dynarmic/src/backend_x64/emit_x64.cpp

/* This file is part of the dynarmic project.
 * Copyright (c) 2016 MerryMage
 * This software may be used and distributed according to the terms of the GNU
 * General Public License version 2 or any later version.
 */

#include <unordered_map>

#include "backend_x64/abi.h"
#include "backend_x64/block_of_code.h"
#include "backend_x64/emit_x64.h"
#include "backend_x64/jitstate.h"
#include "common/assert.h"
#include "common/bit_util.h"
#include "frontend/arm/types.h"
#include "frontend/ir/basic_block.h"
#include "frontend/ir/location_descriptor.h"
#include "frontend/ir/microinstruction.h"
#include "frontend/ir/opcodes.h"

// TODO: Have ARM flags in host flags and not have them use up GPR registers unless necessary.
// TODO: Actually implement that proper instruction selector you've always wanted to sweetheart.

namespace Dynarmic {
namespace BackendX64 {

static Xbyak::Address MJitStateReg(Arm::Reg reg) {
    using namespace Xbyak::util;
    return dword[r15 + offsetof(JitState, Reg) + sizeof(u32) * static_cast<size_t>(reg)];
}

static Xbyak::Address MJitStateExtReg(Arm::ExtReg reg) {
    using namespace Xbyak::util;
    if (Arm::IsSingleExtReg(reg)) {
        size_t index = static_cast<size_t>(reg) - static_cast<size_t>(Arm::ExtReg::S0);
        return dword[r15 + offsetof(JitState, ExtReg) + sizeof(u32) * index];
    }
    if (Arm::IsDoubleExtReg(reg)) {
        size_t index = static_cast<size_t>(reg) - static_cast<size_t>(Arm::ExtReg::D0);
        return qword[r15 + offsetof(JitState, ExtReg) + sizeof(u64) * index];
    }
    ASSERT_MSG(false, "Should never happen.");
}

static Xbyak::Address MJitStateCpsr() {
    using namespace Xbyak::util;
    return dword[r15 + offsetof(JitState, Cpsr)];
}

static void EraseInstruction(IR::Block& block, IR::Inst* inst) {
    block.Instructions().erase(inst);
}

EmitX64::EmitX64(BlockOfCode* code, UserCallbacks cb, Jit* jit_interface)
    : reg_alloc(code), code(code), cb(cb), jit_interface(jit_interface) {
}

EmitX64::BlockDescriptor EmitX64::Emit(IR::Block& block) {
    const IR::LocationDescriptor descriptor = block.Location();

    reg_alloc.Reset();

    code->align();
    const CodePtr code_ptr = code->getCurr();
    EmitX64::BlockDescriptor& block_desc = block_descriptors[descriptor.UniqueHash()];
    block_desc.code_ptr = code_ptr;

    EmitCondPrelude(block);

    for (auto iter = block.begin(); iter != block.end(); ++iter) {
        IR::Inst* inst = &*iter;

        // Call the relevant Emit* member function.
        switch (inst->GetOpcode()) {

#define OPCODE(name, type, ...)                \
        case IR::Opcode::name:                 \
            EmitX64::Emit##name(block, inst);  \
            break;
#include "frontend/ir/opcodes.inc"
#undef OPCODE

        default:
            ASSERT_MSG(false, "Invalid opcode %zu", static_cast<size_t>(inst->GetOpcode()));
            break;
        }

        reg_alloc.EndOfAllocScope();
    }

    EmitAddCycles(block.CycleCount());
    EmitTerminal(block.GetTerminal(), block.Location());
    code->int3();

    reg_alloc.AssertNoMoreUses();

    Patch(descriptor, code_ptr);
    block_desc.size = std::intptr_t(code->getCurr()) - std::intptr_t(code_ptr);
    return block_desc;
}

boost::optional<EmitX64::BlockDescriptor> EmitX64::GetBasicBlock(IR::LocationDescriptor descriptor) const {
    auto iter = block_descriptors.find(descriptor.UniqueHash());
    if (iter == block_descriptors.end())
        return boost::none;
    return boost::make_optional<BlockDescriptor>(iter->second);
}

void EmitX64::EmitBreakpoint(IR::Block&, IR::Inst*) {
    code->int3();
}

void EmitX64::EmitIdentity(IR::Block&, IR::Inst* inst) {
    if (!inst->GetArg(0).IsImmediate()) {
        reg_alloc.RegisterAddDef(inst, inst->GetArg(0));
    }
}

void EmitX64::EmitGetRegister(IR::Block&, IR::Inst* inst) {
    Arm::Reg reg = inst->GetArg(0).GetRegRef();
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateReg(reg));
}

void EmitX64::EmitGetExtendedRegister32(IR::Block&, IR::Inst* inst) {
    Arm::ExtReg reg = inst->GetArg(0).GetExtRegRef();
    ASSERT(Arm::IsSingleExtReg(reg));

    Xbyak::Xmm result = reg_alloc.DefXmm(inst);
    code->movss(result, MJitStateExtReg(reg));
}

void EmitX64::EmitGetExtendedRegister64(IR::Block&, IR::Inst* inst) {
    Arm::ExtReg reg = inst->GetArg(0).GetExtRegRef();
    ASSERT(Arm::IsDoubleExtReg(reg));
    Xbyak::Xmm result = reg_alloc.DefXmm(inst);
    code->movsd(result, MJitStateExtReg(reg));
}

void EmitX64::EmitSetRegister(IR::Block&, IR::Inst* inst) {
    Arm::Reg reg = inst->GetArg(0).GetRegRef();
    IR::Value arg = inst->GetArg(1);
    if (arg.IsImmediate()) {
        code->mov(MJitStateReg(reg), arg.GetU32());
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseGpr(arg).cvt32();
        code->mov(MJitStateReg(reg), to_store);
    }
}

void EmitX64::EmitSetExtendedRegister32(IR::Block&, IR::Inst* inst) {
    Arm::ExtReg reg = inst->GetArg(0).GetExtRegRef();
    ASSERT(Arm::IsSingleExtReg(reg));
    Xbyak::Xmm source = reg_alloc.UseXmm(inst->GetArg(1));
    code->movss(MJitStateExtReg(reg), source);
}

void EmitX64::EmitSetExtendedRegister64(IR::Block&, IR::Inst* inst) {
    Arm::ExtReg reg = inst->GetArg(0).GetExtRegRef();
    ASSERT(Arm::IsDoubleExtReg(reg));
    Xbyak::Xmm source = reg_alloc.UseXmm(inst->GetArg(1));
    code->movsd(MJitStateExtReg(reg), source);
}

void EmitX64::EmitGetCpsr(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateCpsr());
}

void EmitX64::EmitSetCpsr(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 arg = reg_alloc.UseGpr(inst->GetArg(0)).cvt32();
    code->mov(MJitStateCpsr(), arg);
}

void EmitX64::EmitGetNFlag(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateCpsr());
    code->shr(result, 31);
}

void EmitX64::EmitSetNFlag(IR::Block&, IR::Inst* inst) {
    constexpr size_t flag_bit = 31;
    constexpr u32 flag_mask = 1u << flag_bit;
    IR::Value arg = inst->GetArg(0);
    if (arg.IsImmediate()) {
        if (arg.GetU1()) {
            code->or_(MJitStateCpsr(), flag_mask);
        } else {
            code->and_(MJitStateCpsr(), ~flag_mask);
        }
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseScratchGpr(arg).cvt32();

        code->shl(to_store, flag_bit);
        code->and_(MJitStateCpsr(), ~flag_mask);
        code->or_(MJitStateCpsr(), to_store);
    }
}

void EmitX64::EmitGetZFlag(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateCpsr());
    code->shr(result, 30);
    code->and_(result, 1);
}

void EmitX64::EmitSetZFlag(IR::Block&, IR::Inst* inst) {
    constexpr size_t flag_bit = 30;
    constexpr u32 flag_mask = 1u << flag_bit;
    IR::Value arg = inst->GetArg(0);
    if (arg.IsImmediate()) {
        if (arg.GetU1()) {
            code->or_(MJitStateCpsr(), flag_mask);
        } else {
            code->and_(MJitStateCpsr(), ~flag_mask);
        }
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseScratchGpr(arg).cvt32();

        code->shl(to_store, flag_bit);
        code->and_(MJitStateCpsr(), ~flag_mask);
        code->or_(MJitStateCpsr(), to_store);
    }
}

void EmitX64::EmitGetCFlag(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateCpsr());
    code->shr(result, 29);
    code->and_(result, 1);
}

void EmitX64::EmitSetCFlag(IR::Block&, IR::Inst* inst) {
    constexpr size_t flag_bit = 29;
    constexpr u32 flag_mask = 1u << flag_bit;
    IR::Value arg = inst->GetArg(0);
    if (arg.IsImmediate()) {
        if (arg.GetU1()) {
            code->or_(MJitStateCpsr(), flag_mask);
        } else {
            code->and_(MJitStateCpsr(), ~flag_mask);
        }
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseScratchGpr(arg).cvt32();

        code->shl(to_store, flag_bit);
        code->and_(MJitStateCpsr(), ~flag_mask);
        code->or_(MJitStateCpsr(), to_store);
    }
}

void EmitX64::EmitGetVFlag(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateCpsr());
    code->shr(result, 28);
    code->and_(result, 1);
}

void EmitX64::EmitSetVFlag(IR::Block&, IR::Inst* inst) {
    constexpr size_t flag_bit = 28;
    constexpr u32 flag_mask = 1u << flag_bit;
    IR::Value arg = inst->GetArg(0);
    if (arg.IsImmediate()) {
        if (arg.GetU1()) {
            code->or_(MJitStateCpsr(), flag_mask);
        } else {
            code->and_(MJitStateCpsr(), ~flag_mask);
        }
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseScratchGpr(arg).cvt32();

        code->shl(to_store, flag_bit);
        code->and_(MJitStateCpsr(), ~flag_mask);
        code->or_(MJitStateCpsr(), to_store);
    }
}

void EmitX64::EmitOrQFlag(IR::Block&, IR::Inst* inst) {
    constexpr size_t flag_bit = 27;
    constexpr u32 flag_mask = 1u << flag_bit;
    IR::Value arg = inst->GetArg(0);
    if (arg.IsImmediate()) {
        if (arg.GetU1())
            code->or_(MJitStateCpsr(), flag_mask);
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseScratchGpr(arg).cvt32();

        code->shl(to_store, flag_bit);
        code->or_(MJitStateCpsr(), to_store);
    }
}

void EmitX64::EmitGetGEFlags(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    code->mov(result, MJitStateCpsr());
    code->shr(result, 16);
    code->and_(result, 0xF);
}

void EmitX64::EmitSetGEFlags(IR::Block&, IR::Inst* inst) {
    constexpr size_t flag_bit = 16;
    constexpr u32 flag_mask = 0xFu << flag_bit;
    IR::Value arg = inst->GetArg(0);
    if (arg.IsImmediate()) {
        u32 imm = (arg.GetU32() << flag_bit) & flag_mask;
        code->and_(MJitStateCpsr(), ~flag_mask);
        code->or_(MJitStateCpsr(), imm);
    } else {
        Xbyak::Reg32 to_store = reg_alloc.UseScratchGpr(arg).cvt32();

        code->shl(to_store, flag_bit);
        code->and_(to_store, flag_mask);
        code->and_(MJitStateCpsr(), ~flag_mask);
        code->or_(MJitStateCpsr(), to_store);
    }
}

void EmitX64::EmitBXWritePC(IR::Block&, IR::Inst* inst) {
    const u32 T_bit = 1 << 5;
    auto arg = inst->GetArg(0);

    // Pseudocode:
    // if (new_pc & 1) {
    //    new_pc &= 0xFFFFFFFE;
    //    cpsr.T = true;
    // } else {
    //    new_pc &= 0xFFFFFFFC;
    //    cpsr.T = false;
    // }

    if (arg.IsImmediate()) {
        u32 new_pc = arg.GetU32();
        if (Common::Bit<0>(new_pc)) {
            new_pc &= 0xFFFFFFFE;
            code->mov(MJitStateReg(Arm::Reg::PC), new_pc);
            code->or_(MJitStateCpsr(), T_bit);
        } else {
            new_pc &= 0xFFFFFFFC;
            code->mov(MJitStateReg(Arm::Reg::PC), new_pc);
            code->and_(MJitStateCpsr(), ~T_bit);
        }
    } else {
        using Xbyak::util::ptr;

        Xbyak::Reg64 new_pc = reg_alloc.UseScratchGpr(arg);
        Xbyak::Reg64 tmp1 = reg_alloc.ScratchGpr();
        Xbyak::Reg64 tmp2 = reg_alloc.ScratchGpr();

        code->mov(tmp1, MJitStateCpsr());
        code->mov(tmp2, tmp1);
        code->and_(tmp2, u32(~T_bit));         // CPSR.T = 0
        code->or_(tmp1, u32(T_bit));           // CPSR.T = 1
        code->test(new_pc, u32(1));
        code->cmove(tmp1, tmp2);               // CPSR.T = pc & 1
        code->mov(MJitStateCpsr(), tmp1);
        code->lea(tmp2, ptr[new_pc + new_pc * 1]);
        code->or_(tmp2, u32(0xFFFFFFFC));      // tmp2 = pc & 1 ? 0xFFFFFFFE : 0xFFFFFFFC
        code->and_(new_pc, tmp2);
        code->mov(MJitStateReg(Arm::Reg::PC), new_pc);
    }
}

void EmitX64::EmitCallSupervisor(IR::Block&, IR::Inst* inst) {
    auto imm32 = inst->GetArg(0);

    reg_alloc.HostCall(nullptr, imm32);

    code->SwitchMxcsrOnExit();
    code->CallFunction(cb.CallSVC);
    code->SwitchMxcsrOnEntry();
}

static u32 GetFpscrImpl(JitState* jit_state) {
    return jit_state->Fpscr();
}

void EmitX64::EmitGetFpscr(IR::Block&, IR::Inst* inst) {
    reg_alloc.HostCall(inst);
    code->mov(code->ABI_PARAM1, code->r15);

    code->SwitchMxcsrOnExit();
    code->CallFunction(&GetFpscrImpl);
    code->SwitchMxcsrOnEntry();
}

static void SetFpscrImpl(u32 value, JitState* jit_state) {
    jit_state->SetFpscr(value);
}

void EmitX64::EmitSetFpscr(IR::Block&, IR::Inst* inst) {
    auto a = inst->GetArg(0);

    reg_alloc.HostCall(nullptr, a);
    code->mov(code->ABI_PARAM2, code->r15);

    code->SwitchMxcsrOnExit();
    code->CallFunction(&SetFpscrImpl);
    code->SwitchMxcsrOnEntry();
}

void EmitX64::EmitGetFpscrNZCV(IR::Block&, IR::Inst* inst) {
    using namespace Xbyak::util;

    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();

    code->mov(result, dword[r15 + offsetof(JitState, FPSCR_nzcv)]);
}

void EmitX64::EmitSetFpscrNZCV(IR::Block&, IR::Inst* inst) {
    using namespace Xbyak::util;

    Xbyak::Reg32 value = reg_alloc.UseGpr(inst->GetArg(0)).cvt32();

    code->mov(dword[r15 + offsetof(JitState, FPSCR_nzcv)], value);
}

void EmitX64::EmitPushRSB(IR::Block&, IR::Inst* inst) {
    using namespace Xbyak::util;

    ASSERT(inst->GetArg(0).IsImmediate());
    u64 unique_hash_of_target = inst->GetArg(0).GetU64();

    auto iter = block_descriptors.find(unique_hash_of_target);
    CodePtr target_code_ptr = iter != block_descriptors.end()
                            ? iter->second.code_ptr
                            : code->GetReturnFromRunCodeAddress();

    Xbyak::Reg64 code_ptr_reg = reg_alloc.ScratchGpr({HostLoc::RCX});
    Xbyak::Reg64 loc_desc_reg = reg_alloc.ScratchGpr();
    Xbyak::Reg32 index_reg = reg_alloc.ScratchGpr().cvt32();

    code->mov(index_reg, dword[r15 + offsetof(JitState, rsb_ptr)]);
    code->add(index_reg, 1);
    code->and_(index_reg, u32(JitState::RSBSize - 1));

    code->mov(loc_desc_reg, unique_hash_of_target);

    patch_information[unique_hash_of_target].mov_rcx.emplace_back(code->getCurr());
    EmitPatchMovRcx(target_code_ptr);

    Xbyak::Label label;
    for (size_t i = 0; i < JitState::RSBSize; ++i) {
        code->cmp(loc_desc_reg, qword[r15 + offsetof(JitState, rsb_location_descriptors) + i * sizeof(u64)]);
        code->je(label, code->T_SHORT);
    }

    code->mov(dword[r15 + offsetof(JitState, rsb_ptr)], index_reg);
    code->mov(qword[r15 + index_reg.cvt64() * 8 + offsetof(JitState, rsb_location_descriptors)], loc_desc_reg);
    code->mov(qword[r15 + index_reg.cvt64() * 8 + offsetof(JitState, rsb_codeptrs)], code_ptr_reg);
    code->L(label);
}

void EmitX64::EmitGetCarryFromOp(IR::Block&, IR::Inst*) {
    ASSERT_MSG(false, "should never happen");
}

void EmitX64::EmitGetOverflowFromOp(IR::Block&, IR::Inst*) {
    ASSERT_MSG(false, "should never happen");
}

void EmitX64::EmitGetGEFromOp(IR::Block&, IR::Inst*) {
    ASSERT_MSG(false, "should never happen");
}

void EmitX64::EmitPack2x32To1x64(IR::Block&, IR::Inst* inst) {
    OpArg lo;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        lo = result.cvt32();
    } else {
        std::tie(lo, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }
    lo.setBit(32);
    Xbyak::Reg64 hi = reg_alloc.UseScratchGpr(inst->GetArg(1));

    code->shl(hi, 32);
    code->mov(result.cvt32(), *lo); // Zero extend to 64-bits
    code->or_(result, hi);
}

void EmitX64::EmitLeastSignificantWord(IR::Block&, IR::Inst* inst) {
    reg_alloc.RegisterAddDef(inst, inst->GetArg(0));
}

void EmitX64::EmitMostSignificantWord(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);
    Xbyak::Reg64 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);

    code->shr(result, 32);

    if (carry_inst) {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();
        Xbyak::Reg64 carry = reg_alloc.DefGpr(carry_inst);

        code->setc(carry.cvt8());
    }
}

void EmitX64::EmitLeastSignificantHalf(IR::Block&, IR::Inst* inst) {
    reg_alloc.RegisterAddDef(inst, inst->GetArg(0));
}

void EmitX64::EmitLeastSignificantByte(IR::Block&, IR::Inst* inst) {
    reg_alloc.RegisterAddDef(inst, inst->GetArg(0));
}

void EmitX64::EmitMostSignificantBit(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();

    // TODO: Flag optimization

    code->shr(result, 31);
}

void EmitX64::EmitIsZero(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();

    // TODO: Flag optimization

    code->test(result, result);
    code->sete(result.cvt8());
    code->movzx(result, result.cvt8());
}

void EmitX64::EmitIsZero64(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg64 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);

    // TODO: Flag optimization

    code->test(result, result);
    code->sete(result.cvt8());
    code->movzx(result, result.cvt8());
}

void EmitX64::EmitLogicalShiftLeft(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);

    // TODO: Consider using BMI2 instructions like SHLX when arm-in-host flags is implemented.

    if (!carry_inst) {
        if (!inst->GetArg(2).IsImmediate()) {
            // TODO: Remove redundant argument.
            inst->GetArg(2).GetInst()->DecrementRemainingUses();
        }

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            u8 shift = shift_arg.GetU8();

            if (shift <= 31) {
                code->shl(result, shift);
            } else {
                code->xor_(result, result);
            }
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 zero = reg_alloc.ScratchGpr().cvt32();

            // The 32-bit x64 SHL instruction masks the shift count by 0x1F before performing the shift.
            // ARM differs from the behaviour: It does not mask the count, so shifts above 31 result in zeros.

            code->shl(result, shift);
            code->xor_(zero, zero);
            code->cmp(shift, 32);
            code->cmovnb(result, zero);
        }
    } else {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            u8 shift = shift_arg.GetU8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt32();

            if (shift == 0) {
                // There is nothing more to do.
            } else if (shift < 32) {
                code->bt(carry.cvt32(), 0);
                code->shl(result, shift);
                code->setc(carry.cvt8());
            } else if (shift > 32) {
                code->xor_(result, result);
                code->xor_(carry, carry);
            } else {
                code->mov(carry, result);
                code->xor_(result, result);
                code->and_(carry, 1);
            }
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt32();

            // TODO: Optimize this.

            code->inLocalLabel();

            code->cmp(shift, 32);
            code->ja(".Rs_gt32");
            code->je(".Rs_eq32");
            // if (Rs & 0xFF < 32) {
            code->bt(carry.cvt32(), 0); // Set the carry flag for correct behaviour in the case when Rs & 0xFF == 0
            code->shl(result, shift);
            code->setc(carry.cvt8());
            code->jmp(".end");
            // } else if (Rs & 0xFF > 32) {
            code->L(".Rs_gt32");
            code->xor_(result, result);
            code->xor_(carry, carry);
            code->jmp(".end");
            // } else if (Rs & 0xFF == 32) {
            code->L(".Rs_eq32");
            code->mov(carry, result);
            code->and_(carry, 1);
            code->xor_(result, result);
            // }
            code->L(".end");

            code->outLocalLabel();
        }
    }
}

void EmitX64::EmitLogicalShiftRight(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);

    if (!carry_inst) {
        if (!inst->GetArg(2).IsImmediate()) {
            // TODO: Remove redundant argument.
            inst->GetArg(2).GetInst()->DecrementRemainingUses();
        }

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            u8 shift = shift_arg.GetU8();

            if (shift <= 31) {
                code->shr(result, shift);
            } else {
                code->xor_(result, result);
            }
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 zero = reg_alloc.ScratchGpr().cvt32();

            // The 32-bit x64 SHR instruction masks the shift count by 0x1F before performing the shift.
            // ARM differs from the behaviour: It does not mask the count, so shifts above 31 result in zeros.

            code->shr(result, shift);
            code->xor_(zero, zero);
            code->cmp(shift, 32);
            code->cmovnb(result, zero);
        }
    } else {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            u8 shift = shift_arg.GetU8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt32();

            if (shift == 0) {
                // There is nothing more to do.
            } else if (shift < 32) {
                code->shr(result, shift);
                code->setc(carry.cvt8());
            } else if (shift == 32) {
                code->bt(result, 31);
                code->setc(carry.cvt8());
                code->mov(result, 0);
            } else {
                code->xor_(result, result);
                code->xor_(carry, carry);
            }
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt32();

            // TODO: Optimize this.

            code->inLocalLabel();

            code->cmp(shift, 32);
            code->ja(".Rs_gt32");
            code->je(".Rs_eq32");
            // if (Rs & 0xFF == 0) goto end;
            code->test(shift, shift);
            code->jz(".end");
            // if (Rs & 0xFF < 32) {
            code->shr(result, shift);
            code->setc(carry.cvt8());
            code->jmp(".end");
            // } else if (Rs & 0xFF > 32) {
            code->L(".Rs_gt32");
            code->xor_(result, result);
            code->xor_(carry, carry);
            code->jmp(".end");
            // } else if (Rs & 0xFF == 32) {
            code->L(".Rs_eq32");
            code->bt(result, 31);
            code->setc(carry.cvt8());
            code->xor_(result, result);
            // }
            code->L(".end");

            code->outLocalLabel();
        }
    }
}

void EmitX64::EmitLogicalShiftRight64(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg64 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);

    auto shift_arg = inst->GetArg(1);
    ASSERT_MSG(shift_arg.IsImmediate(), "variable 64 bit shifts are not implemented");
    u8 shift = shift_arg.GetU8();
    ASSERT_MSG(shift < 64, "shift width clamping is not implemented");

    code->shr(result.cvt64(), shift);
}

void EmitX64::EmitArithmeticShiftRight(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);

    if (!carry_inst) {
        if (!inst->GetArg(2).IsImmediate()) {
            // TODO: Remove redundant argument.
            inst->GetArg(2).GetInst()->DecrementRemainingUses();
        }

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            u8 shift = shift_arg.GetU8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();

            code->sar(result, u8(shift < 31 ? shift : 31));
        } else {
            Xbyak::Reg32 shift = reg_alloc.UseScratchGpr(shift_arg, {HostLoc::RCX}).cvt32();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg32 const31 = reg_alloc.ScratchGpr().cvt32();

            // The 32-bit x64 SAR instruction masks the shift count by 0x1F before performing the shift.
            // ARM differs from the behaviour: It does not mask the count.

            // We note that all shift values above 31 have the same behaviour as 31 does, so we saturate `shift` to 31.
            code->mov(const31, 31);
            code->movzx(shift, shift.cvt8());
            code->cmp(shift, u32(31));
            code->cmovg(shift, const31);
            code->sar(result, shift.cvt8());
        }
    } else {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            u8 shift = shift_arg.GetU8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg8 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt8();

            if (shift == 0) {
                // There is nothing more to do.
            } else if (shift <= 31) {
                code->sar(result, shift);
                code->setc(carry);
            } else {
                code->sar(result, 31);
                code->bt(result, 31);
                code->setc(carry);
            }
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg8 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt8();

            // TODO: Optimize this.

            code->inLocalLabel();

            code->cmp(shift, u32(31));
            code->ja(".Rs_gt31");
            // if (Rs & 0xFF == 0) goto end;
            code->test(shift, shift);
            code->jz(".end");
            // if (Rs & 0xFF <= 31) {
            code->sar(result, shift);
            code->setc(carry);
            code->jmp(".end");
            // } else if (Rs & 0xFF > 31) {
            code->L(".Rs_gt31");
            code->sar(result, 31); // 31 produces the same results as anything above 31
            code->bt(result, 31);
            code->setc(carry);
            // }
            code->L(".end");

            code->outLocalLabel();
        }
    }
}

void EmitX64::EmitRotateRight(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);

    if (!carry_inst) {
        if (!inst->GetArg(2).IsImmediate()) {
            // TODO: Remove redundant argument.
            inst->GetArg(2).GetInst()->DecrementRemainingUses();
        }

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            u8 shift = shift_arg.GetU8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();

            code->ror(result, u8(shift & 0x1F));
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();

            // x64 ROR instruction does (shift & 0x1F) for us.
            code->ror(result, shift);
        }
    } else {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();

        auto shift_arg = inst->GetArg(1);

        if (shift_arg.IsImmediate()) {
            u8 shift = shift_arg.GetU8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg8 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt8();

            if (shift == 0) {
                // There is nothing more to do.
            } else if ((shift & 0x1F) == 0) {
                code->bt(result, u8(31));
                code->setc(carry);
            } else {
                code->ror(result, shift);
                code->setc(carry);
            }
        } else {
            Xbyak::Reg8 shift = reg_alloc.UseScratchGpr(shift_arg, {HostLoc::RCX}).cvt8();
            Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
            Xbyak::Reg8 carry = reg_alloc.UseDefGpr(inst->GetArg(2), carry_inst).cvt8();

            // TODO: Optimize

            code->inLocalLabel();

            // if (Rs & 0xFF == 0) goto end;
            code->test(shift, shift);
            code->jz(".end");

            code->and_(shift.cvt32(), u32(0x1F));
            code->jz(".zero_1F");
            // if (Rs & 0x1F != 0) {
            code->ror(result, shift);
            code->setc(carry);
            code->jmp(".end");
            // } else {
            code->L(".zero_1F");
            code->bt(result, u8(31));
            code->setc(carry);
            // }
            code->L(".end");

            code->outLocalLabel();
        }
    }
}

void EmitX64::EmitRotateRightExtended(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();
    Xbyak::Reg8 carry = carry_inst
                        ? reg_alloc.UseDefGpr(inst->GetArg(1), carry_inst).cvt8()
                        : reg_alloc.UseGpr(inst->GetArg(1)).cvt8();

    code->bt(carry.cvt32(), 0);
    code->rcr(result, 1);

    if (carry_inst) {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();
        code->setc(carry);
    }
}

const Xbyak::Reg64 INVALID_REG = Xbyak::Reg64(-1);

static Xbyak::Reg8 DoCarry(RegAlloc& reg_alloc, const IR::Value& carry_in, IR::Inst* carry_out) {
    if (carry_in.IsImmediate()) {
        return carry_out ? reg_alloc.DefGpr(carry_out).cvt8() : INVALID_REG.cvt8();
    } else {
        return carry_out ? reg_alloc.UseDefGpr(carry_in, carry_out).cvt8() : reg_alloc.UseGpr(carry_in).cvt8();
    }
}

void EmitX64::EmitAddWithCarry(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);
    auto overflow_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetOverflowFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);
    IR::Value carry_in = inst->GetArg(2);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg8 carry = DoCarry(reg_alloc, carry_in, carry_inst);
    Xbyak::Reg8 overflow = overflow_inst ? reg_alloc.DefGpr(overflow_inst).cvt8() : INVALID_REG.cvt8();

    // TODO: Consider using LEA.

    if (b.IsImmediate()) {
        u32 op_arg = b.GetU32();
        if (carry_in.IsImmediate()) {
            if (carry_in.GetU1()) {
                code->stc();
                code->adc(result, op_arg);
            } else {
                code->add(result, op_arg);
            }
        } else {
            code->bt(carry.cvt32(), 0);
            code->adc(result, op_arg);
        }
    } else {
        OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);
        op_arg.setBit(32);
        if (carry_in.IsImmediate()) {
            if (carry_in.GetU1()) {
                code->stc();
                code->adc(result, *op_arg);
            } else {
                code->add(result, *op_arg);
            }
        } else {
            code->bt(carry.cvt32(), 0);
            code->adc(result, *op_arg);
        }
    }

    if (carry_inst) {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();
        code->setc(carry);
    }
    if (overflow_inst) {
        EraseInstruction(block, overflow_inst);
        inst->DecrementRemainingUses();
        code->seto(overflow);
    }
}

void EmitX64::EmitAdd64(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg64 result = reg_alloc.UseDefGpr(a, inst);
    Xbyak::Reg64 op_arg = reg_alloc.UseGpr(b);

    code->add(result, op_arg);
}

void EmitX64::EmitSubWithCarry(IR::Block& block, IR::Inst* inst) {
    auto carry_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetCarryFromOp);
    auto overflow_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetOverflowFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);
    IR::Value carry_in = inst->GetArg(2);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg8 carry = DoCarry(reg_alloc, carry_in, carry_inst);
    Xbyak::Reg8 overflow = overflow_inst ? reg_alloc.DefGpr(overflow_inst).cvt8() : INVALID_REG.cvt8();

    // TODO: Consider using LEA.
    // TODO: Optimize CMP case.
    // Note that x64 CF is inverse of what the ARM carry flag is here.

    if (b.IsImmediate()) {
        u32 op_arg = b.GetU32();
        if (carry_in.IsImmediate()) {
            if (carry_in.GetU1()) {
                code->sub(result, op_arg);
            } else {
                code->stc();
                code->sbb(result, op_arg);
            }
        } else {
            code->bt(carry.cvt32(), 0);
            code->cmc();
            code->sbb(result, op_arg);
        }
    } else {
        OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);
        op_arg.setBit(32);
        if (carry_in.IsImmediate()) {
            if (carry_in.GetU1()) {
                code->sub(result, *op_arg);
            } else {
                code->stc();
                code->sbb(result, *op_arg);
            }
        } else {
            code->bt(carry.cvt32(), 0);
            code->cmc();
            code->sbb(result, *op_arg);
        }
    }

    if (carry_inst) {
        EraseInstruction(block, carry_inst);
        inst->DecrementRemainingUses();
        code->setnc(carry);
    }
    if (overflow_inst) {
        EraseInstruction(block, overflow_inst);
        inst->DecrementRemainingUses();
        code->seto(overflow);
    }
}

void EmitX64::EmitSub64(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg64 result = reg_alloc.UseDefGpr(a, inst);
    Xbyak::Reg64 op_arg = reg_alloc.UseGpr(b);

    code->sub(result, op_arg);
}

void EmitX64::EmitMul(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);
    if (a.IsImmediate())
        std::swap(a, b);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
    if (b.IsImmediate()) {
        code->imul(result, result, b.GetU32());
    } else {
        OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);
        op_arg.setBit(32);

        code->imul(result, *op_arg);
    }
}

void EmitX64::EmitMul64(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg64 result = reg_alloc.UseDefGpr(a, inst);
    OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);

    code->imul(result, *op_arg);
}

void EmitX64::EmitAnd(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();

    if (b.IsImmediate()) {
        u32 op_arg = b.GetU32();

        code->and_(result, op_arg);
    } else {
        OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);
        op_arg.setBit(32);

        code->and_(result, *op_arg);
    }
}

void EmitX64::EmitEor(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();

    if (b.IsImmediate()) {
        u32 op_arg = b.GetU32();

        code->xor_(result, op_arg);
    } else {
        OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);
        op_arg.setBit(32);

        code->xor_(result, *op_arg);
    }
}

void EmitX64::EmitOr(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();

    if (b.IsImmediate()) {
        u32 op_arg = b.GetU32();

        code->or_(result, op_arg);
    } else {
        OpArg op_arg = reg_alloc.UseOpArg(b, any_gpr);
        op_arg.setBit(32);

        code->or_(result, *op_arg);
    }
}

void EmitX64::EmitNot(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    if (a.IsImmediate()) {
        Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();

        code->mov(result, u32(~a.GetU32()));
    } else {
        Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();

        code->not_(result);
    }
}

void EmitX64::EmitSignExtendWordToLong(IR::Block&, IR::Inst* inst) {
    OpArg source;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        source = result;
    } else {
        std::tie(source, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }

    source.setBit(32);
    code->movsxd(result.cvt64(), *source);
}

void EmitX64::EmitSignExtendHalfToWord(IR::Block&, IR::Inst* inst) {
    OpArg source;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        source = result;
    } else {
        std::tie(source, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }

    source.setBit(16);
    code->movsx(result.cvt32(), *source);
}

void EmitX64::EmitSignExtendByteToWord(IR::Block&, IR::Inst* inst) {
    OpArg source;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        source = result;
    } else {
        std::tie(source, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }

    source.setBit(8);
    code->movsx(result.cvt32(), *source);
}

void EmitX64::EmitZeroExtendWordToLong(IR::Block&, IR::Inst* inst) {
    OpArg source;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        source = result;
    } else {
        std::tie(source, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }

    source.setBit(32);
    code->mov(result.cvt32(), *source); // x64 zeros upper 32 bits on a 32-bit move
}

void EmitX64::EmitZeroExtendHalfToWord(IR::Block&, IR::Inst* inst) {
    OpArg source;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        source = result;
    } else {
        std::tie(source, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }

    source.setBit(16);
    code->movzx(result.cvt32(), *source);
}

void EmitX64::EmitZeroExtendByteToWord(IR::Block&, IR::Inst* inst) {
    OpArg source;
    Xbyak::Reg64 result;
    if (inst->GetArg(0).IsImmediate()) {
        // TODO: Optimize
        result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);
        source = result;
    } else {
        std::tie(source, result) = reg_alloc.UseDefOpArgGpr(inst->GetArg(0), inst);
    }

    source.setBit(8);
    code->movzx(result.cvt32(), *source);
}

void EmitX64::EmitByteReverseWord(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt32();

    code->bswap(result);
}

void EmitX64::EmitByteReverseHalf(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg16 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst).cvt16();

    code->rol(result, 8);
}

void EmitX64::EmitByteReverseDual(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg64 result = reg_alloc.UseDefGpr(inst->GetArg(0), inst);

    code->bswap(result);
}

void EmitX64::EmitCountLeadingZeros(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    if (cpu_info.has(Xbyak::util::Cpu::tLZCNT)) {
        Xbyak::Reg32 source = reg_alloc.UseGpr(a).cvt32();
        Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();

        code->lzcnt(result, source);
    } else {
        Xbyak::Reg32 source = reg_alloc.UseScratchGpr(a).cvt32();
        Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();

        // The result of a bsr of zero is undefined, but zf is set after it.
        code->bsr(result, source);
        code->mov(source, 0xFFFFFFFF);
        code->cmovz(result, source);
        code->neg(result);
        code->add(result, 31);
    }
}

void EmitX64::EmitNegateLowWord(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();

    code->ror(result, 16);
    code->xor(result, 0xFFFF0000);
    code->add(result, 0x00010000);
    code->ror(result, 16);
}

void EmitX64::EmitNegateHighWord(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();

    code->xor(result, 0xFFFF0000);
    code->add(result, 0x00010000);
}

void EmitX64::EmitSignedSaturatedAdd(IR::Block& block, IR::Inst* inst) {
    auto overflow_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetOverflowFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 addend = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 overflow = overflow_inst ? reg_alloc.DefGpr(overflow_inst).cvt32() : reg_alloc.ScratchGpr().cvt32();

    code->mov(overflow, result);
    code->shr(overflow, 31);
    code->add(overflow, 0x7FFFFFFF);
    // overflow now contains 0x7FFFFFFF if a was positive, or 0x80000000 if a was negative
    code->add(result, addend);
    code->cmovo(result, overflow);

    if (overflow_inst) {
        EraseInstruction(block, overflow_inst);
        inst->DecrementRemainingUses();

        code->seto(overflow.cvt8());
    }
}

void EmitX64::EmitSignedSaturatedSub(IR::Block& block, IR::Inst* inst) {
    auto overflow_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetOverflowFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 subend = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 overflow = overflow_inst ? reg_alloc.DefGpr(overflow_inst).cvt32() : reg_alloc.ScratchGpr().cvt32();

    code->mov(overflow, result);
    code->shr(overflow, 31);
    code->add(overflow, 0x7FFFFFFF);
    // overflow now contains 0x7FFFFFFF if a was positive, or 0x80000000 if a was negative
    code->sub(result, subend);
    code->cmovo(result, overflow);

    if (overflow_inst) {
        EraseInstruction(block, overflow_inst);
        inst->DecrementRemainingUses();

        code->seto(overflow.cvt8());
    }
}

void EmitX64::EmitUnsignedSaturation(IR::Block& block, IR::Inst* inst) {
    auto overflow_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetOverflowFromOp);

    IR::Value a = inst->GetArg(0);
    size_t N = inst->GetArg(1).GetU8();
    ASSERT(N <= 31);

    u32 saturated_value = (1u << N) - 1;

    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Reg32 reg_a = reg_alloc.UseGpr(a).cvt32();
    Xbyak::Reg32 overflow = overflow_inst ? reg_alloc.DefGpr(overflow_inst).cvt32() : reg_alloc.ScratchGpr().cvt32();

    // Pseudocode: result = clamp(reg_a, 0, saturated_value);
    code->xor_(overflow, overflow);
    code->cmp(reg_a, saturated_value);
    code->mov(result, saturated_value);
    code->cmovle(result, overflow);
    code->cmovbe(result, reg_a);

    if (overflow_inst) {
        EraseInstruction(block, overflow_inst);
        inst->DecrementRemainingUses();

        code->seta(overflow.cvt8());
    }
}

void EmitX64::EmitSignedSaturation(IR::Block& block, IR::Inst* inst) {
    auto overflow_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetOverflowFromOp);

    IR::Value a = inst->GetArg(0);
    size_t N = inst->GetArg(1).GetU8();
    ASSERT(N >= 1 && N <= 32);

    if (N == 32) {
        reg_alloc.RegisterAddDef(inst, a);
        if (overflow_inst) {
            auto no_overflow = IR::Value(false);
            overflow_inst->ReplaceUsesWith(no_overflow);
        }
        return;
    }

    u32 mask = (1u << N) - 1;
    u32 positive_saturated_value = (1u << (N - 1)) - 1;
    u32 negative_saturated_value = 1u << (N - 1);
    u32 sext_negative_satured_value = Common::SignExtend(N, negative_saturated_value);

    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Reg32 reg_a = reg_alloc.UseGpr(a).cvt32();
    Xbyak::Reg32 overflow = overflow_inst ? reg_alloc.DefGpr(overflow_inst).cvt32() : reg_alloc.ScratchGpr().cvt32();
    Xbyak::Reg32 tmp = reg_alloc.ScratchGpr().cvt32();

    // overflow now contains a value between 0 and mask if it was originally between {negative,positive}_saturated_value.
    code->lea(overflow, code->ptr[reg_a.cvt64() + negative_saturated_value]);

    // Put the appropriate saturated value in result
    code->cmp(reg_a, positive_saturated_value);
    code->mov(tmp, positive_saturated_value);
    code->mov(result, sext_negative_satured_value);
    code->cmovg(result, tmp);

    // Do the saturation
    code->cmp(overflow, mask);
    code->cmovbe(result, reg_a);

    if (overflow_inst) {
        EraseInstruction(block, overflow_inst);
        inst->DecrementRemainingUses();

        code->seta(overflow.cvt8());
    }
}

/**
 * Extracts the most significant bits from each of the packed bytes, and packs them together.
 *
 *     value before:    a-------b-------c-------d-------
 *     value after:     0000000000000000000000000000abcd
 *
 * @param value The register containing the value to operate on. Result will be stored in the same register.
 * @param a_tmp A register which can be used as a scratch register.
 */
static void ExtractMostSignificantBitFromPackedBytes(const Xbyak::util::Cpu& cpu_info, BlockOfCode* code, RegAlloc& reg_alloc, Xbyak::Reg32 value, boost::optional<Xbyak::Reg32> a_tmp = boost::none) {
    if (cpu_info.has(Xbyak::util::Cpu::tBMI2)) {
        Xbyak::Reg32 tmp = a_tmp ? *a_tmp : reg_alloc.ScratchGpr().cvt32();
        code->mov(tmp, 0x80808080);
        code->pext(value, value, tmp);
    } else {
        code->and_(value, 0x80808080);
        code->imul(value, value, 0x00204081);
        code->shr(value, 28);
    }
}

/**
 * Extracts the most significant bits from each of the packed words, duplicates them, and packs them together.
 *
 *     value before:    a---------------b---------------
 *     value after:     0000000000000000000000000000aabb
 *
 * @param value The register containing the value to operate on. Result will be stored in the same register.
 */
static void ExtractAndDuplicateMostSignificantBitFromPackedWords(BlockOfCode* code, Xbyak::Reg32 value) {
    code->and_(value, 0x80008000);
    code->shr(value, 1);
    code->imul(value, value, 0xC003);
    code->shr(value, 28);
}

void EmitX64::EmitPackedAddU8(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseScratchGpr(a).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseScratchGpr(b).cvt32();
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Reg32 reg_ge, tmp;

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
        tmp = reg_alloc.ScratchGpr().cvt32();

        code->mov(reg_ge, reg_a);
        code->and_(reg_ge, reg_b);
    }

    // SWAR Arithmetic
    code->mov(result, reg_a);
    code->xor_(result, reg_b);
    code->and_(result, 0x80808080);
    code->and_(reg_a, 0x7F7F7F7F);
    code->and_(reg_b, 0x7F7F7F7F);
    code->add(reg_a, reg_b);
    if (ge_inst) {
        code->mov(tmp, result);
        code->and_(tmp, reg_a);
        code->or_(reg_ge, tmp);
    }
    code->xor_(result, reg_a);
    if (ge_inst) {
        ExtractMostSignificantBitFromPackedBytes(cpu_info, code, reg_alloc, reg_ge, tmp);
    }
}

void EmitX64::EmitPackedAddS8(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 reg_ge;

    Xbyak::Xmm xmm_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_b = reg_alloc.ScratchXmm();

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
    }

    code->movd(xmm_a, reg_a);
    code->movd(xmm_b, reg_b);
    if (ge_inst) {
        Xbyak::Xmm saturated_sum = reg_alloc.ScratchXmm();
        code->movdqa(saturated_sum, xmm_a);
        code->paddsb(saturated_sum, xmm_b);
        code->movd(reg_ge, saturated_sum);
    }
    code->paddb(xmm_a, xmm_b);
    code->movd(reg_a, xmm_a);
    if (ge_inst) {
        code->not_(reg_ge);
        ExtractMostSignificantBitFromPackedBytes(cpu_info, code, reg_alloc, reg_ge);
    }
}

void EmitX64::EmitPackedAddU16(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseScratchGpr(a).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseScratchGpr(b).cvt32();
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Reg32 reg_ge, tmp;

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();

        code->mov(reg_ge, reg_a);
        code->and_(reg_ge, reg_b);
    }

    // SWAR Arithmetic
    code->mov(result, reg_a);
    code->xor_(result, reg_b);
    code->and_(result, 0x80008000);
    code->and_(reg_a, 0x7FFF7FFF);
    code->and_(reg_b, 0x7FFF7FFF);
    code->add(reg_a, reg_b);
    if (ge_inst) {
        tmp = reg_alloc.ScratchGpr().cvt32();
        code->mov(tmp, result);
        code->and_(tmp, reg_a);
        code->or_(reg_ge, tmp);
    }
    code->xor_(result, reg_a);
    if (ge_inst) {
        ExtractAndDuplicateMostSignificantBitFromPackedWords(code, reg_ge);
    }
}

void EmitX64::EmitPackedAddS16(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 reg_ge;

    Xbyak::Xmm xmm_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_b = reg_alloc.ScratchXmm();

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
    }

    code->movd(xmm_a, reg_a);
    code->movd(xmm_b, reg_b);
    if (ge_inst) {
        Xbyak::Xmm saturated_sum = reg_alloc.ScratchXmm();
        code->movdqa(saturated_sum, xmm_a);
        code->paddsw(saturated_sum, xmm_b);
        code->movd(reg_ge, saturated_sum);
    }
    code->paddw(xmm_a, xmm_b);
    code->movd(reg_a, xmm_a);
    if (ge_inst) {
        code->not_(reg_ge);
        ExtractAndDuplicateMostSignificantBitFromPackedWords(code, reg_ge);
    }
}

void EmitX64::EmitPackedSubU8(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 reg_ge;

    Xbyak::Xmm xmm_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_b = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_ge;

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
        xmm_ge = reg_alloc.ScratchXmm();
    }

    code->movd(xmm_a, reg_a);
    code->movd(xmm_b, reg_b);
    if (ge_inst) {
        code->movdqa(xmm_ge, xmm_a);
        code->pmaxub(xmm_ge, xmm_b);
        code->pcmpeqb(xmm_ge, xmm_a);
        code->movd(reg_ge, xmm_ge);
    }
    code->psubb(xmm_a, xmm_b);
    code->movd(reg_a, xmm_a);

    if (ge_inst) {
        ExtractMostSignificantBitFromPackedBytes(cpu_info, code, reg_alloc, reg_ge);
    }
}


void EmitX64::EmitPackedSubS8(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 reg_ge;

    Xbyak::Xmm xmm_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_b = reg_alloc.ScratchXmm();

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();
        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
    }
    code->movd(xmm_b, reg_b);
    code->movd(xmm_a, reg_a);
    if (ge_inst) {
        Xbyak::Xmm xmm_ge = reg_alloc.ScratchXmm();
        code->movdqa(xmm_ge, xmm_a);
        code->psubsb(xmm_ge, xmm_b);
        code->movd(reg_ge, xmm_ge);
    }
    code->psubb(xmm_a, xmm_b);
    code->movd(reg_a, xmm_a);
    if (ge_inst) {
        code->not_(reg_ge);
        ExtractMostSignificantBitFromPackedBytes(cpu_info, code, reg_alloc, reg_ge);
    }
}

void EmitX64::EmitPackedSubU16(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 reg_ge;

    Xbyak::Xmm xmm_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_b = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_ge;

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
        xmm_ge = reg_alloc.ScratchXmm();
    }

    code->movd(xmm_a, reg_a);
    code->movd(xmm_b, reg_b);
    if (ge_inst) {
        code->movdqa(xmm_ge, xmm_a);
        code->pmaxuw(xmm_ge, xmm_b);
        code->pcmpeqw(xmm_ge, xmm_a);
        code->movd(reg_ge, xmm_ge);
    }
    code->psubw(xmm_a, xmm_b);
    code->movd(reg_a, xmm_a);
    if (ge_inst) {
        ExtractAndDuplicateMostSignificantBitFromPackedWords(code, reg_ge);
    }
}

void EmitX64::EmitPackedSubS16(IR::Block& block, IR::Inst* inst) {
    auto ge_inst = inst->GetAssociatedPseudoOperation(IR::Opcode::GetGEFromOp);

    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 reg_ge;

    Xbyak::Xmm xmm_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_b = reg_alloc.ScratchXmm();

    if (ge_inst) {
        EraseInstruction(block, ge_inst);
        inst->DecrementRemainingUses();

        reg_ge = reg_alloc.DefGpr(ge_inst).cvt32();
    }

    code->movd(xmm_b, reg_b);
    code->movd(xmm_a, reg_a);
    if (ge_inst) {
        Xbyak::Xmm xmm_ge = reg_alloc.ScratchXmm();
        code->movdqa(xmm_ge, xmm_a);
        code->psubsw(xmm_ge, xmm_b);
        code->movd(reg_ge, xmm_ge);
    }
    code->psubw(xmm_a, xmm_b);
    code->movd(reg_a, xmm_a);
    if (ge_inst) {
        code->not_(reg_ge);
        ExtractAndDuplicateMostSignificantBitFromPackedWords(code, reg_ge);
    }
}

void EmitX64::EmitPackedHalvingAddU8(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    // This code path requires SSSE3 because of the PSHUFB instruction.
    // A fallback implementation is provided below.
    if (cpu_info.has(Xbyak::util::Cpu::tSSSE3)) {
        Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
        Xbyak::Reg32 arg = reg_alloc.UseGpr(b).cvt32();

        // Load the operands into Xmm registers
        Xbyak::Xmm xmm_scratch_a = reg_alloc.ScratchXmm();
        Xbyak::Xmm xmm_scratch_b = reg_alloc.ScratchXmm();

        Xbyak::Xmm xmm_mask = reg_alloc.ScratchXmm();
        Xbyak::Reg64 mask = reg_alloc.ScratchGpr();

        code->movd(xmm_scratch_a, result);
        code->movd(xmm_scratch_b, arg);

        // Set the mask to expand the values
        // 0xAABBCCDD becomes 0x00AA00BB00CC00DD
        code->mov(mask, 0x8003800280018000);
        code->movq(xmm_mask, mask);

        // Expand each 8-bit value to 16-bit
        code->pshufb(xmm_scratch_a, xmm_mask);
        code->pshufb(xmm_scratch_b, xmm_mask);

        // Add the individual 16-bit values
        code->paddw(xmm_scratch_a, xmm_scratch_b);

        // Shift the 16-bit values to the right to halve them
        code->psrlw(xmm_scratch_a, 1);

        // Set the mask to pack the values again
        // 0x00AA00BB00CC00DD becomes 0xAABBCCDD
        code->mov(mask, 0x06040200);
        code->movq(xmm_mask, mask);

        // Shuffle them back to 8-bit values
        code->pshufb(xmm_scratch_a, xmm_mask);

        code->movd(result, xmm_scratch_a);
        return;
    }

    // Fallback implementation in case the CPU doesn't support SSSE3
    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 xor_a_b = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Reg32 and_a_b = reg_a;
    Xbyak::Reg32 result = reg_a;

    code->mov(xor_a_b, reg_a);
    code->and(and_a_b, reg_b);
    code->xor(xor_a_b, reg_b);
    code->shr(xor_a_b, 1);
    code->and(xor_a_b, 0x7F7F7F7F);
    code->add(result, xor_a_b);
}

void EmitX64::EmitPackedHalvingAddU16(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 xor_a_b = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Reg32 and_a_b = reg_a;
    Xbyak::Reg32 result = reg_a;

    // This relies on the equality x+y == ((x&y) << 1) + (x^y).
    // Note that x^y always contains the LSB of the result.
    // Since we want to calculate (x+y)/2, we can instead calculate (x&y) + ((x^y)>>1).
    // We mask by 0x7FFF to remove the LSB so that it doesn't leak into the field below.

    code->mov(xor_a_b, reg_a);
    code->and(and_a_b, reg_b);
    code->xor(xor_a_b, reg_b);
    code->shr(xor_a_b, 1);
    code->and(xor_a_b, 0x7FFF7FFF);
    code->add(result, xor_a_b);
}

void EmitX64::EmitPackedHalvingAddS8(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 xor_a_b = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Reg32 and_a_b = reg_a;
    Xbyak::Reg32 result = reg_a;
    Xbyak::Reg32 carry = reg_alloc.ScratchGpr().cvt32();

    // This relies on the equality x+y == ((x&y) << 1) + (x^y).
    // Note that x^y always contains the LSB of the result.
    // Since we want to calculate (x+y)/2, we can instead calculate (x&y) + ((x^y)>>1).
    // We mask by 0x7F to remove the LSB so that it doesn't leak into the field below.
    // carry propagates the sign bit from (x^y)>>1 upwards by one.

    code->mov(xor_a_b, reg_a);
    code->and(and_a_b, reg_b);
    code->xor(xor_a_b, reg_b);
    code->mov(carry, xor_a_b);
    code->and(carry, 0x80808080);
    code->shr(xor_a_b, 1);
    code->and(xor_a_b, 0x7F7F7F7F);
    code->add(result, xor_a_b);
    code->xor(result, carry);
}

void EmitX64::EmitPackedHalvingAddS16(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 reg_a = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 reg_b = reg_alloc.UseGpr(b).cvt32();
    Xbyak::Reg32 xor_a_b = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Reg32 and_a_b = reg_a;
    Xbyak::Reg32 result = reg_a;
    Xbyak::Reg32 carry = reg_alloc.ScratchGpr().cvt32();

    // This relies on the equality x+y == ((x&y) << 1) + (x^y).
    // Note that x^y always contains the LSB of the result.
    // Since we want to calculate (x+y)/2, we can instead calculate (x&y) + ((x^y)>>1).
    // We mask by 0x7FFF to remove the LSB so that it doesn't leak into the field below.
    // carry propagates the sign bit from (x^y)>>1 upwards by one.

    code->mov(xor_a_b, reg_a);
    code->and(and_a_b, reg_b);
    code->xor(xor_a_b, reg_b);
    code->mov(carry, xor_a_b);
    code->and(carry, 0x80008000);
    code->shr(xor_a_b, 1);
    code->and(xor_a_b, 0x7FFF7FFF);
    code->add(result, xor_a_b);
    code->xor(result, carry);
}

void EmitX64::EmitPackedHalvingSubU8(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 minuend = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 subtrahend = reg_alloc.UseScratchGpr(b).cvt32();

    // This relies on the equality x-y == (x^y) - (((x^y)&y) << 1).
    // Note that x^y always contains the LSB of the result.
    // Since we want to calculate (x+y)/2, we can instead calculate ((x^y)>>1) - ((x^y)&y).

    code->xor(minuend, subtrahend);
    code->and(subtrahend, minuend);
    code->shr(minuend, 1);

    // At this point,
    // minuend := (a^b) >> 1
    // subtrahend := (a^b) & b

    // We must now perform a partitioned subtraction.
    // We can do this because minuend contains 7 bit fields.
    // We use the extra bit in minuend as a bit to borrow from; we set this bit.
    // We invert this bit at the end as this tells us if that bit was borrowed from.
    code->or(minuend, 0x80808080);
    code->sub(minuend, subtrahend);
    code->xor(minuend, 0x80808080);

    // minuend now contains the desired result.
}

void EmitX64::EmitPackedHalvingSubU16(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 minuend = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 subtrahend = reg_alloc.UseScratchGpr(b).cvt32();

    // This relies on the equality x-y == (x^y) - (((x^y)&y) << 1).
    // Note that x^y always contains the LSB of the result.
    // Since we want to calculate (x+y)/2, we can instead calculate ((x^y)>>1) - ((x^y)&y).

    code->xor(minuend, subtrahend);
    code->and(subtrahend, minuend);
    code->shr(minuend, 1);

    // At this point,
    // minuend := (a^b) >> 1
    // subtrahend := (a^b) & b

    // We must now perform a partitioned subtraction.
    // We can do this because minuend contains 15 bit fields.
    // We use the extra bit in minuend as a bit to borrow from; we set this bit.
    // We invert this bit at the end as this tells us if that bit was borrowed from.
    code->or(minuend, 0x80008000);
    code->sub(minuend, subtrahend);
    code->xor(minuend, 0x80008000);

    // minuend now contains the desired result.
}

static void EmitPackedOperation(BlockOfCode* code, RegAlloc& reg_alloc, IR::Inst* inst, void (Xbyak::CodeGenerator::*fn)(const Xbyak::Mmx& mmx, const Xbyak::Operand&)) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Reg32 result = reg_alloc.UseDefGpr(a, inst).cvt32();
    Xbyak::Reg32 arg = reg_alloc.UseGpr(b).cvt32();

    Xbyak::Xmm xmm_scratch_a = reg_alloc.ScratchXmm();
    Xbyak::Xmm xmm_scratch_b = reg_alloc.ScratchXmm();

    code->movd(xmm_scratch_a, result);
    code->movd(xmm_scratch_b, arg);

    (code->*fn)(xmm_scratch_a, xmm_scratch_b);

    code->movd(result, xmm_scratch_a);
}

void EmitX64::EmitPackedSaturatedAddU8(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::paddusb);
}

void EmitX64::EmitPackedSaturatedAddS8(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::paddsb);
}

void EmitX64::EmitPackedSaturatedSubU8(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::psubusb);
}

void EmitX64::EmitPackedSaturatedSubS8(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::psubsb);
}

void EmitX64::EmitPackedSaturatedAddU16(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::paddusw);
}

void EmitX64::EmitPackedSaturatedAddS16(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::paddsw);
}

void EmitX64::EmitPackedSaturatedSubU16(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::psubusw);
}

void EmitX64::EmitPackedSaturatedSubS16(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::psubsw);
}

void EmitX64::EmitPackedAbsDiffSumS8(IR::Block&, IR::Inst* inst) {
    EmitPackedOperation(code, reg_alloc, inst, &Xbyak::CodeGenerator::psadbw);
}

static void DenormalsAreZero32(BlockOfCode* code, Xbyak::Xmm xmm_value, Xbyak::Reg32 gpr_scratch) {
    using namespace Xbyak::util;
    Xbyak::Label end;

    // We need to report back whether we've found a denormal on input.
    // SSE doesn't do this for us when SSE's DAZ is enabled.

    code->movd(gpr_scratch, xmm_value);
    code->and_(gpr_scratch, u32(0x7FFFFFFF));
    code->sub(gpr_scratch, u32(1));
    code->cmp(gpr_scratch, u32(0x007FFFFE));
    code->ja(end);
    code->pxor(xmm_value, xmm_value);
    code->mov(dword[r15 + offsetof(JitState, FPSCR_IDC)], u32(1 << 7));
    code->L(end);
}

static void DenormalsAreZero64(BlockOfCode* code, Xbyak::Xmm xmm_value, Xbyak::Reg64 gpr_scratch) {
    using namespace Xbyak::util;
    Xbyak::Label end;

    auto mask = code->MFloatNonSignMask64();
    mask.setBit(64);
    auto penult_denormal = code->MFloatPenultimatePositiveDenormal64();
    penult_denormal.setBit(64);

    code->movq(gpr_scratch, xmm_value);
    code->and_(gpr_scratch, mask);
    code->sub(gpr_scratch, u32(1));
    code->cmp(gpr_scratch, penult_denormal);
    code->ja(end);
    code->pxor(xmm_value, xmm_value);
    code->mov(dword[r15 + offsetof(JitState, FPSCR_IDC)], u32(1 << 7));
    code->L(end);
}

static void FlushToZero32(BlockOfCode* code, Xbyak::Xmm xmm_value, Xbyak::Reg32 gpr_scratch) {
    using namespace Xbyak::util;
    Xbyak::Label end;

    code->movd(gpr_scratch, xmm_value);
    code->and_(gpr_scratch, u32(0x7FFFFFFF));
    code->sub(gpr_scratch, u32(1));
    code->cmp(gpr_scratch, u32(0x007FFFFE));
    code->ja(end);
    code->pxor(xmm_value, xmm_value);
    code->mov(dword[r15 + offsetof(JitState, FPSCR_UFC)], u32(1 << 3));
    code->L(end);
}

static void FlushToZero64(BlockOfCode* code, Xbyak::Xmm xmm_value, Xbyak::Reg64 gpr_scratch) {
    using namespace Xbyak::util;
    Xbyak::Label end;

    auto mask = code->MFloatNonSignMask64();
    mask.setBit(64);
    auto penult_denormal = code->MFloatPenultimatePositiveDenormal64();
    penult_denormal.setBit(64);

    code->movq(gpr_scratch, xmm_value);
    code->and_(gpr_scratch, mask);
    code->sub(gpr_scratch, u32(1));
    code->cmp(gpr_scratch, penult_denormal);
    code->ja(end);
    code->pxor(xmm_value, xmm_value);
    code->mov(dword[r15 + offsetof(JitState, FPSCR_UFC)], u32(1 << 3));
    code->L(end);
}

static void DefaultNaN32(BlockOfCode* code, Xbyak::Xmm xmm_value) {
    Xbyak::Label end;

    code->ucomiss(xmm_value, xmm_value);
    code->jnp(end);
    code->movaps(xmm_value, code->MFloatNaN32());
    code->L(end);
}

static void DefaultNaN64(BlockOfCode* code, Xbyak::Xmm xmm_value) {
    Xbyak::Label end;

    code->ucomisd(xmm_value, xmm_value);
    code->jnp(end);
    code->movaps(xmm_value, code->MFloatNaN64());
    code->L(end);
}

static void ZeroIfNaN64(BlockOfCode* code, Xbyak::Xmm xmm_value, Xbyak::Xmm xmm_scratch) {
    code->pxor(xmm_scratch, xmm_scratch);
    code->cmpordsd(xmm_scratch, xmm_value); // true mask when ordered (i.e.: when not an NaN)
    code->pand(xmm_value, xmm_scratch);
}

static void FPThreeOp32(BlockOfCode* code, RegAlloc& reg_alloc, IR::Block& block, IR::Inst* inst, void (Xbyak::CodeGenerator::*fn)(const Xbyak::Xmm&, const Xbyak::Operand&)) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);
    Xbyak::Xmm operand = reg_alloc.UseXmm(b);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero32(code, result, gpr_scratch);
        DenormalsAreZero32(code, operand, gpr_scratch);
    }
    (code->*fn)(result, operand);
    if (block.Location().FPSCR().FTZ()) {
        FlushToZero32(code, result, gpr_scratch);
    }
    if (block.Location().FPSCR().DN()) {
        DefaultNaN32(code, result);
    }
}

static void FPThreeOp64(BlockOfCode* code, RegAlloc& reg_alloc, IR::Block& block, IR::Inst* inst, void (Xbyak::CodeGenerator::*fn)(const Xbyak::Xmm&, const Xbyak::Operand&)) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);
    Xbyak::Xmm operand = reg_alloc.UseXmm(b);
    Xbyak::Reg64 gpr_scratch = reg_alloc.ScratchGpr();

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero64(code, result, gpr_scratch);
        DenormalsAreZero64(code, operand, gpr_scratch);
    }
    (code->*fn)(result, operand);
    if (block.Location().FPSCR().FTZ()) {
        FlushToZero64(code, result, gpr_scratch);
    }
    if (block.Location().FPSCR().DN()) {
        DefaultNaN64(code, result);
    }
}

static void FPTwoOp32(BlockOfCode* code, RegAlloc& reg_alloc, IR::Block& block, IR::Inst* inst, void (Xbyak::CodeGenerator::*fn)(const Xbyak::Xmm&, const Xbyak::Operand&)) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero32(code, result, gpr_scratch);
    }

    (code->*fn)(result, result);
    if (block.Location().FPSCR().FTZ()) {
        FlushToZero32(code, result, gpr_scratch);
    }
    if (block.Location().FPSCR().DN()) {
        DefaultNaN32(code, result);
    }
}

static void FPTwoOp64(BlockOfCode* code, RegAlloc& reg_alloc, IR::Block& block, IR::Inst* inst, void (Xbyak::CodeGenerator::*fn)(const Xbyak::Xmm&, const Xbyak::Operand&)) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);
    Xbyak::Reg64 gpr_scratch = reg_alloc.ScratchGpr();

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero64(code, result, gpr_scratch);
    }

    (code->*fn)(result, result);
    if (block.Location().FPSCR().FTZ()) {
        FlushToZero64(code, result, gpr_scratch);
    }
    if (block.Location().FPSCR().DN()) {
        DefaultNaN64(code, result);
    }
}

void EmitX64::EmitTransferFromFP32(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg32 result = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Xmm source = reg_alloc.UseXmm(inst->GetArg(0));
    // TODO: Eliminate this.
    code->movd(result, source);
}

void EmitX64::EmitTransferFromFP64(IR::Block&, IR::Inst* inst) {
    Xbyak::Reg64 result = reg_alloc.DefGpr(inst);
    Xbyak::Xmm source = reg_alloc.UseXmm(inst->GetArg(0));
    // TODO: Eliminate this.
    code->movq(result, source);
}

void EmitX64::EmitTransferToFP32(IR::Block&, IR::Inst* inst) {
    if (inst->GetArg(0).IsImmediate() && inst->GetArg(0).GetU32() == 0) {
        Xbyak::Xmm result = reg_alloc.DefXmm(inst);
        code->xorps(result, result);
    } else {
        Xbyak::Xmm result = reg_alloc.DefXmm(inst);
        Xbyak::Reg32 source = reg_alloc.UseGpr(inst->GetArg(0)).cvt32();
        // TODO: Eliminate this.
        code->movd(result, source);
    }
}

void EmitX64::EmitTransferToFP64(IR::Block&, IR::Inst* inst) {
    if (inst->GetArg(0).IsImmediate() && inst->GetArg(0).GetU64() == 0) {
        Xbyak::Xmm result = reg_alloc.DefXmm(inst);
        code->xorpd(result, result);
    } else {
        Xbyak::Xmm result = reg_alloc.DefXmm(inst);
        Xbyak::Reg64 source = reg_alloc.UseGpr(inst->GetArg(0));
        // TODO: Eliminate this.
        code->movq(result, source);
    }
}

void EmitX64::EmitFPAbs32(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);

    code->pand(result, code->MFloatNonSignMask32());
}

void EmitX64::EmitFPAbs64(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);

    code->pand(result, code->MFloatNonSignMask64());
}

void EmitX64::EmitFPNeg32(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);

    code->pxor(result, code->MFloatNegativeZero32());
}

void EmitX64::EmitFPNeg64(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);

    code->pxor(result, code->MFloatNegativeZero64());
}

void EmitX64::EmitFPAdd32(IR::Block& block, IR::Inst* inst) {
    FPThreeOp32(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::addss);
}

void EmitX64::EmitFPAdd64(IR::Block& block, IR::Inst* inst) {
    FPThreeOp64(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::addsd);
}

void EmitX64::EmitFPDiv32(IR::Block& block, IR::Inst* inst) {
    FPThreeOp32(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::divss);
}

void EmitX64::EmitFPDiv64(IR::Block& block, IR::Inst* inst) {
    FPThreeOp64(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::divsd);
}

void EmitX64::EmitFPMul32(IR::Block& block, IR::Inst* inst) {
    FPThreeOp32(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::mulss);
}

void EmitX64::EmitFPMul64(IR::Block& block, IR::Inst* inst) {
    FPThreeOp64(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::mulsd);
}

void EmitX64::EmitFPSqrt32(IR::Block& block, IR::Inst* inst) {
    FPTwoOp32(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::sqrtss);
}

void EmitX64::EmitFPSqrt64(IR::Block& block, IR::Inst* inst) {
    FPTwoOp64(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::sqrtsd);
}

void EmitX64::EmitFPSub32(IR::Block& block, IR::Inst* inst) {
    FPThreeOp32(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::subss);
}

void EmitX64::EmitFPSub64(IR::Block& block, IR::Inst* inst) {
    FPThreeOp64(code, reg_alloc, block, inst, &Xbyak::CodeGenerator::subsd);
}

static void SetFpscrNzcvFromFlags(BlockOfCode* code, RegAlloc& reg_alloc) {
    reg_alloc.ScratchGpr({HostLoc::RAX}); // lahf requires use of ah
    Xbyak::Reg32 nzcv_imm = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Reg32 nzcv = reg_alloc.ScratchGpr().cvt32();

    using namespace Xbyak::util;

    code->lahf();
    code->mov(nzcv_imm, 0x30000000);
    code->cmp(ah, 0b01000111);
    code->cmove(nzcv, nzcv_imm);
    code->mov(nzcv_imm, 0x20000000);
    code->cmp(ah, 0b00000010);
    code->cmove(nzcv, nzcv_imm);
    code->mov(nzcv_imm, 0x80000000);
    code->cmp(ah, 0b00000011);
    code->cmove(nzcv, nzcv_imm);
    code->mov(nzcv_imm, 0x60000000);
    code->cmp(ah, 0b01000010);
    code->cmove(nzcv, nzcv_imm);
    code->mov(dword[r15 + offsetof(JitState, FPSCR_nzcv)], nzcv);
}

void EmitX64::EmitFPCompare32(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);
    bool quiet = inst->GetArg(2).GetU1();

    Xbyak::Xmm reg_a = reg_alloc.UseXmm(a);
    Xbyak::Xmm reg_b = reg_alloc.UseXmm(b);

    if (quiet) {
        code->ucomiss(reg_a, reg_b);
    } else {
        code->comiss(reg_a, reg_b);
    }

    SetFpscrNzcvFromFlags(code, reg_alloc);
}

void EmitX64::EmitFPCompare64(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    IR::Value b = inst->GetArg(1);
    bool quiet = inst->GetArg(2).GetU1();

    Xbyak::Xmm reg_a = reg_alloc.UseXmm(a);
    Xbyak::Xmm reg_b = reg_alloc.UseXmm(b);

    if (quiet) {
        code->ucomisd(reg_a, reg_b);
    } else {
        code->comisd(reg_a, reg_b);
    }

    SetFpscrNzcvFromFlags(code, reg_alloc);
}

void EmitX64::EmitFPSingleToDouble(IR::Block& block, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);
    Xbyak::Reg64 gpr_scratch = reg_alloc.ScratchGpr();

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero32(code, result, gpr_scratch.cvt32());
    }
    code->cvtss2sd(result, result);
    if (block.Location().FPSCR().FTZ()) {
        FlushToZero64(code, result, gpr_scratch);
    }
    if (block.Location().FPSCR().DN()) {
        DefaultNaN64(code, result);
    }
}

void EmitX64::EmitFPDoubleToSingle(IR::Block& block, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);

    Xbyak::Xmm result = reg_alloc.UseDefXmm(a, inst);
    Xbyak::Reg64 gpr_scratch = reg_alloc.ScratchGpr();

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero64(code, result, gpr_scratch);
    }
    code->cvtsd2ss(result, result);
    if (block.Location().FPSCR().FTZ()) {
        FlushToZero32(code, result, gpr_scratch.cvt32());
    }
    if (block.Location().FPSCR().DN()) {
        DefaultNaN32(code, result);
    }
}

void EmitX64::EmitFPSingleToS32(IR::Block& block, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_towards_zero = inst->GetArg(1).GetU1();

    Xbyak::Xmm from = reg_alloc.UseScratchXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Xmm xmm_scratch = reg_alloc.ScratchXmm();

    // ARM saturates on conversion; this differs from x64 which returns a sentinel value.
    // Conversion to double is lossless, and allows for clamping.

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero32(code, from, gpr_scratch);
    }
    code->cvtss2sd(from, from);
    // First time is to set flags
    if (round_towards_zero) {
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
    } else {
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
    }
    // Clamp to output range
    ZeroIfNaN64(code, from, xmm_scratch);
    code->minsd(from, code->MFloatMaxS32());
    code->maxsd(from, code->MFloatMinS32());
    // Second time is for real
    if (round_towards_zero) {
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
    } else {
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
    }
    code->movd(to, gpr_scratch);
}

void EmitX64::EmitFPSingleToU32(IR::Block& block, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_towards_zero = inst->GetArg(1).GetU1();

    Xbyak::Xmm from = reg_alloc.UseScratchXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Xmm xmm_scratch = reg_alloc.ScratchXmm();

    // ARM saturates on conversion; this differs from x64 which returns a sentinel value.
    // Conversion to double is lossless, and allows for accurate clamping.
    //
    // Since SSE2 doesn't provide an unsigned conversion, we shift the range as appropriate.
    //
    // FIXME: Inexact exception not correctly signalled with the below code

    if (block.Location().FPSCR().RMode() != Arm::FPSCR::RoundingMode::TowardsZero && !round_towards_zero) {
        if (block.Location().FPSCR().FTZ()) {
            DenormalsAreZero32(code, from, gpr_scratch);
        }
        code->cvtss2sd(from, from);
        ZeroIfNaN64(code, from, xmm_scratch);
        // Bring into SSE range
        code->addsd(from, code->MFloatMinS32());
        // First time is to set flags
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
        // Clamp to output range
        code->minsd(from, code->MFloatMaxS32());
        code->maxsd(from, code->MFloatMinS32());
        // Actually convert
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
        // Bring back into original range
        code->add(gpr_scratch, u32(2147483648u));
        code->movd(to, gpr_scratch);
    } else {
        Xbyak::Xmm xmm_mask = reg_alloc.ScratchXmm();
        Xbyak::Reg32 gpr_mask = reg_alloc.ScratchGpr().cvt32();

        if (block.Location().FPSCR().FTZ()) {
            DenormalsAreZero32(code, from, gpr_scratch);
        }
        code->cvtss2sd(from, from);
        ZeroIfNaN64(code, from, xmm_scratch);
        // Generate masks if out-of-signed-range
        code->movaps(xmm_mask, code->MFloatMaxS32());
        code->cmpltsd(xmm_mask, from);
        code->movd(gpr_mask, xmm_mask);
        code->pand(xmm_mask, code->MFloatMinS32());
        code->and_(gpr_mask, u32(2147483648u));
        // Bring into range if necessary
        code->addsd(from, xmm_mask);
        // First time is to set flags
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
        // Clamp to output range
        code->minsd(from, code->MFloatMaxS32());
        code->maxsd(from, code->MFloatMinU32());
        // Actually convert
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
        // Bring back into original range if necessary
        code->add(gpr_scratch, gpr_mask);
        code->movd(to, gpr_scratch);
    }
}

void EmitX64::EmitFPDoubleToS32(IR::Block& block, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_towards_zero = inst->GetArg(1).GetU1();

    Xbyak::Xmm from = reg_alloc.UseScratchXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Xmm xmm_scratch = reg_alloc.ScratchXmm();

    // ARM saturates on conversion; this differs from x64 which returns a sentinel value.

    if (block.Location().FPSCR().FTZ()) {
        DenormalsAreZero64(code, from, gpr_scratch.cvt64());
    }
    // First time is to set flags
    if (round_towards_zero) {
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
    } else {
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
    }
    // Clamp to output range
    ZeroIfNaN64(code, from, xmm_scratch);
    code->minsd(from, code->MFloatMaxS32());
    code->maxsd(from, code->MFloatMinS32());
    // Second time is for real
    if (round_towards_zero) {
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
    } else {
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
    }
    code->movd(to, gpr_scratch);
}

void EmitX64::EmitFPDoubleToU32(IR::Block& block, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_towards_zero = inst->GetArg(1).GetU1();

    Xbyak::Xmm from = reg_alloc.UseScratchXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();
    Xbyak::Xmm xmm_scratch = reg_alloc.ScratchXmm();

    // ARM saturates on conversion; this differs from x64 which returns a sentinel value.
    // TODO: Use VCVTPD2UDQ when AVX512VL is available.
    // FIXME: Inexact exception not correctly signalled with the below code

    if (block.Location().FPSCR().RMode() != Arm::FPSCR::RoundingMode::TowardsZero && !round_towards_zero) {
        if (block.Location().FPSCR().FTZ()) {
            DenormalsAreZero64(code, from, gpr_scratch.cvt64());
        }
        ZeroIfNaN64(code, from, xmm_scratch);
        // Bring into SSE range
        code->addsd(from, code->MFloatMinS32());
        // First time is to set flags
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
        // Clamp to output range
        code->minsd(from, code->MFloatMaxS32());
        code->maxsd(from, code->MFloatMinS32());
        // Actually convert
        code->cvtsd2si(gpr_scratch, from); // 32 bit gpr
        // Bring back into original range
        code->add(gpr_scratch, u32(2147483648u));
        code->movd(to, gpr_scratch);
    } else {
        Xbyak::Xmm xmm_mask = reg_alloc.ScratchXmm();
        Xbyak::Reg32 gpr_mask = reg_alloc.ScratchGpr().cvt32();

        if (block.Location().FPSCR().FTZ()) {
            DenormalsAreZero64(code, from, gpr_scratch.cvt64());
        }
        ZeroIfNaN64(code, from, xmm_scratch);
        // Generate masks if out-of-signed-range
        code->movaps(xmm_mask, code->MFloatMaxS32());
        code->cmpltsd(xmm_mask, from);
        code->movd(gpr_mask, xmm_mask);
        code->pand(xmm_mask, code->MFloatMinS32());
        code->and_(gpr_mask, u32(2147483648u));
        // Bring into range if necessary
        code->addsd(from, xmm_mask);
        // First time is to set flags
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
        // Clamp to output range
        code->minsd(from, code->MFloatMaxS32());
        code->maxsd(from, code->MFloatMinU32());
        // Actually convert
        code->cvttsd2si(gpr_scratch, from); // 32 bit gpr
        // Bring back into original range if necessary
        code->add(gpr_scratch, gpr_mask);
        code->movd(to, gpr_scratch);
    }
}

void EmitX64::EmitFPS32ToSingle(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_to_nearest = inst->GetArg(1).GetU1();
    ASSERT_MSG(!round_to_nearest, "round_to_nearest unimplemented");

    Xbyak::Xmm from = reg_alloc.UseXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();

    code->movd(gpr_scratch, from);
    code->cvtsi2ss(to, gpr_scratch);
}

void EmitX64::EmitFPU32ToSingle(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_to_nearest = inst->GetArg(1).GetU1();
    ASSERT_MSG(!round_to_nearest, "round_to_nearest unimplemented");

    Xbyak::Xmm from = reg_alloc.UseXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    // Use a 64-bit register to ensure we don't end up treating the input as signed
    Xbyak::Reg64 gpr_scratch = reg_alloc.ScratchGpr();

    code->movq(gpr_scratch, from);
    code->cvtsi2ss(to, gpr_scratch);
}

void EmitX64::EmitFPS32ToDouble(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_to_nearest = inst->GetArg(1).GetU1();
    ASSERT_MSG(!round_to_nearest, "round_to_nearest unimplemented");

    Xbyak::Xmm from = reg_alloc.UseXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    Xbyak::Reg32 gpr_scratch = reg_alloc.ScratchGpr().cvt32();

    code->movd(gpr_scratch, from);
    code->cvtsi2sd(to, gpr_scratch);
}

void EmitX64::EmitFPU32ToDouble(IR::Block&, IR::Inst* inst) {
    IR::Value a = inst->GetArg(0);
    bool round_to_nearest = inst->GetArg(1).GetU1();
    ASSERT_MSG(!round_to_nearest, "round_to_nearest unimplemented");

    Xbyak::Xmm from = reg_alloc.UseXmm(a);
    Xbyak::Xmm to = reg_alloc.DefXmm(inst);
    // Use a 64-bit register to ensure we don't end up treating the input as signed
    Xbyak::Reg64 gpr_scratch = reg_alloc.ScratchGpr();

    code->movq(gpr_scratch, from);
    code->cvtsi2sd(to, gpr_scratch);
}


void EmitX64::EmitClearExclusive(IR::Block&, IR::Inst*) {
    using namespace Xbyak::util;

    code->mov(code->byte[r15 + offsetof(JitState, exclusive_state)], u8(0));
}

void EmitX64::EmitSetExclusive(IR::Block&, IR::Inst* inst) {
    using namespace Xbyak::util;

    ASSERT(inst->GetArg(1).IsImmediate());
    Xbyak::Reg32 address = reg_alloc.UseGpr(inst->GetArg(0)).cvt32();

    code->mov(code->byte[r15 + offsetof(JitState, exclusive_state)], u8(1));
    code->mov(dword[r15 + offsetof(JitState, exclusive_address)], address);
}

template <typename FunctionPointer>
static void ReadMemory(BlockOfCode* code, RegAlloc& reg_alloc, IR::Inst* inst, UserCallbacks& cb, size_t bit_size, FunctionPointer fn) {
    if (!cb.page_table) {
        reg_alloc.HostCall(inst, inst->GetArg(0));
        code->CallFunction(fn);
        return;
    }

    using namespace Xbyak::util;

    Xbyak::Reg64 result = reg_alloc.DefGpr(inst, { ABI_RETURN });
    Xbyak::Reg32 vaddr = reg_alloc.UseScratchGpr(inst->GetArg(0), { ABI_PARAM1 }).cvt32();
    Xbyak::Reg64 page_index = reg_alloc.ScratchGpr();
    Xbyak::Reg64 page_offset = reg_alloc.ScratchGpr();

    Xbyak::Label abort, end;

    code->mov(rax, reinterpret_cast<u64>(cb.page_table));
    code->mov(page_index.cvt32(), vaddr);
    code->shr(page_index.cvt32(), 12);
    code->mov(rax, qword[rax + page_index * 8]);
    code->test(rax, rax);
    code->jz(abort);
    code->mov(page_offset.cvt32(), vaddr);
    code->and_(page_offset.cvt32(), 4095);
    switch (bit_size) {
    case 8:
        code->movzx(result, code->byte[rax + page_offset]);
        break;
    case 16:
        code->movzx(result, word[rax + page_offset]);
        break;
    case 32:
        code->mov(result.cvt32(), dword[rax + page_offset]);
        break;
    case 64:
        code->mov(result.cvt64(), qword[rax + page_offset]);
        break;
    default:
        ASSERT_MSG(false, "Invalid bit_size");
        break;
    }
    code->jmp(end);
    code->L(abort);
    code->call(code->GetMemoryReadCallback(bit_size));
    code->L(end);
}

template<typename FunctionPointer>
static void WriteMemory(BlockOfCode* code, RegAlloc& reg_alloc, IR::Inst* inst, UserCallbacks& cb, size_t bit_size, FunctionPointer fn) {
    if (!cb.page_table) {
        reg_alloc.HostCall(inst, inst->GetArg(0), inst->GetArg(1));
        code->CallFunction(fn);
        return;
    }

    using namespace Xbyak::util;

    reg_alloc.ScratchGpr({ HostLoc::RAX });
    Xbyak::Reg32 vaddr = reg_alloc.UseScratchGpr(inst->GetArg(0), { ABI_PARAM1 }).cvt32();
    Xbyak::Reg64 value = reg_alloc.UseScratchGpr(inst->GetArg(1), { ABI_PARAM2 });
    Xbyak::Reg64 page_index = reg_alloc.ScratchGpr();
    Xbyak::Reg64 page_offset = reg_alloc.ScratchGpr();

    Xbyak::Label abort, end;

    code->mov(rax, reinterpret_cast<u64>(cb.page_table));
    code->mov(page_index.cvt32(), vaddr);
    code->shr(page_index.cvt32(), 12);
    code->mov(rax, qword[rax + page_index * 8]);
    code->test(rax, rax);
    code->jz(abort);
    code->mov(page_offset.cvt32(), vaddr);
    code->and_(page_offset.cvt32(), 4095);
    switch (bit_size) {
    case 8:
        code->mov(code->byte[rax + page_offset], value.cvt8());
        break;
    case 16:
        code->mov(word[rax + page_offset], value.cvt16());
        break;
    case 32:
        code->mov(dword[rax + page_offset], value.cvt32());
        break;
    case 64:
        code->mov(qword[rax + page_offset], value.cvt64());
        break;
    default:
        ASSERT_MSG(false, "Invalid bit_size");
        break;
    }
    code->jmp(end);
    code->L(abort);
    code->call(code->GetMemoryWriteCallback(bit_size));
    code->L(end);
}

void EmitX64::EmitReadMemory8(IR::Block&, IR::Inst* inst) {
    ReadMemory(code, reg_alloc, inst, cb, 8, cb.MemoryRead8);
}

void EmitX64::EmitReadMemory16(IR::Block&, IR::Inst* inst) {
    ReadMemory(code, reg_alloc, inst, cb, 16, cb.MemoryRead16);
}

void EmitX64::EmitReadMemory32(IR::Block&, IR::Inst* inst) {
    ReadMemory(code, reg_alloc, inst, cb, 32, cb.MemoryRead32);
}

void EmitX64::EmitReadMemory64(IR::Block&, IR::Inst* inst) {
    ReadMemory(code, reg_alloc, inst, cb, 64, cb.MemoryRead64);
}

void EmitX64::EmitWriteMemory8(IR::Block&, IR::Inst* inst) {
    WriteMemory(code, reg_alloc, inst, cb, 8, cb.MemoryWrite8);
}

void EmitX64::EmitWriteMemory16(IR::Block&, IR::Inst* inst) {
    WriteMemory(code, reg_alloc, inst, cb, 16, cb.MemoryWrite16);
}

void EmitX64::EmitWriteMemory32(IR::Block&, IR::Inst* inst) {
    WriteMemory(code, reg_alloc, inst, cb, 32, cb.MemoryWrite32);
}

void EmitX64::EmitWriteMemory64(IR::Block&, IR::Inst* inst) {
    WriteMemory(code, reg_alloc, inst, cb, 64, cb.MemoryWrite64);
}

template <typename FunctionPointer>
static void ExclusiveWrite(BlockOfCode* code, RegAlloc& reg_alloc, IR::Inst* inst, FunctionPointer fn) {
    using namespace Xbyak::util;
    Xbyak::Label end;

    reg_alloc.HostCall(nullptr, inst->GetArg(0), inst->GetArg(1));
    Xbyak::Reg32 passed = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Reg32 tmp = code->ABI_RETURN.cvt32(); // Use one of the unusued HostCall registers.

    code->mov(passed, u32(1));
    code->cmp(code->byte[r15 + offsetof(JitState, exclusive_state)], u8(0));
    code->je(end);
    code->mov(tmp, code->ABI_PARAM1);
    code->xor_(tmp, dword[r15 + offsetof(JitState, exclusive_address)]);
    code->test(tmp, JitState::RESERVATION_GRANULE_MASK);
    code->jne(end);
    code->mov(code->byte[r15 + offsetof(JitState, exclusive_state)], u8(0));
    code->CallFunction(fn);
    code->xor_(passed, passed);
    code->L(end);
}

void EmitX64::EmitExclusiveWriteMemory8(IR::Block&, IR::Inst* inst) {
    ExclusiveWrite(code, reg_alloc, inst, cb.MemoryWrite8);
}

void EmitX64::EmitExclusiveWriteMemory16(IR::Block&, IR::Inst* inst) {
    ExclusiveWrite(code, reg_alloc, inst, cb.MemoryWrite16);
}

void EmitX64::EmitExclusiveWriteMemory32(IR::Block&, IR::Inst* inst) {
    ExclusiveWrite(code, reg_alloc, inst, cb.MemoryWrite32);
}

void EmitX64::EmitExclusiveWriteMemory64(IR::Block&, IR::Inst* inst) {
    using namespace Xbyak::util;
    Xbyak::Label end;

    reg_alloc.HostCall(nullptr, inst->GetArg(0), inst->GetArg(1));
    Xbyak::Reg32 passed = reg_alloc.DefGpr(inst).cvt32();
    Xbyak::Reg64 value_hi = reg_alloc.UseScratchGpr(inst->GetArg(2));
    Xbyak::Reg64 value = code->ABI_PARAM2;
    Xbyak::Reg32 tmp = code->ABI_RETURN.cvt32(); // Use one of the unusued HostCall registers.

    code->mov(passed, u32(1));
    code->cmp(code->byte[r15 + offsetof(JitState, exclusive_state)], u8(0));
    code->je(end);
    code->mov(tmp, code->ABI_PARAM1);
    code->xor_(tmp, dword[r15 + offsetof(JitState, exclusive_address)]);
    code->test(tmp, JitState::RESERVATION_GRANULE_MASK);
    code->jne(end);
    code->mov(code->byte[r15 + offsetof(JitState, exclusive_state)], u8(0));
    code->mov(value.cvt32(), value.cvt32()); // zero extend to 64-bits
    code->shl(value_hi, 32);
    code->or_(value, value_hi);
    code->CallFunction(cb.MemoryWrite64);
    code->xor_(passed, passed);
    code->L(end);
}

void EmitX64::EmitAddCycles(size_t cycles) {
    using namespace Xbyak::util;
    ASSERT(cycles < std::numeric_limits<u32>::max());
    code->sub(qword[r15 + offsetof(JitState, cycles_remaining)], static_cast<u32>(cycles));
}

static Xbyak::Label EmitCond(BlockOfCode* code, Arm::Cond cond) {
    using namespace Xbyak::util;

    Xbyak::Label label;

    const Xbyak::Reg32 cpsr = eax;
    code->mov(cpsr, MJitStateCpsr());

    constexpr size_t n_shift = 31;
    constexpr size_t z_shift = 30;
    constexpr size_t c_shift = 29;
    constexpr size_t v_shift = 28;
    constexpr u32 n_mask = 1u << n_shift;
    constexpr u32 z_mask = 1u << z_shift;
    constexpr u32 c_mask = 1u << c_shift;
    constexpr u32 v_mask = 1u << v_shift;

    switch (cond) {
    case Arm::Cond::EQ: //z
        code->test(cpsr, z_mask);
        code->jnz(label);
        break;
    case Arm::Cond::NE: //!z
        code->test(cpsr, z_mask);
        code->jz(label);
        break;
    case Arm::Cond::CS: //c
        code->test(cpsr, c_mask);
        code->jnz(label);
        break;
    case Arm::Cond::CC: //!c
        code->test(cpsr, c_mask);
        code->jz(label);
        break;
    case Arm::Cond::MI: //n
        code->test(cpsr, n_mask);
        code->jnz(label);
        break;
    case Arm::Cond::PL: //!n
        code->test(cpsr, n_mask);
        code->jz(label);
        break;
    case Arm::Cond::VS: //v
        code->test(cpsr, v_mask);
        code->jnz(label);
        break;
    case Arm::Cond::VC: //!v
        code->test(cpsr, v_mask);
        code->jz(label);
        break;
    case Arm::Cond::HI: { //c & !z
        code->and_(cpsr, z_mask | c_mask);
        code->cmp(cpsr, c_mask);
        code->je(label);
        break;
    }
    case Arm::Cond::LS: { //!c | z
        code->and_(cpsr, z_mask | c_mask);
        code->cmp(cpsr, c_mask);
        code->jne(label);
        break;
    }
    case Arm::Cond::GE: { // n == v
        code->and_(cpsr, n_mask | v_mask);
        code->jz(label);
        code->cmp(cpsr, n_mask | v_mask);
        code->je(label);
        break;
    }
    case Arm::Cond::LT: { // n != v
        Xbyak::Label fail;
        code->and_(cpsr, n_mask | v_mask);
        code->jz(fail);
        code->cmp(cpsr, n_mask | v_mask);
        code->jne(label);
        code->L(fail);
        break;
    }
    case Arm::Cond::GT: { // !z & (n == v)
        const Xbyak::Reg32 tmp1 = ebx;
        const Xbyak::Reg32 tmp2 = esi;
        code->mov(tmp1, cpsr);
        code->mov(tmp2, cpsr);
        code->shr(tmp1, n_shift);
        code->shr(tmp2, v_shift);
        code->shr(cpsr, z_shift);
        code->xor_(tmp1, tmp2);
        code->or_(tmp1, cpsr);
        code->test(tmp1, 1);
        code->jz(label);
        break;
    }
    case Arm::Cond::LE: { // z | (n != v)
        const Xbyak::Reg32 tmp1 = ebx;
        const Xbyak::Reg32 tmp2 = esi;
        code->mov(tmp1, cpsr);
        code->mov(tmp2, cpsr);
        code->shr(tmp1, n_shift);
        code->shr(tmp2, v_shift);
        code->shr(cpsr, z_shift);
        code->xor_(tmp1, tmp2);
        code->or_(tmp1, cpsr);
        code->test(tmp1, 1);
        code->jnz(label);
        break;
    }
    default:
        ASSERT_MSG(false, "Unknown cond %zu", static_cast<size_t>(cond));
        break;
    }

    return label;
}

void EmitX64::EmitCondPrelude(const IR::Block& block) {
    if (block.GetCondition() == Arm::Cond::AL) {
        ASSERT(!block.HasConditionFailedLocation());
        return;
    }

    ASSERT(block.HasConditionFailedLocation());

    Xbyak::Label pass = EmitCond(code, block.GetCondition());
    EmitAddCycles(block.ConditionFailedCycleCount());
    EmitTerminalLinkBlock(IR::Term::LinkBlock{block.ConditionFailedLocation()}, block.Location());
    code->L(pass);
}

void EmitX64::EmitTerminal(IR::Terminal terminal, IR::LocationDescriptor initial_location) {
    switch (terminal.which()) {
    case 1:
        EmitTerminalInterpret(boost::get<IR::Term::Interpret>(terminal), initial_location);
        return;
    case 2:
        EmitTerminalReturnToDispatch(boost::get<IR::Term::ReturnToDispatch>(terminal), initial_location);
        return;
    case 3:
        EmitTerminalLinkBlock(boost::get<IR::Term::LinkBlock>(terminal), initial_location);
        return;
    case 4:
        EmitTerminalLinkBlockFast(boost::get<IR::Term::LinkBlockFast>(terminal), initial_location);
        return;
    case 5:
        EmitTerminalPopRSBHint(boost::get<IR::Term::PopRSBHint>(terminal), initial_location);
        return;
    case 6:
        EmitTerminalIf(boost::get<IR::Term::If>(terminal), initial_location);
        return;
    case 7:
        EmitTerminalCheckHalt(boost::get<IR::Term::CheckHalt>(terminal), initial_location);
        return;
    default:
        ASSERT_MSG(false, "Invalid Terminal. Bad programmer.");
        return;
    }
}

void EmitX64::EmitTerminalInterpret(IR::Term::Interpret terminal, IR::LocationDescriptor initial_location) {
    ASSERT_MSG(terminal.next.TFlag() == initial_location.TFlag(), "Unimplemented");
    ASSERT_MSG(terminal.next.EFlag() == initial_location.EFlag(), "Unimplemented");

    code->mov(code->ABI_PARAM1.cvt32(), terminal.next.PC());
    code->mov(code->ABI_PARAM2, reinterpret_cast<u64>(jit_interface));
    code->mov(code->ABI_PARAM3, reinterpret_cast<u64>(cb.user_arg));
    code->mov(MJitStateReg(Arm::Reg::PC), code->ABI_PARAM1.cvt32());
    code->SwitchMxcsrOnExit();
    code->CallFunction(cb.InterpreterFallback);
    code->ReturnFromRunCode(false); // TODO: Check cycles
}

void EmitX64::EmitTerminalReturnToDispatch(IR::Term::ReturnToDispatch, IR::LocationDescriptor) {
    code->ReturnFromRunCode();
}

void EmitX64::EmitTerminalLinkBlock(IR::Term::LinkBlock terminal, IR::LocationDescriptor initial_location) {
    using namespace Xbyak::util;

    if (terminal.next.TFlag() != initial_location.TFlag()) {
        if (terminal.next.TFlag()) {
            code->or_(MJitStateCpsr(), u32(1 << 5));
        } else {
            code->and_(MJitStateCpsr(), u32(~(1 << 5)));
        }
    }
    if (terminal.next.EFlag() != initial_location.EFlag()) {
        if (terminal.next.EFlag()) {
            code->or_(MJitStateCpsr(), u32(1 << 9));
        } else {
            code->and_(MJitStateCpsr(), u32(~(1 << 9)));
        }
    }

    code->cmp(qword[r15 + offsetof(JitState, cycles_remaining)], 0);

    patch_information[terminal.next.UniqueHash()].jg.emplace_back(code->getCurr());
    if (auto next_bb = GetBasicBlock(terminal.next)) {
        EmitPatchJg(next_bb->code_ptr);
    } else {
        EmitPatchJg();
    }

    code->mov(MJitStateReg(Arm::Reg::PC), terminal.next.PC());
    code->ReturnFromRunCode(); // TODO: Check cycles, Properly do a link
}

void EmitX64::EmitTerminalLinkBlockFast(IR::Term::LinkBlockFast terminal, IR::LocationDescriptor initial_location) {
    using namespace Xbyak::util;

    if (terminal.next.TFlag() != initial_location.TFlag()) {
        if (terminal.next.TFlag()) {
            code->or_(MJitStateCpsr(), u32(1 << 5));
        } else {
            code->and_(MJitStateCpsr(), u32(~(1 << 5)));
        }
    }
    if (terminal.next.EFlag() != initial_location.EFlag()) {
        if (terminal.next.EFlag()) {
            code->or_(MJitStateCpsr(), u32(1 << 9));
        } else {
            code->and_(MJitStateCpsr(), u32(~(1 << 9)));
        }
    }

    patch_information[terminal.next.UniqueHash()].jmp.emplace_back(code->getCurr());
    if (auto next_bb = GetBasicBlock(terminal.next)) {
        EmitPatchJmp(terminal.next, next_bb->code_ptr);
    } else {
        EmitPatchJmp(terminal.next);
    }
}

void EmitX64::EmitTerminalPopRSBHint(IR::Term::PopRSBHint, IR::LocationDescriptor) {
    using namespace Xbyak::util;

    // This calculation has to match up with IREmitter::PushRSB
    code->mov(ebx, MJitStateCpsr());
    code->mov(ecx, MJitStateReg(Arm::Reg::PC));
    code->and_(ebx, u32((1 << 5) | (1 << 9)));
    code->shr(ebx, 2);
    code->or_(ebx, dword[r15 + offsetof(JitState, FPSCR_mode)]);
    code->shl(rbx, 32);
    code->or_(rbx, rcx);

    code->mov(rax, reinterpret_cast<u64>(code->GetReturnFromRunCodeAddress()));
    for (size_t i = 0; i < JitState::RSBSize; ++i) {
        code->cmp(rbx, qword[r15 + offsetof(JitState, rsb_location_descriptors) + i * sizeof(u64)]);
        code->cmove(rax, qword[r15 + offsetof(JitState, rsb_codeptrs) + i * sizeof(u64)]);
    }

    code->jmp(rax);
}

void EmitX64::EmitTerminalIf(IR::Term::If terminal, IR::LocationDescriptor initial_location) {
    Xbyak::Label pass = EmitCond(code, terminal.if_);
    EmitTerminal(terminal.else_, initial_location);
    code->L(pass);
    EmitTerminal(terminal.then_, initial_location);
}

void EmitX64::EmitTerminalCheckHalt(IR::Term::CheckHalt terminal, IR::LocationDescriptor initial_location) {
    using namespace Xbyak::util;

    code->cmp(code->byte[r15 + offsetof(JitState, halt_requested)], u8(0));
    code->jne(code->GetReturnFromRunCodeAddress());
    EmitTerminal(terminal.else_, initial_location);
}

void EmitX64::Patch(const IR::LocationDescriptor& desc, CodePtr bb) {
    const CodePtr save_code_ptr = code->getCurr();
    const PatchInformation& patch_info = patch_information[desc.UniqueHash()];

    for (CodePtr location : patch_info.jg) {
        code->SetCodePtr(location);
        EmitPatchJg(bb);
    }

    for (CodePtr location : patch_info.jmp) {
        code->SetCodePtr(location);
        EmitPatchJmp(desc, bb);
    }

    for (CodePtr location : patch_info.mov_rcx) {
        code->SetCodePtr(location);
        EmitPatchMovRcx(bb);
    }

    code->SetCodePtr(save_code_ptr);
}

void EmitX64::Unpatch(const IR::LocationDescriptor& desc) {
    Patch(desc, nullptr);
}

void EmitX64::EmitPatchJg(CodePtr target_code_ptr) {
    const CodePtr patch_location = code->getCurr();
    if (target_code_ptr) {
        code->jg(target_code_ptr);
    }
    code->EnsurePatchLocationSize(patch_location, 6);
}

void EmitX64::EmitPatchJmp(const IR::LocationDescriptor& target_desc, CodePtr target_code_ptr) {
    const CodePtr patch_location = code->getCurr();
    if (target_code_ptr) {
        code->jmp(target_code_ptr);
    } else {
        code->mov(MJitStateReg(Arm::Reg::PC), target_desc.PC());
        code->jmp(code->GetReturnFromRunCodeAddress());
    }
    code->EnsurePatchLocationSize(patch_location, 13);
}

void EmitX64::EmitPatchMovRcx(CodePtr target_code_ptr) {
    if (!target_code_ptr) {
        target_code_ptr = code->GetReturnFromRunCodeAddress();
    }
    const CodePtr patch_location = code->getCurr();
    code->mov(code->rcx, reinterpret_cast<u64>(target_code_ptr));
    code->EnsurePatchLocationSize(patch_location, 10);
}

void EmitX64::ClearCache() {
    block_descriptors.clear();
    patch_information.clear();
}

} // namespace BackendX64
} // namespace Dynarmic