transaction.c 81.3 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0+
2
/*
3
 * linux/fs/jbd2/transaction.c
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
 *
 * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
 *
 * Copyright 1998 Red Hat corp --- All Rights Reserved
 *
 * 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>
19
#include <linux/jbd2.h>
20 21 22 23 24
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/highmem.h>
25
#include <linux/hrtimer.h>
26
#include <linux/backing-dev.h>
27
#include <linux/bug.h>
28
#include <linux/module.h>
29
#include <linux/sched/mm.h>
30

31 32
#include <trace/events/jbd2.h>

33
static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh);
34
static void __jbd2_journal_unfile_buffer(struct journal_head *jh);
35

36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
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)
{
52 53
	kmem_cache_destroy(transaction_cache);
	transaction_cache = NULL;
54 55 56 57 58 59 60 61 62
}

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

63
/*
64
 * jbd2_get_transaction: obtain a new transaction_t object.
65 66 67 68 69 70 71 72 73 74 75 76 77 78
 *
 * 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 *
79
jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
80 81 82
{
	transaction->t_journal = journal;
	transaction->t_state = T_RUNNING;
83
	transaction->t_start_time = ktime_get();
84 85 86
	transaction->t_tid = journal->j_transaction_sequence++;
	transaction->t_expires = jiffies + journal->j_commit_interval;
	spin_lock_init(&transaction->t_handle_lock);
87
	atomic_set(&transaction->t_updates, 0);
88 89
	atomic_set(&transaction->t_outstanding_credits,
		   atomic_read(&journal->j_reserved_credits));
90
	atomic_set(&transaction->t_handle_count, 0);
91
	INIT_LIST_HEAD(&transaction->t_inode_list);
92
	INIT_LIST_HEAD(&transaction->t_private_list);
93 94

	/* Set up the commit timer for the new transaction. */
95
	journal->j_commit_timer.expires = round_jiffies_up(transaction->t_expires);
96 97 98 99
	add_timer(&journal->j_commit_timer);

	J_ASSERT(journal->j_running_transaction == NULL);
	journal->j_running_transaction = transaction;
100 101
	transaction->t_max_wait = 0;
	transaction->t_start = jiffies;
102
	transaction->t_requested = 0;
103 104 105 106 107 108 109 110 111 112 113 114

	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.
 */

115
/*
116
 * Update transaction's maximum wait time, if debugging is enabled.
117 118 119 120 121 122 123 124
 *
 * 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.
 */
125 126
static inline void update_t_max_wait(transaction_t *transaction,
				     unsigned long ts)
127 128 129 130 131 132 133 134 135 136 137 138 139
{
#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
}

140
/*
141 142 143
 * Wait until running transaction passes to T_FLUSH state and new transaction
 * can thus be started. Also starts the commit if needed. The function expects
 * running transaction to exist and releases j_state_lock.
144 145 146 147 148 149 150 151 152 153 154 155 156 157
 */
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);
158
	jbd2_might_wait_for_commit(journal);
159 160 161 162
	schedule();
	finish_wait(&journal->j_wait_transaction_locked, &wait);
}

163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
/*
 * Wait until running transaction transitions from T_SWITCH to T_FLUSH
 * state and new transaction can thus be started. The function releases
 * j_state_lock.
 */
static void wait_transaction_switching(journal_t *journal)
	__releases(journal->j_state_lock)
{
	DEFINE_WAIT(wait);

	if (WARN_ON(!journal->j_running_transaction ||
		    journal->j_running_transaction->t_state != T_SWITCH))
		return;
	prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
			TASK_UNINTERRUPTIBLE);
	read_unlock(&journal->j_state_lock);
	/*
	 * We don't call jbd2_might_wait_for_commit() here as there's no
	 * waiting for outstanding handles happening anymore in T_SWITCH state
	 * and handling of reserved handles actually relies on that for
	 * correctness.
	 */
	schedule();
	finish_wait(&journal->j_wait_transaction_locked, &wait);
}

189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211
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.
	 */
212 213
	if (t->t_state != T_RUNNING) {
		WARN_ON_ONCE(t->t_state >= T_FLUSH);
214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230
		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);
231 232 233 234 235 236 237 238

		/*
		 * 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);
239
			jbd2_might_wait_for_commit(journal);
240 241 242 243 244 245
			wait_event(journal->j_wait_reserved,
				   atomic_read(&journal->j_reserved_credits) + total <=
				   journal->j_max_transaction_buffers);
			return 1;
		}

246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263
		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);
264
		jbd2_might_wait_for_commit(journal);
265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
		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);
282
		jbd2_might_wait_for_commit(journal);
283 284 285 286 287 288 289 290
		wait_event(journal->j_wait_reserved,
			 atomic_read(&journal->j_reserved_credits) + rsv_blocks
			 <= journal->j_max_transaction_buffers / 2);
		return 1;
	}
	return 0;
}

291 292 293 294 295 296 297
/*
 * 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.
 */

298
static int start_this_handle(journal_t *journal, handle_t *handle,
299
			     gfp_t gfp_mask)
300
{
301
	transaction_t	*transaction, *new_transaction = NULL;
302 303
	int		blocks = handle->h_buffer_credits;
	int		rsv_blocks = 0;
304
	unsigned long ts = jiffies;
305

306 307 308
	if (handle->h_rsv_handle)
		rsv_blocks = handle->h_rsv_handle->h_buffer_credits;

309
	/*
310 311 312
	 * 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.
313
	 */
314 315 316 317 318 319 320
	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);
321
		return -ENOSPC;
322 323 324 325
	}

alloc_transaction:
	if (!journal->j_running_transaction) {
326 327 328 329 330 331
		/*
		 * 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;
332 333
		new_transaction = kmem_cache_zalloc(transaction_cache,
						    gfp_mask);
334
		if (!new_transaction)
335
			return -ENOMEM;
336 337 338 339 340 341 342 343
	}

	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
	 */
344 345
repeat:
	read_lock(&journal->j_state_lock);
346
	BUG_ON(journal->j_flags & JBD2_UNMOUNT);
347
	if (is_journal_aborted(journal) ||
348
	    (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) {
349
		read_unlock(&journal->j_state_lock);
350
		jbd2_journal_free_transaction(new_transaction);
351
		return -EROFS;
352 353
	}

354 355 356 357 358 359
	/*
	 * 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) {
360
		read_unlock(&journal->j_state_lock);
361 362 363 364 365 366
		wait_event(journal->j_wait_transaction_locked,
				journal->j_barrier_count == 0);
		goto repeat;
	}

	if (!journal->j_running_transaction) {
367 368
		read_unlock(&journal->j_state_lock);
		if (!new_transaction)
369
			goto alloc_transaction;
370
		write_lock(&journal->j_state_lock);
371
		if (!journal->j_running_transaction &&
372
		    (handle->h_reserved || !journal->j_barrier_count)) {
373 374
			jbd2_get_transaction(journal, new_transaction);
			new_transaction = NULL;
375
		}
376 377
		write_unlock(&journal->j_state_lock);
		goto repeat;
378 379 380 381
	}

	transaction = journal->j_running_transaction;

382 383 384 385 386
	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 {
387
		/*
388 389
		 * We have handle reserved so we are allowed to join T_LOCKED
		 * transaction and we don't have to check for transaction size
390 391 392
		 * and journal space. But we still have to wait while running
		 * transaction is being switched to a committing one as it
		 * won't wait for any handles anymore.
393
		 */
394 395 396 397
		if (transaction->t_state == T_SWITCH) {
			wait_transaction_switching(journal);
			goto repeat;
		}
398 399
		sub_reserved_credits(journal, blocks);
		handle->h_reserved = 0;
400 401 402
	}

	/* OK, account for the buffers that this operation expects to
403 404
	 * use and add the handle to the running transaction. 
	 */
405
	update_t_max_wait(transaction, ts);
406
	handle->h_transaction = transaction;
407
	handle->h_requested_credits = blocks;
408
	handle->h_start_jiffies = jiffies;
409
	atomic_inc(&transaction->t_updates);
410
	atomic_inc(&transaction->t_handle_count);
411 412
	jbd_debug(4, "Handle %p given %d credits (total %d, free %lu)\n",
		  handle, blocks,
413
		  atomic_read(&transaction->t_outstanding_credits),
414
		  jbd2_log_space_left(journal));
415
	read_unlock(&journal->j_state_lock);
416
	current->journal_info = handle;
417

418
	rwsem_acquire_read(&journal->j_trans_commit_map, 0, 0, _THIS_IP_);
419
	jbd2_journal_free_transaction(new_transaction);
420 421 422 423 424
	/*
	 * 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();
425
	return 0;
426 427 428 429 430
}

/* Allocate a new handle.  This should probably be in a slab... */
static handle_t *new_handle(int nblocks)
{
431
	handle_t *handle = jbd2_alloc_handle(GFP_NOFS);
432 433 434 435 436 437 438 439
	if (!handle)
		return NULL;
	handle->h_buffer_credits = nblocks;
	handle->h_ref = 1;

	return handle;
}

440 441 442
handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
			      gfp_t gfp_mask, unsigned int type,
			      unsigned int line_no)
443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458
{
	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);
459 460 461 462 463 464 465 466 467 468 469 470
	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;
	}
471

472
	err = start_this_handle(journal, handle, gfp_mask);
473
	if (err < 0) {
474 475
		if (handle->h_rsv_handle)
			jbd2_free_handle(handle->h_rsv_handle);
476
		jbd2_free_handle(handle);
477
		return ERR_PTR(err);
478
	}
479 480 481 482 483
	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);
484

485 486
	return handle;
}
487 488 489
EXPORT_SYMBOL(jbd2__journal_start);


490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
/**
 * 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.
 */
509 510
handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
{
511
	return jbd2__journal_start(journal, nblocks, 0, GFP_NOFS, 0, 0);
512 513 514
}
EXPORT_SYMBOL(jbd2_journal_start);

515 516 517 518 519 520 521 522 523 524 525
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);

/**
526
 * int jbd2_journal_start_reserved() - start reserved handle
527
 * @handle: handle to start
528 529
 * @type: for handle statistics
 * @line_no: for handle statistics
530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
 *
 * 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);
565
	if (ret < 0) {
566
		handle->h_journal = journal;
567
		jbd2_journal_free_reserved(handle);
568 569
		return ret;
	}
570 571
	handle->h_type = type;
	handle->h_line_no = line_no;
572
	return 0;
573 574
}
EXPORT_SYMBOL(jbd2_journal_start_reserved);
575 576

/**
577
 * int jbd2_journal_extend() - extend buffer credits.
578 579 580 581 582
 * @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
583
 * a credit for a number of buffer modifications in advance, but can
584 585
 * extend its credit if it needs more.
 *
586
 * jbd2_journal_extend tries to give the running handle more buffer credits.
587 588 589 590 591 592 593 594 595
 * 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.
 */
596
int jbd2_journal_extend(handle_t *handle, int nblocks)
597 598
{
	transaction_t *transaction = handle->h_transaction;
599
	journal_t *journal;
600 601 602 603
	int result;
	int wanted;

	if (is_handle_aborted(handle))
604 605
		return -EROFS;
	journal = transaction->t_journal;
606 607 608

	result = 1;

609
	read_lock(&journal->j_state_lock);
610 611

	/* Don't extend a locked-down transaction! */
612
	if (transaction->t_state != T_RUNNING) {
613 614 615 616 617 618
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction not running\n", handle, nblocks);
		goto error_out;
	}

	spin_lock(&transaction->t_handle_lock);
619 620
	wanted = atomic_add_return(nblocks,
				   &transaction->t_outstanding_credits);
621 622 623 624

	if (wanted > journal->j_max_transaction_buffers) {
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction too large\n", handle, nblocks);
625
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
626 627 628
		goto unlock;
	}

629 630
	if (wanted + (wanted >> JBD2_CONTROL_BLOCKS_SHIFT) >
	    jbd2_log_space_left(journal)) {
631 632
		jbd_debug(3, "denied handle %p %d blocks: "
			  "insufficient log space\n", handle, nblocks);
633
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
634 635 636
		goto unlock;
	}

637
	trace_jbd2_handle_extend(journal->j_fs_dev->bd_dev,
638
				 transaction->t_tid,
639 640 641 642
				 handle->h_type, handle->h_line_no,
				 handle->h_buffer_credits,
				 nblocks);

643
	handle->h_buffer_credits += nblocks;
644
	handle->h_requested_credits += nblocks;
645 646 647 648 649 650
	result = 0;

	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
unlock:
	spin_unlock(&transaction->t_handle_lock);
error_out:
651
	read_unlock(&journal->j_state_lock);
652 653 654 655 656
	return result;
}


/**
657
 * int jbd2_journal_restart() - restart a handle .
658 659
 * @handle:  handle to restart
 * @nblocks: nr credits requested
660
 * @gfp_mask: memory allocation flags (for start_this_handle)
661 662 663 664
 *
 * Restart a handle for a multi-transaction filesystem
 * operation.
 *
665 666
 * 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
667
 * handle's transaction so far and reattach the handle to a new
668
 * transaction capable of guaranteeing the requested number of
669 670
 * credits. We preserve reserved handle if there's any attached to the
 * passed in handle.
671
 */
672
int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask)
673 674
{
	transaction_t *transaction = handle->h_transaction;
675
	journal_t *journal;
676 677
	tid_t		tid;
	int		need_to_start, ret;
678 679 680 681 682

	/* 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;
683
	journal = transaction->t_journal;
684 685 686 687 688

	/*
	 * First unlink the handle from its current transaction, and start the
	 * commit on that.
	 */
689
	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
690 691
	J_ASSERT(journal_current_handle() == handle);

692
	read_lock(&journal->j_state_lock);
693
	spin_lock(&transaction->t_handle_lock);
694 695
	atomic_sub(handle->h_buffer_credits,
		   &transaction->t_outstanding_credits);
696 697 698 699
	if (handle->h_rsv_handle) {
		sub_reserved_credits(journal,
				     handle->h_rsv_handle->h_buffer_credits);
	}
700
	if (atomic_dec_and_test(&transaction->t_updates))
701
		wake_up(&journal->j_wait_updates);
702
	tid = transaction->t_tid;
703
	spin_unlock(&transaction->t_handle_lock);
704 705
	handle->h_transaction = NULL;
	current->journal_info = NULL;
706 707

	jbd_debug(2, "restarting handle %p\n", handle);
708
	need_to_start = !tid_geq(journal->j_commit_request, tid);
709
	read_unlock(&journal->j_state_lock);
710 711
	if (need_to_start)
		jbd2_log_start_commit(journal, tid);
712

713
	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
714
	handle->h_buffer_credits = nblocks;
715 716 717 718 719 720
	/*
	 * 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);
721
	ret = start_this_handle(journal, handle, gfp_mask);
722 723
	return ret;
}
724
EXPORT_SYMBOL(jbd2__journal_restart);
725 726


727 728 729 730 731 732
int jbd2_journal_restart(handle_t *handle, int nblocks)
{
	return jbd2__journal_restart(handle, nblocks, GFP_NOFS);
}
EXPORT_SYMBOL(jbd2_journal_restart);

733
/**
734
 * void jbd2_journal_lock_updates () - establish a transaction barrier.
735 736 737 738 739 740 741 742
 * @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.
 */
743
void jbd2_journal_lock_updates(journal_t *journal)
744 745 746
{
	DEFINE_WAIT(wait);

747 748
	jbd2_might_wait_for_commit(journal);

749
	write_lock(&journal->j_state_lock);
750 751
	++journal->j_barrier_count;

752 753 754 755 756 757 758 759
	/* 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);
	}

760 761 762 763 764 765 766 767
	/* 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);
768 769
		prepare_to_wait(&journal->j_wait_updates, &wait,
				TASK_UNINTERRUPTIBLE);
770
		if (!atomic_read(&transaction->t_updates)) {
771
			spin_unlock(&transaction->t_handle_lock);
772
			finish_wait(&journal->j_wait_updates, &wait);
773 774 775
			break;
		}
		spin_unlock(&transaction->t_handle_lock);
776
		write_unlock(&journal->j_state_lock);
777 778
		schedule();
		finish_wait(&journal->j_wait_updates, &wait);
779
		write_lock(&journal->j_state_lock);
780
	}
781
	write_unlock(&journal->j_state_lock);
782 783 784

	/*
	 * We have now established a barrier against other normal updates, but
785
	 * we also need to barrier against other jbd2_journal_lock_updates() calls
786 787 788 789 790 791 792
	 * to make sure that we serialise special journal-locked operations
	 * too.
	 */
	mutex_lock(&journal->j_barrier);
}

/**
793
 * void jbd2_journal_unlock_updates (journal_t* journal) - release barrier
794 795
 * @journal:  Journal to release the barrier on.
 *
796
 * Release a transaction barrier obtained with jbd2_journal_lock_updates().
797 798 799
 *
 * Should be called without the journal lock held.
 */
800
void jbd2_journal_unlock_updates (journal_t *journal)
801 802 803 804
{
	J_ASSERT(journal->j_barrier_count != 0);

	mutex_unlock(&journal->j_barrier);
805
	write_lock(&journal->j_state_lock);
806
	--journal->j_barrier_count;
807
	write_unlock(&journal->j_state_lock);
808 809 810
	wake_up(&journal->j_wait_transaction_locked);
}

811
static void warn_dirty_buffer(struct buffer_head *bh)
812
{
813
	printk(KERN_WARNING
814
	       "JBD2: Spotted dirty metadata buffer (dev = %pg, blocknr = %llu). "
815 816
	       "There's a risk of filesystem corruption in case of system "
	       "crash.\n",
817
	       bh->b_bdev, (unsigned long long)bh->b_blocknr);
818 819
}

820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843
/* 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;
}

844 845 846 847 848 849 850 851 852 853 854 855 856 857 858
/*
 * 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;
859
	transaction_t *transaction = handle->h_transaction;
860 861 862
	journal_t *journal;
	int error;
	char *frozen_buffer = NULL;
863
	unsigned long start_lock, time_lock;
864 865 866 867 868

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

869
	jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy);
870 871 872 873 874 875 876

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

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

877
 	start_lock = jiffies;
878 879 880
	lock_buffer(bh);
	jbd_lock_bh_state(bh);

881 882 883 884 885 886
	/* 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));

887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
	/* 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);
913
			warn_dirty_buffer(bh);
914 915 916 917 918 919
		}
		/*
		 * 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.
		 */
920 921 922
		JBUFFER_TRACE(jh, "Journalling dirty buffer");
		clear_buffer_dirty(bh);
		set_buffer_jbddirty(bh);
923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941
	}

	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;

942 943 944 945
	/*
	 * this is the first time this transaction is touching this buffer,
	 * reset the modified flag
	 */
946
	jh->b_modified = 0;
947

948 949 950 951 952 953 954 955 956
	/*
	 * 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");
957 958 959 960 961 962
		/*
		 * 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();
963 964 965 966 967
		spin_lock(&journal->j_list_lock);
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
		spin_unlock(&journal->j_list_lock);
		goto done;
	}
968 969 970 971 972 973 974
	/*
	 * 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);
975
		goto attach_next;
976 977
	}

978 979 980
	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);
981

982 983 984 985 986 987 988 989 990 991 992 993 994 995 996
	/*
	 * 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;
	}
997

998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
	/*
	 * 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");
1014
			jbd_unlock_bh_state(bh);
1015 1016
			frozen_buffer = jbd2_alloc(jh2bh(jh)->b_size,
						   GFP_NOFS | __GFP_NOFAIL);
1017
			goto repeat;
1018
		}
1019 1020 1021
		jh->b_frozen_data = frozen_buffer;
		frozen_buffer = NULL;
		jbd2_freeze_jh_data(jh);
1022
	}
1023 1024 1025 1026 1027 1028 1029
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();
1030
	jh->b_next_transaction = transaction;
1031 1032 1033 1034 1035 1036 1037 1038

done:
	jbd_unlock_bh_state(bh);

	/*
	 * If we are about to journal a buffer, then any revoke pending on it is
	 * no longer valid
	 */
1039
	jbd2_journal_cancel_revoke(handle, jh);
1040 1041 1042

out:
	if (unlikely(frozen_buffer))	/* It's usually NULL */
1043
		jbd2_free(frozen_buffer, bh->b_size);
1044 1045 1046 1047 1048

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

1049
/* Fast check whether buffer is already attached to the required transaction */
1050 1051
static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh,
							bool undo)
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
{
	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;
1078 1079 1080
	/* For undo access buffer must have data copied */
	if (undo && !jh->b_committed_data)
		goto out;
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
	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;
}

1102
/**
1103
 * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update.
1104 1105 1106
 * @handle: transaction to add buffer modifications to
 * @bh:     bh to be used for metadata writes
 *
1107
 * Returns: error code or 0 on success.
1108 1109
 *
 * In full data journalling mode the buffer may be of type BJ_AsyncData,
1110
 * because we're ``write()ing`` a buffer which is also part of a shared mapping.
1111 1112
 */

1113
int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh)
1114
{
1115
	struct journal_head *jh;
1116 1117
	int rc;

1118
	if (jbd2_write_access_granted(handle, bh, false))
1119 1120 1121
		return 0;

	jh = jbd2_journal_add_journal_head(bh);
1122 1123 1124 1125
	/* 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);
1126
	jbd2_journal_put_journal_head(jh);
1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143
	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. */

/**
1144
 * int jbd2_journal_get_create_access () - notify intent to use newly created bh
1145 1146 1147 1148 1149
 * @handle: transaction to new buffer to
 * @bh: new buffer.
 *
 * Call this if you create a new bh.
 */
1150
int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
1151 1152
{
	transaction_t *transaction = handle->h_transaction;
1153
	journal_t *journal;
1154
	struct journal_head *jh = jbd2_journal_add_journal_head(bh);
1155 1156 1157 1158 1159 1160
	int err;

	jbd_debug(5, "journal_head %p\n", jh);
	err = -EROFS;
	if (is_handle_aborted(handle))
		goto out;
1161
	journal = transaction->t_journal;
1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181
	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) {
1182 1183 1184 1185 1186 1187 1188 1189 1190
		/*
		 * 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));
1191 1192 1193
		/* first access by this transaction */
		jh->b_modified = 0;

1194
		JBUFFER_TRACE(jh, "file as BJ_Reserved");
1195
		spin_lock(&journal->j_list_lock);
1196
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
1197
		spin_unlock(&journal->j_list_lock);
1198
	} else if (jh->b_transaction == journal->j_committing_transaction) {
1199 1200 1201
		/* first access by this transaction */
		jh->b_modified = 0;

1202
		JBUFFER_TRACE(jh, "set next transaction");
1203
		spin_lock(&journal->j_list_lock);
1204
		jh->b_next_transaction = transaction;
1205
		spin_unlock(&journal->j_list_lock);
1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
	}
	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");
1217
	jbd2_journal_cancel_revoke(handle, jh);
1218
out:
1219
	jbd2_journal_put_journal_head(jh);
1220 1221 1222 1223
	return err;
}

/**
1224
 * int jbd2_journal_get_undo_access() -  Notify intent to modify metadata with
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235
 *     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.
 *
1236
 * To deal with that, jbd2_journal_get_undo_access requests write access to a
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248
 * 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.
 */
1249
int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
1250 1251
{
	int err;
1252
	struct journal_head *jh;
1253 1254
	char *committed_data = NULL;

1255
	if (jbd2_write_access_granted(handle, bh, true))
1256
		return 0;
1257

1258
	jh = jbd2_journal_add_journal_head(bh);
1259 1260
	JBUFFER_TRACE(jh, "entry");

1261 1262 1263 1264 1265 1266 1267 1268 1269 1270
	/*
	 * 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:
1271 1272 1273
	if (!jh->b_committed_data)
		committed_data = jbd2_alloc(jh2bh(jh)->b_size,
					    GFP_NOFS|__GFP_NOFAIL);
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290

	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:
1291
	jbd2_journal_put_journal_head(jh);
1292
	if (unlikely(committed_data))
1293
		jbd2_free(committed_data, bh->b_size);
1294 1295 1296
	return err;
}

Joel Becker's avatar
Joel Becker committed
1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
/**
 * 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)
{
1311
	struct journal_head *jh = jbd2_journal_grab_journal_head(bh);
Joel Becker's avatar
Joel Becker committed
1312

1313 1314
	if (WARN_ON(!jh))
		return;
Joel Becker's avatar
Joel Becker committed
1315
	jh->b_triggers = type;
1316
	jbd2_journal_put_journal_head(jh);
Joel Becker's avatar
Joel Becker committed
1317 1318
}

1319
void jbd2_buffer_frozen_trigger(struct journal_head *jh, void *mapped_data,
Joel Becker's avatar
Joel Becker committed
1320 1321 1322 1323
				struct jbd2_buffer_trigger_type *triggers)
{
	struct buffer_head *bh = jh2bh(jh);

1324
	if (!triggers || !triggers->t_frozen)
Joel Becker's avatar
Joel Becker committed
1325 1326
		return;

1327
	triggers->t_frozen(triggers, bh, mapped_data, bh->b_size);
Joel Becker's avatar
Joel Becker committed
1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338
}

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));
}

1339
/**
1340
 * int jbd2_journal_dirty_metadata() -  mark a buffer as containing dirty metadata
1341 1342 1343 1344 1345 1346
 * @handle: transaction to add buffer to.
 * @bh: buffer to mark
 *
 * mark dirty metadata which needs to be journaled as part of the current
 * transaction.
 *
1347 1348 1349 1350
 * 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.
 *
1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361
 * 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.
 */
1362
int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
1363 1364
{
	transaction_t *transaction = handle->h_transaction;
1365
	journal_t *journal;
1366
	struct journal_head *jh;
1367
	int ret = 0;
1368 1369

	if (is_handle_aborted(handle))
1370
		return -EROFS;
1371 1372 1373
	if (!buffer_jbd(bh))
		return -EUCLEAN;

1374 1375 1376 1377 1378
	/*
	 * We don't grab jh reference here since the buffer must be part
	 * of the running transaction.
	 */
	jh = bh2jh(bh);
1379 1380 1381
	jbd_debug(5, "journal_head %p\n", jh);
	JBUFFER_TRACE(jh, "entry");

1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399
	/*
	 * 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);
1400 1401 1402 1403 1404 1405 1406
			if (jh->b_transaction == transaction &&
			    jh->b_jlist != BJ_Metadata)
				pr_err("JBD2: assertion failure: h_type=%u "
				       "h_line_no=%u block_no=%llu jlist=%u\n",
				       handle->h_type, handle->h_line_no,
				       (unsigned long long) bh->b_blocknr,
				       jh->b_jlist);
1407 1408 1409 1410 1411 1412 1413 1414
			J_ASSERT_JH(jh, jh->b_transaction != transaction ||
					jh->b_jlist == BJ_Metadata);
			jbd_unlock_bh_state(bh);
		}
		goto out;
	}

	journal = transaction->t_journal;
1415 1416 1417 1418 1419 1420 1421 1422
	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
		 */
1423 1424 1425 1426
		if (handle->h_buffer_credits <= 0) {
			ret = -ENOSPC;
			goto out_unlock_bh;
		}
1427
		jh->b_modified = 1;
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439
		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");
1440 1441
		if (unlikely(jh->b_transaction !=
			     journal->j_running_transaction)) {
1442
			printk(KERN_ERR "JBD2: %s: "
1443
			       "jh->b_transaction (%llu, %p, %u) != "
1444
			       "journal->j_running_transaction (%p, %u)\n",
1445 1446 1447 1448 1449 1450 1451 1452 1453
			       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;
		}
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466
		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");
1467 1468 1469 1470 1471 1472 1473 1474
		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",
1475 1476
			       journal->j_devname,
			       (unsigned long long) bh->b_blocknr,
1477
			       transaction, transaction->t_tid,
1478
			       jh->b_transaction,
1479 1480
			       jh->b_transaction ?
			       jh->b_transaction->t_tid : 0,
1481 1482 1483
			       jh->b_next_transaction,
			       jh->b_next_transaction ?
			       jh->b_next_transaction->t_tid : 0,
1484 1485
			       jh->b_jlist);
			WARN_ON(1);
1486 1487
			ret = -EINVAL;
		}
1488 1489 1490 1491 1492 1493
		/* 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: */
1494
	J_ASSERT_JH(jh, jh->b_frozen_data == NULL);
1495 1496 1497

	JBUFFER_TRACE(jh, "file as BJ_Metadata");
	spin_lock(&journal->j_list_lock);
1498
	__jbd2_journal_file_buffer(jh, transaction, BJ_Metadata);
1499 1500 1501 1502 1503
	spin_unlock(&journal->j_list_lock);
out_unlock_bh:
	jbd_unlock_bh_state(bh);
out:
	JBUFFER_TRACE(jh, "exit");
1504
	return ret;
1505 1506 1507
}

/**
1508
 * void jbd2_journal_forget() - bforget() for potentially-journaled buffers.
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523
 * @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.
 */
1524
int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
1525 1526
{
	transaction_t *transaction = handle->h_transaction;
1527
	journal_t *journal;
1528 1529 1530
	struct journal_head *jh;
	int drop_reserve = 0;
	int err = 0;
1531
	int was_modified = 0;
1532

1533 1534 1535 1536
	if (is_handle_aborted(handle))
		return -EROFS;
	journal = transaction->t_journal;

1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552
	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;
	}

1553
	/* keep track of whether or not this transaction modified us */
1554 1555
	was_modified = jh->b_modified;

1556 1557 1558 1559 1560 1561
	/*
	 * The buffer's going from the transaction, we must drop
	 * all references -bzzz
	 */
	jh->b_modified = 0;

1562
	if (jh->b_transaction == transaction) {
1563 1564 1565 1566 1567 1568 1569 1570 1571 1572
		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");

1573 1574 1575 1576 1577 1578
		/*
		 * we only want to drop a reference if this transaction
		 * modified the buffer
		 */
		if (was_modified)
			drop_reserve = 1;
1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591

		/*
		 * 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.
		 */

1592
		spin_lock(&journal->j_list_lock);
1593
		if (jh->b_cp_transaction) {
1594