summaryrefslogtreecommitdiff
path: root/kvm.c
blob: 571567c0fbeae281ca2835d141d9e6dcc79e95f1 (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
/*
 * GPL HEADER START
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 *
 * GPL HEADER END
 *
 * Originally implemented on Linux:
 * Copyright (C) 2006 Qumranet, Inc.
 *
 * Authors:
 *   Avi Kivity   <avi@qumranet.com>
 *   Yaniv Kamay  <yaniv@qumranet.com>
 *
 * Ported to illumos by Joyent
 * Copyright 2011 Joyent, Inc. All rights reserved.
 *
 * Authors:
 *   Max Bruning	<max@joyent.com>
 *   Bryan Cantrill	<bryan@joyent.com>
 *   Robert Mustacchi	<rm@joyent.com>
 */

/*
 * KVM -- Kernel Virtual Machine Driver
 * ------------------------------------
 *
 * The kvm driver's purpose it to provide an interface for accelerating virtual
 * machines. To that end the kernel implements and provides emulation for
 * various pieces of hardware. The kernel also interacts directly with
 * extensions to the x86 instruction set via VT-x and related technologies on
 * Intel processors. The system is designed to support SVM (now marketed as
 * AMD-V); however, it is not currently implemented in the illumos version. KVM
 * does not provide all the pieces necessary for vitalization, nor is that a
 * part of its design.
 *
 * KVM is a psuedo-device presented to userland as a character device. Consumers
 * open the device and interact primarily through ioctl(2) and mmap(2).
 *
 * General Theory
 * --------------
 *
 * A consumer will open up the KVM driver and perform ioctls to set up initial
 * state and create virtual CPUs (VCPU). To run a specific VCPU an ioctl is
 * performed. When the ioctl occurs we use the instruction set extensions to try
 * and run that CPU in the current thread. This is run for as long as possible
 * until an instruction that needs to be emulated by the host, e.g. a write to
 * emulated hardware, or some external event brings us out e.g. an interrupt,
 * the schedular descheduling the thread, etc.. Each VCPU is modeled as a
 * thread.  The KVM driver notes the exit reason and either handles it and
 * emulates it or returns to the guest to handle it. This loop generally follows
 * this flowchart:
 *
 *
 *       Userland                        Kernel
 *                         |
 *    |-----------|        |
 *    | VCPU_RUN  |--------|-----------------|
 *    | ioctl(2)  |        |                 |
 *    |-----------|        |                \|/
 *          ^              |            |---------|
 *          |              |            | Run CPU |
 *          |              |       |--->| for the |
 *          |              |       |    |  guest  |
 *          |              |       |    |---------|
 *          |              |       |         |
 *          |              |       |         |
 *          |              |       |         |
 *          |              |       |         | Stop execution of
 *          |              |       |         | guest
 *          |              |       |         |------------|
 *          |              |  |---------|                 |
 *          |              |  |  Handle |                 |
 *          |              |  |  guest  |                \|/
 *          |              |  |  exit   |               /   \
 *     |---------|         |  |---------|             /       \
 *     |  Handle |         |       ^                /  Can the  \
 *     |  guest  |         |       |--------------/ Kernel handle \
 *     |  exit   |         |           Yes        \   the exit    /
 *     |---------|         |                        \  reason?  /
 *          ^              |                          \       /
 *          |              |                            \   /
 *          |              |                              |
 *          |              |                              | No
 *          |--------------|------------------------------|
 *                         |
 *
 * The data regarding the state of the VCPU and of the overall virtual machine
 * is available via mmap(2) of the file descriptor corresponding to the VCPU of
 * interest.
 *
 * All the memory for the guest is handled in the userspace of the guest. This
 * includes mapping in the BIOS, the program text for the guest, and providing
 * devices. To communicate about this information, get and set kernel device
 * state, and interact in various ways,
 *
 * Kernel Emulated and Assisted Hardware
 * -------------------------------------
 *
 * CPUs
 *
 * Intel and AMD provide hardware acceleration that allows for a CPU to run in
 * various execution and addressing modes:
 *   + Real Mode - 8086 style 16-bit operands and 20-bit addressing
 *   + Protected Mode - 80286 style 32-bit operands and addressing and Virtual
 *   			Memory
 *   + Protected Mode with PAE - Physical Address Extensions to allow 36-bits of
 *				 addressing for physical memory. Only 32-bits of
 *				 addressing for virtual memory are available.
 *
 *   + Long Mode - amd64 style 64-bit operands and 64-bit virtual addressing.
 *		   Currently only 48 bits of physical memory can be addressed.
 *
 *   + System Management mode is unsupported and untested. It may work. It may
 *     cause a panic.
 *
 * Other Hardware
 *
 * The kernel emulates various pieces of additional hardware that are necessary
 * for an x86 system to function. These include:
 *
 *   + i8254 PIT - Intel Programmable Interval Timer
 *   + i8259 PIC - Intel Programmable Interrupt Controller
 *   + Modern APIC architecture consisting of:
 *      - Local APIC
 *      - I/O APIC
 *   + IRQ routing table
 *   + MMU - Memory Management Unit
 *
 * The following diagram shows how the different pieces of emulated hardware fit
 * together. An arrow pointing to something denotes that the pointed to item is
 * contained within the object.
 *
 *                                 Up to KVM_MAX_VCPUS (64) cpus
 *
 *                                         |---------|     |-------|
 *           |-------------|               | Virtual |     | Local |    Per
 *           |             |-------------->| CPU #n  |     | APIC  |<-- VCPU
 *           |   Virtual   |               |---------|     |-------|     |
 *           |   Machine   |                               ^            \|/
 *           |             |-------------->|---------|-----|     |-------------|
 *           |-------------|               | Virtual |           | Registers   |
 *              | | | | |                  | CPU #0  |---------->|             |
 *              | | | | |                  |---------|           | RAX,RIP,ETC |
 *              | | | | |                                        | CR0,CR4,ETC |
 *              | | | | |                                        | CPUID,ETC   |
 *              | | | | |                                        |-------------|
 *              | | | | |
 *              | | | | |
 *              | | | | |
 *              | | | | |
 * |-------|    | | | | |                           |-------------------------|
 * | i8254 |<---| | | | |                           |                         |
 * |  PIT  |      | | | |                           |    Memory Management    |
 * |-------|      | | | |-------------------------->|          Unit           |
 *                | | | |                           |           &&            |
 *                | | | |  |--------------|         |    Shadow Page Table    |
 * |-------|      | | | |->| Input/Output |         |                         |
 * | i8259 |<-----| |      |     APIC     |         |-------------------------|
 * |  PIC  |       \|/     |--------------|
 * |-------|   |---------|
 *             |   IRQ   |
 *             | Routing |
 *             |  Table  |
 *             |---------|
 *
 *
 * Internal Code Layout and Design
 * -------------------------------
 *
 * The KVM code can be broken down into the following broad sections:
 *
 *    + Device driver entry points
 *    + Generic code and driver entry points
 *    + x86 and architecture specific code
 *    + Hardware emulation specific code
 *    + Host CPU specific code
 *
 * Host CPU Specific Code
 *
 * Both Intel and AMD provide a means for accelerating guest operation, VT-X
 * (VMX) and SVM (AMD-V) respectively. However, the instructions, design, and
 * means of interacting with each are different. To get around this there is a
 * generic vector of operations which are implemented by both subsystems. The
 * rest of the code base references these operations via the vector. As a part
 * of attach(9E), the system dynamically determines whether the system
 * should use the VMX or SVM operations.
 *
 * The operations vector is entitled kvm_x86_ops. It's functions are:
 * TODO Functions and descriptions, though there may be too many
 *
 *
 * Hardware Emulation Specific Code
 *
 * Various pieces of hardware are emulated by the kernel in the KVM module as
 * described previously. These are accessed in several ways:
 *
 *    + Userland performs ioctl(2)s to get and set state
 *    + Guests perform PIO to devices
 *    + Guests write to memory locations that correspond to devices
 *
 * To handle memory mapped devices in the guest there is an internal notion of
 * an I/O device. There is an internal notion of an I/O bus. Devices can be
 * registered onto the bus. Currently two buses exist. One for programmed I/O
 * devices and another for memory mapped devices.
 *
 * Code related to IRQs is primairly contained within kvm_irq.c and
 * kvm_irq_conn.c. To facilitate and provide a more generic IRQ system there are
 * two useful sets of notifiers. The notifiers fire a callback when the
 * specified event occurs.  Currently there are two notifiers:
 *
 *
 *    + IRQ Mask Notifier: This fires its callback when an IRQ has been masked
 *			   by an operation.
 *    + IRQ Ack Notifier: This fires its callback when an IRQ has been
 *			  acknowledged.
 *
 * The hardware emulation code is broken down across the following files:
 *
 *    + i8254 PIT implementation: kvm_i8254.c and kvm_i8254.h
 *    + i8259 PIC implementation: kvm_i8259.c
 *    + I/O APIC Implementation: kvm_ioapic.c and kvm_ioapic.h
 *    + Local APIC Implementation: kvm_lapic.c and kvm_lapic.h
 *    + Memory Management Unit: kvm_mmu.c, kvm_mmu.h, and kvm_paging_tmpl.h
 *
 * x86 and Architecture Specific Code
 *
 * The code specific to x86 that is not device specific is broken across two
 * files. The first is kvm_x86.c. This contains most of the x86 specific
 * logic, calls into the CPU specific vector of operations, and serves as a
 * gateway to some device specific portions and memory management code.
 *
 * The other main piece of this is kvm_emulate.c. This file contains code
 * that cannot be handled by the CPU specific instructions and instead need to
 * be handled by kvm, for example an inb or outb instruction.
 *
 * Generic Code
 *
 * The code that is not specific to devices or to x86 specifically can be found
 * in kvm.c. This includes code that interacts directly with different parts of
 * the rest of the kernel; the scheduler, cross calls, etc.
 *
 * Device Driver Entry Points
 *
 * The KVM driver is a psuedo-device that presents as a character device. All of
 * the necessary entry points and related pieces of infrastructure are all
 * located in kvm.c. This includes all of the logic related to open(2),
 * close(2), mmap(2), ioctl(2), and the other necessary driver entry points.
 *
 * Interactions between Userland and the Kernel
 * --------------------------------------------
 *
 * -Opening and cloning / VCPUs
 * -The mmap(2) related pieces.
 * -The general ioctl->arch->x86_ops->vmx
 *
 * Timers and Cyclics
 * ------------------
 *
 * -Timers mapping to cyclics
 *
 * Memory Management
 * -----------------
 *
 * -Current memory model / assumptions (i.e. can't be paged)
 * -Use of kpm
 */

#include <sys/types.h>
#include <sys/param.h>
#include <sys/errno.h>
#include <sys/uio.h>
#include <sys/buf.h>
#include <sys/modctl.h>
#include <sys/open.h>
#include <sys/kmem.h>
#include <sys/poll.h>
#include <sys/conf.h>
#include <sys/cmn_err.h>
#include <sys/stat.h>
#include <sys/ddi.h>
#include <sys/sunddi.h>
#include <sys/atomic.h>
#include <sys/spl.h>
#include <sys/cpuvar.h>
#include <sys/segments.h>
#include <sys/cred.h>
#include <sys/devops.h>
#include <sys/file.h>
#include <sys/mman.h>
#include <sys/stat.h>
#include <sys/vm.h>
#include <sys/proc.h>
#include <vm/seg_kpm.h>
#include <sys/avl.h>
#include <sys/condvar_impl.h>
#include <sys/file.h>
#include <sys/vnode.h>
#include <sys/strsubr.h>
#include <sys/stream.h>
#include <sys/machparam.h>
#include <sys/xc_levels.h>
#include <asm/cpu.h>

#include "kvm_bitops.h"
#include "kvm_vmx.h"
#include "msr-index.h"
#include "kvm_msr.h"
#include "kvm_host.h"
#include "kvm_lapic.h"
#include "processor-flags.h"
#include "hyperv.h"
#include "kvm_apicdef.h"
#include "kvm_iodev.h"
#include "kvm.h"
#include "kvm_x86impl.h"
#include "kvm_irq.h"
#include "kvm_tss.h"
#include "kvm_ioapic.h"
#include "kvm_coalesced_mmio.h"
#include "kvm_i8254.h"
#include "kvm_mmu.h"
#include "kvm_cache_regs.h"

#undef DEBUG

/*
 * The entire state of the kvm device.
 */
typedef struct {
	struct kvm *kds_kvmp;		/* pointer to underlying VM */
	struct kvm_vcpu *kds_vcpu;	/* pointer to VCPU */
} kvm_devstate_t;

/*
 * Globals
 */
page_t *bad_page;
void *bad_page_kma;
pfn_t bad_pfn;

/*
 * Tunables
 */
static int kvm_hiwat = 0x1000000;

/*
 * Internal driver-wide values
 */
static void *kvm_state;		/* DDI state */
static vmem_t *kvm_minor;	/* minor number arena */
static dev_info_t *kvm_dip;	/* global devinfo hanlde */
static minor_t kvm_base_minor;	/* The only minor device that can be opened */
static int kvmid;		/* monotonically increasing, unique per vm */
static int largepages_enabled = 1;
static cpuset_t cpus_hardware_enabled;
static kmutex_t cpus_hardware_enabled_mp;
static volatile uint32_t hardware_enable_failed;
static int kvm_usage_count;
static list_t vm_list;
static kmutex_t kvm_lock;
static int ignore_msrs = 0;
static unsigned long empty_zero_page[PAGESIZE / sizeof (unsigned long)];

int
kvm_xcall_func(kvm_xcall_t func, void *arg)
{
	if (func != NULL)
		(*func)(arg);

	return (0);
}

void
kvm_xcall(processorid_t cpu, kvm_xcall_t func, void *arg)
{
	cpuset_t set;

	CPUSET_ZERO(set);

	if (cpu == KVM_CPUALL) {
		CPUSET_ALL(set);
	} else {
		CPUSET_ADD(set, cpu);
	}

	kpreempt_disable();
	xc_sync((xc_arg_t)func, (xc_arg_t)arg, 0, CPUSET2BV(set),
		(xc_func_t) kvm_xcall_func);
	kpreempt_enable();
}

void
kvm_user_return_notifier_register(struct kvm_vcpu *vcpu,
    struct kvm_user_return_notifier *urn)
{
	vcpu->urn = urn;
}

void
kvm_user_return_notifier_unregister(struct kvm_vcpu *vcpu,
    struct kvm_user_return_notifier *urn)
{
	vcpu->urn = NULL;
}

void
kvm_fire_urn(struct kvm_vcpu *vcpu)
{
	if (vcpu->urn)
		vcpu->urn->on_user_return(vcpu, vcpu->urn);
}

/*
 * Called when we've been asked to save our context. i.e. we're being swapped
 * out.
 */
void
kvm_ctx_save(void *arg)
{
	struct kvm_vcpu *vcpu = arg;
	kvm_arch_vcpu_put(vcpu);
	kvm_fire_urn(vcpu);
}

/*
 * Called when we're being asked to restore our context. i.e. we're returning
 * from being swapped out.
 */
void
kvm_ctx_restore(void *arg)
{
	int cpu;

	cpu = CPU->cpu_seqid;
	struct kvm_vcpu *vcpu = arg;
	kvm_arch_vcpu_load(vcpu, cpu);
}

#ifdef XXX_KVM_DECLARATION
#define	pfn_valid(pfn) ((pfn < physmax) && (pfn != PFN_INVALID))
#else
#define	pfn_valid(pfn) (pfn != PFN_INVALID)
#endif

inline int
kvm_is_mmio_pfn(pfn_t pfn)
{
	if (pfn_valid(pfn)) {
#ifdef XXX
		struct page *page = compound_head(pfn_to_page(pfn));
		return (PageReserved(page));
#else
		XXX_KVM_PROBE;
#endif
		return (0);
	} else
		return (1);
}

/*
 * Switches to specified vcpu, until a matching vcpu_put()
 */
void
vcpu_load(struct kvm_vcpu *vcpu)
{
	int cpu;

	mutex_enter(&vcpu->mutex);
	kpreempt_disable();
	cpu = CPU->cpu_seqid;
	installctx(curthread, vcpu, kvm_ctx_save, kvm_ctx_restore, NULL,
	    NULL, NULL, NULL);
	kvm_arch_vcpu_load(vcpu, cpu);
	kpreempt_enable();
}

struct kvm_vcpu *
kvm_get_vcpu(struct kvm *kvm, int i)
{
	smp_rmb();
	return (kvm->vcpus[i]);
}

void
vcpu_put(struct kvm_vcpu *vcpu)
{
	kpreempt_disable();
	kvm_arch_vcpu_put(vcpu);
	kvm_fire_urn(vcpu);
	removectx(curthread, vcpu, kvm_ctx_save, kvm_ctx_restore, NULL,
	    NULL, NULL, NULL);
	kpreempt_enable();
	mutex_exit(&vcpu->mutex);
}

int
make_all_cpus_request(struct kvm *kvm, unsigned int req)
{
	int i;
	processorid_t me, cpu;
	struct kvm_vcpu *vcpu;

	mutex_enter(&kvm->requests_lock);

	kpreempt_disable();
	me = curthread->t_cpu->cpu_id;
	for (i = 0; i < kvm->online_vcpus; i++) {
		vcpu = kvm->vcpus[i];
		if (!vcpu)
			break;
		if (test_and_set_bit(req, &vcpu->requests))
			continue;
		cpu = vcpu->cpu;
		if (cpu != -1 && cpu != me)
			poke_cpu(cpu);
	}

	kpreempt_enable();

	mutex_exit(&kvm->requests_lock);

	return (1);
}

void
kvm_flush_remote_tlbs(struct kvm *kvm)
{
	if (make_all_cpus_request(kvm, KVM_REQ_TLB_FLUSH))
		KVM_KSTAT_INC(kvm, kvmks_remote_tlb_flush);
}

void
kvm_reload_remote_mmus(struct kvm *kvm)
{
	make_all_cpus_request(kvm, KVM_REQ_MMU_RELOAD);
}

int
kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
{
	int r;

	mutex_init(&vcpu->mutex, NULL, MUTEX_DRIVER, 0);
	vcpu->cpu = -1;
	vcpu->kvm = kvm;
	vcpu->vcpu_id = id;
	vcpu->run = ddi_umem_alloc(PAGESIZE * 2, DDI_UMEM_SLEEP, &vcpu->cookie);

	r = kvm_arch_vcpu_init(vcpu);

	if (r != 0) {
		vcpu->run = NULL;
		ddi_umem_free(vcpu->cookie);
		return (r);
	}

	return (0);
}

void
kvm_vcpu_uninit(struct kvm_vcpu *vcpu)
{
	kvm_arch_vcpu_uninit(vcpu);
	ddi_umem_free(vcpu->cookie);
}

/*
 * Note if we want to implement the kvm mmu notifier components than the
 * following two functions will need to be readdressed.
 */
static int kvm_init_mmu_notifier(struct kvm *kvm)
{
	return (0);
}

static void
kvm_fini_mmu_notifier(struct kvm *kvm)
{
}

static void
kvm_destroy_vm(struct kvm *kvmp)
{
	int ii;
	void *cookie;

	if (kvmp == NULL)
		return;

	if (kvmp->kvm_kstat != NULL)
		kstat_delete(kvmp->kvm_kstat);

	kvm_arch_flush_shadow(kvmp);  /* clean up shadow page tables */

	kvm_arch_destroy_vm_comps(kvmp);
	kvm_free_irq_routing(kvmp);
	kvm_destroy_pic(kvmp);
	kvm_ioapic_destroy(kvmp);
	kvm_coalesced_mmio_free(kvmp);

	list_remove(&vm_list, kvmp);
	/*
	 * XXX: The fact that we're cleaning these up here means that we aren't
	 * properly cleaning them up somewhere else.
	 */
	cookie = NULL;
	while (avl_destroy_nodes(&kvmp->kvm_avlmp, &cookie) != NULL)
		continue;
	avl_destroy(&kvmp->kvm_avlmp);
	mutex_destroy(&kvmp->kvm_avllock);
	mutex_destroy(&kvmp->memslots_lock);
	mutex_destroy(&kvmp->slots_lock);
	mutex_destroy(&kvmp->irq_lock);
	mutex_destroy(&kvmp->lock);
	mutex_destroy(&kvmp->requests_lock);
	mutex_destroy(&kvmp->mmu_lock);
	kvmp->mm = NULL;
	kvm_fini_mmu_notifier(kvmp);

	for (ii = 0; ii < KVM_NR_BUSES; ii++)
		kmem_free(kvmp->buses[ii], sizeof (struct kvm_io_bus));

	rw_destroy(&kvmp->kvm_rwlock);

	/*
	 * These lists are contained by the pic. However, the pic isn't
	 */
	list_destroy(&kvmp->irq_ack_notifier_list);
	list_destroy(&kvmp->mask_notifier_list);

	kvm_arch_destroy_vm(kvmp);
}

static struct kvm *
kvm_create_vm(void)
{
	int rval = 0;
	int i;
	struct kvm *kvmp = kvm_arch_create_vm();

	if (kvmp == NULL)
		return (NULL);

	list_create(&kvmp->mask_notifier_list,
		    sizeof (struct kvm_irq_mask_notifier),
		    offsetof(struct kvm_irq_mask_notifier, link));
	list_create(&kvmp->irq_ack_notifier_list,
		    sizeof (struct kvm_irq_ack_notifier),
		    offsetof(struct kvm_irq_ack_notifier, link));

	kvmp->memslots = kmem_zalloc(sizeof (struct kvm_memslots), KM_SLEEP);

	rw_init(&kvmp->kvm_rwlock, NULL, RW_DRIVER, NULL);

	for (i = 0; i < KVM_NR_BUSES; i++) {
		kvmp->buses[i] =
		    kmem_zalloc(sizeof (struct kvm_io_bus), KM_SLEEP);
	}

	rval = kvm_init_mmu_notifier(kvmp);

	if (rval != DDI_SUCCESS) {
		rw_destroy(&kvmp->kvm_rwlock);
		kvm_arch_destroy_vm(kvmp);
		return (NULL);
	}

	/*
	 * XXX note that the as struct does not contain  a refcnt, may
	 * have to go lower
	 */
	kvmp->mm = curproc->p_as;
	mutex_init(&kvmp->mmu_lock, NULL, MUTEX_DRIVER, NULL);
	mutex_init(&kvmp->requests_lock, NULL, MUTEX_DRIVER, NULL);
	mutex_init(&kvmp->lock, NULL, MUTEX_DRIVER, NULL);
	mutex_init(&kvmp->memslots_lock, NULL, MUTEX_DRIVER, NULL);
	mutex_init(&kvmp->irq_lock, NULL, MUTEX_DRIVER, NULL);
	mutex_init(&kvmp->slots_lock, NULL, MUTEX_DRIVER, NULL);
	mutex_init(&kvmp->kvm_avllock, NULL, MUTEX_DRIVER, NULL);
	avl_create(&kvmp->kvm_avlmp, kvm_avlmmucmp, sizeof (kvm_mmu_page_t),
	    offsetof(kvm_mmu_page_t, kmp_avlnode));

	mutex_enter(&kvm_lock);
	kvmp->kvmid = kvmid++;
	kvmp->users_count = 1;
	list_insert_tail(&vm_list, kvmp);
	mutex_exit(&kvm_lock);

	if ((kvmp->kvm_kstat = kstat_create("kvm", kvmp->kvmid, "vm",
	    "misc", KSTAT_TYPE_NAMED, sizeof (kvm_stats_t) /
	    sizeof (kstat_named_t), KSTAT_FLAG_VIRTUAL)) == NULL) {
		kvm_destroy_vm(kvmp);
		return (NULL);
	}

	kvmp->kvm_kstat->ks_data = &kvmp->kvm_stats;

	KVM_KSTAT_INIT(kvmp, kvmks_pid, "pid");
	kvmp->kvm_stats.kvmks_pid.value.ui64 = kvmp->kvm_pid = curproc->p_pid;

	KVM_KSTAT_INIT(kvmp, kvmks_mmu_pte_write, "mmu-pte-write");
	KVM_KSTAT_INIT(kvmp, kvmks_mmu_pte_updated, "mmu-pte-updated");
	KVM_KSTAT_INIT(kvmp, kvmks_mmu_pte_zapped, "mmu-pte-zapped");
	KVM_KSTAT_INIT(kvmp, kvmks_mmu_flooded, "mmu-flooded");
	KVM_KSTAT_INIT(kvmp, kvmks_mmu_cache_miss, "mmu-cache-miss");
	KVM_KSTAT_INIT(kvmp, kvmks_mmu_recycled, "mmu-recycled");
	KVM_KSTAT_INIT(kvmp, kvmks_remote_tlb_flush, "remote-tlb-flush");
	KVM_KSTAT_INIT(kvmp, kvmks_lpages, "lpages");
	KVM_KSTAT_INIT(kvmp, kvmks_mmu_unsync_page, "mmu-unsync-page");

	kstat_install(kvmp->kvm_kstat);

	kvm_coalesced_mmio_init(kvmp);

	return (kvmp);
}

/*
 * Free any memory in @free but not in @dont.
 */
static void
kvm_free_physmem_slot(struct kvm_memory_slot *free,
    struct kvm_memory_slot *dont)
{
	int i;

	if (!dont || free->rmap != dont->rmap)
		kmem_free(free->rmap, free->npages * sizeof (struct page *));

	if ((!dont || free->dirty_bitmap != dont->dirty_bitmap) &&
	    free->dirty_bitmap)
		kmem_free(free->dirty_bitmap, free->dirty_bitmap_sz);

	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
		if ((!dont || free->lpage_info[i] != dont->lpage_info[i]) &&
		    free->lpage_info[i]) {
			kmem_free(free->lpage_info[i], free->lpage_info_sz[i]);
			free->lpage_info[i] = NULL;
		}
	}

	free->npages = 0;
	free->dirty_bitmap = NULL;
	free->rmap = NULL;
}

void
kvm_free_physmem(struct kvm *kvm)
{
	int ii;
	struct kvm_memslots *slots = kvm->memslots;

	for (ii = 0; ii < slots->nmemslots; ii++)
		kvm_free_physmem_slot(&slots->memslots[ii], NULL);

	kmem_free(kvm->memslots, sizeof (struct kvm_memslots));
}

void
kvm_get_kvm(struct kvm *kvm)
{
	atomic_inc_32(&kvm->users_count);
}

unsigned long
kvm_dirty_bitmap_bytes(struct kvm_memory_slot *memslot)
{
	return (BT_SIZEOFMAP(memslot->npages));
}

/*
 * Allocate some memory and give it an address in the guest physical address
 * space.
 *
 * Discontiguous memory is allowed, mostly for framebuffers.
 *
 * Must be called holding mmap_sem for write.
 */
int
__kvm_set_memory_region(struct kvm *kvmp,
    struct kvm_userspace_memory_region *mem, int user_alloc)
{
	int r, flush_shadow = 0;
	gfn_t base_gfn;
	unsigned long npages;
	unsigned long i;
	struct kvm_memory_slot *memslot;
	struct kvm_memory_slot old, new;
	struct kvm_memslots *slots, *old_memslots;

	r = EINVAL;
	/* General sanity checks */
	if (mem->memory_size & (PAGESIZE - 1))
		goto out;
	if (mem->guest_phys_addr & (PAGESIZE - 1))
		goto out;
	if (user_alloc && (mem->userspace_addr & (PAGESIZE - 1)))
		goto out;
	if (mem->slot >= KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS)
		goto out;
	if (mem->guest_phys_addr + mem->memory_size < mem->guest_phys_addr)
		goto out;

	memslot = &kvmp->memslots->memslots[mem->slot];
	base_gfn = mem->guest_phys_addr >> PAGESHIFT;
	npages = mem->memory_size >> PAGESHIFT;

	if (!npages)
		mem->flags &= ~KVM_MEM_LOG_DIRTY_PAGES;

	new = old = *memslot;

	new.base_gfn = base_gfn;
	new.npages = npages;
	new.flags = mem->flags;

	/* Disallow changing a memory slot's size. */
	r = EINVAL;
	if (npages && old.npages && npages != old.npages)
		goto out_free;

	/* Check for overlaps */
	r = EEXIST;
	for (i = 0; i < KVM_MEMORY_SLOTS; ++i) {
		struct kvm_memory_slot *s = &kvmp->memslots->memslots[i];

		if (s == memslot || !s->npages)
			continue;
		if (!((base_gfn + npages <= s->base_gfn) ||
		    (base_gfn >= s->base_gfn + s->npages)))
			goto out_free;
	}

	/* Free page dirty bitmap if unneeded */
	if (!(new.flags & KVM_MEM_LOG_DIRTY_PAGES))
		new.dirty_bitmap = NULL;

	r = ENOMEM;

	/* Allocate if a slot is being created */
	if (npages && !new.rmap) {
		new.rmap =
		    kmem_zalloc(npages * sizeof (struct page *), KM_SLEEP);

		new.user_alloc = user_alloc;
		new.userspace_addr = mem->userspace_addr;
	}

	if (!npages)
		goto skip_lpage;

	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
		unsigned long ugfn;
		unsigned long j;
		int lpages;
		int level = i + 2;

		/* Avoid unused variable warning if no large pages */
		(void) level;

		if (new.lpage_info[i])
			continue;

		lpages = 1 + (base_gfn + npages - 1) /
		    KVM_PAGES_PER_HPAGE(level);
		lpages -= base_gfn / KVM_PAGES_PER_HPAGE(level);

		new.lpage_info[i] =
		    kmem_zalloc(lpages * sizeof (*new.lpage_info[i]), KM_SLEEP);
		new.lpage_info_sz[i] = lpages * sizeof (*new.lpage_info[i]);

		if (base_gfn % KVM_PAGES_PER_HPAGE(level))
			new.lpage_info[i][0].write_count = 1;
		if ((base_gfn+npages) % KVM_PAGES_PER_HPAGE(level))
			new.lpage_info[i][lpages - 1].write_count = 1;
		ugfn = new.userspace_addr >> PAGESHIFT;
		/*
		 * If the gfn and userspace address are not aligned wrt each
		 * other, or if explicitly asked to, disable large page
		 * support for this slot
		 */
		if ((base_gfn ^ ugfn) & (KVM_PAGES_PER_HPAGE(level) - 1) ||
		    !largepages_enabled)
			for (j = 0; j < lpages; ++j)
				new.lpage_info[i][j].write_count = 1;
	}

skip_lpage:

	/* Allocate page dirty bitmap if needed */
	if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) {
		unsigned long dirty_bytes = kvm_dirty_bitmap_bytes(&new);

		new.dirty_bitmap = kmem_zalloc(dirty_bytes, KM_SLEEP);
		new.dirty_bitmap_sz = dirty_bytes;

		/* destroy any largepage mappings for dirty tracking */
		if (old.npages)
			flush_shadow = 1;
	}

	if (!npages) {
		r = ENOMEM;
		slots = kmem_zalloc(sizeof (kvm_memslots_t), KM_SLEEP);
		memcpy(slots, kvmp->memslots, sizeof (kvm_memslots_t));
		if (mem->slot >= slots->nmemslots)
			slots->nmemslots = mem->slot + 1;
		slots->memslots[mem->slot].flags |= KVM_MEMSLOT_INVALID;

		mutex_enter(&kvmp->memslots_lock);
		old_memslots = kvmp->memslots;
		kvmp->memslots = slots;
		mutex_exit(&kvmp->memslots_lock);

		/*
		 * From this point no new shadow pages pointing to a deleted
		 * memslot will be created.
		 *
		 * validation of sp->gfn happens in:
		 * 	- gfn_to_hva (kvm_read_guest, gfn_to_pfn)
		 * 	- kvm_is_visible_gfn (mmu_check_roots)
		 */
		kvm_arch_flush_shadow(kvmp);

		/* XXX: how many bytes to free??? */
		kmem_free(old_memslots, sizeof (struct kvm_memslots));
	}

	r = kvm_arch_prepare_memory_region(kvmp, &new, old, mem, user_alloc);
	if (r)
		goto out_free;

	r = ENOMEM;
	slots = kmem_zalloc(sizeof (kvm_memslots_t), KM_SLEEP);
	memcpy(slots, kvmp->memslots, sizeof (kvm_memslots_t));

	if (mem->slot >= slots->nmemslots)
		slots->nmemslots = mem->slot + 1;

	/* actual memory is freed via old in kvm_free_physmem_slot below */
	if (!npages) {
		new.rmap = NULL;
		new.dirty_bitmap = NULL;
		for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i)
			new.lpage_info[i] = NULL;
	}

	slots->memslots[mem->slot] = new;
	mutex_enter(&kvmp->memslots_lock);
	old_memslots = kvmp->memslots;
	kvmp->memslots = slots;
	mutex_exit(&kvmp->memslots_lock);

	kvm_arch_commit_memory_region(kvmp, mem, old, user_alloc);

	mutex_enter(&kvmp->memslots_lock);
	kvm_free_physmem_slot(&old, &new);
	mutex_exit(&kvmp->memslots_lock);

	kmem_free(old_memslots, sizeof (struct kvm_memslots));

	if (flush_shadow)
		kvm_arch_flush_shadow(kvmp);

	return (DDI_SUCCESS);

out_free:
	kvm_free_physmem_slot(&new, &old);
out:
	return (r);
}

int
kvm_set_memory_region(kvm_t *kvm,
    kvm_userspace_memory_region_t *mem, int user_alloc)
{
	int r;

	mutex_enter(&kvm->slots_lock);
	r = __kvm_set_memory_region(kvm, mem, user_alloc);
	mutex_exit(&kvm->slots_lock);

	return (r);
}

int
kvm_vm_ioctl_set_memory_region(struct kvm *kvm,
    struct kvm_userspace_memory_region *mem, int user_alloc)
{
	if (mem->slot >= KVM_MEMORY_SLOTS)
		return (EINVAL);

	return (kvm_set_memory_region(kvm, mem, user_alloc));
}

void
kvm_disable_largepages(void)
{
	largepages_enabled = 0;
}

int
is_error_pfn(pfn_t pfn)
{
	return (pfn == bad_pfn);
}

static unsigned long
bad_hva(void)
{
	return (PAGEOFFSET);
}

int
kvm_is_error_hva(unsigned long addr)
{
	return (addr == bad_hva());
}

struct kvm_memory_slot *
gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn)
{
	int i;
	struct kvm_memslots *slots;

	mutex_enter(&kvm->memslots_lock);
	slots = kvm->memslots;

	for (i = 0; i < slots->nmemslots; ++i) {
		struct kvm_memory_slot *memslot = &slots->memslots[i];

		if (gfn >= memslot->base_gfn &&
		    gfn < memslot->base_gfn + memslot->npages) {
			mutex_exit(&kvm->memslots_lock);
			return (memslot);
		}
	}
	mutex_exit(&kvm->memslots_lock);
	return (NULL);
}

struct kvm_memory_slot *
gfn_to_memslot(struct kvm *kvm, gfn_t gfn)
{
	gfn = unalias_gfn(kvm, gfn);
	return (gfn_to_memslot_unaliased(kvm, gfn));
}

int
kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn)
{
	struct kvm_memslots *slots;
	int i;

	gfn = unalias_gfn_instantiation(kvm, gfn);

	mutex_enter(&kvm->memslots_lock);
	slots = kvm->memslots;

	for (i = 0; i < KVM_MEMORY_SLOTS; ++i) {
		struct kvm_memory_slot *memslot = &slots->memslots[i];

		if (memslot->flags & KVM_MEMSLOT_INVALID)
			continue;

		if (gfn >= memslot->base_gfn &&
		    gfn < memslot->base_gfn + memslot->npages) {
			mutex_exit(&kvm->memslots_lock);
			return (1);
		}
	}

	mutex_exit(&kvm->memslots_lock);
	return (0);
}

unsigned long
kvm_host_page_size(struct kvm *kvm, gfn_t gfn)
{
	struct vm_area_struct *vma;
	unsigned long addr, size;

	size = PAGESIZE;

	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr))
		return (PAGESIZE);

#ifdef XXX
	down_read(&current->mm->mmap_sem);
	vma = find_vma(current->mm, addr);
	if (!vma)
		goto out;

	size = vma_kernel_pagesize(vma);

out:
	up_read(&current->mm->mmap_sem);
	return (size);
#else
	XXX_KVM_PROBE;
	return (PAGESIZE);
#endif
}

int
memslot_id(struct kvm *kvm, gfn_t gfn)
{
	int i;
	struct kvm_memslots *slots;
	struct kvm_memory_slot *memslot = NULL;

	gfn = unalias_gfn(kvm, gfn);

	mutex_enter(&kvm->memslots_lock);
	slots = kvm->memslots;
	for (i = 0; i < slots->nmemslots; ++i) {
		memslot = &slots->memslots[i];

		if (gfn >= memslot->base_gfn &&
		    gfn < memslot->base_gfn + memslot->npages)
			break;
	}

	mutex_exit(&kvm->memslots_lock);
	return (memslot - slots->memslots);
}

unsigned long
gfn_to_hva(struct kvm *kvm, gfn_t gfn)
{
	struct kvm_memory_slot *slot;

	gfn = unalias_gfn_instantiation(kvm, gfn);
	slot = gfn_to_memslot_unaliased(kvm, gfn);
	if (!slot || slot->flags & KVM_MEMSLOT_INVALID)
		return (bad_hva());

	return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGESIZE);
}

static pfn_t
hva_to_pfn(struct kvm *kvm, unsigned long addr)
{
	page_t page[1];
	int npages;
	pfn_t pfn;
	proc_t *procp = ttoproc(curthread);
	struct as *as = procp->p_as;

#ifdef XXX

	npages = get_user_pages_fast(addr, 1, 1, page);

	if (unlikely(npages != 1)) {
		struct vm_area_struct *vma;

		down_read(&current->mm->mmap_sem);
		vma = find_vma(current->mm, addr);

		if (vma == NULL || addr < vma->vm_start ||
		    !(vma->vm_flags & VM_PFNMAP)) {
			up_read(&current->mm->mmap_sem);
			get_page(bad_page);
			return (page_to_pfn(bad_page));
		}

		pfn = ((addr - vma->vm_start) >> PAGESHIFT) + vma->vm_pgoff;
		up_read(&current->mm->mmap_sem);
		BUG_ON(!kvm_is_mmio_pfn(pfn));
	} else
		pfn = page_to_pfn(page[0]);
#else
	XXX_KVM_PROBE;
	if (addr < kernelbase)
		pfn = hat_getpfnum(as->a_hat, (caddr_t)addr);
	else
		pfn = hat_getpfnum(kas.a_hat, (caddr_t)addr);
#endif
	return (pfn);
}

pfn_t
gfn_to_pfn(struct kvm *kvm, gfn_t gfn)
{
	unsigned long addr;
	pfn_t pfn;

	addr = gfn_to_hva(kvm, gfn);

	if (kvm_is_error_hva(addr)) {
		get_page(bad_page);
		return (page_to_pfn(bad_page));
	}

	pfn = hva_to_pfn(kvm, addr);

	return (pfn);
}

page_t *
gfn_to_page(struct kvm *kvm, gfn_t gfn)
{
	pfn_t pfn = gfn_to_pfn(kvm, gfn);

	if (!kvm_is_mmio_pfn(pfn))
		return (pfn_to_page(pfn));

	get_page(bad_page);
	return (bad_page);
}

void
kvm_release_pfn_clean(pfn_t pfn)
{
#ifdef XXX
	if (!kvm_is_mmio_pfn(pfn))
		put_page(pfn_to_page(pfn));
#else
	XXX_KVM_PROBE;
#endif
}

void
kvm_release_page_dirty(page_t *page)
{
	kvm_release_pfn_dirty(page_to_pfn(page));
}

void
kvm_release_pfn_dirty(pfn_t pfn)
{
	kvm_set_pfn_dirty(pfn);
	kvm_release_pfn_clean(pfn);
}

void
kvm_set_pfn_dirty(pfn_t pfn)
{
#ifdef XXX
	if (!kvm_is_mmio_pfn(pfn)) {
		struct page *page = pfn_to_page(pfn);
		if (!PageReserved(page))
			SetPageDirty(page); /* XXX - not defined in linux?! */
	}
#else
	XXX_KVM_PROBE;
#endif
}

void
kvm_set_pfn_accessed(struct kvm *kvm, pfn_t pfn)
{
#ifdef XXX
	if (!kvm_is_mmio_pfn(pfn))
		mark_page_accessed(pfn_to_page(pfn));
#else
	XXX_KVM_PROBE;
#endif
}

void
kvm_get_pfn(struct kvm_vcpu *vcpu, pfn_t pfn)
{
	if (!kvm_is_mmio_pfn(pfn))
		get_page(pfn_to_page(pfn));
}

static int
next_segment(unsigned long len, int offset)
{
	if (len > PAGESIZE - offset)
		return (PAGESIZE - offset);
	else
		return (len);
}

int
kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset, int len)
{
	int r = 0;
	unsigned long addr;

	addr = gfn_to_hva(kvm, gfn);

	if (kvm_is_error_hva(addr))
		return (-EFAULT);

	if (addr >= kernelbase) {
		bcopy((caddr_t)(addr + offset), data, len);
	} else {
		r = copyin((caddr_t)(addr + offset), data, len);
	}

	if (r)
		return (-EFAULT);

	return (0);
}

int
kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len)
{
	gfn_t gfn = gpa >> PAGESHIFT;
	int seg;
	int offset = offset_in_page(gpa);
	int ret;
	uintptr_t dp = (uintptr_t)data;

	while ((seg = next_segment(len, offset)) != 0) {
		ret = kvm_read_guest_page(kvm, gfn, (void *)dp, offset, seg);
		if (ret < 0)
			return (ret);
		offset = 0;
		len -= seg;
		dp += seg;
		++gfn;
	}
	return (0);
}

int
kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len)
{
	int r;
	unsigned long addr;
	gfn_t gfn = gpa >> PAGESHIFT;
	int offset = offset_in_page(gpa);

	addr = gfn_to_hva(kvm, gfn);
	if (kvm_is_error_hva(addr))
		return (-EFAULT);

#ifdef XXX
	pagefault_disable();
#else
	XXX_KVM_PROBE;
#endif

	r = copyin((caddr_t)addr + offset, data, len);
#ifdef XXX
	pagefault_enable();
#else
	XXX_KVM_PROBE;
#endif
	if (r)
		return (-EFAULT);

	return (0);
}

int
kvm_write_guest_page(struct kvm *kvm,
    gfn_t gfn, const void *data, int offset, int len)
{
	int r = 0;
	unsigned long addr;

	addr = gfn_to_hva(kvm, gfn);

	if (kvm_is_error_hva(addr))
		return (-EFAULT);

	/* XXX - addr could be user or kernel */
	if (addr >= kernelbase) {
		bcopy(data, (caddr_t)(addr + offset), len);
	} else {
		r = copyout(data, (caddr_t)(addr + offset), len);
	}

	if (r)
		return (-EFAULT);

	mark_page_dirty(kvm, gfn);
	return (0);
}

int
kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data, unsigned long len)
{
	gfn_t gfn = gpa >> PAGESHIFT;
	int seg;
	int offset = offset_in_page(gpa);
	int ret;
	uintptr_t dp = (uintptr_t)data;

	while ((seg = next_segment(len, offset)) != 0) {
		ret = kvm_write_guest_page(kvm, gfn, (void *)dp, offset, seg);
		if (ret < 0)
			return (ret);
		offset = 0;
		len -= seg;
		dp += seg;
		++gfn;
	}

	return (0);
}

int
kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len)
{
	return (kvm_write_guest_page(kvm, gfn, empty_zero_page, offset, len));
}

void
mark_page_dirty(struct kvm *kvm, gfn_t gfn)
{
	struct kvm_memory_slot *memslot;

	gfn = unalias_gfn(kvm, gfn);
	memslot = gfn_to_memslot_unaliased(kvm, gfn);

	if (memslot && memslot->dirty_bitmap) {
		unsigned long rel_gfn = gfn - memslot->base_gfn;
		unsigned long *p = memslot->dirty_bitmap + rel_gfn / 64;
		int offset = rel_gfn % 64;

		/* avoid RMW */
		if (!test_bit(offset, p))
			__set_bit(offset, p);
	}
}

int
kvm_vcpu_is_bsp(struct kvm_vcpu *vcpu)
{
	return (vcpu->kvm->bsp_vcpu_id == vcpu->vcpu_id);
}

/*
 * The vCPU has executed a HLT instruction with in-kernel mode enabled.
 */
void
kvm_vcpu_block(struct kvm_vcpu *vcpu)
{
	for (;;) {
		if (kvm_arch_vcpu_runnable(vcpu)) {
			set_bit(KVM_REQ_UNHALT, &vcpu->requests);
			break;
		}

		if (issig(JUSTLOOKING))
			break;

		mutex_enter(&vcpu->kvcpu_kick_lock);

		if (kvm_cpu_has_pending_timer(vcpu)) {
			mutex_exit(&vcpu->kvcpu_kick_lock);
			break;
		}

		(void) cv_wait_sig_swap(&vcpu->kvcpu_kick_cv,
		    &vcpu->kvcpu_kick_lock);

		mutex_exit(&vcpu->kvcpu_kick_lock);
	}
}

/*
 * Creates some virtual cpus.  Good luck creating more than one.
 */
int
kvm_vm_ioctl_create_vcpu(struct kvm *kvm, uint32_t id, int *rval_p)
{
	int r, i;
	struct kvm_vcpu *vcpu, *v;

	vcpu = kvm_arch_vcpu_create(kvm, id);
	if (vcpu == NULL)
		return (EINVAL);

	r = kvm_arch_vcpu_setup(vcpu);
	if (r)
		return (r);

	mutex_enter(&kvm->lock);

	if (kvm->online_vcpus == KVM_MAX_VCPUS) {
		r = EINVAL;
		goto vcpu_destroy;
	}

	/* kvm_for_each_vcpu(r, v, kvm) */
	for (i = 0; i < kvm->online_vcpus; i++) {
		v = kvm->vcpus[i];
		if (v->vcpu_id == id) {
			r = -EEXIST;
			goto vcpu_destroy;
		}
	}

	/* BUG_ON(kvm->vcpus[atomic_read(&kvm->online_vcpus)]); */

	/* Now it's all set up, let userspace reach it */
	kvm_get_kvm(kvm);

	*rval_p = kvm->online_vcpus;  /* guarantee unique id */
	vcpu->vcpu_id = *rval_p;

	kvm->vcpus[kvm->online_vcpus] = vcpu;

	smp_wmb();

	atomic_inc_32(&kvm->online_vcpus);

	if (kvm->bsp_vcpu_id == id)
		kvm->bsp_vcpu = vcpu;

	mutex_exit(&kvm->lock);
	return (r);

vcpu_destroy:
#ifdef XXX
	mutex_exit(&kvm->lock);
	kvm_arch_vcpu_destroy(vcpu);
#else
	XXX_KVM_PROBE;
#endif
	return (r);
}

static int
kvm_vcpu_ioctl_set_sigmask(struct kvm_vcpu *vcpu, sigset_t *sigset)
{
	if (sigset) {
		vcpu->sigset_active = 1;
		vcpu->sigset = *sigset;
	} else
		vcpu->sigset_active = 0;

	return (0);
}

static int
kvm_dev_ioctl_create_vm(kvm_devstate_t *ksp, intptr_t arg, int *rv)
{
	if (ksp->kds_kvmp != NULL)
		return (EINVAL);

	ksp->kds_kvmp = kvm_create_vm();

	if (ksp->kds_kvmp == NULL) {
		cmn_err(CE_WARN, "Could not create new vm\n");
		return (EIO);
	}
	*rv = ksp->kds_kvmp->kvmid;
	return (DDI_SUCCESS);
}

int
kvm_dev_ioctl_check_extension_generic(long arg, int *rv)
{
	switch (arg) {
	case KVM_CAP_USER_MEMORY:
	case KVM_CAP_DESTROY_MEMORY_REGION_WORKS:
	case KVM_CAP_JOIN_MEMORY_REGIONS_WORKS:
	case KVM_CAP_SET_BOOT_CPU_ID:
	case KVM_CAP_INTERNAL_ERROR_DATA:
		*rv = 1;
		return (DDI_SUCCESS);
	case KVM_CAP_IRQ_ROUTING:
		*rv = KVM_MAX_IRQ_ROUTES;
		return (DDI_SUCCESS);
	default:
		break;
	}
	return (kvm_dev_ioctl_check_extension(arg, rv));
}

static void
hardware_enable(void *junk)
{
	int cpu;
	int r;

	cpu = curthread->t_cpu->cpu_id;

	mutex_enter(&cpus_hardware_enabled_mp);
	if (CPU_IN_SET(cpus_hardware_enabled, cpu)) {
		mutex_exit(&cpus_hardware_enabled_mp);
		return;
	}

	CPUSET_ADD(cpus_hardware_enabled, cpu);
	mutex_exit(&cpus_hardware_enabled_mp);

	r = kvm_arch_hardware_enable(NULL);

	if (r) {
		mutex_enter(&cpus_hardware_enabled_mp);
		CPUSET_DEL(cpus_hardware_enabled, cpu);
		mutex_exit(&cpus_hardware_enabled_mp);
		atomic_inc_32(&hardware_enable_failed);
		cmn_err(CE_WARN, "kvm: enabling virtualization CPU%d failed\n",
			cpu);
	}
}

void
hardware_disable(void *junk)
{
	int cpu = curthread->t_cpu->cpu_id;

	mutex_enter(&cpus_hardware_enabled_mp);
	if (!CPU_IN_SET(cpus_hardware_enabled, cpu)) {
		mutex_exit(&cpus_hardware_enabled_mp);
		return;
	}

	CPUSET_DEL(cpus_hardware_enabled, cpu);
	mutex_exit(&cpus_hardware_enabled_mp);
	kvm_arch_hardware_disable(NULL);
}

static void
hardware_disable_all_nolock(void)
{
	kvm_usage_count--;
	if (!kvm_usage_count)
		on_each_cpu(hardware_disable, NULL, 1);
}

static void
hardware_disable_all(void)
{
	mutex_enter(&kvm_lock);
	hardware_disable_all_nolock();
	mutex_exit(&kvm_lock);
}

static int
hardware_enable_all(void)
{
	int r = 0;

	mutex_enter(&kvm_lock);

	kvm_usage_count++;
	if (kvm_usage_count == 1) {
		hardware_enable_failed = 0;
		on_each_cpu(hardware_enable, NULL, 1);

		if (hardware_enable_failed) {
			hardware_disable_all_nolock();
			r = EBUSY;
		}
	}

	mutex_exit(&kvm_lock);

	return (r);
}

/* kvm_io_bus_write - called under kvm->slots_lock */
int
kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
    int len, const void *val)
{
	int i;
#ifdef XXX_KVM_DECLARATION
	struct kvm_io_bus *bus = rcu_dereference(kvm->buses[bus_idx]);
#else
	struct kvm_io_bus *bus = kvm->buses[bus_idx];
#endif

	for (i = 0; i < bus->dev_count; i++) {
		if (!kvm_iodevice_write(bus->devs[i], addr, len, val))
			return (0);
	}

	return (-EOPNOTSUPP);
}

/* kvm_io_bus_read - called under kvm->slots_lock */
int
kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr,
    int len, void *val)
{
	int i;
#ifdef XXX_KVM_DECLARATION
	struct kvm_io_bus *bus = rcu_dereference(kvm->buses[bus_idx]);
#else
	struct kvm_io_bus *bus = kvm->buses[bus_idx];
#endif

	for (i = 0; i < bus->dev_count; i++) {
		if (!kvm_iodevice_read(bus->devs[i], addr, len, val))
			return (0);
	}

	return (-EOPNOTSUPP);
}

/* Caller must hold slots_lock. */
int
kvm_io_bus_register_dev(struct kvm *kvm,
    enum kvm_bus bus_idx, struct kvm_io_device *dev)
{
	struct kvm_io_bus *new_bus, *bus;

	bus = kvm->buses[bus_idx];
	if (bus->dev_count > NR_IOBUS_DEVS-1)
		return (-ENOSPC);

	new_bus = kmem_zalloc(sizeof (struct kvm_io_bus), KM_SLEEP);
	if (!new_bus)
		return (-ENOMEM);
	memcpy(new_bus, bus, sizeof (struct kvm_io_bus));
	new_bus->devs[new_bus->dev_count++] = dev;
#ifdef XXX
	rcu_assign_pointer(kvm->buses[bus_idx], new_bus);
	synchronize_srcu_expedited(&kvm->srcu);
#else
	XXX_KVM_PROBE;
	kvm->buses[bus_idx] = new_bus;
#endif
	if (bus)
		kmem_free(bus, sizeof (struct kvm_io_bus));

	return (0);
}

/* Caller must hold slots_lock. */
int
kvm_io_bus_unregister_dev(struct kvm *kvm,
    enum kvm_bus bus_idx, struct kvm_io_device *dev)
{
	int i, r;
	struct kvm_io_bus *new_bus, *bus;

	new_bus = kmem_zalloc(sizeof (struct kvm_io_bus), KM_SLEEP);
	if (!new_bus)
		return (-ENOMEM);

	bus = kvm->buses[bus_idx];
	memcpy(new_bus, bus, sizeof (struct kvm_io_bus));

	r = -ENOENT;
	for (i = 0; i < new_bus->dev_count; i++) {
		if (new_bus->devs[i] == dev) {
			r = 0;
			new_bus->devs[i] = new_bus->devs[--new_bus->dev_count];
			break;
		}
	}

	if (r) {
		kmem_free(new_bus, sizeof (struct kvm_io_bus));
		return (r);
	}

#ifdef XXX
	rcu_assign_pointer(kvm->buses[bus_idx], new_bus);
	synchronize_srcu_expedited(&kvm->srcu);
#else
	XXX_KVM_SYNC_PROBE;
	kvm->buses[bus_idx] = new_bus;
#endif
	kmem_free(bus, sizeof (struct kvm_io_bus));
	return (r);
}

int
kvm_init(void *opaque)
{
	int r;
	int cpu;

	r = kvm_arch_init(opaque);

	if (r != DDI_SUCCESS)
		return (r);

	bad_page = alloc_page(KM_SLEEP, &bad_page_kma);
	bad_pfn = bad_page->p_pagenum;

	r = kvm_arch_hardware_setup();

	if (r != DDI_SUCCESS)
		goto out_free;

	r = 0;
	kvm_xcall(KVM_CPUALL, kvm_arch_check_processor_compat, &r);
	if (r < 0)
		goto out_free_1;

	return (0);

out_free_1:
#ifdef XXX
	kvm_arch_hardware_unsetup();
#else
	XXX_KVM_PROBE;
#endif
out_free:
	kmem_free(bad_page_kma, PAGESIZE);
out:
#ifdef XXX
	kvm_arch_exit();
#else
	XXX_KVM_PROBE;
#endif
out_fail:
	return (r);
}


void
kvm_guest_exit(void)
{
#ifdef XXX
	account_system_vtime(current);
	current->flags &= ~PF_VCPU;
#else
	XXX_KVM_PROBE;
#endif
}

void
kvm_guest_enter(void)
{
#ifdef XXX
	account_system_vtime(current);
	current->flags |= PF_VCPU;
#else
	XXX_KVM_PROBE;
#endif
}

/*
 * Find the first cleared bit in a memory region.
 */
unsigned long
find_first_zero_bit(const unsigned long *addr, unsigned long size)
{
	const unsigned long *p = addr;
	unsigned long result = 0;
	unsigned long tmp;

	while (size & ~(64-1)) {
		if (~(tmp = *(p++)))
			goto found;
		result += 64;
		size -= 64;
	}
	if (!size)
		return (result);

	tmp = (*p) | (~0UL << size);
	if (tmp == ~0UL)	/* Are any bits zero? */
		return (result + size);	/* Nope. */
found:
	return (result + ffz(tmp));
}

int
zero_constructor(void *buf, void *arg, int tags)
{
	bzero(buf, (size_t)arg);
	return (0);
}

static int
kvm_attach(dev_info_t *dip, ddi_attach_cmd_t cmd)
{
	minor_t instance;

	if (kpm_enable == 0) {
		cmn_err(CE_WARN, "kvm: kpm_enable must be true\n");
		return (DDI_FAILURE);
	}


	if (cmd != DDI_ATTACH)
		return (DDI_FAILURE);

	if (kvm_dip != NULL)
		return (DDI_FAILURE);

	if (ddi_soft_state_init(&kvm_state, sizeof (kvm_devstate_t), 1) != 0)
		return (DDI_FAILURE);

	instance = ddi_get_instance(dip);
	if (ddi_create_minor_node(dip, "kvm",
	    S_IFCHR, instance, DDI_PSEUDO, 0) == DDI_FAILURE) {
		ddi_soft_state_fini(&kvm_state);
		return (DDI_FAILURE);
	}

	mutex_init(&kvm_lock, NULL, MUTEX_DRIVER, 0);
	if (vmx_init() != DDI_SUCCESS) {
		ddi_soft_state_fini(&kvm_state);
		ddi_remove_minor_node(dip, NULL);
		mutex_destroy(&kvm_lock);
		return (DDI_FAILURE);
	}

	mutex_init(&cpus_hardware_enabled_mp, NULL, MUTEX_DRIVER,
	    (void *)XC_HI_PIL);
	if (hardware_enable_all() != 0) {
		ddi_soft_state_fini(&kvm_state);
		ddi_remove_minor_node(dip, NULL);
		mutex_destroy(&kvm_lock);
		mutex_destroy(&cpus_hardware_enabled_mp);
		vmx_fini();
		return (DDI_FAILURE);
	}

	kvm_dip = dip;
	kvm_base_minor = instance;

	list_create(&vm_list, sizeof (struct kvm),
	    offsetof(struct kvm, vm_list));
	kvm_minor = vmem_create("kvm_minor", (void *)1, UINT32_MAX - 1, 1,
	    NULL, NULL, NULL, 0, VM_SLEEP | VMC_IDENTIFIER);

	ddi_report_dev(dip);

	return (DDI_SUCCESS);
}

static int
kvm_detach(dev_info_t *dip, ddi_detach_cmd_t cmd)
{
	int instance;

	if (cmd != DDI_DETACH)
		return (DDI_FAILURE);

	VERIFY(kvm_dip != NULL && kvm_dip == dip);
	instance = ddi_get_instance(dip);
	VERIFY(instance == kvm_base_minor);
	ddi_prop_remove_all(dip);
	ddi_remove_minor_node(dip, NULL);
	list_destroy(&vm_list);
	vmem_destroy(kvm_minor);
	kvm_dip = NULL;

	hardware_disable_all();
	kvm_arch_hardware_unsetup();
	kvm_arch_exit();
	kmem_free(bad_page_kma, PAGESIZE);

	vmx_fini();
	mmu_destroy_caches();
	mutex_destroy(&cpus_hardware_enabled_mp);
	mutex_destroy(&kvm_lock);
	ddi_soft_state_fini(&kvm_state);

	return (DDI_SUCCESS);
}

/*ARGSUSED*/
static int
kvm_getinfo(dev_info_t *dip, ddi_info_cmd_t infocmd, void *arg, void **result)
{
	kvm_devstate_t *rsp;
	int error = DDI_FAILURE;

	switch (infocmd) {
	case DDI_INFO_DEVT2DEVINFO:
		*result = kvm_dip;
		break;

	case DDI_INFO_DEVT2INSTANCE:
		*result = (void *)((uint64_t)getminor((dev_t)arg));
		error = DDI_SUCCESS;
		break;

	default:
		break;
	}

	return (error);
}

/*ARGSUSED*/
static int
kvm_open(dev_t *devp, int flag, int otype, cred_t *credp)
{
	minor_t minor;
	kvm_devstate_t *ksp;


	if (flag & FEXCL || flag & FNDELAY)
		return (EINVAL);

	if (otype != OTYP_CHR)
		return (EINVAL);

	if (!(flag & FREAD && flag & FWRITE))
		return (EINVAL);

	if (getminor(*devp) != kvm_base_minor)
		return (ENXIO);

	minor = (minor_t)(uintptr_t)vmem_alloc(kvm_minor,
	    1, VM_BESTFIT | VM_SLEEP);

	if (ddi_soft_state_zalloc(kvm_state, minor) != 0) {
		vmem_free(kvm_minor, (void *)(uintptr_t)minor, 1);
		return (ENXIO);
	}

	*devp = makedevice(getmajor(*devp), minor);
	ksp = ddi_get_soft_state(kvm_state, minor);
	VERIFY(ksp != NULL);

	return (0);
}

/*ARGSUSED*/
static int
kvm_close(dev_t dev, int flag, int otyp, cred_t *cred)
{
	kvm_devstate_t *ksp;
	minor_t minor = getminor(dev);
	kvm_t *kvmp;

	VERIFY(getminor(dev) != kvm_base_minor);
	ksp = ddi_get_soft_state(kvm_state, minor);

	if ((kvmp = ksp->kds_kvmp) != NULL) {
		mutex_enter(&kvm_lock);

		if (kvmp->kvm_clones > 0) {
			kvmp->kvm_clones--;
			mutex_exit(&kvm_lock);
		} else {
			kvm_destroy_vm(kvmp);
			mutex_exit(&kvm_lock);
		}
	}

	ddi_soft_state_free(kvm_state, minor);
	vmem_free(kvm_minor, (void *)(uintptr_t)minor, 1);

	return (0);
}

static int
kvm_ioctl(dev_t dev, int cmd, intptr_t arg, int md, cred_t *cr, int *rv)
{
	int rval = DDI_SUCCESS;
	minor_t minor;
	kvm_devstate_t *ksp;
	void *argp = (void *)arg;

	minor = getminor(dev);
	ksp = ddi_get_soft_state(kvm_state, minor);
	if (ksp == NULL)
		return (ENXIO);

	union {
		struct kvm_pit_state ps;
		struct kvm_pit_state2 ps2;
#ifdef XXX_KVM_DECLARATION
		struct kvm_memory_alias alias;
#endif
		struct kvm_pit_config pit_config;
	} u;

	struct {
		int cmd;		/* command */
		void *func;		/* function to call */
		size_t size;		/* size of user-level structure */
		boolean_t copyout;	/* boolean: copy out after func */
		boolean_t vmwide;	/* boolean: ioctl is not per-VCPU */
	} *ioctl, ioctltab[] = {
		{ KVM_RUN, kvm_arch_vcpu_ioctl_run },
		{ KVM_X86_SETUP_MCE, kvm_vcpu_ioctl_x86_setup_mce,
		    sizeof (uint64_t) },
		{ KVM_GET_MSRS, kvm_vcpu_ioctl_get_msrs,
		    sizeof (struct kvm_msrs), B_TRUE },
		{ KVM_SET_MSRS, kvm_vcpu_ioctl_set_msrs,
		    sizeof (struct kvm_msrs) },
		{ KVM_GET_MP_STATE, kvm_arch_vcpu_ioctl_get_mpstate,
		    sizeof (struct kvm_mp_state), B_TRUE },
		{ KVM_SET_MP_STATE, kvm_arch_vcpu_ioctl_set_mpstate,
		    sizeof (struct kvm_mp_state) },
		{ KVM_GET_REGS, kvm_arch_vcpu_ioctl_get_regs,
		    sizeof (struct kvm_regs), B_TRUE },
		{ KVM_SET_REGS, kvm_arch_vcpu_ioctl_set_regs,
		    sizeof (struct kvm_regs) },
		{ KVM_GET_SREGS, kvm_arch_vcpu_ioctl_get_sregs,
		    sizeof (struct kvm_sregs), B_TRUE },
		{ KVM_SET_SREGS, kvm_arch_vcpu_ioctl_set_sregs,
		    sizeof (struct kvm_sregs) },
		{ KVM_GET_FPU, kvm_arch_vcpu_ioctl_get_fpu,
		    sizeof (struct kvm_fpu), B_TRUE },
		{ KVM_SET_FPU, kvm_arch_vcpu_ioctl_set_fpu,
		    sizeof (struct kvm_fpu) },
		{ KVM_GET_CPUID2, kvm_vcpu_ioctl_get_cpuid2,
		    sizeof (struct kvm_cpuid2), B_TRUE },
		{ KVM_SET_CPUID2, kvm_vcpu_ioctl_set_cpuid2,
		    sizeof (struct kvm_cpuid2) },
		{ KVM_GET_LAPIC, kvm_vcpu_ioctl_get_lapic,
		    sizeof (struct kvm_lapic_state), B_TRUE },
		{ KVM_SET_LAPIC, kvm_vcpu_ioctl_set_lapic,
		    sizeof (struct kvm_lapic_state) },
		{ KVM_GET_VCPU_EVENTS, kvm_vcpu_ioctl_x86_get_vcpu_events,
		    sizeof (struct kvm_vcpu_events), B_TRUE },
		{ KVM_SET_VCPU_EVENTS, kvm_vcpu_ioctl_x86_set_vcpu_events,
		    sizeof (struct kvm_vcpu_events) },
		{ KVM_INTERRUPT, kvm_vcpu_ioctl_interrupt,
		    sizeof (struct kvm_interrupt) },
		{ KVM_SET_VAPIC_ADDR, kvm_lapic_set_vapic_addr,
		    sizeof (struct kvm_vapic_addr) },
		{ KVM_GET_PIT2, kvm_vm_ioctl_get_pit2,
		    sizeof (struct kvm_pit_state2), B_TRUE, B_TRUE },
		{ KVM_SET_PIT2, kvm_vm_ioctl_set_pit2,
		    sizeof (struct kvm_pit_state2), B_FALSE, B_TRUE },
		{ 0, NULL }
	};

	for (ioctl = &ioctltab[0]; ioctl->func != NULL; ioctl++) {
		caddr_t buf = NULL;

		if (ioctl->cmd != cmd)
			continue;

		if (ioctl->size != 0) {
			buf = kmem_alloc(ioctl->size, KM_SLEEP);

			if (copyin(argp, buf, ioctl->size) != 0) {
				kmem_free(buf, ioctl->size);
				return (EFAULT);
			}
		}

		if (ioctl->vmwide) {
			kvm_t *kvmp;
			int (*func)(kvm_t *, void *, int *, intptr_t);

			if ((kvmp = ksp->kds_kvmp) == NULL) {
				kmem_free(buf, ioctl->size);
				return (EINVAL);
			}

			func = (int(*)(kvm_t *, void *, int *,
			    intptr_t))ioctl->func;
			rval = func(kvmp, buf, rv, arg);
		} else {
			kvm_vcpu_t *vcpu;
			int (*func)(kvm_vcpu_t *, void *, int *, intptr_t);

			if ((vcpu = ksp->kds_vcpu) == NULL) {
				kmem_free(buf, ioctl->size);
				return (EINVAL);
			}

			func = (int(*)(kvm_vcpu_t *, void *, int *,
			    intptr_t))ioctl->func;
			rval = func(vcpu, buf, rv, arg);
		}

		if (rval == 0 && ioctl->size != 0 && ioctl->copyout) {
			if (copyout(buf, argp, ioctl->size) != 0) {
				kmem_free(buf, ioctl->size);
				return (EFAULT);
			}
		}

		kmem_free(buf, ioctl->size);

		return (rval < 0 ? -rval : rval);
	}

	switch (cmd) {
	case KVM_GET_API_VERSION:
		if (arg != NULL) {
			rval = EINVAL;
			break;
		}
		*rv = KVM_API_VERSION;
		break;

	case KVM_CREATE_VM:
		if (arg != NULL) {
			rval = EINVAL;
			break;
		}

		rval = kvm_dev_ioctl_create_vm(ksp, arg, rv);
		break;

	case KVM_CLONE: {
		dev_t parent = arg;
		kvm_devstate_t *clone;
		struct kvm *kvmp;

		/*
		 * We are not allowed to clone another open if we have created
		 * a virtual machine or virtual CPU with this open.
		 */
		if (ksp->kds_kvmp != NULL || ksp->kds_vcpu != NULL) {
			rval = EBUSY;
			break;
		}

		if (getmajor(parent) != getmajor(dev)) {
			rval = ENODEV;
			break;
		}

		minor = getminor(parent);

		mutex_enter(&kvm_lock);

		if ((clone = ddi_get_soft_state(kvm_state, minor)) == NULL) {
			mutex_exit(&kvm_lock);
			rval = EINVAL;
			break;
		}

		if ((kvmp = clone->kds_kvmp) == NULL) {
			mutex_exit(&kvm_lock);
			rval = ESRCH;
			break;
		}

		kvmp->kvm_clones++;
		ksp->kds_kvmp = kvmp;

		mutex_exit(&kvm_lock);
		break;
	}

	case KVM_CHECK_EXTENSION:
		rval = kvm_dev_ioctl_check_extension_generic(arg, rv);
		break;

	case KVM_GET_VCPU_MMAP_SIZE:
		if (arg != NULL) {
			rval = EINVAL;
			break;
		}
		*rv = ptob(KVM_VCPU_MMAP_LENGTH);
		break;

	case KVM_CREATE_PIT2:
		if (copyin(argp, &u.pit_config,
		    sizeof (struct kvm_pit_config)) != 0) {
			rval = EFAULT;
			break;
		}
		/*FALLTHROUGH*/

	case KVM_CREATE_PIT: {
		struct kvm *kvmp;

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (cmd == KVM_CREATE_PIT) {
			u.pit_config.flags = KVM_PIT_SPEAKER_DUMMY;
		} else {
			ASSERT(cmd == KVM_CREATE_PIT2);
		}

		mutex_enter(&kvmp->slots_lock);

		if (kvmp->arch.vpit != NULL) {
			rval = EEXIST;
		} else if ((kvmp->arch.vpit = kvm_create_pit(kvmp,
		    u.pit_config.flags)) == NULL) {
			rval = ENOMEM;
		}

		mutex_exit(&kvmp->slots_lock);
		break;
	}

	case KVM_CREATE_IRQCHIP: {
		struct kvm_pic *vpic;
		struct kvm *kvmp;

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		mutex_enter(&kvmp->lock);
		rval = EEXIST;
		if (kvmp->arch.vpic)
			goto create_irqchip_unlock;
		rval = ENOMEM;
		vpic = kvm_create_pic(kvmp);
		if (vpic) {
			rval = kvm_ioapic_init(kvmp);
			if (rval) {
				kvm_io_bus_unregister_dev(kvmp,
				    KVM_PIO_BUS, &vpic->dev);
				goto create_irqchip_unlock;
			}
		} else
			goto create_irqchip_unlock;

		smp_wmb();
		kvmp->arch.vpic = vpic;
		smp_wmb();

		rval = kvm_setup_default_irq_routing(kvmp);
		if (rval) {
			mutex_enter(&kvmp->irq_lock);
			kvm_ioapic_destroy(kvmp);
			kvm_destroy_pic(kvmp);
			mutex_exit(&kvmp->irq_lock);
		}
	create_irqchip_unlock:
		mutex_exit(&kvmp->lock);
		break;
	}

	case KVM_X86_GET_MCE_CAP_SUPPORTED: {
		uint64_t mce_cap = KVM_MCE_CAP_SUPPORTED;

		if (copyout(&mce_cap, argp, sizeof (mce_cap)))
			rval = EFAULT;

		break;
	}

	case KVM_SET_IDENTITY_MAP_ADDR: {
		uint64_t addr;

		if (ksp->kds_kvmp == NULL) {
			rval = EINVAL;
			break;
		}

		if (copyin((void *)arg, &addr, sizeof (uint64_t)) != 0) {
			rval = EFAULT;
			break;
		}

		rval = kvm_vm_ioctl_set_identity_map_addr(ksp->kds_kvmp, addr);

		*rv = 0;
		break;
	}

	case KVM_GET_MSR_INDEX_LIST: {
		rval = kvm_vm_ioctl_get_msr_index_list(NULL, arg);
		*rv = 0;
		break;
	}
	case KVM_CREATE_VCPU: {
		uint32_t id = (uintptr_t)arg;
		struct kvm *kvmp;
		struct kvm_vcpu *vcpu;

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (ksp->kds_vcpu != NULL) {
			rval = EEXIST;
			break;
		}

		rval = kvm_vm_ioctl_create_vcpu(ksp->kds_kvmp, id, rv);

		if (rval == 0) {
			ksp->kds_vcpu = kvmp->vcpus[id];
			ASSERT(ksp->kds_vcpu != NULL);
		}

		break;
	}

	case KVM_SET_USER_MEMORY_REGION: {
		struct kvm_userspace_memory_region map;
		struct kvm *kvmp;

		if (copyin(argp, &map, sizeof (map)) != 0) {
			rval = EFAULT;
			break;
		}

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		rval = kvm_vm_ioctl_set_memory_region(kvmp, &map, 1);
		break;
	}

	case KVM_GET_SUPPORTED_CPUID: {
		struct kvm_cpuid2 *cpuid_arg = (struct kvm_cpuid2 *)arg;
		struct kvm_cpuid2 *cpuid;

		cpuid = kmem_zalloc(sizeof (struct kvm_cpuid2), KM_SLEEP);

		if (copyin(argp, cpuid, sizeof (struct kvm_cpuid2)) != 0) {
			kmem_free(cpuid, sizeof (struct kvm_cpuid2));
			rval = EFAULT;
			break;
		}

		if ((rval = kvm_dev_ioctl_get_supported_cpuid(cpuid,
		    cpuid_arg->entries)) != 0) {
			kmem_free(cpuid, sizeof (struct kvm_cpuid2));
			break;
		}

		if (copyout(&cpuid->nent, cpuid_arg, sizeof (int)))
			rval = EFAULT;

		kmem_free(cpuid, sizeof (struct kvm_cpuid2));
		break;
	}
	case KVM_SET_SIGNAL_MASK: {
		struct kvm_signal_mask *sigmask = argp;
		struct kvm_signal_mask kvm_sigmask;
		sigset_t sigset;
		struct kvm_vcpu *vcpu;

		if ((vcpu = ksp->kds_vcpu) == NULL) {
			rval = EINVAL;
			break;
		}

		if (argp == NULL) {
			rval = kvm_vcpu_ioctl_set_sigmask(vcpu, NULL);
			break;
		}

		if (copyin(argp, &kvm_sigmask, sizeof (kvm_sigmask)) != 0) {
			rval = EFAULT;
			break;
		}

		if (kvm_sigmask.len != sizeof (sigset)) {
			rval = EINVAL;
			break;
		}

		if (copyin(sigmask->sigset, &sigset, sizeof (sigset)) != 0) {
			rval = EINVAL;
			break;
		}

		rval = kvm_vcpu_ioctl_set_sigmask(vcpu, &sigset);
		break;
	}

	case KVM_SET_TSS_ADDR: {

		if (ksp->kds_kvmp == NULL) {
			rval = EINVAL;
			break;
		}

		rval = kvm_vm_ioctl_set_tss_addr(ksp->kds_kvmp, arg);
		break;
	}

	case KVM_SET_BOOT_CPU_ID: {
		struct kvm *kvmp;

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (arg >= KVM_MAX_VCPUS) {
			rval = EINVAL;
			break;
		}

		mutex_enter(&kvmp->lock);
		if (kvmp->online_vcpus != 0) {
			rval = EBUSY;
			break;
		} else {
			kvmp->bsp_vcpu_id = arg;
			*rv = kvmp->bsp_vcpu_id;
		}

		mutex_exit(&kvmp->lock);
		break;
	}

	case KVM_REGISTER_COALESCED_MMIO: {
		struct kvm *kvmp;
		struct kvm_coalesced_mmio_zone *zone;
		size_t sz = sizeof (struct kvm_coalesced_mmio_zone);

		zone = kmem_zalloc(sz, KM_SLEEP);

		if (copyin(argp, zone, sz) != 0) {
			kmem_free(zone, sz);
			rval = EFAULT;
			break;
		}

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			kmem_free(zone, sz);
			break;
		}

		rval = kvm_vm_ioctl_register_coalesced_mmio(kvmp, zone);

		kmem_free(zone, sz);
		break;
	}

	case KVM_UNREGISTER_COALESCED_MMIO: {
		struct kvm_coalesced_mmio_zone *zone;
		struct kvm *kvmp;
		size_t sz = sizeof (struct kvm_coalesced_mmio_zone);

		zone = kmem_zalloc(sz, KM_SLEEP);

		if (copyin(argp, zone, sz) != 0) {
			kmem_free(zone, sz);
			break;
		}

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			kmem_free(zone, sz);
			rval = EINVAL;
			break;
		}

		rval = kvm_vm_ioctl_unregister_coalesced_mmio(kvmp, zone);

		kmem_free(zone, sz);
		break;
	}
#ifdef KVM_CAP_IRQ_ROUTING
	case KVM_SET_GSI_ROUTING: {
		struct kvm_irq_routing *route;
		struct kvm *kvmp;
		struct kvm_irq_routing_entry *entries;
		uint32_t nroutes;
		size_t sz = sizeof (kvm_irq_routing_t) + KVM_MAX_IRQ_ROUTES *
		    sizeof (struct kvm_irq_routing_entry);

		/*
		 * Note the route must be allocated on the heap. The sizeof
		 * (kvm_kirq_routing) is approximately 0xc038 currently.
		 */
		route = kmem_zalloc(sz, KM_SLEEP);

		/*
		 * copyin the number of routes, then copyin the routes
		 * themselves.
		 */
		if (copyin(argp, &nroutes, sizeof (nroutes)) != 0) {
			kmem_free(route, sz);
			rval = EFAULT;
			break;
		}

		if (nroutes <= 0) {
			kmem_free(route, sz);
			rval = EINVAL;
			break;
		}

		if (copyin(argp, route,
		    sizeof (struct kvm_irq_routing) + (nroutes - 1) *
		    sizeof (struct kvm_irq_routing_entry)) != 0) {
			kmem_free(route, sz);
			rval = EFAULT;
			break;
		}

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			kmem_free(route, sz);
			rval = EINVAL;
			break;
		}

		if (route->nr >= KVM_MAX_IRQ_ROUTES || route->flags) {
			kmem_free(route, sz);
			rval = EINVAL;
			break;
		}

		rval = kvm_set_irq_routing(kvmp, route->entries,
		    route->nr, route->flags);
		kmem_free(route, sz);
		*rv = 0;
		break;
	}
#endif /* KVM_CAP_IRQ_ROUTING */
	case KVM_IRQ_LINE_STATUS:
	case KVM_IRQ_LINE: {
		struct kvm_irq_level level;
		struct kvm *kvmp;
		size_t sz = sizeof (struct kvm_irq_level);
		int32_t status;

		if (copyin(argp, &level, sz) != 0) {
			rval = EFAULT;
			break;
		}

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (!irqchip_in_kernel(kvmp)) {
			rval = ENXIO;
			break;
		}

		status = kvm_set_irq(kvmp, KVM_USERSPACE_IRQ_SOURCE_ID,
		    level.irq, level.level);

		if (cmd == KVM_IRQ_LINE_STATUS) {
			level.status = status;

			if (copyout(&level, argp, sz) != 0) {
				rval = EFAULT;
				break;
			}
		}

		break;
	}

	case KVM_GET_IRQCHIP: {
		struct kvm *kvmp;
		struct kvm_irqchip chip;
		size_t sz = sizeof (struct kvm_irqchip);
		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (!irqchip_in_kernel(kvmp)) {
			rval = ENXIO;
			break;
		}

		rval = kvm_vm_ioctl_get_irqchip(kvmp, &chip);

		if (rval == 0 && copyout(&chip, argp, sz) != 0) {
			rval = EFAULT;
			break;
		}

		break;
	}

	case KVM_SET_IRQCHIP: {
		struct kvm *kvmp;
		struct kvm_irqchip chip;
		size_t sz = sizeof (struct kvm_irqchip);
		/* 0: PIC master, 1: PIC slave, 2: IOAPIC */

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (copyin(argp, &chip, sizeof (struct kvm_irqchip)) != 0) {
			rval = EFAULT;
			break;
		}

		if (!irqchip_in_kernel(kvmp)) {
			rval = ENXIO;
			break;
		}

		rval = kvm_vm_ioctl_set_irqchip(kvmp, &chip);
		break;
	}
	case KVM_GET_DIRTY_LOG: {
		struct kvm_dirty_log log;
		struct kvm *kvmp;

		if ((kvmp = ksp->kds_kvmp) == NULL) {
			rval = EINVAL;
			break;
		}

		if (copyin(argp, &log, sizeof (struct kvm_dirty_log)) != 0) {
			rval = EFAULT;
			break;
		}

		rval = kvm_vm_ioctl_get_dirty_log(kvmp, &log);
		break;
	}
	case KVM_NMI: {

		if (ksp->kds_kvmp == NULL) {
			rval = EINVAL;
			break;
		}

		if (ksp->kds_vcpu == NULL) {
			rval = EINVAL;
			break;
		}

		rval = kvm_vcpu_ioctl_nmi(ksp->kds_vcpu);
		break;
	}
	case KVM_NET_QUEUE: {
		struct vnode *vn;
		file_t *fp;
		struct stroptions *stropt;
		mblk_t *mp;
		queue_t *q;

		fp = getf(arg);
		if (fp == NULL) {
			rval = EINVAL;
			break;
		}
		ASSERT(fp->f_vnode);

		if (fp->f_vnode->v_stream == NULL) {
			releasef(arg);
			rval = EINVAL;
			break;
		}

		mp = allocb(sizeof (struct stroptions), BPRI_LO);
		if (mp == NULL) {
			releasef(arg);
			rval = ENOMEM;
		}

		/*
		 * XXX This really just shouldn't need to exist, etc. and we
		 * should really get the hiwat value more intelligently at least
		 * a #define or a tunable god forbid. Oh well, as bmc said
		 * earlier:
		 * "I am in blood steeped in so far that I wade no more.
		 * Returning were as tedious as go o'er.
		 *
		 * We'd love to just putmsg on RD(fp->f_vnode->v_stream->sd_wq)
		 * however that would be the stream head. Instead, we need to
		 * get the write version and then go to the next one and then
		 * the opposite end. The doctor may hemorrhage before the
		 * patient.
		 *
		 * Banquo's ghost is waiting to pop up
		 */
		mp->b_datap->db_type = M_SETOPTS;
		stropt = (struct stroptions *)mp->b_rptr;
		stropt->so_flags = SO_HIWAT;
		stropt->so_hiwat = kvm_hiwat;
		q = WR(fp->f_vnode->v_stream->sd_wrq);
		q = RD(q->q_next);
		putnext(q, mp);

		releasef(arg);

		rval = 0;
		*rv = 0;
		break;
	}
	default:
#ifndef XXX
		XXX_KVM_PROBE;
		DTRACE_PROBE1(kvm__xxx__ioctl, int, cmd);
#endif
		rval = EINVAL;  /* x64, others may do other things... */
	}

	if (*rv == -1)
		return (EINVAL);

	return (rval < 0 ? -rval : rval);
}

/* BEGIN CSTYLED */

/*
 * mmap(2), segmap(9E), and devmap(9E)
 *
 * Users call mmap(2). For each call to mmap(2) there is a corresponding call to
 * segmap(9E). segmap(9E) is responsible for making sure that the various
 * requests in the mmap call make sense from the question of protection,
 * offsets, lengths, etc. It then ends by calling the ddi_devmap_segmap() which
 * is what is responsible for making all of the actual mappings.
 *
 * The devmap entry point is called a variable number of times. It is called a
 * number of times until all the maplen values equal the original length of the
 * requested mapping. This allows us to make several different mappings by not
 * honoring the full requested mapping the first time. Each subsequent time it
 * is called with an updated offset and length.
 */


/*
 * We can only create one mapping per dhp. We know whether this is the first
 * time or the second time in based on the requested offset / length. If we only
 * have one page worth, then it's always looking for the shared mmio page. If it
 * is asking for KVM_VCPU_MMAP_LENGTH pages, then it's asking for the shared
 * vcpu pages.
 */
static int
kvm_devmap(dev_t dev, devmap_cookie_t dhp, offset_t off, size_t len,
    size_t *maplen, uint_t model)
{
	int res;
	minor_t instance;
	kvm_devstate_t *ksp;
	kvm_vcpu_t *vcpu;

	instance = getminor(dev);
	ksp = ddi_get_soft_state(kvm_state, instance);
	if (ksp == NULL)
		return (ENXIO);

	/*
	 * Enforce that only 64-bit guests are allowed.
	 */
	if (ddi_model_convert_from(model) == DDI_MODEL_ILP32)
		return (EINVAL);

	/* Double check for betrayl */
	if (ksp->kds_kvmp == NULL)
		return (EINVAL);

	if (ksp->kds_vcpu == NULL)
		return (EINVAL);

	vcpu = ksp->kds_vcpu;

	if (len == PAGESIZE) {
		res = devmap_umem_setup(dhp, kvm_dip, NULL,
		    ksp->kds_kvmp->mmio_cookie, 0, len, PROT_READ | PROT_WRITE |
		    PROT_USER, DEVMAP_DEFAULTS, NULL);
		*maplen = len;
		return (res);
	}

	res = devmap_umem_setup(dhp, kvm_dip, NULL, vcpu->cookie, 0,
	    PAGESIZE*2, PROT_READ | PROT_WRITE | PROT_USER, DEVMAP_DEFAULTS,
	    NULL);

	*maplen = PAGESIZE*2;

	return (res);
}

/*
 * We determine which vcpu we're trying to mmap in based upon the file
 * descriptor that is used. For a given vcpu n the offset to specify it is
 * n*KVM_VCPU_MMAP_LENGTH. Thus the first vcpu is at offset 0. 
 */
static int
kvm_segmap(dev_t dev, off_t off, struct as *asp, caddr_t *addrp, off_t len,
    unsigned int prot, unsigned int maxprot, unsigned int flags,
    cred_t *credp)
{
	kvm_devstate_t *ksp;
	off_t poff;

	if ((ksp = ddi_get_soft_state(kvm_state, getminor(dev))) == NULL)
		return (ENXIO);

	if (prot & PROT_EXEC)
		return (EINVAL);

	if (!(prot & PROT_USER))
	    return (EINVAL);

	if (len != ptob(KVM_VCPU_MMAP_LENGTH))
		return (EINVAL);

	/*
	 * Verify that we have a VCPU
	 */
	if (ksp->kds_vcpu == NULL)
		return (EINVAL);

	/*
	 * We only allow mmaping at a specific cpu
	 */
	if (off != 0)
		return (EINVAL);

	return (ddi_devmap_segmap(dev, off, asp, addrp, len, prot, maxprot,
	    flags, credp));
}


static struct cb_ops kvm_cb_ops = {
	kvm_open,
	kvm_close,	/* close */
	nodev,
	nodev,
	nodev,		/* dump */
	nodev,		/* read */
	nodev,		/* write */
	kvm_ioctl,
	kvm_devmap,
	nodev,		/* mmap */
	kvm_segmap,	/* segmap */
	nochpoll,	/* poll */
	ddi_prop_op,
	NULL,
	D_NEW | D_MP | D_DEVMAP
};

static struct dev_ops kvm_ops = {
	DEVO_REV,
	0,
	kvm_getinfo,
	nulldev,	/* identify */
	nulldev,	/* probe */
	kvm_attach,
	kvm_detach,
	nodev,		/* reset */
	&kvm_cb_ops,
	(struct bus_ops *)0
};

static struct modldrv modldrv = {
	&mod_driverops,
	"kvm driver v0.1",
	&kvm_ops
};

static struct modlinkage modlinkage = {
	MODREV_1,
	{ &modldrv, NULL }
};

int
_init(void)
{

	return (mod_install(&modlinkage));
}

int
_fini(void)
{
	return (mod_remove(&modlinkage));
}

int
_info(struct modinfo *modinfop)
{
	return (mod_info(&modlinkage, modinfop));
}
/* END CSTYLED */