MachineSink.cpp 53.5 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452
//===- MachineSink.cpp - Sinking for machine instructions -----------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This pass moves instructions into successor blocks when possible, so that
// they aren't executed on paths where their results aren't needed.
//
// This pass is not intended to be a replacement or a complete alternative
// for an LLVM-IR-level sinking pass. It is only designed to sink simple
// constructs that are not exposed before lowering and instruction selection.
//
//===----------------------------------------------------------------------===//

#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/PointerIntPair.h"
#include "llvm/ADT/SetVector.h"
#include "llvm/ADT/SmallSet.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/SparseBitVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
#include "llvm/CodeGen/MachineBlockFrequencyInfo.h"
#include "llvm/CodeGen/MachineBranchProbabilityInfo.h"
#include "llvm/CodeGen/MachineDominators.h"
#include "llvm/CodeGen/MachineFunction.h"
#include "llvm/CodeGen/MachineFunctionPass.h"
#include "llvm/CodeGen/MachineInstr.h"
#include "llvm/CodeGen/MachineLoopInfo.h"
#include "llvm/CodeGen/MachineOperand.h"
#include "llvm/CodeGen/MachinePostDominators.h"
#include "llvm/CodeGen/MachineRegisterInfo.h"
#include "llvm/CodeGen/TargetInstrInfo.h"
#include "llvm/CodeGen/TargetRegisterInfo.h"
#include "llvm/CodeGen/TargetSubtargetInfo.h"
#include "llvm/IR/BasicBlock.h"
#include "llvm/IR/DebugInfoMetadata.h"
#include "llvm/IR/LLVMContext.h"
#include "llvm/InitializePasses.h"
#include "llvm/MC/MCRegisterInfo.h"
#include "llvm/Pass.h"
#include "llvm/Support/BranchProbability.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"
#include <algorithm>
#include <cassert>
#include <cstdint>
#include <map>
#include <utility>
#include <vector>

using namespace llvm;

#define DEBUG_TYPE "machine-sink"

static cl::opt<bool>
SplitEdges("machine-sink-split",
           cl::desc("Split critical edges during machine sinking"),
           cl::init(true), cl::Hidden);

static cl::opt<bool>
UseBlockFreqInfo("machine-sink-bfi",
           cl::desc("Use block frequency info to find successors to sink"),
           cl::init(true), cl::Hidden);

static cl::opt<unsigned> SplitEdgeProbabilityThreshold(
    "machine-sink-split-probability-threshold",
    cl::desc(
        "Percentage threshold for splitting single-instruction critical edge. "
        "If the branch threshold is higher than this threshold, we allow "
        "speculative execution of up to 1 instruction to avoid branching to "
        "splitted critical edge"),
    cl::init(40), cl::Hidden);

STATISTIC(NumSunk,      "Number of machine instructions sunk");
STATISTIC(NumSplit,     "Number of critical edges split");
STATISTIC(NumCoalesces, "Number of copies coalesced");
STATISTIC(NumPostRACopySink, "Number of copies sunk after RA");

namespace {

  class MachineSinking : public MachineFunctionPass {
    const TargetInstrInfo *TII;
    const TargetRegisterInfo *TRI;
    MachineRegisterInfo  *MRI;     // Machine register information
    MachineDominatorTree *DT;      // Machine dominator tree
    MachinePostDominatorTree *PDT; // Machine post dominator tree
    MachineLoopInfo *LI;
    MachineBlockFrequencyInfo *MBFI;
    const MachineBranchProbabilityInfo *MBPI;
    AliasAnalysis *AA;

    // Remember which edges have been considered for breaking.
    SmallSet<std::pair<MachineBasicBlock*, MachineBasicBlock*>, 8>
    CEBCandidates;
    // Remember which edges we are about to split.
    // This is different from CEBCandidates since those edges
    // will be split.
    SetVector<std::pair<MachineBasicBlock *, MachineBasicBlock *>> ToSplit;

    SparseBitVector<> RegsToClearKillFlags;

    using AllSuccsCache =
        std::map<MachineBasicBlock *, SmallVector<MachineBasicBlock *, 4>>;

    /// DBG_VALUE pointer and flag. The flag is true if this DBG_VALUE is
    /// post-dominated by another DBG_VALUE of the same variable location.
    /// This is necessary to detect sequences such as:
    ///     %0 = someinst
    ///     DBG_VALUE %0, !123, !DIExpression()
    ///     %1 = anotherinst
    ///     DBG_VALUE %1, !123, !DIExpression()
    /// Where if %0 were to sink, the DBG_VAUE should not sink with it, as that
    /// would re-order assignments.
    using SeenDbgUser = PointerIntPair<MachineInstr *, 1>;

    /// Record of DBG_VALUE uses of vregs in a block, so that we can identify
    /// debug instructions to sink.
    SmallDenseMap<unsigned, TinyPtrVector<SeenDbgUser>> SeenDbgUsers;

    /// Record of debug variables that have had their locations set in the
    /// current block.
    DenseSet<DebugVariable> SeenDbgVars;

  public:
    static char ID; // Pass identification

    MachineSinking() : MachineFunctionPass(ID) {
      initializeMachineSinkingPass(*PassRegistry::getPassRegistry());
    }

    bool runOnMachineFunction(MachineFunction &MF) override;

    void getAnalysisUsage(AnalysisUsage &AU) const override {
      MachineFunctionPass::getAnalysisUsage(AU);
      AU.addRequired<AAResultsWrapperPass>();
      AU.addRequired<MachineDominatorTree>();
      AU.addRequired<MachinePostDominatorTree>();
      AU.addRequired<MachineLoopInfo>();
      AU.addRequired<MachineBranchProbabilityInfo>();
      AU.addPreserved<MachineLoopInfo>();
      if (UseBlockFreqInfo)
        AU.addRequired<MachineBlockFrequencyInfo>();
    }

    void releaseMemory() override {
      CEBCandidates.clear();
    }

  private:
    bool ProcessBlock(MachineBasicBlock &MBB);
    void ProcessDbgInst(MachineInstr &MI);
    bool isWorthBreakingCriticalEdge(MachineInstr &MI,
                                     MachineBasicBlock *From,
                                     MachineBasicBlock *To);

    /// Postpone the splitting of the given critical
    /// edge (\p From, \p To).
    ///
    /// We do not split the edges on the fly. Indeed, this invalidates
    /// the dominance information and thus triggers a lot of updates
    /// of that information underneath.
    /// Instead, we postpone all the splits after each iteration of
    /// the main loop. That way, the information is at least valid
    /// for the lifetime of an iteration.
    ///
    /// \return True if the edge is marked as toSplit, false otherwise.
    /// False can be returned if, for instance, this is not profitable.
    bool PostponeSplitCriticalEdge(MachineInstr &MI,
                                   MachineBasicBlock *From,
                                   MachineBasicBlock *To,
                                   bool BreakPHIEdge);
    bool SinkInstruction(MachineInstr &MI, bool &SawStore,
                         AllSuccsCache &AllSuccessors);

    /// If we sink a COPY inst, some debug users of it's destination may no
    /// longer be dominated by the COPY, and will eventually be dropped.
    /// This is easily rectified by forwarding the non-dominated debug uses
    /// to the copy source.
    void SalvageUnsunkDebugUsersOfCopy(MachineInstr &,
                                       MachineBasicBlock *TargetBlock);
    bool AllUsesDominatedByBlock(unsigned Reg, MachineBasicBlock *MBB,
                                 MachineBasicBlock *DefMBB,
                                 bool &BreakPHIEdge, bool &LocalUse) const;
    MachineBasicBlock *FindSuccToSinkTo(MachineInstr &MI, MachineBasicBlock *MBB,
               bool &BreakPHIEdge, AllSuccsCache &AllSuccessors);
    bool isProfitableToSinkTo(unsigned Reg, MachineInstr &MI,
                              MachineBasicBlock *MBB,
                              MachineBasicBlock *SuccToSinkTo,
                              AllSuccsCache &AllSuccessors);

    bool PerformTrivialForwardCoalescing(MachineInstr &MI,
                                         MachineBasicBlock *MBB);

    SmallVector<MachineBasicBlock *, 4> &
    GetAllSortedSuccessors(MachineInstr &MI, MachineBasicBlock *MBB,
                           AllSuccsCache &AllSuccessors) const;
  };

} // end anonymous namespace

char MachineSinking::ID = 0;

char &llvm::MachineSinkingID = MachineSinking::ID;

INITIALIZE_PASS_BEGIN(MachineSinking, DEBUG_TYPE,
                      "Machine code sinking", false, false)
INITIALIZE_PASS_DEPENDENCY(MachineBranchProbabilityInfo)
INITIALIZE_PASS_DEPENDENCY(MachineDominatorTree)
INITIALIZE_PASS_DEPENDENCY(MachineLoopInfo)
INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
INITIALIZE_PASS_END(MachineSinking, DEBUG_TYPE,
                    "Machine code sinking", false, false)

bool MachineSinking::PerformTrivialForwardCoalescing(MachineInstr &MI,
                                                     MachineBasicBlock *MBB) {
  if (!MI.isCopy())
    return false;

  Register SrcReg = MI.getOperand(1).getReg();
  Register DstReg = MI.getOperand(0).getReg();
  if (!Register::isVirtualRegister(SrcReg) ||
      !Register::isVirtualRegister(DstReg) || !MRI->hasOneNonDBGUse(SrcReg))
    return false;

  const TargetRegisterClass *SRC = MRI->getRegClass(SrcReg);
  const TargetRegisterClass *DRC = MRI->getRegClass(DstReg);
  if (SRC != DRC)
    return false;

  MachineInstr *DefMI = MRI->getVRegDef(SrcReg);
  if (DefMI->isCopyLike())
    return false;
  LLVM_DEBUG(dbgs() << "Coalescing: " << *DefMI);
  LLVM_DEBUG(dbgs() << "*** to: " << MI);
  MRI->replaceRegWith(DstReg, SrcReg);
  MI.eraseFromParent();

  // Conservatively, clear any kill flags, since it's possible that they are no
  // longer correct.
  MRI->clearKillFlags(SrcReg);

  ++NumCoalesces;
  return true;
}

/// AllUsesDominatedByBlock - Return true if all uses of the specified register
/// occur in blocks dominated by the specified block. If any use is in the
/// definition block, then return false since it is never legal to move def
/// after uses.
bool
MachineSinking::AllUsesDominatedByBlock(unsigned Reg,
                                        MachineBasicBlock *MBB,
                                        MachineBasicBlock *DefMBB,
                                        bool &BreakPHIEdge,
                                        bool &LocalUse) const {
  assert(Register::isVirtualRegister(Reg) && "Only makes sense for vregs");

  // Ignore debug uses because debug info doesn't affect the code.
  if (MRI->use_nodbg_empty(Reg))
    return true;

  // BreakPHIEdge is true if all the uses are in the successor MBB being sunken
  // into and they are all PHI nodes. In this case, machine-sink must break
  // the critical edge first. e.g.
  //
  // %bb.1:
  //   Predecessors according to CFG: %bb.0
  //     ...
  //     %def = DEC64_32r %x, implicit-def dead %eflags
  //     ...
  //     JE_4 <%bb.37>, implicit %eflags
  //   Successors according to CFG: %bb.37 %bb.2
  //
  // %bb.2:
  //     %p = PHI %y, %bb.0, %def, %bb.1
  if (all_of(MRI->use_nodbg_operands(Reg), [&](MachineOperand &MO) {
        MachineInstr *UseInst = MO.getParent();
        unsigned OpNo = UseInst->getOperandNo(&MO);
        MachineBasicBlock *UseBlock = UseInst->getParent();
        return UseBlock == MBB && UseInst->isPHI() &&
               UseInst->getOperand(OpNo + 1).getMBB() == DefMBB;
      })) {
    BreakPHIEdge = true;
    return true;
  }

  for (MachineOperand &MO : MRI->use_nodbg_operands(Reg)) {
    // Determine the block of the use.
    MachineInstr *UseInst = MO.getParent();
    unsigned OpNo = &MO - &UseInst->getOperand(0);
    MachineBasicBlock *UseBlock = UseInst->getParent();
    if (UseInst->isPHI()) {
      // PHI nodes use the operand in the predecessor block, not the block with
      // the PHI.
      UseBlock = UseInst->getOperand(OpNo+1).getMBB();
    } else if (UseBlock == DefMBB) {
      LocalUse = true;
      return false;
    }

    // Check that it dominates.
    if (!DT->dominates(MBB, UseBlock))
      return false;
  }

  return true;
}

bool MachineSinking::runOnMachineFunction(MachineFunction &MF) {
  if (skipFunction(MF.getFunction()))
    return false;

  LLVM_DEBUG(dbgs() << "******** Machine Sinking ********\n");

  TII = MF.getSubtarget().getInstrInfo();
  TRI = MF.getSubtarget().getRegisterInfo();
  MRI = &MF.getRegInfo();
  DT = &getAnalysis<MachineDominatorTree>();
  PDT = &getAnalysis<MachinePostDominatorTree>();
  LI = &getAnalysis<MachineLoopInfo>();
  MBFI = UseBlockFreqInfo ? &getAnalysis<MachineBlockFrequencyInfo>() : nullptr;
  MBPI = &getAnalysis<MachineBranchProbabilityInfo>();
  AA = &getAnalysis<AAResultsWrapperPass>().getAAResults();

  bool EverMadeChange = false;

  while (true) {
    bool MadeChange = false;

    // Process all basic blocks.
    CEBCandidates.clear();
    ToSplit.clear();
    for (auto &MBB: MF)
      MadeChange |= ProcessBlock(MBB);

    // If we have anything we marked as toSplit, split it now.
    for (auto &Pair : ToSplit) {
      auto NewSucc = Pair.first->SplitCriticalEdge(Pair.second, *this);
      if (NewSucc != nullptr) {
        LLVM_DEBUG(dbgs() << " *** Splitting critical edge: "
                          << printMBBReference(*Pair.first) << " -- "
                          << printMBBReference(*NewSucc) << " -- "
                          << printMBBReference(*Pair.second) << '\n');
        if (MBFI)
          MBFI->onEdgeSplit(*Pair.first, *NewSucc, *MBPI);

        MadeChange = true;
        ++NumSplit;
      } else
        LLVM_DEBUG(dbgs() << " *** Not legal to break critical edge\n");
    }
    // If this iteration over the code changed anything, keep iterating.
    if (!MadeChange) break;
    EverMadeChange = true;
  }

  // Now clear any kill flags for recorded registers.
  for (auto I : RegsToClearKillFlags)
    MRI->clearKillFlags(I);
  RegsToClearKillFlags.clear();

  return EverMadeChange;
}

bool MachineSinking::ProcessBlock(MachineBasicBlock &MBB) {
  // Can't sink anything out of a block that has less than two successors.
  if (MBB.succ_size() <= 1 || MBB.empty()) return false;

  // Don't bother sinking code out of unreachable blocks. In addition to being
  // unprofitable, it can also lead to infinite looping, because in an
  // unreachable loop there may be nowhere to stop.
  if (!DT->isReachableFromEntry(&MBB)) return false;

  bool MadeChange = false;

  // Cache all successors, sorted by frequency info and loop depth.
  AllSuccsCache AllSuccessors;

  // Walk the basic block bottom-up.  Remember if we saw a store.
  MachineBasicBlock::iterator I = MBB.end();
  --I;
  bool ProcessedBegin, SawStore = false;
  do {
    MachineInstr &MI = *I;  // The instruction to sink.

    // Predecrement I (if it's not begin) so that it isn't invalidated by
    // sinking.
    ProcessedBegin = I == MBB.begin();
    if (!ProcessedBegin)
      --I;

    if (MI.isDebugInstr()) {
      if (MI.isDebugValue())
        ProcessDbgInst(MI);
      continue;
    }

    bool Joined = PerformTrivialForwardCoalescing(MI, &MBB);
    if (Joined) {
      MadeChange = true;
      continue;
    }

    if (SinkInstruction(MI, SawStore, AllSuccessors)) {
      ++NumSunk;
      MadeChange = true;
    }

    // If we just processed the first instruction in the block, we're done.
  } while (!ProcessedBegin);

  SeenDbgUsers.clear();
  SeenDbgVars.clear();

  return MadeChange;
}

void MachineSinking::ProcessDbgInst(MachineInstr &MI) {
  // When we see DBG_VALUEs for registers, record any vreg it reads, so that
  // we know what to sink if the vreg def sinks.
  assert(MI.isDebugValue() && "Expected DBG_VALUE for processing");

  DebugVariable Var(MI.getDebugVariable(), MI.getDebugExpression(),
                    MI.getDebugLoc()->getInlinedAt());
  bool SeenBefore = SeenDbgVars.count(Var) != 0;

  MachineOperand &MO = MI.getDebugOperand(0);
  if (MO.isReg() && MO.getReg().isVirtual())
    SeenDbgUsers[MO.getReg()].push_back(SeenDbgUser(&MI, SeenBefore));

  // Record the variable for any DBG_VALUE, to avoid re-ordering any of them.
  SeenDbgVars.insert(Var);
}

bool MachineSinking::isWorthBreakingCriticalEdge(MachineInstr &MI,
                                                 MachineBasicBlock *From,
                                                 MachineBasicBlock *To) {
  // FIXME: Need much better heuristics.

  // If the pass has already considered breaking this edge (during this pass
  // through the function), then let's go ahead and break it. This means
  // sinking multiple "cheap" instructions into the same block.
  if (!CEBCandidates.insert(std::make_pair(From, To)).second)
    return true;

  if (!MI.isCopy() && !TII->isAsCheapAsAMove(MI))
    return true;

  if (From->isSuccessor(To) && MBPI->getEdgeProbability(From, To) <=
      BranchProbability(SplitEdgeProbabilityThreshold, 100))
    return true;

  // MI is cheap, we probably don't want to break the critical edge for it.
  // However, if this would allow some definitions of its source operands
  // to be sunk then it's probably worth it.
  for (unsigned i = 0, e = MI.getNumOperands(); i != e; ++i) {
    const MachineOperand &MO = MI.getOperand(i);
    if (!MO.isReg() || !MO.isUse())
      continue;
    Register Reg = MO.getReg();
    if (Reg == 0)
      continue;

    // We don't move live definitions of physical registers,
    // so sinking their uses won't enable any opportunities.
    if (Register::isPhysicalRegister(Reg))
      continue;

    // If this instruction is the only user of a virtual register,
    // check if breaking the edge will enable sinking
    // both this instruction and the defining instruction.
    if (MRI->hasOneNonDBGUse(Reg)) {
      // If the definition resides in same MBB,
      // claim it's likely we can sink these together.
      // If definition resides elsewhere, we aren't
      // blocking it from being sunk so don't break the edge.
      MachineInstr *DefMI = MRI->getVRegDef(Reg);
      if (DefMI->getParent() == MI.getParent())
        return true;
    }
  }

  return false;
}

bool MachineSinking::PostponeSplitCriticalEdge(MachineInstr &MI,
                                               MachineBasicBlock *FromBB,
                                               MachineBasicBlock *ToBB,
                                               bool BreakPHIEdge) {
  if (!isWorthBreakingCriticalEdge(MI, FromBB, ToBB))
    return false;

  // Avoid breaking back edge. From == To means backedge for single BB loop.
  if (!SplitEdges || FromBB == ToBB)
    return false;

  // Check for backedges of more "complex" loops.
  if (LI->getLoopFor(FromBB) == LI->getLoopFor(ToBB) &&
      LI->isLoopHeader(ToBB))
    return false;

  // It's not always legal to break critical edges and sink the computation
  // to the edge.
  //
  // %bb.1:
  // v1024
  // Beq %bb.3
  // <fallthrough>
  // %bb.2:
  // ... no uses of v1024
  // <fallthrough>
  // %bb.3:
  // ...
  //       = v1024
  //
  // If %bb.1 -> %bb.3 edge is broken and computation of v1024 is inserted:
  //
  // %bb.1:
  // ...
  // Bne %bb.2
  // %bb.4:
  // v1024 =
  // B %bb.3
  // %bb.2:
  // ... no uses of v1024
  // <fallthrough>
  // %bb.3:
  // ...
  //       = v1024
  //
  // This is incorrect since v1024 is not computed along the %bb.1->%bb.2->%bb.3
  // flow. We need to ensure the new basic block where the computation is
  // sunk to dominates all the uses.
  // It's only legal to break critical edge and sink the computation to the
  // new block if all the predecessors of "To", except for "From", are
  // not dominated by "From". Given SSA property, this means these
  // predecessors are dominated by "To".
  //
  // There is no need to do this check if all the uses are PHI nodes. PHI
  // sources are only defined on the specific predecessor edges.
  if (!BreakPHIEdge) {
    for (MachineBasicBlock::pred_iterator PI = ToBB->pred_begin(),
           E = ToBB->pred_end(); PI != E; ++PI) {
      if (*PI == FromBB)
        continue;
      if (!DT->dominates(ToBB, *PI))
        return false;
    }
  }

  ToSplit.insert(std::make_pair(FromBB, ToBB));

  return true;
}

/// isProfitableToSinkTo - Return true if it is profitable to sink MI.
bool MachineSinking::isProfitableToSinkTo(unsigned Reg, MachineInstr &MI,
                                          MachineBasicBlock *MBB,
                                          MachineBasicBlock *SuccToSinkTo,
                                          AllSuccsCache &AllSuccessors) {
  assert (SuccToSinkTo && "Invalid SinkTo Candidate BB");

  if (MBB == SuccToSinkTo)
    return false;

  // It is profitable if SuccToSinkTo does not post dominate current block.
  if (!PDT->dominates(SuccToSinkTo, MBB))
    return true;

  // It is profitable to sink an instruction from a deeper loop to a shallower
  // loop, even if the latter post-dominates the former (PR21115).
  if (LI->getLoopDepth(MBB) > LI->getLoopDepth(SuccToSinkTo))
    return true;

  // Check if only use in post dominated block is PHI instruction.
  bool NonPHIUse = false;
  for (MachineInstr &UseInst : MRI->use_nodbg_instructions(Reg)) {
    MachineBasicBlock *UseBlock = UseInst.getParent();
    if (UseBlock == SuccToSinkTo && !UseInst.isPHI())
      NonPHIUse = true;
  }
  if (!NonPHIUse)
    return true;

  // If SuccToSinkTo post dominates then also it may be profitable if MI
  // can further profitably sinked into another block in next round.
  bool BreakPHIEdge = false;
  // FIXME - If finding successor is compile time expensive then cache results.
  if (MachineBasicBlock *MBB2 =
          FindSuccToSinkTo(MI, SuccToSinkTo, BreakPHIEdge, AllSuccessors))
    return isProfitableToSinkTo(Reg, MI, SuccToSinkTo, MBB2, AllSuccessors);

  MachineLoop *ML = LI->getLoopFor(MBB);

  // If the instruction is not inside a loop, it is not profitable to sink MI to
  // a post dominate block SuccToSinkTo.
  if (!ML)
    return false;

  // If this instruction is inside a loop and sinking this instruction can make
  // more registers live range shorten, it is still prifitable.
  for (unsigned i = 0, e = MI.getNumOperands(); i != e; ++i) {
    const MachineOperand &MO = MI.getOperand(i);
    // Ignore non-register operands.
    if (!MO.isReg())
      continue;
    Register Reg = MO.getReg();
    if (Reg == 0)
      continue;

    // Don't handle physical register.
    if (Register::isPhysicalRegister(Reg))
      return false;

    // Users for the defs are all dominated by SuccToSinkTo.
    if (MO.isDef()) {
      // This def register's live range is shortened after sinking.
      bool LocalUse = false;
      if (!AllUsesDominatedByBlock(Reg, SuccToSinkTo, MBB, BreakPHIEdge,
                                   LocalUse))
        return false;
    } else {
      MachineInstr *DefMI = MRI->getVRegDef(Reg);
      // DefMI is defined outside of loop. There should be no live range
      // impact for this operand. Defination outside of loop means:
      // 1: defination is outside of loop.
      // 2: defination is in this loop, but it is a PHI in the loop header.
      if (LI->getLoopFor(DefMI->getParent()) != ML ||
          (DefMI->isPHI() && LI->isLoopHeader(DefMI->getParent())))
        continue;
      // DefMI is inside the loop. Mark it as not profitable as sinking MI will
      // enlarge DefMI live range.
      // FIXME: check the register pressure in block SuccToSinkTo, if it is
      // smaller than the limit after sinking, it is still profitable to sink.
      return false;
    }
  }

  // If MI is in loop and all its operands are alive across the whole loop, it
  // is profitable to sink MI.
  return true;
}

/// Get the sorted sequence of successors for this MachineBasicBlock, possibly
/// computing it if it was not already cached.
SmallVector<MachineBasicBlock *, 4> &
MachineSinking::GetAllSortedSuccessors(MachineInstr &MI, MachineBasicBlock *MBB,
                                       AllSuccsCache &AllSuccessors) const {
  // Do we have the sorted successors in cache ?
  auto Succs = AllSuccessors.find(MBB);
  if (Succs != AllSuccessors.end())
    return Succs->second;

  SmallVector<MachineBasicBlock *, 4> AllSuccs(MBB->succ_begin(),
                                               MBB->succ_end());

  // Handle cases where sinking can happen but where the sink point isn't a
  // successor. For example:
  //
  //   x = computation
  //   if () {} else {}
  //   use x
  //
  for (MachineDomTreeNode *DTChild : DT->getNode(MBB)->children()) {
    // DomTree children of MBB that have MBB as immediate dominator are added.
    if (DTChild->getIDom()->getBlock() == MI.getParent() &&
        // Skip MBBs already added to the AllSuccs vector above.
        !MBB->isSuccessor(DTChild->getBlock()))
      AllSuccs.push_back(DTChild->getBlock());
  }

  // Sort Successors according to their loop depth or block frequency info.
  llvm::stable_sort(
      AllSuccs, [this](const MachineBasicBlock *L, const MachineBasicBlock *R) {
        uint64_t LHSFreq = MBFI ? MBFI->getBlockFreq(L).getFrequency() : 0;
        uint64_t RHSFreq = MBFI ? MBFI->getBlockFreq(R).getFrequency() : 0;
        bool HasBlockFreq = LHSFreq != 0 && RHSFreq != 0;
        return HasBlockFreq ? LHSFreq < RHSFreq
                            : LI->getLoopDepth(L) < LI->getLoopDepth(R);
      });

  auto it = AllSuccessors.insert(std::make_pair(MBB, AllSuccs));

  return it.first->second;
}

/// FindSuccToSinkTo - Find a successor to sink this instruction to.
MachineBasicBlock *
MachineSinking::FindSuccToSinkTo(MachineInstr &MI, MachineBasicBlock *MBB,
                                 bool &BreakPHIEdge,
                                 AllSuccsCache &AllSuccessors) {
  assert (MBB && "Invalid MachineBasicBlock!");

  // Loop over all the operands of the specified instruction.  If there is
  // anything we can't handle, bail out.

  // SuccToSinkTo - This is the successor to sink this instruction to, once we
  // decide.
  MachineBasicBlock *SuccToSinkTo = nullptr;
  for (unsigned i = 0, e = MI.getNumOperands(); i != e; ++i) {
    const MachineOperand &MO = MI.getOperand(i);
    if (!MO.isReg()) continue;  // Ignore non-register operands.

    Register Reg = MO.getReg();
    if (Reg == 0) continue;

    if (Register::isPhysicalRegister(Reg)) {
      if (MO.isUse()) {
        // If the physreg has no defs anywhere, it's just an ambient register
        // and we can freely move its uses. Alternatively, if it's allocatable,
        // it could get allocated to something with a def during allocation.
        if (!MRI->isConstantPhysReg(Reg))
          return nullptr;
      } else if (!MO.isDead()) {
        // A def that isn't dead. We can't move it.
        return nullptr;
      }
    } else {
      // Virtual register uses are always safe to sink.
      if (MO.isUse()) continue;

      // If it's not safe to move defs of the register class, then abort.
      if (!TII->isSafeToMoveRegClassDefs(MRI->getRegClass(Reg)))
        return nullptr;

      // Virtual register defs can only be sunk if all their uses are in blocks
      // dominated by one of the successors.
      if (SuccToSinkTo) {
        // If a previous operand picked a block to sink to, then this operand
        // must be sinkable to the same block.
        bool LocalUse = false;
        if (!AllUsesDominatedByBlock(Reg, SuccToSinkTo, MBB,
                                     BreakPHIEdge, LocalUse))
          return nullptr;

        continue;
      }

      // Otherwise, we should look at all the successors and decide which one
      // we should sink to. If we have reliable block frequency information
      // (frequency != 0) available, give successors with smaller frequencies
      // higher priority, otherwise prioritize smaller loop depths.
      for (MachineBasicBlock *SuccBlock :
           GetAllSortedSuccessors(MI, MBB, AllSuccessors)) {
        bool LocalUse = false;
        if (AllUsesDominatedByBlock(Reg, SuccBlock, MBB,
                                    BreakPHIEdge, LocalUse)) {
          SuccToSinkTo = SuccBlock;
          break;
        }
        if (LocalUse)
          // Def is used locally, it's never safe to move this def.
          return nullptr;
      }

      // If we couldn't find a block to sink to, ignore this instruction.
      if (!SuccToSinkTo)
        return nullptr;
      if (!isProfitableToSinkTo(Reg, MI, MBB, SuccToSinkTo, AllSuccessors))
        return nullptr;
    }
  }

  // It is not possible to sink an instruction into its own block.  This can
  // happen with loops.
  if (MBB == SuccToSinkTo)
    return nullptr;

  // It's not safe to sink instructions to EH landing pad. Control flow into
  // landing pad is implicitly defined.
  if (SuccToSinkTo && SuccToSinkTo->isEHPad())
    return nullptr;

  // It ought to be okay to sink instructions into an INLINEASM_BR target, but
  // only if we make sure that MI occurs _before_ an INLINEASM_BR instruction in
  // the source block (which this code does not yet do). So for now, forbid
  // doing so.
  if (SuccToSinkTo && SuccToSinkTo->isInlineAsmBrIndirectTarget())
    return nullptr;

  return SuccToSinkTo;
}

/// Return true if MI is likely to be usable as a memory operation by the
/// implicit null check optimization.
///
/// This is a "best effort" heuristic, and should not be relied upon for
/// correctness.  This returning true does not guarantee that the implicit null
/// check optimization is legal over MI, and this returning false does not
/// guarantee MI cannot possibly be used to do a null check.
static bool SinkingPreventsImplicitNullCheck(MachineInstr &MI,
                                             const TargetInstrInfo *TII,
                                             const TargetRegisterInfo *TRI) {
  using MachineBranchPredicate = TargetInstrInfo::MachineBranchPredicate;

  auto *MBB = MI.getParent();
  if (MBB->pred_size() != 1)
    return false;

  auto *PredMBB = *MBB->pred_begin();
  auto *PredBB = PredMBB->getBasicBlock();

  // Frontends that don't use implicit null checks have no reason to emit
  // branches with make.implicit metadata, and this function should always
  // return false for them.
  if (!PredBB ||
      !PredBB->getTerminator()->getMetadata(LLVMContext::MD_make_implicit))
    return false;

  const MachineOperand *BaseOp;
  int64_t Offset;
  bool OffsetIsScalable;
  if (!TII->getMemOperandWithOffset(MI, BaseOp, Offset, OffsetIsScalable, TRI))
    return false;

  if (!BaseOp->isReg())
    return false;

  if (!(MI.mayLoad() && !MI.isPredicable()))
    return false;

  MachineBranchPredicate MBP;
  if (TII->analyzeBranchPredicate(*PredMBB, MBP, false))
    return false;

  return MBP.LHS.isReg() && MBP.RHS.isImm() && MBP.RHS.getImm() == 0 &&
         (MBP.Predicate == MachineBranchPredicate::PRED_NE ||
          MBP.Predicate == MachineBranchPredicate::PRED_EQ) &&
         MBP.LHS.getReg() == BaseOp->getReg();
}

/// If the sunk instruction is a copy, try to forward the copy instead of
/// leaving an 'undef' DBG_VALUE in the original location. Don't do this if
/// there's any subregister weirdness involved. Returns true if copy
/// propagation occurred.
static bool attemptDebugCopyProp(MachineInstr &SinkInst, MachineInstr &DbgMI) {
  const MachineRegisterInfo &MRI = SinkInst.getMF()->getRegInfo();
  const TargetInstrInfo &TII = *SinkInst.getMF()->getSubtarget().getInstrInfo();

  // Copy DBG_VALUE operand and set the original to undef. We then check to
  // see whether this is something that can be copy-forwarded. If it isn't,
  // continue around the loop.
  MachineOperand &DbgMO = DbgMI.getDebugOperand(0);

  const MachineOperand *SrcMO = nullptr, *DstMO = nullptr;
  auto CopyOperands = TII.isCopyInstr(SinkInst);
  if (!CopyOperands)
    return false;
  SrcMO = CopyOperands->Source;
  DstMO = CopyOperands->Destination;

  // Check validity of forwarding this copy.
  bool PostRA = MRI.getNumVirtRegs() == 0;

  // Trying to forward between physical and virtual registers is too hard.
  if (DbgMO.getReg().isVirtual() != SrcMO->getReg().isVirtual())
    return false;

  // Only try virtual register copy-forwarding before regalloc, and physical
  // register copy-forwarding after regalloc.
  bool arePhysRegs = !DbgMO.getReg().isVirtual();
  if (arePhysRegs != PostRA)
    return false;

  // Pre-regalloc, only forward if all subregisters agree (or there are no
  // subregs at all). More analysis might recover some forwardable copies.
  if (!PostRA && (DbgMO.getSubReg() != SrcMO->getSubReg() ||
                  DbgMO.getSubReg() != DstMO->getSubReg()))
    return false;

  // Post-regalloc, we may be sinking a DBG_VALUE of a sub or super-register
  // of this copy. Only forward the copy if the DBG_VALUE operand exactly
  // matches the copy destination.
  if (PostRA && DbgMO.getReg() != DstMO->getReg())
    return false;

  DbgMO.setReg(SrcMO->getReg());
  DbgMO.setSubReg(SrcMO->getSubReg());
  return true;
}

/// Sink an instruction and its associated debug instructions.
static void performSink(MachineInstr &MI, MachineBasicBlock &SuccToSinkTo,
                        MachineBasicBlock::iterator InsertPos,
                        SmallVectorImpl<MachineInstr *> &DbgValuesToSink) {

  // If we cannot find a location to use (merge with), then we erase the debug
  // location to prevent debug-info driven tools from potentially reporting
  // wrong location information.
  if (!SuccToSinkTo.empty() && InsertPos != SuccToSinkTo.end())
    MI.setDebugLoc(DILocation::getMergedLocation(MI.getDebugLoc(),
                                                 InsertPos->getDebugLoc()));
  else
    MI.setDebugLoc(DebugLoc());

  // Move the instruction.
  MachineBasicBlock *ParentBlock = MI.getParent();
  SuccToSinkTo.splice(InsertPos, ParentBlock, MI,
                      ++MachineBasicBlock::iterator(MI));

  // Sink a copy of debug users to the insert position. Mark the original
  // DBG_VALUE location as 'undef', indicating that any earlier variable
  // location should be terminated as we've optimised away the value at this
  // point.
  for (SmallVectorImpl<MachineInstr *>::iterator DBI = DbgValuesToSink.begin(),
                                                 DBE = DbgValuesToSink.end();
       DBI != DBE; ++DBI) {
    MachineInstr *DbgMI = *DBI;
    MachineInstr *NewDbgMI = DbgMI->getMF()->CloneMachineInstr(*DBI);
    SuccToSinkTo.insert(InsertPos, NewDbgMI);

    if (!attemptDebugCopyProp(MI, *DbgMI))
      DbgMI->setDebugValueUndef();
  }
}

/// SinkInstruction - Determine whether it is safe to sink the specified machine
/// instruction out of its current block into a successor.
bool MachineSinking::SinkInstruction(MachineInstr &MI, bool &SawStore,
                                     AllSuccsCache &AllSuccessors) {
  // Don't sink instructions that the target prefers not to sink.
  if (!TII->shouldSink(MI))
    return false;

  // Check if it's safe to move the instruction.
  if (!MI.isSafeToMove(AA, SawStore))
    return false;

  // Convergent operations may not be made control-dependent on additional
  // values.
  if (MI.isConvergent())
    return false;

  // Don't break implicit null checks.  This is a performance heuristic, and not
  // required for correctness.
  if (SinkingPreventsImplicitNullCheck(MI, TII, TRI))
    return false;

  // FIXME: This should include support for sinking instructions within the
  // block they are currently in to shorten the live ranges.  We often get
  // instructions sunk into the top of a large block, but it would be better to
  // also sink them down before their first use in the block.  This xform has to
  // be careful not to *increase* register pressure though, e.g. sinking
  // "x = y + z" down if it kills y and z would increase the live ranges of y
  // and z and only shrink the live range of x.

  bool BreakPHIEdge = false;
  MachineBasicBlock *ParentBlock = MI.getParent();
  MachineBasicBlock *SuccToSinkTo =
      FindSuccToSinkTo(MI, ParentBlock, BreakPHIEdge, AllSuccessors);

  // If there are no outputs, it must have side-effects.
  if (!SuccToSinkTo)
    return false;

  // If the instruction to move defines a dead physical register which is live
  // when leaving the basic block, don't move it because it could turn into a
  // "zombie" define of that preg. E.g., EFLAGS. (<rdar://problem/8030636>)
  for (unsigned I = 0, E = MI.getNumOperands(); I != E; ++I) {
    const MachineOperand &MO = MI.getOperand(I);
    if (!MO.isReg()) continue;
    Register Reg = MO.getReg();
    if (Reg == 0 || !Register::isPhysicalRegister(Reg))
      continue;
    if (SuccToSinkTo->isLiveIn(Reg))
      return false;
  }

  LLVM_DEBUG(dbgs() << "Sink instr " << MI << "\tinto block " << *SuccToSinkTo);

  // If the block has multiple predecessors, this is a critical edge.
  // Decide if we can sink along it or need to break the edge.
  if (SuccToSinkTo->pred_size() > 1) {
    // We cannot sink a load across a critical edge - there may be stores in
    // other code paths.
    bool TryBreak = false;
    bool store = true;
    if (!MI.isSafeToMove(AA, store)) {
      LLVM_DEBUG(dbgs() << " *** NOTE: Won't sink load along critical edge.\n");
      TryBreak = true;
    }

    // We don't want to sink across a critical edge if we don't dominate the
    // successor. We could be introducing calculations to new code paths.
    if (!TryBreak && !DT->dominates(ParentBlock, SuccToSinkTo)) {
      LLVM_DEBUG(dbgs() << " *** NOTE: Critical edge found\n");
      TryBreak = true;
    }

    // Don't sink instructions into a loop.
    if (!TryBreak && LI->isLoopHeader(SuccToSinkTo)) {
      LLVM_DEBUG(dbgs() << " *** NOTE: Loop header found\n");
      TryBreak = true;
    }

    // Otherwise we are OK with sinking along a critical edge.
    if (!TryBreak)
      LLVM_DEBUG(dbgs() << "Sinking along critical edge.\n");
    else {
      // Mark this edge as to be split.
      // If the edge can actually be split, the next iteration of the main loop
      // will sink MI in the newly created block.
      bool Status =
        PostponeSplitCriticalEdge(MI, ParentBlock, SuccToSinkTo, BreakPHIEdge);
      if (!Status)
        LLVM_DEBUG(dbgs() << " *** PUNTING: Not legal or profitable to "
                             "break critical edge\n");
      // The instruction will not be sunk this time.
      return false;
    }
  }

  if (BreakPHIEdge) {
    // BreakPHIEdge is true if all the uses are in the successor MBB being
    // sunken into and they are all PHI nodes. In this case, machine-sink must
    // break the critical edge first.
    bool Status = PostponeSplitCriticalEdge(MI, ParentBlock,
                                            SuccToSinkTo, BreakPHIEdge);
    if (!Status)
      LLVM_DEBUG(dbgs() << " *** PUNTING: Not legal or profitable to "
                           "break critical edge\n");
    // The instruction will not be sunk this time.
    return false;
  }

  // Determine where to insert into. Skip phi nodes.
  MachineBasicBlock::iterator InsertPos = SuccToSinkTo->begin();
  while (InsertPos != SuccToSinkTo->end() && InsertPos->isPHI())
    ++InsertPos;

  // Collect debug users of any vreg that this inst defines.
  SmallVector<MachineInstr *, 4> DbgUsersToSink;
  for (auto &MO : MI.operands()) {
    if (!MO.isReg() || !MO.isDef() || !MO.getReg().isVirtual())
      continue;
    if (!SeenDbgUsers.count(MO.getReg()))
      continue;

    // Sink any users that don't pass any other DBG_VALUEs for this variable.
    auto &Users = SeenDbgUsers[MO.getReg()];
    for (auto &User : Users) {
      MachineInstr *DbgMI = User.getPointer();
      if (User.getInt()) {
        // This DBG_VALUE would re-order assignments. If we can't copy-propagate
        // it, it can't be recovered. Set it undef.
        if (!attemptDebugCopyProp(MI, *DbgMI))
          DbgMI->setDebugValueUndef();
      } else {
        DbgUsersToSink.push_back(DbgMI);
      }
    }
  }

  // After sinking, some debug users may not be dominated any more. If possible,
  // copy-propagate their operands. As it's expensive, don't do this if there's
  // no debuginfo in the program.
  if (MI.getMF()->getFunction().getSubprogram() && MI.isCopy())
    SalvageUnsunkDebugUsersOfCopy(MI, SuccToSinkTo);

  performSink(MI, *SuccToSinkTo, InsertPos, DbgUsersToSink);

  // Conservatively, clear any kill flags, since it's possible that they are no
  // longer correct.
  // Note that we have to clear the kill flags for any register this instruction
  // uses as we may sink over another instruction which currently kills the
  // used registers.
  for (MachineOperand &MO : MI.operands()) {
    if (MO.isReg() && MO.isUse())
      RegsToClearKillFlags.set(MO.getReg()); // Remember to clear kill flags.
  }

  return true;
}

void MachineSinking::SalvageUnsunkDebugUsersOfCopy(
    MachineInstr &MI, MachineBasicBlock *TargetBlock) {
  assert(MI.isCopy());
  assert(MI.getOperand(1).isReg());

  // Enumerate all users of vreg operands that are def'd. Skip those that will
  // be sunk. For the rest, if they are not dominated by the block we will sink
  // MI into, propagate the copy source to them.
  SmallVector<MachineInstr *, 4> DbgDefUsers;
  const MachineRegisterInfo &MRI = MI.getMF()->getRegInfo();
  for (auto &MO : MI.operands()) {
    if (!MO.isReg() || !MO.isDef() || !MO.getReg().isVirtual())
      continue;
    for (auto &User : MRI.use_instructions(MO.getReg())) {
      if (!User.isDebugValue() || DT->dominates(TargetBlock, User.getParent()))
        continue;

      // If is in same block, will either sink or be use-before-def.
      if (User.getParent() == MI.getParent())
        continue;

      assert(User.getDebugOperand(0).isReg() &&
             "DBG_VALUE user of vreg, but non reg operand?");
      DbgDefUsers.push_back(&User);
    }
  }

  // Point the users of this copy that are no longer dominated, at the source
  // of the copy.
  for (auto *User : DbgDefUsers) {
    User->getDebugOperand(0).setReg(MI.getOperand(1).getReg());
    User->getDebugOperand(0).setSubReg(MI.getOperand(1).getSubReg());
  }
}

//===----------------------------------------------------------------------===//
// This pass is not intended to be a replacement or a complete alternative
// for the pre-ra machine sink pass. It is only designed to sink COPY
// instructions which should be handled after RA.
//
// This pass sinks COPY instructions into a successor block, if the COPY is not
// used in the current block and the COPY is live-in to a single successor
// (i.e., doesn't require the COPY to be duplicated).  This avoids executing the
// copy on paths where their results aren't needed.  This also exposes
// additional opportunites for dead copy elimination and shrink wrapping.
//
// These copies were either not handled by or are inserted after the MachineSink
// pass. As an example of the former case, the MachineSink pass cannot sink
// COPY instructions with allocatable source registers; for AArch64 these type
// of copy instructions are frequently used to move function parameters (PhyReg)
// into virtual registers in the entry block.
//
// For the machine IR below, this pass will sink %w19 in the entry into its
// successor (%bb.1) because %w19 is only live-in in %bb.1.
// %bb.0:
//   %wzr = SUBSWri %w1, 1
//   %w19 = COPY %w0
//   Bcc 11, %bb.2
// %bb.1:
//   Live Ins: %w19
//   BL @fun
//   %w0 = ADDWrr %w0, %w19
//   RET %w0
// %bb.2:
//   %w0 = COPY %wzr
//   RET %w0
// As we sink %w19 (CSR in AArch64) into %bb.1, the shrink-wrapping pass will be
// able to see %bb.0 as a candidate.
//===----------------------------------------------------------------------===//
namespace {

class PostRAMachineSinking : public MachineFunctionPass {
public:
  bool runOnMachineFunction(MachineFunction &MF) override;

  static char ID;
  PostRAMachineSinking() : MachineFunctionPass(ID) {}
  StringRef getPassName() const override { return "PostRA Machine Sink"; }

  void getAnalysisUsage(AnalysisUsage &AU) const override {
    AU.setPreservesCFG();
    MachineFunctionPass::getAnalysisUsage(AU);
  }

  MachineFunctionProperties getRequiredProperties() const override {
    return MachineFunctionProperties().set(
        MachineFunctionProperties::Property::NoVRegs);
  }

private:
  /// Track which register units have been modified and used.
  LiveRegUnits ModifiedRegUnits, UsedRegUnits;

  /// Track DBG_VALUEs of (unmodified) register units. Each DBG_VALUE has an
  /// entry in this map for each unit it touches.
  DenseMap<unsigned, TinyPtrVector<MachineInstr *>> SeenDbgInstrs;

  /// Sink Copy instructions unused in the same block close to their uses in
  /// successors.
  bool tryToSinkCopy(MachineBasicBlock &BB, MachineFunction &MF,
                     const TargetRegisterInfo *TRI, const TargetInstrInfo *TII);
};
} // namespace

char PostRAMachineSinking::ID = 0;
char &llvm::PostRAMachineSinkingID = PostRAMachineSinking::ID;

INITIALIZE_PASS(PostRAMachineSinking, "postra-machine-sink",
                "PostRA Machine Sink", false, false)

static bool aliasWithRegsInLiveIn(MachineBasicBlock &MBB, unsigned Reg,
                                  const TargetRegisterInfo *TRI) {
  LiveRegUnits LiveInRegUnits(*TRI);
  LiveInRegUnits.addLiveIns(MBB);
  return !LiveInRegUnits.available(Reg);
}

static MachineBasicBlock *
getSingleLiveInSuccBB(MachineBasicBlock &CurBB,
                      const SmallPtrSetImpl<MachineBasicBlock *> &SinkableBBs,
                      unsigned Reg, const TargetRegisterInfo *TRI) {
  // Try to find a single sinkable successor in which Reg is live-in.
  MachineBasicBlock *BB = nullptr;
  for (auto *SI : SinkableBBs) {
    if (aliasWithRegsInLiveIn(*SI, Reg, TRI)) {
      // If BB is set here, Reg is live-in to at least two sinkable successors,
      // so quit.
      if (BB)
        return nullptr;
      BB = SI;
    }
  }
  // Reg is not live-in to any sinkable successors.
  if (!BB)
    return nullptr;

  // Check if any register aliased with Reg is live-in in other successors.
  for (auto *SI : CurBB.successors()) {
    if (!SinkableBBs.count(SI) && aliasWithRegsInLiveIn(*SI, Reg, TRI))
      return nullptr;
  }
  return BB;
}

static MachineBasicBlock *
getSingleLiveInSuccBB(MachineBasicBlock &CurBB,
                      const SmallPtrSetImpl<MachineBasicBlock *> &SinkableBBs,
                      ArrayRef<unsigned> DefedRegsInCopy,
                      const TargetRegisterInfo *TRI) {
  MachineBasicBlock *SingleBB = nullptr;
  for (auto DefReg : DefedRegsInCopy) {
    MachineBasicBlock *BB =
        getSingleLiveInSuccBB(CurBB, SinkableBBs, DefReg, TRI);
    if (!BB || (SingleBB && SingleBB != BB))
      return nullptr;
    SingleBB = BB;
  }
  return SingleBB;
}

static void clearKillFlags(MachineInstr *MI, MachineBasicBlock &CurBB,
                           SmallVectorImpl<unsigned> &UsedOpsInCopy,
                           LiveRegUnits &UsedRegUnits,
                           const TargetRegisterInfo *TRI) {
  for (auto U : UsedOpsInCopy) {
    MachineOperand &MO = MI->getOperand(U);
    Register SrcReg = MO.getReg();
    if (!UsedRegUnits.available(SrcReg)) {
      MachineBasicBlock::iterator NI = std::next(MI->getIterator());
      for (MachineInstr &UI : make_range(NI, CurBB.end())) {
        if (UI.killsRegister(SrcReg, TRI)) {
          UI.clearRegisterKills(SrcReg, TRI);
          MO.setIsKill(true);
          break;
        }
      }
    }
  }
}

static void updateLiveIn(MachineInstr *MI, MachineBasicBlock *SuccBB,
                         SmallVectorImpl<unsigned> &UsedOpsInCopy,
                         SmallVectorImpl<unsigned> &DefedRegsInCopy) {
  MachineFunction &MF = *SuccBB->getParent();
  const TargetRegisterInfo *TRI = MF.getSubtarget().getRegisterInfo();
  for (unsigned DefReg : DefedRegsInCopy)
    for (MCSubRegIterator S(DefReg, TRI, true); S.isValid(); ++S)
      SuccBB->removeLiveIn(*S);
  for (auto U : UsedOpsInCopy) {
    Register SrcReg = MI->getOperand(U).getReg();
    LaneBitmask Mask;
    for (MCRegUnitMaskIterator S(SrcReg, TRI); S.isValid(); ++S) {
      Mask |= (*S).second;
    }
    SuccBB->addLiveIn(SrcReg, Mask.any() ? Mask : LaneBitmask::getAll());
  }
  SuccBB->sortUniqueLiveIns();
}

static bool hasRegisterDependency(MachineInstr *MI,
                                  SmallVectorImpl<unsigned> &UsedOpsInCopy,
                                  SmallVectorImpl<unsigned> &DefedRegsInCopy,
                                  LiveRegUnits &ModifiedRegUnits,
                                  LiveRegUnits &UsedRegUnits) {
  bool HasRegDependency = false;
  for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
    MachineOperand &MO = MI->getOperand(i);
    if (!MO.isReg())
      continue;
    Register Reg = MO.getReg();
    if (!Reg)
      continue;
    if (MO.isDef()) {
      if (!ModifiedRegUnits.available(Reg) || !UsedRegUnits.available(Reg)) {
        HasRegDependency = true;
        break;
      }
      DefedRegsInCopy.push_back(Reg);

      // FIXME: instead of isUse(), readsReg() would be a better fix here,
      // For example, we can ignore modifications in reg with undef. However,
      // it's not perfectly clear if skipping the internal read is safe in all
      // other targets.
    } else if (MO.isUse()) {
      if (!ModifiedRegUnits.available(Reg)) {
        HasRegDependency = true;
        break;
      }
      UsedOpsInCopy.push_back(i);
    }
  }
  return HasRegDependency;
}

static SmallSet<unsigned, 4> getRegUnits(unsigned Reg,
                                         const TargetRegisterInfo *TRI) {
  SmallSet<unsigned, 4> RegUnits;
  for (auto RI = MCRegUnitIterator(Reg, TRI); RI.isValid(); ++RI)
    RegUnits.insert(*RI);
  return RegUnits;
}

bool PostRAMachineSinking::tryToSinkCopy(MachineBasicBlock &CurBB,
                                         MachineFunction &MF,
                                         const TargetRegisterInfo *TRI,
                                         const TargetInstrInfo *TII) {
  SmallPtrSet<MachineBasicBlock *, 2> SinkableBBs;
  // FIXME: For now, we sink only to a successor which has a single predecessor
  // so that we can directly sink COPY instructions to the successor without
  // adding any new block or branch instruction.
  for (MachineBasicBlock *SI : CurBB.successors())
    if (!SI->livein_empty() && SI->pred_size() == 1)
      SinkableBBs.insert(SI);

  if (SinkableBBs.empty())
    return false;

  bool Changed = false;

  // Track which registers have been modified and used between the end of the
  // block and the current instruction.
  ModifiedRegUnits.clear();
  UsedRegUnits.clear();
  SeenDbgInstrs.clear();

  for (auto I = CurBB.rbegin(), E = CurBB.rend(); I != E;) {
    MachineInstr *MI = &*I;
    ++I;

    // Track the operand index for use in Copy.
    SmallVector<unsigned, 2> UsedOpsInCopy;
    // Track the register number defed in Copy.
    SmallVector<unsigned, 2> DefedRegsInCopy;

    // We must sink this DBG_VALUE if its operand is sunk. To avoid searching
    // for DBG_VALUEs later, record them when they're encountered.
    if (MI->isDebugValue()) {
      auto &MO = MI->getDebugOperand(0);
      if (MO.isReg() && Register::isPhysicalRegister(MO.getReg())) {
        // Bail if we can already tell the sink would be rejected, rather
        // than needlessly accumulating lots of DBG_VALUEs.
        if (hasRegisterDependency(MI, UsedOpsInCopy, DefedRegsInCopy,
                                  ModifiedRegUnits, UsedRegUnits))
          continue;

        // Record debug use of each reg unit.
        SmallSet<unsigned, 4> Units = getRegUnits(MO.getReg(), TRI);
        for (unsigned Reg : Units)
          SeenDbgInstrs[Reg].push_back(MI);
      }
      continue;
    }

    if (MI->isDebugInstr())
      continue;

    // Do not move any instruction across function call.
    if (MI->isCall())
      return false;

    if (!MI->isCopy() || !MI->getOperand(0).isRenamable()) {
      LiveRegUnits::accumulateUsedDefed(*MI, ModifiedRegUnits, UsedRegUnits,
                                        TRI);
      continue;
    }

    // Don't sink the COPY if it would violate a register dependency.
    if (hasRegisterDependency(MI, UsedOpsInCopy, DefedRegsInCopy,
                              ModifiedRegUnits, UsedRegUnits)) {
      LiveRegUnits::accumulateUsedDefed(*MI, ModifiedRegUnits, UsedRegUnits,
                                        TRI);
      continue;
    }
    assert((!UsedOpsInCopy.empty() && !DefedRegsInCopy.empty()) &&
           "Unexpect SrcReg or DefReg");
    MachineBasicBlock *SuccBB =
        getSingleLiveInSuccBB(CurBB, SinkableBBs, DefedRegsInCopy, TRI);
    // Don't sink if we cannot find a single sinkable successor in which Reg
    // is live-in.
    if (!SuccBB) {
      LiveRegUnits::accumulateUsedDefed(*MI, ModifiedRegUnits, UsedRegUnits,
                                        TRI);
      continue;
    }
    assert((SuccBB->pred_size() == 1 && *SuccBB->pred_begin() == &CurBB) &&
           "Unexpected predecessor");

    // Collect DBG_VALUEs that must sink with this copy. We've previously
    // recorded which reg units that DBG_VALUEs read, if this instruction
    // writes any of those units then the corresponding DBG_VALUEs must sink.
    SetVector<MachineInstr *> DbgValsToSinkSet;
    SmallVector<MachineInstr *, 4> DbgValsToSink;
    for (auto &MO : MI->operands()) {
      if (!MO.isReg() || !MO.isDef())
        continue;

      SmallSet<unsigned, 4> Units = getRegUnits(MO.getReg(), TRI);
      for (unsigned Reg : Units)
        for (auto *MI : SeenDbgInstrs.lookup(Reg))
          DbgValsToSinkSet.insert(MI);
    }
    DbgValsToSink.insert(DbgValsToSink.begin(), DbgValsToSinkSet.begin(),
                         DbgValsToSinkSet.end());

    // Clear the kill flag if SrcReg is killed between MI and the end of the
    // block.
    clearKillFlags(MI, CurBB, UsedOpsInCopy, UsedRegUnits, TRI);
    MachineBasicBlock::iterator InsertPos = SuccBB->getFirstNonPHI();
    performSink(*MI, *SuccBB, InsertPos, DbgValsToSink);
    updateLiveIn(MI, SuccBB, UsedOpsInCopy, DefedRegsInCopy);

    Changed = true;
    ++NumPostRACopySink;
  }
  return Changed;
}

bool PostRAMachineSinking::runOnMachineFunction(MachineFunction &MF) {
  if (skipFunction(MF.getFunction()))
    return false;

  bool Changed = false;
  const TargetRegisterInfo *TRI = MF.getSubtarget().getRegisterInfo();
  const TargetInstrInfo *TII = MF.getSubtarget().getInstrInfo();

  ModifiedRegUnits.init(*TRI);
  UsedRegUnits.init(*TRI);
  for (auto &BB : MF)
    Changed |= tryToSinkCopy(BB, MF, TRI, TII);

  return Changed;
}