aboutsummaryrefslogtreecommitdiffstats
path: root/fs/btrfs/transaction.c
blob: 2cb116099b90207d0f1910176de8a029bd23c93f (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
/*
 * Copyright (C) 2007 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/sched.h>
#include <linux/writeback.h>
#include <linux/pagemap.h>
#include <linux/blkdev.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "locking.h"
#include "tree-log.h"

#define BTRFS_ROOT_TRANS_TAG 0

static noinline void put_transaction(struct btrfs_transaction *transaction)
{
	WARN_ON(transaction->use_count == 0);
	transaction->use_count--;
	if (transaction->use_count == 0) {
		list_del_init(&transaction->list);
		memset(transaction, 0, sizeof(*transaction));
		kmem_cache_free(btrfs_transaction_cachep, transaction);
	}
}

static noinline void switch_commit_root(struct btrfs_root *root)
{
	free_extent_buffer(root->commit_root);
	root->commit_root = btrfs_root_node(root);
}

/*
 * either allocate a new transaction or hop into the existing one
 */
static noinline int join_transaction(struct btrfs_root *root)
{
	struct btrfs_transaction *cur_trans;
	cur_trans = root->fs_info->running_transaction;
	if (!cur_trans) {
		cur_trans = kmem_cache_alloc(btrfs_transaction_cachep,
					     GFP_NOFS);
		BUG_ON(!cur_trans);
		root->fs_info->generation++;
		cur_trans->num_writers = 1;
		cur_trans->num_joined = 0;
		cur_trans->transid = root->fs_info->generation;
		init_waitqueue_head(&cur_trans->writer_wait);
		init_waitqueue_head(&cur_trans->commit_wait);
		cur_trans->in_commit = 0;
		cur_trans->blocked = 0;
		cur_trans->use_count = 1;
		cur_trans->commit_done = 0;
		cur_trans->start_time = get_seconds();

		cur_trans->delayed_refs.root = RB_ROOT;
		cur_trans->delayed_refs.num_entries = 0;
		cur_trans->delayed_refs.num_heads_ready = 0;
		cur_trans->delayed_refs.num_heads = 0;
		cur_trans->delayed_refs.flushing = 0;
		cur_trans->delayed_refs.run_delayed_start = 0;
		spin_lock_init(&cur_trans->delayed_refs.lock);

		INIT_LIST_HEAD(&cur_trans->pending_snapshots);
		list_add_tail(&cur_trans->list, &root->fs_info->trans_list);
		extent_io_tree_init(&cur_trans->dirty_pages,
				     root->fs_info->btree_inode->i_mapping,
				     GFP_NOFS);
		spin_lock(&root->fs_info->new_trans_lock);
		root->fs_info->running_transaction = cur_trans;
		spin_unlock(&root->fs_info->new_trans_lock);
	} else {
		cur_trans->num_writers++;
		cur_trans->num_joined++;
	}

	return 0;
}

/*
 * this does all the record keeping required to make sure that a reference
 * counted root is properly recorded in a given transaction.  This is required
 * to make sure the old root from before we joined the transaction is deleted
 * when the transaction commits
 */
static noinline int record_root_in_trans(struct btrfs_trans_handle *trans,
					 struct btrfs_root *root)
{
	if (root->ref_cows && root->last_trans < trans->transid) {
		WARN_ON(root == root->fs_info->extent_root);
		WARN_ON(root->commit_root != root->node);

		radix_tree_tag_set(&root->fs_info->fs_roots_radix,
			   (unsigned long)root->root_key.objectid,
			   BTRFS_ROOT_TRANS_TAG);
		root->last_trans = trans->transid;
		btrfs_init_reloc_root(trans, root);
	}
	return 0;
}

int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
			       struct btrfs_root *root)
{
	if (!root->ref_cows)
		return 0;

	mutex_lock(&root->fs_info->trans_mutex);
	if (root->last_trans == trans->transid) {
		mutex_unlock(&root->fs_info->trans_mutex);
		return 0;
	}

	record_root_in_trans(trans, root);
	mutex_unlock(&root->fs_info->trans_mutex);
	return 0;
}

/* wait for commit against the current transaction to become unblocked
 * when this is done, it is safe to start a new transaction, but the current
 * transaction might not be fully on disk.
 */
static void wait_current_trans(struct btrfs_root *root)
{
	struct btrfs_transaction *cur_trans;

	cur_trans = root->fs_info->running_transaction;
	if (cur_trans && cur_trans->blocked) {
		DEFINE_WAIT(wait);
		cur_trans->use_count++;
		while (1) {
			prepare_to_wait(&root->fs_info->transaction_wait, &wait,
					TASK_UNINTERRUPTIBLE);
			if (!cur_trans->blocked)
				break;
			mutex_unlock(&root->fs_info->trans_mutex);
			schedule();
			mutex_lock(&root->fs_info->trans_mutex);
		}
		finish_wait(&root->fs_info->transaction_wait, &wait);
		put_transaction(cur_trans);
	}
}

enum btrfs_trans_type {
	TRANS_START,
	TRANS_JOIN,
	TRANS_USERSPACE,
};

static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
					     int num_blocks, int type)
{
	struct btrfs_trans_handle *h =
		kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS);
	int ret;

	mutex_lock(&root->fs_info->trans_mutex);
	if (!root->fs_info->log_root_recovering &&
	    ((type == TRANS_START && !root->fs_info->open_ioctl_trans) ||
	     type == TRANS_USERSPACE))
		wait_current_trans(root);
	ret = join_transaction(root);
	BUG_ON(ret);

	h->transid = root->fs_info->running_transaction->transid;
	h->transaction = root->fs_info->running_transaction;
	h->blocks_reserved = num_blocks;
	h->blocks_used = 0;
	h->block_group = 0;
	h->alloc_exclude_nr = 0;
	h->alloc_exclude_start = 0;
	h->delayed_ref_updates = 0;

	if (!current->journal_info && type != TRANS_USERSPACE)
		current->journal_info = h;

	root->fs_info->running_transaction->use_count++;
	record_root_in_trans(h, root);
	mutex_unlock(&root->fs_info->trans_mutex);
	return h;
}

struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
						   int num_blocks)
{
	return start_transaction(root, num_blocks, TRANS_START);
}
struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root,
						   int num_blocks)
{
	return start_transaction(root, num_blocks, TRANS_JOIN);
}

struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *r,
							 int num_blocks)
{
	return start_transaction(r, num_blocks, TRANS_USERSPACE);
}

/* wait for a transaction commit to be fully complete */
static noinline int wait_for_commit(struct btrfs_root *root,
				    struct btrfs_transaction *commit)
{
	DEFINE_WAIT(wait);
	mutex_lock(&root->fs_info->trans_mutex);
	while (!commit->commit_done) {
		prepare_to_wait(&commit->commit_wait, &wait,
				TASK_UNINTERRUPTIBLE);
		if (commit->commit_done)
			break;
		mutex_unlock(&root->fs_info->trans_mutex);
		schedule();
		mutex_lock(&root->fs_info->trans_mutex);
	}
	mutex_unlock(&root->fs_info->trans_mutex);
	finish_wait(&commit->commit_wait, &wait);
	return 0;
}

#if 0
/*
 * rate limit against the drop_snapshot code.  This helps to slow down new
 * operations if the drop_snapshot code isn't able to keep up.
 */
static void throttle_on_drops(struct btrfs_root *root)
{
	struct btrfs_fs_info *info = root->fs_info;
	int harder_count = 0;

harder:
	if (atomic_read(&info->throttles)) {
		DEFINE_WAIT(wait);
		int thr;
		thr = atomic_read(&info->throttle_gen);

		do {
			prepare_to_wait(&info->transaction_throttle,
					&wait, TASK_UNINTERRUPTIBLE);
			if (!atomic_read(&info->throttles)) {
				finish_wait(&info->transaction_throttle, &wait);
				break;
			}
			schedule();
			finish_wait(&info->transaction_throttle, &wait);
		} while (thr == atomic_read(&info->throttle_gen));
		harder_count++;

		if (root->fs_info->total_ref_cache_size > 1 * 1024 * 1024 &&
		    harder_count < 2)
			goto harder;

		if (root->fs_info->total_ref_cache_size > 5 * 1024 * 1024 &&
		    harder_count < 10)
			goto harder;

		if (root->fs_info->total_ref_cache_size > 10 * 1024 * 1024 &&
		    harder_count < 20)
			goto harder;
	}
}
#endif

void btrfs_throttle(struct btrfs_root *root)
{
	mutex_lock(&root->fs_info->trans_mutex);
	if (!root->fs_info->open_ioctl_trans)
		wait_current_trans(root);
	mutex_unlock(&root->fs_info->trans_mutex);
}

static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
			  struct btrfs_root *root, int throttle)
{
	struct btrfs_transaction *cur_trans;
	struct btrfs_fs_info *info = root->fs_info;
	int count = 0;

	while (count < 4) {
		unsigned long cur = trans->delayed_ref_updates;
		trans->delayed_ref_updates = 0;
		if (cur &&
		    trans->transaction->delayed_refs.num_heads_ready > 64) {
			trans->delayed_ref_updates = 0;

			/*
			 * do a full flush if the transaction is trying
			 * to close
			 */
			if (trans->transaction->delayed_refs.flushing)
				cur = 0;
			btrfs_run_delayed_refs(trans, root, cur);
		} else {
			break;
		}
		count++;
	}

	mutex_lock(&info->trans_mutex);
	cur_trans = info->running_transaction;
	WARN_ON(cur_trans != trans->transaction);
	WARN_ON(cur_trans->num_writers < 1);
	cur_trans->num_writers--;

	if (waitqueue_active(&cur_trans->writer_wait))
		wake_up(&cur_trans->writer_wait);
	put_transaction(cur_trans);
	mutex_unlock(&info->trans_mutex);

	if (current->journal_info == trans)
		current->journal_info = NULL;
	memset(trans, 0, sizeof(*trans));
	kmem_cache_free(btrfs_trans_handle_cachep, trans);

	if (throttle)
		btrfs_run_delayed_iputs(root);

	return 0;
}

int btrfs_end_transaction(struct btrfs_trans_handle *trans,
			  struct btrfs_root *root)
{
	return __btrfs_end_transaction(trans, root, 0);
}

int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
				   struct btrfs_root *root)
{
	return __btrfs_end_transaction(trans, root, 1);
}

/*
 * when btree blocks are allocated, they have some corresponding bits set for
 * them in one of two extent_io trees.  This is used to make sure all of
 * those extents are sent to disk but does not wait on them
 */
int btrfs_write_marked_extents(struct btrfs_root *root,
			       struct extent_io_tree *dirty_pages, int mark)
{
	int ret;
	int err = 0;
	int werr = 0;
	struct page *page;
	struct inode *btree_inode = root->fs_info->btree_inode;
	u64 start = 0;
	u64 end;
	unsigned long index;

	while (1) {
		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
					    mark);
		if (ret)
			break;
		while (start <= end) {
			cond_resched();

			index = start >> PAGE_CACHE_SHIFT;
			start = (u64)(index + 1) << PAGE_CACHE_SHIFT;
			page = find_get_page(btree_inode->i_mapping, index);
			if (!page)
				continue;

			btree_lock_page_hook(page);
			if (!page->mapping) {
				unlock_page(page);
				page_cache_release(page);
				continue;
			}

			if (PageWriteback(page)) {
				if (PageDirty(page))
					wait_on_page_writeback(page);
				else {
					unlock_page(page);
					page_cache_release(page);
					continue;
				}
			}
			err = write_one_page(page, 0);
			if (err)
				werr = err;
			page_cache_release(page);
		}
	}
	if (err)
		werr = err;
	return werr;
}

/*
 * when btree blocks are allocated, they have some corresponding bits set for
 * them in one of two extent_io trees.  This is used to make sure all of
 * those extents are on disk for transaction or log commit.  We wait
 * on all the pages and clear them from the dirty pages state tree
 */
int btrfs_wait_marked_extents(struct btrfs_root *root,
			      struct extent_io_tree *dirty_pages, int mark)
{
	int ret;
	int err = 0;
	int werr = 0;
	struct page *page;
	struct inode *btree_inode = root->fs_info->btree_inode;
	u64 start = 0;
	u64 end;
	unsigned long index;

	while (1) {
		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
					    mark);
		if (ret)
			break;

		clear_extent_bits(dirty_pages, start, end, mark, GFP_NOFS);
		while (start <= end) {
			index = start >> PAGE_CACHE_SHIFT;
			start = (u64)(index + 1) << PAGE_CACHE_SHIFT;
			page = find_get_page(btree_inode->i_mapping, index);
			if (!page)
				continue;
			if (PageDirty(page)) {
				btree_lock_page_hook(page);
				wait_on_page_writeback(page);
				err = write_one_page(page, 0);
				if (err)
					werr = err;
			}
			wait_on_page_writeback(page);
			page_cache_release(page);
			cond_resched();
		}
	}
	if (err)
		werr = err;
	return werr;
}

/*
 * when btree blocks are allocated, they have some corresponding bits set for
 * them in one of two extent_io trees.  This is used to make sure all of
 * those extents are on disk for transaction or log commit
 */
int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
				struct extent_io_tree *dirty_pages, int mark)
{
	int ret;
	int ret2;

	ret = btrfs_write_marked_extents(root, dirty_pages, mark);
	ret2 = btrfs_wait_marked_extents(root, dirty_pages, mark);
	return ret || ret2;
}

int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
				     struct btrfs_root *root)
{
	if (!trans || !trans->transaction) {
		struct inode *btree_inode;
		btree_inode = root->fs_info->btree_inode;
		return filemap_write_and_wait(btree_inode->i_mapping);
	}
	return btrfs_write_and_wait_marked_extents(root,
					   &trans->transaction->dirty_pages,
					   EXTENT_DIRTY);
}

/*
 * this is used to update the root pointer in the tree of tree roots.
 *
 * But, in the case of the extent allocation tree, updating the root
 * pointer may allocate blocks which may change the root of the extent
 * allocation tree.
 *
 * So, this loops and repeats and makes sure the cowonly root didn't
 * change while the root pointer was being updated in the metadata.
 */
static int update_cowonly_root(struct btrfs_trans_handle *trans,
			       struct btrfs_root *root)
{
	int ret;
	u64 old_root_bytenr;
	u64 old_root_used;
	struct btrfs_root *tree_root = root->fs_info->tree_root;

	old_root_used = btrfs_root_used(&root->root_item);
	btrfs_write_dirty_block_groups(trans, root);

	while (1) {
		old_root_bytenr = btrfs_root_bytenr(&root->root_item);
		if (old_root_bytenr == root->node->start &&
		    old_root_used == btrfs_root_used(&root->root_item))
			break;

		btrfs_set_root_node(&root->root_item, root->node);
		ret = btrfs_update_root(trans, tree_root,
					&root->root_key,
					&root->root_item);
		BUG_ON(ret);

		old_root_used = btrfs_root_used(&root->root_item);
		ret = btrfs_write_dirty_block_groups(trans, root);
		BUG_ON(ret);
	}

	if (root != root->fs_info->extent_root)
		switch_commit_root(root);

	return 0;
}

/*
 * update all the cowonly tree roots on disk
 */
static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans,
					 struct btrfs_root *root)
{
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct list_head *next;
	struct extent_buffer *eb;
	int ret;

	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
	BUG_ON(ret);

	eb = btrfs_lock_root_node(fs_info->tree_root);
	btrfs_cow_block(trans, fs_info->tree_root, eb, NULL, 0, &eb);
	btrfs_tree_unlock(eb);
	free_extent_buffer(eb);

	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
	BUG_ON(ret);

	while (!list_empty(&fs_info->dirty_cowonly_roots)) {
		next = fs_info->dirty_cowonly_roots.next;
		list_del_init(next);
		root = list_entry(next, struct btrfs_root, dirty_list);

		update_cowonly_root(trans, root);
	}

	down_write(&fs_info->extent_commit_sem);
	switch_commit_root(fs_info->extent_root);
	up_write(&fs_info->extent_commit_sem);

	return 0;
}

/*
 * dead roots are old snapshots that need to be deleted.  This allocates
 * a dirty root struct and adds it into the list of dead roots that need to
 * be deleted
 */
int btrfs_add_dead_root(struct btrfs_root *root)
{
	mutex_lock(&root->fs_info->trans_mutex);
	list_add(&root->root_list, &root->fs_info->dead_roots);
	mutex_unlock(&root->fs_info->trans_mutex);
	return 0;
}

/*
 * update all the cowonly tree roots on disk
 */
static noinline int commit_fs_roots(struct btrfs_trans_handle *trans,
				    struct btrfs_root *root)
{
	struct btrfs_root *gang[8];
	struct btrfs_fs_info *fs_info = root->fs_info;
	int i;
	int ret;
	int err = 0;

	while (1) {
		ret = radix_tree_gang_lookup_tag(&fs_info->fs_roots_radix,
						 (void **)gang, 0,
						 ARRAY_SIZE(gang),
						 BTRFS_ROOT_TRANS_TAG);
		if (ret == 0)
			break;
		for (i = 0; i < ret; i++) {
			root = gang[i];
			radix_tree_tag_clear(&fs_info->fs_roots_radix,
					(unsigned long)root->root_key.objectid,
					BTRFS_ROOT_TRANS_TAG);

			btrfs_free_log(trans, root);
			btrfs_update_reloc_root(trans, root);

			if (root->commit_root != root->node) {
				switch_commit_root(root);
				btrfs_set_root_node(&root->root_item,
						    root->node);
			}

			err = btrfs_update_root(trans, fs_info->tree_root,
						&root->root_key,
						&root->root_item);
			if (err)
				break;
		}
	}
	return err;
}

/*
 * defrag a given btree.  If cacheonly == 1, this won't read from the disk,
 * otherwise every leaf in the btree is read and defragged.
 */
int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
{
	struct btrfs_fs_info *info = root->fs_info;
	int ret;
	struct btrfs_trans_handle *trans;
	unsigned long nr;

	smp_mb();
	if (root->defrag_running)
		return 0;
	trans = btrfs_start_transaction(root, 1);
	while (1) {
		root->defrag_running = 1;
		ret = btrfs_defrag_leaves(trans, root, cacheonly);
		nr = trans->blocks_used;
		btrfs_end_transaction(trans, root);
		btrfs_btree_balance_dirty(info->tree_root, nr);
		cond_resched();

		trans = btrfs_start_transaction(root, 1);
		if (root->fs_info->closing || ret != -EAGAIN)
			break;
	}
	root->defrag_running = 0;
	smp_mb();
	btrfs_end_transaction(trans, root);
	return 0;
}

#if 0
/*
 * when dropping snapshots, we generate a ton of delayed refs, and it makes
 * sense not to join the transaction while it is trying to flush the current
 * queue of delayed refs out.
 *
 * This is used by the drop snapshot code only
 */
static noinline int wait_transaction_pre_flush(struct btrfs_fs_info *info)
{
	DEFINE_WAIT(wait);

	mutex_lock(&info->trans_mutex);
	while (info->running_transaction &&
	       info->running_transaction->delayed_refs.flushing) {
		prepare_to_wait(&info->transaction_wait, &wait,
				TASK_UNINTERRUPTIBLE);
		mutex_unlock(&info->trans_mutex);

		schedule();

		mutex_lock(&info->trans_mutex);
		finish_wait(&info->transaction_wait, &wait);
	}
	mutex_unlock(&info->trans_mutex);
	return 0;
}

/*
 * Given a list of roots that need to be deleted, call btrfs_drop_snapshot on
 * all of them
 */
int btrfs_drop_dead_root(struct btrfs_root *root)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_root *tree_root = root->fs_info->tree_root;
	unsigned long nr;
	int ret;

	while (1) {
		/*
		 * we don't want to jump in and create a bunch of
		 * delayed refs if the transaction is starting to close
		 */
		wait_transaction_pre_flush(tree_root->fs_info);
		trans = btrfs_start_transaction(tree_root, 1);

		/*
		 * we've joined a transaction, make sure it isn't
		 * closing right now
		 */
		if (trans->transaction->delayed_refs.flushing) {
			btrfs_end_transaction(trans, tree_root);
			continue;
		}

		ret = btrfs_drop_snapshot(trans, root);
		if (ret != -EAGAIN)
			break;

		ret = btrfs_update_root(trans, tree_root,
					&root->root_key,
					&root->root_item);
		if (ret)
			break;

		nr = trans->blocks_used;
		ret = btrfs_end_transaction(trans, tree_root);
		BUG_ON(ret);

		btrfs_btree_balance_dirty(tree_root, nr);
		cond_resched();
	}
	BUG_ON(ret);

	ret = btrfs_del_root(trans, tree_root, &root->root_key);
	BUG_ON(ret);

	nr = trans->blocks_used;
	ret = btrfs_end_transaction(trans, tree_root);
	BUG_ON(ret);

	free_extent_buffer(root->node);
	free_extent_buffer(root->commit_root);
	kfree(root);

	btrfs_btree_balance_dirty(tree_root, nr);
	return ret;
}
#endif

/*
 * new snapshots need to be created at a very specific time in the
 * transaction commit.  This does the actual creation
 */
static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
				   struct btrfs_fs_info *fs_info,
				   struct btrfs_pending_snapshot *pending)
{
	struct btrfs_key key;
	struct btrfs_root_item *new_root_item;
	struct btrfs_root *tree_root = fs_info->tree_root;
	struct btrfs_root *root = pending->root;
	struct btrfs_root *parent_root;
	struct inode *parent_inode;
	struct extent_buffer *tmp;
	struct extent_buffer *old;
	int ret;
	u64 objectid;
	int namelen;
	u64 index = 0;

	parent_inode = pending->dentry->d_parent->d_inode;
	parent_root = BTRFS_I(parent_inode)->root;

	new_root_item = kmalloc(sizeof(*new_root_item), GFP_NOFS);
	if (!new_root_item) {
		ret = -ENOMEM;
		goto fail;
	}
	ret = btrfs_find_free_objectid(trans, tree_root, 0, &objectid);
	if (ret)
		goto fail;

	key.objectid = objectid;
	/* record when the snapshot was created in key.offset */
	key.offset = trans->transid;
	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);

	memcpy(&pending->root_key, &key, sizeof(key));
	pending->root_key.offset = (u64)-1;

	record_root_in_trans(trans, parent_root);
	/*
	 * insert the directory item
	 */
	namelen = strlen(pending->name);
	ret = btrfs_set_inode_index(parent_inode, &index);
	BUG_ON(ret);
	ret = btrfs_insert_dir_item(trans, parent_root,
			    pending->name, namelen,
			    parent_inode->i_ino,
			    &pending->root_key, BTRFS_FT_DIR, index);
	BUG_ON(ret);

	btrfs_i_size_write(parent_inode, parent_inode->i_size + namelen * 2);
	ret = btrfs_update_inode(trans, parent_root, parent_inode);
	BUG_ON(ret);

	record_root_in_trans(trans, root);
	btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
	memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));

	old = btrfs_lock_root_node(root);
	btrfs_cow_block(trans, root, old, NULL, 0, &old);
	btrfs_set_lock_blocking(old);

	btrfs_copy_root(trans, root, old, &tmp, objectid);
	btrfs_tree_unlock(old);
	free_extent_buffer(old);

	btrfs_set_root_node(new_root_item, tmp);
	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
				new_root_item);
	BUG_ON(ret);
	btrfs_tree_unlock(tmp);
	free_extent_buffer(tmp);

	ret = btrfs_add_root_ref(trans, parent_root->fs_info->tree_root,
				 pending->root_key.objectid,
				 parent_root->root_key.objectid,
				 parent_inode->i_ino, index, pending->name,
				 namelen);
	BUG_ON(ret);

fail:
	kfree(new_root_item);
	return ret;
}

/*
 * create all the snapshots we've scheduled for creation
 */
static noinline int create_pending_snapshots(struct btrfs_trans_handle *trans,
					     struct btrfs_fs_info *fs_info)
{
	struct btrfs_pending_snapshot *pending;
	struct list_head *head = &trans->transaction->pending_snapshots;
	int ret;

	list_for_each_entry(pending, head, list) {
		ret = create_pending_snapshot(trans, fs_info, pending);
		BUG_ON(ret);
	}
	return 0;
}

static void update_super_roots(struct btrfs_root *root)
{
	struct btrfs_root_item *root_item;
	struct btrfs_super_block *super;

	super = &root->fs_info->super_copy;

	root_item = &root->fs_info->chunk_root->root_item;
	super->chunk_root = root_item->bytenr;
	super->chunk_root_generation = root_item->generation;
	super->chunk_root_level = root_item->level;

	root_item = &root->fs_info->tree_root->root_item;
	super->root = root_item->bytenr;
	super->generation = root_item->generation;
	super->root_level = root_item->level;
}

int btrfs_transaction_in_commit(struct btrfs_fs_info *info)
{
	int ret = 0;
	spin_lock(&info->new_trans_lock);
	if (info->running_transaction)
		ret = info->running_transaction->in_commit;
	spin_unlock(&info->new_trans_lock);
	return ret;
}

int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
			     struct btrfs_root *root)
{
	unsigned long joined = 0;
	unsigned long timeout = 1;
	struct btrfs_transaction *cur_trans;
	struct btrfs_transaction *prev_trans = NULL;
	DEFINE_WAIT(wait);
	int ret;
	int should_grow = 0;
	unsigned long now = get_seconds();
	int flush_on_commit = btrfs_test_opt(root, FLUSHONCOMMIT);

	btrfs_run_ordered_operations(root, 0);

	/* make a pass through all the delayed refs we have so far
	 * any runnings procs may add more while we are here
	 */
	ret = btrfs_run_delayed_refs(trans, root, 0);
	BUG_ON(ret);

	cur_trans = trans->transaction;
	/*
	 * set the flushing flag so procs in this transaction have to
	 * start sending their work down.
	 */
	cur_trans->delayed_refs.flushing = 1;

	ret = btrfs_run_delayed_refs(trans, root, 0);
	BUG_ON(ret);

	mutex_lock(&root->fs_info->trans_mutex);
	if (cur_trans->in_commit) {
		cur_trans->use_count++;
		mutex_unlock(&root->fs_info->trans_mutex);
		btrfs_end_transaction(trans, root);

		ret = wait_for_commit(root, cur_trans);
		BUG_ON(ret);

		mutex_lock(&root->fs_info->trans_mutex);
		put_transaction(cur_trans);
		mutex_unlock(&root->fs_info->trans_mutex);

		return 0;
	}

	trans->transaction->in_commit = 1;
	trans->transaction->blocked = 1;
	if (cur_trans->list.prev != &root->fs_info->trans_list) {
		prev_trans = list_entry(cur_trans->list.prev,
					struct btrfs_transaction, list);
		if (!prev_trans->commit_done) {
			prev_trans->use_count++;
			mutex_unlock(&root->fs_info->trans_mutex);

			wait_for_commit(root, prev_trans);

			mutex_lock(&root->fs_info->trans_mutex);
			put_transaction(prev_trans);
		}
	}

	if (now < cur_trans->start_time || now - cur_trans->start_time < 1)
		should_grow = 1;

	do {
		int snap_pending = 0;
		joined = cur_trans->num_joined;
		if (!list_empty(&trans->transaction->pending_snapshots))
			snap_pending = 1;

		WARN_ON(cur_trans != trans->transaction);
		prepare_to_wait(&cur_trans->writer_wait, &wait,
				TASK_UNINTERRUPTIBLE);

		if (cur_trans->num_writers > 1)
			timeout = MAX_SCHEDULE_TIMEOUT;
		else if (should_grow)
			timeout = 1;

		mutex_unlock(&root->fs_info->trans_mutex);

		if (flush_on_commit || snap_pending) {
			btrfs_start_delalloc_inodes(root, 1);
			ret = btrfs_wait_ordered_extents(root, 0, 1);
			BUG_ON(ret);
		}

		/*
		 * rename don't use btrfs_join_transaction, so, once we
		 * set the transaction to blocked above, we aren't going
		 * to get any new ordered operations.  We can safely run
		 * it here and no for sure that nothing new will be added
		 * to the list
		 */
		btrfs_run_ordered_operations(root, 1);

		smp_mb();
		if (cur_trans->num_writers > 1 || should_grow)
			schedule_timeout(timeout);

		mutex_lock(&root->fs_info->trans_mutex);
		finish_wait(&cur_trans->writer_wait, &wait);
	} while (cur_trans->num_writers > 1 ||
		 (should_grow && cur_trans->num_joined != joined));

	ret = create_pending_snapshots(trans, root->fs_info);
	BUG_ON(ret);

	ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
	BUG_ON(ret);

	WARN_ON(cur_trans != trans->transaction);

	/* btrfs_commit_tree_roots is responsible for getting the
	 * various roots consistent with each other.  Every pointer
	 * in the tree of tree roots has to point to the most up to date
	 * root for every subvolume and other tree.  So, we have to keep
	 * the tree logging code from jumping in and changing any
	 * of the trees.
	 *
	 * At this point in the commit, there can't be any tree-log
	 * writers, but a little lower down we drop the trans mutex
	 * and let new people in.  By holding the tree_log_mutex
	 * from now until after the super is written, we avoid races
	 * with the tree-log code.
	 */
	mutex_lock(&root->fs_info->tree_log_mutex);

	ret = commit_fs_roots(trans, root);
	BUG_ON(ret);

	/* commit_fs_roots gets rid of all the tree log roots, it is now
	 * safe to free the root of tree log roots
	 */
	btrfs_free_log_root_tree(trans, root->fs_info);

	ret = commit_cowonly_roots(trans, root);
	BUG_ON(ret);

	btrfs_prepare_extent_commit(trans, root);

	cur_trans = root->fs_info->running_transaction;
	spin_lock(&root->fs_info->new_trans_lock);
	root->fs_info->running_transaction = NULL;
	spin_unlock(&root->fs_info->new_trans_lock);

	btrfs_set_root_node(&root->fs_info->tree_root->root_item,
			    root->fs_info->tree_root->node);
	switch_commit_root(root->fs_info->tree_root);

	btrfs_set_root_node(&root->fs_info->chunk_root->root_item,
			    root->fs_info->chunk_root->node);
	switch_commit_root(root->fs_info->chunk_root);

	update_super_roots(root);

	if (!root->fs_info->log_root_recovering) {
		btrfs_set_super_log_root(&root->fs_info->super_copy, 0);
		btrfs_set_super_log_root_level(&root->fs_info->super_copy, 0);
	}

	memcpy(&root->fs_info->super_for_commit, &root->fs_info->super_copy,
	       sizeof(root->fs_info->super_copy));

	trans->transaction->blocked = 0;

	wake_up(&root->fs_info->transaction_wait);

	mutex_unlock(&root->fs_info->trans_mutex);
	ret = btrfs_write_and_wait_transaction(trans, root);
	BUG_ON(ret);
	write_ctree_super(trans, root, 0);

	/*
	 * the super is written, we can safely allow the tree-loggers
	 * to go about their business
	 */
	mutex_unlock(&root->fs_info->tree_log_mutex);

	btrfs_finish_extent_commit(trans, root);

	mutex_lock(&root->fs_info->trans_mutex);

	cur_trans->commit_done = 1;

	root->fs_info->last_trans_committed = cur_trans->transid;

	wake_up(&cur_trans->commit_wait);

	put_transaction(cur_trans);
	put_transaction(cur_trans);

	mutex_unlock(&root->fs_info->trans_mutex);

	if (current->journal_info == trans)
		current->journal_info = NULL;

	kmem_cache_free(btrfs_trans_handle_cachep, trans);

	if (current != root->fs_info->transaction_kthread)
		btrfs_run_delayed_iputs(root);

	return ret;
}

/*
 * interface function to delete all the snapshots we have scheduled for deletion
 */
int btrfs_clean_old_snapshots(struct btrfs_root *root)
{
	LIST_HEAD(list);
	struct btrfs_fs_info *fs_info = root->fs_info;

	mutex_lock(&fs_info->trans_mutex);
	list_splice_init(&fs_info->dead_roots, &list);
	mutex_unlock(&fs_info->trans_mutex);

	while (!list_empty(&list)) {
		root = list_entry(list.next, struct btrfs_root, root_list);
		list_del(&root->root_list);

		if (btrfs_header_backref_rev(root->node) <
		    BTRFS_MIXED_BACKREF_REV)
			btrfs_drop_snapshot(root, 0);
		else
			btrfs_drop_snapshot(root, 1);
	}
	return 0;
}