Loading...
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 | /* * Copyright (c) 2000-2021 Apple Inc. All rights reserved. * * @APPLE_OSREFERENCE_LICENSE_HEADER_START@ * * This file contains Original Code and/or Modifications of Original Code * as defined in and that are subject to the Apple Public Source License * Version 2.0 (the 'License'). You may not use this file except in * compliance with the License. The rights granted to you under the License * may not be used to create, or enable the creation or redistribution of, * unlawful or unlicensed copies of an Apple operating system, or to * circumvent, violate, or enable the circumvention or violation of, any * terms of an Apple operating system software license agreement. * * Please obtain a copy of the License at * http://www.opensource.apple.com/apsl/ and read it before using this file. * * The Original Code and all software distributed under the License are * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT. * Please see the License for the specific language governing rights and * limitations under the License. * * @APPLE_OSREFERENCE_LICENSE_HEADER_END@ */ /* * Mach Operating System * Copyright (c) 1987 Carnegie-Mellon University * All rights reserved. The CMU software License Agreement specifies * the terms and conditions for use and redistribution. */ /* * NOTICE: This file was modified by SPARTA, Inc. in 2006 to introduce * support for mandatory and extensible security protections. This notice * is included in support of clause 2.2 (b) of the Apple Public License, * Version 2.0. */ #include <vm/vm_options.h> #include <kern/ecc.h> #include <kern/task.h> #include <kern/thread.h> #include <kern/debug.h> #include <kern/extmod_statistics.h> #include <mach/mach_traps.h> #include <mach/port.h> #include <mach/sdt.h> #include <mach/task.h> #include <mach/task_access.h> #include <mach/task_special_ports.h> #include <mach/time_value.h> #include <mach/vm_map.h> #include <mach/vm_param.h> #include <mach/vm_prot.h> #include <machine/machine_routines.h> #include <sys/file_internal.h> #include <sys/param.h> #include <sys/systm.h> #include <sys/dir.h> #include <sys/namei.h> #include <sys/proc_internal.h> #include <sys/kauth.h> #include <sys/vm.h> #include <sys/file.h> #include <sys/vnode_internal.h> #include <sys/mount.h> #include <sys/xattr.h> #include <sys/trace.h> #include <sys/kernel.h> #include <sys/ubc_internal.h> #include <sys/user.h> #include <sys/syslog.h> #include <sys/stat.h> #include <sys/sysproto.h> #include <sys/mman.h> #include <sys/sysctl.h> #include <sys/cprotect.h> #include <sys/kpi_socket.h> #include <sys/kas_info.h> #include <sys/socket.h> #include <sys/socketvar.h> #include <sys/random.h> #include <sys/code_signing.h> #if NECP #include <net/necp.h> #endif /* NECP */ #if SKYWALK #include <skywalk/os_channel.h> #endif /* SKYWALK */ #include <security/audit/audit.h> #include <security/mac.h> #include <bsm/audit_kevents.h> #include <kern/kalloc.h> #include <kern/host_statistics.h> #include <vm/vm_map_internal.h> #include <vm/vm_kern_xnu.h> #include <vm/vm_pageout_xnu.h> #include <mach/shared_region.h> #include <vm/vm_shared_region_internal.h> #include <vm/vm_dyld_pager_internal.h> #include <vm/vm_protos_internal.h> #include <vm/vm_compressor_info.h> /* for c_segment_info */ #include <vm/vm_compressor_internal.h> #include <vm/vm_compressor_xnu.h> /* for vm_compressor_serialize_segment_debug_info() */ #include <vm/vm_object_xnu.h> /* for vm_chead_select_t */ #include <vm/vm_memory_entry_xnu.h> #include <vm/vm_iokit.h> #include <vm/vm_reclaim_xnu.h> #if HAS_MTE #include <vm/vm_compressor_xnu.h> #include <vm/vm_mteinfo_internal.h> #endif /* HAS_MTE */ #include <sys/kern_memorystatus.h> #include <sys/kern_memorystatus_freeze.h> #include <sys/proc_internal.h> #include <mach-o/fixup-chains.h> #if CONFIG_MACF #include <security/mac_framework.h> #endif #include <kern/bits.h> #if CONFIG_CSR #include <sys/csr.h> #endif /* CONFIG_CSR */ #include <sys/trust_caches.h> #include <libkern/amfi/amfi.h> #include <IOKit/IOBSD.h> #if VM_MAP_DEBUG_APPLE_PROTECT SYSCTL_INT(_vm, OID_AUTO, map_debug_apple_protect, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_map_debug_apple_protect, 0, ""); #endif /* VM_MAP_DEBUG_APPLE_PROTECT */ #if DEVELOPMENT || DEBUG extern int vm_object_cache_evict_all(void); static int sysctl_vm_object_cache_evict SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, req) (void) vm_object_cache_evict_all(); return 0; } SYSCTL_PROC(_vm, OID_AUTO, object_cache_evict, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_object_cache_evict, "I", ""); static int sysctl_kmem_alloc_contig SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2) vm_offset_t kaddr; kern_return_t kr; int error = 0; int size = 0; error = sysctl_handle_int(oidp, &size, 0, req); if (error || !req->newptr) { return error; } kr = kmem_alloc_contig(kernel_map, &kaddr, (vm_size_t)size, 0, 0, 0, KMA_DATA, VM_KERN_MEMORY_IOKIT); if (kr == KERN_SUCCESS) { kmem_free(kernel_map, kaddr, size); } return error; } SYSCTL_PROC(_vm, OID_AUTO, kmem_alloc_contig, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_kmem_alloc_contig, "I", ""); extern int vm_region_footprint; SYSCTL_INT(_vm, OID_AUTO, region_footprint, CTLFLAG_RW | CTLFLAG_ANYBODY | CTLFLAG_LOCKED, &vm_region_footprint, 0, ""); static int sysctl_kmem_gobj_stats SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) kmem_gobj_stats stats = kmem_get_gobj_stats(); return SYSCTL_OUT(req, &stats, sizeof(stats)); } SYSCTL_PROC(_vm, OID_AUTO, kmem_gobj_stats, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_kmem_gobj_stats, "S,kmem_gobj_stats", ""); #endif /* DEVELOPMENT || DEBUG */ static int sysctl_vm_self_region_footprint SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) int error = 0; int value; value = task_self_region_footprint(); error = SYSCTL_OUT(req, &value, sizeof(int)); if (error) { return error; } if (!req->newptr) { return 0; } error = SYSCTL_IN(req, &value, sizeof(int)); if (error) { return error; } task_self_region_footprint_set(value); return 0; } SYSCTL_PROC(_vm, OID_AUTO, self_region_footprint, CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_ANYBODY | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_self_region_footprint, "I", ""); static int sysctl_vm_self_region_page_size SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) int error = 0; int value; value = (1 << thread_self_region_page_shift()); error = SYSCTL_OUT(req, &value, sizeof(int)); if (error) { return error; } if (!req->newptr) { return 0; } error = SYSCTL_IN(req, &value, sizeof(int)); if (error) { return error; } if (value != 0 && value != 4096 && value != 16384) { return EINVAL; } #if !__ARM_MIXED_PAGE_SIZE__ if (value != vm_map_page_size(current_map())) { return EINVAL; } #endif /* !__ARM_MIXED_PAGE_SIZE__ */ thread_self_region_page_shift_set(bit_first(value)); return 0; } SYSCTL_PROC(_vm, OID_AUTO, self_region_page_size, CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_ANYBODY | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_self_region_page_size, "I", ""); static int sysctl_vm_self_region_info_flags SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) int error = 0; int value; kern_return_t kr; value = task_self_region_info_flags(); error = SYSCTL_OUT(req, &value, sizeof(int)); if (error) { return error; } if (!req->newptr) { return 0; } error = SYSCTL_IN(req, &value, sizeof(int)); if (error) { return error; } kr = task_self_region_info_flags_set(value); if (kr != KERN_SUCCESS) { return EINVAL; } return 0; } SYSCTL_PROC(_vm, OID_AUTO, self_region_info_flags, CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_ANYBODY | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_self_region_info_flags, "I", ""); #if DEVELOPMENT || DEBUG extern int panic_on_unsigned_execute; SYSCTL_INT(_vm, OID_AUTO, panic_on_unsigned_execute, CTLFLAG_RW | CTLFLAG_LOCKED, &panic_on_unsigned_execute, 0, ""); extern int vm_log_xnu_user_debug; SYSCTL_INT(_vm, OID_AUTO, log_xnu_user_debug, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_log_xnu_user_debug, 0, ""); #endif /* DEVELOPMENT || DEBUG */ extern int vm_log_map_delete_permanent_prot_none; SYSCTL_INT(_vm, OID_AUTO, log_map_delete_permanent_prot_none, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_log_map_delete_permanent_prot_none, 0, ""); extern int cs_executable_create_upl; extern int cs_executable_wire; SYSCTL_INT(_vm, OID_AUTO, cs_executable_create_upl, CTLFLAG_RD | CTLFLAG_LOCKED, &cs_executable_create_upl, 0, ""); SYSCTL_INT(_vm, OID_AUTO, cs_executable_wire, CTLFLAG_RD | CTLFLAG_LOCKED, &cs_executable_wire, 0, ""); extern int apple_protect_pager_count; extern int apple_protect_pager_count_mapped; extern unsigned int apple_protect_pager_cache_limit; SYSCTL_INT(_vm, OID_AUTO, apple_protect_pager_count, CTLFLAG_RD | CTLFLAG_LOCKED, &apple_protect_pager_count, 0, ""); SYSCTL_INT(_vm, OID_AUTO, apple_protect_pager_count_mapped, CTLFLAG_RD | CTLFLAG_LOCKED, &apple_protect_pager_count_mapped, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, apple_protect_pager_cache_limit, CTLFLAG_RW | CTLFLAG_LOCKED, &apple_protect_pager_cache_limit, 0, ""); #if DEVELOPMENT || DEBUG extern int radar_20146450; SYSCTL_INT(_vm, OID_AUTO, radar_20146450, CTLFLAG_RW | CTLFLAG_LOCKED, &radar_20146450, 0, ""); extern int macho_printf; SYSCTL_INT(_vm, OID_AUTO, macho_printf, CTLFLAG_RW | CTLFLAG_LOCKED, &macho_printf, 0, ""); extern int apple_protect_pager_data_request_debug; SYSCTL_INT(_vm, OID_AUTO, apple_protect_pager_data_request_debug, CTLFLAG_RW | CTLFLAG_LOCKED, &apple_protect_pager_data_request_debug, 0, ""); extern unsigned int vm_object_copy_delayed_paging_wait_disable; EXPERIMENT_FACTOR_LEGACY_UINT(_vm, vm_object_copy_delayed_paging_wait_disable, &vm_object_copy_delayed_paging_wait_disable, FALSE, TRUE, ""); __enum_closed_decl(vm_submap_test_op, uint32_t, { vsto_make_submap = 1, /* make submap from entries in current_map() * at start..end, offset ignored */ vsto_remap_submap = 2, /* map in current_map() at start..end, * from parent address submap_base_address * and submap address offset */ vsto_end }); static int sysctl_vm_submap_test_ctl SYSCTL_HANDLER_ARGS { int error; struct { vm_submap_test_op op; mach_vm_address_t submap_base_address; mach_vm_address_t start; mach_vm_address_t end; mach_vm_address_t offset; } args; if (req->newlen != sizeof(args)) { return EINVAL; } error = SYSCTL_IN(req, &args, sizeof(args)); if (error) { return error; } switch (args.op) { case vsto_make_submap: vm_map_testing_make_sealed_submap(current_map(), args.start, args.end); break; case vsto_remap_submap: vm_map_testing_remap_submap(current_map(), args.submap_base_address, args.start, args.end, args.offset); break; default: return EINVAL; } return 0; } SYSCTL_PROC(_vm, OID_AUTO, submap_test_ctl, CTLFLAG_WR | CTLFLAG_LOCKED, 0, 0, &sysctl_vm_submap_test_ctl, "-", ""); #if __arm64__ /* These are meant to support the page table accounting unit test. */ extern unsigned int arm_hardware_page_size; extern unsigned int arm_pt_desc_size; extern unsigned int arm_pt_root_size; extern unsigned int inuse_user_tteroot_count; extern unsigned int inuse_kernel_tteroot_count; extern unsigned int inuse_user_ttepages_count; extern unsigned int inuse_kernel_ttepages_count; extern unsigned int inuse_user_ptepages_count; extern unsigned int inuse_kernel_ptepages_count; SYSCTL_UINT(_vm, OID_AUTO, native_hw_pagesize, CTLFLAG_RD | CTLFLAG_LOCKED, &arm_hardware_page_size, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, arm_pt_desc_size, CTLFLAG_RD | CTLFLAG_LOCKED, &arm_pt_desc_size, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, arm_pt_root_size, CTLFLAG_RD | CTLFLAG_LOCKED, &arm_pt_root_size, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, user_tte_root, CTLFLAG_RD | CTLFLAG_LOCKED, &inuse_user_tteroot_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, kernel_tte_root, CTLFLAG_RD | CTLFLAG_LOCKED, &inuse_kernel_tteroot_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, user_tte_pages, CTLFLAG_RD | CTLFLAG_LOCKED, &inuse_user_ttepages_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, kernel_tte_pages, CTLFLAG_RD | CTLFLAG_LOCKED, &inuse_kernel_ttepages_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, user_pte_pages, CTLFLAG_RD | CTLFLAG_LOCKED, &inuse_user_ptepages_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, kernel_pte_pages, CTLFLAG_RD | CTLFLAG_LOCKED, &inuse_kernel_ptepages_count, 0, ""); #if !CONFIG_SPTM extern unsigned int free_page_size_tt_count; extern unsigned int free_tt_count; SYSCTL_UINT(_vm, OID_AUTO, free_1page_tte_root, CTLFLAG_RD | CTLFLAG_LOCKED, &free_page_size_tt_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, free_tte_root, CTLFLAG_RD | CTLFLAG_LOCKED, &free_tt_count, 0, ""); #endif #if DEVELOPMENT || DEBUG extern unsigned long pmap_asid_flushes; SYSCTL_ULONG(_vm, OID_AUTO, pmap_asid_flushes, CTLFLAG_RD | CTLFLAG_LOCKED, &pmap_asid_flushes, ""); extern unsigned long pmap_asid_hits; SYSCTL_ULONG(_vm, OID_AUTO, pmap_asid_hits, CTLFLAG_RD | CTLFLAG_LOCKED, &pmap_asid_hits, ""); extern unsigned long pmap_asid_misses; SYSCTL_ULONG(_vm, OID_AUTO, pmap_asid_misses, CTLFLAG_RD | CTLFLAG_LOCKED, &pmap_asid_misses, ""); extern unsigned long pmap_speculation_restrictions; SYSCTL_ULONG(_vm, OID_AUTO, pmap_speculation_restrictions, CTLFLAG_RD | CTLFLAG_LOCKED, &pmap_speculation_restrictions, ""); #endif #endif /* __arm64__ */ #endif /* DEVELOPMENT || DEBUG */ SYSCTL_INT(_vm, OID_AUTO, vm_do_collapse_compressor, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.do_collapse_compressor, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_do_collapse_compressor_pages, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.do_collapse_compressor_pages, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_do_collapse_terminate, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.do_collapse_terminate, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_do_collapse_terminate_failure, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.do_collapse_terminate_failure, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_should_cow_but_wired, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.should_cow_but_wired, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_create_upl_extra_cow, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.create_upl_extra_cow, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_create_upl_extra_cow_pages, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.create_upl_extra_cow_pages, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_create_upl_lookup_failure_write, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.create_upl_lookup_failure_write, 0, ""); SYSCTL_INT(_vm, OID_AUTO, vm_create_upl_lookup_failure_copy, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_counters.create_upl_lookup_failure_copy, 0, ""); #if VM_SCAN_FOR_SHADOW_CHAIN static int vm_shadow_max_enabled = 0; /* Disabled by default */ extern int proc_shadow_max(void); static int vm_shadow_max SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) int value = 0; if (vm_shadow_max_enabled) { value = proc_shadow_max(); } return SYSCTL_OUT(req, &value, sizeof(value)); } SYSCTL_PROC(_vm, OID_AUTO, vm_shadow_max, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &vm_shadow_max, "I", ""); SYSCTL_INT(_vm, OID_AUTO, vm_shadow_max_enabled, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_shadow_max_enabled, 0, ""); #endif /* VM_SCAN_FOR_SHADOW_CHAIN */ SYSCTL_INT(_vm, OID_AUTO, vm_debug_events, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_debug_events, 0, ""); #if PAGE_SLEEP_WITH_INHERITOR #if DEVELOPMENT || DEBUG extern uint32_t page_worker_table_size; SYSCTL_INT(_vm, OID_AUTO, page_worker_table_size, CTLFLAG_RD | CTLFLAG_LOCKED, &page_worker_table_size, 0, ""); SCALABLE_COUNTER_DECLARE(page_worker_hash_collisions); SYSCTL_SCALABLE_COUNTER(_vm, page_worker_hash_collisions, page_worker_hash_collisions, ""); SCALABLE_COUNTER_DECLARE(page_worker_inheritor_sleeps); SYSCTL_SCALABLE_COUNTER(_vm, page_worker_inheritor_sleeps, page_worker_inheritor_sleeps, ""); #endif /* DEVELOPMENT || DEBUG */ #endif /* PAGE_SLEEP_WITH_INHERITOR */ #if COMPRESSOR_PAGEOUT_CHEADS_MAX_COUNT > 1 extern uint32_t vm_cheads; extern vm_chead_select_t vm_chead_select; extern boolean_t vm_chead_rehint; #if DEVELOPMENT || DEBUG SYSCTL_UINT(_vm, OID_AUTO, compressor_heads, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_cheads, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, compressor_head_select, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_chead_select, 0, ""); SYSCTL_INT(_vm, OID_AUTO, compressor_head_rehint, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_chead_rehint, 0, ""); #endif /* DEVELOPMENT || DEBUG */ EXPERIMENT_FACTOR_UINT(compressor_heads, &vm_cheads, 1, COMPRESSOR_PAGEOUT_CHEADS_MAX_COUNT, ""); EXPERIMENT_FACTOR_UINT(compressor_head_select, &vm_chead_select, CSEL_MIN, CSEL_MAX, ""); EXPERIMENT_FACTOR_INT(compressor_head_rehint, &vm_chead_rehint, 0, 1, ""); #endif /* COMPRESSOR_PAGEOUT_CHEADS_MAX_COUNT > 1 */ /* * Sysctl's related to data/stack execution. See osfmk/vm/vm_map.c */ #if DEVELOPMENT || DEBUG extern int allow_stack_exec, allow_data_exec; SYSCTL_INT(_vm, OID_AUTO, allow_stack_exec, CTLFLAG_RW | CTLFLAG_LOCKED, &allow_stack_exec, 0, ""); SYSCTL_INT(_vm, OID_AUTO, allow_data_exec, CTLFLAG_RW | CTLFLAG_LOCKED, &allow_data_exec, 0, ""); #endif /* DEVELOPMENT || DEBUG */ static const char *prot_values[] = { "none", "read-only", "write-only", "read-write", "execute-only", "read-execute", "write-execute", "read-write-execute" }; void log_stack_execution_failure(addr64_t vaddr, vm_prot_t prot) { printf("Data/Stack execution not permitted: %s[pid %d] at virtual address 0x%qx, protections were %s\n", current_proc()->p_comm, proc_getpid(current_proc()), vaddr, prot_values[prot & VM_PROT_ALL]); } /* * shared_region_unnest_logging: level of logging of unnesting events * 0 - no logging * 1 - throttled logging of unexpected unnesting events (default) * 2 - unthrottled logging of unexpected unnesting events * 3+ - unthrottled logging of all unnesting events */ int shared_region_unnest_logging = 1; SYSCTL_INT(_vm, OID_AUTO, shared_region_unnest_logging, CTLFLAG_RW | CTLFLAG_LOCKED, &shared_region_unnest_logging, 0, ""); int vm_shared_region_unnest_log_interval = 10; int shared_region_unnest_log_count_threshold = 5; #if XNU_TARGET_OS_OSX #if defined (__x86_64__) static int scdir_enforce = 1; #else /* defined (__x86_64__) */ static int scdir_enforce = 0; /* AOT caches live elsewhere */ #endif /* defined (__x86_64__) */ static char *scdir_path[] = { "/System/Library/dyld/", "/System/Volumes/Preboot/Cryptexes/OS/System/Library/dyld", "/System/Cryptexes/OS/System/Library/dyld", NULL }; #else /* XNU_TARGET_OS_OSX */ static int scdir_enforce = 0; static char *scdir_path[] = { "/System/Library/Caches/com.apple.dyld/", "/private/preboot/Cryptexes/OS/System/Library/Caches/com.apple.dyld", "/System/Cryptexes/OS/System/Library/Caches/com.apple.dyld", NULL }; #endif /* XNU_TARGET_OS_OSX */ static char *driverkit_scdir_path[] = { "/System/DriverKit/System/Library/dyld/", #if XNU_TARGET_OS_OSX "/System/Volumes/Preboot/Cryptexes/OS/System/DriverKit/System/Library/dyld", #else "/private/preboot/Cryptexes/OS/System/DriverKit/System/Library/dyld", #endif /* XNU_TARGET_OS_OSX */ "/System/Cryptexes/OS/System/DriverKit/System/Library/dyld", NULL }; #ifndef SECURE_KERNEL static int sysctl_scdir_enforce SYSCTL_HANDLER_ARGS { #if CONFIG_CSR if (csr_check(CSR_ALLOW_UNRESTRICTED_FS) != 0) { printf("Failed attempt to set vm.enforce_shared_cache_dir sysctl\n"); return EPERM; } #endif /* CONFIG_CSR */ return sysctl_handle_int(oidp, arg1, arg2, req); } SYSCTL_PROC(_vm, OID_AUTO, enforce_shared_cache_dir, CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_LOCKED, &scdir_enforce, 0, sysctl_scdir_enforce, "I", ""); #endif /* These log rate throttling state variables aren't thread safe, but * are sufficient unto the task. */ static int64_t last_unnest_log_time = 0; static int shared_region_unnest_log_count = 0; void log_unnest_badness( vm_map_t m, vm_map_offset_t s, vm_map_offset_t e, boolean_t is_nested_map, vm_map_offset_t lowest_unnestable_addr) { struct timeval tv; if (shared_region_unnest_logging == 0) { return; } if (shared_region_unnest_logging <= 2 && is_nested_map && s >= lowest_unnestable_addr) { /* * Unnesting of writable map entries is fine. */ return; } if (shared_region_unnest_logging <= 1) { microtime(&tv); if ((tv.tv_sec - last_unnest_log_time) < vm_shared_region_unnest_log_interval) { if (shared_region_unnest_log_count++ > shared_region_unnest_log_count_threshold) { return; } } else { last_unnest_log_time = tv.tv_sec; shared_region_unnest_log_count = 0; } } DTRACE_VM4(log_unnest_badness, vm_map_t, m, vm_map_offset_t, s, vm_map_offset_t, e, vm_map_offset_t, lowest_unnestable_addr); printf("%s[%d] triggered unnest of range 0x%qx->0x%qx of DYLD shared region in VM map %p. While not abnormal for debuggers, this increases system memory footprint until the target exits.\n", current_proc()->p_comm, proc_getpid(current_proc()), (uint64_t)s, (uint64_t)e, (void *) VM_KERNEL_ADDRPERM(m)); } uint64_t vm_purge_filebacked_pagers(void) { uint64_t pages_purged; pages_purged = 0; pages_purged += apple_protect_pager_purge_all(); pages_purged += shared_region_pager_purge_all(); pages_purged += dyld_pager_purge_all(); #if DEVELOPMENT || DEBUG printf("%s:%d pages purged: %llu\n", __FUNCTION__, __LINE__, pages_purged); #endif /* DEVELOPMENT || DEBUG */ return pages_purged; } int useracc( user_addr_ut addr_u, user_size_ut len_u, int prot) { vm_map_t map; vm_prot_t vm_prot = VM_PROT_WRITE; map = current_map(); if (prot == B_READ) { vm_prot = VM_PROT_READ; } return vm_map_check_protection(map, addr_u, vm_sanitize_compute_ut_end(addr_u, len_u), vm_prot, VM_SANITIZE_CALLER_USERACC); } #if XNU_PLATFORM_MacOSX static __attribute__((always_inline, warn_unused_result)) kern_return_t vslock_sanitize( vm_map_t map, user_addr_ut addr_u, user_size_ut len_u, vm_sanitize_caller_t vm_sanitize_caller, vm_map_offset_t *start, vm_map_offset_t *end, vm_map_size_t *size) { return vm_sanitize_addr_size(addr_u, len_u, vm_sanitize_caller, map, VM_SANITIZE_FLAGS_SIZE_ZERO_SUCCEEDS, start, end, size); } #endif /* XNU_PLATFORM_MacOSX */ int vslock(user_addr_ut addr, user_size_ut len) { kern_return_t kret; #if XNU_PLATFORM_MacOSX /* * Preserve previous behavior on macOS for overflows due to bin * compatibility i.e. return success for overflows without doing * anything. Error compatibility returns VM_ERR_RETURN_NOW (on macOS) * for overflow errors which gets converted to KERN_SUCCESS by * vm_sanitize_get_kr. */ vm_map_offset_t start, end; vm_map_size_t size; kret = vslock_sanitize(current_map(), addr, len, VM_SANITIZE_CALLER_VSLOCK, &start, &end, &size); if (__improbable(kret != KERN_SUCCESS)) { switch (vm_sanitize_get_kr(kret)) { case KERN_SUCCESS: return 0; case KERN_INVALID_ADDRESS: case KERN_NO_SPACE: return ENOMEM; case KERN_PROTECTION_FAILURE: return EACCES; default: return EINVAL; } } #endif /* XNU_PLATFORM_MacOSX */ kret = vm_map_wire_kernel(current_map(), addr, vm_sanitize_compute_ut_end(addr, len), vm_sanitize_wrap_prot(VM_PROT_READ | VM_PROT_WRITE), VM_KERN_MEMORY_BSD, FALSE); switch (kret) { case KERN_SUCCESS: return 0; case KERN_INVALID_ADDRESS: case KERN_NO_SPACE: return ENOMEM; case KERN_PROTECTION_FAILURE: return EACCES; default: return EINVAL; } } int vsunlock(user_addr_ut addr, user_size_ut len, __unused int dirtied) { #if FIXME /* [ */ pmap_t pmap; vm_page_t pg; vm_map_offset_t vaddr; ppnum_t paddr; #endif /* FIXME ] */ kern_return_t kret; vm_map_t map; map = current_map(); #if FIXME /* [ */ if (dirtied) { pmap = get_task_pmap(current_task()); for (vaddr = vm_map_trunc_page(addr, PAGE_MASK); vaddr < vm_map_round_page(addr + len, PAGE_MASK); vaddr += PAGE_SIZE) { paddr = pmap_find_phys(pmap, vaddr); pg = PHYS_TO_VM_PAGE(paddr); vm_page_set_modified(pg); } } #endif /* FIXME ] */ #ifdef lint dirtied++; #endif /* lint */ #if XNU_PLATFORM_MacOSX /* * Preserve previous behavior on macOS for overflows due to bin * compatibility i.e. return success for overflows without doing * anything. Error compatibility returns VM_ERR_RETURN_NOW (on macOS) * for overflow errors which gets converted to KERN_SUCCESS by * vm_sanitize_get_kr. */ vm_map_offset_t start, end; vm_map_size_t size; kret = vslock_sanitize(map, addr, len, VM_SANITIZE_CALLER_VSUNLOCK, &start, &end, &size); if (__improbable(kret != KERN_SUCCESS)) { switch (vm_sanitize_get_kr(kret)) { case KERN_SUCCESS: return 0; case KERN_INVALID_ADDRESS: case KERN_NO_SPACE: return ENOMEM; case KERN_PROTECTION_FAILURE: return EACCES; default: return EINVAL; } } #endif /* XNU_PLATFORM_MacOSX */ kret = vm_map_unwire(map, addr, vm_sanitize_compute_ut_end(addr, len), false); switch (kret) { case KERN_SUCCESS: return 0; case KERN_INVALID_ADDRESS: case KERN_NO_SPACE: return ENOMEM; case KERN_PROTECTION_FAILURE: return EACCES; default: return EINVAL; } } int subyte( user_addr_t addr, int byte) { char character; character = (char)byte; return copyout((void *)&(character), addr, sizeof(char)) == 0 ? 0 : -1; } int suibyte( user_addr_t addr, int byte) { char character; character = (char)byte; return copyout((void *)&(character), addr, sizeof(char)) == 0 ? 0 : -1; } int fubyte(user_addr_t addr) { unsigned char byte; if (copyin(addr, (void *) &byte, sizeof(char))) { return -1; } return byte; } int fuibyte(user_addr_t addr) { unsigned char byte; if (copyin(addr, (void *) &(byte), sizeof(char))) { return -1; } return byte; } int suword( user_addr_t addr, long word) { return copyout((void *) &word, addr, sizeof(int)) == 0 ? 0 : -1; } long fuword(user_addr_t addr) { long word = 0; if (copyin(addr, (void *) &word, sizeof(int))) { return -1; } return word; } /* suiword and fuiword are the same as suword and fuword, respectively */ int suiword( user_addr_t addr, long word) { return copyout((void *) &word, addr, sizeof(int)) == 0 ? 0 : -1; } long fuiword(user_addr_t addr) { long word = 0; if (copyin(addr, (void *) &word, sizeof(int))) { return -1; } return word; } /* * With a 32-bit kernel and mixed 32/64-bit user tasks, this interface allows the * fetching and setting of process-sized size_t and pointer values. */ int sulong(user_addr_t addr, int64_t word) { if (IS_64BIT_PROCESS(current_proc())) { return copyout((void *)&word, addr, sizeof(word)) == 0 ? 0 : -1; } else { return suiword(addr, (long)word); } } int64_t fulong(user_addr_t addr) { int64_t longword; if (IS_64BIT_PROCESS(current_proc())) { if (copyin(addr, (void *)&longword, sizeof(longword)) != 0) { return -1; } return longword; } else { return (int64_t)fuiword(addr); } } int suulong(user_addr_t addr, uint64_t uword) { if (IS_64BIT_PROCESS(current_proc())) { return copyout((void *)&uword, addr, sizeof(uword)) == 0 ? 0 : -1; } else { return suiword(addr, (uint32_t)uword); } } uint64_t fuulong(user_addr_t addr) { uint64_t ulongword; if (IS_64BIT_PROCESS(current_proc())) { if (copyin(addr, (void *)&ulongword, sizeof(ulongword)) != 0) { return -1ULL; } return ulongword; } else { return (uint64_t)fuiword(addr); } } int swapon(__unused proc_t procp, __unused struct swapon_args *uap, __unused int *retval) { return ENOTSUP; } #if defined(SECURE_KERNEL) static int kern_secure_kernel = 1; #else static int kern_secure_kernel = 0; #endif SYSCTL_INT(_kern, OID_AUTO, secure_kernel, CTLFLAG_RD | CTLFLAG_LOCKED, &kern_secure_kernel, 0, ""); SYSCTL_INT(_vm, OID_AUTO, shared_region_trace_level, CTLFLAG_RW | CTLFLAG_LOCKED, &shared_region_trace_level, 0, ""); SYSCTL_INT(_vm, OID_AUTO, shared_region_version, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_version, 0, ""); SYSCTL_INT(_vm, OID_AUTO, shared_region_persistence, CTLFLAG_RW | CTLFLAG_LOCKED, &shared_region_persistence, 0, ""); /* * shared_region_check_np: * * This system call is intended for dyld. * * dyld calls this when any process starts to see if the process's shared * region is already set up and ready to use. * This call returns the base address of the first mapping in the * process's shared region's first mapping. * dyld will then check what's mapped at that address. * * If the shared region is empty, dyld will then attempt to map the shared * cache file in the shared region via the shared_region_map_and_slide_2_np() * system call. * * If something's already mapped in the shared region, dyld will check if it * matches the shared cache it would like to use for that process. * If it matches, evrything's ready and the process can proceed and use the * shared region. * If it doesn't match, dyld will unmap the shared region and map the shared * cache into the process's address space via mmap(). * * A NULL pointer argument can be used by dyld to indicate it has unmapped * the shared region. We will remove the shared_region reference from the task. * * ERROR VALUES * EINVAL no shared region * ENOMEM shared region is empty * EFAULT bad address for "start_address" */ int shared_region_check_np( __unused struct proc *p, struct shared_region_check_np_args *uap, __unused int *retvalp) { vm_shared_region_t shared_region; mach_vm_offset_t start_address = 0; int error = 0; kern_return_t kr = KERN_FAILURE; task_t task = current_task(); SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] -> check_np(0x%llx)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->start_address)); /* * Special value of start_address used to indicate that map_with_linking() should * no longer be allowed in this process */ if (uap->start_address == (task_get_64bit_addr(task) ? DYLD_VM_END_MWL : (uint32_t)DYLD_VM_END_MWL)) { p->p_disallow_map_with_linking = TRUE; return 0; } /* retrieve the current task's shared region */ shared_region = vm_shared_region_get(task); if (shared_region != NULL) { /* * A NULL argument is used by dyld to indicate the task * has unmapped its shared region. */ if (uap->start_address == 0) { /* unmap it first */ vm_shared_region_remove(task, shared_region); vm_shared_region_set(task, NULL); } else { /* retrieve address of its first mapping... */ kr = vm_shared_region_start_address(shared_region, &start_address); if (kr != KERN_SUCCESS) { SHARED_REGION_TRACE_ERROR(("shared_region: %p [%d(%s)] " "check_np(0x%llx) " "vm_shared_region_start_address() returned 0x%x\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->start_address, kr)); error = ENOMEM; } if (error == 0) { /* Insert the shared region submap and various bits of debug info into the task. */ kr = vm_shared_region_update_task(task, shared_region, start_address); if (kr != KERN_SUCCESS) { SHARED_REGION_TRACE_ERROR(("shared_region: %p [%d(%s)] " "check_np(0x%llx) " "vm_shared_update_task() returned 0x%x\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->start_address, kr)); error = ENOMEM; } } #if __has_feature(ptrauth_calls) /* * Remap any section of the shared library that * has authenticated pointers into private memory. */ if ((error == 0) && (vm_shared_region_auth_remap(shared_region) != KERN_SUCCESS)) { SHARED_REGION_TRACE_ERROR(("shared_region: %p [%d(%s)] " "check_np(0x%llx) " "vm_shared_region_auth_remap() failed\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->start_address)); error = ENOMEM; } #endif /* __has_feature(ptrauth_calls) */ /* Give the start address to the caller */ if (error == 0) { error = copyout(&start_address, (user_addr_t) uap->start_address, sizeof(start_address)); if (error != 0) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] " "check_np(0x%llx) " "copyout(0x%llx) error %d\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->start_address, (uint64_t)start_address, error)); } } } vm_shared_region_deallocate(shared_region); } else { /* no shared region ! */ error = EINVAL; } SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] check_np(0x%llx) <- 0x%llx %d\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->start_address, (uint64_t)start_address, error)); return error; } static int shared_region_copyin( struct proc *p, user_addr_t user_addr, unsigned int count, unsigned int element_size, void *kernel_data) { int error = 0; vm_size_t size = count * element_size; error = copyin(user_addr, kernel_data, size); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(): " "copyin(0x%llx, %ld) failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)user_addr, (long)size, error)); } return error; } /* * A reasonable upper limit to prevent overflow of allocation/copyin. */ #define _SR_FILE_MAPPINGS_MAX_FILES 256 /* forward declaration */ __attribute__((noinline)) static void shared_region_map_and_slide_cleanup( struct proc *p, uint32_t files_count, struct _sr_file_mappings *sr_file_mappings, struct vm_shared_region *shared_region); /* * Setup part of _shared_region_map_and_slide(). * It had to be broken out of _shared_region_map_and_slide() to * prevent compiler inlining from blowing out the stack. */ __attribute__((noinline)) static int shared_region_map_and_slide_setup( struct proc *p, uint32_t files_count, struct shared_file_np *files, uint32_t mappings_count, struct shared_file_mapping_slide_np *mappings, struct _sr_file_mappings **sr_file_mappings, struct vm_shared_region **shared_region_ptr, struct vnode *rdir_vp) { int error = 0; struct _sr_file_mappings *srfmp; uint32_t mappings_next; struct vnode_attr va; off_t fs; #if CONFIG_MACF vm_prot_t maxprot = VM_PROT_ALL; #endif uint32_t i; struct vm_shared_region *shared_region = NULL; boolean_t is_driverkit = task_is_driver(current_task()); SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] -> map_and_slide_setup\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); if (files_count > _SR_FILE_MAPPINGS_MAX_FILES) { error = E2BIG; goto done; } if (files_count == 0) { error = EINVAL; goto done; } *sr_file_mappings = kalloc_type(struct _sr_file_mappings, files_count, Z_WAITOK | Z_ZERO); if (*sr_file_mappings == NULL) { error = ENOMEM; goto done; } mappings_next = 0; for (i = 0; i < files_count; i++) { srfmp = &(*sr_file_mappings)[i]; srfmp->fd = files[i].sf_fd; srfmp->mappings_count = files[i].sf_mappings_count; srfmp->mappings = &mappings[mappings_next]; mappings_next += srfmp->mappings_count; if (mappings_next > mappings_count) { error = EINVAL; goto done; } srfmp->slide = files[i].sf_slide; } /* get the process's shared region (setup in vm_map_exec()) */ shared_region = vm_shared_region_get(current_task()); *shared_region_ptr = shared_region; if (shared_region == NULL) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(): " "no shared region\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); error = EINVAL; goto done; } /* * Check the shared region matches the current root * directory of this process. Deny the mapping to * avoid tainting the shared region with something that * doesn't quite belong into it. */ struct vnode *sr_vnode = vm_shared_region_root_dir(shared_region); if (sr_vnode != NULL ? rdir_vp != sr_vnode : rdir_vp != rootvnode) { SHARED_REGION_TRACE_ERROR( ("shared_region: map(%p) root_dir mismatch\n", (void *)VM_KERNEL_ADDRPERM(current_thread()))); error = EPERM; goto done; } for (srfmp = &(*sr_file_mappings)[0]; srfmp < &(*sr_file_mappings)[files_count]; srfmp++) { if (srfmp->mappings_count == 0) { /* no mappings here... */ continue; } /* * A file descriptor of -1 is used to indicate that the data * to be put in the shared region for this mapping comes directly * from the processes address space. Ensure we have proper alignments. */ if (srfmp->fd == -1) { /* only allow one mapping per fd */ if (srfmp->mappings_count > 1) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map data >1 mapping\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); error = EINVAL; goto done; } /* * The destination address and size must be page aligned. */ struct shared_file_mapping_slide_np *mapping = &srfmp->mappings[0]; mach_vm_address_t dest_addr = mapping->sms_address; mach_vm_size_t map_size = mapping->sms_size; if (!vm_map_page_aligned(dest_addr, vm_map_page_mask(current_map()))) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map data destination 0x%llx not aligned\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, dest_addr)); error = EINVAL; goto done; } if (!vm_map_page_aligned(map_size, vm_map_page_mask(current_map()))) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map data size 0x%llx not aligned\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, map_size)); error = EINVAL; goto done; } continue; } /* get file structure from file descriptor */ error = fp_get_ftype(p, srfmp->fd, DTYPE_VNODE, EINVAL, &srfmp->fp); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map: " "fd=%d lookup failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, srfmp->fd, error)); goto done; } /* we need at least read permission on the file */ if (!(srfmp->fp->fp_glob->fg_flag & FREAD)) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map: " "fd=%d not readable\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, srfmp->fd)); error = EPERM; goto done; } /* get vnode from file structure */ error = vnode_getwithref((vnode_t)fp_get_data(srfmp->fp)); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map: " "fd=%d getwithref failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, srfmp->fd, error)); goto done; } srfmp->vp = (struct vnode *)fp_get_data(srfmp->fp); /* make sure the vnode is a regular file */ if (srfmp->vp->v_type != VREG) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "not a file (type=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name, srfmp->vp->v_type)); error = EINVAL; goto done; } #if CONFIG_MACF /* pass in 0 for the offset argument because AMFI does not need the offset * of the shared cache */ error = mac_file_check_mmap(vfs_context_ucred(vfs_context_current()), srfmp->fp->fp_glob, VM_PROT_ALL, MAP_FILE | MAP_PRIVATE | MAP_FIXED, 0, &maxprot); if (error) { goto done; } #endif /* MAC */ #if XNU_TARGET_OS_OSX && defined(__arm64__) /* * Check if the shared cache is in the trust cache; * if so, we can skip the root ownership check. */ #if DEVELOPMENT || DEBUG /* * Skip both root ownership and trust cache check if * enforcement is disabled. */ if (!cs_system_enforcement()) { goto after_root_check; } #endif /* DEVELOPMENT || DEBUG */ struct cs_blob *blob = csvnode_get_blob(srfmp->vp, 0); if (blob == NULL) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "missing CS blob\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); goto root_check; } const uint8_t *cdhash = csblob_get_cdhash(blob); if (cdhash == NULL) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "missing cdhash\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); goto root_check; } bool in_trust_cache = false; TrustCacheQueryToken_t qt; if (query_trust_cache(kTCQueryTypeAll, cdhash, &qt) == KERN_SUCCESS) { TCType_t tc_type = kTCTypeInvalid; TCReturn_t tc_ret = amfi->TrustCache.queryGetTCType(&qt, &tc_type); in_trust_cache = (tc_ret.error == kTCReturnSuccess && (tc_type == kTCTypeCryptex1BootOS || tc_type == kTCTypeStatic || tc_type == kTCTypeEngineering)); } if (!in_trust_cache) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "not in trust cache\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); goto root_check; } goto after_root_check; root_check: #endif /* XNU_TARGET_OS_OSX && defined(__arm64__) */ /* The shared cache file must be owned by root */ VATTR_INIT(&va); VATTR_WANTED(&va, va_uid); error = vnode_getattr(srfmp->vp, &va, vfs_context_current()); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "vnode_getattr(%p) failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), error)); goto done; } if (va.va_uid != 0) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "owned by uid=%d instead of 0\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name, va.va_uid)); error = EPERM; goto done; } #if XNU_TARGET_OS_OSX && defined(__arm64__) after_root_check: #endif /* XNU_TARGET_OS_OSX && defined(__arm64__) */ #if CONFIG_CSR if (csr_check(CSR_ALLOW_UNRESTRICTED_FS) != 0) { VATTR_INIT(&va); VATTR_WANTED(&va, va_flags); error = vnode_getattr(srfmp->vp, &va, vfs_context_current()); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "vnode_getattr(%p) failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), error)); goto done; } if (!(va.va_flags & SF_RESTRICTED)) { /* * CSR is not configured in CSR_ALLOW_UNRESTRICTED_FS mode, and * the shared cache file is NOT SIP-protected, so reject the * mapping request */ SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'), " "vnode is not SIP-protected. \n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); error = EPERM; goto done; } } #else /* CONFIG_CSR */ /* * Devices without SIP/ROSP need to make sure that the shared cache * is either on the root volume or in the preboot cryptex volume. */ assert(rdir_vp != NULL); if (srfmp->vp->v_mount != rdir_vp->v_mount) { vnode_t preboot_vp = NULL; #if XNU_TARGET_OS_OSX #define PREBOOT_CRYPTEX_PATH "/System/Volumes/Preboot/Cryptexes" #else #define PREBOOT_CRYPTEX_PATH "/private/preboot/Cryptexes" #endif error = vnode_lookup(PREBOOT_CRYPTEX_PATH, 0, &preboot_vp, vfs_context_current()); if (error || srfmp->vp->v_mount != preboot_vp->v_mount) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "not on process' root volume nor preboot volume\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); error = EPERM; if (preboot_vp) { (void)vnode_put(preboot_vp); } goto done; } else if (preboot_vp) { (void)vnode_put(preboot_vp); } } #endif /* CONFIG_CSR */ if (scdir_enforce) { char **expected_scdir_path = is_driverkit ? driverkit_scdir_path : scdir_path; struct vnode *scdir_vp = NULL; for (expected_scdir_path = is_driverkit ? driverkit_scdir_path : scdir_path; *expected_scdir_path != NULL; expected_scdir_path++) { /* get vnode for expected_scdir_path */ error = vnode_lookup(*expected_scdir_path, 0, &scdir_vp, vfs_context_current()); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)]: " "vnode_lookup(%s) failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, *expected_scdir_path, error)); continue; } /* check if parent is scdir_vp */ assert(scdir_vp != NULL); if (vnode_parent(srfmp->vp) == scdir_vp) { (void)vnode_put(scdir_vp); scdir_vp = NULL; goto scdir_ok; } (void)vnode_put(scdir_vp); scdir_vp = NULL; } /* nothing matches */ SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "shared cache file not in expected directory\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); error = EPERM; goto done; } scdir_ok: /* get vnode size */ error = vnode_size(srfmp->vp, &fs, vfs_context_current()); if (error) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "vnode_size(%p) failed (error=%d)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), error)); goto done; } srfmp->file_size = fs; /* get the file's memory object handle */ srfmp->file_control = ubc_getobject(srfmp->vp, UBC_HOLDOBJECT); if (srfmp->file_control == MEMORY_OBJECT_CONTROL_NULL) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "no memory object\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name)); error = EINVAL; goto done; } /* check that the mappings are properly covered by code signatures */ if (!cs_system_enforcement()) { /* code signing is not enforced: no need to check */ } else { for (i = 0; i < srfmp->mappings_count; i++) { if (srfmp->mappings[i].sms_init_prot & VM_PROT_ZF) { /* zero-filled mapping: not backed by the file */ continue; } if (ubc_cs_is_range_codesigned(srfmp->vp, srfmp->mappings[i].sms_file_offset, srfmp->mappings[i].sms_size)) { /* this mapping is fully covered by code signatures */ continue; } SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(%p:'%s'): " "mapping #%d/%d [0x%llx:0x%llx:0x%llx:0x%x:0x%x] " "is not code-signed\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (void *)VM_KERNEL_ADDRPERM(srfmp->vp), srfmp->vp->v_name, i, srfmp->mappings_count, srfmp->mappings[i].sms_address, srfmp->mappings[i].sms_size, srfmp->mappings[i].sms_file_offset, srfmp->mappings[i].sms_max_prot, srfmp->mappings[i].sms_init_prot)); error = EINVAL; goto done; } } } done: if (error != 0) { shared_region_map_and_slide_cleanup(p, files_count, *sr_file_mappings, shared_region); *sr_file_mappings = NULL; *shared_region_ptr = NULL; } SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] map_and_slide_setup <- %d\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, error)); return error; } /* * shared_region_map_np() * * This system call is intended for dyld. * * dyld uses this to map a shared cache file into a shared region. * This is usually done only the first time a shared cache is needed. * Subsequent processes will just use the populated shared region without * requiring any further setup. */ static int _shared_region_map_and_slide( struct proc *p, uint32_t files_count, struct shared_file_np *files, uint32_t mappings_count, struct shared_file_mapping_slide_np *mappings) { int error = 0; kern_return_t kr = KERN_SUCCESS; struct _sr_file_mappings *sr_file_mappings = NULL; struct vnode *rdir_vp = NULL; struct vm_shared_region *shared_region = NULL; /* * Get a reference to the current proc's root dir. * Need this to prevent racing with chroot. */ proc_fdlock(p); rdir_vp = p->p_fd.fd_rdir; if (rdir_vp == NULL) { rdir_vp = rootvnode; } assert(rdir_vp != NULL); vnode_get(rdir_vp); proc_fdunlock(p); /* * Turn files, mappings into sr_file_mappings and other setup. */ error = shared_region_map_and_slide_setup(p, files_count, files, mappings_count, mappings, &sr_file_mappings, &shared_region, rdir_vp); if (error != 0) { vnode_put(rdir_vp); return error; } /* map the file(s) into that shared region's submap */ kr = vm_shared_region_map_file(shared_region, files_count, sr_file_mappings); if (kr != KERN_SUCCESS) { SHARED_REGION_TRACE_ERROR(("shared_region: %p [%d(%s)] map(): " "vm_shared_region_map_file() failed kr=0x%x\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, kr)); } /* convert kern_return_t to errno */ switch (kr) { case KERN_SUCCESS: error = 0; break; case KERN_INVALID_ADDRESS: error = EFAULT; break; case KERN_PROTECTION_FAILURE: error = EPERM; break; case KERN_NO_SPACE: error = ENOMEM; break; case KERN_FAILURE: case KERN_INVALID_ARGUMENT: default: error = EINVAL; break; } /* * Mark that this process is now using split libraries. */ if (error == 0 && (p->p_flag & P_NOSHLIB)) { OSBitAndAtomic(~((uint32_t)P_NOSHLIB), &p->p_flag); } vnode_put(rdir_vp); shared_region_map_and_slide_cleanup(p, files_count, sr_file_mappings, shared_region); SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] <- map\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); return error; } /* * Clean up part of _shared_region_map_and_slide() * It had to be broken out of _shared_region_map_and_slide() to * prevent compiler inlining from blowing out the stack. */ __attribute__((noinline)) static void shared_region_map_and_slide_cleanup( struct proc *p, uint32_t files_count, struct _sr_file_mappings *sr_file_mappings, struct vm_shared_region *shared_region) { struct _sr_file_mappings *srfmp; struct vnode_attr va; if (sr_file_mappings != NULL) { for (srfmp = &sr_file_mappings[0]; srfmp < &sr_file_mappings[files_count]; srfmp++) { if (srfmp->vp != NULL) { vnode_lock_spin(srfmp->vp); srfmp->vp->v_flag |= VSHARED_DYLD; vnode_unlock(srfmp->vp); /* update the vnode's access time */ if (!(vnode_vfsvisflags(srfmp->vp) & MNT_NOATIME)) { VATTR_INIT(&va); nanotime(&va.va_access_time); VATTR_SET_ACTIVE(&va, va_access_time); vnode_setattr(srfmp->vp, &va, vfs_context_current()); } #if NAMEDSTREAMS /* * If the shared cache is compressed, it may * have a namedstream vnode instantiated for * for it. That namedstream vnode will also * have to be marked with VSHARED_DYLD. */ if (vnode_hasnamedstreams(srfmp->vp)) { vnode_t svp; if (vnode_getnamedstream(srfmp->vp, &svp, XATTR_RESOURCEFORK_NAME, NS_OPEN, 0, vfs_context_kernel()) == 0) { vnode_lock_spin(svp); svp->v_flag |= VSHARED_DYLD; vnode_unlock(svp); vnode_put(svp); } } #endif /* NAMEDSTREAMS */ /* * release the vnode... * ubc_map() still holds it for us in the non-error case */ (void) vnode_put(srfmp->vp); srfmp->vp = NULL; } if (srfmp->fp != NULL) { /* release the file descriptor */ fp_drop(p, srfmp->fd, srfmp->fp, 0); srfmp->fp = NULL; } } kfree_type(struct _sr_file_mappings, files_count, sr_file_mappings); } if (shared_region != NULL) { vm_shared_region_deallocate(shared_region); } } /* * For each file mapped, we may have mappings for: * TEXT, EXECUTE, LINKEDIT, DATA_CONST, __AUTH, DATA * so let's round up to 8 mappings per file. */ #define SFM_MAX (_SR_FILE_MAPPINGS_MAX_FILES * 8) /* max mapping structs allowed to pass in */ /* * This is the new interface for setting up shared region mappings. * * The slide used for shared regions setup using this interface is done differently * from the old interface. The slide value passed in the shared_files_np represents * a max value. The kernel will choose a random value based on that, then use it * for all shared regions. */ #if defined (__x86_64__) #define SLIDE_AMOUNT_MASK ~FOURK_PAGE_MASK #else #define SLIDE_AMOUNT_MASK ~SIXTEENK_PAGE_MASK #endif static inline __result_use_check kern_return_t shared_region_map_and_slide_2_np_sanitize( struct proc *p, user_addr_t mappings_userspace_addr, unsigned int count, shared_file_mapping_slide_np_t *mappings) { kern_return_t kr; vm_map_t map = current_map(); mach_vm_address_t addr, end; mach_vm_offset_t offset, offset_end; mach_vm_size_t size, offset_size; user_addr_t slide_start, slide_end, slide_size; vm_prot_t cur; vm_prot_t max; user_addr_t user_addr = mappings_userspace_addr; for (size_t i = 0; i < count; i++) { shared_file_mapping_slide_np_ut mapping_u; /* * First we bring each mapping struct into our kernel stack to * avoid TOCTOU. */ kr = shared_region_copyin( p, user_addr, 1, // copy 1 element at a time sizeof(shared_file_mapping_slide_np_ut), &mapping_u); if (__improbable(kr != KERN_SUCCESS)) { return kr; } /* * Then, we sanitize the data on the kernel stack. */ kr = vm_sanitize_addr_size( mapping_u.sms_address_u, mapping_u.sms_size_u, VM_SANITIZE_CALLER_SHARED_REGION_MAP_AND_SLIDE_2_NP, map, (VM_SANITIZE_FLAGS_SIZE_ZERO_FALLTHROUGH | VM_SANITIZE_FLAGS_CHECK_ALIGNED_START | VM_SANITIZE_FLAGS_CHECK_ALIGNED_SIZE), &addr, &end, &size); if (__improbable(kr != KERN_SUCCESS)) { return kr; } kr = vm_sanitize_addr_size( mapping_u.sms_file_offset_u, mapping_u.sms_size_u, VM_SANITIZE_CALLER_SHARED_REGION_MAP_AND_SLIDE_2_NP, PAGE_MASK, (VM_SANITIZE_FLAGS_SIZE_ZERO_FALLTHROUGH | VM_SANITIZE_FLAGS_GET_UNALIGNED_VALUES), &offset, &offset_end, &offset_size); if (__improbable(kr != KERN_SUCCESS)) { return kr; } if (__improbable(0 != (offset & vm_map_page_mask(map)))) { return KERN_INVALID_ARGUMENT; } /* * Unsafe access is immediately followed by wrap to * convert from addr to size. */ mach_vm_size_ut sms_slide_size_u = vm_sanitize_wrap_size( VM_SANITIZE_UNSAFE_UNWRAP( mapping_u.sms_slide_size_u)); kr = vm_sanitize_addr_size( mapping_u.sms_slide_start_u, sms_slide_size_u, VM_SANITIZE_CALLER_SHARED_REGION_MAP_AND_SLIDE_2_NP, map, (VM_SANITIZE_FLAGS_SIZE_ZERO_FALLTHROUGH | VM_SANITIZE_FLAGS_GET_UNALIGNED_VALUES), &slide_start, &slide_end, &slide_size); if (__improbable(kr != KERN_SUCCESS)) { return kr; } kr = vm_sanitize_cur_and_max_prots( mapping_u.sms_init_prot_u, mapping_u.sms_max_prot_u, VM_SANITIZE_CALLER_SHARED_REGION_MAP_AND_SLIDE_2_NP, map, VM_PROT_SFM_EXTENSIONS_MASK | VM_PROT_TPRO, &cur, &max); if (__improbable(kr != KERN_SUCCESS)) { return kr; } /* * Finally, we move the data from the kernel stack to our * caller-allocated kernel heap buffer. */ mappings[i].sms_address = addr; mappings[i].sms_size = size; mappings[i].sms_file_offset = offset; mappings[i].sms_slide_size = slide_size; mappings[i].sms_slide_start = slide_start; mappings[i].sms_max_prot = max; mappings[i].sms_init_prot = cur; if (__improbable(os_add_overflow( user_addr, sizeof(shared_file_mapping_slide_np_ut), &user_addr))) { return KERN_INVALID_ARGUMENT; } } return KERN_SUCCESS; } int shared_region_map_and_slide_2_np( struct proc *p, struct shared_region_map_and_slide_2_np_args *uap, __unused int *retvalp) { unsigned int files_count; struct shared_file_np *shared_files = NULL; unsigned int mappings_count; struct shared_file_mapping_slide_np *mappings = NULL; kern_return_t kr = KERN_SUCCESS; files_count = uap->files_count; mappings_count = uap->mappings_count; SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] -> map_and_slide(0x%llx)\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->mappings_u)); if (files_count == 0) { SHARED_REGION_TRACE_INFO( ("shared_region: %p [%d(%s)] map(): " "no files\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); kr = 0; /* no files to map: we're done ! */ goto done; } else if (files_count <= _SR_FILE_MAPPINGS_MAX_FILES) { shared_files = kalloc_data(files_count * sizeof(shared_files[0]), Z_WAITOK); if (shared_files == NULL) { kr = KERN_RESOURCE_SHORTAGE; goto done; } } else { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(): " "too many files (%d) max %d\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, files_count, _SR_FILE_MAPPINGS_MAX_FILES)); kr = KERN_FAILURE; goto done; } if (mappings_count == 0) { SHARED_REGION_TRACE_INFO( ("shared_region: %p [%d(%s)] map(): " "no mappings\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); kr = 0; /* no mappings: we're done ! */ goto done; } else if (mappings_count <= SFM_MAX) { mappings = kalloc_data(mappings_count * sizeof(mappings[0]), Z_WAITOK); if (mappings == NULL) { kr = KERN_RESOURCE_SHORTAGE; goto done; } } else { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(): " "too many mappings (%d) max %d\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, mappings_count, SFM_MAX)); kr = KERN_FAILURE; goto done; } /* * struct shared_file_np does not have fields that are subject to * sanitization, it is thus copied from userspace as is. */ kr = shared_region_copyin(p, uap->files, files_count, sizeof(shared_files[0]), shared_files); if (kr != KERN_SUCCESS) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] copyin() returned 0x%x\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, kr)); goto done; } kr = shared_region_map_and_slide_2_np_sanitize( p, uap->mappings_u, mappings_count, mappings); if (__improbable(kr != KERN_SUCCESS)) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] sanitize() returned 0x%x\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, kr)); kr = vm_sanitize_get_kr(kr); goto done; } uint32_t max_slide = shared_files[0].sf_slide; uint32_t random_val; uint32_t slide_amount; if (max_slide != 0) { read_random(&random_val, sizeof random_val); slide_amount = ((random_val % max_slide) & SLIDE_AMOUNT_MASK); } else { slide_amount = 0; } #if DEVELOPMENT || DEBUG extern bool bootarg_disable_aslr; if (bootarg_disable_aslr) { slide_amount = 0; } #endif /* DEVELOPMENT || DEBUG */ /* * Fix up the mappings to reflect the desired slide. */ unsigned int f; unsigned int m = 0; unsigned int i; for (f = 0; f < files_count; ++f) { shared_files[f].sf_slide = slide_amount; for (i = 0; i < shared_files[f].sf_mappings_count; ++i, ++m) { if (m >= mappings_count) { SHARED_REGION_TRACE_ERROR( ("shared_region: %p [%d(%s)] map(): " "mapping count argument was too small\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm)); kr = KERN_FAILURE; goto done; } if (__improbable( os_add_overflow( mappings[m].sms_address, slide_amount, &mappings[m].sms_address))) { kr = KERN_INVALID_ARGUMENT; goto done; } if (mappings[m].sms_slide_size != 0) { mach_vm_address_t discard; /* Slide and check that new start/size pairs do not overflow. */ if (__improbable( os_add_overflow( mappings[m].sms_slide_start, slide_amount, &mappings[m].sms_slide_start) || os_add_overflow( mappings[m].sms_slide_start, mappings[m].sms_slide_size, &discard))) { kr = KERN_INVALID_ARGUMENT; goto done; } } } } kr = _shared_region_map_and_slide(p, files_count, shared_files, mappings_count, mappings); done: kfree_data(shared_files, files_count * sizeof(shared_files[0])); kfree_data(mappings, mappings_count * sizeof(mappings[0])); SHARED_REGION_TRACE_DEBUG( ("shared_region: %p [%d(%s)] map_and_slide(0x%llx) <- 0x%x\n", (void *)VM_KERNEL_ADDRPERM(current_thread()), proc_getpid(p), p->p_comm, (uint64_t)uap->mappings_u, kr)); return kr; } SYSCTL_QUAD(_vm, OID_AUTO, vmwls_total_success, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_total_success, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_total_fail, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_total_fail, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_overflow, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_overflow, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_bad_offset, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_bad_offset, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_bad_addr, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_bad_addr, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_bad_prot, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_bad_prot, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_bad_file, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_bad_file, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_bad_shadows, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_bad_shadows, ""); SYSCTL_QUAD(_vm, OID_AUTO, vmwls_bad_cow, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_with_linking_stats.vmwls_bad_cow, ""); /* * A syscall for dyld to use to map data pages that need load time relocation fixups. * The fixups are performed by a custom pager during page-in, so the pages still appear * "clean" and hence are easily discarded under memory pressure. They can be re-paged-in * on demand later, all w/o using the compressor. * * Note these page are treated as MAP_PRIVATE. So if the application dirties any pages while * running, they are COW'd as normal. */ int map_with_linking_np( struct proc *p, struct map_with_linking_np_args *uap, __unused int *retvalp) { uint32_t region_count; uint32_t r; struct mwl_region *regions = NULL; struct mwl_region *rp; uint32_t link_info_size; void *link_info = NULL; /* starts with a struct mwl_info_hdr */ struct mwl_info_hdr *info_hdr = NULL; uint64_t binds_size; int fd; struct fileproc *fp = NULL; struct vnode *vp = NULL; size_t file_size; off_t fs; struct vnode_attr va; memory_object_control_t file_control = NULL; int error; kern_return_t kr = KERN_SUCCESS; /* * Check if dyld has told us it finished with this call. */ if (p->p_disallow_map_with_linking) { printf("%s: [%d(%s)]: map__with_linking() was disabled\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } /* * First we do some sanity checking on what dyld has passed us. */ region_count = uap->region_count; link_info_size = uap->link_info_size; if (region_count == 0) { printf("%s: [%d(%s)]: region_count == 0\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } if (region_count > MWL_MAX_REGION_COUNT) { printf("%s: [%d(%s)]: region_count too big %d\n", __func__, proc_getpid(p), p->p_comm, region_count); kr = KERN_FAILURE; goto done; } if (link_info_size <= MWL_MIN_LINK_INFO_SIZE) { printf("%s: [%d(%s)]: link_info_size too small\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } if (link_info_size >= MWL_MAX_LINK_INFO_SIZE) { printf("%s: [%d(%s)]: link_info_size too big %d\n", __func__, proc_getpid(p), p->p_comm, link_info_size); kr = KERN_FAILURE; goto done; } /* * Allocate and copyin the regions and link info */ regions = kalloc_data(region_count * sizeof(regions[0]), Z_WAITOK); if (regions == NULL) { printf("%s: [%d(%s)]: failed to allocate regions\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_RESOURCE_SHORTAGE; goto done; } kr = shared_region_copyin(p, uap->regions, region_count, sizeof(regions[0]), regions); if (kr != KERN_SUCCESS) { printf("%s: [%d(%s)]: failed to copyin regions kr=%d\n", __func__, proc_getpid(p), p->p_comm, kr); goto done; } link_info = kalloc_data(link_info_size, Z_WAITOK); if (link_info == NULL) { printf("%s: [%d(%s)]: failed to allocate link_info\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_RESOURCE_SHORTAGE; goto done; } kr = shared_region_copyin(p, uap->link_info, 1, link_info_size, link_info); if (kr != KERN_SUCCESS) { printf("%s: [%d(%s)]: failed to copyin link_info kr=%d\n", __func__, proc_getpid(p), p->p_comm, kr); goto done; } /* * Do some verification the data structures. */ info_hdr = (struct mwl_info_hdr *)link_info; if (info_hdr->mwli_version != MWL_INFO_VERS) { printf("%s: [%d(%s)]: unrecognized mwli_version=%d\n", __func__, proc_getpid(p), p->p_comm, info_hdr->mwli_version); kr = KERN_FAILURE; goto done; } if (info_hdr->mwli_binds_offset > link_info_size) { printf("%s: [%d(%s)]: mwli_binds_offset too large %d\n", __func__, proc_getpid(p), p->p_comm, info_hdr->mwli_binds_offset); kr = KERN_FAILURE; goto done; } /* some older devs have s/w page size > h/w page size, no need to support them */ if (info_hdr->mwli_page_size != PAGE_SIZE) { /* no printf, since this is expected on some devices */ kr = KERN_INVALID_ARGUMENT; goto done; } binds_size = (uint64_t)info_hdr->mwli_binds_count * ((info_hdr->mwli_pointer_format == DYLD_CHAINED_PTR_32) ? 4 : 8); if (binds_size > link_info_size - info_hdr->mwli_binds_offset) { printf("%s: [%d(%s)]: mwli_binds_count too large %d\n", __func__, proc_getpid(p), p->p_comm, info_hdr->mwli_binds_count); kr = KERN_FAILURE; goto done; } if (info_hdr->mwli_chains_offset > link_info_size) { printf("%s: [%d(%s)]: mwli_chains_offset too large %d\n", __func__, proc_getpid(p), p->p_comm, info_hdr->mwli_chains_offset); kr = KERN_FAILURE; goto done; } /* * Ensure the chained starts in the link info and make sure the * segment info offsets are within bounds. */ if (info_hdr->mwli_chains_size < sizeof(struct dyld_chained_starts_in_image)) { printf("%s: [%d(%s)]: mwli_chains_size too small %d\n", __func__, proc_getpid(p), p->p_comm, info_hdr->mwli_chains_size); kr = KERN_FAILURE; goto done; } if (info_hdr->mwli_chains_size > link_info_size - info_hdr->mwli_chains_offset) { printf("%s: [%d(%s)]: mwli_chains_size too large %d\n", __func__, proc_getpid(p), p->p_comm, info_hdr->mwli_chains_size); kr = KERN_FAILURE; goto done; } /* Note that more verification of offsets is done in the pager itself */ /* * Ensure we've only been given one FD and verify valid protections. */ fd = regions[0].mwlr_fd; for (r = 0; r < region_count; ++r) { if (regions[r].mwlr_fd != fd) { printf("%s: [%d(%s)]: mwlr_fd mismatch %d and %d\n", __func__, proc_getpid(p), p->p_comm, fd, regions[r].mwlr_fd); kr = KERN_FAILURE; goto done; } /* * Only allow data mappings and not zero fill. Permit TPRO * mappings only when VM_PROT_READ | VM_PROT_WRITE. */ if (regions[r].mwlr_protections & VM_PROT_EXECUTE) { printf("%s: [%d(%s)]: mwlr_protections EXECUTE not allowed\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } if (regions[r].mwlr_protections & VM_PROT_ZF) { printf("%s: [%d(%s)]: region %d, found VM_PROT_ZF not allowed\n", __func__, proc_getpid(p), p->p_comm, r); kr = KERN_FAILURE; goto done; } if ((regions[r].mwlr_protections & VM_PROT_TPRO) && !(regions[r].mwlr_protections & VM_PROT_WRITE)) { printf("%s: [%d(%s)]: region %d, found VM_PROT_TPRO without VM_PROT_WRITE\n", __func__, proc_getpid(p), p->p_comm, r); kr = KERN_FAILURE; goto done; } } /* get file structure from file descriptor */ error = fp_get_ftype(p, fd, DTYPE_VNODE, EINVAL, &fp); if (error) { printf("%s: [%d(%s)]: fp_get_ftype() failed, error %d\n", __func__, proc_getpid(p), p->p_comm, error); kr = KERN_FAILURE; goto done; } /* We need at least read permission on the file */ if (!(fp->fp_glob->fg_flag & FREAD)) { printf("%s: [%d(%s)]: not readable\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } /* Get the vnode from file structure */ vp = (struct vnode *)fp_get_data(fp); error = vnode_getwithref(vp); if (error) { printf("%s: [%d(%s)]: failed to get vnode, error %d\n", __func__, proc_getpid(p), p->p_comm, error); kr = KERN_FAILURE; vp = NULL; /* just to be sure */ goto done; } /* Make sure the vnode is a regular file */ if (vp->v_type != VREG) { printf("%s: [%d(%s)]: vnode not VREG\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } /* get vnode size */ error = vnode_size(vp, &fs, vfs_context_current()); if (error) { goto done; } file_size = fs; /* get the file's memory object handle */ file_control = ubc_getobject(vp, UBC_HOLDOBJECT); if (file_control == MEMORY_OBJECT_CONTROL_NULL) { printf("%s: [%d(%s)]: no memory object\n", __func__, proc_getpid(p), p->p_comm); kr = KERN_FAILURE; goto done; } for (r = 0; r < region_count; ++r) { rp = ®ions[r]; #if CONFIG_MACF vm_prot_t prot = (rp->mwlr_protections & VM_PROT_ALL); error = mac_file_check_mmap(vfs_context_ucred(vfs_context_current()), fp->fp_glob, prot, MAP_FILE | MAP_PRIVATE | MAP_FIXED, rp->mwlr_file_offset, &prot); if (error) { printf("%s: [%d(%s)]: mac_file_check_mmap() failed, region %d, error %d\n", __func__, proc_getpid(p), p->p_comm, r, error); kr = KERN_FAILURE; goto done; } #endif /* MAC */ /* check that the mappings are properly covered by code signatures */ if (cs_system_enforcement()) { if (!ubc_cs_is_range_codesigned(vp, rp->mwlr_file_offset, rp->mwlr_size)) { printf("%s: [%d(%s)]: region %d, not code signed\n", __func__, proc_getpid(p), p->p_comm, r); kr = KERN_FAILURE; goto done; } } } /* update the vnode's access time */ if (!(vnode_vfsvisflags(vp) & MNT_NOATIME)) { VATTR_INIT(&va); nanotime(&va.va_access_time); VATTR_SET_ACTIVE(&va, va_access_time); vnode_setattr(vp, &va, vfs_context_current()); } /* get the VM to do the work */ kr = vm_map_with_linking(proc_task(p), regions, region_count, &link_info, link_info_size, file_control); done: if (fp != NULL) { /* release the file descriptor */ fp_drop(p, fd, fp, 0); } if (vp != NULL) { (void)vnode_put(vp); } if (regions != NULL) { kfree_data(regions, region_count * sizeof(regions[0])); } /* link info is NULL if it is used in the pager, if things worked */ if (link_info != NULL) { kfree_data(link_info, link_info_size); } switch (kr) { case KERN_SUCCESS: return 0; case KERN_RESOURCE_SHORTAGE: return ENOMEM; default: return EINVAL; } } #if DEBUG || DEVELOPMENT SYSCTL_INT(_vm, OID_AUTO, dyld_pager_count, CTLFLAG_RD | CTLFLAG_LOCKED, &dyld_pager_count, 0, ""); SYSCTL_INT(_vm, OID_AUTO, dyld_pager_count_max, CTLFLAG_RD | CTLFLAG_LOCKED, &dyld_pager_count_max, 0, ""); #endif /* DEBUG || DEVELOPMENT */ /* sysctl overflow room */ SYSCTL_INT(_vm, OID_AUTO, pagesize, CTLFLAG_RD | CTLFLAG_LOCKED, (int *) &page_size, 0, "vm page size"); /* vm_page_free_target is provided as a makeshift solution for applications that want to * allocate buffer space, possibly purgeable memory, but not cause inactive pages to be * reclaimed. It allows the app to calculate how much memory is free outside the free target. */ extern unsigned int vm_page_free_target; SYSCTL_INT(_vm, OID_AUTO, vm_page_free_target, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_free_target, 0, "Pageout daemon free target"); SYSCTL_INT(_vm, OID_AUTO, memory_pressure, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_state.vm_memory_pressure, 0, "Memory pressure indicator"); static int vm_ctl_page_free_wanted SYSCTL_HANDLER_ARGS { #pragma unused(oidp, arg1, arg2) unsigned int page_free_wanted; page_free_wanted = mach_vm_ctl_page_free_wanted(); return SYSCTL_OUT(req, &page_free_wanted, sizeof(page_free_wanted)); } SYSCTL_PROC(_vm, OID_AUTO, page_free_wanted, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, vm_ctl_page_free_wanted, "I", ""); extern unsigned int vm_page_purgeable_count; SYSCTL_INT(_vm, OID_AUTO, page_purgeable_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_purgeable_count, 0, "Purgeable page count"); extern unsigned int vm_page_purgeable_wired_count; SYSCTL_INT(_vm, OID_AUTO, page_purgeable_wired_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_purgeable_wired_count, 0, "Wired purgeable page count"); extern unsigned int vm_page_kern_lpage_count; SYSCTL_INT(_vm, OID_AUTO, kern_lpage_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_kern_lpage_count, 0, "kernel used large pages"); SCALABLE_COUNTER_DECLARE(vm_page_grab_count); SYSCTL_SCALABLE_COUNTER(_vm, pages_grabbed, vm_page_grab_count, "Total pages grabbed"); SCALABLE_COUNTER_DECLARE(vm_page_grab_count_kern); SYSCTL_SCALABLE_COUNTER(_vm, pages_grabbed_kern, vm_page_grab_count_kern, "Total pages grabbed (kernel)"); SCALABLE_COUNTER_DECLARE(vm_page_grab_count_iopl); SYSCTL_SCALABLE_COUNTER(_vm, pages_grabbed_iopl, vm_page_grab_count_iopl, "Total pages grabbed (iopl)"); SCALABLE_COUNTER_DECLARE(vm_page_grab_count_upl); SYSCTL_SCALABLE_COUNTER(_vm, pages_grabbed_upl, vm_page_grab_count_upl, "Total pages grabbed (upl)"); #if DEVELOPMENT || DEBUG SCALABLE_COUNTER_DECLARE(vm_page_deactivate_behind_count); SYSCTL_SCALABLE_COUNTER(_vm, pages_deactivated_behind, vm_page_deactivate_behind_count, "Number of pages deactivated behind"); #endif #if DEVELOPMENT || DEBUG #if __ARM_MIXED_PAGE_SIZE__ static int vm_mixed_pagesize_supported = 1; #else static int vm_mixed_pagesize_supported = 0; #endif /*__ARM_MIXED_PAGE_SIZE__ */ SYSCTL_INT(_debug, OID_AUTO, vm_mixed_pagesize_supported, CTLFLAG_ANYBODY | CTLFLAG_RD | CTLFLAG_LOCKED, &vm_mixed_pagesize_supported, 0, "kernel support for mixed pagesize"); SYSCTL_ULONG(_vm, OID_AUTO, pages_freed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_page_pages_freed, "Total pages freed"); SYSCTL_INT(_vm, OID_AUTO, pageout_purged_objects, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_purged_objects, 0, "System purged object count"); SYSCTL_UINT(_vm, OID_AUTO, pageout_cleaned_busy, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_cleaned_busy, 0, "Cleaned pages busy (deactivated)"); SYSCTL_UINT(_vm, OID_AUTO, pageout_cleaned_nolock, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_cleaned_nolock, 0, "Cleaned pages no-lock (deactivated)"); SYSCTL_UINT(_vm, OID_AUTO, pageout_cleaned_volatile_reactivated, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_cleaned_volatile_reactivated, 0, "Cleaned pages volatile reactivated"); SYSCTL_UINT(_vm, OID_AUTO, pageout_cleaned_fault_reactivated, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_cleaned_fault_reactivated, 0, "Cleaned pages fault reactivated"); SYSCTL_UINT(_vm, OID_AUTO, pageout_cleaned_reactivated, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_cleaned_reactivated, 0, "Cleaned pages reactivated"); /* sum of all reactivated AND busy and nolock (even though those actually get reDEactivated */ SYSCTL_ULONG(_vm, OID_AUTO, pageout_cleaned, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_freed_cleaned, "Cleaned pages freed"); SYSCTL_UINT(_vm, OID_AUTO, pageout_cleaned_reference_reactivated, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_cleaned_reference_reactivated, 0, "Cleaned pages reference reactivated"); SYSCTL_UINT(_vm, OID_AUTO, pageout_enqueued_cleaned, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_debug.vm_pageout_enqueued_cleaned, 0, ""); /* sum of next two */ #endif /* DEVELOPMENT || DEBUG */ extern int madvise_free_debug; SYSCTL_INT(_vm, OID_AUTO, madvise_free_debug, CTLFLAG_RW | CTLFLAG_LOCKED, &madvise_free_debug, 0, "zero-fill on madvise(MADV_FREE*)"); extern int madvise_free_debug_sometimes; SYSCTL_INT(_vm, OID_AUTO, madvise_free_debug_sometimes, CTLFLAG_RW | CTLFLAG_LOCKED, &madvise_free_debug_sometimes, 0, "sometimes zero-fill on madvise(MADV_FREE*)"); SYSCTL_INT(_vm, OID_AUTO, page_reusable_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_count, 0, "Reusable page count"); SYSCTL_QUAD(_vm, OID_AUTO, reusable_success, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_pages_success, ""); SYSCTL_QUAD(_vm, OID_AUTO, reusable_failure, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_pages_failure, ""); SYSCTL_QUAD(_vm, OID_AUTO, reusable_pages_shared, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_pages_shared, ""); SYSCTL_QUAD(_vm, OID_AUTO, all_reusable_calls, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.all_reusable_calls, ""); SYSCTL_QUAD(_vm, OID_AUTO, partial_reusable_calls, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.partial_reusable_calls, ""); SYSCTL_QUAD(_vm, OID_AUTO, reuse_success, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reuse_pages_success, ""); SYSCTL_QUAD(_vm, OID_AUTO, reuse_failure, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reuse_pages_failure, ""); SYSCTL_QUAD(_vm, OID_AUTO, all_reuse_calls, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.all_reuse_calls, ""); SYSCTL_QUAD(_vm, OID_AUTO, partial_reuse_calls, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.partial_reuse_calls, ""); SYSCTL_QUAD(_vm, OID_AUTO, can_reuse_success, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.can_reuse_success, ""); SYSCTL_QUAD(_vm, OID_AUTO, can_reuse_failure, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.can_reuse_failure, ""); SYSCTL_QUAD(_vm, OID_AUTO, reusable_reclaimed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_reclaimed, ""); SYSCTL_QUAD(_vm, OID_AUTO, reusable_nonwritable, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_nonwritable, ""); SYSCTL_QUAD(_vm, OID_AUTO, reusable_shared, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.reusable_shared, ""); SYSCTL_QUAD(_vm, OID_AUTO, free_shared, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_stats_reusable.free_shared, ""); extern unsigned int vm_page_free_count, vm_page_speculative_count; SYSCTL_UINT(_vm, OID_AUTO, page_free_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_free_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_speculative_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_speculative_count, 0, ""); extern unsigned int vm_page_cleaned_count; SYSCTL_UINT(_vm, OID_AUTO, page_cleaned_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_cleaned_count, 0, "Cleaned queue size"); extern unsigned int vm_page_pageable_internal_count, vm_page_pageable_external_count; SYSCTL_UINT(_vm, OID_AUTO, page_pageable_internal_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_pageable_internal_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_pageable_external_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_pageable_external_count, 0, ""); /* pageout counts */ SYSCTL_UINT(_vm, OID_AUTO, pageout_inactive_clean, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_state.vm_pageout_inactive_clean, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, pageout_inactive_used, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_state.vm_pageout_inactive_used, 0, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_inactive_dirty_internal, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_inactive_dirty_internal, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_inactive_dirty_external, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_inactive_dirty_external, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_speculative_clean, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_freed_speculative, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_freed_external, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_freed_external, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_freed_speculative, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_freed_speculative, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_freed_cleaned, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_freed_cleaned, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_protected_sharedcache, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_protected_sharedcache, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_forcereclaimed_sharedcache, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_forcereclaimed_sharedcache, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_protected_realtime, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_protected_realtime, ""); SYSCTL_ULONG(_vm, OID_AUTO, pageout_forcereclaimed_realtime, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_pageout_forcereclaimed_realtime, ""); extern unsigned int vm_page_realtime_count; SYSCTL_UINT(_vm, OID_AUTO, page_realtime_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_realtime_count, 0, ""); extern int vm_pageout_protect_realtime; SYSCTL_INT(_vm, OID_AUTO, pageout_protect_realtime, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_pageout_protect_realtime, 0, ""); /* counts of pages prefaulted when entering a memory object */ extern int64_t vm_prefault_nb_pages, vm_prefault_nb_bailout; extern int64_t vm_prefault_nb_no_page, vm_prefault_nb_wrong_page; SYSCTL_QUAD(_vm, OID_AUTO, prefault_nb_pages, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_prefault_nb_pages, ""); SYSCTL_QUAD(_vm, OID_AUTO, prefault_nb_bailout, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_prefault_nb_bailout, ""); SYSCTL_QUAD(_vm, OID_AUTO, prefault_nb_no_page, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_prefault_nb_no_page, ""); SYSCTL_QUAD(_vm, OID_AUTO, prefault_nb_wrong_page, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_prefault_nb_wrong_page, ""); #if defined (__x86_64__) extern unsigned int vm_clump_promote_threshold; SYSCTL_UINT(_vm, OID_AUTO, vm_clump_promote_threshold, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_clump_promote_threshold, 0, "clump size threshold for promotes"); #if DEVELOPMENT || DEBUG extern unsigned long vm_clump_stats[]; SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats1, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[1], "free page allocations from clump of 1 page"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats2, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[2], "free page allocations from clump of 2 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats3, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[3], "free page allocations from clump of 3 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats4, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[4], "free page allocations from clump of 4 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats5, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[5], "free page allocations from clump of 5 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats6, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[6], "free page allocations from clump of 6 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats7, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[7], "free page allocations from clump of 7 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats8, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[8], "free page allocations from clump of 8 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats9, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[9], "free page allocations from clump of 9 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats10, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[10], "free page allocations from clump of 10 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats11, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[11], "free page allocations from clump of 11 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats12, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[12], "free page allocations from clump of 12 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats13, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[13], "free page allocations from clump of 13 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats14, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[14], "free page allocations from clump of 14 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats15, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[15], "free page allocations from clump of 15 pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_stats16, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_stats[16], "free page allocations from clump of 16 pages"); extern unsigned long vm_clump_allocs, vm_clump_inserts, vm_clump_inrange, vm_clump_promotes; SYSCTL_LONG(_vm, OID_AUTO, vm_clump_alloc, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_allocs, "free page allocations"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_inserts, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_inserts, "free page insertions"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_inrange, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_inrange, "free page insertions that are part of vm_pages"); SYSCTL_LONG(_vm, OID_AUTO, vm_clump_promotes, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_clump_promotes, "pages promoted to head"); #endif /* if DEVELOPMENT || DEBUG */ #endif /* #if defined (__x86_64__) */ #if CONFIG_SECLUDED_MEMORY SYSCTL_UINT(_vm, OID_AUTO, num_tasks_can_use_secluded_mem, CTLFLAG_RD | CTLFLAG_LOCKED, &num_tasks_can_use_secluded_mem, 0, ""); extern unsigned int vm_page_secluded_target; extern unsigned int vm_page_secluded_count; extern unsigned int vm_page_secluded_count_free; extern unsigned int vm_page_secluded_count_inuse; extern unsigned int vm_page_secluded_count_over_target; SYSCTL_UINT(_vm, OID_AUTO, page_secluded_target, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded_target, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded_count, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_count_free, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded_count_free, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_count_inuse, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded_count_inuse, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_count_over_target, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded_count_over_target, 0, ""); extern struct vm_page_secluded_data vm_page_secluded; SYSCTL_UINT(_vm, OID_AUTO, page_secluded_eligible, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.eligible_for_secluded, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_success_free, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_success_free, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_success_other, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_success_other, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_failure_locked, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_failure_locked, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_failure_state, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_failure_state, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_failure_realtime, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_failure_realtime, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_failure_dirty, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_failure_dirty, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_for_iokit, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_for_iokit, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, page_secluded_grab_for_iokit_success, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_secluded.grab_for_iokit_success, 0, ""); #endif /* CONFIG_SECLUDED_MEMORY */ #if CONFIG_DEFERRED_RECLAIM #pragma mark Deferred Reclaim SYSCTL_NODE(_vm, OID_AUTO, reclaim, CTLFLAG_RW | CTLFLAG_LOCKED, 0, "Deferred Memory Reclamation"); #if DEVELOPMENT || DEBUG /* * VM reclaim testing */ extern bool vm_deferred_reclamation_block_until_task_has_been_reclaimed(task_t task); static int sysctl_vm_reclaim_wait_for_pid SYSCTL_HANDLER_ARGS { int error = EINVAL, pid = 0; /* * Only send on write */ error = sysctl_handle_int(oidp, &pid, 0, req); if (error || !req->newptr) { return error; } if (pid <= 0) { return EINVAL; } proc_t p = proc_find(pid); if (p == PROC_NULL) { return ESRCH; } task_t t = proc_task(p); if (t == TASK_NULL) { proc_rele(p); return ESRCH; } task_reference(t); proc_rele(p); bool success = vm_deferred_reclamation_block_until_task_has_been_reclaimed(t); if (success) { error = 0; } task_deallocate(t); return error; } SYSCTL_PROC(_vm_reclaim, OID_AUTO, wait_for_pid, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_reclaim_wait_for_pid, "I", "Block until the given pid has been drained by kernel GC"); static int sysctl_vm_reclaim_drain_pid SYSCTL_HANDLER_ARGS { int error = EINVAL; kern_return_t kr; pid_t pid; error = sysctl_handle_int(oidp, &pid, 0, req); /* Only reclaim on write */ if (error || !req->newptr) { return error; } if (pid <= 0) { return EINVAL; } proc_t p = proc_find(pid); if (p == PROC_NULL) { return ESRCH; } task_t t = proc_task(p); if (t == TASK_NULL) { proc_rele(p); return ESRCH; } task_reference(t); proc_rele(p); kr = vm_deferred_reclamation_task_drain(t, RECLAIM_OPTIONS_NONE); task_deallocate(t); return mach_to_bsd_errno(kr); } SYSCTL_PROC(_vm_reclaim, OID_AUTO, drain_pid, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_reclaim_drain_pid, "I", "Drain the deferred reclamation buffer for a pid"); static int proc_filter_reclaimable(proc_t p, __unused void *arg) { task_t task = proc_task(p); return vm_deferred_reclamation_task_has_ring(task); } static int proc_reclaim_drain(proc_t p, __unused void *arg) { kern_return_t kr; task_t task = proc_task(p); kr = vm_deferred_reclamation_task_drain(task, RECLAIM_OPTIONS_NONE); return mach_to_bsd_errno(kr); } static int sysctl_vm_reclaim_drain_all SYSCTL_HANDLER_ARGS { int error; int val; if (!req->newptr) { return EINVAL; } error = sysctl_handle_int(oidp, &val, 0, req); if (error || val == FALSE) { return error; } proc_iterate(PROC_ALLPROCLIST, proc_reclaim_drain, NULL, proc_filter_reclaimable, NULL); return 0; } SYSCTL_PROC(_vm_reclaim, OID_AUTO, drain_all, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_reclaim_drain_all, "I", "Fully reclaim from every deferred reclamation buffer on the system"); extern uint32_t vm_reclaim_buffer_count; extern uint64_t vm_reclaim_gc_epoch; extern uint64_t vm_reclaim_gc_reclaim_count; extern uint64_t vm_reclaim_sampling_period_abs; extern uint64_t vm_reclaim_sampling_period_ns; extern bool vm_reclaim_debug; extern bool vm_reclaim_enabled; extern uint32_t vm_reclaim_autotrim_pct_normal; extern uint32_t vm_reclaim_autotrim_pct_pressure; extern uint32_t vm_reclaim_autotrim_pct_critical; extern uint32_t vm_reclaim_wma_weight_base; extern uint32_t vm_reclaim_wma_weight_cur; extern uint32_t vm_reclaim_wma_denom; extern uint64_t vm_reclaim_abandonment_threshold; SYSCTL_UINT(_vm_reclaim, OID_AUTO, reclaim_buffer_count, CTLFLAG_RD | CTLFLAG_LOCKED, (uint32_t *)&vm_reclaim_buffer_count, 0, "The number of deferred memory buffers currently alive"); SYSCTL_QUAD(_vm_reclaim, OID_AUTO, reclaim_gc_epoch, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_gc_epoch, "Number of times the global GC thread has run"); SYSCTL_QUAD(_vm_reclaim, OID_AUTO, reclaim_gc_reclaim_count, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_gc_reclaim_count, "Number of times the global GC thread has reclaimed from a buffer"); SYSCTL_COMPAT_UINT(_vm_reclaim, OID_AUTO, debug, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_debug, 0, "Debug logs for vm.reclaim"); SYSCTL_COMPAT_UINT(_vm_reclaim, OID_AUTO, enabled, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_enabled, 0, "Whether deferred memory reclamation is enabled on this system"); SYSCTL_UINT(_vm_reclaim, OID_AUTO, autotrim_pct_normal, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_autotrim_pct_normal, 0, "Percentage of a task's lifetime max phys_footprint that must be reclaimable " "to engage auto-trim when the system is operating normally"); SYSCTL_UINT(_vm_reclaim, OID_AUTO, autotrim_pct_pressure, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_autotrim_pct_pressure, 0, "Percentage of a task's lifetime max phys_footprint that must be reclaimable " "to engage auto-trim when the system is under memory pressure"); SYSCTL_UINT(_vm_reclaim, OID_AUTO, autotrim_pct_critical, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_autotrim_pct_critical, 0, "Percentage of a task's lifetime max phys_footprint that must be reclaimable " "to engage auto-trim when the system is under critical memory pressure"); SYSCTL_UINT(_vm_reclaim, OID_AUTO, wma_weight_base, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_wma_weight_base, 0, "Weight applied to historical minimum buffer size samples"); SYSCTL_UINT(_vm_reclaim, OID_AUTO, wma_weight_cur, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_wma_weight_cur, 0, "Weight applied to current sampled minimum buffer size"); SYSCTL_UINT(_vm_reclaim, OID_AUTO, wma_denom, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_wma_denom, 0, "Denominator for weighted moving average calculation"); SYSCTL_QUAD(_vm_reclaim, OID_AUTO, abandonment_threshold, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_reclaim_abandonment_threshold, "The number of sampling periods between accounting updates that may elapse " "before the buffer is considered \"abandoned\""); static int sysctl_vm_reclaim_sampling_period SYSCTL_HANDLER_ARGS { uint64_t new_val_ns; uint64_t old_val_ns = vm_reclaim_sampling_period_ns; int err = sysctl_io_number(req, vm_reclaim_sampling_period_ns, sizeof(vm_reclaim_sampling_period_ns), &new_val_ns, NULL); if (err || !req->newptr) { return err; } if (new_val_ns != old_val_ns) { vm_reclaim_sampling_period_ns = new_val_ns; nanoseconds_to_absolutetime(vm_reclaim_sampling_period_ns, &vm_reclaim_sampling_period_abs); } return 0; } SYSCTL_PROC(_vm_reclaim, OID_AUTO, sampling_period_ns, CTLFLAG_RW | CTLTYPE_QUAD | CTLFLAG_LOCKED, NULL, 0, sysctl_vm_reclaim_sampling_period, "QU", "Interval (nanoseconds) at which to sample the minimum buffer size and " "consider trimming excess"); #endif /* DEVELOPMENT || DEBUG */ #endif /* CONFIG_DEFERRED_RECLAIM */ #include <kern/thread.h> #include <sys/user.h> void vm_pageout_io_throttle(void); void vm_pageout_io_throttle(void) { struct uthread *uthread = current_uthread(); /* * thread is marked as a low priority I/O type * and the I/O we issued while in this cleaning operation * collided with normal I/O operations... we'll * delay in order to mitigate the impact of this * task on the normal operation of the system */ if (uthread->uu_lowpri_window) { throttle_lowpri_io(1); } } int vm_pressure_monitor( __unused struct proc *p, struct vm_pressure_monitor_args *uap, int *retval) { kern_return_t kr; uint32_t pages_reclaimed; uint32_t pages_wanted; kr = mach_vm_pressure_monitor( (boolean_t) uap->wait_for_pressure, uap->nsecs_monitored, (uap->pages_reclaimed) ? &pages_reclaimed : NULL, &pages_wanted); switch (kr) { case KERN_SUCCESS: break; case KERN_ABORTED: return EINTR; default: return EINVAL; } if (uap->pages_reclaimed) { if (copyout((void *)&pages_reclaimed, uap->pages_reclaimed, sizeof(pages_reclaimed)) != 0) { return EFAULT; } } *retval = (int) pages_wanted; return 0; } int kas_info(struct proc *p, struct kas_info_args *uap, int *retval __unused) { #ifndef CONFIG_KAS_INFO (void)p; (void)uap; return ENOTSUP; #else /* CONFIG_KAS_INFO */ int selector = uap->selector; user_addr_t valuep = uap->value; user_addr_t sizep = uap->size; user_size_t size, rsize; int error; if (!kauth_cred_issuser(kauth_cred_get())) { return EPERM; } #if CONFIG_MACF error = mac_system_check_kas_info(kauth_cred_get(), selector); if (error) { return error; } #endif if (IS_64BIT_PROCESS(p)) { user64_size_t size64; error = copyin(sizep, &size64, sizeof(size64)); size = (user_size_t)size64; } else { user32_size_t size32; error = copyin(sizep, &size32, sizeof(size32)); size = (user_size_t)size32; } if (error) { return error; } switch (selector) { case KAS_INFO_KERNEL_TEXT_SLIDE_SELECTOR: { uint64_t slide = vm_kernel_slide; if (sizeof(slide) != size) { return EINVAL; } error = copyout(&slide, valuep, sizeof(slide)); if (error) { return error; } rsize = size; } break; case KAS_INFO_KERNEL_SEGMENT_VMADDR_SELECTOR: { uint32_t i; kernel_mach_header_t *mh = &_mh_execute_header; struct load_command *cmd; cmd = (struct load_command*) &mh[1]; uint64_t *bases; rsize = mh->ncmds * sizeof(uint64_t); /* * Return the size if no data was passed */ if (valuep == 0) { break; } if (rsize > size) { return EINVAL; } bases = kalloc_data(rsize, Z_WAITOK | Z_ZERO); for (i = 0; i < mh->ncmds; i++) { if (cmd->cmd == LC_SEGMENT_KERNEL) { __IGNORE_WCASTALIGN(kernel_segment_command_t * sg = (kernel_segment_command_t *) cmd); bases[i] = (uint64_t)sg->vmaddr; } cmd = (struct load_command *) ((uintptr_t) cmd + cmd->cmdsize); } error = copyout(bases, valuep, rsize); kfree_data(bases, rsize); if (error) { return error; } } break; case KAS_INFO_SPTM_TEXT_SLIDE_SELECTOR: case KAS_INFO_TXM_TEXT_SLIDE_SELECTOR: { #if CONFIG_SPTM const uint64_t slide = (selector == KAS_INFO_SPTM_TEXT_SLIDE_SELECTOR) ? vm_sptm_offsets.slide : vm_txm_offsets.slide; #else const uint64_t slide = 0; #endif if (sizeof(slide) != size) { return EINVAL; } error = copyout(&slide, valuep, sizeof(slide)); if (error) { return error; } rsize = size; } break; default: return EINVAL; } if (IS_64BIT_PROCESS(p)) { user64_size_t size64 = (user64_size_t)rsize; error = copyout(&size64, sizep, sizeof(size64)); } else { user32_size_t size32 = (user32_size_t)rsize; error = copyout(&size32, sizep, sizeof(size32)); } return error; #endif /* CONFIG_KAS_INFO */ } #pragma clang diagnostic push #pragma clang diagnostic ignored "-Wcast-qual" #pragma clang diagnostic ignored "-Wunused-function" static void asserts() { static_assert(sizeof(vm_min_kernel_address) == sizeof(unsigned long)); static_assert(sizeof(vm_max_kernel_address) == sizeof(unsigned long)); } SYSCTL_ULONG(_vm, OID_AUTO, vm_min_kernel_address, CTLFLAG_RD, (unsigned long *) &vm_min_kernel_address, ""); SYSCTL_ULONG(_vm, OID_AUTO, vm_max_kernel_address, CTLFLAG_RD, (unsigned long *) &vm_max_kernel_address, ""); #pragma clang diagnostic pop extern uint32_t vm_page_pages; SYSCTL_UINT(_vm, OID_AUTO, pages, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_pages, 0, ""); extern uint32_t vm_page_busy_absent_skipped; SYSCTL_UINT(_vm, OID_AUTO, page_busy_absent_skipped, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_busy_absent_skipped, 0, ""); extern uint32_t vm_page_upl_tainted; SYSCTL_UINT(_vm, OID_AUTO, upl_pages_tainted, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_upl_tainted, 0, ""); extern uint32_t vm_page_iopl_tainted; SYSCTL_UINT(_vm, OID_AUTO, iopl_pages_tainted, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_page_iopl_tainted, 0, ""); #if __arm64__ && (DEVELOPMENT || DEBUG) extern int vm_footprint_suspend_allowed; SYSCTL_INT(_vm, OID_AUTO, footprint_suspend_allowed, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_footprint_suspend_allowed, 0, ""); extern void pmap_footprint_suspend(vm_map_t map, boolean_t suspend); static int sysctl_vm_footprint_suspend SYSCTL_HANDLER_ARGS { #pragma unused(oidp, arg1, arg2) int error = 0; int new_value; if (req->newptr == USER_ADDR_NULL) { return 0; } error = SYSCTL_IN(req, &new_value, sizeof(int)); if (error) { return error; } if (!vm_footprint_suspend_allowed) { if (new_value != 0) { /* suspends are not allowed... */ return 0; } /* ... but let resumes proceed */ } DTRACE_VM2(footprint_suspend, vm_map_t, current_map(), int, new_value); pmap_footprint_suspend(current_map(), new_value); return 0; } SYSCTL_PROC(_vm, OID_AUTO, footprint_suspend, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_ANYBODY | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &sysctl_vm_footprint_suspend, "I", ""); #endif /* __arm64__ && (DEVELOPMENT || DEBUG) */ extern uint64_t vm_map_corpse_footprint_count; extern uint64_t vm_map_corpse_footprint_size_avg; extern uint64_t vm_map_corpse_footprint_size_max; extern uint64_t vm_map_corpse_footprint_full; extern uint64_t vm_map_corpse_footprint_no_buf; SYSCTL_QUAD(_vm, OID_AUTO, corpse_footprint_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_corpse_footprint_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, corpse_footprint_size_avg, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_corpse_footprint_size_avg, ""); SYSCTL_QUAD(_vm, OID_AUTO, corpse_footprint_size_max, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_corpse_footprint_size_max, ""); SYSCTL_QUAD(_vm, OID_AUTO, corpse_footprint_full, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_corpse_footprint_full, ""); SYSCTL_QUAD(_vm, OID_AUTO, corpse_footprint_no_buf, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_corpse_footprint_no_buf, ""); #if CODE_SIGNING_MONITOR extern uint64_t vm_cs_defer_to_csm; extern uint64_t vm_cs_defer_to_csm_not; SYSCTL_QUAD(_vm, OID_AUTO, cs_defer_to_csm, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_cs_defer_to_csm, ""); SYSCTL_QUAD(_vm, OID_AUTO, cs_defer_to_csm_not, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_cs_defer_to_csm_not, ""); #endif /* CODE_SIGNING_MONITOR */ extern uint64_t shared_region_pager_copied; extern uint64_t shared_region_pager_slid; extern uint64_t shared_region_pager_slid_error; extern uint64_t shared_region_pager_reclaimed; SYSCTL_QUAD(_vm, OID_AUTO, shared_region_pager_copied, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pager_copied, ""); SYSCTL_QUAD(_vm, OID_AUTO, shared_region_pager_slid, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pager_slid, ""); SYSCTL_QUAD(_vm, OID_AUTO, shared_region_pager_slid_error, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pager_slid_error, ""); SYSCTL_QUAD(_vm, OID_AUTO, shared_region_pager_reclaimed, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pager_reclaimed, ""); extern int shared_region_destroy_delay; SYSCTL_INT(_vm, OID_AUTO, shared_region_destroy_delay, CTLFLAG_RW | CTLFLAG_LOCKED, &shared_region_destroy_delay, 0, ""); #if MACH_ASSERT extern int pmap_ledgers_panic_leeway; SYSCTL_INT(_vm, OID_AUTO, pmap_ledgers_panic_leeway, CTLFLAG_RW | CTLFLAG_LOCKED, &pmap_ledgers_panic_leeway, 0, ""); #endif /* MACH_ASSERT */ extern uint64_t vm_map_lookup_and_lock_object_copy_slowly_count; extern uint64_t vm_map_lookup_and_lock_object_copy_slowly_size; extern uint64_t vm_map_lookup_and_lock_object_copy_slowly_max; extern uint64_t vm_map_lookup_and_lock_object_copy_slowly_restart; extern uint64_t vm_map_lookup_and_lock_object_copy_slowly_error; extern uint64_t vm_map_lookup_and_lock_object_copy_strategically_count; extern uint64_t vm_map_lookup_and_lock_object_copy_strategically_size; extern uint64_t vm_map_lookup_and_lock_object_copy_strategically_max; extern uint64_t vm_map_lookup_and_lock_object_copy_strategically_restart; extern uint64_t vm_map_lookup_and_lock_object_copy_strategically_error; extern uint64_t vm_map_lookup_and_lock_object_copy_shadow_count; extern uint64_t vm_map_lookup_and_lock_object_copy_shadow_size; extern uint64_t vm_map_lookup_and_lock_object_copy_shadow_max; SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_slowly_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_slowly_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_slowly_size, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_slowly_size, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_slowly_max, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_slowly_max, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_slowly_restart, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_slowly_restart, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_slowly_error, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_slowly_error, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_strategically_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_strategically_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_strategically_size, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_strategically_size, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_strategically_max, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_strategically_max, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_strategically_restart, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_strategically_restart, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_strategically_error, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_strategically_error, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_shadow_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_shadow_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_shadow_size, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_shadow_size, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_lookup_locked_copy_shadow_max, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_lookup_and_lock_object_copy_shadow_max, ""); extern int vm_protect_privileged_from_untrusted; SYSCTL_INT(_vm, OID_AUTO, protect_privileged_from_untrusted, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_protect_privileged_from_untrusted, 0, ""); extern uint64_t vm_copied_on_read; extern uint64_t vm_copied_on_read_kernel_map; extern uint64_t vm_copied_on_read_platform_map; SYSCTL_QUAD(_vm, OID_AUTO, copied_on_read, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_copied_on_read, ""); SYSCTL_QUAD(_vm, OID_AUTO, copied_on_read_kernel_map, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_copied_on_read_kernel_map, ""); SYSCTL_QUAD(_vm, OID_AUTO, copied_on_read_platform_map, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_copied_on_read_platform_map, ""); extern int vm_shared_region_count; extern int vm_shared_region_peak; SYSCTL_INT(_vm, OID_AUTO, shared_region_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_shared_region_count, 0, ""); SYSCTL_INT(_vm, OID_AUTO, shared_region_peak, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_shared_region_peak, 0, ""); #if DEVELOPMENT || DEBUG extern unsigned int shared_region_pagers_resident_count; SYSCTL_INT(_vm, OID_AUTO, shared_region_pagers_resident_count, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pagers_resident_count, 0, ""); extern unsigned int shared_region_pagers_resident_peak; SYSCTL_INT(_vm, OID_AUTO, shared_region_pagers_resident_peak, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pagers_resident_peak, 0, ""); extern int shared_region_pager_count; SYSCTL_INT(_vm, OID_AUTO, shared_region_pager_count, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_pager_count, 0, ""); #if __has_feature(ptrauth_calls) extern int shared_region_key_count; SYSCTL_INT(_vm, OID_AUTO, shared_region_key_count, CTLFLAG_RD | CTLFLAG_LOCKED, &shared_region_key_count, 0, ""); extern int vm_shared_region_reslide_count; SYSCTL_INT(_vm, OID_AUTO, shared_region_reslide_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_shared_region_reslide_count, 0, ""); #endif /* __has_feature(ptrauth_calls) */ #endif /* DEVELOPMENT || DEBUG */ #if MACH_ASSERT extern int debug4k_filter; SYSCTL_INT(_vm, OID_AUTO, debug4k_filter, CTLFLAG_RW | CTLFLAG_LOCKED, &debug4k_filter, 0, ""); extern int debug4k_panic_on_terminate; SYSCTL_INT(_vm, OID_AUTO, debug4k_panic_on_terminate, CTLFLAG_RW | CTLFLAG_LOCKED, &debug4k_panic_on_terminate, 0, ""); extern int debug4k_panic_on_exception; SYSCTL_INT(_vm, OID_AUTO, debug4k_panic_on_exception, CTLFLAG_RW | CTLFLAG_LOCKED, &debug4k_panic_on_exception, 0, ""); extern int debug4k_panic_on_misaligned_sharing; SYSCTL_INT(_vm, OID_AUTO, debug4k_panic_on_misaligned_sharing, CTLFLAG_RW | CTLFLAG_LOCKED, &debug4k_panic_on_misaligned_sharing, 0, ""); #endif /* MACH_ASSERT */ extern uint64_t vm_map_set_size_limit_count; extern uint64_t vm_map_set_data_limit_count; extern uint64_t vm_map_enter_RLIMIT_AS_count; extern uint64_t vm_map_enter_RLIMIT_DATA_count; SYSCTL_QUAD(_vm, OID_AUTO, map_set_size_limit_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_set_size_limit_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_set_data_limit_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_set_data_limit_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_enter_RLIMIT_AS_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_enter_RLIMIT_AS_count, ""); SYSCTL_QUAD(_vm, OID_AUTO, map_enter_RLIMIT_DATA_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_enter_RLIMIT_DATA_count, ""); extern uint64_t vm_fault_resilient_media_initiate; extern uint64_t vm_fault_resilient_media_retry; extern uint64_t vm_fault_resilient_media_proceed; extern uint64_t vm_fault_resilient_media_release; extern uint64_t vm_fault_resilient_media_abort1; extern uint64_t vm_fault_resilient_media_abort2; SYSCTL_QUAD(_vm, OID_AUTO, fault_resilient_media_initiate, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_initiate, ""); SYSCTL_QUAD(_vm, OID_AUTO, fault_resilient_media_retry, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_retry, ""); SYSCTL_QUAD(_vm, OID_AUTO, fault_resilient_media_proceed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_proceed, ""); SYSCTL_QUAD(_vm, OID_AUTO, fault_resilient_media_release, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_release, ""); SYSCTL_QUAD(_vm, OID_AUTO, fault_resilient_media_abort1, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_abort1, ""); SYSCTL_QUAD(_vm, OID_AUTO, fault_resilient_media_abort2, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_abort2, ""); #if MACH_ASSERT extern int vm_fault_resilient_media_inject_error1_rate; extern int vm_fault_resilient_media_inject_error1; extern int vm_fault_resilient_media_inject_error2_rate; extern int vm_fault_resilient_media_inject_error2; extern int vm_fault_resilient_media_inject_error3_rate; extern int vm_fault_resilient_media_inject_error3; SYSCTL_INT(_vm, OID_AUTO, fault_resilient_media_inject_error1_rate, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_fault_resilient_media_inject_error1_rate, 0, ""); SYSCTL_INT(_vm, OID_AUTO, fault_resilient_media_inject_error1, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_inject_error1, 0, ""); SYSCTL_INT(_vm, OID_AUTO, fault_resilient_media_inject_error2_rate, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_fault_resilient_media_inject_error2_rate, 0, ""); SYSCTL_INT(_vm, OID_AUTO, fault_resilient_media_inject_error2, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_inject_error2, 0, ""); SYSCTL_INT(_vm, OID_AUTO, fault_resilient_media_inject_error3_rate, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_fault_resilient_media_inject_error3_rate, 0, ""); SYSCTL_INT(_vm, OID_AUTO, fault_resilient_media_inject_error3, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_fault_resilient_media_inject_error3, 0, ""); #endif /* MACH_ASSERT */ extern uint64_t pmap_query_page_info_retries; SYSCTL_QUAD(_vm, OID_AUTO, pmap_query_page_info_retries, CTLFLAG_RD | CTLFLAG_LOCKED, &pmap_query_page_info_retries, ""); /* * A sysctl which causes all existing shared regions to become stale. They * will no longer be used by anything new and will be torn down as soon as * the last existing user exits. A write of non-zero value causes that to happen. * This should only be used by launchd, so we check that this is initproc. */ static int shared_region_pivot(__unused struct sysctl_oid *oidp, __unused void *arg1, __unused int arg2, struct sysctl_req *req) { unsigned int value = 0; int changed = 0; int error = sysctl_io_number(req, 0, sizeof(value), &value, &changed); if (error || !changed) { return error; } if (current_proc() != initproc) { return EPERM; } vm_shared_region_pivot(); return 0; } SYSCTL_PROC(_vm, OID_AUTO, shared_region_pivot, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_LOCKED, 0, 0, shared_region_pivot, "I", ""); extern uint64_t vm_object_shadow_forced; extern uint64_t vm_object_shadow_skipped; SYSCTL_QUAD(_vm, OID_AUTO, object_shadow_forced, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_shadow_forced, ""); SYSCTL_QUAD(_vm, OID_AUTO, object_shadow_skipped, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_shadow_skipped, ""); extern uint64_t vm_object_upl_throttle_cnt; SYSCTL_QUAD(_vm, OID_AUTO, object_upl_throttle_cnt, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_upl_throttle_cnt, "The number of times in which a UPL write was throttled due to pageout starvation"); #if HAS_MTE #pragma mark MTE SYSCTL_NODE(_vm, OID_AUTO, mte, CTLFLAG_RD | CTLFLAG_LOCKED, 0, "mte"); /* sysctls for vm.mte.* counters. */ SYSCTL_UINT(_vm_mte, OID_AUTO, tagged, CTLFLAG_RD, &vm_page_tagged_count, 0, "tagged pages in use"); SYSCTL_QUAD(_vm_mte, OID_AUTO, refill_thread_wakeups, CTLFLAG_RD, &vm_mte_refill_thread_wakeups, "the number of times the refill thread was woken up"); /* sysctls for vm.mte.free.* counters. */ SYSCTL_NODE(_vm_mte, OID_AUTO, free, CTLFLAG_RD | CTLFLAG_LOCKED, 0, "free counts"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, total, CTLFLAG_RD, &vm_page_free_count, 0, "total free pages (same as vm.page_free_count)"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, taggable, CTLFLAG_RD, &vm_page_free_taggable_count, 0, "free taggable pages in the MTE free queue"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, claimable, CTLFLAG_RD, &mte_claimable_queue.vmpfq_count, 0, "free tag storage pages on the MTE claimable queue"); SYSCTL_SCALABLE_COUNTER(_vm_mte_free, cpu_untagged, vm_cpu_free_count, "free untagged pages in CPU lists"); SYSCTL_SCALABLE_COUNTER(_vm_mte_free, cpu_claimed, vm_cpu_free_claimed_count, "free claimed pages in CPU lists"); SYSCTL_SCALABLE_COUNTER(_vm_mte_free, cpu_tagged, vm_cpu_free_tagged_count, "free tagged pages in CPU lists"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_untaggable_0, CTLFLAG_RD, &mte_free_queues[MTE_FREE_UNTAGGABLE_0].vmpfq_count, 0, "disabled/pinned/deactivating/claimed (with 16 free pages or less) tag storage pages") SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_untaggable_1, CTLFLAG_RD, &mte_free_queues[MTE_FREE_UNTAGGABLE_1].vmpfq_count, 0, "claimed (with 17 free pages or more) or disabled (with 16 pages or less) tag storage pages") SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_untaggable_2, CTLFLAG_RD, &mte_free_queues[MTE_FREE_UNTAGGABLE_2].vmpfq_count, 0, "disabled (with 17 pages or more) tag storage pages") SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_active_0, CTLFLAG_RD, &mte_free_queues[MTE_FREE_ACTIVE_0].vmpfq_count, 0, "active tag storages with free covered pages (bucket 0)"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_active_1, CTLFLAG_RD, &mte_free_queues[MTE_FREE_ACTIVE_1].vmpfq_count, 0, "active tag storages with free covered pages (bucket 1)"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_active_2, CTLFLAG_RD, &mte_free_queues[MTE_FREE_ACTIVE_2].vmpfq_count, 0, "active tag storages with free covered pages (bucket 2)"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_active_3, CTLFLAG_RD, &mte_free_queues[MTE_FREE_ACTIVE_3].vmpfq_count, 0, "active tag storages with free covered pages (bucket 3)"); SYSCTL_UINT(_vm_mte_free, OID_AUTO, tag_storage_untaggable_activating, CTLFLAG_RD, &mte_free_queues[MTE_FREE_UNTAGGABLE_ACTIVATING].vmpfq_count, 0, "activating/reclaiming tag storages with free covered pages"); /* sysctls for vm.mte.tag_storage.cell_* counters. */ SYSCTL_NODE(_vm_mte, OID_AUTO, cell, CTLFLAG_RW | CTLFLAG_LOCKED, 0, "mte cell"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, disabled, CTLFLAG_RD, &mte_info_lists[MTE_LIST_DISABLED_IDX].count, 0, "free inactive tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, disabled_recursive, CTLFLAG_RD, &vm_page_recursive_tag_storage_count, 0, "recursive tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, disabled_unmanaged, CTLFLAG_RD, &vm_page_unmanaged_tag_storage_count, 0, "unmanaged tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, retired, CTLFLAG_RD, &vm_page_retired_tag_storage_count, 0, "retired tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, pinned, CTLFLAG_RD, &mte_info_lists[MTE_LIST_PINNED_IDX].count, 0, "unreclaimable tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, deactivating, CTLFLAG_RD, &mte_info_lists[MTE_LIST_DEACTIVATING_IDX].count, 0, "deactivating tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, claimed, CTLFLAG_RD, &mte_info_lists[MTE_LIST_CLAIMED_IDX].count, 0, "claimed tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, inactive, CTLFLAG_RD, &mte_info_lists[MTE_LIST_INACTIVE_IDX].count, 0, "free inactive tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, reclaiming, CTLFLAG_RD, &mte_info_lists[MTE_LIST_RECLAIMING_IDX].count, 0, "reclaiming tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, activating, CTLFLAG_RD, &mte_info_lists[MTE_LIST_ACTIVATING_IDX].count, 0, "activating tag storage pages"); SYSCTL_UINT(_vm_mte_cell, OID_AUTO, active_0, CTLFLAG_RD, &mte_info_lists[MTE_LIST_ACTIVE_0_IDX].count, 0, "active tag storage pages with no used page tagged"); static int tag_storage_active SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) uint32_t value = mteinfo_tag_storage_active(false); return SYSCTL_OUT(req, &value, sizeof(value)); } SYSCTL_PROC(_vm_mte_cell, OID_AUTO, active, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &tag_storage_active, "I", "active tag storage pages"); /* sysctls for vm.mte.tag_storage.* counters. */ SYSCTL_NODE(_vm_mte, OID_AUTO, tag_storage, CTLFLAG_RW | CTLFLAG_LOCKED, 0, "mte tag storage"); SYSCTL_UINT(_vm_mte_tag_storage, OID_AUTO, reserved, CTLFLAG_RD, &vm_page_tag_storage_reserved, 0, "free tag storage pages reserve"); SYSCTL_UINT(_vm_mte_tag_storage, OID_AUTO, wired, CTLFLAG_RD, &vm_page_wired_tag_storage_count, 0, "wired tag storage pages"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, activations, CTLFLAG_RD, &vm_page_tag_storage_activation_count, "tag storage activations (inactive/claimed -> active)"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, deactivations, CTLFLAG_RD, &vm_page_tag_storage_deactivation_count, "tag storage deactivations (active -> inactive)"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, reclaims, CTLFLAG_RD, &vm_page_tag_storage_reclaim_success_count, "successful tag storage reclamations"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, reclaims_from_cpu, CTLFLAG_RD, &vm_page_tag_storage_reclaim_from_cpu_count, "successful tag storage reclamations from the cpu free lists"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, reclaim_failures, CTLFLAG_RD, &vm_page_tag_storage_reclaim_failure_count, "failed tag storage reclamations"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, reclaim_wired_failures, CTLFLAG_RD, &vm_page_tag_storage_reclaim_wired_failure_count, "failed tag storage reclamations due to tag storage being wired"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, wire_relocations, CTLFLAG_RD, &vm_page_tag_storage_wire_relocation_count, "tag storage relocations due to wiring"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, reclaim_compressor_failures, CTLFLAG_RD, &vm_page_tag_storage_reclaim_compressor_failure_count, "failed tag storage reclamations due to tag storage used in compressor pool"); SYSCTL_QUAD(_vm_mte_tag_storage, OID_AUTO, compressor_relocations, CTLFLAG_RD, &vm_page_tag_storage_compressor_relocation_count, "tag storage relocations due to compressor pool"); SYSCTL_UINT(_vm_mte_tag_storage, OID_AUTO, free_unmanaged, CTLFLAG_RD, &vm_page_free_unmanaged_tag_storage_count, 0, "number of free unmanaged tag storage pages"); SYSCTL_SCALABLE_COUNTER(_vm_mte_tag_storage, cpu_allocated_claimed, vm_cpu_claimed_count, "claimed tag storage pages allocated"); static int tag_storage_fragmentation SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) uint32_t value = mteinfo_tag_storage_fragmentation(false); return SYSCTL_OUT(req, &value, sizeof(value)); } SYSCTL_PROC(_vm_mte_tag_storage, OID_AUTO, fragmentation, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &tag_storage_fragmentation, "I", "the achievable the fragmentation of the tag storage space (in parts per thousand)"); static int tag_storage_fragmentation_actual SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) uint32_t value = mteinfo_tag_storage_fragmentation(true); return SYSCTL_OUT(req, &value, sizeof(value)); } SYSCTL_PROC(_vm_mte_tag_storage, OID_AUTO, fragmentation_actual, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &tag_storage_fragmentation_actual, "I", "the actual the fragmentation of the tag storage space (in parts per thousand)"); /* sysctls for vm.mte.compresor_* */ extern unsigned int vm_object_no_compressor_pager_for_mte_count; SYSCTL_INT(_vm_mte, OID_AUTO, no_compressor_pager_for_mte, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_no_compressor_pager_for_mte_count, 0, ""); /* sysctls for MTE compression stats */ SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_pages_compressed, compressor_tagged_pages_compressed, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_pages_decompressed, compressor_tagged_pages_decompressed, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_pages_freed, compressor_tagged_pages_freed, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_pages_corrupted, compressor_tagged_pages_corrupted, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_overhead_bytes, compressor_tags_overhead_bytes, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_pages, compressor_tagged_pages, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_ts_pages_used, compressor_tag_storage_pages_in_pool, "the number of tag storage pages used in the compressor"); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_non_ts_pages_used, compressor_non_tag_storage_pages_in_pool, "the number of non-tag storage pages used in the compressor"); #if DEVELOPMENT || DEBUG SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_all_zero, compressor_tags_all_zero, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_same_value, compressor_tags_same_value, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_below_align, compressor_tags_below_align, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_above_align, compressor_tags_above_align, ""); SYSCTL_SCALABLE_COUNTER(_vm_mte, compress_incompressible, compressor_tags_incompressible, ""); #endif /* DEVELOPMENT || DEBUG */ #endif /* HAS_MTE */ SYSCTL_INT(_vm, OID_AUTO, vmtc_total, CTLFLAG_RD | CTLFLAG_LOCKED, &vmtc_total, 0, "total text page corruptions detected"); #if DEBUG || DEVELOPMENT /* * A sysctl that can be used to corrupt a text page with an illegal instruction. * Used for testing text page self healing. */ extern kern_return_t vm_corrupt_text_addr(uintptr_t); static int corrupt_text_addr(__unused struct sysctl_oid *oidp, __unused void *arg1, __unused int arg2, struct sysctl_req *req) { uint64_t value = 0; int error = sysctl_handle_quad(oidp, &value, 0, req); if (error || !req->newptr) { return error; } if (vm_corrupt_text_addr((uintptr_t)value) == KERN_SUCCESS) { return 0; } else { return EINVAL; } } SYSCTL_PROC(_vm, OID_AUTO, corrupt_text_addr, CTLTYPE_QUAD | CTLFLAG_WR | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, corrupt_text_addr, "-", ""); #endif /* DEBUG || DEVELOPMENT */ #if CONFIG_MAP_RANGES /* * vm.malloc_ranges * * space-separated list of <left:right> hexadecimal addresses. */ static int vm_map_malloc_ranges SYSCTL_HANDLER_ARGS { vm_map_t map = current_map(); struct mach_vm_range r1, r2; char str[20 * 4]; int len; mach_vm_offset_t right_hole_max; if (vm_map_get_user_range(map, UMEM_RANGE_ID_DEFAULT, &r1)) { return ENOENT; } if (vm_map_get_user_range(map, UMEM_RANGE_ID_HEAP, &r2)) { return ENOENT; } #if XNU_TARGET_OS_IOS && EXTENDED_USER_VA_SUPPORT right_hole_max = MACH_VM_JUMBO_ADDRESS; #else /* !XNU_TARGET_OS_IOS || !EXTENDED_USER_VA_SUPPORT */ right_hole_max = get_map_max(map); #endif /* XNU_TARGET_OS_IOS && EXTENDED_USER_VA_SUPPORT */ len = scnprintf(str, sizeof(str), "0x%llx:0x%llx 0x%llx:0x%llx", r1.max_address, r2.min_address, r2.max_address, right_hole_max); return SYSCTL_OUT(req, str, len); } SYSCTL_PROC(_vm, OID_AUTO, malloc_ranges, CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_LOCKED | CTLFLAG_MASKED, 0, 0, &vm_map_malloc_ranges, "A", ""); #if DEBUG || DEVELOPMENT static int vm_map_user_range_default SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) struct mach_vm_range range; if (vm_map_get_user_range(current_map(), UMEM_RANGE_ID_DEFAULT, &range) != KERN_SUCCESS) { return EINVAL; } return SYSCTL_OUT(req, &range, sizeof(range)); } static int vm_map_user_range_heap SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) struct mach_vm_range range; if (vm_map_get_user_range(current_map(), UMEM_RANGE_ID_HEAP, &range) != KERN_SUCCESS) { return EINVAL; } return SYSCTL_OUT(req, &range, sizeof(range)); } static int vm_map_user_range_large_file SYSCTL_HANDLER_ARGS { #pragma unused(arg1, arg2, oidp) struct mach_vm_range range; if (vm_map_get_user_range(current_map(), UMEM_RANGE_ID_LARGE_FILE, &range) != KERN_SUCCESS) { return EINVAL; } return SYSCTL_OUT(req, &range, sizeof(range)); } /* * A sysctl that can be used to return ranges for the current VM map. * Used for testing VM ranges. */ SYSCTL_PROC(_vm, OID_AUTO, vm_map_user_range_default, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &vm_map_user_range_default, "S,mach_vm_range", ""); SYSCTL_PROC(_vm, OID_AUTO, vm_map_user_range_heap, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &vm_map_user_range_heap, "S,mach_vm_range", ""); SYSCTL_PROC(_vm, OID_AUTO, vm_map_user_range_large_file, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, &vm_map_user_range_large_file, "S,mach_vm_range", ""); #endif /* DEBUG || DEVELOPMENT */ #endif /* CONFIG_MAP_RANGES */ #if DEBUG || DEVELOPMENT #endif /* DEBUG || DEVELOPMENT */ extern uint64_t vm_map_range_overflows_count; SYSCTL_QUAD(_vm, OID_AUTO, map_range_overflows_count, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_map_range_overflows_count, ""); extern boolean_t vm_map_range_overflows_log; SYSCTL_INT(_vm, OID_AUTO, map_range_oveflows_log, CTLFLAG_RW | CTLFLAG_LOCKED, &vm_map_range_overflows_log, 0, ""); extern uint64_t c_seg_filled_no_contention; extern uint64_t c_seg_filled_contention; extern clock_sec_t c_seg_filled_contention_sec_max; extern clock_nsec_t c_seg_filled_contention_nsec_max; SYSCTL_QUAD(_vm, OID_AUTO, c_seg_filled_no_contention, CTLFLAG_RD | CTLFLAG_LOCKED, &c_seg_filled_no_contention, ""); SYSCTL_QUAD(_vm, OID_AUTO, c_seg_filled_contention, CTLFLAG_RD | CTLFLAG_LOCKED, &c_seg_filled_contention, ""); SYSCTL_ULONG(_vm, OID_AUTO, c_seg_filled_contention_sec_max, CTLFLAG_RD | CTLFLAG_LOCKED, &c_seg_filled_contention_sec_max, ""); SYSCTL_UINT(_vm, OID_AUTO, c_seg_filled_contention_nsec_max, CTLFLAG_RD | CTLFLAG_LOCKED, &c_seg_filled_contention_nsec_max, 0, ""); #if (XNU_TARGET_OS_OSX && __arm64__) extern clock_nsec_t c_process_major_report_over_ms; /* report if over ? ms */ extern int c_process_major_yield_after; /* yield after moving ? segments */ extern uint64_t c_process_major_reports; extern clock_sec_t c_process_major_max_sec; extern clock_nsec_t c_process_major_max_nsec; extern uint32_t c_process_major_peak_segcount; SYSCTL_UINT(_vm, OID_AUTO, c_process_major_report_over_ms, CTLFLAG_RW | CTLFLAG_LOCKED, &c_process_major_report_over_ms, 0, ""); SYSCTL_INT(_vm, OID_AUTO, c_process_major_yield_after, CTLFLAG_RW | CTLFLAG_LOCKED, &c_process_major_yield_after, 0, ""); SYSCTL_QUAD(_vm, OID_AUTO, c_process_major_reports, CTLFLAG_RD | CTLFLAG_LOCKED, &c_process_major_reports, ""); SYSCTL_ULONG(_vm, OID_AUTO, c_process_major_max_sec, CTLFLAG_RD | CTLFLAG_LOCKED, &c_process_major_max_sec, ""); SYSCTL_UINT(_vm, OID_AUTO, c_process_major_max_nsec, CTLFLAG_RD | CTLFLAG_LOCKED, &c_process_major_max_nsec, 0, ""); SYSCTL_UINT(_vm, OID_AUTO, c_process_major_peak_segcount, CTLFLAG_RD | CTLFLAG_LOCKED, &c_process_major_peak_segcount, 0, ""); #endif /* (XNU_TARGET_OS_OSX && __arm64__) */ #if DEVELOPMENT || DEBUG extern int panic_object_not_alive; SYSCTL_INT(_vm, OID_AUTO, panic_object_not_alive, CTLFLAG_RW | CTLFLAG_LOCKED | CTLFLAG_ANYBODY, &panic_object_not_alive, 0, ""); #endif /* DEVELOPMENT || DEBUG */ #if FBDP_DEBUG_OBJECT_NO_PAGER extern int fbdp_no_panic; SYSCTL_INT(_vm, OID_AUTO, fbdp_no_panic, CTLFLAG_RW | CTLFLAG_LOCKED | CTLFLAG_ANYBODY, &fbdp_no_panic, 0, ""); #endif /* MACH_ASSERT */ extern uint64_t cluster_direct_write_wired; SYSCTL_QUAD(_vm, OID_AUTO, cluster_direct_write_wired, CTLFLAG_RD | CTLFLAG_LOCKED, &cluster_direct_write_wired, ""); extern uint64_t vm_object_pageout_not_on_queue; extern uint64_t vm_object_pageout_not_pageable; extern uint64_t vm_object_pageout_pageable; extern uint64_t vm_object_pageout_active_local; SYSCTL_QUAD(_vm, OID_AUTO, object_pageout_not_on_queue, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_pageout_not_on_queue, ""); SYSCTL_QUAD(_vm, OID_AUTO, object_pageout_not_pageable, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_pageout_not_pageable, ""); SYSCTL_QUAD(_vm, OID_AUTO, object_pageout_pageable, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_pageout_pageable, ""); SYSCTL_QUAD(_vm, OID_AUTO, object_pageout_active_local, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_object_pageout_active_local, ""); #if DEVELOPMENT || DEBUG static uint32_t sysctl_compressor_seg_magic(vm_c_serialize_add_data_t with_data) { #if HAS_MTE if (with_data == VM_C_SERIALIZE_DATA_TAGS) { return VM_C_SEGMENT_INFO_MAGIC_WITH_TAGS; } #else #pragma unused(with_data) #endif /* HAS_MTE */ return VM_C_SEGMENT_INFO_MAGIC; } /* The largest possible single segment + its slots is * (sizeof(c_segment_info) + C_SLOT_MAX_INDEX * sizeof(c_slot_info)) + (data of a single segment) */ #define SYSCTL_SEG_BUF_SIZE (8 * 1024 + 64 * 1024) extern uint32_t c_segments_available; struct sysctl_buf_header { uint32_t magic; } __attribute__((packed)); /* This sysctl iterates over the populated c_segments and writes some info about each one and its slots. * instead of doing everything here, the function calls a function vm_compressor.c. */ static int sysctl_compressor_segments_stream(struct sysctl_req *req, vm_c_serialize_add_data_t with_data) { char* buf = kalloc_data(SYSCTL_SEG_BUF_SIZE, Z_WAITOK | Z_ZERO); if (!buf) { return ENOMEM; } size_t offset = 0; int error = 0; int segno = 0; /* 4 byte header to identify the version of the formatting of the data. * This should be incremented if c_segment_info or c_slot_info are changed */ ((struct sysctl_buf_header*)buf)->magic = sysctl_compressor_seg_magic(with_data); offset += sizeof(uint32_t); while (segno < c_segments_available) { size_t left_sz = SYSCTL_SEG_BUF_SIZE - offset; kern_return_t kr = vm_compressor_serialize_segment_debug_info(segno, buf + offset, &left_sz, with_data); if (kr == KERN_NO_SPACE) { /* failed to add another segment, push the current buffer out and try again */ if (offset == 0) { error = EINVAL; /* no space to write but I didn't write anything, shouldn't really happen */ goto out; } /* write out chunk */ error = SYSCTL_OUT(req, buf, offset); if (error) { goto out; } offset = 0; bzero(buf, SYSCTL_SEG_BUF_SIZE); /* zero any reserved bits that are not going to be filled */ /* don't increment segno, need to try again saving the current one */ } else if (kr != KERN_SUCCESS) { error = EINVAL; goto out; } else { offset += left_sz; ++segno; assert(offset <= SYSCTL_SEG_BUF_SIZE); } } if (offset > 0) { /* write last chunk */ error = SYSCTL_OUT(req, buf, offset); } out: kfree_data(buf, SYSCTL_SEG_BUF_SIZE) return error; } static int sysctl_compressor_segments(__unused struct sysctl_oid *oidp, __unused void *arg1, __unused int arg2, struct sysctl_req *req) { return sysctl_compressor_segments_stream(req, VM_C_SERIALIZE_DATA_NONE); } SYSCTL_PROC(_vm, OID_AUTO, compressor_segments, CTLTYPE_STRUCT | CTLFLAG_LOCKED | CTLFLAG_RD, 0, 0, sysctl_compressor_segments, "S", ""); #if HAS_MTE static int sysctl_compressor_segments_data(__unused struct sysctl_oid *oidp, __unused void *arg1, __unused int arg2, struct sysctl_req *req) { return sysctl_compressor_segments_stream(req, VM_C_SERIALIZE_DATA_TAGS); } SYSCTL_PROC(_vm, OID_AUTO, compressor_segments_data, CTLTYPE_STRUCT | CTLFLAG_LOCKED | CTLFLAG_RD, 0, 0, sysctl_compressor_segments_data, "S", ""); #endif /* HAS_MTE */ extern uint32_t vm_compressor_fragmentation_level(void); static int sysctl_compressor_fragmentation_level(__unused struct sysctl_oid *oidp, __unused void *arg1, __unused int arg2, struct sysctl_req *req) { uint32_t value = vm_compressor_fragmentation_level(); return SYSCTL_OUT(req, &value, sizeof(value)); } SYSCTL_PROC(_vm, OID_AUTO, compressor_fragmentation_level, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, sysctl_compressor_fragmentation_level, "IU", ""); extern uint32_t vm_compressor_incore_fragmentation_wasted_pages(void); static int sysctl_compressor_incore_fragmentation_wasted_pages(__unused struct sysctl_oid *oidp, __unused void *arg1, __unused int arg2, struct sysctl_req *req) { uint32_t value = vm_compressor_incore_fragmentation_wasted_pages(); return SYSCTL_OUT(req, &value, sizeof(value)); } SYSCTL_PROC(_vm, OID_AUTO, compressor_incore_fragmentation_wasted_pages, CTLTYPE_INT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0, sysctl_compressor_incore_fragmentation_wasted_pages, "IU", ""); #define SYSCTL_VM_OBJECTS_SLOTMAP_BUF_SIZE (8 * 1024) /* This sysctl iterates over all the entries of the vm_map of the a given process and write some info about the vm_object pointed by the entries. * This can be used for mapping where are all the pages of a process located in the compressor. */ static int sysctl_task_vm_objects_slotmap(__unused struct sysctl_oid *oidp, void *arg1, int arg2, struct sysctl_req *req) { int error = 0; char *buf = NULL; proc_t p = PROC_NULL; task_t task = TASK_NULL; vm_map_t map = VM_MAP_NULL; __block size_t offset = 0; /* go from pid to proc to task to vm_map. see sysctl_procargsx() for another example of this procession */ int *name = arg1; int namelen = arg2; if (namelen < 1) { return EINVAL; } int pid = name[0]; p = proc_find(pid); /* this increments a reference to the proc */ if (p == PROC_NULL) { return EINVAL; } task = proc_task(p); proc_rele(p); /* decrement ref of proc */ p = PROC_NULL; if (task == TASK_NULL) { return EINVAL; } /* convert proc reference to task reference */ task_reference(task); /* task reference to map reference */ map = get_task_map_reference(task); task_deallocate(task); if (map == VM_MAP_NULL) { return EINVAL; /* nothing allocated yet */ } buf = kalloc_data(SYSCTL_VM_OBJECTS_SLOTMAP_BUF_SIZE, Z_WAITOK | Z_ZERO); if (!buf) { error = ENOMEM; goto out; } /* 4 byte header to identify the version of the formatting of the data. * This should be incremented if c_segment_info or c_slot_info are changed */ ((struct sysctl_buf_header*)buf)->magic = VM_MAP_ENTRY_INFO_MAGIC; offset += sizeof(uint32_t); kern_return_t (^write_header)(int) = ^kern_return_t (int nentries) { /* write the header, happens only once at the beginning so we should have enough space */ assert(offset + sizeof(struct vm_map_info_hdr) < SYSCTL_VM_OBJECTS_SLOTMAP_BUF_SIZE); struct vm_map_info_hdr* out_hdr = (struct vm_map_info_hdr*)(buf + offset); out_hdr->vmi_nentries = nentries; offset += sizeof(struct vm_map_info_hdr); return KERN_SUCCESS; }; kern_return_t (^write_entry)(void*) = ^kern_return_t (void* entry) { while (true) { /* try up to 2 times, first try write the the current buffer, otherwise to a new buffer */ size_t left_sz = SYSCTL_VM_OBJECTS_SLOTMAP_BUF_SIZE - offset; kern_return_t kr = vm_map_dump_entry_and_compressor_pager(entry, buf + offset, &left_sz); if (kr == KERN_NO_SPACE) { /* failed to write anything, flush the current buffer and try again */ if (offset == 0) { return KERN_FAILURE; /* no space to write but I didn't write anything yet, shouldn't really happen */ } /* write out chunk */ int out_error = SYSCTL_OUT(req, buf, offset); if (out_error) { return KERN_FAILURE; } offset = 0; bzero(buf, SYSCTL_VM_OBJECTS_SLOTMAP_BUF_SIZE); /* zero any reserved bits that are not going to be filled */ continue; /* need to retry the entry dump again with the cleaned buffer */ } else if (kr != KERN_SUCCESS) { return kr; } offset += left_sz; break; } return KERN_SUCCESS; }; /* this foreach first calls to the first callback with the number of entries, then calls the second for every entry * when the buffer is exhausted, it is flushed to the sysctl and restarted */ kern_return_t kr = vm_map_entries_foreach(map, write_header, write_entry); if (kr != KERN_SUCCESS) { goto out; } if (offset > 0) { /* last chunk */ error = SYSCTL_OUT(req, buf, offset); } out: if (buf != NULL) { kfree_data(buf, SYSCTL_VM_OBJECTS_SLOTMAP_BUF_SIZE) } if (map != NULL) { vm_map_deallocate(map); } return error; } SYSCTL_PROC(_vm, OID_AUTO, task_vm_objects_slotmap, CTLTYPE_NODE | CTLFLAG_LOCKED | CTLFLAG_RD, 0, 0, sysctl_task_vm_objects_slotmap, "S", ""); #pragma mark VM Host Statistics SYSCTL_NODE(_vm, OID_AUTO, stat, CTLFLAG_RW | CTLFLAG_LOCKED, 0, "Host memory statistics"); SYSCTL_SCALABLE_COUNTER(_vm_stat, zero_fills, vm_statistics_zero_fill_count, "Pages zero-filled"); SYSCTL_SCALABLE_COUNTER(_vm_stat, reactivations, vm_statistics_reactivations, "Pages reactivated"); SYSCTL_SCALABLE_COUNTER(_vm_stat, pageins, vm_statistics_pageins, "Pages paged-in (including speculation)"); SYSCTL_SCALABLE_COUNTER(_vm_stat, pageins_requested, vm_statistics_pageins_requested, "Page-ins requested"); SYSCTL_SCALABLE_COUNTER(_vm_stat, pageins_aborted, vm_statistics_pageins_aborted, "Pages aborted during page-in"); SYSCTL_SCALABLE_COUNTER(_vm_stat, pageouts, vm_statistics_pageouts, "Pages paged-out"); SYSCTL_SCALABLE_COUNTER(_vm_stat, faults, vm_statistics_faults, "Pages faulted"); SYSCTL_SCALABLE_COUNTER(_vm_stat, cow_faults, vm_statistics_cow_faults, "Pages faulted due to copy-on-write"); SYSCTL_SCALABLE_COUNTER(_vm_stat, obj_cache_lookups, vm_statistics_lookups, "Pages looked up in the object-cache"); SYSCTL_SCALABLE_COUNTER(_vm_stat, obj_cache_hits, vm_statistics_hits, "Object-cache lookup hits"); SYSCTL_SCALABLE_COUNTER(_vm_stat, purges, vm_statistics_purges, "Pages purged"); SYSCTL_SCALABLE_COUNTER(_vm_stat, decompressions, vm_statistics_decompressions, "Pages decompressed"); SYSCTL_SCALABLE_COUNTER(_vm_stat, compressions, vm_statistics_compressions, "Pages compressed"); SYSCTL_SCALABLE_COUNTER(_vm_stat, swapins, vm_statistics_swapins, "Pages swapped in"); SYSCTL_SCALABLE_COUNTER(_vm_stat, swapouts, vm_statistics_swapouts, "Pages swapped out"); static int systctl_vm_reset_tag SYSCTL_HANDLER_ARGS { #pragma unused(oidp, arg1, arg2) int error; int tag; kern_return_t kr; /* Need to be root */ if (!kauth_cred_issuser(kauth_cred_get())) { return EPERM; } error = SYSCTL_IN(req, &tag, sizeof(tag)); if (error) { return error; } if (tag > VM_MAX_TAG_VALUE) { return EINVAL; } kr = vm_tag_reset_peak((vm_tag_t)tag); return mach_to_bsd_errno(kr); } SYSCTL_PROC(_vm, OID_AUTO, reset_tag, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_MASKED | CTLFLAG_LOCKED, 0, 0, &systctl_vm_reset_tag, "I", ""); static int systctl_vm_reset_all_tags SYSCTL_HANDLER_ARGS { #pragma unused(oidp, arg1, arg2) /* Only reset the values if the sysctl is a write */ if (!req->newptr) { return EINVAL; } /* Need to be root */ if (!kauth_cred_issuser(kauth_cred_get())) { return EPERM; } vm_tag_reset_all_peaks(); return 0; } SYSCTL_PROC(_vm, OID_AUTO, reset_all_tags, CTLTYPE_INT | CTLFLAG_WR | CTLFLAG_MASKED | CTLFLAG_LOCKED, 0, 0, &systctl_vm_reset_all_tags, "I", ""); #endif /* DEVELOPMENT || DEBUG */ SYSCTL_NODE(_vm, OID_AUTO, compressor, CTLFLAG_RD | CTLFLAG_LOCKED, 0, "VM Compressor"); SYSCTL_INT(_vm_compressor, OID_AUTO, mode, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_compressor_mode, 0, ""); SYSCTL_INT(_vm_compressor, OID_AUTO, is_active, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_compressor_is_active, 0, ""); SYSCTL_INT(_vm_compressor, OID_AUTO, is_available, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_compressor_available, 0, ""); SYSCTL_UINT(_vm_compressor, OID_AUTO, pages_compressed, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_pages_compressed, 0, "The amount of uncompressed data stored in the compressor (in pages)"); #if CONFIG_FREEZE SYSCTL_UINT(_vm_compressor, OID_AUTO, pages_compressed_incore, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_pages_compressed_incore, 0, "The amount of uncompressed data stored in the in-core compressor (in pages)"); SYSCTL_UINT(_vm_compressor, OID_AUTO, pages_compressed_incore_late_swapout, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_pages_compressed_incore_late_swapout, 0, "The amount of uncompressed data stored in the in-core compressor and queued for swapout (in pages)"); #endif SYSCTL_UINT(_vm_compressor, OID_AUTO, pages_compressed_limit, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_pages_compressed_limit, 0, "The limit on the amount of uncompressed data the compressor will store (in pages)"); SYSCTL_NODE(_vm_compressor, OID_AUTO, segment, CTLFLAG_RD | CTLFLAG_LOCKED, 0, "VM Compressor Segment Counts"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, total, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_count, 0, "Number of allocated segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, aging, CTLFLAG_RD | CTLFLAG_LOCKED, &c_age_count, 0, "Number of aging segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swappedin_early, CTLFLAG_RD | CTLFLAG_LOCKED, &c_early_swappedin_count, 0, "Number of (early) swapped-in segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swappedin_regular, CTLFLAG_RD | CTLFLAG_LOCKED, &c_regular_swappedin_count, 0, "Number of (regular) swapped-in segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swappedin_late, CTLFLAG_RD | CTLFLAG_LOCKED, &c_late_swappedin_count, 0, "Number of (late) swapped-in segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swapout_early, CTLFLAG_RD | CTLFLAG_LOCKED, &c_early_swapout_count, 0, "Number of (early) ready-to-swap segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swapout_regular, CTLFLAG_RD | CTLFLAG_LOCKED, &c_regular_swapout_count, 0, "Number of (regular) ready-to-swap segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swapout_late, CTLFLAG_RD | CTLFLAG_LOCKED, &c_late_swapout_count, 0, "Number of (late) ready-to-swap segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swapio, CTLFLAG_RD | CTLFLAG_LOCKED, &c_swapio_count, 0, "Number of swapping-out segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swappedout, CTLFLAG_RD | CTLFLAG_LOCKED, &c_swappedout_count, 0, "Number of (non-sparse) swapped-out segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, swappedout_sparse, CTLFLAG_RD | CTLFLAG_LOCKED, &c_swappedout_sparse_count, 0, "Number of (sparse) swapped-out segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, majorcompact, CTLFLAG_RD | CTLFLAG_LOCKED, &c_major_count, 0, "Number of recently-compacted segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, minorcompact, CTLFLAG_RD | CTLFLAG_LOCKED, &c_minor_count, 0, "Number of segments queued for deferred minor compaction"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, filling, CTLFLAG_RD | CTLFLAG_LOCKED, &c_filling_count, 0, "Number of filling segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, empty, CTLFLAG_RD | CTLFLAG_LOCKED, &c_empty_count, 0, "Number of empty segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, bad, CTLFLAG_RD | CTLFLAG_LOCKED, &c_bad_count, 0, "Number of bad segments"); SYSCTL_UINT(_vm_compressor_segment, OID_AUTO, limit, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segments_limit, 0, "Limit on the number of allocated segments"); SYSCTL_NODE(_vm_compressor, OID_AUTO, svp, CTLFLAG_RD | CTLFLAG_LOCKED, 0, "VM Compressor Single-Value"); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, in_hash, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_in_hash, 0, ""); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, hash_succeeded, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_hash_succeeded, 0, ""); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, hash_failed, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_hash_failed, 0, ""); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, zval_compressions, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_zero_compressions, 0, ""); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, zval_decompressions, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_zero_decompressions, 0, ""); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, nzval_compressions, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_nonzero_compressions, 0, ""); SYSCTL_UINT(_vm_compressor_svp, OID_AUTO, nzval_decompressions, CTLFLAG_RD | CTLFLAG_LOCKED, &c_segment_svp_nonzero_decompressions, 0, ""); SYSCTL_NODE(_vm_compressor, OID_AUTO, compactor, CTLFLAG_RD | CTLFLAG_LOCKED, 0, "VM Compressor Compactor"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compactions_completed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compactions_completed, "Major compactions completed"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compactions_considered, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compactions_considered, "Major compactions considered"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compactions_bailed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compactions_bailed, "Major compactions bailed (due to contention)"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compaction_bytes_moved, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compaction_bytes_moved, "Bytes moved between segments during major compactions"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compaction_slots_moved, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compaction_slots_moved, "Slots moved between segments during major compactions"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compaction_bytes_freed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compaction_bytes_freed, "Bytes freed as a result of major compaction"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, major_compaction_segments_freed, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_major_compaction_segments_freed, "Segments freed as a result of major compaction"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, swapouts_queued, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_swapouts_queued, "The number of segments queued for swapout after a major compaction"); SYSCTL_QUAD(_vm_compressor_compactor, OID_AUTO, swapout_bytes_wasted, CTLFLAG_RD | CTLFLAG_LOCKED, &vm_pageout_vminfo.vm_compactor_swapout_bytes_wasted, "The number of unused bytes in segments queued for swapout"); |