llvm.org GIT mirror llvm / b0c326b lib / Target / AArch64 / AArch64LoadStoreOptimizer.cpp
b0c326b

Tree @b0c326b (Download .tar.gz)

AArch64LoadStoreOptimizer.cpp @b0c326braw · history · blame

   1
   2
   3
   4
   5
   6
   7
   8
   9
  10
  11
  12
  13
  14
  15
  16
  17
  18
  19
  20
  21
  22
  23
  24
  25
  26
  27
  28
  29
  30
  31
  32
  33
  34
  35
  36
  37
  38
  39
  40
  41
  42
  43
  44
  45
  46
  47
  48
  49
  50
  51
  52
  53
  54
  55
  56
  57
  58
  59
  60
  61
  62
  63
  64
  65
  66
  67
  68
  69
  70
  71
  72
  73
  74
  75
  76
  77
  78
  79
  80
  81
  82
  83
  84
  85
  86
  87
  88
  89
  90
  91
  92
  93
  94
  95
  96
  97
  98
  99
 100
 101
 102
 103
 104
 105
 106
 107
 108
 109
 110
 111
 112
 113
 114
 115
 116
 117
 118
 119
 120
 121
 122
 123
 124
 125
 126
 127
 128
 129
 130
 131
 132
 133
 134
 135
 136
 137
 138
 139
 140
 141
 142
 143
 144
 145
 146
 147
 148
 149
 150
 151
 152
 153
 154
 155
 156
 157
 158
 159
 160
 161
 162
 163
 164
 165
 166
 167
 168
 169
 170
 171
 172
 173
 174
 175
 176
 177
 178
 179
 180
 181
 182
 183
 184
 185
 186
 187
 188
 189
 190
 191
 192
 193
 194
 195
 196
 197
 198
 199
 200
 201
 202
 203
 204
 205
 206
 207
 208
 209
 210
 211
 212
 213
 214
 215
 216
 217
 218
 219
 220
 221
 222
 223
 224
 225
 226
 227
 228
 229
 230
 231
 232
 233
 234
 235
 236
 237
 238
 239
 240
 241
 242
 243
 244
 245
 246
 247
 248
 249
 250
 251
 252
 253
 254
 255
 256
 257
 258
 259
 260
 261
 262
 263
 264
 265
 266
 267
 268
 269
 270
 271
 272
 273
 274
 275
 276
 277
 278
 279
 280
 281
 282
 283
 284
 285
 286
 287
 288
 289
 290
 291
 292
 293
 294
 295
 296
 297
 298
 299
 300
 301
 302
 303
 304
 305
 306
 307
 308
 309
 310
 311
 312
 313
 314
 315
 316
 317
 318
 319
 320
 321
 322
 323
 324
 325
 326
 327
 328
 329
 330
 331
 332
 333
 334
 335
 336
 337
 338
 339
 340
 341
 342
 343
 344
 345
 346
 347
 348
 349
 350
 351
 352
 353
 354
 355
 356
 357
 358
 359
 360
 361
 362
 363
 364
 365
 366
 367
 368
 369
 370
 371
 372
 373
 374
 375
 376
 377
 378
 379
 380
 381
 382
 383
 384
 385
 386
 387
 388
 389
 390
 391
 392
 393
 394
 395
 396
 397
 398
 399
 400
 401
 402
 403
 404
 405
 406
 407
 408
 409
 410
 411
 412
 413
 414
 415
 416
 417
 418
 419
 420
 421
 422
 423
 424
 425
 426
 427
 428
 429
 430
 431
 432
 433
 434
 435
 436
 437
 438
 439
 440
 441
 442
 443
 444
 445
 446
 447
 448
 449
 450
 451
 452
 453
 454
 455
 456
 457
 458
 459
 460
 461
 462
 463
 464
 465
 466
 467
 468
 469
 470
 471
 472
 473
 474
 475
 476
 477
 478
 479
 480
 481
 482
 483
 484
 485
 486
 487
 488
 489
 490
 491
 492
 493
 494
 495
 496
 497
 498
 499
 500
 501
 502
 503
 504
 505
 506
 507
 508
 509
 510
 511
 512
 513
 514
 515
 516
 517
 518
 519
 520
 521
 522
 523
 524
 525
 526
 527
 528
 529
 530
 531
 532
 533
 534
 535
 536
 537
 538
 539
 540
 541
 542
 543
 544
 545
 546
 547
 548
 549
 550
 551
 552
 553
 554
 555
 556
 557
 558
 559
 560
 561
 562
 563
 564
 565
 566
 567
 568
 569
 570
 571
 572
 573
 574
 575
 576
 577
 578
 579
 580
 581
 582
 583
 584
 585
 586
 587
 588
 589
 590
 591
 592
 593
 594
 595
 596
 597
 598
 599
 600
 601
 602
 603
 604
 605
 606
 607
 608
 609
 610
 611
 612
 613
 614
 615
 616
 617
 618
 619
 620
 621
 622
 623
 624
 625
 626
 627
 628
 629
 630
 631
 632
 633
 634
 635
 636
 637
 638
 639
 640
 641
 642
 643
 644
 645
 646
 647
 648
 649
 650
 651
 652
 653
 654
 655
 656
 657
 658
 659
 660
 661
 662
 663
 664
 665
 666
 667
 668
 669
 670
 671
 672
 673
 674
 675
 676
 677
 678
 679
 680
 681
 682
 683
 684
 685
 686
 687
 688
 689
 690
 691
 692
 693
 694
 695
 696
 697
 698
 699
 700
 701
 702
 703
 704
 705
 706
 707
 708
 709
 710
 711
 712
 713
 714
 715
 716
 717
 718
 719
 720
 721
 722
 723
 724
 725
 726
 727
 728
 729
 730
 731
 732
 733
 734
 735
 736
 737
 738
 739
 740
 741
 742
 743
 744
 745
 746
 747
 748
 749
 750
 751
 752
 753
 754
 755
 756
 757
 758
 759
 760
 761
 762
 763
 764
 765
 766
 767
 768
 769
 770
 771
 772
 773
 774
 775
 776
 777
 778
 779
 780
 781
 782
 783
 784
 785
 786
 787
 788
 789
 790
 791
 792
 793
 794
 795
 796
 797
 798
 799
 800
 801
 802
 803
 804
 805
 806
 807
 808
 809
 810
 811
 812
 813
 814
 815
 816
 817
 818
 819
 820
 821
 822
 823
 824
 825
 826
 827
 828
 829
 830
 831
 832
 833
 834
 835
 836
 837
 838
 839
 840
 841
 842
 843
 844
 845
 846
 847
 848
 849
 850
 851
 852
 853
 854
 855
 856
 857
 858
 859
 860
 861
 862
 863
 864
 865
 866
 867
 868
 869
 870
 871
 872
 873
 874
 875
 876
 877
 878
 879
 880
 881
 882
 883
 884
 885
 886
 887
 888
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
//=- AArch64LoadStoreOptimizer.cpp - AArch64 load/store opt. pass -*- C++ -*-=//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file contains a pass that performs load / store related peephole
// optimizations. This pass should be run after register allocation.
//
//===----------------------------------------------------------------------===//

#include "AArch64InstrInfo.h"
#include "AArch64Subtarget.h"
#include "MCTargetDesc/AArch64AddressingModes.h"
#include "llvm/ADT/BitVector.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineFunctionPass.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineInstrBuilder.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/Target/TargetInstrInfo.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Target/TargetRegisterInfo.h"
using namespace llvm;

#define DEBUG_TYPE "aarch64-ldst-opt"

STATISTIC(NumPairCreated, "Number of load/store pair instructions generated");
STATISTIC(NumPostFolded, "Number of post-index updates folded");
STATISTIC(NumPreFolded, "Number of pre-index updates folded");
STATISTIC(NumUnscaledPairCreated,
          "Number of load/store from unscaled generated");
STATISTIC(NumNarrowLoadsPromoted, "Number of narrow loads promoted");
STATISTIC(NumZeroStoresPromoted, "Number of narrow zero stores promoted");
STATISTIC(NumLoadsFromStoresPromoted, "Number of loads from stores promoted");

// The LdStLimit limits how far we search for load/store pairs.
static cl::opt<unsigned> LdStLimit("aarch64-load-store-scan-limit",
                                   cl::init(20), cl::Hidden);

// The UpdateLimit limits how far we search for update instructions when we form
// pre-/post-index instructions.
static cl::opt<unsigned> UpdateLimit("aarch64-update-scan-limit", cl::init(100),
                                     cl::Hidden);

namespace llvm {
void initializeAArch64LoadStoreOptPass(PassRegistry &);
}

#define AARCH64_LOAD_STORE_OPT_NAME "AArch64 load / store optimization pass"

namespace {

typedef struct LdStPairFlags {
  // If a matching instruction is found, MergeForward is set to true if the
  // merge is to remove the first instruction and replace the second with
  // a pair-wise insn, and false if the reverse is true.
  bool MergeForward;

  // SExtIdx gives the index of the result of the load pair that must be
  // extended. The value of SExtIdx assumes that the paired load produces the
  // value in this order: (I, returned iterator), i.e., -1 means no value has
  // to be extended, 0 means I, and 1 means the returned iterator.
  int SExtIdx;

  LdStPairFlags() : MergeForward(false), SExtIdx(-1) {}

  void setMergeForward(bool V = true) { MergeForward = V; }
  bool getMergeForward() const { return MergeForward; }

  void setSExtIdx(int V) { SExtIdx = V; }
  int getSExtIdx() const { return SExtIdx; }

} LdStPairFlags;

struct AArch64LoadStoreOpt : public MachineFunctionPass {
  static char ID;
  AArch64LoadStoreOpt() : MachineFunctionPass(ID) {
    initializeAArch64LoadStoreOptPass(*PassRegistry::getPassRegistry());
  }

  const AArch64InstrInfo *TII;
  const TargetRegisterInfo *TRI;
  const AArch64Subtarget *Subtarget;

  // Track which registers have been modified and used.
  BitVector ModifiedRegs, UsedRegs;

  // Scan the instructions looking for a load/store that can be combined
  // with the current instruction into a load/store pair.
  // Return the matching instruction if one is found, else MBB->end().
  MachineBasicBlock::iterator findMatchingInsn(MachineBasicBlock::iterator I,
                                               LdStPairFlags &Flags,
                                               unsigned Limit);

  // Scan the instructions looking for a store that writes to the address from
  // which the current load instruction reads. Return true if one is found.
  bool findMatchingStore(MachineBasicBlock::iterator I, unsigned Limit,
                         MachineBasicBlock::iterator &StoreI);

  // Merge the two instructions indicated into a wider instruction.
  MachineBasicBlock::iterator
  mergeNarrowInsns(MachineBasicBlock::iterator I,
                   MachineBasicBlock::iterator MergeMI,
                   const LdStPairFlags &Flags);

  // Merge the two instructions indicated into a single pair-wise instruction.
  MachineBasicBlock::iterator
  mergePairedInsns(MachineBasicBlock::iterator I,
                   MachineBasicBlock::iterator Paired,
                   const LdStPairFlags &Flags);

  // Promote the load that reads directly from the address stored to.
  MachineBasicBlock::iterator
  promoteLoadFromStore(MachineBasicBlock::iterator LoadI,
                       MachineBasicBlock::iterator StoreI);

  // Scan the instruction list to find a base register update that can
  // be combined with the current instruction (a load or store) using
  // pre or post indexed addressing with writeback. Scan forwards.
  MachineBasicBlock::iterator
  findMatchingUpdateInsnForward(MachineBasicBlock::iterator I,
                                int UnscaledOffset, unsigned Limit);

  // Scan the instruction list to find a base register update that can
  // be combined with the current instruction (a load or store) using
  // pre or post indexed addressing with writeback. Scan backwards.
  MachineBasicBlock::iterator
  findMatchingUpdateInsnBackward(MachineBasicBlock::iterator I, unsigned Limit);

  // Find an instruction that updates the base register of the ld/st
  // instruction.
  bool isMatchingUpdateInsn(MachineInstr *MemMI, MachineInstr *MI,
                            unsigned BaseReg, int Offset);

  // Merge a pre- or post-index base register update into a ld/st instruction.
  MachineBasicBlock::iterator
  mergeUpdateInsn(MachineBasicBlock::iterator I,
                  MachineBasicBlock::iterator Update, bool IsPreIdx);

  // Find and merge foldable ldr/str instructions.
  bool tryToMergeLdStInst(MachineBasicBlock::iterator &MBBI);

  // Find and pair ldr/str instructions.
  bool tryToPairLdStInst(MachineBasicBlock::iterator &MBBI);

  // Find and promote load instructions which read directly from store.
  bool tryToPromoteLoadFromStore(MachineBasicBlock::iterator &MBBI);

  // Check if converting two narrow loads into a single wider load with
  // bitfield extracts could be enabled.
  bool enableNarrowLdMerge(MachineFunction &Fn);

  bool optimizeBlock(MachineBasicBlock &MBB, bool enableNarrowLdOpt);

  bool runOnMachineFunction(MachineFunction &Fn) override;

  const char *getPassName() const override {
    return AARCH64_LOAD_STORE_OPT_NAME;
  }
};
char AArch64LoadStoreOpt::ID = 0;
} // namespace

INITIALIZE_PASS(AArch64LoadStoreOpt, "aarch64-ldst-opt",
                AARCH64_LOAD_STORE_OPT_NAME, false, false)

static unsigned getBitExtrOpcode(MachineInstr *MI) {
  switch (MI->getOpcode()) {
  default:
    llvm_unreachable("Unexpected opcode.");
  case AArch64::LDRBBui:
  case AArch64::LDURBBi:
  case AArch64::LDRHHui:
  case AArch64::LDURHHi:
    return AArch64::UBFMWri;
  case AArch64::LDRSBWui:
  case AArch64::LDURSBWi:
  case AArch64::LDRSHWui:
  case AArch64::LDURSHWi:
    return AArch64::SBFMWri;
  }
}

static bool isNarrowStore(unsigned Opc) {
  switch (Opc) {
  default:
    return false;
  case AArch64::STRBBui:
  case AArch64::STURBBi:
  case AArch64::STRHHui:
  case AArch64::STURHHi:
    return true;
  }
}

static bool isNarrowLoad(unsigned Opc) {
  switch (Opc) {
  default:
    return false;
  case AArch64::LDRHHui:
  case AArch64::LDURHHi:
  case AArch64::LDRBBui:
  case AArch64::LDURBBi:
  case AArch64::LDRSHWui:
  case AArch64::LDURSHWi:
  case AArch64::LDRSBWui:
  case AArch64::LDURSBWi:
    return true;
  }
}

static bool isNarrowLoad(MachineInstr *MI) {
  return isNarrowLoad(MI->getOpcode());
}

static bool isNarrowLoadOrStore(unsigned Opc) {
  return isNarrowLoad(Opc) || isNarrowStore(Opc);
}

// Scaling factor for unscaled load or store.
static int getMemScale(MachineInstr *MI) {
  switch (MI->getOpcode()) {
  default:
    llvm_unreachable("Opcode has unknown scale!");
  case AArch64::LDRBBui:
  case AArch64::LDURBBi:
  case AArch64::LDRSBWui:
  case AArch64::LDURSBWi:
  case AArch64::STRBBui:
  case AArch64::STURBBi:
    return 1;
  case AArch64::LDRHHui:
  case AArch64::LDURHHi:
  case AArch64::LDRSHWui:
  case AArch64::LDURSHWi:
  case AArch64::STRHHui:
  case AArch64::STURHHi:
    return 2;
  case AArch64::LDRSui:
  case AArch64::LDURSi:
  case AArch64::LDRSWui:
  case AArch64::LDURSWi:
  case AArch64::LDRWui:
  case AArch64::LDURWi:
  case AArch64::STRSui:
  case AArch64::STURSi:
  case AArch64::STRWui:
  case AArch64::STURWi:
  case AArch64::LDPSi:
  case AArch64::LDPSWi:
  case AArch64::LDPWi:
  case AArch64::STPSi:
  case AArch64::STPWi:
    return 4;
  case AArch64::LDRDui:
  case AArch64::LDURDi:
  case AArch64::LDRXui:
  case AArch64::LDURXi:
  case AArch64::STRDui:
  case AArch64::STURDi:
  case AArch64::STRXui:
  case AArch64::STURXi:
  case AArch64::LDPDi:
  case AArch64::LDPXi:
  case AArch64::STPDi:
  case AArch64::STPXi:
    return 8;
  case AArch64::LDRQui:
  case AArch64::LDURQi:
  case AArch64::STRQui:
  case AArch64::STURQi:
  case AArch64::LDPQi:
  case AArch64::STPQi:
    return 16;
  }
}

static unsigned getMatchingNonSExtOpcode(unsigned Opc,
                                         bool *IsValidLdStrOpc = nullptr) {
  if (IsValidLdStrOpc)
    *IsValidLdStrOpc = true;
  switch (Opc) {
  default:
    if (IsValidLdStrOpc)
      *IsValidLdStrOpc = false;
    return UINT_MAX;
  case AArch64::STRDui:
  case AArch64::STURDi:
  case AArch64::STRQui:
  case AArch64::STURQi:
  case AArch64::STRBBui:
  case AArch64::STURBBi:
  case AArch64::STRHHui:
  case AArch64::STURHHi:
  case AArch64::STRWui:
  case AArch64::STURWi:
  case AArch64::STRXui:
  case AArch64::STURXi:
  case AArch64::LDRDui:
  case AArch64::LDURDi:
  case AArch64::LDRQui:
  case AArch64::LDURQi:
  case AArch64::LDRWui:
  case AArch64::LDURWi:
  case AArch64::LDRXui:
  case AArch64::LDURXi:
  case AArch64::STRSui:
  case AArch64::STURSi:
  case AArch64::LDRSui:
  case AArch64::LDURSi:
  case AArch64::LDRHHui:
  case AArch64::LDURHHi:
  case AArch64::LDRBBui:
  case AArch64::LDURBBi:
    return Opc;
  case AArch64::LDRSWui:
    return AArch64::LDRWui;
  case AArch64::LDURSWi:
    return AArch64::LDURWi;
  case AArch64::LDRSBWui:
    return AArch64::LDRBBui;
  case AArch64::LDRSHWui:
    return AArch64::LDRHHui;
  case AArch64::LDURSBWi:
    return AArch64::LDURBBi;
  case AArch64::LDURSHWi:
    return AArch64::LDURHHi;
  }
}

static unsigned getMatchingWideOpcode(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("Opcode has no wide equivalent!");
  case AArch64::STRBBui:
    return AArch64::STRHHui;
  case AArch64::STRHHui:
    return AArch64::STRWui;
  case AArch64::STURBBi:
    return AArch64::STURHHi;
  case AArch64::STURHHi:
    return AArch64::STURWi;
  case AArch64::STURWi:
    return AArch64::STURXi;
  case AArch64::STRWui:
    return AArch64::STRXui;
  case AArch64::LDRHHui:
  case AArch64::LDRSHWui:
    return AArch64::LDRWui;
  case AArch64::LDURHHi:
  case AArch64::LDURSHWi:
    return AArch64::LDURWi;
  case AArch64::LDRBBui:
  case AArch64::LDRSBWui:
    return AArch64::LDRHHui;
  case AArch64::LDURBBi:
  case AArch64::LDURSBWi:
    return AArch64::LDURHHi;
  }
}

static unsigned getMatchingPairOpcode(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("Opcode has no pairwise equivalent!");
  case AArch64::STRSui:
  case AArch64::STURSi:
    return AArch64::STPSi;
  case AArch64::STRDui:
  case AArch64::STURDi:
    return AArch64::STPDi;
  case AArch64::STRQui:
  case AArch64::STURQi:
    return AArch64::STPQi;
  case AArch64::STRWui:
  case AArch64::STURWi:
    return AArch64::STPWi;
  case AArch64::STRXui:
  case AArch64::STURXi:
    return AArch64::STPXi;
  case AArch64::LDRSui:
  case AArch64::LDURSi:
    return AArch64::LDPSi;
  case AArch64::LDRDui:
  case AArch64::LDURDi:
    return AArch64::LDPDi;
  case AArch64::LDRQui:
  case AArch64::LDURQi:
    return AArch64::LDPQi;
  case AArch64::LDRWui:
  case AArch64::LDURWi:
    return AArch64::LDPWi;
  case AArch64::LDRXui:
  case AArch64::LDURXi:
    return AArch64::LDPXi;
  case AArch64::LDRSWui:
  case AArch64::LDURSWi:
    return AArch64::LDPSWi;
  }
}

static unsigned isMatchingStore(MachineInstr *LoadInst,
                                MachineInstr *StoreInst) {
  unsigned LdOpc = LoadInst->getOpcode();
  unsigned StOpc = StoreInst->getOpcode();
  switch (LdOpc) {
  default:
    llvm_unreachable("Unsupported load instruction!");
  case AArch64::LDRBBui:
    return StOpc == AArch64::STRBBui || StOpc == AArch64::STRHHui ||
           StOpc == AArch64::STRWui || StOpc == AArch64::STRXui;
  case AArch64::LDURBBi:
    return StOpc == AArch64::STURBBi || StOpc == AArch64::STURHHi ||
           StOpc == AArch64::STURWi || StOpc == AArch64::STURXi;
  case AArch64::LDRHHui:
    return StOpc == AArch64::STRHHui || StOpc == AArch64::STRWui ||
           StOpc == AArch64::STRXui;
  case AArch64::LDURHHi:
    return StOpc == AArch64::STURHHi || StOpc == AArch64::STURWi ||
           StOpc == AArch64::STURXi;
  case AArch64::LDRWui:
    return StOpc == AArch64::STRWui || StOpc == AArch64::STRXui;
  case AArch64::LDURWi:
    return StOpc == AArch64::STURWi || StOpc == AArch64::STURXi;
  case AArch64::LDRXui:
    return StOpc == AArch64::STRXui;
  case AArch64::LDURXi:
    return StOpc == AArch64::STURXi;
  }
}

static unsigned getPreIndexedOpcode(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("Opcode has no pre-indexed equivalent!");
  case AArch64::STRSui:
    return AArch64::STRSpre;
  case AArch64::STRDui:
    return AArch64::STRDpre;
  case AArch64::STRQui:
    return AArch64::STRQpre;
  case AArch64::STRBBui:
    return AArch64::STRBBpre;
  case AArch64::STRHHui:
    return AArch64::STRHHpre;
  case AArch64::STRWui:
    return AArch64::STRWpre;
  case AArch64::STRXui:
    return AArch64::STRXpre;
  case AArch64::LDRSui:
    return AArch64::LDRSpre;
  case AArch64::LDRDui:
    return AArch64::LDRDpre;
  case AArch64::LDRQui:
    return AArch64::LDRQpre;
  case AArch64::LDRBBui:
    return AArch64::LDRBBpre;
  case AArch64::LDRHHui:
    return AArch64::LDRHHpre;
  case AArch64::LDRWui:
    return AArch64::LDRWpre;
  case AArch64::LDRXui:
    return AArch64::LDRXpre;
  case AArch64::LDRSWui:
    return AArch64::LDRSWpre;
  case AArch64::LDPSi:
    return AArch64::LDPSpre;
  case AArch64::LDPSWi:
    return AArch64::LDPSWpre;
  case AArch64::LDPDi:
    return AArch64::LDPDpre;
  case AArch64::LDPQi:
    return AArch64::LDPQpre;
  case AArch64::LDPWi:
    return AArch64::LDPWpre;
  case AArch64::LDPXi:
    return AArch64::LDPXpre;
  case AArch64::STPSi:
    return AArch64::STPSpre;
  case AArch64::STPDi:
    return AArch64::STPDpre;
  case AArch64::STPQi:
    return AArch64::STPQpre;
  case AArch64::STPWi:
    return AArch64::STPWpre;
  case AArch64::STPXi:
    return AArch64::STPXpre;
  }
}

static unsigned getPostIndexedOpcode(unsigned Opc) {
  switch (Opc) {
  default:
    llvm_unreachable("Opcode has no post-indexed wise equivalent!");
  case AArch64::STRSui:
    return AArch64::STRSpost;
  case AArch64::STRDui:
    return AArch64::STRDpost;
  case AArch64::STRQui:
    return AArch64::STRQpost;
  case AArch64::STRBBui:
    return AArch64::STRBBpost;
  case AArch64::STRHHui:
    return AArch64::STRHHpost;
  case AArch64::STRWui:
    return AArch64::STRWpost;
  case AArch64::STRXui:
    return AArch64::STRXpost;
  case AArch64::LDRSui:
    return AArch64::LDRSpost;
  case AArch64::LDRDui:
    return AArch64::LDRDpost;
  case AArch64::LDRQui:
    return AArch64::LDRQpost;
  case AArch64::LDRBBui:
    return AArch64::LDRBBpost;
  case AArch64::LDRHHui:
    return AArch64::LDRHHpost;
  case AArch64::LDRWui:
    return AArch64::LDRWpost;
  case AArch64::LDRXui:
    return AArch64::LDRXpost;
  case AArch64::LDRSWui:
    return AArch64::LDRSWpost;
  case AArch64::LDPSi:
    return AArch64::LDPSpost;
  case AArch64::LDPSWi:
    return AArch64::LDPSWpost;
  case AArch64::LDPDi:
    return AArch64::LDPDpost;
  case AArch64::LDPQi:
    return AArch64::LDPQpost;
  case AArch64::LDPWi:
    return AArch64::LDPWpost;
  case AArch64::LDPXi:
    return AArch64::LDPXpost;
  case AArch64::STPSi:
    return AArch64::STPSpost;
  case AArch64::STPDi:
    return AArch64::STPDpost;
  case AArch64::STPQi:
    return AArch64::STPQpost;
  case AArch64::STPWi:
    return AArch64::STPWpost;
  case AArch64::STPXi:
    return AArch64::STPXpost;
  }
}

static bool isPairedLdSt(const MachineInstr *MI) {
  switch (MI->getOpcode()) {
  default:
    return false;
  case AArch64::LDPSi:
  case AArch64::LDPSWi:
  case AArch64::LDPDi:
  case AArch64::LDPQi:
  case AArch64::LDPWi:
  case AArch64::LDPXi:
  case AArch64::STPSi:
  case AArch64::STPDi:
  case AArch64::STPQi:
  case AArch64::STPWi:
  case AArch64::STPXi:
    return true;
  }
}

static const MachineOperand &getLdStRegOp(const MachineInstr *MI,
                                          unsigned PairedRegOp = 0) {
  assert(PairedRegOp < 2 && "Unexpected register operand idx.");
  unsigned Idx = isPairedLdSt(MI) ? PairedRegOp : 0;
  return MI->getOperand(Idx);
}

static const MachineOperand &getLdStBaseOp(const MachineInstr *MI) {
  unsigned Idx = isPairedLdSt(MI) ? 2 : 1;
  return MI->getOperand(Idx);
}

static const MachineOperand &getLdStOffsetOp(const MachineInstr *MI) {
  unsigned Idx = isPairedLdSt(MI) ? 3 : 2;
  return MI->getOperand(Idx);
}

static bool isLdOffsetInRangeOfSt(MachineInstr *LoadInst,
                                  MachineInstr *StoreInst,
                                  const AArch64InstrInfo *TII) {
  assert(isMatchingStore(LoadInst, StoreInst) && "Expect only matched ld/st.");
  int LoadSize = getMemScale(LoadInst);
  int StoreSize = getMemScale(StoreInst);
  int UnscaledStOffset = TII->isUnscaledLdSt(StoreInst)
                             ? getLdStOffsetOp(StoreInst).getImm()
                             : getLdStOffsetOp(StoreInst).getImm() * StoreSize;
  int UnscaledLdOffset = TII->isUnscaledLdSt(LoadInst)
                             ? getLdStOffsetOp(LoadInst).getImm()
                             : getLdStOffsetOp(LoadInst).getImm() * LoadSize;
  return (UnscaledStOffset <= UnscaledLdOffset) &&
         (UnscaledLdOffset + LoadSize <= (UnscaledStOffset + StoreSize));
}

static bool isPromotableZeroStoreOpcode(MachineInstr *MI) {
  unsigned Opc = MI->getOpcode();
  return isNarrowStore(Opc) || Opc == AArch64::STRWui || Opc == AArch64::STURWi;
}

static bool isPromotableZeroStoreInst(MachineInstr *MI) {
  return (isPromotableZeroStoreOpcode(MI)) &&
         getLdStRegOp(MI).getReg() == AArch64::WZR;
}

MachineBasicBlock::iterator
AArch64LoadStoreOpt::mergeNarrowInsns(MachineBasicBlock::iterator I,
                                      MachineBasicBlock::iterator MergeMI,
                                      const LdStPairFlags &Flags) {
  MachineBasicBlock::iterator NextI = I;
  ++NextI;
  // If NextI is the second of the two instructions to be merged, we need
  // to skip one further. Either way we merge will invalidate the iterator,
  // and we don't need to scan the new instruction, as it's a pairwise
  // instruction, which we're not considering for further action anyway.
  if (NextI == MergeMI)
    ++NextI;

  unsigned Opc = I->getOpcode();
  bool IsScaled = !TII->isUnscaledLdSt(Opc);
  int OffsetStride = IsScaled ? 1 : getMemScale(I);

  bool MergeForward = Flags.getMergeForward();
  // Insert our new paired instruction after whichever of the paired
  // instructions MergeForward indicates.
  MachineBasicBlock::iterator InsertionPoint = MergeForward ? MergeMI : I;
  // Also based on MergeForward is from where we copy the base register operand
  // so we get the flags compatible with the input code.
  const MachineOperand &BaseRegOp =
      MergeForward ? getLdStBaseOp(MergeMI) : getLdStBaseOp(I);

  // Which register is Rt and which is Rt2 depends on the offset order.
  MachineInstr *RtMI, *Rt2MI;
  if (getLdStOffsetOp(I).getImm() ==
      getLdStOffsetOp(MergeMI).getImm() + OffsetStride) {
    RtMI = MergeMI;
    Rt2MI = I;
  } else {
    RtMI = I;
    Rt2MI = MergeMI;
  }

  int OffsetImm = getLdStOffsetOp(RtMI).getImm();
  // Change the scaled offset from small to large type.
  if (IsScaled) {
    assert(((OffsetImm & 1) == 0) && "Unexpected offset to merge");
    OffsetImm /= 2;
  }

  DebugLoc DL = I->getDebugLoc();
  MachineBasicBlock *MBB = I->getParent();
  if (isNarrowLoad(Opc)) {
    MachineInstr *RtNewDest = MergeForward ? I : MergeMI;
    // When merging small (< 32 bit) loads for big-endian targets, the order of
    // the component parts gets swapped.
    if (!Subtarget->isLittleEndian())
      std::swap(RtMI, Rt2MI);
    // Construct the new load instruction.
    MachineInstr *NewMemMI, *BitExtMI1, *BitExtMI2;
    NewMemMI =
        BuildMI(*MBB, InsertionPoint, DL, TII->get(getMatchingWideOpcode(Opc)))
            .addOperand(getLdStRegOp(RtNewDest))
            .addOperand(BaseRegOp)
            .addImm(OffsetImm)
            .setMemRefs(I->mergeMemRefsWith(*MergeMI));

    DEBUG(
        dbgs()
        << "Creating the new load and extract. Replacing instructions:\n    ");
    DEBUG(I->print(dbgs()));
    DEBUG(dbgs() << "    ");
    DEBUG(MergeMI->print(dbgs()));
    DEBUG(dbgs() << "  with instructions:\n    ");
    DEBUG((NewMemMI)->print(dbgs()));

    int Width = getMemScale(I) == 1 ? 8 : 16;
    int LSBLow = 0;
    int LSBHigh = Width;
    int ImmsLow = LSBLow + Width - 1;
    int ImmsHigh = LSBHigh + Width - 1;
    MachineInstr *ExtDestMI = MergeForward ? MergeMI : I;
    if ((ExtDestMI == Rt2MI) == Subtarget->isLittleEndian()) {
      // Create the bitfield extract for high bits.
      BitExtMI1 =
          BuildMI(*MBB, InsertionPoint, DL, TII->get(getBitExtrOpcode(Rt2MI)))
              .addOperand(getLdStRegOp(Rt2MI))
              .addReg(getLdStRegOp(RtNewDest).getReg())
              .addImm(LSBHigh)
              .addImm(ImmsHigh);
      // Create the bitfield extract for low bits.
      if (RtMI->getOpcode() == getMatchingNonSExtOpcode(RtMI->getOpcode())) {
        // For unsigned, prefer to use AND for low bits.
        BitExtMI2 = BuildMI(*MBB, InsertionPoint, DL, TII->get(AArch64::ANDWri))
                        .addOperand(getLdStRegOp(RtMI))
                        .addReg(getLdStRegOp(RtNewDest).getReg())
                        .addImm(ImmsLow);
      } else {
        BitExtMI2 =
            BuildMI(*MBB, InsertionPoint, DL, TII->get(getBitExtrOpcode(RtMI)))
                .addOperand(getLdStRegOp(RtMI))
                .addReg(getLdStRegOp(RtNewDest).getReg())
                .addImm(LSBLow)
                .addImm(ImmsLow);
      }
    } else {
      // Create the bitfield extract for low bits.
      if (RtMI->getOpcode() == getMatchingNonSExtOpcode(RtMI->getOpcode())) {
        // For unsigned, prefer to use AND for low bits.
        BitExtMI1 = BuildMI(*MBB, InsertionPoint, DL, TII->get(AArch64::ANDWri))
                        .addOperand(getLdStRegOp(RtMI))
                        .addReg(getLdStRegOp(RtNewDest).getReg())
                        .addImm(ImmsLow);
      } else {
        BitExtMI1 =
            BuildMI(*MBB, InsertionPoint, DL, TII->get(getBitExtrOpcode(RtMI)))
                .addOperand(getLdStRegOp(RtMI))
                .addReg(getLdStRegOp(RtNewDest).getReg())
                .addImm(LSBLow)
                .addImm(ImmsLow);
      }

      // Create the bitfield extract for high bits.
      BitExtMI2 =
          BuildMI(*MBB, InsertionPoint, DL, TII->get(getBitExtrOpcode(Rt2MI)))
              .addOperand(getLdStRegOp(Rt2MI))
              .addReg(getLdStRegOp(RtNewDest).getReg())
              .addImm(LSBHigh)
              .addImm(ImmsHigh);
    }
    DEBUG(dbgs() << "    ");
    DEBUG((BitExtMI1)->print(dbgs()));
    DEBUG(dbgs() << "    ");
    DEBUG((BitExtMI2)->print(dbgs()));
    DEBUG(dbgs() << "\n");

    // Erase the old instructions.
    I->eraseFromParent();
    MergeMI->eraseFromParent();
    return NextI;
  }
  assert(isPromotableZeroStoreInst(I) && "Expected promotable zero store");

  // Construct the new instruction.
  MachineInstrBuilder MIB;
  MIB = BuildMI(*MBB, InsertionPoint, DL, TII->get(getMatchingWideOpcode(Opc)))
            .addReg(isNarrowStore(Opc) ? AArch64::WZR : AArch64::XZR)
            .addOperand(BaseRegOp)
            .addImm(OffsetImm)
            .setMemRefs(I->mergeMemRefsWith(*MergeMI));

  (void)MIB;

  DEBUG(dbgs() << "Creating wider load/store. Replacing instructions:\n    ");
  DEBUG(I->print(dbgs()));
  DEBUG(dbgs() << "    ");
  DEBUG(MergeMI->print(dbgs()));
  DEBUG(dbgs() << "  with instruction:\n    ");
  DEBUG(((MachineInstr *)MIB)->print(dbgs()));
  DEBUG(dbgs() << "\n");

  // Erase the old instructions.
  I->eraseFromParent();
  MergeMI->eraseFromParent();
  return NextI;
}

MachineBasicBlock::iterator
AArch64LoadStoreOpt::mergePairedInsns(MachineBasicBlock::iterator I,
                                      MachineBasicBlock::iterator Paired,
                                      const LdStPairFlags &Flags) {
  MachineBasicBlock::iterator NextI = I;
  ++NextI;
  // If NextI is the second of the two instructions to be merged, we need
  // to skip one further. Either way we merge will invalidate the iterator,
  // and we don't need to scan the new instruction, as it's a pairwise
  // instruction, which we're not considering for further action anyway.
  if (NextI == Paired)
    ++NextI;

  int SExtIdx = Flags.getSExtIdx();
  unsigned Opc =
      SExtIdx == -1 ? I->getOpcode() : getMatchingNonSExtOpcode(I->getOpcode());
  bool IsUnscaled = TII->isUnscaledLdSt(Opc);
  int OffsetStride = IsUnscaled ? getMemScale(I) : 1;

  bool MergeForward = Flags.getMergeForward();
  // Insert our new paired instruction after whichever of the paired
  // instructions MergeForward indicates.
  MachineBasicBlock::iterator InsertionPoint = MergeForward ? Paired : I;
  // Also based on MergeForward is from where we copy the base register operand
  // so we get the flags compatible with the input code.
  const MachineOperand &BaseRegOp =
      MergeForward ? getLdStBaseOp(Paired) : getLdStBaseOp(I);

  int Offset = getLdStOffsetOp(I).getImm();
  int PairedOffset = getLdStOffsetOp(Paired).getImm();
  bool PairedIsUnscaled = TII->isUnscaledLdSt(Paired->getOpcode());
  if (IsUnscaled != PairedIsUnscaled) {
    // We're trying to pair instructions that differ in how they are scaled.  If
    // I is scaled then scale the offset of Paired accordingly.  Otherwise, do
    // the opposite (i.e., make Paired's offset unscaled).
    int MemSize = getMemScale(Paired);
    if (PairedIsUnscaled) {
      // If the unscaled offset isn't a multiple of the MemSize, we can't
      // pair the operations together.
      assert(!(PairedOffset % getMemScale(Paired)) &&
             "Offset should be a multiple of the stride!");
      PairedOffset /= MemSize;
    } else {
      PairedOffset *= MemSize;
    }
  }

  // Which register is Rt and which is Rt2 depends on the offset order.
  MachineInstr *RtMI, *Rt2MI;
  if (Offset == PairedOffset + OffsetStride) {
    RtMI = Paired;
    Rt2MI = I;
    // Here we swapped the assumption made for SExtIdx.
    // I.e., we turn ldp I, Paired into ldp Paired, I.
    // Update the index accordingly.
    if (SExtIdx != -1)
      SExtIdx = (SExtIdx + 1) % 2;
  } else {
    RtMI = I;
    Rt2MI = Paired;
  }
  int OffsetImm = getLdStOffsetOp(RtMI).getImm();
  // Scale the immediate offset, if necessary.
  if (TII->isUnscaledLdSt(RtMI->getOpcode())) {
    assert(!(OffsetImm % getMemScale(RtMI)) &&
           "Unscaled offset cannot be scaled.");
    OffsetImm /= getMemScale(RtMI);
  }

  // Construct the new instruction.
  MachineInstrBuilder MIB;
  DebugLoc DL = I->getDebugLoc();
  MachineBasicBlock *MBB = I->getParent();
  MIB = BuildMI(*MBB, InsertionPoint, DL, TII->get(getMatchingPairOpcode(Opc)))
            .addOperand(getLdStRegOp(RtMI))
            .addOperand(getLdStRegOp(Rt2MI))
            .addOperand(BaseRegOp)
            .addImm(OffsetImm)
            .setMemRefs(I->mergeMemRefsWith(*Paired));

  (void)MIB;

  DEBUG(dbgs() << "Creating pair load/store. Replacing instructions:\n    ");
  DEBUG(I->print(dbgs()));
  DEBUG(dbgs() << "    ");
  DEBUG(Paired->print(dbgs()));
  DEBUG(dbgs() << "  with instruction:\n    ");
  if (SExtIdx != -1) {
    // Generate the sign extension for the proper result of the ldp.
    // I.e., with X1, that would be:
    // %W1<def> = KILL %W1, %X1<imp-def>
    // %X1<def> = SBFMXri %X1<kill>, 0, 31
    MachineOperand &DstMO = MIB->getOperand(SExtIdx);
    // Right now, DstMO has the extended register, since it comes from an
    // extended opcode.
    unsigned DstRegX = DstMO.getReg();
    // Get the W variant of that register.
    unsigned DstRegW = TRI->getSubReg(DstRegX, AArch64::sub_32);
    // Update the result of LDP to use the W instead of the X variant.
    DstMO.setReg(DstRegW);
    DEBUG(((MachineInstr *)MIB)->print(dbgs()));
    DEBUG(dbgs() << "\n");
    // Make the machine verifier happy by providing a definition for
    // the X register.
    // Insert this definition right after the generated LDP, i.e., before
    // InsertionPoint.
    MachineInstrBuilder MIBKill =
        BuildMI(*MBB, InsertionPoint, DL, TII->get(TargetOpcode::KILL), DstRegW)
            .addReg(DstRegW)
            .addReg(DstRegX, RegState::Define);
    MIBKill->getOperand(2).setImplicit();
    // Create the sign extension.
    MachineInstrBuilder MIBSXTW =
        BuildMI(*MBB, InsertionPoint, DL, TII->get(AArch64::SBFMXri), DstRegX)
            .addReg(DstRegX)
            .addImm(0)
            .addImm(31);
    (void)MIBSXTW;
    DEBUG(dbgs() << "  Extend operand:\n    ");
    DEBUG(((MachineInstr *)MIBSXTW)->print(dbgs()));
  } else {
    DEBUG(((MachineInstr *)MIB)->print(dbgs()));
  }
  DEBUG(dbgs() << "\n");

  // Erase the old instructions.
  I->eraseFromParent();
  Paired->eraseFromParent();

  return NextI;
}

MachineBasicBlock::iterator
AArch64LoadStoreOpt::promoteLoadFromStore(MachineBasicBlock::iterator LoadI,
                                          MachineBasicBlock::iterator StoreI) {
  MachineBasicBlock::iterator NextI = LoadI;
  ++NextI;

  int LoadSize = getMemScale(LoadI);
  int StoreSize = getMemScale(StoreI);
  unsigned LdRt = getLdStRegOp(LoadI).getReg();
  unsigned StRt = getLdStRegOp(StoreI).getReg();
  bool IsStoreXReg = TRI->getRegClass(AArch64::GPR64RegClassID)->contains(StRt);

  assert((IsStoreXReg ||
          TRI->getRegClass(AArch64::GPR32RegClassID)->contains(StRt)) &&
         "Unexpected RegClass");

  MachineInstr *BitExtMI;
  if (LoadSize == StoreSize && (LoadSize == 4 || LoadSize == 8)) {
    // Remove the load, if the destination register of the loads is the same
    // register for stored value.
    if (StRt == LdRt && LoadSize == 8) {
      DEBUG(dbgs() << "Remove load instruction:\n    ");
      DEBUG(LoadI->print(dbgs()));
      DEBUG(dbgs() << "\n");
      LoadI->eraseFromParent();
      return NextI;
    }
    // Replace the load with a mov if the load and store are in the same size.
    BitExtMI =
        BuildMI(*LoadI->getParent(), LoadI, LoadI->getDebugLoc(),
                TII->get(IsStoreXReg ? AArch64::ORRXrs : AArch64::ORRWrs), LdRt)
            .addReg(IsStoreXReg ? AArch64::XZR : AArch64::WZR)
            .addReg(StRt)
            .addImm(AArch64_AM::getShifterImm(AArch64_AM::LSL, 0));
  } else {
    // FIXME: Currently we disable this transformation in big-endian targets as
    // performance and correctness are verified only in little-endian.
    if (!Subtarget->isLittleEndian())
      return NextI;
    bool IsUnscaled = TII->isUnscaledLdSt(LoadI);
    assert(IsUnscaled == TII->isUnscaledLdSt(StoreI) &&
           "Unsupported ld/st match");
    assert(LoadSize <= StoreSize && "Invalid load size");
    int UnscaledLdOffset = IsUnscaled
                               ? getLdStOffsetOp(LoadI).getImm()
                               : getLdStOffsetOp(LoadI).getImm() * LoadSize;
    int UnscaledStOffset = IsUnscaled
                               ? getLdStOffsetOp(StoreI).getImm()
                               : getLdStOffsetOp(StoreI).getImm() * StoreSize;
    int Width = LoadSize * 8;
    int Immr = 8 * (UnscaledLdOffset - UnscaledStOffset);
    int Imms = Immr + Width - 1;
    unsigned DestReg = IsStoreXReg
                           ? TRI->getMatchingSuperReg(LdRt, AArch64::sub_32,
                                                      &AArch64::GPR64RegClass)
                           : LdRt;

    assert((UnscaledLdOffset >= UnscaledStOffset &&
            (UnscaledLdOffset + LoadSize) <= UnscaledStOffset + StoreSize) &&
           "Invalid offset");

    Immr = 8 * (UnscaledLdOffset - UnscaledStOffset);
    Imms = Immr + Width - 1;
    if (UnscaledLdOffset == UnscaledStOffset) {
      uint32_t AndMaskEncoded = ((IsStoreXReg ? 1 : 0) << 12) // N
                                | ((Immr) << 6)               // immr
                                | ((Imms) << 0)               // imms
          ;

      BitExtMI =
          BuildMI(*LoadI->getParent(), LoadI, LoadI->getDebugLoc(),
                  TII->get(IsStoreXReg ? AArch64::ANDXri : AArch64::ANDWri),
                  DestReg)
              .addReg(StRt)
              .addImm(AndMaskEncoded);
    } else {
      BitExtMI =
          BuildMI(*LoadI->getParent(), LoadI, LoadI->getDebugLoc(),
                  TII->get(IsStoreXReg ? AArch64::UBFMXri : AArch64::UBFMWri),
                  DestReg)
              .addReg(StRt)
              .addImm(Immr)
              .addImm(Imms);
    }
  }

  DEBUG(dbgs() << "Promoting load by replacing :\n    ");
  DEBUG(StoreI->print(dbgs()));
  DEBUG(dbgs() << "    ");
  DEBUG(LoadI->print(dbgs()));
  DEBUG(dbgs() << "  with instructions:\n    ");
  DEBUG(StoreI->print(dbgs()));
  DEBUG(dbgs() << "    ");
  DEBUG((BitExtMI)->print(dbgs()));
  DEBUG(dbgs() << "\n");

  // Erase the old instructions.
  LoadI->eraseFromParent();
  return NextI;
}

/// trackRegDefsUses - Remember what registers the specified instruction uses
/// and modifies.
static void trackRegDefsUses(const MachineInstr *MI, BitVector &ModifiedRegs,
                             BitVector &UsedRegs,
                             const TargetRegisterInfo *TRI) {
  for (const MachineOperand &MO : MI->operands()) {
    if (MO.isRegMask())
      ModifiedRegs.setBitsNotInMask(MO.getRegMask());

    if (!MO.isReg())
      continue;
    unsigned Reg = MO.getReg();
    if (!Reg)
      continue;
    if (MO.isDef()) {
      for (MCRegAliasIterator AI(Reg, TRI, true); AI.isValid(); ++AI)
        ModifiedRegs.set(*AI);
    } else {
      assert(MO.isUse() && "Reg operand not a def and not a use?!?");
      for (MCRegAliasIterator AI(Reg, TRI, true); AI.isValid(); ++AI)
        UsedRegs.set(*AI);
    }
  }
}

static bool inBoundsForPair(bool IsUnscaled, int Offset, int OffsetStride) {
  // Convert the byte-offset used by unscaled into an "element" offset used
  // by the scaled pair load/store instructions.
  if (IsUnscaled) {
    // If the byte-offset isn't a multiple of the stride, there's no point
    // trying to match it.
    if (Offset % OffsetStride)
      return false;
    Offset /= OffsetStride;
  }
  return Offset <= 63 && Offset >= -64;
}

// Do alignment, specialized to power of 2 and for signed ints,
// avoiding having to do a C-style cast from uint_64t to int when
// using alignTo from include/llvm/Support/MathExtras.h.
// FIXME: Move this function to include/MathExtras.h?
static int alignTo(int Num, int PowOf2) {
  return (Num + PowOf2 - 1) & ~(PowOf2 - 1);
}

static bool mayAlias(MachineInstr *MIa, MachineInstr *MIb,
                     const AArch64InstrInfo *TII) {
  // One of the instructions must modify memory.
  if (!MIa->mayStore() && !MIb->mayStore())
    return false;

  // Both instructions must be memory operations.
  if (!MIa->mayLoadOrStore() && !MIb->mayLoadOrStore())
    return false;

  return !TII->areMemAccessesTriviallyDisjoint(MIa, MIb);
}

static bool mayAlias(MachineInstr *MIa,
                     SmallVectorImpl<MachineInstr *> &MemInsns,
                     const AArch64InstrInfo *TII) {
  for (auto &MIb : MemInsns)
    if (mayAlias(MIa, MIb, TII))
      return true;

  return false;
}

bool AArch64LoadStoreOpt::findMatchingStore(
    MachineBasicBlock::iterator I, unsigned Limit,
    MachineBasicBlock::iterator &StoreI) {
  MachineBasicBlock::iterator B = I->getParent()->begin();
  MachineBasicBlock::iterator MBBI = I;
  MachineInstr *LoadMI = I;
  unsigned BaseReg = getLdStBaseOp(LoadMI).getReg();

  // If the load is the first instruction in the block, there's obviously
  // not any matching store.
  if (MBBI == B)
    return false;

  // Track which registers have been modified and used between the first insn
  // and the second insn.
  ModifiedRegs.reset();
  UsedRegs.reset();

  unsigned Count = 0;
  do {
    --MBBI;
    MachineInstr *MI = MBBI;

    // Don't count DBG_VALUE instructions towards the search limit.
    if (!MI->isDebugValue())
      ++Count;

    // If the load instruction reads directly from the address to which the
    // store instruction writes and the stored value is not modified, we can
    // promote the load. Since we do not handle stores with pre-/post-index,
    // it's unnecessary to check if BaseReg is modified by the store itself.
    if (MI->mayStore() && isMatchingStore(LoadMI, MI) &&
        BaseReg == getLdStBaseOp(MI).getReg() &&
        isLdOffsetInRangeOfSt(LoadMI, MI, TII) &&
        !ModifiedRegs[getLdStRegOp(MI).getReg()]) {
      StoreI = MBBI;
      return true;
    }

    if (MI->isCall())
      return false;

    // Update modified / uses register lists.
    trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);

    // Otherwise, if the base register is modified, we have no match, so
    // return early.
    if (ModifiedRegs[BaseReg])
      return false;

    // If we encounter a store aliased with the load, return early.
    if (MI->mayStore() && mayAlias(LoadMI, MI, TII))
      return false;
  } while (MBBI != B && Count < Limit);
  return false;
}

// Returns true if these two opcodes can be merged or paired.  Otherwise,
// returns false.
static bool canMergeOpc(unsigned OpcA, unsigned OpcB, LdStPairFlags &Flags,
                        const AArch64InstrInfo *TII) {
  // Opcodes match: nothing more to check.
  if (OpcA == OpcB)
    return true;

  // Try to match a sign-extended load/store with a zero-extended load/store.
  bool IsValidLdStrOpc, PairIsValidLdStrOpc;
  unsigned NonSExtOpc = getMatchingNonSExtOpcode(OpcA, &IsValidLdStrOpc);
  assert(IsValidLdStrOpc &&
         "Given Opc should be a Load or Store with an immediate");
  // OpcA will be the first instruction in the pair.
  if (NonSExtOpc == getMatchingNonSExtOpcode(OpcB, &PairIsValidLdStrOpc)) {
    Flags.setSExtIdx(NonSExtOpc == (unsigned)OpcA ? 1 : 0);
    return true;
  }

  // If the second instruction isn't even a load/store, bail out.
  if (!PairIsValidLdStrOpc)
    return false;

  // FIXME: We don't support merging narrow loads/stores with mixed
  // scaled/unscaled offsets.
  if (isNarrowLoadOrStore(OpcA) || isNarrowLoadOrStore(OpcB))
    return false;

  // Try to match an unscaled load/store with a scaled load/store.
  return TII->isUnscaledLdSt(OpcA) != TII->isUnscaledLdSt(OpcB) &&
         getMatchingPairOpcode(OpcA) == getMatchingPairOpcode(OpcB);

  // FIXME: Can we also match a mixed sext/zext unscaled/scaled pair?
}

/// Scan the instructions looking for a load/store that can be combined with the
/// current instruction into a wider equivalent or a load/store pair.
MachineBasicBlock::iterator
AArch64LoadStoreOpt::findMatchingInsn(MachineBasicBlock::iterator I,
                                      LdStPairFlags &Flags, unsigned Limit) {
  MachineBasicBlock::iterator E = I->getParent()->end();
  MachineBasicBlock::iterator MBBI = I;
  MachineInstr *FirstMI = I;
  ++MBBI;

  unsigned Opc = FirstMI->getOpcode();
  bool MayLoad = FirstMI->mayLoad();
  bool IsUnscaled = TII->isUnscaledLdSt(FirstMI);
  unsigned Reg = getLdStRegOp(FirstMI).getReg();
  unsigned BaseReg = getLdStBaseOp(FirstMI).getReg();
  int Offset = getLdStOffsetOp(FirstMI).getImm();
  int OffsetStride = IsUnscaled ? getMemScale(FirstMI) : 1;
  bool IsPromotableZeroStore = isPromotableZeroStoreInst(FirstMI);

  // Track which registers have been modified and used between the first insn
  // (inclusive) and the second insn.
  ModifiedRegs.reset();
  UsedRegs.reset();

  // Remember any instructions that read/write memory between FirstMI and MI.
  SmallVector<MachineInstr *, 4> MemInsns;

  for (unsigned Count = 0; MBBI != E && Count < Limit; ++MBBI) {
    MachineInstr *MI = MBBI;
    // Skip DBG_VALUE instructions. Otherwise debug info can affect the
    // optimization by changing how far we scan.
    if (MI->isDebugValue())
      continue;

    // Now that we know this is a real instruction, count it.
    ++Count;

    Flags.setSExtIdx(-1);
    if (canMergeOpc(Opc, MI->getOpcode(), Flags, TII) &&
        getLdStOffsetOp(MI).isImm()) {
      assert(MI->mayLoadOrStore() && "Expected memory operation.");
      // If we've found another instruction with the same opcode, check to see
      // if the base and offset are compatible with our starting instruction.
      // These instructions all have scaled immediate operands, so we just
      // check for +1/-1. Make sure to check the new instruction offset is
      // actually an immediate and not a symbolic reference destined for
      // a relocation.
      //
      // Pairwise instructions have a 7-bit signed offset field. Single insns
      // have a 12-bit unsigned offset field. To be a valid combine, the
      // final offset must be in range.
      unsigned MIBaseReg = getLdStBaseOp(MI).getReg();
      int MIOffset = getLdStOffsetOp(MI).getImm();
      bool MIIsUnscaled = TII->isUnscaledLdSt(MI);
      if (IsUnscaled != MIIsUnscaled) {
        // We're trying to pair instructions that differ in how they are scaled.
        // If FirstMI is scaled then scale the offset of MI accordingly.
        // Otherwise, do the opposite (i.e., make MI's offset unscaled).
        int MemSize = getMemScale(MI);
        if (MIIsUnscaled) {
          // If the unscaled offset isn't a multiple of the MemSize, we can't
          // pair the operations together: bail and keep looking.
          if (MIOffset % MemSize)
            continue;
          MIOffset /= MemSize;
        } else {
          MIOffset *= MemSize;
        }
      }

      if (BaseReg == MIBaseReg && ((Offset == MIOffset + OffsetStride) ||
                                   (Offset + OffsetStride == MIOffset))) {
        int MinOffset = Offset < MIOffset ? Offset : MIOffset;
        // If this is a volatile load/store that otherwise matched, stop looking
        // as something is going on that we don't have enough information to
        // safely transform. Similarly, stop if we see a hint to avoid pairs.
        if (MI->hasOrderedMemoryRef() || TII->isLdStPairSuppressed(MI))
          return E;
        // If the resultant immediate offset of merging these instructions
        // is out of range for a pairwise instruction, bail and keep looking.
        bool IsNarrowLoad = isNarrowLoad(MI->getOpcode());
        if (!IsNarrowLoad &&
            !inBoundsForPair(IsUnscaled, MinOffset, OffsetStride)) {
          trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);
          MemInsns.push_back(MI);
          continue;
        }

        if (IsNarrowLoad || IsPromotableZeroStore) {
          // If the alignment requirements of the scaled wide load/store
          // instruction can't express the offset of the scaled narrow
          // input, bail and keep looking.
          if (!IsUnscaled && alignTo(MinOffset, 2) != MinOffset) {
            trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);
            MemInsns.push_back(MI);
            continue;
          }
        } else {
          // If the alignment requirements of the paired (scaled) instruction
          // can't express the offset of the unscaled input, bail and keep
          // looking.
          if (IsUnscaled && (alignTo(MinOffset, OffsetStride) != MinOffset)) {
            trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);
            MemInsns.push_back(MI);
            continue;
          }
        }
        // If the destination register of the loads is the same register, bail
        // and keep looking. A load-pair instruction with both destination
        // registers the same is UNPREDICTABLE and will result in an exception.
        // For narrow stores, allow only when the stored value is the same
        // (i.e., WZR).
        if ((MayLoad && Reg == getLdStRegOp(MI).getReg()) ||
            (IsPromotableZeroStore && Reg != getLdStRegOp(MI).getReg())) {
          trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);
          MemInsns.push_back(MI);
          continue;
        }

        // If the Rt of the second instruction was not modified or used between
        // the two instructions and none of the instructions between the second
        // and first alias with the second, we can combine the second into the
        // first.
        if (!ModifiedRegs[getLdStRegOp(MI).getReg()] &&
            !(MI->mayLoad() && UsedRegs[getLdStRegOp(MI).getReg()]) &&
            !mayAlias(MI, MemInsns, TII)) {
          Flags.setMergeForward(false);
          return MBBI;
        }

        // Likewise, if the Rt of the first instruction is not modified or used
        // between the two instructions and none of the instructions between the
        // first and the second alias with the first, we can combine the first
        // into the second.
        if (!ModifiedRegs[getLdStRegOp(FirstMI).getReg()] &&
            !(MayLoad && UsedRegs[getLdStRegOp(FirstMI).getReg()]) &&
            !mayAlias(FirstMI, MemInsns, TII)) {
          Flags.setMergeForward(true);
          return MBBI;
        }
        // Unable to combine these instructions due to interference in between.
        // Keep looking.
      }
    }

    // If the instruction wasn't a matching load or store.  Stop searching if we
    // encounter a call instruction that might modify memory.
    if (MI->isCall())
      return E;

    // Update modified / uses register lists.
    trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);

    // Otherwise, if the base register is modified, we have no match, so
    // return early.
    if (ModifiedRegs[BaseReg])
      return E;

    // Update list of instructions that read/write memory.
    if (MI->mayLoadOrStore())
      MemInsns.push_back(MI);
  }
  return E;
}

MachineBasicBlock::iterator
AArch64LoadStoreOpt::mergeUpdateInsn(MachineBasicBlock::iterator I,
                                     MachineBasicBlock::iterator Update,
                                     bool IsPreIdx) {
  assert((Update->getOpcode() == AArch64::ADDXri ||
          Update->getOpcode() == AArch64::SUBXri) &&
         "Unexpected base register update instruction to merge!");
  MachineBasicBlock::iterator NextI = I;
  // Return the instruction following the merged instruction, which is
  // the instruction following our unmerged load. Unless that's the add/sub
  // instruction we're merging, in which case it's the one after that.
  if (++NextI == Update)
    ++NextI;

  int Value = Update->getOperand(2).getImm();
  assert(AArch64_AM::getShiftValue(Update->getOperand(3).getImm()) == 0 &&
         "Can't merge 1 << 12 offset into pre-/post-indexed load / store");
  if (Update->getOpcode() == AArch64::SUBXri)
    Value = -Value;

  unsigned NewOpc = IsPreIdx ? getPreIndexedOpcode(I->getOpcode())
                             : getPostIndexedOpcode(I->getOpcode());
  MachineInstrBuilder MIB;
  if (!isPairedLdSt(I)) {
    // Non-paired instruction.
    MIB = BuildMI(*I->getParent(), I, I->getDebugLoc(), TII->get(NewOpc))
              .addOperand(getLdStRegOp(Update))
              .addOperand(getLdStRegOp(I))
              .addOperand(getLdStBaseOp(I))
              .addImm(Value)
              .setMemRefs(I->memoperands_begin(), I->memoperands_end());
  } else {
    // Paired instruction.
    int Scale = getMemScale(I);
    MIB = BuildMI(*I->getParent(), I, I->getDebugLoc(), TII->get(NewOpc))
              .addOperand(getLdStRegOp(Update))
              .addOperand(getLdStRegOp(I, 0))
              .addOperand(getLdStRegOp(I, 1))
              .addOperand(getLdStBaseOp(I))
              .addImm(Value / Scale)
              .setMemRefs(I->memoperands_begin(), I->memoperands_end());
  }
  (void)MIB;

  if (IsPreIdx)
    DEBUG(dbgs() << "Creating pre-indexed load/store.");
  else
    DEBUG(dbgs() << "Creating post-indexed load/store.");
  DEBUG(dbgs() << "    Replacing instructions:\n    ");
  DEBUG(I->print(dbgs()));
  DEBUG(dbgs() << "    ");
  DEBUG(Update->print(dbgs()));
  DEBUG(dbgs() << "  with instruction:\n    ");
  DEBUG(((MachineInstr *)MIB)->print(dbgs()));
  DEBUG(dbgs() << "\n");

  // Erase the old instructions for the block.
  I->eraseFromParent();
  Update->eraseFromParent();

  return NextI;
}

bool AArch64LoadStoreOpt::isMatchingUpdateInsn(MachineInstr *MemMI,
                                               MachineInstr *MI,
                                               unsigned BaseReg, int Offset) {
  switch (MI->getOpcode()) {
  default:
    break;
  case AArch64::SUBXri:
    // Negate the offset for a SUB instruction.
    Offset *= -1;
  // FALLTHROUGH
  case AArch64::ADDXri:
    // Make sure it's a vanilla immediate operand, not a relocation or
    // anything else we can't handle.
    if (!MI->getOperand(2).isImm())
      break;
    // Watch out for 1 << 12 shifted value.
    if (AArch64_AM::getShiftValue(MI->getOperand(3).getImm()))
      break;

    // The update instruction source and destination register must be the
    // same as the load/store base register.
    if (MI->getOperand(0).getReg() != BaseReg ||
        MI->getOperand(1).getReg() != BaseReg)
      break;

    bool IsPairedInsn = isPairedLdSt(MemMI);
    int UpdateOffset = MI->getOperand(2).getImm();
    // For non-paired load/store instructions, the immediate must fit in a
    // signed 9-bit integer.
    if (!IsPairedInsn && (UpdateOffset > 255 || UpdateOffset < -256))
      break;

    // For paired load/store instructions, the immediate must be a multiple of
    // the scaling factor.  The scaled offset must also fit into a signed 7-bit
    // integer.
    if (IsPairedInsn) {
      int Scale = getMemScale(MemMI);
      if (UpdateOffset % Scale != 0)
        break;

      int ScaledOffset = UpdateOffset / Scale;
      if (ScaledOffset > 64 || ScaledOffset < -64)
        break;
    }

    // If we have a non-zero Offset, we check that it matches the amount
    // we're adding to the register.
    if (!Offset || Offset == MI->getOperand(2).getImm())
      return true;
    break;
  }
  return false;
}

MachineBasicBlock::iterator AArch64LoadStoreOpt::findMatchingUpdateInsnForward(
    MachineBasicBlock::iterator I, int UnscaledOffset, unsigned Limit) {
  MachineBasicBlock::iterator E = I->getParent()->end();
  MachineInstr *MemMI = I;
  MachineBasicBlock::iterator MBBI = I;

  unsigned BaseReg = getLdStBaseOp(MemMI).getReg();
  int MIUnscaledOffset = getLdStOffsetOp(MemMI).getImm() * getMemScale(MemMI);

  // Scan forward looking for post-index opportunities.  Updating instructions
  // can't be formed if the memory instruction doesn't have the offset we're
  // looking for.
  if (MIUnscaledOffset != UnscaledOffset)
    return E;

  // If the base register overlaps a destination register, we can't
  // merge the update.
  bool IsPairedInsn = isPairedLdSt(MemMI);
  for (unsigned i = 0, e = IsPairedInsn ? 2 : 1; i != e; ++i) {
    unsigned DestReg = getLdStRegOp(MemMI, i).getReg();
    if (DestReg == BaseReg || TRI->isSubRegister(BaseReg, DestReg))
      return E;
  }

  // Track which registers have been modified and used between the first insn
  // (inclusive) and the second insn.
  ModifiedRegs.reset();
  UsedRegs.reset();
  ++MBBI;
  for (unsigned Count = 0; MBBI != E && Count < Limit; ++MBBI) {
    MachineInstr *MI = MBBI;
    // Skip DBG_VALUE instructions.
    if (MI->isDebugValue())
      continue;

    // Now that we know this is a real instruction, count it.
    ++Count;

    // If we found a match, return it.
    if (isMatchingUpdateInsn(I, MI, BaseReg, UnscaledOffset))
      return MBBI;

    // Update the status of what the instruction clobbered and used.
    trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);

    // Otherwise, if the base register is used or modified, we have no match, so
    // return early.
    if (ModifiedRegs[BaseReg] || UsedRegs[BaseReg])
      return E;
  }
  return E;
}

MachineBasicBlock::iterator AArch64LoadStoreOpt::findMatchingUpdateInsnBackward(
    MachineBasicBlock::iterator I, unsigned Limit) {
  MachineBasicBlock::iterator B = I->getParent()->begin();
  MachineBasicBlock::iterator E = I->getParent()->end();
  MachineInstr *MemMI = I;
  MachineBasicBlock::iterator MBBI = I;

  unsigned BaseReg = getLdStBaseOp(MemMI).getReg();
  int Offset = getLdStOffsetOp(MemMI).getImm();

  // If the load/store is the first instruction in the block, there's obviously
  // not any matching update. Ditto if the memory offset isn't zero.
  if (MBBI == B || Offset != 0)
    return E;
  // If the base register overlaps a destination register, we can't
  // merge the update.
  bool IsPairedInsn = isPairedLdSt(MemMI);
  for (unsigned i = 0, e = IsPairedInsn ? 2 : 1; i != e; ++i) {
    unsigned DestReg = getLdStRegOp(MemMI, i).getReg();
    if (DestReg == BaseReg || TRI->isSubRegister(BaseReg, DestReg))
      return E;
  }

  // Track which registers have been modified and used between the first insn
  // (inclusive) and the second insn.
  ModifiedRegs.reset();
  UsedRegs.reset();
  unsigned Count = 0;
  do {
    --MBBI;
    MachineInstr *MI = MBBI;

    // Don't count DBG_VALUE instructions towards the search limit.
    if (!MI->isDebugValue())
      ++Count;

    // If we found a match, return it.
    if (isMatchingUpdateInsn(I, MI, BaseReg, Offset))
      return MBBI;

    // Update the status of what the instruction clobbered and used.
    trackRegDefsUses(MI, ModifiedRegs, UsedRegs, TRI);

    // Otherwise, if the base register is used or modified, we have no match, so
    // return early.
    if (ModifiedRegs[BaseReg] || UsedRegs[BaseReg])
      return E;
  } while (MBBI != B && Count < Limit);
  return E;
}

bool AArch64LoadStoreOpt::tryToPromoteLoadFromStore(
    MachineBasicBlock::iterator &MBBI) {
  MachineInstr *MI = MBBI;
  // If this is a volatile load, don't mess with it.
  if (MI->hasOrderedMemoryRef())
    return false;

  // Make sure this is a reg+imm.
  // FIXME: It is possible to extend it to handle reg+reg cases.
  if (!getLdStOffsetOp(MI).isImm())
    return false;

  // Look backward up to LdStLimit instructions.
  MachineBasicBlock::iterator StoreI;
  if (findMatchingStore(MBBI, LdStLimit, StoreI)) {
    ++NumLoadsFromStoresPromoted;
    // Promote the load. Keeping the iterator straight is a
    // pain, so we let the merge routine tell us what the next instruction
    // is after it's done mucking about.
    MBBI = promoteLoadFromStore(MBBI, StoreI);
    return true;
  }
  return false;
}

// Find narrow loads that can be converted into a single wider load with
// bitfield extract instructions.  Also merge adjacent zero stores into a wider
// store.
bool AArch64LoadStoreOpt::tryToMergeLdStInst(
    MachineBasicBlock::iterator &MBBI) {
  assert((isNarrowLoad(MBBI) || isPromotableZeroStoreOpcode(MBBI)) &&
         "Expected narrow op.");
  MachineInstr *MI = MBBI;
  MachineBasicBlock::iterator E = MI->getParent()->end();

  if (!TII->isCandidateToMergeOrPair(MI))
    return false;

  // For promotable zero stores, the stored value should be WZR.
  if (isPromotableZeroStoreOpcode(MI) &&
      getLdStRegOp(MI).getReg() != AArch64::WZR)
    return false;

  // Look ahead up to LdStLimit instructions for a mergable instruction.
  LdStPairFlags Flags;
  MachineBasicBlock::iterator MergeMI =
      findMatchingInsn(MBBI, Flags, LdStLimit);
  if (MergeMI != E) {
    if (isNarrowLoad(MI)) {
      ++NumNarrowLoadsPromoted;
    } else if (isPromotableZeroStoreInst(MI)) {
      ++NumZeroStoresPromoted;
    }
    // Keeping the iterator straight is a pain, so we let the merge routine tell
    // us what the next instruction is after it's done mucking about.
    MBBI = mergeNarrowInsns(MBBI, MergeMI, Flags);
    return true;
  }
  return false;
}

// Find loads and stores that can be merged into a single load or store pair
// instruction.
bool AArch64LoadStoreOpt::tryToPairLdStInst(MachineBasicBlock::iterator &MBBI) {
  MachineInstr *MI = MBBI;
  MachineBasicBlock::iterator E = MI->getParent()->end();

  if (!TII->isCandidateToMergeOrPair(MI))
    return false;

  // Early exit if the offset is not possible to match. (6 bits of positive
  // range, plus allow an extra one in case we find a later insn that matches
  // with Offset-1)
  bool IsUnscaled = TII->isUnscaledLdSt(MI);
  int Offset = getLdStOffsetOp(MI).getImm();
  int OffsetStride = IsUnscaled ? getMemScale(MI) : 1;
  if (!inBoundsForPair(IsUnscaled, Offset, OffsetStride))
    return false;

  // Look ahead up to LdStLimit instructions for a pairable instruction.
  LdStPairFlags Flags;
  MachineBasicBlock::iterator Paired = findMatchingInsn(MBBI, Flags, LdStLimit);
  if (Paired != E) {
    ++NumPairCreated;
    if (TII->isUnscaledLdSt(MI))
      ++NumUnscaledPairCreated;
    // Keeping the iterator straight is a pain, so we let the merge routine tell
    // us what the next instruction is after it's done mucking about.
    MBBI = mergePairedInsns(MBBI, Paired, Flags);
    return true;
  }
  return false;
}

bool AArch64LoadStoreOpt::optimizeBlock(MachineBasicBlock &MBB,
                                        bool enableNarrowLdOpt) {
  bool Modified = false;
  // Four tranformations to do here:
  // 1) Find loads that directly read from stores and promote them by
  //    replacing with mov instructions. If the store is wider than the load,
  //    the load will be replaced with a bitfield extract.
  //      e.g.,
  //        str w1, [x0, #4]
  //        ldrh w2, [x0, #6]
  //        ; becomes
  //        str w1, [x0, #4]
  //        lsr	w2, w1, #16
  for (MachineBasicBlock::iterator MBBI = MBB.begin(), E = MBB.end();
       MBBI != E;) {
    MachineInstr *MI = MBBI;
    switch (MI->getOpcode()) {
    default:
      // Just move on to the next instruction.
      ++MBBI;
      break;
    // Scaled instructions.
    case AArch64::LDRBBui:
    case AArch64::LDRHHui:
    case AArch64::LDRWui:
    case AArch64::LDRXui:
    // Unscaled instructions.
    case AArch64::LDURBBi:
    case AArch64::LDURHHi:
    case AArch64::LDURWi:
    case AArch64::LDURXi: {
      if (tryToPromoteLoadFromStore(MBBI)) {
        Modified = true;
        break;
      }
      ++MBBI;
      break;
    }
    }
  }
  // 2) Find narrow loads that can be converted into a single wider load
  //    with bitfield extract instructions.
  //      e.g.,
  //        ldrh w0, [x2]
  //        ldrh w1, [x2, #2]
  //        ; becomes
  //        ldr w0, [x2]
  //        ubfx w1, w0, #16, #16
  //        and w0, w0, #ffff
  //
  //    Also merge adjacent zero stores into a wider store.
  //      e.g.,
  //        strh wzr, [x0]
  //        strh wzr, [x0, #2]
  //        ; becomes
  //        str wzr, [x0]
  for (MachineBasicBlock::iterator MBBI = MBB.begin(), E = MBB.end();
       enableNarrowLdOpt && MBBI != E;) {
    MachineInstr *MI = MBBI;
    switch (MI->getOpcode()) {
    default:
      // Just move on to the next instruction.
      ++MBBI;
      break;
    // Scaled instructions.
    case AArch64::LDRBBui:
    case AArch64::LDRHHui:
    case AArch64::LDRSBWui:
    case AArch64::LDRSHWui:
    case AArch64::STRBBui:
    case AArch64::STRHHui:
    case AArch64::STRWui:
    // Unscaled instructions.
    case AArch64::LDURBBi:
    case AArch64::LDURHHi:
    case AArch64::LDURSBWi:
    case AArch64::LDURSHWi:
    case AArch64::STURBBi:
    case AArch64::STURHHi:
    case AArch64::STURWi: {
      if (tryToMergeLdStInst(MBBI)) {
        Modified = true;
        break;
      }
      ++MBBI;
      break;
    }
    }
  }
  // 3) Find loads and stores that can be merged into a single load or store
  //    pair instruction.
  //      e.g.,
  //        ldr x0, [x2]
  //        ldr x1, [x2, #8]
  //        ; becomes
  //        ldp x0, x1, [x2]
  for (MachineBasicBlock::iterator MBBI = MBB.begin(), E = MBB.end();
       MBBI != E;) {
    MachineInstr *MI = MBBI;
    switch (MI->getOpcode()) {
    default:
      // Just move on to the next instruction.
      ++MBBI;
      break;
    // Scaled instructions.
    case AArch64::STRSui:
    case AArch64::STRDui:
    case AArch64::STRQui:
    case AArch64::STRXui:
    case AArch64::STRWui:
    case AArch64::LDRSui:
    case AArch64::LDRDui:
    case AArch64::LDRQui:
    case AArch64::LDRXui:
    case AArch64::LDRWui:
    case AArch64::LDRSWui:
    // Unscaled instructions.
    case AArch64::STURSi:
    case AArch64::STURDi:
    case AArch64::STURQi:
    case AArch64::STURWi:
    case AArch64::STURXi:
    case AArch64::LDURSi:
    case AArch64::LDURDi:
    case AArch64::LDURQi:
    case AArch64::LDURWi:
    case AArch64::LDURXi:
    case AArch64::LDURSWi: {
      if (tryToPairLdStInst(MBBI)) {
        Modified = true;
        break;
      }
      ++MBBI;
      break;
    }
    }
  }
  // 4) Find base register updates that can be merged into the load or store
  //    as a base-reg writeback.
  //      e.g.,
  //        ldr x0, [x2]
  //        add x2, x2, #4
  //        ; becomes
  //        ldr x0, [x2], #4
  for (MachineBasicBlock::iterator MBBI = MBB.begin(), E = MBB.end();
       MBBI != E;) {
    MachineInstr *MI = MBBI;
    // Do update merging. It's simpler to keep this separate from the above
    // switchs, though not strictly necessary.
    unsigned Opc = MI->getOpcode();
    switch (Opc) {
    default:
      // Just move on to the next instruction.
      ++MBBI;
      break;
    // Scaled instructions.
    case AArch64::STRSui:
    case AArch64::STRDui:
    case AArch64::STRQui:
    case AArch64::STRXui:
    case AArch64::STRWui:
    case AArch64::STRHHui:
    case AArch64::STRBBui:
    case AArch64::LDRSui:
    case AArch64::LDRDui:
    case AArch64::LDRQui:
    case AArch64::LDRXui:
    case AArch64::LDRWui:
    case AArch64::LDRHHui:
    case AArch64::LDRBBui:
    // Unscaled instructions.
    case AArch64::STURSi:
    case AArch64::STURDi:
    case AArch64::STURQi:
    case AArch64::STURWi:
    case AArch64::STURXi:
    case AArch64::LDURSi:
    case AArch64::LDURDi:
    case AArch64::LDURQi:
    case AArch64::LDURWi:
    case AArch64::LDURXi:
    // Paired instructions.
    case AArch64::LDPSi:
    case AArch64::LDPSWi:
    case AArch64::LDPDi:
    case AArch64::LDPQi:
    case AArch64::LDPWi:
    case AArch64::LDPXi:
    case AArch64::STPSi:
    case AArch64::STPDi:
    case AArch64::STPQi:
    case AArch64::STPWi:
    case AArch64::STPXi: {
      // Make sure this is a reg+imm (as opposed to an address reloc).
      if (!getLdStOffsetOp(MI).isImm()) {
        ++MBBI;
        break;
      }
      // Look forward to try to form a post-index instruction. For example,
      // ldr x0, [x20]
      // add x20, x20, #32
      //   merged into:
      // ldr x0, [x20], #32
      MachineBasicBlock::iterator Update =
          findMatchingUpdateInsnForward(MBBI, 0, UpdateLimit);
      if (Update != E) {
        // Merge the update into the ld/st.
        MBBI = mergeUpdateInsn(MBBI, Update, /*IsPreIdx=*/false);
        Modified = true;
        ++NumPostFolded;
        break;
      }
      // Don't know how to handle pre/post-index versions, so move to the next
      // instruction.
      if (TII->isUnscaledLdSt(Opc)) {
        ++MBBI;
        break;
      }

      // Look back to try to find a pre-index instruction. For example,
      // add x0, x0, #8
      // ldr x1, [x0]
      //   merged into:
      // ldr x1, [x0, #8]!
      Update = findMatchingUpdateInsnBackward(MBBI, UpdateLimit);
      if (Update != E) {
        // Merge the update into the ld/st.
        MBBI = mergeUpdateInsn(MBBI, Update, /*IsPreIdx=*/true);
        Modified = true;
        ++NumPreFolded;
        break;
      }
      // The immediate in the load/store is scaled by the size of the memory
      // operation. The immediate in the add we're looking for,
      // however, is not, so adjust here.
      int UnscaledOffset = getLdStOffsetOp(MI).getImm() * getMemScale(MI);

      // Look forward to try to find a post-index instruction. For example,
      // ldr x1, [x0, #64]
      // add x0, x0, #64
      //   merged into:
      // ldr x1, [x0, #64]!
      Update = findMatchingUpdateInsnForward(MBBI, UnscaledOffset, UpdateLimit);
      if (Update != E) {
        // Merge the update into the ld/st.
        MBBI = mergeUpdateInsn(MBBI, Update, /*IsPreIdx=*/true);
        Modified = true;
        ++NumPreFolded;
        break;
      }

      // Nothing found. Just move to the next instruction.
      ++MBBI;
      break;
    }
    }
  }

  return Modified;
}

bool AArch64LoadStoreOpt::enableNarrowLdMerge(MachineFunction &Fn) {
  bool ProfitableArch = Subtarget->isCortexA57() || Subtarget->isKryo();
  // FIXME: The benefit from converting narrow loads into a wider load could be
  // microarchitectural as it assumes that a single load with two bitfield
  // extracts is cheaper than two narrow loads. Currently, this conversion is
  // enabled only in cortex-a57 on which performance benefits were verified.
  return ProfitableArch && !Subtarget->requiresStrictAlign();
}

bool AArch64LoadStoreOpt::runOnMachineFunction(MachineFunction &Fn) {
  Subtarget = &static_cast<const AArch64Subtarget &>(Fn.getSubtarget());
  TII = static_cast<const AArch64InstrInfo *>(Subtarget->getInstrInfo());
  TRI = Subtarget->getRegisterInfo();

  // Resize the modified and used register bitfield trackers.  We do this once
  // per function and then clear the bitfield each time we optimize a load or
  // store.
  ModifiedRegs.resize(TRI->getNumRegs());
  UsedRegs.resize(TRI->getNumRegs());

  bool Modified = false;
  bool enableNarrowLdOpt = enableNarrowLdMerge(Fn);
  for (auto &MBB : Fn)
    Modified |= optimizeBlock(MBB, enableNarrowLdOpt);

  return Modified;
}

// FIXME: Do we need/want a pre-alloc pass like ARM has to try to keep
// loads and stores near one another?

// FIXME: When pairing store instructions it's very possible for this pass to
// hoist a store with a KILL marker above another use (without a KILL marker).
// The resulting IR is invalid, but nothing uses the KILL markers after this
// pass, so it's never caused a problem in practice.

/// createAArch64LoadStoreOptimizationPass - returns an instance of the
/// load / store optimization pass.
FunctionPass *llvm::createAArch64LoadStoreOptimizationPass() {
  return new AArch64LoadStoreOpt();
}