aboutsummaryrefslogtreecommitdiff
path: root/block/sheepdog.c
blob: c9236679c61c4098050648edc38fbce7f8a74f69 (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
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
/*
 * Copyright (C) 2009-2010 Nippon Telegraph and Telephone Corporation.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License version
 * 2 as published by the Free Software Foundation.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program. If not, see <http://www.gnu.org/licenses/>.
 *
 * Contributions after 2012-01-13 are licensed under the terms of the
 * GNU GPL, version 2 or (at your option) any later version.
 */

#include "qemu/osdep.h"
#include "qapi-visit.h"
#include "qapi/error.h"
#include "qapi/qmp/qdict.h"
#include "qapi/qobject-input-visitor.h"
#include "qemu/uri.h"
#include "qemu/error-report.h"
#include "qemu/sockets.h"
#include "block/block_int.h"
#include "sysemu/block-backend.h"
#include "qemu/bitops.h"
#include "qemu/cutils.h"

#define SD_PROTO_VER 0x01

#define SD_DEFAULT_ADDR "localhost"
#define SD_DEFAULT_PORT 7000

#define SD_OP_CREATE_AND_WRITE_OBJ  0x01
#define SD_OP_READ_OBJ       0x02
#define SD_OP_WRITE_OBJ      0x03
/* 0x04 is used internally by Sheepdog */

#define SD_OP_NEW_VDI        0x11
#define SD_OP_LOCK_VDI       0x12
#define SD_OP_RELEASE_VDI    0x13
#define SD_OP_GET_VDI_INFO   0x14
#define SD_OP_READ_VDIS      0x15
#define SD_OP_FLUSH_VDI      0x16
#define SD_OP_DEL_VDI        0x17
#define SD_OP_GET_CLUSTER_DEFAULT   0x18

#define SD_FLAG_CMD_WRITE    0x01
#define SD_FLAG_CMD_COW      0x02
#define SD_FLAG_CMD_CACHE    0x04 /* Writeback mode for cache */
#define SD_FLAG_CMD_DIRECT   0x08 /* Don't use cache */

#define SD_RES_SUCCESS       0x00 /* Success */
#define SD_RES_UNKNOWN       0x01 /* Unknown error */
#define SD_RES_NO_OBJ        0x02 /* No object found */
#define SD_RES_EIO           0x03 /* I/O error */
#define SD_RES_VDI_EXIST     0x04 /* Vdi exists already */
#define SD_RES_INVALID_PARMS 0x05 /* Invalid parameters */
#define SD_RES_SYSTEM_ERROR  0x06 /* System error */
#define SD_RES_VDI_LOCKED    0x07 /* Vdi is locked */
#define SD_RES_NO_VDI        0x08 /* No vdi found */
#define SD_RES_NO_BASE_VDI   0x09 /* No base vdi found */
#define SD_RES_VDI_READ      0x0A /* Cannot read requested vdi */
#define SD_RES_VDI_WRITE     0x0B /* Cannot write requested vdi */
#define SD_RES_BASE_VDI_READ 0x0C /* Cannot read base vdi */
#define SD_RES_BASE_VDI_WRITE   0x0D /* Cannot write base vdi */
#define SD_RES_NO_TAG        0x0E /* Requested tag is not found */
#define SD_RES_STARTUP       0x0F /* Sheepdog is on starting up */
#define SD_RES_VDI_NOT_LOCKED   0x10 /* Vdi is not locked */
#define SD_RES_SHUTDOWN      0x11 /* Sheepdog is shutting down */
#define SD_RES_NO_MEM        0x12 /* Cannot allocate memory */
#define SD_RES_FULL_VDI      0x13 /* we already have the maximum vdis */
#define SD_RES_VER_MISMATCH  0x14 /* Protocol version mismatch */
#define SD_RES_NO_SPACE      0x15 /* Server has no room for new objects */
#define SD_RES_WAIT_FOR_FORMAT  0x16 /* Waiting for a format operation */
#define SD_RES_WAIT_FOR_JOIN    0x17 /* Waiting for other nodes joining */
#define SD_RES_JOIN_FAILED   0x18 /* Target node had failed to join sheepdog */
#define SD_RES_HALT          0x19 /* Sheepdog is stopped serving IO request */
#define SD_RES_READONLY      0x1A /* Object is read-only */

/*
 * Object ID rules
 *
 *  0 - 19 (20 bits): data object space
 * 20 - 31 (12 bits): reserved data object space
 * 32 - 55 (24 bits): vdi object space
 * 56 - 59 ( 4 bits): reserved vdi object space
 * 60 - 63 ( 4 bits): object type identifier space
 */

#define VDI_SPACE_SHIFT   32
#define VDI_BIT (UINT64_C(1) << 63)
#define VMSTATE_BIT (UINT64_C(1) << 62)
#define MAX_DATA_OBJS (UINT64_C(1) << 20)
#define MAX_CHILDREN 1024
#define SD_MAX_VDI_LEN 256
#define SD_MAX_VDI_TAG_LEN 256
#define SD_NR_VDIS   (1U << 24)
#define SD_DATA_OBJ_SIZE (UINT64_C(1) << 22)
#define SD_MAX_VDI_SIZE (SD_DATA_OBJ_SIZE * MAX_DATA_OBJS)
#define SD_DEFAULT_BLOCK_SIZE_SHIFT 22
/*
 * For erasure coding, we use at most SD_EC_MAX_STRIP for data strips and
 * (SD_EC_MAX_STRIP - 1) for parity strips
 *
 * SD_MAX_COPIES is sum of number of data strips and parity strips.
 */
#define SD_EC_MAX_STRIP 16
#define SD_MAX_COPIES (SD_EC_MAX_STRIP * 2 - 1)

#define SD_INODE_SIZE (sizeof(SheepdogInode))
#define CURRENT_VDI_ID 0

#define LOCK_TYPE_NORMAL 0
#define LOCK_TYPE_SHARED 1      /* for iSCSI multipath */

typedef struct SheepdogReq {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint32_t opcode_specific[8];
} SheepdogReq;

typedef struct SheepdogRsp {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint32_t result;
    uint32_t opcode_specific[7];
} SheepdogRsp;

typedef struct SheepdogObjReq {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint64_t oid;
    uint64_t cow_oid;
    uint8_t copies;
    uint8_t copy_policy;
    uint8_t reserved[6];
    uint64_t offset;
} SheepdogObjReq;

typedef struct SheepdogObjRsp {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint32_t result;
    uint8_t copies;
    uint8_t copy_policy;
    uint8_t reserved[2];
    uint32_t pad[6];
} SheepdogObjRsp;

typedef struct SheepdogVdiReq {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint64_t vdi_size;
    uint32_t base_vdi_id;
    uint8_t copies;
    uint8_t copy_policy;
    uint8_t store_policy;
    uint8_t block_size_shift;
    uint32_t snapid;
    uint32_t type;
    uint32_t pad[2];
} SheepdogVdiReq;

typedef struct SheepdogVdiRsp {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint32_t result;
    uint32_t rsvd;
    uint32_t vdi_id;
    uint32_t pad[5];
} SheepdogVdiRsp;

typedef struct SheepdogClusterRsp {
    uint8_t proto_ver;
    uint8_t opcode;
    uint16_t flags;
    uint32_t epoch;
    uint32_t id;
    uint32_t data_length;
    uint32_t result;
    uint8_t nr_copies;
    uint8_t copy_policy;
    uint8_t block_size_shift;
    uint8_t __pad1;
    uint32_t __pad2[6];
} SheepdogClusterRsp;

typedef struct SheepdogInode {
    char name[SD_MAX_VDI_LEN];
    char tag[SD_MAX_VDI_TAG_LEN];
    uint64_t ctime;
    uint64_t snap_ctime;
    uint64_t vm_clock_nsec;
    uint64_t vdi_size;
    uint64_t vm_state_size;
    uint16_t copy_policy;
    uint8_t nr_copies;
    uint8_t block_size_shift;
    uint32_t snap_id;
    uint32_t vdi_id;
    uint32_t parent_vdi_id;
    uint32_t child_vdi_id[MAX_CHILDREN];
    uint32_t data_vdi_id[MAX_DATA_OBJS];
} SheepdogInode;

#define SD_INODE_HEADER_SIZE offsetof(SheepdogInode, data_vdi_id)

/*
 * 64 bit FNV-1a non-zero initial basis
 */
#define FNV1A_64_INIT ((uint64_t)0xcbf29ce484222325ULL)

/*
 * 64 bit Fowler/Noll/Vo FNV-1a hash code
 */
static inline uint64_t fnv_64a_buf(void *buf, size_t len, uint64_t hval)
{
    unsigned char *bp = buf;
    unsigned char *be = bp + len;
    while (bp < be) {
        hval ^= (uint64_t) *bp++;
        hval += (hval << 1) + (hval << 4) + (hval << 5) +
            (hval << 7) + (hval << 8) + (hval << 40);
    }
    return hval;
}

static inline bool is_data_obj_writable(SheepdogInode *inode, unsigned int idx)
{
    return inode->vdi_id == inode->data_vdi_id[idx];
}

static inline bool is_data_obj(uint64_t oid)
{
    return !(VDI_BIT & oid);
}

static inline uint64_t data_oid_to_idx(uint64_t oid)
{
    return oid & (MAX_DATA_OBJS - 1);
}

static inline uint32_t oid_to_vid(uint64_t oid)
{
    return (oid & ~VDI_BIT) >> VDI_SPACE_SHIFT;
}

static inline uint64_t vid_to_vdi_oid(uint32_t vid)
{
    return VDI_BIT | ((uint64_t)vid << VDI_SPACE_SHIFT);
}

static inline uint64_t vid_to_vmstate_oid(uint32_t vid, uint32_t idx)
{
    return VMSTATE_BIT | ((uint64_t)vid << VDI_SPACE_SHIFT) | idx;
}

static inline uint64_t vid_to_data_oid(uint32_t vid, uint32_t idx)
{
    return ((uint64_t)vid << VDI_SPACE_SHIFT) | idx;
}

static inline bool is_snapshot(struct SheepdogInode *inode)
{
    return !!inode->snap_ctime;
}

static inline size_t count_data_objs(const struct SheepdogInode *inode)
{
    return DIV_ROUND_UP(inode->vdi_size,
                        (1UL << inode->block_size_shift));
}

#undef DPRINTF
#ifdef DEBUG_SDOG
#define DEBUG_SDOG_PRINT 1
#else
#define DEBUG_SDOG_PRINT 0
#endif
#define DPRINTF(fmt, args...)                                           \
    do {                                                                \
        if (DEBUG_SDOG_PRINT) {                                         \
            fprintf(stderr, "%s %d: " fmt, __func__, __LINE__, ##args); \
        }                                                               \
    } while (0)

typedef struct SheepdogAIOCB SheepdogAIOCB;
typedef struct BDRVSheepdogState BDRVSheepdogState;

typedef struct AIOReq {
    SheepdogAIOCB *aiocb;
    unsigned int iov_offset;

    uint64_t oid;
    uint64_t base_oid;
    uint64_t offset;
    unsigned int data_len;
    uint8_t flags;
    uint32_t id;
    bool create;

    QLIST_ENTRY(AIOReq) aio_siblings;
} AIOReq;

enum AIOCBState {
    AIOCB_WRITE_UDATA,
    AIOCB_READ_UDATA,
    AIOCB_FLUSH_CACHE,
    AIOCB_DISCARD_OBJ,
};

#define AIOCBOverlapping(x, y)                                 \
    (!(x->max_affect_data_idx < y->min_affect_data_idx          \
       || y->max_affect_data_idx < x->min_affect_data_idx))

struct SheepdogAIOCB {
    BDRVSheepdogState *s;

    QEMUIOVector *qiov;

    int64_t sector_num;
    int nb_sectors;

    int ret;
    enum AIOCBState aiocb_type;

    Coroutine *coroutine;
    int nr_pending;

    uint32_t min_affect_data_idx;
    uint32_t max_affect_data_idx;

    /*
     * The difference between affect_data_idx and dirty_data_idx:
     * affect_data_idx represents range of index of all request types.
     * dirty_data_idx represents range of index updated by COW requests.
     * dirty_data_idx is used for updating an inode object.
     */
    uint32_t min_dirty_data_idx;
    uint32_t max_dirty_data_idx;

    QLIST_ENTRY(SheepdogAIOCB) aiocb_siblings;
};

struct BDRVSheepdogState {
    BlockDriverState *bs;
    AioContext *aio_context;

    SheepdogInode inode;

    char name[SD_MAX_VDI_LEN];
    bool is_snapshot;
    uint32_t cache_flags;
    bool discard_supported;

    SocketAddress *addr;
    int fd;

    CoMutex lock;
    Coroutine *co_send;
    Coroutine *co_recv;

    uint32_t aioreq_seq_num;

    /* Every aio request must be linked to either of these queues. */
    QLIST_HEAD(inflight_aio_head, AIOReq) inflight_aio_head;
    QLIST_HEAD(failed_aio_head, AIOReq) failed_aio_head;

    CoQueue overlapping_queue;
    QLIST_HEAD(inflight_aiocb_head, SheepdogAIOCB) inflight_aiocb_head;
};

typedef struct BDRVSheepdogReopenState {
    int fd;
    int cache_flags;
} BDRVSheepdogReopenState;

static const char * sd_strerror(int err)
{
    int i;

    static const struct {
        int err;
        const char *desc;
    } errors[] = {
        {SD_RES_SUCCESS, "Success"},
        {SD_RES_UNKNOWN, "Unknown error"},
        {SD_RES_NO_OBJ, "No object found"},
        {SD_RES_EIO, "I/O error"},
        {SD_RES_VDI_EXIST, "VDI exists already"},
        {SD_RES_INVALID_PARMS, "Invalid parameters"},
        {SD_RES_SYSTEM_ERROR, "System error"},
        {SD_RES_VDI_LOCKED, "VDI is already locked"},
        {SD_RES_NO_VDI, "No vdi found"},
        {SD_RES_NO_BASE_VDI, "No base VDI found"},
        {SD_RES_VDI_READ, "Failed read the requested VDI"},
        {SD_RES_VDI_WRITE, "Failed to write the requested VDI"},
        {SD_RES_BASE_VDI_READ, "Failed to read the base VDI"},
        {SD_RES_BASE_VDI_WRITE, "Failed to write the base VDI"},
        {SD_RES_NO_TAG, "Failed to find the requested tag"},
        {SD_RES_STARTUP, "The system is still booting"},
        {SD_RES_VDI_NOT_LOCKED, "VDI isn't locked"},
        {SD_RES_SHUTDOWN, "The system is shutting down"},
        {SD_RES_NO_MEM, "Out of memory on the server"},
        {SD_RES_FULL_VDI, "We already have the maximum vdis"},
        {SD_RES_VER_MISMATCH, "Protocol version mismatch"},
        {SD_RES_NO_SPACE, "Server has no space for new objects"},
        {SD_RES_WAIT_FOR_FORMAT, "Sheepdog is waiting for a format operation"},
        {SD_RES_WAIT_FOR_JOIN, "Sheepdog is waiting for other nodes joining"},
        {SD_RES_JOIN_FAILED, "Target node had failed to join sheepdog"},
        {SD_RES_HALT, "Sheepdog is stopped serving IO request"},
        {SD_RES_READONLY, "Object is read-only"},
    };

    for (i = 0; i < ARRAY_SIZE(errors); ++i) {
        if (errors[i].err == err) {
            return errors[i].desc;
        }
    }

    return "Invalid error code";
}

/*
 * Sheepdog I/O handling:
 *
 * 1. In sd_co_rw_vector, we send the I/O requests to the server and
 *    link the requests to the inflight_list in the
 *    BDRVSheepdogState.  The function yields while waiting for
 *    receiving the response.
 *
 * 2. We receive the response in aio_read_response, the fd handler to
 *    the sheepdog connection.  We switch back to sd_co_readv/sd_writev
 *    after all the requests belonging to the AIOCB are finished.  If
 *    needed, sd_co_writev will send another requests for the vdi object.
 */

static inline AIOReq *alloc_aio_req(BDRVSheepdogState *s, SheepdogAIOCB *acb,
                                    uint64_t oid, unsigned int data_len,
                                    uint64_t offset, uint8_t flags, bool create,
                                    uint64_t base_oid, unsigned int iov_offset)
{
    AIOReq *aio_req;

    aio_req = g_malloc(sizeof(*aio_req));
    aio_req->aiocb = acb;
    aio_req->iov_offset = iov_offset;
    aio_req->oid = oid;
    aio_req->base_oid = base_oid;
    aio_req->offset = offset;
    aio_req->data_len = data_len;
    aio_req->flags = flags;
    aio_req->id = s->aioreq_seq_num++;
    aio_req->create = create;

    acb->nr_pending++;
    return aio_req;
}

static void wait_for_overlapping_aiocb(BDRVSheepdogState *s, SheepdogAIOCB *acb)
{
    SheepdogAIOCB *cb;

retry:
    QLIST_FOREACH(cb, &s->inflight_aiocb_head, aiocb_siblings) {
        if (AIOCBOverlapping(acb, cb)) {
            qemu_co_queue_wait(&s->overlapping_queue, NULL);
            goto retry;
        }
    }
}

static void sd_aio_setup(SheepdogAIOCB *acb, BDRVSheepdogState *s,
                         QEMUIOVector *qiov, int64_t sector_num, int nb_sectors,
                         int type)
{
    uint32_t object_size;

    object_size = (UINT32_C(1) << s->inode.block_size_shift);

    acb->s = s;

    acb->qiov = qiov;

    acb->sector_num = sector_num;
    acb->nb_sectors = nb_sectors;

    acb->coroutine = qemu_coroutine_self();
    acb->ret = 0;
    acb->nr_pending = 0;

    acb->min_affect_data_idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size;
    acb->max_affect_data_idx = (acb->sector_num * BDRV_SECTOR_SIZE +
                              acb->nb_sectors * BDRV_SECTOR_SIZE) / object_size;

    acb->min_dirty_data_idx = UINT32_MAX;
    acb->max_dirty_data_idx = 0;
    acb->aiocb_type = type;

    if (type == AIOCB_FLUSH_CACHE) {
        return;
    }

    wait_for_overlapping_aiocb(s, acb);
    QLIST_INSERT_HEAD(&s->inflight_aiocb_head, acb, aiocb_siblings);
}

static SocketAddress *sd_socket_address(const char *path,
                                        const char *host, const char *port)
{
    SocketAddress *addr = g_new0(SocketAddress, 1);

    if (path) {
        addr->type = SOCKET_ADDRESS_TYPE_UNIX;
        addr->u.q_unix.path = g_strdup(path);
    } else {
        addr->type = SOCKET_ADDRESS_TYPE_INET;
        addr->u.inet.host = g_strdup(host ?: SD_DEFAULT_ADDR);
        addr->u.inet.port = g_strdup(port ?: stringify(SD_DEFAULT_PORT));
    }

    return addr;
}

static SocketAddress *sd_server_config(QDict *options, Error **errp)
{
    QDict *server = NULL;
    QObject *crumpled_server = NULL;
    Visitor *iv = NULL;
    SocketAddress *saddr = NULL;
    Error *local_err = NULL;

    qdict_extract_subqdict(options, &server, "server.");

    crumpled_server = qdict_crumple(server, errp);
    if (!crumpled_server) {
        goto done;
    }

    /*
     * FIXME .numeric, .to, .ipv4 or .ipv6 don't work with -drive
     * server.type=inet.  .to doesn't matter, it's ignored anyway.
     * That's because when @options come from -blockdev or
     * blockdev_add, members are typed according to the QAPI schema,
     * but when they come from -drive, they're all QString.  The
     * visitor expects the former.
     */
    iv = qobject_input_visitor_new(crumpled_server);
    visit_type_SocketAddress(iv, NULL, &saddr, &local_err);
    if (local_err) {
        error_propagate(errp, local_err);
        goto done;
    }

done:
    visit_free(iv);
    qobject_decref(crumpled_server);
    QDECREF(server);
    return saddr;
}

/* Return -EIO in case of error, file descriptor on success */
static int connect_to_sdog(BDRVSheepdogState *s, Error **errp)
{
    int fd;

    fd = socket_connect(s->addr, NULL, NULL, errp);

    if (s->addr->type == SOCKET_ADDRESS_TYPE_INET && fd >= 0) {
        int ret = socket_set_nodelay(fd);
        if (ret < 0) {
            error_report("%s", strerror(errno));
        }
    }

    if (fd >= 0) {
        qemu_set_nonblock(fd);
    } else {
        fd = -EIO;
    }

    return fd;
}

/* Return 0 on success and -errno in case of error */
static coroutine_fn int send_co_req(int sockfd, SheepdogReq *hdr, void *data,
                                    unsigned int *wlen)
{
    int ret;

    ret = qemu_co_send(sockfd, hdr, sizeof(*hdr));
    if (ret != sizeof(*hdr)) {
        error_report("failed to send a req, %s", strerror(errno));
        return -errno;
    }

    ret = qemu_co_send(sockfd, data, *wlen);
    if (ret != *wlen) {
        error_report("failed to send a req, %s", strerror(errno));
        return -errno;
    }

    return ret;
}

typedef struct SheepdogReqCo {
    int sockfd;
    BlockDriverState *bs;
    AioContext *aio_context;
    SheepdogReq *hdr;
    void *data;
    unsigned int *wlen;
    unsigned int *rlen;
    int ret;
    bool finished;
    Coroutine *co;
} SheepdogReqCo;

static void restart_co_req(void *opaque)
{
    SheepdogReqCo *srco = opaque;

    aio_co_wake(srco->co);
}

static coroutine_fn void do_co_req(void *opaque)
{
    int ret;
    SheepdogReqCo *srco = opaque;
    int sockfd = srco->sockfd;
    SheepdogReq *hdr = srco->hdr;
    void *data = srco->data;
    unsigned int *wlen = srco->wlen;
    unsigned int *rlen = srco->rlen;

    srco->co = qemu_coroutine_self();
    aio_set_fd_handler(srco->aio_context, sockfd, false,
                       NULL, restart_co_req, NULL, srco);

    ret = send_co_req(sockfd, hdr, data, wlen);
    if (ret < 0) {
        goto out;
    }

    aio_set_fd_handler(srco->aio_context, sockfd, false,
                       restart_co_req, NULL, NULL, srco);

    ret = qemu_co_recv(sockfd, hdr, sizeof(*hdr));
    if (ret != sizeof(*hdr)) {
        error_report("failed to get a rsp, %s", strerror(errno));
        ret = -errno;
        goto out;
    }

    if (*rlen > hdr->data_length) {
        *rlen = hdr->data_length;
    }

    if (*rlen) {
        ret = qemu_co_recv(sockfd, data, *rlen);
        if (ret != *rlen) {
            error_report("failed to get the data, %s", strerror(errno));
            ret = -errno;
            goto out;
        }
    }
    ret = 0;
out:
    /* there is at most one request for this sockfd, so it is safe to
     * set each handler to NULL. */
    aio_set_fd_handler(srco->aio_context, sockfd, false,
                       NULL, NULL, NULL, NULL);

    srco->co = NULL;
    srco->ret = ret;
    /* Set srco->finished before reading bs->wakeup.  */
    atomic_mb_set(&srco->finished, true);
    if (srco->bs) {
        bdrv_wakeup(srco->bs);
    }
}

/*
 * Send the request to the sheep in a synchronous manner.
 *
 * Return 0 on success, -errno in case of error.
 */
static int do_req(int sockfd, BlockDriverState *bs, SheepdogReq *hdr,
                  void *data, unsigned int *wlen, unsigned int *rlen)
{
    Coroutine *co;
    SheepdogReqCo srco = {
        .sockfd = sockfd,
        .aio_context = bs ? bdrv_get_aio_context(bs) : qemu_get_aio_context(),
        .bs = bs,
        .hdr = hdr,
        .data = data,
        .wlen = wlen,
        .rlen = rlen,
        .ret = 0,
        .finished = false,
    };

    if (qemu_in_coroutine()) {
        do_co_req(&srco);
    } else {
        co = qemu_coroutine_create(do_co_req, &srco);
        if (bs) {
            bdrv_coroutine_enter(bs, co);
            BDRV_POLL_WHILE(bs, !srco.finished);
        } else {
            qemu_coroutine_enter(co);
            while (!srco.finished) {
                aio_poll(qemu_get_aio_context(), true);
            }
        }
    }

    return srco.ret;
}

static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
                                         struct iovec *iov, int niov,
                                         enum AIOCBState aiocb_type);
static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req);
static int reload_inode(BDRVSheepdogState *s, uint32_t snapid, const char *tag);
static int get_sheep_fd(BDRVSheepdogState *s, Error **errp);
static void co_write_request(void *opaque);

static coroutine_fn void reconnect_to_sdog(void *opaque)
{
    BDRVSheepdogState *s = opaque;
    AIOReq *aio_req, *next;

    aio_set_fd_handler(s->aio_context, s->fd, false, NULL,
                       NULL, NULL, NULL);
    close(s->fd);
    s->fd = -1;

    /* Wait for outstanding write requests to be completed. */
    while (s->co_send != NULL) {
        co_write_request(opaque);
    }

    /* Try to reconnect the sheepdog server every one second. */
    while (s->fd < 0) {
        Error *local_err = NULL;
        s->fd = get_sheep_fd(s, &local_err);
        if (s->fd < 0) {
            DPRINTF("Wait for connection to be established\n");
            error_report_err(local_err);
            co_aio_sleep_ns(bdrv_get_aio_context(s->bs), QEMU_CLOCK_REALTIME,
                            1000000000ULL);
        }
    };

    /*
     * Now we have to resend all the request in the inflight queue.  However,
     * resend_aioreq() can yield and newly created requests can be added to the
     * inflight queue before the coroutine is resumed.  To avoid mixing them, we
     * have to move all the inflight requests to the failed queue before
     * resend_aioreq() is called.
     */
    QLIST_FOREACH_SAFE(aio_req, &s->inflight_aio_head, aio_siblings, next) {
        QLIST_REMOVE(aio_req, aio_siblings);
        QLIST_INSERT_HEAD(&s->failed_aio_head, aio_req, aio_siblings);
    }

    /* Resend all the failed aio requests. */
    while (!QLIST_EMPTY(&s->failed_aio_head)) {
        aio_req = QLIST_FIRST(&s->failed_aio_head);
        QLIST_REMOVE(aio_req, aio_siblings);
        resend_aioreq(s, aio_req);
    }
}

/*
 * Receive responses of the I/O requests.
 *
 * This function is registered as a fd handler, and called from the
 * main loop when s->fd is ready for reading responses.
 */
static void coroutine_fn aio_read_response(void *opaque)
{
    SheepdogObjRsp rsp;
    BDRVSheepdogState *s = opaque;
    int fd = s->fd;
    int ret;
    AIOReq *aio_req = NULL;
    SheepdogAIOCB *acb;
    uint64_t idx;

    /* read a header */
    ret = qemu_co_recv(fd, &rsp, sizeof(rsp));
    if (ret != sizeof(rsp)) {
        error_report("failed to get the header, %s", strerror(errno));
        goto err;
    }

    /* find the right aio_req from the inflight aio list */
    QLIST_FOREACH(aio_req, &s->inflight_aio_head, aio_siblings) {
        if (aio_req->id == rsp.id) {
            break;
        }
    }
    if (!aio_req) {
        error_report("cannot find aio_req %x", rsp.id);
        goto err;
    }

    acb = aio_req->aiocb;

    switch (acb->aiocb_type) {
    case AIOCB_WRITE_UDATA:
        if (!is_data_obj(aio_req->oid)) {
            break;
        }
        idx = data_oid_to_idx(aio_req->oid);

        if (aio_req->create) {
            /*
             * If the object is newly created one, we need to update
             * the vdi object (metadata object).  min_dirty_data_idx
             * and max_dirty_data_idx are changed to include updated
             * index between them.
             */
            if (rsp.result == SD_RES_SUCCESS) {
                s->inode.data_vdi_id[idx] = s->inode.vdi_id;
                acb->max_dirty_data_idx = MAX(idx, acb->max_dirty_data_idx);
                acb->min_dirty_data_idx = MIN(idx, acb->min_dirty_data_idx);
            }
        }
        break;
    case AIOCB_READ_UDATA:
        ret = qemu_co_recvv(fd, acb->qiov->iov, acb->qiov->niov,
                            aio_req->iov_offset, rsp.data_length);
        if (ret != rsp.data_length) {
            error_report("failed to get the data, %s", strerror(errno));
            goto err;
        }
        break;
    case AIOCB_FLUSH_CACHE:
        if (rsp.result == SD_RES_INVALID_PARMS) {
            DPRINTF("disable cache since the server doesn't support it\n");
            s->cache_flags = SD_FLAG_CMD_DIRECT;
            rsp.result = SD_RES_SUCCESS;
        }
        break;
    case AIOCB_DISCARD_OBJ:
        switch (rsp.result) {
        case SD_RES_INVALID_PARMS:
            error_report("server doesn't support discard command");
            rsp.result = SD_RES_SUCCESS;
            s->discard_supported = false;
            break;
        default:
            break;
        }
    }

    /* No more data for this aio_req (reload_inode below uses its own file
     * descriptor handler which doesn't use co_recv).
    */
    s->co_recv = NULL;

    QLIST_REMOVE(aio_req, aio_siblings);
    switch (rsp.result) {
    case SD_RES_SUCCESS:
        break;
    case SD_RES_READONLY:
        if (s->inode.vdi_id == oid_to_vid(aio_req->oid)) {
            ret = reload_inode(s, 0, "");
            if (ret < 0) {
                goto err;
            }
        }
        if (is_data_obj(aio_req->oid)) {
            aio_req->oid = vid_to_data_oid(s->inode.vdi_id,
                                           data_oid_to_idx(aio_req->oid));
        } else {
            aio_req->oid = vid_to_vdi_oid(s->inode.vdi_id);
        }
        resend_aioreq(s, aio_req);
        return;
    default:
        acb->ret = -EIO;
        error_report("%s", sd_strerror(rsp.result));
        break;
    }

    g_free(aio_req);

    if (!--acb->nr_pending) {
        /*
         * We've finished all requests which belong to the AIOCB, so
         * we can switch back to sd_co_readv/writev now.
         */
        aio_co_wake(acb->coroutine);
    }

    return;

err:
    reconnect_to_sdog(opaque);
}

static void co_read_response(void *opaque)
{
    BDRVSheepdogState *s = opaque;

    if (!s->co_recv) {
        s->co_recv = qemu_coroutine_create(aio_read_response, opaque);
    }

    aio_co_enter(s->aio_context, s->co_recv);
}

static void co_write_request(void *opaque)
{
    BDRVSheepdogState *s = opaque;

    aio_co_wake(s->co_send);
}

/*
 * Return a socket descriptor to read/write objects.
 *
 * We cannot use this descriptor for other operations because
 * the block driver may be on waiting response from the server.
 */
static int get_sheep_fd(BDRVSheepdogState *s, Error **errp)
{
    int fd;

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        return fd;
    }

    aio_set_fd_handler(s->aio_context, fd, false,
                       co_read_response, NULL, NULL, s);
    return fd;
}

/*
 * Parse numeric snapshot ID in @str
 * If @str can't be parsed as number, return false.
 * Else, if the number is zero or too large, set *@snapid to zero and
 * return true.
 * Else, set *@snapid to the number and return true.
 */
static bool sd_parse_snapid(const char *str, uint32_t *snapid)
{
    unsigned long ul;
    int ret;

    ret = qemu_strtoul(str, NULL, 10, &ul);
    if (ret == -ERANGE) {
        ul = ret = 0;
    }
    if (ret) {
        return false;
    }
    if (ul > UINT32_MAX) {
        ul = 0;
    }

    *snapid = ul;
    return true;
}

static bool sd_parse_snapid_or_tag(const char *str,
                                   uint32_t *snapid, char tag[])
{
    if (!sd_parse_snapid(str, snapid)) {
        *snapid = 0;
        if (g_strlcpy(tag, str, SD_MAX_VDI_TAG_LEN) >= SD_MAX_VDI_TAG_LEN) {
            return false;
        }
    } else if (!*snapid) {
        return false;
    } else {
        tag[0] = 0;
    }
    return true;
}

typedef struct {
    const char *path;           /* non-null iff transport is tcp */
    const char *host;           /* valid when transport is tcp */
    int port;                   /* valid when transport is tcp */
    char vdi[SD_MAX_VDI_LEN];
    char tag[SD_MAX_VDI_TAG_LEN];
    uint32_t snap_id;
    /* Remainder is only for sd_config_done() */
    URI *uri;
    QueryParams *qp;
} SheepdogConfig;

static void sd_config_done(SheepdogConfig *cfg)
{
    if (cfg->qp) {
        query_params_free(cfg->qp);
    }
    uri_free(cfg->uri);
}

static void sd_parse_uri(SheepdogConfig *cfg, const char *filename,
                         Error **errp)
{
    Error *err = NULL;
    QueryParams *qp = NULL;
    bool is_unix;
    URI *uri;

    memset(cfg, 0, sizeof(*cfg));

    cfg->uri = uri = uri_parse(filename);
    if (!uri) {
        error_setg(&err, "invalid URI");
        goto out;
    }

    /* transport */
    if (!strcmp(uri->scheme, "sheepdog")) {
        is_unix = false;
    } else if (!strcmp(uri->scheme, "sheepdog+tcp")) {
        is_unix = false;
    } else if (!strcmp(uri->scheme, "sheepdog+unix")) {
        is_unix = true;
    } else {
        error_setg(&err, "URI scheme must be 'sheepdog', 'sheepdog+tcp',"
                   " or 'sheepdog+unix'");
        goto out;
    }

    if (uri->path == NULL || !strcmp(uri->path, "/")) {
        error_setg(&err, "missing file path in URI");
        goto out;
    }
    if (g_strlcpy(cfg->vdi, uri->path + 1, SD_MAX_VDI_LEN)
        >= SD_MAX_VDI_LEN) {
        error_setg(&err, "VDI name is too long");
        goto out;
    }

    cfg->qp = qp = query_params_parse(uri->query);

    if (is_unix) {
        /* sheepdog+unix:///vdiname?socket=path */
        if (uri->server || uri->port) {
            error_setg(&err, "URI scheme %s doesn't accept a server address",
                       uri->scheme);
            goto out;
        }
        if (!qp->n) {
            error_setg(&err,
                       "URI scheme %s requires query parameter 'socket'",
                       uri->scheme);
            goto out;
        }
        if (qp->n != 1 || strcmp(qp->p[0].name, "socket")) {
            error_setg(&err, "unexpected query parameters");
            goto out;
        }
        cfg->path = qp->p[0].value;
    } else {
        /* sheepdog[+tcp]://[host:port]/vdiname */
        if (qp->n) {
            error_setg(&err, "unexpected query parameters");
            goto out;
        }
        cfg->host = uri->server;
        cfg->port = uri->port;
    }

    /* snapshot tag */
    if (uri->fragment) {
        if (!sd_parse_snapid_or_tag(uri->fragment,
                                    &cfg->snap_id, cfg->tag)) {
            error_setg(&err, "'%s' is not a valid snapshot ID",
                       uri->fragment);
            goto out;
        }
    } else {
        cfg->snap_id = CURRENT_VDI_ID; /* search current vdi */
    }

out:
    if (err) {
        error_propagate(errp, err);
        sd_config_done(cfg);
    }
}

/*
 * Parse a filename (old syntax)
 *
 * filename must be one of the following formats:
 *   1. [vdiname]
 *   2. [vdiname]:[snapid]
 *   3. [vdiname]:[tag]
 *   4. [hostname]:[port]:[vdiname]
 *   5. [hostname]:[port]:[vdiname]:[snapid]
 *   6. [hostname]:[port]:[vdiname]:[tag]
 *
 * You can boot from the snapshot images by specifying `snapid` or
 * `tag'.
 *
 * You can run VMs outside the Sheepdog cluster by specifying
 * `hostname' and `port' (experimental).
 */
static void parse_vdiname(SheepdogConfig *cfg, const char *filename,
                          Error **errp)
{
    Error *err = NULL;
    char *p, *q, *uri;
    const char *host_spec, *vdi_spec;
    int nr_sep;

    strstart(filename, "sheepdog:", &filename);
    p = q = g_strdup(filename);

    /* count the number of separators */
    nr_sep = 0;
    while (*p) {
        if (*p == ':') {
            nr_sep++;
        }
        p++;
    }
    p = q;

    /* use the first two tokens as host_spec. */
    if (nr_sep >= 2) {
        host_spec = p;
        p = strchr(p, ':');
        p++;
        p = strchr(p, ':');
        *p++ = '\0';
    } else {
        host_spec = "";
    }

    vdi_spec = p;

    p = strchr(vdi_spec, ':');
    if (p) {
        *p++ = '#';
    }

    uri = g_strdup_printf("sheepdog://%s/%s", host_spec, vdi_spec);

    /*
     * FIXME We to escape URI meta-characters, e.g. "x?y=z"
     * produces "sheepdog://x?y=z".  Because of that ...
     */
    sd_parse_uri(cfg, uri, &err);
    if (err) {
        /*
         * ... this can fail, but the error message is misleading.
         * Replace it by the traditional useless one until the
         * escaping is fixed.
         */
        error_free(err);
        error_setg(errp, "Can't parse filename");
    }

    g_free(q);
    g_free(uri);
}

static void sd_parse_filename(const char *filename, QDict *options,
                              Error **errp)
{
    Error *err = NULL;
    SheepdogConfig cfg;
    char buf[32];

    if (strstr(filename, "://")) {
        sd_parse_uri(&cfg, filename, &err);
    } else {
        parse_vdiname(&cfg, filename, &err);
    }
    if (err) {
        error_propagate(errp, err);
        return;
    }

    if (cfg.path) {
        qdict_set_default_str(options, "server.path", cfg.path);
        qdict_set_default_str(options, "server.type", "unix");
    } else {
        qdict_set_default_str(options, "server.type", "inet");
        qdict_set_default_str(options, "server.host",
                              cfg.host ?: SD_DEFAULT_ADDR);
        snprintf(buf, sizeof(buf), "%d", cfg.port ?: SD_DEFAULT_PORT);
        qdict_set_default_str(options, "server.port", buf);
    }
    qdict_set_default_str(options, "vdi", cfg.vdi);
    qdict_set_default_str(options, "tag", cfg.tag);
    if (cfg.snap_id) {
        snprintf(buf, sizeof(buf), "%d", cfg.snap_id);
        qdict_set_default_str(options, "snap-id", buf);
    }

    sd_config_done(&cfg);
}

static int find_vdi_name(BDRVSheepdogState *s, const char *filename,
                         uint32_t snapid, const char *tag, uint32_t *vid,
                         bool lock, Error **errp)
{
    int ret, fd;
    SheepdogVdiReq hdr;
    SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
    unsigned int wlen, rlen = 0;
    char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN];

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        return fd;
    }

    /* This pair of strncpy calls ensures that the buffer is zero-filled,
     * which is desirable since we'll soon be sending those bytes, and
     * don't want the send_req to read uninitialized data.
     */
    strncpy(buf, filename, SD_MAX_VDI_LEN);
    strncpy(buf + SD_MAX_VDI_LEN, tag, SD_MAX_VDI_TAG_LEN);

    memset(&hdr, 0, sizeof(hdr));
    if (lock) {
        hdr.opcode = SD_OP_LOCK_VDI;
        hdr.type = LOCK_TYPE_NORMAL;
    } else {
        hdr.opcode = SD_OP_GET_VDI_INFO;
    }
    wlen = SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN;
    hdr.proto_ver = SD_PROTO_VER;
    hdr.data_length = wlen;
    hdr.snapid = snapid;
    hdr.flags = SD_FLAG_CMD_WRITE;

    ret = do_req(fd, s->bs, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
    if (ret) {
        error_setg_errno(errp, -ret, "cannot get vdi info");
        goto out;
    }

    if (rsp->result != SD_RES_SUCCESS) {
        error_setg(errp, "cannot get vdi info, %s, %s %" PRIu32 " %s",
                   sd_strerror(rsp->result), filename, snapid, tag);
        if (rsp->result == SD_RES_NO_VDI) {
            ret = -ENOENT;
        } else if (rsp->result == SD_RES_VDI_LOCKED) {
            ret = -EBUSY;
        } else {
            ret = -EIO;
        }
        goto out;
    }
    *vid = rsp->vdi_id;

    ret = 0;
out:
    closesocket(fd);
    return ret;
}

static void coroutine_fn add_aio_request(BDRVSheepdogState *s, AIOReq *aio_req,
                                         struct iovec *iov, int niov,
                                         enum AIOCBState aiocb_type)
{
    int nr_copies = s->inode.nr_copies;
    SheepdogObjReq hdr;
    unsigned int wlen = 0;
    int ret;
    uint64_t oid = aio_req->oid;
    unsigned int datalen = aio_req->data_len;
    uint64_t offset = aio_req->offset;
    uint8_t flags = aio_req->flags;
    uint64_t old_oid = aio_req->base_oid;
    bool create = aio_req->create;

    QLIST_INSERT_HEAD(&s->inflight_aio_head, aio_req, aio_siblings);

    if (!nr_copies) {
        error_report("bug");
    }

    memset(&hdr, 0, sizeof(hdr));

    switch (aiocb_type) {
    case AIOCB_FLUSH_CACHE:
        hdr.opcode = SD_OP_FLUSH_VDI;
        break;
    case AIOCB_READ_UDATA:
        hdr.opcode = SD_OP_READ_OBJ;
        hdr.flags = flags;
        break;
    case AIOCB_WRITE_UDATA:
        if (create) {
            hdr.opcode = SD_OP_CREATE_AND_WRITE_OBJ;
        } else {
            hdr.opcode = SD_OP_WRITE_OBJ;
        }
        wlen = datalen;
        hdr.flags = SD_FLAG_CMD_WRITE | flags;
        break;
    case AIOCB_DISCARD_OBJ:
        hdr.opcode = SD_OP_WRITE_OBJ;
        hdr.flags = SD_FLAG_CMD_WRITE | flags;
        s->inode.data_vdi_id[data_oid_to_idx(oid)] = 0;
        offset = offsetof(SheepdogInode,
                          data_vdi_id[data_oid_to_idx(oid)]);
        oid = vid_to_vdi_oid(s->inode.vdi_id);
        wlen = datalen = sizeof(uint32_t);
        break;
    }

    if (s->cache_flags) {
        hdr.flags |= s->cache_flags;
    }

    hdr.oid = oid;
    hdr.cow_oid = old_oid;
    hdr.copies = s->inode.nr_copies;

    hdr.data_length = datalen;
    hdr.offset = offset;

    hdr.id = aio_req->id;

    qemu_co_mutex_lock(&s->lock);
    s->co_send = qemu_coroutine_self();
    aio_set_fd_handler(s->aio_context, s->fd, false,
                       co_read_response, co_write_request, NULL, s);
    socket_set_cork(s->fd, 1);

    /* send a header */
    ret = qemu_co_send(s->fd, &hdr, sizeof(hdr));
    if (ret != sizeof(hdr)) {
        error_report("failed to send a req, %s", strerror(errno));
        goto out;
    }

    if (wlen) {
        ret = qemu_co_sendv(s->fd, iov, niov, aio_req->iov_offset, wlen);
        if (ret != wlen) {
            error_report("failed to send a data, %s", strerror(errno));
        }
    }
out:
    socket_set_cork(s->fd, 0);
    aio_set_fd_handler(s->aio_context, s->fd, false,
                       co_read_response, NULL, NULL, s);
    s->co_send = NULL;
    qemu_co_mutex_unlock(&s->lock);
}

static int read_write_object(int fd, BlockDriverState *bs, char *buf,
                             uint64_t oid, uint8_t copies,
                             unsigned int datalen, uint64_t offset,
                             bool write, bool create, uint32_t cache_flags)
{
    SheepdogObjReq hdr;
    SheepdogObjRsp *rsp = (SheepdogObjRsp *)&hdr;
    unsigned int wlen, rlen;
    int ret;

    memset(&hdr, 0, sizeof(hdr));

    if (write) {
        wlen = datalen;
        rlen = 0;
        hdr.flags = SD_FLAG_CMD_WRITE;
        if (create) {
            hdr.opcode = SD_OP_CREATE_AND_WRITE_OBJ;
        } else {
            hdr.opcode = SD_OP_WRITE_OBJ;
        }
    } else {
        wlen = 0;
        rlen = datalen;
        hdr.opcode = SD_OP_READ_OBJ;
    }

    hdr.flags |= cache_flags;

    hdr.oid = oid;
    hdr.data_length = datalen;
    hdr.offset = offset;
    hdr.copies = copies;

    ret = do_req(fd, bs, (SheepdogReq *)&hdr, buf, &wlen, &rlen);
    if (ret) {
        error_report("failed to send a request to the sheep");
        return ret;
    }

    switch (rsp->result) {
    case SD_RES_SUCCESS:
        return 0;
    default:
        error_report("%s", sd_strerror(rsp->result));
        return -EIO;
    }
}

static int read_object(int fd, BlockDriverState *bs, char *buf,
                       uint64_t oid, uint8_t copies,
                       unsigned int datalen, uint64_t offset,
                       uint32_t cache_flags)
{
    return read_write_object(fd, bs, buf, oid, copies,
                             datalen, offset, false,
                             false, cache_flags);
}

static int write_object(int fd, BlockDriverState *bs, char *buf,
                        uint64_t oid, uint8_t copies,
                        unsigned int datalen, uint64_t offset, bool create,
                        uint32_t cache_flags)
{
    return read_write_object(fd, bs, buf, oid, copies,
                             datalen, offset, true,
                             create, cache_flags);
}

/* update inode with the latest state */
static int reload_inode(BDRVSheepdogState *s, uint32_t snapid, const char *tag)
{
    Error *local_err = NULL;
    SheepdogInode *inode;
    int ret = 0, fd;
    uint32_t vid = 0;

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        return -EIO;
    }

    inode = g_malloc(SD_INODE_HEADER_SIZE);

    ret = find_vdi_name(s, s->name, snapid, tag, &vid, false, &local_err);
    if (ret) {
        error_report_err(local_err);
        goto out;
    }

    ret = read_object(fd, s->bs, (char *)inode, vid_to_vdi_oid(vid),
                      s->inode.nr_copies, SD_INODE_HEADER_SIZE, 0,
                      s->cache_flags);
    if (ret < 0) {
        goto out;
    }

    if (inode->vdi_id != s->inode.vdi_id) {
        memcpy(&s->inode, inode, SD_INODE_HEADER_SIZE);
    }

out:
    g_free(inode);
    closesocket(fd);

    return ret;
}

static void coroutine_fn resend_aioreq(BDRVSheepdogState *s, AIOReq *aio_req)
{
    SheepdogAIOCB *acb = aio_req->aiocb;

    aio_req->create = false;

    /* check whether this request becomes a CoW one */
    if (acb->aiocb_type == AIOCB_WRITE_UDATA && is_data_obj(aio_req->oid)) {
        int idx = data_oid_to_idx(aio_req->oid);

        if (is_data_obj_writable(&s->inode, idx)) {
            goto out;
        }

        if (s->inode.data_vdi_id[idx]) {
            aio_req->base_oid = vid_to_data_oid(s->inode.data_vdi_id[idx], idx);
            aio_req->flags |= SD_FLAG_CMD_COW;
        }
        aio_req->create = true;
    }
out:
    if (is_data_obj(aio_req->oid)) {
        add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
                        acb->aiocb_type);
    } else {
        struct iovec iov;
        iov.iov_base = &s->inode;
        iov.iov_len = sizeof(s->inode);
        add_aio_request(s, aio_req, &iov, 1, AIOCB_WRITE_UDATA);
    }
}

static void sd_detach_aio_context(BlockDriverState *bs)
{
    BDRVSheepdogState *s = bs->opaque;

    aio_set_fd_handler(s->aio_context, s->fd, false, NULL,
                       NULL, NULL, NULL);
}

static void sd_attach_aio_context(BlockDriverState *bs,
                                  AioContext *new_context)
{
    BDRVSheepdogState *s = bs->opaque;

    s->aio_context = new_context;
    aio_set_fd_handler(new_context, s->fd, false,
                       co_read_response, NULL, NULL, s);
}

static QemuOptsList runtime_opts = {
    .name = "sheepdog",
    .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
    .desc = {
        {
            .name = "vdi",
            .type = QEMU_OPT_STRING,
        },
        {
            .name = "snap-id",
            .type = QEMU_OPT_NUMBER,
        },
        {
            .name = "tag",
            .type = QEMU_OPT_STRING,
        },
        { /* end of list */ }
    },
};

static int sd_open(BlockDriverState *bs, QDict *options, int flags,
                   Error **errp)
{
    int ret, fd;
    uint32_t vid = 0;
    BDRVSheepdogState *s = bs->opaque;
    const char *vdi, *snap_id_str, *tag;
    uint64_t snap_id;
    char *buf = NULL;
    QemuOpts *opts;
    Error *local_err = NULL;

    s->bs = bs;
    s->aio_context = bdrv_get_aio_context(bs);

    opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
    qemu_opts_absorb_qdict(opts, options, &local_err);
    if (local_err) {
        error_propagate(errp, local_err);
        ret = -EINVAL;
        goto err_no_fd;
    }

    s->addr = sd_server_config(options, errp);
    if (!s->addr) {
        ret = -EINVAL;
        goto err_no_fd;
    }

    vdi = qemu_opt_get(opts, "vdi");
    snap_id_str = qemu_opt_get(opts, "snap-id");
    snap_id = qemu_opt_get_number(opts, "snap-id", CURRENT_VDI_ID);
    tag = qemu_opt_get(opts, "tag");

    if (!vdi) {
        error_setg(errp, "parameter 'vdi' is missing");
        ret = -EINVAL;
        goto err_no_fd;
    }
    if (strlen(vdi) >= SD_MAX_VDI_LEN) {
        error_setg(errp, "value of parameter 'vdi' is too long");
        ret = -EINVAL;
        goto err_no_fd;
    }

    if (snap_id > UINT32_MAX) {
        snap_id = 0;
    }
    if (snap_id_str && !snap_id) {
        error_setg(errp, "'snap-id=%s' is not a valid snapshot ID",
                   snap_id_str);
        ret = -EINVAL;
        goto err_no_fd;
    }

    if (!tag) {
        tag = "";
    }
    if (tag && strlen(tag) >= SD_MAX_VDI_TAG_LEN) {
        error_setg(errp, "value of parameter 'tag' is too long");
        ret = -EINVAL;
        goto err_no_fd;
    }

    QLIST_INIT(&s->inflight_aio_head);
    QLIST_INIT(&s->failed_aio_head);
    QLIST_INIT(&s->inflight_aiocb_head);

    s->fd = get_sheep_fd(s, errp);
    if (s->fd < 0) {
        ret = s->fd;
        goto err_no_fd;
    }

    ret = find_vdi_name(s, vdi, (uint32_t)snap_id, tag, &vid, true, errp);
    if (ret) {
        goto err;
    }

    /*
     * QEMU block layer emulates writethrough cache as 'writeback + flush', so
     * we always set SD_FLAG_CMD_CACHE (writeback cache) as default.
     */
    s->cache_flags = SD_FLAG_CMD_CACHE;
    if (flags & BDRV_O_NOCACHE) {
        s->cache_flags = SD_FLAG_CMD_DIRECT;
    }
    s->discard_supported = true;

    if (snap_id || tag[0]) {
        DPRINTF("%" PRIx32 " snapshot inode was open.\n", vid);
        s->is_snapshot = true;
    }

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        ret = fd;
        goto err;
    }

    buf = g_malloc(SD_INODE_SIZE);
    ret = read_object(fd, s->bs, buf, vid_to_vdi_oid(vid),
                      0, SD_INODE_SIZE, 0, s->cache_flags);

    closesocket(fd);

    if (ret) {
        error_setg(errp, "Can't read snapshot inode");
        goto err;
    }

    memcpy(&s->inode, buf, sizeof(s->inode));

    bs->total_sectors = s->inode.vdi_size / BDRV_SECTOR_SIZE;
    pstrcpy(s->name, sizeof(s->name), vdi);
    qemu_co_mutex_init(&s->lock);
    qemu_co_queue_init(&s->overlapping_queue);
    qemu_opts_del(opts);
    g_free(buf);
    return 0;

err:
    aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd,
                       false, NULL, NULL, NULL, NULL);
    closesocket(s->fd);
err_no_fd:
    qemu_opts_del(opts);
    g_free(buf);
    return ret;
}

static int sd_reopen_prepare(BDRVReopenState *state, BlockReopenQueue *queue,
                             Error **errp)
{
    BDRVSheepdogState *s = state->bs->opaque;
    BDRVSheepdogReopenState *re_s;
    int ret = 0;

    re_s = state->opaque = g_new0(BDRVSheepdogReopenState, 1);

    re_s->cache_flags = SD_FLAG_CMD_CACHE;
    if (state->flags & BDRV_O_NOCACHE) {
        re_s->cache_flags = SD_FLAG_CMD_DIRECT;
    }

    re_s->fd = get_sheep_fd(s, errp);
    if (re_s->fd < 0) {
        ret = re_s->fd;
        return ret;
    }

    return ret;
}

static void sd_reopen_commit(BDRVReopenState *state)
{
    BDRVSheepdogReopenState *re_s = state->opaque;
    BDRVSheepdogState *s = state->bs->opaque;

    if (s->fd) {
        aio_set_fd_handler(s->aio_context, s->fd, false,
                           NULL, NULL, NULL, NULL);
        closesocket(s->fd);
    }

    s->fd = re_s->fd;
    s->cache_flags = re_s->cache_flags;

    g_free(state->opaque);
    state->opaque = NULL;

    return;
}

static void sd_reopen_abort(BDRVReopenState *state)
{
    BDRVSheepdogReopenState *re_s = state->opaque;
    BDRVSheepdogState *s = state->bs->opaque;

    if (re_s == NULL) {
        return;
    }

    if (re_s->fd) {
        aio_set_fd_handler(s->aio_context, re_s->fd, false,
                           NULL, NULL, NULL, NULL);
        closesocket(re_s->fd);
    }

    g_free(state->opaque);
    state->opaque = NULL;

    return;
}

static int do_sd_create(BDRVSheepdogState *s, uint32_t *vdi_id, int snapshot,
                        Error **errp)
{
    SheepdogVdiReq hdr;
    SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
    int fd, ret;
    unsigned int wlen, rlen = 0;
    char buf[SD_MAX_VDI_LEN];

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        return fd;
    }

    /* FIXME: would it be better to fail (e.g., return -EIO) when filename
     * does not fit in buf?  For now, just truncate and avoid buffer overrun.
     */
    memset(buf, 0, sizeof(buf));
    pstrcpy(buf, sizeof(buf), s->name);

    memset(&hdr, 0, sizeof(hdr));
    hdr.opcode = SD_OP_NEW_VDI;
    hdr.base_vdi_id = s->inode.vdi_id;

    wlen = SD_MAX_VDI_LEN;

    hdr.flags = SD_FLAG_CMD_WRITE;
    hdr.snapid = snapshot;

    hdr.data_length = wlen;
    hdr.vdi_size = s->inode.vdi_size;
    hdr.copy_policy = s->inode.copy_policy;
    hdr.copies = s->inode.nr_copies;
    hdr.block_size_shift = s->inode.block_size_shift;

    ret = do_req(fd, NULL, (SheepdogReq *)&hdr, buf, &wlen, &rlen);

    closesocket(fd);

    if (ret) {
        error_setg_errno(errp, -ret, "create failed");
        return ret;
    }

    if (rsp->result != SD_RES_SUCCESS) {
        error_setg(errp, "%s, %s", sd_strerror(rsp->result), s->inode.name);
        return -EIO;
    }

    if (vdi_id) {
        *vdi_id = rsp->vdi_id;
    }

    return 0;
}

static int sd_prealloc(const char *filename, Error **errp)
{
    BlockBackend *blk = NULL;
    BDRVSheepdogState *base = NULL;
    unsigned long buf_size;
    uint32_t idx, max_idx;
    uint32_t object_size;
    int64_t vdi_size;
    void *buf = NULL;
    int ret;

    blk = blk_new_open(filename, NULL, NULL,
                       BDRV_O_RDWR | BDRV_O_RESIZE | BDRV_O_PROTOCOL, errp);
    if (blk == NULL) {
        ret = -EIO;
        goto out_with_err_set;
    }

    blk_set_allow_write_beyond_eof(blk, true);

    vdi_size = blk_getlength(blk);
    if (vdi_size < 0) {
        ret = vdi_size;
        goto out;
    }

    base = blk_bs(blk)->opaque;
    object_size = (UINT32_C(1) << base->inode.block_size_shift);
    buf_size = MIN(object_size, SD_DATA_OBJ_SIZE);
    buf = g_malloc0(buf_size);

    max_idx = DIV_ROUND_UP(vdi_size, buf_size);

    for (idx = 0; idx < max_idx; idx++) {
        /*
         * The created image can be a cloned image, so we need to read
         * a data from the source image.
         */
        ret = blk_pread(blk, idx * buf_size, buf, buf_size);
        if (ret < 0) {
            goto out;
        }
        ret = blk_pwrite(blk, idx * buf_size, buf, buf_size, 0);
        if (ret < 0) {
            goto out;
        }
    }

    ret = 0;
out:
    if (ret < 0) {
        error_setg_errno(errp, -ret, "Can't pre-allocate");
    }
out_with_err_set:
    if (blk) {
        blk_unref(blk);
    }
    g_free(buf);

    return ret;
}

/*
 * Sheepdog support two kinds of redundancy, full replication and erasure
 * coding.
 *
 * # create a fully replicated vdi with x copies
 * -o redundancy=x (1 <= x <= SD_MAX_COPIES)
 *
 * # create a erasure coded vdi with x data strips and y parity strips
 * -o redundancy=x:y (x must be one of {2,4,8,16} and 1 <= y < SD_EC_MAX_STRIP)
 */
static int parse_redundancy(BDRVSheepdogState *s, const char *opt)
{
    struct SheepdogInode *inode = &s->inode;
    const char *n1, *n2;
    long copy, parity;
    char p[10];

    pstrcpy(p, sizeof(p), opt);
    n1 = strtok(p, ":");
    n2 = strtok(NULL, ":");

    if (!n1) {
        return -EINVAL;
    }

    copy = strtol(n1, NULL, 10);
    /* FIXME fix error checking by switching to qemu_strtol() */
    if (copy > SD_MAX_COPIES || copy < 1) {
        return -EINVAL;
    }
    if (!n2) {
        inode->copy_policy = 0;
        inode->nr_copies = copy;
        return 0;
    }

    if (copy != 2 && copy != 4 && copy != 8 && copy != 16) {
        return -EINVAL;
    }

    parity = strtol(n2, NULL, 10);
    /* FIXME fix error checking by switching to qemu_strtol() */
    if (parity >= SD_EC_MAX_STRIP || parity < 1) {
        return -EINVAL;
    }

    /*
     * 4 bits for parity and 4 bits for data.
     * We have to compress upper data bits because it can't represent 16
     */
    inode->copy_policy = ((copy / 2) << 4) + parity;
    inode->nr_copies = copy + parity;

    return 0;
}

static int parse_block_size_shift(BDRVSheepdogState *s, QemuOpts *opt)
{
    struct SheepdogInode *inode = &s->inode;
    uint64_t object_size;
    int obj_order;

    object_size = qemu_opt_get_size_del(opt, BLOCK_OPT_OBJECT_SIZE, 0);
    if (object_size) {
        if ((object_size - 1) & object_size) {    /* not a power of 2? */
            return -EINVAL;
        }
        obj_order = ctz32(object_size);
        if (obj_order < 20 || obj_order > 31) {
            return -EINVAL;
        }
        inode->block_size_shift = (uint8_t)obj_order;
    }

    return 0;
}

static int sd_create(const char *filename, QemuOpts *opts,
                     Error **errp)
{
    Error *err = NULL;
    int ret = 0;
    uint32_t vid = 0;
    char *backing_file = NULL;
    char *buf = NULL;
    BDRVSheepdogState *s;
    SheepdogConfig cfg;
    uint64_t max_vdi_size;
    bool prealloc = false;

    s = g_new0(BDRVSheepdogState, 1);

    if (strstr(filename, "://")) {
        sd_parse_uri(&cfg, filename, &err);
    } else {
        parse_vdiname(&cfg, filename, &err);
    }
    if (err) {
        error_propagate(errp, err);
        goto out;
    }

    buf = cfg.port ? g_strdup_printf("%d", cfg.port) : NULL;
    s->addr = sd_socket_address(cfg.path, cfg.host, buf);
    g_free(buf);
    strcpy(s->name, cfg.vdi);
    sd_config_done(&cfg);

    s->inode.vdi_size = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
                                 BDRV_SECTOR_SIZE);
    backing_file = qemu_opt_get_del(opts, BLOCK_OPT_BACKING_FILE);
    buf = qemu_opt_get_del(opts, BLOCK_OPT_PREALLOC);
    if (!buf || !strcmp(buf, "off")) {
        prealloc = false;
    } else if (!strcmp(buf, "full")) {
        prealloc = true;
    } else {
        error_setg(errp, "Invalid preallocation mode: '%s'", buf);
        ret = -EINVAL;
        goto out;
    }

    g_free(buf);
    buf = qemu_opt_get_del(opts, BLOCK_OPT_REDUNDANCY);
    if (buf) {
        ret = parse_redundancy(s, buf);
        if (ret < 0) {
            error_setg(errp, "Invalid redundancy mode: '%s'", buf);
            goto out;
        }
    }
    ret = parse_block_size_shift(s, opts);
    if (ret < 0) {
        error_setg(errp, "Invalid object_size."
                         " obect_size needs to be power of 2"
                         " and be limited from 2^20 to 2^31");
        goto out;
    }

    if (backing_file) {
        BlockBackend *blk;
        BDRVSheepdogState *base;
        BlockDriver *drv;

        /* Currently, only Sheepdog backing image is supported. */
        drv = bdrv_find_protocol(backing_file, true, NULL);
        if (!drv || strcmp(drv->protocol_name, "sheepdog") != 0) {
            error_setg(errp, "backing_file must be a sheepdog image");
            ret = -EINVAL;
            goto out;
        }

        blk = blk_new_open(backing_file, NULL, NULL,
                           BDRV_O_PROTOCOL, errp);
        if (blk == NULL) {
            ret = -EIO;
            goto out;
        }

        base = blk_bs(blk)->opaque;

        if (!is_snapshot(&base->inode)) {
            error_setg(errp, "cannot clone from a non snapshot vdi");
            blk_unref(blk);
            ret = -EINVAL;
            goto out;
        }
        s->inode.vdi_id = base->inode.vdi_id;
        blk_unref(blk);
    }

    s->aio_context = qemu_get_aio_context();

    /* if block_size_shift is not specified, get cluster default value */
    if (s->inode.block_size_shift == 0) {
        SheepdogVdiReq hdr;
        SheepdogClusterRsp *rsp = (SheepdogClusterRsp *)&hdr;
        int fd;
        unsigned int wlen = 0, rlen = 0;

        fd = connect_to_sdog(s, errp);
        if (fd < 0) {
            ret = fd;
            goto out;
        }

        memset(&hdr, 0, sizeof(hdr));
        hdr.opcode = SD_OP_GET_CLUSTER_DEFAULT;
        hdr.proto_ver = SD_PROTO_VER;

        ret = do_req(fd, NULL, (SheepdogReq *)&hdr,
                     NULL, &wlen, &rlen);
        closesocket(fd);
        if (ret) {
            error_setg_errno(errp, -ret, "failed to get cluster default");
            goto out;
        }
        if (rsp->result == SD_RES_SUCCESS) {
            s->inode.block_size_shift = rsp->block_size_shift;
        } else {
            s->inode.block_size_shift = SD_DEFAULT_BLOCK_SIZE_SHIFT;
        }
    }

    max_vdi_size = (UINT64_C(1) << s->inode.block_size_shift) * MAX_DATA_OBJS;

    if (s->inode.vdi_size > max_vdi_size) {
        error_setg(errp, "An image is too large."
                         " The maximum image size is %"PRIu64 "GB",
                         max_vdi_size / 1024 / 1024 / 1024);
        ret = -EINVAL;
        goto out;
    }

    ret = do_sd_create(s, &vid, 0, errp);
    if (ret) {
        goto out;
    }

    if (prealloc) {
        ret = sd_prealloc(filename, errp);
    }
out:
    g_free(backing_file);
    g_free(buf);
    g_free(s);
    return ret;
}

static void sd_close(BlockDriverState *bs)
{
    Error *local_err = NULL;
    BDRVSheepdogState *s = bs->opaque;
    SheepdogVdiReq hdr;
    SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
    unsigned int wlen, rlen = 0;
    int fd, ret;

    DPRINTF("%s\n", s->name);

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        return;
    }

    memset(&hdr, 0, sizeof(hdr));

    hdr.opcode = SD_OP_RELEASE_VDI;
    hdr.type = LOCK_TYPE_NORMAL;
    hdr.base_vdi_id = s->inode.vdi_id;
    wlen = strlen(s->name) + 1;
    hdr.data_length = wlen;
    hdr.flags = SD_FLAG_CMD_WRITE;

    ret = do_req(fd, s->bs, (SheepdogReq *)&hdr,
                 s->name, &wlen, &rlen);

    closesocket(fd);

    if (!ret && rsp->result != SD_RES_SUCCESS &&
        rsp->result != SD_RES_VDI_NOT_LOCKED) {
        error_report("%s, %s", sd_strerror(rsp->result), s->name);
    }

    aio_set_fd_handler(bdrv_get_aio_context(bs), s->fd,
                       false, NULL, NULL, NULL, NULL);
    closesocket(s->fd);
    qapi_free_SocketAddress(s->addr);
}

static int64_t sd_getlength(BlockDriverState *bs)
{
    BDRVSheepdogState *s = bs->opaque;

    return s->inode.vdi_size;
}

static int sd_truncate(BlockDriverState *bs, int64_t offset, Error **errp)
{
    BDRVSheepdogState *s = bs->opaque;
    int ret, fd;
    unsigned int datalen;
    uint64_t max_vdi_size;

    max_vdi_size = (UINT64_C(1) << s->inode.block_size_shift) * MAX_DATA_OBJS;
    if (offset < s->inode.vdi_size) {
        error_setg(errp, "shrinking is not supported");
        return -EINVAL;
    } else if (offset > max_vdi_size) {
        error_setg(errp, "too big image size");
        return -EINVAL;
    }

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        return fd;
    }

    /* we don't need to update entire object */
    datalen = SD_INODE_SIZE - sizeof(s->inode.data_vdi_id);
    s->inode.vdi_size = offset;
    ret = write_object(fd, s->bs, (char *)&s->inode,
                       vid_to_vdi_oid(s->inode.vdi_id), s->inode.nr_copies,
                       datalen, 0, false, s->cache_flags);
    close(fd);

    if (ret < 0) {
        error_setg_errno(errp, -ret, "failed to update an inode");
    }

    return ret;
}

/*
 * This function is called after writing data objects.  If we need to
 * update metadata, this sends a write request to the vdi object.
 */
static void coroutine_fn sd_write_done(SheepdogAIOCB *acb)
{
    BDRVSheepdogState *s = acb->s;
    struct iovec iov;
    AIOReq *aio_req;
    uint32_t offset, data_len, mn, mx;

    mn = acb->min_dirty_data_idx;
    mx = acb->max_dirty_data_idx;
    if (mn <= mx) {
        /* we need to update the vdi object. */
        ++acb->nr_pending;
        offset = sizeof(s->inode) - sizeof(s->inode.data_vdi_id) +
            mn * sizeof(s->inode.data_vdi_id[0]);
        data_len = (mx - mn + 1) * sizeof(s->inode.data_vdi_id[0]);

        acb->min_dirty_data_idx = UINT32_MAX;
        acb->max_dirty_data_idx = 0;

        iov.iov_base = &s->inode;
        iov.iov_len = sizeof(s->inode);
        aio_req = alloc_aio_req(s, acb, vid_to_vdi_oid(s->inode.vdi_id),
                                data_len, offset, 0, false, 0, offset);
        add_aio_request(s, aio_req, &iov, 1, AIOCB_WRITE_UDATA);
        if (--acb->nr_pending) {
            qemu_coroutine_yield();
        }
    }
}

/* Delete current working VDI on the snapshot chain */
static bool sd_delete(BDRVSheepdogState *s)
{
    Error *local_err = NULL;
    unsigned int wlen = SD_MAX_VDI_LEN, rlen = 0;
    SheepdogVdiReq hdr = {
        .opcode = SD_OP_DEL_VDI,
        .base_vdi_id = s->inode.vdi_id,
        .data_length = wlen,
        .flags = SD_FLAG_CMD_WRITE,
    };
    SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;
    int fd, ret;

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        return false;
    }

    ret = do_req(fd, s->bs, (SheepdogReq *)&hdr,
                 s->name, &wlen, &rlen);
    closesocket(fd);
    if (ret) {
        return false;
    }
    switch (rsp->result) {
    case SD_RES_NO_VDI:
        error_report("%s was already deleted", s->name);
        /* fall through */
    case SD_RES_SUCCESS:
        break;
    default:
        error_report("%s, %s", sd_strerror(rsp->result), s->name);
        return false;
    }

    return true;
}

/*
 * Create a writable VDI from a snapshot
 */
static int sd_create_branch(BDRVSheepdogState *s)
{
    Error *local_err = NULL;
    int ret, fd;
    uint32_t vid;
    char *buf;
    bool deleted;

    DPRINTF("%" PRIx32 " is snapshot.\n", s->inode.vdi_id);

    buf = g_malloc(SD_INODE_SIZE);

    /*
     * Even If deletion fails, we will just create extra snapshot based on
     * the working VDI which was supposed to be deleted. So no need to
     * false bail out.
     */
    deleted = sd_delete(s);
    ret = do_sd_create(s, &vid, !deleted, &local_err);
    if (ret) {
        error_report_err(local_err);
        goto out;
    }

    DPRINTF("%" PRIx32 " is created.\n", vid);

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        ret = fd;
        goto out;
    }

    ret = read_object(fd, s->bs, buf, vid_to_vdi_oid(vid),
                      s->inode.nr_copies, SD_INODE_SIZE, 0, s->cache_flags);

    closesocket(fd);

    if (ret < 0) {
        goto out;
    }

    memcpy(&s->inode, buf, sizeof(s->inode));

    s->is_snapshot = false;
    ret = 0;
    DPRINTF("%" PRIx32 " was newly created.\n", s->inode.vdi_id);

out:
    g_free(buf);

    return ret;
}

/*
 * Send I/O requests to the server.
 *
 * This function sends requests to the server, links the requests to
 * the inflight_list in BDRVSheepdogState, and exits without
 * waiting the response.  The responses are received in the
 * `aio_read_response' function which is called from the main loop as
 * a fd handler.
 *
 * Returns 1 when we need to wait a response, 0 when there is no sent
 * request and -errno in error cases.
 */
static void coroutine_fn sd_co_rw_vector(SheepdogAIOCB *acb)
{
    int ret = 0;
    unsigned long len, done = 0, total = acb->nb_sectors * BDRV_SECTOR_SIZE;
    unsigned long idx;
    uint32_t object_size;
    uint64_t oid;
    uint64_t offset;
    BDRVSheepdogState *s = acb->s;
    SheepdogInode *inode = &s->inode;
    AIOReq *aio_req;

    if (acb->aiocb_type == AIOCB_WRITE_UDATA && s->is_snapshot) {
        /*
         * In the case we open the snapshot VDI, Sheepdog creates the
         * writable VDI when we do a write operation first.
         */
        ret = sd_create_branch(s);
        if (ret) {
            acb->ret = -EIO;
            return;
        }
    }

    object_size = (UINT32_C(1) << inode->block_size_shift);
    idx = acb->sector_num * BDRV_SECTOR_SIZE / object_size;
    offset = (acb->sector_num * BDRV_SECTOR_SIZE) % object_size;

    /*
     * Make sure we don't free the aiocb before we are done with all requests.
     * This additional reference is dropped at the end of this function.
     */
    acb->nr_pending++;

    while (done != total) {
        uint8_t flags = 0;
        uint64_t old_oid = 0;
        bool create = false;

        oid = vid_to_data_oid(inode->data_vdi_id[idx], idx);

        len = MIN(total - done, object_size - offset);

        switch (acb->aiocb_type) {
        case AIOCB_READ_UDATA:
            if (!inode->data_vdi_id[idx]) {
                qemu_iovec_memset(acb->qiov, done, 0, len);
                goto done;
            }
            break;
        case AIOCB_WRITE_UDATA:
            if (!inode->data_vdi_id[idx]) {
                create = true;
            } else if (!is_data_obj_writable(inode, idx)) {
                /* Copy-On-Write */
                create = true;
                old_oid = oid;
                flags = SD_FLAG_CMD_COW;
            }
            break;
        case AIOCB_DISCARD_OBJ:
            /*
             * We discard the object only when the whole object is
             * 1) allocated 2) trimmed. Otherwise, simply skip it.
             */
            if (len != object_size || inode->data_vdi_id[idx] == 0) {
                goto done;
            }
            break;
        default:
            break;
        }

        if (create) {
            DPRINTF("update ino (%" PRIu32 ") %" PRIu64 " %" PRIu64 " %ld\n",
                    inode->vdi_id, oid,
                    vid_to_data_oid(inode->data_vdi_id[idx], idx), idx);
            oid = vid_to_data_oid(inode->vdi_id, idx);
            DPRINTF("new oid %" PRIx64 "\n", oid);
        }

        aio_req = alloc_aio_req(s, acb, oid, len, offset, flags, create,
                                old_oid,
                                acb->aiocb_type == AIOCB_DISCARD_OBJ ?
                                0 : done);
        add_aio_request(s, aio_req, acb->qiov->iov, acb->qiov->niov,
                        acb->aiocb_type);
    done:
        offset = 0;
        idx++;
        done += len;
    }
    if (--acb->nr_pending) {
        qemu_coroutine_yield();
    }
}

static void sd_aio_complete(SheepdogAIOCB *acb)
{
    if (acb->aiocb_type == AIOCB_FLUSH_CACHE) {
        return;
    }

    QLIST_REMOVE(acb, aiocb_siblings);
    qemu_co_queue_restart_all(&acb->s->overlapping_queue);
}

static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
                        int nb_sectors, QEMUIOVector *qiov)
{
    SheepdogAIOCB acb;
    int ret;
    int64_t offset = (sector_num + nb_sectors) * BDRV_SECTOR_SIZE;
    BDRVSheepdogState *s = bs->opaque;

    if (offset > s->inode.vdi_size) {
        ret = sd_truncate(bs, offset, NULL);
        if (ret < 0) {
            return ret;
        }
    }

    sd_aio_setup(&acb, s, qiov, sector_num, nb_sectors, AIOCB_WRITE_UDATA);
    sd_co_rw_vector(&acb);
    sd_write_done(&acb);
    sd_aio_complete(&acb);

    return acb.ret;
}

static coroutine_fn int sd_co_readv(BlockDriverState *bs, int64_t sector_num,
                       int nb_sectors, QEMUIOVector *qiov)
{
    SheepdogAIOCB acb;
    BDRVSheepdogState *s = bs->opaque;

    sd_aio_setup(&acb, s, qiov, sector_num, nb_sectors, AIOCB_READ_UDATA);
    sd_co_rw_vector(&acb);
    sd_aio_complete(&acb);

    return acb.ret;
}

static int coroutine_fn sd_co_flush_to_disk(BlockDriverState *bs)
{
    BDRVSheepdogState *s = bs->opaque;
    SheepdogAIOCB acb;
    AIOReq *aio_req;

    if (s->cache_flags != SD_FLAG_CMD_CACHE) {
        return 0;
    }

    sd_aio_setup(&acb, s, NULL, 0, 0, AIOCB_FLUSH_CACHE);

    acb.nr_pending++;
    aio_req = alloc_aio_req(s, &acb, vid_to_vdi_oid(s->inode.vdi_id),
                            0, 0, 0, false, 0, 0);
    add_aio_request(s, aio_req, NULL, 0, acb.aiocb_type);

    if (--acb.nr_pending) {
        qemu_coroutine_yield();
    }

    sd_aio_complete(&acb);
    return acb.ret;
}

static int sd_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
{
    Error *local_err = NULL;
    BDRVSheepdogState *s = bs->opaque;
    int ret, fd;
    uint32_t new_vid;
    SheepdogInode *inode;
    unsigned int datalen;

    DPRINTF("sn_info: name %s id_str %s s: name %s vm_state_size %" PRId64 " "
            "is_snapshot %d\n", sn_info->name, sn_info->id_str,
            s->name, sn_info->vm_state_size, s->is_snapshot);

    if (s->is_snapshot) {
        error_report("You can't create a snapshot of a snapshot VDI, "
                     "%s (%" PRIu32 ").", s->name, s->inode.vdi_id);

        return -EINVAL;
    }

    DPRINTF("%s %s\n", sn_info->name, sn_info->id_str);

    s->inode.vm_state_size = sn_info->vm_state_size;
    s->inode.vm_clock_nsec = sn_info->vm_clock_nsec;
    /* It appears that inode.tag does not require a NUL terminator,
     * which means this use of strncpy is ok.
     */
    strncpy(s->inode.tag, sn_info->name, sizeof(s->inode.tag));
    /* we don't need to update entire object */
    datalen = SD_INODE_SIZE - sizeof(s->inode.data_vdi_id);
    inode = g_malloc(datalen);

    /* refresh inode. */
    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        ret = fd;
        goto cleanup;
    }

    ret = write_object(fd, s->bs, (char *)&s->inode,
                       vid_to_vdi_oid(s->inode.vdi_id), s->inode.nr_copies,
                       datalen, 0, false, s->cache_flags);
    if (ret < 0) {
        error_report("failed to write snapshot's inode.");
        goto cleanup;
    }

    ret = do_sd_create(s, &new_vid, 1, &local_err);
    if (ret < 0) {
        error_reportf_err(local_err,
                          "failed to create inode for snapshot: ");
        goto cleanup;
    }

    ret = read_object(fd, s->bs, (char *)inode,
                      vid_to_vdi_oid(new_vid), s->inode.nr_copies, datalen, 0,
                      s->cache_flags);

    if (ret < 0) {
        error_report("failed to read new inode info. %s", strerror(errno));
        goto cleanup;
    }

    memcpy(&s->inode, inode, datalen);
    DPRINTF("s->inode: name %s snap_id %x oid %x\n",
            s->inode.name, s->inode.snap_id, s->inode.vdi_id);

cleanup:
    g_free(inode);
    closesocket(fd);
    return ret;
}

/*
 * We implement rollback(loadvm) operation to the specified snapshot by
 * 1) switch to the snapshot
 * 2) rely on sd_create_branch to delete working VDI and
 * 3) create a new working VDI based on the specified snapshot
 */
static int sd_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
{
    BDRVSheepdogState *s = bs->opaque;
    BDRVSheepdogState *old_s;
    char tag[SD_MAX_VDI_TAG_LEN];
    uint32_t snapid = 0;
    int ret;

    if (!sd_parse_snapid_or_tag(snapshot_id, &snapid, tag)) {
        return -EINVAL;
    }

    old_s = g_new(BDRVSheepdogState, 1);

    memcpy(old_s, s, sizeof(BDRVSheepdogState));

    ret = reload_inode(s, snapid, tag);
    if (ret) {
        goto out;
    }

    ret = sd_create_branch(s);
    if (ret) {
        goto out;
    }

    g_free(old_s);

    return 0;
out:
    /* recover bdrv_sd_state */
    memcpy(s, old_s, sizeof(BDRVSheepdogState));
    g_free(old_s);

    error_report("failed to open. recover old bdrv_sd_state.");

    return ret;
}

#define NR_BATCHED_DISCARD 128

static int remove_objects(BDRVSheepdogState *s, Error **errp)
{
    int fd, i = 0, nr_objs = 0;
    int ret;
    SheepdogInode *inode = &s->inode;

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        return fd;
    }

    nr_objs = count_data_objs(inode);
    while (i < nr_objs) {
        int start_idx, nr_filled_idx;

        while (i < nr_objs && !inode->data_vdi_id[i]) {
            i++;
        }
        start_idx = i;

        nr_filled_idx = 0;
        while (i < nr_objs && nr_filled_idx < NR_BATCHED_DISCARD) {
            if (inode->data_vdi_id[i]) {
                inode->data_vdi_id[i] = 0;
                nr_filled_idx++;
            }

            i++;
        }

        ret = write_object(fd, s->bs,
                           (char *)&inode->data_vdi_id[start_idx],
                           vid_to_vdi_oid(s->inode.vdi_id), inode->nr_copies,
                           (i - start_idx) * sizeof(uint32_t),
                           offsetof(struct SheepdogInode,
                                    data_vdi_id[start_idx]),
                           false, s->cache_flags);
        if (ret < 0) {
            error_setg(errp, "Failed to discard snapshot inode");
            goto out;
        }
    }

    ret = 0;
out:
    closesocket(fd);
    return ret;
}

static int sd_snapshot_delete(BlockDriverState *bs,
                              const char *snapshot_id,
                              const char *name,
                              Error **errp)
{
    /*
     * FIXME should delete the snapshot matching both @snapshot_id and
     * @name, but @name not used here
     */
    unsigned long snap_id = 0;
    char snap_tag[SD_MAX_VDI_TAG_LEN];
    int fd, ret;
    char buf[SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN];
    BDRVSheepdogState *s = bs->opaque;
    unsigned int wlen = SD_MAX_VDI_LEN + SD_MAX_VDI_TAG_LEN, rlen = 0;
    uint32_t vid;
    SheepdogVdiReq hdr = {
        .opcode = SD_OP_DEL_VDI,
        .data_length = wlen,
        .flags = SD_FLAG_CMD_WRITE,
    };
    SheepdogVdiRsp *rsp = (SheepdogVdiRsp *)&hdr;

    ret = remove_objects(s, errp);
    if (ret) {
        return ret;
    }

    memset(buf, 0, sizeof(buf));
    memset(snap_tag, 0, sizeof(snap_tag));
    pstrcpy(buf, SD_MAX_VDI_LEN, s->name);
    /* TODO Use sd_parse_snapid() once this mess is cleaned up */
    ret = qemu_strtoul(snapshot_id, NULL, 10, &snap_id);
    if (ret || snap_id > UINT32_MAX) {
        /*
         * FIXME Since qemu_strtoul() returns -EINVAL when
         * @snapshot_id is null, @snapshot_id is mandatory.  Correct
         * would be to require at least one of @snapshot_id and @name.
         */
        error_setg(errp, "Invalid snapshot ID: %s",
                         snapshot_id ? snapshot_id : "<null>");
        return -EINVAL;
    }

    if (snap_id) {
        hdr.snapid = (uint32_t) snap_id;
    } else {
        /* FIXME I suspect we should use @name here */
        /* FIXME don't truncate silently */
        pstrcpy(snap_tag, sizeof(snap_tag), snapshot_id);
        pstrcpy(buf + SD_MAX_VDI_LEN, SD_MAX_VDI_TAG_LEN, snap_tag);
    }

    ret = find_vdi_name(s, s->name, snap_id, snap_tag, &vid, true, errp);
    if (ret) {
        return ret;
    }

    fd = connect_to_sdog(s, errp);
    if (fd < 0) {
        return fd;
    }

    ret = do_req(fd, s->bs, (SheepdogReq *)&hdr,
                 buf, &wlen, &rlen);
    closesocket(fd);
    if (ret) {
        error_setg_errno(errp, -ret, "Couldn't send request to server");
        return ret;
    }

    switch (rsp->result) {
    case SD_RES_NO_VDI:
        error_setg(errp, "Can't find the snapshot");
        return -ENOENT;
    case SD_RES_SUCCESS:
        break;
    default:
        error_setg(errp, "%s", sd_strerror(rsp->result));
        return -EIO;
    }

    return 0;
}

static int sd_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
{
    Error *local_err = NULL;
    BDRVSheepdogState *s = bs->opaque;
    SheepdogReq req;
    int fd, nr = 1024, ret, max = BITS_TO_LONGS(SD_NR_VDIS) * sizeof(long);
    QEMUSnapshotInfo *sn_tab = NULL;
    unsigned wlen, rlen;
    int found = 0;
    static SheepdogInode inode;
    unsigned long *vdi_inuse;
    unsigned int start_nr;
    uint64_t hval;
    uint32_t vid;

    vdi_inuse = g_malloc(max);

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        ret = fd;
        goto out;
    }

    rlen = max;
    wlen = 0;

    memset(&req, 0, sizeof(req));

    req.opcode = SD_OP_READ_VDIS;
    req.data_length = max;

    ret = do_req(fd, s->bs, &req, vdi_inuse, &wlen, &rlen);

    closesocket(fd);
    if (ret) {
        goto out;
    }

    sn_tab = g_new0(QEMUSnapshotInfo, nr);

    /* calculate a vdi id with hash function */
    hval = fnv_64a_buf(s->name, strlen(s->name), FNV1A_64_INIT);
    start_nr = hval & (SD_NR_VDIS - 1);

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        ret = fd;
        goto out;
    }

    for (vid = start_nr; found < nr; vid = (vid + 1) % SD_NR_VDIS) {
        if (!test_bit(vid, vdi_inuse)) {
            break;
        }

        /* we don't need to read entire object */
        ret = read_object(fd, s->bs, (char *)&inode,
                          vid_to_vdi_oid(vid),
                          0, SD_INODE_SIZE - sizeof(inode.data_vdi_id), 0,
                          s->cache_flags);

        if (ret) {
            continue;
        }

        if (!strcmp(inode.name, s->name) && is_snapshot(&inode)) {
            sn_tab[found].date_sec = inode.snap_ctime >> 32;
            sn_tab[found].date_nsec = inode.snap_ctime & 0xffffffff;
            sn_tab[found].vm_state_size = inode.vm_state_size;
            sn_tab[found].vm_clock_nsec = inode.vm_clock_nsec;

            snprintf(sn_tab[found].id_str, sizeof(sn_tab[found].id_str),
                     "%" PRIu32, inode.snap_id);
            pstrcpy(sn_tab[found].name,
                    MIN(sizeof(sn_tab[found].name), sizeof(inode.tag)),
                    inode.tag);
            found++;
        }
    }

    closesocket(fd);
out:
    *psn_tab = sn_tab;

    g_free(vdi_inuse);

    if (ret < 0) {
        return ret;
    }

    return found;
}

static int do_load_save_vmstate(BDRVSheepdogState *s, uint8_t *data,
                                int64_t pos, int size, int load)
{
    Error *local_err = NULL;
    bool create;
    int fd, ret = 0, remaining = size;
    unsigned int data_len;
    uint64_t vmstate_oid;
    uint64_t offset;
    uint32_t vdi_index;
    uint32_t vdi_id = load ? s->inode.parent_vdi_id : s->inode.vdi_id;
    uint32_t object_size = (UINT32_C(1) << s->inode.block_size_shift);

    fd = connect_to_sdog(s, &local_err);
    if (fd < 0) {
        error_report_err(local_err);
        return fd;
    }

    while (remaining) {
        vdi_index = pos / object_size;
        offset = pos % object_size;

        data_len = MIN(remaining, object_size - offset);

        vmstate_oid = vid_to_vmstate_oid(vdi_id, vdi_index);

        create = (offset == 0);
        if (load) {
            ret = read_object(fd, s->bs, (char *)data, vmstate_oid,
                              s->inode.nr_copies, data_len, offset,
                              s->cache_flags);
        } else {
            ret = write_object(fd, s->bs, (char *)data, vmstate_oid,
                               s->inode.nr_copies, data_len, offset, create,
                               s->cache_flags);
        }

        if (ret < 0) {
            error_report("failed to save vmstate %s", strerror(errno));
            goto cleanup;
        }

        pos += data_len;
        data += data_len;
        remaining -= data_len;
    }
    ret = size;
cleanup:
    closesocket(fd);
    return ret;
}

static int sd_save_vmstate(BlockDriverState *bs, QEMUIOVector *qiov,
                           int64_t pos)
{
    BDRVSheepdogState *s = bs->opaque;
    void *buf;
    int ret;

    buf = qemu_blockalign(bs, qiov->size);
    qemu_iovec_to_buf(qiov, 0, buf, qiov->size);
    ret = do_load_save_vmstate(s, (uint8_t *) buf, pos, qiov->size, 0);
    qemu_vfree(buf);

    return ret;
}

static int sd_load_vmstate(BlockDriverState *bs, QEMUIOVector *qiov,
                           int64_t pos)
{
    BDRVSheepdogState *s = bs->opaque;
    void *buf;
    int ret;

    buf = qemu_blockalign(bs, qiov->size);
    ret = do_load_save_vmstate(s, buf, pos, qiov->size, 1);
    qemu_iovec_from_buf(qiov, 0, buf, qiov->size);
    qemu_vfree(buf);

    return ret;
}


static coroutine_fn int sd_co_pdiscard(BlockDriverState *bs, int64_t offset,
                                      int count)
{
    SheepdogAIOCB acb;
    BDRVSheepdogState *s = bs->opaque;
    QEMUIOVector discard_iov;
    struct iovec iov;
    uint32_t zero = 0;

    if (!s->discard_supported) {
        return 0;
    }

    memset(&discard_iov, 0, sizeof(discard_iov));
    memset(&iov, 0, sizeof(iov));
    iov.iov_base = &zero;
    iov.iov_len = sizeof(zero);
    discard_iov.iov = &iov;
    discard_iov.niov = 1;
    if (!QEMU_IS_ALIGNED(offset | count, BDRV_SECTOR_SIZE)) {
        return -ENOTSUP;
    }
    sd_aio_setup(&acb, s, &discard_iov, offset >> BDRV_SECTOR_BITS,
                 count >> BDRV_SECTOR_BITS, AIOCB_DISCARD_OBJ);
    sd_co_rw_vector(&acb);
    sd_aio_complete(&acb);

    return acb.ret;
}

static coroutine_fn int64_t
sd_co_get_block_status(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
                       int *pnum, BlockDriverState **file)
{
    BDRVSheepdogState *s = bs->opaque;
    SheepdogInode *inode = &s->inode;
    uint32_t object_size = (UINT32_C(1) << inode->block_size_shift);
    uint64_t offset = sector_num * BDRV_SECTOR_SIZE;
    unsigned long start = offset / object_size,
                  end = DIV_ROUND_UP((sector_num + nb_sectors) *
                                     BDRV_SECTOR_SIZE, object_size);
    unsigned long idx;
    int64_t ret = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID | offset;

    for (idx = start; idx < end; idx++) {
        if (inode->data_vdi_id[idx] == 0) {
            break;
        }
    }
    if (idx == start) {
        /* Get the longest length of unallocated sectors */
        ret = 0;
        for (idx = start + 1; idx < end; idx++) {
            if (inode->data_vdi_id[idx] != 0) {
                break;
            }
        }
    }

    *pnum = (idx - start) * object_size / BDRV_SECTOR_SIZE;
    if (*pnum > nb_sectors) {
        *pnum = nb_sectors;
    }
    if (ret > 0 && ret & BDRV_BLOCK_OFFSET_VALID) {
        *file = bs;
    }
    return ret;
}

static int64_t sd_get_allocated_file_size(BlockDriverState *bs)
{
    BDRVSheepdogState *s = bs->opaque;
    SheepdogInode *inode = &s->inode;
    uint32_t object_size = (UINT32_C(1) << inode->block_size_shift);
    unsigned long i, last = DIV_ROUND_UP(inode->vdi_size, object_size);
    uint64_t size = 0;

    for (i = 0; i < last; i++) {
        if (inode->data_vdi_id[i] == 0) {
            continue;
        }
        size += object_size;
    }
    return size;
}

static QemuOptsList sd_create_opts = {
    .name = "sheepdog-create-opts",
    .head = QTAILQ_HEAD_INITIALIZER(sd_create_opts.head),
    .desc = {
        {
            .name = BLOCK_OPT_SIZE,
            .type = QEMU_OPT_SIZE,
            .help = "Virtual disk size"
        },
        {
            .name = BLOCK_OPT_BACKING_FILE,
            .type = QEMU_OPT_STRING,
            .help = "File name of a base image"
        },
        {
            .name = BLOCK_OPT_PREALLOC,
            .type = QEMU_OPT_STRING,
            .help = "Preallocation mode (allowed values: off, full)"
        },
        {
            .name = BLOCK_OPT_REDUNDANCY,
            .type = QEMU_OPT_STRING,
            .help = "Redundancy of the image"
        },
        {
            .name = BLOCK_OPT_OBJECT_SIZE,
            .type = QEMU_OPT_SIZE,
            .help = "Object size of the image"
        },
        { /* end of list */ }
    }
};

static BlockDriver bdrv_sheepdog = {
    .format_name    = "sheepdog",
    .protocol_name  = "sheepdog",
    .instance_size  = sizeof(BDRVSheepdogState),
    .bdrv_parse_filename    = sd_parse_filename,
    .bdrv_file_open = sd_open,
    .bdrv_reopen_prepare    = sd_reopen_prepare,
    .bdrv_reopen_commit     = sd_reopen_commit,
    .bdrv_reopen_abort      = sd_reopen_abort,
    .bdrv_close     = sd_close,
    .bdrv_create    = sd_create,
    .bdrv_has_zero_init = bdrv_has_zero_init_1,
    .bdrv_getlength = sd_getlength,
    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
    .bdrv_truncate  = sd_truncate,

    .bdrv_co_readv  = sd_co_readv,
    .bdrv_co_writev = sd_co_writev,
    .bdrv_co_flush_to_disk  = sd_co_flush_to_disk,
    .bdrv_co_pdiscard = sd_co_pdiscard,
    .bdrv_co_get_block_status = sd_co_get_block_status,

    .bdrv_snapshot_create   = sd_snapshot_create,
    .bdrv_snapshot_goto     = sd_snapshot_goto,
    .bdrv_snapshot_delete   = sd_snapshot_delete,
    .bdrv_snapshot_list     = sd_snapshot_list,

    .bdrv_save_vmstate  = sd_save_vmstate,
    .bdrv_load_vmstate  = sd_load_vmstate,

    .bdrv_detach_aio_context = sd_detach_aio_context,
    .bdrv_attach_aio_context = sd_attach_aio_context,

    .create_opts    = &sd_create_opts,
};

static BlockDriver bdrv_sheepdog_tcp = {
    .format_name    = "sheepdog",
    .protocol_name  = "sheepdog+tcp",
    .instance_size  = sizeof(BDRVSheepdogState),
    .bdrv_parse_filename    = sd_parse_filename,
    .bdrv_file_open = sd_open,
    .bdrv_reopen_prepare    = sd_reopen_prepare,
    .bdrv_reopen_commit     = sd_reopen_commit,
    .bdrv_reopen_abort      = sd_reopen_abort,
    .bdrv_close     = sd_close,
    .bdrv_create    = sd_create,
    .bdrv_has_zero_init = bdrv_has_zero_init_1,
    .bdrv_getlength = sd_getlength,
    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
    .bdrv_truncate  = sd_truncate,

    .bdrv_co_readv  = sd_co_readv,
    .bdrv_co_writev = sd_co_writev,
    .bdrv_co_flush_to_disk  = sd_co_flush_to_disk,
    .bdrv_co_pdiscard = sd_co_pdiscard,
    .bdrv_co_get_block_status = sd_co_get_block_status,

    .bdrv_snapshot_create   = sd_snapshot_create,
    .bdrv_snapshot_goto     = sd_snapshot_goto,
    .bdrv_snapshot_delete   = sd_snapshot_delete,
    .bdrv_snapshot_list     = sd_snapshot_list,

    .bdrv_save_vmstate  = sd_save_vmstate,
    .bdrv_load_vmstate  = sd_load_vmstate,

    .bdrv_detach_aio_context = sd_detach_aio_context,
    .bdrv_attach_aio_context = sd_attach_aio_context,

    .create_opts    = &sd_create_opts,
};

static BlockDriver bdrv_sheepdog_unix = {
    .format_name    = "sheepdog",
    .protocol_name  = "sheepdog+unix",
    .instance_size  = sizeof(BDRVSheepdogState),
    .bdrv_parse_filename    = sd_parse_filename,
    .bdrv_file_open = sd_open,
    .bdrv_reopen_prepare    = sd_reopen_prepare,
    .bdrv_reopen_commit     = sd_reopen_commit,
    .bdrv_reopen_abort      = sd_reopen_abort,
    .bdrv_close     = sd_close,
    .bdrv_create    = sd_create,
    .bdrv_has_zero_init = bdrv_has_zero_init_1,
    .bdrv_getlength = sd_getlength,
    .bdrv_get_allocated_file_size = sd_get_allocated_file_size,
    .bdrv_truncate  = sd_truncate,

    .bdrv_co_readv  = sd_co_readv,
    .bdrv_co_writev = sd_co_writev,
    .bdrv_co_flush_to_disk  = sd_co_flush_to_disk,
    .bdrv_co_pdiscard = sd_co_pdiscard,
    .bdrv_co_get_block_status = sd_co_get_block_status,

    .bdrv_snapshot_create   = sd_snapshot_create,
    .bdrv_snapshot_goto     = sd_snapshot_goto,
    .bdrv_snapshot_delete   = sd_snapshot_delete,
    .bdrv_snapshot_list     = sd_snapshot_list,

    .bdrv_save_vmstate  = sd_save_vmstate,
    .bdrv_load_vmstate  = sd_load_vmstate,

    .bdrv_detach_aio_context = sd_detach_aio_context,
    .bdrv_attach_aio_context = sd_attach_aio_context,

    .create_opts    = &sd_create_opts,
};

static void bdrv_sheepdog_init(void)
{
    bdrv_register(&bdrv_sheepdog);
    bdrv_register(&bdrv_sheepdog_tcp);
    bdrv_register(&bdrv_sheepdog_unix);
}
block_init(bdrv_sheepdog_init);