diff --git a/target-xtensa/helper.c b/target-xtensa/helper.c
index 11c318d..9e18984 100644
--- a/target-xtensa/helper.c
+++ b/target-xtensa/helper.c
@@ -38,6 +38,8 @@
         a1, a2, a3, a4, a5, a6) \
     { .targno = (no), .type = (typ), .group = (grp) },
 
+static void reset_mmu(CPUState *env);
+
 void cpu_reset(CPUXtensaState *env)
 {
     env->exception_taken = 0;
@@ -48,6 +50,7 @@
     env->sregs[VECBASE] = env->config->vecbase;
 
     env->pending_irq_level = 0;
+    reset_mmu(env);
 }
 
 static const XtensaConfig core_config[] = {
@@ -150,7 +153,19 @@
 
 target_phys_addr_t cpu_get_phys_page_debug(CPUState *env, target_ulong addr)
 {
-    return addr;
+    uint32_t paddr;
+    uint32_t page_size;
+    unsigned access;
+
+    if (xtensa_get_physical_addr(env, addr, 0, 0,
+                &paddr, &page_size, &access) == 0) {
+        return paddr;
+    }
+    if (xtensa_get_physical_addr(env, addr, 2, 0,
+                &paddr, &page_size, &access) == 0) {
+        return paddr;
+    }
+    return ~0;
 }
 
 static uint32_t relocated_vector(CPUState *env, uint32_t vector)
@@ -255,3 +270,326 @@
     }
     check_interrupts(env);
 }
+
+static void reset_tlb_mmu_all_ways(CPUState *env,
+        const xtensa_tlb *tlb, xtensa_tlb_entry entry[][MAX_TLB_WAY_SIZE])
+{
+    unsigned wi, ei;
+
+    for (wi = 0; wi < tlb->nways; ++wi) {
+        for (ei = 0; ei < tlb->way_size[wi]; ++ei) {
+            entry[wi][ei].asid = 0;
+            entry[wi][ei].variable = true;
+        }
+    }
+}
+
+static void reset_tlb_mmu_ways56(CPUState *env,
+        const xtensa_tlb *tlb, xtensa_tlb_entry entry[][MAX_TLB_WAY_SIZE])
+{
+    if (!tlb->varway56) {
+        static const xtensa_tlb_entry way5[] = {
+            {
+                .vaddr = 0xd0000000,
+                .paddr = 0,
+                .asid = 1,
+                .attr = 7,
+                .variable = false,
+            }, {
+                .vaddr = 0xd8000000,
+                .paddr = 0,
+                .asid = 1,
+                .attr = 3,
+                .variable = false,
+            }
+        };
+        static const xtensa_tlb_entry way6[] = {
+            {
+                .vaddr = 0xe0000000,
+                .paddr = 0xf0000000,
+                .asid = 1,
+                .attr = 7,
+                .variable = false,
+            }, {
+                .vaddr = 0xf0000000,
+                .paddr = 0xf0000000,
+                .asid = 1,
+                .attr = 3,
+                .variable = false,
+            }
+        };
+        memcpy(entry[5], way5, sizeof(way5));
+        memcpy(entry[6], way6, sizeof(way6));
+    } else {
+        uint32_t ei;
+        for (ei = 0; ei < 8; ++ei) {
+            entry[6][ei].vaddr = ei << 29;
+            entry[6][ei].paddr = ei << 29;
+            entry[6][ei].asid = 1;
+            entry[6][ei].attr = 2;
+        }
+    }
+}
+
+static void reset_tlb_region_way0(CPUState *env,
+        xtensa_tlb_entry entry[][MAX_TLB_WAY_SIZE])
+{
+    unsigned ei;
+
+    for (ei = 0; ei < 8; ++ei) {
+        entry[0][ei].vaddr = ei << 29;
+        entry[0][ei].paddr = ei << 29;
+        entry[0][ei].asid = 1;
+        entry[0][ei].attr = 2;
+        entry[0][ei].variable = true;
+    }
+}
+
+static void reset_mmu(CPUState *env)
+{
+    if (xtensa_option_enabled(env->config, XTENSA_OPTION_MMU)) {
+        env->sregs[RASID] = 0x04030201;
+        env->sregs[ITLBCFG] = 0;
+        env->sregs[DTLBCFG] = 0;
+        env->autorefill_idx = 0;
+        reset_tlb_mmu_all_ways(env, &env->config->itlb, env->itlb);
+        reset_tlb_mmu_all_ways(env, &env->config->dtlb, env->dtlb);
+        reset_tlb_mmu_ways56(env, &env->config->itlb, env->itlb);
+        reset_tlb_mmu_ways56(env, &env->config->dtlb, env->dtlb);
+    } else {
+        reset_tlb_region_way0(env, env->itlb);
+        reset_tlb_region_way0(env, env->dtlb);
+    }
+}
+
+static unsigned get_ring(const CPUState *env, uint8_t asid)
+{
+    unsigned i;
+    for (i = 0; i < 4; ++i) {
+        if (((env->sregs[RASID] >> i * 8) & 0xff) == asid) {
+            return i;
+        }
+    }
+    return 0xff;
+}
+
+/*!
+ * Lookup xtensa TLB for the given virtual address.
+ * See ISA, 4.6.2.2
+ *
+ * \param pwi: [out] way index
+ * \param pei: [out] entry index
+ * \param pring: [out] access ring
+ * \return 0 if ok, exception cause code otherwise
+ */
+int xtensa_tlb_lookup(const CPUState *env, uint32_t addr, bool dtlb,
+        uint32_t *pwi, uint32_t *pei, uint8_t *pring)
+{
+    const xtensa_tlb *tlb = dtlb ?
+        &env->config->dtlb : &env->config->itlb;
+    const xtensa_tlb_entry (*entry)[MAX_TLB_WAY_SIZE] = dtlb ?
+        env->dtlb : env->itlb;
+
+    int nhits = 0;
+    unsigned wi;
+
+    for (wi = 0; wi < tlb->nways; ++wi) {
+        uint32_t vpn;
+        uint32_t ei;
+        split_tlb_entry_spec_way(env, addr, dtlb, &vpn, wi, &ei);
+        if (entry[wi][ei].vaddr == vpn && entry[wi][ei].asid) {
+            unsigned ring = get_ring(env, entry[wi][ei].asid);
+            if (ring < 4) {
+                if (++nhits > 1) {
+                    return dtlb ?
+                        LOAD_STORE_TLB_MULTI_HIT_CAUSE :
+                        INST_TLB_MULTI_HIT_CAUSE;
+                }
+                *pwi = wi;
+                *pei = ei;
+                *pring = ring;
+            }
+        }
+    }
+    return nhits ? 0 :
+        (dtlb ? LOAD_STORE_TLB_MISS_CAUSE : INST_TLB_MISS_CAUSE);
+}
+
+/*!
+ * Convert MMU ATTR to PAGE_{READ,WRITE,EXEC} mask.
+ * See ISA, 4.6.5.10
+ */
+static unsigned mmu_attr_to_access(uint32_t attr)
+{
+    unsigned access = 0;
+    if (attr < 12) {
+        access |= PAGE_READ;
+        if (attr & 0x1) {
+            access |= PAGE_EXEC;
+        }
+        if (attr & 0x2) {
+            access |= PAGE_WRITE;
+        }
+    } else if (attr == 13) {
+        access |= PAGE_READ | PAGE_WRITE;
+    }
+    return access;
+}
+
+/*!
+ * Convert region protection ATTR to PAGE_{READ,WRITE,EXEC} mask.
+ * See ISA, 4.6.3.3
+ */
+static unsigned region_attr_to_access(uint32_t attr)
+{
+    unsigned access = 0;
+    if ((attr < 6 && attr != 3) || attr == 14) {
+        access |= PAGE_READ | PAGE_WRITE;
+    }
+    if (attr > 0 && attr < 6) {
+        access |= PAGE_EXEC;
+    }
+    return access;
+}
+
+static bool is_access_granted(unsigned access, int is_write)
+{
+    switch (is_write) {
+    case 0:
+        return access & PAGE_READ;
+
+    case 1:
+        return access & PAGE_WRITE;
+
+    case 2:
+        return access & PAGE_EXEC;
+
+    default:
+        return 0;
+    }
+}
+
+static int autorefill_mmu(CPUState *env, uint32_t vaddr, bool dtlb,
+        uint32_t *wi, uint32_t *ei, uint8_t *ring);
+
+static int get_physical_addr_mmu(CPUState *env,
+        uint32_t vaddr, int is_write, int mmu_idx,
+        uint32_t *paddr, uint32_t *page_size, unsigned *access)
+{
+    bool dtlb = is_write != 2;
+    uint32_t wi;
+    uint32_t ei;
+    uint8_t ring;
+    int ret = xtensa_tlb_lookup(env, vaddr, dtlb, &wi, &ei, &ring);
+
+    if ((ret == INST_TLB_MISS_CAUSE || ret == LOAD_STORE_TLB_MISS_CAUSE) &&
+            (mmu_idx != 0 || ((vaddr ^ env->sregs[PTEVADDR]) & 0xffc00000)) &&
+            autorefill_mmu(env, vaddr, dtlb, &wi, &ei, &ring) == 0) {
+        ret = 0;
+    }
+    if (ret != 0) {
+        return ret;
+    }
+
+    const xtensa_tlb_entry *entry =
+        xtensa_tlb_get_entry(env, dtlb, wi, ei);
+
+    if (ring < mmu_idx) {
+        return dtlb ?
+            LOAD_STORE_PRIVILEGE_CAUSE :
+            INST_FETCH_PRIVILEGE_CAUSE;
+    }
+
+    *access = mmu_attr_to_access(entry->attr);
+    if (!is_access_granted(*access, is_write)) {
+        return dtlb ?
+            (is_write ?
+             STORE_PROHIBITED_CAUSE :
+             LOAD_PROHIBITED_CAUSE) :
+            INST_FETCH_PROHIBITED_CAUSE;
+    }
+
+    *paddr = entry->paddr | (vaddr & ~xtensa_tlb_get_addr_mask(env, dtlb, wi));
+    *page_size = ~xtensa_tlb_get_addr_mask(env, dtlb, wi) + 1;
+
+    return 0;
+}
+
+static int autorefill_mmu(CPUState *env, uint32_t vaddr, bool dtlb,
+        uint32_t *wi, uint32_t *ei, uint8_t *ring)
+{
+    uint32_t paddr;
+    uint32_t page_size;
+    unsigned access;
+    uint32_t pt_vaddr =
+        (env->sregs[PTEVADDR] | (vaddr >> 10)) & 0xfffffffc;
+    int ret = get_physical_addr_mmu(env, pt_vaddr, 0, 0,
+            &paddr, &page_size, &access);
+
+    qemu_log("%s: trying autorefill(%08x) -> %08x\n", __func__,
+            vaddr, ret ? ~0 : paddr);
+
+    if (ret == 0) {
+        uint32_t vpn;
+        uint32_t pte = ldl_phys(paddr);
+
+        *ring = (pte >> 4) & 0x3;
+        *wi = (++env->autorefill_idx) & 0x3;
+        split_tlb_entry_spec_way(env, vaddr, dtlb, &vpn, *wi, ei);
+        xtensa_tlb_set_entry(env, dtlb, *wi, *ei, vpn, pte);
+        qemu_log("%s: autorefill(%08x): %08x -> %08x\n",
+                __func__, vaddr, vpn, pte);
+    }
+    return ret;
+}
+
+static int get_physical_addr_region(CPUState *env,
+        uint32_t vaddr, int is_write, int mmu_idx,
+        uint32_t *paddr, uint32_t *page_size, unsigned *access)
+{
+    bool dtlb = is_write != 2;
+    uint32_t wi = 0;
+    uint32_t ei = (vaddr >> 29) & 0x7;
+    const xtensa_tlb_entry *entry =
+        xtensa_tlb_get_entry(env, dtlb, wi, ei);
+
+    *access = region_attr_to_access(entry->attr);
+    if (!is_access_granted(*access, is_write)) {
+        return dtlb ?
+            (is_write ?
+             STORE_PROHIBITED_CAUSE :
+             LOAD_PROHIBITED_CAUSE) :
+            INST_FETCH_PROHIBITED_CAUSE;
+    }
+
+    *paddr = entry->paddr | (vaddr & ~REGION_PAGE_MASK);
+    *page_size = ~REGION_PAGE_MASK + 1;
+
+    return 0;
+}
+
+/*!
+ * Convert virtual address to physical addr.
+ * MMU may issue pagewalk and change xtensa autorefill TLB way entry.
+ *
+ * \return 0 if ok, exception cause code otherwise
+ */
+int xtensa_get_physical_addr(CPUState *env,
+        uint32_t vaddr, int is_write, int mmu_idx,
+        uint32_t *paddr, uint32_t *page_size, unsigned *access)
+{
+    if (xtensa_option_enabled(env->config, XTENSA_OPTION_MMU)) {
+        return get_physical_addr_mmu(env, vaddr, is_write, mmu_idx,
+                paddr, page_size, access);
+    } else if (xtensa_option_bits_enabled(env->config,
+                XTENSA_OPTION_BIT(XTENSA_OPTION_REGION_PROTECTION) |
+                XTENSA_OPTION_BIT(XTENSA_OPTION_REGION_TRANSLATION))) {
+        return get_physical_addr_region(env, vaddr, is_write, mmu_idx,
+                paddr, page_size, access);
+    } else {
+        *paddr = vaddr;
+        *page_size = TARGET_PAGE_SIZE;
+        *access = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
+        return 0;
+    }
+}