1 //===- NVPTXLowerAggrCopies.cpp - ------------------------------*- C++ -*--===// 2 // 3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. 4 // See https://llvm.org/LICENSE.txt for license information. 5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception 6 // 7 //===----------------------------------------------------------------------===// 8 // 9 // \file 10 // Lower aggregate copies, memset, memcpy, memmov intrinsics into loops when 11 // the size is large or is not a compile-time constant. 12 // 13 //===----------------------------------------------------------------------===// 14 15 #include "NVPTXLowerAggrCopies.h" 16 #include "llvm/Analysis/TargetTransformInfo.h" 17 #include "llvm/CodeGen/StackProtector.h" 18 #include "llvm/IR/Constants.h" 19 #include "llvm/IR/DataLayout.h" 20 #include "llvm/IR/Function.h" 21 #include "llvm/IR/IRBuilder.h" 22 #include "llvm/IR/Instructions.h" 23 #include "llvm/IR/IntrinsicInst.h" 24 #include "llvm/IR/Intrinsics.h" 25 #include "llvm/IR/LLVMContext.h" 26 #include "llvm/IR/Module.h" 27 #include "llvm/Support/Debug.h" 28 #include "llvm/Transforms/Utils/BasicBlockUtils.h" 29 #include "llvm/Transforms/Utils/LowerMemIntrinsics.h" 30 31 #define DEBUG_TYPE "nvptx" 32 33 using namespace llvm; 34 35 namespace { 36 37 // actual analysis class, which is a functionpass 38 struct NVPTXLowerAggrCopies : public FunctionPass { 39 static char ID; 40 41 NVPTXLowerAggrCopies() : FunctionPass(ID) {} 42 43 void getAnalysisUsage(AnalysisUsage &AU) const override { 44 AU.addPreserved<StackProtector>(); 45 AU.addRequired<TargetTransformInfoWrapperPass>(); 46 } 47 48 bool runOnFunction(Function &F) override; 49 50 static const unsigned MaxAggrCopySize = 128; 51 52 StringRef getPassName() const override { 53 return "Lower aggregate copies/intrinsics into loops"; 54 } 55 }; 56 57 char NVPTXLowerAggrCopies::ID = 0; 58 59 bool NVPTXLowerAggrCopies::runOnFunction(Function &F) { 60 SmallVector<LoadInst *, 4> AggrLoads; 61 SmallVector<MemIntrinsic *, 4> MemCalls; 62 63 const DataLayout &DL = F.getParent()->getDataLayout(); 64 LLVMContext &Context = F.getParent()->getContext(); 65 const TargetTransformInfo &TTI = 66 getAnalysis<TargetTransformInfoWrapperPass>().getTTI(F); 67 68 // Collect all aggregate loads and mem* calls. 69 for (BasicBlock &BB : F) { 70 for (Instruction &I : BB) { 71 if (LoadInst *LI = dyn_cast<LoadInst>(&I)) { 72 if (!LI->hasOneUse()) 73 continue; 74 75 if (DL.getTypeStoreSize(LI->getType()) < MaxAggrCopySize) 76 continue; 77 78 if (StoreInst *SI = dyn_cast<StoreInst>(LI->user_back())) { 79 if (SI->getOperand(0) != LI) 80 continue; 81 AggrLoads.push_back(LI); 82 } 83 } else if (MemIntrinsic *IntrCall = dyn_cast<MemIntrinsic>(&I)) { 84 // Convert intrinsic calls with variable size or with constant size 85 // larger than the MaxAggrCopySize threshold. 86 if (ConstantInt *LenCI = dyn_cast<ConstantInt>(IntrCall->getLength())) { 87 if (LenCI->getZExtValue() >= MaxAggrCopySize) { 88 MemCalls.push_back(IntrCall); 89 } 90 } else { 91 MemCalls.push_back(IntrCall); 92 } 93 } 94 } 95 } 96 97 if (AggrLoads.size() == 0 && MemCalls.size() == 0) { 98 return false; 99 } 100 101 // 102 // Do the transformation of an aggr load/copy/set to a loop 103 // 104 for (LoadInst *LI : AggrLoads) { 105 auto *SI = cast<StoreInst>(*LI->user_begin()); 106 Value *SrcAddr = LI->getOperand(0); 107 Value *DstAddr = SI->getOperand(1); 108 unsigned NumLoads = DL.getTypeStoreSize(LI->getType()); 109 ConstantInt *CopyLen = 110 ConstantInt::get(Type::getInt32Ty(Context), NumLoads); 111 112 createMemCpyLoopKnownSize(/* ConvertedInst */ SI, 113 /* SrcAddr */ SrcAddr, /* DstAddr */ DstAddr, 114 /* CopyLen */ CopyLen, 115 /* SrcAlign */ LI->getAlign(), 116 /* DestAlign */ SI->getAlign(), 117 /* SrcIsVolatile */ LI->isVolatile(), 118 /* DstIsVolatile */ SI->isVolatile(), 119 /* CanOverlap */ true, TTI); 120 121 SI->eraseFromParent(); 122 LI->eraseFromParent(); 123 } 124 125 // Transform mem* intrinsic calls. 126 for (MemIntrinsic *MemCall : MemCalls) { 127 if (MemCpyInst *Memcpy = dyn_cast<MemCpyInst>(MemCall)) { 128 expandMemCpyAsLoop(Memcpy, TTI); 129 } else if (MemMoveInst *Memmove = dyn_cast<MemMoveInst>(MemCall)) { 130 expandMemMoveAsLoop(Memmove); 131 } else if (MemSetInst *Memset = dyn_cast<MemSetInst>(MemCall)) { 132 expandMemSetAsLoop(Memset); 133 } 134 MemCall->eraseFromParent(); 135 } 136 137 return true; 138 } 139 140 } // namespace 141 142 namespace llvm { 143 void initializeNVPTXLowerAggrCopiesPass(PassRegistry &); 144 } 145 146 INITIALIZE_PASS(NVPTXLowerAggrCopies, "nvptx-lower-aggr-copies", 147 "Lower aggregate copies, and llvm.mem* intrinsics into loops", 148 false, false) 149 150 FunctionPass *llvm::createLowerAggrCopies() { 151 return new NVPTXLowerAggrCopies(); 152 } 153