aboutsummaryrefslogtreecommitdiffstats
path: root/lib/CodeGen/AtomicExpandPass.cpp
blob: 4b64be053dfd31fa928895369e86ff7dab515cfa (plain)
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
//===-- AtomicExpandPass.cpp - Expand atomic instructions -------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file contains a pass (at IR level) to replace atomic instructions with
// either (intrinsic-based) load-linked/store-conditional loops or AtomicCmpXchg.
//
//===----------------------------------------------------------------------===//

#include "llvm/CodeGen/Passes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/IR/InstIterator.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/Intrinsics.h"
#include "llvm/IR/Module.h"
#include "llvm/Support/Debug.h"
#include "llvm/Target/TargetLowering.h"
#include "llvm/Target/TargetMachine.h"
#include "llvm/Target/TargetSubtargetInfo.h"

using namespace llvm;

#define DEBUG_TYPE "atomic-expand"

namespace {
  class AtomicExpand: public FunctionPass {
    const TargetMachine *TM;
    const TargetLowering *TLI;
  public:
    static char ID; // Pass identification, replacement for typeid
    explicit AtomicExpand(const TargetMachine *TM = nullptr)
      : FunctionPass(ID), TM(TM), TLI(nullptr) {
      initializeAtomicExpandPass(*PassRegistry::getPassRegistry());
    }

    bool runOnFunction(Function &F) override;

  private:
    bool bracketInstWithFences(Instruction *I, AtomicOrdering Order,
                               bool IsStore, bool IsLoad);
    bool expandAtomicLoad(LoadInst *LI);
    bool expandAtomicLoadToLL(LoadInst *LI);
    bool expandAtomicLoadToCmpXchg(LoadInst *LI);
    bool expandAtomicStore(StoreInst *SI);
    bool expandAtomicRMW(AtomicRMWInst *AI);
    bool expandAtomicRMWToLLSC(AtomicRMWInst *AI);
    bool expandAtomicRMWToCmpXchg(AtomicRMWInst *AI);
    bool expandAtomicCmpXchg(AtomicCmpXchgInst *CI);
    bool isIdempotentRMW(AtomicRMWInst *AI);
    bool simplifyIdempotentRMW(AtomicRMWInst *AI);
  };
}

char AtomicExpand::ID = 0;
char &llvm::AtomicExpandID = AtomicExpand::ID;
INITIALIZE_TM_PASS(AtomicExpand, "atomic-expand",
    "Expand Atomic calls in terms of either load-linked & store-conditional or cmpxchg",
    false, false)

FunctionPass *llvm::createAtomicExpandPass(const TargetMachine *TM) {
  return new AtomicExpand(TM);
}

bool AtomicExpand::runOnFunction(Function &F) {
  if (!TM || !TM->getSubtargetImpl(F)->enableAtomicExpand())
    return false;
  TLI = TM->getSubtargetImpl(F)->getTargetLowering();

  SmallVector<Instruction *, 1> AtomicInsts;

  // Changing control-flow while iterating through it is a bad idea, so gather a
  // list of all atomic instructions before we start.
  for (inst_iterator I = inst_begin(F), E = inst_end(F); I != E; ++I) {
    if (I->isAtomic())
      AtomicInsts.push_back(&*I);
  }

  bool MadeChange = false;
  for (auto I : AtomicInsts) {
    auto LI = dyn_cast<LoadInst>(I);
    auto SI = dyn_cast<StoreInst>(I);
    auto RMWI = dyn_cast<AtomicRMWInst>(I);
    auto CASI = dyn_cast<AtomicCmpXchgInst>(I);
    assert((LI || SI || RMWI || CASI || isa<FenceInst>(I)) &&
           "Unknown atomic instruction");

    auto FenceOrdering = Monotonic;
    bool IsStore, IsLoad;
    if (TLI->getInsertFencesForAtomic()) {
      if (LI && isAtLeastAcquire(LI->getOrdering())) {
        FenceOrdering = LI->getOrdering();
        LI->setOrdering(Monotonic);
        IsStore = false;
        IsLoad = true;
      } else if (SI && isAtLeastRelease(SI->getOrdering())) {
        FenceOrdering = SI->getOrdering();
        SI->setOrdering(Monotonic);
        IsStore = true;
        IsLoad = false;
      } else if (RMWI && (isAtLeastRelease(RMWI->getOrdering()) ||
                          isAtLeastAcquire(RMWI->getOrdering()))) {
        FenceOrdering = RMWI->getOrdering();
        RMWI->setOrdering(Monotonic);
        IsStore = IsLoad = true;
      } else if (CASI && !TLI->hasLoadLinkedStoreConditional() &&
                 (isAtLeastRelease(CASI->getSuccessOrdering()) ||
                  isAtLeastAcquire(CASI->getSuccessOrdering()))) {
        // If a compare and swap is lowered to LL/SC, we can do smarter fence
        // insertion, with a stronger one on the success path than on the
        // failure path. As a result, fence insertion is directly done by
        // expandAtomicCmpXchg in that case.
        FenceOrdering = CASI->getSuccessOrdering();
        CASI->setSuccessOrdering(Monotonic);
        CASI->setFailureOrdering(Monotonic);
        IsStore = IsLoad = true;
      }

      if (FenceOrdering != Monotonic) {
        MadeChange |= bracketInstWithFences(I, FenceOrdering, IsStore, IsLoad);
      }
    }

    if (LI && TLI->shouldExpandAtomicLoadInIR(LI)) {
      MadeChange |= expandAtomicLoad(LI);
    } else if (SI && TLI->shouldExpandAtomicStoreInIR(SI)) {
      MadeChange |= expandAtomicStore(SI);
    } else if (RMWI) {
      // There are two different ways of expanding RMW instructions:
      // - into a load if it is idempotent
      // - into a Cmpxchg/LL-SC loop otherwise
      // we try them in that order.
      MadeChange |=
          (isIdempotentRMW(RMWI) && simplifyIdempotentRMW(RMWI)) ||
          (TLI->shouldExpandAtomicRMWInIR(RMWI) && expandAtomicRMW(RMWI));
    } else if (CASI && TLI->hasLoadLinkedStoreConditional()) {
      MadeChange |= expandAtomicCmpXchg(CASI);
    }
  }
  return MadeChange;
}

bool AtomicExpand::bracketInstWithFences(Instruction *I, AtomicOrdering Order,
                                         bool IsStore, bool IsLoad) {
  IRBuilder<> Builder(I);

  auto LeadingFence = TLI->emitLeadingFence(Builder, Order, IsStore, IsLoad);

  auto TrailingFence = TLI->emitTrailingFence(Builder, Order, IsStore, IsLoad);
  // The trailing fence is emitted before the instruction instead of after
  // because there is no easy way of setting Builder insertion point after
  // an instruction. So we must erase it from the BB, and insert it back
  // in the right place.
  // We have a guard here because not every atomic operation generates a
  // trailing fence.
  if (TrailingFence) {
    TrailingFence->removeFromParent();
    TrailingFence->insertAfter(I);
  }

  return (LeadingFence || TrailingFence);
}

bool AtomicExpand::expandAtomicLoad(LoadInst *LI) {
  if (TLI->hasLoadLinkedStoreConditional())
    return expandAtomicLoadToLL(LI);
  else
    return expandAtomicLoadToCmpXchg(LI);
}

bool AtomicExpand::expandAtomicLoadToLL(LoadInst *LI) {
  IRBuilder<> Builder(LI);

  // On some architectures, load-linked instructions are atomic for larger
  // sizes than normal loads. For example, the only 64-bit load guaranteed
  // to be single-copy atomic by ARM is an ldrexd (A3.5.3).
  Value *Val =
      TLI->emitLoadLinked(Builder, LI->getPointerOperand(), LI->getOrdering());

  LI->replaceAllUsesWith(Val);
  LI->eraseFromParent();

  return true;
}

bool AtomicExpand::expandAtomicLoadToCmpXchg(LoadInst *LI) {
  IRBuilder<> Builder(LI);
  AtomicOrdering Order = LI->getOrdering();
  Value *Addr = LI->getPointerOperand();
  Type *Ty = cast<PointerType>(Addr->getType())->getElementType();
  Constant *DummyVal = Constant::getNullValue(Ty);

  Value *Pair = Builder.CreateAtomicCmpXchg(
      Addr, DummyVal, DummyVal, Order,
      AtomicCmpXchgInst::getStrongestFailureOrdering(Order));
  Value *Loaded = Builder.CreateExtractValue(Pair, 0, "loaded");

  LI->replaceAllUsesWith(Loaded);
  LI->eraseFromParent();

  return true;
}

bool AtomicExpand::expandAtomicStore(StoreInst *SI) {
  // This function is only called on atomic stores that are too large to be
  // atomic if implemented as a native store. So we replace them by an
  // atomic swap, that can be implemented for example as a ldrex/strex on ARM
  // or lock cmpxchg8/16b on X86, as these are atomic for larger sizes.
  // It is the responsibility of the target to only return true in
  // shouldExpandAtomicRMW in cases where this is required and possible.
  IRBuilder<> Builder(SI);
  AtomicRMWInst *AI =
      Builder.CreateAtomicRMW(AtomicRMWInst::Xchg, SI->getPointerOperand(),
                              SI->getValueOperand(), SI->getOrdering());
  SI->eraseFromParent();

  // Now we have an appropriate swap instruction, lower it as usual.
  return expandAtomicRMW(AI);
}

bool AtomicExpand::expandAtomicRMW(AtomicRMWInst *AI) {
  if (TLI->hasLoadLinkedStoreConditional())
    return expandAtomicRMWToLLSC(AI);
  else
    return expandAtomicRMWToCmpXchg(AI);
}

/// Emit IR to implement the given atomicrmw operation on values in registers,
/// returning the new value.
static Value *performAtomicOp(AtomicRMWInst::BinOp Op, IRBuilder<> &Builder,
                              Value *Loaded, Value *Inc) {
  Value *NewVal;
  switch (Op) {
  case AtomicRMWInst::Xchg:
    return Inc;
  case AtomicRMWInst::Add:
    return Builder.CreateAdd(Loaded, Inc, "new");
  case AtomicRMWInst::Sub:
    return Builder.CreateSub(Loaded, Inc, "new");
  case AtomicRMWInst::And:
    return Builder.CreateAnd(Loaded, Inc, "new");
  case AtomicRMWInst::Nand:
    return Builder.CreateNot(Builder.CreateAnd(Loaded, Inc), "new");
  case AtomicRMWInst::Or:
    return Builder.CreateOr(Loaded, Inc, "new");
  case AtomicRMWInst::Xor:
    return Builder.CreateXor(Loaded, Inc, "new");
  case AtomicRMWInst::Max:
    NewVal = Builder.CreateICmpSGT(Loaded, Inc);
    return Builder.CreateSelect(NewVal, Loaded, Inc, "new");
  case AtomicRMWInst::Min:
    NewVal = Builder.CreateICmpSLE(Loaded, Inc);
    return Builder.CreateSelect(NewVal, Loaded, Inc, "new");
  case AtomicRMWInst::UMax:
    NewVal = Builder.CreateICmpUGT(Loaded, Inc);
    return Builder.CreateSelect(NewVal, Loaded, Inc, "new");
  case AtomicRMWInst::UMin:
    NewVal = Builder.CreateICmpULE(Loaded, Inc);
    return Builder.CreateSelect(NewVal, Loaded, Inc, "new");
  default:
    llvm_unreachable("Unknown atomic op");
  }
}

bool AtomicExpand::expandAtomicRMWToLLSC(AtomicRMWInst *AI) {
  AtomicOrdering MemOpOrder = AI->getOrdering();
  Value *Addr = AI->getPointerOperand();
  BasicBlock *BB = AI->getParent();
  Function *F = BB->getParent();
  LLVMContext &Ctx = F->getContext();

  // Given: atomicrmw some_op iN* %addr, iN %incr ordering
  //
  // The standard expansion we produce is:
  //     [...]
  //     fence?
  // atomicrmw.start:
  //     %loaded = @load.linked(%addr)
  //     %new = some_op iN %loaded, %incr
  //     %stored = @store_conditional(%new, %addr)
  //     %try_again = icmp i32 ne %stored, 0
  //     br i1 %try_again, label %loop, label %atomicrmw.end
  // atomicrmw.end:
  //     fence?
  //     [...]
  BasicBlock *ExitBB = BB->splitBasicBlock(AI, "atomicrmw.end");
  BasicBlock *LoopBB =  BasicBlock::Create(Ctx, "atomicrmw.start", F, ExitBB);

  // This grabs the DebugLoc from AI.
  IRBuilder<> Builder(AI);

  // The split call above "helpfully" added a branch at the end of BB (to the
  // wrong place), but we might want a fence too. It's easiest to just remove
  // the branch entirely.
  std::prev(BB->end())->eraseFromParent();
  Builder.SetInsertPoint(BB);
  Builder.CreateBr(LoopBB);

  // Start the main loop block now that we've taken care of the preliminaries.
  Builder.SetInsertPoint(LoopBB);
  Value *Loaded = TLI->emitLoadLinked(Builder, Addr, MemOpOrder);

  Value *NewVal =
      performAtomicOp(AI->getOperation(), Builder, Loaded, AI->getValOperand());

  Value *StoreSuccess =
      TLI->emitStoreConditional(Builder, NewVal, Addr, MemOpOrder);
  Value *TryAgain = Builder.CreateICmpNE(
      StoreSuccess, ConstantInt::get(IntegerType::get(Ctx, 32), 0), "tryagain");
  Builder.CreateCondBr(TryAgain, LoopBB, ExitBB);

  Builder.SetInsertPoint(ExitBB, ExitBB->begin());

  AI->replaceAllUsesWith(Loaded);
  AI->eraseFromParent();

  return true;
}

bool AtomicExpand::expandAtomicRMWToCmpXchg(AtomicRMWInst *AI) {
  AtomicOrdering MemOpOrder =
      AI->getOrdering() == Unordered ? Monotonic : AI->getOrdering();
  Value *Addr = AI->getPointerOperand();
  BasicBlock *BB = AI->getParent();
  Function *F = BB->getParent();
  LLVMContext &Ctx = F->getContext();

  // Given: atomicrmw some_op iN* %addr, iN %incr ordering
  //
  // The standard expansion we produce is:
  //     [...]
  //     %init_loaded = load atomic iN* %addr
  //     br label %loop
  // loop:
  //     %loaded = phi iN [ %init_loaded, %entry ], [ %new_loaded, %loop ]
  //     %new = some_op iN %loaded, %incr
  //     %pair = cmpxchg iN* %addr, iN %loaded, iN %new
  //     %new_loaded = extractvalue { iN, i1 } %pair, 0
  //     %success = extractvalue { iN, i1 } %pair, 1
  //     br i1 %success, label %atomicrmw.end, label %loop
  // atomicrmw.end:
  //     [...]
  BasicBlock *ExitBB = BB->splitBasicBlock(AI, "atomicrmw.end");
  BasicBlock *LoopBB = BasicBlock::Create(Ctx, "atomicrmw.start", F, ExitBB);

  // This grabs the DebugLoc from AI.
  IRBuilder<> Builder(AI);

  // The split call above "helpfully" added a branch at the end of BB (to the
  // wrong place), but we want a load. It's easiest to just remove
  // the branch entirely.
  std::prev(BB->end())->eraseFromParent();
  Builder.SetInsertPoint(BB);
  LoadInst *InitLoaded = Builder.CreateLoad(Addr);
  // Atomics require at least natural alignment.
  InitLoaded->setAlignment(AI->getType()->getPrimitiveSizeInBits());
  Builder.CreateBr(LoopBB);

  // Start the main loop block now that we've taken care of the preliminaries.
  Builder.SetInsertPoint(LoopBB);
  PHINode *Loaded = Builder.CreatePHI(AI->getType(), 2, "loaded");
  Loaded->addIncoming(InitLoaded, BB);

  Value *NewVal =
      performAtomicOp(AI->getOperation(), Builder, Loaded, AI->getValOperand());

  Value *Pair = Builder.CreateAtomicCmpXchg(
      Addr, Loaded, NewVal, MemOpOrder,
      AtomicCmpXchgInst::getStrongestFailureOrdering(MemOpOrder));
  Value *NewLoaded = Builder.CreateExtractValue(Pair, 0, "newloaded");
  Loaded->addIncoming(NewLoaded, LoopBB);

  Value *Success = Builder.CreateExtractValue(Pair, 1, "success");
  Builder.CreateCondBr(Success, ExitBB, LoopBB);

  Builder.SetInsertPoint(ExitBB, ExitBB->begin());

  AI->replaceAllUsesWith(NewLoaded);
  AI->eraseFromParent();

  return true;
}

bool AtomicExpand::expandAtomicCmpXchg(AtomicCmpXchgInst *CI) {
  AtomicOrdering SuccessOrder = CI->getSuccessOrdering();
  AtomicOrdering FailureOrder = CI->getFailureOrdering();
  Value *Addr = CI->getPointerOperand();
  BasicBlock *BB = CI->getParent();
  Function *F = BB->getParent();
  LLVMContext &Ctx = F->getContext();
  // If getInsertFencesForAtomic() returns true, then the target does not want
  // to deal with memory orders, and emitLeading/TrailingFence should take care
  // of everything. Otherwise, emitLeading/TrailingFence are no-op and we
  // should preserve the ordering.
  AtomicOrdering MemOpOrder =
      TLI->getInsertFencesForAtomic() ? Monotonic : SuccessOrder;

  // Given: cmpxchg some_op iN* %addr, iN %desired, iN %new success_ord fail_ord
  //
  // The full expansion we produce is:
  //     [...]
  //     fence?
  // cmpxchg.start:
  //     %loaded = @load.linked(%addr)
  //     %should_store = icmp eq %loaded, %desired
  //     br i1 %should_store, label %cmpxchg.trystore,
  //                          label %cmpxchg.failure
  // cmpxchg.trystore:
  //     %stored = @store_conditional(%new, %addr)
  //     %success = icmp eq i32 %stored, 0
  //     br i1 %success, label %cmpxchg.success, label %loop/%cmpxchg.failure
  // cmpxchg.success:
  //     fence?
  //     br label %cmpxchg.end
  // cmpxchg.failure:
  //     fence?
  //     br label %cmpxchg.end
  // cmpxchg.end:
  //     %success = phi i1 [true, %cmpxchg.success], [false, %cmpxchg.failure]
  //     %restmp = insertvalue { iN, i1 } undef, iN %loaded, 0
  //     %res = insertvalue { iN, i1 } %restmp, i1 %success, 1
  //     [...]
  BasicBlock *ExitBB = BB->splitBasicBlock(CI, "cmpxchg.end");
  auto FailureBB = BasicBlock::Create(Ctx, "cmpxchg.failure", F, ExitBB);
  auto SuccessBB = BasicBlock::Create(Ctx, "cmpxchg.success", F, FailureBB);
  auto TryStoreBB = BasicBlock::Create(Ctx, "cmpxchg.trystore", F, SuccessBB);
  auto LoopBB = BasicBlock::Create(Ctx, "cmpxchg.start", F, TryStoreBB);

  // This grabs the DebugLoc from CI
  IRBuilder<> Builder(CI);

  // The split call above "helpfully" added a branch at the end of BB (to the
  // wrong place), but we might want a fence too. It's easiest to just remove
  // the branch entirely.
  std::prev(BB->end())->eraseFromParent();
  Builder.SetInsertPoint(BB);
  TLI->emitLeadingFence(Builder, SuccessOrder, /*IsStore=*/true,
                        /*IsLoad=*/true);
  Builder.CreateBr(LoopBB);

  // Start the main loop block now that we've taken care of the preliminaries.
  Builder.SetInsertPoint(LoopBB);
  Value *Loaded = TLI->emitLoadLinked(Builder, Addr, MemOpOrder);
  Value *ShouldStore =
      Builder.CreateICmpEQ(Loaded, CI->getCompareOperand(), "should_store");

  // If the the cmpxchg doesn't actually need any ordering when it fails, we can
  // jump straight past that fence instruction (if it exists).
  Builder.CreateCondBr(ShouldStore, TryStoreBB, FailureBB);

  Builder.SetInsertPoint(TryStoreBB);
  Value *StoreSuccess = TLI->emitStoreConditional(
      Builder, CI->getNewValOperand(), Addr, MemOpOrder);
  StoreSuccess = Builder.CreateICmpEQ(
      StoreSuccess, ConstantInt::get(Type::getInt32Ty(Ctx), 0), "success");
  Builder.CreateCondBr(StoreSuccess, SuccessBB,
                       CI->isWeak() ? FailureBB : LoopBB);

  // Make sure later instructions don't get reordered with a fence if necessary.
  Builder.SetInsertPoint(SuccessBB);
  TLI->emitTrailingFence(Builder, SuccessOrder, /*IsStore=*/true,
                         /*IsLoad=*/true);
  Builder.CreateBr(ExitBB);

  Builder.SetInsertPoint(FailureBB);
  TLI->emitTrailingFence(Builder, FailureOrder, /*IsStore=*/true,
                         /*IsLoad=*/true);
  Builder.CreateBr(ExitBB);

  // Finally, we have control-flow based knowledge of whether the cmpxchg
  // succeeded or not. We expose this to later passes by converting any
  // subsequent "icmp eq/ne %loaded, %oldval" into a use of an appropriate PHI.

  // Setup the builder so we can create any PHIs we need.
  Builder.SetInsertPoint(ExitBB, ExitBB->begin());
  PHINode *Success = Builder.CreatePHI(Type::getInt1Ty(Ctx), 2);
  Success->addIncoming(ConstantInt::getTrue(Ctx), SuccessBB);
  Success->addIncoming(ConstantInt::getFalse(Ctx), FailureBB);

  // Look for any users of the cmpxchg that are just comparing the loaded value
  // against the desired one, and replace them with the CFG-derived version.
  SmallVector<ExtractValueInst *, 2> PrunedInsts;
  for (auto User : CI->users()) {
    ExtractValueInst *EV = dyn_cast<ExtractValueInst>(User);
    if (!EV)
      continue;

    assert(EV->getNumIndices() == 1 && EV->getIndices()[0] <= 1 &&
           "weird extraction from { iN, i1 }");

    if (EV->getIndices()[0] == 0)
      EV->replaceAllUsesWith(Loaded);
    else
      EV->replaceAllUsesWith(Success);

    PrunedInsts.push_back(EV);
  }

  // We can remove the instructions now we're no longer iterating through them.
  for (auto EV : PrunedInsts)
    EV->eraseFromParent();

  if (!CI->use_empty()) {
    // Some use of the full struct return that we don't understand has happened,
    // so we've got to reconstruct it properly.
    Value *Res;
    Res = Builder.CreateInsertValue(UndefValue::get(CI->getType()), Loaded, 0);
    Res = Builder.CreateInsertValue(Res, Success, 1);

    CI->replaceAllUsesWith(Res);
  }

  CI->eraseFromParent();
  return true;
}

bool AtomicExpand::isIdempotentRMW(AtomicRMWInst* RMWI) {
  auto C = dyn_cast<ConstantInt>(RMWI->getValOperand());
  if(!C)
    return false;

  AtomicRMWInst::BinOp Op = RMWI->getOperation();
  switch(Op) {
    case AtomicRMWInst::Add:
    case AtomicRMWInst::Sub:
    case AtomicRMWInst::Or:
    case AtomicRMWInst::Xor:
      return C->isZero();
    case AtomicRMWInst::And:
      return C->isMinusOne();
    // FIXME: we could also treat Min/Max/UMin/UMax by the INT_MIN/INT_MAX/...
    default:
      return false;
  }
}

bool AtomicExpand::simplifyIdempotentRMW(AtomicRMWInst* RMWI) {
  if (auto ResultingLoad = TLI->lowerIdempotentRMWIntoFencedLoad(RMWI)) {
    if (TLI->shouldExpandAtomicLoadInIR(ResultingLoad))
      expandAtomicLoad(ResultingLoad);
    return true;
  }
  return false;
}