transaction.c 79.7 KB
Newer Older
1
/*
2
 * linux/fs/jbd2/transaction.c
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
 *
 * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
 *
 * Copyright 1998 Red Hat corp --- All Rights Reserved
 *
 * This file is part of the Linux kernel and is made available under
 * the terms of the GNU General Public License, version 2, or at your
 * option, any later version, incorporated herein by reference.
 *
 * Generic filesystem transaction handling code; part of the ext2fs
 * journaling system.
 *
 * This file manages transactions (compound commits managed by the
 * journaling code) and handles (individual atomic operations by the
 * filesystem).
 */

#include <linux/time.h>
#include <linux/fs.h>
22
#include <linux/jbd2.h>
23 24 25 26 27
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/highmem.h>
J
Josef Bacik 已提交
28
#include <linux/hrtimer.h>
29
#include <linux/backing-dev.h>
30
#include <linux/bug.h>
31
#include <linux/module.h>
32
#include <linux/sched/mm.h>
33

34 35
#include <trace/events/jbd2.h>

36
static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh);
37
static void __jbd2_journal_unfile_buffer(struct journal_head *jh);
38

39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
static struct kmem_cache *transaction_cache;
int __init jbd2_journal_init_transaction_cache(void)
{
	J_ASSERT(!transaction_cache);
	transaction_cache = kmem_cache_create("jbd2_transaction_s",
					sizeof(transaction_t),
					0,
					SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
					NULL);
	if (transaction_cache)
		return 0;
	return -ENOMEM;
}

void jbd2_journal_destroy_transaction_cache(void)
{
	if (transaction_cache) {
		kmem_cache_destroy(transaction_cache);
		transaction_cache = NULL;
	}
}

void jbd2_journal_free_transaction(transaction_t *transaction)
{
	if (unlikely(ZERO_OR_NULL_PTR(transaction)))
		return;
	kmem_cache_free(transaction_cache, transaction);
}

68
/*
69
 * jbd2_get_transaction: obtain a new transaction_t object.
70 71 72 73 74 75 76 77 78 79 80 81 82 83
 *
 * Simply allocate and initialise a new transaction.  Create it in
 * RUNNING state and add it to the current journal (which should not
 * have an existing running transaction: we only make a new transaction
 * once we have started to commit the old one).
 *
 * Preconditions:
 *	The journal MUST be locked.  We don't perform atomic mallocs on the
 *	new transaction	and we can't block without protecting against other
 *	processes trying to touch the journal while it is in transition.
 *
 */

static transaction_t *
84
jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
85 86 87
{
	transaction->t_journal = journal;
	transaction->t_state = T_RUNNING;
J
Josef Bacik 已提交
88
	transaction->t_start_time = ktime_get();
89 90 91
	transaction->t_tid = journal->j_transaction_sequence++;
	transaction->t_expires = jiffies + journal->j_commit_interval;
	spin_lock_init(&transaction->t_handle_lock);
92
	atomic_set(&transaction->t_updates, 0);
J
Jan Kara 已提交
93 94
	atomic_set(&transaction->t_outstanding_credits,
		   atomic_read(&journal->j_reserved_credits));
95
	atomic_set(&transaction->t_handle_count, 0);
96
	INIT_LIST_HEAD(&transaction->t_inode_list);
97
	INIT_LIST_HEAD(&transaction->t_private_list);
98 99

	/* Set up the commit timer for the new transaction. */
100
	journal->j_commit_timer.expires = round_jiffies_up(transaction->t_expires);
101 102 103 104
	add_timer(&journal->j_commit_timer);

	J_ASSERT(journal->j_running_transaction == NULL);
	journal->j_running_transaction = transaction;
105 106
	transaction->t_max_wait = 0;
	transaction->t_start = jiffies;
107
	transaction->t_requested = 0;
108 109 110 111 112 113 114 115 116 117 118 119

	return transaction;
}

/*
 * Handle management.
 *
 * A handle_t is an object which represents a single atomic update to a
 * filesystem, and which tracks all of the modifications which form part
 * of that one update.
 */

120
/*
121
 * Update transaction's maximum wait time, if debugging is enabled.
122 123 124 125 126 127 128 129
 *
 * In order for t_max_wait to be reliable, it must be protected by a
 * lock.  But doing so will mean that start_this_handle() can not be
 * run in parallel on SMP systems, which limits our scalability.  So
 * unless debugging is enabled, we no longer update t_max_wait, which
 * means that maximum wait time reported by the jbd2_run_stats
 * tracepoint will always be zero.
 */
130 131
static inline void update_t_max_wait(transaction_t *transaction,
				     unsigned long ts)
132 133 134 135 136 137 138 139 140 141 142 143 144
{
#ifdef CONFIG_JBD2_DEBUG
	if (jbd2_journal_enable_debug &&
	    time_after(transaction->t_start, ts)) {
		ts = jbd2_time_diff(ts, transaction->t_start);
		spin_lock(&transaction->t_handle_lock);
		if (ts > transaction->t_max_wait)
			transaction->t_max_wait = ts;
		spin_unlock(&transaction->t_handle_lock);
	}
#endif
}

J
Jan Kara 已提交
145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162
/*
 * Wait until running transaction passes T_LOCKED state. Also starts the commit
 * if needed. The function expects running transaction to exist and releases
 * j_state_lock.
 */
static void wait_transaction_locked(journal_t *journal)
	__releases(journal->j_state_lock)
{
	DEFINE_WAIT(wait);
	int need_to_start;
	tid_t tid = journal->j_running_transaction->t_tid;

	prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
			TASK_UNINTERRUPTIBLE);
	need_to_start = !tid_geq(journal->j_commit_request, tid);
	read_unlock(&journal->j_state_lock);
	if (need_to_start)
		jbd2_log_start_commit(journal, tid);
163
	jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
	schedule();
	finish_wait(&journal->j_wait_transaction_locked, &wait);
}

static void sub_reserved_credits(journal_t *journal, int blocks)
{
	atomic_sub(blocks, &journal->j_reserved_credits);
	wake_up(&journal->j_wait_reserved);
}

/*
 * Wait until we can add credits for handle to the running transaction.  Called
 * with j_state_lock held for reading. Returns 0 if handle joined the running
 * transaction. Returns 1 if we had to wait, j_state_lock is dropped, and
 * caller must retry.
 */
static int add_transaction_credits(journal_t *journal, int blocks,
				   int rsv_blocks)
{
	transaction_t *t = journal->j_running_transaction;
	int needed;
	int total = blocks + rsv_blocks;

	/*
	 * If the current transaction is locked down for commit, wait
	 * for the lock to be released.
	 */
	if (t->t_state == T_LOCKED) {
		wait_transaction_locked(journal);
		return 1;
	}

	/*
	 * If there is not enough space left in the log to write all
	 * potential buffers requested by this operation, we need to
	 * stall pending a log checkpoint to free some more log space.
	 */
	needed = atomic_add_return(total, &t->t_outstanding_credits);
	if (needed > journal->j_max_transaction_buffers) {
		/*
		 * If the current transaction is already too large,
		 * then start to commit it: we can then go back and
		 * attach this handle to a new transaction.
		 */
		atomic_sub(total, &t->t_outstanding_credits);
209 210 211 212 213 214 215 216

		/*
		 * Is the number of reserved credits in the current transaction too
		 * big to fit this handle? Wait until reserved credits are freed.
		 */
		if (atomic_read(&journal->j_reserved_credits) + total >
		    journal->j_max_transaction_buffers) {
			read_unlock(&journal->j_state_lock);
217
			jbd2_might_wait_for_commit(journal);
218 219 220 221 222 223
			wait_event(journal->j_wait_reserved,
				   atomic_read(&journal->j_reserved_credits) + total <=
				   journal->j_max_transaction_buffers);
			return 1;
		}

J
Jan Kara 已提交
224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241
		wait_transaction_locked(journal);
		return 1;
	}

	/*
	 * The commit code assumes that it can get enough log space
	 * without forcing a checkpoint.  This is *critical* for
	 * correctness: a checkpoint of a buffer which is also
	 * associated with a committing transaction creates a deadlock,
	 * so commit simply cannot force through checkpoints.
	 *
	 * We must therefore ensure the necessary space in the journal
	 * *before* starting to dirty potentially checkpointed buffers
	 * in the new transaction.
	 */
	if (jbd2_log_space_left(journal) < jbd2_space_needed(journal)) {
		atomic_sub(total, &t->t_outstanding_credits);
		read_unlock(&journal->j_state_lock);
242
		jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
		write_lock(&journal->j_state_lock);
		if (jbd2_log_space_left(journal) < jbd2_space_needed(journal))
			__jbd2_log_wait_for_space(journal);
		write_unlock(&journal->j_state_lock);
		return 1;
	}

	/* No reservation? We are done... */
	if (!rsv_blocks)
		return 0;

	needed = atomic_add_return(rsv_blocks, &journal->j_reserved_credits);
	/* We allow at most half of a transaction to be reserved */
	if (needed > journal->j_max_transaction_buffers / 2) {
		sub_reserved_credits(journal, rsv_blocks);
		atomic_sub(total, &t->t_outstanding_credits);
		read_unlock(&journal->j_state_lock);
260
		jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
261 262 263 264 265 266 267 268
		wait_event(journal->j_wait_reserved,
			 atomic_read(&journal->j_reserved_credits) + rsv_blocks
			 <= journal->j_max_transaction_buffers / 2);
		return 1;
	}
	return 0;
}

269 270 271 272 273 274 275
/*
 * start_this_handle: Given a handle, deal with any locking or stalling
 * needed to make sure that there is enough journal space for the handle
 * to begin.  Attach the handle to a transaction and set up the
 * transaction's buffer credits.
 */

276
static int start_this_handle(journal_t *journal, handle_t *handle,
D
Dan Carpenter 已提交
277
			     gfp_t gfp_mask)
278
{
279
	transaction_t	*transaction, *new_transaction = NULL;
J
Jan Kara 已提交
280 281
	int		blocks = handle->h_buffer_credits;
	int		rsv_blocks = 0;
282
	unsigned long ts = jiffies;
283

284 285 286
	if (handle->h_rsv_handle)
		rsv_blocks = handle->h_rsv_handle->h_buffer_credits;

J
Jan Kara 已提交
287
	/*
288 289 290
	 * Limit the number of reserved credits to 1/2 of maximum transaction
	 * size and limit the number of total credits to not exceed maximum
	 * transaction size per operation.
J
Jan Kara 已提交
291
	 */
292 293 294 295 296 297 298
	if ((rsv_blocks > journal->j_max_transaction_buffers / 2) ||
	    (rsv_blocks + blocks > journal->j_max_transaction_buffers)) {
		printk(KERN_ERR "JBD2: %s wants too many credits "
		       "credits:%d rsv_credits:%d max:%d\n",
		       current->comm, blocks, rsv_blocks,
		       journal->j_max_transaction_buffers);
		WARN_ON(1);
299
		return -ENOSPC;
300 301 302 303
	}

alloc_transaction:
	if (!journal->j_running_transaction) {
304 305 306 307 308 309
		/*
		 * If __GFP_FS is not present, then we may be being called from
		 * inside the fs writeback layer, so we MUST NOT fail.
		 */
		if ((gfp_mask & __GFP_FS) == 0)
			gfp_mask |= __GFP_NOFAIL;
310 311
		new_transaction = kmem_cache_zalloc(transaction_cache,
						    gfp_mask);
312
		if (!new_transaction)
313
			return -ENOMEM;
314 315 316 317 318 319 320 321
	}

	jbd_debug(3, "New handle %p going live.\n", handle);

	/*
	 * We need to hold j_state_lock until t_updates has been incremented,
	 * for proper journal barrier handling
	 */
322 323
repeat:
	read_lock(&journal->j_state_lock);
324
	BUG_ON(journal->j_flags & JBD2_UNMOUNT);
325
	if (is_journal_aborted(journal) ||
326
	    (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) {
327
		read_unlock(&journal->j_state_lock);
328
		jbd2_journal_free_transaction(new_transaction);
329
		return -EROFS;
330 331
	}

J
Jan Kara 已提交
332 333 334 335 336 337
	/*
	 * Wait on the journal's transaction barrier if necessary. Specifically
	 * we allow reserved handles to proceed because otherwise commit could
	 * deadlock on page writeback not being able to complete.
	 */
	if (!handle->h_reserved && journal->j_barrier_count) {
338
		read_unlock(&journal->j_state_lock);
339 340 341 342 343 344
		wait_event(journal->j_wait_transaction_locked,
				journal->j_barrier_count == 0);
		goto repeat;
	}

	if (!journal->j_running_transaction) {
345 346
		read_unlock(&journal->j_state_lock);
		if (!new_transaction)
347
			goto alloc_transaction;
348
		write_lock(&journal->j_state_lock);
349
		if (!journal->j_running_transaction &&
J
Jan Kara 已提交
350
		    (handle->h_reserved || !journal->j_barrier_count)) {
351 352
			jbd2_get_transaction(journal, new_transaction);
			new_transaction = NULL;
353
		}
354 355
		write_unlock(&journal->j_state_lock);
		goto repeat;
356 357 358 359
	}

	transaction = journal->j_running_transaction;

J
Jan Kara 已提交
360 361 362 363 364
	if (!handle->h_reserved) {
		/* We may have dropped j_state_lock - restart in that case */
		if (add_transaction_credits(journal, blocks, rsv_blocks))
			goto repeat;
	} else {
365
		/*
J
Jan Kara 已提交
366 367 368
		 * We have handle reserved so we are allowed to join T_LOCKED
		 * transaction and we don't have to check for transaction size
		 * and journal space.
369
		 */
J
Jan Kara 已提交
370 371
		sub_reserved_credits(journal, blocks);
		handle->h_reserved = 0;
372 373 374
	}

	/* OK, account for the buffers that this operation expects to
375 376
	 * use and add the handle to the running transaction. 
	 */
377
	update_t_max_wait(transaction, ts);
378
	handle->h_transaction = transaction;
J
Jan Kara 已提交
379
	handle->h_requested_credits = blocks;
380
	handle->h_start_jiffies = jiffies;
381
	atomic_inc(&transaction->t_updates);
382
	atomic_inc(&transaction->t_handle_count);
J
Jan Kara 已提交
383 384
	jbd_debug(4, "Handle %p given %d credits (total %d, free %lu)\n",
		  handle, blocks,
385
		  atomic_read(&transaction->t_outstanding_credits),
386
		  jbd2_log_space_left(journal));
387
	read_unlock(&journal->j_state_lock);
388
	current->journal_info = handle;
389

390
	rwsem_acquire_read(&journal->j_trans_commit_map, 0, 0, _THIS_IP_);
391
	jbd2_journal_free_transaction(new_transaction);
392 393 394 395 396
	/*
	 * Ensure that no allocations done while the transaction is open are
	 * going to recurse back to the fs layer.
	 */
	handle->saved_alloc_context = memalloc_nofs_save();
397
	return 0;
398 399 400 401 402
}

/* Allocate a new handle.  This should probably be in a slab... */
static handle_t *new_handle(int nblocks)
{
M
Mingming Cao 已提交
403
	handle_t *handle = jbd2_alloc_handle(GFP_NOFS);
404 405 406 407 408 409 410 411
	if (!handle)
		return NULL;
	handle->h_buffer_credits = nblocks;
	handle->h_ref = 1;

	return handle;
}

J
Jan Kara 已提交
412 413 414
handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
			      gfp_t gfp_mask, unsigned int type,
			      unsigned int line_no)
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
{
	handle_t *handle = journal_current_handle();
	int err;

	if (!journal)
		return ERR_PTR(-EROFS);

	if (handle) {
		J_ASSERT(handle->h_transaction->t_journal == journal);
		handle->h_ref++;
		return handle;
	}

	handle = new_handle(nblocks);
	if (!handle)
		return ERR_PTR(-ENOMEM);
J
Jan Kara 已提交
431 432 433 434 435 436 437 438 439 440 441 442
	if (rsv_blocks) {
		handle_t *rsv_handle;

		rsv_handle = new_handle(rsv_blocks);
		if (!rsv_handle) {
			jbd2_free_handle(handle);
			return ERR_PTR(-ENOMEM);
		}
		rsv_handle->h_reserved = 1;
		rsv_handle->h_journal = journal;
		handle->h_rsv_handle = rsv_handle;
	}
443

444
	err = start_this_handle(journal, handle, gfp_mask);
445
	if (err < 0) {
J
Jan Kara 已提交
446 447
		if (handle->h_rsv_handle)
			jbd2_free_handle(handle->h_rsv_handle);
M
Mingming Cao 已提交
448
		jbd2_free_handle(handle);
449
		return ERR_PTR(err);
450
	}
451 452 453 454 455
	handle->h_type = type;
	handle->h_line_no = line_no;
	trace_jbd2_handle_start(journal->j_fs_dev->bd_dev,
				handle->h_transaction->t_tid, type,
				line_no, nblocks);
456

457 458
	return handle;
}
459 460 461
EXPORT_SYMBOL(jbd2__journal_start);


462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480
/**
 * handle_t *jbd2_journal_start() - Obtain a new handle.
 * @journal: Journal to start transaction on.
 * @nblocks: number of block buffer we might modify
 *
 * We make sure that the transaction can guarantee at least nblocks of
 * modified buffers in the log.  We block until the log can guarantee
 * that much space. Additionally, if rsv_blocks > 0, we also create another
 * handle with rsv_blocks reserved blocks in the journal. This handle is
 * is stored in h_rsv_handle. It is not attached to any particular transaction
 * and thus doesn't block transaction commit. If the caller uses this reserved
 * handle, it has to set h_rsv_handle to NULL as otherwise jbd2_journal_stop()
 * on the parent handle will dispose the reserved one. Reserved handle has to
 * be converted to a normal handle using jbd2_journal_start_reserved() before
 * it can be used.
 *
 * Return a pointer to a newly allocated handle, or an ERR_PTR() value
 * on failure.
 */
481 482
handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
{
J
Jan Kara 已提交
483
	return jbd2__journal_start(journal, nblocks, 0, GFP_NOFS, 0, 0);
484 485 486
}
EXPORT_SYMBOL(jbd2_journal_start);

J
Jan Kara 已提交
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
void jbd2_journal_free_reserved(handle_t *handle)
{
	journal_t *journal = handle->h_journal;

	WARN_ON(!handle->h_reserved);
	sub_reserved_credits(journal, handle->h_buffer_credits);
	jbd2_free_handle(handle);
}
EXPORT_SYMBOL(jbd2_journal_free_reserved);

/**
 * int jbd2_journal_start_reserved(handle_t *handle) - start reserved handle
 * @handle: handle to start
 *
 * Start handle that has been previously reserved with jbd2_journal_reserve().
 * This attaches @handle to the running transaction (or creates one if there's
 * not transaction running). Unlike jbd2_journal_start() this function cannot
 * block on journal commit, checkpointing, or similar stuff. It can block on
 * memory allocation or frozen journal though.
 *
 * Return 0 on success, non-zero on error - handle is freed in that case.
 */
int jbd2_journal_start_reserved(handle_t *handle, unsigned int type,
				unsigned int line_no)
{
	journal_t *journal = handle->h_journal;
	int ret = -EIO;

	if (WARN_ON(!handle->h_reserved)) {
		/* Someone passed in normal handle? Just stop it. */
		jbd2_journal_stop(handle);
		return ret;
	}
	/*
	 * Usefulness of mixing of reserved and unreserved handles is
	 * questionable. So far nobody seems to need it so just error out.
	 */
	if (WARN_ON(current->journal_info)) {
		jbd2_journal_free_reserved(handle);
		return ret;
	}

	handle->h_journal = NULL;
	/*
	 * GFP_NOFS is here because callers are likely from writeback or
	 * similarly constrained call sites
	 */
	ret = start_this_handle(journal, handle, GFP_NOFS);
535
	if (ret < 0) {
J
Jan Kara 已提交
536
		jbd2_journal_free_reserved(handle);
537 538
		return ret;
	}
J
Jan Kara 已提交
539 540
	handle->h_type = type;
	handle->h_line_no = line_no;
541
	return 0;
J
Jan Kara 已提交
542 543
}
EXPORT_SYMBOL(jbd2_journal_start_reserved);
544 545

/**
546
 * int jbd2_journal_extend() - extend buffer credits.
547 548 549 550 551
 * @handle:  handle to 'extend'
 * @nblocks: nr blocks to try to extend by.
 *
 * Some transactions, such as large extends and truncates, can be done
 * atomically all at once or in several stages.  The operation requests
552
 * a credit for a number of buffer modifications in advance, but can
553 554
 * extend its credit if it needs more.
 *
555
 * jbd2_journal_extend tries to give the running handle more buffer credits.
556 557 558 559 560 561 562 563 564
 * It does not guarantee that allocation - this is a best-effort only.
 * The calling process MUST be able to deal cleanly with a failure to
 * extend here.
 *
 * Return 0 on success, non-zero on failure.
 *
 * return code < 0 implies an error
 * return code > 0 implies normal transaction-full status.
 */
565
int jbd2_journal_extend(handle_t *handle, int nblocks)
566 567
{
	transaction_t *transaction = handle->h_transaction;
568
	journal_t *journal;
569 570 571 572
	int result;
	int wanted;

	if (is_handle_aborted(handle))
573 574
		return -EROFS;
	journal = transaction->t_journal;
575 576 577

	result = 1;

578
	read_lock(&journal->j_state_lock);
579 580

	/* Don't extend a locked-down transaction! */
581
	if (transaction->t_state != T_RUNNING) {
582 583 584 585 586 587
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction not running\n", handle, nblocks);
		goto error_out;
	}

	spin_lock(&transaction->t_handle_lock);
588 589
	wanted = atomic_add_return(nblocks,
				   &transaction->t_outstanding_credits);
590 591 592 593

	if (wanted > journal->j_max_transaction_buffers) {
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction too large\n", handle, nblocks);
594
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
595 596 597
		goto unlock;
	}

598 599
	if (wanted + (wanted >> JBD2_CONTROL_BLOCKS_SHIFT) >
	    jbd2_log_space_left(journal)) {
600 601
		jbd_debug(3, "denied handle %p %d blocks: "
			  "insufficient log space\n", handle, nblocks);
602
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
603 604 605
		goto unlock;
	}

606
	trace_jbd2_handle_extend(journal->j_fs_dev->bd_dev,
607
				 transaction->t_tid,
608 609 610 611
				 handle->h_type, handle->h_line_no,
				 handle->h_buffer_credits,
				 nblocks);

612
	handle->h_buffer_credits += nblocks;
613
	handle->h_requested_credits += nblocks;
614 615 616 617 618 619
	result = 0;

	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
unlock:
	spin_unlock(&transaction->t_handle_lock);
error_out:
620
	read_unlock(&journal->j_state_lock);
621 622 623 624 625
	return result;
}


/**
626
 * int jbd2_journal_restart() - restart a handle .
627 628 629 630 631 632
 * @handle:  handle to restart
 * @nblocks: nr credits requested
 *
 * Restart a handle for a multi-transaction filesystem
 * operation.
 *
633 634
 * If the jbd2_journal_extend() call above fails to grant new buffer credits
 * to a running handle, a call to jbd2_journal_restart will commit the
635
 * handle's transaction so far and reattach the handle to a new
636
 * transaction capable of guaranteeing the requested number of
J
Jan Kara 已提交
637 638
 * credits. We preserve reserved handle if there's any attached to the
 * passed in handle.
639
 */
D
Dan Carpenter 已提交
640
int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask)
641 642
{
	transaction_t *transaction = handle->h_transaction;
643
	journal_t *journal;
644 645
	tid_t		tid;
	int		need_to_start, ret;
646 647 648 649 650

	/* If we've had an abort of any type, don't even think about
	 * actually doing the restart! */
	if (is_handle_aborted(handle))
		return 0;
651
	journal = transaction->t_journal;
652 653 654 655 656

	/*
	 * First unlink the handle from its current transaction, and start the
	 * commit on that.
	 */
657
	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
658 659
	J_ASSERT(journal_current_handle() == handle);

660
	read_lock(&journal->j_state_lock);
661
	spin_lock(&transaction->t_handle_lock);
662 663
	atomic_sub(handle->h_buffer_credits,
		   &transaction->t_outstanding_credits);
J
Jan Kara 已提交
664 665 666 667
	if (handle->h_rsv_handle) {
		sub_reserved_credits(journal,
				     handle->h_rsv_handle->h_buffer_credits);
	}
668
	if (atomic_dec_and_test(&transaction->t_updates))
669
		wake_up(&journal->j_wait_updates);
670
	tid = transaction->t_tid;
671
	spin_unlock(&transaction->t_handle_lock);
672 673
	handle->h_transaction = NULL;
	current->journal_info = NULL;
674 675

	jbd_debug(2, "restarting handle %p\n", handle);
676
	need_to_start = !tid_geq(journal->j_commit_request, tid);
677
	read_unlock(&journal->j_state_lock);
678 679
	if (need_to_start)
		jbd2_log_start_commit(journal, tid);
680

681
	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
682
	handle->h_buffer_credits = nblocks;
683 684 685 686 687 688
	/*
	 * Restore the original nofs context because the journal restart
	 * is basically the same thing as journal stop and start.
	 * start_this_handle will start a new nofs context.
	 */
	memalloc_nofs_restore(handle->saved_alloc_context);
689
	ret = start_this_handle(journal, handle, gfp_mask);
690 691
	return ret;
}
692
EXPORT_SYMBOL(jbd2__journal_restart);
693 694


695 696 697 698 699 700
int jbd2_journal_restart(handle_t *handle, int nblocks)
{
	return jbd2__journal_restart(handle, nblocks, GFP_NOFS);
}
EXPORT_SYMBOL(jbd2_journal_restart);

701
/**
702
 * void jbd2_journal_lock_updates () - establish a transaction barrier.
703 704 705 706 707 708 709 710
 * @journal:  Journal to establish a barrier on.
 *
 * This locks out any further updates from being started, and blocks
 * until all existing updates have completed, returning only once the
 * journal is in a quiescent state with no updates running.
 *
 * The journal lock should not be held on entry.
 */
711
void jbd2_journal_lock_updates(journal_t *journal)
712 713 714
{
	DEFINE_WAIT(wait);

715 716
	jbd2_might_wait_for_commit(journal);

717
	write_lock(&journal->j_state_lock);
718 719
	++journal->j_barrier_count;

J
Jan Kara 已提交
720 721 722 723 724 725 726 727
	/* Wait until there are no reserved handles */
	if (atomic_read(&journal->j_reserved_credits)) {
		write_unlock(&journal->j_state_lock);
		wait_event(journal->j_wait_reserved,
			   atomic_read(&journal->j_reserved_credits) == 0);
		write_lock(&journal->j_state_lock);
	}

728 729 730 731 732 733 734 735
	/* Wait until there are no running updates */
	while (1) {
		transaction_t *transaction = journal->j_running_transaction;

		if (!transaction)
			break;

		spin_lock(&transaction->t_handle_lock);
736 737
		prepare_to_wait(&journal->j_wait_updates, &wait,
				TASK_UNINTERRUPTIBLE);
738
		if (!atomic_read(&transaction->t_updates)) {
739
			spin_unlock(&transaction->t_handle_lock);
740
			finish_wait(&journal->j_wait_updates, &wait);
741 742 743
			break;
		}
		spin_unlock(&transaction->t_handle_lock);
744
		write_unlock(&journal->j_state_lock);
745 746
		schedule();
		finish_wait(&journal->j_wait_updates, &wait);
747
		write_lock(&journal->j_state_lock);
748
	}
749
	write_unlock(&journal->j_state_lock);
750 751 752

	/*
	 * We have now established a barrier against other normal updates, but
753
	 * we also need to barrier against other jbd2_journal_lock_updates() calls
754 755 756 757 758 759 760
	 * to make sure that we serialise special journal-locked operations
	 * too.
	 */
	mutex_lock(&journal->j_barrier);
}

/**
761
 * void jbd2_journal_unlock_updates (journal_t* journal) - release barrier
762 763
 * @journal:  Journal to release the barrier on.
 *
764
 * Release a transaction barrier obtained with jbd2_journal_lock_updates().
765 766 767
 *
 * Should be called without the journal lock held.
 */
768
void jbd2_journal_unlock_updates (journal_t *journal)
769 770 771 772
{
	J_ASSERT(journal->j_barrier_count != 0);

	mutex_unlock(&journal->j_barrier);
773
	write_lock(&journal->j_state_lock);
774
	--journal->j_barrier_count;
775
	write_unlock(&journal->j_state_lock);
776 777 778
	wake_up(&journal->j_wait_transaction_locked);
}

779
static void warn_dirty_buffer(struct buffer_head *bh)
780
{
781
	printk(KERN_WARNING
782
	       "JBD2: Spotted dirty metadata buffer (dev = %pg, blocknr = %llu). "
783 784
	       "There's a risk of filesystem corruption in case of system "
	       "crash.\n",
785
	       bh->b_bdev, (unsigned long long)bh->b_blocknr);
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
/* Call t_frozen trigger and copy buffer data into jh->b_frozen_data. */
static void jbd2_freeze_jh_data(struct journal_head *jh)
{
	struct page *page;
	int offset;
	char *source;
	struct buffer_head *bh = jh2bh(jh);

	J_EXPECT_JH(jh, buffer_uptodate(bh), "Possible IO failure.\n");
	page = bh->b_page;
	offset = offset_in_page(bh->b_data);
	source = kmap_atomic(page);
	/* Fire data frozen trigger just before we copy the data */
	jbd2_buffer_frozen_trigger(jh, source + offset, jh->b_triggers);
	memcpy(jh->b_frozen_data, source + offset, bh->b_size);
	kunmap_atomic(source);

	/*
	 * Now that the frozen data is saved off, we need to store any matching
	 * triggers.
	 */
	jh->b_frozen_triggers = jh->b_triggers;
}

812 813 814 815 816 817 818 819 820 821 822 823 824 825 826
/*
 * If the buffer is already part of the current transaction, then there
 * is nothing we need to do.  If it is already part of a prior
 * transaction which we are still committing to disk, then we need to
 * make sure that we do not overwrite the old copy: we do copy-out to
 * preserve the copy going to disk.  We also account the buffer against
 * the handle's metadata buffer credits (unless the buffer is already
 * part of the transaction, that is).
 *
 */
static int
do_get_write_access(handle_t *handle, struct journal_head *jh,
			int force_copy)
{
	struct buffer_head *bh;
827
	transaction_t *transaction = handle->h_transaction;
828 829 830
	journal_t *journal;
	int error;
	char *frozen_buffer = NULL;
831
	unsigned long start_lock, time_lock;
832 833 834 835 836

	if (is_handle_aborted(handle))
		return -EROFS;
	journal = transaction->t_journal;

837
	jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy);
838 839 840 841 842 843 844

	JBUFFER_TRACE(jh, "entry");
repeat:
	bh = jh2bh(jh);

	/* @@@ Need to check for errors here at some point. */

845
 	start_lock = jiffies;
846 847 848
	lock_buffer(bh);
	jbd_lock_bh_state(bh);

849 850 851 852 853 854
	/* If it takes too long to lock the buffer, trace it */
	time_lock = jbd2_time_diff(start_lock, jiffies);
	if (time_lock > HZ/10)
		trace_jbd2_lock_buffer_stall(bh->b_bdev->bd_dev,
			jiffies_to_msecs(time_lock));

855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880
	/* We now hold the buffer lock so it is safe to query the buffer
	 * state.  Is the buffer dirty?
	 *
	 * If so, there are two possibilities.  The buffer may be
	 * non-journaled, and undergoing a quite legitimate writeback.
	 * Otherwise, it is journaled, and we don't expect dirty buffers
	 * in that state (the buffers should be marked JBD_Dirty
	 * instead.)  So either the IO is being done under our own
	 * control and this is a bug, or it's a third party IO such as
	 * dump(8) (which may leave the buffer scheduled for read ---
	 * ie. locked but not dirty) or tune2fs (which may actually have
	 * the buffer dirtied, ugh.)  */

	if (buffer_dirty(bh)) {
		/*
		 * First question: is this buffer already part of the current
		 * transaction or the existing committing transaction?
		 */
		if (jh->b_transaction) {
			J_ASSERT_JH(jh,
				jh->b_transaction == transaction ||
				jh->b_transaction ==
					journal->j_committing_transaction);
			if (jh->b_next_transaction)
				J_ASSERT_JH(jh, jh->b_next_transaction ==
							transaction);
881
			warn_dirty_buffer(bh);
882 883 884 885 886 887
		}
		/*
		 * In any case we need to clean the dirty flag and we must
		 * do it under the buffer lock to be sure we don't race
		 * with running write-out.
		 */
888 889 890
		JBUFFER_TRACE(jh, "Journalling dirty buffer");
		clear_buffer_dirty(bh);
		set_buffer_jbddirty(bh);
891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909
	}

	unlock_buffer(bh);

	error = -EROFS;
	if (is_handle_aborted(handle)) {
		jbd_unlock_bh_state(bh);
		goto out;
	}
	error = 0;

	/*
	 * The buffer is already part of this transaction if b_transaction or
	 * b_next_transaction points to it
	 */
	if (jh->b_transaction == transaction ||
	    jh->b_next_transaction == transaction)
		goto done;

910 911 912 913 914 915
	/*
	 * this is the first time this transaction is touching this buffer,
	 * reset the modified flag
	 */
       jh->b_modified = 0;

916 917 918 919 920 921 922 923 924
	/*
	 * If the buffer is not journaled right now, we need to make sure it
	 * doesn't get written to disk before the caller actually commits the
	 * new data
	 */
	if (!jh->b_transaction) {
		JBUFFER_TRACE(jh, "no transaction");
		J_ASSERT_JH(jh, !jh->b_next_transaction);
		JBUFFER_TRACE(jh, "file as BJ_Reserved");
925 926 927 928 929 930
		/*
		 * Make sure all stores to jh (b_modified, b_frozen_data) are
		 * visible before attaching it to the running transaction.
		 * Paired with barrier in jbd2_write_access_granted()
		 */
		smp_wmb();
931 932 933 934 935
		spin_lock(&journal->j_list_lock);
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
		spin_unlock(&journal->j_list_lock);
		goto done;
	}
936 937 938 939 940 941 942
	/*
	 * If there is already a copy-out version of this buffer, then we don't
	 * need to make another one
	 */
	if (jh->b_frozen_data) {
		JBUFFER_TRACE(jh, "has frozen data");
		J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
943
		goto attach_next;
944 945
	}

946 947 948
	JBUFFER_TRACE(jh, "owned by older transaction");
	J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
	J_ASSERT_JH(jh, jh->b_transaction == journal->j_committing_transaction);
949

950 951 952 953 954 955 956 957 958 959 960 961 962 963 964
	/*
	 * There is one case we have to be very careful about.  If the
	 * committing transaction is currently writing this buffer out to disk
	 * and has NOT made a copy-out, then we cannot modify the buffer
	 * contents at all right now.  The essence of copy-out is that it is
	 * the extra copy, not the primary copy, which gets journaled.  If the
	 * primary copy is already going to disk then we cannot do copy-out
	 * here.
	 */
	if (buffer_shadow(bh)) {
		JBUFFER_TRACE(jh, "on shadow: sleep");
		jbd_unlock_bh_state(bh);
		wait_on_bit_io(&bh->b_state, BH_Shadow, TASK_UNINTERRUPTIBLE);
		goto repeat;
	}
965

966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981
	/*
	 * Only do the copy if the currently-owning transaction still needs it.
	 * If buffer isn't on BJ_Metadata list, the committing transaction is
	 * past that stage (here we use the fact that BH_Shadow is set under
	 * bh_state lock together with refiling to BJ_Shadow list and at this
	 * point we know the buffer doesn't have BH_Shadow set).
	 *
	 * Subtle point, though: if this is a get_undo_access, then we will be
	 * relying on the frozen_data to contain the new value of the
	 * committed_data record after the transaction, so we HAVE to force the
	 * frozen_data copy in that case.
	 */
	if (jh->b_jlist == BJ_Metadata || force_copy) {
		JBUFFER_TRACE(jh, "generate frozen data");
		if (!frozen_buffer) {
			JBUFFER_TRACE(jh, "allocate memory for buffer");
982
			jbd_unlock_bh_state(bh);
983 984
			frozen_buffer = jbd2_alloc(jh2bh(jh)->b_size,
						   GFP_NOFS | __GFP_NOFAIL);
985
			goto repeat;
986
		}
987 988 989
		jh->b_frozen_data = frozen_buffer;
		frozen_buffer = NULL;
		jbd2_freeze_jh_data(jh);
990
	}
991 992 993 994 995 996 997
attach_next:
	/*
	 * Make sure all stores to jh (b_modified, b_frozen_data) are visible
	 * before attaching it to the running transaction. Paired with barrier
	 * in jbd2_write_access_granted()
	 */
	smp_wmb();
998
	jh->b_next_transaction = transaction;
999 1000 1001 1002 1003 1004 1005 1006

done:
	jbd_unlock_bh_state(bh);

	/*
	 * If we are about to journal a buffer, then any revoke pending on it is
	 * no longer valid
	 */
1007
	jbd2_journal_cancel_revoke(handle, jh);
1008 1009 1010

out:
	if (unlikely(frozen_buffer))	/* It's usually NULL */
M
Mingming Cao 已提交
1011
		jbd2_free(frozen_buffer, bh->b_size);
1012 1013 1014 1015 1016

	JBUFFER_TRACE(jh, "exit");
	return error;
}

1017
/* Fast check whether buffer is already attached to the required transaction */
1018 1019
static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh,
							bool undo)
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045
{
	struct journal_head *jh;
	bool ret = false;

	/* Dirty buffers require special handling... */
	if (buffer_dirty(bh))
		return false;

	/*
	 * RCU protects us from dereferencing freed pages. So the checks we do
	 * are guaranteed not to oops. However the jh slab object can get freed
	 * & reallocated while we work with it. So we have to be careful. When
	 * we see jh attached to the running transaction, we know it must stay
	 * so until the transaction is committed. Thus jh won't be freed and
	 * will be attached to the same bh while we run.  However it can
	 * happen jh gets freed, reallocated, and attached to the transaction
	 * just after we get pointer to it from bh. So we have to be careful
	 * and recheck jh still belongs to our bh before we return success.
	 */
	rcu_read_lock();
	if (!buffer_jbd(bh))
		goto out;
	/* This should be bh2jh() but that doesn't work with inline functions */
	jh = READ_ONCE(bh->b_private);
	if (!jh)
		goto out;
1046 1047 1048
	/* For undo access buffer must have data copied */
	if (undo && !jh->b_committed_data)
		goto out;
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069
	if (jh->b_transaction != handle->h_transaction &&
	    jh->b_next_transaction != handle->h_transaction)
		goto out;
	/*
	 * There are two reasons for the barrier here:
	 * 1) Make sure to fetch b_bh after we did previous checks so that we
	 * detect when jh went through free, realloc, attach to transaction
	 * while we were checking. Paired with implicit barrier in that path.
	 * 2) So that access to bh done after jbd2_write_access_granted()
	 * doesn't get reordered and see inconsistent state of concurrent
	 * do_get_write_access().
	 */
	smp_mb();
	if (unlikely(jh->b_bh != bh))
		goto out;
	ret = true;
out:
	rcu_read_unlock();
	return ret;
}

1070
/**
1071
 * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update.
1072 1073 1074
 * @handle: transaction to add buffer modifications to
 * @bh:     bh to be used for metadata writes
 *
1075
 * Returns: error code or 0 on success.
1076 1077
 *
 * In full data journalling mode the buffer may be of type BJ_AsyncData,
1078
 * because we're ``write()ing`` a buffer which is also part of a shared mapping.
1079 1080
 */

1081
int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh)
1082
{
1083
	struct journal_head *jh;
1084 1085
	int rc;

1086
	if (jbd2_write_access_granted(handle, bh, false))
1087 1088 1089
		return 0;

	jh = jbd2_journal_add_journal_head(bh);
1090 1091 1092 1093
	/* We do not want to get caught playing with fields which the
	 * log thread also manipulates.  Make sure that the buffer
	 * completes any outstanding IO before proceeding. */
	rc = do_get_write_access(handle, jh, 0);
1094
	jbd2_journal_put_journal_head(jh);
1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
	return rc;
}


/*
 * When the user wants to journal a newly created buffer_head
 * (ie. getblk() returned a new buffer and we are going to populate it
 * manually rather than reading off disk), then we need to keep the
 * buffer_head locked until it has been completely filled with new
 * data.  In this case, we should be able to make the assertion that
 * the bh is not already part of an existing transaction.
 *
 * The buffer should already be locked by the caller by this point.
 * There is no lock ranking violation: it was a newly created,
 * unlocked buffer beforehand. */

/**
1112
 * int jbd2_journal_get_create_access () - notify intent to use newly created bh
1113 1114 1115 1116 1117
 * @handle: transaction to new buffer to
 * @bh: new buffer.
 *
 * Call this if you create a new bh.
 */
1118
int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
1119 1120
{
	transaction_t *transaction = handle->h_transaction;
1121
	journal_t *journal;
1122
	struct journal_head *jh = jbd2_journal_add_journal_head(bh);
1123 1124 1125 1126 1127 1128
	int err;

	jbd_debug(5, "journal_head %p\n", jh);
	err = -EROFS;
	if (is_handle_aborted(handle))
		goto out;
1129
	journal = transaction->t_journal;
1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149
	err = 0;

	JBUFFER_TRACE(jh, "entry");
	/*
	 * The buffer may already belong to this transaction due to pre-zeroing
	 * in the filesystem's new_block code.  It may also be on the previous,
	 * committing transaction's lists, but it HAS to be in Forget state in
	 * that case: the transaction must have deleted the buffer for it to be
	 * reused here.
	 */
	jbd_lock_bh_state(bh);
	J_ASSERT_JH(jh, (jh->b_transaction == transaction ||
		jh->b_transaction == NULL ||
		(jh->b_transaction == journal->j_committing_transaction &&
			  jh->b_jlist == BJ_Forget)));

	J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
	J_ASSERT_JH(jh, buffer_locked(jh2bh(jh)));

	if (jh->b_transaction == NULL) {
1150 1151 1152 1153 1154 1155 1156 1157 1158
		/*
		 * Previous jbd2_journal_forget() could have left the buffer
		 * with jbddirty bit set because it was being committed. When
		 * the commit finished, we've filed the buffer for
		 * checkpointing and marked it dirty. Now we are reallocating
		 * the buffer so the transaction freeing it must have
		 * committed and so it's safe to clear the dirty bit.
		 */
		clear_buffer_dirty(jh2bh(jh));
1159 1160 1161
		/* first access by this transaction */
		jh->b_modified = 0;

1162
		JBUFFER_TRACE(jh, "file as BJ_Reserved");
1163
		spin_lock(&journal->j_list_lock);
1164
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
1165
		spin_unlock(&journal->j_list_lock);
1166
	} else if (jh->b_transaction == journal->j_committing_transaction) {
1167 1168 1169
		/* first access by this transaction */
		jh->b_modified = 0;

1170
		JBUFFER_TRACE(jh, "set next transaction");
1171
		spin_lock(&journal->j_list_lock);
1172
		jh->b_next_transaction = transaction;
1173
		spin_unlock(&journal->j_list_lock);
1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184
	}
	jbd_unlock_bh_state(bh);

	/*
	 * akpm: I added this.  ext3_alloc_branch can pick up new indirect
	 * blocks which contain freed but then revoked metadata.  We need
	 * to cancel the revoke in case we end up freeing it yet again
	 * and the reallocating as data - this would cause a second revoke,
	 * which hits an assertion error.
	 */
	JBUFFER_TRACE(jh, "cancelling revoke");
1185
	jbd2_journal_cancel_revoke(handle, jh);
1186
out:
1187
	jbd2_journal_put_journal_head(jh);
1188 1189 1190 1191
	return err;
}

/**
1192
 * int jbd2_journal_get_undo_access() -  Notify intent to modify metadata with
1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203
 *     non-rewindable consequences
 * @handle: transaction
 * @bh: buffer to undo
 *
 * Sometimes there is a need to distinguish between metadata which has
 * been committed to disk and that which has not.  The ext3fs code uses
 * this for freeing and allocating space, we have to make sure that we
 * do not reuse freed space until the deallocation has been committed,
 * since if we overwrote that space we would make the delete
 * un-rewindable in case of a crash.
 *
1204
 * To deal with that, jbd2_journal_get_undo_access requests write access to a
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
 * buffer for parts of non-rewindable operations such as delete
 * operations on the bitmaps.  The journaling code must keep a copy of
 * the buffer's contents prior to the undo_access call until such time
 * as we know that the buffer has definitely been committed to disk.
 *
 * We never need to know which transaction the committed data is part
 * of, buffers touched here are guaranteed to be dirtied later and so
 * will be committed to a new transaction in due course, at which point
 * we can discard the old committed data pointer.
 *
 * Returns error number or 0 on success.
 */
1217
int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
1218 1219
{
	int err;
1220
	struct journal_head *jh;
1221 1222 1223
	char *committed_data = NULL;

	JBUFFER_TRACE(jh, "entry");
1224
	if (jbd2_write_access_granted(handle, bh, true))
1225
		return 0;
1226

1227
	jh = jbd2_journal_add_journal_head(bh);
1228 1229 1230 1231 1232 1233 1234 1235 1236 1237
	/*
	 * Do this first --- it can drop the journal lock, so we want to
	 * make sure that obtaining the committed_data is done
	 * atomically wrt. completion of any outstanding commits.
	 */
	err = do_get_write_access(handle, jh, 1);
	if (err)
		goto out;

repeat:
1238 1239 1240
	if (!jh->b_committed_data)
		committed_data = jbd2_alloc(jh2bh(jh)->b_size,
					    GFP_NOFS|__GFP_NOFAIL);
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257

	jbd_lock_bh_state(bh);
	if (!jh->b_committed_data) {
		/* Copy out the current buffer contents into the
		 * preserved, committed copy. */
		JBUFFER_TRACE(jh, "generate b_committed data");
		if (!committed_data) {
			jbd_unlock_bh_state(bh);
			goto repeat;
		}

		jh->b_committed_data = committed_data;
		committed_data = NULL;
		memcpy(jh->b_committed_data, bh->b_data, bh->b_size);
	}
	jbd_unlock_bh_state(bh);
out:
1258
	jbd2_journal_put_journal_head(jh);
1259
	if (unlikely(committed_data))
M
Mingming Cao 已提交
1260
		jbd2_free(committed_data, bh->b_size);
1261 1262 1263
	return err;
}

J
Joel Becker 已提交
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277
/**
 * void jbd2_journal_set_triggers() - Add triggers for commit writeout
 * @bh: buffer to trigger on
 * @type: struct jbd2_buffer_trigger_type containing the trigger(s).
 *
 * Set any triggers on this journal_head.  This is always safe, because
 * triggers for a committing buffer will be saved off, and triggers for
 * a running transaction will match the buffer in that transaction.
 *
 * Call with NULL to clear the triggers.
 */
void jbd2_journal_set_triggers(struct buffer_head *bh,
			       struct jbd2_buffer_trigger_type *type)
{
1278
	struct journal_head *jh = jbd2_journal_grab_journal_head(bh);
J
Joel Becker 已提交
1279

1280 1281
	if (WARN_ON(!jh))
		return;
J
Joel Becker 已提交
1282
	jh->b_triggers = type;
1283
	jbd2_journal_put_journal_head(jh);
J
Joel Becker 已提交
1284 1285
}

1286
void jbd2_buffer_frozen_trigger(struct journal_head *jh, void *mapped_data,
J
Joel Becker 已提交
1287 1288 1289 1290
				struct jbd2_buffer_trigger_type *triggers)
{
	struct buffer_head *bh = jh2bh(jh);

1291
	if (!triggers || !triggers->t_frozen)
J
Joel Becker 已提交
1292 1293
		return;

1294
	triggers->t_frozen(triggers, bh, mapped_data, bh->b_size);
J
Joel Becker 已提交
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
}

void jbd2_buffer_abort_trigger(struct journal_head *jh,
			       struct jbd2_buffer_trigger_type *triggers)
{
	if (!triggers || !triggers->t_abort)
		return;

	triggers->t_abort(triggers, jh2bh(jh));
}

1306
/**
1307
 * int jbd2_journal_dirty_metadata() -  mark a buffer as containing dirty metadata
1308 1309 1310 1311 1312 1313
 * @handle: transaction to add buffer to.
 * @bh: buffer to mark
 *
 * mark dirty metadata which needs to be journaled as part of the current
 * transaction.
 *
1314 1315 1316 1317
 * The buffer must have previously had jbd2_journal_get_write_access()
 * called so that it has a valid journal_head attached to the buffer
 * head.
 *
1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328
 * The buffer is placed on the transaction's metadata list and is marked
 * as belonging to the transaction.
 *
 * Returns error number or 0 on success.
 *
 * Special care needs to be taken if the buffer already belongs to the
 * current committing transaction (in which case we should have frozen
 * data present for that commit).  In that case, we don't relink the
 * buffer: that only gets done when the old transaction finally
 * completes its commit.
 */
1329
int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
1330 1331
{
	transaction_t *transaction = handle->h_transaction;
1332
	journal_t *journal;
1333
	struct journal_head *jh;
1334
	int ret = 0;
1335 1336

	if (is_handle_aborted(handle))
1337
		return -EROFS;
1338
	if (!buffer_jbd(bh)) {
1339 1340 1341
		ret = -EUCLEAN;
		goto out;
	}
1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
	/*
	 * We don't grab jh reference here since the buffer must be part
	 * of the running transaction.
	 */
	jh = bh2jh(bh);
	/*
	 * This and the following assertions are unreliable since we may see jh
	 * in inconsistent state unless we grab bh_state lock. But this is
	 * crucial to catch bugs so let's do a reliable check until the
	 * lockless handling is fully proven.
	 */
	if (jh->b_transaction != transaction &&
	    jh->b_next_transaction != transaction) {
		jbd_lock_bh_state(bh);
		J_ASSERT_JH(jh, jh->b_transaction == transaction ||
				jh->b_next_transaction == transaction);
		jbd_unlock_bh_state(bh);
	}
	if (jh->b_modified == 1) {
		/* If it's in our transaction it must be in BJ_Metadata list. */
		if (jh->b_transaction == transaction &&
		    jh->b_jlist != BJ_Metadata) {
			jbd_lock_bh_state(bh);
			J_ASSERT_JH(jh, jh->b_transaction != transaction ||
					jh->b_jlist == BJ_Metadata);
			jbd_unlock_bh_state(bh);
		}
		goto out;
	}

	journal = transaction->t_journal;
1373 1374
	jbd_debug(5, "journal_head %p\n", jh);
	JBUFFER_TRACE(jh, "entry");
1375 1376 1377 1378 1379 1380 1381 1382 1383 1384

	jbd_lock_bh_state(bh);

	if (jh->b_modified == 0) {
		/*
		 * This buffer's got modified and becoming part
		 * of the transaction. This needs to be done
		 * once a transaction -bzzz
		 */
		jh->b_modified = 1;
1385 1386 1387 1388
		if (handle->h_buffer_credits <= 0) {
			ret = -ENOSPC;
			goto out_unlock_bh;
		}
1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400
		handle->h_buffer_credits--;
	}

	/*
	 * fastpath, to avoid expensive locking.  If this buffer is already
	 * on the running transaction's metadata list there is nothing to do.
	 * Nobody can take it off again because there is a handle open.
	 * I _think_ we're OK here with SMP barriers - a mistaken decision will
	 * result in this test being false, so we go in and take the locks.
	 */
	if (jh->b_transaction == transaction && jh->b_jlist == BJ_Metadata) {
		JBUFFER_TRACE(jh, "fastpath");
1401 1402
		if (unlikely(jh->b_transaction !=
			     journal->j_running_transaction)) {
1403
			printk(KERN_ERR "JBD2: %s: "
1404
			       "jh->b_transaction (%llu, %p, %u) != "
1405
			       "journal->j_running_transaction (%p, %u)\n",
1406 1407 1408 1409 1410 1411 1412 1413 1414
			       journal->j_devname,
			       (unsigned long long) bh->b_blocknr,
			       jh->b_transaction,
			       jh->b_transaction ? jh->b_transaction->t_tid : 0,
			       journal->j_running_transaction,
			       journal->j_running_transaction ?
			       journal->j_running_transaction->t_tid : 0);
			ret = -EINVAL;
		}
1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427
		goto out_unlock_bh;
	}

	set_buffer_jbddirty(bh);

	/*
	 * Metadata already on the current transaction list doesn't
	 * need to be filed.  Metadata on another transaction's list must
	 * be committing, and will be refiled once the commit completes:
	 * leave it alone for now.
	 */
	if (jh->b_transaction != transaction) {
		JBUFFER_TRACE(jh, "already on other transaction");
1428 1429 1430 1431 1432 1433 1434 1435
		if (unlikely(((jh->b_transaction !=
			       journal->j_committing_transaction)) ||
			     (jh->b_next_transaction != transaction))) {
			printk(KERN_ERR "jbd2_journal_dirty_metadata: %s: "
			       "bad jh for block %llu: "
			       "transaction (%p, %u), "
			       "jh->b_transaction (%p, %u), "
			       "jh->b_next_transaction (%p, %u), jlist %u\n",
1436 1437
			       journal->j_devname,
			       (unsigned long long) bh->b_blocknr,
1438
			       transaction, transaction->t_tid,
1439
			       jh->b_transaction,
1440 1441
			       jh->b_transaction ?
			       jh->b_transaction->t_tid : 0,
1442 1443 1444
			       jh->b_next_transaction,
			       jh->b_next_transaction ?
			       jh->b_next_transaction->t_tid : 0,
1445 1446
			       jh->b_jlist);
			WARN_ON(1);
1447 1448
			ret = -EINVAL;
		}
1449 1450 1451 1452 1453 1454
		/* And this case is illegal: we can't reuse another
		 * transaction's data buffer, ever. */
		goto out_unlock_bh;
	}

	/* That test should have eliminated the following case: */
1455
	J_ASSERT_JH(jh, jh->b_frozen_data == NULL);
1456 1457 1458

	JBUFFER_TRACE(jh, "file as BJ_Metadata");
	spin_lock(&journal->j_list_lock);
1459
	__jbd2_journal_file_buffer(jh, transaction, BJ_Metadata);
1460 1461 1462 1463 1464
	spin_unlock(&journal->j_list_lock);
out_unlock_bh:
	jbd_unlock_bh_state(bh);
out:
	JBUFFER_TRACE(jh, "exit");
1465
	return ret;
1466 1467 1468
}

/**
1469
 * void jbd2_journal_forget() - bforget() for potentially-journaled buffers.
1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484
 * @handle: transaction handle
 * @bh:     bh to 'forget'
 *
 * We can only do the bforget if there are no commits pending against the
 * buffer.  If the buffer is dirty in the current running transaction we
 * can safely unlink it.
 *
 * bh may not be a journalled buffer at all - it may be a non-JBD
 * buffer which came off the hashtable.  Check for this.
 *
 * Decrements bh->b_count by one.
 *
 * Allow this call even if the handle has aborted --- it may be part of
 * the caller's cleanup after an abort.
 */
1485
int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
1486 1487
{
	transaction_t *transaction = handle->h_transaction;
1488
	journal_t *journal;
1489 1490 1491
	struct journal_head *jh;
	int drop_reserve = 0;
	int err = 0;
1492
	int was_modified = 0;
1493

1494 1495 1496 1497
	if (is_handle_aborted(handle))
		return -EROFS;
	journal = transaction->t_journal;

1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513
	BUFFER_TRACE(bh, "entry");

	jbd_lock_bh_state(bh);

	if (!buffer_jbd(bh))
		goto not_jbd;
	jh = bh2jh(bh);

	/* Critical error: attempting to delete a bitmap buffer, maybe?
	 * Don't do any jbd operations, and return an error. */
	if (!J_EXPECT_JH(jh, !jh->b_committed_data,
			 "inconsistent data on disk")) {
		err = -EIO;
		goto not_jbd;
	}

1514
	/* keep track of whether or not this transaction modified us */
1515 1516
	was_modified = jh->b_modified;

1517 1518 1519 1520 1521 1522
	/*
	 * The buffer's going from the transaction, we must drop
	 * all references -bzzz
	 */
	jh->b_modified = 0;

1523
	if (jh->b_transaction == transaction) {
1524 1525 1526 1527 1528 1529 1530 1531 1532 1533
		J_ASSERT_JH(jh, !jh->b_frozen_data);

		/* If we are forgetting a buffer which is already part
		 * of this transaction, then we can just drop it from
		 * the transaction immediately. */
		clear_buffer_dirty(bh);
		clear_buffer_jbddirty(bh);

		JBUFFER_TRACE(jh, "belongs to current transaction: unfile");

1534 1535 1536 1537 1538 1539
		/*
		 * we only want to drop a reference if this transaction
		 * modified the buffer
		 */
		if (was_modified)
			drop_reserve = 1;
1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552

		/*
		 * We are no longer going to journal this buffer.
		 * However, the commit of this transaction is still
		 * important to the buffer: the delete that we are now
		 * processing might obsolete an old log entry, so by
		 * committing, we can satisfy the buffer's checkpoint.
		 *
		 * So, if we have a checkpoint on the buffer, we should
		 * now refile the buffer on our BJ_Forget list so that
		 * we know to remove the checkpoint after we commit.
		 */

1553
		spin_lock(&journal->j_list_lock);
1554
		if (jh->b_cp_transaction) {
1555 1556
			__jbd2_journal_temp_unlink_buffer(jh);
			__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
1557
		} else {
1558
			__jbd2_journal_unfile_buffer(jh);
1559 1560 1561 1562 1563 1564 1565
			if (!buffer_jbd(bh)) {
				spin_unlock(&journal->j_list_lock);
				jbd_unlock_bh_state(bh);
				__bforget(bh);
				goto drop;
			}
		}
1566
		spin_unlock(&journal->j_list_lock);
1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577
	} else if (jh->b_transaction) {
		J_ASSERT_JH(jh, (jh->b_transaction ==
				 journal->j_committing_transaction));
		/* However, if the buffer is still owned by a prior
		 * (committing) transaction, we can't drop it yet... */
		JBUFFER_TRACE(jh, "belongs to older transaction");
		/* ... but we CAN drop it from the new transaction if we
		 * have also modified it since the original commit. */

		if (jh->b_next_transaction) {
			J_ASSERT(jh->b_next_transaction == transaction);
1578
			spin_lock(&journal->j_list_lock);
1579
			jh->b_next_transaction = NULL;
1580
			spin_unlock(&journal->j_list_lock);
1581 1582 1583 1584 1585 1586 1587

			/*
			 * only drop a reference if this transaction modified
			 * the buffer
			 */
			if (was_modified)
				drop_reserve = 1;
1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602
		}
	}

not_jbd:
	jbd_unlock_bh_state(bh);
	__brelse(bh);
drop:
	if (drop_reserve) {
		/* no need to reserve log space for this block -bzzz */
		handle->h_buffer_credits++;
	}
	return err;
}

/**
1603
 * int jbd2_journal_stop() - complete a transaction
1604
 * @handle: transaction to complete.
1605 1606 1607 1608 1609 1610 1611 1612
 *
 * All done for a particular handle.
 *
 * There is not much action needed here.  We just return any remaining
 * buffer credits to the transaction and remove the handle.  The only
 * complication is that we need to start a commit operation if the
 * filesystem is marked for synchronous update.
 *
1613
 * jbd2_journal_stop itself will not usually return an error, but it may
1614
 * do so in unusual circumstances.  In particular, expect it to
1615
 * return -EIO if a jbd2_journal_abort has been executed since the
1616 1617
 * transaction began.
 */
1618
int jbd2_journal_stop(handle_t *handle)
1619 1620
{
	transaction_t *transaction = handle->h_transaction;
1621 1622
	journal_t *journal;
	int err = 0, wait_for_commit = 0;
1623
	tid_t tid;
1624 1625
	pid_t pid;

1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641
	if (!transaction) {
		/*
		 * Handle is already detached from the transaction so
		 * there is nothing to do other than decrease a refcount,
		 * or free the handle if refcount drops to zero
		 */
		if (--handle->h_ref > 0) {
			jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
							 handle->h_ref);
			return err;
		} else {
			if (handle->h_rsv_handle)
				jbd2_free_handle(handle->h_rsv_handle);
			goto free_and_exit;
		}
	}
1642 1643
	journal = transaction->t_journal;

1644 1645 1646 1647
	J_ASSERT(journal_current_handle() == handle);

	if (is_handle_aborted(handle))
		err = -EIO;
1648
	else
1649
		J_ASSERT(atomic_read(&transaction->t_updates) > 0);
1650 1651 1652 1653 1654 1655 1656 1657

	if (--handle->h_ref > 0) {
		jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
			  handle->h_ref);
		return err;
	}

	jbd_debug(4, "Handle %p going down\n", handle);
1658
	trace_jbd2_handle_stats(journal->j_fs_dev->bd_dev,
1659
				transaction->t_tid,
1660 1661 1662 1663 1664
				handle->h_type, handle->h_line_no,
				jiffies - handle->h_start_jiffies,
				handle->h_sync, handle->h_requested_credits,
				(handle->h_requested_credits -
				 handle->h_buffer_credits));
1665 1666 1667 1668

	/*
	 * Implement synchronous transaction batching.  If the handle
	 * was synchronous, don't force a commit immediately.  Let's
J
Josef Bacik 已提交
1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
	 * yield and let another thread piggyback onto this
	 * transaction.  Keep doing that while new threads continue to
	 * arrive.  It doesn't cost much - we're about to run a commit
	 * and sleep on IO anyway.  Speeds up many-threaded, many-dir
	 * operations by 30x or more...
	 *
	 * We try and optimize the sleep time against what the
	 * underlying disk can do, instead of having a static sleep
	 * time.  This is useful for the case where our storage is so
	 * fast that it is more optimal to go ahead and force a flush
	 * and wait for the transaction to be committed than it is to
	 * wait for an arbitrary amount of time for new writers to
	 * join the transaction.  We achieve this by measuring how
	 * long it takes to commit a transaction, and compare it with
	 * how long this transaction has been running, and if run time
	 * < commit time then we sleep for the delta and commit.  This
	 * greatly helps super fast disks that would see slowdowns as
	 * more threads started doing fsyncs.
1687
	 *
J
Josef Bacik 已提交
1688 1689 1690 1691
	 * But don't do this if this process was the most recent one
	 * to perform a synchronous write.  We do this to detect the
	 * case where a single process is doing a stream of sync
	 * writes.  No point in waiting for joiners in that case.
1692 1693
	 *
	 * Setting max_batch_time to 0 disables this completely.
1694 1695
	 */
	pid = current->pid;
1696 1697
	if (handle->h_sync && journal->j_last_sync_writer != pid &&
	    journal->j_max_batch_time) {
J
Josef Bacik 已提交
1698 1699
		u64 commit_time, trans_time;

1700
		journal->j_last_sync_writer = pid;
J
Josef Bacik 已提交
1701

1702
		read_lock(&journal->j_state_lock);
J
Josef Bacik 已提交
1703
		commit_time = journal->j_average_commit_time;
1704
		read_unlock(&journal->j_state_lock);
J
Josef Bacik 已提交
1705 1706 1707 1708

		trans_time = ktime_to_ns(ktime_sub(ktime_get(),
						   transaction->t_start_time));

1709 1710
		commit_time = max_t(u64, commit_time,
				    1000*journal->j_min_batch_time);
J
Josef Bacik 已提交
1711
		commit_time = min_t(u64, commit_time,
1712
				    1000*journal->j_max_batch_time);
J
Josef Bacik 已提交
1713 1714 1715 1716 1717 1718 1719

		if (trans_time < commit_time) {
			ktime_t expires = ktime_add_ns(ktime_get(),
						       commit_time);
			set_current_state(TASK_UNINTERRUPTIBLE);
			schedule_hrtimeout(&expires, HRTIMER_MODE_ABS);
		}
1720 1721
	}

1722 1723
	if (handle->h_sync)
		transaction->t_synchronous_commit = 1;
1724
	current->journal_info = NULL;
1725 1726
	atomic_sub(handle->h_buffer_credits,
		   &transaction->t_outstanding_credits);
1727 1728 1729 1730 1731 1732 1733 1734

	/*
	 * If the handle is marked SYNC, we need to set another commit
	 * going!  We also want to force a commit if the current
	 * transaction is occupying too much of the log, or if the
	 * transaction is too old now.
	 */
	if (handle->h_sync ||
1735 1736 1737
	    (atomic_read(&transaction->t_outstanding_credits) >
	     journal->j_max_transaction_buffers) ||
	    time_after_eq(jiffies, transaction->t_expires)) {
1738 1739 1740 1741 1742 1743 1744
		/* Do this even for aborted journals: an abort still
		 * completes the commit thread, it just doesn't write
		 * anything to disk. */

		jbd_debug(2, "transaction too old, requesting commit for "
					"handle %p\n", handle);
		/* This is non-blocking */
1745
		jbd2_log_start_commit(journal, transaction->t_tid);
1746 1747

		/*
1748
		 * Special case: JBD2_SYNC synchronous updates require us
1749 1750 1751
		 * to wait for the commit to complete.
		 */
		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
1752
			wait_for_commit = 1;
1753 1754
	}

1755 1756
	/*
	 * Once we drop t_updates, if it goes to zero the transaction
L
Lucas De Marchi 已提交
1757
	 * could start committing on us and eventually disappear.  So
1758 1759 1760 1761 1762 1763 1764 1765 1766 1767
	 * once we do this, we must not dereference transaction
	 * pointer again.
	 */
	tid = transaction->t_tid;
	if (atomic_dec_and_test(&transaction->t_updates)) {
		wake_up(&journal->j_wait_updates);
		if (journal->j_barrier_count)
			wake_up(&journal->j_wait_transaction_locked);
	}

1768
	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
1769

1770 1771 1772
	if (wait_for_commit)
		err = jbd2_log_wait_commit(journal, tid);

J
Jan Kara 已提交
1773 1774
	if (handle->h_rsv_handle)
		jbd2_journal_free_reserved(handle->h_rsv_handle);
1775
free_and_exit:
1776 1777 1778 1779 1780
	/*
	 * Scope of the GFP_NOFS context is over here and so we can restore the
	 * original alloc context.
	 */
	memalloc_nofs_restore(handle->saved_alloc_context);
M
Mingming Cao 已提交
1781
	jbd2_free_handle(handle);
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 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840
	return err;
}

/*
 *
 * List management code snippets: various functions for manipulating the
 * transaction buffer lists.
 *
 */

/*
 * Append a buffer to a transaction list, given the transaction's list head
 * pointer.
 *
 * j_list_lock is held.
 *
 * jbd_lock_bh_state(jh2bh(jh)) is held.
 */

static inline void
__blist_add_buffer(struct journal_head **list, struct journal_head *jh)
{
	if (!*list) {
		jh->b_tnext = jh->b_tprev = jh;
		*list = jh;
	} else {
		/* Insert at the tail of the list to preserve order */
		struct journal_head *first = *list, *last = first->b_tprev;
		jh->b_tprev = last;
		jh->b_tnext = first;
		last->b_tnext = first->b_tprev = jh;
	}
}

/*
 * Remove a buffer from a transaction list, given the transaction's list
 * head pointer.
 *
 * Called with j_list_lock held, and the journal may not be locked.
 *
 * jbd_lock_bh_state(jh2bh(jh)) is held.
 */

static inline void
__blist_del_buffer(struct journal_head **list, struct journal_head *jh)
{
	if (*list == jh) {
		*list = jh->b_tnext;
		if (*list == jh)
			*list = NULL;
	}
	jh->b_tprev->b_tnext = jh->b_tnext;
	jh->b_tnext->b_tprev = jh->b_tprev;
}

/*
 * Remove a buffer from the appropriate transaction list.
 *
 * Note that this function can *change* the value of
1841 1842 1843 1844
 * bh->b_transaction->t_buffers, t_forget, t_shadow_list, t_log_list or
 * t_reserved_list.  If the caller is holding onto a copy of one of these
 * pointers, it could go bad.  Generally the caller needs to re-read the
 * pointer from the transaction_t.
1845
 *
1846
 * Called under j_list_lock.
1847
 */
1848
static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860
{
	struct journal_head **list = NULL;
	transaction_t *transaction;
	struct buffer_head *bh = jh2bh(jh);

	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
	transaction = jh->b_transaction;
	if (transaction)
		assert_spin_locked(&transaction->t_journal->j_list_lock);

	J_ASSERT_JH(jh, jh->b_jlist < BJ_Types);
	if (jh->b_jlist != BJ_None)
1861
		J_ASSERT_JH(jh, transaction != NULL);
1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883

	switch (jh->b_jlist) {
	case BJ_None:
		return;
	case BJ_Metadata:
		transaction->t_nr_buffers--;
		J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0);
		list = &transaction->t_buffers;
		break;
	case BJ_Forget:
		list = &transaction->t_forget;
		break;
	case BJ_Shadow:
		list = &transaction->t_shadow_list;
		break;
	case BJ_Reserved:
		list = &transaction->t_reserved_list;
		break;
	}

	__blist_del_buffer(list, jh);
	jh->b_jlist = BJ_None;
1884 1885 1886
	if (transaction && is_journal_aborted(transaction->t_journal))
		clear_buffer_jbddirty(bh);
	else if (test_clear_buffer_jbddirty(bh))
1887 1888 1889
		mark_buffer_dirty(bh);	/* Expose it to the VM */
}

1890 1891 1892 1893 1894 1895 1896 1897
/*
 * Remove buffer from all transactions.
 *
 * Called with bh_state lock and j_list_lock
 *
 * jh and bh may be already freed when this function returns.
 */
static void __jbd2_journal_unfile_buffer(struct journal_head *jh)
1898
{
1899
	__jbd2_journal_temp_unlink_buffer(jh);
1900
	jh->b_transaction = NULL;
1901
	jbd2_journal_put_journal_head(jh);
1902 1903
}

1904
void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh)
1905
{
1906 1907 1908 1909 1910
	struct buffer_head *bh = jh2bh(jh);

	/* Get reference so that buffer cannot be freed before we unlock it */
	get_bh(bh);
	jbd_lock_bh_state(bh);
1911
	spin_lock(&journal->j_list_lock);
1912
	__jbd2_journal_unfile_buffer(jh);
1913
	spin_unlock(&journal->j_list_lock);
1914 1915
	jbd_unlock_bh_state(bh);
	__brelse(bh);
1916 1917 1918
}

/*
1919
 * Called from jbd2_journal_try_to_free_buffers().
1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932
 *
 * Called under jbd_lock_bh_state(bh)
 */
static void
__journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh)
{
	struct journal_head *jh;

	jh = bh2jh(bh);

	if (buffer_locked(bh) || buffer_dirty(bh))
		goto out;

1933
	if (jh->b_next_transaction != NULL || jh->b_transaction != NULL)
1934 1935 1936
		goto out;

	spin_lock(&journal->j_list_lock);
1937
	if (jh->b_cp_transaction != NULL) {
1938
		/* written-back checkpointed metadata buffer */
1939 1940
		JBUFFER_TRACE(jh, "remove from checkpoint list");
		__jbd2_journal_remove_checkpoint(jh);
1941 1942 1943 1944 1945 1946 1947
	}
	spin_unlock(&journal->j_list_lock);
out:
	return;
}

/**
1948
 * int jbd2_journal_try_to_free_buffers() - try to free page buffers.
1949 1950
 * @journal: journal for operation
 * @page: to try and free
1951
 * @gfp_mask: we use the mask to detect how hard should we try to release
1952 1953
 * buffers. If __GFP_DIRECT_RECLAIM and __GFP_FS is set, we wait for commit
 * code to release the buffers.
1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966
 *
 *
 * For all the buffers on this page,
 * if they are fully written out ordered data, move them onto BUF_CLEAN
 * so try_to_free_buffers() can reap them.
 *
 * This function returns non-zero if we wish try_to_free_buffers()
 * to be called. We do this if the page is releasable by try_to_free_buffers().
 * We also do it if the page has locked or dirty buffers and the caller wants
 * us to perform sync or async writeout.
 *
 * This complicates JBD locking somewhat.  We aren't protected by the
 * BKL here.  We wish to remove the buffer from its committing or
1967
 * running transaction's ->t_datalist via __jbd2_journal_unfile_buffer.
1968 1969 1970 1971
 *
 * This may *change* the value of transaction_t->t_datalist, so anyone
 * who looks at t_datalist needs to lock against this function.
 *
1972 1973
 * Even worse, someone may be doing a jbd2_journal_dirty_data on this
 * buffer.  So we need to lock against that.  jbd2_journal_dirty_data()
1974 1975 1976 1977 1978 1979 1980 1981
 * will come out of the lock with the buffer dirty, which makes it
 * ineligible for release here.
 *
 * Who else is affected by this?  hmm...  Really the only contender
 * is do_get_write_access() - it could be looking at the buffer while
 * journal_try_to_free_buffer() is changing its state.  But that
 * cannot happen because we never reallocate freed data as metadata
 * while the data is part of a transaction.  Yes?
1982 1983
 *
 * Return 0 on failure, 1 on success
1984
 */
1985
int jbd2_journal_try_to_free_buffers(journal_t *journal,
1986
				struct page *page, gfp_t gfp_mask)
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
{
	struct buffer_head *head;
	struct buffer_head *bh;
	int ret = 0;

	J_ASSERT(PageLocked(page));

	head = page_buffers(page);
	bh = head;
	do {
		struct journal_head *jh;

		/*
		 * We take our own ref against the journal_head here to avoid
		 * having to add tons of locking around each instance of
2002
		 * jbd2_journal_put_journal_head().
2003
		 */
2004
		jh = jbd2_journal_grab_journal_head(bh);
2005 2006 2007 2008 2009
		if (!jh)
			continue;

		jbd_lock_bh_state(bh);
		__journal_try_to_free_buffer(journal, bh);
2010
		jbd2_journal_put_journal_head(jh);
2011 2012 2013 2014
		jbd_unlock_bh_state(bh);
		if (buffer_jbd(bh))
			goto busy;
	} while ((bh = bh->b_this_page) != head);
2015

2016
	ret = try_to_free_buffers(page);
2017

2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040
busy:
	return ret;
}

/*
 * This buffer is no longer needed.  If it is on an older transaction's
 * checkpoint list we need to record it on this transaction's forget list
 * to pin this buffer (and hence its checkpointing transaction) down until
 * this transaction commits.  If the buffer isn't on a checkpoint list, we
 * release it.
 * Returns non-zero if JBD no longer has an interest in the buffer.
 *
 * Called under j_list_lock.
 *
 * Called under jbd_lock_bh_state(bh).
 */
static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
{
	int may_free = 1;
	struct buffer_head *bh = jh2bh(jh);

	if (jh->b_cp_transaction) {
		JBUFFER_TRACE(jh, "on running+cp transaction");
2041
		__jbd2_journal_temp_unlink_buffer(jh);
2042 2043 2044 2045 2046 2047
		/*
		 * We don't want to write the buffer anymore, clear the
		 * bit so that we don't confuse checks in
		 * __journal_file_buffer
		 */
		clear_buffer_dirty(bh);
2048
		__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
2049 2050 2051
		may_free = 0;
	} else {
		JBUFFER_TRACE(jh, "on running transaction");
2052
		__jbd2_journal_unfile_buffer(jh);
2053 2054 2055 2056 2057
	}
	return may_free;
}

/*
2058
 * jbd2_journal_invalidatepage
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 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103
 *
 * This code is tricky.  It has a number of cases to deal with.
 *
 * There are two invariants which this code relies on:
 *
 * i_size must be updated on disk before we start calling invalidatepage on the
 * data.
 *
 *  This is done in ext3 by defining an ext3_setattr method which
 *  updates i_size before truncate gets going.  By maintaining this
 *  invariant, we can be sure that it is safe to throw away any buffers
 *  attached to the current transaction: once the transaction commits,
 *  we know that the data will not be needed.
 *
 *  Note however that we can *not* throw away data belonging to the
 *  previous, committing transaction!
 *
 * Any disk blocks which *are* part of the previous, committing
 * transaction (and which therefore cannot be discarded immediately) are
 * not going to be reused in the new running transaction
 *
 *  The bitmap committed_data images guarantee this: any block which is
 *  allocated in one transaction and removed in the next will be marked
 *  as in-use in the committed_data bitmap, so cannot be reused until
 *  the next transaction to delete the block commits.  This means that
 *  leaving committing buffers dirty is quite safe: the disk blocks
 *  cannot be reallocated to a different file and so buffer aliasing is
 *  not possible.
 *
 *
 * The above applies mainly to ordered data mode.  In writeback mode we
 * don't make guarantees about the order in which data hits disk --- in
 * particular we don't guarantee that new dirty data is flushed before
 * transaction commit --- so it is always safe just to discard data
 * immediately in that mode.  --sct
 */

/*
 * The journal_unmap_buffer helper function returns zero if the buffer
 * concerned remains pinned as an anonymous buffer belonging to an older
 * transaction.
 *
 * We're outside-transaction here.  Either or both of j_running_transaction
 * and j_committing_transaction may be NULL.
 */
2104 2105
static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
				int partial_page)
2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121
{
	transaction_t *transaction;
	struct journal_head *jh;
	int may_free = 1;

	BUFFER_TRACE(bh, "entry");

	/*
	 * It is safe to proceed here without the j_list_lock because the
	 * buffers cannot be stolen by try_to_free_buffers as long as we are
	 * holding the page lock. --sct
	 */

	if (!buffer_jbd(bh))
		goto zap_buffer_unlocked;

2122
	/* OK, we have data buffer in journaled mode */
2123
	write_lock(&journal->j_state_lock);
2124 2125 2126
	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);

2127
	jh = jbd2_journal_grab_journal_head(bh);
2128 2129 2130
	if (!jh)
		goto zap_buffer_no_jh;

2131 2132 2133 2134 2135 2136 2137 2138 2139 2140
	/*
	 * We cannot remove the buffer from checkpoint lists until the
	 * transaction adding inode to orphan list (let's call it T)
	 * is committed.  Otherwise if the transaction changing the
	 * buffer would be cleaned from the journal before T is
	 * committed, a crash will cause that the correct contents of
	 * the buffer will be lost.  On the other hand we have to
	 * clear the buffer dirty bit at latest at the moment when the
	 * transaction marking the buffer as freed in the filesystem
	 * structures is committed because from that moment on the
2141
	 * block can be reallocated and used by a different page.
2142 2143 2144
	 * Since the block hasn't been freed yet but the inode has
	 * already been added to orphan list, it is safe for us to add
	 * the buffer to BJ_Forget list of the newest transaction.
2145 2146 2147 2148 2149 2150 2151 2152
	 *
	 * Also we have to clear buffer_mapped flag of a truncated buffer
	 * because the buffer_head may be attached to the page straddling
	 * i_size (can happen only when blocksize < pagesize) and thus the
	 * buffer_head can be reused when the file is extended again. So we end
	 * up keeping around invalidated buffers attached to transactions'
	 * BJ_Forget list just to stop checkpointing code from cleaning up
	 * the transaction this buffer was modified in.
2153
	 */
2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166
	transaction = jh->b_transaction;
	if (transaction == NULL) {
		/* First case: not on any transaction.  If it
		 * has no checkpoint link, then we can zap it:
		 * it's a writeback-mode buffer so we don't care
		 * if it hits disk safely. */
		if (!jh->b_cp_transaction) {
			JBUFFER_TRACE(jh, "not on any transaction: zap");
			goto zap_buffer;
		}

		if (!buffer_dirty(bh)) {
			/* bdflush has written it.  We can drop it now */
2167
			__jbd2_journal_remove_checkpoint(jh);
2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179
			goto zap_buffer;
		}

		/* OK, it must be in the journal but still not
		 * written fully to disk: it's metadata or
		 * journaled data... */

		if (journal->j_running_transaction) {
			/* ... and once the current transaction has
			 * committed, the buffer won't be needed any
			 * longer. */
			JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget");
2180
			may_free = __dispose_buffer(jh,
2181
					journal->j_running_transaction);
2182
			goto zap_buffer;
2183 2184 2185 2186 2187 2188 2189
		} else {
			/* There is no currently-running transaction. So the
			 * orphan record which we wrote for this file must have
			 * passed into commit.  We must attach this buffer to
			 * the committing transaction, if it exists. */
			if (journal->j_committing_transaction) {
				JBUFFER_TRACE(jh, "give to committing trans");
2190
				may_free = __dispose_buffer(jh,
2191
					journal->j_committing_transaction);
2192
				goto zap_buffer;
2193 2194 2195 2196
			} else {
				/* The orphan record's transaction has
				 * committed.  We can cleanse this buffer */
				clear_buffer_jbddirty(bh);
2197
				__jbd2_journal_remove_checkpoint(jh);
2198 2199 2200 2201
				goto zap_buffer;
			}
		}
	} else if (transaction == journal->j_committing_transaction) {
2202
		JBUFFER_TRACE(jh, "on committing transaction");
2203
		/*
2204
		 * The buffer is committing, we simply cannot touch
2205 2206 2207 2208 2209 2210 2211 2212
		 * it. If the page is straddling i_size we have to wait
		 * for commit and try again.
		 */
		if (partial_page) {
			jbd2_journal_put_journal_head(jh);
			spin_unlock(&journal->j_list_lock);
			jbd_unlock_bh_state(bh);
			write_unlock(&journal->j_state_lock);
2213
			return -EBUSY;
2214 2215 2216 2217 2218 2219
		}
		/*
		 * OK, buffer won't be reachable after truncate. We just set
		 * j_next_transaction to the running transaction (if there is
		 * one) and mark buffer as freed so that commit code knows it
		 * should clear dirty bits when it is done with the buffer.
2220
		 */
2221
		set_buffer_freed(bh);
2222 2223
		if (journal->j_running_transaction && buffer_jbddirty(bh))
			jh->b_next_transaction = journal->j_running_transaction;
2224
		jbd2_journal_put_journal_head(jh);
2225 2226
		spin_unlock(&journal->j_list_lock);
		jbd_unlock_bh_state(bh);
2227
		write_unlock(&journal->j_state_lock);
2228 2229 2230 2231 2232 2233 2234 2235 2236
		return 0;
	} else {
		/* Good, the buffer belongs to the running transaction.
		 * We are writing our own transaction's data, not any
		 * previous one's, so it is safe to throw it away
		 * (remember that we expect the filesystem to have set
		 * i_size already for this truncate so recovery will not
		 * expose the disk blocks we are discarding here.) */
		J_ASSERT_JH(jh, transaction == journal->j_running_transaction);
2237
		JBUFFER_TRACE(jh, "on running transaction");
2238 2239 2240 2241
		may_free = __dispose_buffer(jh, transaction);
	}

zap_buffer:
2242 2243 2244 2245 2246 2247 2248 2249 2250
	/*
	 * This is tricky. Although the buffer is truncated, it may be reused
	 * if blocksize < pagesize and it is attached to the page straddling
	 * EOF. Since the buffer might have been added to BJ_Forget list of the
	 * running transaction, journal_get_write_access() won't clear
	 * b_modified and credit accounting gets confused. So clear b_modified
	 * here.
	 */
	jh->b_modified = 0;
2251
	jbd2_journal_put_journal_head(jh);
2252 2253 2254
zap_buffer_no_jh:
	spin_unlock(&journal->j_list_lock);
	jbd_unlock_bh_state(bh);
2255
	write_unlock(&journal->j_state_lock);
2256 2257 2258 2259 2260 2261
zap_buffer_unlocked:
	clear_buffer_dirty(bh);
	J_ASSERT_BH(bh, !buffer_jbddirty(bh));
	clear_buffer_mapped(bh);
	clear_buffer_req(bh);
	clear_buffer_new(bh);
2262 2263
	clear_buffer_delay(bh);
	clear_buffer_unwritten(bh);
2264 2265 2266 2267 2268
	bh->b_bdev = NULL;
	return may_free;
}

/**
2269
 * void jbd2_journal_invalidatepage()
2270 2271
 * @journal: journal to use for flush...
 * @page:    page to flush
2272 2273
 * @offset:  start of the range to invalidate
 * @length:  length of the range to invalidate
2274
 *
2275 2276 2277 2278
 * Reap page buffers containing data after in the specified range in page.
 * Can return -EBUSY if buffers are part of the committing transaction and
 * the page is straddling i_size. Caller then has to wait for current commit
 * and try again.
2279
 */
2280 2281
int jbd2_journal_invalidatepage(journal_t *journal,
				struct page *page,
2282 2283
				unsigned int offset,
				unsigned int length)
2284 2285
{
	struct buffer_head *head, *bh, *next;
2286
	unsigned int stop = offset + length;
2287
	unsigned int curr_off = 0;
2288
	int partial_page = (offset || length < PAGE_SIZE);
2289
	int may_free = 1;
2290
	int ret = 0;
2291 2292 2293 2294

	if (!PageLocked(page))
		BUG();
	if (!page_has_buffers(page))
2295
		return 0;
2296

2297
	BUG_ON(stop > PAGE_SIZE || stop < length);
2298

2299 2300 2301 2302 2303 2304 2305 2306 2307
	/* We will potentially be playing with lists other than just the
	 * data lists (especially for journaled data mode), so be
	 * cautious in our locking. */

	head = bh = page_buffers(page);
	do {
		unsigned int next_off = curr_off + bh->b_size;
		next = bh->b_this_page;

2308 2309 2310
		if (next_off > stop)
			return 0;

2311 2312 2313
		if (offset <= curr_off) {
			/* This block is wholly outside the truncation point */
			lock_buffer(bh);
2314
			ret = journal_unmap_buffer(journal, bh, partial_page);
2315
			unlock_buffer(bh);
2316 2317 2318
			if (ret < 0)
				return ret;
			may_free &= ret;
2319 2320 2321 2322 2323 2324
		}
		curr_off = next_off;
		bh = next;

	} while (bh != head);

2325
	if (!partial_page) {
2326 2327 2328
		if (may_free && try_to_free_buffers(page))
			J_ASSERT(!page_has_buffers(page));
	}
2329
	return 0;
2330 2331 2332 2333 2334
}

/*
 * File a buffer on the given transaction list.
 */
2335
void __jbd2_journal_file_buffer(struct journal_head *jh,
2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346
			transaction_t *transaction, int jlist)
{
	struct journal_head **list = NULL;
	int was_dirty = 0;
	struct buffer_head *bh = jh2bh(jh);

	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
	assert_spin_locked(&transaction->t_journal->j_list_lock);

	J_ASSERT_JH(jh, jh->b_jlist < BJ_Types);
	J_ASSERT_JH(jh, jh->b_transaction == transaction ||
2347
				jh->b_transaction == NULL);
2348 2349 2350 2351 2352 2353

	if (jh->b_transaction && jh->b_jlist == jlist)
		return;

	if (jlist == BJ_Metadata || jlist == BJ_Reserved ||
	    jlist == BJ_Shadow || jlist == BJ_Forget) {
2354 2355 2356 2357 2358 2359 2360 2361 2362
		/*
		 * For metadata buffers, we track dirty bit in buffer_jbddirty
		 * instead of buffer_dirty. We should not see a dirty bit set
		 * here because we clear it in do_get_write_access but e.g.
		 * tune2fs can modify the sb and set the dirty bit at any time
		 * so we try to gracefully handle that.
		 */
		if (buffer_dirty(bh))
			warn_dirty_buffer(bh);
2363 2364 2365 2366 2367 2368
		if (test_clear_buffer_dirty(bh) ||
		    test_clear_buffer_jbddirty(bh))
			was_dirty = 1;
	}

	if (jh->b_transaction)
2369
		__jbd2_journal_temp_unlink_buffer(jh);
2370 2371
	else
		jbd2_journal_grab_journal_head(bh);
2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400
	jh->b_transaction = transaction;

	switch (jlist) {
	case BJ_None:
		J_ASSERT_JH(jh, !jh->b_committed_data);
		J_ASSERT_JH(jh, !jh->b_frozen_data);
		return;
	case BJ_Metadata:
		transaction->t_nr_buffers++;
		list = &transaction->t_buffers;
		break;
	case BJ_Forget:
		list = &transaction->t_forget;
		break;
	case BJ_Shadow:
		list = &transaction->t_shadow_list;
		break;
	case BJ_Reserved:
		list = &transaction->t_reserved_list;
		break;
	}

	__blist_add_buffer(list, jh);
	jh->b_jlist = jlist;

	if (was_dirty)
		set_buffer_jbddirty(bh);
}

2401
void jbd2_journal_file_buffer(struct journal_head *jh,
2402 2403 2404 2405
				transaction_t *transaction, int jlist)
{
	jbd_lock_bh_state(jh2bh(jh));
	spin_lock(&transaction->t_journal->j_list_lock);
2406
	__jbd2_journal_file_buffer(jh, transaction, jlist);
2407 2408 2409 2410 2411 2412 2413 2414 2415 2416
	spin_unlock(&transaction->t_journal->j_list_lock);
	jbd_unlock_bh_state(jh2bh(jh));
}

/*
 * Remove a buffer from its current buffer list in preparation for
 * dropping it from its current transaction entirely.  If the buffer has
 * already started to be used by a subsequent transaction, refile the
 * buffer on that transaction's metadata list.
 *
2417
 * Called under j_list_lock
2418
 * Called under jbd_lock_bh_state(jh2bh(jh))
2419 2420
 *
 * jh and bh may be already free when this function returns
2421
 */
2422
void __jbd2_journal_refile_buffer(struct journal_head *jh)
2423
{
2424
	int was_dirty, jlist;
2425 2426 2427 2428 2429 2430 2431 2432
	struct buffer_head *bh = jh2bh(jh);

	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
	if (jh->b_transaction)
		assert_spin_locked(&jh->b_transaction->t_journal->j_list_lock);

	/* If the buffer is now unused, just drop it. */
	if (jh->b_next_transaction == NULL) {
2433
		__jbd2_journal_unfile_buffer(jh);
2434 2435 2436 2437 2438 2439 2440 2441 2442
		return;
	}

	/*
	 * It has been modified by a later transaction: add it to the new
	 * transaction's metadata list.
	 */

	was_dirty = test_clear_buffer_jbddirty(bh);
2443
	__jbd2_journal_temp_unlink_buffer(jh);
2444 2445 2446 2447 2448
	/*
	 * We set b_transaction here because b_next_transaction will inherit
	 * our jh reference and thus __jbd2_journal_file_buffer() must not
	 * take a new one.
	 */
2449 2450
	jh->b_transaction = jh->b_next_transaction;
	jh->b_next_transaction = NULL;
2451 2452 2453 2454 2455 2456 2457
	if (buffer_freed(bh))
		jlist = BJ_Forget;
	else if (jh->b_modified)
		jlist = BJ_Metadata;
	else
		jlist = BJ_Reserved;
	__jbd2_journal_file_buffer(jh, jh->b_transaction, jlist);
2458 2459 2460 2461 2462 2463 2464
	J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING);

	if (was_dirty)
		set_buffer_jbddirty(bh);
}

/*
2465 2466 2467 2468
 * __jbd2_journal_refile_buffer() with necessary locking added. We take our
 * bh reference so that we can safely unlock bh.
 *
 * The jh and bh may be freed by this call.
2469
 */
2470
void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh)
2471 2472 2473
{
	struct buffer_head *bh = jh2bh(jh);

2474 2475
	/* Get reference so that buffer cannot be freed before we unlock it */
	get_bh(bh);
2476 2477
	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);
2478
	__jbd2_journal_refile_buffer(jh);
2479 2480 2481 2482
	jbd_unlock_bh_state(bh);
	spin_unlock(&journal->j_list_lock);
	__brelse(bh);
}
2483 2484 2485 2486

/*
 * File inode in the inode list of the handle's transaction
 */
2487 2488
static int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode,
				   unsigned long flags)
2489 2490
{
	transaction_t *transaction = handle->h_transaction;
2491
	journal_t *journal;
2492 2493

	if (is_handle_aborted(handle))
2494 2495
		return -EROFS;
	journal = transaction->t_journal;
2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512

	jbd_debug(4, "Adding inode %lu, tid:%d\n", jinode->i_vfs_inode->i_ino,
			transaction->t_tid);

	/*
	 * First check whether inode isn't already on the transaction's
	 * lists without taking the lock. Note that this check is safe
	 * without the lock as we cannot race with somebody removing inode
	 * from the transaction. The reason is that we remove inode from the
	 * transaction only in journal_release_jbd_inode() and when we commit
	 * the transaction. We are guarded from the first case by holding
	 * a reference to the inode. We are safe against the second case
	 * because if jinode->i_transaction == transaction, commit code
	 * cannot touch the transaction because we hold reference to it,
	 * and if jinode->i_next_transaction == transaction, commit code
	 * will only file the inode where we want it.
	 */
2513 2514 2515
	if ((jinode->i_transaction == transaction ||
	    jinode->i_next_transaction == transaction) &&
	    (jinode->i_flags & flags) == flags)
2516 2517 2518
		return 0;

	spin_lock(&journal->j_list_lock);
2519 2520
	jinode->i_flags |= flags;
	/* Is inode already attached where we need it? */
2521 2522 2523 2524
	if (jinode->i_transaction == transaction ||
	    jinode->i_next_transaction == transaction)
		goto done;

2525 2526 2527 2528 2529 2530 2531
	/*
	 * We only ever set this variable to 1 so the test is safe. Since
	 * t_need_data_flush is likely to be set, we do the test to save some
	 * cacheline bouncing
	 */
	if (!transaction->t_need_data_flush)
		transaction->t_need_data_flush = 1;
2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550
	/* On some different transaction's list - should be
	 * the committing one */
	if (jinode->i_transaction) {
		J_ASSERT(jinode->i_next_transaction == NULL);
		J_ASSERT(jinode->i_transaction ==
					journal->j_committing_transaction);
		jinode->i_next_transaction = transaction;
		goto done;
	}
	/* Not on any transaction list... */
	J_ASSERT(!jinode->i_next_transaction);
	jinode->i_transaction = transaction;
	list_add(&jinode->i_list, &transaction->t_inode_list);
done:
	spin_unlock(&journal->j_list_lock);

	return 0;
}

2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561
int jbd2_journal_inode_add_write(handle_t *handle, struct jbd2_inode *jinode)
{
	return jbd2_journal_file_inode(handle, jinode,
				       JI_WRITE_DATA | JI_WAIT_DATA);
}

int jbd2_journal_inode_add_wait(handle_t *handle, struct jbd2_inode *jinode)
{
	return jbd2_journal_file_inode(handle, jinode, JI_WAIT_DATA);
}

2562
/*
2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580
 * File truncate and transaction commit interact with each other in a
 * non-trivial way.  If a transaction writing data block A is
 * committing, we cannot discard the data by truncate until we have
 * written them.  Otherwise if we crashed after the transaction with
 * write has committed but before the transaction with truncate has
 * committed, we could see stale data in block A.  This function is a
 * helper to solve this problem.  It starts writeout of the truncated
 * part in case it is in the committing transaction.
 *
 * Filesystem code must call this function when inode is journaled in
 * ordered mode before truncation happens and after the inode has been
 * placed on orphan list with the new inode size. The second condition
 * avoids the race that someone writes new data and we start
 * committing the transaction after this function has been called but
 * before a transaction for truncate is started (and furthermore it
 * allows us to optimize the case where the addition to orphan list
 * happens in the same transaction as write --- we don't have to write
 * any data in such case).
2581
 */
2582 2583
int jbd2_journal_begin_ordered_truncate(journal_t *journal,
					struct jbd2_inode *jinode,
2584 2585
					loff_t new_size)
{
2586
	transaction_t *inode_trans, *commit_trans;
2587 2588
	int ret = 0;

2589 2590
	/* This is a quick check to avoid locking if not necessary */
	if (!jinode->i_transaction)
2591
		goto out;
2592 2593 2594
	/* Locks are here just to force reading of recent values, it is
	 * enough that the transaction was not committing before we started
	 * a transaction adding the inode to orphan list */
2595
	read_lock(&journal->j_state_lock);
2596
	commit_trans = journal->j_committing_transaction;
2597
	read_unlock(&journal->j_state_lock);
2598 2599 2600 2601 2602
	spin_lock(&journal->j_list_lock);
	inode_trans = jinode->i_transaction;
	spin_unlock(&journal->j_list_lock);
	if (inode_trans == commit_trans) {
		ret = filemap_fdatawrite_range(jinode->i_vfs_inode->i_mapping,
2603 2604 2605 2606 2607 2608 2609
			new_size, LLONG_MAX);
		if (ret)
			jbd2_journal_abort(journal, ret);
	}
out:
	return ret;
}