diff --git a/llvm/lib/IR/AutoUpgrade.cpp b/llvm/lib/IR/AutoUpgrade.cpp index 5beefaa1ec701..53de9eef516b3 100644 --- a/llvm/lib/IR/AutoUpgrade.cpp +++ b/llvm/lib/IR/AutoUpgrade.cpp @@ -2245,6 +2245,1696 @@ void llvm::UpgradeInlineAsmString(std::string *AsmStr) { } } +static Value *upgradeX86IntrinsicCall(StringRef Name, CallBase *CI, Function *F, + IRBuilder<> &Builder) { + LLVMContext &C = F->getContext(); + Value *Rep = nullptr; + + if (Name.starts_with("sse4a.movnt.")) { + SmallVector Elts; + Elts.push_back( + ConstantAsMetadata::get(ConstantInt::get(Type::getInt32Ty(C), 1))); + MDNode *Node = MDNode::get(C, Elts); + + Value *Arg0 = CI->getArgOperand(0); + Value *Arg1 = CI->getArgOperand(1); + + // Nontemporal (unaligned) store of the 0'th element of the float/double + // vector. + Type *SrcEltTy = cast(Arg1->getType())->getElementType(); + PointerType *EltPtrTy = PointerType::getUnqual(SrcEltTy); + Value *Addr = Builder.CreateBitCast(Arg0, EltPtrTy, "cast"); + Value *Extract = + Builder.CreateExtractElement(Arg1, (uint64_t)0, "extractelement"); + + StoreInst *SI = Builder.CreateAlignedStore(Extract, Addr, Align(1)); + SI->setMetadata(LLVMContext::MD_nontemporal, Node); + } else if (Name.starts_with("avx.movnt.") || + Name.starts_with("avx512.storent.")) { + SmallVector Elts; + Elts.push_back( + ConstantAsMetadata::get(ConstantInt::get(Type::getInt32Ty(C), 1))); + MDNode *Node = MDNode::get(C, Elts); + + Value *Arg0 = CI->getArgOperand(0); + Value *Arg1 = CI->getArgOperand(1); + + // Convert the type of the pointer to a pointer to the stored type. + Value *BC = Builder.CreateBitCast( + Arg0, PointerType::getUnqual(Arg1->getType()), "cast"); + StoreInst *SI = Builder.CreateAlignedStore( + Arg1, BC, + Align(Arg1->getType()->getPrimitiveSizeInBits().getFixedValue() / 8)); + SI->setMetadata(LLVMContext::MD_nontemporal, Node); + } else if (Name == "sse2.storel.dq") { + Value *Arg0 = CI->getArgOperand(0); + Value *Arg1 = CI->getArgOperand(1); + + auto *NewVecTy = FixedVectorType::get(Type::getInt64Ty(C), 2); + Value *BC0 = Builder.CreateBitCast(Arg1, NewVecTy, "cast"); + Value *Elt = Builder.CreateExtractElement(BC0, (uint64_t)0); + Value *BC = Builder.CreateBitCast( + Arg0, PointerType::getUnqual(Elt->getType()), "cast"); + Builder.CreateAlignedStore(Elt, BC, Align(1)); + } else if (Name.starts_with("sse.storeu.") || + Name.starts_with("sse2.storeu.") || + Name.starts_with("avx.storeu.")) { + Value *Arg0 = CI->getArgOperand(0); + Value *Arg1 = CI->getArgOperand(1); + + Arg0 = Builder.CreateBitCast(Arg0, PointerType::getUnqual(Arg1->getType()), + "cast"); + Builder.CreateAlignedStore(Arg1, Arg0, Align(1)); + } else if (Name == "avx512.mask.store.ss") { + Value *Mask = Builder.CreateAnd(CI->getArgOperand(2), Builder.getInt8(1)); + upgradeMaskedStore(Builder, CI->getArgOperand(0), CI->getArgOperand(1), + Mask, false); + } else if (Name.starts_with("avx512.mask.store")) { + // "avx512.mask.storeu." or "avx512.mask.store." + bool Aligned = Name[17] != 'u'; // "avx512.mask.storeu". + upgradeMaskedStore(Builder, CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2), Aligned); + } else if (Name.starts_with("sse2.pcmp") || Name.starts_with("avx2.pcmp")) { + // Upgrade packed integer vector compare intrinsics to compare instructions. + // "sse2.pcpmpeq." "sse2.pcmpgt." "avx2.pcmpeq." or "avx2.pcmpgt." + bool CmpEq = Name[9] == 'e'; + Rep = Builder.CreateICmp(CmpEq ? ICmpInst::ICMP_EQ : ICmpInst::ICMP_SGT, + CI->getArgOperand(0), CI->getArgOperand(1)); + Rep = Builder.CreateSExt(Rep, CI->getType(), ""); + } else if (Name.starts_with("avx512.broadcastm")) { + Type *ExtTy = Type::getInt32Ty(C); + if (CI->getOperand(0)->getType()->isIntegerTy(8)) + ExtTy = Type::getInt64Ty(C); + unsigned NumElts = CI->getType()->getPrimitiveSizeInBits() / + ExtTy->getPrimitiveSizeInBits(); + Rep = Builder.CreateZExt(CI->getArgOperand(0), ExtTy); + Rep = Builder.CreateVectorSplat(NumElts, Rep); + } else if (Name == "sse.sqrt.ss" || Name == "sse2.sqrt.sd") { + Value *Vec = CI->getArgOperand(0); + Value *Elt0 = Builder.CreateExtractElement(Vec, (uint64_t)0); + Function *Intr = Intrinsic::getDeclaration(F->getParent(), Intrinsic::sqrt, + Elt0->getType()); + Elt0 = Builder.CreateCall(Intr, Elt0); + Rep = Builder.CreateInsertElement(Vec, Elt0, (uint64_t)0); + } else if (Name.starts_with("avx.sqrt.p") || + Name.starts_with("sse2.sqrt.p") || + Name.starts_with("sse.sqrt.p")) { + Rep = + Builder.CreateCall(Intrinsic::getDeclaration( + F->getParent(), Intrinsic::sqrt, CI->getType()), + {CI->getArgOperand(0)}); + } else if (Name.starts_with("avx512.mask.sqrt.p")) { + if (CI->arg_size() == 4 && + (!isa(CI->getArgOperand(3)) || + cast(CI->getArgOperand(3))->getZExtValue() != 4)) { + Intrinsic::ID IID = Name[18] == 's' ? Intrinsic::x86_avx512_sqrt_ps_512 + : Intrinsic::x86_avx512_sqrt_pd_512; + + Value *Args[] = {CI->getArgOperand(0), CI->getArgOperand(3)}; + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), + Args); + } else { + Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), + Intrinsic::sqrt, + CI->getType()), + {CI->getArgOperand(0)}); + } + Rep = + emitX86Select(Builder, CI->getArgOperand(2), Rep, CI->getArgOperand(1)); + } else if (Name.starts_with("avx512.ptestm") || + Name.starts_with("avx512.ptestnm")) { + Value *Op0 = CI->getArgOperand(0); + Value *Op1 = CI->getArgOperand(1); + Value *Mask = CI->getArgOperand(2); + Rep = Builder.CreateAnd(Op0, Op1); + llvm::Type *Ty = Op0->getType(); + Value *Zero = llvm::Constant::getNullValue(Ty); + ICmpInst::Predicate Pred = Name.starts_with("avx512.ptestm") + ? ICmpInst::ICMP_NE + : ICmpInst::ICMP_EQ; + Rep = Builder.CreateICmp(Pred, Rep, Zero); + Rep = applyX86MaskOn1BitsVec(Builder, Rep, Mask); + } else if (Name.starts_with("avx512.mask.pbroadcast")) { + unsigned NumElts = cast(CI->getArgOperand(1)->getType()) + ->getNumElements(); + Rep = Builder.CreateVectorSplat(NumElts, CI->getArgOperand(0)); + Rep = + emitX86Select(Builder, CI->getArgOperand(2), Rep, CI->getArgOperand(1)); + } else if (Name.starts_with("avx512.kunpck")) { + unsigned NumElts = CI->getType()->getScalarSizeInBits(); + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), NumElts); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), NumElts); + int Indices[64]; + for (unsigned i = 0; i != NumElts; ++i) + Indices[i] = i; + + // First extract half of each vector. This gives better codegen than + // doing it in a single shuffle. + LHS = Builder.CreateShuffleVector(LHS, LHS, ArrayRef(Indices, NumElts / 2)); + RHS = Builder.CreateShuffleVector(RHS, RHS, ArrayRef(Indices, NumElts / 2)); + // Concat the vectors. + // NOTE: Operands have to be swapped to match intrinsic definition. + Rep = Builder.CreateShuffleVector(RHS, LHS, ArrayRef(Indices, NumElts)); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.kand.w") { + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); + Rep = Builder.CreateAnd(LHS, RHS); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.kandn.w") { + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); + LHS = Builder.CreateNot(LHS); + Rep = Builder.CreateAnd(LHS, RHS); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.kor.w") { + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); + Rep = Builder.CreateOr(LHS, RHS); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.kxor.w") { + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); + Rep = Builder.CreateXor(LHS, RHS); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.kxnor.w") { + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); + LHS = Builder.CreateNot(LHS); + Rep = Builder.CreateXor(LHS, RHS); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.knot.w") { + Rep = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Rep = Builder.CreateNot(Rep); + Rep = Builder.CreateBitCast(Rep, CI->getType()); + } else if (Name == "avx512.kortestz.w" || Name == "avx512.kortestc.w") { + Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); + Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); + Rep = Builder.CreateOr(LHS, RHS); + Rep = Builder.CreateBitCast(Rep, Builder.getInt16Ty()); + Value *C; + if (Name[14] == 'c') + C = ConstantInt::getAllOnesValue(Builder.getInt16Ty()); + else + C = ConstantInt::getNullValue(Builder.getInt16Ty()); + Rep = Builder.CreateICmpEQ(Rep, C); + Rep = Builder.CreateZExt(Rep, Builder.getInt32Ty()); + } else if (Name == "sse.add.ss" || Name == "sse2.add.sd" || + Name == "sse.sub.ss" || Name == "sse2.sub.sd" || + Name == "sse.mul.ss" || Name == "sse2.mul.sd" || + Name == "sse.div.ss" || Name == "sse2.div.sd") { + Type *I32Ty = Type::getInt32Ty(C); + Value *Elt0 = Builder.CreateExtractElement(CI->getArgOperand(0), + ConstantInt::get(I32Ty, 0)); + Value *Elt1 = Builder.CreateExtractElement(CI->getArgOperand(1), + ConstantInt::get(I32Ty, 0)); + Value *EltOp; + if (Name.contains(".add.")) + EltOp = Builder.CreateFAdd(Elt0, Elt1); + else if (Name.contains(".sub.")) + EltOp = Builder.CreateFSub(Elt0, Elt1); + else if (Name.contains(".mul.")) + EltOp = Builder.CreateFMul(Elt0, Elt1); + else + EltOp = Builder.CreateFDiv(Elt0, Elt1); + Rep = Builder.CreateInsertElement(CI->getArgOperand(0), EltOp, + ConstantInt::get(I32Ty, 0)); + } else if (Name.starts_with("avx512.mask.pcmp")) { + // "avx512.mask.pcmpeq." or "avx512.mask.pcmpgt." + bool CmpEq = Name[16] == 'e'; + Rep = upgradeMaskedCompare(Builder, *CI, CmpEq ? 0 : 6, true); + } else if (Name.starts_with("avx512.mask.vpshufbitqmb.")) { + Type *OpTy = CI->getArgOperand(0)->getType(); + unsigned VecWidth = OpTy->getPrimitiveSizeInBits(); + Intrinsic::ID IID; + switch (VecWidth) { + default: + llvm_unreachable("Unexpected intrinsic"); + case 128: + IID = Intrinsic::x86_avx512_vpshufbitqmb_128; + break; + case 256: + IID = Intrinsic::x86_avx512_vpshufbitqmb_256; + break; + case 512: + IID = Intrinsic::x86_avx512_vpshufbitqmb_512; + break; + } + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getOperand(0), CI->getArgOperand(1)}); + Rep = applyX86MaskOn1BitsVec(Builder, Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.fpclass.p")) { + Type *OpTy = CI->getArgOperand(0)->getType(); + unsigned VecWidth = OpTy->getPrimitiveSizeInBits(); + unsigned EltWidth = OpTy->getScalarSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && EltWidth == 32) + IID = Intrinsic::x86_avx512_fpclass_ps_128; + else if (VecWidth == 256 && EltWidth == 32) + IID = Intrinsic::x86_avx512_fpclass_ps_256; + else if (VecWidth == 512 && EltWidth == 32) + IID = Intrinsic::x86_avx512_fpclass_ps_512; + else if (VecWidth == 128 && EltWidth == 64) + IID = Intrinsic::x86_avx512_fpclass_pd_128; + else if (VecWidth == 256 && EltWidth == 64) + IID = Intrinsic::x86_avx512_fpclass_pd_256; + else if (VecWidth == 512 && EltWidth == 64) + IID = Intrinsic::x86_avx512_fpclass_pd_512; + else + llvm_unreachable("Unexpected intrinsic"); + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getOperand(0), CI->getArgOperand(1)}); + Rep = applyX86MaskOn1BitsVec(Builder, Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.cmp.p")) { + SmallVector Args(CI->args()); + Type *OpTy = Args[0]->getType(); + unsigned VecWidth = OpTy->getPrimitiveSizeInBits(); + unsigned EltWidth = OpTy->getScalarSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && EltWidth == 32) + IID = Intrinsic::x86_avx512_mask_cmp_ps_128; + else if (VecWidth == 256 && EltWidth == 32) + IID = Intrinsic::x86_avx512_mask_cmp_ps_256; + else if (VecWidth == 512 && EltWidth == 32) + IID = Intrinsic::x86_avx512_mask_cmp_ps_512; + else if (VecWidth == 128 && EltWidth == 64) + IID = Intrinsic::x86_avx512_mask_cmp_pd_128; + else if (VecWidth == 256 && EltWidth == 64) + IID = Intrinsic::x86_avx512_mask_cmp_pd_256; + else if (VecWidth == 512 && EltWidth == 64) + IID = Intrinsic::x86_avx512_mask_cmp_pd_512; + else + llvm_unreachable("Unexpected intrinsic"); + + Value *Mask = Constant::getAllOnesValue(CI->getType()); + if (VecWidth == 512) + std::swap(Mask, Args.back()); + Args.push_back(Mask); + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), + Args); + } else if (Name.starts_with("avx512.mask.cmp.")) { + // Integer compare intrinsics. + unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); + Rep = upgradeMaskedCompare(Builder, *CI, Imm, true); + } else if (Name.starts_with("avx512.mask.ucmp.")) { + unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); + Rep = upgradeMaskedCompare(Builder, *CI, Imm, false); + } else if (Name.starts_with("avx512.cvtb2mask.") || + Name.starts_with("avx512.cvtw2mask.") || + Name.starts_with("avx512.cvtd2mask.") || + Name.starts_with("avx512.cvtq2mask.")) { + Value *Op = CI->getArgOperand(0); + Value *Zero = llvm::Constant::getNullValue(Op->getType()); + Rep = Builder.CreateICmp(ICmpInst::ICMP_SLT, Op, Zero); + Rep = applyX86MaskOn1BitsVec(Builder, Rep, nullptr); + } else if (Name == "ssse3.pabs.b.128" || Name == "ssse3.pabs.w.128" || + Name == "ssse3.pabs.d.128" || Name.starts_with("avx2.pabs") || + Name.starts_with("avx512.mask.pabs")) { + Rep = upgradeAbs(Builder, *CI); + } else if (Name == "sse41.pmaxsb" || Name == "sse2.pmaxs.w" || + Name == "sse41.pmaxsd" || Name.starts_with("avx2.pmaxs") || + Name.starts_with("avx512.mask.pmaxs")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::smax); + } else if (Name == "sse2.pmaxu.b" || Name == "sse41.pmaxuw" || + Name == "sse41.pmaxud" || Name.starts_with("avx2.pmaxu") || + Name.starts_with("avx512.mask.pmaxu")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::umax); + } else if (Name == "sse41.pminsb" || Name == "sse2.pmins.w" || + Name == "sse41.pminsd" || Name.starts_with("avx2.pmins") || + Name.starts_with("avx512.mask.pmins")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::smin); + } else if (Name == "sse2.pminu.b" || Name == "sse41.pminuw" || + Name == "sse41.pminud" || Name.starts_with("avx2.pminu") || + Name.starts_with("avx512.mask.pminu")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::umin); + } else if (Name == "sse2.pmulu.dq" || Name == "avx2.pmulu.dq" || + Name == "avx512.pmulu.dq.512" || + Name.starts_with("avx512.mask.pmulu.dq.")) { + Rep = upgradePMULDQ(Builder, *CI, /*Signed*/ false); + } else if (Name == "sse41.pmuldq" || Name == "avx2.pmul.dq" || + Name == "avx512.pmul.dq.512" || + Name.starts_with("avx512.mask.pmul.dq.")) { + Rep = upgradePMULDQ(Builder, *CI, /*Signed*/ true); + } else if (Name == "sse.cvtsi2ss" || Name == "sse2.cvtsi2sd" || + Name == "sse.cvtsi642ss" || Name == "sse2.cvtsi642sd") { + Rep = + Builder.CreateSIToFP(CI->getArgOperand(1), + cast(CI->getType())->getElementType()); + Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); + } else if (Name == "avx512.cvtusi2sd") { + Rep = + Builder.CreateUIToFP(CI->getArgOperand(1), + cast(CI->getType())->getElementType()); + Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); + } else if (Name == "sse2.cvtss2sd") { + Rep = Builder.CreateExtractElement(CI->getArgOperand(1), (uint64_t)0); + Rep = Builder.CreateFPExt( + Rep, cast(CI->getType())->getElementType()); + Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); + } else if (Name == "sse2.cvtdq2pd" || Name == "sse2.cvtdq2ps" || + Name == "avx.cvtdq2.pd.256" || Name == "avx.cvtdq2.ps.256" || + Name.starts_with("avx512.mask.cvtdq2pd.") || + Name.starts_with("avx512.mask.cvtudq2pd.") || + Name.starts_with("avx512.mask.cvtdq2ps.") || + Name.starts_with("avx512.mask.cvtudq2ps.") || + Name.starts_with("avx512.mask.cvtqq2pd.") || + Name.starts_with("avx512.mask.cvtuqq2pd.") || + Name == "avx512.mask.cvtqq2ps.256" || + Name == "avx512.mask.cvtqq2ps.512" || + Name == "avx512.mask.cvtuqq2ps.256" || + Name == "avx512.mask.cvtuqq2ps.512" || Name == "sse2.cvtps2pd" || + Name == "avx.cvt.ps2.pd.256" || + Name == "avx512.mask.cvtps2pd.128" || + Name == "avx512.mask.cvtps2pd.256") { + auto *DstTy = cast(CI->getType()); + Rep = CI->getArgOperand(0); + auto *SrcTy = cast(Rep->getType()); + + unsigned NumDstElts = DstTy->getNumElements(); + if (NumDstElts < SrcTy->getNumElements()) { + assert(NumDstElts == 2 && "Unexpected vector size"); + Rep = Builder.CreateShuffleVector(Rep, Rep, ArrayRef{0, 1}); + } + + bool IsPS2PD = SrcTy->getElementType()->isFloatTy(); + bool IsUnsigned = Name.contains("cvtu"); + if (IsPS2PD) + Rep = Builder.CreateFPExt(Rep, DstTy, "cvtps2pd"); + else if (CI->arg_size() == 4 && + (!isa(CI->getArgOperand(3)) || + cast(CI->getArgOperand(3))->getZExtValue() != 4)) { + Intrinsic::ID IID = IsUnsigned ? Intrinsic::x86_avx512_uitofp_round + : Intrinsic::x86_avx512_sitofp_round; + Function *F = + Intrinsic::getDeclaration(CI->getModule(), IID, {DstTy, SrcTy}); + Rep = Builder.CreateCall(F, {Rep, CI->getArgOperand(3)}); + } else { + Rep = IsUnsigned ? Builder.CreateUIToFP(Rep, DstTy, "cvt") + : Builder.CreateSIToFP(Rep, DstTy, "cvt"); + } + + if (CI->arg_size() >= 3) + Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, + CI->getArgOperand(1)); + } else if (Name.starts_with("avx512.mask.vcvtph2ps.") || + Name.starts_with("vcvtph2ps.")) { + auto *DstTy = cast(CI->getType()); + Rep = CI->getArgOperand(0); + auto *SrcTy = cast(Rep->getType()); + unsigned NumDstElts = DstTy->getNumElements(); + if (NumDstElts != SrcTy->getNumElements()) { + assert(NumDstElts == 4 && "Unexpected vector size"); + Rep = Builder.CreateShuffleVector(Rep, Rep, ArrayRef{0, 1, 2, 3}); + } + Rep = Builder.CreateBitCast( + Rep, FixedVectorType::get(Type::getHalfTy(C), NumDstElts)); + Rep = Builder.CreateFPExt(Rep, DstTy, "cvtph2ps"); + if (CI->arg_size() >= 3) + Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, + CI->getArgOperand(1)); + } else if (Name.starts_with("avx512.mask.load")) { + // "avx512.mask.loadu." or "avx512.mask.load." + bool Aligned = Name[16] != 'u'; // "avx512.mask.loadu". + Rep = upgradeMaskedLoad(Builder, CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2), Aligned); + } else if (Name.starts_with("avx512.mask.expand.load.")) { + auto *ResultTy = cast(CI->getType()); + Type *PtrTy = ResultTy->getElementType(); + + // Cast the pointer to element type. + Value *Ptr = Builder.CreateBitCast(CI->getOperand(0), + llvm::PointerType::getUnqual(PtrTy)); + + Value *MaskVec = getX86MaskVec(Builder, CI->getArgOperand(2), + ResultTy->getNumElements()); + + Function *ELd = Intrinsic::getDeclaration( + F->getParent(), Intrinsic::masked_expandload, ResultTy); + Rep = Builder.CreateCall(ELd, {Ptr, MaskVec, CI->getOperand(1)}); + } else if (Name.starts_with("avx512.mask.compress.store.")) { + auto *ResultTy = cast(CI->getArgOperand(1)->getType()); + Type *PtrTy = ResultTy->getElementType(); + + // Cast the pointer to element type. + Value *Ptr = Builder.CreateBitCast(CI->getOperand(0), + llvm::PointerType::getUnqual(PtrTy)); + + Value *MaskVec = + getX86MaskVec(Builder, CI->getArgOperand(2), + cast(ResultTy)->getNumElements()); + + Function *CSt = Intrinsic::getDeclaration( + F->getParent(), Intrinsic::masked_compressstore, ResultTy); + Rep = Builder.CreateCall(CSt, {CI->getArgOperand(1), Ptr, MaskVec}); + } else if (Name.starts_with("avx512.mask.compress.") || + Name.starts_with("avx512.mask.expand.")) { + auto *ResultTy = cast(CI->getType()); + + Value *MaskVec = getX86MaskVec(Builder, CI->getArgOperand(2), + ResultTy->getNumElements()); + + bool IsCompress = Name[12] == 'c'; + Intrinsic::ID IID = IsCompress ? Intrinsic::x86_avx512_mask_compress + : Intrinsic::x86_avx512_mask_expand; + Function *Intr = Intrinsic::getDeclaration(F->getParent(), IID, ResultTy); + Rep = Builder.CreateCall(Intr, + {CI->getOperand(0), CI->getOperand(1), MaskVec}); + } else if (Name.starts_with("xop.vpcom")) { + bool IsSigned; + if (Name.ends_with("ub") || Name.ends_with("uw") || Name.ends_with("ud") || + Name.ends_with("uq")) + IsSigned = false; + else if (Name.ends_with("b") || Name.ends_with("w") || + Name.ends_with("d") || Name.ends_with("q")) + IsSigned = true; + else + llvm_unreachable("Unknown suffix"); + + unsigned Imm; + if (CI->arg_size() == 3) { + Imm = cast(CI->getArgOperand(2))->getZExtValue(); + } else { + Name = Name.substr(9); // strip off "xop.vpcom" + if (Name.starts_with("lt")) + Imm = 0; + else if (Name.starts_with("le")) + Imm = 1; + else if (Name.starts_with("gt")) + Imm = 2; + else if (Name.starts_with("ge")) + Imm = 3; + else if (Name.starts_with("eq")) + Imm = 4; + else if (Name.starts_with("ne")) + Imm = 5; + else if (Name.starts_with("false")) + Imm = 6; + else if (Name.starts_with("true")) + Imm = 7; + else + llvm_unreachable("Unknown condition"); + } + + Rep = upgradeX86vpcom(Builder, *CI, Imm, IsSigned); + } else if (Name.starts_with("xop.vpcmov")) { + Value *Sel = CI->getArgOperand(2); + Value *NotSel = Builder.CreateNot(Sel); + Value *Sel0 = Builder.CreateAnd(CI->getArgOperand(0), Sel); + Value *Sel1 = Builder.CreateAnd(CI->getArgOperand(1), NotSel); + Rep = Builder.CreateOr(Sel0, Sel1); + } else if (Name.starts_with("xop.vprot") || Name.starts_with("avx512.prol") || + Name.starts_with("avx512.mask.prol")) { + Rep = upgradeX86Rotate(Builder, *CI, false); + } else if (Name.starts_with("avx512.pror") || + Name.starts_with("avx512.mask.pror")) { + Rep = upgradeX86Rotate(Builder, *CI, true); + } else if (Name.starts_with("avx512.vpshld.") || + Name.starts_with("avx512.mask.vpshld") || + Name.starts_with("avx512.maskz.vpshld")) { + bool ZeroMask = Name[11] == 'z'; + Rep = upgradeX86ConcatShift(Builder, *CI, false, ZeroMask); + } else if (Name.starts_with("avx512.vpshrd.") || + Name.starts_with("avx512.mask.vpshrd") || + Name.starts_with("avx512.maskz.vpshrd")) { + bool ZeroMask = Name[11] == 'z'; + Rep = upgradeX86ConcatShift(Builder, *CI, true, ZeroMask); + } else if (Name == "sse42.crc32.64.8") { + Function *CRC32 = Intrinsic::getDeclaration( + F->getParent(), Intrinsic::x86_sse42_crc32_32_8); + Value *Trunc0 = + Builder.CreateTrunc(CI->getArgOperand(0), Type::getInt32Ty(C)); + Rep = Builder.CreateCall(CRC32, {Trunc0, CI->getArgOperand(1)}); + Rep = Builder.CreateZExt(Rep, CI->getType(), ""); + } else if (Name.starts_with("avx.vbroadcast.s") || + Name.starts_with("avx512.vbroadcast.s")) { + // Replace broadcasts with a series of insertelements. + auto *VecTy = cast(CI->getType()); + Type *EltTy = VecTy->getElementType(); + unsigned EltNum = VecTy->getNumElements(); + Value *Load = Builder.CreateLoad(EltTy, CI->getArgOperand(0)); + Type *I32Ty = Type::getInt32Ty(C); + Rep = PoisonValue::get(VecTy); + for (unsigned I = 0; I < EltNum; ++I) + Rep = Builder.CreateInsertElement(Rep, Load, ConstantInt::get(I32Ty, I)); + } else if (Name.starts_with("sse41.pmovsx") || + Name.starts_with("sse41.pmovzx") || + Name.starts_with("avx2.pmovsx") || + Name.starts_with("avx2.pmovzx") || + Name.starts_with("avx512.mask.pmovsx") || + Name.starts_with("avx512.mask.pmovzx")) { + auto *DstTy = cast(CI->getType()); + unsigned NumDstElts = DstTy->getNumElements(); + + // Extract a subvector of the first NumDstElts lanes and sign/zero extend. + SmallVector ShuffleMask(NumDstElts); + for (unsigned i = 0; i != NumDstElts; ++i) + ShuffleMask[i] = i; + + Value *SV = Builder.CreateShuffleVector(CI->getArgOperand(0), ShuffleMask); + + bool DoSext = Name.contains("pmovsx"); + Rep = + DoSext ? Builder.CreateSExt(SV, DstTy) : Builder.CreateZExt(SV, DstTy); + // If there are 3 arguments, it's a masked intrinsic so we need a select. + if (CI->arg_size() == 3) + Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, + CI->getArgOperand(1)); + } else if (Name == "avx512.mask.pmov.qd.256" || + Name == "avx512.mask.pmov.qd.512" || + Name == "avx512.mask.pmov.wb.256" || + Name == "avx512.mask.pmov.wb.512") { + Type *Ty = CI->getArgOperand(1)->getType(); + Rep = Builder.CreateTrunc(CI->getArgOperand(0), Ty); + Rep = + emitX86Select(Builder, CI->getArgOperand(2), Rep, CI->getArgOperand(1)); + } else if (Name.starts_with("avx.vbroadcastf128") || + Name == "avx2.vbroadcasti128") { + // Replace vbroadcastf128/vbroadcasti128 with a vector load+shuffle. + Type *EltTy = cast(CI->getType())->getElementType(); + unsigned NumSrcElts = 128 / EltTy->getPrimitiveSizeInBits(); + auto *VT = FixedVectorType::get(EltTy, NumSrcElts); + Value *Op = Builder.CreatePointerCast(CI->getArgOperand(0), + PointerType::getUnqual(VT)); + Value *Load = Builder.CreateAlignedLoad(VT, Op, Align(1)); + if (NumSrcElts == 2) + Rep = Builder.CreateShuffleVector(Load, ArrayRef{0, 1, 0, 1}); + else + Rep = Builder.CreateShuffleVector(Load, + ArrayRef{0, 1, 2, 3, 0, 1, 2, 3}); + } else if (Name.starts_with("avx512.mask.shuf.i") || + Name.starts_with("avx512.mask.shuf.f")) { + unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); + Type *VT = CI->getType(); + unsigned NumLanes = VT->getPrimitiveSizeInBits() / 128; + unsigned NumElementsInLane = 128 / VT->getScalarSizeInBits(); + unsigned ControlBitsMask = NumLanes - 1; + unsigned NumControlBits = NumLanes / 2; + SmallVector ShuffleMask(0); + + for (unsigned l = 0; l != NumLanes; ++l) { + unsigned LaneMask = (Imm >> (l * NumControlBits)) & ControlBitsMask; + // We actually need the other source. + if (l >= NumLanes / 2) + LaneMask += NumLanes; + for (unsigned i = 0; i != NumElementsInLane; ++i) + ShuffleMask.push_back(LaneMask * NumElementsInLane + i); + } + Rep = Builder.CreateShuffleVector(CI->getArgOperand(0), + CI->getArgOperand(1), ShuffleMask); + Rep = + emitX86Select(Builder, CI->getArgOperand(4), Rep, CI->getArgOperand(3)); + } else if (Name.starts_with("avx512.mask.broadcastf") || + Name.starts_with("avx512.mask.broadcasti")) { + unsigned NumSrcElts = cast(CI->getArgOperand(0)->getType()) + ->getNumElements(); + unsigned NumDstElts = + cast(CI->getType())->getNumElements(); + + SmallVector ShuffleMask(NumDstElts); + for (unsigned i = 0; i != NumDstElts; ++i) + ShuffleMask[i] = i % NumSrcElts; + + Rep = Builder.CreateShuffleVector(CI->getArgOperand(0), + CI->getArgOperand(0), ShuffleMask); + Rep = + emitX86Select(Builder, CI->getArgOperand(2), Rep, CI->getArgOperand(1)); + } else if (Name.starts_with("avx2.pbroadcast") || + Name.starts_with("avx2.vbroadcast") || + Name.starts_with("avx512.pbroadcast") || + Name.starts_with("avx512.mask.broadcast.s")) { + // Replace vp?broadcasts with a vector shuffle. + Value *Op = CI->getArgOperand(0); + ElementCount EC = cast(CI->getType())->getElementCount(); + Type *MaskTy = VectorType::get(Type::getInt32Ty(C), EC); + SmallVector M; + ShuffleVectorInst::getShuffleMask(Constant::getNullValue(MaskTy), M); + Rep = Builder.CreateShuffleVector(Op, M); + + if (CI->arg_size() == 3) + Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, + CI->getArgOperand(1)); + } else if (Name.starts_with("sse2.padds.") || + Name.starts_with("avx2.padds.") || + Name.starts_with("avx512.padds.") || + Name.starts_with("avx512.mask.padds.")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::sadd_sat); + } else if (Name.starts_with("sse2.psubs.") || + Name.starts_with("avx2.psubs.") || + Name.starts_with("avx512.psubs.") || + Name.starts_with("avx512.mask.psubs.")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::ssub_sat); + } else if (Name.starts_with("sse2.paddus.") || + Name.starts_with("avx2.paddus.") || + Name.starts_with("avx512.mask.paddus.")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::uadd_sat); + } else if (Name.starts_with("sse2.psubus.") || + Name.starts_with("avx2.psubus.") || + Name.starts_with("avx512.mask.psubus.")) { + Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::usub_sat); + } else if (Name.starts_with("avx512.mask.palignr.")) { + Rep = upgradeX86ALIGNIntrinsics(Builder, CI->getArgOperand(0), + CI->getArgOperand(1), CI->getArgOperand(2), + CI->getArgOperand(3), CI->getArgOperand(4), + false); + } else if (Name.starts_with("avx512.mask.valign.")) { + Rep = upgradeX86ALIGNIntrinsics( + Builder, CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2), CI->getArgOperand(3), CI->getArgOperand(4), true); + } else if (Name == "sse2.psll.dq" || Name == "avx2.psll.dq") { + // 128/256-bit shift left specified in bits. + unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); + Rep = upgradeX86PSLLDQIntrinsics(Builder, CI->getArgOperand(0), + Shift / 8); // Shift is in bits. + } else if (Name == "sse2.psrl.dq" || Name == "avx2.psrl.dq") { + // 128/256-bit shift right specified in bits. + unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); + Rep = upgradeX86PSRLDQIntrinsics(Builder, CI->getArgOperand(0), + Shift / 8); // Shift is in bits. + } else if (Name == "sse2.psll.dq.bs" || Name == "avx2.psll.dq.bs" || + Name == "avx512.psll.dq.512") { + // 128/256/512-bit shift left specified in bytes. + unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); + Rep = upgradeX86PSLLDQIntrinsics(Builder, CI->getArgOperand(0), Shift); + } else if (Name == "sse2.psrl.dq.bs" || Name == "avx2.psrl.dq.bs" || + Name == "avx512.psrl.dq.512") { + // 128/256/512-bit shift right specified in bytes. + unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); + Rep = upgradeX86PSRLDQIntrinsics(Builder, CI->getArgOperand(0), Shift); + } else if (Name == "sse41.pblendw" || Name.starts_with("sse41.blendp") || + Name.starts_with("avx.blend.p") || Name == "avx2.pblendw" || + Name.starts_with("avx2.pblendd.")) { + Value *Op0 = CI->getArgOperand(0); + Value *Op1 = CI->getArgOperand(1); + unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); + auto *VecTy = cast(CI->getType()); + unsigned NumElts = VecTy->getNumElements(); + + SmallVector Idxs(NumElts); + for (unsigned i = 0; i != NumElts; ++i) + Idxs[i] = ((Imm >> (i % 8)) & 1) ? i + NumElts : i; + + Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); + } else if (Name.starts_with("avx.vinsertf128.") || + Name == "avx2.vinserti128" || + Name.starts_with("avx512.mask.insert")) { + Value *Op0 = CI->getArgOperand(0); + Value *Op1 = CI->getArgOperand(1); + unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); + unsigned DstNumElts = + cast(CI->getType())->getNumElements(); + unsigned SrcNumElts = + cast(Op1->getType())->getNumElements(); + unsigned Scale = DstNumElts / SrcNumElts; + + // Mask off the high bits of the immediate value; hardware ignores those. + Imm = Imm % Scale; + + // Extend the second operand into a vector the size of the destination. + SmallVector Idxs(DstNumElts); + for (unsigned i = 0; i != SrcNumElts; ++i) + Idxs[i] = i; + for (unsigned i = SrcNumElts; i != DstNumElts; ++i) + Idxs[i] = SrcNumElts; + Rep = Builder.CreateShuffleVector(Op1, Idxs); + + // Insert the second operand into the first operand. + + // Note that there is no guarantee that instruction lowering will actually + // produce a vinsertf128 instruction for the created shuffles. In + // particular, the 0 immediate case involves no lane changes, so it can + // be handled as a blend. + + // Example of shuffle mask for 32-bit elements: + // Imm = 1 + // Imm = 0 + + // First fill with identify mask. + for (unsigned i = 0; i != DstNumElts; ++i) + Idxs[i] = i; + // Then replace the elements where we need to insert. + for (unsigned i = 0; i != SrcNumElts; ++i) + Idxs[i + Imm * SrcNumElts] = i + DstNumElts; + Rep = Builder.CreateShuffleVector(Op0, Rep, Idxs); + + // If the intrinsic has a mask operand, handle that. + if (CI->arg_size() == 5) + Rep = emitX86Select(Builder, CI->getArgOperand(4), Rep, + CI->getArgOperand(3)); + } else if (Name.starts_with("avx.vextractf128.") || + Name == "avx2.vextracti128" || + Name.starts_with("avx512.mask.vextract")) { + Value *Op0 = CI->getArgOperand(0); + unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); + unsigned DstNumElts = + cast(CI->getType())->getNumElements(); + unsigned SrcNumElts = + cast(Op0->getType())->getNumElements(); + unsigned Scale = SrcNumElts / DstNumElts; + + // Mask off the high bits of the immediate value; hardware ignores those. + Imm = Imm % Scale; + + // Get indexes for the subvector of the input vector. + SmallVector Idxs(DstNumElts); + for (unsigned i = 0; i != DstNumElts; ++i) { + Idxs[i] = i + (Imm * DstNumElts); + } + Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); + + // If the intrinsic has a mask operand, handle that. + if (CI->arg_size() == 4) + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, + CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.perm.df.") || + Name.starts_with("avx512.mask.perm.di.")) { + Value *Op0 = CI->getArgOperand(0); + unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); + auto *VecTy = cast(CI->getType()); + unsigned NumElts = VecTy->getNumElements(); + + SmallVector Idxs(NumElts); + for (unsigned i = 0; i != NumElts; ++i) + Idxs[i] = (i & ~0x3) + ((Imm >> (2 * (i & 0x3))) & 3); + + Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); + + if (CI->arg_size() == 4) + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, + CI->getArgOperand(2)); + } else if (Name.starts_with("avx.vperm2f128.") || Name == "avx2.vperm2i128") { + // The immediate permute control byte looks like this: + // [1:0] - select 128 bits from sources for low half of destination + // [2] - ignore + // [3] - zero low half of destination + // [5:4] - select 128 bits from sources for high half of destination + // [6] - ignore + // [7] - zero high half of destination + + uint8_t Imm = cast(CI->getArgOperand(2))->getZExtValue(); + + unsigned NumElts = cast(CI->getType())->getNumElements(); + unsigned HalfSize = NumElts / 2; + SmallVector ShuffleMask(NumElts); + + // Determine which operand(s) are actually in use for this instruction. + Value *V0 = (Imm & 0x02) ? CI->getArgOperand(1) : CI->getArgOperand(0); + Value *V1 = (Imm & 0x20) ? CI->getArgOperand(1) : CI->getArgOperand(0); + + // If needed, replace operands based on zero mask. + V0 = (Imm & 0x08) ? ConstantAggregateZero::get(CI->getType()) : V0; + V1 = (Imm & 0x80) ? ConstantAggregateZero::get(CI->getType()) : V1; + + // Permute low half of result. + unsigned StartIndex = (Imm & 0x01) ? HalfSize : 0; + for (unsigned i = 0; i < HalfSize; ++i) + ShuffleMask[i] = StartIndex + i; + + // Permute high half of result. + StartIndex = (Imm & 0x10) ? HalfSize : 0; + for (unsigned i = 0; i < HalfSize; ++i) + ShuffleMask[i + HalfSize] = NumElts + StartIndex + i; + + Rep = Builder.CreateShuffleVector(V0, V1, ShuffleMask); + + } else if (Name.starts_with("avx.vpermil.") || Name == "sse2.pshuf.d" || + Name.starts_with("avx512.mask.vpermil.p") || + Name.starts_with("avx512.mask.pshuf.d.")) { + Value *Op0 = CI->getArgOperand(0); + unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); + auto *VecTy = cast(CI->getType()); + unsigned NumElts = VecTy->getNumElements(); + // Calculate the size of each index in the immediate. + unsigned IdxSize = 64 / VecTy->getScalarSizeInBits(); + unsigned IdxMask = ((1 << IdxSize) - 1); + + SmallVector Idxs(NumElts); + // Lookup the bits for this element, wrapping around the immediate every + // 8-bits. Elements are grouped into sets of 2 or 4 elements so we need + // to offset by the first index of each group. + for (unsigned i = 0; i != NumElts; ++i) + Idxs[i] = ((Imm >> ((i * IdxSize) % 8)) & IdxMask) | (i & ~IdxMask); + + Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); + + if (CI->arg_size() == 4) + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, + CI->getArgOperand(2)); + } else if (Name == "sse2.pshufl.w" || + Name.starts_with("avx512.mask.pshufl.w.")) { + Value *Op0 = CI->getArgOperand(0); + unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); + unsigned NumElts = cast(CI->getType())->getNumElements(); + + SmallVector Idxs(NumElts); + for (unsigned l = 0; l != NumElts; l += 8) { + for (unsigned i = 0; i != 4; ++i) + Idxs[i + l] = ((Imm >> (2 * i)) & 0x3) + l; + for (unsigned i = 4; i != 8; ++i) + Idxs[i + l] = i + l; + } + + Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); + + if (CI->arg_size() == 4) + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, + CI->getArgOperand(2)); + } else if (Name == "sse2.pshufh.w" || + Name.starts_with("avx512.mask.pshufh.w.")) { + Value *Op0 = CI->getArgOperand(0); + unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); + unsigned NumElts = cast(CI->getType())->getNumElements(); + + SmallVector Idxs(NumElts); + for (unsigned l = 0; l != NumElts; l += 8) { + for (unsigned i = 0; i != 4; ++i) + Idxs[i + l] = i + l; + for (unsigned i = 0; i != 4; ++i) + Idxs[i + l + 4] = ((Imm >> (2 * i)) & 0x3) + 4 + l; + } + + Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); + + if (CI->arg_size() == 4) + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, + CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.shuf.p")) { + Value *Op0 = CI->getArgOperand(0); + Value *Op1 = CI->getArgOperand(1); + unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); + unsigned NumElts = cast(CI->getType())->getNumElements(); + + unsigned NumLaneElts = 128 / CI->getType()->getScalarSizeInBits(); + unsigned HalfLaneElts = NumLaneElts / 2; + + SmallVector Idxs(NumElts); + for (unsigned i = 0; i != NumElts; ++i) { + // Base index is the starting element of the lane. + Idxs[i] = i - (i % NumLaneElts); + // If we are half way through the lane switch to the other source. + if ((i % NumLaneElts) >= HalfLaneElts) + Idxs[i] += NumElts; + // Now select the specific element. By adding HalfLaneElts bits from + // the immediate. Wrapping around the immediate every 8-bits. + Idxs[i] += (Imm >> ((i * HalfLaneElts) % 8)) & ((1 << HalfLaneElts) - 1); + } + + Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); + + Rep = + emitX86Select(Builder, CI->getArgOperand(4), Rep, CI->getArgOperand(3)); + } else if (Name.starts_with("avx512.mask.movddup") || + Name.starts_with("avx512.mask.movshdup") || + Name.starts_with("avx512.mask.movsldup")) { + Value *Op0 = CI->getArgOperand(0); + unsigned NumElts = cast(CI->getType())->getNumElements(); + unsigned NumLaneElts = 128 / CI->getType()->getScalarSizeInBits(); + + unsigned Offset = 0; + if (Name.starts_with("avx512.mask.movshdup.")) + Offset = 1; + + SmallVector Idxs(NumElts); + for (unsigned l = 0; l != NumElts; l += NumLaneElts) + for (unsigned i = 0; i != NumLaneElts; i += 2) { + Idxs[i + l + 0] = i + l + Offset; + Idxs[i + l + 1] = i + l + Offset; + } + + Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); + + Rep = + emitX86Select(Builder, CI->getArgOperand(2), Rep, CI->getArgOperand(1)); + } else if (Name.starts_with("avx512.mask.punpckl") || + Name.starts_with("avx512.mask.unpckl.")) { + Value *Op0 = CI->getArgOperand(0); + Value *Op1 = CI->getArgOperand(1); + int NumElts = cast(CI->getType())->getNumElements(); + int NumLaneElts = 128 / CI->getType()->getScalarSizeInBits(); + + SmallVector Idxs(NumElts); + for (int l = 0; l != NumElts; l += NumLaneElts) + for (int i = 0; i != NumLaneElts; ++i) + Idxs[i + l] = l + (i / 2) + NumElts * (i % 2); + + Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); + + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.punpckh") || + Name.starts_with("avx512.mask.unpckh.")) { + Value *Op0 = CI->getArgOperand(0); + Value *Op1 = CI->getArgOperand(1); + int NumElts = cast(CI->getType())->getNumElements(); + int NumLaneElts = 128 / CI->getType()->getScalarSizeInBits(); + + SmallVector Idxs(NumElts); + for (int l = 0; l != NumElts; l += NumLaneElts) + for (int i = 0; i != NumLaneElts; ++i) + Idxs[i + l] = (NumLaneElts / 2) + l + (i / 2) + NumElts * (i % 2); + + Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); + + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.and.") || + Name.starts_with("avx512.mask.pand.")) { + VectorType *FTy = cast(CI->getType()); + VectorType *ITy = VectorType::getInteger(FTy); + Rep = Builder.CreateAnd(Builder.CreateBitCast(CI->getArgOperand(0), ITy), + Builder.CreateBitCast(CI->getArgOperand(1), ITy)); + Rep = Builder.CreateBitCast(Rep, FTy); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.andn.") || + Name.starts_with("avx512.mask.pandn.")) { + VectorType *FTy = cast(CI->getType()); + VectorType *ITy = VectorType::getInteger(FTy); + Rep = Builder.CreateNot(Builder.CreateBitCast(CI->getArgOperand(0), ITy)); + Rep = Builder.CreateAnd(Rep, + Builder.CreateBitCast(CI->getArgOperand(1), ITy)); + Rep = Builder.CreateBitCast(Rep, FTy); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.or.") || + Name.starts_with("avx512.mask.por.")) { + VectorType *FTy = cast(CI->getType()); + VectorType *ITy = VectorType::getInteger(FTy); + Rep = Builder.CreateOr(Builder.CreateBitCast(CI->getArgOperand(0), ITy), + Builder.CreateBitCast(CI->getArgOperand(1), ITy)); + Rep = Builder.CreateBitCast(Rep, FTy); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.xor.") || + Name.starts_with("avx512.mask.pxor.")) { + VectorType *FTy = cast(CI->getType()); + VectorType *ITy = VectorType::getInteger(FTy); + Rep = Builder.CreateXor(Builder.CreateBitCast(CI->getArgOperand(0), ITy), + Builder.CreateBitCast(CI->getArgOperand(1), ITy)); + Rep = Builder.CreateBitCast(Rep, FTy); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.padd.")) { + Rep = Builder.CreateAdd(CI->getArgOperand(0), CI->getArgOperand(1)); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.psub.")) { + Rep = Builder.CreateSub(CI->getArgOperand(0), CI->getArgOperand(1)); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.pmull.")) { + Rep = Builder.CreateMul(CI->getArgOperand(0), CI->getArgOperand(1)); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.add.p")) { + if (Name.ends_with(".512")) { + Intrinsic::ID IID; + if (Name[17] == 's') + IID = Intrinsic::x86_avx512_add_ps_512; + else + IID = Intrinsic::x86_avx512_add_pd_512; + + Rep = Builder.CreateCall( + Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getArgOperand(0), CI->getArgOperand(1), CI->getArgOperand(4)}); + } else { + Rep = Builder.CreateFAdd(CI->getArgOperand(0), CI->getArgOperand(1)); + } + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.div.p")) { + if (Name.ends_with(".512")) { + Intrinsic::ID IID; + if (Name[17] == 's') + IID = Intrinsic::x86_avx512_div_ps_512; + else + IID = Intrinsic::x86_avx512_div_pd_512; + + Rep = Builder.CreateCall( + Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getArgOperand(0), CI->getArgOperand(1), CI->getArgOperand(4)}); + } else { + Rep = Builder.CreateFDiv(CI->getArgOperand(0), CI->getArgOperand(1)); + } + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.mul.p")) { + if (Name.ends_with(".512")) { + Intrinsic::ID IID; + if (Name[17] == 's') + IID = Intrinsic::x86_avx512_mul_ps_512; + else + IID = Intrinsic::x86_avx512_mul_pd_512; + + Rep = Builder.CreateCall( + Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getArgOperand(0), CI->getArgOperand(1), CI->getArgOperand(4)}); + } else { + Rep = Builder.CreateFMul(CI->getArgOperand(0), CI->getArgOperand(1)); + } + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.sub.p")) { + if (Name.ends_with(".512")) { + Intrinsic::ID IID; + if (Name[17] == 's') + IID = Intrinsic::x86_avx512_sub_ps_512; + else + IID = Intrinsic::x86_avx512_sub_pd_512; + + Rep = Builder.CreateCall( + Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getArgOperand(0), CI->getArgOperand(1), CI->getArgOperand(4)}); + } else { + Rep = Builder.CreateFSub(CI->getArgOperand(0), CI->getArgOperand(1)); + } + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if ((Name.starts_with("avx512.mask.max.p") || + Name.starts_with("avx512.mask.min.p")) && + Name.drop_front(18) == ".512") { + bool IsDouble = Name[17] == 'd'; + bool IsMin = Name[13] == 'i'; + static const Intrinsic::ID MinMaxTbl[2][2] = { + {Intrinsic::x86_avx512_max_ps_512, Intrinsic::x86_avx512_max_pd_512}, + {Intrinsic::x86_avx512_min_ps_512, Intrinsic::x86_avx512_min_pd_512}}; + Intrinsic::ID IID = MinMaxTbl[IsMin][IsDouble]; + + Rep = Builder.CreateCall( + Intrinsic::getDeclaration(F->getParent(), IID), + {CI->getArgOperand(0), CI->getArgOperand(1), CI->getArgOperand(4)}); + Rep = + emitX86Select(Builder, CI->getArgOperand(3), Rep, CI->getArgOperand(2)); + } else if (Name.starts_with("avx512.mask.lzcnt.")) { + Rep = + Builder.CreateCall(Intrinsic::getDeclaration( + F->getParent(), Intrinsic::ctlz, CI->getType()), + {CI->getArgOperand(0), Builder.getInt1(false)}); + Rep = + emitX86Select(Builder, CI->getArgOperand(2), Rep, CI->getArgOperand(1)); + } else if (Name.starts_with("avx512.mask.psll")) { + bool IsImmediate = Name[16] == 'i' || (Name.size() > 18 && Name[18] == 'i'); + bool IsVariable = Name[16] == 'v'; + char Size = Name[16] == '.' ? Name[17] + : Name[17] == '.' ? Name[18] + : Name[18] == '.' ? Name[19] + : Name[20]; + + Intrinsic::ID IID; + if (IsVariable && Name[17] != '.') { + if (Size == 'd' && Name[17] == '2') // avx512.mask.psllv2.di + IID = Intrinsic::x86_avx2_psllv_q; + else if (Size == 'd' && Name[17] == '4') // avx512.mask.psllv4.di + IID = Intrinsic::x86_avx2_psllv_q_256; + else if (Size == 's' && Name[17] == '4') // avx512.mask.psllv4.si + IID = Intrinsic::x86_avx2_psllv_d; + else if (Size == 's' && Name[17] == '8') // avx512.mask.psllv8.si + IID = Intrinsic::x86_avx2_psllv_d_256; + else if (Size == 'h' && Name[17] == '8') // avx512.mask.psllv8.hi + IID = Intrinsic::x86_avx512_psllv_w_128; + else if (Size == 'h' && Name[17] == '1') // avx512.mask.psllv16.hi + IID = Intrinsic::x86_avx512_psllv_w_256; + else if (Name[17] == '3' && Name[18] == '2') // avx512.mask.psllv32hi + IID = Intrinsic::x86_avx512_psllv_w_512; + else + llvm_unreachable("Unexpected size"); + } else if (Name.ends_with(".128")) { + if (Size == 'd') // avx512.mask.psll.d.128, avx512.mask.psll.di.128 + IID = IsImmediate ? Intrinsic::x86_sse2_pslli_d + : Intrinsic::x86_sse2_psll_d; + else if (Size == 'q') // avx512.mask.psll.q.128, avx512.mask.psll.qi.128 + IID = IsImmediate ? Intrinsic::x86_sse2_pslli_q + : Intrinsic::x86_sse2_psll_q; + else if (Size == 'w') // avx512.mask.psll.w.128, avx512.mask.psll.wi.128 + IID = IsImmediate ? Intrinsic::x86_sse2_pslli_w + : Intrinsic::x86_sse2_psll_w; + else + llvm_unreachable("Unexpected size"); + } else if (Name.ends_with(".256")) { + if (Size == 'd') // avx512.mask.psll.d.256, avx512.mask.psll.di.256 + IID = IsImmediate ? Intrinsic::x86_avx2_pslli_d + : Intrinsic::x86_avx2_psll_d; + else if (Size == 'q') // avx512.mask.psll.q.256, avx512.mask.psll.qi.256 + IID = IsImmediate ? Intrinsic::x86_avx2_pslli_q + : Intrinsic::x86_avx2_psll_q; + else if (Size == 'w') // avx512.mask.psll.w.256, avx512.mask.psll.wi.256 + IID = IsImmediate ? Intrinsic::x86_avx2_pslli_w + : Intrinsic::x86_avx2_psll_w; + else + llvm_unreachable("Unexpected size"); + } else { + if (Size == 'd') // psll.di.512, pslli.d, psll.d, psllv.d.512 + IID = IsImmediate ? Intrinsic::x86_avx512_pslli_d_512 + : IsVariable ? Intrinsic::x86_avx512_psllv_d_512 + : Intrinsic::x86_avx512_psll_d_512; + else if (Size == 'q') // psll.qi.512, pslli.q, psll.q, psllv.q.512 + IID = IsImmediate ? Intrinsic::x86_avx512_pslli_q_512 + : IsVariable ? Intrinsic::x86_avx512_psllv_q_512 + : Intrinsic::x86_avx512_psll_q_512; + else if (Size == 'w') // psll.wi.512, pslli.w, psll.w + IID = IsImmediate ? Intrinsic::x86_avx512_pslli_w_512 + : Intrinsic::x86_avx512_psll_w_512; + else + llvm_unreachable("Unexpected size"); + } + + Rep = upgradeX86MaskedShift(Builder, *CI, IID); + } else if (Name.starts_with("avx512.mask.psrl")) { + bool IsImmediate = Name[16] == 'i' || (Name.size() > 18 && Name[18] == 'i'); + bool IsVariable = Name[16] == 'v'; + char Size = Name[16] == '.' ? Name[17] + : Name[17] == '.' ? Name[18] + : Name[18] == '.' ? Name[19] + : Name[20]; + + Intrinsic::ID IID; + if (IsVariable && Name[17] != '.') { + if (Size == 'd' && Name[17] == '2') // avx512.mask.psrlv2.di + IID = Intrinsic::x86_avx2_psrlv_q; + else if (Size == 'd' && Name[17] == '4') // avx512.mask.psrlv4.di + IID = Intrinsic::x86_avx2_psrlv_q_256; + else if (Size == 's' && Name[17] == '4') // avx512.mask.psrlv4.si + IID = Intrinsic::x86_avx2_psrlv_d; + else if (Size == 's' && Name[17] == '8') // avx512.mask.psrlv8.si + IID = Intrinsic::x86_avx2_psrlv_d_256; + else if (Size == 'h' && Name[17] == '8') // avx512.mask.psrlv8.hi + IID = Intrinsic::x86_avx512_psrlv_w_128; + else if (Size == 'h' && Name[17] == '1') // avx512.mask.psrlv16.hi + IID = Intrinsic::x86_avx512_psrlv_w_256; + else if (Name[17] == '3' && Name[18] == '2') // avx512.mask.psrlv32hi + IID = Intrinsic::x86_avx512_psrlv_w_512; + else + llvm_unreachable("Unexpected size"); + } else if (Name.ends_with(".128")) { + if (Size == 'd') // avx512.mask.psrl.d.128, avx512.mask.psrl.di.128 + IID = IsImmediate ? Intrinsic::x86_sse2_psrli_d + : Intrinsic::x86_sse2_psrl_d; + else if (Size == 'q') // avx512.mask.psrl.q.128, avx512.mask.psrl.qi.128 + IID = IsImmediate ? Intrinsic::x86_sse2_psrli_q + : Intrinsic::x86_sse2_psrl_q; + else if (Size == 'w') // avx512.mask.psrl.w.128, avx512.mask.psrl.wi.128 + IID = IsImmediate ? Intrinsic::x86_sse2_psrli_w + : Intrinsic::x86_sse2_psrl_w; + else + llvm_unreachable("Unexpected size"); + } else if (Name.ends_with(".256")) { + if (Size == 'd') // avx512.mask.psrl.d.256, avx512.mask.psrl.di.256 + IID = IsImmediate ? Intrinsic::x86_avx2_psrli_d + : Intrinsic::x86_avx2_psrl_d; + else if (Size == 'q') // avx512.mask.psrl.q.256, avx512.mask.psrl.qi.256 + IID = IsImmediate ? Intrinsic::x86_avx2_psrli_q + : Intrinsic::x86_avx2_psrl_q; + else if (Size == 'w') // avx512.mask.psrl.w.256, avx512.mask.psrl.wi.256 + IID = IsImmediate ? Intrinsic::x86_avx2_psrli_w + : Intrinsic::x86_avx2_psrl_w; + else + llvm_unreachable("Unexpected size"); + } else { + if (Size == 'd') // psrl.di.512, psrli.d, psrl.d, psrl.d.512 + IID = IsImmediate ? Intrinsic::x86_avx512_psrli_d_512 + : IsVariable ? Intrinsic::x86_avx512_psrlv_d_512 + : Intrinsic::x86_avx512_psrl_d_512; + else if (Size == 'q') // psrl.qi.512, psrli.q, psrl.q, psrl.q.512 + IID = IsImmediate ? Intrinsic::x86_avx512_psrli_q_512 + : IsVariable ? Intrinsic::x86_avx512_psrlv_q_512 + : Intrinsic::x86_avx512_psrl_q_512; + else if (Size == 'w') // psrl.wi.512, psrli.w, psrl.w) + IID = IsImmediate ? Intrinsic::x86_avx512_psrli_w_512 + : Intrinsic::x86_avx512_psrl_w_512; + else + llvm_unreachable("Unexpected size"); + } + + Rep = upgradeX86MaskedShift(Builder, *CI, IID); + } else if (Name.starts_with("avx512.mask.psra")) { + bool IsImmediate = Name[16] == 'i' || (Name.size() > 18 && Name[18] == 'i'); + bool IsVariable = Name[16] == 'v'; + char Size = Name[16] == '.' ? Name[17] + : Name[17] == '.' ? Name[18] + : Name[18] == '.' ? Name[19] + : Name[20]; + + Intrinsic::ID IID; + if (IsVariable && Name[17] != '.') { + if (Size == 's' && Name[17] == '4') // avx512.mask.psrav4.si + IID = Intrinsic::x86_avx2_psrav_d; + else if (Size == 's' && Name[17] == '8') // avx512.mask.psrav8.si + IID = Intrinsic::x86_avx2_psrav_d_256; + else if (Size == 'h' && Name[17] == '8') // avx512.mask.psrav8.hi + IID = Intrinsic::x86_avx512_psrav_w_128; + else if (Size == 'h' && Name[17] == '1') // avx512.mask.psrav16.hi + IID = Intrinsic::x86_avx512_psrav_w_256; + else if (Name[17] == '3' && Name[18] == '2') // avx512.mask.psrav32hi + IID = Intrinsic::x86_avx512_psrav_w_512; + else + llvm_unreachable("Unexpected size"); + } else if (Name.ends_with(".128")) { + if (Size == 'd') // avx512.mask.psra.d.128, avx512.mask.psra.di.128 + IID = IsImmediate ? Intrinsic::x86_sse2_psrai_d + : Intrinsic::x86_sse2_psra_d; + else if (Size == 'q') // avx512.mask.psra.q.128, avx512.mask.psra.qi.128 + IID = IsImmediate ? Intrinsic::x86_avx512_psrai_q_128 + : IsVariable ? Intrinsic::x86_avx512_psrav_q_128 + : Intrinsic::x86_avx512_psra_q_128; + else if (Size == 'w') // avx512.mask.psra.w.128, avx512.mask.psra.wi.128 + IID = IsImmediate ? Intrinsic::x86_sse2_psrai_w + : Intrinsic::x86_sse2_psra_w; + else + llvm_unreachable("Unexpected size"); + } else if (Name.ends_with(".256")) { + if (Size == 'd') // avx512.mask.psra.d.256, avx512.mask.psra.di.256 + IID = IsImmediate ? Intrinsic::x86_avx2_psrai_d + : Intrinsic::x86_avx2_psra_d; + else if (Size == 'q') // avx512.mask.psra.q.256, avx512.mask.psra.qi.256 + IID = IsImmediate ? Intrinsic::x86_avx512_psrai_q_256 + : IsVariable ? Intrinsic::x86_avx512_psrav_q_256 + : Intrinsic::x86_avx512_psra_q_256; + else if (Size == 'w') // avx512.mask.psra.w.256, avx512.mask.psra.wi.256 + IID = IsImmediate ? Intrinsic::x86_avx2_psrai_w + : Intrinsic::x86_avx2_psra_w; + else + llvm_unreachable("Unexpected size"); + } else { + if (Size == 'd') // psra.di.512, psrai.d, psra.d, psrav.d.512 + IID = IsImmediate ? Intrinsic::x86_avx512_psrai_d_512 + : IsVariable ? Intrinsic::x86_avx512_psrav_d_512 + : Intrinsic::x86_avx512_psra_d_512; + else if (Size == 'q') // psra.qi.512, psrai.q, psra.q + IID = IsImmediate ? Intrinsic::x86_avx512_psrai_q_512 + : IsVariable ? Intrinsic::x86_avx512_psrav_q_512 + : Intrinsic::x86_avx512_psra_q_512; + else if (Size == 'w') // psra.wi.512, psrai.w, psra.w + IID = IsImmediate ? Intrinsic::x86_avx512_psrai_w_512 + : Intrinsic::x86_avx512_psra_w_512; + else + llvm_unreachable("Unexpected size"); + } + + Rep = upgradeX86MaskedShift(Builder, *CI, IID); + } else if (Name.starts_with("avx512.mask.move.s")) { + Rep = upgradeMaskedMove(Builder, *CI); + } else if (Name.starts_with("avx512.cvtmask2")) { + Rep = upgradeMaskToInt(Builder, *CI); + } else if (Name.ends_with(".movntdqa")) { + MDNode *Node = MDNode::get( + C, ConstantAsMetadata::get(ConstantInt::get(Type::getInt32Ty(C), 1))); + + Value *Ptr = CI->getArgOperand(0); + + // Convert the type of the pointer to a pointer to the stored type. + Value *BC = Builder.CreateBitCast( + Ptr, PointerType::getUnqual(CI->getType()), "cast"); + LoadInst *LI = Builder.CreateAlignedLoad( + CI->getType(), BC, + Align(CI->getType()->getPrimitiveSizeInBits().getFixedValue() / 8)); + LI->setMetadata(LLVMContext::MD_nontemporal, Node); + Rep = LI; + } else if (Name.starts_with("fma.vfmadd.") || + Name.starts_with("fma.vfmsub.") || + Name.starts_with("fma.vfnmadd.") || + Name.starts_with("fma.vfnmsub.")) { + bool NegMul = Name[6] == 'n'; + bool NegAcc = NegMul ? Name[8] == 's' : Name[7] == 's'; + bool IsScalar = NegMul ? Name[12] == 's' : Name[11] == 's'; + + Value *Ops[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + + if (IsScalar) { + Ops[0] = Builder.CreateExtractElement(Ops[0], (uint64_t)0); + Ops[1] = Builder.CreateExtractElement(Ops[1], (uint64_t)0); + Ops[2] = Builder.CreateExtractElement(Ops[2], (uint64_t)0); + } + + if (NegMul && !IsScalar) + Ops[0] = Builder.CreateFNeg(Ops[0]); + if (NegMul && IsScalar) + Ops[1] = Builder.CreateFNeg(Ops[1]); + if (NegAcc) + Ops[2] = Builder.CreateFNeg(Ops[2]); + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), + Intrinsic::fma, + Ops[0]->getType()), + Ops); + + if (IsScalar) + Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); + } else if (Name.starts_with("fma4.vfmadd.s")) { + Value *Ops[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + + Ops[0] = Builder.CreateExtractElement(Ops[0], (uint64_t)0); + Ops[1] = Builder.CreateExtractElement(Ops[1], (uint64_t)0); + Ops[2] = Builder.CreateExtractElement(Ops[2], (uint64_t)0); + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), + Intrinsic::fma, + Ops[0]->getType()), + Ops); + + Rep = Builder.CreateInsertElement(Constant::getNullValue(CI->getType()), + Rep, (uint64_t)0); + } else if (Name.starts_with("avx512.mask.vfmadd.s") || + Name.starts_with("avx512.maskz.vfmadd.s") || + Name.starts_with("avx512.mask3.vfmadd.s") || + Name.starts_with("avx512.mask3.vfmsub.s") || + Name.starts_with("avx512.mask3.vfnmsub.s")) { + bool IsMask3 = Name[11] == '3'; + bool IsMaskZ = Name[11] == 'z'; + // Drop the "avx512.mask." to make it easier. + Name = Name.drop_front(IsMask3 || IsMaskZ ? 13 : 12); + bool NegMul = Name[2] == 'n'; + bool NegAcc = NegMul ? Name[4] == 's' : Name[3] == 's'; + + Value *A = CI->getArgOperand(0); + Value *B = CI->getArgOperand(1); + Value *C = CI->getArgOperand(2); + + if (NegMul && (IsMask3 || IsMaskZ)) + A = Builder.CreateFNeg(A); + if (NegMul && !(IsMask3 || IsMaskZ)) + B = Builder.CreateFNeg(B); + if (NegAcc) + C = Builder.CreateFNeg(C); + + A = Builder.CreateExtractElement(A, (uint64_t)0); + B = Builder.CreateExtractElement(B, (uint64_t)0); + C = Builder.CreateExtractElement(C, (uint64_t)0); + + if (!isa(CI->getArgOperand(4)) || + cast(CI->getArgOperand(4))->getZExtValue() != 4) { + Value *Ops[] = {A, B, C, CI->getArgOperand(4)}; + + Intrinsic::ID IID; + if (Name.back() == 'd') + IID = Intrinsic::x86_avx512_vfmadd_f64; + else + IID = Intrinsic::x86_avx512_vfmadd_f32; + Function *FMA = Intrinsic::getDeclaration(CI->getModule(), IID); + Rep = Builder.CreateCall(FMA, Ops); + } else { + Function *FMA = Intrinsic::getDeclaration(CI->getModule(), Intrinsic::fma, + A->getType()); + Rep = Builder.CreateCall(FMA, {A, B, C}); + } + + Value *PassThru = IsMaskZ ? Constant::getNullValue(Rep->getType()) + : IsMask3 ? C + : A; + + // For Mask3 with NegAcc, we need to create a new extractelement that + // avoids the negation above. + if (NegAcc && IsMask3) + PassThru = + Builder.CreateExtractElement(CI->getArgOperand(2), (uint64_t)0); + + Rep = emitX86ScalarSelect(Builder, CI->getArgOperand(3), Rep, PassThru); + Rep = Builder.CreateInsertElement(CI->getArgOperand(IsMask3 ? 2 : 0), Rep, + (uint64_t)0); + } else if (Name.starts_with("avx512.mask.vfmadd.p") || + Name.starts_with("avx512.mask.vfnmadd.p") || + Name.starts_with("avx512.mask.vfnmsub.p") || + Name.starts_with("avx512.mask3.vfmadd.p") || + Name.starts_with("avx512.mask3.vfmsub.p") || + Name.starts_with("avx512.mask3.vfnmsub.p") || + Name.starts_with("avx512.maskz.vfmadd.p")) { + bool IsMask3 = Name[11] == '3'; + bool IsMaskZ = Name[11] == 'z'; + // Drop the "avx512.mask." to make it easier. + Name = Name.drop_front(IsMask3 || IsMaskZ ? 13 : 12); + bool NegMul = Name[2] == 'n'; + bool NegAcc = NegMul ? Name[4] == 's' : Name[3] == 's'; + + Value *A = CI->getArgOperand(0); + Value *B = CI->getArgOperand(1); + Value *C = CI->getArgOperand(2); + + if (NegMul && (IsMask3 || IsMaskZ)) + A = Builder.CreateFNeg(A); + if (NegMul && !(IsMask3 || IsMaskZ)) + B = Builder.CreateFNeg(B); + if (NegAcc) + C = Builder.CreateFNeg(C); + + if (CI->arg_size() == 5 && + (!isa(CI->getArgOperand(4)) || + cast(CI->getArgOperand(4))->getZExtValue() != 4)) { + Intrinsic::ID IID; + // Check the character before ".512" in string. + if (Name[Name.size() - 5] == 's') + IID = Intrinsic::x86_avx512_vfmadd_ps_512; + else + IID = Intrinsic::x86_avx512_vfmadd_pd_512; + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), + {A, B, C, CI->getArgOperand(4)}); + } else { + Function *FMA = Intrinsic::getDeclaration(CI->getModule(), Intrinsic::fma, + A->getType()); + Rep = Builder.CreateCall(FMA, {A, B, C}); + } + + Value *PassThru = IsMaskZ ? llvm::Constant::getNullValue(CI->getType()) + : IsMask3 ? CI->getArgOperand(2) + : CI->getArgOperand(0); + + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); + } else if (Name.starts_with("fma.vfmsubadd.p")) { + unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); + unsigned EltWidth = CI->getType()->getScalarSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && EltWidth == 32) + IID = Intrinsic::x86_fma_vfmaddsub_ps; + else if (VecWidth == 256 && EltWidth == 32) + IID = Intrinsic::x86_fma_vfmaddsub_ps_256; + else if (VecWidth == 128 && EltWidth == 64) + IID = Intrinsic::x86_fma_vfmaddsub_pd; + else if (VecWidth == 256 && EltWidth == 64) + IID = Intrinsic::x86_fma_vfmaddsub_pd_256; + else + llvm_unreachable("Unexpected intrinsic"); + + Value *Ops[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + Ops[2] = Builder.CreateFNeg(Ops[2]); + Rep = + Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), Ops); + } else if (Name.starts_with("avx512.mask.vfmaddsub.p") || + Name.starts_with("avx512.mask3.vfmaddsub.p") || + Name.starts_with("avx512.maskz.vfmaddsub.p") || + Name.starts_with("avx512.mask3.vfmsubadd.p")) { + bool IsMask3 = Name[11] == '3'; + bool IsMaskZ = Name[11] == 'z'; + // Drop the "avx512.mask." to make it easier. + Name = Name.drop_front(IsMask3 || IsMaskZ ? 13 : 12); + bool IsSubAdd = Name[3] == 's'; + if (CI->arg_size() == 5) { + Intrinsic::ID IID; + // Check the character before ".512" in string. + if (Name[Name.size() - 5] == 's') + IID = Intrinsic::x86_avx512_vfmaddsub_ps_512; + else + IID = Intrinsic::x86_avx512_vfmaddsub_pd_512; + + Value *Ops[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2), CI->getArgOperand(4)}; + if (IsSubAdd) + Ops[2] = Builder.CreateFNeg(Ops[2]); + + Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), + Ops); + } else { + int NumElts = cast(CI->getType())->getNumElements(); + + Value *Ops[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + + Function *FMA = Intrinsic::getDeclaration(CI->getModule(), Intrinsic::fma, + Ops[0]->getType()); + Value *Odd = Builder.CreateCall(FMA, Ops); + Ops[2] = Builder.CreateFNeg(Ops[2]); + Value *Even = Builder.CreateCall(FMA, Ops); + + if (IsSubAdd) + std::swap(Even, Odd); + + SmallVector Idxs(NumElts); + for (int i = 0; i != NumElts; ++i) + Idxs[i] = i + (i % 2) * NumElts; + + Rep = Builder.CreateShuffleVector(Even, Odd, Idxs); + } + + Value *PassThru = IsMaskZ ? llvm::Constant::getNullValue(CI->getType()) + : IsMask3 ? CI->getArgOperand(2) + : CI->getArgOperand(0); + + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); + } else if (Name.starts_with("avx512.mask.pternlog.") || + Name.starts_with("avx512.maskz.pternlog.")) { + bool ZeroMask = Name[11] == 'z'; + unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); + unsigned EltWidth = CI->getType()->getScalarSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && EltWidth == 32) + IID = Intrinsic::x86_avx512_pternlog_d_128; + else if (VecWidth == 256 && EltWidth == 32) + IID = Intrinsic::x86_avx512_pternlog_d_256; + else if (VecWidth == 512 && EltWidth == 32) + IID = Intrinsic::x86_avx512_pternlog_d_512; + else if (VecWidth == 128 && EltWidth == 64) + IID = Intrinsic::x86_avx512_pternlog_q_128; + else if (VecWidth == 256 && EltWidth == 64) + IID = Intrinsic::x86_avx512_pternlog_q_256; + else if (VecWidth == 512 && EltWidth == 64) + IID = Intrinsic::x86_avx512_pternlog_q_512; + else + llvm_unreachable("Unexpected intrinsic"); + + Value *Args[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2), CI->getArgOperand(3)}; + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), + Args); + Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) + : CI->getArgOperand(0); + Rep = emitX86Select(Builder, CI->getArgOperand(4), Rep, PassThru); + } else if (Name.starts_with("avx512.mask.vpmadd52") || + Name.starts_with("avx512.maskz.vpmadd52")) { + bool ZeroMask = Name[11] == 'z'; + bool High = Name[20] == 'h' || Name[21] == 'h'; + unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && !High) + IID = Intrinsic::x86_avx512_vpmadd52l_uq_128; + else if (VecWidth == 256 && !High) + IID = Intrinsic::x86_avx512_vpmadd52l_uq_256; + else if (VecWidth == 512 && !High) + IID = Intrinsic::x86_avx512_vpmadd52l_uq_512; + else if (VecWidth == 128 && High) + IID = Intrinsic::x86_avx512_vpmadd52h_uq_128; + else if (VecWidth == 256 && High) + IID = Intrinsic::x86_avx512_vpmadd52h_uq_256; + else if (VecWidth == 512 && High) + IID = Intrinsic::x86_avx512_vpmadd52h_uq_512; + else + llvm_unreachable("Unexpected intrinsic"); + + Value *Args[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), + Args); + Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) + : CI->getArgOperand(0); + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); + } else if (Name.starts_with("avx512.mask.vpermi2var.") || + Name.starts_with("avx512.mask.vpermt2var.") || + Name.starts_with("avx512.maskz.vpermt2var.")) { + bool ZeroMask = Name[11] == 'z'; + bool IndexForm = Name[17] == 'i'; + Rep = upgradeX86VPERMT2Intrinsics(Builder, *CI, ZeroMask, IndexForm); + } else if (Name.starts_with("avx512.mask.vpdpbusd.") || + Name.starts_with("avx512.maskz.vpdpbusd.") || + Name.starts_with("avx512.mask.vpdpbusds.") || + Name.starts_with("avx512.maskz.vpdpbusds.")) { + bool ZeroMask = Name[11] == 'z'; + bool IsSaturating = Name[ZeroMask ? 21 : 20] == 's'; + unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && !IsSaturating) + IID = Intrinsic::x86_avx512_vpdpbusd_128; + else if (VecWidth == 256 && !IsSaturating) + IID = Intrinsic::x86_avx512_vpdpbusd_256; + else if (VecWidth == 512 && !IsSaturating) + IID = Intrinsic::x86_avx512_vpdpbusd_512; + else if (VecWidth == 128 && IsSaturating) + IID = Intrinsic::x86_avx512_vpdpbusds_128; + else if (VecWidth == 256 && IsSaturating) + IID = Intrinsic::x86_avx512_vpdpbusds_256; + else if (VecWidth == 512 && IsSaturating) + IID = Intrinsic::x86_avx512_vpdpbusds_512; + else + llvm_unreachable("Unexpected intrinsic"); + + Value *Args[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), + Args); + Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) + : CI->getArgOperand(0); + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); + } else if (Name.starts_with("avx512.mask.vpdpwssd.") || + Name.starts_with("avx512.maskz.vpdpwssd.") || + Name.starts_with("avx512.mask.vpdpwssds.") || + Name.starts_with("avx512.maskz.vpdpwssds.")) { + bool ZeroMask = Name[11] == 'z'; + bool IsSaturating = Name[ZeroMask ? 21 : 20] == 's'; + unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); + Intrinsic::ID IID; + if (VecWidth == 128 && !IsSaturating) + IID = Intrinsic::x86_avx512_vpdpwssd_128; + else if (VecWidth == 256 && !IsSaturating) + IID = Intrinsic::x86_avx512_vpdpwssd_256; + else if (VecWidth == 512 && !IsSaturating) + IID = Intrinsic::x86_avx512_vpdpwssd_512; + else if (VecWidth == 128 && IsSaturating) + IID = Intrinsic::x86_avx512_vpdpwssds_128; + else if (VecWidth == 256 && IsSaturating) + IID = Intrinsic::x86_avx512_vpdpwssds_256; + else if (VecWidth == 512 && IsSaturating) + IID = Intrinsic::x86_avx512_vpdpwssds_512; + else + llvm_unreachable("Unexpected intrinsic"); + + Value *Args[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), + Args); + Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) + : CI->getArgOperand(0); + Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); + } else if (Name == "addcarryx.u32" || Name == "addcarryx.u64" || + Name == "addcarry.u32" || Name == "addcarry.u64" || + Name == "subborrow.u32" || Name == "subborrow.u64") { + Intrinsic::ID IID; + if (Name[0] == 'a' && Name.back() == '2') + IID = Intrinsic::x86_addcarry_32; + else if (Name[0] == 'a' && Name.back() == '4') + IID = Intrinsic::x86_addcarry_64; + else if (Name[0] == 's' && Name.back() == '2') + IID = Intrinsic::x86_subborrow_32; + else if (Name[0] == 's' && Name.back() == '4') + IID = Intrinsic::x86_subborrow_64; + else + llvm_unreachable("Unexpected intrinsic"); + + // Make a call with 3 operands. + Value *Args[] = {CI->getArgOperand(0), CI->getArgOperand(1), + CI->getArgOperand(2)}; + Value *NewCall = Builder.CreateCall( + Intrinsic::getDeclaration(CI->getModule(), IID), Args); + + // Extract the second result and store it. + Value *Data = Builder.CreateExtractValue(NewCall, 1); + // Cast the pointer to the right type. + Value *Ptr = Builder.CreateBitCast( + CI->getArgOperand(3), llvm::PointerType::getUnqual(Data->getType())); + Builder.CreateAlignedStore(Data, Ptr, Align(1)); + // Replace the original call result with the first result of the new call. + Value *CF = Builder.CreateExtractValue(NewCall, 0); + + CI->replaceAllUsesWith(CF); + Rep = nullptr; + } else if (Name.starts_with("avx512.mask.") && + upgradeAVX512MaskToSelect(Name, Builder, *CI, Rep)) { + // Rep will be updated by the call in the condition. + } + + return Rep; +} + static Value *upgradeARMIntrinsicCall(StringRef Name, CallBase *CI, Function *F, IRBuilder<> &Builder) { if (Name == "mve.vctp64.old") { @@ -2497,1766 +4187,10 @@ void llvm::UpgradeIntrinsicCall(CallBase *CI, Function *NewFn) { bool IsARM = Name.consume_front("arm."); bool IsAMDGCN = Name.consume_front("amdgcn."); bool IsDbg = Name.consume_front("dbg."); - - if (IsX86 && Name.starts_with("sse4a.movnt.")) { - SmallVector Elts; - Elts.push_back( - ConstantAsMetadata::get(ConstantInt::get(Type::getInt32Ty(C), 1))); - MDNode *Node = MDNode::get(C, Elts); - - Value *Arg0 = CI->getArgOperand(0); - Value *Arg1 = CI->getArgOperand(1); - - // Nontemporal (unaligned) store of the 0'th element of the float/double - // vector. - Type *SrcEltTy = cast(Arg1->getType())->getElementType(); - PointerType *EltPtrTy = PointerType::getUnqual(SrcEltTy); - Value *Addr = Builder.CreateBitCast(Arg0, EltPtrTy, "cast"); - Value *Extract = - Builder.CreateExtractElement(Arg1, (uint64_t)0, "extractelement"); - - StoreInst *SI = Builder.CreateAlignedStore(Extract, Addr, Align(1)); - SI->setMetadata(LLVMContext::MD_nontemporal, Node); - - // Remove intrinsic. - CI->eraseFromParent(); - return; - } - - if (IsX86 && (Name.starts_with("avx.movnt.") || - Name.starts_with("avx512.storent."))) { - SmallVector Elts; - Elts.push_back( - ConstantAsMetadata::get(ConstantInt::get(Type::getInt32Ty(C), 1))); - MDNode *Node = MDNode::get(C, Elts); - - Value *Arg0 = CI->getArgOperand(0); - Value *Arg1 = CI->getArgOperand(1); - - // Convert the type of the pointer to a pointer to the stored type. - Value *BC = Builder.CreateBitCast(Arg0, - PointerType::getUnqual(Arg1->getType()), - "cast"); - StoreInst *SI = Builder.CreateAlignedStore( - Arg1, BC, - Align(Arg1->getType()->getPrimitiveSizeInBits().getFixedValue() / 8)); - SI->setMetadata(LLVMContext::MD_nontemporal, Node); - - // Remove intrinsic. - CI->eraseFromParent(); - return; - } - - if (IsX86 && Name == "sse2.storel.dq") { - Value *Arg0 = CI->getArgOperand(0); - Value *Arg1 = CI->getArgOperand(1); - - auto *NewVecTy = FixedVectorType::get(Type::getInt64Ty(C), 2); - Value *BC0 = Builder.CreateBitCast(Arg1, NewVecTy, "cast"); - Value *Elt = Builder.CreateExtractElement(BC0, (uint64_t)0); - Value *BC = Builder.CreateBitCast(Arg0, - PointerType::getUnqual(Elt->getType()), - "cast"); - Builder.CreateAlignedStore(Elt, BC, Align(1)); - - // Remove intrinsic. - CI->eraseFromParent(); - return; - } - - if (IsX86 && (Name.starts_with("sse.storeu.") || - Name.starts_with("sse2.storeu.") || - Name.starts_with("avx.storeu."))) { - Value *Arg0 = CI->getArgOperand(0); - Value *Arg1 = CI->getArgOperand(1); - - Arg0 = Builder.CreateBitCast(Arg0, - PointerType::getUnqual(Arg1->getType()), - "cast"); - Builder.CreateAlignedStore(Arg1, Arg0, Align(1)); - - // Remove intrinsic. - CI->eraseFromParent(); - return; - } - - if (IsX86 && Name == "avx512.mask.store.ss") { - Value *Mask = Builder.CreateAnd(CI->getArgOperand(2), Builder.getInt8(1)); - upgradeMaskedStore(Builder, CI->getArgOperand(0), CI->getArgOperand(1), - Mask, false); - - // Remove intrinsic. - CI->eraseFromParent(); - return; - } - - if (IsX86 && Name.starts_with("avx512.mask.store")) { - // "avx512.mask.storeu." or "avx512.mask.store." - bool Aligned = Name[17] != 'u'; // "avx512.mask.storeu". - upgradeMaskedStore(Builder, CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2), Aligned); - - // Remove intrinsic. - CI->eraseFromParent(); - return; - } - Value *Rep = nullptr; - // Upgrade packed integer vector compare intrinsics to compare instructions. - if (IsX86 && (Name.starts_with("sse2.pcmp") || - Name.starts_with("avx2.pcmp"))) { - // "sse2.pcpmpeq." "sse2.pcmpgt." "avx2.pcmpeq." or "avx2.pcmpgt." - bool CmpEq = Name[9] == 'e'; - Rep = Builder.CreateICmp(CmpEq ? ICmpInst::ICMP_EQ : ICmpInst::ICMP_SGT, - CI->getArgOperand(0), CI->getArgOperand(1)); - Rep = Builder.CreateSExt(Rep, CI->getType(), ""); - } else if (IsX86 && (Name.starts_with("avx512.broadcastm"))) { - Type *ExtTy = Type::getInt32Ty(C); - if (CI->getOperand(0)->getType()->isIntegerTy(8)) - ExtTy = Type::getInt64Ty(C); - unsigned NumElts = CI->getType()->getPrimitiveSizeInBits() / - ExtTy->getPrimitiveSizeInBits(); - Rep = Builder.CreateZExt(CI->getArgOperand(0), ExtTy); - Rep = Builder.CreateVectorSplat(NumElts, Rep); - } else if (IsX86 && (Name == "sse.sqrt.ss" || - Name == "sse2.sqrt.sd")) { - Value *Vec = CI->getArgOperand(0); - Value *Elt0 = Builder.CreateExtractElement(Vec, (uint64_t)0); - Function *Intr = Intrinsic::getDeclaration(F->getParent(), - Intrinsic::sqrt, Elt0->getType()); - Elt0 = Builder.CreateCall(Intr, Elt0); - Rep = Builder.CreateInsertElement(Vec, Elt0, (uint64_t)0); - } else if (IsX86 && (Name.starts_with("avx.sqrt.p") || - Name.starts_with("sse2.sqrt.p") || - Name.starts_with("sse.sqrt.p"))) { - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), - Intrinsic::sqrt, - CI->getType()), - {CI->getArgOperand(0)}); - } else if (IsX86 && (Name.starts_with("avx512.mask.sqrt.p"))) { - if (CI->arg_size() == 4 && - (!isa(CI->getArgOperand(3)) || - cast(CI->getArgOperand(3))->getZExtValue() != 4)) { - Intrinsic::ID IID = Name[18] == 's' ? Intrinsic::x86_avx512_sqrt_ps_512 - : Intrinsic::x86_avx512_sqrt_pd_512; - - Value *Args[] = { CI->getArgOperand(0), CI->getArgOperand(3) }; - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), - IID), Args); - } else { - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), - Intrinsic::sqrt, - CI->getType()), - {CI->getArgOperand(0)}); - } - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("avx512.ptestm") || - Name.starts_with("avx512.ptestnm"))) { - Value *Op0 = CI->getArgOperand(0); - Value *Op1 = CI->getArgOperand(1); - Value *Mask = CI->getArgOperand(2); - Rep = Builder.CreateAnd(Op0, Op1); - llvm::Type *Ty = Op0->getType(); - Value *Zero = llvm::Constant::getNullValue(Ty); - ICmpInst::Predicate Pred = - Name.starts_with("avx512.ptestm") ? ICmpInst::ICMP_NE : ICmpInst::ICMP_EQ; - Rep = Builder.CreateICmp(Pred, Rep, Zero); - Rep = applyX86MaskOn1BitsVec(Builder, Rep, Mask); - } else if (IsX86 && (Name.starts_with("avx512.mask.pbroadcast"))){ - unsigned NumElts = cast(CI->getArgOperand(1)->getType()) - ->getNumElements(); - Rep = Builder.CreateVectorSplat(NumElts, CI->getArgOperand(0)); - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("avx512.kunpck"))) { - unsigned NumElts = CI->getType()->getScalarSizeInBits(); - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), NumElts); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), NumElts); - int Indices[64]; - for (unsigned i = 0; i != NumElts; ++i) - Indices[i] = i; - - // First extract half of each vector. This gives better codegen than - // doing it in a single shuffle. - LHS = - Builder.CreateShuffleVector(LHS, LHS, ArrayRef(Indices, NumElts / 2)); - RHS = - Builder.CreateShuffleVector(RHS, RHS, ArrayRef(Indices, NumElts / 2)); - // Concat the vectors. - // NOTE: Operands have to be swapped to match intrinsic definition. - Rep = Builder.CreateShuffleVector(RHS, LHS, ArrayRef(Indices, NumElts)); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && Name == "avx512.kand.w") { - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); - Rep = Builder.CreateAnd(LHS, RHS); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && Name == "avx512.kandn.w") { - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); - LHS = Builder.CreateNot(LHS); - Rep = Builder.CreateAnd(LHS, RHS); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && Name == "avx512.kor.w") { - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); - Rep = Builder.CreateOr(LHS, RHS); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && Name == "avx512.kxor.w") { - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); - Rep = Builder.CreateXor(LHS, RHS); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && Name == "avx512.kxnor.w") { - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); - LHS = Builder.CreateNot(LHS); - Rep = Builder.CreateXor(LHS, RHS); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && Name == "avx512.knot.w") { - Rep = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Rep = Builder.CreateNot(Rep); - Rep = Builder.CreateBitCast(Rep, CI->getType()); - } else if (IsX86 && - (Name == "avx512.kortestz.w" || Name == "avx512.kortestc.w")) { - Value *LHS = getX86MaskVec(Builder, CI->getArgOperand(0), 16); - Value *RHS = getX86MaskVec(Builder, CI->getArgOperand(1), 16); - Rep = Builder.CreateOr(LHS, RHS); - Rep = Builder.CreateBitCast(Rep, Builder.getInt16Ty()); - Value *C; - if (Name[14] == 'c') - C = ConstantInt::getAllOnesValue(Builder.getInt16Ty()); - else - C = ConstantInt::getNullValue(Builder.getInt16Ty()); - Rep = Builder.CreateICmpEQ(Rep, C); - Rep = Builder.CreateZExt(Rep, Builder.getInt32Ty()); - } else if (IsX86 && (Name == "sse.add.ss" || Name == "sse2.add.sd" || - Name == "sse.sub.ss" || Name == "sse2.sub.sd" || - Name == "sse.mul.ss" || Name == "sse2.mul.sd" || - Name == "sse.div.ss" || Name == "sse2.div.sd")) { - Type *I32Ty = Type::getInt32Ty(C); - Value *Elt0 = Builder.CreateExtractElement(CI->getArgOperand(0), - ConstantInt::get(I32Ty, 0)); - Value *Elt1 = Builder.CreateExtractElement(CI->getArgOperand(1), - ConstantInt::get(I32Ty, 0)); - Value *EltOp; - if (Name.contains(".add.")) - EltOp = Builder.CreateFAdd(Elt0, Elt1); - else if (Name.contains(".sub.")) - EltOp = Builder.CreateFSub(Elt0, Elt1); - else if (Name.contains(".mul.")) - EltOp = Builder.CreateFMul(Elt0, Elt1); - else - EltOp = Builder.CreateFDiv(Elt0, Elt1); - Rep = Builder.CreateInsertElement(CI->getArgOperand(0), EltOp, - ConstantInt::get(I32Ty, 0)); - } else if (IsX86 && Name.starts_with("avx512.mask.pcmp")) { - // "avx512.mask.pcmpeq." or "avx512.mask.pcmpgt." - bool CmpEq = Name[16] == 'e'; - Rep = upgradeMaskedCompare(Builder, *CI, CmpEq ? 0 : 6, true); - } else if (IsX86 && Name.starts_with("avx512.mask.vpshufbitqmb.")) { - Type *OpTy = CI->getArgOperand(0)->getType(); - unsigned VecWidth = OpTy->getPrimitiveSizeInBits(); - Intrinsic::ID IID; - switch (VecWidth) { - default: llvm_unreachable("Unexpected intrinsic"); - case 128: IID = Intrinsic::x86_avx512_vpshufbitqmb_128; break; - case 256: IID = Intrinsic::x86_avx512_vpshufbitqmb_256; break; - case 512: IID = Intrinsic::x86_avx512_vpshufbitqmb_512; break; - } - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getOperand(0), CI->getArgOperand(1) }); - Rep = applyX86MaskOn1BitsVec(Builder, Rep, CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.fpclass.p")) { - Type *OpTy = CI->getArgOperand(0)->getType(); - unsigned VecWidth = OpTy->getPrimitiveSizeInBits(); - unsigned EltWidth = OpTy->getScalarSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && EltWidth == 32) - IID = Intrinsic::x86_avx512_fpclass_ps_128; - else if (VecWidth == 256 && EltWidth == 32) - IID = Intrinsic::x86_avx512_fpclass_ps_256; - else if (VecWidth == 512 && EltWidth == 32) - IID = Intrinsic::x86_avx512_fpclass_ps_512; - else if (VecWidth == 128 && EltWidth == 64) - IID = Intrinsic::x86_avx512_fpclass_pd_128; - else if (VecWidth == 256 && EltWidth == 64) - IID = Intrinsic::x86_avx512_fpclass_pd_256; - else if (VecWidth == 512 && EltWidth == 64) - IID = Intrinsic::x86_avx512_fpclass_pd_512; - else - llvm_unreachable("Unexpected intrinsic"); - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getOperand(0), CI->getArgOperand(1) }); - Rep = applyX86MaskOn1BitsVec(Builder, Rep, CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.cmp.p")) { - SmallVector Args(CI->args()); - Type *OpTy = Args[0]->getType(); - unsigned VecWidth = OpTy->getPrimitiveSizeInBits(); - unsigned EltWidth = OpTy->getScalarSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && EltWidth == 32) - IID = Intrinsic::x86_avx512_mask_cmp_ps_128; - else if (VecWidth == 256 && EltWidth == 32) - IID = Intrinsic::x86_avx512_mask_cmp_ps_256; - else if (VecWidth == 512 && EltWidth == 32) - IID = Intrinsic::x86_avx512_mask_cmp_ps_512; - else if (VecWidth == 128 && EltWidth == 64) - IID = Intrinsic::x86_avx512_mask_cmp_pd_128; - else if (VecWidth == 256 && EltWidth == 64) - IID = Intrinsic::x86_avx512_mask_cmp_pd_256; - else if (VecWidth == 512 && EltWidth == 64) - IID = Intrinsic::x86_avx512_mask_cmp_pd_512; - else - llvm_unreachable("Unexpected intrinsic"); - - Value *Mask = Constant::getAllOnesValue(CI->getType()); - if (VecWidth == 512) - std::swap(Mask, Args.back()); - Args.push_back(Mask); - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - Args); - } else if (IsX86 && Name.starts_with("avx512.mask.cmp.")) { - // Integer compare intrinsics. - unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); - Rep = upgradeMaskedCompare(Builder, *CI, Imm, true); - } else if (IsX86 && Name.starts_with("avx512.mask.ucmp.")) { - unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); - Rep = upgradeMaskedCompare(Builder, *CI, Imm, false); - } else if (IsX86 && (Name.starts_with("avx512.cvtb2mask.") || - Name.starts_with("avx512.cvtw2mask.") || - Name.starts_with("avx512.cvtd2mask.") || - Name.starts_with("avx512.cvtq2mask."))) { - Value *Op = CI->getArgOperand(0); - Value *Zero = llvm::Constant::getNullValue(Op->getType()); - Rep = Builder.CreateICmp(ICmpInst::ICMP_SLT, Op, Zero); - Rep = applyX86MaskOn1BitsVec(Builder, Rep, nullptr); - } else if(IsX86 && (Name == "ssse3.pabs.b.128" || - Name == "ssse3.pabs.w.128" || - Name == "ssse3.pabs.d.128" || - Name.starts_with("avx2.pabs") || - Name.starts_with("avx512.mask.pabs"))) { - Rep = upgradeAbs(Builder, *CI); - } else if (IsX86 && (Name == "sse41.pmaxsb" || - Name == "sse2.pmaxs.w" || - Name == "sse41.pmaxsd" || - Name.starts_with("avx2.pmaxs") || - Name.starts_with("avx512.mask.pmaxs"))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::smax); - } else if (IsX86 && (Name == "sse2.pmaxu.b" || - Name == "sse41.pmaxuw" || - Name == "sse41.pmaxud" || - Name.starts_with("avx2.pmaxu") || - Name.starts_with("avx512.mask.pmaxu"))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::umax); - } else if (IsX86 && (Name == "sse41.pminsb" || - Name == "sse2.pmins.w" || - Name == "sse41.pminsd" || - Name.starts_with("avx2.pmins") || - Name.starts_with("avx512.mask.pmins"))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::smin); - } else if (IsX86 && (Name == "sse2.pminu.b" || - Name == "sse41.pminuw" || - Name == "sse41.pminud" || - Name.starts_with("avx2.pminu") || - Name.starts_with("avx512.mask.pminu"))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::umin); - } else if (IsX86 && (Name == "sse2.pmulu.dq" || - Name == "avx2.pmulu.dq" || - Name == "avx512.pmulu.dq.512" || - Name.starts_with("avx512.mask.pmulu.dq."))) { - Rep = upgradePMULDQ(Builder, *CI, /*Signed*/false); - } else if (IsX86 && (Name == "sse41.pmuldq" || - Name == "avx2.pmul.dq" || - Name == "avx512.pmul.dq.512" || - Name.starts_with("avx512.mask.pmul.dq."))) { - Rep = upgradePMULDQ(Builder, *CI, /*Signed*/true); - } else if (IsX86 && (Name == "sse.cvtsi2ss" || - Name == "sse2.cvtsi2sd" || - Name == "sse.cvtsi642ss" || - Name == "sse2.cvtsi642sd")) { - Rep = Builder.CreateSIToFP( - CI->getArgOperand(1), - cast(CI->getType())->getElementType()); - Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); - } else if (IsX86 && Name == "avx512.cvtusi2sd") { - Rep = Builder.CreateUIToFP( - CI->getArgOperand(1), - cast(CI->getType())->getElementType()); - Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); - } else if (IsX86 && Name == "sse2.cvtss2sd") { - Rep = Builder.CreateExtractElement(CI->getArgOperand(1), (uint64_t)0); - Rep = Builder.CreateFPExt( - Rep, cast(CI->getType())->getElementType()); - Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, (uint64_t)0); - } else if (IsX86 && (Name == "sse2.cvtdq2pd" || - Name == "sse2.cvtdq2ps" || - Name == "avx.cvtdq2.pd.256" || - Name == "avx.cvtdq2.ps.256" || - Name.starts_with("avx512.mask.cvtdq2pd.") || - Name.starts_with("avx512.mask.cvtudq2pd.") || - Name.starts_with("avx512.mask.cvtdq2ps.") || - Name.starts_with("avx512.mask.cvtudq2ps.") || - Name.starts_with("avx512.mask.cvtqq2pd.") || - Name.starts_with("avx512.mask.cvtuqq2pd.") || - Name == "avx512.mask.cvtqq2ps.256" || - Name == "avx512.mask.cvtqq2ps.512" || - Name == "avx512.mask.cvtuqq2ps.256" || - Name == "avx512.mask.cvtuqq2ps.512" || - Name == "sse2.cvtps2pd" || - Name == "avx.cvt.ps2.pd.256" || - Name == "avx512.mask.cvtps2pd.128" || - Name == "avx512.mask.cvtps2pd.256")) { - auto *DstTy = cast(CI->getType()); - Rep = CI->getArgOperand(0); - auto *SrcTy = cast(Rep->getType()); - - unsigned NumDstElts = DstTy->getNumElements(); - if (NumDstElts < SrcTy->getNumElements()) { - assert(NumDstElts == 2 && "Unexpected vector size"); - Rep = Builder.CreateShuffleVector(Rep, Rep, ArrayRef{0, 1}); - } - - bool IsPS2PD = SrcTy->getElementType()->isFloatTy(); - bool IsUnsigned = Name.contains("cvtu"); - if (IsPS2PD) - Rep = Builder.CreateFPExt(Rep, DstTy, "cvtps2pd"); - else if (CI->arg_size() == 4 && - (!isa(CI->getArgOperand(3)) || - cast(CI->getArgOperand(3))->getZExtValue() != 4)) { - Intrinsic::ID IID = IsUnsigned ? Intrinsic::x86_avx512_uitofp_round - : Intrinsic::x86_avx512_sitofp_round; - Function *F = Intrinsic::getDeclaration(CI->getModule(), IID, - { DstTy, SrcTy }); - Rep = Builder.CreateCall(F, { Rep, CI->getArgOperand(3) }); - } else { - Rep = IsUnsigned ? Builder.CreateUIToFP(Rep, DstTy, "cvt") - : Builder.CreateSIToFP(Rep, DstTy, "cvt"); - } - - if (CI->arg_size() >= 3) - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("avx512.mask.vcvtph2ps.") || - Name.starts_with("vcvtph2ps."))) { - auto *DstTy = cast(CI->getType()); - Rep = CI->getArgOperand(0); - auto *SrcTy = cast(Rep->getType()); - unsigned NumDstElts = DstTy->getNumElements(); - if (NumDstElts != SrcTy->getNumElements()) { - assert(NumDstElts == 4 && "Unexpected vector size"); - Rep = Builder.CreateShuffleVector(Rep, Rep, ArrayRef{0, 1, 2, 3}); - } - Rep = Builder.CreateBitCast( - Rep, FixedVectorType::get(Type::getHalfTy(C), NumDstElts)); - Rep = Builder.CreateFPExt(Rep, DstTy, "cvtph2ps"); - if (CI->arg_size() >= 3) - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && Name.starts_with("avx512.mask.load")) { - // "avx512.mask.loadu." or "avx512.mask.load." - bool Aligned = Name[16] != 'u'; // "avx512.mask.loadu". - Rep = - upgradeMaskedLoad(Builder, CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2), Aligned); - } else if (IsX86 && Name.starts_with("avx512.mask.expand.load.")) { - auto *ResultTy = cast(CI->getType()); - Type *PtrTy = ResultTy->getElementType(); - - // Cast the pointer to element type. - Value *Ptr = Builder.CreateBitCast(CI->getOperand(0), - llvm::PointerType::getUnqual(PtrTy)); - - Value *MaskVec = getX86MaskVec(Builder, CI->getArgOperand(2), - ResultTy->getNumElements()); - - Function *ELd = Intrinsic::getDeclaration(F->getParent(), - Intrinsic::masked_expandload, - ResultTy); - Rep = Builder.CreateCall(ELd, { Ptr, MaskVec, CI->getOperand(1) }); - } else if (IsX86 && Name.starts_with("avx512.mask.compress.store.")) { - auto *ResultTy = cast(CI->getArgOperand(1)->getType()); - Type *PtrTy = ResultTy->getElementType(); - - // Cast the pointer to element type. - Value *Ptr = Builder.CreateBitCast(CI->getOperand(0), - llvm::PointerType::getUnqual(PtrTy)); - - Value *MaskVec = - getX86MaskVec(Builder, CI->getArgOperand(2), - cast(ResultTy)->getNumElements()); - - Function *CSt = Intrinsic::getDeclaration(F->getParent(), - Intrinsic::masked_compressstore, - ResultTy); - Rep = Builder.CreateCall(CSt, { CI->getArgOperand(1), Ptr, MaskVec }); - } else if (IsX86 && (Name.starts_with("avx512.mask.compress.") || - Name.starts_with("avx512.mask.expand."))) { - auto *ResultTy = cast(CI->getType()); - - Value *MaskVec = getX86MaskVec(Builder, CI->getArgOperand(2), - ResultTy->getNumElements()); - - bool IsCompress = Name[12] == 'c'; - Intrinsic::ID IID = IsCompress ? Intrinsic::x86_avx512_mask_compress - : Intrinsic::x86_avx512_mask_expand; - Function *Intr = Intrinsic::getDeclaration(F->getParent(), IID, ResultTy); - Rep = Builder.CreateCall(Intr, { CI->getOperand(0), CI->getOperand(1), - MaskVec }); - } else if (IsX86 && Name.starts_with("xop.vpcom")) { - bool IsSigned; - if (Name.ends_with("ub") || Name.ends_with("uw") || Name.ends_with("ud") || - Name.ends_with("uq")) - IsSigned = false; - else if (Name.ends_with("b") || Name.ends_with("w") || Name.ends_with("d") || - Name.ends_with("q")) - IsSigned = true; - else - llvm_unreachable("Unknown suffix"); - - unsigned Imm; - if (CI->arg_size() == 3) { - Imm = cast(CI->getArgOperand(2))->getZExtValue(); - } else { - Name = Name.substr(9); // strip off "xop.vpcom" - if (Name.starts_with("lt")) - Imm = 0; - else if (Name.starts_with("le")) - Imm = 1; - else if (Name.starts_with("gt")) - Imm = 2; - else if (Name.starts_with("ge")) - Imm = 3; - else if (Name.starts_with("eq")) - Imm = 4; - else if (Name.starts_with("ne")) - Imm = 5; - else if (Name.starts_with("false")) - Imm = 6; - else if (Name.starts_with("true")) - Imm = 7; - else - llvm_unreachable("Unknown condition"); - } - - Rep = upgradeX86vpcom(Builder, *CI, Imm, IsSigned); - } else if (IsX86 && Name.starts_with("xop.vpcmov")) { - Value *Sel = CI->getArgOperand(2); - Value *NotSel = Builder.CreateNot(Sel); - Value *Sel0 = Builder.CreateAnd(CI->getArgOperand(0), Sel); - Value *Sel1 = Builder.CreateAnd(CI->getArgOperand(1), NotSel); - Rep = Builder.CreateOr(Sel0, Sel1); - } else if (IsX86 && (Name.starts_with("xop.vprot") || - Name.starts_with("avx512.prol") || - Name.starts_with("avx512.mask.prol"))) { - Rep = upgradeX86Rotate(Builder, *CI, false); - } else if (IsX86 && (Name.starts_with("avx512.pror") || - Name.starts_with("avx512.mask.pror"))) { - Rep = upgradeX86Rotate(Builder, *CI, true); - } else if (IsX86 && (Name.starts_with("avx512.vpshld.") || - Name.starts_with("avx512.mask.vpshld") || - Name.starts_with("avx512.maskz.vpshld"))) { - bool ZeroMask = Name[11] == 'z'; - Rep = upgradeX86ConcatShift(Builder, *CI, false, ZeroMask); - } else if (IsX86 && (Name.starts_with("avx512.vpshrd.") || - Name.starts_with("avx512.mask.vpshrd") || - Name.starts_with("avx512.maskz.vpshrd"))) { - bool ZeroMask = Name[11] == 'z'; - Rep = upgradeX86ConcatShift(Builder, *CI, true, ZeroMask); - } else if (IsX86 && Name == "sse42.crc32.64.8") { - Function *CRC32 = Intrinsic::getDeclaration(F->getParent(), - Intrinsic::x86_sse42_crc32_32_8); - Value *Trunc0 = Builder.CreateTrunc(CI->getArgOperand(0), Type::getInt32Ty(C)); - Rep = Builder.CreateCall(CRC32, {Trunc0, CI->getArgOperand(1)}); - Rep = Builder.CreateZExt(Rep, CI->getType(), ""); - } else if (IsX86 && (Name.starts_with("avx.vbroadcast.s") || - Name.starts_with("avx512.vbroadcast.s"))) { - // Replace broadcasts with a series of insertelements. - auto *VecTy = cast(CI->getType()); - Type *EltTy = VecTy->getElementType(); - unsigned EltNum = VecTy->getNumElements(); - Value *Load = Builder.CreateLoad(EltTy, CI->getArgOperand(0)); - Type *I32Ty = Type::getInt32Ty(C); - Rep = PoisonValue::get(VecTy); - for (unsigned I = 0; I < EltNum; ++I) - Rep = Builder.CreateInsertElement(Rep, Load, - ConstantInt::get(I32Ty, I)); - } else if (IsX86 && (Name.starts_with("sse41.pmovsx") || - Name.starts_with("sse41.pmovzx") || - Name.starts_with("avx2.pmovsx") || - Name.starts_with("avx2.pmovzx") || - Name.starts_with("avx512.mask.pmovsx") || - Name.starts_with("avx512.mask.pmovzx"))) { - auto *DstTy = cast(CI->getType()); - unsigned NumDstElts = DstTy->getNumElements(); - - // Extract a subvector of the first NumDstElts lanes and sign/zero extend. - SmallVector ShuffleMask(NumDstElts); - for (unsigned i = 0; i != NumDstElts; ++i) - ShuffleMask[i] = i; - - Value *SV = - Builder.CreateShuffleVector(CI->getArgOperand(0), ShuffleMask); - - bool DoSext = Name.contains("pmovsx"); - Rep = DoSext ? Builder.CreateSExt(SV, DstTy) - : Builder.CreateZExt(SV, DstTy); - // If there are 3 arguments, it's a masked intrinsic so we need a select. - if (CI->arg_size() == 3) - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (Name == "avx512.mask.pmov.qd.256" || - Name == "avx512.mask.pmov.qd.512" || - Name == "avx512.mask.pmov.wb.256" || - Name == "avx512.mask.pmov.wb.512") { - Type *Ty = CI->getArgOperand(1)->getType(); - Rep = Builder.CreateTrunc(CI->getArgOperand(0), Ty); - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("avx.vbroadcastf128") || - Name == "avx2.vbroadcasti128")) { - // Replace vbroadcastf128/vbroadcasti128 with a vector load+shuffle. - Type *EltTy = cast(CI->getType())->getElementType(); - unsigned NumSrcElts = 128 / EltTy->getPrimitiveSizeInBits(); - auto *VT = FixedVectorType::get(EltTy, NumSrcElts); - Value *Op = Builder.CreatePointerCast(CI->getArgOperand(0), - PointerType::getUnqual(VT)); - Value *Load = Builder.CreateAlignedLoad(VT, Op, Align(1)); - if (NumSrcElts == 2) - Rep = Builder.CreateShuffleVector(Load, ArrayRef{0, 1, 0, 1}); - else - Rep = Builder.CreateShuffleVector( - Load, ArrayRef{0, 1, 2, 3, 0, 1, 2, 3}); - } else if (IsX86 && (Name.starts_with("avx512.mask.shuf.i") || - Name.starts_with("avx512.mask.shuf.f"))) { - unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); - Type *VT = CI->getType(); - unsigned NumLanes = VT->getPrimitiveSizeInBits() / 128; - unsigned NumElementsInLane = 128 / VT->getScalarSizeInBits(); - unsigned ControlBitsMask = NumLanes - 1; - unsigned NumControlBits = NumLanes / 2; - SmallVector ShuffleMask(0); - - for (unsigned l = 0; l != NumLanes; ++l) { - unsigned LaneMask = (Imm >> (l * NumControlBits)) & ControlBitsMask; - // We actually need the other source. - if (l >= NumLanes / 2) - LaneMask += NumLanes; - for (unsigned i = 0; i != NumElementsInLane; ++i) - ShuffleMask.push_back(LaneMask * NumElementsInLane + i); - } - Rep = Builder.CreateShuffleVector(CI->getArgOperand(0), - CI->getArgOperand(1), ShuffleMask); - Rep = emitX86Select(Builder, CI->getArgOperand(4), Rep, - CI->getArgOperand(3)); - }else if (IsX86 && (Name.starts_with("avx512.mask.broadcastf") || - Name.starts_with("avx512.mask.broadcasti"))) { - unsigned NumSrcElts = - cast(CI->getArgOperand(0)->getType()) - ->getNumElements(); - unsigned NumDstElts = - cast(CI->getType())->getNumElements(); - - SmallVector ShuffleMask(NumDstElts); - for (unsigned i = 0; i != NumDstElts; ++i) - ShuffleMask[i] = i % NumSrcElts; - - Rep = Builder.CreateShuffleVector(CI->getArgOperand(0), - CI->getArgOperand(0), - ShuffleMask); - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("avx2.pbroadcast") || - Name.starts_with("avx2.vbroadcast") || - Name.starts_with("avx512.pbroadcast") || - Name.starts_with("avx512.mask.broadcast.s"))) { - // Replace vp?broadcasts with a vector shuffle. - Value *Op = CI->getArgOperand(0); - ElementCount EC = cast(CI->getType())->getElementCount(); - Type *MaskTy = VectorType::get(Type::getInt32Ty(C), EC); - SmallVector M; - ShuffleVectorInst::getShuffleMask(Constant::getNullValue(MaskTy), M); - Rep = Builder.CreateShuffleVector(Op, M); - - if (CI->arg_size() == 3) - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("sse2.padds.") || - Name.starts_with("avx2.padds.") || - Name.starts_with("avx512.padds.") || - Name.starts_with("avx512.mask.padds."))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::sadd_sat); - } else if (IsX86 && (Name.starts_with("sse2.psubs.") || - Name.starts_with("avx2.psubs.") || - Name.starts_with("avx512.psubs.") || - Name.starts_with("avx512.mask.psubs."))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::ssub_sat); - } else if (IsX86 && (Name.starts_with("sse2.paddus.") || - Name.starts_with("avx2.paddus.") || - Name.starts_with("avx512.mask.paddus."))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::uadd_sat); - } else if (IsX86 && (Name.starts_with("sse2.psubus.") || - Name.starts_with("avx2.psubus.") || - Name.starts_with("avx512.mask.psubus."))) { - Rep = upgradeX86BinaryIntrinsics(Builder, *CI, Intrinsic::usub_sat); - } else if (IsX86 && Name.starts_with("avx512.mask.palignr.")) { - Rep = upgradeX86ALIGNIntrinsics( - Builder, CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2), CI->getArgOperand(3), CI->getArgOperand(4), - false); - } else if (IsX86 && Name.starts_with("avx512.mask.valign.")) { - Rep = upgradeX86ALIGNIntrinsics( - Builder, CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2), CI->getArgOperand(3), CI->getArgOperand(4), - true); - } else if (IsX86 && (Name == "sse2.psll.dq" || - Name == "avx2.psll.dq")) { - // 128/256-bit shift left specified in bits. - unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); - Rep = upgradeX86PSLLDQIntrinsics(Builder, CI->getArgOperand(0), - Shift / 8); // Shift is in bits. - } else if (IsX86 && (Name == "sse2.psrl.dq" || - Name == "avx2.psrl.dq")) { - // 128/256-bit shift right specified in bits. - unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); - Rep = upgradeX86PSRLDQIntrinsics(Builder, CI->getArgOperand(0), - Shift / 8); // Shift is in bits. - } else if (IsX86 && (Name == "sse2.psll.dq.bs" || - Name == "avx2.psll.dq.bs" || - Name == "avx512.psll.dq.512")) { - // 128/256/512-bit shift left specified in bytes. - unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); - Rep = upgradeX86PSLLDQIntrinsics(Builder, CI->getArgOperand(0), Shift); - } else if (IsX86 && (Name == "sse2.psrl.dq.bs" || - Name == "avx2.psrl.dq.bs" || - Name == "avx512.psrl.dq.512")) { - // 128/256/512-bit shift right specified in bytes. - unsigned Shift = cast(CI->getArgOperand(1))->getZExtValue(); - Rep = upgradeX86PSRLDQIntrinsics(Builder, CI->getArgOperand(0), Shift); - } else if (IsX86 && (Name == "sse41.pblendw" || - Name.starts_with("sse41.blendp") || - Name.starts_with("avx.blend.p") || - Name == "avx2.pblendw" || - Name.starts_with("avx2.pblendd."))) { - Value *Op0 = CI->getArgOperand(0); - Value *Op1 = CI->getArgOperand(1); - unsigned Imm = cast (CI->getArgOperand(2))->getZExtValue(); - auto *VecTy = cast(CI->getType()); - unsigned NumElts = VecTy->getNumElements(); - - SmallVector Idxs(NumElts); - for (unsigned i = 0; i != NumElts; ++i) - Idxs[i] = ((Imm >> (i%8)) & 1) ? i + NumElts : i; - - Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); - } else if (IsX86 && (Name.starts_with("avx.vinsertf128.") || - Name == "avx2.vinserti128" || - Name.starts_with("avx512.mask.insert"))) { - Value *Op0 = CI->getArgOperand(0); - Value *Op1 = CI->getArgOperand(1); - unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); - unsigned DstNumElts = - cast(CI->getType())->getNumElements(); - unsigned SrcNumElts = - cast(Op1->getType())->getNumElements(); - unsigned Scale = DstNumElts / SrcNumElts; - - // Mask off the high bits of the immediate value; hardware ignores those. - Imm = Imm % Scale; - - // Extend the second operand into a vector the size of the destination. - SmallVector Idxs(DstNumElts); - for (unsigned i = 0; i != SrcNumElts; ++i) - Idxs[i] = i; - for (unsigned i = SrcNumElts; i != DstNumElts; ++i) - Idxs[i] = SrcNumElts; - Rep = Builder.CreateShuffleVector(Op1, Idxs); - - // Insert the second operand into the first operand. - - // Note that there is no guarantee that instruction lowering will actually - // produce a vinsertf128 instruction for the created shuffles. In - // particular, the 0 immediate case involves no lane changes, so it can - // be handled as a blend. - - // Example of shuffle mask for 32-bit elements: - // Imm = 1 - // Imm = 0 - - // First fill with identify mask. - for (unsigned i = 0; i != DstNumElts; ++i) - Idxs[i] = i; - // Then replace the elements where we need to insert. - for (unsigned i = 0; i != SrcNumElts; ++i) - Idxs[i + Imm * SrcNumElts] = i + DstNumElts; - Rep = Builder.CreateShuffleVector(Op0, Rep, Idxs); - - // If the intrinsic has a mask operand, handle that. - if (CI->arg_size() == 5) - Rep = emitX86Select(Builder, CI->getArgOperand(4), Rep, - CI->getArgOperand(3)); - } else if (IsX86 && (Name.starts_with("avx.vextractf128.") || - Name == "avx2.vextracti128" || - Name.starts_with("avx512.mask.vextract"))) { - Value *Op0 = CI->getArgOperand(0); - unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); - unsigned DstNumElts = - cast(CI->getType())->getNumElements(); - unsigned SrcNumElts = - cast(Op0->getType())->getNumElements(); - unsigned Scale = SrcNumElts / DstNumElts; - - // Mask off the high bits of the immediate value; hardware ignores those. - Imm = Imm % Scale; - - // Get indexes for the subvector of the input vector. - SmallVector Idxs(DstNumElts); - for (unsigned i = 0; i != DstNumElts; ++i) { - Idxs[i] = i + (Imm * DstNumElts); - } - Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); - - // If the intrinsic has a mask operand, handle that. - if (CI->arg_size() == 4) - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (!IsX86 && Name == "stackprotectorcheck") { - Rep = nullptr; - } else if (IsX86 && (Name.starts_with("avx512.mask.perm.df.") || - Name.starts_with("avx512.mask.perm.di."))) { - Value *Op0 = CI->getArgOperand(0); - unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); - auto *VecTy = cast(CI->getType()); - unsigned NumElts = VecTy->getNumElements(); - - SmallVector Idxs(NumElts); - for (unsigned i = 0; i != NumElts; ++i) - Idxs[i] = (i & ~0x3) + ((Imm >> (2 * (i & 0x3))) & 3); - - Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); - - if (CI->arg_size() == 4) - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx.vperm2f128.") || - Name == "avx2.vperm2i128")) { - // The immediate permute control byte looks like this: - // [1:0] - select 128 bits from sources for low half of destination - // [2] - ignore - // [3] - zero low half of destination - // [5:4] - select 128 bits from sources for high half of destination - // [6] - ignore - // [7] - zero high half of destination - - uint8_t Imm = cast(CI->getArgOperand(2))->getZExtValue(); - - unsigned NumElts = cast(CI->getType())->getNumElements(); - unsigned HalfSize = NumElts / 2; - SmallVector ShuffleMask(NumElts); - - // Determine which operand(s) are actually in use for this instruction. - Value *V0 = (Imm & 0x02) ? CI->getArgOperand(1) : CI->getArgOperand(0); - Value *V1 = (Imm & 0x20) ? CI->getArgOperand(1) : CI->getArgOperand(0); - - // If needed, replace operands based on zero mask. - V0 = (Imm & 0x08) ? ConstantAggregateZero::get(CI->getType()) : V0; - V1 = (Imm & 0x80) ? ConstantAggregateZero::get(CI->getType()) : V1; - - // Permute low half of result. - unsigned StartIndex = (Imm & 0x01) ? HalfSize : 0; - for (unsigned i = 0; i < HalfSize; ++i) - ShuffleMask[i] = StartIndex + i; - - // Permute high half of result. - StartIndex = (Imm & 0x10) ? HalfSize : 0; - for (unsigned i = 0; i < HalfSize; ++i) - ShuffleMask[i + HalfSize] = NumElts + StartIndex + i; - - Rep = Builder.CreateShuffleVector(V0, V1, ShuffleMask); - - } else if (IsX86 && (Name.starts_with("avx.vpermil.") || - Name == "sse2.pshuf.d" || - Name.starts_with("avx512.mask.vpermil.p") || - Name.starts_with("avx512.mask.pshuf.d."))) { - Value *Op0 = CI->getArgOperand(0); - unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); - auto *VecTy = cast(CI->getType()); - unsigned NumElts = VecTy->getNumElements(); - // Calculate the size of each index in the immediate. - unsigned IdxSize = 64 / VecTy->getScalarSizeInBits(); - unsigned IdxMask = ((1 << IdxSize) - 1); - - SmallVector Idxs(NumElts); - // Lookup the bits for this element, wrapping around the immediate every - // 8-bits. Elements are grouped into sets of 2 or 4 elements so we need - // to offset by the first index of each group. - for (unsigned i = 0; i != NumElts; ++i) - Idxs[i] = ((Imm >> ((i * IdxSize) % 8)) & IdxMask) | (i & ~IdxMask); - - Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); - - if (CI->arg_size() == 4) - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name == "sse2.pshufl.w" || - Name.starts_with("avx512.mask.pshufl.w."))) { - Value *Op0 = CI->getArgOperand(0); - unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); - unsigned NumElts = cast(CI->getType())->getNumElements(); - - SmallVector Idxs(NumElts); - for (unsigned l = 0; l != NumElts; l += 8) { - for (unsigned i = 0; i != 4; ++i) - Idxs[i + l] = ((Imm >> (2 * i)) & 0x3) + l; - for (unsigned i = 4; i != 8; ++i) - Idxs[i + l] = i + l; - } - - Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); - - if (CI->arg_size() == 4) - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name == "sse2.pshufh.w" || - Name.starts_with("avx512.mask.pshufh.w."))) { - Value *Op0 = CI->getArgOperand(0); - unsigned Imm = cast(CI->getArgOperand(1))->getZExtValue(); - unsigned NumElts = cast(CI->getType())->getNumElements(); - - SmallVector Idxs(NumElts); - for (unsigned l = 0; l != NumElts; l += 8) { - for (unsigned i = 0; i != 4; ++i) - Idxs[i + l] = i + l; - for (unsigned i = 0; i != 4; ++i) - Idxs[i + l + 4] = ((Imm >> (2 * i)) & 0x3) + 4 + l; - } - - Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); - - if (CI->arg_size() == 4) - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.shuf.p")) { - Value *Op0 = CI->getArgOperand(0); - Value *Op1 = CI->getArgOperand(1); - unsigned Imm = cast(CI->getArgOperand(2))->getZExtValue(); - unsigned NumElts = cast(CI->getType())->getNumElements(); - - unsigned NumLaneElts = 128/CI->getType()->getScalarSizeInBits(); - unsigned HalfLaneElts = NumLaneElts / 2; - - SmallVector Idxs(NumElts); - for (unsigned i = 0; i != NumElts; ++i) { - // Base index is the starting element of the lane. - Idxs[i] = i - (i % NumLaneElts); - // If we are half way through the lane switch to the other source. - if ((i % NumLaneElts) >= HalfLaneElts) - Idxs[i] += NumElts; - // Now select the specific element. By adding HalfLaneElts bits from - // the immediate. Wrapping around the immediate every 8-bits. - Idxs[i] += (Imm >> ((i * HalfLaneElts) % 8)) & ((1 << HalfLaneElts) - 1); - } - - Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); - - Rep = emitX86Select(Builder, CI->getArgOperand(4), Rep, - CI->getArgOperand(3)); - } else if (IsX86 && (Name.starts_with("avx512.mask.movddup") || - Name.starts_with("avx512.mask.movshdup") || - Name.starts_with("avx512.mask.movsldup"))) { - Value *Op0 = CI->getArgOperand(0); - unsigned NumElts = cast(CI->getType())->getNumElements(); - unsigned NumLaneElts = 128/CI->getType()->getScalarSizeInBits(); - - unsigned Offset = 0; - if (Name.starts_with("avx512.mask.movshdup.")) - Offset = 1; - - SmallVector Idxs(NumElts); - for (unsigned l = 0; l != NumElts; l += NumLaneElts) - for (unsigned i = 0; i != NumLaneElts; i += 2) { - Idxs[i + l + 0] = i + l + Offset; - Idxs[i + l + 1] = i + l + Offset; - } - - Rep = Builder.CreateShuffleVector(Op0, Op0, Idxs); - - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && (Name.starts_with("avx512.mask.punpckl") || - Name.starts_with("avx512.mask.unpckl."))) { - Value *Op0 = CI->getArgOperand(0); - Value *Op1 = CI->getArgOperand(1); - int NumElts = cast(CI->getType())->getNumElements(); - int NumLaneElts = 128/CI->getType()->getScalarSizeInBits(); - - SmallVector Idxs(NumElts); - for (int l = 0; l != NumElts; l += NumLaneElts) - for (int i = 0; i != NumLaneElts; ++i) - Idxs[i + l] = l + (i / 2) + NumElts * (i % 2); - - Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); - - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx512.mask.punpckh") || - Name.starts_with("avx512.mask.unpckh."))) { - Value *Op0 = CI->getArgOperand(0); - Value *Op1 = CI->getArgOperand(1); - int NumElts = cast(CI->getType())->getNumElements(); - int NumLaneElts = 128/CI->getType()->getScalarSizeInBits(); - - SmallVector Idxs(NumElts); - for (int l = 0; l != NumElts; l += NumLaneElts) - for (int i = 0; i != NumLaneElts; ++i) - Idxs[i + l] = (NumLaneElts / 2) + l + (i / 2) + NumElts * (i % 2); - - Rep = Builder.CreateShuffleVector(Op0, Op1, Idxs); - - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx512.mask.and.") || - Name.starts_with("avx512.mask.pand."))) { - VectorType *FTy = cast(CI->getType()); - VectorType *ITy = VectorType::getInteger(FTy); - Rep = Builder.CreateAnd(Builder.CreateBitCast(CI->getArgOperand(0), ITy), - Builder.CreateBitCast(CI->getArgOperand(1), ITy)); - Rep = Builder.CreateBitCast(Rep, FTy); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx512.mask.andn.") || - Name.starts_with("avx512.mask.pandn."))) { - VectorType *FTy = cast(CI->getType()); - VectorType *ITy = VectorType::getInteger(FTy); - Rep = Builder.CreateNot(Builder.CreateBitCast(CI->getArgOperand(0), ITy)); - Rep = Builder.CreateAnd(Rep, - Builder.CreateBitCast(CI->getArgOperand(1), ITy)); - Rep = Builder.CreateBitCast(Rep, FTy); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx512.mask.or.") || - Name.starts_with("avx512.mask.por."))) { - VectorType *FTy = cast(CI->getType()); - VectorType *ITy = VectorType::getInteger(FTy); - Rep = Builder.CreateOr(Builder.CreateBitCast(CI->getArgOperand(0), ITy), - Builder.CreateBitCast(CI->getArgOperand(1), ITy)); - Rep = Builder.CreateBitCast(Rep, FTy); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx512.mask.xor.") || - Name.starts_with("avx512.mask.pxor."))) { - VectorType *FTy = cast(CI->getType()); - VectorType *ITy = VectorType::getInteger(FTy); - Rep = Builder.CreateXor(Builder.CreateBitCast(CI->getArgOperand(0), ITy), - Builder.CreateBitCast(CI->getArgOperand(1), ITy)); - Rep = Builder.CreateBitCast(Rep, FTy); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.padd.")) { - Rep = Builder.CreateAdd(CI->getArgOperand(0), CI->getArgOperand(1)); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.psub.")) { - Rep = Builder.CreateSub(CI->getArgOperand(0), CI->getArgOperand(1)); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.pmull.")) { - Rep = Builder.CreateMul(CI->getArgOperand(0), CI->getArgOperand(1)); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.add.p")) { - if (Name.ends_with(".512")) { - Intrinsic::ID IID; - if (Name[17] == 's') - IID = Intrinsic::x86_avx512_add_ps_512; - else - IID = Intrinsic::x86_avx512_add_pd_512; - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(4) }); - } else { - Rep = Builder.CreateFAdd(CI->getArgOperand(0), CI->getArgOperand(1)); - } - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.div.p")) { - if (Name.ends_with(".512")) { - Intrinsic::ID IID; - if (Name[17] == 's') - IID = Intrinsic::x86_avx512_div_ps_512; - else - IID = Intrinsic::x86_avx512_div_pd_512; - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(4) }); - } else { - Rep = Builder.CreateFDiv(CI->getArgOperand(0), CI->getArgOperand(1)); - } - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.mul.p")) { - if (Name.ends_with(".512")) { - Intrinsic::ID IID; - if (Name[17] == 's') - IID = Intrinsic::x86_avx512_mul_ps_512; - else - IID = Intrinsic::x86_avx512_mul_pd_512; - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(4) }); - } else { - Rep = Builder.CreateFMul(CI->getArgOperand(0), CI->getArgOperand(1)); - } - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.sub.p")) { - if (Name.ends_with(".512")) { - Intrinsic::ID IID; - if (Name[17] == 's') - IID = Intrinsic::x86_avx512_sub_ps_512; - else - IID = Intrinsic::x86_avx512_sub_pd_512; - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(4) }); - } else { - Rep = Builder.CreateFSub(CI->getArgOperand(0), CI->getArgOperand(1)); - } - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && (Name.starts_with("avx512.mask.max.p") || - Name.starts_with("avx512.mask.min.p")) && - Name.drop_front(18) == ".512") { - bool IsDouble = Name[17] == 'd'; - bool IsMin = Name[13] == 'i'; - static const Intrinsic::ID MinMaxTbl[2][2] = { - { Intrinsic::x86_avx512_max_ps_512, Intrinsic::x86_avx512_max_pd_512 }, - { Intrinsic::x86_avx512_min_ps_512, Intrinsic::x86_avx512_min_pd_512 } - }; - Intrinsic::ID IID = MinMaxTbl[IsMin][IsDouble]; - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(4) }); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, - CI->getArgOperand(2)); - } else if (IsX86 && Name.starts_with("avx512.mask.lzcnt.")) { - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), - Intrinsic::ctlz, - CI->getType()), - { CI->getArgOperand(0), Builder.getInt1(false) }); - Rep = emitX86Select(Builder, CI->getArgOperand(2), Rep, - CI->getArgOperand(1)); - } else if (IsX86 && Name.starts_with("avx512.mask.psll")) { - bool IsImmediate = Name[16] == 'i' || - (Name.size() > 18 && Name[18] == 'i'); - bool IsVariable = Name[16] == 'v'; - char Size = Name[16] == '.' ? Name[17] : - Name[17] == '.' ? Name[18] : - Name[18] == '.' ? Name[19] : - Name[20]; - - Intrinsic::ID IID; - if (IsVariable && Name[17] != '.') { - if (Size == 'd' && Name[17] == '2') // avx512.mask.psllv2.di - IID = Intrinsic::x86_avx2_psllv_q; - else if (Size == 'd' && Name[17] == '4') // avx512.mask.psllv4.di - IID = Intrinsic::x86_avx2_psllv_q_256; - else if (Size == 's' && Name[17] == '4') // avx512.mask.psllv4.si - IID = Intrinsic::x86_avx2_psllv_d; - else if (Size == 's' && Name[17] == '8') // avx512.mask.psllv8.si - IID = Intrinsic::x86_avx2_psllv_d_256; - else if (Size == 'h' && Name[17] == '8') // avx512.mask.psllv8.hi - IID = Intrinsic::x86_avx512_psllv_w_128; - else if (Size == 'h' && Name[17] == '1') // avx512.mask.psllv16.hi - IID = Intrinsic::x86_avx512_psllv_w_256; - else if (Name[17] == '3' && Name[18] == '2') // avx512.mask.psllv32hi - IID = Intrinsic::x86_avx512_psllv_w_512; - else - llvm_unreachable("Unexpected size"); - } else if (Name.ends_with(".128")) { - if (Size == 'd') // avx512.mask.psll.d.128, avx512.mask.psll.di.128 - IID = IsImmediate ? Intrinsic::x86_sse2_pslli_d - : Intrinsic::x86_sse2_psll_d; - else if (Size == 'q') // avx512.mask.psll.q.128, avx512.mask.psll.qi.128 - IID = IsImmediate ? Intrinsic::x86_sse2_pslli_q - : Intrinsic::x86_sse2_psll_q; - else if (Size == 'w') // avx512.mask.psll.w.128, avx512.mask.psll.wi.128 - IID = IsImmediate ? Intrinsic::x86_sse2_pslli_w - : Intrinsic::x86_sse2_psll_w; - else - llvm_unreachable("Unexpected size"); - } else if (Name.ends_with(".256")) { - if (Size == 'd') // avx512.mask.psll.d.256, avx512.mask.psll.di.256 - IID = IsImmediate ? Intrinsic::x86_avx2_pslli_d - : Intrinsic::x86_avx2_psll_d; - else if (Size == 'q') // avx512.mask.psll.q.256, avx512.mask.psll.qi.256 - IID = IsImmediate ? Intrinsic::x86_avx2_pslli_q - : Intrinsic::x86_avx2_psll_q; - else if (Size == 'w') // avx512.mask.psll.w.256, avx512.mask.psll.wi.256 - IID = IsImmediate ? Intrinsic::x86_avx2_pslli_w - : Intrinsic::x86_avx2_psll_w; - else - llvm_unreachable("Unexpected size"); - } else { - if (Size == 'd') // psll.di.512, pslli.d, psll.d, psllv.d.512 - IID = IsImmediate ? Intrinsic::x86_avx512_pslli_d_512 : - IsVariable ? Intrinsic::x86_avx512_psllv_d_512 : - Intrinsic::x86_avx512_psll_d_512; - else if (Size == 'q') // psll.qi.512, pslli.q, psll.q, psllv.q.512 - IID = IsImmediate ? Intrinsic::x86_avx512_pslli_q_512 : - IsVariable ? Intrinsic::x86_avx512_psllv_q_512 : - Intrinsic::x86_avx512_psll_q_512; - else if (Size == 'w') // psll.wi.512, pslli.w, psll.w - IID = IsImmediate ? Intrinsic::x86_avx512_pslli_w_512 - : Intrinsic::x86_avx512_psll_w_512; - else - llvm_unreachable("Unexpected size"); - } - - Rep = upgradeX86MaskedShift(Builder, *CI, IID); - } else if (IsX86 && Name.starts_with("avx512.mask.psrl")) { - bool IsImmediate = Name[16] == 'i' || - (Name.size() > 18 && Name[18] == 'i'); - bool IsVariable = Name[16] == 'v'; - char Size = Name[16] == '.' ? Name[17] : - Name[17] == '.' ? Name[18] : - Name[18] == '.' ? Name[19] : - Name[20]; - - Intrinsic::ID IID; - if (IsVariable && Name[17] != '.') { - if (Size == 'd' && Name[17] == '2') // avx512.mask.psrlv2.di - IID = Intrinsic::x86_avx2_psrlv_q; - else if (Size == 'd' && Name[17] == '4') // avx512.mask.psrlv4.di - IID = Intrinsic::x86_avx2_psrlv_q_256; - else if (Size == 's' && Name[17] == '4') // avx512.mask.psrlv4.si - IID = Intrinsic::x86_avx2_psrlv_d; - else if (Size == 's' && Name[17] == '8') // avx512.mask.psrlv8.si - IID = Intrinsic::x86_avx2_psrlv_d_256; - else if (Size == 'h' && Name[17] == '8') // avx512.mask.psrlv8.hi - IID = Intrinsic::x86_avx512_psrlv_w_128; - else if (Size == 'h' && Name[17] == '1') // avx512.mask.psrlv16.hi - IID = Intrinsic::x86_avx512_psrlv_w_256; - else if (Name[17] == '3' && Name[18] == '2') // avx512.mask.psrlv32hi - IID = Intrinsic::x86_avx512_psrlv_w_512; - else - llvm_unreachable("Unexpected size"); - } else if (Name.ends_with(".128")) { - if (Size == 'd') // avx512.mask.psrl.d.128, avx512.mask.psrl.di.128 - IID = IsImmediate ? Intrinsic::x86_sse2_psrli_d - : Intrinsic::x86_sse2_psrl_d; - else if (Size == 'q') // avx512.mask.psrl.q.128, avx512.mask.psrl.qi.128 - IID = IsImmediate ? Intrinsic::x86_sse2_psrli_q - : Intrinsic::x86_sse2_psrl_q; - else if (Size == 'w') // avx512.mask.psrl.w.128, avx512.mask.psrl.wi.128 - IID = IsImmediate ? Intrinsic::x86_sse2_psrli_w - : Intrinsic::x86_sse2_psrl_w; - else - llvm_unreachable("Unexpected size"); - } else if (Name.ends_with(".256")) { - if (Size == 'd') // avx512.mask.psrl.d.256, avx512.mask.psrl.di.256 - IID = IsImmediate ? Intrinsic::x86_avx2_psrli_d - : Intrinsic::x86_avx2_psrl_d; - else if (Size == 'q') // avx512.mask.psrl.q.256, avx512.mask.psrl.qi.256 - IID = IsImmediate ? Intrinsic::x86_avx2_psrli_q - : Intrinsic::x86_avx2_psrl_q; - else if (Size == 'w') // avx512.mask.psrl.w.256, avx512.mask.psrl.wi.256 - IID = IsImmediate ? Intrinsic::x86_avx2_psrli_w - : Intrinsic::x86_avx2_psrl_w; - else - llvm_unreachable("Unexpected size"); - } else { - if (Size == 'd') // psrl.di.512, psrli.d, psrl.d, psrl.d.512 - IID = IsImmediate ? Intrinsic::x86_avx512_psrli_d_512 : - IsVariable ? Intrinsic::x86_avx512_psrlv_d_512 : - Intrinsic::x86_avx512_psrl_d_512; - else if (Size == 'q') // psrl.qi.512, psrli.q, psrl.q, psrl.q.512 - IID = IsImmediate ? Intrinsic::x86_avx512_psrli_q_512 : - IsVariable ? Intrinsic::x86_avx512_psrlv_q_512 : - Intrinsic::x86_avx512_psrl_q_512; - else if (Size == 'w') // psrl.wi.512, psrli.w, psrl.w) - IID = IsImmediate ? Intrinsic::x86_avx512_psrli_w_512 - : Intrinsic::x86_avx512_psrl_w_512; - else - llvm_unreachable("Unexpected size"); - } - - Rep = upgradeX86MaskedShift(Builder, *CI, IID); - } else if (IsX86 && Name.starts_with("avx512.mask.psra")) { - bool IsImmediate = Name[16] == 'i' || - (Name.size() > 18 && Name[18] == 'i'); - bool IsVariable = Name[16] == 'v'; - char Size = Name[16] == '.' ? Name[17] : - Name[17] == '.' ? Name[18] : - Name[18] == '.' ? Name[19] : - Name[20]; - - Intrinsic::ID IID; - if (IsVariable && Name[17] != '.') { - if (Size == 's' && Name[17] == '4') // avx512.mask.psrav4.si - IID = Intrinsic::x86_avx2_psrav_d; - else if (Size == 's' && Name[17] == '8') // avx512.mask.psrav8.si - IID = Intrinsic::x86_avx2_psrav_d_256; - else if (Size == 'h' && Name[17] == '8') // avx512.mask.psrav8.hi - IID = Intrinsic::x86_avx512_psrav_w_128; - else if (Size == 'h' && Name[17] == '1') // avx512.mask.psrav16.hi - IID = Intrinsic::x86_avx512_psrav_w_256; - else if (Name[17] == '3' && Name[18] == '2') // avx512.mask.psrav32hi - IID = Intrinsic::x86_avx512_psrav_w_512; - else - llvm_unreachable("Unexpected size"); - } else if (Name.ends_with(".128")) { - if (Size == 'd') // avx512.mask.psra.d.128, avx512.mask.psra.di.128 - IID = IsImmediate ? Intrinsic::x86_sse2_psrai_d - : Intrinsic::x86_sse2_psra_d; - else if (Size == 'q') // avx512.mask.psra.q.128, avx512.mask.psra.qi.128 - IID = IsImmediate ? Intrinsic::x86_avx512_psrai_q_128 : - IsVariable ? Intrinsic::x86_avx512_psrav_q_128 : - Intrinsic::x86_avx512_psra_q_128; - else if (Size == 'w') // avx512.mask.psra.w.128, avx512.mask.psra.wi.128 - IID = IsImmediate ? Intrinsic::x86_sse2_psrai_w - : Intrinsic::x86_sse2_psra_w; - else - llvm_unreachable("Unexpected size"); - } else if (Name.ends_with(".256")) { - if (Size == 'd') // avx512.mask.psra.d.256, avx512.mask.psra.di.256 - IID = IsImmediate ? Intrinsic::x86_avx2_psrai_d - : Intrinsic::x86_avx2_psra_d; - else if (Size == 'q') // avx512.mask.psra.q.256, avx512.mask.psra.qi.256 - IID = IsImmediate ? Intrinsic::x86_avx512_psrai_q_256 : - IsVariable ? Intrinsic::x86_avx512_psrav_q_256 : - Intrinsic::x86_avx512_psra_q_256; - else if (Size == 'w') // avx512.mask.psra.w.256, avx512.mask.psra.wi.256 - IID = IsImmediate ? Intrinsic::x86_avx2_psrai_w - : Intrinsic::x86_avx2_psra_w; - else - llvm_unreachable("Unexpected size"); - } else { - if (Size == 'd') // psra.di.512, psrai.d, psra.d, psrav.d.512 - IID = IsImmediate ? Intrinsic::x86_avx512_psrai_d_512 : - IsVariable ? Intrinsic::x86_avx512_psrav_d_512 : - Intrinsic::x86_avx512_psra_d_512; - else if (Size == 'q') // psra.qi.512, psrai.q, psra.q - IID = IsImmediate ? Intrinsic::x86_avx512_psrai_q_512 : - IsVariable ? Intrinsic::x86_avx512_psrav_q_512 : - Intrinsic::x86_avx512_psra_q_512; - else if (Size == 'w') // psra.wi.512, psrai.w, psra.w - IID = IsImmediate ? Intrinsic::x86_avx512_psrai_w_512 - : Intrinsic::x86_avx512_psra_w_512; - else - llvm_unreachable("Unexpected size"); - } - - Rep = upgradeX86MaskedShift(Builder, *CI, IID); - } else if (IsX86 && Name.starts_with("avx512.mask.move.s")) { - Rep = upgradeMaskedMove(Builder, *CI); - } else if (IsX86 && Name.starts_with("avx512.cvtmask2")) { - Rep = upgradeMaskToInt(Builder, *CI); - } else if (IsX86 && Name.ends_with(".movntdqa")) { - MDNode *Node = MDNode::get( - C, ConstantAsMetadata::get(ConstantInt::get(Type::getInt32Ty(C), 1))); - - Value *Ptr = CI->getArgOperand(0); - - // Convert the type of the pointer to a pointer to the stored type. - Value *BC = Builder.CreateBitCast( - Ptr, PointerType::getUnqual(CI->getType()), "cast"); - LoadInst *LI = Builder.CreateAlignedLoad( - CI->getType(), BC, - Align(CI->getType()->getPrimitiveSizeInBits().getFixedValue() / 8)); - LI->setMetadata(LLVMContext::MD_nontemporal, Node); - Rep = LI; - } else if (IsX86 && (Name.starts_with("fma.vfmadd.") || - Name.starts_with("fma.vfmsub.") || - Name.starts_with("fma.vfnmadd.") || - Name.starts_with("fma.vfnmsub."))) { - bool NegMul = Name[6] == 'n'; - bool NegAcc = NegMul ? Name[8] == 's' : Name[7] == 's'; - bool IsScalar = NegMul ? Name[12] == 's' : Name[11] == 's'; - - Value *Ops[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2) }; - - if (IsScalar) { - Ops[0] = Builder.CreateExtractElement(Ops[0], (uint64_t)0); - Ops[1] = Builder.CreateExtractElement(Ops[1], (uint64_t)0); - Ops[2] = Builder.CreateExtractElement(Ops[2], (uint64_t)0); - } - - if (NegMul && !IsScalar) - Ops[0] = Builder.CreateFNeg(Ops[0]); - if (NegMul && IsScalar) - Ops[1] = Builder.CreateFNeg(Ops[1]); - if (NegAcc) - Ops[2] = Builder.CreateFNeg(Ops[2]); - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), - Intrinsic::fma, - Ops[0]->getType()), - Ops); - - if (IsScalar) - Rep = Builder.CreateInsertElement(CI->getArgOperand(0), Rep, - (uint64_t)0); - } else if (IsX86 && Name.starts_with("fma4.vfmadd.s")) { - Value *Ops[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2) }; - - Ops[0] = Builder.CreateExtractElement(Ops[0], (uint64_t)0); - Ops[1] = Builder.CreateExtractElement(Ops[1], (uint64_t)0); - Ops[2] = Builder.CreateExtractElement(Ops[2], (uint64_t)0); - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), - Intrinsic::fma, - Ops[0]->getType()), - Ops); - - Rep = Builder.CreateInsertElement(Constant::getNullValue(CI->getType()), - Rep, (uint64_t)0); - } else if (IsX86 && (Name.starts_with("avx512.mask.vfmadd.s") || - Name.starts_with("avx512.maskz.vfmadd.s") || - Name.starts_with("avx512.mask3.vfmadd.s") || - Name.starts_with("avx512.mask3.vfmsub.s") || - Name.starts_with("avx512.mask3.vfnmsub.s"))) { - bool IsMask3 = Name[11] == '3'; - bool IsMaskZ = Name[11] == 'z'; - // Drop the "avx512.mask." to make it easier. - Name = Name.drop_front(IsMask3 || IsMaskZ ? 13 : 12); - bool NegMul = Name[2] == 'n'; - bool NegAcc = NegMul ? Name[4] == 's' : Name[3] == 's'; - - Value *A = CI->getArgOperand(0); - Value *B = CI->getArgOperand(1); - Value *C = CI->getArgOperand(2); - - if (NegMul && (IsMask3 || IsMaskZ)) - A = Builder.CreateFNeg(A); - if (NegMul && !(IsMask3 || IsMaskZ)) - B = Builder.CreateFNeg(B); - if (NegAcc) - C = Builder.CreateFNeg(C); - - A = Builder.CreateExtractElement(A, (uint64_t)0); - B = Builder.CreateExtractElement(B, (uint64_t)0); - C = Builder.CreateExtractElement(C, (uint64_t)0); - - if (!isa(CI->getArgOperand(4)) || - cast(CI->getArgOperand(4))->getZExtValue() != 4) { - Value *Ops[] = { A, B, C, CI->getArgOperand(4) }; - - Intrinsic::ID IID; - if (Name.back() == 'd') - IID = Intrinsic::x86_avx512_vfmadd_f64; - else - IID = Intrinsic::x86_avx512_vfmadd_f32; - Function *FMA = Intrinsic::getDeclaration(CI->getModule(), IID); - Rep = Builder.CreateCall(FMA, Ops); - } else { - Function *FMA = Intrinsic::getDeclaration(CI->getModule(), - Intrinsic::fma, - A->getType()); - Rep = Builder.CreateCall(FMA, { A, B, C }); - } - - Value *PassThru = IsMaskZ ? Constant::getNullValue(Rep->getType()) : - IsMask3 ? C : A; - - // For Mask3 with NegAcc, we need to create a new extractelement that - // avoids the negation above. - if (NegAcc && IsMask3) - PassThru = Builder.CreateExtractElement(CI->getArgOperand(2), - (uint64_t)0); - - Rep = emitX86ScalarSelect(Builder, CI->getArgOperand(3), Rep, PassThru); - Rep = Builder.CreateInsertElement(CI->getArgOperand(IsMask3 ? 2 : 0), - Rep, (uint64_t)0); - } else if (IsX86 && (Name.starts_with("avx512.mask.vfmadd.p") || - Name.starts_with("avx512.mask.vfnmadd.p") || - Name.starts_with("avx512.mask.vfnmsub.p") || - Name.starts_with("avx512.mask3.vfmadd.p") || - Name.starts_with("avx512.mask3.vfmsub.p") || - Name.starts_with("avx512.mask3.vfnmsub.p") || - Name.starts_with("avx512.maskz.vfmadd.p"))) { - bool IsMask3 = Name[11] == '3'; - bool IsMaskZ = Name[11] == 'z'; - // Drop the "avx512.mask." to make it easier. - Name = Name.drop_front(IsMask3 || IsMaskZ ? 13 : 12); - bool NegMul = Name[2] == 'n'; - bool NegAcc = NegMul ? Name[4] == 's' : Name[3] == 's'; - - Value *A = CI->getArgOperand(0); - Value *B = CI->getArgOperand(1); - Value *C = CI->getArgOperand(2); - - if (NegMul && (IsMask3 || IsMaskZ)) - A = Builder.CreateFNeg(A); - if (NegMul && !(IsMask3 || IsMaskZ)) - B = Builder.CreateFNeg(B); - if (NegAcc) - C = Builder.CreateFNeg(C); - - if (CI->arg_size() == 5 && - (!isa(CI->getArgOperand(4)) || - cast(CI->getArgOperand(4))->getZExtValue() != 4)) { - Intrinsic::ID IID; - // Check the character before ".512" in string. - if (Name[Name.size()-5] == 's') - IID = Intrinsic::x86_avx512_vfmadd_ps_512; - else - IID = Intrinsic::x86_avx512_vfmadd_pd_512; - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - { A, B, C, CI->getArgOperand(4) }); - } else { - Function *FMA = Intrinsic::getDeclaration(CI->getModule(), - Intrinsic::fma, - A->getType()); - Rep = Builder.CreateCall(FMA, { A, B, C }); - } - - Value *PassThru = IsMaskZ ? llvm::Constant::getNullValue(CI->getType()) : - IsMask3 ? CI->getArgOperand(2) : - CI->getArgOperand(0); - - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); - } else if (IsX86 && Name.starts_with("fma.vfmsubadd.p")) { - unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); - unsigned EltWidth = CI->getType()->getScalarSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && EltWidth == 32) - IID = Intrinsic::x86_fma_vfmaddsub_ps; - else if (VecWidth == 256 && EltWidth == 32) - IID = Intrinsic::x86_fma_vfmaddsub_ps_256; - else if (VecWidth == 128 && EltWidth == 64) - IID = Intrinsic::x86_fma_vfmaddsub_pd; - else if (VecWidth == 256 && EltWidth == 64) - IID = Intrinsic::x86_fma_vfmaddsub_pd_256; - else - llvm_unreachable("Unexpected intrinsic"); - - Value *Ops[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2) }; - Ops[2] = Builder.CreateFNeg(Ops[2]); - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - Ops); - } else if (IsX86 && (Name.starts_with("avx512.mask.vfmaddsub.p") || - Name.starts_with("avx512.mask3.vfmaddsub.p") || - Name.starts_with("avx512.maskz.vfmaddsub.p") || - Name.starts_with("avx512.mask3.vfmsubadd.p"))) { - bool IsMask3 = Name[11] == '3'; - bool IsMaskZ = Name[11] == 'z'; - // Drop the "avx512.mask." to make it easier. - Name = Name.drop_front(IsMask3 || IsMaskZ ? 13 : 12); - bool IsSubAdd = Name[3] == 's'; - if (CI->arg_size() == 5) { - Intrinsic::ID IID; - // Check the character before ".512" in string. - if (Name[Name.size()-5] == 's') - IID = Intrinsic::x86_avx512_vfmaddsub_ps_512; - else - IID = Intrinsic::x86_avx512_vfmaddsub_pd_512; - - Value *Ops[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2), CI->getArgOperand(4) }; - if (IsSubAdd) - Ops[2] = Builder.CreateFNeg(Ops[2]); - - Rep = Builder.CreateCall(Intrinsic::getDeclaration(F->getParent(), IID), - Ops); - } else { - int NumElts = cast(CI->getType())->getNumElements(); - - Value *Ops[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2) }; - - Function *FMA = Intrinsic::getDeclaration(CI->getModule(), Intrinsic::fma, - Ops[0]->getType()); - Value *Odd = Builder.CreateCall(FMA, Ops); - Ops[2] = Builder.CreateFNeg(Ops[2]); - Value *Even = Builder.CreateCall(FMA, Ops); - - if (IsSubAdd) - std::swap(Even, Odd); - - SmallVector Idxs(NumElts); - for (int i = 0; i != NumElts; ++i) - Idxs[i] = i + (i % 2) * NumElts; - - Rep = Builder.CreateShuffleVector(Even, Odd, Idxs); - } - - Value *PassThru = IsMaskZ ? llvm::Constant::getNullValue(CI->getType()) : - IsMask3 ? CI->getArgOperand(2) : - CI->getArgOperand(0); - - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); - } else if (IsX86 && (Name.starts_with("avx512.mask.pternlog.") || - Name.starts_with("avx512.maskz.pternlog."))) { - bool ZeroMask = Name[11] == 'z'; - unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); - unsigned EltWidth = CI->getType()->getScalarSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && EltWidth == 32) - IID = Intrinsic::x86_avx512_pternlog_d_128; - else if (VecWidth == 256 && EltWidth == 32) - IID = Intrinsic::x86_avx512_pternlog_d_256; - else if (VecWidth == 512 && EltWidth == 32) - IID = Intrinsic::x86_avx512_pternlog_d_512; - else if (VecWidth == 128 && EltWidth == 64) - IID = Intrinsic::x86_avx512_pternlog_q_128; - else if (VecWidth == 256 && EltWidth == 64) - IID = Intrinsic::x86_avx512_pternlog_q_256; - else if (VecWidth == 512 && EltWidth == 64) - IID = Intrinsic::x86_avx512_pternlog_q_512; - else - llvm_unreachable("Unexpected intrinsic"); - - Value *Args[] = { CI->getArgOperand(0) , CI->getArgOperand(1), - CI->getArgOperand(2), CI->getArgOperand(3) }; - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), - Args); - Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) - : CI->getArgOperand(0); - Rep = emitX86Select(Builder, CI->getArgOperand(4), Rep, PassThru); - } else if (IsX86 && (Name.starts_with("avx512.mask.vpmadd52") || - Name.starts_with("avx512.maskz.vpmadd52"))) { - bool ZeroMask = Name[11] == 'z'; - bool High = Name[20] == 'h' || Name[21] == 'h'; - unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && !High) - IID = Intrinsic::x86_avx512_vpmadd52l_uq_128; - else if (VecWidth == 256 && !High) - IID = Intrinsic::x86_avx512_vpmadd52l_uq_256; - else if (VecWidth == 512 && !High) - IID = Intrinsic::x86_avx512_vpmadd52l_uq_512; - else if (VecWidth == 128 && High) - IID = Intrinsic::x86_avx512_vpmadd52h_uq_128; - else if (VecWidth == 256 && High) - IID = Intrinsic::x86_avx512_vpmadd52h_uq_256; - else if (VecWidth == 512 && High) - IID = Intrinsic::x86_avx512_vpmadd52h_uq_512; - else - llvm_unreachable("Unexpected intrinsic"); - - Value *Args[] = { CI->getArgOperand(0) , CI->getArgOperand(1), - CI->getArgOperand(2) }; - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), - Args); - Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) - : CI->getArgOperand(0); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); - } else if (IsX86 && (Name.starts_with("avx512.mask.vpermi2var.") || - Name.starts_with("avx512.mask.vpermt2var.") || - Name.starts_with("avx512.maskz.vpermt2var."))) { - bool ZeroMask = Name[11] == 'z'; - bool IndexForm = Name[17] == 'i'; - Rep = upgradeX86VPERMT2Intrinsics(Builder, *CI, ZeroMask, IndexForm); - } else if (IsX86 && (Name.starts_with("avx512.mask.vpdpbusd.") || - Name.starts_with("avx512.maskz.vpdpbusd.") || - Name.starts_with("avx512.mask.vpdpbusds.") || - Name.starts_with("avx512.maskz.vpdpbusds."))) { - bool ZeroMask = Name[11] == 'z'; - bool IsSaturating = Name[ZeroMask ? 21 : 20] == 's'; - unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && !IsSaturating) - IID = Intrinsic::x86_avx512_vpdpbusd_128; - else if (VecWidth == 256 && !IsSaturating) - IID = Intrinsic::x86_avx512_vpdpbusd_256; - else if (VecWidth == 512 && !IsSaturating) - IID = Intrinsic::x86_avx512_vpdpbusd_512; - else if (VecWidth == 128 && IsSaturating) - IID = Intrinsic::x86_avx512_vpdpbusds_128; - else if (VecWidth == 256 && IsSaturating) - IID = Intrinsic::x86_avx512_vpdpbusds_256; - else if (VecWidth == 512 && IsSaturating) - IID = Intrinsic::x86_avx512_vpdpbusds_512; - else - llvm_unreachable("Unexpected intrinsic"); - - Value *Args[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2) }; - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), - Args); - Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) - : CI->getArgOperand(0); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); - } else if (IsX86 && (Name.starts_with("avx512.mask.vpdpwssd.") || - Name.starts_with("avx512.maskz.vpdpwssd.") || - Name.starts_with("avx512.mask.vpdpwssds.") || - Name.starts_with("avx512.maskz.vpdpwssds."))) { - bool ZeroMask = Name[11] == 'z'; - bool IsSaturating = Name[ZeroMask ? 21 : 20] == 's'; - unsigned VecWidth = CI->getType()->getPrimitiveSizeInBits(); - Intrinsic::ID IID; - if (VecWidth == 128 && !IsSaturating) - IID = Intrinsic::x86_avx512_vpdpwssd_128; - else if (VecWidth == 256 && !IsSaturating) - IID = Intrinsic::x86_avx512_vpdpwssd_256; - else if (VecWidth == 512 && !IsSaturating) - IID = Intrinsic::x86_avx512_vpdpwssd_512; - else if (VecWidth == 128 && IsSaturating) - IID = Intrinsic::x86_avx512_vpdpwssds_128; - else if (VecWidth == 256 && IsSaturating) - IID = Intrinsic::x86_avx512_vpdpwssds_256; - else if (VecWidth == 512 && IsSaturating) - IID = Intrinsic::x86_avx512_vpdpwssds_512; - else - llvm_unreachable("Unexpected intrinsic"); - - Value *Args[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2) }; - Rep = Builder.CreateCall(Intrinsic::getDeclaration(CI->getModule(), IID), - Args); - Value *PassThru = ZeroMask ? ConstantAggregateZero::get(CI->getType()) - : CI->getArgOperand(0); - Rep = emitX86Select(Builder, CI->getArgOperand(3), Rep, PassThru); - } else if (IsX86 && (Name == "addcarryx.u32" || Name == "addcarryx.u64" || - Name == "addcarry.u32" || Name == "addcarry.u64" || - Name == "subborrow.u32" || Name == "subborrow.u64")) { - Intrinsic::ID IID; - if (Name[0] == 'a' && Name.back() == '2') - IID = Intrinsic::x86_addcarry_32; - else if (Name[0] == 'a' && Name.back() == '4') - IID = Intrinsic::x86_addcarry_64; - else if (Name[0] == 's' && Name.back() == '2') - IID = Intrinsic::x86_subborrow_32; - else if (Name[0] == 's' && Name.back() == '4') - IID = Intrinsic::x86_subborrow_64; - else - llvm_unreachable("Unexpected intrinsic"); - - // Make a call with 3 operands. - Value *Args[] = { CI->getArgOperand(0), CI->getArgOperand(1), - CI->getArgOperand(2)}; - Value *NewCall = Builder.CreateCall( - Intrinsic::getDeclaration(CI->getModule(), IID), - Args); - - // Extract the second result and store it. - Value *Data = Builder.CreateExtractValue(NewCall, 1); - // Cast the pointer to the right type. - Value *Ptr = Builder.CreateBitCast(CI->getArgOperand(3), - llvm::PointerType::getUnqual(Data->getType())); - Builder.CreateAlignedStore(Data, Ptr, Align(1)); - // Replace the original call result with the first result of the new call. - Value *CF = Builder.CreateExtractValue(NewCall, 0); - CI->replaceAllUsesWith(CF); + if (!IsX86 && Name == "stackprotectorcheck") { Rep = nullptr; - } else if (IsX86 && Name.starts_with("avx512.mask.") && - upgradeAVX512MaskToSelect(Name, Builder, *CI, Rep)) { - // Rep will be updated by the call in the condition. } else if (IsNVVM && (Name == "abs.i" || Name == "abs.ll")) { Value *Arg = CI->getArgOperand(0); Value *Neg = Builder.CreateNeg(Arg, "neg"); @@ -4332,6 +4266,8 @@ void llvm::UpgradeIntrinsicCall(CallBase *CI, Function *NewFn) { Rep = Builder.CreateBitCast(Rep, F->getReturnType()); } } + } else if (IsX86) { + Rep = upgradeX86IntrinsicCall(Name, CI, F, Builder); } else if (IsARM) { Rep = upgradeARMIntrinsicCall(Name, CI, F, Builder); } else if (IsAMDGCN) {