summaryrefslogblamecommitdiffstats
path: root/plugins/api.c
blob: bbdc5a4eb46da159477497d1e8b403fbbf2d64d3 (plain) (tree)








































                                                                      

                          

                        
                               

                      
                      






































































































































































                                                                                





                                                                 




























                                                                   


















                                                                             




                                                                             

      
                                                                     

                     
                        






















                                                                                          





























                                                                    







                                                
/*
 * QEMU Plugin API
 *
 * This provides the API that is available to the plugins to interact
 * with QEMU. We have to be careful not to expose internal details of
 * how QEMU works so we abstract out things like translation and
 * instructions to anonymous data types:
 *
 *  qemu_plugin_tb
 *  qemu_plugin_insn
 *
 * Which can then be passed back into the API to do additional things.
 * As such all the public functions in here are exported in
 * qemu-plugin.h.
 *
 * The general life-cycle of a plugin is:
 *
 *  - plugin is loaded, public qemu_plugin_install called
 *    - the install func registers callbacks for events
 *    - usually an atexit_cb is registered to dump info at the end
 *  - when a registered event occurs the plugin is called
 *     - some events pass additional info
 *     - during translation the plugin can decide to instrument any
 *       instruction
 *  - when QEMU exits all the registered atexit callbacks are called
 *
 * Copyright (C) 2017, Emilio G. Cota <cota@braap.org>
 * Copyright (C) 2019, Linaro
 *
 * License: GNU GPL, version 2 or later.
 *   See the COPYING file in the top-level directory.
 *
 * SPDX-License-Identifier: GPL-2.0-or-later
 *
 */

#include "qemu/osdep.h"
#include "qemu/plugin.h"
#include "cpu.h"
#include "sysemu/sysemu.h"
#include "tcg/tcg.h"
#include "exec/exec-all.h"
#include "disas/disas.h"
#include "plugin.h"
#ifndef CONFIG_USER_ONLY
#include "qemu/plugin-memory.h"
#include "hw/boards.h"
#endif
#include "trace/mem.h"

/* Uninstall and Reset handlers */

void qemu_plugin_uninstall(qemu_plugin_id_t id, qemu_plugin_simple_cb_t cb)
{
    plugin_reset_uninstall(id, cb, false);
}

void qemu_plugin_reset(qemu_plugin_id_t id, qemu_plugin_simple_cb_t cb)
{
    plugin_reset_uninstall(id, cb, true);
}

/*
 * Plugin Register Functions
 *
 * This allows the plugin to register callbacks for various events
 * during the translation.
 */

void qemu_plugin_register_vcpu_init_cb(qemu_plugin_id_t id,
                                       qemu_plugin_vcpu_simple_cb_t cb)
{
    plugin_register_cb(id, QEMU_PLUGIN_EV_VCPU_INIT, cb);
}

void qemu_plugin_register_vcpu_exit_cb(qemu_plugin_id_t id,
                                       qemu_plugin_vcpu_simple_cb_t cb)
{
    plugin_register_cb(id, QEMU_PLUGIN_EV_VCPU_EXIT, cb);
}

void qemu_plugin_register_vcpu_tb_exec_cb(struct qemu_plugin_tb *tb,
                                          qemu_plugin_vcpu_udata_cb_t cb,
                                          enum qemu_plugin_cb_flags flags,
                                          void *udata)
{
    plugin_register_dyn_cb__udata(&tb->cbs[PLUGIN_CB_REGULAR],
                                  cb, flags, udata);
}

void qemu_plugin_register_vcpu_tb_exec_inline(struct qemu_plugin_tb *tb,
                                              enum qemu_plugin_op op,
                                              void *ptr, uint64_t imm)
{
    plugin_register_inline_op(&tb->cbs[PLUGIN_CB_INLINE], 0, op, ptr, imm);
}

void qemu_plugin_register_vcpu_insn_exec_cb(struct qemu_plugin_insn *insn,
                                            qemu_plugin_vcpu_udata_cb_t cb,
                                            enum qemu_plugin_cb_flags flags,
                                            void *udata)
{
    plugin_register_dyn_cb__udata(&insn->cbs[PLUGIN_CB_INSN][PLUGIN_CB_REGULAR],
        cb, flags, udata);
}

void qemu_plugin_register_vcpu_insn_exec_inline(struct qemu_plugin_insn *insn,
                                                enum qemu_plugin_op op,
                                                void *ptr, uint64_t imm)
{
    plugin_register_inline_op(&insn->cbs[PLUGIN_CB_INSN][PLUGIN_CB_INLINE],
                              0, op, ptr, imm);
}



void qemu_plugin_register_vcpu_mem_cb(struct qemu_plugin_insn *insn,
                                      qemu_plugin_vcpu_mem_cb_t cb,
                                      enum qemu_plugin_cb_flags flags,
                                      enum qemu_plugin_mem_rw rw,
                                      void *udata)
{
    plugin_register_vcpu_mem_cb(&insn->cbs[PLUGIN_CB_MEM][PLUGIN_CB_REGULAR],
                                cb, flags, rw, udata);
}

void qemu_plugin_register_vcpu_mem_inline(struct qemu_plugin_insn *insn,
                                          enum qemu_plugin_mem_rw rw,
                                          enum qemu_plugin_op op, void *ptr,
                                          uint64_t imm)
{
    plugin_register_inline_op(&insn->cbs[PLUGIN_CB_MEM][PLUGIN_CB_INLINE],
        rw, op, ptr, imm);
}

void qemu_plugin_register_vcpu_tb_trans_cb(qemu_plugin_id_t id,
                                           qemu_plugin_vcpu_tb_trans_cb_t cb)
{
    plugin_register_cb(id, QEMU_PLUGIN_EV_VCPU_TB_TRANS, cb);
}

void qemu_plugin_register_vcpu_syscall_cb(qemu_plugin_id_t id,
                                          qemu_plugin_vcpu_syscall_cb_t cb)
{
    plugin_register_cb(id, QEMU_PLUGIN_EV_VCPU_SYSCALL, cb);
}

void
qemu_plugin_register_vcpu_syscall_ret_cb(qemu_plugin_id_t id,
                                         qemu_plugin_vcpu_syscall_ret_cb_t cb)
{
    plugin_register_cb(id, QEMU_PLUGIN_EV_VCPU_SYSCALL_RET, cb);
}

/*
 * Plugin Queries
 *
 * These are queries that the plugin can make to gauge information
 * from our opaque data types. We do not want to leak internal details
 * here just information useful to the plugin.
 */

/*
 * Translation block information:
 *
 * A plugin can query the virtual address of the start of the block
 * and the number of instructions in it. It can also get access to
 * each translated instruction.
 */

size_t qemu_plugin_tb_n_insns(const struct qemu_plugin_tb *tb)
{
    return tb->n;
}

uint64_t qemu_plugin_tb_vaddr(const struct qemu_plugin_tb *tb)
{
    return tb->vaddr;
}

struct qemu_plugin_insn *
qemu_plugin_tb_get_insn(const struct qemu_plugin_tb *tb, size_t idx)
{
    if (unlikely(idx >= tb->n)) {
        return NULL;
    }
    return g_ptr_array_index(tb->insns, idx);
}

/*
 * Instruction information
 *
 * These queries allow the plugin to retrieve information about each
 * instruction being translated.
 */

const void *qemu_plugin_insn_data(const struct qemu_plugin_insn *insn)
{
    return insn->data->data;
}

size_t qemu_plugin_insn_size(const struct qemu_plugin_insn *insn)
{
    return insn->data->len;
}

uint64_t qemu_plugin_insn_vaddr(const struct qemu_plugin_insn *insn)
{
    return insn->vaddr;
}

void *qemu_plugin_insn_haddr(const struct qemu_plugin_insn *insn)
{
    return insn->haddr;
}

char *qemu_plugin_insn_disas(const struct qemu_plugin_insn *insn)
{
    CPUState *cpu = current_cpu;
    return plugin_disas(cpu, insn->vaddr, insn->data->len);
}

/*
 * The memory queries allow the plugin to query information about a
 * memory access.
 */

unsigned qemu_plugin_mem_size_shift(qemu_plugin_meminfo_t info)
{
    return info & TRACE_MEM_SZ_SHIFT_MASK;
}

bool qemu_plugin_mem_is_sign_extended(qemu_plugin_meminfo_t info)
{
    return !!(info & TRACE_MEM_SE);
}

bool qemu_plugin_mem_is_big_endian(qemu_plugin_meminfo_t info)
{
    return !!(info & TRACE_MEM_BE);
}

bool qemu_plugin_mem_is_store(qemu_plugin_meminfo_t info)
{
    return !!(info & TRACE_MEM_ST);
}

/*
 * Virtual Memory queries
 */

#ifdef CONFIG_SOFTMMU
static __thread struct qemu_plugin_hwaddr hwaddr_info;

struct qemu_plugin_hwaddr *qemu_plugin_get_hwaddr(qemu_plugin_meminfo_t info,
                                                  uint64_t vaddr)
{
    CPUState *cpu = current_cpu;
    unsigned int mmu_idx = info >> TRACE_MEM_MMU_SHIFT;
    hwaddr_info.is_store = info & TRACE_MEM_ST;

    if (!tlb_plugin_lookup(cpu, vaddr, mmu_idx,
                           info & TRACE_MEM_ST, &hwaddr_info)) {
        error_report("invalid use of qemu_plugin_get_hwaddr");
        return NULL;
    }

    return &hwaddr_info;
}
#else
struct qemu_plugin_hwaddr *qemu_plugin_get_hwaddr(qemu_plugin_meminfo_t info,
                                                  uint64_t vaddr)
{
    return NULL;
}
#endif

bool qemu_plugin_hwaddr_is_io(const struct qemu_plugin_hwaddr *haddr)
{
#ifdef CONFIG_SOFTMMU
    return haddr->is_io;
#else
    return false;
#endif
}

uint64_t qemu_plugin_hwaddr_device_offset(const struct qemu_plugin_hwaddr *haddr)
{
#ifdef CONFIG_SOFTMMU
    if (haddr) {
        if (!haddr->is_io) {
            ram_addr_t ram_addr = qemu_ram_addr_from_host((void *) haddr->v.ram.hostaddr);
            if (ram_addr == RAM_ADDR_INVALID) {
                error_report("Bad ram pointer %"PRIx64"", haddr->v.ram.hostaddr);
                abort();
            }
            return ram_addr;
        } else {
            return haddr->v.io.offset;
        }
    }
#endif
    return 0;
}

/*
 * Queries to the number and potential maximum number of vCPUs there
 * will be. This helps the plugin dimension per-vcpu arrays.
 */

#ifndef CONFIG_USER_ONLY
static MachineState * get_ms(void)
{
    return MACHINE(qdev_get_machine());
}
#endif

int qemu_plugin_n_vcpus(void)
{
#ifdef CONFIG_USER_ONLY
    return -1;
#else
    return get_ms()->smp.cpus;
#endif
}

int qemu_plugin_n_max_vcpus(void)
{
#ifdef CONFIG_USER_ONLY
    return -1;
#else
    return get_ms()->smp.max_cpus;
#endif
}

/*
 * Plugin output
 */
void qemu_plugin_outs(const char *string)
{
    qemu_log_mask(CPU_LOG_PLUGIN, "%s", string);
}
id='n1484' href='#n1484'>1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359
/*
 * Tiny Code Generator for QEMU
 *
 * Copyright (c) 2008 Andrzej Zaborowski
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */

#include "elf.h"
#include "../tcg-pool.c.inc"

int arm_arch = __ARM_ARCH;

#ifndef use_idiv_instructions
bool use_idiv_instructions;
#endif

/* ??? Ought to think about changing CONFIG_SOFTMMU to always defined.  */
#ifdef CONFIG_SOFTMMU
# define USING_SOFTMMU 1
#else
# define USING_SOFTMMU 0
#endif

#ifdef CONFIG_DEBUG_TCG
static const char * const tcg_target_reg_names[TCG_TARGET_NB_REGS] = {
    "%r0",
    "%r1",
    "%r2",
    "%r3",
    "%r4",
    "%r5",
    "%r6",
    "%r7",
    "%r8",
    "%r9",
    "%r10",
    "%r11",
    "%r12",
    "%r13",
    "%r14",
    "%pc",
};
#endif

static const int tcg_target_reg_alloc_order[] = {
    TCG_REG_R4,
    TCG_REG_R5,
    TCG_REG_R6,
    TCG_REG_R7,
    TCG_REG_R8,
    TCG_REG_R9,
    TCG_REG_R10,
    TCG_REG_R11,
    TCG_REG_R13,
    TCG_REG_R0,
    TCG_REG_R1,
    TCG_REG_R2,
    TCG_REG_R3,
    TCG_REG_R12,
    TCG_REG_R14,
};

static const int tcg_target_call_iarg_regs[4] = {
    TCG_REG_R0, TCG_REG_R1, TCG_REG_R2, TCG_REG_R3
};
static const int tcg_target_call_oarg_regs[2] = {
    TCG_REG_R0, TCG_REG_R1
};

#define TCG_REG_TMP  TCG_REG_R12

enum arm_cond_code_e {
    COND_EQ = 0x0,
    COND_NE = 0x1,
    COND_CS = 0x2,	/* Unsigned greater or equal */
    COND_CC = 0x3,	/* Unsigned less than */
    COND_MI = 0x4,	/* Negative */
    COND_PL = 0x5,	/* Zero or greater */
    COND_VS = 0x6,	/* Overflow */
    COND_VC = 0x7,	/* No overflow */
    COND_HI = 0x8,	/* Unsigned greater than */
    COND_LS = 0x9,	/* Unsigned less or equal */
    COND_GE = 0xa,
    COND_LT = 0xb,
    COND_GT = 0xc,
    COND_LE = 0xd,
    COND_AL = 0xe,
};

#define TO_CPSR (1 << 20)

#define SHIFT_IMM_LSL(im)	(((im) << 7) | 0x00)
#define SHIFT_IMM_LSR(im)	(((im) << 7) | 0x20)
#define SHIFT_IMM_ASR(im)	(((im) << 7) | 0x40)
#define SHIFT_IMM_ROR(im)	(((im) << 7) | 0x60)
#define SHIFT_REG_LSL(rs)	(((rs) << 8) | 0x10)
#define SHIFT_REG_LSR(rs)	(((rs) << 8) | 0x30)
#define SHIFT_REG_ASR(rs)	(((rs) << 8) | 0x50)
#define SHIFT_REG_ROR(rs)	(((rs) << 8) | 0x70)

typedef enum {
    ARITH_AND = 0x0 << 21,
    ARITH_EOR = 0x1 << 21,
    ARITH_SUB = 0x2 << 21,
    ARITH_RSB = 0x3 << 21,
    ARITH_ADD = 0x4 << 21,
    ARITH_ADC = 0x5 << 21,
    ARITH_SBC = 0x6 << 21,
    ARITH_RSC = 0x7 << 21,
    ARITH_TST = 0x8 << 21 | TO_CPSR,
    ARITH_CMP = 0xa << 21 | TO_CPSR,
    ARITH_CMN = 0xb << 21 | TO_CPSR,
    ARITH_ORR = 0xc << 21,
    ARITH_MOV = 0xd << 21,
    ARITH_BIC = 0xe << 21,
    ARITH_MVN = 0xf << 21,

    INSN_CLZ       = 0x016f0f10,
    INSN_RBIT      = 0x06ff0f30,

    INSN_LDR_IMM   = 0x04100000,
    INSN_LDR_REG   = 0x06100000,
    INSN_STR_IMM   = 0x04000000,
    INSN_STR_REG   = 0x06000000,

    INSN_LDRH_IMM  = 0x005000b0,
    INSN_LDRH_REG  = 0x001000b0,
    INSN_LDRSH_IMM = 0x005000f0,
    INSN_LDRSH_REG = 0x001000f0,
    INSN_STRH_IMM  = 0x004000b0,
    INSN_STRH_REG  = 0x000000b0,

    INSN_LDRB_IMM  = 0x04500000,
    INSN_LDRB_REG  = 0x06500000,
    INSN_LDRSB_IMM = 0x005000d0,
    INSN_LDRSB_REG = 0x001000d0,
    INSN_STRB_IMM  = 0x04400000,
    INSN_STRB_REG  = 0x06400000,

    INSN_LDRD_IMM  = 0x004000d0,
    INSN_LDRD_REG  = 0x000000d0,
    INSN_STRD_IMM  = 0x004000f0,
    INSN_STRD_REG  = 0x000000f0,

    INSN_DMB_ISH   = 0xf57ff05b,
    INSN_DMB_MCR   = 0xee070fba,

    /* Architected nop introduced in v6k.  */
    /* ??? This is an MSR (imm) 0,0,0 insn.  Anyone know if this
       also Just So Happened to do nothing on pre-v6k so that we
       don't need to conditionalize it?  */
    INSN_NOP_v6k   = 0xe320f000,
    /* Otherwise the assembler uses mov r0,r0 */
    INSN_NOP_v4    = (COND_AL << 28) | ARITH_MOV,
} ARMInsn;

#define INSN_NOP   (use_armv7_instructions ? INSN_NOP_v6k : INSN_NOP_v4)

static const uint8_t tcg_cond_to_arm_cond[] = {
    [TCG_COND_EQ] = COND_EQ,
    [TCG_COND_NE] = COND_NE,
    [TCG_COND_LT] = COND_LT,
    [TCG_COND_GE] = COND_GE,
    [TCG_COND_LE] = COND_LE,
    [TCG_COND_GT] = COND_GT,
    /* unsigned */
    [TCG_COND_LTU] = COND_CC,
    [TCG_COND_GEU] = COND_CS,
    [TCG_COND_LEU] = COND_LS,
    [TCG_COND_GTU] = COND_HI,
};

static inline bool reloc_pc24(tcg_insn_unit *code_ptr, tcg_insn_unit *target)
{
    ptrdiff_t offset = (tcg_ptr_byte_diff(target, code_ptr) - 8) >> 2;
    if (offset == sextract32(offset, 0, 24)) {
        *code_ptr = (*code_ptr & ~0xffffff) | (offset & 0xffffff);
        return true;
    }
    return false;
}

static inline bool reloc_pc13(tcg_insn_unit *code_ptr, tcg_insn_unit *target)
{
    ptrdiff_t offset = tcg_ptr_byte_diff(target, code_ptr) - 8;

    if (offset >= -0xfff && offset <= 0xfff) {
        tcg_insn_unit insn = *code_ptr;
        bool u = (offset >= 0);
        if (!u) {
            offset = -offset;
        }
        insn = deposit32(insn, 23, 1, u);
        insn = deposit32(insn, 0, 12, offset);
        *code_ptr = insn;
        return true;
    }
    return false;
}

static bool patch_reloc(tcg_insn_unit *code_ptr, int type,
                        intptr_t value, intptr_t addend)
{
    tcg_debug_assert(addend == 0);

    if (type == R_ARM_PC24) {
        return reloc_pc24(code_ptr, (tcg_insn_unit *)value);
    } else if (type == R_ARM_PC13) {
        return reloc_pc13(code_ptr, (tcg_insn_unit *)value);
    } else {
        g_assert_not_reached();
    }
}

#define TCG_CT_CONST_ARM  0x100
#define TCG_CT_CONST_INV  0x200
#define TCG_CT_CONST_NEG  0x400
#define TCG_CT_CONST_ZERO 0x800

/* parse target specific constraints */
static const char *target_parse_constraint(TCGArgConstraint *ct,
                                           const char *ct_str, TCGType type)
{
    switch (*ct_str++) {
    case 'I':
        ct->ct |= TCG_CT_CONST_ARM;
        break;
    case 'K':
        ct->ct |= TCG_CT_CONST_INV;
        break;
    case 'N': /* The gcc constraint letter is L, already used here.  */
        ct->ct |= TCG_CT_CONST_NEG;
        break;
    case 'Z':
        ct->ct |= TCG_CT_CONST_ZERO;
        break;

    case 'r':
        ct->regs = 0xffff;
        break;

    /* qemu_ld address */
    case 'l':
        ct->regs = 0xffff;
#ifdef CONFIG_SOFTMMU
        /* r0-r2,lr will be overwritten when reading the tlb entry,
           so don't use these. */
        tcg_regset_reset_reg(ct->regs, TCG_REG_R0);
        tcg_regset_reset_reg(ct->regs, TCG_REG_R1);
        tcg_regset_reset_reg(ct->regs, TCG_REG_R2);
        tcg_regset_reset_reg(ct->regs, TCG_REG_R3);
        tcg_regset_reset_reg(ct->regs, TCG_REG_R14);
#endif
        break;

    /* qemu_st address & data */
    case 's':
        ct->regs = 0xffff;
        /* r0-r2 will be overwritten when reading the tlb entry (softmmu only)
           and r0-r1 doing the byte swapping, so don't use these. */
        tcg_regset_reset_reg(ct->regs, TCG_REG_R0);
        tcg_regset_reset_reg(ct->regs, TCG_REG_R1);
#if defined(CONFIG_SOFTMMU)
        /* Avoid clashes with registers being used for helper args */
        tcg_regset_reset_reg(ct->regs, TCG_REG_R2);
#if TARGET_LONG_BITS == 64
        /* Avoid clashes with registers being used for helper args */
        tcg_regset_reset_reg(ct->regs, TCG_REG_R3);
#endif
        tcg_regset_reset_reg(ct->regs, TCG_REG_R14);
#endif
        break;

    default:
        return NULL;
    }
    return ct_str;
}

static inline uint32_t rotl(uint32_t val, int n)
{
  return (val << n) | (val >> (32 - n));
}

/* ARM immediates for ALU instructions are made of an unsigned 8-bit
   right-rotated by an even amount between 0 and 30. */
static inline int encode_imm(uint32_t imm)
{
    int shift;

    /* simple case, only lower bits */
    if ((imm & ~0xff) == 0)
        return 0;
    /* then try a simple even shift */
    shift = ctz32(imm) & ~1;
    if (((imm >> shift) & ~0xff) == 0)
        return 32 - shift;
    /* now try harder with rotations */
    if ((rotl(imm, 2) & ~0xff) == 0)
        return 2;
    if ((rotl(imm, 4) & ~0xff) == 0)
        return 4;
    if ((rotl(imm, 6) & ~0xff) == 0)
        return 6;
    /* imm can't be encoded */
    return -1;
}

static inline int check_fit_imm(uint32_t imm)
{
    return encode_imm(imm) >= 0;
}

/* Test if a constant matches the constraint.
 * TODO: define constraints for:
 *
 * ldr/str offset:   between -0xfff and 0xfff
 * ldrh/strh offset: between -0xff and 0xff
 * mov operand2:     values represented with x << (2 * y), x < 0x100
 * add, sub, eor...: ditto
 */
static inline int tcg_target_const_match(tcg_target_long val, TCGType type,
                                         const TCGArgConstraint *arg_ct)
{
    int ct;
    ct = arg_ct->ct;
    if (ct & TCG_CT_CONST) {
        return 1;
    } else if ((ct & TCG_CT_CONST_ARM) && check_fit_imm(val)) {
        return 1;
    } else if ((ct & TCG_CT_CONST_INV) && check_fit_imm(~val)) {
        return 1;
    } else if ((ct & TCG_CT_CONST_NEG) && check_fit_imm(-val)) {
        return 1;
    } else if ((ct & TCG_CT_CONST_ZERO) && val == 0) {
        return 1;
    } else {
        return 0;
    }
}

static inline void tcg_out_b(TCGContext *s, int cond, int32_t offset)
{
    tcg_out32(s, (cond << 28) | 0x0a000000 |
                    (((offset - 8) >> 2) & 0x00ffffff));
}

static inline void tcg_out_bl(TCGContext *s, int cond, int32_t offset)
{
    tcg_out32(s, (cond << 28) | 0x0b000000 |
                    (((offset - 8) >> 2) & 0x00ffffff));
}

static inline void tcg_out_blx(TCGContext *s, int cond, int rn)
{
    tcg_out32(s, (cond << 28) | 0x012fff30 | rn);
}

static inline void tcg_out_blx_imm(TCGContext *s, int32_t offset)
{
    tcg_out32(s, 0xfa000000 | ((offset & 2) << 23) |
                (((offset - 8) >> 2) & 0x00ffffff));
}

static inline void tcg_out_dat_reg(TCGContext *s,
                int cond, int opc, int rd, int rn, int rm, int shift)
{
    tcg_out32(s, (cond << 28) | (0 << 25) | opc |
                    (rn << 16) | (rd << 12) | shift | rm);
}

static inline void tcg_out_nop(TCGContext *s)
{
    tcg_out32(s, INSN_NOP);
}

static inline void tcg_out_mov_reg(TCGContext *s, int cond, int rd, int rm)
{
    /* Simple reg-reg move, optimising out the 'do nothing' case */
    if (rd != rm) {
        tcg_out_dat_reg(s, cond, ARITH_MOV, rd, 0, rm, SHIFT_IMM_LSL(0));
    }
}

static inline void tcg_out_bx(TCGContext *s, int cond, TCGReg rn)
{
    /* Unless the C portion of QEMU is compiled as thumb, we don't
       actually need true BX semantics; merely a branch to an address
       held in a register.  */
    if (use_armv5t_instructions) {
        tcg_out32(s, (cond << 28) | 0x012fff10 | rn);
    } else {
        tcg_out_mov_reg(s, cond, TCG_REG_PC, rn);
    }
}

static inline void tcg_out_dat_imm(TCGContext *s,
                int cond, int opc, int rd, int rn, int im)
{
    tcg_out32(s, (cond << 28) | (1 << 25) | opc |
                    (rn << 16) | (rd << 12) | im);
}

/* Note that this routine is used for both LDR and LDRH formats, so we do
   not wish to include an immediate shift at this point.  */
static void tcg_out_memop_r(TCGContext *s, int cond, ARMInsn opc, TCGReg rt,
                            TCGReg rn, TCGReg rm, bool u, bool p, bool w)
{
    tcg_out32(s, (cond << 28) | opc | (u << 23) | (p << 24)
              | (w << 21) | (rn << 16) | (rt << 12) | rm);
}

static void tcg_out_memop_8(TCGContext *s, int cond, ARMInsn opc, TCGReg rt,
                            TCGReg rn, int imm8, bool p, bool w)
{
    bool u = 1;
    if (imm8 < 0) {
        imm8 = -imm8;
        u = 0;
    }
    tcg_out32(s, (cond << 28) | opc | (u << 23) | (p << 24) | (w << 21) |
              (rn << 16) | (rt << 12) | ((imm8 & 0xf0) << 4) | (imm8 & 0xf));
}

static void tcg_out_memop_12(TCGContext *s, int cond, ARMInsn opc, TCGReg rt,
                             TCGReg rn, int imm12, bool p, bool w)
{
    bool u = 1;
    if (imm12 < 0) {
        imm12 = -imm12;
        u = 0;
    }
    tcg_out32(s, (cond << 28) | opc | (u << 23) | (p << 24) | (w << 21) |
              (rn << 16) | (rt << 12) | imm12);
}

static inline void tcg_out_ld32_12(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, int imm12)
{
    tcg_out_memop_12(s, cond, INSN_LDR_IMM, rt, rn, imm12, 1, 0);
}

static inline void tcg_out_st32_12(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, int imm12)
{
    tcg_out_memop_12(s, cond, INSN_STR_IMM, rt, rn, imm12, 1, 0);
}

static inline void tcg_out_ld32_r(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDR_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_st32_r(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_STR_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_ldrd_8(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, int imm8)
{
    tcg_out_memop_8(s, cond, INSN_LDRD_IMM, rt, rn, imm8, 1, 0);
}

static inline void tcg_out_ldrd_r(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDRD_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_ldrd_rwb(TCGContext *s, int cond, TCGReg rt,
                                    TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDRD_REG, rt, rn, rm, 1, 1, 1);
}

static inline void tcg_out_strd_8(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, int imm8)
{
    tcg_out_memop_8(s, cond, INSN_STRD_IMM, rt, rn, imm8, 1, 0);
}

static inline void tcg_out_strd_r(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_STRD_REG, rt, rn, rm, 1, 1, 0);
}

/* Register pre-increment with base writeback.  */
static inline void tcg_out_ld32_rwb(TCGContext *s, int cond, TCGReg rt,
                                    TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDR_REG, rt, rn, rm, 1, 1, 1);
}

static inline void tcg_out_st32_rwb(TCGContext *s, int cond, TCGReg rt,
                                    TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_STR_REG, rt, rn, rm, 1, 1, 1);
}

static inline void tcg_out_ld16u_8(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, int imm8)
{
    tcg_out_memop_8(s, cond, INSN_LDRH_IMM, rt, rn, imm8, 1, 0);
}

static inline void tcg_out_st16_8(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, int imm8)
{
    tcg_out_memop_8(s, cond, INSN_STRH_IMM, rt, rn, imm8, 1, 0);
}

static inline void tcg_out_ld16u_r(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDRH_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_st16_r(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_STRH_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_ld16s_8(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, int imm8)
{
    tcg_out_memop_8(s, cond, INSN_LDRSH_IMM, rt, rn, imm8, 1, 0);
}

static inline void tcg_out_ld16s_r(TCGContext *s, int cond, TCGReg rt,
                                   TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDRSH_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_ld8_12(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, int imm12)
{
    tcg_out_memop_12(s, cond, INSN_LDRB_IMM, rt, rn, imm12, 1, 0);
}

static inline void tcg_out_st8_12(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, int imm12)
{
    tcg_out_memop_12(s, cond, INSN_STRB_IMM, rt, rn, imm12, 1, 0);
}

static inline void tcg_out_ld8_r(TCGContext *s, int cond, TCGReg rt,
                                 TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDRB_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_st8_r(TCGContext *s, int cond, TCGReg rt,
                                 TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_STRB_REG, rt, rn, rm, 1, 1, 0);
}

static inline void tcg_out_ld8s_8(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, int imm8)
{
    tcg_out_memop_8(s, cond, INSN_LDRSB_IMM, rt, rn, imm8, 1, 0);
}

static inline void tcg_out_ld8s_r(TCGContext *s, int cond, TCGReg rt,
                                  TCGReg rn, TCGReg rm)
{
    tcg_out_memop_r(s, cond, INSN_LDRSB_REG, rt, rn, rm, 1, 1, 0);
}

static void tcg_out_movi_pool(TCGContext *s, int cond, int rd, uint32_t arg)
{
    new_pool_label(s, arg, R_ARM_PC13, s->code_ptr, 0);
    tcg_out_ld32_12(s, cond, rd, TCG_REG_PC, 0);
}

static void tcg_out_movi32(TCGContext *s, int cond, int rd, uint32_t arg)
{
    int rot, diff, opc, sh1, sh2;
    uint32_t tt0, tt1, tt2;

    /* Check a single MOV/MVN before anything else.  */
    rot = encode_imm(arg);
    if (rot >= 0) {
        tcg_out_dat_imm(s, cond, ARITH_MOV, rd, 0,
                        rotl(arg, rot) | (rot << 7));
        return;
    }
    rot = encode_imm(~arg);
    if (rot >= 0) {
        tcg_out_dat_imm(s, cond, ARITH_MVN, rd, 0,
                        rotl(~arg, rot) | (rot << 7));
        return;
    }

    /* Check for a pc-relative address.  This will usually be the TB,
       or within the TB, which is immediately before the code block.  */
    diff = arg - ((intptr_t)s->code_ptr + 8);
    if (diff >= 0) {
        rot = encode_imm(diff);
        if (rot >= 0) {
            tcg_out_dat_imm(s, cond, ARITH_ADD, rd, TCG_REG_PC,
                            rotl(diff, rot) | (rot << 7));
            return;
        }
    } else {
        rot = encode_imm(-diff);
        if (rot >= 0) {
            tcg_out_dat_imm(s, cond, ARITH_SUB, rd, TCG_REG_PC,
                            rotl(-diff, rot) | (rot << 7));
            return;
        }
    }

    /* Use movw + movt.  */
    if (use_armv7_instructions) {
        /* movw */
        tcg_out32(s, (cond << 28) | 0x03000000 | (rd << 12)
                  | ((arg << 4) & 0x000f0000) | (arg & 0xfff));
        if (arg & 0xffff0000) {
            /* movt */
            tcg_out32(s, (cond << 28) | 0x03400000 | (rd << 12)
                      | ((arg >> 12) & 0x000f0000) | ((arg >> 16) & 0xfff));
        }
        return;
    }

    /* Look for sequences of two insns.  If we have lots of 1's, we can
       shorten the sequence by beginning with mvn and then clearing
       higher bits with eor.  */
    tt0 = arg;
    opc = ARITH_MOV;
    if (ctpop32(arg) > 16) {
        tt0 = ~arg;
        opc = ARITH_MVN;
    }
    sh1 = ctz32(tt0) & ~1;
    tt1 = tt0 & ~(0xff << sh1);
    sh2 = ctz32(tt1) & ~1;
    tt2 = tt1 & ~(0xff << sh2);
    if (tt2 == 0) {
        rot = ((32 - sh1) << 7) & 0xf00;
        tcg_out_dat_imm(s, cond, opc, rd,  0, ((tt0 >> sh1) & 0xff) | rot);
        rot = ((32 - sh2) << 7) & 0xf00;
        tcg_out_dat_imm(s, cond, ARITH_EOR, rd, rd,
                        ((tt0 >> sh2) & 0xff) | rot);
        return;
    }

    /* Otherwise, drop it into the constant pool.  */
    tcg_out_movi_pool(s, cond, rd, arg);
}

static inline void tcg_out_dat_rI(TCGContext *s, int cond, int opc, TCGArg dst,
                                  TCGArg lhs, TCGArg rhs, int rhs_is_const)
{
    /* Emit either the reg,imm or reg,reg form of a data-processing insn.
     * rhs must satisfy the "rI" constraint.
     */
    if (rhs_is_const) {
        int rot = encode_imm(rhs);
        tcg_debug_assert(rot >= 0);
        tcg_out_dat_imm(s, cond, opc, dst, lhs, rotl(rhs, rot) | (rot << 7));
    } else {
        tcg_out_dat_reg(s, cond, opc, dst, lhs, rhs, SHIFT_IMM_LSL(0));
    }
}

static void tcg_out_dat_rIK(TCGContext *s, int cond, int opc, int opinv,
                            TCGReg dst, TCGReg lhs, TCGArg rhs,
                            bool rhs_is_const)
{
    /* Emit either the reg,imm or reg,reg form of a data-processing insn.
     * rhs must satisfy the "rIK" constraint.
     */
    if (rhs_is_const) {
        int rot = encode_imm(rhs);
        if (rot < 0) {
            rhs = ~rhs;
            rot = encode_imm(rhs);
            tcg_debug_assert(rot >= 0);
            opc = opinv;
        }
        tcg_out_dat_imm(s, cond, opc, dst, lhs, rotl(rhs, rot) | (rot << 7));
    } else {
        tcg_out_dat_reg(s, cond, opc, dst, lhs, rhs, SHIFT_IMM_LSL(0));
    }
}

static void tcg_out_dat_rIN(TCGContext *s, int cond, int opc, int opneg,
                            TCGArg dst, TCGArg lhs, TCGArg rhs,
                            bool rhs_is_const)
{
    /* Emit either the reg,imm or reg,reg form of a data-processing insn.
     * rhs must satisfy the "rIN" constraint.
     */
    if (rhs_is_const) {
        int rot = encode_imm(rhs);
        if (rot < 0) {
            rhs = -rhs;
            rot = encode_imm(rhs);
            tcg_debug_assert(rot >= 0);
            opc = opneg;
        }
        tcg_out_dat_imm(s, cond, opc, dst, lhs, rotl(rhs, rot) | (rot << 7));
    } else {
        tcg_out_dat_reg(s, cond, opc, dst, lhs, rhs, SHIFT_IMM_LSL(0));
    }
}

static inline void tcg_out_mul32(TCGContext *s, int cond, TCGReg rd,
                                 TCGReg rn, TCGReg rm)
{
    /* if ArchVersion() < 6 && d == n then UNPREDICTABLE;  */
    if (!use_armv6_instructions && rd == rn) {
        if (rd == rm) {
            /* rd == rn == rm; copy an input to tmp first.  */
            tcg_out_mov_reg(s, cond, TCG_REG_TMP, rn);
            rm = rn = TCG_REG_TMP;
        } else {
            rn = rm;
            rm = rd;
        }
    }
    /* mul */
    tcg_out32(s, (cond << 28) | 0x90 | (rd << 16) | (rm << 8) | rn);
}

static inline void tcg_out_umull32(TCGContext *s, int cond, TCGReg rd0,
                                   TCGReg rd1, TCGReg rn, TCGReg rm)
{
    /* if ArchVersion() < 6 && (dHi == n || dLo == n) then UNPREDICTABLE;  */
    if (!use_armv6_instructions && (rd0 == rn || rd1 == rn)) {
        if (rd0 == rm || rd1 == rm) {
            tcg_out_mov_reg(s, cond, TCG_REG_TMP, rn);
            rn = TCG_REG_TMP;
        } else {
            TCGReg t = rn;
            rn = rm;
            rm = t;
        }
    }
    /* umull */
    tcg_out32(s, (cond << 28) | 0x00800090 |
              (rd1 << 16) | (rd0 << 12) | (rm << 8) | rn);
}

static inline void tcg_out_smull32(TCGContext *s, int cond, TCGReg rd0,
                                   TCGReg rd1, TCGReg rn, TCGReg rm)
{
    /* if ArchVersion() < 6 && (dHi == n || dLo == n) then UNPREDICTABLE;  */
    if (!use_armv6_instructions && (rd0 == rn || rd1 == rn)) {
        if (rd0 == rm || rd1 == rm) {
            tcg_out_mov_reg(s, cond, TCG_REG_TMP, rn);
            rn = TCG_REG_TMP;
        } else {
            TCGReg t = rn;
            rn = rm;
            rm = t;
        }
    }
    /* smull */
    tcg_out32(s, (cond << 28) | 0x00c00090 |
              (rd1 << 16) | (rd0 << 12) | (rm << 8) | rn);
}

static inline void tcg_out_sdiv(TCGContext *s, int cond, int rd, int rn, int rm)
{
    tcg_out32(s, 0x0710f010 | (cond << 28) | (rd << 16) | rn | (rm << 8));
}

static inline void tcg_out_udiv(TCGContext *s, int cond, int rd, int rn, int rm)
{
    tcg_out32(s, 0x0730f010 | (cond << 28) | (rd << 16) | rn | (rm << 8));
}

static inline void tcg_out_ext8s(TCGContext *s, int cond,
                                 int rd, int rn)
{
    if (use_armv6_instructions) {
        /* sxtb */
        tcg_out32(s, 0x06af0070 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rn, SHIFT_IMM_LSL(24));
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rd, SHIFT_IMM_ASR(24));
    }
}

static inline void tcg_out_ext8u(TCGContext *s, int cond,
                                 int rd, int rn)
{
    tcg_out_dat_imm(s, cond, ARITH_AND, rd, rn, 0xff);
}

static inline void tcg_out_ext16s(TCGContext *s, int cond,
                                  int rd, int rn)
{
    if (use_armv6_instructions) {
        /* sxth */
        tcg_out32(s, 0x06bf0070 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rn, SHIFT_IMM_LSL(16));
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rd, SHIFT_IMM_ASR(16));
    }
}

static inline void tcg_out_ext16u(TCGContext *s, int cond,
                                  int rd, int rn)
{
    if (use_armv6_instructions) {
        /* uxth */
        tcg_out32(s, 0x06ff0070 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rn, SHIFT_IMM_LSL(16));
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rd, SHIFT_IMM_LSR(16));
    }
}

static inline void tcg_out_bswap16s(TCGContext *s, int cond, int rd, int rn)
{
    if (use_armv6_instructions) {
        /* revsh */
        tcg_out32(s, 0x06ff0fb0 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        TCG_REG_TMP, 0, rn, SHIFT_IMM_LSL(24));
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        TCG_REG_TMP, 0, TCG_REG_TMP, SHIFT_IMM_ASR(16));
        tcg_out_dat_reg(s, cond, ARITH_ORR,
                        rd, TCG_REG_TMP, rn, SHIFT_IMM_LSR(8));
    }
}

static inline void tcg_out_bswap16(TCGContext *s, int cond, int rd, int rn)
{
    if (use_armv6_instructions) {
        /* rev16 */
        tcg_out32(s, 0x06bf0fb0 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        TCG_REG_TMP, 0, rn, SHIFT_IMM_LSL(24));
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        TCG_REG_TMP, 0, TCG_REG_TMP, SHIFT_IMM_LSR(16));
        tcg_out_dat_reg(s, cond, ARITH_ORR,
                        rd, TCG_REG_TMP, rn, SHIFT_IMM_LSR(8));
    }
}

/* swap the two low bytes assuming that the two high input bytes and the
   two high output bit can hold any value. */
static inline void tcg_out_bswap16st(TCGContext *s, int cond, int rd, int rn)
{
    if (use_armv6_instructions) {
        /* rev16 */
        tcg_out32(s, 0x06bf0fb0 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        TCG_REG_TMP, 0, rn, SHIFT_IMM_LSR(8));
        tcg_out_dat_imm(s, cond, ARITH_AND, TCG_REG_TMP, TCG_REG_TMP, 0xff);
        tcg_out_dat_reg(s, cond, ARITH_ORR,
                        rd, TCG_REG_TMP, rn, SHIFT_IMM_LSL(8));
    }
}

static inline void tcg_out_bswap32(TCGContext *s, int cond, int rd, int rn)
{
    if (use_armv6_instructions) {
        /* rev */
        tcg_out32(s, 0x06bf0f30 | (cond << 28) | (rd << 12) | rn);
    } else {
        tcg_out_dat_reg(s, cond, ARITH_EOR,
                        TCG_REG_TMP, rn, rn, SHIFT_IMM_ROR(16));
        tcg_out_dat_imm(s, cond, ARITH_BIC,
                        TCG_REG_TMP, TCG_REG_TMP, 0xff | 0x800);
        tcg_out_dat_reg(s, cond, ARITH_MOV,
                        rd, 0, rn, SHIFT_IMM_ROR(8));
        tcg_out_dat_reg(s, cond, ARITH_EOR,
                        rd, rd, TCG_REG_TMP, SHIFT_IMM_LSR(8));
    }
}

static inline void tcg_out_deposit(TCGContext *s, int cond, TCGReg rd,
                                   TCGArg a1, int ofs, int len, bool const_a1)
{
    if (const_a1) {
        /* bfi becomes bfc with rn == 15.  */
        a1 = 15;
    }
    /* bfi/bfc */
    tcg_out32(s, 0x07c00010 | (cond << 28) | (rd << 12) | a1
              | (ofs << 7) | ((ofs + len - 1) << 16));
}

static inline void tcg_out_extract(TCGContext *s, int cond, TCGReg rd,
                                   TCGArg a1, int ofs, int len)
{
    /* ubfx */
    tcg_out32(s, 0x07e00050 | (cond << 28) | (rd << 12) | a1
              | (ofs << 7) | ((len - 1) << 16));
}

static inline void tcg_out_sextract(TCGContext *s, int cond, TCGReg rd,
                                    TCGArg a1, int ofs, int len)
{
    /* sbfx */
    tcg_out32(s, 0x07a00050 | (cond << 28) | (rd << 12) | a1
              | (ofs << 7) | ((len - 1) << 16));
}

static inline void tcg_out_ld32u(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xfff || offset < -0xfff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_ld32_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_ld32_12(s, cond, rd, rn, offset);
}

static inline void tcg_out_st32(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xfff || offset < -0xfff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_st32_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_st32_12(s, cond, rd, rn, offset);
}

static inline void tcg_out_ld16u(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xff || offset < -0xff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_ld16u_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_ld16u_8(s, cond, rd, rn, offset);
}

static inline void tcg_out_ld16s(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xff || offset < -0xff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_ld16s_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_ld16s_8(s, cond, rd, rn, offset);
}

static inline void tcg_out_st16(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xff || offset < -0xff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_st16_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_st16_8(s, cond, rd, rn, offset);
}

static inline void tcg_out_ld8u(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xfff || offset < -0xfff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_ld8_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_ld8_12(s, cond, rd, rn, offset);
}

static inline void tcg_out_ld8s(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xff || offset < -0xff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_ld8s_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_ld8s_8(s, cond, rd, rn, offset);
}

static inline void tcg_out_st8(TCGContext *s, int cond,
                int rd, int rn, int32_t offset)
{
    if (offset > 0xfff || offset < -0xfff) {
        tcg_out_movi32(s, cond, TCG_REG_TMP, offset);
        tcg_out_st8_r(s, cond, rd, rn, TCG_REG_TMP);
    } else
        tcg_out_st8_12(s, cond, rd, rn, offset);
}

/* The _goto case is normally between TBs within the same code buffer, and
 * with the code buffer limited to 16MB we wouldn't need the long case.
 * But we also use it for the tail-call to the qemu_ld/st helpers, which does.
 */
static void tcg_out_goto(TCGContext *s, int cond, tcg_insn_unit *addr)
{
    intptr_t addri = (intptr_t)addr;
    ptrdiff_t disp = tcg_pcrel_diff(s, addr);

    if ((addri & 1) == 0 && disp - 8 < 0x01fffffd && disp - 8 > -0x01fffffd) {
        tcg_out_b(s, cond, disp);
        return;
    }
    tcg_out_movi_pool(s, cond, TCG_REG_PC, addri);
}

/* The call case is mostly used for helpers - so it's not unreasonable
 * for them to be beyond branch range */
static void tcg_out_call(TCGContext *s, tcg_insn_unit *addr)
{
    intptr_t addri = (intptr_t)addr;
    ptrdiff_t disp = tcg_pcrel_diff(s, addr);

    if (disp - 8 < 0x02000000 && disp - 8 >= -0x02000000) {
        if (addri & 1) {
            /* Use BLX if the target is in Thumb mode */
            if (!use_armv5t_instructions) {
                tcg_abort();
            }
            tcg_out_blx_imm(s, disp);
        } else {
            tcg_out_bl(s, COND_AL, disp);
        }
    } else if (use_armv7_instructions) {
        tcg_out_movi32(s, COND_AL, TCG_REG_TMP, addri);
        tcg_out_blx(s, COND_AL, TCG_REG_TMP);
    } else {
        /* ??? Know that movi_pool emits exactly 1 insn.  */
        tcg_out_dat_imm(s, COND_AL, ARITH_ADD, TCG_REG_R14, TCG_REG_PC, 0);
        tcg_out_movi_pool(s, COND_AL, TCG_REG_PC, addri);
    }
}

static inline void tcg_out_goto_label(TCGContext *s, int cond, TCGLabel *l)
{
    if (l->has_value) {
        tcg_out_goto(s, cond, l->u.value_ptr);
    } else {
        tcg_out_reloc(s, s->code_ptr, R_ARM_PC24, l, 0);
        tcg_out_b(s, cond, 0);
    }
}

static inline void tcg_out_mb(TCGContext *s, TCGArg a0)
{
    if (use_armv7_instructions) {
        tcg_out32(s, INSN_DMB_ISH);
    } else if (use_armv6_instructions) {
        tcg_out32(s, INSN_DMB_MCR);
    }
}

static TCGCond tcg_out_cmp2(TCGContext *s, const TCGArg *args,
                            const int *const_args)
{
    TCGReg al = args[0];
    TCGReg ah = args[1];
    TCGArg bl = args[2];
    TCGArg bh = args[3];
    TCGCond cond = args[4];
    int const_bl = const_args[2];
    int const_bh = const_args[3];

    switch (cond) {
    case TCG_COND_EQ:
    case TCG_COND_NE:
    case TCG_COND_LTU:
    case TCG_COND_LEU:
    case TCG_COND_GTU:
    case TCG_COND_GEU:
        /* We perform a conditional comparision.  If the high half is
           equal, then overwrite the flags with the comparison of the
           low half.  The resulting flags cover the whole.  */
        tcg_out_dat_rI(s, COND_AL, ARITH_CMP, 0, ah, bh, const_bh);
        tcg_out_dat_rI(s, COND_EQ, ARITH_CMP, 0, al, bl, const_bl);
        return cond;

    case TCG_COND_LT:
    case TCG_COND_GE:
        /* We perform a double-word subtraction and examine the result.
           We do not actually need the result of the subtract, so the
           low part "subtract" is a compare.  For the high half we have
           no choice but to compute into a temporary.  */
        tcg_out_dat_rI(s, COND_AL, ARITH_CMP, 0, al, bl, const_bl);
        tcg_out_dat_rI(s, COND_AL, ARITH_SBC | TO_CPSR,
                       TCG_REG_TMP, ah, bh, const_bh);
        return cond;

    case TCG_COND_LE:
    case TCG_COND_GT:
        /* Similar, but with swapped arguments, via reversed subtract.  */
        tcg_out_dat_rI(s, COND_AL, ARITH_RSB | TO_CPSR,
                       TCG_REG_TMP, al, bl, const_bl);
        tcg_out_dat_rI(s, COND_AL, ARITH_RSC | TO_CPSR,
                       TCG_REG_TMP, ah, bh, const_bh);
        return tcg_swap_cond(cond);

    default:
        g_assert_not_reached();
    }
}

#ifdef CONFIG_SOFTMMU
#include "../tcg-ldst.c.inc"

/* helper signature: helper_ret_ld_mmu(CPUState *env, target_ulong addr,
 *                                     int mmu_idx, uintptr_t ra)
 */
static void * const qemu_ld_helpers[16] = {
    [MO_UB]   = helper_ret_ldub_mmu,
    [MO_SB]   = helper_ret_ldsb_mmu,

    [MO_LEUW] = helper_le_lduw_mmu,
    [MO_LEUL] = helper_le_ldul_mmu,
    [MO_LEQ]  = helper_le_ldq_mmu,
    [MO_LESW] = helper_le_ldsw_mmu,
    [MO_LESL] = helper_le_ldul_mmu,

    [MO_BEUW] = helper_be_lduw_mmu,
    [MO_BEUL] = helper_be_ldul_mmu,
    [MO_BEQ]  = helper_be_ldq_mmu,
    [MO_BESW] = helper_be_ldsw_mmu,
    [MO_BESL] = helper_be_ldul_mmu,
};

/* helper signature: helper_ret_st_mmu(CPUState *env, target_ulong addr,
 *                                     uintxx_t val, int mmu_idx, uintptr_t ra)
 */
static void * const qemu_st_helpers[16] = {
    [MO_UB]   = helper_ret_stb_mmu,
    [MO_LEUW] = helper_le_stw_mmu,
    [MO_LEUL] = helper_le_stl_mmu,
    [MO_LEQ]  = helper_le_stq_mmu,
    [MO_BEUW] = helper_be_stw_mmu,
    [MO_BEUL] = helper_be_stl_mmu,
    [MO_BEQ]  = helper_be_stq_mmu,
};

/* Helper routines for marshalling helper function arguments into
 * the correct registers and stack.
 * argreg is where we want to put this argument, arg is the argument itself.
 * Return value is the updated argreg ready for the next call.
 * Note that argreg 0..3 is real registers, 4+ on stack.
 *
 * We provide routines for arguments which are: immediate, 32 bit
 * value in register, 16 and 8 bit values in register (which must be zero
 * extended before use) and 64 bit value in a lo:hi register pair.
 */
#define DEFINE_TCG_OUT_ARG(NAME, ARGTYPE, MOV_ARG, EXT_ARG)                \
static TCGReg NAME(TCGContext *s, TCGReg argreg, ARGTYPE arg)              \
{                                                                          \
    if (argreg < 4) {                                                      \
        MOV_ARG(s, COND_AL, argreg, arg);                                  \
    } else {                                                               \
        int ofs = (argreg - 4) * 4;                                        \
        EXT_ARG;                                                           \
        tcg_debug_assert(ofs + 4 <= TCG_STATIC_CALL_ARGS_SIZE);            \
        tcg_out_st32_12(s, COND_AL, arg, TCG_REG_CALL_STACK, ofs);         \
    }                                                                      \
    return argreg + 1;                                                     \
}

DEFINE_TCG_OUT_ARG(tcg_out_arg_imm32, uint32_t, tcg_out_movi32,
    (tcg_out_movi32(s, COND_AL, TCG_REG_TMP, arg), arg = TCG_REG_TMP))
DEFINE_TCG_OUT_ARG(tcg_out_arg_reg8, TCGReg, tcg_out_ext8u,
    (tcg_out_ext8u(s, COND_AL, TCG_REG_TMP, arg), arg = TCG_REG_TMP))
DEFINE_TCG_OUT_ARG(tcg_out_arg_reg16, TCGReg, tcg_out_ext16u,
    (tcg_out_ext16u(s, COND_AL, TCG_REG_TMP, arg), arg = TCG_REG_TMP))
DEFINE_TCG_OUT_ARG(tcg_out_arg_reg32, TCGReg, tcg_out_mov_reg, )

static TCGReg tcg_out_arg_reg64(TCGContext *s, TCGReg argreg,
                                TCGReg arglo, TCGReg arghi)
{
    /* 64 bit arguments must go in even/odd register pairs
     * and in 8-aligned stack slots.
     */
    if (argreg & 1) {
        argreg++;
    }
    if (use_armv6_instructions && argreg >= 4
        && (arglo & 1) == 0 && arghi == arglo + 1) {
        tcg_out_strd_8(s, COND_AL, arglo,
                       TCG_REG_CALL_STACK, (argreg - 4) * 4);
        return argreg + 2;
    } else {
        argreg = tcg_out_arg_reg32(s, argreg, arglo);
        argreg = tcg_out_arg_reg32(s, argreg, arghi);
        return argreg;
    }
}

#define TLB_SHIFT	(CPU_TLB_ENTRY_BITS + CPU_TLB_BITS)

/* We expect to use an 9-bit sign-magnitude negative offset from ENV.  */
QEMU_BUILD_BUG_ON(TLB_MASK_TABLE_OFS(0) > 0);
QEMU_BUILD_BUG_ON(TLB_MASK_TABLE_OFS(0) < -256);

/* These offsets are built into the LDRD below.  */
QEMU_BUILD_BUG_ON(offsetof(CPUTLBDescFast, mask) != 0);
QEMU_BUILD_BUG_ON(offsetof(CPUTLBDescFast, table) != 4);

/* Load and compare a TLB entry, leaving the flags set.  Returns the register
   containing the addend of the tlb entry.  Clobbers R0, R1, R2, TMP.  */

static TCGReg tcg_out_tlb_read(TCGContext *s, TCGReg addrlo, TCGReg addrhi,
                               MemOp opc, int mem_index, bool is_load)
{
    int cmp_off = (is_load ? offsetof(CPUTLBEntry, addr_read)
                   : offsetof(CPUTLBEntry, addr_write));
    int fast_off = TLB_MASK_TABLE_OFS(mem_index);
    int mask_off = fast_off + offsetof(CPUTLBDescFast, mask);
    int table_off = fast_off + offsetof(CPUTLBDescFast, table);
    unsigned s_bits = opc & MO_SIZE;
    unsigned a_bits = get_alignment_bits(opc);

    /*
     * We don't support inline unaligned acceses, but we can easily
     * support overalignment checks.
     */
    if (a_bits < s_bits) {
        a_bits = s_bits;
    }

    /* Load env_tlb(env)->f[mmu_idx].{mask,table} into {r0,r1}.  */
    if (use_armv6_instructions) {
        tcg_out_ldrd_8(s, COND_AL, TCG_REG_R0, TCG_AREG0, fast_off);
    } else {
        tcg_out_ld(s, TCG_TYPE_I32, TCG_REG_R0, TCG_AREG0, mask_off);
        tcg_out_ld(s, TCG_TYPE_I32, TCG_REG_R1, TCG_AREG0, table_off);
    }

    /* Extract the tlb index from the address into R0.  */
    tcg_out_dat_reg(s, COND_AL, ARITH_AND, TCG_REG_R0, TCG_REG_R0, addrlo,
                    SHIFT_IMM_LSR(TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS));

    /*
     * Add the tlb_table pointer, creating the CPUTLBEntry address in R1.
     * Load the tlb comparator into R2/R3 and the fast path addend into R1.
     */
    if (cmp_off == 0) {
        if (use_armv6_instructions && TARGET_LONG_BITS == 64) {
            tcg_out_ldrd_rwb(s, COND_AL, TCG_REG_R2, TCG_REG_R1, TCG_REG_R0);
        } else {
            tcg_out_ld32_rwb(s, COND_AL, TCG_REG_R2, TCG_REG_R1, TCG_REG_R0);
        }
    } else {
        tcg_out_dat_reg(s, COND_AL, ARITH_ADD,
                        TCG_REG_R1, TCG_REG_R1, TCG_REG_R0, 0);
        if (use_armv6_instructions && TARGET_LONG_BITS == 64) {
            tcg_out_ldrd_8(s, COND_AL, TCG_REG_R2, TCG_REG_R1, cmp_off);
        } else {
            tcg_out_ld32_12(s, COND_AL, TCG_REG_R2, TCG_REG_R1, cmp_off);
        }
    }
    if (!use_armv6_instructions && TARGET_LONG_BITS == 64) {
        tcg_out_ld32_12(s, COND_AL, TCG_REG_R3, TCG_REG_R1, cmp_off + 4);
    }

    /* Load the tlb addend.  */
    tcg_out_ld32_12(s, COND_AL, TCG_REG_R1, TCG_REG_R1,
                    offsetof(CPUTLBEntry, addend));

    /*
     * Check alignment, check comparators.
     * Do this in no more than 3 insns.  Use MOVW for v7, if possible,
     * to reduce the number of sequential conditional instructions.
     * Almost all guests have at least 4k pages, which means that we need
     * to clear at least 9 bits even for an 8-byte memory, which means it
     * isn't worth checking for an immediate operand for BIC.
     */
    if (use_armv7_instructions && TARGET_PAGE_BITS <= 16) {
        tcg_target_ulong mask = ~(TARGET_PAGE_MASK | ((1 << a_bits) - 1));

        tcg_out_movi32(s, COND_AL, TCG_REG_TMP, mask);
        tcg_out_dat_reg(s, COND_AL, ARITH_BIC, TCG_REG_TMP,
                        addrlo, TCG_REG_TMP, 0);
        tcg_out_dat_reg(s, COND_AL, ARITH_CMP, 0, TCG_REG_R2, TCG_REG_TMP, 0);
    } else {
        if (a_bits) {
            tcg_out_dat_imm(s, COND_AL, ARITH_TST, 0, addrlo,
                            (1 << a_bits) - 1);
        }
        tcg_out_dat_reg(s, COND_AL, ARITH_MOV, TCG_REG_TMP, 0, addrlo,
                        SHIFT_IMM_LSR(TARGET_PAGE_BITS));
        tcg_out_dat_reg(s, (a_bits ? COND_EQ : COND_AL), ARITH_CMP,
                        0, TCG_REG_R2, TCG_REG_TMP,
                        SHIFT_IMM_LSL(TARGET_PAGE_BITS));
    }

    if (TARGET_LONG_BITS == 64) {
        tcg_out_dat_reg(s, COND_EQ, ARITH_CMP, 0, TCG_REG_R3, addrhi, 0);
    }

    return TCG_REG_R1;
}

/* Record the context of a call to the out of line helper code for the slow
   path for a load or store, so that we can later generate the correct
   helper code.  */
static void add_qemu_ldst_label(TCGContext *s, bool is_ld, TCGMemOpIdx oi,
                                TCGReg datalo, TCGReg datahi, TCGReg addrlo,
                                TCGReg addrhi, tcg_insn_unit *raddr,
                                tcg_insn_unit *label_ptr)
{
    TCGLabelQemuLdst *label = new_ldst_label(s);

    label->is_ld = is_ld;
    label->oi = oi;
    label->datalo_reg = datalo;
    label->datahi_reg = datahi;
    label->addrlo_reg = addrlo;
    label->addrhi_reg = addrhi;
    label->raddr = raddr;
    label->label_ptr[0] = label_ptr;
}

static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *lb)
{
    TCGReg argreg, datalo, datahi;
    TCGMemOpIdx oi = lb->oi;
    MemOp opc = get_memop(oi);
    void *func;

    if (!reloc_pc24(lb->label_ptr[0], s->code_ptr)) {
        return false;
    }

    argreg = tcg_out_arg_reg32(s, TCG_REG_R0, TCG_AREG0);
    if (TARGET_LONG_BITS == 64) {
        argreg = tcg_out_arg_reg64(s, argreg, lb->addrlo_reg, lb->addrhi_reg);
    } else {
        argreg = tcg_out_arg_reg32(s, argreg, lb->addrlo_reg);
    }
    argreg = tcg_out_arg_imm32(s, argreg, oi);
    argreg = tcg_out_arg_reg32(s, argreg, TCG_REG_R14);

    /* For armv6 we can use the canonical unsigned helpers and minimize
       icache usage.  For pre-armv6, use the signed helpers since we do
       not have a single insn sign-extend.  */
    if (use_armv6_instructions) {
        func = qemu_ld_helpers[opc & (MO_BSWAP | MO_SIZE)];
    } else {
        func = qemu_ld_helpers[opc & (MO_BSWAP | MO_SSIZE)];
        if (opc & MO_SIGN) {
            opc = MO_UL;
        }
    }
    tcg_out_call(s, func);

    datalo = lb->datalo_reg;
    datahi = lb->datahi_reg;
    switch (opc & MO_SSIZE) {
    case MO_SB:
        tcg_out_ext8s(s, COND_AL, datalo, TCG_REG_R0);
        break;
    case MO_SW:
        tcg_out_ext16s(s, COND_AL, datalo, TCG_REG_R0);
        break;
    default:
        tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0);
        break;
    case MO_Q:
        if (datalo != TCG_REG_R1) {
            tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0);
            tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1);
        } else if (datahi != TCG_REG_R0) {
            tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1);
            tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0);
        } else {
            tcg_out_mov_reg(s, COND_AL, TCG_REG_TMP, TCG_REG_R0);
            tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1);
            tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_TMP);
        }
        break;
    }

    tcg_out_goto(s, COND_AL, lb->raddr);
    return true;
}

static bool tcg_out_qemu_st_slow_path(TCGContext *s, TCGLabelQemuLdst *lb)
{
    TCGReg argreg, datalo, datahi;
    TCGMemOpIdx oi = lb->oi;
    MemOp opc = get_memop(oi);

    if (!reloc_pc24(lb->label_ptr[0], s->code_ptr)) {
        return false;
    }

    argreg = TCG_REG_R0;
    argreg = tcg_out_arg_reg32(s, argreg, TCG_AREG0);
    if (TARGET_LONG_BITS == 64) {
        argreg = tcg_out_arg_reg64(s, argreg, lb->addrlo_reg, lb->addrhi_reg);
    } else {
        argreg = tcg_out_arg_reg32(s, argreg, lb->addrlo_reg);
    }

    datalo = lb->datalo_reg;
    datahi = lb->datahi_reg;
    switch (opc & MO_SIZE) {
    case MO_8:
        argreg = tcg_out_arg_reg8(s, argreg, datalo);
        break;
    case MO_16:
        argreg = tcg_out_arg_reg16(s, argreg, datalo);
        break;
    case MO_32:
    default:
        argreg = tcg_out_arg_reg32(s, argreg, datalo);
        break;
    case MO_64:
        argreg = tcg_out_arg_reg64(s, argreg, datalo, datahi);
        break;
    }

    argreg = tcg_out_arg_imm32(s, argreg, oi);
    argreg = tcg_out_arg_reg32(s, argreg, TCG_REG_R14);

    /* Tail-call to the helper, which will return to the fast path.  */
    tcg_out_goto(s, COND_AL, qemu_st_helpers[opc & (MO_BSWAP | MO_SIZE)]);
    return true;
}
#endif /* SOFTMMU */

static inline void tcg_out_qemu_ld_index(TCGContext *s, MemOp opc,
                                         TCGReg datalo, TCGReg datahi,
                                         TCGReg addrlo, TCGReg addend)
{
    MemOp bswap = opc & MO_BSWAP;

    switch (opc & MO_SSIZE) {
    case MO_UB:
        tcg_out_ld8_r(s, COND_AL, datalo, addrlo, addend);
        break;
    case MO_SB:
        tcg_out_ld8s_r(s, COND_AL, datalo, addrlo, addend);
        break;
    case MO_UW:
        tcg_out_ld16u_r(s, COND_AL, datalo, addrlo, addend);
        if (bswap) {
            tcg_out_bswap16(s, COND_AL, datalo, datalo);
        }
        break;
    case MO_SW:
        if (bswap) {
            tcg_out_ld16u_r(s, COND_AL, datalo, addrlo, addend);
            tcg_out_bswap16s(s, COND_AL, datalo, datalo);
        } else {
            tcg_out_ld16s_r(s, COND_AL, datalo, addrlo, addend);
        }
        break;
    case MO_UL:
    default:
        tcg_out_ld32_r(s, COND_AL, datalo, addrlo, addend);
        if (bswap) {
            tcg_out_bswap32(s, COND_AL, datalo, datalo);
        }
        break;
    case MO_Q:
        {
            TCGReg dl = (bswap ? datahi : datalo);
            TCGReg dh = (bswap ? datalo : datahi);

            /* Avoid ldrd for user-only emulation, to handle unaligned.  */
            if (USING_SOFTMMU && use_armv6_instructions
                && (dl & 1) == 0 && dh == dl + 1) {
                tcg_out_ldrd_r(s, COND_AL, dl, addrlo, addend);
            } else if (dl != addend) {
                tcg_out_ld32_rwb(s, COND_AL, dl, addend, addrlo);
                tcg_out_ld32_12(s, COND_AL, dh, addend, 4);
            } else {
                tcg_out_dat_reg(s, COND_AL, ARITH_ADD, TCG_REG_TMP,
                                addend, addrlo, SHIFT_IMM_LSL(0));
                tcg_out_ld32_12(s, COND_AL, dl, TCG_REG_TMP, 0);
                tcg_out_ld32_12(s, COND_AL, dh, TCG_REG_TMP, 4);
            }
            if (bswap) {
                tcg_out_bswap32(s, COND_AL, dl, dl);
                tcg_out_bswap32(s, COND_AL, dh, dh);
            }
        }
        break;
    }
}

static inline void tcg_out_qemu_ld_direct(TCGContext *s, MemOp opc,
                                          TCGReg datalo, TCGReg datahi,
                                          TCGReg addrlo)
{
    MemOp bswap = opc & MO_BSWAP;

    switch (opc & MO_SSIZE) {
    case MO_UB:
        tcg_out_ld8_12(s, COND_AL, datalo, addrlo, 0);
        break;
    case MO_SB:
        tcg_out_ld8s_8(s, COND_AL, datalo, addrlo, 0);
        break;
    case MO_UW:
        tcg_out_ld16u_8(s, COND_AL, datalo, addrlo, 0);
        if (bswap) {
            tcg_out_bswap16(s, COND_AL, datalo, datalo);
        }
        break;
    case MO_SW:
        if (bswap) {
            tcg_out_ld16u_8(s, COND_AL, datalo, addrlo, 0);
            tcg_out_bswap16s(s, COND_AL, datalo, datalo);
        } else {
            tcg_out_ld16s_8(s, COND_AL, datalo, addrlo, 0);
        }
        break;
    case MO_UL:
    default:
        tcg_out_ld32_12(s, COND_AL, datalo, addrlo, 0);
        if (bswap) {
            tcg_out_bswap32(s, COND_AL, datalo, datalo);
        }
        break;
    case MO_Q:
        {
            TCGReg dl = (bswap ? datahi : datalo);
            TCGReg dh = (bswap ? datalo : datahi);

            /* Avoid ldrd for user-only emulation, to handle unaligned.  */
            if (USING_SOFTMMU && use_armv6_instructions
                && (dl & 1) == 0 && dh == dl + 1) {
                tcg_out_ldrd_8(s, COND_AL, dl, addrlo, 0);
            } else if (dl == addrlo) {
                tcg_out_ld32_12(s, COND_AL, dh, addrlo, bswap ? 0 : 4);
                tcg_out_ld32_12(s, COND_AL, dl, addrlo, bswap ? 4 : 0);
            } else {
                tcg_out_ld32_12(s, COND_AL, dl, addrlo, bswap ? 4 : 0);
                tcg_out_ld32_12(s, COND_AL, dh, addrlo, bswap ? 0 : 4);
            }
            if (bswap) {
                tcg_out_bswap32(s, COND_AL, dl, dl);
                tcg_out_bswap32(s, COND_AL, dh, dh);
            }
        }
        break;
    }
}

static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, bool is64)
{
    TCGReg addrlo, datalo, datahi, addrhi __attribute__((unused));
    TCGMemOpIdx oi;
    MemOp opc;
#ifdef CONFIG_SOFTMMU
    int mem_index;
    TCGReg addend;
    tcg_insn_unit *label_ptr;
#endif

    datalo = *args++;
    datahi = (is64 ? *args++ : 0);
    addrlo = *args++;
    addrhi = (TARGET_LONG_BITS == 64 ? *args++ : 0);
    oi = *args++;
    opc = get_memop(oi);

#ifdef CONFIG_SOFTMMU
    mem_index = get_mmuidx(oi);
    addend = tcg_out_tlb_read(s, addrlo, addrhi, opc, mem_index, 1);

    /* This a conditional BL only to load a pointer within this opcode into LR
       for the slow path.  We will not be using the value for a tail call.  */
    label_ptr = s->code_ptr;
    tcg_out_bl(s, COND_NE, 0);

    tcg_out_qemu_ld_index(s, opc, datalo, datahi, addrlo, addend);

    add_qemu_ldst_label(s, true, oi, datalo, datahi, addrlo, addrhi,
                        s->code_ptr, label_ptr);
#else /* !CONFIG_SOFTMMU */
    if (guest_base) {
        tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_TMP, guest_base);
        tcg_out_qemu_ld_index(s, opc, datalo, datahi, addrlo, TCG_REG_TMP);
    } else {
        tcg_out_qemu_ld_direct(s, opc, datalo, datahi, addrlo);
    }
#endif
}

static inline void tcg_out_qemu_st_index(TCGContext *s, int cond, MemOp opc,
                                         TCGReg datalo, TCGReg datahi,
                                         TCGReg addrlo, TCGReg addend)
{
    MemOp bswap = opc & MO_BSWAP;

    switch (opc & MO_SIZE) {
    case MO_8:
        tcg_out_st8_r(s, cond, datalo, addrlo, addend);
        break;
    case MO_16:
        if (bswap) {
            tcg_out_bswap16st(s, cond, TCG_REG_R0, datalo);
            tcg_out_st16_r(s, cond, TCG_REG_R0, addrlo, addend);
        } else {
            tcg_out_st16_r(s, cond, datalo, addrlo, addend);
        }
        break;
    case MO_32:
    default:
        if (bswap) {
            tcg_out_bswap32(s, cond, TCG_REG_R0, datalo);
            tcg_out_st32_r(s, cond, TCG_REG_R0, addrlo, addend);
        } else {
            tcg_out_st32_r(s, cond, datalo, addrlo, addend);
        }
        break;
    case MO_64:
        /* Avoid strd for user-only emulation, to handle unaligned.  */
        if (bswap) {
            tcg_out_bswap32(s, cond, TCG_REG_R0, datahi);
            tcg_out_st32_rwb(s, cond, TCG_REG_R0, addend, addrlo);
            tcg_out_bswap32(s, cond, TCG_REG_R0, datalo);
            tcg_out_st32_12(s, cond, TCG_REG_R0, addend, 4);
        } else if (USING_SOFTMMU && use_armv6_instructions
                   && (datalo & 1) == 0 && datahi == datalo + 1) {
            tcg_out_strd_r(s, cond, datalo, addrlo, addend);
        } else {
            tcg_out_st32_rwb(s, cond, datalo, addend, addrlo);
            tcg_out_st32_12(s, cond, datahi, addend, 4);
        }
        break;
    }
}

static inline void tcg_out_qemu_st_direct(TCGContext *s, MemOp opc,
                                          TCGReg datalo, TCGReg datahi,
                                          TCGReg addrlo)
{
    MemOp bswap = opc & MO_BSWAP;

    switch (opc & MO_SIZE) {
    case MO_8:
        tcg_out_st8_12(s, COND_AL, datalo, addrlo, 0);
        break;
    case MO_16:
        if (bswap) {
            tcg_out_bswap16st(s, COND_AL, TCG_REG_R0, datalo);
            tcg_out_st16_8(s, COND_AL, TCG_REG_R0, addrlo, 0);
        } else {
            tcg_out_st16_8(s, COND_AL, datalo, addrlo, 0);
        }
        break;
    case MO_32:
    default:
        if (bswap) {
            tcg_out_bswap32(s, COND_AL, TCG_REG_R0, datalo);
            tcg_out_st32_12(s, COND_AL, TCG_REG_R0, addrlo, 0);
        } else {
            tcg_out_st32_12(s, COND_AL, datalo, addrlo, 0);
        }
        break;
    case MO_64:
        /* Avoid strd for user-only emulation, to handle unaligned.  */
        if (bswap) {
            tcg_out_bswap32(s, COND_AL, TCG_REG_R0, datahi);
            tcg_out_st32_12(s, COND_AL, TCG_REG_R0, addrlo, 0);
            tcg_out_bswap32(s, COND_AL, TCG_REG_R0, datalo);
            tcg_out_st32_12(s, COND_AL, TCG_REG_R0, addrlo, 4);
        } else if (USING_SOFTMMU && use_armv6_instructions
                   && (datalo & 1) == 0 && datahi == datalo + 1) {
            tcg_out_strd_8(s, COND_AL, datalo, addrlo, 0);
        } else {
            tcg_out_st32_12(s, COND_AL, datalo, addrlo, 0);
            tcg_out_st32_12(s, COND_AL, datahi, addrlo, 4);
        }
        break;
    }
}

static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, bool is64)
{
    TCGReg addrlo, datalo, datahi, addrhi __attribute__((unused));
    TCGMemOpIdx oi;
    MemOp opc;
#ifdef CONFIG_SOFTMMU
    int mem_index;
    TCGReg addend;
    tcg_insn_unit *label_ptr;
#endif

    datalo = *args++;
    datahi = (is64 ? *args++ : 0);
    addrlo = *args++;
    addrhi = (TARGET_LONG_BITS == 64 ? *args++ : 0);
    oi = *args++;
    opc = get_memop(oi);

#ifdef CONFIG_SOFTMMU
    mem_index = get_mmuidx(oi);
    addend = tcg_out_tlb_read(s, addrlo, addrhi, opc, mem_index, 0);

    tcg_out_qemu_st_index(s, COND_EQ, opc, datalo, datahi, addrlo, addend);

    /* The conditional call must come last, as we're going to return here.  */
    label_ptr = s->code_ptr;
    tcg_out_bl(s, COND_NE, 0);

    add_qemu_ldst_label(s, false, oi, datalo, datahi, addrlo, addrhi,
                        s->code_ptr, label_ptr);
#else /* !CONFIG_SOFTMMU */
    if (guest_base) {
        tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_TMP, guest_base);
        tcg_out_qemu_st_index(s, COND_AL, opc, datalo,
                              datahi, addrlo, TCG_REG_TMP);
    } else {
        tcg_out_qemu_st_direct(s, opc, datalo, datahi, addrlo);
    }
#endif
}

static void tcg_out_epilogue(TCGContext *s);

static inline void tcg_out_op(TCGContext *s, TCGOpcode opc,
                const TCGArg *args, const int *const_args)
{
    TCGArg a0, a1, a2, a3, a4, a5;
    int c;

    switch (opc) {
    case INDEX_op_exit_tb:
        tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_R0, args[0]);
        tcg_out_epilogue(s);
        break;
    case INDEX_op_goto_tb:
        {
            /* Indirect jump method */
            intptr_t ptr, dif, dil;
            TCGReg base = TCG_REG_PC;

            tcg_debug_assert(s->tb_jmp_insn_offset == 0);
            ptr = (intptr_t)(s->tb_jmp_target_addr + args[0]);
            dif = ptr - ((intptr_t)s->code_ptr + 8);
            dil = sextract32(dif, 0, 12);
            if (dif != dil) {
                /* The TB is close, but outside the 12 bits addressable by
                   the load.  We can extend this to 20 bits with a sub of a
                   shifted immediate from pc.  In the vastly unlikely event
                   the code requires more than 1MB, we'll use 2 insns and
                   be no worse off.  */
                base = TCG_REG_R0;
                tcg_out_movi32(s, COND_AL, base, ptr - dil);
            }
            tcg_out_ld32_12(s, COND_AL, TCG_REG_PC, base, dil);
            set_jmp_reset_offset(s, args[0]);
        }
        break;
    case INDEX_op_goto_ptr:
        tcg_out_bx(s, COND_AL, args[0]);
        break;
    case INDEX_op_br:
        tcg_out_goto_label(s, COND_AL, arg_label(args[0]));
        break;

    case INDEX_op_ld8u_i32:
        tcg_out_ld8u(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_ld8s_i32:
        tcg_out_ld8s(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_ld16u_i32:
        tcg_out_ld16u(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_ld16s_i32:
        tcg_out_ld16s(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_ld_i32:
        tcg_out_ld32u(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_st8_i32:
        tcg_out_st8(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_st16_i32:
        tcg_out_st16(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_st_i32:
        tcg_out_st32(s, COND_AL, args[0], args[1], args[2]);
        break;

    case INDEX_op_movcond_i32:
        /* Constraints mean that v2 is always in the same register as dest,
         * so we only need to do "if condition passed, move v1 to dest".
         */
        tcg_out_dat_rIN(s, COND_AL, ARITH_CMP, ARITH_CMN, 0,
                        args[1], args[2], const_args[2]);
        tcg_out_dat_rIK(s, tcg_cond_to_arm_cond[args[5]], ARITH_MOV,
                        ARITH_MVN, args[0], 0, args[3], const_args[3]);
        break;
    case INDEX_op_add_i32:
        tcg_out_dat_rIN(s, COND_AL, ARITH_ADD, ARITH_SUB,
                        args[0], args[1], args[2], const_args[2]);
        break;
    case INDEX_op_sub_i32:
        if (const_args[1]) {
            if (const_args[2]) {
                tcg_out_movi32(s, COND_AL, args[0], args[1] - args[2]);
            } else {
                tcg_out_dat_rI(s, COND_AL, ARITH_RSB,
                               args[0], args[2], args[1], 1);
            }
        } else {
            tcg_out_dat_rIN(s, COND_AL, ARITH_SUB, ARITH_ADD,
                            args[0], args[1], args[2], const_args[2]);
        }
        break;
    case INDEX_op_and_i32:
        tcg_out_dat_rIK(s, COND_AL, ARITH_AND, ARITH_BIC,
                        args[0], args[1], args[2], const_args[2]);
        break;
    case INDEX_op_andc_i32:
        tcg_out_dat_rIK(s, COND_AL, ARITH_BIC, ARITH_AND,
                        args[0], args[1], args[2], const_args[2]);
        break;
    case INDEX_op_or_i32:
        c = ARITH_ORR;
        goto gen_arith;
    case INDEX_op_xor_i32:
        c = ARITH_EOR;
        /* Fall through.  */
    gen_arith:
        tcg_out_dat_rI(s, COND_AL, c, args[0], args[1], args[2], const_args[2]);
        break;
    case INDEX_op_add2_i32:
        a0 = args[0], a1 = args[1], a2 = args[2];
        a3 = args[3], a4 = args[4], a5 = args[5];
        if (a0 == a3 || (a0 == a5 && !const_args[5])) {
            a0 = TCG_REG_TMP;
        }
        tcg_out_dat_rIN(s, COND_AL, ARITH_ADD | TO_CPSR, ARITH_SUB | TO_CPSR,
                        a0, a2, a4, const_args[4]);
        tcg_out_dat_rIK(s, COND_AL, ARITH_ADC, ARITH_SBC,
                        a1, a3, a5, const_args[5]);
        tcg_out_mov_reg(s, COND_AL, args[0], a0);
        break;
    case INDEX_op_sub2_i32:
        a0 = args[0], a1 = args[1], a2 = args[2];
        a3 = args[3], a4 = args[4], a5 = args[5];
        if ((a0 == a3 && !const_args[3]) || (a0 == a5 && !const_args[5])) {
            a0 = TCG_REG_TMP;
        }
        if (const_args[2]) {
            if (const_args[4]) {
                tcg_out_movi32(s, COND_AL, a0, a4);
                a4 = a0;
            }
            tcg_out_dat_rI(s, COND_AL, ARITH_RSB | TO_CPSR, a0, a4, a2, 1);
        } else {
            tcg_out_dat_rIN(s, COND_AL, ARITH_SUB | TO_CPSR,
                            ARITH_ADD | TO_CPSR, a0, a2, a4, const_args[4]);
        }
        if (const_args[3]) {
            if (const_args[5]) {
                tcg_out_movi32(s, COND_AL, a1, a5);
                a5 = a1;
            }
            tcg_out_dat_rI(s, COND_AL, ARITH_RSC, a1, a5, a3, 1);
        } else {
            tcg_out_dat_rIK(s, COND_AL, ARITH_SBC, ARITH_ADC,
                            a1, a3, a5, const_args[5]);
        }
        tcg_out_mov_reg(s, COND_AL, args[0], a0);
        break;
    case INDEX_op_neg_i32:
        tcg_out_dat_imm(s, COND_AL, ARITH_RSB, args[0], args[1], 0);
        break;
    case INDEX_op_not_i32:
        tcg_out_dat_reg(s, COND_AL,
                        ARITH_MVN, args[0], 0, args[1], SHIFT_IMM_LSL(0));
        break;
    case INDEX_op_mul_i32:
        tcg_out_mul32(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_mulu2_i32:
        tcg_out_umull32(s, COND_AL, args[0], args[1], args[2], args[3]);
        break;
    case INDEX_op_muls2_i32:
        tcg_out_smull32(s, COND_AL, args[0], args[1], args[2], args[3]);
        break;
    /* XXX: Perhaps args[2] & 0x1f is wrong */
    case INDEX_op_shl_i32:
        c = const_args[2] ?
                SHIFT_IMM_LSL(args[2] & 0x1f) : SHIFT_REG_LSL(args[2]);
        goto gen_shift32;
    case INDEX_op_shr_i32:
        c = const_args[2] ? (args[2] & 0x1f) ? SHIFT_IMM_LSR(args[2] & 0x1f) :
                SHIFT_IMM_LSL(0) : SHIFT_REG_LSR(args[2]);
        goto gen_shift32;
    case INDEX_op_sar_i32:
        c = const_args[2] ? (args[2] & 0x1f) ? SHIFT_IMM_ASR(args[2] & 0x1f) :
                SHIFT_IMM_LSL(0) : SHIFT_REG_ASR(args[2]);
        goto gen_shift32;
    case INDEX_op_rotr_i32:
        c = const_args[2] ? (args[2] & 0x1f) ? SHIFT_IMM_ROR(args[2] & 0x1f) :
                SHIFT_IMM_LSL(0) : SHIFT_REG_ROR(args[2]);
        /* Fall through.  */
    gen_shift32:
        tcg_out_dat_reg(s, COND_AL, ARITH_MOV, args[0], 0, args[1], c);
        break;

    case INDEX_op_rotl_i32:
        if (const_args[2]) {
            tcg_out_dat_reg(s, COND_AL, ARITH_MOV, args[0], 0, args[1],
                            ((0x20 - args[2]) & 0x1f) ?
                            SHIFT_IMM_ROR((0x20 - args[2]) & 0x1f) :
                            SHIFT_IMM_LSL(0));
        } else {
            tcg_out_dat_imm(s, COND_AL, ARITH_RSB, TCG_REG_TMP, args[2], 0x20);
            tcg_out_dat_reg(s, COND_AL, ARITH_MOV, args[0], 0, args[1],
                            SHIFT_REG_ROR(TCG_REG_TMP));
        }
        break;

    case INDEX_op_ctz_i32:
        tcg_out_dat_reg(s, COND_AL, INSN_RBIT, TCG_REG_TMP, 0, args[1], 0);
        a1 = TCG_REG_TMP;
        goto do_clz;

    case INDEX_op_clz_i32:
        a1 = args[1];
    do_clz:
        a0 = args[0];
        a2 = args[2];
        c = const_args[2];
        if (c && a2 == 32) {
            tcg_out_dat_reg(s, COND_AL, INSN_CLZ, a0, 0, a1, 0);
            break;
        }
        tcg_out_dat_imm(s, COND_AL, ARITH_CMP, 0, a1, 0);
        tcg_out_dat_reg(s, COND_NE, INSN_CLZ, a0, 0, a1, 0);
        if (c || a0 != a2) {
            tcg_out_dat_rIK(s, COND_EQ, ARITH_MOV, ARITH_MVN, a0, 0, a2, c);
        }
        break;

    case INDEX_op_brcond_i32:
        tcg_out_dat_rIN(s, COND_AL, ARITH_CMP, ARITH_CMN, 0,
                       args[0], args[1], const_args[1]);
        tcg_out_goto_label(s, tcg_cond_to_arm_cond[args[2]],
                           arg_label(args[3]));
        break;
    case INDEX_op_setcond_i32:
        tcg_out_dat_rIN(s, COND_AL, ARITH_CMP, ARITH_CMN, 0,
                        args[1], args[2], const_args[2]);
        tcg_out_dat_imm(s, tcg_cond_to_arm_cond[args[3]],
                        ARITH_MOV, args[0], 0, 1);
        tcg_out_dat_imm(s, tcg_cond_to_arm_cond[tcg_invert_cond(args[3])],
                        ARITH_MOV, args[0], 0, 0);
        break;

    case INDEX_op_brcond2_i32:
        c = tcg_out_cmp2(s, args, const_args);
        tcg_out_goto_label(s, tcg_cond_to_arm_cond[c], arg_label(args[5]));
        break;
    case INDEX_op_setcond2_i32:
        c = tcg_out_cmp2(s, args + 1, const_args + 1);
        tcg_out_dat_imm(s, tcg_cond_to_arm_cond[c], ARITH_MOV, args[0], 0, 1);
        tcg_out_dat_imm(s, tcg_cond_to_arm_cond[tcg_invert_cond(c)],
                        ARITH_MOV, args[0], 0, 0);
        break;

    case INDEX_op_qemu_ld_i32:
        tcg_out_qemu_ld(s, args, 0);
        break;
    case INDEX_op_qemu_ld_i64:
        tcg_out_qemu_ld(s, args, 1);
        break;
    case INDEX_op_qemu_st_i32:
        tcg_out_qemu_st(s, args, 0);
        break;
    case INDEX_op_qemu_st_i64:
        tcg_out_qemu_st(s, args, 1);
        break;

    case INDEX_op_bswap16_i32:
        tcg_out_bswap16(s, COND_AL, args[0], args[1]);
        break;
    case INDEX_op_bswap32_i32:
        tcg_out_bswap32(s, COND_AL, args[0], args[1]);
        break;

    case INDEX_op_ext8s_i32:
        tcg_out_ext8s(s, COND_AL, args[0], args[1]);
        break;
    case INDEX_op_ext16s_i32:
        tcg_out_ext16s(s, COND_AL, args[0], args[1]);
        break;
    case INDEX_op_ext16u_i32:
        tcg_out_ext16u(s, COND_AL, args[0], args[1]);
        break;

    case INDEX_op_deposit_i32:
        tcg_out_deposit(s, COND_AL, args[0], args[2],
                        args[3], args[4], const_args[2]);
        break;
    case INDEX_op_extract_i32:
        tcg_out_extract(s, COND_AL, args[0], args[1], args[2], args[3]);
        break;
    case INDEX_op_sextract_i32:
        tcg_out_sextract(s, COND_AL, args[0], args[1], args[2], args[3]);
        break;
    case INDEX_op_extract2_i32:
        /* ??? These optimization vs zero should be generic.  */
        /* ??? But we can't substitute 2 for 1 in the opcode stream yet.  */
        if (const_args[1]) {
            if (const_args[2]) {
                tcg_out_movi(s, TCG_TYPE_REG, args[0], 0);
            } else {
                tcg_out_dat_reg(s, COND_AL, ARITH_MOV, args[0], 0,
                                args[2], SHIFT_IMM_LSL(32 - args[3]));
            }
        } else if (const_args[2]) {
            tcg_out_dat_reg(s, COND_AL, ARITH_MOV, args[0], 0,
                            args[1], SHIFT_IMM_LSR(args[3]));
        } else {
            /* We can do extract2 in 2 insns, vs the 3 required otherwise.  */
            tcg_out_dat_reg(s, COND_AL, ARITH_MOV, TCG_REG_TMP, 0,
                            args[2], SHIFT_IMM_LSL(32 - args[3]));
            tcg_out_dat_reg(s, COND_AL, ARITH_ORR, args[0], TCG_REG_TMP,
                            args[1], SHIFT_IMM_LSR(args[3]));
        }
        break;

    case INDEX_op_div_i32:
        tcg_out_sdiv(s, COND_AL, args[0], args[1], args[2]);
        break;
    case INDEX_op_divu_i32:
        tcg_out_udiv(s, COND_AL, args[0], args[1], args[2]);
        break;

    case INDEX_op_mb:
        tcg_out_mb(s, args[0]);
        break;

    case INDEX_op_mov_i32:  /* Always emitted via tcg_out_mov.  */
    case INDEX_op_movi_i32: /* Always emitted via tcg_out_movi.  */
    case INDEX_op_call:     /* Always emitted via tcg_out_call.  */
    default:
        tcg_abort();
    }
}

static const TCGTargetOpDef *tcg_target_op_def(TCGOpcode op)
{
    static const TCGTargetOpDef r = { .args_ct_str = { "r" } };
    static const TCGTargetOpDef r_r = { .args_ct_str = { "r", "r" } };
    static const TCGTargetOpDef s_s = { .args_ct_str = { "s", "s" } };
    static const TCGTargetOpDef r_l = { .args_ct_str = { "r", "l" } };
    static const TCGTargetOpDef r_r_r = { .args_ct_str = { "r", "r", "r" } };
    static const TCGTargetOpDef r_r_l = { .args_ct_str = { "r", "r", "l" } };
    static const TCGTargetOpDef r_l_l = { .args_ct_str = { "r", "l", "l" } };
    static const TCGTargetOpDef s_s_s = { .args_ct_str = { "s", "s", "s" } };
    static const TCGTargetOpDef r_r_ri = { .args_ct_str = { "r", "r", "ri" } };
    static const TCGTargetOpDef r_r_rI = { .args_ct_str = { "r", "r", "rI" } };
    static const TCGTargetOpDef r_r_rIN
        = { .args_ct_str = { "r", "r", "rIN" } };
    static const TCGTargetOpDef r_r_rIK
        = { .args_ct_str = { "r", "r", "rIK" } };
    static const TCGTargetOpDef r_r_r_r
        = { .args_ct_str = { "r", "r", "r", "r" } };
    static const TCGTargetOpDef r_r_l_l
        = { .args_ct_str = { "r", "r", "l", "l" } };
    static const TCGTargetOpDef s_s_s_s
        = { .args_ct_str = { "s", "s", "s", "s" } };
    static const TCGTargetOpDef br
        = { .args_ct_str = { "r", "rIN" } };
    static const TCGTargetOpDef ext2
        = { .args_ct_str = { "r", "rZ", "rZ" } };
    static const TCGTargetOpDef dep
        = { .args_ct_str = { "r", "0", "rZ" } };
    static const TCGTargetOpDef movc
        = { .args_ct_str = { "r", "r", "rIN", "rIK", "0" } };
    static const TCGTargetOpDef add2
        = { .args_ct_str = { "r", "r", "r", "r", "rIN", "rIK" } };
    static const TCGTargetOpDef sub2
        = { .args_ct_str = { "r", "r", "rI", "rI", "rIN", "rIK" } };
    static const TCGTargetOpDef br2
        = { .args_ct_str = { "r", "r", "rI", "rI" } };
    static const TCGTargetOpDef setc2
        = { .args_ct_str = { "r", "r", "r", "rI", "rI" } };

    switch (op) {
    case INDEX_op_goto_ptr:
        return &r;

    case INDEX_op_ld8u_i32:
    case INDEX_op_ld8s_i32:
    case INDEX_op_ld16u_i32:
    case INDEX_op_ld16s_i32:
    case INDEX_op_ld_i32:
    case INDEX_op_st8_i32:
    case INDEX_op_st16_i32:
    case INDEX_op_st_i32:
    case INDEX_op_neg_i32:
    case INDEX_op_not_i32:
    case INDEX_op_bswap16_i32:
    case INDEX_op_bswap32_i32:
    case INDEX_op_ext8s_i32:
    case INDEX_op_ext16s_i32:
    case INDEX_op_ext16u_i32:
    case INDEX_op_extract_i32:
    case INDEX_op_sextract_i32:
        return &r_r;

    case INDEX_op_add_i32:
    case INDEX_op_sub_i32:
    case INDEX_op_setcond_i32:
        return &r_r_rIN;
    case INDEX_op_and_i32:
    case INDEX_op_andc_i32:
    case INDEX_op_clz_i32:
    case INDEX_op_ctz_i32:
        return &r_r_rIK;
    case INDEX_op_mul_i32:
    case INDEX_op_div_i32:
    case INDEX_op_divu_i32:
        return &r_r_r;
    case INDEX_op_mulu2_i32:
    case INDEX_op_muls2_i32:
        return &r_r_r_r;
    case INDEX_op_or_i32:
    case INDEX_op_xor_i32:
        return &r_r_rI;
    case INDEX_op_shl_i32:
    case INDEX_op_shr_i32:
    case INDEX_op_sar_i32:
    case INDEX_op_rotl_i32:
    case INDEX_op_rotr_i32:
        return &r_r_ri;

    case INDEX_op_brcond_i32:
        return &br;
    case INDEX_op_deposit_i32:
        return &dep;
    case INDEX_op_extract2_i32:
        return &ext2;
    case INDEX_op_movcond_i32:
        return &movc;
    case INDEX_op_add2_i32:
        return &add2;
    case INDEX_op_sub2_i32:
        return &sub2;
    case INDEX_op_brcond2_i32:
        return &br2;
    case INDEX_op_setcond2_i32:
        return &setc2;

    case INDEX_op_qemu_ld_i32:
        return TARGET_LONG_BITS == 32 ? &r_l : &r_l_l;
    case INDEX_op_qemu_ld_i64:
        return TARGET_LONG_BITS == 32 ? &r_r_l : &r_r_l_l;
    case INDEX_op_qemu_st_i32:
        return TARGET_LONG_BITS == 32 ? &s_s : &s_s_s;
    case INDEX_op_qemu_st_i64:
        return TARGET_LONG_BITS == 32 ? &s_s_s : &s_s_s_s;

    default:
        return NULL;
    }
}

static void tcg_target_init(TCGContext *s)
{
    /* Only probe for the platform and capabilities if we havn't already
       determined maximum values at compile time.  */
#ifndef use_idiv_instructions
    {
        unsigned long hwcap = qemu_getauxval(AT_HWCAP);
        use_idiv_instructions = (hwcap & HWCAP_ARM_IDIVA) != 0;
    }
#endif
    if (__ARM_ARCH < 7) {
        const char *pl = (const char *)qemu_getauxval(AT_PLATFORM);
        if (pl != NULL && pl[0] == 'v' && pl[1] >= '4' && pl[1] <= '9') {
            arm_arch = pl[1] - '0';
        }
    }

    tcg_target_available_regs[TCG_TYPE_I32] = 0xffff;

    tcg_target_call_clobber_regs = 0;
    tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R0);
    tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R1);
    tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R2);
    tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R3);
    tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R12);
    tcg_regset_set_reg(tcg_target_call_clobber_regs, TCG_REG_R14);

    s->reserved_regs = 0;
    tcg_regset_set_reg(s->reserved_regs, TCG_REG_CALL_STACK);
    tcg_regset_set_reg(s->reserved_regs, TCG_REG_TMP);
    tcg_regset_set_reg(s->reserved_regs, TCG_REG_PC);
}

static inline void tcg_out_ld(TCGContext *s, TCGType type, TCGReg arg,
                              TCGReg arg1, intptr_t arg2)
{
    tcg_out_ld32u(s, COND_AL, arg, arg1, arg2);
}

static inline void tcg_out_st(TCGContext *s, TCGType