summaryrefslogtreecommitdiffstats
path: root/fs/btrfs/backref.c
blob: 33cba1abf8b6025c79d2de96575b54f8f350c373 (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
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
/*
 * Copyright (C) 2011 STRATO.  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/mm.h>
#include <linux/rbtree.h>
#include <trace/events/btrfs.h>
#include "ctree.h"
#include "disk-io.h"
#include "backref.h"
#include "ulist.h"
#include "transaction.h"
#include "delayed-ref.h"
#include "locking.h"

/* Just an arbitrary number so we can be sure this happened */
#define BACKREF_FOUND_SHARED 6

struct extent_inode_elem {
	u64 inum;
	u64 offset;
	struct extent_inode_elem *next;
};

static int check_extent_in_eb(const struct btrfs_key *key,
			      const struct extent_buffer *eb,
			      const struct btrfs_file_extent_item *fi,
			      u64 extent_item_pos,
			      struct extent_inode_elem **eie)
{
	u64 offset = 0;
	struct extent_inode_elem *e;

	if (!btrfs_file_extent_compression(eb, fi) &&
	    !btrfs_file_extent_encryption(eb, fi) &&
	    !btrfs_file_extent_other_encoding(eb, fi)) {
		u64 data_offset;
		u64 data_len;

		data_offset = btrfs_file_extent_offset(eb, fi);
		data_len = btrfs_file_extent_num_bytes(eb, fi);

		if (extent_item_pos < data_offset ||
		    extent_item_pos >= data_offset + data_len)
			return 1;
		offset = extent_item_pos - data_offset;
	}

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

	e->next = *eie;
	e->inum = key->objectid;
	e->offset = key->offset + offset;
	*eie = e;

	return 0;
}

static void free_inode_elem_list(struct extent_inode_elem *eie)
{
	struct extent_inode_elem *eie_next;

	for (; eie; eie = eie_next) {
		eie_next = eie->next;
		kfree(eie);
	}
}

static int find_extent_in_eb(const struct extent_buffer *eb,
			     u64 wanted_disk_byte, u64 extent_item_pos,
			     struct extent_inode_elem **eie)
{
	u64 disk_byte;
	struct btrfs_key key;
	struct btrfs_file_extent_item *fi;
	int slot;
	int nritems;
	int extent_type;
	int ret;

	/*
	 * from the shared data ref, we only have the leaf but we need
	 * the key. thus, we must look into all items and see that we
	 * find one (some) with a reference to our extent item.
	 */
	nritems = btrfs_header_nritems(eb);
	for (slot = 0; slot < nritems; ++slot) {
		btrfs_item_key_to_cpu(eb, &key, slot);
		if (key.type != BTRFS_EXTENT_DATA_KEY)
			continue;
		fi = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
		extent_type = btrfs_file_extent_type(eb, fi);
		if (extent_type == BTRFS_FILE_EXTENT_INLINE)
			continue;
		/* don't skip BTRFS_FILE_EXTENT_PREALLOC, we can handle that */
		disk_byte = btrfs_file_extent_disk_bytenr(eb, fi);
		if (disk_byte != wanted_disk_byte)
			continue;

		ret = check_extent_in_eb(&key, eb, fi, extent_item_pos, eie);
		if (ret < 0)
			return ret;
	}

	return 0;
}

struct preftree {
	struct rb_root root;
	unsigned int count;
};

#define PREFTREE_INIT	{ .root = RB_ROOT, .count = 0 }

struct preftrees {
	struct preftree direct;    /* BTRFS_SHARED_[DATA|BLOCK]_REF_KEY */
	struct preftree indirect;  /* BTRFS_[TREE_BLOCK|EXTENT_DATA]_REF_KEY */
	struct preftree indirect_missing_keys;
};

/*
 * Checks for a shared extent during backref search.
 *
 * The share_count tracks prelim_refs (direct and indirect) having a
 * ref->count >0:
 *  - incremented when a ref->count transitions to >0
 *  - decremented when a ref->count transitions to <1
 */
struct share_check {
	u64 root_objectid;
	u64 inum;
	int share_count;
};

static inline int extent_is_shared(struct share_check *sc)
{
	return (sc && sc->share_count > 1) ? BACKREF_FOUND_SHARED : 0;
}

static struct kmem_cache *btrfs_prelim_ref_cache;

int __init btrfs_prelim_ref_init(void)
{
	btrfs_prelim_ref_cache = kmem_cache_create("btrfs_prelim_ref",
					sizeof(struct prelim_ref),
					0,
					SLAB_MEM_SPREAD,
					NULL);
	if (!btrfs_prelim_ref_cache)
		return -ENOMEM;
	return 0;
}

void btrfs_prelim_ref_exit(void)
{
	kmem_cache_destroy(btrfs_prelim_ref_cache);
}

static void free_pref(struct prelim_ref *ref)
{
	kmem_cache_free(btrfs_prelim_ref_cache, ref);
}

/*
 * Return 0 when both refs are for the same block (and can be merged).
 * A -1 return indicates ref1 is a 'lower' block than ref2, while 1
 * indicates a 'higher' block.
 */
static int prelim_ref_compare(struct prelim_ref *ref1,
			      struct prelim_ref *ref2)
{
	if (ref1->level < ref2->level)
		return -1;
	if (ref1->level > ref2->level)
		return 1;
	if (ref1->root_id < ref2->root_id)
		return -1;
	if (ref1->root_id > ref2->root_id)
		return 1;
	if (ref1->key_for_search.type < ref2->key_for_search.type)
		return -1;
	if (ref1->key_for_search.type > ref2->key_for_search.type)
		return 1;
	if (ref1->key_for_search.objectid < ref2->key_for_search.objectid)
		return -1;
	if (ref1->key_for_search.objectid > ref2->key_for_search.objectid)
		return 1;
	if (ref1->key_for_search.offset < ref2->key_for_search.offset)
		return -1;
	if (ref1->key_for_search.offset > ref2->key_for_search.offset)
		return 1;
	if (ref1->parent < ref2->parent)
		return -1;
	if (ref1->parent > ref2->parent)
		return 1;

	return 0;
}

void update_share_count(struct share_check *sc, int oldcount, int newcount)
{
	if ((!sc) || (oldcount == 0 && newcount < 1))
		return;

	if (oldcount > 0 && newcount < 1)
		sc->share_count--;
	else if (oldcount < 1 && newcount > 0)
		sc->share_count++;
}

/*
 * Add @newref to the @root rbtree, merging identical refs.
 *
 * Callers should assume that newref has been freed after calling.
 */
static void prelim_ref_insert(const struct btrfs_fs_info *fs_info,
			      struct preftree *preftree,
			      struct prelim_ref *newref,
			      struct share_check *sc)
{
	struct rb_root *root;
	struct rb_node **p;
	struct rb_node *parent = NULL;
	struct prelim_ref *ref;
	int result;

	root = &preftree->root;
	p = &root->rb_node;

	while (*p) {
		parent = *p;
		ref = rb_entry(parent, struct prelim_ref, rbnode);
		result = prelim_ref_compare(ref, newref);
		if (result < 0) {
			p = &(*p)->rb_left;
		} else if (result > 0) {
			p = &(*p)->rb_right;
		} else {
			/* Identical refs, merge them and free @newref */
			struct extent_inode_elem *eie = ref->inode_list;

			while (eie && eie->next)
				eie = eie->next;

			if (!eie)
				ref->inode_list = newref->inode_list;
			else
				eie->next = newref->inode_list;
			trace_btrfs_prelim_ref_merge(fs_info, ref, newref,
						     preftree->count);
			/*
			 * A delayed ref can have newref->count < 0.
			 * The ref->count is updated to follow any
			 * BTRFS_[ADD|DROP]_DELAYED_REF actions.
			 */
			update_share_count(sc, ref->count,
					   ref->count + newref->count);
			ref->count += newref->count;
			free_pref(newref);
			return;
		}
	}

	update_share_count(sc, 0, newref->count);
	preftree->count++;
	trace_btrfs_prelim_ref_insert(fs_info, newref, NULL, preftree->count);
	rb_link_node(&newref->rbnode, parent, p);
	rb_insert_color(&newref->rbnode, root);
}

/*
 * Release the entire tree.  We don't care about internal consistency so
 * just free everything and then reset the tree root.
 */
static void prelim_release(struct preftree *preftree)
{
	struct prelim_ref *ref, *next_ref;

	rbtree_postorder_for_each_entry_safe(ref, next_ref, &preftree->root,
					     rbnode)
		free_pref(ref);

	preftree->root = RB_ROOT;
	preftree->count = 0;
}

/*
 * the rules for all callers of this function are:
 * - obtaining the parent is the goal
 * - if you add a key, you must know that it is a correct key
 * - if you cannot add the parent or a correct key, then we will look into the
 *   block later to set a correct key
 *
 * delayed refs
 * ============
 *        backref type | shared | indirect | shared | indirect
 * information         |   tree |     tree |   data |     data
 * --------------------+--------+----------+--------+----------
 *      parent logical |    y   |     -    |    -   |     -
 *      key to resolve |    -   |     y    |    y   |     y
 *  tree block logical |    -   |     -    |    -   |     -
 *  root for resolving |    y   |     y    |    y   |     y
 *
 * - column 1:       we've the parent -> done
 * - column 2, 3, 4: we use the key to find the parent
 *
 * on disk refs (inline or keyed)
 * ==============================
 *        backref type | shared | indirect | shared | indirect
 * information         |   tree |     tree |   data |     data
 * --------------------+--------+----------+--------+----------
 *      parent logical |    y   |     -    |    y   |     -
 *      key to resolve |    -   |     -    |    -   |     y
 *  tree block logical |    y   |     y    |    y   |     y
 *  root for resolving |    -   |     y    |    y   |     y
 *
 * - column 1, 3: we've the parent -> done
 * - column 2:    we take the first key from the block to find the parent
 *                (see add_missing_keys)
 * - column 4:    we use the key to find the parent
 *
 * additional information that's available but not required to find the parent
 * block might help in merging entries to gain some speed.
 */
static int add_prelim_ref(const struct btrfs_fs_info *fs_info,
			  struct preftree *preftree, u64 root_id,
			  const struct btrfs_key *key, int level, u64 parent,
			  u64 wanted_disk_byte, int count,
			  struct share_check *sc, gfp_t gfp_mask)
{
	struct prelim_ref *ref;

	if (root_id == BTRFS_DATA_RELOC_TREE_OBJECTID)
		return 0;

	ref = kmem_cache_alloc(btrfs_prelim_ref_cache, gfp_mask);
	if (!ref)
		return -ENOMEM;

	ref->root_id = root_id;
	if (key) {
		ref->key_for_search = *key;
		/*
		 * We can often find data backrefs with an offset that is too
		 * large (>= LLONG_MAX, maximum allowed file offset) due to
		 * underflows when subtracting a file's offset with the data
		 * offset of its corresponding extent data item. This can
		 * happen for example in the clone ioctl.
		 * So if we detect such case we set the search key's offset to
		 * zero to make sure we will find the matching file extent item
		 * at add_all_parents(), otherwise we will miss it because the
		 * offset taken form the backref is much larger then the offset
		 * of the file extent item. This can make us scan a very large
		 * number of file extent items, but at least it will not make
		 * us miss any.
		 * This is an ugly workaround for a behaviour that should have
		 * never existed, but it does and a fix for the clone ioctl
		 * would touch a lot of places, cause backwards incompatibility
		 * and would not fix the problem for extents cloned with older
		 * kernels.
		 */
		if (ref->key_for_search.type == BTRFS_EXTENT_DATA_KEY &&
		    ref->key_for_search.offset >= LLONG_MAX)
			ref->key_for_search.offset = 0;
	} else {
		memset(&ref->key_for_search, 0, sizeof(ref->key_for_search));
	}

	ref->inode_list = NULL;
	ref->level = level;
	ref->count = count;
	ref->parent = parent;
	ref->wanted_disk_byte = wanted_disk_byte;
	prelim_ref_insert(fs_info, preftree, ref, sc);
	return extent_is_shared(sc);
}

/* direct refs use root == 0, key == NULL */
static int add_direct_ref(const struct btrfs_fs_info *fs_info,
			  struct preftrees *preftrees, int level, u64 parent,
			  u64 wanted_disk_byte, int count,
			  struct share_check *sc, gfp_t gfp_mask)
{
	return add_prelim_ref(fs_info, &preftrees->direct, 0, NULL, level,
			      parent, wanted_disk_byte, count, sc, gfp_mask);
}

/* indirect refs use parent == 0 */
static int add_indirect_ref(const struct btrfs_fs_info *fs_info,
			    struct preftrees *preftrees, u64 root_id,
			    const struct btrfs_key *key, int level,
			    u64 wanted_disk_byte, int count,
			    struct share_check *sc, gfp_t gfp_mask)
{
	struct preftree *tree = &preftrees->indirect;

	if (!key)
		tree = &preftrees->indirect_missing_keys;
	return add_prelim_ref(fs_info, tree, root_id, key, level, 0,
			      wanted_disk_byte, count, sc, gfp_mask);
}

static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path,
			   struct ulist *parents, struct prelim_ref *ref,
			   int level, u64 time_seq, const u64 *extent_item_pos,
			   u64 total_refs)
{
	int ret = 0;
	int slot;
	struct extent_buffer *eb;
	struct btrfs_key key;
	struct btrfs_key *key_for_search = &ref->key_for_search;
	struct btrfs_file_extent_item *fi;
	struct extent_inode_elem *eie = NULL, *old = NULL;
	u64 disk_byte;
	u64 wanted_disk_byte = ref->wanted_disk_byte;
	u64 count = 0;

	if (level != 0) {
		eb = path->nodes[level];
		ret = ulist_add(parents, eb->start, 0, GFP_NOFS);
		if (ret < 0)
			return ret;
		return 0;
	}

	/*
	 * We normally enter this function with the path already pointing to
	 * the first item to check. But sometimes, we may enter it with
	 * slot==nritems. In that case, go to the next leaf before we continue.
	 */
	if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
		if (time_seq == SEQ_LAST)
			ret = btrfs_next_leaf(root, path);
		else
			ret = btrfs_next_old_leaf(root, path, time_seq);
	}

	while (!ret && count < total_refs) {
		eb = path->nodes[0];
		slot = path->slots[0];

		btrfs_item_key_to_cpu(eb, &key, slot);

		if (key.objectid != key_for_search->objectid ||
		    key.type != BTRFS_EXTENT_DATA_KEY)
			break;

		fi = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item);
		disk_byte = btrfs_file_extent_disk_bytenr(eb, fi);

		if (disk_byte == wanted_disk_byte) {
			eie = NULL;
			old = NULL;
			count++;
			if (extent_item_pos) {
				ret = check_extent_in_eb(&key, eb, fi,
						*extent_item_pos,
						&eie);
				if (ret < 0)
					break;
			}
			if (ret > 0)
				goto next;
			ret = ulist_add_merge_ptr(parents, eb->start,
						  eie, (void **)&old, GFP_NOFS);
			if (ret < 0)
				break;
			if (!ret && extent_item_pos) {
				while (old->next)
					old = old->next;
				old->next = eie;
			}
			eie = NULL;
		}
next:
		if (time_seq == SEQ_LAST)
			ret = btrfs_next_item(root, path);
		else
			ret = btrfs_next_old_item(root, path, time_seq);
	}

	if (ret > 0)
		ret = 0;
	else if (ret < 0)
		free_inode_elem_list(eie);
	return ret;
}

/*
 * resolve an indirect backref in the form (root_id, key, level)
 * to a logical address
 */
static int resolve_indirect_ref(struct btrfs_fs_info *fs_info,
				struct btrfs_path *path, u64 time_seq,
				struct prelim_ref *ref, struct ulist *parents,
				const u64 *extent_item_pos, u64 total_refs)
{
	struct btrfs_root *root;
	struct btrfs_key root_key;
	struct extent_buffer *eb;
	int ret = 0;
	int root_level;
	int level = ref->level;
	int index;

	root_key.objectid = ref->root_id;
	root_key.type = BTRFS_ROOT_ITEM_KEY;
	root_key.offset = (u64)-1;

	index = srcu_read_lock(&fs_info->subvol_srcu);

	root = btrfs_get_fs_root(fs_info, &root_key, false);
	if (IS_ERR(root)) {
		srcu_read_unlock(&fs_info->subvol_srcu, index);
		ret = PTR_ERR(root);
		goto out;
	}

	if (btrfs_is_testing(fs_info)) {
		srcu_read_unlock(&fs_info->subvol_srcu, index);
		ret = -ENOENT;
		goto out;
	}

	if (path->search_commit_root)
		root_level = btrfs_header_level(root->commit_root);
	else if (time_seq == SEQ_LAST)
		root_level = btrfs_header_level(root->node);
	else
		root_level = btrfs_old_root_level(root, time_seq);

	if (root_level + 1 == level) {
		srcu_read_unlock(&fs_info->subvol_srcu, index);
		goto out;
	}

	path->lowest_level = level;
	if (time_seq == SEQ_LAST)
		ret = btrfs_search_slot(NULL, root, &ref->key_for_search, path,
					0, 0);
	else
		ret = btrfs_search_old_slot(root, &ref->key_for_search, path,
					    time_seq);

	/* root node has been locked, we can release @subvol_srcu safely here */
	srcu_read_unlock(&fs_info->subvol_srcu, index);

	btrfs_debug(fs_info,
		"search slot in root %llu (level %d, ref count %d) returned %d for key (%llu %u %llu)",
		 ref->root_id, level, ref->count, ret,
		 ref->key_for_search.objectid, ref->key_for_search.type,
		 ref->key_for_search.offset);
	if (ret < 0)
		goto out;

	eb = path->nodes[level];
	while (!eb) {
		if (WARN_ON(!level)) {
			ret = 1;
			goto out;
		}
		level--;
		eb = path->nodes[level];
	}

	ret = add_all_parents(root, path, parents, ref, level, time_seq,
			      extent_item_pos, total_refs);
out:
	path->lowest_level = 0;
	btrfs_release_path(path);
	return ret;
}

static struct extent_inode_elem *
unode_aux_to_inode_list(struct ulist_node *node)
{
	if (!node)
		return NULL;
	return (struct extent_inode_elem *)(uintptr_t)node->aux;
}

/*
 * We maintain three seperate rbtrees: one for direct refs, one for
 * indirect refs which have a key, and one for indirect refs which do not
 * have a key. Each tree does merge on insertion.
 *
 * Once all of the references are located, we iterate over the tree of
 * indirect refs with missing keys. An appropriate key is located and
 * the ref is moved onto the tree for indirect refs. After all missing
 * keys are thus located, we iterate over the indirect ref tree, resolve
 * each reference, and then insert the resolved reference onto the
 * direct tree (merging there too).
 *
 * New backrefs (i.e., for parent nodes) are added to the appropriate
 * rbtree as they are encountered. The new backrefs are subsequently
 * resolved as above.
 */
static int resolve_indirect_refs(struct btrfs_fs_info *fs_info,
				 struct btrfs_path *path, u64 time_seq,
				 struct preftrees *preftrees,
				 const u64 *extent_item_pos, u64 total_refs,
				 struct share_check *sc)
{
	int err;
	int ret = 0;
	struct ulist *parents;
	struct ulist_node *node;
	struct ulist_iterator uiter;
	struct rb_node *rnode;

	parents = ulist_alloc(GFP_NOFS);
	if (!parents)
		return -ENOMEM;

	/*
	 * We could trade memory usage for performance here by iterating
	 * the tree, allocating new refs for each insertion, and then
	 * freeing the entire indirect tree when we're done.  In some test
	 * cases, the tree can grow quite large (~200k objects).
	 */
	while ((rnode = rb_first(&preftrees->indirect.root))) {
		struct prelim_ref *ref;

		ref = rb_entry(rnode, struct prelim_ref, rbnode);
		if (WARN(ref->parent,
			 "BUG: direct ref found in indirect tree")) {
			ret = -EINVAL;
			goto out;
		}

		rb_erase(&ref->rbnode, &preftrees->indirect.root);
		preftrees->indirect.count--;

		if (ref->count == 0) {
			free_pref(ref);
			continue;
		}

		if (sc && sc->root_objectid &&
		    ref->root_id != sc->root_objectid) {
			free_pref(ref);
			ret = BACKREF_FOUND_SHARED;
			goto out;
		}
		err = resolve_indirect_ref(fs_info, path, time_seq, ref,
					   parents, extent_item_pos,
					   total_refs);
		/*
		 * we can only tolerate ENOENT,otherwise,we should catch error
		 * and return directly.
		 */
		if (err == -ENOENT) {
			prelim_ref_insert(fs_info, &preftrees->direct, ref,
					  NULL);
			continue;
		} else if (err) {
			free_pref(ref);
			ret = err;
			goto out;
		}

		/* we put the first parent into the ref at hand */
		ULIST_ITER_INIT(&uiter);
		node = ulist_next(parents, &uiter);
		ref->parent = node ? node->val : 0;
		ref->inode_list = unode_aux_to_inode_list(node);

		/* Add a prelim_ref(s) for any other parent(s). */
		while ((node = ulist_next(parents, &uiter))) {
			struct prelim_ref *new_ref;

			new_ref = kmem_cache_alloc(btrfs_prelim_ref_cache,
						   GFP_NOFS);
			if (!new_ref) {
				free_pref(ref);
				ret = -ENOMEM;
				goto out;
			}
			memcpy(new_ref, ref, sizeof(*ref));
			new_ref->parent = node->val;
			new_ref->inode_list = unode_aux_to_inode_list(node);
			prelim_ref_insert(fs_info, &preftrees->direct,
					  new_ref, NULL);
		}

		/*
		 * Now it's a direct ref, put it in the the direct tree. We must
		 * do this last because the ref could be merged/freed here.
		 */
		prelim_ref_insert(fs_info, &preftrees->direct, ref, NULL);

		ulist_reinit(parents);
		cond_resched();
	}
out:
	ulist_free(parents);
	return ret;
}

/*
 * read tree blocks and add keys where required.
 */
static int add_missing_keys(struct btrfs_fs_info *fs_info,
			    struct preftrees *preftrees)
{
	struct prelim_ref *ref;
	struct extent_buffer *eb;
	struct preftree *tree = &preftrees->indirect_missing_keys;
	struct rb_node *node;

	while ((node = rb_first(&tree->root))) {
		ref = rb_entry(node, struct prelim_ref, rbnode);
		rb_erase(node, &tree->root);

		BUG_ON(ref->parent);	/* should not be a direct ref */
		BUG_ON(ref->key_for_search.type);
		BUG_ON(!ref->wanted_disk_byte);

		eb = read_tree_block(fs_info, ref->wanted_disk_byte, 0);
		if (IS_ERR(eb)) {
			free_pref(ref);
			return PTR_ERR(eb);
		} else if (!extent_buffer_uptodate(eb)) {
			free_pref(ref);
			free_extent_buffer(eb);
			return -EIO;
		}
		btrfs_tree_read_lock(eb);
		if (btrfs_header_level(eb) == 0)
			btrfs_item_key_to_cpu(eb, &ref->key_for_search, 0);
		else
			btrfs_node_key_to_cpu(eb, &ref->key_for_search, 0);
		btrfs_tree_read_unlock(eb);
		free_extent_buffer(eb);
		prelim_ref_insert(fs_info, &preftrees->indirect, ref, NULL);
		cond_resched();
	}
	return 0;
}

/*
 * add all currently queued delayed refs from this head whose seq nr is
 * smaller or equal that seq to the list
 */
static int add_delayed_refs(const struct btrfs_fs_info *fs_info,
			    struct btrfs_delayed_ref_head *head, u64 seq,
			    struct preftrees *preftrees, u64 *total_refs,
			    struct share_check *sc)
{
	struct btrfs_delayed_ref_node *node;
	struct btrfs_delayed_extent_op *extent_op = head->extent_op;
	struct btrfs_key key;
	struct btrfs_key tmp_op_key;
	struct btrfs_key *op_key = NULL;
	int count;
	int ret = 0;

	if (extent_op && extent_op->update_key) {
		btrfs_disk_key_to_cpu(&tmp_op_key, &extent_op->key);
		op_key = &tmp_op_key;
	}

	spin_lock(&head->lock);
	list_for_each_entry(node, &head->ref_list, list) {
		if (node->seq > seq)
			continue;

		switch (node->action) {
		case BTRFS_ADD_DELAYED_EXTENT:
		case BTRFS_UPDATE_DELAYED_HEAD:
			WARN_ON(1);
			continue;
		case BTRFS_ADD_DELAYED_REF:
			count = node->ref_mod;
			break;
		case BTRFS_DROP_DELAYED_REF:
			count = node->ref_mod * -1;
			break;
		default:
			BUG_ON(1);
		}
		*total_refs += count;
		switch (node->type) {
		case BTRFS_TREE_BLOCK_REF_KEY: {
			/* NORMAL INDIRECT METADATA backref */
			struct btrfs_delayed_tree_ref *ref;

			ref = btrfs_delayed_node_to_tree_ref(node);
			ret = add_indirect_ref(fs_info, preftrees, ref->root,
					       &tmp_op_key, ref->level + 1,
					       node->bytenr, count, sc,
					       GFP_ATOMIC);
			break;
		}
		case BTRFS_SHARED_BLOCK_REF_KEY: {
			/* SHARED DIRECT METADATA backref */
			struct btrfs_delayed_tree_ref *ref;

			ref = btrfs_delayed_node_to_tree_ref(node);

			ret = add_direct_ref(fs_info, preftrees, ref->level + 1,
					     ref->parent, node->bytenr, count,
					     sc, GFP_ATOMIC);
			break;
		}
		case BTRFS_EXTENT_DATA_REF_KEY: {
			/* NORMAL INDIRECT DATA backref */
			struct btrfs_delayed_data_ref *ref;
			ref = btrfs_delayed_node_to_data_ref(node);

			key.objectid = ref->objectid;
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = ref->offset;

			/*
			 * Found a inum that doesn't match our known inum, we
			 * know it's shared.
			 */
			if (sc && sc->inum && ref->objectid != sc->inum) {
				ret = BACKREF_FOUND_SHARED;
				goto out;
			}

			ret = add_indirect_ref(fs_info, preftrees, ref->root,
					       &key, 0, node->bytenr, count, sc,
					       GFP_ATOMIC);
			break;
		}
		case BTRFS_SHARED_DATA_REF_KEY: {
			/* SHARED DIRECT FULL backref */
			struct btrfs_delayed_data_ref *ref;

			ref = btrfs_delayed_node_to_data_ref(node);

			ret = add_direct_ref(fs_info, preftrees, 0, ref->parent,
					     node->bytenr, count, sc,
					     GFP_ATOMIC);
			break;
		}
		default:
			WARN_ON(1);
		}
		/*
		 * We must ignore BACKREF_FOUND_SHARED until all delayed
		 * refs have been checked.
		 */
		if (ret && (ret != BACKREF_FOUND_SHARED))
			break;
	}
	if (!ret)
		ret = extent_is_shared(sc);
out:
	spin_unlock(&head->lock);
	return ret;
}

/*
 * add all inline backrefs for bytenr to the list
 *
 * Returns 0 on success, <0 on error, or BACKREF_FOUND_SHARED.
 */
static int add_inline_refs(const struct btrfs_fs_info *fs_info,
			   struct btrfs_path *path, u64 bytenr,
			   int *info_level, struct preftrees *preftrees,
			   u64 *total_refs, struct share_check *sc)
{
	int ret = 0;
	int slot;
	struct extent_buffer *leaf;
	struct btrfs_key key;
	struct btrfs_key found_key;
	unsigned long ptr;
	unsigned long end;
	struct btrfs_extent_item *ei;
	u64 flags;
	u64 item_size;

	/*
	 * enumerate all inline refs
	 */
	leaf = path->nodes[0];
	slot = path->slots[0];

	item_size = btrfs_item_size_nr(leaf, slot);
	BUG_ON(item_size < sizeof(*ei));

	ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
	flags = btrfs_extent_flags(leaf, ei);
	*total_refs += btrfs_extent_refs(leaf, ei);
	btrfs_item_key_to_cpu(leaf, &found_key, slot);

	ptr = (unsigned long)(ei + 1);
	end = (unsigned long)ei + item_size;

	if (found_key.type == BTRFS_EXTENT_ITEM_KEY &&
	    flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
		struct btrfs_tree_block_info *info;

		info = (struct btrfs_tree_block_info *)ptr;
		*info_level = btrfs_tree_block_level(leaf, info);
		ptr += sizeof(struct btrfs_tree_block_info);
		BUG_ON(ptr > end);
	} else if (found_key.type == BTRFS_METADATA_ITEM_KEY) {
		*info_level = found_key.offset;
	} else {
		BUG_ON(!(flags & BTRFS_EXTENT_FLAG_DATA));
	}

	while (ptr < end) {
		struct btrfs_extent_inline_ref *iref;
		u64 offset;
		int type;

		iref = (struct btrfs_extent_inline_ref *)ptr;
		type = btrfs_get_extent_inline_ref_type(leaf, iref,
							BTRFS_REF_TYPE_ANY);
		if (type == BTRFS_REF_TYPE_INVALID)
			return -EINVAL;

		offset = btrfs_extent_inline_ref_offset(leaf, iref);

		switch (type) {
		case BTRFS_SHARED_BLOCK_REF_KEY:
			ret = add_direct_ref(fs_info, preftrees,
					     *info_level + 1, offset,
					     bytenr, 1, NULL, GFP_NOFS);
			break;
		case BTRFS_SHARED_DATA_REF_KEY: {
			struct btrfs_shared_data_ref *sdref;
			int count;

			sdref = (struct btrfs_shared_data_ref *)(iref + 1);
			count = btrfs_shared_data_ref_count(leaf, sdref);

			ret = add_direct_ref(fs_info, preftrees, 0, offset,
					     bytenr, count, sc, GFP_NOFS);
			break;
		}
		case BTRFS_TREE_BLOCK_REF_KEY:
			ret = add_indirect_ref(fs_info, preftrees, offset,
					       NULL, *info_level + 1,
					       bytenr, 1, NULL, GFP_NOFS);
			break;
		case BTRFS_EXTENT_DATA_REF_KEY: {
			struct btrfs_extent_data_ref *dref;
			int count;
			u64 root;

			dref = (struct btrfs_extent_data_ref *)(&iref->offset);
			count = btrfs_extent_data_ref_count(leaf, dref);
			key.objectid = btrfs_extent_data_ref_objectid(leaf,
								      dref);
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = btrfs_extent_data_ref_offset(leaf, dref);

			if (sc && sc->inum && key.objectid != sc->inum) {
				ret = BACKREF_FOUND_SHARED;
				break;
			}

			root = btrfs_extent_data_ref_root(leaf, dref);

			ret = add_indirect_ref(fs_info, preftrees, root,
					       &key, 0, bytenr, count,
					       sc, GFP_NOFS);
			break;
		}
		default:
			WARN_ON(1);
		}
		if (ret)
			return ret;
		ptr += btrfs_extent_inline_ref_size(type);
	}

	return 0;
}

/*
 * add all non-inline backrefs for bytenr to the list
 *
 * Returns 0 on success, <0 on error, or BACKREF_FOUND_SHARED.
 */
static int add_keyed_refs(struct btrfs_fs_info *fs_info,
			  struct btrfs_path *path, u64 bytenr,
			  int info_level, struct preftrees *preftrees,
			  struct share_check *sc)
{
	struct btrfs_root *extent_root = fs_info->extent_root;
	int ret;
	int slot;
	struct extent_buffer *leaf;
	struct btrfs_key key;

	while (1) {
		ret = btrfs_next_item(extent_root, path);
		if (ret < 0)
			break;
		if (ret) {
			ret = 0;
			break;
		}

		slot = path->slots[0];
		leaf = path->nodes[0];
		btrfs_item_key_to_cpu(leaf, &key, slot);

		if (key.objectid != bytenr)
			break;
		if (key.type < BTRFS_TREE_BLOCK_REF_KEY)
			continue;
		if (key.type > BTRFS_SHARED_DATA_REF_KEY)
			break;

		switch (key.type) {
		case BTRFS_SHARED_BLOCK_REF_KEY:
			/* SHARED DIRECT METADATA backref */
			ret = add_direct_ref(fs_info, preftrees,
					     info_level + 1, key.offset,
					     bytenr, 1, NULL, GFP_NOFS);
			break;
		case BTRFS_SHARED_DATA_REF_KEY: {
			/* SHARED DIRECT FULL backref */
			struct btrfs_shared_data_ref *sdref;
			int count;

			sdref = btrfs_item_ptr(leaf, slot,
					      struct btrfs_shared_data_ref);
			count = btrfs_shared_data_ref_count(leaf, sdref);
			ret = add_direct_ref(fs_info, preftrees, 0,
					     key.offset, bytenr, count,
					     sc, GFP_NOFS);
			break;
		}
		case BTRFS_TREE_BLOCK_REF_KEY:
			/* NORMAL INDIRECT METADATA backref */
			ret = add_indirect_ref(fs_info, preftrees, key.offset,
					       NULL, info_level + 1, bytenr,
					       1, NULL, GFP_NOFS);
			break;
		case BTRFS_EXTENT_DATA_REF_KEY: {
			/* NORMAL INDIRECT DATA backref */
			struct btrfs_extent_data_ref *dref;
			int count;
			u64 root;

			dref = btrfs_item_ptr(leaf, slot,
					      struct btrfs_extent_data_ref);
			count = btrfs_extent_data_ref_count(leaf, dref);
			key.objectid = btrfs_extent_data_ref_objectid(leaf,
								      dref);
			key.type = BTRFS_EXTENT_DATA_KEY;
			key.offset = btrfs_extent_data_ref_offset(leaf, dref);

			if (sc && sc->inum && key.objectid != sc->inum) {
				ret = BACKREF_FOUND_SHARED;
				break;
			}

			root = btrfs_extent_data_ref_root(leaf, dref);
			ret = add_indirect_ref(fs_info, preftrees, root,
					       &key, 0, bytenr, count,
					       sc, GFP_NOFS);
			break;
		}
		default:
			WARN_ON(1);
		}
		if (ret)
			return ret;

	}

	return ret;
}

/*
 * this adds all existing backrefs (inline backrefs, backrefs and delayed
 * refs) for the given bytenr to the refs list, merges duplicates and resolves
 * indirect refs to their parent bytenr.
 * When roots are found, they're added to the roots list
 *
 * If time_seq is set to SEQ_LAST, it will not search delayed_refs, and behave
 * much like trans == NULL case, the difference only lies in it will not
 * commit root.
 * The special case is for qgroup to search roots in commit_transaction().
 *
 * @sc - if !NULL, then immediately return BACKREF_FOUND_SHARED when a
 * shared extent is detected.
 *
 * Otherwise this returns 0 for success and <0 for an error.
 *
 * FIXME some caching might speed things up
 */
static int find_parent_nodes(struct btrfs_trans_handle *trans,
			     struct btrfs_fs_info *fs_info, u64 bytenr,
			     u64 time_seq, struct ulist *refs,
			     struct ulist *roots, const u64 *extent_item_pos,
			     struct share_check *sc)
{
	struct btrfs_key key;
	struct btrfs_path *path;
	struct btrfs_delayed_ref_root *delayed_refs = NULL;
	struct btrfs_delayed_ref_head *head;
	int info_level = 0;
	int ret;
	struct prelim_ref *ref;
	struct rb_node *node;
	struct extent_inode_elem *eie = NULL;
	/* total of both direct AND indirect refs! */
	u64 total_refs = 0;
	struct preftrees preftrees = {
		.direct = PREFTREE_INIT,
		.indirect = PREFTREE_INIT,
		.indirect_missing_keys = PREFTREE_INIT
	};

	key.objectid = bytenr;
	key.offset = (u64)-1;
	if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
		key.type = BTRFS_METADATA_ITEM_KEY;
	else
		key.type = BTRFS_EXTENT_ITEM_KEY;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;
	if (!trans) {
		path->search_commit_root = 1;
		path->skip_locking = 1;
	}

	if (time_seq == SEQ_LAST)
		path->skip_locking = 1;

	/*
	 * grab both a lock on the path and a lock on the delayed ref head.
	 * We need both to get a consistent picture of how the refs look
	 * at a specified point in time
	 */
again:
	head = NULL;

	ret = btrfs_search_slot(trans, fs_info->extent_root, &key, path, 0, 0);
	if (ret < 0)
		goto out;
	BUG_ON(ret == 0);

#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
	if (trans && likely(trans->type != __TRANS_DUMMY) &&
	    time_seq != SEQ_LAST) {
#else
	if (trans && time_seq != SEQ_LAST) {
#endif
		/*
		 * look if there are updates for this ref queued and lock the
		 * head
		 */
		delayed_refs = &trans->transaction->delayed_refs;
		spin_lock(&delayed_refs->lock);
		head = btrfs_find_delayed_ref_head(delayed_refs, bytenr);
		if (head) {
			if (!mutex_trylock(&head->mutex)) {
				refcount_inc(&head->refs);
				spin_unlock(&delayed_refs->lock);

				btrfs_release_path(path);

				/*
				 * Mutex was contended, block until it's
				 * released and try again
				 */
				mutex_lock(&head->mutex);
				mutex_unlock(&head->mutex);
				btrfs_put_delayed_ref_head(head);
				goto again;
			}
			spin_unlock(&delayed_refs->lock);
			ret = add_delayed_refs(fs_info, head, time_seq,
					       &preftrees, &total_refs, sc);
			mutex_unlock(&head->mutex);
			if (ret)
				goto out;
		} else {
			spin_unlock(&delayed_refs->lock);
		}
	}

	if (path->slots[0]) {
		struct extent_buffer *leaf;
		int slot;

		path->slots[0]--;
		leaf = path->nodes[0];
		slot = path->slots[0];
		btrfs_item_key_to_cpu(leaf, &key, slot);
		if (key.objectid == bytenr &&
		    (key.type == BTRFS_EXTENT_ITEM_KEY ||
		     key.type == BTRFS_METADATA_ITEM_KEY)) {
			ret = add_inline_refs(fs_info, path, bytenr,
					      &info_level, &preftrees,
					      &total_refs, sc);
			if (ret)
				goto out;
			ret = add_keyed_refs(fs_info, path, bytenr, info_level,
					     &preftrees, sc);
			if (ret)
				goto out;
		}
	}

	btrfs_release_path(path);

	ret = add_missing_keys(fs_info, &preftrees);
	if (ret)
		goto out;

	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect_missing_keys.root));

	ret = resolve_indirect_refs(fs_info, path, time_seq, &preftrees,
				    extent_item_pos, total_refs, sc);
	if (ret)
		goto out;

	WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect.root));

	/*
	 * This walks the tree of merged and resolved refs. Tree blocks are
	 * read in as needed. Unique entries are added to the ulist, and
	 * the list of found roots is updated.
	 *
	 * We release the entire tree in one go before returning.
	 */
	node = rb_first(&preftrees.direct.root);
	while (node) {
		ref = rb_entry(node, struct prelim_ref, rbnode);
		node = rb_next(&ref->rbnode);
		WARN_ON(ref->count < 0);
		if (roots && ref->count && ref->root_id && ref->parent == 0) {
			if (sc && sc->root_objectid &&
			    ref->root_id != sc->root_objectid) {
				ret = BACKREF_FOUND_SHARED;
				goto out;
			}

			/* no parent == root of tree */
			ret = ulist_add(roots, ref->root_id, 0, GFP_NOFS);
			if (ret < 0)
				goto out;
		}
		if (ref->count && ref->parent) {
			if (extent_item_pos && !ref->inode_list &&
			    ref->level == 0) {
				struct extent_buffer *eb;

				eb = read_tree_block(fs_info, ref->parent, 0);
				if (IS_ERR(eb)) {
					ret = PTR_ERR(eb);
					goto out;
				} else if (!extent_buffer_uptodate(eb)) {
					free_extent_buffer(eb);
					ret = -EIO;
					goto out;
				}
				btrfs_tree_read_lock(eb);
				btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
				ret = find_extent_in_eb(eb, bytenr,
							*extent_item_pos, &eie);
				btrfs_tree_read_unlock_blocking(eb);
				free_extent_buffer(eb);
				if (ret < 0)
					goto out;
				ref->inode_list = eie;
			}
			ret = ulist_add_merge_ptr(refs, ref->parent,
						  ref->inode_list,
						  (void **)&eie, GFP_NOFS);
			if (ret < 0)
				goto out;
			if (!ret && extent_item_pos) {
				/*
				 * we've recorded that parent, so we must extend
				 * its inode list here
				 */
				BUG_ON(!eie);
				while (eie->next)
					eie = eie->next;
				eie->next = ref->inode_list;
			}
			eie = NULL;
		}
		cond_resched();
	}

out:
	btrfs_free_path(path);

	prelim_release(&preftrees.direct);
	prelim_release(&preftrees.indirect);
	prelim_release(&preftrees.indirect_missing_keys);

	if (ret < 0)
		free_inode_elem_list(eie);
	return ret;
}

static void free_leaf_list(struct ulist *blocks)
{
	struct ulist_node *node = NULL;
	struct extent_inode_elem *eie;
	struct ulist_iterator uiter;

	ULIST_ITER_INIT(&uiter);
	while ((node = ulist_next(blocks, &uiter))) {
		if (!node->aux)
			continue;
		eie = unode_aux_to_inode_list(node);
		free_inode_elem_list(eie);
		node->aux = 0;
	}

	ulist_free(blocks);
}

/*
 * Finds all leafs with a reference to the specified combination of bytenr and
 * offset. key_list_head will point to a list of corresponding keys (caller must
 * free each list element). The leafs will be stored in the leafs ulist, which
 * must be freed with ulist_free.
 *
 * returns 0 on success, <0 on error
 */
static int btrfs_find_all_leafs(struct btrfs_trans_handle *trans,
				struct btrfs_fs_info *fs_info, u64 bytenr,
				u64 time_seq, struct ulist **leafs,
				const u64 *extent_item_pos)
{
	int ret;

	*leafs = ulist_alloc(GFP_NOFS);
	if (!*leafs)
		return -ENOMEM;

	ret = find_parent_nodes(trans, fs_info, bytenr, time_seq,
				*leafs, NULL, extent_item_pos, NULL);
	if (ret < 0 && ret != -ENOENT) {
		free_leaf_list(*leafs);
		return ret;
	}

	return 0;
}

/*
 * walk all backrefs for a given extent to find all roots that reference this
 * extent. Walking a backref means finding all extents that reference this
 * extent and in turn walk the backrefs of those, too. Naturally this is a
 * recursive process, but here it is implemented in an iterative fashion: We
 * find all referencing extents for the extent in question and put them on a
 * list. In turn, we find all referencing extents for those, further appending
 * to the list. The way we iterate the list allows adding more elements after
 * the current while iterating. The process stops when we reach the end of the
 * list. Found roots are added to the roots list.
 *
 * returns 0 on success, < 0 on error.
 */
static int btrfs_find_all_roots_safe(struct btrfs_trans_handle *trans,
				     struct btrfs_fs_info *fs_info, u64 bytenr,
				     u64 time_seq, struct ulist **roots)
{
	struct ulist *tmp;
	struct ulist_node *node = NULL;
	struct ulist_iterator uiter;
	int ret;

	tmp = ulist_alloc(GFP_NOFS);
	if (!tmp)
		return -ENOMEM;
	*roots = ulist_alloc(GFP_NOFS);
	if (!*roots) {
		ulist_free(tmp);
		return -ENOMEM;
	}

	ULIST_ITER_INIT(&uiter);
	while (1) {
		ret = find_parent_nodes(trans, fs_info, bytenr, time_seq,
					tmp, *roots, NULL, NULL);
		if (ret < 0 && ret != -ENOENT) {
			ulist_free(tmp);
			ulist_free(*roots);
			return ret;
		}
		node = ulist_next(tmp, &uiter);
		if (!node)
			break;
		bytenr = node->val;
		cond_resched();
	}

	ulist_free(tmp);
	return 0;
}

int btrfs_find_all_roots(struct btrfs_trans_handle *trans,
			 struct btrfs_fs_info *fs_info, u64 bytenr,
			 u64 time_seq, struct ulist **roots)
{
	int ret;

	if (!trans)
		down_read(&fs_info->commit_root_sem);
	ret = btrfs_find_all_roots_safe(trans, fs_info, bytenr,
					time_seq, roots);
	if (!trans)
		up_read(&fs_info->commit_root_sem);
	return ret;
}

/**
 * btrfs_check_shared - tell us whether an extent is shared
 *
 * btrfs_check_shared uses the backref walking code but will short
 * circuit as soon as it finds a root or inode that doesn't match the
 * one passed in. This provides a significant performance benefit for
 * callers (such as fiemap) which want to know whether the extent is
 * shared but do not need a ref count.
 *
 * This attempts to allocate a transaction in order to account for
 * delayed refs, but continues on even when the alloc fails.
 *
 * Return: 0 if extent is not shared, 1 if it is shared, < 0 on error.
 */
int btrfs_check_shared(struct btrfs_root *root, u64 inum, u64 bytenr)
{
	struct btrfs_fs_info *fs_info = root->fs_info;
	struct btrfs_trans_handle *trans;
	struct ulist *tmp = NULL;
	struct ulist *roots = NULL;
	struct ulist_iterator uiter;
	struct ulist_node *node;
	struct seq_list elem = SEQ_LIST_INIT(elem);
	int ret = 0;
	struct share_check shared = {
		.root_objectid = root->objectid,
		.inum = inum,
		.share_count = 0,
	};

	tmp = ulist_alloc(GFP_NOFS);
	roots = ulist_alloc(GFP_NOFS);
	if (!tmp || !roots) {
		ulist_free(tmp);
		ulist_free(roots);
		return -ENOMEM;
	}

	trans = btrfs_join_transaction(root);
	if (IS_ERR(trans)) {
		trans = NULL;
		down_read(&fs_info->commit_root_sem);
	} else {
		btrfs_get_tree_mod_seq(fs_info, &elem);
	}

	ULIST_ITER_INIT(&uiter);
	while (1) {
		ret = find_parent_nodes(trans, fs_info, bytenr, elem.seq, tmp,
					roots, NULL, &shared);
		if (ret == BACKREF_FOUND_SHARED) {
			/* this is the only condition under which we return 1 */
			ret = 1;
			break;
		}
		if (ret < 0 && ret != -ENOENT)
			break;
		ret = 0;
		node = ulist_next(tmp, &uiter);
		if (!node)
			break;
		bytenr = node->val;
		cond_resched();
	}

	if (trans) {
		btrfs_put_tree_mod_seq(fs_info, &elem);
		btrfs_end_transaction(trans);
	} else {
		up_read(&fs_info->commit_root_sem);
	}
	ulist_free(tmp);
	ulist_free(roots);
	return ret;
}

int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid,
			  u64 start_off, struct btrfs_path *path,
			  struct btrfs_inode_extref **ret_extref,
			  u64 *found_off)
{
	int ret, slot;
	struct btrfs_key key;
	struct btrfs_key found_key;
	struct btrfs_inode_extref *extref;
	const struct extent_buffer *leaf;
	unsigned long ptr;

	key.objectid = inode_objectid;
	key.type = BTRFS_INODE_EXTREF_KEY;
	key.offset = start_off;

	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
	if (ret < 0)
		return ret;

	while (1) {
		leaf = path->nodes[0];
		slot = path->slots[0];
		if (slot >= btrfs_header_nritems(leaf)) {
			/*
			 * If the item at offset is not found,
			 * btrfs_search_slot will point us to the slot
			 * where it should be inserted. In our case
			 * that will be the slot directly before the
			 * next INODE_REF_KEY_V2 item. In the case
			 * that we're pointing to the last slot in a
			 * leaf, we must move one leaf over.
			 */
			ret = btrfs_next_leaf(root, path);
			if (ret) {
				if (ret >= 1)
					ret = -ENOENT;
				break;
			}
			continue;
		}

		btrfs_item_key_to_cpu(leaf, &found_key, slot);

		/*
		 * Check that we're still looking at an extended ref key for
		 * this particular objectid. If we have different
		 * objectid or type then there are no more to be found
		 * in the tree and we can exit.
		 */
		ret = -ENOENT;
		if (found_key.objectid != inode_objectid)
			break;
		if (found_key.type != BTRFS_INODE_EXTREF_KEY)
			break;

		ret = 0;
		ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
		extref = (struct btrfs_inode_extref *)ptr;
		*ret_extref = extref;
		if (found_off)
			*found_off = found_key.offset;
		break;
	}

	return ret;
}

/*
 * this iterates to turn a name (from iref/extref) into a full filesystem path.
 * Elements of the path are separated by '/' and the path is guaranteed to be
 * 0-terminated. the path is only given within the current file system.
 * Therefore, it never starts with a '/'. the caller is responsible to provide
 * "size" bytes in "dest". the dest buffer will be filled backwards. finally,
 * the start point of the resulting string is returned. this pointer is within
 * dest, normally.
 * in case the path buffer would overflow, the pointer is decremented further
 * as if output was written to the buffer, though no more output is actually
 * generated. that way, the caller can determine how much space would be
 * required for the path to fit into the buffer. in that case, the returned
 * value will be smaller than dest. callers must check this!
 */
char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
			u32 name_len, unsigned long name_off,
			struct extent_buffer *eb_in, u64 parent,
			char *dest, u32 size)
{
	int slot;
	u64 next_inum;
	int ret;
	s64 bytes_left = ((s64)size) - 1;
	struct extent_buffer *eb = eb_in;
	struct btrfs_key found_key;
	int leave_spinning = path->leave_spinning;
	struct btrfs_inode_ref *iref;

	if (bytes_left >= 0)
		dest[bytes_left] = '\0';

	path->leave_spinning = 1;
	while (1) {
		bytes_left -= name_len;
		if (bytes_left >= 0)
			read_extent_buffer(eb, dest + bytes_left,
					   name_off, name_len);
		if (eb != eb_in) {
			if (!path->skip_locking)
				btrfs_tree_read_unlock_blocking(eb);
			free_extent_buffer(eb);
		}
		ret = btrfs_find_item(fs_root, path, parent, 0,
				BTRFS_INODE_REF_KEY, &found_key);
		if (ret > 0)
			ret = -ENOENT;
		if (ret)
			break;

		next_inum = found_key.offset;

		/* regular exit ahead */
		if (parent == next_inum)
			break;

		slot = path->slots[0];
		eb = path->nodes[0];
		/* make sure we can use eb after releasing the path */
		if (eb != eb_in) {
			if (!path->skip_locking)
				btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
			path->nodes[0] = NULL;
			path->locks[0] = 0;
		}
		btrfs_release_path(path);
		iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref);

		name_len = btrfs_inode_ref_name_len(eb, iref);
		name_off = (unsigned long)(iref + 1);

		parent = next_inum;
		--bytes_left;
		if (bytes_left >= 0)
			dest[bytes_left] = '/';
	}

	btrfs_release_path(path);
	path->leave_spinning = leave_spinning;

	if (ret)
		return ERR_PTR(ret);

	return dest + bytes_left;
}

/*
 * this makes the path point to (logical EXTENT_ITEM *)
 * returns BTRFS_EXTENT_FLAG_DATA for data, BTRFS_EXTENT_FLAG_TREE_BLOCK for
 * tree blocks and <0 on error.
 */
int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
			struct btrfs_path *path, struct btrfs_key *found_key,
			u64 *flags_ret)
{
	int ret;
	u64 flags;
	u64 size = 0;
	u32 item_size;
	const struct extent_buffer *eb;
	struct btrfs_extent_item *ei;
	struct btrfs_key key;

	if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
		key.type = BTRFS_METADATA_ITEM_KEY;
	else
		key.type = BTRFS_EXTENT_ITEM_KEY;
	key.objectid = logical;
	key.offset = (u64)-1;

	ret = btrfs_search_slot(NULL, fs_info->extent_root, &key, path, 0, 0);
	if (ret < 0)
		return ret;

	ret = btrfs_previous_extent_item(fs_info->extent_root, path, 0);
	if (ret) {
		if (ret > 0)
			ret = -ENOENT;
		return ret;
	}
	btrfs_item_key_to_cpu(path->nodes[0], found_key, path->slots[0]);
	if (found_key->type == BTRFS_METADATA_ITEM_KEY)
		size = fs_info->nodesize;
	else if (found_key->type == BTRFS_EXTENT_ITEM_KEY)
		size = found_key->offset;

	if (found_key->objectid > logical ||
	    found_key->objectid + size <= logical) {
		btrfs_debug(fs_info,
			"logical %llu is not within any extent", logical);
		return -ENOENT;
	}

	eb = path->nodes[0];
	item_size = btrfs_item_size_nr(eb, path->slots[0]);
	BUG_ON(item_size < sizeof(*ei));

	ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
	flags = btrfs_extent_flags(eb, ei);

	btrfs_debug(fs_info,
		"logical %llu is at position %llu within the extent (%llu EXTENT_ITEM %llu) flags %#llx size %u",
		 logical, logical - found_key->objectid, found_key->objectid,
		 found_key->offset, flags, item_size);

	WARN_ON(!flags_ret);
	if (flags_ret) {
		if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK)
			*flags_ret = BTRFS_EXTENT_FLAG_TREE_BLOCK;
		else if (flags & BTRFS_EXTENT_FLAG_DATA)
			*flags_ret = BTRFS_EXTENT_FLAG_DATA;
		else
			BUG_ON(1);
		return 0;
	}

	return -EIO;
}

/*
 * helper function to iterate extent inline refs. ptr must point to a 0 value
 * for the first call and may be modified. it is used to track state.
 * if more refs exist, 0 is returned and the next call to
 * get_extent_inline_ref must pass the modified ptr parameter to get the
 * next ref. after the last ref was processed, 1 is returned.
 * returns <0 on error
 */
static int get_extent_inline_ref(unsigned long *ptr,
				 const struct extent_buffer *eb,
				 const struct btrfs_key *key,
				 const struct btrfs_extent_item *ei,
				 u32 item_size,
				 struct btrfs_extent_inline_ref **out_eiref,
				 int *out_type)
{
	unsigned long end;
	u64 flags;
	struct btrfs_tree_block_info *info;

	if (!*ptr) {
		/* first call */
		flags = btrfs_extent_flags(eb, ei);
		if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
			if (key->type == BTRFS_METADATA_ITEM_KEY) {
				/* a skinny metadata extent */
				*out_eiref =
				     (struct btrfs_extent_inline_ref *)(ei + 1);
			} else {
				WARN_ON(key->type != BTRFS_EXTENT_ITEM_KEY);
				info = (struct btrfs_tree_block_info *)(ei + 1);
				*out_eiref =
				   (struct btrfs_extent_inline_ref *)(info + 1);
			}
		} else {
			*out_eiref = (struct btrfs_extent_inline_ref *)(ei + 1);
		}
		*ptr = (unsigned long)*out_eiref;
		if ((unsigned long)(*ptr) >= (unsigned long)ei + item_size)
			return -ENOENT;
	}

	end = (unsigned long)ei + item_size;
	*out_eiref = (struct btrfs_extent_inline_ref *)(*ptr);
	*out_type = btrfs_get_extent_inline_ref_type(eb, *out_eiref,
						     BTRFS_REF_TYPE_ANY);
	if (*out_type == BTRFS_REF_TYPE_INVALID)
		return -EINVAL;

	*ptr += btrfs_extent_inline_ref_size(*out_type);
	WARN_ON(*ptr > end);
	if (*ptr == end)
		return 1; /* last */

	return 0;
}

/*
 * reads the tree block backref for an extent. tree level and root are returned
 * through out_level and out_root. ptr must point to a 0 value for the first
 * call and may be modified (see get_extent_inline_ref comment).
 * returns 0 if data was provided, 1 if there was no more data to provide or
 * <0 on error.
 */
int tree_backref_for_extent(unsigned long *ptr, struct extent_buffer *eb,
			    struct btrfs_key *key, struct btrfs_extent_item *ei,
			    u32 item_size, u64 *out_root, u8 *out_level)
{
	int ret;
	int type;
	struct btrfs_extent_inline_ref *eiref;

	if (*ptr == (unsigned long)-1)
		return 1;

	while (1) {
		ret = get_extent_inline_ref(ptr, eb, key, ei, item_size,
					      &eiref, &type);
		if (ret < 0)
			return ret;

		if (type == BTRFS_TREE_BLOCK_REF_KEY ||
		    type == BTRFS_SHARED_BLOCK_REF_KEY)
			break;

		if (ret == 1)
			return 1;
	}

	/* we can treat both ref types equally here */
	*out_root = btrfs_extent_inline_ref_offset(eb, eiref);

	if (key->type == BTRFS_EXTENT_ITEM_KEY) {
		struct btrfs_tree_block_info *info;

		info = (struct btrfs_tree_block_info *)(ei + 1);
		*out_level = btrfs_tree_block_level(eb, info);
	} else {
		ASSERT(key->type == BTRFS_METADATA_ITEM_KEY);
		*out_level = (u8)key->offset;
	}

	if (ret == 1)
		*ptr = (unsigned long)-1;

	return 0;
}

static int iterate_leaf_refs(struct btrfs_fs_info *fs_info,
			     struct extent_inode_elem *inode_list,
			     u64 root, u64 extent_item_objectid,
			     iterate_extent_inodes_t *iterate, void *ctx)
{
	struct extent_inode_elem *eie;
	int ret = 0;

	for (eie = inode_list; eie; eie = eie->next) {
		btrfs_debug(fs_info,
			    "ref for %llu resolved, key (%llu EXTEND_DATA %llu), root %llu",
			    extent_item_objectid, eie->inum,
			    eie->offset, root);
		ret = iterate(eie->inum, eie->offset, root, ctx);
		if (ret) {
			btrfs_debug(fs_info,
				    "stopping iteration for %llu due to ret=%d",
				    extent_item_objectid, ret);
			break;
		}
	}

	return ret;
}

/*
 * calls iterate() for every inode that references the extent identified by
 * the given parameters.
 * when the iterator function returns a non-zero value, iteration stops.
 */
int iterate_extent_inodes(struct btrfs_fs_info *fs_info,
				u64 extent_item_objectid, u64 extent_item_pos,
				int search_commit_root,
				iterate_extent_inodes_t *iterate, void *ctx)
{
	int ret;
	struct btrfs_trans_handle *trans = NULL;
	struct ulist *refs = NULL;
	struct ulist *roots = NULL;
	struct ulist_node *ref_node = NULL;
	struct ulist_node *root_node = NULL;
	struct seq_list tree_mod_seq_elem = SEQ_LIST_INIT(tree_mod_seq_elem);
	struct ulist_iterator ref_uiter;
	struct ulist_iterator root_uiter;

	btrfs_debug(fs_info, "resolving all inodes for extent %llu",
			extent_item_objectid);

	if (!search_commit_root) {
		trans = btrfs_join_transaction(fs_info->extent_root);
		if (IS_ERR(trans))
			return PTR_ERR(trans);
		btrfs_get_tree_mod_seq(fs_info, &tree_mod_seq_elem);
	} else {
		down_read(&fs_info->commit_root_sem);
	}

	ret = btrfs_find_all_leafs(trans, fs_info, extent_item_objectid,
				   tree_mod_seq_elem.seq, &refs,
				   &extent_item_pos);
	if (ret)
		goto out;

	ULIST_ITER_INIT(&ref_uiter);
	while (!ret && (ref_node = ulist_next(refs, &ref_uiter))) {
		ret = btrfs_find_all_roots_safe(trans, fs_info, ref_node->val,
						tree_mod_seq_elem.seq, &roots);
		if (ret)
			break;
		ULIST_ITER_INIT(&root_uiter);
		while (!ret && (root_node = ulist_next(roots, &root_uiter))) {
			btrfs_debug(fs_info,
				    "root %llu references leaf %llu, data list %#llx",
				    root_node->val, ref_node->val,
				    ref_node->aux);
			ret = iterate_leaf_refs(fs_info,
						(struct extent_inode_elem *)
						(uintptr_t)ref_node->aux,
						root_node->val,
						extent_item_objectid,
						iterate, ctx);
		}
		ulist_free(roots);
	}

	free_leaf_list(refs);
out:
	if (!search_commit_root) {
		btrfs_put_tree_mod_seq(fs_info, &tree_mod_seq_elem);
		btrfs_end_transaction(trans);
	} else {
		up_read(&fs_info->commit_root_sem);
	}

	return ret;
}

int iterate_inodes_from_logical(u64 logical, struct btrfs_fs_info *fs_info,
				struct btrfs_path *path,
				iterate_extent_inodes_t *iterate, void *ctx)
{
	int ret;
	u64 extent_item_pos;
	u64 flags = 0;
	struct btrfs_key found_key;
	int search_commit_root = path->search_commit_root;

	ret = extent_from_logical(fs_info, logical, path, &found_key, &flags);
	btrfs_release_path(path);
	if (ret < 0)
		return ret;
	if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK)
		return -EINVAL;

	extent_item_pos = logical - found_key.objectid;
	ret = iterate_extent_inodes(fs_info, found_key.objectid,
					extent_item_pos, search_commit_root,
					iterate, ctx);

	return ret;
}

typedef int (iterate_irefs_t)(u64 parent, u32 name_len, unsigned long name_off,
			      struct extent_buffer *eb, void *ctx);

static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
			      struct btrfs_path *path,
			      iterate_irefs_t *iterate, void *ctx)
{
	int ret = 0;
	int slot;
	u32 cur;
	u32 len;
	u32 name_len;
	u64 parent = 0;
	int found = 0;
	struct extent_buffer *eb;
	struct btrfs_item *item;
	struct btrfs_inode_ref *iref;
	struct btrfs_key found_key;

	while (!ret) {
		ret = btrfs_find_item(fs_root, path, inum,
				parent ? parent + 1 : 0, BTRFS_INODE_REF_KEY,
				&found_key);

		if (ret < 0)
			break;
		if (ret) {
			ret = found ? 0 : -ENOENT;
			break;
		}
		++found;

		parent = found_key.offset;
		slot = path->slots[0];
		eb = btrfs_clone_extent_buffer(path->nodes[0]);
		if (!eb) {
			ret = -ENOMEM;
			break;
		}
		extent_buffer_get(eb);
		btrfs_tree_read_lock(eb);
		btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
		btrfs_release_path(path);

		item = btrfs_item_nr(slot);
		iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref);

		for (cur = 0; cur < btrfs_item_size(eb, item); cur += len) {
			name_len = btrfs_inode_ref_name_len(eb, iref);
			/* path must be released before calling iterate()! */
			btrfs_debug(fs_root->fs_info,
				"following ref at offset %u for inode %llu in tree %llu",
				cur, found_key.objectid, fs_root->objectid);
			ret = iterate(parent, name_len,
				      (unsigned long)(iref + 1), eb, ctx);
			if (ret)
				break;
			len = sizeof(*iref) + name_len;
			iref = (struct btrfs_inode_ref *)((char *)iref + len);
		}
		btrfs_tree_read_unlock_blocking(eb);
		free_extent_buffer(eb);
	}

	btrfs_release_path(path);

	return ret;
}

static int iterate_inode_extrefs(u64 inum, struct btrfs_root *fs_root,
				 struct btrfs_path *path,
				 iterate_irefs_t *iterate, void *ctx)
{
	int ret;
	int slot;
	u64 offset = 0;
	u64 parent;
	int found = 0;
	struct extent_buffer *eb;
	struct btrfs_inode_extref *extref;
	u32 item_size;
	u32 cur_offset;
	unsigned long ptr;

	while (1) {
		ret = btrfs_find_one_extref(fs_root, inum, offset, path, &extref,
					    &offset);
		if (ret < 0)
			break;
		if (ret) {
			ret = found ? 0 : -ENOENT;
			break;
		}
		++found;

		slot = path->slots[0];
		eb = btrfs_clone_extent_buffer(path->nodes[0]);
		if (!eb) {
			ret = -ENOMEM;
			break;
		}
		extent_buffer_get(eb);

		btrfs_tree_read_lock(eb);
		btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
		btrfs_release_path(path);

		item_size = btrfs_item_size_nr(eb, slot);
		ptr = btrfs_item_ptr_offset(eb, slot);
		cur_offset = 0;

		while (cur_offset < item_size) {
			u32 name_len;

			extref = (struct btrfs_inode_extref *)(ptr + cur_offset);
			parent = btrfs_inode_extref_parent(eb, extref);
			name_len = btrfs_inode_extref_name_len(eb, extref);
			ret = iterate(parent, name_len,
				      (unsigned long)&extref->name, eb, ctx);
			if (ret)
				break;

			cur_offset += btrfs_inode_extref_name_len(eb, extref);
			cur_offset += sizeof(*extref);
		}
		btrfs_tree_read_unlock_blocking(eb);
		free_extent_buffer(eb);

		offset++;
	}

	btrfs_release_path(path);

	return ret;
}

static int iterate_irefs(u64 inum, struct btrfs_root *fs_root,
			 struct btrfs_path *path, iterate_irefs_t *iterate,
			 void *ctx)
{
	int ret;
	int found_refs = 0;

	ret = iterate_inode_refs(inum, fs_root, path, iterate, ctx);
	if (!ret)
		++found_refs;
	else if (ret != -ENOENT)
		return ret;

	ret = iterate_inode_extrefs(inum, fs_root, path, iterate, ctx);
	if (ret == -ENOENT && found_refs)
		return 0;

	return ret;
}

/*
 * returns 0 if the path could be dumped (probably truncated)
 * returns <0 in case of an error
 */
static int inode_to_path(u64 inum, u32 name_len, unsigned long name_off,
			 struct extent_buffer *eb, void *ctx)
{
	struct inode_fs_paths *ipath = ctx;
	char *fspath;
	char *fspath_min;
	int i = ipath->fspath->elem_cnt;
	const int s_ptr = sizeof(char *);
	u32 bytes_left;

	bytes_left = ipath->fspath->bytes_left > s_ptr ?
					ipath->fspath->bytes_left - s_ptr : 0;

	fspath_min = (char *)ipath->fspath->val + (i + 1) * s_ptr;
	fspath = btrfs_ref_to_path(ipath->fs_root, ipath->btrfs_path, name_len,
				   name_off, eb, inum, fspath_min, bytes_left);
	if (IS_ERR(fspath))
		return PTR_ERR(fspath);

	if (fspath > fspath_min) {
		ipath->fspath->val[i] = (u64)(unsigned long)fspath;
		++ipath->fspath->elem_cnt;
		ipath->fspath->bytes_left = fspath - fspath_min;
	} else {
		++ipath->fspath->elem_missed;
		ipath->fspath->bytes_missing += fspath_min - fspath;
		ipath->fspath->bytes_left = 0;
	}

	return 0;
}

/*
 * this dumps all file system paths to the inode into the ipath struct, provided
 * is has been created large enough. each path is zero-terminated and accessed
 * from ipath->fspath->val[i].
 * when it returns, there are ipath->fspath->elem_cnt number of paths available
 * in ipath->fspath->val[]. when the allocated space wasn't sufficient, the
 * number of missed paths is recorded in ipath->fspath->elem_missed, otherwise,
 * it's zero. ipath->fspath->bytes_missing holds the number of bytes that would
 * have been needed to return all paths.
 */
int paths_from_inode(u64 inum, struct inode_fs_paths *ipath)
{
	return iterate_irefs(inum, ipath->fs_root, ipath->btrfs_path,
			     inode_to_path, ipath);
}

struct btrfs_data_container *init_data_container(u32 total_bytes)
{
	struct btrfs_data_container *data;
	size_t alloc_bytes;

	alloc_bytes = max_t(size_t, total_bytes, sizeof(*data));
	data = kvmalloc(alloc_bytes, GFP_KERNEL);
	if (!data)
		return ERR_PTR(-ENOMEM);

	if (total_bytes >= sizeof(*data)) {
		data->bytes_left = total_bytes - sizeof(*data);
		data->bytes_missing = 0;
	} else {
		data->bytes_missing = sizeof(*data) - total_bytes;
		data->bytes_left = 0;
	}

	data->elem_cnt = 0;
	data->elem_missed = 0;

	return data;
}

/*
 * allocates space to return multiple file system paths for an inode.
 * total_bytes to allocate are passed, note that space usable for actual path
 * information will be total_bytes - sizeof(struct inode_fs_paths).
 * the returned pointer must be freed with free_ipath() in the end.
 */
struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root,
					struct btrfs_path *path)
{
	struct inode_fs_paths *ifp;
	struct btrfs_data_container *fspath;

	fspath = init_data_container(total_bytes);
	if (IS_ERR(fspath))
		return (void *)fspath;

	ifp = kmalloc(sizeof(*ifp), GFP_KERNEL);
	if (!ifp) {
		kvfree(fspath);
		return ERR_PTR(-ENOMEM);
	}

	ifp->btrfs_path = path;
	ifp->fspath = fspath;
	ifp->fs_root = fs_root;

	return ifp;
}

void free_ipath(struct inode_fs_paths *ipath)
{
	if (!ipath)
		return;
	kvfree(ipath->fspath);
	kfree(ipath);
}