Merge remote-tracking branch 'agraf/xen-next' into staging

17 files changed, 826 insertions, 185 deletions

diff --git a/Makefile.target b/Makefile.target
index b1a0f6d..760aa02 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -218,6 +218,8 @@ obj-$(CONFIG_NO_XEN) += xen-stub.o
 obj-i386-$(CONFIG_XEN_MAPCACHE) += xen-mapcache.o
 obj-$(CONFIG_NO_XEN_MAPCACHE) += xen-mapcache-stub.o
 
+obj-i386-$(CONFIG_XEN) += xen_platform.o
+
 # Inter-VM PCI shared memory
 CONFIG_IVSHMEM =
 ifeq ($(CONFIG_KVM), y)
diff --git a/configure b/configure
index 44c092a..b63b49f 100755
--- a/configure
+++ b/configure
@@ -1210,6 +1210,7 @@ int main(void) {
   xc = xc_interface_open(0, 0, 0);
   xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
   xc_gnttab_open(NULL, 0);
+  xc_domain_add_to_physmap(0, 0, XENMAPSPACE_gmfn, 0, 0);
   return 0;
 }
 EOF
@@ -1228,10 +1229,14 @@ EOF
 # error HVM_MAX_VCPUS not defined
 #endif
 int main(void) {
+  struct xen_add_to_physmap xatp = {
+    .domid = 0, .space = XENMAPSPACE_gmfn, .idx = 0, .gpfn = 0,
+  };
   xs_daemon_open();
   xc_interface_open();
   xc_gnttab_open();
   xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
+  xc_memory_op(0, XENMEM_add_to_physmap, &xatp);
   return 0;
 }
 EOF
@@ -1240,7 +1245,29 @@ EOF
     xen_ctrl_version=400
     xen=yes
 
-  # Xen 3.3.0, 3.4.0
+  # Xen 3.4.0
+  elif (
+      cat > $TMPC <<EOF
+#include <xenctrl.h>
+#include <xs.h>
+int main(void) {
+  struct xen_add_to_physmap xatp = {
+    .domid = 0, .space = XENMAPSPACE_gmfn, .idx = 0, .gpfn = 0,
+  };
+  xs_daemon_open();
+  xc_interface_open();
+  xc_gnttab_open();
+  xc_hvm_set_mem_type(0, 0, HVMMEM_ram_ro, 0, 0);
+  xc_memory_op(0, XENMEM_add_to_physmap, &xatp);
+  return 0;
+}
+EOF
+      compile_prog "" "$xen_libs"
+    ) ; then
+    xen_ctrl_version=340
+    xen=yes
+
+  # Xen 3.3.0
   elif (
       cat > $TMPC <<EOF
 #include <xenctrl.h>
diff --git a/cpu-common.h b/cpu-common.h
index 9f59172..b027e43 100644
--- a/cpu-common.h
+++ b/cpu-common.h
@@ -65,6 +65,7 @@ void qemu_ram_free_from_ptr(ram_addr_t addr);
 void qemu_ram_remap(ram_addr_t addr, ram_addr_t length);
 /* This should only be used for ram local to a device.  */
 void *qemu_get_ram_ptr(ram_addr_t addr);
+void *qemu_ram_ptr_length(target_phys_addr_t addr, target_phys_addr_t *size);
 /* Same but slower, to use for migration, where the order of
  * RAMBlocks must not change. */
 void *qemu_safe_ram_ptr(ram_addr_t addr);
diff --git a/exec.c b/exec.c
index 09928a3..7f14332 100644
--- a/exec.c
+++ b/exec.c
@@ -53,6 +53,7 @@
 #endif
 #else /* !CONFIG_USER_ONLY */
 #include "xen-mapcache.h"
+#include "trace.h"
 #endif
 
 //#define DEBUG_TB_INVALIDATE
@@ -3084,11 +3085,12 @@ void *qemu_get_ram_ptr(ram_addr_t addr)
             if (xen_mapcache_enabled()) {
                 /* We need to check if the requested address is in the RAM
                  * because we don't want to map the entire memory in QEMU.
+                 * In that case just map until the end of the page.
                  */
                 if (block->offset == 0) {
-                    return qemu_map_cache(addr, 0, 1);
+                    return qemu_map_cache(addr, 0, 0);
                 } else if (block->host == NULL) {
-                    block->host = xen_map_block(block->offset, block->length);
+                    block->host = qemu_map_cache(block->offset, block->length, 1);
                 }
             }
             return block->host + (addr - block->offset);
@@ -3113,11 +3115,12 @@ void *qemu_safe_ram_ptr(ram_addr_t addr)
             if (xen_mapcache_enabled()) {
                 /* We need to check if the requested address is in the RAM
                  * because we don't want to map the entire memory in QEMU.
+                 * In that case just map until the end of the page.
                  */
                 if (block->offset == 0) {
-                    return qemu_map_cache(addr, 0, 1);
+                    return qemu_map_cache(addr, 0, 0);
                 } else if (block->host == NULL) {
-                    block->host = xen_map_block(block->offset, block->length);
+                    block->host = qemu_map_cache(block->offset, block->length, 1);
                 }
             }
             return block->host + (addr - block->offset);
@@ -3130,32 +3133,46 @@ void *qemu_safe_ram_ptr(ram_addr_t addr)
     return NULL;
 }
 
-void qemu_put_ram_ptr(void *addr)
+/* Return a host pointer to guest's ram. Similar to qemu_get_ram_ptr
+ * but takes a size argument */
+void *qemu_ram_ptr_length(target_phys_addr_t addr, target_phys_addr_t *size)
 {
-    trace_qemu_put_ram_ptr(addr);
-
-    if (xen_mapcache_enabled()) {
+    if (xen_mapcache_enabled())
+        return qemu_map_cache(addr, *size, 1);
+    else {
         RAMBlock *block;
 
         QLIST_FOREACH(block, &ram_list.blocks, next) {
-            if (addr == block->host) {
-                break;
+            if (addr - block->offset < block->length) {
+                if (addr - block->offset + *size > block->length)
+                    *size = block->length - addr + block->offset;
+                return block->host + (addr - block->offset);
             }
         }
-        if (block && block->host) {
-            xen_unmap_block(block->host, block->length);
-            block->host = NULL;
-        } else {
-            qemu_map_cache_unlock(addr);
-        }
+
+        fprintf(stderr, "Bad ram offset %" PRIx64 "\n", (uint64_t)addr);
+        abort();
+
+        *size = 0;
+        return NULL;
     }
 }
 
+void qemu_put_ram_ptr(void *addr)
+{
+    trace_qemu_put_ram_ptr(addr);
+}
+
 int qemu_ram_addr_from_host(void *ptr, ram_addr_t *ram_addr)
 {
     RAMBlock *block;
     uint8_t *host = ptr;
 
+    if (xen_mapcache_enabled()) {
+        *ram_addr = qemu_ram_addr_from_mapcache(ptr);
+        return 0;
+    }
+
     QLIST_FOREACH(block, &ram_list.blocks, next) {
         /* This case append when the block is not mapped. */
         if (block->host == NULL) {
@@ -3167,11 +3184,6 @@ int qemu_ram_addr_from_host(void *ptr, ram_addr_t *ram_addr)
         }
     }
 
-    if (xen_mapcache_enabled()) {
-        *ram_addr = qemu_ram_addr_from_mapcache(ptr);
-        return 0;
-    }
-
     return -1;
 }
 
@@ -4003,14 +4015,12 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
                               int is_write)
 {
     target_phys_addr_t len = *plen;
-    target_phys_addr_t done = 0;
+    target_phys_addr_t todo = 0;
     int l;
-    uint8_t *ret = NULL;
-    uint8_t *ptr;
     target_phys_addr_t page;
     unsigned long pd;
     PhysPageDesc *p;
-    unsigned long addr1;
+    target_phys_addr_t addr1 = addr;
 
     while (len > 0) {
         page = addr & TARGET_PAGE_MASK;
@@ -4025,7 +4035,7 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
         }
 
         if ((pd & ~TARGET_PAGE_MASK) != IO_MEM_RAM) {
-            if (done || bounce.buffer) {
+            if (todo || bounce.buffer) {
                 break;
             }
             bounce.buffer = qemu_memalign(TARGET_PAGE_SIZE, TARGET_PAGE_SIZE);
@@ -4034,23 +4044,17 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
             if (!is_write) {
                 cpu_physical_memory_read(addr, bounce.buffer, l);
             }
-            ptr = bounce.buffer;
-        } else {
-            addr1 = (pd & TARGET_PAGE_MASK) + (addr & ~TARGET_PAGE_MASK);
-            ptr = qemu_get_ram_ptr(addr1);
-        }
-        if (!done) {
-            ret = ptr;
-        } else if (ret + done != ptr) {
-            break;
+
+            *plen = l;
+            return bounce.buffer;
         }
 
         len -= l;
         addr += l;
-        done += l;
+        todo += l;
     }
-    *plen = done;
-    return ret;
+    *plen = todo;
+    return qemu_ram_ptr_length(addr1, plen);
 }
 
 /* Unmaps a memory region previously mapped by cpu_physical_memory_map().
@@ -4080,13 +4084,7 @@ void cpu_physical_memory_unmap(void *buffer, target_phys_addr_t len,
             }
         }
         if (xen_mapcache_enabled()) {
-            uint8_t *buffer1 = buffer;
-            uint8_t *end_buffer = buffer + len;
-
-            while (buffer1 < end_buffer) {
-                qemu_put_ram_ptr(buffer1);
-                buffer1 += TARGET_PAGE_SIZE;
-            }
+            qemu_invalidate_entry(buffer);
         }
         return;
     }
diff --git a/hw/cirrus_vga.c b/hw/cirrus_vga.c
index 722cac7..3c5043e 100644
--- a/hw/cirrus_vga.c
+++ b/hw/cirrus_vga.c
@@ -3088,8 +3088,11 @@ static void pci_cirrus_write_config(PCIDevice *d,
     CirrusVGAState *s = &pvs->cirrus_vga;
 
     pci_default_write_config(d, address, val, len);
-    if (s->vga.map_addr && d->io_regions[0].addr == PCI_BAR_UNMAPPED)
+    if (s->vga.map_addr && d->io_regions[0].addr == PCI_BAR_UNMAPPED) {
         s->vga.map_addr = 0;
+        s->vga.lfb_addr = 0;
+        s->vga.lfb_end = 0;
+    }
     cirrus_update_memory_access(s);
 }
 
diff --git a/hw/hw.h b/hw/hw.h
index 56447a7..9dd7096 100644
--- a/hw/hw.h
+++ b/hw/hw.h
@@ -780,6 +780,9 @@ extern const VMStateDescription vmstate_ptimer;
 #define VMSTATE_INT32_LE(_f, _s)                                   \
     VMSTATE_SINGLE(_f, _s, 0, vmstate_info_int32_le, int32_t)
 
+#define VMSTATE_UINT8_TEST(_f, _s, _t)                               \
+    VMSTATE_SINGLE_TEST(_f, _s, _t, 0, vmstate_info_uint8, uint8_t)
+
 #define VMSTATE_UINT16_TEST(_f, _s, _t)                               \
     VMSTATE_SINGLE_TEST(_f, _s, _t, 0, vmstate_info_uint16, uint16_t)
 
diff --git a/hw/pc.h b/hw/pc.h
index 0dcbee7..6d5730b 100644
--- a/hw/pc.h
+++ b/hw/pc.h
@@ -176,7 +176,6 @@ struct PCII440FXState;
 typedef struct PCII440FXState PCII440FXState;
 
 PCIBus *i440fx_init(PCII440FXState **pi440fx_state, int *piix_devfn, qemu_irq *pic, ram_addr_t ram_size);
-PCIBus *i440fx_xen_init(PCII440FXState **pi440fx_state, int *piix3_devfn, qemu_irq *pic, ram_addr_t ram_size);
 void i440fx_init_memory_mappings(PCII440FXState *d);
 
 /* piix4.c */
diff --git a/hw/pc_piix.c b/hw/pc_piix.c
index 9a22a8a..8dbeb0c 100644
--- a/hw/pc_piix.c
+++ b/hw/pc_piix.c
@@ -124,11 +124,7 @@ static void pc_init1(ram_addr_t ram_size,
     isa_irq = qemu_allocate_irqs(isa_irq_handler, isa_irq_state, 24);
 
     if (pci_enabled) {
-        if (!xen_enabled()) {
-            pci_bus = i440fx_init(&i440fx_state, &piix3_devfn, isa_irq, ram_size);
-        } else {
-            pci_bus = i440fx_xen_init(&i440fx_state, &piix3_devfn, isa_irq, ram_size);
-        }
+        pci_bus = i440fx_init(&i440fx_state, &piix3_devfn, isa_irq, ram_size);
     } else {
         pci_bus = NULL;
         i440fx_state = NULL;
@@ -140,6 +136,10 @@ static void pc_init1(ram_addr_t ram_size,
 
     pc_vga_init(pci_enabled? pci_bus: NULL);
 
+    if (xen_enabled()) {
+        pci_create_simple(pci_bus, -1, "xen-platform");
+    }
+
     /* init basic PC hardware */
     pc_basic_device_init(isa_irq, &rtc_state, xen_enabled());
 
diff --git a/hw/pci_ids.h b/hw/pci_ids.h
index d9457ed..d94578c 100644
--- a/hw/pci_ids.h
+++ b/hw/pci_ids.h
@@ -109,3 +109,5 @@
 #define PCI_DEVICE_ID_INTEL_82371AB      0x7111
 #define PCI_DEVICE_ID_INTEL_82371AB_2    0x7112
 #define PCI_DEVICE_ID_INTEL_82371AB_3    0x7113
+
+#define PCI_VENDOR_ID_XENSOURCE          0x5853
diff --git a/hw/piix_pci.c b/hw/piix_pci.c
index 85a320e..3e2698d 100644
--- a/hw/piix_pci.c
+++ b/hw/piix_pci.c
@@ -40,6 +40,7 @@ typedef PCIHostState I440FXState;
 
 #define PIIX_NUM_PIC_IRQS       16      /* i8259 * 2 */
 #define PIIX_NUM_PIRQS          4ULL    /* PIRQ[A-D] */
+#define XEN_PIIX_NUM_PIRQS      128ULL
 #define PIIX_PIRQC              0x60
 
 typedef struct PIIX3State {
@@ -78,6 +79,8 @@ struct PCII440FXState {
 #define I440FX_SMRAM    0x72
 
 static void piix3_set_irq(void *opaque, int pirq, int level);
+static void piix3_write_config_xen(PCIDevice *dev,
+                               uint32_t address, uint32_t val, int len);
 
 /* return the global irq number corresponding to a given device irq
    pin. We could also use the bus number to have a more precise
@@ -173,13 +176,6 @@ static void i440fx_write_config(PCIDevice *dev,
     }
 }
 
-static void i440fx_write_config_xen(PCIDevice *dev,
-                                    uint32_t address, uint32_t val, int len)
-{
-    xen_piix_pci_write_config_client(address, val, len);
-    i440fx_write_config(dev, address, val, len);
-}
-
 static int i440fx_load_old(QEMUFile* f, void *opaque, int version_id)
 {
     PCII440FXState *d = opaque;
@@ -267,8 +263,21 @@ static PCIBus *i440fx_common_init(const char *device_name,
     d = pci_create_simple(b, 0, device_name);
     *pi440fx_state = DO_UPCAST(PCII440FXState, dev, d);
 
-    piix3 = DO_UPCAST(PIIX3State, dev,
-                      pci_create_simple_multifunction(b, -1, true, "PIIX3"));
+    /* Xen supports additional interrupt routes from the PCI devices to
+     * the IOAPIC: the four pins of each PCI device on the bus are also
+     * connected to the IOAPIC directly.
+     * These additional routes can be discovered through ACPI. */
+    if (xen_enabled()) {
+        piix3 = DO_UPCAST(PIIX3State, dev,
+                pci_create_simple_multifunction(b, -1, true, "PIIX3-xen"));
+        pci_bus_irqs(b, xen_piix3_set_irq, xen_pci_slot_get_pirq,
+                piix3, XEN_PIIX_NUM_PIRQS);
+    } else {
+        piix3 = DO_UPCAST(PIIX3State, dev,
+                pci_create_simple_multifunction(b, -1, true, "PIIX3"));
+        pci_bus_irqs(b, piix3_set_irq, pci_slot_get_pirq, piix3,
+                PIIX_NUM_PIRQS);
+    }
     piix3->pic = pic;
 
     (*pi440fx_state)->piix3 = piix3;
@@ -289,21 +298,6 @@ PCIBus *i440fx_init(PCII440FXState **pi440fx_state, int *piix3_devfn,
     PCIBus *b;
 
     b = i440fx_common_init("i440FX", pi440fx_state, piix3_devfn, pic, ram_size);
-    pci_bus_irqs(b, piix3_set_irq, pci_slot_get_pirq, (*pi440fx_state)->piix3,
-                 PIIX_NUM_PIRQS);
-
-    return b;
-}
-
-PCIBus *i440fx_xen_init(PCII440FXState **pi440fx_state, int *piix3_devfn,
-                        qemu_irq *pic, ram_addr_t ram_size)
-{
-    PCIBus *b;
-
-    b = i440fx_common_init("i440FX-xen", pi440fx_state, piix3_devfn, pic, ram_size);
-    pci_bus_irqs(b, xen_piix3_set_irq, xen_pci_slot_get_pirq,
-                 (*pi440fx_state)->piix3, PIIX_NUM_PIRQS);
-
     return b;
 }
 
@@ -365,6 +359,13 @@ static void piix3_write_config(PCIDevice *dev,
     }
 }
 
+static void piix3_write_config_xen(PCIDevice *dev,
+                               uint32_t address, uint32_t val, int len)
+{
+    xen_piix_pci_write_config_client(address, val, len);
+    piix3_write_config(dev, address, val, len);
+}
+
 static void piix3_reset(void *opaque)
 {
     PIIX3State *d = opaque;
@@ -465,14 +466,6 @@ static PCIDeviceInfo i440fx_info[] = {
         .init         = i440fx_initfn,
         .config_write = i440fx_write_config,
     },{
-        .qdev.name    = "i440FX-xen",
-        .qdev.desc    = "Host bridge",
-        .qdev.size    = sizeof(PCII440FXState),
-        .qdev.vmsd    = &vmstate_i440fx,
-        .qdev.no_user = 1,
-        .init         = i440fx_initfn,
-        .config_write = i440fx_write_config_xen,
-    },{
         .qdev.name    = "PIIX3",
         .qdev.desc    = "ISA bridge",
         .qdev.size    = sizeof(PIIX3State),
@@ -482,6 +475,15 @@ static PCIDeviceInfo i440fx_info[] = {
         .init         = piix3_initfn,
         .config_write = piix3_write_config,
     },{
+        .qdev.name    = "PIIX3-xen",
+        .qdev.desc    = "ISA bridge",
+        .qdev.size    = sizeof(PIIX3State),
+        .qdev.vmsd    = &vmstate_piix3,
+        .qdev.no_user = 1,
+        .no_hotplug   = 1,
+        .init         = piix3_initfn,
+        .config_write = piix3_write_config_xen,
+    },{
         /* end of list */
     }
 };
diff --git a/hw/xen_common.h b/hw/xen_common.h
index a1958a0..2c79af6 100644
--- a/hw/xen_common.h
+++ b/hw/xen_common.h
@@ -71,6 +71,20 @@ static inline int xc_domain_populate_physmap_exact
         (xc_handle, domid, nr_extents, extent_order, mem_flags, extent_start);
 }
 
+static inline int xc_domain_add_to_physmap(int xc_handle, uint32_t domid,
+                                           unsigned int space, unsigned long idx,
+                                           xen_pfn_t gpfn)
+{
+    struct xen_add_to_physmap xatp = {
+        .domid = domid,
+        .space = space,
+        .idx = idx,
+        .gpfn = gpfn,
+    };
+
+    return xc_memory_op(xc_handle, XENMEM_add_to_physmap, &xatp);
+}
+
 
 /* Xen 4.1 */
 #else
diff --git a/hw/xen_platform.c b/hw/xen_platform.c
new file mode 100644
index 0000000..b167eee
--- /dev/null
+++ b/hw/xen_platform.c
@@ -0,0 +1,340 @@
+/*
+ * XEN platform pci device, formerly known as the event channel device
+ *
+ * Copyright (c) 2003-2004 Intel Corp.
+ * Copyright (c) 2006 XenSource
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include <assert.h>
+
+#include "hw.h"
+#include "pc.h"
+#include "pci.h"
+#include "irq.h"
+#include "xen_common.h"
+#include "net.h"
+#include "xen_backend.h"
+#include "rwhandler.h"
+#include "trace.h"
+
+#include <xenguest.h>
+
+//#define DEBUG_PLATFORM
+
+#ifdef DEBUG_PLATFORM
+#define DPRINTF(fmt, ...) do { \
+    fprintf(stderr, "xen_platform: " fmt, ## __VA_ARGS__); \
+} while (0)
+#else
+#define DPRINTF(fmt, ...) do { } while (0)
+#endif
+
+#define PFFLAG_ROM_LOCK 1 /* Sets whether ROM memory area is RW or RO */
+
+typedef struct PCIXenPlatformState {
+    PCIDevice  pci_dev;
+    uint8_t flags; /* used only for version_id == 2 */
+    int drivers_blacklisted;
+    uint16_t driver_product_version;
+
+    /* Log from guest drivers */
+    char log_buffer[4096];
+    int log_buffer_off;
+} PCIXenPlatformState;
+
+#define XEN_PLATFORM_IOPORT 0x10
+
+/* Send bytes to syslog */
+static void log_writeb(PCIXenPlatformState *s, char val)
+{
+    if (val == '\n' || s->log_buffer_off == sizeof(s->log_buffer) - 1) {
+        /* Flush buffer */
+        s->log_buffer[s->log_buffer_off] = 0;
+        trace_xen_platform_log(s->log_buffer);
+        s->log_buffer_off = 0;
+    } else {
+        s->log_buffer[s->log_buffer_off++] = val;
+    }
+}
+
+/* Xen Platform, Fixed IOPort */
+
+static void platform_fixed_ioport_writew(void *opaque, uint32_t addr, uint32_t val)
+{
+    PCIXenPlatformState *s = opaque;
+
+    switch (addr - XEN_PLATFORM_IOPORT) {
+    case 0:
+        /* TODO: */
+        /* Unplug devices.  Value is a bitmask of which devices to
+           unplug, with bit 0 the IDE devices, bit 1 the network
+           devices, and bit 2 the non-primary-master IDE devices. */
+        break;
+    case 2:
+        switch (val) {
+        case 1:
+            DPRINTF("Citrix Windows PV drivers loaded in guest\n");
+            break;
+        case 0:
+            DPRINTF("Guest claimed to be running PV product 0?\n");
+            break;
+        default:
+            DPRINTF("Unknown PV product %d loaded in guest\n", val);
+            break;
+        }
+        s->driver_product_version = val;
+        break;
+    }
+}
+
+static void platform_fixed_ioport_writel(void *opaque, uint32_t addr,
+                                         uint32_t val)
+{
+    switch (addr - XEN_PLATFORM_IOPORT) {
+    case 0:
+        /* PV driver version */
+        break;
+    }
+}
+
+static void platform_fixed_ioport_writeb(void *opaque, uint32_t addr, uint32_t val)
+{
+    PCIXenPlatformState *s = opaque;
+
+    switch (addr - XEN_PLATFORM_IOPORT) {
+    case 0: /* Platform flags */ {
+        hvmmem_type_t mem_type = (val & PFFLAG_ROM_LOCK) ?
+            HVMMEM_ram_ro : HVMMEM_ram_rw;
+        if (xc_hvm_set_mem_type(xen_xc, xen_domid, mem_type, 0xc0, 0x40)) {
+            DPRINTF("unable to change ro/rw state of ROM memory area!\n");
+        } else {
+            s->flags = val & PFFLAG_ROM_LOCK;
+            DPRINTF("changed ro/rw state of ROM memory area. now is %s state.\n",
+                    (mem_type == HVMMEM_ram_ro ? "ro":"rw"));
+        }
+        break;
+    }
+    case 2:
+        log_writeb(s, val);
+        break;
+    }
+}
+
+static uint32_t platform_fixed_ioport_readw(void *opaque, uint32_t addr)
+{
+    PCIXenPlatformState *s = opaque;
+
+    switch (addr - XEN_PLATFORM_IOPORT) {
+    case 0:
+        if (s->drivers_blacklisted) {
+            /* The drivers will recognise this magic number and refuse
+             * to do anything. */
+            return 0xd249;
+        } else {
+            /* Magic value so that you can identify the interface. */
+            return 0x49d2;
+        }
+    default:
+        return 0xffff;
+    }
+}
+
+static uint32_t platform_fixed_ioport_readb(void *opaque, uint32_t addr)
+{
+    PCIXenPlatformState *s = opaque;
+
+    switch (addr - XEN_PLATFORM_IOPORT) {
+    case 0:
+        /* Platform flags */
+        return s->flags;
+    case 2:
+        /* Version number */
+        return 1;
+    default:
+        return 0xff;
+    }
+}
+
+static void platform_fixed_ioport_reset(void *opaque)
+{
+    PCIXenPlatformState *s = opaque;
+
+    platform_fixed_ioport_writeb(s, XEN_PLATFORM_IOPORT, 0);
+}
+
+static void platform_fixed_ioport_init(PCIXenPlatformState* s)
+{
+    register_ioport_write(XEN_PLATFORM_IOPORT, 16, 4, platform_fixed_ioport_writel, s);
+    register_ioport_write(XEN_PLATFORM_IOPORT, 16, 2, platform_fixed_ioport_writew, s);
+    register_ioport_write(XEN_PLATFORM_IOPORT, 16, 1, platform_fixed_ioport_writeb, s);
+    register_ioport_read(XEN_PLATFORM_IOPORT, 16, 2, platform_fixed_ioport_readw, s);
+    register_ioport_read(XEN_PLATFORM_IOPORT, 16, 1, platform_fixed_ioport_readb, s);
+}
+
+/* Xen Platform PCI Device */
+
+static uint32_t xen_platform_ioport_readb(void *opaque, uint32_t addr)
+{
+    addr &= 0xff;
+
+    if (addr == 0) {
+        return platform_fixed_ioport_readb(opaque, XEN_PLATFORM_IOPORT);
+    } else {
+        return ~0u;
+    }
+}
+
+static void xen_platform_ioport_writeb(void *opaque, uint32_t addr, uint32_t val)
+{
+    PCIXenPlatformState *s = opaque;
+
+    addr &= 0xff;
+    val  &= 0xff;
+
+    switch (addr) {
+    case 0: /* Platform flags */
+        platform_fixed_ioport_writeb(opaque, XEN_PLATFORM_IOPORT, val);
+        break;
+    case 8:
+        log_writeb(s, val);
+        break;
+    default:
+        break;
+    }
+}
+
+static void platform_ioport_map(PCIDevice *pci_dev, int region_num, pcibus_t addr, pcibus_t size, int type)
+{
+    PCIXenPlatformState *d = DO_UPCAST(PCIXenPlatformState, pci_dev, pci_dev);
+
+    register_ioport_write(addr, size, 1, xen_platform_ioport_writeb, d);
+    register_ioport_read(addr, size, 1, xen_platform_ioport_readb, d);
+}
+
+static uint32_t platform_mmio_read(ReadWriteHandler *handler, pcibus_t addr, int len)
+{
+    DPRINTF("Warning: attempted read from physical address "
+            "0x" TARGET_FMT_plx " in xen platform mmio space\n", addr);
+
+    return 0;
+}
+
+static void platform_mmio_write(ReadWriteHandler *handler, pcibus_t addr,
+                                uint32_t val, int len)
+{
+    DPRINTF("Warning: attempted write of 0x%x to physical "
+            "address 0x" TARGET_FMT_plx " in xen platform mmio space\n",
+            val, addr);
+}
+
+static ReadWriteHandler platform_mmio_handler = {
+    .read = &platform_mmio_read,
+    .write = &platform_mmio_write,
+};
+
+static void platform_mmio_map(PCIDevice *d, int region_num,
+                              pcibus_t addr, pcibus_t size, int type)
+{
+    int mmio_io_addr;
+
+    mmio_io_addr = cpu_register_io_memory_simple(&platform_mmio_handler,
+                                                 DEVICE_NATIVE_ENDIAN);
+
+    cpu_register_physical_memory(addr, size, mmio_io_addr);
+}
+
+static int xen_platform_post_load(void *opaque, int version_id)
+{
+    PCIXenPlatformState *s = opaque;
+
+    platform_fixed_ioport_writeb(s, XEN_PLATFORM_IOPORT, s->flags);
+
+    return 0;
+}
+
+static const VMStateDescription vmstate_xen_platform = {
+    .name = "platform",
+    .version_id = 4,
+    .minimum_version_id = 4,
+    .minimum_version_id_old = 4,
+    .post_load = xen_platform_post_load,
+    .fields = (VMStateField []) {
+        VMSTATE_PCI_DEVICE(pci_dev, PCIXenPlatformState),
+        VMSTATE_UINT8(flags, PCIXenPlatformState),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
+static int xen_platform_initfn(PCIDevice *dev)
+{
+    PCIXenPlatformState *d = DO_UPCAST(PCIXenPlatformState, pci_dev, dev);
+    uint8_t *pci_conf;
+
+    pci_conf = d->pci_dev.config;
+
+    pci_config_set_vendor_id(pci_conf, PCI_VENDOR_ID_XENSOURCE);
+    pci_config_set_device_id(pci_conf, 0x0001);
+    pci_set_word(pci_conf + PCI_SUBSYSTEM_VENDOR_ID, PCI_VENDOR_ID_XENSOURCE);
+    pci_set_word(pci_conf + PCI_SUBSYSTEM_ID, 0x0001);
+
+    pci_set_word(pci_conf + PCI_COMMAND, PCI_COMMAND_IO | PCI_COMMAND_MEMORY);
+
+    pci_config_set_revision(pci_conf, 1);
+    pci_config_set_prog_interface(pci_conf, 0);
+
+    pci_config_set_class(pci_conf, PCI_CLASS_OTHERS << 8 | 0x80);
+
+    pci_conf[PCI_INTERRUPT_PIN] = 1;
+
+    pci_register_bar(&d->pci_dev, 0, 0x100,
+            PCI_BASE_ADDRESS_SPACE_IO, platform_ioport_map);
+
+    /* reserve 16MB mmio address for share memory*/
+    pci_register_bar(&d->pci_dev, 1, 0x1000000,
+            PCI_BASE_ADDRESS_MEM_PREFETCH, platform_mmio_map);
+
+    platform_fixed_ioport_init(d);
+
+    return 0;
+}
+
+static void platform_reset(DeviceState *dev)
+{
+    PCIXenPlatformState *s = DO_UPCAST(PCIXenPlatformState, pci_dev.qdev, dev);
+
+    platform_fixed_ioport_reset(s);
+}
+
+static PCIDeviceInfo xen_platform_info = {
+    .init = xen_platform_initfn,
+    .qdev.name = "xen-platform",
+    .qdev.desc = "XEN platform pci device",
+    .qdev.size = sizeof(PCIXenPlatformState),
+    .qdev.vmsd = &vmstate_xen_platform,
+    .qdev.reset = platform_reset,
+};
+
+static void xen_platform_register(void)
+{
+    pci_qdev_register(&xen_platform_info);
+}
+
+device_init(xen_platform_register);
diff --git a/trace-events b/trace-events
index f1230f1..bebf612f 100644
--- a/trace-events
+++ b/trace-events
@@ -396,6 +396,7 @@ disable milkymist_vgafb_memory_write(uint32_t addr, uint32_t value) "addr %08x v
 
 # xen-all.c
 disable xen_ram_alloc(unsigned long ram_addr, unsigned long size) "requested: %#lx, size %#lx"
+disable xen_client_set_memory(uint64_t start_addr, unsigned long size, unsigned long phys_offset, bool log_dirty) "%#"PRIx64" size %#lx, offset %#lx, log_dirty %i"
 
 # xen-mapcache.c
 disable qemu_map_cache(uint64_t phys_addr) "want %#"PRIx64""
@@ -406,3 +407,6 @@ disable xen_unmap_block(void* addr, unsigned long size) "%p, size %#lx"
 
 # exec.c
 disable qemu_put_ram_ptr(void* addr) "%p"
+
+# hw/xen_platform.c
+disable xen_platform_log(char *s) "xen platform: %s"
diff --git a/xen-all.c b/xen-all.c
index 0eac202..fe75ddd 100644
--- a/xen-all.c
+++ b/xen-all.c
@@ -13,6 +13,7 @@
 #include "hw/xen_common.h"
 #include "hw/xen_backend.h"
 
+#include "range.h"
 #include "xen-mapcache.h"
 #include "trace.h"
 
@@ -54,6 +55,14 @@ static inline ioreq_t *xen_vcpu_ioreq(shared_iopage_t *shared_page, int vcpu)
 
 #define BUFFER_IO_MAX_DELAY  100
 
+typedef struct XenPhysmap {
+    target_phys_addr_t start_addr;
+    ram_addr_t size;
+    target_phys_addr_t phys_offset;
+
+    QLIST_ENTRY(XenPhysmap) list;
+} XenPhysmap;
+
 typedef struct XenIOState {
     shared_iopage_t *shared_page;
     buffered_iopage_t *buffered_io_page;
@@ -66,6 +75,9 @@ typedef struct XenIOState {
     int send_vcpu;
 
     struct xs_handle *xenstore;
+    CPUPhysMemoryClient client;
+    QLIST_HEAD(, XenPhysmap) physmap;
+    const XenPhysmap *log_for_dirtybit;
 
     Notifier exit;
 } XenIOState;
@@ -178,6 +190,270 @@ void xen_ram_alloc(ram_addr_t ram_addr, ram_addr_t size)
     qemu_free(pfn_list);
 }
 
+static XenPhysmap *get_physmapping(XenIOState *state,
+                                   target_phys_addr_t start_addr, ram_addr_t size)
+{
+    XenPhysmap *physmap = NULL;
+
+    start_addr &= TARGET_PAGE_MASK;
+
+    QLIST_FOREACH(physmap, &state->physmap, list) {
+        if (range_covers_byte(physmap->start_addr, physmap->size, start_addr)) {
+            return physmap;
+        }
+    }
+    return NULL;
+}
+
+#if CONFIG_XEN_CTRL_INTERFACE_VERSION >= 340
+static int xen_add_to_physmap(XenIOState *state,
+                              target_phys_addr_t start_addr,
+                              ram_addr_t size,
+                              target_phys_addr_t phys_offset)
+{
+    unsigned long i = 0;
+    int rc = 0;
+    XenPhysmap *physmap = NULL;
+    target_phys_addr_t pfn, start_gpfn;
+    RAMBlock *block;
+
+    if (get_physmapping(state, start_addr, size)) {
+        return 0;
+    }
+    if (size <= 0) {
+        return -1;
+    }
+
+    /* Xen can only handle a single dirty log region for now and we want
+     * the linear framebuffer to be that region.
+     * Avoid tracking any regions that is not videoram and avoid tracking
+     * the legacy vga region. */
+    QLIST_FOREACH(block, &ram_list.blocks, next) {
+        if (!strcmp(block->idstr, "vga.vram") && block->offset == phys_offset
+                && start_addr > 0xbffff) {
+            goto go_physmap;
+        }
+    }
+    return -1;
+
+go_physmap:
+    DPRINTF("mapping vram to %llx - %llx, from %llx\n",
+            start_addr, start_addr + size, phys_offset);
+
+    pfn = phys_offset >> TARGET_PAGE_BITS;
+    start_gpfn = start_addr >> TARGET_PAGE_BITS;
+    for (i = 0; i < size >> TARGET_PAGE_BITS; i++) {
+        unsigned long idx = pfn + i;
+        xen_pfn_t gpfn = start_gpfn + i;
+
+        rc = xc_domain_add_to_physmap(xen_xc, xen_domid, XENMAPSPACE_gmfn, idx, gpfn);
+        if (rc) {
+            DPRINTF("add_to_physmap MFN %"PRI_xen_pfn" to PFN %"
+                    PRI_xen_pfn" failed: %d\n", idx, gpfn, rc);
+            return -rc;
+        }
+    }
+
+    physmap = qemu_malloc(sizeof (XenPhysmap));
+
+    physmap->start_addr = start_addr;
+    physmap->size = size;
+    physmap->phys_offset = phys_offset;
+
+    QLIST_INSERT_HEAD(&state->physmap, physmap, list);
+
+    xc_domain_pin_memory_cacheattr(xen_xc, xen_domid,
+                                   start_addr >> TARGET_PAGE_BITS,
+                                   (start_addr + size) >> TARGET_PAGE_BITS,
+                                   XEN_DOMCTL_MEM_CACHEATTR_WB);
+    return 0;
+}
+
+static int xen_remove_from_physmap(XenIOState *state,
+                                   target_phys_addr_t start_addr,
+                                   ram_addr_t size)
+{
+    unsigned long i = 0;
+    int rc = 0;
+    XenPhysmap *physmap = NULL;
+    target_phys_addr_t phys_offset = 0;
+
+    physmap = get_physmapping(state, start_addr, size);
+    if (physmap == NULL) {
+        return -1;
+    }
+
+    phys_offset = physmap->phys_offset;
+    size = physmap->size;
+
+    DPRINTF("unmapping vram to %llx - %llx, from %llx\n",
+            phys_offset, phys_offset + size, start_addr);
+
+    size >>= TARGET_PAGE_BITS;
+    start_addr >>= TARGET_PAGE_BITS;
+    phys_offset >>= TARGET_PAGE_BITS;
+    for (i = 0; i < size; i++) {
+        unsigned long idx = start_addr + i;
+        xen_pfn_t gpfn = phys_offset + i;
+
+        rc = xc_domain_add_to_physmap(xen_xc, xen_domid, XENMAPSPACE_gmfn, idx, gpfn);
+        if (rc) {
+            fprintf(stderr, "add_to_physmap MFN %"PRI_xen_pfn" to PFN %"
+                    PRI_xen_pfn" failed: %d\n", idx, gpfn, rc);
+            return -rc;
+        }
+    }
+
+    QLIST_REMOVE(physmap, list);
+    if (state->log_for_dirtybit == physmap) {
+        state->log_for_dirtybit = NULL;
+    }
+    free(physmap);
+
+    return 0;
+}
+
+#else
+static int xen_add_to_physmap(XenIOState *state,
+                              target_phys_addr_t start_addr,
+                              ram_addr_t size,
+                              target_phys_addr_t phys_offset)
+{
+    return -ENOSYS;
+}
+
+static int xen_remove_from_physmap(XenIOState *state,
+                                   target_phys_addr_t start_addr,
+                                   ram_addr_t size)
+{
+    return -ENOSYS;
+}
+#endif
+
+static void xen_client_set_memory(struct CPUPhysMemoryClient *client,
+                                  target_phys_addr_t start_addr,
+                                  ram_addr_t size,
+                                  ram_addr_t phys_offset,
+                                  bool log_dirty)
+{
+    XenIOState *state = container_of(client, XenIOState, client);
+    ram_addr_t flags = phys_offset & ~TARGET_PAGE_MASK;
+    hvmmem_type_t mem_type;
+
+    if (!(start_addr != phys_offset
+          && ( (log_dirty && flags < IO_MEM_UNASSIGNED)
+               || (!log_dirty && flags == IO_MEM_UNASSIGNED)))) {
+        return;
+    }
+
+    trace_xen_client_set_memory(start_addr, size, phys_offset, log_dirty);
+
+    start_addr &= TARGET_PAGE_MASK;
+    size = TARGET_PAGE_ALIGN(size);
+    phys_offset &= TARGET_PAGE_MASK;
+
+    switch (flags) {
+    case IO_MEM_RAM:
+        xen_add_to_physmap(state, start_addr, size, phys_offset);
+        break;
+    case IO_MEM_ROM:
+        mem_type = HVMMEM_ram_ro;
+        if (xc_hvm_set_mem_type(xen_xc, xen_domid, mem_type,
+                                start_addr >> TARGET_PAGE_BITS,
+                                size >> TARGET_PAGE_BITS)) {
+            DPRINTF("xc_hvm_set_mem_type error, addr: "TARGET_FMT_plx"\n",
+                    start_addr);
+        }
+        break;
+    case IO_MEM_UNASSIGNED:
+        if (xen_remove_from_physmap(state, start_addr, size) < 0) {
+            DPRINTF("physmapping does not exist at "TARGET_FMT_plx"\n", start_addr);
+        }
+        break;
+    }
+}
+
+static int xen_sync_dirty_bitmap(XenIOState *state,
+                                 target_phys_addr_t start_addr,
+                                 ram_addr_t size)
+{
+    target_phys_addr_t npages = size >> TARGET_PAGE_BITS;
+    target_phys_addr_t vram_offset = 0;
+    const int width = sizeof(unsigned long) * 8;
+    unsigned long bitmap[(npages + width - 1) / width];
+    int rc, i, j;
+    const XenPhysmap *physmap = NULL;
+
+    physmap = get_physmapping(state, start_addr, size);
+    if (physmap == NULL) {
+        /* not handled */
+        return -1;
+    }
+
+    if (state->log_for_dirtybit == NULL) {
+        state->log_for_dirtybit = physmap;
+    } else if (state->log_for_dirtybit != physmap) {
+        return -1;
+    }
+    vram_offset = physmap->phys_offset;
+
+    rc = xc_hvm_track_dirty_vram(xen_xc, xen_domid,
+                                 start_addr >> TARGET_PAGE_BITS, npages,
+                                 bitmap);
+    if (rc) {
+        return rc;
+    }
+
+    for (i = 0; i < ARRAY_SIZE(bitmap); i++) {
+        unsigned long map = bitmap[i];
+        while (map != 0) {
+            j = ffsl(map) - 1;
+            map &= ~(1ul << j);
+            cpu_physical_memory_set_dirty(vram_offset + (i * width + j) * TARGET_PAGE_SIZE);
+        };
+    }
+
+    return 0;
+}
+
+static int xen_log_start(CPUPhysMemoryClient *client, target_phys_addr_t phys_addr, ram_addr_t size)
+{
+    XenIOState *state = container_of(client, XenIOState, client);
+
+    return xen_sync_dirty_bitmap(state, phys_addr, size);
+}
+
+static int xen_log_stop(CPUPhysMemoryClient *client, target_phys_addr_t phys_addr, ram_addr_t size)
+{
+    XenIOState *state = container_of(client, XenIOState, client);
+
+    state->log_for_dirtybit = NULL;
+    /* Disable dirty bit tracking */
+    return xc_hvm_track_dirty_vram(xen_xc, xen_domid, 0, 0, NULL);
+}
+
+static int xen_client_sync_dirty_bitmap(struct CPUPhysMemoryClient *client,
+                                        target_phys_addr_t start_addr,
+                                        target_phys_addr_t end_addr)
+{
+    XenIOState *state = container_of(client, XenIOState, client);
+
+    return xen_sync_dirty_bitmap(state, start_addr, end_addr - start_addr);
+}
+
+static int xen_client_migration_log(struct CPUPhysMemoryClient *client,
+                                    int enable)
+{
+    return 0;
+}
+
+static CPUPhysMemoryClient xen_cpu_phys_memory_client = {
+    .set_memory = xen_client_set_memory,
+    .sync_dirty_bitmap = xen_client_sync_dirty_bitmap,
+    .migration_log = xen_client_migration_log,
+    .log_start = xen_log_start,
+    .log_stop = xen_log_stop,
+};
 
 /* VCPU Operations, MMIO, IO ring ... */
 
@@ -581,6 +857,11 @@ int xen_hvm_init(void)
 
     qemu_add_vm_change_state_handler(xen_vm_change_state_handler, state);
 
+    state->client = xen_cpu_phys_memory_client;
+    QLIST_INIT(&state->physmap);
+    cpu_register_phys_memory_client(&state->client);
+    state->log_for_dirtybit = NULL;
+
     return 0;
 }
 
diff --git a/xen-mapcache-stub.c b/xen-mapcache-stub.c
index 7c14b3d..8a2380a 100644
--- a/xen-mapcache-stub.c
+++ b/xen-mapcache-stub.c
@@ -22,10 +22,6 @@ uint8_t *qemu_map_cache(target_phys_addr_t phys_addr, target_phys_addr_t size, u
     return qemu_get_ram_ptr(phys_addr);
 }
 
-void qemu_map_cache_unlock(void *buffer)
-{
-}
-
 ram_addr_t qemu_ram_addr_from_mapcache(void *ptr)
 {
     return -1;
@@ -38,7 +34,3 @@ void qemu_invalidate_map_cache(void)
 void qemu_invalidate_entry(uint8_t *buffer)
 {
 }
-uint8_t *xen_map_block(target_phys_addr_t phys_addr, target_phys_addr_t size)
-{
-    return NULL;
-}
diff --git a/xen-mapcache.c b/xen-mapcache.c
index 349cc62..fac47cd 100644
--- a/xen-mapcache.c
+++ b/xen-mapcache.c
@@ -43,14 +43,16 @@
 typedef struct MapCacheEntry {
     target_phys_addr_t paddr_index;
     uint8_t *vaddr_base;
-    DECLARE_BITMAP(valid_mapping, MCACHE_BUCKET_SIZE >> XC_PAGE_SHIFT);
+    unsigned long *valid_mapping;
     uint8_t lock;
+    target_phys_addr_t size;
     struct MapCacheEntry *next;
 } MapCacheEntry;
 
 typedef struct MapCacheRev {
     uint8_t *vaddr_req;
     target_phys_addr_t paddr_index;
+    target_phys_addr_t size;
     QTAILQ_ENTRY(MapCacheRev) next;
 } MapCacheRev;
 
@@ -68,6 +70,15 @@ typedef struct MapCache {
 
 static MapCache *mapcache;
 
+static inline int test_bits(int nr, int size, const unsigned long *addr)
+{
+    unsigned long res = find_next_zero_bit(addr, size + nr, nr);
+    if (res >= nr + size)
+        return 1;
+    else
+        return 0;
+}
+
 void qemu_map_cache_init(void)
 {
     unsigned long size;
@@ -115,11 +126,15 @@ static void qemu_remap_bucket(MapCacheEntry *entry,
     err = qemu_mallocz(nb_pfn * sizeof (int));
 
     if (entry->vaddr_base != NULL) {
-        if (munmap(entry->vaddr_base, size) != 0) {
+        if (munmap(entry->vaddr_base, entry->size) != 0) {
             perror("unmap fails");
             exit(-1);
         }
     }
+    if (entry->valid_mapping != NULL) {
+        qemu_free(entry->valid_mapping);
+        entry->valid_mapping = NULL;
+    }
 
     for (i = 0; i < nb_pfn; i++) {
         pfns[i] = (address_index << (MCACHE_BUCKET_SHIFT-XC_PAGE_SHIFT)) + i;
@@ -134,6 +149,9 @@ static void qemu_remap_bucket(MapCacheEntry *entry,
 
     entry->vaddr_base = vaddr_base;
     entry->paddr_index = address_index;
+    entry->size = size;
+    entry->valid_mapping = (unsigned long *) qemu_mallocz(sizeof(unsigned long) *
+            BITS_TO_LONGS(size >> XC_PAGE_SHIFT));
 
     bitmap_zero(entry->valid_mapping, nb_pfn);
     for (i = 0; i < nb_pfn; i++) {
@@ -151,32 +169,47 @@ uint8_t *qemu_map_cache(target_phys_addr_t phys_addr, target_phys_addr_t size, u
     MapCacheEntry *entry, *pentry = NULL;
     target_phys_addr_t address_index  = phys_addr >> MCACHE_BUCKET_SHIFT;
     target_phys_addr_t address_offset = phys_addr & (MCACHE_BUCKET_SIZE - 1);
+    target_phys_addr_t __size = size;
 
     trace_qemu_map_cache(phys_addr);
 
-    if (address_index == mapcache->last_address_index && !lock) {
+    if (address_index == mapcache->last_address_index && !lock && !__size) {
         trace_qemu_map_cache_return(mapcache->last_address_vaddr + address_offset);
         return mapcache->last_address_vaddr + address_offset;
     }
 
+    /* size is always a multiple of MCACHE_BUCKET_SIZE */
+    if ((address_offset + (__size % MCACHE_BUCKET_SIZE)) > MCACHE_BUCKET_SIZE)
+        __size += MCACHE_BUCKET_SIZE;
+    if (__size % MCACHE_BUCKET_SIZE)
+        __size += MCACHE_BUCKET_SIZE - (__size % MCACHE_BUCKET_SIZE);
+    if (!__size)
+        __size = MCACHE_BUCKET_SIZE;
+
     entry = &mapcache->entry[address_index % mapcache->nr_buckets];
 
-    while (entry && entry->lock && entry->paddr_index != address_index && entry->vaddr_base) {
+    while (entry && entry->lock && entry->vaddr_base &&
+            (entry->paddr_index != address_index || entry->size != __size ||
+             !test_bits(address_offset >> XC_PAGE_SHIFT, size >> XC_PAGE_SHIFT,
+                 entry->valid_mapping))) {
         pentry = entry;
         entry = entry->next;
     }
     if (!entry) {
         entry = qemu_mallocz(sizeof (MapCacheEntry));
         pentry->next = entry;
-        qemu_remap_bucket(entry, size ? : MCACHE_BUCKET_SIZE, address_index);
+        qemu_remap_bucket(entry, __size, address_index);
     } else if (!entry->lock) {
         if (!entry->vaddr_base || entry->paddr_index != address_index ||
-            !test_bit(address_offset >> XC_PAGE_SHIFT, entry->valid_mapping)) {
-            qemu_remap_bucket(entry, size ? : MCACHE_BUCKET_SIZE, address_index);
+                entry->size != __size ||
+                !test_bits(address_offset >> XC_PAGE_SHIFT, size >> XC_PAGE_SHIFT,
+                    entry->valid_mapping)) {
+            qemu_remap_bucket(entry, __size, address_index);
         }
     }
 
-    if (!test_bit(address_offset >> XC_PAGE_SHIFT, entry->valid_mapping)) {
+    if(!test_bits(address_offset >> XC_PAGE_SHIFT, size >> XC_PAGE_SHIFT,
+                entry->valid_mapping)) {
         mapcache->last_address_index = -1;
         trace_qemu_map_cache_return(NULL);
         return NULL;
@@ -189,6 +222,7 @@ uint8_t *qemu_map_cache(target_phys_addr_t phys_addr, target_phys_addr_t size, u
         entry->lock++;
         reventry->vaddr_req = mapcache->last_address_vaddr + address_offset;
         reventry->paddr_index = mapcache->last_address_index;
+        reventry->size = entry->size;
         QTAILQ_INSERT_HEAD(&mapcache->locked_entries, reventry, next);
     }
 
@@ -196,48 +230,18 @@ uint8_t *qemu_map_cache(target_phys_addr_t phys_addr, target_phys_addr_t size, u
     return mapcache->last_address_vaddr + address_offset;
 }
 
-void qemu_map_cache_unlock(void *buffer)
-{
-    MapCacheEntry *entry = NULL, *pentry = NULL;
-    MapCacheRev *reventry;
-    target_phys_addr_t paddr_index;
-    int found = 0;
-
-    QTAILQ_FOREACH(reventry, &mapcache->locked_entries, next) {
-        if (reventry->vaddr_req == buffer) {
-            paddr_index = reventry->paddr_index;
-            found = 1;
-            break;
-        }
-    }
-    if (!found) {
-        return;
-    }
-    QTAILQ_REMOVE(&mapcache->locked_entries, reventry, next);
-    qemu_free(reventry);
-
-    entry = &mapcache->entry[paddr_index % mapcache->nr_buckets];
-    while (entry && entry->paddr_index != paddr_index) {
-        pentry = entry;
-        entry = entry->next;
-    }
-    if (!entry) {
-        return;
-    }
-    if (entry->lock > 0) {
-        entry->lock--;
-    }
-}
-
 ram_addr_t qemu_ram_addr_from_mapcache(void *ptr)
 {
+    MapCacheEntry *entry = NULL, *pentry = NULL;
     MapCacheRev *reventry;
     target_phys_addr_t paddr_index;
+    target_phys_addr_t size;
     int found = 0;
 
     QTAILQ_FOREACH(reventry, &mapcache->locked_entries, next) {
         if (reventry->vaddr_req == ptr) {
             paddr_index = reventry->paddr_index;
+            size = reventry->size;
             found = 1;
             break;
         }
@@ -252,7 +256,17 @@ ram_addr_t qemu_ram_addr_from_mapcache(void *ptr)
         return 0;
     }
 
-    return paddr_index << MCACHE_BUCKET_SHIFT;
+    entry = &mapcache->entry[paddr_index % mapcache->nr_buckets];
+    while (entry && (entry->paddr_index != paddr_index || entry->size != size)) {
+        pentry = entry;
+        entry = entry->next;
+    }
+    if (!entry) {
+        DPRINTF("Trying to find address %p that is not in the mapcache!\n", ptr);
+        return 0;
+    }
+    return (reventry->paddr_index << MCACHE_BUCKET_SHIFT) +
+        ((unsigned long) ptr - (unsigned long) entry->vaddr_base);
 }
 
 void qemu_invalidate_entry(uint8_t *buffer)
@@ -260,6 +274,7 @@ void qemu_invalidate_entry(uint8_t *buffer)
     MapCacheEntry *entry = NULL, *pentry = NULL;
     MapCacheRev *reventry;
     target_phys_addr_t paddr_index;
+    target_phys_addr_t size;
     int found = 0;
 
     if (mapcache->last_address_vaddr == buffer) {
@@ -269,6 +284,7 @@ void qemu_invalidate_entry(uint8_t *buffer)
     QTAILQ_FOREACH(reventry, &mapcache->locked_entries, next) {
         if (reventry->vaddr_req == buffer) {
             paddr_index = reventry->paddr_index;
+            size = reventry->size;
             found = 1;
             break;
         }
@@ -284,7 +300,7 @@ void qemu_invalidate_entry(uint8_t *buffer)
     qemu_free(reventry);
 
     entry = &mapcache->entry[paddr_index % mapcache->nr_buckets];
-    while (entry && entry->paddr_index != paddr_index) {
+    while (entry && (entry->paddr_index != paddr_index || entry->size != size)) {
         pentry = entry;
         entry = entry->next;
     }
@@ -298,10 +314,11 @@ void qemu_invalidate_entry(uint8_t *buffer)
     }
 
     pentry->next = entry->next;
-    if (munmap(entry->vaddr_base, MCACHE_BUCKET_SIZE) != 0) {
+    if (munmap(entry->vaddr_base, entry->size) != 0) {
         perror("unmap fails");
         exit(-1);
     }
+    qemu_free(entry->valid_mapping);
     qemu_free(entry);
 }
 
@@ -328,13 +345,16 @@ void qemu_invalidate_map_cache(void)
             continue;
         }
 
-        if (munmap(entry->vaddr_base, MCACHE_BUCKET_SIZE) != 0) {
+        if (munmap(entry->vaddr_base, entry->size) != 0) {
             perror("unmap fails");
             exit(-1);
         }
 
         entry->paddr_index = 0;
         entry->vaddr_base = NULL;
+        entry->size = 0;
+        qemu_free(entry->valid_mapping);
+        entry->valid_mapping = NULL;
     }
 
     mapcache->last_address_index = -1;
@@ -342,34 +362,3 @@ void qemu_invalidate_map_cache(void)
 
     mapcache_unlock();
 }
-
-uint8_t *xen_map_block(target_phys_addr_t phys_addr, target_phys_addr_t size)
-{
-    uint8_t *vaddr_base;
-    xen_pfn_t *pfns;
-    int *err;
-    unsigned int i;
-    target_phys_addr_t nb_pfn = size >> XC_PAGE_SHIFT;
-
-    trace_xen_map_block(phys_addr, size);
-    phys_addr >>= XC_PAGE_SHIFT;
-
-    pfns = qemu_mallocz(nb_pfn * sizeof (xen_pfn_t));
-    err = qemu_mallocz(nb_pfn * sizeof (int));
-
-    for (i = 0; i < nb_pfn; i++) {
-        pfns[i] = phys_addr + i;
-    }
-
-    vaddr_base = xc_map_foreign_bulk(xen_xc, xen_domid, PROT_READ|PROT_WRITE,
-                                     pfns, err, nb_pfn);
-    if (vaddr_base == NULL) {
-        perror("xc_map_foreign_bulk");
-        exit(-1);
-    }
-
-    qemu_free(pfns);
-    qemu_free(err);
-
-    return vaddr_base;
-}
diff --git a/xen-mapcache.h b/xen-mapcache.h
index 339444c..6216cc3 100644
--- a/xen-mapcache.h
+++ b/xen-mapcache.h
@@ -9,28 +9,12 @@
 #ifndef XEN_MAPCACHE_H
 #define XEN_MAPCACHE_H
 
-#include <sys/mman.h>
-#include "trace.h"
-
 void     qemu_map_cache_init(void);
 uint8_t  *qemu_map_cache(target_phys_addr_t phys_addr, target_phys_addr_t size, uint8_t lock);
-void     qemu_map_cache_unlock(void *phys_addr);
 ram_addr_t qemu_ram_addr_from_mapcache(void *ptr);
 void     qemu_invalidate_entry(uint8_t *buffer);
 void     qemu_invalidate_map_cache(void);
 
-uint8_t *xen_map_block(target_phys_addr_t phys_addr, target_phys_addr_t size);
-
-static inline void xen_unmap_block(void *addr, ram_addr_t size)
-{
-    trace_xen_unmap_block(addr, size);
-
-    if (munmap(addr, size) != 0) {
-        hw_error("xen_unmap_block: %s", strerror(errno));
-    }
-}
-
-
 #define mapcache_lock()   ((void)0)
 #define mapcache_unlock() ((void)0)