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 | /* * Copyright (c) 2000-2020 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@ */ /* * Copyright (c) 1982, 1986, 1989, 1991, 1993 * The Regents of the University of California. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by the University of * California, Berkeley and its contributors. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * From: @(#)uipc_usrreq.c 8.3 (Berkeley) 1/4/94 */ /* * NOTICE: This file was modified by SPARTA, Inc. in 2005 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 <sys/param.h> #include <sys/systm.h> #include <sys/kernel.h> #include <sys/domain.h> #include <sys/fcntl.h> #include <sys/malloc.h> /* XXX must be before <sys/file.h> */ #include <sys/file_internal.h> #include <sys/guarded.h> #include <sys/filedesc.h> #include <sys/lock.h> #include <sys/mbuf.h> #include <sys/namei.h> #include <sys/proc_internal.h> #include <sys/kauth.h> #include <sys/protosw.h> #include <sys/socket.h> #include <sys/socketvar.h> #include <sys/stat.h> #include <sys/sysctl.h> #include <sys/un.h> #include <sys/unpcb.h> #include <sys/vnode_internal.h> #include <sys/kdebug.h> #include <sys/mcache.h> #include <kern/zalloc.h> #include <kern/locks.h> #include <kern/task.h> #if CONFIG_MACF #include <security/mac_framework.h> #endif /* CONFIG_MACF */ #include <mach/vm_param.h> /* * Maximum number of FDs that can be passed in an mbuf */ #define UIPC_MAX_CMSG_FD 512 ZONE_DECLARE(unp_zone, "unpzone", sizeof(struct unpcb), ZC_NONE); static unp_gen_t unp_gencnt; static u_int unp_count; static LCK_ATTR_DECLARE(unp_mtx_attr, 0, 0); static LCK_GRP_DECLARE(unp_mtx_grp, "unp_list"); static LCK_RW_DECLARE_ATTR(unp_list_mtx, &unp_mtx_grp, &unp_mtx_attr); static LCK_MTX_DECLARE_ATTR(unp_disconnect_lock, &unp_mtx_grp, &unp_mtx_attr); static LCK_MTX_DECLARE_ATTR(unp_connect_lock, &unp_mtx_grp, &unp_mtx_attr); static LCK_MTX_DECLARE_ATTR(uipc_lock, &unp_mtx_grp, &unp_mtx_attr); static u_int disconnect_in_progress; static struct unp_head unp_shead, unp_dhead; static int unp_defer, unp_gcing, unp_gcwait; static thread_t unp_gcthread = NULL; static LIST_HEAD(, fileglob) unp_msghead = LIST_HEAD_INITIALIZER(unp_msghead); /* * mDNSResponder tracing. When enabled, endpoints connected to * /var/run/mDNSResponder will be traced; during each send on * the traced socket, we log the PID and process name of the * sending process. We also print out a bit of info related * to the data itself; this assumes ipc_msg_hdr in dnssd_ipc.h * of mDNSResponder stays the same. */ #define MDNSRESPONDER_PATH "/var/run/mDNSResponder" static int unpst_tracemdns; /* enable tracing */ #define MDNS_IPC_MSG_HDR_VERSION_1 1 struct mdns_ipc_msg_hdr { uint32_t version; uint32_t datalen; uint32_t ipc_flags; uint32_t op; union { void *context; uint32_t u32[2]; } __attribute__((packed)); uint32_t reg_index; } __attribute__((packed)); /* * Unix communications domain. * * TODO: * SEQPACKET, RDM * rethink name space problems * need a proper out-of-band * lock pushdown */ static struct sockaddr sun_noname = { .sa_len = sizeof(sun_noname), .sa_family = AF_LOCAL, .sa_data = { 0 } }; static ino_t unp_ino; /* prototype for fake inode numbers */ static int unp_attach(struct socket *); static void unp_detach(struct unpcb *); static int unp_bind(struct unpcb *, struct sockaddr *, proc_t); static int unp_connect(struct socket *, struct sockaddr *, proc_t); static void unp_disconnect(struct unpcb *); static void unp_shutdown(struct unpcb *); static void unp_drop(struct unpcb *, int); __private_extern__ void unp_gc(void); static void unp_scan(struct mbuf *, void (*)(struct fileglob *, void *arg), void *arg); static void unp_mark(struct fileglob *, __unused void *); static void unp_discard(struct fileglob *, void *); static int unp_internalize(struct mbuf *, proc_t); static int unp_listen(struct unpcb *, proc_t); static void unpcb_to_compat(struct unpcb *, struct unpcb_compat *); static void unp_get_locks_in_order(struct socket *so, struct socket *conn_so); static void unp_get_locks_in_order(struct socket *so, struct socket *conn_so) { if (so < conn_so) { socket_lock(conn_so, 1); } else { struct unpcb *unp = sotounpcb(so); unp->unp_flags |= UNP_DONTDISCONNECT; unp->rw_thrcount++; socket_unlock(so, 0); /* Get the locks in the correct order */ socket_lock(conn_so, 1); socket_lock(so, 0); unp->rw_thrcount--; if (unp->rw_thrcount == 0) { unp->unp_flags &= ~UNP_DONTDISCONNECT; wakeup(unp); } } } static int uipc_abort(struct socket *so) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } unp_drop(unp, ECONNABORTED); unp_detach(unp); sofree(so); return 0; } static int uipc_accept(struct socket *so, struct sockaddr **nam) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } /* * Pass back name of connected socket, * if it was bound and we are still connected * (our peer may have closed already!). */ if (unp->unp_conn && unp->unp_conn->unp_addr) { *nam = dup_sockaddr((struct sockaddr *) unp->unp_conn->unp_addr, 1); } else { *nam = dup_sockaddr((struct sockaddr *)&sun_noname, 1); } return 0; } /* * Returns: 0 Success * EISCONN * unp_attach: */ static int uipc_attach(struct socket *so, __unused int proto, __unused proc_t p) { struct unpcb *unp = sotounpcb(so); if (unp != 0) { return EISCONN; } return unp_attach(so); } static int uipc_bind(struct socket *so, struct sockaddr *nam, proc_t p) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } return unp_bind(unp, nam, p); } /* * Returns: 0 Success * EINVAL * unp_connect:??? [See elsewhere in this file] */ static int uipc_connect(struct socket *so, struct sockaddr *nam, proc_t p) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } return unp_connect(so, nam, p); } /* * Returns: 0 Success * EINVAL * unp_connect2:EPROTOTYPE Protocol wrong type for socket * unp_connect2:EINVAL Invalid argument */ static int uipc_connect2(struct socket *so1, struct socket *so2) { struct unpcb *unp = sotounpcb(so1); if (unp == 0) { return EINVAL; } return unp_connect2(so1, so2); } /* control is EOPNOTSUPP */ static int uipc_detach(struct socket *so) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } LCK_MTX_ASSERT(&unp->unp_mtx, LCK_MTX_ASSERT_OWNED); unp_detach(unp); return 0; } static int uipc_disconnect(struct socket *so) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } unp_disconnect(unp); return 0; } /* * Returns: 0 Success * EINVAL */ static int uipc_listen(struct socket *so, __unused proc_t p) { struct unpcb *unp = sotounpcb(so); if (unp == 0 || unp->unp_vnode == 0) { return EINVAL; } return unp_listen(unp, p); } static int uipc_peeraddr(struct socket *so, struct sockaddr **nam) { struct unpcb *unp = sotounpcb(so); if (unp == NULL) { return EINVAL; } if (unp->unp_conn != NULL && unp->unp_conn->unp_addr != NULL) { *nam = dup_sockaddr((struct sockaddr *) unp->unp_conn->unp_addr, 1); } else { *nam = dup_sockaddr((struct sockaddr *)&sun_noname, 1); } return 0; } static int uipc_rcvd(struct socket *so, __unused int flags) { struct unpcb *unp = sotounpcb(so); struct socket *so2; if (unp == 0) { return EINVAL; } switch (so->so_type) { case SOCK_DGRAM: panic("uipc_rcvd DGRAM?"); /*NOTREACHED*/ case SOCK_STREAM: #define rcv (&so->so_rcv) #define snd (&so2->so_snd) if (unp->unp_conn == 0) { break; } so2 = unp->unp_conn->unp_socket; unp_get_locks_in_order(so, so2); /* * Adjust backpressure on sender * and wakeup any waiting to write. */ snd->sb_mbmax += unp->unp_mbcnt - rcv->sb_mbcnt; unp->unp_mbcnt = rcv->sb_mbcnt; snd->sb_hiwat += unp->unp_cc - rcv->sb_cc; unp->unp_cc = rcv->sb_cc; if (sb_notify(&so2->so_snd)) { sowakeup(so2, &so2->so_snd, so); } socket_unlock(so2, 1); #undef snd #undef rcv break; default: panic("uipc_rcvd unknown socktype"); } return 0; } /* pru_rcvoob is EOPNOTSUPP */ /* * Returns: 0 Success * EINVAL * EOPNOTSUPP * EPIPE * ENOTCONN * EISCONN * unp_internalize:EINVAL * unp_internalize:EBADF * unp_connect:EAFNOSUPPORT Address family not supported * unp_connect:EINVAL Invalid argument * unp_connect:ENOTSOCK Not a socket * unp_connect:ECONNREFUSED Connection refused * unp_connect:EISCONN Socket is connected * unp_connect:EPROTOTYPE Protocol wrong type for socket * unp_connect:??? * sbappendaddr:ENOBUFS [5th argument, contents modified] * sbappendaddr:??? [whatever a filter author chooses] */ static int uipc_send(struct socket *so, int flags, struct mbuf *m, struct sockaddr *nam, struct mbuf *control, proc_t p) { int error = 0; struct unpcb *unp = sotounpcb(so); struct socket *so2; if (unp == 0) { error = EINVAL; goto release; } if (flags & PRUS_OOB) { error = EOPNOTSUPP; goto release; } if (control) { /* release lock to avoid deadlock (4436174) */ socket_unlock(so, 0); error = unp_internalize(control, p); socket_lock(so, 0); if (error) { goto release; } } switch (so->so_type) { case SOCK_DGRAM: { struct sockaddr *from; if (nam) { if (unp->unp_conn) { error = EISCONN; break; } error = unp_connect(so, nam, p); if (error) { so->so_state &= ~SS_ISCONNECTING; break; } } else { if (unp->unp_conn == 0) { error = ENOTCONN; break; } } so2 = unp->unp_conn->unp_socket; if (so != so2) { unp_get_locks_in_order(so, so2); } if (unp->unp_addr) { from = (struct sockaddr *)unp->unp_addr; } else { from = &sun_noname; } /* * sbappendaddr() will fail when the receiver runs out of * space; in contrast to SOCK_STREAM, we will lose messages * for the SOCK_DGRAM case when the receiver's queue overflows. * SB_UNIX on the socket buffer implies that the callee will * not free the control message, if any, because we would need * to call unp_dispose() on it. */ if (sbappendaddr(&so2->so_rcv, from, m, control, &error)) { control = NULL; if (sb_notify(&so2->so_rcv)) { sowakeup(so2, &so2->so_rcv, so); } } else if (control != NULL && error == 0) { /* A socket filter took control; don't touch it */ control = NULL; } if (so != so2) { socket_unlock(so2, 1); } m = NULL; if (nam) { unp_disconnect(unp); } break; } case SOCK_STREAM: { int didreceive = 0; #define rcv (&so2->so_rcv) #define snd (&so->so_snd) /* Connect if not connected yet. */ /* * Note: A better implementation would complain * if not equal to the peer's address. */ if ((so->so_state & SS_ISCONNECTED) == 0) { if (nam) { error = unp_connect(so, nam, p); if (error) { so->so_state &= ~SS_ISCONNECTING; break; /* XXX */ } } else { error = ENOTCONN; break; } } if (so->so_state & SS_CANTSENDMORE) { error = EPIPE; break; } if (unp->unp_conn == 0) { panic("uipc_send connected but no connection?"); } so2 = unp->unp_conn->unp_socket; unp_get_locks_in_order(so, so2); /* Check socket state again as we might have unlocked the socket * while trying to get the locks in order */ if ((so->so_state & SS_CANTSENDMORE)) { error = EPIPE; socket_unlock(so2, 1); break; } if (unp->unp_flags & UNP_TRACE_MDNS) { struct mdns_ipc_msg_hdr hdr; if (mbuf_copydata(m, 0, sizeof(hdr), &hdr) == 0 && hdr.version == ntohl(MDNS_IPC_MSG_HDR_VERSION_1)) { printf("%s[mDNSResponder] pid=%d (%s): op=0x%x\n", __func__, p->p_pid, p->p_comm, ntohl(hdr.op)); } } /* * Send to paired receive port, and then reduce send buffer * hiwater marks to maintain backpressure. Wake up readers. * SB_UNIX flag will allow new record to be appended to the * receiver's queue even when it is already full. It is * possible, however, that append might fail. In that case, * we will need to call unp_dispose() on the control message; * the callee will not free it since SB_UNIX is set. */ didreceive = control ? sbappendcontrol(rcv, m, control, &error) : sbappend(rcv, m); snd->sb_mbmax -= rcv->sb_mbcnt - unp->unp_conn->unp_mbcnt; unp->unp_conn->unp_mbcnt = rcv->sb_mbcnt; if ((int32_t)snd->sb_hiwat >= (int32_t)(rcv->sb_cc - unp->unp_conn->unp_cc)) { snd->sb_hiwat -= rcv->sb_cc - unp->unp_conn->unp_cc; } else { snd->sb_hiwat = 0; } unp->unp_conn->unp_cc = rcv->sb_cc; if (didreceive) { control = NULL; if (sb_notify(&so2->so_rcv)) { sowakeup(so2, &so2->so_rcv, so); } } else if (control != NULL && error == 0) { /* A socket filter took control; don't touch it */ control = NULL; } socket_unlock(so2, 1); m = NULL; #undef snd #undef rcv } break; default: panic("uipc_send unknown socktype"); } /* * SEND_EOF is equivalent to a SEND followed by * a SHUTDOWN. */ if (flags & PRUS_EOF) { socantsendmore(so); unp_shutdown(unp); } if (control && error != 0) { socket_unlock(so, 0); unp_dispose(control); socket_lock(so, 0); } release: if (control) { m_freem(control); } if (m) { m_freem(m); } return error; } static int uipc_sense(struct socket *so, void *ub, int isstat64) { struct unpcb *unp = sotounpcb(so); struct socket *so2; blksize_t blksize; if (unp == 0) { return EINVAL; } blksize = so->so_snd.sb_hiwat; if (so->so_type == SOCK_STREAM && unp->unp_conn != 0) { so2 = unp->unp_conn->unp_socket; blksize += so2->so_rcv.sb_cc; } if (unp->unp_ino == 0) { unp->unp_ino = unp_ino++; } if (isstat64 != 0) { struct stat64 *sb64; sb64 = (struct stat64 *)ub; sb64->st_blksize = blksize; sb64->st_dev = NODEV; sb64->st_ino = (ino64_t)unp->unp_ino; } else { struct stat *sb; sb = (struct stat *)ub; sb->st_blksize = blksize; sb->st_dev = NODEV; sb->st_ino = (ino_t)(uintptr_t)unp->unp_ino; } return 0; } /* * Returns: 0 Success * EINVAL * * Notes: This is not strictly correct, as unp_shutdown() also calls * socantrcvmore(). These should maybe both be conditionalized * on the 'how' argument in soshutdown() as called from the * shutdown() system call. */ static int uipc_shutdown(struct socket *so) { struct unpcb *unp = sotounpcb(so); if (unp == 0) { return EINVAL; } socantsendmore(so); unp_shutdown(unp); return 0; } /* * Returns: 0 Success * EINVAL Invalid argument */ static int uipc_sockaddr(struct socket *so, struct sockaddr **nam) { struct unpcb *unp = sotounpcb(so); if (unp == NULL) { return EINVAL; } if (unp->unp_addr != NULL) { *nam = dup_sockaddr((struct sockaddr *)unp->unp_addr, 1); } else { *nam = dup_sockaddr((struct sockaddr *)&sun_noname, 1); } return 0; } struct pr_usrreqs uipc_usrreqs = { .pru_abort = uipc_abort, .pru_accept = uipc_accept, .pru_attach = uipc_attach, .pru_bind = uipc_bind, .pru_connect = uipc_connect, .pru_connect2 = uipc_connect2, .pru_detach = uipc_detach, .pru_disconnect = uipc_disconnect, .pru_listen = uipc_listen, .pru_peeraddr = uipc_peeraddr, .pru_rcvd = uipc_rcvd, .pru_send = uipc_send, .pru_sense = uipc_sense, .pru_shutdown = uipc_shutdown, .pru_sockaddr = uipc_sockaddr, .pru_sosend = sosend, .pru_soreceive = soreceive, }; int uipc_ctloutput(struct socket *so, struct sockopt *sopt) { struct unpcb *unp = sotounpcb(so); int error = 0; pid_t peerpid; proc_t p; task_t t; struct socket *peerso; switch (sopt->sopt_dir) { case SOPT_GET: switch (sopt->sopt_name) { case LOCAL_PEERCRED: if (unp->unp_flags & UNP_HAVEPC) { error = sooptcopyout(sopt, &unp->unp_peercred, sizeof(unp->unp_peercred)); } else { if (so->so_type == SOCK_STREAM) { error = ENOTCONN; } else { error = EINVAL; } } break; case LOCAL_PEERPID: case LOCAL_PEEREPID: if (unp->unp_conn == NULL) { error = ENOTCONN; break; } peerso = unp->unp_conn->unp_socket; if (peerso == NULL) { panic("peer is connected but has no socket?"); } unp_get_locks_in_order(so, peerso); if (sopt->sopt_name == LOCAL_PEEREPID && peerso->so_flags & SOF_DELEGATED) { peerpid = peerso->e_pid; } else { peerpid = peerso->last_pid; } socket_unlock(peerso, 1); error = sooptcopyout(sopt, &peerpid, sizeof(peerpid)); break; case LOCAL_PEERUUID: case LOCAL_PEEREUUID: if (unp->unp_conn == NULL) { error = ENOTCONN; break; } peerso = unp->unp_conn->unp_socket; if (peerso == NULL) { panic("peer is connected but has no socket?"); } unp_get_locks_in_order(so, peerso); if (sopt->sopt_name == LOCAL_PEEREUUID && peerso->so_flags & SOF_DELEGATED) { error = sooptcopyout(sopt, &peerso->e_uuid, sizeof(peerso->e_uuid)); } else { error = sooptcopyout(sopt, &peerso->last_uuid, sizeof(peerso->last_uuid)); } socket_unlock(peerso, 1); break; case LOCAL_PEERTOKEN: if (unp->unp_conn == NULL) { error = ENOTCONN; break; } peerso = unp->unp_conn->unp_socket; if (peerso == NULL) { panic("peer is connected but has no socket?"); } unp_get_locks_in_order(so, peerso); peerpid = peerso->last_pid; p = proc_find(peerpid); if (p != PROC_NULL) { t = proc_task(p); if (t != TASK_NULL) { audit_token_t peertoken; mach_msg_type_number_t count = TASK_AUDIT_TOKEN_COUNT; if (task_info(t, TASK_AUDIT_TOKEN, (task_info_t)&peertoken, &count) == KERN_SUCCESS) { error = sooptcopyout(sopt, &peertoken, sizeof(peertoken)); } else { error = EINVAL; } } else { error = EINVAL; } proc_rele(p); } else { error = EINVAL; } socket_unlock(peerso, 1); break; default: error = EOPNOTSUPP; break; } break; case SOPT_SET: default: error = EOPNOTSUPP; break; } return error; } /* * Both send and receive buffers are allocated PIPSIZ bytes of buffering * for stream sockets, although the total for sender and receiver is * actually only PIPSIZ. * Datagram sockets really use the sendspace as the maximum datagram size, * and don't really want to reserve the sendspace. Their recvspace should * be large enough for at least one max-size datagram plus address. */ #ifndef PIPSIZ #define PIPSIZ 8192 #endif static u_int32_t unpst_sendspace = PIPSIZ; static u_int32_t unpst_recvspace = PIPSIZ; static u_int32_t unpdg_sendspace = 2 * 1024; /* really max datagram size */ static u_int32_t unpdg_recvspace = 4 * 1024; static int unp_rights; /* file descriptors in flight */ static int unp_disposed; /* discarded file descriptors */ SYSCTL_DECL(_net_local_stream); SYSCTL_INT(_net_local_stream, OID_AUTO, sendspace, CTLFLAG_RW | CTLFLAG_LOCKED, &unpst_sendspace, 0, ""); SYSCTL_INT(_net_local_stream, OID_AUTO, recvspace, CTLFLAG_RW | CTLFLAG_LOCKED, &unpst_recvspace, 0, ""); SYSCTL_INT(_net_local_stream, OID_AUTO, tracemdns, CTLFLAG_RW | CTLFLAG_LOCKED, &unpst_tracemdns, 0, ""); SYSCTL_DECL(_net_local_dgram); SYSCTL_INT(_net_local_dgram, OID_AUTO, maxdgram, CTLFLAG_RW | CTLFLAG_LOCKED, &unpdg_sendspace, 0, ""); SYSCTL_INT(_net_local_dgram, OID_AUTO, recvspace, CTLFLAG_RW | CTLFLAG_LOCKED, &unpdg_recvspace, 0, ""); SYSCTL_DECL(_net_local); SYSCTL_INT(_net_local, OID_AUTO, inflight, CTLFLAG_RD | CTLFLAG_LOCKED, &unp_rights, 0, ""); /* * Returns: 0 Success * ENOBUFS * soreserve:ENOBUFS */ static int unp_attach(struct socket *so) { struct unpcb *unp; int error = 0; if (so->so_snd.sb_hiwat == 0 || so->so_rcv.sb_hiwat == 0) { switch (so->so_type) { case SOCK_STREAM: error = soreserve(so, unpst_sendspace, unpst_recvspace); break; case SOCK_DGRAM: error = soreserve(so, unpdg_sendspace, unpdg_recvspace); break; default: panic("unp_attach"); } if (error) { return error; } } unp = (struct unpcb *)zalloc(unp_zone); if (unp == NULL) { return ENOBUFS; } bzero(unp, sizeof(*unp)); lck_mtx_init(&unp->unp_mtx, &unp_mtx_grp, &unp_mtx_attr); lck_rw_lock_exclusive(&unp_list_mtx); LIST_INIT(&unp->unp_refs); unp->unp_socket = so; unp->unp_gencnt = ++unp_gencnt; unp_count++; LIST_INSERT_HEAD(so->so_type == SOCK_DGRAM ? &unp_dhead : &unp_shead, unp, unp_link); lck_rw_done(&unp_list_mtx); so->so_pcb = (caddr_t)unp; /* * Mark AF_UNIX socket buffers accordingly so that: * * a. In the SOCK_STREAM case, socket buffer append won't fail due to * the lack of space; this essentially loosens the sbspace() check, * since there is disconnect between sosend() and uipc_send() with * respect to flow control that might result in our dropping the * data in uipc_send(). By setting this, we allow for slightly * more records to be appended to the receiving socket to avoid * losing data (which we can't afford in the SOCK_STREAM case). * Flow control still takes place since we adjust the sender's * hiwat during each send. This doesn't affect the SOCK_DGRAM * case and append would still fail when the queue overflows. * * b. In the presence of control messages containing internalized * file descriptors, the append routines will not free them since * we'd need to undo the work first via unp_dispose(). */ so->so_rcv.sb_flags |= SB_UNIX; so->so_snd.sb_flags |= SB_UNIX; return 0; } static void unp_detach(struct unpcb *unp) { int so_locked = 1; lck_rw_lock_exclusive(&unp_list_mtx); LIST_REMOVE(unp, unp_link); --unp_count; ++unp_gencnt; lck_rw_done(&unp_list_mtx); if (unp->unp_vnode) { struct vnode *tvp = NULL; socket_unlock(unp->unp_socket, 0); /* Holding unp_connect_lock will avoid a race between * a thread closing the listening socket and a thread * connecting to it. */ lck_mtx_lock(&unp_connect_lock); socket_lock(unp->unp_socket, 0); if (unp->unp_vnode) { tvp = unp->unp_vnode; unp->unp_vnode->v_socket = NULL; unp->unp_vnode = NULL; } lck_mtx_unlock(&unp_connect_lock); if (tvp != NULL) { vnode_rele(tvp); /* drop the usecount */ } } if (unp->unp_conn) { unp_disconnect(unp); } while (unp->unp_refs.lh_first) { struct unpcb *unp2 = NULL; /* This datagram socket is connected to one or more * sockets. In order to avoid a race condition between removing * this reference and closing the connected socket, we need * to check disconnect_in_progress */ if (so_locked == 1) { socket_unlock(unp->unp_socket, 0); so_locked = 0; } lck_mtx_lock(&unp_disconnect_lock); while (disconnect_in_progress != 0) { (void)msleep((caddr_t)&disconnect_in_progress, &unp_disconnect_lock, PSOCK, "disconnect", NULL); } disconnect_in_progress = 1; lck_mtx_unlock(&unp_disconnect_lock); /* Now we are sure that any unpcb socket disconnect is not happening */ if (unp->unp_refs.lh_first != NULL) { unp2 = unp->unp_refs.lh_first; socket_lock(unp2->unp_socket, 1); } lck_mtx_lock(&unp_disconnect_lock); disconnect_in_progress = 0; wakeup(&disconnect_in_progress); lck_mtx_unlock(&unp_disconnect_lock); if (unp2 != NULL) { /* We already locked this socket and have a reference on it */ unp_drop(unp2, ECONNRESET); socket_unlock(unp2->unp_socket, 1); } } if (so_locked == 0) { socket_lock(unp->unp_socket, 0); so_locked = 1; } soisdisconnected(unp->unp_socket); /* makes sure we're getting dealloced */ unp->unp_socket->so_flags |= SOF_PCBCLEARING; } /* * Returns: 0 Success * EAFNOSUPPORT * EINVAL * EADDRINUSE * namei:??? [anything namei can return] * vnode_authorize:??? [anything vnode_authorize can return] * * Notes: p at this point is the current process, as this function is * only called by sobind(). */ static int unp_bind( struct unpcb *unp, struct sockaddr *nam, proc_t p) { struct sockaddr_un *soun = (struct sockaddr_un *)nam; struct vnode *vp, *dvp; struct vnode_attr va; vfs_context_t ctx = vfs_context_current(); int error, namelen; struct nameidata nd; struct socket *so = unp->unp_socket; char buf[SOCK_MAXADDRLEN]; if (nam->sa_family != 0 && nam->sa_family != AF_UNIX) { return EAFNOSUPPORT; } /* * Check if the socket is already bound to an address */ if (unp->unp_vnode != NULL) { return EINVAL; } /* * Check if the socket may have been shut down */ if ((so->so_state & (SS_CANTRCVMORE | SS_CANTSENDMORE)) == (SS_CANTRCVMORE | SS_CANTSENDMORE)) { return EINVAL; } namelen = soun->sun_len - offsetof(struct sockaddr_un, sun_path); if (namelen <= 0) { return EINVAL; } /* * Note: sun_path is not a zero terminated "C" string */ if (namelen >= SOCK_MAXADDRLEN) { return EINVAL; } bcopy(soun->sun_path, buf, namelen); buf[namelen] = 0; socket_unlock(so, 0); NDINIT(&nd, CREATE, OP_MKFIFO, FOLLOW | LOCKPARENT, UIO_SYSSPACE, CAST_USER_ADDR_T(buf), ctx); /* SHOULD BE ABLE TO ADOPT EXISTING AND wakeup() ALA FIFO's */ error = namei(&nd); if (error) { socket_lock(so, 0); return error; } dvp = nd.ni_dvp; vp = nd.ni_vp; if (vp != NULL) { /* * need to do this before the vnode_put of dvp * since we may have to release an fs_nodelock */ nameidone(&nd); vnode_put(dvp); vnode_put(vp); socket_lock(so, 0); return EADDRINUSE; } VATTR_INIT(&va); VATTR_SET(&va, va_type, VSOCK); VATTR_SET(&va, va_mode, (ACCESSPERMS & ~p->p_fd->fd_cmask)); #if CONFIG_MACF error = mac_vnode_check_create(ctx, nd.ni_dvp, &nd.ni_cnd, &va); if (error == 0) #endif /* CONFIG_MACF */ #if CONFIG_MACF_SOCKET_SUBSET error = mac_vnode_check_uipc_bind(ctx, nd.ni_dvp, &nd.ni_cnd, &va); if (error == 0) #endif /* MAC_SOCKET_SUBSET */ /* authorize before creating */ error = vnode_authorize(dvp, NULL, KAUTH_VNODE_ADD_FILE, ctx); if (!error) { /* create the socket */ error = vn_create(dvp, &vp, &nd, &va, 0, 0, NULL, ctx); } nameidone(&nd); vnode_put(dvp); if (error) { socket_lock(so, 0); return error; } socket_lock(so, 0); if (unp->unp_vnode != NULL) { vnode_put(vp); /* drop the iocount */ return EINVAL; } error = vnode_ref(vp); /* gain a longterm reference */ if (error) { vnode_put(vp); /* drop the iocount */ return error; } vp->v_socket = unp->unp_socket; unp->unp_vnode = vp; unp->unp_addr = (struct sockaddr_un *)dup_sockaddr(nam, 1); vnode_put(vp); /* drop the iocount */ return 0; } /* * Returns: 0 Success * EAFNOSUPPORT Address family not supported * EINVAL Invalid argument * ENOTSOCK Not a socket * ECONNREFUSED Connection refused * EPROTOTYPE Protocol wrong type for socket * EISCONN Socket is connected * unp_connect2:EPROTOTYPE Protocol wrong type for socket * unp_connect2:EINVAL Invalid argument * namei:??? [anything namei can return] * vnode_authorize:???? [anything vnode_authorize can return] * * Notes: p at this point is the current process, as this function is * only called by sosend(), sendfile(), and soconnectlock(). */ static int unp_connect(struct socket *so, struct sockaddr *nam, __unused proc_t p) { struct sockaddr_un *soun = (struct sockaddr_un *)nam; struct vnode *vp; struct socket *so2, *so3, *list_so = NULL; struct unpcb *unp, *unp2, *unp3; vfs_context_t ctx = vfs_context_current(); int error, len; struct nameidata nd; char buf[SOCK_MAXADDRLEN]; if (nam->sa_family != 0 && nam->sa_family != AF_UNIX) { return EAFNOSUPPORT; } unp = sotounpcb(so); so2 = so3 = NULL; len = nam->sa_len - offsetof(struct sockaddr_un, sun_path); if (len <= 0) { return EINVAL; } /* * Note: sun_path is not a zero terminated "C" string */ if (len >= SOCK_MAXADDRLEN) { return EINVAL; } soisconnecting(so); bcopy(soun->sun_path, buf, len); buf[len] = 0; socket_unlock(so, 0); NDINIT(&nd, LOOKUP, OP_LOOKUP, FOLLOW | LOCKLEAF, UIO_SYSSPACE, CAST_USER_ADDR_T(buf), ctx); error = namei(&nd); if (error) { socket_lock(so, 0); return error; } nameidone(&nd); vp = nd.ni_vp; if (vp->v_type != VSOCK) { error = ENOTSOCK; socket_lock(so, 0); goto out; } #if CONFIG_MACF_SOCKET_SUBSET error = mac_vnode_check_uipc_connect(ctx, vp, so); if (error) { socket_lock(so, 0); goto out; } #endif /* MAC_SOCKET_SUBSET */ error = vnode_authorize(vp, NULL, KAUTH_VNODE_WRITE_DATA, ctx); if (error) { socket_lock(so, 0); goto out; } lck_mtx_lock(&unp_connect_lock); if (vp->v_socket == 0) { lck_mtx_unlock(&unp_connect_lock); error = ECONNREFUSED; socket_lock(so, 0); goto out; } socket_lock(vp->v_socket, 1); /* Get a reference on the listening socket */ so2 = vp->v_socket; lck_mtx_unlock(&unp_connect_lock); if (so2->so_pcb == NULL) { error = ECONNREFUSED; if (so != so2) { socket_unlock(so2, 1); socket_lock(so, 0); } else { /* Release the reference held for the listen socket */ VERIFY(so2->so_usecount > 0); so2->so_usecount--; } goto out; } if (so < so2) { socket_unlock(so2, 0); socket_lock(so, 0); socket_lock(so2, 0); } else if (so > so2) { socket_lock(so, 0); } /* * Check if socket was connected while we were trying to * get the socket locks in order. * XXX - probably shouldn't return an error for SOCK_DGRAM */ if ((so->so_state & SS_ISCONNECTED) != 0) { error = EISCONN; goto decref_out; } if (so->so_type != so2->so_type) { error = EPROTOTYPE; goto decref_out; } if (so->so_proto->pr_flags & PR_CONNREQUIRED) { /* Release the incoming socket but keep a reference */ socket_unlock(so, 0); if ((so2->so_options & SO_ACCEPTCONN) == 0 || (so3 = sonewconn(so2, 0, nam)) == 0) { error = ECONNREFUSED; if (so != so2) { socket_unlock(so2, 1); socket_lock(so, 0); } else { socket_lock(so, 0); /* Release the reference held for * listen socket. */ VERIFY(so2->so_usecount > 0); so2->so_usecount--; } goto out; } unp2 = sotounpcb(so2); unp3 = sotounpcb(so3); if (unp2->unp_addr) { unp3->unp_addr = (struct sockaddr_un *) dup_sockaddr((struct sockaddr *)unp2->unp_addr, 1); } /* * unp_peercred management: * * The connecter's (client's) credentials are copied * from its process structure at the time of connect() * (which is now). */ cru2x(vfs_context_ucred(ctx), &unp3->unp_peercred); unp3->unp_flags |= UNP_HAVEPC; /* * The receiver's (server's) credentials are copied * from the unp_peercred member of socket on which the * former called listen(); unp_listen() cached that * process's credentials at that time so we can use * them now. */ KASSERT(unp2->unp_flags & UNP_HAVEPCCACHED, ("unp_connect: listener without cached peercred")); /* Here we need to have both so and so2 locks and so2 * is already locked. Lock ordering is required. */ if (so < so2) { socket_unlock(so2, 0); socket_lock(so, 0); socket_lock(so2, 0); } else { socket_lock(so, 0); } /* Check again if the socket state changed when its lock was released */ if ((so->so_state & SS_ISCONNECTED) != 0) { error = EISCONN; socket_unlock(so2, 1); socket_lock(so3, 0); sofreelastref(so3, 1); goto out; } memcpy(&unp->unp_peercred, &unp2->unp_peercred, sizeof(unp->unp_peercred)); unp->unp_flags |= UNP_HAVEPC; /* Hold the reference on listening socket until the end */ socket_unlock(so2, 0); list_so = so2; /* Lock ordering doesn't matter because so3 was just created */ socket_lock(so3, 1); so2 = so3; /* * Enable tracing for mDNSResponder endpoints. (The use * of sizeof instead of strlen below takes the null * terminating character into account.) */ if (unpst_tracemdns && !strncmp(soun->sun_path, MDNSRESPONDER_PATH, sizeof(MDNSRESPONDER_PATH))) { unp->unp_flags |= UNP_TRACE_MDNS; unp2->unp_flags |= UNP_TRACE_MDNS; } } error = unp_connect2(so, so2); decref_out: if (so2 != NULL) { if (so != so2) { socket_unlock(so2, 1); } else { /* Release the extra reference held for the listen socket. * This is possible only for SOCK_DGRAM sockets. We refuse * connecting to the same socket for SOCK_STREAM sockets. */ VERIFY(so2->so_usecount > 0); so2->so_usecount--; } } if (list_so != NULL) { socket_lock(list_so, 0); socket_unlock(list_so, 1); } out: LCK_MTX_ASSERT(&unp->unp_mtx, LCK_MTX_ASSERT_OWNED); vnode_put(vp); return error; } /* * Returns: 0 Success * EPROTOTYPE Protocol wrong type for socket * EINVAL Invalid argument */ int unp_connect2(struct socket *so, struct socket *so2) { struct unpcb *unp = sotounpcb(so); struct unpcb *unp2; if (so2->so_type != so->so_type) { return EPROTOTYPE; } unp2 = sotounpcb(so2); LCK_MTX_ASSERT(&unp->unp_mtx, LCK_MTX_ASSERT_OWNED); LCK_MTX_ASSERT(&unp2->unp_mtx, LCK_MTX_ASSERT_OWNED); /* Verify both sockets are still opened */ if (unp == 0 || unp2 == 0) { return EINVAL; } unp->unp_conn = unp2; so2->so_usecount++; switch (so->so_type) { case SOCK_DGRAM: LIST_INSERT_HEAD(&unp2->unp_refs, unp, unp_reflink); if (so != so2) { /* Avoid lock order reversals due to drop/acquire in soisconnected. */ /* Keep an extra reference on so2 that will be dropped * soon after getting the locks in order */ socket_unlock(so2, 0); soisconnected(so); unp_get_locks_in_order(so, so2); VERIFY(so2->so_usecount > 0); so2->so_usecount--; } else { soisconnected(so); } break; case SOCK_STREAM: /* This takes care of socketpair */ if (!(unp->unp_flags & UNP_HAVEPC) && !(unp2->unp_flags & UNP_HAVEPC)) { cru2x(kauth_cred_get(), &unp->unp_peercred); unp->unp_flags |= UNP_HAVEPC; cru2x(kauth_cred_get(), &unp2->unp_peercred); unp2->unp_flags |= UNP_HAVEPC; } unp2->unp_conn = unp; so->so_usecount++; /* Avoid lock order reversals due to drop/acquire in soisconnected. */ socket_unlock(so, 0); soisconnected(so2); /* Keep an extra reference on so2, that will be dropped soon after * getting the locks in order again. */ socket_unlock(so2, 0); socket_lock(so, 0); soisconnected(so); unp_get_locks_in_order(so, so2); /* Decrement the extra reference left before */ VERIFY(so2->so_usecount > 0); so2->so_usecount--; break; default: panic("unknown socket type %d in unp_connect2", so->so_type); } LCK_MTX_ASSERT(&unp->unp_mtx, LCK_MTX_ASSERT_OWNED); LCK_MTX_ASSERT(&unp2->unp_mtx, LCK_MTX_ASSERT_OWNED); return 0; } static void unp_disconnect(struct unpcb *unp) { struct unpcb *unp2 = NULL; struct socket *so2 = NULL, *so; struct socket *waitso; int so_locked = 1, strdisconn = 0; so = unp->unp_socket; if (unp->unp_conn == NULL) { return; } lck_mtx_lock(&unp_disconnect_lock); while (disconnect_in_progress != 0) { if (so_locked == 1) { socket_unlock(so, 0); so_locked = 0; } (void)msleep((caddr_t)&disconnect_in_progress, &unp_disconnect_lock, PSOCK, "disconnect", NULL); } disconnect_in_progress = 1; lck_mtx_unlock(&unp_disconnect_lock); if (so_locked == 0) { socket_lock(so, 0); so_locked = 1; } unp2 = unp->unp_conn; if (unp2 == 0 || unp2->unp_socket == NULL) { goto out; } so2 = unp2->unp_socket; try_again: if (so == so2) { if (so_locked == 0) { socket_lock(so, 0); } waitso = so; } else if (so < so2) { if (so_locked == 0) { socket_lock(so, 0); } socket_lock(so2, 1); waitso = so2; } else { if (so_locked == 1) { socket_unlock(so, 0); } socket_lock(so2, 1); socket_lock(so, 0); waitso = so; } so_locked = 1; LCK_MTX_ASSERT(&unp->unp_mtx, LCK_MTX_ASSERT_OWNED); LCK_MTX_ASSERT(&unp2->unp_mtx, LCK_MTX_ASSERT_OWNED); /* Check for the UNP_DONTDISCONNECT flag, if it * is set, release both sockets and go to sleep */ if ((((struct unpcb *)waitso->so_pcb)->unp_flags & UNP_DONTDISCONNECT) != 0) { if (so != so2) { socket_unlock(so2, 1); } so_locked = 0; (void)msleep(waitso->so_pcb, &unp->unp_mtx, PSOCK | PDROP, "unpdisconnect", NULL); goto try_again; } if (unp->unp_conn == NULL) { panic("unp_conn became NULL after sleep"); } unp->unp_conn = NULL; VERIFY(so2->so_usecount > 0); so2->so_usecount--; if (unp->unp_flags & UNP_TRACE_MDNS) { unp->unp_flags &= ~UNP_TRACE_MDNS; } switch (unp->unp_socket->so_type) { case SOCK_DGRAM: LIST_REMOVE(unp, unp_reflink); unp->unp_socket->so_state &= ~SS_ISCONNECTED; if (so != so2) { socket_unlock(so2, 1); } break; case SOCK_STREAM: unp2->unp_conn = NULL; VERIFY(so->so_usecount > 0); so->so_usecount--; /* Set the socket state correctly but do a wakeup later when * we release all locks except the socket lock, this will avoid * a deadlock. */ unp->unp_socket->so_state &= ~(SS_ISCONNECTING | SS_ISCONNECTED | SS_ISDISCONNECTING); unp->unp_socket->so_state |= (SS_CANTRCVMORE | SS_CANTSENDMORE | SS_ISDISCONNECTED); unp2->unp_socket->so_state &= ~(SS_ISCONNECTING | SS_ISCONNECTED | SS_ISDISCONNECTING); unp->unp_socket->so_state |= (SS_CANTRCVMORE | SS_CANTSENDMORE | SS_ISDISCONNECTED); if (unp2->unp_flags & UNP_TRACE_MDNS) { unp2->unp_flags &= ~UNP_TRACE_MDNS; } strdisconn = 1; break; default: panic("unknown socket type %d", so->so_type); } out: lck_mtx_lock(&unp_disconnect_lock); disconnect_in_progress = 0; wakeup(&disconnect_in_progress); lck_mtx_unlock(&unp_disconnect_lock); if (strdisconn) { socket_unlock(so, 0); soisdisconnected(so2); socket_unlock(so2, 1); socket_lock(so, 0); soisdisconnected(so); } LCK_MTX_ASSERT(&unp->unp_mtx, LCK_MTX_ASSERT_OWNED); return; } /* * unpcb_to_compat copies specific bits of a unpcb to a unpcb_compat format. * The unpcb_compat data structure is passed to user space and must not change. */ static void unpcb_to_compat(struct unpcb *up, struct unpcb_compat *cp) { #if defined(__LP64__) cp->unp_link.le_next = (u_int32_t) VM_KERNEL_ADDRPERM(up->unp_link.le_next); cp->unp_link.le_prev = (u_int32_t) VM_KERNEL_ADDRPERM(up->unp_link.le_prev); #else cp->unp_link.le_next = (struct unpcb_compat *) VM_KERNEL_ADDRPERM(up->unp_link.le_next); cp->unp_link.le_prev = (struct unpcb_compat **) VM_KERNEL_ADDRPERM(up->unp_link.le_prev); #endif cp->unp_socket = (_UNPCB_PTR(struct socket *)) VM_KERNEL_ADDRPERM(up->unp_socket); cp->unp_vnode = (_UNPCB_PTR(struct vnode *)) VM_KERNEL_ADDRPERM(up->unp_vnode); cp->unp_ino = up->unp_ino; cp->unp_conn = (_UNPCB_PTR(struct unpcb_compat *)) VM_KERNEL_ADDRPERM(up->unp_conn); cp->unp_refs = (u_int32_t)VM_KERNEL_ADDRPERM(up->unp_refs.lh_first); #if defined(__LP64__) cp->unp_reflink.le_next = (u_int32_t)VM_KERNEL_ADDRPERM(up->unp_reflink.le_next); cp->unp_reflink.le_prev = (u_int32_t)VM_KERNEL_ADDRPERM(up->unp_reflink.le_prev); #else cp->unp_reflink.le_next = (struct unpcb_compat *)VM_KERNEL_ADDRPERM(up->unp_reflink.le_next); cp->unp_reflink.le_prev = (struct unpcb_compat **)VM_KERNEL_ADDRPERM(up->unp_reflink.le_prev); #endif cp->unp_addr = (_UNPCB_PTR(struct sockaddr_un *)) VM_KERNEL_ADDRPERM(up->unp_addr); cp->unp_cc = up->unp_cc; cp->unp_mbcnt = up->unp_mbcnt; cp->unp_gencnt = up->unp_gencnt; } static int unp_pcblist SYSCTL_HANDLER_ARGS { #pragma unused(oidp,arg2) int error, i, n; struct unpcb *unp, **unp_list; unp_gen_t gencnt; struct xunpgen xug; struct unp_head *head; lck_rw_lock_shared(&unp_list_mtx); head = ((intptr_t)arg1 == SOCK_DGRAM ? &unp_dhead : &unp_shead); /* * The process of preparing the PCB list is too time-consuming and * resource-intensive to repeat twice on every request. */ if (req->oldptr == USER_ADDR_NULL) { n = unp_count; req->oldidx = 2 * sizeof(xug) + (n + n / 8) * sizeof(struct xunpcb); lck_rw_done(&unp_list_mtx); return 0; } if (req->newptr != USER_ADDR_NULL) { lck_rw_done(&unp_list_mtx); return EPERM; } /* * OK, now we're committed to doing something. */ gencnt = unp_gencnt; n = unp_count; bzero(&xug, sizeof(xug)); xug.xug_len = sizeof(xug); xug.xug_count = n; xug.xug_gen = gencnt; xug.xug_sogen = so_gencnt; error = SYSCTL_OUT(req, &xug, sizeof(xug)); if (error) { lck_rw_done(&unp_list_mtx); return error; } /* * We are done if there is no pcb */ if (n == 0) { lck_rw_done(&unp_list_mtx); return 0; } size_t unp_list_len = n * sizeof(*unp_list); unp_list = kheap_alloc(KHEAP_TEMP, unp_list_len, Z_WAITOK); if (unp_list == 0) { lck_rw_done(&unp_list_mtx); return ENOMEM; } for (unp = head->lh_first, i = 0; unp && i < n; unp = unp->unp_link.le_next) { if (unp->unp_gencnt <= gencnt) { unp_list[i++] = unp; } } n = i; /* in case we lost some during malloc */ error = 0; for (i = 0; i < n; i++) { unp = unp_list[i]; if (unp->unp_gencnt <= gencnt) { struct xunpcb xu; bzero(&xu, sizeof(xu)); xu.xu_len = sizeof(xu); xu.xu_unpp = (_UNPCB_PTR(struct unpcb_compat *)) VM_KERNEL_ADDRPERM(unp); /* * XXX - need more locking here to protect against * connect/disconnect races for SMP. */ if (unp->unp_addr) { bcopy(unp->unp_addr, &xu.xu_au, unp->unp_addr->sun_len); } if (unp->unp_conn && unp->unp_conn->unp_addr) { bcopy(unp->unp_conn->unp_addr, &xu.xu_cau, unp->unp_conn->unp_addr->sun_len); } unpcb_to_compat(unp, &xu.xu_unp); sotoxsocket(unp->unp_socket, &xu.xu_socket); error = SYSCTL_OUT(req, &xu, sizeof(xu)); } } if (!error) { /* * Give the user an updated idea of our state. * If the generation differs from what we told * her before, she knows that something happened * while we were processing this request, and it * might be necessary to retry. */ bzero(&xug, sizeof(xug)); xug.xug_len = sizeof(xug); xug.xug_gen = unp_gencnt; xug.xug_sogen = so_gencnt; xug.xug_count = unp_count; error = SYSCTL_OUT(req, &xug, sizeof(xug)); } kheap_free(KHEAP_TEMP, unp_list, unp_list_len); lck_rw_done(&unp_list_mtx); return error; } SYSCTL_PROC(_net_local_dgram, OID_AUTO, pcblist, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, (caddr_t)(long)SOCK_DGRAM, 0, unp_pcblist, "S,xunpcb", "List of active local datagram sockets"); SYSCTL_PROC(_net_local_stream, OID_AUTO, pcblist, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, (caddr_t)(long)SOCK_STREAM, 0, unp_pcblist, "S,xunpcb", "List of active local stream sockets"); #if XNU_TARGET_OS_OSX static int unp_pcblist64 SYSCTL_HANDLER_ARGS { #pragma unused(oidp,arg2) int error, i, n; struct unpcb *unp, **unp_list; unp_gen_t gencnt; struct xunpgen xug; struct unp_head *head; lck_rw_lock_shared(&unp_list_mtx); head = ((intptr_t)arg1 == SOCK_DGRAM ? &unp_dhead : &unp_shead); /* * The process of preparing the PCB list is too time-consuming and * resource-intensive to repeat twice on every request. */ if (req->oldptr == USER_ADDR_NULL) { n = unp_count; req->oldidx = 2 * sizeof(xug) + (n + n / 8) * (sizeof(struct xunpcb64)); lck_rw_done(&unp_list_mtx); return 0; } if (req->newptr != USER_ADDR_NULL) { lck_rw_done(&unp_list_mtx); return EPERM; } /* * OK, now we're committed to doing something. */ gencnt = unp_gencnt; n = unp_count; bzero(&xug, sizeof(xug)); xug.xug_len = sizeof(xug); xug.xug_count = n; xug.xug_gen = gencnt; xug.xug_sogen = so_gencnt; error = SYSCTL_OUT(req, &xug, sizeof(xug)); if (error) { lck_rw_done(&unp_list_mtx); return error; } /* * We are done if there is no pcb */ if (n == 0) { lck_rw_done(&unp_list_mtx); return 0; } size_t unp_list_size = n * sizeof(*unp_list); unp_list = kheap_alloc(KHEAP_TEMP, unp_list_size, Z_WAITOK); if (unp_list == 0) { lck_rw_done(&unp_list_mtx); return ENOMEM; } for (unp = head->lh_first, i = 0; unp && i < n; unp = unp->unp_link.le_next) { if (unp->unp_gencnt <= gencnt) { unp_list[i++] = unp; } } n = i; /* in case we lost some during malloc */ error = 0; for (i = 0; i < n; i++) { unp = unp_list[i]; if (unp->unp_gencnt <= gencnt) { struct xunpcb64 xu; size_t xu_len = sizeof(struct xunpcb64); bzero(&xu, xu_len); xu.xu_len = (u_int32_t)xu_len; xu.xu_unpp = (u_int64_t)VM_KERNEL_ADDRPERM(unp); xu.xunp_link.le_next = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_link.le_next); xu.xunp_link.le_prev = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_link.le_prev); xu.xunp_socket = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_socket); xu.xunp_vnode = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_vnode); xu.xunp_ino = unp->unp_ino; xu.xunp_conn = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_conn); xu.xunp_refs = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_refs.lh_first); xu.xunp_reflink.le_next = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_reflink.le_next); xu.xunp_reflink.le_prev = (u_int64_t) VM_KERNEL_ADDRPERM(unp->unp_reflink.le_prev); xu.xunp_cc = unp->unp_cc; xu.xunp_mbcnt = unp->unp_mbcnt; xu.xunp_gencnt = unp->unp_gencnt; if (unp->unp_socket) { sotoxsocket64(unp->unp_socket, &xu.xu_socket); } /* * XXX - need more locking here to protect against * connect/disconnect races for SMP. */ if (unp->unp_addr) { bcopy(unp->unp_addr, &xu.xu_au, unp->unp_addr->sun_len); } if (unp->unp_conn && unp->unp_conn->unp_addr) { bcopy(unp->unp_conn->unp_addr, &xu.xu_cau, unp->unp_conn->unp_addr->sun_len); } error = SYSCTL_OUT(req, &xu, xu_len); } } if (!error) { /* * Give the user an updated idea of our state. * If the generation differs from what we told * her before, she knows that something happened * while we were processing this request, and it * might be necessary to retry. */ bzero(&xug, sizeof(xug)); xug.xug_len = sizeof(xug); xug.xug_gen = unp_gencnt; xug.xug_sogen = so_gencnt; xug.xug_count = unp_count; error = SYSCTL_OUT(req, &xug, sizeof(xug)); } kheap_free(KHEAP_TEMP, unp_list, unp_list_size); lck_rw_done(&unp_list_mtx); return error; } SYSCTL_PROC(_net_local_dgram, OID_AUTO, pcblist64, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, (caddr_t)(long)SOCK_DGRAM, 0, unp_pcblist64, "S,xunpcb64", "List of active local datagram sockets 64 bit"); SYSCTL_PROC(_net_local_stream, OID_AUTO, pcblist64, CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, (caddr_t)(long)SOCK_STREAM, 0, unp_pcblist64, "S,xunpcb64", "List of active local stream sockets 64 bit"); #endif /* XNU_TARGET_OS_OSX */ static void unp_shutdown(struct unpcb *unp) { struct socket *so = unp->unp_socket; struct socket *so2; if (unp->unp_socket->so_type == SOCK_STREAM && unp->unp_conn) { so2 = unp->unp_conn->unp_socket; unp_get_locks_in_order(so, so2); socantrcvmore(so2); socket_unlock(so2, 1); } } static void unp_drop(struct unpcb *unp, int errno) { struct socket *so = unp->unp_socket; so->so_error = (u_short)errno; unp_disconnect(unp); } /* always called under uipc_lock */ static void unp_gc_wait(void) { if (unp_gcthread == current_thread()) { return; } while (unp_gcing != 0) { unp_gcwait = 1; msleep(&unp_gcing, &uipc_lock, 0, "unp_gc_wait", NULL); } } /* * fg_insertuipc_mark * * Description: Mark fileglob for insertion onto message queue if needed * Also takes fileglob reference * * Parameters: fg Fileglob pointer to insert * * Returns: true, if the fileglob needs to be inserted onto msg queue * * Locks: Takes and drops fg_lock, potentially many times */ static boolean_t fg_insertuipc_mark(struct fileglob * fg) { boolean_t insert = FALSE; lck_mtx_lock_spin(&fg->fg_lock); while (fg->fg_lflags & FG_RMMSGQ) { lck_mtx_convert_spin(&fg->fg_lock); fg->fg_lflags |= FG_WRMMSGQ; msleep(&fg->fg_lflags, &fg->fg_lock, 0, "fg_insertuipc", NULL); } os_ref_retain_raw(&fg->fg_count, &f_refgrp); fg->fg_msgcount++; if (fg->fg_msgcount == 1) { fg->fg_lflags |= FG_INSMSGQ; insert = TRUE; } lck_mtx_unlock(&fg->fg_lock); return insert; } /* * fg_insertuipc * * Description: Insert marked fileglob onto message queue * * Parameters: fg Fileglob pointer to insert * * Returns: void * * Locks: Takes and drops fg_lock & uipc_lock * DO NOT call this function with proc_fdlock held as unp_gc() * can potentially try to acquire proc_fdlock, which can result * in a deadlock if this function is in unp_gc_wait(). */ static void fg_insertuipc(struct fileglob * fg) { if (fg->fg_lflags & FG_INSMSGQ) { lck_mtx_lock_spin(&uipc_lock); unp_gc_wait(); LIST_INSERT_HEAD(&unp_msghead, fg, f_msglist); lck_mtx_unlock(&uipc_lock); lck_mtx_lock(&fg->fg_lock); fg->fg_lflags &= ~FG_INSMSGQ; if (fg->fg_lflags & FG_WINSMSGQ) { fg->fg_lflags &= ~FG_WINSMSGQ; wakeup(&fg->fg_lflags); } lck_mtx_unlock(&fg->fg_lock); } } /* * fg_removeuipc_mark * * Description: Mark the fileglob for removal from message queue if needed * Also releases fileglob message queue reference * * Parameters: fg Fileglob pointer to remove * * Returns: true, if the fileglob needs to be removed from msg queue * * Locks: Takes and drops fg_lock, potentially many times */ static boolean_t fg_removeuipc_mark(struct fileglob * fg) { boolean_t remove = FALSE; lck_mtx_lock_spin(&fg->fg_lock); while (fg->fg_lflags & FG_INSMSGQ) { lck_mtx_convert_spin(&fg->fg_lock); fg->fg_lflags |= FG_WINSMSGQ; msleep(&fg->fg_lflags, &fg->fg_lock, 0, "fg_removeuipc", NULL); } fg->fg_msgcount--; if (fg->fg_msgcount == 0) { fg->fg_lflags |= FG_RMMSGQ; remove = TRUE; } lck_mtx_unlock(&fg->fg_lock); return remove; } /* * fg_removeuipc * * Description: Remove marked fileglob from message queue * * Parameters: fg Fileglob pointer to remove * * Returns: void * * Locks: Takes and drops fg_lock & uipc_lock * DO NOT call this function with proc_fdlock held as unp_gc() * can potentially try to acquire proc_fdlock, which can result * in a deadlock if this function is in unp_gc_wait(). */ static void fg_removeuipc(struct fileglob * fg) { if (fg->fg_lflags & FG_RMMSGQ) { lck_mtx_lock_spin(&uipc_lock); unp_gc_wait(); LIST_REMOVE(fg, f_msglist); lck_mtx_unlock(&uipc_lock); lck_mtx_lock(&fg->fg_lock); fg->fg_lflags &= ~FG_RMMSGQ; if (fg->fg_lflags & FG_WRMMSGQ) { fg->fg_lflags &= ~FG_WRMMSGQ; wakeup(&fg->fg_lflags); } lck_mtx_unlock(&fg->fg_lock); } } /* * Returns: 0 Success * EMSGSIZE The new fd's will not fit * ENOBUFS Cannot alloc struct fileproc */ int unp_externalize(struct mbuf *rights) { proc_t p = current_proc(); /* XXX */ int i; struct cmsghdr *cm = mtod(rights, struct cmsghdr *); struct fileglob **rp = (struct fileglob **)(cm + 1); int *fds = (int *)(cm + 1); struct fileproc *fp; struct fileproc **fileproc_l; int newfds = (cm->cmsg_len - sizeof(*cm)) / sizeof(int); int f, error = 0; fileproc_l = kheap_alloc(KHEAP_TEMP, newfds * sizeof(struct fileproc *), Z_WAITOK); if (fileproc_l == NULL) { error = ENOMEM; goto discard; } proc_fdlock(p); /* * if the new FD's will not fit, then we free them all */ if (!fdavail(p, newfds)) { proc_fdunlock(p); error = EMSGSIZE; goto discard; } /* * now change each pointer to an fd in the global table to * an integer that is the index to the local fd table entry * that we set up to point to the global one we are transferring. * XXX (1) this assumes a pointer and int are the same size, * XXX or the mbuf can hold the expansion * XXX (2) allocation failures should be non-fatal */ for (i = 0; i < newfds; i++) { if (fdalloc(p, 0, &f)) { panic("unp_externalize:fdalloc"); } fp = fileproc_alloc_init(NULL); if (fp == NULL) { panic("unp_externalize:fileproc_alloc_init"); } fp->fp_glob = rp[i]; if (fg_removeuipc_mark(rp[i])) { /* * Take an iocount on the fp for completing the * removal from the global msg queue */ os_ref_retain_locked(&fp->fp_iocount); fileproc_l[i] = fp; } else { fileproc_l[i] = NULL; } procfdtbl_releasefd(p, f, fp); fds[i] = f; } proc_fdunlock(p); for (i = 0; i < newfds; i++) { if (fileproc_l[i] != NULL) { VERIFY(fileproc_l[i]->fp_glob != NULL && (fileproc_l[i]->fp_glob->fg_lflags & FG_RMMSGQ)); VERIFY(fds[i] >= 0); fg_removeuipc(fileproc_l[i]->fp_glob); /* Drop the iocount */ fp_drop(p, fds[i], fileproc_l[i], 0); fileproc_l[i] = NULL; } if (fds[i] != 0) { (void) OSAddAtomic(-1, &unp_rights); } } discard: kheap_free(KHEAP_TEMP, fileproc_l, newfds * sizeof(struct fileproc *)); if (error) { for (i = 0; i < newfds; i++) { unp_discard(*rp, p); *rp++ = NULL; } } return error; } void unp_init(void) { _CASSERT(UIPC_MAX_CMSG_FD >= (MCLBYTES / sizeof(int))); LIST_INIT(&unp_dhead); LIST_INIT(&unp_shead); } #ifndef MIN #define MIN(a, b) (((a) < (b)) ? (a) : (b)) #endif /* * Returns: 0 Success * EINVAL * EBADF */ static int unp_internalize(struct mbuf *control, proc_t p) { struct cmsghdr *cm = mtod(control, struct cmsghdr *); int *fds; struct fileglob **rp; struct fileproc *fp; int i, error; int oldfds; uint8_t fg_ins[UIPC_MAX_CMSG_FD / 8]; /* 64bit: cmsg_len is 'uint32_t', m_len is 'long' */ if (cm->cmsg_type != SCM_RIGHTS || cm->cmsg_level != SOL_SOCKET || (socklen_t)cm->cmsg_len != (socklen_t)control->m_len) { return EINVAL; } oldfds = (cm->cmsg_len - sizeof(*cm)) / sizeof(int); bzero(fg_ins, sizeof(fg_ins)); proc_fdlock(p); fds = (int *)(cm + 1); for (i = 0; i < oldfds; i++) { struct fileproc *tmpfp; if ((tmpfp = fp_get_noref_locked(p, fds[i])) == NULL) { proc_fdunlock(p); return EBADF; } else if (!fg_sendable(tmpfp->fp_glob)) { proc_fdunlock(p); return EINVAL; } else if (fp_isguarded(tmpfp, GUARD_SOCKET_IPC)) { error = fp_guard_exception(p, fds[i], tmpfp, kGUARD_EXC_SOCKET_IPC); proc_fdunlock(p); return error; } } rp = (struct fileglob **)(cm + 1); /* On K64 we need to walk backwards because a fileglob * is twice the size of an fd * and doing them in-order would result in stomping over unprocessed fd's */ for (i = (oldfds - 1); i >= 0; i--) { fp = fp_get_noref_locked(p, fds[i]); if (fg_insertuipc_mark(fp->fp_glob)) { fg_ins[i / 8] |= 0x80 >> (i % 8); } rp[i] = fp->fp_glob; } proc_fdunlock(p); for (i = 0; i < oldfds; i++) { if (fg_ins[i / 8] & (0x80 >> (i % 8))) { VERIFY(rp[i]->fg_lflags & FG_INSMSGQ); fg_insertuipc(rp[i]); } (void) OSAddAtomic(1, &unp_rights); } return 0; } __private_extern__ void unp_gc(void) { struct fileglob *fg, *nextfg; struct socket *so; static struct fileglob **extra_ref; struct fileglob **fpp; int nunref, i; int need_gcwakeup = 0; lck_mtx_lock(&uipc_lock); if (unp_gcing) { lck_mtx_unlock(&uipc_lock); return; } unp_gcing = 1; unp_defer = 0; unp_gcthread = current_thread(); lck_mtx_unlock(&uipc_lock); /* * before going through all this, set all FDs to * be NOT defered and NOT externally accessible */ for (fg = unp_msghead.lh_first; fg != 0; fg = fg->f_msglist.le_next) { os_atomic_andnot(&fg->fg_flag, FMARK | FDEFER, relaxed); } do { for (fg = unp_msghead.lh_first; fg != 0; fg = fg->f_msglist.le_next) { lck_mtx_lock(&fg->fg_lock); /* * If the file is not open, skip it */ if (os_ref_get_count_raw(&fg->fg_count) == 0) { lck_mtx_unlock(&fg->fg_lock); continue; } /* * If we already marked it as 'defer' in a * previous pass, then try process it this time * and un-mark it */ if (fg->fg_flag & FDEFER) { os_atomic_andnot(&fg->fg_flag, FDEFER, relaxed); unp_defer--; } else { /* * if it's not defered, then check if it's * already marked.. if so skip it */ if (fg->fg_flag & FMARK) { lck_mtx_unlock(&fg->fg_lock); continue; } /* * If all references are from messages * in transit, then skip it. it's not * externally accessible. */ if (os_ref_get_count_raw(&fg->fg_count) == fg->fg_msgcount) { lck_mtx_unlock(&fg->fg_lock); continue; } /* * If it got this far then it must be * externally accessible. */ os_atomic_or(&fg->fg_flag, FMARK, relaxed); } /* * either it was defered, or it is externally * accessible and not already marked so. * Now check if it is possibly one of OUR sockets. */ if (FILEGLOB_DTYPE(fg) != DTYPE_SOCKET || (so = (struct socket *)fg->fg_data) == 0) { lck_mtx_unlock(&fg->fg_lock); continue; } if (so->so_proto->pr_domain != localdomain || (so->so_proto->pr_flags & PR_RIGHTS) == 0) { lck_mtx_unlock(&fg->fg_lock); continue; } #ifdef notdef if (so->so_rcv.sb_flags & SB_LOCK) { /* * This is problematical; it's not clear * we need to wait for the sockbuf to be * unlocked (on a uniprocessor, at least), * and it's also not clear what to do * if sbwait returns an error due to receipt * of a signal. If sbwait does return * an error, we'll go into an infinite * loop. Delete all of this for now. */ (void) sbwait(&so->so_rcv); goto restart; } #endif /* * So, Ok, it's one of our sockets and it IS externally * accessible (or was defered). Now we look * to see if we hold any file descriptors in its * message buffers. Follow those links and mark them * as accessible too. * * In case a file is passed onto itself we need to * release the file lock. */ lck_mtx_unlock(&fg->fg_lock); unp_scan(so->so_rcv.sb_mb, unp_mark, 0); } } while (unp_defer); /* * We grab an extra reference to each of the file table entries * that are not otherwise accessible and then free the rights * that are stored in messages on them. * * The bug in the orginal code is a little tricky, so I'll describe * what's wrong with it here. * * It is incorrect to simply unp_discard each entry for fg_msgcount * times -- consider the case of sockets A and B that contain * references to each other. On a last close of some other socket, * we trigger a gc since the number of outstanding rights (unp_rights) * is non-zero. If during the sweep phase the gc code un_discards, * we end up doing a (full) closef on the descriptor. A closef on A * results in the following chain. Closef calls soo_close, which * calls soclose. Soclose calls first (through the switch * uipc_usrreq) unp_detach, which re-invokes unp_gc. Unp_gc simply * returns because the previous instance had set unp_gcing, and * we return all the way back to soclose, which marks the socket * with SS_NOFDREF, and then calls sofree. Sofree calls sorflush * to free up the rights that are queued in messages on the socket A, * i.e., the reference on B. The sorflush calls via the dom_dispose * switch unp_dispose, which unp_scans with unp_discard. This second * instance of unp_discard just calls closef on B. * * Well, a similar chain occurs on B, resulting in a sorflush on B, * which results in another closef on A. Unfortunately, A is already * being closed, and the descriptor has already been marked with * SS_NOFDREF, and soclose panics at this point. * * Here, we first take an extra reference to each inaccessible * descriptor. Then, we call sorflush ourself, since we know * it is a Unix domain socket anyhow. After we destroy all the * rights carried in messages, we do a last closef to get rid * of our extra reference. This is the last close, and the * unp_detach etc will shut down the socket. * * 91/09/19, bsy@cs.cmu.edu */ size_t extra_ref_size = nfiles * sizeof(struct fileglob *); extra_ref = kheap_alloc(KHEAP_TEMP, extra_ref_size, Z_WAITOK); if (extra_ref == NULL) { goto bail; } for (nunref = 0, fg = unp_msghead.lh_first, fpp = extra_ref; fg != 0; fg = nextfg) { lck_mtx_lock(&fg->fg_lock); nextfg = fg->f_msglist.le_next; /* * If it's not open, skip it */ if (os_ref_get_count_raw(&fg->fg_count) == 0) { lck_mtx_unlock(&fg->fg_lock); continue; } /* * If all refs are from msgs, and it's not marked accessible * then it must be referenced from some unreachable cycle * of (shut-down) FDs, so include it in our * list of FDs to remove */ if (fg->fg_flag & FMARK) { lck_mtx_unlock(&fg->fg_lock); continue; } if (os_ref_get_count_raw(&fg->fg_count) == fg->fg_msgcount) { os_ref_retain_raw(&fg->fg_count, &f_refgrp); *fpp++ = fg; nunref++; } lck_mtx_unlock(&fg->fg_lock); } /* * for each FD on our hit list, do the following two things */ for (i = nunref, fpp = extra_ref; --i >= 0; ++fpp) { struct fileglob *tfg; tfg = *fpp; if (FILEGLOB_DTYPE(tfg) == DTYPE_SOCKET && tfg->fg_data != NULL) { so = (struct socket *)(tfg->fg_data); socket_lock(so, 0); sorflush(so); socket_unlock(so, 0); } } for (i = nunref, fpp = extra_ref; --i >= 0; ++fpp) { fg_drop(PROC_NULL, *fpp); } kheap_free(KHEAP_TEMP, extra_ref, extra_ref_size); bail: lck_mtx_lock(&uipc_lock); unp_gcing = 0; unp_gcthread = NULL; if (unp_gcwait != 0) { unp_gcwait = 0; need_gcwakeup = 1; } lck_mtx_unlock(&uipc_lock); if (need_gcwakeup != 0) { wakeup(&unp_gcing); } } void unp_dispose(struct mbuf *m) { if (m) { unp_scan(m, unp_discard, NULL); } } /* * Returns: 0 Success */ static int unp_listen(struct unpcb *unp, proc_t p) { kauth_cred_t safecred = kauth_cred_proc_ref(p); cru2x(safecred, &unp->unp_peercred); kauth_cred_unref(&safecred); unp->unp_flags |= UNP_HAVEPCCACHED; return 0; } static void unp_scan(struct mbuf *m0, void (*op)(struct fileglob *, void *arg), void *arg) { struct mbuf *m; struct fileglob **rp; struct cmsghdr *cm; int i; int qfds; while (m0) { for (m = m0; m; m = m->m_next) { if (m->m_type == MT_CONTROL && (size_t)m->m_len >= sizeof(*cm)) { cm = mtod(m, struct cmsghdr *); if (cm->cmsg_level != SOL_SOCKET || cm->cmsg_type != SCM_RIGHTS) { continue; } qfds = (cm->cmsg_len - sizeof(*cm)) / sizeof(int); rp = (struct fileglob **)(cm + 1); for (i = 0; i < qfds; i++) { (*op)(*rp++, arg); } break; /* XXX, but saves time */ } } m0 = m0->m_act; } } static void unp_mark(struct fileglob *fg, __unused void *arg) { uint32_t oflags, nflags; os_atomic_rmw_loop(&fg->fg_flag, oflags, nflags, relaxed, { if (oflags & FMARK) { os_atomic_rmw_loop_give_up(return ); } nflags = oflags | FMARK | FDEFER; }); unp_defer++; } static void unp_discard(struct fileglob *fg, void *p) { if (p == NULL) { p = current_proc(); /* XXX */ } (void) OSAddAtomic(1, &unp_disposed); if (fg_removeuipc_mark(fg)) { VERIFY(fg->fg_lflags & FG_RMMSGQ); fg_removeuipc(fg); } (void) OSAddAtomic(-1, &unp_rights); (void) fg_drop(p, fg); } int unp_lock(struct socket *so, int refcount, void * lr) { void * lr_saved; if (lr == 0) { lr_saved = (void *) __builtin_return_address(0); } else { lr_saved = lr; } if (so->so_pcb) { lck_mtx_lock(&((struct unpcb *)so->so_pcb)->unp_mtx); } else { panic("unp_lock: so=%p NO PCB! lr=%p ref=0x%x\n", so, lr_saved, so->so_usecount); } if (so->so_usecount < 0) { panic("unp_lock: so=%p so_pcb=%p lr=%p ref=0x%x\n", so, so->so_pcb, lr_saved, so->so_usecount); } if (refcount) { VERIFY(so->so_usecount > 0); so->so_usecount++; } so->lock_lr[so->next_lock_lr] = lr_saved; so->next_lock_lr = (so->next_lock_lr + 1) % SO_LCKDBG_MAX; return 0; } int unp_unlock(struct socket *so, int refcount, void * lr) { void * lr_saved; lck_mtx_t * mutex_held = NULL; struct unpcb *unp = sotounpcb(so); if (lr == 0) { lr_saved = (void *) __builtin_return_address(0); } else { lr_saved = lr; } if (refcount) { so->so_usecount--; } if (so->so_usecount < 0) { panic("unp_unlock: so=%p usecount=%x\n", so, so->so_usecount); } if (so->so_pcb == NULL) { panic("unp_unlock: so=%p NO PCB usecount=%x\n", so, so->so_usecount); } else { mutex_held = &((struct unpcb *)so->so_pcb)->unp_mtx; } LCK_MTX_ASSERT(mutex_held, LCK_MTX_ASSERT_OWNED); so->unlock_lr[so->next_unlock_lr] = lr_saved; so->next_unlock_lr = (so->next_unlock_lr + 1) % SO_LCKDBG_MAX; if (so->so_usecount == 0 && (so->so_flags & SOF_PCBCLEARING)) { sofreelastref(so, 1); if (unp->unp_addr) { FREE(unp->unp_addr, M_SONAME); } lck_mtx_unlock(mutex_held); lck_mtx_destroy(&unp->unp_mtx, &unp_mtx_grp); zfree(unp_zone, unp); unp_gc(); } else { lck_mtx_unlock(mutex_held); } return 0; } lck_mtx_t * unp_getlock(struct socket *so, __unused int flags) { struct unpcb *unp = (struct unpcb *)so->so_pcb; if (so->so_pcb) { if (so->so_usecount < 0) { panic("unp_getlock: so=%p usecount=%x\n", so, so->so_usecount); } return &unp->unp_mtx; } else { panic("unp_getlock: so=%p NULL so_pcb\n", so); return so->so_proto->pr_domain->dom_mtx; } } |