		unsigned int tail;
		tail = current_tail(sdp);

		gfs2_ail1_empty(sdp, 0);
		if (sdp->sd_log_tail != tail)
			log_pull_tail(sdp, tail);
	}
D
David Teigland 已提交
621 622 623
}

/**
624
 * gfs2_log_flush - flush incore transaction(s)
D
David Teigland 已提交
625 626 627 628 629
 * @sdp: the filesystem
 * @gl: The glock structure to flush.  If NULL, flush the whole incore log
 *
 */

630
void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
D
David Teigland 已提交
631 632 633
{
	struct gfs2_ail *ai;

634
	down_write(&sdp->sd_log_flush_lock);
635 636 637 638 639

	if (gl) {
		gfs2_log_lock(sdp);
		if (list_empty(&gl->gl_le.le_list)) {
			gfs2_log_unlock(sdp);
640
			up_write(&sdp->sd_log_flush_lock);
641 642 643 644 645
			return;
		}
		gfs2_log_unlock(sdp);
	}

646 647 648
	ai = kzalloc(sizeof(struct gfs2_ail), GFP_NOFS | __GFP_NOFAIL);
	INIT_LIST_HEAD(&ai->ai_ail1_list);
	INIT_LIST_HEAD(&ai->ai_ail2_list);
D
David Teigland 已提交
649

650 651 652 653
	gfs2_assert_withdraw(sdp,
			     sdp->sd_log_num_buf + sdp->sd_log_num_jdata ==
			     sdp->sd_log_commited_buf +
			     sdp->sd_log_commited_databuf);
D
David Teigland 已提交
654 655 656 657 658 659 660 661 662 663
	gfs2_assert_withdraw(sdp,
			sdp->sd_log_num_revoke == sdp->sd_log_commited_revoke);

	sdp->sd_log_flush_head = sdp->sd_log_head;
	sdp->sd_log_flush_wrapped = 0;
	ai->ai_first = sdp->sd_log_flush_head;

	lops_before_commit(sdp);
	if (!list_empty(&sdp->sd_log_flush_list))
		log_flush_commit(sdp);
664 665 666 667
	else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle){
		gfs2_log_lock(sdp);
		sdp->sd_log_blks_free--; /* Adjust for unreserved buffer */
		gfs2_log_unlock(sdp);
D
David Teigland 已提交
668
		log_write_header(sdp, 0, PULL);
669
	}
D
David Teigland 已提交
670
	lops_after_commit(sdp, ai);
671

672 673
	gfs2_log_lock(sdp);
	sdp->sd_log_head = sdp->sd_log_flush_head;
S
Steven Whitehouse 已提交
674 675
	sdp->sd_log_blks_reserved = 0;
	sdp->sd_log_commited_buf = 0;
676
	sdp->sd_log_commited_databuf = 0;
S
Steven Whitehouse 已提交
677
	sdp->sd_log_commited_revoke = 0;
D
David Teigland 已提交
678 679 680 681 682 683 684 685

	if (!list_empty(&ai->ai_ail1_list)) {
		list_add(&ai->ai_list, &sdp->sd_ail1_list);
		ai = NULL;
	}
	gfs2_log_unlock(sdp);

	sdp->sd_vfs->s_dirt = 0;
686
	up_write(&sdp->sd_log_flush_lock);
D
David Teigland 已提交
687 688 689 690 691 692

	kfree(ai);
}

static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
{
693
	unsigned int reserved;
D
David Teigland 已提交
694 695 696 697 698
	unsigned int old;

	gfs2_log_lock(sdp);

	sdp->sd_log_commited_buf += tr->tr_num_buf_new - tr->tr_num_buf_rm;
699 700 701 702
	sdp->sd_log_commited_databuf += tr->tr_num_databuf_new -
		tr->tr_num_databuf_rm;
	gfs2_assert_withdraw(sdp, (((int)sdp->sd_log_commited_buf) >= 0) ||
			     (((int)sdp->sd_log_commited_databuf) >= 0));
D
David Teigland 已提交
703 704
	sdp->sd_log_commited_revoke += tr->tr_num_revoke - tr->tr_num_revoke_rm;
	gfs2_assert_withdraw(sdp, ((int)sdp->sd_log_commited_revoke) >= 0);
705
	reserved = calc_reserved(sdp);
D
David Teigland 已提交
706 707 708 709
	old = sdp->sd_log_blks_free;
	sdp->sd_log_blks_free += tr->tr_reserved -
				 (reserved - sdp->sd_log_blks_reserved);

710
	gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free >= old);
711 712
	gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free <=
			     sdp->sd_jdesc->jd_blocks);
D
David Teigland 已提交
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732

	sdp->sd_log_blks_reserved = reserved;

	gfs2_log_unlock(sdp);
}

/**
 * gfs2_log_commit - Commit a transaction to the log
 * @sdp: the filesystem
 * @tr: the transaction
 *
 * Returns: errno
 */

void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
{
	log_refund(sdp, tr);
	lops_incore_commit(sdp, tr);

	sdp->sd_vfs->s_dirt = 1;
733
	up_read(&sdp->sd_log_flush_lock);
D
David Teigland 已提交
734 735

	gfs2_log_lock(sdp);
736 737 738
	if (sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks))
		wake_up_process(sdp->sd_logd_process);
	gfs2_log_unlock(sdp);
D
David Teigland 已提交
739 740 741 742 743 744 745 746 747 748
}

/**
 * gfs2_log_shutdown - write a shutdown header into a journal
 * @sdp: the filesystem
 *
 */

void gfs2_log_shutdown(struct gfs2_sbd *sdp)
{
749
	down_write(&sdp->sd_log_flush_lock);
D
David Teigland 已提交
750 751 752 753

	gfs2_assert_withdraw(sdp, !sdp->sd_log_blks_reserved);
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_gl);
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_buf);
754
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_jdata);
D
David Teigland 已提交
755 756 757 758 759 760 761 762
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_rg);
	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_databuf);
	gfs2_assert_withdraw(sdp, list_empty(&sdp->sd_ail1_list));

	sdp->sd_log_flush_head = sdp->sd_log_head;
	sdp->sd_log_flush_wrapped = 0;

763 764
	log_write_header(sdp, GFS2_LOG_HEAD_UNMOUNT,
			 (sdp->sd_log_tail == current_tail(sdp)) ? 0 : PULL);
D
David Teigland 已提交
765

766 767 768
	gfs2_assert_warn(sdp, sdp->sd_log_blks_free == sdp->sd_jdesc->jd_blocks);
	gfs2_assert_warn(sdp, sdp->sd_log_head == sdp->sd_log_tail);
	gfs2_assert_warn(sdp, list_empty(&sdp->sd_ail2_list));
D
David Teigland 已提交
769 770 771 772

	sdp->sd_log_head = sdp->sd_log_flush_head;
	sdp->sd_log_tail = sdp->sd_log_head;

773
	up_write(&sdp->sd_log_flush_lock);
D
David Teigland 已提交
774 775
}

776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793

/**
 * gfs2_meta_syncfs - sync all the buffers in a filesystem
 * @sdp: the filesystem
 *
 */

void gfs2_meta_syncfs(struct gfs2_sbd *sdp)
{
	gfs2_log_flush(sdp, NULL);
	for (;;) {
		gfs2_ail1_start(sdp, DIO_ALL);
		if (gfs2_ail1_empty(sdp, DIO_ALL))
			break;
		msleep(10);
	}
}