aboutsummaryrefslogtreecommitdiffstats
path: root/module/zfs/arc.c
blob: c1b86717427adcc22bbbd05e353b90fadc8ef563 (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
/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
 */

/*
 * DVA-based Adjustable Replacement Cache
 *
 * While much of the theory of operation used here is
 * based on the self-tuning, low overhead replacement cache
 * presented by Megiddo and Modha at FAST 2003, there are some
 * significant differences:
 *
 * 1. The Megiddo and Modha model assumes any page is evictable.
 * Pages in its cache cannot be "locked" into memory.  This makes
 * the eviction algorithm simple: evict the last page in the list.
 * This also make the performance characteristics easy to reason
 * about.  Our cache is not so simple.  At any given moment, some
 * subset of the blocks in the cache are un-evictable because we
 * have handed out a reference to them.  Blocks are only evictable
 * when there are no external references active.  This makes
 * eviction far more problematic:  we choose to evict the evictable
 * blocks that are the "lowest" in the list.
 *
 * There are times when it is not possible to evict the requested
 * space.  In these circumstances we are unable to adjust the cache
 * size.  To prevent the cache growing unbounded at these times we
 * implement a "cache throttle" that slows the flow of new data
 * into the cache until we can make space available.
 *
 * 2. The Megiddo and Modha model assumes a fixed cache size.
 * Pages are evicted when the cache is full and there is a cache
 * miss.  Our model has a variable sized cache.  It grows with
 * high use, but also tries to react to memory pressure from the
 * operating system: decreasing its size when system memory is
 * tight.
 *
 * 3. The Megiddo and Modha model assumes a fixed page size. All
 * elements of the cache are therefor exactly the same size.  So
 * when adjusting the cache size following a cache miss, its simply
 * a matter of choosing a single page to evict.  In our model, we
 * have variable sized cache blocks (rangeing from 512 bytes to
 * 128K bytes).  We therefor choose a set of blocks to evict to make
 * space for a cache miss that approximates as closely as possible
 * the space used by the new block.
 *
 * See also:  "ARC: A Self-Tuning, Low Overhead Replacement Cache"
 * by N. Megiddo & D. Modha, FAST 2003
 */

/*
 * The locking model:
 *
 * A new reference to a cache buffer can be obtained in two
 * ways: 1) via a hash table lookup using the DVA as a key,
 * or 2) via one of the ARC lists.  The arc_read() interface
 * uses method 1, while the internal arc algorithms for
 * adjusting the cache use method 2.  We therefor provide two
 * types of locks: 1) the hash table lock array, and 2) the
 * arc list locks.
 *
 * Buffers do not have their own mutexs, rather they rely on the
 * hash table mutexs for the bulk of their protection (i.e. most
 * fields in the arc_buf_hdr_t are protected by these mutexs).
 *
 * buf_hash_find() returns the appropriate mutex (held) when it
 * locates the requested buffer in the hash table.  It returns
 * NULL for the mutex if the buffer was not in the table.
 *
 * buf_hash_remove() expects the appropriate hash mutex to be
 * already held before it is invoked.
 *
 * Each arc state also has a mutex which is used to protect the
 * buffer list associated with the state.  When attempting to
 * obtain a hash table lock while holding an arc list lock you
 * must use: mutex_tryenter() to avoid deadlock.  Also note that
 * the active state mutex must be held before the ghost state mutex.
 *
 * Arc buffers may have an associated eviction callback function.
 * This function will be invoked prior to removing the buffer (e.g.
 * in arc_do_user_evicts()).  Note however that the data associated
 * with the buffer may be evicted prior to the callback.  The callback
 * must be made with *no locks held* (to prevent deadlock).  Additionally,
 * the users of callbacks must ensure that their private data is
 * protected from simultaneous callbacks from arc_buf_evict()
 * and arc_do_user_evicts().
 *
 * Note that the majority of the performance stats are manipulated
 * with atomic operations.
 *
 * The L2ARC uses the l2arc_buflist_mtx global mutex for the following:
 *
 *	- L2ARC buflist creation
 *	- L2ARC buflist eviction
 *	- L2ARC write completion, which walks L2ARC buflists
 *	- ARC header destruction, as it removes from L2ARC buflists
 *	- ARC header release, as it removes from L2ARC buflists
 */

#include <sys/spa.h>
#include <sys/zio.h>
#include <sys/zfs_context.h>
#include <sys/arc.h>
#include <sys/refcount.h>
#include <sys/vdev.h>
#include <sys/vdev_impl.h>
#ifdef _KERNEL
#include <sys/vmsystm.h>
#include <vm/anon.h>
#include <sys/fs/swapnode.h>
#include <sys/dnlc.h>
#endif
#include <sys/callb.h>
#include <sys/kstat.h>
#include <zfs_fletcher.h>

static kmutex_t		arc_reclaim_thr_lock;
static kcondvar_t	arc_reclaim_thr_cv;	/* used to signal reclaim thr */
static uint8_t		arc_thread_exit;

extern int zfs_write_limit_shift;
extern uint64_t zfs_write_limit_max;
extern kmutex_t zfs_write_limit_lock;

#define	ARC_REDUCE_DNLC_PERCENT	3
uint_t arc_reduce_dnlc_percent = ARC_REDUCE_DNLC_PERCENT;

typedef enum arc_reclaim_strategy {
	ARC_RECLAIM_AGGR,		/* Aggressive reclaim strategy */
	ARC_RECLAIM_CONS		/* Conservative reclaim strategy */
} arc_reclaim_strategy_t;

/* number of seconds before growing cache again */
static int		arc_grow_retry = 60;

/* shift of arc_c for calculating both min and max arc_p */
static int		arc_p_min_shift = 4;

/* log2(fraction of arc to reclaim) */
static int		arc_shrink_shift = 5;

/*
 * minimum lifespan of a prefetch block in clock ticks
 * (initialized in arc_init())
 */
static int		arc_min_prefetch_lifespan;

static int arc_dead;

/*
 * The arc has filled available memory and has now warmed up.
 */
static boolean_t arc_warm;

/*
 * These tunables are for performance analysis.
 */
uint64_t zfs_arc_max;
uint64_t zfs_arc_min;
uint64_t zfs_arc_meta_limit = 0;
int zfs_arc_grow_retry = 0;
int zfs_arc_shrink_shift = 0;
int zfs_arc_p_min_shift = 0;

/*
 * Note that buffers can be in one of 6 states:
 *	ARC_anon	- anonymous (discussed below)
 *	ARC_mru		- recently used, currently cached
 *	ARC_mru_ghost	- recentely used, no longer in cache
 *	ARC_mfu		- frequently used, currently cached
 *	ARC_mfu_ghost	- frequently used, no longer in cache
 *	ARC_l2c_only	- exists in L2ARC but not other states
 * When there are no active references to the buffer, they are
 * are linked onto a list in one of these arc states.  These are
 * the only buffers that can be evicted or deleted.  Within each
 * state there are multiple lists, one for meta-data and one for
 * non-meta-data.  Meta-data (indirect blocks, blocks of dnodes,
 * etc.) is tracked separately so that it can be managed more
 * explicitly: favored over data, limited explicitly.
 *
 * Anonymous buffers are buffers that are not associated with
 * a DVA.  These are buffers that hold dirty block copies
 * before they are written to stable storage.  By definition,
 * they are "ref'd" and are considered part of arc_mru
 * that cannot be freed.  Generally, they will aquire a DVA
 * as they are written and migrate onto the arc_mru list.
 *
 * The ARC_l2c_only state is for buffers that are in the second
 * level ARC but no longer in any of the ARC_m* lists.  The second
 * level ARC itself may also contain buffers that are in any of
 * the ARC_m* states - meaning that a buffer can exist in two
 * places.  The reason for the ARC_l2c_only state is to keep the
 * buffer header in the hash table, so that reads that hit the
 * second level ARC benefit from these fast lookups.
 */

typedef struct arc_state {
	list_t	arcs_list[ARC_BUFC_NUMTYPES];	/* list of evictable buffers */
	uint64_t arcs_lsize[ARC_BUFC_NUMTYPES];	/* amount of evictable data */
	uint64_t arcs_size;	/* total amount of data in this state */
	kmutex_t arcs_mtx;
} arc_state_t;

/* The 6 states: */
static arc_state_t ARC_anon;
static arc_state_t ARC_mru;
static arc_state_t ARC_mru_ghost;
static arc_state_t ARC_mfu;
static arc_state_t ARC_mfu_ghost;
static arc_state_t ARC_l2c_only;

typedef struct arc_stats {
	kstat_named_t arcstat_hits;
	kstat_named_t arcstat_misses;
	kstat_named_t arcstat_demand_data_hits;
	kstat_named_t arcstat_demand_data_misses;
	kstat_named_t arcstat_demand_metadata_hits;
	kstat_named_t arcstat_demand_metadata_misses;
	kstat_named_t arcstat_prefetch_data_hits;
	kstat_named_t arcstat_prefetch_data_misses;
	kstat_named_t arcstat_prefetch_metadata_hits;
	kstat_named_t arcstat_prefetch_metadata_misses;
	kstat_named_t arcstat_mru_hits;
	kstat_named_t arcstat_mru_ghost_hits;
	kstat_named_t arcstat_mfu_hits;
	kstat_named_t arcstat_mfu_ghost_hits;
	kstat_named_t arcstat_deleted;
	kstat_named_t arcstat_recycle_miss;
	kstat_named_t arcstat_mutex_miss;
	kstat_named_t arcstat_evict_skip;
	kstat_named_t arcstat_evict_l2_cached;
	kstat_named_t arcstat_evict_l2_eligible;
	kstat_named_t arcstat_evict_l2_ineligible;
	kstat_named_t arcstat_hash_elements;
	kstat_named_t arcstat_hash_elements_max;
	kstat_named_t arcstat_hash_collisions;
	kstat_named_t arcstat_hash_chains;
	kstat_named_t arcstat_hash_chain_max;
	kstat_named_t arcstat_p;
	kstat_named_t arcstat_c;
	kstat_named_t arcstat_c_min;
	kstat_named_t arcstat_c_max;
	kstat_named_t arcstat_size;
	kstat_named_t arcstat_hdr_size;
	kstat_named_t arcstat_data_size;
	kstat_named_t arcstat_other_size;
	kstat_named_t arcstat_l2_hits;
	kstat_named_t arcstat_l2_misses;
	kstat_named_t arcstat_l2_feeds;
	kstat_named_t arcstat_l2_rw_clash;
	kstat_named_t arcstat_l2_read_bytes;
	kstat_named_t arcstat_l2_write_bytes;
	kstat_named_t arcstat_l2_writes_sent;
	kstat_named_t arcstat_l2_writes_done;
	kstat_named_t arcstat_l2_writes_error;
	kstat_named_t arcstat_l2_writes_hdr_miss;
	kstat_named_t arcstat_l2_evict_lock_retry;
	kstat_named_t arcstat_l2_evict_reading;
	kstat_named_t arcstat_l2_free_on_write;
	kstat_named_t arcstat_l2_abort_lowmem;
	kstat_named_t arcstat_l2_cksum_bad;
	kstat_named_t arcstat_l2_io_error;
	kstat_named_t arcstat_l2_size;
	kstat_named_t arcstat_l2_hdr_size;
	kstat_named_t arcstat_memory_throttle_count;
} arc_stats_t;

static arc_stats_t arc_stats = {
	{ "hits",			KSTAT_DATA_UINT64 },
	{ "misses",			KSTAT_DATA_UINT64 },
	{ "demand_data_hits",		KSTAT_DATA_UINT64 },
	{ "demand_data_misses",		KSTAT_DATA_UINT64 },
	{ "demand_metadata_hits",	KSTAT_DATA_UINT64 },
	{ "demand_metadata_misses",	KSTAT_DATA_UINT64 },
	{ "prefetch_data_hits",		KSTAT_DATA_UINT64 },
	{ "prefetch_data_misses",	KSTAT_DATA_UINT64 },
	{ "prefetch_metadata_hits",	KSTAT_DATA_UINT64 },
	{ "prefetch_metadata_misses",	KSTAT_DATA_UINT64 },
	{ "mru_hits",			KSTAT_DATA_UINT64 },
	{ "mru_ghost_hits",		KSTAT_DATA_UINT64 },
	{ "mfu_hits",			KSTAT_DATA_UINT64 },
	{ "mfu_ghost_hits",		KSTAT_DATA_UINT64 },
	{ "deleted",			KSTAT_DATA_UINT64 },
	{ "recycle_miss",		KSTAT_DATA_UINT64 },
	{ "mutex_miss",			KSTAT_DATA_UINT64 },
	{ "evict_skip",			KSTAT_DATA_UINT64 },
	{ "evict_l2_cached",		KSTAT_DATA_UINT64 },
	{ "evict_l2_eligible",		KSTAT_DATA_UINT64 },
	{ "evict_l2_ineligible",	KSTAT_DATA_UINT64 },
	{ "hash_elements",		KSTAT_DATA_UINT64 },
	{ "hash_elements_max",		KSTAT_DATA_UINT64 },
	{ "hash_collisions",		KSTAT_DATA_UINT64 },
	{ "hash_chains",		KSTAT_DATA_UINT64 },
	{ "hash_chain_max",		KSTAT_DATA_UINT64 },
	{ "p",				KSTAT_DATA_UINT64 },
	{ "c",				KSTAT_DATA_UINT64 },
	{ "c_min",			KSTAT_DATA_UINT64 },
	{ "c_max",			KSTAT_DATA_UINT64 },
	{ "size",			KSTAT_DATA_UINT64 },
	{ "hdr_size",			KSTAT_DATA_UINT64 },
	{ "data_size",			KSTAT_DATA_UINT64 },
	{ "other_size",			KSTAT_DATA_UINT64 },
	{ "l2_hits",			KSTAT_DATA_UINT64 },
	{ "l2_misses",			KSTAT_DATA_UINT64 },
	{ "l2_feeds",			KSTAT_DATA_UINT64 },
	{ "l2_rw_clash",		KSTAT_DATA_UINT64 },
	{ "l2_read_bytes",		KSTAT_DATA_UINT64 },
	{ "l2_write_bytes",		KSTAT_DATA_UINT64 },
	{ "l2_writes_sent",		KSTAT_DATA_UINT64 },
	{ "l2_writes_done",		KSTAT_DATA_UINT64 },
	{ "l2_writes_error",		KSTAT_DATA_UINT64 },
	{ "l2_writes_hdr_miss",		KSTAT_DATA_UINT64 },
	{ "l2_evict_lock_retry",	KSTAT_DATA_UINT64 },
	{ "l2_evict_reading",		KSTAT_DATA_UINT64 },
	{ "l2_free_on_write",		KSTAT_DATA_UINT64 },
	{ "l2_abort_lowmem",		KSTAT_DATA_UINT64 },
	{ "l2_cksum_bad",		KSTAT_DATA_UINT64 },
	{ "l2_io_error",		KSTAT_DATA_UINT64 },
	{ "l2_size",			KSTAT_DATA_UINT64 },
	{ "l2_hdr_size",		KSTAT_DATA_UINT64 },
	{ "memory_throttle_count",	KSTAT_DATA_UINT64 }
};

#define	ARCSTAT(stat)	(arc_stats.stat.value.ui64)

#define	ARCSTAT_INCR(stat, val) \
	atomic_add_64(&arc_stats.stat.value.ui64, (val));

#define	ARCSTAT_BUMP(stat)	ARCSTAT_INCR(stat, 1)
#define	ARCSTAT_BUMPDOWN(stat)	ARCSTAT_INCR(stat, -1)

#define	ARCSTAT_MAX(stat, val) {					\
	uint64_t m;							\
	while ((val) > (m = arc_stats.stat.value.ui64) &&		\
	    (m != atomic_cas_64(&arc_stats.stat.value.ui64, m, (val))))	\
		continue;						\
}

#define	ARCSTAT_MAXSTAT(stat) \
	ARCSTAT_MAX(stat##_max, arc_stats.stat.value.ui64)

/*
 * We define a macro to allow ARC hits/misses to be easily broken down by
 * two separate conditions, giving a total of four different subtypes for
 * each of hits and misses (so eight statistics total).
 */
#define	ARCSTAT_CONDSTAT(cond1, stat1, notstat1, cond2, stat2, notstat2, stat) \
	if (cond1) {							\
		if (cond2) {						\
			ARCSTAT_BUMP(arcstat_##stat1##_##stat2##_##stat); \
		} else {						\
			ARCSTAT_BUMP(arcstat_##stat1##_##notstat2##_##stat); \
		}							\
	} else {							\
		if (cond2) {						\
			ARCSTAT_BUMP(arcstat_##notstat1##_##stat2##_##stat); \
		} else {						\
			ARCSTAT_BUMP(arcstat_##notstat1##_##notstat2##_##stat);\
		}							\
	}

kstat_t			*arc_ksp;
static arc_state_t	*arc_anon;
static arc_state_t	*arc_mru;
static arc_state_t	*arc_mru_ghost;
static arc_state_t	*arc_mfu;
static arc_state_t	*arc_mfu_ghost;
static arc_state_t	*arc_l2c_only;

/*
 * There are several ARC variables that are critical to export as kstats --
 * but we don't want to have to grovel around in the kstat whenever we wish to
 * manipulate them.  For these variables, we therefore define them to be in
 * terms of the statistic variable.  This assures that we are not introducing
 * the possibility of inconsistency by having shadow copies of the variables,
 * while still allowing the code to be readable.
 */
#define	arc_size	ARCSTAT(arcstat_size)	/* actual total arc size */
#define	arc_p		ARCSTAT(arcstat_p)	/* target size of MRU */
#define	arc_c		ARCSTAT(arcstat_c)	/* target size of cache */
#define	arc_c_min	ARCSTAT(arcstat_c_min)	/* min target cache size */
#define	arc_c_max	ARCSTAT(arcstat_c_max)	/* max target cache size */

static int		arc_no_grow;	/* Don't try to grow cache size */
static uint64_t		arc_tempreserve;
static uint64_t		arc_loaned_bytes;
static uint64_t		arc_meta_used;
static uint64_t		arc_meta_limit;
static uint64_t		arc_meta_max = 0;

typedef struct l2arc_buf_hdr l2arc_buf_hdr_t;

typedef struct arc_callback arc_callback_t;

struct arc_callback {
	void			*acb_private;
	arc_done_func_t		*acb_done;
	arc_buf_t		*acb_buf;
	zio_t			*acb_zio_dummy;
	arc_callback_t		*acb_next;
};

typedef struct arc_write_callback arc_write_callback_t;

struct arc_write_callback {
	void		*awcb_private;
	arc_done_func_t	*awcb_ready;
	arc_done_func_t	*awcb_done;
	arc_buf_t	*awcb_buf;
};

struct arc_buf_hdr {
	/* protected by hash lock */
	dva_t			b_dva;
	uint64_t		b_birth;
	uint64_t		b_cksum0;

	kmutex_t		b_freeze_lock;
	zio_cksum_t		*b_freeze_cksum;
	void			*b_thawed;

	arc_buf_hdr_t		*b_hash_next;
	arc_buf_t		*b_buf;
	uint32_t		b_flags;
	uint32_t		b_datacnt;

	arc_callback_t		*b_acb;
	kcondvar_t		b_cv;

	/* immutable */
	arc_buf_contents_t	b_type;
	uint64_t		b_size;
	uint64_t		b_spa;

	/* protected by arc state mutex */
	arc_state_t		*b_state;
	list_node_t		b_arc_node;

	/* updated atomically */
	clock_t			b_arc_access;

	/* self protecting */
	refcount_t		b_refcnt;

	l2arc_buf_hdr_t		*b_l2hdr;
	list_node_t		b_l2node;
};

static arc_buf_t *arc_eviction_list;
static kmutex_t arc_eviction_mtx;
static arc_buf_hdr_t arc_eviction_hdr;
static void arc_get_data_buf(arc_buf_t *buf);
static void arc_access(arc_buf_hdr_t *buf, kmutex_t *hash_lock);
static int arc_evict_needed(arc_buf_contents_t type);
static void arc_evict_ghost(arc_state_t *state, uint64_t spa, int64_t bytes);

static boolean_t l2arc_write_eligible(uint64_t spa_guid, arc_buf_hdr_t *ab);

#define	GHOST_STATE(state)	\
	((state) == arc_mru_ghost || (state) == arc_mfu_ghost ||	\
	(state) == arc_l2c_only)

/*
 * Private ARC flags.  These flags are private ARC only flags that will show up
 * in b_flags in the arc_hdr_buf_t.  Some flags are publicly declared, and can
 * be passed in as arc_flags in things like arc_read.  However, these flags
 * should never be passed and should only be set by ARC code.  When adding new
 * public flags, make sure not to smash the private ones.
 */

#define	ARC_IN_HASH_TABLE	(1 << 9)	/* this buffer is hashed */
#define	ARC_IO_IN_PROGRESS	(1 << 10)	/* I/O in progress for buf */
#define	ARC_IO_ERROR		(1 << 11)	/* I/O failed for buf */
#define	ARC_FREED_IN_READ	(1 << 12)	/* buf freed while in read */
#define	ARC_BUF_AVAILABLE	(1 << 13)	/* block not in active use */
#define	ARC_INDIRECT		(1 << 14)	/* this is an indirect block */
#define	ARC_FREE_IN_PROGRESS	(1 << 15)	/* hdr about to be freed */
#define	ARC_L2_WRITING		(1 << 16)	/* L2ARC write in progress */
#define	ARC_L2_EVICTED		(1 << 17)	/* evicted during I/O */
#define	ARC_L2_WRITE_HEAD	(1 << 18)	/* head of write list */

#define	HDR_IN_HASH_TABLE(hdr)	((hdr)->b_flags & ARC_IN_HASH_TABLE)
#define	HDR_IO_IN_PROGRESS(hdr)	((hdr)->b_flags & ARC_IO_IN_PROGRESS)
#define	HDR_IO_ERROR(hdr)	((hdr)->b_flags & ARC_IO_ERROR)
#define	HDR_PREFETCH(hdr)	((hdr)->b_flags & ARC_PREFETCH)
#define	HDR_FREED_IN_READ(hdr)	((hdr)->b_flags & ARC_FREED_IN_READ)
#define	HDR_BUF_AVAILABLE(hdr)	((hdr)->b_flags & ARC_BUF_AVAILABLE)
#define	HDR_FREE_IN_PROGRESS(hdr)	((hdr)->b_flags & ARC_FREE_IN_PROGRESS)
#define	HDR_L2CACHE(hdr)	((hdr)->b_flags & ARC_L2CACHE)
#define	HDR_L2_READING(hdr)	((hdr)->b_flags & ARC_IO_IN_PROGRESS &&	\
				    (hdr)->b_l2hdr != NULL)
#define	HDR_L2_WRITING(hdr)	((hdr)->b_flags & ARC_L2_WRITING)
#define	HDR_L2_EVICTED(hdr)	((hdr)->b_flags & ARC_L2_EVICTED)
#define	HDR_L2_WRITE_HEAD(hdr)	((hdr)->b_flags & ARC_L2_WRITE_HEAD)

/*
 * Other sizes
 */

#define	HDR_SIZE ((int64_t)sizeof (arc_buf_hdr_t))
#define	L2HDR_SIZE ((int64_t)sizeof (l2arc_buf_hdr_t))

/*
 * Hash table routines
 */

#define	HT_LOCK_PAD	64

struct ht_lock {
	kmutex_t	ht_lock;
#ifdef _KERNEL
	unsigned char	pad[(HT_LOCK_PAD - sizeof (kmutex_t))];
#endif
};

#define	BUF_LOCKS 256
typedef struct buf_hash_table {
	uint64_t ht_mask;
	arc_buf_hdr_t **ht_table;
	struct ht_lock ht_locks[BUF_LOCKS];
} buf_hash_table_t;

static buf_hash_table_t buf_hash_table;

#define	BUF_HASH_INDEX(spa, dva, birth) \
	(buf_hash(spa, dva, birth) & buf_hash_table.ht_mask)
#define	BUF_HASH_LOCK_NTRY(idx) (buf_hash_table.ht_locks[idx & (BUF_LOCKS-1)])
#define	BUF_HASH_LOCK(idx)	(&(BUF_HASH_LOCK_NTRY(idx).ht_lock))
#define	HDR_LOCK(hdr) \
	(BUF_HASH_LOCK(BUF_HASH_INDEX(hdr->b_spa, &hdr->b_dva, hdr->b_birth)))

uint64_t zfs_crc64_table[256];

/*
 * Level 2 ARC
 */

#define	L2ARC_WRITE_SIZE	(8 * 1024 * 1024)	/* initial write max */
#define	L2ARC_HEADROOM		2		/* num of writes */
#define	L2ARC_FEED_SECS		1		/* caching interval secs */
#define	L2ARC_FEED_MIN_MS	200		/* min caching interval ms */

#define	l2arc_writes_sent	ARCSTAT(arcstat_l2_writes_sent)
#define	l2arc_writes_done	ARCSTAT(arcstat_l2_writes_done)

/*
 * L2ARC Performance Tunables
 */
uint64_t l2arc_write_max = L2ARC_WRITE_SIZE;	/* default max write size */
uint64_t l2arc_write_boost = L2ARC_WRITE_SIZE;	/* extra write during warmup */
uint64_t l2arc_headroom = L2ARC_HEADROOM;	/* number of dev writes */
uint64_t l2arc_feed_secs = L2ARC_FEED_SECS;	/* interval seconds */
uint64_t l2arc_feed_min_ms = L2ARC_FEED_MIN_MS;	/* min interval milliseconds */
boolean_t l2arc_noprefetch = B_TRUE;		/* don't cache prefetch bufs */
boolean_t l2arc_feed_again = B_TRUE;		/* turbo warmup */
boolean_t l2arc_norw = B_TRUE;			/* no reads during writes */

/*
 * L2ARC Internals
 */
typedef struct l2arc_dev {
	vdev_t			*l2ad_vdev;	/* vdev */
	spa_t			*l2ad_spa;	/* spa */
	uint64_t		l2ad_hand;	/* next write location */
	uint64_t		l2ad_write;	/* desired write size, bytes */
	uint64_t		l2ad_boost;	/* warmup write boost, bytes */
	uint64_t		l2ad_start;	/* first addr on device */
	uint64_t		l2ad_end;	/* last addr on device */
	uint64_t		l2ad_evict;	/* last addr eviction reached */
	boolean_t		l2ad_first;	/* first sweep through */
	boolean_t		l2ad_writing;	/* currently writing */
	list_t			*l2ad_buflist;	/* buffer list */
	list_node_t		l2ad_node;	/* device list node */
} l2arc_dev_t;

static list_t L2ARC_dev_list;			/* device list */
static list_t *l2arc_dev_list;			/* device list pointer */
static kmutex_t l2arc_dev_mtx;			/* device list mutex */
static l2arc_dev_t *l2arc_dev_last;		/* last device used */
static kmutex_t l2arc_buflist_mtx;		/* mutex for all buflists */
static list_t L2ARC_free_on_write;		/* free after write buf list */
static list_t *l2arc_free_on_write;		/* free after write list ptr */
static kmutex_t l2arc_free_on_write_mtx;	/* mutex for list */
static uint64_t l2arc_ndev;			/* number of devices */

typedef struct l2arc_read_callback {
	arc_buf_t	*l2rcb_buf;		/* read buffer */
	spa_t		*l2rcb_spa;		/* spa */
	blkptr_t	l2rcb_bp;		/* original blkptr */
	zbookmark_t	l2rcb_zb;		/* original bookmark */
	int		l2rcb_flags;		/* original flags */
} l2arc_read_callback_t;

typedef struct l2arc_write_callback {
	l2arc_dev_t	*l2wcb_dev;		/* device info */
	arc_buf_hdr_t	*l2wcb_head;		/* head of write buflist */
} l2arc_write_callback_t;

struct l2arc_buf_hdr {
	/* protected by arc_buf_hdr  mutex */
	l2arc_dev_t	*b_dev;			/* L2ARC device */
	uint64_t	b_daddr;		/* disk address, offset byte */
};

typedef struct l2arc_data_free {
	/* protected by l2arc_free_on_write_mtx */
	void		*l2df_data;
	size_t		l2df_size;
	void		(*l2df_func)(void *, size_t);
	list_node_t	l2df_list_node;
} l2arc_data_free_t;

static kmutex_t l2arc_feed_thr_lock;
static kcondvar_t l2arc_feed_thr_cv;
static uint8_t l2arc_thread_exit;

static void l2arc_read_done(zio_t *zio);
static void l2arc_hdr_stat_add(void);
static void l2arc_hdr_stat_remove(void);

static uint64_t
buf_hash(uint64_t spa, const dva_t *dva, uint64_t birth)
{
	uint8_t *vdva = (uint8_t *)dva;
	uint64_t crc = -1ULL;
	int i;

	ASSERT(zfs_crc64_table[128] == ZFS_CRC64_POLY);

	for (i = 0; i < sizeof (dva_t); i++)
		crc = (crc >> 8) ^ zfs_crc64_table[(crc ^ vdva[i]) & 0xFF];

	crc ^= (spa>>8) ^ birth;

	return (crc);
}

#define	BUF_EMPTY(buf)						\
	((buf)->b_dva.dva_word[0] == 0 &&			\
	(buf)->b_dva.dva_word[1] == 0 &&			\
	(buf)->b_birth == 0)

#define	BUF_EQUAL(spa, dva, birth, buf)				\
	((buf)->b_dva.dva_word[0] == (dva)->dva_word[0]) &&	\
	((buf)->b_dva.dva_word[1] == (dva)->dva_word[1]) &&	\
	((buf)->b_birth == birth) && ((buf)->b_spa == spa)

static void
buf_discard_identity(arc_buf_hdr_t *hdr)
{
	hdr->b_dva.dva_word[0] = 0;
	hdr->b_dva.dva_word[1] = 0;
	hdr->b_birth = 0;
	hdr->b_cksum0 = 0;
}

static arc_buf_hdr_t *
buf_hash_find(uint64_t spa, const dva_t *dva, uint64_t birth, kmutex_t **lockp)
{
	uint64_t idx = BUF_HASH_INDEX(spa, dva, birth);
	kmutex_t *hash_lock = BUF_HASH_LOCK(idx);
	arc_buf_hdr_t *buf;

	mutex_enter(hash_lock);
	for (buf = buf_hash_table.ht_table[idx]; buf != NULL;
	    buf = buf->b_hash_next) {
		if (BUF_EQUAL(spa, dva, birth, buf)) {
			*lockp = hash_lock;
			return (buf);
		}
	}
	mutex_exit(hash_lock);
	*lockp = NULL;
	return (NULL);
}

/*
 * Insert an entry into the hash table.  If there is already an element
 * equal to elem in the hash table, then the already existing element
 * will be returned and the new element will not be inserted.
 * Otherwise returns NULL.
 */
static arc_buf_hdr_t *
buf_hash_insert(arc_buf_hdr_t *buf, kmutex_t **lockp)
{
	uint64_t idx = BUF_HASH_INDEX(buf->b_spa, &buf->b_dva, buf->b_birth);
	kmutex_t *hash_lock = BUF_HASH_LOCK(idx);
	arc_buf_hdr_t *fbuf;
	uint32_t i;

	ASSERT(!HDR_IN_HASH_TABLE(buf));
	*lockp = hash_lock;
	mutex_enter(hash_lock);
	for (fbuf = buf_hash_table.ht_table[idx], i = 0; fbuf != NULL;
	    fbuf = fbuf->b_hash_next, i++) {
		if (BUF_EQUAL(buf->b_spa, &buf->b_dva, buf->b_birth, fbuf))
			return (fbuf);
	}

	buf->b_hash_next = buf_hash_table.ht_table[idx];
	buf_hash_table.ht_table[idx] = buf;
	buf->b_flags |= ARC_IN_HASH_TABLE;

	/* collect some hash table performance data */
	if (i > 0) {
		ARCSTAT_BUMP(arcstat_hash_collisions);
		if (i == 1)
			ARCSTAT_BUMP(arcstat_hash_chains);

		ARCSTAT_MAX(arcstat_hash_chain_max, i);
	}

	ARCSTAT_BUMP(arcstat_hash_elements);
	ARCSTAT_MAXSTAT(arcstat_hash_elements);

	return (NULL);
}

static void
buf_hash_remove(arc_buf_hdr_t *buf)
{
	arc_buf_hdr_t *fbuf, **bufp;
	uint64_t idx = BUF_HASH_INDEX(buf->b_spa, &buf->b_dva, buf->b_birth);

	ASSERT(MUTEX_HELD(BUF_HASH_LOCK(idx)));
	ASSERT(HDR_IN_HASH_TABLE(buf));

	bufp = &buf_hash_table.ht_table[idx];
	while ((fbuf = *bufp) != buf) {
		ASSERT(fbuf != NULL);
		bufp = &fbuf->b_hash_next;
	}
	*bufp = buf->b_hash_next;
	buf->b_hash_next = NULL;
	buf->b_flags &= ~ARC_IN_HASH_TABLE;

	/* collect some hash table performance data */
	ARCSTAT_BUMPDOWN(arcstat_hash_elements);

	if (buf_hash_table.ht_table[idx] &&
	    buf_hash_table.ht_table[idx]->b_hash_next == NULL)
		ARCSTAT_BUMPDOWN(arcstat_hash_chains);
}

/*
 * Global data structures and functions for the buf kmem cache.
 */
static kmem_cache_t *hdr_cache;
static kmem_cache_t *buf_cache;

static void
buf_fini(void)
{
	int i;

	kmem_free(buf_hash_table.ht_table,
	    (buf_hash_table.ht_mask + 1) * sizeof (void *));
	for (i = 0; i < BUF_LOCKS; i++)
		mutex_destroy(&buf_hash_table.ht_locks[i].ht_lock);
	kmem_cache_destroy(hdr_cache);
	kmem_cache_destroy(buf_cache);
}

/*
 * Constructor callback - called when the cache is empty
 * and a new buf is requested.
 */
/* ARGSUSED */
static int
hdr_cons(void *vbuf, void *unused, int kmflag)
{
	arc_buf_hdr_t *buf = vbuf;

	bzero(buf, sizeof (arc_buf_hdr_t));
	refcount_create(&buf->b_refcnt);
	cv_init(&buf->b_cv, NULL, CV_DEFAULT, NULL);
	mutex_init(&buf->b_freeze_lock, NULL, MUTEX_DEFAULT, NULL);
	arc_space_consume(sizeof (arc_buf_hdr_t), ARC_SPACE_HDRS);

	return (0);
}

/* ARGSUSED */
static int
buf_cons(void *vbuf, void *unused, int kmflag)
{
	arc_buf_t *buf = vbuf;

	bzero(buf, sizeof (arc_buf_t));
	mutex_init(&buf->b_evict_lock, NULL, MUTEX_DEFAULT, NULL);
	rw_init(&buf->b_data_lock, NULL, RW_DEFAULT, NULL);
	arc_space_consume(sizeof (arc_buf_t), ARC_SPACE_HDRS);

	return (0);
}

/*
 * Destructor callback - called when a cached buf is
 * no longer required.
 */
/* ARGSUSED */
static void
hdr_dest(void *vbuf, void *unused)
{
	arc_buf_hdr_t *buf = vbuf;

	ASSERT(BUF_EMPTY(buf));
	refcount_destroy(&buf->b_refcnt);
	cv_destroy(&buf->b_cv);
	mutex_destroy(&buf->b_freeze_lock);
	arc_space_return(sizeof (arc_buf_hdr_t), ARC_SPACE_HDRS);
}

/* ARGSUSED */
static void
buf_dest(void *vbuf, void *unused)
{
	arc_buf_t *buf = vbuf;

	mutex_destroy(&buf->b_evict_lock);
	rw_destroy(&buf->b_data_lock);
	arc_space_return(sizeof (arc_buf_t), ARC_SPACE_HDRS);
}

/*
 * Reclaim callback -- invoked when memory is low.
 */
/* ARGSUSED */
static void
hdr_recl(void *unused)
{
	dprintf("hdr_recl called\n");
	/*
	 * umem calls the reclaim func when we destroy the buf cache,
	 * which is after we do arc_fini().
	 */
	if (!arc_dead)
		cv_signal(&arc_reclaim_thr_cv);
}

static void
buf_init(void)
{
	uint64_t *ct;
	uint64_t hsize = 1ULL << 12;
	int i, j;

	/*
	 * The hash table is big enough to fill all of physical memory
	 * with an average 64K block size.  The table will take up
	 * totalmem*sizeof(void*)/64K (eg. 128KB/GB with 8-byte pointers).
	 */
	while (hsize * 65536 < physmem * PAGESIZE)
		hsize <<= 1;
retry:
	buf_hash_table.ht_mask = hsize - 1;
	buf_hash_table.ht_table =
	    kmem_zalloc(hsize * sizeof (void*), KM_NOSLEEP);
	if (buf_hash_table.ht_table == NULL) {
		ASSERT(hsize > (1ULL << 8));
		hsize >>= 1;
		goto retry;
	}

	hdr_cache = kmem_cache_create("arc_buf_hdr_t", sizeof (arc_buf_hdr_t),
	    0, hdr_cons, hdr_dest, hdr_recl, NULL, NULL, 0);
	buf_cache = kmem_cache_create("arc_buf_t", sizeof (arc_buf_t),
	    0, buf_cons, buf_dest, NULL, NULL, NULL, 0);

	for (i = 0; i < 256; i++)
		for (ct = zfs_crc64_table + i, *ct = i, j = 8; j > 0; j--)
			*ct = (*ct >> 1) ^ (-(*ct & 1) & ZFS_CRC64_POLY);

	for (i = 0; i < BUF_LOCKS; i++) {
		mutex_init(&buf_hash_table.ht_locks[i].ht_lock,
		    NULL, MUTEX_DEFAULT, NULL);
	}
}

#define	ARC_MINTIME	(hz>>4) /* 62 ms */

static void
arc_cksum_verify(arc_buf_t *buf)
{
	zio_cksum_t zc;

	if (!(zfs_flags & ZFS_DEBUG_MODIFY))
		return;

	mutex_enter(&buf->b_hdr->b_freeze_lock);
	if (buf->b_hdr->b_freeze_cksum == NULL ||
	    (buf->b_hdr->b_flags & ARC_IO_ERROR)) {
		mutex_exit(&buf->b_hdr->b_freeze_lock);
		return;
	}
	fletcher_2_native(buf->b_data, buf->b_hdr->b_size, &zc);
	if (!ZIO_CHECKSUM_EQUAL(*buf->b_hdr->b_freeze_cksum, zc))
		panic("buffer modified while frozen!");
	mutex_exit(&buf->b_hdr->b_freeze_lock);
}

static int
arc_cksum_equal(arc_buf_t *buf)
{
	zio_cksum_t zc;
	int equal;

	mutex_enter(&buf->b_hdr->b_freeze_lock);
	fletcher_2_native(buf->b_data, buf->b_hdr->b_size, &zc);
	equal = ZIO_CHECKSUM_EQUAL(*buf->b_hdr->b_freeze_cksum, zc);
	mutex_exit(&buf->b_hdr->b_freeze_lock);

	return (equal);
}

static void
arc_cksum_compute(arc_buf_t *buf, boolean_t force)
{
	if (!force && !(zfs_flags & ZFS_DEBUG_MODIFY))
		return;

	mutex_enter(&buf->b_hdr->b_freeze_lock);
	if (buf->b_hdr->b_freeze_cksum != NULL) {
		mutex_exit(&buf->b_hdr->b_freeze_lock);
		return;
	}
	buf->b_hdr->b_freeze_cksum = kmem_alloc(sizeof (zio_cksum_t), KM_SLEEP);
	fletcher_2_native(buf->b_data, buf->b_hdr->b_size,
	    buf->b_hdr->b_freeze_cksum);
	mutex_exit(&buf->b_hdr->b_freeze_lock);
}

void
arc_buf_thaw(arc_buf_t *buf)
{
	if (zfs_flags & ZFS_DEBUG_MODIFY) {
		if (buf->b_hdr->b_state != arc_anon)
			panic("modifying non-anon buffer!");
		if (buf->b_hdr->b_flags & ARC_IO_IN_PROGRESS)
			panic("modifying buffer while i/o in progress!");
		arc_cksum_verify(buf);
	}

	mutex_enter(&buf->b_hdr->b_freeze_lock);
	if (buf->b_hdr->b_freeze_cksum != NULL) {
		kmem_free(buf->b_hdr->b_freeze_cksum, sizeof (zio_cksum_t));
		buf->b_hdr->b_freeze_cksum = NULL;
	}

	if (zfs_flags & ZFS_DEBUG_MODIFY) {
		if (buf->b_hdr->b_thawed)
			kmem_free(buf->b_hdr->b_thawed, 1);
		buf->b_hdr->b_thawed = kmem_alloc(1, KM_SLEEP);
	}

	mutex_exit(&buf->b_hdr->b_freeze_lock);
}

void
arc_buf_freeze(arc_buf_t *buf)
{
	kmutex_t *hash_lock;

	if (!(zfs_flags & ZFS_DEBUG_MODIFY))
		return;

	hash_lock = HDR_LOCK(buf->b_hdr);
	mutex_enter(hash_lock);

	ASSERT(buf->b_hdr->b_freeze_cksum != NULL ||
	    buf->b_hdr->b_state == arc_anon);
	arc_cksum_compute(buf, B_FALSE);
	mutex_exit(hash_lock);
}

static void
add_reference(arc_buf_hdr_t *ab, kmutex_t *hash_lock, void *tag)
{
	ASSERT(MUTEX_HELD(hash_lock));

	if ((refcount_add(&ab->b_refcnt, tag) == 1) &&
	    (ab->b_state != arc_anon)) {
		uint64_t delta = ab->b_size * ab->b_datacnt;
		list_t *list = &ab->b_state->arcs_list[ab->b_type];
		uint64_t *size = &ab->b_state->arcs_lsize[ab->b_type];

		ASSERT(!MUTEX_HELD(&ab->b_state->arcs_mtx));
		mutex_enter(&ab->b_state->arcs_mtx);
		ASSERT(list_link_active(&ab->b_arc_node));
		list_remove(list, ab);
		if (GHOST_STATE(ab->b_state)) {
			ASSERT3U(ab->b_datacnt, ==, 0);
			ASSERT3P(ab->b_buf, ==, NULL);
			delta = ab->b_size;
		}
		ASSERT(delta > 0);
		ASSERT3U(*size, >=, delta);
		atomic_add_64(size, -delta);
		mutex_exit(&ab->b_state->arcs_mtx);
		/* remove the prefetch flag if we get a reference */
		if (ab->b_flags & ARC_PREFETCH)
			ab->b_flags &= ~ARC_PREFETCH;
	}
}

static int
remove_reference(arc_buf_hdr_t *ab, kmutex_t *hash_lock, void *tag)
{
	int cnt;
	arc_state_t *state = ab->b_state;

	ASSERT(state == arc_anon || MUTEX_HELD(hash_lock));
	ASSERT(!GHOST_STATE(state));

	if (((cnt = refcount_remove(&ab->b_refcnt, tag)) == 0) &&
	    (state != arc_anon)) {
		uint64_t *size = &state->arcs_lsize[ab->b_type];

		ASSERT(!MUTEX_HELD(&state->arcs_mtx));
		mutex_enter(&state->arcs_mtx);
		ASSERT(!list_link_active(&ab->b_arc_node));
		list_insert_head(&state->arcs_list[ab->b_type], ab);
		ASSERT(ab->b_datacnt > 0);
		atomic_add_64(size, ab->b_size * ab->b_datacnt);
		mutex_exit(&state->arcs_mtx);
	}
	return (cnt);
}

/*
 * Move the supplied buffer to the indicated state.  The mutex
 * for the buffer must be held by the caller.
 */
static void
arc_change_state(arc_state_t *new_state, arc_buf_hdr_t *ab, kmutex_t *hash_lock)
{
	arc_state_t *old_state = ab->b_state;
	int64_t refcnt = refcount_count(&ab->b_refcnt);
	uint64_t from_delta, to_delta;

	ASSERT(MUTEX_HELD(hash_lock));
	ASSERT(new_state != old_state);
	ASSERT(refcnt == 0 || ab->b_datacnt > 0);
	ASSERT(ab->b_datacnt == 0 || !GHOST_STATE(new_state));
	ASSERT(ab->b_datacnt <= 1 || old_state != arc_anon);

	from_delta = to_delta = ab->b_datacnt * ab->b_size;

	/*
	 * If this buffer is evictable, transfer it from the
	 * old state list to the new state list.
	 */
	if (refcnt == 0) {
		if (old_state != arc_anon) {
			int use_mutex = !MUTEX_HELD(&old_state->arcs_mtx);
			uint64_t *size = &old_state->arcs_lsize[ab->b_type];

			if (use_mutex)
				mutex_enter(&old_state->arcs_mtx);

			ASSERT(list_link_active(&ab->b_arc_node));
			list_remove(&old_state->arcs_list[ab->b_type], ab);

			/*
			 * If prefetching out of the ghost cache,
			 * we will have a non-zero datacnt.
			 */
			if (GHOST_STATE(old_state) && ab->b_datacnt == 0) {
				/* ghost elements have a ghost size */
				ASSERT(ab->b_buf == NULL);
				from_delta = ab->b_size;
			}
			ASSERT3U(*size, >=, from_delta);
			atomic_add_64(size, -from_delta);

			if (use_mutex)
				mutex_exit(&old_state->arcs_mtx);
		}
		if (new_state != arc_anon) {
			int use_mutex = !MUTEX_HELD(&new_state->arcs_mtx);
			uint64_t *size = &new_state->arcs_lsize[ab->b_type];

			if (use_mutex)
				mutex_enter(&new_state->arcs_mtx);

			list_insert_head(&new_state->arcs_list[ab->b_type], ab);

			/* ghost elements have a ghost size */
			if (GHOST_STATE(new_state)) {
				ASSERT(ab->b_datacnt == 0);
				ASSERT(ab->b_buf == NULL);
				to_delta = ab->b_size;
			}
			atomic_add_64(size, to_delta);

			if (use_mutex)
				mutex_exit(&new_state->arcs_mtx);
		}
	}

	ASSERT(!BUF_EMPTY(ab));
	if (new_state == arc_anon && HDR_IN_HASH_TABLE(ab))
		buf_hash_remove(ab);

	/* adjust state sizes */
	if (to_delta)
		atomic_add_64(&new_state->arcs_size, to_delta);
	if (from_delta) {
		ASSERT3U(old_state->arcs_size, >=, from_delta);
		atomic_add_64(&old_state->arcs_size, -from_delta);
	}
	ab->b_state = new_state;

	/* adjust l2arc hdr stats */
	if (new_state == arc_l2c_only)
		l2arc_hdr_stat_add();
	else if (old_state == arc_l2c_only)
		l2arc_hdr_stat_remove();
}

void
arc_space_consume(uint64_t space, arc_space_type_t type)
{
	ASSERT(type >= 0 && type < ARC_SPACE_NUMTYPES);

	switch (type) {
	default:
		break;
	case ARC_SPACE_DATA:
		ARCSTAT_INCR(arcstat_data_size, space);
		break;
	case ARC_SPACE_OTHER:
		ARCSTAT_INCR(arcstat_other_size, space);
		break;
	case ARC_SPACE_HDRS:
		ARCSTAT_INCR(arcstat_hdr_size, space);
		break;
	case ARC_SPACE_L2HDRS:
		ARCSTAT_INCR(arcstat_l2_hdr_size, space);
		break;
	}

	atomic_add_64(&arc_meta_used, space);
	atomic_add_64(&arc_size, space);
}

void
arc_space_return(uint64_t space, arc_space_type_t type)
{
	ASSERT(type >= 0 && type < ARC_SPACE_NUMTYPES);

	switch (type) {
	default:
		break;
	case ARC_SPACE_DATA:
		ARCSTAT_INCR(arcstat_data_size, -space);
		break;
	case ARC_SPACE_OTHER:
		ARCSTAT_INCR(arcstat_other_size, -space);
		break;
	case ARC_SPACE_HDRS:
		ARCSTAT_INCR(arcstat_hdr_size, -space);
		break;
	case ARC_SPACE_L2HDRS:
		ARCSTAT_INCR(arcstat_l2_hdr_size, -space);
		break;
	}

	ASSERT(arc_meta_used >= space);
	if (arc_meta_max < arc_meta_used)
		arc_meta_max = arc_meta_used;
	atomic_add_64(&arc_meta_used, -space);
	ASSERT(arc_size >= space);
	atomic_add_64(&arc_size, -space);
}

void *
arc_data_buf_alloc(uint64_t size)
{
	if (arc_evict_needed(ARC_BUFC_DATA))
		cv_signal(&arc_reclaim_thr_cv);
	atomic_add_64(&arc_size, size);
	return (zio_data_buf_alloc(size));
}

void
arc_data_buf_free(void *buf, uint64_t size)
{
	zio_data_buf_free(buf, size);
	ASSERT(arc_size >= size);
	atomic_add_64(&arc_size, -size);
}

arc_buf_t *
arc_buf_alloc(spa_t *spa, int size, void *tag, arc_buf_contents_t type)
{
	arc_buf_hdr_t *hdr;
	arc_buf_t *buf;

	ASSERT3U(size, >, 0);
	hdr = kmem_cache_alloc(hdr_cache, KM_PUSHPAGE);
	ASSERT(BUF_EMPTY(hdr));
	hdr->b_size = size;
	hdr->b_type = type;
	hdr->b_spa = spa_guid(spa);
	hdr->b_state = arc_anon;
	hdr->b_arc_access = 0;
	buf = kmem_cache_alloc(buf_cache, KM_PUSHPAGE);
	buf->b_hdr = hdr;
	buf->b_data = NULL;
	buf->b_efunc = NULL;
	buf->b_private = NULL;
	buf->b_next = NULL;
	hdr->b_buf = buf;
	arc_get_data_buf(buf);
	hdr->b_datacnt = 1;
	hdr->b_flags = 0;
	ASSERT(refcount_is_zero(&hdr->b_refcnt));
	(void) refcount_add(&hdr->b_refcnt, tag);

	return (buf);
}

static char *arc_onloan_tag = "onloan";

/*
 * Loan out an anonymous arc buffer. Loaned buffers are not counted as in
 * flight data by arc_tempreserve_space() until they are "returned". Loaned
 * buffers must be returned to the arc before they can be used by the DMU or
 * freed.
 */
arc_buf_t *
arc_loan_buf(spa_t *spa, int size)
{
	arc_buf_t *buf;

	buf = arc_buf_alloc(spa, size, arc_onloan_tag, ARC_BUFC_DATA);

	atomic_add_64(&arc_loaned_bytes, size);
	return (buf);
}

/*
 * Return a loaned arc buffer to the arc.
 */
void
arc_return_buf(arc_buf_t *buf, void *tag)
{
	arc_buf_hdr_t *hdr = buf->b_hdr;

	ASSERT(buf->b_data != NULL);
	(void) refcount_add(&hdr->b_refcnt, tag);
	(void) refcount_remove(&hdr->b_refcnt, arc_onloan_tag);

	atomic_add_64(&arc_loaned_bytes, -hdr->b_size);
}

/* Detach an arc_buf from a dbuf (tag) */
void
arc_loan_inuse_buf(arc_buf_t *buf, void *tag)
{
	arc_buf_hdr_t *hdr;

	ASSERT(buf->b_data != NULL);
	hdr = buf->b_hdr;
	(void) refcount_add(&hdr->b_refcnt, arc_onloan_tag);
	(void) refcount_remove(&hdr->b_refcnt, tag);
	buf->b_efunc = NULL;
	buf->b_private = NULL;

	atomic_add_64(&arc_loaned_bytes, hdr->b_size);
}

static arc_buf_t *
arc_buf_clone(arc_buf_t *from)
{
	arc_buf_t *buf;
	arc_buf_hdr_t *hdr = from->b_hdr;
	uint64_t size = hdr->b_size;

	ASSERT(hdr->b_state != arc_anon);

	buf = kmem_cache_alloc(buf_cache, KM_PUSHPAGE);
	buf->b_hdr = hdr;
	buf->b_data = NULL;
	buf->b_efunc = NULL;
	buf->b_private = NULL;
	buf->b_next = hdr->b_buf;
	hdr->b_buf = buf;
	arc_get_data_buf(buf);
	bcopy(from->b_data, buf->b_data, size);
	hdr->b_datacnt += 1;
	return (buf);
}

void
arc_buf_add_ref(arc_buf_t *buf, void* tag)
{
	arc_buf_hdr_t *hdr;
	kmutex_t *hash_lock;

	/*
	 * Check to see if this buffer is evicted.  Callers
	 * must verify b_data != NULL to know if the add_ref
	 * was successful.
	 */
	mutex_enter(&buf->b_evict_lock);
	if (buf->b_data == NULL) {
		mutex_exit(&buf->b_evict_lock);
		return;
	}
	hash_lock = HDR_LOCK(buf->b_hdr);
	mutex_enter(hash_lock);
	hdr = buf->b_hdr;
	ASSERT3P(hash_lock, ==, HDR_LOCK(hdr));
	mutex_exit(&buf->b_evict_lock);

	ASSERT(hdr->b_state == arc_mru || hdr->b_state == arc_mfu);
	add_reference(hdr, hash_lock, tag);
	DTRACE_PROBE1(arc__hit, arc_buf_hdr_t *, hdr);
	arc_access(hdr, hash_lock);
	mutex_exit(hash_lock);
	ARCSTAT_BUMP(arcstat_hits);
	ARCSTAT_CONDSTAT(!(hdr->b_flags & ARC_PREFETCH),
	    demand, prefetch, hdr->b_type != ARC_BUFC_METADATA,
	    data, metadata, hits);
}

/*
 * Free the arc data buffer.  If it is an l2arc write in progress,
 * the buffer is placed on l2arc_free_on_write to be freed later.
 */
static void
arc_buf_data_free(arc_buf_hdr_t *hdr, void (*free_func)(void *, size_t),
    void *data, size_t size)
{
	if (HDR_L2_WRITING(hdr)) {
		l2arc_data_free_t *df;
		df = kmem_alloc(sizeof (l2arc_data_free_t), KM_SLEEP);
		df->l2df_data = data;
		df->l2df_size = size;
		df->l2df_func = free_func;
		mutex_enter(&l2arc_free_on_write_mtx);
		list_insert_head(l2arc_free_on_write, df);
		mutex_exit(&l2arc_free_on_write_mtx);
		ARCSTAT_BUMP(arcstat_l2_free_on_write);
	} else {
		free_func(data, size);
	}
}

static void
arc_buf_destroy(arc_buf_t *buf, boolean_t recycle, boolean_t all)
{
	arc_buf_t **bufp;

	/* free up data associated with the buf */
	if (buf->b_data) {
		arc_state_t *state = buf->b_hdr->b_state;
		uint64_t size = buf->b_hdr->b_size;
		arc_buf_contents_t type = buf->b_hdr->b_type;

		arc_cksum_verify(buf);

		if (!recycle) {
			if (type == ARC_BUFC_METADATA) {
				arc_buf_data_free(buf->b_hdr, zio_buf_free,
				    buf->b_data, size);
				arc_space_return(size, ARC_SPACE_DATA);
			} else {
				ASSERT(type == ARC_BUFC_DATA);
				arc_buf_data_free(buf->b_hdr,
				    zio_data_buf_free, buf->b_data, size);
				ARCSTAT_INCR(arcstat_data_size, -size);
				atomic_add_64(&arc_size, -size);
			}
		}
		if (list_link_active(&buf->b_hdr->b_arc_node)) {
			uint64_t *cnt = &state->arcs_lsize[type];

			ASSERT(refcount_is_zero(&buf->b_hdr->b_refcnt));
			ASSERT(state != arc_anon);

			ASSERT3U(*cnt, >=, size);
			atomic_add_64(cnt, -size);
		}
		ASSERT3U(state->arcs_size, >=, size);
		atomic_add_64(&state->arcs_size, -size);
		buf->b_data = NULL;
		ASSERT(buf->b_hdr->b_datacnt > 0);
		buf->b_hdr->b_datacnt -= 1;
	}

	/* only remove the buf if requested */
	if (!all)
		return;

	/* remove the buf from the hdr list */
	for (bufp = &buf->b_hdr->b_buf; *bufp != buf; bufp = &(*bufp)->b_next)
		continue;
	*bufp = buf->b_next;
	buf->b_next = NULL;

	ASSERT(buf->b_efunc == NULL);

	/* clean up the buf */
	buf->b_hdr = NULL;
	kmem_cache_free(buf_cache, buf);
}

static void
arc_hdr_destroy(arc_buf_hdr_t *hdr)
{
	l2arc_buf_hdr_t *l2hdr = hdr->b_l2hdr;

	ASSERT(refcount_is_zero(&hdr->b_refcnt));
	ASSERT3P(hdr->b_state, ==, arc_anon);
	ASSERT(!HDR_IO_IN_PROGRESS(hdr));

	if (l2hdr != NULL) {
		boolean_t buflist_held = MUTEX_HELD(&l2arc_buflist_mtx);
		/*
		 * To prevent arc_free() and l2arc_evict() from
		 * attempting to free the same buffer at the same time,
		 * a FREE_IN_PROGRESS flag is given to arc_free() to
		 * give it priority.  l2arc_evict() can't destroy this
		 * header while we are waiting on l2arc_buflist_mtx.
		 *
		 * The hdr may be removed from l2ad_buflist before we
		 * grab l2arc_buflist_mtx, so b_l2hdr is rechecked.
		 */
		if (!buflist_held) {
			mutex_enter(&l2arc_buflist_mtx);
			l2hdr = hdr->b_l2hdr;
		}

		if (l2hdr != NULL) {
			list_remove(l2hdr->b_dev->l2ad_buflist, hdr);
			ARCSTAT_INCR(arcstat_l2_size, -hdr->b_size);
			kmem_free(l2hdr, sizeof (l2arc_buf_hdr_t));
			if (hdr->b_state == arc_l2c_only)
				l2arc_hdr_stat_remove();
			hdr->b_l2hdr = NULL;
		}

		if (!buflist_held)
			mutex_exit(&l2arc_buflist_mtx);
	}

	if (!BUF_EMPTY(hdr)) {
		ASSERT(!HDR_IN_HASH_TABLE(hdr));
		buf_discard_identity(hdr);
	}
	while (hdr->b_buf) {
		arc_buf_t *buf = hdr->b_buf;

		if (buf->b_efunc) {
			mutex_enter(&arc_eviction_mtx);
			mutex_enter(&buf->b_evict_lock);
			ASSERT(buf->b_hdr != NULL);
			arc_buf_destroy(hdr->b_buf, FALSE, FALSE);
			hdr->b_buf = buf->b_next;
			buf->b_hdr = &arc_eviction_hdr;
			buf->b_next = arc_eviction_list;
			arc_eviction_list = buf;
			mutex_exit(&buf->b_evict_lock);
			mutex_exit(&arc_eviction_mtx);
		} else {
			arc_buf_destroy(hdr->b_buf, FALSE, TRUE);
		}
	}
	if (hdr->b_freeze_cksum != NULL) {
		kmem_free(hdr->b_freeze_cksum, sizeof (zio_cksum_t));
		hdr->b_freeze_cksum = NULL;
	}
	if (hdr->b_thawed) {
		kmem_free(hdr->b_thawed, 1);
		hdr->b_thawed = NULL;
	}

	ASSERT(!list_link_active(&hdr->b_arc_node));
	ASSERT3P(hdr->b_hash_next, ==, NULL);
	ASSERT3P(hdr->b_acb, ==, NULL);
	kmem_cache_free(hdr_cache, hdr);
}

void
arc_buf_free(arc_buf_t *buf, void *tag)
{
	arc_buf_hdr_t *hdr = buf->b_hdr;
	int hashed = hdr->b_state != arc_anon;

	ASSERT(buf->b_efunc == NULL);
	ASSERT(buf->b_data != NULL);

	if (hashed) {
		kmutex_t *hash_lock = HDR_LOCK(hdr);

		mutex_enter(hash_lock);
		hdr = buf->b_hdr;
		ASSERT3P(hash_lock, ==, HDR_LOCK(hdr));

		(void) remove_reference(hdr, hash_lock, tag);
		if (hdr->b_datacnt > 1) {
			arc_buf_destroy(buf, FALSE, TRUE);
		} else {
			ASSERT(buf == hdr->b_buf);
			ASSERT(buf->b_efunc == NULL);
			hdr->b_flags |= ARC_BUF_AVAILABLE;
		}
		mutex_exit(hash_lock);
	} else if (HDR_IO_IN_PROGRESS(hdr)) {
		int destroy_hdr;
		/*
		 * We are in the middle of an async write.  Don't destroy
		 * this buffer unless the write completes before we finish
		 * decrementing the reference count.
		 */
		mutex_enter(&arc_eviction_mtx);
		(void) remove_reference(hdr, NULL, tag);
		ASSERT(refcount_is_zero(&hdr->b_refcnt));
		destroy_hdr = !HDR_IO_IN_PROGRESS(hdr);
		mutex_exit(&arc_eviction_mtx);
		if (destroy_hdr)
			arc_hdr_destroy(hdr);
	} else {
		if (remove_reference(hdr, NULL, tag) > 0)
			arc_buf_destroy(buf, FALSE, TRUE);
		else
			arc_hdr_destroy(hdr);
	}
}

int
arc_buf_remove_ref(arc_buf_t *buf, void* tag)
{
	arc_buf_hdr_t *hdr = buf->b_hdr;
	kmutex_t *hash_lock = HDR_LOCK(hdr);
	int no_callback = (buf->b_efunc == NULL);

	if (hdr->b_state == arc_anon) {
		ASSERT(hdr->b_datacnt == 1);
		arc_buf_free(buf, tag);
		return (no_callback);
	}

	mutex_enter(hash_lock);
	hdr = buf->b_hdr;
	ASSERT3P(hash_lock, ==, HDR_LOCK(hdr));
	ASSERT(hdr->b_state != arc_anon);
	ASSERT(buf->b_data != NULL);

	(void) remove_reference(hdr, hash_lock, tag);
	if (hdr->b_datacnt > 1) {
		if (no_callback)
			arc_buf_destroy(buf, FALSE, TRUE);
	} else if (no_callback) {
		ASSERT(hdr->b_buf == buf && buf->b_next == NULL);
		ASSERT(buf->b_efunc == NULL);
		hdr->b_flags |= ARC_BUF_AVAILABLE;
	}
	ASSERT(no_callback || hdr->b_datacnt > 1 ||
	    refcount_is_zero(&hdr->b_refcnt));
	mutex_exit(hash_lock);
	return (no_callback);
}

int
arc_buf_size(arc_buf_t *buf)
{
	return (buf->b_hdr->b_size);
}

/*
 * Evict buffers from list until we've removed the specified number of
 * bytes.  Move the removed buffers to the appropriate evict state.
 * If the recycle flag is set, then attempt to "recycle" a buffer:
 * - look for a buffer to evict that is `bytes' long.
 * - return the data block from this buffer rather than freeing it.
 * This flag is used by callers that are trying to make space for a
 * new buffer in a full arc cache.
 *
 * This function makes a "best effort".  It skips over any buffers
 * it can't get a hash_lock on, and so may not catch all candidates.
 * It may also return without evicting as much space as requested.
 */
static void *
arc_evict(arc_state_t *state, uint64_t spa, int64_t bytes, boolean_t recycle,
    arc_buf_contents_t type)
{
	arc_state_t *evicted_state;
	uint64_t bytes_evicted = 0, skipped = 0, missed = 0;
	arc_buf_hdr_t *ab, *ab_prev = NULL;
	list_t *list = &state->arcs_list[type];
	kmutex_t *hash_lock;
	boolean_t have_lock;
	void *stolen = NULL;

	ASSERT(state == arc_mru || state == arc_mfu);

	evicted_state = (state == arc_mru) ? arc_mru_ghost : arc_mfu_ghost;

	mutex_enter(&state->arcs_mtx);
	mutex_enter(&evicted_state->arcs_mtx);

	for (ab = list_tail(list); ab; ab = ab_prev) {
		ab_prev = list_prev(list, ab);
		/* prefetch buffers have a minimum lifespan */
		if (HDR_IO_IN_PROGRESS(ab) ||
		    (spa && ab->b_spa != spa) ||
		    (ab->b_flags & (ARC_PREFETCH|ARC_INDIRECT) &&
		    ddi_get_lbolt() - ab->b_arc_access <
		    arc_min_prefetch_lifespan)) {
			skipped++;
			continue;
		}
		/* "lookahead" for better eviction candidate */
		if (recycle && ab->b_size != bytes &&
		    ab_prev && ab_prev->b_size == bytes)
			continue;
		hash_lock = HDR_LOCK(ab);
		have_lock = MUTEX_HELD(hash_lock);
		if (have_lock || mutex_tryenter(hash_lock)) {
			ASSERT3U(refcount_count(&ab->b_refcnt), ==, 0);
			ASSERT(ab->b_datacnt > 0);
			while (ab->b_buf) {
				arc_buf_t *buf = ab->b_buf;
				if (!mutex_tryenter(&buf->b_evict_lock)) {
					missed += 1;
					break;
				}
				if (buf->b_data) {
					bytes_evicted += ab->b_size;
					if (recycle && ab->b_type == type &&
					    ab->b_size == bytes &&
					    !HDR_L2_WRITING(ab)) {
						stolen = buf->b_data;
						recycle = FALSE;
					}
				}
				if (buf->b_efunc) {
					mutex_enter(&arc_eviction_mtx);
					arc_buf_destroy(buf,
					    buf->b_data == stolen, FALSE);
					ab->b_buf = buf->b_next;
					buf->b_hdr = &arc_eviction_hdr;
					buf->b_next = arc_eviction_list;
					arc_eviction_list = buf;
					mutex_exit(&arc_eviction_mtx);
					mutex_exit(&buf->b_evict_lock);
				} else {
					mutex_exit(&buf->b_evict_lock);
					arc_buf_destroy(buf,
					    buf->b_data == stolen, TRUE);
				}
			}

			if (ab->b_l2hdr) {
				ARCSTAT_INCR(arcstat_evict_l2_cached,
				    ab->b_size);
			} else {
				if (l2arc_write_eligible(ab->b_spa, ab)) {
					ARCSTAT_INCR(arcstat_evict_l2_eligible,
					    ab->b_size);
				} else {
					ARCSTAT_INCR(
					    arcstat_evict_l2_ineligible,
					    ab->b_size);
				}
			}

			if (ab->b_datacnt == 0) {
				arc_change_state(evicted_state, ab, hash_lock);
				ASSERT(HDR_IN_HASH_TABLE(ab));
				ab->b_flags |= ARC_IN_HASH_TABLE;
				ab->b_flags &= ~ARC_BUF_AVAILABLE;
				DTRACE_PROBE1(arc__evict, arc_buf_hdr_t *, ab);
			}
			if (!have_lock)
				mutex_exit(hash_lock);
			if (bytes >= 0 && bytes_evicted >= bytes)
				break;
		} else {
			missed += 1;
		}
	}

	mutex_exit(&evicted_state->arcs_mtx);
	mutex_exit(&state->arcs_mtx);

	if (bytes_evicted < bytes)
		dprintf("only evicted %lld bytes from %x",
		    (longlong_t)bytes_evicted, state);

	if (skipped)
		ARCSTAT_INCR(arcstat_evict_skip, skipped);

	if (missed)
		ARCSTAT_INCR(arcstat_mutex_miss, missed);

	/*
	 * We have just evicted some date into the ghost state, make
	 * sure we also adjust the ghost state size if necessary.
	 */
	if (arc_no_grow &&
	    arc_mru_ghost->arcs_size + arc_mfu_ghost->arcs_size > arc_c) {
		int64_t mru_over = arc_anon->arcs_size + arc_mru->arcs_size +
		    arc_mru_ghost->arcs_size - arc_c;

		if (mru_over > 0 && arc_mru_ghost->arcs_lsize[type] > 0) {
			int64_t todelete =
			    MIN(arc_mru_ghost->arcs_lsize[type], mru_over);
			arc_evict_ghost(arc_mru_ghost, 0, todelete);
		} else if (arc_mfu_ghost->arcs_lsize[type] > 0) {
			int64_t todelete = MIN(arc_mfu_ghost->arcs_lsize[type],
			    arc_mru_ghost->arcs_size +
			    arc_mfu_ghost->arcs_size - arc_c);
			arc_evict_ghost(arc_mfu_ghost, 0, todelete);
		}
	}

	return (stolen);
}

/*
 * Remove buffers from list until we've removed the specified number of
 * bytes.  Destroy the buffers that are removed.
 */
static void
arc_evict_ghost(arc_state_t *state, uint64_t spa, int64_t bytes)
{
	arc_buf_hdr_t *ab, *ab_prev;
	arc_buf_hdr_t marker;
	list_t *list = &state->arcs_list[ARC_BUFC_DATA];
	kmutex_t *hash_lock;
	uint64_t bytes_deleted = 0;
	uint64_t bufs_skipped = 0;

	ASSERT(GHOST_STATE(state));
	bzero(&marker, sizeof(marker));
top:
	mutex_enter(&state->arcs_mtx);
	for (ab = list_tail(list); ab; ab = ab_prev) {
		ab_prev = list_prev(list, ab);
		if (spa && ab->b_spa != spa)
			continue;

		/* ignore markers */
		if (ab->b_spa == 0)
			continue;

		hash_lock = HDR_LOCK(ab);
		/* caller may be trying to modify this buffer, skip it */
		if (MUTEX_HELD(hash_lock))
			continue;
		if (mutex_tryenter(hash_lock)) {
			ASSERT(!HDR_IO_IN_PROGRESS(ab));
			ASSERT(ab->b_buf == NULL);
			ARCSTAT_BUMP(arcstat_deleted);
			bytes_deleted += ab->b_size;

			if (ab->b_l2hdr != NULL) {
				/*
				 * This buffer is cached on the 2nd Level ARC;
				 * don't destroy the header.
				 */
				arc_change_state(arc_l2c_only, ab, hash_lock);
				mutex_exit(hash_lock);
			} else {
				arc_change_state(arc_anon, ab, hash_lock);
				mutex_exit(hash_lock);
				arc_hdr_destroy(ab);
			}

			DTRACE_PROBE1(arc__delete, arc_buf_hdr_t *, ab);
			if (bytes >= 0 && bytes_deleted >= bytes)
				break;
		} else if (bytes < 0) {
			/*
			 * Insert a list marker and then wait for the
			 * hash lock to become available. Once its
			 * available, restart from where we left off.
			 */
			list_insert_after(list, ab, &marker);
			mutex_exit(&state->arcs_mtx);
			mutex_enter(hash_lock);
			mutex_exit(hash_lock);
			mutex_enter(&state->arcs_mtx);
			ab_prev = list_prev(list, &marker);
			list_remove(list, &marker);
		} else
			bufs_skipped += 1;
	}
	mutex_exit(&state->arcs_mtx);

	if (list == &state->arcs_list[ARC_BUFC_DATA] &&
	    (bytes < 0 || bytes_deleted < bytes)) {
		list = &state->arcs_list[ARC_BUFC_METADATA];
		goto top;
	}

	if (bufs_skipped) {
		ARCSTAT_INCR(arcstat_mutex_miss, bufs_skipped);
		ASSERT(bytes >= 0);
	}

	if (bytes_deleted < bytes)
		dprintf("only deleted %lld bytes from %p",
		    (longlong_t)bytes_deleted, state);
}

static void
arc_adjust(void)
{
	int64_t adjustment, delta;

	/*
	 * Adjust MRU size
	 */

	adjustment = MIN((int64_t)(arc_size - arc_c),
	    (int64_t)(arc_anon->arcs_size + arc_mru->arcs_size + arc_meta_used -
	    arc_p));

	if (adjustment > 0 && arc_mru->arcs_lsize[ARC_BUFC_DATA] > 0) {
		delta = MIN(arc_mru->arcs_lsize[ARC_BUFC_DATA], adjustment);
		(void) arc_evict(arc_mru, 0, delta, FALSE, ARC_BUFC_DATA);
		adjustment -= delta;
	}

	if (adjustment > 0 && arc_mru->arcs_lsize[ARC_BUFC_METADATA] > 0) {
		delta = MIN(arc_mru->arcs_lsize[ARC_BUFC_METADATA], adjustment);
		(void) arc_evict(arc_mru, 0, delta, FALSE,
		    ARC_BUFC_METADATA);
	}

	/*
	 * Adjust MFU size
	 */

	adjustment = arc_size - arc_c;

	if (adjustment > 0 && arc_mfu->arcs_lsize[ARC_BUFC_DATA] > 0) {
		delta = MIN(adjustment, arc_mfu->arcs_lsize[ARC_BUFC_DATA]);
		(void) arc_evict(arc_mfu, 0, delta, FALSE, ARC_BUFC_DATA);
		adjustment -= delta;
	}

	if (adjustment > 0 && arc_mfu->arcs_lsize[ARC_BUFC_METADATA] > 0) {
		int64_t delta = MIN(adjustment,
		    arc_mfu->arcs_lsize[ARC_BUFC_METADATA]);
		(void) arc_evict(arc_mfu, 0, delta, FALSE,
		    ARC_BUFC_METADATA);
	}

	/*
	 * Adjust ghost lists
	 */

	adjustment = arc_mru->arcs_size + arc_mru_ghost->arcs_size - arc_c;

	if (adjustment > 0 && arc_mru_ghost->arcs_size > 0) {
		delta = MIN(arc_mru_ghost->arcs_size, adjustment);
		arc_evict_ghost(arc_mru_ghost, 0, delta);
	}

	adjustment =
	    arc_mru_ghost->arcs_size + arc_mfu_ghost->arcs_size - arc_c;

	if (adjustment > 0 && arc_mfu_ghost->arcs_size > 0) {
		delta = MIN(arc_mfu_ghost->arcs_size, adjustment);
		arc_evict_ghost(arc_mfu_ghost, 0, delta);
	}
}

static void
arc_do_user_evicts(void)
{
	mutex_enter(&arc_eviction_mtx);
	while (arc_eviction_list != NULL) {
		arc_buf_t *buf = arc_eviction_list;
		arc_eviction_list = buf->b_next;
		mutex_enter(&buf->b_evict_lock);
		buf->b_hdr = NULL;
		mutex_exit(&buf->b_evict_lock);
		mutex_exit(&arc_eviction_mtx);

		if (buf->b_efunc != NULL)
			VERIFY(buf->b_efunc(buf) == 0);

		buf->b_efunc = NULL;
		buf->b_private = NULL;
		kmem_cache_free(buf_cache, buf);
		mutex_enter(&arc_eviction_mtx);
	}
	mutex_exit(&arc_eviction_mtx);
}

/*
 * Flush all *evictable* data from the cache for the given spa.
 * NOTE: this will not touch "active" (i.e. referenced) data.
 */
void
arc_flush(spa_t *spa)
{
	uint64_t guid = 0;

	if (spa)
		guid = spa_guid(spa);

	while (list_head(&arc_mru->arcs_list[ARC_BUFC_DATA])) {
		(void) arc_evict(arc_mru, guid, -1, FALSE, ARC_BUFC_DATA);
		if (spa)
			break;
	}
	while (list_head(&arc_mru->arcs_list[ARC_BUFC_METADATA])) {
		(void) arc_evict(arc_mru, guid, -1, FALSE, ARC_BUFC_METADATA);
		if (spa)
			break;
	}
	while (list_head(&arc_mfu->arcs_list[ARC_BUFC_DATA])) {
		(void) arc_evict(arc_mfu, guid, -1, FALSE, ARC_BUFC_DATA);
		if (spa)
			break;
	}
	while (list_head(&arc_mfu->arcs_list[ARC_BUFC_METADATA])) {
		(void) arc_evict(arc_mfu, guid, -1, FALSE, ARC_BUFC_METADATA);
		if (spa)
			break;
	}

	arc_evict_ghost(arc_mru_ghost, guid, -1);
	arc_evict_ghost(arc_mfu_ghost, guid, -1);

	mutex_enter(&arc_reclaim_thr_lock);
	arc_do_user_evicts();
	mutex_exit(&arc_reclaim_thr_lock);
	ASSERT(spa || arc_eviction_list == NULL);
}

void
arc_shrink(void)
{
	if (arc_c > arc_c_min) {
		uint64_t to_free;

#ifdef _KERNEL
		to_free = MAX(arc_c >> arc_shrink_shift, ptob(needfree));
#else
		to_free = arc_c >> arc_shrink_shift;
#endif
		if (arc_c > arc_c_min + to_free)
			atomic_add_64(&arc_c, -to_free);
		else
			arc_c = arc_c_min;

		atomic_add_64(&arc_p, -(arc_p >> arc_shrink_shift));
		if (arc_c > arc_size)
			arc_c = MAX(arc_size, arc_c_min);
		if (arc_p > arc_c)
			arc_p = (arc_c >> 1);
		ASSERT(arc_c >= arc_c_min);
		ASSERT((int64_t)arc_p >= 0);
	}

	if (arc_size > arc_c)
		arc_adjust();
}

static int
arc_reclaim_needed(void)
{
	uint64_t extra;

#ifdef _KERNEL

	if (needfree)
		return (1);

	/*
	 * take 'desfree' extra pages, so we reclaim sooner, rather than later
	 */
	extra = desfree;

	/*
	 * check that we're out of range of the pageout scanner.  It starts to
	 * schedule paging if freemem is less than lotsfree and needfree.
	 * lotsfree is the high-water mark for pageout, and needfree is the
	 * number of needed free pages.  We add extra pages here to make sure
	 * the scanner doesn't start up while we're freeing memory.
	 */
	if (freemem < lotsfree + needfree + extra)
		return (1);

	/*
	 * check to make sure that swapfs has enough space so that anon
	 * reservations can still succeed. anon_resvmem() checks that the
	 * availrmem is greater than swapfs_minfree, and the number of reserved
	 * swap pages.  We also add a bit of extra here just to prevent
	 * circumstances from getting really dire.
	 */
	if (availrmem < swapfs_minfree + swapfs_reserve + extra)
		return (1);

#if defined(__i386)
	/*
	 * If we're on an i386 platform, it's possible that we'll exhaust the
	 * kernel heap space before we ever run out of available physical
	 * memory.  Most checks of the size of the heap_area compare against
	 * tune.t_minarmem, which is the minimum available real memory that we
	 * can have in the system.  However, this is generally fixed at 25 pages
	 * which is so low that it's useless.  In this comparison, we seek to
	 * calculate the total heap-size, and reclaim if more than 3/4ths of the
	 * heap is allocated.  (Or, in the calculation, if less than 1/4th is
	 * free)
	 */
	if (btop(vmem_size(heap_arena, VMEM_FREE)) <
	    (btop(vmem_size(heap_arena, VMEM_FREE | VMEM_ALLOC)) >> 2))
		return (1);
#endif

#else
	if (spa_get_random(100) == 0)
		return (1);
#endif
	return (0);
}

static void
arc_kmem_reap_now(arc_reclaim_strategy_t strat)
{
	size_t			i;
	kmem_cache_t		*prev_cache = NULL;
	kmem_cache_t		*prev_data_cache = NULL;
	extern kmem_cache_t	*zio_buf_cache[];
	extern kmem_cache_t	*zio_data_buf_cache[];

#ifdef _KERNEL
	if (arc_meta_used >= arc_meta_limit) {
		/*
		 * We are exceeding our meta-data cache limit.
		 * Purge some DNLC entries to release holds on meta-data.
		 */
		dnlc_reduce_cache((void *)(uintptr_t)arc_reduce_dnlc_percent);
	}
#if defined(__i386)
	/*
	 * Reclaim unused memory from all kmem caches.
	 */
	kmem_reap();
#endif
#endif

	/*
	 * An aggressive reclamation will shrink the cache size as well as
	 * reap free buffers from the arc kmem caches.
	 */
	if (strat == ARC_RECLAIM_AGGR)
		arc_shrink();

	for (i = 0; i < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT; i++) {
		if (zio_buf_cache[i] != prev_cache) {
			prev_cache = zio_buf_cache[i];
			kmem_cache_reap_now(zio_buf_cache[i]);
		}
		if (zio_data_buf_cache[i] != prev_data_cache) {
			prev_data_cache = zio_data_buf_cache[i];
			kmem_cache_reap_now(zio_data_buf_cache[i]);
		}
	}
	kmem_cache_reap_now(buf_cache);
	kmem_cache_reap_now(hdr_cache);
}

static void
arc_reclaim_thread(void)
{
	clock_t			growtime = 0;
	arc_reclaim_strategy_t	last_reclaim = ARC_RECLAIM_CONS;
	callb_cpr_t		cpr;

	CALLB_CPR_INIT(&cpr, &arc_reclaim_thr_lock, callb_generic_cpr, FTAG);

	mutex_enter(&arc_reclaim_thr_lock);
	while (arc_thread_exit == 0) {
		if (arc_reclaim_needed()) {

			if (arc_no_grow) {
				if (last_reclaim == ARC_RECLAIM_CONS) {
					last_reclaim = ARC_RECLAIM_AGGR;
				} else {
					last_reclaim = ARC_RECLAIM_CONS;
				}
			} else {
				arc_no_grow = TRUE;
				last_reclaim = ARC_RECLAIM_AGGR;
				membar_producer();
			}

			/* reset the growth delay for every reclaim */
			growtime = ddi_get_lbolt() + (arc_grow_retry * hz);

			arc_kmem_reap_now(last_reclaim);
			arc_warm = B_TRUE;

		} else if (arc_no_grow && ddi_get_lbolt() >= growtime) {
			arc_no_grow = FALSE;
		}

		arc_adjust();

		if (arc_eviction_list != NULL)
			arc_do_user_evicts();

		/* block until needed, or one second, whichever is shorter */
		CALLB_CPR_SAFE_BEGIN(&cpr);
		(void) cv_timedwait(&arc_reclaim_thr_cv,
		    &arc_reclaim_thr_lock, (ddi_get_lbolt() + hz));
		CALLB_CPR_SAFE_END(&cpr, &arc_reclaim_thr_lock);
	}

	arc_thread_exit = 0;
	cv_broadcast(&arc_reclaim_thr_cv);
	CALLB_CPR_EXIT(&cpr);		/* drops arc_reclaim_thr_lock */
	thread_exit();
}

/*
 * Adapt arc info given the number of bytes we are trying to add and
 * the state that we are comming from.  This function is only called
 * when we are adding new content to the cache.
 */
static void
arc_adapt(int bytes, arc_state_t *state)
{
	int mult;
	uint64_t arc_p_min = (arc_c >> arc_p_min_shift);

	if (state == arc_l2c_only)
		return;

	ASSERT(bytes > 0);
	/*
	 * Adapt the target size of the MRU list:
	 *	- if we just hit in the MRU ghost list, then increase
	 *	  the target size of the MRU list.
	 *	- if we just hit in the MFU ghost list, then increase
	 *	  the target size of the MFU list by decreasing the
	 *	  target size of the MRU list.
	 */
	if (state == arc_mru_ghost) {
		mult = ((arc_mru_ghost->arcs_size >= arc_mfu_ghost->arcs_size) ?
		    1 : (arc_mfu_ghost->arcs_size/arc_mru_ghost->arcs_size));
		mult = MIN(mult, 10); /* avoid wild arc_p adjustment */

		arc_p = MIN(arc_c - arc_p_min, arc_p + bytes * mult);
	} else if (state == arc_mfu_ghost) {
		uint64_t delta;

		mult = ((arc_mfu_ghost->arcs_size >= arc_mru_ghost->arcs_size) ?
		    1 : (arc_mru_ghost->arcs_size/arc_mfu_ghost->arcs_size));
		mult = MIN(mult, 10);

		delta = MIN(bytes * mult, arc_p);
		arc_p = MAX(arc_p_min, arc_p - delta);
	}
	ASSERT((int64_t)arc_p >= 0);

	if (arc_reclaim_needed()) {
		cv_signal(&arc_reclaim_thr_cv);
		return;
	}

	if (arc_no_grow)
		return;

	if (arc_c >= arc_c_max)
		return;

	/*
	 * If we're within (2 * maxblocksize) bytes of the target
	 * cache size, increment the target cache size
	 */
	if (arc_size > arc_c - (2ULL << SPA_MAXBLOCKSHIFT)) {
		atomic_add_64(&arc_c, (int64_t)bytes);
		if (arc_c > arc_c_max)
			arc_c = arc_c_max;
		else if (state == arc_anon)
			atomic_add_64(&arc_p, (int64_t)bytes);
		if (arc_p > arc_c)
			arc_p = arc_c;
	}
	ASSERT((int64_t)arc_p >= 0);
}

/*
 * Check if the cache has reached its limits and eviction is required
 * prior to insert.
 */
static int
arc_evict_needed(arc_buf_contents_t type)
{
	if (type == ARC_BUFC_METADATA && arc_meta_used >= arc_meta_limit)
		return (1);

#ifdef _KERNEL
	/*
	 * If zio data pages are being allocated out of a separate heap segment,
	 * then enforce that the size of available vmem for this area remains
	 * above about 1/32nd free.
	 */
	if (type == ARC_BUFC_DATA && zio_arena != NULL &&
	    vmem_size(zio_arena, VMEM_FREE) <
	    (vmem_size(zio_arena, VMEM_ALLOC) >> 5))
		return (1);
#endif

	if (arc_reclaim_needed())
		return (1);

	return (arc_size > arc_c);
}

/*
 * The buffer, supplied as the first argument, needs a data block.
 * So, if we are at cache max, determine which cache should be victimized.
 * We have the following cases:
 *
 * 1. Insert for MRU, p > sizeof(arc_anon + arc_mru) ->
 * In this situation if we're out of space, but the resident size of the MFU is
 * under the limit, victimize the MFU cache to satisfy this insertion request.
 *
 * 2. Insert for MRU, p <= sizeof(arc_anon + arc_mru) ->
 * Here, we've used up all of the available space for the MRU, so we need to
 * evict from our own cache instead.  Evict from the set of resident MRU
 * entries.
 *
 * 3. Insert for MFU (c - p) > sizeof(arc_mfu) ->
 * c minus p represents the MFU space in the cache, since p is the size of the
 * cache that is dedicated to the MRU.  In this situation there's still space on
 * the MFU side, so the MRU side needs to be victimized.
 *
 * 4. Insert for MFU (c - p) < sizeof(arc_mfu) ->
 * MFU's resident set is consuming more space than it has been allotted.  In
 * this situation, we must victimize our own cache, the MFU, for this insertion.
 */
static void
arc_get_data_buf(arc_buf_t *buf)
{
	arc_state_t		*state = buf->b_hdr->b_state;
	uint64_t		size = buf->b_hdr->b_size;
	arc_buf_contents_t	type = buf->b_hdr->b_type;

	arc_adapt(size, state);

	/*
	 * We have not yet reached cache maximum size,
	 * just allocate a new buffer.
	 */
	if (!arc_evict_needed(type)) {
		if (type == ARC_BUFC_METADATA) {
			buf->b_data = zio_buf_alloc(size);
			arc_space_consume(size, ARC_SPACE_DATA);
		} else {
			ASSERT(type == ARC_BUFC_DATA);
			buf->b_data = zio_data_buf_alloc(size);
			ARCSTAT_INCR(arcstat_data_size, size);
			atomic_add_64(&arc_size, size);
		}
		goto out;
	}

	/*
	 * If we are prefetching from the mfu ghost list, this buffer
	 * will end up on the mru list; so steal space from there.
	 */
	if (state == arc_mfu_ghost)
		state = buf->b_hdr->b_flags & ARC_PREFETCH ? arc_mru : arc_mfu;
	else if (state == arc_mru_ghost)
		state = arc_mru;

	if (state == arc_mru || state == arc_anon) {
		uint64_t mru_used = arc_anon->arcs_size + arc_mru->arcs_size;
		state = (arc_mfu->arcs_lsize[type] >= size &&
		    arc_p > mru_used) ? arc_mfu : arc_mru;
	} else {
		/* MFU cases */
		uint64_t mfu_space = arc_c - arc_p;
		state =  (arc_mru->arcs_lsize[type] >= size &&
		    mfu_space > arc_mfu->arcs_size) ? arc_mru : arc_mfu;
	}
	if ((buf->b_data = arc_evict(state, 0, size, TRUE, type)) == NULL) {
		if (type == ARC_BUFC_METADATA) {
			buf->b_data = zio_buf_alloc(size);
			arc_space_consume(size, ARC_SPACE_DATA);
		} else {
			ASSERT(type == ARC_BUFC_DATA);
			buf->b_data = zio_data_buf_alloc(size);
			ARCSTAT_INCR(arcstat_data_size, size);
			atomic_add_64(&arc_size, size);
		}
		ARCSTAT_BUMP(arcstat_recycle_miss);
	}
	ASSERT(buf->b_data != NULL);
out:
	/*
	 * Update the state size.  Note that ghost states have a
	 * "ghost size" and so don't need to be updated.
	 */
	if (!GHOST_STATE(buf->b_hdr->b_state)) {
		arc_buf_hdr_t *hdr = buf->b_hdr;

		atomic_add_64(&hdr->b_state->arcs_size, size);
		if (list_link_active(&hdr->b_arc_node)) {
			ASSERT(refcount_is_zero(&hdr->b_refcnt));
			atomic_add_64(&hdr->b_state->arcs_lsize[type], size);
		}
		/*
		 * If we are growing the cache, and we are adding anonymous
		 * data, and we have outgrown arc_p, update arc_p
		 */
		if (arc_size < arc_c && hdr->b_state == arc_anon &&
		    arc_anon->arcs_size + arc_mru->arcs_size > arc_p)
			arc_p = MIN(arc_c, arc_p + size);
	}
}

/*
 * This routine is called whenever a buffer is accessed.
 * NOTE: the hash lock is dropped in this function.
 */
static void
arc_access(arc_buf_hdr_t *buf, kmutex_t *hash_lock)
{
	clock_t now;

	ASSERT(MUTEX_HELD(hash_lock));

	if (buf->b_state == arc_anon) {
		/*
		 * This buffer is not in the cache, and does not
		 * appear in our "ghost" list.  Add the new buffer
		 * to the MRU state.
		 */

		ASSERT(buf->b_arc_access == 0);
		buf->b_arc_access = ddi_get_lbolt();
		DTRACE_PROBE1(new_state__mru, arc_buf_hdr_t *, buf);
		arc_change_state(arc_mru, buf, hash_lock);

	} else if (buf->b_state == arc_mru) {
		now = ddi_get_lbolt();

		/*
		 * If this buffer is here because of a prefetch, then either:
		 * - clear the flag if this is a "referencing" read
		 *   (any subsequent access will bump this into the MFU state).
		 * or
		 * - move the buffer to the head of the list if this is
		 *   another prefetch (to make it less likely to be evicted).
		 */
		if ((buf->b_flags & ARC_PREFETCH) != 0) {
			if (refcount_count(&buf->b_refcnt) == 0) {
				ASSERT(list_link_active(&buf->b_arc_node));
			} else {
				buf->b_flags &= ~ARC_PREFETCH;
				ARCSTAT_BUMP(arcstat_mru_hits);
			}
			buf->b_arc_access = now;
			return;
		}

		/*
		 * This buffer has been "accessed" only once so far,
		 * but it is still in the cache. Move it to the MFU
		 * state.
		 */
		if (now > buf->b_arc_access + ARC_MINTIME) {
			/*
			 * More than 125ms have passed since we
			 * instantiated this buffer.  Move it to the
			 * most frequently used state.
			 */
			buf->b_arc_access = now;
			DTRACE_PROBE1(new_state__mfu, arc_buf_hdr_t *, buf);
			arc_change_state(arc_mfu, buf, hash_lock);
		}
		ARCSTAT_BUMP(arcstat_mru_hits);
	} else if (buf->b_state == arc_mru_ghost) {
		arc_state_t	*new_state;
		/*
		 * This buffer has been "accessed" recently, but
		 * was evicted from the cache.  Move it to the
		 * MFU state.
		 */

		if (buf->b_flags & ARC_PREFETCH) {
			new_state = arc_mru;
			if (refcount_count(&buf->b_refcnt) > 0)
				buf->b_flags &= ~ARC_PREFETCH;
			DTRACE_PROBE1(new_state__mru, arc_buf_hdr_t *, buf);
		} else {
			new_state = arc_mfu;
			DTRACE_PROBE1(new_state__mfu, arc_buf_hdr_t *, buf);
		}

		buf->b_arc_access = ddi_get_lbolt();
		arc_change_state(new_state, buf, hash_lock);

		ARCSTAT_BUMP(arcstat_mru_ghost_hits);
	} else if (buf->b_state == arc_mfu) {
		/*
		 * This buffer has been accessed more than once and is
		 * still in the cache.  Keep it in the MFU state.
		 *
		 * NOTE: an add_reference() that occurred when we did
		 * the arc_read() will have kicked this off the list.
		 * If it was a prefetch, we will explicitly move it to
		 * the head of the list now.
		 */
		if ((buf->b_flags & ARC_PREFETCH) != 0) {
			ASSERT(refcount_count(&buf->b_refcnt) == 0);
			ASSERT(list_link_active(&buf->b_arc_node));
		}
		ARCSTAT_BUMP(arcstat_mfu_hits);
		buf->b_arc_access = ddi_get_lbolt();
	} else if (buf->b_state == arc_mfu_ghost) {
		arc_state_t	*new_state = arc_mfu;
		/*
		 * This buffer has been accessed more than once but has
		 * been evicted from the cache.  Move it back to the
		 * MFU state.
		 */

		if (buf->b_flags & ARC_PREFETCH) {
			/*
			 * This is a prefetch access...
			 * move this block back to the MRU state.
			 */
			ASSERT3U(refcount_count(&buf->b_refcnt), ==, 0);
			new_state = arc_mru;
		}

		buf->b_arc_access = ddi_get_lbolt();
		DTRACE_PROBE1(new_state__mfu, arc_buf_hdr_t *, buf);
		arc_change_state(new_state, buf, hash_lock);

		ARCSTAT_BUMP(arcstat_mfu_ghost_hits);
	} else if (buf->b_state == arc_l2c_only) {
		/*
		 * This buffer is on the 2nd Level ARC.
		 */

		buf->b_arc_access = ddi_get_lbolt();
		DTRACE_PROBE1(new_state__mfu, arc_buf_hdr_t *, buf);
		arc_change_state(arc_mfu, buf, hash_lock);
	} else {
		ASSERT(!"invalid arc state");
	}
}

/* a generic arc_done_func_t which you can use */
/* ARGSUSED */
void
arc_bcopy_func(zio_t *zio, arc_buf_t *buf, void *arg)
{
	if (zio == NULL || zio->io_error == 0)
		bcopy(buf->b_data, arg, buf->b_hdr->b_size);
	VERIFY(arc_buf_remove_ref(buf, arg) == 1);
}

/* a generic arc_done_func_t */
void
arc_getbuf_func(zio_t *zio, arc_buf_t *buf, void *arg)
{
	arc_buf_t **bufp = arg;
	if (zio && zio->io_error) {
		VERIFY(arc_buf_remove_ref(buf, arg) == 1);
		*bufp = NULL;
	} else {
		*bufp = buf;
		ASSERT(buf->b_data);
	}
}

static void
arc_read_done(zio_t *zio)
{
	arc_buf_hdr_t	*hdr, *found;
	arc_buf_t	*buf;
	arc_buf_t	*abuf;	/* buffer we're assigning to callback */
	kmutex_t	*hash_lock;
	arc_callback_t	*callback_list, *acb;
	int		freeable = FALSE;

	buf = zio->io_private;
	hdr = buf->b_hdr;

	/*
	 * The hdr was inserted into hash-table and removed from lists
	 * prior to starting I/O.  We should find this header, since
	 * it's in the hash table, and it should be legit since it's
	 * not possible to evict it during the I/O.  The only possible
	 * reason for it not to be found is if we were freed during the
	 * read.
	 */
	found = buf_hash_find(hdr->b_spa, &hdr->b_dva, hdr->b_birth,
	    &hash_lock);

	ASSERT((found == NULL && HDR_FREED_IN_READ(hdr) && hash_lock == NULL) ||
	    (found == hdr && DVA_EQUAL(&hdr->b_dva, BP_IDENTITY(zio->io_bp))) ||
	    (found == hdr && HDR_L2_READING(hdr)));

	hdr->b_flags &= ~ARC_L2_EVICTED;
	if (l2arc_noprefetch && (hdr->b_flags & ARC_PREFETCH))
		hdr->b_flags &= ~ARC_L2CACHE;

	/* byteswap if necessary */
	callback_list = hdr->b_acb;
	ASSERT(callback_list != NULL);
	if (BP_SHOULD_BYTESWAP(zio->io_bp) && zio->io_error == 0) {
		arc_byteswap_func_t *func = BP_GET_LEVEL(zio->io_bp) > 0 ?
		    byteswap_uint64_array :
		    dmu_ot[BP_GET_TYPE(zio->io_bp)].ot_byteswap;
		func(buf->b_data, hdr->b_size);
	}

	arc_cksum_compute(buf, B_FALSE);

	if (hash_lock && zio->io_error == 0 && hdr->b_state == arc_anon) {
		/*
		 * Only call arc_access on anonymous buffers.  This is because
		 * if we've issued an I/O for an evicted buffer, we've already
		 * called arc_access (to prevent any simultaneous readers from
		 * getting confused).
		 */
		arc_access(hdr, hash_lock);
	}

	/* create copies of the data buffer for the callers */
	abuf = buf;
	for (acb = callback_list; acb; acb = acb->acb_next) {
		if (acb->acb_done) {
			if (abuf == NULL)
				abuf = arc_buf_clone(buf);
			acb->acb_buf = abuf;
			abuf = NULL;
		}
	}
	hdr->b_acb = NULL;
	hdr->b_flags &= ~ARC_IO_IN_PROGRESS;
	ASSERT(!HDR_BUF_AVAILABLE(hdr));
	if (abuf == buf) {
		ASSERT(buf->b_efunc == NULL);
		ASSERT(hdr->b_datacnt == 1);
		hdr->b_flags |= ARC_BUF_AVAILABLE;
	}

	ASSERT(refcount_is_zero(&hdr->b_refcnt) || callback_list != NULL);

	if (zio->io_error != 0) {
		hdr->b_flags |= ARC_IO_ERROR;
		if (hdr->b_state != arc_anon)
			arc_change_state(arc_anon, hdr, hash_lock);
		if (HDR_IN_HASH_TABLE(hdr))
			buf_hash_remove(hdr);
		freeable = refcount_is_zero(&hdr->b_refcnt);
	}

	/*
	 * Broadcast before we drop the hash_lock to avoid the possibility
	 * that the hdr (and hence the cv) might be freed before we get to
	 * the cv_broadcast().
	 */
	cv_broadcast(&hdr->b_cv);

	if (hash_lock) {
		mutex_exit(hash_lock);
	} else {
		/*
		 * This block was freed while we waited for the read to
		 * complete.  It has been removed from the hash table and
		 * moved to the anonymous state (so that it won't show up
		 * in the cache).
		 */
		ASSERT3P(hdr->b_state, ==, arc_anon);
		freeable = refcount_is_zero(&hdr->b_refcnt);
	}

	/* execute each callback and free its structure */
	while ((acb = callback_list) != NULL) {
		if (acb->acb_done)
			acb->acb_done(zio, acb->acb_buf, acb->acb_private);

		if (acb->acb_zio_dummy != NULL) {
			acb->acb_zio_dummy->io_error = zio->io_error;
			zio_nowait(acb->acb_zio_dummy);
		}

		callback_list = acb->acb_next;
		kmem_free(acb, sizeof (arc_callback_t));
	}

	if (freeable)
		arc_hdr_destroy(hdr);
}

/*
 * "Read" the block block at the specified DVA (in bp) via the
 * cache.  If the block is found in the cache, invoke the provided
 * callback immediately and return.  Note that the `zio' parameter
 * in the callback will be NULL in this case, since no IO was
 * required.  If the block is not in the cache pass the read request
 * on to the spa with a substitute callback function, so that the
 * requested block will be added to the cache.
 *
 * If a read request arrives for a block that has a read in-progress,
 * either wait for the in-progress read to complete (and return the
 * results); or, if this is a read with a "done" func, add a record
 * to the read to invoke the "done" func when the read completes,
 * and return; or just return.
 *
 * arc_read_done() will invoke all the requested "done" functions
 * for readers of this block.
 *
 * Normal callers should use arc_read and pass the arc buffer and offset
 * for the bp.  But if you know you don't need locking, you can use
 * arc_read_bp.
 */
int
arc_read(zio_t *pio, spa_t *spa, const blkptr_t *bp, arc_buf_t *pbuf,
    arc_done_func_t *done, void *private, int priority, int zio_flags,
    uint32_t *arc_flags, const zbookmark_t *zb)
{
	int err;

	if (pbuf == NULL) {
		/*
		 * XXX This happens from traverse callback funcs, for
		 * the objset_phys_t block.
		 */
		return (arc_read_nolock(pio, spa, bp, done, private, priority,
		    zio_flags, arc_flags, zb));
	}

	ASSERT(!refcount_is_zero(&pbuf->b_hdr->b_refcnt));
	ASSERT3U((char *)bp - (char *)pbuf->b_data, <, pbuf->b_hdr->b_size);
	rw_enter(&pbuf->b_data_lock, RW_READER);

	err = arc_read_nolock(pio, spa, bp, done, private, priority,
	    zio_flags, arc_flags, zb);
	rw_exit(&pbuf->b_data_lock);

	return (err);
}

int
arc_read_nolock(zio_t *pio, spa_t *spa, const blkptr_t *bp,
    arc_done_func_t *done, void *private, int priority, int zio_flags,
    uint32_t *arc_flags, const zbookmark_t *zb)
{
	arc_buf_hdr_t *hdr;
	arc_buf_t *buf;
	kmutex_t *hash_lock;
	zio_t *rzio;
	uint64_t guid = spa_guid(spa);

top:
	hdr = buf_hash_find(guid, BP_IDENTITY(bp), BP_PHYSICAL_BIRTH(bp),
	    &hash_lock);
	if (hdr && hdr->b_datacnt > 0) {

		*arc_flags |= ARC_CACHED;

		if (HDR_IO_IN_PROGRESS(hdr)) {

			if (*arc_flags & ARC_WAIT) {
				cv_wait(&hdr->b_cv, hash_lock);
				mutex_exit(hash_lock);
				goto top;
			}
			ASSERT(*arc_flags & ARC_NOWAIT);

			if (done) {
				arc_callback_t	*acb = NULL;

				acb = kmem_zalloc(sizeof (arc_callback_t),
				    KM_SLEEP);
				acb->acb_done = done;
				acb->acb_private = private;
				if (pio != NULL)
					acb->acb_zio_dummy = zio_null(pio,
					    spa, NULL, NULL, NULL, zio_flags);

				ASSERT(acb->acb_done != NULL);
				acb->acb_next = hdr->b_acb;
				hdr->b_acb = acb;
				add_reference(hdr, hash_lock, private);
				mutex_exit(hash_lock);
				return (0);
			}
			mutex_exit(hash_lock);
			return (0);
		}

		ASSERT(hdr->b_state == arc_mru || hdr->b_state == arc_mfu);

		if (done) {
			add_reference(hdr, hash_lock, private);
			/*
			 * If this block is already in use, create a new
			 * copy of the data so that we will be guaranteed
			 * that arc_release() will always succeed.
			 */
			buf = hdr->b_buf;
			ASSERT(buf);
			ASSERT(buf->b_data);
			if (HDR_BUF_AVAILABLE(hdr)) {
				ASSERT(buf->b_efunc == NULL);
				hdr->b_flags &= ~ARC_BUF_AVAILABLE;
			} else {
				buf = arc_buf_clone(buf);
			}

		} else if (*arc_flags & ARC_PREFETCH &&
		    refcount_count(&hdr->b_refcnt) == 0) {
			hdr->b_flags |= ARC_PREFETCH;
		}
		DTRACE_PROBE1(arc__hit, arc_buf_hdr_t *, hdr);
		arc_access(hdr, hash_lock);
		if (*arc_flags & ARC_L2CACHE)
			hdr->b_flags |= ARC_L2CACHE;
		mutex_exit(hash_lock);
		ARCSTAT_BUMP(arcstat_hits);
		ARCSTAT_CONDSTAT(!(hdr->b_flags & ARC_PREFETCH),
		    demand, prefetch, hdr->b_type != ARC_BUFC_METADATA,
		    data, metadata, hits);

		if (done)
			done(NULL, buf, private);
	} else {
		uint64_t size = BP_GET_LSIZE(bp);
		arc_callback_t	*acb;
		vdev_t *vd = NULL;
		uint64_t addr;
		boolean_t devw = B_FALSE;

		if (hdr == NULL) {
			/* this block is not in the cache */
			arc_buf_hdr_t	*exists;
			arc_buf_contents_t type = BP_GET_BUFC_TYPE(bp);
			buf = arc_buf_alloc(spa, size, private, type);
			hdr = buf->b_hdr;
			hdr->b_dva = *BP_IDENTITY(bp);
			hdr->b_birth = BP_PHYSICAL_BIRTH(bp);
			hdr->b_cksum0 = bp->blk_cksum.zc_word[0];
			exists = buf_hash_insert(hdr, &hash_lock);
			if (exists) {
				/* somebody beat us to the hash insert */
				mutex_exit(hash_lock);
				buf_discard_identity(hdr);
				(void) arc_buf_remove_ref(buf, private);
				goto top; /* restart the IO request */
			}
			/* if this is a prefetch, we don't have a reference */
			if (*arc_flags & ARC_PREFETCH) {
				(void) remove_reference(hdr, hash_lock,
				    private);
				hdr->b_flags |= ARC_PREFETCH;
			}
			if (*arc_flags & ARC_L2CACHE)
				hdr->b_flags |= ARC_L2CACHE;
			if (BP_GET_LEVEL(bp) > 0)
				hdr->b_flags |= ARC_INDIRECT;
		} else {
			/* this block is in the ghost cache */
			ASSERT(GHOST_STATE(hdr->b_state));
			ASSERT(!HDR_IO_IN_PROGRESS(hdr));
			ASSERT3U(refcount_count(&hdr->b_refcnt), ==, 0);
			ASSERT(hdr->b_buf == NULL);

			/* if this is a prefetch, we don't have a reference */
			if (*arc_flags & ARC_PREFETCH)
				hdr->b_flags |= ARC_PREFETCH;
			else
				add_reference(hdr, hash_lock, private);
			if (*arc_flags & ARC_L2CACHE)
				hdr->b_flags |= ARC_L2CACHE;
			buf = kmem_cache_alloc(buf_cache, KM_PUSHPAGE);
			buf->b_hdr = hdr;
			buf->b_data = NULL;
			buf->b_efunc = NULL;
			buf->b_private = NULL;
			buf->b_next = NULL;
			hdr->b_buf = buf;
			ASSERT(hdr->b_datacnt == 0);
			hdr->b_datacnt = 1;
			arc_get_data_buf(buf);
			arc_access(hdr, hash_lock);
		}

		ASSERT(!GHOST_STATE(hdr->b_state));

		acb = kmem_zalloc(sizeof (arc_callback_t), KM_SLEEP);
		acb->acb_done = done;
		acb->acb_private = private;

		ASSERT(hdr->b_acb == NULL);
		hdr->b_acb = acb;
		hdr->b_flags |= ARC_IO_IN_PROGRESS;

		if (HDR_L2CACHE(hdr) && hdr->b_l2hdr != NULL &&
		    (vd = hdr->b_l2hdr->b_dev->l2ad_vdev) != NULL) {
			devw = hdr->b_l2hdr->b_dev->l2ad_writing;
			addr = hdr->b_l2hdr->b_daddr;
			/*
			 * Lock out device removal.
			 */
			if (vdev_is_dead(vd) ||
			    !spa_config_tryenter(spa, SCL_L2ARC, vd, RW_READER))
				vd = NULL;
		}

		mutex_exit(hash_lock);

		ASSERT3U(hdr->b_size, ==, size);
		DTRACE_PROBE4(arc__miss, arc_buf_hdr_t *, hdr, blkptr_t *, bp,
		    uint64_t, size, zbookmark_t *, zb);
		ARCSTAT_BUMP(arcstat_misses);
		ARCSTAT_CONDSTAT(!(hdr->b_flags & ARC_PREFETCH),
		    demand, prefetch, hdr->b_type != ARC_BUFC_METADATA,
		    data, metadata, misses);

		if (vd != NULL && l2arc_ndev != 0 && !(l2arc_norw && devw)) {
			/*
			 * Read from the L2ARC if the following are true:
			 * 1. The L2ARC vdev was previously cached.
			 * 2. This buffer still has L2ARC metadata.
			 * 3. This buffer isn't currently writing to the L2ARC.
			 * 4. The L2ARC entry wasn't evicted, which may
			 *    also have invalidated the vdev.
			 * 5. This isn't prefetch and l2arc_noprefetch is set.
			 */
			if (hdr->b_l2hdr != NULL &&
			    !HDR_L2_WRITING(hdr) && !HDR_L2_EVICTED(hdr) &&
			    !(l2arc_noprefetch && HDR_PREFETCH(hdr))) {
				l2arc_read_callback_t *cb;

				DTRACE_PROBE1(l2arc__hit, arc_buf_hdr_t *, hdr);
				ARCSTAT_BUMP(arcstat_l2_hits);

				cb = kmem_zalloc(sizeof (l2arc_read_callback_t),
				    KM_SLEEP);
				cb->l2rcb_buf = buf;
				cb->l2rcb_spa = spa;
				cb->l2rcb_bp = *bp;
				cb->l2rcb_zb = *zb;
				cb->l2rcb_flags = zio_flags;

				/*
				 * l2arc read.  The SCL_L2ARC lock will be
				 * released by l2arc_read_done().
				 */
				rzio = zio_read_phys(pio, vd, addr, size,
				    buf->b_data, ZIO_CHECKSUM_OFF,
				    l2arc_read_done, cb, priority, zio_flags |
				    ZIO_FLAG_DONT_CACHE | ZIO_FLAG_CANFAIL |
				    ZIO_FLAG_DONT_PROPAGATE |
				    ZIO_FLAG_DONT_RETRY, B_FALSE);
				DTRACE_PROBE2(l2arc__read, vdev_t *, vd,
				    zio_t *, rzio);
				ARCSTAT_INCR(arcstat_l2_read_bytes, size);

				if (*arc_flags & ARC_NOWAIT) {
					zio_nowait(rzio);
					return (0);
				}

				ASSERT(*arc_flags & ARC_WAIT);
				if (zio_wait(rzio) == 0)
					return (0);

				/* l2arc read error; goto zio_read() */
			} else {
				DTRACE_PROBE1(l2arc__miss,
				    arc_buf_hdr_t *, hdr);
				ARCSTAT_BUMP(arcstat_l2_misses);
				if (HDR_L2_WRITING(hdr))
					ARCSTAT_BUMP(arcstat_l2_rw_clash);
				spa_config_exit(spa, SCL_L2ARC, vd);
			}
		} else {
			if (vd != NULL)
				spa_config_exit(spa, SCL_L2ARC, vd);
			if (l2arc_ndev != 0) {
				DTRACE_PROBE1(l2arc__miss,
				    arc_buf_hdr_t *, hdr);
				ARCSTAT_BUMP(arcstat_l2_misses);
			}
		}

		rzio = zio_read(pio, spa, bp, buf->b_data, size,
		    arc_read_done, buf, priority, zio_flags, zb);

		if (*arc_flags & ARC_WAIT)
			return (zio_wait(rzio));

		ASSERT(*arc_flags & ARC_NOWAIT);
		zio_nowait(rzio);
	}
	return (0);
}

void
arc_set_callback(arc_buf_t *buf, arc_evict_func_t *func, void *private)
{
	ASSERT(buf->b_hdr != NULL);
	ASSERT(buf->b_hdr->b_state != arc_anon);
	ASSERT(!refcount_is_zero(&buf->b_hdr->b_refcnt) || func == NULL);
	ASSERT(buf->b_efunc == NULL);
	ASSERT(!HDR_BUF_AVAILABLE(buf->b_hdr));

	buf->b_efunc = func;
	buf->b_private = private;
}

/*
 * This is used by the DMU to let the ARC know that a buffer is
 * being evicted, so the ARC should clean up.  If this arc buf
 * is not yet in the evicted state, it will be put there.
 */
int
arc_buf_evict(arc_buf_t *buf)
{
	arc_buf_hdr_t *hdr;
	kmutex_t *hash_lock;
	arc_buf_t **bufp;

	mutex_enter(&buf->b_evict_lock);
	hdr = buf->b_hdr;
	if (hdr == NULL) {
		/*
		 * We are in arc_do_user_evicts().
		 */
		ASSERT(buf->b_data == NULL);
		mutex_exit(&buf->b_evict_lock);
		return (0);
	} else if (buf->b_data == NULL) {
		arc_buf_t copy = *buf; /* structure assignment */
		/*
		 * We are on the eviction list; process this buffer now
		 * but let arc_do_user_evicts() do the reaping.
		 */
		buf->b_efunc = NULL;
		mutex_exit(&buf->b_evict_lock);
		VERIFY(copy.b_efunc(&copy) == 0);
		return (1);
	}
	hash_lock = HDR_LOCK(hdr);
	mutex_enter(hash_lock);
	hdr = buf->b_hdr;
	ASSERT3P(hash_lock, ==, HDR_LOCK(hdr));

	ASSERT3U(refcount_count(&hdr->b_refcnt), <, hdr->b_datacnt);
	ASSERT(hdr->b_state == arc_mru || hdr->b_state == arc_mfu);

	/*
	 * Pull this buffer off of the hdr
	 */
	bufp = &hdr->b_buf;
	while (*bufp != buf)
		bufp = &(*bufp)->b_next;
	*bufp = buf->b_next;

	ASSERT(buf->b_data != NULL);
	arc_buf_destroy(buf, FALSE, FALSE);

	if (hdr->b_datacnt == 0) {
		arc_state_t *old_state = hdr->b_state;
		arc_state_t *evicted_state;

		ASSERT(hdr->b_buf == NULL);
		ASSERT(refcount_is_zero(&hdr->b_refcnt));

		evicted_state =
		    (old_state == arc_mru) ? arc_mru_ghost : arc_mfu_ghost;

		mutex_enter(&old_state->arcs_mtx);
		mutex_enter(&evicted_state->arcs_mtx);

		arc_change_state(evicted_state, hdr, hash_lock);
		ASSERT(HDR_IN_HASH_TABLE(hdr));
		hdr->b_flags |= ARC_IN_HASH_TABLE;
		hdr->b_flags &= ~ARC_BUF_AVAILABLE;

		mutex_exit(&evicted_state->arcs_mtx);
		mutex_exit(&old_state->arcs_mtx);
	}
	mutex_exit(hash_lock);
	mutex_exit(&buf->b_evict_lock);

	VERIFY(buf->b_efunc(buf) == 0);
	buf->b_efunc = NULL;
	buf->b_private = NULL;
	buf->b_hdr = NULL;
	buf->b_next = NULL;
	kmem_cache_free(buf_cache, buf);
	return (1);
}

/*
 * Release this buffer from the cache.  This must be done
 * after a read and prior to modifying the buffer contents.
 * If the buffer has more than one reference, we must make
 * a new hdr for the buffer.
 */
void
arc_release(arc_buf_t *buf, void *tag)
{
	arc_buf_hdr_t *hdr;
	kmutex_t *hash_lock = NULL;
	l2arc_buf_hdr_t *l2hdr;
	uint64_t buf_size;

	/*
	 * It would be nice to assert that if it's DMU metadata (level >
	 * 0 || it's the dnode file), then it must be syncing context.
	 * But we don't know that information at this level.
	 */

	mutex_enter(&buf->b_evict_lock);
	hdr = buf->b_hdr;

	/* this buffer is not on any list */
	ASSERT(refcount_count(&hdr->b_refcnt) > 0);

	if (hdr->b_state == arc_anon) {
		/* this buffer is already released */
		ASSERT(buf->b_efunc == NULL);
	} else {
		hash_lock = HDR_LOCK(hdr);
		mutex_enter(hash_lock);
		hdr = buf->b_hdr;
		ASSERT3P(hash_lock, ==, HDR_LOCK(hdr));
	}

	l2hdr = hdr->b_l2hdr;
	if (l2hdr) {
		mutex_enter(&l2arc_buflist_mtx);
		hdr->b_l2hdr = NULL;
		buf_size = hdr->b_size;
	}

	/*
	 * Do we have more than one buf?
	 */
	if (hdr->b_datacnt > 1) {
		arc_buf_hdr_t *nhdr;
		arc_buf_t **bufp;
		uint64_t blksz = hdr->b_size;
		uint64_t spa = hdr->b_spa;
		arc_buf_contents_t type = hdr->b_type;
		uint32_t flags = hdr->b_flags;

		ASSERT(hdr->b_buf != buf || buf->b_next != NULL);
		/*
		 * Pull the data off of this hdr and attach it to
		 * a new anonymous hdr.
		 */
		(void) remove_reference(hdr, hash_lock, tag);
		bufp = &hdr->b_buf;
		while (*bufp != buf)
			bufp = &(*bufp)->b_next;
		*bufp = buf->b_next;
		buf->b_next = NULL;

		ASSERT3U(hdr->b_state->arcs_size, >=, hdr->b_size);
		atomic_add_64(&hdr->b_state->arcs_size, -hdr->b_size);
		if (refcount_is_zero(&hdr->b_refcnt)) {
			uint64_t *size = &hdr->b_state->arcs_lsize[hdr->b_type];
			ASSERT3U(*size, >=, hdr->b_size);
			atomic_add_64(size, -hdr->b_size);
		}
		hdr->b_datacnt -= 1;
		arc_cksum_verify(buf);

		mutex_exit(hash_lock);

		nhdr = kmem_cache_alloc(hdr_cache, KM_PUSHPAGE);
		nhdr->b_size = blksz;
		nhdr->b_spa = spa;
		nhdr->b_type = type;
		nhdr->b_buf = buf;
		nhdr->b_state = arc_anon;
		nhdr->b_arc_access = 0;
		nhdr->b_flags = flags & ARC_L2_WRITING;
		nhdr->b_l2hdr = NULL;
		nhdr->b_datacnt = 1;
		nhdr->b_freeze_cksum = NULL;
		(void) refcount_add(&nhdr->b_refcnt, tag);
		buf->b_hdr = nhdr;
		mutex_exit(&buf->b_evict_lock);
		atomic_add_64(&arc_anon->arcs_size, blksz);
	} else {
		mutex_exit(&buf->b_evict_lock);
		ASSERT(refcount_count(&hdr->b_refcnt) == 1);
		ASSERT(!list_link_active(&hdr->b_arc_node));
		ASSERT(!HDR_IO_IN_PROGRESS(hdr));
		if (hdr->b_state != arc_anon)
			arc_change_state(arc_anon, hdr, hash_lock);
		hdr->b_arc_access = 0;
		if (hash_lock)
			mutex_exit(hash_lock);

		buf_discard_identity(hdr);
		arc_buf_thaw(buf);
	}
	buf->b_efunc = NULL;
	buf->b_private = NULL;

	if (l2hdr) {
		list_remove(l2hdr->b_dev->l2ad_buflist, hdr);
		kmem_free(l2hdr, sizeof (l2arc_buf_hdr_t));
		ARCSTAT_INCR(arcstat_l2_size, -buf_size);
		mutex_exit(&l2arc_buflist_mtx);
	}
}

/*
 * Release this buffer.  If it does not match the provided BP, fill it
 * with that block's contents.
 */
/* ARGSUSED */
int
arc_release_bp(arc_buf_t *buf, void *tag, blkptr_t *bp, spa_t *spa,
    zbookmark_t *zb)
{
	arc_release(buf, tag);
	return (0);
}

int
arc_released(arc_buf_t *buf)
{
	int released;

	mutex_enter(&buf->b_evict_lock);
	released = (buf->b_data != NULL && buf->b_hdr->b_state == arc_anon);
	mutex_exit(&buf->b_evict_lock);
	return (released);
}

int
arc_has_callback(arc_buf_t *buf)
{
	int callback;

	mutex_enter(&buf->b_evict_lock);
	callback = (buf->b_efunc != NULL);
	mutex_exit(&buf->b_evict_lock);
	return (callback);
}

#ifdef ZFS_DEBUG
int
arc_referenced(arc_buf_t *buf)
{
	int referenced;

	mutex_enter(&buf->b_evict_lock);
	referenced = (refcount_count(&buf->b_hdr->b_refcnt));
	mutex_exit(&buf->b_evict_lock);
	return (referenced);
}
#endif

static void
arc_write_ready(zio_t *zio)
{
	arc_write_callback_t *callback = zio->io_private;
	arc_buf_t *buf = callback->awcb_buf;
	arc_buf_hdr_t *hdr = buf->b_hdr;

	ASSERT(!refcount_is_zero(&buf->b_hdr->b_refcnt));
	callback->awcb_ready(zio, buf, callback->awcb_private);

	/*
	 * If the IO is already in progress, then this is a re-write
	 * attempt, so we need to thaw and re-compute the cksum.
	 * It is the responsibility of the callback to handle the
	 * accounting for any re-write attempt.
	 */
	if (HDR_IO_IN_PROGRESS(hdr)) {
		mutex_enter(&hdr->b_freeze_lock);
		if (hdr->b_freeze_cksum != NULL) {
			kmem_free(hdr->b_freeze_cksum, sizeof (zio_cksum_t));
			hdr->b_freeze_cksum = NULL;
		}
		mutex_exit(&hdr->b_freeze_lock);
	}
	arc_cksum_compute(buf, B_FALSE);
	hdr->b_flags |= ARC_IO_IN_PROGRESS;
}

static void
arc_write_done(zio_t *zio)
{
	arc_write_callback_t *callback = zio->io_private;
	arc_buf_t *buf = callback->awcb_buf;
	arc_buf_hdr_t *hdr = buf->b_hdr;

	ASSERT(hdr->b_acb == NULL);

	if (zio->io_error == 0) {
		hdr->b_dva = *BP_IDENTITY(zio->io_bp);
		hdr->b_birth = BP_PHYSICAL_BIRTH(zio->io_bp);
		hdr->b_cksum0 = zio->io_bp->blk_cksum.zc_word[0];
	} else {
		ASSERT(BUF_EMPTY(hdr));
	}

	/*
	 * If the block to be written was all-zero, we may have
	 * compressed it away.  In this case no write was performed
	 * so there will be no dva/birth/checksum.  The buffer must
	 * therefore remain anonymous (and uncached).
	 */
	if (!BUF_EMPTY(hdr)) {
		arc_buf_hdr_t *exists;
		kmutex_t *hash_lock;

		ASSERT(zio->io_error == 0);

		arc_cksum_verify(buf);

		exists = buf_hash_insert(hdr, &hash_lock);
		if (exists) {
			/*
			 * This can only happen if we overwrite for
			 * sync-to-convergence, because we remove
			 * buffers from the hash table when we arc_free().
			 */
			if (zio->io_flags & ZIO_FLAG_IO_REWRITE) {
				if (!BP_EQUAL(&zio->io_bp_orig, zio->io_bp))
					panic("bad overwrite, hdr=%p exists=%p",
					    (void *)hdr, (void *)exists);
				ASSERT(refcount_is_zero(&exists->b_refcnt));
				arc_change_state(arc_anon, exists, hash_lock);
				mutex_exit(hash_lock);
				arc_hdr_destroy(exists);
				exists = buf_hash_insert(hdr, &hash_lock);
				ASSERT3P(exists, ==, NULL);
			} else {
				/* Dedup */
				ASSERT(hdr->b_datacnt == 1);
				ASSERT(hdr->b_state == arc_anon);
				ASSERT(BP_GET_DEDUP(zio->io_bp));
				ASSERT(BP_GET_LEVEL(zio->io_bp) == 0);
			}
		}
		hdr->b_flags &= ~ARC_IO_IN_PROGRESS;
		/* if it's not anon, we are doing a scrub */
		if (!exists && hdr->b_state == arc_anon)
			arc_access(hdr, hash_lock);
		mutex_exit(hash_lock);
	} else {
		hdr->b_flags &= ~ARC_IO_IN_PROGRESS;
	}

	ASSERT(!refcount_is_zero(&hdr->b_refcnt));
	callback->awcb_done(zio, buf, callback->awcb_private);

	kmem_free(callback, sizeof (arc_write_callback_t));
}

zio_t *
arc_write(zio_t *pio, spa_t *spa, uint64_t txg,
    blkptr_t *bp, arc_buf_t *buf, boolean_t l2arc, const zio_prop_t *zp,
    arc_done_func_t *ready, arc_done_func_t *done, void *private,
    int priority, int zio_flags, const zbookmark_t *zb)
{
	arc_buf_hdr_t *hdr = buf->b_hdr;
	arc_write_callback_t *callback;
	zio_t *zio;

	ASSERT(ready != NULL);
	ASSERT(done != NULL);
	ASSERT(!HDR_IO_ERROR(hdr));
	ASSERT((hdr->b_flags & ARC_IO_IN_PROGRESS) == 0);
	ASSERT(hdr->b_acb == NULL);
	if (l2arc)
		hdr->b_flags |= ARC_L2CACHE;
	callback = kmem_zalloc(sizeof (arc_write_callback_t), KM_SLEEP);
	callback->awcb_ready = ready;
	callback->awcb_done = done;
	callback->awcb_private = private;
	callback->awcb_buf = buf;

	zio = zio_write(pio, spa, txg, bp, buf->b_data, hdr->b_size, zp,
	    arc_write_ready, arc_write_done, callback, priority, zio_flags, zb);

	return (zio);
}

static int
arc_memory_throttle(uint64_t reserve, uint64_t inflight_data, uint64_t txg)
{
#ifdef _KERNEL
	uint64_t available_memory = ptob(freemem);
	static uint64_t page_load = 0;
	static uint64_t last_txg = 0;

#if defined(__i386)
	available_memory =
	    MIN(available_memory, vmem_size(heap_arena, VMEM_FREE));
#endif
	if (available_memory >= zfs_write_limit_max)
		return (0);

	if (txg > last_txg) {
		last_txg = txg;
		page_load = 0;
	}
	/*
	 * If we are in pageout, we know that memory is already tight,
	 * the arc is already going to be evicting, so we just want to
	 * continue to let page writes occur as quickly as possible.
	 */
	if (curproc == proc_pageout) {
		if (page_load > MAX(ptob(minfree), available_memory) / 4)
			return (ERESTART);
		/* Note: reserve is inflated, so we deflate */
		page_load += reserve / 8;
		return (0);
	} else if (page_load > 0 && arc_reclaim_needed()) {
		/* memory is low, delay before restarting */
		ARCSTAT_INCR(arcstat_memory_throttle_count, 1);
		return (EAGAIN);
	}
	page_load = 0;

	if (arc_size > arc_c_min) {
		uint64_t evictable_memory =
		    arc_mru->arcs_lsize[ARC_BUFC_DATA] +
		    arc_mru->arcs_lsize[ARC_BUFC_METADATA] +
		    arc_mfu->arcs_lsize[ARC_BUFC_DATA] +
		    arc_mfu->arcs_lsize[ARC_BUFC_METADATA];
		available_memory += MIN(evictable_memory, arc_size - arc_c_min);
	}

	if (inflight_data > available_memory / 4) {
		ARCSTAT_INCR(arcstat_memory_throttle_count, 1);
		return (ERESTART);
	}
#endif
	return (0);
}

void
arc_tempreserve_clear(uint64_t reserve)
{
	atomic_add_64(&arc_tempreserve, -reserve);
	ASSERT((int64_t)arc_tempreserve >= 0);
}

int
arc_tempreserve_space(uint64_t reserve, uint64_t txg)
{
	int error;
	uint64_t anon_size;

#ifdef ZFS_DEBUG
	/*
	 * Once in a while, fail for no reason.  Everything should cope.
	 */
	if (spa_get_random(10000) == 0) {
		dprintf("forcing random failure\n");
		return (ERESTART);
	}
#endif
	if (reserve > arc_c/4 && !arc_no_grow)
		arc_c = MIN(arc_c_max, reserve * 4);
	if (reserve > arc_c)
		return (ENOMEM);

	/*
	 * Don't count loaned bufs as in flight dirty data to prevent long
	 * network delays from blocking transactions that are ready to be
	 * assigned to a txg.
	 */
	anon_size = MAX((int64_t)(arc_anon->arcs_size - arc_loaned_bytes), 0);

	/*
	 * Writes will, almost always, require additional memory allocations
	 * in order to compress/encrypt/etc the data.  We therefor need to
	 * make sure that there is sufficient available memory for this.
	 */
	if ((error = arc_memory_throttle(reserve, anon_size, txg)))
		return (error);

	/*
	 * Throttle writes when the amount of dirty data in the cache
	 * gets too large.  We try to keep the cache less than half full
	 * of dirty blocks so that our sync times don't grow too large.
	 * Note: if two requests come in concurrently, we might let them
	 * both succeed, when one of them should fail.  Not a huge deal.
	 */

	if (reserve + arc_tempreserve + anon_size > arc_c / 2 &&
	    anon_size > arc_c / 4) {
		dprintf("failing, arc_tempreserve=%lluK anon_meta=%lluK "
		    "anon_data=%lluK tempreserve=%lluK arc_c=%lluK\n",
		    arc_tempreserve>>10,
		    arc_anon->arcs_lsize[ARC_BUFC_METADATA]>>10,
		    arc_anon->arcs_lsize[ARC_BUFC_DATA]>>10,
		    reserve>>10, arc_c>>10);
		return (ERESTART);
	}
	atomic_add_64(&arc_tempreserve, reserve);
	return (0);
}

void
arc_init(void)
{
	mutex_init(&arc_reclaim_thr_lock, NULL, MUTEX_DEFAULT, NULL);
	cv_init(&arc_reclaim_thr_cv, NULL, CV_DEFAULT, NULL);

	/* Convert seconds to clock ticks */
	arc_min_prefetch_lifespan = 1 * hz;

	/* Start out with 1/8 of all memory */
	arc_c = physmem * PAGESIZE / 8;

#ifdef _KERNEL
	/*
	 * On architectures where the physical memory can be larger
	 * than the addressable space (intel in 32-bit mode), we may
	 * need to limit the cache to 1/8 of VM size.
	 */
	arc_c = MIN(arc_c, vmem_size(heap_arena, VMEM_ALLOC | VMEM_FREE) / 8);
#endif

	/* set min cache to 1/32 of all memory, or 64MB, whichever is more */
	arc_c_min = MAX(arc_c / 4, 64<<20);
	/* set max to 3/4 of all memory, or all but 1GB, whichever is more */
	if (arc_c * 8 >= 1<<30)
		arc_c_max = (arc_c * 8) - (1<<30);
	else
		arc_c_max = arc_c_min;
	arc_c_max = MAX(arc_c * 6, arc_c_max);

	/*
	 * Allow the tunables to override our calculations if they are
	 * reasonable (ie. over 64MB)
	 */
	if (zfs_arc_max > 64<<20 && zfs_arc_max < physmem * PAGESIZE)
		arc_c_max = zfs_arc_max;
	if (zfs_arc_min > 64<<20 && zfs_arc_min <= arc_c_max)
		arc_c_min = zfs_arc_min;

	arc_c = arc_c_max;
	arc_p = (arc_c >> 1);

	/* limit meta-data to 1/4 of the arc capacity */
	arc_meta_limit = arc_c_max / 4;

	/* Allow the tunable to override if it is reasonable */
	if (zfs_arc_meta_limit > 0 && zfs_arc_meta_limit <= arc_c_max)
		arc_meta_limit = zfs_arc_meta_limit;

	if (arc_c_min < arc_meta_limit / 2 && zfs_arc_min == 0)
		arc_c_min = arc_meta_limit / 2;

	if (zfs_arc_grow_retry > 0)
		arc_grow_retry = zfs_arc_grow_retry;

	if (zfs_arc_shrink_shift > 0)
		arc_shrink_shift = zfs_arc_shrink_shift;

	if (zfs_arc_p_min_shift > 0)
		arc_p_min_shift = zfs_arc_p_min_shift;

	/* if kmem_flags are set, lets try to use less memory */
	if (kmem_debugging())
		arc_c = arc_c / 2;
	if (arc_c < arc_c_min)
		arc_c = arc_c_min;

	arc_anon = &ARC_anon;
	arc_mru = &ARC_mru;
	arc_mru_ghost = &ARC_mru_ghost;
	arc_mfu = &ARC_mfu;
	arc_mfu_ghost = &ARC_mfu_ghost;
	arc_l2c_only = &ARC_l2c_only;
	arc_size = 0;

	mutex_init(&arc_anon->arcs_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&arc_mru->arcs_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&arc_mru_ghost->arcs_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&arc_mfu->arcs_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&arc_mfu_ghost->arcs_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&arc_l2c_only->arcs_mtx, NULL, MUTEX_DEFAULT, NULL);

	list_create(&arc_mru->arcs_list[ARC_BUFC_METADATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mru->arcs_list[ARC_BUFC_DATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mru_ghost->arcs_list[ARC_BUFC_METADATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mru_ghost->arcs_list[ARC_BUFC_DATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mfu->arcs_list[ARC_BUFC_METADATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mfu->arcs_list[ARC_BUFC_DATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mfu_ghost->arcs_list[ARC_BUFC_METADATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_mfu_ghost->arcs_list[ARC_BUFC_DATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_l2c_only->arcs_list[ARC_BUFC_METADATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));
	list_create(&arc_l2c_only->arcs_list[ARC_BUFC_DATA],
	    sizeof (arc_buf_hdr_t), offsetof(arc_buf_hdr_t, b_arc_node));

	buf_init();

	arc_thread_exit = 0;
	arc_eviction_list = NULL;
	mutex_init(&arc_eviction_mtx, NULL, MUTEX_DEFAULT, NULL);
	bzero(&arc_eviction_hdr, sizeof (arc_buf_hdr_t));

	arc_ksp = kstat_create("zfs", 0, "arcstats", "misc", KSTAT_TYPE_NAMED,
	    sizeof (arc_stats) / sizeof (kstat_named_t), KSTAT_FLAG_VIRTUAL);

	if (arc_ksp != NULL) {
		arc_ksp->ks_data = &arc_stats;
		kstat_install(arc_ksp);
	}

	(void) thread_create(NULL, 0, arc_reclaim_thread, NULL, 0, &p0,
	    TS_RUN, minclsyspri);

	arc_dead = FALSE;
	arc_warm = B_FALSE;

	if (zfs_write_limit_max == 0)
		zfs_write_limit_max = ptob(physmem) >> zfs_write_limit_shift;
	else
		zfs_write_limit_shift = 0;
	mutex_init(&zfs_write_limit_lock, NULL, MUTEX_DEFAULT, NULL);
}

void
arc_fini(void)
{
	mutex_enter(&arc_reclaim_thr_lock);
	arc_thread_exit = 1;
	while (arc_thread_exit != 0)
		cv_wait(&arc_reclaim_thr_cv, &arc_reclaim_thr_lock);
	mutex_exit(&arc_reclaim_thr_lock);

	arc_flush(NULL);

	arc_dead = TRUE;

	if (arc_ksp != NULL) {
		kstat_delete(arc_ksp);
		arc_ksp = NULL;
	}

	mutex_destroy(&arc_eviction_mtx);
	mutex_destroy(&arc_reclaim_thr_lock);
	cv_destroy(&arc_reclaim_thr_cv);

	list_destroy(&arc_mru->arcs_list[ARC_BUFC_METADATA]);
	list_destroy(&arc_mru_ghost->arcs_list[ARC_BUFC_METADATA]);
	list_destroy(&arc_mfu->arcs_list[ARC_BUFC_METADATA]);
	list_destroy(&arc_mfu_ghost->arcs_list[ARC_BUFC_METADATA]);
	list_destroy(&arc_mru->arcs_list[ARC_BUFC_DATA]);
	list_destroy(&arc_mru_ghost->arcs_list[ARC_BUFC_DATA]);
	list_destroy(&arc_mfu->arcs_list[ARC_BUFC_DATA]);
	list_destroy(&arc_mfu_ghost->arcs_list[ARC_BUFC_DATA]);

	mutex_destroy(&arc_anon->arcs_mtx);
	mutex_destroy(&arc_mru->arcs_mtx);
	mutex_destroy(&arc_mru_ghost->arcs_mtx);
	mutex_destroy(&arc_mfu->arcs_mtx);
	mutex_destroy(&arc_mfu_ghost->arcs_mtx);
	mutex_destroy(&arc_l2c_only->arcs_mtx);

	mutex_destroy(&zfs_write_limit_lock);

	buf_fini();

	ASSERT(arc_loaned_bytes == 0);
}

/*
 * Level 2 ARC
 *
 * The level 2 ARC (L2ARC) is a cache layer in-between main memory and disk.
 * It uses dedicated storage devices to hold cached data, which are populated
 * using large infrequent writes.  The main role of this cache is to boost
 * the performance of random read workloads.  The intended L2ARC devices
 * include short-stroked disks, solid state disks, and other media with
 * substantially faster read latency than disk.
 *
 *                 +-----------------------+
 *                 |         ARC           |
 *                 +-----------------------+
 *                    |         ^     ^
 *                    |         |     |
 *      l2arc_feed_thread()    arc_read()
 *                    |         |     |
 *                    |  l2arc read   |
 *                    V         |     |
 *               +---------------+    |
 *               |     L2ARC     |    |
 *               +---------------+    |
 *                   |    ^           |
 *          l2arc_write() |           |
 *                   |    |           |
 *                   V    |           |
 *                 +-------+      +-------+
 *                 | vdev  |      | vdev  |
 *                 | cache |      | cache |
 *                 +-------+      +-------+
 *                 +=========+     .-----.
 *                 :  L2ARC  :    |-_____-|
 *                 : devices :    | Disks |
 *                 +=========+    `-_____-'
 *
 * Read requests are satisfied from the following sources, in order:
 *
 *	1) ARC
 *	2) vdev cache of L2ARC devices
 *	3) L2ARC devices
 *	4) vdev cache of disks
 *	5) disks
 *
 * Some L2ARC device types exhibit extremely slow write performance.
 * To accommodate for this there are some significant differences between
 * the L2ARC and traditional cache design:
 *
 * 1. There is no eviction path from the ARC to the L2ARC.  Evictions from
 * the ARC behave as usual, freeing buffers and placing headers on ghost
 * lists.  The ARC does not send buffers to the L2ARC during eviction as
 * this would add inflated write latencies for all ARC memory pressure.
 *
 * 2. The L2ARC attempts to cache data from the ARC before it is evicted.
 * It does this by periodically scanning buffers from the eviction-end of
 * the MFU and MRU ARC lists, copying them to the L2ARC devices if they are
 * not already there.  It scans until a headroom of buffers is satisfied,
 * which itself is a buffer for ARC eviction.  The thread that does this is
 * l2arc_feed_thread(), illustrated below; example sizes are included to
 * provide a better sense of ratio than this diagram:
 *
 *	       head -->                        tail
 *	        +---------------------+----------+
 *	ARC_mfu |:::::#:::::::::::::::|o#o###o###|-->.   # already on L2ARC
 *	        +---------------------+----------+   |   o L2ARC eligible
 *	ARC_mru |:#:::::::::::::::::::|#o#ooo####|-->|   : ARC buffer
 *	        +---------------------+----------+   |
 *	             15.9 Gbytes      ^ 32 Mbytes    |
 *	                           headroom          |
 *	                                      l2arc_feed_thread()
 *	                                             |
 *	                 l2arc write hand <--[oooo]--'
 *	                         |           8 Mbyte
 *	                         |          write max
 *	                         V
 *		  +==============================+
 *	L2ARC dev |####|#|###|###|    |####| ... |
 *	          +==============================+
 *	                     32 Gbytes
 *
 * 3. If an ARC buffer is copied to the L2ARC but then hit instead of
 * evicted, then the L2ARC has cached a buffer much sooner than it probably
 * needed to, potentially wasting L2ARC device bandwidth and storage.  It is
 * safe to say that this is an uncommon case, since buffers at the end of
 * the ARC lists have moved there due to inactivity.
 *
 * 4. If the ARC evicts faster than the L2ARC can maintain a headroom,
 * then the L2ARC simply misses copying some buffers.  This serves as a
 * pressure valve to prevent heavy read workloads from both stalling the ARC
 * with waits and clogging the L2ARC with writes.  This also helps prevent
 * the potential for the L2ARC to churn if it attempts to cache content too
 * quickly, such as during backups of the entire pool.
 *
 * 5. After system boot and before the ARC has filled main memory, there are
 * no evictions from the ARC and so the tails of the ARC_mfu and ARC_mru
 * lists can remain mostly static.  Instead of searching from tail of these
 * lists as pictured, the l2arc_feed_thread() will search from the list heads
 * for eligible buffers, greatly increasing its chance of finding them.
 *
 * The L2ARC device write speed is also boosted during this time so that
 * the L2ARC warms up faster.  Since there have been no ARC evictions yet,
 * there are no L2ARC reads, and no fear of degrading read performance
 * through increased writes.
 *
 * 6. Writes to the L2ARC devices are grouped and sent in-sequence, so that
 * the vdev queue can aggregate them into larger and fewer writes.  Each
 * device is written to in a rotor fashion, sweeping writes through
 * available space then repeating.
 *
 * 7. The L2ARC does not store dirty content.  It never needs to flush
 * write buffers back to disk based storage.
 *
 * 8. If an ARC buffer is written (and dirtied) which also exists in the
 * L2ARC, the now stale L2ARC buffer is immediately dropped.
 *
 * The performance of the L2ARC can be tweaked by a number of tunables, which
 * may be necessary for different workloads:
 *
 *	l2arc_write_max		max write bytes per interval
 *	l2arc_write_boost	extra write bytes during device warmup
 *	l2arc_noprefetch	skip caching prefetched buffers
 *	l2arc_headroom		number of max device writes to precache
 *	l2arc_feed_secs		seconds between L2ARC writing
 *
 * Tunables may be removed or added as future performance improvements are
 * integrated, and also may become zpool properties.
 *
 * There are three key functions that control how the L2ARC warms up:
 *
 *	l2arc_write_eligible()	check if a buffer is eligible to cache
 *	l2arc_write_size()	calculate how much to write
 *	l2arc_write_interval()	calculate sleep delay between writes
 *
 * These three functions determine what to write, how much, and how quickly
 * to send writes.
 */

static boolean_t
l2arc_write_eligible(uint64_t spa_guid, arc_buf_hdr_t *ab)
{
	/*
	 * A buffer is *not* eligible for the L2ARC if it:
	 * 1. belongs to a different spa.
	 * 2. is already cached on the L2ARC.
	 * 3. has an I/O in progress (it may be an incomplete read).
	 * 4. is flagged not eligible (zfs property).
	 */
	if (ab->b_spa != spa_guid || ab->b_l2hdr != NULL ||
	    HDR_IO_IN_PROGRESS(ab) || !HDR_L2CACHE(ab))
		return (B_FALSE);

	return (B_TRUE);
}

static uint64_t
l2arc_write_size(l2arc_dev_t *dev)
{
	uint64_t size;

	size = dev->l2ad_write;

	if (arc_warm == B_FALSE)
		size += dev->l2ad_boost;

	return (size);

}

static clock_t
l2arc_write_interval(clock_t began, uint64_t wanted, uint64_t wrote)
{
	clock_t interval, next, now;

	/*
	 * If the ARC lists are busy, increase our write rate; if the
	 * lists are stale, idle back.  This is achieved by checking
	 * how much we previously wrote - if it was more than half of
	 * what we wanted, schedule the next write much sooner.
	 */
	if (l2arc_feed_again && wrote > (wanted / 2))
		interval = (hz * l2arc_feed_min_ms) / 1000;
	else
		interval = hz * l2arc_feed_secs;

	now = ddi_get_lbolt();
	next = MAX(now, MIN(now + interval, began + interval));

	return (next);
}

static void
l2arc_hdr_stat_add(void)
{
	ARCSTAT_INCR(arcstat_l2_hdr_size, HDR_SIZE + L2HDR_SIZE);
	ARCSTAT_INCR(arcstat_hdr_size, -HDR_SIZE);
}

static void
l2arc_hdr_stat_remove(void)
{
	ARCSTAT_INCR(arcstat_l2_hdr_size, -(HDR_SIZE + L2HDR_SIZE));
	ARCSTAT_INCR(arcstat_hdr_size, HDR_SIZE);
}

/*
 * Cycle through L2ARC devices.  This is how L2ARC load balances.
 * If a device is returned, this also returns holding the spa config lock.
 */
static l2arc_dev_t *
l2arc_dev_get_next(void)
{
	l2arc_dev_t *first, *next = NULL;

	/*
	 * Lock out the removal of spas (spa_namespace_lock), then removal
	 * of cache devices (l2arc_dev_mtx).  Once a device has been selected,
	 * both locks will be dropped and a spa config lock held instead.
	 */
	mutex_enter(&spa_namespace_lock);
	mutex_enter(&l2arc_dev_mtx);

	/* if there are no vdevs, there is nothing to do */
	if (l2arc_ndev == 0)
		goto out;

	first = NULL;
	next = l2arc_dev_last;
	do {
		/* loop around the list looking for a non-faulted vdev */
		if (next == NULL) {
			next = list_head(l2arc_dev_list);
		} else {
			next = list_next(l2arc_dev_list, next);
			if (next == NULL)
				next = list_head(l2arc_dev_list);
		}

		/* if we have come back to the start, bail out */
		if (first == NULL)
			first = next;
		else if (next == first)
			break;

	} while (vdev_is_dead(next->l2ad_vdev));

	/* if we were unable to find any usable vdevs, return NULL */
	if (vdev_is_dead(next->l2ad_vdev))
		next = NULL;

	l2arc_dev_last = next;

out:
	mutex_exit(&l2arc_dev_mtx);

	/*
	 * Grab the config lock to prevent the 'next' device from being
	 * removed while we are writing to it.
	 */
	if (next != NULL)
		spa_config_enter(next->l2ad_spa, SCL_L2ARC, next, RW_READER);
	mutex_exit(&spa_namespace_lock);

	return (next);
}

/*
 * Free buffers that were tagged for destruction.
 */
static void
l2arc_do_free_on_write(void)
{
	list_t *buflist;
	l2arc_data_free_t *df, *df_prev;

	mutex_enter(&l2arc_free_on_write_mtx);
	buflist = l2arc_free_on_write;

	for (df = list_tail(buflist); df; df = df_prev) {
		df_prev = list_prev(buflist, df);
		ASSERT(df->l2df_data != NULL);
		ASSERT(df->l2df_func != NULL);
		df->l2df_func(df->l2df_data, df->l2df_size);
		list_remove(buflist, df);
		kmem_free(df, sizeof (l2arc_data_free_t));
	}

	mutex_exit(&l2arc_free_on_write_mtx);
}

/*
 * A write to a cache device has completed.  Update all headers to allow
 * reads from these buffers to begin.
 */
static void
l2arc_write_done(zio_t *zio)
{
	l2arc_write_callback_t *cb;
	l2arc_dev_t *dev;
	list_t *buflist;
	arc_buf_hdr_t *head, *ab, *ab_prev;
	l2arc_buf_hdr_t *abl2;
	kmutex_t *hash_lock;

	cb = zio->io_private;
	ASSERT(cb != NULL);
	dev = cb->l2wcb_dev;
	ASSERT(dev != NULL);
	head = cb->l2wcb_head;
	ASSERT(head != NULL);
	buflist = dev->l2ad_buflist;
	ASSERT(buflist != NULL);
	DTRACE_PROBE2(l2arc__iodone, zio_t *, zio,
	    l2arc_write_callback_t *, cb);

	if (zio->io_error != 0)
		ARCSTAT_BUMP(arcstat_l2_writes_error);

	mutex_enter(&l2arc_buflist_mtx);

	/*
	 * All writes completed, or an error was hit.
	 */
	for (ab = list_prev(buflist, head); ab; ab = ab_prev) {
		ab_prev = list_prev(buflist, ab);

		hash_lock = HDR_LOCK(ab);
		if (!mutex_tryenter(hash_lock)) {
			/*
			 * This buffer misses out.  It may be in a stage
			 * of eviction.  Its ARC_L2_WRITING flag will be
			 * left set, denying reads to this buffer.
			 */
			ARCSTAT_BUMP(arcstat_l2_writes_hdr_miss);
			continue;
		}

		if (zio->io_error != 0) {
			/*
			 * Error - drop L2ARC entry.
			 */
			list_remove(buflist, ab);
			abl2 = ab->b_l2hdr;
			ab->b_l2hdr = NULL;
			kmem_free(abl2, sizeof (l2arc_buf_hdr_t));
			ARCSTAT_INCR(arcstat_l2_size, -ab->b_size);
		}

		/*
		 * Allow ARC to begin reads to this L2ARC entry.
		 */
		ab->b_flags &= ~ARC_L2_WRITING;

		mutex_exit(hash_lock);
	}

	atomic_inc_64(&l2arc_writes_done);
	list_remove(buflist, head);
	kmem_cache_free(hdr_cache, head);
	mutex_exit(&l2arc_buflist_mtx);

	l2arc_do_free_on_write();

	kmem_free(cb, sizeof (l2arc_write_callback_t));
}

/*
 * A read to a cache device completed.  Validate buffer contents before
 * handing over to the regular ARC routines.
 */
static void
l2arc_read_done(zio_t *zio)
{
	l2arc_read_callback_t *cb;
	arc_buf_hdr_t *hdr;
	arc_buf_t *buf;
	kmutex_t *hash_lock;
	int equal;

	ASSERT(zio->io_vd != NULL);
	ASSERT(zio->io_flags & ZIO_FLAG_DONT_PROPAGATE);

	spa_config_exit(zio->io_spa, SCL_L2ARC, zio->io_vd);

	cb = zio->io_private;
	ASSERT(cb != NULL);
	buf = cb->l2rcb_buf;
	ASSERT(buf != NULL);

	hash_lock = HDR_LOCK(buf->b_hdr);
	mutex_enter(hash_lock);
	hdr = buf->b_hdr;
	ASSERT3P(hash_lock, ==, HDR_LOCK(hdr));

	/*
	 * Check this survived the L2ARC journey.
	 */
	equal = arc_cksum_equal(buf);
	if (equal && zio->io_error == 0 && !HDR_L2_EVICTED(hdr)) {
		mutex_exit(hash_lock);
		zio->io_private = buf;
		zio->io_bp_copy = cb->l2rcb_bp;	/* XXX fix in L2ARC 2.0	*/
		zio->io_bp = &zio->io_bp_copy;	/* XXX fix in L2ARC 2.0	*/
		arc_read_done(zio);
	} else {
		mutex_exit(hash_lock);
		/*
		 * Buffer didn't survive caching.  Increment stats and
		 * reissue to the original storage device.
		 */
		if (zio->io_error != 0) {
			ARCSTAT_BUMP(arcstat_l2_io_error);
		} else {
			zio->io_error = EIO;
		}
		if (!equal)
			ARCSTAT_BUMP(arcstat_l2_cksum_bad);

		/*
		 * If there's no waiter, issue an async i/o to the primary
		 * storage now.  If there *is* a waiter, the caller must
		 * issue the i/o in a context where it's OK to block.
		 */
		if (zio->io_waiter == NULL) {
			zio_t *pio = zio_unique_parent(zio);

			ASSERT(!pio || pio->io_child_type == ZIO_CHILD_LOGICAL);

			zio_nowait(zio_read(pio, cb->l2rcb_spa, &cb->l2rcb_bp,
			    buf->b_data, zio->io_size, arc_read_done, buf,
			    zio->io_priority, cb->l2rcb_flags, &cb->l2rcb_zb));
		}
	}

	kmem_free(cb, sizeof (l2arc_read_callback_t));
}

/*
 * This is the list priority from which the L2ARC will search for pages to
 * cache.  This is used within loops (0..3) to cycle through lists in the
 * desired order.  This order can have a significant effect on cache
 * performance.
 *
 * Currently the metadata lists are hit first, MFU then MRU, followed by
 * the data lists.  This function returns a locked list, and also returns
 * the lock pointer.
 */
static list_t *
l2arc_list_locked(int list_num, kmutex_t **lock)
{
	list_t *list;

	ASSERT(list_num >= 0 && list_num <= 3);

	switch (list_num) {
	case 0:
		list = &arc_mfu->arcs_list[ARC_BUFC_METADATA];
		*lock = &arc_mfu->arcs_mtx;
		break;
	case 1:
		list = &arc_mru->arcs_list[ARC_BUFC_METADATA];
		*lock = &arc_mru->arcs_mtx;
		break;
	case 2:
		list = &arc_mfu->arcs_list[ARC_BUFC_DATA];
		*lock = &arc_mfu->arcs_mtx;
		break;
	case 3:
		list = &arc_mru->arcs_list[ARC_BUFC_DATA];
		*lock = &arc_mru->arcs_mtx;
		break;
	}

	ASSERT(!(MUTEX_HELD(*lock)));
	mutex_enter(*lock);
	return (list);
}

/*
 * Evict buffers from the device write hand to the distance specified in
 * bytes.  This distance may span populated buffers, it may span nothing.
 * This is clearing a region on the L2ARC device ready for writing.
 * If the 'all' boolean is set, every buffer is evicted.
 */
static void
l2arc_evict(l2arc_dev_t *dev, uint64_t distance, boolean_t all)
{
	list_t *buflist;
	l2arc_buf_hdr_t *abl2;
	arc_buf_hdr_t *ab, *ab_prev;
	kmutex_t *hash_lock;
	uint64_t taddr;

	buflist = dev->l2ad_buflist;

	if (buflist == NULL)
		return;

	if (!all && dev->l2ad_first) {
		/*
		 * This is the first sweep through the device.  There is
		 * nothing to evict.
		 */
		return;
	}

	if (dev->l2ad_hand >= (dev->l2ad_end - (2 * distance))) {
		/*
		 * When nearing the end of the device, evict to the end
		 * before the device write hand jumps to the start.
		 */
		taddr = dev->l2ad_end;
	} else {
		taddr = dev->l2ad_hand + distance;
	}
	DTRACE_PROBE4(l2arc__evict, l2arc_dev_t *, dev, list_t *, buflist,
	    uint64_t, taddr, boolean_t, all);

top:
	mutex_enter(&l2arc_buflist_mtx);
	for (ab = list_tail(buflist); ab; ab = ab_prev) {
		ab_prev = list_prev(buflist, ab);

		hash_lock = HDR_LOCK(ab);
		if (!mutex_tryenter(hash_lock)) {
			/*
			 * Missed the hash lock.  Retry.
			 */
			ARCSTAT_BUMP(arcstat_l2_evict_lock_retry);
			mutex_exit(&l2arc_buflist_mtx);
			mutex_enter(hash_lock);
			mutex_exit(hash_lock);
			goto top;
		}

		if (HDR_L2_WRITE_HEAD(ab)) {
			/*
			 * We hit a write head node.  Leave it for
			 * l2arc_write_done().
			 */
			list_remove(buflist, ab);
			mutex_exit(hash_lock);
			continue;
		}

		if (!all && ab->b_l2hdr != NULL &&
		    (ab->b_l2hdr->b_daddr > taddr ||
		    ab->b_l2hdr->b_daddr < dev->l2ad_hand)) {
			/*
			 * We've evicted to the target address,
			 * or the end of the device.
			 */
			mutex_exit(hash_lock);
			break;
		}

		if (HDR_FREE_IN_PROGRESS(ab)) {
			/*
			 * Already on the path to destruction.
			 */
			mutex_exit(hash_lock);
			continue;
		}

		if (ab->b_state == arc_l2c_only) {
			ASSERT(!HDR_L2_READING(ab));
			/*
			 * This doesn't exist in the ARC.  Destroy.
			 * arc_hdr_destroy() will call list_remove()
			 * and decrement arcstat_l2_size.
			 */
			arc_change_state(arc_anon, ab, hash_lock);
			arc_hdr_destroy(ab);
		} else {
			/*
			 * Invalidate issued or about to be issued
			 * reads, since we may be about to write
			 * over this location.
			 */
			if (HDR_L2_READING(ab)) {
				ARCSTAT_BUMP(arcstat_l2_evict_reading);
				ab->b_flags |= ARC_L2_EVICTED;
			}

			/*
			 * Tell ARC this no longer exists in L2ARC.
			 */
			if (ab->b_l2hdr != NULL) {
				abl2 = ab->b_l2hdr;
				ab->b_l2hdr = NULL;
				kmem_free(abl2, sizeof (l2arc_buf_hdr_t));
				ARCSTAT_INCR(arcstat_l2_size, -ab->b_size);
			}
			list_remove(buflist, ab);

			/*
			 * This may have been leftover after a
			 * failed write.
			 */
			ab->b_flags &= ~ARC_L2_WRITING;
		}
		mutex_exit(hash_lock);
	}
	mutex_exit(&l2arc_buflist_mtx);

	vdev_space_update(dev->l2ad_vdev, -(taddr - dev->l2ad_evict), 0, 0);
	dev->l2ad_evict = taddr;
}

/*
 * Find and write ARC buffers to the L2ARC device.
 *
 * An ARC_L2_WRITING flag is set so that the L2ARC buffers are not valid
 * for reading until they have completed writing.
 */
static uint64_t
l2arc_write_buffers(spa_t *spa, l2arc_dev_t *dev, uint64_t target_sz)
{
	arc_buf_hdr_t *ab, *ab_prev, *head;
	l2arc_buf_hdr_t *hdrl2;
	list_t *list;
	uint64_t passed_sz, write_sz, buf_sz, headroom;
	void *buf_data;
	kmutex_t *hash_lock, *list_lock;
	boolean_t have_lock, full;
	l2arc_write_callback_t *cb;
	zio_t *pio, *wzio;
	uint64_t guid = spa_guid(spa);
	int try;

	ASSERT(dev->l2ad_vdev != NULL);

	pio = NULL;
	write_sz = 0;
	full = B_FALSE;
	head = kmem_cache_alloc(hdr_cache, KM_PUSHPAGE);
	head->b_flags |= ARC_L2_WRITE_HEAD;

	/*
	 * Copy buffers for L2ARC writing.
	 */
	mutex_enter(&l2arc_buflist_mtx);
	for (try = 0; try <= 3; try++) {
		list = l2arc_list_locked(try, &list_lock);
		passed_sz = 0;

		/*
		 * L2ARC fast warmup.
		 *
		 * Until the ARC is warm and starts to evict, read from the
		 * head of the ARC lists rather than the tail.
		 */
		headroom = target_sz * l2arc_headroom;
		if (arc_warm == B_FALSE)
			ab = list_head(list);
		else
			ab = list_tail(list);

		for (; ab; ab = ab_prev) {
			if (arc_warm == B_FALSE)
				ab_prev = list_next(list, ab);
			else
				ab_prev = list_prev(list, ab);

			hash_lock = HDR_LOCK(ab);
			have_lock = MUTEX_HELD(hash_lock);
			if (!have_lock && !mutex_tryenter(hash_lock)) {
				/*
				 * Skip this buffer rather than waiting.
				 */
				continue;
			}

			passed_sz += ab->b_size;
			if (passed_sz > headroom) {
				/*
				 * Searched too far.
				 */
				mutex_exit(hash_lock);
				break;
			}

			if (!l2arc_write_eligible(guid, ab)) {
				mutex_exit(hash_lock);
				continue;
			}

			if ((write_sz + ab->b_size) > target_sz) {
				full = B_TRUE;
				mutex_exit(hash_lock);
				break;
			}

			if (pio == NULL) {
				/*
				 * Insert a dummy header on the buflist so
				 * l2arc_write_done() can find where the
				 * write buffers begin without searching.
				 */
				list_insert_head(dev->l2ad_buflist, head);

				cb = kmem_alloc(
				    sizeof (l2arc_write_callback_t), KM_SLEEP);
				cb->l2wcb_dev = dev;
				cb->l2wcb_head = head;
				pio = zio_root(spa, l2arc_write_done, cb,
				    ZIO_FLAG_CANFAIL);
			}

			/*
			 * Create and add a new L2ARC header.
			 */
			hdrl2 = kmem_zalloc(sizeof (l2arc_buf_hdr_t), KM_SLEEP);
			hdrl2->b_dev = dev;
			hdrl2->b_daddr = dev->l2ad_hand;

			ab->b_flags |= ARC_L2_WRITING;
			ab->b_l2hdr = hdrl2;
			list_insert_head(dev->l2ad_buflist, ab);
			buf_data = ab->b_buf->b_data;
			buf_sz = ab->b_size;

			/*
			 * Compute and store the buffer cksum before
			 * writing.  On debug the cksum is verified first.
			 */
			arc_cksum_verify(ab->b_buf);
			arc_cksum_compute(ab->b_buf, B_TRUE);

			mutex_exit(hash_lock);

			wzio = zio_write_phys(pio, dev->l2ad_vdev,
			    dev->l2ad_hand, buf_sz, buf_data, ZIO_CHECKSUM_OFF,
			    NULL, NULL, ZIO_PRIORITY_ASYNC_WRITE,
			    ZIO_FLAG_CANFAIL, B_FALSE);

			DTRACE_PROBE2(l2arc__write, vdev_t *, dev->l2ad_vdev,
			    zio_t *, wzio);
			(void) zio_nowait(wzio);

			/*
			 * Keep the clock hand suitably device-aligned.
			 */
			buf_sz = vdev_psize_to_asize(dev->l2ad_vdev, buf_sz);

			write_sz += buf_sz;
			dev->l2ad_hand += buf_sz;
		}

		mutex_exit(list_lock);

		if (full == B_TRUE)
			break;
	}
	mutex_exit(&l2arc_buflist_mtx);

	if (pio == NULL) {
		ASSERT3U(write_sz, ==, 0);
		kmem_cache_free(hdr_cache, head);
		return (0);
	}

	ASSERT3U(write_sz, <=, target_sz);
	ARCSTAT_BUMP(arcstat_l2_writes_sent);
	ARCSTAT_INCR(arcstat_l2_write_bytes, write_sz);
	ARCSTAT_INCR(arcstat_l2_size, write_sz);
	vdev_space_update(dev->l2ad_vdev, write_sz, 0, 0);

	/*
	 * Bump device hand to the device start if it is approaching the end.
	 * l2arc_evict() will already have evicted ahead for this case.
	 */
	if (dev->l2ad_hand >= (dev->l2ad_end - target_sz)) {
		vdev_space_update(dev->l2ad_vdev,
		    dev->l2ad_end - dev->l2ad_hand, 0, 0);
		dev->l2ad_hand = dev->l2ad_start;
		dev->l2ad_evict = dev->l2ad_start;
		dev->l2ad_first = B_FALSE;
	}

	dev->l2ad_writing = B_TRUE;
	(void) zio_wait(pio);
	dev->l2ad_writing = B_FALSE;

	return (write_sz);
}

/*
 * This thread feeds the L2ARC at regular intervals.  This is the beating
 * heart of the L2ARC.
 */
static void
l2arc_feed_thread(void)
{
	callb_cpr_t cpr;
	l2arc_dev_t *dev;
	spa_t *spa;
	uint64_t size, wrote;
	clock_t begin, next = ddi_get_lbolt();

	CALLB_CPR_INIT(&cpr, &l2arc_feed_thr_lock, callb_generic_cpr, FTAG);

	mutex_enter(&l2arc_feed_thr_lock);

	while (l2arc_thread_exit == 0) {
		CALLB_CPR_SAFE_BEGIN(&cpr);
		(void) cv_timedwait(&l2arc_feed_thr_cv, &l2arc_feed_thr_lock,
		    next);
		CALLB_CPR_SAFE_END(&cpr, &l2arc_feed_thr_lock);
		next = ddi_get_lbolt() + hz;

		/*
		 * Quick check for L2ARC devices.
		 */
		mutex_enter(&l2arc_dev_mtx);
		if (l2arc_ndev == 0) {
			mutex_exit(&l2arc_dev_mtx);
			continue;
		}
		mutex_exit(&l2arc_dev_mtx);
		begin = ddi_get_lbolt();

		/*
		 * This selects the next l2arc device to write to, and in
		 * doing so the next spa to feed from: dev->l2ad_spa.   This
		 * will return NULL if there are now no l2arc devices or if
		 * they are all faulted.
		 *
		 * If a device is returned, its spa's config lock is also
		 * held to prevent device removal.  l2arc_dev_get_next()
		 * will grab and release l2arc_dev_mtx.
		 */
		if ((dev = l2arc_dev_get_next()) == NULL)
			continue;

		spa = dev->l2ad_spa;
		ASSERT(spa != NULL);

		/*
		 * If the pool is read-only then force the feed thread to
		 * sleep a little longer.
		 */
		if (!spa_writeable(spa)) {
			next = ddi_get_lbolt() + 5 * l2arc_feed_secs * hz;
			spa_config_exit(spa, SCL_L2ARC, dev);
			continue;
		}

		/*
		 * Avoid contributing to memory pressure.
		 */
		if (arc_reclaim_needed()) {
			ARCSTAT_BUMP(arcstat_l2_abort_lowmem);
			spa_config_exit(spa, SCL_L2ARC, dev);
			continue;
		}

		ARCSTAT_BUMP(arcstat_l2_feeds);

		size = l2arc_write_size(dev);

		/*
		 * Evict L2ARC buffers that will be overwritten.
		 */
		l2arc_evict(dev, size, B_FALSE);

		/*
		 * Write ARC buffers.
		 */
		wrote = l2arc_write_buffers(spa, dev, size);

		/*
		 * Calculate interval between writes.
		 */
		next = l2arc_write_interval(begin, size, wrote);
		spa_config_exit(spa, SCL_L2ARC, dev);
	}

	l2arc_thread_exit = 0;
	cv_broadcast(&l2arc_feed_thr_cv);
	CALLB_CPR_EXIT(&cpr);		/* drops l2arc_feed_thr_lock */
	thread_exit();
}

boolean_t
l2arc_vdev_present(vdev_t *vd)
{
	l2arc_dev_t *dev;

	mutex_enter(&l2arc_dev_mtx);
	for (dev = list_head(l2arc_dev_list); dev != NULL;
	    dev = list_next(l2arc_dev_list, dev)) {
		if (dev->l2ad_vdev == vd)
			break;
	}
	mutex_exit(&l2arc_dev_mtx);

	return (dev != NULL);
}

/*
 * Add a vdev for use by the L2ARC.  By this point the spa has already
 * validated the vdev and opened it.
 */
void
l2arc_add_vdev(spa_t *spa, vdev_t *vd)
{
	l2arc_dev_t *adddev;

	ASSERT(!l2arc_vdev_present(vd));

	/*
	 * Create a new l2arc device entry.
	 */
	adddev = kmem_zalloc(sizeof (l2arc_dev_t), KM_SLEEP);
	adddev->l2ad_spa = spa;
	adddev->l2ad_vdev = vd;
	adddev->l2ad_write = l2arc_write_max;
	adddev->l2ad_boost = l2arc_write_boost;
	adddev->l2ad_start = VDEV_LABEL_START_SIZE;
	adddev->l2ad_end = VDEV_LABEL_START_SIZE + vdev_get_min_asize(vd);
	adddev->l2ad_hand = adddev->l2ad_start;
	adddev->l2ad_evict = adddev->l2ad_start;
	adddev->l2ad_first = B_TRUE;
	adddev->l2ad_writing = B_FALSE;
	ASSERT3U(adddev->l2ad_write, >, 0);

	/*
	 * This is a list of all ARC buffers that are still valid on the
	 * device.
	 */
	adddev->l2ad_buflist = kmem_zalloc(sizeof (list_t), KM_SLEEP);
	list_create(adddev->l2ad_buflist, sizeof (arc_buf_hdr_t),
	    offsetof(arc_buf_hdr_t, b_l2node));

	vdev_space_update(vd, 0, 0, adddev->l2ad_end - adddev->l2ad_hand);

	/*
	 * Add device to global list
	 */
	mutex_enter(&l2arc_dev_mtx);
	list_insert_head(l2arc_dev_list, adddev);
	atomic_inc_64(&l2arc_ndev);
	mutex_exit(&l2arc_dev_mtx);
}

/*
 * Remove a vdev from the L2ARC.
 */
void
l2arc_remove_vdev(vdev_t *vd)
{
	l2arc_dev_t *dev, *nextdev, *remdev = NULL;

	/*
	 * Find the device by vdev
	 */
	mutex_enter(&l2arc_dev_mtx);
	for (dev = list_head(l2arc_dev_list); dev; dev = nextdev) {
		nextdev = list_next(l2arc_dev_list, dev);
		if (vd == dev->l2ad_vdev) {
			remdev = dev;
			break;
		}
	}
	ASSERT(remdev != NULL);

	/*
	 * Remove device from global list
	 */
	list_remove(l2arc_dev_list, remdev);
	l2arc_dev_last = NULL;		/* may have been invalidated */
	atomic_dec_64(&l2arc_ndev);
	mutex_exit(&l2arc_dev_mtx);

	/*
	 * Clear all buflists and ARC references.  L2ARC device flush.
	 */
	l2arc_evict(remdev, 0, B_TRUE);
	list_destroy(remdev->l2ad_buflist);
	kmem_free(remdev->l2ad_buflist, sizeof (list_t));
	kmem_free(remdev, sizeof (l2arc_dev_t));
}

void
l2arc_init(void)
{
	l2arc_thread_exit = 0;
	l2arc_ndev = 0;
	l2arc_writes_sent = 0;
	l2arc_writes_done = 0;

	mutex_init(&l2arc_feed_thr_lock, NULL, MUTEX_DEFAULT, NULL);
	cv_init(&l2arc_feed_thr_cv, NULL, CV_DEFAULT, NULL);
	mutex_init(&l2arc_dev_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&l2arc_buflist_mtx, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&l2arc_free_on_write_mtx, NULL, MUTEX_DEFAULT, NULL);

	l2arc_dev_list = &L2ARC_dev_list;
	l2arc_free_on_write = &L2ARC_free_on_write;
	list_create(l2arc_dev_list, sizeof (l2arc_dev_t),
	    offsetof(l2arc_dev_t, l2ad_node));
	list_create(l2arc_free_on_write, sizeof (l2arc_data_free_t),
	    offsetof(l2arc_data_free_t, l2df_list_node));
}

void
l2arc_fini(void)
{
	/*
	 * This is called from dmu_fini(), which is called from spa_fini();
	 * Because of this, we can assume that all l2arc devices have
	 * already been removed when the pools themselves were removed.
	 */

	l2arc_do_free_on_write();

	mutex_destroy(&l2arc_feed_thr_lock);
	cv_destroy(&l2arc_feed_thr_cv);
	mutex_destroy(&l2arc_dev_mtx);
	mutex_destroy(&l2arc_buflist_mtx);
	mutex_destroy(&l2arc_free_on_write_mtx);

	list_destroy(l2arc_dev_list);
	list_destroy(l2arc_free_on_write);
}

void
l2arc_start(void)
{
	if (!(spa_mode_global & FWRITE))
		return;

	(void) thread_create(NULL, 0, l2arc_feed_thread, NULL, 0, &p0,
	    TS_RUN, minclsyspri);
}

void
l2arc_stop(void)
{
	if (!(spa_mode_global & FWRITE))
		return;

	mutex_enter(&l2arc_feed_thr_lock);
	cv_signal(&l2arc_feed_thr_cv);	/* kick thread out of startup */
	l2arc_thread_exit = 1;
	while (l2arc_thread_exit != 0)
		cv_wait(&l2arc_feed_thr_cv, &l2arc_feed_thr_lock);
	mutex_exit(&l2arc_feed_thr_lock);
}