transaction.c 81.3 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0+
2
/*
3
 * linux/fs/jbd2/transaction.c
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
 *
 * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
 *
 * Copyright 1998 Red Hat corp --- All Rights Reserved
 *
 * Generic filesystem transaction handling code; part of the ext2fs
 * journaling system.
 *
 * This file manages transactions (compound commits managed by the
 * journaling code) and handles (individual atomic operations by the
 * filesystem).
 */

#include <linux/time.h>
#include <linux/fs.h>
19
#include <linux/jbd2.h>
20 21 22 23 24
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/highmem.h>
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 67 68 69 70 71 72 73 74 75
/*
 * We reserve t_outstanding_credits >> JBD2_CONTROL_BLOCKS_SHIFT for
 * transaction descriptor blocks.
 */
#define JBD2_CONTROL_BLOCKS_SHIFT 5

static int jbd2_descriptor_blocks_per_trans(journal_t *journal)
{
	return journal->j_max_transaction_buffers >> JBD2_CONTROL_BLOCKS_SHIFT;
}

76
/*
77
 * jbd2_get_transaction: obtain a new transaction_t object.
78
 *
79
 * Simply initialise a new transaction. Initialize it in
80 81 82 83 84 85 86 87 88 89 90
 * 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.
 *
 */

91 92
static void jbd2_get_transaction(journal_t *journal,
				transaction_t *transaction)
93 94 95
{
	transaction->t_journal = journal;
	transaction->t_state = T_RUNNING;
J
Josef Bacik 已提交
96
	transaction->t_start_time = ktime_get();
97 98 99
	transaction->t_tid = journal->j_transaction_sequence++;
	transaction->t_expires = jiffies + journal->j_commit_interval;
	spin_lock_init(&transaction->t_handle_lock);
100
	atomic_set(&transaction->t_updates, 0);
J
Jan Kara 已提交
101
	atomic_set(&transaction->t_outstanding_credits,
102
		   jbd2_descriptor_blocks_per_trans(journal) +
J
Jan Kara 已提交
103
		   atomic_read(&journal->j_reserved_credits));
104
	atomic_set(&transaction->t_handle_count, 0);
105
	INIT_LIST_HEAD(&transaction->t_inode_list);
106
	INIT_LIST_HEAD(&transaction->t_private_list);
107 108

	/* Set up the commit timer for the new transaction. */
109
	journal->j_commit_timer.expires = round_jiffies_up(transaction->t_expires);
110 111 112 113
	add_timer(&journal->j_commit_timer);

	J_ASSERT(journal->j_running_transaction == NULL);
	journal->j_running_transaction = transaction;
114 115
	transaction->t_max_wait = 0;
	transaction->t_start = jiffies;
116
	transaction->t_requested = 0;
117 118 119 120 121 122 123 124 125 126
}

/*
 * 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.
 */

127
/*
128
 * Update transaction's maximum wait time, if debugging is enabled.
129 130 131 132 133 134 135 136
 *
 * 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.
 */
137 138
static inline void update_t_max_wait(transaction_t *transaction,
				     unsigned long ts)
139 140 141 142 143 144 145 146 147 148 149 150 151
{
#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
152
/*
153 154 155
 * Wait until running transaction passes to T_FLUSH state and new transaction
 * can thus be started. Also starts the commit if needed. The function expects
 * running transaction to exist and releases j_state_lock.
J
Jan Kara 已提交
156 157 158 159 160 161 162 163 164 165 166 167 168 169
 */
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);
170
	jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
171 172 173 174
	schedule();
	finish_wait(&journal->j_wait_transaction_locked, &wait);
}

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
/*
 * Wait until running transaction transitions from T_SWITCH to T_FLUSH
 * state and new transaction can thus be started. The function releases
 * j_state_lock.
 */
static void wait_transaction_switching(journal_t *journal)
	__releases(journal->j_state_lock)
{
	DEFINE_WAIT(wait);

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

J
Jan Kara 已提交
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
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.
	 */
224 225
	if (t->t_state != T_RUNNING) {
		WARN_ON_ONCE(t->t_state >= T_FLUSH);
J
Jan Kara 已提交
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
		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);
243 244 245 246 247 248 249 250

		/*
		 * 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);
251
			jbd2_might_wait_for_commit(journal);
252 253 254 255 256 257
			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
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275
		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);
276
		jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293
		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);
294
		jbd2_might_wait_for_commit(journal);
J
Jan Kara 已提交
295 296 297 298 299 300 301 302
		wait_event(journal->j_wait_reserved,
			 atomic_read(&journal->j_reserved_credits) + rsv_blocks
			 <= journal->j_max_transaction_buffers / 2);
		return 1;
	}
	return 0;
}

303 304 305 306 307 308 309
/*
 * 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.
 */

310
static int start_this_handle(journal_t *journal, handle_t *handle,
D
Dan Carpenter 已提交
311
			     gfp_t gfp_mask)
312
{
313
	transaction_t	*transaction, *new_transaction = NULL;
J
Jan Kara 已提交
314 315
	int		blocks = handle->h_buffer_credits;
	int		rsv_blocks = 0;
316
	unsigned long ts = jiffies;
317

318 319 320
	if (handle->h_rsv_handle)
		rsv_blocks = handle->h_rsv_handle->h_buffer_credits;

J
Jan Kara 已提交
321
	/*
322 323 324
	 * 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
325
	 */
326 327 328 329 330 331 332
	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);
333
		return -ENOSPC;
334 335 336 337
	}

alloc_transaction:
	if (!journal->j_running_transaction) {
338 339 340 341 342 343
		/*
		 * 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;
344 345
		new_transaction = kmem_cache_zalloc(transaction_cache,
						    gfp_mask);
346
		if (!new_transaction)
347
			return -ENOMEM;
348 349 350 351 352 353 354 355
	}

	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
	 */
356 357
repeat:
	read_lock(&journal->j_state_lock);
358
	BUG_ON(journal->j_flags & JBD2_UNMOUNT);
359
	if (is_journal_aborted(journal) ||
360
	    (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) {
361
		read_unlock(&journal->j_state_lock);
362
		jbd2_journal_free_transaction(new_transaction);
363
		return -EROFS;
364 365
	}

J
Jan Kara 已提交
366 367 368 369 370 371
	/*
	 * 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) {
372
		read_unlock(&journal->j_state_lock);
373 374 375 376 377 378
		wait_event(journal->j_wait_transaction_locked,
				journal->j_barrier_count == 0);
		goto repeat;
	}

	if (!journal->j_running_transaction) {
379 380
		read_unlock(&journal->j_state_lock);
		if (!new_transaction)
381
			goto alloc_transaction;
382
		write_lock(&journal->j_state_lock);
383
		if (!journal->j_running_transaction &&
J
Jan Kara 已提交
384
		    (handle->h_reserved || !journal->j_barrier_count)) {
385 386
			jbd2_get_transaction(journal, new_transaction);
			new_transaction = NULL;
387
		}
388 389
		write_unlock(&journal->j_state_lock);
		goto repeat;
390 391 392 393
	}

	transaction = journal->j_running_transaction;

J
Jan Kara 已提交
394 395 396 397 398
	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 {
399
		/*
J
Jan Kara 已提交
400 401
		 * We have handle reserved so we are allowed to join T_LOCKED
		 * transaction and we don't have to check for transaction size
402 403 404
		 * and journal space. But we still have to wait while running
		 * transaction is being switched to a committing one as it
		 * won't wait for any handles anymore.
405
		 */
406 407 408 409
		if (transaction->t_state == T_SWITCH) {
			wait_transaction_switching(journal);
			goto repeat;
		}
J
Jan Kara 已提交
410 411
		sub_reserved_credits(journal, blocks);
		handle->h_reserved = 0;
412 413 414
	}

	/* OK, account for the buffers that this operation expects to
415 416
	 * use and add the handle to the running transaction. 
	 */
417
	update_t_max_wait(transaction, ts);
418
	handle->h_transaction = transaction;
J
Jan Kara 已提交
419
	handle->h_requested_credits = blocks;
420
	handle->h_start_jiffies = jiffies;
421
	atomic_inc(&transaction->t_updates);
422
	atomic_inc(&transaction->t_handle_count);
J
Jan Kara 已提交
423 424
	jbd_debug(4, "Handle %p given %d credits (total %d, free %lu)\n",
		  handle, blocks,
425
		  atomic_read(&transaction->t_outstanding_credits),
426
		  jbd2_log_space_left(journal));
427
	read_unlock(&journal->j_state_lock);
428
	current->journal_info = handle;
429

430
	rwsem_acquire_read(&journal->j_trans_commit_map, 0, 0, _THIS_IP_);
431
	jbd2_journal_free_transaction(new_transaction);
432 433 434 435 436
	/*
	 * 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();
437
	return 0;
438 439 440 441 442
}

/* Allocate a new handle.  This should probably be in a slab... */
static handle_t *new_handle(int nblocks)
{
M
Mingming Cao 已提交
443
	handle_t *handle = jbd2_alloc_handle(GFP_NOFS);
444 445 446 447 448 449 450 451
	if (!handle)
		return NULL;
	handle->h_buffer_credits = nblocks;
	handle->h_ref = 1;

	return handle;
}

J
Jan Kara 已提交
452 453 454
handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks,
			      gfp_t gfp_mask, unsigned int type,
			      unsigned int line_no)
455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
{
	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
471 472 473 474 475 476 477 478 479 480 481 482
	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;
	}
483

484
	err = start_this_handle(journal, handle, gfp_mask);
485
	if (err < 0) {
J
Jan Kara 已提交
486 487
		if (handle->h_rsv_handle)
			jbd2_free_handle(handle->h_rsv_handle);
M
Mingming Cao 已提交
488
		jbd2_free_handle(handle);
489
		return ERR_PTR(err);
490
	}
491 492 493 494 495
	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);
496

497 498
	return handle;
}
499 500 501
EXPORT_SYMBOL(jbd2__journal_start);


502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520
/**
 * 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.
 */
521 522
handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
{
J
Jan Kara 已提交
523
	return jbd2__journal_start(journal, nblocks, 0, GFP_NOFS, 0, 0);
524 525 526
}
EXPORT_SYMBOL(jbd2_journal_start);

527
static void __jbd2_journal_unreserve_handle(handle_t *handle)
J
Jan Kara 已提交
528 529 530 531 532
{
	journal_t *journal = handle->h_journal;

	WARN_ON(!handle->h_reserved);
	sub_reserved_credits(journal, handle->h_buffer_credits);
533 534 535 536 537
}

void jbd2_journal_free_reserved(handle_t *handle)
{
	__jbd2_journal_unreserve_handle(handle);
J
Jan Kara 已提交
538 539 540 541 542
	jbd2_free_handle(handle);
}
EXPORT_SYMBOL(jbd2_journal_free_reserved);

/**
543
 * int jbd2_journal_start_reserved() - start reserved handle
J
Jan Kara 已提交
544
 * @handle: handle to start
545 546
 * @type: for handle statistics
 * @line_no: for handle statistics
J
Jan Kara 已提交
547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581
 *
 * 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);
582
	if (ret < 0) {
583
		handle->h_journal = journal;
J
Jan Kara 已提交
584
		jbd2_journal_free_reserved(handle);
585 586
		return ret;
	}
J
Jan Kara 已提交
587 588
	handle->h_type = type;
	handle->h_line_no = line_no;
589 590 591
	trace_jbd2_handle_start(journal->j_fs_dev->bd_dev,
				handle->h_transaction->t_tid, type,
				line_no, handle->h_buffer_credits);
592
	return 0;
J
Jan Kara 已提交
593 594
}
EXPORT_SYMBOL(jbd2_journal_start_reserved);
595 596

/**
597
 * int jbd2_journal_extend() - extend buffer credits.
598 599 600 601 602
 * @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
603
 * a credit for a number of buffer modifications in advance, but can
604 605
 * extend its credit if it needs more.
 *
606
 * jbd2_journal_extend tries to give the running handle more buffer credits.
607 608 609 610 611 612 613 614 615
 * 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.
 */
616
int jbd2_journal_extend(handle_t *handle, int nblocks)
617 618
{
	transaction_t *transaction = handle->h_transaction;
619
	journal_t *journal;
620 621 622 623
	int result;
	int wanted;

	if (is_handle_aborted(handle))
624 625
		return -EROFS;
	journal = transaction->t_journal;
626 627 628

	result = 1;

629
	read_lock(&journal->j_state_lock);
630 631

	/* Don't extend a locked-down transaction! */
632
	if (transaction->t_state != T_RUNNING) {
633 634 635 636 637 638
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction not running\n", handle, nblocks);
		goto error_out;
	}

	spin_lock(&transaction->t_handle_lock);
639 640
	wanted = atomic_add_return(nblocks,
				   &transaction->t_outstanding_credits);
641 642 643 644

	if (wanted > journal->j_max_transaction_buffers) {
		jbd_debug(3, "denied handle %p %d blocks: "
			  "transaction too large\n", handle, nblocks);
645
		atomic_sub(nblocks, &transaction->t_outstanding_credits);
646 647 648
		goto unlock;
	}

649
	trace_jbd2_handle_extend(journal->j_fs_dev->bd_dev,
650
				 transaction->t_tid,
651 652 653 654
				 handle->h_type, handle->h_line_no,
				 handle->h_buffer_credits,
				 nblocks);

655
	handle->h_buffer_credits += nblocks;
656
	handle->h_requested_credits += nblocks;
657 658 659 660 661 662
	result = 0;

	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
unlock:
	spin_unlock(&transaction->t_handle_lock);
error_out:
663
	read_unlock(&journal->j_state_lock);
664 665 666
	return result;
}

667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688
static void stop_this_handle(handle_t *handle)
{
	transaction_t *transaction = handle->h_transaction;
	journal_t *journal = transaction->t_journal;

	J_ASSERT(journal_current_handle() == handle);
	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
	current->journal_info = NULL;
	atomic_sub(handle->h_buffer_credits,
		   &transaction->t_outstanding_credits);
	if (handle->h_rsv_handle)
		__jbd2_journal_unreserve_handle(handle->h_rsv_handle);
	if (atomic_dec_and_test(&transaction->t_updates))
		wake_up(&journal->j_wait_updates);

	rwsem_release(&journal->j_trans_commit_map, 1, _THIS_IP_);
	/*
	 * 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);
}
689 690

/**
691
 * int jbd2_journal_restart() - restart a handle .
692 693
 * @handle:  handle to restart
 * @nblocks: nr credits requested
694
 * @gfp_mask: memory allocation flags (for start_this_handle)
695 696 697 698
 *
 * Restart a handle for a multi-transaction filesystem
 * operation.
 *
699 700
 * 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
701
 * handle's transaction so far and reattach the handle to a new
702
 * transaction capable of guaranteeing the requested number of
J
Jan Kara 已提交
703 704
 * credits. We preserve reserved handle if there's any attached to the
 * passed in handle.
705
 */
D
Dan Carpenter 已提交
706
int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask)
707 708
{
	transaction_t *transaction = handle->h_transaction;
709
	journal_t *journal;
710
	tid_t		tid;
711
	int		need_to_start;
712 713 714 715 716

	/* 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;
717
	journal = transaction->t_journal;
718
	tid = transaction->t_tid;
719 720 721 722 723

	/*
	 * First unlink the handle from its current transaction, and start the
	 * commit on that.
	 */
724 725
	jbd_debug(2, "restarting handle %p\n", handle);
	stop_this_handle(handle);
726
	handle->h_transaction = NULL;
727

728 729 730 731 732
	/*
	 * TODO: If we use READ_ONCE / WRITE_ONCE for j_commit_request we can
 	 * get rid of pointless j_state_lock traffic like this.
	 */
	read_lock(&journal->j_state_lock);
733
	need_to_start = !tid_geq(journal->j_commit_request, tid);
734
	read_unlock(&journal->j_state_lock);
735 736
	if (need_to_start)
		jbd2_log_start_commit(journal, tid);
737
	handle->h_buffer_credits = nblocks;
738
	return start_this_handle(journal, handle, gfp_mask);
739
}
740
EXPORT_SYMBOL(jbd2__journal_restart);
741 742


743 744 745 746 747 748
int jbd2_journal_restart(handle_t *handle, int nblocks)
{
	return jbd2__journal_restart(handle, nblocks, GFP_NOFS);
}
EXPORT_SYMBOL(jbd2_journal_restart);

749
/**
750
 * void jbd2_journal_lock_updates () - establish a transaction barrier.
751 752 753 754 755 756 757 758
 * @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.
 */
759
void jbd2_journal_lock_updates(journal_t *journal)
760 761 762
{
	DEFINE_WAIT(wait);

763 764
	jbd2_might_wait_for_commit(journal);

765
	write_lock(&journal->j_state_lock);
766 767
	++journal->j_barrier_count;

J
Jan Kara 已提交
768 769 770 771 772 773 774 775
	/* 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);
	}

776 777 778 779 780 781 782 783
	/* 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);
784 785
		prepare_to_wait(&journal->j_wait_updates, &wait,
				TASK_UNINTERRUPTIBLE);
786
		if (!atomic_read(&transaction->t_updates)) {
787
			spin_unlock(&transaction->t_handle_lock);
788
			finish_wait(&journal->j_wait_updates, &wait);
789 790 791
			break;
		}
		spin_unlock(&transaction->t_handle_lock);
792
		write_unlock(&journal->j_state_lock);
793 794
		schedule();
		finish_wait(&journal->j_wait_updates, &wait);
795
		write_lock(&journal->j_state_lock);
796
	}
797
	write_unlock(&journal->j_state_lock);
798 799 800

	/*
	 * We have now established a barrier against other normal updates, but
801
	 * we also need to barrier against other jbd2_journal_lock_updates() calls
802 803 804 805 806 807 808
	 * to make sure that we serialise special journal-locked operations
	 * too.
	 */
	mutex_lock(&journal->j_barrier);
}

/**
809
 * void jbd2_journal_unlock_updates (journal_t* journal) - release barrier
810 811
 * @journal:  Journal to release the barrier on.
 *
812
 * Release a transaction barrier obtained with jbd2_journal_lock_updates().
813 814 815
 *
 * Should be called without the journal lock held.
 */
816
void jbd2_journal_unlock_updates (journal_t *journal)
817 818 819 820
{
	J_ASSERT(journal->j_barrier_count != 0);

	mutex_unlock(&journal->j_barrier);
821
	write_lock(&journal->j_state_lock);
822
	--journal->j_barrier_count;
823
	write_unlock(&journal->j_state_lock);
824 825 826
	wake_up(&journal->j_wait_transaction_locked);
}

827
static void warn_dirty_buffer(struct buffer_head *bh)
828
{
829
	printk(KERN_WARNING
830
	       "JBD2: Spotted dirty metadata buffer (dev = %pg, blocknr = %llu). "
831 832
	       "There's a risk of filesystem corruption in case of system "
	       "crash.\n",
833
	       bh->b_bdev, (unsigned long long)bh->b_blocknr);
834 835
}

836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859
/* 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;
}

860 861 862 863 864 865 866 867 868 869 870 871 872 873 874
/*
 * 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;
875
	transaction_t *transaction = handle->h_transaction;
876 877 878
	journal_t *journal;
	int error;
	char *frozen_buffer = NULL;
879
	unsigned long start_lock, time_lock;
880 881 882 883 884

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

885
	jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy);
886 887 888 889 890 891 892

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

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

893
 	start_lock = jiffies;
894 895 896
	lock_buffer(bh);
	jbd_lock_bh_state(bh);

897 898 899 900 901 902
	/* 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));

903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928
	/* 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);
929
			warn_dirty_buffer(bh);
930 931 932 933 934 935
		}
		/*
		 * 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.
		 */
936 937 938
		JBUFFER_TRACE(jh, "Journalling dirty buffer");
		clear_buffer_dirty(bh);
		set_buffer_jbddirty(bh);
939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957
	}

	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;

958 959 960 961
	/*
	 * this is the first time this transaction is touching this buffer,
	 * reset the modified flag
	 */
962
	jh->b_modified = 0;
963

964 965 966 967 968 969 970 971 972
	/*
	 * 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");
973 974 975 976 977 978
		/*
		 * 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();
979 980 981 982 983
		spin_lock(&journal->j_list_lock);
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
		spin_unlock(&journal->j_list_lock);
		goto done;
	}
984 985 986 987 988 989 990
	/*
	 * 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);
991
		goto attach_next;
992 993
	}

994 995 996
	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);
997

998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012
	/*
	 * There is one case we have to be very careful about.  If the
	 * committing transaction is currently writing this buffer out to disk
	 * and has NOT made a copy-out, then we cannot modify the buffer
	 * contents at all right now.  The essence of copy-out is that it is
	 * the extra copy, not the primary copy, which gets journaled.  If the
	 * primary copy is already going to disk then we cannot do copy-out
	 * here.
	 */
	if (buffer_shadow(bh)) {
		JBUFFER_TRACE(jh, "on shadow: sleep");
		jbd_unlock_bh_state(bh);
		wait_on_bit_io(&bh->b_state, BH_Shadow, TASK_UNINTERRUPTIBLE);
		goto repeat;
	}
1013

1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029
	/*
	 * 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");
1030
			jbd_unlock_bh_state(bh);
1031 1032
			frozen_buffer = jbd2_alloc(jh2bh(jh)->b_size,
						   GFP_NOFS | __GFP_NOFAIL);
1033
			goto repeat;
1034
		}
1035 1036 1037
		jh->b_frozen_data = frozen_buffer;
		frozen_buffer = NULL;
		jbd2_freeze_jh_data(jh);
1038
	}
1039 1040 1041 1042 1043 1044 1045
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();
1046
	jh->b_next_transaction = transaction;
1047 1048 1049 1050 1051 1052 1053 1054

done:
	jbd_unlock_bh_state(bh);

	/*
	 * If we are about to journal a buffer, then any revoke pending on it is
	 * no longer valid
	 */
1055
	jbd2_journal_cancel_revoke(handle, jh);
1056 1057 1058

out:
	if (unlikely(frozen_buffer))	/* It's usually NULL */
M
Mingming Cao 已提交
1059
		jbd2_free(frozen_buffer, bh->b_size);
1060 1061 1062 1063 1064

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

1065
/* Fast check whether buffer is already attached to the required transaction */
1066 1067
static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh,
							bool undo)
1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093
{
	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;
1094 1095 1096
	/* For undo access buffer must have data copied */
	if (undo && !jh->b_committed_data)
		goto out;
1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117
	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;
}

1118
/**
1119
 * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update.
1120 1121 1122
 * @handle: transaction to add buffer modifications to
 * @bh:     bh to be used for metadata writes
 *
1123
 * Returns: error code or 0 on success.
1124 1125
 *
 * In full data journalling mode the buffer may be of type BJ_AsyncData,
1126
 * because we're ``write()ing`` a buffer which is also part of a shared mapping.
1127 1128
 */

1129
int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh)
1130
{
1131
	struct journal_head *jh;
1132 1133
	int rc;

1134
	if (jbd2_write_access_granted(handle, bh, false))
1135 1136 1137
		return 0;

	jh = jbd2_journal_add_journal_head(bh);
1138 1139 1140 1141
	/* 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);
1142
	jbd2_journal_put_journal_head(jh);
1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159
	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. */

/**
1160
 * int jbd2_journal_get_create_access () - notify intent to use newly created bh
1161 1162 1163 1164 1165
 * @handle: transaction to new buffer to
 * @bh: new buffer.
 *
 * Call this if you create a new bh.
 */
1166
int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
1167 1168
{
	transaction_t *transaction = handle->h_transaction;
1169
	journal_t *journal;
1170
	struct journal_head *jh = jbd2_journal_add_journal_head(bh);
1171 1172 1173 1174 1175 1176
	int err;

	jbd_debug(5, "journal_head %p\n", jh);
	err = -EROFS;
	if (is_handle_aborted(handle))
		goto out;
1177
	journal = transaction->t_journal;
1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
	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) {
1198 1199 1200 1201 1202 1203 1204 1205 1206
		/*
		 * 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));
1207 1208 1209
		/* first access by this transaction */
		jh->b_modified = 0;

1210
		JBUFFER_TRACE(jh, "file as BJ_Reserved");
1211
		spin_lock(&journal->j_list_lock);
1212
		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
1213
		spin_unlock(&journal->j_list_lock);
1214
	} else if (jh->b_transaction == journal->j_committing_transaction) {
1215 1216 1217
		/* first access by this transaction */
		jh->b_modified = 0;

1218
		JBUFFER_TRACE(jh, "set next transaction");
1219
		spin_lock(&journal->j_list_lock);
1220
		jh->b_next_transaction = transaction;
1221
		spin_unlock(&journal->j_list_lock);
1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232
	}
	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");
1233
	jbd2_journal_cancel_revoke(handle, jh);
1234
out:
1235
	jbd2_journal_put_journal_head(jh);
1236 1237 1238 1239
	return err;
}

/**
1240
 * int jbd2_journal_get_undo_access() -  Notify intent to modify metadata with
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
 *     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.
 *
1252
 * To deal with that, jbd2_journal_get_undo_access requests write access to a
1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264
 * 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.
 */
1265
int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
1266 1267
{
	int err;
1268
	struct journal_head *jh;
1269 1270
	char *committed_data = NULL;

1271
	if (jbd2_write_access_granted(handle, bh, true))
1272
		return 0;
1273

1274
	jh = jbd2_journal_add_journal_head(bh);
1275 1276
	JBUFFER_TRACE(jh, "entry");

1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
	/*
	 * 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:
1287 1288 1289
	if (!jh->b_committed_data)
		committed_data = jbd2_alloc(jh2bh(jh)->b_size,
					    GFP_NOFS|__GFP_NOFAIL);
1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306

	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:
1307
	jbd2_journal_put_journal_head(jh);
1308
	if (unlikely(committed_data))
M
Mingming Cao 已提交
1309
		jbd2_free(committed_data, bh->b_size);
1310 1311 1312
	return err;
}

J
Joel Becker 已提交
1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326
/**
 * 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)
{
1327
	struct journal_head *jh = jbd2_journal_grab_journal_head(bh);
J
Joel Becker 已提交
1328

1329 1330
	if (WARN_ON(!jh))
		return;
J
Joel Becker 已提交
1331
	jh->b_triggers = type;
1332
	jbd2_journal_put_journal_head(jh);
J
Joel Becker 已提交
1333 1334
}

1335
void jbd2_buffer_frozen_trigger(struct journal_head *jh, void *mapped_data,
J
Joel Becker 已提交
1336 1337 1338 1339
				struct jbd2_buffer_trigger_type *triggers)
{
	struct buffer_head *bh = jh2bh(jh);

1340
	if (!triggers || !triggers->t_frozen)
J
Joel Becker 已提交
1341 1342
		return;

1343
	triggers->t_frozen(triggers, bh, mapped_data, bh->b_size);
J
Joel Becker 已提交
1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354
}

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));
}

1355
/**
1356
 * int jbd2_journal_dirty_metadata() -  mark a buffer as containing dirty metadata
1357 1358 1359 1360 1361 1362
 * @handle: transaction to add buffer to.
 * @bh: buffer to mark
 *
 * mark dirty metadata which needs to be journaled as part of the current
 * transaction.
 *
1363 1364 1365 1366
 * 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.
 *
1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377
 * 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.
 */
1378
int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
1379 1380
{
	transaction_t *transaction = handle->h_transaction;
1381
	journal_t *journal;
1382
	struct journal_head *jh;
1383
	int ret = 0;
1384 1385

	if (is_handle_aborted(handle))
1386
		return -EROFS;
1387 1388 1389
	if (!buffer_jbd(bh))
		return -EUCLEAN;

1390 1391 1392 1393 1394
	/*
	 * We don't grab jh reference here since the buffer must be part
	 * of the running transaction.
	 */
	jh = bh2jh(bh);
1395 1396 1397
	jbd_debug(5, "journal_head %p\n", jh);
	JBUFFER_TRACE(jh, "entry");

1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415
	/*
	 * 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);
1416 1417 1418 1419 1420 1421 1422
			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);
1423 1424 1425 1426 1427 1428 1429 1430
			J_ASSERT_JH(jh, jh->b_transaction != transaction ||
					jh->b_jlist == BJ_Metadata);
			jbd_unlock_bh_state(bh);
		}
		goto out;
	}

	journal = transaction->t_journal;
1431 1432 1433 1434 1435 1436 1437 1438
	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
		 */
1439 1440 1441 1442
		if (handle->h_buffer_credits <= 0) {
			ret = -ENOSPC;
			goto out_unlock_bh;
		}
1443
		jh->b_modified = 1;
1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455
		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");
1456 1457
		if (unlikely(jh->b_transaction !=
			     journal->j_running_transaction)) {
1458
			printk(KERN_ERR "JBD2: %s: "
1459
			       "jh->b_transaction (%llu, %p, %u) != "
1460
			       "journal->j_running_transaction (%p, %u)\n",
1461 1462 1463 1464 1465 1466 1467 1468 1469
			       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;
		}
1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482
		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");
1483 1484 1485 1486 1487 1488 1489 1490
		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",
1491 1492
			       journal->j_devname,
			       (unsigned long long) bh->b_blocknr,
1493
			       transaction, transaction->t_tid,
1494
			       jh->b_transaction,
1495 1496
			       jh->b_transaction ?
			       jh->b_transaction->t_tid : 0,
1497 1498 1499
			       jh->b_next_transaction,
			       jh->b_next_transaction ?
			       jh->b_next_transaction->t_tid : 0,
1500 1501
			       jh->b_jlist);
			WARN_ON(1);
1502 1503
			ret = -EINVAL;
		}
1504 1505 1506 1507 1508 1509
		/* 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: */
1510
	J_ASSERT_JH(jh, jh->b_frozen_data == NULL);
1511 1512 1513

	JBUFFER_TRACE(jh, "file as BJ_Metadata");
	spin_lock(&journal->j_list_lock);
1514
	__jbd2_journal_file_buffer(jh, transaction, BJ_Metadata);
1515 1516 1517 1518 1519
	spin_unlock(&journal->j_list_lock);
out_unlock_bh:
	jbd_unlock_bh_state(bh);
out:
	JBUFFER_TRACE(jh, "exit");
1520
	return ret;
1521 1522 1523
}

/**
1524
 * void jbd2_journal_forget() - bforget() for potentially-journaled buffers.
1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539
 * @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.
 */
1540
int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
1541 1542
{
	transaction_t *transaction = handle->h_transaction;
1543
	journal_t *journal;
1544 1545 1546
	struct journal_head *jh;
	int drop_reserve = 0;
	int err = 0;
1547
	int was_modified = 0;
1548

1549 1550 1551 1552
	if (is_handle_aborted(handle))
		return -EROFS;
	journal = transaction->t_journal;

1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568
	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;
	}

1569
	/* keep track of whether or not this transaction modified us */
1570 1571
	was_modified = jh->b_modified;

1572 1573 1574 1575 1576 1577
	/*
	 * The buffer's going from the transaction, we must drop
	 * all references -bzzz
	 */
	jh->b_modified = 0;

1578
	if (jh->b_transaction == transaction) {
1579 1580 1581 1582 1583 1584 1585 1586 1587 1588
		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");

1589 1590 1591 1592 1593 1594
		/*
		 * we only want to drop a reference if this transaction
		 * modified the buffer
		 */
		if (was_modified)
			drop_reserve = 1;
1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607

		/*
		 * 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.
		 */

1608
		spin_lock(&journal->j_list_lock);
1609
		if (jh->b_cp_transaction) {
1610 1611
			__jbd2_journal_temp_unlink_buffer(jh);
			__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
1612
		} else {
1613
			__jbd2_journal_unfile_buffer(jh);
1614 1615
			if (!buffer_jbd(bh)) {
				spin_unlock(&journal->j_list_lock);
1616
				goto not_jbd;
1617 1618
			}
		}
1619
		spin_unlock(&journal->j_list_lock);
1620 1621 1622 1623 1624 1625
	} 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");
1626 1627 1628 1629 1630 1631
		/* ... 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. */
1632

1633 1634 1635
		set_buffer_freed(bh);

		if (!jh->b_next_transaction) {
1636
			spin_lock(&journal->j_list_lock);
1637
			jh->b_next_transaction = transaction;
1638
			spin_unlock(&journal->j_list_lock);
1639 1640
		} else {
			J_ASSERT(jh->b_next_transaction == transaction);
1641 1642 1643 1644 1645 1646 1647

			/*
			 * only drop a reference if this transaction modified
			 * the buffer
			 */
			if (was_modified)
				drop_reserve = 1;
1648
		}
1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680
	} 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);
1681 1682 1683 1684 1685 1686 1687 1688 1689 1690
	}

	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;
1691 1692 1693 1694 1695

not_jbd:
	jbd_unlock_bh_state(bh);
	__bforget(bh);
	goto drop;
1696 1697 1698
}

/**
1699
 * int jbd2_journal_stop() - complete a transaction
1700
 * @handle: transaction to complete.
1701 1702 1703 1704 1705 1706 1707 1708
 *
 * 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.
 *
1709
 * jbd2_journal_stop itself will not usually return an error, but it may
1710
 * do so in unusual circumstances.  In particular, expect it to
1711
 * return -EIO if a jbd2_journal_abort has been executed since the
1712 1713
 * transaction began.
 */
1714
int jbd2_journal_stop(handle_t *handle)
1715 1716
{
	transaction_t *transaction = handle->h_transaction;
1717 1718
	journal_t *journal;
	int err = 0, wait_for_commit = 0;
1719
	tid_t tid;
1720 1721
	pid_t pid;

J
Jan Kara 已提交
1722 1723 1724 1725 1726 1727 1728
	if (--handle->h_ref > 0) {
		jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
						 handle->h_ref);
		if (is_handle_aborted(handle))
			return -EIO;
		return 0;
	}
1729 1730
	if (!transaction) {
		/*
J
Jan Kara 已提交
1731 1732
		 * Handle is already detached from the transaction so there is
		 * nothing to do other than free the handle.
1733
		 */
1734
		memalloc_nofs_restore(handle->saved_alloc_context);
J
Jan Kara 已提交
1735
		goto free_and_exit;
1736
	}
1737
	journal = transaction->t_journal;
J
Jan Kara 已提交
1738
	tid = transaction->t_tid;
1739

1740 1741 1742 1743
	if (is_handle_aborted(handle))
		err = -EIO;

	jbd_debug(4, "Handle %p going down\n", handle);
1744
	trace_jbd2_handle_stats(journal->j_fs_dev->bd_dev,
J
Jan Kara 已提交
1745
				tid, handle->h_type, handle->h_line_no,
1746 1747 1748 1749
				jiffies - handle->h_start_jiffies,
				handle->h_sync, handle->h_requested_credits,
				(handle->h_requested_credits -
				 handle->h_buffer_credits));
1750 1751 1752 1753

	/*
	 * Implement synchronous transaction batching.  If the handle
	 * was synchronous, don't force a commit immediately.  Let's
J
Josef Bacik 已提交
1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771
	 * 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.
1772
	 *
J
Josef Bacik 已提交
1773 1774 1775 1776
	 * 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.
1777 1778
	 *
	 * Setting max_batch_time to 0 disables this completely.
1779 1780
	 */
	pid = current->pid;
1781 1782
	if (handle->h_sync && journal->j_last_sync_writer != pid &&
	    journal->j_max_batch_time) {
J
Josef Bacik 已提交
1783 1784
		u64 commit_time, trans_time;

1785
		journal->j_last_sync_writer = pid;
J
Josef Bacik 已提交
1786

1787
		read_lock(&journal->j_state_lock);
J
Josef Bacik 已提交
1788
		commit_time = journal->j_average_commit_time;
1789
		read_unlock(&journal->j_state_lock);
J
Josef Bacik 已提交
1790 1791 1792 1793

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

1794 1795
		commit_time = max_t(u64, commit_time,
				    1000*journal->j_min_batch_time);
J
Josef Bacik 已提交
1796
		commit_time = min_t(u64, commit_time,
1797
				    1000*journal->j_max_batch_time);
J
Josef Bacik 已提交
1798 1799 1800 1801 1802 1803 1804

		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);
		}
1805 1806
	}

1807 1808
	if (handle->h_sync)
		transaction->t_synchronous_commit = 1;
1809 1810 1811

	/*
	 * If the handle is marked SYNC, we need to set another commit
1812 1813
	 * going!  We also want to force a commit if the transaction is too
	 * old now.
1814 1815
	 */
	if (handle->h_sync ||
1816
	    time_after_eq(jiffies, transaction->t_expires)) {
1817 1818 1819 1820 1821 1822 1823
		/* 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 */
J
Jan Kara 已提交
1824
		jbd2_log_start_commit(journal, tid);
1825 1826

		/*
1827
		 * Special case: JBD2_SYNC synchronous updates require us
1828 1829 1830
		 * to wait for the commit to complete.
		 */
		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
1831
			wait_for_commit = 1;
1832 1833
	}

1834
	/*
1835 1836 1837 1838
	 * Once stop_this_handle() drops t_updates, the transaction could start
	 * committing on us and eventually disappear.  So we must not
	 * dereference transaction pointer again after calling
	 * stop_this_handle().
1839
	 */
1840
	stop_this_handle(handle);
1841

1842 1843 1844
	if (wait_for_commit)
		err = jbd2_log_wait_commit(journal, tid);

1845
free_and_exit:
1846 1847
	if (handle->h_rsv_handle)
		jbd2_free_handle(handle->h_rsv_handle);
M
Mingming Cao 已提交
1848
	jbd2_free_handle(handle);
1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 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
	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
1908 1909 1910 1911
 * 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.
1912
 *
1913
 * Called under j_list_lock.
1914
 */
1915
static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927
{
	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)
1928
		J_ASSERT_JH(jh, transaction != NULL);
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950

	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;
1951 1952 1953
	if (transaction && is_journal_aborted(transaction->t_journal))
		clear_buffer_jbddirty(bh);
	else if (test_clear_buffer_jbddirty(bh))
1954 1955 1956
		mark_buffer_dirty(bh);	/* Expose it to the VM */
}

1957 1958 1959 1960 1961 1962 1963 1964
/*
 * 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)
1965
{
1966
	__jbd2_journal_temp_unlink_buffer(jh);
1967
	jh->b_transaction = NULL;
1968
	jbd2_journal_put_journal_head(jh);
1969 1970
}

1971
void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh)
1972
{
1973 1974 1975 1976 1977
	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);
1978
	spin_lock(&journal->j_list_lock);
1979
	__jbd2_journal_unfile_buffer(jh);
1980
	spin_unlock(&journal->j_list_lock);
1981 1982
	jbd_unlock_bh_state(bh);
	__brelse(bh);
1983 1984 1985
}

/*
1986
 * Called from jbd2_journal_try_to_free_buffers().
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
 *
 * 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;

2000
	if (jh->b_next_transaction != NULL || jh->b_transaction != NULL)
2001 2002 2003
		goto out;

	spin_lock(&journal->j_list_lock);
2004
	if (jh->b_cp_transaction != NULL) {
2005
		/* written-back checkpointed metadata buffer */
2006 2007
		JBUFFER_TRACE(jh, "remove from checkpoint list");
		__jbd2_journal_remove_checkpoint(jh);
2008 2009 2010 2011 2012 2013 2014
	}
	spin_unlock(&journal->j_list_lock);
out:
	return;
}

/**
2015
 * int jbd2_journal_try_to_free_buffers() - try to free page buffers.
2016 2017
 * @journal: journal for operation
 * @page: to try and free
2018
 * @gfp_mask: we use the mask to detect how hard should we try to release
2019 2020
 * buffers. If __GFP_DIRECT_RECLAIM and __GFP_FS is set, we wait for commit
 * code to release the buffers.
2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033
 *
 *
 * 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
2034
 * running transaction's ->t_datalist via __jbd2_journal_unfile_buffer.
2035 2036 2037 2038
 *
 * This may *change* the value of transaction_t->t_datalist, so anyone
 * who looks at t_datalist needs to lock against this function.
 *
2039 2040
 * 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()
2041 2042 2043 2044 2045 2046 2047 2048
 * 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?
2049 2050
 *
 * Return 0 on failure, 1 on success
2051
 */
2052
int jbd2_journal_try_to_free_buffers(journal_t *journal,
2053
				struct page *page, gfp_t gfp_mask)
2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068
{
	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
2069
		 * jbd2_journal_put_journal_head().
2070
		 */
2071
		jh = jbd2_journal_grab_journal_head(bh);
2072 2073 2074 2075 2076
		if (!jh)
			continue;

		jbd_lock_bh_state(bh);
		__journal_try_to_free_buffer(journal, bh);
2077
		jbd2_journal_put_journal_head(jh);
2078 2079 2080 2081
		jbd_unlock_bh_state(bh);
		if (buffer_jbd(bh))
			goto busy;
	} while ((bh = bh->b_this_page) != head);
2082

2083
	ret = try_to_free_buffers(page);
2084

2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107
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");
2108
		__jbd2_journal_temp_unlink_buffer(jh);
2109 2110 2111 2112 2113 2114
		/*
		 * 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);
2115
		__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
2116 2117 2118
		may_free = 0;
	} else {
		JBUFFER_TRACE(jh, "on running transaction");
2119
		__jbd2_journal_unfile_buffer(jh);
2120 2121 2122 2123 2124
	}
	return may_free;
}

/*
2125
 * jbd2_journal_invalidatepage
2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170
 *
 * 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.
 */
2171 2172
static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
				int partial_page)
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188
{
	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;

2189
	/* OK, we have data buffer in journaled mode */
2190
	write_lock(&journal->j_state_lock);
2191 2192 2193
	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);

2194
	jh = jbd2_journal_grab_journal_head(bh);
2195 2196 2197
	if (!jh)
		goto zap_buffer_no_jh;

2198 2199 2200 2201 2202 2203 2204 2205 2206 2207
	/*
	 * 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
2208
	 * block can be reallocated and used by a different page.
2209 2210 2211
	 * 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.
2212 2213 2214 2215 2216 2217 2218 2219
	 *
	 * 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.
2220
	 */
2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233
	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 */
2234
			__jbd2_journal_remove_checkpoint(jh);
2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246
			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");
2247
			may_free = __dispose_buffer(jh,
2248
					journal->j_running_transaction);
2249
			goto zap_buffer;
2250 2251 2252 2253 2254 2255 2256
		} 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");
2257
				may_free = __dispose_buffer(jh,
2258
					journal->j_committing_transaction);
2259
				goto zap_buffer;
2260 2261 2262 2263
			} else {
				/* The orphan record's transaction has
				 * committed.  We can cleanse this buffer */
				clear_buffer_jbddirty(bh);
2264
				__jbd2_journal_remove_checkpoint(jh);
2265 2266 2267 2268
				goto zap_buffer;
			}
		}
	} else if (transaction == journal->j_committing_transaction) {
2269
		JBUFFER_TRACE(jh, "on committing transaction");
2270
		/*
2271
		 * The buffer is committing, we simply cannot touch
2272 2273 2274 2275 2276 2277 2278 2279
		 * 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);
2280
			return -EBUSY;
2281 2282 2283 2284 2285 2286
		}
		/*
		 * OK, buffer won't be reachable after truncate. We just set
		 * j_next_transaction to the running transaction (if there is
		 * one) and mark buffer as freed so that commit code knows it
		 * should clear dirty bits when it is done with the buffer.
2287
		 */
2288
		set_buffer_freed(bh);
2289 2290
		if (journal->j_running_transaction && buffer_jbddirty(bh))
			jh->b_next_transaction = journal->j_running_transaction;
2291
		jbd2_journal_put_journal_head(jh);
2292 2293
		spin_unlock(&journal->j_list_lock);
		jbd_unlock_bh_state(bh);
2294
		write_unlock(&journal->j_state_lock);
2295 2296 2297 2298 2299 2300 2301 2302 2303
		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);
2304
		JBUFFER_TRACE(jh, "on running transaction");
2305 2306 2307 2308
		may_free = __dispose_buffer(jh, transaction);
	}

zap_buffer:
2309 2310 2311 2312 2313 2314 2315 2316 2317
	/*
	 * 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;
2318
	jbd2_journal_put_journal_head(jh);
2319 2320 2321
zap_buffer_no_jh:
	spin_unlock(&journal->j_list_lock);
	jbd_unlock_bh_state(bh);
2322
	write_unlock(&journal->j_state_lock);
2323 2324 2325 2326 2327 2328
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);
2329 2330
	clear_buffer_delay(bh);
	clear_buffer_unwritten(bh);
2331 2332 2333 2334 2335
	bh->b_bdev = NULL;
	return may_free;
}

/**
2336
 * void jbd2_journal_invalidatepage()
2337 2338
 * @journal: journal to use for flush...
 * @page:    page to flush
2339 2340
 * @offset:  start of the range to invalidate
 * @length:  length of the range to invalidate
2341
 *
2342 2343 2344 2345
 * 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.
2346
 */
2347 2348
int jbd2_journal_invalidatepage(journal_t *journal,
				struct page *page,
2349 2350
				unsigned int offset,
				unsigned int length)
2351 2352
{
	struct buffer_head *head, *bh, *next;
2353
	unsigned int stop = offset + length;
2354
	unsigned int curr_off = 0;
2355
	int partial_page = (offset || length < PAGE_SIZE);
2356
	int may_free = 1;
2357
	int ret = 0;
2358 2359 2360 2361

	if (!PageLocked(page))
		BUG();
	if (!page_has_buffers(page))
2362
		return 0;
2363

2364
	BUG_ON(stop > PAGE_SIZE || stop < length);
2365

2366 2367 2368 2369 2370 2371 2372 2373 2374
	/* 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;

2375 2376 2377
		if (next_off > stop)
			return 0;

2378 2379 2380
		if (offset <= curr_off) {
			/* This block is wholly outside the truncation point */
			lock_buffer(bh);
2381
			ret = journal_unmap_buffer(journal, bh, partial_page);
2382
			unlock_buffer(bh);
2383 2384 2385
			if (ret < 0)
				return ret;
			may_free &= ret;
2386 2387 2388 2389 2390 2391
		}
		curr_off = next_off;
		bh = next;

	} while (bh != head);

2392
	if (!partial_page) {
2393 2394 2395
		if (may_free && try_to_free_buffers(page))
			J_ASSERT(!page_has_buffers(page));
	}
2396
	return 0;
2397 2398 2399 2400 2401
}

/*
 * File a buffer on the given transaction list.
 */
2402
void __jbd2_journal_file_buffer(struct journal_head *jh,
2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413
			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 ||
2414
				jh->b_transaction == NULL);
2415 2416 2417 2418 2419 2420

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

	if (jlist == BJ_Metadata || jlist == BJ_Reserved ||
	    jlist == BJ_Shadow || jlist == BJ_Forget) {
2421 2422 2423 2424 2425 2426 2427 2428 2429
		/*
		 * 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);
2430 2431 2432 2433 2434 2435
		if (test_clear_buffer_dirty(bh) ||
		    test_clear_buffer_jbddirty(bh))
			was_dirty = 1;
	}

	if (jh->b_transaction)
2436
		__jbd2_journal_temp_unlink_buffer(jh);
2437 2438
	else
		jbd2_journal_grab_journal_head(bh);
2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467
	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);
}

2468
void jbd2_journal_file_buffer(struct journal_head *jh,
2469 2470 2471 2472
				transaction_t *transaction, int jlist)
{
	jbd_lock_bh_state(jh2bh(jh));
	spin_lock(&transaction->t_journal->j_list_lock);
2473
	__jbd2_journal_file_buffer(jh, transaction, jlist);
2474 2475 2476 2477 2478 2479 2480 2481 2482 2483
	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.
 *
2484
 * Called under j_list_lock
2485
 * Called under jbd_lock_bh_state(jh2bh(jh))
2486 2487
 *
 * jh and bh may be already free when this function returns
2488
 */
2489
void __jbd2_journal_refile_buffer(struct journal_head *jh)
2490
{
2491
	int was_dirty, jlist;
2492 2493 2494 2495 2496 2497 2498 2499
	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) {
2500
		__jbd2_journal_unfile_buffer(jh);
2501 2502 2503 2504 2505 2506 2507 2508 2509
		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);
2510
	__jbd2_journal_temp_unlink_buffer(jh);
2511 2512 2513 2514 2515
	/*
	 * 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.
	 */
2516 2517
	jh->b_transaction = jh->b_next_transaction;
	jh->b_next_transaction = NULL;
2518 2519 2520 2521 2522 2523 2524
	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);
2525 2526 2527 2528 2529 2530 2531
	J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING);

	if (was_dirty)
		set_buffer_jbddirty(bh);
}

/*
2532 2533 2534 2535
 * __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.
2536
 */
2537
void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh)
2538 2539 2540
{
	struct buffer_head *bh = jh2bh(jh);

2541 2542
	/* Get reference so that buffer cannot be freed before we unlock it */
	get_bh(bh);
2543 2544
	jbd_lock_bh_state(bh);
	spin_lock(&journal->j_list_lock);
2545
	__jbd2_journal_refile_buffer(jh);
2546 2547 2548 2549
	jbd_unlock_bh_state(bh);
	spin_unlock(&journal->j_list_lock);
	__brelse(bh);
}
2550 2551 2552 2553

/*
 * File inode in the inode list of the handle's transaction
 */
2554
static int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode,
2555
		unsigned long flags, loff_t start_byte, loff_t end_byte)
2556 2557
{
	transaction_t *transaction = handle->h_transaction;
2558
	journal_t *journal;
2559 2560

	if (is_handle_aborted(handle))
2561 2562
		return -EROFS;
	journal = transaction->t_journal;
2563 2564 2565 2566 2567

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

	spin_lock(&journal->j_list_lock);
2568
	jinode->i_flags |= flags;
2569 2570 2571 2572 2573 2574 2575 2576 2577

	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;
	}

2578
	/* Is inode already attached where we need it? */
2579 2580 2581 2582
	if (jinode->i_transaction == transaction ||
	    jinode->i_next_transaction == transaction)
		goto done;

2583 2584 2585 2586 2587 2588 2589
	/*
	 * 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;
2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608
	/* 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;
}

2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621
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);
2622 2623
}

2624
/*
2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642
 * 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).
2643
 */
2644 2645
int jbd2_journal_begin_ordered_truncate(journal_t *journal,
					struct jbd2_inode *jinode,
2646 2647
					loff_t new_size)
{
2648
	transaction_t *inode_trans, *commit_trans;
2649 2650
	int ret = 0;

2651 2652
	/* This is a quick check to avoid locking if not necessary */
	if (!jinode->i_transaction)
2653
		goto out;
2654 2655 2656
	/* 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 */
2657
	read_lock(&journal->j_state_lock);
2658
	commit_trans = journal->j_committing_transaction;
2659
	read_unlock(&journal->j_state_lock);
2660 2661 2662 2663 2664
	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,
2665 2666 2667 2668 2669 2670 2671
			new_size, LLONG_MAX);
		if (ret)
			jbd2_journal_abort(journal, ret);
	}
out:
	return ret;
}