summary refs log tree commit diff stats
path: root/target
diff options
context:
space:
mode:
Diffstat (limited to 'target')
-rw-r--r--target/arm/Kconfig9
-rw-r--r--target/arm/cortex-regs.c11
-rw-r--r--target/arm/cpu.c9
-rw-r--r--target/arm/cpu.h4
-rw-r--r--target/arm/debug_helper.c11
-rw-r--r--target/arm/gdbstub64.c2
-rw-r--r--target/arm/helper.c15
-rw-r--r--target/arm/helper.h8
-rw-r--r--target/arm/internals.h12
-rw-r--r--target/arm/kvm.c35
-rw-r--r--target/arm/kvm64.c5
-rw-r--r--target/arm/kvm_arm.h19
-rw-r--r--target/arm/ptw.c95
-rw-r--r--target/arm/tcg/a64.decode152
-rw-r--r--target/arm/tcg/arm_ldst.h (renamed from target/arm/arm_ldst.h)0
-rw-r--r--target/arm/tcg/helper-a64.h (renamed from target/arm/helper-a64.h)0
-rw-r--r--target/arm/tcg/helper-mve.h (renamed from target/arm/helper-mve.h)0
-rw-r--r--target/arm/tcg/helper-sme.h (renamed from target/arm/helper-sme.h)0
-rw-r--r--target/arm/tcg/helper-sve.h (renamed from target/arm/helper-sve.h)0
-rw-r--r--target/arm/tcg/meson.build1
-rw-r--r--target/arm/tcg/pauth_helper.c6
-rw-r--r--target/arm/tcg/sve_helper.c6
-rw-r--r--target/arm/tcg/sve_ldst_internal.h (renamed from target/arm/sve_ldst_internal.h)0
-rw-r--r--target/arm/tcg/translate-a32.h (renamed from target/arm/translate-a32.h)0
-rw-r--r--target/arm/tcg/translate-a64.c1285
-rw-r--r--target/arm/tcg/translate.h5
-rw-r--r--target/i386/cpu.c4
-rw-r--r--target/i386/cpu.h3
-rw-r--r--target/i386/ops_sse.h8
-rw-r--r--target/i386/tcg/decode-new.c.inc90
-rw-r--r--target/i386/tcg/emit.c.inc2
-rw-r--r--target/openrisc/cpu.c4
-rw-r--r--target/openrisc/fpu_helper.c13
-rw-r--r--target/openrisc/sys_helper.c45
-rw-r--r--target/openrisc/translate.c72
-rw-r--r--target/s390x/cpu.c8
-rw-r--r--target/s390x/tcg/insn-data.h.inc2
-rw-r--r--target/s390x/tcg/translate.c81
38 files changed, 1105 insertions, 917 deletions
diff --git a/target/arm/Kconfig b/target/arm/Kconfig
index 39f05b6420..5947366f6e 100644
--- a/target/arm/Kconfig
+++ b/target/arm/Kconfig
@@ -1,13 +1,8 @@
 config ARM
     bool
+    select ARM_COMPATIBLE_SEMIHOSTING if TCG
+    select ARM_V7M if TCG
 
 config AARCH64
     bool
     select ARM
-
-# This config exists just so we can make SEMIHOSTING default when TCG
-# is selected without also changing it for other architectures.
-config ARM_SEMIHOSTING
-    bool
-    default y if TCG && ARM
-    select ARM_COMPATIBLE_SEMIHOSTING
diff --git a/target/arm/cortex-regs.c b/target/arm/cortex-regs.c
index 17708480e7..ae817b08dd 100644
--- a/target/arm/cortex-regs.c
+++ b/target/arm/cortex-regs.c
@@ -15,8 +15,15 @@ static uint64_t l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri)
 {
     ARMCPU *cpu = env_archcpu(env);
 
-    /* Number of cores is in [25:24]; otherwise we RAZ */
-    return (cpu->core_count - 1) << 24;
+    /*
+     * Number of cores is in [25:24]; otherwise we RAZ.
+     * If the board didn't configure the CPUs into clusters,
+     * we default to "all CPUs in one cluster", which might be
+     * more than the 4 that the hardware permits and which is
+     * all you can report in this two-bit field. Saturate to
+     * 0b11 (== 4 CPUs) rather than overflowing the field.
+     */
+    return MIN(cpu->core_count - 1, 3) << 24;
 }
 
 static const ARMCPRegInfo cortex_a72_a57_a53_cp_reginfo[] = {
diff --git a/target/arm/cpu.c b/target/arm/cpu.c
index 5182ed0c91..f6a88e52ac 100644
--- a/target/arm/cpu.c
+++ b/target/arm/cpu.c
@@ -1480,6 +1480,7 @@ void arm_cpu_post_init(Object *obj)
                                      qdev_prop_allow_set_link_before_realize,
                                      OBJ_PROP_LINK_STRONG);
         }
+        cpu->has_mte = true;
     }
 #endif
 }
@@ -1616,7 +1617,7 @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp)
         }
         if (cpu->tag_memory) {
             error_setg(errp,
-                       "Cannot enable %s when guest CPUs has MTE enabled",
+                       "Cannot enable %s when guest CPUs has tag memory enabled",
                        current_accel_name());
             return;
         }
@@ -1996,10 +1997,10 @@ static void arm_cpu_realizefn(DeviceState *dev, Error **errp)
     }
 
 #ifndef CONFIG_USER_ONLY
-    if (cpu->tag_memory == NULL && cpu_isar_feature(aa64_mte, cpu)) {
+    if (!cpu->has_mte && cpu_isar_feature(aa64_mte, cpu)) {
         /*
-         * Disable the MTE feature bits if we do not have tag-memory
-         * provided by the machine.
+         * Disable the MTE feature bits if we do not have the feature
+         * setup by the machine.
          */
         cpu->isar.id_aa64pfr1 =
             FIELD_DP64(cpu->isar.id_aa64pfr1, ID_AA64PFR1, MTE, 0);
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index d469a2637b..c3463e39bc 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -935,6 +935,9 @@ struct ArchCPU {
      */
     uint32_t psci_conduit;
 
+    /* CPU has Memory Tag Extension */
+    bool has_mte;
+
     /* For v8M, initial value of the Secure VTOR */
     uint32_t init_svtor;
     /* For v8M, initial value of the Non-secure VTOR */
@@ -1053,6 +1056,7 @@ struct ArchCPU {
     bool prop_pauth;
     bool prop_pauth_impdef;
     bool prop_lpa2;
+    OnOffAuto prop_mte;
 
     /* DCZ blocksize, in log_2(words), ie low 4 bits of DCZID_EL0 */
     uint32_t dcz_blocksize;
diff --git a/target/arm/debug_helper.c b/target/arm/debug_helper.c
index dfc8b2a1a5..d41cc643b1 100644
--- a/target/arm/debug_helper.c
+++ b/target/arm/debug_helper.c
@@ -949,8 +949,10 @@ static const ARMCPRegInfo debug_cp_reginfo[] = {
       .access = PL0_R, .accessfn = access_tdcc,
       .type = ARM_CP_CONST, .resetvalue = 0 },
     /*
-     * OSDTRRX_EL1/OSDTRTX_EL1 are used for save and restore of DBGDTRRX_EL0.
-     * It is a component of the Debug Communications Channel, which is not implemented.
+     * These registers belong to the Debug Communications Channel,
+     * which is not implemented. However we implement RAZ/WI behaviour
+     * with trapping to prevent spurious SIGILLs if the guest OS does
+     * access them as the support cannot be probed for.
      */
     { .name = "OSDTRRX_EL1", .state = ARM_CP_STATE_BOTH, .cp = 14,
       .opc0 = 2, .opc1 = 0, .crn = 0, .crm = 0, .opc2 = 2,
@@ -960,6 +962,11 @@ static const ARMCPRegInfo debug_cp_reginfo[] = {
       .opc0 = 2, .opc1 = 0, .crn = 0, .crm = 3, .opc2 = 2,
       .access = PL1_RW, .accessfn = access_tdcc,
       .type = ARM_CP_CONST, .resetvalue = 0 },
+    /* DBGDTRTX_EL0/DBGDTRRX_EL0 depend on direction */
+    { .name = "DBGDTR_EL0", .state = ARM_CP_STATE_BOTH, .cp = 14,
+      .opc0 = 2, .opc1 = 3, .crn = 0, .crm = 5, .opc2 = 0,
+      .access = PL0_RW, .accessfn = access_tdcc,
+      .type = ARM_CP_CONST, .resetvalue = 0 },
     /*
      * OSECCR_EL1 provides a mechanism for an operating system
      * to access the contents of EDECCR. EDECCR is not implemented though,
diff --git a/target/arm/gdbstub64.c b/target/arm/gdbstub64.c
index c1f7e8c934..d7b79a6589 100644
--- a/target/arm/gdbstub64.c
+++ b/target/arm/gdbstub64.c
@@ -233,7 +233,7 @@ int aarch64_gdb_get_pauth_reg(CPUARMState *env, GByteArray *buf, int reg)
             ARMMMUIdx mmu_idx = arm_stage1_mmu_idx(env);
             ARMVAParameters param;
 
-            param = aa64_va_parameters(env, -is_high, mmu_idx, is_data);
+            param = aa64_va_parameters(env, -is_high, mmu_idx, is_data, false);
             return gdb_get_reg64(buf, pauth_ptr_mask(param));
         }
     default:
diff --git a/target/arm/helper.c b/target/arm/helper.c
index 2297626bfb..0b7fd2e7e6 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -4904,7 +4904,7 @@ static TLBIRange tlbi_aa64_get_range(CPUARMState *env, ARMMMUIdx mmuidx,
     unsigned int page_size_granule, page_shift, num, scale, exponent;
     /* Extract one bit to represent the va selector in use. */
     uint64_t select = sextract64(value, 36, 1);
-    ARMVAParameters param = aa64_va_parameters(env, select, mmuidx, true);
+    ARMVAParameters param = aa64_va_parameters(env, select, mmuidx, true, false);
     TLBIRange ret = { };
     ARMGranuleSize gran;
 
@@ -11193,7 +11193,8 @@ static ARMGranuleSize sanitize_gran_size(ARMCPU *cpu, ARMGranuleSize gran,
 }
 
 ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
-                                   ARMMMUIdx mmu_idx, bool data)
+                                   ARMMMUIdx mmu_idx, bool data,
+                                   bool el1_is_aa32)
 {
     uint64_t tcr = regime_tcr(env, mmu_idx);
     bool epd, hpd, tsz_oob, ds, ha, hd;
@@ -11289,6 +11290,16 @@ ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
         }
     }
 
+    if (stage2 && el1_is_aa32) {
+        /*
+         * For AArch32 EL1 the min txsz (and thus max IPA size) requirements
+         * are loosened: a configured IPA of 40 bits is permitted even if
+         * the implemented PA is less than that (and so a 40 bit IPA would
+         * fault for an AArch64 EL1). See R_DTLMN.
+         */
+        min_tsz = MIN(min_tsz, 24);
+    }
+
     if (tsz > max_tsz) {
         tsz = max_tsz;
         tsz_oob = true;
diff --git a/target/arm/helper.h b/target/arm/helper.h
index 018b00ea75..3335c2b10b 100644
--- a/target/arm/helper.h
+++ b/target/arm/helper.h
@@ -1039,9 +1039,9 @@ DEF_HELPER_FLAGS_5(gvec_uclamp_d, TCG_CALL_NO_RWG,
                    void, ptr, ptr, ptr, ptr, i32)
 
 #ifdef TARGET_AARCH64
-#include "helper-a64.h"
-#include "helper-sve.h"
-#include "helper-sme.h"
+#include "tcg/helper-a64.h"
+#include "tcg/helper-sve.h"
+#include "tcg/helper-sme.h"
 #endif
 
-#include "helper-mve.h"
+#include "tcg/helper-mve.h"
diff --git a/target/arm/internals.h b/target/arm/internals.h
index 0df8f3b8bc..c869d18c38 100644
--- a/target/arm/internals.h
+++ b/target/arm/internals.h
@@ -1091,8 +1091,18 @@ typedef struct ARMVAParameters {
     ARMGranuleSize gran : 2;
 } ARMVAParameters;
 
+/**
+ * aa64_va_parameters: Return parameters for an AArch64 virtual address
+ * @env: CPU
+ * @va: virtual address to look up
+ * @mmu_idx: determines translation regime to use
+ * @data: true if this is a data access
+ * @el1_is_aa32: true if we are asking about stage 2 when EL1 is AArch32
+ *  (ignored if @mmu_idx is for a stage 1 regime; only affects tsz/tsz_oob)
+ */
 ARMVAParameters aa64_va_parameters(CPUARMState *env, uint64_t va,
-                                   ARMMMUIdx mmu_idx, bool data);
+                                   ARMMMUIdx mmu_idx, bool data,
+                                   bool el1_is_aa32);
 
 int aa64_va_parameter_tbi(uint64_t tcr, ARMMMUIdx mmu_idx);
 int aa64_va_parameter_tbid(uint64_t tcr, ARMMMUIdx mmu_idx);
diff --git a/target/arm/kvm.c b/target/arm/kvm.c
index 84da49332c..9553488ecd 100644
--- a/target/arm/kvm.c
+++ b/target/arm/kvm.c
@@ -31,6 +31,7 @@
 #include "hw/boards.h"
 #include "hw/irq.h"
 #include "qemu/log.h"
+#include "migration/blocker.h"
 
 const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
     KVM_CAP_LAST_INFO
@@ -1064,3 +1065,37 @@ bool kvm_arch_cpu_check_are_resettable(void)
 void kvm_arch_accel_class_init(ObjectClass *oc)
 {
 }
+
+void kvm_arm_enable_mte(Object *cpuobj, Error **errp)
+{
+    static bool tried_to_enable;
+    static bool succeeded_to_enable;
+    Error *mte_migration_blocker = NULL;
+    int ret;
+
+    if (!tried_to_enable) {
+        /*
+         * MTE on KVM is enabled on a per-VM basis (and retrying doesn't make
+         * sense), and we only want a single migration blocker as well.
+         */
+        tried_to_enable = true;
+
+        ret = kvm_vm_enable_cap(kvm_state, KVM_CAP_ARM_MTE, 0);
+        if (ret) {
+            error_setg_errno(errp, -ret, "Failed to enable KVM_CAP_ARM_MTE");
+            return;
+        }
+
+        /* TODO: add proper migration support with MTE enabled */
+        error_setg(&mte_migration_blocker,
+                   "Live migration disabled due to MTE enabled");
+        if (migrate_add_blocker(mte_migration_blocker, errp)) {
+            error_free(mte_migration_blocker);
+            return;
+        }
+        succeeded_to_enable = true;
+    }
+    if (succeeded_to_enable) {
+        object_property_set_bool(cpuobj, "has_mte", true, NULL);
+    }
+}
diff --git a/target/arm/kvm64.c b/target/arm/kvm64.c
index 810db33ccb..1893f38793 100644
--- a/target/arm/kvm64.c
+++ b/target/arm/kvm64.c
@@ -756,6 +756,11 @@ bool kvm_arm_steal_time_supported(void)
     return kvm_check_extension(kvm_state, KVM_CAP_STEAL_TIME);
 }
 
+bool kvm_arm_mte_supported(void)
+{
+    return kvm_check_extension(kvm_state, KVM_CAP_ARM_MTE);
+}
+
 QEMU_BUILD_BUG_ON(KVM_ARM64_SVE_VQ_MIN != 1);
 
 uint32_t kvm_arm_sve_get_vls(CPUState *cs)
diff --git a/target/arm/kvm_arm.h b/target/arm/kvm_arm.h
index 330fbe5c72..2083547bf6 100644
--- a/target/arm/kvm_arm.h
+++ b/target/arm/kvm_arm.h
@@ -314,6 +314,13 @@ bool kvm_arm_pmu_supported(void);
 bool kvm_arm_sve_supported(void);
 
 /**
+ * kvm_arm_mte_supported:
+ *
+ * Returns: true if KVM can enable MTE, and false otherwise.
+ */
+bool kvm_arm_mte_supported(void);
+
+/**
  * kvm_arm_get_max_vm_ipa_size:
  * @ms: Machine state handle
  * @fixed_ipa: True when the IPA limit is fixed at 40. This is the case
@@ -377,6 +384,8 @@ void kvm_arm_pvtime_init(CPUState *cs, uint64_t ipa);
 
 int kvm_arm_set_irq(int cpu, int irqtype, int irq, int level);
 
+void kvm_arm_enable_mte(Object *cpuobj, Error **errp);
+
 #else
 
 /*
@@ -403,6 +412,11 @@ static inline bool kvm_arm_steal_time_supported(void)
     return false;
 }
 
+static inline bool kvm_arm_mte_supported(void)
+{
+    return false;
+}
+
 /*
  * These functions should never actually be called without KVM support.
  */
@@ -451,6 +465,11 @@ static inline uint32_t kvm_arm_sve_get_vls(CPUState *cs)
     g_assert_not_reached();
 }
 
+static inline void kvm_arm_enable_mte(Object *cpuobj, Error **errp)
+{
+    g_assert_not_reached();
+}
+
 #endif
 
 static inline const char *gic_class_name(void)
diff --git a/target/arm/ptw.c b/target/arm/ptw.c
index bd75da8dbc..69c05cd9da 100644
--- a/target/arm/ptw.c
+++ b/target/arm/ptw.c
@@ -103,6 +103,37 @@ ARMMMUIdx arm_stage1_mmu_idx(CPUARMState *env)
     return stage_1_mmu_idx(arm_mmu_idx(env));
 }
 
+/*
+ * Return where we should do ptw loads from for a stage 2 walk.
+ * This depends on whether the address we are looking up is a
+ * Secure IPA or a NonSecure IPA, which we know from whether this is
+ * Stage2 or Stage2_S.
+ * If this is the Secure EL1&0 regime we need to check the NSW and SW bits.
+ */
+static ARMMMUIdx ptw_idx_for_stage_2(CPUARMState *env, ARMMMUIdx stage2idx)
+{
+    bool s2walk_secure;
+
+    /*
+     * We're OK to check the current state of the CPU here because
+     * (1) we always invalidate all TLBs when the SCR_EL3.NS bit changes
+     * (2) there's no way to do a lookup that cares about Stage 2 for a
+     * different security state to the current one for AArch64, and AArch32
+     * never has a secure EL2. (AArch32 ATS12NSO[UP][RW] allow EL3 to do
+     * an NS stage 1+2 lookup while the NS bit is 0.)
+     */
+    if (!arm_is_secure_below_el3(env) || !arm_el_is_aa64(env, 3)) {
+        return ARMMMUIdx_Phys_NS;
+    }
+    if (stage2idx == ARMMMUIdx_Stage2_S) {
+        s2walk_secure = !(env->cp15.vstcr_el2 & VSTCR_SW);
+    } else {
+        s2walk_secure = !(env->cp15.vtcr_el2 & VTCR_NSW);
+    }
+    return s2walk_secure ? ARMMMUIdx_Phys_S : ARMMMUIdx_Phys_NS;
+
+}
+
 static bool regime_translation_big_endian(CPUARMState *env, ARMMMUIdx mmu_idx)
 {
     return (regime_sctlr(env, mmu_idx) & SCTLR_EE) != 0;
@@ -220,7 +251,6 @@ static bool S1_ptw_translate(CPUARMState *env, S1Translate *ptw,
     ARMMMUIdx mmu_idx = ptw->in_mmu_idx;
     ARMMMUIdx s2_mmu_idx = ptw->in_ptw_idx;
     uint8_t pte_attrs;
-    bool pte_secure;
 
     ptw->out_virt = addr;
 
@@ -232,8 +262,8 @@ static bool S1_ptw_translate(CPUARMState *env, S1Translate *ptw,
         if (regime_is_stage2(s2_mmu_idx)) {
             S1Translate s2ptw = {
                 .in_mmu_idx = s2_mmu_idx,
-                .in_ptw_idx = is_secure ? ARMMMUIdx_Phys_S : ARMMMUIdx_Phys_NS,
-                .in_secure = is_secure,
+                .in_ptw_idx = ptw_idx_for_stage_2(env, s2_mmu_idx),
+                .in_secure = s2_mmu_idx == ARMMMUIdx_Stage2_S,
                 .in_debug = true,
             };
             GetPhysAddrResult s2 = { };
@@ -244,12 +274,12 @@ static bool S1_ptw_translate(CPUARMState *env, S1Translate *ptw,
             }
             ptw->out_phys = s2.f.phys_addr;
             pte_attrs = s2.cacheattrs.attrs;
-            pte_secure = s2.f.attrs.secure;
+            ptw->out_secure = s2.f.attrs.secure;
         } else {
             /* Regime is physical. */
             ptw->out_phys = addr;
             pte_attrs = 0;
-            pte_secure = is_secure;
+            ptw->out_secure = s2_mmu_idx == ARMMMUIdx_Phys_S;
         }
         ptw->out_host = NULL;
         ptw->out_rw = false;
@@ -270,7 +300,7 @@ static bool S1_ptw_translate(CPUARMState *env, S1Translate *ptw,
         ptw->out_phys = full->phys_addr | (addr & ~TARGET_PAGE_MASK);
         ptw->out_rw = full->prot & PAGE_WRITE;
         pte_attrs = full->pte_attrs;
-        pte_secure = full->attrs.secure;
+        ptw->out_secure = full->attrs.secure;
 #else
         g_assert_not_reached();
 #endif
@@ -293,11 +323,6 @@ static bool S1_ptw_translate(CPUARMState *env, S1Translate *ptw,
         }
     }
 
-    /* Check if page table walk is to secure or non-secure PA space. */
-    ptw->out_secure = (is_secure
-                       && !(pte_secure
-                            ? env->cp15.vstcr_el2 & VSTCR_SW
-                            : env->cp15.vtcr_el2 & VTCR_NSW));
     ptw->out_be = regime_translation_big_endian(env, mmu_idx);
     return true;
 
@@ -1110,17 +1135,6 @@ static int check_s2_mmu_setup(ARMCPU *cpu, bool is_aa64, uint64_t tcr,
     sl0 = extract32(tcr, 6, 2);
     if (is_aa64) {
         /*
-         * AArch64.S2InvalidTxSZ: While we checked tsz_oob near the top of
-         * get_phys_addr_lpae, that used aa64_va_parameters which apply
-         * to aarch64.  If Stage1 is aarch32, the min_txsz is larger.
-         * See AArch64.S2MinTxSZ, where min_tsz is 24, translated to
-         * inputsize is 64 - 24 = 40.
-         */
-        if (iasize < 40 && !arm_el_is_aa64(&cpu->env, 1)) {
-            goto fail;
-        }
-
-        /*
          * AArch64.S2InvalidSL: Interpretation of SL depends on the page size,
          * so interleave AArch64.S2StartLevel.
          */
@@ -1259,7 +1273,8 @@ static bool get_phys_addr_lpae(CPUARMState *env, S1Translate *ptw,
         int ps;
 
         param = aa64_va_parameters(env, address, mmu_idx,
-                                   access_type != MMU_INST_FETCH);
+                                   access_type != MMU_INST_FETCH,
+                                   !arm_el_is_aa64(env, 1));
         level = 0;
 
         /*
@@ -1415,17 +1430,18 @@ static bool get_phys_addr_lpae(CPUARMState *env, S1Translate *ptw,
     descaddrmask &= ~indexmask_grainsize;
 
     /*
-     * Secure accesses start with the page table in secure memory and
+     * Secure stage 1 accesses start with the page table in secure memory and
      * can be downgraded to non-secure at any step. Non-secure accesses
      * remain non-secure. We implement this by just ORing in the NSTable/NS
      * bits at each step.
+     * Stage 2 never gets this kind of downgrade.
      */
     tableattrs = is_secure ? 0 : (1 << 4);
 
  next_level:
     descaddr |= (address >> (stride * (4 - level))) & indexmask;
     descaddr &= ~7ULL;
-    nstable = extract32(tableattrs, 4, 1);
+    nstable = !regime_is_stage2(mmu_idx) && extract32(tableattrs, 4, 1);
     if (nstable) {
         /*
          * Stage2_S -> Stage2 or Phys_S -> Phys_NS
@@ -2725,7 +2741,7 @@ static bool get_phys_addr_twostage(CPUARMState *env, S1Translate *ptw,
     hwaddr ipa;
     int s1_prot, s1_lgpgsz;
     bool is_secure = ptw->in_secure;
-    bool ret, ipa_secure, s2walk_secure;
+    bool ret, ipa_secure;
     ARMCacheAttrs cacheattrs1;
     bool is_el0;
     uint64_t hcr;
@@ -2739,20 +2755,11 @@ static bool get_phys_addr_twostage(CPUARMState *env, S1Translate *ptw,
 
     ipa = result->f.phys_addr;
     ipa_secure = result->f.attrs.secure;
-    if (is_secure) {
-        /* Select TCR based on the NS bit from the S1 walk. */
-        s2walk_secure = !(ipa_secure
-                          ? env->cp15.vstcr_el2 & VSTCR_SW
-                          : env->cp15.vtcr_el2 & VTCR_NSW);
-    } else {
-        assert(!ipa_secure);
-        s2walk_secure = false;
-    }
 
     is_el0 = ptw->in_mmu_idx == ARMMMUIdx_Stage1_E0;
-    ptw->in_mmu_idx = s2walk_secure ? ARMMMUIdx_Stage2_S : ARMMMUIdx_Stage2;
-    ptw->in_ptw_idx = s2walk_secure ? ARMMMUIdx_Phys_S : ARMMMUIdx_Phys_NS;
-    ptw->in_secure = s2walk_secure;
+    ptw->in_mmu_idx = ipa_secure ? ARMMMUIdx_Stage2_S : ARMMMUIdx_Stage2;
+    ptw->in_secure = ipa_secure;
+    ptw->in_ptw_idx = ptw_idx_for_stage_2(env, ptw->in_mmu_idx);
 
     /*
      * S1 is done, now do S2 translation.
@@ -2860,6 +2867,16 @@ static bool get_phys_addr_with_struct(CPUARMState *env, S1Translate *ptw,
         ptw->in_ptw_idx = is_secure ? ARMMMUIdx_Stage2_S : ARMMMUIdx_Stage2;
         break;
 
+    case ARMMMUIdx_Stage2:
+    case ARMMMUIdx_Stage2_S:
+        /*
+         * Second stage lookup uses physical for ptw; whether this is S or
+         * NS may depend on the SW/NSW bits if this is a stage 2 lookup for
+         * the Secure EL2&0 regime.
+         */
+        ptw->in_ptw_idx = ptw_idx_for_stage_2(env, mmu_idx);
+        break;
+
     case ARMMMUIdx_E10_0:
         s1_mmu_idx = ARMMMUIdx_Stage1_E0;
         goto do_twostage;
@@ -2883,7 +2900,7 @@ static bool get_phys_addr_with_struct(CPUARMState *env, S1Translate *ptw,
         /* fall through */
 
     default:
-        /* Single stage and second stage uses physical for ptw. */
+        /* Single stage uses physical for ptw. */
         ptw->in_ptw_idx = is_secure ? ARMMMUIdx_Phys_S : ARMMMUIdx_Phys_NS;
         break;
     }
diff --git a/target/arm/tcg/a64.decode b/target/arm/tcg/a64.decode
new file mode 100644
index 0000000000..12a310d0a3
--- /dev/null
+++ b/target/arm/tcg/a64.decode
@@ -0,0 +1,152 @@
+# AArch64 A64 allowed instruction decoding
+#
+#  Copyright (c) 2023 Linaro, Ltd
+#
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 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/>.
+
+#
+# This file is processed by scripts/decodetree.py
+#
+
+&r               rn
+&ri              rd imm
+&rri_sf          rd rn imm sf
+&i               imm
+
+
+### Data Processing - Immediate
+
+# PC-rel addressing
+
+%imm_pcrel      5:s19 29:2
+@pcrel          . .. ..... ................... rd:5     &ri imm=%imm_pcrel
+
+ADR             0 .. 10000 ................... .....    @pcrel
+ADRP            1 .. 10000 ................... .....    @pcrel
+
+# Add/subtract (immediate)
+
+%imm12_sh12     10:12 !function=shl_12
+@addsub_imm     sf:1 .. ...... . imm:12 rn:5 rd:5
+@addsub_imm12   sf:1 .. ...... . ............ rn:5 rd:5 imm=%imm12_sh12
+
+ADD_i           . 00 100010 0 ............ ..... .....  @addsub_imm
+ADD_i           . 00 100010 1 ............ ..... .....  @addsub_imm12
+ADDS_i          . 01 100010 0 ............ ..... .....  @addsub_imm
+ADDS_i          . 01 100010 1 ............ ..... .....  @addsub_imm12
+
+SUB_i           . 10 100010 0 ............ ..... .....  @addsub_imm
+SUB_i           . 10 100010 1 ............ ..... .....  @addsub_imm12
+SUBS_i          . 11 100010 0 ............ ..... .....  @addsub_imm
+SUBS_i          . 11 100010 1 ............ ..... .....  @addsub_imm12
+
+# Add/subtract (immediate with tags)
+
+&rri_tag        rd rn uimm6 uimm4
+@addsub_imm_tag . .. ...... . uimm6:6 .. uimm4:4 rn:5 rd:5 &rri_tag
+
+ADDG_i          1 00 100011 0 ...... 00 .... ..... ..... @addsub_imm_tag
+SUBG_i          1 10 100011 0 ...... 00 .... ..... ..... @addsub_imm_tag
+
+# Logical (immediate)
+
+&rri_log        rd rn sf dbm
+@logic_imm_64   1 .. ...... dbm:13 rn:5 rd:5            &rri_log sf=1
+@logic_imm_32   0 .. ...... 0 dbm:12 rn:5 rd:5          &rri_log sf=0
+
+AND_i           . 00 100100 . ...... ...... ..... ..... @logic_imm_64
+AND_i           . 00 100100 . ...... ...... ..... ..... @logic_imm_32
+ORR_i           . 01 100100 . ...... ...... ..... ..... @logic_imm_64
+ORR_i           . 01 100100 . ...... ...... ..... ..... @logic_imm_32
+EOR_i           . 10 100100 . ...... ...... ..... ..... @logic_imm_64
+EOR_i           . 10 100100 . ...... ...... ..... ..... @logic_imm_32
+ANDS_i          . 11 100100 . ...... ...... ..... ..... @logic_imm_64
+ANDS_i          . 11 100100 . ...... ...... ..... ..... @logic_imm_32
+
+# Move wide (immediate)
+
+&movw           rd sf imm hw
+@movw_64        1 .. ...... hw:2   imm:16 rd:5          &movw sf=1
+@movw_32        0 .. ...... 0 hw:1 imm:16 rd:5          &movw sf=0
+
+MOVN            . 00 100101 .. ................ .....   @movw_64
+MOVN            . 00 100101 .. ................ .....   @movw_32
+MOVZ            . 10 100101 .. ................ .....   @movw_64
+MOVZ            . 10 100101 .. ................ .....   @movw_32
+MOVK            . 11 100101 .. ................ .....   @movw_64
+MOVK            . 11 100101 .. ................ .....   @movw_32
+
+# Bitfield
+
+&bitfield       rd rn sf immr imms
+@bitfield_64    1 .. ...... 1 immr:6 imms:6 rn:5 rd:5      &bitfield sf=1
+@bitfield_32    0 .. ...... 0 0 immr:5 0 imms:5 rn:5 rd:5  &bitfield sf=0
+
+SBFM            . 00 100110 . ...... ...... ..... ..... @bitfield_64
+SBFM            . 00 100110 . ...... ...... ..... ..... @bitfield_32
+BFM             . 01 100110 . ...... ...... ..... ..... @bitfield_64
+BFM             . 01 100110 . ...... ...... ..... ..... @bitfield_32
+UBFM            . 10 100110 . ...... ...... ..... ..... @bitfield_64
+UBFM            . 10 100110 . ...... ...... ..... ..... @bitfield_32
+
+# Extract
+
+&extract        rd rn rm imm sf
+
+EXTR            1 00 100111 1 0 rm:5 imm:6 rn:5 rd:5     &extract sf=1
+EXTR            0 00 100111 0 0 rm:5 0 imm:5 rn:5 rd:5   &extract sf=0
+
+# Branches
+
+%imm26   0:s26 !function=times_4
+@branch         . ..... .......................... &i imm=%imm26
+
+B               0 00101 .......................... @branch
+BL              1 00101 .......................... @branch
+
+%imm19   5:s19 !function=times_4
+&cbz     rt imm sf nz
+
+CBZ             sf:1 011010 nz:1 ................... rt:5 &cbz imm=%imm19
+
+%imm14     5:s14 !function=times_4
+%imm31_19  31:1 19:5
+&tbz       rt imm nz bitpos
+
+TBZ             . 011011 nz:1 ..... .............. rt:5 &tbz  imm=%imm14 bitpos=%imm31_19
+
+B_cond          0101010 0 ................... 0 cond:4 imm=%imm19
+
+BR              1101011 0000 11111 000000 rn:5 00000 &r
+BLR             1101011 0001 11111 000000 rn:5 00000 &r
+RET             1101011 0010 11111 000000 rn:5 00000 &r
+
+&braz       rn m
+BRAZ            1101011 0000 11111 00001 m:1 rn:5 11111 &braz   # BRAAZ, BRABZ
+BLRAZ           1101011 0001 11111 00001 m:1 rn:5 11111 &braz   # BLRAAZ, BLRABZ
+
+&reta       m
+RETA            1101011 0010 11111 00001 m:1 11111 11111 &reta  # RETAA, RETAB
+
+&bra        rn rm m
+BRA             1101011 1000 11111 00001 m:1 rn:5 rm:5 &bra # BRAA, BRAB
+BLRA            1101011 1001 11111 00001 m:1 rn:5 rm:5 &bra # BLRAA, BLRAB
+
+ERET            1101011 0100 11111 000000 11111 00000
+ERETA           1101011 0100 11111 00001 m:1 11111 11111 &reta  # ERETAA, ERETAB
+
+# We don't need to decode DRPS because it always UNDEFs except when
+# the processor is in halting debug state (which we don't implement).
+# The pattern is listed here as documentation.
+# DRPS            1101011 0101 11111 000000 11111 00000
diff --git a/target/arm/arm_ldst.h b/target/arm/tcg/arm_ldst.h
index cee0548a1c..cee0548a1c 100644
--- a/target/arm/arm_ldst.h
+++ b/target/arm/tcg/arm_ldst.h
diff --git a/target/arm/helper-a64.h b/target/arm/tcg/helper-a64.h
index ff56807247..ff56807247 100644
--- a/target/arm/helper-a64.h
+++ b/target/arm/tcg/helper-a64.h
diff --git a/target/arm/helper-mve.h b/target/arm/tcg/helper-mve.h
index 76bd25006d..76bd25006d 100644
--- a/target/arm/helper-mve.h
+++ b/target/arm/tcg/helper-mve.h
diff --git a/target/arm/helper-sme.h b/target/arm/tcg/helper-sme.h
index 27eef49a11..27eef49a11 100644
--- a/target/arm/helper-sme.h
+++ b/target/arm/tcg/helper-sme.h
diff --git a/target/arm/helper-sve.h b/target/arm/tcg/helper-sve.h
index cc4e1d8948..cc4e1d8948 100644
--- a/target/arm/helper-sve.h
+++ b/target/arm/tcg/helper-sve.h
diff --git a/target/arm/tcg/meson.build b/target/arm/tcg/meson.build
index 4d99f6dacb..130ed62fcd 100644
--- a/target/arm/tcg/meson.build
+++ b/target/arm/tcg/meson.build
@@ -13,6 +13,7 @@ gen = [
   decodetree.process('a32-uncond.decode', extra_args: '--static-decode=disas_a32_uncond'),
   decodetree.process('t32.decode', extra_args: '--static-decode=disas_t32'),
   decodetree.process('t16.decode', extra_args: ['-w', '16', '--static-decode=disas_t16']),
+  decodetree.process('a64.decode', extra_args: ['--static-decode=disas_a64']),
 ]
 
 arm_ss.add(gen)
diff --git a/target/arm/tcg/pauth_helper.c b/target/arm/tcg/pauth_helper.c
index de067fa716..62af569341 100644
--- a/target/arm/tcg/pauth_helper.c
+++ b/target/arm/tcg/pauth_helper.c
@@ -293,7 +293,7 @@ static uint64_t pauth_addpac(CPUARMState *env, uint64_t ptr, uint64_t modifier,
                              ARMPACKey *key, bool data)
 {
     ARMMMUIdx mmu_idx = arm_stage1_mmu_idx(env);
-    ARMVAParameters param = aa64_va_parameters(env, ptr, mmu_idx, data);
+    ARMVAParameters param = aa64_va_parameters(env, ptr, mmu_idx, data, false);
     uint64_t pac, ext_ptr, ext, test;
     int bot_bit, top_bit;
 
@@ -355,7 +355,7 @@ static uint64_t pauth_auth(CPUARMState *env, uint64_t ptr, uint64_t modifier,
                            ARMPACKey *key, bool data, int keynumber)
 {
     ARMMMUIdx mmu_idx = arm_stage1_mmu_idx(env);
-    ARMVAParameters param = aa64_va_parameters(env, ptr, mmu_idx, data);
+    ARMVAParameters param = aa64_va_parameters(env, ptr, mmu_idx, data, false);
     int bot_bit, top_bit;
     uint64_t pac, orig_ptr, test;
 
@@ -379,7 +379,7 @@ static uint64_t pauth_auth(CPUARMState *env, uint64_t ptr, uint64_t modifier,
 static uint64_t pauth_strip(CPUARMState *env, uint64_t ptr, bool data)
 {
     ARMMMUIdx mmu_idx = arm_stage1_mmu_idx(env);
-    ARMVAParameters param = aa64_va_parameters(env, ptr, mmu_idx, data);
+    ARMVAParameters param = aa64_va_parameters(env, ptr, mmu_idx, data, false);
 
     return pauth_original_ptr(ptr, param);
 }
diff --git a/target/arm/tcg/sve_helper.c b/target/arm/tcg/sve_helper.c
index ccf5e5beca..0097522470 100644
--- a/target/arm/tcg/sve_helper.c
+++ b/target/arm/tcg/sve_helper.c
@@ -6727,6 +6727,7 @@ void sve_ldff1_z(CPUARMState *env, void *vd, uint64_t *vg, void *vm,
     intptr_t reg_off;
     SVEHostPage info;
     target_ulong addr, in_page;
+    ARMVectorReg scratch;
 
     /* Skip to the first true predicate.  */
     reg_off = find_next_active(vg, 0, reg_max, esz);
@@ -6736,6 +6737,11 @@ void sve_ldff1_z(CPUARMState *env, void *vd, uint64_t *vg, void *vm,
         return;
     }
 
+    /* Protect against overlap between vd and vm. */
+    if (unlikely(vd == vm)) {
+        vm = memcpy(&scratch, vm, reg_max);
+    }
+
     /*
      * Probe the first element, allowing faults.
      */
diff --git a/target/arm/sve_ldst_internal.h b/target/arm/tcg/sve_ldst_internal.h
index 4f159ec4ad..4f159ec4ad 100644
--- a/target/arm/sve_ldst_internal.h
+++ b/target/arm/tcg/sve_ldst_internal.h
diff --git a/target/arm/translate-a32.h b/target/arm/tcg/translate-a32.h
index 48a15379d2..48a15379d2 100644
--- a/target/arm/translate-a32.h
+++ b/target/arm/tcg/translate-a32.h
diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c
index dff391bfe2..741a608739 100644
--- a/target/arm/tcg/translate-a64.c
+++ b/target/arm/tcg/translate-a64.c
@@ -56,6 +56,13 @@ enum a64_shift_type {
     A64_SHIFT_TYPE_ROR = 3
 };
 
+/*
+ * Include the generated decoders.
+ */
+
+#include "decode-sme-fa64.c.inc"
+#include "decode-a64.c.inc"
+
 /* Table based decoder typedefs - used when the relevant bits for decode
  * are too awkwardly scattered across the instruction (eg SIMD).
  */
@@ -675,83 +682,102 @@ static inline void gen_logic_CC(int sf, TCGv_i64 result)
 }
 
 /* dest = T0 + T1; compute C, N, V and Z flags */
-static void gen_add_CC(int sf, TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
+static void gen_add64_CC(TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
 {
-    if (sf) {
-        TCGv_i64 result, flag, tmp;
-        result = tcg_temp_new_i64();
-        flag = tcg_temp_new_i64();
-        tmp = tcg_temp_new_i64();
+    TCGv_i64 result, flag, tmp;
+    result = tcg_temp_new_i64();
+    flag = tcg_temp_new_i64();
+    tmp = tcg_temp_new_i64();
 
-        tcg_gen_movi_i64(tmp, 0);
-        tcg_gen_add2_i64(result, flag, t0, tmp, t1, tmp);
+    tcg_gen_movi_i64(tmp, 0);
+    tcg_gen_add2_i64(result, flag, t0, tmp, t1, tmp);
 
-        tcg_gen_extrl_i64_i32(cpu_CF, flag);
+    tcg_gen_extrl_i64_i32(cpu_CF, flag);
 
-        gen_set_NZ64(result);
+    gen_set_NZ64(result);
 
-        tcg_gen_xor_i64(flag, result, t0);
-        tcg_gen_xor_i64(tmp, t0, t1);
-        tcg_gen_andc_i64(flag, flag, tmp);
-        tcg_gen_extrh_i64_i32(cpu_VF, flag);
+    tcg_gen_xor_i64(flag, result, t0);
+    tcg_gen_xor_i64(tmp, t0, t1);
+    tcg_gen_andc_i64(flag, flag, tmp);
+    tcg_gen_extrh_i64_i32(cpu_VF, flag);
 
-        tcg_gen_mov_i64(dest, result);
-    } else {
-        /* 32 bit arithmetic */
-        TCGv_i32 t0_32 = tcg_temp_new_i32();
-        TCGv_i32 t1_32 = tcg_temp_new_i32();
-        TCGv_i32 tmp = tcg_temp_new_i32();
+    tcg_gen_mov_i64(dest, result);
+}
 
-        tcg_gen_movi_i32(tmp, 0);
-        tcg_gen_extrl_i64_i32(t0_32, t0);
-        tcg_gen_extrl_i64_i32(t1_32, t1);
-        tcg_gen_add2_i32(cpu_NF, cpu_CF, t0_32, tmp, t1_32, tmp);
-        tcg_gen_mov_i32(cpu_ZF, cpu_NF);
-        tcg_gen_xor_i32(cpu_VF, cpu_NF, t0_32);
-        tcg_gen_xor_i32(tmp, t0_32, t1_32);
-        tcg_gen_andc_i32(cpu_VF, cpu_VF, tmp);
-        tcg_gen_extu_i32_i64(dest, cpu_NF);
+static void gen_add32_CC(TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
+{
+    TCGv_i32 t0_32 = tcg_temp_new_i32();
+    TCGv_i32 t1_32 = tcg_temp_new_i32();
+    TCGv_i32 tmp = tcg_temp_new_i32();
+
+    tcg_gen_movi_i32(tmp, 0);
+    tcg_gen_extrl_i64_i32(t0_32, t0);
+    tcg_gen_extrl_i64_i32(t1_32, t1);
+    tcg_gen_add2_i32(cpu_NF, cpu_CF, t0_32, tmp, t1_32, tmp);
+    tcg_gen_mov_i32(cpu_ZF, cpu_NF);
+    tcg_gen_xor_i32(cpu_VF, cpu_NF, t0_32);
+    tcg_gen_xor_i32(tmp, t0_32, t1_32);
+    tcg_gen_andc_i32(cpu_VF, cpu_VF, tmp);
+    tcg_gen_extu_i32_i64(dest, cpu_NF);
+}
+
+static void gen_add_CC(int sf, TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
+{
+    if (sf) {
+        gen_add64_CC(dest, t0, t1);
+    } else {
+        gen_add32_CC(dest, t0, t1);
     }
 }
 
 /* dest = T0 - T1; compute C, N, V and Z flags */
-static void gen_sub_CC(int sf, TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
+static void gen_sub64_CC(TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
 {
-    if (sf) {
-        /* 64 bit arithmetic */
-        TCGv_i64 result, flag, tmp;
+    /* 64 bit arithmetic */
+    TCGv_i64 result, flag, tmp;
 
-        result = tcg_temp_new_i64();
-        flag = tcg_temp_new_i64();
-        tcg_gen_sub_i64(result, t0, t1);
+    result = tcg_temp_new_i64();
+    flag = tcg_temp_new_i64();
+    tcg_gen_sub_i64(result, t0, t1);
 
-        gen_set_NZ64(result);
+    gen_set_NZ64(result);
 
-        tcg_gen_setcond_i64(TCG_COND_GEU, flag, t0, t1);
-        tcg_gen_extrl_i64_i32(cpu_CF, flag);
+    tcg_gen_setcond_i64(TCG_COND_GEU, flag, t0, t1);
+    tcg_gen_extrl_i64_i32(cpu_CF, flag);
 
-        tcg_gen_xor_i64(flag, result, t0);
-        tmp = tcg_temp_new_i64();
-        tcg_gen_xor_i64(tmp, t0, t1);
-        tcg_gen_and_i64(flag, flag, tmp);
-        tcg_gen_extrh_i64_i32(cpu_VF, flag);
-        tcg_gen_mov_i64(dest, result);
-    } else {
-        /* 32 bit arithmetic */
-        TCGv_i32 t0_32 = tcg_temp_new_i32();
-        TCGv_i32 t1_32 = tcg_temp_new_i32();
-        TCGv_i32 tmp;
+    tcg_gen_xor_i64(flag, result, t0);
+    tmp = tcg_temp_new_i64();
+    tcg_gen_xor_i64(tmp, t0, t1);
+    tcg_gen_and_i64(flag, flag, tmp);
+    tcg_gen_extrh_i64_i32(cpu_VF, flag);
+    tcg_gen_mov_i64(dest, result);
+}
 
-        tcg_gen_extrl_i64_i32(t0_32, t0);
-        tcg_gen_extrl_i64_i32(t1_32, t1);
-        tcg_gen_sub_i32(cpu_NF, t0_32, t1_32);
-        tcg_gen_mov_i32(cpu_ZF, cpu_NF);
-        tcg_gen_setcond_i32(TCG_COND_GEU, cpu_CF, t0_32, t1_32);
-        tcg_gen_xor_i32(cpu_VF, cpu_NF, t0_32);
-        tmp = tcg_temp_new_i32();
-        tcg_gen_xor_i32(tmp, t0_32, t1_32);
-        tcg_gen_and_i32(cpu_VF, cpu_VF, tmp);
-        tcg_gen_extu_i32_i64(dest, cpu_NF);
+static void gen_sub32_CC(TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
+{
+    /* 32 bit arithmetic */
+    TCGv_i32 t0_32 = tcg_temp_new_i32();
+    TCGv_i32 t1_32 = tcg_temp_new_i32();
+    TCGv_i32 tmp;
+
+    tcg_gen_extrl_i64_i32(t0_32, t0);
+    tcg_gen_extrl_i64_i32(t1_32, t1);
+    tcg_gen_sub_i32(cpu_NF, t0_32, t1_32);
+    tcg_gen_mov_i32(cpu_ZF, cpu_NF);
+    tcg_gen_setcond_i32(TCG_COND_GEU, cpu_CF, t0_32, t1_32);
+    tcg_gen_xor_i32(cpu_VF, cpu_NF, t0_32);
+    tmp = tcg_temp_new_i32();
+    tcg_gen_xor_i32(tmp, t0_32, t1_32);
+    tcg_gen_and_i32(cpu_VF, cpu_VF, tmp);
+    tcg_gen_extu_i32_i64(dest, cpu_NF);
+}
+
+static void gen_sub_CC(int sf, TCGv_i64 dest, TCGv_i64 t0, TCGv_i64 t1)
+{
+    if (sf) {
+        gen_sub64_CC(dest, t0, t1);
+    } else {
+        gen_sub32_CC(dest, t0, t1);
     }
 }
 
@@ -1293,116 +1319,279 @@ static inline AArch64DecodeFn *lookup_disas_fn(const AArch64DecodeTable *table,
  * match up with those in the manual.
  */
 
-/* Unconditional branch (immediate)
- *   31  30       26 25                                  0
- * +----+-----------+-------------------------------------+
- * | op | 0 0 1 0 1 |                 imm26               |
- * +----+-----------+-------------------------------------+
- */
-static void disas_uncond_b_imm(DisasContext *s, uint32_t insn)
+static bool trans_B(DisasContext *s, arg_i *a)
 {
-    int64_t diff = sextract32(insn, 0, 26) * 4;
-
-    if (insn & (1U << 31)) {
-        /* BL Branch with link */
-        gen_pc_plus_diff(s, cpu_reg(s, 30), curr_insn_len(s));
-    }
+    reset_btype(s);
+    gen_goto_tb(s, 0, a->imm);
+    return true;
+}
 
-    /* B Branch / BL Branch with link */
+static bool trans_BL(DisasContext *s, arg_i *a)
+{
+    gen_pc_plus_diff(s, cpu_reg(s, 30), curr_insn_len(s));
     reset_btype(s);
-    gen_goto_tb(s, 0, diff);
+    gen_goto_tb(s, 0, a->imm);
+    return true;
 }
 
-/* Compare and branch (immediate)
- *   31  30         25  24  23                  5 4      0
- * +----+-------------+----+---------------------+--------+
- * | sf | 0 1 1 0 1 0 | op |         imm19       |   Rt   |
- * +----+-------------+----+---------------------+--------+
- */
-static void disas_comp_b_imm(DisasContext *s, uint32_t insn)
+
+static bool trans_CBZ(DisasContext *s, arg_cbz *a)
 {
-    unsigned int sf, op, rt;
-    int64_t diff;
     DisasLabel match;
     TCGv_i64 tcg_cmp;
 
-    sf = extract32(insn, 31, 1);
-    op = extract32(insn, 24, 1); /* 0: CBZ; 1: CBNZ */
-    rt = extract32(insn, 0, 5);
-    diff = sextract32(insn, 5, 19) * 4;
-
-    tcg_cmp = read_cpu_reg(s, rt, sf);
+    tcg_cmp = read_cpu_reg(s, a->rt, a->sf);
     reset_btype(s);
 
     match = gen_disas_label(s);
-    tcg_gen_brcondi_i64(op ? TCG_COND_NE : TCG_COND_EQ,
+    tcg_gen_brcondi_i64(a->nz ? TCG_COND_NE : TCG_COND_EQ,
                         tcg_cmp, 0, match.label);
     gen_goto_tb(s, 0, 4);
     set_disas_label(s, match);
-    gen_goto_tb(s, 1, diff);
+    gen_goto_tb(s, 1, a->imm);
+    return true;
 }
 
-/* Test and branch (immediate)
- *   31  30         25  24  23   19 18          5 4    0
- * +----+-------------+----+-------+-------------+------+
- * | b5 | 0 1 1 0 1 1 | op |  b40  |    imm14    |  Rt  |
- * +----+-------------+----+-------+-------------+------+
- */
-static void disas_test_b_imm(DisasContext *s, uint32_t insn)
+static bool trans_TBZ(DisasContext *s, arg_tbz *a)
 {
-    unsigned int bit_pos, op, rt;
-    int64_t diff;
     DisasLabel match;
     TCGv_i64 tcg_cmp;
 
-    bit_pos = (extract32(insn, 31, 1) << 5) | extract32(insn, 19, 5);
-    op = extract32(insn, 24, 1); /* 0: TBZ; 1: TBNZ */
-    diff = sextract32(insn, 5, 14) * 4;
-    rt = extract32(insn, 0, 5);
-
     tcg_cmp = tcg_temp_new_i64();
-    tcg_gen_andi_i64(tcg_cmp, cpu_reg(s, rt), (1ULL << bit_pos));
+    tcg_gen_andi_i64(tcg_cmp, cpu_reg(s, a->rt), 1ULL << a->bitpos);
 
     reset_btype(s);
 
     match = gen_disas_label(s);
-    tcg_gen_brcondi_i64(op ? TCG_COND_NE : TCG_COND_EQ,
+    tcg_gen_brcondi_i64(a->nz ? TCG_COND_NE : TCG_COND_EQ,
                         tcg_cmp, 0, match.label);
     gen_goto_tb(s, 0, 4);
     set_disas_label(s, match);
-    gen_goto_tb(s, 1, diff);
+    gen_goto_tb(s, 1, a->imm);
+    return true;
 }
 
-/* Conditional branch (immediate)
- *  31           25  24  23                  5   4  3    0
- * +---------------+----+---------------------+----+------+
- * | 0 1 0 1 0 1 0 | o1 |         imm19       | o0 | cond |
- * +---------------+----+---------------------+----+------+
- */
-static void disas_cond_b_imm(DisasContext *s, uint32_t insn)
+static bool trans_B_cond(DisasContext *s, arg_B_cond *a)
 {
-    unsigned int cond;
-    int64_t diff;
-
-    if ((insn & (1 << 4)) || (insn & (1 << 24))) {
-        unallocated_encoding(s);
-        return;
-    }
-    diff = sextract32(insn, 5, 19) * 4;
-    cond = extract32(insn, 0, 4);
-
     reset_btype(s);
-    if (cond < 0x0e) {
+    if (a->cond < 0x0e) {
         /* genuinely conditional branches */
         DisasLabel match = gen_disas_label(s);
-        arm_gen_test_cc(cond, match.label);
+        arm_gen_test_cc(a->cond, match.label);
         gen_goto_tb(s, 0, 4);
         set_disas_label(s, match);
-        gen_goto_tb(s, 1, diff);
+        gen_goto_tb(s, 1, a->imm);
     } else {
         /* 0xe and 0xf are both "always" conditions */
-        gen_goto_tb(s, 0, diff);
+        gen_goto_tb(s, 0, a->imm);
+    }
+    return true;
+}
+
+static void set_btype_for_br(DisasContext *s, int rn)
+{
+    if (dc_isar_feature(aa64_bti, s)) {
+        /* BR to {x16,x17} or !guard -> 1, else 3.  */
+        set_btype(s, rn == 16 || rn == 17 || !s->guarded_page ? 1 : 3);
+    }
+}
+
+static void set_btype_for_blr(DisasContext *s)
+{
+    if (dc_isar_feature(aa64_bti, s)) {
+        /* BLR sets BTYPE to 2, regardless of source guarded page.  */
+        set_btype(s, 2);
+    }
+}
+
+static bool trans_BR(DisasContext *s, arg_r *a)
+{
+    gen_a64_set_pc(s, cpu_reg(s, a->rn));
+    set_btype_for_br(s, a->rn);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_BLR(DisasContext *s, arg_r *a)
+{
+    TCGv_i64 dst = cpu_reg(s, a->rn);
+    TCGv_i64 lr = cpu_reg(s, 30);
+    if (dst == lr) {
+        TCGv_i64 tmp = tcg_temp_new_i64();
+        tcg_gen_mov_i64(tmp, dst);
+        dst = tmp;
+    }
+    gen_pc_plus_diff(s, lr, curr_insn_len(s));
+    gen_a64_set_pc(s, dst);
+    set_btype_for_blr(s);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_RET(DisasContext *s, arg_r *a)
+{
+    gen_a64_set_pc(s, cpu_reg(s, a->rn));
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static TCGv_i64 auth_branch_target(DisasContext *s, TCGv_i64 dst,
+                                   TCGv_i64 modifier, bool use_key_a)
+{
+    TCGv_i64 truedst;
+    /*
+     * Return the branch target for a BRAA/RETA/etc, which is either
+     * just the destination dst, or that value with the pauth check
+     * done and the code removed from the high bits.
+     */
+    if (!s->pauth_active) {
+        return dst;
+    }
+
+    truedst = tcg_temp_new_i64();
+    if (use_key_a) {
+        gen_helper_autia(truedst, cpu_env, dst, modifier);
+    } else {
+        gen_helper_autib(truedst, cpu_env, dst, modifier);
+    }
+    return truedst;
+}
+
+static bool trans_BRAZ(DisasContext *s, arg_braz *a)
+{
+    TCGv_i64 dst;
+
+    if (!dc_isar_feature(aa64_pauth, s)) {
+        return false;
+    }
+
+    dst = auth_branch_target(s, cpu_reg(s, a->rn), tcg_constant_i64(0), !a->m);
+    gen_a64_set_pc(s, dst);
+    set_btype_for_br(s, a->rn);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_BLRAZ(DisasContext *s, arg_braz *a)
+{
+    TCGv_i64 dst, lr;
+
+    if (!dc_isar_feature(aa64_pauth, s)) {
+        return false;
+    }
+
+    dst = auth_branch_target(s, cpu_reg(s, a->rn), tcg_constant_i64(0), !a->m);
+    lr = cpu_reg(s, 30);
+    if (dst == lr) {
+        TCGv_i64 tmp = tcg_temp_new_i64();
+        tcg_gen_mov_i64(tmp, dst);
+        dst = tmp;
+    }
+    gen_pc_plus_diff(s, lr, curr_insn_len(s));
+    gen_a64_set_pc(s, dst);
+    set_btype_for_blr(s);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_RETA(DisasContext *s, arg_reta *a)
+{
+    TCGv_i64 dst;
+
+    dst = auth_branch_target(s, cpu_reg(s, 30), cpu_X[31], !a->m);
+    gen_a64_set_pc(s, dst);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_BRA(DisasContext *s, arg_bra *a)
+{
+    TCGv_i64 dst;
+
+    if (!dc_isar_feature(aa64_pauth, s)) {
+        return false;
+    }
+    dst = auth_branch_target(s, cpu_reg(s,a->rn), cpu_reg_sp(s, a->rm), !a->m);
+    gen_a64_set_pc(s, dst);
+    set_btype_for_br(s, a->rn);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_BLRA(DisasContext *s, arg_bra *a)
+{
+    TCGv_i64 dst, lr;
+
+    if (!dc_isar_feature(aa64_pauth, s)) {
+        return false;
+    }
+    dst = auth_branch_target(s, cpu_reg(s, a->rn), cpu_reg_sp(s, a->rm), !a->m);
+    lr = cpu_reg(s, 30);
+    if (dst == lr) {
+        TCGv_i64 tmp = tcg_temp_new_i64();
+        tcg_gen_mov_i64(tmp, dst);
+        dst = tmp;
+    }
+    gen_pc_plus_diff(s, lr, curr_insn_len(s));
+    gen_a64_set_pc(s, dst);
+    set_btype_for_blr(s);
+    s->base.is_jmp = DISAS_JUMP;
+    return true;
+}
+
+static bool trans_ERET(DisasContext *s, arg_ERET *a)
+{
+    TCGv_i64 dst;
+
+    if (s->current_el == 0) {
+        return false;
+    }
+    if (s->fgt_eret) {
+        gen_exception_insn_el(s, 0, EXCP_UDEF, 0, 2);
+        return true;
+    }
+    dst = tcg_temp_new_i64();
+    tcg_gen_ld_i64(dst, cpu_env,
+                   offsetof(CPUARMState, elr_el[s->current_el]));
+
+    if (tb_cflags(s->base.tb) & CF_USE_ICOUNT) {
+        gen_io_start();
     }
+
+    gen_helper_exception_return(cpu_env, dst);
+    /* Must exit loop to check un-masked IRQs */
+    s->base.is_jmp = DISAS_EXIT;
+    return true;
+}
+
+static bool trans_ERETA(DisasContext *s, arg_reta *a)
+{
+    TCGv_i64 dst;
+
+    if (!dc_isar_feature(aa64_pauth, s)) {
+        return false;
+    }
+    if (s->current_el == 0) {
+        return false;
+    }
+    /* The FGT trap takes precedence over an auth trap. */
+    if (s->fgt_eret) {
+        gen_exception_insn_el(s, 0, EXCP_UDEF, a->m ? 3 : 2, 2);
+        return true;
+    }
+    dst = tcg_temp_new_i64();
+    tcg_gen_ld_i64(dst, cpu_env,
+                   offsetof(CPUARMState, elr_el[s->current_el]));
+
+    dst = auth_branch_target(s, dst, cpu_X[31], !a->m);
+    if (tb_cflags(s->base.tb) & CF_USE_ICOUNT) {
+        gen_io_start();
+    }
+
+    gen_helper_exception_return(cpu_env, dst);
+    /* Must exit loop to check un-masked IRQs */
+    s->base.is_jmp = DISAS_EXIT;
+    return true;
 }
 
 /* HINT instruction group, including various allocated HINTs */
@@ -2173,233 +2362,10 @@ static void disas_exc(DisasContext *s, uint32_t insn)
     }
 }
 
-/* Unconditional branch (register)
- *  31           25 24   21 20   16 15   10 9    5 4     0
- * +---------------+-------+-------+-------+------+-------+
- * | 1 1 0 1 0 1 1 |  opc  |  op2  |  op3  |  Rn  |  op4  |
- * +---------------+-------+-------+-------+------+-------+
- */
-static void disas_uncond_b_reg(DisasContext *s, uint32_t insn)
-{
-    unsigned int opc, op2, op3, rn, op4;
-    unsigned btype_mod = 2;   /* 0: BR, 1: BLR, 2: other */
-    TCGv_i64 dst;
-    TCGv_i64 modifier;
-
-    opc = extract32(insn, 21, 4);
-    op2 = extract32(insn, 16, 5);
-    op3 = extract32(insn, 10, 6);
-    rn = extract32(insn, 5, 5);
-    op4 = extract32(insn, 0, 5);
-
-    if (op2 != 0x1f) {
-        goto do_unallocated;
-    }
-
-    switch (opc) {
-    case 0: /* BR */
-    case 1: /* BLR */
-    case 2: /* RET */
-        btype_mod = opc;
-        switch (op3) {
-        case 0:
-            /* BR, BLR, RET */
-            if (op4 != 0) {
-                goto do_unallocated;
-            }
-            dst = cpu_reg(s, rn);
-            break;
-
-        case 2:
-        case 3:
-            if (!dc_isar_feature(aa64_pauth, s)) {
-                goto do_unallocated;
-            }
-            if (opc == 2) {
-                /* RETAA, RETAB */
-                if (rn != 0x1f || op4 != 0x1f) {
-                    goto do_unallocated;
-                }
-                rn = 30;
-                modifier = cpu_X[31];
-            } else {
-                /* BRAAZ, BRABZ, BLRAAZ, BLRABZ */
-                if (op4 != 0x1f) {
-                    goto do_unallocated;
-                }
-                modifier = tcg_constant_i64(0);
-            }
-            if (s->pauth_active) {
-                dst = tcg_temp_new_i64();
-                if (op3 == 2) {
-                    gen_helper_autia(dst, cpu_env, cpu_reg(s, rn), modifier);
-                } else {
-                    gen_helper_autib(dst, cpu_env, cpu_reg(s, rn), modifier);
-                }
-            } else {
-                dst = cpu_reg(s, rn);
-            }
-            break;
-
-        default:
-            goto do_unallocated;
-        }
-        /* BLR also needs to load return address */
-        if (opc == 1) {
-            TCGv_i64 lr = cpu_reg(s, 30);
-            if (dst == lr) {
-                TCGv_i64 tmp = tcg_temp_new_i64();
-                tcg_gen_mov_i64(tmp, dst);
-                dst = tmp;
-            }
-            gen_pc_plus_diff(s, lr, curr_insn_len(s));
-        }
-        gen_a64_set_pc(s, dst);
-        break;
-
-    case 8: /* BRAA */
-    case 9: /* BLRAA */
-        if (!dc_isar_feature(aa64_pauth, s)) {
-            goto do_unallocated;
-        }
-        if ((op3 & ~1) != 2) {
-            goto do_unallocated;
-        }
-        btype_mod = opc & 1;
-        if (s->pauth_active) {
-            dst = tcg_temp_new_i64();
-            modifier = cpu_reg_sp(s, op4);
-            if (op3 == 2) {
-                gen_helper_autia(dst, cpu_env, cpu_reg(s, rn), modifier);
-            } else {
-                gen_helper_autib(dst, cpu_env, cpu_reg(s, rn), modifier);
-            }
-        } else {
-            dst = cpu_reg(s, rn);
-        }
-        /* BLRAA also needs to load return address */
-        if (opc == 9) {
-            TCGv_i64 lr = cpu_reg(s, 30);
-            if (dst == lr) {
-                TCGv_i64 tmp = tcg_temp_new_i64();
-                tcg_gen_mov_i64(tmp, dst);
-                dst = tmp;
-            }
-            gen_pc_plus_diff(s, lr, curr_insn_len(s));
-        }
-        gen_a64_set_pc(s, dst);
-        break;
-
-    case 4: /* ERET */
-        if (s->current_el == 0) {
-            goto do_unallocated;
-        }
-        switch (op3) {
-        case 0: /* ERET */
-            if (op4 != 0) {
-                goto do_unallocated;
-            }
-            if (s->fgt_eret) {
-                gen_exception_insn_el(s, 0, EXCP_UDEF, syn_erettrap(op3), 2);
-                return;
-            }
-            dst = tcg_temp_new_i64();
-            tcg_gen_ld_i64(dst, cpu_env,
-                           offsetof(CPUARMState, elr_el[s->current_el]));
-            break;
-
-        case 2: /* ERETAA */
-        case 3: /* ERETAB */
-            if (!dc_isar_feature(aa64_pauth, s)) {
-                goto do_unallocated;
-            }
-            if (rn != 0x1f || op4 != 0x1f) {
-                goto do_unallocated;
-            }
-            /* The FGT trap takes precedence over an auth trap. */
-            if (s->fgt_eret) {
-                gen_exception_insn_el(s, 0, EXCP_UDEF, syn_erettrap(op3), 2);
-                return;
-            }
-            dst = tcg_temp_new_i64();
-            tcg_gen_ld_i64(dst, cpu_env,
-                           offsetof(CPUARMState, elr_el[s->current_el]));
-            if (s->pauth_active) {
-                modifier = cpu_X[31];
-                if (op3 == 2) {
-                    gen_helper_autia(dst, cpu_env, dst, modifier);
-                } else {
-                    gen_helper_autib(dst, cpu_env, dst, modifier);
-                }
-            }
-            break;
-
-        default:
-            goto do_unallocated;
-        }
-        if (tb_cflags(s->base.tb) & CF_USE_ICOUNT) {
-            gen_io_start();
-        }
-
-        gen_helper_exception_return(cpu_env, dst);
-        /* Must exit loop to check un-masked IRQs */
-        s->base.is_jmp = DISAS_EXIT;
-        return;
-
-    case 5: /* DRPS */
-        if (op3 != 0 || op4 != 0 || rn != 0x1f) {
-            goto do_unallocated;
-        } else {
-            unallocated_encoding(s);
-        }
-        return;
-
-    default:
-    do_unallocated:
-        unallocated_encoding(s);
-        return;
-    }
-
-    switch (btype_mod) {
-    case 0: /* BR */
-        if (dc_isar_feature(aa64_bti, s)) {
-            /* BR to {x16,x17} or !guard -> 1, else 3.  */
-            set_btype(s, rn == 16 || rn == 17 || !s->guarded_page ? 1 : 3);
-        }
-        break;
-
-    case 1: /* BLR */
-        if (dc_isar_feature(aa64_bti, s)) {
-            /* BLR sets BTYPE to 2, regardless of source guarded page.  */
-            set_btype(s, 2);
-        }
-        break;
-
-    default: /* RET or none of the above.  */
-        /* BTYPE will be set to 0 by normal end-of-insn processing.  */
-        break;
-    }
-
-    s->base.is_jmp = DISAS_JUMP;
-}
-
 /* Branches, exception generating and system instructions */
 static void disas_b_exc_sys(DisasContext *s, uint32_t insn)
 {
     switch (extract32(insn, 25, 7)) {
-    case 0x0a: case 0x0b:
-    case 0x4a: case 0x4b: /* Unconditional branch (immediate) */
-        disas_uncond_b_imm(s, insn);
-        break;
-    case 0x1a: case 0x5a: /* Compare & branch (immediate) */
-        disas_comp_b_imm(s, insn);
-        break;
-    case 0x1b: case 0x5b: /* Test & branch (immediate) */
-        disas_test_b_imm(s, insn);
-        break;
-    case 0x2a: /* Conditional branch (immediate) */
-        disas_cond_b_imm(s, insn);
-        break;
     case 0x6a: /* Exception generation / System */
         if (insn & (1 << 24)) {
             if (extract32(insn, 22, 2) == 0) {
@@ -2411,9 +2377,6 @@ static void disas_b_exc_sys(DisasContext *s, uint32_t insn)
             disas_exc(s, insn);
         }
         break;
-    case 0x6b: /* Unconditional branch (register) */
-        disas_uncond_b_reg(s, insn);
-        break;
     default:
         unallocated_encoding(s);
         break;
@@ -4172,132 +4135,82 @@ static void disas_ldst(DisasContext *s, uint32_t insn)
     }
 }
 
-/* PC-rel. addressing
- *   31  30   29 28       24 23                5 4    0
- * +----+-------+-----------+-------------------+------+
- * | op | immlo | 1 0 0 0 0 |       immhi       |  Rd  |
- * +----+-------+-----------+-------------------+------+
- */
-static void disas_pc_rel_adr(DisasContext *s, uint32_t insn)
-{
-    unsigned int page, rd;
-    int64_t offset;
+typedef void ArithTwoOp(TCGv_i64, TCGv_i64, TCGv_i64);
 
-    page = extract32(insn, 31, 1);
-    /* SignExtend(immhi:immlo) -> offset */
-    offset = sextract64(insn, 5, 19);
-    offset = offset << 2 | extract32(insn, 29, 2);
-    rd = extract32(insn, 0, 5);
+static bool gen_rri(DisasContext *s, arg_rri_sf *a,
+                    bool rd_sp, bool rn_sp, ArithTwoOp *fn)
+{
+    TCGv_i64 tcg_rn = rn_sp ? cpu_reg_sp(s, a->rn) : cpu_reg(s, a->rn);
+    TCGv_i64 tcg_rd = rd_sp ? cpu_reg_sp(s, a->rd) : cpu_reg(s, a->rd);
+    TCGv_i64 tcg_imm = tcg_constant_i64(a->imm);
 
-    if (page) {
-        /* ADRP (page based) */
-        offset <<= 12;
-        /* The page offset is ok for CF_PCREL. */
-        offset -= s->pc_curr & 0xfff;
+    fn(tcg_rd, tcg_rn, tcg_imm);
+    if (!a->sf) {
+        tcg_gen_ext32u_i64(tcg_rd, tcg_rd);
     }
-
-    gen_pc_plus_diff(s, cpu_reg(s, rd), offset);
+    return true;
 }
 
 /*
- * Add/subtract (immediate)
- *
- *  31 30 29 28         23 22 21         10 9   5 4   0
- * +--+--+--+-------------+--+-------------+-----+-----+
- * |sf|op| S| 1 0 0 0 1 0 |sh|    imm12    |  Rn | Rd  |
- * +--+--+--+-------------+--+-------------+-----+-----+
- *
- *    sf: 0 -> 32bit, 1 -> 64bit
- *    op: 0 -> add  , 1 -> sub
- *     S: 1 -> set flags
- *    sh: 1 -> LSL imm by 12
+ * PC-rel. addressing
  */
-static void disas_add_sub_imm(DisasContext *s, uint32_t insn)
-{
-    int rd = extract32(insn, 0, 5);
-    int rn = extract32(insn, 5, 5);
-    uint64_t imm = extract32(insn, 10, 12);
-    bool shift = extract32(insn, 22, 1);
-    bool setflags = extract32(insn, 29, 1);
-    bool sub_op = extract32(insn, 30, 1);
-    bool is_64bit = extract32(insn, 31, 1);
-
-    TCGv_i64 tcg_rn = cpu_reg_sp(s, rn);
-    TCGv_i64 tcg_rd = setflags ? cpu_reg(s, rd) : cpu_reg_sp(s, rd);
-    TCGv_i64 tcg_result;
 
-    if (shift) {
-        imm <<= 12;
-    }
+static bool trans_ADR(DisasContext *s, arg_ri *a)
+{
+    gen_pc_plus_diff(s, cpu_reg(s, a->rd), a->imm);
+    return true;
+}
 
-    tcg_result = tcg_temp_new_i64();
-    if (!setflags) {
-        if (sub_op) {
-            tcg_gen_subi_i64(tcg_result, tcg_rn, imm);
-        } else {
-            tcg_gen_addi_i64(tcg_result, tcg_rn, imm);
-        }
-    } else {
-        TCGv_i64 tcg_imm = tcg_constant_i64(imm);
-        if (sub_op) {
-            gen_sub_CC(is_64bit, tcg_result, tcg_rn, tcg_imm);
-        } else {
-            gen_add_CC(is_64bit, tcg_result, tcg_rn, tcg_imm);
-        }
-    }
+static bool trans_ADRP(DisasContext *s, arg_ri *a)
+{
+    int64_t offset = (int64_t)a->imm << 12;
 
-    if (is_64bit) {
-        tcg_gen_mov_i64(tcg_rd, tcg_result);
-    } else {
-        tcg_gen_ext32u_i64(tcg_rd, tcg_result);
-    }
+    /* The page offset is ok for CF_PCREL. */
+    offset -= s->pc_curr & 0xfff;
+    gen_pc_plus_diff(s, cpu_reg(s, a->rd), offset);
+    return true;
 }
 
 /*
+ * Add/subtract (immediate)
+ */
+TRANS(ADD_i, gen_rri, a, 1, 1, tcg_gen_add_i64)
+TRANS(SUB_i, gen_rri, a, 1, 1, tcg_gen_sub_i64)
+TRANS(ADDS_i, gen_rri, a, 0, 1, a->sf ? gen_add64_CC : gen_add32_CC)
+TRANS(SUBS_i, gen_rri, a, 0, 1, a->sf ? gen_sub64_CC : gen_sub32_CC)
+
+/*
  * Add/subtract (immediate, with tags)
- *
- *  31 30 29 28         23 22 21     16 14      10 9   5 4   0
- * +--+--+--+-------------+--+---------+--+-------+-----+-----+
- * |sf|op| S| 1 0 0 0 1 1 |o2|  uimm6  |o3| uimm4 |  Rn | Rd  |
- * +--+--+--+-------------+--+---------+--+-------+-----+-----+
- *
- *    op: 0 -> add, 1 -> sub
  */
-static void disas_add_sub_imm_with_tags(DisasContext *s, uint32_t insn)
+
+static bool gen_add_sub_imm_with_tags(DisasContext *s, arg_rri_tag *a,
+                                      bool sub_op)
 {
-    int rd = extract32(insn, 0, 5);
-    int rn = extract32(insn, 5, 5);
-    int uimm4 = extract32(insn, 10, 4);
-    int uimm6 = extract32(insn, 16, 6);
-    bool sub_op = extract32(insn, 30, 1);
     TCGv_i64 tcg_rn, tcg_rd;
     int imm;
 
-    /* Test all of sf=1, S=0, o2=0, o3=0.  */
-    if ((insn & 0xa040c000u) != 0x80000000u ||
-        !dc_isar_feature(aa64_mte_insn_reg, s)) {
-        unallocated_encoding(s);
-        return;
-    }
-
-    imm = uimm6 << LOG2_TAG_GRANULE;
+    imm = a->uimm6 << LOG2_TAG_GRANULE;
     if (sub_op) {
         imm = -imm;
     }
 
-    tcg_rn = cpu_reg_sp(s, rn);
-    tcg_rd = cpu_reg_sp(s, rd);
+    tcg_rn = cpu_reg_sp(s, a->rn);
+    tcg_rd = cpu_reg_sp(s, a->rd);
 
     if (s->ata) {
         gen_helper_addsubg(tcg_rd, cpu_env, tcg_rn,
                            tcg_constant_i32(imm),
-                           tcg_constant_i32(uimm4));
+                           tcg_constant_i32(a->uimm4));
     } else {
         tcg_gen_addi_i64(tcg_rd, tcg_rn, imm);
         gen_address_with_allocation_tag0(tcg_rd, tcg_rd);
     }
+    return true;
 }
 
+TRANS_FEAT(ADDG_i, aa64_mte_insn_reg, gen_add_sub_imm_with_tags, a, false)
+TRANS_FEAT(SUBG_i, aa64_mte_insn_reg, gen_add_sub_imm_with_tags, a, true)
+
 /* The input should be a value in the bottom e bits (with higher
  * bits zero); returns that value replicated into every element
  * of size e in a 64 bit integer.
@@ -4312,14 +4225,12 @@ static uint64_t bitfield_replicate(uint64_t mask, unsigned int e)
     return mask;
 }
 
-/* Return a value with the bottom len bits set (where 0 < len <= 64) */
-static inline uint64_t bitmask64(unsigned int length)
-{
-    assert(length > 0 && length <= 64);
-    return ~0ULL >> (64 - length);
-}
+/*
+ * Logical (immediate)
+ */
 
-/* Simplified variant of pseudocode DecodeBitMasks() for the case where we
+/*
+ * Simplified variant of pseudocode DecodeBitMasks() for the case where we
  * only require the wmask. Returns false if the imms/immr/immn are a reserved
  * value (ie should cause a guest UNDEF exception), and true if they are
  * valid, in which case the decoded bit pattern is written to result.
@@ -4374,10 +4285,10 @@ bool logic_imm_decode_wmask(uint64_t *result, unsigned int immn,
     /* Create the value of one element: s+1 set bits rotated
      * by r within the element (which is e bits wide)...
      */
-    mask = bitmask64(s + 1);
+    mask = MAKE_64BIT_MASK(0, s + 1);
     if (r) {
         mask = (mask >> r) | (mask << (e - r));
-        mask &= bitmask64(e);
+        mask &= MAKE_64BIT_MASK(0, e);
     }
     /* ...then replicate the element over the whole 64 bit value */
     mask = bitfield_replicate(mask, e);
@@ -4385,295 +4296,215 @@ bool logic_imm_decode_wmask(uint64_t *result, unsigned int immn,
     return true;
 }
 
-/* Logical (immediate)
- *   31  30 29 28         23 22  21  16 15  10 9    5 4    0
- * +----+-----+-------------+---+------+------+------+------+
- * | sf | opc | 1 0 0 1 0 0 | N | immr | imms |  Rn  |  Rd  |
- * +----+-----+-------------+---+------+------+------+------+
- */
-static void disas_logic_imm(DisasContext *s, uint32_t insn)
+static bool gen_rri_log(DisasContext *s, arg_rri_log *a, bool set_cc,
+                        void (*fn)(TCGv_i64, TCGv_i64, int64_t))
 {
-    unsigned int sf, opc, is_n, immr, imms, rn, rd;
     TCGv_i64 tcg_rd, tcg_rn;
-    uint64_t wmask;
-    bool is_and = false;
-
-    sf = extract32(insn, 31, 1);
-    opc = extract32(insn, 29, 2);
-    is_n = extract32(insn, 22, 1);
-    immr = extract32(insn, 16, 6);
-    imms = extract32(insn, 10, 6);
-    rn = extract32(insn, 5, 5);
-    rd = extract32(insn, 0, 5);
-
-    if (!sf && is_n) {
-        unallocated_encoding(s);
-        return;
-    }
+    uint64_t imm;
 
-    if (opc == 0x3) { /* ANDS */
-        tcg_rd = cpu_reg(s, rd);
-    } else {
-        tcg_rd = cpu_reg_sp(s, rd);
+    /* Some immediate field values are reserved. */
+    if (!logic_imm_decode_wmask(&imm, extract32(a->dbm, 12, 1),
+                                extract32(a->dbm, 0, 6),
+                                extract32(a->dbm, 6, 6))) {
+        return false;
     }
-    tcg_rn = cpu_reg(s, rn);
-
-    if (!logic_imm_decode_wmask(&wmask, is_n, imms, immr)) {
-        /* some immediate field values are reserved */
-        unallocated_encoding(s);
-        return;
+    if (!a->sf) {
+        imm &= 0xffffffffull;
     }
 
-    if (!sf) {
-        wmask &= 0xffffffff;
-    }
+    tcg_rd = set_cc ? cpu_reg(s, a->rd) : cpu_reg_sp(s, a->rd);
+    tcg_rn = cpu_reg(s, a->rn);
 
-    switch (opc) {
-    case 0x3: /* ANDS */
-    case 0x0: /* AND */
-        tcg_gen_andi_i64(tcg_rd, tcg_rn, wmask);
-        is_and = true;
-        break;
-    case 0x1: /* ORR */
-        tcg_gen_ori_i64(tcg_rd, tcg_rn, wmask);
-        break;
-    case 0x2: /* EOR */
-        tcg_gen_xori_i64(tcg_rd, tcg_rn, wmask);
-        break;
-    default:
-        assert(FALSE); /* must handle all above */
-        break;
+    fn(tcg_rd, tcg_rn, imm);
+    if (set_cc) {
+        gen_logic_CC(a->sf, tcg_rd);
     }
-
-    if (!sf && !is_and) {
-        /* zero extend final result; we know we can skip this for AND
-         * since the immediate had the high 32 bits clear.
-         */
+    if (!a->sf) {
         tcg_gen_ext32u_i64(tcg_rd, tcg_rd);
     }
-
-    if (opc == 3) { /* ANDS */
-        gen_logic_CC(sf, tcg_rd);
-    }
+    return true;
 }
 
+TRANS(AND_i, gen_rri_log, a, false, tcg_gen_andi_i64)
+TRANS(ORR_i, gen_rri_log, a, false, tcg_gen_ori_i64)
+TRANS(EOR_i, gen_rri_log, a, false, tcg_gen_xori_i64)
+TRANS(ANDS_i, gen_rri_log, a, true, tcg_gen_andi_i64)
+
 /*
  * Move wide (immediate)
- *
- *  31 30 29 28         23 22 21 20             5 4    0
- * +--+-----+-------------+-----+----------------+------+
- * |sf| opc | 1 0 0 1 0 1 |  hw |  imm16         |  Rd  |
- * +--+-----+-------------+-----+----------------+------+
- *
- * sf: 0 -> 32 bit, 1 -> 64 bit
- * opc: 00 -> N, 10 -> Z, 11 -> K
- * hw: shift/16 (0,16, and sf only 32, 48)
  */
-static void disas_movw_imm(DisasContext *s, uint32_t insn)
+
+static bool trans_MOVZ(DisasContext *s, arg_movw *a)
 {
-    int rd = extract32(insn, 0, 5);
-    uint64_t imm = extract32(insn, 5, 16);
-    int sf = extract32(insn, 31, 1);
-    int opc = extract32(insn, 29, 2);
-    int pos = extract32(insn, 21, 2) << 4;
-    TCGv_i64 tcg_rd = cpu_reg(s, rd);
+    int pos = a->hw << 4;
+    tcg_gen_movi_i64(cpu_reg(s, a->rd), (uint64_t)a->imm << pos);
+    return true;
+}
 
-    if (!sf && (pos >= 32)) {
-        unallocated_encoding(s);
-        return;
+static bool trans_MOVN(DisasContext *s, arg_movw *a)
+{
+    int pos = a->hw << 4;
+    uint64_t imm = a->imm;
+
+    imm = ~(imm << pos);
+    if (!a->sf) {
+        imm = (uint32_t)imm;
     }
+    tcg_gen_movi_i64(cpu_reg(s, a->rd), imm);
+    return true;
+}
 
-    switch (opc) {
-    case 0: /* MOVN */
-    case 2: /* MOVZ */
-        imm <<= pos;
-        if (opc == 0) {
-            imm = ~imm;
-        }
-        if (!sf) {
-            imm &= 0xffffffffu;
-        }
-        tcg_gen_movi_i64(tcg_rd, imm);
-        break;
-    case 3: /* MOVK */
-        tcg_gen_deposit_i64(tcg_rd, tcg_rd, tcg_constant_i64(imm), pos, 16);
-        if (!sf) {
-            tcg_gen_ext32u_i64(tcg_rd, tcg_rd);
-        }
-        break;
-    default:
-        unallocated_encoding(s);
-        break;
+static bool trans_MOVK(DisasContext *s, arg_movw *a)
+{
+    int pos = a->hw << 4;
+    TCGv_i64 tcg_rd, tcg_im;
+
+    tcg_rd = cpu_reg(s, a->rd);
+    tcg_im = tcg_constant_i64(a->imm);
+    tcg_gen_deposit_i64(tcg_rd, tcg_rd, tcg_im, pos, 16);
+    if (!a->sf) {
+        tcg_gen_ext32u_i64(tcg_rd, tcg_rd);
     }
+    return true;
 }
 
-/* Bitfield
- *   31  30 29 28         23 22  21  16 15  10 9    5 4    0
- * +----+-----+-------------+---+------+------+------+------+
- * | sf | opc | 1 0 0 1 1 0 | N | immr | imms |  Rn  |  Rd  |
- * +----+-----+-------------+---+------+------+------+------+
+/*
+ * Bitfield
  */
-static void disas_bitfield(DisasContext *s, uint32_t insn)
+
+static bool trans_SBFM(DisasContext *s, arg_SBFM *a)
 {
-    unsigned int sf, n, opc, ri, si, rn, rd, bitsize, pos, len;
-    TCGv_i64 tcg_rd, tcg_tmp;
+    TCGv_i64 tcg_rd = cpu_reg(s, a->rd);
+    TCGv_i64 tcg_tmp = read_cpu_reg(s, a->rn, 1);
+    unsigned int bitsize = a->sf ? 64 : 32;
+    unsigned int ri = a->immr;
+    unsigned int si = a->imms;
+    unsigned int pos, len;
 
-    sf = extract32(insn, 31, 1);
-    opc = extract32(insn, 29, 2);
-    n = extract32(insn, 22, 1);
-    ri = extract32(insn, 16, 6);
-    si = extract32(insn, 10, 6);
-    rn = extract32(insn, 5, 5);
-    rd = extract32(insn, 0, 5);
-    bitsize = sf ? 64 : 32;
+    if (si >= ri) {
+        /* Wd<s-r:0> = Wn<s:r> */
+        len = (si - ri) + 1;
+        tcg_gen_sextract_i64(tcg_rd, tcg_tmp, ri, len);
+        if (!a->sf) {
+            tcg_gen_ext32u_i64(tcg_rd, tcg_rd);
+        }
+    } else {
+        /* Wd<32+s-r,32-r> = Wn<s:0> */
+        len = si + 1;
+        pos = (bitsize - ri) & (bitsize - 1);
 
-    if (sf != n || ri >= bitsize || si >= bitsize || opc > 2) {
-        unallocated_encoding(s);
-        return;
+        if (len < ri) {
+            /*
+             * Sign extend the destination field from len to fill the
+             * balance of the word.  Let the deposit below insert all
+             * of those sign bits.
+             */
+            tcg_gen_sextract_i64(tcg_tmp, tcg_tmp, 0, len);
+            len = ri;
+        }
+
+        /*
+         * We start with zero, and we haven't modified any bits outside
+         * bitsize, therefore no final zero-extension is unneeded for !sf.
+         */
+        tcg_gen_deposit_z_i64(tcg_rd, tcg_tmp, pos, len);
     }
+    return true;
+}
 
-    tcg_rd = cpu_reg(s, rd);
+static bool trans_UBFM(DisasContext *s, arg_UBFM *a)
+{
+    TCGv_i64 tcg_rd = cpu_reg(s, a->rd);
+    TCGv_i64 tcg_tmp = read_cpu_reg(s, a->rn, 1);
+    unsigned int bitsize = a->sf ? 64 : 32;
+    unsigned int ri = a->immr;
+    unsigned int si = a->imms;
+    unsigned int pos, len;
 
-    /* Suppress the zero-extend for !sf.  Since RI and SI are constrained
-       to be smaller than bitsize, we'll never reference data outside the
-       low 32-bits anyway.  */
-    tcg_tmp = read_cpu_reg(s, rn, 1);
+    tcg_rd = cpu_reg(s, a->rd);
+    tcg_tmp = read_cpu_reg(s, a->rn, 1);
 
-    /* Recognize simple(r) extractions.  */
     if (si >= ri) {
         /* Wd<s-r:0> = Wn<s:r> */
         len = (si - ri) + 1;
-        if (opc == 0) { /* SBFM: ASR, SBFX, SXTB, SXTH, SXTW */
-            tcg_gen_sextract_i64(tcg_rd, tcg_tmp, ri, len);
-            goto done;
-        } else if (opc == 2) { /* UBFM: UBFX, LSR, UXTB, UXTH */
-            tcg_gen_extract_i64(tcg_rd, tcg_tmp, ri, len);
-            return;
-        }
-        /* opc == 1, BFXIL fall through to deposit */
-        tcg_gen_shri_i64(tcg_tmp, tcg_tmp, ri);
-        pos = 0;
+        tcg_gen_extract_i64(tcg_rd, tcg_tmp, ri, len);
     } else {
-        /* Handle the ri > si case with a deposit
-         * Wd<32+s-r,32-r> = Wn<s:0>
-         */
+        /* Wd<32+s-r,32-r> = Wn<s:0> */
         len = si + 1;
         pos = (bitsize - ri) & (bitsize - 1);
+        tcg_gen_deposit_z_i64(tcg_rd, tcg_tmp, pos, len);
     }
+    return true;
+}
 
-    if (opc == 0 && len < ri) {
-        /* SBFM: sign extend the destination field from len to fill
-           the balance of the word.  Let the deposit below insert all
-           of those sign bits.  */
-        tcg_gen_sextract_i64(tcg_tmp, tcg_tmp, 0, len);
-        len = ri;
-    }
+static bool trans_BFM(DisasContext *s, arg_BFM *a)
+{
+    TCGv_i64 tcg_rd = cpu_reg(s, a->rd);
+    TCGv_i64 tcg_tmp = read_cpu_reg(s, a->rn, 1);
+    unsigned int bitsize = a->sf ? 64 : 32;
+    unsigned int ri = a->immr;
+    unsigned int si = a->imms;
+    unsigned int pos, len;
+
+    tcg_rd = cpu_reg(s, a->rd);
+    tcg_tmp = read_cpu_reg(s, a->rn, 1);
 
-    if (opc == 1) { /* BFM, BFXIL */
-        tcg_gen_deposit_i64(tcg_rd, tcg_rd, tcg_tmp, pos, len);
+    if (si >= ri) {
+        /* Wd<s-r:0> = Wn<s:r> */
+        tcg_gen_shri_i64(tcg_tmp, tcg_tmp, ri);
+        len = (si - ri) + 1;
+        pos = 0;
     } else {
-        /* SBFM or UBFM: We start with zero, and we haven't modified
-           any bits outside bitsize, therefore the zero-extension
-           below is unneeded.  */
-        tcg_gen_deposit_z_i64(tcg_rd, tcg_tmp, pos, len);
-        return;
+        /* Wd<32+s-r,32-r> = Wn<s:0> */
+        len = si + 1;
+        pos = (bitsize - ri) & (bitsize - 1);
     }
 
- done:
-    if (!sf) { /* zero extend final result */
+    tcg_gen_deposit_i64(tcg_rd, tcg_rd, tcg_tmp, pos, len);
+    if (!a->sf) {
         tcg_gen_ext32u_i64(tcg_rd, tcg_rd);
     }
+    return true;
 }
 
-/* Extract
- *   31  30  29 28         23 22   21  20  16 15    10 9    5 4    0
- * +----+------+-------------+---+----+------+--------+------+------+
- * | sf | op21 | 1 0 0 1 1 1 | N | o0 |  Rm  |  imms  |  Rn  |  Rd  |
- * +----+------+-------------+---+----+------+--------+------+------+
- */
-static void disas_extract(DisasContext *s, uint32_t insn)
+static bool trans_EXTR(DisasContext *s, arg_extract *a)
 {
-    unsigned int sf, n, rm, imm, rn, rd, bitsize, op21, op0;
+    TCGv_i64 tcg_rd, tcg_rm, tcg_rn;
 
-    sf = extract32(insn, 31, 1);
-    n = extract32(insn, 22, 1);
-    rm = extract32(insn, 16, 5);
-    imm = extract32(insn, 10, 6);
-    rn = extract32(insn, 5, 5);
-    rd = extract32(insn, 0, 5);
-    op21 = extract32(insn, 29, 2);
-    op0 = extract32(insn, 21, 1);
-    bitsize = sf ? 64 : 32;
+    tcg_rd = cpu_reg(s, a->rd);
 
-    if (sf != n || op21 || op0 || imm >= bitsize) {
-        unallocated_encoding(s);
+    if (unlikely(a->imm == 0)) {
+        /*
+         * tcg shl_i32/shl_i64 is undefined for 32/64 bit shifts,
+         * so an extract from bit 0 is a special case.
+         */
+        if (a->sf) {
+            tcg_gen_mov_i64(tcg_rd, cpu_reg(s, a->rm));
+        } else {
+            tcg_gen_ext32u_i64(tcg_rd, cpu_reg(s, a->rm));
+        }
     } else {
-        TCGv_i64 tcg_rd, tcg_rm, tcg_rn;
+        tcg_rm = cpu_reg(s, a->rm);
+        tcg_rn = cpu_reg(s, a->rn);
 
-        tcg_rd = cpu_reg(s, rd);
-
-        if (unlikely(imm == 0)) {
-            /* tcg shl_i32/shl_i64 is undefined for 32/64 bit shifts,
-             * so an extract from bit 0 is a special case.
-             */
-            if (sf) {
-                tcg_gen_mov_i64(tcg_rd, cpu_reg(s, rm));
-            } else {
-                tcg_gen_ext32u_i64(tcg_rd, cpu_reg(s, rm));
-            }
+        if (a->sf) {
+            /* Specialization to ROR happens in EXTRACT2.  */
+            tcg_gen_extract2_i64(tcg_rd, tcg_rm, tcg_rn, a->imm);
         } else {
-            tcg_rm = cpu_reg(s, rm);
-            tcg_rn = cpu_reg(s, rn);
+            TCGv_i32 t0 = tcg_temp_new_i32();
 
-            if (sf) {
-                /* Specialization to ROR happens in EXTRACT2.  */
-                tcg_gen_extract2_i64(tcg_rd, tcg_rm, tcg_rn, imm);
+            tcg_gen_extrl_i64_i32(t0, tcg_rm);
+            if (a->rm == a->rn) {
+                tcg_gen_rotri_i32(t0, t0, a->imm);
             } else {
-                TCGv_i32 t0 = tcg_temp_new_i32();
-
-                tcg_gen_extrl_i64_i32(t0, tcg_rm);
-                if (rm == rn) {
-                    tcg_gen_rotri_i32(t0, t0, imm);
-                } else {
-                    TCGv_i32 t1 = tcg_temp_new_i32();
-                    tcg_gen_extrl_i64_i32(t1, tcg_rn);
-                    tcg_gen_extract2_i32(t0, t0, t1, imm);
-                }
-                tcg_gen_extu_i32_i64(tcg_rd, t0);
+                TCGv_i32 t1 = tcg_temp_new_i32();
+                tcg_gen_extrl_i64_i32(t1, tcg_rn);
+                tcg_gen_extract2_i32(t0, t0, t1, a->imm);
             }
+            tcg_gen_extu_i32_i64(tcg_rd, t0);
         }
     }
-}
-
-/* Data processing - immediate */
-static void disas_data_proc_imm(DisasContext *s, uint32_t insn)
-{
-    switch (extract32(insn, 23, 6)) {
-    case 0x20: case 0x21: /* PC-rel. addressing */
-        disas_pc_rel_adr(s, insn);
-        break;
-    case 0x22: /* Add/subtract (immediate) */
-        disas_add_sub_imm(s, insn);
-        break;
-    case 0x23: /* Add/subtract (immediate, with tags) */
-        disas_add_sub_imm_with_tags(s, insn);
-        break;
-    case 0x24: /* Logical (immediate) */
-        disas_logic_imm(s, insn);
-        break;
-    case 0x25: /* Move wide (immediate) */
-        disas_movw_imm(s, insn);
-        break;
-    case 0x26: /* Bitfield */
-        disas_bitfield(s, insn);
-        break;
-    case 0x27: /* Extract */
-        disas_extract(s, insn);
-        break;
-    default:
-        unallocated_encoding(s);
-        break;
-    }
+    return true;
 }
 
 /* Shift a TCGv src by TCGv shift_amount, put result in dst.
@@ -14100,12 +13931,6 @@ static void disas_data_proc_simd_fp(DisasContext *s, uint32_t insn)
     }
 }
 
-/*
- * Include the generated SME FA64 decoder.
- */
-
-#include "decode-sme-fa64.c.inc"
-
 static bool trans_OK(DisasContext *s, arg_OK *a)
 {
     return true;
@@ -14200,6 +14025,33 @@ static bool btype_destination_ok(uint32_t insn, bool bt, int btype)
     return false;
 }
 
+/* C3.1 A64 instruction index by encoding */
+static void disas_a64_legacy(DisasContext *s, uint32_t insn)
+{
+    switch (extract32(insn, 25, 4)) {
+    case 0xa: case 0xb: /* Branch, exception generation and system insns */
+        disas_b_exc_sys(s, insn);
+        break;
+    case 0x4:
+    case 0x6:
+    case 0xc:
+    case 0xe:      /* Loads and stores */
+        disas_ldst(s, insn);
+        break;
+    case 0x5:
+    case 0xd:      /* Data processing - register */
+        disas_data_proc_reg(s, insn);
+        break;
+    case 0x7:
+    case 0xf:      /* Data processing - SIMD and floating point */
+        disas_data_proc_simd_fp(s, insn);
+        break;
+    default:
+        unallocated_encoding(s);
+        break;
+    }
+}
+
 static void aarch64_tr_init_disas_context(DisasContextBase *dcbase,
                                           CPUState *cpu)
 {
@@ -14401,43 +14253,10 @@ static void aarch64_tr_translate_insn(DisasContextBase *dcbase, CPUState *cpu)
         disas_sme_fa64(s, insn);
     }
 
-    switch (extract32(insn, 25, 4)) {
-    case 0x0:
-        if (!extract32(insn, 31, 1) || !disas_sme(s, insn)) {
-            unallocated_encoding(s);
-        }
-        break;
-    case 0x1: case 0x3: /* UNALLOCATED */
-        unallocated_encoding(s);
-        break;
-    case 0x2:
-        if (!disas_sve(s, insn)) {
-            unallocated_encoding(s);
-        }
-        break;
-    case 0x8: case 0x9: /* Data processing - immediate */
-        disas_data_proc_imm(s, insn);
-        break;
-    case 0xa: case 0xb: /* Branch, exception generation and system insns */
-        disas_b_exc_sys(s, insn);
-        break;
-    case 0x4:
-    case 0x6:
-    case 0xc:
-    case 0xe:      /* Loads and stores */
-        disas_ldst(s, insn);
-        break;
-    case 0x5:
-    case 0xd:      /* Data processing - register */
-        disas_data_proc_reg(s, insn);
-        break;
-    case 0x7:
-    case 0xf:      /* Data processing - SIMD and floating point */
-        disas_data_proc_simd_fp(s, insn);
-        break;
-    default:
-        assert(FALSE); /* all 15 cases should be handled above */
-        break;
+    if (!disas_a64(s, insn) &&
+        !disas_sme(s, insn) &&
+        !disas_sve(s, insn)) {
+        disas_a64_legacy(s, insn);
     }
 
     /*
diff --git a/target/arm/tcg/translate.h b/target/arm/tcg/translate.h
index f02d4685b4..a9d1f4adc2 100644
--- a/target/arm/tcg/translate.h
+++ b/target/arm/tcg/translate.h
@@ -220,6 +220,11 @@ static inline int rsub_8(DisasContext *s, int x)
     return 8 - x;
 }
 
+static inline int shl_12(DisasContext *s, int x)
+{
+    return x << 12;
+}
+
 static inline int neon_3same_fp_size(DisasContext *s, int x)
 {
     /* Convert 0==fp32, 1==fp16 into a MO_* value */
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 4187759f10..a61cd6d99d 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -863,7 +863,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
             "tsx-ldtrk", NULL, NULL /* pconfig */, "arch-lbr",
             NULL, NULL, "amx-bf16", "avx512-fp16",
             "amx-tile", "amx-int8", "spec-ctrl", "stibp",
-            NULL, "arch-capabilities", "core-capability", "ssbd",
+            "flush-l1d", "arch-capabilities", "core-capability", "ssbd",
         },
         .cpuid = {
             .eax = 7,
@@ -1050,7 +1050,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
             "ssb-no", "mds-no", "pschange-mc-no", "tsx-ctrl",
             "taa-no", NULL, NULL, NULL,
             NULL, NULL, NULL, NULL,
-            NULL, NULL, NULL, NULL,
+            NULL, "fb-clear", NULL, NULL,
             NULL, NULL, NULL, NULL,
             NULL, NULL, NULL, NULL,
             NULL, NULL, NULL, NULL,
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 8ade71ab55..7201a71de8 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -899,6 +899,8 @@ uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
 #define CPUID_7_0_EDX_SPEC_CTRL         (1U << 26)
 /* Single Thread Indirect Branch Predictors */
 #define CPUID_7_0_EDX_STIBP             (1U << 27)
+/* Flush L1D cache */
+#define CPUID_7_0_EDX_FLUSH_L1D         (1U << 28)
 /* Arch Capabilities */
 #define CPUID_7_0_EDX_ARCH_CAPABILITIES (1U << 29)
 /* Core Capability */
@@ -1016,6 +1018,7 @@ uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
 #define MSR_ARCH_CAP_PSCHANGE_MC_NO     (1U << 6)
 #define MSR_ARCH_CAP_TSX_CTRL_MSR       (1U << 7)
 #define MSR_ARCH_CAP_TAA_NO             (1U << 8)
+#define MSR_ARCH_CAP_FB_CLEAR           (1U << 17)
 
 #define MSR_CORE_CAP_SPLIT_LOCK_DETECT  (1U << 5)
 
diff --git a/target/i386/ops_sse.h b/target/i386/ops_sse.h
index 0bd6bfad8a..fb63af7afa 100644
--- a/target/i386/ops_sse.h
+++ b/target/i386/ops_sse.h
@@ -2497,6 +2497,14 @@ void helper_vpermdq_ymm(Reg *d, Reg *v, Reg *s, uint32_t order)
     d->Q(1) = r1;
     d->Q(2) = r2;
     d->Q(3) = r3;
+    if (order & 0x8) {
+        d->Q(0) = 0;
+        d->Q(1) = 0;
+    }
+    if (order & 0x80) {
+        d->Q(2) = 0;
+        d->Q(3) = 0;
+    }
 }
 
 void helper_vpermq_ymm(Reg *d, Reg *s, uint32_t order)
diff --git a/target/i386/tcg/decode-new.c.inc b/target/i386/tcg/decode-new.c.inc
index 4fdd87750b..46afd9960b 100644
--- a/target/i386/tcg/decode-new.c.inc
+++ b/target/i386/tcg/decode-new.c.inc
@@ -237,7 +237,7 @@ static void decode_group14(DisasContext *s, CPUX86State *env, X86OpEntry *entry,
 static void decode_0F6F(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
 {
     static const X86OpEntry opcodes_0F6F[4] = {
-        X86_OP_ENTRY3(MOVDQ,       P,q, None,None, Q,q, vex1 mmx),  /* movq */
+        X86_OP_ENTRY3(MOVDQ,       P,q, None,None, Q,q, vex5 mmx),  /* movq */
         X86_OP_ENTRY3(MOVDQ,       V,x, None,None, W,x, vex1),      /* movdqa */
         X86_OP_ENTRY3(MOVDQ,       V,x, None,None, W,x, vex4_unal), /* movdqu */
         {},
@@ -274,9 +274,9 @@ static void decode_0F78(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
 {
     static const X86OpEntry opcodes_0F78[4] = {
         {},
-        X86_OP_ENTRY3(EXTRQ_i,       V,x, None,None, I,w,  cpuid(SSE4A)),
+        X86_OP_ENTRY3(EXTRQ_i,       V,x, None,None, I,w,  cpuid(SSE4A)), /* AMD extension */
         {},
-        X86_OP_ENTRY3(INSERTQ_i,     V,x, U,x, I,w,        cpuid(SSE4A)),
+        X86_OP_ENTRY3(INSERTQ_i,     V,x, U,x, I,w,        cpuid(SSE4A)), /* AMD extension */
     };
     *entry = *decode_by_prefix(s, opcodes_0F78);
 }
@@ -284,9 +284,9 @@ static void decode_0F78(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
 static void decode_0F79(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
 {
     if (s->prefix & PREFIX_REPNZ) {
-        entry->gen = gen_INSERTQ_r;
+        entry->gen = gen_INSERTQ_r; /* AMD extension */
     } else if (s->prefix & PREFIX_DATA) {
-        entry->gen = gen_EXTRQ_r;
+        entry->gen = gen_EXTRQ_r; /* AMD extension */
     } else {
         entry->gen = NULL;
     };
@@ -306,7 +306,7 @@ static void decode_0F7E(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
 static void decode_0F7F(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
 {
     static const X86OpEntry opcodes_0F7F[4] = {
-        X86_OP_ENTRY3(MOVDQ,       W,x, None,None, V,x, vex1 mmx), /* movq */
+        X86_OP_ENTRY3(MOVDQ,       W,x, None,None, V,x, vex5 mmx), /* movq */
         X86_OP_ENTRY3(MOVDQ,       W,x, None,None, V,x, vex1), /* movdqa */
         X86_OP_ENTRY3(MOVDQ,       W,x, None,None, V,x, vex4_unal), /* movdqu */
         {},
@@ -639,15 +639,15 @@ static void decode_0F10(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
     static const X86OpEntry opcodes_0F10_reg[4] = {
         X86_OP_ENTRY3(MOVDQ,   V,x,  None,None, W,x, vex4_unal), /* MOVUPS */
         X86_OP_ENTRY3(MOVDQ,   V,x,  None,None, W,x, vex4_unal), /* MOVUPD */
-        X86_OP_ENTRY3(VMOVSS,  V,x,  H,x,       W,x, vex4),
-        X86_OP_ENTRY3(VMOVLPx, V,x,  H,x,       W,x, vex4), /* MOVSD */
+        X86_OP_ENTRY3(VMOVSS,  V,x,  H,x,       W,x, vex5),
+        X86_OP_ENTRY3(VMOVLPx, V,x,  H,x,       W,x, vex5), /* MOVSD */
     };
 
     static const X86OpEntry opcodes_0F10_mem[4] = {
         X86_OP_ENTRY3(MOVDQ,      V,x,  None,None, W,x,  vex4_unal), /* MOVUPS */
         X86_OP_ENTRY3(MOVDQ,      V,x,  None,None, W,x,  vex4_unal), /* MOVUPD */
-        X86_OP_ENTRY3(VMOVSS_ld,  V,x,  H,x,       M,ss, vex4),
-        X86_OP_ENTRY3(VMOVSD_ld,  V,x,  H,x,       M,sd, vex4),
+        X86_OP_ENTRY3(VMOVSS_ld,  V,x,  H,x,       M,ss, vex5),
+        X86_OP_ENTRY3(VMOVSD_ld,  V,x,  H,x,       M,sd, vex5),
     };
 
     if ((get_modrm(s, env) >> 6) == 3) {
@@ -660,17 +660,17 @@ static void decode_0F10(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
 static void decode_0F11(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
 {
     static const X86OpEntry opcodes_0F11_reg[4] = {
-        X86_OP_ENTRY3(MOVDQ,   W,x,  None,None, V,x, vex4), /* MOVPS */
-        X86_OP_ENTRY3(MOVDQ,   W,x,  None,None, V,x, vex4), /* MOVPD */
-        X86_OP_ENTRY3(VMOVSS,  W,x,  H,x,       V,x, vex4),
-        X86_OP_ENTRY3(VMOVLPx, W,x,  H,x,       V,q, vex4), /* MOVSD */
+        X86_OP_ENTRY3(MOVDQ,   W,x,  None,None, V,x, vex4), /* MOVUPS */
+        X86_OP_ENTRY3(MOVDQ,   W,x,  None,None, V,x, vex4), /* MOVUPD */
+        X86_OP_ENTRY3(VMOVSS,  W,x,  H,x,       V,x, vex5),
+        X86_OP_ENTRY3(VMOVLPx, W,x,  H,x,       V,q, vex5), /* MOVSD */
     };
 
     static const X86OpEntry opcodes_0F11_mem[4] = {
-        X86_OP_ENTRY3(MOVDQ,      W,x,  None,None, V,x, vex4), /* MOVPS */
-        X86_OP_ENTRY3(MOVDQ,      W,x,  None,None, V,x, vex4), /* MOVPD */
-        X86_OP_ENTRY3(VMOVSS_st,  M,ss, None,None, V,x, vex4),
-        X86_OP_ENTRY3(VMOVLPx_st, M,sd, None,None, V,x, vex4), /* MOVSD */
+        X86_OP_ENTRY3(MOVDQ,      W,x,  None,None, V,x, vex4), /* MOVUPS */
+        X86_OP_ENTRY3(MOVDQ,      W,x,  None,None, V,x, vex4), /* MOVUPD */
+        X86_OP_ENTRY3(VMOVSS_st,  M,ss, None,None, V,x, vex5),
+        X86_OP_ENTRY3(VMOVLPx_st, M,sd, None,None, V,x, vex5), /* MOVSD */
     };
 
     if ((get_modrm(s, env) >> 6) == 3) {
@@ -687,16 +687,16 @@ static void decode_0F12(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
          * Use dq for operand for compatibility with gen_MOVSD and
          * to allow VEX128 only.
          */
-        X86_OP_ENTRY3(VMOVLPx_ld, V,dq, H,dq,      M,q, vex4), /* MOVLPS */
-        X86_OP_ENTRY3(VMOVLPx_ld, V,dq, H,dq,      M,q, vex4), /* MOVLPD */
+        X86_OP_ENTRY3(VMOVLPx_ld, V,dq, H,dq,      M,q, vex5), /* MOVLPS */
+        X86_OP_ENTRY3(VMOVLPx_ld, V,dq, H,dq,      M,q, vex5), /* MOVLPD */
         X86_OP_ENTRY3(VMOVSLDUP,  V,x,  None,None, W,x, vex4 cpuid(SSE3)),
-        X86_OP_ENTRY3(VMOVDDUP,   V,x,  None,None, WM,q, vex4 cpuid(SSE3)), /* qq if VEX.256 */
+        X86_OP_ENTRY3(VMOVDDUP,   V,x,  None,None, WM,q, vex5 cpuid(SSE3)), /* qq if VEX.256 */
     };
     static const X86OpEntry opcodes_0F12_reg[4] = {
-        X86_OP_ENTRY3(VMOVHLPS,  V,dq, H,dq,       U,dq, vex4),
-        X86_OP_ENTRY3(VMOVLPx,   W,x,  H,x,        U,q,  vex4), /* MOVLPD */
+        X86_OP_ENTRY3(VMOVHLPS,  V,dq, H,dq,       U,dq, vex7),
+        X86_OP_ENTRY3(VMOVLPx,   W,x,  H,x,        U,q,  vex5), /* MOVLPD */
         X86_OP_ENTRY3(VMOVSLDUP, V,x,  None,None,  U,x,  vex4 cpuid(SSE3)),
-        X86_OP_ENTRY3(VMOVDDUP,  V,x,  None,None,  U,x,  vex4 cpuid(SSE3)),
+        X86_OP_ENTRY3(VMOVDDUP,  V,x,  None,None,  U,x,  vex5 cpuid(SSE3)),
     };
 
     if ((get_modrm(s, env) >> 6) == 3) {
@@ -716,15 +716,15 @@ static void decode_0F16(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
          * Operand 1 technically only reads the low 64 bits, but uses dq so that
          * it is easier to check for op0 == op1 in an endianness-neutral manner.
          */
-        X86_OP_ENTRY3(VMOVHPx_ld, V,dq, H,dq,      M,q, vex4), /* MOVHPS */
-        X86_OP_ENTRY3(VMOVHPx_ld, V,dq, H,dq,      M,q, vex4), /* MOVHPD */
+        X86_OP_ENTRY3(VMOVHPx_ld, V,dq, H,dq,      M,q, vex5), /* MOVHPS */
+        X86_OP_ENTRY3(VMOVHPx_ld, V,dq, H,dq,      M,q, vex5), /* MOVHPD */
         X86_OP_ENTRY3(VMOVSHDUP,  V,x,  None,None, W,x, vex4 cpuid(SSE3)),
         {},
     };
     static const X86OpEntry opcodes_0F16_reg[4] = {
         /* Same as above, operand 1 could be Hq if it wasn't for big-endian.  */
-        X86_OP_ENTRY3(VMOVLHPS,  V,dq, H,dq,      U,q, vex4),
-        X86_OP_ENTRY3(VMOVHPx,   V,x,  H,x,       U,x, vex4), /* MOVHPD */
+        X86_OP_ENTRY3(VMOVLHPS,  V,dq, H,dq,      U,q, vex7),
+        X86_OP_ENTRY3(VMOVHPx,   V,x,  H,x,       U,x, vex5), /* MOVHPD */
         X86_OP_ENTRY3(VMOVSHDUP, V,x,  None,None, U,x, vex4 cpuid(SSE3)),
         {},
     };
@@ -750,8 +750,9 @@ static void decode_0F2A(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
 static void decode_0F2B(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
 {
     static const X86OpEntry opcodes_0F2B[4] = {
-        X86_OP_ENTRY3(MOVDQ,      M,x,  None,None, V,x, vex4), /* MOVNTPS */
-        X86_OP_ENTRY3(MOVDQ,      M,x,  None,None, V,x, vex4), /* MOVNTPD */
+        X86_OP_ENTRY3(MOVDQ,      M,x,  None,None, V,x, vex1), /* MOVNTPS */
+        X86_OP_ENTRY3(MOVDQ,      M,x,  None,None, V,x, vex1), /* MOVNTPD */
+        /* AMD extensions */
         X86_OP_ENTRY3(VMOVSS_st,  M,ss, None,None, V,x, vex4 cpuid(SSE4A)), /* MOVNTSS */
         X86_OP_ENTRY3(VMOVLPx_st, M,sd, None,None, V,x, vex4 cpuid(SSE4A)), /* MOVNTSD */
     };
@@ -783,6 +784,17 @@ static void decode_0F2D(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
     *entry = *decode_by_prefix(s, opcodes_0F2D);
 }
 
+static void decode_VxCOMISx(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
+{
+    /*
+     * VUCOMISx and VCOMISx are different and use no-prefix and 0x66 for SS and SD
+     * respectively.  Scalar values usually are associated with 0xF2 and 0xF3, for
+     * which X86_VEX_REPScalar exists, but here it has to be decoded by hand.
+     */
+    entry->s1 = entry->s2 = (s->prefix & PREFIX_DATA ? X86_SIZE_sd : X86_SIZE_ss);
+    entry->gen = (*b == 0x2E ? gen_VUCOMI : gen_VCOMI);
+}
+
 static void decode_sse_unary(DisasContext *s, CPUX86State *env, X86OpEntry *entry, uint8_t *b)
 {
     if (!(s->prefix & (PREFIX_REPZ | PREFIX_REPNZ))) {
@@ -813,7 +825,7 @@ static void decode_0FE6(DisasContext *s, CPUX86State *env, X86OpEntry *entry, ui
     static const X86OpEntry opcodes_0FE6[4] = {
         {},
         X86_OP_ENTRY2(VCVTTPD2DQ,  V,x, W,x,      vex2),
-        X86_OP_ENTRY2(VCVTDQ2PD,   V,x, W,x,      vex2),
+        X86_OP_ENTRY2(VCVTDQ2PD,   V,x, W,x,      vex5),
         X86_OP_ENTRY2(VCVTPD2DQ,   V,x, W,x,      vex2),
     };
     *entry = *decode_by_prefix(s, opcodes_0FE6);
@@ -831,17 +843,17 @@ static const X86OpEntry opcodes_0F[256] = {
     [0x10] = X86_OP_GROUP0(0F10),
     [0x11] = X86_OP_GROUP0(0F11),
     [0x12] = X86_OP_GROUP0(0F12),
-    [0x13] = X86_OP_ENTRY3(VMOVLPx_st,  M,q, None,None, V,q,  vex4 p_00_66),
+    [0x13] = X86_OP_ENTRY3(VMOVLPx_st,  M,q, None,None, V,q,  vex5 p_00_66),
     [0x14] = X86_OP_ENTRY3(VUNPCKLPx,   V,x, H,x, W,x,        vex4 p_00_66),
     [0x15] = X86_OP_ENTRY3(VUNPCKHPx,   V,x, H,x, W,x,        vex4 p_00_66),
     [0x16] = X86_OP_GROUP0(0F16),
     /* Incorrectly listed as Mq,Vq in the manual */
-    [0x17] = X86_OP_ENTRY3(VMOVHPx_st,  M,q, None,None, V,dq, vex4 p_00_66),
+    [0x17] = X86_OP_ENTRY3(VMOVHPx_st,  M,q, None,None, V,dq, vex5 p_00_66),
 
     [0x50] = X86_OP_ENTRY3(MOVMSK,     G,y, None,None, U,x, vex7 p_00_66),
-    [0x51] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
-    [0x52] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex4_rep5 p_00_f3),
-    [0x53] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex4_rep5 p_00_f3),
+    [0x51] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2), /* sqrtps */
+    [0x52] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex4_rep5 p_00_f3), /* rsqrtps */
+    [0x53] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex4_rep5 p_00_f3), /* rcpps */
     [0x54] = X86_OP_ENTRY3(PAND,       V,x, H,x, W,x,  vex4 p_00_66), /* vand */
     [0x55] = X86_OP_ENTRY3(PANDN,      V,x, H,x, W,x,  vex4 p_00_66), /* vandn */
     [0x56] = X86_OP_ENTRY3(POR,        V,x, H,x, W,x,  vex4 p_00_66), /* vor */
@@ -871,15 +883,15 @@ static const X86OpEntry opcodes_0F[256] = {
     [0x2B] = X86_OP_GROUP0(0F2B),
     [0x2C] = X86_OP_GROUP0(0F2C),
     [0x2D] = X86_OP_GROUP0(0F2D),
-    [0x2E] = X86_OP_ENTRY3(VUCOMI,     None,None, V,x, W,x,  vex4 p_00_66),
-    [0x2F] = X86_OP_ENTRY3(VCOMI,      None,None, V,x, W,x,  vex4 p_00_66),
+    [0x2E] = X86_OP_GROUP3(VxCOMISx,   None,None, V,x, W,x,  vex3 p_00_66), /* VUCOMISS/SD */
+    [0x2F] = X86_OP_GROUP3(VxCOMISx,   None,None, V,x, W,x,  vex3 p_00_66), /* VCOMISS/SD */
 
     [0x38] = X86_OP_GROUP0(0F38),
     [0x3a] = X86_OP_GROUP0(0F3A),
 
     [0x58] = X86_OP_ENTRY3(VADD,       V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
     [0x59] = X86_OP_ENTRY3(VMUL,       V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
-    [0x5a] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
+    [0x5a] = X86_OP_GROUP3(sse_unary,  V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2), /* CVTPS2PD */
     [0x5b] = X86_OP_GROUP0(0F5B),
     [0x5c] = X86_OP_ENTRY3(VSUB,       V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
     [0x5d] = X86_OP_ENTRY3(VMIN,       V,x, H,x, W,x, vex2_rep3 p_00_66_f3_f2),
diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc
index 95fb4f52fa..4fe8dec427 100644
--- a/target/i386/tcg/emit.c.inc
+++ b/target/i386/tcg/emit.c.inc
@@ -2285,7 +2285,7 @@ static void gen_VZEROALL(DisasContext *s, CPUX86State *env, X86DecodedInsn *deco
 {
     TCGv_ptr ptr = tcg_temp_new_ptr();
 
-    tcg_gen_addi_ptr(ptr, cpu_env, offsetof(CPUX86State, xmm_t0));
+    tcg_gen_addi_ptr(ptr, cpu_env, offsetof(CPUX86State, xmm_regs));
     gen_helper_memset(ptr, ptr, tcg_constant_i32(0),
                       tcg_constant_ptr(CPU_NB_REGS * sizeof(ZMMReg)));
 }
diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c
index 0ce4f796fa..61d748cfdc 100644
--- a/target/openrisc/cpu.c
+++ b/target/openrisc/cpu.c
@@ -22,6 +22,7 @@
 #include "qemu/qemu-print.h"
 #include "cpu.h"
 #include "exec/exec-all.h"
+#include "fpu/softfloat-helpers.h"
 #include "tcg/tcg.h"
 
 static void openrisc_cpu_set_pc(CPUState *cs, vaddr value)
@@ -90,6 +91,9 @@ static void openrisc_cpu_reset_hold(Object *obj)
     s->exception_index = -1;
     cpu_set_fpcsr(&cpu->env, 0);
 
+    set_float_detect_tininess(float_tininess_before_rounding,
+                              &cpu->env.fp_status);
+
 #ifndef CONFIG_USER_ONLY
     cpu->env.picmr = 0x00000000;
     cpu->env.picsr = 0x00000000;
diff --git a/target/openrisc/fpu_helper.c b/target/openrisc/fpu_helper.c
index f9e34fa2cc..8b81d2f62f 100644
--- a/target/openrisc/fpu_helper.c
+++ b/target/openrisc/fpu_helper.c
@@ -20,8 +20,8 @@
 
 #include "qemu/osdep.h"
 #include "cpu.h"
+#include "exec/exec-all.h"
 #include "exec/helper-proto.h"
-#include "exception.h"
 #include "fpu/softfloat.h"
 
 static int ieee_ex_to_openrisc(int fexcp)
@@ -45,6 +45,15 @@ static int ieee_ex_to_openrisc(int fexcp)
     return ret;
 }
 
+static G_NORETURN
+void do_fpe(CPUOpenRISCState *env, uintptr_t pc)
+{
+    CPUState *cs = env_cpu(env);
+
+    cs->exception_index = EXCP_FPE;
+    cpu_loop_exit_restore(cs, pc);
+}
+
 void HELPER(update_fpcsr)(CPUOpenRISCState *env)
 {
     int tmp = get_float_exception_flags(&env->fp_status);
@@ -55,7 +64,7 @@ void HELPER(update_fpcsr)(CPUOpenRISCState *env)
         if (tmp) {
             env->fpcsr |= tmp;
             if (env->fpcsr & FPCSR_FPEE) {
-                helper_exception(env, EXCP_FPE);
+                do_fpe(env, GETPC());
             }
         }
     }
diff --git a/target/openrisc/sys_helper.c b/target/openrisc/sys_helper.c
index ec145960e3..ccdee3b8be 100644
--- a/target/openrisc/sys_helper.c
+++ b/target/openrisc/sys_helper.c
@@ -29,17 +29,37 @@
 
 #define TO_SPR(group, number) (((group) << 11) + (number))
 
+static inline bool is_user(CPUOpenRISCState *env)
+{
+#ifdef CONFIG_USER_ONLY
+    return true;
+#else
+    return (env->sr & SR_SM) == 0;
+#endif
+}
+
 void HELPER(mtspr)(CPUOpenRISCState *env, target_ulong spr, target_ulong rb)
 {
-#ifndef CONFIG_USER_ONLY
     OpenRISCCPU *cpu = env_archcpu(env);
+#ifndef CONFIG_USER_ONLY
     CPUState *cs = env_cpu(env);
     target_ulong mr;
     int idx;
 #endif
 
+    /* Handle user accessible SPRs first.  */
     switch (spr) {
+    case TO_SPR(0, 20): /* FPCSR */
+        cpu_set_fpcsr(env, rb);
+        return;
+    }
+
+    if (is_user(env)) {
+        raise_exception(cpu, EXCP_ILLEGAL);
+    }
+
 #ifndef CONFIG_USER_ONLY
+    switch (spr) {
     case TO_SPR(0, 11): /* EVBAR */
         env->evbar = rb;
         break;
@@ -187,27 +207,33 @@ void HELPER(mtspr)(CPUOpenRISCState *env, target_ulong spr, target_ulong rb)
         cpu_openrisc_timer_update(cpu);
         qemu_mutex_unlock_iothread();
         break;
-#endif
-
-    case TO_SPR(0, 20): /* FPCSR */
-        cpu_set_fpcsr(env, rb);
-        break;
     }
+#endif
 }
 
 target_ulong HELPER(mfspr)(CPUOpenRISCState *env, target_ulong rd,
                            target_ulong spr)
 {
+    OpenRISCCPU *cpu = env_archcpu(env);
 #ifndef CONFIG_USER_ONLY
     uint64_t data[TARGET_INSN_START_WORDS];
     MachineState *ms = MACHINE(qdev_get_machine());
-    OpenRISCCPU *cpu = env_archcpu(env);
     CPUState *cs = env_cpu(env);
     int idx;
 #endif
 
+    /* Handle user accessible SPRs first.  */
     switch (spr) {
+    case TO_SPR(0, 20): /* FPCSR */
+        return env->fpcsr;
+    }
+
+    if (is_user(env)) {
+        raise_exception(cpu, EXCP_ILLEGAL);
+    }
+
 #ifndef CONFIG_USER_ONLY
+    switch (spr) {
     case TO_SPR(0, 0): /* VR */
         return env->vr;
 
@@ -324,11 +350,8 @@ target_ulong HELPER(mfspr)(CPUOpenRISCState *env, target_ulong rd,
         cpu_openrisc_count_update(cpu);
         qemu_mutex_unlock_iothread();
         return cpu_openrisc_count_get(cpu);
-#endif
-
-    case TO_SPR(0, 20): /* FPCSR */
-        return env->fpcsr;
     }
+#endif
 
     /* for rd is passed in, if rd unchanged, just keep it back.  */
     return rd;
diff --git a/target/openrisc/translate.c b/target/openrisc/translate.c
index 76e53c78d4..43ba0cc1ad 100644
--- a/target/openrisc/translate.c
+++ b/target/openrisc/translate.c
@@ -819,45 +819,12 @@ static bool trans_l_xori(DisasContext *dc, arg_rri *a)
 
 static bool trans_l_mfspr(DisasContext *dc, arg_l_mfspr *a)
 {
-    check_r0_write(dc, a->d);
-
-    if (is_user(dc)) {
-        gen_illegal_exception(dc);
-    } else {
-        TCGv spr = tcg_temp_new();
-
-        if (tb_cflags(dc->base.tb) & CF_USE_ICOUNT) {
-            gen_io_start();
-            if (dc->delayed_branch) {
-                tcg_gen_mov_tl(cpu_pc, jmp_pc);
-                tcg_gen_discard_tl(jmp_pc);
-            } else {
-                tcg_gen_movi_tl(cpu_pc, dc->base.pc_next + 4);
-            }
-            dc->base.is_jmp = DISAS_EXIT;
-        }
+    TCGv spr = tcg_temp_new();
 
-        tcg_gen_ori_tl(spr, cpu_R(dc, a->a), a->k);
-        gen_helper_mfspr(cpu_R(dc, a->d), cpu_env, cpu_R(dc, a->d), spr);
-    }
-    return true;
-}
-
-static bool trans_l_mtspr(DisasContext *dc, arg_l_mtspr *a)
-{
-    if (is_user(dc)) {
-        gen_illegal_exception(dc);
-    } else {
-        TCGv spr;
+    check_r0_write(dc, a->d);
 
-        if (tb_cflags(dc->base.tb) & CF_USE_ICOUNT) {
-            gen_io_start();
-        }
-        /* For SR, we will need to exit the TB to recognize the new
-         * exception state.  For NPC, in theory this counts as a branch
-         * (although the SPR only exists for use by an ICE).  Save all
-         * of the cpu state first, allowing it to be overwritten.
-         */
+    if (tb_cflags(dc->base.tb) & CF_USE_ICOUNT) {
+        gen_io_start();
         if (dc->delayed_branch) {
             tcg_gen_mov_tl(cpu_pc, jmp_pc);
             tcg_gen_discard_tl(jmp_pc);
@@ -865,11 +832,36 @@ static bool trans_l_mtspr(DisasContext *dc, arg_l_mtspr *a)
             tcg_gen_movi_tl(cpu_pc, dc->base.pc_next + 4);
         }
         dc->base.is_jmp = DISAS_EXIT;
+    }
+
+    tcg_gen_ori_tl(spr, cpu_R(dc, a->a), a->k);
+    gen_helper_mfspr(cpu_R(dc, a->d), cpu_env, cpu_R(dc, a->d), spr);
+    return true;
+}
+
+static bool trans_l_mtspr(DisasContext *dc, arg_l_mtspr *a)
+{
+    TCGv spr = tcg_temp_new();
 
-        spr = tcg_temp_new();
-        tcg_gen_ori_tl(spr, cpu_R(dc, a->a), a->k);
-        gen_helper_mtspr(cpu_env, spr, cpu_R(dc, a->b));
+    if (tb_cflags(dc->base.tb) & CF_USE_ICOUNT) {
+        gen_io_start();
     }
+    /*
+     * For SR, we will need to exit the TB to recognize the new
+     * exception state.  For NPC, in theory this counts as a branch
+     * (although the SPR only exists for use by an ICE).  Save all
+     * of the cpu state first, allowing it to be overwritten.
+     */
+    if (dc->delayed_branch) {
+        tcg_gen_mov_tl(cpu_pc, jmp_pc);
+        tcg_gen_discard_tl(jmp_pc);
+    } else {
+        tcg_gen_movi_tl(cpu_pc, dc->base.pc_next + 4);
+    }
+    dc->base.is_jmp = DISAS_EXIT;
+
+    tcg_gen_ori_tl(spr, cpu_R(dc, a->a), a->k);
+    gen_helper_mtspr(cpu_env, spr, cpu_R(dc, a->b));
     return true;
 }
 
diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c
index 40fdeaa905..df167493c3 100644
--- a/target/s390x/cpu.c
+++ b/target/s390x/cpu.c
@@ -140,6 +140,13 @@ static bool s390_cpu_has_work(CPUState *cs)
     return s390_cpu_has_int(cpu);
 }
 
+static void s390_query_cpu_fast(CPUState *cpu, CpuInfoFast *value)
+{
+    S390CPU *s390_cpu = S390_CPU(cpu);
+
+    value->u.s390x.cpu_state = s390_cpu->env.cpu_state;
+}
+
 /* S390CPUClass::reset() */
 static void s390_cpu_reset(CPUState *s, cpu_reset_type type)
 {
@@ -332,6 +339,7 @@ static void s390_cpu_class_init(ObjectClass *oc, void *data)
     cc->class_by_name = s390_cpu_class_by_name,
     cc->has_work = s390_cpu_has_work;
     cc->dump_state = s390_cpu_dump_state;
+    cc->query_cpu_fast = s390_query_cpu_fast;
     cc->set_pc = s390_cpu_set_pc;
     cc->get_pc = s390_cpu_get_pc;
     cc->gdb_read_register = s390_cpu_gdb_read_register;
diff --git a/target/s390x/tcg/insn-data.h.inc b/target/s390x/tcg/insn-data.h.inc
index 597d968b0e..1f1ac742a9 100644
--- a/target/s390x/tcg/insn-data.h.inc
+++ b/target/s390x/tcg/insn-data.h.inc
@@ -606,7 +606,7 @@
     F(0xed04, LDEB,    RXE,   Z,   0, m2_32u, new, f1, ldeb, 0, IF_BFP)
     F(0xed05, LXDB,    RXE,   Z,   0, m2_64, new_x, x1, lxdb, 0, IF_BFP)
     F(0xed06, LXEB,    RXE,   Z,   0, m2_32u, new_x, x1, lxeb, 0, IF_BFP)
-    F(0xb324, LDER,    RXE,   Z,   0, e2, new, f1, lde, 0, IF_AFP1)
+    F(0xb324, LDER,    RRE,   Z,   0, e2, new, f1, lde, 0, IF_AFP1)
     F(0xed24, LDE,     RXE,   Z,   0, m2_32u, new, f1, lde, 0, IF_AFP1)
 /* LOAD ROUNDED */
     F(0xb344, LEDBR,   RRF_e, Z,   0, f2, new, e1, ledb, 0, IF_BFP)
diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c
index a05205beb1..d6670e6a87 100644
--- a/target/s390x/tcg/translate.c
+++ b/target/s390x/tcg/translate.c
@@ -1534,18 +1534,51 @@ static DisasJumpType op_bal(DisasContext *s, DisasOps *o)
     }
 }
 
+/*
+ * Disassemble the target of a branch. The results are returned in a form
+ * suitable for passing into help_branch():
+ *
+ * - bool IS_IMM reflects whether the target is fixed or computed. Non-EXECUTEd
+ *   branches, whose DisasContext *S contains the relative immediate field RI,
+ *   are considered fixed. All the other branches are considered computed.
+ * - int IMM is the value of RI.
+ * - TCGv_i64 CDEST is the address of the computed target.
+ */
+#define disas_jdest(s, ri, is_imm, imm, cdest) do {                            \
+    if (have_field(s, ri)) {                                                   \
+        if (unlikely(s->ex_value)) {                                           \
+            cdest = tcg_temp_new_i64();                                        \
+            tcg_gen_ld_i64(cdest, cpu_env, offsetof(CPUS390XState, ex_target));\
+            tcg_gen_addi_i64(cdest, cdest, (int64_t)get_field(s, ri) * 2);     \
+            is_imm = false;                                                    \
+        } else {                                                               \
+            is_imm = true;                                                     \
+        }                                                                      \
+    } else {                                                                   \
+        is_imm = false;                                                        \
+    }                                                                          \
+    imm = is_imm ? get_field(s, ri) : 0;                                       \
+} while (false)
+
 static DisasJumpType op_basi(DisasContext *s, DisasOps *o)
 {
+    DisasCompare c;
+    bool is_imm;
+    int imm;
+
     pc_to_link_info(o->out, s, s->pc_tmp);
-    return help_goto_direct(s, s->base.pc_next + (int64_t)get_field(s, i2) * 2);
+
+    disas_jdest(s, i2, is_imm, imm, o->in2);
+    disas_jcc(s, &c, 0xf);
+    return help_branch(s, &c, is_imm, imm, o->in2);
 }
 
 static DisasJumpType op_bc(DisasContext *s, DisasOps *o)
 {
     int m1 = get_field(s, m1);
-    bool is_imm = have_field(s, i2);
-    int imm = is_imm ? get_field(s, i2) : 0;
     DisasCompare c;
+    bool is_imm;
+    int imm;
 
     /* BCR with R2 = 0 causes no branching */
     if (have_field(s, r2) && get_field(s, r2) == 0) {
@@ -1562,6 +1595,7 @@ static DisasJumpType op_bc(DisasContext *s, DisasOps *o)
         return DISAS_NEXT;
     }
 
+    disas_jdest(s, i2, is_imm, imm, o->in2);
     disas_jcc(s, &c, m1);
     return help_branch(s, &c, is_imm, imm, o->in2);
 }
@@ -1569,10 +1603,10 @@ static DisasJumpType op_bc(DisasContext *s, DisasOps *o)
 static DisasJumpType op_bct32(DisasContext *s, DisasOps *o)
 {
     int r1 = get_field(s, r1);
-    bool is_imm = have_field(s, i2);
-    int imm = is_imm ? get_field(s, i2) : 0;
     DisasCompare c;
+    bool is_imm;
     TCGv_i64 t;
+    int imm;
 
     c.cond = TCG_COND_NE;
     c.is_64 = false;
@@ -1584,6 +1618,7 @@ static DisasJumpType op_bct32(DisasContext *s, DisasOps *o)
     c.u.s32.b = tcg_constant_i32(0);
     tcg_gen_extrl_i64_i32(c.u.s32.a, t);
 
+    disas_jdest(s, i2, is_imm, imm, o->in2);
     return help_branch(s, &c, is_imm, imm, o->in2);
 }
 
@@ -1611,9 +1646,9 @@ static DisasJumpType op_bcth(DisasContext *s, DisasOps *o)
 static DisasJumpType op_bct64(DisasContext *s, DisasOps *o)
 {
     int r1 = get_field(s, r1);
-    bool is_imm = have_field(s, i2);
-    int imm = is_imm ? get_field(s, i2) : 0;
     DisasCompare c;
+    bool is_imm;
+    int imm;
 
     c.cond = TCG_COND_NE;
     c.is_64 = true;
@@ -1622,6 +1657,7 @@ static DisasJumpType op_bct64(DisasContext *s, DisasOps *o)
     c.u.s64.a = regs[r1];
     c.u.s64.b = tcg_constant_i64(0);
 
+    disas_jdest(s, i2, is_imm, imm, o->in2);
     return help_branch(s, &c, is_imm, imm, o->in2);
 }
 
@@ -1629,10 +1665,10 @@ static DisasJumpType op_bx32(DisasContext *s, DisasOps *o)
 {
     int r1 = get_field(s, r1);
     int r3 = get_field(s, r3);
-    bool is_imm = have_field(s, i2);
-    int imm = is_imm ? get_field(s, i2) : 0;
     DisasCompare c;
+    bool is_imm;
     TCGv_i64 t;
+    int imm;
 
     c.cond = (s->insn->data ? TCG_COND_LE : TCG_COND_GT);
     c.is_64 = false;
@@ -1645,6 +1681,7 @@ static DisasJumpType op_bx32(DisasContext *s, DisasOps *o)
     tcg_gen_extrl_i64_i32(c.u.s32.b, regs[r3 | 1]);
     store_reg32_i64(r1, t);
 
+    disas_jdest(s, i2, is_imm, imm, o->in2);
     return help_branch(s, &c, is_imm, imm, o->in2);
 }
 
@@ -1652,9 +1689,9 @@ static DisasJumpType op_bx64(DisasContext *s, DisasOps *o)
 {
     int r1 = get_field(s, r1);
     int r3 = get_field(s, r3);
-    bool is_imm = have_field(s, i2);
-    int imm = is_imm ? get_field(s, i2) : 0;
     DisasCompare c;
+    bool is_imm;
+    int imm;
 
     c.cond = (s->insn->data ? TCG_COND_LE : TCG_COND_GT);
     c.is_64 = true;
@@ -1668,6 +1705,7 @@ static DisasJumpType op_bx64(DisasContext *s, DisasOps *o)
     tcg_gen_add_i64(regs[r1], regs[r1], regs[r3]);
     c.u.s64.a = regs[r1];
 
+    disas_jdest(s, i2, is_imm, imm, o->in2);
     return help_branch(s, &c, is_imm, imm, o->in2);
 }
 
@@ -1685,10 +1723,9 @@ static DisasJumpType op_cj(DisasContext *s, DisasOps *o)
     c.u.s64.a = o->in1;
     c.u.s64.b = o->in2;
 
-    is_imm = have_field(s, i4);
-    if (is_imm) {
-        imm = get_field(s, i4);
-    } else {
+    o->out = NULL;
+    disas_jdest(s, i4, is_imm, imm, o->out);
+    if (!is_imm && !o->out) {
         imm = 0;
         o->out = get_address(s, 0, get_field(s, b4),
                              get_field(s, d4));
@@ -5764,15 +5801,13 @@ static void in2_a2(DisasContext *s, DisasOps *o)
 
 static TCGv gen_ri2(DisasContext *s)
 {
-    int64_t delta = (int64_t)get_field(s, i2) * 2;
-    TCGv ri2;
+    TCGv ri2 = NULL;
+    bool is_imm;
+    int imm;
 
-    if (unlikely(s->ex_value)) {
-        ri2 = tcg_temp_new_i64();
-        tcg_gen_ld_i64(ri2, cpu_env, offsetof(CPUS390XState, ex_target));
-        tcg_gen_addi_i64(ri2, ri2, delta);
-    } else {
-        ri2 = tcg_constant_i64(s->base.pc_next + delta);
+    disas_jdest(s, i2, is_imm, imm, ri2);
+    if (is_imm) {
+        ri2 = tcg_constant_i64(s->base.pc_next + imm * 2);
     }
 
     return ri2;