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
|
/*
* Copyright (c) 1991, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* The Mach Operating System project at Carnegie-Mellon University.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* from: @(#)vm_map.c 8.3 (Berkeley) 1/12/94
*
*
* Copyright (c) 1987, 1990 Carnegie-Mellon University.
* All rights reserved.
*
* Authors: Avadis Tevanian, Jr., Michael Wayne Young
*
* Permission to use, copy, modify and distribute this software and
* its documentation is hereby granted, provided that both the copyright
* notice and this permission notice appear in all copies of the
* software, derivative works or modified versions, and any portions
* thereof, and that both notices appear in supporting documentation.
*
* CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
* CONDITION. CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND
* FOR ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
*
* Carnegie Mellon requests users of this software to return to
*
* Software Distribution Coordinator or Software.Distribution@CS.CMU.EDU
* School of Computer Science
* Carnegie Mellon University
* Pittsburgh PA 15213-3890
*
* any improvements or extensions that they make and grant Carnegie the
* rights to redistribute these changes.
*/
/*
* Virtual memory mapping module.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/ktr.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/vmmeter.h>
#include <sys/mman.h>
#include <sys/vnode.h>
#include <sys/resourcevar.h>
#include <sys/file.h>
#include <sys/sysent.h>
#include <sys/shm.h>
#include <vm/vm.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <vm/vm_page.h>
#include <vm/vm_object.h>
#include <vm/vm_pager.h>
#include <vm/vm_kern.h>
#include <vm/vm_extern.h>
#include <vm/swap_pager.h>
#include <vm/uma.h>
/*
* Virtual memory maps provide for the mapping, protection,
* and sharing of virtual memory objects. In addition,
* this module provides for an efficient virtual copy of
* memory from one map to another.
*
* Synchronization is required prior to most operations.
*
* Maps consist of an ordered doubly-linked list of simple
* entries; a single hint is used to speed up lookups.
*
* Since portions of maps are specified by start/end addresses,
* which may not align with existing map entries, all
* routines merely "clip" entries to these start/end values.
* [That is, an entry is split into two, bordering at a
* start or end value.] Note that these clippings may not
* always be necessary (as the two resulting entries are then
* not changed); however, the clipping is done for convenience.
*
* As mentioned above, virtual copy operations are performed
* by copying VM object references from one map to
* another, and then marking both regions as copy-on-write.
*/
/*
* vm_map_startup:
*
* Initialize the vm_map module. Must be called before
* any other vm_map routines.
*
* Map and entry structures are allocated from the general
* purpose memory pool with some exceptions:
*
* - The kernel map and kmem submap are allocated statically.
* - Kernel map entries are allocated out of a static pool.
*
* These restrictions are necessary since malloc() uses the
* maps and requires map entries.
*/
static struct mtx map_sleep_mtx;
static uma_zone_t mapentzone;
static uma_zone_t kmapentzone;
static uma_zone_t mapzone;
static uma_zone_t vmspace_zone;
static struct vm_object kmapentobj;
static void vmspace_zinit(void *mem, int size);
static void vmspace_zfini(void *mem, int size);
static void vm_map_zinit(void *mem, int size);
static void vm_map_zfini(void *mem, int size);
static void _vm_map_init(vm_map_t map, vm_offset_t min, vm_offset_t max);
#ifdef INVARIANTS
static void vm_map_zdtor(void *mem, int size, void *arg);
static void vmspace_zdtor(void *mem, int size, void *arg);
#endif
void
vm_map_startup(void)
{
mtx_init(&map_sleep_mtx, "vm map sleep mutex", NULL, MTX_DEF);
mapzone = uma_zcreate("MAP", sizeof(struct vm_map), NULL,
#ifdef INVARIANTS
vm_map_zdtor,
#else
NULL,
#endif
vm_map_zinit, vm_map_zfini, UMA_ALIGN_PTR, UMA_ZONE_NOFREE);
uma_prealloc(mapzone, MAX_KMAP);
kmapentzone = uma_zcreate("KMAP ENTRY", sizeof(struct vm_map_entry),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR,
UMA_ZONE_MTXCLASS | UMA_ZONE_VM);
uma_prealloc(kmapentzone, MAX_KMAPENT);
mapentzone = uma_zcreate("MAP ENTRY", sizeof(struct vm_map_entry),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
uma_prealloc(mapentzone, MAX_MAPENT);
}
static void
vmspace_zfini(void *mem, int size)
{
struct vmspace *vm;
vm = (struct vmspace *)mem;
pmap_release(vmspace_pmap(vm));
vm_map_zfini(&vm->vm_map, sizeof(vm->vm_map));
}
static void
vmspace_zinit(void *mem, int size)
{
struct vmspace *vm;
vm = (struct vmspace *)mem;
vm_map_zinit(&vm->vm_map, sizeof(vm->vm_map));
pmap_pinit(vmspace_pmap(vm));
}
static void
vm_map_zfini(void *mem, int size)
{
vm_map_t map;
map = (vm_map_t)mem;
mtx_destroy(&map->system_mtx);
lockdestroy(&map->lock);
}
static void
vm_map_zinit(void *mem, int size)
{
vm_map_t map;
map = (vm_map_t)mem;
map->nentries = 0;
map->size = 0;
map->infork = 0;
mtx_init(&map->system_mtx, "system map", NULL, MTX_DEF | MTX_DUPOK);
lockinit(&map->lock, PVM, "thrd_sleep", 0, LK_NOPAUSE);
}
#ifdef INVARIANTS
static void
vmspace_zdtor(void *mem, int size, void *arg)
{
struct vmspace *vm;
vm = (struct vmspace *)mem;
vm_map_zdtor(&vm->vm_map, sizeof(vm->vm_map), arg);
}
static void
vm_map_zdtor(void *mem, int size, void *arg)
{
vm_map_t map;
map = (vm_map_t)mem;
KASSERT(map->nentries == 0,
("map %p nentries == %d on free.",
map, map->nentries));
KASSERT(map->size == 0,
("map %p size == %lu on free.",
map, (unsigned long)map->size));
KASSERT(map->infork == 0,
("map %p infork == %d on free.",
map, map->infork));
}
#endif /* INVARIANTS */
/*
* Allocate a vmspace structure, including a vm_map and pmap,
* and initialize those structures. The refcnt is set to 1.
*/
struct vmspace *
vmspace_alloc(min, max)
vm_offset_t min, max;
{
struct vmspace *vm;
vm = uma_zalloc(vmspace_zone, M_WAITOK);
CTR1(KTR_VM, "vmspace_alloc: %p", vm);
_vm_map_init(&vm->vm_map, min, max);
vm->vm_map.pmap = vmspace_pmap(vm); /* XXX */
vm->vm_refcnt = 1;
vm->vm_shm = NULL;
vm->vm_swrss = 0;
vm->vm_tsize = 0;
vm->vm_dsize = 0;
vm->vm_ssize = 0;
vm->vm_taddr = 0;
vm->vm_daddr = 0;
vm->vm_maxsaddr = 0;
vm->vm_exitingcnt = 0;
return (vm);
}
void
vm_init2(void)
{
uma_zone_set_obj(kmapentzone, &kmapentobj, lmin(cnt.v_page_count,
(VM_MAX_KERNEL_ADDRESS - KERNBASE) / PAGE_SIZE) / 8 +
maxproc * 2 + maxfiles);
vmspace_zone = uma_zcreate("VMSPACE", sizeof(struct vmspace), NULL,
#ifdef INVARIANTS
vmspace_zdtor,
#else
NULL,
#endif
vmspace_zinit, vmspace_zfini, UMA_ALIGN_PTR, UMA_ZONE_NOFREE);
pmap_init2();
}
static __inline void
vmspace_dofree(struct vmspace *vm)
{
CTR1(KTR_VM, "vmspace_free: %p", vm);
/*
* Make sure any SysV shm is freed, it might not have been in
* exit1().
*/
shmexit(vm);
/*
* Lock the map, to wait out all other references to it.
* Delete all of the mappings and pages they hold, then call
* the pmap module to reclaim anything left.
*/
vm_map_lock(&vm->vm_map);
(void) vm_map_delete(&vm->vm_map, vm->vm_map.min_offset,
vm->vm_map.max_offset);
vm_map_unlock(&vm->vm_map);
uma_zfree(vmspace_zone, vm);
}
void
vmspace_free(struct vmspace *vm)
{
GIANT_REQUIRED;
if (vm->vm_refcnt == 0)
panic("vmspace_free: attempt to free already freed vmspace");
--vm->vm_refcnt;
if (vm->vm_refcnt == 0 && vm->vm_exitingcnt == 0)
vmspace_dofree(vm);
}
void
vmspace_exitfree(struct proc *p)
{
struct vmspace *vm;
GIANT_REQUIRED;
vm = p->p_vmspace;
p->p_vmspace = NULL;
/*
* cleanup by parent process wait()ing on exiting child. vm_refcnt
* may not be 0 (e.g. fork() and child exits without exec()ing).
* exitingcnt may increment above 0 and drop back down to zero
* several times while vm_refcnt is held non-zero. vm_refcnt
* may also increment above 0 and drop back down to zero several
* times while vm_exitingcnt is held non-zero.
*
* The last wait on the exiting child's vmspace will clean up
* the remainder of the vmspace.
*/
--vm->vm_exitingcnt;
if (vm->vm_refcnt == 0 && vm->vm_exitingcnt == 0)
vmspace_dofree(vm);
}
void
_vm_map_lock(vm_map_t map, const char *file, int line)
{
int error;
if (map->system_map)
_mtx_lock_flags(&map->system_mtx, 0, file, line);
else {
error = lockmgr(&map->lock, LK_EXCLUSIVE, NULL, curthread);
KASSERT(error == 0, ("%s: failed to get lock", __func__));
}
map->timestamp++;
}
void
_vm_map_unlock(vm_map_t map, const char *file, int line)
{
if (map->system_map)
_mtx_unlock_flags(&map->system_mtx, 0, file, line);
else
lockmgr(&map->lock, LK_RELEASE, NULL, curthread);
}
void
_vm_map_lock_read(vm_map_t map, const char *file, int line)
{
int error;
if (map->system_map)
_mtx_lock_flags(&map->system_mtx, 0, file, line);
else {
error = lockmgr(&map->lock, LK_EXCLUSIVE, NULL, curthread);
KASSERT(error == 0, ("%s: failed to get lock", __func__));
}
}
void
_vm_map_unlock_read(vm_map_t map, const char *file, int line)
{
if (map->system_map)
_mtx_unlock_flags(&map->system_mtx, 0, file, line);
else
lockmgr(&map->lock, LK_RELEASE, NULL, curthread);
}
int
_vm_map_trylock(vm_map_t map, const char *file, int line)
{
int error;
error = map->system_map ?
!_mtx_trylock(&map->system_mtx, 0, file, line) :
lockmgr(&map->lock, LK_EXCLUSIVE | LK_NOWAIT, NULL, curthread);
if (error == 0)
map->timestamp++;
return (error == 0);
}
int
_vm_map_trylock_read(vm_map_t map, const char *file, int line)
{
int error;
error = map->system_map ?
!_mtx_trylock(&map->system_mtx, 0, file, line) :
lockmgr(&map->lock, LK_EXCLUSIVE | LK_NOWAIT, NULL, curthread);
return (error == 0);
}
int
_vm_map_lock_upgrade(vm_map_t map, const char *file, int line)
{
if (map->system_map) {
#ifdef INVARIANTS
_mtx_assert(&map->system_mtx, MA_OWNED, file, line);
#endif
} else
KASSERT(lockstatus(&map->lock, curthread) == LK_EXCLUSIVE,
("%s: lock not held", __func__));
map->timestamp++;
return (0);
}
void
_vm_map_lock_downgrade(vm_map_t map, const char *file, int line)
{
if (map->system_map) {
#ifdef INVARIANTS
_mtx_assert(&map->system_mtx, MA_OWNED, file, line);
#endif
} else
KASSERT(lockstatus(&map->lock, curthread) == LK_EXCLUSIVE,
("%s: lock not held", __func__));
}
/*
* vm_map_unlock_and_wait:
*/
int
vm_map_unlock_and_wait(vm_map_t map, boolean_t user_wait)
{
mtx_lock(&map_sleep_mtx);
vm_map_unlock(map);
return (msleep(&map->root, &map_sleep_mtx, PDROP | PVM, "vmmaps", 0));
}
/*
* vm_map_wakeup:
*/
void
vm_map_wakeup(vm_map_t map)
{
/*
* Acquire and release map_sleep_mtx to prevent a wakeup()
* from being performed (and lost) between the vm_map_unlock()
* and the msleep() in vm_map_unlock_and_wait().
*/
mtx_lock(&map_sleep_mtx);
mtx_unlock(&map_sleep_mtx);
wakeup(&map->root);
}
long
vmspace_resident_count(struct vmspace *vmspace)
{
return pmap_resident_count(vmspace_pmap(vmspace));
}
long
vmspace_wired_count(struct vmspace *vmspace)
{
return pmap_wired_count(vmspace_pmap(vmspace));
}
/*
* vm_map_create:
*
* Creates and returns a new empty VM map with
* the given physical map structure, and having
* the given lower and upper address bounds.
*/
vm_map_t
vm_map_create(pmap_t pmap, vm_offset_t min, vm_offset_t max)
{
vm_map_t result;
result = uma_zalloc(mapzone, M_WAITOK);
CTR1(KTR_VM, "vm_map_create: %p", result);
_vm_map_init(result, min, max);
result->pmap = pmap;
return (result);
}
/*
* Initialize an existing vm_map structure
* such as that in the vmspace structure.
* The pmap is set elsewhere.
*/
static void
_vm_map_init(vm_map_t map, vm_offset_t min, vm_offset_t max)
{
map->header.next = map->header.prev = &map->header;
map->needs_wakeup = FALSE;
map->system_map = 0;
map->min_offset = min;
map->max_offset = max;
map->first_free = &map->header;
map->flags = 0;
map->root = NULL;
map->timestamp = 0;
}
void
vm_map_init(vm_map_t map, vm_offset_t min, vm_offset_t max)
{
_vm_map_init(map, min, max);
mtx_init(&map->system_mtx, "system map", NULL, MTX_DEF | MTX_DUPOK);
lockinit(&map->lock, PVM, "thrd_sleep", 0, LK_NOPAUSE);
}
/*
* vm_map_entry_dispose: [ internal use only ]
*
* Inverse of vm_map_entry_create.
*/
static void
vm_map_entry_dispose(vm_map_t map, vm_map_entry_t entry)
{
uma_zfree(map->system_map ? kmapentzone : mapentzone, entry);
}
/*
* vm_map_entry_create: [ internal use only ]
*
* Allocates a VM map entry for insertion.
* No entry fields are filled in.
*/
static vm_map_entry_t
vm_map_entry_create(vm_map_t map)
{
vm_map_entry_t new_entry;
if (map->system_map)
new_entry = uma_zalloc(kmapentzone, M_NOWAIT);
else
new_entry = uma_zalloc(mapentzone, M_WAITOK);
if (new_entry == NULL)
panic("vm_map_entry_create: kernel resources exhausted");
return (new_entry);
}
/*
* vm_map_entry_set_behavior:
*
* Set the expected access behavior, either normal, random, or
* sequential.
*/
static __inline void
vm_map_entry_set_behavior(vm_map_entry_t entry, u_char behavior)
{
entry->eflags = (entry->eflags & ~MAP_ENTRY_BEHAV_MASK) |
(behavior & MAP_ENTRY_BEHAV_MASK);
}
/*
* vm_map_entry_splay:
*
* Implements Sleator and Tarjan's top-down splay algorithm. Returns
* the vm_map_entry containing the given address. If, however, that
* address is not found in the vm_map, returns a vm_map_entry that is
* adjacent to the address, coming before or after it.
*/
static vm_map_entry_t
vm_map_entry_splay(vm_offset_t address, vm_map_entry_t root)
{
struct vm_map_entry dummy;
vm_map_entry_t lefttreemax, righttreemin, y;
if (root == NULL)
return (root);
lefttreemax = righttreemin = &dummy;
for (;; root = y) {
if (address < root->start) {
if ((y = root->left) == NULL)
break;
if (address < y->start) {
/* Rotate right. */
root->left = y->right;
y->right = root;
root = y;
if ((y = root->left) == NULL)
break;
}
/* Link into the new root's right tree. */
righttreemin->left = root;
righttreemin = root;
} else if (address >= root->end) {
if ((y = root->right) == NULL)
break;
if (address >= y->end) {
/* Rotate left. */
root->right = y->left;
y->left = root;
root = y;
if ((y = root->right) == NULL)
break;
}
/* Link into the new root's left tree. */
lefttreemax->right = root;
lefttreemax = root;
} else
break;
}
/* Assemble the new root. */
lefttreemax->right = root->left;
righttreemin->left = root->right;
root->left = dummy.right;
root->right = dummy.left;
return (root);
}
/*
* vm_map_entry_{un,}link:
*
* Insert/remove entries from maps.
*/
static void
vm_map_entry_link(vm_map_t map,
vm_map_entry_t after_where,
vm_map_entry_t entry)
{
CTR4(KTR_VM,
"vm_map_entry_link: map %p, nentries %d, entry %p, after %p", map,
map->nentries, entry, after_where);
map->nentries++;
entry->prev = after_where;
entry->next = after_where->next;
entry->next->prev = entry;
after_where->next = entry;
if (after_where != &map->header) {
if (after_where != map->root)
vm_map_entry_splay(after_where->start, map->root);
entry->right = after_where->right;
entry->left = after_where;
after_where->right = NULL;
} else {
entry->right = map->root;
entry->left = NULL;
}
map->root = entry;
}
static void
vm_map_entry_unlink(vm_map_t map,
vm_map_entry_t entry)
{
vm_map_entry_t next, prev, root;
if (entry != map->root)
vm_map_entry_splay(entry->start, map->root);
if (entry->left == NULL)
root = entry->right;
else {
root = vm_map_entry_splay(entry->start, entry->left);
root->right = entry->right;
}
map->root = root;
prev = entry->prev;
next = entry->next;
next->prev = prev;
prev->next = next;
map->nentries--;
CTR3(KTR_VM, "vm_map_entry_unlink: map %p, nentries %d, entry %p", map,
map->nentries, entry);
}
/*
* vm_map_lookup_entry: [ internal use only ]
*
* Finds the map entry containing (or
* immediately preceding) the specified address
* in the given map; the entry is returned
* in the "entry" parameter. The boolean
* result indicates whether the address is
* actually contained in the map.
*/
boolean_t
vm_map_lookup_entry(
vm_map_t map,
vm_offset_t address,
vm_map_entry_t *entry) /* OUT */
{
vm_map_entry_t cur;
cur = vm_map_entry_splay(address, map->root);
if (cur == NULL)
*entry = &map->header;
else {
map->root = cur;
if (address >= cur->start) {
*entry = cur;
if (cur->end > address)
return (TRUE);
} else
*entry = cur->prev;
}
return (FALSE);
}
/*
* vm_map_insert:
*
* Inserts the given whole VM object into the target
* map at the specified address range. The object's
* size should match that of the address range.
*
* Requires that the map be locked, and leaves it so.
*
* If object is non-NULL, ref count must be bumped by caller
* prior to making call to account for the new entry.
*/
int
vm_map_insert(vm_map_t map, vm_object_t object, vm_ooffset_t offset,
vm_offset_t start, vm_offset_t end, vm_prot_t prot, vm_prot_t max,
int cow)
{
vm_map_entry_t new_entry;
vm_map_entry_t prev_entry;
vm_map_entry_t temp_entry;
vm_eflags_t protoeflags;
/*
* Check that the start and end points are not bogus.
*/
if ((start < map->min_offset) || (end > map->max_offset) ||
(start >= end))
return (KERN_INVALID_ADDRESS);
/*
* Find the entry prior to the proposed starting address; if it's part
* of an existing entry, this range is bogus.
*/
if (vm_map_lookup_entry(map, start, &temp_entry))
return (KERN_NO_SPACE);
prev_entry = temp_entry;
/*
* Assert that the next entry doesn't overlap the end point.
*/
if ((prev_entry->next != &map->header) &&
(prev_entry->next->start < end))
return (KERN_NO_SPACE);
protoeflags = 0;
if (cow & MAP_COPY_ON_WRITE)
protoeflags |= MAP_ENTRY_COW|MAP_ENTRY_NEEDS_COPY;
if (cow & MAP_NOFAULT) {
protoeflags |= MAP_ENTRY_NOFAULT;
KASSERT(object == NULL,
("vm_map_insert: paradoxical MAP_NOFAULT request"));
}
if (cow & MAP_DISABLE_SYNCER)
protoeflags |= MAP_ENTRY_NOSYNC;
if (cow & MAP_DISABLE_COREDUMP)
protoeflags |= MAP_ENTRY_NOCOREDUMP;
if (object != NULL) {
/*
* OBJ_ONEMAPPING must be cleared unless this mapping
* is trivially proven to be the only mapping for any
* of the object's pages. (Object granularity
* reference counting is insufficient to recognize
* aliases with precision.)
*/
VM_OBJECT_LOCK(object);
if (object->ref_count > 1 || object->shadow_count != 0)
vm_object_clear_flag(object, OBJ_ONEMAPPING);
VM_OBJECT_UNLOCK(object);
}
else if ((prev_entry != &map->header) &&
(prev_entry->eflags == protoeflags) &&
(prev_entry->end == start) &&
(prev_entry->wired_count == 0) &&
((prev_entry->object.vm_object == NULL) ||
vm_object_coalesce(prev_entry->object.vm_object,
OFF_TO_IDX(prev_entry->offset),
(vm_size_t)(prev_entry->end - prev_entry->start),
(vm_size_t)(end - prev_entry->end)))) {
/*
* We were able to extend the object. Determine if we
* can extend the previous map entry to include the
* new range as well.
*/
if ((prev_entry->inheritance == VM_INHERIT_DEFAULT) &&
(prev_entry->protection == prot) &&
(prev_entry->max_protection == max)) {
map->size += (end - prev_entry->end);
prev_entry->end = end;
vm_map_simplify_entry(map, prev_entry);
return (KERN_SUCCESS);
}
/*
* If we can extend the object but cannot extend the
* map entry, we have to create a new map entry. We
* must bump the ref count on the extended object to
* account for it. object may be NULL.
*/
object = prev_entry->object.vm_object;
offset = prev_entry->offset +
(prev_entry->end - prev_entry->start);
vm_object_reference(object);
}
/*
* NOTE: if conditionals fail, object can be NULL here. This occurs
* in things like the buffer map where we manage kva but do not manage
* backing objects.
*/
/*
* Create a new entry
*/
new_entry = vm_map_entry_create(map);
new_entry->start = start;
new_entry->end = end;
new_entry->eflags = protoeflags;
new_entry->object.vm_object = object;
new_entry->offset = offset;
new_entry->avail_ssize = 0;
new_entry->inheritance = VM_INHERIT_DEFAULT;
new_entry->protection = prot;
new_entry->max_protection = max;
new_entry->wired_count = 0;
/*
* Insert the new entry into the list
*/
vm_map_entry_link(map, prev_entry, new_entry);
map->size += new_entry->end - new_entry->start;
/*
* Update the free space hint
*/
if ((map->first_free == prev_entry) &&
(prev_entry->end >= new_entry->start)) {
map->first_free = new_entry;
}
#if 0
/*
* Temporarily removed to avoid MAP_STACK panic, due to
* MAP_STACK being a huge hack. Will be added back in
* when MAP_STACK (and the user stack mapping) is fixed.
*/
/*
* It may be possible to simplify the entry
*/
vm_map_simplify_entry(map, new_entry);
#endif
if (cow & (MAP_PREFAULT|MAP_PREFAULT_PARTIAL)) {
vm_map_pmap_enter(map, start, prot,
object, OFF_TO_IDX(offset), end - start,
cow & MAP_PREFAULT_PARTIAL);
}
return (KERN_SUCCESS);
}
/*
* Find sufficient space for `length' bytes in the given map, starting at
* `start'. The map must be locked. Returns 0 on success, 1 on no space.
*/
int
vm_map_findspace(
vm_map_t map,
vm_offset_t start,
vm_size_t length,
vm_offset_t *addr)
{
vm_map_entry_t entry, next;
vm_offset_t end;
if (start < map->min_offset)
start = map->min_offset;
if (start > map->max_offset)
return (1);
/*
* Look for the first possible address; if there's already something
* at this address, we have to start after it.
*/
if (start == map->min_offset) {
if ((entry = map->first_free) != &map->header)
start = entry->end;
} else {
vm_map_entry_t tmp;
if (vm_map_lookup_entry(map, start, &tmp))
start = tmp->end;
entry = tmp;
}
/*
* Look through the rest of the map, trying to fit a new region in the
* gap between existing regions, or after the very last region.
*/
for (;; start = (entry = next)->end) {
/*
* Find the end of the proposed new region. Be sure we didn't
* go beyond the end of the map, or wrap around the address;
* if so, we lose. Otherwise, if this is the last entry, or
* if the proposed new region fits before the next entry, we
* win.
*/
end = start + length;
if (end > map->max_offset || end < start)
return (1);
next = entry->next;
if (next == &map->header || next->start >= end)
break;
}
*addr = start;
if (map == kernel_map) {
vm_offset_t ksize;
if ((ksize = round_page(start + length)) > kernel_vm_end) {
pmap_growkernel(ksize);
}
}
return (0);
}
/*
* vm_map_find finds an unallocated region in the target address
* map with the given length. The search is defined to be
* first-fit from the specified address; the region found is
* returned in the same parameter.
*
* If object is non-NULL, ref count must be bumped by caller
* prior to making call to account for the new entry.
*/
int
vm_map_find(vm_map_t map, vm_object_t object, vm_ooffset_t offset,
vm_offset_t *addr, /* IN/OUT */
vm_size_t length, boolean_t find_space, vm_prot_t prot,
vm_prot_t max, int cow)
{
vm_offset_t start;
int result, s = 0;
start = *addr;
if (map == kmem_map)
s = splvm();
vm_map_lock(map);
if (find_space) {
if (vm_map_findspace(map, start, length, addr)) {
vm_map_unlock(map);
if (map == kmem_map)
splx(s);
return (KERN_NO_SPACE);
}
start = *addr;
}
result = vm_map_insert(map, object, offset,
start, start + length, prot, max, cow);
vm_map_unlock(map);
if (map == kmem_map)
splx(s);
return (result);
}
/*
* vm_map_simplify_entry:
*
* Simplify the given map entry by merging with either neighbor. This
* routine also has the ability to merge with both neighbors.
*
* The map must be locked.
*
* This routine guarentees that the passed entry remains valid (though
* possibly extended). When merging, this routine may delete one or
* both neighbors.
*/
void
vm_map_simplify_entry(vm_map_t map, vm_map_entry_t entry)
{
vm_map_entry_t next, prev;
vm_size_t prevsize, esize;
if (entry->eflags & (MAP_ENTRY_IN_TRANSITION | MAP_ENTRY_IS_SUB_MAP))
return;
prev = entry->prev;
if (prev != &map->header) {
prevsize = prev->end - prev->start;
if ( (prev->end == entry->start) &&
(prev->object.vm_object == entry->object.vm_object) &&
(!prev->object.vm_object ||
(prev->offset + prevsize == entry->offset)) &&
(prev->eflags == entry->eflags) &&
(prev->protection == entry->protection) &&
(prev->max_protection == entry->max_protection) &&
(prev->inheritance == entry->inheritance) &&
(prev->wired_count == entry->wired_count)) {
if (map->first_free == prev)
map->first_free = entry;
vm_map_entry_unlink(map, prev);
entry->start = prev->start;
entry->offset = prev->offset;
if (prev->object.vm_object)
vm_object_deallocate(prev->object.vm_object);
vm_map_entry_dispose(map, prev);
}
}
next = entry->next;
if (next != &map->header) {
esize = entry->end - entry->start;
if ((entry->end == next->start) &&
(next->object.vm_object == entry->object.vm_object) &&
(!entry->object.vm_object ||
(entry->offset + esize == next->offset)) &&
(next->eflags == entry->eflags) &&
(next->protection == entry->protection) &&
(next->max_protection == entry->max_protection) &&
(next->inheritance == entry->inheritance) &&
(next->wired_count == entry->wired_count)) {
if (map->first_free == next)
map->first_free = entry;
vm_map_entry_unlink(map, next);
entry->end = next->end;
if (next->object.vm_object)
vm_object_deallocate(next->object.vm_object);
vm_map_entry_dispose(map, next);
}
}
}
/*
* vm_map_clip_start: [ internal use only ]
*
* Asserts that the given entry begins at or after
* the specified address; if necessary,
* it splits the entry into two.
*/
#define vm_map_clip_start(map, entry, startaddr) \
{ \
if (startaddr > entry->start) \
_vm_map_clip_start(map, entry, startaddr); \
}
/*
* This routine is called only when it is known that
* the entry must be split.
*/
static void
_vm_map_clip_start(vm_map_t map, vm_map_entry_t entry, vm_offset_t start)
{
vm_map_entry_t new_entry;
/*
* Split off the front portion -- note that we must insert the new
* entry BEFORE this one, so that this entry has the specified
* starting address.
*/
vm_map_simplify_entry(map, entry);
/*
* If there is no object backing this entry, we might as well create
* one now. If we defer it, an object can get created after the map
* is clipped, and individual objects will be created for the split-up
* map. This is a bit of a hack, but is also about the best place to
* put this improvement.
*/
if (entry->object.vm_object == NULL && !map->system_map) {
vm_object_t object;
object = vm_object_allocate(OBJT_DEFAULT,
atop(entry->end - entry->start));
entry->object.vm_object = object;
entry->offset = 0;
}
new_entry = vm_map_entry_create(map);
*new_entry = *entry;
new_entry->end = start;
entry->offset += (start - entry->start);
entry->start = start;
vm_map_entry_link(map, entry->prev, new_entry);
if ((entry->eflags & MAP_ENTRY_IS_SUB_MAP) == 0) {
vm_object_reference(new_entry->object.vm_object);
}
}
/*
* vm_map_clip_end: [ internal use only ]
*
* Asserts that the given entry ends at or before
* the specified address; if necessary,
* it splits the entry into two.
*/
#define vm_map_clip_end(map, entry, endaddr) \
{ \
if ((endaddr) < (entry->end)) \
_vm_map_clip_end((map), (entry), (endaddr)); \
}
/*
* This routine is called only when it is known that
* the entry must be split.
*/
static void
_vm_map_clip_end(vm_map_t map, vm_map_entry_t entry, vm_offset_t end)
{
vm_map_entry_t new_entry;
/*
* If there is no object backing this entry, we might as well create
* one now. If we defer it, an object can get created after the map
* is clipped, and individual objects will be created for the split-up
* map. This is a bit of a hack, but is also about the best place to
* put this improvement.
*/
if (entry->object.vm_object == NULL && !map->system_map) {
vm_object_t object;
object = vm_object_allocate(OBJT_DEFAULT,
atop(entry->end - entry->start));
entry->object.vm_object = object;
entry->offset = 0;
}
/*
* Create a new entry and insert it AFTER the specified entry
*/
new_entry = vm_map_entry_create(map);
*new_entry = *entry;
new_entry->start = entry->end = end;
new_entry->offset += (end - entry->start);
vm_map_entry_link(map, entry, new_entry);
if ((entry->eflags & MAP_ENTRY_IS_SUB_MAP) == 0) {
vm_object_reference(new_entry->object.vm_object);
}
}
/*
* VM_MAP_RANGE_CHECK: [ internal use only ]
*
* Asserts that the starting and ending region
* addresses fall within the valid range of the map.
*/
#define VM_MAP_RANGE_CHECK(map, start, end) \
{ \
if (start < vm_map_min(map)) \
start = vm_map_min(map); \
if (end > vm_map_max(map)) \
end = vm_map_max(map); \
if (start > end) \
start = end; \
}
/*
* vm_map_submap: [ kernel use only ]
*
* Mark the given range as handled by a subordinate map.
*
* This range must have been created with vm_map_find,
* and no other operations may have been performed on this
* range prior to calling vm_map_submap.
*
* Only a limited number of operations can be performed
* within this rage after calling vm_map_submap:
* vm_fault
* [Don't try vm_map_copy!]
*
* To remove a submapping, one must first remove the
* range from the superior map, and then destroy the
* submap (if desired). [Better yet, don't try it.]
*/
int
vm_map_submap(
vm_map_t map,
vm_offset_t start,
vm_offset_t end,
vm_map_t submap)
{
vm_map_entry_t entry;
int result = KERN_INVALID_ARGUMENT;
vm_map_lock(map);
VM_MAP_RANGE_CHECK(map, start, end);
if (vm_map_lookup_entry(map, start, &entry)) {
vm_map_clip_start(map, entry, start);
} else
entry = entry->next;
vm_map_clip_end(map, entry, end);
if ((entry->start == start) && (entry->end == end) &&
((entry->eflags & MAP_ENTRY_COW) == 0) &&
(entry->object.vm_object == NULL)) {
entry->object.sub_map = submap;
entry->eflags |= MAP_ENTRY_IS_SUB_MAP;
result = KERN_SUCCESS;
}
vm_map_unlock(map);
return (result);
}
/*
* The maximum number of pages to map
*/
#define MAX_INIT_PT 96
/*
* vm_map_pmap_enter:
*
* Preload read-only mappings for the given object into the specified
* map. This eliminates the soft faults on process startup and
* immediately after an mmap(2).
*/
void
vm_map_pmap_enter(vm_map_t map, vm_offset_t addr, vm_prot_t prot,
vm_object_t object, vm_pindex_t pindex, vm_size_t size, int flags)
{
vm_offset_t tmpidx;
int psize;
vm_page_t p, mpte;
if ((prot & VM_PROT_READ) == 0 || object == NULL)
return;
mtx_lock(&Giant);
VM_OBJECT_LOCK(object);
if (object->type == OBJT_DEVICE) {
pmap_object_init_pt(map->pmap, addr, object, pindex, size);
goto unlock_return;
}
psize = atop(size);
if (object->type != OBJT_VNODE ||
((flags & MAP_PREFAULT_PARTIAL) && (psize > MAX_INIT_PT) &&
(object->resident_page_count > MAX_INIT_PT))) {
goto unlock_return;
}
if (psize + pindex > object->size) {
if (object->size < pindex)
goto unlock_return;
psize = object->size - pindex;
}
mpte = NULL;
if ((p = TAILQ_FIRST(&object->memq)) != NULL) {
if (p->pindex < pindex) {
p = vm_page_splay(pindex, object->root);
if ((object->root = p)->pindex < pindex)
p = TAILQ_NEXT(p, listq);
}
}
/*
* Assert: the variable p is either (1) the page with the
* least pindex greater than or equal to the parameter pindex
* or (2) NULL.
*/
for (;
p != NULL && (tmpidx = p->pindex - pindex) < psize;
p = TAILQ_NEXT(p, listq)) {
/*
* don't allow an madvise to blow away our really
* free pages allocating pv entries.
*/
if ((flags & MAP_PREFAULT_MADVISE) &&
cnt.v_free_count < cnt.v_free_reserved) {
break;
}
vm_page_lock_queues();
if ((p->valid & VM_PAGE_BITS_ALL) == VM_PAGE_BITS_ALL &&
(p->busy == 0) &&
(p->flags & (PG_BUSY | PG_FICTITIOUS)) == 0) {
if ((p->queue - p->pc) == PQ_CACHE)
vm_page_deactivate(p);
vm_page_busy(p);
vm_page_unlock_queues();
VM_OBJECT_UNLOCK(object);
mpte = pmap_enter_quick(map->pmap,
addr + ptoa(tmpidx), p, mpte);
VM_OBJECT_LOCK(object);
vm_page_lock_queues();
vm_page_wakeup(p);
}
vm_page_unlock_queues();
}
unlock_return:
VM_OBJECT_UNLOCK(object);
mtx_unlock(&Giant);
}
/*
* vm_map_protect:
*
* Sets the protection of the specified address
* region in the target map. If "set_max" is
* specified, the maximum protection is to be set;
* otherwise, only the current protection is affected.
*/
int
vm_map_protect(vm_map_t map, vm_offset_t start, vm_offset_t end,
vm_prot_t new_prot, boolean_t set_max)
{
vm_map_entry_t current;
vm_map_entry_t entry;
vm_map_lock(map);
VM_MAP_RANGE_CHECK(map, start, end);
if (vm_map_lookup_entry(map, start, &entry)) {
vm_map_clip_start(map, entry, start);
} else {
entry = entry->next;
}
/*
* Make a first pass to check for protection violations.
*/
current = entry;
while ((current != &map->header) && (current->start < end)) {
if (current->eflags & MAP_ENTRY_IS_SUB_MAP) {
vm_map_unlock(map);
return (KERN_INVALID_ARGUMENT);
}
if ((new_prot & current->max_protection) != new_prot) {
vm_map_unlock(map);
return (KERN_PROTECTION_FAILURE);
}
current = current->next;
}
/*
* Go back and fix up protections. [Note that clipping is not
* necessary the second time.]
*/
current = entry;
while ((current != &map->header) && (current->start < end)) {
vm_prot_t old_prot;
vm_map_clip_end(map, current, end);
old_prot = current->protection;
if (set_max)
current->protection =
(current->max_protection = new_prot) &
old_prot;
else
current->protection = new_prot;
/*
* Update physical map if necessary. Worry about copy-on-write
* here -- CHECK THIS XXX
*/
if (current->protection != old_prot) {
mtx_lock(&Giant);
#define MASK(entry) (((entry)->eflags & MAP_ENTRY_COW) ? ~VM_PROT_WRITE : \
VM_PROT_ALL)
pmap_protect(map->pmap, current->start,
current->end,
current->protection & MASK(current));
#undef MASK
mtx_unlock(&Giant);
}
vm_map_simplify_entry(map, current);
current = current->next;
}
vm_map_unlock(map);
return (KERN_SUCCESS);
}
/*
* vm_map_madvise:
*
* This routine traverses a processes map handling the madvise
* system call. Advisories are classified as either those effecting
* the vm_map_entry structure, or those effecting the underlying
* objects.
*/
int
vm_map_madvise(
vm_map_t map,
vm_offset_t start,
vm_offset_t end,
int behav)
{
vm_map_entry_t current, entry;
int modify_map = 0;
/*
* Some madvise calls directly modify the vm_map_entry, in which case
* we need to use an exclusive lock on the map and we need to perform
* various clipping operations. Otherwise we only need a read-lock
* on the map.
*/
switch(behav) {
case MADV_NORMAL:
case MADV_SEQUENTIAL:
case MADV_RANDOM:
case MADV_NOSYNC:
case MADV_AUTOSYNC:
case MADV_NOCORE:
case MADV_CORE:
modify_map = 1;
vm_map_lock(map);
break;
case MADV_WILLNEED:
case MADV_DONTNEED:
case MADV_FREE:
vm_map_lock_read(map);
break;
default:
return (KERN_INVALID_ARGUMENT);
}
/*
* Locate starting entry and clip if necessary.
*/
VM_MAP_RANGE_CHECK(map, start, end);
if (vm_map_lookup_entry(map, start, &entry)) {
if (modify_map)
vm_map_clip_start(map, entry, start);
} else {
entry = entry->next;
}
if (modify_map) {
/*
* madvise behaviors that are implemented in the vm_map_entry.
*
* We clip the vm_map_entry so that behavioral changes are
* limited to the specified address range.
*/
for (current = entry;
(current != &map->header) && (current->start < end);
current = current->next
) {
if (current->eflags & MAP_ENTRY_IS_SUB_MAP)
continue;
vm_map_clip_end(map, current, end);
switch (behav) {
case MADV_NORMAL:
vm_map_entry_set_behavior(current, MAP_ENTRY_BEHAV_NORMAL);
break;
case MADV_SEQUENTIAL:
vm_map_entry_set_behavior(current, MAP_ENTRY_BEHAV_SEQUENTIAL);
break;
case MADV_RANDOM:
vm_map_entry_set_behavior(current, MAP_ENTRY_BEHAV_RANDOM);
break;
case MADV_NOSYNC:
current->eflags |= MAP_ENTRY_NOSYNC;
break;
case MADV_AUTOSYNC:
current->eflags &= ~MAP_ENTRY_NOSYNC;
break;
case MADV_NOCORE:
current->eflags |= MAP_ENTRY_NOCOREDUMP;
break;
case MADV_CORE:
current->eflags &= ~MAP_ENTRY_NOCOREDUMP;
break;
default:
break;
}
vm_map_simplify_entry(map, current);
}
vm_map_unlock(map);
} else {
vm_pindex_t pindex;
int count;
/*
* madvise behaviors that are implemented in the underlying
* vm_object.
*
* Since we don't clip the vm_map_entry, we have to clip
* the vm_object pindex and count.
*/
for (current = entry;
(current != &map->header) && (current->start < end);
current = current->next
) {
vm_offset_t useStart;
if (current->eflags & MAP_ENTRY_IS_SUB_MAP)
continue;
pindex = OFF_TO_IDX(current->offset);
count = atop(current->end - current->start);
useStart = current->start;
if (current->start < start) {
pindex += atop(start - current->start);
count -= atop(start - current->start);
useStart = start;
}
if (current->end > end)
count -= atop(current->end - end);
if (count <= 0)
continue;
vm_object_madvise(current->object.vm_object,
pindex, count, behav);
if (behav == MADV_WILLNEED) {
vm_map_pmap_enter(map,
useStart,
current->protection,
current->object.vm_object,
pindex,
(count << PAGE_SHIFT),
MAP_PREFAULT_MADVISE
);
}
}
vm_map_unlock_read(map);
}
return (0);
}
/*
* vm_map_inherit:
*
* Sets the inheritance of the specified address
* range in the target map. Inheritance
* affects how the map will be shared with
* child maps at the time of vm_map_fork.
*/
int
vm_map_inherit(vm_map_t map, vm_offset_t start, vm_offset_t end,
vm_inherit_t new_inheritance)
{
vm_map_entry_t entry;
vm_map_entry_t temp_entry;
switch (new_inheritance) {
case VM_INHERIT_NONE:
case VM_INHERIT_COPY:
case VM_INHERIT_SHARE:
break;
default:
return (KERN_INVALID_ARGUMENT);
}
vm_map_lock(map);
VM_MAP_RANGE_CHECK(map, start, end);
if (vm_map_lookup_entry(map, start, &temp_entry)) {
entry = temp_entry;
vm_map_clip_start(map, entry, start);
} else
entry = temp_entry->next;
while ((entry != &map->header) && (entry->start < end)) {
vm_map_clip_end(map, entry, end);
entry->inheritance = new_inheritance;
vm_map_simplify_entry(map, entry);
entry = entry->next;
}
vm_map_unlock(map);
return (KERN_SUCCESS);
}
/*
* vm_map_unwire:
*
* Implements both kernel and user unwiring.
*/
int
vm_map_unwire(vm_map_t map, vm_offset_t start, vm_offset_t end,
int flags)
{
vm_map_entry_t entry, first_entry, tmp_entry;
vm_offset_t saved_start;
unsigned int last_timestamp;
int rv;
boolean_t need_wakeup, result, user_unwire;
user_unwire = (flags & VM_MAP_WIRE_USER) ? TRUE : FALSE;
vm_map_lock(map);
VM_MAP_RANGE_CHECK(map, start, end);
if (!vm_map_lookup_entry(map, start, &first_entry)) {
if (flags & VM_MAP_WIRE_HOLESOK)
first_entry = first_entry->next;
else {
vm_map_unlock(map);
return (KERN_INVALID_ADDRESS);
}
}
last_timestamp = map->timestamp;
entry = first_entry;
while (entry != &map->header && entry->start < end) {
if (entry->eflags & MAP_ENTRY_IN_TRANSITION) {
/*
* We have not yet clipped the entry.
*/
saved_start = (start >= entry->start) ? start :
entry->start;
entry->eflags |= MAP_ENTRY_NEEDS_WAKEUP;
if (vm_map_unlock_and_wait(map, user_unwire)) {
/*
* Allow interruption of user unwiring?
*/
}
vm_map_lock(map);
if (last_timestamp+1 != map->timestamp) {
/*
* Look again for the entry because the map was
* modified while it was unlocked.
* Specifically, the entry may have been
* clipped, merged, or deleted.
*/
if (!vm_map_lookup_entry(map, saved_start,
&tmp_entry)) {
if (flags & VM_MAP_WIRE_HOLESOK)
tmp_entry = tmp_entry->next;
else {
if (saved_start == start) {
/*
* First_entry has been deleted.
*/
vm_map_unlock(map);
return (KERN_INVALID_ADDRESS);
}
end = saved_start;
rv = KERN_INVALID_ADDRESS;
goto done;
}
}
if (entry == first_entry)
first_entry = tmp_entry;
else
first_entry = NULL;
entry = tmp_entry;
}
last_timestamp = map->timestamp;
continue;
}
vm_map_clip_start(map, entry, start);
vm_map_clip_end(map, entry, end);
/*
* Mark the entry in case the map lock is released. (See
* above.)
*/
entry->eflags |= MAP_ENTRY_IN_TRANSITION;
/*
* Check the map for holes in the specified region.
* If VM_MAP_WIRE_HOLESOK was specified, skip this check.
*/
if (((flags & VM_MAP_WIRE_HOLESOK) == 0) &&
(entry->end < end && (entry->next == &map->header ||
entry->next->start > entry->end))) {
end = entry->end;
rv = KERN_INVALID_ADDRESS;
goto done;
}
/*
* If system unwiring, require that the entry is system wired.
*/
if (!user_unwire && entry->wired_count < ((entry->eflags &
MAP_ENTRY_USER_WIRED) ? 2 : 1)) {
end = entry->end;
rv = KERN_INVALID_ARGUMENT;
goto done;
}
entry = entry->next;
}
rv = KERN_SUCCESS;
done:
need_wakeup = FALSE;
if (first_entry == NULL) {
result = vm_map_lookup_entry(map, start, &first_entry);
if (!result && (flags & VM_MAP_WIRE_HOLESOK))
first_entry = first_entry->next;
else
KASSERT(result, ("vm_map_unwire: lookup failed"));
}
entry = first_entry;
while (entry != &map->header && entry->start < end) {
if (rv == KERN_SUCCESS && (!user_unwire ||
(entry->eflags & MAP_ENTRY_USER_WIRED))) {
if (user_unwire)
entry->eflags &= ~MAP_ENTRY_USER_WIRED;
entry->wired_count--;
if (entry->wired_count == 0) {
/*
* Retain the map lock.
*/
vm_fault_unwire(map, entry->start, entry->end,
entry->object.vm_object != NULL &&
entry->object.vm_object->type == OBJT_DEVICE);
}
}
KASSERT(entry->eflags & MAP_ENTRY_IN_TRANSITION,
("vm_map_unwire: in-transition flag missing"));
entry->eflags &= ~MAP_ENTRY_IN_TRANSITION;
if (entry->eflags & MAP_ENTRY_NEEDS_WAKEUP) {
entry->eflags &= ~MAP_ENTRY_NEEDS_WAKEUP;
need_wakeup = TRUE;
}
vm_map_simplify_entry(map, entry);
entry = entry->next;
}
vm_map_unlock(map);
if (need_wakeup)
vm_map_wakeup(map);
return (rv);
}
/*
* vm_map_wire:
*
* Implements both kernel and user wiring.
*/
int
vm_map_wire(vm_map_t map, vm_offset_t start, vm_offset_t end,
int flags)
{
vm_map_entry_t entry, first_entry, tmp_entry;
vm_offset_t saved_end, saved_start;
unsigned int last_timestamp;
int rv;
boolean_t fictitious, need_wakeup, result, user_wire;
user_wire = (flags & VM_MAP_WIRE_USER) ? TRUE : FALSE;
vm_map_lock(map);
VM_MAP_RANGE_CHECK(map, start, end);
if (!vm_map_lookup_entry(map, start, &first_entry)) {
if (flags & VM_MAP_WIRE_HOLESOK)
first_entry = first_entry->next;
else {
vm_map_unlock(map);
return (KERN_INVALID_ADDRESS);
}
}
last_timestamp = map->timestamp;
entry = first_entry;
while (entry != &map->header && entry->start < end) {
if (entry->eflags & MAP_ENTRY_IN_TRANSITION) {
/*
* We have not yet clipped the entry.
*/
saved_start = (start >= entry->start) ? start :
entry->start;
entry->eflags |= MAP_ENTRY_NEEDS_WAKEUP;
if (vm_map_unlock_and_wait(map, user_wire)) {
/*
* Allow interruption of user wiring?
*/
}
vm_map_lock(map);
if (last_timestamp + 1 != map->timestamp) {
/*
* Look again for the entry because the map was
* modified while it was unlocked.
* Specifically, the entry may have been
* clipped, merged, or deleted.
*/
if (!vm_map_lookup_entry(map, saved_start,
&tmp_entry)) {
if (flags & VM_MAP_WIRE_HOLESOK)
tmp_entry = tmp_entry->next;
else {
if (saved_start == start) {
/*
* first_entry has been deleted.
*/
vm_map_unlock(map);
return (KERN_INVALID_ADDRESS);
}
end = saved_start;
rv = KERN_INVALID_ADDRESS;
goto done;
}
}
if (entry == first_entry)
first_entry = tmp_entry;
else
first_entry = NULL;
entry = tmp_entry;
}
last_timestamp = map->timestamp;
continue;
}
vm_map_clip_start(map, entry, start);
vm_map_clip_end(map, entry, end);
/*
* Mark the entry in case the map lock is released. (See
* above.)
*/
entry->eflags |= MAP_ENTRY_IN_TRANSITION;
/*
*
*/
if (entry->wired_count == 0) {
entry->wired_count++;
saved_start = entry->start;
saved_end = entry->end;
fictitious = entry->object.vm_object != NULL &&
entry->object.vm_object->type == OBJT_DEVICE;
/*
* Release the map lock, relying on the in-transition
* mark.
*/
vm_map_unlock(map);
rv = vm_fault_wire(map, saved_start, saved_end,
user_wire, fictitious);
vm_map_lock(map);
if (last_timestamp + 1 != map->timestamp) {
/*
* Look again for the entry because the map was
* modified while it was unlocked. The entry
* may have been clipped, but NOT merged or
* deleted.
*/
result = vm_map_lookup_entry(map, saved_start,
&tmp_entry);
KASSERT(result, ("vm_map_wire: lookup failed"));
if (entry == first_entry)
first_entry = tmp_entry;
else
first_entry = NULL;
entry = tmp_entry;
while (entry->end < saved_end) {
if (rv != KERN_SUCCESS) {
KASSERT(entry->wired_count == 1,
("vm_map_wire: bad count"));
entry->wired_count = -1;
}
entry = entry->next;
}
}
last_timestamp = map->timestamp;
if (rv != KERN_SUCCESS) {
KASSERT(entry->wired_count == 1,
("vm_map_wire: bad count"));
/*
* Assign an out-of-range value to represent
* the failure to wire this entry.
*/
entry->wired_count = -1;
end = entry->end;
goto done;
}
} else if (!user_wire ||
(entry->eflags & MAP_ENTRY_USER_WIRED) == 0) {
entry->wired_count++;
}
/*
* Check the map for holes in the specified region.
* If VM_MAP_WIRE_HOLESOK was specified, skip this check.
*/
if (((flags & VM_MAP_WIRE_HOLESOK) == 0) &&
(entry->end < end && (entry->next == &map->header ||
entry->next->start > entry->end))) {
end = entry->end;
rv = KERN_INVALID_ADDRESS;
goto done;
}
entry = entry->next;
}
rv = KERN_SUCCESS;
done:
need_wakeup = FALSE;
if (first_entry == NULL) {
result = vm_map_lookup_entry(map, start, &first_entry);
if (!result && (flags & VM_MAP_WIRE_HOLESOK))
first_entry = first_entry->next;
else
KASSERT(result, ("vm_map_wire: lookup failed"));
}
entry = first_entry;
while (entry != &map->header && entry->start < end) {
if (rv == KERN_SUCCESS) {
if (user_wire)
entry->eflags |= MAP_ENTRY_USER_WIRED;
} else if (entry->wired_count == -1) {
/*
* Wiring failed on this entry. Thus, unwiring is
* unnecessary.
*/
entry->wired_count = 0;
} else {
if (!user_wire ||
(entry->eflags & MAP_ENTRY_USER_WIRED) == 0)
entry->wired_count--;
if (entry->wired_count == 0) {
/*
* Retain the map lock.
*/
vm_fault_unwire(map, entry->start, entry->end,
entry->object.vm_object != NULL &&
entry->object.vm_object->type == OBJT_DEVICE);
}
}
KASSERT(entry->eflags & MAP_ENTRY_IN_TRANSITION,
("vm_map_wire: in-transition flag missing"));
entry->eflags &= ~MAP_ENTRY_IN_TRANSITION;
if (entry->eflags & MAP_ENTRY_NEEDS_WAKEUP) {
entry->eflags &= ~MAP_ENTRY_NEEDS_WAKEUP;
need_wakeup = TRUE;
}
vm_map_simplify_entry(map, entry);
entry = entry->next;
}
vm_map_unlock(map);
if (need_wakeup)
vm_map_wakeup(map);
return (rv);
}
/*
* vm_map_sync
*
* Push any dirty cached pages in the address range to their pager.
* If syncio is TRUE, dirty pages are written synchronously.
* If invalidate is TRUE, any cached pages are freed as well.
*
* If the size of the region from start to end is zero, we are
* supposed to flush all modified pages within the region containing
* start. Unfortunately, a region can be split or coalesced with
* neighboring regions, making it difficult to determine what the
* original region was. Therefore, we approximate this requirement by
* flushing the current region containing start.
*
* Returns an error if any part of the specified range is not mapped.
*/
int
vm_map_sync(
vm_map_t map,
vm_offset_t start,
vm_offset_t end,
boolean_t syncio,
boolean_t invalidate)
{
vm_map_entry_t current;
vm_map_entry_t entry;
vm_size_t size;
vm_object_t object;
vm_ooffset_t offset;
vm_map_lock_read(map);
VM_MAP_RANGE_CHECK(map, start, end);
if (!vm_map_lookup_entry(map, start, &entry)) {
vm_map_unlock_read(map);
return (KERN_INVALID_ADDRESS);
} else if (start == end) {
start = entry->start;
end = entry->end;
}
/*
* Make a first pass to check for user-wired memory and holes.
*/
for (current = entry; current->start < end; current = current->next) {
if (invalidate && (current->eflags & MAP_ENTRY_USER_WIRED)) {
vm_map_unlock_read(map);
return (KERN_INVALID_ARGUMENT);
}
if (end > current->end &&
(current->next == &map->header ||
current->end != current->next->start)) {
vm_map_unlock_read(map);
return (KERN_INVALID_ADDRESS);
}
}
if (invalidate) {
mtx_lock(&Giant);
pmap_remove(map->pmap, start, end);
mtx_unlock(&Giant);
}
/*
* Make a second pass, cleaning/uncaching pages from the indicated
* objects as we go.
*/
for (current = entry; current->start < end; current = current->next) {
offset = current->offset + (start - current->start);
size = (end <= current->end ? end : current->end) - start;
if (current->eflags & MAP_ENTRY_IS_SUB_MAP) {
vm_map_t smap;
vm_map_entry_t tentry;
vm_size_t tsize;
smap = current->object.sub_map;
vm_map_lock_read(smap);
(void) vm_map_lookup_entry(smap, offset, &tentry);
tsize = tentry->end - offset;
if (tsize < size)
size = tsize;
object = tentry->object.vm_object;
offset = tentry->offset + (offset - tentry->start);
vm_map_unlock_read(smap);
} else {
object = current->object.vm_object;
}
vm_object_sync(object, offset, size, syncio, invalidate);
start += size;
}
vm_map_unlock_read(map);
return (KERN_SUCCESS);
}
/*
* vm_map_entry_unwire: [ internal use only ]
*
* Make the region specified by this entry pageable.
*
* The map in question should be locked.
* [This is the reason for this routine's existence.]
*/
static void
vm_map_entry_unwire(vm_map_t map, vm_map_entry_t entry)
{
vm_fault_unwire(map, entry->start, entry->end,
entry->object.vm_object != NULL &&
entry->object.vm_object->type == OBJT_DEVICE);
entry->wired_count = 0;
}
/*
* vm_map_entry_delete: [ internal use only ]
*
* Deallocate the given entry from the target map.
*/
static void
vm_map_entry_delete(vm_map_t map, vm_map_entry_t entry)
{
vm_object_t object;
vm_pindex_t offidxstart, offidxend, count;
vm_map_entry_unlink(map, entry);
map->size -= entry->end - entry->start;
if ((entry->eflags & MAP_ENTRY_IS_SUB_MAP) == 0 &&
(object = entry->object.vm_object) != NULL) {
count = OFF_TO_IDX(entry->end - entry->start);
offidxstart = OFF_TO_IDX(entry->offset);
offidxend = offidxstart + count;
VM_OBJECT_LOCK(object);
if (object->ref_count != 1 &&
((object->flags & (OBJ_NOSPLIT|OBJ_ONEMAPPING)) == OBJ_ONEMAPPING ||
object == kernel_object || object == kmem_object) &&
(object->type == OBJT_DEFAULT || object->type == OBJT_SWAP)) {
vm_object_collapse(object);
vm_object_page_remove(object, offidxstart, offidxend, FALSE);
if (object->type == OBJT_SWAP)
swap_pager_freespace(object, offidxstart, count);
if (offidxend >= object->size &&
offidxstart < object->size)
object->size = offidxstart;
}
VM_OBJECT_UNLOCK(object);
vm_object_deallocate(object);
}
vm_map_entry_dispose(map, entry);
}
/*
* vm_map_delete: [ internal use only ]
*
* Deallocates the given address range from the target
* map.
*/
int
vm_map_delete(vm_map_t map, vm_offset_t start, vm_offset_t end)
{
vm_map_entry_t entry;
vm_map_entry_t first_entry;
/*
* Find the start of the region, and clip it
*/
if (!vm_map_lookup_entry(map, start, &first_entry))
entry = first_entry->next;
else {
entry = first_entry;
vm_map_clip_start(map, entry, start);
}
/*
* Save the free space hint
*/
if (entry == &map->header) {
map->first_free = &map->header;
} else if (map->first_free->start >= start) {
map->first_free = entry->prev;
}
/*
* Step through all entries in this region
*/
while ((entry != &map->header) && (entry->start < end)) {
vm_map_entry_t next;
/*
* Wait for wiring or unwiring of an entry to complete.
*/
if ((entry->eflags & MAP_ENTRY_IN_TRANSITION) != 0) {
unsigned int last_timestamp;
vm_offset_t saved_start;
vm_map_entry_t tmp_entry;
saved_start = entry->start;
entry->eflags |= MAP_ENTRY_NEEDS_WAKEUP;
last_timestamp = map->timestamp;
(void) vm_map_unlock_and_wait(map, FALSE);
vm_map_lock(map);
if (last_timestamp + 1 != map->timestamp) {
/*
* Look again for the entry because the map was
* modified while it was unlocked.
* Specifically, the entry may have been
* clipped, merged, or deleted.
*/
if (!vm_map_lookup_entry(map, saved_start,
&tmp_entry))
entry = tmp_entry->next;
else {
entry = tmp_entry;
vm_map_clip_start(map, entry,
saved_start);
}
}
continue;
}
vm_map_clip_end(map, entry, end);
next = entry->next;
/*
* Unwire before removing addresses from the pmap; otherwise,
* unwiring will put the entries back in the pmap.
*/
if (entry->wired_count != 0) {
vm_map_entry_unwire(map, entry);
}
if (!map->system_map)
mtx_lock(&Giant);
pmap_remove(map->pmap, entry->start, entry->end);
if (!map->system_map)
mtx_unlock(&Giant);
/*
* Delete the entry (which may delete the object) only after
* removing all pmap entries pointing to its pages.
* (Otherwise, its page frames may be reallocated, and any
* modify bits will be set in the wrong object!)
*/
vm_map_entry_delete(map, entry);
entry = next;
}
return (KERN_SUCCESS);
}
/*
* vm_map_remove:
*
* Remove the given address range from the target map.
* This is the exported form of vm_map_delete.
*/
int
vm_map_remove(vm_map_t map, vm_offset_t start, vm_offset_t end)
{
int result, s = 0;
if (map == kmem_map)
s = splvm();
vm_map_lock(map);
VM_MAP_RANGE_CHECK(map, start, end);
result = vm_map_delete(map, start, end);
vm_map_unlock(map);
if (map == kmem_map)
splx(s);
return (result);
}
/*
* vm_map_check_protection:
*
* Assert that the target map allows the specified privilege on the
* entire address region given. The entire region must be allocated.
*
* WARNING! This code does not and should not check whether the
* contents of the region is accessible. For example a smaller file
* might be mapped into a larger address space.
*
* NOTE! This code is also called by munmap().
*
* The map must be locked. A read lock is sufficient.
*/
boolean_t
vm_map_check_protection(vm_map_t map, vm_offset_t start, vm_offset_t end,
vm_prot_t protection)
{
vm_map_entry_t entry;
vm_map_entry_t tmp_entry;
if (!vm_map_lookup_entry(map, start, &tmp_entry))
return (FALSE);
entry = tmp_entry;
while (start < end) {
if (entry == &map->header)
return (FALSE);
/*
* No holes allowed!
*/
if (start < entry->start)
return (FALSE);
/*
* Check protection associated with entry.
*/
if ((entry->protection & protection) != protection)
return (FALSE);
/* go to next entry */
start = entry->end;
entry = entry->next;
}
return (TRUE);
}
/*
* vm_map_copy_entry:
*
* Copies the contents of the source entry to the destination
* entry. The entries *must* be aligned properly.
*/
static void
vm_map_copy_entry(
vm_map_t src_map,
vm_map_t dst_map,
vm_map_entry_t src_entry,
vm_map_entry_t dst_entry)
{
vm_object_t src_object;
if ((dst_entry->eflags|src_entry->eflags) & MAP_ENTRY_IS_SUB_MAP)
return;
if (src_entry->wired_count == 0) {
/*
* If the source entry is marked needs_copy, it is already
* write-protected.
*/
if ((src_entry->eflags & MAP_ENTRY_NEEDS_COPY) == 0) {
pmap_protect(src_map->pmap,
src_entry->start,
src_entry->end,
src_entry->protection & ~VM_PROT_WRITE);
}
/*
* Make a copy of the object.
*/
if ((src_object = src_entry->object.vm_object) != NULL) {
VM_OBJECT_LOCK(src_object);
if ((src_object->handle == NULL) &&
(src_object->type == OBJT_DEFAULT ||
src_object->type == OBJT_SWAP)) {
vm_object_collapse(src_object);
if ((src_object->flags & (OBJ_NOSPLIT|OBJ_ONEMAPPING)) == OBJ_ONEMAPPING) {
vm_object_split(src_entry);
src_object = src_entry->object.vm_object;
}
}
vm_object_reference_locked(src_object);
vm_object_clear_flag(src_object, OBJ_ONEMAPPING);
VM_OBJECT_UNLOCK(src_object);
dst_entry->object.vm_object = src_object;
src_entry->eflags |= (MAP_ENTRY_COW|MAP_ENTRY_NEEDS_COPY);
dst_entry->eflags |= (MAP_ENTRY_COW|MAP_ENTRY_NEEDS_COPY);
dst_entry->offset = src_entry->offset;
} else {
dst_entry->object.vm_object = NULL;
dst_entry->offset = 0;
}
pmap_copy(dst_map->pmap, src_map->pmap, dst_entry->start,
dst_entry->end - dst_entry->start, src_entry->start);
} else {
/*
* Of course, wired down pages can't be set copy-on-write.
* Cause wired pages to be copied into the new map by
* simulating faults (the new pages are pageable)
*/
vm_fault_copy_entry(dst_map, src_map, dst_entry, src_entry);
}
}
/*
* vmspace_map_entry_forked:
* Update the newly-forked vmspace each time a map entry is inherited
* or copied. The values for vm_dsize and vm_tsize are approximate
* (and mostly-obsolete ideas in the face of mmap(2) et al.)
*/
static void
vmspace_map_entry_forked(const struct vmspace *vm1, struct vmspace *vm2,
vm_map_entry_t entry)
{
vm_size_t entrysize;
vm_offset_t newend;
entrysize = entry->end - entry->start;
vm2->vm_map.size += entrysize;
if (entry->eflags & (MAP_ENTRY_GROWS_DOWN | MAP_ENTRY_GROWS_UP)) {
vm2->vm_ssize += btoc(entrysize);
} else if (entry->start >= (vm_offset_t)vm1->vm_daddr &&
entry->start < (vm_offset_t)vm1->vm_daddr + ctob(vm1->vm_dsize)) {
newend = MIN(entry->end,
(vm_offset_t)vm1->vm_daddr + ctob(vm1->vm_dsize));
vm2->vm_dsize += btoc(newend - entry->start);
} else if (entry->start >= (vm_offset_t)vm1->vm_taddr &&
entry->start < (vm_offset_t)vm1->vm_taddr + ctob(vm1->vm_tsize)) {
newend = MIN(entry->end,
(vm_offset_t)vm1->vm_taddr + ctob(vm1->vm_tsize));
vm2->vm_tsize += btoc(newend - entry->start);
}
}
/*
* vmspace_fork:
* Create a new process vmspace structure and vm_map
* based on those of an existing process. The new map
* is based on the old map, according to the inheritance
* values on the regions in that map.
*
* XXX It might be worth coalescing the entries added to the new vmspace.
*
* The source map must not be locked.
*/
struct vmspace *
vmspace_fork(struct vmspace *vm1)
{
struct vmspace *vm2;
vm_map_t old_map = &vm1->vm_map;
vm_map_t new_map;
vm_map_entry_t old_entry;
vm_map_entry_t new_entry;
vm_object_t object;
GIANT_REQUIRED;
vm_map_lock(old_map);
old_map->infork = 1;
vm2 = vmspace_alloc(old_map->min_offset, old_map->max_offset);
vm2->vm_taddr = vm1->vm_taddr;
vm2->vm_daddr = vm1->vm_daddr;
vm2->vm_maxsaddr = vm1->vm_maxsaddr;
new_map = &vm2->vm_map; /* XXX */
new_map->timestamp = 1;
/* Do not inherit the MAP_WIREFUTURE property. */
if ((new_map->flags & MAP_WIREFUTURE) == MAP_WIREFUTURE)
new_map->flags &= ~MAP_WIREFUTURE;
old_entry = old_map->header.next;
while (old_entry != &old_map->header) {
if (old_entry->eflags & MAP_ENTRY_IS_SUB_MAP)
panic("vm_map_fork: encountered a submap");
switch (old_entry->inheritance) {
case VM_INHERIT_NONE:
break;
case VM_INHERIT_SHARE:
/*
* Clone the entry, creating the shared object if necessary.
*/
object = old_entry->object.vm_object;
if (object == NULL) {
object = vm_object_allocate(OBJT_DEFAULT,
atop(old_entry->end - old_entry->start));
old_entry->object.vm_object = object;
old_entry->offset = (vm_offset_t) 0;
}
/*
* Add the reference before calling vm_object_shadow
* to insure that a shadow object is created.
*/
vm_object_reference(object);
if (old_entry->eflags & MAP_ENTRY_NEEDS_COPY) {
vm_object_shadow(&old_entry->object.vm_object,
&old_entry->offset,
atop(old_entry->end - old_entry->start));
old_entry->eflags &= ~MAP_ENTRY_NEEDS_COPY;
/* Transfer the second reference too. */
vm_object_reference(
old_entry->object.vm_object);
vm_object_deallocate(object);
object = old_entry->object.vm_object;
}
VM_OBJECT_LOCK(object);
vm_object_clear_flag(object, OBJ_ONEMAPPING);
VM_OBJECT_UNLOCK(object);
/*
* Clone the entry, referencing the shared object.
*/
new_entry = vm_map_entry_create(new_map);
*new_entry = *old_entry;
new_entry->eflags &= ~MAP_ENTRY_USER_WIRED;
new_entry->wired_count = 0;
/*
* Insert the entry into the new map -- we know we're
* inserting at the end of the new map.
*/
vm_map_entry_link(new_map, new_map->header.prev,
new_entry);
vmspace_map_entry_forked(vm1, vm2, new_entry);
/*
* Update the physical map
*/
pmap_copy(new_map->pmap, old_map->pmap,
new_entry->start,
(old_entry->end - old_entry->start),
old_entry->start);
break;
case VM_INHERIT_COPY:
/*
* Clone the entry and link into the map.
*/
new_entry = vm_map_entry_create(new_map);
*new_entry = *old_entry;
new_entry->eflags &= ~MAP_ENTRY_USER_WIRED;
new_entry->wired_count = 0;
new_entry->object.vm_object = NULL;
vm_map_entry_link(new_map, new_map->header.prev,
new_entry);
vmspace_map_entry_forked(vm1, vm2, new_entry);
vm_map_copy_entry(old_map, new_map, old_entry,
new_entry);
break;
}
old_entry = old_entry->next;
}
old_map->infork = 0;
vm_map_unlock(old_map);
return (vm2);
}
int
vm_map_stack(vm_map_t map, vm_offset_t addrbos, vm_size_t max_ssize,
vm_prot_t prot, vm_prot_t max, int cow)
{
vm_map_entry_t new_entry, prev_entry;
vm_offset_t bot, top;
vm_size_t init_ssize;
int orient, rv;
rlim_t vmemlim;
/*
* The stack orientation is piggybacked with the cow argument.
* Extract it into orient and mask the cow argument so that we
* don't pass it around further.
* NOTE: We explicitly allow bi-directional stacks.
*/
orient = cow & (MAP_STACK_GROWS_DOWN|MAP_STACK_GROWS_UP);
cow &= ~orient;
KASSERT(orient != 0, ("No stack grow direction"));
if (addrbos < vm_map_min(map) || addrbos > map->max_offset)
return (KERN_NO_SPACE);
init_ssize = (max_ssize < sgrowsiz) ? max_ssize : sgrowsiz;
PROC_LOCK(curthread->td_proc);
vmemlim = lim_cur(curthread->td_proc, RLIMIT_VMEM);
PROC_UNLOCK(curthread->td_proc);
vm_map_lock(map);
/* If addr is already mapped, no go */
if (vm_map_lookup_entry(map, addrbos, &prev_entry)) {
vm_map_unlock(map);
return (KERN_NO_SPACE);
}
/* If we would blow our VMEM resource limit, no go */
if (map->size + init_ssize > vmemlim) {
vm_map_unlock(map);
return (KERN_NO_SPACE);
}
/*
* If we can't accomodate max_ssize in the current mapping, no go.
* However, we need to be aware that subsequent user mappings might
* map into the space we have reserved for stack, and currently this
* space is not protected.
*
* Hopefully we will at least detect this condition when we try to
* grow the stack.
*/
if ((prev_entry->next != &map->header) &&
(prev_entry->next->start < addrbos + max_ssize)) {
vm_map_unlock(map);
return (KERN_NO_SPACE);
}
/*
* We initially map a stack of only init_ssize. We will grow as
* needed later. Depending on the orientation of the stack (i.e.
* the grow direction) we either map at the top of the range, the
* bottom of the range or in the middle.
*
* Note: we would normally expect prot and max to be VM_PROT_ALL,
* and cow to be 0. Possibly we should eliminate these as input
* parameters, and just pass these values here in the insert call.
*/
if (orient == MAP_STACK_GROWS_DOWN)
bot = addrbos + max_ssize - init_ssize;
else if (orient == MAP_STACK_GROWS_UP)
bot = addrbos;
else
bot = round_page(addrbos + max_ssize/2 - init_ssize/2);
top = bot + init_ssize;
rv = vm_map_insert(map, NULL, 0, bot, top, prot, max, cow);
/* Now set the avail_ssize amount. */
if (rv == KERN_SUCCESS) {
if (prev_entry != &map->header)
vm_map_clip_end(map, prev_entry, bot);
new_entry = prev_entry->next;
if (new_entry->end != top || new_entry->start != bot)
panic("Bad entry start/end for new stack entry");
new_entry->avail_ssize = max_ssize - init_ssize;
if (orient & MAP_STACK_GROWS_DOWN)
new_entry->eflags |= MAP_ENTRY_GROWS_DOWN;
if (orient & MAP_STACK_GROWS_UP)
new_entry->eflags |= MAP_ENTRY_GROWS_UP;
}
vm_map_unlock(map);
return (rv);
}
/* Attempts to grow a vm stack entry. Returns KERN_SUCCESS if the
* desired address is already mapped, or if we successfully grow
* the stack. Also returns KERN_SUCCESS if addr is outside the
* stack range (this is strange, but preserves compatibility with
* the grow function in vm_machdep.c).
*/
int
vm_map_growstack(struct proc *p, vm_offset_t addr)
{
vm_map_entry_t next_entry, prev_entry;
vm_map_entry_t new_entry, stack_entry;
struct vmspace *vm = p->p_vmspace;
vm_map_t map = &vm->vm_map;
vm_offset_t end;
size_t grow_amount, max_grow;
rlim_t stacklim, vmemlim;
int is_procstack, rv;
Retry:
PROC_LOCK(p);
stacklim = lim_cur(p, RLIMIT_STACK);
vmemlim = lim_cur(p, RLIMIT_VMEM);
PROC_UNLOCK(p);
vm_map_lock_read(map);
/* If addr is already in the entry range, no need to grow.*/
if (vm_map_lookup_entry(map, addr, &prev_entry)) {
vm_map_unlock_read(map);
return (KERN_SUCCESS);
}
next_entry = prev_entry->next;
if (!(prev_entry->eflags & MAP_ENTRY_GROWS_UP)) {
/*
* This entry does not grow upwards. Since the address lies
* beyond this entry, the next entry (if one exists) has to
* be a downward growable entry. The entry list header is
* never a growable entry, so it suffices to check the flags.
*/
if (!(next_entry->eflags & MAP_ENTRY_GROWS_DOWN)) {
vm_map_unlock_read(map);
return (KERN_SUCCESS);
}
stack_entry = next_entry;
} else {
/*
* This entry grows upward. If the next entry does not at
* least grow downwards, this is the entry we need to grow.
* otherwise we have two possible choices and we have to
* select one.
*/
if (next_entry->eflags & MAP_ENTRY_GROWS_DOWN) {
/*
* We have two choices; grow the entry closest to
* the address to minimize the amount of growth.
*/
if (addr - prev_entry->end <= next_entry->start - addr)
stack_entry = prev_entry;
else
stack_entry = next_entry;
} else
stack_entry = prev_entry;
}
if (stack_entry == next_entry) {
KASSERT(stack_entry->eflags & MAP_ENTRY_GROWS_DOWN, ("foo"));
KASSERT(addr < stack_entry->start, ("foo"));
end = (prev_entry != &map->header) ? prev_entry->end :
stack_entry->start - stack_entry->avail_ssize;
grow_amount = roundup(stack_entry->start - addr, PAGE_SIZE);
max_grow = stack_entry->start - end;
} else {
KASSERT(stack_entry->eflags & MAP_ENTRY_GROWS_UP, ("foo"));
KASSERT(addr >= stack_entry->end, ("foo"));
end = (next_entry != &map->header) ? next_entry->start :
stack_entry->end + stack_entry->avail_ssize;
grow_amount = roundup(addr + 1 - stack_entry->end, PAGE_SIZE);
max_grow = end - stack_entry->end;
}
if (grow_amount > stack_entry->avail_ssize) {
vm_map_unlock_read(map);
return (KERN_NO_SPACE);
}
/*
* If there is no longer enough space between the entries nogo, and
* adjust the available space. Note: this should only happen if the
* user has mapped into the stack area after the stack was created,
* and is probably an error.
*
* This also effectively destroys any guard page the user might have
* intended by limiting the stack size.
*/
if (grow_amount > max_grow) {
if (vm_map_lock_upgrade(map))
goto Retry;
stack_entry->avail_ssize = max_grow;
vm_map_unlock(map);
return (KERN_NO_SPACE);
}
is_procstack = (addr >= (vm_offset_t)vm->vm_maxsaddr) ? 1 : 0;
/*
* If this is the main process stack, see if we're over the stack
* limit.
*/
if (is_procstack && (ctob(vm->vm_ssize) + grow_amount > stacklim)) {
vm_map_unlock_read(map);
return (KERN_NO_SPACE);
}
/* Round up the grow amount modulo SGROWSIZ */
grow_amount = roundup (grow_amount, sgrowsiz);
if (grow_amount > stack_entry->avail_ssize)
grow_amount = stack_entry->avail_ssize;
if (is_procstack && (ctob(vm->vm_ssize) + grow_amount > stacklim)) {
grow_amount = stacklim - ctob(vm->vm_ssize);
}
/* If we would blow our VMEM resource limit, no go */
if (map->size + grow_amount > vmemlim) {
vm_map_unlock_read(map);
return (KERN_NO_SPACE);
}
if (vm_map_lock_upgrade(map))
goto Retry;
if (stack_entry == next_entry) {
/*
* Growing downward.
*/
/* Get the preliminary new entry start value */
addr = stack_entry->start - grow_amount;
/*
* If this puts us into the previous entry, cut back our
* growth to the available space. Also, see the note above.
*/
if (addr < end) {
stack_entry->avail_ssize = max_grow;
addr = end;
}
rv = vm_map_insert(map, NULL, 0, addr, stack_entry->start,
p->p_sysent->sv_stackprot, VM_PROT_ALL, 0);
/* Adjust the available stack space by the amount we grew. */
if (rv == KERN_SUCCESS) {
if (prev_entry != &map->header)
vm_map_clip_end(map, prev_entry, addr);
new_entry = prev_entry->next;
KASSERT(new_entry == stack_entry->prev, ("foo"));
KASSERT(new_entry->end == stack_entry->start, ("foo"));
KASSERT(new_entry->start == addr, ("foo"));
grow_amount = new_entry->end - new_entry->start;
new_entry->avail_ssize = stack_entry->avail_ssize -
grow_amount;
stack_entry->eflags &= ~MAP_ENTRY_GROWS_DOWN;
new_entry->eflags |= MAP_ENTRY_GROWS_DOWN;
}
} else {
/*
* Growing upward.
*/
addr = stack_entry->end + grow_amount;
/*
* If this puts us into the next entry, cut back our growth
* to the available space. Also, see the note above.
*/
if (addr > end) {
stack_entry->avail_ssize = end - stack_entry->end;
addr = end;
}
grow_amount = addr - stack_entry->end;
/* Grow the underlying object if applicable. */
if (stack_entry->object.vm_object == NULL ||
vm_object_coalesce(stack_entry->object.vm_object,
OFF_TO_IDX(stack_entry->offset),
(vm_size_t)(stack_entry->end - stack_entry->start),
(vm_size_t)grow_amount)) {
map->size += (addr - stack_entry->end);
/* Update the current entry. */
stack_entry->end = addr;
stack_entry->avail_ssize -= grow_amount;
rv = KERN_SUCCESS;
if (next_entry != &map->header)
vm_map_clip_start(map, next_entry, addr);
} else
rv = KERN_FAILURE;
}
if (rv == KERN_SUCCESS && is_procstack)
vm->vm_ssize += btoc(grow_amount);
vm_map_unlock(map);
/*
* Heed the MAP_WIREFUTURE flag if it was set for this process.
*/
if (rv == KERN_SUCCESS && (map->flags & MAP_WIREFUTURE)) {
vm_map_wire(map,
(stack_entry == next_entry) ? addr : addr - grow_amount,
(stack_entry == next_entry) ? stack_entry->start : addr,
(p->p_flag & P_SYSTEM)
? VM_MAP_WIRE_SYSTEM|VM_MAP_WIRE_NOHOLES
: VM_MAP_WIRE_USER|VM_MAP_WIRE_NOHOLES);
}
return (rv);
}
/*
* Unshare the specified VM space for exec. If other processes are
* mapped to it, then create a new one. The new vmspace is null.
*/
void
vmspace_exec(struct proc *p, vm_offset_t minuser, vm_offset_t maxuser)
{
struct vmspace *oldvmspace = p->p_vmspace;
struct vmspace *newvmspace;
GIANT_REQUIRED;
newvmspace = vmspace_alloc(minuser, maxuser);
newvmspace->vm_swrss = oldvmspace->vm_swrss;
/*
* This code is written like this for prototype purposes. The
* goal is to avoid running down the vmspace here, but let the
* other process's that are still using the vmspace to finally
* run it down. Even though there is little or no chance of blocking
* here, it is a good idea to keep this form for future mods.
*/
p->p_vmspace = newvmspace;
if (p == curthread->td_proc) /* XXXKSE ? */
pmap_activate(curthread);
vmspace_free(oldvmspace);
}
/*
* Unshare the specified VM space for forcing COW. This
* is called by rfork, for the (RFMEM|RFPROC) == 0 case.
*/
void
vmspace_unshare(struct proc *p)
{
struct vmspace *oldvmspace = p->p_vmspace;
struct vmspace *newvmspace;
GIANT_REQUIRED;
if (oldvmspace->vm_refcnt == 1)
return;
newvmspace = vmspace_fork(oldvmspace);
p->p_vmspace = newvmspace;
if (p == curthread->td_proc) /* XXXKSE ? */
pmap_activate(curthread);
vmspace_free(oldvmspace);
}
/*
* vm_map_lookup:
*
* Finds the VM object, offset, and
* protection for a given virtual address in the
* specified map, assuming a page fault of the
* type specified.
*
* Leaves the map in question locked for read; return
* values are guaranteed until a vm_map_lookup_done
* call is performed. Note that the map argument
* is in/out; the returned map must be used in
* the call to vm_map_lookup_done.
*
* A handle (out_entry) is returned for use in
* vm_map_lookup_done, to make that fast.
*
* If a lookup is requested with "write protection"
* specified, the map may be changed to perform virtual
* copying operations, although the data referenced will
* remain the same.
*/
int
vm_map_lookup(vm_map_t *var_map, /* IN/OUT */
vm_offset_t vaddr,
vm_prot_t fault_typea,
vm_map_entry_t *out_entry, /* OUT */
vm_object_t *object, /* OUT */
vm_pindex_t *pindex, /* OUT */
vm_prot_t *out_prot, /* OUT */
boolean_t *wired) /* OUT */
{
vm_map_entry_t entry;
vm_map_t map = *var_map;
vm_prot_t prot;
vm_prot_t fault_type = fault_typea;
RetryLookup:;
/*
* Lookup the faulting address.
*/
vm_map_lock_read(map);
#define RETURN(why) \
{ \
vm_map_unlock_read(map); \
return (why); \
}
/*
* If the map has an interesting hint, try it before calling full
* blown lookup routine.
*/
entry = map->root;
*out_entry = entry;
if (entry == NULL ||
(vaddr < entry->start) || (vaddr >= entry->end)) {
/*
* Entry was either not a valid hint, or the vaddr was not
* contained in the entry, so do a full lookup.
*/
if (!vm_map_lookup_entry(map, vaddr, out_entry))
RETURN(KERN_INVALID_ADDRESS);
entry = *out_entry;
}
/*
* Handle submaps.
*/
if (entry->eflags & MAP_ENTRY_IS_SUB_MAP) {
vm_map_t old_map = map;
*var_map = map = entry->object.sub_map;
vm_map_unlock_read(old_map);
goto RetryLookup;
}
/*
* Check whether this task is allowed to have this page.
* Note the special case for MAP_ENTRY_COW
* pages with an override. This is to implement a forced
* COW for debuggers.
*/
if (fault_type & VM_PROT_OVERRIDE_WRITE)
prot = entry->max_protection;
else
prot = entry->protection;
fault_type &= (VM_PROT_READ|VM_PROT_WRITE|VM_PROT_EXECUTE);
if ((fault_type & prot) != fault_type) {
RETURN(KERN_PROTECTION_FAILURE);
}
if ((entry->eflags & MAP_ENTRY_USER_WIRED) &&
(entry->eflags & MAP_ENTRY_COW) &&
(fault_type & VM_PROT_WRITE) &&
(fault_typea & VM_PROT_OVERRIDE_WRITE) == 0) {
RETURN(KERN_PROTECTION_FAILURE);
}
/*
* If this page is not pageable, we have to get it for all possible
* accesses.
*/
*wired = (entry->wired_count != 0);
if (*wired)
prot = fault_type = entry->protection;
/*
* If the entry was copy-on-write, we either ...
*/
if (entry->eflags & MAP_ENTRY_NEEDS_COPY) {
/*
* If we want to write the page, we may as well handle that
* now since we've got the map locked.
*
* If we don't need to write the page, we just demote the
* permissions allowed.
*/
if (fault_type & VM_PROT_WRITE) {
/*
* Make a new object, and place it in the object
* chain. Note that no new references have appeared
* -- one just moved from the map to the new
* object.
*/
if (vm_map_lock_upgrade(map))
goto RetryLookup;
vm_object_shadow(
&entry->object.vm_object,
&entry->offset,
atop(entry->end - entry->start));
entry->eflags &= ~MAP_ENTRY_NEEDS_COPY;
vm_map_lock_downgrade(map);
} else {
/*
* We're attempting to read a copy-on-write page --
* don't allow writes.
*/
prot &= ~VM_PROT_WRITE;
}
}
/*
* Create an object if necessary.
*/
if (entry->object.vm_object == NULL &&
!map->system_map) {
if (vm_map_lock_upgrade(map))
goto RetryLookup;
entry->object.vm_object = vm_object_allocate(OBJT_DEFAULT,
atop(entry->end - entry->start));
entry->offset = 0;
vm_map_lock_downgrade(map);
}
/*
* Return the object/offset from this entry. If the entry was
* copy-on-write or empty, it has been fixed up.
*/
*pindex = OFF_TO_IDX((vaddr - entry->start) + entry->offset);
*object = entry->object.vm_object;
/*
* Return whether this is the only map sharing this data.
*/
*out_prot = prot;
return (KERN_SUCCESS);
#undef RETURN
}
/*
* vm_map_lookup_done:
*
* Releases locks acquired by a vm_map_lookup
* (according to the handle returned by that lookup).
*/
void
vm_map_lookup_done(vm_map_t map, vm_map_entry_t entry)
{
/*
* Unlock the main-level map
*/
vm_map_unlock_read(map);
}
#include "opt_ddb.h"
#ifdef DDB
#include <sys/kernel.h>
#include <ddb/ddb.h>
/*
* vm_map_print: [ debug ]
*/
DB_SHOW_COMMAND(map, vm_map_print)
{
static int nlines;
/* XXX convert args. */
vm_map_t map = (vm_map_t)addr;
boolean_t full = have_addr;
vm_map_entry_t entry;
db_iprintf("Task map %p: pmap=%p, nentries=%d, version=%u\n",
(void *)map,
(void *)map->pmap, map->nentries, map->timestamp);
nlines++;
if (!full && db_indent)
return;
db_indent += 2;
for (entry = map->header.next; entry != &map->header;
entry = entry->next) {
db_iprintf("map entry %p: start=%p, end=%p\n",
(void *)entry, (void *)entry->start, (void *)entry->end);
nlines++;
{
static char *inheritance_name[4] =
{"share", "copy", "none", "donate_copy"};
db_iprintf(" prot=%x/%x/%s",
entry->protection,
entry->max_protection,
inheritance_name[(int)(unsigned char)entry->inheritance]);
if (entry->wired_count != 0)
db_printf(", wired");
}
if (entry->eflags & MAP_ENTRY_IS_SUB_MAP) {
db_printf(", share=%p, offset=0x%jx\n",
(void *)entry->object.sub_map,
(uintmax_t)entry->offset);
nlines++;
if ((entry->prev == &map->header) ||
(entry->prev->object.sub_map !=
entry->object.sub_map)) {
db_indent += 2;
vm_map_print((db_expr_t)(intptr_t)
entry->object.sub_map,
full, 0, (char *)0);
db_indent -= 2;
}
} else {
db_printf(", object=%p, offset=0x%jx",
(void *)entry->object.vm_object,
(uintmax_t)entry->offset);
if (entry->eflags & MAP_ENTRY_COW)
db_printf(", copy (%s)",
(entry->eflags & MAP_ENTRY_NEEDS_COPY) ? "needed" : "done");
db_printf("\n");
nlines++;
if ((entry->prev == &map->header) ||
(entry->prev->object.vm_object !=
entry->object.vm_object)) {
db_indent += 2;
vm_object_print((db_expr_t)(intptr_t)
entry->object.vm_object,
full, 0, (char *)0);
nlines += 4;
db_indent -= 2;
}
}
}
db_indent -= 2;
if (db_indent == 0)
nlines = 0;
}
DB_SHOW_COMMAND(procvm, procvm)
{
struct proc *p;
if (have_addr) {
p = (struct proc *) addr;
} else {
p = curproc;
}
db_printf("p = %p, vmspace = %p, map = %p, pmap = %p\n",
(void *)p, (void *)p->p_vmspace, (void *)&p->p_vmspace->vm_map,
(void *)vmspace_pmap(p->p_vmspace));
vm_map_print((db_expr_t)(intptr_t)&p->p_vmspace->vm_map, 1, 0, NULL);
}
#endif /* DDB */
|