/*
   SPARC translation

   Copyright (C) 2003 Thomas M. Ogrisegg <tom@fnord.at>
   Copyright (C) 2003-2005 Fabrice Bellard

   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 "disas/disas.h"
#include "exec/helper-proto.h"
#include "exec/exec-all.h"
#include "tcg-op.h"
#include "exec/cpu_ldst.h"

#include "exec/helper-gen.h"

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


#define DEBUG_DISAS

#define DYNAMIC_PC  1 /* dynamic pc value */
#define JUMP_PC     2 /* dynamic pc value which takes only two values
                         according to jump_pc[T2] */

#define DISAS_EXIT  DISAS_TARGET_0

/* global register indexes */
static TCGv_ptr cpu_regwptr;
static TCGv cpu_cc_src, cpu_cc_src2, cpu_cc_dst;
static TCGv_i32 cpu_cc_op;
static TCGv_i32 cpu_psr;
static TCGv cpu_fsr, cpu_pc, cpu_npc;
static TCGv cpu_regs[32];
static TCGv cpu_y;
#ifndef CONFIG_USER_ONLY
static TCGv cpu_tbr;
#endif
static TCGv cpu_cond;
#ifdef TARGET_SPARC64
static TCGv_i32 cpu_xcc, cpu_fprs;
static TCGv cpu_gsr;
static TCGv cpu_tick_cmpr, cpu_stick_cmpr, cpu_hstick_cmpr;
static TCGv cpu_hintp, cpu_htba, cpu_hver, cpu_ssr, cpu_ver;
#else
static TCGv cpu_wim;
#endif
/* Floating point registers */
static TCGv_i64 cpu_fpr[TARGET_DPREGS];

#include "exec/gen-icount.h"

typedef struct DisasContext {
    DisasContextBase base;
    target_ulong pc;    /* current Program Counter: integer or DYNAMIC_PC */
    target_ulong npc;   /* next PC: integer or DYNAMIC_PC or JUMP_PC */
    target_ulong jump_pc[2]; /* used when JUMP_PC pc value is used */
    int mem_idx;
    bool fpu_enabled;
    bool address_mask_32bit;
#ifndef CONFIG_USER_ONLY
    bool supervisor;
#ifdef TARGET_SPARC64
    bool hypervisor;
#endif
#endif

    uint32_t cc_op;  /* current CC operation */
    sparc_def_t *def;
    TCGv_i32 t32[3];
    TCGv ttl[5];
    int n_t32;
    int n_ttl;
#ifdef TARGET_SPARC64
    int fprs_dirty;
    int asi;
#endif
} DisasContext;

typedef struct {
    TCGCond cond;
    bool is_bool;
    bool g1, g2;
    TCGv c1, c2;
} DisasCompare;

// This function uses non-native bit order
#define GET_FIELD(X, FROM, TO)                                  \
    ((X) >> (31 - (TO)) & ((1 << ((TO) - (FROM) + 1)) - 1))

// This function uses the order in the manuals, i.e. bit 0 is 2^0
#define GET_FIELD_SP(X, FROM, TO)               \
    GET_FIELD(X, 31 - (TO), 31 - (FROM))

#define GET_FIELDs(x,a,b) sign_extend (GET_FIELD(x,a,b), (b) - (a) + 1)
#define GET_FIELD_SPs(x,a,b) sign_extend (GET_FIELD_SP(x,a,b), ((b) - (a) + 1))

#ifdef TARGET_SPARC64
#define DFPREG(r) (((r & 1) << 5) | (r & 0x1e))
#define QFPREG(r) (((r & 1) << 5) | (r & 0x1c))
#else
#define DFPREG(r) (r & 0x1e)
#define QFPREG(r) (r & 0x1c)
#endif

#define UA2005_HTRAP_MASK 0xff
#define V8_TRAP_MASK 0x7f

static int sign_extend(int x, int len)
{
    len = 32 - len;
    return (x << len) >> len;
}

#define IS_IMM (insn & (1<<13))

static inline TCGv_i32 get_temp_i32(DisasContext *dc)
{
    TCGv_i32 t;
    assert(dc->n_t32 < ARRAY_SIZE(dc->t32));
    dc->t32[dc->n_t32++] = t = tcg_temp_new_i32();
    return t;
}

static inline TCGv get_temp_tl(DisasContext *dc)
{
    TCGv t;
    assert(dc->n_ttl < ARRAY_SIZE(dc->ttl));
    dc->ttl[dc->n_ttl++] = t = tcg_temp_new();
    return t;
}

static inline void gen_update_fprs_dirty(DisasContext *dc, int rd)
{
#if defined(TARGET_SPARC64)
    int bit = (rd < 32) ? 1 : 2;
    /* If we know we've already set this bit within the TB,
       we can avoid setting it again.  */
    if (!(dc->fprs_dirty & bit)) {
        dc->fprs_dirty |= bit;
        tcg_gen_ori_i32(cpu_fprs, cpu_fprs, bit);
    }
#endif
}

/* floating point registers moves */
static TCGv_i32 gen_load_fpr_F(DisasContext *dc, unsigned int src)
{
#if TCG_TARGET_REG_BITS == 32
    if (src & 1) {
        return TCGV_LOW(cpu_fpr[src / 2]);
    } else {
        return TCGV_HIGH(cpu_fpr[src / 2]);
    }
#else
    TCGv_i32 ret = get_temp_i32(dc);
    if (src & 1) {
        tcg_gen_extrl_i64_i32(ret, cpu_fpr[src / 2]);
    } else {
        tcg_gen_extrh_i64_i32(ret, cpu_fpr[src / 2]);
    }
    return ret;
#endif
}

static void gen_store_fpr_F(DisasContext *dc, unsigned int dst, TCGv_i32 v)
{
#if TCG_TARGET_REG_BITS == 32
    if (dst & 1) {
        tcg_gen_mov_i32(TCGV_LOW(cpu_fpr[dst / 2]), v);
    } else {
        tcg_gen_mov_i32(TCGV_HIGH(cpu_fpr[dst / 2]), v);
    }
#else
    TCGv_i64 t = (TCGv_i64)v;
    tcg_gen_deposit_i64(cpu_fpr[dst / 2], cpu_fpr[dst / 2], t,
                        (dst & 1 ? 0 : 32), 32);
#endif
    gen_update_fprs_dirty(dc, dst);
}

static TCGv_i32 gen_dest_fpr_F(DisasContext *dc)
{
    return get_temp_i32(dc);
}

static TCGv_i64 gen_load_fpr_D(DisasContext *dc, unsigned int src)
{
    src = DFPREG(src);
    return cpu_fpr[src / 2];
}

static void gen_store_fpr_D(DisasContext *dc, unsigned int dst, TCGv_i64 v)
{
    dst = DFPREG(dst);
    tcg_gen_mov_i64(cpu_fpr[dst / 2], v);
    gen_update_fprs_dirty(dc, dst);
}

static TCGv_i64 gen_dest_fpr_D(DisasContext *dc, unsigned int dst)
{
    return cpu_fpr[DFPREG(dst) / 2];
}

static void gen_op_load_fpr_QT0(unsigned int src)
{
    tcg_gen_st_i64(cpu_fpr[src / 2], cpu_env, offsetof(CPUSPARCState, qt0) +
                   offsetof(CPU_QuadU, ll.upper));
    tcg_gen_st_i64(cpu_fpr[src/2 + 1], cpu_env, offsetof(CPUSPARCState, qt0) +
                   offsetof(CPU_QuadU, ll.lower));
}

static void gen_op_load_fpr_QT1(unsigned int src)
{
    tcg_gen_st_i64(cpu_fpr[src / 2], cpu_env, offsetof(CPUSPARCState, qt1) +
                   offsetof(CPU_QuadU, ll.upper));
    tcg_gen_st_i64(cpu_fpr[src/2 + 1], cpu_env, offsetof(CPUSPARCState, qt1) +
                   offsetof(CPU_QuadU, ll.lower));
}

static void gen_op_store_QT0_fpr(unsigned int dst)
{
    tcg_gen_ld_i64(cpu_fpr[dst / 2], cpu_env, offsetof(CPUSPARCState, qt0) +
                   offsetof(CPU_QuadU, ll.upper));
    tcg_gen_ld_i64(cpu_fpr[dst/2 + 1], cpu_env, offsetof(CPUSPARCState, qt0) +
                   offsetof(CPU_QuadU, ll.lower));
}

static void gen_store_fpr_Q(DisasContext *dc, unsigned int dst,
                            TCGv_i64 v1, TCGv_i64 v2)
{
    dst = QFPREG(dst);

    tcg_gen_mov_i64(cpu_fpr[dst / 2], v1);
    tcg_gen_mov_i64(cpu_fpr[dst / 2 + 1], v2);
    gen_update_fprs_dirty(dc, dst);
}

#ifdef TARGET_SPARC64
static TCGv_i64 gen_load_fpr_Q0(DisasContext *dc, unsigned int src)
{
    src = QFPREG(src);
    return cpu_fpr[src / 2];
}

static TCGv_i64 gen_load_fpr_Q1(DisasContext *dc, unsigned int src)
{
    src = QFPREG(src);
    return cpu_fpr[src / 2 + 1];
}

static void gen_move_Q(DisasContext *dc, unsigned int rd, unsigned int rs)
{
    rd = QFPREG(rd);
    rs = QFPREG(rs);

    tcg_gen_mov_i64(cpu_fpr[rd / 2], cpu_fpr[rs / 2]);
    tcg_gen_mov_i64(cpu_fpr[rd / 2 + 1], cpu_fpr[rs / 2 + 1]);
    gen_update_fprs_dirty(dc, rd);
}
#endif

/* moves */
#ifdef CONFIG_USER_ONLY
#define supervisor(dc) 0
#ifdef TARGET_SPARC64
#define hypervisor(dc) 0
#endif
#else
#ifdef TARGET_SPARC64
#define hypervisor(dc) (dc->hypervisor)
#define supervisor(dc) (dc->supervisor | dc->hypervisor)
#else
#define supervisor(dc) (dc->supervisor)
#endif
#endif

#ifdef TARGET_SPARC64
#ifndef TARGET_ABI32
#define AM_CHECK(dc) ((dc)->address_mask_32bit)
#else
#define AM_CHECK(dc) (1)
#endif
#endif

static inline void gen_address_mask(DisasContext *dc, TCGv addr)
{
#ifdef TARGET_SPARC64
    if (AM_CHECK(dc))
        tcg_gen_andi_tl(addr, addr, 0xffffffffULL);
#endif
}

static inline TCGv gen_load_gpr(DisasContext *dc, int reg)
{
    if (reg > 0) {
        assert(reg < 32);
        return cpu_regs[reg];
    } else {
        TCGv t = get_temp_tl(dc);
        tcg_gen_movi_tl(t, 0);
        return t;
    }
}

static inline void gen_store_gpr(DisasContext *dc, int reg, TCGv v)
{
    if (reg > 0) {
        assert(reg < 32);
        tcg_gen_mov_tl(cpu_regs[reg], v);
    }
}

static inline TCGv gen_dest_gpr(DisasContext *dc, int reg)
{
    if (reg > 0) {
        assert(reg < 32);
        return cpu_regs[reg];
    } else {
        return get_temp_tl(dc);
    }
}

static inline bool use_goto_tb(DisasContext *s, target_ulong pc,
                               target_ulong npc)
{
    if (unlikely(s->base.singlestep_enabled || singlestep)) {
        return false;
    }

#ifndef CONFIG_USER_ONLY
    return (pc & TARGET_PAGE_MASK) == (s->base.tb->pc & TARGET_PAGE_MASK) &&
           (npc & TARGET_PAGE_MASK) == (s->base.tb->pc & TARGET_PAGE_MASK);
#else
    return true;
#endif
}

static inline void gen_goto_tb(DisasContext *s, int tb_num,
                               target_ulong pc, target_ulong npc)
{
    if (use_goto_tb(s, pc, npc))  {
        /* jump to same page: we can use a direct jump */
        tcg_gen_goto_tb(tb_num);
        tcg_gen_movi_tl(cpu_pc, pc);
        tcg_gen_movi_tl(cpu_npc, npc);
        tcg_gen_exit_tb(s->base.tb, tb_num);
    } else {
        /* jump to another page: currently not optimized */
        tcg_gen_movi_tl(cpu_pc, pc);
        tcg_gen_movi_tl(cpu_npc, npc);
        tcg_gen_exit_tb(NULL, 0);
    }
}

// XXX suboptimal
static inline void gen_mov_reg_N(TCGv reg, TCGv_i32 src)
{
    tcg_gen_extu_i32_tl(reg, src);
    tcg_gen_extract_tl(reg, reg, PSR_NEG_SHIFT, 1);
}

static inline void gen_mov_reg_Z(TCGv reg, TCGv_i32 src)
{
    tcg_gen_extu_i32_tl(reg, src);
    tcg_gen_extract_tl(reg, reg, PSR_ZERO_SHIFT, 1);
}

static inline void gen_mov_reg_V(TCGv reg, TCGv_i32 src)
{
    tcg_gen_extu_i32_tl(reg, src);
    tcg_gen_extract_tl(reg, reg, PSR_OVF_SHIFT, 1);
}

static inline void gen_mov_reg_C(TCGv reg, TCGv_i32 src)
{
    tcg_gen_extu_i32_tl(reg, src);
    tcg_gen_extract_tl(reg, reg, PSR_CARRY_SHIFT, 1);
}

static inline void gen_op_add_cc(TCGv dst, TCGv src1, TCGv src2)
{
    tcg_gen_mov_tl(cpu_cc_src, src1);
    tcg_gen_mov_tl(cpu_cc_src2, src2);
    tcg_gen_add_tl(cpu_cc_dst, cpu_cc_src, cpu_cc_src2);
    tcg_gen_mov_tl(dst, cpu_cc_dst);
}

static TCGv_i32 gen_add32_carry32(void)
{
    TCGv_i32 carry_32, cc_src1_32, cc_src2_32;

    /* Carry is computed from a previous add: (dst < src)  */
#if TARGET_LONG_BITS == 64
    cc_src1_32 = tcg_temp_new_i32();
    cc_src2_32 = tcg_temp_new_i32();
    tcg_gen_extrl_i64_i32(cc_src1_32, cpu_cc_dst);
    tcg_gen_extrl_i64_i32(cc_src2_32, cpu_cc_src);
#else
    cc_src1_32 = cpu_cc_dst;
    cc_src2_32 = cpu_cc_src;
#endif

    carry_32 = tcg_temp_new_i32();
    tcg_gen_setcond_i32(TCG_COND_LTU, carry_32, cc_src1_32, cc_src2_32);

#if TARGET_LONG_BITS == 64
    tcg_temp_free_i32(cc_src1_32);
    tcg_temp_free_i32(cc_src2_32);
#endif

    return carry_32;
}

static TCGv_i32 gen_sub32_carry32(void)
{
    TCGv_i32 carry_32, cc_src1_32, cc_src2_32;

    /* Carry is computed from a previous borrow: (src1 < src2)  */
#if TARGET_LONG_BITS == 64
    cc_src1_32 = tcg_temp_new_i32();
    cc_src2_32 = tcg_temp_new_i32();
    tcg_gen_extrl_i64_i32(cc_src1_32, cpu_cc_src);
    tcg_gen_extrl_i64_i32(cc_src2_32, cpu_cc_src2);
#else
    cc_src1_32 = cpu_cc_src;
    cc_src2_32 = cpu_cc_src2;
#endif

    carry_32 = tcg_temp_new_i32();
    tcg_gen_setcond_i32(TCG_COND_LTU, carry_32, cc_src1_32, cc_src2_32);

#if TARGET_LONG_BITS == 64
    tcg_temp_free_i32(cc_src1_32);
    tcg_temp_free_i32(cc_src2_32);
#endif

    return carry_32;
}

static void gen_op_addx_int(DisasContext *dc, TCGv dst, TCGv src1,
                            TCGv src2, int update_cc)
{
    TCGv_i32 carry_32;
    TCGv carry;

    switch (dc->cc_op) {
    case CC_OP_DIV:
    case CC_OP_LOGIC:
        /* Carry is known to be zero.  Fall back to plain ADD.  */
        if (update_cc) {
            gen_op_add_cc(dst, src1, src2);
        } else {
            tcg_gen_add_tl(dst, src1, src2);
        }
        return;

    case CC_OP_ADD:
    case CC_OP_TADD:
    case CC_OP_TADDTV:
        if (TARGET_LONG_BITS == 32) {
            /* We can re-use the host's hardware carry generation by using
               an ADD2 opcode.  We discard the low part of the output.
               Ideally we'd combine this operation with the add that
               generated the carry in the first place.  */
            carry = tcg_temp_new();
            tcg_gen_add2_tl(carry, dst, cpu_cc_src, src1, cpu_cc_src2, src2);
            tcg_temp_free(carry);
            goto add_done;
        }
        carry_32 = gen_add32_carry32();
        break;

    case CC_OP_SUB:
    case CC_OP_TSUB:
    case CC_OP_TSUBTV:
        carry_32 = gen_sub32_carry32();
        break;

    default:
        /* We need external help to produce the carry.  */
        carry_32 = tcg_temp_new_i32();
        gen_helper_compute_C_icc(carry_32, cpu_env);
        break;
    }

#if TARGET_LONG_BITS == 64
    carry = tcg_temp_new();
    tcg_gen_extu_i32_i64(carry, carry_32);
#else
    carry = carry_32;
#endif

    tcg_gen_add_tl(dst, src1, src2);
    tcg_gen_add_tl(dst, dst, carry);

    tcg_temp_free_i32(carry_32);
#if TARGET_LONG_BITS == 64
    tcg_temp_free(carry);
#endif

 add_done:
    if (update_cc) {
        tcg_gen_mov_tl(cpu_cc_src, src1);
        tcg_gen_mov_tl(cpu_cc_src2, src2);
        tcg_gen_mov_tl(cpu_cc_dst, dst);
        tcg_gen_movi_i32(cpu_cc_op, CC_OP_ADDX);
        dc->cc_op = CC_OP_ADDX;
    }
}

static inline void gen_op_sub_cc(TCGv dst, TCGv src1, TCGv src2)
{
    tcg_gen_mov_tl(cpu_cc_src, src1);
    tcg_gen_mov_tl(cpu_cc_src2, src2);
    tcg_gen_sub_tl(cpu_cc_dst, cpu_cc_src, cpu_cc_src2);
    tcg_gen_mov_tl(dst, cpu_cc_dst);
}

static void gen_op_subx_int(DisasContext *dc, TCGv dst, TCGv src1,
                            TCGv src2, int update_cc)
{
    TCGv_i32 carry_32;
    TCGv carry;

    switch (dc->cc_op) {
    case CC_OP_DIV:
    case CC_OP_LOGIC:
        /* Carry is known to be zero.  Fall back to plain SUB.  */
        if (update_cc) {
            gen_op_sub_cc(dst, src1, src2);
        } else {
            tcg_gen_sub_tl(dst, src1, src2);
        }
        return;

    case CC_OP_ADD:
    case CC_OP_TADD:
    case CC_OP_TADDTV:
        carry_32 = gen_add32_carry32();
        break;

    case CC_OP_SUB:
    case CC_OP_TSUB:
    case CC_OP_TSUBTV:
        if (TARGET_LONG_BITS == 32) {
            /* We can re-use the host's hardware carry generation by using
               a SUB2 opcode.  We discard the low part of the output.
               Ideally we'd combine this operation with the add that
               generated the carry in the first place.  */
            carry = tcg_temp_new();
            tcg_gen_sub2_tl(carry, dst, cpu_cc_src, src1, cpu_cc_src2, src2);
            tcg_temp_free(carry);
            goto sub_done;
        }
        carry_32 = gen_sub32_carry32();
        break;

    default:
        /* We need external help to produce the carry.  */
        carry_32 = tcg_temp_new_i32();
        gen_helper_compute_C_icc(carry_32, cpu_env);
        break;
    }

#if TARGET_LONG_BITS == 64
    carry = tcg_temp_new();
    tcg_gen_extu_i32_i64(carry, carry_32);
#else
    carry = carry_32;
#endif

    tcg_gen_sub_tl(dst, src1, src2);
    tcg_gen_sub_tl(dst, dst, carry);

    tcg_temp_free_i32(carry_32);
#if TARGET_LONG_BITS == 64
    tcg_temp_free(carry);
#endif

 sub_done:
    if (update_cc) {
        tcg_gen_mov_tl(cpu_cc_src, src1);
        tcg_gen_mov_tl(cpu_cc_src2, src2);
        tcg_gen_mov_tl(cpu_cc_dst, dst);
        tcg_gen_movi_i32(cpu_cc_op, CC_OP_SUBX);
        dc->cc_op = CC_OP_SUBX;
    }
}

static inline void gen_op_mulscc(TCGv dst, TCGv src1, TCGv src2)
{
    TCGv r_temp, zero, t0;

    r_temp = tcg_temp_new();
    t0 = tcg_temp_new();

    /* old op:
    if (!(env->y & 1))
        T1 = 0;
    */
    zero = tcg_const_tl(0);
    tcg_gen_andi_tl(cpu_cc_src, src1, 0xffffffff);
    tcg_gen_andi_tl(r_temp, cpu_y, 0x1);
    tcg_gen_andi_tl(cpu_cc_src2, src2, 0xffffffff);
    tcg_gen_movcond_tl(TCG_COND_EQ, cpu_cc_src2, r_temp, zero,
                       zero, cpu_cc_src2);
    tcg_temp_free(zero);

    // b2 = T0 & 1;
    // env->y = (b2 << 31) | (env->y >> 1);
    tcg_gen_extract_tl(t0, cpu_y, 1, 31);
    tcg_gen_deposit_tl(cpu_y, t0, cpu_cc_src, 31, 1);

    // b1 = N ^ V;
    gen_mov_reg_N(t0, cpu_psr);
    gen_mov_reg_V(r_temp, cpu_psr);
    tcg_gen_xor_tl(t0, t0, r_temp);
    tcg_temp_free(r_temp);

    // T0 = (b1 << 31) | (T0 >> 1);
    // src1 = T0;
    tcg_gen_shli_tl(t0, t0, 31);
    tcg_gen_shri_tl(cpu_cc_src, cpu_cc_src, 1);
    tcg_gen_or_tl(cpu_cc_src, cpu_cc_src, t0);
    tcg_temp_free(t0);

    tcg_gen_add_tl(cpu_cc_dst, cpu_cc_src, cpu_cc_src2);

    tcg_gen_mov_tl(dst, cpu_cc_dst);
}

static inline void gen_op_multiply(TCGv dst, TCGv src1, TCGv src2, int sign_ext)
{
#if TARGET_LONG_BITS == 32
    if (sign_ext) {
        tcg_gen_muls2_tl(dst, cpu_y, src1, src2);
    } else {
        tcg_gen_mulu2_tl(dst, cpu_y, src1, src2);
    }
#else
    TCGv t0 = tcg_temp_new_i64();
    TCGv t1 = tcg_temp_new_i64();

    if (sign_ext) {
        tcg_gen_ext32s_i64(t0, src1);
        tcg_gen_ext32s_i64(t1, src2);
    } else {
        tcg_gen_ext32u_i64(t0, src1);
        tcg_gen_ext32u_i64(t1, src2);
    }

    tcg_gen_mul_i64(dst, t0, t1);
    tcg_temp_free(t0);
    tcg_temp_free(t1);

    tcg_gen_shri_i64(cpu_y, dst, 32);
#endif
}

static inline void gen_op_umul(TCGv dst, TCGv src1, TCGv src2)
{
    /* zero-extend truncated operands before multiplication */
    gen_op_multiply(dst, src1, src2, 0);
}

static inline void gen_op_smul(TCGv dst, TCGv src1, TCGv src2)
{
    /* sign-extend truncated operands before multiplication */
    gen_op_multiply(dst, src1, src2, 1);
}

// 1
static inline void gen_op_eval_ba(TCGv dst)
{
    tcg_gen_movi_tl(dst, 1);
}

// Z
static inline void gen_op_eval_be(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_Z(dst, src);
}

// Z | (N ^ V)
static inline void gen_op_eval_ble(TCGv dst, TCGv_i32 src)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_N(t0, src);
    gen_mov_reg_V(dst, src);
    tcg_gen_xor_tl(dst, dst, t0);
    gen_mov_reg_Z(t0, src);
    tcg_gen_or_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// N ^ V
static inline void gen_op_eval_bl(TCGv dst, TCGv_i32 src)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_V(t0, src);
    gen_mov_reg_N(dst, src);
    tcg_gen_xor_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// C | Z
static inline void gen_op_eval_bleu(TCGv dst, TCGv_i32 src)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_Z(t0, src);
    gen_mov_reg_C(dst, src);
    tcg_gen_or_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// C
static inline void gen_op_eval_bcs(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_C(dst, src);
}

// V
static inline void gen_op_eval_bvs(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_V(dst, src);
}

// 0
static inline void gen_op_eval_bn(TCGv dst)
{
    tcg_gen_movi_tl(dst, 0);
}

// N
static inline void gen_op_eval_bneg(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_N(dst, src);
}

// !Z
static inline void gen_op_eval_bne(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_Z(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !(Z | (N ^ V))
static inline void gen_op_eval_bg(TCGv dst, TCGv_i32 src)
{
    gen_op_eval_ble(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !(N ^ V)
static inline void gen_op_eval_bge(TCGv dst, TCGv_i32 src)
{
    gen_op_eval_bl(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !(C | Z)
static inline void gen_op_eval_bgu(TCGv dst, TCGv_i32 src)
{
    gen_op_eval_bleu(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !C
static inline void gen_op_eval_bcc(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_C(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !N
static inline void gen_op_eval_bpos(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_N(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !V
static inline void gen_op_eval_bvc(TCGv dst, TCGv_i32 src)
{
    gen_mov_reg_V(dst, src);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

/*
  FPSR bit field FCC1 | FCC0:
   0 =
   1 <
   2 >
   3 unordered
*/
static inline void gen_mov_reg_FCC0(TCGv reg, TCGv src,
                                    unsigned int fcc_offset)
{
    tcg_gen_shri_tl(reg, src, FSR_FCC0_SHIFT + fcc_offset);
    tcg_gen_andi_tl(reg, reg, 0x1);
}

static inline void gen_mov_reg_FCC1(TCGv reg, TCGv src,
                                    unsigned int fcc_offset)
{
    tcg_gen_shri_tl(reg, src, FSR_FCC1_SHIFT + fcc_offset);
    tcg_gen_andi_tl(reg, reg, 0x1);
}

// !0: FCC0 | FCC1
static inline void gen_op_eval_fbne(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_or_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// 1 or 2: FCC0 ^ FCC1
static inline void gen_op_eval_fblg(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_xor_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// 1 or 3: FCC0
static inline void gen_op_eval_fbul(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    gen_mov_reg_FCC0(dst, src, fcc_offset);
}

// 1: FCC0 & !FCC1
static inline void gen_op_eval_fbl(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_andc_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// 2 or 3: FCC1
static inline void gen_op_eval_fbug(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    gen_mov_reg_FCC1(dst, src, fcc_offset);
}

// 2: !FCC0 & FCC1
static inline void gen_op_eval_fbg(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_andc_tl(dst, t0, dst);
    tcg_temp_free(t0);
}

// 3: FCC0 & FCC1
static inline void gen_op_eval_fbu(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_and_tl(dst, dst, t0);
    tcg_temp_free(t0);
}

// 0: !(FCC0 | FCC1)
static inline void gen_op_eval_fbe(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_or_tl(dst, dst, t0);
    tcg_gen_xori_tl(dst, dst, 0x1);
    tcg_temp_free(t0);
}

// 0 or 3: !(FCC0 ^ FCC1)
static inline void gen_op_eval_fbue(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_xor_tl(dst, dst, t0);
    tcg_gen_xori_tl(dst, dst, 0x1);
    tcg_temp_free(t0);
}

// 0 or 2: !FCC0
static inline void gen_op_eval_fbge(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !1: !(FCC0 & !FCC1)
static inline void gen_op_eval_fbuge(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_andc_tl(dst, dst, t0);
    tcg_gen_xori_tl(dst, dst, 0x1);
    tcg_temp_free(t0);
}

// 0 or 1: !FCC1
static inline void gen_op_eval_fble(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    gen_mov_reg_FCC1(dst, src, fcc_offset);
    tcg_gen_xori_tl(dst, dst, 0x1);
}

// !2: !(!FCC0 & FCC1)
static inline void gen_op_eval_fbule(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_andc_tl(dst, t0, dst);
    tcg_gen_xori_tl(dst, dst, 0x1);
    tcg_temp_free(t0);
}

// !3: !(FCC0 & FCC1)
static inline void gen_op_eval_fbo(TCGv dst, TCGv src,
                                    unsigned int fcc_offset)
{
    TCGv t0 = tcg_temp_new();
    gen_mov_reg_FCC0(dst, src, fcc_offset);
    gen_mov_reg_FCC1(t0, src, fcc_offset);
    tcg_gen_and_tl(dst, dst, t0);
    tcg_gen_xori_tl(dst, dst, 0x1);
    tcg_temp_free(t0);
}

static inline void gen_branch2(DisasContext *dc, target_ulong pc1,
                               target_ulong pc2, TCGv r_cond)
{
    TCGLabel *l1 = gen_new_label();

    tcg_gen_brcondi_tl(TCG_COND_EQ, r_cond, 0, l1);

    gen_goto_tb(dc, 0, pc1, pc1 + 4);

    gen_set_label(l1);
    gen_goto_tb(dc, 1, pc2, pc2 + 4);
}

static void gen_branch_a(DisasContext *dc, target_ulong pc1)
{
    TCGLabel *l1 = gen_new_label();
    target_ulong npc = dc->npc;

    tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_cond, 0, l1);

    gen_goto_tb(dc, 0, npc, pc1);

    gen_set_label(l1);
    gen_goto_tb(dc, 1, npc + 4, npc + 8);

    dc->base.is_jmp = DISAS_NORETURN;
}

static void gen_branch_n(DisasContext *dc, target_ulong pc1)
{
    target_ulong npc = dc->npc;

    if (likely(npc != DYNAMIC_PC)) {
        dc->pc = npc;
        dc->jump_pc[0] = pc1;
        dc->jump_pc[1] = npc + 4;
        dc->npc = JUMP_PC;
    } else {
        TCGv t, z;

        tcg_gen_mov_tl(cpu_pc, cpu_npc);

        tcg_gen_addi_tl(cpu_npc, cpu_npc, 4);
        t = tcg_const_tl(pc1);
        z = tcg_const_tl(0);
        tcg_gen_movcond_tl(TCG_COND_NE, cpu_npc, cpu_cond, z, t, cpu_npc);
        tcg_temp_free(t);
        tcg_temp_free(z);

        dc->pc = DYNAMIC_PC;
    }
}

static inline void gen_generic_branch(DisasContext *dc)
{
    TCGv npc0 = tcg_const_tl(dc->jump_pc[0]);
    TCGv npc1 = tcg_const_tl(dc->jump_pc[1]);
    TCGv zero = tcg_const_tl(0);

    tcg_gen_movcond_tl(TCG_COND_NE, cpu_npc, cpu_cond, zero, npc0, npc1);

    tcg_temp_free(npc0);
    tcg_temp_free(npc1);
    tcg_temp_free(zero);
}

/* call this function before using the condition register as it may
   have been set for a jump */
static inline void flush_cond(DisasContext *dc)
{
    if (dc->npc == JUMP_PC) {
        gen_generic_branch(dc);
        dc->npc = DYNAMIC_PC;
    }
}

static inline void save_npc(DisasContext *dc)
{
    if (dc->npc == JUMP_PC) {
        gen_generic_branch(dc);
        dc->npc = DYNAMIC_PC;
    } else if (dc->npc != DYNAMIC_PC) {
        tcg_gen_movi_tl(cpu_npc, dc->npc);
    }
}

static inline void update_psr(DisasContext *dc)
{
    if (dc->cc_op != CC_OP_FLAGS) {
        dc->cc_op = CC_OP_FLAGS;
        gen_helper_compute_psr(cpu_env);
    }
}

static inline void save_state(DisasContext *dc)
{
    tcg_gen_movi_tl(cpu_pc, dc->pc);
    save_npc(dc);
}

static void gen_exception(DisasContext *dc, int which)
{
    TCGv_i32 t;

    save_state(dc);
    t = tcg_const_i32(which);
    gen_helper_raise_exception(cpu_env, t);
    tcg_temp_free_i32(t);
    dc->base.is_jmp = DISAS_NORETURN;
}

static void gen_check_align(TCGv addr, int mask)
{
    TCGv_i32 r_mask = tcg_const_i32(mask);
    gen_helper_check_align(cpu_env, addr, r_mask);
    tcg_temp_free_i32(r_mask);
}

static inline void gen_mov_pc_npc(DisasContext *dc)
{
    if (dc->npc == JUMP_PC) {
        gen_generic_branch(dc);
        tcg_gen_mov_tl(cpu_pc, cpu_npc);
        dc->pc = DYNAMIC_PC;
    } else if (dc->npc == DYNAMIC_PC) {
        tcg_gen_mov_tl(cpu_pc, cpu_npc);
        dc->pc = DYNAMIC_PC;
    } else {
        dc->pc = dc->npc;
    }
}

static inline void gen_op_next_insn(void)
{
    tcg_gen_mov_tl(cpu_pc, cpu_npc);
    tcg_gen_addi_tl(cpu_npc, cpu_npc, 4);
}

static void free_compare(DisasCompare *cmp)
{
    if (!cmp->g1) {
        tcg_temp_free(cmp->c1);
    }
    if (!cmp->g2) {
        tcg_temp_free(cmp->c2);
    }
}

static void gen_compare(DisasCompare *cmp, bool xcc, unsigned int cond,
                        DisasContext *dc)
{
    static int subcc_cond[16] = {
        TCG_COND_NEVER,
        TCG_COND_EQ,
        TCG_COND_LE,
        TCG_COND_LT,
        TCG_COND_LEU,
        TCG_COND_LTU,
        -1, /* neg */
        -1, /* overflow */
        TCG_COND_ALWAYS,
        TCG_COND_NE,
        TCG_COND_GT,
        TCG_COND_GE,
        TCG_COND_GTU,
        TCG_COND_GEU,
        -1, /* pos */
        -1, /* no overflow */
    };

    static int logic_cond[16] = {
        TCG_COND_NEVER,
        TCG_COND_EQ,     /* eq:  Z */
        TCG_COND_LE,     /* le:  Z | (N ^ V) -> Z | N */
        TCG_COND_LT,     /* lt:  N ^ V -> N */
        TCG_COND_EQ,     /* leu: C | Z -> Z */
        TCG_COND_NEVER,  /* ltu: C -> 0 */
        TCG_COND_LT,     /* neg: N */
        TCG_COND_NEVER,  /* vs:  V -> 0 */
        TCG_COND_ALWAYS,
        TCG_COND_NE,     /* ne:  !Z */
        TCG_COND_GT,     /* gt:  !(Z | (N ^ V)) -> !(Z | N) */
        TCG_COND_GE,     /* ge:  !(N ^ V) -> !N */
        TCG_COND_NE,     /* gtu: !(C | Z) -> !Z */
        TCG_COND_ALWAYS, /* geu: !C -> 1 */
        TCG_COND_GE,     /* pos: !N */
        TCG_COND_ALWAYS, /* vc:  !V -> 1 */
    };

    TCGv_i32 r_src;
    TCGv r_dst;

#ifdef TARGET_SPARC64
    if (xcc) {
        r_src = cpu_xcc;
    } else {
        r_src = cpu_psr;
    }
#else
    r_src = cpu_psr;
#endif

    switch (dc->cc_op) {
    case CC_OP_LOGIC:
        cmp->cond = logic_cond[cond];
    do_compare_dst_0:
        cmp->is_bool = false;
        cmp->g2 = false;
        cmp->c2 = tcg_const_tl(0);
#ifdef TARGET_SPARC64
        if (!xcc) {
            cmp->g1 = false;
            cmp->c1 = tcg_temp_new();
            tcg_gen_ext32s_tl(cmp->c1, cpu_cc_dst);
            break;
        }
#endif
        cmp->g1 = true;
        cmp->c1 = cpu_cc_dst;
        break;

    case CC_OP_SUB:
        switch (cond) {
        case 6:  /* neg */
        case 14: /* pos */
            cmp->cond = (cond == 6 ? TCG_COND_LT : TCG_COND_GE);
            goto do_compare_dst_0;

        case 7: /* overflow */
        case 15: /* !overflow */
            goto do_dynamic;

        default:
            cmp->cond = subcc_cond[cond];
            cmp->is_bool = false;
#ifdef TARGET_SPARC64
            if (!xcc) {
                /* Note that sign-extension works for unsigned compares as
                   long as both operands are sign-extended.  */
                cmp->g1 = cmp->g2 = false;
                cmp->c1 = tcg_temp_new();
                cmp->c2 = tcg_temp_new();
                tcg_gen_ext32s_tl(cmp->c1, cpu_cc_src);
                tcg_gen_ext32s_tl(cmp->c2, cpu_cc_src2);
                break;
            }
#endif
            cmp->g1 = cmp->g2 = true;
            cmp->c1 = cpu_cc_src;
            cmp->c2 = cpu_cc_src2;
            break;
        }
        break;

    default:
    do_dynamic:
        gen_helper_compute_psr(cpu_env);
        dc->cc_op = CC_OP_FLAGS;
        /* FALLTHRU */

    case CC_OP_FLAGS:
        /* We're going to generate a boolean result.  */
        cmp->cond = TCG_COND_NE;
        cmp->is_bool = true;
        cmp->g1 = cmp->g2 = false;
        cmp->c1 = r_dst = tcg_temp_new();
        cmp->c2 = tcg_const_tl(0);

        switch (cond) {
        case 0x0:
            gen_op_eval_bn(r_dst);
            break;
        case 0x1:
            gen_op_eval_be(r_dst, r_src);
            break;
        case 0x2:
            gen_op_eval_ble(r_dst, r_src);
            break;
        case 0x3:
            gen_op_eval_bl(r_dst, r_src);
            break;
        case 0x4:
            gen_op_eval_bleu(r_dst, r_src);
            break;
        case 0x5:
            gen_op_eval_bcs(r_dst, r_src);
            break;
        case 0x6:
            gen_op_eval_bneg(r_dst, r_src);
            break;
        case 0x7:
            gen_op_eval_bvs(r_dst, r_src);
            break;
        case 0x8:
            gen_op_eval_ba(r_dst);
            break;
        case 0x9:
            gen_op_eval_bne(r_dst, r_src);
            break;
        case 0xa:
            gen_op_eval_bg(r_dst, r_src);
            break;
        case 0xb:
            gen_op_eval_bge(r_dst, r_src);
            break;
        case 0xc:
            gen_op_eval_bgu(r_dst, r_src);
            break;
        case 0xd:
            gen_op_eval_bcc(r_dst, r_src);
            break;
        case 0xe:
            gen_op_eval_bpos(r_dst, r_src);
            break;
        case 0xf:
            gen_op_eval_bvc(r_dst, r_src);
            break;
        }
        break;
    }
}

static void gen_fcompare(DisasCompare *cmp, unsigned int cc, unsigned int cond)
{
    unsigned int offset;
    TCGv r_dst;

    /* For now we still generate a straight boolean result.  */
    cmp->cond = TCG_COND_NE;
    cmp->is_bool = true;
    cmp->g1 = cmp->g2 = false;
    cmp->c1 = r_dst = tcg_temp_new();
    cmp->c2 = tcg_const_tl(0);

    switch (cc) {
    default:
    case 0x0:
        offset = 0;
        break;
    case 0x1:
        offset = 32 - 10;
        break;
    case 0x2:
        offset = 34 - 10;
        break;
    case 0x3:
        offset = 36 - 10;
        break;
    }

    switch (cond) {
    case 0x0:
        gen_op_eval_bn(r_dst);
        break;
    case 0x1:
        gen_op_eval_fbne(r_dst, cpu_fsr, offset);
        break;
    case 0x2:
        gen_op_eval_fblg(r_dst, cpu_fsr, offset);
        break;
    case 0x3:
        gen_op_eval_fbul(r_dst, cpu_fsr, offset);
        break;
    case 0x4:
        gen_op_eval_fbl(r_dst, cpu_fsr, offset);
        break;
    case 0x5:
        gen_op_eval_fbug(r_dst, cpu_fsr, offset);
        break;
    case 0x6:
        gen_op_eval_fbg(r_dst, cpu_fsr, offset);
        break;
    case 0x7:
        gen_op_eval_fbu(r_dst, cpu_fsr, offset);
        break;
    case 0x8:
        gen_op_eval_ba(r_dst);
        break;
    case 0x9:
        gen_op_eval_fbe(r_dst, cpu_fsr, offset);
        break;
    case 0xa:
        gen_op_eval_fbue(r_dst, cpu_fsr, offset);
        break;
    case 0xb:
        gen_op_eval_fbge(r_dst, cpu_fsr, offset);
        break;
    case 0xc:
        gen_op_eval_fbuge(r_dst, cpu_fsr, offset);
        break;
    case 0xd:
        gen_op_eval_fble(r_dst, cpu_fsr, offset);
        break;
    case 0xe:
        gen_op_eval_fbule(r_dst, cpu_fsr, offset);
        break;
    case 0xf:
        gen_op_eval_fbo(r_dst, cpu_fsr, offset);
        break;
    }
}

static void gen_cond(TCGv r_dst, unsigned int cc, unsigned int cond,
                     DisasContext *dc)
{
    DisasCompare cmp;
    gen_compare(&cmp, cc, cond, dc);

    /* The interface is to return a boolean in r_dst.  */
    if (cmp.is_bool) {
        tcg_gen_mov_tl(r_dst, cmp.c1);
    } else {
        tcg_gen_setcond_tl(cmp.cond, r_dst, cmp.c1, cmp.c2);
    }

    free_compare(&cmp);
}

static void gen_fcond(TCGv r_dst, unsigned int cc, unsigned int cond)
{
    DisasCompare cmp;
    gen_fcompare(&cmp, cc, cond);

    /* The interface is to return a boolean in r_dst.  */
    if (cmp.is_bool) {
        tcg_gen_mov_tl(r_dst, cmp.c1);
    } else {
        tcg_gen_setcond_tl(cmp.cond, r_dst, cmp.c1, cmp.c2);
    }

    free_compare(&cmp);
}

#ifdef TARGET_SPARC64
// Inverted logic
static const int gen_tcg_cond_reg[8] = {
    -1,
    TCG_COND_NE,
    TCG_COND_GT,
    TCG_COND_GE,
    -1,
    TCG_COND_EQ,
    TCG_COND_LE,
    TCG_COND_LT,
};

static void gen_compare_reg(DisasCompare *cmp, int cond, TCGv r_src)
{
    cmp->cond = tcg_invert_cond(gen_tcg_cond_reg[cond]);
    cmp->is_bool = false;
    cmp->g1 = true;
    cmp->g2 = false;
    cmp->c1 = r_src;
    cmp->c2 = tcg_const_tl(0);
}

static inline void gen_cond_reg(TCGv r_dst, int cond, TCGv r_src)
{
    DisasCompare cmp;
    gen_compare_reg(&cmp, cond, r_src);

    /* The interface is to return a boolean in r_dst.  */
    tcg_gen_setcond_tl(cmp.cond, r_dst, cmp.c1, cmp.c2);

    free_compare(&cmp);
}
#endif

static void do_branch(DisasContext *dc, int32_t offset, uint32_t insn, int cc)
{
    unsigned int cond = GET_FIELD(insn, 3, 6), a = (insn & (1 << 29));
    target_ulong target = dc->pc + offset;

#ifdef TARGET_SPARC64
    if (unlikely(AM_CHECK(dc))) {
        target &= 0xffffffffULL;
    }
#endif
    if (cond == 0x0) {
        /* unconditional not taken */
        if (a) {
            dc->pc = dc->npc + 4;
            dc->npc = dc->pc + 4;
        } else {
            dc->pc = dc->npc;
            dc->npc = dc->pc + 4;
        }
    } else if (cond == 0x8) {
        /* unconditional taken */
        if (a) {
            dc->pc = target;
            dc->npc = dc->pc + 4;
        } else {
            dc->pc = dc->npc;
            dc->npc = target;
            tcg_gen_mov_tl(cpu_pc, cpu_npc);
        }
    } else {
        flush_cond(dc);
        gen_cond(cpu_cond, cc, cond, dc);
        if (a) {
            gen_branch_a(dc, target);
        } else {
            gen_branch_n(dc, target);
        }
    }
}

static void do_fbranch(DisasContext *dc, int32_t offset, uint32_t insn, int cc)
{
    unsigned int cond = GET_FIELD(insn, 3, 6), a = (insn & (1 << 29));
    target_ulong target = dc->pc + offset;

#ifdef TARGET_SPARC64
    if (unlikely(AM_CHECK(dc))) {
        target &= 0xffffffffULL;
    }
#endif
    if (cond == 0x0) {
        /* unconditional not taken */
        if (a) {
            dc->pc = dc->npc + 4;
            dc->npc = dc->pc + 4;
        } else {
            dc->pc = dc->npc;
            dc->npc = dc->pc + 4;
        }
    } else if (cond == 0x8) {
        /* unconditional taken */
        if (a) {
            dc->pc = target;
            dc->npc = dc->pc + 4;
        } else {
            dc->pc = dc->npc;
            dc->npc = target;
            tcg_gen_mov_tl(cpu_pc, cpu_npc);
        }
    } else {
        flush_cond(dc);
        gen_fcond(cpu_cond, cc, cond);
        if (a) {
            gen_branch_a(dc, target);
        } else {
            gen_branch_n(dc, target);
        }
    }
}

#ifdef TARGET_SPARC64
static void do_branch_reg(DisasContext *dc, int32_t offset, uint32_t insn,
                          TCGv r_reg)
{
    unsigned int cond = GET_FIELD_SP(insn, 25, 27), a = (insn & (1 << 29));
    target_ulong target = dc->pc + offset;

    if (unlikely(AM_CHECK(dc))) {
        target &= 0xffffffffULL;
    }
    flush_cond(dc);
    gen_cond_reg(cpu_cond, cond, r_reg);
    if (a) {
        gen_branch_a(dc, target);
    } else {
        gen_branch_n(dc, target);
    }
}

static inline void gen_op_fcmps(int fccno, TCGv_i32 r_rs1, TCGv_i32 r_rs2)
{
    switch (fccno) {
    case 0:
        gen_helper_fcmps(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 1:
        gen_helper_fcmps_fcc1(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 2:
        gen_helper_fcmps_fcc2(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 3:
        gen_helper_fcmps_fcc3(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    }
}

static inline void gen_op_fcmpd(int fccno, TCGv_i64 r_rs1, TCGv_i64 r_rs2)
{
    switch (fccno) {
    case 0:
        gen_helper_fcmpd(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 1:
        gen_helper_fcmpd_fcc1(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 2:
        gen_helper_fcmpd_fcc2(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 3:
        gen_helper_fcmpd_fcc3(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    }
}

static inline void gen_op_fcmpq(int fccno)
{
    switch (fccno) {
    case 0:
        gen_helper_fcmpq(cpu_fsr, cpu_env);
        break;
    case 1:
        gen_helper_fcmpq_fcc1(cpu_fsr, cpu_env);
        break;
    case 2:
        gen_helper_fcmpq_fcc2(cpu_fsr, cpu_env);
        break;
    case 3:
        gen_helper_fcmpq_fcc3(cpu_fsr, cpu_env);
        break;
    }
}

static inline void gen_op_fcmpes(int fccno, TCGv_i32 r_rs1, TCGv_i32 r_rs2)
{
    switch (fccno) {
    case 0:
        gen_helper_fcmpes(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 1:
        gen_helper_fcmpes_fcc1(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 2:
        gen_helper_fcmpes_fcc2(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 3:
        gen_helper_fcmpes_fcc3(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    }
}

static inline void gen_op_fcmped(int fccno, TCGv_i64 r_rs1, TCGv_i64 r_rs2)
{
    switch (fccno) {
    case 0:
        gen_helper_fcmped(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 1:
        gen_helper_fcmped_fcc1(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 2:
        gen_helper_fcmped_fcc2(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    case 3:
        gen_helper_fcmped_fcc3(cpu_fsr, cpu_env, r_rs1, r_rs2);
        break;
    }
}

static inline void gen_op_fcmpeq(int fccno)
{
    switch (fccno) {
    case 0:
        gen_helper_fcmpeq(cpu_fsr, cpu_env);
        break;
    case 1:
        gen_helper_fcmpeq_fcc1(cpu_fsr, cpu_env);
        break;
    case 2:
        gen_helper_fcmpeq_fcc2(cpu_fsr, cpu_env);
        break;
    case 3:
        gen_helper_fcmpeq_fcc3(cpu_fsr, cpu_env);
        break;
    }
}

#else

static inline void gen_op_fcmps(int fccno, TCGv r_rs1, TCGv r_rs2)
{
    gen_helper_fcmps(cpu_fsr, cpu_env, r_rs1, r_rs2);
}

static inline void gen_op_fcmpd(int fccno, TCGv_i64 r_rs1, TCGv_i64 r_rs2)
{
    gen_helper_fcmpd(cpu_fsr, cpu_env, r_rs1, r_rs2);
}

static inline void gen_op_fcmpq(int fccno)
{
    gen_helper_fcmpq(cpu_fsr, cpu_env);
}

static inline void gen_op_fcmpes(int fccno, TCGv r_rs1, TCGv r_rs2)
{
    gen_helper_fcmpes(cpu_fsr, cpu_env, r_rs1, r_rs2);
}

static inline void gen_op_fcmped(int fccno, TCGv_i64 r_rs1, TCGv_i64 r_rs2)
{
    gen_helper_fcmped(cpu_fsr, cpu_env, r_rs1, r_rs2);
}

static inline void gen_op_fcmpeq(int fccno)
{
    gen_helper_fcmpeq(cpu_fsr, cpu_env);
}
#endif

static void gen_op_fpexception_im(DisasContext *dc, int fsr_flags)
{
    tcg_gen_andi_tl(cpu_fsr, cpu_fsr, FSR_FTT_NMASK);
    tcg_gen_ori_tl(cpu_fsr, cpu_fsr, fsr_flags);
    gen_exception(dc, TT_FP_EXCP);
}

static int gen_trap_ifnofpu(DisasContext *dc)
{
#if !defined(CONFIG_USER_ONLY)
    if (!dc->fpu_enabled) {
        gen_exception(dc, TT_NFPU_INSN);
        return 1;
    }
#endif
    return 0;
}

static inline void gen_op_clear_ieee_excp_and_FTT(void)
{
    tcg_gen_andi_tl(cpu_fsr, cpu_fsr, FSR_FTT_CEXC_NMASK);
}

static inline void gen_fop_FF(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_i32, TCGv_ptr, TCGv_i32))
{
    TCGv_i32 dst, src;

    src = gen_load_fpr_F(dc, rs);
    dst = gen_dest_fpr_F(dc);

    gen(dst, cpu_env, src);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_F(dc, rd, dst);
}

static inline void gen_ne_fop_FF(DisasContext *dc, int rd, int rs,
                                 void (*gen)(TCGv_i32, TCGv_i32))
{
    TCGv_i32 dst, src;

    src = gen_load_fpr_F(dc, rs);
    dst = gen_dest_fpr_F(dc);

    gen(dst, src);

    gen_store_fpr_F(dc, rd, dst);
}

static inline void gen_fop_FFF(DisasContext *dc, int rd, int rs1, int rs2,
                        void (*gen)(TCGv_i32, TCGv_ptr, TCGv_i32, TCGv_i32))
{
    TCGv_i32 dst, src1, src2;

    src1 = gen_load_fpr_F(dc, rs1);
    src2 = gen_load_fpr_F(dc, rs2);
    dst = gen_dest_fpr_F(dc);

    gen(dst, cpu_env, src1, src2);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_F(dc, rd, dst);
}

#ifdef TARGET_SPARC64
static inline void gen_ne_fop_FFF(DisasContext *dc, int rd, int rs1, int rs2,
                                  void (*gen)(TCGv_i32, TCGv_i32, TCGv_i32))
{
    TCGv_i32 dst, src1, src2;

    src1 = gen_load_fpr_F(dc, rs1);
    src2 = gen_load_fpr_F(dc, rs2);
    dst = gen_dest_fpr_F(dc);

    gen(dst, src1, src2);

    gen_store_fpr_F(dc, rd, dst);
}
#endif

static inline void gen_fop_DD(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_i64, TCGv_ptr, TCGv_i64))
{
    TCGv_i64 dst, src;

    src = gen_load_fpr_D(dc, rs);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_env, src);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_D(dc, rd, dst);
}

#ifdef TARGET_SPARC64
static inline void gen_ne_fop_DD(DisasContext *dc, int rd, int rs,
                                 void (*gen)(TCGv_i64, TCGv_i64))
{
    TCGv_i64 dst, src;

    src = gen_load_fpr_D(dc, rs);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, src);

    gen_store_fpr_D(dc, rd, dst);
}
#endif

static inline void gen_fop_DDD(DisasContext *dc, int rd, int rs1, int rs2,
                        void (*gen)(TCGv_i64, TCGv_ptr, TCGv_i64, TCGv_i64))
{
    TCGv_i64 dst, src1, src2;

    src1 = gen_load_fpr_D(dc, rs1);
    src2 = gen_load_fpr_D(dc, rs2);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_env, src1, src2);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_D(dc, rd, dst);
}

#ifdef TARGET_SPARC64
static inline void gen_ne_fop_DDD(DisasContext *dc, int rd, int rs1, int rs2,
                                  void (*gen)(TCGv_i64, TCGv_i64, TCGv_i64))
{
    TCGv_i64 dst, src1, src2;

    src1 = gen_load_fpr_D(dc, rs1);
    src2 = gen_load_fpr_D(dc, rs2);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, src1, src2);

    gen_store_fpr_D(dc, rd, dst);
}

static inline void gen_gsr_fop_DDD(DisasContext *dc, int rd, int rs1, int rs2,
                           void (*gen)(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_i64))
{
    TCGv_i64 dst, src1, src2;

    src1 = gen_load_fpr_D(dc, rs1);
    src2 = gen_load_fpr_D(dc, rs2);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_gsr, src1, src2);

    gen_store_fpr_D(dc, rd, dst);
}

static inline void gen_ne_fop_DDDD(DisasContext *dc, int rd, int rs1, int rs2,
                           void (*gen)(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_i64))
{
    TCGv_i64 dst, src0, src1, src2;

    src1 = gen_load_fpr_D(dc, rs1);
    src2 = gen_load_fpr_D(dc, rs2);
    src0 = gen_load_fpr_D(dc, rd);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, src0, src1, src2);

    gen_store_fpr_D(dc, rd, dst);
}
#endif

static inline void gen_fop_QQ(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_ptr))
{
    gen_op_load_fpr_QT1(QFPREG(rs));

    gen(cpu_env);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_op_store_QT0_fpr(QFPREG(rd));
    gen_update_fprs_dirty(dc, QFPREG(rd));
}

#ifdef TARGET_SPARC64
static inline void gen_ne_fop_QQ(DisasContext *dc, int rd, int rs,
                                 void (*gen)(TCGv_ptr))
{
    gen_op_load_fpr_QT1(QFPREG(rs));

    gen(cpu_env);

    gen_op_store_QT0_fpr(QFPREG(rd));
    gen_update_fprs_dirty(dc, QFPREG(rd));
}
#endif

static inline void gen_fop_QQQ(DisasContext *dc, int rd, int rs1, int rs2,
                               void (*gen)(TCGv_ptr))
{
    gen_op_load_fpr_QT0(QFPREG(rs1));
    gen_op_load_fpr_QT1(QFPREG(rs2));

    gen(cpu_env);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_op_store_QT0_fpr(QFPREG(rd));
    gen_update_fprs_dirty(dc, QFPREG(rd));
}

static inline void gen_fop_DFF(DisasContext *dc, int rd, int rs1, int rs2,
                        void (*gen)(TCGv_i64, TCGv_ptr, TCGv_i32, TCGv_i32))
{
    TCGv_i64 dst;
    TCGv_i32 src1, src2;

    src1 = gen_load_fpr_F(dc, rs1);
    src2 = gen_load_fpr_F(dc, rs2);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_env, src1, src2);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_D(dc, rd, dst);
}

static inline void gen_fop_QDD(DisasContext *dc, int rd, int rs1, int rs2,
                               void (*gen)(TCGv_ptr, TCGv_i64, TCGv_i64))
{
    TCGv_i64 src1, src2;

    src1 = gen_load_fpr_D(dc, rs1);
    src2 = gen_load_fpr_D(dc, rs2);

    gen(cpu_env, src1, src2);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_op_store_QT0_fpr(QFPREG(rd));
    gen_update_fprs_dirty(dc, QFPREG(rd));
}

#ifdef TARGET_SPARC64
static inline void gen_fop_DF(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_i64, TCGv_ptr, TCGv_i32))
{
    TCGv_i64 dst;
    TCGv_i32 src;

    src = gen_load_fpr_F(dc, rs);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_env, src);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_D(dc, rd, dst);
}
#endif

static inline void gen_ne_fop_DF(DisasContext *dc, int rd, int rs,
                                 void (*gen)(TCGv_i64, TCGv_ptr, TCGv_i32))
{
    TCGv_i64 dst;
    TCGv_i32 src;

    src = gen_load_fpr_F(dc, rs);
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_env, src);

    gen_store_fpr_D(dc, rd, dst);
}

static inline void gen_fop_FD(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_i32, TCGv_ptr, TCGv_i64))
{
    TCGv_i32 dst;
    TCGv_i64 src;

    src = gen_load_fpr_D(dc, rs);
    dst = gen_dest_fpr_F(dc);

    gen(dst, cpu_env, src);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_F(dc, rd, dst);
}

static inline void gen_fop_FQ(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_i32, TCGv_ptr))
{
    TCGv_i32 dst;

    gen_op_load_fpr_QT1(QFPREG(rs));
    dst = gen_dest_fpr_F(dc);

    gen(dst, cpu_env);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_F(dc, rd, dst);
}

static inline void gen_fop_DQ(DisasContext *dc, int rd, int rs,
                              void (*gen)(TCGv_i64, TCGv_ptr))
{
    TCGv_i64 dst;

    gen_op_load_fpr_QT1(QFPREG(rs));
    dst = gen_dest_fpr_D(dc, rd);

    gen(dst, cpu_env);
    gen_helper_check_ieee_exceptions(cpu_fsr, cpu_env);

    gen_store_fpr_D(dc, rd, dst);
}

static inline void gen_ne_fop_QF(DisasContext *dc, int rd, int rs,
                                 void (*gen)(TCGv_ptr, TCGv_i32))
{
    TCGv_i32 src;

    src = gen_load_fpr_F(dc, rs);

    gen(cpu_env, src);

    gen_op_store_QT0_fpr(QFPREG(rd));
    gen_update_fprs_dirty(dc, QFPREG(rd));
}

static inline void gen_ne_fop_QD(DisasContext *dc, int rd, int rs,
                                 void (*gen)(TCGv_ptr, TCGv_i64))
{
    TCGv_i64 src;

    src = gen_load_fpr_D(dc, rs);

    gen(cpu_env, src);

    gen_op_store_QT0_fpr(QFPREG(rd));
    gen_update_fprs_dirty(dc, QFPREG(rd));
}

static void gen_swap(DisasContext *dc, TCGv dst, TCGv src,
                     TCGv addr, int mmu_idx, MemOp memop)
{
    gen_address_mask(dc, addr);
    tcg_gen_atomic_xchg_tl(dst, addr, src, mmu_idx, memop);
}

static void gen_ldstub(DisasContext *dc, TCGv dst, TCGv addr, int mmu_idx)
{
    TCGv m1 = tcg_const_tl(0xff);
    gen_address_mask(dc, addr);
    tcg_gen_atomic_xchg_tl(dst, addr, m1, mmu_idx, MO_UB);
    tcg_temp_free(m1);
}

/* asi moves */
#if !defined(CONFIG_USER_ONLY) || defined(TARGET_SPARC64)
typedef enum {
    GET_ASI_HELPER,
    GET_ASI_EXCP,
    GET_ASI_DIRECT,
    GET_ASI_DTWINX,
    GET_ASI_BLOCK,
    GET_ASI_SHORT,
    GET_ASI_BCOPY,
    GET_ASI_BFILL,
} ASIType;

typedef struct {
    ASIType type;
    int asi;
    int mem_idx;
    MemOp memop;
} DisasASI;

static DisasASI get_asi(DisasContext *dc, int insn, MemOp memop)
{
    int asi = GET_FIELD(insn, 19, 26);
    ASIType type = GET_ASI_HELPER;
    int mem_idx = dc->mem_idx;

#ifndef TARGET_SPARC64
    /* Before v9, all asis are immediate and privileged.  */
    if (IS_IMM) {
        gen_exception(dc, TT_ILL_INSN);
        type = GET_ASI_EXCP;
    } else if (supervisor(dc)
               /* Note that LEON accepts ASI_USERDATA in user mode, for
                  use with CASA.  Also note that previous versions of
                  QEMU allowed (and old versions of gcc emitted) ASI_P
                  for LEON, which is incorrect.  */
               || (asi == ASI_USERDATA
                   && (dc->def->features & CPU_FEATURE_CASA))) {
        switch (asi) {
        case ASI_USERDATA:   /* User data access */
            mem_idx = MMU_USER_IDX;
            type = GET_ASI_DIRECT;
            break;
        case ASI_KERNELDATA: /* Supervisor data access */
            mem_idx = MMU_KERNEL_IDX;
            type = GET_ASI_DIRECT;
            break;
        case ASI_M_BYPASS:    /* MMU passthrough */
        case ASI_LEON_BYPASS: /* LEON MMU passthrough */
            mem_idx = MMU_PHYS_IDX;
            type = GET_ASI_DIRECT;
            break;
        case ASI_M_BCOPY: /* Block copy, sta access */
            mem_idx = MMU_KERNEL_IDX;
            type = GET_ASI_BCOPY;
            break;
        case ASI_M_BFILL: /* Block fill, stda access */
            mem_idx = MMU_KERNEL_IDX;
            type = GET_ASI_BFILL;
            break;
        }

        /* MMU_PHYS_IDX is used when the MMU is disabled to passthrough the
         * permissions check in get_physical_address(..).
         */
        mem_idx = (dc->mem_idx == MMU_PHYS_IDX) ? MMU_PHYS_IDX : mem_idx;
    } else {
        gen_exception(dc, TT_PRIV_INSN);
        type = GET_ASI_EXCP;
    }
#else
    if (IS_IMM) {
        asi = dc->asi;
    }
    /* With v9, all asis below 0x80 are privileged.  */
    /* ??? We ought to check cpu_has_hypervisor, but we didn't copy
       down that bit into DisasContext.  For the moment that's ok,
       since the direct implementations below doesn't have any ASIs
       in the restricted [0x30, 0x7f] range, and the check will be
       done properly in the helper.  */
    if (!supervisor(dc) && asi < 0x80) {
        gen_exception(dc, TT_PRIV_ACT);
        type = GET_ASI_EXCP;
    } else {
        switch (asi) {
        case ASI_REAL:      /* Bypass */
        case ASI_REAL_IO:   /* Bypass, non-cacheable */
        case ASI_REAL_L:    /* Bypass LE */
        case ASI_REAL_IO_L: /* Bypass, non-cacheable LE */
        case ASI_TWINX_REAL:   /* Real address, twinx */
        case ASI_TWINX_REAL_L: /* Real address, twinx, LE */
        case ASI_QUAD_LDD_PHYS:
        case ASI_QUAD_LDD_PHYS_L:
            mem_idx = MMU_PHYS_IDX;
            break;
        case ASI_N:  /* Nucleus */
        case ASI_NL: /* Nucleus LE */
        case ASI_TWINX_N:
        case ASI_TWINX_NL:
        case ASI_NUCLEUS_QUAD_LDD:
        case ASI_NUCLEUS_QUAD_LDD_L:
            if (hypervisor(dc)) {
                mem_idx = MMU_PHYS_IDX;
            } else {
                mem_idx = MMU_NUCLEUS_IDX;
            }
            break;
        case ASI_AIUP:  /* As if user primary */
        case ASI_AIUPL: /* As if user primary LE */
        case ASI_TWINX_AIUP:
        case ASI_TWINX_AIUP_L:
        case ASI_BLK_AIUP_4V:
        case ASI_BLK_AIUP_L_4V:
        case ASI_BLK_AIUP:
        case ASI_BLK_AIUPL:
            mem_idx = MMU_USER_IDX;
            break;
        case ASI_AIUS:  /* As if user secondary */
        case ASI_AIUSL: /* As if user secondary LE */
        case ASI_TWINX_AIUS:
        case ASI_TWINX_AIUS_L:
        case ASI_BLK_AIUS_4V:
        case ASI_BLK_AIUS_L_4V:
        case ASI_BLK_AIUS:
        case ASI_BLK_AIUSL:
            mem_idx = MMU_USER_SECONDARY_IDX;
            break;
        case ASI_S:  /* Secondary */
        case ASI_SL: /* Secondary LE */
        case ASI_TWINX_S:
        case ASI_TWINX_SL:
        case ASI_BLK_COMMIT_S:
        case ASI_BLK_S:
        case ASI_BLK_SL:
        case ASI_FL8_S:
        case ASI_FL8_SL:
        case ASI_FL16_S:
        case ASI_FL16_SL:
            if (mem_idx == MMU_USER_IDX) {
                mem_idx = MMU_USER_SECONDARY_IDX;
            } else if (mem_idx == MMU_KERNEL_IDX) {
                mem_idx = MMU_KERNEL_SECONDARY_IDX;
            }
            break;
        case ASI_P:  /* Primary */
        case ASI_PL: /* Primary LE */
        case ASI_TWINX_P:
        case ASI_TWINX_PL:
        case ASI_BLK_COMMIT_P:
        case ASI_BLK_P:
        case ASI_BLK_PL:
        case ASI_FL8_P:
        case ASI_FL8_PL:
        case ASI_FL16_P:
        case ASI_FL16_PL:
            break;
        }
        switch (asi) {
        case ASI_REAL:
        case ASI_REAL_IO:
        case ASI_REAL_L:
        case ASI_REAL_IO_L:
        case ASI_N:
        case ASI_NL:
        case ASI_AIUP:
        case ASI_AIUPL:
        case ASI_AIUS:
        case ASI_AIUSL:
        case ASI_S:
        case ASI_SL:
        case ASI_P:
        case ASI_PL:
            type = GET_ASI_DIRECT;
            break;
        case ASI_TWINX_REAL:
        case ASI_TWINX_REAL_L:
        case ASI_TWINX_N:
        case ASI_TWINX_NL:
        case ASI_TWINX_AIUP:
        case ASI_TWINX_AIUP_L:
        case ASI_TWINX_AIUS:
        case ASI_TWINX_AIUS_L:
        case ASI_TWINX_P:
        case ASI_TWINX_PL:
        case ASI_TWINX_S:
        case ASI_TWINX_SL:
        case ASI_QUAD_LDD_PHYS:
        case ASI_QUAD_LDD_PHYS_L:
        case ASI_NUCLEUS_QUAD_LDD:
        case ASI_NUCLEUS_QUAD_LDD_L:
            type = GET_ASI_DTWINX;
            break;
        case ASI_BLK_COMMIT_P:
        case ASI_BLK_COMMIT_S:
        case ASI_BLK_AIUP_4V:
        case ASI_BLK_AIUP_L_4V:
        case ASI_BLK_AIUP:
        case ASI_BLK_AIUPL:
        case ASI_BLK_AIUS_4V:
        case ASI_BLK_AIUS_L_4V:
        case ASI_BLK_AIUS:
        case ASI_BLK_AIUSL:
        case ASI_BLK_S:
        case ASI_BLK_SL:
        case ASI_BLK_P:
        case ASI_BLK_PL:
            type = GET_ASI_BLOCK;
            break;
        case ASI_FL8_S:
        case ASI_FL8_SL:
        case ASI_FL8_P:
        case ASI_FL8_PL:
            memop = MO_UB;
            type = GET_ASI_SHORT;
            break;
        case ASI_FL16_S:
        case ASI_FL16_SL:
        case ASI_FL16_P:
        case ASI_FL16_PL:
            memop = MO_TEUW;
            type = GET_ASI_SHORT;
            break;
        }
        /* The little-endian asis all have bit 3 set.  */
        if (asi & 8) {
            memop ^= MO_BSWAP;
        }
    }
#endif

    return (DisasASI){ type, asi, mem_idx, memop };
}

static void gen_ld_asi(DisasContext *dc, TCGv dst, TCGv addr,
                       int insn, MemOp memop)
{
    DisasASI da = get_asi(dc, insn, memop);

    switch (da.type) {
    case GET_ASI_EXCP:
        break;
    case GET_ASI_DTWINX: /* Reserved for ldda.  */
        gen_exception(dc, TT_ILL_INSN);
        break;
    case GET_ASI_DIRECT:
        gen_address_mask(dc, addr);
        tcg_gen_qemu_ld_tl(dst, addr, da.mem_idx, da.memop);
        break;
    default:
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(memop);

            save_state(dc);
#ifdef TARGET_SPARC64
            gen_helper_ld_asi(dst, cpu_env, addr, r_asi, r_mop);
#else
            {
                TCGv_i64 t64 = tcg_temp_new_i64();
                gen_helper_ld_asi(t64, cpu_env, addr, r_asi, r_mop);
                tcg_gen_trunc_i64_tl(dst, t64);
                tcg_temp_free_i64(t64);
            }
#endif
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);
        }
        break;
    }
}

static void gen_st_asi(DisasContext *dc, TCGv src, TCGv addr,
                       int insn, MemOp memop)
{
    DisasASI da = get_asi(dc, insn, memop);

    switch (da.type) {
    case GET_ASI_EXCP:
        break;
    case GET_ASI_DTWINX: /* Reserved for stda.  */
#ifndef TARGET_SPARC64
        gen_exception(dc, TT_ILL_INSN);
        break;
#else
        if (!(dc->def->features & CPU_FEATURE_HYPV)) {
            /* Pre OpenSPARC CPUs don't have these */
            gen_exception(dc, TT_ILL_INSN);
            return;
        }
        /* in OpenSPARC T1+ CPUs TWINX ASIs in store instructions
         * are ST_BLKINIT_ ASIs */
        /* fall through */
#endif
    case GET_ASI_DIRECT:
        gen_address_mask(dc, addr);
        tcg_gen_qemu_st_tl(src, addr, da.mem_idx, da.memop);
        break;
#if !defined(TARGET_SPARC64) && !defined(CONFIG_USER_ONLY)
    case GET_ASI_BCOPY:
        /* Copy 32 bytes from the address in SRC to ADDR.  */
        /* ??? The original qemu code suggests 4-byte alignment, dropping
           the low bits, but the only place I can see this used is in the
           Linux kernel with 32 byte alignment, which would make more sense
           as a cacheline-style operation.  */
        {
            TCGv saddr = tcg_temp_new();
            TCGv daddr = tcg_temp_new();
            TCGv four = tcg_const_tl(4);
            TCGv_i32 tmp = tcg_temp_new_i32();
            int i;

            tcg_gen_andi_tl(saddr, src, -4);
            tcg_gen_andi_tl(daddr, addr, -4);
            for (i = 0; i < 32; i += 4) {
                /* Since the loads and stores are paired, allow the
                   copy to happen in the host endianness.  */
                tcg_gen_qemu_ld_i32(tmp, saddr, da.mem_idx, MO_UL);
                tcg_gen_qemu_st_i32(tmp, daddr, da.mem_idx, MO_UL);
                tcg_gen_add_tl(saddr, saddr, four);
                tcg_gen_add_tl(daddr, daddr, four);
            }

            tcg_temp_free(saddr);
            tcg_temp_free(daddr);
            tcg_temp_free(four);
            tcg_temp_free_i32(tmp);
        }
        break;
#endif
    default:
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(memop & MO_SIZE);

            save_state(dc);
#ifdef TARGET_SPARC64
            gen_helper_st_asi(cpu_env, addr, src, r_asi, r_mop);
#else
            {
                TCGv_i64 t64 = tcg_temp_new_i64();
                tcg_gen_extu_tl_i64(t64, src);
                gen_helper_st_asi(cpu_env, addr, t64, r_asi, r_mop);
                tcg_temp_free_i64(t64);
            }
#endif
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);

            /* A write to a TLB register may alter page maps.  End the TB. */
            dc->npc = DYNAMIC_PC;
        }
        break;
    }
}

static void gen_swap_asi(DisasContext *dc, TCGv dst, TCGv src,
                         TCGv addr, int insn)
{
    DisasASI da = get_asi(dc, insn, MO_TEUL);

    switch (da.type) {
    case GET_ASI_EXCP:
        break;
    case GET_ASI_DIRECT:
        gen_swap(dc, dst, src, addr, da.mem_idx, da.memop);
        break;
    default:
        /* ??? Should be DAE_invalid_asi.  */
        gen_exception(dc, TT_DATA_ACCESS);
        break;
    }
}

static void gen_cas_asi(DisasContext *dc, TCGv addr, TCGv cmpv,
                        int insn, int rd)
{
    DisasASI da = get_asi(dc, insn, MO_TEUL);
    TCGv oldv;

    switch (da.type) {
    case GET_ASI_EXCP:
        return;
    case GET_ASI_DIRECT:
        oldv = tcg_temp_new();
        tcg_gen_atomic_cmpxchg_tl(oldv, addr, cmpv, gen_load_gpr(dc, rd),
                                  da.mem_idx, da.memop);
        gen_store_gpr(dc, rd, oldv);
        tcg_temp_free(oldv);
        break;
    default:
        /* ??? Should be DAE_invalid_asi.  */
        gen_exception(dc, TT_DATA_ACCESS);
        break;
    }
}

static void gen_ldstub_asi(DisasContext *dc, TCGv dst, TCGv addr, int insn)
{
    DisasASI da = get_asi(dc, insn, MO_UB);

    switch (da.type) {
    case GET_ASI_EXCP:
        break;
    case GET_ASI_DIRECT:
        gen_ldstub(dc, dst, addr, da.mem_idx);
        break;
    default:
        /* ??? In theory, this should be raise DAE_invalid_asi.
           But the SS-20 roms do ldstuba [%l0] #ASI_M_CTL, %o1.  */
        if (tb_cflags(dc->base.tb) & CF_PARALLEL) {
            gen_helper_exit_atomic(cpu_env);
        } else {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(MO_UB);
            TCGv_i64 s64, t64;

            save_state(dc);
            t64 = tcg_temp_new_i64();
            gen_helper_ld_asi(t64, cpu_env, addr, r_asi, r_mop);

            s64 = tcg_const_i64(0xff);
            gen_helper_st_asi(cpu_env, addr, s64, r_asi, r_mop);
            tcg_temp_free_i64(s64);
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);

            tcg_gen_trunc_i64_tl(dst, t64);
            tcg_temp_free_i64(t64);

            /* End the TB.  */
            dc->npc = DYNAMIC_PC;
        }
        break;
    }
}
#endif

#ifdef TARGET_SPARC64
static void gen_ldf_asi(DisasContext *dc, TCGv addr,
                        int insn, int size, int rd)
{
    DisasASI da = get_asi(dc, insn, (size == 4 ? MO_TEUL : MO_TEQ));
    TCGv_i32 d32;
    TCGv_i64 d64;

    switch (da.type) {
    case GET_ASI_EXCP:
        break;

    case GET_ASI_DIRECT:
        gen_address_mask(dc, addr);
        switch (size) {
        case 4:
            d32 = gen_dest_fpr_F(dc);
            tcg_gen_qemu_ld_i32(d32, addr, da.mem_idx, da.memop);
            gen_store_fpr_F(dc, rd, d32);
            break;
        case 8:
            tcg_gen_qemu_ld_i64(cpu_fpr[rd / 2], addr, da.mem_idx,
                                da.memop | MO_ALIGN_4);
            break;
        case 16:
            d64 = tcg_temp_new_i64();
            tcg_gen_qemu_ld_i64(d64, addr, da.mem_idx, da.memop | MO_ALIGN_4);
            tcg_gen_addi_tl(addr, addr, 8);
            tcg_gen_qemu_ld_i64(cpu_fpr[rd/2+1], addr, da.mem_idx,
                                da.memop | MO_ALIGN_4);
            tcg_gen_mov_i64(cpu_fpr[rd / 2], d64);
            tcg_temp_free_i64(d64);
            break;
        default:
            g_assert_not_reached();
        }
        break;

    case GET_ASI_BLOCK:
        /* Valid for lddfa on aligned registers only.  */
        if (size == 8 && (rd & 7) == 0) {
            MemOp memop;
            TCGv eight;
            int i;

            gen_address_mask(dc, addr);

            /* The first operation checks required alignment.  */
            memop = da.memop | MO_ALIGN_64;
            eight = tcg_const_tl(8);
            for (i = 0; ; ++i) {
                tcg_gen_qemu_ld_i64(cpu_fpr[rd / 2 + i], addr,
                                    da.mem_idx, memop);
                if (i == 7) {
                    break;
                }
                tcg_gen_add_tl(addr, addr, eight);
                memop = da.memop;
            }
            tcg_temp_free(eight);
        } else {
            gen_exception(dc, TT_ILL_INSN);
        }
        break;

    case GET_ASI_SHORT:
        /* Valid for lddfa only.  */
        if (size == 8) {
            gen_address_mask(dc, addr);
            tcg_gen_qemu_ld_i64(cpu_fpr[rd / 2], addr, da.mem_idx, da.memop);
        } else {
            gen_exception(dc, TT_ILL_INSN);
        }
        break;

    default:
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(da.memop);

            save_state(dc);
            /* According to the table in the UA2011 manual, the only
               other asis that are valid for ldfa/lddfa/ldqfa are
               the NO_FAULT asis.  We still need a helper for these,
               but we can just use the integer asi helper for them.  */
            switch (size) {
            case 4:
                d64 = tcg_temp_new_i64();
                gen_helper_ld_asi(d64, cpu_env, addr, r_asi, r_mop);
                d32 = gen_dest_fpr_F(dc);
                tcg_gen_extrl_i64_i32(d32, d64);
                tcg_temp_free_i64(d64);
                gen_store_fpr_F(dc, rd, d32);
                break;
            case 8:
                gen_helper_ld_asi(cpu_fpr[rd / 2], cpu_env, addr, r_asi, r_mop);
                break;
            case 16:
                d64 = tcg_temp_new_i64();
                gen_helper_ld_asi(d64, cpu_env, addr, r_asi, r_mop);
                tcg_gen_addi_tl(addr, addr, 8);
                gen_helper_ld_asi(cpu_fpr[rd/2+1], cpu_env, addr, r_asi, r_mop);
                tcg_gen_mov_i64(cpu_fpr[rd / 2], d64);
                tcg_temp_free_i64(d64);
                break;
            default:
                g_assert_not_reached();
            }
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);
        }
        break;
    }
}

static void gen_stf_asi(DisasContext *dc, TCGv addr,
                        int insn, int size, int rd)
{
    DisasASI da = get_asi(dc, insn, (size == 4 ? MO_TEUL : MO_TEQ));
    TCGv_i32 d32;

    switch (da.type) {
    case GET_ASI_EXCP:
        break;

    case GET_ASI_DIRECT:
        gen_address_mask(dc, addr);
        switch (size) {
        case 4:
            d32 = gen_load_fpr_F(dc, rd);
            tcg_gen_qemu_st_i32(d32, addr, da.mem_idx, da.memop);
            break;
        case 8:
            tcg_gen_qemu_st_i64(cpu_fpr[rd / 2], addr, da.mem_idx,
                                da.memop | MO_ALIGN_4);
            break;
        case 16:
            /* Only 4-byte alignment required.  However, it is legal for the
               cpu to signal the alignment fault, and the OS trap handler is
               required to fix it up.  Requiring 16-byte alignment here avoids
               having to probe the second page before performing the first
               write.  */
            tcg_gen_qemu_st_i64(cpu_fpr[rd / 2], addr, da.mem_idx,
                                da.memop | MO_ALIGN_16);
            tcg_gen_addi_tl(addr, addr, 8);
            tcg_gen_qemu_st_i64(cpu_fpr[rd/2+1], addr, da.mem_idx, da.memop);
            break;
        default:
            g_assert_not_reached();
        }
        break;

    case GET_ASI_BLOCK:
        /* Valid for stdfa on aligned registers only.  */
        if (size == 8 && (rd & 7) == 0) {
            MemOp memop;
            TCGv eight;
            int i;

            gen_address_mask(dc, addr);

            /* The first operation checks required alignment.  */
            memop = da.memop | MO_ALIGN_64;
            eight = tcg_const_tl(8);
            for (i = 0; ; ++i) {
                tcg_gen_qemu_st_i64(cpu_fpr[rd / 2 + i], addr,
                                    da.mem_idx, memop);
                if (i == 7) {
                    break;
                }
                tcg_gen_add_tl(addr, addr, eight);
                memop = da.memop;
            }
            tcg_temp_free(eight);
        } else {
            gen_exception(dc, TT_ILL_INSN);
        }
        break;

    case GET_ASI_SHORT:
        /* Valid for stdfa only.  */
        if (size == 8) {
            gen_address_mask(dc, addr);
            tcg_gen_qemu_st_i64(cpu_fpr[rd / 2], addr, da.mem_idx, da.memop);
        } else {
            gen_exception(dc, TT_ILL_INSN);
        }
        break;

    default:
        /* According to the table in the UA2011 manual, the only
           other asis that are valid for ldfa/lddfa/ldqfa are
           the PST* asis, which aren't currently handled.  */
        gen_exception(dc, TT_ILL_INSN);
        break;
    }
}

static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd)
{
    DisasASI da = get_asi(dc, insn, MO_TEQ);
    TCGv_i64 hi = gen_dest_gpr(dc, rd);
    TCGv_i64 lo = gen_dest_gpr(dc, rd + 1);

    switch (da.type) {
    case GET_ASI_EXCP:
        return;

    case GET_ASI_DTWINX:
        gen_address_mask(dc, addr);
        tcg_gen_qemu_ld_i64(hi, addr, da.mem_idx, da.memop | MO_ALIGN_16);
        tcg_gen_addi_tl(addr, addr, 8);
        tcg_gen_qemu_ld_i64(lo, addr, da.mem_idx, da.memop);
        break;

    case GET_ASI_DIRECT:
        {
            TCGv_i64 tmp = tcg_temp_new_i64();

            gen_address_mask(dc, addr);
            tcg_gen_qemu_ld_i64(tmp, addr, da.mem_idx, da.memop);

            /* Note that LE ldda acts as if each 32-bit register
               result is byte swapped.  Having just performed one
               64-bit bswap, we need now to swap the writebacks.  */
            if ((da.memop & MO_BSWAP) == MO_TE) {
                tcg_gen_extr32_i64(lo, hi, tmp);
            } else {
                tcg_gen_extr32_i64(hi, lo, tmp);
            }
            tcg_temp_free_i64(tmp);
        }
        break;

    default:
        /* ??? In theory we've handled all of the ASIs that are valid
           for ldda, and this should raise DAE_invalid_asi.  However,
           real hardware allows others.  This can be seen with e.g.
           FreeBSD 10.3 wrt ASI_IC_TAG.  */
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(da.memop);
            TCGv_i64 tmp = tcg_temp_new_i64();

            save_state(dc);
            gen_helper_ld_asi(tmp, cpu_env, addr, r_asi, r_mop);
            tcg_temp_free_i32(r_asi);
            tcg_temp_free_i32(r_mop);

            /* See above.  */
            if ((da.memop & MO_BSWAP) == MO_TE) {
                tcg_gen_extr32_i64(lo, hi, tmp);
            } else {
                tcg_gen_extr32_i64(hi, lo, tmp);
            }
            tcg_temp_free_i64(tmp);
        }
        break;
    }

    gen_store_gpr(dc, rd, hi);
    gen_store_gpr(dc, rd + 1, lo);
}

static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr,
                         int insn, int rd)
{
    DisasASI da = get_asi(dc, insn, MO_TEQ);
    TCGv lo = gen_load_gpr(dc, rd + 1);

    switch (da.type) {
    case GET_ASI_EXCP:
        break;

    case GET_ASI_DTWINX:
        gen_address_mask(dc, addr);
        tcg_gen_qemu_st_i64(hi, addr, da.mem_idx, da.memop | MO_ALIGN_16);
        tcg_gen_addi_tl(addr, addr, 8);
        tcg_gen_qemu_st_i64(lo, addr, da.mem_idx, da.memop);
        break;

    case GET_ASI_DIRECT:
        {
            TCGv_i64 t64 = tcg_temp_new_i64();

            /* Note that LE stda acts as if each 32-bit register result is
               byte swapped.  We will perform one 64-bit LE store, so now
               we must swap the order of the construction.  */
            if ((da.memop & MO_BSWAP) == MO_TE) {
                tcg_gen_concat32_i64(t64, lo, hi);
            } else {
                tcg_gen_concat32_i64(t64, hi, lo);
            }
            gen_address_mask(dc, addr);
            tcg_gen_qemu_st_i64(t64, addr, da.mem_idx, da.memop);
            tcg_temp_free_i64(t64);
        }
        break;

    default:
        /* ??? In theory we've handled all of the ASIs that are valid
           for stda, and this should raise DAE_invalid_asi.  */
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(da.memop);
            TCGv_i64 t64 = tcg_temp_new_i64();

            /* See above.  */
            if ((da.memop & MO_BSWAP) == MO_TE) {
                tcg_gen_concat32_i64(t64, lo, hi);
            } else {
                tcg_gen_concat32_i64(t64, hi, lo);
            }

            save_state(dc);
            gen_helper_st_asi(cpu_env, addr, t64, r_asi, r_mop);
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);
            tcg_temp_free_i64(t64);
        }
        break;
    }
}

static void gen_casx_asi(DisasContext *dc, TCGv addr, TCGv cmpv,
                         int insn, int rd)
{
    DisasASI da = get_asi(dc, insn, MO_TEQ);
    TCGv oldv;

    switch (da.type) {
    case GET_ASI_EXCP:
        return;
    case GET_ASI_DIRECT:
        oldv = tcg_temp_new();
        tcg_gen_atomic_cmpxchg_tl(oldv, addr, cmpv, gen_load_gpr(dc, rd),
                                  da.mem_idx, da.memop);
        gen_store_gpr(dc, rd, oldv);
        tcg_temp_free(oldv);
        break;
    default:
        /* ??? Should be DAE_invalid_asi.  */
        gen_exception(dc, TT_DATA_ACCESS);
        break;
    }
}

#elif !defined(CONFIG_USER_ONLY)
static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd)
{
    /* ??? Work around an apparent bug in Ubuntu gcc 4.8.2-10ubuntu2+12,
       whereby "rd + 1" elicits "error: array subscript is above array".
       Since we have already asserted that rd is even, the semantics
       are unchanged.  */
    TCGv lo = gen_dest_gpr(dc, rd | 1);
    TCGv hi = gen_dest_gpr(dc, rd);
    TCGv_i64 t64 = tcg_temp_new_i64();
    DisasASI da = get_asi(dc, insn, MO_TEQ);

    switch (da.type) {
    case GET_ASI_EXCP:
        tcg_temp_free_i64(t64);
        return;
    case GET_ASI_DIRECT:
        gen_address_mask(dc, addr);
        tcg_gen_qemu_ld_i64(t64, addr, da.mem_idx, da.memop);
        break;
    default:
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(MO_Q);

            save_state(dc);
            gen_helper_ld_asi(t64, cpu_env, addr, r_asi, r_mop);
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);
        }
        break;
    }

    tcg_gen_extr_i64_i32(lo, hi, t64);
    tcg_temp_free_i64(t64);
    gen_store_gpr(dc, rd | 1, lo);
    gen_store_gpr(dc, rd, hi);
}

static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr,
                         int insn, int rd)
{
    DisasASI da = get_asi(dc, insn, MO_TEQ);
    TCGv lo = gen_load_gpr(dc, rd + 1);
    TCGv_i64 t64 = tcg_temp_new_i64();

    tcg_gen_concat_tl_i64(t64, lo, hi);

    switch (da.type) {
    case GET_ASI_EXCP:
        break;
    case GET_ASI_DIRECT:
        gen_address_mask(dc, addr);
        tcg_gen_qemu_st_i64(t64, addr, da.mem_idx, da.memop);
        break;
    case GET_ASI_BFILL:
        /* Store 32 bytes of T64 to ADDR.  */
        /* ??? The original qemu code suggests 8-byte alignment, dropping
           the low bits, but the only place I can see this used is in the
           Linux kernel with 32 byte alignment, which would make more sense
           as a cacheline-style operation.  */
        {
            TCGv d_addr = tcg_temp_new();
            TCGv eight = tcg_const_tl(8);
            int i;

            tcg_gen_andi_tl(d_addr, addr, -8);
            for (i = 0; i < 32; i += 8) {
                tcg_gen_qemu_st_i64(t64, d_addr, da.mem_idx, da.memop);
                tcg_gen_add_tl(d_addr, d_addr, eight);
            }

            tcg_temp_free(d_addr);
            tcg_temp_free(eight);
        }
        break;
    default:
        {
            TCGv_i32 r_asi = tcg_const_i32(da.asi);
            TCGv_i32 r_mop = tcg_const_i32(MO_Q);

            save_state(dc);
            gen_helper_st_asi(cpu_env, addr, t64, r_asi, r_mop);
            tcg_temp_free_i32(r_mop);
            tcg_temp_free_i32(r_asi);
        }
        break;
    }

    tcg_temp_free_i64(t64);
}
#endif

static TCGv get_src1(DisasContext *dc, unsigned int insn)
{
    unsigned int rs1 = GET_FIELD(insn, 13, 17);
    return gen_load_gpr(dc, rs1);
}

static TCGv get_src2(DisasContext *dc, unsigned int insn)
{
    if (IS_IMM) { /* immediate */
        target_long simm = GET_FIELDs(insn, 19, 31);
        TCGv t = get_temp_tl(dc);
        tcg_gen_movi_tl(t, simm);
        return t;
    } else {      /* register */
        unsigned int rs2 = GET_FIELD(insn, 27, 31);
        return gen_load_gpr(dc, rs2);
    }
}

#ifdef TARGET_SPARC64
static void gen_fmovs(DisasContext *dc, DisasCompare *cmp, int rd, int rs)
{
    TCGv_i32 c32, zero, dst, s1, s2;

    /* We have two choices here: extend the 32 bit data and use movcond_i64,
       or fold the comparison down to 32 bits and use movcond_i32.  Choose
       the later.  */
    c32 = tcg_temp_new_i32();
    if (cmp->is_bool) {
        tcg_gen_extrl_i64_i32(c32, cmp->c1);
    } else {
        TCGv_i64 c64 = tcg_temp_new_i64();
        tcg_gen_setcond_i64(cmp->cond, c64, cmp->c1, cmp->c2);
        tcg_gen_extrl_i64_i32(c32, c64);
        tcg_temp_free_i64(c64);
    }

    s1 = gen_load_fpr_F(dc, rs);
    s2 = gen_load_fpr_F(dc, rd);
    dst = gen_dest_fpr_F(dc);
    zero = tcg_const_i32(0);

    tcg_gen_movcond_i32(TCG_COND_NE, dst, c32, zero, s1, s2);

    tcg_temp_free_i32(c32);
    tcg_temp_free_i32(zero);
    gen_store_fpr_F(dc, rd, dst);
}

static void gen_fmovd(DisasContext *dc, DisasCompare *cmp, int rd, int rs)
{
    TCGv_i64 dst = gen_dest_fpr_D(dc, rd);
    tcg_gen_movcond_i64(cmp->cond, dst, cmp->c1, cmp->c2,
                        gen_load_fpr_D(dc, rs),
                        gen_load_fpr_D(dc, rd));
    gen_store_fpr_D(dc, rd, dst);
}

static void gen_fmovq(DisasContext *dc, DisasCompare *cmp, int rd, int rs)
{
    int qd = QFPREG(rd);
    int qs = QFPREG(rs);

    tcg_gen_movcond_i64(cmp->cond, cpu_fpr[qd / 2], cmp->c1, cmp->c2,
                        cpu_fpr[qs / 2], cpu_fpr[qd / 2]);
    tcg_gen_movcond_i64(cmp->cond, cpu_fpr[qd / 2 + 1], cmp->c1, cmp->c2,
                        cpu_fpr[qs / 2 + 1], cpu_fpr[qd / 2 + 1]);

    gen_update_fprs_dirty(dc, qd);
}

#ifndef CONFIG_USER_ONLY
static inline void gen_load_trap_state_at_tl(TCGv_ptr r_tsptr, TCGv_env cpu_env)
{
    TCGv_i32 r_tl = tcg_temp_new_i32();

    /* load env->tl into r_tl */
    tcg_gen_ld_i32(r_tl, cpu_env, offsetof(CPUSPARCState, tl));

    /* tl = [0 ... MAXTL_MASK] where MAXTL_MASK must be power of 2 */
    tcg_gen_andi_i32(r_tl, r_tl, MAXTL_MASK);

    /* calculate offset to current trap state from env->ts, reuse r_tl */
    tcg_gen_muli_i32(r_tl, r_tl, sizeof (trap_state));
    tcg_gen_addi_ptr(r_tsptr, cpu_env, offsetof(CPUSPARCState, ts));

    /* tsptr = env->ts[env->tl & MAXTL_MASK] */
    {