summaryrefslogtreecommitdiffstats
path: root/fs/btrfs/delayed-inode.c
blob: dd6f566a383f00e83c9f36125ee4ecd3fc0e3541 (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
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
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
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
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
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
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
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (C) 2011 Fujitsu.  All rights reserved.
 * Written by Miao Xie <miaox@cn.fujitsu.com>
 */

#include <linux/slab.h>
#include <linux/iversion.h>
#include "ctree.h"
#include "fs.h"
#include "messages.h"
#include "misc.h"
#include "delayed-inode.h"
#include "disk-io.h"
#include "transaction.h"
#include "qgroup.h"
#include "locking.h"
#include "inode-item.h"
#include "space-info.h"
#include "accessors.h"
#include "file-item.h"

#define BTRFS_DELAYED_WRITEBACK		512
#define BTRFS_DELAYED_BACKGROUND	128
#define BTRFS_DELAYED_BATCH		16

static struct kmem_cache *delayed_node_cache;

int __init btrfs_delayed_inode_init(void)
{
	delayed_node_cache = KMEM_CACHE(btrfs_delayed_node, 0);
	if (!delayed_node_cache)
		return -ENOMEM;
	return 0;
}

void __cold btrfs_delayed_inode_exit(void)
{
	kmem_cache_destroy(delayed_node_cache);
}

void btrfs_init_delayed_root(struct btrfs_delayed_root *delayed_root)
{
	atomic_set(&delayed_root->items, 0);
	atomic_set(&delayed_root->items_seq, 0);
	delayed_root->nodes = 0;
	spin_lock_init(&delayed_root->lock);
	init_waitqueue_head(&delayed_root->wait);
	INIT_LIST_HEAD(&delayed_root->node_list);
	INIT_LIST_HEAD(&delayed_root->prepare_list);
}

static inline void btrfs_init_delayed_node(
				struct btrfs_delayed_node *delayed_node,
				struct btrfs_root *root, u64 inode_id)
{
	delayed_node->root = root;
	delayed_node->inode_id = inode_id;
	refcount_set(&delayed_node->refs, 0);
	delayed_node->ins_root = RB_ROOT_CACHED;
	delayed_node->del_root = RB_ROOT_CACHED;
	mutex_init(&delayed_node->mutex);
	INIT_LIST_HEAD(&delayed_node->n_list);
	INIT_LIST_HEAD(&delayed_node->p_list);
}

static struct btrfs_delayed_node *btrfs_get_delayed_node(
		struct btrfs_inode *btrfs_inode)
{
	struct btrfs_root *root = btrfs_inode->root;
	u64 ino = btrfs_ino(btrfs_inode);
	struct btrfs_delayed_node *node;

	node = READ_ONCE(btrfs_inode->delayed_node);
	if (node) {
		refcount_inc(&node->refs);
		return node;
	}

	spin_lock(&root->inode_lock);
	node = xa_load(&root->delayed_nodes, ino);

	if (node) {
		if (btrfs_inode->delayed_node) {
			refcount_inc(&node->refs);	/* can be accessed */
			BUG_ON(btrfs_inode->delayed_node != node);
			spin_unlock(&root->inode_lock);
			return node;
		}

		/*
		 * It's possible that we're racing into the middle of removing
		 * this node from the xarray.  In this case, the refcount
		 * was zero and it should never go back to one.  Just return
		 * NULL like it was never in the xarray at all; our release
		 * function is in the process of removing it.
		 *
		 * Some implementations of refcount_inc refuse to bump the
		 * refcount once it has hit zero.  If we don't do this dance
		 * here, refcount_inc() may decide to just WARN_ONCE() instead
		 * of actually bumping the refcount.
		 *
		 * If this node is properly in the xarray, we want to bump the
		 * refcount twice, once for the inode and once for this get
		 * operation.
		 */
		if (refcount_inc_not_zero(&node->refs)) {
			refcount_inc(&node->refs);
			btrfs_inode->delayed_node = node;
		} else {
			node = NULL;
		}

		spin_unlock(&root->inode_lock);
		return node;
	}
	spin_unlock(&root->inode_lock);

	return NULL;
}

/* Will return either the node or PTR_ERR(-ENOMEM) */
static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node(
		struct btrfs_inode *btrfs_inode)
{
	struct btrfs_delayed_node *node;
	struct btrfs_root *root = btrfs_inode->root;
	u64 ino = btrfs_ino(btrfs_inode);
	int ret;
	void *ptr;

again:
	node = btrfs_get_delayed_node(btrfs_inode);
	if (node)
		return node;

	node = kmem_cache_zalloc(delayed_node_cache, GFP_NOFS);
	if (!node)
		return ERR_PTR(-ENOMEM);
	btrfs_init_delayed_node(node, root, ino);

	/* Cached in the inode and can be accessed. */
	refcount_set(&node->refs, 2);

	/* Allocate and reserve the slot, from now it can return a NULL from xa_load(). */
	ret = xa_reserve(&root->delayed_nodes, ino, GFP_NOFS);
	if (ret == -ENOMEM) {
		kmem_cache_free(delayed_node_cache, node);
		return ERR_PTR(-ENOMEM);
	}
	spin_lock(&root->inode_lock);
	ptr = xa_load(&root->delayed_nodes, ino);
	if (ptr) {
		/* Somebody inserted it, go back and read it. */
		spin_unlock(&root->inode_lock);
		kmem_cache_free(delayed_node_cache, node);
		node = NULL;
		goto again;
	}
	ptr = xa_store(&root->delayed_nodes, ino, node, GFP_ATOMIC);
	ASSERT(xa_err(ptr) != -EINVAL);
	ASSERT(xa_err(ptr) != -ENOMEM);
	ASSERT(ptr == NULL);
	btrfs_inode->delayed_node = node;
	spin_unlock(&root->inode_lock);

	return node;
}

/*
 * Call it when holding delayed_node->mutex
 *
 * If mod = 1, add this node into the prepared list.
 */
static void btrfs_queue_delayed_node(struct btrfs_delayed_root *root,
				     struct btrfs_delayed_node *node,
				     int mod)
{
	spin_lock(&root->lock);
	if (test_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags)) {
		if (!list_empty(&node->p_list))
			list_move_tail(&node->p_list, &root->prepare_list);
		else if (mod)
			list_add_tail(&node->p_list, &root->prepare_list);
	} else {
		list_add_tail(&node->n_list, &root->node_list);
		list_add_tail(&node->p_list, &root->prepare_list);
		refcount_inc(&node->refs);	/* inserted into list */
		root->nodes++;
		set_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags);
	}
	spin_unlock(&root->lock);
}

/* Call it when holding delayed_node->mutex */
static void btrfs_dequeue_delayed_node(struct btrfs_delayed_root *root,
				       struct btrfs_delayed_node *node)
{
	spin_lock(&root->lock);
	if (test_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags)) {
		root->nodes--;
		refcount_dec(&node->refs);	/* not in the list */
		list_del_init(&node->n_list);
		if (!list_empty(&node->p_list))
			list_del_init(&node->p_list);
		clear_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags);
	}
	spin_unlock(&root->lock);
}

static struct btrfs_delayed_node *btrfs_first_delayed_node(
			struct btrfs_delayed_root *delayed_root)
{
	struct list_head *p;
	struct btrfs_delayed_node *node = NULL;

	spin_lock(&delayed_root->lock);
	if (list_empty(&delayed_root->node_list))
		goto out;

	p = delayed_root->node_list.next;
	node = list_entry(p, struct btrfs_delayed_node, n_list);
	refcount_inc(&node->refs);
out:
	spin_unlock(&delayed_root->lock);

	return node;
}

static struct btrfs_delayed_node *btrfs_next_delayed_node(
						struct btrfs_delayed_node *node)
{
	struct btrfs_delayed_root *delayed_root;
	struct list_head *p;
	struct btrfs_delayed_node *next = NULL;

	delayed_root = node->root->fs_info->delayed_root;
	spin_lock(&delayed_root->lock);
	if (!test_bit(BTRFS_DELAYED_NODE_IN_LIST, &node->flags)) {
		/* not in the list */
		if (list_empty(&delayed_root->node_list))
			goto out;
		p = delayed_root->node_list.next;
	} else if (list_is_last(&node->n_list, &delayed_root->node_list))
		goto out;
	else
		p = node->n_list.next;

	next = list_entry(p, struct btrfs_delayed_node, n_list);
	refcount_inc(&next->refs);
out:
	spin_unlock(&delayed_root->lock);

	return next;
}

static void __btrfs_release_delayed_node(
				struct btrfs_delayed_node *delayed_node,
				int mod)
{
	struct btrfs_delayed_root *delayed_root;

	if (!delayed_node)
		return;

	delayed_root = delayed_node->root->fs_info->delayed_root;

	mutex_lock(&delayed_node->mutex);
	if (delayed_node->count)
		btrfs_queue_delayed_node(delayed_root, delayed_node, mod);
	else
		btrfs_dequeue_delayed_node(delayed_root, delayed_node);
	mutex_unlock(&delayed_node->mutex);

	if (refcount_dec_and_test(&delayed_node->refs)) {
		struct btrfs_root *root = delayed_node->root;

		spin_lock(&root->inode_lock);
		/*
		 * Once our refcount goes to zero, nobody is allowed to bump it
		 * back up.  We can delete it now.
		 */
		ASSERT(refcount_read(&delayed_node->refs) == 0);
		xa_erase(&root->delayed_nodes, delayed_node->inode_id);
		spin_unlock(&root->inode_lock);
		kmem_cache_free(delayed_node_cache, delayed_node);
	}
}

static inline void btrfs_release_delayed_node(struct btrfs_delayed_node *node)
{
	__btrfs_release_delayed_node(node, 0);
}

static struct btrfs_delayed_node *btrfs_first_prepared_delayed_node(
					struct btrfs_delayed_root *delayed_root)
{
	struct list_head *p;
	struct btrfs_delayed_node *node = NULL;

	spin_lock(&delayed_root->lock);
	if (list_empty(&delayed_root->prepare_list))
		goto out;

	p = delayed_root->prepare_list.next;
	list_del_init(p);
	node = list_entry(p, struct btrfs_delayed_node, p_list);
	refcount_inc(&node->refs);
out:
	spin_unlock(&delayed_root->lock);

	return node;
}

static inline void btrfs_release_prepared_delayed_node(
					struct btrfs_delayed_node *node)
{
	__btrfs_release_delayed_node(node, 1);
}

static struct btrfs_delayed_item *btrfs_alloc_delayed_item(u16 data_len,
					   struct btrfs_delayed_node *node,
					   enum btrfs_delayed_item_type type)
{
	struct btrfs_delayed_item *item;

	item = kmalloc(struct_size(item, data, data_len), GFP_NOFS);
	if (item) {
		item->data_len = data_len;
		item->type = type;
		item->bytes_reserved = 0;
		item->delayed_node = node;
		RB_CLEAR_NODE(&item->rb_node);
		INIT_LIST_HEAD(&item->log_list);
		item->logged = false;
		refcount_set(&item->refs, 1);
	}
	return item;
}

/*
 * Look up the delayed item by key.
 *
 * @delayed_node: pointer to the delayed node
 * @index:	  the dir index value to lookup (offset of a dir index key)
 *
 * Note: if we don't find the right item, we will return the prev item and
 * the next item.
 */
static struct btrfs_delayed_item *__btrfs_lookup_delayed_item(
				struct rb_root *root,
				u64 index)
{
	struct rb_node *node = root->rb_node;
	struct btrfs_delayed_item *delayed_item = NULL;

	while (node) {
		delayed_item = rb_entry(node, struct btrfs_delayed_item,
					rb_node);
		if (delayed_item->index < index)
			node = node->rb_right;
		else if (delayed_item->index > index)
			node = node->rb_left;
		else
			return delayed_item;
	}

	return NULL;
}

static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
				    struct btrfs_delayed_item *ins)
{
	struct rb_node **p, *node;
	struct rb_node *parent_node = NULL;
	struct rb_root_cached *root;
	struct btrfs_delayed_item *item;
	bool leftmost = true;

	if (ins->type == BTRFS_DELAYED_INSERTION_ITEM)
		root = &delayed_node->ins_root;
	else
		root = &delayed_node->del_root;

	p = &root->rb_root.rb_node;
	node = &ins->rb_node;

	while (*p) {
		parent_node = *p;
		item = rb_entry(parent_node, struct btrfs_delayed_item,
				 rb_node);

		if (item->index < ins->index) {
			p = &(*p)->rb_right;
			leftmost = false;
		} else if (item->index > ins->index) {
			p = &(*p)->rb_left;
		} else {
			return -EEXIST;
		}
	}

	rb_link_node(node, parent_node, p);
	rb_insert_color_cached(node, root, leftmost);

	if (ins->type == BTRFS_DELAYED_INSERTION_ITEM &&
	    ins->index >= delayed_node->index_cnt)
		delayed_node->index_cnt = ins->index + 1;

	delayed_node->count++;
	atomic_inc(&delayed_node->root->fs_info->delayed_root->items);
	return 0;
}

static void finish_one_item(struct btrfs_delayed_root *delayed_root)
{
	int seq = atomic_inc_return(&delayed_root->items_seq);

	/* atomic_dec_return implies a barrier */
	if ((atomic_dec_return(&delayed_root->items) <
	    BTRFS_DELAYED_BACKGROUND || seq % BTRFS_DELAYED_BATCH == 0))
		cond_wake_up_nomb(&delayed_root->wait);
}

static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)
{
	struct btrfs_delayed_node *delayed_node = delayed_item->delayed_node;
	struct rb_root_cached *root;
	struct btrfs_delayed_root *delayed_root;

	/* Not inserted, ignore it. */
	if (RB_EMPTY_NODE(&delayed_item->rb_node))
		return;

	/* If it's in a rbtree, then we need to have delayed node locked. */
	lockdep_assert_held(&delayed_node->mutex);

	delayed_root = delayed_node->root->fs_info->delayed_root;

	if (delayed_item->type == BTRFS_DELAYED_INSERTION_ITEM)
		root = &delayed_node->ins_root;
	else
		root = &delayed_node->del_root;

	rb_erase_cached(&delayed_item->rb_node, root);
	RB_CLEAR_NODE(&delayed_item->rb_node);
	delayed_node->count--;

	finish_one_item(delayed_root);
}

static void btrfs_release_delayed_item(struct btrfs_delayed_item *item)
{
	if (item) {
		__btrfs_remove_delayed_item(item);
		if (refcount_dec_and_test(&item->refs))
			kfree(item);
	}
}

static struct btrfs_delayed_item *__btrfs_first_delayed_insertion_item(
					struct btrfs_delayed_node *delayed_node)
{
	struct rb_node *p;
	struct btrfs_delayed_item *item = NULL;

	p = rb_first_cached(&delayed_node->ins_root);
	if (p)
		item = rb_entry(p, struct btrfs_delayed_item, rb_node);

	return item;
}

static struct btrfs_delayed_item *__btrfs_first_delayed_deletion_item(
					struct btrfs_delayed_node *delayed_node)
{
	struct rb_node *p;
	struct btrfs_delayed_item *item = NULL;

	p = rb_first_cached(&delayed_node->del_root);
	if (p)
		item = rb_entry(p, struct btrfs_delayed_item, rb_node);

	return item;
}

static struct btrfs_delayed_item *__btrfs_next_delayed_item(
						struct btrfs_delayed_item *item)
{
	struct rb_node *p;
	struct btrfs_delayed_item *next = NULL;

	p = rb_next(&item->rb_node);
	if (p)
		next = rb_entry(p, struct btrfs_delayed_item, rb_node);

	return next;
}

static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
					       struct btrfs_delayed_item *item)
{
	struct btrfs_block_rsv *src_rsv;
	struct btrfs_block_rsv *dst_rsv;
	struct btrfs_fs_info *fs_info = trans->fs_info;
	u64 num_bytes;
	int ret;

	if (!trans->bytes_reserved)
		return 0;

	src_rsv = trans->block_rsv;
	dst_rsv = &fs_info->delayed_block_rsv;

	num_bytes = btrfs_calc_insert_metadata_size(fs_info, 1);

	/*
	 * Here we migrate space rsv from transaction rsv, since have already
	 * reserved space when starting a transaction.  So no need to reserve
	 * qgroup space here.
	 */
	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, true);
	if (!ret) {
		trace_btrfs_space_reservation(fs_info, "delayed_item",
					      item->delayed_node->inode_id,
					      num_bytes, 1);
		/*
		 * For insertions we track reserved metadata space by accounting
		 * for the number of leaves that will be used, based on the delayed
		 * node's curr_index_batch_size and index_item_leaves fields.
		 */
		if (item->type == BTRFS_DELAYED_DELETION_ITEM)
			item->bytes_reserved = num_bytes;
	}

	return ret;
}

static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
						struct btrfs_delayed_item *item)
{
	struct btrfs_block_rsv *rsv;
	struct btrfs_fs_info *fs_info = root->fs_info;

	if (!item->bytes_reserved)
		return;

	rsv = &fs_info->delayed_block_rsv;
	/*
	 * Check btrfs_delayed_item_reserve_metadata() to see why we don't need
	 * to release/reserve qgroup space.
	 */
	trace_btrfs_space_reservation(fs_info, "delayed_item",
				      item->delayed_node->inode_id,
				      item->bytes_reserved, 0);
	btrfs_block_rsv_release(fs_info, rsv, item->bytes_reserved, NULL);
}

static void btrfs_delayed_item_release_leaves(struct btrfs_delayed_node *node,
					      unsigned int num_leaves)
{
	struct btrfs_fs_info *fs_info = node->root->fs_info;
	const u64 bytes = btrfs_calc_insert_metadata_size(fs_info, num_leaves);

	/* There are no space reservations during log replay, bail out. */
	if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags))
		return;

	trace_btrfs_space_reservation(fs_info, "delayed_item", node->inode_id,
				      bytes, 0);
	btrfs_block_rsv_release(fs_info, &fs_info->delayed_block_rsv, bytes, NULL);
}

static int btrfs_delayed_inode_reserve_metadata(
					struct btrfs_trans_handle *trans,
					struct btrfs_root *root,
					struct btrfs_delayed_node *node)
{
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_block_rsv *src_rsv;
	struct btrfs_block_rsv *dst_rsv;
	u64 num_bytes;
	int ret;

	src_rsv = trans->block_rsv;
	dst_rsv = &fs_info->delayed_block_rsv;

	num_bytes = btrfs_calc_metadata_size(fs_info, 1);

	/*
	 * btrfs_dirty_inode will update the inode under btrfs_join_transaction
	 * which doesn't reserve space for speed.  This is a problem since we
	 * still need to reserve space for this update, so try to reserve the
	 * space.
	 *
	 * Now if src_rsv == delalloc_block_rsv we'll let it just steal since
	 * we always reserve enough to update the inode item.
	 */
	if (!src_rsv || (!trans->bytes_reserved &&
			 src_rsv->type != BTRFS_BLOCK_RSV_DELALLOC)) {
		ret = btrfs_qgroup_reserve_meta(root, num_bytes,
					  BTRFS_QGROUP_RSV_META_PREALLOC, true);
		if (ret < 0)
			return ret;
		ret = btrfs_block_rsv_add(fs_info, dst_rsv, num_bytes,
					  BTRFS_RESERVE_NO_FLUSH);
		/* NO_FLUSH could only fail with -ENOSPC */
		ASSERT(ret == 0 || ret == -ENOSPC);
		if (ret)
			btrfs_qgroup_free_meta_prealloc(root, num_bytes);
	} else {
		ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, true);
	}

	if (!ret) {
		trace_btrfs_space_reservation(fs_info, "delayed_inode",
					      node->inode_id, num_bytes, 1);
		node->bytes_reserved = num_bytes;
	}

	return ret;
}

static void btrfs_delayed_inode_release_metadata(struct btrfs_fs_info *fs_info,
						struct btrfs_delayed_node *node,
						bool qgroup_free)
{
	struct btrfs_block_rsv *rsv;

	if (!node->bytes_reserved)
		return;

	rsv = &fs_info->delayed_block_rsv;
	trace_btrfs_space_reservation(fs_info, "delayed_inode",
				      node->inode_id, node->bytes_reserved, 0);
	btrfs_block_rsv_release(fs_info, rsv, node->bytes_reserved, NULL);
	if (qgroup_free)
		btrfs_qgroup_free_meta_prealloc(node->root,
				node->bytes_reserved);
	else
		btrfs_qgroup_convert_reserved_meta(node->root,
				node->bytes_reserved);
	node->bytes_reserved = 0;
}

/*
 * Insert a single delayed item or a batch of delayed items, as many as possible
 * that fit in a leaf. The delayed items (dir index keys) are sorted by their key
 * in the rbtree, and if there's a gap between two consecutive dir index items,
 * then it means at some point we had delayed dir indexes to add but they got
 * removed (by btrfs_delete_delayed_dir_index()) before we attempted to flush them
 * into the subvolume tree. Dir index keys also have their offsets coming from a
 * monotonically increasing counter, so we can't get new keys with an offset that
 * fits within a gap between delayed dir index items.
 */
static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans,
				     struct btrfs_root *root,
				     struct btrfs_path *path,
				     struct btrfs_delayed_item *first_item)
{
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_delayed_node *node = first_item->delayed_node;
	LIST_HEAD(item_list);
	struct btrfs_delayed_item *curr;
	struct btrfs_delayed_item *next;
	const int max_size = BTRFS_LEAF_DATA_SIZE(fs_info);
	struct btrfs_item_batch batch;
	struct btrfs_key first_key;
	const u32 first_data_size = first_item->data_len;
	int total_size;
	char *ins_data = NULL;
	int ret;
	bool continuous_keys_only = false;

	lockdep_assert_held(&node->mutex);

	/*
	 * During normal operation the delayed index offset is continuously
	 * increasing, so we can batch insert all items as there will not be any
	 * overlapping keys in the tree.
	 *
	 * The exception to this is log replay, where we may have interleaved
	 * offsets in the tree, so our batch needs to be continuous keys only in
	 * order to ensure we do not end up with out of order items in our leaf.
	 */
	if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags))
		continuous_keys_only = true;

	/*
	 * For delayed items to insert, we track reserved metadata bytes based
	 * on the number of leaves that we will use.
	 * See btrfs_insert_delayed_dir_index() and
	 * btrfs_delayed_item_reserve_metadata()).
	 */
	ASSERT(first_item->bytes_reserved == 0);

	list_add_tail(&first_item->tree_list, &item_list);
	batch.total_data_size = first_data_size;
	batch.nr = 1;
	total_size = first_data_size + sizeof(struct btrfs_item);
	curr = first_item;

	while (true) {
		int next_size;

		next = __btrfs_next_delayed_item(curr);
		if (!next)
			break;

		/*
		 * We cannot allow gaps in the key space if we're doing log
		 * replay.
		 */
		if (continuous_keys_only && (next->index != curr->index + 1))
			break;

		ASSERT(next->bytes_reserved == 0);

		next_size = next->data_len + sizeof(struct btrfs_item);
		if (total_size + next_size > max_size)
			break;

		list_add_tail(&next->tree_list, &item_list);
		batch.nr++;
		total_size += next_size;
		batch.total_data_size += next->data_len;
		curr = next;
	}

	if (batch.nr == 1) {
		first_key.objectid = node->inode_id;
		first_key.type = BTRFS_DIR_INDEX_KEY;
		first_key.offset = first_item->index;
		batch.keys = &first_key;
		batch.data_sizes = &first_data_size;
	} else {
		struct btrfs_key *ins_keys;
		u32 *ins_sizes;
		int i = 0;

		ins_data = kmalloc(batch.nr * sizeof(u32) +
				   batch.nr * sizeof(struct btrfs_key), GFP_NOFS);
		if (!ins_data) {
			ret = -ENOMEM;
			goto out;
		}
		ins_sizes = (u32 *)ins_data;
		ins_keys = (struct btrfs_key *)(ins_data + batch.nr * sizeof(u32));
		batch.keys = ins_keys;
		batch.data_sizes = ins_sizes;
		list_for_each_entry(curr, &item_list, tree_list) {
			ins_keys[i].objectid = node->inode_id;
			ins_keys[i].type = BTRFS_DIR_INDEX_KEY;
			ins_keys[i].offset = curr->index;
			ins_sizes[i] = curr->data_len;
			i++;
		}
	}

	ret = btrfs_insert_empty_items(trans, root, path, &batch);
	if (ret)
		goto out;

	list_for_each_entry(curr, &item_list, tree_list) {
		char *data_ptr;

		data_ptr = btrfs_item_ptr(path->nodes[0], path->slots[0], char);
		write_extent_buffer(path->nodes[0], &curr->data,
				    (unsigned long)data_ptr, curr->data_len);
		path->slots[0]++;
	}

	/*
	 * Now release our path before releasing the delayed items and their
	 * metadata reservations, so that we don't block other tasks for more
	 * time than needed.
	 */
	btrfs_release_path(path);

	ASSERT(node->index_item_leaves > 0);

	/*
	 * For normal operations we will batch an entire leaf's worth of delayed
	 * items, so if there are more items to process we can decrement
	 * index_item_leaves by 1 as we inserted 1 leaf's worth of items.
	 *
	 * However for log replay we may not have inserted an entire leaf's
	 * worth of items, we may have not had continuous items, so decrementing
	 * here would mess up the index_item_leaves accounting.  For this case
	 * only clean up the accounting when there are no items left.
	 */
	if (next && !continuous_keys_only) {
		/*
		 * We inserted one batch of items into a leaf a there are more
		 * items to flush in a future batch, now release one unit of
		 * metadata space from the delayed block reserve, corresponding
		 * the leaf we just flushed to.
		 */
		btrfs_delayed_item_release_leaves(node, 1);
		node->index_item_leaves--;
	} else if (!next) {
		/*
		 * There are no more items to insert. We can have a number of
		 * reserved leaves > 1 here - this happens when many dir index
		 * items are added and then removed before they are flushed (file
		 * names with a very short life, never span a transaction). So
		 * release all remaining leaves.
		 */
		btrfs_delayed_item_release_leaves(node, node->index_item_leaves);
		node->index_item_leaves = 0;
	}

	list_for_each_entry_safe(curr, next, &item_list, tree_list) {
		list_del(&curr->tree_list);
		btrfs_release_delayed_item(curr);
	}
out:
	kfree(ins_data);
	return ret;
}

static int btrfs_insert_delayed_items(struct btrfs_trans_handle *trans,
				      struct btrfs_path *path,
				      struct btrfs_root *root,
				      struct btrfs_delayed_node *node)
{
	int ret = 0;

	while (ret == 0) {
		struct btrfs_delayed_item *curr;

		mutex_lock(&node->mutex);
		curr = __btrfs_first_delayed_insertion_item(node);
		if (!curr) {
			mutex_unlock(&node->mutex);
			break;
		}
		ret = btrfs_insert_delayed_item(trans, root, path, curr);
		mutex_unlock(&node->mutex);
	}

	return ret;
}

static int btrfs_batch_delete_items(struct btrfs_trans_handle *trans,
				    struct btrfs_root *root,
				    struct btrfs_path *path,
				    struct btrfs_delayed_item *item)
{
	const u64 ino = item->delayed_node->inode_id;
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_delayed_item *curr, *next;
	struct extent_buffer *leaf = path->nodes[0];
	LIST_HEAD(batch_list);
	int nitems, slot, last_slot;
	int ret;
	u64 total_reserved_size = item->bytes_reserved;

	ASSERT(leaf != NULL);

	slot = path->slots[0];
	last_slot = btrfs_header_nritems(leaf) - 1;
	/*
	 * Our caller always gives us a path pointing to an existing item, so
	 * this can not happen.
	 */
	ASSERT(slot <= last_slot);
	if (WARN_ON(slot > last_slot))
		return -ENOENT;

	nitems = 1;
	curr = item;
	list_add_tail(&curr->tree_list, &batch_list);

	/*
	 * Keep checking if the next delayed item matches the next item in the
	 * leaf - if so, we can add it to the batch of items to delete from the
	 * leaf.
	 */
	while (slot < last_slot) {
		struct btrfs_key key;

		next = __btrfs_next_delayed_item(curr);
		if (!next)
			break;

		slot++;
		btrfs_item_key_to_cpu(leaf, &key, slot);
		if (key.objectid != ino ||
		    key.type != BTRFS_DIR_INDEX_KEY ||
		    key.offset != next->index)
			break;
		nitems++;
		curr = next;
		list_add_tail(&curr->tree_list, &batch_list);
		total_reserved_size += curr->bytes_reserved;
	}

	ret = btrfs_del_items(trans, root, path, path->slots[0], nitems);
	if (ret)
		return ret;

	/* In case of BTRFS_FS_LOG_RECOVERING items won't have reserved space */
	if (total_reserved_size > 0) {
		/*
		 * Check btrfs_delayed_item_reserve_metadata() to see why we
		 * don't need to release/reserve qgroup space.
		 */
		trace_btrfs_space_reservation(fs_info, "delayed_item", ino,
					      total_reserved_size, 0);
		btrfs_block_rsv_release(fs_info, &fs_info->delayed_block_rsv,
					total_reserved_size, NULL);
	}

	list_for_each_entry_safe(curr, next, &batch_list, tree_list) {
		list_del(&curr->tree_list);
		btrfs_release_delayed_item(curr);
	}

	return 0;
}

static int btrfs_delete_delayed_items(struct btrfs_trans_handle *trans,
				      struct btrfs_path *path,
				      struct btrfs_root *root,
				      struct btrfs_delayed_node *node)
{
	struct btrfs_key key;
	int ret = 0;

	key.objectid = node->inode_id;
	key.type = BTRFS_DIR_INDEX_KEY;

	while (ret == 0) {
		struct btrfs_delayed_item *item;

		mutex_lock(&node->mutex);
		item = __btrfs_first_delayed_deletion_item(node);
		if (!item) {
			mutex_unlock(&node->mutex);
			break;
		}

		key.offset = item->index;
		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
		if (ret > 0) {
			/*
			 * There's no matching item in the leaf. This means we
			 * have already deleted this item in a past run of the
			 * delayed items. We ignore errors when running delayed
			 * items from an async context, through a work queue job
			 * running btrfs_async_run_delayed_root(), and don't
			 * release delayed items that failed to complete. This
			 * is because we will retry later, and at transaction
			 * commit time we always run delayed items and will
			 * then deal with errors if they fail to run again.
			 *
			 * So just release delayed items for which we can't find
			 * an item in the tree, and move to the next item.
			 */
			btrfs_release_path(path);
			btrfs_release_delayed_item(item);
			ret = 0;
		} else if (ret == 0) {
			ret = btrfs_batch_delete_items(trans, root, path, item);
			btrfs_release_path(path);
		}

		/*
		 * We unlock and relock on each iteration, this is to prevent
		 * blocking other tasks for too long while we are being run from
		 * the async context (work queue job). Those tasks are typically
		 * running system calls like creat/mkdir/rename/unlink/etc which
		 * need to add delayed items to this delayed node.
		 */
		mutex_unlock(&node->mutex);
	}

	return ret;
}

static void btrfs_release_delayed_inode(struct btrfs_delayed_node *delayed_node)
{
	struct btrfs_delayed_root *delayed_root;

	if (delayed_node &&
	    test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags)) {
		ASSERT(delayed_node->root);
		clear_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags);
		delayed_node->count--;

		delayed_root = delayed_node->root->fs_info->delayed_root;
		finish_one_item(delayed_root);
	}
}

static void btrfs_release_delayed_iref(struct btrfs_delayed_node *delayed_node)
{

	if (test_and_clear_bit(BTRFS_DELAYED_NODE_DEL_IREF, &delayed_node->flags)) {
		struct btrfs_delayed_root *delayed_root;

		ASSERT(delayed_node->root);
		delayed_node->count--;

		delayed_root = delayed_node->root->fs_info->delayed_root;
		finish_one_item(delayed_root);
	}
}

static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
					struct btrfs_root *root,
					struct btrfs_path *path,
					struct btrfs_delayed_node *node)
{
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_key key;
	struct btrfs_inode_item *inode_item;
	struct extent_buffer *leaf;
	int mod;
	int ret;

	key.objectid = node->inode_id;
	key.type = BTRFS_INODE_ITEM_KEY;
	key.offset = 0;

	if (test_bit(BTRFS_DELAYED_NODE_DEL_IREF, &node->flags))
		mod = -1;
	else
		mod = 1;

	ret = btrfs_lookup_inode(trans, root, path, &key, mod);
	if (ret > 0)
		ret = -ENOENT;
	if (ret < 0)
		goto out;

	leaf = path->nodes[0];
	inode_item = btrfs_item_ptr(leaf, path->slots[0],
				    struct btrfs_inode_item);
	write_extent_buffer(leaf, &node->inode_item, (unsigned long)inode_item,
			    sizeof(struct btrfs_inode_item));
	btrfs_mark_buffer_dirty(trans, leaf);

	if (!test_bit(BTRFS_DELAYED_NODE_DEL_IREF, &node->flags))
		goto out;

	/*
	 * Now we're going to delete the INODE_REF/EXTREF, which should be the
	 * only one ref left.  Check if the next item is an INODE_REF/EXTREF.
	 *
	 * But if we're the last item already, release and search for the last
	 * INODE_REF/EXTREF.
	 */
	if (path->slots[0] + 1 >= btrfs_header_nritems(leaf)) {
		key.objectid = node->inode_id;
		key.type = BTRFS_INODE_EXTREF_KEY;
		key.offset = (u64)-1;

		btrfs_release_path(path);
		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
		if (ret < 0)
			goto err_out;
		ASSERT(ret > 0);
		ASSERT(path->slots[0] > 0);
		ret = 0;
		path->slots[0]--;
		leaf = path->nodes[0];
	} else {
		path->slots[0]++;
	}
	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
	if (key.objectid != node->inode_id)
		goto out;
	if (key.type != BTRFS_INODE_REF_KEY &&
	    key.type != BTRFS_INODE_EXTREF_KEY)
		goto out;

	/*
	 * Delayed iref deletion is for the inode who has only one link,
	 * so there is only one iref. The case that several irefs are
	 * in the same item doesn't exist.
	 */
	ret = btrfs_del_item(trans, root, path);
out:
	btrfs_release_delayed_iref(node);
	btrfs_release_path(path);
err_out:
	btrfs_delayed_inode_release_metadata(fs_info, node, (ret < 0));
	btrfs_release_delayed_inode(node);

	/*
	 * If we fail to update the delayed inode we need to abort the
	 * transaction, because we could leave the inode with the improper
	 * counts behind.
	 */
	if (ret && ret != -ENOENT)
		btrfs_abort_transaction(trans, ret);

	return ret;
}

static inline int btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
					     struct btrfs_root *root,
					     struct btrfs_path *path,
					     struct btrfs_delayed_node *node)
{
	int ret;

	mutex_lock(&node->mutex);
	if (!test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &node->flags)) {
		mutex_unlock(&node->mutex);
		return 0;
	}

	ret = __btrfs_update_delayed_inode(trans, root, path, node);
	mutex_unlock(&node->mutex);
	return ret;
}

static inline int
__btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
				   struct btrfs_path *path,
				   struct btrfs_delayed_node *node)
{
	int ret;

	ret = btrfs_insert_delayed_items(trans, path, node->root, node);
	if (ret)
		return ret;

	ret = btrfs_delete_delayed_items(trans, path, node->root, node);
	if (ret)
		return ret;

	ret = btrfs_update_delayed_inode(trans, node->root, path, node);
	return ret;
}

/*
 * Called when committing the transaction.
 * Returns 0 on success.
 * Returns < 0 on error and returns with an aborted transaction with any
 * outstanding delayed items cleaned up.
 */
static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans, int nr)
{
	struct btrfs_fs_info *fs_info = trans->fs_info;
	struct btrfs_delayed_root *delayed_root;
	struct btrfs_delayed_node *curr_node, *prev_node;
	struct btrfs_path *path;
	struct btrfs_block_rsv *block_rsv;
	int ret = 0;
	bool count = (nr > 0);

	if (TRANS_ABORTED(trans))
		return -EIO;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

	block_rsv = trans->block_rsv;
	trans->block_rsv = &fs_info->delayed_block_rsv;

	delayed_root = fs_info->delayed_root;

	curr_node = btrfs_first_delayed_node(delayed_root);
	while (curr_node && (!count || nr--)) {
		ret = __btrfs_commit_inode_delayed_items(trans, path,
							 curr_node);
		if (ret) {
			btrfs_abort_transaction(trans, ret);
			break;
		}

		prev_node = curr_node;
		curr_node = btrfs_next_delayed_node(curr_node);
		/*
		 * See the comment below about releasing path before releasing
		 * node. If the commit of delayed items was successful the path
		 * should always be released, but in case of an error, it may
		 * point to locked extent buffers (a leaf at the very least).
		 */
		ASSERT(path->nodes[0] == NULL);
		btrfs_release_delayed_node(prev_node);
	}

	/*
	 * Release the path to avoid a potential deadlock and lockdep splat when
	 * releasing the delayed node, as that requires taking the delayed node's
	 * mutex. If another task starts running delayed items before we take
	 * the mutex, it will first lock the mutex and then it may try to lock
	 * the same btree path (leaf).
	 */
	btrfs_free_path(path);

	if (curr_node)
		btrfs_release_delayed_node(curr_node);
	trans->block_rsv = block_rsv;

	return ret;
}

int btrfs_run_delayed_items(struct btrfs_trans_handle *trans)
{
	return __btrfs_run_delayed_items(trans, -1);
}

int btrfs_run_delayed_items_nr(struct btrfs_trans_handle *trans, int nr)
{
	return __btrfs_run_delayed_items(trans, nr);
}

int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
				     struct btrfs_inode *inode)
{
	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
	struct btrfs_path *path;
	struct btrfs_block_rsv *block_rsv;
	int ret;

	if (!delayed_node)
		return 0;

	mutex_lock(&delayed_node->mutex);
	if (!delayed_node->count) {
		mutex_unlock(&delayed_node->mutex);
		btrfs_release_delayed_node(delayed_node);
		return 0;
	}
	mutex_unlock(&delayed_node->mutex);

	path = btrfs_alloc_path();
	if (!path) {
		btrfs_release_delayed_node(delayed_node);
		return -ENOMEM;
	}

	block_rsv = trans->block_rsv;
	trans->block_rsv = &delayed_node->root->fs_info->delayed_block_rsv;

	ret = __btrfs_commit_inode_delayed_items(trans, path, delayed_node);

	btrfs_release_delayed_node(delayed_node);
	btrfs_free_path(path);
	trans->block_rsv = block_rsv;

	return ret;
}

int btrfs_commit_inode_delayed_inode(struct btrfs_inode *inode)
{
	struct btrfs_fs_info *fs_info = inode->root->fs_info;
	struct btrfs_trans_handle *trans;
	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
	struct btrfs_path *path;
	struct btrfs_block_rsv *block_rsv;
	int ret;

	if (!delayed_node)
		return 0;

	mutex_lock(&delayed_node->mutex);
	if (!test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags)) {
		mutex_unlock(&delayed_node->mutex);
		btrfs_release_delayed_node(delayed_node);
		return 0;
	}
	mutex_unlock(&delayed_node->mutex);

	trans = btrfs_join_transaction(delayed_node->root);
	if (IS_ERR(trans)) {
		ret = PTR_ERR(trans);
		goto out;
	}

	path = btrfs_alloc_path();
	if (!path) {
		ret = -ENOMEM;
		goto trans_out;
	}

	block_rsv = trans->block_rsv;
	trans->block_rsv = &fs_info->delayed_block_rsv;

	mutex_lock(&delayed_node->mutex);
	if (test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags))
		ret = __btrfs_update_delayed_inode(trans, delayed_node->root,
						   path, delayed_node);
	else
		ret = 0;
	mutex_unlock(&delayed_node->mutex);

	btrfs_free_path(path);
	trans->block_rsv = block_rsv;
trans_out:
	btrfs_end_transaction(trans);
	btrfs_btree_balance_dirty(fs_info);
out:
	btrfs_release_delayed_node(delayed_node);

	return ret;
}

void btrfs_remove_delayed_node(struct btrfs_inode *inode)
{
	struct btrfs_delayed_node *delayed_node;

	delayed_node = READ_ONCE(inode->delayed_node);
	if (!delayed_node)
		return;

	inode->delayed_node = NULL;
	btrfs_release_delayed_node(delayed_node);
}

struct btrfs_async_delayed_work {
	struct btrfs_delayed_root *delayed_root;
	int nr;
	struct btrfs_work work;
};

static void btrfs_async_run_delayed_root(struct btrfs_work *work)
{
	struct btrfs_async_delayed_work *async_work;
	struct btrfs_delayed_root *delayed_root;
	struct btrfs_trans_handle *trans;
	struct btrfs_path *path;
	struct btrfs_delayed_node *delayed_node = NULL;
	struct btrfs_root *root;
	struct btrfs_block_rsv *block_rsv;
	int total_done = 0;

	async_work = container_of(work, struct btrfs_async_delayed_work, work);
	delayed_root = async_work->delayed_root;

	path = btrfs_alloc_path();
	if (!path)
		goto out;

	do {
		if (atomic_read(&delayed_root->items) <
		    BTRFS_DELAYED_BACKGROUND / 2)
			break;

		delayed_node = btrfs_first_prepared_delayed_node(delayed_root);
		if (!delayed_node)
			break;

		root = delayed_node->root;

		trans = btrfs_join_transaction(root);
		if (IS_ERR(trans)) {
			btrfs_release_path(path);
			btrfs_release_prepared_delayed_node(delayed_node);
			total_done++;
			continue;
		}

		block_rsv = trans->block_rsv;
		trans->block_rsv = &root->fs_info->delayed_block_rsv;

		__btrfs_commit_inode_delayed_items(trans, path, delayed_node);

		trans->block_rsv = block_rsv;
		btrfs_end_transaction(trans);
		btrfs_btree_balance_dirty_nodelay(root->fs_info);

		btrfs_release_path(path);
		btrfs_release_prepared_delayed_node(delayed_node);
		total_done++;

	} while ((async_work->nr == 0 && total_done < BTRFS_DELAYED_WRITEBACK)
		 || total_done < async_work->nr);

	btrfs_free_path(path);
out:
	wake_up(&delayed_root->wait);
	kfree(async_work);
}


static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
				     struct btrfs_fs_info *fs_info, int nr)
{
	struct btrfs_async_delayed_work *async_work;

	async_work = kmalloc(sizeof(*async_work), GFP_NOFS);
	if (!async_work)
		return -ENOMEM;

	async_work->delayed_root = delayed_root;
	btrfs_init_work(&async_work->work, btrfs_async_run_delayed_root, NULL);
	async_work->nr = nr;

	btrfs_queue_work(fs_info->delayed_workers, &async_work->work);
	return 0;
}

void btrfs_assert_delayed_root_empty(struct btrfs_fs_info *fs_info)
{
	WARN_ON(btrfs_first_delayed_node(fs_info->delayed_root));
}

static int could_end_wait(struct btrfs_delayed_root *delayed_root, int seq)
{
	int val = atomic_read(&delayed_root->items_seq);

	if (val < seq || val >= seq + BTRFS_DELAYED_BATCH)
		return 1;

	if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND)
		return 1;

	return 0;
}

void btrfs_balance_delayed_items(struct btrfs_fs_info *fs_info)
{
	struct btrfs_delayed_root *delayed_root = fs_info->delayed_root;

	if ((atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND) ||
		btrfs_workqueue_normal_congested(fs_info->delayed_workers))
		return;

	if (atomic_read(&delayed_root->items) >= BTRFS_DELAYED_WRITEBACK) {
		int seq;
		int ret;

		seq = atomic_read(&delayed_root->items_seq);

		ret = btrfs_wq_run_delayed_node(delayed_root, fs_info, 0);
		if (ret)
			return;

		wait_event_interruptible(delayed_root->wait,
					 could_end_wait(delayed_root, seq));
		return;
	}

	btrfs_wq_run_delayed_node(delayed_root, fs_info, BTRFS_DELAYED_BATCH);
}

static void btrfs_release_dir_index_item_space(struct btrfs_trans_handle *trans)
{
	struct btrfs_fs_info *fs_info = trans->fs_info;
	const u64 bytes = btrfs_calc_insert_metadata_size(fs_info, 1);

	if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags))
		return;

	/*
	 * Adding the new dir index item does not require touching another
	 * leaf, so we can release 1 unit of metadata that was previously
	 * reserved when starting the transaction. This applies only to
	 * the case where we had a transaction start and excludes the
	 * transaction join case (when replaying log trees).
	 */
	trace_btrfs_space_reservation(fs_info, "transaction",
				      trans->transid, bytes, 0);
	btrfs_block_rsv_release(fs_info, trans->block_rsv, bytes, NULL);
	ASSERT(trans->bytes_reserved >= bytes);
	trans->bytes_reserved -= bytes;
}

/* Will return 0, -ENOMEM or -EEXIST (index number collision, unexpected). */
int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
				   const char *name, int name_len,
				   struct btrfs_inode *dir,
				   struct btrfs_disk_key *disk_key, u8 flags,
				   u64 index)
{
	struct btrfs_fs_info *fs_info = trans->fs_info;
	const unsigned int leaf_data_size = BTRFS_LEAF_DATA_SIZE(fs_info);
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_delayed_item *delayed_item;
	struct btrfs_dir_item *dir_item;
	bool reserve_leaf_space;
	u32 data_len;
	int ret;

	delayed_node = btrfs_get_or_create_delayed_node(dir);
	if (IS_ERR(delayed_node))
		return PTR_ERR(delayed_node);

	delayed_item = btrfs_alloc_delayed_item(sizeof(*dir_item) + name_len,
						delayed_node,
						BTRFS_DELAYED_INSERTION_ITEM);
	if (!delayed_item) {
		ret = -ENOMEM;
		goto release_node;
	}

	delayed_item->index = index;

	dir_item = (struct btrfs_dir_item *)delayed_item->data;
	dir_item->location = *disk_key;
	btrfs_set_stack_dir_transid(dir_item, trans->transid);
	btrfs_set_stack_dir_data_len(dir_item, 0);
	btrfs_set_stack_dir_name_len(dir_item, name_len);
	btrfs_set_stack_dir_flags(dir_item, flags);
	memcpy((char *)(dir_item + 1), name, name_len);

	data_len = delayed_item->data_len + sizeof(struct btrfs_item);

	mutex_lock(&delayed_node->mutex);

	/*
	 * First attempt to insert the delayed item. This is to make the error
	 * handling path simpler in case we fail (-EEXIST). There's no risk of
	 * any other task coming in and running the delayed item before we do
	 * the metadata space reservation below, because we are holding the
	 * delayed node's mutex and that mutex must also be locked before the
	 * node's delayed items can be run.
	 */
	ret = __btrfs_add_delayed_item(delayed_node, delayed_item);
	if (unlikely(ret)) {
		btrfs_err(trans->fs_info,
"error adding delayed dir index item, name: %.*s, index: %llu, root: %llu, dir: %llu, dir->index_cnt: %llu, delayed_node->index_cnt: %llu, error: %d",
			  name_len, name, index, btrfs_root_id(delayed_node->root),
			  delayed_node->inode_id, dir->index_cnt,
			  delayed_node->index_cnt, ret);
		btrfs_release_delayed_item(delayed_item);
		btrfs_release_dir_index_item_space(trans);
		mutex_unlock(&delayed_node->mutex);
		goto release_node;
	}

	if (delayed_node->index_item_leaves == 0 ||
	    delayed_node->curr_index_batch_size + data_len > leaf_data_size) {
		delayed_node->curr_index_batch_size = data_len;
		reserve_leaf_space = true;
	} else {
		delayed_node->curr_index_batch_size += data_len;
		reserve_leaf_space = false;
	}

	if (reserve_leaf_space) {
		ret = btrfs_delayed_item_reserve_metadata(trans, delayed_item);
		/*
		 * Space was reserved for a dir index item insertion when we
		 * started the transaction, so getting a failure here should be
		 * impossible.
		 */
		if (WARN_ON(ret)) {
			btrfs_release_delayed_item(delayed_item);
			mutex_unlock(&delayed_node->mutex);
			goto release_node;
		}

		delayed_node->index_item_leaves++;
	} else {
		btrfs_release_dir_index_item_space(trans);
	}
	mutex_unlock(&delayed_node->mutex);

release_node:
	btrfs_release_delayed_node(delayed_node);
	return ret;
}

static int btrfs_delete_delayed_insertion_item(struct btrfs_fs_info *fs_info,
					       struct btrfs_delayed_node *node,
					       u64 index)
{
	struct btrfs_delayed_item *item;

	mutex_lock(&node->mutex);
	item = __btrfs_lookup_delayed_item(&node->ins_root.rb_root, index);
	if (!item) {
		mutex_unlock(&node->mutex);
		return 1;
	}

	/*
	 * For delayed items to insert, we track reserved metadata bytes based
	 * on the number of leaves that we will use.
	 * See btrfs_insert_delayed_dir_index() and
	 * btrfs_delayed_item_reserve_metadata()).
	 */
	ASSERT(item->bytes_reserved == 0);
	ASSERT(node->index_item_leaves > 0);

	/*
	 * If there's only one leaf reserved, we can decrement this item from the
	 * current batch, otherwise we can not because we don't know which leaf
	 * it belongs to. With the current limit on delayed items, we rarely
	 * accumulate enough dir index items to fill more than one leaf (even
	 * when using a leaf size of 4K).
	 */
	if (node->index_item_leaves == 1) {
		const u32 data_len = item->data_len + sizeof(struct btrfs_item);

		ASSERT(node->curr_index_batch_size >= data_len);
		node->curr_index_batch_size -= data_len;
	}

	btrfs_release_delayed_item(item);

	/* If we now have no more dir index items, we can release all leaves. */
	if (RB_EMPTY_ROOT(&node->ins_root.rb_root)) {
		btrfs_delayed_item_release_leaves(node, node->index_item_leaves);
		node->index_item_leaves = 0;
	}

	mutex_unlock(&node->mutex);
	return 0;
}

int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
				   struct btrfs_inode *dir, u64 index)
{
	struct btrfs_delayed_node *node;
	struct btrfs_delayed_item *item;
	int ret;

	node = btrfs_get_or_create_delayed_node(dir);
	if (IS_ERR(node))
		return PTR_ERR(node);

	ret = btrfs_delete_delayed_insertion_item(trans->fs_info, node, index);
	if (!ret)
		goto end;

	item = btrfs_alloc_delayed_item(0, node, BTRFS_DELAYED_DELETION_ITEM);
	if (!item) {
		ret = -ENOMEM;
		goto end;
	}

	item->index = index;

	ret = btrfs_delayed_item_reserve_metadata(trans, item);
	/*
	 * we have reserved enough space when we start a new transaction,
	 * so reserving metadata failure is impossible.
	 */
	if (ret < 0) {
		btrfs_err(trans->fs_info,
"metadata reservation failed for delayed dir item deltiona, should have been reserved");
		btrfs_release_delayed_item(item);
		goto end;
	}

	mutex_lock(&node->mutex);
	ret = __btrfs_add_delayed_item(node, item);
	if (unlikely(ret)) {
		btrfs_err(trans->fs_info,
			  "err add delayed dir index item(index: %llu) into the deletion tree of the delayed node(root id: %llu, inode id: %llu, errno: %d)",
			  index, node->root->root_key.objectid,
			  node->inode_id, ret);
		btrfs_delayed_item_release_metadata(dir->root, item);
		btrfs_release_delayed_item(item);
	}
	mutex_unlock(&node->mutex);
end:
	btrfs_release_delayed_node(node);
	return ret;
}

int btrfs_inode_delayed_dir_index_count(struct btrfs_inode *inode)
{
	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);

	if (!delayed_node)
		return -ENOENT;

	/*
	 * Since we have held i_mutex of this directory, it is impossible that
	 * a new directory index is added into the delayed node and index_cnt
	 * is updated now. So we needn't lock the delayed node.
	 */
	if (!delayed_node->index_cnt) {
		btrfs_release_delayed_node(delayed_node);
		return -EINVAL;
	}

	inode->index_cnt = delayed_node->index_cnt;
	btrfs_release_delayed_node(delayed_node);
	return 0;
}

bool btrfs_readdir_get_delayed_items(struct inode *inode,
				     u64 last_index,
				     struct list_head *ins_list,
				     struct list_head *del_list)
{
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_delayed_item *item;

	delayed_node = btrfs_get_delayed_node(BTRFS_I(inode));
	if (!delayed_node)
		return false;

	/*
	 * We can only do one readdir with delayed items at a time because of
	 * item->readdir_list.
	 */
	btrfs_inode_unlock(BTRFS_I(inode), BTRFS_ILOCK_SHARED);
	btrfs_inode_lock(BTRFS_I(inode), 0);

	mutex_lock(&delayed_node->mutex);
	item = __btrfs_first_delayed_insertion_item(delayed_node);
	while (item && item->index <= last_index) {
		refcount_inc(&item->refs);
		list_add_tail(&item->readdir_list, ins_list);
		item = __btrfs_next_delayed_item(item);
	}

	item = __btrfs_first_delayed_deletion_item(delayed_node);
	while (item && item->index <= last_index) {
		refcount_inc(&item->refs);
		list_add_tail(&item->readdir_list, del_list);
		item = __btrfs_next_delayed_item(item);
	}
	mutex_unlock(&delayed_node->mutex);
	/*
	 * This delayed node is still cached in the btrfs inode, so refs
	 * must be > 1 now, and we needn't check it is going to be freed
	 * or not.
	 *
	 * Besides that, this function is used to read dir, we do not
	 * insert/delete delayed items in this period. So we also needn't
	 * requeue or dequeue this delayed node.
	 */
	refcount_dec(&delayed_node->refs);

	return true;
}

void btrfs_readdir_put_delayed_items(struct inode *inode,
				     struct list_head *ins_list,
				     struct list_head *del_list)
{
	struct btrfs_delayed_item *curr, *next;

	list_for_each_entry_safe(curr, next, ins_list, readdir_list) {
		list_del(&curr->readdir_list);
		if (refcount_dec_and_test(&curr->refs))
			kfree(curr);
	}

	list_for_each_entry_safe(curr, next, del_list, readdir_list) {
		list_del(&curr->readdir_list);
		if (refcount_dec_and_test(&curr->refs))
			kfree(curr);
	}

	/*
	 * The VFS is going to do up_read(), so we need to downgrade back to a
	 * read lock.
	 */
	downgrade_write(&inode->i_rwsem);
}

int btrfs_should_delete_dir_index(struct list_head *del_list,
				  u64 index)
{
	struct btrfs_delayed_item *curr;
	int ret = 0;

	list_for_each_entry(curr, del_list, readdir_list) {
		if (curr->index > index)
			break;
		if (curr->index == index) {
			ret = 1;
			break;
		}
	}
	return ret;
}

/*
 * Read dir info stored in the delayed tree.
 */
int btrfs_readdir_delayed_dir_index(struct dir_context *ctx,
				    struct list_head *ins_list)
{
	struct btrfs_dir_item *di;
	struct btrfs_delayed_item *curr, *next;
	struct btrfs_key location;
	char *name;
	int name_len;
	int over = 0;
	unsigned char d_type;

	/*
	 * Changing the data of the delayed item is impossible. So
	 * we needn't lock them. And we have held i_mutex of the
	 * directory, nobody can delete any directory indexes now.
	 */
	list_for_each_entry_safe(curr, next, ins_list, readdir_list) {
		list_del(&curr->readdir_list);

		if (curr->index < ctx->pos) {
			if (refcount_dec_and_test(&curr->refs))
				kfree(curr);
			continue;
		}

		ctx->pos = curr->index;

		di = (struct btrfs_dir_item *)curr->data;
		name = (char *)(di + 1);
		name_len = btrfs_stack_dir_name_len(di);

		d_type = fs_ftype_to_dtype(btrfs_dir_flags_to_ftype(di->type));
		btrfs_disk_key_to_cpu(&location, &di->location);

		over = !dir_emit(ctx, name, name_len,
			       location.objectid, d_type);

		if (refcount_dec_and_test(&curr->refs))
			kfree(curr);

		if (over)
			return 1;
		ctx->pos++;
	}
	return 0;
}

static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
				  struct btrfs_inode_item *inode_item,
				  struct inode *inode)
{
	u64 flags;

	btrfs_set_stack_inode_uid(inode_item, i_uid_read(inode));
	btrfs_set_stack_inode_gid(inode_item, i_gid_read(inode));
	btrfs_set_stack_inode_size(inode_item, BTRFS_I(inode)->disk_i_size);
	btrfs_set_stack_inode_mode(inode_item, inode->i_mode);
	btrfs_set_stack_inode_nlink(inode_item, inode->i_nlink);
	btrfs_set_stack_inode_nbytes(inode_item, inode_get_bytes(inode));
	btrfs_set_stack_inode_generation(inode_item,
					 BTRFS_I(inode)->generation);
	btrfs_set_stack_inode_sequence(inode_item,
				       inode_peek_iversion(inode));
	btrfs_set_stack_inode_transid(inode_item, trans->transid);
	btrfs_set_stack_inode_rdev(inode_item, inode->i_rdev);
	flags = btrfs_inode_combine_flags(BTRFS_I(inode)->flags,
					  BTRFS_I(inode)->ro_flags);
	btrfs_set_stack_inode_flags(inode_item, flags);
	btrfs_set_stack_inode_block_group(inode_item, 0);

	btrfs_set_stack_timespec_sec(&inode_item->atime,
				     inode_get_atime_sec(inode));
	btrfs_set_stack_timespec_nsec(&inode_item->atime,
				      inode_get_atime_nsec(inode));

	btrfs_set_stack_timespec_sec(&inode_item->mtime,
				     inode_get_mtime_sec(inode));
	btrfs_set_stack_timespec_nsec(&inode_item->mtime,
				      inode_get_mtime_nsec(inode));

	btrfs_set_stack_timespec_sec(&inode_item->ctime,
				     inode_get_ctime_sec(inode));
	btrfs_set_stack_timespec_nsec(&inode_item->ctime,
				      inode_get_ctime_nsec(inode));

	btrfs_set_stack_timespec_sec(&inode_item->otime, BTRFS_I(inode)->i_otime_sec);
	btrfs_set_stack_timespec_nsec(&inode_item->otime, BTRFS_I(inode)->i_otime_nsec);
}

int btrfs_fill_inode(struct inode *inode, u32 *rdev)
{
	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_inode_item *inode_item;

	delayed_node = btrfs_get_delayed_node(BTRFS_I(inode));
	if (!delayed_node)
		return -ENOENT;

	mutex_lock(&delayed_node->mutex);
	if (!test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags)) {
		mutex_unlock(&delayed_node->mutex);
		btrfs_release_delayed_node(delayed_node);
		return -ENOENT;
	}

	inode_item = &delayed_node->inode_item;

	i_uid_write(inode, btrfs_stack_inode_uid(inode_item));
	i_gid_write(inode, btrfs_stack_inode_gid(inode_item));
	btrfs_i_size_write(BTRFS_I(inode), btrfs_stack_inode_size(inode_item));
	btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0,
			round_up(i_size_read(inode), fs_info->sectorsize));
	inode->i_mode = btrfs_stack_inode_mode(inode_item);
	set_nlink(inode, btrfs_stack_inode_nlink(inode_item));
	inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item));
	BTRFS_I(inode)->generation = btrfs_stack_inode_generation(inode_item);
        BTRFS_I(inode)->last_trans = btrfs_stack_inode_transid(inode_item);

	inode_set_iversion_queried(inode,
				   btrfs_stack_inode_sequence(inode_item));
	inode->i_rdev = 0;
	*rdev = btrfs_stack_inode_rdev(inode_item);
	btrfs_inode_split_flags(btrfs_stack_inode_flags(inode_item),
				&BTRFS_I(inode)->flags, &BTRFS_I(inode)->ro_flags);

	inode_set_atime(inode, btrfs_stack_timespec_sec(&inode_item->atime),
			btrfs_stack_timespec_nsec(&inode_item->atime));

	inode_set_mtime(inode, btrfs_stack_timespec_sec(&inode_item->mtime),
			btrfs_stack_timespec_nsec(&inode_item->mtime));

	inode_set_ctime(inode, btrfs_stack_timespec_sec(&inode_item->ctime),
			btrfs_stack_timespec_nsec(&inode_item->ctime));

	BTRFS_I(inode)->i_otime_sec = btrfs_stack_timespec_sec(&inode_item->otime);
	BTRFS_I(inode)->i_otime_nsec = btrfs_stack_timespec_nsec(&inode_item->otime);

	inode->i_generation = BTRFS_I(inode)->generation;
	BTRFS_I(inode)->index_cnt = (u64)-1;

	mutex_unlock(&delayed_node->mutex);
	btrfs_release_delayed_node(delayed_node);
	return 0;
}

int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
			       struct btrfs_inode *inode)
{
	struct btrfs_root *root = inode->root;
	struct btrfs_delayed_node *delayed_node;
	int ret = 0;

	delayed_node = btrfs_get_or_create_delayed_node(inode);
	if (IS_ERR(delayed_node))
		return PTR_ERR(delayed_node);

	mutex_lock(&delayed_node->mutex);
	if (test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags)) {
		fill_stack_inode_item(trans, &delayed_node->inode_item,
				      &inode->vfs_inode);
		goto release_node;
	}

	ret = btrfs_delayed_inode_reserve_metadata(trans, root, delayed_node);
	if (ret)
		goto release_node;

	fill_stack_inode_item(trans, &delayed_node->inode_item, &inode->vfs_inode);
	set_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags);
	delayed_node->count++;
	atomic_inc(&root->fs_info->delayed_root->items);
release_node:
	mutex_unlock(&delayed_node->mutex);
	btrfs_release_delayed_node(delayed_node);
	return ret;
}

int btrfs_delayed_delete_inode_ref(struct btrfs_inode *inode)
{
	struct btrfs_fs_info *fs_info = inode->root->fs_info;
	struct btrfs_delayed_node *delayed_node;

	/*
	 * we don't do delayed inode updates during log recovery because it
	 * leads to enospc problems.  This means we also can't do
	 * delayed inode refs
	 */
	if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags))
		return -EAGAIN;

	delayed_node = btrfs_get_or_create_delayed_node(inode);
	if (IS_ERR(delayed_node))
		return PTR_ERR(delayed_node);

	/*
	 * We don't reserve space for inode ref deletion is because:
	 * - We ONLY do async inode ref deletion for the inode who has only
	 *   one link(i_nlink == 1), it means there is only one inode ref.
	 *   And in most case, the inode ref and the inode item are in the
	 *   same leaf, and we will deal with them at the same time.
	 *   Since we are sure we will reserve the space for the inode item,
	 *   it is unnecessary to reserve space for inode ref deletion.
	 * - If the inode ref and the inode item are not in the same leaf,
	 *   We also needn't worry about enospc problem, because we reserve
	 *   much more space for the inode update than it needs.
	 * - At the worst, we can steal some space from the global reservation.
	 *   It is very rare.
	 */
	mutex_lock(&delayed_node->mutex);
	if (test_bit(BTRFS_DELAYED_NODE_DEL_IREF, &delayed_node->flags))
		goto release_node;

	set_bit(BTRFS_DELAYED_NODE_DEL_IREF, &delayed_node->flags);
	delayed_node->count++;
	atomic_inc(&fs_info->delayed_root->items);
release_node:
	mutex_unlock(&delayed_node->mutex);
	btrfs_release_delayed_node(delayed_node);
	return 0;
}

static void __btrfs_kill_delayed_node(struct btrfs_delayed_node *delayed_node)
{
	struct btrfs_root *root = delayed_node->root;
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_delayed_item *curr_item, *prev_item;

	mutex_lock(&delayed_node->mutex);
	curr_item = __btrfs_first_delayed_insertion_item(delayed_node);
	while (curr_item) {
		prev_item = curr_item;
		curr_item = __btrfs_next_delayed_item(prev_item);
		btrfs_release_delayed_item(prev_item);
	}

	if (delayed_node->index_item_leaves > 0) {
		btrfs_delayed_item_release_leaves(delayed_node,
					  delayed_node->index_item_leaves);
		delayed_node->index_item_leaves = 0;
	}

	curr_item = __btrfs_first_delayed_deletion_item(delayed_node);
	while (curr_item) {
		btrfs_delayed_item_release_metadata(root, curr_item);
		prev_item = curr_item;
		curr_item = __btrfs_next_delayed_item(prev_item);
		btrfs_release_delayed_item(prev_item);
	}

	btrfs_release_delayed_iref(delayed_node);

	if (test_bit(BTRFS_DELAYED_NODE_INODE_DIRTY, &delayed_node->flags)) {
		btrfs_delayed_inode_release_metadata(fs_info, delayed_node, false);
		btrfs_release_delayed_inode(delayed_node);
	}
	mutex_unlock(&delayed_node->mutex);
}

void btrfs_kill_delayed_inode_items(struct btrfs_inode *inode)
{
	struct btrfs_delayed_node *delayed_node;

	delayed_node = btrfs_get_delayed_node(inode);
	if (!delayed_node)
		return;

	__btrfs_kill_delayed_node(delayed_node);
	btrfs_release_delayed_node(delayed_node);
}

void btrfs_kill_all_delayed_nodes(struct btrfs_root *root)
{
	unsigned long index = 0;
	struct btrfs_delayed_node *delayed_nodes[8];

	while (1) {
		struct btrfs_delayed_node *node;
		int count;

		spin_lock(&root->inode_lock);
		if (xa_empty(&root->delayed_nodes)) {
			spin_unlock(&root->inode_lock);
			return;
		}

		count = 0;
		xa_for_each_start(&root->delayed_nodes, index, node, index) {
			/*
			 * Don't increase refs in case the node is dead and
			 * about to be removed from the tree in the loop below
			 */
			if (refcount_inc_not_zero(&node->refs)) {
				delayed_nodes[count] = node;
				count++;
			}
			if (count >= ARRAY_SIZE(delayed_nodes))
				break;
		}
		spin_unlock(&root->inode_lock);
		index++;

		for (int i = 0; i < count; i++) {
			__btrfs_kill_delayed_node(delayed_nodes[i]);
			btrfs_release_delayed_node(delayed_nodes[i]);
		}
	}
}

void btrfs_destroy_delayed_inodes(struct btrfs_fs_info *fs_info)
{
	struct btrfs_delayed_node *curr_node, *prev_node;

	curr_node = btrfs_first_delayed_node(fs_info->delayed_root);
	while (curr_node) {
		__btrfs_kill_delayed_node(curr_node);

		prev_node = curr_node;
		curr_node = btrfs_next_delayed_node(curr_node);
		btrfs_release_delayed_node(prev_node);
	}
}

void btrfs_log_get_delayed_items(struct btrfs_inode *inode,
				 struct list_head *ins_list,
				 struct list_head *del_list)
{
	struct btrfs_delayed_node *node;
	struct btrfs_delayed_item *item;

	node = btrfs_get_delayed_node(inode);
	if (!node)
		return;

	mutex_lock(&node->mutex);
	item = __btrfs_first_delayed_insertion_item(node);
	while (item) {
		/*
		 * It's possible that the item is already in a log list. This
		 * can happen in case two tasks are trying to log the same
		 * directory. For example if we have tasks A and task B:
		 *
		 * Task A collected the delayed items into a log list while
		 * under the inode's log_mutex (at btrfs_log_inode()), but it
		 * only releases the items after logging the inodes they point
		 * to (if they are new inodes), which happens after unlocking
		 * the log mutex;
		 *
		 * Task B enters btrfs_log_inode() and acquires the log_mutex
		 * of the same directory inode, before task B releases the
		 * delayed items. This can happen for example when logging some
		 * inode we need to trigger logging of its parent directory, so
		 * logging two files that have the same parent directory can
		 * lead to this.
		 *
		 * If this happens, just ignore delayed items already in a log
		 * list. All the tasks logging the directory are under a log
		 * transaction and whichever finishes first can not sync the log
		 * before the other completes and leaves the log transaction.
		 */
		if (!item->logged && list_empty(&item->log_list)) {
			refcount_inc(&item->refs);
			list_add_tail(&item->log_list, ins_list);
		}
		item = __btrfs_next_delayed_item(item);
	}

	item = __btrfs_first_delayed_deletion_item(node);
	while (item) {
		/* It may be non-empty, for the same reason mentioned above. */
		if (!item->logged && list_empty(&item->log_list)) {
			refcount_inc(&item->refs);
			list_add_tail(&item->log_list, del_list);
		}
		item = __btrfs_next_delayed_item(item);
	}
	mutex_unlock(&node->mutex);

	/*
	 * We are called during inode logging, which means the inode is in use
	 * and can not be evicted before we finish logging the inode. So we never
	 * have the last reference on the delayed inode.
	 * Also, we don't use btrfs_release_delayed_node() because that would
	 * requeue the delayed inode (change its order in the list of prepared
	 * nodes) and we don't want to do such change because we don't create or
	 * delete delayed items.
	 */
	ASSERT(refcount_read(&node->refs) > 1);
	refcount_dec(&node->refs);
}

void btrfs_log_put_delayed_items(struct btrfs_inode *inode,
				 struct list_head *ins_list,
				 struct list_head *del_list)
{
	struct btrfs_delayed_node *node;
	struct btrfs_delayed_item *item;
	struct btrfs_delayed_item *next;

	node = btrfs_get_delayed_node(inode);
	if (!node)
		return;

	mutex_lock(&node->mutex);

	list_for_each_entry_safe(item, next, ins_list, log_list) {
		item->logged = true;
		list_del_init(&item->log_list);
		if (refcount_dec_and_test(&item->refs))
			kfree(item);
	}

	list_for_each_entry_safe(item, next, del_list, log_list) {
		item->logged = true;
		list_del_init(&item->log_list);
		if (refcount_dec_and_test(&item->refs))
			kfree(item);
	}

	mutex_unlock(&node->mutex);

	/*
	 * We are called during inode logging, which means the inode is in use
	 * and can not be evicted before we finish logging the inode. So we never
	 * have the last reference on the delayed inode.
	 * Also, we don't use btrfs_release_delayed_node() because that would
	 * requeue the delayed inode (change its order in the list of prepared
	 * nodes) and we don't want to do such change because we don't create or
	 * delete delayed items.
	 */
	ASSERT(refcount_read(&node->refs) > 1);
	refcount_dec(&node->refs);
}