Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-20190903' into staging

Allow page table bit to swap endianness.
Reorganize watchpoints out of i/o path.
Return host address from probe_write / probe_access.

# gpg: Signature made Tue 03 Sep 2019 16:47:50 BST
# gpg:                using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg:                issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [full]
# Primary key fingerprint: 7A48 1E78 868B 4DB6 A85A  05C0 64DF 38E8 AF7E 215F

* remotes/rth/tags/pull-tcg-20190903: (36 commits)
  tcg: Factor out probe_write() logic into probe_access()
  tcg: Make probe_write() return a pointer to the host page
  s390x/tcg: Pass a size to probe_write() in do_csst()
  hppa/tcg: Call probe_write() also for CONFIG_USER_ONLY
  mips/tcg: Call probe_write() for CONFIG_USER_ONLY as well
  tcg: Enforce single page access in probe_write()
  tcg: Factor out CONFIG_USER_ONLY probe_write() from s390x code
  s390x/tcg: Fix length calculation in probe_write_access()
  s390x/tcg: Use guest_addr_valid() instead of h2g_valid() in probe_write_access()
  tcg: Check for watchpoints in probe_write()
  cputlb: Handle watchpoints via TLB_WATCHPOINT
  cputlb: Remove double-alignment in store_helper
  cputlb: Fix size operand for tlb_fill on unaligned store
  exec: Factor out cpu_watchpoint_address_matches
  cputlb: Fold TLB_RECHECK into TLB_INVALID_MASK
  exec: Factor out core logic of check_watchpoint()
  exec: Move user-only watchpoint stubs inline
  target/sparc: sun4u Invert Endian TTE bit
  target/sparc: Add TLB entry with attributes
  cputlb: Byte swap memory transaction attribute
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

2 files changed, 262 insertions, 182 deletions

diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index bb9897b..abae796 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -710,6 +710,7 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
     hwaddr iotlb, xlat, sz, paddr_page;
     target_ulong vaddr_page;
     int asidx = cpu_asidx_from_attrs(cpu, attrs);
+    int wp_flags;
 
     assert_cpu_is_self(cpu);
 
@@ -732,11 +733,12 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
 
     address = vaddr_page;
     if (size < TARGET_PAGE_SIZE) {
-        /*
-         * Slow-path the TLB entries; we will repeat the MMU check and TLB
-         * fill on every access.
-         */
-        address |= TLB_RECHECK;
+        /* Repeat the MMU check and TLB fill on every access.  */
+        address |= TLB_INVALID_MASK;
+    }
+    if (attrs.byte_swap) {
+        /* Force the access through the I/O slow path.  */
+        address |= TLB_MMIO;
     }
     if (!memory_region_is_ram(section->mr) &&
         !memory_region_is_romd(section->mr)) {
@@ -751,6 +753,8 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
     code_address = address;
     iotlb = memory_region_section_get_iotlb(cpu, section, vaddr_page,
                                             paddr_page, xlat, prot, &address);
+    wp_flags = cpu_watchpoint_address_matches(cpu, vaddr_page,
+                                              TARGET_PAGE_SIZE);
 
     index = tlb_index(env, mmu_idx, vaddr_page);
     te = tlb_entry(env, mmu_idx, vaddr_page);
@@ -804,6 +808,9 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
     tn.addend = addend - vaddr_page;
     if (prot & PAGE_READ) {
         tn.addr_read = address;
+        if (wp_flags & BP_MEM_READ) {
+            tn.addr_read |= TLB_WATCHPOINT;
+        }
     } else {
         tn.addr_read = -1;
     }
@@ -830,6 +837,9 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
         if (prot & PAGE_WRITE_INV) {
             tn.addr_write |= TLB_INVALID_MASK;
         }
+        if (wp_flags & BP_MEM_WRITE) {
+            tn.addr_write |= TLB_WATCHPOINT;
+        }
     }
 
     copy_tlb_helper_locked(te, &tn);
@@ -881,7 +891,7 @@ static void tlb_fill(CPUState *cpu, target_ulong addr, int size,
 
 static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
                          int mmu_idx, target_ulong addr, uintptr_t retaddr,
-                         MMUAccessType access_type, int size)
+                         MMUAccessType access_type, MemOp op)
 {
     CPUState *cpu = env_cpu(env);
     hwaddr mr_offset;
@@ -891,6 +901,10 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
     bool locked = false;
     MemTxResult r;
 
+    if (iotlbentry->attrs.byte_swap) {
+        op ^= MO_BSWAP;
+    }
+
     section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs);
     mr = section->mr;
     mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr;
@@ -906,14 +920,13 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
         qemu_mutex_lock_iothread();
         locked = true;
     }
-    r = memory_region_dispatch_read(mr, mr_offset,
-                                    &val, size, iotlbentry->attrs);
+    r = memory_region_dispatch_read(mr, mr_offset, &val, op, iotlbentry->attrs);
     if (r != MEMTX_OK) {
         hwaddr physaddr = mr_offset +
             section->offset_within_address_space -
             section->offset_within_region;
 
-        cpu_transaction_failed(cpu, physaddr, addr, size, access_type,
+        cpu_transaction_failed(cpu, physaddr, addr, memop_size(op), access_type,
                                mmu_idx, iotlbentry->attrs, r, retaddr);
     }
     if (locked) {
@@ -925,7 +938,7 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
 
 static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
                       int mmu_idx, uint64_t val, target_ulong addr,
-                      uintptr_t retaddr, int size)
+                      uintptr_t retaddr, MemOp op)
 {
     CPUState *cpu = env_cpu(env);
     hwaddr mr_offset;
@@ -934,6 +947,10 @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
     bool locked = false;
     MemTxResult r;
 
+    if (iotlbentry->attrs.byte_swap) {
+        op ^= MO_BSWAP;
+    }
+
     section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs);
     mr = section->mr;
     mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr;
@@ -947,15 +964,15 @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
         qemu_mutex_lock_iothread();
         locked = true;
     }
-    r = memory_region_dispatch_write(mr, mr_offset,
-                                     val, size, iotlbentry->attrs);
+    r = memory_region_dispatch_write(mr, mr_offset, val, op, iotlbentry->attrs);
     if (r != MEMTX_OK) {
         hwaddr physaddr = mr_offset +
             section->offset_within_address_space -
             section->offset_within_region;
 
-        cpu_transaction_failed(cpu, physaddr, addr, size, MMU_DATA_STORE,
-                               mmu_idx, iotlbentry->attrs, r, retaddr);
+        cpu_transaction_failed(cpu, physaddr, addr, memop_size(op),
+                               MMU_DATA_STORE, mmu_idx, iotlbentry->attrs, r,
+                               retaddr);
     }
     if (locked) {
         qemu_mutex_unlock_iothread();
@@ -1015,10 +1032,15 @@ static bool victim_tlb_hit(CPUArchState *env, size_t mmu_idx, size_t index,
   victim_tlb_hit(env, mmu_idx, index, offsetof(CPUTLBEntry, TY), \
                  (ADDR) & TARGET_PAGE_MASK)
 
-/* NOTE: this function can trigger an exception */
-/* NOTE2: the returned address is not exactly the physical address: it
- * is actually a ram_addr_t (in system mode; the user mode emulation
- * version of this function returns a guest virtual address).
+/*
+ * Return a ram_addr_t for the virtual address for execution.
+ *
+ * Return -1 if we can't translate and execute from an entire page
+ * of RAM.  This will force us to execute by loading and translating
+ * one insn at a time, without caching.
+ *
+ * NOTE: This function will trigger an exception if the page is
+ * not executable.
  */
 tb_page_addr_t get_page_addr_code(CPUArchState *env, target_ulong addr)
 {
@@ -1032,19 +1054,20 @@ tb_page_addr_t get_page_addr_code(CPUArchState *env, target_ulong addr)
             tlb_fill(env_cpu(env), addr, 0, MMU_INST_FETCH, mmu_idx, 0);
             index = tlb_index(env, mmu_idx, addr);
             entry = tlb_entry(env, mmu_idx, addr);
+
+            if (unlikely(entry->addr_code & TLB_INVALID_MASK)) {
+                /*
+                 * The MMU protection covers a smaller range than a target
+                 * page, so we must redo the MMU check for every insn.
+                 */
+                return -1;
+            }
         }
         assert(tlb_hit(entry->addr_code, addr));
     }
 
-    if (unlikely(entry->addr_code & (TLB_RECHECK | TLB_MMIO))) {
-        /*
-         * Return -1 if we can't translate and execute from an entire
-         * page of RAM here, which will cause us to execute by loading
-         * and translating one insn at a time, without caching:
-         *  - TLB_RECHECK: means the MMU protection covers a smaller range
-         *    than a target page, so we must redo the MMU check every insn
-         *  - TLB_MMIO: region is not backed by RAM
-         */
+    if (unlikely(entry->addr_code & TLB_MMIO)) {
+        /* The region is not backed by RAM.  */
         return -1;
     }
 
@@ -1052,25 +1075,70 @@ tb_page_addr_t get_page_addr_code(CPUArchState *env, target_ulong addr)
     return qemu_ram_addr_from_host_nofail(p);
 }
 
-/* Probe for whether the specified guest write access is permitted.
- * If it is not permitted then an exception will be taken in the same
- * way as if this were a real write access (and we will not return).
- * Otherwise the function will return, and there will be a valid
- * entry in the TLB for this access.
+/*
+ * Probe for whether the specified guest access is permitted. If it is not
+ * permitted then an exception will be taken in the same way as if this
+ * were a real access (and we will not return).
+ * If the size is 0 or the page requires I/O access, returns NULL; otherwise,
+ * returns the address of the host page similar to tlb_vaddr_to_host().
  */
-void probe_write(CPUArchState *env, target_ulong addr, int size, int mmu_idx,
-                 uintptr_t retaddr)
+void *probe_access(CPUArchState *env, target_ulong addr, int size,
+                   MMUAccessType access_type, int mmu_idx, uintptr_t retaddr)
 {
     uintptr_t index = tlb_index(env, mmu_idx, addr);
     CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
+    target_ulong tlb_addr;
+    size_t elt_ofs;
+    int wp_access;
 
-    if (!tlb_hit(tlb_addr_write(entry), addr)) {
-        /* TLB entry is for a different page */
-        if (!VICTIM_TLB_HIT(addr_write, addr)) {
-            tlb_fill(env_cpu(env), addr, size, MMU_DATA_STORE,
-                     mmu_idx, retaddr);
+    g_assert(-(addr | TARGET_PAGE_MASK) >= size);
+
+    switch (access_type) {
+    case MMU_DATA_LOAD:
+        elt_ofs = offsetof(CPUTLBEntry, addr_read);
+        wp_access = BP_MEM_READ;
+        break;
+    case MMU_DATA_STORE:
+        elt_ofs = offsetof(CPUTLBEntry, addr_write);
+        wp_access = BP_MEM_WRITE;
+        break;
+    case MMU_INST_FETCH:
+        elt_ofs = offsetof(CPUTLBEntry, addr_code);
+        wp_access = BP_MEM_READ;
+        break;
+    default:
+        g_assert_not_reached();
+    }
+    tlb_addr = tlb_read_ofs(entry, elt_ofs);
+
+    if (unlikely(!tlb_hit(tlb_addr, addr))) {
+        if (!victim_tlb_hit(env, mmu_idx, index, elt_ofs,
+                            addr & TARGET_PAGE_MASK)) {
+            tlb_fill(env_cpu(env), addr, size, access_type, mmu_idx, retaddr);
+            /* TLB resize via tlb_fill may have moved the entry. */
+            index = tlb_index(env, mmu_idx, addr);
+            entry = tlb_entry(env, mmu_idx, addr);
         }
+        tlb_addr = tlb_read_ofs(entry, elt_ofs);
+    }
+
+    if (!size) {
+        return NULL;
     }
+
+    /* Handle watchpoints.  */
+    if (tlb_addr & TLB_WATCHPOINT) {
+        cpu_check_watchpoint(env_cpu(env), addr, size,
+                             env_tlb(env)->d[mmu_idx].iotlb[index].attrs,
+                             wp_access, retaddr);
+    }
+
+    if (tlb_addr & (TLB_NOTDIRTY | TLB_MMIO)) {
+        /* I/O access */
+        return NULL;
+    }
+
+    return (void *)((uintptr_t)addr + entry->addend);
 }
 
 void *tlb_vaddr_to_host(CPUArchState *env, abi_ptr addr,
@@ -1133,7 +1201,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
     uintptr_t index = tlb_index(env, mmu_idx, addr);
     CPUTLBEntry *tlbe = tlb_entry(env, mmu_idx, addr);
     target_ulong tlb_addr = tlb_addr_write(tlbe);
-    TCGMemOp mop = get_memop(oi);
+    MemOp mop = get_memop(oi);
     int a_bits = get_alignment_bits(mop);
     int s_bits = mop & MO_SIZE;
     void *hostaddr;
@@ -1169,7 +1237,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
     }
 
     /* Notice an IO access or a needs-MMU-lookup access */
-    if (unlikely(tlb_addr & (TLB_MMIO | TLB_RECHECK))) {
+    if (unlikely(tlb_addr & TLB_MMIO)) {
         /* There's really nothing that can be done to
            support this apart from stop-the-world.  */
         goto stop_the_world;
@@ -1201,37 +1269,6 @@ static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
     cpu_loop_exit_atomic(env_cpu(env), retaddr);
 }
 
-#ifdef TARGET_WORDS_BIGENDIAN
-#define NEED_BE_BSWAP 0
-#define NEED_LE_BSWAP 1
-#else
-#define NEED_BE_BSWAP 1
-#define NEED_LE_BSWAP 0
-#endif
-
-/*
- * Byte Swap Helper
- *
- * This should all dead code away depending on the build host and
- * access type.
- */
-
-static inline uint64_t handle_bswap(uint64_t val, int size, bool big_endian)
-{
-    if ((big_endian && NEED_BE_BSWAP) || (!big_endian && NEED_LE_BSWAP)) {
-        switch (size) {
-        case 1: return val;
-        case 2: return bswap16(val);
-        case 4: return bswap32(val);
-        case 8: return bswap64(val);
-        default:
-            g_assert_not_reached();
-        }
-    } else {
-        return val;
-    }
-}
-
 /*
  * Load Helpers
  *
@@ -1246,7 +1283,7 @@ typedef uint64_t FullLoadHelper(CPUArchState *env, target_ulong addr,
 
 static inline uint64_t __attribute__((always_inline))
 load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi,
-            uintptr_t retaddr, size_t size, bool big_endian, bool code_read,
+            uintptr_t retaddr, MemOp op, bool code_read,
             FullLoadHelper *full_load)
 {
     uintptr_t mmu_idx = get_mmuidx(oi);
@@ -1260,6 +1297,7 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi,
     unsigned a_bits = get_alignment_bits(get_memop(oi));
     void *haddr;
     uint64_t res;
+    size_t size = memop_size(op);
 
     /* Handle CPU specific unaligned behaviour */
     if (addr & ((1 << a_bits) - 1)) {
@@ -1277,37 +1315,36 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi,
             entry = tlb_entry(env, mmu_idx, addr);
         }
         tlb_addr = code_read ? entry->addr_code : entry->addr_read;
+        tlb_addr &= ~TLB_INVALID_MASK;
     }
 
-    /* Handle an IO access.  */
+    /* Handle anything that isn't just a straight memory access.  */
     if (unlikely(tlb_addr & ~TARGET_PAGE_MASK)) {
+        CPUIOTLBEntry *iotlbentry;
+
+        /* For anything that is unaligned, recurse through full_load.  */
         if ((addr & (size - 1)) != 0) {
             goto do_unaligned_access;
         }
 
-        if (tlb_addr & TLB_RECHECK) {
-            /*
-             * This is a TLB_RECHECK access, where the MMU protection
-             * covers a smaller range than a target page, and we must
-             * repeat the MMU check here. This tlb_fill() call might
-             * longjump out if this access should cause a guest exception.
-             */
-            tlb_fill(env_cpu(env), addr, size,
-                     access_type, mmu_idx, retaddr);
-            index = tlb_index(env, mmu_idx, addr);
-            entry = tlb_entry(env, mmu_idx, addr);
+        iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index];
+
+        /* Handle watchpoints.  */
+        if (unlikely(tlb_addr & TLB_WATCHPOINT)) {
+            /* On watchpoint hit, this will longjmp out.  */
+            cpu_check_watchpoint(env_cpu(env), addr, size,
+                                 iotlbentry->attrs, BP_MEM_READ, retaddr);
 
-            tlb_addr = code_read ? entry->addr_code : entry->addr_read;
-            tlb_addr &= ~TLB_RECHECK;
-            if (!(tlb_addr & ~TARGET_PAGE_MASK)) {
-                /* RAM access */
+            /* The backing page may or may not require I/O.  */
+            tlb_addr &= ~TLB_WATCHPOINT;
+            if ((tlb_addr & ~TARGET_PAGE_MASK) == 0) {
                 goto do_aligned_access;
             }
         }
 
-        res = io_readx(env, &env_tlb(env)->d[mmu_idx].iotlb[index],
-                       mmu_idx, addr, retaddr, access_type, size);
-        return handle_bswap(res, size, big_endian);
+        /* Handle I/O access.  */
+        return io_readx(env, iotlbentry, mmu_idx, addr,
+                        retaddr, access_type, op);
     }
 
     /* Handle slow unaligned access (it spans two pages or IO).  */
@@ -1324,7 +1361,7 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi,
         r2 = full_load(env, addr2, oi, retaddr);
         shift = (addr & (size - 1)) * 8;
 
-        if (big_endian) {
+        if (memop_big_endian(op)) {
             /* Big-endian combine.  */
             res = (r1 << shift) | (r2 >> ((size * 8) - shift));
         } else {
@@ -1336,30 +1373,27 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi,
 
  do_aligned_access:
     haddr = (void *)((uintptr_t)addr + entry->addend);
-    switch (size) {
-    case 1:
+    switch (op) {
+    case MO_UB:
         res = ldub_p(haddr);
         break;
-    case 2:
-        if (big_endian) {
-            res = lduw_be_p(haddr);
-        } else {
-            res = lduw_le_p(haddr);
-        }
+    case MO_BEUW:
+        res = lduw_be_p(haddr);
         break;
-    case 4:
-        if (big_endian) {
-            res = (uint32_t)ldl_be_p(haddr);
-        } else {
-            res = (uint32_t)ldl_le_p(haddr);
-        }
+    case MO_LEUW:
+        res = lduw_le_p(haddr);
         break;
-    case 8:
-        if (big_endian) {
-            res = ldq_be_p(haddr);
-        } else {
-            res = ldq_le_p(haddr);
-        }
+    case MO_BEUL:
+        res = (uint32_t)ldl_be_p(haddr);
+        break;
+    case MO_LEUL:
+        res = (uint32_t)ldl_le_p(haddr);
+        break;
+    case MO_BEQ:
+        res = ldq_be_p(haddr);
+        break;
+    case MO_LEQ:
+        res = ldq_le_p(haddr);
         break;
     default:
         g_assert_not_reached();
@@ -1381,8 +1415,7 @@ load_helper(CPUArchState *env, target_ulong addr, TCGMemOpIdx oi,
 static uint64_t full_ldub_mmu(CPUArchState *env, target_ulong addr,
                               TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 1, false, false,
-                       full_ldub_mmu);
+    return load_helper(env, addr, oi, retaddr, MO_UB, false, full_ldub_mmu);
 }
 
 tcg_target_ulong helper_ret_ldub_mmu(CPUArchState *env, target_ulong addr,
@@ -1394,7 +1427,7 @@ tcg_target_ulong helper_ret_ldub_mmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_le_lduw_mmu(CPUArchState *env, target_ulong addr,
                                  TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 2, false, false,
+    return load_helper(env, addr, oi, retaddr, MO_LEUW, false,
                        full_le_lduw_mmu);
 }
 
@@ -1407,7 +1440,7 @@ tcg_target_ulong helper_le_lduw_mmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_be_lduw_mmu(CPUArchState *env, target_ulong addr,
                                  TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 2, true, false,
+    return load_helper(env, addr, oi, retaddr, MO_BEUW, false,
                        full_be_lduw_mmu);
 }
 
@@ -1420,7 +1453,7 @@ tcg_target_ulong helper_be_lduw_mmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_le_ldul_mmu(CPUArchState *env, target_ulong addr,
                                  TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 4, false, false,
+    return load_helper(env, addr, oi, retaddr, MO_LEUL, false,
                        full_le_ldul_mmu);
 }
 
@@ -1433,7 +1466,7 @@ tcg_target_ulong helper_le_ldul_mmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_be_ldul_mmu(CPUArchState *env, target_ulong addr,
                                  TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 4, true, false,
+    return load_helper(env, addr, oi, retaddr, MO_BEUL, false,
                        full_be_ldul_mmu);
 }
 
@@ -1446,14 +1479,14 @@ tcg_target_ulong helper_be_ldul_mmu(CPUArchState *env, target_ulong addr,
 uint64_t helper_le_ldq_mmu(CPUArchState *env, target_ulong addr,
                            TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 8, false, false,
+    return load_helper(env, addr, oi, retaddr, MO_LEQ, false,
                        helper_le_ldq_mmu);
 }
 
 uint64_t helper_be_ldq_mmu(CPUArchState *env, target_ulong addr,
                            TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 8, true, false,
+    return load_helper(env, addr, oi, retaddr, MO_BEQ, false,
                        helper_be_ldq_mmu);
 }
 
@@ -1499,7 +1532,7 @@ tcg_target_ulong helper_be_ldsl_mmu(CPUArchState *env, target_ulong addr,
 
 static inline void __attribute__((always_inline))
 store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
-             TCGMemOpIdx oi, uintptr_t retaddr, size_t size, bool big_endian)
+             TCGMemOpIdx oi, uintptr_t retaddr, MemOp op)
 {
     uintptr_t mmu_idx = get_mmuidx(oi);
     uintptr_t index = tlb_index(env, mmu_idx, addr);
@@ -1508,6 +1541,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
     const size_t tlb_off = offsetof(CPUTLBEntry, addr_write);
     unsigned a_bits = get_alignment_bits(get_memop(oi));
     void *haddr;
+    size_t size = memop_size(op);
 
     /* Handle CPU specific unaligned behaviour */
     if (addr & ((1 << a_bits) - 1)) {
@@ -1527,35 +1561,32 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
         tlb_addr = tlb_addr_write(entry) & ~TLB_INVALID_MASK;
     }
 
-    /* Handle an IO access.  */
+    /* Handle anything that isn't just a straight memory access.  */
     if (unlikely(tlb_addr & ~TARGET_PAGE_MASK)) {
+        CPUIOTLBEntry *iotlbentry;
+
+        /* For anything that is unaligned, recurse through byte stores.  */
         if ((addr & (size - 1)) != 0) {
             goto do_unaligned_access;
         }
 
-        if (tlb_addr & TLB_RECHECK) {
-            /*
-             * This is a TLB_RECHECK access, where the MMU protection
-             * covers a smaller range than a target page, and we must
-             * repeat the MMU check here. This tlb_fill() call might
-             * longjump out if this access should cause a guest exception.
-             */
-            tlb_fill(env_cpu(env), addr, size, MMU_DATA_STORE,
-                     mmu_idx, retaddr);
-            index = tlb_index(env, mmu_idx, addr);
-            entry = tlb_entry(env, mmu_idx, addr);
+        iotlbentry = &env_tlb(env)->d[mmu_idx].iotlb[index];
+
+        /* Handle watchpoints.  */
+        if (unlikely(tlb_addr & TLB_WATCHPOINT)) {
+            /* On watchpoint hit, this will longjmp out.  */
+            cpu_check_watchpoint(env_cpu(env), addr, size,
+                                 iotlbentry->attrs, BP_MEM_WRITE, retaddr);
 
-            tlb_addr = tlb_addr_write(entry);
-            tlb_addr &= ~TLB_RECHECK;
-            if (!(tlb_addr & ~TARGET_PAGE_MASK)) {
-                /* RAM access */
+            /* The backing page may or may not require I/O.  */
+            tlb_addr &= ~TLB_WATCHPOINT;
+            if ((tlb_addr & ~TARGET_PAGE_MASK) == 0) {
                 goto do_aligned_access;
             }
         }
 
-        io_writex(env, &env_tlb(env)->d[mmu_idx].iotlb[index], mmu_idx,
-                  handle_bswap(val, size, big_endian),
-                  addr, retaddr, size);
+        /* Handle I/O access.  */
+        io_writex(env, iotlbentry, mmu_idx, val, addr, retaddr, op);
         return;
     }
 
@@ -1567,6 +1598,8 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
         uintptr_t index2;
         CPUTLBEntry *entry2;
         target_ulong page2, tlb_addr2;
+        size_t size2;
+
     do_unaligned_access:
         /*
          * Ensure the second page is in the TLB.  Note that the first page
@@ -1574,14 +1607,33 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
          * cannot evict the first.
          */
         page2 = (addr + size) & TARGET_PAGE_MASK;
+        size2 = (addr + size) & ~TARGET_PAGE_MASK;
         index2 = tlb_index(env, mmu_idx, page2);
         entry2 = tlb_entry(env, mmu_idx, page2);
         tlb_addr2 = tlb_addr_write(entry2);
-        if (!tlb_hit_page(tlb_addr2, page2)
-            && !victim_tlb_hit(env, mmu_idx, index2, tlb_off,
-                               page2 & TARGET_PAGE_MASK)) {
-            tlb_fill(env_cpu(env), page2, size, MMU_DATA_STORE,
-                     mmu_idx, retaddr);
+        if (!tlb_hit_page(tlb_addr2, page2)) {
+            if (!victim_tlb_hit(env, mmu_idx, index2, tlb_off, page2)) {
+                tlb_fill(env_cpu(env), page2, size2, MMU_DATA_STORE,
+                         mmu_idx, retaddr);
+                index2 = tlb_index(env, mmu_idx, page2);
+                entry2 = tlb_entry(env, mmu_idx, page2);
+            }
+            tlb_addr2 = tlb_addr_write(entry2);
+        }
+
+        /*
+         * Handle watchpoints.  Since this may trap, all checks
+         * must happen before any store.
+         */
+        if (unlikely(tlb_addr & TLB_WATCHPOINT)) {
+            cpu_check_watchpoint(env_cpu(env), addr, size - size2,
+                                 env_tlb(env)->d[mmu_idx].iotlb[index].attrs,
+                                 BP_MEM_WRITE, retaddr);
+        }
+        if (unlikely(tlb_addr2 & TLB_WATCHPOINT)) {
+            cpu_check_watchpoint(env_cpu(env), page2, size2,
+                                 env_tlb(env)->d[mmu_idx].iotlb[index2].attrs,
+                                 BP_MEM_WRITE, retaddr);
         }
 
         /*
@@ -1591,7 +1643,7 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
          */
         for (i = 0; i < size; ++i) {
             uint8_t val8;
-            if (big_endian) {
+            if (memop_big_endian(op)) {
                 /* Big-endian extract.  */
                 val8 = val >> (((size - 1) * 8) - (i * 8));
             } else {
@@ -1605,30 +1657,27 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
 
  do_aligned_access:
     haddr = (void *)((uintptr_t)addr + entry->addend);
-    switch (size) {
-    case 1:
+    switch (op) {
+    case MO_UB:
         stb_p(haddr, val);
         break;
-    case 2:
-        if (big_endian) {
-            stw_be_p(haddr, val);
-        } else {
-            stw_le_p(haddr, val);
-        }
+    case MO_BEUW:
+        stw_be_p(haddr, val);
         break;
-    case 4:
-        if (big_endian) {
-            stl_be_p(haddr, val);
-        } else {
-            stl_le_p(haddr, val);
-        }
+    case MO_LEUW:
+        stw_le_p(haddr, val);
         break;
-    case 8:
-        if (big_endian) {
-            stq_be_p(haddr, val);
-        } else {
-            stq_le_p(haddr, val);
-        }
+    case MO_BEUL:
+        stl_be_p(haddr, val);
+        break;
+    case MO_LEUL:
+        stl_le_p(haddr, val);
+        break;
+    case MO_BEQ:
+        stq_be_p(haddr, val);
+        break;
+    case MO_LEQ:
+        stq_le_p(haddr, val);
         break;
     default:
         g_assert_not_reached();
@@ -1639,43 +1688,43 @@ store_helper(CPUArchState *env, target_ulong addr, uint64_t val,
 void helper_ret_stb_mmu(CPUArchState *env, target_ulong addr, uint8_t val,
                         TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 1, false);
+    store_helper(env, addr, val, oi, retaddr, MO_UB);
 }
 
 void helper_le_stw_mmu(CPUArchState *env, target_ulong addr, uint16_t val,
                        TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 2, false);
+    store_helper(env, addr, val, oi, retaddr, MO_LEUW);
 }
 
 void helper_be_stw_mmu(CPUArchState *env, target_ulong addr, uint16_t val,
                        TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 2, true);
+    store_helper(env, addr, val, oi, retaddr, MO_BEUW);
 }
 
 void helper_le_stl_mmu(CPUArchState *env, target_ulong addr, uint32_t val,
                        TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 4, false);
+    store_helper(env, addr, val, oi, retaddr, MO_LEUL);
 }
 
 void helper_be_stl_mmu(CPUArchState *env, target_ulong addr, uint32_t val,
                        TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 4, true);
+    store_helper(env, addr, val, oi, retaddr, MO_BEUL);
 }
 
 void helper_le_stq_mmu(CPUArchState *env, target_ulong addr, uint64_t val,
                        TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 8, false);
+    store_helper(env, addr, val, oi, retaddr, MO_LEQ);
 }
 
 void helper_be_stq_mmu(CPUArchState *env, target_ulong addr, uint64_t val,
                        TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    store_helper(env, addr, val, oi, retaddr, 8, true);
+    store_helper(env, addr, val, oi, retaddr, MO_BEQ);
 }
 
 /* First set of helpers allows passing in of OI and RETADDR.  This makes
@@ -1740,8 +1789,7 @@ void helper_be_stq_mmu(CPUArchState *env, target_ulong addr, uint64_t val,
 static uint64_t full_ldub_cmmu(CPUArchState *env, target_ulong addr,
                                TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 1, false, true,
-                       full_ldub_cmmu);
+    return load_helper(env, addr, oi, retaddr, MO_8, true, full_ldub_cmmu);
 }
 
 uint8_t helper_ret_ldb_cmmu(CPUArchState *env, target_ulong addr,
@@ -1753,7 +1801,7 @@ uint8_t helper_ret_ldb_cmmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_le_lduw_cmmu(CPUArchState *env, target_ulong addr,
                                   TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 2, false, true,
+    return load_helper(env, addr, oi, retaddr, MO_LEUW, true,
                        full_le_lduw_cmmu);
 }
 
@@ -1766,7 +1814,7 @@ uint16_t helper_le_ldw_cmmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_be_lduw_cmmu(CPUArchState *env, target_ulong addr,
                                   TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 2, true, true,
+    return load_helper(env, addr, oi, retaddr, MO_BEUW, true,
                        full_be_lduw_cmmu);
 }
 
@@ -1779,7 +1827,7 @@ uint16_t helper_be_ldw_cmmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_le_ldul_cmmu(CPUArchState *env, target_ulong addr,
                                   TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 4, false, true,
+    return load_helper(env, addr, oi, retaddr, MO_LEUL, true,
                        full_le_ldul_cmmu);
 }
 
@@ -1792,7 +1840,7 @@ uint32_t helper_le_ldl_cmmu(CPUArchState *env, target_ulong addr,
 static uint64_t full_be_ldul_cmmu(CPUArchState *env, target_ulong addr,
                                   TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 4, true, true,
+    return load_helper(env, addr, oi, retaddr, MO_BEUL, true,
                        full_be_ldul_cmmu);
 }
 
@@ -1805,13 +1853,13 @@ uint32_t helper_be_ldl_cmmu(CPUArchState *env, target_ulong addr,
 uint64_t helper_le_ldq_cmmu(CPUArchState *env, target_ulong addr,
                             TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 8, false, true,
+    return load_helper(env, addr, oi, retaddr, MO_LEQ, true,
                        helper_le_ldq_cmmu);
 }
 
 uint64_t helper_be_ldq_cmmu(CPUArchState *env, target_ulong addr,
                             TCGMemOpIdx oi, uintptr_t retaddr)
 {
-    return load_helper(env, addr, oi, retaddr, 8, true, true,
+    return load_helper(env, addr, oi, retaddr, MO_BEQ, true,
                        helper_be_ldq_cmmu);
 }
diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c
index 897d157..71c4bf6 100644
--- a/accel/tcg/user-exec.c
+++ b/accel/tcg/user-exec.c
@@ -188,6 +188,38 @@ static inline int handle_cpu_signal(uintptr_t pc, siginfo_t *info,
     g_assert_not_reached();
 }
 
+void *probe_access(CPUArchState *env, target_ulong addr, int size,
+                   MMUAccessType access_type, int mmu_idx, uintptr_t retaddr)
+{
+    int flags;
+
+    g_assert(-(addr | TARGET_PAGE_MASK) >= size);
+
+    switch (access_type) {
+    case MMU_DATA_STORE:
+        flags = PAGE_WRITE;
+        break;
+    case MMU_DATA_LOAD:
+        flags = PAGE_READ;
+        break;
+    case MMU_INST_FETCH:
+        flags = PAGE_EXEC;
+        break;
+    default:
+        g_assert_not_reached();
+    }
+
+    if (!guest_addr_valid(addr) || page_check_range(addr, size, flags) < 0) {
+        CPUState *cpu = env_cpu(env);
+        CPUClass *cc = CPU_GET_CLASS(cpu);
+        cc->tlb_fill(cpu, addr, size, access_type, MMU_USER_IDX, false,
+                     retaddr);
+        g_assert_not_reached();
+    }
+
+    return size ? g2h(addr) : NULL;
+}
+
 #if defined(__i386__)
 
 #if defined(__NetBSD__)