transaction.c 82.4 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>
J
Josef Bacik 已提交
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
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);
C
Chengguang Xu 已提交
45 46 47 48 49
	if (!transaction_cache) {
		pr_emerg("JBD2: failed to create transaction cache\n");
		return -ENOMEM;
	}
	return 0;
50 51 52 53
}

void jbd2_journal_destroy_transaction_cache(void)
{
54 55
	kmem_cache_destroy(transaction_cache);
	transaction_cache = NULL;
56 57 58 59 60 61 62 63 64
}

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

65
/*
66
 * jbd2_get_transaction: obtain a new transaction_t object.
67 68 69 70 71 72 73 74 75 76 77 78 79 80
 *
 * 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 *
81
jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
82 83 84
{
	transaction->t_journal = journal;
	transaction->t_state = T_RUNNING;
J
Josef Bacik 已提交
85
	transaction->t_start_time = ktime_get();
86 87 88
	transaction->t_tid = journal->j_transaction_sequence++;
	transaction->t_expires = jiffies + journal->j_commit_interval;
	spin_lock_init(&transaction->t_handle_lock);
89
	atomic_set(&transaction->t_updates, 0);
J
Jan Kara 已提交
90 91
	atomic_set(&transaction->t_outstanding_credits,
		   atomic_read(&journal->j_reserved_credits));
92
	atomic_set(&transaction->t_handle_count, 0);
93
	INIT_LIST_HEAD(&transaction->t_inode_list);
94
	INIT_LIST_HEAD(&transaction->t_private_list);
95 96

	/* Set up the commit timer for the new transaction. */
97
	journal->j_commit_timer.expires = round_jiffies_up(transaction->t_expires);
98 99 100 101
	add_timer(&journal->j_commit_timer);

	J_ASSERT(journal->j_running_transaction == NULL);
	journal->j_running_transaction = transaction;
102 103
	transaction->t_max_wait = 0;
	transaction->t_start = jiffies;
104
	transaction->t_requested = 0;
105 106 107 108 109 110 111 112 113 114 115 116

	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.
 */

117
/*
118
 * Update transaction's maximum wait time, if debugging is enabled.
119 120 121 122 123 124 125 126
 *
 * 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.
 */
127 128
static inline void update_t_max_wait(transaction_t *transaction,
				     unsigned long ts)
129 130 131 132 133 134 135 136 137 138 139 140 141
{
#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
142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
/*
 * 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);
160
	jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205
	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);
206 207 208 209 210 211 212 213

		/*
		 * 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);
214
			jbd2_might_wait_for_commit(journal);
215 216 217 218 219 220
			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
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238
		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);
239
		jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256
		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);
257
		jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
258 259 260 261 262 263 264 265
		wait_event(journal->j_wait_reserved,
			 atomic_read(&journal->j_reserved_credits) + rsv_blocks
			 <= journal->j_max_transaction_buffers / 2);
		return 1;
	}
	return 0;
}

266 267 268 269 270 271 272
/*
 * 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.
 */

273
static int start_this_handle(journal_t *journal, handle_t *handle,
D
Dan Carpenter 已提交
274
			     gfp_t gfp_mask)
275
{
276
	transaction_t	*transaction, *new_transaction = NULL;
J
Jan Kara 已提交
277 278
	int		blocks = handle->h_buffer_credits;
	int		rsv_blocks = 0;
279
	unsigned long ts = jiffies;
280

281 282 283
	if (handle->h_rsv_handle)
		rsv_blocks = handle->h_rsv_handle->h_buffer_credits;

J
Jan Kara 已提交
284
	/*
285 286 287
	 * 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
288
	 */
289 290 291 292 293 294 295
	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);
296
		return -ENOSPC;
297 298 299 300
	}

alloc_transaction:
	if (!journal->j_running_transaction) {
301 302 303 304 305 306
		/*
		 * 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;
307 308
		new_transaction = kmem_cache_zalloc(transaction_cache,
						    gfp_mask);
309
		if (!new_transaction)
310
			return -ENOMEM;
311 312 313 314 315 316 317 318
	}

	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
	 */
319 320
repeat:
	read_lock(&journal->j_state_lock);
321
	BUG_ON(journal->j_flags & JBD2_UNMOUNT);
322
	if (is_journal_aborted(journal) ||
323
	    (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) {
324
		read_unlock(&journal->j_state_lock);
325
		jbd2_journal_free_transaction(new_transaction);
326
		return -EROFS;
327 328
	}

J
Jan Kara 已提交
329 330 331 332 333 334
	/*
	 * 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) {
335
		read_unlock(&journal->j_state_lock);
336 337 338 339 340 341
		wait_event(journal->j_wait_transaction_locked,
				journal->j_barrier_count == 0);
		goto repeat;
	}

	if (!journal->j_running_transaction) {
342 343
		read_unlock(&journal->j_state_lock);
		if (!new_transaction)
344
			goto alloc_transaction;
345
		write_lock(&journal->j_state_lock);
346
		if (!journal->j_running_transaction &&
J
Jan Kara 已提交
347
		    (handle->h_reserved || !journal->j_barrier_count)) {
348 349
			jbd2_get_transaction(journal, new_transaction);
			new_transaction = NULL;
350
		}
351 352
		write_unlock(&journal->j_state_lock);
		goto repeat;
353 354 355 356
	}

	transaction = journal->j_running_transaction;

J
Jan Kara 已提交
357 358 359 360 361
	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 {
362
		/*
J
Jan Kara 已提交
363 364 365
		 * 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.
366
		 */
J
Jan Kara 已提交
367 368
		sub_reserved_credits(journal, blocks);
		handle->h_reserved = 0;
369 370 371
	}

	/* OK, account for the buffers that this operation expects to
372 373
	 * use and add the handle to the running transaction. 
	 */
374
	update_t_max_wait(transaction, ts);
375
	handle->h_transaction = transaction;
J
Jan Kara 已提交
376
	handle->h_requested_credits = blocks;
377
	handle->h_start_jiffies = jiffies;
378
	atomic_inc(&transaction->t_updates);
379
	atomic_inc(&transaction->t_handle_count);
J
Jan Kara 已提交
380 381
	jbd_debug(4, "Handle %p given %d credits (total %d, free %lu)\n",
		  handle, blocks,
382
		  atomic_read(&transaction->t_outstanding_credits),
383
		  jbd2_log_space_left(journal));
384
	read_unlock(&journal->j_state_lock);
385
	current->journal_info = handle;
386

387
	rwsem_acquire_read(&journal->j_trans_commit_map, 0, 0, _THIS_IP_);
388
	jbd2_journal_free_transaction(new_transaction);
389 390 391 392 393
	/*
	 * 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();
394
	return 0;
395 396 397 398 399
}

/* Allocate a new handle.  This should probably be in a slab... */
static handle_t *new_handle(int nblocks)
{
M
Mingming Cao 已提交
400
	handle_t *handle = jbd2_alloc_handle(GFP_NOFS);
401 402 403 404
	if (!handle)
		return NULL;
	handle->h_buffer_credits = nblocks;
	handle->h_ref = 1;
405
	handle->h_pre_start_jiffies = jiffies;
X
Xiaoguang Wang 已提交
406
#ifdef CONFIG_SCHEDSTATS
407 408
	handle->h_sched_wait_sum = current->se.statistics.wait_sum;
	handle->h_io_wait_sum = current->se.statistics.iowait_sum;
X
Xiaoguang Wang 已提交
409
#endif
410 411 412 413

	return handle;
}

J
Jan Kara 已提交
414 415 416
handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
			      gfp_t gfp_mask, unsigned int type,
			      unsigned int line_no)
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
{
	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
433 434 435 436 437 438 439 440 441 442 443 444
	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;
	}
445

446
	err = start_this_handle(journal, handle, gfp_mask);
447
	if (err < 0) {
J
Jan Kara 已提交
448 449
		if (handle->h_rsv_handle)
			jbd2_free_handle(handle->h_rsv_handle);
M
Mingming Cao 已提交
450
		jbd2_free_handle(handle);
451
		return ERR_PTR(err);
452
	}
453 454 455 456 457
	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);
458

459 460
	return handle;
}
461 462 463
EXPORT_SYMBOL(jbd2__journal_start);


464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
/**
 * 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.
 */
483 484
handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
{
J
Jan Kara 已提交
485
	return jbd2__journal_start(journal, nblocks, 0, GFP_NOFS, 0, 0);
486 487 488
}
EXPORT_SYMBOL(jbd2_journal_start);

J
Jan Kara 已提交
489 490 491 492 493 494 495 496 497 498 499
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);

/**
500
 * int jbd2_journal_start_reserved() - start reserved handle
J
Jan Kara 已提交
501
 * @handle: handle to start
502 503
 * @type: for handle statistics
 * @line_no: for handle statistics
J
Jan Kara 已提交
504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538
 *
 * 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);
539
	if (ret < 0) {
540
		handle->h_journal = journal;
J
Jan Kara 已提交
541
		jbd2_journal_free_reserved(handle);
542 543
		return ret;
	}
J
Jan Kara 已提交
544 545
	handle->h_type = type;
	handle->h_line_no = line_no;
546
	return 0;
J
Jan Kara 已提交
547 548
}
EXPORT_SYMBOL(jbd2_journal_start_reserved);
549 550

/**
551
 * int jbd2_journal_extend() - extend buffer credits.
552 553 554 555 556
 * @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
557
 * a credit for a number of buffer modifications in advance, but can
558 559
 * extend its credit if it needs more.
 *
560
 * jbd2_journal_extend tries to give the running handle more buffer credits.
561 562 563 564 565 566 567 568 569
 * 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.
 */
570
int jbd2_journal_extend(handle_t *handle, int nblocks)
571 572
{
	transaction_t *transaction = handle->h_transaction;
573
	journal_t *journal;
574 575 576 577
	int result;
	int wanted;

	if (is_handle_aborted(handle))
578 579
		return -EROFS;
	journal = transaction->t_journal;
580 581 582

	result = 1;

583
	read_lock(&journal->j_state_lock);
584 585

	/* Don't extend a locked-down transaction! */
586
	if (transaction->t_state != T_RUNNING) {
587 588 589 590 591 592
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction not running\n", handle, nblocks);
		goto error_out;
	}

	spin_lock(&transaction->t_handle_lock);
593 594
	wanted = atomic_add_return(nblocks,
				   &transaction->t_outstanding_credits);
595 596 597 598

	if (wanted > journal->j_max_transaction_buffers) {
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction too large\n", handle, nblocks);
599
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
600 601 602
		goto unlock;
	}

603 604
	if (wanted + (wanted >> JBD2_CONTROL_BLOCKS_SHIFT) >
	    jbd2_log_space_left(journal)) {
605 606
		jbd_debug(3, "denied handle %p %d blocks: "
			  "insufficient log space\n", handle, nblocks);
607
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
608 609 610
		goto unlock;
	}

611
	trace_jbd2_handle_extend(journal->j_fs_dev->bd_dev,
612
				 transaction->t_tid,
613 614 615 616
				 handle->h_type, handle->h_line_no,
				 handle->h_buffer_credits,
				 nblocks);

617
	handle->h_buffer_credits += nblocks;
618
	handle->h_requested_credits += nblocks;
619 620 621 622 623 624
	result = 0;

	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
unlock:
	spin_unlock(&transaction->t_handle_lock);
error_out:
625
	read_unlock(&journal->j_state_lock);
626 627 628 629 630
	return result;
}


/**
631
 * int jbd2_journal_restart() - restart a handle .
632 633
 * @handle:  handle to restart
 * @nblocks: nr credits requested
634
 * @gfp_mask: memory allocation flags (for start_this_handle)
635 636 637 638
 *
 * Restart a handle for a multi-transaction filesystem
 * operation.
 *
639 640
 * 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
641
 * handle's transaction so far and reattach the handle to a new
642
 * transaction capable of guaranteeing the requested number of
J
Jan Kara 已提交
643 644
 * credits. We preserve reserved handle if there's any attached to the
 * passed in handle.
645
 */
D
Dan Carpenter 已提交
646
int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask)
647 648
{
	transaction_t *transaction = handle->h_transaction;
649
	journal_t *journal;
650 651
	tid_t		tid;
	int		need_to_start, ret;
652 653 654 655 656

	/* 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;
657
	journal = transaction->t_journal;
658 659 660 661 662

	/*
	 * First unlink the handle from its current transaction, and start the
	 * commit on that.
	 */
663
	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
664 665
	J_ASSERT(journal_current_handle() == handle);

666
	read_lock(&journal->j_state_lock);
667
	spin_lock(&transaction->t_handle_lock);
668 669
	atomic_sub(handle->h_buffer_credits,
		   &transaction->t_outstanding_credits);
J
Jan Kara 已提交
670 671 672 673
	if (handle->h_rsv_handle) {
		sub_reserved_credits(journal,
				     handle->h_rsv_handle->h_buffer_credits);
	}
674
	if (atomic_dec_and_test(&transaction->t_updates))
675
		wake_up(&journal->j_wait_updates);
676
	tid = transaction->t_tid;
677
	spin_unlock(&transaction->t_handle_lock);
678 679
	handle->h_transaction = NULL;
	current->journal_info = NULL;
680 681

	jbd_debug(2, "restarting handle %p\n", handle);
682
	need_to_start = !tid_geq(journal->j_commit_request, tid);
683
	read_unlock(&journal->j_state_lock);
684 685
	if (need_to_start)
		jbd2_log_start_commit(journal, tid);
686

687
	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
688
	handle->h_buffer_credits = nblocks;
689 690 691 692 693 694
	/*
	 * 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);
695
	ret = start_this_handle(journal, handle, gfp_mask);
696 697
	return ret;
}
698
EXPORT_SYMBOL(jbd2__journal_restart);
699 700


701 702 703 704 705 706
int jbd2_journal_restart(handle_t *handle, int nblocks)
{
	return jbd2__journal_restart(handle, nblocks, GFP_NOFS);
}
EXPORT_SYMBOL(jbd2_journal_restart);

707
/**
708
 * void jbd2_journal_lock_updates () - establish a transaction barrier.
709 710 711 712 713 714 715 716
 * @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.
 */
717
void jbd2_journal_lock_updates(journal_t *journal)
718 719 720
{
	DEFINE_WAIT(wait);

721 722
	jbd2_might_wait_for_commit(journal);

723
	write_lock(&journal->j_state_lock);
724 725
	++journal->j_barrier_count;

J
Jan Kara 已提交
726 727 728 729 730 731 732 733
	/* 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);
	}

734 735 736 737 738 739 740 741
	/* 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);
742 743
		prepare_to_wait(&journal->j_wait_updates, &wait,
				TASK_UNINTERRUPTIBLE);
744
		if (!atomic_read(&transaction->t_updates)) {
745
			spin_unlock(&transaction->t_handle_lock);
746
			finish_wait(&journal->j_wait_updates, &wait);
747 748 749
			break;
		}
		spin_unlock(&transaction->t_handle_lock);
750
		write_unlock(&journal->j_state_lock);
751 752
		schedule();
		finish_wait(&journal->j_wait_updates, &wait);
753
		write_lock(&journal->j_state_lock);
754
	}
755
	write_unlock(&journal->j_state_lock);
756 757 758

	/*
	 * We have now established a barrier against other normal updates, but
759
	 * we also need to barrier against other jbd2_journal_lock_updates() calls
760 761 762 763 764 765 766
	 * to make sure that we serialise special journal-locked operations
	 * too.
	 */
	mutex_lock(&journal->j_barrier);
}

/**
767
 * void jbd2_journal_unlock_updates (journal_t* journal) - release barrier
768 769
 * @journal:  Journal to release the barrier on.
 *
770
 * Release a transaction barrier obtained with jbd2_journal_lock_updates().
771 772 773
 *
 * Should be called without the journal lock held.
 */
774
void jbd2_journal_unlock_updates (journal_t *journal)
775 776 777 778
{
	J_ASSERT(journal->j_barrier_count != 0);

	mutex_unlock(&journal->j_barrier);
779
	write_lock(&journal->j_state_lock);
780
	--journal->j_barrier_count;
781
	write_unlock(&journal->j_state_lock);
782 783 784
	wake_up(&journal->j_wait_transaction_locked);
}

785
static void warn_dirty_buffer(struct buffer_head *bh)
786
{
787
	printk(KERN_WARNING
788
	       "JBD2: Spotted dirty metadata buffer (dev = %pg, blocknr = %llu). "
789 790
	       "There's a risk of filesystem corruption in case of system "
	       "crash.\n",
791
	       bh->b_bdev, (unsigned long long)bh->b_blocknr);
792 793
}

794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817
/* 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;
}

818 819 820 821 822 823 824 825 826 827 828 829 830 831 832
/*
 * 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;
833
	transaction_t *transaction = handle->h_transaction;
834 835 836
	journal_t *journal;
	int error;
	char *frozen_buffer = NULL;
837
	unsigned long start_lock, time_lock;
838 839 840 841 842

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

843
	jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy);
844 845 846 847 848 849 850

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

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

851
 	start_lock = jiffies;
852 853 854
	lock_buffer(bh);
	jbd_lock_bh_state(bh);

855 856 857 858 859 860
	/* 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));

861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886
	/* 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);
887
			warn_dirty_buffer(bh);
888 889 890 891 892 893
		}
		/*
		 * 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.
		 */
894 895 896
		JBUFFER_TRACE(jh, "Journalling dirty buffer");
		clear_buffer_dirty(bh);
		set_buffer_jbddirty(bh);
897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915
	}

	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;

916 917 918 919 920 921
	/*
	 * this is the first time this transaction is touching this buffer,
	 * reset the modified flag
	 */
       jh->b_modified = 0;

922 923 924 925 926 927 928 929 930
	/*
	 * 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");
931 932 933 934 935 936
		/*
		 * 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();
937 938 939 940 941
		spin_lock(&journal->j_list_lock);
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
		spin_unlock(&journal->j_list_lock);
		goto done;
	}
942 943 944 945 946 947 948
	/*
	 * 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);
949
		goto attach_next;
950 951
	}

952 953 954
	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);
955

956 957 958 959 960 961 962 963 964 965 966 967
	/*
	 * 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);
968
		task_set_wait_res(TASK_WAIT_PAGE, bh->b_page);
969
		wait_on_bit_io(&bh->b_state, BH_Shadow, TASK_UNINTERRUPTIBLE);
970
		task_clear_wait_res();
971 972
		goto repeat;
	}
973

974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
	/*
	 * 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");
990
			jbd_unlock_bh_state(bh);
991 992
			frozen_buffer = jbd2_alloc(jh2bh(jh)->b_size,
						   GFP_NOFS | __GFP_NOFAIL);
993
			goto repeat;
994
		}
995 996 997
		jh->b_frozen_data = frozen_buffer;
		frozen_buffer = NULL;
		jbd2_freeze_jh_data(jh);
998
	}
999 1000 1001 1002 1003 1004 1005
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();
1006
	jh->b_next_transaction = transaction;
1007 1008 1009 1010 1011 1012 1013 1014

done:
	jbd_unlock_bh_state(bh);

	/*
	 * If we are about to journal a buffer, then any revoke pending on it is
	 * no longer valid
	 */
1015
	jbd2_journal_cancel_revoke(handle, jh);
1016 1017 1018

out:
	if (unlikely(frozen_buffer))	/* It's usually NULL */
M
Mingming Cao 已提交
1019
		jbd2_free(frozen_buffer, bh->b_size);
1020 1021 1022 1023 1024

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

1025
/* Fast check whether buffer is already attached to the required transaction */
1026 1027
static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh,
							bool undo)
1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053
{
	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;
1054 1055 1056
	/* For undo access buffer must have data copied */
	if (undo && !jh->b_committed_data)
		goto out;
1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
	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;
}

1078
/**
1079
 * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update.
1080 1081 1082
 * @handle: transaction to add buffer modifications to
 * @bh:     bh to be used for metadata writes
 *
1083
 * Returns: error code or 0 on success.
1084 1085
 *
 * In full data journalling mode the buffer may be of type BJ_AsyncData,
1086
 * because we're ``write()ing`` a buffer which is also part of a shared mapping.
1087 1088
 */

1089
int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh)
1090
{
1091
	struct journal_head *jh;
1092 1093
	int rc;

1094
	if (jbd2_write_access_granted(handle, bh, false))
1095 1096 1097
		return 0;

	jh = jbd2_journal_add_journal_head(bh);
1098 1099 1100 1101
	/* 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);
1102
	jbd2_journal_put_journal_head(jh);
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119
	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. */

/**
1120
 * int jbd2_journal_get_create_access () - notify intent to use newly created bh
1121 1122 1123 1124 1125
 * @handle: transaction to new buffer to
 * @bh: new buffer.
 *
 * Call this if you create a new bh.
 */
1126
int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
1127 1128
{
	transaction_t *transaction = handle->h_transaction;
1129
	journal_t *journal;
1130
	struct journal_head *jh = jbd2_journal_add_journal_head(bh);
1131 1132 1133 1134 1135 1136
	int err;

	jbd_debug(5, "journal_head %p\n", jh);
	err = -EROFS;
	if (is_handle_aborted(handle))
		goto out;
1137
	journal = transaction->t_journal;
1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
	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) {
1158 1159 1160 1161 1162 1163 1164 1165 1166
		/*
		 * 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));
1167 1168 1169
		/* first access by this transaction */
		jh->b_modified = 0;

1170
		JBUFFER_TRACE(jh, "file as BJ_Reserved");
1171
		spin_lock(&journal->j_list_lock);
1172
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
1173
		spin_unlock(&journal->j_list_lock);
1174
	} else if (jh->b_transaction == journal->j_committing_transaction) {
1175 1176 1177
		/* first access by this transaction */
		jh->b_modified = 0;

1178
		JBUFFER_TRACE(jh, "set next transaction");
1179
		spin_lock(&journal->j_list_lock);
1180
		jh->b_next_transaction = transaction;
1181
		spin_unlock(&journal->j_list_lock);
1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192
	}
	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");
1193
	jbd2_journal_cancel_revoke(handle, jh);
1194
out:
1195
	jbd2_journal_put_journal_head(jh);
1196 1197 1198 1199
	return err;
}

/**
1200
 * int jbd2_journal_get_undo_access() -  Notify intent to modify metadata with
1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211
 *     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.
 *
1212
 * To deal with that, jbd2_journal_get_undo_access requests write access to a
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
 * 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.
 */
1225
int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
1226 1227
{
	int err;
1228
	struct journal_head *jh;
1229 1230
	char *committed_data = NULL;

1231
	if (jbd2_write_access_granted(handle, bh, true))
1232
		return 0;
1233

1234
	jh = jbd2_journal_add_journal_head(bh);
1235 1236
	JBUFFER_TRACE(jh, "entry");

1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
	/*
	 * 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:
1247 1248 1249
	if (!jh->b_committed_data)
		committed_data = jbd2_alloc(jh2bh(jh)->b_size,
					    GFP_NOFS|__GFP_NOFAIL);
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266

	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:
1267
	jbd2_journal_put_journal_head(jh);
1268
	if (unlikely(committed_data))
M
Mingming Cao 已提交
1269
		jbd2_free(committed_data, bh->b_size);
1270 1271 1272
	return err;
}

J
Joel Becker 已提交
1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
/**
 * 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)
{
1287
	struct journal_head *jh = jbd2_journal_grab_journal_head(bh);
J
Joel Becker 已提交
1288

1289 1290
	if (WARN_ON(!jh))
		return;
J
Joel Becker 已提交
1291
	jh->b_triggers = type;
1292
	jbd2_journal_put_journal_head(jh);
J
Joel Becker 已提交
1293 1294
}

1295
void jbd2_buffer_frozen_trigger(struct journal_head *jh, void *mapped_data,
J
Joel Becker 已提交
1296 1297 1298 1299
				struct jbd2_buffer_trigger_type *triggers)
{
	struct buffer_head *bh = jh2bh(jh);

1300
	if (!triggers || !triggers->t_frozen)
J
Joel Becker 已提交
1301 1302
		return;

1303
	triggers->t_frozen(triggers, bh, mapped_data, bh->b_size);
J
Joel Becker 已提交
1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
}

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));
}

1315
/**
1316
 * int jbd2_journal_dirty_metadata() -  mark a buffer as containing dirty metadata
1317 1318 1319 1320 1321 1322
 * @handle: transaction to add buffer to.
 * @bh: buffer to mark
 *
 * mark dirty metadata which needs to be journaled as part of the current
 * transaction.
 *
1323 1324 1325 1326
 * 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.
 *
1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337
 * 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.
 */
1338
int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
1339 1340
{
	transaction_t *transaction = handle->h_transaction;
1341
	journal_t *journal;
1342
	struct journal_head *jh;
1343
	int ret = 0;
1344 1345

	if (is_handle_aborted(handle))
1346
		return -EROFS;
1347 1348 1349
	if (!buffer_jbd(bh))
		return -EUCLEAN;

1350 1351 1352 1353 1354
	/*
	 * We don't grab jh reference here since the buffer must be part
	 * of the running transaction.
	 */
	jh = bh2jh(bh);
1355 1356 1357
	jbd_debug(5, "journal_head %p\n", jh);
	JBUFFER_TRACE(jh, "entry");

1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
	/*
	 * 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);
1376 1377 1378 1379 1380 1381 1382
			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);
1383 1384 1385 1386 1387 1388 1389 1390
			J_ASSERT_JH(jh, jh->b_transaction != transaction ||
					jh->b_jlist == BJ_Metadata);
			jbd_unlock_bh_state(bh);
		}
		goto out;
	}

	journal = transaction->t_journal;
1391 1392 1393 1394 1395 1396 1397 1398
	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
		 */
1399 1400 1401 1402
		if (handle->h_buffer_credits <= 0) {
			ret = -ENOSPC;
			goto out_unlock_bh;
		}
1403
		jh->b_modified = 1;
1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415
		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");
1416 1417
		if (unlikely(jh->b_transaction !=
			     journal->j_running_transaction)) {
1418
			printk(KERN_ERR "JBD2: %s: "
1419
			       "jh->b_transaction (%llu, %p, %u) != "
1420
			       "journal->j_running_transaction (%p, %u)\n",
1421 1422 1423 1424 1425 1426 1427 1428 1429
			       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;
		}
1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442
		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");
1443 1444 1445 1446 1447 1448 1449 1450
		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",
1451 1452
			       journal->j_devname,
			       (unsigned long long) bh->b_blocknr,
1453
			       transaction, transaction->t_tid,
1454
			       jh->b_transaction,
1455 1456
			       jh->b_transaction ?
			       jh->b_transaction->t_tid : 0,
1457 1458 1459
			       jh->b_next_transaction,
			       jh->b_next_transaction ?
			       jh->b_next_transaction->t_tid : 0,
1460 1461
			       jh->b_jlist);
			WARN_ON(1);
1462 1463
			ret = -EINVAL;
		}
1464 1465 1466 1467 1468 1469
		/* 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: */
1470
	J_ASSERT_JH(jh, jh->b_frozen_data == NULL);
1471 1472 1473

	JBUFFER_TRACE(jh, "file as BJ_Metadata");
	spin_lock(&journal->j_list_lock);
1474
	__jbd2_journal_file_buffer(jh, transaction, BJ_Metadata);
1475 1476 1477 1478 1479
	spin_unlock(&journal->j_list_lock);
out_unlock_bh:
	jbd_unlock_bh_state(bh);
out:
	JBUFFER_TRACE(jh, "exit");
1480
	return ret;
1481 1482 1483
}

/**
1484
 * void jbd2_journal_forget() - bforget() for potentially-journaled buffers.
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499
 * @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.
 */
1500
int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
1501 1502
{
	transaction_t *transaction = handle->h_transaction;
1503
	journal_t *journal;
1504 1505 1506
	struct journal_head *jh;
	int drop_reserve = 0;
	int err = 0;
1507
	int was_modified = 0;
1508

1509 1510 1511 1512
	if (is_handle_aborted(handle))
		return -EROFS;
	journal = transaction->t_journal;

1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528
	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;
	}

1529
	/* keep track of whether or not this transaction modified us */
1530 1531
	was_modified = jh->b_modified;

1532 1533 1534 1535 1536 1537
	/*
	 * The buffer's going from the transaction, we must drop
	 * all references -bzzz
	 */
	jh->b_modified = 0;

1538
	if (jh->b_transaction == transaction) {
1539 1540 1541 1542 1543 1544 1545 1546 1547 1548
		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");

1549 1550 1551 1552 1553 1554
		/*
		 * we only want to drop a reference if this transaction
		 * modified the buffer
		 */
		if (was_modified)
			drop_reserve = 1;
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567

		/*
		 * 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.
		 */

1568
		spin_lock(&journal->j_list_lock);
1569
		if (jh->b_cp_transaction) {
1570 1571
			__jbd2_journal_temp_unlink_buffer(jh);
			__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
1572
		} else {
1573
			__jbd2_journal_unfile_buffer(jh);
1574 1575
			if (!buffer_jbd(bh)) {
				spin_unlock(&journal->j_list_lock);
1576
				goto not_jbd;
1577 1578
			}
		}
1579
		spin_unlock(&journal->j_list_lock);
1580 1581 1582 1583 1584 1585
	} 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");
1586 1587 1588 1589 1590 1591
		/* ... but we CAN drop it from the new transaction through
		 * marking the buffer as freed and set j_next_transaction to
		 * the new transaction, so that not only the commit code
		 * knows it should clear dirty bits when it is done with the
		 * buffer, but also the buffer can be checkpointed only
		 * after the new transaction commits. */
1592

1593 1594 1595
		set_buffer_freed(bh);

		if (!jh->b_next_transaction) {
1596
			spin_lock(&journal->j_list_lock);
1597
			jh->b_next_transaction = transaction;
1598
			spin_unlock(&journal->j_list_lock);
1599 1600
		} else {
			J_ASSERT(jh->b_next_transaction == transaction);
1601 1602 1603 1604 1605 1606 1607

			/*
			 * only drop a reference if this transaction modified
			 * the buffer
			 */
			if (was_modified)
				drop_reserve = 1;
1608
		}
1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640
	} else {
		/*
		 * Finally, if the buffer is not belongs to any
		 * transaction, we can just drop it now if it has no
		 * checkpoint.
		 */
		spin_lock(&journal->j_list_lock);
		if (!jh->b_cp_transaction) {
			JBUFFER_TRACE(jh, "belongs to none transaction");
			spin_unlock(&journal->j_list_lock);
			goto not_jbd;
		}

		/*
		 * Otherwise, if the buffer has been written to disk,
		 * it is safe to remove the checkpoint and drop it.
		 */
		if (!buffer_dirty(bh)) {
			__jbd2_journal_remove_checkpoint(jh);
			spin_unlock(&journal->j_list_lock);
			goto not_jbd;
		}

		/*
		 * The buffer is still not written to disk, we should
		 * attach this buffer to current transaction so that the
		 * buffer can be checkpointed only after the current
		 * transaction commits.
		 */
		clear_buffer_dirty(bh);
		__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
		spin_unlock(&journal->j_list_lock);
1641 1642 1643 1644 1645 1646 1647 1648 1649 1650
	}

	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;
1651 1652 1653 1654 1655

not_jbd:
	jbd_unlock_bh_state(bh);
	__bforget(bh);
	goto drop;
1656 1657 1658
}

/**
1659
 * int jbd2_journal_stop() - complete a transaction
1660
 * @handle: transaction to complete.
1661 1662 1663 1664 1665 1666 1667 1668
 *
 * 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.
 *
1669
 * jbd2_journal_stop itself will not usually return an error, but it may
1670
 * do so in unusual circumstances.  In particular, expect it to
1671
 * return -EIO if a jbd2_journal_abort has been executed since the
1672 1673
 * transaction began.
 */
1674
int jbd2_journal_stop(handle_t *handle)
1675 1676
{
	transaction_t *transaction = handle->h_transaction;
1677 1678
	journal_t *journal;
	int err = 0, wait_for_commit = 0;
1679
	tid_t tid;
1680 1681
	pid_t pid;

1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697
	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;
		}
	}
1698 1699
	journal = transaction->t_journal;

1700 1701 1702 1703
	J_ASSERT(journal_current_handle() == handle);

	if (is_handle_aborted(handle))
		err = -EIO;
1704
	else
1705
		J_ASSERT(atomic_read(&transaction->t_updates) > 0);
1706 1707 1708 1709 1710 1711 1712 1713

	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);
1714
	trace_jbd2_handle_stats(journal->j_fs_dev->bd_dev,
1715
				transaction->t_tid,
1716 1717 1718 1719 1720
				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));
1721 1722 1723 1724

	/*
	 * Implement synchronous transaction batching.  If the handle
	 * was synchronous, don't force a commit immediately.  Let's
J
Josef Bacik 已提交
1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742
	 * 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.
1743
	 *
J
Josef Bacik 已提交
1744 1745 1746 1747
	 * 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.
1748 1749
	 *
	 * Setting max_batch_time to 0 disables this completely.
1750 1751
	 */
	pid = current->pid;
1752 1753
	if (handle->h_sync && journal->j_last_sync_writer != pid &&
	    journal->j_max_batch_time) {
J
Josef Bacik 已提交
1754 1755
		u64 commit_time, trans_time;

1756
		journal->j_last_sync_writer = pid;
J
Josef Bacik 已提交
1757

1758
		read_lock(&journal->j_state_lock);
J
Josef Bacik 已提交
1759
		commit_time = journal->j_average_commit_time;
1760
		read_unlock(&journal->j_state_lock);
J
Josef Bacik 已提交
1761 1762 1763 1764

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

1765 1766
		commit_time = max_t(u64, commit_time,
				    1000*journal->j_min_batch_time);
J
Josef Bacik 已提交
1767
		commit_time = min_t(u64, commit_time,
1768
				    1000*journal->j_max_batch_time);
J
Josef Bacik 已提交
1769 1770 1771 1772 1773 1774 1775

		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);
		}
1776 1777
	}

1778 1779
	if (handle->h_sync)
		transaction->t_synchronous_commit = 1;
1780
	current->journal_info = NULL;
1781 1782
	atomic_sub(handle->h_buffer_credits,
		   &transaction->t_outstanding_credits);
1783 1784 1785 1786 1787 1788 1789 1790

	/*
	 * 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 ||
1791 1792 1793
	    (atomic_read(&transaction->t_outstanding_credits) >
	     journal->j_max_transaction_buffers) ||
	    time_after_eq(jiffies, transaction->t_expires)) {
1794 1795 1796 1797 1798 1799 1800
		/* 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 */
1801
		jbd2_log_start_commit(journal, transaction->t_tid);
1802 1803

		/*
1804
		 * Special case: JBD2_SYNC synchronous updates require us
1805 1806 1807
		 * to wait for the commit to complete.
		 */
		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
1808
			wait_for_commit = 1;
1809 1810
	}

1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825
	do {
		unsigned long transaction_locked_time, delta;
		unsigned long journal_space_wait;
		u64 sched_wait, io_wait;

		transaction_locked_time = READ_ONCE(transaction->t_locked_time);
		if (!transaction_locked_time)
			break;

		delta = jiffies_to_msecs(jiffies - transaction_locked_time);
		if (delta < READ_ONCE(journal->j_stall_thresh))
			break;

		journal_space_wait = handle->h_start_jiffies -
					handle->h_pre_start_jiffies;
X
Xiaoguang Wang 已提交
1826
#ifdef CONFIG_SCHEDSTATS
1827 1828 1829 1830
		sched_wait = current->se.statistics.wait_sum -
					handle->h_sched_wait_sum;
		io_wait = current->se.statistics.iowait_sum -
					handle->h_io_wait_sum;
X
Xiaoguang Wang 已提交
1831 1832 1833 1834
#else
		sched_wait = 0;
		io_wait = 0;
#endif
1835 1836 1837 1838 1839 1840
		trace_jbd2_slow_handle_stats(journal->j_fs_dev->bd_dev,
			transaction->t_tid, 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,
			delta, jiffies_to_msecs(journal_space_wait),
J
Joseph Qi 已提交
1841 1842
			div_u64(sched_wait, NSEC_PER_MSEC),
			div_u64(io_wait, NSEC_PER_MSEC));
1843 1844
	} while (0);

1845 1846
	/*
	 * Once we drop t_updates, if it goes to zero the transaction
L
Lucas De Marchi 已提交
1847
	 * could start committing on us and eventually disappear.  So
1848 1849 1850 1851 1852 1853 1854 1855 1856 1857
	 * 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);
	}

1858
	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
1859

1860 1861 1862
	if (wait_for_commit)
		err = jbd2_log_wait_commit(journal, tid);

J
Jan Kara 已提交
1863 1864
	if (handle->h_rsv_handle)
		jbd2_journal_free_reserved(handle->h_rsv_handle);
1865
free_and_exit:
1866 1867 1868 1869 1870
	/*
	 * 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
1871
	jbd2_free_handle(handle);
1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930
	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
1931 1932 1933 1934
 * 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.
1935
 *
1936
 * Called under j_list_lock.
1937
 */
1938
static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950
{
	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)
1951
		J_ASSERT_JH(jh, transaction != NULL);
1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973

	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;
1974 1975 1976
	if (transaction && is_journal_aborted(transaction->t_journal))
		clear_buffer_jbddirty(bh);
	else if (test_clear_buffer_jbddirty(bh))
1977 1978 1979
		mark_buffer_dirty(bh);	/* Expose it to the VM */
}

1980 1981 1982 1983 1984 1985 1986 1987
/*
 * 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)
1988
{
1989
	__jbd2_journal_temp_unlink_buffer(jh);
1990
	jh->b_transaction = NULL;
1991
	jbd2_journal_put_journal_head(jh);
1992 1993
}

1994
void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh)
1995
{
1996 1997 1998 1999 2000
	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);
2001
	spin_lock(&journal->j_list_lock);
2002
	__jbd2_journal_unfile_buffer(jh);
2003
	spin_unlock(&journal->j_list_lock);
2004 2005
	jbd_unlock_bh_state(bh);
	__brelse(bh);
2006 2007 2008
}

/*
2009
 * Called from jbd2_journal_try_to_free_buffers().
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022
 *
 * 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;

2023
	if (jh->b_next_transaction != NULL || jh->b_transaction != NULL)
2024 2025 2026
		goto out;

	spin_lock(&journal->j_list_lock);
2027
	if (jh->b_cp_transaction != NULL) {
2028
		/* written-back checkpointed metadata buffer */
2029 2030
		JBUFFER_TRACE(jh, "remove from checkpoint list");
		__jbd2_journal_remove_checkpoint(jh);
2031 2032 2033 2034 2035 2036 2037
	}
	spin_unlock(&journal->j_list_lock);
out:
	return;
}

/**
2038
 * int jbd2_journal_try_to_free_buffers() - try to free page buffers.
2039 2040
 * @journal: journal for operation
 * @page: to try and free
2041
 * @gfp_mask: we use the mask to detect how hard should we try to release
2042 2043
 * buffers. If __GFP_DIRECT_RECLAIM and __GFP_FS is set, we wait for commit
 * code to release the buffers.
2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056
 *
 *
 * 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
2057
 * running transaction's ->t_datalist via __jbd2_journal_unfile_buffer.
2058 2059 2060 2061
 *
 * This may *change* the value of transaction_t->t_datalist, so anyone
 * who looks at t_datalist needs to lock against this function.
 *
2062 2063
 * 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()
2064 2065 2066 2067 2068 2069 2070 2071
 * 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?
2072 2073
 *
 * Return 0 on failure, 1 on success
2074
 */
2075
int jbd2_journal_try_to_free_buffers(journal_t *journal,
2076
				struct page *page, gfp_t gfp_mask)
2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091
{
	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
2092
		 * jbd2_journal_put_journal_head().
2093
		 */
2094
		jh = jbd2_journal_grab_journal_head(bh);
2095 2096 2097 2098 2099
		if (!jh)
			continue;

		jbd_lock_bh_state(bh);
		__journal_try_to_free_buffer(journal, bh);
2100
		jbd2_journal_put_journal_head(jh);
2101 2102 2103 2104
		jbd_unlock_bh_state(bh);
		if (buffer_jbd(bh))
			goto busy;
	} while ((bh = bh->b_this_page) != head);
2105

2106
	ret = try_to_free_buffers(page);
2107

2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130
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");
2131
		__jbd2_journal_temp_unlink_buffer(jh);
2132 2133 2134 2135 2136 2137
		/*
		 * 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);
2138
		__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
2139 2140 2141
		may_free = 0;
	} else {
		JBUFFER_TRACE(jh, "on running transaction");
2142
		__jbd2_journal_unfile_buffer(jh);
2143 2144 2145 2146 2147
	}
	return may_free;
}

/*
2148
 * jbd2_journal_invalidatepage
2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193
 *
 * 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.
 */
2194 2195
static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
				int partial_page)
2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211
{
	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;

2212
	/* OK, we have data buffer in journaled mode */
2213
	write_lock(&journal->j_state_lock);
2214 2215 2216
	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);

2217
	jh = jbd2_journal_grab_journal_head(bh);
2218 2219 2220
	if (!jh)
		goto zap_buffer_no_jh;

2221 2222 2223 2224 2225 2226 2227 2228 2229 2230
	/*
	 * 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
2231
	 * block can be reallocated and used by a different page.
2232 2233 2234
	 * 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.
2235 2236 2237 2238 2239 2240 2241 2242
	 *
	 * 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.
2243
	 */
2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256
	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 */
2257
			__jbd2_journal_remove_checkpoint(jh);
2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269
			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");
2270
			may_free = __dispose_buffer(jh,
2271
					journal->j_running_transaction);
2272
			goto zap_buffer;
2273 2274 2275 2276 2277 2278 2279
		} 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");
2280
				may_free = __dispose_buffer(jh,
2281
					journal->j_committing_transaction);
2282
				goto zap_buffer;
2283 2284 2285 2286
			} else {
				/* The orphan record's transaction has
				 * committed.  We can cleanse this buffer */
				clear_buffer_jbddirty(bh);
2287
				__jbd2_journal_remove_checkpoint(jh);
2288 2289 2290 2291
				goto zap_buffer;
			}
		}
	} else if (transaction == journal->j_committing_transaction) {
2292
		JBUFFER_TRACE(jh, "on committing transaction");
2293
		/*
2294
		 * The buffer is committing, we simply cannot touch
2295 2296 2297 2298 2299 2300 2301 2302
		 * 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);
2303
			return -EBUSY;
2304 2305
		}
		/*
2306 2307 2308 2309 2310
		 * OK, buffer won't be reachable after truncate. We just clear
		 * b_modified to not confuse transaction credit accounting, and
		 * 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.
2311
		 */
2312
		set_buffer_freed(bh);
2313 2314
		if (journal->j_running_transaction && buffer_jbddirty(bh))
			jh->b_next_transaction = journal->j_running_transaction;
2315
		jh->b_modified = 0;
2316
		jbd2_journal_put_journal_head(jh);
2317 2318
		spin_unlock(&journal->j_list_lock);
		jbd_unlock_bh_state(bh);
2319
		write_unlock(&journal->j_state_lock);
2320 2321 2322 2323 2324 2325 2326 2327 2328
		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);
2329
		JBUFFER_TRACE(jh, "on running transaction");
2330 2331 2332 2333
		may_free = __dispose_buffer(jh, transaction);
	}

zap_buffer:
2334 2335 2336 2337 2338 2339 2340 2341 2342
	/*
	 * 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;
2343
	jbd2_journal_put_journal_head(jh);
2344 2345 2346
zap_buffer_no_jh:
	spin_unlock(&journal->j_list_lock);
	jbd_unlock_bh_state(bh);
2347
	write_unlock(&journal->j_state_lock);
2348 2349 2350 2351 2352 2353
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);
2354 2355
	clear_buffer_delay(bh);
	clear_buffer_unwritten(bh);
2356 2357 2358 2359 2360
	bh->b_bdev = NULL;
	return may_free;
}

/**
2361
 * void jbd2_journal_invalidatepage()
2362 2363
 * @journal: journal to use for flush...
 * @page:    page to flush
2364 2365
 * @offset:  start of the range to invalidate
 * @length:  length of the range to invalidate
2366
 *
2367 2368 2369 2370
 * 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.
2371
 */
2372 2373
int jbd2_journal_invalidatepage(journal_t *journal,
				struct page *page,
2374 2375
				unsigned int offset,
				unsigned int length)
2376 2377
{
	struct buffer_head *head, *bh, *next;
2378
	unsigned int stop = offset + length;
2379
	unsigned int curr_off = 0;
2380
	int partial_page = (offset || length < PAGE_SIZE);
2381
	int may_free = 1;
2382
	int ret = 0;
2383 2384 2385 2386

	if (!PageLocked(page))
		BUG();
	if (!page_has_buffers(page))
2387
		return 0;
2388

2389
	BUG_ON(stop > PAGE_SIZE || stop < length);
2390

2391 2392 2393 2394 2395 2396 2397 2398 2399
	/* 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;

2400 2401 2402
		if (next_off > stop)
			return 0;

2403 2404 2405
		if (offset <= curr_off) {
			/* This block is wholly outside the truncation point */
			lock_buffer(bh);
2406
			ret = journal_unmap_buffer(journal, bh, partial_page);
2407
			unlock_buffer(bh);
2408 2409 2410
			if (ret < 0)
				return ret;
			may_free &= ret;
2411 2412 2413 2414 2415 2416
		}
		curr_off = next_off;
		bh = next;

	} while (bh != head);

2417
	if (!partial_page) {
2418 2419 2420
		if (may_free && try_to_free_buffers(page))
			J_ASSERT(!page_has_buffers(page));
	}
2421
	return 0;
2422 2423 2424 2425 2426
}

/*
 * File a buffer on the given transaction list.
 */
2427
void __jbd2_journal_file_buffer(struct journal_head *jh,
2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438
			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 ||
2439
				jh->b_transaction == NULL);
2440 2441 2442 2443 2444 2445

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

	if (jlist == BJ_Metadata || jlist == BJ_Reserved ||
	    jlist == BJ_Shadow || jlist == BJ_Forget) {
2446 2447 2448 2449 2450 2451 2452 2453 2454
		/*
		 * 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);
2455 2456 2457 2458 2459 2460
		if (test_clear_buffer_dirty(bh) ||
		    test_clear_buffer_jbddirty(bh))
			was_dirty = 1;
	}

	if (jh->b_transaction)
2461
		__jbd2_journal_temp_unlink_buffer(jh);
2462 2463
	else
		jbd2_journal_grab_journal_head(bh);
2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492
	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);
}

2493
void jbd2_journal_file_buffer(struct journal_head *jh,
2494 2495 2496 2497
				transaction_t *transaction, int jlist)
{
	jbd_lock_bh_state(jh2bh(jh));
	spin_lock(&transaction->t_journal->j_list_lock);
2498
	__jbd2_journal_file_buffer(jh, transaction, jlist);
2499 2500 2501 2502 2503 2504 2505 2506 2507 2508
	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.
 *
2509
 * Called under j_list_lock
2510
 * Called under jbd_lock_bh_state(jh2bh(jh))
2511 2512
 *
 * jh and bh may be already free when this function returns
2513
 */
2514
void __jbd2_journal_refile_buffer(struct journal_head *jh)
2515
{
2516
	int was_dirty, jlist;
2517 2518 2519 2520 2521 2522 2523 2524
	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) {
2525
		__jbd2_journal_unfile_buffer(jh);
2526 2527 2528 2529 2530 2531 2532 2533 2534
		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);
2535
	__jbd2_journal_temp_unlink_buffer(jh);
2536 2537 2538 2539 2540
	/*
	 * 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.
	 */
2541 2542
	jh->b_transaction = jh->b_next_transaction;
	jh->b_next_transaction = NULL;
2543 2544 2545 2546 2547 2548 2549
	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);
2550 2551 2552 2553 2554 2555 2556
	J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING);

	if (was_dirty)
		set_buffer_jbddirty(bh);
}

/*
2557 2558 2559 2560
 * __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.
2561
 */
2562
void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh)
2563 2564 2565
{
	struct buffer_head *bh = jh2bh(jh);

2566 2567
	/* Get reference so that buffer cannot be freed before we unlock it */
	get_bh(bh);
2568 2569
	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);
2570
	__jbd2_journal_refile_buffer(jh);
2571 2572 2573 2574
	jbd_unlock_bh_state(bh);
	spin_unlock(&journal->j_list_lock);
	__brelse(bh);
}
2575 2576 2577 2578

/*
 * File inode in the inode list of the handle's transaction
 */
2579
static int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode,
2580
		unsigned long flags, loff_t start_byte, loff_t end_byte)
2581 2582
{
	transaction_t *transaction = handle->h_transaction;
2583
	journal_t *journal;
2584 2585

	if (is_handle_aborted(handle))
2586 2587
		return -EROFS;
	journal = transaction->t_journal;
2588 2589 2590 2591 2592

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

	spin_lock(&journal->j_list_lock);
2593
	jinode->i_flags |= flags;
2594 2595 2596 2597 2598 2599 2600 2601 2602

	if (jinode->i_dirty_end) {
		jinode->i_dirty_start = min(jinode->i_dirty_start, start_byte);
		jinode->i_dirty_end = max(jinode->i_dirty_end, end_byte);
	} else {
		jinode->i_dirty_start = start_byte;
		jinode->i_dirty_end = end_byte;
	}

2603
	/* Is inode already attached where we need it? */
2604 2605 2606 2607
	if (jinode->i_transaction == transaction ||
	    jinode->i_next_transaction == transaction)
		goto done;

2608 2609 2610 2611 2612 2613 2614
	/*
	 * 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;
2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633
	/* 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;
}

2634 2635 2636
int jbd2_journal_inode_add_write(handle_t *handle, struct jbd2_inode *jinode)
{
	return jbd2_journal_file_inode(handle, jinode,
2637
			JI_WRITE_DATA | JI_WAIT_DATA, 0, LLONG_MAX);
2638 2639 2640 2641
}

int jbd2_journal_inode_add_wait(handle_t *handle, struct jbd2_inode *jinode)
{
2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658
	return jbd2_journal_file_inode(handle, jinode, JI_WAIT_DATA, 0,
			LLONG_MAX);
}

int jbd2_journal_inode_ranged_write(handle_t *handle,
		struct jbd2_inode *jinode, loff_t start_byte, loff_t length)
{
	return jbd2_journal_file_inode(handle, jinode,
			JI_WRITE_DATA | JI_WAIT_DATA, start_byte,
			start_byte + length - 1);
}

int jbd2_journal_inode_ranged_wait(handle_t *handle, struct jbd2_inode *jinode,
		loff_t start_byte, loff_t length)
{
	return jbd2_journal_file_inode(handle, jinode, JI_WAIT_DATA,
			start_byte, start_byte + length - 1);
2659 2660
}

2661
/*
2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679
 * 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).
2680
 */
2681 2682
int jbd2_journal_begin_ordered_truncate(journal_t *journal,
					struct jbd2_inode *jinode,
2683 2684
					loff_t new_size)
{
2685
	transaction_t *inode_trans, *commit_trans;
2686 2687
	int ret = 0;

2688 2689
	/* This is a quick check to avoid locking if not necessary */
	if (!jinode->i_transaction)
2690
		goto out;
2691 2692 2693
	/* 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 */
2694
	read_lock(&journal->j_state_lock);
2695
	commit_trans = journal->j_committing_transaction;
2696
	read_unlock(&journal->j_state_lock);
2697 2698 2699 2700 2701
	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,
2702 2703 2704 2705 2706 2707 2708
			new_size, LLONG_MAX);
		if (ret)
			jbd2_journal_abort(journal, ret);
	}
out:
	return ret;
}