summaryrefslogtreecommitdiff
path: root/storage/innobase/buf/buf0buf.cc
blob: 802e67de1b0f0d1681d63d86b4de796bae5488f3 (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
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
6133
6134
6135
6136
6137
6138
6139
6140
6141
6142
6143
6144
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
6164
6165
6166
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
6210
6211
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
6245
6246
6247
6248
6249
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
6261
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
6344
6345
6346
6347
6348
6349
6350
6351
6352
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
6629
6630
6631
6632
6633
6634
6635
6636
6637
6638
6639
6640
6641
6642
6643
6644
6645
6646
6647
6648
6649
6650
6651
6652
6653
6654
6655
6656
6657
6658
6659
6660
6661
6662
6663
6664
6665
6666
6667
6668
6669
6670
6671
6672
6673
6674
6675
6676
6677
6678
6679
6680
6681
6682
6683
6684
6685
6686
6687
6688
6689
6690
6691
6692
6693
6694
6695
6696
6697
6698
6699
6700
6701
6702
6703
6704
6705
6706
6707
6708
6709
6710
6711
6712
6713
6714
6715
6716
6717
6718
6719
6720
6721
6722
6723
6724
6725
6726
6727
6728
6729
6730
6731
6732
6733
6734
6735
6736
6737
6738
6739
6740
6741
6742
6743
6744
6745
6746
6747
6748
6749
6750
6751
6752
6753
6754
6755
6756
6757
6758
6759
6760
6761
6762
6763
6764
6765
6766
6767
6768
6769
6770
6771
6772
6773
6774
6775
6776
6777
6778
6779
6780
6781
6782
6783
6784
6785
6786
6787
6788
6789
6790
6791
6792
6793
6794
6795
6796
6797
6798
6799
6800
6801
6802
6803
6804
6805
6806
6807
6808
6809
6810
6811
6812
6813
6814
6815
6816
6817
6818
6819
6820
6821
6822
6823
6824
6825
6826
6827
6828
6829
6830
6831
6832
6833
6834
6835
6836
6837
6838
6839
6840
6841
6842
6843
6844
6845
6846
6847
6848
6849
6850
6851
6852
6853
6854
6855
6856
6857
6858
6859
6860
6861
6862
6863
6864
6865
6866
6867
6868
6869
6870
6871
6872
6873
6874
6875
6876
6877
6878
6879
6880
6881
6882
6883
6884
6885
6886
6887
6888
6889
6890
6891
6892
6893
6894
6895
6896
6897
6898
6899
6900
6901
6902
6903
6904
6905
6906
6907
6908
6909
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
6931
6932
6933
6934
6935
6936
6937
6938
6939
6940
6941
6942
6943
6944
6945
6946
6947
6948
6949
6950
6951
6952
6953
6954
6955
6956
6957
6958
6959
6960
6961
6962
6963
6964
6965
6966
6967
6968
6969
6970
6971
6972
6973
6974
6975
6976
6977
6978
6979
6980
6981
6982
6983
6984
6985
6986
6987
6988
6989
6990
6991
6992
6993
6994
6995
6996
6997
6998
6999
7000
7001
7002
7003
7004
7005
7006
7007
7008
7009
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
7026
7027
7028
7029
7030
7031
7032
7033
7034
7035
7036
7037
7038
7039
7040
7041
7042
7043
7044
7045
7046
7047
7048
7049
7050
7051
7052
7053
7054
7055
7056
7057
7058
7059
7060
7061
7062
7063
7064
7065
7066
7067
7068
7069
7070
7071
7072
7073
7074
7075
7076
7077
7078
7079
7080
7081
7082
7083
7084
7085
7086
7087
7088
7089
7090
7091
7092
7093
7094
7095
7096
7097
7098
7099
7100
7101
7102
7103
7104
7105
7106
7107
7108
7109
7110
7111
7112
7113
7114
7115
7116
7117
7118
7119
7120
7121
7122
7123
7124
7125
7126
7127
7128
7129
7130
7131
7132
7133
7134
7135
7136
7137
7138
7139
7140
7141
7142
7143
7144
7145
7146
7147
7148
7149
7150
7151
7152
7153
7154
7155
7156
7157
7158
7159
7160
7161
7162
7163
7164
7165
7166
7167
7168
7169
7170
7171
7172
7173
7174
7175
7176
7177
7178
7179
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
7224
7225
7226
7227
7228
7229
7230
7231
7232
7233
7234
7235
7236
7237
7238
7239
7240
7241
7242
7243
7244
7245
7246
7247
7248
7249
7250
7251
7252
7253
7254
7255
7256
7257
7258
7259
7260
7261
7262
7263
7264
7265
7266
7267
7268
7269
7270
7271
7272
7273
7274
7275
7276
7277
7278
7279
7280
7281
7282
7283
7284
7285
7286
7287
7288
7289
7290
7291
7292
7293
7294
7295
7296
7297
7298
7299
7300
7301
7302
7303
7304
7305
7306
7307
7308
7309
7310
7311
7312
7313
7314
7315
7316
7317
7318
7319
7320
7321
7322
7323
7324
7325
7326
7327
7328
7329
7330
7331
7332
7333
7334
7335
7336
7337
7338
7339
7340
7341
7342
7343
7344
7345
7346
7347
7348
7349
7350
7351
7352
7353
7354
7355
7356
7357
7358
7359
7360
7361
7362
7363
7364
7365
7366
7367
7368
7369
7370
7371
7372
7373
7374
7375
7376
7377
7378
7379
7380
7381
7382
7383
7384
7385
7386
7387
7388
7389
7390
7391
7392
7393
7394
7395
7396
7397
7398
7399
7400
7401
7402
7403
7404
7405
7406
7407
7408
7409
7410
7411
7412
7413
7414
7415
7416
7417
7418
7419
7420
7421
7422
7423
7424
7425
7426
7427
7428
7429
7430
7431
7432
7433
7434
7435
7436
7437
7438
7439
7440
7441
7442
7443
7444
7445
7446
7447
7448
7449
7450
7451
7452
7453
7454
7455
7456
7457
7458
7459
7460
7461
7462
7463
7464
7465
7466
7467
7468
7469
7470
7471
7472
7473
7474
7475
7476
7477
7478
7479
7480
7481
7482
7483
7484
7485
7486
7487
7488
7489
7490
7491
7492
7493
7494
7495
7496
7497
7498
7499
7500
7501
/*****************************************************************************

Copyright (c) 1995, 2018, Oracle and/or its affiliates. All Rights Reserved.
Copyright (c) 2008, Google Inc.
Copyright (c) 2013, 2020, MariaDB Corporation.

Portions of this file contain modifications contributed and copyrighted by
Google, Inc. Those modifications are gratefully acknowledged and are described
briefly in the InnoDB documentation. The contributions by Google are
incorporated with their permission, and subject to the conditions contained in
the file COPYING.Google.

This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; version 2 of the License.

This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1335 USA

*****************************************************************************/

/**************************************************//**
@file buf/buf0buf.cc
The database buffer buf_pool

Created 11/5/1995 Heikki Tuuri
*******************************************************/

#include "mtr0types.h"
#include "mach0data.h"
#include "page0size.h"
#include "buf0buf.h"
#include <string.h>

#ifdef UNIV_NONINL
#include "buf0buf.ic"
#endif

#ifndef UNIV_INNOCHECKSUM
#include "mem0mem.h"
#include "btr0btr.h"
#include "fil0fil.h"
#include "fil0crypt.h"
#include "buf0buddy.h"
#include "lock0lock.h"
#include "sync0rw.h"
#include "btr0sea.h"
#include "ibuf0ibuf.h"
#include "trx0undo.h"
#include "trx0purge.h"
#include "log0log.h"
#include "dict0stats_bg.h"
#include "srv0srv.h"
#include "srv0start.h"
#include "dict0dict.h"
#include "log0recv.h"
#include "srv0mon.h"
#include "log0crypt.h"
#endif /* !UNIV_INNOCHECKSUM */
#include "page0zip.h"
#include "sync0sync.h"
#include "buf0dump.h"
#include <new>
#include <map>
#include <sstream>
#ifndef UNIV_INNOCHECKSUM
#include "fil0pagecompress.h"
#include "fsp0pagecompress.h"
#endif
#include "ut0byte.h"
#include <new>

#ifdef UNIV_LINUX
#include <stdlib.h>
#endif

#ifdef HAVE_LZO
#include "lzo/lzo1x.h"
#endif

using st_::span;

#ifdef HAVE_LIBNUMA
#include <numa.h>
#include <numaif.h>
struct set_numa_interleave_t
{
	set_numa_interleave_t()
	{
		if (srv_numa_interleave) {

			struct bitmask *numa_mems_allowed = numa_get_mems_allowed();
			ib::info() << "Setting NUMA memory policy to"
				" MPOL_INTERLEAVE";
			if (set_mempolicy(MPOL_INTERLEAVE,
					  numa_mems_allowed->maskp,
					  numa_mems_allowed->size) != 0) {

				ib::warn() << "Failed to set NUMA memory"
					" policy to MPOL_INTERLEAVE: "
					<< strerror(errno);
			}
			numa_bitmask_free(numa_mems_allowed);
		}
	}

	~set_numa_interleave_t()
	{
		if (srv_numa_interleave) {

			ib::info() << "Setting NUMA memory policy to"
				" MPOL_DEFAULT";
			if (set_mempolicy(MPOL_DEFAULT, NULL, 0) != 0) {
				ib::warn() << "Failed to set NUMA memory"
					" policy to MPOL_DEFAULT: "
					<< strerror(errno);
			}
		}
	}
};

#define NUMA_MEMPOLICY_INTERLEAVE_IN_SCOPE set_numa_interleave_t scoped_numa
#else
#define NUMA_MEMPOLICY_INTERLEAVE_IN_SCOPE
#endif /* HAVE_LIBNUMA */

#ifdef HAVE_SNAPPY
#include "snappy-c.h"
#endif

/*
		IMPLEMENTATION OF THE BUFFER POOL
		=================================

Performance improvement:
------------------------
Thread scheduling in NT may be so slow that the OS wait mechanism should
not be used even in waiting for disk reads to complete.
Rather, we should put waiting query threads to the queue of
waiting jobs, and let the OS thread do something useful while the i/o
is processed. In this way we could remove most OS thread switches in
an i/o-intensive benchmark like TPC-C.

A possibility is to put a user space thread library between the database
and NT. User space thread libraries might be very fast.

SQL Server 7.0 can be configured to use 'fibers' which are lightweight
threads in NT. These should be studied.

		Buffer frames and blocks
		------------------------
Following the terminology of Gray and Reuter, we call the memory
blocks where file pages are loaded buffer frames. For each buffer
frame there is a control block, or shortly, a block, in the buffer
control array. The control info which does not need to be stored
in the file along with the file page, resides in the control block.

		Buffer pool struct
		------------------
The buffer buf_pool contains a single mutex which protects all the
control data structures of the buf_pool. The content of a buffer frame is
protected by a separate read-write lock in its control block, though.
These locks can be locked and unlocked without owning the buf_pool->mutex.
The OS events in the buf_pool struct can be waited for without owning the
buf_pool->mutex.

The buf_pool->mutex is a hot-spot in main memory, causing a lot of
memory bus traffic on multiprocessor systems when processors
alternately access the mutex. On our Pentium, the mutex is accessed
maybe every 10 microseconds. We gave up the solution to have mutexes
for each control block, for instance, because it seemed to be
complicated.

A solution to reduce mutex contention of the buf_pool->mutex is to
create a separate mutex for the page hash table. On Pentium,
accessing the hash table takes 2 microseconds, about half
of the total buf_pool->mutex hold time.

		Control blocks
		--------------

The control block contains, for instance, the bufferfix count
which is incremented when a thread wants a file page to be fixed
in a buffer frame. The bufferfix operation does not lock the
contents of the frame, however. For this purpose, the control
block contains a read-write lock.

The buffer frames have to be aligned so that the start memory
address of a frame is divisible by the universal page size, which
is a power of two.

We intend to make the buffer buf_pool size on-line reconfigurable,
that is, the buf_pool size can be changed without closing the database.
Then the database administarator may adjust it to be bigger
at night, for example. The control block array must
contain enough control blocks for the maximum buffer buf_pool size
which is used in the particular database.
If the buf_pool size is cut, we exploit the virtual memory mechanism of
the OS, and just refrain from using frames at high addresses. Then the OS
can swap them to disk.

The control blocks containing file pages are put to a hash table
according to the file address of the page.
We could speed up the access to an individual page by using
"pointer swizzling": we could replace the page references on
non-leaf index pages by direct pointers to the page, if it exists
in the buf_pool. We could make a separate hash table where we could
chain all the page references in non-leaf pages residing in the buf_pool,
using the page reference as the hash key,
and at the time of reading of a page update the pointers accordingly.
Drawbacks of this solution are added complexity and,
possibly, extra space required on non-leaf pages for memory pointers.
A simpler solution is just to speed up the hash table mechanism
in the database, using tables whose size is a power of 2.

		Lists of blocks
		---------------

There are several lists of control blocks.

The free list (buf_pool->free) contains blocks which are currently not
used.

The common LRU list contains all the blocks holding a file page
except those for which the bufferfix count is non-zero.
The pages are in the LRU list roughly in the order of the last
access to the page, so that the oldest pages are at the end of the
list. We also keep a pointer to near the end of the LRU list,
which we can use when we want to artificially age a page in the
buf_pool. This is used if we know that some page is not needed
again for some time: we insert the block right after the pointer,
causing it to be replaced sooner than would normally be the case.
Currently this aging mechanism is used for read-ahead mechanism
of pages, and it can also be used when there is a scan of a full
table which cannot fit in the memory. Putting the pages near the
end of the LRU list, we make sure that most of the buf_pool stays
in the main memory, undisturbed.

The unzip_LRU list contains a subset of the common LRU list.  The
blocks on the unzip_LRU list hold a compressed file page and the
corresponding uncompressed page frame.  A block is in unzip_LRU if and
only if the predicate buf_page_belongs_to_unzip_LRU(&block->page)
holds.  The blocks in unzip_LRU will be in same order as they are in
the common LRU list.  That is, each manipulation of the common LRU
list will result in the same manipulation of the unzip_LRU list.

The chain of modified blocks (buf_pool->flush_list) contains the blocks
holding file pages that have been modified in the memory
but not written to disk yet. The block with the oldest modification
which has not yet been written to disk is at the end of the chain.
The access to this list is protected by buf_pool->flush_list_mutex.

The chain of unmodified compressed blocks (buf_pool->zip_clean)
contains the control blocks (buf_page_t) of those compressed pages
that are not in buf_pool->flush_list and for which no uncompressed
page has been allocated in the buffer pool.  The control blocks for
uncompressed pages are accessible via buf_block_t objects that are
reachable via buf_pool->chunks[].

The chains of free memory blocks (buf_pool->zip_free[]) are used by
the buddy allocator (buf0buddy.cc) to keep track of currently unused
memory blocks of size sizeof(buf_page_t)..UNIV_PAGE_SIZE / 2.  These
blocks are inside the UNIV_PAGE_SIZE-sized memory blocks of type
BUF_BLOCK_MEMORY that the buddy allocator requests from the buffer
pool.  The buddy allocator is solely used for allocating control
blocks for compressed pages (buf_page_t) and compressed page frames.

		Loading a file page
		-------------------

First, a victim block for replacement has to be found in the
buf_pool. It is taken from the free list or searched for from the
end of the LRU-list. An exclusive lock is reserved for the frame,
the io_fix field is set in the block fixing the block in buf_pool,
and the io-operation for loading the page is queued. The io-handler thread
releases the X-lock on the frame and resets the io_fix field
when the io operation completes.

A thread may request the above operation using the function
buf_page_get(). It may then continue to request a lock on the frame.
The lock is granted when the io-handler releases the x-lock.

		Read-ahead
		----------

The read-ahead mechanism is intended to be intelligent and
isolated from the semantically higher levels of the database
index management. From the higher level we only need the
information if a file page has a natural successor or
predecessor page. On the leaf level of a B-tree index,
these are the next and previous pages in the natural
order of the pages.

Let us first explain the read-ahead mechanism when the leafs
of a B-tree are scanned in an ascending or descending order.
When a read page is the first time referenced in the buf_pool,
the buffer manager checks if it is at the border of a so-called
linear read-ahead area. The tablespace is divided into these
areas of size 64 blocks, for example. So if the page is at the
border of such an area, the read-ahead mechanism checks if
all the other blocks in the area have been accessed in an
ascending or descending order. If this is the case, the system
looks at the natural successor or predecessor of the page,
checks if that is at the border of another area, and in this case
issues read-requests for all the pages in that area. Maybe
we could relax the condition that all the pages in the area
have to be accessed: if data is deleted from a table, there may
appear holes of unused pages in the area.

A different read-ahead mechanism is used when there appears
to be a random access pattern to a file.
If a new page is referenced in the buf_pool, and several pages
of its random access area (for instance, 32 consecutive pages
in a tablespace) have recently been referenced, we may predict
that the whole area may be needed in the near future, and issue
the read requests for the whole area.
*/

#ifndef UNIV_INNOCHECKSUM
/** Value in microseconds */
static const int WAIT_FOR_READ	= 100;
static const int WAIT_FOR_WRITE = 100;
/** Number of attempts made to read in a page in the buffer pool */
static const ulint	BUF_PAGE_READ_MAX_RETRIES = 100;
/** Number of pages to read ahead */
static const ulint	BUF_READ_AHEAD_PAGES = 64;
/** The maximum portion of the buffer pool that can be used for the
read-ahead buffer.  (Divide buf_pool size by this amount) */
static const ulint	BUF_READ_AHEAD_PORTION = 32;

/** The buffer pools of the database */
buf_pool_t*	buf_pool_ptr;

/** true when resizing buffer pool is in the critical path. */
volatile bool	buf_pool_resizing;

/** true when withdrawing buffer pool pages might cause page relocation */
volatile bool	buf_pool_withdrawing;

/** the clock is incremented every time a pointer to a page may become obsolete;
if the withdrwa clock has not changed, the pointer is still valid in buffer
pool. if changed, the pointer might not be in buffer pool any more. */
volatile ulint	buf_withdraw_clock;

/** Map of buffer pool chunks by its first frame address
This is newly made by initialization of buffer pool and buf_resize_thread.
Currently, no need mutex protection for update. */
typedef std::map<
	const byte*,
	buf_chunk_t*,
	std::less<const byte*>,
	ut_allocator<std::pair<const byte* const, buf_chunk_t*> > >
	buf_pool_chunk_map_t;

static buf_pool_chunk_map_t*			buf_chunk_map_reg;

/** Chunk map to be used to lookup.
The map pointed by this should not be updated */
static buf_pool_chunk_map_t*	buf_chunk_map_ref = NULL;

#ifdef UNIV_DEBUG
/** Disable resizing buffer pool to make assertion code not expensive. */
my_bool			buf_disable_resize_buffer_pool_debug = TRUE;
#endif /* UNIV_DEBUG */

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
/** This is used to insert validation operations in execution
in the debug version */
static ulint	buf_dbg_counter	= 0;
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

#if defined UNIV_PFS_MUTEX || defined UNIV_PFS_RWLOCK
# ifndef PFS_SKIP_BUFFER_MUTEX_RWLOCK

/* Buffer block mutexes and rwlocks can be registered
in one group rather than individually. If PFS_GROUP_BUFFER_SYNC
is defined, register buffer block mutex and rwlock
in one group after their initialization. */
#  define PFS_GROUP_BUFFER_SYNC

/* This define caps the number of mutexes/rwlocks can
be registered with performance schema. Developers can
modify this define if necessary. Please note, this would
be effective only if PFS_GROUP_BUFFER_SYNC is defined. */
#  define PFS_MAX_BUFFER_MUTEX_LOCK_REGISTER	ULINT_MAX

# endif /* !PFS_SKIP_BUFFER_MUTEX_RWLOCK */
#endif /* UNIV_PFS_MUTEX || UNIV_PFS_RWLOCK */

/** Macro to determine whether the read of write counter is used depending
on the io_type */
#define MONITOR_RW_COUNTER(io_type, counter)		\
	((io_type == BUF_IO_READ)			\
	 ? (counter##_READ)				\
	 : (counter##_WRITTEN))


/** Reserve a buffer slot for encryption, decryption or page compression.
@param[in,out]	buf_pool	buffer pool
@return reserved buffer slot */
static buf_tmp_buffer_t* buf_pool_reserve_tmp_slot(buf_pool_t* buf_pool)
{
	for (ulint i = 0; i < buf_pool->tmp_arr->n_slots; i++) {
		buf_tmp_buffer_t* slot = &buf_pool->tmp_arr->slots[i];
		if (slot->acquire()) {
			return slot;
		}
	}

	/* We assume that free slot is found */
	ut_error;
	return NULL;
}

/** Reserve a buffer for encryption, decryption or decompression.
@param[in,out]	slot	reserved slot */
static void buf_tmp_reserve_crypt_buf(buf_tmp_buffer_t* slot)
{
	if (!slot->crypt_buf) {
		slot->crypt_buf = static_cast<byte*>(
			aligned_malloc(srv_page_size, srv_page_size));
	}
}

/** Reserve a buffer for compression.
@param[in,out]	slot	reserved slot */
static void buf_tmp_reserve_compression_buf(buf_tmp_buffer_t* slot)
{
	if (!slot->comp_buf) {
		/* Both snappy and lzo compression methods require that
		output buffer used for compression is bigger than input
		buffer. Increase the allocated buffer size accordingly. */
		ulint size = srv_page_size;
#ifdef HAVE_LZO
		size += LZO1X_1_15_MEM_COMPRESS;
#elif defined HAVE_SNAPPY
		size = snappy_max_compressed_length(size);
#endif
		slot->comp_buf = static_cast<byte*>(
			aligned_malloc(size, srv_page_size));
	}
}

/** Registers a chunk to buf_pool_chunk_map
@param[in]	chunk	chunk of buffers */
static
void
buf_pool_register_chunk(
	buf_chunk_t*	chunk)
{
	buf_chunk_map_reg->insert(buf_pool_chunk_map_t::value_type(
		chunk->blocks->frame, chunk));
}

/** Decrypt a page for temporary tablespace.
@param[in,out]	tmp_frame	Temporary buffer
@param[in]	src_frame	Page to decrypt
@return true if temporary tablespace decrypted, false if not */
static bool buf_tmp_page_decrypt(byte* tmp_frame, byte* src_frame)
{
	if (buf_is_zeroes(span<const byte>(src_frame, srv_page_size))) {
		return true;
	}

	/* read space & lsn */
	uint header_len = FIL_PAGE_DATA;

	/* Copy FIL page header, it is not encrypted */
	memcpy(tmp_frame, src_frame, header_len);

	/* Calculate the offset where decryption starts */
	const byte* src = src_frame + header_len;
	byte* dst = tmp_frame + header_len;
	uint srclen = uint(srv_page_size)
		- header_len - FIL_PAGE_DATA_END;
	ulint offset = mach_read_from_4(src_frame + FIL_PAGE_OFFSET);

	if (!log_tmp_block_decrypt(src, srclen, dst,
				   (offset * srv_page_size))) {
		return false;
	}

	memcpy(tmp_frame + srv_page_size - FIL_PAGE_DATA_END,
	       src_frame + srv_page_size - FIL_PAGE_DATA_END,
	       FIL_PAGE_DATA_END);

	memcpy(src_frame, tmp_frame, srv_page_size);
	srv_stats.pages_decrypted.inc();
	srv_stats.n_temp_blocks_decrypted.inc();

	return true; /* page was decrypted */
}

/** Decrypt a page.
@param[in,out]	bpage	Page control block
@param[in,out]	space	tablespace
@return whether the operation was successful */
static bool buf_page_decrypt_after_read(buf_page_t* bpage, fil_space_t* space)
{
	ut_ad(space->n_pending_ios > 0);
	ut_ad(space->id == bpage->id.space());

	byte* dst_frame = bpage->zip.data ? bpage->zip.data :
		((buf_block_t*) bpage)->frame;
	bool page_compressed = fil_page_is_compressed(dst_frame);
	buf_pool_t* buf_pool = buf_pool_from_bpage(bpage);

	if (bpage->id.page_no() == 0) {
		/* File header pages are not encrypted/compressed */
		return (true);
	}

	if (space->purpose == FIL_TYPE_TEMPORARY
	    && innodb_encrypt_temporary_tables) {
		buf_tmp_buffer_t* slot = buf_pool_reserve_tmp_slot(buf_pool);
		buf_tmp_reserve_crypt_buf(slot);

		if (!buf_tmp_page_decrypt(slot->crypt_buf, dst_frame)) {
			slot->release();
			ib::error() << "Encrypted page " << bpage->id
				    << " in file " << space->chain.start->name;
			return false;
		}

		slot->release();
		return true;
	}

	/* Page is encrypted if encryption information is found from
	tablespace and page contains used key_version. This is true
	also for pages first compressed and then encrypted. */

	buf_tmp_buffer_t* slot;

	if (page_compressed) {
		/* the page we read is unencrypted */
		/* Find free slot from temporary memory array */
decompress:
		slot = buf_pool_reserve_tmp_slot(buf_pool);
		/* For decompression, use crypt_buf. */
		buf_tmp_reserve_crypt_buf(slot);
decompress_with_slot:
		ut_d(fil_page_type_validate(dst_frame));

		ulint write_size = fil_page_decompress(slot->crypt_buf,
						       dst_frame);
		slot->release();

		ut_ad(!write_size || fil_page_type_validate(dst_frame));
		ut_ad(space->n_pending_ios > 0);
		return write_size != 0;
	}

	if (space->crypt_data
	    && mach_read_from_4(FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION
			       + dst_frame)) {
		/* Verify encryption checksum before we even try to
		decrypt. */
		if (!fil_space_verify_crypt_checksum(dst_frame, bpage->size)) {
decrypt_failed:
			ib::error() << "Encrypted page " << bpage->id
				    << " in file " << space->chain.start->name
				    << " looks corrupted; key_version="
				    << mach_read_from_4(
					    FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION
					    + dst_frame);
			return false;
		}

		/* Find free slot from temporary memory array */
		slot = buf_pool_reserve_tmp_slot(buf_pool);
		buf_tmp_reserve_crypt_buf(slot);

		ut_d(fil_page_type_validate(dst_frame));

		/* decrypt using crypt_buf to dst_frame */
		if (!fil_space_decrypt(space, slot->crypt_buf, dst_frame)) {
			slot->release();
			goto decrypt_failed;
		}

		ut_d(fil_page_type_validate(dst_frame));

		if (fil_page_is_compressed_encrypted(dst_frame)) {
			goto decompress_with_slot;
		}

		slot->release();
	} else if (fil_page_is_compressed_encrypted(dst_frame)) {
		goto decompress;
	}

	ut_ad(space->n_pending_ios > 0);
	return true;
}

/* prototypes for new functions added to ha_innodb.cc */
trx_t* innobase_get_trx();

/********************************************************************//**
Gets the smallest oldest_modification lsn for any page in the pool. Returns
zero if all modified pages have been flushed to disk.
@return oldest modification in pool, zero if none */
lsn_t
buf_pool_get_oldest_modification(void)
/*==================================*/
{
	lsn_t		lsn = 0;
	lsn_t		oldest_lsn = 0;

	/* When we traverse all the flush lists we don't want another
	thread to add a dirty page to any flush list. */
	log_flush_order_mutex_enter();

	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		buf_flush_list_mutex_enter(buf_pool);

		buf_page_t*	bpage;

		/* We don't let log-checkpoint halt because pages from system
		temporary are not yet flushed to the disk. Anyway, object
		residing in system temporary doesn't generate REDO logging. */
		for (bpage = UT_LIST_GET_LAST(buf_pool->flush_list);
		     bpage != NULL
			&& fsp_is_system_temporary(bpage->id.space());
		     bpage = UT_LIST_GET_PREV(list, bpage)) {
			/* Do nothing. */
		}

		if (bpage != NULL) {
			ut_ad(bpage->in_flush_list);
			lsn = bpage->oldest_modification;
		}

		buf_flush_list_mutex_exit(buf_pool);

		if (!oldest_lsn || oldest_lsn > lsn) {
			oldest_lsn = lsn;
		}
	}

	log_flush_order_mutex_exit();

	/* The returned answer may be out of date: the flush_list can
	change after the mutex has been released. */

	return(oldest_lsn);
}

/********************************************************************//**
Get total buffer pool statistics. */
void
buf_get_total_list_len(
/*===================*/
	ulint*		LRU_len,	/*!< out: length of all LRU lists */
	ulint*		free_len,	/*!< out: length of all free lists */
	ulint*		flush_list_len)	/*!< out: length of all flush lists */
{
	ulint		i;

	*LRU_len = 0;
	*free_len = 0;
	*flush_list_len = 0;

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		*LRU_len += UT_LIST_GET_LEN(buf_pool->LRU);
		*free_len += UT_LIST_GET_LEN(buf_pool->free);
		*flush_list_len += UT_LIST_GET_LEN(buf_pool->flush_list);
	}
}

/********************************************************************//**
Get total list size in bytes from all buffer pools. */
void
buf_get_total_list_size_in_bytes(
/*=============================*/
	buf_pools_list_size_t*	buf_pools_list_size)	/*!< out: list sizes
							in all buffer pools */
{
	ut_ad(buf_pools_list_size);
	memset(buf_pools_list_size, 0, sizeof(*buf_pools_list_size));

	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);
		/* We don't need mutex protection since this is
		for statistics purpose */
		buf_pools_list_size->LRU_bytes += buf_pool->stat.LRU_bytes;
		buf_pools_list_size->unzip_LRU_bytes +=
			UT_LIST_GET_LEN(buf_pool->unzip_LRU) * UNIV_PAGE_SIZE;
		buf_pools_list_size->flush_list_bytes +=
			buf_pool->stat.flush_list_bytes;
	}
}

/********************************************************************//**
Get total buffer pool statistics. */
void
buf_get_total_stat(
/*===============*/
	buf_pool_stat_t*	tot_stat)	/*!< out: buffer pool stats */
{
	ulint			i;

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

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_stat_t*buf_stat;
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		buf_stat = &buf_pool->stat;
		tot_stat->n_page_gets += buf_stat->n_page_gets;
		tot_stat->n_pages_read += buf_stat->n_pages_read;
		tot_stat->n_pages_written += buf_stat->n_pages_written;
		tot_stat->n_pages_created += buf_stat->n_pages_created;
		tot_stat->n_ra_pages_read_rnd += buf_stat->n_ra_pages_read_rnd;
		tot_stat->n_ra_pages_read += buf_stat->n_ra_pages_read;
		tot_stat->n_ra_pages_evicted += buf_stat->n_ra_pages_evicted;
		tot_stat->n_pages_made_young += buf_stat->n_pages_made_young;

		tot_stat->n_pages_not_made_young +=
			buf_stat->n_pages_not_made_young;
	}
}

/********************************************************************//**
Allocates a buffer block.
@return own: the allocated block, in state BUF_BLOCK_MEMORY */
buf_block_t*
buf_block_alloc(
/*============*/
	buf_pool_t*	buf_pool)	/*!< in/out: buffer pool instance,
					or NULL for round-robin selection
					of the buffer pool */
{
	buf_block_t*	block;
	ulint		index;
	static ulint	buf_pool_index;

	if (buf_pool == NULL) {
		/* We are allocating memory from any buffer pool, ensure
		we spread the grace on all buffer pool instances. */
		index = buf_pool_index++ % srv_buf_pool_instances;
		buf_pool = buf_pool_from_array(index);
	}

	block = buf_LRU_get_free_block(buf_pool);

	buf_block_set_state(block, BUF_BLOCK_MEMORY);

	return(block);
}
#endif /* !UNIV_INNOCHECKSUM */

/** Checks if the page is in crc32 checksum format.
@param[in]	read_buf		database page
@param[in]	checksum_field1		new checksum field
@param[in]	checksum_field2		old checksum field
@return true if the page is in crc32 checksum format. */
bool
buf_page_is_checksum_valid_crc32(
	const byte*			read_buf,
	ulint				checksum_field1,
	ulint				checksum_field2)
{
	const uint32_t	crc32 = buf_calc_page_crc32(read_buf);

#ifdef UNIV_INNOCHECKSUM
	if (log_file
	    && srv_checksum_algorithm == SRV_CHECKSUM_ALGORITHM_STRICT_CRC32) {
		fprintf(log_file, "page::%llu;"
			" crc32 calculated = %u;"
			" recorded checksum field1 = " ULINTPF " recorded"
			" checksum field2 =" ULINTPF "\n", cur_page_num,
			crc32, checksum_field1, checksum_field2);
	}
#endif /* UNIV_INNOCHECKSUM */

	if (checksum_field1 != checksum_field2) {
		return false;
	}

	return checksum_field1 == crc32
#ifdef INNODB_BUG_ENDIAN_CRC32
		|| checksum_field1 == buf_calc_page_crc32(read_buf, true)
#endif
		;
}

/** Checks if the page is in innodb checksum format.
@param[in]	read_buf	database page
@param[in]	checksum_field1	new checksum field
@param[in]	checksum_field2	old checksum field
@return true if the page is in innodb checksum format. */
bool
buf_page_is_checksum_valid_innodb(
	const byte*			read_buf,
	ulint				checksum_field1,
	ulint				checksum_field2)
{
	/* There are 2 valid formulas for
	checksum_field2 (old checksum field) which algo=innodb could have
	written to the page:

	1. Very old versions of InnoDB only stored 8 byte lsn to the
	start and the end of the page.

	2. Newer InnoDB versions store the old formula checksum
	(buf_calc_page_old_checksum()). */

	ulint	old_checksum = buf_calc_page_old_checksum(read_buf);
	ulint	new_checksum = buf_calc_page_new_checksum(read_buf);

#ifdef UNIV_INNOCHECKSUM
	if (log_file
	    && srv_checksum_algorithm == SRV_CHECKSUM_ALGORITHM_INNODB) {
		fprintf(log_file, "page::%llu;"
			" old style: calculated ="
			" " ULINTPF "; recorded = " ULINTPF "\n",
			cur_page_num, old_checksum,
			checksum_field2);
		fprintf(log_file, "page::%llu;"
			" new style: calculated ="
			" " ULINTPF "; crc32 = %u; recorded = " ULINTPF "\n",
			cur_page_num, new_checksum,
			buf_calc_page_crc32(read_buf), checksum_field1);
	}

	if (log_file
	    && srv_checksum_algorithm == SRV_CHECKSUM_ALGORITHM_STRICT_INNODB) {
		fprintf(log_file, "page::%llu;"
			" old style: calculated ="
			" " ULINTPF "; recorded checksum = " ULINTPF "\n",
			cur_page_num, old_checksum,
			checksum_field2);
		fprintf(log_file, "page::%llu;"
			" new style: calculated ="
			" " ULINTPF "; recorded checksum  = " ULINTPF "\n",
			cur_page_num, new_checksum,
			checksum_field1);
	}
#endif /* UNIV_INNOCHECKSUM */


	if (checksum_field2 != mach_read_from_4(read_buf + FIL_PAGE_LSN)
	    && checksum_field2 != old_checksum) {
		DBUG_LOG("checksum",
			 "Page checksum crc32 not valid"
			 << " field1 " << checksum_field1
			 << " field2 " << checksum_field2
			 << " crc32 " << buf_calc_page_old_checksum(read_buf)
			 << " lsn " << mach_read_from_4(
				 read_buf + FIL_PAGE_LSN));
		return(false);
	}

	/* old field is fine, check the new field */

	/* InnoDB versions < 4.0.14 and < 4.1.1 stored the space id
	(always equal to 0), to FIL_PAGE_SPACE_OR_CHKSUM */

	if (checksum_field1 != 0 && checksum_field1 != new_checksum) {
		DBUG_LOG("checksum",
			 "Page checksum crc32 not valid"
			 << " field1 " << checksum_field1
			 << " field2 " << checksum_field2
			 << " crc32 " << buf_calc_page_new_checksum(read_buf)
			 << " lsn " << mach_read_from_4(
				 read_buf + FIL_PAGE_LSN));
		return(false);
	}

	return(true);
}

/** Checks if the page is in none checksum format.
@param[in]	read_buf	database page
@param[in]	checksum_field1	new checksum field
@param[in]	checksum_field2	old checksum field
@return true if the page is in none checksum format. */
bool
buf_page_is_checksum_valid_none(
	const byte*			read_buf,
	ulint				checksum_field1,
	ulint				checksum_field2)
{
#ifndef DBUG_OFF
	if (checksum_field1 != checksum_field2
	    && checksum_field1 != BUF_NO_CHECKSUM_MAGIC) {
		DBUG_LOG("checksum",
			 "Page checksum crc32 not valid"
			 << " field1 " << checksum_field1
			 << " field2 " << checksum_field2
			 << " crc32 " << BUF_NO_CHECKSUM_MAGIC
			 << " lsn " << mach_read_from_4(read_buf
							+ FIL_PAGE_LSN));
	}
#endif /* DBUG_OFF */

#ifdef UNIV_INNOCHECKSUM
	if (log_file
	    && srv_checksum_algorithm == SRV_CHECKSUM_ALGORITHM_STRICT_NONE) {
		fprintf(log_file,
			"page::%llu; none checksum: calculated"
			" = %lu; recorded checksum_field1 = " ULINTPF
			" recorded checksum_field2 = " ULINTPF "\n",
			cur_page_num, BUF_NO_CHECKSUM_MAGIC,
			checksum_field1, checksum_field2);
	}
#endif /* UNIV_INNOCHECKSUM */

	return(checksum_field1 == checksum_field2
	       && checksum_field1 == BUF_NO_CHECKSUM_MAGIC);
}

#ifdef INNODB_BUG_ENDIAN_CRC32
/** Validate the CRC-32C checksum of a page.
@param[in]	page		buffer page (srv_page_size bytes)
@param[in]	checksum	CRC-32C checksum stored on page
@return	computed checksum */
static uint32_t buf_page_check_crc32(const byte* page, uint32_t checksum)
{
	uint32_t crc32 = buf_calc_page_crc32(page);

	if (checksum != crc32) {
		crc32 = buf_calc_page_crc32(page, true);
	}

	return crc32;
}
#else /* INNODB_BUG_ENDIAN_CRC32 */
/** Validate the CRC-32C checksum of a page.
@param[in]	page		buffer page (srv_page_size bytes)
@param[in]	checksum	CRC-32C checksum stored on page
@return	computed checksum */
# define buf_page_check_crc32(page, checksum) buf_calc_page_crc32(page)
#endif /* INNODB_BUG_ENDIAN_CRC32 */


/** Check if a buffer is all zeroes.
@param[in]	buf	data to check
@return whether the buffer is all zeroes */
bool buf_is_zeroes(span<const byte> buf)
{
  ut_ad(buf.size() <= sizeof field_ref_zero);
  return memcmp(buf.data(), field_ref_zero, buf.size()) == 0;
}

/** Check if a page is corrupt.
@param[in]	check_lsn	whether the LSN should be checked
@param[in]	read_buf	database page
@param[in]	page_size	page size
@param[in]	space		tablespace
@return whether the page is corrupted */
bool
buf_page_is_corrupted(
	bool			check_lsn,
	const byte*		read_buf,
	const page_size_t&	page_size,
#ifndef UNIV_INNOCHECKSUM
	const fil_space_t* 	space)
#else
	const void* 	 	space)
#endif
{
	ut_ad(page_size.logical() == srv_page_size);
#ifndef UNIV_INNOCHECKSUM
	DBUG_EXECUTE_IF("buf_page_import_corrupt_failure", return(true); );
#endif
	size_t		checksum_field1 = 0;
	size_t		checksum_field2 = 0;
	uint32_t	crc32 = 0;
	bool		crc32_inited = false;

	ulint page_type = mach_read_from_2(read_buf + FIL_PAGE_TYPE);

	/* We can trust page type if page compression is set on tablespace
	flags because page compression flag means file must have been
	created with 10.1 (later than 5.5 code base). In 10.1 page
	compressed tables do not contain post compression checksum and
	FIL_PAGE_END_LSN_OLD_CHKSUM field stored. Note that space can
	be null if we are in fil_check_first_page() and first page
	is not compressed or encrypted. Page checksum is verified
	after decompression (i.e. normally pages are already
	decompressed at this stage). */
	if ((page_type == FIL_PAGE_PAGE_COMPRESSED ||
	     page_type == FIL_PAGE_PAGE_COMPRESSED_ENCRYPTED)
#ifndef UNIV_INNOCHECKSUM
	    && space && FSP_FLAGS_HAS_PAGE_COMPRESSION(space->flags)
#endif
	) {
		return(false);
	}

	if (!page_size.is_compressed()
	    && memcmp(read_buf + FIL_PAGE_LSN + 4,
		      read_buf + page_size.logical()
		      - FIL_PAGE_END_LSN_OLD_CHKSUM + 4, 4)) {

		/* Stored log sequence numbers at the start and the end
		of page do not match */

		return(true);
	}

#ifndef UNIV_INNOCHECKSUM
	if (check_lsn && recv_lsn_checks_on) {
		lsn_t		current_lsn;
		const lsn_t	page_lsn
			= mach_read_from_8(read_buf + FIL_PAGE_LSN);

		/* Since we are going to reset the page LSN during the import
		phase it makes no sense to spam the log with error messages. */

		if (log_peek_lsn(&current_lsn) && current_lsn < page_lsn) {

			const ulint	space_id = mach_read_from_4(
				read_buf + FIL_PAGE_SPACE_ID);
			const ulint	page_no = mach_read_from_4(
				read_buf + FIL_PAGE_OFFSET);

			ib::error() << "Page " << page_id_t(space_id, page_no)
				<< " log sequence number " << page_lsn
				<< " is in the future! Current system"
				<< " log sequence number "
				<< current_lsn << ".";

			ib::error() << "Your database may be corrupt or"
				" you may have copied the InnoDB"
				" tablespace but not the InnoDB"
				" log files. "
				<< FORCE_RECOVERY_MSG;

		}
	}
#endif /* !UNIV_INNOCHECKSUM */

	/* Check whether the checksum fields have correct values */

	const srv_checksum_algorithm_t curr_algo =
		static_cast<srv_checksum_algorithm_t>(srv_checksum_algorithm);

	if (curr_algo == SRV_CHECKSUM_ALGORITHM_NONE) {
		return(false);
	}

	if (page_size.is_compressed()) {
		return(!page_zip_verify_checksum(read_buf,
						 page_size.physical()));
	}

	checksum_field1 = mach_read_from_4(
		read_buf + FIL_PAGE_SPACE_OR_CHKSUM);

	checksum_field2 = mach_read_from_4(
		read_buf + page_size.logical() - FIL_PAGE_END_LSN_OLD_CHKSUM);

#if FIL_PAGE_LSN % 8
#error "FIL_PAGE_LSN must be 64 bit aligned"
#endif

	/* A page filled with NUL bytes is considered not corrupted.
	The FIL_PAGE_FILE_FLUSH_LSN field may be written nonzero for
	the first page of the system tablespace.
	Ignore it for the system tablespace. */
	if (!checksum_field1 && !checksum_field2) {
		/* Checksum fields can have valid value as zero.
		If the page is not empty then do the checksum
		calculation for the page. */
		bool all_zeroes = true;
		for (size_t i = 0; i < srv_page_size; i++) {
#ifndef UNIV_INNOCHECKSUM
			if (i == FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION
			    && (!space || !space->id)) {
				i += 8;
			}
#endif
			if (read_buf[i]) {
				all_zeroes = false;
				break;
			}
		}

		if (all_zeroes) {
			return false;
		}
	}

	switch (curr_algo) {
	case SRV_CHECKSUM_ALGORITHM_STRICT_CRC32:
		return !buf_page_is_checksum_valid_crc32(
			read_buf, checksum_field1, checksum_field2);
	case SRV_CHECKSUM_ALGORITHM_STRICT_INNODB:
		return !buf_page_is_checksum_valid_innodb(
			read_buf, checksum_field1, checksum_field2);
	case SRV_CHECKSUM_ALGORITHM_STRICT_NONE:
		return !buf_page_is_checksum_valid_none(
			read_buf, checksum_field1, checksum_field2);
	case SRV_CHECKSUM_ALGORITHM_CRC32:
	case SRV_CHECKSUM_ALGORITHM_INNODB:
		if (buf_page_is_checksum_valid_none(read_buf,
			checksum_field1, checksum_field2)) {
#ifdef UNIV_INNOCHECKSUM
			if (log_file) {
				fprintf(log_file, "page::%llu;"
					" old style: calculated = %u;"
					" recorded = " ULINTPF ";\n",
					cur_page_num,
					buf_calc_page_old_checksum(read_buf),
					checksum_field2);
				fprintf(log_file, "page::%llu;"
					" new style: calculated = %u;"
					" crc32 = %u; recorded = " ULINTPF ";\n",
					cur_page_num,
					buf_calc_page_new_checksum(read_buf),
					buf_calc_page_crc32(read_buf),
					checksum_field1);
			}
#endif /* UNIV_INNOCHECKSUM */
			return false;
		}

		/* Very old versions of InnoDB only stored 8 byte lsn to the
		start and the end of the page. */

		/* Since innodb_checksum_algorithm is not strict_* allow
		any of the algos to match for the old field */

		if (checksum_field2
		    != mach_read_from_4(read_buf + FIL_PAGE_LSN)
		    && checksum_field2 != BUF_NO_CHECKSUM_MAGIC) {

			if (curr_algo == SRV_CHECKSUM_ALGORITHM_CRC32) {
				DBUG_EXECUTE_IF(
					"page_intermittent_checksum_mismatch", {
					static int page_counter;
					if (page_counter++ == 2) {
						checksum_field2++;
					}
				});

				crc32 = buf_page_check_crc32(read_buf,
							     checksum_field2);
				crc32_inited = true;

				if (checksum_field2 != crc32
				    && checksum_field2
				       != buf_calc_page_old_checksum(read_buf)) {
					return true;
				}
			} else {
				ut_ad(curr_algo
				      == SRV_CHECKSUM_ALGORITHM_INNODB);

				if (checksum_field2
				    != buf_calc_page_old_checksum(read_buf)) {
					crc32 = buf_page_check_crc32(
						read_buf, checksum_field2);
					crc32_inited = true;

					if (checksum_field2 != crc32) {
						return true;
					}
				}
			}
		}

		if (checksum_field1 == 0
		    || checksum_field1 == BUF_NO_CHECKSUM_MAGIC) {
		} else if (curr_algo == SRV_CHECKSUM_ALGORITHM_CRC32) {
			if (!crc32_inited) {
				crc32 = buf_page_check_crc32(
					read_buf, checksum_field2);
				crc32_inited = true;
			}

			if (checksum_field1 != crc32
			    && checksum_field1
			    != buf_calc_page_new_checksum(read_buf)) {
				return true;
			}
		} else {
			ut_ad(curr_algo == SRV_CHECKSUM_ALGORITHM_INNODB);

			if (checksum_field1
			    != buf_calc_page_new_checksum(read_buf)) {

				if (!crc32_inited) {
					crc32 = buf_page_check_crc32(
						read_buf, checksum_field2);
					crc32_inited = true;
				}

				if (checksum_field1 != crc32) {
					return true;
				}
			}
		}

		if (crc32_inited
		    && ((checksum_field1 == crc32
			 && checksum_field2 != crc32)
			|| (checksum_field1 != crc32
			    && checksum_field2 == crc32))) {
			return true;
		}

		break;
	case SRV_CHECKSUM_ALGORITHM_NONE:
		/* should have returned false earlier */
		break;
	}

	return false;
}

#ifndef UNIV_INNOCHECKSUM
/** Dump a page to stderr.
@param[in]	read_buf	database page
@param[in]	page_size	page size */
UNIV_INTERN
void
buf_page_print(const byte* read_buf, const page_size_t& page_size)
{
	dict_index_t*	index;

#ifndef UNIV_DEBUG
	ib::info() << "Page dump in ascii and hex ("
		<< page_size.physical() << " bytes):";

	ut_print_buf(stderr, read_buf, page_size.physical());
	fputs("\nInnoDB: End of page dump\n", stderr);
#endif

	if (page_size.is_compressed()) {
		/* Print compressed page. */
		ib::info() << "Compressed page type ("
			<< fil_page_get_type(read_buf)
			<< "); stored checksum in field1 "
			<< mach_read_from_4(
				read_buf + FIL_PAGE_SPACE_OR_CHKSUM)
			<< "; calculated checksums for field1: "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_CRC32)
			<< " "
			<< page_zip_calc_checksum(
				read_buf, page_size.physical(),
				SRV_CHECKSUM_ALGORITHM_CRC32)
#ifdef INNODB_BUG_ENDIAN_CRC32
			<< "/"
			<< page_zip_calc_checksum(
				read_buf, page_size.physical(),
				SRV_CHECKSUM_ALGORITHM_CRC32, true)
#endif
			<< ", "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_INNODB)
			<< " "
			<< page_zip_calc_checksum(
				read_buf, page_size.physical(),
				SRV_CHECKSUM_ALGORITHM_INNODB)
			<< ", "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_NONE)
			<< " "
			<< page_zip_calc_checksum(
				read_buf, page_size.physical(),
				SRV_CHECKSUM_ALGORITHM_NONE)
			<< "; page LSN "
			<< mach_read_from_8(read_buf + FIL_PAGE_LSN)
			<< "; page number (if stored to page"
			<< " already) "
			<< mach_read_from_4(read_buf + FIL_PAGE_OFFSET)
			<< "; space id (if stored to page already) "
			<< mach_read_from_4(
				read_buf + FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID);

	} else {
		const uint32_t	crc32 = buf_calc_page_crc32(read_buf);
#ifdef INNODB_BUG_ENDIAN_CRC32
		const uint32_t	crc32_legacy = buf_calc_page_crc32(read_buf,
								   true);
#endif /* INNODB_BUG_ENDIAN_CRC32 */
		ulint page_type = fil_page_get_type(read_buf);

		ib::info() << "Uncompressed page, stored checksum in field1 "
			<< mach_read_from_4(
				read_buf + FIL_PAGE_SPACE_OR_CHKSUM)
			<< ", calculated checksums for field1: "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_CRC32) << " "
			<< crc32
#ifdef INNODB_BUG_ENDIAN_CRC32
			<< "/" << crc32_legacy
#endif
			<< ", "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_INNODB) << " "
			<< buf_calc_page_new_checksum(read_buf)
			<< ", "
			<< " page type " << page_type << " == "
			<< fil_get_page_type_name(page_type) << "."
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_NONE) << " "
			<< BUF_NO_CHECKSUM_MAGIC
			<< ", stored checksum in field2 "
			<< mach_read_from_4(read_buf + page_size.logical()
					    - FIL_PAGE_END_LSN_OLD_CHKSUM)
			<< ", calculated checksums for field2: "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_CRC32) << " "
			<< crc32
#ifdef INNODB_BUG_ENDIAN_CRC32
			<< "/" << crc32_legacy
#endif
			<< ", "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_INNODB) << " "
			<< buf_calc_page_old_checksum(read_buf)
			<< ", "
			<< buf_checksum_algorithm_name(
				SRV_CHECKSUM_ALGORITHM_NONE) << " "
			<< BUF_NO_CHECKSUM_MAGIC
			<< ",  page LSN "
			<< mach_read_from_4(read_buf + FIL_PAGE_LSN)
			<< " "
			<< mach_read_from_4(read_buf + FIL_PAGE_LSN + 4)
			<< ", low 4 bytes of LSN at page end "
			<< mach_read_from_4(read_buf + page_size.logical()
					    - FIL_PAGE_END_LSN_OLD_CHKSUM + 4)
			<< ", page number (if stored to page already) "
			<< mach_read_from_4(read_buf + FIL_PAGE_OFFSET)
			<< ", space id (if created with >= MySQL-4.1.1"
			   " and stored already) "
			<< mach_read_from_4(
				read_buf + FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID);
	}

	if (mach_read_from_2(read_buf + TRX_UNDO_PAGE_HDR + TRX_UNDO_PAGE_TYPE)
	    == TRX_UNDO_INSERT) {
		fprintf(stderr,
			"InnoDB: Page may be an insert undo log page\n");
	} else if (mach_read_from_2(read_buf + TRX_UNDO_PAGE_HDR
				    + TRX_UNDO_PAGE_TYPE)
		   == TRX_UNDO_UPDATE) {
		fprintf(stderr,
			"InnoDB: Page may be an update undo log page\n");
	}

	switch (fil_page_get_type(read_buf)) {
		index_id_t	index_id;
	case FIL_PAGE_INDEX:
	case FIL_PAGE_RTREE:
		index_id = btr_page_get_index_id(read_buf);
		ib::info() << "Page may be an index page where"
			" index id is " << index_id;

		index = dict_index_find_on_id_low(index_id);
		if (index) {
			ib::info()
				<< "Index " << index_id
				<< " is " << index->name
				<< " in table " << index->table->name;
		}
		break;
	case FIL_PAGE_INODE:
		fputs("InnoDB: Page may be an 'inode' page\n", stderr);
		break;
	case FIL_PAGE_IBUF_FREE_LIST:
		fputs("InnoDB: Page may be an insert buffer free list page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_ALLOCATED:
		fputs("InnoDB: Page may be a freshly allocated page\n",
		      stderr);
		break;
	case FIL_PAGE_IBUF_BITMAP:
		fputs("InnoDB: Page may be an insert buffer bitmap page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_SYS:
		fputs("InnoDB: Page may be a system page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_TRX_SYS:
		fputs("InnoDB: Page may be a transaction system page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_FSP_HDR:
		fputs("InnoDB: Page may be a file space header page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_XDES:
		fputs("InnoDB: Page may be an extent descriptor page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_BLOB:
		fputs("InnoDB: Page may be a BLOB page\n",
		      stderr);
		break;
	case FIL_PAGE_TYPE_ZBLOB:
	case FIL_PAGE_TYPE_ZBLOB2:
		fputs("InnoDB: Page may be a compressed BLOB page\n",
		      stderr);
		break;
	}
}

# ifdef PFS_GROUP_BUFFER_SYNC
extern mysql_pfs_key_t	buffer_block_mutex_key;

/********************************************************************//**
This function registers mutexes and rwlocks in buffer blocks with
performance schema. If PFS_MAX_BUFFER_MUTEX_LOCK_REGISTER is
defined to be a value less than chunk->size, then only mutexes
and rwlocks in the first PFS_MAX_BUFFER_MUTEX_LOCK_REGISTER
blocks are registered. */
static
void
pfs_register_buffer_block(
/*======================*/
	buf_chunk_t*	chunk)		/*!< in/out: chunk of buffers */
{
	buf_block_t*    block;
	ulint		num_to_register;

	block = chunk->blocks;

	num_to_register = ut_min(
		chunk->size, PFS_MAX_BUFFER_MUTEX_LOCK_REGISTER);

	for (ulint i = 0; i < num_to_register; i++) {
#  ifdef UNIV_PFS_MUTEX
		BPageMutex*	mutex;

		mutex = &block->mutex;
		mutex->pfs_add(buffer_block_mutex_key);
#  endif /* UNIV_PFS_MUTEX */

		rw_lock_t*	rwlock;

#  ifdef UNIV_PFS_RWLOCK
		rwlock = &block->lock;
		ut_a(!rwlock->pfs_psi);
		rwlock->pfs_psi = (PSI_server)
			? PSI_server->init_rwlock(buf_block_lock_key, rwlock)
			: NULL;

#   ifdef UNIV_DEBUG
		rwlock = &block->debug_latch;
		ut_a(!rwlock->pfs_psi);
		rwlock->pfs_psi = (PSI_server)
			? PSI_server->init_rwlock(buf_block_debug_latch_key,
						  rwlock)
			: NULL;
#   endif /* UNIV_DEBUG */

#  endif /* UNIV_PFS_RWLOCK */
		block++;
	}
}
# endif /* PFS_GROUP_BUFFER_SYNC */

/********************************************************************//**
Initializes a buffer control block when the buf_pool is created. */
static
void
buf_block_init(
/*===========*/
	buf_pool_t*	buf_pool,	/*!< in: buffer pool instance */
	buf_block_t*	block,		/*!< in: pointer to control block */
	byte*		frame)		/*!< in: pointer to buffer frame */
{
	UNIV_MEM_DESC(frame, UNIV_PAGE_SIZE);

	/* This function should only be executed at database startup or by
	buf_pool_resize(). Either way, adaptive hash index must not exist. */
	assert_block_ahi_empty_on_init(block);

	block->frame = frame;

	block->page.buf_pool_index = buf_pool_index(buf_pool);
	block->page.flush_type = BUF_FLUSH_LRU;
	block->page.state = BUF_BLOCK_NOT_USED;
	block->page.buf_fix_count = 0;
	block->page.io_fix = BUF_IO_NONE;
	block->page.flush_observer = NULL;
	block->page.real_size = 0;
	block->modify_clock = 0;
	block->page.slot = NULL;

	ut_d(block->page.file_page_was_freed = FALSE);

#ifdef BTR_CUR_HASH_ADAPT
	block->index = NULL;
#endif /* BTR_CUR_HASH_ADAPT */
	block->skip_flush_check = false;

	ut_d(block->page.in_page_hash = FALSE);
	ut_d(block->page.in_zip_hash = FALSE);
	ut_d(block->page.in_flush_list = FALSE);
	ut_d(block->page.in_free_list = FALSE);
	ut_d(block->page.in_LRU_list = FALSE);
	ut_d(block->in_unzip_LRU_list = FALSE);
	ut_d(block->in_withdraw_list = FALSE);

	page_zip_des_init(&block->page.zip);

	mutex_create(LATCH_ID_BUF_BLOCK_MUTEX, &block->mutex);

#if defined PFS_SKIP_BUFFER_MUTEX_RWLOCK || defined PFS_GROUP_BUFFER_SYNC
	/* If PFS_SKIP_BUFFER_MUTEX_RWLOCK is defined, skip registration
	of buffer block rwlock with performance schema.

	If PFS_GROUP_BUFFER_SYNC is defined, skip the registration
	since buffer block rwlock will be registered later in
	pfs_register_buffer_block(). */

	rw_lock_create(PFS_NOT_INSTRUMENTED, &block->lock, SYNC_LEVEL_VARYING);

	ut_d(rw_lock_create(PFS_NOT_INSTRUMENTED, &block->debug_latch,
			    SYNC_LEVEL_VARYING));

#else /* PFS_SKIP_BUFFER_MUTEX_RWLOCK || PFS_GROUP_BUFFER_SYNC */

	rw_lock_create(buf_block_lock_key, &block->lock, SYNC_LEVEL_VARYING);

	ut_d(rw_lock_create(buf_block_debug_latch_key,
			    &block->debug_latch, SYNC_LEVEL_VARYING));

#endif /* PFS_SKIP_BUFFER_MUTEX_RWLOCK || PFS_GROUP_BUFFER_SYNC */

	block->lock.is_block_lock = 1;

	ut_ad(rw_lock_validate(&(block->lock)));
}

/********************************************************************//**
Allocates a chunk of buffer frames.
@return chunk, or NULL on failure */
static
buf_chunk_t*
buf_chunk_init(
/*===========*/
	buf_pool_t*	buf_pool,	/*!< in: buffer pool instance */
	buf_chunk_t*	chunk,		/*!< out: chunk of buffers */
	ulint		mem_size)	/*!< in: requested size in bytes */
{
	buf_block_t*	block;
	byte*		frame;
	ulint		i;

	/* Round down to a multiple of page size,
	although it already should be. */
	mem_size = ut_2pow_round(mem_size, UNIV_PAGE_SIZE);
	/* Reserve space for the block descriptors. */
	mem_size += ut_2pow_round((mem_size / UNIV_PAGE_SIZE) * (sizeof *block)
				  + (UNIV_PAGE_SIZE - 1), UNIV_PAGE_SIZE);

	DBUG_EXECUTE_IF("ib_buf_chunk_init_fails", return(NULL););

	chunk->mem = buf_pool->allocator.allocate_large(mem_size,
							&chunk->mem_pfx);

	if (UNIV_UNLIKELY(chunk->mem == NULL)) {

		return(NULL);
	}

#ifdef HAVE_LIBNUMA
	if (srv_numa_interleave) {
		struct bitmask *numa_mems_allowed = numa_get_mems_allowed();
		int	st = mbind(chunk->mem, chunk->mem_size(),
				   MPOL_INTERLEAVE,
				   numa_mems_allowed->maskp,
				   numa_mems_allowed->size,
				   MPOL_MF_MOVE);
		if (st != 0) {
			ib::warn() << "Failed to set NUMA memory policy of"
				" buffer pool page frames to MPOL_INTERLEAVE"
				" (error: " << strerror(errno) << ").";
		}
		numa_bitmask_free(numa_mems_allowed);
	}
#endif /* HAVE_LIBNUMA */


	/* Allocate the block descriptors from
	the start of the memory block. */
	chunk->blocks = (buf_block_t*) chunk->mem;

	/* Align a pointer to the first frame.  Note that when
	opt_large_page_size is smaller than UNIV_PAGE_SIZE,
	we may allocate one fewer block than requested.  When
	it is bigger, we may allocate more blocks than requested. */

	frame = (byte*) ut_align(chunk->mem, UNIV_PAGE_SIZE);
	chunk->size = chunk->mem_pfx.m_size / UNIV_PAGE_SIZE
		- (frame != chunk->mem);

	/* Subtract the space needed for block descriptors. */
	{
		ulint	size = chunk->size;

		while (frame < (byte*) (chunk->blocks + size)) {
			frame += UNIV_PAGE_SIZE;
			size--;
		}

		chunk->size = size;
	}

	/* Init block structs and assign frames for them. Then we
	assign the frames to the first blocks (we already mapped the
	memory above). */

	block = chunk->blocks;

	for (i = chunk->size; i--; ) {

		buf_block_init(buf_pool, block, frame);
		UNIV_MEM_INVALID(block->frame, UNIV_PAGE_SIZE);

		/* Add the block to the free list */
		UT_LIST_ADD_LAST(buf_pool->free, &block->page);

		ut_d(block->page.in_free_list = TRUE);
		ut_ad(buf_pool_from_block(block) == buf_pool);

		block++;
		frame += UNIV_PAGE_SIZE;
	}

	buf_pool_register_chunk(chunk);

#ifdef PFS_GROUP_BUFFER_SYNC
	pfs_register_buffer_block(chunk);
#endif /* PFS_GROUP_BUFFER_SYNC */
	return(chunk);
}

#ifdef UNIV_DEBUG
/*********************************************************************//**
Finds a block in the given buffer chunk that points to a
given compressed page.
@return buffer block pointing to the compressed page, or NULL */
static
buf_block_t*
buf_chunk_contains_zip(
/*===================*/
	buf_chunk_t*	chunk,	/*!< in: chunk being checked */
	const void*	data)	/*!< in: pointer to compressed page */
{
	buf_block_t*	block;
	ulint		i;

	block = chunk->blocks;

	for (i = chunk->size; i--; block++) {
		if (block->page.zip.data == data) {

			return(block);
		}
	}

	return(NULL);
}

/*********************************************************************//**
Finds a block in the buffer pool that points to a
given compressed page.
@return buffer block pointing to the compressed page, or NULL */
buf_block_t*
buf_pool_contains_zip(
/*==================*/
	buf_pool_t*	buf_pool,	/*!< in: buffer pool instance */
	const void*	data)		/*!< in: pointer to compressed page */
{
	ulint		n;
	buf_chunk_t*	chunk = buf_pool->chunks;

	ut_ad(buf_pool);
	ut_ad(buf_pool_mutex_own(buf_pool));
	for (n = buf_pool->n_chunks; n--; chunk++) {

		buf_block_t* block = buf_chunk_contains_zip(chunk, data);

		if (block) {
			return(block);
		}
	}

	return(NULL);
}
#endif /* UNIV_DEBUG */

/*********************************************************************//**
Checks that all file pages in the buffer chunk are in a replaceable state.
@return address of a non-free block, or NULL if all freed */
static
const buf_block_t*
buf_chunk_not_freed(
/*================*/
	buf_chunk_t*	chunk)	/*!< in: chunk being checked */
{
	buf_block_t*	block;
	ulint		i;

	block = chunk->blocks;

	for (i = chunk->size; i--; block++) {
		ibool	ready;

		switch (buf_block_get_state(block)) {
		case BUF_BLOCK_POOL_WATCH:
		case BUF_BLOCK_ZIP_PAGE:
		case BUF_BLOCK_ZIP_DIRTY:
			/* The uncompressed buffer pool should never
			contain compressed block descriptors. */
			ut_error;
			break;
		case BUF_BLOCK_NOT_USED:
		case BUF_BLOCK_READY_FOR_USE:
		case BUF_BLOCK_MEMORY:
		case BUF_BLOCK_REMOVE_HASH:
			/* Skip blocks that are not being used for
			file pages. */
			break;
		case BUF_BLOCK_FILE_PAGE:
			if (srv_read_only_mode) {
				/* The page cleaner is disabled in
				read-only mode.  No pages can be
				dirtied, so all of them must be clean. */
				ut_ad(block->page.oldest_modification
				      == block->page.newest_modification);
				ut_ad(block->page.oldest_modification == 0
				      || block->page.oldest_modification
				      == recv_sys->recovered_lsn
				      || srv_force_recovery
				      == SRV_FORCE_NO_LOG_REDO);
				ut_ad(block->page.buf_fix_count == 0);
				ut_ad(block->page.io_fix == BUF_IO_NONE);
				break;
			}

			buf_page_mutex_enter(block);
			ready = buf_flush_ready_for_replace(&block->page);
			buf_page_mutex_exit(block);

			if (!ready) {
				return(block);
			}

			break;
		}
	}

	return(NULL);
}

/********************************************************************//**
Set buffer pool size variables after resizing it */
static
void
buf_pool_set_sizes(void)
/*====================*/
{
	ulint	i;
	ulint	curr_size = 0;

	buf_pool_mutex_enter_all();

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);
		curr_size += buf_pool->curr_pool_size;
	}

	srv_buf_pool_curr_size = curr_size;
	srv_buf_pool_old_size = srv_buf_pool_size;
	srv_buf_pool_base_size = srv_buf_pool_size;

	buf_pool_mutex_exit_all();
}

/********************************************************************//**
Initialize a buffer pool instance.
@return DB_SUCCESS if all goes well. */
static
ulint
buf_pool_init_instance(
/*===================*/
	buf_pool_t*	buf_pool,	/*!< in: buffer pool instance */
	ulint		buf_pool_size,	/*!< in: size in bytes */
	ulint		instance_no)	/*!< in: id of the instance */
{
	ulint		i;
	ulint		chunk_size;
	buf_chunk_t*	chunk;

	ut_ad(buf_pool_size % srv_buf_pool_chunk_unit == 0);

	/* 1. Initialize general fields
	------------------------------- */
	mutex_create(LATCH_ID_BUF_POOL, &buf_pool->mutex);

	mutex_create(LATCH_ID_BUF_POOL_ZIP, &buf_pool->zip_mutex);

	new(&buf_pool->allocator)
		ut_allocator<unsigned char>(mem_key_buf_buf_pool);

	buf_pool_mutex_enter(buf_pool);

	if (buf_pool_size > 0) {
		buf_pool->n_chunks
			= buf_pool_size / srv_buf_pool_chunk_unit;
		chunk_size = srv_buf_pool_chunk_unit;

		buf_pool->chunks =
			reinterpret_cast<buf_chunk_t*>(ut_zalloc_nokey(
				buf_pool->n_chunks * sizeof(*chunk)));
		buf_pool->chunks_old = NULL;

		UT_LIST_INIT(buf_pool->LRU, &buf_page_t::LRU);
		UT_LIST_INIT(buf_pool->free, &buf_page_t::list);
		UT_LIST_INIT(buf_pool->withdraw, &buf_page_t::list);
		buf_pool->withdraw_target = 0;
		UT_LIST_INIT(buf_pool->flush_list, &buf_page_t::list);
		UT_LIST_INIT(buf_pool->unzip_LRU, &buf_block_t::unzip_LRU);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
		UT_LIST_INIT(buf_pool->zip_clean, &buf_page_t::list);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

		for (i = 0; i < UT_ARR_SIZE(buf_pool->zip_free); ++i) {
			UT_LIST_INIT(
				buf_pool->zip_free[i], &buf_buddy_free_t::list);
		}

		buf_pool->curr_size = 0;
		chunk = buf_pool->chunks;

		do {
			if (!buf_chunk_init(buf_pool, chunk, chunk_size)) {
				while (--chunk >= buf_pool->chunks) {
					buf_block_t*	block = chunk->blocks;

					for (i = chunk->size; i--; block++) {
						mutex_free(&block->mutex);
						rw_lock_free(&block->lock);

						ut_d(rw_lock_free(
							&block->debug_latch));
					}

					buf_pool->allocator.deallocate_large(
						chunk->mem, &chunk->mem_pfx);
				}
				ut_free(buf_pool->chunks);
				buf_pool_mutex_exit(buf_pool);

				return(DB_ERROR);
			}

			buf_pool->curr_size += chunk->size;
		} while (++chunk < buf_pool->chunks + buf_pool->n_chunks);

		buf_pool->instance_no = instance_no;
		buf_pool->read_ahead_area =
			ut_min(BUF_READ_AHEAD_PAGES,
			       ut_2_power_up(buf_pool->curr_size /
					     BUF_READ_AHEAD_PORTION));
		buf_pool->curr_pool_size = buf_pool->curr_size * UNIV_PAGE_SIZE;

		buf_pool->old_size = buf_pool->curr_size;
		buf_pool->n_chunks_new = buf_pool->n_chunks;

		/* Number of locks protecting page_hash must be a
		power of two */
		srv_n_page_hash_locks = static_cast<ulong>(
			 ut_2_power_up(srv_n_page_hash_locks));
		ut_a(srv_n_page_hash_locks != 0);
		ut_a(srv_n_page_hash_locks <= MAX_PAGE_HASH_LOCKS);

		buf_pool->page_hash = ib_create(
			2 * buf_pool->curr_size,
			LATCH_ID_HASH_TABLE_RW_LOCK,
			srv_n_page_hash_locks, MEM_HEAP_FOR_PAGE_HASH);

		buf_pool->page_hash_old = NULL;

		buf_pool->zip_hash = hash_create(2 * buf_pool->curr_size);

		buf_pool->last_printout_time = time(NULL);
	}
	/* 2. Initialize flushing fields
	-------------------------------- */

	mutex_create(LATCH_ID_FLUSH_LIST, &buf_pool->flush_list_mutex);

	for (i = BUF_FLUSH_LRU; i < BUF_FLUSH_N_TYPES; i++) {
		buf_pool->no_flush[i] = os_event_create(0);
	}

	buf_pool->watch = (buf_page_t*) ut_zalloc_nokey(
		sizeof(*buf_pool->watch) * BUF_POOL_WATCH_SIZE);
	for (i = 0; i < BUF_POOL_WATCH_SIZE; i++) {
		buf_pool->watch[i].buf_pool_index
			= unsigned(buf_pool->instance_no);
	}

	/* All fields are initialized by ut_zalloc_nokey(). */

	buf_pool->try_LRU_scan = TRUE;

	/* Initialize the hazard pointer for flush_list batches */
	new(&buf_pool->flush_hp)
		FlushHp(buf_pool, &buf_pool->flush_list_mutex);

	/* Initialize the hazard pointer for LRU batches */
	new(&buf_pool->lru_hp) LRUHp(buf_pool, &buf_pool->mutex);

	/* Initialize the iterator for LRU scan search */
	new(&buf_pool->lru_scan_itr) LRUItr(buf_pool, &buf_pool->mutex);

	/* Initialize the iterator for single page scan search */
	new(&buf_pool->single_scan_itr) LRUItr(buf_pool, &buf_pool->mutex);

	/* Initialize the temporal memory array and slots */
	buf_pool->tmp_arr = (buf_tmp_array_t *)ut_malloc_nokey(sizeof(buf_tmp_array_t));
	memset(buf_pool->tmp_arr, 0, sizeof(buf_tmp_array_t));
	ulint n_slots = (srv_n_read_io_threads + srv_n_write_io_threads) * (8 * OS_AIO_N_PENDING_IOS_PER_THREAD);
	buf_pool->tmp_arr->n_slots = n_slots;
	buf_pool->tmp_arr->slots = (buf_tmp_buffer_t*)ut_malloc_nokey(sizeof(buf_tmp_buffer_t) * n_slots);
	memset(buf_pool->tmp_arr->slots, 0, (sizeof(buf_tmp_buffer_t) * n_slots));

	buf_pool_mutex_exit(buf_pool);

	DBUG_EXECUTE_IF("buf_pool_init_instance_force_oom",
		return(DB_ERROR); );

	return(DB_SUCCESS);
}

/********************************************************************//**
free one buffer pool instance */
static
void
buf_pool_free_instance(
/*===================*/
	buf_pool_t*	buf_pool)	/* in,own: buffer pool instance
					to free */
{
	buf_chunk_t*	chunk;
	buf_chunk_t*	chunks;
	buf_page_t*	bpage;
	buf_page_t*	prev_bpage = 0;

	mutex_free(&buf_pool->mutex);
	mutex_free(&buf_pool->zip_mutex);
	mutex_free(&buf_pool->flush_list_mutex);

	if (buf_pool->flush_rbt) {
		rbt_free(buf_pool->flush_rbt);
		buf_pool->flush_rbt = NULL;
	}

	for (bpage = UT_LIST_GET_LAST(buf_pool->LRU);
	     bpage != NULL;
	     bpage = prev_bpage) {

		prev_bpage = UT_LIST_GET_PREV(LRU, bpage);
		buf_page_state	state = buf_page_get_state(bpage);

		ut_ad(buf_page_in_file(bpage));
		ut_ad(bpage->in_LRU_list);

		if (state != BUF_BLOCK_FILE_PAGE) {
			/* We must not have any dirty block except
			when doing a fast shutdown. */
			ut_ad(state == BUF_BLOCK_ZIP_PAGE
			      || srv_fast_shutdown == 2);
			buf_page_free_descriptor(bpage);
		}
	}

	ut_free(buf_pool->watch);
	buf_pool->watch = NULL;

	chunks = buf_pool->chunks;
	chunk = chunks + buf_pool->n_chunks;

	while (--chunk >= chunks) {
		buf_block_t*	block = chunk->blocks;

		for (ulint i = chunk->size; i--; block++) {
			mutex_free(&block->mutex);
			rw_lock_free(&block->lock);

			ut_d(rw_lock_free(&block->debug_latch));
		}

		buf_pool->allocator.deallocate_large(
			chunk->mem, &chunk->mem_pfx);
	}

	for (ulint i = BUF_FLUSH_LRU; i < BUF_FLUSH_N_TYPES; ++i) {
		os_event_destroy(buf_pool->no_flush[i]);
	}

	ut_free(buf_pool->chunks);
	ha_clear(buf_pool->page_hash);
	hash_table_free(buf_pool->page_hash);
	hash_table_free(buf_pool->zip_hash);

	/* Free all used temporary slots */
	if (buf_pool->tmp_arr) {
		for(ulint i = 0; i < buf_pool->tmp_arr->n_slots; i++) {
			buf_tmp_buffer_t* slot = &(buf_pool->tmp_arr->slots[i]);
			if (slot && slot->crypt_buf) {
				aligned_free(slot->crypt_buf);
				slot->crypt_buf = NULL;
			}

			if (slot && slot->comp_buf) {
				aligned_free(slot->comp_buf);
				slot->comp_buf = NULL;
			}
		}

		ut_free(buf_pool->tmp_arr->slots);
		ut_free(buf_pool->tmp_arr);
		buf_pool->tmp_arr = NULL;
	}

	buf_pool->allocator.~ut_allocator();
}

/********************************************************************//**
Creates the buffer pool.
@return DB_SUCCESS if success, DB_ERROR if not enough memory or error */
dberr_t
buf_pool_init(
/*==========*/
	ulint	total_size,	/*!< in: size of the total pool in bytes */
	ulint	n_instances)	/*!< in: number of instances */
{
	ulint		i;
	const ulint	size	= total_size / n_instances;

	ut_ad(n_instances > 0);
	ut_ad(n_instances <= MAX_BUFFER_POOLS);
	ut_ad(n_instances == srv_buf_pool_instances);

	NUMA_MEMPOLICY_INTERLEAVE_IN_SCOPE;

	buf_pool_resizing = false;
	buf_pool_withdrawing = false;
	buf_withdraw_clock = 0;

	buf_pool_ptr = (buf_pool_t*) ut_zalloc_nokey(
		n_instances * sizeof *buf_pool_ptr);

	buf_chunk_map_reg = UT_NEW_NOKEY(buf_pool_chunk_map_t());

	for (i = 0; i < n_instances; i++) {
		buf_pool_t*	ptr	= &buf_pool_ptr[i];

		if (buf_pool_init_instance(ptr, size, i) != DB_SUCCESS) {

			/* Free all the instances created so far. */
			buf_pool_free(i);

			return(DB_ERROR);
		}
	}

	buf_chunk_map_ref = buf_chunk_map_reg;

	buf_pool_set_sizes();
	buf_LRU_old_ratio_update(100 * 3/ 8, FALSE);

	btr_search_sys_create(buf_pool_get_curr_size() / sizeof(void*) / 64);

	return(DB_SUCCESS);
}

/********************************************************************//**
Frees the buffer pool at shutdown.  This must not be invoked before
freeing all mutexes. */
void
buf_pool_free(
/*==========*/
	ulint	n_instances)	/*!< in: numbere of instances to free */
{
	for (ulint i = 0; i < n_instances; i++) {
		buf_pool_free_instance(buf_pool_from_array(i));
	}

	UT_DELETE(buf_chunk_map_reg);
	buf_chunk_map_reg = buf_chunk_map_ref = NULL;

	ut_free(buf_pool_ptr);
	buf_pool_ptr = NULL;
}

/** Reallocate a control block.
@param[in]	buf_pool	buffer pool instance
@param[in]	block		pointer to control block
@retval false	if failed because of no free blocks. */
static
bool
buf_page_realloc(
	buf_pool_t*	buf_pool,
	buf_block_t*	block)
{
	buf_block_t*	new_block;

	ut_ad(buf_pool_withdrawing);
	ut_ad(buf_pool_mutex_own(buf_pool));
	ut_ad(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);

	new_block = buf_LRU_get_free_only(buf_pool);

	if (new_block == NULL) {
		return(false); /* free_list was not enough */
	}

	rw_lock_t*	hash_lock = buf_page_hash_lock_get(buf_pool, block->page.id);

	rw_lock_x_lock(hash_lock);
	mutex_enter(&block->mutex);

	if (buf_page_can_relocate(&block->page)) {
		mutex_enter(&new_block->mutex);

		memcpy(new_block->frame, block->frame, srv_page_size);
		new (&new_block->page) buf_page_t(block->page);

		/* relocate LRU list */
		ut_ad(block->page.in_LRU_list);
		ut_ad(!block->page.in_zip_hash);
		ut_d(block->page.in_LRU_list = FALSE);

		buf_LRU_adjust_hp(buf_pool, &block->page);

		buf_page_t*	prev_b = UT_LIST_GET_PREV(LRU, &block->page);
		UT_LIST_REMOVE(buf_pool->LRU, &block->page);

		if (prev_b != NULL) {
			UT_LIST_INSERT_AFTER(buf_pool->LRU, prev_b, &new_block->page);
		} else {
			UT_LIST_ADD_FIRST(buf_pool->LRU, &new_block->page);
		}

		if (buf_pool->LRU_old == &block->page) {
			buf_pool->LRU_old = &new_block->page;
		}

		ut_ad(new_block->page.in_LRU_list);

		/* relocate unzip_LRU list */
		if (block->page.zip.data != NULL) {
			ut_ad(block->in_unzip_LRU_list);
			ut_d(new_block->in_unzip_LRU_list = TRUE);
			UNIV_MEM_DESC(&new_block->page.zip.data,
				      page_zip_get_size(&new_block->page.zip));

			buf_block_t*	prev_block = UT_LIST_GET_PREV(unzip_LRU, block);
			UT_LIST_REMOVE(buf_pool->unzip_LRU, block);

			ut_d(block->in_unzip_LRU_list = FALSE);
			block->page.zip.data = NULL;
			page_zip_set_size(&block->page.zip, 0);

			if (prev_block != NULL) {
				UT_LIST_INSERT_AFTER(buf_pool->unzip_LRU, prev_block, new_block);
			} else {
				UT_LIST_ADD_FIRST(buf_pool->unzip_LRU, new_block);
			}
		} else {
			ut_ad(!block->in_unzip_LRU_list);
			ut_d(new_block->in_unzip_LRU_list = FALSE);
		}

		/* relocate buf_pool->page_hash */
		ut_ad(block->page.in_page_hash);
		ut_ad(&block->page == buf_page_hash_get_low(buf_pool,
							    block->page.id));
		ut_d(block->page.in_page_hash = FALSE);
		ulint	fold = block->page.id.fold();
		ut_ad(fold == new_block->page.id.fold());
		HASH_REPLACE(buf_page_t, hash, buf_pool->page_hash, fold,
			     &block->page, &new_block->page);

		ut_ad(new_block->page.in_page_hash);

		buf_block_modify_clock_inc(block);
		memset(block->frame + FIL_PAGE_OFFSET, 0xff, 4);
		memset(block->frame + FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID, 0xff, 4);
		UNIV_MEM_INVALID(block->frame, UNIV_PAGE_SIZE);
		buf_block_set_state(block, BUF_BLOCK_REMOVE_HASH);
		block->page.id
		    = page_id_t(ULINT32_UNDEFINED, ULINT32_UNDEFINED);

		/* Relocate buf_pool->flush_list. */
		if (block->page.oldest_modification) {
			buf_flush_relocate_on_flush_list(
				&block->page, &new_block->page);
		}

		/* set other flags of buf_block_t */

#ifdef BTR_CUR_HASH_ADAPT
		/* This code should only be executed by buf_pool_resize(),
		while the adaptive hash index is disabled. */
		assert_block_ahi_empty(block);
		assert_block_ahi_empty_on_init(new_block);
		ut_ad(!block->index);
		new_block->index	= NULL;
		new_block->n_hash_helps	= 0;
		new_block->n_fields	= 1;
		new_block->left_side	= TRUE;
#endif /* BTR_CUR_HASH_ADAPT */

		new_block->lock_hash_val = block->lock_hash_val;
		ut_ad(new_block->lock_hash_val == lock_rec_hash(
			new_block->page.id.space(),
			new_block->page.id.page_no()));

		rw_lock_x_unlock(hash_lock);
		mutex_exit(&new_block->mutex);

		/* free block */
		buf_block_set_state(block, BUF_BLOCK_MEMORY);
		buf_LRU_block_free_non_file_page(block);

		mutex_exit(&block->mutex);
	} else {
		rw_lock_x_unlock(hash_lock);
		mutex_exit(&block->mutex);

		/* free new_block */
		mutex_enter(&new_block->mutex);
		buf_LRU_block_free_non_file_page(new_block);
		mutex_exit(&new_block->mutex);
	}

	return(true); /* free_list was enough */
}

/** Sets the global variable that feeds MySQL's innodb_buffer_pool_resize_status
to the specified string. The format and the following parameters are the
same as the ones used for printf(3).
@param[in]	fmt	format
@param[in]	...	extra parameters according to fmt */
static
void
buf_resize_status(
	const char*	fmt,
	...)
{
	va_list	ap;

	va_start(ap, fmt);

	vsnprintf(
		export_vars.innodb_buffer_pool_resize_status,
		sizeof(export_vars.innodb_buffer_pool_resize_status),
		fmt, ap);

	va_end(ap);

	ib::info() << export_vars.innodb_buffer_pool_resize_status;
}

/** Determines if a block is intended to be withdrawn.
@param[in]	buf_pool	buffer pool instance
@param[in]	block		pointer to control block
@retval true	if will be withdrawn */
bool
buf_block_will_withdrawn(
	buf_pool_t*		buf_pool,
	const buf_block_t*	block)
{
	ut_ad(buf_pool->curr_size < buf_pool->old_size);
	ut_ad(!buf_pool_resizing || buf_pool_mutex_own(buf_pool));

	const buf_chunk_t*	chunk
		= buf_pool->chunks + buf_pool->n_chunks_new;
	const buf_chunk_t*	echunk
		= buf_pool->chunks + buf_pool->n_chunks;

	while (chunk < echunk) {
		if (block >= chunk->blocks
		    && block < chunk->blocks + chunk->size) {
			return(true);
		}
		++chunk;
	}

	return(false);
}

/** Determines if a frame is intended to be withdrawn.
@param[in]	buf_pool	buffer pool instance
@param[in]	ptr		pointer to a frame
@retval true	if will be withdrawn */
bool
buf_frame_will_withdrawn(
	buf_pool_t*	buf_pool,
	const byte*	ptr)
{
	ut_ad(buf_pool->curr_size < buf_pool->old_size);
	ut_ad(!buf_pool_resizing || buf_pool_mutex_own(buf_pool));

	const buf_chunk_t*	chunk
		= buf_pool->chunks + buf_pool->n_chunks_new;
	const buf_chunk_t*	echunk
		= buf_pool->chunks + buf_pool->n_chunks;

	while (chunk < echunk) {
		if (ptr >= chunk->blocks->frame
		    && ptr < (chunk->blocks + chunk->size - 1)->frame
			     + UNIV_PAGE_SIZE) {
			return(true);
		}
		++chunk;
	}

	return(false);
}

/** Withdraw the buffer pool blocks from end of the buffer pool instance
until withdrawn by buf_pool->withdraw_target.
@param[in]	buf_pool	buffer pool instance
@retval true	if retry is needed */
static
bool
buf_pool_withdraw_blocks(
	buf_pool_t*	buf_pool)
{
	buf_block_t*	block;
	ulint		loop_count = 0;
	ulint		i = buf_pool_index(buf_pool);

	ib::info() << "buffer pool " << i
		<< " : start to withdraw the last "
		<< buf_pool->withdraw_target << " blocks.";

	/* Minimize buf_pool->zip_free[i] lists */
	buf_pool_mutex_enter(buf_pool);
	buf_buddy_condense_free(buf_pool);
	buf_pool_mutex_exit(buf_pool);

	while (UT_LIST_GET_LEN(buf_pool->withdraw)
	       < buf_pool->withdraw_target) {

		/* try to withdraw from free_list */
		ulint	count1 = 0;

		buf_pool_mutex_enter(buf_pool);
		block = reinterpret_cast<buf_block_t*>(
			UT_LIST_GET_FIRST(buf_pool->free));
		while (block != NULL
		       && UT_LIST_GET_LEN(buf_pool->withdraw)
			  < buf_pool->withdraw_target) {
			ut_ad(block->page.in_free_list);
			ut_ad(!block->page.in_flush_list);
			ut_ad(!block->page.in_LRU_list);
			ut_a(!buf_page_in_file(&block->page));

			buf_block_t*	next_block;
			next_block = reinterpret_cast<buf_block_t*>(
				UT_LIST_GET_NEXT(
					list, &block->page));

			if (buf_block_will_withdrawn(buf_pool, block)) {
				/* This should be withdrawn */
				UT_LIST_REMOVE(
					buf_pool->free,
					&block->page);
				UT_LIST_ADD_LAST(
					buf_pool->withdraw,
					&block->page);
				ut_d(block->in_withdraw_list = TRUE);
				count1++;
			}

			block = next_block;
		}
		buf_pool_mutex_exit(buf_pool);

		/* reserve free_list length */
		if (UT_LIST_GET_LEN(buf_pool->withdraw)
		    < buf_pool->withdraw_target) {
			ulint	scan_depth;
			flush_counters_t n;

			/* cap scan_depth with current LRU size. */
			buf_pool_mutex_enter(buf_pool);
			scan_depth = UT_LIST_GET_LEN(buf_pool->LRU);
			buf_pool_mutex_exit(buf_pool);

			scan_depth = ut_min(
				ut_max(buf_pool->withdraw_target
				       - UT_LIST_GET_LEN(buf_pool->withdraw),
				       static_cast<ulint>(srv_LRU_scan_depth)),
				scan_depth);

			buf_flush_do_batch(buf_pool, BUF_FLUSH_LRU,
				scan_depth, 0, &n);
			buf_flush_wait_batch_end(buf_pool, BUF_FLUSH_LRU);

			if (n.flushed) {
				MONITOR_INC_VALUE_CUMULATIVE(
					MONITOR_LRU_BATCH_FLUSH_TOTAL_PAGE,
					MONITOR_LRU_BATCH_FLUSH_COUNT,
					MONITOR_LRU_BATCH_FLUSH_PAGES,
					n.flushed);
			}
		}

		/* relocate blocks/buddies in withdrawn area */
		ulint	count2 = 0;

		buf_pool_mutex_enter(buf_pool);
		buf_page_t*	bpage;
		bpage = UT_LIST_GET_FIRST(buf_pool->LRU);
		while (bpage != NULL) {
			BPageMutex*	block_mutex;
			buf_page_t*	next_bpage;

			block_mutex = buf_page_get_mutex(bpage);
			mutex_enter(block_mutex);

			next_bpage = UT_LIST_GET_NEXT(LRU, bpage);

			if (bpage->zip.data != NULL
			    && buf_frame_will_withdrawn(
				buf_pool,
				static_cast<byte*>(bpage->zip.data))) {

				if (buf_page_can_relocate(bpage)) {
					mutex_exit(block_mutex);
					buf_pool_mutex_exit_forbid(buf_pool);
					if(!buf_buddy_realloc(
						buf_pool, bpage->zip.data,
						page_zip_get_size(
							&bpage->zip))) {

						/* failed to allocate block */
						buf_pool_mutex_exit_allow(
							buf_pool);
						break;
					}
					buf_pool_mutex_exit_allow(buf_pool);
					mutex_enter(block_mutex);
					count2++;
				}
				/* NOTE: if the page is in use,
				not reallocated yet */
			}

			if (buf_page_get_state(bpage)
			    == BUF_BLOCK_FILE_PAGE
			    && buf_block_will_withdrawn(
				buf_pool,
				reinterpret_cast<buf_block_t*>(bpage))) {

				if (buf_page_can_relocate(bpage)) {
					mutex_exit(block_mutex);
					buf_pool_mutex_exit_forbid(buf_pool);
					if(!buf_page_realloc(
						buf_pool,
						reinterpret_cast<buf_block_t*>(
							bpage))) {
						/* failed to allocate block */
						buf_pool_mutex_exit_allow(
							buf_pool);
						break;
					}
					buf_pool_mutex_exit_allow(buf_pool);
					count2++;
				} else {
					mutex_exit(block_mutex);
				}
				/* NOTE: if the page is in use,
				not reallocated yet */
			} else {
				mutex_exit(block_mutex);
			}

			bpage = next_bpage;
		}
		buf_pool_mutex_exit(buf_pool);

		buf_resize_status(
			"buffer pool %lu : withdrawing blocks. (%lu/%lu)",
			i, UT_LIST_GET_LEN(buf_pool->withdraw),
			buf_pool->withdraw_target);

		ib::info() << "buffer pool " << i << " : withdrew "
			<< count1 << " blocks from free list."
			<< " Tried to relocate " << count2 << " pages ("
			<< UT_LIST_GET_LEN(buf_pool->withdraw) << "/"
			<< buf_pool->withdraw_target << ").";

		if (++loop_count >= 10) {
			/* give up for now.
			retried after user threads paused. */

			ib::info() << "buffer pool " << i
				<< " : will retry to withdraw later.";

			/* need retry later */
			return(true);
		}
	}

	/* confirm withdrawn enough */
	const buf_chunk_t*	chunk
		= buf_pool->chunks + buf_pool->n_chunks_new;
	const buf_chunk_t*	echunk
		= buf_pool->chunks + buf_pool->n_chunks;

	while (chunk < echunk) {
		block = chunk->blocks;
		for (ulint j = chunk->size; j--; block++) {
			/* If !=BUF_BLOCK_NOT_USED block in the
			withdrawn area, it means corruption
			something */
			ut_a(buf_block_get_state(block)
				== BUF_BLOCK_NOT_USED);
			ut_ad(block->in_withdraw_list);
		}
		++chunk;
	}

	ib::info() << "buffer pool " << i << " : withdrawn target "
		<< UT_LIST_GET_LEN(buf_pool->withdraw) << " blocks.";

	/* retry is not needed */
	++buf_withdraw_clock;

	return(false);
}

/** resize page_hash and zip_hash for a buffer pool instance.
@param[in]	buf_pool	buffer pool instance */
static
void
buf_pool_resize_hash(
	buf_pool_t*	buf_pool)
{
	hash_table_t*	new_hash_table;

	ut_ad(buf_pool->page_hash_old == NULL);

	/* recreate page_hash */
	new_hash_table = ib_recreate(
		buf_pool->page_hash, 2 * buf_pool->curr_size);

	for (ulint i = 0; i < hash_get_n_cells(buf_pool->page_hash); i++) {
		buf_page_t*	bpage;

		bpage = static_cast<buf_page_t*>(
			HASH_GET_FIRST(
				buf_pool->page_hash, i));

		while (bpage) {
			buf_page_t*	prev_bpage = bpage;
			ulint		fold;

			bpage = static_cast<buf_page_t*>(
				HASH_GET_NEXT(
					hash, prev_bpage));

			fold = prev_bpage->id.fold();

			HASH_DELETE(buf_page_t, hash,
				buf_pool->page_hash, fold,
				prev_bpage);

			HASH_INSERT(buf_page_t, hash,
				new_hash_table, fold,
				prev_bpage);
		}
	}

	buf_pool->page_hash_old = buf_pool->page_hash;
	buf_pool->page_hash = new_hash_table;

	/* recreate zip_hash */
	new_hash_table = hash_create(2 * buf_pool->curr_size);

	for (ulint i = 0; i < hash_get_n_cells(buf_pool->zip_hash); i++) {
		buf_page_t*	bpage;

		bpage = static_cast<buf_page_t*>(
			HASH_GET_FIRST(buf_pool->zip_hash, i));

		while (bpage) {
			buf_page_t*	prev_bpage = bpage;
			ulint		fold;

			bpage = static_cast<buf_page_t*>(
				HASH_GET_NEXT(
					hash, prev_bpage));

			fold = BUF_POOL_ZIP_FOLD(
				reinterpret_cast<buf_block_t*>(
					prev_bpage));

			HASH_DELETE(buf_page_t, hash,
				buf_pool->zip_hash, fold,
				prev_bpage);

			HASH_INSERT(buf_page_t, hash,
				new_hash_table, fold,
				prev_bpage);
		}
	}

	hash_table_free(buf_pool->zip_hash);
	buf_pool->zip_hash = new_hash_table;
}

/** Resize the buffer pool based on srv_buf_pool_size from
srv_buf_pool_old_size. */
static
void
buf_pool_resize()
{
	buf_pool_t*	buf_pool;
	ulint		new_instance_size;
	bool		warning = false;

	NUMA_MEMPOLICY_INTERLEAVE_IN_SCOPE;

	ut_ad(!buf_pool_resizing);
	ut_ad(!buf_pool_withdrawing);
	ut_ad(srv_buf_pool_chunk_unit > 0);

	new_instance_size = srv_buf_pool_size / srv_buf_pool_instances;
	new_instance_size /= UNIV_PAGE_SIZE;

	buf_resize_status("Resizing buffer pool from " ULINTPF " to "
			  ULINTPF " (unit=" ULINTPF ").",
			  srv_buf_pool_old_size, srv_buf_pool_size,
			  srv_buf_pool_chunk_unit);

	/* set new limit for all buffer pool for resizing */
	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool = buf_pool_from_array(i);
		buf_pool_mutex_enter(buf_pool);

		ut_ad(buf_pool->curr_size == buf_pool->old_size);
		ut_ad(buf_pool->n_chunks_new == buf_pool->n_chunks);
		ut_ad(UT_LIST_GET_LEN(buf_pool->withdraw) == 0);
		ut_ad(buf_pool->flush_rbt == NULL);

		buf_pool->curr_size = new_instance_size;

		buf_pool->n_chunks_new = new_instance_size * UNIV_PAGE_SIZE
			/ srv_buf_pool_chunk_unit;

		buf_pool_mutex_exit(buf_pool);
	}
#ifdef BTR_CUR_HASH_ADAPT
	/* disable AHI if needed */
	bool	btr_search_disabled = false;

	buf_resize_status("Disabling adaptive hash index.");

	btr_search_s_lock_all();
	if (btr_search_enabled) {
		btr_search_s_unlock_all();
		btr_search_disabled = true;
	} else {
		btr_search_s_unlock_all();
	}

	btr_search_disable();

	if (btr_search_disabled) {
		ib::info() << "disabled adaptive hash index.";
	}
#endif /* BTR_CUR_HASH_ADAPT */

	/* set withdraw target */
	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool = buf_pool_from_array(i);
		if (buf_pool->curr_size < buf_pool->old_size) {
			ulint	withdraw_target = 0;

			const buf_chunk_t*	chunk
				= buf_pool->chunks + buf_pool->n_chunks_new;
			const buf_chunk_t*	echunk
				= buf_pool->chunks + buf_pool->n_chunks;

			while (chunk < echunk) {
				withdraw_target += chunk->size;
				++chunk;
			}

			ut_ad(buf_pool->withdraw_target == 0);
			buf_pool->withdraw_target = withdraw_target;
			buf_pool_withdrawing = true;
		}
	}

	buf_resize_status("Withdrawing blocks to be shrunken.");

	time_t		withdraw_started = time(NULL);
	ulint		message_interval = 60;
	ulint		retry_interval = 1;

withdraw_retry:
	bool	should_retry_withdraw = false;

	/* wait for the number of blocks fit to the new size (if needed)*/
	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool = buf_pool_from_array(i);
		if (buf_pool->curr_size < buf_pool->old_size) {

			should_retry_withdraw |=
				buf_pool_withdraw_blocks(buf_pool);
		}
	}

	if (srv_shutdown_state != SRV_SHUTDOWN_NONE) {
		/* abort to resize for shutdown. */
		buf_pool_withdrawing = false;
		return;
	}

	/* abort buffer pool load */
	buf_load_abort();

	const time_t current_time = time(NULL);

	if (should_retry_withdraw
	    && difftime(current_time, withdraw_started) >= message_interval) {

		if (message_interval > 900) {
			message_interval = 1800;
		} else {
			message_interval *= 2;
		}

		lock_mutex_enter();
		trx_sys_mutex_enter();
		bool	found = false;
		for (trx_t* trx = UT_LIST_GET_FIRST(trx_sys->mysql_trx_list);
		     trx != NULL;
		     trx = UT_LIST_GET_NEXT(mysql_trx_list, trx)) {
			if (trx->state != TRX_STATE_NOT_STARTED
			    && trx->mysql_thd != NULL
			    && withdraw_started > trx->start_time) {
				if (!found) {
					ib::warn() <<
						"The following trx might hold"
						" the blocks in buffer pool to"
					        " be withdrawn. Buffer pool"
						" resizing can complete only"
						" after all the transactions"
						" below release the blocks.";
					found = true;
				}

				lock_trx_print_wait_and_mvcc_state(
					stderr, trx, current_time);
			}
		}
		trx_sys_mutex_exit();
		lock_mutex_exit();

		withdraw_started = current_time;
	}

	if (should_retry_withdraw) {
		ib::info() << "Will retry to withdraw " << retry_interval
			<< " seconds later.";
		os_thread_sleep(retry_interval * 1000000);

		if (retry_interval > 5) {
			retry_interval = 10;
		} else {
			retry_interval *= 2;
		}

		goto withdraw_retry;
	}

	buf_pool_withdrawing = false;

	buf_resize_status("Latching whole of buffer pool.");

#ifndef DBUG_OFF
	{
		bool	should_wait = true;

		while (should_wait) {
			should_wait = false;
			DBUG_EXECUTE_IF(
				"ib_buf_pool_resize_wait_before_resize",
				should_wait = true; os_thread_sleep(10000););
		}
	}
#endif /* !DBUG_OFF */

	if (srv_shutdown_state != SRV_SHUTDOWN_NONE) {
		return;
	}

	/* Indicate critical path */
	buf_pool_resizing = true;

	/* Acquire all buf_pool_mutex/hash_lock */
	for (ulint i = 0; i < srv_buf_pool_instances; ++i) {
		buf_pool_t*	buf_pool = buf_pool_from_array(i);

		buf_pool_mutex_enter(buf_pool);
	}
	for (ulint i = 0; i < srv_buf_pool_instances; ++i) {
		buf_pool_t*	buf_pool = buf_pool_from_array(i);

		hash_lock_x_all(buf_pool->page_hash);
	}

	buf_chunk_map_reg = UT_NEW_NOKEY(buf_pool_chunk_map_t());

	/* add/delete chunks */
	for (ulint i = 0; i < srv_buf_pool_instances; ++i) {
		buf_pool_t*	buf_pool = buf_pool_from_array(i);
		buf_chunk_t*	chunk;
		buf_chunk_t*	echunk;

		buf_resize_status("buffer pool %lu :"
			" resizing with chunks %lu to %lu.",
			i, buf_pool->n_chunks, buf_pool->n_chunks_new);

		if (buf_pool->n_chunks_new < buf_pool->n_chunks) {
			/* delete chunks */
			chunk = buf_pool->chunks
				+ buf_pool->n_chunks_new;
			echunk = buf_pool->chunks + buf_pool->n_chunks;

			ulint	sum_freed = 0;

			while (chunk < echunk) {
				buf_block_t*	block = chunk->blocks;

				for (ulint j = chunk->size;
				     j--; block++) {
					mutex_free(&block->mutex);
					rw_lock_free(&block->lock);

					ut_d(rw_lock_free(
						&block->debug_latch));
				}

				buf_pool->allocator.deallocate_large(
					chunk->mem, &chunk->mem_pfx);

				sum_freed += chunk->size;

				++chunk;
			}

			/* discard withdraw list */
			UT_LIST_INIT(buf_pool->withdraw,
				     &buf_page_t::list);
			buf_pool->withdraw_target = 0;

			ib::info() << "buffer pool " << i << " : "
				<< buf_pool->n_chunks - buf_pool->n_chunks_new
				<< " chunks (" << sum_freed
				<< " blocks) were freed.";

			buf_pool->n_chunks = buf_pool->n_chunks_new;
		}

		{
			/* reallocate buf_pool->chunks */
			const ulint	new_chunks_size
				= buf_pool->n_chunks_new * sizeof(*chunk);

			buf_chunk_t*	new_chunks
				= reinterpret_cast<buf_chunk_t*>(
					ut_zalloc_nokey_nofatal(new_chunks_size));

			DBUG_EXECUTE_IF("buf_pool_resize_chunk_null",
					ut_free(new_chunks);
					new_chunks = NULL;);

			if (new_chunks == NULL) {
				ib::error() << "buffer pool " << i
					<< " : failed to allocate"
					" the chunk array.";
				buf_pool->n_chunks_new
					= buf_pool->n_chunks;
				warning = true;
				buf_pool->chunks_old = NULL;
				for (ulint j = 0; j < buf_pool->n_chunks_new; j++) {
					buf_pool_register_chunk(&(buf_pool->chunks[j]));
				}
				goto calc_buf_pool_size;
			}

			ulint	n_chunks_copy = ut_min(buf_pool->n_chunks_new,
						       buf_pool->n_chunks);

			memcpy(new_chunks, buf_pool->chunks,
			       n_chunks_copy * sizeof(*chunk));

			for (ulint j = 0; j < n_chunks_copy; j++) {
				buf_pool_register_chunk(&new_chunks[j]);
			}

			buf_pool->chunks_old = buf_pool->chunks;
			buf_pool->chunks = new_chunks;
		}


		if (buf_pool->n_chunks_new > buf_pool->n_chunks) {
			/* add chunks */
			chunk = buf_pool->chunks + buf_pool->n_chunks;
			echunk = buf_pool->chunks
				+ buf_pool->n_chunks_new;

			ulint	sum_added = 0;
			ulint	n_chunks = buf_pool->n_chunks;

			while (chunk < echunk) {
				ulong	unit = srv_buf_pool_chunk_unit;

				if (!buf_chunk_init(buf_pool, chunk, unit)) {

					ib::error() << "buffer pool " << i
						<< " : failed to allocate"
						" new memory.";

					warning = true;

					buf_pool->n_chunks_new
						= n_chunks;

					break;
				}

				sum_added += chunk->size;

				++n_chunks;
				++chunk;
			}

			ib::info() << "buffer pool " << i << " : "
				<< buf_pool->n_chunks_new - buf_pool->n_chunks
				<< " chunks (" << sum_added
				<< " blocks) were added.";

			buf_pool->n_chunks = n_chunks;
		}
calc_buf_pool_size:

		/* recalc buf_pool->curr_size */
		ulint	new_size = 0;

		chunk = buf_pool->chunks;
		do {
			new_size += chunk->size;
		} while (++chunk < buf_pool->chunks
				   + buf_pool->n_chunks);

		buf_pool->curr_size = new_size;
		buf_pool->n_chunks_new = buf_pool->n_chunks;

		if (buf_pool->chunks_old) {
			ut_free(buf_pool->chunks_old);
			buf_pool->chunks_old = NULL;
		}
	}

	buf_pool_chunk_map_t*	chunk_map_old = buf_chunk_map_ref;
	buf_chunk_map_ref = buf_chunk_map_reg;

	/* set instance sizes */
	{
		ulint	curr_size = 0;

		for (ulint i = 0; i < srv_buf_pool_instances; i++) {
			buf_pool = buf_pool_from_array(i);

			ut_ad(UT_LIST_GET_LEN(buf_pool->withdraw) == 0);

			buf_pool->read_ahead_area =
				ut_min(BUF_READ_AHEAD_PAGES,
				       ut_2_power_up(buf_pool->curr_size /
						      BUF_READ_AHEAD_PORTION));
			buf_pool->curr_pool_size
				= buf_pool->curr_size * UNIV_PAGE_SIZE;
			curr_size += buf_pool->curr_pool_size;
			buf_pool->old_size = buf_pool->curr_size;
		}
		srv_buf_pool_curr_size = curr_size;
		innodb_set_buf_pool_size(buf_pool_size_align(curr_size));
	}

	const bool	new_size_too_diff
		= srv_buf_pool_base_size > srv_buf_pool_size * 2
			|| srv_buf_pool_base_size * 2 < srv_buf_pool_size;

	/* Normalize page_hash and zip_hash,
	if the new size is too different */
	if (!warning && new_size_too_diff) {

		buf_resize_status("Resizing hash tables.");

		for (ulint i = 0; i < srv_buf_pool_instances; ++i) {
			buf_pool_t*	buf_pool = buf_pool_from_array(i);

			buf_pool_resize_hash(buf_pool);

			ib::info() << "buffer pool " << i
				<< " : hash tables were resized.";
		}
	}

	/* Release all buf_pool_mutex/page_hash */
	for (ulint i = 0; i < srv_buf_pool_instances; ++i) {
		buf_pool_t*	buf_pool = buf_pool_from_array(i);

		hash_unlock_x_all(buf_pool->page_hash);
		buf_pool_mutex_exit(buf_pool);

		if (buf_pool->page_hash_old != NULL) {
			hash_table_free(buf_pool->page_hash_old);
			buf_pool->page_hash_old = NULL;
		}
	}

	UT_DELETE(chunk_map_old);

	buf_pool_resizing = false;

	/* Normalize other components, if the new size is too different */
	if (!warning && new_size_too_diff) {
		srv_buf_pool_base_size = srv_buf_pool_size;

		buf_resize_status("Resizing also other hash tables.");

		/* normalize lock_sys */
		srv_lock_table_size = 5 * (srv_buf_pool_size / UNIV_PAGE_SIZE);
		lock_sys_resize(srv_lock_table_size);

		/* normalize dict_sys */
		dict_resize();

		ib::info() << "Resized hash tables at lock_sys,"
#ifdef BTR_CUR_HASH_ADAPT
			" adaptive hash index,"
#endif /* BTR_CUR_HASH_ADAPT */
			" dictionary.";
	}

	/* normalize ibuf->max_size */
	ibuf_max_size_update(srv_change_buffer_max_size);

	if (srv_buf_pool_old_size != srv_buf_pool_size) {

		ib::info() << "Completed to resize buffer pool from "
			<< srv_buf_pool_old_size
			<< " to " << srv_buf_pool_size << ".";
		srv_buf_pool_old_size = srv_buf_pool_size;
	}

#ifdef BTR_CUR_HASH_ADAPT
	/* enable AHI if needed */
	if (btr_search_disabled) {
		btr_search_enable(true);
		ib::info() << "Re-enabled adaptive hash index.";
	}
#endif /* BTR_CUR_HASH_ADAPT */

	char	now[32];

	ut_sprintf_timestamp(now);
	if (!warning) {
		buf_resize_status("Completed resizing buffer pool at %s.",
			now);
	} else {
		buf_resize_status("Resizing buffer pool failed,"
			" finished resizing at %s.", now);
	}

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(buf_validate());
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

	return;
}

/** This is the thread for resizing buffer pool. It waits for an event and
when waked up either performs a resizing and sleeps again.
@return	this function does not return, calls os_thread_exit()
*/
extern "C"
os_thread_ret_t
DECLARE_THREAD(buf_resize_thread)(void*)
{
	my_thread_init();

	while (srv_shutdown_state == SRV_SHUTDOWN_NONE) {
		os_event_wait(srv_buf_resize_event);
		os_event_reset(srv_buf_resize_event);

		if (srv_shutdown_state != SRV_SHUTDOWN_NONE) {
			break;
		}

		buf_pool_mutex_enter_all();
		if (srv_buf_pool_old_size == srv_buf_pool_size) {
			buf_pool_mutex_exit_all();
			std::ostringstream sout;
			sout << "Size did not change (old size = new size = "
				<< srv_buf_pool_size << ". Nothing to do.";
			buf_resize_status(sout.str().c_str());

			/* nothing to do */
			continue;
		}
		buf_pool_mutex_exit_all();

		buf_pool_resize();
	}

	srv_buf_resize_thread_active = false;

	my_thread_end();
	os_thread_exit();

	OS_THREAD_DUMMY_RETURN;
}

/********************************************************************//**
Relocate a buffer control block.  Relocates the block on the LRU list
and in buf_pool->page_hash.  Does not relocate bpage->list.
The caller must take care of relocating bpage->list. */
static
void
buf_relocate(
/*=========*/
	buf_page_t*	bpage,	/*!< in/out: control block being relocated;
				buf_page_get_state(bpage) must be
				BUF_BLOCK_ZIP_DIRTY or BUF_BLOCK_ZIP_PAGE */
	buf_page_t*	dpage)	/*!< in/out: destination control block */
{
	buf_page_t*	b;
	buf_pool_t*	buf_pool = buf_pool_from_bpage(bpage);

	ut_ad(buf_pool_mutex_own(buf_pool));
	ut_ad(buf_page_hash_lock_held_x(buf_pool, bpage));
	ut_ad(mutex_own(buf_page_get_mutex(bpage)));
	ut_a(buf_page_get_io_fix(bpage) == BUF_IO_NONE);
	ut_a(bpage->buf_fix_count == 0);
	ut_ad(bpage->in_LRU_list);
	ut_ad(!bpage->in_zip_hash);
	ut_ad(bpage->in_page_hash);
	ut_ad(bpage == buf_page_hash_get_low(buf_pool, bpage->id));

	ut_ad(!buf_pool_watch_is_sentinel(buf_pool, bpage));
#ifdef UNIV_DEBUG
	switch (buf_page_get_state(bpage)) {
	case BUF_BLOCK_POOL_WATCH:
	case BUF_BLOCK_NOT_USED:
	case BUF_BLOCK_READY_FOR_USE:
	case BUF_BLOCK_FILE_PAGE:
	case BUF_BLOCK_MEMORY:
	case BUF_BLOCK_REMOVE_HASH:
		ut_error;
	case BUF_BLOCK_ZIP_DIRTY:
	case BUF_BLOCK_ZIP_PAGE:
		break;
	}
#endif /* UNIV_DEBUG */

	new (dpage) buf_page_t(*bpage);

	/* Important that we adjust the hazard pointer before
	removing bpage from LRU list. */
	buf_LRU_adjust_hp(buf_pool, bpage);

	ut_d(bpage->in_LRU_list = FALSE);
	ut_d(bpage->in_page_hash = FALSE);

	/* relocate buf_pool->LRU */
	b = UT_LIST_GET_PREV(LRU, bpage);
	UT_LIST_REMOVE(buf_pool->LRU, bpage);

	if (b != NULL) {
		UT_LIST_INSERT_AFTER(buf_pool->LRU, b, dpage);
	} else {
		UT_LIST_ADD_FIRST(buf_pool->LRU, dpage);
	}

	if (UNIV_UNLIKELY(buf_pool->LRU_old == bpage)) {
		buf_pool->LRU_old = dpage;
#ifdef UNIV_LRU_DEBUG
		/* buf_pool->LRU_old must be the first item in the LRU list
		whose "old" flag is set. */
		ut_a(buf_pool->LRU_old->old);
		ut_a(!UT_LIST_GET_PREV(LRU, buf_pool->LRU_old)
		     || !UT_LIST_GET_PREV(LRU, buf_pool->LRU_old)->old);
		ut_a(!UT_LIST_GET_NEXT(LRU, buf_pool->LRU_old)
		     || UT_LIST_GET_NEXT(LRU, buf_pool->LRU_old)->old);
	} else {
		/* Check that the "old" flag is consistent in
		the block and its neighbours. */
		buf_page_set_old(dpage, buf_page_is_old(dpage));
#endif /* UNIV_LRU_DEBUG */
	}

        ut_d(CheckInLRUList::validate(buf_pool));

	/* relocate buf_pool->page_hash */
	ulint	fold = bpage->id.fold();
	ut_ad(fold == dpage->id.fold());
	HASH_REPLACE(buf_page_t, hash, buf_pool->page_hash, fold, bpage,
		     dpage);
}

/** Hazard Pointer implementation. */

/** Set current value
@param bpage	buffer block to be set as hp */
void
HazardPointer::set(buf_page_t* bpage)
{
	ut_ad(mutex_own(m_mutex));
	ut_ad(!bpage || buf_pool_from_bpage(bpage) == m_buf_pool);
	ut_ad(!bpage || buf_page_in_file(bpage));

	m_hp = bpage;
}

/** Checks if a bpage is the hp
@param bpage    buffer block to be compared
@return true if it is hp */

bool
HazardPointer::is_hp(const buf_page_t* bpage)
{
	ut_ad(mutex_own(m_mutex));
	ut_ad(!m_hp || buf_pool_from_bpage(m_hp) == m_buf_pool);
	ut_ad(!bpage || buf_pool_from_bpage(bpage) == m_buf_pool);

	return(bpage == m_hp);
}

/** Adjust the value of hp. This happens when some other thread working
on the same list attempts to remove the hp from the list.
@param bpage	buffer block to be compared */

void
FlushHp::adjust(const buf_page_t* bpage)
{
	ut_ad(bpage != NULL);

	/** We only support reverse traversal for now. */
	if (is_hp(bpage)) {
		m_hp = UT_LIST_GET_PREV(list, m_hp);
	}

	ut_ad(!m_hp || m_hp->in_flush_list);
}

/** Adjust the value of hp. This happens when some other thread working
on the same list attempts to remove the hp from the list.
@param bpage	buffer block to be compared */

void
LRUHp::adjust(const buf_page_t* bpage)
{
	ut_ad(bpage);

	/** We only support reverse traversal for now. */
	if (is_hp(bpage)) {
		m_hp = UT_LIST_GET_PREV(LRU, m_hp);
	}

	ut_ad(!m_hp || m_hp->in_LRU_list);
}

/** Selects from where to start a scan. If we have scanned too deep into
the LRU list it resets the value to the tail of the LRU list.
@return buf_page_t from where to start scan. */

buf_page_t*
LRUItr::start()
{
	ut_ad(mutex_own(m_mutex));

	if (!m_hp || m_hp->old) {
		m_hp = UT_LIST_GET_LAST(m_buf_pool->LRU);
	}

	return(m_hp);
}

/** Determine if a block is a sentinel for a buffer pool watch.
@param[in]	buf_pool	buffer pool instance
@param[in]	bpage		block
@return TRUE if a sentinel for a buffer pool watch, FALSE if not */
ibool
buf_pool_watch_is_sentinel(
	const buf_pool_t*	buf_pool,
	const buf_page_t*	bpage)
{
	/* We must also own the appropriate hash lock. */
	ut_ad(buf_page_hash_lock_held_s_or_x(buf_pool, bpage));
	ut_ad(buf_page_in_file(bpage));

	if (bpage < &buf_pool->watch[0]
	    || bpage >= &buf_pool->watch[BUF_POOL_WATCH_SIZE]) {

		ut_ad(buf_page_get_state(bpage) != BUF_BLOCK_ZIP_PAGE
		      || bpage->zip.data != NULL);

		return(FALSE);
	}

	ut_ad(buf_page_get_state(bpage) == BUF_BLOCK_ZIP_PAGE);
	ut_ad(!bpage->in_zip_hash);
	ut_ad(bpage->in_page_hash);
	ut_ad(bpage->zip.data == NULL);
	return(TRUE);
}

/** Add watch for the given page to be read in. Caller must have
appropriate hash_lock for the bpage. This function may release the
hash_lock and reacquire it.
@param[in]	page_id		page id
@param[in,out]	hash_lock	hash_lock currently latched
@return NULL if watch set, block if the page is in the buffer pool */
static
buf_page_t*
buf_pool_watch_set(
	const page_id_t		page_id,
	rw_lock_t**		hash_lock)
{
	buf_page_t*	bpage;
	ulint		i;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);

	ut_ad(*hash_lock == buf_page_hash_lock_get(buf_pool, page_id));

	ut_ad(rw_lock_own(*hash_lock, RW_LOCK_X));

	bpage = buf_page_hash_get_low(buf_pool, page_id);

	if (bpage != NULL) {
page_found:
		if (!buf_pool_watch_is_sentinel(buf_pool, bpage)) {
			/* The page was loaded meanwhile. */
			return(bpage);
		}

		/* Add to an existing watch. */
		buf_block_fix(bpage);
		return(NULL);
	}

	/* From this point this function becomes fairly heavy in terms
	of latching. We acquire the buf_pool mutex as well as all the
	hash_locks. buf_pool mutex is needed because any changes to
	the page_hash must be covered by it and hash_locks are needed
	because we don't want to read any stale information in
	buf_pool->watch[]. However, it is not in the critical code path
	as this function will be called only by the purge thread. */

	/* To obey latching order first release the hash_lock. */
	rw_lock_x_unlock(*hash_lock);

	buf_pool_mutex_enter(buf_pool);
	hash_lock_x_all(buf_pool->page_hash);

	/* If not own buf_pool_mutex, page_hash can be changed. */
	*hash_lock = buf_page_hash_lock_get(buf_pool, page_id);

	/* We have to recheck that the page
	was not loaded or a watch set by some other
	purge thread. This is because of the small
	time window between when we release the
	hash_lock to acquire buf_pool mutex above. */

	bpage = buf_page_hash_get_low(buf_pool, page_id);
	if (UNIV_LIKELY_NULL(bpage)) {
		buf_pool_mutex_exit(buf_pool);
		hash_unlock_x_all_but(buf_pool->page_hash, *hash_lock);
		goto page_found;
	}

	/* The maximum number of purge threads should never exceed
	BUF_POOL_WATCH_SIZE. So there is no way for purge thread
	instance to hold a watch when setting another watch. */
	for (i = 0; i < BUF_POOL_WATCH_SIZE; i++) {
		bpage = &buf_pool->watch[i];

		ut_ad(bpage->access_time == 0);
		ut_ad(bpage->newest_modification == 0);
		ut_ad(bpage->oldest_modification == 0);
		ut_ad(bpage->zip.data == NULL);
		ut_ad(!bpage->in_zip_hash);

		switch (bpage->state) {
		case BUF_BLOCK_POOL_WATCH:
			ut_ad(!bpage->in_page_hash);
			ut_ad(bpage->buf_fix_count == 0);

			/* bpage is pointing to buf_pool->watch[],
			which is protected by buf_pool->mutex.
			Normally, buf_page_t objects are protected by
			buf_block_t::mutex or buf_pool->zip_mutex or both. */

			bpage->state = BUF_BLOCK_ZIP_PAGE;
			bpage->id = page_id;
			bpage->buf_fix_count = 1;

			ut_d(bpage->in_page_hash = TRUE);
			HASH_INSERT(buf_page_t, hash, buf_pool->page_hash,
				    page_id.fold(), bpage);

			buf_pool_mutex_exit(buf_pool);
			/* Once the sentinel is in the page_hash we can
			safely release all locks except just the
			relevant hash_lock */
			hash_unlock_x_all_but(buf_pool->page_hash,
						*hash_lock);

			return(NULL);
		case BUF_BLOCK_ZIP_PAGE:
			ut_ad(bpage->in_page_hash);
			ut_ad(bpage->buf_fix_count > 0);
			break;
		default:
			ut_error;
		}
	}

	/* Allocation failed.  Either the maximum number of purge
	threads should never exceed BUF_POOL_WATCH_SIZE, or this code
	should be modified to return a special non-NULL value and the
	caller should purge the record directly. */
	ut_error;

	/* Fix compiler warning */
	return(NULL);
}

/** Remove the sentinel block for the watch before replacing it with a
real block. buf_page_watch_clear() or buf_page_watch_occurred() will notice
that the block has been replaced with the real block.
@param[in,out]	buf_pool	buffer pool instance
@param[in,out]	watch		sentinel for watch
@return reference count, to be added to the replacement block */
static
void
buf_pool_watch_remove(
	buf_pool_t*	buf_pool,
	buf_page_t*	watch)
{
#ifdef UNIV_DEBUG
	/* We must also own the appropriate hash_bucket mutex. */
	rw_lock_t* hash_lock = buf_page_hash_lock_get(buf_pool, watch->id);
	ut_ad(rw_lock_own(hash_lock, RW_LOCK_X));
#endif /* UNIV_DEBUG */

	ut_ad(buf_pool_mutex_own(buf_pool));

	HASH_DELETE(buf_page_t, hash, buf_pool->page_hash, watch->id.fold(),
		    watch);
	ut_d(watch->in_page_hash = FALSE);
	watch->buf_fix_count = 0;
	watch->state = BUF_BLOCK_POOL_WATCH;
}

/** Stop watching if the page has been read in.
buf_pool_watch_set(same_page_id) must have returned NULL before.
@param[in]	page_id	page id */
void buf_pool_watch_unset(const page_id_t page_id)
{
	buf_page_t*	bpage;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);

	/* We only need to have buf_pool mutex in case where we end
	up calling buf_pool_watch_remove but to obey latching order
	we acquire it here before acquiring hash_lock. This should
	not cause too much grief as this function is only ever
	called from the purge thread. */
	buf_pool_mutex_enter(buf_pool);

	rw_lock_t*	hash_lock = buf_page_hash_lock_get(buf_pool, page_id);
	rw_lock_x_lock(hash_lock);

	/* The page must exist because buf_pool_watch_set()
	increments buf_fix_count. */
	bpage = buf_page_hash_get_low(buf_pool, page_id);

	if (buf_block_unfix(bpage) == 0
	    && buf_pool_watch_is_sentinel(buf_pool, bpage)) {
		buf_pool_watch_remove(buf_pool, bpage);
	}

	buf_pool_mutex_exit(buf_pool);
	rw_lock_x_unlock(hash_lock);
}

/** Check if the page has been read in.
This may only be called after buf_pool_watch_set(same_page_id)
has returned NULL and before invoking buf_pool_watch_unset(same_page_id).
@param[in]	page_id	page id
@return false if the given page was not read in, true if it was */
bool buf_pool_watch_occurred(const page_id_t page_id)
{
	bool		ret;
	buf_page_t*	bpage;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);
	rw_lock_t*	hash_lock = buf_page_hash_lock_get(buf_pool, page_id);

	rw_lock_s_lock(hash_lock);

	/* If not own buf_pool_mutex, page_hash can be changed. */
	hash_lock = buf_page_hash_lock_s_confirm(hash_lock, buf_pool, page_id);

	/* The page must exist because buf_pool_watch_set()
	increments buf_fix_count. */
	bpage = buf_page_hash_get_low(buf_pool, page_id);

	ret = !buf_pool_watch_is_sentinel(buf_pool, bpage);
	rw_lock_s_unlock(hash_lock);

	return(ret);
}

/********************************************************************//**
Moves a page to the start of the buffer pool LRU list. This high-level
function can be used to prevent an important page from slipping out of
the buffer pool. */
void
buf_page_make_young(
/*================*/
	buf_page_t*	bpage)	/*!< in: buffer block of a file page */
{
	buf_pool_t*	buf_pool = buf_pool_from_bpage(bpage);

	buf_pool_mutex_enter(buf_pool);

	ut_a(buf_page_in_file(bpage));

	buf_LRU_make_block_young(bpage);

	buf_pool_mutex_exit(buf_pool);
}

/********************************************************************//**
Moves a page to the start of the buffer pool LRU list if it is too old.
This high-level function can be used to prevent an important page from
slipping out of the buffer pool. */
static
void
buf_page_make_young_if_needed(
/*==========================*/
	buf_page_t*	bpage)		/*!< in/out: buffer block of a
					file page */
{
#ifdef UNIV_DEBUG
	buf_pool_t*	buf_pool = buf_pool_from_bpage(bpage);
	ut_ad(!buf_pool_mutex_own(buf_pool));
#endif /* UNIV_DEBUG */
	ut_a(buf_page_in_file(bpage));

	if (buf_page_peek_if_too_old(bpage)) {
		buf_page_make_young(bpage);
	}
}

#ifdef UNIV_DEBUG

/** Sets file_page_was_freed TRUE if the page is found in the buffer pool.
This function should be called when we free a file page and want the
debug version to check that it is not accessed any more unless
reallocated.
@param[in]	page_id	page id
@return control block if found in page hash table, otherwise NULL */
buf_page_t* buf_page_set_file_page_was_freed(const page_id_t page_id)
{
	buf_page_t*	bpage;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);
	rw_lock_t*	hash_lock;

	bpage = buf_page_hash_get_s_locked(buf_pool, page_id, &hash_lock);

	if (bpage) {
		BPageMutex*	block_mutex = buf_page_get_mutex(bpage);
		ut_ad(!buf_pool_watch_is_sentinel(buf_pool, bpage));
		mutex_enter(block_mutex);
		rw_lock_s_unlock(hash_lock);
		/* bpage->file_page_was_freed can already hold
		when this code is invoked from dict_drop_index_tree() */
		bpage->file_page_was_freed = TRUE;
		mutex_exit(block_mutex);
	}

	return(bpage);
}

/** Sets file_page_was_freed FALSE if the page is found in the buffer pool.
This function should be called when we free a file page and want the
debug version to check that it is not accessed any more unless
reallocated.
@param[in]	page_id	page id
@return control block if found in page hash table, otherwise NULL */
buf_page_t* buf_page_reset_file_page_was_freed(const page_id_t page_id)
{
	buf_page_t*	bpage;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);
	rw_lock_t*	hash_lock;

	bpage = buf_page_hash_get_s_locked(buf_pool, page_id, &hash_lock);
	if (bpage) {
		BPageMutex*	block_mutex = buf_page_get_mutex(bpage);
		ut_ad(!buf_pool_watch_is_sentinel(buf_pool, bpage));
		mutex_enter(block_mutex);
		rw_lock_s_unlock(hash_lock);
		bpage->file_page_was_freed = FALSE;
		mutex_exit(block_mutex);
	}

	return(bpage);
}
#endif /* UNIV_DEBUG */

/** Attempts to discard the uncompressed frame of a compressed page.
The caller should not be holding any mutexes when this function is called.
@param[in]	page_id	page id */
static void buf_block_try_discard_uncompressed(const page_id_t page_id)
{
	buf_page_t*	bpage;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);

	/* Since we need to acquire buf_pool mutex to discard
	the uncompressed frame and because page_hash mutex resides
	below buf_pool mutex in sync ordering therefore we must
	first release the page_hash mutex. This means that the
	block in question can move out of page_hash. Therefore
	we need to check again if the block is still in page_hash. */
	buf_pool_mutex_enter(buf_pool);

	bpage = buf_page_hash_get(buf_pool, page_id);

	if (bpage) {
		buf_LRU_free_page(bpage, false);
	}

	buf_pool_mutex_exit(buf_pool);
}

/** Get read access to a compressed page (usually of type
FIL_PAGE_TYPE_ZBLOB or FIL_PAGE_TYPE_ZBLOB2).
The page must be released with buf_page_release_zip().
NOTE: the page is not protected by any latch.  Mutual exclusion has to
be implemented at a higher level.  In other words, all possible
accesses to a given page through this function must be protected by
the same set of mutexes or latches.
@param[in]	page_id		page id
@param[in]	page_size	page size
@return pointer to the block */
buf_page_t*
buf_page_get_zip(
	const page_id_t		page_id,
	const page_size_t&	page_size)
{
	buf_page_t*	bpage;
	BPageMutex*	block_mutex;
	rw_lock_t*	hash_lock;
	ibool		discard_attempted = FALSE;
	ibool		must_read;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);

	buf_pool->stat.n_page_gets++;

	for (;;) {
lookup:

		/* The following call will also grab the page_hash
		mutex if the page is found. */
		bpage = buf_page_hash_get_s_locked(buf_pool, page_id,
						   &hash_lock);
		if (bpage) {
			ut_ad(!buf_pool_watch_is_sentinel(buf_pool, bpage));
			break;
		}

		/* Page not in buf_pool: needs to be read from file */

		ut_ad(!hash_lock);
		dberr_t err = buf_read_page(page_id, page_size);

		if (UNIV_UNLIKELY(err != DB_SUCCESS)) {
			ib::error() << "Reading compressed page " << page_id
				<< " failed with error: " << err;

			goto err_exit;
		}

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
		ut_a(++buf_dbg_counter % 5771 || buf_validate());
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */
	}

	ut_ad(buf_page_hash_lock_held_s(buf_pool, bpage));

	if (!bpage->zip.data) {
		/* There is no compressed page. */
err_exit:
		rw_lock_s_unlock(hash_lock);
		return(NULL);
	}

	ut_ad(!buf_pool_watch_is_sentinel(buf_pool, bpage));

	switch (buf_page_get_state(bpage)) {
	case BUF_BLOCK_ZIP_PAGE:
	case BUF_BLOCK_ZIP_DIRTY:
		buf_block_fix(bpage);
		block_mutex = &buf_pool->zip_mutex;
		goto got_block;
	case BUF_BLOCK_FILE_PAGE:
		/* Discard the uncompressed page frame if possible. */
		if (!discard_attempted) {
			rw_lock_s_unlock(hash_lock);
			buf_block_try_discard_uncompressed(page_id);
			discard_attempted = TRUE;
			goto lookup;
		}

		buf_block_buf_fix_inc((buf_block_t*) bpage,
				      __FILE__, __LINE__);

		block_mutex = &((buf_block_t*) bpage)->mutex;
		goto got_block;
	default:
		break;
	}

	ut_error;
	goto err_exit;

got_block:
	mutex_enter(block_mutex);
	must_read = buf_page_get_io_fix(bpage) == BUF_IO_READ;

	rw_lock_s_unlock(hash_lock);

	ut_ad(!bpage->file_page_was_freed);

	buf_page_set_accessed(bpage);

	mutex_exit(block_mutex);

	buf_page_make_young_if_needed(bpage);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(++buf_dbg_counter % 5771 || buf_validate());
	ut_a(bpage->buf_fix_count > 0);
	ut_a(buf_page_in_file(bpage));
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

	if (must_read) {
		/* Let us wait until the read operation
		completes */

		for (;;) {
			enum buf_io_fix	io_fix;

			mutex_enter(block_mutex);
			io_fix = buf_page_get_io_fix(bpage);
			mutex_exit(block_mutex);

			if (io_fix == BUF_IO_READ) {

				os_thread_sleep(WAIT_FOR_READ);
			} else {
				break;
			}
		}
	}

	return(bpage);
}

/********************************************************************//**
Initialize some fields of a control block. */
UNIV_INLINE
void
buf_block_init_low(
/*===============*/
	buf_block_t*	block)	/*!< in: block to init */
{
	block->skip_flush_check = false;
#ifdef BTR_CUR_HASH_ADAPT
	/* No adaptive hash index entries may point to a previously
	unused (and now freshly allocated) block. */
	assert_block_ahi_empty_on_init(block);
	block->index		= NULL;

	block->n_hash_helps	= 0;
	block->n_fields		= 1;
	block->n_bytes		= 0;
	block->left_side	= TRUE;
#endif /* BTR_CUR_HASH_ADAPT */
}

/********************************************************************//**
Decompress a block.
@return TRUE if successful */
ibool
buf_zip_decompress(
/*===============*/
	buf_block_t*	block,	/*!< in/out: block */
	ibool		check)	/*!< in: TRUE=verify the page checksum */
{
	const byte*	frame = block->page.zip.data;
	ulint		size = page_zip_get_size(&block->page.zip);
	/* The tablespace will not be found if this function is called
	during IMPORT. */
	fil_space_t* space = fil_space_acquire_for_io(block->page.id.space());
	const unsigned key_version = mach_read_from_4(
		frame + FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION);
	fil_space_crypt_t* crypt_data = space ? space->crypt_data : NULL;
	const bool encrypted = crypt_data
		&& crypt_data->type != CRYPT_SCHEME_UNENCRYPTED
		&& (!crypt_data->is_default_encryption()
		    || srv_encrypt_tables);

	ut_ad(block->page.size.is_compressed());
	ut_a(block->page.id.space() != 0);

	if (UNIV_UNLIKELY(check && !page_zip_verify_checksum(frame, size))) {

		ib::error() << "Compressed page checksum mismatch for "
			<< (space ? space->chain.start->name : "")
			<< block->page.id << ": stored: "
			<< mach_read_from_4(frame + FIL_PAGE_SPACE_OR_CHKSUM)
			<< ", crc32: "
			<< page_zip_calc_checksum(
				frame, size, SRV_CHECKSUM_ALGORITHM_CRC32)
#ifdef INNODB_BUG_ENDIAN_CRC32
			<< "/"
			<< page_zip_calc_checksum(
				frame, size, SRV_CHECKSUM_ALGORITHM_CRC32,
				true)
#endif
			<< " innodb: "
			<< page_zip_calc_checksum(
				frame, size, SRV_CHECKSUM_ALGORITHM_INNODB)
			<< ", none: "
			<< page_zip_calc_checksum(
				frame, size, SRV_CHECKSUM_ALGORITHM_NONE)
			<< " (algorithm: " << srv_checksum_algorithm << ")";

		goto err_exit;
	}

	switch (fil_page_get_type(frame)) {
	case FIL_PAGE_INDEX:
	case FIL_PAGE_RTREE:
		if (page_zip_decompress(&block->page.zip,
					block->frame, TRUE)) {
			if (space) {
				fil_space_release_for_io(space);
			}
			return(TRUE);
		}

		ib::error() << "Unable to decompress "
			<< (space ? space->chain.start->name : "")
			<< block->page.id;
		goto err_exit;
	case FIL_PAGE_TYPE_ALLOCATED:
	case FIL_PAGE_INODE:
	case FIL_PAGE_IBUF_BITMAP:
	case FIL_PAGE_TYPE_FSP_HDR:
	case FIL_PAGE_TYPE_XDES:
	case FIL_PAGE_TYPE_ZBLOB:
	case FIL_PAGE_TYPE_ZBLOB2:
		/* Copy to uncompressed storage. */
		memcpy(block->frame, frame, block->page.size.physical());
		if (space) {
			fil_space_release_for_io(space);
		}

		return(TRUE);
	}

	ib::error() << "Unknown compressed page type "
		<< fil_page_get_type(frame)
		<< " in " << (space ? space->chain.start->name : "")
		<< block->page.id;

err_exit:
	if (encrypted) {
		ib::info() << "Row compressed page could be encrypted"
			" with key_version " << key_version;
		dict_set_encrypted_by_space(block->page.id.space());
	} else {
		dict_set_corrupted_by_space(block->page.id.space());
	}

	if (space) {
		fil_space_release_for_io(space);
	}

	return(FALSE);
}

#ifdef BTR_CUR_HASH_ADAPT
/** Get a buffer block from an adaptive hash index pointer.
This function does not return if the block is not identified.
@param[in]	ptr	pointer to within a page frame
@return pointer to block, never NULL */
buf_block_t*
buf_block_from_ahi(const byte* ptr)
{
	buf_pool_chunk_map_t::iterator it;

	buf_pool_chunk_map_t*	chunk_map = buf_chunk_map_ref;
	ut_ad(buf_chunk_map_ref == buf_chunk_map_reg);
	ut_ad(!buf_pool_resizing);

	buf_chunk_t*	chunk;
	it = chunk_map->upper_bound(ptr);

	ut_a(it != chunk_map->begin());

	if (it == chunk_map->end()) {
		chunk = chunk_map->rbegin()->second;
	} else {
		chunk = (--it)->second;
	}

	ulint		offs = ptr - chunk->blocks->frame;

	offs >>= UNIV_PAGE_SIZE_SHIFT;

	ut_a(offs < chunk->size);

	buf_block_t*	block = &chunk->blocks[offs];

	/* The function buf_chunk_init() invokes buf_block_init() so that
	block[n].frame == block->frame + n * UNIV_PAGE_SIZE.  Check it. */
	ut_ad(block->frame == page_align(ptr));
	/* Read the state of the block without holding a mutex.
	A state transition from BUF_BLOCK_FILE_PAGE to
	BUF_BLOCK_REMOVE_HASH is possible during this execution. */
	ut_d(const buf_page_state state = buf_block_get_state(block));
	ut_ad(state == BUF_BLOCK_FILE_PAGE || state == BUF_BLOCK_REMOVE_HASH);
	return(block);
}
#endif /* BTR_CUR_HASH_ADAPT */

/********************************************************************//**
Find out if a pointer belongs to a buf_block_t. It can be a pointer to
the buf_block_t itself or a member of it. This functions checks one of
the buffer pool instances.
@return TRUE if ptr belongs to a buf_block_t struct */
static
ibool
buf_pointer_is_block_field_instance(
/*================================*/
	buf_pool_t*	buf_pool,	/*!< in: buffer pool instance */
	const void*	ptr)		/*!< in: pointer not dereferenced */
{
	const buf_chunk_t*		chunk	= buf_pool->chunks;
	const buf_chunk_t* const	echunk	= chunk + ut_min(
		buf_pool->n_chunks, buf_pool->n_chunks_new);

	/* TODO: protect buf_pool->chunks with a mutex (the older pointer will
	currently remain while during buf_pool_resize()) */
	while (chunk < echunk) {
		if (ptr >= (void*) chunk->blocks
		    && ptr < (void*) (chunk->blocks + chunk->size)) {

			return(TRUE);
		}

		chunk++;
	}

	return(FALSE);
}

/********************************************************************//**
Find out if a pointer belongs to a buf_block_t. It can be a pointer to
the buf_block_t itself or a member of it
@return TRUE if ptr belongs to a buf_block_t struct */
ibool
buf_pointer_is_block_field(
/*=======================*/
	const void*	ptr)	/*!< in: pointer not dereferenced */
{
	ulint	i;

	for (i = 0; i < srv_buf_pool_instances; i++) {
		ibool	found;

		found = buf_pointer_is_block_field_instance(
			buf_pool_from_array(i), ptr);
		if (found) {
			return(TRUE);
		}
	}

	return(FALSE);
}

/********************************************************************//**
Find out if a buffer block was created by buf_chunk_init().
@return TRUE if "block" has been added to buf_pool->free by buf_chunk_init() */
static
ibool
buf_block_is_uncompressed(
/*======================*/
	buf_pool_t*		buf_pool,	/*!< in: buffer pool instance */
	const buf_block_t*	block)		/*!< in: pointer to block,
						not dereferenced */
{
	if ((((ulint) block) % sizeof *block) != 0) {
		/* The pointer should be aligned. */
		return(FALSE);
	}

	return(buf_pointer_is_block_field_instance(buf_pool, (void*) block));
}

#if defined UNIV_DEBUG || defined UNIV_IBUF_DEBUG
/********************************************************************//**
Return true if probe is enabled.
@return true if probe enabled. */
static
bool
buf_debug_execute_is_force_flush()
/*==============================*/
{
	DBUG_EXECUTE_IF("ib_buf_force_flush", return(true); );

	/* This is used during queisce testing, we want to ensure maximum
	buffering by the change buffer. */

	if (srv_ibuf_disable_background_merge) {
		return(true);
	}

	return(false);
}
#endif /* UNIV_DEBUG || UNIV_IBUF_DEBUG */

/** Wait for the block to be read in.
@param[in]	block	The block to check */
static
void
buf_wait_for_read(
	buf_block_t*	block)
{
	/* Note:

	We are using the block->lock to check for IO state (and a dirty read).
	We set the IO_READ state under the protection of the hash_lock
	(and block->mutex). This is safe because another thread can only
	access the block (and check for IO state) after the block has been
	added to the page hashtable. */

	if (buf_block_get_io_fix(block) == BUF_IO_READ) {

		/* Wait until the read operation completes */

		BPageMutex*	mutex = buf_page_get_mutex(&block->page);

		for (;;) {
			buf_io_fix	io_fix;

			mutex_enter(mutex);

			io_fix = buf_block_get_io_fix(block);

			mutex_exit(mutex);

			if (io_fix == BUF_IO_READ) {
				/* Wait by temporaly s-latch */
				rw_lock_s_lock(&block->lock);
				rw_lock_s_unlock(&block->lock);
			} else {
				break;
			}
		}
	}
}

#ifdef BTR_CUR_HASH_ADAPT
/** If a stale adaptive hash index exists on the block, drop it.
Multiple executions of btr_search_drop_page_hash_index() on the
same block must be prevented by exclusive page latch. */
ATTRIBUTE_COLD
static void buf_defer_drop_ahi(buf_block_t *block, mtr_memo_type_t fix_type)
{
  switch (fix_type) {
  case MTR_MEMO_BUF_FIX:
    /* We do not drop the adaptive hash index, because safely doing
    so would require acquiring block->lock, and that is not safe
    to acquire in some RW_NO_LATCH access paths. Those code paths
    should have no business accessing the adaptive hash index anyway. */
    break;
  case MTR_MEMO_PAGE_S_FIX:
    /* Temporarily release our S-latch. */
    rw_lock_s_unlock(&block->lock);
    rw_lock_x_lock(&block->lock);
    if (dict_index_t *index= block->index)
      if (index->freed())
        btr_search_drop_page_hash_index(block);
    rw_lock_x_unlock(&block->lock);
    rw_lock_s_lock(&block->lock);
    break;
  case MTR_MEMO_PAGE_SX_FIX:
    rw_lock_sx_unlock(&block->lock);
    rw_lock_x_lock(&block->lock);
    if (dict_index_t *index= block->index)
      if (index->freed())
        btr_search_drop_page_hash_index(block);
    rw_lock_x_unlock(&block->lock);
    rw_lock_sx_lock(&block->lock);
    break;
  default:
    ut_ad(fix_type == MTR_MEMO_PAGE_X_FIX);
    btr_search_drop_page_hash_index(block);
  }
}
#endif /* BTR_CUR_HASH_ADAPT */

/** Lock the page with the given latch type.
@param[in,out]	block		block to be locked
@param[in]	rw_latch	RW_S_LATCH, RW_X_LATCH, RW_NO_LATCH
@param[in]	mtr		mini-transaction
@param[in]	file		file name
@param[in]	line		line where called
@return pointer to locked block */
static buf_block_t* buf_page_mtr_lock(buf_block_t *block,
                                      ulint rw_latch,
                                      mtr_t* mtr,
                                      const char *file,
                                      unsigned line)
{
  mtr_memo_type_t fix_type;
  switch (rw_latch)
  {
  case RW_NO_LATCH:
    fix_type= MTR_MEMO_BUF_FIX;
    goto done;
  case RW_S_LATCH:
    rw_lock_s_lock_inline(&block->lock, 0, file, line);
    fix_type= MTR_MEMO_PAGE_S_FIX;
    break;
  case RW_SX_LATCH:
    rw_lock_sx_lock_inline(&block->lock, 0, file, line);
    fix_type= MTR_MEMO_PAGE_SX_FIX;
    break;
  default:
    ut_ad(rw_latch == RW_X_LATCH);
    rw_lock_x_lock_inline(&block->lock, 0, file, line);
    fix_type= MTR_MEMO_PAGE_X_FIX;
    break;
  }

#ifdef BTR_CUR_HASH_ADAPT
  {
    dict_index_t *index= block->index;
    if (index && index->freed())
      buf_defer_drop_ahi(block, fix_type);
  }
#endif /* BTR_CUR_HASH_ADAPT */

done:
  mtr_memo_push(mtr, block, fix_type);
  return block;
}

/** This is the low level function used to get access to a database page.
@param[in]	page_id		page id
@param[in]	rw_latch	RW_S_LATCH, RW_X_LATCH, RW_NO_LATCH
@param[in]	guess		guessed block or NULL
@param[in]	mode		BUF_GET, BUF_GET_IF_IN_POOL,
BUF_PEEK_IF_IN_POOL, BUF_GET_NO_LATCH, or BUF_GET_IF_IN_POOL_OR_WATCH
@param[in]	file		file name
@param[in]	line		line where called
@param[in]	mtr		mini-transaction
@return pointer to the block or NULL */
buf_block_t*
buf_page_get_low(
	const page_id_t		page_id,
	const page_size_t&	page_size,
	ulint			rw_latch,
	buf_block_t*		guess,
	ulint			mode,
	const char*		file,
	unsigned		line,
	mtr_t*			mtr,
	dberr_t*		err)
{
	buf_block_t*	block;
	unsigned	access_time;
	rw_lock_t*	hash_lock;
	buf_block_t*	fix_block;
	ulint		retries = 0;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);

	ut_ad((mtr == NULL) == (mode == BUF_EVICT_IF_IN_POOL));
	ut_ad(!mtr || mtr->is_active());
	ut_ad((rw_latch == RW_S_LATCH)
	      || (rw_latch == RW_X_LATCH)
	      || (rw_latch == RW_SX_LATCH)
	      || (rw_latch == RW_NO_LATCH));

	if (err) {
		*err = DB_SUCCESS;
	}

#ifdef UNIV_DEBUG
	switch (mode) {
	case BUF_EVICT_IF_IN_POOL:
		/* After DISCARD TABLESPACE, the tablespace would not exist,
		but in IMPORT TABLESPACE, PageConverter::operator() must
		replace any old pages, which were not evicted during DISCARD.
		Skip the assertion on space_page_size. */
		break;
	case BUF_PEEK_IF_IN_POOL:
	case BUF_GET_IF_IN_POOL:
		/* The caller may pass a dummy page size,
		because it does not really matter. */
		break;
	default:
		ut_error;
	case BUF_GET_NO_LATCH:
		ut_ad(rw_latch == RW_NO_LATCH);
		/* fall through */
	case BUF_GET:
	case BUF_GET_IF_IN_POOL_OR_WATCH:
	case BUF_GET_POSSIBLY_FREED:
		bool			found;
		const page_size_t&	space_page_size
			= fil_space_get_page_size(page_id.space(), &found);
		ut_ad(found);
		ut_ad(page_size.equals_to(space_page_size));
	}
#endif /* UNIV_DEBUG */

	ut_ad(!mtr || !ibuf_inside(mtr)
	      || ibuf_page_low(page_id, page_size, FALSE, file, line, NULL));

	buf_pool->stat.n_page_gets++;
	hash_lock = buf_page_hash_lock_get(buf_pool, page_id);
loop:
	block = guess;

	rw_lock_s_lock(hash_lock);

	/* If not own buf_pool_mutex, page_hash can be changed. */
	hash_lock = buf_page_hash_lock_s_confirm(hash_lock, buf_pool, page_id);

	if (block != NULL) {

		/* If the guess is a compressed page descriptor that
		has been allocated by buf_page_alloc_descriptor(),
		it may have been freed by buf_relocate(). */

		if (!buf_block_is_uncompressed(buf_pool, block)
		    || page_id != block->page.id
		    || buf_block_get_state(block) != BUF_BLOCK_FILE_PAGE) {

			/* Our guess was bogus or things have changed
			since. */
			block = guess = NULL;
		} else {
			ut_ad(!block->page.in_zip_hash);
		}
	}

	if (block == NULL) {
		block = (buf_block_t*) buf_page_hash_get_low(buf_pool, page_id);
	}

	if (!block || buf_pool_watch_is_sentinel(buf_pool, &block->page)) {
		rw_lock_s_unlock(hash_lock);
		block = NULL;
	}

	if (block == NULL) {

		/* Page not in buf_pool: needs to be read from file */

		if (mode == BUF_GET_IF_IN_POOL_OR_WATCH) {
			rw_lock_x_lock(hash_lock);

			/* If not own buf_pool_mutex,
			page_hash can be changed. */
			hash_lock = buf_page_hash_lock_x_confirm(
				hash_lock, buf_pool, page_id);

			block = (buf_block_t*) buf_pool_watch_set(
				page_id, &hash_lock);

			if (block) {
				/* We can release hash_lock after we
				increment the fix count to make
				sure that no state change takes place. */
				fix_block = block;

				if (fsp_is_system_temporary(page_id.space())) {
					/* For temporary tablespace,
					the mutex is being used for
					synchronization between user
					thread and flush thread,
					instead of block->lock. See
					buf_flush_page() for the flush
					thread counterpart. */

					BPageMutex*	fix_mutex
						= buf_page_get_mutex(
							&fix_block->page);
					mutex_enter(fix_mutex);
					buf_block_fix(fix_block);
					mutex_exit(fix_mutex);
				} else {
					buf_block_fix(fix_block);
				}

				/* Now safe to release page_hash mutex */
				rw_lock_x_unlock(hash_lock);
				goto got_block;
			}

			rw_lock_x_unlock(hash_lock);
		}

		switch (mode) {
		case BUF_GET_IF_IN_POOL:
		case BUF_GET_IF_IN_POOL_OR_WATCH:
		case BUF_PEEK_IF_IN_POOL:
		case BUF_EVICT_IF_IN_POOL:
			ut_ad(!rw_lock_own_flagged(
				      hash_lock,
				      RW_LOCK_FLAG_X | RW_LOCK_FLAG_S));
			return(NULL);
		}

		/* The call path is buf_read_page() ->
		buf_read_page_low() (fil_io()) ->
		buf_page_io_complete() ->
		buf_decrypt_after_read(). Here fil_space_t* is used
		and we decrypt -> buf_page_check_corrupt() where page
		checksums are compared. Decryption, decompression as
		well as error handling takes place at a lower level.
		Here we only need to know whether the page really is
		corrupted, or if an encrypted page with a valid
		checksum cannot be decypted. */

		dberr_t local_err = buf_read_page(page_id, page_size);

		if (local_err == DB_SUCCESS) {
			buf_read_ahead_random(page_id, page_size,
					      ibuf_inside(mtr));

			retries = 0;
		} else if (mode == BUF_GET_POSSIBLY_FREED) {
			if (err) {
				*err = local_err;
			}
			return NULL;
		} else if (retries < BUF_PAGE_READ_MAX_RETRIES) {
			++retries;

			DBUG_EXECUTE_IF(
				"innodb_page_corruption_retries",
				retries = BUF_PAGE_READ_MAX_RETRIES;
			);
		} else {
			if (err) {
				*err = local_err;
			}

			/* Pages whose encryption key is unavailable or used
			key, encryption algorithm or encryption method is
			incorrect are marked as encrypted in
			buf_page_check_corrupt(). Unencrypted page could be
			corrupted in a way where the key_id field is
			nonzero. There is no checksum on field
			FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION. */
			if (local_err == DB_DECRYPTION_FAILED) {
				return (NULL);
			}

			if (local_err == DB_PAGE_CORRUPTED
			    && srv_force_recovery) {
				return NULL;
			}

			/* Try to set table as corrupted instead of
			asserting. */
			if (page_id.space() != TRX_SYS_SPACE &&
			    dict_set_corrupted_by_space(page_id.space())) {
				return (NULL);
			}

			ib::fatal() << "Unable to read page " << page_id
				<< " into the buffer pool after "
				<< BUF_PAGE_READ_MAX_RETRIES
				<< ". The most probable cause"
				" of this error may be that the"
				" table has been corrupted."
				" See https://mariadb.com/kb/en/library/innodb-recovery-modes/";
		}

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
		ut_a(fsp_skip_sanity_check(page_id.space())
		     || ++buf_dbg_counter % 5771
		     || buf_validate());
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */
		goto loop;
	} else {
		fix_block = block;
	}

	if (fsp_is_system_temporary(page_id.space())) {
		/* For temporary tablespace, the mutex is being used
		for synchronization between user thread and flush
		thread, instead of block->lock. See buf_flush_page()
		for the flush thread counterpart. */
		BPageMutex*	fix_mutex = buf_page_get_mutex(
			&fix_block->page);
		mutex_enter(fix_mutex);
		buf_block_fix(fix_block);
		mutex_exit(fix_mutex);
	} else {
		buf_block_fix(fix_block);
	}

	/* Now safe to release page_hash mutex */
	rw_lock_s_unlock(hash_lock);

got_block:

	switch (mode) {
	case BUF_GET_IF_IN_POOL:
	case BUF_PEEK_IF_IN_POOL:
	case BUF_EVICT_IF_IN_POOL:
		buf_page_t*	fix_page = &fix_block->page;
		BPageMutex*	fix_mutex = buf_page_get_mutex(fix_page);
		mutex_enter(fix_mutex);
		const bool	must_read
			= (buf_page_get_io_fix(fix_page) == BUF_IO_READ);
		mutex_exit(fix_mutex);

		if (must_read) {
			/* The page is being read to buffer pool,
			but we cannot wait around for the read to
			complete. */
			buf_block_unfix(fix_block);

			return(NULL);
		}
	}

	switch (buf_block_get_state(fix_block)) {
		buf_page_t*	bpage;

	case BUF_BLOCK_FILE_PAGE:
		bpage = &block->page;
		if (fsp_is_system_temporary(page_id.space())
		    && buf_page_get_io_fix(bpage) != BUF_IO_NONE) {
			/* This suggests that the page is being flushed.
			Avoid returning reference to this page.
			Instead wait for the flush action to complete. */
			buf_block_unfix(fix_block);
			os_thread_sleep(WAIT_FOR_WRITE);
			goto loop;
		}

		if (UNIV_UNLIKELY(mode == BUF_EVICT_IF_IN_POOL)) {
evict_from_pool:
			ut_ad(!fix_block->page.oldest_modification);
			buf_pool_mutex_enter(buf_pool);
			buf_block_unfix(fix_block);

			if (!buf_LRU_free_page(&fix_block->page, true)) {
				ut_ad(0);
			}

			buf_pool_mutex_exit(buf_pool);
			return(NULL);
		}

		break;

	case BUF_BLOCK_ZIP_PAGE:
	case BUF_BLOCK_ZIP_DIRTY:
		if (mode == BUF_PEEK_IF_IN_POOL) {
			/* This mode is only used for dropping an
			adaptive hash index.  There cannot be an
			adaptive hash index for a compressed-only
			page, so do not bother decompressing the page. */
			buf_block_unfix(fix_block);

			return(NULL);
		}

		bpage = &block->page;

		/* Note: We have already buffer fixed this block. */
		if (bpage->buf_fix_count > 1
		    || buf_page_get_io_fix(bpage) != BUF_IO_NONE) {

			/* This condition often occurs when the buffer
			is not buffer-fixed, but I/O-fixed by
			buf_page_init_for_read(). */
			buf_block_unfix(fix_block);

			/* The block is buffer-fixed or I/O-fixed.
			Try again later. */
			os_thread_sleep(WAIT_FOR_READ);

			goto loop;
		}

		if (UNIV_UNLIKELY(mode == BUF_EVICT_IF_IN_POOL)) {
			goto evict_from_pool;
		}

		/* Buffer-fix the block so that it cannot be evicted
		or relocated while we are attempting to allocate an
		uncompressed page. */

		block = buf_LRU_get_free_block(buf_pool);

		buf_pool_mutex_enter(buf_pool);

		/* If not own buf_pool_mutex, page_hash can be changed. */
		hash_lock = buf_page_hash_lock_get(buf_pool, page_id);

		rw_lock_x_lock(hash_lock);

		/* Buffer-fixing prevents the page_hash from changing. */
		ut_ad(bpage == buf_page_hash_get_low(buf_pool, page_id));

		buf_block_unfix(fix_block);

		buf_page_mutex_enter(block);
		mutex_enter(&buf_pool->zip_mutex);

		fix_block = block;

		if (bpage->buf_fix_count > 0
		    || buf_page_get_io_fix(bpage) != BUF_IO_NONE) {

			mutex_exit(&buf_pool->zip_mutex);
			/* The block was buffer-fixed or I/O-fixed while
			buf_pool->mutex was not held by this thread.
			Free the block that was allocated and retry.
			This should be extremely unlikely, for example,
			if buf_page_get_zip() was invoked. */

			buf_LRU_block_free_non_file_page(block);
			buf_pool_mutex_exit(buf_pool);
			rw_lock_x_unlock(hash_lock);
			buf_page_mutex_exit(block);

			/* Try again */
			goto loop;
		}

		/* Move the compressed page from bpage to block,
		and uncompress it. */

		/* Note: this is the uncompressed block and it is not
		accessible by other threads yet because it is not in
		any list or hash table */
		buf_relocate(bpage, &block->page);

		buf_block_init_low(block);

		/* Set after buf_relocate(). */
		block->page.buf_fix_count = 1;

		block->lock_hash_val = lock_rec_hash(page_id.space(),
						     page_id.page_no());

		UNIV_MEM_DESC(&block->page.zip.data,
			      page_zip_get_size(&block->page.zip));

		if (buf_page_get_state(&block->page) == BUF_BLOCK_ZIP_PAGE) {
#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
			UT_LIST_REMOVE(buf_pool->zip_clean, &block->page);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */
			ut_ad(!block->page.in_flush_list);
		} else {
			/* Relocate buf_pool->flush_list. */
			buf_flush_relocate_on_flush_list(bpage, &block->page);
		}

		/* Buffer-fix, I/O-fix, and X-latch the block
		for the duration of the decompression.
		Also add the block to the unzip_LRU list. */
		block->page.state = BUF_BLOCK_FILE_PAGE;

		/* Insert at the front of unzip_LRU list */
		buf_unzip_LRU_add_block(block, FALSE);

		buf_block_set_io_fix(block, BUF_IO_READ);
		rw_lock_x_lock_inline(&block->lock, 0, file, line);

		UNIV_MEM_INVALID(bpage, sizeof *bpage);

		rw_lock_x_unlock(hash_lock);
		buf_pool->n_pend_unzip++;
		mutex_exit(&buf_pool->zip_mutex);
		buf_pool_mutex_exit(buf_pool);

		access_time = buf_page_is_accessed(&block->page);

		buf_page_mutex_exit(block);

		buf_page_free_descriptor(bpage);

		/* Decompress the page while not holding
		buf_pool->mutex or block->mutex. */

		{
			bool	success = buf_zip_decompress(block, false);

			if (!success) {
				buf_pool_mutex_enter(buf_pool);
				buf_page_mutex_enter(fix_block);
				buf_block_set_io_fix(fix_block, BUF_IO_NONE);
				buf_page_mutex_exit(fix_block);

				--buf_pool->n_pend_unzip;
				buf_block_unfix(fix_block);
				buf_pool_mutex_exit(buf_pool);
				rw_lock_x_unlock(&fix_block->lock);

				if (err) {
					*err = DB_PAGE_CORRUPTED;
				}
				return NULL;
			}
		}

		if (!access_time && !recv_no_ibuf_operations) {
			ibuf_merge_or_delete_for_page(
				block, page_id, &page_size, TRUE);
		}

		buf_pool_mutex_enter(buf_pool);

		buf_page_mutex_enter(fix_block);

		buf_block_set_io_fix(fix_block, BUF_IO_NONE);

		buf_page_mutex_exit(fix_block);

		--buf_pool->n_pend_unzip;

		buf_pool_mutex_exit(buf_pool);

		rw_lock_x_unlock(&block->lock);

		break;

	case BUF_BLOCK_POOL_WATCH:
	case BUF_BLOCK_NOT_USED:
	case BUF_BLOCK_READY_FOR_USE:
	case BUF_BLOCK_MEMORY:
	case BUF_BLOCK_REMOVE_HASH:
		ut_error;
		break;
	}

	ut_ad(block == fix_block);
	ut_ad(fix_block->page.buf_fix_count > 0);

	ut_ad(!rw_lock_own_flagged(hash_lock,
				   RW_LOCK_FLAG_X | RW_LOCK_FLAG_S));

	ut_ad(buf_block_get_state(fix_block) == BUF_BLOCK_FILE_PAGE);

#if defined UNIV_DEBUG || defined UNIV_IBUF_DEBUG

	if ((mode == BUF_GET_IF_IN_POOL || mode == BUF_GET_IF_IN_POOL_OR_WATCH)
	    && (ibuf_debug || buf_debug_execute_is_force_flush())) {

		/* Try to evict the block from the buffer pool, to use the
		insert buffer (change buffer) as much as possible. */

		buf_pool_mutex_enter(buf_pool);

		buf_block_unfix(fix_block);

		/* Now we are only holding the buf_pool->mutex,
		not block->mutex or hash_lock. Blocks cannot be
		relocated or enter or exit the buf_pool while we
		are holding the buf_pool->mutex. */

		if (buf_LRU_free_page(&fix_block->page, true)) {

			buf_pool_mutex_exit(buf_pool);

			/* If not own buf_pool_mutex,
			page_hash can be changed. */
			hash_lock = buf_page_hash_lock_get(buf_pool, page_id);

			rw_lock_x_lock(hash_lock);

			/* If not own buf_pool_mutex,
			page_hash can be changed. */
			hash_lock = buf_page_hash_lock_x_confirm(
				hash_lock, buf_pool, page_id);

			if (mode == BUF_GET_IF_IN_POOL_OR_WATCH) {
				/* Set the watch, as it would have
				been set if the page were not in the
				buffer pool in the first place. */
				block = (buf_block_t*) buf_pool_watch_set(
					page_id, &hash_lock);
			} else {
				block = (buf_block_t*) buf_page_hash_get_low(
					buf_pool, page_id);
			}

			rw_lock_x_unlock(hash_lock);

			if (block != NULL) {
				/* Either the page has been read in or
				a watch was set on that in the window
				where we released the buf_pool::mutex
				and before we acquire the hash_lock
				above. Try again. */
				guess = block;

				goto loop;
			}

			return(NULL);
		}

		buf_page_mutex_enter(fix_block);

		if (buf_flush_page_try(buf_pool, fix_block)) {
			guess = fix_block;

			goto loop;
		}

		buf_page_mutex_exit(fix_block);

		buf_block_fix(fix_block);

		/* Failed to evict the page; change it directly */

		buf_pool_mutex_exit(buf_pool);
	}
#endif /* UNIV_DEBUG || UNIV_IBUF_DEBUG */

	ut_ad(fix_block->page.buf_fix_count > 0);

#ifdef UNIV_DEBUG
	/* We have already buffer fixed the page, and we are committed to
	returning this page to the caller. Register for debugging.
	Avoid debug latching if page/block belongs to system temporary
	tablespace (Not much needed for table with single threaded access.). */
	if (!fsp_is_system_temporary(page_id.space())) {
		ibool   ret;
		ret = rw_lock_s_lock_nowait(
			&fix_block->debug_latch, file, line);
		ut_a(ret);
	}
#endif /* UNIV_DEBUG */

	/* While tablespace is reinited the indexes are already freed but the
	blocks related to it still resides in buffer pool. Trying to remove
	such blocks from buffer pool would invoke removal of AHI entries
	associated with these blocks. Logic to remove AHI entry will try to
	load the block but block is already in free state. Handle the said case
	with mode = BUF_PEEK_IF_IN_POOL that is invoked from
	"btr_search_drop_page_hash_when_freed". */
	ut_ad(mode == BUF_GET_POSSIBLY_FREED
	      || mode == BUF_PEEK_IF_IN_POOL
	      || !fix_block->page.file_page_was_freed);

	/* Check if this is the first access to the page */
	access_time = buf_page_is_accessed(&fix_block->page);

	/* This is a heuristic and we don't care about ordering issues. */
	if (access_time == 0) {
		buf_page_mutex_enter(fix_block);

		buf_page_set_accessed(&fix_block->page);

		buf_page_mutex_exit(fix_block);
	}

	if (mode != BUF_PEEK_IF_IN_POOL) {
		buf_page_make_young_if_needed(&fix_block->page);
	}

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(fsp_skip_sanity_check(page_id.space())
	     || ++buf_dbg_counter % 5771
	     || buf_validate());
	ut_a(buf_block_get_state(fix_block) == BUF_BLOCK_FILE_PAGE);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

	/* We have to wait here because the IO_READ state was set
	under the protection of the hash_lock and not the block->mutex
	and block->lock. */
	buf_wait_for_read(fix_block);

	if (fix_block->page.id != page_id) {

		buf_block_unfix(fix_block);

#ifdef UNIV_DEBUG
		if (!fsp_is_system_temporary(page_id.space())) {
			rw_lock_s_unlock(&fix_block->debug_latch);
		}
#endif /* UNIV_DEBUG */

		if (err) {
			*err = DB_PAGE_CORRUPTED;
		}

		return NULL;
	}

	fix_block = buf_page_mtr_lock(fix_block, rw_latch, mtr, file, line);

	if (mode != BUF_PEEK_IF_IN_POOL && !access_time) {
		/* In the case of a first access, try to apply linear
		read-ahead */

		buf_read_ahead_linear(page_id, page_size, ibuf_inside(mtr));
	}

	ut_ad(!rw_lock_own_flagged(hash_lock,
				   RW_LOCK_FLAG_X | RW_LOCK_FLAG_S));

	return(fix_block);
}

/** This is the general function used to get access to a database page.
It does page initialization and applies the buffered redo logs.
@param[in]	page_id		page id
@param[in]	rw_latch	RW_S_LATCH, RW_X_LATCH, RW_NO_LATCH
@param[in]	guess		guessed block or NULL
@param[in]	mode		BUF_GET, BUF_GET_IF_IN_POOL,
BUF_PEEK_IF_IN_POOL, BUF_GET_NO_LATCH, or BUF_GET_IF_IN_POOL_OR_WATCH
@param[in]	file		file name
@param[in]	line		line where called
@param[in]	mtr		mini-transaction
@param[out]	err		DB_SUCCESS or error code
@return pointer to the block or NULL */
buf_block_t*
buf_page_get_gen(
	const page_id_t		page_id,
	const page_size_t&	page_size,
	ulint			rw_latch,
	buf_block_t*		guess,
	ulint			mode,
	const char*		file,
	unsigned		line,
	mtr_t*			mtr,
	dberr_t*		err)
{
  if (buf_block_t *block = recv_recovery_create_page(page_id))
  {
    buf_block_fix(block);
    ut_ad(rw_lock_s_lock_nowait(&block->debug_latch, file, line));
    block= buf_page_mtr_lock(block, rw_latch, mtr, file, line);
    return block;
  }

  return buf_page_get_low(page_id, page_size, rw_latch,
                          guess, mode, file, line, mtr, err);
}

/********************************************************************//**
This is the general function used to get optimistic access to a database
page.
@return TRUE if success */
ibool
buf_page_optimistic_get(
/*====================*/
	ulint		rw_latch,/*!< in: RW_S_LATCH, RW_X_LATCH */
	buf_block_t*	block,	/*!< in: guessed buffer block */
	ib_uint64_t	modify_clock,/*!< in: modify clock value */
	const char*	file,	/*!< in: file name */
	unsigned	line,	/*!< in: line where called */
	mtr_t*		mtr)	/*!< in: mini-transaction */
{
	buf_pool_t*	buf_pool;
	unsigned	access_time;
	ibool		success;

	ut_ad(block);
	ut_ad(mtr);
	ut_ad(mtr->is_active());
	ut_ad((rw_latch == RW_S_LATCH) || (rw_latch == RW_X_LATCH));

	buf_page_mutex_enter(block);

	if (UNIV_UNLIKELY(buf_block_get_state(block) != BUF_BLOCK_FILE_PAGE)) {

		buf_page_mutex_exit(block);

		return(FALSE);
	}

	buf_block_buf_fix_inc(block, file, line);

	access_time = buf_page_is_accessed(&block->page);

	buf_page_set_accessed(&block->page);

	buf_page_mutex_exit(block);

	buf_page_make_young_if_needed(&block->page);

	ut_ad(!ibuf_inside(mtr)
	      || ibuf_page(block->page.id, block->page.size, NULL));

	mtr_memo_type_t	fix_type;

	switch (rw_latch) {
	case RW_S_LATCH:
		success = rw_lock_s_lock_nowait(&block->lock, file, line);

		fix_type = MTR_MEMO_PAGE_S_FIX;
		break;
	case RW_X_LATCH:
		success = rw_lock_x_lock_func_nowait_inline(
			&block->lock, file, line);

		fix_type = MTR_MEMO_PAGE_X_FIX;
		break;
	default:
		ut_error; /* RW_SX_LATCH is not implemented yet */
	}

	if (!success) {
		buf_block_buf_fix_dec(block);
		return(FALSE);
	}

	if (modify_clock != block->modify_clock) {

		buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK);

		if (rw_latch == RW_S_LATCH) {
			rw_lock_s_unlock(&block->lock);
		} else {
			rw_lock_x_unlock(&block->lock);
		}

		buf_block_buf_fix_dec(block);
		return(FALSE);
	}

	mtr_memo_push(mtr, block, fix_type);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(fsp_skip_sanity_check(block->page.id.space())
	     || ++buf_dbg_counter % 5771
	     || buf_validate());
	ut_a(block->page.buf_fix_count > 0);
	ut_a(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

	ut_d(buf_page_mutex_enter(block));
	ut_ad(!block->page.file_page_was_freed);
	ut_d(buf_page_mutex_exit(block));

	if (!access_time) {
		/* In the case of a first access, try to apply linear
		read-ahead */
		buf_read_ahead_linear(block->page.id, block->page.size,
				      ibuf_inside(mtr));
	}

	buf_pool = buf_pool_from_block(block);
	buf_pool->stat.n_page_gets++;

	return(TRUE);
}

/********************************************************************//**
This is used to get access to a known database page, when no waiting can be
done. For example, if a search in an adaptive hash index leads us to this
frame.
@return TRUE if success */
ibool
buf_page_get_known_nowait(
/*======================*/
	ulint		rw_latch,/*!< in: RW_S_LATCH, RW_X_LATCH */
	buf_block_t*	block,	/*!< in: the known page */
	ulint		mode,	/*!< in: BUF_MAKE_YOUNG or BUF_KEEP_OLD */
	const char*	file,	/*!< in: file name */
	unsigned	line,	/*!< in: line where called */
	mtr_t*		mtr)	/*!< in: mini-transaction */
{
	buf_pool_t*	buf_pool;
	ibool		success;

	ut_ad(mtr->is_active());
	ut_ad((rw_latch == RW_S_LATCH) || (rw_latch == RW_X_LATCH));

	buf_page_mutex_enter(block);

	if (buf_block_get_state(block) == BUF_BLOCK_REMOVE_HASH) {
		/* Another thread is just freeing the block from the LRU list
		of the buffer pool: do not try to access this page; this
		attempt to access the page can only come through the hash
		index because when the buffer block state is ..._REMOVE_HASH,
		we have already removed it from the page address hash table
		of the buffer pool. */

		buf_page_mutex_exit(block);

		return(FALSE);
	}

	ut_a(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);

	buf_block_buf_fix_inc(block, file, line);

	buf_page_set_accessed(&block->page);

	buf_page_mutex_exit(block);

	buf_pool = buf_pool_from_block(block);

#ifdef BTR_CUR_HASH_ADAPT
	if (mode == BUF_MAKE_YOUNG) {
		buf_page_make_young_if_needed(&block->page);
	}
#endif /* BTR_CUR_HASH_ADAPT */

	ut_ad(!ibuf_inside(mtr) || mode == BUF_KEEP_OLD);

	mtr_memo_type_t	fix_type;

	switch (rw_latch) {
	case RW_S_LATCH:
		success = rw_lock_s_lock_nowait(&block->lock, file, line);
		fix_type = MTR_MEMO_PAGE_S_FIX;
		break;
	case RW_X_LATCH:
		success = rw_lock_x_lock_func_nowait_inline(
			&block->lock, file, line);

		fix_type = MTR_MEMO_PAGE_X_FIX;
		break;
	default:
		ut_error; /* RW_SX_LATCH is not implemented yet */
	}

	if (!success) {
		buf_block_buf_fix_dec(block);
		return(FALSE);
	}

	mtr_memo_push(mtr, block, fix_type);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(++buf_dbg_counter % 5771 || buf_validate());
	ut_a(block->page.buf_fix_count > 0);
	ut_a(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

#ifdef UNIV_DEBUG
	if (mode != BUF_KEEP_OLD) {
		/* If mode == BUF_KEEP_OLD, we are executing an I/O
		completion routine.  Avoid a bogus assertion failure
		when ibuf_merge_or_delete_for_page() is processing a
		page that was just freed due to DROP INDEX, or
		deleting a record from SYS_INDEXES. This check will be
		skipped in recv_recover_page() as well. */

# ifdef BTR_CUR_HASH_ADAPT
		ut_ad(!block->page.file_page_was_freed
		      || (block->index && block->index->freed()));
# else /* BTR_CUR_HASH_ADAPT */
		ut_ad(!block->page.file_page_was_freed);
# endif /* BTR_CUR_HASH_ADAPT */
	}
#endif /* UNIV_DEBUG */

	buf_pool->stat.n_page_gets++;

	return(TRUE);
}

/** Given a tablespace id and page number tries to get that page. If the
page is not in the buffer pool it is not loaded and NULL is returned.
Suitable for using when holding the lock_sys_t::mutex.
@param[in]	page_id	page id
@param[in]	file	file name
@param[in]	line	line where called
@param[in]	mtr	mini-transaction
@return pointer to a page or NULL */
buf_block_t*
buf_page_try_get_func(
	const page_id_t		page_id,
	const char*		file,
	unsigned		line,
	mtr_t*			mtr)
{
	buf_block_t*	block;
	ibool		success;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);
	rw_lock_t*	hash_lock;

	ut_ad(mtr);
	ut_ad(mtr->is_active());

	block = buf_block_hash_get_s_locked(buf_pool, page_id, &hash_lock);

	if (!block || buf_block_get_state(block) != BUF_BLOCK_FILE_PAGE) {
		if (block) {
			rw_lock_s_unlock(hash_lock);
		}
		return(NULL);
	}

	ut_ad(!buf_pool_watch_is_sentinel(buf_pool, &block->page));

	buf_page_mutex_enter(block);
	rw_lock_s_unlock(hash_lock);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);
	ut_a(page_id == block->page.id);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

	buf_block_buf_fix_inc(block, file, line);
	buf_page_mutex_exit(block);

	mtr_memo_type_t	fix_type = MTR_MEMO_PAGE_S_FIX;
	success = rw_lock_s_lock_nowait(&block->lock, file, line);

	if (!success) {
		/* Let us try to get an X-latch. If the current thread
		is holding an X-latch on the page, we cannot get an
		S-latch. */

		fix_type = MTR_MEMO_PAGE_X_FIX;
		success = rw_lock_x_lock_func_nowait_inline(&block->lock,
							    file, line);
	}

	if (!success) {
		buf_block_buf_fix_dec(block);
		return(NULL);
	}

	mtr_memo_push(mtr, block, fix_type);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(fsp_skip_sanity_check(block->page.id.space())
	     || ++buf_dbg_counter % 5771
	     || buf_validate());
	ut_a(block->page.buf_fix_count > 0);
	ut_a(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

	ut_d(buf_page_mutex_enter(block));
	ut_d(ut_a(!block->page.file_page_was_freed));
	ut_d(buf_page_mutex_exit(block));

	buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK);

	buf_pool->stat.n_page_gets++;

	return(block);
}

/********************************************************************//**
Initialize some fields of a control block. */
UNIV_INLINE
void
buf_page_init_low(
/*==============*/
	buf_page_t*	bpage)	/*!< in: block to init */
{
	bpage->flush_type = BUF_FLUSH_LRU;
	bpage->io_fix = BUF_IO_NONE;
	bpage->buf_fix_count = 0;
	bpage->old = 0;
	bpage->freed_page_clock = 0;
	bpage->access_time = 0;
	bpage->newest_modification = 0;
	bpage->oldest_modification = 0;
	bpage->real_size = 0;
	bpage->slot = NULL;

	HASH_INVALIDATE(bpage, hash);

	ut_d(bpage->file_page_was_freed = FALSE);
}

/** Inits a page to the buffer buf_pool.
@param[in,out]	buf_pool	buffer pool
@param[in]	page_id		page id
@param[in,out]	block		block to init */
static
void
buf_page_init(
	buf_pool_t*		buf_pool,
	const page_id_t		page_id,
	const page_size_t&	page_size,
	buf_block_t*		block)
{
	buf_page_t*	hash_page;

	ut_ad(buf_pool == buf_pool_get(page_id));
	ut_ad(buf_pool_mutex_own(buf_pool));

	ut_ad(buf_page_mutex_own(block));
	ut_a(buf_block_get_state(block) != BUF_BLOCK_FILE_PAGE);

	ut_ad(rw_lock_own(buf_page_hash_lock_get(buf_pool, page_id),
			  RW_LOCK_X));

	/* Set the state of the block */
	buf_block_set_file_page(block, page_id);

#ifdef UNIV_DEBUG_VALGRIND
	if (is_system_tablespace(page_id.space())) {
		/* Silence valid Valgrind warnings about uninitialized
		data being written to data files.  There are some unused
		bytes on some pages that InnoDB does not initialize. */
		UNIV_MEM_VALID(block->frame, UNIV_PAGE_SIZE);
	}
#endif /* UNIV_DEBUG_VALGRIND */

	buf_block_init_low(block);

	block->lock_hash_val = lock_rec_hash(page_id.space(),
					     page_id.page_no());

	buf_page_init_low(&block->page);

	/* Insert into the hash table of file pages */

	hash_page = buf_page_hash_get_low(buf_pool, page_id);

	if (hash_page == NULL) {
		/* Block not found in hash table */
	} else if (UNIV_LIKELY(buf_pool_watch_is_sentinel(buf_pool,
							  hash_page))) {
		/* Preserve the reference count. */
		ib_uint32_t	buf_fix_count = hash_page->buf_fix_count;

		ut_a(buf_fix_count > 0);

		my_atomic_add32((int32*) &block->page.buf_fix_count, buf_fix_count);

		buf_pool_watch_remove(buf_pool, hash_page);
	} else {
		ib::fatal() << "Page already foudn in the hash table: "
			    << page_id;
	}

	ut_ad(!block->page.in_zip_hash);
	ut_ad(!block->page.in_page_hash);
	ut_d(block->page.in_page_hash = TRUE);

	block->page.id = page_id;
	block->page.size.copy_from(page_size);

	HASH_INSERT(buf_page_t, hash, buf_pool->page_hash,
		    page_id.fold(), &block->page);

	if (page_size.is_compressed()) {
		page_zip_set_size(&block->page.zip, page_size.physical());
	}
}

/** Initialize a page for read to the buffer buf_pool. If the page is
(1) already in buf_pool, or
(2) if we specify to read only ibuf pages and the page is not an ibuf page, or
(3) if the space is deleted or being deleted,
then this function does nothing.
Sets the io_fix flag to BUF_IO_READ and sets a non-recursive exclusive lock
on the buffer frame. The io-handler must take care that the flag is cleared
and the lock released later.
@param[out]	err			DB_SUCCESS or DB_TABLESPACE_DELETED
@param[in]	mode			BUF_READ_IBUF_PAGES_ONLY, ...
@param[in]	page_id			page id
@param[in]	unzip			whether the uncompressed page is
					requested (for ROW_FORMAT=COMPRESSED)
@return pointer to the block
@retval	NULL	in case of an error */
buf_page_t*
buf_page_init_for_read(
	dberr_t*		err,
	ulint			mode,
	const page_id_t		page_id,
	const page_size_t&	page_size,
	bool			unzip)
{
	buf_block_t*	block;
	buf_page_t*	bpage	= NULL;
	buf_page_t*	watch_page;
	rw_lock_t*	hash_lock;
	mtr_t		mtr;
	ibool		lru	= FALSE;
	void*		data;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);

	ut_ad(buf_pool);

	*err = DB_SUCCESS;

	if (mode == BUF_READ_IBUF_PAGES_ONLY) {
		/* It is a read-ahead within an ibuf routine */

		ut_ad(!ibuf_bitmap_page(page_id, page_size));

		ibuf_mtr_start(&mtr);

		if (!recv_no_ibuf_operations &&
		    !ibuf_page(page_id, page_size, &mtr)) {

			ibuf_mtr_commit(&mtr);

			return(NULL);
		}
	} else {
		ut_ad(mode == BUF_READ_ANY_PAGE);
	}

	if (page_size.is_compressed() && !unzip && !recv_recovery_is_on()) {
		block = NULL;
	} else {
		block = buf_LRU_get_free_block(buf_pool);
		ut_ad(block);
		ut_ad(buf_pool_from_block(block) == buf_pool);
	}

	buf_pool_mutex_enter(buf_pool);

	hash_lock = buf_page_hash_lock_get(buf_pool, page_id);
	rw_lock_x_lock(hash_lock);

	watch_page = buf_page_hash_get_low(buf_pool, page_id);
	if (watch_page && !buf_pool_watch_is_sentinel(buf_pool, watch_page)) {
		/* The page is already in the buffer pool. */
		watch_page = NULL;
		rw_lock_x_unlock(hash_lock);
		if (block) {
			buf_page_mutex_enter(block);
			buf_LRU_block_free_non_file_page(block);
			buf_page_mutex_exit(block);
		}

		bpage = NULL;
		goto func_exit;
	}

	if (block) {
		bpage = &block->page;

		buf_page_mutex_enter(block);

		ut_ad(buf_pool_from_bpage(bpage) == buf_pool);

		buf_page_init(buf_pool, page_id, page_size, block);

		/* Note: We are using the hash_lock for protection. This is
		safe because no other thread can lookup the block from the
		page hashtable yet. */

		buf_page_set_io_fix(bpage, BUF_IO_READ);

		rw_lock_x_unlock(hash_lock);

		/* The block must be put to the LRU list, to the old blocks */
		buf_LRU_add_block(bpage, TRUE/* to old blocks */);

		/* We set a pass-type x-lock on the frame because then
		the same thread which called for the read operation
		(and is running now at this point of code) can wait
		for the read to complete by waiting for the x-lock on
		the frame; if the x-lock were recursive, the same
		thread would illegally get the x-lock before the page
		read is completed.  The x-lock is cleared by the
		io-handler thread. */

		rw_lock_x_lock_gen(&block->lock, BUF_IO_READ);

		if (page_size.is_compressed()) {
			/* buf_pool->mutex may be released and
			reacquired by buf_buddy_alloc().  Thus, we
			must release block->mutex in order not to
			break the latching order in the reacquisition
			of buf_pool->mutex.  We also must defer this
			operation until after the block descriptor has
			been added to buf_pool->LRU and
			buf_pool->page_hash. */
			buf_page_mutex_exit(block);
			data = buf_buddy_alloc(buf_pool, page_size.physical(),
					       &lru);
			buf_page_mutex_enter(block);
			block->page.zip.data = (page_zip_t*) data;

			/* To maintain the invariant
			block->in_unzip_LRU_list
			== buf_page_belongs_to_unzip_LRU(&block->page)
			we have to add this block to unzip_LRU
			after block->page.zip.data is set. */
			ut_ad(buf_page_belongs_to_unzip_LRU(&block->page));
			buf_unzip_LRU_add_block(block, TRUE);
		}

		buf_page_mutex_exit(block);
	} else {
		rw_lock_x_unlock(hash_lock);

		/* The compressed page must be allocated before the
		control block (bpage), in order to avoid the
		invocation of buf_buddy_relocate_block() on
		uninitialized data. */
		data = buf_buddy_alloc(buf_pool, page_size.physical(), &lru);

		rw_lock_x_lock(hash_lock);

		/* If buf_buddy_alloc() allocated storage from the LRU list,
		it released and reacquired buf_pool->mutex.  Thus, we must
		check the page_hash again, as it may have been modified. */
		if (UNIV_UNLIKELY(lru)) {

			watch_page = buf_page_hash_get_low(buf_pool, page_id);

			if (UNIV_UNLIKELY(watch_page
			    && !buf_pool_watch_is_sentinel(buf_pool,
							   watch_page))) {

				/* The block was added by some other thread. */
				rw_lock_x_unlock(hash_lock);
				watch_page = NULL;
				buf_buddy_free(buf_pool, data,
					       page_size.physical());

				bpage = NULL;
				goto func_exit;
			}
		}

		bpage = buf_page_alloc_descriptor();

		/* Initialize the buf_pool pointer. */
		bpage->buf_pool_index = buf_pool_index(buf_pool);

		page_zip_des_init(&bpage->zip);
		page_zip_set_size(&bpage->zip, page_size.physical());
		bpage->zip.data = (page_zip_t*) data;

		bpage->size.copy_from(page_size);

		mutex_enter(&buf_pool->zip_mutex);
		UNIV_MEM_DESC(bpage->zip.data, bpage->size.physical());

		buf_page_init_low(bpage);

		bpage->state = BUF_BLOCK_ZIP_PAGE;
		bpage->id = page_id;
		bpage->flush_observer = NULL;

		ut_d(bpage->in_page_hash = FALSE);
		ut_d(bpage->in_zip_hash = FALSE);
		ut_d(bpage->in_flush_list = FALSE);
		ut_d(bpage->in_free_list = FALSE);
		ut_d(bpage->in_LRU_list = FALSE);

		ut_d(bpage->in_page_hash = TRUE);

		if (watch_page != NULL) {

			/* Preserve the reference count. */
			ib_uint32_t	buf_fix_count;

			buf_fix_count = watch_page->buf_fix_count;

			ut_a(buf_fix_count > 0);

			my_atomic_add32((int32*) &bpage->buf_fix_count, buf_fix_count);

			ut_ad(buf_pool_watch_is_sentinel(buf_pool, watch_page));
			buf_pool_watch_remove(buf_pool, watch_page);
		}

		HASH_INSERT(buf_page_t, hash, buf_pool->page_hash,
			    bpage->id.fold(), bpage);

		rw_lock_x_unlock(hash_lock);

		/* The block must be put to the LRU list, to the old blocks.
		The zip size is already set into the page zip */
		buf_LRU_add_block(bpage, TRUE/* to old blocks */);
#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
		buf_LRU_insert_zip_clean(bpage);
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

		buf_page_set_io_fix(bpage, BUF_IO_READ);

		mutex_exit(&buf_pool->zip_mutex);
	}

	buf_pool->n_pend_reads++;
func_exit:
	buf_pool_mutex_exit(buf_pool);

	if (mode == BUF_READ_IBUF_PAGES_ONLY) {

		ibuf_mtr_commit(&mtr);
	}

	ut_ad(!rw_lock_own_flagged(hash_lock,
				   RW_LOCK_FLAG_X | RW_LOCK_FLAG_S));
	ut_ad(!bpage || buf_page_in_file(bpage));

	return(bpage);
}

/** Initializes a page to the buffer buf_pool. The page is usually not read
from a file even if it cannot be found in the buffer buf_pool. This is one
of the functions which perform to a block a state transition NOT_USED =>
FILE_PAGE (the other is buf_page_get_gen).
@param[in]	page_id		page id
@param[in]	page_size	page size
@param[in]	mtr		mini-transaction
@return pointer to the block, page bufferfixed */
buf_block_t*
buf_page_create(
	const page_id_t		page_id,
	const page_size_t&	page_size,
	mtr_t*			mtr)
{
	buf_frame_t*	frame;
	buf_block_t*	block;
	buf_block_t*	free_block	= NULL;
	buf_pool_t*	buf_pool = buf_pool_get(page_id);
	rw_lock_t*	hash_lock;

	ut_ad(mtr->is_active());
	ut_ad(page_id.space() != 0 || !page_size.is_compressed());

	free_block = buf_LRU_get_free_block(buf_pool);

	buf_pool_mutex_enter(buf_pool);

	hash_lock = buf_page_hash_lock_get(buf_pool, page_id);
	rw_lock_x_lock(hash_lock);

	block = (buf_block_t*) buf_page_hash_get_low(buf_pool, page_id);

	if (block
	    && buf_page_in_file(&block->page)
	    && !buf_pool_watch_is_sentinel(buf_pool, &block->page)) {
		ut_d(block->page.file_page_was_freed = FALSE);

		/* Page can be found in buf_pool */
		buf_pool_mutex_exit(buf_pool);
		rw_lock_x_unlock(hash_lock);

		buf_block_free(free_block);
#ifdef BTR_CUR_HASH_ADAPT
		if (block->page.state == BUF_BLOCK_FILE_PAGE
		    && UNIV_LIKELY_NULL(block->index)) {
			btr_search_drop_page_hash_index(block);
		}
#endif /* BTR_CUR_HASH_ADAPT */

		if (!recv_recovery_is_on()) {
			return buf_page_get_with_no_latch(page_id, page_size,
							  mtr);
		}

		mutex_exit(&recv_sys->mutex);
		block = buf_page_get_with_no_latch(page_id, page_size, mtr);
		mutex_enter(&recv_sys->mutex);
		return block;
	}

	/* If we get here, the page was not in buf_pool: init it there */

	DBUG_PRINT("ib_buf", ("create page %u:%u",
			      page_id.space(), page_id.page_no()));

	block = free_block;

	buf_page_mutex_enter(block);

	buf_page_init(buf_pool, page_id, page_size, block);

	rw_lock_x_unlock(hash_lock);

	/* The block must be put to the LRU list */
	buf_LRU_add_block(&block->page, FALSE);

	buf_block_buf_fix_inc(block, __FILE__, __LINE__);
	buf_pool->stat.n_pages_created++;

	if (page_size.is_compressed()) {
		void*	data;
		ibool	lru;

		/* Prevent race conditions during buf_buddy_alloc(),
		which may release and reacquire buf_pool->mutex,
		by IO-fixing and X-latching the block. */

		buf_page_set_io_fix(&block->page, BUF_IO_READ);
		rw_lock_x_lock(&block->lock);

		buf_page_mutex_exit(block);
		/* buf_pool->mutex may be released and reacquired by
		buf_buddy_alloc().  Thus, we must release block->mutex
		in order not to break the latching order in
		the reacquisition of buf_pool->mutex.  We also must
		defer this operation until after the block descriptor
		has been added to buf_pool->LRU and buf_pool->page_hash. */
		data = buf_buddy_alloc(buf_pool, page_size.physical(), &lru);
		buf_page_mutex_enter(block);
		block->page.zip.data = (page_zip_t*) data;

		/* To maintain the invariant
		block->in_unzip_LRU_list
		== buf_page_belongs_to_unzip_LRU(&block->page)
		we have to add this block to unzip_LRU after
		block->page.zip.data is set. */
		ut_ad(buf_page_belongs_to_unzip_LRU(&block->page));
		buf_unzip_LRU_add_block(block, FALSE);

		buf_page_set_io_fix(&block->page, BUF_IO_NONE);
		rw_lock_x_unlock(&block->lock);
	}

	buf_pool_mutex_exit(buf_pool);

	mtr_memo_push(mtr, block, MTR_MEMO_BUF_FIX);

	buf_page_set_accessed(&block->page);

	buf_page_mutex_exit(block);

	/* Delete possible entries for the page from the insert buffer:
	such can exist if the page belonged to an index which was dropped */
	if (!recv_recovery_is_on()) {
		ibuf_merge_or_delete_for_page(NULL, page_id, &page_size, TRUE);
	}

	frame = block->frame;

	memset(frame + FIL_PAGE_PREV, 0xff, 4);
	memset(frame + FIL_PAGE_NEXT, 0xff, 4);
	mach_write_to_2(frame + FIL_PAGE_TYPE, FIL_PAGE_TYPE_ALLOCATED);

	/* FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION is only used on the
	following pages:
	(1) The first page of the InnoDB system tablespace (page 0:0)
	(2) FIL_RTREE_SPLIT_SEQ_NUM on R-tree pages
	(3) key_version on encrypted pages (not page 0:0) */

	memset(frame + FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION, 0, 8);
	memset(frame + FIL_PAGE_LSN, 0, 8);

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
	ut_a(++buf_dbg_counter % 5771 || buf_validate());
#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */
	return(block);
}

/********************************************************************//**
Monitor the buffer page read/write activity, and increment corresponding
counter value if MONITOR_MODULE_BUF_PAGE (module_buf_page) module is
enabled. */
static
void
buf_page_monitor(
/*=============*/
	const buf_page_t*	bpage,	/*!< in: pointer to the block */
	enum buf_io_fix		io_type)/*!< in: io_fix types */
{
	const byte*	frame;
	monitor_id_t	counter;

	/* If the counter module is not turned on, just return */
	if (!MONITOR_IS_ON(MONITOR_MODULE_BUF_PAGE)) {
		return;
	}

	ut_a(io_type == BUF_IO_READ || io_type == BUF_IO_WRITE);

	frame = bpage->zip.data
		? bpage->zip.data
		: ((buf_block_t*) bpage)->frame;

	switch (fil_page_get_type(frame)) {
		ulint	level;

	case FIL_PAGE_INDEX:
	case FIL_PAGE_RTREE:
		level = btr_page_get_level_low(frame);

		/* Check if it is an index page for insert buffer */
		if (btr_page_get_index_id(frame)
		    == (index_id_t)(DICT_IBUF_ID_MIN + IBUF_SPACE_ID)) {
			if (level == 0) {
				counter = MONITOR_RW_COUNTER(
					io_type, MONITOR_INDEX_IBUF_LEAF_PAGE);
			} else {
				counter = MONITOR_RW_COUNTER(
					io_type,
					MONITOR_INDEX_IBUF_NON_LEAF_PAGE);
			}
		} else {
			if (level == 0) {
				counter = MONITOR_RW_COUNTER(
					io_type, MONITOR_INDEX_LEAF_PAGE);
			} else {
				counter = MONITOR_RW_COUNTER(
					io_type, MONITOR_INDEX_NON_LEAF_PAGE);
			}
		}
		break;

	case FIL_PAGE_UNDO_LOG:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_UNDO_LOG_PAGE);
		break;

	case FIL_PAGE_INODE:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_INODE_PAGE);
		break;

	case FIL_PAGE_IBUF_FREE_LIST:
		counter = MONITOR_RW_COUNTER(io_type,
					     MONITOR_IBUF_FREELIST_PAGE);
		break;

	case FIL_PAGE_IBUF_BITMAP:
		counter = MONITOR_RW_COUNTER(io_type,
					     MONITOR_IBUF_BITMAP_PAGE);
		break;

	case FIL_PAGE_TYPE_SYS:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_SYSTEM_PAGE);
		break;

	case FIL_PAGE_TYPE_TRX_SYS:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_TRX_SYSTEM_PAGE);
		break;

	case FIL_PAGE_TYPE_FSP_HDR:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_FSP_HDR_PAGE);
		break;

	case FIL_PAGE_TYPE_XDES:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_XDES_PAGE);
		break;

	case FIL_PAGE_TYPE_BLOB:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_BLOB_PAGE);
		break;

	case FIL_PAGE_TYPE_ZBLOB:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_ZBLOB_PAGE);
		break;

	case FIL_PAGE_TYPE_ZBLOB2:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_ZBLOB2_PAGE);
		break;

	default:
		counter = MONITOR_RW_COUNTER(io_type, MONITOR_OTHER_PAGE);
	}

	MONITOR_INC_NOCHECK(counter);
}

/** Mark a table corrupted.
@param[in]	bpage	corrupted page
@param[in]	space	tablespace of the corrupted page */
ATTRIBUTE_COLD
static void buf_mark_space_corrupt(buf_page_t* bpage, const fil_space_t& space)
{
	/* If block is not encrypted find the table with specified
	space id, and mark it corrupted. Encrypted tables
	are marked unusable later e.g. in ::open(). */
	if (!space.crypt_data
	    || space.crypt_data->type == CRYPT_SCHEME_UNENCRYPTED) {
		dict_set_corrupted_by_space(bpage->id.space());
	} else {
		dict_set_encrypted_by_space(bpage->id.space());
	}
}

/** Mark a table corrupted.
@param[in]	bpage	Corrupted page
@param[in]	space	Corrupted page belongs to tablespace
Also remove the bpage from LRU list. */
static
void
buf_corrupt_page_release(buf_page_t* bpage, const fil_space_t* space)
{
	buf_pool_t*	buf_pool = buf_pool_from_bpage(bpage);
	const ibool	uncompressed = (buf_page_get_state(bpage)
					== BUF_BLOCK_FILE_PAGE);
	page_id_t	old_page_id = bpage->id;

	/* First unfix and release lock on the bpage */
	buf_pool_mutex_enter(buf_pool);
	mutex_enter(buf_page_get_mutex(bpage));
	ut_ad(buf_page_get_io_fix(bpage) == BUF_IO_READ);
	ut_ad(bpage->id.space() == space->id);

	/* buf_fix_count can be greater than zero. Because other thread
	can wait in buf_page_wait_read() for the page to be read. */

	bpage->id.set_corrupt_id();
	/* Set BUF_IO_NONE before we remove the block from LRU list */
	buf_page_set_io_fix(bpage, BUF_IO_NONE);

	if (uncompressed) {
		rw_lock_x_unlock_gen(
			&((buf_block_t*) bpage)->lock,
			BUF_IO_READ);
	}

	mutex_exit(buf_page_get_mutex(bpage));

	if (!srv_force_recovery) {
		buf_mark_space_corrupt(bpage, *space);
	}

	/* After this point bpage can't be referenced. */
	buf_LRU_free_one_page(bpage, old_page_id);

	ut_ad(buf_pool->n_pend_reads > 0);
	buf_pool->n_pend_reads--;

	buf_pool_mutex_exit(buf_pool);
}

/** Check if page is maybe compressed, encrypted or both when we encounter
corrupted page. Note that we can't be 100% sure if page is corrupted
or decrypt/decompress just failed.
@param[in,out]	bpage		page
@param[in,out]	space		tablespace from fil_space_acquire_for_io()
@return	whether the operation succeeded
@retval	DB_SUCCESS		if page has been read and is not corrupted
@retval	DB_PAGE_CORRUPTED	if page based on checksum check is corrupted
@retval	DB_DECRYPTION_FAILED	if page post encryption checksum matches but
after decryption normal page checksum does not match.
@retval	DB_TABLESPACE_DELETED	if accessed tablespace is not found */
static dberr_t buf_page_check_corrupt(buf_page_t* bpage, fil_space_t* space)
{
	ut_ad(space->n_pending_ios > 0);

	byte* dst_frame = (bpage->zip.data) ? bpage->zip.data :
		((buf_block_t*) bpage)->frame;
	dberr_t err = DB_SUCCESS;

	/* In buf_decrypt_after_read we have either decrypted the page if
	page post encryption checksum matches and used key_id is found
	from the encryption plugin. If checksum did not match page was
	not decrypted and it could be either encrypted and corrupted
	or corrupted or good page. If we decrypted, there page could
	still be corrupted if used key does not match. */
	const bool seems_encrypted = mach_read_from_4(
		dst_frame + FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION)
		&& space->crypt_data
		&& space->crypt_data->type != CRYPT_SCHEME_UNENCRYPTED;

	/* If traditional checksums match, we assume that page is
	not anymore encrypted. */
	if (buf_page_is_corrupted(
		true, dst_frame, bpage->size, space)) {
		err = DB_PAGE_CORRUPTED;
	}

	if (seems_encrypted && err == DB_PAGE_CORRUPTED
	    && bpage->id.page_no() != 0) {
		err = DB_DECRYPTION_FAILED;

		ib::error()
			<< "The page " << bpage->id << " in file '"
			<< space->chain.start->name
			<< "' cannot be decrypted.";

		ib::info()
			<< "However key management plugin or used key_version "
			<< mach_read_from_4(dst_frame
					    + FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION)
			<< " is not found or"
			" used encryption algorithm or method does not match.";

		if (bpage->id.space() != TRX_SYS_SPACE) {
			ib::info()
				<< "Marking tablespace as missing."
				" You may drop this table or"
				" install correct key management plugin"
				" and key file.";
		}
	}

	return (err);
}

/** Complete a read or write request of a file page to or from the buffer pool.
@param[in,out]	bpage	page to complete
@param[in]	dblwr	whether the doublewrite buffer was used (on write)
@param[in]	evict	whether or not to evict the page from LRU list
@return whether the operation succeeded
@retval	DB_SUCCESS		always when writing, or if a read page was OK
@retval	DB_TABLESPACE_DELETED	if the tablespace does not exist
@retval	DB_PAGE_CORRUPTED	if the checksum fails on a page read
@retval	DB_DECRYPTION_FAILED	if page post encryption checksum matches but
				after decryption normal page checksum does
				not match */
UNIV_INTERN
dberr_t
buf_page_io_complete(buf_page_t* bpage, bool dblwr, bool evict)
{
	enum buf_io_fix	io_type;
	buf_pool_t*	buf_pool = buf_pool_from_bpage(bpage);
	const bool	uncompressed = (buf_page_get_state(bpage)
					== BUF_BLOCK_FILE_PAGE);
	ut_a(buf_page_in_file(bpage));

	/* We do not need protect io_fix here by mutex to read
	it because this is the only function where we can change the value
	from BUF_IO_READ or BUF_IO_WRITE to some other value, and our code
	ensures that this is the only thread that handles the i/o for this
	block. */

	io_type = buf_page_get_io_fix(bpage);
	ut_ad(io_type == BUF_IO_READ || io_type == BUF_IO_WRITE);
	ut_ad(bpage->size.is_compressed() == (bpage->zip.data != NULL));
	ut_ad(uncompressed || bpage->zip.data);

	if (io_type == BUF_IO_READ) {
		ulint	read_page_no = 0;
		ulint	read_space_id = 0;
		byte*	frame = bpage->zip.data
			? bpage->zip.data
			: reinterpret_cast<buf_block_t*>(bpage)->frame;
		ut_ad(frame);
		fil_space_t* space = fil_space_acquire_for_io(
			bpage->id.space());
		if (!space) {
			return DB_TABLESPACE_DELETED;
		}

		dberr_t	err;

		if (!buf_page_decrypt_after_read(bpage, space)) {
			err = DB_DECRYPTION_FAILED;
			goto database_corrupted;
		}

		if (bpage->zip.data && uncompressed) {
			my_atomic_addlint(&buf_pool->n_pend_unzip, 1);
			ibool ok = buf_zip_decompress((buf_block_t*) bpage,
						      FALSE);
			my_atomic_addlint(&buf_pool->n_pend_unzip, -1);

			if (!ok) {
				ib::info() << "Page "
					   << bpage->id
					   << " zip_decompress failure.";

				err = DB_PAGE_CORRUPTED;
				goto database_corrupted;
			}
		}

		/* If this page is not uninitialized and not in the
		doublewrite buffer, then the page number and space id
		should be the same as in block. */
		read_page_no = mach_read_from_4(frame + FIL_PAGE_OFFSET);
		read_space_id = mach_read_from_4(
			frame + FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID);

		if (bpage->id.space() == TRX_SYS_SPACE
		    && buf_dblwr_page_inside(bpage->id.page_no())) {

			ib::error() << "Reading page " << bpage->id
				<< ", which is in the doublewrite buffer!";

		} else if (read_space_id == 0 && read_page_no == 0) {
			/* This is likely an uninitialized page. */
		} else if ((bpage->id.space() != TRX_SYS_SPACE
			    && bpage->id.space() != read_space_id)
			   || bpage->id.page_no() != read_page_no) {
			/* We did not compare space_id to read_space_id
			in the system tablespace, because the field
			was written as garbage before MySQL 4.1.1,
			which did not support innodb_file_per_table. */

			ib::error() << "Space id and page no stored in "
				"the page, read in are "
				<< page_id_t(read_space_id, read_page_no)
				<< ", should be " << bpage->id;
		}

		err = buf_page_check_corrupt(bpage, space);

		if (err != DB_SUCCESS) {
database_corrupted:
			/* Not a real corruption if it was triggered by
			error injection */
			DBUG_EXECUTE_IF(
				"buf_page_import_corrupt_failure",
				if (!is_predefined_tablespace(
					    bpage->id.space())) {
					buf_corrupt_page_release(bpage, space);
					ib::info() << "Simulated IMPORT "
						"corruption";
					fil_space_release_for_io(space);
					return(err);
				}
				err = DB_SUCCESS;
				goto page_not_corrupt;
			);

			if (uncompressed && bpage->zip.data) {
				memset(reinterpret_cast<buf_block_t*>(bpage)
				       ->frame, 0, srv_page_size);
			}

			if (err == DB_PAGE_CORRUPTED) {
				ib::error()
					<< "Database page corruption on disk"
					" or a failed file read of tablespace "
					<< space->name << " page " << bpage->id
					<< ". You may have to recover from "
					<< "a backup.";

				buf_page_print(frame, bpage->size);

				ib::info()
					<< "It is also possible that your"
					" operating system has corrupted"
					" its own file cache and rebooting"
					" your computer removes the error."
					" If the corrupt page is an index page."
					" You can also try to fix the"
					" corruption by dumping, dropping,"
					" and reimporting the corrupt table."
					" You can use CHECK TABLE to scan"
					" your table for corruption. "
					<< FORCE_RECOVERY_MSG;
			}

			if (!srv_force_recovery) {

				/* If page space id is larger than TRX_SYS_SPACE
				(0), we will attempt to mark the corresponding
				table as corrupted instead of crashing server */
				if (bpage->id.space() == TRX_SYS_SPACE) {
					ib::fatal() << "Aborting because of"
						" a corrupt database page.";
				}

				buf_corrupt_page_release(bpage, space);
				fil_space_release_for_io(space);
				return(err);
			}
		}

		DBUG_EXECUTE_IF("buf_page_import_corrupt_failure",
				page_not_corrupt: bpage = bpage; );

		if (err == DB_PAGE_CORRUPTED
		    || err == DB_DECRYPTION_FAILED) {
			const page_id_t corrupt_page_id = bpage->id;

			buf_corrupt_page_release(bpage, space);

			if (recv_recovery_is_on()) {
				recv_recover_corrupt_page(corrupt_page_id);
			}

			fil_space_release_for_io(space);
			return err;
		}

		if (recv_recovery_is_on()) {
			recv_recover_page(bpage);
		}

		/* If space is being truncated then avoid ibuf operation.
		During re-init we have already freed ibuf entries. */
		if (uncompressed
		    && !recv_no_ibuf_operations
		    && (bpage->id.space() == 0
			|| !is_predefined_tablespace(bpage->id.space()))
		    && !srv_is_tablespace_truncated(bpage->id.space())
		    && fil_page_get_type(frame) == FIL_PAGE_INDEX
		    && page_is_leaf(frame)) {

			ibuf_merge_or_delete_for_page(
				(buf_block_t*) bpage, bpage->id,
				&bpage->size, TRUE);
		}

		fil_space_release_for_io(space);
	} else {
		/* io_type == BUF_IO_WRITE */
		if (bpage->slot) {
			/* Mark slot free */
			bpage->slot->release();
			bpage->slot = NULL;
		}
	}

	BPageMutex* block_mutex = buf_page_get_mutex(bpage);
	buf_pool_mutex_enter(buf_pool);
	mutex_enter(block_mutex);

	/* Because this thread which does the unlocking is not the same that
	did the locking, we use a pass value != 0 in unlock, which simply
	removes the newest lock debug record, without checking the thread
	id. */

	buf_page_set_io_fix(bpage, BUF_IO_NONE);
	buf_page_monitor(bpage, io_type);

	if (io_type == BUF_IO_READ) {
		/* NOTE that the call to ibuf may have moved the ownership of
		the x-latch to this OS thread: do not let this confuse you in
		debugging! */

		ut_ad(buf_pool->n_pend_reads > 0);
		buf_pool->n_pend_reads--;
		buf_pool->stat.n_pages_read++;

		if (uncompressed) {
			rw_lock_x_unlock_gen(&((buf_block_t*) bpage)->lock,
					     BUF_IO_READ);
		}

		mutex_exit(block_mutex);
	} else {
		/* Write means a flush operation: call the completion
		routine in the flush system */

		buf_flush_write_complete(bpage, dblwr);

		if (uncompressed) {
			rw_lock_sx_unlock_gen(&((buf_block_t*) bpage)->lock,
					      BUF_IO_WRITE);
		}

		buf_pool->stat.n_pages_written++;

		/* We decide whether or not to evict the page from the
		LRU list based on the flush_type.
		* BUF_FLUSH_LIST: don't evict
		* BUF_FLUSH_LRU: always evict
		* BUF_FLUSH_SINGLE_PAGE: eviction preference is passed
		by the caller explicitly. */
		if (buf_page_get_flush_type(bpage) == BUF_FLUSH_LRU) {
			evict = true;
		}

		mutex_exit(block_mutex);

		if (evict) {
			buf_LRU_free_page(bpage, true);
		}
	}

	DBUG_PRINT("ib_buf", ("%s page %u:%u",
			      io_type == BUF_IO_READ ? "read" : "wrote",
			      bpage->id.space(), bpage->id.page_no()));

	buf_pool_mutex_exit(buf_pool);

	return DB_SUCCESS;
}

/*********************************************************************//**
Asserts that all file pages in the buffer are in a replaceable state.
@return TRUE */
static
ibool
buf_all_freed_instance(
/*===================*/
	buf_pool_t*	buf_pool)	/*!< in: buffer pool instancce */
{
	ulint		i;
	buf_chunk_t*	chunk;

	ut_ad(buf_pool);

	buf_pool_mutex_enter(buf_pool);

	chunk = buf_pool->chunks;

	for (i = buf_pool->n_chunks; i--; chunk++) {

		if (const buf_block_t* block = buf_chunk_not_freed(chunk)) {
			ib::fatal() << "Page " << block->page.id
				<< " still fixed or dirty";
		}
	}

	buf_pool_mutex_exit(buf_pool);

	return(TRUE);
}

/** Refreshes the statistics used to print per-second averages.
@param[in,out]	buf_pool	buffer pool instance */
static
void
buf_refresh_io_stats(
	buf_pool_t*	buf_pool)
{
	buf_pool->last_printout_time = time(NULL);
	buf_pool->old_stat = buf_pool->stat;
}

/*********************************************************************//**
Invalidates file pages in one buffer pool instance */
static
void
buf_pool_invalidate_instance(
/*=========================*/
	buf_pool_t*	buf_pool)	/*!< in: buffer pool instance */
{
	ulint		i;

	buf_pool_mutex_enter(buf_pool);

	for (i = BUF_FLUSH_LRU; i < BUF_FLUSH_N_TYPES; i++) {

		/* As this function is called during startup and
		during redo application phase during recovery, InnoDB
		is single threaded (apart from IO helper threads) at
		this stage. No new write batch can be in intialization
		stage at this point. */
		ut_ad(buf_pool->init_flush[i] == FALSE);

		/* However, it is possible that a write batch that has
		been posted earlier is still not complete. For buffer
		pool invalidation to proceed we must ensure there is NO
		write activity happening. */
		if (buf_pool->n_flush[i] > 0) {
			buf_flush_t	type = static_cast<buf_flush_t>(i);

			buf_pool_mutex_exit(buf_pool);
			buf_flush_wait_batch_end(buf_pool, type);
			buf_pool_mutex_enter(buf_pool);
		}
	}

	buf_pool_mutex_exit(buf_pool);

	ut_ad(buf_all_freed_instance(buf_pool));

	buf_pool_mutex_enter(buf_pool);

	while (buf_LRU_scan_and_free_block(buf_pool, true)) {
	}

	ut_ad(UT_LIST_GET_LEN(buf_pool->LRU) == 0);
	ut_ad(UT_LIST_GET_LEN(buf_pool->unzip_LRU) == 0);

	buf_pool->freed_page_clock = 0;
	buf_pool->LRU_old = NULL;
	buf_pool->LRU_old_len = 0;

	memset(&buf_pool->stat, 0x00, sizeof(buf_pool->stat));
	buf_refresh_io_stats(buf_pool);

	buf_pool_mutex_exit(buf_pool);
}

/*********************************************************************//**
Invalidates the file pages in the buffer pool when an archive recovery is
completed. All the file pages buffered must be in a replaceable state when
this function is called: not latched and not modified. */
void
buf_pool_invalidate(void)
/*=====================*/
{
	ulint   i;

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_invalidate_instance(buf_pool_from_array(i));
	}
}

#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
/*********************************************************************//**
Validates data in one buffer pool instance
@return TRUE */
static
ibool
buf_pool_validate_instance(
/*=======================*/
	buf_pool_t*	buf_pool)	/*!< in: buffer pool instance */
{
	buf_page_t*	b;
	buf_chunk_t*	chunk;
	ulint		i;
	ulint		n_lru_flush	= 0;
	ulint		n_page_flush	= 0;
	ulint		n_list_flush	= 0;
	ulint		n_lru		= 0;
	ulint		n_flush		= 0;
	ulint		n_free		= 0;
	ulint		n_zip		= 0;

	ut_ad(buf_pool);

	buf_pool_mutex_enter(buf_pool);
	hash_lock_x_all(buf_pool->page_hash);

	chunk = buf_pool->chunks;

	/* Check the uncompressed blocks. */

	for (i = buf_pool->n_chunks; i--; chunk++) {

		ulint		j;
		buf_block_t*	block = chunk->blocks;

		for (j = chunk->size; j--; block++) {

			buf_page_mutex_enter(block);

			switch (buf_block_get_state(block)) {
			case BUF_BLOCK_POOL_WATCH:
			case BUF_BLOCK_ZIP_PAGE:
			case BUF_BLOCK_ZIP_DIRTY:
				/* These should only occur on
				zip_clean, zip_free[], or flush_list. */
				ut_error;
				break;

			case BUF_BLOCK_FILE_PAGE:
				ut_a(buf_page_hash_get_low(
						buf_pool, block->page.id)
				     == &block->page);

				switch (buf_page_get_io_fix(&block->page)) {
				case BUF_IO_NONE:
					break;

				case BUF_IO_WRITE:
					switch (buf_page_get_flush_type(
							&block->page)) {
					case BUF_FLUSH_LRU:
						n_lru_flush++;
						goto assert_s_latched;
					case BUF_FLUSH_SINGLE_PAGE:
						n_page_flush++;
assert_s_latched:
						ut_a(rw_lock_is_locked(
							     &block->lock,
								     RW_LOCK_S)
						     || rw_lock_is_locked(
								&block->lock,
								RW_LOCK_SX));
						break;
					case BUF_FLUSH_LIST:
						n_list_flush++;
						break;
					default:
						ut_error;
					}

					break;

				case BUF_IO_READ:

					ut_a(rw_lock_is_locked(&block->lock,
							       RW_LOCK_X));
					break;

				case BUF_IO_PIN:
					break;
				}

				n_lru++;
				break;

			case BUF_BLOCK_NOT_USED:
				n_free++;
				break;

			case BUF_BLOCK_READY_FOR_USE:
			case BUF_BLOCK_MEMORY:
			case BUF_BLOCK_REMOVE_HASH:
				/* do nothing */
				break;
			}

			buf_page_mutex_exit(block);
		}
	}

	mutex_enter(&buf_pool->zip_mutex);

	/* Check clean compressed-only blocks. */

	for (b = UT_LIST_GET_FIRST(buf_pool->zip_clean); b;
	     b = UT_LIST_GET_NEXT(list, b)) {
		ut_a(buf_page_get_state(b) == BUF_BLOCK_ZIP_PAGE);
		switch (buf_page_get_io_fix(b)) {
		case BUF_IO_NONE:
		case BUF_IO_PIN:
			/* All clean blocks should be I/O-unfixed. */
			break;
		case BUF_IO_READ:
			/* In buf_LRU_free_page(), we temporarily set
			b->io_fix = BUF_IO_READ for a newly allocated
			control block in order to prevent
			buf_page_get_gen() from decompressing the block. */
			break;
		default:
			ut_error;
			break;
		}

		/* It is OK to read oldest_modification here because
		we have acquired buf_pool->zip_mutex above which acts
		as the 'block->mutex' for these bpages. */
		ut_a(!b->oldest_modification);
		ut_a(buf_page_hash_get_low(buf_pool, b->id) == b);
		n_lru++;
		n_zip++;
	}

	/* Check dirty blocks. */

	buf_flush_list_mutex_enter(buf_pool);
	for (b = UT_LIST_GET_FIRST(buf_pool->flush_list); b;
	     b = UT_LIST_GET_NEXT(list, b)) {
		ut_ad(b->in_flush_list);
		ut_a(b->oldest_modification);
		n_flush++;

		switch (buf_page_get_state(b)) {
		case BUF_BLOCK_ZIP_DIRTY:
			n_lru++;
			n_zip++;
			switch (buf_page_get_io_fix(b)) {
			case BUF_IO_NONE:
			case BUF_IO_READ:
			case BUF_IO_PIN:
				break;
			case BUF_IO_WRITE:
				switch (buf_page_get_flush_type(b)) {
				case BUF_FLUSH_LRU:
					n_lru_flush++;
					break;
				case BUF_FLUSH_SINGLE_PAGE:
					n_page_flush++;
					break;
				case BUF_FLUSH_LIST:
					n_list_flush++;
					break;
				default:
					ut_error;
				}
				break;
			}
			break;
		case BUF_BLOCK_FILE_PAGE:
			/* uncompressed page */
			break;
		case BUF_BLOCK_POOL_WATCH:
		case BUF_BLOCK_ZIP_PAGE:
		case BUF_BLOCK_NOT_USED:
		case BUF_BLOCK_READY_FOR_USE:
		case BUF_BLOCK_MEMORY:
		case BUF_BLOCK_REMOVE_HASH:
			ut_error;
			break;
		}
		ut_a(buf_page_hash_get_low(buf_pool, b->id) == b);
	}

	ut_a(UT_LIST_GET_LEN(buf_pool->flush_list) == n_flush);

	hash_unlock_x_all(buf_pool->page_hash);
	buf_flush_list_mutex_exit(buf_pool);

	mutex_exit(&buf_pool->zip_mutex);

	if (buf_pool->curr_size == buf_pool->old_size
	    && n_lru + n_free > buf_pool->curr_size + n_zip) {

		ib::fatal() << "n_LRU " << n_lru << ", n_free " << n_free
			<< ", pool " << buf_pool->curr_size
			<< " zip " << n_zip << ". Aborting...";
	}

	ut_a(UT_LIST_GET_LEN(buf_pool->LRU) == n_lru);
	if (buf_pool->curr_size == buf_pool->old_size
	    && UT_LIST_GET_LEN(buf_pool->free) != n_free) {

		ib::fatal() << "Free list len "
			<< UT_LIST_GET_LEN(buf_pool->free)
			<< ", free blocks " << n_free << ". Aborting...";
	}

	ut_a(buf_pool->n_flush[BUF_FLUSH_LIST] == n_list_flush);
	ut_a(buf_pool->n_flush[BUF_FLUSH_LRU] == n_lru_flush);
	ut_a(buf_pool->n_flush[BUF_FLUSH_SINGLE_PAGE] == n_page_flush);

	buf_pool_mutex_exit(buf_pool);

	ut_a(buf_LRU_validate());
	ut_a(buf_flush_validate(buf_pool));

	return(TRUE);
}

/*********************************************************************//**
Validates the buffer buf_pool data structure.
@return TRUE */
ibool
buf_validate(void)
/*==============*/
{
	ulint	i;

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		buf_pool_validate_instance(buf_pool);
	}
	return(TRUE);
}

#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */

#if defined UNIV_DEBUG_PRINT || defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
/*********************************************************************//**
Prints info of the buffer buf_pool data structure for one instance. */
static
void
buf_print_instance(
/*===============*/
	buf_pool_t*	buf_pool)
{
	index_id_t*	index_ids;
	ulint*		counts;
	ulint		size;
	ulint		i;
	ulint		j;
	index_id_t	id;
	ulint		n_found;
	buf_chunk_t*	chunk;
	dict_index_t*	index;

	ut_ad(buf_pool);

	size = buf_pool->curr_size;

	index_ids = static_cast<index_id_t*>(
		ut_malloc_nokey(size * sizeof *index_ids));

	counts = static_cast<ulint*>(ut_malloc_nokey(sizeof(ulint) * size));

	buf_pool_mutex_enter(buf_pool);
	buf_flush_list_mutex_enter(buf_pool);

	ib::info() << *buf_pool;

	buf_flush_list_mutex_exit(buf_pool);

	/* Count the number of blocks belonging to each index in the buffer */

	n_found = 0;

	chunk = buf_pool->chunks;

	for (i = buf_pool->n_chunks; i--; chunk++) {
		buf_block_t*	block		= chunk->blocks;
		ulint		n_blocks	= chunk->size;

		for (; n_blocks--; block++) {
			const buf_frame_t* frame = block->frame;

			if (fil_page_index_page_check(frame)) {

				id = btr_page_get_index_id(frame);

				/* Look for the id in the index_ids array */
				j = 0;

				while (j < n_found) {

					if (index_ids[j] == id) {
						counts[j]++;

						break;
					}
					j++;
				}

				if (j == n_found) {
					n_found++;
					index_ids[j] = id;
					counts[j] = 1;
				}
			}
		}
	}

	buf_pool_mutex_exit(buf_pool);

	for (i = 0; i < n_found; i++) {
		index = dict_index_get_if_in_cache(index_ids[i]);

		if (!index) {
			ib::info() << "Block count for index "
				<< index_ids[i] << " in buffer is about "
				<< counts[i];
		} else {
			ib::info() << "Block count for index " << index_ids[i]
				<< " in buffer is about " << counts[i]
				<< ", index " << index->name
				<< " of table " << index->table->name;
		}
	}

	ut_free(index_ids);
	ut_free(counts);

	ut_a(buf_pool_validate_instance(buf_pool));
}

/*********************************************************************//**
Prints info of the buffer buf_pool data structure. */
void
buf_print(void)
/*===========*/
{
	ulint   i;

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);
		buf_print_instance(buf_pool);
	}
}
#endif /* UNIV_DEBUG_PRINT || UNIV_DEBUG || UNIV_BUF_DEBUG */

#ifdef UNIV_DEBUG
/*********************************************************************//**
Returns the number of latched pages in the buffer pool.
@return number of latched pages */
static
ulint
buf_get_latched_pages_number_instance(
/*==================================*/
	buf_pool_t*	buf_pool)	/*!< in: buffer pool instance */
{
	buf_page_t*	b;
	ulint		i;
	buf_chunk_t*	chunk;
	ulint		fixed_pages_number = 0;

	buf_pool_mutex_enter(buf_pool);

	chunk = buf_pool->chunks;

	for (i = buf_pool->n_chunks; i--; chunk++) {
		buf_block_t*	block;
		ulint		j;

		block = chunk->blocks;

		for (j = chunk->size; j--; block++) {
			if (buf_block_get_state(block)
			    != BUF_BLOCK_FILE_PAGE) {

				continue;
			}

			buf_page_mutex_enter(block);

			if (block->page.buf_fix_count != 0
			    || buf_page_get_io_fix(&block->page)
			    != BUF_IO_NONE) {
				fixed_pages_number++;
			}

			buf_page_mutex_exit(block);
		}
	}

	mutex_enter(&buf_pool->zip_mutex);

	/* Traverse the lists of clean and dirty compressed-only blocks. */

	for (b = UT_LIST_GET_FIRST(buf_pool->zip_clean); b;
	     b = UT_LIST_GET_NEXT(list, b)) {
		ut_a(buf_page_get_state(b) == BUF_BLOCK_ZIP_PAGE);
		ut_a(buf_page_get_io_fix(b) != BUF_IO_WRITE);

		if (b->buf_fix_count != 0
		    || buf_page_get_io_fix(b) != BUF_IO_NONE) {
			fixed_pages_number++;
		}
	}

	buf_flush_list_mutex_enter(buf_pool);
	for (b = UT_LIST_GET_FIRST(buf_pool->flush_list); b;
	     b = UT_LIST_GET_NEXT(list, b)) {
		ut_ad(b->in_flush_list);

		switch (buf_page_get_state(b)) {
		case BUF_BLOCK_ZIP_DIRTY:
			if (b->buf_fix_count != 0
			    || buf_page_get_io_fix(b) != BUF_IO_NONE) {
				fixed_pages_number++;
			}
			break;
		case BUF_BLOCK_FILE_PAGE:
			/* uncompressed page */
			break;
		case BUF_BLOCK_POOL_WATCH:
		case BUF_BLOCK_ZIP_PAGE:
		case BUF_BLOCK_NOT_USED:
		case BUF_BLOCK_READY_FOR_USE:
		case BUF_BLOCK_MEMORY:
		case BUF_BLOCK_REMOVE_HASH:
			ut_error;
			break;
		}
	}

	buf_flush_list_mutex_exit(buf_pool);
	mutex_exit(&buf_pool->zip_mutex);
	buf_pool_mutex_exit(buf_pool);

	return(fixed_pages_number);
}

/*********************************************************************//**
Returns the number of latched pages in all the buffer pools.
@return number of latched pages */
ulint
buf_get_latched_pages_number(void)
/*==============================*/
{
	ulint	i;
	ulint	total_latched_pages = 0;

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		total_latched_pages += buf_get_latched_pages_number_instance(
			buf_pool);
	}

	return(total_latched_pages);
}

#endif /* UNIV_DEBUG */

/*********************************************************************//**
Returns the number of pending buf pool read ios.
@return number of pending read I/O operations */
ulint
buf_get_n_pending_read_ios(void)
/*============================*/
{
	ulint	pend_ios = 0;

	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		pend_ios += buf_pool_from_array(i)->n_pend_reads;
	}

	return(pend_ios);
}

/*********************************************************************//**
Returns the ratio in percents of modified pages in the buffer pool /
database pages in the buffer pool.
@return modified page percentage ratio */
double
buf_get_modified_ratio_pct(void)
/*============================*/
{
	double		ratio;
	ulint		lru_len = 0;
	ulint		free_len = 0;
	ulint		flush_list_len = 0;

	buf_get_total_list_len(&lru_len, &free_len, &flush_list_len);

	ratio = static_cast<double>(100 * flush_list_len)
		/ (1 + lru_len + free_len);

	/* 1 + is there to avoid division by zero */

	return(ratio);
}

/*******************************************************************//**
Aggregates a pool stats information with the total buffer pool stats  */
static
void
buf_stats_aggregate_pool_info(
/*==========================*/
	buf_pool_info_t*	total_info,	/*!< in/out: the buffer pool
						info to store aggregated
						result */
	const buf_pool_info_t*	pool_info)	/*!< in: individual buffer pool
						stats info */
{
	ut_a(total_info && pool_info);

	/* Nothing to copy if total_info is the same as pool_info */
	if (total_info == pool_info) {
		return;
	}

	total_info->pool_size += pool_info->pool_size;
	total_info->lru_len += pool_info->lru_len;
	total_info->old_lru_len += pool_info->old_lru_len;
	total_info->free_list_len += pool_info->free_list_len;
	total_info->flush_list_len += pool_info->flush_list_len;
	total_info->n_pend_unzip += pool_info->n_pend_unzip;
	total_info->n_pend_reads += pool_info->n_pend_reads;
	total_info->n_pending_flush_lru += pool_info->n_pending_flush_lru;
	total_info->n_pending_flush_list += pool_info->n_pending_flush_list;
	total_info->n_pages_made_young += pool_info->n_pages_made_young;
	total_info->n_pages_not_made_young += pool_info->n_pages_not_made_young;
	total_info->n_pages_read += pool_info->n_pages_read;
	total_info->n_pages_created += pool_info->n_pages_created;
	total_info->n_pages_written += pool_info->n_pages_written;
	total_info->n_page_gets += pool_info->n_page_gets;
	total_info->n_ra_pages_read_rnd += pool_info->n_ra_pages_read_rnd;
	total_info->n_ra_pages_read += pool_info->n_ra_pages_read;
	total_info->n_ra_pages_evicted += pool_info->n_ra_pages_evicted;
	total_info->page_made_young_rate += pool_info->page_made_young_rate;
	total_info->page_not_made_young_rate +=
		pool_info->page_not_made_young_rate;
	total_info->pages_read_rate += pool_info->pages_read_rate;
	total_info->pages_created_rate += pool_info->pages_created_rate;
	total_info->pages_written_rate += pool_info->pages_written_rate;
	total_info->n_page_get_delta += pool_info->n_page_get_delta;
	total_info->page_read_delta += pool_info->page_read_delta;
	total_info->young_making_delta += pool_info->young_making_delta;
	total_info->not_young_making_delta += pool_info->not_young_making_delta;
	total_info->pages_readahead_rnd_rate += pool_info->pages_readahead_rnd_rate;
	total_info->pages_readahead_rate += pool_info->pages_readahead_rate;
	total_info->pages_evicted_rate += pool_info->pages_evicted_rate;
	total_info->unzip_lru_len += pool_info->unzip_lru_len;
	total_info->io_sum += pool_info->io_sum;
	total_info->io_cur += pool_info->io_cur;
	total_info->unzip_sum += pool_info->unzip_sum;
	total_info->unzip_cur += pool_info->unzip_cur;
}
/*******************************************************************//**
Collect buffer pool stats information for a buffer pool. Also
record aggregated stats if there are more than one buffer pool
in the server */
void
buf_stats_get_pool_info(
/*====================*/
	buf_pool_t*		buf_pool,	/*!< in: buffer pool */
	ulint			pool_id,	/*!< in: buffer pool ID */
	buf_pool_info_t*	all_pool_info)	/*!< in/out: buffer pool info
						to fill */
{
	buf_pool_info_t*	pool_info;
	time_t			current_time;
	double			time_elapsed;

	/* Find appropriate pool_info to store stats for this buffer pool */
	pool_info = &all_pool_info[pool_id];

	buf_pool_mutex_enter(buf_pool);
	buf_flush_list_mutex_enter(buf_pool);

	pool_info->pool_unique_id = pool_id;

	pool_info->pool_size = buf_pool->curr_size;

	pool_info->lru_len = UT_LIST_GET_LEN(buf_pool->LRU);

	pool_info->old_lru_len = buf_pool->LRU_old_len;

	pool_info->free_list_len = UT_LIST_GET_LEN(buf_pool->free);

	pool_info->flush_list_len = UT_LIST_GET_LEN(buf_pool->flush_list);

	pool_info->n_pend_unzip = UT_LIST_GET_LEN(buf_pool->unzip_LRU);

	pool_info->n_pend_reads = buf_pool->n_pend_reads;

	pool_info->n_pending_flush_lru =
		 (buf_pool->n_flush[BUF_FLUSH_LRU]
		  + buf_pool->init_flush[BUF_FLUSH_LRU]);

	pool_info->n_pending_flush_list =
		 (buf_pool->n_flush[BUF_FLUSH_LIST]
		  + buf_pool->init_flush[BUF_FLUSH_LIST]);

	pool_info->n_pending_flush_single_page =
		 (buf_pool->n_flush[BUF_FLUSH_SINGLE_PAGE]
		  + buf_pool->init_flush[BUF_FLUSH_SINGLE_PAGE]);

	buf_flush_list_mutex_exit(buf_pool);

	current_time = time(NULL);
	time_elapsed = 0.001 + difftime(current_time,
					buf_pool->last_printout_time);

	pool_info->n_pages_made_young = buf_pool->stat.n_pages_made_young;

	pool_info->n_pages_not_made_young =
		buf_pool->stat.n_pages_not_made_young;

	pool_info->n_pages_read = buf_pool->stat.n_pages_read;

	pool_info->n_pages_created = buf_pool->stat.n_pages_created;

	pool_info->n_pages_written = buf_pool->stat.n_pages_written;

	pool_info->n_page_gets = buf_pool->stat.n_page_gets;

	pool_info->n_ra_pages_read_rnd = buf_pool->stat.n_ra_pages_read_rnd;
	pool_info->n_ra_pages_read = buf_pool->stat.n_ra_pages_read;

	pool_info->n_ra_pages_evicted = buf_pool->stat.n_ra_pages_evicted;

	pool_info->page_made_young_rate =
		 (buf_pool->stat.n_pages_made_young
		  - buf_pool->old_stat.n_pages_made_young) / time_elapsed;

	pool_info->page_not_made_young_rate =
		 (buf_pool->stat.n_pages_not_made_young
		  - buf_pool->old_stat.n_pages_not_made_young) / time_elapsed;

	pool_info->pages_read_rate =
		(buf_pool->stat.n_pages_read
		  - buf_pool->old_stat.n_pages_read) / time_elapsed;

	pool_info->pages_created_rate =
		(buf_pool->stat.n_pages_created
		 - buf_pool->old_stat.n_pages_created) / time_elapsed;

	pool_info->pages_written_rate =
		(buf_pool->stat.n_pages_written
		 - buf_pool->old_stat.n_pages_written) / time_elapsed;

	pool_info->n_page_get_delta = buf_pool->stat.n_page_gets
				      - buf_pool->old_stat.n_page_gets;

	if (pool_info->n_page_get_delta) {
		pool_info->page_read_delta = buf_pool->stat.n_pages_read
					     - buf_pool->old_stat.n_pages_read;

		pool_info->young_making_delta =
			buf_pool->stat.n_pages_made_young
			- buf_pool->old_stat.n_pages_made_young;

		pool_info->not_young_making_delta =
			buf_pool->stat.n_pages_not_made_young
			- buf_pool->old_stat.n_pages_not_made_young;
	}
	pool_info->pages_readahead_rnd_rate =
		 (buf_pool->stat.n_ra_pages_read_rnd
		  - buf_pool->old_stat.n_ra_pages_read_rnd) / time_elapsed;


	pool_info->pages_readahead_rate =
		 (buf_pool->stat.n_ra_pages_read
		  - buf_pool->old_stat.n_ra_pages_read) / time_elapsed;

	pool_info->pages_evicted_rate =
		(buf_pool->stat.n_ra_pages_evicted
		 - buf_pool->old_stat.n_ra_pages_evicted) / time_elapsed;

	pool_info->unzip_lru_len = UT_LIST_GET_LEN(buf_pool->unzip_LRU);

	pool_info->io_sum = buf_LRU_stat_sum.io;

	pool_info->io_cur = buf_LRU_stat_cur.io;

	pool_info->unzip_sum = buf_LRU_stat_sum.unzip;

	pool_info->unzip_cur = buf_LRU_stat_cur.unzip;

	buf_refresh_io_stats(buf_pool);
	buf_pool_mutex_exit(buf_pool);
}

/*********************************************************************//**
Prints info of the buffer i/o. */
static
void
buf_print_io_instance(
/*==================*/
	buf_pool_info_t*pool_info,	/*!< in: buffer pool info */
	FILE*		file)		/*!< in/out: buffer where to print */
{
	ut_ad(pool_info);

	fprintf(file,
		"Buffer pool size   " ULINTPF "\n"
		"Free buffers       " ULINTPF "\n"
		"Database pages     " ULINTPF "\n"
		"Old database pages " ULINTPF "\n"
		"Modified db pages  " ULINTPF "\n"
		"Percent of dirty pages(LRU & free pages): %.3f\n"
		"Max dirty pages percent: %.3f\n"
		"Pending reads " ULINTPF "\n"
		"Pending writes: LRU " ULINTPF ", flush list " ULINTPF
		", single page " ULINTPF "\n",
		pool_info->pool_size,
		pool_info->free_list_len,
		pool_info->lru_len,
		pool_info->old_lru_len,
		pool_info->flush_list_len,
		(((double) pool_info->flush_list_len) /
		  (pool_info->lru_len + pool_info->free_list_len + 1.0)) * 100.0,
		srv_max_buf_pool_modified_pct,
		pool_info->n_pend_reads,
		pool_info->n_pending_flush_lru,
		pool_info->n_pending_flush_list,
		pool_info->n_pending_flush_single_page);

	fprintf(file,
		"Pages made young " ULINTPF ", not young " ULINTPF "\n"
		"%.2f youngs/s, %.2f non-youngs/s\n"
		"Pages read " ULINTPF ", created " ULINTPF
		", written " ULINTPF "\n"
		"%.2f reads/s, %.2f creates/s, %.2f writes/s\n",
		pool_info->n_pages_made_young,
		pool_info->n_pages_not_made_young,
		pool_info->page_made_young_rate,
		pool_info->page_not_made_young_rate,
		pool_info->n_pages_read,
		pool_info->n_pages_created,
		pool_info->n_pages_written,
		pool_info->pages_read_rate,
		pool_info->pages_created_rate,
		pool_info->pages_written_rate);

	if (pool_info->n_page_get_delta) {
		double hit_rate = double(pool_info->page_read_delta)
			/ pool_info->n_page_get_delta;

		if (hit_rate > 1) {
			hit_rate = 1;
		}

		fprintf(file,
			"Buffer pool hit rate " ULINTPF " / 1000,"
			" young-making rate " ULINTPF " / 1000 not "
			ULINTPF " / 1000\n",
			ulint(1000 * (1 - hit_rate)),
			ulint(1000 * double(pool_info->young_making_delta)
			      / pool_info->n_page_get_delta),
			ulint(1000 * double(pool_info->not_young_making_delta)
			      / pool_info->n_page_get_delta));
	} else {
		fputs("No buffer pool page gets since the last printout\n",
		      file);
	}

	/* Statistics about read ahead algorithm */
	fprintf(file, "Pages read ahead %.2f/s,"
		" evicted without access %.2f/s,"
		" Random read ahead %.2f/s\n",

		pool_info->pages_readahead_rate,
		pool_info->pages_evicted_rate,
		pool_info->pages_readahead_rnd_rate);

	/* Print some values to help us with visualizing what is
	happening with LRU eviction. */
	fprintf(file,
		"LRU len: " ULINTPF ", unzip_LRU len: " ULINTPF "\n"
		"I/O sum[" ULINTPF "]:cur[" ULINTPF "], "
		"unzip sum[" ULINTPF "]:cur[" ULINTPF "]\n",
		pool_info->lru_len, pool_info->unzip_lru_len,
		pool_info->io_sum, pool_info->io_cur,
		pool_info->unzip_sum, pool_info->unzip_cur);
}

/*********************************************************************//**
Prints info of the buffer i/o. */
void
buf_print_io(
/*=========*/
	FILE*	file)	/*!< in/out: buffer where to print */
{
	ulint			i;
	buf_pool_info_t*	pool_info;
	buf_pool_info_t*	pool_info_total;

	/* If srv_buf_pool_instances is greater than 1, allocate
	one extra buf_pool_info_t, the last one stores
	aggregated/total values from all pools */
	if (srv_buf_pool_instances > 1) {
		pool_info = (buf_pool_info_t*) ut_zalloc_nokey((
			srv_buf_pool_instances + 1) * sizeof *pool_info);

		pool_info_total = &pool_info[srv_buf_pool_instances];
	} else {
		ut_a(srv_buf_pool_instances == 1);

		pool_info_total = pool_info =
			static_cast<buf_pool_info_t*>(
				ut_zalloc_nokey(sizeof *pool_info));
	}

	for (i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		/* Fetch individual buffer pool info and calculate
		aggregated stats along the way */
		buf_stats_get_pool_info(buf_pool, i, pool_info);

		/* If we have more than one buffer pool, store
		the aggregated stats  */
		if (srv_buf_pool_instances > 1) {
			buf_stats_aggregate_pool_info(pool_info_total,
						      &pool_info[i]);
		}
	}

	/* Print the aggreate buffer pool info */
	buf_print_io_instance(pool_info_total, file);

	/* If there are more than one buffer pool, print each individual pool
	info */
	if (srv_buf_pool_instances > 1) {
		fputs("----------------------\n"
		"INDIVIDUAL BUFFER POOL INFO\n"
		"----------------------\n", file);

		for (i = 0; i < srv_buf_pool_instances; i++) {
			fprintf(file, "---BUFFER POOL " ULINTPF "\n", i);
			buf_print_io_instance(&pool_info[i], file);
		}
	}

	ut_free(pool_info);
}

/**********************************************************************//**
Refreshes the statistics used to print per-second averages. */
void
buf_refresh_io_stats_all(void)
/*==========================*/
{
	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		buf_refresh_io_stats(buf_pool);
	}
}

/**********************************************************************//**
Check if all pages in all buffer pools are in a replacable state.
@return FALSE if not */
ibool
buf_all_freed(void)
/*===============*/
{
	for (ulint i = 0; i < srv_buf_pool_instances; i++) {
		buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		if (!buf_all_freed_instance(buf_pool)) {
			return(FALSE);
		}
	}

	return(TRUE);
}

/*********************************************************************//**
Checks that there currently are no pending i/o-operations for the buffer
pool.
@return number of pending i/o */
ulint
buf_pool_check_no_pending_io(void)
/*==============================*/
{
	ulint		i;
	ulint		pending_io = 0;

	buf_pool_mutex_enter_all();

	for (i = 0; i < srv_buf_pool_instances; i++) {
		const buf_pool_t*	buf_pool;

		buf_pool = buf_pool_from_array(i);

		pending_io += buf_pool->n_pend_reads
			      + buf_pool->n_flush[BUF_FLUSH_LRU]
			      + buf_pool->n_flush[BUF_FLUSH_SINGLE_PAGE]
			      + buf_pool->n_flush[BUF_FLUSH_LIST];

	}

	buf_pool_mutex_exit_all();

	return(pending_io);
}

/** Print the given page_id_t object.
@param[in,out]	out	the output stream
@param[in]	page_id	the page_id_t object to be printed
@return the output stream */
std::ostream&
operator<<(
	std::ostream&		out,
	const page_id_t		page_id)
{
	out << "[page id: space=" << page_id.m_space
		<< ", page number=" << page_id.m_page_no << "]";
	return(out);
}

#if defined UNIV_DEBUG_PRINT || defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
/** Print the given buf_pool_t object.
@param[in,out]	out		the output stream
@param[in]	buf_pool	the buf_pool_t object to be printed
@return the output stream */
std::ostream&
operator<<(
	std::ostream&		out,
	const buf_pool_t&	buf_pool)
{
	out << "[buffer pool instance: "
		<< "buf_pool size=" << buf_pool.curr_size
		<< ", database pages=" << UT_LIST_GET_LEN(buf_pool.LRU)
		<< ", free pages=" << UT_LIST_GET_LEN(buf_pool.free)
		<< ", modified database pages="
		<< UT_LIST_GET_LEN(buf_pool.flush_list)
		<< ", n pending decompressions=" << buf_pool.n_pend_unzip
		<< ", n pending reads=" << buf_pool.n_pend_reads
		<< ", n pending flush LRU=" << buf_pool.n_flush[BUF_FLUSH_LRU]
		<< " list=" << buf_pool.n_flush[BUF_FLUSH_LIST]
		<< " single page=" << buf_pool.n_flush[BUF_FLUSH_SINGLE_PAGE]
		<< ", pages made young=" << buf_pool.stat.n_pages_made_young
		<< ", not young=" << buf_pool.stat.n_pages_not_made_young
		<< ", pages read=" << buf_pool.stat.n_pages_read
		<< ", created=" << buf_pool.stat.n_pages_created
		<< ", written=" << buf_pool.stat.n_pages_written << "]";
	return(out);
}
#endif /* UNIV_DEBUG_PRINT || UNIV_DEBUG || UNIV_BUF_DEBUG */

/** Encrypt a buffer of temporary tablespace
@param[in]	offset		Page offset
@param[in]	src_frame	Page to encrypt
@param[in,out]	dst_frame	Output buffer
@return encrypted buffer or NULL */
static byte* buf_tmp_page_encrypt(
	ulint	offset,
	byte*	src_frame,
	byte*	dst_frame)
{
	uint header_len = FIL_PAGE_DATA;
	/* FIL page header is not encrypted */
	memcpy(dst_frame, src_frame, header_len);

	/* Calculate the start offset in a page */
	uint unencrypted_bytes = header_len + FIL_PAGE_DATA_END;
	uint srclen = srv_page_size - unencrypted_bytes;
	const byte* src = src_frame + header_len;
	byte* dst = dst_frame + header_len;

	if (!log_tmp_block_encrypt(src, srclen, dst, (offset * srv_page_size),
				   true)) {
		return NULL;
	}

	memcpy(dst_frame + srv_page_size - FIL_PAGE_DATA_END,
	       src_frame + srv_page_size - FIL_PAGE_DATA_END,
	       FIL_PAGE_DATA_END);

	/* Handle post encryption checksum */
	mach_write_to_4(dst_frame + FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION + 4,
			buf_calc_page_crc32(dst_frame));

	srv_stats.pages_encrypted.inc();
	srv_stats.n_temp_blocks_encrypted.inc();
	return dst_frame;
}

/** Encryption and page_compression hook that is called just before
a page is written to disk.
@param[in,out]	space		tablespace
@param[in,out]	bpage		buffer page
@param[in]	src_frame	physical page frame that is being encrypted
@return	page frame to be written to file
(may be src_frame or an encrypted/compressed copy of it) */
UNIV_INTERN
byte*
buf_page_encrypt_before_write(
	fil_space_t*	space,
	buf_page_t*	bpage,
	byte*		src_frame)
{
	ut_ad(space->id == bpage->id.space());
	bpage->real_size = UNIV_PAGE_SIZE;

	fil_page_type_validate(src_frame);

	switch (bpage->id.page_no()) {
	case 0:
		/* Page 0 of a tablespace is not encrypted/compressed */
		return src_frame;
	case TRX_SYS_PAGE_NO:
		if (bpage->id.space() == TRX_SYS_SPACE) {
			/* don't encrypt/compress page as it contains
			address to dblwr buffer */
			return src_frame;
		}
	}

	fil_space_crypt_t* crypt_data = space->crypt_data;

	bool encrypted, page_compressed;

	if (space->purpose == FIL_TYPE_TEMPORARY) {
		ut_ad(!crypt_data);
		encrypted = innodb_encrypt_temporary_tables;
		page_compressed = false;
	} else {
		encrypted = crypt_data
			&& !crypt_data->not_encrypted()
			&& crypt_data->type != CRYPT_SCHEME_UNENCRYPTED
			&& (!crypt_data->is_default_encryption()
			    || srv_encrypt_tables);

		page_compressed = FSP_FLAGS_HAS_PAGE_COMPRESSION(space->flags);
	}

	if (!encrypted && !page_compressed) {
		/* No need to encrypt or page compress the page.
		Clear key-version & crypt-checksum. */
		memset(src_frame + FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION, 0, 8);
		return src_frame;
	}

	ut_ad(!bpage->size.is_compressed() || !page_compressed);
	buf_pool_t* buf_pool = buf_pool_from_bpage(bpage);
	/* Find free slot from temporary memory array */
	buf_tmp_buffer_t* slot = buf_pool_reserve_tmp_slot(buf_pool);
	slot->out_buf = NULL;
	bpage->slot = slot;

	buf_tmp_reserve_crypt_buf(slot);
	byte *dst_frame = slot->crypt_buf;

	if (!page_compressed) {
not_compressed:
		byte* tmp;
		if (space->purpose == FIL_TYPE_TEMPORARY) {
			/* Encrypt temporary tablespace page content */
			tmp = buf_tmp_page_encrypt(bpage->id.page_no(),
						   src_frame, dst_frame);
		} else {
			/* Encrypt page content */
			tmp = fil_space_encrypt(
					space, bpage->id.page_no(),
					bpage->newest_modification,
					src_frame, dst_frame);
		}

		bpage->real_size = UNIV_PAGE_SIZE;
		slot->out_buf = dst_frame = tmp;

		ut_d(fil_page_type_validate(tmp));
	} else {
		ut_ad(space->purpose != FIL_TYPE_TEMPORARY);
		/* First we compress the page content */
		buf_tmp_reserve_compression_buf(slot);
		byte* tmp = slot->comp_buf;
		ulint out_len = fil_page_compress(
			src_frame, tmp,
			fsp_flags_get_page_compression_level(space->flags),
			fil_space_get_block_size(space, bpage->id.page_no()),
			encrypted);
		if (!out_len) {
			goto not_compressed;
		}

		bpage->real_size = out_len;

		/* Workaround for MDEV-15527. */
		memset(tmp + out_len, 0 , srv_page_size - out_len);
		ut_d(fil_page_type_validate(tmp));

		if (encrypted) {
			/* And then we encrypt the page content */
			tmp = fil_space_encrypt(space,
						bpage->id.page_no(),
						bpage->newest_modification,
						tmp,
						dst_frame);
		}

		slot->out_buf = dst_frame = tmp;
	}

	ut_d(fil_page_type_validate(dst_frame));

	// return dst_frame which will be written
	return dst_frame;
}

/**
Should we punch hole to deallocate unused portion of the page.
@param[in]	bpage		Page control block
@return true if punch hole should be used, false if not */
bool
buf_page_should_punch_hole(
	const buf_page_t* bpage)
{
	return (bpage->real_size != bpage->size.physical());
}

/**
Calculate the length of trim (punch_hole) operation.
@param[in]	bpage		Page control block
@param[in]	write_length	Write length
@return length of the trim or zero. */
ulint
buf_page_get_trim_length(
	const buf_page_t*	bpage,
	ulint			write_length)
{
	return (bpage->size.physical() - write_length);
}
#endif /* !UNIV_INNOCHECKSUM */