llvm.org GIT mirror llvm / 1a4e1d8 lib / Target / PowerPC / PPCLoopPreIncPrep.cpp
1a4e1d8

Tree @1a4e1d8 (Download .tar.gz)

PPCLoopPreIncPrep.cpp @1a4e1d8raw · 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
//===------ PPCLoopPreIncPrep.cpp - Loop Pre-Inc. AM Prep. Pass -----------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file implements a pass to prepare loops for pre-increment addressing
// modes. Additional PHIs are created for loop induction variables used by
// load/store instructions so that the pre-increment forms can be used.
// Generically, this means transforming loops like this:
//   for (int i = 0; i < n; ++i)
//     array[i] = c;
// to look like this:
//   T *p = array[-1];
//   for (int i = 0; i < n; ++i)
//     *++p = c;
//===----------------------------------------------------------------------===//

#define DEBUG_TYPE "ppc-loop-preinc-prep"
#include "PPC.h"
#include "PPCTargetMachine.h"
#include "llvm/ADT/DepthFirstIterator.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/SmallSet.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/CodeMetrics.h"
#include "llvm/Analysis/InstructionSimplify.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/ScalarEvolution.h"
#include "llvm/Analysis/ScalarEvolutionExpander.h"
#include "llvm/Analysis/ScalarEvolutionExpressions.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/IR/CFG.h"
#include "llvm/IR/Dominators.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/IntrinsicInst.h"
#include "llvm/IR/Module.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Transforms/Scalar.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
#include "llvm/Transforms/Utils/Local.h"
#include "llvm/Transforms/Utils/LoopUtils.h"
#include "llvm/Transforms/Utils/ValueMapper.h"
using namespace llvm;

// By default, we limit this to creating 16 PHIs (which is a little over half
// of the allocatable register set).
static cl::opt<unsigned> MaxVars("ppc-preinc-prep-max-vars",
                                 cl::Hidden, cl::init(16),
  cl::desc("Potential PHI threshold for PPC preinc loop prep"));

namespace llvm {
  void initializePPCLoopPreIncPrepPass(PassRegistry&);
}

namespace {

  class PPCLoopPreIncPrep : public FunctionPass {
  public:
    static char ID; // Pass ID, replacement for typeid
    PPCLoopPreIncPrep() : FunctionPass(ID), TM(nullptr) {
      initializePPCLoopPreIncPrepPass(*PassRegistry::getPassRegistry());
    }
    PPCLoopPreIncPrep(PPCTargetMachine &TM) : FunctionPass(ID), TM(&TM) {
      initializePPCLoopPreIncPrepPass(*PassRegistry::getPassRegistry());
    }

    void getAnalysisUsage(AnalysisUsage &AU) const override {
      AU.addPreserved<DominatorTreeWrapperPass>();
      AU.addRequired<LoopInfoWrapperPass>();
      AU.addPreserved<LoopInfoWrapperPass>();
      AU.addRequired<ScalarEvolutionWrapperPass>();
    }

    bool runOnFunction(Function &F) override;

    bool runOnLoop(Loop *L);
    void simplifyLoopLatch(Loop *L);
    bool rotateLoop(Loop *L);

  private:
    PPCTargetMachine *TM;
    DominatorTree *DT;
    LoopInfo *LI;
    ScalarEvolution *SE;
    bool PreserveLCSSA;
  };
}

char PPCLoopPreIncPrep::ID = 0;
static const char *name = "Prepare loop for pre-inc. addressing modes";
INITIALIZE_PASS_BEGIN(PPCLoopPreIncPrep, DEBUG_TYPE, name, false, false)
INITIALIZE_PASS_DEPENDENCY(LoopInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(ScalarEvolutionWrapperPass)
INITIALIZE_PASS_END(PPCLoopPreIncPrep, DEBUG_TYPE, name, false, false)

FunctionPass *llvm::createPPCLoopPreIncPrepPass(PPCTargetMachine &TM) {
  return new PPCLoopPreIncPrep(TM);
}

namespace {
  struct BucketElement {
    BucketElement(const SCEVConstant *O, Instruction *I) : Offset(O), Instr(I) {}
    BucketElement(Instruction *I) : Offset(nullptr), Instr(I) {}

    const SCEVConstant *Offset;
    Instruction *Instr;
  };

  struct Bucket {
    Bucket(const SCEV *B, Instruction *I) : BaseSCEV(B),
                                            Elements(1, BucketElement(I)) {}

    const SCEV *BaseSCEV;
    SmallVector<BucketElement, 16> Elements;
  };
}

static bool IsPtrInBounds(Value *BasePtr) {
  Value *StrippedBasePtr = BasePtr;
  while (BitCastInst *BC = dyn_cast<BitCastInst>(StrippedBasePtr))
    StrippedBasePtr = BC->getOperand(0);
  if (GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(StrippedBasePtr))
    return GEP->isInBounds();

  return false;
}

static Value *GetPointerOperand(Value *MemI) {
  if (LoadInst *LMemI = dyn_cast<LoadInst>(MemI)) {
    return LMemI->getPointerOperand();
  } else if (StoreInst *SMemI = dyn_cast<StoreInst>(MemI)) {
    return SMemI->getPointerOperand();
  } else if (IntrinsicInst *IMemI = dyn_cast<IntrinsicInst>(MemI)) {
    if (IMemI->getIntrinsicID() == Intrinsic::prefetch)
      return IMemI->getArgOperand(0);
  }

  return 0;
}

bool PPCLoopPreIncPrep::runOnFunction(Function &F) {
  if (skipFunction(F))
    return false;

  LI = &getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
  SE = &getAnalysis<ScalarEvolutionWrapperPass>().getSE();
  auto *DTWP = getAnalysisIfAvailable<DominatorTreeWrapperPass>();
  DT = DTWP ? &DTWP->getDomTree() : nullptr;
  PreserveLCSSA = mustPreserveAnalysisID(LCSSAID);

  bool MadeChange = false;

  for (auto I = LI->begin(), IE = LI->end(); I != IE; ++I)
    for (auto L = df_begin(*I), LE = df_end(*I); L != LE; ++L)
      MadeChange |= runOnLoop(*L);

  return MadeChange;
}

bool PPCLoopPreIncPrep::runOnLoop(Loop *L) {
  bool MadeChange = false;

  // Only prep. the inner-most loop
  if (!L->empty())
    return MadeChange;

  DEBUG(dbgs() << "PIP: Examining: " << *L << "\n");

  BasicBlock *Header = L->getHeader();

  const PPCSubtarget *ST =
    TM ? TM->getSubtargetImpl(*Header->getParent()) : nullptr;

  unsigned HeaderLoopPredCount =
    std::distance(pred_begin(Header), pred_end(Header));

  // Collect buckets of comparable addresses used by loads and stores.
  SmallVector<Bucket, 16> Buckets;
  for (Loop::block_iterator I = L->block_begin(), IE = L->block_end();
       I != IE; ++I) {
    for (BasicBlock::iterator J = (*I)->begin(), JE = (*I)->end();
        J != JE; ++J) {
      Value *PtrValue;
      Instruction *MemI;

      if (LoadInst *LMemI = dyn_cast<LoadInst>(J)) {
        MemI = LMemI;
        PtrValue = LMemI->getPointerOperand();
      } else if (StoreInst *SMemI = dyn_cast<StoreInst>(J)) {
        MemI = SMemI;
        PtrValue = SMemI->getPointerOperand();
      } else if (IntrinsicInst *IMemI = dyn_cast<IntrinsicInst>(J)) {
        if (IMemI->getIntrinsicID() == Intrinsic::prefetch) {
          MemI = IMemI;
          PtrValue = IMemI->getArgOperand(0);
        } else continue;
      } else continue;

      unsigned PtrAddrSpace = PtrValue->getType()->getPointerAddressSpace();
      if (PtrAddrSpace)
        continue;

      // There are no update forms for Altivec vector load/stores.
      if (ST && ST->hasAltivec() &&
          PtrValue->getType()->getPointerElementType()->isVectorTy())
        continue;

      if (L->isLoopInvariant(PtrValue))
        continue;

      const SCEV *LSCEV = SE->getSCEVAtScope(PtrValue, L);
      if (const SCEVAddRecExpr *LARSCEV = dyn_cast<SCEVAddRecExpr>(LSCEV)) {
        if (LARSCEV->getLoop() != L)
          continue;
      } else {
        continue;
      }

      bool FoundBucket = false;
      for (auto &B : Buckets) {
        const SCEV *Diff = SE->getMinusSCEV(LSCEV, B.BaseSCEV);
        if (const auto *CDiff = dyn_cast<SCEVConstant>(Diff)) {
          B.Elements.push_back(BucketElement(CDiff, MemI));
          FoundBucket = true;
          break;
        }
      }

      if (!FoundBucket) {
        if (Buckets.size() == MaxVars)
          return MadeChange;
        Buckets.push_back(Bucket(LSCEV, MemI));
      }
    }
  }

  if (Buckets.empty())
    return MadeChange;

  BasicBlock *LoopPredecessor = L->getLoopPredecessor();
  // If there is no loop predecessor, or the loop predecessor's terminator
  // returns a value (which might contribute to determining the loop's
  // iteration space), insert a new preheader for the loop.
  if (!LoopPredecessor ||
      !LoopPredecessor->getTerminator()->getType()->isVoidTy()) {
    LoopPredecessor = InsertPreheaderForLoop(L, DT, LI, PreserveLCSSA);
    if (LoopPredecessor)
      MadeChange = true;
  }
  if (!LoopPredecessor)
    return MadeChange;

  DEBUG(dbgs() << "PIP: Found " << Buckets.size() << " buckets\n");

  SmallSet<BasicBlock *, 16> BBChanged;
  for (unsigned i = 0, e = Buckets.size(); i != e; ++i) {
    // The base address of each bucket is transformed into a phi and the others
    // are rewritten as offsets of that variable.

    // We have a choice now of which instruction's memory operand we use as the
    // base for the generated PHI. Always picking the first instruction in each
    // bucket does not work well, specifically because that instruction might
    // be a prefetch (and there are no pre-increment dcbt variants). Otherwise,
    // the choice is somewhat arbitrary, because the backend will happily
    // generate direct offsets from both the pre-incremented and
    // post-incremented pointer values. Thus, we'll pick the first non-prefetch
    // instruction in each bucket, and adjust the recurrence and other offsets
    // accordingly. 
    for (int j = 0, je = Buckets[i].Elements.size(); j != je; ++j) {
      if (auto *II = dyn_cast<IntrinsicInst>(Buckets[i].Elements[j].Instr))
        if (II->getIntrinsicID() == Intrinsic::prefetch)
          continue;

      // If we'd otherwise pick the first element anyway, there's nothing to do.
      if (j == 0)
        break;

      // If our chosen element has no offset from the base pointer, there's
      // nothing to do.
      if (!Buckets[i].Elements[j].Offset ||
          Buckets[i].Elements[j].Offset->isZero())
        break;

      const SCEV *Offset = Buckets[i].Elements[j].Offset;
      Buckets[i].BaseSCEV = SE->getAddExpr(Buckets[i].BaseSCEV, Offset);
      for (auto &E : Buckets[i].Elements) {
        if (E.Offset)
          E.Offset = cast<SCEVConstant>(SE->getMinusSCEV(E.Offset, Offset));
        else
          E.Offset = cast<SCEVConstant>(SE->getNegativeSCEV(Offset));
      }

      std::swap(Buckets[i].Elements[j], Buckets[i].Elements[0]);
      break;
    }

    const SCEVAddRecExpr *BasePtrSCEV =
      cast<SCEVAddRecExpr>(Buckets[i].BaseSCEV);
    if (!BasePtrSCEV->isAffine())
      continue;

    DEBUG(dbgs() << "PIP: Transforming: " << *BasePtrSCEV << "\n");
    assert(BasePtrSCEV->getLoop() == L &&
           "AddRec for the wrong loop?");

    // The instruction corresponding to the Bucket's BaseSCEV must be the first
    // in the vector of elements.
    Instruction *MemI = Buckets[i].Elements.begin()->Instr;
    Value *BasePtr = GetPointerOperand(MemI);
    assert(BasePtr && "No pointer operand");

    Type *I8Ty = Type::getInt8Ty(MemI->getParent()->getContext());
    Type *I8PtrTy = Type::getInt8PtrTy(MemI->getParent()->getContext(),
      BasePtr->getType()->getPointerAddressSpace());

    const SCEV *BasePtrStartSCEV = BasePtrSCEV->getStart();
    if (!SE->isLoopInvariant(BasePtrStartSCEV, L))
      continue;

    const SCEVConstant *BasePtrIncSCEV =
      dyn_cast<SCEVConstant>(BasePtrSCEV->getStepRecurrence(*SE));
    if (!BasePtrIncSCEV)
      continue;
    BasePtrStartSCEV = SE->getMinusSCEV(BasePtrStartSCEV, BasePtrIncSCEV);
    if (!isSafeToExpand(BasePtrStartSCEV, *SE))
      continue;

    DEBUG(dbgs() << "PIP: New start is: " << *BasePtrStartSCEV << "\n");

    PHINode *NewPHI = PHINode::Create(I8PtrTy, HeaderLoopPredCount,
      MemI->hasName() ? MemI->getName() + ".phi" : "",
      Header->getFirstNonPHI());

    SCEVExpander SCEVE(*SE, Header->getModule()->getDataLayout(), "pistart");
    Value *BasePtrStart = SCEVE.expandCodeFor(BasePtrStartSCEV, I8PtrTy,
      LoopPredecessor->getTerminator());

    // Note that LoopPredecessor might occur in the predecessor list multiple
    // times, and we need to add it the right number of times.
    for (pred_iterator PI = pred_begin(Header), PE = pred_end(Header);
         PI != PE; ++PI) {
      if (*PI != LoopPredecessor)
        continue;

      NewPHI->addIncoming(BasePtrStart, LoopPredecessor);
    }

    Instruction *InsPoint = &*Header->getFirstInsertionPt();
    GetElementPtrInst *PtrInc = GetElementPtrInst::Create(
        I8Ty, NewPHI, BasePtrIncSCEV->getValue(),
        MemI->hasName() ? MemI->getName() + ".inc" : "", InsPoint);
    PtrInc->setIsInBounds(IsPtrInBounds(BasePtr));
    for (pred_iterator PI = pred_begin(Header), PE = pred_end(Header);
         PI != PE; ++PI) {
      if (*PI == LoopPredecessor)
        continue;

      NewPHI->addIncoming(PtrInc, *PI);
    }

    Instruction *NewBasePtr;
    if (PtrInc->getType() != BasePtr->getType())
      NewBasePtr = new BitCastInst(PtrInc, BasePtr->getType(),
        PtrInc->hasName() ? PtrInc->getName() + ".cast" : "", InsPoint);
    else
      NewBasePtr = PtrInc;

    if (Instruction *IDel = dyn_cast<Instruction>(BasePtr))
      BBChanged.insert(IDel->getParent());
    BasePtr->replaceAllUsesWith(NewBasePtr);
    RecursivelyDeleteTriviallyDeadInstructions(BasePtr);

    // Keep track of the replacement pointer values we've inserted so that we
    // don't generate more pointer values than necessary.
    SmallPtrSet<Value *, 16> NewPtrs;
    NewPtrs.insert( NewBasePtr);

    for (auto I = std::next(Buckets[i].Elements.begin()),
         IE = Buckets[i].Elements.end(); I != IE; ++I) {
      Value *Ptr = GetPointerOperand(I->Instr);
      assert(Ptr && "No pointer operand");
      if (NewPtrs.count(Ptr))
        continue;

      Instruction *RealNewPtr;
      if (!I->Offset || I->Offset->getValue()->isZero()) {
        RealNewPtr = NewBasePtr;
      } else {
        Instruction *PtrIP = dyn_cast<Instruction>(Ptr);
        if (PtrIP && isa<Instruction>(NewBasePtr) &&
            cast<Instruction>(NewBasePtr)->getParent() == PtrIP->getParent())
          PtrIP = 0;
        else if (isa<PHINode>(PtrIP))
          PtrIP = &*PtrIP->getParent()->getFirstInsertionPt();
        else if (!PtrIP)
          PtrIP = I->Instr;

        GetElementPtrInst *NewPtr = GetElementPtrInst::Create(
            I8Ty, PtrInc, I->Offset->getValue(),
            I->Instr->hasName() ? I->Instr->getName() + ".off" : "", PtrIP);
        if (!PtrIP)
          NewPtr->insertAfter(cast<Instruction>(PtrInc));
        NewPtr->setIsInBounds(IsPtrInBounds(Ptr));
        RealNewPtr = NewPtr;
      }

      if (Instruction *IDel = dyn_cast<Instruction>(Ptr))
        BBChanged.insert(IDel->getParent());

      Instruction *ReplNewPtr;
      if (Ptr->getType() != RealNewPtr->getType()) {
        ReplNewPtr = new BitCastInst(RealNewPtr, Ptr->getType(),
          Ptr->hasName() ? Ptr->getName() + ".cast" : "");
        ReplNewPtr->insertAfter(RealNewPtr);
      } else
        ReplNewPtr = RealNewPtr;

      Ptr->replaceAllUsesWith(ReplNewPtr);
      RecursivelyDeleteTriviallyDeadInstructions(Ptr);

      NewPtrs.insert(RealNewPtr);
    }

    MadeChange = true;
  }

  for (Loop::block_iterator I = L->block_begin(), IE = L->block_end();
       I != IE; ++I) {
    if (BBChanged.count(*I))
      DeleteDeadPHIs(*I);
  }

  return MadeChange;
}