BasicTargetTransformInfo.cpp revision 04b84c2f92b8c9cf863853eca33f47f9fbd80fd1 
1//===- BasicTargetTransformInfo.cpp - Basic target-independent TTI impl ---===//
2//
3//                     The LLVM Compiler Infrastructure
4//
5// This file is distributed under the University of Illinois Open Source
6// License. See LICENSE.TXT for details.
7//
8//===----------------------------------------------------------------------===//
9/// \file
10/// This file provides the implementation of a basic TargetTransformInfo pass
11/// predicated on the target abstractions present in the target independent
12/// code generator. It uses these (primarily TargetLowering) to model as much
13/// of the TTI query interface as possible. It is included by most targets so
14/// that they can specialize only a small subset of the query space.
15///
16//===----------------------------------------------------------------------===//
17
18#define DEBUG_TYPE "basictti"
19#include "llvm/CodeGen/Passes.h"
20#include "llvm/Analysis/TargetTransformInfo.h"
21#include "llvm/Target/TargetLowering.h"
22#include <utility>
23
24using namespace llvm;
25
26namespace {
27
28class BasicTTI : public ImmutablePass, public TargetTransformInfo {
29  const TargetMachine *TM;
30
31  /// Estimate the overhead of scalarizing an instruction. Insert and Extract
32  /// are set if the result needs to be inserted and/or extracted from vectors.
33  unsigned getScalarizationOverhead(Type *Ty, bool Insert, bool Extract) const;
34
35  const TargetLoweringBase *getTLI() const { return TM->getTargetLowering(); }
36
37public:
38  BasicTTI() : ImmutablePass(ID), TM(0) {
39    llvm_unreachable("This pass cannot be directly constructed");
40  }
41
42  BasicTTI(const TargetMachine *TM) : ImmutablePass(ID), TM(TM) {
43    initializeBasicTTIPass(*PassRegistry::getPassRegistry());
44  }
45
46  virtual void initializePass() {
47    pushTTIStack(this);
48  }
49
50  virtual void finalizePass() {
51    popTTIStack();
52  }
53
54  virtual void getAnalysisUsage(AnalysisUsage &AU) const {
55    TargetTransformInfo::getAnalysisUsage(AU);
56  }
57
58  /// Pass identification.
59  static char ID;
60
61  /// Provide necessary pointer adjustments for the two base classes.
62  virtual void *getAdjustedAnalysisPointer(const void *ID) {
63    if (ID == &TargetTransformInfo::ID)
64      return (TargetTransformInfo*)this;
65    return this;
66  }
67
68  /// \name Scalar TTI Implementations
69  /// @{
70
71  virtual bool isLegalAddImmediate(int64_t imm) const;
72  virtual bool isLegalICmpImmediate(int64_t imm) const;
73  virtual bool isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV,
74                                     int64_t BaseOffset, bool HasBaseReg,
75                                     int64_t Scale) const;
76  virtual int getScalingFactorCost(Type *Ty, GlobalValue *BaseGV,
77                                   int64_t BaseOffset, bool HasBaseReg,
78                                   int64_t Scale) const;
79  virtual bool isTruncateFree(Type *Ty1, Type *Ty2) const;
80  virtual bool isTypeLegal(Type *Ty) const;
81  virtual unsigned getJumpBufAlignment() const;
82  virtual unsigned getJumpBufSize() const;
83  virtual bool shouldBuildLookupTables() const;
84
85  /// @}
86
87  /// \name Vector TTI Implementations
88  /// @{
89
90  virtual unsigned getNumberOfRegisters(bool Vector) const;
91  virtual unsigned getMaximumUnrollFactor() const;
92  virtual unsigned getRegisterBitWidth(bool Vector) const;
93  virtual unsigned getArithmeticInstrCost(unsigned Opcode, Type *Ty,
94                                          OperandValueKind,
95                                          OperandValueKind) const;
96  virtual unsigned getShuffleCost(ShuffleKind Kind, Type *Tp,
97                                  int Index, Type *SubTp) const;
98  virtual unsigned getCastInstrCost(unsigned Opcode, Type *Dst,
99                                    Type *Src) const;
100  virtual unsigned getCFInstrCost(unsigned Opcode) const;
101  virtual unsigned getCmpSelInstrCost(unsigned Opcode, Type *ValTy,
102                                      Type *CondTy) const;
103  virtual unsigned getVectorInstrCost(unsigned Opcode, Type *Val,
104                                      unsigned Index) const;
105  virtual unsigned getMemoryOpCost(unsigned Opcode, Type *Src,
106                                   unsigned Alignment,
107                                   unsigned AddressSpace) const;
108  virtual unsigned getIntrinsicInstrCost(Intrinsic::ID, Type *RetTy,
109                                         ArrayRef<Type*> Tys) const;
110  virtual unsigned getNumberOfParts(Type *Tp) const;
111  virtual unsigned getAddressComputationCost(Type *Ty) const;
112
113  /// @}
114};
115
116}
117
118INITIALIZE_AG_PASS(BasicTTI, TargetTransformInfo, "basictti",
119                   "Target independent code generator's TTI", true, true, false)
120char BasicTTI::ID = 0;
121
122ImmutablePass *
123llvm::createBasicTargetTransformInfoPass(const TargetMachine *TM) {
124  return new BasicTTI(TM);
125}
126
127
128bool BasicTTI::isLegalAddImmediate(int64_t imm) const {
129  return getTLI()->isLegalAddImmediate(imm);
130}
131
132bool BasicTTI::isLegalICmpImmediate(int64_t imm) const {
133  return getTLI()->isLegalICmpImmediate(imm);
134}
135
136bool BasicTTI::isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV,
137                                     int64_t BaseOffset, bool HasBaseReg,
138                                     int64_t Scale) const {
139  TargetLoweringBase::AddrMode AM;
140  AM.BaseGV = BaseGV;
141  AM.BaseOffs = BaseOffset;
142  AM.HasBaseReg = HasBaseReg;
143  AM.Scale = Scale;
144  return getTLI()->isLegalAddressingMode(AM, Ty);
145}
146
147int BasicTTI::getScalingFactorCost(Type *Ty, GlobalValue *BaseGV,
148                                   int64_t BaseOffset, bool HasBaseReg,
149                                   int64_t Scale) const {
150  TargetLoweringBase::AddrMode AM;
151  AM.BaseGV = BaseGV;
152  AM.BaseOffs = BaseOffset;
153  AM.HasBaseReg = HasBaseReg;
154  AM.Scale = Scale;
155  return getTLI()->getScalingFactorCost(AM, Ty);
156}
157
158bool BasicTTI::isTruncateFree(Type *Ty1, Type *Ty2) const {
159  return getTLI()->isTruncateFree(Ty1, Ty2);
160}
161
162bool BasicTTI::isTypeLegal(Type *Ty) const {
163  EVT T = getTLI()->getValueType(Ty);
164  return getTLI()->isTypeLegal(T);
165}
166
167unsigned BasicTTI::getJumpBufAlignment() const {
168  return getTLI()->getJumpBufAlignment();
169}
170
171unsigned BasicTTI::getJumpBufSize() const {
172  return getTLI()->getJumpBufSize();
173}
174
175bool BasicTTI::shouldBuildLookupTables() const {
176  const TargetLoweringBase *TLI = getTLI();
177  return TLI->supportJumpTables() &&
178      (TLI->isOperationLegalOrCustom(ISD::BR_JT, MVT::Other) ||
179       TLI->isOperationLegalOrCustom(ISD::BRIND, MVT::Other));
180}
181
182//===----------------------------------------------------------------------===//
183//
184// Calls used by the vectorizers.
185//
186//===----------------------------------------------------------------------===//
187
188unsigned BasicTTI::getScalarizationOverhead(Type *Ty, bool Insert,
189                                            bool Extract) const {
190  assert (Ty->isVectorTy() && "Can only scalarize vectors");
191  unsigned Cost = 0;
192
193  for (int i = 0, e = Ty->getVectorNumElements(); i < e; ++i) {
194    if (Insert)
195      Cost += TopTTI->getVectorInstrCost(Instruction::InsertElement, Ty, i);
196    if (Extract)
197      Cost += TopTTI->getVectorInstrCost(Instruction::ExtractElement, Ty, i);
198  }
199
200  return Cost;
201}
202
203unsigned BasicTTI::getNumberOfRegisters(bool Vector) const {
204  return 1;
205}
206
207unsigned BasicTTI::getRegisterBitWidth(bool Vector) const {
208  return 32;
209}
210
211unsigned BasicTTI::getMaximumUnrollFactor() const {
212  return 1;
213}
214
215unsigned BasicTTI::getArithmeticInstrCost(unsigned Opcode, Type *Ty,
216                                          OperandValueKind,
217                                          OperandValueKind) const {
218  // Check if any of the operands are vector operands.
219  const TargetLoweringBase *TLI = getTLI();
220  int ISD = TLI->InstructionOpcodeToISD(Opcode);
221  assert(ISD && "Invalid opcode");
222
223  std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(Ty);
224
225  bool IsFloat = Ty->getScalarType()->isFloatingPointTy();
226  // Assume that floating point arithmetic operations cost twice as much as
227  // integer operations.
228  unsigned OpCost = (IsFloat ? 2 : 1);
229
230  if (TLI->isOperationLegalOrPromote(ISD, LT.second)) {
231    // The operation is legal. Assume it costs 1.
232    // If the type is split to multiple registers, assume that there is some
233    // overhead to this.
234    // TODO: Once we have extract/insert subvector cost we need to use them.
235    if (LT.first > 1)
236      return LT.first * 2 * OpCost;
237    return LT.first * 1 * OpCost;
238  }
239
240  if (!TLI->isOperationExpand(ISD, LT.second)) {
241    // If the operation is custom lowered then assume
242    // thare the code is twice as expensive.
243    return LT.first * 2 * OpCost;
244  }
245
246  // Else, assume that we need to scalarize this op.
247  if (Ty->isVectorTy()) {
248    unsigned Num = Ty->getVectorNumElements();
249    unsigned Cost = TopTTI->getArithmeticInstrCost(Opcode, Ty->getScalarType());
250    // return the cost of multiple scalar invocation plus the cost of inserting
251    // and extracting the values.
252    return getScalarizationOverhead(Ty, true, true) + Num * Cost;
253  }
254
255  // We don't know anything about this scalar instruction.
256  return OpCost;
257}
258
259unsigned BasicTTI::getShuffleCost(ShuffleKind Kind, Type *Tp, int Index,
260                                  Type *SubTp) const {
261  return 1;
262}
263
264unsigned BasicTTI::getCastInstrCost(unsigned Opcode, Type *Dst,
265                                    Type *Src) const {
266  const TargetLoweringBase *TLI = getTLI();
267  int ISD = TLI->InstructionOpcodeToISD(Opcode);
268  assert(ISD && "Invalid opcode");
269
270  std::pair<unsigned, MVT> SrcLT = TLI->getTypeLegalizationCost(Src);
271  std::pair<unsigned, MVT> DstLT = TLI->getTypeLegalizationCost(Dst);
272
273  // Check for NOOP conversions.
274  if (SrcLT.first == DstLT.first &&
275      SrcLT.second.getSizeInBits() == DstLT.second.getSizeInBits()) {
276
277      // Bitcast between types that are legalized to the same type are free.
278      if (Opcode == Instruction::BitCast || Opcode == Instruction::Trunc)
279        return 0;
280  }
281
282  if (Opcode == Instruction::Trunc &&
283      TLI->isTruncateFree(SrcLT.second, DstLT.second))
284    return 0;
285
286  if (Opcode == Instruction::ZExt &&
287      TLI->isZExtFree(SrcLT.second, DstLT.second))
288    return 0;
289
290  // If the cast is marked as legal (or promote) then assume low cost.
291  if (TLI->isOperationLegalOrPromote(ISD, DstLT.second))
292    return 1;
293
294  // Handle scalar conversions.
295  if (!Src->isVectorTy() && !Dst->isVectorTy()) {
296
297    // Scalar bitcasts are usually free.
298    if (Opcode == Instruction::BitCast)
299      return 0;
300
301    // Just check the op cost. If the operation is legal then assume it costs 1.
302    if (!TLI->isOperationExpand(ISD, DstLT.second))
303      return  1;
304
305    // Assume that illegal scalar instruction are expensive.
306    return 4;
307  }
308
309  // Check vector-to-vector casts.
310  if (Dst->isVectorTy() && Src->isVectorTy()) {
311
312    // If the cast is between same-sized registers, then the check is simple.
313    if (SrcLT.first == DstLT.first &&
314        SrcLT.second.getSizeInBits() == DstLT.second.getSizeInBits()) {
315
316      // Assume that Zext is done using AND.
317      if (Opcode == Instruction::ZExt)
318        return 1;
319
320      // Assume that sext is done using SHL and SRA.
321      if (Opcode == Instruction::SExt)
322        return 2;
323
324      // Just check the op cost. If the operation is legal then assume it costs
325      // 1 and multiply by the type-legalization overhead.
326      if (!TLI->isOperationExpand(ISD, DstLT.second))
327        return SrcLT.first * 1;
328    }
329
330    // If we are converting vectors and the operation is illegal, or
331    // if the vectors are legalized to different types, estimate the
332    // scalarization costs.
333    unsigned Num = Dst->getVectorNumElements();
334    unsigned Cost = TopTTI->getCastInstrCost(Opcode, Dst->getScalarType(),
335                                             Src->getScalarType());
336
337    // Return the cost of multiple scalar invocation plus the cost of
338    // inserting and extracting the values.
339    return getScalarizationOverhead(Dst, true, true) + Num * Cost;
340  }
341
342  // We already handled vector-to-vector and scalar-to-scalar conversions. This
343  // is where we handle bitcast between vectors and scalars. We need to assume
344  //  that the conversion is scalarized in one way or another.
345  if (Opcode == Instruction::BitCast)
346    // Illegal bitcasts are done by storing and loading from a stack slot.
347    return (Src->isVectorTy()? getScalarizationOverhead(Src, false, true):0) +
348           (Dst->isVectorTy()? getScalarizationOverhead(Dst, true, false):0);
349
350  llvm_unreachable("Unhandled cast");
351 }
352
353unsigned BasicTTI::getCFInstrCost(unsigned Opcode) const {
354  // Branches are assumed to be predicted.
355  return 0;
356}
357
358unsigned BasicTTI::getCmpSelInstrCost(unsigned Opcode, Type *ValTy,
359                                      Type *CondTy) const {
360  const TargetLoweringBase *TLI = getTLI();
361  int ISD = TLI->InstructionOpcodeToISD(Opcode);
362  assert(ISD && "Invalid opcode");
363
364  // Selects on vectors are actually vector selects.
365  if (ISD == ISD::SELECT) {
366    assert(CondTy && "CondTy must exist");
367    if (CondTy->isVectorTy())
368      ISD = ISD::VSELECT;
369  }
370
371  std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(ValTy);
372
373  if (!TLI->isOperationExpand(ISD, LT.second)) {
374    // The operation is legal. Assume it costs 1. Multiply
375    // by the type-legalization overhead.
376    return LT.first * 1;
377  }
378
379  // Otherwise, assume that the cast is scalarized.
380  if (ValTy->isVectorTy()) {
381    unsigned Num = ValTy->getVectorNumElements();
382    if (CondTy)
383      CondTy = CondTy->getScalarType();
384    unsigned Cost = TopTTI->getCmpSelInstrCost(Opcode, ValTy->getScalarType(),
385                                               CondTy);
386
387    // Return the cost of multiple scalar invocation plus the cost of inserting
388    // and extracting the values.
389    return getScalarizationOverhead(ValTy, true, false) + Num * Cost;
390  }
391
392  // Unknown scalar opcode.
393  return 1;
394}
395
396unsigned BasicTTI::getVectorInstrCost(unsigned Opcode, Type *Val,
397                                      unsigned Index) const {
398  return 1;
399}
400
401unsigned BasicTTI::getMemoryOpCost(unsigned Opcode, Type *Src,
402                                   unsigned Alignment,
403                                   unsigned AddressSpace) const {
404  assert(!Src->isVoidTy() && "Invalid type");
405  std::pair<unsigned, MVT> LT = getTLI()->getTypeLegalizationCost(Src);
406
407  // Assume that all loads of legal types cost 1.
408  return LT.first;
409}
410
411unsigned BasicTTI::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
412                                         ArrayRef<Type *> Tys) const {
413  unsigned ISD = 0;
414  switch (IID) {
415  default: {
416    // Assume that we need to scalarize this intrinsic.
417    unsigned ScalarizationCost = 0;
418    unsigned ScalarCalls = 1;
419    if (RetTy->isVectorTy()) {
420      ScalarizationCost = getScalarizationOverhead(RetTy, true, false);
421      ScalarCalls = std::max(ScalarCalls, RetTy->getVectorNumElements());
422    }
423    for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) {
424      if (Tys[i]->isVectorTy()) {
425        ScalarizationCost += getScalarizationOverhead(Tys[i], false, true);
426        ScalarCalls = std::max(ScalarCalls, RetTy->getVectorNumElements());
427      }
428    }
429
430    return ScalarCalls + ScalarizationCost;
431  }
432  // Look for intrinsics that can be lowered directly or turned into a scalar
433  // intrinsic call.
434  case Intrinsic::sqrt:    ISD = ISD::FSQRT;  break;
435  case Intrinsic::sin:     ISD = ISD::FSIN;   break;
436  case Intrinsic::cos:     ISD = ISD::FCOS;   break;
437  case Intrinsic::exp:     ISD = ISD::FEXP;   break;
438  case Intrinsic::exp2:    ISD = ISD::FEXP2;  break;
439  case Intrinsic::log:     ISD = ISD::FLOG;   break;
440  case Intrinsic::log10:   ISD = ISD::FLOG10; break;
441  case Intrinsic::log2:    ISD = ISD::FLOG2;  break;
442  case Intrinsic::fabs:    ISD = ISD::FABS;   break;
443  case Intrinsic::floor:   ISD = ISD::FFLOOR; break;
444  case Intrinsic::ceil:    ISD = ISD::FCEIL;  break;
445  case Intrinsic::trunc:   ISD = ISD::FTRUNC; break;
446  case Intrinsic::nearbyint:
447                           ISD = ISD::FNEARBYINT; break;
448  case Intrinsic::rint:    ISD = ISD::FRINT;  break;
449  case Intrinsic::pow:     ISD = ISD::FPOW;   break;
450  case Intrinsic::fma:     ISD = ISD::FMA;    break;
451  case Intrinsic::fmuladd: ISD = ISD::FMA;    break; // FIXME: mul + add?
452  }
453
454  const TargetLoweringBase *TLI = getTLI();
455  std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(RetTy);
456
457  if (TLI->isOperationLegalOrPromote(ISD, LT.second)) {
458    // The operation is legal. Assume it costs 1.
459    // If the type is split to multiple registers, assume that thre is some
460    // overhead to this.
461    // TODO: Once we have extract/insert subvector cost we need to use them.
462    if (LT.first > 1)
463      return LT.first * 2;
464    return LT.first * 1;
465  }
466
467  if (!TLI->isOperationExpand(ISD, LT.second)) {
468    // If the operation is custom lowered then assume
469    // thare the code is twice as expensive.
470    return LT.first * 2;
471  }
472
473  // Else, assume that we need to scalarize this intrinsic. For math builtins
474  // this will emit a costly libcall, adding call overhead and spills. Make it
475  // very expensive.
476  if (RetTy->isVectorTy()) {
477    unsigned Num = RetTy->getVectorNumElements();
478    unsigned Cost = TopTTI->getIntrinsicInstrCost(IID, RetTy->getScalarType(),
479                                                  Tys);
480    return 10 * Cost * Num;
481  }
482
483  // This is going to be turned into a library call, make it expensive.
484  return 10;
485}
486
487unsigned BasicTTI::getNumberOfParts(Type *Tp) const {
488  std::pair<unsigned, MVT> LT = getTLI()->getTypeLegalizationCost(Tp);
489  return LT.first;
490}
491
492unsigned BasicTTI::getAddressComputationCost(Type *Ty) const {
493  return 0;
494}
495