/* * Mesa 3-D graphics library * Version: 6.5.1 * * Copyright (C) 1999-2006 Brian Paul All Rights Reserved. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), * to deal in the Software without restriction, including without limitation * the rights to use, copy, modify, merge, publish, distribute, sublicense, * and/or sell copies of the Software, and to permit persons to whom the * Software is furnished to do so, subject to the following conditions: * * The above copyright notice and this permission notice shall be included * in all copies or substantial portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ /* * XXX There's probably some work to do in order to make this file * truly reusable outside of Mesa. First, the glheader.h include must go. */ #include "glheader.h" #include "glthread.h" /* * This file should still compile even when THREADS is not defined. * This is to make things easier to deal with on the makefile scene.. */ #ifdef THREADS #include /* * Error messages */ #define INIT_TSD_ERROR "_glthread_: failed to allocate key for thread specific data" #define GET_TSD_ERROR "_glthread_: failed to get thread specific data" #define SET_TSD_ERROR "_glthread_: thread failed to set thread specific data" /* * Magic number to determine if a TSD object has been initialized. * Kind of a hack but there doesn't appear to be a better cross-platform * solution. */ #define INIT_MAGIC 0xff8adc98 /* * POSIX Threads -- The best way to go if your platform supports them. * Solaris >= 2.5 have POSIX threads, IRIX >= 6.4 reportedly * has them, and many of the free Unixes now have them. * Be sure to use appropriate -mt or -D_REENTRANT type * compile flags when building. */ #ifdef PTHREADS unsigned long _glthread_GetID(void) { return (unsigned long) pthread_self(); } void _glthread_InitTSD(_glthread_TSD *tsd) { if (pthread_key_create(&tsd->key, NULL/*free*/) != 0) { perror(INIT_TSD_ERROR); exit(-1); } tsd->initMagic = INIT_MAGIC; } void * _glthread_GetTSD(_glthread_TSD *tsd) { if (tsd->initMagic != (int) INIT_MAGIC) { _glthread_InitTSD(tsd); } return pthread_getspecific(tsd->key); } void _glthread_SetTSD(_glthread_TSD *tsd, void *ptr) { if (tsd->initMagic != (int) INIT_MAGIC) { _glthread_InitTSD(tsd); } if (pthread_setspecific(tsd->key, ptr) != 0) { perror(SET_TSD_ERROR); exit(-1); } } #endif /* PTHREADS */ /* * Solaris/Unix International Threads -- Use only if POSIX threads * aren't available on your Unix platform. Solaris 2.[34] are examples * of platforms where this is the case. Be sure to use -mt and/or * -D_REENTRANT when compiling. */ #ifdef SOLARIS_THREADS #define USE_LOCK_FOR_KEY /* undef this to try a version without lock for the global key... */ unsigned long _glthread_GetID(void) { abort(); /* XXX not implemented yet */ return (unsigned long) 0; } void _glthread_InitTSD(_glthread_TSD *tsd) { if ((errno = mutex_init(&tsd->keylock, 0, NULL)) != 0 || (errno = thr_keycreate(&(tsd->key), free)) != 0) { perror(INIT_TSD_ERROR); exit(-1); } tsd->initMagic = INIT_MAGIC; } void * _glthread_GetTSD(_glthread_TSD *tsd) { void* ret; if (tsd->initMagic != INIT_MAGIC) { _glthread_InitTSD(tsd); } #ifdef USE_LOCK_FOR_KEY mutex_lock(&tsd->keylock); thr_getspecific(tsd->key, &ret); mutex_unlock(&tsd->keylock); #else if ((errno = thr_getspecific(tsd->key, &ret)) != 0) { perror(GET_TSD_ERROR); exit(-1); } #endif return ret; } void _glthread_SetTSD(_glthread_TSD *tsd, void *ptr) { if (tsd->initMagic != INIT_MAGIC) { _glthread_InitTSD(tsd); } if ((errno = thr_setspecific(tsd->key, ptr)) != 0) { perror(SET_TSD_ERROR); exit(-1); } } #undef USE_LOCK_FOR_KEY #endif /* SOLARIS_THREADS */ /* * Win32 Threads. The only available option for Windows 95/NT. * Be sure that you compile using the Multithreaded runtime, otherwise * bad things will happen. */ #ifdef WIN32_THREADS void FreeTSD(_glthread_TSD *p) { if (p->initMagic==INIT_MAGIC) { TlsFree(p->key); p->initMagic=0; } } void InsteadOf_exit(int nCode) { DWORD dwErr=GetLastError(); } unsigned long _glthread_GetID(void) { return GetCurrentThreadId(); } void _glthread_InitTSD(_glthread_TSD *tsd) { tsd->key = TlsAlloc(); if (tsd->key == TLS_OUT_OF_INDEXES) { perror("Mesa:_glthread_InitTSD"); InsteadOf_exit(-1); } tsd->initMagic = INIT_MAGIC; } void * _glthread_GetTSD(_glthread_TSD *tsd) { if (tsd->initMagic != INIT_MAGIC) { _glthread_InitTSD(tsd); } return TlsGetValue(tsd->key); } void _glthread_SetTSD(_glthread_TSD *tsd, void *ptr) { /* the following code assumes that the _glthread_TSD has been initialized to zero at creation */ if (tsd->initMagic != INIT_MAGIC) { _glthread_InitTSD(tsd); } if (TlsSetValue(tsd->key, ptr) == 0) { perror("Mesa:_glthread_SetTSD"); InsteadOf_exit(-1); } } #endif /* WIN32_THREADS */ /* * XFree86 has its own thread wrapper, Xthreads.h * We wrap it again for GL. */ #ifdef USE_XTHREADS unsigned long _glthread_GetID(void) { return (unsigned long) xthread_self(); } void _glthread_InitTSD(_glthread_TSD *tsd) { if (xthread_key_create(&tsd->key, NULL) != 0) { perror(INIT_TSD_ERROR); exit(-1); } tsd->initMagic = INIT_MAGIC; } void * _glthread_GetTSD(_glthread_TSD *tsd) { void *ptr; if (tsd->initMagic != INIT_MAGIC) { _glthread_InitTSD(tsd); } xthread_get_specific(tsd->key, &ptr); return ptr; } void _glthread_SetTSD(_glthread_TSD *tsd, void *ptr) { if (tsd->initMagic != INIT_MAGIC) { _glthread_InitTSD(tsd); } xthread_set_specific(tsd->key, ptr); } #endif /* XTHREAD */ /* * BeOS threads */ #ifdef BEOS_THREADS unsigned long _glthread_GetID(void) { return (unsigned long) find_thread(NULL); } void _glthread_InitTSD(_glthread_TSD *tsd) { tsd->key = tls_allocate(); tsd->initMagic = INIT_MAGIC; } void * _glthread_GetTSD(_glthread_TSD *tsd) { if (tsd->initMagic != (int) INIT_MAGIC) { _glthread_InitTSD(tsd); } return tls_get(tsd->key); } void _glthread_SetTSD(_glthread_TSD *tsd, void *ptr) { if (tsd->initMagic != (int) INIT_MAGIC) { _glthread_InitTSD(tsd); } tls_set(tsd->key, ptr); } #endif /* BEOS_THREADS */ #else /* THREADS */ /* * no-op functions */ unsigned long _glthread_GetID(void) { return 0; } void _glthread_InitTSD(_glthread_TSD *tsd) { (void) tsd; } void * _glthread_GetTSD(_glthread_TSD *tsd) { (void) tsd; return NULL; } void _glthread_SetTSD(_glthread_TSD *tsd, void *ptr) { (void) tsd; (void) ptr; } #endif /* THREADS */ 8'>98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 
/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
 * Copyright 2011 Nexenta Systems, Inc. All rights reserved.
 * Copyright (c) 2011, 2015 by Delphix. All rights reserved.
 * Copyright (c) 2014, Joyent, Inc. All rights reserved.
 * Copyright 2014 HybridCluster. All rights reserved.
 * Copyright 2016 RackTop Systems.
 * Copyright (c) 2016 Actifio, Inc. All rights reserved.
 */

#include <sys/dmu.h>
#include <sys/dmu_impl.h>
#include <sys/dmu_tx.h>
#include <sys/dbuf.h>
#include <sys/dnode.h>
#include <sys/zfs_context.h>
#include <sys/dmu_objset.h>
#include <sys/dmu_traverse.h>
#include <sys/dsl_dataset.h>
#include <sys/dsl_dir.h>
#include <sys/dsl_prop.h>
#include <sys/dsl_pool.h>
#include <sys/dsl_synctask.h>
#include <sys/spa_impl.h>
#include <sys/zfs_ioctl.h>
#include <sys/zap.h>
#include <sys/zio_checksum.h>
#include <sys/zfs_znode.h>
#include <zfs_fletcher.h>
#include <sys/avl.h>
#include <sys/ddt.h>
#include <sys/zfs_onexit.h>
#include <sys/dmu_send.h>
#include <sys/dsl_destroy.h>
#include <sys/blkptr.h>
#include <sys/dsl_bookmark.h>
#include <sys/zfeature.h>
#include <sys/bqueue.h>
#include <sys/zvol.h>
#include <sys/policy.h>

/* Set this tunable to TRUE to replace corrupt data with 0x2f5baddb10c */
int zfs_send_corrupt_data = B_FALSE;
int zfs_send_queue_length = SPA_MAXBLOCKSIZE;
int zfs_recv_queue_length = SPA_MAXBLOCKSIZE;
/* Set this tunable to FALSE to disable setting of DRR_FLAG_FREERECORDS */
int zfs_send_set_freerecords_bit = B_TRUE;

static char *dmu_recv_tag = "dmu_recv_tag";
const char *recv_clone_name = "%recv";

#define	BP_SPAN(datablkszsec, indblkshift, level) \
	(((uint64_t)datablkszsec) << (SPA_MINBLOCKSHIFT + \
	(level) * (indblkshift - SPA_BLKPTRSHIFT)))

static void byteswap_record(dmu_replay_record_t *drr);

struct send_thread_arg {
	bqueue_t	q;
	dsl_dataset_t	*ds;		/* Dataset to traverse */
	uint64_t	fromtxg;	/* Traverse from this txg */
	int		flags;		/* flags to pass to traverse_dataset */
	int		error_code;
	boolean_t	cancel;
	zbookmark_phys_t resume;
};

struct send_block_record {
	boolean_t		eos_marker; /* Marks the end of the stream */
	blkptr_t		bp;
	zbookmark_phys_t	zb;
	uint8_t			indblkshift;
	uint16_t		datablkszsec;
	bqueue_node_t		ln;
};

typedef struct dump_bytes_io {
	dmu_sendarg_t	*dbi_dsp;
	void		*dbi_buf;
	int		dbi_len;
} dump_bytes_io_t;

static void
dump_bytes_cb(void *arg)
{
	dump_bytes_io_t *dbi = (dump_bytes_io_t *)arg;
	dmu_sendarg_t *dsp = dbi->dbi_dsp;
	dsl_dataset_t *ds = dmu_objset_ds(dsp->dsa_os);
	ssize_t resid; /* have to get resid to get detailed errno */

	/*
	 * The code does not rely on len being a multiple of 8.  We keep
	 * this assertion because of the corresponding assertion in
	 * receive_read().  Keeping this assertion ensures that we do not
	 * inadvertently break backwards compatibility (causing the assertion
	 * in receive_read() to trigger on old software). Newer feature flags
	 * (such as raw send) may break this assertion since they were
	 * introduced after the requirement was made obsolete.
	 */

	ASSERT(dbi->dbi_len % 8 == 0 ||
	    (dsp->dsa_featureflags & DMU_BACKUP_FEATURE_RAW) != 0);

	dsp->dsa_err = vn_rdwr(UIO_WRITE, dsp->dsa_vp,
	    (caddr_t)dbi->dbi_buf, dbi->dbi_len,
	    0, UIO_SYSSPACE, FAPPEND, RLIM64_INFINITY, CRED(), &resid);

	mutex_enter(&ds->ds_sendstream_lock);
	*dsp->dsa_off += dbi->dbi_len;
	mutex_exit(&ds->ds_sendstream_lock);
}

static int
dump_bytes(dmu_sendarg_t *dsp, void *buf, int len)
{
	dump_bytes_io_t dbi;

	dbi.dbi_dsp = dsp;
	dbi.dbi_buf = buf;
	dbi.dbi_len = len;

#if defined(HAVE_LARGE_STACKS)
	dump_bytes_cb(&dbi);
#else
	/*
	 * The vn_rdwr() call is performed in a taskq to ensure that there is
	 * always enough stack space to write safely to the target filesystem.
	 * The ZIO_TYPE_FREE threads are used because there can be a lot of
	 * them and they are used in vdev_file.c for a similar purpose.
	 */
	spa_taskq_dispatch_sync(dmu_objset_spa(dsp->dsa_os), ZIO_TYPE_FREE,
	    ZIO_TASKQ_ISSUE, dump_bytes_cb, &dbi, TQ_SLEEP);
#endif /* HAVE_LARGE_STACKS */

	return (dsp->dsa_err);
}

/*
 * For all record types except BEGIN, fill in the checksum (overlaid in
 * drr_u.drr_checksum.drr_checksum).  The checksum verifies everything
 * up to the start of the checksum itself.
 */
static int
dump_record(dmu_sendarg_t *dsp, void *payload, int payload_len)
{
	ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum),
	    ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t));
	(void) fletcher_4_incremental_native(dsp->dsa_drr,
	    offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum),
	    &dsp->dsa_zc);
	if (dsp->dsa_drr->drr_type == DRR_BEGIN) {
		dsp->dsa_sent_begin = B_TRUE;
	} else {
		ASSERT(ZIO_CHECKSUM_IS_ZERO(&dsp->dsa_drr->drr_u.
		    drr_checksum.drr_checksum));
		dsp->dsa_drr->drr_u.drr_checksum.drr_checksum = dsp->dsa_zc;
	}
	if (dsp->dsa_drr->drr_type == DRR_END) {
		dsp->dsa_sent_end = B_TRUE;
	}
	(void) fletcher_4_incremental_native(&dsp->dsa_drr->
	    drr_u.drr_checksum.drr_checksum,
	    sizeof (zio_cksum_t), &dsp->dsa_zc);
	if (dump_bytes(dsp, dsp->dsa_drr, sizeof (dmu_replay_record_t)) != 0)
		return (SET_ERROR(EINTR));
	if (payload_len != 0) {
		(void) fletcher_4_incremental_native(payload, payload_len,
		    &dsp->dsa_zc);
		if (dump_bytes(dsp, payload, payload_len) != 0)
			return (SET_ERROR(EINTR));
	}
	return (0);
}

/*
 * Fill in the drr_free struct, or perform aggregation if the previous record is
 * also a free record, and the two are adjacent.
 *
 * Note that we send free records even for a full send, because we want to be
 * able to receive a full send as a clone, which requires a list of all the free
 * and freeobject records that were generated on the source.
 */
static int
dump_free(dmu_sendarg_t *dsp, uint64_t object, uint64_t offset,
    uint64_t length)
{
	struct drr_free *drrf = &(dsp->dsa_drr->drr_u.drr_free);

	/*
	 * When we receive a free record, dbuf_free_range() assumes
	 * that the receiving system doesn't have any dbufs in the range
	 * being freed.  This is always true because there is a one-record
	 * constraint: we only send one WRITE record for any given
	 * object,offset.  We know that the one-record constraint is
	 * true because we always send data in increasing order by
	 * object,offset.
	 *
	 * If the increasing-order constraint ever changes, we should find
	 * another way to assert that the one-record constraint is still
	 * satisfied.
	 */
	ASSERT(object > dsp->dsa_last_data_object ||
	    (object == dsp->dsa_last_data_object &&
	    offset > dsp->dsa_last_data_offset));

	/*
	 * If there is a pending op, but it's not PENDING_FREE, push it out,
	 * since free block aggregation can only be done for blocks of the
	 * same type (i.e., DRR_FREE records can only be aggregated with
	 * other DRR_FREE records.  DRR_FREEOBJECTS records can only be
	 * aggregated with other DRR_FREEOBJECTS records.
	 */
	if (dsp->dsa_pending_op != PENDING_NONE &&
	    dsp->dsa_pending_op != PENDING_FREE) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}

	if (dsp->dsa_pending_op == PENDING_FREE) {
		/*
		 * There should never be a PENDING_FREE if length is
		 * DMU_OBJECT_END (because dump_dnode is the only place where
		 * this function is called with a DMU_OBJECT_END, and only after
		 * flushing any pending record).
		 */
		ASSERT(length != DMU_OBJECT_END);
		/*
		 * Check to see whether this free block can be aggregated
		 * with pending one.
		 */
		if (drrf->drr_object == object && drrf->drr_offset +
		    drrf->drr_length == offset) {
			if (offset + length < offset)
				drrf->drr_length = DMU_OBJECT_END;
			else
				drrf->drr_length += length;
			return (0);
		} else {
			/* not a continuation.  Push out pending record */
			if (dump_record(dsp, NULL, 0) != 0)
				return (SET_ERROR(EINTR));
			dsp->dsa_pending_op = PENDING_NONE;
		}
	}
	/* create a FREE record and make it pending */
	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_FREE;
	drrf->drr_object = object;
	drrf->drr_offset = offset;
	if (offset + length < offset)
		drrf->drr_length = DMU_OBJECT_END;
	else
		drrf->drr_length = length;
	drrf->drr_toguid = dsp->dsa_toguid;
	if (length == DMU_OBJECT_END) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
	} else {
		dsp->dsa_pending_op = PENDING_FREE;
	}

	return (0);
}

static int
dump_write(dmu_sendarg_t *dsp, dmu_object_type_t type, uint64_t object,
    uint64_t offset, int lsize, int psize, const blkptr_t *bp, void *data)
{
	uint64_t payload_size;
	boolean_t raw = (dsp->dsa_featureflags & DMU_BACKUP_FEATURE_RAW);
	struct drr_write *drrw = &(dsp->dsa_drr->drr_u.drr_write);

	/*
	 * We send data in increasing object, offset order.
	 * See comment in dump_free() for details.
	 */
	ASSERT(object > dsp->dsa_last_data_object ||
	    (object == dsp->dsa_last_data_object &&
	    offset > dsp->dsa_last_data_offset));
	dsp->dsa_last_data_object = object;
	dsp->dsa_last_data_offset = offset + lsize - 1;

	/*
	 * If there is any kind of pending aggregation (currently either
	 * a grouping of free objects or free blocks), push it out to
	 * the stream, since aggregation can't be done across operations
	 * of different types.
	 */
	if (dsp->dsa_pending_op != PENDING_NONE) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}
	/* write a WRITE record */
	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_WRITE;
	drrw->drr_object = object;
	drrw->drr_type = type;
	drrw->drr_offset = offset;
	drrw->drr_toguid = dsp->dsa_toguid;
	drrw->drr_logical_size = lsize;

	/* only set the compression fields if the buf is compressed or raw */
	if (raw || lsize != psize) {
		ASSERT(!BP_IS_EMBEDDED(bp));
		ASSERT3S(psize, >, 0);

		if (raw) {
			ASSERT(BP_IS_PROTECTED(bp));

			/*
			 * This is a raw protected block so we need to pass
			 * along everything the receiving side will need to
			 * interpret this block, including the byteswap, salt,
			 * IV, and MAC.
			 */
			if (BP_SHOULD_BYTESWAP(bp))
				drrw->drr_flags |= DRR_RAW_BYTESWAP;
			zio_crypt_decode_params_bp(bp, drrw->drr_salt,
			    drrw->drr_iv);
			zio_crypt_decode_mac_bp(bp, drrw->drr_mac);
		} else {
			/* this is a compressed block */
			ASSERT(dsp->dsa_featureflags &
			    DMU_BACKUP_FEATURE_COMPRESSED);
			ASSERT(!BP_SHOULD_BYTESWAP(bp));
			ASSERT(!DMU_OT_IS_METADATA(BP_GET_TYPE(bp)));
			ASSERT3U(BP_GET_COMPRESS(bp), !=, ZIO_COMPRESS_OFF);
			ASSERT3S(lsize, >=, psize);
		}

		/* set fields common to compressed and raw sends */
		drrw->drr_compressiontype = BP_GET_COMPRESS(bp);
		drrw->drr_compressed_size = psize;
		payload_size = drrw->drr_compressed_size;
	} else {
		payload_size = drrw->drr_logical_size;
	}

	if (bp == NULL || BP_IS_EMBEDDED(bp) || (BP_IS_PROTECTED(bp) && !raw)) {
		/*
		 * There's no pre-computed checksum for partial-block writes,
		 * embedded BP's, or encrypted BP's that are being sent as
		 * plaintext, so (like fletcher4-checkummed blocks) userland
		 * will have to compute a dedup-capable checksum itself.
		 */
		drrw->drr_checksumtype = ZIO_CHECKSUM_OFF;
	} else {
		drrw->drr_checksumtype = BP_GET_CHECKSUM(bp);
		if (zio_checksum_table[drrw->drr_checksumtype].ci_flags &
		    ZCHECKSUM_FLAG_DEDUP)
			drrw->drr_flags |= DRR_CHECKSUM_DEDUP;
		DDK_SET_LSIZE(&drrw->drr_key, BP_GET_LSIZE(bp));
		DDK_SET_PSIZE(&drrw->drr_key, BP_GET_PSIZE(bp));
		DDK_SET_COMPRESS(&drrw->drr_key, BP_GET_COMPRESS(bp));
		DDK_SET_CRYPT(&drrw->drr_key, BP_IS_PROTECTED(bp));
		drrw->drr_key.ddk_cksum = bp->blk_cksum;
	}

	if (dump_record(dsp, data, payload_size) != 0)
		return (SET_ERROR(EINTR));
	return (0);
}

static int
dump_write_embedded(dmu_sendarg_t *dsp, uint64_t object, uint64_t offset,
    int blksz, const blkptr_t *bp)
{
	char buf[BPE_PAYLOAD_SIZE];
	struct drr_write_embedded *drrw =
	    &(dsp->dsa_drr->drr_u.drr_write_embedded);

	if (dsp->dsa_pending_op != PENDING_NONE) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}

	ASSERT(BP_IS_EMBEDDED(bp));

	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_WRITE_EMBEDDED;
	drrw->drr_object = object;
	drrw->drr_offset = offset;
	drrw->drr_length = blksz;
	drrw->drr_toguid = dsp->dsa_toguid;
	drrw->drr_compression = BP_GET_COMPRESS(bp);
	drrw->drr_etype = BPE_GET_ETYPE(bp);
	drrw->drr_lsize = BPE_GET_LSIZE(bp);
	drrw->drr_psize = BPE_GET_PSIZE(bp);

	decode_embedded_bp_compressed(bp, buf);

	if (dump_record(dsp, buf, P2ROUNDUP(drrw->drr_psize, 8)) != 0)
		return (SET_ERROR(EINTR));
	return (0);
}

static int
dump_spill(dmu_sendarg_t *dsp, const blkptr_t *bp, uint64_t object, void *data)
{
	struct drr_spill *drrs = &(dsp->dsa_drr->drr_u.drr_spill);
	uint64_t blksz = BP_GET_LSIZE(bp);

	if (dsp->dsa_pending_op != PENDING_NONE) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}

	/* write a SPILL record */
	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_SPILL;
	drrs->drr_object = object;
	drrs->drr_length = blksz;
	drrs->drr_toguid = dsp->dsa_toguid;

	/* handle raw send fields */
	if (dsp->dsa_featureflags & DMU_BACKUP_FEATURE_RAW) {
		ASSERT(BP_IS_PROTECTED(bp));

		if (BP_SHOULD_BYTESWAP(bp))
			drrs->drr_flags |= DRR_RAW_BYTESWAP;
		drrs->drr_compressiontype = BP_GET_COMPRESS(bp);
		drrs->drr_compressed_size = BP_GET_PSIZE(bp);
		zio_crypt_decode_params_bp(bp, drrs->drr_salt, drrs->drr_iv);
		zio_crypt_decode_mac_bp(bp, drrs->drr_mac);
	}

	if (dump_record(dsp, data, blksz) != 0)
		return (SET_ERROR(EINTR));
	return (0);
}

static int
dump_freeobjects(dmu_sendarg_t *dsp, uint64_t firstobj, uint64_t numobjs)
{
	struct drr_freeobjects *drrfo = &(dsp->dsa_drr->drr_u.drr_freeobjects);
	uint64_t maxobj = DNODES_PER_BLOCK *
	    (DMU_META_DNODE(dsp->dsa_os)->dn_maxblkid + 1);

	/*
	 * ZoL < 0.7 does not handle large FREEOBJECTS records correctly,
	 * leading to zfs recv never completing. to avoid this issue, don't
	 * send FREEOBJECTS records for object IDs which cannot exist on the
	 * receiving side.
	 */
	if (maxobj > 0) {
		if (maxobj < firstobj)
			return (0);

		if (maxobj < firstobj + numobjs)
			numobjs = maxobj - firstobj;
	}

	/*
	 * If there is a pending op, but it's not PENDING_FREEOBJECTS,
	 * push it out, since free block aggregation can only be done for
	 * blocks of the same type (i.e., DRR_FREE records can only be
	 * aggregated with other DRR_FREE records.  DRR_FREEOBJECTS records
	 * can only be aggregated with other DRR_FREEOBJECTS records.
	 */
	if (dsp->dsa_pending_op != PENDING_NONE &&
	    dsp->dsa_pending_op != PENDING_FREEOBJECTS) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}
	if (dsp->dsa_pending_op == PENDING_FREEOBJECTS) {
		/*
		 * See whether this free object array can be aggregated
		 * with pending one
		 */
		if (drrfo->drr_firstobj + drrfo->drr_numobjs == firstobj) {
			drrfo->drr_numobjs += numobjs;
			return (0);
		} else {
			/* can't be aggregated.  Push out pending record */
			if (dump_record(dsp, NULL, 0) != 0)
				return (SET_ERROR(EINTR));
			dsp->dsa_pending_op = PENDING_NONE;
		}
	}

	/* write a FREEOBJECTS record */
	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_FREEOBJECTS;
	drrfo->drr_firstobj = firstobj;
	drrfo->drr_numobjs = numobjs;
	drrfo->drr_toguid = dsp->dsa_toguid;

	dsp->dsa_pending_op = PENDING_FREEOBJECTS;

	return (0);
}

static int
dump_dnode(dmu_sendarg_t *dsp, const blkptr_t *bp, uint64_t object,
    dnode_phys_t *dnp)
{
	struct drr_object *drro = &(dsp->dsa_drr->drr_u.drr_object);
	int bonuslen;

	if (object < dsp->dsa_resume_object) {
		/*
		 * Note: when resuming, we will visit all the dnodes in
		 * the block of dnodes that we are resuming from.  In
		 * this case it's unnecessary to send the dnodes prior to
		 * the one we are resuming from.  We should be at most one
		 * block's worth of dnodes behind the resume point.
		 */
		ASSERT3U(dsp->dsa_resume_object - object, <,
		    1 << (DNODE_BLOCK_SHIFT - DNODE_SHIFT));
		return (0);
	}

	if (dnp == NULL || dnp->dn_type == DMU_OT_NONE)
		return (dump_freeobjects(dsp, object, 1));

	if (dsp->dsa_pending_op != PENDING_NONE) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}

	/* write an OBJECT record */
	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_OBJECT;
	drro->drr_object = object;
	drro->drr_type = dnp->dn_type;
	drro->drr_bonustype = dnp->dn_bonustype;
	drro->drr_blksz = dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT;
	drro->drr_bonuslen = dnp->dn_bonuslen;
	drro->drr_dn_slots = dnp->dn_extra_slots + 1;
	drro->drr_checksumtype = dnp->dn_checksum;
	drro->drr_compress = dnp->dn_compress;
	drro->drr_toguid = dsp->dsa_toguid;

	if (!(dsp->dsa_featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) &&
	    drro->drr_blksz > SPA_OLD_MAXBLOCKSIZE)
		drro->drr_blksz = SPA_OLD_MAXBLOCKSIZE;

	bonuslen = P2ROUNDUP(dnp->dn_bonuslen, 8);

	if ((dsp->dsa_featureflags & DMU_BACKUP_FEATURE_RAW)) {
		ASSERT(BP_IS_ENCRYPTED(bp));

		if (BP_SHOULD_BYTESWAP(bp))
			drro->drr_flags |= DRR_RAW_BYTESWAP;

		/* needed for reconstructing dnp on recv side */
		drro->drr_maxblkid = dnp->dn_maxblkid;
		drro->drr_indblkshift = dnp->dn_indblkshift;
		drro->drr_nlevels = dnp->dn_nlevels;
		drro->drr_nblkptr = dnp->dn_nblkptr;

		/*
		 * Since we encrypt the entire bonus area, the (raw) part
		 * beyond the bonuslen is actually nonzero, so we need
		 * to send it.
		 */
		if (bonuslen != 0) {
			drro->drr_raw_bonuslen = DN_MAX_BONUS_LEN(dnp);
			bonuslen = drro->drr_raw_bonuslen;
		}
	}

	if (dump_record(dsp, DN_BONUS(dnp), bonuslen) != 0)
		return (SET_ERROR(EINTR));

	/* Free anything past the end of the file. */
	if (dump_free(dsp, object, (dnp->dn_maxblkid + 1) *
	    (dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT), DMU_OBJECT_END) != 0)
		return (SET_ERROR(EINTR));
	if (dsp->dsa_err != 0)
		return (SET_ERROR(EINTR));
	return (0);
}

static int
dump_object_range(dmu_sendarg_t *dsp, const blkptr_t *bp, uint64_t firstobj,
    uint64_t numslots)
{
	struct drr_object_range *drror =
	    &(dsp->dsa_drr->drr_u.drr_object_range);

	/* we only use this record type for raw sends */
	ASSERT(BP_IS_PROTECTED(bp));
	ASSERT(dsp->dsa_featureflags & DMU_BACKUP_FEATURE_RAW);
	ASSERT3U(BP_GET_COMPRESS(bp), ==, ZIO_COMPRESS_OFF);
	ASSERT3U(BP_GET_TYPE(bp), ==, DMU_OT_DNODE);
	ASSERT0(BP_GET_LEVEL(bp));

	if (dsp->dsa_pending_op != PENDING_NONE) {
		if (dump_record(dsp, NULL, 0) != 0)
			return (SET_ERROR(EINTR));
		dsp->dsa_pending_op = PENDING_NONE;
	}

	bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t));
	dsp->dsa_drr->drr_type = DRR_OBJECT_RANGE;
	drror->drr_firstobj = firstobj;
	drror->drr_numslots = numslots;
	drror->drr_toguid = dsp->dsa_toguid;
	if (BP_SHOULD_BYTESWAP(bp))
		drror->drr_flags |= DRR_RAW_BYTESWAP;
	zio_crypt_decode_params_bp(bp, drror->drr_salt, drror->drr_iv);
	zio_crypt_decode_mac_bp(bp, drror->drr_mac);

	if (dump_record(dsp, NULL, 0) != 0)
		return (SET_ERROR(EINTR));
	return (0);
}

static boolean_t
backup_do_embed(dmu_sendarg_t *dsp, const blkptr_t *bp)
{
	if (!BP_IS_EMBEDDED(bp))
		return (B_FALSE);

	/*
	 * Compression function must be legacy, or explicitly enabled.
	 */
	if ((BP_GET_COMPRESS(bp) >= ZIO_COMPRESS_LEGACY_FUNCTIONS &&
	    !(dsp->dsa_featureflags & DMU_BACKUP_FEATURE_LZ4)))
		return (B_FALSE);

	/*
	 * Embed type must be explicitly enabled.
	 */
	switch (BPE_GET_ETYPE(bp)) {
	case BP_EMBEDDED_TYPE_DATA:
		if (dsp->dsa_featureflags & DMU_BACKUP_FEATURE_EMBED_DATA)
			return (B_TRUE);
		break;
	default:
		return (B_FALSE);
	}
	return (B_FALSE);
}

/*
 * This is the callback function to traverse_dataset that acts as the worker
 * thread for dmu_send_impl.
 */
/*ARGSUSED*/
static int
send_cb(spa_t *spa, zilog_t *zilog, const blkptr_t *bp,
    const zbookmark_phys_t *zb, const struct dnode_phys *dnp, void *arg)
{
	struct send_thread_arg *sta = arg;
	struct send_block_record *record;
	uint64_t record_size;
	int err = 0;

	ASSERT(zb->zb_object == DMU_META_DNODE_OBJECT ||
	    zb->zb_object >= sta->resume.zb_object);
	ASSERT3P(sta->ds, !=, NULL);

	if (sta->cancel)
		return (SET_ERROR(EINTR));

	if (bp == NULL) {
		ASSERT3U(zb->zb_level, ==, ZB_DNODE_LEVEL);
		return (0);
	} else if (zb->zb_level < 0) {
		return (0);
	}

	record = kmem_zalloc(sizeof (struct send_block_record), KM_SLEEP);
	record->eos_marker = B_FALSE;
	record->bp = *bp;
	record->zb = *zb;
	record->indblkshift = dnp->dn_indblkshift;
	record->datablkszsec = dnp->dn_datablkszsec;
	record_size = dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT;
	bqueue_enqueue(&sta->q, record, record_size);

	return (err);
}

/*
 * This function kicks off the traverse_dataset.  It also handles setting the
 * error code of the thread in case something goes wrong, and pushes the End of
 * Stream record when the traverse_dataset call has finished.  If there is no
 * dataset to traverse, the thread immediately pushes End of Stream marker.
 */
static void
send_traverse_thread(void *arg)
{
	struct send_thread_arg *st_arg = arg;
	int err;
	struct send_block_record *data;
	fstrans_cookie_t cookie = spl_fstrans_mark();

	if (st_arg->ds != NULL) {
		err = traverse_dataset_resume(st_arg->ds,
		    st_arg->fromtxg, &st_arg->resume,
		    st_arg->flags, send_cb, st_arg);

		if (err != EINTR)
			st_arg->error_code = err;
	}
	data = kmem_zalloc(sizeof (*data), KM_SLEEP);
	data->eos_marker = B_TRUE;
	bqueue_enqueue(&st_arg->q, data, 1);
	spl_fstrans_unmark(cookie);
	thread_exit();
}

/*
 * This function actually handles figuring out what kind of record needs to be
 * dumped, reading the data (which has hopefully been prefetched), and calling
 * the appropriate helper function.
 */
static int
do_dump(dmu_sendarg_t *dsa, struct send_block_record *data)
{
	dsl_dataset_t *ds = dmu_objset_ds(dsa->dsa_os);
	const blkptr_t *bp = &data->bp;
	const zbookmark_phys_t *zb = &data->zb;
	uint8_t indblkshift = data->indblkshift;
	uint16_t dblkszsec = data->datablkszsec;
	spa_t *spa = ds->ds_dir->dd_pool->dp_spa;
	dmu_object_type_t type = bp ? BP_GET_TYPE(bp) : DMU_OT_NONE;
	int err = 0;

	ASSERT3U(zb->zb_level, >=, 0);

	ASSERT(zb->zb_object == DMU_META_DNODE_OBJECT ||
	    zb->zb_object >= dsa->dsa_resume_object);

	/*
	 * All bps of an encrypted os should have the encryption bit set.
	 * If this is not true it indicates tampering and we report an error.
	 */
	if (dsa->dsa_os->os_encrypted &&
	    !BP_IS_HOLE(bp) && !BP_USES_CRYPT(bp)) {
		spa_log_error(spa, zb);
		zfs_panic_recover("unencrypted block in encrypted "
		    "object set %llu", ds->ds_object);
		return (SET_ERROR(EIO));
	}

	if (zb->zb_object != DMU_META_DNODE_OBJECT &&
	    DMU_OBJECT_IS_SPECIAL(zb->zb_object)) {
		return (0);
	} else if (BP_IS_HOLE(bp) &&
	    zb->zb_object == DMU_META_DNODE_OBJECT) {
		uint64_t span = BP_SPAN(dblkszsec, indblkshift, zb->zb_level);
		uint64_t dnobj = (zb->zb_blkid * span) >> DNODE_SHIFT;
		err = dump_freeobjects(dsa, dnobj, span >> DNODE_SHIFT);
	} else if (BP_IS_HOLE(bp)) {
		uint64_t span = BP_SPAN(dblkszsec, indblkshift, zb->zb_level);
		uint64_t offset = zb->zb_blkid * span;
		/* Don't dump free records for offsets > DMU_OBJECT_END */
		if (zb->zb_blkid == 0 || span <= DMU_OBJECT_END / zb->zb_blkid)
			err = dump_free(dsa, zb->zb_object, offset, span);
	} else if (zb->zb_level > 0 || type == DMU_OT_OBJSET) {
		return (0);
	} else if (type == DMU_OT_DNODE) {
		int epb = BP_GET_LSIZE(bp) >> DNODE_SHIFT;
		arc_flags_t aflags = ARC_FLAG_WAIT;
		arc_buf_t *abuf;
		enum zio_flag zioflags = ZIO_FLAG_CANFAIL;

		if (dsa->dsa_featureflags & DMU_BACKUP_FEATURE_RAW) {
			ASSERT(BP_IS_ENCRYPTED(bp));
			ASSERT3U(BP_GET_COMPRESS(bp), ==, ZIO_COMPRESS_OFF);
			zioflags |= ZIO_FLAG_RAW;
		}

		ASSERT0(zb->zb_level);

		if (arc_read(NULL, spa, bp, arc_getbuf_func, &abuf,
		    ZIO_PRIORITY_ASYNC_READ, zioflags, &aflags, zb) != 0)
			return (SET_ERROR(EIO));

		dnode_phys_t *blk = abuf->b_data;
		uint64_t dnobj = zb->zb_blkid * epb;

		/*
		 * Raw sends require sending encryption parameters for the
		 * block of dnodes. Regular sends do not need to send this
		 * info.
		 */
		if (dsa->dsa_featureflags & DMU_BACKUP_FEATURE_RAW) {
			ASSERT(arc_is_encrypted(abuf));
			err = dump_object_range(dsa, bp, dnobj, epb);
		}

		if (err == 0) {
			for (int i = 0; i < epb;
			    i += blk[i].dn_extra_slots + 1) {
				err = dump_dnode(dsa, bp, dnobj + i, blk + i);
				if (err != 0)
					break;
			}
		}
		arc_buf_destroy(abuf, &abuf);
	} else if (type == DMU_OT_SA) {
		arc_flags_t aflags = ARC_FLAG_WAIT;
		arc_buf_t *abuf;
		enum zio_flag zioflags = ZIO_FLAG_CANFAIL;

		if (dsa->dsa_featureflags & DMU_BACKUP_FEATURE_RAW) {
			ASSERT(BP_IS_PROTECTED(bp));
			zioflags |= ZIO_FLAG_RAW;
		}

		if (arc_read(NULL, spa, bp, arc_getbuf_func, &abuf,
		    ZIO_PRIORITY_ASYNC_READ, zioflags, &aflags, zb) != 0)
			return (SET_ERROR(EIO));

		err = dump_spill(dsa, bp, zb->zb_object, abuf->b_data);
		arc_buf_destroy(abuf, &abuf);
	} else if (backup_do_embed(dsa, bp)) {
		/* it's an embedded level-0 block of a regular object */
		int blksz = dblkszsec << SPA_MINBLOCKSHIFT;
		ASSERT0(zb->zb_level);
		err = dump_write_embedded(dsa, zb->zb_object,
		    zb->zb_blkid * blksz, blksz, bp);
	} else {
		/* it's a level-0 block of a regular object */
		arc_flags_t aflags = ARC_FLAG_WAIT;
		arc_buf_t *abuf;
		int blksz = dblkszsec << SPA_MINBLOCKSHIFT;
		uint64_t offset;

		/*
		 * If we have large blocks stored on disk but the send flags
		 * don't allow us to send large blocks, we split the data from
		 * the arc buf into chunks.
		 */
		boolean_t split_large_blocks = blksz > SPA_OLD_MAXBLOCKSIZE &&
		    !(dsa->dsa_featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS);

		/*
		 * Raw sends require that we always get raw data as it exists
		 * on disk, so we assert that we are not splitting blocks here.
		 */
		boolean_t request_raw =
		    (dsa->dsa_featureflags & DMU_BACKUP_FEATURE_RAW) != 0;

		/*
		 * We should only request compressed data from the ARC if all
		 * the following are true:
		 *  - stream compression was requested
		 *  - we aren't splitting large blocks into smaller chunks
		 *  - the data won't need to be byteswapped before sending
		 *  - this isn't an embedded block
		 *  - this isn't metadata (if receiving on a different endian
		 *    system it can be byteswapped more easily)
		 */
		boolean_t request_compressed =
		    (dsa->dsa_featureflags & DMU_BACKUP_FEATURE_COMPRESSED) &&
		    !split_large_blocks && !BP_SHOULD_BYTESWAP(bp) &&
		    !BP_IS_EMBEDDED(bp) && !DMU_OT_IS_METADATA(BP_GET_TYPE(bp));

		IMPLY(request_raw, !split_large_blocks);
		IMPLY(request_raw, BP_IS_PROTECTED(bp));
		ASSERT0(zb->zb_level);
		ASSERT(zb->zb_object > dsa->dsa_resume_object ||
		    (zb->zb_object == dsa->dsa_resume_object &&
		    zb->zb_blkid * blksz >= dsa->dsa_resume_offset));

		ASSERT3U(blksz, ==, BP_GET_LSIZE(bp));

		enum zio_flag zioflags = ZIO_FLAG_CANFAIL;
		if (request_raw)
			zioflags |= ZIO_FLAG_RAW;
		else if (request_compressed)
			zioflags |= ZIO_FLAG_RAW_COMPRESS;

		if (arc_read(NULL, spa, bp, arc_getbuf_func, &abuf,
		    ZIO_PRIORITY_ASYNC_READ, zioflags, &aflags, zb) != 0) {
			if (zfs_send_corrupt_data) {
				/* Send a block filled with 0x"zfs badd bloc" */
				abuf = arc_alloc_buf(spa, &abuf, ARC_BUFC_DATA,
				    blksz);
				uint64_t *ptr;
				for (ptr = abuf->b_data;
				    (char *)ptr < (char *)abuf->b_data + blksz;
				    ptr++)
					*ptr = 0x2f5baddb10cULL;
			} else {
				return (SET_ERROR(EIO));
			}
		}

		offset = zb->zb_blkid * blksz;

		if (split_large_blocks) {
			ASSERT0(arc_is_encrypted(abuf));
			ASSERT3U(arc_get_compression(abuf), ==,
			    ZIO_COMPRESS_OFF);
			char *buf = abuf->b_data;
			while (blksz > 0 && err == 0) {
				int n = MIN(blksz, SPA_OLD_MAXBLOCKSIZE);
				err = dump_write(dsa, type, zb->zb_object,
				    offset, n, n, NULL, buf);
				offset += n;
				buf += n;
				blksz -= n;
			}
		} else {
			err = dump_write(dsa, type, zb->zb_object, offset,
			    blksz, arc_buf_size(abuf), bp, abuf->b_data);
		}
		arc_buf_destroy(abuf, &abuf);
	}

	ASSERT(err == 0 || err == EINTR);
	return (err);
}

/*
 * Pop the new data off the queue, and free the old data.
 */
static struct send_block_record *
get_next_record(bqueue_t *bq, struct send_block_record *data)
{
	struct send_block_record *tmp = bqueue_dequeue(bq);
	kmem_free(data, sizeof (*data));
	return (tmp);
}

/*
 * Actually do the bulk of the work in a zfs send.
 *
 * Note: Releases dp using the specified tag.
 */
static int
dmu_send_impl(void *tag, dsl_pool_t *dp, dsl_dataset_t *to_ds,
    zfs_bookmark_phys_t *ancestor_zb, boolean_t is_clone,
    boolean_t embedok, boolean_t large_block_ok, boolean_t compressok,
    boolean_t rawok, int outfd, uint64_t resumeobj, uint64_t resumeoff,
    vnode_t *vp, offset_t *off)
{
	objset_t *os;
	dmu_replay_record_t *drr;
	dmu_sendarg_t *dsp;
	int err;
	uint64_t fromtxg = 0;
	uint64_t featureflags = 0;
	struct send_thread_arg to_arg;
	void *payload = NULL;
	size_t payload_len = 0;
	struct send_block_record *to_data;

	err = dmu_objset_from_ds(to_ds, &os);
	if (err != 0) {
		dsl_pool_rele(dp, tag);
		return (err);
	}

	/*
	 * If this is a non-raw send of an encrypted ds, we can ensure that
	 * the objset_phys_t is authenticated. This is safe because this is
	 * either a snapshot or we have owned the dataset, ensuring that
	 * it can't be modified.
	 */
	if (!rawok && os->os_encrypted &&
	    arc_is_unauthenticated(os->os_phys_buf)) {
		zbookmark_phys_t zb;

		SET_BOOKMARK(&zb, to_ds->ds_object, ZB_ROOT_OBJECT,
		    ZB_ROOT_LEVEL, ZB_ROOT_BLKID);
		err = arc_untransform(os->os_phys_buf, os->os_spa,
		    &zb, B_FALSE);
		if (err != 0) {
			dsl_pool_rele(dp, tag);
			return (err);
		}

		ASSERT0(arc_is_unauthenticated(os->os_phys_buf));
	}

	drr = kmem_zalloc(sizeof (dmu_replay_record_t), KM_SLEEP);
	drr->drr_type = DRR_BEGIN;
	drr->drr_u.drr_begin.drr_magic = DMU_BACKUP_MAGIC;
	DMU_SET_STREAM_HDRTYPE(drr->drr_u.drr_begin.drr_versioninfo,
	    DMU_SUBSTREAM);

	bzero(&to_arg, sizeof (to_arg));

#ifdef _KERNEL
	if (dmu_objset_type(os) == DMU_OST_ZFS) {
		uint64_t version;
		if (zfs_get_zplprop(os, ZFS_PROP_VERSION, &version) != 0) {
			kmem_free(drr, sizeof (dmu_replay_record_t));
			dsl_pool_rele(dp, tag);
			return (SET_ERROR(EINVAL));
		}
		if (version >= ZPL_VERSION_SA) {
			featureflags |= DMU_BACKUP_FEATURE_SA_SPILL;
		}
	}
#endif

	/* raw sends imply large_block_ok */
	if ((large_block_ok || rawok) &&
	    to_ds->ds_feature_inuse[SPA_FEATURE_LARGE_BLOCKS])
		featureflags |= DMU_BACKUP_FEATURE_LARGE_BLOCKS;
	if (to_ds->ds_feature_inuse[SPA_FEATURE_LARGE_DNODE])
		featureflags |= DMU_BACKUP_FEATURE_LARGE_DNODE;

	/* encrypted datasets will not have embedded blocks */
	if ((embedok || rawok) && !os->os_encrypted &&
	    spa_feature_is_active(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA)) {
		featureflags |= DMU_BACKUP_FEATURE_EMBED_DATA;
	}

	/* raw send implies compressok */
	if (compressok || rawok)
		featureflags |= DMU_BACKUP_FEATURE_COMPRESSED;
	if (rawok && os->os_encrypted)
		featureflags |= DMU_BACKUP_FEATURE_RAW;

	if ((featureflags &
	    (DMU_BACKUP_FEATURE_EMBED_DATA | DMU_BACKUP_FEATURE_COMPRESSED |
	    DMU_BACKUP_FEATURE_RAW)) != 0 &&
	    spa_feature_is_active(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS)) {
		featureflags |= DMU_BACKUP_FEATURE_LZ4;
	}

	if (resumeobj != 0 || resumeoff != 0) {
		featureflags |= DMU_BACKUP_FEATURE_RESUMING;
	}

	DMU_SET_FEATUREFLAGS(drr->drr_u.drr_begin.drr_versioninfo,
	    featureflags);

	drr->drr_u.drr_begin.drr_creation_time =
	    dsl_dataset_phys(to_ds)->ds_creation_time;
	drr->drr_u.drr_begin.drr_type = dmu_objset_type(os);
	if (is_clone)
		drr->drr_u.drr_begin.drr_flags |= DRR_FLAG_CLONE;
	drr->drr_u.drr_begin.drr_toguid = dsl_dataset_phys(to_ds)->ds_guid;
	if (dsl_dataset_phys(to_ds)->ds_flags & DS_FLAG_CI_DATASET)
		drr->drr_u.drr_begin.drr_flags |= DRR_FLAG_CI_DATA;
	if (zfs_send_set_freerecords_bit)
		drr->drr_u.drr_begin.drr_flags |= DRR_FLAG_FREERECORDS;

	if (ancestor_zb != NULL) {
		drr->drr_u.drr_begin.drr_fromguid =
		    ancestor_zb->zbm_guid;
		fromtxg = ancestor_zb->zbm_creation_txg;
	}
	dsl_dataset_name(to_ds, drr->drr_u.drr_begin.drr_toname);
	if (!to_ds->ds_is_snapshot) {
		(void) strlcat(drr->drr_u.drr_begin.drr_toname, "@--head--",
		    sizeof (drr->drr_u.drr_begin.drr_toname));
	}

	dsp = kmem_zalloc(sizeof (dmu_sendarg_t), KM_SLEEP);

	dsp->dsa_drr = drr;
	dsp->dsa_vp = vp;
	dsp->dsa_outfd = outfd;
	dsp->dsa_proc = curproc;
	dsp->dsa_os = os;
	dsp->dsa_off = off;
	dsp->dsa_toguid = dsl_dataset_phys(to_ds)->ds_guid;
	dsp->dsa_pending_op = PENDING_NONE;
	dsp->dsa_featureflags = featureflags;
	dsp->dsa_resume_object = resumeobj;
	dsp->dsa_resume_offset = resumeoff;

	mutex_enter(&to_ds->ds_sendstream_lock);
	list_insert_head(&to_ds->ds_sendstreams, dsp);
	mutex_exit(&to_ds->ds_sendstream_lock);

	dsl_dataset_long_hold(to_ds, FTAG);
	dsl_pool_rele(dp, tag);

	/* handle features that require a DRR_BEGIN payload */
	if (featureflags &
	    (DMU_BACKUP_FEATURE_RESUMING | DMU_BACKUP_FEATURE_RAW)) {
		nvlist_t *keynvl = NULL;
		nvlist_t *nvl = fnvlist_alloc();

		if (featureflags & DMU_BACKUP_FEATURE_RESUMING) {
			dmu_object_info_t to_doi;
			err = dmu_object_info(os, resumeobj, &to_doi);
			if (err != 0) {
				fnvlist_free(nvl);
				goto out;
			}

			SET_BOOKMARK(&to_arg.resume, to_ds->ds_object,
			    resumeobj, 0,
			    resumeoff / to_doi.doi_data_block_size);

			fnvlist_add_uint64(nvl, "resume_object", resumeobj);
			fnvlist_add_uint64(nvl, "resume_offset", resumeoff);
		}

		if (featureflags & DMU_BACKUP_FEATURE_RAW) {
			ASSERT(os->os_encrypted);

			err = dsl_crypto_populate_key_nvlist(to_ds, &keynvl);
			if (err != 0) {
				fnvlist_free(nvl);
				goto out;
			}

			fnvlist_add_nvlist(nvl, "crypt_keydata", keynvl);
		}

		payload = fnvlist_pack(nvl, &payload_len);
		drr->drr_payloadlen = payload_len;
		fnvlist_free(keynvl);
		fnvlist_free(nvl);
	}

	err = dump_record(dsp, payload, payload_len);
	fnvlist_pack_free(payload, payload_len);
	if (err != 0) {
		err = dsp->dsa_err;
		goto out;
	}

	err = bqueue_init(&to_arg.q,
	    MAX(zfs_send_queue_length, 2 * zfs_max_recordsize),
	    offsetof(struct send_block_record, ln));
	to_arg.error_code = 0;
	to_arg.cancel = B_FALSE;
	to_arg.ds = to_ds;
	to_arg.fromtxg = fromtxg;
	to_arg.flags = TRAVERSE_PRE | TRAVERSE_PREFETCH;
	if (rawok)
		to_arg.flags |= TRAVERSE_NO_DECRYPT;
	(void) thread_create(NULL, 0, send_traverse_thread, &to_arg, 0, curproc,
	    TS_RUN, minclsyspri);

	to_data = bqueue_dequeue(&to_arg.q);

	while (!to_data->eos_marker && err == 0) {
		err = do_dump(dsp, to_data);
		to_data = get_next_record(&to_arg.q, to_data);
		if (issig(JUSTLOOKING) && issig(FORREAL))
			err = EINTR;
	}

	if (err != 0) {
		to_arg.cancel = B_TRUE;
		while (!to_data->eos_marker) {
			to_data = get_next_record(&to_arg.q, to_data);
		}
	}
	kmem_free(to_data, sizeof (*to_data));

	bqueue_destroy(&to_arg.q);

	if (err == 0 && to_arg.error_code != 0)
		err = to_arg.error_code;

	if (err != 0)
		goto out;

	if (dsp->dsa_pending_op != PENDING_NONE)
		if (dump_record(dsp, NULL, 0) != 0)
			err = SET_ERROR(EINTR);

	if (err != 0) {
		if (err == EINTR && dsp->dsa_err != 0)
			err = dsp->dsa_err;
		goto out;
	}

	bzero(drr, sizeof (dmu_replay_record_t));
	drr->drr_type = DRR_END;
	drr->drr_u.drr_end.drr_checksum = dsp->dsa_zc;
	drr->drr_u.drr_end.drr_toguid = dsp->dsa_toguid;

	if (dump_record(dsp, NULL, 0) != 0)
		err = dsp->dsa_err;
out:
	mutex_enter(&to_ds->ds_sendstream_lock);
	list_remove(&to_ds->ds_sendstreams, dsp);
	mutex_exit(&to_ds->ds_sendstream_lock);

	VERIFY(err != 0 || (dsp->dsa_sent_begin && dsp->dsa_sent_end));

	kmem_free(drr, sizeof (dmu_replay_record_t));
	kmem_free(dsp, sizeof (dmu_sendarg_t));

	dsl_dataset_long_rele(to_ds, FTAG);

	return (err);
}

int
dmu_send_obj(const char *pool, uint64_t tosnap, uint64_t fromsnap,
    boolean_t embedok, boolean_t large_block_ok, boolean_t compressok,
    boolean_t rawok, int outfd, vnode_t *vp, offset_t *off)
{
	dsl_pool_t *dp;
	dsl_dataset_t *ds;
	dsl_dataset_t *fromds = NULL;
	ds_hold_flags_t dsflags = (rawok) ? 0 : DS_HOLD_FLAG_DECRYPT;
	int err;

	err = dsl_pool_hold(pool, FTAG, &dp);
	if (err != 0)
		return (err);

	err = dsl_dataset_hold_obj_flags(dp, tosnap, dsflags, FTAG, &ds);
	if (err != 0) {
		dsl_pool_rele(dp, FTAG);
		return (err);
	}

	if (fromsnap != 0) {
		zfs_bookmark_phys_t zb;
		boolean_t is_clone;

		err = dsl_dataset_hold_obj(dp, fromsnap, FTAG, &fromds);
		if (err != 0) {
			dsl_dataset_rele_flags(ds, dsflags, FTAG);
			dsl_pool_rele(dp, FTAG);
			return (err);
		}
		if (!dsl_dataset_is_before(ds, fromds, 0))
			err = SET_ERROR(EXDEV);
		zb.zbm_creation_time =
		    dsl_dataset_phys(fromds)->ds_creation_time;
		zb.zbm_creation_txg = dsl_dataset_phys(fromds)->ds_creation_txg;
		zb.zbm_guid = dsl_dataset_phys(fromds)->ds_guid;
		is_clone = (fromds->ds_dir != ds->ds_dir);
		dsl_dataset_rele(fromds, FTAG);
		err = dmu_send_impl(FTAG, dp, ds, &zb, is_clone,
		    embedok, large_block_ok, compressok, rawok, outfd,
		    0, 0, vp, off);
	} else {
		err = dmu_send_impl(FTAG, dp, ds, NULL, B_FALSE,
		    embedok, large_block_ok, compressok, rawok, outfd,
		    0, 0, vp, off);
	}
	dsl_dataset_rele_flags(ds, dsflags, FTAG);
	return (err);
}

int
dmu_send(const char *tosnap, const char *fromsnap, boolean_t embedok,
    boolean_t large_block_ok, boolean_t compressok, boolean_t rawok,
    int outfd, uint64_t resumeobj, uint64_t resumeoff, vnode_t *vp,
    offset_t *off)
{
	dsl_pool_t *dp;
	dsl_dataset_t *ds;
	int err;
	ds_hold_flags_t dsflags = (rawok) ? 0 : DS_HOLD_FLAG_DECRYPT;
	boolean_t owned = B_FALSE;

	if (fromsnap != NULL && strpbrk(fromsnap, "@#") == NULL)
		return (SET_ERROR(EINVAL));

	err = dsl_pool_hold(tosnap, FTAG, &dp);
	if (err != 0)
		return (err);

	if (strchr(tosnap, '@') == NULL && spa_writeable(dp->dp_spa)) {
		/*
		 * We are sending a filesystem or volume.  Ensure
		 * that it doesn't change by owning the dataset.
		 */
		err = dsl_dataset_own(dp, tosnap, dsflags, FTAG, &ds);
		owned = B_TRUE;
	} else {
		err = dsl_dataset_hold_flags(dp, tosnap, dsflags, FTAG, &ds);
	}
	if (err != 0) {
		dsl_pool_rele(dp, FTAG);
		return (err);
	}

	if (fromsnap != NULL) {
		zfs_bookmark_phys_t zb;
		boolean_t is_clone = B_FALSE;
		int fsnamelen = strchr(tosnap, '@') - tosnap;

		/*
		 * If the fromsnap is in a different filesystem, then
		 * mark the send stream as a clone.
		 */
		if (strncmp(tosnap, fromsnap, fsnamelen) != 0 ||
		    (fromsnap[fsnamelen] != '@' &&
		    fromsnap[fsnamelen] != '#')) {
			is_clone = B_TRUE;
		}

		if (strchr(fromsnap, '@')) {
			dsl_dataset_t *fromds;
			err = dsl_dataset_hold(dp, fromsnap, FTAG, &fromds);
			if (err == 0) {
				if (!dsl_dataset_is_before(ds, fromds, 0))
					err = SET_ERROR(EXDEV);
				zb.zbm_creation_time =
				    dsl_dataset_phys(fromds)->ds_creation_time;
				zb.zbm_creation_txg =
				    dsl_dataset_phys(fromds)->ds_creation_txg;
				zb.zbm_guid = dsl_dataset_phys(fromds)->ds_guid;
				is_clone = (ds->ds_dir != fromds->ds_dir);
				dsl_dataset_rele(fromds, FTAG);
			}
		} else {
			err = dsl_bookmark_lookup(dp, fromsnap, ds, &zb);
		}
		if (err != 0) {
			if (owned)
				dsl_dataset_disown(ds, dsflags, FTAG);
			else
				dsl_dataset_rele_flags(ds, dsflags, FTAG);

			dsl_pool_rele(dp, FTAG);
			return (err);
		}
		err = dmu_send_impl(FTAG, dp, ds, &zb, is_clone,
		    embedok, large_block_ok, compressok, rawok,
		    outfd, resumeobj, resumeoff, vp, off);
	} else {
		err = dmu_send_impl(FTAG, dp, ds, NULL, B_FALSE,
		    embedok, large_block_ok, compressok, rawok,
		    outfd, resumeobj, resumeoff, vp, off);
	}
	if (owned)
		dsl_dataset_disown(ds, dsflags, FTAG);
	else
		dsl_dataset_rele_flags(ds, dsflags, FTAG);

	return (err);
}

static int
dmu_adjust_send_estimate_for_indirects(dsl_dataset_t *ds, uint64_t uncompressed,
    uint64_t compressed, boolean_t stream_compressed, uint64_t *sizep)
{
	int err;
	uint64_t size;
	/*
	 * Assume that space (both on-disk and in-stream) is dominated by
	 * data.  We will adjust for indirect blocks and the copies property,
	 * but ignore per-object space used (eg, dnodes and DRR_OBJECT records).
	 */

	uint64_t recordsize;
	uint64_t record_count;
	objset_t *os;
	VERIFY0(dmu_objset_from_ds(ds, &os));

	/* Assume all (uncompressed) blocks are recordsize. */
	if (os->os_phys->os_type == DMU_OST_ZVOL) {
		err = dsl_prop_get_int_ds(ds,
		    zfs_prop_to_name(ZFS_PROP_VOLBLOCKSIZE), &recordsize);
	} else {
		err = dsl_prop_get_int_ds(ds,
		    zfs_prop_to_name(ZFS_PROP_RECORDSIZE), &recordsize);
	}
	if (err != 0)
		return (err);
	record_count = uncompressed / recordsize;

	/*
	 * If we're estimating a send size for a compressed stream, use the
	 * compressed data size to estimate the stream size. Otherwise, use the
	 * uncompressed data size.
	 */
	size = stream_compressed ? compressed : uncompressed;

	/*
	 * Subtract out approximate space used by indirect blocks.
	 * Assume most space is used by data blocks (non-indirect, non-dnode).
	 * Assume no ditto blocks or internal fragmentation.
	 *
	 * Therefore, space used by indirect blocks is sizeof(blkptr_t) per
	 * block.
	 */
	size -= record_count * sizeof (blkptr_t);

	/* Add in the space for the record associated with each block. */
	size += record_count * sizeof (dmu_replay_record_t);

	*sizep = size;

	return (0);
}

int
dmu_send_estimate(dsl_dataset_t *ds, dsl_dataset_t *fromds,
    boolean_t stream_compressed, uint64_t *sizep)
{
	int err;
	uint64_t uncomp, comp;

	ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool));

	/* tosnap must be a snapshot */
	if (!ds->ds_is_snapshot)
		return (SET_ERROR(EINVAL));

	/* fromsnap, if provided, must be a snapshot */
	if (fromds != NULL && !fromds->ds_is_snapshot)
		return (SET_ERROR(EINVAL));

	/*
	 * fromsnap must be an earlier snapshot from the same fs as tosnap,
	 * or the origin's fs.
	 */
	if (fromds != NULL && !dsl_dataset_is_before(ds, fromds, 0))
		return (SET_ERROR(EXDEV));

	/* Get compressed and uncompressed size estimates of changed data. */
	if (fromds == NULL) {
		uncomp = dsl_dataset_phys(ds)->ds_uncompressed_bytes;
		comp = dsl_dataset_phys(ds)->ds_compressed_bytes;
	} else {
		uint64_t used;
		err = dsl_dataset_space_written(fromds, ds,
		    &used, &comp, &uncomp);
		if (err != 0)
			return (err);
	}

	err = dmu_adjust_send_estimate_for_indirects(ds, uncomp, comp,
	    stream_compressed, sizep);
	/*
	 * Add the size of the BEGIN and END records to the estimate.
	 */
	*sizep += 2 * sizeof (dmu_replay_record_t);
	return (err);
}

struct calculate_send_arg {
	uint64_t uncompressed;
	uint64_t compressed;
};

/*
 * Simple callback used to traverse the blocks of a snapshot and sum their
 * uncompressed and compressed sizes.
 */
/* ARGSUSED */
static int
dmu_calculate_send_traversal(spa_t *spa, zilog_t *zilog, const blkptr_t *bp,
    const zbookmark_phys_t *zb, const dnode_phys_t *dnp, void *arg)
{
	struct calculate_send_arg *space = arg;
	if (bp != NULL && !BP_IS_HOLE(bp)) {
		space->uncompressed += BP_GET_UCSIZE(bp);
		space->compressed += BP_GET_PSIZE(bp);
	}
	return (0);
}

/*
 * Given a desination snapshot and a TXG, calculate the approximate size of a
 * send stream sent from that TXG. from_txg may be zero, indicating that the
 * whole snapshot will be sent.
 */
int
dmu_send_estimate_from_txg(dsl_dataset_t *ds, uint64_t from_txg,
    boolean_t stream_compressed, uint64_t *sizep)
{
	int err;
	struct calculate_send_arg size = { 0 };

	ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool));

	/* tosnap must be a snapshot */
	if (!dsl_dataset_is_snapshot(ds))
		return (SET_ERROR(EINVAL));

	/* verify that from_txg is before the provided snapshot was taken */
	if (from_txg >= dsl_dataset_phys(ds)->ds_creation_txg) {
		return (SET_ERROR(EXDEV));
	}
	/*
	 * traverse the blocks of the snapshot with birth times after
	 * from_txg, summing their uncompressed size
	 */
	err = traverse_dataset(ds, from_txg,
	    TRAVERSE_POST | TRAVERSE_NO_DECRYPT,
	    dmu_calculate_send_traversal, &size);

	if (err)
		return (err);

	err = dmu_adjust_send_estimate_for_indirects(ds, size.uncompressed,
	    size.compressed, stream_compressed, sizep);
	return (err);
}

typedef struct dmu_recv_begin_arg {
	const char *drba_origin;
	dmu_recv_cookie_t *drba_cookie;
	cred_t *drba_cred;
	uint64_t drba_snapobj;
} dmu_recv_begin_arg_t;

static int
recv_begin_check_existing_impl(dmu_recv_begin_arg_t *drba, dsl_dataset_t *ds,
    uint64_t fromguid)
{
	uint64_t val;
	int error;
	dsl_pool_t *dp = ds->ds_dir->dd_pool;
	struct drr_begin *drrb = drba->drba_cookie->drc_drrb;
	uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo);
	boolean_t encrypted = ds->ds_dir->dd_crypto_obj != 0;
	boolean_t raw = (featureflags & DMU_BACKUP_FEATURE_RAW) != 0;

	/* temporary clone name must not exist */
	error = zap_lookup(dp->dp_meta_objset,
	    dsl_dir_phys(ds->ds_dir)->dd_child_dir_zapobj, recv_clone_name,
	    8, 1, &val);
	if (error != ENOENT)
		return (error == 0 ? EBUSY : error);

	/* new snapshot name must not exist */
	error = zap_lookup(dp->dp_meta_objset,
	    dsl_dataset_phys(ds)->ds_snapnames_zapobj,
	    drba->drba_cookie->drc_tosnap, 8, 1, &val);
	if (error != ENOENT)
		return (error == 0 ? EEXIST : error);

	/*
	 * Check snapshot limit before receiving. We'll recheck again at the
	 * end, but might as well abort before receiving if we're already over
	 * the limit.
	 *
	 * Note that we do not check the file system limit with
	 * dsl_dir_fscount_check because the temporary %clones don't count
	 * against that limit.
	 */
	error = dsl_fs_ss_limit_check(ds->ds_dir, 1, ZFS_PROP_SNAPSHOT_LIMIT,
	    NULL, drba->drba_cred);
	if (error != 0)
		return (error);

	if (fromguid != 0) {
		dsl_dataset_t *snap;
		uint64_t obj = dsl_dataset_phys(ds)->ds_prev_snap_obj;

		/* Can't perform a raw receive on top of a non-raw receive */
		if (!encrypted && raw)
			return (SET_ERROR(EINVAL));

		/* Find snapshot in this dir that matches fromguid. */
		while (obj != 0) {
			error = dsl_dataset_hold_obj(dp, obj, FTAG,
			    &snap);
			if (error != 0)
				return (SET_ERROR(ENODEV));
			if (snap->ds_dir != ds->ds_dir) {
				dsl_dataset_rele(snap, FTAG);
				return (SET_ERROR(ENODEV));
			}
			if (dsl_dataset_phys(snap)->ds_guid == fromguid)
				break;
			obj = dsl_dataset_phys(snap)->ds_prev_snap_obj;
			dsl_dataset_rele(snap, FTAG);
		}
		if (obj == 0)
			return (SET_ERROR(ENODEV));

		if (drba->drba_cookie->drc_force) {
			drba->drba_snapobj = obj;
		} else {
			/*
			 * If we are not forcing, there must be no
			 * changes since fromsnap.
			 */
			if (dsl_dataset_modified_since_snap(ds, snap)) {
				dsl_dataset_rele(snap, FTAG);
				return (SET_ERROR(ETXTBSY));
			}
			drba->drba_snapobj = ds->ds_prev->ds_object;
		}

		dsl_dataset_rele(snap, FTAG);
	} else {
		/* if full, then must be forced */
		if (!drba->drba_cookie->drc_force)
			return (SET_ERROR(EEXIST));

		/*
		 * We don't support using zfs recv -F to blow away
		 * encrypted filesystems. This would require the
		 * dsl dir to point to the old encryption key and
		 * the new one at the same time during the receive.
		 */
		if ((!encrypted && raw) || encrypted)
			return (SET_ERROR(EINVAL));

		drba->drba_snapobj = 0;
	}

	return (0);

}

static int
dmu_recv_begin_check(void *arg, dmu_tx_t *tx)
{
	dmu_recv_begin_arg_t *drba = arg;
	dsl_pool_t *dp = dmu_tx_pool(tx);
	struct drr_begin *drrb = drba->drba_cookie->drc_drrb;
	uint64_t fromguid = drrb->drr_fromguid;
	int flags = drrb->drr_flags;
	ds_hold_flags_t dsflags = 0;
	int error;
	uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo);
	dsl_dataset_t *ds;
	const char *tofs = drba->drba_cookie->drc_tofs;

	/* already checked */
	ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC);
	ASSERT(!(featureflags & DMU_BACKUP_FEATURE_RESUMING));

	if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) ==
	    DMU_COMPOUNDSTREAM ||
	    drrb->drr_type >= DMU_OST_NUMTYPES ||
	    ((flags & DRR_FLAG_CLONE) && drba->drba_origin == NULL))
		return (SET_ERROR(EINVAL));

	/* Verify pool version supports SA if SA_SPILL feature set */
	if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) &&
	    spa_version(dp->dp_spa) < SPA_VERSION_SA)
		return (SET_ERROR(ENOTSUP));

	if (drba->drba_cookie->drc_resumable &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EXTENSIBLE_DATASET))
		return (SET_ERROR(ENOTSUP));

	/*
	 * The receiving code doesn't know how to translate a WRITE_EMBEDDED
	 * record to a plain WRITE record, so the pool must have the
	 * EMBEDDED_DATA feature enabled if the stream has WRITE_EMBEDDED
	 * records.  Same with WRITE_EMBEDDED records that use LZ4 compression.
	 */
	if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA))
		return (SET_ERROR(ENOTSUP));
	if ((featureflags & DMU_BACKUP_FEATURE_LZ4) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS))
		return (SET_ERROR(ENOTSUP));

	/*
	 * The receiving code doesn't know how to translate large blocks
	 * to smaller ones, so the pool must have the LARGE_BLOCKS
	 * feature enabled if the stream has LARGE_BLOCKS. Same with
	 * large dnodes.
	 */
	if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS))
		return (SET_ERROR(ENOTSUP));
	if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_DNODE))
		return (SET_ERROR(ENOTSUP));

	if ((featureflags & DMU_BACKUP_FEATURE_RAW)) {
		/* raw receives require the encryption feature */
		if (!spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_ENCRYPTION))
			return (SET_ERROR(ENOTSUP));
	} else {
		dsflags |= DS_HOLD_FLAG_DECRYPT;
	}

	error = dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds);
	if (error == 0) {
		/* target fs already exists; recv into temp clone */

		/* Can't recv a clone into an existing fs */
		if (flags & DRR_FLAG_CLONE || drba->drba_origin) {
			dsl_dataset_rele_flags(ds, dsflags, FTAG);
			return (SET_ERROR(EINVAL));
		}

		error = recv_begin_check_existing_impl(drba, ds, fromguid);
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
	} else if (error == ENOENT) {
		/* target fs does not exist; must be a full backup or clone */
		char buf[ZFS_MAX_DATASET_NAME_LEN];

		/*
		 * If it's a non-clone incremental, we are missing the
		 * target fs, so fail the recv.
		 */
		if (fromguid != 0 && !(flags & DRR_FLAG_CLONE ||
		    drba->drba_origin))
			return (SET_ERROR(ENOENT));

		/*
		 * If we're receiving a full send as a clone, and it doesn't
		 * contain all the necessary free records and freeobject
		 * records, reject it.
		 */
		if (fromguid == 0 && drba->drba_origin &&
		    !(flags & DRR_FLAG_FREERECORDS))
			return (SET_ERROR(EINVAL));

		/* Open the parent of tofs */
		ASSERT3U(strlen(tofs), <, sizeof (buf));
		(void) strlcpy(buf, tofs, strrchr(tofs, '/') - tofs + 1);
		error = dsl_dataset_hold_flags(dp, buf, dsflags, FTAG, &ds);
		if (error != 0)
			return (error);

		/*
		 * Check filesystem and snapshot limits before receiving. We'll
		 * recheck snapshot limits again at the end (we create the
		 * filesystems and increment those counts during begin_sync).
		 */
		error = dsl_fs_ss_limit_check(ds->ds_dir, 1,
		    ZFS_PROP_FILESYSTEM_LIMIT, NULL, drba->drba_cred);
		if (error != 0) {
			dsl_dataset_rele_flags(ds, dsflags, FTAG);
			return (error);
		}

		error = dsl_fs_ss_limit_check(ds->ds_dir, 1,
		    ZFS_PROP_SNAPSHOT_LIMIT, NULL, drba->drba_cred);
		if (error != 0) {
			dsl_dataset_rele_flags(ds, dsflags, FTAG);
			return (error);
		}

		if (drba->drba_origin != NULL) {
			dsl_dataset_t *origin;

			error = dsl_dataset_hold_flags(dp, drba->drba_origin,
			    dsflags, FTAG, &origin);
			if (error != 0) {
				dsl_dataset_rele_flags(ds, dsflags, FTAG);
				return (error);
			}
			if (!origin->ds_is_snapshot) {
				dsl_dataset_rele_flags(origin, dsflags, FTAG);
				dsl_dataset_rele_flags(ds, dsflags, FTAG);
				return (SET_ERROR(EINVAL));
			}
			if (dsl_dataset_phys(origin)->ds_guid != fromguid &&
			    fromguid != 0) {
				dsl_dataset_rele_flags(origin, dsflags, FTAG);
				dsl_dataset_rele_flags(ds, dsflags, FTAG);
				return (SET_ERROR(ENODEV));
			}
			dsl_dataset_rele_flags(origin,
			    dsflags, FTAG);
		}
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		error = 0;
	}
	return (error);
}

static void
dmu_recv_begin_sync(void *arg, dmu_tx_t *tx)
{
	dmu_recv_begin_arg_t *drba = arg;
	dsl_pool_t *dp = dmu_tx_pool(tx);
	objset_t *mos = dp->dp_meta_objset;
	struct drr_begin *drrb = drba->drba_cookie->drc_drrb;
	const char *tofs = drba->drba_cookie->drc_tofs;
	uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo);
	dsl_dataset_t *ds, *newds;
	objset_t *os;
	uint64_t dsobj;
	ds_hold_flags_t dsflags = 0;
	int error;
	uint64_t crflags = 0;
	dsl_crypto_params_t *dcpp = NULL;
	dsl_crypto_params_t dcp = { 0 };

	if (drrb->drr_flags & DRR_FLAG_CI_DATA)
		crflags |= DS_FLAG_CI_DATASET;
	if ((featureflags & DMU_BACKUP_FEATURE_RAW) == 0) {
		dsflags |= DS_HOLD_FLAG_DECRYPT;
	} else {
		dcp.cp_cmd = DCP_CMD_RAW_RECV;
	}

	error = dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds);
	if (error == 0) {
		/* create temporary clone */
		dsl_dataset_t *snap = NULL;

		if (drba->drba_snapobj != 0) {
			VERIFY0(dsl_dataset_hold_obj(dp,
			    drba->drba_snapobj, FTAG, &snap));
		} else {
			/* we use the dcp whenever we are not making a clone */
			dcpp = &dcp;
		}

		dsobj = dsl_dataset_create_sync(ds->ds_dir, recv_clone_name,
		    snap, crflags, drba->drba_cred, dcpp, tx);
		if (drba->drba_snapobj != 0)
			dsl_dataset_rele(snap, FTAG);
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
	} else {
		dsl_dir_t *dd;
		const char *tail;
		dsl_dataset_t *origin = NULL;

		VERIFY0(dsl_dir_hold(dp, tofs, FTAG, &dd, &tail));

		if (drba->drba_origin != NULL) {
			VERIFY0(dsl_dataset_hold(dp, drba->drba_origin,
			    FTAG, &origin));
		} else {
			/* we use the dcp whenever we are not making a clone */
			dcpp = &dcp;
		}

		/* Create new dataset. */
		dsobj = dsl_dataset_create_sync(dd, strrchr(tofs, '/') + 1,
		    origin, crflags, drba->drba_cred, dcpp, tx);
		if (origin != NULL)
			dsl_dataset_rele(origin, FTAG);
		dsl_dir_rele(dd, FTAG);
		drba->drba_cookie->drc_newfs = B_TRUE;
	}
	VERIFY0(dsl_dataset_own_obj(dp, dsobj, dsflags, dmu_recv_tag, &newds));
	VERIFY0(dmu_objset_from_ds(newds, &os));

	if (drba->drba_cookie->drc_resumable) {
		dsl_dataset_zapify(newds, tx);
		if (drrb->drr_fromguid != 0) {
			VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_FROMGUID,
			    8, 1, &drrb->drr_fromguid, tx));
		}
		VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TOGUID,
		    8, 1, &drrb->drr_toguid, tx));
		VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TONAME,
		    1, strlen(drrb->drr_toname) + 1, drrb->drr_toname, tx));
		uint64_t one = 1;
		uint64_t zero = 0;
		VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OBJECT,
		    8, 1, &one, tx));
		VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OFFSET,
		    8, 1, &zero, tx));
		VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_BYTES,
		    8, 1, &zero, tx));
		if (featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) {
			VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_LARGEBLOCK,
			    8, 1, &one, tx));
		}
		if (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) {
			VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_EMBEDOK,
			    8, 1, &one, tx));
		}
		if (featureflags & DMU_BACKUP_FEATURE_COMPRESSED) {
			VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_COMPRESSOK,
			    8, 1, &one, tx));
		}
		if (featureflags & DMU_BACKUP_FEATURE_RAW) {
			VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_RAWOK,
			    8, 1, &one, tx));
		}
	}

	/*
	 * Usually the os->os_encrypted value is tied to the presence of a
	 * DSL Crypto Key object in the dd. However, that will not be received
	 * until dmu_recv_stream(), so we set the value manually for now.
	 */
	if (featureflags & DMU_BACKUP_FEATURE_RAW) {
		os->os_encrypted = B_TRUE;
		drba->drba_cookie->drc_raw = B_TRUE;
	}

	dmu_buf_will_dirty(newds->ds_dbuf, tx);
	dsl_dataset_phys(newds)->ds_flags |= DS_FLAG_INCONSISTENT;

	/*
	 * If we actually created a non-clone, we need to create the objset
	 * in our new dataset. If this is a raw send we postpone this until
	 * dmu_recv_stream() so that we can allocate the metadnode with the
	 * properties from the DRR_BEGIN payload.
	 */
	rrw_enter(&newds->ds_bp_rwlock, RW_READER, FTAG);
	if (BP_IS_HOLE(dsl_dataset_get_blkptr(newds)) &&
	    (featureflags & DMU_BACKUP_FEATURE_RAW) == 0) {
		(void) dmu_objset_create_impl(dp->dp_spa,
		    newds, dsl_dataset_get_blkptr(newds), drrb->drr_type, tx);
	}
	rrw_exit(&newds->ds_bp_rwlock, FTAG);

	drba->drba_cookie->drc_ds = newds;

	spa_history_log_internal_ds(newds, "receive", tx, "");
}

static int
dmu_recv_resume_begin_check(void *arg, dmu_tx_t *tx)
{
	dmu_recv_begin_arg_t *drba = arg;
	dsl_pool_t *dp = dmu_tx_pool(tx);
	struct drr_begin *drrb = drba->drba_cookie->drc_drrb;
	int error;
	ds_hold_flags_t dsflags = 0;
	uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo);
	dsl_dataset_t *ds;
	const char *tofs = drba->drba_cookie->drc_tofs;

	/* already checked */
	ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC);
	ASSERT(featureflags & DMU_BACKUP_FEATURE_RESUMING);

	if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) ==
	    DMU_COMPOUNDSTREAM ||
	    drrb->drr_type >= DMU_OST_NUMTYPES)
		return (SET_ERROR(EINVAL));

	/* Verify pool version supports SA if SA_SPILL feature set */
	if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) &&
	    spa_version(dp->dp_spa) < SPA_VERSION_SA)
		return (SET_ERROR(ENOTSUP));

	/*
	 * The receiving code doesn't know how to translate a WRITE_EMBEDDED
	 * record to a plain WRITE record, so the pool must have the
	 * EMBEDDED_DATA feature enabled if the stream has WRITE_EMBEDDED
	 * records.  Same with WRITE_EMBEDDED records that use LZ4 compression.
	 */
	if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA))
		return (SET_ERROR(ENOTSUP));
	if ((featureflags & DMU_BACKUP_FEATURE_LZ4) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS))
		return (SET_ERROR(ENOTSUP));

	/*
	 * The receiving code doesn't know how to translate large blocks
	 * to smaller ones, so the pool must have the LARGE_BLOCKS
	 * feature enabled if the stream has LARGE_BLOCKS. Same with
	 * large dnodes.
	 */
	if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS))
		return (SET_ERROR(ENOTSUP));
	if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) &&
	    !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_DNODE))
		return (SET_ERROR(ENOTSUP));

	/* 6 extra bytes for /%recv */
	char recvname[ZFS_MAX_DATASET_NAME_LEN + 6];
	(void) snprintf(recvname, sizeof (recvname), "%s/%s",
	    tofs, recv_clone_name);

	if ((featureflags & DMU_BACKUP_FEATURE_RAW) == 0)
		dsflags |= DS_HOLD_FLAG_DECRYPT;

	if (dsl_dataset_hold_flags(dp, recvname, dsflags, FTAG, &ds) != 0) {
		/* %recv does not exist; continue in tofs */
		error = dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds);
		if (error != 0)
			return (error);
	}

	/* check that ds is marked inconsistent */
	if (!DS_IS_INCONSISTENT(ds)) {
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		return (SET_ERROR(EINVAL));
	}

	/* check that there is resuming data, and that the toguid matches */
	if (!dsl_dataset_is_zapified(ds)) {
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		return (SET_ERROR(EINVAL));
	}
	uint64_t val;
	error = zap_lookup(dp->dp_meta_objset, ds->ds_object,
	    DS_FIELD_RESUME_TOGUID, sizeof (val), 1, &val);
	if (error != 0 || drrb->drr_toguid != val) {
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		return (SET_ERROR(EINVAL));
	}

	/*
	 * Check if the receive is still running.  If so, it will be owned.
	 * Note that nothing else can own the dataset (e.g. after the receive
	 * fails) because it will be marked inconsistent.
	 */
	if (dsl_dataset_has_owner(ds)) {
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		return (SET_ERROR(EBUSY));
	}

	/* There should not be any snapshots of this fs yet. */
	if (ds->ds_prev != NULL && ds->ds_prev->ds_dir == ds->ds_dir) {
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		return (SET_ERROR(EINVAL));
	}

	/*
	 * Note: resume point will be checked when we process the first WRITE
	 * record.
	 */

	/* check that the origin matches */
	val = 0;
	(void) zap_lookup(dp->dp_meta_objset, ds->ds_object,
	    DS_FIELD_RESUME_FROMGUID, sizeof (val), 1, &val);
	if (drrb->drr_fromguid != val) {
		dsl_dataset_rele_flags(ds, dsflags, FTAG);
		return (SET_ERROR(EINVAL));
	}

	dsl_dataset_rele_flags(ds, dsflags, FTAG);
	return (0);
}

static void
dmu_recv_resume_begin_sync(void *arg, dmu_tx_t *tx)
{
	dmu_recv_begin_arg_t *drba = arg;
	dsl_pool_t *dp = dmu_tx_pool(tx);
	const char *tofs = drba->drba_cookie->drc_tofs;
	struct drr_begin *drrb = drba->drba_cookie->drc_drrb;
	uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo);
	dsl_dataset_t *ds;
	objset_t *os;
	ds_hold_flags_t dsflags = 0;
	uint64_t dsobj;
	/* 6 extra bytes for /%recv */
	char recvname[ZFS_MAX_DATASET_NAME_LEN + 6];

	(void) snprintf(recvname, sizeof (recvname), "%s/%s",
	    tofs, recv_clone_name);

	if (featureflags & DMU_BACKUP_FEATURE_RAW) {
		drba->drba_cookie->drc_raw = B_TRUE;
	} else {
		dsflags |= DS_HOLD_FLAG_DECRYPT;
	}

	if (dsl_dataset_hold_flags(dp, recvname, dsflags, FTAG, &ds) != 0) {
		/* %recv does not exist; continue in tofs */
		VERIFY0(dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds));
		drba->drba_cookie->drc_newfs = B_TRUE;
	}

	/* clear the inconsistent flag so that we can own it */
	ASSERT(DS_IS_INCONSISTENT(ds));
	dmu_buf_will_dirty(ds->ds_dbuf, tx);
	dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT;
	dsobj = ds->ds_object;
	dsl_dataset_rele_flags(ds, dsflags, FTAG);

	VERIFY0(dsl_dataset_own_obj(dp, dsobj, dsflags, dmu_recv_tag, &ds));
	VERIFY0(dmu_objset_from_ds(ds, &os));

	dmu_buf_will_dirty(ds->ds_dbuf, tx);
	dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_INCONSISTENT;

	rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG);
	ASSERT(!BP_IS_HOLE(dsl_dataset_get_blkptr(ds)) ||
	    drba->drba_cookie->drc_raw);
	rrw_exit(&ds->ds_bp_rwlock, FTAG);

	drba->drba_cookie->drc_ds = ds;

	spa_history_log_internal_ds(ds, "resume receive", tx, "");
}

/*
 * NB: callers *MUST* call dmu_recv_stream() if dmu_recv_begin()
 * succeeds; otherwise we will leak the holds on the datasets.
 */
int
dmu_recv_begin(char *tofs, char *tosnap, dmu_replay_record_t *drr_begin,
    boolean_t force, boolean_t resumable, char *origin, dmu_recv_cookie_t *drc)
{
	dmu_recv_begin_arg_t drba = { 0 };

	bzero(drc, sizeof (dmu_recv_cookie_t));
	drc->drc_drr_begin = drr_begin;
	drc->drc_drrb = &drr_begin->drr_u.drr_begin;
	drc->drc_tosnap = tosnap;
	drc->drc_tofs = tofs;
	drc->drc_force = force;
	drc->drc_resumable = resumable;
	drc->drc_cred = CRED();
	drc->drc_clone = (origin != NULL);

	if (drc->drc_drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) {
		drc->drc_byteswap = B_TRUE;
		(void) fletcher_4_incremental_byteswap(drr_begin,
		    sizeof (dmu_replay_record_t), &drc->drc_cksum);
		byteswap_record(drr_begin);
	} else if (drc->drc_drrb->drr_magic == DMU_BACKUP_MAGIC) {
		(void) fletcher_4_incremental_native(drr_begin,
		    sizeof (dmu_replay_record_t), &drc->drc_cksum);
	} else {
		return (SET_ERROR(EINVAL));
	}

	drba.drba_origin = origin;
	drba.drba_cookie = drc;
	drba.drba_cred = CRED();

	if (DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo) &
	    DMU_BACKUP_FEATURE_RESUMING) {
		return (dsl_sync_task(tofs,
		    dmu_recv_resume_begin_check, dmu_recv_resume_begin_sync,
		    &drba, 5, ZFS_SPACE_CHECK_NORMAL));
	} else  {
		return (dsl_sync_task(tofs,
		    dmu_recv_begin_check, dmu_recv_begin_sync,
		    &drba, 5, ZFS_SPACE_CHECK_NORMAL));
	}
}

struct receive_record_arg {
	dmu_replay_record_t header;
	void *payload; /* Pointer to a buffer containing the payload */
	/*
	 * If the record is a write, pointer to the arc_buf_t containing the
	 * payload.
	 */
	arc_buf_t *arc_buf;
	int payload_size;
	uint64_t bytes_read; /* bytes read from stream when record created */
	boolean_t eos_marker; /* Marks the end of the stream */
	bqueue_node_t node;
};

struct receive_writer_arg {
	objset_t *os;
	boolean_t byteswap;
	bqueue_t q;

	/*
	 * These three args are used to signal to the main thread that we're
	 * done.
	 */
	kmutex_t mutex;
	kcondvar_t cv;
	boolean_t done;

	int err;
	/* A map from guid to dataset to help handle dedup'd streams. */
	avl_tree_t *guid_to_ds_map;
	boolean_t resumable;
	boolean_t raw;
	uint64_t last_object;
	uint64_t last_offset;
	uint64_t max_object; /* highest object ID referenced in stream */
	uint64_t bytes_read; /* bytes read when current record created */

	/* Encryption parameters for the last received DRR_OBJECT_RANGE */
	uint64_t or_firstobj;
	uint64_t or_numslots;
	uint8_t or_salt[ZIO_DATA_SALT_LEN];
	uint8_t or_iv[ZIO_DATA_IV_LEN];
	uint8_t or_mac[ZIO_DATA_MAC_LEN];
	boolean_t or_byteorder;
};

struct objlist {
	list_t list; /* List of struct receive_objnode. */
	/*
	 * Last object looked up. Used to assert that objects are being looked
	 * up in ascending order.
	 */
	uint64_t last_lookup;
};

struct receive_objnode {
	list_node_t node;
	uint64_t object;
};

struct receive_arg  {
	objset_t *os;
	vnode_t *vp; /* The vnode to read the stream from */
	uint64_t voff; /* The current offset in the stream */
	uint64_t bytes_read;
	/*
	 * A record that has had its payload read in, but hasn't yet been handed
	 * off to the worker thread.
	 */
	struct receive_record_arg *rrd;
	/* A record that has had its header read in, but not its payload. */
	struct receive_record_arg *next_rrd;
	zio_cksum_t cksum;
	zio_cksum_t prev_cksum;
	int err;
	boolean_t byteswap;
	boolean_t raw;
	uint64_t featureflags;
	/* Sorted list of objects not to issue prefetches for. */
	struct objlist ignore_objlist;
};

typedef struct guid_map_entry {
	uint64_t	guid;
	boolean_t	raw;
	dsl_dataset_t	*gme_ds;
	avl_node_t	avlnode;
} guid_map_entry_t;

static int
guid_compare(const void *arg1, const void *arg2)
{
	const guid_map_entry_t *gmep1 = (const guid_map_entry_t *)arg1;
	const guid_map_entry_t *gmep2 = (const guid_map_entry_t *)arg2;

	return (AVL_CMP(gmep1->guid, gmep2->guid));
}

static void
free_guid_map_onexit(void *arg)
{
	avl_tree_t *ca = arg;
	void *cookie = NULL;
	guid_map_entry_t *gmep;

	while ((gmep = avl_destroy_nodes(ca, &cookie)) != NULL) {
		dsl_dataset_long_rele(gmep->gme_ds, gmep);
		dsl_dataset_rele_flags(gmep->gme_ds,
		    (gmep->raw) ? 0 : DS_HOLD_FLAG_DECRYPT, gmep);
		kmem_free(gmep, sizeof (guid_map_entry_t));
	}
	avl_destroy(ca);
	kmem_free(ca, sizeof (avl_tree_t));
}

static int
receive_read(struct receive_arg *ra, int len, void *buf)
{
	int done = 0;

	/*
	 * The code doesn't rely on this (lengths being multiples of 8).  See
	 * comment in dump_bytes.
	 */
	ASSERT(len % 8 == 0 ||
	    (ra->featureflags & DMU_BACKUP_FEATURE_RAW) != 0);

	while (done < len) {
		ssize_t resid;

		ra->err = vn_rdwr(UIO_READ, ra->vp,
		    (char *)buf + done, len - done,
		    ra->voff, UIO_SYSSPACE, FAPPEND,
		    RLIM64_INFINITY, CRED(), &resid);

		if (resid == len - done) {
			/*
			 * Note: ECKSUM indicates that the receive
			 * was interrupted and can potentially be resumed.
			 */
			ra->err = SET_ERROR(ECKSUM);
		}
		ra->voff += len - done - resid;
		done = len - resid;
		if (ra->err != 0)
			return (ra->err);
	}

	ra->bytes_read += len;

	ASSERT3U(done, ==, len);
	return (0);
}

noinline static void
byteswap_record(dmu_replay_record_t *drr)
{
#define	DO64(X) (drr->drr_u.X = BSWAP_64(drr->drr_u.X))
#define	DO32(X) (drr->drr_u.X = BSWAP_32(drr->drr_u.X))
	drr->drr_type = BSWAP_32(drr->drr_type);
	drr->drr_payloadlen = BSWAP_32(drr->drr_payloadlen);

	switch (drr->drr_type) {
	case DRR_BEGIN:
		DO64(drr_begin.drr_magic);
		DO64(drr_begin.drr_versioninfo);
		DO64(drr_begin.drr_creation_time);
		DO32(drr_begin.drr_type);
		DO32(drr_begin.drr_flags);
		DO64(drr_begin.drr_toguid);
		DO64(drr_begin.drr_fromguid);
		break;
	case DRR_OBJECT:
		DO64(drr_object.drr_object);
		DO32(drr_object.drr_type);
		DO32(drr_object.drr_bonustype);
		DO32(drr_object.drr_blksz);
		DO32(drr_object.drr_bonuslen);
		DO32(drr_object.drr_raw_bonuslen);
		DO64(drr_object.drr_toguid);
		DO64(drr_object.drr_maxblkid);
		break;
	case DRR_FREEOBJECTS:
		DO64(drr_freeobjects.drr_firstobj);
		DO64(drr_freeobjects.drr_numobjs);
		DO64(drr_freeobjects.drr_toguid);
		break;
	case DRR_WRITE:
		DO64(drr_write.drr_object);
		DO32(drr_write.drr_type);
		DO64(drr_write.drr_offset);
		DO64(drr_write.drr_logical_size);
		DO64(drr_write.drr_toguid);
		ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write.drr_key.ddk_cksum);
		DO64(drr_write.drr_key.ddk_prop);
		DO64(drr_write.drr_compressed_size);
		break;
	case DRR_WRITE_BYREF:
		DO64(drr_write_byref.drr_object);
		DO64(drr_write_byref.drr_offset);
		DO64(drr_write_byref.drr_length);
		DO64(drr_write_byref.drr_toguid);
		DO64(drr_write_byref.drr_refguid);
		DO64(drr_write_byref.drr_refobject);
		DO64(drr_write_byref.drr_refoffset);
		ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write_byref.
		    drr_key.ddk_cksum);
		DO64(drr_write_byref.drr_key.ddk_prop);
		break;
	case DRR_WRITE_EMBEDDED:
		DO64(drr_write_embedded.drr_object);
		DO64(drr_write_embedded.drr_offset);
		DO64(drr_write_embedded.drr_length);
		DO64(drr_write_embedded.drr_toguid);
		DO32(drr_write_embedded.drr_lsize);
		DO32(drr_write_embedded.drr_psize);
		break;
	case DRR_FREE:
		DO64(drr_free.drr_object);
		DO64(drr_free.drr_offset);
		DO64(drr_free.drr_length);
		DO64(drr_free.drr_toguid);
		break;
	case DRR_SPILL:
		DO64(drr_spill.drr_object);
		DO64(drr_spill.drr_length);
		DO64(drr_spill.drr_toguid);
		DO64(drr_spill.drr_compressed_size);
		DO32(drr_spill.drr_type);
		break;
	case DRR_OBJECT_RANGE:
		DO64(drr_object_range.drr_firstobj);
		DO64(drr_object_range.drr_numslots);
		DO64(drr_object_range.drr_toguid);
		break;
	case DRR_END:
		DO64(drr_end.drr_toguid);
		ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_end.drr_checksum);
		break;
	default:
		break;
	}

	if (drr->drr_type != DRR_BEGIN) {
		ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_checksum.drr_checksum);
	}

#undef DO64
#undef DO32
}

static inline uint8_t
deduce_nblkptr(dmu_object_type_t bonus_type, uint64_t bonus_size)
{
	if (bonus_type == DMU_OT_SA) {
		return (1);
	} else {
		return (1 +
		    ((DN_OLD_MAX_BONUSLEN -
		    MIN(DN_OLD_MAX_BONUSLEN, bonus_size)) >> SPA_BLKPTRSHIFT));
	}
}

static void
save_resume_state(struct receive_writer_arg *rwa,
    uint64_t object, uint64_t offset, dmu_tx_t *tx)
{
	int txgoff = dmu_tx_get_txg(tx) & TXG_MASK;

	if (!rwa->resumable)
		return;

	/*
	 * We use ds_resume_bytes[] != 0 to indicate that we need to
	 * update this on disk, so it must not be 0.
	 */
	ASSERT(rwa->bytes_read != 0);

	/*
	 * We only resume from write records, which have a valid
	 * (non-meta-dnode) object number.
	 */
	ASSERT(object != 0);

	/*
	 * For resuming to work correctly, we must receive records in order,
	 * sorted by object,offset.  This is checked by the callers, but
	 * assert it here for good measure.
	 */
	ASSERT3U(object, >=, rwa->os->os_dsl_dataset->ds_resume_object[txgoff]);
	ASSERT(object != rwa->os->os_dsl_dataset->ds_resume_object[txgoff] ||
	    offset >= rwa->os->os_dsl_dataset->ds_resume_offset[txgoff]);
	ASSERT3U(rwa->bytes_read, >=,
	    rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff]);

	rwa->os->os_dsl_dataset->ds_resume_object[txgoff] = object;
	rwa->os->os_dsl_dataset->ds_resume_offset[txgoff] = offset;
	rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff] = rwa->bytes_read;
}

noinline static int
receive_object(struct receive_writer_arg *rwa, struct drr_object *drro,
    void *data)
{
	dmu_object_info_t doi;
	dmu_tx_t *tx;
	uint64_t object;
	int err;

	if (drro->drr_type == DMU_OT_NONE ||
	    !DMU_OT_IS_VALID(drro->drr_type) ||
	    !DMU_OT_IS_VALID(drro->drr_bonustype) ||
	    drro->drr_checksumtype >= ZIO_CHECKSUM_FUNCTIONS ||
	    drro->drr_compress >= ZIO_COMPRESS_FUNCTIONS ||
	    P2PHASE(drro->drr_blksz, SPA_MINBLOCKSIZE) ||
	    drro->drr_blksz < SPA_MINBLOCKSIZE ||
	    drro->drr_blksz > spa_maxblocksize(dmu_objset_spa(rwa->os)) ||
	    drro->drr_bonuslen >
	    DN_BONUS_SIZE(spa_maxdnodesize(dmu_objset_spa(rwa->os))) ||
	    drro->drr_dn_slots >
	    (spa_maxdnodesize(dmu_objset_spa(rwa->os)) >> DNODE_SHIFT))  {
		return (SET_ERROR(EINVAL));
	}

	if (rwa->raw) {
		/*
		 * We should have received a DRR_OBJECT_RANGE record
		 * containing this block and stored it in rwa.
		 */
		if (drro->drr_object < rwa->or_firstobj ||
		    drro->drr_object >= rwa->or_firstobj + rwa->or_numslots ||
		    drro->drr_raw_bonuslen < drro->drr_bonuslen ||
		    drro->drr_indblkshift > SPA_MAXBLOCKSHIFT ||
		    drro->drr_nlevels > DN_MAX_LEVELS ||
		    drro->drr_nblkptr > DN_MAX_NBLKPTR ||
		    DN_SLOTS_TO_BONUSLEN(drro->drr_dn_slots) <
		    drro->drr_raw_bonuslen)
			return (SET_ERROR(EINVAL));
	} else {
		if (drro->drr_flags != 0 || drro->drr_raw_bonuslen != 0 ||
		    drro->drr_indblkshift != 0 || drro->drr_nlevels != 0 ||
		    drro->drr_nblkptr != 0)
			return (SET_ERROR(EINVAL));
	}

	err = dmu_object_info(rwa->os, drro->drr_object, &doi);
	if (err != 0 && err != ENOENT && err != EEXIST)
		return (SET_ERROR(EINVAL));

	if (drro->drr_object > rwa->max_object)
		rwa->max_object = drro->drr_object;

	/*
	 * If we are losing blkptrs or changing the block size this must
	 * be a new file instance.  We must clear out the previous file
	 * contents before we can change this type of metadata in the dnode.
	 * Raw receives will also check that the indirect structure of the
	 * dnode hasn't changed.
	 */
	if (err == 0) {
		uint32_t indblksz = drro->drr_indblkshift ?
		    1ULL << drro->drr_indblkshift : 0;
		int nblkptr = deduce_nblkptr(drro->drr_bonustype,
		    drro->drr_bonuslen);

		object = drro->drr_object;

		/* nblkptr will be bounded by the bonus size and type */
		if (rwa->raw && nblkptr != drro->drr_nblkptr)
			return (SET_ERROR(EINVAL));

		if (rwa->raw &&
		    (drro->drr_blksz != doi.doi_data_block_size ||
		    nblkptr < doi.doi_nblkptr ||
		    indblksz != doi.doi_metadata_block_size ||
		    drro->drr_nlevels < doi.doi_indirection ||
		    drro->drr_dn_slots != doi.doi_dnodesize >> DNODE_SHIFT)) {
			err = dmu_free_long_range_raw(rwa->os,
			    drro->drr_object, 0, DMU_OBJECT_END);
			if (err != 0)
				return (SET_ERROR(EINVAL));
		} else if (drro->drr_blksz != doi.doi_data_block_size ||
		    nblkptr < doi.doi_nblkptr ||
		    drro->drr_dn_slots != doi.doi_dnodesize >> DNODE_SHIFT) {
			err = dmu_free_long_range(rwa->os, drro->drr_object,
			    0, DMU_OBJECT_END);
			if (err != 0)
				return (SET_ERROR(EINVAL));
		}

		/*
		 * The dmu does not currently support decreasing nlevels
		 * on an object. For non-raw sends, this does not matter
		 * and the new object can just use the previous one's nlevels.
		 * For raw sends, however, the structure of the received dnode
		 * (including nlevels) must match that of the send side.
		 * Therefore, instead of using dmu_object_reclaim(), we must
		 * free the object completely and call dmu_object_claim_dnsize()
		 * instead.
		 */
		if ((rwa->raw && drro->drr_nlevels < doi.doi_indirection) ||
		    drro->drr_dn_slots != doi.doi_dnodesize >> DNODE_SHIFT) {
			if (rwa->raw) {
				err = dmu_free_long_object_raw(rwa->os,
				    drro->drr_object);
			} else {
				err = dmu_free_long_object(rwa->os,
				    drro->drr_object);
			}
			if (err != 0)
				return (SET_ERROR(EINVAL));

			txg_wait_synced(dmu_objset_pool(rwa->os), 0);
			object = DMU_NEW_OBJECT;
		}
	} else if (err == EEXIST) {
		/*
		 * The object requested is currently an interior slot of a
		 * multi-slot dnode. This will be resolved when the next txg
		 * is synced out, since the send stream will have told us
		 * to free this slot when we freed the associated dnode
		 * earlier in the stream.
		 */
		txg_wait_synced(dmu_objset_pool(rwa->os), 0);
		object = drro->drr_object;
	} else {
		/* object is free and we are about to allocate a new one */
		object = DMU_NEW_OBJECT;
	}

	/*
	 * If this is a multi-slot dnode there is a chance that this
	 * object will expand into a slot that is already used by
	 * another object from the previous snapshot. We must free
	 * these objects before we attempt to allocate the new dnode.
	 */
	if (drro->drr_dn_slots > 1) {
		boolean_t need_sync = B_FALSE;

		for (uint64_t slot = drro->drr_object + 1;
		    slot < drro->drr_object + drro->drr_dn_slots;
		    slot++) {
			dmu_object_info_t slot_doi;

			err = dmu_object_info(rwa->os, slot, &slot_doi);
			if (err == ENOENT || err == EEXIST)
				continue;
			else if (err != 0)
				return (err);

			if (rwa->raw)
				err = dmu_free_long_object_raw(rwa->os, slot);
			else
				err = dmu_free_long_object(rwa->os, slot);

			if (err != 0)
				return (err);

			need_sync = B_TRUE;
		}

		if (need_sync)
			txg_wait_synced(dmu_objset_pool(rwa->os), 0);
	}

	tx = dmu_tx_create(rwa->os);
	dmu_tx_hold_bonus(tx, object);
	dmu_tx_hold_write(tx, object, 0, 0);
	err = dmu_tx_assign(tx, TXG_WAIT);
	if (err != 0) {
		dmu_tx_abort(tx);
		return (err);
	}

	if (object == DMU_NEW_OBJECT) {
		/* currently free, want to be allocated */
		err = dmu_object_claim_dnsize(rwa->os, drro->drr_object,
		    drro->drr_type, drro->drr_blksz,
		    drro->drr_bonustype, drro->drr_bonuslen,
		    drro->drr_dn_slots << DNODE_SHIFT, tx);
	} else if (drro->drr_type != doi.doi_type ||
	    drro->drr_blksz != doi.doi_data_block_size ||
	    drro->drr_bonustype != doi.doi_bonus_type ||
	    drro->drr_bonuslen != doi.doi_bonus_size) {
		/* currently allocated, but with different properties */
		err = dmu_object_reclaim(rwa->os, drro->drr_object,
		    drro->drr_type, drro->drr_blksz,
		    drro->drr_bonustype, drro->drr_bonuslen, tx);
	}
	if (err != 0) {
		dmu_tx_commit(tx);
		return (SET_ERROR(EINVAL));
	}

	if (rwa->raw) {
		/*
		 * Convert the buffer associated with this range of dnodes
		 * to a raw buffer. This ensures that it will be written out
		 * as a raw buffer when we fill in the dnode object. Since we
		 * are committing this tx now, it is possible for the dnode
		 * block to end up on-disk with the incorrect MAC. Despite
		 * this, the dataset is marked as inconsistent so no other
		 * code paths (apart from scrubs) will attempt to read this
		 * data. Scrubs will not be effected by this either since
		 * scrubs only read raw data and do not attempt to check
		 * the MAC.
		 */
		err = dmu_convert_mdn_block_to_raw(rwa->os, rwa->or_firstobj,
		    rwa->or_byteorder, rwa->or_salt, rwa->or_iv, rwa->or_mac,
		    tx);
		if (err != 0) {
			dmu_tx_commit(tx);
			return (SET_ERROR(EINVAL));
		}
	}

	dmu_object_set_checksum(rwa->os, drro->drr_object,
	    drro->drr_checksumtype, tx);
	dmu_object_set_compress(rwa->os, drro->drr_object,
	    drro->drr_compress, tx);

	/* handle more restrictive dnode structuring for raw recvs */
	if (rwa->raw) {
		/*
		 * Set the indirect block shift and nlevels. This will not fail
		 * because we ensured all of the blocks were free earlier if
		 * this is a new object.
		 */
		VERIFY0(dmu_object_set_blocksize(rwa->os, drro->drr_object,
		    drro->drr_blksz, drro->drr_indblkshift, tx));
		VERIFY0(dmu_object_set_nlevels(rwa->os, drro->drr_object,
		    drro->drr_nlevels, tx));
		VERIFY0(dmu_object_set_maxblkid(rwa->os, drro->drr_object,
		    drro->drr_maxblkid, tx));
	}

	if (data != NULL) {
		dmu_buf_t *db;
		uint32_t flags = DMU_READ_NO_PREFETCH;

		if (rwa->raw)
			flags |= DMU_READ_NO_DECRYPT;

		VERIFY0(dmu_bonus_hold_impl(rwa->os, drro->drr_object,
		    FTAG, flags, &db));
		dmu_buf_will_dirty(db, tx);

		ASSERT3U(db->db_size, >=, drro->drr_bonuslen);
		bcopy(data, db->db_data, DRR_OBJECT_PAYLOAD_SIZE(drro));

		/*
		 * Raw bonus buffers have their byteorder determined by the
		 * DRR_OBJECT_RANGE record.
		 */
		if (rwa->byteswap && !rwa->raw) {
			dmu_object_byteswap_t byteswap =
			    DMU_OT_BYTESWAP(drro->drr_bonustype);
			dmu_ot_byteswap[byteswap].ob_func(db->db_data,
			    DRR_OBJECT_PAYLOAD_SIZE(drro));
		}
		dmu_buf_rele(db, FTAG);
	}
	dmu_tx_commit(tx);

	return (0);
}

/* ARGSUSED */
noinline static int
receive_freeobjects(struct receive_writer_arg *rwa,
    struct drr_freeobjects *drrfo)
{
	uint64_t obj;
	int next_err = 0;

	if (drrfo->drr_firstobj + drrfo->drr_numobjs < drrfo->drr_firstobj)
		return (SET_ERROR(EINVAL));

	for (obj = drrfo->drr_firstobj == 0 ? 1 : drrfo->drr_firstobj;
	    obj < drrfo->drr_firstobj + drrfo->drr_numobjs && next_err == 0;
	    next_err = dmu_object_next(rwa->os, &obj, FALSE, 0)) {
		dmu_object_info_t doi;
		int err;

		err = dmu_object_info(rwa->os, obj, &doi);
		if (err == ENOENT)
			continue;
		else if (err != 0)
			return (err);

		if (rwa->raw)
			err = dmu_free_long_object_raw(rwa->os, obj);
		else
			err = dmu_free_long_object(rwa->os, obj);

		if (err != 0)
			return (err);

		if (obj > rwa->max_object)
			rwa->max_object = obj;
	}
	if (next_err != ESRCH)
		return (next_err);
	return (0);
}

noinline static int
receive_write(struct receive_writer_arg *rwa, struct drr_write *drrw,
    arc_buf_t *abuf)
{
	int err;
	dmu_tx_t *tx;
	dnode_t *dn;

	if (drrw->drr_offset + drrw->drr_logical_size < drrw->drr_offset ||
	    !DMU_OT_IS_VALID(drrw->drr_type))
		return (SET_ERROR(EINVAL));

	/*
	 * For resuming to work, records must be in increasing order
	 * by (object, offset).
	 */
	if (drrw->drr_object < rwa->last_object ||
	    (drrw->drr_object == rwa->last_object &&
	    drrw->drr_offset < rwa->last_offset)) {
		return (SET_ERROR(EINVAL));
	}
	rwa->last_object = drrw->drr_object;
	rwa->last_offset = drrw->drr_offset;

	if (rwa->last_object > rwa->max_object)
		rwa->max_object = rwa->last_object;

	if (dmu_object_info(rwa->os, drrw->drr_object, NULL) != 0)
		return (SET_ERROR(EINVAL));

	tx = dmu_tx_create(rwa->os);
	dmu_tx_hold_write(tx, drrw->drr_object,
	    drrw->drr_offset, drrw->drr_logical_size);
	err = dmu_tx_assign(tx, TXG_WAIT);
	if (err != 0) {
		dmu_tx_abort(tx);
		return (err);
	}

	if (rwa->raw)
		VERIFY0(dmu_object_dirty_raw(rwa->os, drrw->drr_object, tx));

	if (rwa->byteswap && !arc_is_encrypted(abuf) &&
	    arc_get_compression(abuf) == ZIO_COMPRESS_OFF) {
		dmu_object_byteswap_t byteswap =
		    DMU_OT_BYTESWAP(drrw->drr_type);
		dmu_ot_byteswap[byteswap].ob_func(abuf->b_data,
		    DRR_WRITE_PAYLOAD_SIZE(drrw));
	}

	VERIFY0(dnode_hold(rwa->os, drrw->drr_object, FTAG, &dn));
	dmu_assign_arcbuf_by_dnode(dn, drrw->drr_offset, abuf, tx);
	dnode_rele(dn, FTAG);

	/*
	 * Note: If the receive fails, we want the resume stream to start
	 * with the same record that we last successfully received (as opposed
	 * to the next record), so that we can verify that we are
	 * resuming from the correct location.
	 */
	save_resume_state(rwa, drrw->drr_object, drrw->drr_offset, tx);
	dmu_tx_commit(tx);

	return (0);
}

/*
 * Handle a DRR_WRITE_BYREF record.  This record is used in dedup'ed
 * streams to refer to a copy of the data that is already on the
 * system because it came in earlier in the stream.  This function
 * finds the earlier copy of the data, and uses that copy instead of
 * data from the stream to fulfill this write.
 */
static int
receive_write_byref(struct receive_writer_arg *rwa,
    struct drr_write_byref *drrwbr)
{
	dmu_tx_t *tx;
	int err;
	guid_map_entry_t gmesrch;
	guid_map_entry_t *gmep;
	avl_index_t where;
	objset_t *ref_os = NULL;
	int flags = DMU_READ_PREFETCH;
	dmu_buf_t *dbp;

	if (drrwbr->drr_offset + drrwbr->drr_length < drrwbr->drr_offset)
		return (SET_ERROR(EINVAL));

	/*
	 * If the GUID of the referenced dataset is different from the
	 * GUID of the target dataset, find the referenced dataset.
	 */
	if (drrwbr->drr_toguid != drrwbr->drr_refguid) {
		gmesrch.guid = drrwbr->drr_refguid;
		if ((gmep = avl_find(rwa->guid_to_ds_map, &gmesrch,
		    &where)) == NULL) {
			return (SET_ERROR(EINVAL));
		}
		if (dmu_objset_from_ds(gmep->gme_ds, &ref_os))
			return (SET_ERROR(EINVAL));
	} else {
		ref_os = rwa->os;
	}

	if (drrwbr->drr_object > rwa->max_object)
		rwa->max_object = drrwbr->drr_object;

	if (rwa->raw)
		flags |= DMU_READ_NO_DECRYPT;

	/* may return either a regular db or an encrypted one */
	err = dmu_buf_hold(ref_os, drrwbr->drr_refobject,
	    drrwbr->drr_refoffset, FTAG, &dbp, flags);
	if (err != 0)
		return (err);

	tx = dmu_tx_create(rwa->os);

	dmu_tx_hold_write(tx, drrwbr->drr_object,
	    drrwbr->drr_offset, drrwbr->drr_length);
	err = dmu_tx_assign(tx, TXG_WAIT);
	if (err != 0) {
		dmu_tx_abort(tx);
		return (err);
	}

	if (rwa->raw) {
		VERIFY0(dmu_object_dirty_raw(rwa->os, drrwbr->drr_object, tx));
		dmu_copy_from_buf(rwa->os, drrwbr->drr_object,
		    drrwbr->drr_offset, dbp, tx);
	} else {
		dmu_write(rwa->os, drrwbr->drr_object,
		    drrwbr->drr_offset, drrwbr->drr_length, dbp->db_data, tx);
	}
	dmu_buf_rele(dbp, FTAG);

	/* See comment in restore_write. */
	save_resume_state(rwa, drrwbr->drr_object, drrwbr->drr_offset, tx);
	dmu_tx_commit(tx);
	return (0);
}

static int
receive_write_embedded(struct receive_writer_arg *rwa,
    struct drr_write_embedded *drrwe, void *data)
{
	dmu_tx_t *tx;
	int err;

	if (drrwe->drr_offset + drrwe->drr_length < drrwe->drr_offset)
		return (SET_ERROR(EINVAL));

	if (drrwe->drr_psize > BPE_PAYLOAD_SIZE)
		return (SET_ERROR(EINVAL));

	if (drrwe->drr_etype >= NUM_BP_EMBEDDED_TYPES)
		return (SET_ERROR(EINVAL));
	if (drrwe->drr_compression >= ZIO_COMPRESS_FUNCTIONS)
		return (SET_ERROR(EINVAL));
	if (rwa->raw)
		return (SET_ERROR(EINVAL));

	if (drrwe->drr_object > rwa->max_object)
		rwa->max_object = drrwe->drr_object;

	tx = dmu_tx_create(rwa->os);

	dmu_tx_hold_write(tx, drrwe->drr_object,
	    drrwe->drr_offset, drrwe->drr_length);
	err = dmu_tx_assign(tx, TXG_WAIT);
	if (err != 0) {
		dmu_tx_abort(tx);
		return (err);
	}

	dmu_write_embedded(rwa->os, drrwe->drr_object,
	    drrwe->drr_offset, data, drrwe->drr_etype,
	    drrwe->drr_compression, drrwe->drr_lsize, drrwe->drr_psize,
	    rwa->byteswap ^ ZFS_HOST_BYTEORDER, tx);

	/* See comment in restore_write. */
	save_resume_state(rwa, drrwe->drr_object, drrwe->drr_offset, tx);
	dmu_tx_commit(tx);
	return (0);
}

static int
receive_spill(struct receive_writer_arg *rwa, struct drr_spill *drrs,
    arc_buf_t *abuf)
{
	dmu_tx_t *tx;
	dmu_buf_t *db, *db_spill;
	int err;

	if (drrs->drr_length < SPA_MINBLOCKSIZE ||
	    drrs->drr_length > spa_maxblocksize(dmu_objset_spa(rwa->os)))
		return (SET_ERROR(EINVAL));

	if (rwa->raw) {
		if (!DMU_OT_IS_VALID(drrs->drr_type) ||
		    drrs->drr_compressiontype >= ZIO_COMPRESS_FUNCTIONS ||
		    drrs->drr_compressed_size == 0)
			return (SET_ERROR(EINVAL));
	}

	if (dmu_object_info(rwa->os, drrs->drr_object, NULL) != 0)
		return (SET_ERROR(EINVAL));

	if (drrs->drr_object > rwa->max_object)
		rwa->max_object = drrs->drr_object;

	VERIFY0(dmu_bonus_hold(rwa->os, drrs->drr_object, FTAG, &db));
	if ((err = dmu_spill_hold_by_bonus(db, FTAG, &db_spill)) != 0) {
		dmu_buf_rele(db, FTAG);
		return (err);
	}

	tx = dmu_tx_create(rwa->os);

	dmu_tx_hold_spill(tx, db->db_object);

	err = dmu_tx_assign(tx, TXG_WAIT);
	if (err != 0) {
		dmu_buf_rele(db, FTAG);
		dmu_buf_rele(db_spill, FTAG);
		dmu_tx_abort(tx);
		return (err);
	}

	if (rwa->raw) {
		VERIFY0(dmu_object_dirty_raw(rwa->os, drrs->drr_object, tx));
		dmu_buf_will_change_crypt_params(db_spill, tx);
	} else {
		dmu_buf_will_dirty(db_spill, tx);
	}

	if (db_spill->db_size < drrs->drr_length)
		VERIFY(0 == dbuf_spill_set_blksz(db_spill,
		    drrs->drr_length, tx));

	if (rwa->byteswap && !arc_is_encrypted(abuf) &&
	    arc_get_compression(abuf) == ZIO_COMPRESS_OFF) {
		dmu_object_byteswap_t byteswap =
		    DMU_OT_BYTESWAP(drrs->drr_type);
		dmu_ot_byteswap[byteswap].ob_func(abuf->b_data,
		    DRR_SPILL_PAYLOAD_SIZE(drrs));
	}

	dbuf_assign_arcbuf((dmu_buf_impl_t *)db_spill, abuf, tx);

	dmu_buf_rele(db, FTAG);
	dmu_buf_rele(db_spill, FTAG);

	dmu_tx_commit(tx);
	return (0);
}

/* ARGSUSED */
noinline static int
receive_free(struct receive_writer_arg *rwa, struct drr_free *drrf)
{
	int err;

	if (drrf->drr_length != DMU_OBJECT_END &&
	    drrf->drr_offset + drrf->drr_length < drrf->drr_offset)
		return (SET_ERROR(EINVAL));

	if (dmu_object_info(rwa->os, drrf->drr_object, NULL) != 0)
		return (SET_ERROR(EINVAL));

	if (drrf->drr_object > rwa->max_object)
		rwa->max_object = drrf->drr_object;

	if (rwa->raw) {
		err = dmu_free_long_range_raw(rwa->os, drrf->drr_object,
		    drrf->drr_offset, drrf->drr_length);
	} else {
		err = dmu_free_long_range(rwa->os, drrf->drr_object,
		    drrf->drr_offset, drrf->drr_length);
	}

	return (err);
}

static int
receive_object_range(struct receive_writer_arg *rwa,
    struct drr_object_range *drror)
{
	/*
	 * By default, we assume this block is in our native format
	 * (ZFS_HOST_BYTEORDER). We then take into account whether
	 * the send stream is byteswapped (rwa->byteswap). Finally,
	 * we need to byteswap again if this particular block was
	 * in non-native format on the send side.
	 */
	boolean_t byteorder = ZFS_HOST_BYTEORDER ^ rwa->byteswap ^
	    !!DRR_IS_RAW_BYTESWAPPED(drror->drr_flags);

	/*
	 * Since dnode block sizes are constant, we should not need to worry
	 * about making sure that the dnode block size is the same on the
	 * sending and receiving sides for the time being. For non-raw sends,
	 * this does not matter (and in fact we do not send a DRR_OBJECT_RANGE
	 * record at all). Raw sends require this record type because the
	 * encryption parameters are used to protect an entire block of bonus
	 * buffers. If the size of dnode blocks ever becomes variable,
	 * handling will need to be added to ensure that dnode block sizes
	 * match on the sending and receiving side.
	 */
	if (drror->drr_numslots != DNODES_PER_BLOCK ||
	    P2PHASE(drror->drr_firstobj, DNODES_PER_BLOCK) != 0 ||
	    !rwa->raw)
		return (SET_ERROR(EINVAL));

	if (drror->drr_firstobj > rwa->max_object)
		rwa->max_object = drror->drr_firstobj;

	/*
	 * The DRR_OBJECT_RANGE handling must be deferred to receive_object()
	 * so that the encryption parameters are set with each object that is
	 * written into that block.
	 */
	rwa->or_firstobj = drror->drr_firstobj;
	rwa->or_numslots = drror->drr_numslots;
	bcopy(drror->drr_salt, rwa->or_salt, ZIO_DATA_SALT_LEN);
	bcopy(drror->drr_iv, rwa->or_iv, ZIO_DATA_IV_LEN);
	bcopy(drror->drr_mac, rwa->or_mac, ZIO_DATA_MAC_LEN);
	rwa->or_byteorder = byteorder;

	return (0);
}

/* used to destroy the drc_ds on error */
static void
dmu_recv_cleanup_ds(dmu_recv_cookie_t *drc)
{
	dsl_dataset_t *ds = drc->drc_ds;
	ds_hold_flags_t dsflags = (drc->drc_raw) ? 0 : DS_HOLD_FLAG_DECRYPT;

	/*
	 * Wait for the txg sync before cleaning up the receive. For
	 * resumable receives, this ensures that our resume state has
	 * been written out to disk. For raw receives, this ensures
	 * that the user accounting code will not attempt to do anything
	 * after we stopped receiving the dataset.
	 */
	txg_wait_synced(ds->ds_dir->dd_pool, 0);

	rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG);
	if (drc->drc_resumable && !BP_IS_HOLE(dsl_dataset_get_blkptr(ds))) {
		rrw_exit(&ds->ds_bp_rwlock, FTAG);
		dsl_dataset_disown(ds, dsflags, dmu_recv_tag);
	} else {
		char name[ZFS_MAX_DATASET_NAME_LEN];
		rrw_exit(&ds->ds_bp_rwlock, FTAG);
		dsl_dataset_name(ds, name);
		dsl_dataset_disown(ds, dsflags, dmu_recv_tag);
		(void) dsl_destroy_head(name);
	}
}

static void
receive_cksum(struct receive_arg *ra, int len, void *buf)
{
	if (ra->byteswap) {
		(void) fletcher_4_incremental_byteswap(buf, len, &ra->cksum);
	} else {
		(void) fletcher_4_incremental_native(buf, len, &ra->cksum);
	}
}

/*
 * Read the payload into a buffer of size len, and update the current record's
 * payload field.
 * Allocate ra->next_rrd and read the next record's header into
 * ra->next_rrd->header.
 * Verify checksum of payload and next record.
 */
static int
receive_read_payload_and_next_header(struct receive_arg *ra, int len, void *buf)
{
	int err;
	zio_cksum_t cksum_orig;
	zio_cksum_t *cksump;

	if (len != 0) {
		ASSERT3U(len, <=, SPA_MAXBLOCKSIZE);
		err = receive_read(ra, len, buf);
		if (err != 0)
			return (err);
		receive_cksum(ra, len, buf);

		/* note: rrd is NULL when reading the begin record's payload */
		if (ra->rrd != NULL) {
			ra->rrd->payload = buf;
			ra->rrd->payload_size = len;
			ra->rrd->bytes_read = ra->bytes_read;
		}
	}

	ra->prev_cksum = ra->cksum;

	ra->next_rrd = kmem_zalloc(sizeof (*ra->next_rrd), KM_SLEEP);
	err = receive_read(ra, sizeof (ra->next_rrd->header),
	    &ra->next_rrd->header);
	ra->next_rrd->bytes_read = ra->bytes_read;

	if (err != 0) {
		kmem_free(ra->next_rrd, sizeof (*ra->next_rrd));
		ra->next_rrd = NULL;
		return (err);
	}
	if (ra->next_rrd->header.drr_type == DRR_BEGIN) {
		kmem_free(ra->next_rrd, sizeof (*ra->next_rrd));
		ra->next_rrd = NULL;
		return (SET_ERROR(EINVAL));
	}

	/*
	 * Note: checksum is of everything up to but not including the
	 * checksum itself.
	 */
	ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum),
	    ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t));
	receive_cksum(ra,
	    offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum),
	    &ra->next_rrd->header);

	cksum_orig = ra->next_rrd->header.drr_u.drr_checksum.drr_checksum;
	cksump = &ra->next_rrd->header.drr_u.drr_checksum.drr_checksum;

	if (ra->byteswap)
		byteswap_record(&ra->next_rrd->header);

	if ((!ZIO_CHECKSUM_IS_ZERO(cksump)) &&
	    !ZIO_CHECKSUM_EQUAL(ra->cksum, *cksump)) {
		kmem_free(ra->next_rrd, sizeof (*ra->next_rrd));
		ra->next_rrd = NULL;
		return (SET_ERROR(ECKSUM));
	}

	receive_cksum(ra, sizeof (cksum_orig), &cksum_orig);

	return (0);
}

static void
objlist_create(struct objlist *list)
{
	list_create(&list->list, sizeof (struct receive_objnode),
	    offsetof(struct receive_objnode, node));
	list->last_lookup = 0;
}

static void
objlist_destroy(struct objlist *list)
{
	for (struct receive_objnode *n = list_remove_head(&list->list);
	    n != NULL; n = list_remove_head(&list->list)) {
		kmem_free(n, sizeof (*n));
	}
	list_destroy(&list->list);
}

/*
 * This function looks through the objlist to see if the specified object number
 * is contained in the objlist.  In the process, it will remove all object
 * numbers in the list that are smaller than the specified object number.  Thus,
 * any lookup of an object number smaller than a previously looked up object
 * number will always return false; therefore, all lookups should be done in
 * ascending order.
 */
static boolean_t
objlist_exists(struct objlist *list, uint64_t object)
{
	struct receive_objnode *node = list_head(&list->list);
	ASSERT3U(object, >=, list->last_lookup);
	list->last_lookup = object;
	while (node != NULL && node->object < object) {
		VERIFY3P(node, ==, list_remove_head(&list->list));
		kmem_free(node, sizeof (*node));
		node = list_head(&list->list);
	}
	return (node != NULL && node->object == object);
}

/*
 * The objlist is a list of object numbers stored in ascending order.  However,
 * the insertion of new object numbers does not seek out the correct location to
 * store a new object number; instead, it appends it to the list for simplicity.
 * Thus, any users must take care to only insert new object numbers in ascending
 * order.
 */
static void
objlist_insert(struct objlist *list, uint64_t object)
{
	struct receive_objnode *node = kmem_zalloc(sizeof (*node), KM_SLEEP);
	node->object = object;
#ifdef ZFS_DEBUG
	{
	struct receive_objnode *last_object = list_tail(&list->list);
	uint64_t last_objnum = (last_object != NULL ? last_object->object : 0);
	ASSERT3U(node->object, >, last_objnum);
	}
#endif
	list_insert_tail(&list->list, node);
}

/*
 * Issue the prefetch reads for any necessary indirect blocks.
 *
 * We use the object ignore list to tell us whether or not to issue prefetches
 * for a given object.  We do this for both correctness (in case the blocksize
 * of an object has changed) and performance (if the object doesn't exist, don't
 * needlessly try to issue prefetches).  We also trim the list as we go through
 * the stream to prevent it from growing to an unbounded size.
 *
 * The object numbers within will always be in sorted order, and any write
 * records we see will also be in sorted order, but they're not sorted with
 * respect to each other (i.e. we can get several object records before
 * receiving each object's write records).  As a result, once we've reached a
 * given object number, we can safely remove any reference to lower object
 * numbers in the ignore list. In practice, we receive up to 32 object records
 * before receiving write records, so the list can have up to 32 nodes in it.
 */
/* ARGSUSED */
static void
receive_read_prefetch(struct receive_arg *ra,
    uint64_t object, uint64_t offset, uint64_t length)
{
	if (!objlist_exists(&ra->ignore_objlist, object)) {
		dmu_prefetch(ra->os, object, 1, offset, length,
		    ZIO_PRIORITY_SYNC_READ);
	}
}

/*
 * Read records off the stream, issuing any necessary prefetches.
 */
static int
receive_read_record(struct receive_arg *ra)
{
	int err;

	switch (ra->rrd->header.drr_type) {
	case DRR_OBJECT:
	{
		struct drr_object *drro = &ra->rrd->header.drr_u.drr_object;
		uint32_t size = DRR_OBJECT_PAYLOAD_SIZE(drro);
		void *buf = kmem_zalloc(size, KM_SLEEP);
		dmu_object_info_t doi;

		err = receive_read_payload_and_next_header(ra, size, buf);
		if (err != 0) {
			kmem_free(buf, size);
			return (err);
		}
		err = dmu_object_info(ra->os, drro->drr_object, &doi);
		/*
		 * See receive_read_prefetch for an explanation why we're
		 * storing this object in the ignore_obj_list.
		 */
		if (err == ENOENT || err == EEXIST ||
		    (err == 0 && doi.doi_data_block_size != drro->drr_blksz)) {
			objlist_insert(&ra->ignore_objlist, drro->drr_object);
			err = 0;
		}
		return (err);
	}
	case DRR_FREEOBJECTS:
	{
		err = receive_read_payload_and_next_header(ra, 0, NULL);
		return (err);
	}
	case DRR_WRITE:
	{
		struct drr_write *drrw = &ra->rrd->header.drr_u.drr_write;
		arc_buf_t *abuf;
		boolean_t is_meta = DMU_OT_IS_METADATA(drrw->drr_type);

		if (ra->raw) {
			boolean_t byteorder = ZFS_HOST_BYTEORDER ^
			    !!DRR_IS_RAW_BYTESWAPPED(drrw->drr_flags) ^
			    ra->byteswap;

			abuf = arc_loan_raw_buf(dmu_objset_spa(ra->os),
			    drrw->drr_object, byteorder, drrw->drr_salt,
			    drrw->drr_iv, drrw->drr_mac, drrw->drr_type,
			    drrw->drr_compressed_size, drrw->drr_logical_size,
			    drrw->drr_compressiontype);
		} else if (DRR_WRITE_COMPRESSED(drrw)) {
			ASSERT3U(drrw->drr_compressed_size, >, 0);
			ASSERT3U(drrw->drr_logical_size, >=,
			    drrw->drr_compressed_size);
			ASSERT(!is_meta);
			abuf = arc_loan_compressed_buf(
			    dmu_objset_spa(ra->os),
			    drrw->drr_compressed_size, drrw->drr_logical_size,
			    drrw->drr_compressiontype);
		} else {
			abuf = arc_loan_buf(dmu_objset_spa(ra->os),
			    is_meta, drrw->drr_logical_size);
		}

		err = receive_read_payload_and_next_header(ra,
		    DRR_WRITE_PAYLOAD_SIZE(drrw), abuf->b_data);
		if (err != 0) {
			dmu_return_arcbuf(abuf);
			return (err);
		}
		ra->rrd->arc_buf = abuf;
		receive_read_prefetch(ra, drrw->drr_object, drrw->drr_offset,
		    drrw->drr_logical_size);
		return (err);
	}
	case DRR_WRITE_BYREF:
	{
		struct drr_write_byref *drrwb =
		    &ra->rrd->header.drr_u.drr_write_byref;
		err = receive_read_payload_and_next_header(ra, 0, NULL);
		receive_read_prefetch(ra, drrwb->drr_object, drrwb->drr_offset,
		    drrwb->drr_length);
		return (err);
	}
	case DRR_WRITE_EMBEDDED:
	{
		struct drr_write_embedded *drrwe =
		    &ra->rrd->header.drr_u.drr_write_embedded;
		uint32_t size = P2ROUNDUP(drrwe->drr_psize, 8);
		void *buf = kmem_zalloc(size, KM_SLEEP);

		err = receive_read_payload_and_next_header(ra, size, buf);
		if (err != 0) {
			kmem_free(buf, size);
			return (err);
		}

		receive_read_prefetch(ra, drrwe->drr_object, drrwe->drr_offset,
		    drrwe->drr_length);
		return (err);
	}
	case DRR_FREE:
	{
		/*
		 * It might be beneficial to prefetch indirect blocks here, but
		 * we don't really have the data to decide for sure.
		 */
		err = receive_read_payload_and_next_header(ra, 0, NULL);
		return (err);
	}
	case DRR_END:
	{
		struct drr_end *drre = &ra->rrd->header.drr_u.drr_end;
		if (!ZIO_CHECKSUM_EQUAL(ra->prev_cksum, drre->drr_checksum))
			return (SET_ERROR(ECKSUM));
		return (0);
	}
	case DRR_SPILL:
	{
		struct drr_spill *drrs = &ra->rrd->header.drr_u.drr_spill;
		arc_buf_t *abuf;
		int len = DRR_SPILL_PAYLOAD_SIZE(drrs);

		/* DRR_SPILL records are either raw or uncompressed */
		if (ra->raw) {
			boolean_t byteorder = ZFS_HOST_BYTEORDER ^
			    !!DRR_IS_RAW_BYTESWAPPED(drrs->drr_flags) ^
			    ra->byteswap;

			abuf = arc_loan_raw_buf(dmu_objset_spa(ra->os),
			    drrs->drr_object, byteorder, drrs->drr_salt,
			    drrs->drr_iv, drrs->drr_mac, drrs->drr_type,
			    drrs->drr_compressed_size, drrs->drr_length,
			    drrs->drr_compressiontype);
		} else {
			abuf = arc_loan_buf(dmu_objset_spa(ra->os),
			    DMU_OT_IS_METADATA(drrs->drr_type),
			    drrs->drr_length);
		}

		err = receive_read_payload_and_next_header(ra, len,
		    abuf->b_data);
		if (err != 0) {
			dmu_return_arcbuf(abuf);
			return (err);
		}
		ra->rrd->arc_buf = abuf;
		return (err);
	}
	case DRR_OBJECT_RANGE:
	{
		err = receive_read_payload_and_next_header(ra, 0, NULL);
		return (err);
	}
	default:
		return (SET_ERROR(EINVAL));
	}
}

static void
dprintf_drr(struct receive_record_arg *rrd, int err)
{
#ifdef ZFS_DEBUG
	switch (rrd->header.drr_type) {
	case DRR_OBJECT:
	{
		struct drr_object *drro = &rrd->header.drr_u.drr_object;
		dprintf("drr_type = OBJECT obj = %llu type = %u "
		    "bonustype = %u blksz = %u bonuslen = %u cksumtype = %u "
		    "compress = %u dn_slots = %u err = %d\n",
		    drro->drr_object, drro->drr_type,  drro->drr_bonustype,
		    drro->drr_blksz, drro->drr_bonuslen,
		    drro->drr_checksumtype, drro->drr_compress,
		    drro->drr_dn_slots, err);
		break;
	}
	case DRR_FREEOBJECTS:
	{
		struct drr_freeobjects *drrfo =
		    &rrd->header.drr_u.drr_freeobjects;
		dprintf("drr_type = FREEOBJECTS firstobj = %llu "
		    "numobjs = %llu err = %d\n",
		    drrfo->drr_firstobj, drrfo->drr_numobjs, err);
		break;
	}
	case DRR_WRITE:
	{
		struct drr_write *drrw = &rrd->header.drr_u.drr_write;
		dprintf("drr_type = WRITE obj = %llu type = %u offset = %llu "
		    "lsize = %llu cksumtype = %u cksumflags = %u "
		    "compress = %u psize = %llu err = %d\n",
		    drrw->drr_object, drrw->drr_type, drrw->drr_offset,
		    drrw->drr_logical_size, drrw->drr_checksumtype,
		    drrw->drr_flags, drrw->drr_compressiontype,
		    drrw->drr_compressed_size, err);
		break;
	}
	case DRR_WRITE_BYREF:
	{
		struct drr_write_byref *drrwbr =
		    &rrd->header.drr_u.drr_write_byref;
		dprintf("drr_type = WRITE_BYREF obj = %llu offset = %llu "
		    "length = %llu toguid = %llx refguid = %llx "
		    "refobject = %llu refoffset = %llu cksumtype = %u "
		    "cksumflags = %u err = %d\n",
		    drrwbr->drr_object, drrwbr->drr_offset,
		    drrwbr->drr_length, drrwbr->drr_toguid,
		    drrwbr->drr_refguid, drrwbr->drr_refobject,
		    drrwbr->drr_refoffset, drrwbr->drr_checksumtype,
		    drrwbr->drr_flags, err);
		break;
	}
	case DRR_WRITE_EMBEDDED:
	{
		struct drr_write_embedded *drrwe =
		    &rrd->header.drr_u.drr_write_embedded;
		dprintf("drr_type = WRITE_EMBEDDED obj = %llu offset = %llu "
		    "length = %llu compress = %u etype = %u lsize = %u "
		    "psize = %u err = %d\n",
		    drrwe->drr_object, drrwe->drr_offset, drrwe->drr_length,
		    drrwe->drr_compression, drrwe->drr_etype,
		    drrwe->drr_lsize, drrwe->drr_psize, err);
		break;
	}
	case DRR_FREE:
	{
		struct drr_free *drrf = &rrd->header.drr_u.drr_free;
		dprintf("drr_type = FREE obj = %llu offset = %llu "
		    "length = %lld err = %d\n",
		    drrf->drr_object, drrf->drr_offset, drrf->drr_length,
		    err);
		break;
	}
	case DRR_SPILL:
	{
		struct drr_spill *drrs = &rrd->header.drr_u.drr_spill;
		dprintf("drr_type = SPILL obj = %llu length = %llu "
		    "err = %d\n", drrs->drr_object, drrs->drr_length, err);
		break;
	}
	default:
		return;
	}
#endif
}

/*
 * Commit the records to the pool.
 */
static int
receive_process_record(struct receive_writer_arg *rwa,
    struct receive_record_arg *rrd)
{
	int err;

	/* Processing in order, therefore bytes_read should be increasing. */
	ASSERT3U(rrd->bytes_read, >=, rwa->bytes_read);
	rwa->bytes_read = rrd->bytes_read;

	switch (rrd->header.drr_type) {
	case DRR_OBJECT:
	{
		struct drr_object *drro = &rrd->header.drr_u.drr_object;
		err = receive_object(rwa, drro, rrd->payload);
		kmem_free(rrd->payload, rrd->payload_size);
		rrd->payload = NULL;
		break;
	}
	case DRR_FREEOBJECTS:
	{
		struct drr_freeobjects *drrfo =
		    &rrd->header.drr_u.drr_freeobjects;
		err = receive_freeobjects(rwa, drrfo);
		break;
	}
	case DRR_WRITE:
	{
		struct drr_write *drrw = &rrd->header.drr_u.drr_write;
		err = receive_write(rwa, drrw, rrd->arc_buf);
		/* if receive_write() is successful, it consumes the arc_buf */
		if (err != 0)
			dmu_return_arcbuf(rrd->arc_buf);
		rrd->arc_buf = NULL;
		rrd->payload = NULL;
		break;
	}
	case DRR_WRITE_BYREF:
	{
		struct drr_write_byref *drrwbr =
		    &rrd->header.drr_u.drr_write_byref;
		err = receive_write_byref(rwa, drrwbr);
		break;
	}
	case DRR_WRITE_EMBEDDED:
	{
		struct drr_write_embedded *drrwe =
		    &rrd->header.drr_u.drr_write_embedded;
		err = receive_write_embedded(rwa, drrwe, rrd->payload);
		kmem_free(rrd->payload, rrd->payload_size);
		rrd->payload = NULL;
		break;
	}
	case DRR_FREE:
	{
		struct drr_free *drrf = &rrd->header.drr_u.drr_free;
		err = receive_free(rwa, drrf);
		break;
	}
	case DRR_SPILL:
	{
		struct drr_spill *drrs = &rrd->header.drr_u.drr_spill;
		err = receive_spill(rwa, drrs, rrd->arc_buf);
		/* if receive_spill() is successful, it consumes the arc_buf */
		if (err != 0)
			dmu_return_arcbuf(rrd->arc_buf);
		rrd->arc_buf = NULL;
		rrd->payload = NULL;
		break;
	}
	case DRR_OBJECT_RANGE:
	{
		struct drr_object_range *drror =
		    &rrd->header.drr_u.drr_object_range;
		return (receive_object_range(rwa, drror));
	}
	default:
		return (SET_ERROR(EINVAL));
	}

	if (err != 0)
		dprintf_drr(rrd, err);

	return (err);
}

/*
 * dmu_recv_stream's worker thread; pull records off the queue, and then call
 * receive_process_record  When we're done, signal the main thread and exit.
 */
static void
receive_writer_thread(void *arg)
{
	struct receive_writer_arg *rwa = arg;
	struct receive_record_arg *rrd;
	fstrans_cookie_t cookie = spl_fstrans_mark();

	for (rrd = bqueue_dequeue(&rwa->q); !rrd->eos_marker;
	    rrd = bqueue_dequeue(&rwa->q)) {
		/*
		 * If there's an error, the main thread will stop putting things
		 * on the queue, but we need to clear everything in it before we
		 * can exit.
		 */
		if (rwa->err == 0) {
			rwa->err = receive_process_record(rwa, rrd);
		} else if (rrd->arc_buf != NULL) {
			dmu_return_arcbuf(rrd->arc_buf);
			rrd->arc_buf = NULL;
			rrd->payload = NULL;
		} else if (rrd->payload != NULL) {
			kmem_free(rrd->payload, rrd->payload_size);
			rrd->payload = NULL;
		}
		kmem_free(rrd, sizeof (*rrd));
	}
	kmem_free(rrd, sizeof (*rrd));
	mutex_enter(&rwa->mutex);
	rwa->done = B_TRUE;
	cv_signal(&rwa->cv);
	mutex_exit(&rwa->mutex);
	spl_fstrans_unmark(cookie);
	thread_exit();
}

static int
resume_check(struct receive_arg *ra, nvlist_t *begin_nvl)
{
	uint64_t val;
	objset_t *mos = dmu_objset_pool(ra->os)->dp_meta_objset;
	uint64_t dsobj = dmu_objset_id(ra->os);
	uint64_t resume_obj, resume_off;

	if (nvlist_lookup_uint64(begin_nvl,
	    "resume_object", &resume_obj) != 0 ||
	    nvlist_lookup_uint64(begin_nvl,
	    "resume_offset", &resume_off) != 0) {
		return (SET_ERROR(EINVAL));
	}
	VERIFY0(zap_lookup(mos, dsobj,
	    DS_FIELD_RESUME_OBJECT, sizeof (val), 1, &val));
	if (resume_obj != val)
		return (SET_ERROR(EINVAL));
	VERIFY0(zap_lookup(mos, dsobj,
	    DS_FIELD_RESUME_OFFSET, sizeof (val), 1, &val));
	if (resume_off != val)
		return (SET_ERROR(EINVAL));

	return (0);
}

/*
 * Read in the stream's records, one by one, and apply them to the pool.  There
 * are two threads involved; the thread that calls this function will spin up a
 * worker thread, read the records off the stream one by one, and issue
 * prefetches for any necessary indirect blocks.  It will then push the records
 * onto an internal blocking queue.  The worker thread will pull the records off
 * the queue, and actually write the data into the DMU.  This way, the worker
 * thread doesn't have to wait for reads to complete, since everything it needs
 * (the indirect blocks) will be prefetched.
 *
 * NB: callers *must* call dmu_recv_end() if this succeeds.
 */
int
dmu_recv_stream(dmu_recv_cookie_t *drc, vnode_t *vp, offset_t *voffp,
    int cleanup_fd, uint64_t *action_handlep)
{
	int err = 0;
	struct receive_arg *ra;
	struct receive_writer_arg *rwa;
	int featureflags;
	uint32_t payloadlen;
	void *payload;
	nvlist_t *begin_nvl = NULL;

	ra = kmem_zalloc(sizeof (*ra), KM_SLEEP);
	rwa = kmem_zalloc(sizeof (*rwa), KM_SLEEP);

	ra->byteswap = drc->drc_byteswap;
	ra->raw = drc->drc_raw;
	ra->cksum = drc->drc_cksum;
	ra->vp = vp;
	ra->voff = *voffp;

	if (dsl_dataset_is_zapified(drc->drc_ds)) {
		(void) zap_lookup(drc->drc_ds->ds_dir->dd_pool->dp_meta_objset,
		    drc->drc_ds->ds_object, DS_FIELD_RESUME_BYTES,
		    sizeof (ra->bytes_read), 1, &ra->bytes_read);
	}

	objlist_create(&ra->ignore_objlist);

	/* these were verified in dmu_recv_begin */
	ASSERT3U(DMU_GET_STREAM_HDRTYPE(drc->drc_drrb->drr_versioninfo), ==,
	    DMU_SUBSTREAM);
	ASSERT3U(drc->drc_drrb->drr_type, <, DMU_OST_NUMTYPES);

	/*
	 * Open the objset we are modifying.
	 */
	VERIFY0(dmu_objset_from_ds(drc->drc_ds, &ra->os));

	ASSERT(dsl_dataset_phys(drc->drc_ds)->ds_flags & DS_FLAG_INCONSISTENT);

	featureflags = DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo);
	ra->featureflags = featureflags;

	/* embedded data is incompatible with encrypted datasets */
	if (ra->os->os_encrypted &&
	    (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA)) {
		err = SET_ERROR(EINVAL);
		goto out;
	}

	/* if this stream is dedup'ed, set up the avl tree for guid mapping */
	if (featureflags & DMU_BACKUP_FEATURE_DEDUP) {
		minor_t minor;

		if (cleanup_fd == -1) {
			err = SET_ERROR(EBADF);
			goto out;
		}
		err = zfs_onexit_fd_hold(cleanup_fd, &minor);
		if (err != 0) {
			cleanup_fd = -1;
			goto out;
		}

		if (*action_handlep == 0) {
			rwa->guid_to_ds_map =
			    kmem_alloc(sizeof (avl_tree_t), KM_SLEEP);
			avl_create(rwa->guid_to_ds_map, guid_compare,
			    sizeof (guid_map_entry_t),
			    offsetof(guid_map_entry_t, avlnode));
			err = zfs_onexit_add_cb(minor,
			    free_guid_map_onexit, rwa->guid_to_ds_map,
			    action_handlep);
			if (err != 0)
				goto out;
		} else {
			err = zfs_onexit_cb_data(minor, *action_handlep,
			    (void **)&rwa->guid_to_ds_map);
			if (err != 0)
				goto out;
		}

		drc->drc_guid_to_ds_map = rwa->guid_to_ds_map;
	}

	payloadlen = drc->drc_drr_begin->drr_payloadlen;
	payload = NULL;
	if (payloadlen != 0)
		payload = kmem_alloc(payloadlen, KM_SLEEP);

	err = receive_read_payload_and_next_header(ra, payloadlen, payload);
	if (err != 0) {
		if (payloadlen != 0)
			kmem_free(payload, payloadlen);
		goto out;
	}
	if (payloadlen != 0) {
		err = nvlist_unpack(payload, payloadlen, &begin_nvl, KM_SLEEP);
		kmem_free(payload, payloadlen);
		if (err != 0)
			goto out;
	}

	/* handle DSL encryption key payload */
	if (featureflags & DMU_BACKUP_FEATURE_RAW) {
		nvlist_t *keynvl = NULL;

		ASSERT(ra->os->os_encrypted);
		ASSERT(drc->drc_raw);

		err = nvlist_lookup_nvlist(begin_nvl, "crypt_keydata", &keynvl);
		if (err != 0)
			goto out;

		/*
		 * If this is a new dataset we set the key immediately.
		 * Otherwise we don't want to change the key until we
		 * are sure the rest of the receive succeeded so we stash
		 * the keynvl away until then.
		 */
		err = dsl_crypto_recv_raw(spa_name(ra->os->os_spa),
		    drc->drc_ds->ds_object, drc->drc_drrb->drr_type,
		    keynvl, drc->drc_newfs);
		if (err != 0)
			goto out;

		if (!drc->drc_newfs)
			drc->drc_keynvl = fnvlist_dup(keynvl);
	}

	if (featureflags & DMU_BACKUP_FEATURE_RESUMING) {
		err = resume_check(ra, begin_nvl);
		if (err != 0)
			goto out;
	}

	(void) bqueue_init(&rwa->q,
	    MAX(zfs_recv_queue_length, 2 * zfs_max_recordsize),
	    offsetof(struct receive_record_arg, node));
	cv_init(&rwa->cv, NULL, CV_DEFAULT, NULL);
	mutex_init(&rwa->mutex, NULL, MUTEX_DEFAULT, NULL);
	rwa->os = ra->os;
	rwa->byteswap = drc->drc_byteswap;
	rwa->resumable = drc->drc_resumable;
	rwa->raw = drc->drc_raw;

	(void) thread_create(NULL, 0, receive_writer_thread, rwa, 0, curproc,
	    TS_RUN, minclsyspri);
	/*
	 * We're reading rwa->err without locks, which is safe since we are the
	 * only reader, and the worker thread is the only writer.  It's ok if we
	 * miss a write for an iteration or two of the loop, since the writer
	 * thread will keep freeing records we send it until we send it an eos
	 * marker.
	 *
	 * We can leave this loop in 3 ways:  First, if rwa->err is
	 * non-zero.  In that case, the writer thread will free the rrd we just
	 * pushed.  Second, if  we're interrupted; in that case, either it's the
	 * first loop and ra->rrd was never allocated, or it's later and ra->rrd
	 * has been handed off to the writer thread who will free it.  Finally,
	 * if receive_read_record fails or we're at the end of the stream, then
	 * we free ra->rrd and exit.
	 */
	while (rwa->err == 0) {
		if (issig(JUSTLOOKING) && issig(FORREAL)) {
			err = SET_ERROR(EINTR);
			break;
		}

		ASSERT3P(ra->rrd, ==, NULL);
		ra->rrd = ra->next_rrd;
		ra->next_rrd = NULL;
		/* Allocates and loads header into ra->next_rrd */
		err = receive_read_record(ra);

		if (ra->rrd->header.drr_type == DRR_END || err != 0) {
			kmem_free(ra->rrd, sizeof (*ra->rrd));
			ra->rrd = NULL;
			break;
		}

		bqueue_enqueue(&rwa->q, ra->rrd,
		    sizeof (struct receive_record_arg) + ra->rrd->payload_size);
		ra->rrd = NULL;
	}
	if (ra->next_rrd == NULL)
		ra->next_rrd = kmem_zalloc(sizeof (*ra->next_rrd), KM_SLEEP);
	ra->next_rrd->eos_marker = B_TRUE;
	bqueue_enqueue(&rwa->q, ra->next_rrd, 1);

	mutex_enter(&rwa->mutex);
	while (!rwa->done) {
		cv_wait(&rwa->cv, &rwa->mutex);
	}
	mutex_exit(&rwa->mutex);

	/*
	 * If we are receiving a full stream as a clone, all object IDs which
	 * are greater than the maximum ID referenced in the stream are
	 * by definition unused and must be freed.
	 */
	if (drc->drc_clone && drc->drc_drrb->drr_fromguid == 0) {
		uint64_t obj = rwa->max_object + 1;
		int free_err = 0;
		int next_err = 0;

		while (next_err == 0) {
			if (drc->drc_raw) {
				free_err = dmu_free_long_object_raw(rwa->os,
				    obj);
			} else {
				free_err = dmu_free_long_object(rwa->os, obj);
			}
			if (free_err != 0 && free_err != ENOENT)
				break;

			next_err = dmu_object_next(rwa->os, &obj, FALSE, 0);
		}

		if (err == 0) {
			if (free_err != 0 && free_err != ENOENT)
				err = free_err;
			else if (next_err != ESRCH)
				err = next_err;
		}
	}

	cv_destroy(&rwa->cv);
	mutex_destroy(&rwa->mutex);
	bqueue_destroy(&rwa->q);
	if (err == 0)
		err = rwa->err;

out:
	nvlist_free(begin_nvl);
	if ((featureflags & DMU_BACKUP_FEATURE_DEDUP) && (cleanup_fd != -1))
		zfs_onexit_fd_rele(cleanup_fd);

	if (err != 0) {
		/*
		 * Clean up references. If receive is not resumable,
		 * destroy what we created, so we don't leave it in
		 * the inconsistent state.
		 */
		dmu_recv_cleanup_ds(drc);
		nvlist_free(drc->drc_keynvl);
	}

	*voffp = ra->voff;
	objlist_destroy(&ra->ignore_objlist);
	kmem_free(ra, sizeof (*ra));
	kmem_free(rwa, sizeof (*rwa));
	return (err);
}

static int
dmu_recv_end_check(void *arg, dmu_tx_t *tx)
{
	dmu_recv_cookie_t *drc = arg;
	dsl_pool_t *dp = dmu_tx_pool(tx);
	int error;

	ASSERT3P(drc->drc_ds->ds_owner, ==, dmu_recv_tag);

	if (!drc->drc_newfs) {
		dsl_dataset_t *origin_head;

		error = dsl_dataset_hold(dp, drc->drc_tofs, FTAG, &origin_head);
		if (error != 0)
			return (error);
		if (drc->drc_force) {
			/*
			 * We will destroy any snapshots in tofs (i.e. before
			 * origin_head) that are after the origin (which is
			 * the snap before drc_ds, because drc_ds can not
			 * have any snaps of its own).
			 */
			uint64_t obj;

			obj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj;
			while (obj !=
			    dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj) {
				dsl_dataset_t *snap;
				error = dsl_dataset_hold_obj(dp, obj, FTAG,
				    &snap);
				if (error != 0)
					break;
				if (snap->ds_dir != origin_head->ds_dir)
					error = SET_ERROR(EINVAL);
				if (error == 0)  {
					error = dsl_destroy_snapshot_check_impl(
					    snap, B_FALSE);
				}
				obj = dsl_dataset_phys(snap)->ds_prev_snap_obj;
				dsl_dataset_rele(snap, FTAG);
				if (error != 0)
					break;
			}
			if (error != 0) {
				dsl_dataset_rele(origin_head, FTAG);
				return (error);
			}
		}
		if (drc->drc_keynvl != NULL) {
			error = dsl_crypto_recv_raw_key_check(drc->drc_ds,
			    drc->drc_keynvl, tx);
			if (error != 0) {
				dsl_dataset_rele(origin_head, FTAG);
				return (error);
			}
		}

		error = dsl_dataset_clone_swap_check_impl(drc->drc_ds,
		    origin_head, drc->drc_force, drc->drc_owner, tx);
		if (error != 0) {
			dsl_dataset_rele(origin_head, FTAG);
			return (error);
		}
		error = dsl_dataset_snapshot_check_impl(origin_head,
		    drc->drc_tosnap, tx, B_TRUE, 1, drc->drc_cred);
		dsl_dataset_rele(origin_head, FTAG);
		if (error != 0)
			return (error);

		error = dsl_destroy_head_check_impl(drc->drc_ds, 1);
	} else {
		error = dsl_dataset_snapshot_check_impl(drc->drc_ds,
		    drc->drc_tosnap, tx, B_TRUE, 1, drc->drc_cred);
	}
	return (error);
}

static void
dmu_recv_end_sync(void *arg, dmu_tx_t *tx)
{
	dmu_recv_cookie_t *drc = arg;
	dsl_pool_t *dp = dmu_tx_pool(tx);
	boolean_t encrypted = drc->drc_ds->ds_dir->dd_crypto_obj != 0;

	spa_history_log_internal_ds(drc->drc_ds, "finish receiving",
	    tx, "snap=%s", drc->drc_tosnap);

	if (!drc->drc_newfs) {
		dsl_dataset_t *origin_head;

		VERIFY0(dsl_dataset_hold(dp, drc->drc_tofs, FTAG,
		    &origin_head));

		if (drc->drc_force) {
			/*
			 * Destroy any snapshots of drc_tofs (origin_head)
			 * after the origin (the snap before drc_ds).
			 */
			uint64_t obj;

			obj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj;
			while (obj !=
			    dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj) {
				dsl_dataset_t *snap;
				VERIFY0(dsl_dataset_hold_obj(dp, obj, FTAG,
				    &snap));
				ASSERT3P(snap->ds_dir, ==, origin_head->ds_dir);
				obj = dsl_dataset_phys(snap)->ds_prev_snap_obj;
				dsl_destroy_snapshot_sync_impl(snap,
				    B_FALSE, tx);
				dsl_dataset_rele(snap, FTAG);
			}
		}
		if (drc->drc_keynvl != NULL) {
			dsl_crypto_recv_raw_key_sync(drc->drc_ds,
			    drc->drc_keynvl, tx);
			nvlist_free(drc->drc_keynvl);
			drc->drc_keynvl = NULL;
		}

		VERIFY3P(drc->drc_ds->ds_prev, ==, origin_head->ds_prev);

		dsl_dataset_clone_swap_sync_impl(drc->drc_ds,
		    origin_head, tx);
		dsl_dataset_snapshot_sync_impl(origin_head,
		    drc->drc_tosnap, tx);

		/* set snapshot's creation time and guid */
		dmu_buf_will_dirty(origin_head->ds_prev->ds_dbuf, tx);
		dsl_dataset_phys(origin_head->ds_prev)->ds_creation_time =
		    drc->drc_drrb->drr_creation_time;
		dsl_dataset_phys(origin_head->ds_prev)->ds_guid =
		    drc->drc_drrb->drr_toguid;
		dsl_dataset_phys(origin_head->ds_prev)->ds_flags &=
		    ~DS_FLAG_INCONSISTENT;

		dmu_buf_will_dirty(origin_head->ds_dbuf, tx);
		dsl_dataset_phys(origin_head)->ds_flags &=
		    ~DS_FLAG_INCONSISTENT;

		drc->drc_newsnapobj =
		    dsl_dataset_phys(origin_head)->ds_prev_snap_obj;

		dsl_dataset_rele(origin_head, FTAG);
		dsl_destroy_head_sync_impl(drc->drc_ds, tx);

		if (drc->drc_owner != NULL)
			VERIFY3P(origin_head->ds_owner, ==, drc->drc_owner);
	} else {
		dsl_dataset_t *ds = drc->drc_ds;

		dsl_dataset_snapshot_sync_impl(ds, drc->drc_tosnap, tx);

		/* set snapshot's creation time and guid */
		dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx);
		dsl_dataset_phys(ds->ds_prev)->ds_creation_time =
		    drc->drc_drrb->drr_creation_time;
		dsl_dataset_phys(ds->ds_prev)->ds_guid =
		    drc->drc_drrb->drr_toguid;
		dsl_dataset_phys(ds->ds_prev)->ds_flags &=
		    ~DS_FLAG_INCONSISTENT;

		dmu_buf_will_dirty(ds->ds_dbuf, tx);
		dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT;
		if (dsl_dataset_has_resume_receive_state(ds)) {
			(void) zap_remove(dp->dp_meta_objset, ds->ds_object,
			    DS_FIELD_RESUME_FROMGUID, tx);
			(void) zap_remove(dp->dp_meta_objset, ds->ds_object,
			    DS_FIELD_RESUME_OBJECT, tx);
			(void) zap_remove(dp->dp_meta_objset, ds->ds_object,
			    DS_FIELD_RESUME_OFFSET, tx);
			(void) zap_remove(dp->dp_meta_objset, ds->ds_object,
			    DS_FIELD_RESUME_BYTES, tx);
			(void) zap_remove(dp->dp_meta_objset, ds->ds_object,
			    DS_FIELD_RESUME_TOGUID, tx);
			(void) zap_remove(dp->dp_meta_objset, ds->ds_object,
			    DS_FIELD_RESUME_TONAME, tx);
		}
		drc->drc_newsnapobj =
		    dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj;
	}
	zvol_create_minors(dp->dp_spa, drc->drc_tofs, B_TRUE);

	/*
	 * Release the hold from dmu_recv_begin.  This must be done before
	 * we return to open context, so that when we free the dataset's dnode
	 * we can evict its bonus buffer. Since the dataset may be destroyed
	 * at this point (and therefore won't have a valid pointer to the spa)
	 * we release the key mapping manually here while we do have a valid
	 * pointer, if it exists.
	 */
	if (!drc->drc_raw && encrypted) {
		(void) spa_keystore_remove_mapping(dmu_tx_pool(tx)->dp_spa,
		    drc->drc_ds->ds_object, drc->drc_ds);
	}
	dsl_dataset_disown(drc->drc_ds, 0, dmu_recv_tag);
	drc->drc_ds = NULL;
}

static int
add_ds_to_guidmap(const char *name, avl_tree_t *guid_map, uint64_t snapobj,
    boolean_t raw)
{
	dsl_pool_t *dp;
	dsl_dataset_t *snapds;
	guid_map_entry_t *gmep;
	ds_hold_flags_t dsflags = (raw) ? 0 : DS_HOLD_FLAG_DECRYPT;
	int err;

	ASSERT(guid_map != NULL);

	err = dsl_pool_hold(name, FTAG, &dp);
	if (err != 0)
		return (err);
	gmep = kmem_alloc(sizeof (*gmep), KM_SLEEP);
	err = dsl_dataset_hold_obj_flags(dp, snapobj, dsflags, gmep, &snapds);
	if (err == 0) {
		gmep->guid = dsl_dataset_phys(snapds)->ds_guid;
		gmep->raw = raw;
		gmep->gme_ds = snapds;
		avl_add(guid_map, gmep);
		dsl_dataset_long_hold(snapds, gmep);
	} else {
		kmem_free(gmep, sizeof (*gmep));
	}

	dsl_pool_rele(dp, FTAG);
	return (err);
}

static int dmu_recv_end_modified_blocks = 3;

static int
dmu_recv_existing_end(dmu_recv_cookie_t *drc)
{
#ifdef _KERNEL
	/*
	 * We will be destroying the ds; make sure its origin is unmounted if
	 * necessary.
	 */
	char name[ZFS_MAX_DATASET_NAME_LEN];
	dsl_dataset_name(drc->drc_ds, name);
	zfs_destroy_unmount_origin(name);
#endif

	return (dsl_sync_task(drc->drc_tofs,
	    dmu_recv_end_check, dmu_recv_end_sync, drc,
	    dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL));
}

static int
dmu_recv_new_end(dmu_recv_cookie_t *drc)
{
	return (dsl_sync_task(drc->drc_tofs,
	    dmu_recv_end_check, dmu_recv_end_sync, drc,
	    dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL));
}

int
dmu_recv_end(dmu_recv_cookie_t *drc, void *owner)
{
	int error;

	drc->drc_owner = owner;

	if (drc->drc_newfs)
		error = dmu_recv_new_end(drc);
	else
		error = dmu_recv_existing_end(drc);

	if (error != 0) {
		dmu_recv_cleanup_ds(drc);
		nvlist_free(drc->drc_keynvl);
	} else if (drc->drc_guid_to_ds_map != NULL) {
		(void) add_ds_to_guidmap(drc->drc_tofs, drc->drc_guid_to_ds_map,
		    drc->drc_newsnapobj, drc->drc_raw);
	}
	return (error);
}

/*
 * Return TRUE if this objset is currently being received into.
 */
boolean_t
dmu_objset_is_receiving(objset_t *os)
{
	return (os->os_dsl_dataset != NULL &&
	    os->os_dsl_dataset->ds_owner == dmu_recv_tag);
}

#if defined(_KERNEL)
module_param(zfs_send_corrupt_data, int, 0644);
MODULE_PARM_DESC(zfs_send_corrupt_data, "Allow sending corrupt data");

module_param(zfs_send_queue_length, int, 0644);
MODULE_PARM_DESC(zfs_send_queue_length, "Maximum send queue length");

module_param(zfs_recv_queue_length, int, 0644);
MODULE_PARM_DESC(zfs_recv_queue_length, "Maximum receive queue length");
#endif
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-14 16:40:18 +0000