llvm.org GIT mirror llvm / bfd9349 lib / DebugInfo / DWARF / DWARFVerifier.cpp
bfd9349

Tree @bfd9349 (Download .tar.gz)

DWARFVerifier.cpp @bfd9349raw · 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
//===- DWARFVerifier.cpp --------------------------------------------------===//
//
// 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
//
//===----------------------------------------------------------------------===//
#include "llvm/DebugInfo/DWARF/DWARFVerifier.h"
#include "llvm/ADT/SmallSet.h"
#include "llvm/DebugInfo/DWARF/DWARFCompileUnit.h"
#include "llvm/DebugInfo/DWARF/DWARFContext.h"
#include "llvm/DebugInfo/DWARF/DWARFDebugLine.h"
#include "llvm/DebugInfo/DWARF/DWARFDie.h"
#include "llvm/DebugInfo/DWARF/DWARFExpression.h"
#include "llvm/DebugInfo/DWARF/DWARFFormValue.h"
#include "llvm/DebugInfo/DWARF/DWARFSection.h"
#include "llvm/Support/DJB.h"
#include "llvm/Support/FormatVariadic.h"
#include "llvm/Support/WithColor.h"
#include "llvm/Support/raw_ostream.h"
#include <map>
#include <set>
#include <vector>

using namespace llvm;
using namespace dwarf;
using namespace object;

DWARFVerifier::DieRangeInfo::address_range_iterator
DWARFVerifier::DieRangeInfo::insert(const DWARFAddressRange &R) {
  auto Begin = Ranges.begin();
  auto End = Ranges.end();
  auto Pos = std::lower_bound(Begin, End, R);

  if (Pos != End) {
    if (Pos->intersects(R))
      return Pos;
    if (Pos != Begin) {
      auto Iter = Pos - 1;
      if (Iter->intersects(R))
        return Iter;
    }
  }

  Ranges.insert(Pos, R);
  return Ranges.end();
}

DWARFVerifier::DieRangeInfo::die_range_info_iterator
DWARFVerifier::DieRangeInfo::insert(const DieRangeInfo &RI) {
  auto End = Children.end();
  auto Iter = Children.begin();
  while (Iter != End) {
    if (Iter->intersects(RI))
      return Iter;
    ++Iter;
  }
  Children.insert(RI);
  return Children.end();
}

bool DWARFVerifier::DieRangeInfo::contains(const DieRangeInfo &RHS) const {
  auto I1 = Ranges.begin(), E1 = Ranges.end();
  auto I2 = RHS.Ranges.begin(), E2 = RHS.Ranges.end();
  if (I2 == E2)
    return true;

  DWARFAddressRange R = *I2;
  while (I1 != E1) {
    bool Covered = I1->LowPC <= R.LowPC;
    if (R.LowPC == R.HighPC || (Covered && R.HighPC <= I1->HighPC)) {
      if (++I2 == E2)
        return true;
      R = *I2;
      continue;
    }
    if (!Covered)
      return false;
    if (R.LowPC < I1->HighPC)
      R.LowPC = I1->HighPC;
    ++I1;
  }
  return false;
}

bool DWARFVerifier::DieRangeInfo::intersects(const DieRangeInfo &RHS) const {
  auto I1 = Ranges.begin(), E1 = Ranges.end();
  auto I2 = RHS.Ranges.begin(), E2 = RHS.Ranges.end();
  while (I1 != E1 && I2 != E2) {
    if (I1->intersects(*I2))
      return true;
    if (I1->LowPC < I2->LowPC)
      ++I1;
    else
      ++I2;
  }
  return false;
}

bool DWARFVerifier::verifyUnitHeader(const DWARFDataExtractor DebugInfoData,
                                     uint32_t *Offset, unsigned UnitIndex,
                                     uint8_t &UnitType, bool &isUnitDWARF64) {
  uint64_t AbbrOffset, Length;
  uint8_t AddrSize = 0;
  uint16_t Version;
  bool Success = true;

  bool ValidLength = false;
  bool ValidVersion = false;
  bool ValidAddrSize = false;
  bool ValidType = true;
  bool ValidAbbrevOffset = true;

  uint32_t OffsetStart = *Offset;
  Length = DebugInfoData.getU32(Offset);
  if (Length == UINT32_MAX) {
    Length = DebugInfoData.getU64(Offset);
    isUnitDWARF64 = true;
  }
  Version = DebugInfoData.getU16(Offset);

  if (Version >= 5) {
    UnitType = DebugInfoData.getU8(Offset);
    AddrSize = DebugInfoData.getU8(Offset);
    AbbrOffset = isUnitDWARF64 ? DebugInfoData.getU64(Offset) : DebugInfoData.getU32(Offset);
    ValidType = dwarf::isUnitType(UnitType);
  } else {
    UnitType = 0;
    AbbrOffset = isUnitDWARF64 ? DebugInfoData.getU64(Offset) : DebugInfoData.getU32(Offset);
    AddrSize = DebugInfoData.getU8(Offset);
  }

  if (!DCtx.getDebugAbbrev()->getAbbreviationDeclarationSet(AbbrOffset))
    ValidAbbrevOffset = false;

  ValidLength = DebugInfoData.isValidOffset(OffsetStart + Length + 3);
  ValidVersion = DWARFContext::isSupportedVersion(Version);
  ValidAddrSize = AddrSize == 4 || AddrSize == 8;
  if (!ValidLength || !ValidVersion || !ValidAddrSize || !ValidAbbrevOffset ||
      !ValidType) {
    Success = false;
    error() << format("Units[%d] - start offset: 0x%08x \n", UnitIndex,
                      OffsetStart);
    if (!ValidLength)
      note() << "The length for this unit is too "
                "large for the .debug_info provided.\n";
    if (!ValidVersion)
      note() << "The 16 bit unit header version is not valid.\n";
    if (!ValidType)
      note() << "The unit type encoding is not valid.\n";
    if (!ValidAbbrevOffset)
      note() << "The offset into the .debug_abbrev section is "
                "not valid.\n";
    if (!ValidAddrSize)
      note() << "The address size is unsupported.\n";
  }
  *Offset = OffsetStart + Length + (isUnitDWARF64 ? 12 : 4);
  return Success;
}

unsigned DWARFVerifier::verifyUnitContents(DWARFUnit &Unit) {
  unsigned NumUnitErrors = 0;
  unsigned NumDies = Unit.getNumDIEs();
  for (unsigned I = 0; I < NumDies; ++I) {
    auto Die = Unit.getDIEAtIndex(I);

    if (Die.getTag() == DW_TAG_null)
      continue;

    for (auto AttrValue : Die.attributes()) {
      NumUnitErrors += verifyDebugInfoAttribute(Die, AttrValue);
      NumUnitErrors += verifyDebugInfoForm(Die, AttrValue);
    }

    NumUnitErrors += verifyDebugInfoCallSite(Die);
  }

  DWARFDie Die = Unit.getUnitDIE(/* ExtractUnitDIEOnly = */ false);
  if (!Die) {
    error() << "Compilation unit without DIE.\n";
    NumUnitErrors++;
    return NumUnitErrors;
  }

  if (!dwarf::isUnitType(Die.getTag())) {
    error() << "Compilation unit root DIE is not a unit DIE: "
            << dwarf::TagString(Die.getTag()) << ".\n";
    NumUnitErrors++;
  }

  uint8_t UnitType = Unit.getUnitType();
  if (!DWARFUnit::isMatchingUnitTypeAndTag(UnitType, Die.getTag())) {
    error() << "Compilation unit type (" << dwarf::UnitTypeString(UnitType)
            << ") and root DIE (" << dwarf::TagString(Die.getTag())
            << ") do not match.\n";
    NumUnitErrors++;
  }

  DieRangeInfo RI;
  NumUnitErrors += verifyDieRanges(Die, RI);

  return NumUnitErrors;
}

unsigned DWARFVerifier::verifyDebugInfoCallSite(const DWARFDie &Die) {
  if (Die.getTag() != DW_TAG_call_site)
    return 0;

  DWARFDie Curr = Die.getParent();
  for (; Curr.isValid() && !Curr.isSubprogramDIE(); Curr = Die.getParent()) {
    if (Curr.getTag() == DW_TAG_inlined_subroutine) {
      error() << "Call site entry nested within inlined subroutine:";
      Curr.dump(OS);
      return 1;
    }
  }

  if (!Curr.isValid()) {
    error() << "Call site entry not nested within a valid subprogram:";
    Die.dump(OS);
    return 1;
  }

  Optional<DWARFFormValue> CallAttr =
      Curr.find({DW_AT_call_all_calls, DW_AT_call_all_source_calls,
                 DW_AT_call_all_tail_calls});
  if (!CallAttr) {
    error() << "Subprogram with call site entry has no DW_AT_call attribute:";
    Curr.dump(OS);
    Die.dump(OS, /*indent*/ 1);
    return 1;
  }

  return 0;
}

unsigned DWARFVerifier::verifyAbbrevSection(const DWARFDebugAbbrev *Abbrev) {
  unsigned NumErrors = 0;
  if (Abbrev) {
    const DWARFAbbreviationDeclarationSet *AbbrDecls =
        Abbrev->getAbbreviationDeclarationSet(0);
    for (auto AbbrDecl : *AbbrDecls) {
      SmallDenseSet<uint16_t> AttributeSet;
      for (auto Attribute : AbbrDecl.attributes()) {
        auto Result = AttributeSet.insert(Attribute.Attr);
        if (!Result.second) {
          error() << "Abbreviation declaration contains multiple "
                  << AttributeString(Attribute.Attr) << " attributes.\n";
          AbbrDecl.dump(OS);
          ++NumErrors;
        }
      }
    }
  }
  return NumErrors;
}

bool DWARFVerifier::handleDebugAbbrev() {
  OS << "Verifying .debug_abbrev...\n";

  const DWARFObject &DObj = DCtx.getDWARFObj();
  unsigned NumErrors = 0;
  if (!DObj.getAbbrevSection().empty())
    NumErrors += verifyAbbrevSection(DCtx.getDebugAbbrev());
  if (!DObj.getAbbrevDWOSection().empty())
    NumErrors += verifyAbbrevSection(DCtx.getDebugAbbrevDWO());

  return NumErrors == 0;
}

unsigned DWARFVerifier::verifyUnitSection(const DWARFSection &S,
                                          DWARFSectionKind SectionKind) {
  const DWARFObject &DObj = DCtx.getDWARFObj();
  DWARFDataExtractor DebugInfoData(DObj, S, DCtx.isLittleEndian(), 0);
  unsigned NumDebugInfoErrors = 0;
  uint32_t OffsetStart = 0, Offset = 0, UnitIdx = 0;
  uint8_t UnitType = 0;
  bool isUnitDWARF64 = false;
  bool isHeaderChainValid = true;
  bool hasDIE = DebugInfoData.isValidOffset(Offset);
  DWARFUnitVector TypeUnitVector;
  DWARFUnitVector CompileUnitVector;
  while (hasDIE) {
    OffsetStart = Offset;
    if (!verifyUnitHeader(DebugInfoData, &Offset, UnitIdx, UnitType,
                          isUnitDWARF64)) {
      isHeaderChainValid = false;
      if (isUnitDWARF64)
        break;
    } else {
      DWARFUnitHeader Header;
      Header.extract(DCtx, DebugInfoData, &OffsetStart, SectionKind);
      DWARFUnit *Unit;
      switch (UnitType) {
      case dwarf::DW_UT_type:
      case dwarf::DW_UT_split_type: {
        Unit = TypeUnitVector.addUnit(llvm::make_unique<DWARFTypeUnit>(
            DCtx, S, Header, DCtx.getDebugAbbrev(), &DObj.getRangeSection(),
            &DObj.getLocSection(), DObj.getStringSection(),
            DObj.getStringOffsetSection(), &DObj.getAppleObjCSection(),
            DObj.getLineSection(), DCtx.isLittleEndian(), false,
            TypeUnitVector));
        break;
      }
      case dwarf::DW_UT_skeleton:
      case dwarf::DW_UT_split_compile:
      case dwarf::DW_UT_compile:
      case dwarf::DW_UT_partial:
      // UnitType = 0 means that we are verifying a compile unit in DWARF v4.
      case 0: {
        Unit = CompileUnitVector.addUnit(llvm::make_unique<DWARFCompileUnit>(
            DCtx, S, Header, DCtx.getDebugAbbrev(), &DObj.getRangeSection(),
            &DObj.getLocSection(), DObj.getStringSection(),
            DObj.getStringOffsetSection(), &DObj.getAppleObjCSection(),
            DObj.getLineSection(), DCtx.isLittleEndian(), false,
            CompileUnitVector));
        break;
      }
      default: { llvm_unreachable("Invalid UnitType."); }
      }
      NumDebugInfoErrors += verifyUnitContents(*Unit);
    }
    hasDIE = DebugInfoData.isValidOffset(Offset);
    ++UnitIdx;
  }
  if (UnitIdx == 0 && !hasDIE) {
    warn() << "Section is empty.\n";
    isHeaderChainValid = true;
  }
  if (!isHeaderChainValid)
    ++NumDebugInfoErrors;
  NumDebugInfoErrors += verifyDebugInfoReferences();
  return NumDebugInfoErrors;
}

bool DWARFVerifier::handleDebugInfo() {
  const DWARFObject &DObj = DCtx.getDWARFObj();
  unsigned NumErrors = 0;

  OS << "Verifying .debug_info Unit Header Chain...\n";
  DObj.forEachInfoSections([&](const DWARFSection &S) {
    NumErrors += verifyUnitSection(S, DW_SECT_INFO);
  });

  OS << "Verifying .debug_types Unit Header Chain...\n";
  DObj.forEachTypesSections([&](const DWARFSection &S) {
    NumErrors += verifyUnitSection(S, DW_SECT_TYPES);
  });
  return NumErrors == 0;
}

unsigned DWARFVerifier::verifyDieRanges(const DWARFDie &Die,
                                        DieRangeInfo &ParentRI) {
  unsigned NumErrors = 0;

  if (!Die.isValid())
    return NumErrors;

  auto RangesOrError = Die.getAddressRanges();
  if (!RangesOrError) {
    // FIXME: Report the error.
    ++NumErrors;
    llvm::consumeError(RangesOrError.takeError());
    return NumErrors;
  }

  DWARFAddressRangesVector Ranges = RangesOrError.get();
  // Build RI for this DIE and check that ranges within this DIE do not
  // overlap.
  DieRangeInfo RI(Die);

  // TODO support object files better
  //
  // Some object file formats (i.e. non-MachO) support COMDAT.  ELF in
  // particular does so by placing each function into a section.  The DWARF data
  // for the function at that point uses a section relative DW_FORM_addrp for
  // the DW_AT_low_pc and a DW_FORM_data4 for the offset as the DW_AT_high_pc.
  // In such a case, when the Die is the CU, the ranges will overlap, and we
  // will flag valid conflicting ranges as invalid.
  //
  // For such targets, we should read the ranges from the CU and partition them
  // by the section id.  The ranges within a particular section should be
  // disjoint, although the ranges across sections may overlap.  We would map
  // the child die to the entity that it references and the section with which
  // it is associated.  The child would then be checked against the range
  // information for the associated section.
  //
  // For now, simply elide the range verification for the CU DIEs if we are
  // processing an object file.

  if (!IsObjectFile || IsMachOObject || Die.getTag() != DW_TAG_compile_unit) {
    for (auto Range : Ranges) {
      if (!Range.valid()) {
        ++NumErrors;
        error() << "Invalid address range " << Range << "\n";
        continue;
      }

      // Verify that ranges don't intersect.
      const auto IntersectingRange = RI.insert(Range);
      if (IntersectingRange != RI.Ranges.end()) {
        ++NumErrors;
        error() << "DIE has overlapping address ranges: " << Range << " and "
                << *IntersectingRange << "\n";
        break;
      }
    }
  }

  // Verify that children don't intersect.
  const auto IntersectingChild = ParentRI.insert(RI);
  if (IntersectingChild != ParentRI.Children.end()) {
    ++NumErrors;
    error() << "DIEs have overlapping address ranges:";
    dump(Die);
    dump(IntersectingChild->Die) << '\n';
  }

  // Verify that ranges are contained within their parent.
  bool ShouldBeContained = !Ranges.empty() && !ParentRI.Ranges.empty() &&
                           !(Die.getTag() == DW_TAG_subprogram &&
                             ParentRI.Die.getTag() == DW_TAG_subprogram);
  if (ShouldBeContained && !ParentRI.contains(RI)) {
    ++NumErrors;
    error() << "DIE address ranges are not contained in its parent's ranges:";
    dump(ParentRI.Die);
    dump(Die, 2) << '\n';
  }

  // Recursively check children.
  for (DWARFDie Child : Die)
    NumErrors += verifyDieRanges(Child, RI);

  return NumErrors;
}

unsigned DWARFVerifier::verifyDebugInfoAttribute(const DWARFDie &Die,
                                                 DWARFAttribute &AttrValue) {
  unsigned NumErrors = 0;
  auto ReportError = [&](const Twine &TitleMsg) {
    ++NumErrors;
    error() << TitleMsg << '\n';
    dump(Die) << '\n';
  };

  const DWARFObject &DObj = DCtx.getDWARFObj();
  const auto Attr = AttrValue.Attr;
  switch (Attr) {
  case DW_AT_ranges:
    // Make sure the offset in the DW_AT_ranges attribute is valid.
    if (auto SectionOffset = AttrValue.Value.getAsSectionOffset()) {
      if (*SectionOffset >= DObj.getRangeSection().Data.size())
        ReportError("DW_AT_ranges offset is beyond .debug_ranges bounds:");
      break;
    }
    ReportError("DIE has invalid DW_AT_ranges encoding:");
    break;
  case DW_AT_stmt_list:
    // Make sure the offset in the DW_AT_stmt_list attribute is valid.
    if (auto SectionOffset = AttrValue.Value.getAsSectionOffset()) {
      if (*SectionOffset >= DObj.getLineSection().Data.size())
        ReportError("DW_AT_stmt_list offset is beyond .debug_line bounds: " +
                    llvm::formatv("{0:x8}", *SectionOffset));
      break;
    }
    ReportError("DIE has invalid DW_AT_stmt_list encoding:");
    break;
  case DW_AT_location: {
    auto VerifyLocationExpr = [&](StringRef D) {
      DWARFUnit *U = Die.getDwarfUnit();
      DataExtractor Data(D, DCtx.isLittleEndian(), 0);
      DWARFExpression Expression(Data, U->getVersion(),
                                 U->getAddressByteSize());
      bool Error = llvm::any_of(Expression, [](DWARFExpression::Operation &Op) {
        return Op.isError();
      });
      if (Error || !Expression.verify(U))
        ReportError("DIE contains invalid DWARF expression:");
    };
    if (Optional<ArrayRef<uint8_t>> Expr = AttrValue.Value.getAsBlock()) {
      // Verify inlined location.
      VerifyLocationExpr(llvm::toStringRef(*Expr));
    } else if (auto LocOffset = AttrValue.Value.getAsSectionOffset()) {
      // Verify location list.
      if (auto DebugLoc = DCtx.getDebugLoc())
        if (auto LocList = DebugLoc->getLocationListAtOffset(*LocOffset))
          for (const auto &Entry : LocList->Entries)
            VerifyLocationExpr({Entry.Loc.data(), Entry.Loc.size()});
    }
    break;
  }
  case DW_AT_specification:
  case DW_AT_abstract_origin: {
    if (auto ReferencedDie = Die.getAttributeValueAsReferencedDie(Attr)) {
      auto DieTag = Die.getTag();
      auto RefTag = ReferencedDie.getTag();
      if (DieTag == RefTag)
        break;
      if (DieTag == DW_TAG_inlined_subroutine && RefTag == DW_TAG_subprogram)
        break;
      if (DieTag == DW_TAG_variable && RefTag == DW_TAG_member)
        break;
      ReportError("DIE with tag " + TagString(DieTag) + " has " +
                  AttributeString(Attr) +
                  " that points to DIE with "
                  "incompatible tag " +
                  TagString(RefTag));
    }
    break;
  }
  case DW_AT_type: {
    DWARFDie TypeDie = Die.getAttributeValueAsReferencedDie(DW_AT_type);
    if (TypeDie && !isType(TypeDie.getTag())) {
      ReportError("DIE has " + AttributeString(Attr) +
                  " with incompatible tag " + TagString(TypeDie.getTag()));
    }
    break;
  }
  default:
    break;
  }
  return NumErrors;
}

unsigned DWARFVerifier::verifyDebugInfoForm(const DWARFDie &Die,
                                            DWARFAttribute &AttrValue) {
  const DWARFObject &DObj = DCtx.getDWARFObj();
  auto DieCU = Die.getDwarfUnit();
  unsigned NumErrors = 0;
  const auto Form = AttrValue.Value.getForm();
  switch (Form) {
  case DW_FORM_ref1:
  case DW_FORM_ref2:
  case DW_FORM_ref4:
  case DW_FORM_ref8:
  case DW_FORM_ref_udata: {
    // Verify all CU relative references are valid CU offsets.
    Optional<uint64_t> RefVal = AttrValue.Value.getAsReference();
    assert(RefVal);
    if (RefVal) {
      auto CUSize = DieCU->getNextUnitOffset() - DieCU->getOffset();
      auto CUOffset = AttrValue.Value.getRawUValue();
      if (CUOffset >= CUSize) {
        ++NumErrors;
        error() << FormEncodingString(Form) << " CU offset "
                << format("0x%08" PRIx64, CUOffset)
                << " is invalid (must be less than CU size of "
                << format("0x%08" PRIx32, CUSize) << "):\n";
        Die.dump(OS, 0, DumpOpts);
        dump(Die) << '\n';
      } else {
        // Valid reference, but we will verify it points to an actual
        // DIE later.
        ReferenceToDIEOffsets[*RefVal].insert(Die.getOffset());
      }
    }
    break;
  }
  case DW_FORM_ref_addr: {
    // Verify all absolute DIE references have valid offsets in the
    // .debug_info section.
    Optional<uint64_t> RefVal = AttrValue.Value.getAsReference();
    assert(RefVal);
    if (RefVal) {
      if (*RefVal >= DieCU->getInfoSection().Data.size()) {
        ++NumErrors;
        error() << "DW_FORM_ref_addr offset beyond .debug_info "
                   "bounds:\n";
        dump(Die) << '\n';
      } else {
        // Valid reference, but we will verify it points to an actual
        // DIE later.
        ReferenceToDIEOffsets[*RefVal].insert(Die.getOffset());
      }
    }
    break;
  }
  case DW_FORM_strp: {
    auto SecOffset = AttrValue.Value.getAsSectionOffset();
    assert(SecOffset); // DW_FORM_strp is a section offset.
    if (SecOffset && *SecOffset >= DObj.getStringSection().size()) {
      ++NumErrors;
      error() << "DW_FORM_strp offset beyond .debug_str bounds:\n";
      dump(Die) << '\n';
    }
    break;
  }
  case DW_FORM_strx:
  case DW_FORM_strx1:
  case DW_FORM_strx2:
  case DW_FORM_strx3:
  case DW_FORM_strx4: {
    auto Index = AttrValue.Value.getRawUValue();
    auto DieCU = Die.getDwarfUnit();
    // Check that we have a valid DWARF v5 string offsets table.
    if (!DieCU->getStringOffsetsTableContribution()) {
      ++NumErrors;
      error() << FormEncodingString(Form)
              << " used without a valid string offsets table:\n";
      dump(Die) << '\n';
      break;
    }
    // Check that the index is within the bounds of the section.
    unsigned ItemSize = DieCU->getDwarfStringOffsetsByteSize();
    // Use a 64-bit type to calculate the offset to guard against overflow.
    uint64_t Offset =
        (uint64_t)DieCU->getStringOffsetsBase() + Index * ItemSize;
    if (DObj.getStringOffsetSection().Data.size() < Offset + ItemSize) {
      ++NumErrors;
      error() << FormEncodingString(Form) << " uses index "
              << format("%" PRIu64, Index) << ", which is too large:\n";
      dump(Die) << '\n';
      break;
    }
    // Check that the string offset is valid.
    uint64_t StringOffset = *DieCU->getStringOffsetSectionItem(Index);
    if (StringOffset >= DObj.getStringSection().size()) {
      ++NumErrors;
      error() << FormEncodingString(Form) << " uses index "
              << format("%" PRIu64, Index)
              << ", but the referenced string"
                 " offset is beyond .debug_str bounds:\n";
      dump(Die) << '\n';
    }
    break;
  }
  default:
    break;
  }
  return NumErrors;
}

unsigned DWARFVerifier::verifyDebugInfoReferences() {
  // Take all references and make sure they point to an actual DIE by
  // getting the DIE by offset and emitting an error
  OS << "Verifying .debug_info references...\n";
  unsigned NumErrors = 0;
  for (const std::pair<uint64_t, std::set<uint32_t>> &Pair :
       ReferenceToDIEOffsets) {
    if (DCtx.getDIEForOffset(Pair.first))
      continue;
    ++NumErrors;
    error() << "invalid DIE reference " << format("0x%08" PRIx64, Pair.first)
            << ". Offset is in between DIEs:\n";
    for (auto Offset : Pair.second)
      dump(DCtx.getDIEForOffset(Offset)) << '\n';
    OS << "\n";
  }
  return NumErrors;
}

void DWARFVerifier::verifyDebugLineStmtOffsets() {
  std::map<uint64_t, DWARFDie> StmtListToDie;
  for (const auto &CU : DCtx.compile_units()) {
    auto Die = CU->getUnitDIE();
    // Get the attribute value as a section offset. No need to produce an
    // error here if the encoding isn't correct because we validate this in
    // the .debug_info verifier.
    auto StmtSectionOffset = toSectionOffset(Die.find(DW_AT_stmt_list));
    if (!StmtSectionOffset)
      continue;
    const uint32_t LineTableOffset = *StmtSectionOffset;
    auto LineTable = DCtx.getLineTableForUnit(CU.get());
    if (LineTableOffset < DCtx.getDWARFObj().getLineSection().Data.size()) {
      if (!LineTable) {
        ++NumDebugLineErrors;
        error() << ".debug_line[" << format("0x%08" PRIx32, LineTableOffset)
                << "] was not able to be parsed for CU:\n";
        dump(Die) << '\n';
        continue;
      }
    } else {
      // Make sure we don't get a valid line table back if the offset is wrong.
      assert(LineTable == nullptr);
      // Skip this line table as it isn't valid. No need to create an error
      // here because we validate this in the .debug_info verifier.
      continue;
    }
    auto Iter = StmtListToDie.find(LineTableOffset);
    if (Iter != StmtListToDie.end()) {
      ++NumDebugLineErrors;
      error() << "two compile unit DIEs, "
              << format("0x%08" PRIx32, Iter->second.getOffset()) << " and "
              << format("0x%08" PRIx32, Die.getOffset())
              << ", have the same DW_AT_stmt_list section offset:\n";
      dump(Iter->second);
      dump(Die) << '\n';
      // Already verified this line table before, no need to do it again.
      continue;
    }
    StmtListToDie[LineTableOffset] = Die;
  }
}

void DWARFVerifier::verifyDebugLineRows() {
  for (const auto &CU : DCtx.compile_units()) {
    auto Die = CU->getUnitDIE();
    auto LineTable = DCtx.getLineTableForUnit(CU.get());
    // If there is no line table we will have created an error in the
    // .debug_info verifier or in verifyDebugLineStmtOffsets().
    if (!LineTable)
      continue;

    // Verify prologue.
    uint32_t MaxDirIndex = LineTable->Prologue.IncludeDirectories.size();
    uint32_t FileIndex = 1;
    StringMap<uint16_t> FullPathMap;
    for (const auto &FileName : LineTable->Prologue.FileNames) {
      // Verify directory index.
      if (FileName.DirIdx > MaxDirIndex) {
        ++NumDebugLineErrors;
        error() << ".debug_line["
                << format("0x%08" PRIx64,
                          *toSectionOffset(Die.find(DW_AT_stmt_list)))
                << "].prologue.file_names[" << FileIndex
                << "].dir_idx contains an invalid index: " << FileName.DirIdx
                << "\n";
      }

      // Check file paths for duplicates.
      std::string FullPath;
      const bool HasFullPath = LineTable->getFileNameByIndex(
          FileIndex, CU->getCompilationDir(),
          DILineInfoSpecifier::FileLineInfoKind::AbsoluteFilePath, FullPath);
      assert(HasFullPath && "Invalid index?");
      (void)HasFullPath;
      auto It = FullPathMap.find(FullPath);
      if (It == FullPathMap.end())
        FullPathMap[FullPath] = FileIndex;
      else if (It->second != FileIndex) {
        warn() << ".debug_line["
               << format("0x%08" PRIx64,
                         *toSectionOffset(Die.find(DW_AT_stmt_list)))
               << "].prologue.file_names[" << FileIndex
               << "] is a duplicate of file_names[" << It->second << "]\n";
      }

      FileIndex++;
    }

    // Verify rows.
    uint64_t PrevAddress = 0;
    uint32_t RowIndex = 0;
    for (const auto &Row : LineTable->Rows) {
      // Verify row address.
      if (Row.Address.Address < PrevAddress) {
        ++NumDebugLineErrors;
        error() << ".debug_line["
                << format("0x%08" PRIx64,
                          *toSectionOffset(Die.find(DW_AT_stmt_list)))
                << "] row[" << RowIndex
                << "] decreases in address from previous row:\n";

        DWARFDebugLine::Row::dumpTableHeader(OS);
        if (RowIndex > 0)
          LineTable->Rows[RowIndex - 1].dump(OS);
        Row.dump(OS);
        OS << '\n';
      }

      // Verify file index.
      if (!LineTable->hasFileAtIndex(Row.File)) {
        ++NumDebugLineErrors;
        bool isDWARF5 = LineTable->Prologue.getVersion() >= 5;
        error() << ".debug_line["
                << format("0x%08" PRIx64,
                          *toSectionOffset(Die.find(DW_AT_stmt_list)))
                << "][" << RowIndex << "] has invalid file index " << Row.File
                << " (valid values are [" << (isDWARF5 ? "0," : "1,")
                << LineTable->Prologue.FileNames.size()
                << (isDWARF5 ? ")" : "]") << "):\n";
        DWARFDebugLine::Row::dumpTableHeader(OS);
        Row.dump(OS);
        OS << '\n';
      }
      if (Row.EndSequence)
        PrevAddress = 0;
      else
        PrevAddress = Row.Address.Address;
      ++RowIndex;
    }
  }
}

DWARFVerifier::DWARFVerifier(raw_ostream &S, DWARFContext &D,
                             DIDumpOptions DumpOpts)
    : OS(S), DCtx(D), DumpOpts(std::move(DumpOpts)), IsObjectFile(false),
      IsMachOObject(false) {
  if (const auto *F = DCtx.getDWARFObj().getFile()) {
    IsObjectFile = F->isRelocatableObject();
    IsMachOObject = F->isMachO();
  }
}

bool DWARFVerifier::handleDebugLine() {
  NumDebugLineErrors = 0;
  OS << "Verifying .debug_line...\n";
  verifyDebugLineStmtOffsets();
  verifyDebugLineRows();
  return NumDebugLineErrors == 0;
}

unsigned DWARFVerifier::verifyAppleAccelTable(const DWARFSection *AccelSection,
                                              DataExtractor *StrData,
                                              const char *SectionName) {
  unsigned NumErrors = 0;
  DWARFDataExtractor AccelSectionData(DCtx.getDWARFObj(), *AccelSection,
                                      DCtx.isLittleEndian(), 0);
  AppleAcceleratorTable AccelTable(AccelSectionData, *StrData);

  OS << "Verifying " << SectionName << "...\n";

  // Verify that the fixed part of the header is not too short.
  if (!AccelSectionData.isValidOffset(AccelTable.getSizeHdr())) {
    error() << "Section is too small to fit a section header.\n";
    return 1;
  }

  // Verify that the section is not too short.
  if (Error E = AccelTable.extract()) {
    error() << toString(std::move(E)) << '\n';
    return 1;
  }

  // Verify that all buckets have a valid hash index or are empty.
  uint32_t NumBuckets = AccelTable.getNumBuckets();
  uint32_t NumHashes = AccelTable.getNumHashes();

  uint32_t BucketsOffset =
      AccelTable.getSizeHdr() + AccelTable.getHeaderDataLength();
  uint32_t HashesBase = BucketsOffset + NumBuckets * 4;
  uint32_t OffsetsBase = HashesBase + NumHashes * 4;
  for (uint32_t BucketIdx = 0; BucketIdx < NumBuckets; ++BucketIdx) {
    uint32_t HashIdx = AccelSectionData.getU32(&BucketsOffset);
    if (HashIdx >= NumHashes && HashIdx != UINT32_MAX) {
      error() << format("Bucket[%d] has invalid hash index: %u.\n", BucketIdx,
                        HashIdx);
      ++NumErrors;
    }
  }
  uint32_t NumAtoms = AccelTable.getAtomsDesc().size();
  if (NumAtoms == 0) {
    error() << "No atoms: failed to read HashData.\n";
    return 1;
  }
  if (!AccelTable.validateForms()) {
    error() << "Unsupported form: failed to read HashData.\n";
    return 1;
  }

  for (uint32_t HashIdx = 0; HashIdx < NumHashes; ++HashIdx) {
    uint32_t HashOffset = HashesBase + 4 * HashIdx;
    uint32_t DataOffset = OffsetsBase + 4 * HashIdx;
    uint32_t Hash = AccelSectionData.getU32(&HashOffset);
    uint32_t HashDataOffset = AccelSectionData.getU32(&DataOffset);
    if (!AccelSectionData.isValidOffsetForDataOfSize(HashDataOffset,
                                                     sizeof(uint64_t))) {
      error() << format("Hash[%d] has invalid HashData offset: 0x%08x.\n",
                        HashIdx, HashDataOffset);
      ++NumErrors;
    }

    uint32_t StrpOffset;
    uint32_t StringOffset;
    uint32_t StringCount = 0;
    unsigned Offset;
    unsigned Tag;
    while ((StrpOffset = AccelSectionData.getU32(&HashDataOffset)) != 0) {
      const uint32_t NumHashDataObjects =
          AccelSectionData.getU32(&HashDataOffset);
      for (uint32_t HashDataIdx = 0; HashDataIdx < NumHashDataObjects;
           ++HashDataIdx) {
        std::tie(Offset, Tag) = AccelTable.readAtoms(HashDataOffset);
        auto Die = DCtx.getDIEForOffset(Offset);
        if (!Die) {
          const uint32_t BucketIdx =
              NumBuckets ? (Hash % NumBuckets) : UINT32_MAX;
          StringOffset = StrpOffset;
          const char *Name = StrData->getCStr(&StringOffset);
          if (!Name)
            Name = "<NULL>";

          error() << format(
              "%s Bucket[%d] Hash[%d] = 0x%08x "
              "Str[%u] = 0x%08x "
              "DIE[%d] = 0x%08x is not a valid DIE offset for \"%s\".\n",
              SectionName, BucketIdx, HashIdx, Hash, StringCount, StrpOffset,
              HashDataIdx, Offset, Name);

          ++NumErrors;
          continue;
        }
        if ((Tag != dwarf::DW_TAG_null) && (Die.getTag() != Tag)) {
          error() << "Tag " << dwarf::TagString(Tag)
                  << " in accelerator table does not match Tag "
                  << dwarf::TagString(Die.getTag()) << " of DIE[" << HashDataIdx
                  << "].\n";
          ++NumErrors;
        }
      }
      ++StringCount;
    }
  }
  return NumErrors;
}

unsigned
DWARFVerifier::verifyDebugNamesCULists(const DWARFDebugNames &AccelTable) {
  // A map from CU offset to the (first) Name Index offset which claims to index
  // this CU.
  DenseMap<uint32_t, uint32_t> CUMap;
  const uint32_t NotIndexed = std::numeric_limits<uint32_t>::max();

  CUMap.reserve(DCtx.getNumCompileUnits());
  for (const auto &CU : DCtx.compile_units())
    CUMap[CU->getOffset()] = NotIndexed;

  unsigned NumErrors = 0;
  for (const DWARFDebugNames::NameIndex &NI : AccelTable) {
    if (NI.getCUCount() == 0) {
      error() << formatv("Name Index @ {0:x} does not index any CU\n",
                         NI.getUnitOffset());
      ++NumErrors;
      continue;
    }
    for (uint32_t CU = 0, End = NI.getCUCount(); CU < End; ++CU) {
      uint32_t Offset = NI.getCUOffset(CU);
      auto Iter = CUMap.find(Offset);

      if (Iter == CUMap.end()) {
        error() << formatv(
            "Name Index @ {0:x} references a non-existing CU @ {1:x}\n",
            NI.getUnitOffset(), Offset);
        ++NumErrors;
        continue;
      }

      if (Iter->second != NotIndexed) {
        error() << formatv("Name Index @ {0:x} references a CU @ {1:x}, but "
                           "this CU is already indexed by Name Index @ {2:x}\n",
                           NI.getUnitOffset(), Offset, Iter->second);
        continue;
      }
      Iter->second = NI.getUnitOffset();
    }
  }

  for (const auto &KV : CUMap) {
    if (KV.second == NotIndexed)
      warn() << formatv("CU @ {0:x} not covered by any Name Index\n", KV.first);
  }

  return NumErrors;
}

unsigned
DWARFVerifier::verifyNameIndexBuckets(const DWARFDebugNames::NameIndex &NI,
                                      const DataExtractor &StrData) {
  struct BucketInfo {
    uint32_t Bucket;
    uint32_t Index;

    constexpr BucketInfo(uint32_t Bucket, uint32_t Index)
        : Bucket(Bucket), Index(Index) {}
    bool operator<(const BucketInfo &RHS) const { return Index < RHS.Index; };
  };

  uint32_t NumErrors = 0;
  if (NI.getBucketCount() == 0) {
    warn() << formatv("Name Index @ {0:x} does not contain a hash table.\n",
                      NI.getUnitOffset());
    return NumErrors;
  }

  // Build up a list of (Bucket, Index) pairs. We use this later to verify that
  // each Name is reachable from the appropriate bucket.
  std::vector<BucketInfo> BucketStarts;
  BucketStarts.reserve(NI.getBucketCount() + 1);
  for (uint32_t Bucket = 0, End = NI.getBucketCount(); Bucket < End; ++Bucket) {
    uint32_t Index = NI.getBucketArrayEntry(Bucket);
    if (Index > NI.getNameCount()) {
      error() << formatv("Bucket {0} of Name Index @ {1:x} contains invalid "
                         "value {2}. Valid range is [0, {3}].\n",
                         Bucket, NI.getUnitOffset(), Index, NI.getNameCount());
      ++NumErrors;
      continue;
    }
    if (Index > 0)
      BucketStarts.emplace_back(Bucket, Index);
  }

  // If there were any buckets with invalid values, skip further checks as they
  // will likely produce many errors which will only confuse the actual root
  // problem.
  if (NumErrors > 0)
    return NumErrors;

  // Sort the list in the order of increasing "Index" entries.
  array_pod_sort(BucketStarts.begin(), BucketStarts.end());

  // Insert a sentinel entry at the end, so we can check that the end of the
  // table is covered in the loop below.
  BucketStarts.emplace_back(NI.getBucketCount(), NI.getNameCount() + 1);

  // Loop invariant: NextUncovered is the (1-based) index of the first Name
  // which is not reachable by any of the buckets we processed so far (and
  // hasn't been reported as uncovered).
  uint32_t NextUncovered = 1;
  for (const BucketInfo &B : BucketStarts) {
    // Under normal circumstances B.Index be equal to NextUncovered, but it can
    // be less if a bucket points to names which are already known to be in some
    // bucket we processed earlier. In that case, we won't trigger this error,
    // but report the mismatched hash value error instead. (We know the hash
    // will not match because we have already verified that the name's hash
    // puts it into the previous bucket.)
    if (B.Index > NextUncovered) {
      error() << formatv("Name Index @ {0:x}: Name table entries [{1}, {2}] "
                         "are not covered by the hash table.\n",
                         NI.getUnitOffset(), NextUncovered, B.Index - 1);
      ++NumErrors;
    }
    uint32_t Idx = B.Index;

    // The rest of the checks apply only to non-sentinel entries.
    if (B.Bucket == NI.getBucketCount())
      break;

    // This triggers if a non-empty bucket points to a name with a mismatched
    // hash. Clients are likely to interpret this as an empty bucket, because a
    // mismatched hash signals the end of a bucket, but if this is indeed an
    // empty bucket, the producer should have signalled this by marking the
    // bucket as empty.
    uint32_t FirstHash = NI.getHashArrayEntry(Idx);
    if (FirstHash % NI.getBucketCount() != B.Bucket) {
      error() << formatv(
          "Name Index @ {0:x}: Bucket {1} is not empty but points to a "
          "mismatched hash value {2:x} (belonging to bucket {3}).\n",
          NI.getUnitOffset(), B.Bucket, FirstHash,
          FirstHash % NI.getBucketCount());
      ++NumErrors;
    }

    // This find the end of this bucket and also verifies that all the hashes in
    // this bucket are correct by comparing the stored hashes to the ones we
    // compute ourselves.
    while (Idx <= NI.getNameCount()) {
      uint32_t Hash = NI.getHashArrayEntry(Idx);
      if (Hash % NI.getBucketCount() != B.Bucket)
        break;

      const char *Str = NI.getNameTableEntry(Idx).getString();
      if (caseFoldingDjbHash(Str) != Hash) {
        error() << formatv("Name Index @ {0:x}: String ({1}) at index {2} "
                           "hashes to {3:x}, but "
                           "the Name Index hash is {4:x}\n",
                           NI.getUnitOffset(), Str, Idx,
                           caseFoldingDjbHash(Str), Hash);
        ++NumErrors;
      }

      ++Idx;
    }
    NextUncovered = std::max(NextUncovered, Idx);
  }
  return NumErrors;
}

unsigned DWARFVerifier::verifyNameIndexAttribute(
    const DWARFDebugNames::NameIndex &NI, const DWARFDebugNames::Abbrev &Abbr,
    DWARFDebugNames::AttributeEncoding AttrEnc) {
  StringRef FormName = dwarf::FormEncodingString(AttrEnc.Form);
  if (FormName.empty()) {
    error() << formatv("NameIndex @ {0:x}: Abbreviation {1:x}: {2} uses an "
                       "unknown form: {3}.\n",
                       NI.getUnitOffset(), Abbr.Code, AttrEnc.Index,
                       AttrEnc.Form);
    return 1;
  }

  if (AttrEnc.Index == DW_IDX_type_hash) {
    if (AttrEnc.Form != dwarf::DW_FORM_data8) {
      error() << formatv(
          "NameIndex @ {0:x}: Abbreviation {1:x}: DW_IDX_type_hash "
          "uses an unexpected form {2} (should be {3}).\n",
          NI.getUnitOffset(), Abbr.Code, AttrEnc.Form, dwarf::DW_FORM_data8);
      return 1;
    }
  }

  // A list of known index attributes and their expected form classes.
  // DW_IDX_type_hash is handled specially in the check above, as it has a
  // specific form (not just a form class) we should expect.
  struct FormClassTable {
    dwarf::Index Index;
    DWARFFormValue::FormClass Class;
    StringLiteral ClassName;
  };
  static constexpr FormClassTable Table[] = {
      {dwarf::DW_IDX_compile_unit, DWARFFormValue::FC_Constant, {"constant"}},
      {dwarf::DW_IDX_type_unit, DWARFFormValue::FC_Constant, {"constant"}},
      {dwarf::DW_IDX_die_offset, DWARFFormValue::FC_Reference, {"reference"}},
      {dwarf::DW_IDX_parent, DWARFFormValue::FC_Constant, {"constant"}},
  };

  ArrayRef<FormClassTable> TableRef(Table);
  auto Iter = find_if(TableRef, [AttrEnc](const FormClassTable &T) {
    return T.Index == AttrEnc.Index;
  });
  if (Iter == TableRef.end()) {
    warn() << formatv("NameIndex @ {0:x}: Abbreviation {1:x} contains an "
                      "unknown index attribute: {2}.\n",
                      NI.getUnitOffset(), Abbr.Code, AttrEnc.Index);
    return 0;
  }

  if (!DWARFFormValue(AttrEnc.Form).isFormClass(Iter->Class)) {
    error() << formatv("NameIndex @ {0:x}: Abbreviation {1:x}: {2} uses an "
                       "unexpected form {3} (expected form class {4}).\n",
                       NI.getUnitOffset(), Abbr.Code, AttrEnc.Index,
                       AttrEnc.Form, Iter->ClassName);
    return 1;
  }
  return 0;
}

unsigned
DWARFVerifier::verifyNameIndexAbbrevs(const DWARFDebugNames::NameIndex &NI) {
  if (NI.getLocalTUCount() + NI.getForeignTUCount() > 0) {
    warn() << formatv("Name Index @ {0:x}: Verifying indexes of type units is "
                      "not currently supported.\n",
                      NI.getUnitOffset());
    return 0;
  }

  unsigned NumErrors = 0;
  for (const auto &Abbrev : NI.getAbbrevs()) {
    StringRef TagName = dwarf::TagString(Abbrev.Tag);
    if (TagName.empty()) {
      warn() << formatv("NameIndex @ {0:x}: Abbreviation {1:x} references an "
                        "unknown tag: {2}.\n",
                        NI.getUnitOffset(), Abbrev.Code, Abbrev.Tag);
    }
    SmallSet<unsigned, 5> Attributes;
    for (const auto &AttrEnc : Abbrev.Attributes) {
      if (!Attributes.insert(AttrEnc.Index).second) {
        error() << formatv("NameIndex @ {0:x}: Abbreviation {1:x} contains "
                           "multiple {2} attributes.\n",
                           NI.getUnitOffset(), Abbrev.Code, AttrEnc.Index);
        ++NumErrors;
        continue;
      }
      NumErrors += verifyNameIndexAttribute(NI, Abbrev, AttrEnc);
    }

    if (NI.getCUCount() > 1 && !Attributes.count(dwarf::DW_IDX_compile_unit)) {
      error() << formatv("NameIndex @ {0:x}: Indexing multiple compile units "
                         "and abbreviation {1:x} has no {2} attribute.\n",
                         NI.getUnitOffset(), Abbrev.Code,
                         dwarf::DW_IDX_compile_unit);
      ++NumErrors;
    }
    if (!Attributes.count(dwarf::DW_IDX_die_offset)) {
      error() << formatv(
          "NameIndex @ {0:x}: Abbreviation {1:x} has no {2} attribute.\n",
          NI.getUnitOffset(), Abbrev.Code, dwarf::DW_IDX_die_offset);
      ++NumErrors;
    }
  }
  return NumErrors;
}

static SmallVector<StringRef, 2> getNames(const DWARFDie &DIE,
                                          bool IncludeLinkageName = true) {
  SmallVector<StringRef, 2> Result;
  if (const char *Str = DIE.getName(DINameKind::ShortName))
    Result.emplace_back(Str);
  else if (DIE.getTag() == dwarf::DW_TAG_namespace)
    Result.emplace_back("(anonymous namespace)");

  if (IncludeLinkageName) {
    if (const char *Str = DIE.getName(DINameKind::LinkageName)) {
      if (Result.empty() || Result[0] != Str)
        Result.emplace_back(Str);
    }
  }

  return Result;
}

unsigned DWARFVerifier::verifyNameIndexEntries(
    const DWARFDebugNames::NameIndex &NI,
    const DWARFDebugNames::NameTableEntry &NTE) {
  // Verifying type unit indexes not supported.
  if (NI.getLocalTUCount() + NI.getForeignTUCount() > 0)
    return 0;

  const char *CStr = NTE.getString();
  if (!CStr) {
    error() << formatv(
        "Name Index @ {0:x}: Unable to get string associated with name {1}.\n",
        NI.getUnitOffset(), NTE.getIndex());
    return 1;
  }
  StringRef Str(CStr);

  unsigned NumErrors = 0;
  unsigned NumEntries = 0;
  uint32_t EntryID = NTE.getEntryOffset();
  uint32_t NextEntryID = EntryID;
  Expected<DWARFDebugNames::Entry> EntryOr = NI.getEntry(&NextEntryID);
  for (; EntryOr; ++NumEntries, EntryID = NextEntryID,
                                EntryOr = NI.getEntry(&NextEntryID)) {
    uint32_t CUIndex = *EntryOr->getCUIndex();
    if (CUIndex > NI.getCUCount()) {
      error() << formatv("Name Index @ {0:x}: Entry @ {1:x} contains an "
                         "invalid CU index ({2}).\n",
                         NI.getUnitOffset(), EntryID, CUIndex);
      ++NumErrors;
      continue;
    }
    uint32_t CUOffset = NI.getCUOffset(CUIndex);
    uint64_t DIEOffset = CUOffset + *EntryOr->getDIEUnitOffset();
    DWARFDie DIE = DCtx.getDIEForOffset(DIEOffset);
    if (!DIE) {
      error() << formatv("Name Index @ {0:x}: Entry @ {1:x} references a "
                         "non-existing DIE @ {2:x}.\n",
                         NI.getUnitOffset(), EntryID, DIEOffset);
      ++NumErrors;
      continue;
    }
    if (DIE.getDwarfUnit()->getOffset() != CUOffset) {
      error() << formatv("Name Index @ {0:x}: Entry @ {1:x}: mismatched CU of "
                         "DIE @ {2:x}: index - {3:x}; debug_info - {4:x}.\n",
                         NI.getUnitOffset(), EntryID, DIEOffset, CUOffset,
                         DIE.getDwarfUnit()->getOffset());
      ++NumErrors;
    }
    if (DIE.getTag() != EntryOr->tag()) {
      error() << formatv("Name Index @ {0:x}: Entry @ {1:x}: mismatched Tag of "
                         "DIE @ {2:x}: index - {3}; debug_info - {4}.\n",
                         NI.getUnitOffset(), EntryID, DIEOffset, EntryOr->tag(),
                         DIE.getTag());
      ++NumErrors;
    }

    auto EntryNames = getNames(DIE);
    if (!is_contained(EntryNames, Str)) {
      error() << formatv("Name Index @ {0:x}: Entry @ {1:x}: mismatched Name "
                         "of DIE @ {2:x}: index - {3}; debug_info - {4}.\n",
                         NI.getUnitOffset(), EntryID, DIEOffset, Str,
                         make_range(EntryNames.begin(), EntryNames.end()));
      ++NumErrors;
    }
  }
  handleAllErrors(EntryOr.takeError(),
                  [&](const DWARFDebugNames::SentinelError &) {
                    if (NumEntries > 0)
                      return;
                    error() << formatv("Name Index @ {0:x}: Name {1} ({2}) is "
                                       "not associated with any entries.\n",
                                       NI.getUnitOffset(), NTE.getIndex(), Str);
                    ++NumErrors;
                  },
                  [&](const ErrorInfoBase &Info) {
                    error()
                        << formatv("Name Index @ {0:x}: Name {1} ({2}): {3}\n",
                                   NI.getUnitOffset(), NTE.getIndex(), Str,
                                   Info.message());
                    ++NumErrors;
                  });
  return NumErrors;
}

static bool isVariableIndexable(const DWARFDie &Die, DWARFContext &DCtx) {
  Optional<DWARFFormValue> Location = Die.findRecursively(DW_AT_location);
  if (!Location)
    return false;

  auto ContainsInterestingOperators = [&](StringRef D) {
    DWARFUnit *U = Die.getDwarfUnit();
    DataExtractor Data(D, DCtx.isLittleEndian(), U->getAddressByteSize());
    DWARFExpression Expression(Data, U->getVersion(), U->getAddressByteSize());
    return any_of(Expression, [](DWARFExpression::Operation &Op) {
      return !Op.isError() && (Op.getCode() == DW_OP_addr ||
                               Op.getCode() == DW_OP_form_tls_address ||
                               Op.getCode() == DW_OP_GNU_push_tls_address);
    });
  };

  if (Optional<ArrayRef<uint8_t>> Expr = Location->getAsBlock()) {
    // Inlined location.
    if (ContainsInterestingOperators(toStringRef(*Expr)))
      return true;
  } else if (Optional<uint64_t> Offset = Location->getAsSectionOffset()) {
    // Location list.
    if (const DWARFDebugLoc *DebugLoc = DCtx.getDebugLoc()) {
      if (const DWARFDebugLoc::LocationList *LocList =
              DebugLoc->getLocationListAtOffset(*Offset)) {
        if (any_of(LocList->Entries, [&](const DWARFDebugLoc::Entry &E) {
              return ContainsInterestingOperators({E.Loc.data(), E.Loc.size()});
            }))
          return true;
      }
    }
  }
  return false;
}

unsigned DWARFVerifier::verifyNameIndexCompleteness(
    const DWARFDie &Die, const DWARFDebugNames::NameIndex &NI) {

  // First check, if the Die should be indexed. The code follows the DWARF v5
  // wording as closely as possible.

  // "All non-defining declarations (that is, debugging information entries
  // with a DW_AT_declaration attribute) are excluded."
  if (Die.find(DW_AT_declaration))
    return 0;

  // "DW_TAG_namespace debugging information entries without a DW_AT_name
  // attribute are included with the name “(anonymous namespace)”.
  // All other debugging information entries without a DW_AT_name attribute
  // are excluded."
  // "If a subprogram or inlined subroutine is included, and has a
  // DW_AT_linkage_name attribute, there will be an additional index entry for
  // the linkage name."
  auto IncludeLinkageName = Die.getTag() == DW_TAG_subprogram ||
                            Die.getTag() == DW_TAG_inlined_subroutine;
  auto EntryNames = getNames(Die, IncludeLinkageName);
  if (EntryNames.empty())
    return 0;

  // We deviate from the specification here, which says:
  // "The name index must contain an entry for each debugging information entry
  // that defines a named subprogram, label, variable, type, or namespace,
  // subject to ..."
  // Instead whitelisting all TAGs representing a "type" or a "subprogram", to
  // make sure we catch any missing items, we instead blacklist all TAGs that we
  // know shouldn't be indexed.
  switch (Die.getTag()) {
  // Compile units and modules have names but shouldn't be indexed.
  case DW_TAG_compile_unit:
  case DW_TAG_module:
    return 0;

  // Function and template parameters are not globally visible, so we shouldn't
  // index them.
  case DW_TAG_formal_parameter:
  case DW_TAG_template_value_parameter:
  case DW_TAG_template_type_parameter:
  case DW_TAG_GNU_template_parameter_pack:
  case DW_TAG_GNU_template_template_param:
    return 0;

  // Object members aren't globally visible.
  case DW_TAG_member:
    return 0;

  // According to a strict reading of the specification, enumerators should not
  // be indexed (and LLVM currently does not do that). However, this causes
  // problems for the debuggers, so we may need to reconsider this.
  case DW_TAG_enumerator:
    return 0;

  // Imported declarations should not be indexed according to the specification
  // and LLVM currently does not do that.
  case DW_TAG_imported_declaration:
    return 0;

  // "DW_TAG_subprogram, DW_TAG_inlined_subroutine, and DW_TAG_label debugging
  // information entries without an address attribute (DW_AT_low_pc,
  // DW_AT_high_pc, DW_AT_ranges, or DW_AT_entry_pc) are excluded."
  case DW_TAG_subprogram:
  case DW_TAG_inlined_subroutine:
  case DW_TAG_label:
    if (Die.findRecursively(
            {DW_AT_low_pc, DW_AT_high_pc, DW_AT_ranges, DW_AT_entry_pc}))
      break;
    return 0;

  // "DW_TAG_variable debugging information entries with a DW_AT_location
  // attribute that includes a DW_OP_addr or DW_OP_form_tls_address operator are
  // included; otherwise, they are excluded."
  //
  // LLVM extension: We also add DW_OP_GNU_push_tls_address to this list.
  case DW_TAG_variable:
    if (isVariableIndexable(Die, DCtx))
      break;
    return 0;

  default:
    break;
  }

  // Now we know that our Die should be present in the Index. Let's check if
  // that's the case.
  unsigned NumErrors = 0;
  uint64_t DieUnitOffset = Die.getOffset() - Die.getDwarfUnit()->getOffset();
  for (StringRef Name : EntryNames) {
    if (none_of(NI.equal_range(Name), [&](const DWARFDebugNames::Entry &E) {
          return E.getDIEUnitOffset() == DieUnitOffset;
        })) {
      error() << formatv("Name Index @ {0:x}: Entry for DIE @ {1:x} ({2}) with "
                         "name {3} missing.\n",
                         NI.getUnitOffset(), Die.getOffset(), Die.getTag(),
                         Name);
      ++NumErrors;
    }
  }
  return NumErrors;
}

unsigned DWARFVerifier::verifyDebugNames(const DWARFSection &AccelSection,
                                         const DataExtractor &StrData) {
  unsigned NumErrors = 0;
  DWARFDataExtractor AccelSectionData(DCtx.getDWARFObj(), AccelSection,
                                      DCtx.isLittleEndian(), 0);
  DWARFDebugNames AccelTable(AccelSectionData, StrData);

  OS << "Verifying .debug_names...\n";

  // This verifies that we can read individual name indices and their
  // abbreviation tables.
  if (Error E = AccelTable.extract()) {
    error() << toString(std::move(E)) << '\n';
    return 1;
  }

  NumErrors += verifyDebugNamesCULists(AccelTable);
  for (const auto &NI : AccelTable)
    NumErrors += verifyNameIndexBuckets(NI, StrData);
  for (const auto &NI : AccelTable)
    NumErrors += verifyNameIndexAbbrevs(NI);

  // Don't attempt Entry validation if any of the previous checks found errors
  if (NumErrors > 0)
    return NumErrors;
  for (const auto &NI : AccelTable)
    for (DWARFDebugNames::NameTableEntry NTE : NI)
      NumErrors += verifyNameIndexEntries(NI, NTE);

  if (NumErrors > 0)
    return NumErrors;

  for (const std::unique_ptr<DWARFUnit> &U : DCtx.compile_units()) {
    if (const DWARFDebugNames::NameIndex *NI =
            AccelTable.getCUNameIndex(U->getOffset())) {
      auto *CU = cast<DWARFCompileUnit>(U.get());
      for (const DWARFDebugInfoEntry &Die : CU->dies())
        NumErrors += verifyNameIndexCompleteness(DWARFDie(CU, &Die), *NI);
    }
  }
  return NumErrors;
}

bool DWARFVerifier::handleAccelTables() {
  const DWARFObject &D = DCtx.getDWARFObj();
  DataExtractor StrData(D.getStringSection(), DCtx.isLittleEndian(), 0);
  unsigned NumErrors = 0;
  if (!D.getAppleNamesSection().Data.empty())
    NumErrors += verifyAppleAccelTable(&D.getAppleNamesSection(), &StrData,
                                       ".apple_names");
  if (!D.getAppleTypesSection().Data.empty())
    NumErrors += verifyAppleAccelTable(&D.getAppleTypesSection(), &StrData,
                                       ".apple_types");
  if (!D.getAppleNamespacesSection().Data.empty())
    NumErrors += verifyAppleAccelTable(&D.getAppleNamespacesSection(), &StrData,
                                       ".apple_namespaces");
  if (!D.getAppleObjCSection().Data.empty())
    NumErrors += verifyAppleAccelTable(&D.getAppleObjCSection(), &StrData,
                                       ".apple_objc");

  if (!D.getDebugNamesSection().Data.empty())
    NumErrors += verifyDebugNames(D.getDebugNamesSection(), StrData);
  return NumErrors == 0;
}

raw_ostream &DWARFVerifier::error() const { return WithColor::error(OS); }

raw_ostream &DWARFVerifier::warn() const { return WithColor::warning(OS); }

raw_ostream &DWARFVerifier::note() const { return WithColor::note(OS); }

raw_ostream &DWARFVerifier::dump(const DWARFDie &Die, unsigned indent) const {
  Die.dump(OS, indent, DumpOpts);
  return OS;
}