1 //===----------------------- AlignmentFromAssumptions.cpp -----------------===// 2 // Set Load/Store Alignments From Assumptions 3 // 4 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. 5 // See https://llvm.org/LICENSE.txt for license information. 6 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception 7 // 8 //===----------------------------------------------------------------------===// 9 // 10 // This file implements a ScalarEvolution-based transformation to set 11 // the alignments of load, stores and memory intrinsics based on the truth 12 // expressions of assume intrinsics. The primary motivation is to handle 13 // complex alignment assumptions that apply to vector loads and stores that 14 // appear after vectorization and unrolling. 15 // 16 //===----------------------------------------------------------------------===// 17 18 #include "llvm/IR/Instructions.h" 19 #include "llvm/InitializePasses.h" 20 #include "llvm/Transforms/Scalar/AlignmentFromAssumptions.h" 21 #include "llvm/ADT/SmallPtrSet.h" 22 #include "llvm/ADT/Statistic.h" 23 #include "llvm/Analysis/AliasAnalysis.h" 24 #include "llvm/Analysis/AssumptionCache.h" 25 #include "llvm/Analysis/GlobalsModRef.h" 26 #include "llvm/Analysis/LoopInfo.h" 27 #include "llvm/Analysis/ScalarEvolutionExpressions.h" 28 #include "llvm/Analysis/ValueTracking.h" 29 #include "llvm/IR/Constant.h" 30 #include "llvm/IR/Dominators.h" 31 #include "llvm/IR/Instruction.h" 32 #include "llvm/IR/IntrinsicInst.h" 33 #include "llvm/IR/Intrinsics.h" 34 #include "llvm/IR/Module.h" 35 #include "llvm/Support/Debug.h" 36 #include "llvm/Support/raw_ostream.h" 37 #include "llvm/Transforms/Scalar.h" 38 39 #define AA_NAME "alignment-from-assumptions" 40 #define DEBUG_TYPE AA_NAME 41 using namespace llvm; 42 43 STATISTIC(NumLoadAlignChanged, 44 "Number of loads changed by alignment assumptions"); 45 STATISTIC(NumStoreAlignChanged, 46 "Number of stores changed by alignment assumptions"); 47 STATISTIC(NumMemIntAlignChanged, 48 "Number of memory intrinsics changed by alignment assumptions"); 49 50 namespace { 51 struct AlignmentFromAssumptions : public FunctionPass { 52 static char ID; // Pass identification, replacement for typeid 53 AlignmentFromAssumptions() : FunctionPass(ID) { 54 initializeAlignmentFromAssumptionsPass(*PassRegistry::getPassRegistry()); 55 } 56 57 bool runOnFunction(Function &F) override; 58 59 void getAnalysisUsage(AnalysisUsage &AU) const override { 60 AU.addRequired<AssumptionCacheTracker>(); 61 AU.addRequired<ScalarEvolutionWrapperPass>(); 62 AU.addRequired<DominatorTreeWrapperPass>(); 63 64 AU.setPreservesCFG(); 65 AU.addPreserved<AAResultsWrapperPass>(); 66 AU.addPreserved<GlobalsAAWrapperPass>(); 67 AU.addPreserved<LoopInfoWrapperPass>(); 68 AU.addPreserved<DominatorTreeWrapperPass>(); 69 AU.addPreserved<ScalarEvolutionWrapperPass>(); 70 } 71 72 AlignmentFromAssumptionsPass Impl; 73 }; 74 } 75 76 char AlignmentFromAssumptions::ID = 0; 77 static const char aip_name[] = "Alignment from assumptions"; 78 INITIALIZE_PASS_BEGIN(AlignmentFromAssumptions, AA_NAME, 79 aip_name, false, false) 80 INITIALIZE_PASS_DEPENDENCY(AssumptionCacheTracker) 81 INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass) 82 INITIALIZE_PASS_DEPENDENCY(ScalarEvolutionWrapperPass) 83 INITIALIZE_PASS_END(AlignmentFromAssumptions, AA_NAME, 84 aip_name, false, false) 85 86 FunctionPass *llvm::createAlignmentFromAssumptionsPass() { 87 return new AlignmentFromAssumptions(); 88 } 89 90 // Given an expression for the (constant) alignment, AlignSCEV, and an 91 // expression for the displacement between a pointer and the aligned address, 92 // DiffSCEV, compute the alignment of the displaced pointer if it can be reduced 93 // to a constant. Using SCEV to compute alignment handles the case where 94 // DiffSCEV is a recurrence with constant start such that the aligned offset 95 // is constant. e.g. {16,+,32} % 32 -> 16. 96 static MaybeAlign getNewAlignmentDiff(const SCEV *DiffSCEV, 97 const SCEV *AlignSCEV, 98 ScalarEvolution *SE) { 99 // DiffUnits = Diff % int64_t(Alignment) 100 const SCEV *DiffUnitsSCEV = SE->getURemExpr(DiffSCEV, AlignSCEV); 101 102 LLVM_DEBUG(dbgs() << "\talignment relative to " << *AlignSCEV << " is " 103 << *DiffUnitsSCEV << " (diff: " << *DiffSCEV << ")\n"); 104 105 if (const SCEVConstant *ConstDUSCEV = 106 dyn_cast<SCEVConstant>(DiffUnitsSCEV)) { 107 int64_t DiffUnits = ConstDUSCEV->getValue()->getSExtValue(); 108 109 // If the displacement is an exact multiple of the alignment, then the 110 // displaced pointer has the same alignment as the aligned pointer, so 111 // return the alignment value. 112 if (!DiffUnits) 113 return cast<SCEVConstant>(AlignSCEV)->getValue()->getAlignValue(); 114 115 // If the displacement is not an exact multiple, but the remainder is a 116 // constant, then return this remainder (but only if it is a power of 2). 117 uint64_t DiffUnitsAbs = std::abs(DiffUnits); 118 if (isPowerOf2_64(DiffUnitsAbs)) 119 return Align(DiffUnitsAbs); 120 } 121 122 return None; 123 } 124 125 // There is an address given by an offset OffSCEV from AASCEV which has an 126 // alignment AlignSCEV. Use that information, if possible, to compute a new 127 // alignment for Ptr. 128 static Align getNewAlignment(const SCEV *AASCEV, const SCEV *AlignSCEV, 129 const SCEV *OffSCEV, Value *Ptr, 130 ScalarEvolution *SE) { 131 const SCEV *PtrSCEV = SE->getSCEV(Ptr); 132 // On a platform with 32-bit allocas, but 64-bit flat/global pointer sizes 133 // (*cough* AMDGPU), the effective SCEV type of AASCEV and PtrSCEV 134 // may disagree. Trunc/extend so they agree. 135 PtrSCEV = SE->getTruncateOrZeroExtend( 136 PtrSCEV, SE->getEffectiveSCEVType(AASCEV->getType())); 137 const SCEV *DiffSCEV = SE->getMinusSCEV(PtrSCEV, AASCEV); 138 if (isa<SCEVCouldNotCompute>(DiffSCEV)) 139 return Align(1); 140 141 // On 32-bit platforms, DiffSCEV might now have type i32 -- we've always 142 // sign-extended OffSCEV to i64, so make sure they agree again. 143 DiffSCEV = SE->getNoopOrSignExtend(DiffSCEV, OffSCEV->getType()); 144 145 // What we really want to know is the overall offset to the aligned 146 // address. This address is displaced by the provided offset. 147 DiffSCEV = SE->getAddExpr(DiffSCEV, OffSCEV); 148 149 LLVM_DEBUG(dbgs() << "AFI: alignment of " << *Ptr << " relative to " 150 << *AlignSCEV << " and offset " << *OffSCEV 151 << " using diff " << *DiffSCEV << "\n"); 152 153 if (MaybeAlign NewAlignment = getNewAlignmentDiff(DiffSCEV, AlignSCEV, SE)) { 154 LLVM_DEBUG(dbgs() << "\tnew alignment: " << DebugStr(NewAlignment) << "\n"); 155 return *NewAlignment; 156 } 157 158 if (const SCEVAddRecExpr *DiffARSCEV = dyn_cast<SCEVAddRecExpr>(DiffSCEV)) { 159 // The relative offset to the alignment assumption did not yield a constant, 160 // but we should try harder: if we assume that a is 32-byte aligned, then in 161 // for (i = 0; i < 1024; i += 4) r += a[i]; not all of the loads from a are 162 // 32-byte aligned, but instead alternate between 32 and 16-byte alignment. 163 // As a result, the new alignment will not be a constant, but can still 164 // be improved over the default (of 4) to 16. 165 166 const SCEV *DiffStartSCEV = DiffARSCEV->getStart(); 167 const SCEV *DiffIncSCEV = DiffARSCEV->getStepRecurrence(*SE); 168 169 LLVM_DEBUG(dbgs() << "\ttrying start/inc alignment using start " 170 << *DiffStartSCEV << " and inc " << *DiffIncSCEV << "\n"); 171 172 // Now compute the new alignment using the displacement to the value in the 173 // first iteration, and also the alignment using the per-iteration delta. 174 // If these are the same, then use that answer. Otherwise, use the smaller 175 // one, but only if it divides the larger one. 176 MaybeAlign NewAlignment = getNewAlignmentDiff(DiffStartSCEV, AlignSCEV, SE); 177 MaybeAlign NewIncAlignment = 178 getNewAlignmentDiff(DiffIncSCEV, AlignSCEV, SE); 179 180 LLVM_DEBUG(dbgs() << "\tnew start alignment: " << DebugStr(NewAlignment) 181 << "\n"); 182 LLVM_DEBUG(dbgs() << "\tnew inc alignment: " << DebugStr(NewIncAlignment) 183 << "\n"); 184 185 if (!NewAlignment || !NewIncAlignment) 186 return Align(1); 187 188 const Align NewAlign = *NewAlignment; 189 const Align NewIncAlign = *NewIncAlignment; 190 if (NewAlign > NewIncAlign) { 191 LLVM_DEBUG(dbgs() << "\tnew start/inc alignment: " 192 << DebugStr(NewIncAlign) << "\n"); 193 return NewIncAlign; 194 } 195 if (NewIncAlign > NewAlign) { 196 LLVM_DEBUG(dbgs() << "\tnew start/inc alignment: " << DebugStr(NewAlign) 197 << "\n"); 198 return NewAlign; 199 } 200 assert(NewIncAlign == NewAlign); 201 LLVM_DEBUG(dbgs() << "\tnew start/inc alignment: " << DebugStr(NewAlign) 202 << "\n"); 203 return NewAlign; 204 } 205 206 return Align(1); 207 } 208 209 bool AlignmentFromAssumptionsPass::extractAlignmentInfo(CallInst *I, 210 unsigned Idx, 211 Value *&AAPtr, 212 const SCEV *&AlignSCEV, 213 const SCEV *&OffSCEV) { 214 Type *Int64Ty = Type::getInt64Ty(I->getContext()); 215 OperandBundleUse AlignOB = I->getOperandBundleAt(Idx); 216 if (AlignOB.getTagName() != "align") 217 return false; 218 assert(AlignOB.Inputs.size() >= 2); 219 AAPtr = AlignOB.Inputs[0].get(); 220 // TODO: Consider accumulating the offset to the base. 221 AAPtr = AAPtr->stripPointerCastsSameRepresentation(); 222 AlignSCEV = SE->getSCEV(AlignOB.Inputs[1].get()); 223 AlignSCEV = SE->getTruncateOrZeroExtend(AlignSCEV, Int64Ty); 224 if (!isa<SCEVConstant>(AlignSCEV)) 225 // Added to suppress a crash because consumer doesn't expect non-constant 226 // alignments in the assume bundle. TODO: Consider generalizing caller. 227 return false; 228 if (AlignOB.Inputs.size() == 3) 229 OffSCEV = SE->getSCEV(AlignOB.Inputs[2].get()); 230 else 231 OffSCEV = SE->getZero(Int64Ty); 232 OffSCEV = SE->getTruncateOrZeroExtend(OffSCEV, Int64Ty); 233 return true; 234 } 235 236 bool AlignmentFromAssumptionsPass::processAssumption(CallInst *ACall, 237 unsigned Idx) { 238 Value *AAPtr; 239 const SCEV *AlignSCEV, *OffSCEV; 240 if (!extractAlignmentInfo(ACall, Idx, AAPtr, AlignSCEV, OffSCEV)) 241 return false; 242 243 // Skip ConstantPointerNull and UndefValue. Assumptions on these shouldn't 244 // affect other users. 245 if (isa<ConstantData>(AAPtr)) 246 return false; 247 248 const SCEV *AASCEV = SE->getSCEV(AAPtr); 249 250 // Apply the assumption to all other users of the specified pointer. 251 SmallPtrSet<Instruction *, 32> Visited; 252 SmallVector<Instruction*, 16> WorkList; 253 for (User *J : AAPtr->users()) { 254 if (J == ACall) 255 continue; 256 257 if (Instruction *K = dyn_cast<Instruction>(J)) 258 WorkList.push_back(K); 259 } 260 261 while (!WorkList.empty()) { 262 Instruction *J = WorkList.pop_back_val(); 263 if (LoadInst *LI = dyn_cast<LoadInst>(J)) { 264 if (!isValidAssumeForContext(ACall, J, DT)) 265 continue; 266 Align NewAlignment = getNewAlignment(AASCEV, AlignSCEV, OffSCEV, 267 LI->getPointerOperand(), SE); 268 if (NewAlignment > LI->getAlign()) { 269 LI->setAlignment(NewAlignment); 270 ++NumLoadAlignChanged; 271 } 272 } else if (StoreInst *SI = dyn_cast<StoreInst>(J)) { 273 if (!isValidAssumeForContext(ACall, J, DT)) 274 continue; 275 Align NewAlignment = getNewAlignment(AASCEV, AlignSCEV, OffSCEV, 276 SI->getPointerOperand(), SE); 277 if (NewAlignment > SI->getAlign()) { 278 SI->setAlignment(NewAlignment); 279 ++NumStoreAlignChanged; 280 } 281 } else if (MemIntrinsic *MI = dyn_cast<MemIntrinsic>(J)) { 282 if (!isValidAssumeForContext(ACall, J, DT)) 283 continue; 284 Align NewDestAlignment = 285 getNewAlignment(AASCEV, AlignSCEV, OffSCEV, MI->getDest(), SE); 286 287 LLVM_DEBUG(dbgs() << "\tmem inst: " << DebugStr(NewDestAlignment) 288 << "\n";); 289 if (NewDestAlignment > *MI->getDestAlign()) { 290 MI->setDestAlignment(NewDestAlignment); 291 ++NumMemIntAlignChanged; 292 } 293 294 // For memory transfers, there is also a source alignment that 295 // can be set. 296 if (MemTransferInst *MTI = dyn_cast<MemTransferInst>(MI)) { 297 Align NewSrcAlignment = 298 getNewAlignment(AASCEV, AlignSCEV, OffSCEV, MTI->getSource(), SE); 299 300 LLVM_DEBUG(dbgs() << "\tmem trans: " << DebugStr(NewSrcAlignment) 301 << "\n";); 302 303 if (NewSrcAlignment > *MTI->getSourceAlign()) { 304 MTI->setSourceAlignment(NewSrcAlignment); 305 ++NumMemIntAlignChanged; 306 } 307 } 308 } 309 310 // Now that we've updated that use of the pointer, look for other uses of 311 // the pointer to update. 312 Visited.insert(J); 313 for (User *UJ : J->users()) { 314 Instruction *K = cast<Instruction>(UJ); 315 if (!Visited.count(K)) 316 WorkList.push_back(K); 317 } 318 } 319 320 return true; 321 } 322 323 bool AlignmentFromAssumptions::runOnFunction(Function &F) { 324 if (skipFunction(F)) 325 return false; 326 327 auto &AC = getAnalysis<AssumptionCacheTracker>().getAssumptionCache(F); 328 ScalarEvolution *SE = &getAnalysis<ScalarEvolutionWrapperPass>().getSE(); 329 DominatorTree *DT = &getAnalysis<DominatorTreeWrapperPass>().getDomTree(); 330 331 return Impl.runImpl(F, AC, SE, DT); 332 } 333 334 bool AlignmentFromAssumptionsPass::runImpl(Function &F, AssumptionCache &AC, 335 ScalarEvolution *SE_, 336 DominatorTree *DT_) { 337 SE = SE_; 338 DT = DT_; 339 340 bool Changed = false; 341 for (auto &AssumeVH : AC.assumptions()) 342 if (AssumeVH) { 343 CallInst *Call = cast<CallInst>(AssumeVH); 344 for (unsigned Idx = 0; Idx < Call->getNumOperandBundles(); Idx++) 345 Changed |= processAssumption(Call, Idx); 346 } 347 348 return Changed; 349 } 350 351 PreservedAnalyses 352 AlignmentFromAssumptionsPass::run(Function &F, FunctionAnalysisManager &AM) { 353 354 AssumptionCache &AC = AM.getResult<AssumptionAnalysis>(F); 355 ScalarEvolution &SE = AM.getResult<ScalarEvolutionAnalysis>(F); 356 DominatorTree &DT = AM.getResult<DominatorTreeAnalysis>(F); 357 if (!runImpl(F, AC, &SE, &DT)) 358 return PreservedAnalyses::all(); 359 360 PreservedAnalyses PA; 361 PA.preserveSet<CFGAnalyses>(); 362 PA.preserve<ScalarEvolutionAnalysis>(); 363 return PA; 364 } 365