X86SchedBroadwell.td 69.4 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742
//=- X86SchedBroadwell.td - X86 Broadwell Scheduling ---------*- tablegen -*-=//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This file defines the machine model for Broadwell to support instruction
// scheduling and other instruction cost heuristics.
//
//===----------------------------------------------------------------------===//

def BroadwellModel : SchedMachineModel {
  // All x86 instructions are modeled as a single micro-op, and BW can decode 4
  // instructions per cycle.
  let IssueWidth = 4;
  let MicroOpBufferSize = 192; // Based on the reorder buffer.
  let LoadLatency = 5;
  let MispredictPenalty = 16;

  // Based on the LSD (loop-stream detector) queue size and benchmarking data.
  let LoopMicroOpBufferSize = 50;

  // This flag is set to allow the scheduler to assign a default model to
  // unrecognized opcodes.
  let CompleteModel = 0;
}

let SchedModel = BroadwellModel in {

// Broadwell can issue micro-ops to 8 different ports in one cycle.

// Ports 0, 1, 5, and 6 handle all computation.
// Port 4 gets the data half of stores. Store data can be available later than
// the store address, but since we don't model the latency of stores, we can
// ignore that.
// Ports 2 and 3 are identical. They handle loads and the address half of
// stores. Port 7 can handle address calculations.
def BWPort0 : ProcResource<1>;
def BWPort1 : ProcResource<1>;
def BWPort2 : ProcResource<1>;
def BWPort3 : ProcResource<1>;
def BWPort4 : ProcResource<1>;
def BWPort5 : ProcResource<1>;
def BWPort6 : ProcResource<1>;
def BWPort7 : ProcResource<1>;

// Many micro-ops are capable of issuing on multiple ports.
def BWPort01  : ProcResGroup<[BWPort0, BWPort1]>;
def BWPort23  : ProcResGroup<[BWPort2, BWPort3]>;
def BWPort237 : ProcResGroup<[BWPort2, BWPort3, BWPort7]>;
def BWPort04  : ProcResGroup<[BWPort0, BWPort4]>;
def BWPort05  : ProcResGroup<[BWPort0, BWPort5]>;
def BWPort06  : ProcResGroup<[BWPort0, BWPort6]>;
def BWPort15  : ProcResGroup<[BWPort1, BWPort5]>;
def BWPort16  : ProcResGroup<[BWPort1, BWPort6]>;
def BWPort56  : ProcResGroup<[BWPort5, BWPort6]>;
def BWPort015 : ProcResGroup<[BWPort0, BWPort1, BWPort5]>;
def BWPort056 : ProcResGroup<[BWPort0, BWPort5, BWPort6]>;
def BWPort0156: ProcResGroup<[BWPort0, BWPort1, BWPort5, BWPort6]>;

// 60 Entry Unified Scheduler
def BWPortAny : ProcResGroup<[BWPort0, BWPort1, BWPort2, BWPort3, BWPort4,
                              BWPort5, BWPort6, BWPort7]> {
  let BufferSize=60;
}

// Integer division issued on port 0.
def BWDivider : ProcResource<1>;
// FP division and sqrt on port 0.
def BWFPDivider : ProcResource<1>;

// Integer loads are 5 cycles, so ReadAfterLd registers needn't be available until 5
// cycles after the memory operand.
def : ReadAdvance<ReadAfterLd, 5>;

// Vector loads are 5/5/6 cycles, so ReadAfterVec*Ld registers needn't be available
// until 5/5/6 cycles after the memory operand.
def : ReadAdvance<ReadAfterVecLd, 5>;
def : ReadAdvance<ReadAfterVecXLd, 5>;
def : ReadAdvance<ReadAfterVecYLd, 6>;

def : ReadAdvance<ReadInt2Fpu, 0>;

// Many SchedWrites are defined in pairs with and without a folded load.
// Instructions with folded loads are usually micro-fused, so they only appear
// as two micro-ops when queued in the reservation station.
// This multiclass defines the resource usage for variants with and without
// folded loads.
multiclass BWWriteResPair<X86FoldableSchedWrite SchedRW,
                          list<ProcResourceKind> ExePorts,
                          int Lat, list<int> Res = [1], int UOps = 1,
                          int LoadLat = 5> {
  // Register variant is using a single cycle on ExePort.
  def : WriteRes<SchedRW, ExePorts> {
    let Latency = Lat;
    let ResourceCycles = Res;
    let NumMicroOps = UOps;
  }

  // Memory variant also uses a cycle on port 2/3 and adds LoadLat cycles to
  // the latency (default = 5).
  def : WriteRes<SchedRW.Folded, !listconcat([BWPort23], ExePorts)> {
    let Latency = !add(Lat, LoadLat);
    let ResourceCycles = !listconcat([1], Res);
    let NumMicroOps = !add(UOps, 1);
  }
}

// A folded store needs a cycle on port 4 for the store data, and an extra port
// 2/3/7 cycle to recompute the address.
def : WriteRes<WriteRMW, [BWPort237,BWPort4]>;

// Arithmetic.
defm : BWWriteResPair<WriteALU,    [BWPort0156], 1>; // Simple integer ALU op.
defm : BWWriteResPair<WriteADC,    [BWPort06], 1>; // Integer ALU + flags op.

// Integer multiplication.
defm : BWWriteResPair<WriteIMul8,     [BWPort1],   3>;
defm : BWWriteResPair<WriteIMul16,    [BWPort1,BWPort06,BWPort0156], 4, [1,1,2], 4>;
defm : X86WriteRes<WriteIMul16Imm,    [BWPort1,BWPort0156], 4, [1,1], 2>;
defm : X86WriteRes<WriteIMul16ImmLd,  [BWPort1,BWPort0156,BWPort23], 8, [1,1,1], 3>;
defm : BWWriteResPair<WriteIMul16Reg, [BWPort1],   3>;
defm : BWWriteResPair<WriteIMul32,    [BWPort1,BWPort06,BWPort0156], 4, [1,1,1], 3>;
defm : BWWriteResPair<WriteIMul32Imm, [BWPort1],   3>;
defm : BWWriteResPair<WriteIMul32Reg, [BWPort1],   3>;
defm : BWWriteResPair<WriteIMul64,    [BWPort1,BWPort5], 4, [1,1], 2>;
defm : BWWriteResPair<WriteIMul64Imm, [BWPort1],   3>;
defm : BWWriteResPair<WriteIMul64Reg, [BWPort1],   3>;
def : WriteRes<WriteIMulH, []> { let Latency = 3; }

// TODO: Why isn't the BWDivider used consistently?
defm : X86WriteRes<WriteDiv8,      [BWPort0, BWDivider], 25, [1, 10], 1>;
defm : X86WriteRes<WriteDiv16,     [BWPort0,BWPort1,BWPort5,BWPort6,BWPort01,BWPort0156], 80, [7,7,3,3,1,11], 32>;
defm : X86WriteRes<WriteDiv32,     [BWPort0,BWPort1,BWPort5,BWPort6,BWPort01,BWPort0156], 80, [7,7,3,3,1,11], 32>;
defm : X86WriteRes<WriteDiv64,     [BWPort0,BWPort1,BWPort5,BWPort6,BWPort01,BWPort0156], 80, [7,7,3,3,1,11], 32>;
defm : X86WriteRes<WriteDiv8Ld,    [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 34, [2,2,2,1,1], 8>;
defm : X86WriteRes<WriteDiv16Ld,   [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 34, [2,2,2,1,1], 8>;
defm : X86WriteRes<WriteDiv32Ld,   [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 34, [2,2,2,1,1], 8>;
defm : X86WriteRes<WriteDiv64Ld,   [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 34, [2,2,2,1,1], 8>;

defm : X86WriteRes<WriteIDiv8,     [BWPort0, BWDivider], 25, [1,10], 1>;
defm : X86WriteRes<WriteIDiv16,    [BWPort0, BWDivider], 25, [1,10], 1>;
defm : X86WriteRes<WriteIDiv32,    [BWPort0, BWDivider], 25, [1,10], 1>;
defm : X86WriteRes<WriteIDiv64,    [BWPort0, BWDivider], 25, [1,10], 1>;
defm : X86WriteRes<WriteIDiv8Ld,   [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 35, [2,2,2,1,1], 8>;
defm : X86WriteRes<WriteIDiv16Ld,  [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 35, [2,2,2,1,1], 8>;
defm : X86WriteRes<WriteIDiv32Ld,  [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 35, [2,2,2,1,1], 8>;
defm : X86WriteRes<WriteIDiv64Ld,  [BWPort0,BWPort1,BWPort5,BWPort23,BWPort0156], 35, [2,2,2,1,1], 8>;

defm : X86WriteRes<WriteCMPXCHG,[BWPort06, BWPort0156], 5, [2, 3], 5>;
defm : X86WriteRes<WriteCMPXCHGRMW,[BWPort23, BWPort06, BWPort0156, BWPort237, BWPort4], 8, [1, 2, 1, 1, 1], 6>;
defm : X86WriteRes<WriteBSWAP32,   [BWPort15], 1, [1], 1>;
defm : X86WriteRes<WriteBSWAP64,   [BWPort06, BWPort15], 2, [1, 1], 2>;
defm : X86WriteRes<WriteXCHG,      [BWPort0156], 2, [3], 3>;

defm : BWWriteResPair<WriteCRC32, [BWPort1],   3>;

def : WriteRes<WriteLEA, [BWPort15]>; // LEA instructions can't fold loads.

defm : BWWriteResPair<WriteCMOV,  [BWPort06], 1>; // Conditional move.
defm : X86WriteRes<WriteFCMOV, [BWPort1], 3, [1], 1>; // x87 conditional move.

def  : WriteRes<WriteSETCC, [BWPort06]>; // Setcc.
def  : WriteRes<WriteSETCCStore, [BWPort06,BWPort4,BWPort237]> {
  let Latency = 2;
  let NumMicroOps = 3;
}

defm : X86WriteRes<WriteLAHFSAHF,        [BWPort06], 1, [1], 1>;
defm : X86WriteRes<WriteBitTest,         [BWPort06], 1, [1], 1>; // Bit Test instrs
defm : X86WriteRes<WriteBitTestImmLd,    [BWPort06,BWPort23], 6, [1,1], 2>;
defm : X86WriteRes<WriteBitTestRegLd,    [BWPort0156,BWPort23], 6, [1,1], 2>;
defm : X86WriteRes<WriteBitTestSet,      [BWPort06], 1, [1], 1>; // Bit Test + Set instrs
defm : X86WriteRes<WriteBitTestSetImmLd, [BWPort06,BWPort23], 5, [1,1], 3>;
defm : X86WriteRes<WriteBitTestSetRegLd, [BWPort0156,BWPort23], 5, [1,1], 2>;

// Bit counts.
defm : BWWriteResPair<WriteBSF, [BWPort1], 3>;
defm : BWWriteResPair<WriteBSR, [BWPort1], 3>;
defm : BWWriteResPair<WriteLZCNT,          [BWPort1], 3>;
defm : BWWriteResPair<WriteTZCNT,          [BWPort1], 3>;
defm : BWWriteResPair<WritePOPCNT,         [BWPort1], 3>;

// Integer shifts and rotates.
defm : BWWriteResPair<WriteShift,    [BWPort06],  1>;
defm : BWWriteResPair<WriteShiftCL,  [BWPort06,BWPort0156],  3, [2,1], 3>;
defm : BWWriteResPair<WriteRotate,   [BWPort06],  1, [1], 1>;
defm : BWWriteResPair<WriteRotateCL, [BWPort06,BWPort0156],  3, [2,1], 3>;

// SHLD/SHRD.
defm : X86WriteRes<WriteSHDrri, [BWPort1], 3, [1], 1>;
defm : X86WriteRes<WriteSHDrrcl,[BWPort1,BWPort06,BWPort0156], 6, [1, 1, 2], 4>;
defm : X86WriteRes<WriteSHDmri, [BWPort1,BWPort23,BWPort237,BWPort0156], 9, [1, 1, 1, 1], 4>;
defm : X86WriteRes<WriteSHDmrcl,[BWPort1,BWPort23,BWPort237,BWPort06,BWPort0156], 11, [1, 1, 1, 1, 2], 6>;

// BMI1 BEXTR/BLS, BMI2 BZHI
defm : BWWriteResPair<WriteBEXTR, [BWPort06,BWPort15], 2, [1,1], 2>;
defm : BWWriteResPair<WriteBLS,   [BWPort15], 1>;
defm : BWWriteResPair<WriteBZHI,  [BWPort15], 1>;

// Loads, stores, and moves, not folded with other operations.
defm : X86WriteRes<WriteLoad,    [BWPort23], 5, [1], 1>;
defm : X86WriteRes<WriteStore,   [BWPort237, BWPort4], 1, [1,1], 1>;
defm : X86WriteRes<WriteStoreNT, [BWPort237, BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteMove,    [BWPort0156], 1, [1], 1>;

// Idioms that clear a register, like xorps %xmm0, %xmm0.
// These can often bypass execution ports completely.
def : WriteRes<WriteZero,  []>;

// Treat misc copies as a move.
def : InstRW<[WriteMove], (instrs COPY)>;

// Branches don't produce values, so they have no latency, but they still
// consume resources. Indirect branches can fold loads.
defm : BWWriteResPair<WriteJump,  [BWPort06],   1>;

// Floating point. This covers both scalar and vector operations.
defm : X86WriteRes<WriteFLD0,          [BWPort01], 1, [1], 1>;
defm : X86WriteRes<WriteFLD1,          [BWPort01], 1, [2], 2>;
defm : X86WriteRes<WriteFLDC,          [BWPort01], 1, [2], 2>;
defm : X86WriteRes<WriteFLoad,         [BWPort23], 5, [1], 1>;
defm : X86WriteRes<WriteFLoadX,        [BWPort23], 5, [1], 1>;
defm : X86WriteRes<WriteFLoadY,        [BWPort23], 6, [1], 1>;
defm : X86WriteRes<WriteFMaskedLoad,   [BWPort23,BWPort5], 7, [1,2], 3>;
defm : X86WriteRes<WriteFMaskedLoadY,  [BWPort23,BWPort5], 8, [1,2], 3>;
defm : X86WriteRes<WriteFStore,        [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteFStoreX,       [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteFStoreY,       [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteFStoreNT,      [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteFStoreNTX,     [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteFStoreNTY,     [BWPort237,BWPort4], 1, [1,1], 2>;

defm : X86WriteRes<WriteFMaskedStore32,  [BWPort0,BWPort4,BWPort237,BWPort15], 5, [1,1,1,1], 4>;
defm : X86WriteRes<WriteFMaskedStore32Y, [BWPort0,BWPort4,BWPort237,BWPort15], 5, [1,1,1,1], 4>;
defm : X86WriteRes<WriteFMaskedStore64,  [BWPort0,BWPort4,BWPort237,BWPort15], 5, [1,1,1,1], 4>;
defm : X86WriteRes<WriteFMaskedStore64Y, [BWPort0,BWPort4,BWPort237,BWPort15], 5, [1,1,1,1], 4>;

defm : X86WriteRes<WriteFMove,         [BWPort5], 1, [1], 1>;
defm : X86WriteRes<WriteFMoveX,        [BWPort5], 1, [1], 1>;
defm : X86WriteRes<WriteFMoveY,        [BWPort5], 1, [1], 1>;

defm : BWWriteResPair<WriteFAdd,    [BWPort1],  3, [1], 1, 5>; // Floating point add/sub.
defm : BWWriteResPair<WriteFAddX,   [BWPort1],  3, [1], 1, 5>; // Floating point add/sub (XMM).
defm : BWWriteResPair<WriteFAddY,   [BWPort1],  3, [1], 1, 6>; // Floating point add/sub (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFAddZ>;
defm : BWWriteResPair<WriteFAdd64,  [BWPort1],  3, [1], 1, 5>; // Floating point double add/sub.
defm : BWWriteResPair<WriteFAdd64X, [BWPort1],  3, [1], 1, 5>; // Floating point double add/sub (XMM).
defm : BWWriteResPair<WriteFAdd64Y, [BWPort1],  3, [1], 1, 6>; // Floating point double add/sub (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFAdd64Z>;

defm : BWWriteResPair<WriteFCmp,    [BWPort1],  3, [1], 1, 5>; // Floating point compare.
defm : BWWriteResPair<WriteFCmpX,   [BWPort1],  3, [1], 1, 5>; // Floating point compare (XMM).
defm : BWWriteResPair<WriteFCmpY,   [BWPort1],  3, [1], 1, 6>; // Floating point compare (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFCmpZ>;
defm : BWWriteResPair<WriteFCmp64,  [BWPort1],  3, [1], 1, 5>; // Floating point double compare.
defm : BWWriteResPair<WriteFCmp64X, [BWPort1],  3, [1], 1, 5>; // Floating point double compare (XMM).
defm : BWWriteResPair<WriteFCmp64Y, [BWPort1],  3, [1], 1, 6>; // Floating point double compare (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFCmp64Z>;

defm : BWWriteResPair<WriteFCom,    [BWPort1],  3>; // Floating point compare to flags.

defm : BWWriteResPair<WriteFMul,    [BWPort01], 3, [1], 1, 5>; // Floating point multiplication.
defm : BWWriteResPair<WriteFMulX,   [BWPort01], 3, [1], 1, 5>; // Floating point multiplication (XMM).
defm : BWWriteResPair<WriteFMulY,   [BWPort01], 3, [1], 1, 6>; // Floating point multiplication (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFMulZ>;
defm : BWWriteResPair<WriteFMul64,  [BWPort01], 3, [1], 1, 5>; // Floating point double multiplication.
defm : BWWriteResPair<WriteFMul64X, [BWPort01], 3, [1], 1, 5>; // Floating point double multiplication (XMM).
defm : BWWriteResPair<WriteFMul64Y, [BWPort01], 3, [1], 1, 6>; // Floating point double multiplication (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFMul64Z>;

//defm : BWWriteResPair<WriteFDiv,     [BWPort0,BWFPDivider], 11, [1,3], 1, 5>; // Floating point division.
defm : BWWriteResPair<WriteFDivX,    [BWPort0,BWFPDivider], 11, [1,5], 1, 5>; // Floating point division (XMM).
defm : BWWriteResPair<WriteFDivY,    [BWPort0,BWPort015,BWFPDivider], 17, [2,1,10], 3, 6>; // Floating point division (YMM).
defm : X86WriteResPairUnsupported<WriteFDivZ>;
//defm : BWWriteResPair<WriteFDiv64,   [BWPort0,BWFPDivider], 14, [1,8], 1, 5>; // Floating point division.
defm : BWWriteResPair<WriteFDiv64X,  [BWPort0,BWFPDivider], 14, [1,8], 1, 5>; // Floating point division (XMM).
defm : BWWriteResPair<WriteFDiv64Y,  [BWPort0,BWPort015,BWFPDivider], 23, [2,1,16], 3, 6>; // Floating point division (YMM).
defm : X86WriteResPairUnsupported<WriteFDiv64Z>;

defm : X86WriteRes<WriteFSqrt,       [BWPort0,BWFPDivider], 11, [1,4], 1>; // Floating point square root.
defm : X86WriteRes<WriteFSqrtLd,     [BWPort0,BWPort23,BWFPDivider], 16, [1,1,7], 2>;
defm : BWWriteResPair<WriteFSqrtX,   [BWPort0,BWFPDivider], 11, [1,7], 1, 5>; // Floating point square root (XMM).
defm : BWWriteResPair<WriteFSqrtY,   [BWPort0,BWPort015,BWFPDivider], 21, [2,1,14], 3, 6>; // Floating point square root (YMM).
defm : X86WriteResPairUnsupported<WriteFSqrtZ>;
defm : X86WriteRes<WriteFSqrt64,     [BWPort0,BWFPDivider], 16, [1,8], 1>; // Floating point double square root.
defm : X86WriteRes<WriteFSqrt64Ld,   [BWPort0,BWPort23,BWFPDivider], 21, [1,1,14], 2>;
defm : BWWriteResPair<WriteFSqrt64X, [BWPort0,BWFPDivider], 16, [1,14],1, 5>; // Floating point double square root (XMM).
defm : BWWriteResPair<WriteFSqrt64Y, [BWPort0,BWPort015,BWFPDivider], 29, [2,1,28], 3, 6>; // Floating point double square root (YMM).
defm : X86WriteResPairUnsupported<WriteFSqrt64Z>;
defm : BWWriteResPair<WriteFSqrt80,  [BWPort0,BWFPDivider], 23, [1,9]>; // Floating point long double square root.

defm : BWWriteResPair<WriteFRcp,   [BWPort0],  5, [1], 1, 5>; // Floating point reciprocal estimate.
defm : BWWriteResPair<WriteFRcpX,  [BWPort0],  5, [1], 1, 5>; // Floating point reciprocal estimate (XMM).
defm : BWWriteResPair<WriteFRcpY,  [BWPort0,BWPort015], 11, [2,1], 3, 6>; // Floating point reciprocal estimate (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFRcpZ>;

defm : BWWriteResPair<WriteFRsqrt, [BWPort0],  5, [1], 1, 5>; // Floating point reciprocal square root estimate.
defm : BWWriteResPair<WriteFRsqrtX,[BWPort0],  5, [1], 1, 5>; // Floating point reciprocal square root estimate (XMM).
defm : BWWriteResPair<WriteFRsqrtY,[BWPort0,BWPort015], 11, [2,1], 3, 6>; // Floating point reciprocal square root estimate (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFRsqrtZ>;

defm : BWWriteResPair<WriteFMA,    [BWPort01], 5, [1], 1, 5>; // Fused Multiply Add.
defm : BWWriteResPair<WriteFMAX,   [BWPort01], 5, [1], 1, 5>; // Fused Multiply Add (XMM).
defm : BWWriteResPair<WriteFMAY,   [BWPort01], 5, [1], 1, 6>; // Fused Multiply Add (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFMAZ>;
defm : BWWriteResPair<WriteDPPD,   [BWPort0,BWPort1,BWPort5],  9, [1,1,1], 3, 5>; // Floating point double dot product.
defm : BWWriteResPair<WriteDPPS,   [BWPort0,BWPort1,BWPort5], 14, [2,1,1], 4, 5>; // Floating point single dot product.
defm : BWWriteResPair<WriteDPPSY,  [BWPort0,BWPort1,BWPort5], 14, [2,1,1], 4, 6>; // Floating point single dot product (YMM).
defm : X86WriteResPairUnsupported<WriteDPPSZ>;
defm : BWWriteResPair<WriteFSign,     [BWPort5], 1>; // Floating point fabs/fchs.
defm : X86WriteRes<WriteFRnd,            [BWPort23],  6, [1],   1>; // Floating point rounding.
defm : X86WriteRes<WriteFRndY,           [BWPort23],  6, [1],   1>; // Floating point rounding (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFRndZ>;
defm : X86WriteRes<WriteFRndLd,  [BWPort1,BWPort23], 11, [2,1], 3>;
defm : X86WriteRes<WriteFRndYLd, [BWPort1,BWPort23], 12, [2,1], 3>;
defm : BWWriteResPair<WriteFLogic,    [BWPort5], 1, [1], 1, 5>; // Floating point and/or/xor logicals.
defm : BWWriteResPair<WriteFLogicY,   [BWPort5], 1, [1], 1, 6>; // Floating point and/or/xor logicals (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFLogicZ>;
defm : BWWriteResPair<WriteFTest,     [BWPort0], 1, [1], 1, 5>; // Floating point TEST instructions.
defm : BWWriteResPair<WriteFTestY,    [BWPort0], 1, [1], 1, 6>; // Floating point TEST instructions (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFTestZ>;
defm : BWWriteResPair<WriteFShuffle,  [BWPort5], 1, [1], 1, 5>; // Floating point vector shuffles.
defm : BWWriteResPair<WriteFShuffleY, [BWPort5], 1, [1], 1, 6>; // Floating point vector shuffles (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteFShuffleZ>;
defm : BWWriteResPair<WriteFVarShuffle,  [BWPort5], 1, [1], 1, 5>; // Floating point vector variable shuffles.
defm : BWWriteResPair<WriteFVarShuffleY, [BWPort5], 1, [1], 1, 6>; // Floating point vector variable shuffles.
defm : X86WriteResPairUnsupported<WriteFVarShuffleZ>;
defm : BWWriteResPair<WriteFBlend,  [BWPort015], 1, [1], 1, 5>; // Floating point vector blends.
defm : BWWriteResPair<WriteFBlendY, [BWPort015], 1, [1], 1, 6>; // Floating point vector blends.
defm : X86WriteResPairUnsupported<WriteFBlendZ>;
defm : BWWriteResPair<WriteFVarBlend,  [BWPort5], 2, [2], 2, 5>; // Fp vector variable blends.
defm : BWWriteResPair<WriteFVarBlendY, [BWPort5], 2, [2], 2, 6>; // Fp vector variable blends.
defm : X86WriteResPairUnsupported<WriteFVarBlendZ>;

// FMA Scheduling helper class.
// class FMASC { X86FoldableSchedWrite Sched = WriteFAdd; }

// Vector integer operations.
defm : X86WriteRes<WriteVecLoad,         [BWPort23], 5, [1], 1>;
defm : X86WriteRes<WriteVecLoadX,        [BWPort23], 5, [1], 1>;
defm : X86WriteRes<WriteVecLoadY,        [BWPort23], 6, [1], 1>;
defm : X86WriteRes<WriteVecLoadNT,       [BWPort23], 5, [1], 1>;
defm : X86WriteRes<WriteVecLoadNTY,      [BWPort23], 6, [1], 1>;
defm : X86WriteRes<WriteVecMaskedLoad,   [BWPort23,BWPort5], 7, [1,2], 3>;
defm : X86WriteRes<WriteVecMaskedLoadY,  [BWPort23,BWPort5], 8, [1,2], 3>;
defm : X86WriteRes<WriteVecStore,        [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteVecStoreX,       [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteVecStoreY,       [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteVecStoreNT,      [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteVecStoreNTY,     [BWPort237,BWPort4], 1, [1,1], 2>;
defm : X86WriteRes<WriteVecMaskedStore,  [BWPort0,BWPort4,BWPort237,BWPort15], 5, [1,1,1,1], 4>;
defm : X86WriteRes<WriteVecMaskedStoreY, [BWPort0,BWPort4,BWPort237,BWPort15], 5, [1,1,1,1], 4>;
defm : X86WriteRes<WriteVecMove,         [BWPort015], 1, [1], 1>;
defm : X86WriteRes<WriteVecMoveX,        [BWPort015], 1, [1], 1>;
defm : X86WriteRes<WriteVecMoveY,        [BWPort015], 1, [1], 1>;
defm : X86WriteRes<WriteVecMoveToGpr,    [BWPort0], 1, [1], 1>;
defm : X86WriteRes<WriteVecMoveFromGpr,  [BWPort5], 1, [1], 1>;

defm : X86WriteRes<WriteEMMS,            [BWPort01,BWPort15,BWPort015,BWPort0156], 31, [8,1,21,1], 31>;

defm : BWWriteResPair<WriteVecALU,   [BWPort15], 1, [1], 1, 5>; // Vector integer ALU op, no logicals.
defm : BWWriteResPair<WriteVecALUX,  [BWPort15], 1, [1], 1, 5>; // Vector integer ALU op, no logicals.
defm : BWWriteResPair<WriteVecALUY,  [BWPort15], 1, [1], 1, 6>; // Vector integer ALU op, no logicals (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVecALUZ>;
defm : BWWriteResPair<WriteVecLogic, [BWPort015], 1, [1], 1, 5>; // Vector integer and/or/xor.
defm : BWWriteResPair<WriteVecLogicX,[BWPort015], 1, [1], 1, 5>; // Vector integer and/or/xor.
defm : BWWriteResPair<WriteVecLogicY,[BWPort015], 1, [1], 1, 6>; // Vector integer and/or/xor (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVecLogicZ>;
defm : BWWriteResPair<WriteVecTest,  [BWPort0,BWPort5], 2, [1,1], 2, 5>; // Vector integer TEST instructions.
defm : BWWriteResPair<WriteVecTestY, [BWPort0,BWPort5], 4, [1,1], 2, 6>; // Vector integer TEST instructions (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVecTestZ>;
defm : BWWriteResPair<WriteVecIMul,  [BWPort0],  5, [1], 1, 5>; // Vector integer multiply.
defm : BWWriteResPair<WriteVecIMulX, [BWPort0],  5, [1], 1, 5>; // Vector integer multiply.
defm : BWWriteResPair<WriteVecIMulY, [BWPort0],  5, [1], 1, 6>; // Vector integer multiply.
defm : X86WriteResPairUnsupported<WriteVecIMulZ>;
defm : BWWriteResPair<WritePMULLD,   [BWPort0], 10, [2], 2, 5>; // Vector PMULLD.
defm : BWWriteResPair<WritePMULLDY,  [BWPort0], 10, [2], 2, 6>; // Vector PMULLD (YMM/ZMM).
defm : X86WriteResPairUnsupported<WritePMULLDZ>;
defm : BWWriteResPair<WriteShuffle,  [BWPort5], 1, [1], 1, 5>; // Vector shuffles.
defm : BWWriteResPair<WriteShuffleX, [BWPort5], 1, [1], 1, 5>; // Vector shuffles.
defm : BWWriteResPair<WriteShuffleY, [BWPort5], 1, [1], 1, 6>; // Vector shuffles (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteShuffleZ>;
defm : BWWriteResPair<WriteVarShuffle, [BWPort5], 1, [1], 1, 5>; // Vector variable shuffles.
defm : BWWriteResPair<WriteVarShuffleX,[BWPort5], 1, [1], 1, 5>; // Vector variable shuffles.
defm : BWWriteResPair<WriteVarShuffleY,[BWPort5], 1, [1], 1, 6>; // Vector variable shuffles (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVarShuffleZ>;
defm : BWWriteResPair<WriteBlend,  [BWPort5], 1, [1], 1, 5>; // Vector blends.
defm : BWWriteResPair<WriteBlendY, [BWPort5], 1, [1], 1, 6>; // Vector blends (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteBlendZ>;
defm : BWWriteResPair<WriteVarBlend,  [BWPort5], 2, [2], 2, 5>; // Vector variable blends.
defm : BWWriteResPair<WriteVarBlendY, [BWPort5], 2, [2], 2, 6>; // Vector variable blends (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVarBlendZ>;
defm : BWWriteResPair<WriteMPSAD,  [BWPort0, BWPort5], 7, [1, 2], 3, 5>; // Vector MPSAD.
defm : BWWriteResPair<WriteMPSADY, [BWPort0, BWPort5], 7, [1, 2], 3, 6>; // Vector MPSAD.
defm : X86WriteResPairUnsupported<WriteMPSADZ>;
defm : BWWriteResPair<WritePSADBW,   [BWPort0], 5, [1], 1, 5>; // Vector PSADBW.
defm : BWWriteResPair<WritePSADBWX,  [BWPort0], 5, [1], 1, 5>; // Vector PSADBW.
defm : BWWriteResPair<WritePSADBWY,  [BWPort0], 5, [1], 1, 6>; // Vector PSADBW (YMM/ZMM).
defm : X86WriteResPairUnsupported<WritePSADBWZ>;
defm : BWWriteResPair<WritePHMINPOS, [BWPort0], 5>; // Vector PHMINPOS.

// Vector integer shifts.
defm : BWWriteResPair<WriteVecShift,     [BWPort0], 1, [1], 1, 5>;
defm : BWWriteResPair<WriteVecShiftX,    [BWPort0,BWPort5],  2, [1,1], 2, 5>;
defm : X86WriteRes<WriteVecShiftY,       [BWPort0,BWPort5],  4, [1,1], 2>;
defm : X86WriteRes<WriteVecShiftYLd,     [BWPort0,BWPort23], 7, [1,1], 2>;
defm : X86WriteResPairUnsupported<WriteVecShiftZ>;

defm : BWWriteResPair<WriteVecShiftImm,  [BWPort0],  1, [1], 1, 5>;
defm : BWWriteResPair<WriteVecShiftImmX, [BWPort0],  1, [1], 1, 5>; // Vector integer immediate shifts (XMM).
defm : BWWriteResPair<WriteVecShiftImmY, [BWPort0],  1, [1], 1, 6>; // Vector integer immediate shifts (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVecShiftImmZ>;
defm : BWWriteResPair<WriteVarVecShift,  [BWPort0, BWPort5], 3, [2,1], 3, 5>; // Variable vector shifts.
defm : BWWriteResPair<WriteVarVecShiftY, [BWPort0, BWPort5], 3, [2,1], 3, 6>; // Variable vector shifts (YMM/ZMM).
defm : X86WriteResPairUnsupported<WriteVarVecShiftZ>;

// Vector insert/extract operations.
def : WriteRes<WriteVecInsert, [BWPort5]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [2];
}
def : WriteRes<WriteVecInsertLd, [BWPort5,BWPort23]> {
  let Latency = 6;
  let NumMicroOps = 2;
}

def : WriteRes<WriteVecExtract, [BWPort0,BWPort5]> {
  let Latency = 2;
  let NumMicroOps = 2;
}
def : WriteRes<WriteVecExtractSt, [BWPort4,BWPort5,BWPort237]> {
  let Latency = 2;
  let NumMicroOps = 3;
}

// Conversion between integer and float.
defm : BWWriteResPair<WriteCvtSS2I,   [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPS2I,   [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPS2IY,  [BWPort1], 3>;
defm : X86WriteResPairUnsupported<WriteCvtPS2IZ>;
defm : BWWriteResPair<WriteCvtSD2I,   [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPD2I,   [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPD2IY,  [BWPort1], 3>;
defm : X86WriteResPairUnsupported<WriteCvtPD2IZ>;

defm : BWWriteResPair<WriteCvtI2SS,   [BWPort1], 4>;
defm : BWWriteResPair<WriteCvtI2PS,   [BWPort1], 4>;
defm : BWWriteResPair<WriteCvtI2PSY,  [BWPort1], 4>;
defm : X86WriteResPairUnsupported<WriteCvtI2PSZ>;
defm : BWWriteResPair<WriteCvtI2SD,   [BWPort1], 4>;
defm : BWWriteResPair<WriteCvtI2PD,   [BWPort1], 4>;
defm : BWWriteResPair<WriteCvtI2PDY,  [BWPort1], 4>;
defm : X86WriteResPairUnsupported<WriteCvtI2PDZ>;

defm : BWWriteResPair<WriteCvtSS2SD,  [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPS2PD,  [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPS2PDY, [BWPort1], 3>;
defm : X86WriteResPairUnsupported<WriteCvtPS2PDZ>;
defm : BWWriteResPair<WriteCvtSD2SS,  [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPD2PS,  [BWPort1], 3>;
defm : BWWriteResPair<WriteCvtPD2PSY, [BWPort1], 3>;
defm : X86WriteResPairUnsupported<WriteCvtPD2PSZ>;

defm : X86WriteRes<WriteCvtPH2PS,     [BWPort0,BWPort5], 2, [1,1], 2>;
defm : X86WriteRes<WriteCvtPH2PSY,    [BWPort0,BWPort5], 2, [1,1], 2>;
defm : X86WriteResUnsupported<WriteCvtPH2PSZ>;
defm : X86WriteRes<WriteCvtPH2PSLd,  [BWPort0,BWPort23], 6, [1,1], 2>;
defm : X86WriteRes<WriteCvtPH2PSYLd, [BWPort0,BWPort23], 6, [1,1], 2>;
defm : X86WriteResUnsupported<WriteCvtPH2PSZLd>;

defm : X86WriteRes<WriteCvtPS2PH,    [BWPort1,BWPort5], 4, [1,1], 2>;
defm : X86WriteRes<WriteCvtPS2PHY,   [BWPort1,BWPort5], 6, [1,1], 2>;
defm : X86WriteResUnsupported<WriteCvtPS2PHZ>;
defm : X86WriteRes<WriteCvtPS2PHSt,  [BWPort1,BWPort4,BWPort237], 5, [1,1,1], 3>;
defm : X86WriteRes<WriteCvtPS2PHYSt, [BWPort1,BWPort4,BWPort237], 7, [1,1,1], 3>;
defm : X86WriteResUnsupported<WriteCvtPS2PHZSt>;

// Strings instructions.

// Packed Compare Implicit Length Strings, Return Mask
def : WriteRes<WritePCmpIStrM, [BWPort0]> {
  let Latency = 11;
  let NumMicroOps = 3;
  let ResourceCycles = [3];
}
def : WriteRes<WritePCmpIStrMLd, [BWPort0, BWPort23]> {
  let Latency = 16;
  let NumMicroOps = 4;
  let ResourceCycles = [3,1];
}

// Packed Compare Explicit Length Strings, Return Mask
def : WriteRes<WritePCmpEStrM, [BWPort0, BWPort5, BWPort015, BWPort0156]> {
  let Latency = 19;
  let NumMicroOps = 9;
  let ResourceCycles = [4,3,1,1];
}
def : WriteRes<WritePCmpEStrMLd, [BWPort0, BWPort5, BWPort23, BWPort015, BWPort0156]> {
  let Latency = 24;
  let NumMicroOps = 10;
  let ResourceCycles = [4,3,1,1,1];
}

// Packed Compare Implicit Length Strings, Return Index
def : WriteRes<WritePCmpIStrI, [BWPort0]> {
  let Latency = 11;
  let NumMicroOps = 3;
  let ResourceCycles = [3];
}
def : WriteRes<WritePCmpIStrILd, [BWPort0, BWPort23]> {
  let Latency = 16;
  let NumMicroOps = 4;
  let ResourceCycles = [3,1];
}

// Packed Compare Explicit Length Strings, Return Index
def : WriteRes<WritePCmpEStrI, [BWPort0, BWPort5, BWPort0156]> {
  let Latency = 18;
  let NumMicroOps = 8;
  let ResourceCycles = [4,3,1];
}
def : WriteRes<WritePCmpEStrILd, [BWPort0, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 23;
  let NumMicroOps = 9;
  let ResourceCycles = [4,3,1,1];
}

// MOVMSK Instructions.
def : WriteRes<WriteFMOVMSK,    [BWPort0]> { let Latency = 3; }
def : WriteRes<WriteVecMOVMSK,  [BWPort0]> { let Latency = 3; }
def : WriteRes<WriteVecMOVMSKY, [BWPort0]> { let Latency = 3; }
def : WriteRes<WriteMMXMOVMSK,  [BWPort0]> { let Latency = 1; }

// AES instructions.
def : WriteRes<WriteAESDecEnc, [BWPort5]> { // Decryption, encryption.
  let Latency = 7;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def : WriteRes<WriteAESDecEncLd, [BWPort5, BWPort23]> {
  let Latency = 12;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}

def : WriteRes<WriteAESIMC, [BWPort5]> { // InvMixColumn.
  let Latency = 14;
  let NumMicroOps = 2;
  let ResourceCycles = [2];
}
def : WriteRes<WriteAESIMCLd, [BWPort5, BWPort23]> {
  let Latency = 19;
  let NumMicroOps = 3;
  let ResourceCycles = [2,1];
}

def : WriteRes<WriteAESKeyGen, [BWPort0, BWPort5, BWPort015]> { // Key Generation.
  let Latency = 29;
  let NumMicroOps = 11;
  let ResourceCycles = [2,7,2];
}
def : WriteRes<WriteAESKeyGenLd, [BWPort0, BWPort5, BWPort23, BWPort015]> {
  let Latency = 33;
  let NumMicroOps = 11;
  let ResourceCycles = [2,7,1,1];
}

// Carry-less multiplication instructions.
defm : BWWriteResPair<WriteCLMul,  [BWPort0], 5>;

// Catch-all for expensive system instructions.
def : WriteRes<WriteSystem,     [BWPort0156]> { let Latency = 100; } // def WriteSystem : SchedWrite;

// AVX2.
defm : BWWriteResPair<WriteFShuffle256, [BWPort5], 3, [1], 1, 6>; // Fp 256-bit width vector shuffles.
defm : BWWriteResPair<WriteFVarShuffle256, [BWPort5], 3, [1], 1, 6>; // Fp 256-bit width vector variable shuffles.
defm : BWWriteResPair<WriteShuffle256, [BWPort5], 3, [1], 1, 6>;  // 256-bit width vector shuffles.
defm : BWWriteResPair<WriteVarShuffle256, [BWPort5], 3, [1], 1, 6>;  // 256-bit width vector variable shuffles.

// Old microcoded instructions that nobody use.
def : WriteRes<WriteMicrocoded, [BWPort0156]> { let Latency = 100; } // def WriteMicrocoded : SchedWrite;

// Fence instructions.
def : WriteRes<WriteFence,  [BWPort23, BWPort4]>;

// Load/store MXCSR.
def : WriteRes<WriteLDMXCSR, [BWPort0,BWPort23,BWPort0156]> { let Latency = 7; let NumMicroOps = 3; let ResourceCycles = [1,1,1]; }
def : WriteRes<WriteSTMXCSR, [BWPort4,BWPort5,BWPort237]> { let Latency = 2; let NumMicroOps = 3; let ResourceCycles = [1,1,1]; }

// Nop, not very useful expect it provides a model for nops!
def : WriteRes<WriteNop, []>;

////////////////////////////////////////////////////////////////////////////////
// Horizontal add/sub  instructions.
////////////////////////////////////////////////////////////////////////////////

defm : BWWriteResPair<WriteFHAdd,   [BWPort1,BWPort5], 5, [1,2], 3, 5>;
defm : BWWriteResPair<WriteFHAddY,  [BWPort1,BWPort5], 5, [1,2], 3, 6>;
defm : BWWriteResPair<WritePHAdd,  [BWPort5,BWPort15], 3, [2,1], 3, 5>;
defm : BWWriteResPair<WritePHAddX, [BWPort5,BWPort15], 3, [2,1], 3, 5>;
defm : BWWriteResPair<WritePHAddY, [BWPort5,BWPort15], 3, [2,1], 3, 6>;

// Remaining instrs.

def BWWriteResGroup1 : SchedWriteRes<[BWPort0]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup1], (instregex "VPSLLVQ(Y?)rr",
                                           "VPSRLVQ(Y?)rr")>;

def BWWriteResGroup2 : SchedWriteRes<[BWPort1]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup2], (instregex "COM(P?)_FST0r",
                                           "UCOM_F(P?)r")>;

def BWWriteResGroup3 : SchedWriteRes<[BWPort5]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup3], (instrs MMX_MOVQ2DQrr)>;

def BWWriteResGroup4 : SchedWriteRes<[BWPort6]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup4], (instregex "JMP(16|32|64)r")>;

def BWWriteResGroup5 : SchedWriteRes<[BWPort01]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup5], (instrs FINCSTP, FNOP)>;

def BWWriteResGroup6 : SchedWriteRes<[BWPort06]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup6], (instrs CDQ, CQO)>;

def BWWriteResGroup7 : SchedWriteRes<[BWPort15]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup7], (instregex "ANDN(32|64)rr")>;

def BWWriteResGroup8 : SchedWriteRes<[BWPort015]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup8], (instregex "VPBLENDD(Y?)rri")>;

def BWWriteResGroup9 : SchedWriteRes<[BWPort0156]> {
  let Latency = 1;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup9], (instrs SGDT64m,
                                        SIDT64m,
                                        SMSW16m,
                                        STRm,
                                        SYSCALL)>;

def BWWriteResGroup10 : SchedWriteRes<[BWPort4,BWPort237]> {
  let Latency = 1;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup10], (instrs FBSTPm)>;
def: InstRW<[BWWriteResGroup10], (instregex "ST_FP(32|64|80)m")>;

def BWWriteResGroup12 : SchedWriteRes<[BWPort01]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [2];
}
def: InstRW<[BWWriteResGroup12], (instrs FDECSTP)>;

def BWWriteResGroup14 : SchedWriteRes<[BWPort0156]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [2];
}
def: InstRW<[BWWriteResGroup14], (instrs LFENCE,
                                         MFENCE,
                                         WAIT,
                                         XGETBV)>;

def BWWriteResGroup15 : SchedWriteRes<[BWPort0,BWPort5]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup15], (instregex "(V?)CVTPS2PDrr",
                                            "(V?)CVTSS2SDrr")>;

def BWWriteResGroup16 : SchedWriteRes<[BWPort6,BWPort0156]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup16], (instregex "CLFLUSH")>;

def BWWriteResGroup17 : SchedWriteRes<[BWPort01,BWPort015]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup17], (instrs MMX_MOVDQ2Qrr)>;

def BWWriteResGroup18 : SchedWriteRes<[BWPort237,BWPort0156]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup18], (instrs SFENCE)>;

def BWWriteResGroup20 : SchedWriteRes<[BWPort06,BWPort0156]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup20], (instrs CWD,
                                         JCXZ, JECXZ, JRCXZ,
                                         ADC8i8, SBB8i8,
                                         ADC16i16, SBB16i16,
                                         ADC32i32, SBB32i32,
                                         ADC64i32, SBB64i32)>;

def BWWriteResGroup22 : SchedWriteRes<[BWPort4,BWPort6,BWPort237]> {
  let Latency = 2;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup22], (instrs FNSTCW16m)>;

def BWWriteResGroup24 : SchedWriteRes<[BWPort4,BWPort237,BWPort15]> {
  let Latency = 2;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup24], (instregex "MOVBE(16|32|64)mr")>;

def BWWriteResGroup25 : SchedWriteRes<[BWPort4,BWPort237,BWPort0156]> {
  let Latency = 2;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup25], (instrs PUSH16r, PUSH32r, PUSH64r, PUSH64i8,
                                         STOSB, STOSL, STOSQ, STOSW)>;
def: InstRW<[BWWriteResGroup25], (instregex "PUSH(16|32|64)rmr")>;

def BWWriteResGroup27 : SchedWriteRes<[BWPort1]> {
  let Latency = 3;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup27], (instrs MMX_CVTPI2PSirr)>;
def: InstRW<[BWWriteResGroup27], (instregex "P(DEP|EXT)(32|64)rr",
                                            "(V?)CVTDQ2PS(Y?)rr")>;

def BWWriteResGroup28 : SchedWriteRes<[BWPort5]> {
  let Latency = 3;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup28], (instrs VPBROADCASTBrr,
                                         VPBROADCASTWrr)>;

def BWWriteResGroup33 : SchedWriteRes<[BWPort5,BWPort0156]> {
  let Latency = 3;
  let NumMicroOps = 3;
  let ResourceCycles = [2,1];
}
def: InstRW<[BWWriteResGroup33], (instrs MMX_PACKSSDWirr,
                                         MMX_PACKSSWBirr,
                                         MMX_PACKUSWBirr)>;

def BWWriteResGroup34 : SchedWriteRes<[BWPort6,BWPort0156]> {
  let Latency = 3;
  let NumMicroOps = 3;
  let ResourceCycles = [1,2];
}
def: InstRW<[BWWriteResGroup34], (instregex "CLD")>;

def BWWriteResGroup35 : SchedWriteRes<[BWPort06,BWPort0156]> {
  let Latency = 3;
  let NumMicroOps = 3;
  let ResourceCycles = [1,2];
}
def: InstRW<[BWWriteResGroup35], (instregex "RCL(8|16|32|64)r(1|i)",
                                            "RCR(8|16|32|64)r(1|i)")>;

def BWWriteResGroup37 : SchedWriteRes<[BWPort4,BWPort6,BWPort237,BWPort0156]> {
  let Latency = 3;
  let NumMicroOps = 4;
  let ResourceCycles = [1,1,1,1];
}
def: InstRW<[BWWriteResGroup37], (instregex "CALL(16|32|64)r")>;

def BWWriteResGroup38 : SchedWriteRes<[BWPort4,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 3;
  let NumMicroOps = 4;
  let ResourceCycles = [1,1,1,1];
}
def: InstRW<[BWWriteResGroup38], (instrs CALL64pcrel32)>;

def BWWriteResGroup39 : SchedWriteRes<[BWPort0,BWPort1]> {
  let Latency = 4;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup39], (instregex "(V?)CVT(T?)SD2SI64rr",
                                            "(V?)CVT(T?)SD2SIrr",
                                            "(V?)CVT(T?)SS2SI64rr",
                                            "(V?)CVT(T?)SS2SIrr")>;

def BWWriteResGroup40 : SchedWriteRes<[BWPort0,BWPort5]> {
  let Latency = 4;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup40], (instrs VCVTPS2PDYrr)>;

def BWWriteResGroup41 : SchedWriteRes<[BWPort0,BWPort0156]> {
  let Latency = 4;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup41], (instrs FNSTSW16r)>;

def BWWriteResGroup42 : SchedWriteRes<[BWPort1,BWPort5]> {
  let Latency = 4;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup42], (instrs MMX_CVTPI2PDirr)>;
def: InstRW<[BWWriteResGroup42], (instregex "MMX_CVT(T?)PD2PIirr",
                                            "MMX_CVT(T?)PS2PIirr",
                                            "(V?)CVTDQ2PDrr",
                                            "(V?)CVTPD2PSrr",
                                            "(V?)CVTSD2SSrr",
                                            "(V?)CVTSI642SDrr",
                                            "(V?)CVTSI2SDrr",
                                            "(V?)CVTSI2SSrr",
                                            "(V?)CVT(T?)PD2DQrr")>;

def BWWriteResGroup43 : SchedWriteRes<[BWPort0,BWPort4,BWPort237]> {
  let Latency = 4;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup43], (instrs FNSTSWm)>;

def BWWriteResGroup44 : SchedWriteRes<[BWPort1,BWPort4,BWPort237]> {
  let Latency = 4;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup44], (instregex "IST(T?)_FP(16|32|64)m",
                                            "IST_F(16|32)m")>;

def BWWriteResGroup45 : SchedWriteRes<[BWPort0156]> {
  let Latency = 4;
  let NumMicroOps = 4;
  let ResourceCycles = [4];
}
def: InstRW<[BWWriteResGroup45], (instrs FNCLEX)>;

def BWWriteResGroup46 : SchedWriteRes<[]> {
  let Latency = 0;
  let NumMicroOps = 4;
  let ResourceCycles = [];
}
def: InstRW<[BWWriteResGroup46], (instrs VZEROUPPER)>;

def BWWriteResGroup47 : SchedWriteRes<[BWPort0]> {
  let Latency = 5;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup47], (instregex "MUL_(FPrST0|FST0r|FrST0)")>;

def BWWriteResGroup49 : SchedWriteRes<[BWPort23]> {
  let Latency = 5;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup49], (instregex "MOVSX(16|32|64)rm(8|16|32)",
                                            "MOVZX(16|32|64)rm(8|16)")>;
def: InstRW<[BWWriteResGroup49], (instrs VBROADCASTSSrm,
                                         VMOVDDUPrm, MOVDDUPrm,
                                         VMOVSHDUPrm, MOVSHDUPrm,
                                         VMOVSLDUPrm, MOVSLDUPrm,
                                         VPBROADCASTDrm,
                                         VPBROADCASTQrm)>;

def BWWriteResGroup50 : SchedWriteRes<[BWPort1,BWPort5]> {
  let Latency = 5;
  let NumMicroOps = 3;
  let ResourceCycles = [1,2];
}
def: InstRW<[BWWriteResGroup50], (instregex "(V?)CVTSI642SSrr")>;

def BWWriteResGroup51 : SchedWriteRes<[BWPort1,BWPort6,BWPort06]> {
  let Latency = 5;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup51], (instregex "STR(16|32|64)r")>;

def BWWriteResGroup54 : SchedWriteRes<[BWPort6,BWPort0156]> {
  let Latency = 5;
  let NumMicroOps = 5;
  let ResourceCycles = [1,4];
}
def: InstRW<[BWWriteResGroup54], (instrs PAUSE)>;

def BWWriteResGroup55 : SchedWriteRes<[BWPort06,BWPort0156]> {
  let Latency = 5;
  let NumMicroOps = 5;
  let ResourceCycles = [1,4];
}
def: InstRW<[BWWriteResGroup55], (instrs XSETBV)>;

def BWWriteResGroup57 : SchedWriteRes<[BWPort4,BWPort237,BWPort0156]> {
  let Latency = 5;
  let NumMicroOps = 6;
  let ResourceCycles = [1,1,4];
}
def: InstRW<[BWWriteResGroup57], (instregex "PUSHF(16|64)")>;

def BWWriteResGroup58 : SchedWriteRes<[BWPort23]> {
  let Latency = 6;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup58], (instregex "LD_F(32|64|80)m")>;
def: InstRW<[BWWriteResGroup58], (instrs VBROADCASTF128,
                                         VBROADCASTI128,
                                         VBROADCASTSDYrm,
                                         VBROADCASTSSYrm,
                                         VMOVDDUPYrm,
                                         VMOVSHDUPYrm,
                                         VMOVSLDUPYrm,
                                         VPBROADCASTDYrm,
                                         VPBROADCASTQYrm)>;

def BWWriteResGroup59 : SchedWriteRes<[BWPort0,BWPort23]> {
  let Latency = 6;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup59], (instrs CVTPS2PDrm, VCVTPS2PDrm,
                                         CVTSS2SDrm, VCVTSS2SDrm,
                                         CVTSS2SDrm_Int, VCVTSS2SDrm_Int,
                                         VPSLLVQrm,
                                         VPSRLVQrm)>;

def BWWriteResGroup60 : SchedWriteRes<[BWPort1,BWPort5]> {
  let Latency = 6;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup60], (instrs VCVTDQ2PDYrr,
                                         VCVTPD2PSYrr,
                                         VCVTPD2DQYrr,
                                         VCVTTPD2DQYrr)>;

def BWWriteResGroup62 : SchedWriteRes<[BWPort6,BWPort23]> {
  let Latency = 6;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup62], (instrs FARJMP64)>;
def: InstRW<[BWWriteResGroup62], (instregex "JMP(16|32|64)m")>;

def BWWriteResGroup64 : SchedWriteRes<[BWPort23,BWPort15]> {
  let Latency = 6;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup64], (instregex "ANDN(32|64)rm",
                                            "MOVBE(16|32|64)rm")>;

def BWWriteResGroup65 : SchedWriteRes<[BWPort23,BWPort015]> {
  let Latency = 6;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup65], (instrs VINSERTF128rm,
                                         VINSERTI128rm,
                                         VPBLENDDrmi)>;

def BWWriteResGroup66 : SchedWriteRes<[BWPort23,BWPort0156]> {
  let Latency = 6;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup66], (instrs POP16r, POP32r, POP64r)>;
def: InstRW<[BWWriteResGroup66], (instregex "POP(16|32|64)rmr")>;

def BWWriteResGroup68 : SchedWriteRes<[BWPort1,BWPort6,BWPort06,BWPort0156]> {
  let Latency = 6;
  let NumMicroOps = 4;
  let ResourceCycles = [1,1,1,1];
}
def: InstRW<[BWWriteResGroup68], (instregex "SLDT(16|32|64)r")>;

def BWWriteResGroup69 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort06]> {
  let Latency = 6;
  let NumMicroOps = 4;
  let ResourceCycles = [1,1,1,1];
}
def: InstRW<[BWWriteResGroup69], (instregex "SAR(8|16|32|64)m(1|i)",
                                            "SHL(8|16|32|64)m(1|i)",
                                            "SHR(8|16|32|64)m(1|i)")>;

def BWWriteResGroup70 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort0156]> {
  let Latency = 6;
  let NumMicroOps = 4;
  let ResourceCycles = [1,1,1,1];
}
def: InstRW<[BWWriteResGroup70], (instregex "POP(16|32|64)rmm",
                                            "PUSH(16|32|64)rmm")>;

def BWWriteResGroup71 : SchedWriteRes<[BWPort6,BWPort0156]> {
  let Latency = 6;
  let NumMicroOps = 6;
  let ResourceCycles = [1,5];
}
def: InstRW<[BWWriteResGroup71], (instrs STD)>;

def BWWriteResGroup73 : SchedWriteRes<[BWPort0,BWPort23]> {
  let Latency = 7;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup73], (instrs VPSLLVQYrm,
                                         VPSRLVQYrm)>;

def BWWriteResGroup74 : SchedWriteRes<[BWPort1,BWPort23]> {
  let Latency = 7;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup74], (instregex "FCOM(P?)(32|64)m")>;

def BWWriteResGroup77 : SchedWriteRes<[BWPort23,BWPort015]> {
  let Latency = 7;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup77], (instrs VPBLENDDYrmi)>;

def BWWriteResGroup79 : SchedWriteRes<[BWPort5,BWPort23]> {
  let Latency = 7;
  let NumMicroOps = 3;
  let ResourceCycles = [2,1];
}
def: InstRW<[BWWriteResGroup79], (instrs MMX_PACKSSDWirm,
                                         MMX_PACKSSWBirm,
                                         MMX_PACKUSWBirm)>;

def BWWriteResGroup80 : SchedWriteRes<[BWPort23,BWPort0156]> {
  let Latency = 7;
  let NumMicroOps = 3;
  let ResourceCycles = [1,2];
}
def: InstRW<[BWWriteResGroup80], (instrs LEAVE, LEAVE64,
                                         SCASB, SCASL, SCASQ, SCASW)>;

def BWWriteResGroup82 : SchedWriteRes<[BWPort0,BWPort01,BWPort23]> {
  let Latency = 7;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup82], (instrs FLDCW16m)>;

def BWWriteResGroup84 : SchedWriteRes<[BWPort6,BWPort23,BWPort0156]> {
  let Latency = 7;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup84], (instrs LRETQ, RETQ)>;

def BWWriteResGroup87 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort06]> {
  let Latency = 7;
  let NumMicroOps = 5;
  let ResourceCycles = [1,1,1,2];
}
def: InstRW<[BWWriteResGroup87], (instregex "ROL(8|16|32|64)m(1|i)",
                                            "ROR(8|16|32|64)m(1|i)")>;

def BWWriteResGroup87_1 : SchedWriteRes<[BWPort06]> {
  let Latency = 2;
  let NumMicroOps = 2;
  let ResourceCycles = [2];
}
def: InstRW<[BWWriteResGroup87_1], (instrs ROL8r1, ROL16r1, ROL32r1, ROL64r1,
                                           ROR8r1, ROR16r1, ROR32r1, ROR64r1)>;

def BWWriteResGroup88 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort0156]> {
  let Latency = 7;
  let NumMicroOps = 5;
  let ResourceCycles = [1,1,1,2];
}
def: InstRW<[BWWriteResGroup88], (instregex "XADD(8|16|32|64)rm")>;

def BWWriteResGroup89 : SchedWriteRes<[BWPort4,BWPort6,BWPort23,BWPort237,BWPort0156]> {
  let Latency = 7;
  let NumMicroOps = 5;
  let ResourceCycles = [1,1,1,1,1];
}
def: InstRW<[BWWriteResGroup89], (instregex "CALL(16|32|64)m")>;
def: InstRW<[BWWriteResGroup89], (instrs FARCALL64)>;

def BWWriteResGroup90 : SchedWriteRes<[BWPort6,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 7;
  let NumMicroOps = 7;
  let ResourceCycles = [2,2,1,2];
}
def: InstRW<[BWWriteResGroup90], (instrs LOOP)>;

def BWWriteResGroup91 : SchedWriteRes<[BWPort1,BWPort23]> {
  let Latency = 8;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup91], (instrs MMX_CVTPI2PSirm,
                                         CVTDQ2PSrm,
                                         VCVTDQ2PSrm)>;
def: InstRW<[BWWriteResGroup91], (instregex "P(DEP|EXT)(32|64)rm")>;

def BWWriteResGroup92 : SchedWriteRes<[BWPort5,BWPort23]> {
  let Latency = 8;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup92], (instrs VPMOVSXBDYrm,
                                         VPMOVSXBQYrm,
                                         VPMOVSXBWYrm,
                                         VPMOVSXDQYrm,
                                         VPMOVSXWDYrm,
                                         VPMOVSXWQYrm,
                                         VPMOVZXWDYrm)>;

def BWWriteResGroup97 : SchedWriteRes<[BWPort23,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 8;
  let NumMicroOps = 5;
  let ResourceCycles = [1,1,1,2];
}
def: InstRW<[BWWriteResGroup97], (instregex "RCL(8|16|32|64)m(1|i)",
                                            "RCR(8|16|32|64)m(1|i)")>;

def BWWriteResGroup99 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort0156]> {
  let Latency = 8;
  let NumMicroOps = 6;
  let ResourceCycles = [1,1,1,3];
}
def: InstRW<[BWWriteResGroup99], (instregex "XCHG(8|16|32|64)rm")>;

def BWWriteResGroup100 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 8;
  let NumMicroOps = 6;
  let ResourceCycles = [1,1,1,2,1];
}
def : SchedAlias<WriteADCRMW, BWWriteResGroup100>;
def: InstRW<[BWWriteResGroup100], (instregex "ROL(8|16|32|64)mCL",
                                             "ROR(8|16|32|64)mCL",
                                             "SAR(8|16|32|64)mCL",
                                             "SHL(8|16|32|64)mCL",
                                             "SHR(8|16|32|64)mCL")>;

def BWWriteResGroup101 : SchedWriteRes<[BWPort1,BWPort23]> {
  let Latency = 9;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup101], (instregex "(ADD|SUB|SUBR)_F(32|64)m",
                                             "ILD_F(16|32|64)m")>;
def: InstRW<[BWWriteResGroup101], (instrs VCVTPS2DQYrm,
                                          VCVTTPS2DQYrm)>;

def BWWriteResGroup105 : SchedWriteRes<[BWPort0,BWPort1,BWPort23]> {
  let Latency = 9;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup105], (instregex "(V?)CVTSS2SI(64)?rm",
                                             "(V?)CVT(T?)SD2SI64rm",
                                             "(V?)CVT(T?)SD2SIrm",
                                             "VCVTTSS2SI64rm",
                                             "(V?)CVTTSS2SIrm")>;

def BWWriteResGroup106 : SchedWriteRes<[BWPort0,BWPort5,BWPort23]> {
  let Latency = 9;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup106], (instrs VCVTPS2PDYrm)>;

def BWWriteResGroup107 : SchedWriteRes<[BWPort1,BWPort5,BWPort23]> {
  let Latency = 9;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup107], (instrs CVTPD2PSrm,
                                          CVTPD2DQrm,
                                          CVTTPD2DQrm,
                                          MMX_CVTPI2PDirm)>;
def: InstRW<[BWWriteResGroup107], (instregex "MMX_CVT(T?)PD2PIirm",
                                             "(V?)CVTDQ2PDrm",
                                             "(V?)CVTSD2SSrm")>;

def BWWriteResGroup108 : SchedWriteRes<[BWPort5,BWPort23,BWPort015]> {
  let Latency = 9;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup108], (instregex "VPBROADCASTB(Y?)rm",
                                             "VPBROADCASTW(Y?)rm")>;

def BWWriteResGroup112 : SchedWriteRes<[BWPort23,BWPort06,BWPort0156]> {
  let Latency = 9;
  let NumMicroOps = 5;
  let ResourceCycles = [1,1,3];
}
def: InstRW<[BWWriteResGroup112], (instrs RDRAND16r, RDRAND32r, RDRAND64r)>;

def BWWriteResGroup113 : SchedWriteRes<[BWPort1,BWPort6,BWPort23,BWPort0156]> {
  let Latency = 9;
  let NumMicroOps = 5;
  let ResourceCycles = [1,2,1,1];
}
def: InstRW<[BWWriteResGroup113], (instregex "LAR(16|32|64)rm",
                                             "LSL(16|32|64)rm")>;

def BWWriteResGroup115 : SchedWriteRes<[BWPort0,BWPort23]> {
  let Latency = 10;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup115], (instregex "(V?)PCMPGTQrm")>;

def BWWriteResGroup117 : SchedWriteRes<[BWPort1,BWPort23]> {
  let Latency = 10;
  let NumMicroOps = 3;
  let ResourceCycles = [2,1];
}
def: InstRW<[BWWriteResGroup117], (instregex "FICOM(P?)(16|32)m")>;

def BWWriteResGroup120 : SchedWriteRes<[BWPort0,BWPort1,BWPort5,BWPort23]> {
  let Latency = 10;
  let NumMicroOps = 4;
  let ResourceCycles = [1,1,1,1];
}
def: InstRW<[BWWriteResGroup120], (instregex "CVTTSS2SI64rm")>;

def BWWriteResGroup122_1 : SchedWriteRes<[BWPort0,BWFPDivider]> {
  let Latency = 11;
  let NumMicroOps = 1;
  let ResourceCycles = [1,3]; // Really 2.5 cycle throughput
}
def : SchedAlias<WriteFDiv, BWWriteResGroup122_1>; // TODO - convert to ZnWriteResFpuPair

def BWWriteResGroup123 : SchedWriteRes<[BWPort0,BWPort23]> {
  let Latency = 11;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup123], (instregex "MUL_F(32|64)m")>;
def: InstRW<[BWWriteResGroup123], (instrs VPCMPGTQYrm)>;

def BWWriteResGroup128 : SchedWriteRes<[BWPort1,BWPort5,BWPort23]> {
  let Latency = 11;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup128], (instrs VCVTDQ2PDYrm)>;

def BWWriteResGroup131 : SchedWriteRes<[BWPort1,BWPort06,BWPort0156]> {
  let Latency = 11;
  let NumMicroOps = 7;
  let ResourceCycles = [2,2,3];
}
def: InstRW<[BWWriteResGroup131], (instregex "RCL(16|32|64)rCL",
                                             "RCR(16|32|64)rCL")>;

def BWWriteResGroup132 : SchedWriteRes<[BWPort1,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 11;
  let NumMicroOps = 9;
  let ResourceCycles = [1,4,1,3];
}
def: InstRW<[BWWriteResGroup132], (instrs RCL8rCL)>;

def BWWriteResGroup133 : SchedWriteRes<[BWPort06,BWPort0156]> {
  let Latency = 11;
  let NumMicroOps = 11;
  let ResourceCycles = [2,9];
}
def: InstRW<[BWWriteResGroup133], (instrs LOOPE)>;
def: InstRW<[BWWriteResGroup133], (instrs LOOPNE)>;

def BWWriteResGroup135 : SchedWriteRes<[BWPort1,BWPort23]> {
  let Latency = 12;
  let NumMicroOps = 3;
  let ResourceCycles = [2,1];
}
def: InstRW<[BWWriteResGroup135], (instregex "(ADD|SUB|SUBR)_FI(16|32)m")>;

def BWWriteResGroup139_1 : SchedWriteRes<[BWPort0,BWFPDivider]> {
  let Latency = 14;
  let NumMicroOps = 1;
  let ResourceCycles = [1,4];
}
def : SchedAlias<WriteFDiv64, BWWriteResGroup139_1>; // TODO - convert to ZnWriteResFpuPair

def BWWriteResGroup141 : SchedWriteRes<[BWPort0,BWPort1,BWPort23]> {
  let Latency = 14;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup141], (instregex "MUL_FI(16|32)m")>;

def BWWriteResGroup144 : SchedWriteRes<[BWPort1,BWPort6,BWPort23,BWPort0156]> {
  let Latency = 14;
  let NumMicroOps = 8;
  let ResourceCycles = [2,2,1,3];
}
def: InstRW<[BWWriteResGroup144], (instregex "LAR(16|32|64)rr")>;

def BWWriteResGroup145 : SchedWriteRes<[BWPort1,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 14;
  let NumMicroOps = 10;
  let ResourceCycles = [2,3,1,4];
}
def: InstRW<[BWWriteResGroup145], (instrs RCR8rCL)>;

def BWWriteResGroup146 : SchedWriteRes<[BWPort0,BWPort1,BWPort6,BWPort0156]> {
  let Latency = 14;
  let NumMicroOps = 12;
  let ResourceCycles = [2,1,4,5];
}
def: InstRW<[BWWriteResGroup146], (instrs XCH_F)>;

def BWWriteResGroup147 : SchedWriteRes<[BWPort0]> {
  let Latency = 15;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup147], (instregex "DIVR_(FPrST0|FST0r|FrST0)")>;

def BWWriteResGroup149 : SchedWriteRes<[BWPort1,BWPort23,BWPort237,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 15;
  let NumMicroOps = 10;
  let ResourceCycles = [1,1,1,4,1,2];
}
def: InstRW<[BWWriteResGroup149], (instregex "RCL(8|16|32|64)mCL")>;

def BWWriteResGroup150 : SchedWriteRes<[BWPort0,BWPort23,BWFPDivider]> {
  let Latency = 16;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1,5];
}
def : SchedAlias<WriteFDivLd, BWWriteResGroup150>; // TODO - convert to ZnWriteResFpuPair

def BWWriteResGroup153 : SchedWriteRes<[BWPort4,BWPort23,BWPort237,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 16;
  let NumMicroOps = 14;
  let ResourceCycles = [1,1,1,4,2,5];
}
def: InstRW<[BWWriteResGroup153], (instrs CMPXCHG8B)>;

def BWWriteResGroup154 : SchedWriteRes<[BWPort5,BWPort6]> {
  let Latency = 8;
  let NumMicroOps = 20;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup154], (instrs VZEROALL)>;

def BWWriteResGroup159 : SchedWriteRes<[BWPort5,BWPort6,BWPort06,BWPort0156]> {
  let Latency = 18;
  let NumMicroOps = 8;
  let ResourceCycles = [1,1,1,5];
}
def: InstRW<[BWWriteResGroup159], (instrs CPUID)>;
def: InstRW<[BWWriteResGroup159], (instrs RDTSC)>;

def BWWriteResGroup160 : SchedWriteRes<[BWPort1,BWPort23,BWPort237,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 18;
  let NumMicroOps = 11;
  let ResourceCycles = [2,1,1,3,1,3];
}
def: InstRW<[BWWriteResGroup160], (instregex "RCR(8|16|32|64)mCL")>;

def BWWriteResGroup161 : SchedWriteRes<[BWPort0,BWPort23,BWFPDivider]> {
  let Latency = 19;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1,8];
}
def : SchedAlias<WriteFDiv64Ld, BWWriteResGroup161>; // TODO - convert to ZnWriteResFpuPair

def BWWriteResGroup165 : SchedWriteRes<[BWPort0]> {
  let Latency = 20;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}
def: InstRW<[BWWriteResGroup165], (instregex "DIV_(FPrST0|FST0r|FrST0)")>;

def BWWriteResGroup167 : SchedWriteRes<[BWPort4,BWPort5,BWPort6,BWPort23,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 20;
  let NumMicroOps = 8;
  let ResourceCycles = [1,1,1,1,1,1,2];
}
def: InstRW<[BWWriteResGroup167], (instrs INSB, INSL, INSW)>;

def BWWriteResGroup169 : SchedWriteRes<[BWPort0,BWPort23]> {
  let Latency = 21;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup169], (instregex "DIV_F(32|64)m")>;

def BWWriteResGroup171 : SchedWriteRes<[BWPort0,BWPort4,BWPort5,BWPort23,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 21;
  let NumMicroOps = 19;
  let ResourceCycles = [2,1,4,1,1,4,6];
}
def: InstRW<[BWWriteResGroup171], (instrs CMPXCHG16B)>;

def BWWriteResGroup172 : SchedWriteRes<[BWPort6,BWPort23,BWPort0156]> {
  let Latency = 22;
  let NumMicroOps = 18;
  let ResourceCycles = [1,1,16];
}
def: InstRW<[BWWriteResGroup172], (instrs POPF64)>;

def BWWriteResGroup176 : SchedWriteRes<[BWPort6,BWPort23,BWPort0156]> {
  let Latency = 23;
  let NumMicroOps = 19;
  let ResourceCycles = [3,1,15];
}
def: InstRW<[BWWriteResGroup176], (instregex "XRSTOR(64)?")>;

def BWWriteResGroup177 : SchedWriteRes<[BWPort0,BWPort1,BWPort23]> {
  let Latency = 24;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup177], (instregex "DIV_FI(16|32)m")>;

def BWWriteResGroup180 : SchedWriteRes<[BWPort0,BWPort23]> {
  let Latency = 26;
  let NumMicroOps = 2;
  let ResourceCycles = [1,1];
}
def: InstRW<[BWWriteResGroup180], (instregex "DIVR_F(32|64)m")>;

def BWWriteResGroup182 : SchedWriteRes<[BWPort0,BWPort1,BWPort23]> {
  let Latency = 29;
  let NumMicroOps = 3;
  let ResourceCycles = [1,1,1];
}
def: InstRW<[BWWriteResGroup182], (instregex "DIVR_FI(16|32)m")>;

def BWWriteResGroup183_1 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 22;
  let NumMicroOps = 7;
  let ResourceCycles = [1,3,2,1];
}
def: InstRW<[BWWriteResGroup183_1], (instrs VGATHERQPDrm)>;

def BWWriteResGroup183_2 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 23;
  let NumMicroOps = 9;
  let ResourceCycles = [1,3,4,1];
}
def: InstRW<[BWWriteResGroup183_2], (instrs VGATHERQPDYrm)>;

def BWWriteResGroup183_3 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 24;
  let NumMicroOps = 9;
  let ResourceCycles = [1,5,2,1];
}
def: InstRW<[BWWriteResGroup183_3], (instrs VGATHERQPSYrm)>;

def BWWriteResGroup183_4 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 25;
  let NumMicroOps = 7;
  let ResourceCycles = [1,3,2,1];
}
def: InstRW<[BWWriteResGroup183_4], (instrs VGATHERDPDrm,
                                            VGATHERDPSrm)>;

def BWWriteResGroup183_5 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 26;
  let NumMicroOps = 9;
  let ResourceCycles = [1,5,2,1];
}
def: InstRW<[BWWriteResGroup183_5], (instrs VGATHERDPDYrm)>;

def BWWriteResGroup183_6 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 26;
  let NumMicroOps = 14;
  let ResourceCycles = [1,4,8,1];
}
def: InstRW<[BWWriteResGroup183_6], (instrs VGATHERDPSYrm)>;

def BWWriteResGroup183_7 : SchedWriteRes<[BWPort4, BWPort5, BWPort23, BWPort0156]> {
  let Latency = 27;
  let NumMicroOps = 9;
  let ResourceCycles = [1,5,2,1];
}
def: InstRW<[BWWriteResGroup183_7], (instrs VGATHERQPSrm)>;

def BWWriteResGroup185 : SchedWriteRes<[BWPort4,BWPort6,BWPort23,BWPort237,BWPort0156]> {
  let Latency = 29;
  let NumMicroOps = 27;
  let ResourceCycles = [1,5,1,1,19];
}
def: InstRW<[BWWriteResGroup185], (instrs XSAVE64)>;

def BWWriteResGroup186 : SchedWriteRes<[BWPort4,BWPort6,BWPort23,BWPort237,BWPort0156]> {
  let Latency = 30;
  let NumMicroOps = 28;
  let ResourceCycles = [1,6,1,1,19];
}
def: InstRW<[BWWriteResGroup186], (instrs XSAVE)>;
def: InstRW<[BWWriteResGroup186], (instregex "XSAVEC", "XSAVES", "XSAVEOPT")>;

def BWWriteResGroup191 : SchedWriteRes<[BWPort5,BWPort6,BWPort23,BWPort06,BWPort0156]> {
  let Latency = 34;
  let NumMicroOps = 23;
  let ResourceCycles = [1,5,3,4,10];
}
def: InstRW<[BWWriteResGroup191], (instregex "IN(8|16|32)ri",
                                             "IN(8|16|32)rr")>;

def BWWriteResGroup194 : SchedWriteRes<[BWPort5,BWPort6,BWPort23,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 35;
  let NumMicroOps = 23;
  let ResourceCycles = [1,5,2,1,4,10];
}
def: InstRW<[BWWriteResGroup194], (instregex "OUT(8|16|32)ir",
                                             "OUT(8|16|32)rr")>;

def BWWriteResGroup196 : SchedWriteRes<[BWPort5,BWPort0156]> {
  let Latency = 42;
  let NumMicroOps = 22;
  let ResourceCycles = [2,20];
}
def: InstRW<[BWWriteResGroup196], (instrs RDTSCP)>;

def BWWriteResGroup197 : SchedWriteRes<[BWPort0,BWPort01,BWPort23,BWPort05,BWPort06,BWPort015,BWPort0156]> {
  let Latency = 60;
  let NumMicroOps = 64;
  let ResourceCycles = [2,2,8,1,10,2,39];
}
def: InstRW<[BWWriteResGroup197], (instrs FLDENVm)>;

def BWWriteResGroup198 : SchedWriteRes<[BWPort0,BWPort6,BWPort23,BWPort05,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 63;
  let NumMicroOps = 88;
  let ResourceCycles = [4,4,31,1,2,1,45];
}
def: InstRW<[BWWriteResGroup198], (instrs FXRSTOR64)>;

def BWWriteResGroup199 : SchedWriteRes<[BWPort0,BWPort6,BWPort23,BWPort05,BWPort06,BWPort15,BWPort0156]> {
  let Latency = 63;
  let NumMicroOps = 90;
  let ResourceCycles = [4,2,33,1,2,1,47];
}
def: InstRW<[BWWriteResGroup199], (instrs FXRSTOR)>;

def BWWriteResGroup200 : SchedWriteRes<[BWPort5,BWPort01,BWPort0156]> {
  let Latency = 75;
  let NumMicroOps = 15;
  let ResourceCycles = [6,3,6];
}
def: InstRW<[BWWriteResGroup200], (instrs FNINIT)>;

def BWWriteResGroup202 : SchedWriteRes<[BWPort0,BWPort1,BWPort4,BWPort5,BWPort6,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 115;
  let NumMicroOps = 100;
  let ResourceCycles = [9,9,11,8,1,11,21,30];
}
def: InstRW<[BWWriteResGroup202], (instrs FSTENVm)>;

def: InstRW<[WriteZero], (instrs CLC)>;


// Intruction variants handled by the renamer. These might not need execution
// ports in certain conditions.
// See Agner's Fog "The microarchitecture of Intel, AMD and VIA CPUs",
// section "Haswell and Broadwell Pipeline" > "Register allocation and
// renaming".
// These can be investigated with llvm-exegesis, e.g.
// echo 'pxor %mm0, %mm0' | /tmp/llvm-exegesis -mode=uops -snippets-file=-
// echo 'vxorpd %xmm0, %xmm0, %xmm1' | /tmp/llvm-exegesis -mode=uops -snippets-file=-

def BWWriteZeroLatency : SchedWriteRes<[]> {
  let Latency = 0;
}

def BWWriteZeroIdiom : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteALU]>
]>;
def : InstRW<[BWWriteZeroIdiom], (instrs SUB32rr, SUB64rr,
                                         XOR32rr, XOR64rr)>;

def BWWriteFZeroIdiom : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteFLogic]>
]>;
def : InstRW<[BWWriteFZeroIdiom], (instrs XORPSrr, VXORPSrr, XORPDrr,
                                          VXORPDrr)>;

def BWWriteFZeroIdiomY : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteFLogicY]>
]>;
def : InstRW<[BWWriteFZeroIdiomY], (instrs VXORPSYrr, VXORPDYrr)>;

def BWWriteVZeroIdiomLogicX : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteVecLogicX]>
]>;
def : InstRW<[BWWriteVZeroIdiomLogicX], (instrs PXORrr, VPXORrr)>;

def BWWriteVZeroIdiomLogicY : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteVecLogicY]>
]>;
def : InstRW<[BWWriteVZeroIdiomLogicY], (instrs VPXORYrr)>;

def BWWriteVZeroIdiomALUX : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteVecALUX]>
]>;
def : InstRW<[BWWriteVZeroIdiomALUX], (instrs PSUBBrr, VPSUBBrr,
                                              PSUBDrr, VPSUBDrr,
                                              PSUBQrr, VPSUBQrr,
                                              PSUBWrr, VPSUBWrr,
                                              PCMPGTBrr, VPCMPGTBrr,
                                              PCMPGTDrr, VPCMPGTDrr,
                                              PCMPGTWrr, VPCMPGTWrr)>;

def BWWriteVZeroIdiomALUY : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [WriteVecALUY]>
]>;
def : InstRW<[BWWriteVZeroIdiomALUY], (instrs VPSUBBYrr,
                                              VPSUBDYrr,
                                              VPSUBQYrr,
                                              VPSUBWYrr,
                                              VPCMPGTBYrr,
                                              VPCMPGTDYrr,
                                              VPCMPGTWYrr)>;

def BWWritePCMPGTQ : SchedWriteRes<[BWPort0]> {
  let Latency = 5;
  let NumMicroOps = 1;
  let ResourceCycles = [1];
}

def BWWriteVZeroIdiomPCMPGTQ : SchedWriteVariant<[
    SchedVar<MCSchedPredicate<ZeroIdiomPredicate>, [BWWriteZeroLatency]>,
    SchedVar<NoSchedPred,                          [BWWritePCMPGTQ]>
]>;
def : InstRW<[BWWriteVZeroIdiomPCMPGTQ], (instrs PCMPGTQrr, VPCMPGTQrr,
                                                 VPCMPGTQYrr)>;


// CMOVs that use both Z and C flag require an extra uop.
def BWWriteCMOVA_CMOVBErr : SchedWriteRes<[BWPort06,BWPort0156]> {
  let Latency = 2;
  let ResourceCycles = [1,1];
  let NumMicroOps = 2;
}

def BWWriteCMOVA_CMOVBErm : SchedWriteRes<[BWPort23,BWPort06,BWPort0156]> {
  let Latency = 7;
  let ResourceCycles = [1,1,1];
  let NumMicroOps = 3;
}

def BWCMOVA_CMOVBErr :  SchedWriteVariant<[
  SchedVar<MCSchedPredicate<IsCMOVArr_Or_CMOVBErr>, [BWWriteCMOVA_CMOVBErr]>,
  SchedVar<NoSchedPred,                             [WriteCMOV]>
]>;

def BWCMOVA_CMOVBErm :  SchedWriteVariant<[
  SchedVar<MCSchedPredicate<IsCMOVArm_Or_CMOVBErm>, [BWWriteCMOVA_CMOVBErm]>,
  SchedVar<NoSchedPred,                             [WriteCMOV.Folded]>
]>;

def : InstRW<[BWCMOVA_CMOVBErr], (instrs CMOV16rr, CMOV32rr, CMOV64rr)>;
def : InstRW<[BWCMOVA_CMOVBErm], (instrs CMOV16rm, CMOV32rm, CMOV64rm)>;

// SETCCs that use both Z and C flag require an extra uop.
def BWWriteSETA_SETBEr : SchedWriteRes<[BWPort06,BWPort0156]> {
  let Latency = 2;
  let ResourceCycles = [1,1];
  let NumMicroOps = 2;
}

def BWWriteSETA_SETBEm : SchedWriteRes<[BWPort4,BWPort237,BWPort06,BWPort0156]> {
  let Latency = 3;
  let ResourceCycles = [1,1,1,1];
  let NumMicroOps = 4;
}

def BWSETA_SETBErr :  SchedWriteVariant<[
  SchedVar<MCSchedPredicate<IsSETAr_Or_SETBEr>, [BWWriteSETA_SETBEr]>,
  SchedVar<NoSchedPred,                         [WriteSETCC]>
]>;

def BWSETA_SETBErm :  SchedWriteVariant<[
  SchedVar<MCSchedPredicate<IsSETAm_Or_SETBEm>, [BWWriteSETA_SETBEm]>,
  SchedVar<NoSchedPred,                         [WriteSETCCStore]>
]>;

def : InstRW<[BWSETA_SETBErr], (instrs SETCCr)>;
def : InstRW<[BWSETA_SETBErm], (instrs SETCCm)>;

} // SchedModel