summaryrefslogtreecommitdiffstats
path: root/fs/ocfs2/dlmglue.c
blob: e6220137bf691600be76ac97d86868347269d4d1 (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
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
/* -*- mode: c; c-basic-offset: 8; -*-
 * vim: noexpandtab sw=8 ts=8 sts=0:
 *
 * dlmglue.c
 *
 * Code which implements an OCFS2 specific interface to our DLM.
 *
 * Copyright (C) 2003, 2004 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License 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 021110-1307, USA.
 */

#include <linux/types.h>
#include <linux/slab.h>
#include <linux/highmem.h>
#include <linux/mm.h>
#include <linux/smp_lock.h>
#include <linux/crc32.h>
#include <linux/kthread.h>
#include <linux/pagemap.h>
#include <linux/debugfs.h>
#include <linux/seq_file.h>

#include <cluster/heartbeat.h>
#include <cluster/nodemanager.h>
#include <cluster/tcp.h>

#include <dlm/dlmapi.h>

#define MLOG_MASK_PREFIX ML_DLM_GLUE
#include <cluster/masklog.h>

#include "ocfs2.h"

#include "alloc.h"
#include "dcache.h"
#include "dlmglue.h"
#include "extent_map.h"
#include "file.h"
#include "heartbeat.h"
#include "inode.h"
#include "journal.h"
#include "slot_map.h"
#include "super.h"
#include "uptodate.h"
#include "vote.h"

#include "buffer_head_io.h"

struct ocfs2_mask_waiter {
	struct list_head	mw_item;
	int			mw_status;
	struct completion	mw_complete;
	unsigned long		mw_mask;
	unsigned long		mw_goal;
};

static struct ocfs2_super *ocfs2_get_dentry_osb(struct ocfs2_lock_res *lockres);
static struct ocfs2_super *ocfs2_get_inode_osb(struct ocfs2_lock_res *lockres);

/*
 * Return value from ->downconvert_worker functions.
 *
 * These control the precise actions of ocfs2_unblock_lock()
 * and ocfs2_process_blocked_lock()
 *
 */
enum ocfs2_unblock_action {
	UNBLOCK_CONTINUE	= 0, /* Continue downconvert */
	UNBLOCK_CONTINUE_POST	= 1, /* Continue downconvert, fire
				      * ->post_unlock callback */
	UNBLOCK_STOP_POST	= 2, /* Do not downconvert, fire
				      * ->post_unlock() callback. */
};

struct ocfs2_unblock_ctl {
	int requeue;
	enum ocfs2_unblock_action unblock_action;
};

static int ocfs2_check_meta_downconvert(struct ocfs2_lock_res *lockres,
					int new_level);
static void ocfs2_set_meta_lvb(struct ocfs2_lock_res *lockres);

static int ocfs2_data_convert_worker(struct ocfs2_lock_res *lockres,
				     int blocking);

static int ocfs2_dentry_convert_worker(struct ocfs2_lock_res *lockres,
				       int blocking);

static void ocfs2_dentry_post_unlock(struct ocfs2_super *osb,
				     struct ocfs2_lock_res *lockres);

/*
 * OCFS2 Lock Resource Operations
 *
 * These fine tune the behavior of the generic dlmglue locking infrastructure.
 *
 * The most basic of lock types can point ->l_priv to their respective
 * struct ocfs2_super and allow the default actions to manage things.
 *
 * Right now, each lock type also needs to implement an init function,
 * and trivial lock/unlock wrappers. ocfs2_simple_drop_lockres()
 * should be called when the lock is no longer needed (i.e., object
 * destruction time).
 */
struct ocfs2_lock_res_ops {
	/*
	 * Translate an ocfs2_lock_res * into an ocfs2_super *. Define
	 * this callback if ->l_priv is not an ocfs2_super pointer
	 */
	struct ocfs2_super * (*get_osb)(struct ocfs2_lock_res *);

	/*
	 * Optionally called in the downconvert (or "vote") thread
	 * after a successful downconvert. The lockres will not be
	 * referenced after this callback is called, so it is safe to
	 * free memory, etc.
	 *
	 * The exact semantics of when this is called are controlled
	 * by ->downconvert_worker()
	 */
	void (*post_unlock)(struct ocfs2_super *, struct ocfs2_lock_res *);

	/*
	 * Allow a lock type to add checks to determine whether it is
	 * safe to downconvert a lock. Return 0 to re-queue the
	 * downconvert at a later time, nonzero to continue.
	 *
	 * For most locks, the default checks that there are no
	 * incompatible holders are sufficient.
	 *
	 * Called with the lockres spinlock held.
	 */
	int (*check_downconvert)(struct ocfs2_lock_res *, int);

	/*
	 * Allows a lock type to populate the lock value block. This
	 * is called on downconvert, and when we drop a lock.
	 *
	 * Locks that want to use this should set LOCK_TYPE_USES_LVB
	 * in the flags field.
	 *
	 * Called with the lockres spinlock held.
	 */
	void (*set_lvb)(struct ocfs2_lock_res *);

	/*
	 * Called from the downconvert thread when it is determined
	 * that a lock will be downconverted. This is called without
	 * any locks held so the function can do work that might
	 * schedule (syncing out data, etc).
	 *
	 * This should return any one of the ocfs2_unblock_action
	 * values, depending on what it wants the thread to do.
	 */
	int (*downconvert_worker)(struct ocfs2_lock_res *, int);

	/*
	 * LOCK_TYPE_* flags which describe the specific requirements
	 * of a lock type. Descriptions of each individual flag follow.
	 */
	int flags;
};

/*
 * Some locks want to "refresh" potentially stale data when a
 * meaningful (PRMODE or EXMODE) lock level is first obtained. If this
 * flag is set, the OCFS2_LOCK_NEEDS_REFRESH flag will be set on the
 * individual lockres l_flags member from the ast function. It is
 * expected that the locking wrapper will clear the
 * OCFS2_LOCK_NEEDS_REFRESH flag when done.
 */
#define LOCK_TYPE_REQUIRES_REFRESH 0x1

/*
 * Indicate that a lock type makes use of the lock value block. The
 * ->set_lvb lock type callback must be defined.
 */
#define LOCK_TYPE_USES_LVB		0x2

static struct ocfs2_lock_res_ops ocfs2_inode_rw_lops = {
	.get_osb	= ocfs2_get_inode_osb,
	.flags		= 0,
};

static struct ocfs2_lock_res_ops ocfs2_inode_meta_lops = {
	.get_osb	= ocfs2_get_inode_osb,
	.check_downconvert = ocfs2_check_meta_downconvert,
	.set_lvb	= ocfs2_set_meta_lvb,
	.flags		= LOCK_TYPE_REQUIRES_REFRESH|LOCK_TYPE_USES_LVB,
};

static struct ocfs2_lock_res_ops ocfs2_inode_data_lops = {
	.get_osb	= ocfs2_get_inode_osb,
	.downconvert_worker = ocfs2_data_convert_worker,
	.flags		= 0,
};

static struct ocfs2_lock_res_ops ocfs2_super_lops = {
	.flags		= LOCK_TYPE_REQUIRES_REFRESH,
};

static struct ocfs2_lock_res_ops ocfs2_rename_lops = {
	.flags		= 0,
};

static struct ocfs2_lock_res_ops ocfs2_dentry_lops = {
	.get_osb	= ocfs2_get_dentry_osb,
	.post_unlock	= ocfs2_dentry_post_unlock,
	.downconvert_worker = ocfs2_dentry_convert_worker,
	.flags		= 0,
};

static inline int ocfs2_is_inode_lock(struct ocfs2_lock_res *lockres)
{
	return lockres->l_type == OCFS2_LOCK_TYPE_META ||
		lockres->l_type == OCFS2_LOCK_TYPE_DATA ||
		lockres->l_type == OCFS2_LOCK_TYPE_RW;
}

static inline struct inode *ocfs2_lock_res_inode(struct ocfs2_lock_res *lockres)
{
	BUG_ON(!ocfs2_is_inode_lock(lockres));

	return (struct inode *) lockres->l_priv;
}

static inline struct ocfs2_dentry_lock *ocfs2_lock_res_dl(struct ocfs2_lock_res *lockres)
{
	BUG_ON(lockres->l_type != OCFS2_LOCK_TYPE_DENTRY);

	return (struct ocfs2_dentry_lock *)lockres->l_priv;
}

static inline struct ocfs2_super *ocfs2_get_lockres_osb(struct ocfs2_lock_res *lockres)
{
	if (lockres->l_ops->get_osb)
		return lockres->l_ops->get_osb(lockres);

	return (struct ocfs2_super *)lockres->l_priv;
}

static int ocfs2_lock_create(struct ocfs2_super *osb,
			     struct ocfs2_lock_res *lockres,
			     int level,
			     int dlm_flags);
static inline int ocfs2_may_continue_on_blocked_lock(struct ocfs2_lock_res *lockres,
						     int wanted);
static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
				 struct ocfs2_lock_res *lockres,
				 int level);
static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres);
static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres);
static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres);
static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres, int level);
static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb,
					struct ocfs2_lock_res *lockres);
static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
						int convert);
#define ocfs2_log_dlm_error(_func, _stat, _lockres) do {	\
	mlog(ML_ERROR, "Dlm error \"%s\" while calling %s on "	\
		"resource %s: %s\n", dlm_errname(_stat), _func,	\
		_lockres->l_name, dlm_errmsg(_stat));		\
} while (0)
static void ocfs2_vote_on_unlock(struct ocfs2_super *osb,
				 struct ocfs2_lock_res *lockres);
static int ocfs2_meta_lock_update(struct inode *inode,
				  struct buffer_head **bh);
static void ocfs2_drop_osb_locks(struct ocfs2_super *osb);
static inline int ocfs2_highest_compat_lock_level(int level);

static void ocfs2_build_lock_name(enum ocfs2_lock_type type,
				  u64 blkno,
				  u32 generation,
				  char *name)
{
	int len;

	mlog_entry_void();

	BUG_ON(type >= OCFS2_NUM_LOCK_TYPES);

	len = snprintf(name, OCFS2_LOCK_ID_MAX_LEN, "%c%s%016llx%08x",
		       ocfs2_lock_type_char(type), OCFS2_LOCK_ID_PAD,
		       (long long)blkno, generation);

	BUG_ON(len != (OCFS2_LOCK_ID_MAX_LEN - 1));

	mlog(0, "built lock resource with name: %s\n", name);

	mlog_exit_void();
}

static DEFINE_SPINLOCK(ocfs2_dlm_tracking_lock);

static void ocfs2_add_lockres_tracking(struct ocfs2_lock_res *res,
				       struct ocfs2_dlm_debug *dlm_debug)
{
	mlog(0, "Add tracking for lockres %s\n", res->l_name);

	spin_lock(&ocfs2_dlm_tracking_lock);
	list_add(&res->l_debug_list, &dlm_debug->d_lockres_tracking);
	spin_unlock(&ocfs2_dlm_tracking_lock);
}

static void ocfs2_remove_lockres_tracking(struct ocfs2_lock_res *res)
{
	spin_lock(&ocfs2_dlm_tracking_lock);
	if (!list_empty(&res->l_debug_list))
		list_del_init(&res->l_debug_list);
	spin_unlock(&ocfs2_dlm_tracking_lock);
}

static void ocfs2_lock_res_init_common(struct ocfs2_super *osb,
				       struct ocfs2_lock_res *res,
				       enum ocfs2_lock_type type,
				       struct ocfs2_lock_res_ops *ops,
				       void *priv)
{
	res->l_type          = type;
	res->l_ops           = ops;
	res->l_priv          = priv;

	res->l_level         = LKM_IVMODE;
	res->l_requested     = LKM_IVMODE;
	res->l_blocking      = LKM_IVMODE;
	res->l_action        = OCFS2_AST_INVALID;
	res->l_unlock_action = OCFS2_UNLOCK_INVALID;

	res->l_flags         = OCFS2_LOCK_INITIALIZED;

	ocfs2_add_lockres_tracking(res, osb->osb_dlm_debug);
}

void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res)
{
	/* This also clears out the lock status block */
	memset(res, 0, sizeof(struct ocfs2_lock_res));
	spin_lock_init(&res->l_lock);
	init_waitqueue_head(&res->l_event);
	INIT_LIST_HEAD(&res->l_blocked_list);
	INIT_LIST_HEAD(&res->l_mask_waiters);
}

void ocfs2_inode_lock_res_init(struct ocfs2_lock_res *res,
			       enum ocfs2_lock_type type,
			       unsigned int generation,
			       struct inode *inode)
{
	struct ocfs2_lock_res_ops *ops;

	switch(type) {
		case OCFS2_LOCK_TYPE_RW:
			ops = &ocfs2_inode_rw_lops;
			break;
		case OCFS2_LOCK_TYPE_META:
			ops = &ocfs2_inode_meta_lops;
			break;
		case OCFS2_LOCK_TYPE_DATA:
			ops = &ocfs2_inode_data_lops;
			break;
		default:
			mlog_bug_on_msg(1, "type: %d\n", type);
			ops = NULL; /* thanks, gcc */
			break;
	};

	ocfs2_build_lock_name(type, OCFS2_I(inode)->ip_blkno,
			      generation, res->l_name);
	ocfs2_lock_res_init_common(OCFS2_SB(inode->i_sb), res, type, ops, inode);
}

static struct ocfs2_super *ocfs2_get_inode_osb(struct ocfs2_lock_res *lockres)
{
	struct inode *inode = ocfs2_lock_res_inode(lockres);

	return OCFS2_SB(inode->i_sb);
}

static __u64 ocfs2_get_dentry_lock_ino(struct ocfs2_lock_res *lockres)
{
	__be64 inode_blkno_be;

	memcpy(&inode_blkno_be, &lockres->l_name[OCFS2_DENTRY_LOCK_INO_START],
	       sizeof(__be64));

	return be64_to_cpu(inode_blkno_be);
}

static struct ocfs2_super *ocfs2_get_dentry_osb(struct ocfs2_lock_res *lockres)
{
	struct ocfs2_dentry_lock *dl = lockres->l_priv;

	return OCFS2_SB(dl->dl_inode->i_sb);
}

void ocfs2_dentry_lock_res_init(struct ocfs2_dentry_lock *dl,
				u64 parent, struct inode *inode)
{
	int len;
	u64 inode_blkno = OCFS2_I(inode)->ip_blkno;
	__be64 inode_blkno_be = cpu_to_be64(inode_blkno);
	struct ocfs2_lock_res *lockres = &dl->dl_lockres;

	ocfs2_lock_res_init_once(lockres);

	/*
	 * Unfortunately, the standard lock naming scheme won't work
	 * here because we have two 16 byte values to use. Instead,
	 * we'll stuff the inode number as a binary value. We still
	 * want error prints to show something without garbling the
	 * display, so drop a null byte in there before the inode
	 * number. A future version of OCFS2 will likely use all
	 * binary lock names. The stringified names have been a
	 * tremendous aid in debugging, but now that the debugfs
	 * interface exists, we can mangle things there if need be.
	 *
	 * NOTE: We also drop the standard "pad" value (the total lock
	 * name size stays the same though - the last part is all
	 * zeros due to the memset in ocfs2_lock_res_init_once()
	 */
	len = snprintf(lockres->l_name, OCFS2_DENTRY_LOCK_INO_START,
		       "%c%016llx",
		       ocfs2_lock_type_char(OCFS2_LOCK_TYPE_DENTRY),
		       (long long)parent);

	BUG_ON(len != (OCFS2_DENTRY_LOCK_INO_START - 1));

	memcpy(&lockres->l_name[OCFS2_DENTRY_LOCK_INO_START], &inode_blkno_be,
	       sizeof(__be64));

	ocfs2_lock_res_init_common(OCFS2_SB(inode->i_sb), lockres,
				   OCFS2_LOCK_TYPE_DENTRY, &ocfs2_dentry_lops,
				   dl);
}

static void ocfs2_super_lock_res_init(struct ocfs2_lock_res *res,
				      struct ocfs2_super *osb)
{
	/* Superblock lockres doesn't come from a slab so we call init
	 * once on it manually.  */
	ocfs2_lock_res_init_once(res);
	ocfs2_build_lock_name(OCFS2_LOCK_TYPE_SUPER, OCFS2_SUPER_BLOCK_BLKNO,
			      0, res->l_name);
	ocfs2_lock_res_init_common(osb, res, OCFS2_LOCK_TYPE_SUPER,
				   &ocfs2_super_lops, osb);
}

static void ocfs2_rename_lock_res_init(struct ocfs2_lock_res *res,
				       struct ocfs2_super *osb)
{
	/* Rename lockres doesn't come from a slab so we call init
	 * once on it manually.  */
	ocfs2_lock_res_init_once(res);
	ocfs2_build_lock_name(OCFS2_LOCK_TYPE_RENAME, 0, 0, res->l_name);
	ocfs2_lock_res_init_common(osb, res, OCFS2_LOCK_TYPE_RENAME,
				   &ocfs2_rename_lops, osb);
}

void ocfs2_lock_res_free(struct ocfs2_lock_res *res)
{
	mlog_entry_void();

	if (!(res->l_flags & OCFS2_LOCK_INITIALIZED))
		return;

	ocfs2_remove_lockres_tracking(res);

	mlog_bug_on_msg(!list_empty(&res->l_blocked_list),
			"Lockres %s is on the blocked list\n",
			res->l_name);
	mlog_bug_on_msg(!list_empty(&res->l_mask_waiters),
			"Lockres %s has mask waiters pending\n",
			res->l_name);
	mlog_bug_on_msg(spin_is_locked(&res->l_lock),
			"Lockres %s is locked\n",
			res->l_name);
	mlog_bug_on_msg(res->l_ro_holders,
			"Lockres %s has %u ro holders\n",
			res->l_name, res->l_ro_holders);
	mlog_bug_on_msg(res->l_ex_holders,
			"Lockres %s has %u ex holders\n",
			res->l_name, res->l_ex_holders);

	/* Need to clear out the lock status block for the dlm */
	memset(&res->l_lksb, 0, sizeof(res->l_lksb));

	res->l_flags = 0UL;
	mlog_exit_void();
}

static inline void ocfs2_inc_holders(struct ocfs2_lock_res *lockres,
				     int level)
{
	mlog_entry_void();

	BUG_ON(!lockres);

	switch(level) {
	case LKM_EXMODE:
		lockres->l_ex_holders++;
		break;
	case LKM_PRMODE:
		lockres->l_ro_holders++;
		break;
	default:
		BUG();
	}

	mlog_exit_void();
}

static inline void ocfs2_dec_holders(struct ocfs2_lock_res *lockres,
				     int level)
{
	mlog_entry_void();

	BUG_ON(!lockres);

	switch(level) {
	case LKM_EXMODE:
		BUG_ON(!lockres->l_ex_holders);
		lockres->l_ex_holders--;
		break;
	case LKM_PRMODE:
		BUG_ON(!lockres->l_ro_holders);
		lockres->l_ro_holders--;
		break;
	default:
		BUG();
	}
	mlog_exit_void();
}

/* WARNING: This function lives in a world where the only three lock
 * levels are EX, PR, and NL. It *will* have to be adjusted when more
 * lock types are added. */
static inline int ocfs2_highest_compat_lock_level(int level)
{
	int new_level = LKM_EXMODE;

	if (level == LKM_EXMODE)
		new_level = LKM_NLMODE;
	else if (level == LKM_PRMODE)
		new_level = LKM_PRMODE;
	return new_level;
}

static void lockres_set_flags(struct ocfs2_lock_res *lockres,
			      unsigned long newflags)
{
	struct list_head *pos, *tmp;
	struct ocfs2_mask_waiter *mw;

 	assert_spin_locked(&lockres->l_lock);

	lockres->l_flags = newflags;

	list_for_each_safe(pos, tmp, &lockres->l_mask_waiters) {
		mw = list_entry(pos, struct ocfs2_mask_waiter, mw_item);
		if ((lockres->l_flags & mw->mw_mask) != mw->mw_goal)
			continue;

		list_del_init(&mw->mw_item);
		mw->mw_status = 0;
		complete(&mw->mw_complete);
	}
}
static void lockres_or_flags(struct ocfs2_lock_res *lockres, unsigned long or)
{
	lockres_set_flags(lockres, lockres->l_flags | or);
}
static void lockres_clear_flags(struct ocfs2_lock_res *lockres,
				unsigned long clear)
{
	lockres_set_flags(lockres, lockres->l_flags & ~clear);
}

static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres)
{
	mlog_entry_void();

	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BUSY));
	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_ATTACHED));
	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
	BUG_ON(lockres->l_blocking <= LKM_NLMODE);

	lockres->l_level = lockres->l_requested;
	if (lockres->l_level <=
	    ocfs2_highest_compat_lock_level(lockres->l_blocking)) {
		lockres->l_blocking = LKM_NLMODE;
		lockres_clear_flags(lockres, OCFS2_LOCK_BLOCKED);
	}
	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);

	mlog_exit_void();
}

static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres)
{
	mlog_entry_void();

	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BUSY));
	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_ATTACHED));

	/* Convert from RO to EX doesn't really need anything as our
	 * information is already up to data. Convert from NL to
	 * *anything* however should mark ourselves as needing an
	 * update */
	if (lockres->l_level == LKM_NLMODE &&
	    lockres->l_ops->flags & LOCK_TYPE_REQUIRES_REFRESH)
		lockres_or_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);

	lockres->l_level = lockres->l_requested;
	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);

	mlog_exit_void();
}

static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres)
{
	mlog_entry_void();

	BUG_ON((!lockres->l_flags & OCFS2_LOCK_BUSY));
	BUG_ON(lockres->l_flags & OCFS2_LOCK_ATTACHED);

	if (lockres->l_requested > LKM_NLMODE &&
	    !(lockres->l_flags & OCFS2_LOCK_LOCAL) &&
	    lockres->l_ops->flags & LOCK_TYPE_REQUIRES_REFRESH)
		lockres_or_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);

	lockres->l_level = lockres->l_requested;
	lockres_or_flags(lockres, OCFS2_LOCK_ATTACHED);
	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);

	mlog_exit_void();
}

static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres,
				     int level)
{
	int needs_downconvert = 0;
	mlog_entry_void();

	assert_spin_locked(&lockres->l_lock);

	lockres_or_flags(lockres, OCFS2_LOCK_BLOCKED);

	if (level > lockres->l_blocking) {
		/* only schedule a downconvert if we haven't already scheduled
		 * one that goes low enough to satisfy the level we're
		 * blocking.  this also catches the case where we get
		 * duplicate BASTs */
		if (ocfs2_highest_compat_lock_level(level) <
		    ocfs2_highest_compat_lock_level(lockres->l_blocking))
			needs_downconvert = 1;

		lockres->l_blocking = level;
	}

	mlog_exit(needs_downconvert);
	return needs_downconvert;
}

static void ocfs2_blocking_ast(void *opaque, int level)
{
	struct ocfs2_lock_res *lockres = opaque;
	struct ocfs2_super *osb = ocfs2_get_lockres_osb(lockres);
	int needs_downconvert;
	unsigned long flags;

	BUG_ON(level <= LKM_NLMODE);

	mlog(0, "BAST fired for lockres %s, blocking %d, level %d type %s\n",
	     lockres->l_name, level, lockres->l_level,
	     ocfs2_lock_type_string(lockres->l_type));

	spin_lock_irqsave(&lockres->l_lock, flags);
	needs_downconvert = ocfs2_generic_handle_bast(lockres, level);
	if (needs_downconvert)
		ocfs2_schedule_blocked_lock(osb, lockres);
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	wake_up(&lockres->l_event);

	ocfs2_kick_vote_thread(osb);
}

static void ocfs2_locking_ast(void *opaque)
{
	struct ocfs2_lock_res *lockres = opaque;
	struct dlm_lockstatus *lksb = &lockres->l_lksb;
	unsigned long flags;

	spin_lock_irqsave(&lockres->l_lock, flags);

	if (lksb->status != DLM_NORMAL) {
		mlog(ML_ERROR, "lockres %s: lksb status value of %u!\n",
		     lockres->l_name, lksb->status);
		spin_unlock_irqrestore(&lockres->l_lock, flags);
		return;
	}

	switch(lockres->l_action) {
	case OCFS2_AST_ATTACH:
		ocfs2_generic_handle_attach_action(lockres);
		lockres_clear_flags(lockres, OCFS2_LOCK_LOCAL);
		break;
	case OCFS2_AST_CONVERT:
		ocfs2_generic_handle_convert_action(lockres);
		break;
	case OCFS2_AST_DOWNCONVERT:
		ocfs2_generic_handle_downconvert_action(lockres);
		break;
	default:
		mlog(ML_ERROR, "lockres %s: ast fired with invalid action: %u "
		     "lockres flags = 0x%lx, unlock action: %u\n",
		     lockres->l_name, lockres->l_action, lockres->l_flags,
		     lockres->l_unlock_action);
		BUG();
	}

	/* set it to something invalid so if we get called again we
	 * can catch it. */
	lockres->l_action = OCFS2_AST_INVALID;

	wake_up(&lockres->l_event);
	spin_unlock_irqrestore(&lockres->l_lock, flags);
}

static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
						int convert)
{
	unsigned long flags;

	mlog_entry_void();
	spin_lock_irqsave(&lockres->l_lock, flags);
	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
	if (convert)
		lockres->l_action = OCFS2_AST_INVALID;
	else
		lockres->l_unlock_action = OCFS2_UNLOCK_INVALID;
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	wake_up(&lockres->l_event);
	mlog_exit_void();
}

/* Note: If we detect another process working on the lock (i.e.,
 * OCFS2_LOCK_BUSY), we'll bail out returning 0. It's up to the caller
 * to do the right thing in that case.
 */
static int ocfs2_lock_create(struct ocfs2_super *osb,
			     struct ocfs2_lock_res *lockres,
			     int level,
			     int dlm_flags)
{
	int ret = 0;
	enum dlm_status status = DLM_NORMAL;
	unsigned long flags;

	mlog_entry_void();

	mlog(0, "lock %s, level = %d, flags = %d\n", lockres->l_name, level,
	     dlm_flags);

	spin_lock_irqsave(&lockres->l_lock, flags);
	if ((lockres->l_flags & OCFS2_LOCK_ATTACHED) ||
	    (lockres->l_flags & OCFS2_LOCK_BUSY)) {
		spin_unlock_irqrestore(&lockres->l_lock, flags);
		goto bail;
	}

	lockres->l_action = OCFS2_AST_ATTACH;
	lockres->l_requested = level;
	lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	status = dlmlock(osb->dlm,
			 level,
			 &lockres->l_lksb,
			 dlm_flags,
			 lockres->l_name,
			 OCFS2_LOCK_ID_MAX_LEN - 1,
			 ocfs2_locking_ast,
			 lockres,
			 ocfs2_blocking_ast);
	if (status != DLM_NORMAL) {
		ocfs2_log_dlm_error("dlmlock", status, lockres);
		ret = -EINVAL;
		ocfs2_recover_from_dlm_error(lockres, 1);
	}

	mlog(0, "lock %s, successfull return from dlmlock\n", lockres->l_name);

bail:
	mlog_exit(ret);
	return ret;
}

static inline int ocfs2_check_wait_flag(struct ocfs2_lock_res *lockres,
					int flag)
{
	unsigned long flags;
	int ret;

	spin_lock_irqsave(&lockres->l_lock, flags);
	ret = lockres->l_flags & flag;
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	return ret;
}

static inline void ocfs2_wait_on_busy_lock(struct ocfs2_lock_res *lockres)

{
	wait_event(lockres->l_event,
		   !ocfs2_check_wait_flag(lockres, OCFS2_LOCK_BUSY));
}

static inline void ocfs2_wait_on_refreshing_lock(struct ocfs2_lock_res *lockres)

{
	wait_event(lockres->l_event,
		   !ocfs2_check_wait_flag(lockres, OCFS2_LOCK_REFRESHING));
}

/* predict what lock level we'll be dropping down to on behalf
 * of another node, and return true if the currently wanted
 * level will be compatible with it. */
static inline int ocfs2_may_continue_on_blocked_lock(struct ocfs2_lock_res *lockres,
						     int wanted)
{
	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));

	return wanted <= ocfs2_highest_compat_lock_level(lockres->l_blocking);
}

static void ocfs2_init_mask_waiter(struct ocfs2_mask_waiter *mw)
{
	INIT_LIST_HEAD(&mw->mw_item);
	init_completion(&mw->mw_complete);
}

static int ocfs2_wait_for_mask(struct ocfs2_mask_waiter *mw)
{
	wait_for_completion(&mw->mw_complete);
	/* Re-arm the completion in case we want to wait on it again */
	INIT_COMPLETION(mw->mw_complete);
	return mw->mw_status;
}

static void lockres_add_mask_waiter(struct ocfs2_lock_res *lockres,
				    struct ocfs2_mask_waiter *mw,
				    unsigned long mask,
				    unsigned long goal)
{
	BUG_ON(!list_empty(&mw->mw_item));

	assert_spin_locked(&lockres->l_lock);

	list_add_tail(&mw->mw_item, &lockres->l_mask_waiters);
	mw->mw_mask = mask;
	mw->mw_goal = goal;
}

/* returns 0 if the mw that was removed was already satisfied, -EBUSY
 * if the mask still hadn't reached its goal */
static int lockres_remove_mask_waiter(struct ocfs2_lock_res *lockres,
				      struct ocfs2_mask_waiter *mw)
{
	unsigned long flags;
	int ret = 0;

	spin_lock_irqsave(&lockres->l_lock, flags);
	if (!list_empty(&mw->mw_item)) {
		if ((lockres->l_flags & mw->mw_mask) != mw->mw_goal)
			ret = -EBUSY;

		list_del_init(&mw->mw_item);
		init_completion(&mw->mw_complete);
	}
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	return ret;

}

static int ocfs2_cluster_lock(struct ocfs2_super *osb,
			      struct ocfs2_lock_res *lockres,
			      int level,
			      int lkm_flags,
			      int arg_flags)
{
	struct ocfs2_mask_waiter mw;
	enum dlm_status status;
	int wait, catch_signals = !(osb->s_mount_opt & OCFS2_MOUNT_NOINTR);
	int ret = 0; /* gcc doesn't realize wait = 1 guarantees ret is set */
	unsigned long flags;

	mlog_entry_void();

	ocfs2_init_mask_waiter(&mw);

	if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB)
		lkm_flags |= LKM_VALBLK;

again:
	wait = 0;

	if (catch_signals && signal_pending(current)) {
		ret = -ERESTARTSYS;
		goto out;
	}

	spin_lock_irqsave(&lockres->l_lock, flags);

	mlog_bug_on_msg(lockres->l_flags & OCFS2_LOCK_FREEING,
			"Cluster lock called on freeing lockres %s! flags "
			"0x%lx\n", lockres->l_name, lockres->l_flags);

	/* We only compare against the currently granted level
	 * here. If the lock is blocked waiting on a downconvert,
	 * we'll get caught below. */
	if (lockres->l_flags & OCFS2_LOCK_BUSY &&
	    level > lockres->l_level) {
		/* is someone sitting in dlm_lock? If so, wait on
		 * them. */
		lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BUSY, 0);
		wait = 1;
		goto unlock;
	}

	if (!(lockres->l_flags & OCFS2_LOCK_ATTACHED)) {
		/* lock has not been created yet. */
		spin_unlock_irqrestore(&lockres->l_lock, flags);

		ret = ocfs2_lock_create(osb, lockres, LKM_NLMODE, 0);
		if (ret < 0) {
			mlog_errno(ret);
			goto out;
		}
		goto again;
	}

	if (lockres->l_flags & OCFS2_LOCK_BLOCKED &&
	    !ocfs2_may_continue_on_blocked_lock(lockres, level)) {
		/* is the lock is currently blocked on behalf of
		 * another node */
		lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_BLOCKED, 0);
		wait = 1;
		goto unlock;
	}

	if (level > lockres->l_level) {
		if (lockres->l_action != OCFS2_AST_INVALID)
			mlog(ML_ERROR, "lockres %s has action %u pending\n",
			     lockres->l_name, lockres->l_action);

		lockres->l_action = OCFS2_AST_CONVERT;
		lockres->l_requested = level;
		lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
		spin_unlock_irqrestore(&lockres->l_lock, flags);

		BUG_ON(level == LKM_IVMODE);
		BUG_ON(level == LKM_NLMODE);

		mlog(0, "lock %s, convert from %d to level = %d\n",
		     lockres->l_name, lockres->l_level, level);

		/* call dlm_lock to upgrade lock now */
		status = dlmlock(osb->dlm,
				 level,
				 &lockres->l_lksb,
				 lkm_flags|LKM_CONVERT,
				 lockres->l_name,
				 OCFS2_LOCK_ID_MAX_LEN - 1,
				 ocfs2_locking_ast,
				 lockres,
				 ocfs2_blocking_ast);
		if (status != DLM_NORMAL) {
			if ((lkm_flags & LKM_NOQUEUE) &&
			    (status == DLM_NOTQUEUED))
				ret = -EAGAIN;
			else {
				ocfs2_log_dlm_error("dlmlock", status,
						    lockres);
				ret = -EINVAL;
			}
			ocfs2_recover_from_dlm_error(lockres, 1);
			goto out;
		}

		mlog(0, "lock %s, successfull return from dlmlock\n",
		     lockres->l_name);

		/* At this point we've gone inside the dlm and need to
		 * complete our work regardless. */
		catch_signals = 0;

		/* wait for busy to clear and carry on */
		goto again;
	}

	/* Ok, if we get here then we're good to go. */
	ocfs2_inc_holders(lockres, level);

	ret = 0;
unlock:
	spin_unlock_irqrestore(&lockres->l_lock, flags);
out:
	/*
	 * This is helping work around a lock inversion between the page lock
	 * and dlm locks.  One path holds the page lock while calling aops
	 * which block acquiring dlm locks.  The voting thread holds dlm
	 * locks while acquiring page locks while down converting data locks.
	 * This block is helping an aop path notice the inversion and back
	 * off to unlock its page lock before trying the dlm lock again.
	 */
	if (wait && arg_flags & OCFS2_LOCK_NONBLOCK &&
	    mw.mw_mask & (OCFS2_LOCK_BUSY|OCFS2_LOCK_BLOCKED)) {
		wait = 0;
		if (lockres_remove_mask_waiter(lockres, &mw))
			ret = -EAGAIN;
		else
			goto again;
	}
	if (wait) {
		ret = ocfs2_wait_for_mask(&mw);
		if (ret == 0)
			goto again;
		mlog_errno(ret);
	}

	mlog_exit(ret);
	return ret;
}

static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
				 struct ocfs2_lock_res *lockres,
				 int level)
{
	unsigned long flags;

	mlog_entry_void();
	spin_lock_irqsave(&lockres->l_lock, flags);
	ocfs2_dec_holders(lockres, level);
	ocfs2_vote_on_unlock(osb, lockres);
	spin_unlock_irqrestore(&lockres->l_lock, flags);
	mlog_exit_void();
}

static int ocfs2_create_new_lock(struct ocfs2_super *osb,
				 struct ocfs2_lock_res *lockres,
				 int ex,
				 int local)
{
	int level =  ex ? LKM_EXMODE : LKM_PRMODE;
	unsigned long flags;
	int lkm_flags = local ? LKM_LOCAL : 0;

	spin_lock_irqsave(&lockres->l_lock, flags);
	BUG_ON(lockres->l_flags & OCFS2_LOCK_ATTACHED);
	lockres_or_flags(lockres, OCFS2_LOCK_LOCAL);
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	return ocfs2_lock_create(osb, lockres, level, lkm_flags);
}

/* Grants us an EX lock on the data and metadata resources, skipping
 * the normal cluster directory lookup. Use this ONLY on newly created
 * inodes which other nodes can't possibly see, and which haven't been
 * hashed in the inode hash yet. This can give us a good performance
 * increase as it'll skip the network broadcast normally associated
 * with creating a new lock resource. */
int ocfs2_create_new_inode_locks(struct inode *inode)
{
	int ret;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	BUG_ON(!inode);
	BUG_ON(!ocfs2_inode_is_new(inode));

	mlog_entry_void();

	mlog(0, "Inode %llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno);

	/* NOTE: That we don't increment any of the holder counts, nor
	 * do we add anything to a journal handle. Since this is
	 * supposed to be a new inode which the cluster doesn't know
	 * about yet, there is no need to.  As far as the LVB handling
	 * is concerned, this is basically like acquiring an EX lock
	 * on a resource which has an invalid one -- we'll set it
	 * valid when we release the EX. */

	ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_rw_lockres, 1, 1);
	if (ret) {
		mlog_errno(ret);
		goto bail;
	}

	/*
	 * We don't want to use LKM_LOCAL on a meta data lock as they
	 * don't use a generation in their lock names.
	 */
	ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_meta_lockres, 1, 0);
	if (ret) {
		mlog_errno(ret);
		goto bail;
	}

	ret = ocfs2_create_new_lock(osb, &OCFS2_I(inode)->ip_data_lockres, 1, 1);
	if (ret) {
		mlog_errno(ret);
		goto bail;
	}

bail:
	mlog_exit(ret);
	return ret;
}

int ocfs2_rw_lock(struct inode *inode, int write)
{
	int status, level;
	struct ocfs2_lock_res *lockres;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	BUG_ON(!inode);

	mlog_entry_void();

	mlog(0, "inode %llu take %s RW lock\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
	     write ? "EXMODE" : "PRMODE");

	if (ocfs2_mount_local(osb))
		return 0;

	lockres = &OCFS2_I(inode)->ip_rw_lockres;

	level = write ? LKM_EXMODE : LKM_PRMODE;

	status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, level, 0,
				    0);
	if (status < 0)
		mlog_errno(status);

	mlog_exit(status);
	return status;
}

void ocfs2_rw_unlock(struct inode *inode, int write)
{
	int level = write ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_rw_lockres;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	mlog_entry_void();

	mlog(0, "inode %llu drop %s RW lock\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
	     write ? "EXMODE" : "PRMODE");

	if (!ocfs2_mount_local(osb))
		ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);

	mlog_exit_void();
}

int ocfs2_data_lock_full(struct inode *inode,
			 int write,
			 int arg_flags)
{
	int status = 0, level;
	struct ocfs2_lock_res *lockres;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	BUG_ON(!inode);

	mlog_entry_void();

	mlog(0, "inode %llu take %s DATA lock\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
	     write ? "EXMODE" : "PRMODE");

	/* We'll allow faking a readonly data lock for
	 * rodevices. */
	if (ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb))) {
		if (write) {
			status = -EROFS;
			mlog_errno(status);
		}
		goto out;
	}

	if (ocfs2_mount_local(osb))
		goto out;

	lockres = &OCFS2_I(inode)->ip_data_lockres;

	level = write ? LKM_EXMODE : LKM_PRMODE;

	status = ocfs2_cluster_lock(OCFS2_SB(inode->i_sb), lockres, level,
				    0, arg_flags);
	if (status < 0 && status != -EAGAIN)
		mlog_errno(status);

out:
	mlog_exit(status);
	return status;
}

/* see ocfs2_meta_lock_with_page() */
int ocfs2_data_lock_with_page(struct inode *inode,
			      int write,
			      struct page *page)
{
	int ret;

	ret = ocfs2_data_lock_full(inode, write, OCFS2_LOCK_NONBLOCK);
	if (ret == -EAGAIN) {
		unlock_page(page);
		if (ocfs2_data_lock(inode, write) == 0)
			ocfs2_data_unlock(inode, write);
		ret = AOP_TRUNCATED_PAGE;
	}

	return ret;
}

static void ocfs2_vote_on_unlock(struct ocfs2_super *osb,
				 struct ocfs2_lock_res *lockres)
{
	int kick = 0;

	mlog_entry_void();

	/* If we know that another node is waiting on our lock, kick
	 * the vote thread * pre-emptively when we reach a release
	 * condition. */
	if (lockres->l_flags & OCFS2_LOCK_BLOCKED) {
		switch(lockres->l_blocking) {
		case LKM_EXMODE:
			if (!lockres->l_ex_holders && !lockres->l_ro_holders)
				kick = 1;
			break;
		case LKM_PRMODE:
			if (!lockres->l_ex_holders)
				kick = 1;
			break;
		default:
			BUG();
		}
	}

	if (kick)
		ocfs2_kick_vote_thread(osb);

	mlog_exit_void();
}

void ocfs2_data_unlock(struct inode *inode,
		       int write)
{
	int level = write ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_data_lockres;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	mlog_entry_void();

	mlog(0, "inode %llu drop %s DATA lock\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
	     write ? "EXMODE" : "PRMODE");

	if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)) &&
	    !ocfs2_mount_local(osb))
		ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);

	mlog_exit_void();
}

#define OCFS2_SEC_BITS   34
#define OCFS2_SEC_SHIFT  (64 - 34)
#define OCFS2_NSEC_MASK  ((1ULL << OCFS2_SEC_SHIFT) - 1)

/* LVB only has room for 64 bits of time here so we pack it for
 * now. */
static u64 ocfs2_pack_timespec(struct timespec *spec)
{
	u64 res;
	u64 sec = spec->tv_sec;
	u32 nsec = spec->tv_nsec;

	res = (sec << OCFS2_SEC_SHIFT) | (nsec & OCFS2_NSEC_MASK);

	return res;
}

/* Call this with the lockres locked. I am reasonably sure we don't
 * need ip_lock in this function as anyone who would be changing those
 * values is supposed to be blocked in ocfs2_meta_lock right now. */
static void __ocfs2_stuff_meta_lvb(struct inode *inode)
{
	struct ocfs2_inode_info *oi = OCFS2_I(inode);
	struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres;
	struct ocfs2_meta_lvb *lvb;

	mlog_entry_void();

	lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;

	/*
	 * Invalidate the LVB of a deleted inode - this way other
	 * nodes are forced to go to disk and discover the new inode
	 * status.
	 */
	if (oi->ip_flags & OCFS2_INODE_DELETED) {
		lvb->lvb_version = 0;
		goto out;
	}

	lvb->lvb_version   = OCFS2_LVB_VERSION;
	lvb->lvb_isize	   = cpu_to_be64(i_size_read(inode));
	lvb->lvb_iclusters = cpu_to_be32(oi->ip_clusters);
	lvb->lvb_iuid      = cpu_to_be32(inode->i_uid);
	lvb->lvb_igid      = cpu_to_be32(inode->i_gid);
	lvb->lvb_imode     = cpu_to_be16(inode->i_mode);
	lvb->lvb_inlink    = cpu_to_be16(inode->i_nlink);
	lvb->lvb_iatime_packed  =
		cpu_to_be64(ocfs2_pack_timespec(&inode->i_atime));
	lvb->lvb_ictime_packed =
		cpu_to_be64(ocfs2_pack_timespec(&inode->i_ctime));
	lvb->lvb_imtime_packed =
		cpu_to_be64(ocfs2_pack_timespec(&inode->i_mtime));
	lvb->lvb_iattr    = cpu_to_be32(oi->ip_attr);
	lvb->lvb_igeneration = cpu_to_be32(inode->i_generation);

out:
	mlog_meta_lvb(0, lockres);

	mlog_exit_void();
}

static void ocfs2_unpack_timespec(struct timespec *spec,
				  u64 packed_time)
{
	spec->tv_sec = packed_time >> OCFS2_SEC_SHIFT;
	spec->tv_nsec = packed_time & OCFS2_NSEC_MASK;
}

static void ocfs2_refresh_inode_from_lvb(struct inode *inode)
{
	struct ocfs2_inode_info *oi = OCFS2_I(inode);
	struct ocfs2_lock_res *lockres = &oi->ip_meta_lockres;
	struct ocfs2_meta_lvb *lvb;

	mlog_entry_void();

	mlog_meta_lvb(0, lockres);

	lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;

	/* We're safe here without the lockres lock... */
	spin_lock(&oi->ip_lock);
	oi->ip_clusters = be32_to_cpu(lvb->lvb_iclusters);
	i_size_write(inode, be64_to_cpu(lvb->lvb_isize));

	oi->ip_attr = be32_to_cpu(lvb->lvb_iattr);
	ocfs2_set_inode_flags(inode);

	/* fast-symlinks are a special case */
	if (S_ISLNK(inode->i_mode) && !oi->ip_clusters)
		inode->i_blocks = 0;
	else
		inode->i_blocks =
			ocfs2_align_bytes_to_sectors(i_size_read(inode));

	inode->i_uid     = be32_to_cpu(lvb->lvb_iuid);
	inode->i_gid     = be32_to_cpu(lvb->lvb_igid);
	inode->i_mode    = be16_to_cpu(lvb->lvb_imode);
	inode->i_nlink   = be16_to_cpu(lvb->lvb_inlink);
	ocfs2_unpack_timespec(&inode->i_atime,
			      be64_to_cpu(lvb->lvb_iatime_packed));
	ocfs2_unpack_timespec(&inode->i_mtime,
			      be64_to_cpu(lvb->lvb_imtime_packed));
	ocfs2_unpack_timespec(&inode->i_ctime,
			      be64_to_cpu(lvb->lvb_ictime_packed));
	spin_unlock(&oi->ip_lock);

	mlog_exit_void();
}

static inline int ocfs2_meta_lvb_is_trustable(struct inode *inode,
					      struct ocfs2_lock_res *lockres)
{
	struct ocfs2_meta_lvb *lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;

	if (lvb->lvb_version == OCFS2_LVB_VERSION
	    && be32_to_cpu(lvb->lvb_igeneration) == inode->i_generation)
		return 1;
	return 0;
}

/* Determine whether a lock resource needs to be refreshed, and
 * arbitrate who gets to refresh it.
 *
 *   0 means no refresh needed.
 *
 *   > 0 means you need to refresh this and you MUST call
 *   ocfs2_complete_lock_res_refresh afterwards. */
static int ocfs2_should_refresh_lock_res(struct ocfs2_lock_res *lockres)
{
	unsigned long flags;
	int status = 0;

	mlog_entry_void();

refresh_check:
	spin_lock_irqsave(&lockres->l_lock, flags);
	if (!(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH)) {
		spin_unlock_irqrestore(&lockres->l_lock, flags);
		goto bail;
	}

	if (lockres->l_flags & OCFS2_LOCK_REFRESHING) {
		spin_unlock_irqrestore(&lockres->l_lock, flags);

		ocfs2_wait_on_refreshing_lock(lockres);
		goto refresh_check;
	}

	/* Ok, I'll be the one to refresh this lock. */
	lockres_or_flags(lockres, OCFS2_LOCK_REFRESHING);
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	status = 1;
bail:
	mlog_exit(status);
	return status;
}

/* If status is non zero, I'll mark it as not being in refresh
 * anymroe, but i won't clear the needs refresh flag. */
static inline void ocfs2_complete_lock_res_refresh(struct ocfs2_lock_res *lockres,
						   int status)
{
	unsigned long flags;
	mlog_entry_void();

	spin_lock_irqsave(&lockres->l_lock, flags);
	lockres_clear_flags(lockres, OCFS2_LOCK_REFRESHING);
	if (!status)
		lockres_clear_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	wake_up(&lockres->l_event);

	mlog_exit_void();
}

/* may or may not return a bh if it went to disk. */
static int ocfs2_meta_lock_update(struct inode *inode,
				  struct buffer_head **bh)
{
	int status = 0;
	struct ocfs2_inode_info *oi = OCFS2_I(inode);
	struct ocfs2_lock_res *lockres = NULL;
	struct ocfs2_dinode *fe;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	mlog_entry_void();

	spin_lock(&oi->ip_lock);
	if (oi->ip_flags & OCFS2_INODE_DELETED) {
		mlog(0, "Orphaned inode %llu was deleted while we "
		     "were waiting on a lock. ip_flags = 0x%x\n",
		     (unsigned long long)oi->ip_blkno, oi->ip_flags);
		spin_unlock(&oi->ip_lock);
		status = -ENOENT;
		goto bail;
	}
	spin_unlock(&oi->ip_lock);

	if (!ocfs2_mount_local(osb)) {
		lockres = &oi->ip_meta_lockres;

		if (!ocfs2_should_refresh_lock_res(lockres))
			goto bail;
	}

	/* This will discard any caching information we might have had
	 * for the inode metadata. */
	ocfs2_metadata_cache_purge(inode);

	/* will do nothing for inode types that don't use the extent
	 * map (directories, bitmap files, etc) */
	ocfs2_extent_map_trunc(inode, 0);

	if (lockres && ocfs2_meta_lvb_is_trustable(inode, lockres)) {
		mlog(0, "Trusting LVB on inode %llu\n",
		     (unsigned long long)oi->ip_blkno);
		ocfs2_refresh_inode_from_lvb(inode);
	} else {
		/* Boo, we have to go to disk. */
		/* read bh, cast, ocfs2_refresh_inode */
		status = ocfs2_read_block(OCFS2_SB(inode->i_sb), oi->ip_blkno,
					  bh, OCFS2_BH_CACHED, inode);
		if (status < 0) {
			mlog_errno(status);
			goto bail_refresh;
		}
		fe = (struct ocfs2_dinode *) (*bh)->b_data;

		/* This is a good chance to make sure we're not
		 * locking an invalid object.
		 *
		 * We bug on a stale inode here because we checked
		 * above whether it was wiped from disk. The wiping
		 * node provides a guarantee that we receive that
		 * message and can mark the inode before dropping any
		 * locks associated with it. */
		if (!OCFS2_IS_VALID_DINODE(fe)) {
			OCFS2_RO_ON_INVALID_DINODE(inode->i_sb, fe);
			status = -EIO;
			goto bail_refresh;
		}
		mlog_bug_on_msg(inode->i_generation !=
				le32_to_cpu(fe->i_generation),
				"Invalid dinode %llu disk generation: %u "
				"inode->i_generation: %u\n",
				(unsigned long long)oi->ip_blkno,
				le32_to_cpu(fe->i_generation),
				inode->i_generation);
		mlog_bug_on_msg(le64_to_cpu(fe->i_dtime) ||
				!(fe->i_flags & cpu_to_le32(OCFS2_VALID_FL)),
				"Stale dinode %llu dtime: %llu flags: 0x%x\n",
				(unsigned long long)oi->ip_blkno,
				(unsigned long long)le64_to_cpu(fe->i_dtime),
				le32_to_cpu(fe->i_flags));

		ocfs2_refresh_inode(inode, fe);
	}

	status = 0;
bail_refresh:
	if (lockres)
		ocfs2_complete_lock_res_refresh(lockres, status);
bail:
	mlog_exit(status);
	return status;
}

static int ocfs2_assign_bh(struct inode *inode,
			   struct buffer_head **ret_bh,
			   struct buffer_head *passed_bh)
{
	int status;

	if (passed_bh) {
		/* Ok, the update went to disk for us, use the
		 * returned bh. */
		*ret_bh = passed_bh;
		get_bh(*ret_bh);

		return 0;
	}

	status = ocfs2_read_block(OCFS2_SB(inode->i_sb),
				  OCFS2_I(inode)->ip_blkno,
				  ret_bh,
				  OCFS2_BH_CACHED,
				  inode);
	if (status < 0)
		mlog_errno(status);

	return status;
}

/*
 * returns < 0 error if the callback will never be called, otherwise
 * the result of the lock will be communicated via the callback.
 */
int ocfs2_meta_lock_full(struct inode *inode,
			 struct buffer_head **ret_bh,
			 int ex,
			 int arg_flags)
{
	int status, level, dlm_flags, acquired;
	struct ocfs2_lock_res *lockres = NULL;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	struct buffer_head *local_bh = NULL;

	BUG_ON(!inode);

	mlog_entry_void();

	mlog(0, "inode %llu, take %s META lock\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
	     ex ? "EXMODE" : "PRMODE");

	status = 0;
	acquired = 0;
	/* We'll allow faking a readonly metadata lock for
	 * rodevices. */
	if (ocfs2_is_hard_readonly(osb)) {
		if (ex)
			status = -EROFS;
		goto bail;
	}

	if (ocfs2_mount_local(osb))
		goto local;

	if (!(arg_flags & OCFS2_META_LOCK_RECOVERY))
		wait_event(osb->recovery_event,
			   ocfs2_node_map_is_empty(osb, &osb->recovery_map));

	acquired = 0;
	lockres = &OCFS2_I(inode)->ip_meta_lockres;
	level = ex ? LKM_EXMODE : LKM_PRMODE;
	dlm_flags = 0;
	if (arg_flags & OCFS2_META_LOCK_NOQUEUE)
		dlm_flags |= LKM_NOQUEUE;

	status = ocfs2_cluster_lock(osb, lockres, level, dlm_flags, arg_flags);
	if (status < 0) {
		if (status != -EAGAIN && status != -EIOCBRETRY)
			mlog_errno(status);
		goto bail;
	}

	/* Notify the error cleanup path to drop the cluster lock. */
	acquired = 1;

	/* We wait twice because a node may have died while we were in
	 * the lower dlm layers. The second time though, we've
	 * committed to owning this lock so we don't allow signals to
	 * abort the operation. */
	if (!(arg_flags & OCFS2_META_LOCK_RECOVERY))
		wait_event(osb->recovery_event,
			   ocfs2_node_map_is_empty(osb, &osb->recovery_map));

local:
	/*
	 * We only see this flag if we're being called from
	 * ocfs2_read_locked_inode(). It means we're locking an inode
	 * which hasn't been populated yet, so clear the refresh flag
	 * and let the caller handle it.
	 */
	if (inode->i_state & I_NEW) {
		status = 0;
		if (lockres)
			ocfs2_complete_lock_res_refresh(lockres, 0);
		goto bail;
	}

	/* This is fun. The caller may want a bh back, or it may
	 * not. ocfs2_meta_lock_update definitely wants one in, but
	 * may or may not read one, depending on what's in the
	 * LVB. The result of all of this is that we've *only* gone to
	 * disk if we have to, so the complexity is worthwhile. */
	status = ocfs2_meta_lock_update(inode, &local_bh);
	if (status < 0) {
		if (status != -ENOENT)
			mlog_errno(status);
		goto bail;
	}

	if (ret_bh) {
		status = ocfs2_assign_bh(inode, ret_bh, local_bh);
		if (status < 0) {
			mlog_errno(status);
			goto bail;
		}
	}

bail:
	if (status < 0) {
		if (ret_bh && (*ret_bh)) {
			brelse(*ret_bh);
			*ret_bh = NULL;
		}
		if (acquired)
			ocfs2_meta_unlock(inode, ex);
	}

	if (local_bh)
		brelse(local_bh);

	mlog_exit(status);
	return status;
}

/*
 * This is working around a lock inversion between tasks acquiring DLM locks
 * while holding a page lock and the vote thread which blocks dlm lock acquiry
 * while acquiring page locks.
 *
 * ** These _with_page variantes are only intended to be called from aop
 * methods that hold page locks and return a very specific *positive* error
 * code that aop methods pass up to the VFS -- test for errors with != 0. **
 *
 * The DLM is called such that it returns -EAGAIN if it would have blocked
 * waiting for the vote thread.  In that case we unlock our page so the vote
 * thread can make progress.  Once we've done this we have to return
 * AOP_TRUNCATED_PAGE so the aop method that called us can bubble that back up
 * into the VFS who will then immediately retry the aop call.
 *
 * We do a blocking lock and immediate unlock before returning, though, so that
 * the lock has a great chance of being cached on this node by the time the VFS
 * calls back to retry the aop.    This has a potential to livelock as nodes
 * ping locks back and forth, but that's a risk we're willing to take to avoid
 * the lock inversion simply.
 */
int ocfs2_meta_lock_with_page(struct inode *inode,
			      struct buffer_head **ret_bh,
			      int ex,
			      struct page *page)
{
	int ret;

	ret = ocfs2_meta_lock_full(inode, ret_bh, ex, OCFS2_LOCK_NONBLOCK);
	if (ret == -EAGAIN) {
		unlock_page(page);
		if (ocfs2_meta_lock(inode, ret_bh, ex) == 0)
			ocfs2_meta_unlock(inode, ex);
		ret = AOP_TRUNCATED_PAGE;
	}

	return ret;
}

int ocfs2_meta_lock_atime(struct inode *inode,
			  struct vfsmount *vfsmnt,
			  int *level)
{
	int ret;

	mlog_entry_void();
	ret = ocfs2_meta_lock(inode, NULL, 0);
	if (ret < 0) {
		mlog_errno(ret);
		return ret;
	}

	/*
	 * If we should update atime, we will get EX lock,
	 * otherwise we just get PR lock.
	 */
	if (ocfs2_should_update_atime(inode, vfsmnt)) {
		struct buffer_head *bh = NULL;

		ocfs2_meta_unlock(inode, 0);
		ret = ocfs2_meta_lock(inode, &bh, 1);
		if (ret < 0) {
			mlog_errno(ret);
			return ret;
		}
		*level = 1;
		if (ocfs2_should_update_atime(inode, vfsmnt))
			ocfs2_update_inode_atime(inode, bh);
		if (bh)
			brelse(bh);
	} else
		*level = 0;

	mlog_exit(ret);
	return ret;
}

void ocfs2_meta_unlock(struct inode *inode,
		       int ex)
{
	int level = ex ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_meta_lockres;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	mlog_entry_void();

	mlog(0, "inode %llu drop %s META lock\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
	     ex ? "EXMODE" : "PRMODE");

	if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)) &&
	    !ocfs2_mount_local(osb))
		ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);

	mlog_exit_void();
}

int ocfs2_super_lock(struct ocfs2_super *osb,
		     int ex)
{
	int status = 0;
	int level = ex ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_lock_res *lockres = &osb->osb_super_lockres;
	struct buffer_head *bh;
	struct ocfs2_slot_info *si = osb->slot_info;

	mlog_entry_void();

	if (ocfs2_is_hard_readonly(osb))
		return -EROFS;

	if (ocfs2_mount_local(osb))
		goto bail;

	status = ocfs2_cluster_lock(osb, lockres, level, 0, 0);
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}

	/* The super block lock path is really in the best position to
	 * know when resources covered by the lock need to be
	 * refreshed, so we do it here. Of course, making sense of
	 * everything is up to the caller :) */
	status = ocfs2_should_refresh_lock_res(lockres);
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}
	if (status) {
		bh = si->si_bh;
		status = ocfs2_read_block(osb, bh->b_blocknr, &bh, 0,
					  si->si_inode);
		if (status == 0)
			ocfs2_update_slot_info(si);

		ocfs2_complete_lock_res_refresh(lockres, status);

		if (status < 0)
			mlog_errno(status);
	}
bail:
	mlog_exit(status);
	return status;
}

void ocfs2_super_unlock(struct ocfs2_super *osb,
			int ex)
{
	int level = ex ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_lock_res *lockres = &osb->osb_super_lockres;

	if (!ocfs2_mount_local(osb))
		ocfs2_cluster_unlock(osb, lockres, level);
}

int ocfs2_rename_lock(struct ocfs2_super *osb)
{
	int status;
	struct ocfs2_lock_res *lockres = &osb->osb_rename_lockres;

	if (ocfs2_is_hard_readonly(osb))
		return -EROFS;

	if (ocfs2_mount_local(osb))
		return 0;

	status = ocfs2_cluster_lock(osb, lockres, LKM_EXMODE, 0, 0);
	if (status < 0)
		mlog_errno(status);

	return status;
}

void ocfs2_rename_unlock(struct ocfs2_super *osb)
{
	struct ocfs2_lock_res *lockres = &osb->osb_rename_lockres;

	if (!ocfs2_mount_local(osb))
		ocfs2_cluster_unlock(osb, lockres, LKM_EXMODE);
}

int ocfs2_dentry_lock(struct dentry *dentry, int ex)
{
	int ret;
	int level = ex ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_dentry_lock *dl = dentry->d_fsdata;
	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);

	BUG_ON(!dl);

	if (ocfs2_is_hard_readonly(osb))
		return -EROFS;

	if (ocfs2_mount_local(osb))
		return 0;

	ret = ocfs2_cluster_lock(osb, &dl->dl_lockres, level, 0, 0);
	if (ret < 0)
		mlog_errno(ret);

	return ret;
}

void ocfs2_dentry_unlock(struct dentry *dentry, int ex)
{
	int level = ex ? LKM_EXMODE : LKM_PRMODE;
	struct ocfs2_dentry_lock *dl = dentry->d_fsdata;
	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);

	if (!ocfs2_mount_local(osb))
		ocfs2_cluster_unlock(osb, &dl->dl_lockres, level);
}

/* Reference counting of the dlm debug structure. We want this because
 * open references on the debug inodes can live on after a mount, so
 * we can't rely on the ocfs2_super to always exist. */
static void ocfs2_dlm_debug_free(struct kref *kref)
{
	struct ocfs2_dlm_debug *dlm_debug;

	dlm_debug = container_of(kref, struct ocfs2_dlm_debug, d_refcnt);

	kfree(dlm_debug);
}

void ocfs2_put_dlm_debug(struct ocfs2_dlm_debug *dlm_debug)
{
	if (dlm_debug)
		kref_put(&dlm_debug->d_refcnt, ocfs2_dlm_debug_free);
}

static void ocfs2_get_dlm_debug(struct ocfs2_dlm_debug *debug)
{
	kref_get(&debug->d_refcnt);
}

struct ocfs2_dlm_debug *ocfs2_new_dlm_debug(void)
{
	struct ocfs2_dlm_debug *dlm_debug;

	dlm_debug = kmalloc(sizeof(struct ocfs2_dlm_debug), GFP_KERNEL);
	if (!dlm_debug) {
		mlog_errno(-ENOMEM);
		goto out;
	}

	kref_init(&dlm_debug->d_refcnt);
	INIT_LIST_HEAD(&dlm_debug->d_lockres_tracking);
	dlm_debug->d_locking_state = NULL;
out:
	return dlm_debug;
}

/* Access to this is arbitrated for us via seq_file->sem. */
struct ocfs2_dlm_seq_priv {
	struct ocfs2_dlm_debug *p_dlm_debug;
	struct ocfs2_lock_res p_iter_res;
	struct ocfs2_lock_res p_tmp_res;
};

static struct ocfs2_lock_res *ocfs2_dlm_next_res(struct ocfs2_lock_res *start,
						 struct ocfs2_dlm_seq_priv *priv)
{
	struct ocfs2_lock_res *iter, *ret = NULL;
	struct ocfs2_dlm_debug *dlm_debug = priv->p_dlm_debug;

	assert_spin_locked(&ocfs2_dlm_tracking_lock);

	list_for_each_entry(iter, &start->l_debug_list, l_debug_list) {
		/* discover the head of the list */
		if (&iter->l_debug_list == &dlm_debug->d_lockres_tracking) {
			mlog(0, "End of list found, %p\n", ret);
			break;
		}

		/* We track our "dummy" iteration lockres' by a NULL
		 * l_ops field. */
		if (iter->l_ops != NULL) {
			ret = iter;
			break;
		}
	}

	return ret;
}

static void *ocfs2_dlm_seq_start(struct seq_file *m, loff_t *pos)
{
	struct ocfs2_dlm_seq_priv *priv = m->private;
	struct ocfs2_lock_res *iter;

	spin_lock(&ocfs2_dlm_tracking_lock);
	iter = ocfs2_dlm_next_res(&priv->p_iter_res, priv);
	if (iter) {
		/* Since lockres' have the lifetime of their container
		 * (which can be inodes, ocfs2_supers, etc) we want to
		 * copy this out to a temporary lockres while still
		 * under the spinlock. Obviously after this we can't
		 * trust any pointers on the copy returned, but that's
		 * ok as the information we want isn't typically held
		 * in them. */
		priv->p_tmp_res = *iter;
		iter = &priv->p_tmp_res;
	}
	spin_unlock(&ocfs2_dlm_tracking_lock);

	return iter;
}

static void ocfs2_dlm_seq_stop(struct seq_file *m, void *v)
{
}

static void *ocfs2_dlm_seq_next(struct seq_file *m, void *v, loff_t *pos)
{
	struct ocfs2_dlm_seq_priv *priv = m->private;
	struct ocfs2_lock_res *iter = v;
	struct ocfs2_lock_res *dummy = &priv->p_iter_res;

	spin_lock(&ocfs2_dlm_tracking_lock);
	iter = ocfs2_dlm_next_res(iter, priv);
	list_del_init(&dummy->l_debug_list);
	if (iter) {
		list_add(&dummy->l_debug_list, &iter->l_debug_list);
		priv->p_tmp_res = *iter;
		iter = &priv->p_tmp_res;
	}
	spin_unlock(&ocfs2_dlm_tracking_lock);

	return iter;
}

/* So that debugfs.ocfs2 can determine which format is being used */
#define OCFS2_DLM_DEBUG_STR_VERSION 1
static int ocfs2_dlm_seq_show(struct seq_file *m, void *v)
{
	int i;
	char *lvb;
	struct ocfs2_lock_res *lockres = v;

	if (!lockres)
		return -EINVAL;

	seq_printf(m, "0x%x\t", OCFS2_DLM_DEBUG_STR_VERSION);

	if (lockres->l_type == OCFS2_LOCK_TYPE_DENTRY)
		seq_printf(m, "%.*s%08x\t", OCFS2_DENTRY_LOCK_INO_START - 1,
			   lockres->l_name,
			   (unsigned int)ocfs2_get_dentry_lock_ino(lockres));
	else
		seq_printf(m, "%.*s\t", OCFS2_LOCK_ID_MAX_LEN, lockres->l_name);

	seq_printf(m, "%d\t"
		   "0x%lx\t"
		   "0x%x\t"
		   "0x%x\t"
		   "%u\t"
		   "%u\t"
		   "%d\t"
		   "%d\t",
		   lockres->l_level,
		   lockres->l_flags,
		   lockres->l_action,
		   lockres->l_unlock_action,
		   lockres->l_ro_holders,
		   lockres->l_ex_holders,
		   lockres->l_requested,
		   lockres->l_blocking);

	/* Dump the raw LVB */
	lvb = lockres->l_lksb.lvb;
	for(i = 0; i < DLM_LVB_LEN; i++)
		seq_printf(m, "0x%x\t", lvb[i]);

	/* End the line */
	seq_printf(m, "\n");
	return 0;
}

static struct seq_operations ocfs2_dlm_seq_ops = {
	.start =	ocfs2_dlm_seq_start,
	.stop =		ocfs2_dlm_seq_stop,
	.next =		ocfs2_dlm_seq_next,
	.show =		ocfs2_dlm_seq_show,
};

static int ocfs2_dlm_debug_release(struct inode *inode, struct file *file)
{
	struct seq_file *seq = (struct seq_file *) file->private_data;
	struct ocfs2_dlm_seq_priv *priv = seq->private;
	struct ocfs2_lock_res *res = &priv->p_iter_res;

	ocfs2_remove_lockres_tracking(res);
	ocfs2_put_dlm_debug(priv->p_dlm_debug);
	return seq_release_private(inode, file);
}

static int ocfs2_dlm_debug_open(struct inode *inode, struct file *file)
{
	int ret;
	struct ocfs2_dlm_seq_priv *priv;
	struct seq_file *seq;
	struct ocfs2_super *osb;

	priv = kzalloc(sizeof(struct ocfs2_dlm_seq_priv), GFP_KERNEL);
	if (!priv) {
		ret = -ENOMEM;
		mlog_errno(ret);
		goto out;
	}
	osb = inode->i_private;
	ocfs2_get_dlm_debug(osb->osb_dlm_debug);
	priv->p_dlm_debug = osb->osb_dlm_debug;
	INIT_LIST_HEAD(&priv->p_iter_res.l_debug_list);

	ret = seq_open(file, &ocfs2_dlm_seq_ops);
	if (ret) {
		kfree(priv);
		mlog_errno(ret);
		goto out;
	}

	seq = (struct seq_file *) file->private_data;
	seq->private = priv;

	ocfs2_add_lockres_tracking(&priv->p_iter_res,
				   priv->p_dlm_debug);

out:
	return ret;
}

static const struct file_operations ocfs2_dlm_debug_fops = {
	.open =		ocfs2_dlm_debug_open,
	.release =	ocfs2_dlm_debug_release,
	.read =		seq_read,
	.llseek =	seq_lseek,
};

static int ocfs2_dlm_init_debug(struct ocfs2_super *osb)
{
	int ret = 0;
	struct ocfs2_dlm_debug *dlm_debug = osb->osb_dlm_debug;

	dlm_debug->d_locking_state = debugfs_create_file("locking_state",
							 S_IFREG|S_IRUSR,
							 osb->osb_debug_root,
							 osb,
							 &ocfs2_dlm_debug_fops);
	if (!dlm_debug->d_locking_state) {
		ret = -EINVAL;
		mlog(ML_ERROR,
		     "Unable to create locking state debugfs file.\n");
		goto out;
	}

	ocfs2_get_dlm_debug(dlm_debug);
out:
	return ret;
}

static void ocfs2_dlm_shutdown_debug(struct ocfs2_super *osb)
{
	struct ocfs2_dlm_debug *dlm_debug = osb->osb_dlm_debug;

	if (dlm_debug) {
		debugfs_remove(dlm_debug->d_locking_state);
		ocfs2_put_dlm_debug(dlm_debug);
	}
}

int ocfs2_dlm_init(struct ocfs2_super *osb)
{
	int status = 0;
	u32 dlm_key;
	struct dlm_ctxt *dlm = NULL;

	mlog_entry_void();

	if (ocfs2_mount_local(osb))
		goto local;

	status = ocfs2_dlm_init_debug(osb);
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}

	/* launch vote thread */
	osb->vote_task = kthread_run(ocfs2_vote_thread, osb, "ocfs2vote");
	if (IS_ERR(osb->vote_task)) {
		status = PTR_ERR(osb->vote_task);
		osb->vote_task = NULL;
		mlog_errno(status);
		goto bail;
	}

	/* used by the dlm code to make message headers unique, each
	 * node in this domain must agree on this. */
	dlm_key = crc32_le(0, osb->uuid_str, strlen(osb->uuid_str));

	/* for now, uuid == domain */
	dlm = dlm_register_domain(osb->uuid_str, dlm_key);
	if (IS_ERR(dlm)) {
		status = PTR_ERR(dlm);
		mlog_errno(status);
		goto bail;
	}

	dlm_register_eviction_cb(dlm, &osb->osb_eviction_cb);

local:
	ocfs2_super_lock_res_init(&osb->osb_super_lockres, osb);
	ocfs2_rename_lock_res_init(&osb->osb_rename_lockres, osb);

	osb->dlm = dlm;

	status = 0;
bail:
	if (status < 0) {
		ocfs2_dlm_shutdown_debug(osb);
		if (osb->vote_task)
			kthread_stop(osb->vote_task);
	}

	mlog_exit(status);
	return status;
}

void ocfs2_dlm_shutdown(struct ocfs2_super *osb)
{
	mlog_entry_void();

	dlm_unregister_eviction_cb(&osb->osb_eviction_cb);

	ocfs2_drop_osb_locks(osb);

	if (osb->vote_task) {
		kthread_stop(osb->vote_task);
		osb->vote_task = NULL;
	}

	ocfs2_lock_res_free(&osb->osb_super_lockres);
	ocfs2_lock_res_free(&osb->osb_rename_lockres);

	dlm_unregister_domain(osb->dlm);
	osb->dlm = NULL;

	ocfs2_dlm_shutdown_debug(osb);

	mlog_exit_void();
}

static void ocfs2_unlock_ast(void *opaque, enum dlm_status status)
{
	struct ocfs2_lock_res *lockres = opaque;
	unsigned long flags;

	mlog_entry_void();

	mlog(0, "UNLOCK AST called on lock %s, action = %d\n", lockres->l_name,
	     lockres->l_unlock_action);

	spin_lock_irqsave(&lockres->l_lock, flags);
	/* We tried to cancel a convert request, but it was already
	 * granted. All we want to do here is clear our unlock
	 * state. The wake_up call done at the bottom is redundant
	 * (ocfs2_prepare_cancel_convert doesn't sleep on this) but doesn't
	 * hurt anything anyway */
	if (status == DLM_CANCELGRANT &&
	    lockres->l_unlock_action == OCFS2_UNLOCK_CANCEL_CONVERT) {
		mlog(0, "Got cancelgrant for %s\n", lockres->l_name);

		/* We don't clear the busy flag in this case as it
		 * should have been cleared by the ast which the dlm
		 * has called. */
		goto complete_unlock;
	}

	if (status != DLM_NORMAL) {
		mlog(ML_ERROR, "Dlm passes status %d for lock %s, "
		     "unlock_action %d\n", status, lockres->l_name,
		     lockres->l_unlock_action);
		spin_unlock_irqrestore(&lockres->l_lock, flags);
		return;
	}

	switch(lockres->l_unlock_action) {
	case OCFS2_UNLOCK_CANCEL_CONVERT:
		mlog(0, "Cancel convert success for %s\n", lockres->l_name);
		lockres->l_action = OCFS2_AST_INVALID;
		break;
	case OCFS2_UNLOCK_DROP_LOCK:
		lockres->l_level = LKM_IVMODE;
		break;
	default:
		BUG();
	}

	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
complete_unlock:
	lockres->l_unlock_action = OCFS2_UNLOCK_INVALID;
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	wake_up(&lockres->l_event);

	mlog_exit_void();
}

static int ocfs2_drop_lock(struct ocfs2_super *osb,
			   struct ocfs2_lock_res *lockres)
{
	enum dlm_status status;
	unsigned long flags;
	int lkm_flags = 0;

	/* We didn't get anywhere near actually using this lockres. */
	if (!(lockres->l_flags & OCFS2_LOCK_INITIALIZED))
		goto out;

	if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB)
		lkm_flags |= LKM_VALBLK;

	spin_lock_irqsave(&lockres->l_lock, flags);

	mlog_bug_on_msg(!(lockres->l_flags & OCFS2_LOCK_FREEING),
			"lockres %s, flags 0x%lx\n",
			lockres->l_name, lockres->l_flags);

	while (lockres->l_flags & OCFS2_LOCK_BUSY) {
		mlog(0, "waiting on busy lock \"%s\": flags = %lx, action = "
		     "%u, unlock_action = %u\n",
		     lockres->l_name, lockres->l_flags, lockres->l_action,
		     lockres->l_unlock_action);

		spin_unlock_irqrestore(&lockres->l_lock, flags);

		/* XXX: Today we just wait on any busy
		 * locks... Perhaps we need to cancel converts in the
		 * future? */
		ocfs2_wait_on_busy_lock(lockres);

		spin_lock_irqsave(&lockres->l_lock, flags);
	}

	if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB) {
		if (lockres->l_flags & OCFS2_LOCK_ATTACHED &&
		    lockres->l_level == LKM_EXMODE &&
		    !(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH))
			lockres->l_ops->set_lvb(lockres);
	}

	if (lockres->l_flags & OCFS2_LOCK_BUSY)
		mlog(ML_ERROR, "destroying busy lock: \"%s\"\n",
		     lockres->l_name);
	if (lockres->l_flags & OCFS2_LOCK_BLOCKED)
		mlog(0, "destroying blocked lock: \"%s\"\n", lockres->l_name);

	if (!(lockres->l_flags & OCFS2_LOCK_ATTACHED)) {
		spin_unlock_irqrestore(&lockres->l_lock, flags);
		goto out;
	}

	lockres_clear_flags(lockres, OCFS2_LOCK_ATTACHED);

	/* make sure we never get here while waiting for an ast to
	 * fire. */
	BUG_ON(lockres->l_action != OCFS2_AST_INVALID);

	/* is this necessary? */
	lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
	lockres->l_unlock_action = OCFS2_UNLOCK_DROP_LOCK;
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	mlog(0, "lock %s\n", lockres->l_name);

	status = dlmunlock(osb->dlm, &lockres->l_lksb, lkm_flags,
			   ocfs2_unlock_ast, lockres);
	if (status != DLM_NORMAL) {
		ocfs2_log_dlm_error("dlmunlock", status, lockres);
		mlog(ML_ERROR, "lockres flags: %lu\n", lockres->l_flags);
		dlm_print_one_lock(lockres->l_lksb.lockid);
		BUG();
	}
	mlog(0, "lock %s, successfull return from dlmunlock\n",
	     lockres->l_name);

	ocfs2_wait_on_busy_lock(lockres);
out:
	mlog_exit(0);
	return 0;
}

/* Mark the lockres as being dropped. It will no longer be
 * queued if blocking, but we still may have to wait on it
 * being dequeued from the vote thread before we can consider
 * it safe to drop. 
 *
 * You can *not* attempt to call cluster_lock on this lockres anymore. */
void ocfs2_mark_lockres_freeing(struct ocfs2_lock_res *lockres)
{
	int status;
	struct ocfs2_mask_waiter mw;
	unsigned long flags;

	ocfs2_init_mask_waiter(&mw);

	spin_lock_irqsave(&lockres->l_lock, flags);
	lockres->l_flags |= OCFS2_LOCK_FREEING;
	while (lockres->l_flags & OCFS2_LOCK_QUEUED) {
		lockres_add_mask_waiter(lockres, &mw, OCFS2_LOCK_QUEUED, 0);
		spin_unlock_irqrestore(&lockres->l_lock, flags);

		mlog(0, "Waiting on lockres %s\n", lockres->l_name);

		status = ocfs2_wait_for_mask(&mw);
		if (status)
			mlog_errno(status);

		spin_lock_irqsave(&lockres->l_lock, flags);
	}
	spin_unlock_irqrestore(&lockres->l_lock, flags);
}

void ocfs2_simple_drop_lockres(struct ocfs2_super *osb,
			       struct ocfs2_lock_res *lockres)
{
	int ret;

	ocfs2_mark_lockres_freeing(lockres);
	ret = ocfs2_drop_lock(osb, lockres);
	if (ret)
		mlog_errno(ret);
}

static void ocfs2_drop_osb_locks(struct ocfs2_super *osb)
{
	ocfs2_simple_drop_lockres(osb, &osb->osb_super_lockres);
	ocfs2_simple_drop_lockres(osb, &osb->osb_rename_lockres);
}

int ocfs2_drop_inode_locks(struct inode *inode)
{
	int status, err;

	mlog_entry_void();

	/* No need to call ocfs2_mark_lockres_freeing here -
	 * ocfs2_clear_inode has done it for us. */

	err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb),
			      &OCFS2_I(inode)->ip_data_lockres);
	if (err < 0)
		mlog_errno(err);

	status = err;

	err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb),
			      &OCFS2_I(inode)->ip_meta_lockres);
	if (err < 0)
		mlog_errno(err);
	if (err < 0 && !status)
		status = err;

	err = ocfs2_drop_lock(OCFS2_SB(inode->i_sb),
			      &OCFS2_I(inode)->ip_rw_lockres);
	if (err < 0)
		mlog_errno(err);
	if (err < 0 && !status)
		status = err;

	mlog_exit(status);
	return status;
}

static void ocfs2_prepare_downconvert(struct ocfs2_lock_res *lockres,
				      int new_level)
{
	assert_spin_locked(&lockres->l_lock);

	BUG_ON(lockres->l_blocking <= LKM_NLMODE);

	if (lockres->l_level <= new_level) {
		mlog(ML_ERROR, "lockres->l_level (%u) <= new_level (%u)\n",
		     lockres->l_level, new_level);
		BUG();
	}

	mlog(0, "lock %s, new_level = %d, l_blocking = %d\n",
	     lockres->l_name, new_level, lockres->l_blocking);

	lockres->l_action = OCFS2_AST_DOWNCONVERT;
	lockres->l_requested = new_level;
	lockres_or_flags(lockres, OCFS2_LOCK_BUSY);
}

static int ocfs2_downconvert_lock(struct ocfs2_super *osb,
				  struct ocfs2_lock_res *lockres,
				  int new_level,
				  int lvb)
{
	int ret, dlm_flags = LKM_CONVERT;
	enum dlm_status status;

	mlog_entry_void();

	if (lvb)
		dlm_flags |= LKM_VALBLK;

	status = dlmlock(osb->dlm,
			 new_level,
			 &lockres->l_lksb,
			 dlm_flags,
			 lockres->l_name,
			 OCFS2_LOCK_ID_MAX_LEN - 1,
			 ocfs2_locking_ast,
			 lockres,
			 ocfs2_blocking_ast);
	if (status != DLM_NORMAL) {
		ocfs2_log_dlm_error("dlmlock", status, lockres);
		ret = -EINVAL;
		ocfs2_recover_from_dlm_error(lockres, 1);
		goto bail;
	}

	ret = 0;
bail:
	mlog_exit(ret);
	return ret;
}

/* returns 1 when the caller should unlock and call dlmunlock */
static int ocfs2_prepare_cancel_convert(struct ocfs2_super *osb,
				        struct ocfs2_lock_res *lockres)
{
	assert_spin_locked(&lockres->l_lock);

	mlog_entry_void();
	mlog(0, "lock %s\n", lockres->l_name);

	if (lockres->l_unlock_action == OCFS2_UNLOCK_CANCEL_CONVERT) {
		/* If we're already trying to cancel a lock conversion
		 * then just drop the spinlock and allow the caller to
		 * requeue this lock. */

		mlog(0, "Lockres %s, skip convert\n", lockres->l_name);
		return 0;
	}

	/* were we in a convert when we got the bast fire? */
	BUG_ON(lockres->l_action != OCFS2_AST_CONVERT &&
	       lockres->l_action != OCFS2_AST_DOWNCONVERT);
	/* set things up for the unlockast to know to just
	 * clear out the ast_action and unset busy, etc. */
	lockres->l_unlock_action = OCFS2_UNLOCK_CANCEL_CONVERT;

	mlog_bug_on_msg(!(lockres->l_flags & OCFS2_LOCK_BUSY),
			"lock %s, invalid flags: 0x%lx\n",
			lockres->l_name, lockres->l_flags);

	return 1;
}

static int ocfs2_cancel_convert(struct ocfs2_super *osb,
				struct ocfs2_lock_res *lockres)
{
	int ret;
	enum dlm_status status;

	mlog_entry_void();
	mlog(0, "lock %s\n", lockres->l_name);

	ret = 0;
	status = dlmunlock(osb->dlm,
			   &lockres->l_lksb,
			   LKM_CANCEL,
			   ocfs2_unlock_ast,
			   lockres);
	if (status != DLM_NORMAL) {
		ocfs2_log_dlm_error("dlmunlock", status, lockres);
		ret = -EINVAL;
		ocfs2_recover_from_dlm_error(lockres, 0);
	}

	mlog(0, "lock %s return from dlmunlock\n", lockres->l_name);

	mlog_exit(ret);
	return ret;
}

static int ocfs2_unblock_lock(struct ocfs2_super *osb,
			      struct ocfs2_lock_res *lockres,
			      struct ocfs2_unblock_ctl *ctl)
{
	unsigned long flags;
	int blocking;
	int new_level;
	int ret = 0;
	int set_lvb = 0;

	mlog_entry_void();

	spin_lock_irqsave(&lockres->l_lock, flags);

	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));

recheck:
	if (lockres->l_flags & OCFS2_LOCK_BUSY) {
		ctl->requeue = 1;
		ret = ocfs2_prepare_cancel_convert(osb, lockres);
		spin_unlock_irqrestore(&lockres->l_lock, flags);
		if (ret) {
			ret = ocfs2_cancel_convert(osb, lockres);
			if (ret < 0)
				mlog_errno(ret);
		}
		goto leave;
	}

	/* if we're blocking an exclusive and we have *any* holders,
	 * then requeue. */
	if ((lockres->l_blocking == LKM_EXMODE)
	    && (lockres->l_ex_holders || lockres->l_ro_holders))
		goto leave_requeue;

	/* If it's a PR we're blocking, then only
	 * requeue if we've got any EX holders */
	if (lockres->l_blocking == LKM_PRMODE &&
	    lockres->l_ex_holders)
		goto leave_requeue;

	/*
	 * Can we get a lock in this state if the holder counts are
	 * zero? The meta data unblock code used to check this.
	 */
	if ((lockres->l_ops->flags & LOCK_TYPE_REQUIRES_REFRESH)
	    && (lockres->l_flags & OCFS2_LOCK_REFRESHING))
		goto leave_requeue;

	new_level = ocfs2_highest_compat_lock_level(lockres->l_blocking);

	if (lockres->l_ops->check_downconvert
	    && !lockres->l_ops->check_downconvert(lockres, new_level))
		goto leave_requeue;

	/* If we get here, then we know that there are no more
	 * incompatible holders (and anyone asking for an incompatible
	 * lock is blocked). We can now downconvert the lock */
	if (!lockres->l_ops->downconvert_worker)
		goto downconvert;

	/* Some lockres types want to do a bit of work before
	 * downconverting a lock. Allow that here. The worker function
	 * may sleep, so we save off a copy of what we're blocking as
	 * it may change while we're not holding the spin lock. */
	blocking = lockres->l_blocking;
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	ctl->unblock_action = lockres->l_ops->downconvert_worker(lockres, blocking);

	if (ctl->unblock_action == UNBLOCK_STOP_POST)
		goto leave;

	spin_lock_irqsave(&lockres->l_lock, flags);
	if (blocking != lockres->l_blocking) {
		/* If this changed underneath us, then we can't drop
		 * it just yet. */
		goto recheck;
	}

downconvert:
	ctl->requeue = 0;

	if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB) {
		if (lockres->l_level == LKM_EXMODE)
			set_lvb = 1;

		/*
		 * We only set the lvb if the lock has been fully
		 * refreshed - otherwise we risk setting stale
		 * data. Otherwise, there's no need to actually clear
		 * out the lvb here as it's value is still valid.
		 */
		if (set_lvb && !(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH))
			lockres->l_ops->set_lvb(lockres);
	}

	ocfs2_prepare_downconvert(lockres, new_level);
	spin_unlock_irqrestore(&lockres->l_lock, flags);
	ret = ocfs2_downconvert_lock(osb, lockres, new_level, set_lvb);
leave:
	mlog_exit(ret);
	return ret;

leave_requeue:
	spin_unlock_irqrestore(&lockres->l_lock, flags);
	ctl->requeue = 1;

	mlog_exit(0);
	return 0;
}

static int ocfs2_data_convert_worker(struct ocfs2_lock_res *lockres,
				     int blocking)
{
	struct inode *inode;
	struct address_space *mapping;

       	inode = ocfs2_lock_res_inode(lockres);
	mapping = inode->i_mapping;

	if (filemap_fdatawrite(mapping)) {
		mlog(ML_ERROR, "Could not sync inode %llu for downconvert!",
		     (unsigned long long)OCFS2_I(inode)->ip_blkno);
	}
	sync_mapping_buffers(mapping);
	if (blocking == LKM_EXMODE) {
		truncate_inode_pages(mapping, 0);
		unmap_mapping_range(mapping, 0, 0, 0);
	} else {
		/* We only need to wait on the I/O if we're not also
		 * truncating pages because truncate_inode_pages waits
		 * for us above. We don't truncate pages if we're
		 * blocking anything < EXMODE because we want to keep
		 * them around in that case. */
		filemap_fdatawait(mapping);
	}

	return UNBLOCK_CONTINUE;
}

static int ocfs2_check_meta_downconvert(struct ocfs2_lock_res *lockres,
					int new_level)
{
	struct inode *inode = ocfs2_lock_res_inode(lockres);
	int checkpointed = ocfs2_inode_fully_checkpointed(inode);

	BUG_ON(new_level != LKM_NLMODE && new_level != LKM_PRMODE);
	BUG_ON(lockres->l_level != LKM_EXMODE && !checkpointed);

	if (checkpointed)
		return 1;

	ocfs2_start_checkpoint(OCFS2_SB(inode->i_sb));
	return 0;
}

static void ocfs2_set_meta_lvb(struct ocfs2_lock_res *lockres)
{
	struct inode *inode = ocfs2_lock_res_inode(lockres);

	__ocfs2_stuff_meta_lvb(inode);
}

/*
 * Does the final reference drop on our dentry lock. Right now this
 * happens in the vote thread, but we could choose to simplify the
 * dlmglue API and push these off to the ocfs2_wq in the future.
 */
static void ocfs2_dentry_post_unlock(struct ocfs2_super *osb,
				     struct ocfs2_lock_res *lockres)
{
	struct ocfs2_dentry_lock *dl = ocfs2_lock_res_dl(lockres);
	ocfs2_dentry_lock_put(osb, dl);
}

/*
 * d_delete() matching dentries before the lock downconvert.
 *
 * At this point, any process waiting to destroy the
 * dentry_lock due to last ref count is stopped by the
 * OCFS2_LOCK_QUEUED flag.
 *
 * We have two potential problems
 *
 * 1) If we do the last reference drop on our dentry_lock (via dput)
 *    we'll wind up in ocfs2_release_dentry_lock(), waiting on
 *    the downconvert to finish. Instead we take an elevated
 *    reference and push the drop until after we've completed our
 *    unblock processing.
 *
 * 2) There might be another process with a final reference,
 *    waiting on us to finish processing. If this is the case, we
 *    detect it and exit out - there's no more dentries anyway.
 */
static int ocfs2_dentry_convert_worker(struct ocfs2_lock_res *lockres,
				       int blocking)
{
	struct ocfs2_dentry_lock *dl = ocfs2_lock_res_dl(lockres);
	struct ocfs2_inode_info *oi = OCFS2_I(dl->dl_inode);
	struct dentry *dentry;
	unsigned long flags;
	int extra_ref = 0;

	/*
	 * This node is blocking another node from getting a read
	 * lock. This happens when we've renamed within a
	 * directory. We've forced the other nodes to d_delete(), but
	 * we never actually dropped our lock because it's still
	 * valid. The downconvert code will retain a PR for this node,
	 * so there's no further work to do.
	 */
	if (blocking == LKM_PRMODE)
		return UNBLOCK_CONTINUE;

	/*
	 * Mark this inode as potentially orphaned. The code in
	 * ocfs2_delete_inode() will figure out whether it actually
	 * needs to be freed or not.
	 */
	spin_lock(&oi->ip_lock);
	oi->ip_flags |= OCFS2_INODE_MAYBE_ORPHANED;
	spin_unlock(&oi->ip_lock);

	/*
	 * Yuck. We need to make sure however that the check of
	 * OCFS2_LOCK_FREEING and the extra reference are atomic with
	 * respect to a reference decrement or the setting of that
	 * flag.
	 */
	spin_lock_irqsave(&lockres->l_lock, flags);
	spin_lock(&dentry_attach_lock);
	if (!(lockres->l_flags & OCFS2_LOCK_FREEING)
	    && dl->dl_count) {
		dl->dl_count++;
		extra_ref = 1;
	}
	spin_unlock(&dentry_attach_lock);
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	mlog(0, "extra_ref = %d\n", extra_ref);

	/*
	 * We have a process waiting on us in ocfs2_dentry_iput(),
	 * which means we can't have any more outstanding
	 * aliases. There's no need to do any more work.
	 */
	if (!extra_ref)
		return UNBLOCK_CONTINUE;

	spin_lock(&dentry_attach_lock);
	while (1) {
		dentry = ocfs2_find_local_alias(dl->dl_inode,
						dl->dl_parent_blkno, 1);
		if (!dentry)
			break;
		spin_unlock(&dentry_attach_lock);

		mlog(0, "d_delete(%.*s);\n", dentry->d_name.len,
		     dentry->d_name.name);

		/*
		 * The following dcache calls may do an
		 * iput(). Normally we don't want that from the
		 * downconverting thread, but in this case it's ok
		 * because the requesting node already has an
		 * exclusive lock on the inode, so it can't be queued
		 * for a downconvert.
		 */
		d_delete(dentry);
		dput(dentry);

		spin_lock(&dentry_attach_lock);
	}
	spin_unlock(&dentry_attach_lock);

	/*
	 * If we are the last holder of this dentry lock, there is no
	 * reason to downconvert so skip straight to the unlock.
	 */
	if (dl->dl_count == 1)
		return UNBLOCK_STOP_POST;

	return UNBLOCK_CONTINUE_POST;
}

void ocfs2_process_blocked_lock(struct ocfs2_super *osb,
				struct ocfs2_lock_res *lockres)
{
	int status;
	struct ocfs2_unblock_ctl ctl = {0, 0,};
	unsigned long flags;

	/* Our reference to the lockres in this function can be
	 * considered valid until we remove the OCFS2_LOCK_QUEUED
	 * flag. */

	mlog_entry_void();

	BUG_ON(!lockres);
	BUG_ON(!lockres->l_ops);

	mlog(0, "lockres %s blocked.\n", lockres->l_name);

	/* Detect whether a lock has been marked as going away while
	 * the vote thread was processing other things. A lock can
	 * still be marked with OCFS2_LOCK_FREEING after this check,
	 * but short circuiting here will still save us some
	 * performance. */
	spin_lock_irqsave(&lockres->l_lock, flags);
	if (lockres->l_flags & OCFS2_LOCK_FREEING)
		goto unqueue;
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	status = ocfs2_unblock_lock(osb, lockres, &ctl);
	if (status < 0)
		mlog_errno(status);

	spin_lock_irqsave(&lockres->l_lock, flags);
unqueue:
	if (lockres->l_flags & OCFS2_LOCK_FREEING || !ctl.requeue) {
		lockres_clear_flags(lockres, OCFS2_LOCK_QUEUED);
	} else
		ocfs2_schedule_blocked_lock(osb, lockres);

	mlog(0, "lockres %s, requeue = %s.\n", lockres->l_name,
	     ctl.requeue ? "yes" : "no");
	spin_unlock_irqrestore(&lockres->l_lock, flags);

	if (ctl.unblock_action != UNBLOCK_CONTINUE
	    && lockres->l_ops->post_unlock)
		lockres->l_ops->post_unlock(osb, lockres);

	mlog_exit_void();
}

static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb,
					struct ocfs2_lock_res *lockres)
{
	mlog_entry_void();

	assert_spin_locked(&lockres->l_lock);

	if (lockres->l_flags & OCFS2_LOCK_FREEING) {
		/* Do not schedule a lock for downconvert when it's on
		 * the way to destruction - any nodes wanting access
		 * to the resource will get it soon. */
		mlog(0, "Lockres %s won't be scheduled: flags 0x%lx\n",
		     lockres->l_name, lockres->l_flags);
		return;
	}

	lockres_or_flags(lockres, OCFS2_LOCK_QUEUED);

	spin_lock(&osb->vote_task_lock);
	if (list_empty(&lockres->l_blocked_list)) {
		list_add_tail(&lockres->l_blocked_list,
			      &osb->blocked_lock_list);
		osb->blocked_lock_count++;
	}
	spin_unlock(&osb->vote_task_lock);

	mlog_exit_void();
}

/* This aids in debugging situations where a bad LVB might be involved. */
void ocfs2_dump_meta_lvb_info(u64 level,
			      const char *function,
			      unsigned int line,
			      struct ocfs2_lock_res *lockres)
{
	struct ocfs2_meta_lvb *lvb = (struct ocfs2_meta_lvb *) lockres->l_lksb.lvb;

	mlog(level, "LVB information for %s (called from %s:%u):\n",
	     lockres->l_name, function, line);
	mlog(level, "version: %u, clusters: %u, generation: 0x%x\n",
	     lvb->lvb_version, be32_to_cpu(lvb->lvb_iclusters),
	     be32_to_cpu(lvb->lvb_igeneration));
	mlog(level, "size: %llu, uid %u, gid %u, mode 0x%x\n",
	     (unsigned long long)be64_to_cpu(lvb->lvb_isize),
	     be32_to_cpu(lvb->lvb_iuid), be32_to_cpu(lvb->lvb_igid),
	     be16_to_cpu(lvb->lvb_imode));
	mlog(level, "nlink %u, atime_packed 0x%llx, ctime_packed 0x%llx, "
	     "mtime_packed 0x%llx iattr 0x%x\n", be16_to_cpu(lvb->lvb_inlink),
	     (long long)be64_to_cpu(lvb->lvb_iatime_packed),
	     (long long)be64_to_cpu(lvb->lvb_ictime_packed),
	     (long long)be64_to_cpu(lvb->lvb_imtime_packed),
	     be32_to_cpu(lvb->lvb_iattr));
}