summaryrefslogtreecommitdiff
path: root/Assemble.c
blob: 704b8293c07fecbe491f08cbff037096e6a12eb5 (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
/*
 * mdadm - manage Linux "md" devices aka RAID arrays.
 *
 * Copyright (C) 2001-2016 Neil Brown <neilb@suse.com>
 *
 *
 *    This program is free software; you can redistribute it and/or modify
 *    it under the terms of the GNU General Public License as published by
 *    the Free Software Foundation; either version 2 of the License, or
 *    (at your option) any later version.
 *
 *    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  02111-1307  USA
 *
 *    Author: Neil Brown
 *    Email: <neilb@suse.de>
 */

#include	"mdadm.h"
#include	<ctype.h>

mapping_t assemble_statuses[] = {
	{ "but cannot be started", INCR_NO },
	{ "but not safe to start", INCR_UNSAFE },
	{ "and started", INCR_YES },
	{ NULL, INCR_ALREADY }
};


/**
 * struct assembly_array_info - General, meaningful information for assembly.
 * @name: Array name.
 * @new_cnt: Count of drives known to be members, recently added.
 * @preexist_cnt: Count of member drives in pre-assembled array.
 * @exp_cnt: Count of known expansion targets.
 *
 * FIXME: @exp_new_cnt for recently added expansion targets.
 */
struct assembly_array_info {
	char *name;
	int new_cnt;
	int preexist_cnt;
	int exp_cnt;
};

/**
 * set_array_assembly_status() - generate status of assembly for an array.
 * @c: Global settings.
 * @result: Pointer to status mask.
 * @status: Status to be set/printed.
 * @arr: Array information.
 *
 *  Print status message to user or set it in @result if it is not NULL.
 */
static void set_array_assembly_status(struct context *c,
				   int *result, int status,
				   struct assembly_array_info *arr)
{
	int raid_disks = arr->preexist_cnt + arr->new_cnt;
	char *status_msg = map_num(assemble_statuses, status);

	if (c->export && result)
		*result |= status;

	if (c->export || c->verbose < 0)
		return;

	pr_err("%s has been assembled with %d device%s", arr->name,
	       raid_disks, raid_disks == 1 ? "":"s");
	if (arr->preexist_cnt > 0)
		fprintf(stderr, " (%d new)", arr->new_cnt);
	if (arr->exp_cnt)
		fprintf(stderr, " ( + %d for expansion)", arr->exp_cnt);
	if (status_msg)
		fprintf(stderr, " %s", status_msg);
	fprintf(stderr, ".\n");
}

static int name_matches(char *found, char *required, char *homehost, int require_homehost)
{
	/* See if the name found matches the required name, possibly
	 * prefixed with 'homehost'
	 */
	char *sep;
	unsigned int l;

	if (strcmp(found, required)==0)
		return 1;
	sep = strchr(found, ':');
	if (!sep)
		return 0;
	l = sep - found;
	if (strncmp(found, "any:", 4) == 0 ||
	    (homehost && strcmp(homehost, "any") == 0) ||
	    !require_homehost ||
	    (homehost && strlen(homehost) == l &&
	     strncmp(found, homehost, l) == 0)) {
		/* matching homehost */
		if (strcmp(sep+1, required) == 0)
			return 1;
	}
	return 0;
}

static int is_member_busy(char *metadata_version)
{
	/* check if the given member array is active */
	struct mdstat_ent *mdstat = mdstat_read(0, 0);
	struct mdstat_ent *ent;
	int busy = 0;

	for (ent = mdstat; ent; ent = ent->next) {
		if (ent->metadata_version == NULL)
			continue;
		if (strncmp(ent->metadata_version, "external:", 9) != 0)
			continue;
		if (!is_subarray(&ent->metadata_version[9]))
			continue;
		/* Skip first char - it can be '/' or '-' */
		if (strcmp(&ent->metadata_version[10], metadata_version+1) == 0) {
			busy = 1;
			break;
		}
	}
	free_mdstat(mdstat);

	return busy;
}

static int ident_matches(struct mddev_ident *ident,
			 struct mdinfo *content,
			 struct supertype *tst,
			 char *homehost, int require_homehost,
			 char *update, char *devname)
{

	if (ident->uuid_set && (!update || strcmp(update, "uuid")!= 0) &&
	    same_uuid(content->uuid, ident->uuid, tst->ss->swapuuid)==0 &&
	    memcmp(content->uuid, uuid_zero, sizeof(int[4])) != 0) {
		if (devname)
			pr_err("%s has wrong uuid.\n", devname);
		return 0;
	}
	if (ident->name[0] && (!update || strcmp(update, "name")!= 0) &&
	    name_matches(content->name, ident->name, homehost, require_homehost)==0) {
		if (devname)
			pr_err("%s has wrong name.\n", devname);
		return 0;
	}
	if (ident->super_minor != UnSet &&
	    ident->super_minor != content->array.md_minor) {
		if (devname)
			pr_err("%s has wrong super-minor.\n",
			       devname);
		return 0;
	}
	if (ident->level != UnSet &&
	    ident->level != content->array.level) {
		if (devname)
			pr_err("%s has wrong raid level.\n",
			       devname);
		return 0;
	}
	if (ident->raid_disks != UnSet &&
	    content->array.raid_disks != 0 && /* metadata doesn't know how many to expect */
	    ident->raid_disks!= content->array.raid_disks) {
		if (devname)
			pr_err("%s requires wrong number of drives.\n",
			       devname);
		return 0;
	}
	if (ident->member && ident->member[0]) {
		/* content->text_version must match */
		char *s = strchr(content->text_version+1, '/');
		if (s == NULL) {
			if (devname)
				pr_err("%s is not a container and one is required.\n",
				       devname);
			return 0;
		} else if (strcmp(ident->member, s+1) != 0) {
			if (devname)
				pr_err("skipping wrong member %s is %s\n",
				       content->text_version, devname);
			return 0;
		}
	}
	return 1;
}

static int select_devices(struct mddev_dev *devlist,
			  struct mddev_ident *ident,
			  struct supertype **stp,
			  struct mdinfo **contentp,
			  struct context *c,
			  int inargv, int auto_assem)
{
	struct mddev_dev *tmpdev;
	int num_devs;
	struct supertype *st = *stp;
	struct mdinfo *content = NULL;
	int report_mismatch = ((inargv && c->verbose >= 0) || c->verbose > 0);
	struct domainlist *domains = NULL;
	dev_t rdev;

	tmpdev = devlist; num_devs = 0;
	while (tmpdev) {
		if (tmpdev->used)
			tmpdev->used = 2;
		else
			num_devs++;
		tmpdev->disposition = 0;
		tmpdev = tmpdev->next;
	}

	/* first walk the list of devices to find a consistent set
	 * that match the criterea, if that is possible.
	 * We flag the ones we like with 'used'.
	 */
	for (tmpdev = devlist;
	     tmpdev;
	     tmpdev = tmpdev ? tmpdev->next : NULL) {
		char *devname = tmpdev->devname;
		int dfd;
		struct supertype *tst;
		struct dev_policy *pol = NULL;
		int found_container = 0;

		if (tmpdev->used > 1)
			continue;

		if (ident->container) {
			if (ident->container[0] == '/' &&
			    !same_dev(ident->container, devname)) {
				if (report_mismatch)
					pr_err("%s is not the container required (%s)\n",
					       devname, ident->container);
				continue;
			}
		} else if (ident->devices &&
			   !match_oneof(ident->devices, devname)) {
			/* Note that we ignore the "device=" identifier if a
			 * "container=" is given.  Checking both is unnecessarily
			 * complicated.
			 */
			if (report_mismatch)
				pr_err("%s is not one of %s\n", devname, ident->devices);
			continue;
		}

		tst = dup_super(st);

		dfd = dev_open(devname, O_RDONLY);
		if (dfd < 0) {
			if (report_mismatch)
				pr_err("cannot open device %s: %s\n",
				       devname, strerror(errno));
			tmpdev->used = 2;
		} else if (!fstat_is_blkdev(dfd, devname, &rdev)) {
			tmpdev->used = 2;
		} else if (must_be_container(dfd)) {
			if (st) {
				/* already found some components, this cannot
				 * be another one.
				 */
				if (report_mismatch)
					pr_err("%s is a container, but we are looking for components\n",
					       devname);
				tmpdev->used = 2;
			} if (!tst && (tst = super_by_fd(dfd, NULL)) == NULL) {
				if (report_mismatch)
					pr_err("not a recognisable container: %s\n",
					       devname);
				tmpdev->used = 2;
			} else if (!tst->ss->load_container ||
				   tst->ss->load_container(tst, dfd, NULL)) {
				if (report_mismatch)
					pr_err("no correct container type: %s\n",
					       devname);
				tmpdev->used = 2;
			} else if (auto_assem &&
				   !conf_test_metadata(tst->ss->name,
						       (pol = devid_policy(rdev)),
						       tst->ss->match_home(tst, c->homehost) == 1)) {
				if (report_mismatch)
					pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
					       devname, tst->ss->name);
				tmpdev->used = 2;
			} else
				found_container = 1;
		} else {
			if (!tst && (tst = guess_super(dfd)) == NULL) {
				if (report_mismatch)
					pr_err("no recogniseable superblock on %s\n",
					       devname);
				tmpdev->used = 2;
			} else if ((tst->ignore_hw_compat = 0),
				   tst->ss->load_super(tst, dfd,
						       report_mismatch ? devname : NULL)) {
				if (report_mismatch)
					pr_err("no RAID superblock on %s\n",
					       devname);
				tmpdev->used = 2;
			} else if (tst->ss->compare_super == NULL) {
				if (report_mismatch)
					pr_err("Cannot assemble %s metadata on %s\n",
					       tst->ss->name, devname);
				tmpdev->used = 2;
			} else if (auto_assem && st == NULL &&
				   !conf_test_metadata(tst->ss->name,
						       (pol = devid_policy(rdev)),
						       tst->ss->match_home(tst, c->homehost) == 1)) {
				if (report_mismatch)
					pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
					       devname, tst->ss->name);
				tmpdev->used = 2;
			}
		}
		if (dfd >= 0) close(dfd);
		if (tmpdev->used == 2) {
			if (auto_assem || !inargv)
				/* Ignore unrecognised devices during auto-assembly */
				goto loop;
			if (ident->name[0] ||
			    ident->super_minor != UnSet)
				/* Ignore unrecognised device if looking for
				 * specific array */
				goto loop;
			if (ident->uuid_set)
				/* ignore unrecognized device if looking for
				 * specific uuid
				 */
				goto loop;

			pr_err("%s has no superblock - assembly aborted\n",
			       devname);
			if (st)
				st->ss->free_super(st);
			dev_policy_free(pol);
			domain_free(domains);
			if (tst)
				tst->ss->free_super(tst);
			return -1;
		}

		if (found_container) {
			/* tmpdev is a container.  We need to be either
			 * looking for a member, or auto-assembling
			 */
			/* should be safe to try an exclusive open now, we
			 * have rejected anything that some other mdadm might
			 * be looking at
			 */
			dfd = dev_open(devname, O_RDONLY | O_EXCL);
			if (dfd < 0) {
				if (report_mismatch)
					pr_err("%s is busy - skipping\n", devname);
				goto loop;
			}
			close(dfd);

			if (ident->container && ident->container[0] != '/') {
				/* we have a uuid */
				int uuid[4];

				content = *contentp;
				tst->ss->getinfo_super(tst, content, NULL);

				if (!parse_uuid(ident->container, uuid) ||
				    !same_uuid(content->uuid, uuid, tst->ss->swapuuid)) {
					if (report_mismatch)
						pr_err("%s has wrong UUID to be required container\n",
						       devname);
					goto loop;
				}
			}
			/* It is worth looking inside this container.
			 */
			if (c->verbose > 0)
				pr_err("looking in container %s\n",
				       devname);

			for (content = tst->ss->container_content(tst, NULL);
			     content;
			     content = content->next) {

				if (!ident_matches(ident, content, tst,
						   c->homehost, c->require_homehost,
						   c->update,
						   report_mismatch ? devname : NULL))
					/* message already printed */;
				else if (is_member_busy(content->text_version)) {
					if (report_mismatch)
						pr_err("member %s in %s is already assembled\n",
						       content->text_version,
						       devname);
				} else if (content->array.state & (1<<MD_SB_BLOCK_VOLUME)) {
					/* do not assemble arrays with unsupported configurations */
					pr_err("Cannot activate member %s in %s.\n",
					       content->text_version,
					       devname);
				} else
					break;
			}
			if (!content) {
				tmpdev->used = 2;
				goto loop; /* empty container */
			}

			st = tst; tst = NULL;
			if (!auto_assem && inargv && tmpdev->next != NULL) {
				pr_err("%s is a container, but is not only device given: confused and aborting\n",
				       devname);
				st->ss->free_super(st);
				dev_policy_free(pol);
				domain_free(domains);
				return -1;
			}
			if (c->verbose > 0)
				pr_err("found match on member %s in %s\n",
				       content->text_version, devname);

			/* make sure we finished the loop */
			tmpdev = NULL;
			goto loop;
		} else {
			content = *contentp;
			tst->ss->getinfo_super(tst, content, NULL);

			if (!ident_matches(ident, content, tst,
					   c->homehost, c->require_homehost,
					   c->update,
					   report_mismatch ? devname : NULL))
				goto loop;

			if (auto_assem) {
				/* Never auto-assemble things that conflict
				 * with mdadm.conf in some way
				 */
				struct mddev_ident *match;
				int rv = 0;

				match = conf_match(tst, content, devname,
						   report_mismatch ? c->verbose : -1,
						   &rv);
				if (!match && rv == 2)
					goto loop;
				if (match && match->devname &&
				    strcasecmp(match->devname, "<ignore>") == 0) {
					if (report_mismatch)
						pr_err("%s is a member of an explicitly ignored array\n",
						       devname);
					goto loop;
				}
				if (match && !ident_matches(match, content, tst,
							    c->homehost, c->require_homehost,
							    c->update,
							    report_mismatch ? devname : NULL))
					/* Array exists  in mdadm.conf but some
					 * details don't match, so reject it
					 */
					goto loop;
			}

			/* should be safe to try an exclusive open now, we
			 * have rejected anything that some other mdadm might
			 * be looking at
			 */
			dfd = dev_open(devname, O_RDONLY | O_EXCL);
			if (dfd < 0) {
				if (report_mismatch)
					pr_err("%s is busy - skipping\n", devname);
				goto loop;
			}
			close(dfd);

			if (st == NULL)
				st = dup_super(tst);
			if (st->minor_version == -1)
				st->minor_version = tst->minor_version;

			if (memcmp(content->uuid, uuid_zero,
				   sizeof(int[4])) == 0) {
				/* this is a floating spare.  It cannot define
				 * an array unless there are no more arrays of
				 * this type to be found.  It can be included
				 * in an array of this type though.
				 */
				tmpdev->used = 3;
				goto loop;
			}

			if (st->ss != tst->ss ||
			    st->minor_version != tst->minor_version ||
			    st->ss->compare_super(st, tst, 1) != 0) {
				/* Some mismatch. If exactly one array matches this host,
				 * we can resolve on that one.
				 * Or, if we are auto assembling, we just ignore the second
				 * for now.
				 */
				if (auto_assem)
					goto loop;
				if (c->homehost) {
					int first = st->ss->match_home(st, c->homehost);
					int last = tst->ss->match_home(tst, c->homehost);
					if (first != last &&
					    (first == 1 || last == 1)) {
						/* We can do something */
						if (first) {/* just ignore this one */
							if (report_mismatch)
								pr_err("%s misses out due to wrong homehost\n",
								       devname);
							goto loop;
						} else { /* reject all those sofar */
							struct mddev_dev *td;
							if (report_mismatch)
								pr_err("%s overrides previous devices due to good homehost\n",
								       devname);
							for (td=devlist; td != tmpdev; td=td->next)
								if (td->used == 1)
									td->used = 0;
							tmpdev->used = 1;
							goto loop;
						}
					}
				}
				pr_err("superblock on %s doesn't match others - assembly aborted\n",
				       devname);
				tst->ss->free_super(tst);
				st->ss->free_super(st);
				dev_policy_free(pol);
				domain_free(domains);
				return -1;
			}
			tmpdev->used = 1;
		}
	loop:
		/* Collect domain information from members only */
		if (tmpdev && tmpdev->used == 1) {
			if (!pol)
				pol = devid_policy(rdev);
			domain_merge(&domains, pol, tst?tst->ss->name:NULL);
		}
		dev_policy_free(pol);
		pol = NULL;
		if (tst)
			tst->ss->free_super(tst);
	}

	/* Check if we found some imsm spares but no members */
	if ((auto_assem ||
	     (ident->uuid_set &&
	      memcmp(uuid_zero, ident->uuid,sizeof(uuid_zero)) == 0)) &&
	    (!st || !st->sb))
		for (tmpdev = devlist; tmpdev; tmpdev = tmpdev->next) {
			if (tmpdev->used != 3)
				continue;
			tmpdev->used = 1;
			content = *contentp;

			if (!st->sb) {
				/* we need sb from one of the spares */
				int dfd = dev_open(tmpdev->devname, O_RDONLY);
				if (dfd < 0 ||
				    st->ss->load_super(st, dfd, NULL))
					tmpdev->used = 2;
				close_fd(&dfd);
			}
		}

	/* Now reject spares that don't match domains of identified members */
	for (tmpdev = devlist; tmpdev; tmpdev = tmpdev->next) {
		if (tmpdev->used != 3)
			continue;
		if (!stat_is_blkdev(tmpdev->devname, &rdev)) {
			tmpdev->used = 2;
		} else {
			struct dev_policy *pol = devid_policy(rdev);
			int dt = domain_test(domains, pol, NULL);
			if (inargv && dt != 0)
				/* take this spare as domains match
				 * if there are any */
				tmpdev->used = 1;
			else if (!inargv && dt == 1)
				/* device wasn't explicitly listed, so need
				 * explicit domain match - which we have */
				tmpdev->used = 1;
			else
				/* if domains don't match mark as unused */
				tmpdev->used = 0;
			dev_policy_free(pol);
		}
	}
	domain_free(domains);
	*stp = st;
	if (st && st->sb && content == *contentp)
		st->ss->getinfo_super(st, content, NULL);
	*contentp = content;

	return num_devs;
}

struct devs {
	char *devname;
	int uptodate; /* set once we decide that this device is as
		       * recent as everything else in the array.
		       */
	int included; /* set if the device is already in the array
		       * due to a previous '-I'
		       */
	struct mdinfo i;
};

static int load_devices(struct devs *devices, char *devmap,
			struct mddev_ident *ident, struct supertype **stp,
			struct mddev_dev *devlist, struct context *c,
			struct mdinfo *content,
			int mdfd, char *mddev,
			int *most_recentp, int *bestcntp, int **bestp,
			int inargv)
{
	struct mddev_dev *tmpdev;
	int devcnt = 0;
	int nextspare = 0;
	int bitmap_done = 0;
	int most_recent = -1;
	int bestcnt = 0;
	int *best = *bestp;
	struct supertype *st = *stp;

	for (tmpdev = devlist; tmpdev; tmpdev=tmpdev->next) {
		char *devname = tmpdev->devname;
		struct stat stb;
		struct supertype *tst;
		int i;
		int dfd;
		int disk_state;

		if (tmpdev->used != 1)
			continue;
		/* looks like a good enough match to update the super block if needed */
		if (c->update) {
			/* prepare useful information in info structures */
			struct stat stb2;
			int err;
			fstat(mdfd, &stb2);

			if (strcmp(c->update, "uuid") == 0 && !ident->uuid_set)
				random_uuid((__u8 *)ident->uuid);

			if (strcmp(c->update, "ppl") == 0 &&
			    ident->bitmap_fd >= 0) {
				pr_err("PPL is not compatible with bitmap\n");
				close(mdfd);
				free(devices);
				free(devmap);
				return -1;
			}

			dfd = dev_open(devname,
				       tmpdev->disposition == 'I'
				       ? O_RDWR : (O_RDWR|O_EXCL));

			tst = dup_super(st);
			if (dfd < 0 || tst->ss->load_super(tst, dfd, NULL) != 0) {
				pr_err("cannot re-read metadata from %s - aborting\n",
				       devname);
				if (dfd >= 0)
					close(dfd);
				close(mdfd);
				free(devices);
				free(devmap);
				tst->ss->free_super(tst);
				free(tst);
				*stp = st;
				return -1;
			}
			tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);

			memcpy(content->uuid, ident->uuid, 16);
			strcpy(content->name, ident->name);
			content->array.md_minor = minor(stb2.st_rdev);

			if (strcmp(c->update, "byteorder") == 0)
				err = 0;
			else if (strcmp(c->update, "home-cluster") == 0) {
				tst->cluster_name = c->homecluster;
				err = tst->ss->write_bitmap(tst, dfd, NameUpdate);
			} else if (strcmp(c->update, "nodes") == 0) {
				tst->nodes = c->nodes;
				err = tst->ss->write_bitmap(tst, dfd, NodeNumUpdate);
			} else if (strcmp(c->update, "revert-reshape") == 0 &&
				   c->invalid_backup)
				err = tst->ss->update_super(tst, content,
							    "revert-reshape-nobackup",
							    devname, c->verbose,
							    ident->uuid_set,
							    c->homehost);
			else
				err = tst->ss->update_super(tst, content, c->update,
							    devname, c->verbose,
							    ident->uuid_set,
							    c->homehost);
			if (err < 0) {
				if (err == -1)
					pr_err("--update=%s not understood for %s metadata\n",
					       c->update, tst->ss->name);
				tst->ss->free_super(tst);
				free(tst);
				close(mdfd);
				close(dfd);
				free(devices);
				free(devmap);
				*stp = st;
				return -1;
			}
			if (strcmp(c->update, "uuid")==0 &&
			    !ident->uuid_set) {
				ident->uuid_set = 1;
				memcpy(ident->uuid, content->uuid, 16);
			}
			if (tst->ss->store_super(tst, dfd))
				pr_err("Could not re-write superblock on %s.\n",
				       devname);

			if (strcmp(c->update, "uuid")==0 &&
			    ident->bitmap_fd >= 0 && !bitmap_done) {
				if (bitmap_update_uuid(ident->bitmap_fd,
						       content->uuid,
						       tst->ss->swapuuid) != 0)
					pr_err("Could not update uuid on external bitmap.\n");
				else
					bitmap_done = 1;
			}
		} else {
			dfd = dev_open(devname,
				       tmpdev->disposition == 'I'
				       ? O_RDWR : (O_RDWR|O_EXCL));
			tst = dup_super(st);

			if (dfd < 0 || tst->ss->load_super(tst, dfd, NULL) != 0) {
				pr_err("cannot re-read metadata from %s - aborting\n",
				       devname);
				if (dfd >= 0)
					close(dfd);
				close(mdfd);
				free(devices);
				free(devmap);
				tst->ss->free_super(tst);
				free(tst);
				*stp = st;
				return -1;
			}
			tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);
		}

		fstat(dfd, &stb);
		close(dfd);

		if (c->verbose > 0)
			pr_err("%s is identified as a member of %s, slot %d%s.\n",
			       devname, mddev, content->disk.raid_disk,
			       (content->disk.state & (1<<MD_DISK_REPLACEMENT)) ? " replacement":"");
		devices[devcnt].devname = devname;
		devices[devcnt].uptodate = 0;
		devices[devcnt].included = (tmpdev->disposition == 'I');
		devices[devcnt].i = *content;
		devices[devcnt].i.disk.major = major(stb.st_rdev);
		devices[devcnt].i.disk.minor = minor(stb.st_rdev);

		disk_state = devices[devcnt].i.disk.state & ~((1<<MD_DISK_FAILFAST) |
							      (1<<MD_DISK_WRITEMOSTLY));
		if (disk_state == ((1<<MD_DISK_ACTIVE) | (1<<MD_DISK_SYNC))) {
			if (most_recent < 0 ||
			    devices[devcnt].i.events
			    > devices[most_recent].i.events) {
				struct supertype *tmp = tst;
				tst = st;
				st = tmp;
				most_recent = devcnt;
			}
		}
		tst->ss->free_super(tst);
		free(tst);

		if (content->array.level == LEVEL_MULTIPATH)
			/* with multipath, the raid_disk from the superblock is meaningless */
			i = devcnt;
		else
			i = devices[devcnt].i.disk.raid_disk;
		if (i+1 == 0 || i == MD_DISK_ROLE_JOURNAL) {
			if (nextspare < content->array.raid_disks*2)
				nextspare = content->array.raid_disks*2;
			i = nextspare++;
		} else {
			/* i is raid_disk - double it so there is room for
			 * replacements */
			i *= 2;
			if (devices[devcnt].i.disk.state & (1<<MD_DISK_REPLACEMENT))
				i++;
			if (i >= content->array.raid_disks*2 &&
			    i >= nextspare)
				nextspare = i+1;
		}
		if (i < 10000) {
			if (i >= bestcnt) {
				int newbestcnt = i+10;
				int *newbest = xmalloc(sizeof(int)*newbestcnt);
				int c;
				for (c=0; c < newbestcnt; c++)
					if (c < bestcnt)
						newbest[c] = best[c];
					else
						newbest[c] = -1;
				if (best)free(best);
				best = newbest;
				bestcnt = newbestcnt;
			}
			if (best[i] >=0 &&
			    devices[best[i]].i.events ==
			    devices[devcnt].i.events &&
			    (devices[best[i]].i.disk.minor !=
			     devices[devcnt].i.disk.minor) &&
			    st->ss == &super0 &&
			    content->array.level != LEVEL_MULTIPATH) {
				/* two different devices with identical superblock.
				 * Could be a mis-detection caused by overlapping
				 * partitions.  fail-safe.
				 */
				pr_err("WARNING %s and %s appear to have very similar superblocks.\n"
				       "      If they are really different, please --zero the superblock on one\n"
				       "      If they are the same or overlap, please remove one from %s.\n",
				       devices[best[i]].devname, devname,
				       inargv ? "the list" :
				       "the\n      DEVICE list in mdadm.conf"
					);
				close(mdfd);
				free(devices);
				free(devmap);
				*stp = st;
				return -1;
			}
			if (best[i] == -1 || (devices[best[i]].i.events
					      < devices[devcnt].i.events))
				best[i] = devcnt;
			else if (st->ss == &super_imsm)
				best[i+1] = devcnt;
		}
		devcnt++;
	}
	if (most_recent >= 0)
		*most_recentp = most_recent;
	*bestcntp = bestcnt;
	*bestp = best;
	*stp = st;
	return devcnt;
}

static int force_array(struct mdinfo *content,
		       struct devs *devices,
		       int *best, int bestcnt, char *avail,
		       int most_recent,
		       struct supertype *st,
		       struct context *c)
{
	int okcnt = 0;
	while (!enough(content->array.level, content->array.raid_disks,
		       content->array.layout, 1,
		       avail) ||
	       (content->reshape_active && content->delta_disks > 0 &&
		!enough(content->array.level, (content->array.raid_disks
					       - content->delta_disks),
			content->new_layout, 1, avail))) {
		/* Choose the newest best drive which is
		 * not up-to-date, update the superblock
		 * and add it.
		 */
		int fd;
		struct supertype *tst;
		unsigned long long current_events;
		int chosen_drive = -1;
		int i;

		for (i = 0;
		     i < content->array.raid_disks * 2 && i < bestcnt;
		     i += 2) {
			int j = best[i];
			if (j < 0)
				continue;
			if (devices[j].uptodate)
				continue;
			if (devices[j].i.recovery_start != MaxSector) {
				int delta;
				if (!devices[j].i.reshape_active ||
				    devices[j].i.delta_disks <= 0)
					continue;
				/* When increasing number of devices, an
				 * added device also appears to be
				 * recovering.  It is safe to include it
				 * as long as it won't be a source of
				 * data.
				 * For now, just allow for last data
				 * devices in RAID4 or last devices in RAID4/5/6.
				 */
				delta = devices[j].i.delta_disks;
				if (devices[j].i.array.level >= 4 &&
				    devices[j].i.array.level <= 6 &&
				    i/2 >= content->array.raid_disks - delta)
					/* OK */;
				else if (devices[j].i.array.level == 4 &&
					 i/2 >= content->array.raid_disks - delta - 1)
					/* OK */;
				else
					continue;
			} else if (devices[j].i.reshape_active !=
			    content->reshape_active ||
			    (devices[j].i.reshape_active &&
			    devices[j].i.reshape_progress !=
			    content->reshape_progress))
				/* Here, it may be a source of data. If two
				 * devices claim different progresses, it
				 * means that reshape boundaries differ for
				 * their own devices. Kernel will only treat
				 * the first one as reshape progress and
				 * go on. It may cause disaster, so avoid it.
				 */
				continue;
			if (chosen_drive < 0 ||
			     devices[j].i.events
			    > devices[chosen_drive].i.events)
				chosen_drive = j;
		}
		if (chosen_drive < 0)
			break;
		current_events = devices[chosen_drive].i.events;
	add_another:
		if (c->verbose >= 0)
			pr_err("forcing event count in %s(%d) from %d up to %d\n",
			       devices[chosen_drive].devname,
			       devices[chosen_drive].i.disk.raid_disk,
			       (int)(devices[chosen_drive].i.events),
			       (int)(devices[most_recent].i.events));
		fd = dev_open(devices[chosen_drive].devname,
			      devices[chosen_drive].included ? O_RDWR
			      : (O_RDWR|O_EXCL));
		if (fd < 0) {
			pr_err("Couldn't open %s for write - not updating\n",
			       devices[chosen_drive].devname);
			devices[chosen_drive].i.events = 0;
			continue;
		}
		tst = dup_super(st);
		if (tst->ss->load_super(tst,fd, NULL)) {
			close(fd);
			pr_err("RAID superblock disappeared from %s - not updating.\n",
			       devices[chosen_drive].devname);
			devices[chosen_drive].i.events = 0;
			continue;
		}
		content->events = devices[most_recent].i.events;
		tst->ss->update_super(tst, content, "force-one",
				      devices[chosen_drive].devname, c->verbose,
				      0, NULL);

		if (tst->ss->store_super(tst, fd)) {
			close(fd);
			pr_err("Could not re-write superblock on %s\n",
			       devices[chosen_drive].devname);
			devices[chosen_drive].i.events = 0;
			tst->ss->free_super(tst);
			continue;
		}
		close(fd);
		devices[chosen_drive].i.events = devices[most_recent].i.events;
		devices[chosen_drive].uptodate = 1;
		avail[chosen_drive] = 1;
		okcnt++;
		tst->ss->free_super(tst);
		/* If there are any other drives of the same vintage,
		 * add them in as well.  We can't lose and we might gain
		 */
		for (i = 0;
		     i < content->array.raid_disks * 2 && i < bestcnt ;
		     i += 2) {
			int j = best[i];
			if (j >= 0 &&
			    !devices[j].uptodate &&
			    devices[j].i.recovery_start == MaxSector &&
			    devices[j].i.events == current_events &&
			    ((!devices[j].i.reshape_active &&
			    !content->reshape_active) ||
			    (devices[j].i.reshape_active ==
			    content->reshape_active &&
			    devices[j].i.reshape_progress ==
			    content->reshape_progress))) {
				chosen_drive = j;
				goto add_another;
			}
		}
	}
	return okcnt;
}

static int start_array(int mdfd,
		       char *mddev,
		       struct mdinfo *content,
		       struct supertype *st,
		       struct mddev_ident *ident,
		       int *best, int bestcnt,
		       int chosen_drive,
		       struct devs *devices,
		       unsigned int okcnt,
		       unsigned int sparecnt,
		       unsigned int rebuilding_cnt,
		       unsigned int journalcnt,
		       struct context *c,
		       int clean, char *avail,
		       int start_partial_ok,
		       int err_ok,
		       int was_forced
	)
{
	int rv;
	int i;
	unsigned int req_cnt;

	if (content->journal_device_required && (content->journal_clean == 0)) {
		if (!c->force) {
			pr_err("Not safe to assemble with missing or stale journal device, consider --force.\n");
			return 1;
		}
		pr_err("Journal is missing or stale, starting array read only.\n");
		c->readonly = 1;
	}

	if (content->consistency_policy == CONSISTENCY_POLICY_PPL)
		clean = 1;

	rv = set_array_info(mdfd, st, content);
	if (rv && !err_ok) {
		pr_err("failed to set array info for %s: %s\n",
		       mddev, strerror(errno));
		return 1;
	}
	if (ident->bitmap_fd >= 0) {
		if (ioctl(mdfd, SET_BITMAP_FILE, ident->bitmap_fd) != 0) {
			pr_err("SET_BITMAP_FILE failed.\n");
			return 1;
		}
	} else if (ident->bitmap_file) {
		/* From config file */
		int bmfd = open(ident->bitmap_file, O_RDWR);
		if (bmfd < 0) {
			pr_err("Could not open bitmap file %s\n",
			       ident->bitmap_file);
			return 1;
		}
		if (ioctl(mdfd, SET_BITMAP_FILE, bmfd) != 0) {
			pr_err("Failed to set bitmapfile for %s\n", mddev);
			close(bmfd);
			return 1;
		}
		close(bmfd);
	}

	/* First, add the raid disks, but add the chosen one last */
	for (i = 0; i <= bestcnt; i++) {
		int j;
		if (i < bestcnt) {
			j = best[i];
			if (j == chosen_drive)
				continue;
		} else
			j = chosen_drive;

		if (j >= 0 && !devices[j].included) {
			int dfd;

			dfd = dev_open(devices[j].devname, O_RDWR|O_EXCL);
			if (dfd >= 0) {
				remove_partitions(dfd);
				close(dfd);
			}
			rv = add_disk(mdfd, st, content, &devices[j].i);

			if (rv) {
				pr_err("failed to add %s to %s: %s\n",
				       devices[j].devname, mddev,
				       strerror(errno));
				if (errno == EINVAL && content->array.level == 0 &&
				    content->array.layout != 0) {
					cont_err("Possibly your kernel doesn't support RAID0 layouts.\n");
					cont_err("Please upgrade.\n");
				}
				if (i < content->array.raid_disks * 2 ||
				    i == bestcnt)
					okcnt--;
				else
					sparecnt--;
			} else if (c->verbose > 0) {
				pr_err("added %s to %s as %d%s%s\n",
				       devices[j].devname, mddev,
				       devices[j].i.disk.raid_disk,
				       devices[j].uptodate?"":
				       " (possibly out of date)",
				       (devices[j].i.disk.state &
					(1<<MD_DISK_REPLACEMENT)) ?
				       " replacement":"");
			}
		} else if (j >= 0) {
			if (c->verbose > 0)
				pr_err("%s is already in %s as %d\n",
				       devices[j].devname, mddev,
				       devices[j].i.disk.raid_disk);
		} else if (c->verbose > 0 &&
			   i < content->array.raid_disks * 2 && (i & 1) == 0)
			pr_err("no uptodate device for slot %d of %s\n",
			       i/2, mddev);
	}

	if (content->array.level == LEVEL_CONTAINER) {
		sysfs_rules_apply(mddev, content);
		if (c->verbose >= 0) {
			pr_err("Container %s has been assembled with %d drive%s",
			       mddev, okcnt + sparecnt + journalcnt,
			       okcnt + sparecnt + journalcnt == 1 ? "" : "s");
			if (okcnt < (unsigned)content->array.raid_disks)
				fprintf(stderr, " (out of %d)\n",
					content->array.raid_disks);
			else
				fprintf(stderr, "\n");
		}

		if (st->ss->validate_container) {
			struct mdinfo *devices_list;
			struct mdinfo *info_devices;
			unsigned int count;

			devices_list = NULL;
			info_devices = xmalloc(sizeof(struct mdinfo) *
					       (okcnt + sparecnt));
			for (count = 0; count < okcnt + sparecnt; count++) {
				info_devices[count] = devices[count].i;
				info_devices[count].next = devices_list;
				devices_list = &info_devices[count];
			}
			if (st->ss->validate_container(devices_list))
				pr_err("Mismatch detected!\n");
			free(info_devices);
		}

		st->ss->free_super(st);
		sysfs_uevent(content, "change");
		if (err_ok && okcnt < (unsigned)content->array.raid_disks)
			/* Was partial, is still partial, so signal an error
			 * to ensure we don't retry */
			return 1;
		return 0;
	}

	/* Get number of in-sync devices according to the superblock.
	 * We must have this number to start the array without -s or -R
	 */
	req_cnt = content->array.working_disks;

	if (c->runstop == 1 ||
	    (c->runstop <= 0 &&
	     (enough(content->array.level, content->array.raid_disks,
		     content->array.layout, clean, avail) &&
	       (okcnt + rebuilding_cnt >= req_cnt || start_partial_ok)))) {
		/* This array is good-to-go.
		 * If a reshape is in progress then we might need to
		 * continue monitoring it.  In that case we start
		 * it read-only and let the grow code make it writable.
		 */
		int rv;

		if (content->reshape_active &&
		    !(content->reshape_active & RESHAPE_NO_BACKUP) &&
		    content->delta_disks <= 0) {
			if (!c->backup_file) {
				pr_err("%s: Need a backup file to complete reshape of this array.\n",
				       mddev);
				pr_err("Please provided one with \"--backup-file=...\"\n");
				if (c->update &&
				    strcmp(c->update, "revert-reshape") == 0)
					pr_err("(Don't specify --update=revert-reshape again, that part succeeded.)\n");
				return 1;
			}
			rv = sysfs_set_str(content, NULL,
					   "array_state", "readonly");
			if (rv == 0)
				rv = Grow_continue(mdfd, st, content,
						   c->backup_file, 0,
						   c->freeze_reshape);
		} else if (c->readonly &&
			   sysfs_attribute_available(content, NULL,
						     "array_state")) {
			rv = sysfs_set_str(content, NULL,
					   "array_state", "readonly");
		} else
			rv = ioctl(mdfd, RUN_ARRAY, NULL);
		reopen_mddev(mdfd); /* drop O_EXCL */
		if (rv == 0) {
			sysfs_rules_apply(mddev, content);
			if (c->verbose >= 0) {
				pr_err("%s has been started with %d drive%s",
				       mddev, okcnt, okcnt==1?"":"s");
				if (okcnt < (unsigned)content->array.raid_disks)
					fprintf(stderr, " (out of %d)",
						content->array.raid_disks);
				if (rebuilding_cnt)
					fprintf(stderr, "%s %d rebuilding",
						sparecnt?",":" and",
						rebuilding_cnt);
				if (sparecnt)
					fprintf(stderr, " and %d spare%s",
						sparecnt,
						sparecnt == 1 ? "" : "s");
				if (content->journal_clean)
					fprintf(stderr, " and %d journal",
						journalcnt);
				fprintf(stderr, ".\n");
			}
			if (content->reshape_active &&
			    content->array.level >= 4 &&
			    content->array.level <= 6) {
				/* might need to increase the size
				 * of the stripe cache - default is 256
				 */
				int chunk_size = content->array.chunk_size;

				if (content->reshape_active &&
				    content->new_chunk > chunk_size)
					chunk_size = content->new_chunk;
				if (256 < 4 * ((chunk_size+4065)/4096)) {
					struct mdinfo *sra;

					sra = sysfs_read(mdfd, NULL, 0);
					if (sra)
						sysfs_set_num(sra, NULL,
							      "stripe_cache_size",
							      (4 * chunk_size / 4096) + 1);
					sysfs_free(sra);
				}
			}
			if (okcnt < (unsigned)content->array.raid_disks) {
				/* If any devices did not get added
				 * because the kernel rejected them based
				 * on event count, try adding them
				 * again providing the action policy is
				 * 're-add' or greater.  The bitmap
				 * might allow them to be included, or
				 * they will become spares.
				 */
				for (i = 0; i < bestcnt; i++) {
					int j = best[i];
					if (j >= 0 && !devices[j].uptodate) {
						if (!disk_action_allows(&devices[j].i, st->ss->name, act_re_add))
							continue;
						rv = add_disk(mdfd, st, content,
							      &devices[j].i);
						if (rv == 0 && c->verbose >= 0)
							pr_err("%s has been re-added.\n",
							       devices[j].devname);
					}
				}
			}
			if (content->array.level == 6 &&
			    okcnt + 1 == (unsigned)content->array.raid_disks &&
			    was_forced) {
				struct mdinfo *sra;

				sra = sysfs_read(mdfd, NULL, 0);
				if (sra)
					sysfs_set_str(sra, NULL,
						      "sync_action", "repair");
				sysfs_free(sra);
			}
			return 0;
		}
		pr_err("failed to RUN_ARRAY %s: %s\n", mddev, strerror(errno));
		if (errno == 524 /* ENOTSUP */ &&
		    content->array.level == 0 && content->array.layout == 0)
			cont_err("Please use --update=layout-original or --update=layout-alternate\n");

		if (!enough(content->array.level, content->array.raid_disks,
			    content->array.layout, 1, avail))
			pr_err("Not enough devices to start the array.\n");
		else if (!enough(content->array.level,
				 content->array.raid_disks,
				 content->array.layout, clean, avail))
			pr_err("Not enough devices to start the array while not clean - consider --force.\n");

		return 1;
	}
	if (c->runstop == -1) {
		pr_err("%s assembled from %d drive%s",
		       mddev, okcnt, okcnt == 1 ? "" : "s");
		if (okcnt != (unsigned)content->array.raid_disks)
			fprintf(stderr, " (out of %d)",
				content->array.raid_disks);
		fprintf(stderr, ", but not started.\n");
		return 2;
	}
	if (c->verbose >= -1) {
		pr_err("%s assembled from %d drive%s",
		       mddev, okcnt, okcnt == 1 ? "" : "s");
		if (rebuilding_cnt)
			fprintf(stderr, "%s %d rebuilding",
				sparecnt ? "," : " and", rebuilding_cnt);
		if (sparecnt)
			fprintf(stderr, " and %d spare%s", sparecnt,
				sparecnt == 1 ? "" : "s");
		if (!enough(content->array.level, content->array.raid_disks,
			    content->array.layout, 1, avail))
			fprintf(stderr, " - not enough to start the array.\n");
		else if (!enough(content->array.level,
				 content->array.raid_disks,
				 content->array.layout, clean, avail))
			fprintf(stderr, " - not enough to start the array while not clean - consider --force.\n");
		else {
			if (req_cnt == (unsigned)content->array.raid_disks)
				fprintf(stderr, " - need all %d to start it",
					req_cnt);
			else
				fprintf(stderr, " - need %d to start", req_cnt);
			fprintf(stderr, " (use --run to insist).\n");
		}
	}
	return 1;
}

int Assemble(struct supertype *st, char *mddev,
	     struct mddev_ident *ident,
	     struct mddev_dev *devlist,
	     struct context *c)
{
	/*
	 * The task of Assemble is to find a collection of
	 * devices that should (according to their superblocks)
	 * form an array, and to give this collection to the MD driver.
	 * In Linux-2.4 and later, this involves submitting a
	 * SET_ARRAY_INFO ioctl with no arg - to prepare
	 * the array - and then submit a number of
	 * ADD_NEW_DISK ioctls to add disks into
	 * the array.  Finally RUN_ARRAY might
	 * be submitted to start the array.
	 *
	 * Much of the work of Assemble is in finding and/or
	 * checking the disks to make sure they look right.
	 *
	 * If mddev is not set, then scan must be set and we
	 *  read through the config file for dev+uuid mapping
	 *  We recurse, setting mddev, for each device that
	 *    - isn't running
	 *    - has a valid uuid (or any uuid if !uuidset)
	 *
	 * If mddev is set, we try to determine state of md.
	 *   check version - must be at least 0.90.0
	 *   check kernel version.  must be at least 2.4.
	 *    If not, we can possibly fall back on START_ARRAY
	 *   Try to GET_ARRAY_INFO.
	 *     If possible, give up
	 *     If not, try to STOP_ARRAY just to make sure
	 *
	 * If !uuidset and scan, look in conf-file for uuid
	 *       If not found, give up
	 * If !devlist and scan and uuidset, get list of devs from conf-file
	 *
	 * For each device:
	 *   Check superblock - discard if bad
	 *   Check uuid (set if we don't have one) - discard if no match
	 *   Check superblock similarity if we have a superblock - discard if different
	 *   Record events, devicenum
	 * This should give us a list of devices for the array
	 * We should collect the most recent event number
	 *
	 * Count disks with recent enough event count
	 * While force && !enough disks
	 *    Choose newest rejected disks, update event count
	 *     mark clean and rewrite superblock
	 * If recent kernel:
	 *    SET_ARRAY_INFO
	 *    foreach device with recent events : ADD_NEW_DISK
	 *    if runstop == 1 || "enough" disks and runstop==0 -> RUN_ARRAY
	 * If old kernel:
	 *    Check the device numbers in superblock are right
	 *    update superblock if any changes
	 *    START_ARRAY
	 *
	 */
	int rv = -1;
	int mdfd = -1;
	int clean;
	int auto_assem = (mddev == NULL && !ident->uuid_set &&
			  ident->super_minor == UnSet && ident->name[0] == 0 &&
			  (ident->container == NULL || ident->member == NULL));
	struct devs *devices = NULL;
	char *devmap;
	int *best = NULL; /* indexed by raid_disk */
	int bestcnt = 0;
	int devcnt;
	unsigned int okcnt, sparecnt, rebuilding_cnt, replcnt, journalcnt;
	int journal_clean = 0;
	int i;
	int was_forced = 0;
	int most_recent = 0;
	int chosen_drive;
	int change = 0;
	int inargv = 0;
	int start_partial_ok = (c->runstop >= 0) &&
		(c->force || devlist==NULL || auto_assem);
	int num_devs;
	struct mddev_dev *tmpdev;
	struct mdinfo info;
	struct mdinfo *content = NULL;
	struct mdinfo *pre_exist = NULL;
	char *avail;
	char *name = NULL;
	char chosen_name[1024];
	struct map_ent *map = NULL;
	struct map_ent *mp;

	/*
	 * If any subdevs are listed, then any that don't
	 * match ident are discarded.  Remainder must all match and
	 * become the array.
	 * If no subdevs, then we scan all devices in the config file, but
	 * there must be something in the identity
	 */

	if (!devlist &&
	    ident->uuid_set == 0 &&
	    (ident->super_minor < 0 || ident->super_minor == UnSet) &&
	    ident->name[0] == 0 &&
	    (ident->container == NULL || ident->member == NULL) &&
	    ident->devices == NULL) {
		pr_err("No identity information available for %s - cannot assemble.\n",
		       mddev ? mddev : "further assembly");
		return 1;
	}

	if (devlist == NULL)
		devlist = conf_get_devs();
	else if (mddev)
		inargv = 1;

try_again:
	/* We come back here when doing auto-assembly and attempting some
	 * set of devices failed.  Those are now marked as ->used==2 and
	 * we ignore them and try again
	 */
	if (!st && ident->st)
		st = ident->st;
	if (c->verbose>0)
		pr_err("looking for devices for %s\n",
		       mddev ? mddev : "further assembly");

	content = &info;
	if (st && c->force)
		st->ignore_hw_compat = 1;
	num_devs = select_devices(devlist, ident, &st, &content, c,
				  inargv, auto_assem);
	if (num_devs < 0)
		return 1;

	if (!st || !st->sb || !content)
		return 2;

	/* We have a full set of devices - we now need to find the
	 * array device.
	 * However there is a risk that we are racing with "mdadm -I"
	 * and the array is already partially assembled - we will have
	 * rejected any devices already in this address.
	 * So we take a lock on the map file - to prevent further races -
	 * and look for the uuid in there.  If found and the array is
	 * active, we abort.  If found and the array is not active
	 * we commit to that md device and add all the contained devices
	 * to our list.  We flag them so that we don't try to re-add,
	 * but can remove if they turn out to not be wanted.
	 */
	if (map_lock(&map))
		pr_err("failed to get exclusive lock on mapfile - continue anyway...\n");
	if (c->update && strcmp(c->update,"uuid") == 0)
		mp = NULL;
	else
		mp = map_by_uuid(&map, content->uuid);
	if (mp) {
		struct mdinfo *dv;
		/* array already exists. */
		pre_exist = sysfs_read(-1, mp->devnm, GET_LEVEL|GET_DEVS);
		if (pre_exist->array.level != UnSet) {
			pr_err("Found some drive for an array that is already active: %s\n",
			       mp->path);
			pr_err("giving up.\n");
			goto out;
		}
		for (dv = pre_exist->devs; dv; dv = dv->next) {
			/* We want to add this device to our list,
			 * but it could already be there if "mdadm -I"
			 * started *after* we checked for O_EXCL.
			 * If we add it to the top of the list
			 * it will be preferred over later copies.
			 */
			struct mddev_dev *newdev;
			char *devname = map_dev(dv->disk.major,
						dv->disk.minor,
						0);
			if (!devname)
				continue;
			newdev = xmalloc(sizeof(*newdev));
			newdev->devname = devname;
			newdev->disposition = 'I';
			newdev->used = 1;
			newdev->next = devlist;
			devlist = newdev;
			num_devs++;
		}
		strcpy(chosen_name, mp->path);
		if (c->verbose > 0 || mddev == NULL ||
		    strcmp(mddev, chosen_name) != 0)
			pr_err("Merging with already-assembled %s\n",
			       chosen_name);
		mdfd = open_dev_excl(mp->devnm);
	} else {
		int trustworthy = FOREIGN;
		name = content->name;
		switch (st->ss->match_home(st, c->homehost)
			?: st->ss->match_home(st, "any")) {
		case 1:
			trustworthy = LOCAL;
			name = strchr(content->name, ':');
			if (name)
				name++;
			else
				name = content->name;
			break;
		}
		if (mddev && map_by_name(&map, mddev) != NULL) {
			pr_err("Cannot create device with %s because is in use\n", mddev);
			goto out;
		}
		if (!auto_assem)
			/* If the array is listed in mdadm.conf or on
			 * command line, then we trust the name
			 * even if the array doesn't look local
			 */
			trustworthy = LOCAL;

		if (name[0] == 0 &&
		    content->array.level == LEVEL_CONTAINER) {
			name = content->text_version;
			trustworthy = METADATA;
		}

		if (name[0] && trustworthy != LOCAL &&
		    ! c->require_homehost &&
		    conf_name_is_free(name))
			trustworthy = LOCAL;

		if (trustworthy == LOCAL &&
		    strchr(name, ':'))
			/* Ignore 'host:' prefix of name */
			name = strchr(name, ':')+1;

		mdfd = create_mddev(mddev, name, ident->autof, trustworthy,
				    chosen_name, 0);
	}
	if (mdfd < 0) {
		st->ss->free_super(st);
		if (auto_assem)
			goto try_again;
		goto out;
	}
	mddev = chosen_name;
	if (pre_exist == NULL) {
		if (mddev_busy(fd2devnm(mdfd))) {
			pr_err("%s already active, cannot restart it!\n",
			       mddev);
			for (tmpdev = devlist ;
			     tmpdev && tmpdev->used != 1;
			     tmpdev = tmpdev->next)
				;
			if (tmpdev && auto_assem)
				pr_err("%s needed for %s...\n",
				       mddev, tmpdev->devname);
			close(mdfd);
			mdfd = -3;
			st->ss->free_super(st);
			if (auto_assem)
				goto try_again;
			goto out;
		}
		/* just incase it was started but has no content */
		ioctl(mdfd, STOP_ARRAY, NULL);
	}

	if (content != &info) {
		/* This is a member of a container.  Try starting the array. */
		int err;
		err = assemble_container_content(st, mdfd, content, c,
						 chosen_name, NULL);
		close(mdfd);
		return err;
	}

	/* Ok, no bad inconsistancy, we can try updating etc */
	devices = xcalloc(num_devs, sizeof(*devices));
	devmap = xcalloc(num_devs, content->array.raid_disks);
	devcnt = load_devices(devices, devmap, ident, &st, devlist,
			      c, content, mdfd, mddev,
			      &most_recent, &bestcnt, &best, inargv);
	if (devcnt < 0) {
		mdfd = -3;
		/*
		 * devices is already freed in load_devices, so set devices
		 * to NULL to avoid double free devices.
		 */
		devices = NULL;
		goto out;
	}

	if (devcnt == 0) {
		pr_err("no devices found for %s\n",
		       mddev);
		if (st)
			st->ss->free_super(st);
		free(devmap);
		goto out;
	}

	if (c->update && strcmp(c->update, "byteorder")==0)
		st->minor_version = 90;

	st->ss->getinfo_super(st, content, NULL);
	clean = content->array.state & 1;

	/* now we have some devices that might be suitable.
	 * I wonder how many
	 */
	avail = xcalloc(content->array.raid_disks, 1);
	okcnt = 0;
	replcnt = 0;
	sparecnt=0;
	journalcnt=0;
	rebuilding_cnt=0;
	for (i=0; i< bestcnt; i++) {
		int j = best[i];
		int event_margin = 1; /* always allow a difference of '1'
				       * like the kernel does
				       */
		if (j < 0) continue;
		/* note: we ignore error flags in multipath arrays
		 * as they don't make sense
		 */
		if (content->array.level != LEVEL_MULTIPATH) {
			if (devices[j].i.disk.state & (1<<MD_DISK_JOURNAL)) {
				if (content->journal_device_required)
					journalcnt++;
				else	/* unexpected journal, mark as faulty */
					devices[j].i.disk.state |= (1<<MD_DISK_FAULTY);
			} else if (!(devices[j].i.disk.state & (1<<MD_DISK_ACTIVE))) {
				if (!(devices[j].i.disk.state
				      & (1<<MD_DISK_FAULTY))) {
					devices[j].uptodate = 1;
					sparecnt++;
				}
				continue;
			}
		}
		/* If this device thinks that 'most_recent' has failed, then
		 * we must reject this device.
		 */
		if (j != most_recent && !c->force &&
		    content->array.raid_disks > 0 &&
		    devices[most_recent].i.disk.raid_disk >= 0 &&
		    devmap[j * content->array.raid_disks + devices[most_recent].i.disk.raid_disk] == 0) {
			if (c->verbose > -1)
				pr_err("ignoring %s as it reports %s as failed\n",
				       devices[j].devname, devices[most_recent].devname);
			best[i] = -1;
			continue;
		}
		/* Require event counter to be same as, or just less than,
		 * most recent.  If it is bigger, it must be a stray spare and
		 * should be ignored.
		 */
		if (devices[j].i.events+event_margin >=
		    devices[most_recent].i.events &&
		    devices[j].i.events <=
		    devices[most_recent].i.events
			) {
			devices[j].uptodate = 1;
			if (devices[j].i.disk.state & (1<<MD_DISK_JOURNAL))
				journal_clean = 1;
			if (i < content->array.raid_disks * 2) {
				if (devices[j].i.recovery_start == MaxSector ||
				    (content->reshape_active &&
				     i >= content->array.raid_disks - content->delta_disks)) {
					if (!avail[i/2]) {
						okcnt++;
						avail[i/2]=1;
					} else
						replcnt++;
				} else
					rebuilding_cnt++;
			} else if (devices[j].i.disk.raid_disk != MD_DISK_ROLE_JOURNAL)
				sparecnt++;
		}
	}
	free(devmap);
	if (c->force) {
		int force_ok = force_array(content, devices, best, bestcnt,
					   avail, most_recent, st, c);
		okcnt += force_ok;
		if (force_ok)
			was_forced = 1;
	}
	/* Now we want to look at the superblock which the kernel will base things on
	 * and compare the devices that we think are working with the devices that the
	 * superblock thinks are working.
	 * If there are differences and --force is given, then update this chosen
	 * superblock.
	 */
	chosen_drive = -1;
	st->ss->free_super(st);
	for (i=0; chosen_drive < 0 && i<bestcnt; i+=2) {
		int j = best[i];
		int fd;

		if (j<0)
			continue;
		if (!devices[j].uptodate)
			continue;
		if (devices[j].i.events < devices[most_recent].i.events)
			continue;
		chosen_drive = j;
		if ((fd=dev_open(devices[j].devname,
				 devices[j].included ? O_RDONLY
				 : (O_RDONLY|O_EXCL)))< 0) {
			pr_err("Cannot open %s: %s\n",
			       devices[j].devname, strerror(errno));
			goto out;
		}
		if (st->ss->load_super(st,fd, NULL)) {
			close(fd);
			pr_err("RAID superblock has disappeared from %s\n",
			       devices[j].devname);
			goto out;
		}
		close(fd);
	}
	if (st->sb == NULL) {
		pr_err("No suitable drives found for %s\n", mddev);
		goto out;
	}
	st->ss->getinfo_super(st, content, NULL);
	if (sysfs_init(content, mdfd, NULL)) {
		pr_err("Unable to initialize sysfs\n");
		goto out;
	}

	/* after reload context, store journal_clean in context */
	content->journal_clean = journal_clean;
	for (i=0; i<bestcnt; i++) {
		int j = best[i];
		unsigned int desired_state;

		if (j < 0)
			continue;
		if (devices[j].i.disk.raid_disk == MD_DISK_ROLE_JOURNAL)
			desired_state = (1<<MD_DISK_JOURNAL);
		else if (i >= content->array.raid_disks * 2)
			desired_state = 0;
		else if (i & 1)
			desired_state = (1<<MD_DISK_ACTIVE) | (1<<MD_DISK_REPLACEMENT);
		else
			desired_state = (1<<MD_DISK_ACTIVE) | (1<<MD_DISK_SYNC);

		desired_state |= devices[j].i.disk.state & ((1<<MD_DISK_FAILFAST) |
							    (1<<MD_DISK_WRITEMOSTLY));

		if (!devices[j].uptodate)
			continue;

		devices[j].i.disk.state = desired_state;
		if (!(devices[j].i.array.state & 1))
			clean = 0;

		if (st->ss->update_super(st, &devices[j].i, "assemble", NULL,
					 c->verbose, 0, NULL)) {
			if (c->force) {
				if (c->verbose >= 0)
					pr_err("clearing FAULTY flag for device %d in %s for %s\n",
					       j, mddev, devices[j].devname);
				change = 1;
			} else {
				if (c->verbose >= -1)
					pr_err("device %d in %s has wrong state in superblock, but %s seems ok\n",
					       i, mddev, devices[j].devname);
			}
		}
#if 0
		if (!(super.disks[i].i.disk.state & (1 << MD_DISK_FAULTY))) {
			pr_err("devices %d of %s is not marked FAULTY in superblock, but cannot be found\n",
			       i, mddev);
		}
#endif
	}
	if (c->force && !clean &&
	    !enough(content->array.level, content->array.raid_disks,
		    content->array.layout, clean,
		    avail)) {
		change += st->ss->update_super(st, content, "force-array",
					       devices[chosen_drive].devname, c->verbose,
					       0, NULL);
		was_forced = 1;
		clean = 1;
	}

	if (change) {
		int fd;
		fd = dev_open(devices[chosen_drive].devname,
			      devices[chosen_drive].included ?
			      O_RDWR : (O_RDWR|O_EXCL));
		if (fd < 0) {
			pr_err("Could not open %s for write - cannot Assemble array.\n",
			       devices[chosen_drive].devname);
			goto out;
		}
		if (st->ss->store_super(st, fd)) {
			close(fd);
			pr_err("Could not re-write superblock on %s\n",
			       devices[chosen_drive].devname);
			goto out;
		}
		if (c->verbose >= 0)
			pr_err("Marking array %s as 'clean'\n",
			       mddev);
		close(fd);
	}

	/* If we are in the middle of a reshape we may need to restore saved data
	 * that was moved aside due to the reshape overwriting live data
	 * The code of doing this lives in Grow.c
	 */
	if (content->reshape_active &&
	    !(content->reshape_active & RESHAPE_NO_BACKUP)) {
		int err = 0;
		int *fdlist = xmalloc(sizeof(int)* bestcnt);
		if (c->verbose > 0)
			pr_err("%s has an active reshape - checking if critical section needs to be restored\n",
			       chosen_name);
		if (!c->backup_file)
			c->backup_file = locate_backup(content->sys_name);
		enable_fds(bestcnt/2);
		for (i = 0; i < bestcnt/2; i++) {
			int j = best[i*2];
			if (j >= 0) {
				fdlist[i] = dev_open(devices[j].devname,
						     devices[j].included
						     ? O_RDWR : (O_RDWR|O_EXCL));
				if (fdlist[i] < 0) {
					pr_err("Could not open %s for write - cannot Assemble array.\n",
					       devices[j].devname);
					err = 1;
					break;
				}
			} else
				fdlist[i] = -1;
		}
		if (!err) {
			if (st->ss->external && st->ss->recover_backup)
				err = st->ss->recover_backup(st, content);
			else
				err = Grow_restart(st, content, fdlist, bestcnt/2,
						   c->backup_file, c->verbose > 0);
			if (err && c->invalid_backup) {
				if (c->verbose > 0)
					pr_err("continuing without restoring backup\n");
				err = 0;
			}
		}
		while (i>0) {
			i--;
			if (fdlist[i]>=0) close(fdlist[i]);
		}
		free(fdlist);
		if (err) {
			pr_err("Failed to restore critical section for reshape, sorry.\n");
			if (c->backup_file == NULL)
				cont_err("Possibly you needed to specify the --backup-file\n");
			goto out;
		}
	}

	/* Almost ready to actually *do* something */
	/* First, fill in the map, so that udev can find our name
	 * as soon as we become active.
	 */
	if (c->update && strcmp(c->update, "metadata")==0) {
		content->array.major_version = 1;
		content->array.minor_version = 0;
		strcpy(content->text_version, "1.0");
	}

	map_update(&map, fd2devnm(mdfd), content->text_version,
		   content->uuid, chosen_name);

	rv = start_array(mdfd, mddev, content,
			 st, ident, best, bestcnt,
			 chosen_drive, devices, okcnt, sparecnt,
			 rebuilding_cnt, journalcnt,
			 c,
			 clean, avail, start_partial_ok,
			 pre_exist != NULL,
			 was_forced);
	if (rv == 1 && !pre_exist)
		ioctl(mdfd, STOP_ARRAY, NULL);
	free(devices);
out:
	map_unlock(&map);
	if (rv == 0) {
		wait_for(chosen_name, mdfd);
		close(mdfd);
		if (auto_assem) {
			int usecs = 1;
			/* There is a nasty race with 'mdadm --monitor'.
			 * If it opens this device before we close it,
			 * it gets an incomplete open on which IO
			 * doesn't work and the capacity is
			 * wrong.
			 * If we reopen (to check for layered devices)
			 * before --monitor closes, we loose.
			 *
			 * So: wait upto 1 second for there to be
			 * a non-zero capacity.
			 */
			while (usecs < 1000) {
				mdfd = open(mddev, O_RDONLY);
				if (mdfd >= 0) {
					unsigned long long size;
					if (get_dev_size(mdfd, NULL, &size) &&
					    size > 0)
						break;
					close(mdfd);
				}
				usleep(usecs);
				usecs <<= 1;
			}
		}
	} else if (mdfd >= 0)
		close(mdfd);

	/* '2' means 'OK, but not started yet' */
	if (rv == -1) {
		free(devices);
		return 1;
	}
	return rv == 2 ? 0 : rv;
}

int assemble_container_content(struct supertype *st, int mdfd,
			       struct mdinfo *content, struct context *c,
			       char *chosen_name, int *result)
{
	struct mdinfo *dev, *sra, *dev2;
	struct assembly_array_info array = {chosen_name, 0, 0, 0};
	int old_raid_disks;
	int start_reshape;
	char *avail;
	int err;
	int is_raid456, is_clean, all_disks;

	if (sysfs_init(content, mdfd, NULL)) {
		pr_err("Unable to initialize sysfs\n");
		return 1;
	}

	sra = sysfs_read(mdfd, NULL, GET_VERSION|GET_DEVS);
	if (sra == NULL || strcmp(sra->text_version, content->text_version) != 0) {
		if (content->array.major_version == -1 &&
		    content->array.minor_version == -2 &&
		    c->readonly &&
		    content->text_version[0] == '/')
			content->text_version[0] = '-';
		if (sysfs_set_array(content, 9003) != 0) {
			sysfs_free(sra);
			return 1;
		}
	}

	/* There are two types of reshape: container wide or sub-array specific
	 * Check if metadata requests blocking container wide reshapes
	 */
	start_reshape = (content->reshape_active &&
			 !((content->reshape_active == CONTAINER_RESHAPE) &&
			   (content->array.state & (1<<MD_SB_BLOCK_CONTAINER_RESHAPE))));

	/* Block subarray here if it is under reshape now
	 * Do not allow for any changes in this array
	 */
	if (st->ss->external && content->recovery_blocked && start_reshape)
		block_subarray(content);

	for (dev2 = sra->devs; dev2; dev2 = dev2->next) {
		for (dev = content->devs; dev; dev = dev->next)
			if (dev2->disk.major == dev->disk.major &&
			    dev2->disk.minor == dev->disk.minor)
				break;
		if (dev)
			continue;
		/* Don't want this one any more */
		if (sysfs_set_str(sra, dev2, "slot", "none") < 0 &&
		    errno == EBUSY) {
			pr_err("Cannot remove old device %s: not updating %s\n", dev2->sys_name, sra->sys_name);
			sysfs_free(sra);
			return 1;
		}
		sysfs_set_str(sra, dev2, "state", "remove");
	}
	old_raid_disks = content->array.raid_disks - content->delta_disks;
	avail = xcalloc(content->array.raid_disks, 1);
	for (dev = content->devs; dev; dev = dev->next) {
		if (dev->disk.raid_disk >= 0)
			avail[dev->disk.raid_disk] = 1;
		if (sysfs_add_disk(content, dev, 1) == 0) {
			if (dev->disk.raid_disk >= old_raid_disks &&
			    content->reshape_active)
				array.exp_cnt++;
			else
				array.new_cnt++;
		} else if (errno == EEXIST)
			array.preexist_cnt++;
	}
	sysfs_free(sra);

	all_disks = array.new_cnt + array.exp_cnt + array.preexist_cnt;

	map_update(NULL, fd2devnm(mdfd), content->text_version,
		   content->uuid, chosen_name);

	if (content->consistency_policy == CONSISTENCY_POLICY_PPL &&
	    st->ss->validate_ppl) {
		content->array.state |= 1;
		err = 0;

		for (dev = content->devs; dev; dev = dev->next) {
			int dfd;
			char *devpath;
			int ret;

			ret = st->ss->validate_ppl(st, content, dev);
			if (ret == 0)
				continue;

			if (ret < 0) {
				err = 1;
				break;
			}

			if (!c->force) {
				pr_err("%s contains invalid PPL - consider --force or --update-subarray with --update=no-ppl\n",
					chosen_name);
				content->array.state &= ~1;
				avail[dev->disk.raid_disk] = 0;
				break;
			}

			/* have --force - overwrite the invalid ppl */
			devpath = map_dev(dev->disk.major, dev->disk.minor, 0);
			dfd = dev_open(devpath, O_RDWR);
			if (dfd < 0) {
				pr_err("Failed to open %s\n", devpath);
				err = 1;
				break;
			}

			err = st->ss->write_init_ppl(st, content, dfd);
			close(dfd);

			if (err)
				break;
		}

		if (err) {
			free(avail);
			return err;
		}
	} else if (c->force) {
		/* Set the array as 'clean' so that we can proceed with starting
		 * it even if we don't have all devices. Mdmon doesn't care
		 * if the dirty flag is set in metadata, it will start managing
		 * it anyway.
		 * This is really important for raid456 (RWH case), other levels
		 * are started anyway.
		 */
		content->array.state |= 1;
	}

	is_raid456 = (content->array.level >= 4 && content->array.level <= 6);
	is_clean = content->array.state & 1;

	if (enough(content->array.level, content->array.raid_disks,
		   content->array.layout, is_clean, avail) == 0) {
		set_array_assembly_status(c, result, INCR_NO, &array);

		if (c->verbose >= 0 && is_raid456 && !is_clean)
			pr_err("Consider --force to start dirty degraded array\n");

		free(avail);
		return 1;
	}
	free(avail);

	if (c->runstop <= 0 && all_disks < content->array.working_disks) {

		set_array_assembly_status(c, result, INCR_UNSAFE, &array);

		if (c->verbose >= 0 && c->force)
			pr_err("Consider --run to start array as degraded.\n");
		return 1;
	}

	if (is_raid456 && content->resync_start != MaxSector && c->force &&
	    all_disks < content->array.raid_disks) {

		content->resync_start = MaxSector;
		err = sysfs_set_num(content, NULL, "resync_start", MaxSector);
		if (err)
			return 1;

		pr_err("%s array state forced to clean. It may cause data corruption.\n",
		       chosen_name);
	}

	/*
	 * Before activating the array, perform extra steps required
	 * to configure the internal write-intent bitmap.
	 */
	if (content->consistency_policy == CONSISTENCY_POLICY_BITMAP &&
	    st->ss->set_bitmap)
		st->ss->set_bitmap(st, content);

	if (start_reshape) {
		int spare = content->array.raid_disks + array.exp_cnt;
		if (restore_backup(st, content,
				   array.new_cnt,
				   spare, &c->backup_file, c->verbose) == 1)
			return 1;

		if (content->reshape_progress == 0) {
			/* If reshape progress is 0 - we are assembling the
			 * array that was stopped, before reshape has started.
			 * Array needs to be started as active, Grow_continue()
			 * will start the reshape.
			 */
			sysfs_set_num(content, NULL, "reshape_position",
				      MaxSector);
			err = sysfs_set_str(content, NULL,
					    "array_state", "active");
			sysfs_set_num(content, NULL, "reshape_position", 0);
		} else {
			err = sysfs_set_str(content, NULL,
					    "array_state", "readonly");
		}

		if (err)
			return 1;

		if (st->ss->external) {
			if (!mdmon_running(st->container_devnm))
				start_mdmon(st->container_devnm);
			ping_monitor(st->container_devnm);
			if (mdmon_running(st->container_devnm) &&
			    st->update_tail == NULL)
				st->update_tail = &st->updates;
		}

		err = Grow_continue(mdfd, st, content, c->backup_file,
				    0, c->freeze_reshape);
	} else switch(content->array.level) {
		case LEVEL_LINEAR:
		case LEVEL_MULTIPATH:
		case 0:
			err = sysfs_set_str(content, NULL, "array_state",
					    c->readonly ? "readonly" : "active");
			break;
		default:
			err = sysfs_set_str(content, NULL, "array_state",
					    "readonly");
			/* start mdmon if needed. */
			if (!err) {
				if (!mdmon_running(st->container_devnm))
					start_mdmon(st->container_devnm);
				ping_monitor(st->container_devnm);
			}
			break;
		}
	if (!err)
		sysfs_set_safemode(content, content->safe_mode_delay);

	/* Block subarray here if it is not reshaped now
	 * It has be blocked a little later to allow mdmon to switch in
	 * in to R/W state
	 */
	if (st->ss->external && content->recovery_blocked &&
	    !start_reshape)
		block_subarray(content);

	if (err)
		set_array_assembly_status(c, result, INCR_NO, &array);
	else {
		set_array_assembly_status(c, result, INCR_YES, &array);
		wait_for(chosen_name, mdfd);
		sysfs_rules_apply(chosen_name, content);
	}

	return err;
	/* FIXME should have an O_EXCL and wait for read-auto */
}