/*
 *  ARM translation
 *
 *  Copyright (c) 2003 Fabrice Bellard
 *  Copyright (c) 2005-2007 CodeSourcery
 *  Copyright (c) 2007 OpenedHand, Ltd.
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2 of the License, or (at your option) any later version.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, see <http://www.gnu.org/licenses/>.
 */
#include "qemu/osdep.h"

#include "cpu.h"
#include "internals.h"
#include "disas/disas.h"
#include "exec/exec-all.h"
#include "tcg-op.h"
#include "qemu/log.h"
#include "qemu/bitops.h"
#include "arm_ldst.h"
#include "exec/semihost.h"

#include "exec/helper-proto.h"
#include "exec/helper-gen.h"

#include "trace-tcg.h"
#include "exec/log.h"


#define ENABLE_ARCH_4T    arm_dc_feature(s, ARM_FEATURE_V4T)
#define ENABLE_ARCH_5     arm_dc_feature(s, ARM_FEATURE_V5)
/* currently all emulated v5 cores are also v5TE, so don't bother */
#define ENABLE_ARCH_5TE   arm_dc_feature(s, ARM_FEATURE_V5)
#define ENABLE_ARCH_5J    0
#define ENABLE_ARCH_6     arm_dc_feature(s, ARM_FEATURE_V6)
#define ENABLE_ARCH_6K    arm_dc_feature(s, ARM_FEATURE_V6K)
#define ENABLE_ARCH_6T2   arm_dc_feature(s, ARM_FEATURE_THUMB2)
#define ENABLE_ARCH_7     arm_dc_feature(s, ARM_FEATURE_V7)
#define ENABLE_ARCH_8     arm_dc_feature(s, ARM_FEATURE_V8)

#define ARCH(x) do { if (!ENABLE_ARCH_##x) goto illegal_op; } while(0)

#include "translate.h"

#if defined(CONFIG_USER_ONLY)
#define IS_USER(s) 1
#else
#define IS_USER(s) (s->user)
#endif

TCGv_env cpu_env;
/* We reuse the same 64-bit temporaries for efficiency.  */
static TCGv_i64 cpu_V0, cpu_V1, cpu_M0;
static TCGv_i32 cpu_R[16];
TCGv_i32 cpu_CF, cpu_NF, cpu_VF, cpu_ZF;
TCGv_i64 cpu_exclusive_addr;
TCGv_i64 cpu_exclusive_val;

/* FIXME:  These should be removed.  */
static TCGv_i32 cpu_F0s, cpu_F1s;
static TCGv_i64 cpu_F0d, cpu_F1d;

#include "exec/gen-icount.h"

static const char *regnames[] =
    { "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7",
      "r8", "r9", "r10", "r11", "r12", "r13", "r14", "pc" };

/* initialize TCG globals.  */
void arm_translate_init(void)
{
    int i;

    cpu_env = tcg_global_reg_new_ptr(TCG_AREG0, "env");
    tcg_ctx.tcg_env = cpu_env;

    for (i = 0; i < 16; i++) {
        cpu_R[i] = tcg_global_mem_new_i32(cpu_env,
                                          offsetof(CPUARMState, regs[i]),
                                          regnames[i]);
    }
    cpu_CF = tcg_global_mem_new_i32(cpu_env, offsetof(CPUARMState, CF), "CF");
    cpu_NF = tcg_global_mem_new_i32(cpu_env, offsetof(CPUARMState, NF), "NF");
    cpu_VF = tcg_global_mem_new_i32(cpu_env, offsetof(CPUARMState, VF), "VF");
    cpu_ZF = tcg_global_mem_new_i32(cpu_env, offsetof(CPUARMState, ZF), "ZF");

    cpu_exclusive_addr = tcg_global_mem_new_i64(cpu_env,
        offsetof(CPUARMState, exclusive_addr), "exclusive_addr");
    cpu_exclusive_val = tcg_global_mem_new_i64(cpu_env,
        offsetof(CPUARMState, exclusive_val), "exclusive_val");

    a64_translate_init();
}

static inline ARMMMUIdx get_a32_user_mem_index(DisasContext *s)
{
    /* Return the mmu_idx to use for A32/T32 "unprivileged load/store"
     * insns:
     *  if PL2, UNPREDICTABLE (we choose to implement as if PL0)
     *  otherwise, access as if at PL0.
     */
    switch (s->mmu_idx) {
    case ARMMMUIdx_S1E2:        /* this one is UNPREDICTABLE */
    case ARMMMUIdx_S12NSE0:
    case ARMMMUIdx_S12NSE1:
        return ARMMMUIdx_S12NSE0;
    case ARMMMUIdx_S1E3:
    case ARMMMUIdx_S1SE0:
    case ARMMMUIdx_S1SE1:
        return ARMMMUIdx_S1SE0;
    case ARMMMUIdx_S2NS:
    default:
        g_assert_not_reached();
    }
}

static inline TCGv_i32 load_cpu_offset(int offset)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_ld_i32(tmp, cpu_env, offset);
    return tmp;
}

#define load_cpu_field(name) load_cpu_offset(offsetof(CPUARMState, name))

static inline void store_cpu_offset(TCGv_i32 var, int offset)
{
    tcg_gen_st_i32(var, cpu_env, offset);
    tcg_temp_free_i32(var);
}

#define store_cpu_field(var, name) \
    store_cpu_offset(var, offsetof(CPUARMState, name))

/* Set a variable to the value of a CPU register.  */
static void load_reg_var(DisasContext *s, TCGv_i32 var, int reg)
{
    if (reg == 15) {
        uint32_t addr;
        /* normally, since we updated PC, we need only to add one insn */
        if (s->thumb)
            addr = (long)s->pc + 2;
        else
            addr = (long)s->pc + 4;
        tcg_gen_movi_i32(var, addr);
    } else {
        tcg_gen_mov_i32(var, cpu_R[reg]);
    }
}

/* Create a new temporary and set it to the value of a CPU register.  */
static inline TCGv_i32 load_reg(DisasContext *s, int reg)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    load_reg_var(s, tmp, reg);
    return tmp;
}

/* Set a CPU register.  The source must be a temporary and will be
   marked as dead.  */
static void store_reg(DisasContext *s, int reg, TCGv_i32 var)
{
    if (reg == 15) {
        /* In Thumb mode, we must ignore bit 0.
         * In ARM mode, for ARMv4 and ARMv5, it is UNPREDICTABLE if bits [1:0]
         * are not 0b00, but for ARMv6 and above, we must ignore bits [1:0].
         * We choose to ignore [1:0] in ARM mode for all architecture versions.
         */
        tcg_gen_andi_i32(var, var, s->thumb ? ~1 : ~3);
        s->is_jmp = DISAS_JUMP;
    }
    tcg_gen_mov_i32(cpu_R[reg], var);
    tcg_temp_free_i32(var);
}

/* Value extensions.  */
#define gen_uxtb(var) tcg_gen_ext8u_i32(var, var)
#define gen_uxth(var) tcg_gen_ext16u_i32(var, var)
#define gen_sxtb(var) tcg_gen_ext8s_i32(var, var)
#define gen_sxth(var) tcg_gen_ext16s_i32(var, var)

#define gen_sxtb16(var) gen_helper_sxtb16(var, var)
#define gen_uxtb16(var) gen_helper_uxtb16(var, var)


static inline void gen_set_cpsr(TCGv_i32 var, uint32_t mask)
{
    TCGv_i32 tmp_mask = tcg_const_i32(mask);
    gen_helper_cpsr_write(cpu_env, var, tmp_mask);
    tcg_temp_free_i32(tmp_mask);
}
/* Set NZCV flags from the high 4 bits of var.  */
#define gen_set_nzcv(var) gen_set_cpsr(var, CPSR_NZCV)

static void gen_exception_internal(int excp)
{
    TCGv_i32 tcg_excp = tcg_const_i32(excp);

    assert(excp_is_internal(excp));
    gen_helper_exception_internal(cpu_env, tcg_excp);
    tcg_temp_free_i32(tcg_excp);
}

static void gen_exception(int excp, uint32_t syndrome, uint32_t target_el)
{
    TCGv_i32 tcg_excp = tcg_const_i32(excp);
    TCGv_i32 tcg_syn = tcg_const_i32(syndrome);
    TCGv_i32 tcg_el = tcg_const_i32(target_el);

    gen_helper_exception_with_syndrome(cpu_env, tcg_excp,
                                       tcg_syn, tcg_el);

    tcg_temp_free_i32(tcg_el);
    tcg_temp_free_i32(tcg_syn);
    tcg_temp_free_i32(tcg_excp);
}

static void gen_ss_advance(DisasContext *s)
{
    /* If the singlestep state is Active-not-pending, advance to
     * Active-pending.
     */
    if (s->ss_active) {
        s->pstate_ss = 0;
        gen_helper_clear_pstate_ss(cpu_env);
    }
}

static void gen_step_complete_exception(DisasContext *s)
{
    /* We just completed step of an insn. Move from Active-not-pending
     * to Active-pending, and then also take the swstep exception.
     * This corresponds to making the (IMPDEF) choice to prioritize
     * swstep exceptions over asynchronous exceptions taken to an exception
     * level where debug is disabled. This choice has the advantage that
     * we do not need to maintain internal state corresponding to the
     * ISV/EX syndrome bits between completion of the step and generation
     * of the exception, and our syndrome information is always correct.
     */
    gen_ss_advance(s);
    gen_exception(EXCP_UDEF, syn_swstep(s->ss_same_el, 1, s->is_ldex),
                  default_exception_el(s));
    s->is_jmp = DISAS_EXC;
}

static void gen_smul_dual(TCGv_i32 a, TCGv_i32 b)
{
    TCGv_i32 tmp1 = tcg_temp_new_i32();
    TCGv_i32 tmp2 = tcg_temp_new_i32();
    tcg_gen_ext16s_i32(tmp1, a);
    tcg_gen_ext16s_i32(tmp2, b);
    tcg_gen_mul_i32(tmp1, tmp1, tmp2);
    tcg_temp_free_i32(tmp2);
    tcg_gen_sari_i32(a, a, 16);
    tcg_gen_sari_i32(b, b, 16);
    tcg_gen_mul_i32(b, b, a);
    tcg_gen_mov_i32(a, tmp1);
    tcg_temp_free_i32(tmp1);
}

/* Byteswap each halfword.  */
static void gen_rev16(TCGv_i32 var)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_shri_i32(tmp, var, 8);
    tcg_gen_andi_i32(tmp, tmp, 0x00ff00ff);
    tcg_gen_shli_i32(var, var, 8);
    tcg_gen_andi_i32(var, var, 0xff00ff00);
    tcg_gen_or_i32(var, var, tmp);
    tcg_temp_free_i32(tmp);
}

/* Byteswap low halfword and sign extend.  */
static void gen_revsh(TCGv_i32 var)
{
    tcg_gen_ext16u_i32(var, var);
    tcg_gen_bswap16_i32(var, var);
    tcg_gen_ext16s_i32(var, var);
}

/* Unsigned bitfield extract.  */
static void gen_ubfx(TCGv_i32 var, int shift, uint32_t mask)
{
    if (shift)
        tcg_gen_shri_i32(var, var, shift);
    tcg_gen_andi_i32(var, var, mask);
}

/* Signed bitfield extract.  */
static void gen_sbfx(TCGv_i32 var, int shift, int width)
{
    uint32_t signbit;

    if (shift)
        tcg_gen_sari_i32(var, var, shift);
    if (shift + width < 32) {
        signbit = 1u << (width - 1);
        tcg_gen_andi_i32(var, var, (1u << width) - 1);
        tcg_gen_xori_i32(var, var, signbit);
        tcg_gen_subi_i32(var, var, signbit);
    }
}

/* Return (b << 32) + a. Mark inputs as dead */
static TCGv_i64 gen_addq_msw(TCGv_i64 a, TCGv_i32 b)
{
    TCGv_i64 tmp64 = tcg_temp_new_i64();

    tcg_gen_extu_i32_i64(tmp64, b);
    tcg_temp_free_i32(b);
    tcg_gen_shli_i64(tmp64, tmp64, 32);
    tcg_gen_add_i64(a, tmp64, a);

    tcg_temp_free_i64(tmp64);
    return a;
}

/* Return (b << 32) - a. Mark inputs as dead. */
static TCGv_i64 gen_subq_msw(TCGv_i64 a, TCGv_i32 b)
{
    TCGv_i64 tmp64 = tcg_temp_new_i64();

    tcg_gen_extu_i32_i64(tmp64, b);
    tcg_temp_free_i32(b);
    tcg_gen_shli_i64(tmp64, tmp64, 32);
    tcg_gen_sub_i64(a, tmp64, a);

    tcg_temp_free_i64(tmp64);
    return a;
}

/* 32x32->64 multiply.  Marks inputs as dead.  */
static TCGv_i64 gen_mulu_i64_i32(TCGv_i32 a, TCGv_i32 b)
{
    TCGv_i32 lo = tcg_temp_new_i32();
    TCGv_i32 hi = tcg_temp_new_i32();
    TCGv_i64 ret;

    tcg_gen_mulu2_i32(lo, hi, a, b);
    tcg_temp_free_i32(a);
    tcg_temp_free_i32(b);

    ret = tcg_temp_new_i64();
    tcg_gen_concat_i32_i64(ret, lo, hi);
    tcg_temp_free_i32(lo);
    tcg_temp_free_i32(hi);

    return ret;
}

static TCGv_i64 gen_muls_i64_i32(TCGv_i32 a, TCGv_i32 b)
{
    TCGv_i32 lo = tcg_temp_new_i32();
    TCGv_i32 hi = tcg_temp_new_i32();
    TCGv_i64 ret;

    tcg_gen_muls2_i32(lo, hi, a, b);
    tcg_temp_free_i32(a);
    tcg_temp_free_i32(b);

    ret = tcg_temp_new_i64();
    tcg_gen_concat_i32_i64(ret, lo, hi);
    tcg_temp_free_i32(lo);
    tcg_temp_free_i32(hi);

    return ret;
}

/* Swap low and high halfwords.  */
static void gen_swap_half(TCGv_i32 var)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_shri_i32(tmp, var, 16);
    tcg_gen_shli_i32(var, var, 16);
    tcg_gen_or_i32(var, var, tmp);
    tcg_temp_free_i32(tmp);
}

/* Dual 16-bit add.  Result placed in t0 and t1 is marked as dead.
    tmp = (t0 ^ t1) & 0x8000;
    t0 &= ~0x8000;
    t1 &= ~0x8000;
    t0 = (t0 + t1) ^ tmp;
 */

static void gen_add16(TCGv_i32 t0, TCGv_i32 t1)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_xor_i32(tmp, t0, t1);
    tcg_gen_andi_i32(tmp, tmp, 0x8000);
    tcg_gen_andi_i32(t0, t0, ~0x8000);
    tcg_gen_andi_i32(t1, t1, ~0x8000);
    tcg_gen_add_i32(t0, t0, t1);
    tcg_gen_xor_i32(t0, t0, tmp);
    tcg_temp_free_i32(tmp);
    tcg_temp_free_i32(t1);
}

/* Set CF to the top bit of var.  */
static void gen_set_CF_bit31(TCGv_i32 var)
{
    tcg_gen_shri_i32(cpu_CF, var, 31);
}

/* Set N and Z flags from var.  */
static inline void gen_logic_CC(TCGv_i32 var)
{
    tcg_gen_mov_i32(cpu_NF, var);
    tcg_gen_mov_i32(cpu_ZF, var);
}

/* T0 += T1 + CF.  */
static void gen_adc(TCGv_i32 t0, TCGv_i32 t1)
{
    tcg_gen_add_i32(t0, t0, t1);
    tcg_gen_add_i32(t0, t0, cpu_CF);
}

/* dest = T0 + T1 + CF. */
static void gen_add_carry(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    tcg_gen_add_i32(dest, t0, t1);
    tcg_gen_add_i32(dest, dest, cpu_CF);
}

/* dest = T0 - T1 + CF - 1.  */
static void gen_sub_carry(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    tcg_gen_sub_i32(dest, t0, t1);
    tcg_gen_add_i32(dest, dest, cpu_CF);
    tcg_gen_subi_i32(dest, dest, 1);
}

/* dest = T0 + T1. Compute C, N, V and Z flags */
static void gen_add_CC(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_movi_i32(tmp, 0);
    tcg_gen_add2_i32(cpu_NF, cpu_CF, t0, tmp, t1, tmp);
    tcg_gen_mov_i32(cpu_ZF, cpu_NF);
    tcg_gen_xor_i32(cpu_VF, cpu_NF, t0);
    tcg_gen_xor_i32(tmp, t0, t1);
    tcg_gen_andc_i32(cpu_VF, cpu_VF, tmp);
    tcg_temp_free_i32(tmp);
    tcg_gen_mov_i32(dest, cpu_NF);
}

/* dest = T0 + T1 + CF.  Compute C, N, V and Z flags */
static void gen_adc_CC(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    if (TCG_TARGET_HAS_add2_i32) {
        tcg_gen_movi_i32(tmp, 0);
        tcg_gen_add2_i32(cpu_NF, cpu_CF, t0, tmp, cpu_CF, tmp);
        tcg_gen_add2_i32(cpu_NF, cpu_CF, cpu_NF, cpu_CF, t1, tmp);
    } else {
        TCGv_i64 q0 = tcg_temp_new_i64();
        TCGv_i64 q1 = tcg_temp_new_i64();
        tcg_gen_extu_i32_i64(q0, t0);
        tcg_gen_extu_i32_i64(q1, t1);
        tcg_gen_add_i64(q0, q0, q1);
        tcg_gen_extu_i32_i64(q1, cpu_CF);
        tcg_gen_add_i64(q0, q0, q1);
        tcg_gen_extr_i64_i32(cpu_NF, cpu_CF, q0);
        tcg_temp_free_i64(q0);
        tcg_temp_free_i64(q1);
    }
    tcg_gen_mov_i32(cpu_ZF, cpu_NF);
    tcg_gen_xor_i32(cpu_VF, cpu_NF, t0);
    tcg_gen_xor_i32(tmp, t0, t1);
    tcg_gen_andc_i32(cpu_VF, cpu_VF, tmp);
    tcg_temp_free_i32(tmp);
    tcg_gen_mov_i32(dest, cpu_NF);
}

/* dest = T0 - T1. Compute C, N, V and Z flags */
static void gen_sub_CC(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    TCGv_i32 tmp;
    tcg_gen_sub_i32(cpu_NF, t0, t1);
    tcg_gen_mov_i32(cpu_ZF, cpu_NF);
    tcg_gen_setcond_i32(TCG_COND_GEU, cpu_CF, t0, t1);
    tcg_gen_xor_i32(cpu_VF, cpu_NF, t0);
    tmp = tcg_temp_new_i32();
    tcg_gen_xor_i32(tmp, t0, t1);
    tcg_gen_and_i32(cpu_VF, cpu_VF, tmp);
    tcg_temp_free_i32(tmp);
    tcg_gen_mov_i32(dest, cpu_NF);
}

/* dest = T0 + ~T1 + CF.  Compute C, N, V and Z flags */
static void gen_sbc_CC(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_not_i32(tmp, t1);
    gen_adc_CC(dest, t0, tmp);
    tcg_temp_free_i32(tmp);
}

#define GEN_SHIFT(name)                                               \
static void gen_##name(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)       \
{                                                                     \
    TCGv_i32 tmp1, tmp2, tmp3;                                        \
    tmp1 = tcg_temp_new_i32();                                        \
    tcg_gen_andi_i32(tmp1, t1, 0xff);                                 \
    tmp2 = tcg_const_i32(0);                                          \
    tmp3 = tcg_const_i32(0x1f);                                       \
    tcg_gen_movcond_i32(TCG_COND_GTU, tmp2, tmp1, tmp3, tmp2, t0);    \
    tcg_temp_free_i32(tmp3);                                          \
    tcg_gen_andi_i32(tmp1, tmp1, 0x1f);                               \
    tcg_gen_##name##_i32(dest, tmp2, tmp1);                           \
    tcg_temp_free_i32(tmp2);                                          \
    tcg_temp_free_i32(tmp1);                                          \
}
GEN_SHIFT(shl)
GEN_SHIFT(shr)
#undef GEN_SHIFT

static void gen_sar(TCGv_i32 dest, TCGv_i32 t0, TCGv_i32 t1)
{
    TCGv_i32 tmp1, tmp2;
    tmp1 = tcg_temp_new_i32();
    tcg_gen_andi_i32(tmp1, t1, 0xff);
    tmp2 = tcg_const_i32(0x1f);
    tcg_gen_movcond_i32(TCG_COND_GTU, tmp1, tmp1, tmp2, tmp2, tmp1);
    tcg_temp_free_i32(tmp2);
    tcg_gen_sar_i32(dest, t0, tmp1);
    tcg_temp_free_i32(tmp1);
}

static void tcg_gen_abs_i32(TCGv_i32 dest, TCGv_i32 src)
{
    TCGv_i32 c0 = tcg_const_i32(0);
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_neg_i32(tmp, src);
    tcg_gen_movcond_i32(TCG_COND_GT, dest, src, c0, src, tmp);
    tcg_temp_free_i32(c0);
    tcg_temp_free_i32(tmp);
}

static void shifter_out_im(TCGv_i32 var, int shift)
{
    if (shift == 0) {
        tcg_gen_andi_i32(cpu_CF, var, 1);
    } else {
        tcg_gen_shri_i32(cpu_CF, var, shift);
        if (shift != 31) {
            tcg_gen_andi_i32(cpu_CF, cpu_CF, 1);
        }
    }
}

/* Shift by immediate.  Includes special handling for shift == 0.  */
static inline void gen_arm_shift_im(TCGv_i32 var, int shiftop,
                                    int shift, int flags)
{
    switch (shiftop) {
    case 0: /* LSL */
        if (shift != 0) {
            if (flags)
                shifter_out_im(var, 32 - shift);
            tcg_gen_shli_i32(var, var, shift);
        }
        break;
    case 1: /* LSR */
        if (shift == 0) {
            if (flags) {
                tcg_gen_shri_i32(cpu_CF, var, 31);
            }
            tcg_gen_movi_i32(var, 0);
        } else {
            if (flags)
                shifter_out_im(var, shift - 1);
            tcg_gen_shri_i32(var, var, shift);
        }
        break;
    case 2: /* ASR */
        if (shift == 0)
            shift = 32;
        if (flags)
            shifter_out_im(var, shift - 1);
        if (shift == 32)
          shift = 31;
        tcg_gen_sari_i32(var, var, shift);
        break;
    case 3: /* ROR/RRX */
        if (shift != 0) {
            if (flags)
                shifter_out_im(var, shift - 1);
            tcg_gen_rotri_i32(var, var, shift); break;
        } else {
            TCGv_i32 tmp = tcg_temp_new_i32();
            tcg_gen_shli_i32(tmp, cpu_CF, 31);
            if (flags)
                shifter_out_im(var, 0);
            tcg_gen_shri_i32(var, var, 1);
            tcg_gen_or_i32(var, var, tmp);
            tcg_temp_free_i32(tmp);
        }
    }
};

static inline void gen_arm_shift_reg(TCGv_i32 var, int shiftop,
                                     TCGv_i32 shift, int flags)
{
    if (flags) {
        switch (shiftop) {
        case 0: gen_helper_shl_cc(var, cpu_env, var, shift); break;
        case 1: gen_helper_shr_cc(var, cpu_env, var, shift); break;
        case 2: gen_helper_sar_cc(var, cpu_env, var, shift); break;
        case 3: gen_helper_ror_cc(var, cpu_env, var, shift); break;
        }
    } else {
        switch (shiftop) {
        case 0:
            gen_shl(var, var, shift);
            break;
        case 1:
            gen_shr(var, var, shift);
            break;
        case 2:
            gen_sar(var, var, shift);
            break;
        case 3: tcg_gen_andi_i32(shift, shift, 0x1f);
                tcg_gen_rotr_i32(var, var, shift); break;
        }
    }
    tcg_temp_free_i32(shift);
}

#define PAS_OP(pfx) \
    switch (op2) {  \
    case 0: gen_pas_helper(glue(pfx,add16)); break; \
    case 1: gen_pas_helper(glue(pfx,addsubx)); break; \
    case 2: gen_pas_helper(glue(pfx,subaddx)); break; \
    case 3: gen_pas_helper(glue(pfx,sub16)); break; \
    case 4: gen_pas_helper(glue(pfx,add8)); break; \
    case 7: gen_pas_helper(glue(pfx,sub8)); break; \
    }
static void gen_arm_parallel_addsub(int op1, int op2, TCGv_i32 a, TCGv_i32 b)
{
    TCGv_ptr tmp;

    switch (op1) {
#define gen_pas_helper(name) glue(gen_helper_,name)(a, a, b, tmp)
    case 1:
        tmp = tcg_temp_new_ptr();
        tcg_gen_addi_ptr(tmp, cpu_env, offsetof(CPUARMState, GE));
        PAS_OP(s)
        tcg_temp_free_ptr(tmp);
        break;
    case 5:
        tmp = tcg_temp_new_ptr();
        tcg_gen_addi_ptr(tmp, cpu_env, offsetof(CPUARMState, GE));
        PAS_OP(u)
        tcg_temp_free_ptr(tmp);
        break;
#undef gen_pas_helper
#define gen_pas_helper(name) glue(gen_helper_,name)(a, a, b)
    case 2:
        PAS_OP(q);
        break;
    case 3:
        PAS_OP(sh);
        break;
    case 6:
        PAS_OP(uq);
        break;
    case 7:
        PAS_OP(uh);
        break;
#undef gen_pas_helper
    }
}
#undef PAS_OP

/* For unknown reasons Arm and Thumb-2 use arbitrarily different encodings.  */
#define PAS_OP(pfx) \
    switch (op1) {  \
    case 0: gen_pas_helper(glue(pfx,add8)); break; \
    case 1: gen_pas_helper(glue(pfx,add16)); break; \
    case 2: gen_pas_helper(glue(pfx,addsubx)); break; \
    case 4: gen_pas_helper(glue(pfx,sub8)); break; \
    case 5: gen_pas_helper(glue(pfx,sub16)); break; \
    case 6: gen_pas_helper(glue(pfx,subaddx)); break; \
    }
static void gen_thumb2_parallel_addsub(int op1, int op2, TCGv_i32 a, TCGv_i32 b)
{
    TCGv_ptr tmp;

    switch (op2) {
#define gen_pas_helper(name) glue(gen_helper_,name)(a, a, b, tmp)
    case 0:
        tmp = tcg_temp_new_ptr();
        tcg_gen_addi_ptr(tmp, cpu_env, offsetof(CPUARMState, GE));
        PAS_OP(s)
        tcg_temp_free_ptr(tmp);
        break;
    case 4:
        tmp = tcg_temp_new_ptr();
        tcg_gen_addi_ptr(tmp, cpu_env, offsetof(CPUARMState, GE));
        PAS_OP(u)
        tcg_temp_free_ptr(tmp);
        break;
#undef gen_pas_helper
#define gen_pas_helper(name) glue(gen_helper_,name)(a, a, b)
    case 1:
        PAS_OP(q);
        break;
    case 2:
        PAS_OP(sh);
        break;
    case 5:
        PAS_OP(uq);
        break;
    case 6:
        PAS_OP(uh);
        break;
#undef gen_pas_helper
    }
}
#undef PAS_OP

/*
 * Generate a conditional based on ARM condition code cc.
 * This is common between ARM and Aarch64 targets.
 */
void arm_test_cc(DisasCompare *cmp, int cc)
{
    TCGv_i32 value;
    TCGCond cond;
    bool global = true;

    switch (cc) {
    case 0: /* eq: Z */
    case 1: /* ne: !Z */
        cond = TCG_COND_EQ;
        value = cpu_ZF;
        break;

    case 2: /* cs: C */
    case 3: /* cc: !C */
        cond = TCG_COND_NE;
        value = cpu_CF;
        break;

    case 4: /* mi: N */
    case 5: /* pl: !N */
        cond = TCG_COND_LT;
        value = cpu_NF;
        break;

    case 6: /* vs: V */
    case 7: /* vc: !V */
        cond = TCG_COND_LT;
        value = cpu_VF;
        break;

    case 8: /* hi: C && !Z */
    case 9: /* ls: !C || Z -> !(C && !Z) */
        cond = TCG_COND_NE;
        value = tcg_temp_new_i32();
        global = false;
        /* CF is 1 for C, so -CF is an all-bits-set mask for C;
           ZF is non-zero for !Z; so AND the two subexpressions.  */
        tcg_gen_neg_i32(value, cpu_CF);
        tcg_gen_and_i32(value, value, cpu_ZF);
        break;

    case 10: /* ge: N == V -> N ^ V == 0 */
    case 11: /* lt: N != V -> N ^ V != 0 */
        /* Since we're only interested in the sign bit, == 0 is >= 0.  */
        cond = TCG_COND_GE;
        value = tcg_temp_new_i32();
        global = false;
        tcg_gen_xor_i32(value, cpu_VF, cpu_NF);
        break;

    case 12: /* gt: !Z && N == V */
    case 13: /* le: Z || N != V */
        cond = TCG_COND_NE;
        value = tcg_temp_new_i32();
        global = false;
        /* (N == V) is equal to the sign bit of ~(NF ^ VF).  Propagate
         * the sign bit then AND with ZF to yield the result.  */
        tcg_gen_xor_i32(value, cpu_VF, cpu_NF);
        tcg_gen_sari_i32(value, value, 31);
        tcg_gen_andc_i32(value, cpu_ZF, value);
        break;

    case 14: /* always */
    case 15: /* always */
        /* Use the ALWAYS condition, which will fold early.
         * It doesn't matter what we use for the value.  */
        cond = TCG_COND_ALWAYS;
        value = cpu_ZF;
        goto no_invert;

    default:
        fprintf(stderr, "Bad condition code 0x%x\n", cc);
        abort();
    }

    if (cc & 1) {
        cond = tcg_invert_cond(cond);
    }

 no_invert:
    cmp->cond = cond;
    cmp->value = value;
    cmp->value_global = global;
}

void arm_free_cc(DisasCompare *cmp)
{
    if (!cmp->value_global) {
        tcg_temp_free_i32(cmp->value);
    }
}

void arm_jump_cc(DisasCompare *cmp, TCGLabel *label)
{
    tcg_gen_brcondi_i32(cmp->cond, cmp->value, 0, label);
}

void arm_gen_test_cc(int cc, TCGLabel *label)
{
    DisasCompare cmp;
    arm_test_cc(&cmp, cc);
    arm_jump_cc(&cmp, label);
    arm_free_cc(&cmp);
}

static const uint8_t table_logic_cc[16] = {
    1, /* and */
    1, /* xor */
    0, /* sub */
    0, /* rsb */
    0, /* add */
    0, /* adc */
    0, /* sbc */
    0, /* rsc */
    1, /* andl */
    1, /* xorl */
    0, /* cmp */
    0, /* cmn */
    1, /* orr */
    1, /* mov */
    1, /* bic */
    1, /* mvn */
};

/* Set PC and Thumb state from an immediate address.  */
static inline void gen_bx_im(DisasContext *s, uint32_t addr)
{
    TCGv_i32 tmp;

    s->is_jmp = DISAS_JUMP;
    if (s->thumb != (addr & 1)) {
        tmp = tcg_temp_new_i32();
        tcg_gen_movi_i32(tmp, addr & 1);
        tcg_gen_st_i32(tmp, cpu_env, offsetof(CPUARMState, thumb));
        tcg_temp_free_i32(tmp);
    }
    tcg_gen_movi_i32(cpu_R[15], addr & ~1);
}

/* Set PC and Thumb state from var.  var is marked as dead.  */
static inline void gen_bx(DisasContext *s, TCGv_i32 var)
{
    s->is_jmp = DISAS_JUMP;
    tcg_gen_andi_i32(cpu_R[15], var, ~1);
    tcg_gen_andi_i32(var, var, 1);
    store_cpu_field(var, thumb);
}

/* Variant of store_reg which uses branch&exchange logic when storing
   to r15 in ARM architecture v7 and above. The source must be a temporary
   and will be marked as dead. */
static inline void store_reg_bx(DisasContext *s, int reg, TCGv_i32 var)
{
    if (reg == 15 && ENABLE_ARCH_7) {
        gen_bx(s, var);
    } else {
        store_reg(s, reg, var);
    }
}

/* Variant of store_reg which uses branch&exchange logic when storing
 * to r15 in ARM architecture v5T and above. This is used for storing
 * the results of a LDR/LDM/POP into r15, and corresponds to the cases
 * in the ARM ARM which use the LoadWritePC() pseudocode function. */
static inline void store_reg_from_load(DisasContext *s, int reg, TCGv_i32 var)
{
    if (reg == 15 && ENABLE_ARCH_5) {
        gen_bx(s, var);
    } else {
        store_reg(s, reg, var);
    }
}

#ifdef CONFIG_USER_ONLY
#define IS_USER_ONLY 1
#else
#define IS_USER_ONLY 0
#endif

/* Abstractions of "generate code to do a guest load/store for
 * AArch32", where a vaddr is always 32 bits (and is zero
 * extended if we're a 64 bit core) and  data is also
 * 32 bits unless specifically doing a 64 bit access.
 * These functions work like tcg_gen_qemu_{ld,st}* except
 * that the address argument is TCGv_i32 rather than TCGv.
 */

static inline TCGv gen_aa32_addr(DisasContext *s, TCGv_i32 a32, TCGMemOp op)
{
    TCGv addr = tcg_temp_new();
    tcg_gen_extu_i32_tl(addr, a32);

    /* Not needed for user-mode BE32, where we use MO_BE instead.  */
    if (!IS_USER_ONLY && s->sctlr_b && (op & MO_SIZE) < MO_32) {
        tcg_gen_xori_tl(addr, addr, 4 - (1 << (op & MO_SIZE)));
    }
    return addr;
}

static void gen_aa32_ld_i32(DisasContext *s, TCGv_i32 val, TCGv_i32 a32,
                            int index, TCGMemOp opc)
{
    TCGv addr = gen_aa32_addr(s, a32, opc);
    tcg_gen_qemu_ld_i32(val, addr, index, opc);
    tcg_temp_free(addr);
}

static void gen_aa32_st_i32(DisasContext *s, TCGv_i32 val, TCGv_i32 a32,
                            int index, TCGMemOp opc)
{
    TCGv addr = gen_aa32_addr(s, a32, opc);
    tcg_gen_qemu_st_i32(val, addr, index, opc);
    tcg_temp_free(addr);
}

#define DO_GEN_LD(SUFF, OPC)                                             \
static inline void gen_aa32_ld##SUFF(DisasContext *s, TCGv_i32 val,      \
                                     TCGv_i32 a32, int index)            \
{                                                                        \
    gen_aa32_ld_i32(s, val, a32, index, OPC | s->be_data);               \
}

#define DO_GEN_ST(SUFF, OPC)                                             \
static inline void gen_aa32_st##SUFF(DisasContext *s, TCGv_i32 val,      \
                                     TCGv_i32 a32, int index)            \
{                                                                        \
    gen_aa32_st_i32(s, val, a32, index, OPC | s->be_data);               \
}

static inline void gen_aa32_frob64(DisasContext *s, TCGv_i64 val)
{
    /* Not needed for user-mode BE32, where we use MO_BE instead.  */
    if (!IS_USER_ONLY && s->sctlr_b) {
        tcg_gen_rotri_i64(val, val, 32);
    }
}

static void gen_aa32_ld_i64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32,
                            int index, TCGMemOp opc)
{
    TCGv addr = gen_aa32_addr(s, a32, opc);
    tcg_gen_qemu_ld_i64(val, addr, index, opc);
    gen_aa32_frob64(s, val);
    tcg_temp_free(addr);
}

static inline void gen_aa32_ld64(DisasContext *s, TCGv_i64 val,
                                 TCGv_i32 a32, int index)
{
    gen_aa32_ld_i64(s, val, a32, index, MO_Q | s->be_data);
}

static void gen_aa32_st_i64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32,
                            int index, TCGMemOp opc)
{
    TCGv addr = gen_aa32_addr(s, a32, opc);

    /* Not needed for user-mode BE32, where we use MO_BE instead.  */
    if (!IS_USER_ONLY && s->sctlr_b) {
        TCGv_i64 tmp = tcg_temp_new_i64();
        tcg_gen_rotri_i64(tmp, val, 32);
        tcg_gen_qemu_st_i64(tmp, addr, index, opc);

        tcg_temp_free_i64(tmp);
    } else {
        tcg_gen_qemu_st_i64(val, addr, index, opc);
    }
    tcg_temp_free(addr);
}

static inline void gen_aa32_st64(DisasContext *s, TCGv_i64 val,
                                 TCGv_i32 a32, int index)
{
    gen_aa32_st_i64(s, val, a32, index, MO_Q | s->be_data);
}

DO_GEN_LD(8s, MO_SB)
DO_GEN_LD(8u, MO_UB)
DO_GEN_LD(16s, MO_SW)
DO_GEN_LD(16u, MO_UW)
DO_GEN_LD(32u, MO_UL)
DO_GEN_ST(8, MO_UB)
DO_GEN_ST(16, MO_UW)
DO_GEN_ST(32, MO_UL)

static inline void gen_set_pc_im(DisasContext *s, target_ulong val)
{
    tcg_gen_movi_i32(cpu_R[15], val);
}

static inline void gen_hvc(DisasContext *s, int imm16)
{
    /* The pre HVC helper handles cases when HVC gets trapped
     * as an undefined insn by runtime configuration (ie before
     * the insn really executes).
     */
    gen_set_pc_im(s, s->pc - 4);
    gen_helper_pre_hvc(cpu_env);
    /* Otherwise we will treat this as a real exception which
     * happens after execution of the insn. (The distinction matters
     * for the PC value reported to the exception handler and also
     * for single stepping.)
     */
    s->svc_imm = imm16;
    gen_set_pc_im(s, s->pc);
    s->is_jmp = DISAS_HVC;
}

static inline void gen_smc(DisasContext *s)
{
    /* As with HVC, we may take an exception either before or after
     * the insn executes.
     */
    TCGv_i32 tmp;

    gen_set_pc_im(s, s->pc - 4);
    tmp = tcg_const_i32(syn_aa32_smc());
    gen_helper_pre_smc(cpu_env, tmp);
    tcg_temp_free_i32(tmp);
    gen_set_pc_im(s, s->pc);
    s->is_jmp = DISAS_SMC;
}

static inline void
gen_set_condexec (DisasContext *s)
{
    if (s->condexec_mask) {
        uint32_t val = (s->condexec_cond << 4) | (s->condexec_mask >> 1);
        TCGv_i32 tmp = tcg_temp_new_i32();
        tcg_gen_movi_i32(tmp, val);
        store_cpu_field(tmp, condexec_bits);
    }
}

static void gen_exception_internal_insn(DisasContext *s, int offset, int excp)
{
    gen_set_condexec(s);
    gen_set_pc_im(s, s->pc - offset);
    gen_exception_internal(excp);
    s->is_jmp = DISAS_JUMP;
}

static void gen_exception_insn(DisasContext *s, int offset, int excp,
                               int syn, uint32_t target_el)
{
    gen_set_condexec(s);
    gen_set_pc_im(s, s->pc - offset);
    gen_exception(excp, syn, target_el);
    s->is_jmp = DISAS_JUMP;
}

/* Force a TB lookup after an instruction that changes the CPU state.  */
static inline void gen_lookup_tb(DisasContext *s)
{
    tcg_gen_movi_i32(cpu_R[15], s->pc & ~1);
    s->is_jmp = DISAS_JUMP;
}

static inline void gen_hlt(DisasContext *s, int imm)
{
    /* HLT. This has two purposes.
     * Architecturally, it is an external halting debug instruction.
     * Since QEMU doesn't implement external debug, we treat this as
     * it is required for halting debug disabled: it will UNDEF.
     * Secondly, "HLT 0x3C" is a T32 semihosting trap instruction,
     * and "HLT 0xF000" is an A32 semihosting syscall. These traps
     * must trigger semihosting even for ARMv7 and earlier, where
     * HLT was an undefined encoding.
     * In system mode, we don't allow userspace access to
     * semihosting, to provide some semblance of security
     * (and for consistency with our 32-bit semihosting).
     */
    if (semihosting_enabled() &&
#ifndef CONFIG_USER_ONLY
        s->current_el != 0 &&
#endif
        (imm == (s->thumb ? 0x3c : 0xf000))) {
        gen_exception_internal_insn(s, 0, EXCP_SEMIHOST);
        return;
    }

    gen_exception_insn(s, s->thumb ? 2 : 4, EXCP_UDEF, syn_uncategorized(),
                       default_exception_el(s));
}

static inline void gen_add_data_offset(DisasContext *s, unsigned int insn,
                                       TCGv_i32 var)
{
    int val, rm, shift, shiftop;
    TCGv_i32 offset;

    if (!(insn & (1 << 25))) {
        /* immediate */
        val = insn & 0xfff;
        if (!(insn & (1 << 23)))
            val = -val;
        if (val != 0)
            tcg_gen_addi_i32(var, var, val);
    } else {
        /* shift/register */
        rm = (insn) & 0xf;
        shift = (insn >> 7) & 0x1f;
        shiftop = (insn >> 5) & 3;
        offset = load_reg(s, rm);
        gen_arm_shift_im(offset, shiftop, shift, 0);
        if (!(insn & (1 << 23)))
            tcg_gen_sub_i32(var, var, offset);
        else
            tcg_gen_add_i32(var, var, offset);
        tcg_temp_free_i32(offset);
    }
}

static inline void gen_add_datah_offset(DisasContext *s, unsigned int insn,
                                        int extra, TCGv_i32 var)
{
    int val, rm;
    TCGv_i32 offset;

    if (insn & (1 << 22)) {
        /* immediate */
        val = (insn & 0xf) | ((insn >> 4) & 0xf0);
        if (!(insn & (1 << 23)))
            val = -val;
        val += extra;
        if (val != 0)
            tcg_gen_addi_i32(var, var, val);
    } else {
        /* register */
        if (extra)
            tcg_gen_addi_i32(var, var, extra);
        rm = (insn) & 0xf;
        offset = load_reg(s, rm);
        if (!(insn & (1 << 23)))
            tcg_gen_sub_i32(var, var, offset);
        else
            tcg_gen_add_i32(var, var, offset);
        tcg_temp_free_i32(offset);
    }
}

static TCGv_ptr get_fpstatus_ptr(int neon)
{
    TCGv_ptr statusptr = tcg_temp_new_ptr();
    int offset;
    if (neon) {
        offset = offsetof(CPUARMState, vfp.standard_fp_status);
    } else {
        offset = offsetof(CPUARMState, vfp.fp_status);
    }
    tcg_gen_addi_ptr(statusptr, cpu_env, offset);
    return statusptr;
}

#define VFP_OP2(name)                                                 \
static inline void gen_vfp_##name(int dp)                             \
{                                                                     \
    TCGv_ptr fpst = get_fpstatus_ptr(0);                              \
    if (dp) {                                                         \
        gen_helper_vfp_##name##d(cpu_F0d, cpu_F0d, cpu_F1d, fpst);    \
    } else {                                                          \
        gen_helper_vfp_##name##s(cpu_F0s, cpu_F0s, cpu_F1s, fpst);    \
    }                                                                 \
    tcg_temp_free_ptr(fpst);                                          \
}

VFP_OP2(add)
VFP_OP2(sub)
VFP_OP2(mul)
VFP_OP2(div)

#undef VFP_OP2

static inline void gen_vfp_F1_mul(int dp)
{
    /* Like gen_vfp_mul() but put result in F1 */
    TCGv_ptr fpst = get_fpstatus_ptr(0);
    if (dp) {
        gen_helper_vfp_muld(cpu_F1d, cpu_F0d, cpu_F1d, fpst);
    } else {
        gen_helper_vfp_muls(cpu_F1s, cpu_F0s, cpu_F1s, fpst);
    }
    tcg_temp_free_ptr(fpst);
}

static inline void gen_vfp_F1_neg(int dp)
{
    /* Like gen_vfp_neg() but put result in F1 */
    if (dp) {
        gen_helper_vfp_negd(cpu_F1d, cpu_F0d);
    } else {
        gen_helper_vfp_negs(cpu_F1s, cpu_F0s);
    }
}

static inline void gen_vfp_abs(int dp)
{
    if (dp)
        gen_helper_vfp_absd(cpu_F0d, cpu_F0d);
    else
        gen_helper_vfp_abss(cpu_F0s, cpu_F0s);
}

static inline void gen_vfp_neg(int dp)
{
    if (dp)
        gen_helper_vfp_negd(cpu_F0d, cpu_F0d);
    else
        gen_helper_vfp_negs(cpu_F0s, cpu_F0s);
}

static inline void gen_vfp_sqrt(int dp)
{
    if (dp)
        gen_helper_vfp_sqrtd(cpu_F0d, cpu_F0d, cpu_env);
    else
        gen_helper_vfp_sqrts(cpu_F0s, cpu_F0s, cpu_env);
}

static inline void gen_vfp_cmp(int dp)
{
    if (dp)
        gen_helper_vfp_cmpd(cpu_F0d, cpu_F1d, cpu_env);
    else
        gen_helper_vfp_cmps(cpu_F0s, cpu_F1s, cpu_env);
}

static inline void gen_vfp_cmpe(int dp)
{
    if (dp)
        gen_helper_vfp_cmped(cpu_F0d, cpu_F1d, cpu_env);
    else
        gen_helper_vfp_cmpes(cpu_F0s, cpu_F1s, cpu_env);
}

static inline void gen_vfp_F1_ld0(int dp)
{
    if (dp)
        tcg_gen_movi_i64(cpu_F1d, 0);
    else
        tcg_gen_movi_i32(cpu_F1s, 0);
}

#define VFP_GEN_ITOF(name) \
static inline void gen_vfp_##name(int dp, int neon) \
{ \
    TCGv_ptr statusptr = get_fpstatus_ptr(neon); \
    if (dp) { \
        gen_helper_vfp_##name##d(cpu_F0d, cpu_F0s, statusptr); \
    } else { \
        gen_helper_vfp_##name##s(cpu_F0s, cpu_F0s, statusptr); \
    } \
    tcg_temp_free_ptr(statusptr); \
}

VFP_GEN_ITOF(uito)
VFP_GEN_ITOF(sito)
#undef VFP_GEN_ITOF

#define VFP_GEN_FTOI(name) \
static inline void gen_vfp_##name(int dp, int neon) \
{ \
    TCGv_ptr statusptr = get_fpstatus_ptr(neon); \
    if (dp) { \
        gen_helper_vfp_##name##d(cpu_F0s, cpu_F0d, statusptr); \
    } else { \
        gen_helper_vfp_##name##s(cpu_F0s, cpu_F0s, statusptr); \
    } \
    tcg_temp_free_ptr(statusptr); \
}

VFP_GEN_FTOI(toui)
VFP_GEN_FTOI(touiz)
VFP_GEN_FTOI(tosi)
VFP_GEN_FTOI(tosiz)
#undef VFP_GEN_FTOI

#define VFP_GEN_FIX(name, round) \
static inline void gen_vfp_##name(int dp, int shift, int neon) \
{ \
    TCGv_i32 tmp_shift = tcg_const_i32(shift); \
    TCGv_ptr statusptr = get_fpstatus_ptr(neon); \
    if (dp) { \
        gen_helper_vfp_##name##d##round(cpu_F0d, cpu_F0d, tmp_shift, \
                                        statusptr); \
    } else { \
        gen_helper_vfp_##name##s##round(cpu_F0s, cpu_F0s, tmp_shift, \
                                        statusptr); \
    } \
    tcg_temp_free_i32(tmp_shift); \
    tcg_temp_free_ptr(statusptr); \
}
VFP_GEN_FIX(tosh, _round_to_zero)
VFP_GEN_FIX(tosl, _round_to_zero)
VFP_GEN_FIX(touh, _round_to_zero)
VFP_GEN_FIX(toul, _round_to_zero)
VFP_GEN_FIX(shto, )
VFP_GEN_FIX(slto, )
VFP_GEN_FIX(uhto, )
VFP_GEN_FIX(ulto, )
#undef VFP_GEN_FIX

static inline void gen_vfp_ld(DisasContext *s, int dp, TCGv_i32 addr)
{
    if (dp) {
        gen_aa32_ld64(s, cpu_F0d, addr, get_mem_index(s));
    } else {
        gen_aa32_ld32u(s, cpu_F0s, addr, get_mem_index(s));
    }
}

static inline void gen_vfp_st(DisasContext *s, int dp, TCGv_i32 addr)
{
    if (dp) {
        gen_aa32_st64(s, cpu_F0d, addr, get_mem_index(s));
    } else {
        gen_aa32_st32(s, cpu_F0s, addr, get_mem_index(s));
    }
}

static inline long
vfp_reg_offset (int dp, int reg)
{
    if (dp)
        return offsetof(CPUARMState, vfp.regs[reg]);
    else if (reg & 1) {
        return offsetof(CPUARMState, vfp.regs[reg >> 1])
          + offsetof(CPU_DoubleU, l.upper);
    } else {
        return offsetof(CPUARMState, vfp.regs[reg >> 1])
          + offsetof(CPU_DoubleU, l.lower);
    }
}

/* Return the offset of a 32-bit piece of a NEON register.
   zero is the least significant end of the register.  */
static inline long
neon_reg_offset (int reg, int n)
{
    int sreg;
    sreg = reg * 2 + n;
    return vfp_reg_offset(0, sreg);
}

static TCGv_i32 neon_load_reg(int reg, int pass)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_ld_i32(tmp, cpu_env, neon_reg_offset(reg, pass));
    return tmp;
}

static void neon_store_reg(int reg, int pass, TCGv_i32 var)
{
    tcg_gen_st_i32(var, cpu_env, neon_reg_offset(reg, pass));
    tcg_temp_free_i32(var);
}

static inline void neon_load_reg64(TCGv_i64 var, int reg)
{
    tcg_gen_ld_i64(var, cpu_env, vfp_reg_offset(1, reg));
}

static inline void neon_store_reg64(TCGv_i64 var, int reg)
{
    tcg_gen_st_i64(var, cpu_env, vfp_reg_offset(1, reg));
}

#define tcg_gen_ld_f32 tcg_gen_ld_i32
#define tcg_gen_ld_f64 tcg_gen_ld_i64
#define tcg_gen_st_f32 tcg_gen_st_i32
#define tcg_gen_st_f64 tcg_gen_st_i64

static inline void gen_mov_F0_vreg(int dp, int reg)
{
    if (dp)
        tcg_gen_ld_f64(cpu_F0d, cpu_env, vfp_reg_offset(dp, reg));
    else
        tcg_gen_ld_f32(cpu_F0s, cpu_env, vfp_reg_offset(dp, reg));
}

static inline void gen_mov_F1_vreg(int dp, int reg)
{
    if (dp)
        tcg_gen_ld_f64(cpu_F1d, cpu_env, vfp_reg_offset(dp, reg));
    else
        tcg_gen_ld_f32(cpu_F1s, cpu_env, vfp_reg_offset(dp, reg));
}

static inline void gen_mov_vreg_F0(int dp, int reg)
{
    if (dp)
        tcg_gen_st_f64(cpu_F0d, cpu_env, vfp_reg_offset(dp, reg));
    else
        tcg_gen_st_f32(cpu_F0s, cpu_env, vfp_reg_offset(dp, reg));
}

#define ARM_CP_RW_BIT   (1 << 20)

static inline void iwmmxt_load_reg(TCGv_i64 var, int reg)
{
    tcg_gen_ld_i64(var, cpu_env, offsetof(CPUARMState, iwmmxt.regs[reg]));
}

static inline void iwmmxt_store_reg(TCGv_i64 var, int reg)
{
    tcg_gen_st_i64(var, cpu_env, offsetof(CPUARMState, iwmmxt.regs[reg]));
}

static inline TCGv_i32 iwmmxt_load_creg(int reg)
{
    TCGv_i32 var = tcg_temp_new_i32();
    tcg_gen_ld_i32(var, cpu_env, offsetof(CPUARMState, iwmmxt.cregs[reg]));
    return var;
}

static inline void iwmmxt_store_creg(int reg, TCGv_i32 var)
{
    tcg_gen_st_i32(var, cpu_env, offsetof(CPUARMState, iwmmxt.cregs[reg]));
    tcg_temp_free_i32(var);
}

static inline void gen_op_iwmmxt_movq_wRn_M0(int rn)
{
    iwmmxt_store_reg(cpu_M0, rn);
}

static inline void gen_op_iwmmxt_movq_M0_wRn(int rn)
{
    iwmmxt_load_reg(cpu_M0, rn);
}

static inline void gen_op_iwmmxt_orq_M0_wRn(int rn)
{
    iwmmxt_load_reg(cpu_V1, rn);
    tcg_gen_or_i64(cpu_M0, cpu_M0, cpu_V1);
}

static inline void gen_op_iwmmxt_andq_M0_wRn(int rn)
{
    iwmmxt_load_reg(cpu_V1, rn);
    tcg_gen_and_i64(cpu_M0, cpu_M0, cpu_V1);
}

static inline void gen_op_iwmmxt_xorq_M0_wRn(int rn)
{
    iwmmxt_load_reg(cpu_V1, rn);
    tcg_gen_xor_i64(cpu_M0, cpu_M0, cpu_V1);
}

#define IWMMXT_OP(name) \
static inline void gen_op_iwmmxt_##name##_M0_wRn(int rn) \
{ \
    iwmmxt_load_reg(cpu_V1, rn); \
    gen_helper_iwmmxt_##name(cpu_M0, cpu_M0, cpu_V1); \
}

#define IWMMXT_OP_ENV(name) \
static inline void gen_op_iwmmxt_##name##_M0_wRn(int rn) \
{ \
    iwmmxt_load_reg(cpu_V1, rn); \
    gen_helper_iwmmxt_##name(cpu_M0, cpu_env, cpu_M0, cpu_V1); \
}

#define IWMMXT_OP_ENV_SIZE(name) \
IWMMXT_OP_ENV(name##b) \
IWMMXT_OP_ENV(name##w) \
IWMMXT_OP_ENV(name##l)

#define IWMMXT_OP_ENV1(name) \
static inline void gen_op_iwmmxt_##name##_M0(void) \
{ \
    gen_helper_iwmmxt_##name(cpu_M0, cpu_env, cpu_M0); \
}

IWMMXT_OP(maddsq)
IWMMXT_OP(madduq)
IWMMXT_OP(sadb)
IWMMXT_OP(sadw)
IWMMXT_OP(mulslw)
IWMMXT_OP(mulshw)
IWMMXT_OP(mululw)
IWMMXT_OP(muluhw)
IWMMXT_OP(macsw)
IWMMXT_OP(macuw)

IWMMXT_OP_ENV_SIZE(unpackl)
IWMMXT_OP_ENV_SIZE(unpackh)

IWMMXT_OP_ENV1(unpacklub)
IWMMXT_OP_ENV1(unpackluw)
IWMMXT_OP_ENV1(unpacklul)
IWMMXT_OP_ENV1(unpackhub)
IWMMXT_OP_ENV1(unpackhuw)
IWMMXT_OP_ENV1(unpackhul)
IWMMXT_OP_ENV1(unpacklsb)
IWMMXT_OP_ENV1(unpacklsw)
IWMMXT_OP_ENV1(unpacklsl)
IWMMXT_OP_ENV1(unpackhsb)
IWMMXT_OP_ENV1(unpackhsw)
IWMMXT_OP_ENV1(unpackhsl)

IWMMXT_OP_ENV_SIZE(cmpeq)
IWMMXT_OP_ENV_SIZE(cmpgtu)
IWMMXT_OP_ENV_SIZE(cmpgts)

IWMMXT_OP_ENV_SIZE(mins)
IWMMXT_OP_ENV_SIZE(minu)
IWMMXT_OP_ENV_SIZE(maxs)
IWMMXT_OP_ENV_SIZE(maxu)

IWMMXT_OP_ENV_SIZE(subn)
IWMMXT_OP_ENV_SIZE(addn)
IWMMXT_OP_ENV_SIZE(subu)
IWMMXT_OP_ENV_SIZE(addu)
IWMMXT_OP_ENV_SIZE(subs)
IWMMXT_OP_ENV_SIZE(adds)

IWMMXT_OP_ENV(avgb0)
IWMMXT_OP_ENV(avgb1)
IWMMXT_OP_ENV(avgw0)
IWMMXT_OP_ENV(avgw1)

IWMMXT_OP_ENV(packuw)
IWMMXT_OP_ENV(packul)
IWMMXT_OP_ENV(packuq)
IWMMXT_OP_ENV(packsw)
IWMMXT_OP_ENV(packsl)
IWMMXT_OP_ENV(packsq)

static void gen_op_iwmmxt_set_mup(void)
{
    TCGv_i32 tmp;
    tmp = load_cpu_field(iwmmxt.cregs[ARM_IWMMXT_wCon]);
    tcg_gen_ori_i32(tmp, tmp, 2);
    store_cpu_field(tmp, iwmmxt.cregs[ARM_IWMMXT_wCon]);
}

static void gen_op_iwmmxt_set_cup(void)
{
    TCGv_i32 tmp;
    tmp = load_cpu_field(iwmmxt.cregs[ARM_IWMMXT_wCon]);
    tcg_gen_ori_i32(tmp, tmp, 1);
    store_cpu_field(tmp, iwmmxt.cregs[ARM_IWMMXT_wCon]);
}

static void gen_op_iwmmxt_setpsr_nz(void)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    gen_helper_iwmmxt_setpsr_nz(tmp, cpu_M0);
    store_cpu_field(tmp, iwmmxt.cregs[ARM_IWMMXT_wCASF]);
}

static inline void gen_op_iwmmxt_addl_M0_wRn(int rn)
{
    iwmmxt_load_reg(cpu_V1, rn);
    tcg_gen_ext32u_i64(cpu_V1, cpu_V1);
    tcg_gen_add_i64(cpu_M0, cpu_M0, cpu_V1);
}

static inline int gen_iwmmxt_address(DisasContext *s, uint32_t insn,
                                     TCGv_i32 dest)
{
    int rd;
    uint32_t offset;
    TCGv_i32 tmp;

    rd = (insn >> 16) & 0xf;
    tmp = load_reg(s, rd);

    offset = (insn & 0xff) << ((insn >> 7) & 2);
    if (insn & (1 << 24)) {
        /* Pre indexed */
        if (insn & (1 << 23))
            tcg_gen_addi_i32(tmp, tmp, offset);
        else
            tcg_gen_addi_i32(tmp, tmp, -offset);
        tcg_gen_mov_i32(dest, tmp);
        if (insn & (1 << 21))
            store_reg(s, rd, tmp);
        else
            tcg_temp_free_i32(tmp);
    } else if (insn & (1 << 21)) {
        /* Post indexed */
        tcg_gen_mov_i32(dest, tmp);
        if (insn & (1 << 23))
            tcg_gen_addi_i32(tmp, tmp, offset);
        else
            tcg_gen_addi_i32(tmp, tmp, -offset);
        store_reg(s, rd, tmp);
    } else if (!(insn & (1 << 23)))
        return 1;
    return 0;
}

static inline int gen_iwmmxt_shift(uint32_t insn, uint32_t mask, TCGv_i32 dest)
{
    int rd = (insn >> 0) & 0xf;
    TCGv_i32 tmp;

    if (insn & (1 << 8)) {
        if (rd < ARM_IWMMXT_wCGR0 || rd > ARM_IWMMXT_wCGR3) {
            return 1;
        } else {
            tmp = iwmmxt_load_creg(rd);
        }
    } else {
        tmp = tcg_temp_new_i32();
        iwmmxt_load_reg(cpu_V0, rd);
        tcg_gen_extrl_i64_i32(tmp, cpu_V0);
    }
    tcg_gen_andi_i32(tmp, tmp, mask);
    tcg_gen_mov_i32(dest, tmp);
    tcg_temp_free_i32(tmp);
    return 0;
}

/* Disassemble an iwMMXt instruction.  Returns nonzero if an error occurred
   (ie. an undefined instruction).  */
static int disas_iwmmxt_insn(DisasContext *s, uint32_t insn)
{
    int rd, wrd;
    int rdhi, rdlo, rd0, rd1, i;
    TCGv_i32 addr;
    TCGv_i32 tmp, tmp2, tmp3;

    if ((insn & 0x0e000e00) == 0x0c000000) {
        if ((insn & 0x0fe00ff0) == 0x0c400000) {
            wrd = insn & 0xf;
            rdlo = (insn >> 12) & 0xf;
            rdhi = (insn >> 16) & 0xf;
            if (insn & ARM_CP_RW_BIT) {                 /* TMRRC */
                iwmmxt_load_reg(cpu_V0, wrd);
                tcg_gen_extrl_i64_i32(cpu_R[rdlo], cpu_V0);
                tcg_gen_shri_i64(cpu_V0, cpu_V0, 32);
                tcg_gen_extrl_i64_i32(cpu_R[rdhi], cpu_V0);
            } else {                                    /* TMCRR */
                tcg_gen_concat_i32_i64(cpu_V0, cpu_R[rdlo], cpu_R[rdhi]);
                iwmmxt_store_reg(cpu_V0, wrd);
                gen_op_iwmmxt_set_mup();
            }
            return 0;
        }

        wrd = (insn >> 12) & 0xf;
        addr = tcg_temp_new_i32();
        if (gen_iwmmxt_address(s, insn, addr)) {
            tcg_temp_free_i32(addr);
            return 1;
        }
        if (insn & ARM_CP_RW_BIT) {
            if ((insn >> 28) == 0xf) {                  /* WLDRW wCx */
                tmp = tcg_temp_new_i32();
                gen_aa32_ld32u(s, tmp, addr, get_mem_index(s));
                iwmmxt_store_creg(wrd, tmp);
            } else {
                i = 1;
                if (insn & (1 << 8)) {
                    if (insn & (1 << 22)) {             /* WLDRD */
                        gen_aa32_ld64(s, cpu_M0, addr, get_mem_index(s));
                        i = 0;
                    } else {                            /* WLDRW wRd */
                        tmp = tcg_temp_new_i32();
                        gen_aa32_ld32u(s, tmp, addr, get_mem_index(s));
                    }
                } else {
                    tmp = tcg_temp_new_i32();
                    if (insn & (1 << 22)) {             /* WLDRH */
                        gen_aa32_ld16u(s, tmp, addr, get_mem_index(s));
                    } else {                            /* WLDRB */
                        gen_aa32_ld8u(s, tmp, addr, get_mem_index(s));
                    }
                }
                if (i) {
                    tcg_gen_extu_i32_i64(cpu_M0, tmp);
                    tcg_temp_free_i32(tmp);
                }
                gen_op_iwmmxt_movq_wRn_M0(wrd);
            }
        } else {
            if ((insn >> 28) == 0xf) {                  /* WSTRW wCx */
                tmp = iwmmxt_load_creg(wrd);
                gen_aa32_st32(s, tmp, addr, get_mem_index(s));
            } else {
                gen_op_iwmmxt_movq_M0_wRn(wrd);
                tmp = tcg_temp_new_i32();
                if (insn & (1 << 8)) {
                    if (insn & (1 << 22)) {             /* WSTRD */
                        gen_aa32_st64(s, cpu_M0, addr, get_mem_index(s));
                    } else {                            /* WSTRW wRd */
                        tcg_gen_extrl_i64_i32(tmp, cpu_M0);
                        gen_aa32_st32(s, tmp, addr, get_mem_index(s));
                    }
                } else {
                    if (insn & (1 << 22)) {             /* WSTRH */
                        tcg_gen_extrl_i64_i32(tmp, cpu_M0);
                        gen_aa32_st16(s, tmp, addr, get_mem_index(s));
                    } else {                            /* WSTRB */
                        tcg_gen_extrl_i64_i32(tmp, cpu_M0);
                        gen_aa32_st8(s, tmp, addr, get_mem_index(s));
                    }
                }
            }
            tcg_temp_free_i32(tmp);
        }
        tcg_temp_free_i32(addr);
        return 0;
    }

    if ((insn & 0x0f000000) != 0x0e000000)
        return 1;

    switch (((insn >> 12) & 0xf00) | ((insn >> 4) & 0xff)) {
    case 0x000:                                         /* WOR */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 0) & 0xf;
        rd1 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        gen_op_iwmmxt_orq_M0_wRn(rd1);
        gen_op_iwmmxt_setpsr_nz();
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x011:                                         /* TMCR */
        if (insn & 0xf)
            return 1;
        rd = (insn >> 12) & 0xf;
        wrd = (insn >> 16) & 0xf;
        switch (wrd) {
        case ARM_IWMMXT_wCID:
        case ARM_IWMMXT_wCASF:
            break;
        case ARM_IWMMXT_wCon:
            gen_op_iwmmxt_set_cup();
            /* Fall through.  */
        case ARM_IWMMXT_wCSSF:
            tmp = iwmmxt_load_creg(wrd);
            tmp2 = load_reg(s, rd);
            tcg_gen_andc_i32(tmp, tmp, tmp2);
            tcg_temp_free_i32(tmp2);
            iwmmxt_store_creg(wrd, tmp);
            break;
        case ARM_IWMMXT_wCGR0:
        case ARM_IWMMXT_wCGR1:
        case ARM_IWMMXT_wCGR2:
        case ARM_IWMMXT_wCGR3:
            gen_op_iwmmxt_set_cup();
            tmp = load_reg(s, rd);
            iwmmxt_store_creg(wrd, tmp);
            break;
        default:
            return 1;
        }
        break;
    case 0x100:                                         /* WXOR */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 0) & 0xf;
        rd1 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        gen_op_iwmmxt_xorq_M0_wRn(rd1);
        gen_op_iwmmxt_setpsr_nz();
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x111:                                         /* TMRC */
        if (insn & 0xf)
            return 1;
        rd = (insn >> 12) & 0xf;
        wrd = (insn >> 16) & 0xf;
        tmp = iwmmxt_load_creg(wrd);
        store_reg(s, rd, tmp);
        break;
    case 0x300:                                         /* WANDN */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 0) & 0xf;
        rd1 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tcg_gen_neg_i64(cpu_M0, cpu_M0);
        gen_op_iwmmxt_andq_M0_wRn(rd1);
        gen_op_iwmmxt_setpsr_nz();
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x200:                                         /* WAND */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 0) & 0xf;
        rd1 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        gen_op_iwmmxt_andq_M0_wRn(rd1);
        gen_op_iwmmxt_setpsr_nz();
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x810: case 0xa10:                             /* WMADD */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 0) & 0xf;
        rd1 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        if (insn & (1 << 21))
            gen_op_iwmmxt_maddsq_M0_wRn(rd1);
        else
            gen_op_iwmmxt_madduq_M0_wRn(rd1);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x10e: case 0x50e: case 0x90e: case 0xd0e:     /* WUNPCKIL */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            gen_op_iwmmxt_unpacklb_M0_wRn(rd1);
            break;
        case 1:
            gen_op_iwmmxt_unpacklw_M0_wRn(rd1);
            break;
        case 2:
            gen_op_iwmmxt_unpackll_M0_wRn(rd1);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x10c: case 0x50c: case 0x90c: case 0xd0c:     /* WUNPCKIH */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            gen_op_iwmmxt_unpackhb_M0_wRn(rd1);
            break;
        case 1:
            gen_op_iwmmxt_unpackhw_M0_wRn(rd1);
            break;
        case 2:
            gen_op_iwmmxt_unpackhl_M0_wRn(rd1);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x012: case 0x112: case 0x412: case 0x512:     /* WSAD */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        if (insn & (1 << 22))
            gen_op_iwmmxt_sadw_M0_wRn(rd1);
        else
            gen_op_iwmmxt_sadb_M0_wRn(rd1);
        if (!(insn & (1 << 20)))
            gen_op_iwmmxt_addl_M0_wRn(wrd);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x010: case 0x110: case 0x210: case 0x310:     /* WMUL */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        if (insn & (1 << 21)) {
            if (insn & (1 << 20))
                gen_op_iwmmxt_mulshw_M0_wRn(rd1);
            else
                gen_op_iwmmxt_mulslw_M0_wRn(rd1);
        } else {
            if (insn & (1 << 20))
                gen_op_iwmmxt_muluhw_M0_wRn(rd1);
            else
                gen_op_iwmmxt_mululw_M0_wRn(rd1);
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x410: case 0x510: case 0x610: case 0x710:     /* WMAC */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        if (insn & (1 << 21))
            gen_op_iwmmxt_macsw_M0_wRn(rd1);
        else
            gen_op_iwmmxt_macuw_M0_wRn(rd1);
        if (!(insn & (1 << 20))) {
            iwmmxt_load_reg(cpu_V1, wrd);
            tcg_gen_add_i64(cpu_M0, cpu_M0, cpu_V1);
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x006: case 0x406: case 0x806: case 0xc06:     /* WCMPEQ */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            gen_op_iwmmxt_cmpeqb_M0_wRn(rd1);
            break;
        case 1:
            gen_op_iwmmxt_cmpeqw_M0_wRn(rd1);
            break;
        case 2:
            gen_op_iwmmxt_cmpeql_M0_wRn(rd1);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x800: case 0x900: case 0xc00: case 0xd00:     /* WAVG2 */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        if (insn & (1 << 22)) {
            if (insn & (1 << 20))
                gen_op_iwmmxt_avgw1_M0_wRn(rd1);
            else
                gen_op_iwmmxt_avgw0_M0_wRn(rd1);
        } else {
            if (insn & (1 << 20))
                gen_op_iwmmxt_avgb1_M0_wRn(rd1);
            else
                gen_op_iwmmxt_avgb0_M0_wRn(rd1);
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x802: case 0x902: case 0xa02: case 0xb02:     /* WALIGNR */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = iwmmxt_load_creg(ARM_IWMMXT_wCGR0 + ((insn >> 20) & 3));
        tcg_gen_andi_i32(tmp, tmp, 7);
        iwmmxt_load_reg(cpu_V1, rd1);
        gen_helper_iwmmxt_align(cpu_M0, cpu_M0, cpu_V1, tmp);
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x601: case 0x605: case 0x609: case 0x60d:     /* TINSR */
        if (((insn >> 6) & 3) == 3)
            return 1;
        rd = (insn >> 12) & 0xf;
        wrd = (insn >> 16) & 0xf;
        tmp = load_reg(s, rd);
        gen_op_iwmmxt_movq_M0_wRn(wrd);

        switch ((insn >> 6) & 3) {
        case 0:
            tmp2 = tcg_const_i32(0xff);
            tmp3 = tcg_const_i32((insn & 7) << 3);
            break;
        case 1:
            tmp2 = tcg_const_i32(0xffff);
            tmp3 = tcg_const_i32((insn & 3) << 4);
            break;
        case 2:
            tmp2 = tcg_const_i32(0xffffffff);
            tmp3 = tcg_const_i32((insn & 1) << 5);
            break;
        default:
            TCGV_UNUSED_I32(tmp2);
            TCGV_UNUSED_I32(tmp3);
        }
        gen_helper_iwmmxt_insr(cpu_M0, cpu_M0, tmp, tmp2, tmp3);
        tcg_temp_free_i32(tmp3);
        tcg_temp_free_i32(tmp2);
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x107: case 0x507: case 0x907: case 0xd07:     /* TEXTRM */
        rd = (insn >> 12) & 0xf;
        wrd = (insn >> 16) & 0xf;
        if (rd == 15 || ((insn >> 22) & 3) == 3)
            return 1;
        gen_op_iwmmxt_movq_M0_wRn(wrd);
        tmp = tcg_temp_new_i32();
        switch ((insn >> 22) & 3) {
        case 0:
            tcg_gen_shri_i64(cpu_M0, cpu_M0, (insn & 7) << 3);
            tcg_gen_extrl_i64_i32(tmp, cpu_M0);
            if (insn & 8) {
                tcg_gen_ext8s_i32(tmp, tmp);
            } else {
                tcg_gen_andi_i32(tmp, tmp, 0xff);
            }
            break;
        case 1:
            tcg_gen_shri_i64(cpu_M0, cpu_M0, (insn & 3) << 4);
            tcg_gen_extrl_i64_i32(tmp, cpu_M0);
            if (insn & 8) {
                tcg_gen_ext16s_i32(tmp, tmp);
            } else {
                tcg_gen_andi_i32(tmp, tmp, 0xffff);
            }
            break;
        case 2:
            tcg_gen_shri_i64(cpu_M0, cpu_M0, (insn & 1) << 5);
            tcg_gen_extrl_i64_i32(tmp, cpu_M0);
            break;
        }
        store_reg(s, rd, tmp);
        break;
    case 0x117: case 0x517: case 0x917: case 0xd17:     /* TEXTRC */
        if ((insn & 0x000ff008) != 0x0003f000 || ((insn >> 22) & 3) == 3)
            return 1;
        tmp = iwmmxt_load_creg(ARM_IWMMXT_wCASF);
        switch ((insn >> 22) & 3) {
        case 0:
            tcg_gen_shri_i32(tmp, tmp, ((insn & 7) << 2) + 0);
            break;
        case 1:
            tcg_gen_shri_i32(tmp, tmp, ((insn & 3) << 3) + 4);
            break;
        case 2:
            tcg_gen_shri_i32(tmp, tmp, ((insn & 1) << 4) + 12);
            break;
        }
        tcg_gen_shli_i32(tmp, tmp, 28);
        gen_set_nzcv(tmp);
        tcg_temp_free_i32(tmp);
        break;
    case 0x401: case 0x405: case 0x409: case 0x40d:     /* TBCST */
        if (((insn >> 6) & 3) == 3)
            return 1;
        rd = (insn >> 12) & 0xf;
        wrd = (insn >> 16) & 0xf;
        tmp = load_reg(s, rd);
        switch ((insn >> 6) & 3) {
        case 0:
            gen_helper_iwmmxt_bcstb(cpu_M0, tmp);
            break;
        case 1:
            gen_helper_iwmmxt_bcstw(cpu_M0, tmp);
            break;
        case 2:
            gen_helper_iwmmxt_bcstl(cpu_M0, tmp);
            break;
        }
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x113: case 0x513: case 0x913: case 0xd13:     /* TANDC */
        if ((insn & 0x000ff00f) != 0x0003f000 || ((insn >> 22) & 3) == 3)
            return 1;
        tmp = iwmmxt_load_creg(ARM_IWMMXT_wCASF);
        tmp2 = tcg_temp_new_i32();
        tcg_gen_mov_i32(tmp2, tmp);
        switch ((insn >> 22) & 3) {
        case 0:
            for (i = 0; i < 7; i ++) {
                tcg_gen_shli_i32(tmp2, tmp2, 4);
                tcg_gen_and_i32(tmp, tmp, tmp2);
            }
            break;
        case 1:
            for (i = 0; i < 3; i ++) {
                tcg_gen_shli_i32(tmp2, tmp2, 8);
                tcg_gen_and_i32(tmp, tmp, tmp2);
            }
            break;
        case 2:
            tcg_gen_shli_i32(tmp2, tmp2, 16);
            tcg_gen_and_i32(tmp, tmp, tmp2);
            break;
        }
        gen_set_nzcv(tmp);
        tcg_temp_free_i32(tmp2);
        tcg_temp_free_i32(tmp);
        break;
    case 0x01c: case 0x41c: case 0x81c: case 0xc1c:     /* WACC */
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            gen_helper_iwmmxt_addcb(cpu_M0, cpu_M0);
            break;
        case 1:
            gen_helper_iwmmxt_addcw(cpu_M0, cpu_M0);
            break;
        case 2:
            gen_helper_iwmmxt_addcl(cpu_M0, cpu_M0);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x115: case 0x515: case 0x915: case 0xd15:     /* TORC */
        if ((insn & 0x000ff00f) != 0x0003f000 || ((insn >> 22) & 3) == 3)
            return 1;
        tmp = iwmmxt_load_creg(ARM_IWMMXT_wCASF);
        tmp2 = tcg_temp_new_i32();
        tcg_gen_mov_i32(tmp2, tmp);
        switch ((insn >> 22) & 3) {
        case 0:
            for (i = 0; i < 7; i ++) {
                tcg_gen_shli_i32(tmp2, tmp2, 4);
                tcg_gen_or_i32(tmp, tmp, tmp2);
            }
            break;
        case 1:
            for (i = 0; i < 3; i ++) {
                tcg_gen_shli_i32(tmp2, tmp2, 8);
                tcg_gen_or_i32(tmp, tmp, tmp2);
            }
            break;
        case 2:
            tcg_gen_shli_i32(tmp2, tmp2, 16);
            tcg_gen_or_i32(tmp, tmp, tmp2);
            break;
        }
        gen_set_nzcv(tmp);
        tcg_temp_free_i32(tmp2);
        tcg_temp_free_i32(tmp);
        break;
    case 0x103: case 0x503: case 0x903: case 0xd03:     /* TMOVMSK */
        rd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        if ((insn & 0xf) != 0 || ((insn >> 22) & 3) == 3)
            return 1;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_temp_new_i32();
        switch ((insn >> 22) & 3) {
        case 0:
            gen_helper_iwmmxt_msbb(tmp, cpu_M0);
            break;
        case 1:
            gen_helper_iwmmxt_msbw(tmp, cpu_M0);
            break;
        case 2:
            gen_helper_iwmmxt_msbl(tmp, cpu_M0);
            break;
        }
        store_reg(s, rd, tmp);
        break;
    case 0x106: case 0x306: case 0x506: case 0x706:     /* WCMPGT */
    case 0x906: case 0xb06: case 0xd06: case 0xf06:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            if (insn & (1 << 21))
                gen_op_iwmmxt_cmpgtsb_M0_wRn(rd1);
            else
                gen_op_iwmmxt_cmpgtub_M0_wRn(rd1);
            break;
        case 1:
            if (insn & (1 << 21))
                gen_op_iwmmxt_cmpgtsw_M0_wRn(rd1);
            else
                gen_op_iwmmxt_cmpgtuw_M0_wRn(rd1);
            break;
        case 2:
            if (insn & (1 << 21))
                gen_op_iwmmxt_cmpgtsl_M0_wRn(rd1);
            else
                gen_op_iwmmxt_cmpgtul_M0_wRn(rd1);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x00e: case 0x20e: case 0x40e: case 0x60e:     /* WUNPCKEL */
    case 0x80e: case 0xa0e: case 0xc0e: case 0xe0e:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            if (insn & (1 << 21))
                gen_op_iwmmxt_unpacklsb_M0();
            else
                gen_op_iwmmxt_unpacklub_M0();
            break;
        case 1:
            if (insn & (1 << 21))
                gen_op_iwmmxt_unpacklsw_M0();
            else
                gen_op_iwmmxt_unpackluw_M0();
            break;
        case 2:
            if (insn & (1 << 21))
                gen_op_iwmmxt_unpacklsl_M0();
            else
                gen_op_iwmmxt_unpacklul_M0();
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x00c: case 0x20c: case 0x40c: case 0x60c:     /* WUNPCKEH */
    case 0x80c: case 0xa0c: case 0xc0c: case 0xe0c:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            if (insn & (1 << 21))
                gen_op_iwmmxt_unpackhsb_M0();
            else
                gen_op_iwmmxt_unpackhub_M0();
            break;
        case 1:
            if (insn & (1 << 21))
                gen_op_iwmmxt_unpackhsw_M0();
            else
                gen_op_iwmmxt_unpackhuw_M0();
            break;
        case 2:
            if (insn & (1 << 21))
                gen_op_iwmmxt_unpackhsl_M0();
            else
                gen_op_iwmmxt_unpackhul_M0();
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x204: case 0x604: case 0xa04: case 0xe04:     /* WSRL */
    case 0x214: case 0x614: case 0xa14: case 0xe14:
        if (((insn >> 22) & 3) == 0)
            return 1;
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_temp_new_i32();
        if (gen_iwmmxt_shift(insn, 0xff, tmp)) {
            tcg_temp_free_i32(tmp);
            return 1;
        }
        switch ((insn >> 22) & 3) {
        case 1:
            gen_helper_iwmmxt_srlw(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 2:
            gen_helper_iwmmxt_srll(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 3:
            gen_helper_iwmmxt_srlq(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        }
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x004: case 0x404: case 0x804: case 0xc04:     /* WSRA */
    case 0x014: case 0x414: case 0x814: case 0xc14:
        if (((insn >> 22) & 3) == 0)
            return 1;
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_temp_new_i32();
        if (gen_iwmmxt_shift(insn, 0xff, tmp)) {
            tcg_temp_free_i32(tmp);
            return 1;
        }
        switch ((insn >> 22) & 3) {
        case 1:
            gen_helper_iwmmxt_sraw(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 2:
            gen_helper_iwmmxt_sral(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 3:
            gen_helper_iwmmxt_sraq(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        }
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x104: case 0x504: case 0x904: case 0xd04:     /* WSLL */
    case 0x114: case 0x514: case 0x914: case 0xd14:
        if (((insn >> 22) & 3) == 0)
            return 1;
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_temp_new_i32();
        if (gen_iwmmxt_shift(insn, 0xff, tmp)) {
            tcg_temp_free_i32(tmp);
            return 1;
        }
        switch ((insn >> 22) & 3) {
        case 1:
            gen_helper_iwmmxt_sllw(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 2:
            gen_helper_iwmmxt_slll(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 3:
            gen_helper_iwmmxt_sllq(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        }
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x304: case 0x704: case 0xb04: case 0xf04:     /* WROR */
    case 0x314: case 0x714: case 0xb14: case 0xf14:
        if (((insn >> 22) & 3) == 0)
            return 1;
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_temp_new_i32();
        switch ((insn >> 22) & 3) {
        case 1:
            if (gen_iwmmxt_shift(insn, 0xf, tmp)) {
                tcg_temp_free_i32(tmp);
                return 1;
            }
            gen_helper_iwmmxt_rorw(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 2:
            if (gen_iwmmxt_shift(insn, 0x1f, tmp)) {
                tcg_temp_free_i32(tmp);
                return 1;
            }
            gen_helper_iwmmxt_rorl(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        case 3:
            if (gen_iwmmxt_shift(insn, 0x3f, tmp)) {
                tcg_temp_free_i32(tmp);
                return 1;
            }
            gen_helper_iwmmxt_rorq(cpu_M0, cpu_env, cpu_M0, tmp);
            break;
        }
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x116: case 0x316: case 0x516: case 0x716:     /* WMIN */
    case 0x916: case 0xb16: case 0xd16: case 0xf16:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            if (insn & (1 << 21))
                gen_op_iwmmxt_minsb_M0_wRn(rd1);
            else
                gen_op_iwmmxt_minub_M0_wRn(rd1);
            break;
        case 1:
            if (insn & (1 << 21))
                gen_op_iwmmxt_minsw_M0_wRn(rd1);
            else
                gen_op_iwmmxt_minuw_M0_wRn(rd1);
            break;
        case 2:
            if (insn & (1 << 21))
                gen_op_iwmmxt_minsl_M0_wRn(rd1);
            else
                gen_op_iwmmxt_minul_M0_wRn(rd1);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x016: case 0x216: case 0x416: case 0x616:     /* WMAX */
    case 0x816: case 0xa16: case 0xc16: case 0xe16:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 0:
            if (insn & (1 << 21))
                gen_op_iwmmxt_maxsb_M0_wRn(rd1);
            else
                gen_op_iwmmxt_maxub_M0_wRn(rd1);
            break;
        case 1:
            if (insn & (1 << 21))
                gen_op_iwmmxt_maxsw_M0_wRn(rd1);
            else
                gen_op_iwmmxt_maxuw_M0_wRn(rd1);
            break;
        case 2:
            if (insn & (1 << 21))
                gen_op_iwmmxt_maxsl_M0_wRn(rd1);
            else
                gen_op_iwmmxt_maxul_M0_wRn(rd1);
            break;
        case 3:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x002: case 0x102: case 0x202: case 0x302:     /* WALIGNI */
    case 0x402: case 0x502: case 0x602: case 0x702:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_const_i32((insn >> 20) & 3);
        iwmmxt_load_reg(cpu_V1, rd1);
        gen_helper_iwmmxt_align(cpu_M0, cpu_M0, cpu_V1, tmp);
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    case 0x01a: case 0x11a: case 0x21a: case 0x31a:     /* WSUB */
    case 0x41a: case 0x51a: case 0x61a: case 0x71a:
    case 0x81a: case 0x91a: case 0xa1a: case 0xb1a:
    case 0xc1a: case 0xd1a: case 0xe1a: case 0xf1a:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 20) & 0xf) {
        case 0x0:
            gen_op_iwmmxt_subnb_M0_wRn(rd1);
            break;
        case 0x1:
            gen_op_iwmmxt_subub_M0_wRn(rd1);
            break;
        case 0x3:
            gen_op_iwmmxt_subsb_M0_wRn(rd1);
            break;
        case 0x4:
            gen_op_iwmmxt_subnw_M0_wRn(rd1);
            break;
        case 0x5:
            gen_op_iwmmxt_subuw_M0_wRn(rd1);
            break;
        case 0x7:
            gen_op_iwmmxt_subsw_M0_wRn(rd1);
            break;
        case 0x8:
            gen_op_iwmmxt_subnl_M0_wRn(rd1);
            break;
        case 0x9:
            gen_op_iwmmxt_subul_M0_wRn(rd1);
            break;
        case 0xb:
            gen_op_iwmmxt_subsl_M0_wRn(rd1);
            break;
        default:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x01e: case 0x11e: case 0x21e: case 0x31e:     /* WSHUFH */
    case 0x41e: case 0x51e: case 0x61e: case 0x71e:
    case 0x81e: case 0x91e: case 0xa1e: case 0xb1e:
    case 0xc1e: case 0xd1e: case 0xe1e: case 0xf1e:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        tmp = tcg_const_i32(((insn >> 16) & 0xf0) | (insn & 0x0f));
        gen_helper_iwmmxt_shufh(cpu_M0, cpu_env, cpu_M0, tmp);
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x018: case 0x118: case 0x218: case 0x318:     /* WADD */
    case 0x418: case 0x518: case 0x618: case 0x718:
    case 0x818: case 0x918: case 0xa18: case 0xb18:
    case 0xc18: case 0xd18: case 0xe18: case 0xf18:
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 20) & 0xf) {
        case 0x0:
            gen_op_iwmmxt_addnb_M0_wRn(rd1);
            break;
        case 0x1:
            gen_op_iwmmxt_addub_M0_wRn(rd1);
            break;
        case 0x3:
            gen_op_iwmmxt_addsb_M0_wRn(rd1);
            break;
        case 0x4:
            gen_op_iwmmxt_addnw_M0_wRn(rd1);
            break;
        case 0x5:
            gen_op_iwmmxt_adduw_M0_wRn(rd1);
            break;
        case 0x7:
            gen_op_iwmmxt_addsw_M0_wRn(rd1);
            break;
        case 0x8:
            gen_op_iwmmxt_addnl_M0_wRn(rd1);
            break;
        case 0x9:
            gen_op_iwmmxt_addul_M0_wRn(rd1);
            break;
        case 0xb:
            gen_op_iwmmxt_addsl_M0_wRn(rd1);
            break;
        default:
            return 1;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x008: case 0x108: case 0x208: case 0x308:     /* WPACK */
    case 0x408: case 0x508: case 0x608: case 0x708:
    case 0x808: case 0x908: case 0xa08: case 0xb08:
    case 0xc08: case 0xd08: case 0xe08: case 0xf08:
        if (!(insn & (1 << 20)) || ((insn >> 22) & 3) == 0)
            return 1;
        wrd = (insn >> 12) & 0xf;
        rd0 = (insn >> 16) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        gen_op_iwmmxt_movq_M0_wRn(rd0);
        switch ((insn >> 22) & 3) {
        case 1:
            if (insn & (1 << 21))
                gen_op_iwmmxt_packsw_M0_wRn(rd1);
            else
                gen_op_iwmmxt_packuw_M0_wRn(rd1);
            break;
        case 2:
            if (insn & (1 << 21))
                gen_op_iwmmxt_packsl_M0_wRn(rd1);
            else
                gen_op_iwmmxt_packul_M0_wRn(rd1);
            break;
        case 3:
            if (insn & (1 << 21))
                gen_op_iwmmxt_packsq_M0_wRn(rd1);
            else
                gen_op_iwmmxt_packuq_M0_wRn(rd1);
            break;
        }
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        gen_op_iwmmxt_set_cup();
        break;
    case 0x201: case 0x203: case 0x205: case 0x207:
    case 0x209: case 0x20b: case 0x20d: case 0x20f:
    case 0x211: case 0x213: case 0x215: case 0x217:
    case 0x219: case 0x21b: case 0x21d: case 0x21f:
        wrd = (insn >> 5) & 0xf;
        rd0 = (insn >> 12) & 0xf;
        rd1 = (insn >> 0) & 0xf;
        if (rd0 == 0xf || rd1 == 0xf)
            return 1;
        gen_op_iwmmxt_movq_M0_wRn(wrd);
        tmp = load_reg(s, rd0);
        tmp2 = load_reg(s, rd1);
        switch ((insn >> 16) & 0xf) {
        case 0x0:                                       /* TMIA */
            gen_helper_iwmmxt_muladdsl(cpu_M0, cpu_M0, tmp, tmp2);
            break;
        case 0x8:                                       /* TMIAPH */
            gen_helper_iwmmxt_muladdsw(cpu_M0, cpu_M0, tmp, tmp2);
            break;
        case 0xc: case 0xd: case 0xe: case 0xf:         /* TMIAxy */
            if (insn & (1 << 16))
                tcg_gen_shri_i32(tmp, tmp, 16);
            if (insn & (1 << 17))
                tcg_gen_shri_i32(tmp2, tmp2, 16);
            gen_helper_iwmmxt_muladdswl(cpu_M0, cpu_M0, tmp, tmp2);
            break;
        default:
            tcg_temp_free_i32(tmp2);
            tcg_temp_free_i32(tmp);
            return 1;
        }
        tcg_temp_free_i32(tmp2);
        tcg_temp_free_i32(tmp);
        gen_op_iwmmxt_movq_wRn_M0(wrd);
        gen_op_iwmmxt_set_mup();
        break;
    default:
        return 1;
    }

    return 0;
}

/* Disassemble an XScale DSP instruction.  Returns nonzero if an error occurred
   (ie. an undefined instruction).  */
static int disas_dsp_insn(DisasContext *s, uint32_t insn)
{
    int acc, rd0, rd1, rdhi, rdlo;
    TCGv_i32 tmp, tmp2;

    if ((insn & 0x0ff00f10) == 0x0e200010) {
        /* Multiply with Internal Accumulate Format */
        rd0 = (insn >> 12) & 0xf;
        rd1 = insn & 0xf;
        acc = (insn >> 5) & 7;

        if (acc != 0)
            return 1;

        tmp = load_reg(s, rd0);
        tmp2 = load_reg(s, rd1);
        switch ((insn >> 16) & 0xf) {
        case 0x0:                                       /* MIA */
            gen_helper_iwmmxt_muladdsl(cpu_M0, cpu_M0, tmp, tmp2);
            break;
        case 0x8:                                       /* MIAPH */
            gen_helper_iwmmxt_muladdsw(cpu_M0, cpu_M0, tmp, tmp2);
            break;
        case 0xc:                                       /* MIABB */
        case 0xd:                                       /* MIABT */
        case 0xe:                                       /* MIATB */
        case 0xf:                                       /* MIATT */
            if (insn & (1 << 16))
                tcg_gen_shri_i32(tmp, tmp, 16);
            if (insn & (1 << 17))
                tcg_gen_shri_i32(tmp2, tmp2, 16);
            gen_helper_iwmmxt_muladdswl(cpu_M0, cpu_M0, tmp, tmp2);
            break;
        default:
            return 1;
        }
        tcg_temp_free_i32(tmp2);
        tcg_temp_free_i32(tmp);

        gen_op_iwmmxt_movq_wRn_M0(acc);
        return 0;
    }

    if ((insn & 0x0fe00ff8) == 0x0c400000) {
        /* Internal Accumulator Access Format */
        rdhi = (insn >> 16) & 0xf;
        rdlo = (insn >> 12) & 0xf;
        acc = insn & 7;

        if (acc != 0)
            return 1;

        if (insn & ARM_CP_RW_BIT) {                     /* MRA */
            iwmmxt_load_reg(cpu_V0, acc);
            tcg_gen_extrl_i64_i32(cpu_R[rdlo], cpu_V0);
            tcg_gen_shri_i64(cpu_V0, cpu_V0, 32);
            tcg_gen_extrl_i64_i32(cpu_R[rdhi], cpu_V0);
            tcg_gen_andi_i32(cpu_R[rdhi], cpu_R[rdhi], (1 << (40 - 32)) - 1);
        } else {                                        /* MAR */
            tcg_gen_concat_i32_i64(cpu_V0, cpu_R[rdlo], cpu_R[rdhi]);
            iwmmxt_store_reg(cpu_V0, acc);
        }
        return 0;
    }

    return 1;
}

#define VFP_REG_SHR(x, n) (((n) > 0) ? (x) >> (n) : (x) << -(n))
#define VFP_SREG(insn, bigbit, smallbit) \
  ((VFP_REG_SHR(insn, bigbit - 1) & 0x1e) | (((insn) >> (smallbit)) & 1))
#define VFP_DREG(reg, insn, bigbit, smallbit) do { \
    if (arm_dc_feature(s, ARM_FEATURE_VFP3)) { \
        reg = (((insn) >> (bigbit)) & 0x0f) \
              | (((insn) >> ((smallbit) - 4)) & 0x10); \
    } else { \
        if (insn & (1 << (smallbit))) \
            return 1; \
        reg = ((insn) >> (bigbit)) & 0x0f; \
    }} while (0)

#define VFP_SREG_D(insn) VFP_SREG(insn, 12, 22)
#define VFP_DREG_D(reg, insn) VFP_DREG(reg, insn, 12, 22)
#define VFP_SREG_N(insn) VFP_SREG(insn, 16,  7)
#define VFP_DREG_N(reg, insn) VFP_DREG(reg, insn, 16,  7)
#define VFP_SREG_M(insn) VFP_SREG(insn,  0,  5)
#define VFP_DREG_M(reg, insn) VFP_DREG(reg, insn,  0,  5)

/* Move between integer and VFP cores.  */
static TCGv_i32 gen_vfp_mrs(void)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_mov_i32(tmp, cpu_F0s);
    return tmp;
}

static void gen_vfp_msr(TCGv_i32 tmp)
{
    tcg_gen_mov_i32(cpu_F0s, tmp);
    tcg_temp_free_i32(tmp);
}

static void gen_neon_dup_u8(TCGv_i32 var, int shift)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    if (shift)
        tcg_gen_shri_i32(var, var, shift);
    tcg_gen_ext8u_i32(var, var);
    tcg_gen_shli_i32(tmp, var, 8);
    tcg_gen_or_i32(var, var, tmp);
    tcg_gen_shli_i32(tmp, var, 16);
    tcg_gen_or_i32(var, var, tmp);
    tcg_temp_free_i32(tmp);
}

static void gen_neon_dup_low16(TCGv_i32 var)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_ext16u_i32(var, var);
    tcg_gen_shli_i32(tmp, var, 16);
    tcg_gen_or_i32(var, var, tmp);
    tcg_temp_free_i32(tmp);
}

static void gen_neon_dup_high16(TCGv_i32 var)
{
    TCGv_i32 tmp = tcg_temp_new_i32();
    tcg_gen_andi_i32(var, var, 0xffff0000);
    tcg_gen_shri_i32(tmp, var, 16);
    tcg_gen_or_i32(var, var, tmp);
    tcg_temp_free_i32(tmp);
}

static TCGv_i32 gen_load_and_replicate(DisasContext *s, TCGv_i32 addr, int size)
{
    /* Load a single Neon element and replicate into a 32 bit TCG reg */
    TCGv_i32 tmp = tcg_temp_new_i32();
    switch (size) {
    case 0:
        gen_aa32_ld8u(s, tmp, addr, get_mem_index(s));
        gen_neon_dup_u8(tmp, 0);
        break;
    case 1:
        gen_aa32_ld16u(s, tmp, addr, get_mem_index(s));
        gen_neon_dup_low16(tmp);
        break;
    case 2:
        gen_aa32_ld32u(s, tmp, addr, get_mem_index(s));
        break;
    default: /* Avoid compiler warnings.  */
        abort();
    }
    return tmp;
}

static int handle_vsel(uint32_t insn, uint32_t rd, uint32_t rn, uint32_t rm,
                       uint32_t dp)
{
    uint32_t cc = extract32(insn, 20, 2);

    if (dp) {
        TCGv_i64 frn, frm, dest;
        TCGv_i64 tmp, zero, zf, nf, vf;

        zero = tcg_const_i64(0);

        frn = tcg_temp_new_i64();
        frm = tcg_temp_new_i64();
        dest = tcg_temp_new_i64();

        zf = tcg_temp_new_i64();
        nf = tcg_temp_new_i64();
        vf = tcg_temp_new_i64();

        tcg_gen_extu_i32_i64(zf, cpu_ZF);
        tcg_gen_ext_i32_i64(nf, cpu_NF);
        tcg_gen_ext_i32_i64(vf, cpu_VF);

        tcg_gen_ld_f64(frn, cpu_env, vfp_reg_offset(dp, rn));
        tcg_gen_ld_f64(frm, cpu_env, vfp_reg_offset(dp, rm));
        switch (cc) {
        case 0: /* eq: Z */
            tcg_gen_movcond_i64(TCG_COND_EQ, dest, zf, zero,
                                frn, frm);
            break;
        case 1: /* vs: V */
            tcg_gen_movcond_i64(TCG_COND_LT, dest, vf, zero,
                                frn, frm);
            break;
        case 2: /* ge: N == V -> N ^ V == 0 */
            tmp = tcg_temp_new_i64();
            tcg_gen_xor_i64(tmp, vf, nf);
            tcg_gen_movcond_i64(TCG_COND_GE, dest, tmp, zero,
                                frn, frm);
            tcg_temp_free_i64(tmp);
            break;
        case 3: /* gt: !Z && N == V */
            tcg_gen_movcond_i64(TCG_COND_NE, dest, zf, zero,
                                frn, frm);
            tmp = tcg_temp_new_i64();
            tcg_gen_xor_i64(tmp, vf, nf);
            tcg_gen_movcond_i64(TCG_COND_GE, dest, tmp, zero,
                                dest, frm);
            tcg_temp_free_i64(tmp);
            break;
        }
        tcg_gen_st_f64(dest, cpu_env, vfp_reg_offset(dp, rd));
        tcg_temp_free_i64(frn);
        tcg_temp_free_i64(frm);
        tcg_temp_free_i64(dest);

        tcg_temp_free_i64(zf);
        tcg_temp_free_i64(nf);
        tcg_temp_free_i64(vf);

        tcg_temp_free_i64(zero);
    } else {
        TCGv_i32 frn, frm, dest;
        TCGv_i32 tmp, zero;

        zero = tcg_const_i32(0);

        frn = tcg_temp_new_i32();
        frm = tcg_temp_new_i32();
        dest = tcg_temp_new_i32();
        tcg_gen_ld_f32(frn, cpu_env, vfp_reg_offset(dp, rn));
        tcg_gen_ld_f32(frm, cpu_env, vfp_reg_offset(dp, rm));
        switch (cc) {
        case 0: /* eq: Z */
            tcg_gen_movcond_i32(TCG_COND_EQ, dest, cpu_ZF, zero,
                                frn, frm);
            break;
        case 1: /* vs: V */
            tcg_gen_movcond_i32(TCG_COND_LT, dest, cpu_VF, zero,
                                frn, frm);
            break;
        case 2: /* ge: N == V -> N ^ V == 0 */
            tmp = tcg_temp_new_i32();
            tcg_gen_xor_i32(tmp, cpu_VF, cpu_NF);
            tcg_gen_movcond_i32(TCG_COND_GE, dest, tmp, zero,
                                frn, frm);
            tcg_temp_free_i32(tmp);
            break;
        case 3: /* gt: !Z && N == V */
            tcg_gen_movcond_i32(TCG_COND_NE, dest, cpu_ZF, zero,
                                frn, frm);
            tmp = tcg_temp_new_i32();
            tcg_gen_xor_i32(tmp, cpu_VF, cpu_NF);
            tcg_gen_movcond_i32(TCG_COND_GE, dest, tmp, zero,
                                dest, frm);
            tcg_temp_free_i32(tmp);
            break;
        }
        tcg_gen_st_f32(dest, cpu_env, vfp_reg_offset(dp, rd));
        tcg_temp_free_i32(frn);
        tcg_temp_free_i32(frm);
        tcg_temp_free_i32(dest);

        tcg_temp_free_i32(zero);
    }

    return 0;
}

static int handle_vminmaxnm(uint32_t insn, uint32_t rd, uint32_t rn,
                            uint32_t rm, uint32_t dp)
{
    uint32_t vmin = extract32(insn, 6, 1);
    TCGv_ptr fpst = get_fpstatus_ptr(0);

    if (dp) {
        TCGv_i64 frn, frm, dest;

        frn = tcg_temp_new_i64();
        frm = tcg_temp_new_i64();
        dest = tcg_temp_new_i64();

        tcg_gen_ld_f64(frn, cpu_env, vfp_reg_offset(dp, rn));
        tcg_gen_ld_f64(frm, cpu_env, vfp_reg_offset(dp, rm));
        if (vmin) {
            gen_helper_vfp_minnumd(dest, frn, frm, fpst);
        } else {
            gen_helper_vfp_maxnumd(dest, frn, frm, fpst);
        }
        tcg_gen_st_f64(dest, cpu_env, vfp_reg_offset(dp, rd));
        tcg_temp_free_i64(frn);
        tcg_temp_free_i64(frm);
        tcg_temp_free_i64(dest);
    } else {
        TCGv_i32 frn, frm, dest;

        frn = tcg_temp_new_i32();
        frm = tcg_temp_new_i32();
        dest = tcg_temp_new_i32();

        tcg_gen_ld_f32(frn, cpu_env, vfp_reg_offset(dp, rn));
        tcg_gen_ld_f32(frm, cpu_env, vfp_reg_offset(dp, rm));
        if (vmin) {
            gen_helper_vfp_minnums(dest, frn, frm, fpst);
        } else {
            gen_helper_vfp_maxnums(dest, frn, frm, fpst);
        }
        tcg_gen_st_f32(dest, cpu_env, vfp_reg_offset(dp, rd));
        tcg_temp_free_i32(frn);
        tcg_temp_free_i32(frm);
        tcg_temp_free_i32(dest);
    }

    tcg_temp_free_ptr(fpst);
    return 0;
}

static int handle_vrint(uint32_t insn, uint32_t rd, uint32_t rm, uint32_t dp,
                        int rounding)
{
    TCGv_ptr fpst = get_fpstatus_ptr(0);
    TCGv_i32 tcg_rmode;

    tcg_rmode = tcg_const_i32(arm_rmode_to_sf(rounding));
    gen_helper_set_rmode(tcg_rmode, tcg_rmode, cpu_env);

    if (dp) {
        TCGv_i64 tcg_op;
        TCGv_i64 tcg_res;
        tcg_op = tcg_temp_new_i64();
        tcg_res = tcg_temp_new_i64();
        tcg_gen_ld_f64(tcg_op, cpu_env, vfp_reg_offset(dp, rm));
        gen_helper_rintd(tcg_res, tcg_op, fpst);
        tcg_gen_st_f64(tcg_res, cpu_env, vfp_reg_offset(dp, rd));
        tcg_temp_free_i64(tcg_op);
        tcg_temp_free_i64(tcg_res);
    } else {
        TCGv_i32 tcg_op;
        TCGv_i32 tcg_res;
        tcg_op = tcg_temp_new_i32();
        tcg_res = tcg_temp_new_i32();
        tcg_gen_ld_f32(tcg_op, cpu_env, vfp_reg_offset(dp, rm));
        gen_helper_rints(tcg_res, tcg_op, fpst);
        tcg_gen_st_f32(tcg_res, cpu_env, vfp_reg_offset(dp, rd));