+bool AArch64FastISel::selectFRem(const Instruction *I) {
+ MVT RetVT;
+ if (!isTypeLegal(I->getType(), RetVT))
+ return false;
+
+ RTLIB::Libcall LC;
+ switch (RetVT.SimpleTy) {
+ default:
+ return false;
+ case MVT::f32:
+ LC = RTLIB::REM_F32;
+ break;
+ case MVT::f64:
+ LC = RTLIB::REM_F64;
+ break;
+ }
+
+ ArgListTy Args;
+ Args.reserve(I->getNumOperands());
+
+ // Populate the argument list.
+ for (auto &Arg : I->operands()) {
+ ArgListEntry Entry;
+ Entry.Val = Arg;
+ Entry.Ty = Arg->getType();
+ Args.push_back(Entry);
+ }
+
+ CallLoweringInfo CLI;
+ MCContext &Ctx = MF->getContext();
+ CLI.setCallee(DL, Ctx, TLI.getLibcallCallingConv(LC), I->getType(),
+ TLI.getLibcallName(LC), std::move(Args));
+ if (!lowerCallTo(CLI))
+ return false;
+ updateValueMap(I, CLI.ResultReg);
+ return true;
+}
+
+bool AArch64FastISel::selectSDiv(const Instruction *I) {
+ MVT VT;
+ if (!isTypeLegal(I->getType(), VT))
+ return false;
+
+ if (!isa<ConstantInt>(I->getOperand(1)))
+ return selectBinaryOp(I, ISD::SDIV);
+
+ const APInt &C = cast<ConstantInt>(I->getOperand(1))->getValue();
+ if ((VT != MVT::i32 && VT != MVT::i64) || !C ||
+ !(C.isPowerOf2() || (-C).isPowerOf2()))
+ return selectBinaryOp(I, ISD::SDIV);
+
+ unsigned Lg2 = C.countTrailingZeros();
+ unsigned Src0Reg = getRegForValue(I->getOperand(0));
+ if (!Src0Reg)
+ return false;
+ bool Src0IsKill = hasTrivialKill(I->getOperand(0));
+
+ if (cast<BinaryOperator>(I)->isExact()) {
+ unsigned ResultReg = emitASR_ri(VT, VT, Src0Reg, Src0IsKill, Lg2);
+ if (!ResultReg)
+ return false;
+ updateValueMap(I, ResultReg);
+ return true;
+ }
+
+ int64_t Pow2MinusOne = (1ULL << Lg2) - 1;
+ unsigned AddReg = emitAdd_ri_(VT, Src0Reg, /*IsKill=*/false, Pow2MinusOne);
+ if (!AddReg)
+ return false;
+
+ // (Src0 < 0) ? Pow2 - 1 : 0;
+ if (!emitICmp_ri(VT, Src0Reg, /*IsKill=*/false, 0))
+ return false;
+
+ unsigned SelectOpc;
+ const TargetRegisterClass *RC;
+ if (VT == MVT::i64) {
+ SelectOpc = AArch64::CSELXr;
+ RC = &AArch64::GPR64RegClass;
+ } else {
+ SelectOpc = AArch64::CSELWr;
+ RC = &AArch64::GPR32RegClass;
+ }
+ unsigned SelectReg =
+ fastEmitInst_rri(SelectOpc, RC, AddReg, /*IsKill=*/true, Src0Reg,
+ Src0IsKill, AArch64CC::LT);
+ if (!SelectReg)
+ return false;
+
+ // Divide by Pow2 --> ashr. If we're dividing by a negative value we must also
+ // negate the result.
+ unsigned ZeroReg = (VT == MVT::i64) ? AArch64::XZR : AArch64::WZR;
+ unsigned ResultReg;
+ if (C.isNegative())
+ ResultReg = emitAddSub_rs(/*UseAdd=*/false, VT, ZeroReg, /*IsKill=*/true,
+ SelectReg, /*IsKill=*/true, AArch64_AM::ASR, Lg2);
+ else
+ ResultReg = emitASR_ri(VT, VT, SelectReg, /*IsKill=*/true, Lg2);
+
+ if (!ResultReg)
+ return false;
+
+ updateValueMap(I, ResultReg);
+ return true;
+}
+
+/// This is mostly a copy of the existing FastISel getRegForGEPIndex code. We
+/// have to duplicate it for AArch64, because otherwise we would fail during the
+/// sign-extend emission.
+std::pair<unsigned, bool> AArch64FastISel::getRegForGEPIndex(const Value *Idx) {
+ unsigned IdxN = getRegForValue(Idx);
+ if (IdxN == 0)
+ // Unhandled operand. Halt "fast" selection and bail.
+ return std::pair<unsigned, bool>(0, false);
+
+ bool IdxNIsKill = hasTrivialKill(Idx);
+
+ // If the index is smaller or larger than intptr_t, truncate or extend it.
+ MVT PtrVT = TLI.getPointerTy(DL);
+ EVT IdxVT = EVT::getEVT(Idx->getType(), /*HandleUnknown=*/false);
+ if (IdxVT.bitsLT(PtrVT)) {
+ IdxN = emitIntExt(IdxVT.getSimpleVT(), IdxN, PtrVT, /*IsZExt=*/false);
+ IdxNIsKill = true;
+ } else if (IdxVT.bitsGT(PtrVT))
+ llvm_unreachable("AArch64 FastISel doesn't support types larger than i64");
+ return std::pair<unsigned, bool>(IdxN, IdxNIsKill);
+}
+
+/// This is mostly a copy of the existing FastISel GEP code, but we have to
+/// duplicate it for AArch64, because otherwise we would bail out even for
+/// simple cases. This is because the standard fastEmit functions don't cover
+/// MUL at all and ADD is lowered very inefficientily.
+bool AArch64FastISel::selectGetElementPtr(const Instruction *I) {
+ unsigned N = getRegForValue(I->getOperand(0));
+ if (!N)
+ return false;
+ bool NIsKill = hasTrivialKill(I->getOperand(0));
+
+ // Keep a running tab of the total offset to coalesce multiple N = N + Offset
+ // into a single N = N + TotalOffset.
+ uint64_t TotalOffs = 0;
+ Type *Ty = I->getOperand(0)->getType();
+ MVT VT = TLI.getPointerTy(DL);
+ for (auto OI = std::next(I->op_begin()), E = I->op_end(); OI != E; ++OI) {
+ const Value *Idx = *OI;
+ if (auto *StTy = dyn_cast<StructType>(Ty)) {
+ unsigned Field = cast<ConstantInt>(Idx)->getZExtValue();
+ // N = N + Offset
+ if (Field)
+ TotalOffs += DL.getStructLayout(StTy)->getElementOffset(Field);
+ Ty = StTy->getElementType(Field);
+ } else {
+ Ty = cast<SequentialType>(Ty)->getElementType();
+ // If this is a constant subscript, handle it quickly.
+ if (const auto *CI = dyn_cast<ConstantInt>(Idx)) {
+ if (CI->isZero())
+ continue;
+ // N = N + Offset
+ TotalOffs +=
+ DL.getTypeAllocSize(Ty) * cast<ConstantInt>(CI)->getSExtValue();
+ continue;
+ }
+ if (TotalOffs) {
+ N = emitAdd_ri_(VT, N, NIsKill, TotalOffs);
+ if (!N)
+ return false;
+ NIsKill = true;
+ TotalOffs = 0;
+ }
+
+ // N = N + Idx * ElementSize;
+ uint64_t ElementSize = DL.getTypeAllocSize(Ty);
+ std::pair<unsigned, bool> Pair = getRegForGEPIndex(Idx);
+ unsigned IdxN = Pair.first;
+ bool IdxNIsKill = Pair.second;
+ if (!IdxN)
+ return false;
+
+ if (ElementSize != 1) {
+ unsigned C = fastEmit_i(VT, VT, ISD::Constant, ElementSize);
+ if (!C)
+ return false;
+ IdxN = emitMul_rr(VT, IdxN, IdxNIsKill, C, true);
+ if (!IdxN)
+ return false;
+ IdxNIsKill = true;
+ }
+ N = fastEmit_rr(VT, VT, ISD::ADD, N, NIsKill, IdxN, IdxNIsKill);
+ if (!N)
+ return false;
+ }
+ }
+ if (TotalOffs) {
+ N = emitAdd_ri_(VT, N, NIsKill, TotalOffs);
+ if (!N)
+ return false;
+ }
+ updateValueMap(I, N);
+ return true;
+}
+