1 //===- BasicTTIImpl.h -------------------------------------------*- C++ -*-===// 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 // 10 /// \file 11 /// This file provides a helper that implements much of the TTI interface in 12 /// terms of the target-independent code generator and TargetLowering 13 /// interfaces. 14 // 15 //===----------------------------------------------------------------------===// 16 17 #ifndef LLVM_CODEGEN_BASICTTIIMPL_H 18 #define LLVM_CODEGEN_BASICTTIIMPL_H 19 20 #include "llvm/ADT/APInt.h" 21 #include "llvm/ADT/ArrayRef.h" 22 #include "llvm/ADT/BitVector.h" 23 #include "llvm/ADT/SmallPtrSet.h" 24 #include "llvm/ADT/SmallVector.h" 25 #include "llvm/Analysis/LoopInfo.h" 26 #include "llvm/Analysis/TargetTransformInfo.h" 27 #include "llvm/Analysis/TargetTransformInfoImpl.h" 28 #include "llvm/CodeGen/ISDOpcodes.h" 29 #include "llvm/CodeGen/TargetLowering.h" 30 #include "llvm/CodeGen/TargetSubtargetInfo.h" 31 #include "llvm/CodeGen/ValueTypes.h" 32 #include "llvm/IR/BasicBlock.h" 33 #include "llvm/IR/CallSite.h" 34 #include "llvm/IR/Constant.h" 35 #include "llvm/IR/Constants.h" 36 #include "llvm/IR/DataLayout.h" 37 #include "llvm/IR/DerivedTypes.h" 38 #include "llvm/IR/InstrTypes.h" 39 #include "llvm/IR/Instruction.h" 40 #include "llvm/IR/Instructions.h" 41 #include "llvm/IR/Intrinsics.h" 42 #include "llvm/IR/Operator.h" 43 #include "llvm/IR/Type.h" 44 #include "llvm/IR/Value.h" 45 #include "llvm/MC/MCSchedule.h" 46 #include "llvm/Support/Casting.h" 47 #include "llvm/Support/CommandLine.h" 48 #include "llvm/Support/ErrorHandling.h" 49 #include "llvm/Support/MachineValueType.h" 50 #include "llvm/Support/MathExtras.h" 51 #include <algorithm> 52 #include <cassert> 53 #include <cstdint> 54 #include <limits> 55 #include <utility> 56 57 namespace llvm { 58 59 class Function; 60 class GlobalValue; 61 class LLVMContext; 62 class ScalarEvolution; 63 class SCEV; 64 class TargetMachine; 65 66 extern cl::opt<unsigned> PartialUnrollingThreshold; 67 68 /// Base class which can be used to help build a TTI implementation. 69 /// 70 /// This class provides as much implementation of the TTI interface as is 71 /// possible using the target independent parts of the code generator. 72 /// 73 /// In order to subclass it, your class must implement a getST() method to 74 /// return the subtarget, and a getTLI() method to return the target lowering. 75 /// We need these methods implemented in the derived class so that this class 76 /// doesn't have to duplicate storage for them. 77 template <typename T> 78 class BasicTTIImplBase : public TargetTransformInfoImplCRTPBase<T> { 79 private: 80 using BaseT = TargetTransformInfoImplCRTPBase<T>; 81 using TTI = TargetTransformInfo; 82 83 /// Estimate a cost of Broadcast as an extract and sequence of insert 84 /// operations. getBroadcastShuffleOverhead(Type * Ty)85 unsigned getBroadcastShuffleOverhead(Type *Ty) { 86 assert(Ty->isVectorTy() && "Can only shuffle vectors"); 87 unsigned Cost = 0; 88 // Broadcast cost is equal to the cost of extracting the zero'th element 89 // plus the cost of inserting it into every element of the result vector. 90 Cost += static_cast<T *>(this)->getVectorInstrCost( 91 Instruction::ExtractElement, Ty, 0); 92 93 for (int i = 0, e = Ty->getVectorNumElements(); i < e; ++i) { 94 Cost += static_cast<T *>(this)->getVectorInstrCost( 95 Instruction::InsertElement, Ty, i); 96 } 97 return Cost; 98 } 99 100 /// Estimate a cost of shuffle as a sequence of extract and insert 101 /// operations. getPermuteShuffleOverhead(Type * Ty)102 unsigned getPermuteShuffleOverhead(Type *Ty) { 103 assert(Ty->isVectorTy() && "Can only shuffle vectors"); 104 unsigned Cost = 0; 105 // Shuffle cost is equal to the cost of extracting element from its argument 106 // plus the cost of inserting them onto the result vector. 107 108 // e.g. <4 x float> has a mask of <0,5,2,7> i.e we need to extract from 109 // index 0 of first vector, index 1 of second vector,index 2 of first 110 // vector and finally index 3 of second vector and insert them at index 111 // <0,1,2,3> of result vector. 112 for (int i = 0, e = Ty->getVectorNumElements(); i < e; ++i) { 113 Cost += static_cast<T *>(this) 114 ->getVectorInstrCost(Instruction::InsertElement, Ty, i); 115 Cost += static_cast<T *>(this) 116 ->getVectorInstrCost(Instruction::ExtractElement, Ty, i); 117 } 118 return Cost; 119 } 120 121 /// Estimate a cost of subvector extraction as a sequence of extract and 122 /// insert operations. getExtractSubvectorOverhead(Type * Ty,int Index,Type * SubTy)123 unsigned getExtractSubvectorOverhead(Type *Ty, int Index, Type *SubTy) { 124 assert(Ty && Ty->isVectorTy() && SubTy && SubTy->isVectorTy() && 125 "Can only extract subvectors from vectors"); 126 int NumSubElts = SubTy->getVectorNumElements(); 127 assert((Index + NumSubElts) <= (int)Ty->getVectorNumElements() && 128 "SK_ExtractSubvector index out of range"); 129 130 unsigned Cost = 0; 131 // Subvector extraction cost is equal to the cost of extracting element from 132 // the source type plus the cost of inserting them into the result vector 133 // type. 134 for (int i = 0; i != NumSubElts; ++i) { 135 Cost += static_cast<T *>(this)->getVectorInstrCost( 136 Instruction::ExtractElement, Ty, i + Index); 137 Cost += static_cast<T *>(this)->getVectorInstrCost( 138 Instruction::InsertElement, SubTy, i); 139 } 140 return Cost; 141 } 142 143 /// Estimate a cost of subvector insertion as a sequence of extract and 144 /// insert operations. getInsertSubvectorOverhead(Type * Ty,int Index,Type * SubTy)145 unsigned getInsertSubvectorOverhead(Type *Ty, int Index, Type *SubTy) { 146 assert(Ty && Ty->isVectorTy() && SubTy && SubTy->isVectorTy() && 147 "Can only insert subvectors into vectors"); 148 int NumSubElts = SubTy->getVectorNumElements(); 149 assert((Index + NumSubElts) <= (int)Ty->getVectorNumElements() && 150 "SK_InsertSubvector index out of range"); 151 152 unsigned Cost = 0; 153 // Subvector insertion cost is equal to the cost of extracting element from 154 // the source type plus the cost of inserting them into the result vector 155 // type. 156 for (int i = 0; i != NumSubElts; ++i) { 157 Cost += static_cast<T *>(this)->getVectorInstrCost( 158 Instruction::ExtractElement, SubTy, i); 159 Cost += static_cast<T *>(this)->getVectorInstrCost( 160 Instruction::InsertElement, Ty, i + Index); 161 } 162 return Cost; 163 } 164 165 /// Local query method delegates up to T which *must* implement this! getST()166 const TargetSubtargetInfo *getST() const { 167 return static_cast<const T *>(this)->getST(); 168 } 169 170 /// Local query method delegates up to T which *must* implement this! getTLI()171 const TargetLoweringBase *getTLI() const { 172 return static_cast<const T *>(this)->getTLI(); 173 } 174 getISDIndexedMode(TTI::MemIndexedMode M)175 static ISD::MemIndexedMode getISDIndexedMode(TTI::MemIndexedMode M) { 176 switch (M) { 177 case TTI::MIM_Unindexed: 178 return ISD::UNINDEXED; 179 case TTI::MIM_PreInc: 180 return ISD::PRE_INC; 181 case TTI::MIM_PreDec: 182 return ISD::PRE_DEC; 183 case TTI::MIM_PostInc: 184 return ISD::POST_INC; 185 case TTI::MIM_PostDec: 186 return ISD::POST_DEC; 187 } 188 llvm_unreachable("Unexpected MemIndexedMode"); 189 } 190 191 protected: BasicTTIImplBase(const TargetMachine * TM,const DataLayout & DL)192 explicit BasicTTIImplBase(const TargetMachine *TM, const DataLayout &DL) 193 : BaseT(DL) {} 194 195 using TargetTransformInfoImplBase::DL; 196 197 public: 198 /// \name Scalar TTI Implementations 199 /// @{ allowsMisalignedMemoryAccesses(LLVMContext & Context,unsigned BitWidth,unsigned AddressSpace,unsigned Alignment,bool * Fast)200 bool allowsMisalignedMemoryAccesses(LLVMContext &Context, 201 unsigned BitWidth, unsigned AddressSpace, 202 unsigned Alignment, bool *Fast) const { 203 EVT E = EVT::getIntegerVT(Context, BitWidth); 204 return getTLI()->allowsMisalignedMemoryAccesses(E, AddressSpace, Alignment, Fast); 205 } 206 hasBranchDivergence()207 bool hasBranchDivergence() { return false; } 208 isSourceOfDivergence(const Value * V)209 bool isSourceOfDivergence(const Value *V) { return false; } 210 isAlwaysUniform(const Value * V)211 bool isAlwaysUniform(const Value *V) { return false; } 212 getFlatAddressSpace()213 unsigned getFlatAddressSpace() { 214 // Return an invalid address space. 215 return -1; 216 } 217 isLegalAddImmediate(int64_t imm)218 bool isLegalAddImmediate(int64_t imm) { 219 return getTLI()->isLegalAddImmediate(imm); 220 } 221 isLegalICmpImmediate(int64_t imm)222 bool isLegalICmpImmediate(int64_t imm) { 223 return getTLI()->isLegalICmpImmediate(imm); 224 } 225 226 bool isLegalAddressingMode(Type *Ty, GlobalValue *BaseGV, int64_t BaseOffset, 227 bool HasBaseReg, int64_t Scale, 228 unsigned AddrSpace, Instruction *I = nullptr) { 229 TargetLoweringBase::AddrMode AM; 230 AM.BaseGV = BaseGV; 231 AM.BaseOffs = BaseOffset; 232 AM.HasBaseReg = HasBaseReg; 233 AM.Scale = Scale; 234 return getTLI()->isLegalAddressingMode(DL, AM, Ty, AddrSpace, I); 235 } 236 isIndexedLoadLegal(TTI::MemIndexedMode M,Type * Ty,const DataLayout & DL)237 bool isIndexedLoadLegal(TTI::MemIndexedMode M, Type *Ty, 238 const DataLayout &DL) const { 239 EVT VT = getTLI()->getValueType(DL, Ty); 240 return getTLI()->isIndexedLoadLegal(getISDIndexedMode(M), VT); 241 } 242 isIndexedStoreLegal(TTI::MemIndexedMode M,Type * Ty,const DataLayout & DL)243 bool isIndexedStoreLegal(TTI::MemIndexedMode M, Type *Ty, 244 const DataLayout &DL) const { 245 EVT VT = getTLI()->getValueType(DL, Ty); 246 return getTLI()->isIndexedStoreLegal(getISDIndexedMode(M), VT); 247 } 248 isLSRCostLess(TTI::LSRCost C1,TTI::LSRCost C2)249 bool isLSRCostLess(TTI::LSRCost C1, TTI::LSRCost C2) { 250 return TargetTransformInfoImplBase::isLSRCostLess(C1, C2); 251 } 252 getScalingFactorCost(Type * Ty,GlobalValue * BaseGV,int64_t BaseOffset,bool HasBaseReg,int64_t Scale,unsigned AddrSpace)253 int getScalingFactorCost(Type *Ty, GlobalValue *BaseGV, int64_t BaseOffset, 254 bool HasBaseReg, int64_t Scale, unsigned AddrSpace) { 255 TargetLoweringBase::AddrMode AM; 256 AM.BaseGV = BaseGV; 257 AM.BaseOffs = BaseOffset; 258 AM.HasBaseReg = HasBaseReg; 259 AM.Scale = Scale; 260 return getTLI()->getScalingFactorCost(DL, AM, Ty, AddrSpace); 261 } 262 isTruncateFree(Type * Ty1,Type * Ty2)263 bool isTruncateFree(Type *Ty1, Type *Ty2) { 264 return getTLI()->isTruncateFree(Ty1, Ty2); 265 } 266 isProfitableToHoist(Instruction * I)267 bool isProfitableToHoist(Instruction *I) { 268 return getTLI()->isProfitableToHoist(I); 269 } 270 useAA()271 bool useAA() const { return getST()->useAA(); } 272 isTypeLegal(Type * Ty)273 bool isTypeLegal(Type *Ty) { 274 EVT VT = getTLI()->getValueType(DL, Ty); 275 return getTLI()->isTypeLegal(VT); 276 } 277 getGEPCost(Type * PointeeType,const Value * Ptr,ArrayRef<const Value * > Operands)278 int getGEPCost(Type *PointeeType, const Value *Ptr, 279 ArrayRef<const Value *> Operands) { 280 return BaseT::getGEPCost(PointeeType, Ptr, Operands); 281 } 282 getExtCost(const Instruction * I,const Value * Src)283 int getExtCost(const Instruction *I, const Value *Src) { 284 if (getTLI()->isExtFree(I)) 285 return TargetTransformInfo::TCC_Free; 286 287 if (isa<ZExtInst>(I) || isa<SExtInst>(I)) 288 if (const LoadInst *LI = dyn_cast<LoadInst>(Src)) 289 if (getTLI()->isExtLoad(LI, I, DL)) 290 return TargetTransformInfo::TCC_Free; 291 292 return TargetTransformInfo::TCC_Basic; 293 } 294 getIntrinsicCost(Intrinsic::ID IID,Type * RetTy,ArrayRef<const Value * > Arguments)295 unsigned getIntrinsicCost(Intrinsic::ID IID, Type *RetTy, 296 ArrayRef<const Value *> Arguments) { 297 return BaseT::getIntrinsicCost(IID, RetTy, Arguments); 298 } 299 getIntrinsicCost(Intrinsic::ID IID,Type * RetTy,ArrayRef<Type * > ParamTys)300 unsigned getIntrinsicCost(Intrinsic::ID IID, Type *RetTy, 301 ArrayRef<Type *> ParamTys) { 302 if (IID == Intrinsic::cttz) { 303 if (getTLI()->isCheapToSpeculateCttz()) 304 return TargetTransformInfo::TCC_Basic; 305 return TargetTransformInfo::TCC_Expensive; 306 } 307 308 if (IID == Intrinsic::ctlz) { 309 if (getTLI()->isCheapToSpeculateCtlz()) 310 return TargetTransformInfo::TCC_Basic; 311 return TargetTransformInfo::TCC_Expensive; 312 } 313 314 return BaseT::getIntrinsicCost(IID, RetTy, ParamTys); 315 } 316 getEstimatedNumberOfCaseClusters(const SwitchInst & SI,unsigned & JumpTableSize)317 unsigned getEstimatedNumberOfCaseClusters(const SwitchInst &SI, 318 unsigned &JumpTableSize) { 319 /// Try to find the estimated number of clusters. Note that the number of 320 /// clusters identified in this function could be different from the actural 321 /// numbers found in lowering. This function ignore switches that are 322 /// lowered with a mix of jump table / bit test / BTree. This function was 323 /// initially intended to be used when estimating the cost of switch in 324 /// inline cost heuristic, but it's a generic cost model to be used in other 325 /// places (e.g., in loop unrolling). 326 unsigned N = SI.getNumCases(); 327 const TargetLoweringBase *TLI = getTLI(); 328 const DataLayout &DL = this->getDataLayout(); 329 330 JumpTableSize = 0; 331 bool IsJTAllowed = TLI->areJTsAllowed(SI.getParent()->getParent()); 332 333 // Early exit if both a jump table and bit test are not allowed. 334 if (N < 1 || (!IsJTAllowed && DL.getIndexSizeInBits(0u) < N)) 335 return N; 336 337 APInt MaxCaseVal = SI.case_begin()->getCaseValue()->getValue(); 338 APInt MinCaseVal = MaxCaseVal; 339 for (auto CI : SI.cases()) { 340 const APInt &CaseVal = CI.getCaseValue()->getValue(); 341 if (CaseVal.sgt(MaxCaseVal)) 342 MaxCaseVal = CaseVal; 343 if (CaseVal.slt(MinCaseVal)) 344 MinCaseVal = CaseVal; 345 } 346 347 // Check if suitable for a bit test 348 if (N <= DL.getIndexSizeInBits(0u)) { 349 SmallPtrSet<const BasicBlock *, 4> Dests; 350 for (auto I : SI.cases()) 351 Dests.insert(I.getCaseSuccessor()); 352 353 if (TLI->isSuitableForBitTests(Dests.size(), N, MinCaseVal, MaxCaseVal, 354 DL)) 355 return 1; 356 } 357 358 // Check if suitable for a jump table. 359 if (IsJTAllowed) { 360 if (N < 2 || N < TLI->getMinimumJumpTableEntries()) 361 return N; 362 uint64_t Range = 363 (MaxCaseVal - MinCaseVal) 364 .getLimitedValue(std::numeric_limits<uint64_t>::max() - 1) + 1; 365 // Check whether a range of clusters is dense enough for a jump table 366 if (TLI->isSuitableForJumpTable(&SI, N, Range)) { 367 JumpTableSize = Range; 368 return 1; 369 } 370 } 371 return N; 372 } 373 getJumpBufAlignment()374 unsigned getJumpBufAlignment() { return getTLI()->getJumpBufAlignment(); } 375 getJumpBufSize()376 unsigned getJumpBufSize() { return getTLI()->getJumpBufSize(); } 377 shouldBuildLookupTables()378 bool shouldBuildLookupTables() { 379 const TargetLoweringBase *TLI = getTLI(); 380 return TLI->isOperationLegalOrCustom(ISD::BR_JT, MVT::Other) || 381 TLI->isOperationLegalOrCustom(ISD::BRIND, MVT::Other); 382 } 383 haveFastSqrt(Type * Ty)384 bool haveFastSqrt(Type *Ty) { 385 const TargetLoweringBase *TLI = getTLI(); 386 EVT VT = TLI->getValueType(DL, Ty); 387 return TLI->isTypeLegal(VT) && 388 TLI->isOperationLegalOrCustom(ISD::FSQRT, VT); 389 } 390 isFCmpOrdCheaperThanFCmpZero(Type * Ty)391 bool isFCmpOrdCheaperThanFCmpZero(Type *Ty) { 392 return true; 393 } 394 getFPOpCost(Type * Ty)395 unsigned getFPOpCost(Type *Ty) { 396 // Check whether FADD is available, as a proxy for floating-point in 397 // general. 398 const TargetLoweringBase *TLI = getTLI(); 399 EVT VT = TLI->getValueType(DL, Ty); 400 if (TLI->isOperationLegalOrCustomOrPromote(ISD::FADD, VT)) 401 return TargetTransformInfo::TCC_Basic; 402 return TargetTransformInfo::TCC_Expensive; 403 } 404 getOperationCost(unsigned Opcode,Type * Ty,Type * OpTy)405 unsigned getOperationCost(unsigned Opcode, Type *Ty, Type *OpTy) { 406 const TargetLoweringBase *TLI = getTLI(); 407 switch (Opcode) { 408 default: break; 409 case Instruction::Trunc: 410 if (TLI->isTruncateFree(OpTy, Ty)) 411 return TargetTransformInfo::TCC_Free; 412 return TargetTransformInfo::TCC_Basic; 413 case Instruction::ZExt: 414 if (TLI->isZExtFree(OpTy, Ty)) 415 return TargetTransformInfo::TCC_Free; 416 return TargetTransformInfo::TCC_Basic; 417 } 418 419 return BaseT::getOperationCost(Opcode, Ty, OpTy); 420 } 421 getInliningThresholdMultiplier()422 unsigned getInliningThresholdMultiplier() { return 1; } 423 getUnrollingPreferences(Loop * L,ScalarEvolution & SE,TTI::UnrollingPreferences & UP)424 void getUnrollingPreferences(Loop *L, ScalarEvolution &SE, 425 TTI::UnrollingPreferences &UP) { 426 // This unrolling functionality is target independent, but to provide some 427 // motivation for its intended use, for x86: 428 429 // According to the Intel 64 and IA-32 Architectures Optimization Reference 430 // Manual, Intel Core models and later have a loop stream detector (and 431 // associated uop queue) that can benefit from partial unrolling. 432 // The relevant requirements are: 433 // - The loop must have no more than 4 (8 for Nehalem and later) branches 434 // taken, and none of them may be calls. 435 // - The loop can have no more than 18 (28 for Nehalem and later) uops. 436 437 // According to the Software Optimization Guide for AMD Family 15h 438 // Processors, models 30h-4fh (Steamroller and later) have a loop predictor 439 // and loop buffer which can benefit from partial unrolling. 440 // The relevant requirements are: 441 // - The loop must have fewer than 16 branches 442 // - The loop must have less than 40 uops in all executed loop branches 443 444 // The number of taken branches in a loop is hard to estimate here, and 445 // benchmarking has revealed that it is better not to be conservative when 446 // estimating the branch count. As a result, we'll ignore the branch limits 447 // until someone finds a case where it matters in practice. 448 449 unsigned MaxOps; 450 const TargetSubtargetInfo *ST = getST(); 451 if (PartialUnrollingThreshold.getNumOccurrences() > 0) 452 MaxOps = PartialUnrollingThreshold; 453 else if (ST->getSchedModel().LoopMicroOpBufferSize > 0) 454 MaxOps = ST->getSchedModel().LoopMicroOpBufferSize; 455 else 456 return; 457 458 // Scan the loop: don't unroll loops with calls. 459 for (Loop::block_iterator I = L->block_begin(), E = L->block_end(); I != E; 460 ++I) { 461 BasicBlock *BB = *I; 462 463 for (BasicBlock::iterator J = BB->begin(), JE = BB->end(); J != JE; ++J) 464 if (isa<CallInst>(J) || isa<InvokeInst>(J)) { 465 ImmutableCallSite CS(&*J); 466 if (const Function *F = CS.getCalledFunction()) { 467 if (!static_cast<T *>(this)->isLoweredToCall(F)) 468 continue; 469 } 470 471 return; 472 } 473 } 474 475 // Enable runtime and partial unrolling up to the specified size. 476 // Enable using trip count upper bound to unroll loops. 477 UP.Partial = UP.Runtime = UP.UpperBound = true; 478 UP.PartialThreshold = MaxOps; 479 480 // Avoid unrolling when optimizing for size. 481 UP.OptSizeThreshold = 0; 482 UP.PartialOptSizeThreshold = 0; 483 484 // Set number of instructions optimized when "back edge" 485 // becomes "fall through" to default value of 2. 486 UP.BEInsns = 2; 487 } 488 getInstructionLatency(const Instruction * I)489 int getInstructionLatency(const Instruction *I) { 490 if (isa<LoadInst>(I)) 491 return getST()->getSchedModel().DefaultLoadLatency; 492 493 return BaseT::getInstructionLatency(I); 494 } 495 496 /// @} 497 498 /// \name Vector TTI Implementations 499 /// @{ 500 getNumberOfRegisters(bool Vector)501 unsigned getNumberOfRegisters(bool Vector) { return Vector ? 0 : 1; } 502 getRegisterBitWidth(bool Vector)503 unsigned getRegisterBitWidth(bool Vector) const { return 32; } 504 505 /// Estimate the overhead of scalarizing an instruction. Insert and Extract 506 /// are set if the result needs to be inserted and/or extracted from vectors. getScalarizationOverhead(Type * Ty,bool Insert,bool Extract)507 unsigned getScalarizationOverhead(Type *Ty, bool Insert, bool Extract) { 508 assert(Ty->isVectorTy() && "Can only scalarize vectors"); 509 unsigned Cost = 0; 510 511 for (int i = 0, e = Ty->getVectorNumElements(); i < e; ++i) { 512 if (Insert) 513 Cost += static_cast<T *>(this) 514 ->getVectorInstrCost(Instruction::InsertElement, Ty, i); 515 if (Extract) 516 Cost += static_cast<T *>(this) 517 ->getVectorInstrCost(Instruction::ExtractElement, Ty, i); 518 } 519 520 return Cost; 521 } 522 523 /// Estimate the overhead of scalarizing an instructions unique 524 /// non-constant operands. The types of the arguments are ordinarily 525 /// scalar, in which case the costs are multiplied with VF. getOperandsScalarizationOverhead(ArrayRef<const Value * > Args,unsigned VF)526 unsigned getOperandsScalarizationOverhead(ArrayRef<const Value *> Args, 527 unsigned VF) { 528 unsigned Cost = 0; 529 SmallPtrSet<const Value*, 4> UniqueOperands; 530 for (const Value *A : Args) { 531 if (!isa<Constant>(A) && UniqueOperands.insert(A).second) { 532 Type *VecTy = nullptr; 533 if (A->getType()->isVectorTy()) { 534 VecTy = A->getType(); 535 // If A is a vector operand, VF should be 1 or correspond to A. 536 assert((VF == 1 || VF == VecTy->getVectorNumElements()) && 537 "Vector argument does not match VF"); 538 } 539 else 540 VecTy = VectorType::get(A->getType(), VF); 541 542 Cost += getScalarizationOverhead(VecTy, false, true); 543 } 544 } 545 546 return Cost; 547 } 548 getScalarizationOverhead(Type * VecTy,ArrayRef<const Value * > Args)549 unsigned getScalarizationOverhead(Type *VecTy, ArrayRef<const Value *> Args) { 550 assert(VecTy->isVectorTy()); 551 552 unsigned Cost = 0; 553 554 Cost += getScalarizationOverhead(VecTy, true, false); 555 if (!Args.empty()) 556 Cost += getOperandsScalarizationOverhead(Args, 557 VecTy->getVectorNumElements()); 558 else 559 // When no information on arguments is provided, we add the cost 560 // associated with one argument as a heuristic. 561 Cost += getScalarizationOverhead(VecTy, false, true); 562 563 return Cost; 564 } 565 getMaxInterleaveFactor(unsigned VF)566 unsigned getMaxInterleaveFactor(unsigned VF) { return 1; } 567 568 unsigned getArithmeticInstrCost( 569 unsigned Opcode, Type *Ty, 570 TTI::OperandValueKind Opd1Info = TTI::OK_AnyValue, 571 TTI::OperandValueKind Opd2Info = TTI::OK_AnyValue, 572 TTI::OperandValueProperties Opd1PropInfo = TTI::OP_None, 573 TTI::OperandValueProperties Opd2PropInfo = TTI::OP_None, 574 ArrayRef<const Value *> Args = ArrayRef<const Value *>()) { 575 // Check if any of the operands are vector operands. 576 const TargetLoweringBase *TLI = getTLI(); 577 int ISD = TLI->InstructionOpcodeToISD(Opcode); 578 assert(ISD && "Invalid opcode"); 579 580 std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(DL, Ty); 581 582 bool IsFloat = Ty->isFPOrFPVectorTy(); 583 // Assume that floating point arithmetic operations cost twice as much as 584 // integer operations. 585 unsigned OpCost = (IsFloat ? 2 : 1); 586 587 if (TLI->isOperationLegalOrPromote(ISD, LT.second)) { 588 // The operation is legal. Assume it costs 1. 589 // TODO: Once we have extract/insert subvector cost we need to use them. 590 return LT.first * OpCost; 591 } 592 593 if (!TLI->isOperationExpand(ISD, LT.second)) { 594 // If the operation is custom lowered, then assume that the code is twice 595 // as expensive. 596 return LT.first * 2 * OpCost; 597 } 598 599 // Else, assume that we need to scalarize this op. 600 // TODO: If one of the types get legalized by splitting, handle this 601 // similarly to what getCastInstrCost() does. 602 if (Ty->isVectorTy()) { 603 unsigned Num = Ty->getVectorNumElements(); 604 unsigned Cost = static_cast<T *>(this) 605 ->getArithmeticInstrCost(Opcode, Ty->getScalarType()); 606 // Return the cost of multiple scalar invocation plus the cost of 607 // inserting and extracting the values. 608 return getScalarizationOverhead(Ty, Args) + Num * Cost; 609 } 610 611 // We don't know anything about this scalar instruction. 612 return OpCost; 613 } 614 getShuffleCost(TTI::ShuffleKind Kind,Type * Tp,int Index,Type * SubTp)615 unsigned getShuffleCost(TTI::ShuffleKind Kind, Type *Tp, int Index, 616 Type *SubTp) { 617 switch (Kind) { 618 case TTI::SK_Broadcast: 619 return getBroadcastShuffleOverhead(Tp); 620 case TTI::SK_Select: 621 case TTI::SK_Reverse: 622 case TTI::SK_Transpose: 623 case TTI::SK_PermuteSingleSrc: 624 case TTI::SK_PermuteTwoSrc: 625 return getPermuteShuffleOverhead(Tp); 626 case TTI::SK_ExtractSubvector: 627 return getExtractSubvectorOverhead(Tp, Index, SubTp); 628 case TTI::SK_InsertSubvector: 629 return getInsertSubvectorOverhead(Tp, Index, SubTp); 630 } 631 llvm_unreachable("Unknown TTI::ShuffleKind"); 632 } 633 634 unsigned getCastInstrCost(unsigned Opcode, Type *Dst, Type *Src, 635 const Instruction *I = nullptr) { 636 const TargetLoweringBase *TLI = getTLI(); 637 int ISD = TLI->InstructionOpcodeToISD(Opcode); 638 assert(ISD && "Invalid opcode"); 639 std::pair<unsigned, MVT> SrcLT = TLI->getTypeLegalizationCost(DL, Src); 640 std::pair<unsigned, MVT> DstLT = TLI->getTypeLegalizationCost(DL, Dst); 641 642 // Check for NOOP conversions. 643 if (SrcLT.first == DstLT.first && 644 SrcLT.second.getSizeInBits() == DstLT.second.getSizeInBits()) { 645 646 // Bitcast between types that are legalized to the same type are free. 647 if (Opcode == Instruction::BitCast || Opcode == Instruction::Trunc) 648 return 0; 649 } 650 651 if (Opcode == Instruction::Trunc && 652 TLI->isTruncateFree(SrcLT.second, DstLT.second)) 653 return 0; 654 655 if (Opcode == Instruction::ZExt && 656 TLI->isZExtFree(SrcLT.second, DstLT.second)) 657 return 0; 658 659 if (Opcode == Instruction::AddrSpaceCast && 660 TLI->isNoopAddrSpaceCast(Src->getPointerAddressSpace(), 661 Dst->getPointerAddressSpace())) 662 return 0; 663 664 // If this is a zext/sext of a load, return 0 if the corresponding 665 // extending load exists on target. 666 if ((Opcode == Instruction::ZExt || Opcode == Instruction::SExt) && 667 I && isa<LoadInst>(I->getOperand(0))) { 668 EVT ExtVT = EVT::getEVT(Dst); 669 EVT LoadVT = EVT::getEVT(Src); 670 unsigned LType = 671 ((Opcode == Instruction::ZExt) ? ISD::ZEXTLOAD : ISD::SEXTLOAD); 672 if (TLI->isLoadExtLegal(LType, ExtVT, LoadVT)) 673 return 0; 674 } 675 676 // If the cast is marked as legal (or promote) then assume low cost. 677 if (SrcLT.first == DstLT.first && 678 TLI->isOperationLegalOrPromote(ISD, DstLT.second)) 679 return 1; 680 681 // Handle scalar conversions. 682 if (!Src->isVectorTy() && !Dst->isVectorTy()) { 683 // Scalar bitcasts are usually free. 684 if (Opcode == Instruction::BitCast) 685 return 0; 686 687 // Just check the op cost. If the operation is legal then assume it costs 688 // 1. 689 if (!TLI->isOperationExpand(ISD, DstLT.second)) 690 return 1; 691 692 // Assume that illegal scalar instruction are expensive. 693 return 4; 694 } 695 696 // Check vector-to-vector casts. 697 if (Dst->isVectorTy() && Src->isVectorTy()) { 698 // If the cast is between same-sized registers, then the check is simple. 699 if (SrcLT.first == DstLT.first && 700 SrcLT.second.getSizeInBits() == DstLT.second.getSizeInBits()) { 701 702 // Assume that Zext is done using AND. 703 if (Opcode == Instruction::ZExt) 704 return 1; 705 706 // Assume that sext is done using SHL and SRA. 707 if (Opcode == Instruction::SExt) 708 return 2; 709 710 // Just check the op cost. If the operation is legal then assume it 711 // costs 712 // 1 and multiply by the type-legalization overhead. 713 if (!TLI->isOperationExpand(ISD, DstLT.second)) 714 return SrcLT.first * 1; 715 } 716 717 // If we are legalizing by splitting, query the concrete TTI for the cost 718 // of casting the original vector twice. We also need to factor in the 719 // cost of the split itself. Count that as 1, to be consistent with 720 // TLI->getTypeLegalizationCost(). 721 if ((TLI->getTypeAction(Src->getContext(), TLI->getValueType(DL, Src)) == 722 TargetLowering::TypeSplitVector) || 723 (TLI->getTypeAction(Dst->getContext(), TLI->getValueType(DL, Dst)) == 724 TargetLowering::TypeSplitVector)) { 725 Type *SplitDst = VectorType::get(Dst->getVectorElementType(), 726 Dst->getVectorNumElements() / 2); 727 Type *SplitSrc = VectorType::get(Src->getVectorElementType(), 728 Src->getVectorNumElements() / 2); 729 T *TTI = static_cast<T *>(this); 730 return TTI->getVectorSplitCost() + 731 (2 * TTI->getCastInstrCost(Opcode, SplitDst, SplitSrc, I)); 732 } 733 734 // In other cases where the source or destination are illegal, assume 735 // the operation will get scalarized. 736 unsigned Num = Dst->getVectorNumElements(); 737 unsigned Cost = static_cast<T *>(this)->getCastInstrCost( 738 Opcode, Dst->getScalarType(), Src->getScalarType(), I); 739 740 // Return the cost of multiple scalar invocation plus the cost of 741 // inserting and extracting the values. 742 return getScalarizationOverhead(Dst, true, true) + Num * Cost; 743 } 744 745 // We already handled vector-to-vector and scalar-to-scalar conversions. 746 // This 747 // is where we handle bitcast between vectors and scalars. We need to assume 748 // that the conversion is scalarized in one way or another. 749 if (Opcode == Instruction::BitCast) 750 // Illegal bitcasts are done by storing and loading from a stack slot. 751 return (Src->isVectorTy() ? getScalarizationOverhead(Src, false, true) 752 : 0) + 753 (Dst->isVectorTy() ? getScalarizationOverhead(Dst, true, false) 754 : 0); 755 756 llvm_unreachable("Unhandled cast"); 757 } 758 getExtractWithExtendCost(unsigned Opcode,Type * Dst,VectorType * VecTy,unsigned Index)759 unsigned getExtractWithExtendCost(unsigned Opcode, Type *Dst, 760 VectorType *VecTy, unsigned Index) { 761 return static_cast<T *>(this)->getVectorInstrCost( 762 Instruction::ExtractElement, VecTy, Index) + 763 static_cast<T *>(this)->getCastInstrCost(Opcode, Dst, 764 VecTy->getElementType()); 765 } 766 getCFInstrCost(unsigned Opcode)767 unsigned getCFInstrCost(unsigned Opcode) { 768 // Branches are assumed to be predicted. 769 return 0; 770 } 771 getCmpSelInstrCost(unsigned Opcode,Type * ValTy,Type * CondTy,const Instruction * I)772 unsigned getCmpSelInstrCost(unsigned Opcode, Type *ValTy, Type *CondTy, 773 const Instruction *I) { 774 const TargetLoweringBase *TLI = getTLI(); 775 int ISD = TLI->InstructionOpcodeToISD(Opcode); 776 assert(ISD && "Invalid opcode"); 777 778 // Selects on vectors are actually vector selects. 779 if (ISD == ISD::SELECT) { 780 assert(CondTy && "CondTy must exist"); 781 if (CondTy->isVectorTy()) 782 ISD = ISD::VSELECT; 783 } 784 std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(DL, ValTy); 785 786 if (!(ValTy->isVectorTy() && !LT.second.isVector()) && 787 !TLI->isOperationExpand(ISD, LT.second)) { 788 // The operation is legal. Assume it costs 1. Multiply 789 // by the type-legalization overhead. 790 return LT.first * 1; 791 } 792 793 // Otherwise, assume that the cast is scalarized. 794 // TODO: If one of the types get legalized by splitting, handle this 795 // similarly to what getCastInstrCost() does. 796 if (ValTy->isVectorTy()) { 797 unsigned Num = ValTy->getVectorNumElements(); 798 if (CondTy) 799 CondTy = CondTy->getScalarType(); 800 unsigned Cost = static_cast<T *>(this)->getCmpSelInstrCost( 801 Opcode, ValTy->getScalarType(), CondTy, I); 802 803 // Return the cost of multiple scalar invocation plus the cost of 804 // inserting and extracting the values. 805 return getScalarizationOverhead(ValTy, true, false) + Num * Cost; 806 } 807 808 // Unknown scalar opcode. 809 return 1; 810 } 811 getVectorInstrCost(unsigned Opcode,Type * Val,unsigned Index)812 unsigned getVectorInstrCost(unsigned Opcode, Type *Val, unsigned Index) { 813 std::pair<unsigned, MVT> LT = 814 getTLI()->getTypeLegalizationCost(DL, Val->getScalarType()); 815 816 return LT.first; 817 } 818 819 unsigned getMemoryOpCost(unsigned Opcode, Type *Src, unsigned Alignment, 820 unsigned AddressSpace, const Instruction *I = nullptr) { 821 assert(!Src->isVoidTy() && "Invalid type"); 822 std::pair<unsigned, MVT> LT = getTLI()->getTypeLegalizationCost(DL, Src); 823 824 // Assuming that all loads of legal types cost 1. 825 unsigned Cost = LT.first; 826 827 if (Src->isVectorTy() && 828 Src->getPrimitiveSizeInBits() < LT.second.getSizeInBits()) { 829 // This is a vector load that legalizes to a larger type than the vector 830 // itself. Unless the corresponding extending load or truncating store is 831 // legal, then this will scalarize. 832 TargetLowering::LegalizeAction LA = TargetLowering::Expand; 833 EVT MemVT = getTLI()->getValueType(DL, Src); 834 if (Opcode == Instruction::Store) 835 LA = getTLI()->getTruncStoreAction(LT.second, MemVT); 836 else 837 LA = getTLI()->getLoadExtAction(ISD::EXTLOAD, LT.second, MemVT); 838 839 if (LA != TargetLowering::Legal && LA != TargetLowering::Custom) { 840 // This is a vector load/store for some illegal type that is scalarized. 841 // We must account for the cost of building or decomposing the vector. 842 Cost += getScalarizationOverhead(Src, Opcode != Instruction::Store, 843 Opcode == Instruction::Store); 844 } 845 } 846 847 return Cost; 848 } 849 850 unsigned getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy, 851 unsigned Factor, 852 ArrayRef<unsigned> Indices, 853 unsigned Alignment, unsigned AddressSpace, 854 bool UseMaskForCond = false, 855 bool UseMaskForGaps = false) { 856 VectorType *VT = dyn_cast<VectorType>(VecTy); 857 assert(VT && "Expect a vector type for interleaved memory op"); 858 859 unsigned NumElts = VT->getNumElements(); 860 assert(Factor > 1 && NumElts % Factor == 0 && "Invalid interleave factor"); 861 862 unsigned NumSubElts = NumElts / Factor; 863 VectorType *SubVT = VectorType::get(VT->getElementType(), NumSubElts); 864 865 // Firstly, the cost of load/store operation. 866 unsigned Cost; 867 if (UseMaskForCond || UseMaskForGaps) 868 Cost = static_cast<T *>(this)->getMaskedMemoryOpCost( 869 Opcode, VecTy, Alignment, AddressSpace); 870 else 871 Cost = static_cast<T *>(this)->getMemoryOpCost(Opcode, VecTy, Alignment, 872 AddressSpace); 873 874 // Legalize the vector type, and get the legalized and unlegalized type 875 // sizes. 876 MVT VecTyLT = getTLI()->getTypeLegalizationCost(DL, VecTy).second; 877 unsigned VecTySize = 878 static_cast<T *>(this)->getDataLayout().getTypeStoreSize(VecTy); 879 unsigned VecTyLTSize = VecTyLT.getStoreSize(); 880 881 // Return the ceiling of dividing A by B. 882 auto ceil = [](unsigned A, unsigned B) { return (A + B - 1) / B; }; 883 884 // Scale the cost of the memory operation by the fraction of legalized 885 // instructions that will actually be used. We shouldn't account for the 886 // cost of dead instructions since they will be removed. 887 // 888 // E.g., An interleaved load of factor 8: 889 // %vec = load <16 x i64>, <16 x i64>* %ptr 890 // %v0 = shufflevector %vec, undef, <0, 8> 891 // 892 // If <16 x i64> is legalized to 8 v2i64 loads, only 2 of the loads will be 893 // used (those corresponding to elements [0:1] and [8:9] of the unlegalized 894 // type). The other loads are unused. 895 // 896 // We only scale the cost of loads since interleaved store groups aren't 897 // allowed to have gaps. 898 if (Opcode == Instruction::Load && VecTySize > VecTyLTSize) { 899 // The number of loads of a legal type it will take to represent a load 900 // of the unlegalized vector type. 901 unsigned NumLegalInsts = ceil(VecTySize, VecTyLTSize); 902 903 // The number of elements of the unlegalized type that correspond to a 904 // single legal instruction. 905 unsigned NumEltsPerLegalInst = ceil(NumElts, NumLegalInsts); 906 907 // Determine which legal instructions will be used. 908 BitVector UsedInsts(NumLegalInsts, false); 909 for (unsigned Index : Indices) 910 for (unsigned Elt = 0; Elt < NumSubElts; ++Elt) 911 UsedInsts.set((Index + Elt * Factor) / NumEltsPerLegalInst); 912 913 // Scale the cost of the load by the fraction of legal instructions that 914 // will be used. 915 Cost *= UsedInsts.count() / NumLegalInsts; 916 } 917 918 // Then plus the cost of interleave operation. 919 if (Opcode == Instruction::Load) { 920 // The interleave cost is similar to extract sub vectors' elements 921 // from the wide vector, and insert them into sub vectors. 922 // 923 // E.g. An interleaved load of factor 2 (with one member of index 0): 924 // %vec = load <8 x i32>, <8 x i32>* %ptr 925 // %v0 = shuffle %vec, undef, <0, 2, 4, 6> ; Index 0 926 // The cost is estimated as extract elements at 0, 2, 4, 6 from the 927 // <8 x i32> vector and insert them into a <4 x i32> vector. 928 929 assert(Indices.size() <= Factor && 930 "Interleaved memory op has too many members"); 931 932 for (unsigned Index : Indices) { 933 assert(Index < Factor && "Invalid index for interleaved memory op"); 934 935 // Extract elements from loaded vector for each sub vector. 936 for (unsigned i = 0; i < NumSubElts; i++) 937 Cost += static_cast<T *>(this)->getVectorInstrCost( 938 Instruction::ExtractElement, VT, Index + i * Factor); 939 } 940 941 unsigned InsSubCost = 0; 942 for (unsigned i = 0; i < NumSubElts; i++) 943 InsSubCost += static_cast<T *>(this)->getVectorInstrCost( 944 Instruction::InsertElement, SubVT, i); 945 946 Cost += Indices.size() * InsSubCost; 947 } else { 948 // The interleave cost is extract all elements from sub vectors, and 949 // insert them into the wide vector. 950 // 951 // E.g. An interleaved store of factor 2: 952 // %v0_v1 = shuffle %v0, %v1, <0, 4, 1, 5, 2, 6, 3, 7> 953 // store <8 x i32> %interleaved.vec, <8 x i32>* %ptr 954 // The cost is estimated as extract all elements from both <4 x i32> 955 // vectors and insert into the <8 x i32> vector. 956 957 unsigned ExtSubCost = 0; 958 for (unsigned i = 0; i < NumSubElts; i++) 959 ExtSubCost += static_cast<T *>(this)->getVectorInstrCost( 960 Instruction::ExtractElement, SubVT, i); 961 Cost += ExtSubCost * Factor; 962 963 for (unsigned i = 0; i < NumElts; i++) 964 Cost += static_cast<T *>(this) 965 ->getVectorInstrCost(Instruction::InsertElement, VT, i); 966 } 967 968 if (!UseMaskForCond) 969 return Cost; 970 971 Type *I8Type = Type::getInt8Ty(VT->getContext()); 972 VectorType *MaskVT = VectorType::get(I8Type, NumElts); 973 SubVT = VectorType::get(I8Type, NumSubElts); 974 975 // The Mask shuffling cost is extract all the elements of the Mask 976 // and insert each of them Factor times into the wide vector: 977 // 978 // E.g. an interleaved group with factor 3: 979 // %mask = icmp ult <8 x i32> %vec1, %vec2 980 // %interleaved.mask = shufflevector <8 x i1> %mask, <8 x i1> undef, 981 // <24 x i32> <0,0,0,1,1,1,2,2,2,3,3,3,4,4,4,5,5,5,6,6,6,7,7,7> 982 // The cost is estimated as extract all mask elements from the <8xi1> mask 983 // vector and insert them factor times into the <24xi1> shuffled mask 984 // vector. 985 for (unsigned i = 0; i < NumSubElts; i++) 986 Cost += static_cast<T *>(this)->getVectorInstrCost( 987 Instruction::ExtractElement, SubVT, i); 988 989 for (unsigned i = 0; i < NumElts; i++) 990 Cost += static_cast<T *>(this)->getVectorInstrCost( 991 Instruction::InsertElement, MaskVT, i); 992 993 // The Gaps mask is invariant and created outside the loop, therefore the 994 // cost of creating it is not accounted for here. However if we have both 995 // a MaskForGaps and some other mask that guards the execution of the 996 // memory access, we need to account for the cost of And-ing the two masks 997 // inside the loop. 998 if (UseMaskForGaps) 999 Cost += static_cast<T *>(this)->getArithmeticInstrCost( 1000 BinaryOperator::And, MaskVT); 1001 1002 return Cost; 1003 } 1004 1005 /// Get intrinsic cost based on arguments. 1006 unsigned getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy, 1007 ArrayRef<Value *> Args, FastMathFlags FMF, 1008 unsigned VF = 1) { 1009 unsigned RetVF = (RetTy->isVectorTy() ? RetTy->getVectorNumElements() : 1); 1010 assert((RetVF == 1 || VF == 1) && "VF > 1 and RetVF is a vector type"); 1011 auto *ConcreteTTI = static_cast<T *>(this); 1012 1013 switch (IID) { 1014 default: { 1015 // Assume that we need to scalarize this intrinsic. 1016 SmallVector<Type *, 4> Types; 1017 for (Value *Op : Args) { 1018 Type *OpTy = Op->getType(); 1019 assert(VF == 1 || !OpTy->isVectorTy()); 1020 Types.push_back(VF == 1 ? OpTy : VectorType::get(OpTy, VF)); 1021 } 1022 1023 if (VF > 1 && !RetTy->isVoidTy()) 1024 RetTy = VectorType::get(RetTy, VF); 1025 1026 // Compute the scalarization overhead based on Args for a vector 1027 // intrinsic. A vectorizer will pass a scalar RetTy and VF > 1, while 1028 // CostModel will pass a vector RetTy and VF is 1. 1029 unsigned ScalarizationCost = std::numeric_limits<unsigned>::max(); 1030 if (RetVF > 1 || VF > 1) { 1031 ScalarizationCost = 0; 1032 if (!RetTy->isVoidTy()) 1033 ScalarizationCost += getScalarizationOverhead(RetTy, true, false); 1034 ScalarizationCost += getOperandsScalarizationOverhead(Args, VF); 1035 } 1036 1037 return ConcreteTTI->getIntrinsicInstrCost(IID, RetTy, Types, FMF, 1038 ScalarizationCost); 1039 } 1040 case Intrinsic::masked_scatter: { 1041 assert(VF == 1 && "Can't vectorize types here."); 1042 Value *Mask = Args[3]; 1043 bool VarMask = !isa<Constant>(Mask); 1044 unsigned Alignment = cast<ConstantInt>(Args[2])->getZExtValue(); 1045 return ConcreteTTI->getGatherScatterOpCost( 1046 Instruction::Store, Args[0]->getType(), Args[1], VarMask, Alignment); 1047 } 1048 case Intrinsic::masked_gather: { 1049 assert(VF == 1 && "Can't vectorize types here."); 1050 Value *Mask = Args[2]; 1051 bool VarMask = !isa<Constant>(Mask); 1052 unsigned Alignment = cast<ConstantInt>(Args[1])->getZExtValue(); 1053 return ConcreteTTI->getGatherScatterOpCost(Instruction::Load, RetTy, 1054 Args[0], VarMask, Alignment); 1055 } 1056 case Intrinsic::experimental_vector_reduce_add: 1057 case Intrinsic::experimental_vector_reduce_mul: 1058 case Intrinsic::experimental_vector_reduce_and: 1059 case Intrinsic::experimental_vector_reduce_or: 1060 case Intrinsic::experimental_vector_reduce_xor: 1061 case Intrinsic::experimental_vector_reduce_fadd: 1062 case Intrinsic::experimental_vector_reduce_fmul: 1063 case Intrinsic::experimental_vector_reduce_smax: 1064 case Intrinsic::experimental_vector_reduce_smin: 1065 case Intrinsic::experimental_vector_reduce_fmax: 1066 case Intrinsic::experimental_vector_reduce_fmin: 1067 case Intrinsic::experimental_vector_reduce_umax: 1068 case Intrinsic::experimental_vector_reduce_umin: 1069 return getIntrinsicInstrCost(IID, RetTy, Args[0]->getType(), FMF); 1070 case Intrinsic::fshl: 1071 case Intrinsic::fshr: { 1072 Value *X = Args[0]; 1073 Value *Y = Args[1]; 1074 Value *Z = Args[2]; 1075 TTI::OperandValueProperties OpPropsX, OpPropsY, OpPropsZ, OpPropsBW; 1076 TTI::OperandValueKind OpKindX = TTI::getOperandInfo(X, OpPropsX); 1077 TTI::OperandValueKind OpKindY = TTI::getOperandInfo(Y, OpPropsY); 1078 TTI::OperandValueKind OpKindZ = TTI::getOperandInfo(Z, OpPropsZ); 1079 TTI::OperandValueKind OpKindBW = TTI::OK_UniformConstantValue; 1080 OpPropsBW = isPowerOf2_32(RetTy->getScalarSizeInBits()) ? TTI::OP_PowerOf2 1081 : TTI::OP_None; 1082 // fshl: (X << (Z % BW)) | (Y >> (BW - (Z % BW))) 1083 // fshr: (X << (BW - (Z % BW))) | (Y >> (Z % BW)) 1084 unsigned Cost = 0; 1085 Cost += ConcreteTTI->getArithmeticInstrCost(BinaryOperator::Or, RetTy); 1086 Cost += ConcreteTTI->getArithmeticInstrCost(BinaryOperator::Sub, RetTy); 1087 Cost += ConcreteTTI->getArithmeticInstrCost(BinaryOperator::Shl, RetTy, 1088 OpKindX, OpKindZ, OpPropsX); 1089 Cost += ConcreteTTI->getArithmeticInstrCost(BinaryOperator::LShr, RetTy, 1090 OpKindY, OpKindZ, OpPropsY); 1091 // Non-constant shift amounts requires a modulo. 1092 if (OpKindZ != TTI::OK_UniformConstantValue && 1093 OpKindZ != TTI::OK_NonUniformConstantValue) 1094 Cost += ConcreteTTI->getArithmeticInstrCost(BinaryOperator::URem, RetTy, 1095 OpKindZ, OpKindBW, OpPropsZ, 1096 OpPropsBW); 1097 // For non-rotates (X != Y) we must add shift-by-zero handling costs. 1098 if (X != Y) { 1099 Type *CondTy = Type::getInt1Ty(RetTy->getContext()); 1100 if (RetVF > 1) 1101 CondTy = VectorType::get(CondTy, RetVF); 1102 Cost += ConcreteTTI->getCmpSelInstrCost(BinaryOperator::ICmp, RetTy, 1103 CondTy, nullptr); 1104 Cost += ConcreteTTI->getCmpSelInstrCost(BinaryOperator::Select, RetTy, 1105 CondTy, nullptr); 1106 } 1107 return Cost; 1108 } 1109 } 1110 } 1111 1112 /// Get intrinsic cost based on argument types. 1113 /// If ScalarizationCostPassed is std::numeric_limits<unsigned>::max(), the 1114 /// cost of scalarizing the arguments and the return value will be computed 1115 /// based on types. 1116 unsigned getIntrinsicInstrCost( 1117 Intrinsic::ID IID, Type *RetTy, ArrayRef<Type *> Tys, FastMathFlags FMF, 1118 unsigned ScalarizationCostPassed = std::numeric_limits<unsigned>::max()) { 1119 SmallVector<unsigned, 2> ISDs; 1120 unsigned SingleCallCost = 10; // Library call cost. Make it expensive. 1121 switch (IID) { 1122 default: { 1123 // Assume that we need to scalarize this intrinsic. 1124 unsigned ScalarizationCost = ScalarizationCostPassed; 1125 unsigned ScalarCalls = 1; 1126 Type *ScalarRetTy = RetTy; 1127 if (RetTy->isVectorTy()) { 1128 if (ScalarizationCostPassed == std::numeric_limits<unsigned>::max()) 1129 ScalarizationCost = getScalarizationOverhead(RetTy, true, false); 1130 ScalarCalls = std::max(ScalarCalls, RetTy->getVectorNumElements()); 1131 ScalarRetTy = RetTy->getScalarType(); 1132 } 1133 SmallVector<Type *, 4> ScalarTys; 1134 for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) { 1135 Type *Ty = Tys[i]; 1136 if (Ty->isVectorTy()) { 1137 if (ScalarizationCostPassed == std::numeric_limits<unsigned>::max()) 1138 ScalarizationCost += getScalarizationOverhead(Ty, false, true); 1139 ScalarCalls = std::max(ScalarCalls, Ty->getVectorNumElements()); 1140 Ty = Ty->getScalarType(); 1141 } 1142 ScalarTys.push_back(Ty); 1143 } 1144 if (ScalarCalls == 1) 1145 return 1; // Return cost of a scalar intrinsic. Assume it to be cheap. 1146 1147 unsigned ScalarCost = static_cast<T *>(this)->getIntrinsicInstrCost( 1148 IID, ScalarRetTy, ScalarTys, FMF); 1149 1150 return ScalarCalls * ScalarCost + ScalarizationCost; 1151 } 1152 // Look for intrinsics that can be lowered directly or turned into a scalar 1153 // intrinsic call. 1154 case Intrinsic::sqrt: 1155 ISDs.push_back(ISD::FSQRT); 1156 break; 1157 case Intrinsic::sin: 1158 ISDs.push_back(ISD::FSIN); 1159 break; 1160 case Intrinsic::cos: 1161 ISDs.push_back(ISD::FCOS); 1162 break; 1163 case Intrinsic::exp: 1164 ISDs.push_back(ISD::FEXP); 1165 break; 1166 case Intrinsic::exp2: 1167 ISDs.push_back(ISD::FEXP2); 1168 break; 1169 case Intrinsic::log: 1170 ISDs.push_back(ISD::FLOG); 1171 break; 1172 case Intrinsic::log10: 1173 ISDs.push_back(ISD::FLOG10); 1174 break; 1175 case Intrinsic::log2: 1176 ISDs.push_back(ISD::FLOG2); 1177 break; 1178 case Intrinsic::fabs: 1179 ISDs.push_back(ISD::FABS); 1180 break; 1181 case Intrinsic::canonicalize: 1182 ISDs.push_back(ISD::FCANONICALIZE); 1183 break; 1184 case Intrinsic::minnum: 1185 ISDs.push_back(ISD::FMINNUM); 1186 if (FMF.noNaNs()) 1187 ISDs.push_back(ISD::FMINIMUM); 1188 break; 1189 case Intrinsic::maxnum: 1190 ISDs.push_back(ISD::FMAXNUM); 1191 if (FMF.noNaNs()) 1192 ISDs.push_back(ISD::FMAXIMUM); 1193 break; 1194 case Intrinsic::copysign: 1195 ISDs.push_back(ISD::FCOPYSIGN); 1196 break; 1197 case Intrinsic::floor: 1198 ISDs.push_back(ISD::FFLOOR); 1199 break; 1200 case Intrinsic::ceil: 1201 ISDs.push_back(ISD::FCEIL); 1202 break; 1203 case Intrinsic::trunc: 1204 ISDs.push_back(ISD::FTRUNC); 1205 break; 1206 case Intrinsic::nearbyint: 1207 ISDs.push_back(ISD::FNEARBYINT); 1208 break; 1209 case Intrinsic::rint: 1210 ISDs.push_back(ISD::FRINT); 1211 break; 1212 case Intrinsic::round: 1213 ISDs.push_back(ISD::FROUND); 1214 break; 1215 case Intrinsic::pow: 1216 ISDs.push_back(ISD::FPOW); 1217 break; 1218 case Intrinsic::fma: 1219 ISDs.push_back(ISD::FMA); 1220 break; 1221 case Intrinsic::fmuladd: 1222 ISDs.push_back(ISD::FMA); 1223 break; 1224 // FIXME: We should return 0 whenever getIntrinsicCost == TCC_Free. 1225 case Intrinsic::lifetime_start: 1226 case Intrinsic::lifetime_end: 1227 case Intrinsic::sideeffect: 1228 return 0; 1229 case Intrinsic::masked_store: 1230 return static_cast<T *>(this) 1231 ->getMaskedMemoryOpCost(Instruction::Store, Tys[0], 0, 0); 1232 case Intrinsic::masked_load: 1233 return static_cast<T *>(this) 1234 ->getMaskedMemoryOpCost(Instruction::Load, RetTy, 0, 0); 1235 case Intrinsic::experimental_vector_reduce_add: 1236 return static_cast<T *>(this)->getArithmeticReductionCost( 1237 Instruction::Add, Tys[0], /*IsPairwiseForm=*/false); 1238 case Intrinsic::experimental_vector_reduce_mul: 1239 return static_cast<T *>(this)->getArithmeticReductionCost( 1240 Instruction::Mul, Tys[0], /*IsPairwiseForm=*/false); 1241 case Intrinsic::experimental_vector_reduce_and: 1242 return static_cast<T *>(this)->getArithmeticReductionCost( 1243 Instruction::And, Tys[0], /*IsPairwiseForm=*/false); 1244 case Intrinsic::experimental_vector_reduce_or: 1245 return static_cast<T *>(this)->getArithmeticReductionCost( 1246 Instruction::Or, Tys[0], /*IsPairwiseForm=*/false); 1247 case Intrinsic::experimental_vector_reduce_xor: 1248 return static_cast<T *>(this)->getArithmeticReductionCost( 1249 Instruction::Xor, Tys[0], /*IsPairwiseForm=*/false); 1250 case Intrinsic::experimental_vector_reduce_fadd: 1251 return static_cast<T *>(this)->getArithmeticReductionCost( 1252 Instruction::FAdd, Tys[0], /*IsPairwiseForm=*/false); 1253 case Intrinsic::experimental_vector_reduce_fmul: 1254 return static_cast<T *>(this)->getArithmeticReductionCost( 1255 Instruction::FMul, Tys[0], /*IsPairwiseForm=*/false); 1256 case Intrinsic::experimental_vector_reduce_smax: 1257 case Intrinsic::experimental_vector_reduce_smin: 1258 case Intrinsic::experimental_vector_reduce_fmax: 1259 case Intrinsic::experimental_vector_reduce_fmin: 1260 return static_cast<T *>(this)->getMinMaxReductionCost( 1261 Tys[0], CmpInst::makeCmpResultType(Tys[0]), /*IsPairwiseForm=*/false, 1262 /*IsSigned=*/true); 1263 case Intrinsic::experimental_vector_reduce_umax: 1264 case Intrinsic::experimental_vector_reduce_umin: 1265 return static_cast<T *>(this)->getMinMaxReductionCost( 1266 Tys[0], CmpInst::makeCmpResultType(Tys[0]), /*IsPairwiseForm=*/false, 1267 /*IsSigned=*/false); 1268 case Intrinsic::ctpop: 1269 ISDs.push_back(ISD::CTPOP); 1270 // In case of legalization use TCC_Expensive. This is cheaper than a 1271 // library call but still not a cheap instruction. 1272 SingleCallCost = TargetTransformInfo::TCC_Expensive; 1273 break; 1274 // FIXME: ctlz, cttz, ... 1275 } 1276 1277 const TargetLoweringBase *TLI = getTLI(); 1278 std::pair<unsigned, MVT> LT = TLI->getTypeLegalizationCost(DL, RetTy); 1279 1280 SmallVector<unsigned, 2> LegalCost; 1281 SmallVector<unsigned, 2> CustomCost; 1282 for (unsigned ISD : ISDs) { 1283 if (TLI->isOperationLegalOrPromote(ISD, LT.second)) { 1284 if (IID == Intrinsic::fabs && LT.second.isFloatingPoint() && 1285 TLI->isFAbsFree(LT.second)) { 1286 return 0; 1287 } 1288 1289 // The operation is legal. Assume it costs 1. 1290 // If the type is split to multiple registers, assume that there is some 1291 // overhead to this. 1292 // TODO: Once we have extract/insert subvector cost we need to use them. 1293 if (LT.first > 1) 1294 LegalCost.push_back(LT.first * 2); 1295 else 1296 LegalCost.push_back(LT.first * 1); 1297 } else if (!TLI->isOperationExpand(ISD, LT.second)) { 1298 // If the operation is custom lowered then assume 1299 // that the code is twice as expensive. 1300 CustomCost.push_back(LT.first * 2); 1301 } 1302 } 1303 1304 auto MinLegalCostI = std::min_element(LegalCost.begin(), LegalCost.end()); 1305 if (MinLegalCostI != LegalCost.end()) 1306 return *MinLegalCostI; 1307 1308 auto MinCustomCostI = std::min_element(CustomCost.begin(), CustomCost.end()); 1309 if (MinCustomCostI != CustomCost.end()) 1310 return *MinCustomCostI; 1311 1312 // If we can't lower fmuladd into an FMA estimate the cost as a floating 1313 // point mul followed by an add. 1314 if (IID == Intrinsic::fmuladd) 1315 return static_cast<T *>(this) 1316 ->getArithmeticInstrCost(BinaryOperator::FMul, RetTy) + 1317 static_cast<T *>(this) 1318 ->getArithmeticInstrCost(BinaryOperator::FAdd, RetTy); 1319 1320 // Else, assume that we need to scalarize this intrinsic. For math builtins 1321 // this will emit a costly libcall, adding call overhead and spills. Make it 1322 // very expensive. 1323 if (RetTy->isVectorTy()) { 1324 unsigned ScalarizationCost = 1325 ((ScalarizationCostPassed != std::numeric_limits<unsigned>::max()) 1326 ? ScalarizationCostPassed 1327 : getScalarizationOverhead(RetTy, true, false)); 1328 unsigned ScalarCalls = RetTy->getVectorNumElements(); 1329 SmallVector<Type *, 4> ScalarTys; 1330 for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) { 1331 Type *Ty = Tys[i]; 1332 if (Ty->isVectorTy()) 1333 Ty = Ty->getScalarType(); 1334 ScalarTys.push_back(Ty); 1335 } 1336 unsigned ScalarCost = static_cast<T *>(this)->getIntrinsicInstrCost( 1337 IID, RetTy->getScalarType(), ScalarTys, FMF); 1338 for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) { 1339 if (Tys[i]->isVectorTy()) { 1340 if (ScalarizationCostPassed == std::numeric_limits<unsigned>::max()) 1341 ScalarizationCost += getScalarizationOverhead(Tys[i], false, true); 1342 ScalarCalls = std::max(ScalarCalls, Tys[i]->getVectorNumElements()); 1343 } 1344 } 1345 1346 return ScalarCalls * ScalarCost + ScalarizationCost; 1347 } 1348 1349 // This is going to be turned into a library call, make it expensive. 1350 return SingleCallCost; 1351 } 1352 1353 /// Compute a cost of the given call instruction. 1354 /// 1355 /// Compute the cost of calling function F with return type RetTy and 1356 /// argument types Tys. F might be nullptr, in this case the cost of an 1357 /// arbitrary call with the specified signature will be returned. 1358 /// This is used, for instance, when we estimate call of a vector 1359 /// counterpart of the given function. 1360 /// \param F Called function, might be nullptr. 1361 /// \param RetTy Return value types. 1362 /// \param Tys Argument types. 1363 /// \returns The cost of Call instruction. getCallInstrCost(Function * F,Type * RetTy,ArrayRef<Type * > Tys)1364 unsigned getCallInstrCost(Function *F, Type *RetTy, ArrayRef<Type *> Tys) { 1365 return 10; 1366 } 1367 getNumberOfParts(Type * Tp)1368 unsigned getNumberOfParts(Type *Tp) { 1369 std::pair<unsigned, MVT> LT = getTLI()->getTypeLegalizationCost(DL, Tp); 1370 return LT.first; 1371 } 1372 getAddressComputationCost(Type * Ty,ScalarEvolution *,const SCEV *)1373 unsigned getAddressComputationCost(Type *Ty, ScalarEvolution *, 1374 const SCEV *) { 1375 return 0; 1376 } 1377 1378 /// Try to calculate arithmetic and shuffle op costs for reduction operations. 1379 /// We're assuming that reduction operation are performing the following way: 1380 /// 1. Non-pairwise reduction 1381 /// %val1 = shufflevector<n x t> %val, <n x t> %undef, 1382 /// <n x i32> <i32 n/2, i32 n/2 + 1, ..., i32 n, i32 undef, ..., i32 undef> 1383 /// \----------------v-------------/ \----------v------------/ 1384 /// n/2 elements n/2 elements 1385 /// %red1 = op <n x t> %val, <n x t> val1 1386 /// After this operation we have a vector %red1 where only the first n/2 1387 /// elements are meaningful, the second n/2 elements are undefined and can be 1388 /// dropped. All other operations are actually working with the vector of 1389 /// length n/2, not n, though the real vector length is still n. 1390 /// %val2 = shufflevector<n x t> %red1, <n x t> %undef, 1391 /// <n x i32> <i32 n/4, i32 n/4 + 1, ..., i32 n/2, i32 undef, ..., i32 undef> 1392 /// \----------------v-------------/ \----------v------------/ 1393 /// n/4 elements 3*n/4 elements 1394 /// %red2 = op <n x t> %red1, <n x t> val2 - working with the vector of 1395 /// length n/2, the resulting vector has length n/4 etc. 1396 /// 2. Pairwise reduction: 1397 /// Everything is the same except for an additional shuffle operation which 1398 /// is used to produce operands for pairwise kind of reductions. 1399 /// %val1 = shufflevector<n x t> %val, <n x t> %undef, 1400 /// <n x i32> <i32 0, i32 2, ..., i32 n-2, i32 undef, ..., i32 undef> 1401 /// \-------------v----------/ \----------v------------/ 1402 /// n/2 elements n/2 elements 1403 /// %val2 = shufflevector<n x t> %val, <n x t> %undef, 1404 /// <n x i32> <i32 1, i32 3, ..., i32 n-1, i32 undef, ..., i32 undef> 1405 /// \-------------v----------/ \----------v------------/ 1406 /// n/2 elements n/2 elements 1407 /// %red1 = op <n x t> %val1, <n x t> val2 1408 /// Again, the operation is performed on <n x t> vector, but the resulting 1409 /// vector %red1 is <n/2 x t> vector. 1410 /// 1411 /// The cost model should take into account that the actual length of the 1412 /// vector is reduced on each iteration. getArithmeticReductionCost(unsigned Opcode,Type * Ty,bool IsPairwise)1413 unsigned getArithmeticReductionCost(unsigned Opcode, Type *Ty, 1414 bool IsPairwise) { 1415 assert(Ty->isVectorTy() && "Expect a vector type"); 1416 Type *ScalarTy = Ty->getVectorElementType(); 1417 unsigned NumVecElts = Ty->getVectorNumElements(); 1418 unsigned NumReduxLevels = Log2_32(NumVecElts); 1419 unsigned ArithCost = 0; 1420 unsigned ShuffleCost = 0; 1421 auto *ConcreteTTI = static_cast<T *>(this); 1422 std::pair<unsigned, MVT> LT = 1423 ConcreteTTI->getTLI()->getTypeLegalizationCost(DL, Ty); 1424 unsigned LongVectorCount = 0; 1425 unsigned MVTLen = 1426 LT.second.isVector() ? LT.second.getVectorNumElements() : 1; 1427 while (NumVecElts > MVTLen) { 1428 NumVecElts /= 2; 1429 Type *SubTy = VectorType::get(ScalarTy, NumVecElts); 1430 // Assume the pairwise shuffles add a cost. 1431 ShuffleCost += (IsPairwise + 1) * 1432 ConcreteTTI->getShuffleCost(TTI::SK_ExtractSubvector, Ty, 1433 NumVecElts, SubTy); 1434 ArithCost += ConcreteTTI->getArithmeticInstrCost(Opcode, SubTy); 1435 Ty = SubTy; 1436 ++LongVectorCount; 1437 } 1438 1439 NumReduxLevels -= LongVectorCount; 1440 1441 // The minimal length of the vector is limited by the real length of vector 1442 // operations performed on the current platform. That's why several final 1443 // reduction operations are performed on the vectors with the same 1444 // architecture-dependent length. 1445 1446 // Non pairwise reductions need one shuffle per reduction level. Pairwise 1447 // reductions need two shuffles on every level, but the last one. On that 1448 // level one of the shuffles is <0, u, u, ...> which is identity. 1449 unsigned NumShuffles = NumReduxLevels; 1450 if (IsPairwise && NumReduxLevels >= 1) 1451 NumShuffles += NumReduxLevels - 1; 1452 ShuffleCost += NumShuffles * 1453 ConcreteTTI->getShuffleCost(TTI::SK_PermuteSingleSrc, Ty, 1454 0, Ty); 1455 ArithCost += NumReduxLevels * 1456 ConcreteTTI->getArithmeticInstrCost(Opcode, Ty); 1457 return ShuffleCost + ArithCost + 1458 ConcreteTTI->getVectorInstrCost(Instruction::ExtractElement, Ty, 0); 1459 } 1460 1461 /// Try to calculate op costs for min/max reduction operations. 1462 /// \param CondTy Conditional type for the Select instruction. getMinMaxReductionCost(Type * Ty,Type * CondTy,bool IsPairwise,bool)1463 unsigned getMinMaxReductionCost(Type *Ty, Type *CondTy, bool IsPairwise, 1464 bool) { 1465 assert(Ty->isVectorTy() && "Expect a vector type"); 1466 Type *ScalarTy = Ty->getVectorElementType(); 1467 Type *ScalarCondTy = CondTy->getVectorElementType(); 1468 unsigned NumVecElts = Ty->getVectorNumElements(); 1469 unsigned NumReduxLevels = Log2_32(NumVecElts); 1470 unsigned CmpOpcode; 1471 if (Ty->isFPOrFPVectorTy()) { 1472 CmpOpcode = Instruction::FCmp; 1473 } else { 1474 assert(Ty->isIntOrIntVectorTy() && 1475 "expecting floating point or integer type for min/max reduction"); 1476 CmpOpcode = Instruction::ICmp; 1477 } 1478 unsigned MinMaxCost = 0; 1479 unsigned ShuffleCost = 0; 1480 auto *ConcreteTTI = static_cast<T *>(this); 1481 std::pair<unsigned, MVT> LT = 1482 ConcreteTTI->getTLI()->getTypeLegalizationCost(DL, Ty); 1483 unsigned LongVectorCount = 0; 1484 unsigned MVTLen = 1485 LT.second.isVector() ? LT.second.getVectorNumElements() : 1; 1486 while (NumVecElts > MVTLen) { 1487 NumVecElts /= 2; 1488 Type *SubTy = VectorType::get(ScalarTy, NumVecElts); 1489 CondTy = VectorType::get(ScalarCondTy, NumVecElts); 1490 1491 // Assume the pairwise shuffles add a cost. 1492 ShuffleCost += (IsPairwise + 1) * 1493 ConcreteTTI->getShuffleCost(TTI::SK_ExtractSubvector, Ty, 1494 NumVecElts, SubTy); 1495 MinMaxCost += 1496 ConcreteTTI->getCmpSelInstrCost(CmpOpcode, SubTy, CondTy, nullptr) + 1497 ConcreteTTI->getCmpSelInstrCost(Instruction::Select, SubTy, CondTy, 1498 nullptr); 1499 Ty = SubTy; 1500 ++LongVectorCount; 1501 } 1502 1503 NumReduxLevels -= LongVectorCount; 1504 1505 // The minimal length of the vector is limited by the real length of vector 1506 // operations performed on the current platform. That's why several final 1507 // reduction opertions are perfomed on the vectors with the same 1508 // architecture-dependent length. 1509 1510 // Non pairwise reductions need one shuffle per reduction level. Pairwise 1511 // reductions need two shuffles on every level, but the last one. On that 1512 // level one of the shuffles is <0, u, u, ...> which is identity. 1513 unsigned NumShuffles = NumReduxLevels; 1514 if (IsPairwise && NumReduxLevels >= 1) 1515 NumShuffles += NumReduxLevels - 1; 1516 ShuffleCost += NumShuffles * 1517 ConcreteTTI->getShuffleCost(TTI::SK_PermuteSingleSrc, Ty, 1518 0, Ty); 1519 MinMaxCost += 1520 NumReduxLevels * 1521 (ConcreteTTI->getCmpSelInstrCost(CmpOpcode, Ty, CondTy, nullptr) + 1522 ConcreteTTI->getCmpSelInstrCost(Instruction::Select, Ty, CondTy, 1523 nullptr)); 1524 // The last min/max should be in vector registers and we counted it above. 1525 // So just need a single extractelement. 1526 return ShuffleCost + MinMaxCost + 1527 ConcreteTTI->getVectorInstrCost(Instruction::ExtractElement, Ty, 0); 1528 } 1529 getVectorSplitCost()1530 unsigned getVectorSplitCost() { return 1; } 1531 1532 /// @} 1533 }; 1534 1535 /// Concrete BasicTTIImpl that can be used if no further customization 1536 /// is needed. 1537 class BasicTTIImpl : public BasicTTIImplBase<BasicTTIImpl> { 1538 using BaseT = BasicTTIImplBase<BasicTTIImpl>; 1539 1540 friend class BasicTTIImplBase<BasicTTIImpl>; 1541 1542 const TargetSubtargetInfo *ST; 1543 const TargetLoweringBase *TLI; 1544 getST()1545 const TargetSubtargetInfo *getST() const { return ST; } getTLI()1546 const TargetLoweringBase *getTLI() const { return TLI; } 1547 1548 public: 1549 explicit BasicTTIImpl(const TargetMachine *TM, const Function &F); 1550 }; 1551 1552 } // end namespace llvm 1553 1554 #endif // LLVM_CODEGEN_BASICTTIIMPL_H 1555