summary refs log tree commit diff stats
path: root/hw
diff options
context:
space:
mode:
Diffstat (limited to 'hw')
-rw-r--r--hw/acpi/ich9.c14
-rw-r--r--hw/acpi/piix4.c10
-rw-r--r--hw/alpha/dp264.c7
-rw-r--r--hw/alpha/typhoon.c8
-rw-r--r--hw/arm/Makefile.objs1
-rw-r--r--hw/arm/boot.c34
-rw-r--r--hw/arm/nseries.c5
-rw-r--r--hw/arm/omap_sx1.c2
-rw-r--r--hw/arm/pxa2xx.c248
-rw-r--r--hw/arm/pxa2xx_pic.c2
-rw-r--r--hw/arm/sysbus-fdt.c174
-rw-r--r--hw/arm/vexpress.c2
-rw-r--r--hw/arm/virt-acpi-build.c43
-rw-r--r--hw/arm/virt.c166
-rw-r--r--hw/block/fdc.c332
-rw-r--r--hw/block/pflash_cfi01.c204
-rw-r--r--hw/char/parallel.c25
-rw-r--r--hw/char/serial.c41
-rw-r--r--hw/core/machine.c9
-rw-r--r--hw/core/nmi.c20
-rw-r--r--hw/core/qdev-properties-system.c5
-rw-r--r--hw/display/Makefile.objs4
-rw-r--r--hw/display/cg3.c2
-rw-r--r--hw/display/exynos4210_fimd.c22
-rw-r--r--hw/display/framebuffer.c4
-rw-r--r--hw/display/g364fb.c3
-rw-r--r--hw/display/qxl.c11
-rw-r--r--hw/display/sm501.c2
-rw-r--r--hw/display/tc6393xb.c2
-rw-r--r--hw/display/tcx.c3
-rw-r--r--hw/display/vga-pci.c97
-rw-r--r--hw/display/vga.c11
-rw-r--r--hw/display/vga_int.h6
-rw-r--r--hw/display/virtio-gpu-pci.c68
-rw-r--r--hw/display/virtio-gpu.c918
-rw-r--r--hw/display/virtio-vga.c175
-rw-r--r--hw/display/vmware_vga.c2
-rw-r--r--hw/dma/rc4030.c462
-rw-r--r--hw/gpio/pl061.c2
-rw-r--r--hw/i386/acpi-build.c1
-rw-r--r--hw/i386/pc.c36
-rw-r--r--hw/i386/pc_piix.c13
-rw-r--r--hw/i386/pc_q35.c7
-rw-r--r--hw/ide/core.c32
-rw-r--r--hw/ide/macio.c271
-rw-r--r--hw/ide/pci.c21
-rw-r--r--hw/input/pckbd.c22
-rw-r--r--hw/input/ps2.c11
-rw-r--r--hw/intc/Makefile.objs1
-rw-r--r--hw/intc/apic.c9
-rw-r--r--hw/intc/apic_common.c10
-rw-r--r--hw/intc/arm_gic.c2
-rw-r--r--hw/intc/arm_gicv2m.c192
-rw-r--r--hw/intc/exynos4210_gic.c7
-rw-r--r--hw/isa/i82378.c7
-rw-r--r--hw/isa/isa-bus.c29
-rw-r--r--hw/isa/lpc_ich9.c34
-rw-r--r--hw/lm32/lm32_boards.c10
-rw-r--r--hw/lm32/milkymist.c5
-rw-r--r--hw/mips/Makefile.objs3
-rw-r--r--hw/mips/mips_jazz.c53
-rw-r--r--hw/mips/mips_malta.c15
-rw-r--r--hw/misc/macio/macio.c71
-rw-r--r--hw/net/Makefile.objs1
-rw-r--r--hw/net/cadence_gem.c2
-rw-r--r--hw/net/dp8393x.c369
-rw-r--r--hw/net/e1000.c11
-rw-r--r--hw/net/pcnet.c8
-rw-r--r--hw/net/rocker/qmp-norocker.c50
-rw-r--r--hw/net/rocker/rocker.c68
-rw-r--r--hw/net/rocker/rocker_fp.c29
-rw-r--r--hw/net/rocker/rocker_fp.h2
-rw-r--r--hw/net/rocker/rocker_hw.h1
-rw-r--r--hw/net/rocker/rocker_of_dpa.c312
-rw-r--r--hw/net/rtl8139.c11
-rw-r--r--hw/net/vmxnet3.c12
-rw-r--r--hw/nvram/fw_cfg.c55
-rw-r--r--hw/pci-host/pam.c20
-rw-r--r--hw/pci-host/piix.c54
-rw-r--r--hw/pci-host/q35.c142
-rw-r--r--hw/pci/msi.c4
-rw-r--r--hw/pci/pci-stub.c14
-rw-r--r--hw/pci/pci.c4
-rw-r--r--hw/pci/pcie_aer.c39
-rw-r--r--hw/ppc/Makefile.objs2
-rw-r--r--hw/ppc/e500.c1
-rw-r--r--hw/ppc/mac_newworld.c2
-rw-r--r--hw/ppc/mac_oldworld.c2
-rw-r--r--hw/ppc/prep.c5
-rw-r--r--hw/ppc/spapr.c49
-rw-r--r--hw/ppc/spapr_drc.c744
-rw-r--r--hw/ppc/spapr_events.c338
-rw-r--r--hw/ppc/spapr_iommu.c46
-rw-r--r--hw/ppc/spapr_pci.c513
-rw-r--r--hw/ppc/spapr_rtas.c361
-rw-r--r--hw/ppc/spapr_vio.c2
-rw-r--r--hw/s390x/s390-virtio-ccw.c1
-rw-r--r--hw/s390x/s390-virtio.c1
-rw-r--r--hw/s390x/virtio-ccw.c64
-rw-r--r--hw/s390x/virtio-ccw.h18
-rw-r--r--hw/scsi/scsi-bus.c11
-rw-r--r--hw/sd/pxa2xx_mmci.c68
-rw-r--r--hw/sh4/r2d.c12
-rw-r--r--hw/sparc/sun4m.c8
-rw-r--r--hw/sparc64/sun4u.c2
-rw-r--r--hw/timer/arm_timer.c6
-rw-r--r--hw/timer/hpet.c11
-rw-r--r--hw/timer/mc146818rtc.c23
-rw-r--r--hw/unicore32/puv3.c8
-rw-r--r--hw/usb/hcd-ohci.c11
-rw-r--r--hw/usb/redirect.c34
-rw-r--r--hw/vfio/Makefile.objs2
-rw-r--r--hw/vfio/calxeda-xgmac.c55
-rw-r--r--hw/vfio/platform.c615
-rw-r--r--hw/virtio/dataplane/vring.c2
-rw-r--r--hw/virtio/vhost.c9
-rw-r--r--hw/virtio/virtio-pci.h14
-rw-r--r--hw/virtio/virtio.c16
-rw-r--r--hw/watchdog/Makefile.objs1
-rw-r--r--hw/watchdog/watchdog.c10
-rw-r--r--hw/watchdog/wdt_diag288.c122
-rw-r--r--hw/xen/xen_backend.c4
-rw-r--r--hw/xen/xen_pt.c51
-rw-r--r--hw/xen/xen_pt.h7
-rw-r--r--hw/xen/xen_pt_config_init.c235
-rw-r--r--hw/xen/xen_pt_msi.c12
126 files changed, 6856 insertions, 1858 deletions
diff --git a/hw/acpi/ich9.c b/hw/acpi/ich9.c
index 799351ea44..8a64ffb38f 100644
--- a/hw/acpi/ich9.c
+++ b/hw/acpi/ich9.c
@@ -94,7 +94,8 @@ static void ich9_smi_writel(void *opaque, hwaddr addr, uint64_t val,
     ICH9LPCPMRegs *pm = opaque;
     switch (addr) {
     case 0:
-        pm->smi_en = val;
+        pm->smi_en &= ~pm->smi_en_wmask;
+        pm->smi_en |= (val & pm->smi_en_wmask);
         break;
     }
 }
@@ -151,6 +152,7 @@ static const VMStateDescription vmstate_memhp_state = {
     .version_id = 1,
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
+    .needed = vmstate_test_use_memhp,
     .fields      = (VMStateField[]) {
         VMSTATE_MEMORY_HOTPLUG(acpi_memory_hotplug, ICH9LPCPMRegs),
         VMSTATE_END_OF_LIST()
@@ -174,12 +176,9 @@ const VMStateDescription vmstate_ich9_pm = {
         VMSTATE_UINT32(smi_sts, ICH9LPCPMRegs),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_memhp_state,
-            .needed = vmstate_test_use_memhp,
-        },
-        VMSTATE_END_OF_LIST()
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_memhp_state,
+        NULL
     }
 };
 
@@ -198,6 +197,7 @@ static void pm_reset(void *opaque)
          * support SMM mode. */
         pm->smi_en |= ICH9_PMIO_SMI_EN_APMC_EN;
     }
+    pm->smi_en_wmask = ~0;
 
     acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index b730ca6ced..3bd1d5a865 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -260,6 +260,7 @@ static const VMStateDescription vmstate_memhp_state = {
     .version_id = 1,
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
+    .needed = vmstate_test_use_memhp,
     .fields      = (VMStateField[]) {
         VMSTATE_MEMORY_HOTPLUG(acpi_memory_hotplug, PIIX4PMState),
         VMSTATE_END_OF_LIST()
@@ -298,12 +299,9 @@ static const VMStateDescription vmstate_acpi = {
                             vmstate_test_use_acpi_pci_hotplug),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_memhp_state,
-            .needed = vmstate_test_use_memhp,
-        },
-        VMSTATE_END_OF_LIST()
+    .subsections = (const VMStateDescription*[]) {
+         &vmstate_memhp_state,
+         NULL
     }
 };
 
diff --git a/hw/alpha/dp264.c b/hw/alpha/dp264.c
index 9fe7e8b5cb..f86e7bb830 100644
--- a/hw/alpha/dp264.c
+++ b/hw/alpha/dp264.c
@@ -55,7 +55,7 @@ static void clipper_init(MachineState *machine)
     ISABus *isa_bus;
     qemu_irq rtc_irq;
     long size, i;
-    const char *palcode_filename;
+    char *palcode_filename;
     uint64_t palcode_entry, palcode_low, palcode_high;
     uint64_t kernel_entry, kernel_low, kernel_high;
 
@@ -101,8 +101,8 @@ static void clipper_init(MachineState *machine)
     /* Load PALcode.  Given that this is not "real" cpu palcode,
        but one explicitly written for the emulation, we might as
        well load it directly from and ELF image.  */
-    palcode_filename = (bios_name ? bios_name : "palcode-clipper");
-    palcode_filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, palcode_filename);
+    palcode_filename = qemu_find_file(QEMU_FILE_TYPE_BIOS,
+                                bios_name ? bios_name : "palcode-clipper");
     if (palcode_filename == NULL) {
         hw_error("no palcode provided\n");
         exit(1);
@@ -114,6 +114,7 @@ static void clipper_init(MachineState *machine)
         hw_error("could not load palcode '%s'\n", palcode_filename);
         exit(1);
     }
+    g_free(palcode_filename);
 
     /* Start all cpus at the PALcode RESET entry point.  */
     for (i = 0; i < smp_cpus; ++i) {
diff --git a/hw/alpha/typhoon.c b/hw/alpha/typhoon.c
index 7df842dff7..421162e1d4 100644
--- a/hw/alpha/typhoon.c
+++ b/hw/alpha/typhoon.c
@@ -841,7 +841,7 @@ PCIBus *typhoon_init(ram_addr_t ram_size, ISABus **isa_bus,
         }
     }
 
-    *p_rtc_irq = *qemu_allocate_irqs(typhoon_set_timer_irq, s, 1);
+    *p_rtc_irq = qemu_allocate_irq(typhoon_set_timer_irq, s, 0);
 
     /* Main memory region, 0x00.0000.0000.  Real hardware supports 32GB,
        but the address space hole reserved at this point is 8TB.  */
@@ -918,11 +918,11 @@ PCIBus *typhoon_init(ram_addr_t ram_size, ISABus **isa_bus,
     /* Init the ISA bus.  */
     /* ??? Technically there should be a cy82c693ub pci-isa bridge.  */
     {
-        qemu_irq isa_pci_irq, *isa_irqs;
+        qemu_irq *isa_irqs;
 
         *isa_bus = isa_bus_new(NULL, get_system_memory(), &s->pchip.reg_io);
-        isa_pci_irq = *qemu_allocate_irqs(typhoon_set_isa_irq, s, 1);
-        isa_irqs = i8259_init(*isa_bus, isa_pci_irq);
+        isa_irqs = i8259_init(*isa_bus,
+                              qemu_allocate_irq(typhoon_set_isa_irq, s, 0));
         isa_bus_irqs(*isa_bus, isa_irqs);
     }
 
diff --git a/hw/arm/Makefile.objs b/hw/arm/Makefile.objs
index 4b09caf594..cf346c1d0a 100644
--- a/hw/arm/Makefile.objs
+++ b/hw/arm/Makefile.objs
@@ -5,6 +5,7 @@ obj-y += omap_sx1.o palm.o realview.o spitz.o stellaris.o
 obj-y += tosa.o versatilepb.o vexpress.o virt.o xilinx_zynq.o z2.o
 obj-$(CONFIG_ACPI) += virt-acpi-build.o
 obj-y += netduino2.o
+obj-y += sysbus-fdt.o
 
 obj-y += armv7m.o exynos4210.o pxa2xx.o pxa2xx_gpio.o pxa2xx_pic.o
 obj-$(CONFIG_DIGIC) += digic.o
diff --git a/hw/arm/boot.c b/hw/arm/boot.c
index fa6950352c..1e7fd28daa 100644
--- a/hw/arm/boot.c
+++ b/hw/arm/boot.c
@@ -557,7 +557,7 @@ static void load_image_to_fw_cfg(FWCfgState *fw_cfg, uint16_t size_key,
     fw_cfg_add_bytes(fw_cfg, data_key, data, size);
 }
 
-void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
+static void arm_load_kernel_notify(Notifier *notifier, void *data)
 {
     CPUState *cs;
     int kernel_size;
@@ -568,15 +568,11 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
     hwaddr entry, kernel_load_offset;
     int big_endian;
     static const ARMInsnFixup *primary_loader;
-
-    /* CPU objects (unlike devices) are not automatically reset on system
-     * reset, so we must always register a handler to do so. If we're
-     * actually loading a kernel, the handler is also responsible for
-     * arranging that we start it correctly.
-     */
-    for (cs = CPU(cpu); cs; cs = CPU_NEXT(cs)) {
-        qemu_register_reset(do_cpu_reset, ARM_CPU(cs));
-    }
+    ArmLoadKernelNotifier *n = DO_UPCAST(ArmLoadKernelNotifier,
+                                         notifier, notifier);
+    ARMCPU *cpu = n->cpu;
+    struct arm_boot_info *info =
+        container_of(n, struct arm_boot_info, load_kernel_notifier);
 
     /* Load the kernel.  */
     if (!info->kernel_filename || info->firmware_loaded) {
@@ -775,3 +771,21 @@ void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
         ARM_CPU(cs)->env.boot_info = info;
     }
 }
+
+void arm_load_kernel(ARMCPU *cpu, struct arm_boot_info *info)
+{
+    CPUState *cs;
+
+    info->load_kernel_notifier.cpu = cpu;
+    info->load_kernel_notifier.notifier.notify = arm_load_kernel_notify;
+    qemu_add_machine_init_done_notifier(&info->load_kernel_notifier.notifier);
+
+    /* CPU objects (unlike devices) are not automatically reset on system
+     * reset, so we must always register a handler to do so. If we're
+     * actually loading a kernel, the handler is also responsible for
+     * arranging that we start it correctly.
+     */
+    for (cs = CPU(cpu); cs; cs = CPU_NEXT(cs)) {
+        qemu_register_reset(do_cpu_reset, ARM_CPU(cs));
+    }
+}
diff --git a/hw/arm/nseries.c b/hw/arm/nseries.c
index d243159664..a659e8525d 100644
--- a/hw/arm/nseries.c
+++ b/hw/arm/nseries.c
@@ -133,9 +133,8 @@ static void n800_mmc_cs_cb(void *opaque, int line, int level)
 
 static void n8x0_gpio_setup(struct n800_s *s)
 {
-    qemu_irq *mmc_cs = qemu_allocate_irqs(n800_mmc_cs_cb, s->mpu->mmc, 1);
-    qdev_connect_gpio_out(s->mpu->gpio, N8X0_MMC_CS_GPIO, mmc_cs[0]);
-
+    qdev_connect_gpio_out(s->mpu->gpio, N8X0_MMC_CS_GPIO,
+                          qemu_allocate_irq(n800_mmc_cs_cb, s->mpu->mmc, 0));
     qemu_irq_lower(qdev_get_gpio_in(s->mpu->gpio, N800_BAT_COVER_GPIO));
 }
 
diff --git a/hw/arm/omap_sx1.c b/hw/arm/omap_sx1.c
index 671e02c4ed..4b0f7f9c42 100644
--- a/hw/arm/omap_sx1.c
+++ b/hw/arm/omap_sx1.c
@@ -103,7 +103,6 @@ static void sx1_init(MachineState *machine, const int version)
     struct omap_mpu_state_s *mpu;
     MemoryRegion *address_space = get_system_memory();
     MemoryRegion *flash = g_new(MemoryRegion, 1);
-    MemoryRegion *flash_1 = g_new(MemoryRegion, 1);
     MemoryRegion *cs = g_new(MemoryRegion, 4);
     static uint32_t cs0val = 0x00213090;
     static uint32_t cs1val = 0x00215070;
@@ -165,6 +164,7 @@ static void sx1_init(MachineState *machine, const int version)
 
     if ((version == 1) &&
             (dinfo = drive_get(IF_PFLASH, 0, fl_idx)) != NULL) {
+        MemoryRegion *flash_1 = g_new(MemoryRegion, 1);
         memory_region_init_ram(flash_1, NULL, "omap_sx1.flash1-0", flash1_size,
                                &error_abort);
         vmstate_register_ram_global(flash_1);
diff --git a/hw/arm/pxa2xx.c b/hw/arm/pxa2xx.c
index f921a5680c..ec353f79c4 100644
--- a/hw/arm/pxa2xx.c
+++ b/hw/arm/pxa2xx.c
@@ -334,10 +334,10 @@ static uint64_t pxa2xx_cpccnt_read(CPUARMState *env, const ARMCPRegInfo *ri)
 static const ARMCPRegInfo pxa_cp_reginfo[] = {
     /* cp14 crm==1: perf registers */
     { .name = "CPPMNC", .cp = 14, .crn = 0, .crm = 1, .opc1 = 0, .opc2 = 0,
-      .access = PL1_RW,
+      .access = PL1_RW, .type = ARM_CP_IO,
       .readfn = pxa2xx_cppmnc_read, .writefn = pxa2xx_cppmnc_write },
     { .name = "CPCCNT", .cp = 14, .crn = 1, .crm = 1, .opc1 = 0, .opc2 = 0,
-      .access = PL1_RW,
+      .access = PL1_RW, .type = ARM_CP_IO,
       .readfn = pxa2xx_cpccnt_read, .writefn = arm_cp_write_ignore },
     { .name = "CPINTEN", .cp = 14, .crn = 4, .crm = 1, .opc1 = 0, .opc2 = 0,
       .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
@@ -356,11 +356,11 @@ static const ARMCPRegInfo pxa_cp_reginfo[] = {
       .access = PL1_RW, .type = ARM_CP_CONST, .resetvalue = 0 },
     /* cp14 crn==6: CLKCFG */
     { .name = "CLKCFG", .cp = 14, .crn = 6, .crm = 0, .opc1 = 0, .opc2 = 0,
-      .access = PL1_RW,
+      .access = PL1_RW, .type = ARM_CP_IO,
       .readfn = pxa2xx_clkcfg_read, .writefn = pxa2xx_clkcfg_write },
     /* cp14 crn==7: PWRMODE */
     { .name = "PWRMODE", .cp = 14, .crn = 7, .crm = 0, .opc1 = 0, .opc2 = 0,
-      .access = PL1_RW,
+      .access = PL1_RW, .type = ARM_CP_IO,
       .readfn = arm_cp_read_zero, .writefn = pxa2xx_pwrmode_write },
     REGINFO_SENTINEL
 };
@@ -457,7 +457,7 @@ typedef struct {
 
     MemoryRegion iomem;
     qemu_irq irq;
-    int enable;
+    uint32_t enable;
     SSIBus *bus;
 
     uint32_t sscr[2];
@@ -470,10 +470,39 @@ typedef struct {
     uint8_t ssacd;
 
     uint32_t rx_fifo[16];
-    int rx_level;
-    int rx_start;
+    uint32_t rx_level;
+    uint32_t rx_start;
 } PXA2xxSSPState;
 
+static bool pxa2xx_ssp_vmstate_validate(void *opaque, int version_id)
+{
+    PXA2xxSSPState *s = opaque;
+
+    return s->rx_start < sizeof(s->rx_fifo);
+}
+
+static const VMStateDescription vmstate_pxa2xx_ssp = {
+    .name = "pxa2xx-ssp",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT32(enable, PXA2xxSSPState),
+        VMSTATE_UINT32_ARRAY(sscr, PXA2xxSSPState, 2),
+        VMSTATE_UINT32(sspsp, PXA2xxSSPState),
+        VMSTATE_UINT32(ssto, PXA2xxSSPState),
+        VMSTATE_UINT32(ssitr, PXA2xxSSPState),
+        VMSTATE_UINT32(sssr, PXA2xxSSPState),
+        VMSTATE_UINT8(sstsa, PXA2xxSSPState),
+        VMSTATE_UINT8(ssrsa, PXA2xxSSPState),
+        VMSTATE_UINT8(ssacd, PXA2xxSSPState),
+        VMSTATE_UINT32(rx_level, PXA2xxSSPState),
+        VMSTATE_UINT32(rx_start, PXA2xxSSPState),
+        VMSTATE_VALIDATE("fifo is 16 bytes", pxa2xx_ssp_vmstate_validate),
+        VMSTATE_UINT32_ARRAY(rx_fifo, PXA2xxSSPState, 16),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
 #define SSCR0	0x00	/* SSP Control register 0 */
 #define SSCR1	0x04	/* SSP Control register 1 */
 #define SSSR	0x08	/* SSP Status register */
@@ -705,55 +734,20 @@ static const MemoryRegionOps pxa2xx_ssp_ops = {
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
-static void pxa2xx_ssp_save(QEMUFile *f, void *opaque)
+static void pxa2xx_ssp_reset(DeviceState *d)
 {
-    PXA2xxSSPState *s = (PXA2xxSSPState *) opaque;
-    int i;
-
-    qemu_put_be32(f, s->enable);
+    PXA2xxSSPState *s = PXA2XX_SSP(d);
 
-    qemu_put_be32s(f, &s->sscr[0]);
-    qemu_put_be32s(f, &s->sscr[1]);
-    qemu_put_be32s(f, &s->sspsp);
-    qemu_put_be32s(f, &s->ssto);
-    qemu_put_be32s(f, &s->ssitr);
-    qemu_put_be32s(f, &s->sssr);
-    qemu_put_8s(f, &s->sstsa);
-    qemu_put_8s(f, &s->ssrsa);
-    qemu_put_8s(f, &s->ssacd);
-
-    qemu_put_byte(f, s->rx_level);
-    for (i = 0; i < s->rx_level; i ++)
-        qemu_put_byte(f, s->rx_fifo[(s->rx_start + i) & 0xf]);
-}
-
-static int pxa2xx_ssp_load(QEMUFile *f, void *opaque, int version_id)
-{
-    PXA2xxSSPState *s = (PXA2xxSSPState *) opaque;
-    int i, v;
-
-    s->enable = qemu_get_be32(f);
-
-    qemu_get_be32s(f, &s->sscr[0]);
-    qemu_get_be32s(f, &s->sscr[1]);
-    qemu_get_be32s(f, &s->sspsp);
-    qemu_get_be32s(f, &s->ssto);
-    qemu_get_be32s(f, &s->ssitr);
-    qemu_get_be32s(f, &s->sssr);
-    qemu_get_8s(f, &s->sstsa);
-    qemu_get_8s(f, &s->ssrsa);
-    qemu_get_8s(f, &s->ssacd);
-
-    v = qemu_get_byte(f);
-    if (v < 0 || v > ARRAY_SIZE(s->rx_fifo)) {
-        return -EINVAL;
-    }
-    s->rx_level = v;
-    s->rx_start = 0;
-    for (i = 0; i < s->rx_level; i ++)
-        s->rx_fifo[i] = qemu_get_byte(f);
-
-    return 0;
+    s->enable = 0;
+    s->sscr[0] = s->sscr[1] = 0;
+    s->sspsp = 0;
+    s->ssto = 0;
+    s->ssitr = 0;
+    s->sssr = 0;
+    s->sstsa = 0;
+    s->ssrsa = 0;
+    s->ssacd = 0;
+    s->rx_start = s->rx_level = 0;
 }
 
 static int pxa2xx_ssp_init(SysBusDevice *sbd)
@@ -766,8 +760,6 @@ static int pxa2xx_ssp_init(SysBusDevice *sbd)
     memory_region_init_io(&s->iomem, OBJECT(s), &pxa2xx_ssp_ops, s,
                           "pxa2xx-ssp", 0x1000);
     sysbus_init_mmio(sbd, &s->iomem);
-    register_savevm(dev, "pxa2xx_ssp", -1, 0,
-                    pxa2xx_ssp_save, pxa2xx_ssp_load, s);
 
     s->bus = ssi_create_bus(dev, "ssi");
     return 0;
@@ -1759,24 +1751,33 @@ static PXA2xxI2SState *pxa2xx_i2s_init(MemoryRegion *sysmem,
 }
 
 /* PXA Fast Infra-red Communications Port */
+#define TYPE_PXA2XX_FIR "pxa2xx-fir"
+#define PXA2XX_FIR(obj) OBJECT_CHECK(PXA2xxFIrState, (obj), TYPE_PXA2XX_FIR)
+
 struct PXA2xxFIrState {
+    /*< private >*/
+    SysBusDevice parent_obj;
+    /*< public >*/
+
     MemoryRegion iomem;
     qemu_irq irq;
     qemu_irq rx_dma;
     qemu_irq tx_dma;
-    int enable;
+    uint32_t enable;
     CharDriverState *chr;
 
     uint8_t control[3];
     uint8_t status[2];
 
-    int rx_len;
-    int rx_start;
+    uint32_t rx_len;
+    uint32_t rx_start;
     uint8_t rx_fifo[64];
 };
 
-static void pxa2xx_fir_reset(PXA2xxFIrState *s)
+static void pxa2xx_fir_reset(DeviceState *d)
 {
+    PXA2xxFIrState *s = PXA2XX_FIR(d);
+
     s->control[0] = 0x00;
     s->control[1] = 0x00;
     s->control[2] = 0x00;
@@ -1953,73 +1954,94 @@ static void pxa2xx_fir_event(void *opaque, int event)
 {
 }
 
-static void pxa2xx_fir_save(QEMUFile *f, void *opaque)
+static void pxa2xx_fir_instance_init(Object *obj)
 {
-    PXA2xxFIrState *s = (PXA2xxFIrState *) opaque;
-    int i;
+    PXA2xxFIrState *s = PXA2XX_FIR(obj);
+    SysBusDevice *sbd = SYS_BUS_DEVICE(obj);
 
-    qemu_put_be32(f, s->enable);
-
-    qemu_put_8s(f, &s->control[0]);
-    qemu_put_8s(f, &s->control[1]);
-    qemu_put_8s(f, &s->control[2]);
-    qemu_put_8s(f, &s->status[0]);
-    qemu_put_8s(f, &s->status[1]);
-
-    qemu_put_byte(f, s->rx_len);
-    for (i = 0; i < s->rx_len; i ++)
-        qemu_put_byte(f, s->rx_fifo[(s->rx_start + i) & 63]);
+    memory_region_init_io(&s->iomem, NULL, &pxa2xx_fir_ops, s,
+                          "pxa2xx-fir", 0x1000);
+    sysbus_init_mmio(sbd, &s->iomem);
+    sysbus_init_irq(sbd, &s->irq);
+    sysbus_init_irq(sbd, &s->rx_dma);
+    sysbus_init_irq(sbd, &s->tx_dma);
 }
 
-static int pxa2xx_fir_load(QEMUFile *f, void *opaque, int version_id)
+static void pxa2xx_fir_realize(DeviceState *dev, Error **errp)
 {
-    PXA2xxFIrState *s = (PXA2xxFIrState *) opaque;
-    int i;
-
-    s->enable = qemu_get_be32(f);
+    PXA2xxFIrState *s = PXA2XX_FIR(dev);
 
-    qemu_get_8s(f, &s->control[0]);
-    qemu_get_8s(f, &s->control[1]);
-    qemu_get_8s(f, &s->control[2]);
-    qemu_get_8s(f, &s->status[0]);
-    qemu_get_8s(f, &s->status[1]);
+    if (s->chr) {
+        qemu_chr_fe_claim_no_fail(s->chr);
+        qemu_chr_add_handlers(s->chr, pxa2xx_fir_is_empty,
+                        pxa2xx_fir_rx, pxa2xx_fir_event, s);
+    }
+}
 
-    s->rx_len = qemu_get_byte(f);
-    s->rx_start = 0;
-    for (i = 0; i < s->rx_len; i ++)
-        s->rx_fifo[i] = qemu_get_byte(f);
+static bool pxa2xx_fir_vmstate_validate(void *opaque, int version_id)
+{
+    PXA2xxFIrState *s = opaque;
 
-    return 0;
+    return s->rx_start < ARRAY_SIZE(s->rx_fifo);
 }
 
-static PXA2xxFIrState *pxa2xx_fir_init(MemoryRegion *sysmem,
-                hwaddr base,
-                qemu_irq irq, qemu_irq rx_dma, qemu_irq tx_dma,
-                CharDriverState *chr)
-{
-    PXA2xxFIrState *s = (PXA2xxFIrState *)
-            g_malloc0(sizeof(PXA2xxFIrState));
+static const VMStateDescription pxa2xx_fir_vmsd = {
+    .name = "pxa2xx-fir",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT32(enable, PXA2xxFIrState),
+        VMSTATE_UINT8_ARRAY(control, PXA2xxFIrState, 3),
+        VMSTATE_UINT8_ARRAY(status, PXA2xxFIrState, 2),
+        VMSTATE_UINT32(rx_len, PXA2xxFIrState),
+        VMSTATE_UINT32(rx_start, PXA2xxFIrState),
+        VMSTATE_VALIDATE("fifo is 64 bytes", pxa2xx_fir_vmstate_validate),
+        VMSTATE_UINT8_ARRAY(rx_fifo, PXA2xxFIrState, 64),
+        VMSTATE_END_OF_LIST()
+    }
+};
 
-    s->irq = irq;
-    s->rx_dma = rx_dma;
-    s->tx_dma = tx_dma;
-    s->chr = chr;
+static Property pxa2xx_fir_properties[] = {
+    DEFINE_PROP_CHR("chardev", PXA2xxFIrState, chr),
+    DEFINE_PROP_END_OF_LIST(),
+};
 
-    pxa2xx_fir_reset(s);
+static void pxa2xx_fir_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
 
-    memory_region_init_io(&s->iomem, NULL, &pxa2xx_fir_ops, s, "pxa2xx-fir", 0x1000);
-    memory_region_add_subregion(sysmem, base, &s->iomem);
+    dc->realize = pxa2xx_fir_realize;
+    dc->vmsd = &pxa2xx_fir_vmsd;
+    dc->props = pxa2xx_fir_properties;
+    dc->reset = pxa2xx_fir_reset;
+}
 
-    if (chr) {
-        qemu_chr_fe_claim_no_fail(chr);
-        qemu_chr_add_handlers(chr, pxa2xx_fir_is_empty,
-                        pxa2xx_fir_rx, pxa2xx_fir_event, s);
-    }
+static const TypeInfo pxa2xx_fir_info = {
+    .name = TYPE_PXA2XX_FIR,
+    .parent = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(PXA2xxFIrState),
+    .class_init = pxa2xx_fir_class_init,
+    .instance_init = pxa2xx_fir_instance_init,
+};
 
-    register_savevm(NULL, "pxa2xx_fir", 0, 0, pxa2xx_fir_save,
-                    pxa2xx_fir_load, s);
+static PXA2xxFIrState *pxa2xx_fir_init(MemoryRegion *sysmem,
+                                       hwaddr base,
+                                       qemu_irq irq, qemu_irq rx_dma,
+                                       qemu_irq tx_dma,
+                                       CharDriverState *chr)
+{
+    DeviceState *dev;
+    SysBusDevice *sbd;
 
-    return s;
+    dev = qdev_create(NULL, TYPE_PXA2XX_FIR);
+    qdev_prop_set_chr(dev, "chardev", chr);
+    qdev_init_nofail(dev);
+    sbd = SYS_BUS_DEVICE(dev);
+    sysbus_mmio_map(sbd, 0, base);
+    sysbus_connect_irq(sbd, 0, irq);
+    sysbus_connect_irq(sbd, 1, rx_dma);
+    sysbus_connect_irq(sbd, 2, tx_dma);
+    return PXA2XX_FIR(dev);
 }
 
 static void pxa2xx_reset(void *opaque, int line, int level)
@@ -2306,8 +2328,11 @@ PXA2xxState *pxa255_init(MemoryRegion *address_space, unsigned int sdram_size)
 static void pxa2xx_ssp_class_init(ObjectClass *klass, void *data)
 {
     SysBusDeviceClass *sdc = SYS_BUS_DEVICE_CLASS(klass);
+    DeviceClass *dc = DEVICE_CLASS(klass);
 
     sdc->init = pxa2xx_ssp_init;
+    dc->reset = pxa2xx_ssp_reset;
+    dc->vmsd = &vmstate_pxa2xx_ssp;
 }
 
 static const TypeInfo pxa2xx_ssp_info = {
@@ -2323,6 +2348,7 @@ static void pxa2xx_register_types(void)
     type_register_static(&pxa2xx_ssp_info);
     type_register_static(&pxa2xx_i2c_info);
     type_register_static(&pxa2xx_rtc_sysbus_info);
+    type_register_static(&pxa2xx_fir_info);
 }
 
 type_init(pxa2xx_register_types)
diff --git a/hw/arm/pxa2xx_pic.c b/hw/arm/pxa2xx_pic.c
index 9cfc714874..d41ac93416 100644
--- a/hw/arm/pxa2xx_pic.c
+++ b/hw/arm/pxa2xx_pic.c
@@ -232,7 +232,7 @@ static void pxa2xx_pic_cp_write(CPUARMState *env, const ARMCPRegInfo *ri,
 
 #define REGINFO_FOR_PIC_CP(NAME, CRN) \
     { .name = NAME, .cp = 6, .crn = CRN, .crm = 0, .opc1 = 0, .opc2 = 0, \
-      .access = PL1_RW, \
+      .access = PL1_RW, .type = ARM_CP_IO, \
       .readfn = pxa2xx_pic_cp_read, .writefn = pxa2xx_pic_cp_write }
 
 static const ARMCPRegInfo pxa_pic_cp_reginfo[] = {
diff --git a/hw/arm/sysbus-fdt.c b/hw/arm/sysbus-fdt.c
new file mode 100644
index 0000000000..3038b94b4a
--- /dev/null
+++ b/hw/arm/sysbus-fdt.c
@@ -0,0 +1,174 @@
+/*
+ * ARM Platform Bus device tree generation helpers
+ *
+ * Copyright (c) 2014 Linaro Limited
+ *
+ * Authors:
+ *  Alex Graf <agraf@suse.de>
+ *  Eric Auger <eric.auger@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+#include "hw/arm/sysbus-fdt.h"
+#include "qemu/error-report.h"
+#include "sysemu/device_tree.h"
+#include "hw/platform-bus.h"
+#include "sysemu/sysemu.h"
+
+/*
+ * internal struct that contains the information to create dynamic
+ * sysbus device node
+ */
+typedef struct PlatformBusFDTData {
+    void *fdt; /* device tree handle */
+    int irq_start; /* index of the first IRQ usable by platform bus devices */
+    const char *pbus_node_name; /* name of the platform bus node */
+    PlatformBusDevice *pbus;
+} PlatformBusFDTData;
+
+/*
+ * struct used when calling the machine init done notifier
+ * that constructs the fdt nodes of platform bus devices
+ */
+typedef struct PlatformBusFDTNotifierParams {
+    Notifier notifier;
+    ARMPlatformBusFDTParams *fdt_params;
+} PlatformBusFDTNotifierParams;
+
+/* struct that associates a device type name and a node creation function */
+typedef struct NodeCreationPair {
+    const char *typename;
+    int (*add_fdt_node_fn)(SysBusDevice *sbdev, void *opaque);
+} NodeCreationPair;
+
+/* list of supported dynamic sysbus devices */
+static const NodeCreationPair add_fdt_node_functions[] = {
+    {"", NULL}, /* last element */
+};
+
+/**
+ * add_fdt_node - add the device tree node of a dynamic sysbus device
+ *
+ * @sbdev: handle to the sysbus device
+ * @opaque: handle to the PlatformBusFDTData
+ *
+ * Checks the sysbus type belongs to the list of device types that
+ * are dynamically instantiable and if so call the node creation
+ * function.
+ */
+static int add_fdt_node(SysBusDevice *sbdev, void *opaque)
+{
+    int i, ret;
+
+    for (i = 0; i < ARRAY_SIZE(add_fdt_node_functions); i++) {
+        if (!strcmp(object_get_typename(OBJECT(sbdev)),
+                    add_fdt_node_functions[i].typename)) {
+            ret = add_fdt_node_functions[i].add_fdt_node_fn(sbdev, opaque);
+            assert(!ret);
+            return 0;
+        }
+    }
+    error_report("Device %s can not be dynamically instantiated",
+                     qdev_fw_name(DEVICE(sbdev)));
+    exit(1);
+}
+
+/**
+ * add_all_platform_bus_fdt_nodes - create all the platform bus nodes
+ *
+ * builds the parent platform bus node and all the nodes of dynamic
+ * sysbus devices attached to it.
+ */
+static void add_all_platform_bus_fdt_nodes(ARMPlatformBusFDTParams *fdt_params)
+{
+    const char platcomp[] = "qemu,platform\0simple-bus";
+    PlatformBusDevice *pbus;
+    DeviceState *dev;
+    gchar *node;
+    uint64_t addr, size;
+    int irq_start, dtb_size;
+    struct arm_boot_info *info = fdt_params->binfo;
+    const ARMPlatformBusSystemParams *params = fdt_params->system_params;
+    const char *intc = fdt_params->intc;
+    void *fdt = info->get_dtb(info, &dtb_size);
+
+    /*
+     * If the user provided a dtb, we assume the dynamic sysbus nodes
+     * already are integrated there. This corresponds to a use case where
+     * the dynamic sysbus nodes are complex and their generation is not yet
+     * supported. In that case the user can take charge of the guest dt
+     * while qemu takes charge of the qom stuff.
+     */
+    if (info->dtb_filename) {
+        return;
+    }
+
+    assert(fdt);
+
+    node = g_strdup_printf("/platform@%"PRIx64, params->platform_bus_base);
+    addr = params->platform_bus_base;
+    size = params->platform_bus_size;
+    irq_start = params->platform_bus_first_irq;
+
+    /* Create a /platform node that we can put all devices into */
+    qemu_fdt_add_subnode(fdt, node);
+    qemu_fdt_setprop(fdt, node, "compatible", platcomp, sizeof(platcomp));
+
+    /* Our platform bus region is less than 32bits, so 1 cell is enough for
+     * address and size
+     */
+    qemu_fdt_setprop_cells(fdt, node, "#size-cells", 1);
+    qemu_fdt_setprop_cells(fdt, node, "#address-cells", 1);
+    qemu_fdt_setprop_cells(fdt, node, "ranges", 0, addr >> 32, addr, size);
+
+    qemu_fdt_setprop_phandle(fdt, node, "interrupt-parent", intc);
+
+    dev = qdev_find_recursive(sysbus_get_default(), TYPE_PLATFORM_BUS_DEVICE);
+    pbus = PLATFORM_BUS_DEVICE(dev);
+
+    /* We can only create dt nodes for dynamic devices when they're ready */
+    assert(pbus->done_gathering);
+
+    PlatformBusFDTData data = {
+        .fdt = fdt,
+        .irq_start = irq_start,
+        .pbus_node_name = node,
+        .pbus = pbus,
+    };
+
+    /* Loop through all dynamic sysbus devices and create their node */
+    foreach_dynamic_sysbus_device(add_fdt_node, &data);
+
+    g_free(node);
+}
+
+static void platform_bus_fdt_notify(Notifier *notifier, void *data)
+{
+    PlatformBusFDTNotifierParams *p = DO_UPCAST(PlatformBusFDTNotifierParams,
+                                                notifier, notifier);
+
+    add_all_platform_bus_fdt_nodes(p->fdt_params);
+    g_free(p->fdt_params);
+    g_free(p);
+}
+
+void arm_register_platform_bus_fdt_creator(ARMPlatformBusFDTParams *fdt_params)
+{
+    PlatformBusFDTNotifierParams *p = g_new(PlatformBusFDTNotifierParams, 1);
+
+    p->fdt_params = fdt_params;
+    p->notifier.notify = platform_bus_fdt_notify;
+    qemu_add_machine_init_done_notifier(&p->notifier);
+}
diff --git a/hw/arm/vexpress.c b/hw/arm/vexpress.c
index 8f1a5ea992..da217884e6 100644
--- a/hw/arm/vexpress.c
+++ b/hw/arm/vexpress.c
@@ -525,7 +525,7 @@ static pflash_t *ve_pflash_cfi01_register(hwaddr base, const char *name,
     qdev_prop_set_uint64(dev, "sector-length", VEXPRESS_FLASH_SECT_SIZE);
     qdev_prop_set_uint8(dev, "width", 4);
     qdev_prop_set_uint8(dev, "device-width", 2);
-    qdev_prop_set_uint8(dev, "big-endian", 0);
+    qdev_prop_set_bit(dev, "big-endian", false);
     qdev_prop_set_uint16(dev, "id0", 0x89);
     qdev_prop_set_uint16(dev, "id1", 0x18);
     qdev_prop_set_uint16(dev, "id2", 0x00);
diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
index a9373ccaca..d5a8b9c017 100644
--- a/hw/arm/virt-acpi-build.c
+++ b/hw/arm/virt-acpi-build.c
@@ -84,6 +84,12 @@ static void acpi_dsdt_add_uart(Aml *scope, const MemMapEntry *uart_memmap,
                aml_interrupt(AML_CONSUMER, AML_LEVEL, AML_ACTIVE_HIGH,
                              AML_EXCLUSIVE, uart_irq));
     aml_append(dev, aml_name_decl("_CRS", crs));
+
+    /* The _ADR entry is used to link this device to the UART described
+     * in the SPCR table, i.e. SPCR.base_address.address == _ADR.
+     */
+    aml_append(dev, aml_name_decl("_ADR", aml_int(uart_memmap->base)));
+
     aml_append(scope, dev);
 }
 
@@ -334,6 +340,38 @@ build_rsdp(GArray *rsdp_table, GArray *linker, unsigned rsdt)
 }
 
 static void
+build_spcr(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info)
+{
+    AcpiSerialPortConsoleRedirection *spcr;
+    const MemMapEntry *uart_memmap = &guest_info->memmap[VIRT_UART];
+    int irq = guest_info->irqmap[VIRT_UART] + ARM_SPI_BASE;
+
+    spcr = acpi_data_push(table_data, sizeof(*spcr));
+
+    spcr->interface_type = 0x3;    /* ARM PL011 UART */
+
+    spcr->base_address.space_id = AML_SYSTEM_MEMORY;
+    spcr->base_address.bit_width = 8;
+    spcr->base_address.bit_offset = 0;
+    spcr->base_address.access_width = 1;
+    spcr->base_address.address = cpu_to_le64(uart_memmap->base);
+
+    spcr->interrupt_types = (1 << 3); /* Bit[3] ARMH GIC interrupt */
+    spcr->gsi = cpu_to_le32(irq);  /* Global System Interrupt */
+
+    spcr->baud = 3;                /* Baud Rate: 3 = 9600 */
+    spcr->parity = 0;              /* No Parity */
+    spcr->stopbits = 1;            /* 1 Stop bit */
+    spcr->flowctrl = (1 << 1);     /* Bit[1] = RTS/CTS hardware flow control */
+    spcr->term_type = 0;           /* Terminal Type: 0 = VT100 */
+
+    spcr->pci_device_id = 0xffff;  /* PCI Device ID: not a PCI device */
+    spcr->pci_vendor_id = 0xffff;  /* PCI Vendor ID: not a PCI device */
+
+    build_header(linker, table_data, (void *)spcr, "SPCR", sizeof(*spcr), 2);
+}
+
+static void
 build_mcfg(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info)
 {
     AcpiTableMcfg *mcfg;
@@ -514,7 +552,7 @@ void virt_acpi_build(VirtGuestInfo *guest_info, AcpiBuildTables *tables)
     dsdt = tables_blob->len;
     build_dsdt(tables_blob, tables->linker, guest_info);
 
-    /* FADT MADT GTDT pointed to by RSDT */
+    /* FADT MADT GTDT SPCR pointed to by RSDT */
     acpi_add_table(table_offsets, tables_blob);
     build_fadt(tables_blob, tables->linker, dsdt);
 
@@ -527,6 +565,9 @@ void virt_acpi_build(VirtGuestInfo *guest_info, AcpiBuildTables *tables)
     acpi_add_table(table_offsets, tables_blob);
     build_mcfg(tables_blob, tables->linker, guest_info);
 
+    acpi_add_table(table_offsets, tables_blob);
+    build_spcr(tables_blob, tables->linker, guest_info);
+
     /* RSDT is pointed to by RSDP */
     rsdt = tables_blob->len;
     build_rsdt(tables_blob, tables->linker, table_offsets);
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 05db8cb2f7..f1e85c8e92 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -45,9 +45,11 @@
 #include "qemu/error-report.h"
 #include "hw/pci-host/gpex.h"
 #include "hw/arm/virt-acpi-build.h"
+#include "hw/arm/sysbus-fdt.h"
+#include "hw/platform-bus.h"
 
 /* Number of external interrupt lines to configure the GIC with */
-#define NUM_IRQS 128
+#define NUM_IRQS 256
 
 #define GIC_FDT_IRQ_TYPE_SPI 0
 #define GIC_FDT_IRQ_TYPE_PPI 1
@@ -60,6 +62,10 @@
 #define GIC_FDT_IRQ_PPI_CPU_START 8
 #define GIC_FDT_IRQ_PPI_CPU_WIDTH 8
 
+#define PLATFORM_BUS_NUM_IRQS 64
+
+static ARMPlatformBusSystemParams platform_bus_params;
+
 typedef struct VirtBoardInfo {
     struct arm_boot_info bootinfo;
     const char *cpu_model;
@@ -69,6 +75,8 @@ typedef struct VirtBoardInfo {
     void *fdt;
     int fdt_size;
     uint32_t clock_phandle;
+    uint32_t gic_phandle;
+    uint32_t v2m_phandle;
 } VirtBoardInfo;
 
 typedef struct {
@@ -103,20 +111,22 @@ typedef struct {
  */
 static const MemMapEntry a15memmap[] = {
     /* Space up to 0x8000000 is reserved for a boot ROM */
-    [VIRT_FLASH] =      {          0, 0x08000000 },
-    [VIRT_CPUPERIPHS] = { 0x08000000, 0x00020000 },
+    [VIRT_FLASH] =              {          0, 0x08000000 },
+    [VIRT_CPUPERIPHS] =         { 0x08000000, 0x00020000 },
     /* GIC distributor and CPU interfaces sit inside the CPU peripheral space */
-    [VIRT_GIC_DIST] =   { 0x08000000, 0x00010000 },
-    [VIRT_GIC_CPU] =    { 0x08010000, 0x00010000 },
-    [VIRT_UART] =       { 0x09000000, 0x00001000 },
-    [VIRT_RTC] =        { 0x09010000, 0x00001000 },
-    [VIRT_FW_CFG] =     { 0x09020000, 0x0000000a },
-    [VIRT_MMIO] =       { 0x0a000000, 0x00000200 },
+    [VIRT_GIC_DIST] =           { 0x08000000, 0x00010000 },
+    [VIRT_GIC_CPU] =            { 0x08010000, 0x00010000 },
+    [VIRT_GIC_V2M] =            { 0x08020000, 0x00001000 },
+    [VIRT_UART] =               { 0x09000000, 0x00001000 },
+    [VIRT_RTC] =                { 0x09010000, 0x00001000 },
+    [VIRT_FW_CFG] =             { 0x09020000, 0x0000000a },
+    [VIRT_MMIO] =               { 0x0a000000, 0x00000200 },
     /* ...repeating for a total of NUM_VIRTIO_TRANSPORTS, each of that size */
-    [VIRT_PCIE_MMIO] =  { 0x10000000, 0x2eff0000 },
-    [VIRT_PCIE_PIO] =   { 0x3eff0000, 0x00010000 },
-    [VIRT_PCIE_ECAM] =  { 0x3f000000, 0x01000000 },
-    [VIRT_MEM] =        { 0x40000000, 30ULL * 1024 * 1024 * 1024 },
+    [VIRT_PLATFORM_BUS] =       { 0x0c000000, 0x02000000 },
+    [VIRT_PCIE_MMIO] =          { 0x10000000, 0x2eff0000 },
+    [VIRT_PCIE_PIO] =           { 0x3eff0000, 0x00010000 },
+    [VIRT_PCIE_ECAM] =          { 0x3f000000, 0x01000000 },
+    [VIRT_MEM] =                { 0x40000000, 30ULL * 1024 * 1024 * 1024 },
 };
 
 static const int a15irqmap[] = {
@@ -124,6 +134,8 @@ static const int a15irqmap[] = {
     [VIRT_RTC] = 2,
     [VIRT_PCIE] = 3, /* ... to 6 */
     [VIRT_MMIO] = 16, /* ...to 16 + NUM_VIRTIO_TRANSPORTS - 1 */
+    [VIRT_GIC_V2M] = 48, /* ...to 48 + NUM_GICV2M_SPIS - 1 */
+    [VIRT_PLATFORM_BUS] = 112, /* ...to 112 + PLATFORM_BUS_NUM_IRQS -1 */
 };
 
 static VirtBoardInfo machines[] = {
@@ -133,6 +145,11 @@ static VirtBoardInfo machines[] = {
         .irqmap = a15irqmap,
     },
     {
+        .cpu_model = "cortex-a53",
+        .memmap = a15memmap,
+        .irqmap = a15irqmap,
+    },
+    {
         .cpu_model = "cortex-a57",
         .memmap = a15memmap,
         .irqmap = a15irqmap,
@@ -294,17 +311,28 @@ static void fdt_add_cpu_nodes(const VirtBoardInfo *vbi)
                                         "enable-method", "psci");
         }
 
-        qemu_fdt_setprop_cell(vbi->fdt, nodename, "reg", cpu);
+        qemu_fdt_setprop_cell(vbi->fdt, nodename, "reg", armcpu->mp_affinity);
         g_free(nodename);
     }
 }
 
-static uint32_t fdt_add_gic_node(const VirtBoardInfo *vbi)
+static void fdt_add_v2m_gic_node(VirtBoardInfo *vbi)
 {
-    uint32_t gic_phandle;
+    vbi->v2m_phandle = qemu_fdt_alloc_phandle(vbi->fdt);
+    qemu_fdt_add_subnode(vbi->fdt, "/intc/v2m");
+    qemu_fdt_setprop_string(vbi->fdt, "/intc/v2m", "compatible",
+                            "arm,gic-v2m-frame");
+    qemu_fdt_setprop(vbi->fdt, "/intc/v2m", "msi-controller", NULL, 0);
+    qemu_fdt_setprop_sized_cells(vbi->fdt, "/intc/v2m", "reg",
+                                 2, vbi->memmap[VIRT_GIC_V2M].base,
+                                 2, vbi->memmap[VIRT_GIC_V2M].size);
+    qemu_fdt_setprop_cell(vbi->fdt, "/intc/v2m", "phandle", vbi->v2m_phandle);
+}
 
-    gic_phandle = qemu_fdt_alloc_phandle(vbi->fdt);
-    qemu_fdt_setprop_cell(vbi->fdt, "/", "interrupt-parent", gic_phandle);
+static void fdt_add_gic_node(VirtBoardInfo *vbi)
+{
+    vbi->gic_phandle = qemu_fdt_alloc_phandle(vbi->fdt);
+    qemu_fdt_setprop_cell(vbi->fdt, "/", "interrupt-parent", vbi->gic_phandle);
 
     qemu_fdt_add_subnode(vbi->fdt, "/intc");
     /* 'cortex-a15-gic' means 'GIC v2' */
@@ -317,12 +345,32 @@ static uint32_t fdt_add_gic_node(const VirtBoardInfo *vbi)
                                      2, vbi->memmap[VIRT_GIC_DIST].size,
                                      2, vbi->memmap[VIRT_GIC_CPU].base,
                                      2, vbi->memmap[VIRT_GIC_CPU].size);
-    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "phandle", gic_phandle);
+    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "#address-cells", 0x2);
+    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "#size-cells", 0x2);
+    qemu_fdt_setprop(vbi->fdt, "/intc", "ranges", NULL, 0);
+    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "phandle", vbi->gic_phandle);
+}
+
+static void create_v2m(VirtBoardInfo *vbi, qemu_irq *pic)
+{
+    int i;
+    int irq = vbi->irqmap[VIRT_GIC_V2M];
+    DeviceState *dev;
+
+    dev = qdev_create(NULL, "arm-gicv2m");
+    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, vbi->memmap[VIRT_GIC_V2M].base);
+    qdev_prop_set_uint32(dev, "base-spi", irq);
+    qdev_prop_set_uint32(dev, "num-spi", NUM_GICV2M_SPIS);
+    qdev_init_nofail(dev);
+
+    for (i = 0; i < NUM_GICV2M_SPIS; i++) {
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]);
+    }
 
-    return gic_phandle;
+    fdt_add_v2m_gic_node(vbi);
 }
 
-static uint32_t create_gic(const VirtBoardInfo *vbi, qemu_irq *pic)
+static void create_gic(VirtBoardInfo *vbi, qemu_irq *pic)
 {
     /* We create a standalone GIC v2 */
     DeviceState *gicdev;
@@ -371,7 +419,9 @@ static uint32_t create_gic(const VirtBoardInfo *vbi, qemu_irq *pic)
         pic[i] = qdev_get_gpio_in(gicdev, i);
     }
 
-    return fdt_add_gic_node(vbi);
+    fdt_add_gic_node(vbi);
+
+    create_v2m(vbi, pic);
 }
 
 static void create_uart(const VirtBoardInfo *vbi, qemu_irq *pic)
@@ -510,7 +560,7 @@ static void create_one_flash(const char *name, hwaddr flashbase,
     qdev_prop_set_uint64(dev, "sector-length", sectorlength);
     qdev_prop_set_uint8(dev, "width", 4);
     qdev_prop_set_uint8(dev, "device-width", 2);
-    qdev_prop_set_uint8(dev, "big-endian", 0);
+    qdev_prop_set_bit(dev, "big-endian", false);
     qdev_prop_set_uint16(dev, "id0", 0x89);
     qdev_prop_set_uint16(dev, "id1", 0x18);
     qdev_prop_set_uint16(dev, "id2", 0x00);
@@ -587,7 +637,7 @@ static void create_pcie_irq_map(const VirtBoardInfo *vbi, uint32_t gic_phandle,
                                 int first_irq, const char *nodename)
 {
     int devfn, pin;
-    uint32_t full_irq_map[4 * 4 * 8] = { 0 };
+    uint32_t full_irq_map[4 * 4 * 10] = { 0 };
     uint32_t *irq_map = full_irq_map;
 
     for (devfn = 0; devfn <= 0x18; devfn += 0x8) {
@@ -600,13 +650,13 @@ static void create_pcie_irq_map(const VirtBoardInfo *vbi, uint32_t gic_phandle,
             uint32_t map[] = {
                 devfn << 8, 0, 0,                           /* devfn */
                 pin + 1,                                    /* PCI pin */
-                gic_phandle, irq_type, irq_nr, irq_level }; /* GIC irq */
+                gic_phandle, 0, 0, irq_type, irq_nr, irq_level }; /* GIC irq */
 
             /* Convert map to big endian */
-            for (i = 0; i < 8; i++) {
+            for (i = 0; i < 10; i++) {
                 irq_map[i] = cpu_to_be32(map[i]);
             }
-            irq_map += 8;
+            irq_map += 10;
         }
     }
 
@@ -618,8 +668,7 @@ static void create_pcie_irq_map(const VirtBoardInfo *vbi, uint32_t gic_phandle,
                            0x7           /* PCI irq */);
 }
 
-static void create_pcie(const VirtBoardInfo *vbi, qemu_irq *pic,
-                        uint32_t gic_phandle)
+static void create_pcie(const VirtBoardInfo *vbi, qemu_irq *pic)
 {
     hwaddr base_mmio = vbi->memmap[VIRT_PCIE_MMIO].base;
     hwaddr size_mmio = vbi->memmap[VIRT_PCIE_MMIO].size;
@@ -676,6 +725,8 @@ static void create_pcie(const VirtBoardInfo *vbi, qemu_irq *pic,
     qemu_fdt_setprop_cells(vbi->fdt, nodename, "bus-range", 0,
                            nr_pcie_buses - 1);
 
+    qemu_fdt_setprop_cells(vbi->fdt, nodename, "msi-parent", vbi->v2m_phandle);
+
     qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                  2, base_ecam, 2, size_ecam);
     qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "ranges",
@@ -685,11 +736,52 @@ static void create_pcie(const VirtBoardInfo *vbi, qemu_irq *pic,
                                  2, base_mmio, 2, size_mmio);
 
     qemu_fdt_setprop_cell(vbi->fdt, nodename, "#interrupt-cells", 1);
-    create_pcie_irq_map(vbi, gic_phandle, irq, nodename);
+    create_pcie_irq_map(vbi, vbi->gic_phandle, irq, nodename);
 
     g_free(nodename);
 }
 
+static void create_platform_bus(VirtBoardInfo *vbi, qemu_irq *pic)
+{
+    DeviceState *dev;
+    SysBusDevice *s;
+    int i;
+    ARMPlatformBusFDTParams *fdt_params = g_new(ARMPlatformBusFDTParams, 1);
+    MemoryRegion *sysmem = get_system_memory();
+
+    platform_bus_params.platform_bus_base = vbi->memmap[VIRT_PLATFORM_BUS].base;
+    platform_bus_params.platform_bus_size = vbi->memmap[VIRT_PLATFORM_BUS].size;
+    platform_bus_params.platform_bus_first_irq = vbi->irqmap[VIRT_PLATFORM_BUS];
+    platform_bus_params.platform_bus_num_irqs = PLATFORM_BUS_NUM_IRQS;
+
+    fdt_params->system_params = &platform_bus_params;
+    fdt_params->binfo = &vbi->bootinfo;
+    fdt_params->intc = "/intc";
+    /*
+     * register a machine init done notifier that creates the device tree
+     * nodes of the platform bus and its children dynamic sysbus devices
+     */
+    arm_register_platform_bus_fdt_creator(fdt_params);
+
+    dev = qdev_create(NULL, TYPE_PLATFORM_BUS_DEVICE);
+    dev->id = TYPE_PLATFORM_BUS_DEVICE;
+    qdev_prop_set_uint32(dev, "num_irqs",
+        platform_bus_params.platform_bus_num_irqs);
+    qdev_prop_set_uint32(dev, "mmio_size",
+        platform_bus_params.platform_bus_size);
+    qdev_init_nofail(dev);
+    s = SYS_BUS_DEVICE(dev);
+
+    for (i = 0; i < platform_bus_params.platform_bus_num_irqs; i++) {
+        int irqn = platform_bus_params.platform_bus_first_irq + i;
+        sysbus_connect_irq(s, i, pic[irqn]);
+    }
+
+    memory_region_add_subregion(sysmem,
+                                platform_bus_params.platform_bus_base,
+                                sysbus_mmio_get_region(s, 0));
+}
+
 static void *machvirt_dtb(const struct arm_boot_info *binfo, int *fdt_size)
 {
     const VirtBoardInfo *board = (const VirtBoardInfo *)binfo;
@@ -717,7 +809,6 @@ static void machvirt_init(MachineState *machine)
     VirtBoardInfo *vbi;
     VirtGuestInfoState *guest_info_state = g_malloc0(sizeof *guest_info_state);
     VirtGuestInfo *guest_info = &guest_info_state->info;
-    uint32_t gic_phandle;
     char **cpustr;
 
     if (!cpu_model) {
@@ -794,13 +885,13 @@ static void machvirt_init(MachineState *machine)
 
     create_flash(vbi);
 
-    gic_phandle = create_gic(vbi, pic);
+    create_gic(vbi, pic);
 
     create_uart(vbi, pic);
 
     create_rtc(vbi, pic);
 
-    create_pcie(vbi, pic, gic_phandle);
+    create_pcie(vbi, pic);
 
     /* Create mmio transports, so the user can create virtio backends
      * (which will be automatically plugged in to the transports). If
@@ -828,6 +919,14 @@ static void machvirt_init(MachineState *machine)
     vbi->bootinfo.get_dtb = machvirt_dtb;
     vbi->bootinfo.firmware_loaded = bios_name || drive_get(IF_PFLASH, 0, 0);
     arm_load_kernel(ARM_CPU(first_cpu), &vbi->bootinfo);
+
+    /*
+     * arm_load_kernel machine init done notifier registration must
+     * happen before the platform_bus_create call. In this latter,
+     * another notifier is registered which adds platform bus nodes.
+     * Notifiers are executed in registration reverse order.
+     */
+    create_platform_bus(vbi, pic);
 }
 
 static bool virt_get_secure(Object *obj, Error **errp)
@@ -866,6 +965,7 @@ static void virt_class_init(ObjectClass *oc, void *data)
     mc->desc = "ARM Virtual Machine",
     mc->init = machvirt_init;
     mc->max_cpus = 8;
+    mc->has_dynamic_sysbus = true;
 }
 
 static const TypeInfo machvirt_info = {
diff --git a/hw/block/fdc.c b/hw/block/fdc.c
index d8a8edd936..5e1b67ee43 100644
--- a/hw/block/fdc.c
+++ b/hw/block/fdc.c
@@ -324,7 +324,7 @@ static void fd_revalidate(FDrive *drv)
 /* Intel 82078 floppy disk controller emulation          */
 
 static void fdctrl_reset(FDCtrl *fdctrl, int do_irq);
-static void fdctrl_reset_fifo(FDCtrl *fdctrl);
+static void fdctrl_to_command_phase(FDCtrl *fdctrl);
 static int fdctrl_transfer_handler (void *opaque, int nchan,
                                     int dma_pos, int dma_len);
 static void fdctrl_raise_irq(FDCtrl *fdctrl);
@@ -495,6 +495,33 @@ enum {
     FD_DIR_DSKCHG   = 0x80,
 };
 
+/*
+ * See chapter 5.0 "Controller phases" of the spec:
+ *
+ * Command phase:
+ * The host writes a command and its parameters into the FIFO. The command
+ * phase is completed when all parameters for the command have been supplied,
+ * and execution phase is entered.
+ *
+ * Execution phase:
+ * Data transfers, either DMA or non-DMA. For non-DMA transfers, the FIFO
+ * contains the payload now, otherwise it's unused. When all bytes of the
+ * required data have been transferred, the state is switched to either result
+ * phase (if the command produces status bytes) or directly back into the
+ * command phase for the next command.
+ *
+ * Result phase:
+ * The host reads out the FIFO, which contains one or more result bytes now.
+ */
+enum {
+    /* Only for migration: reconstruct phase from registers like qemu 2.3 */
+    FD_PHASE_RECONSTRUCT    = 0,
+
+    FD_PHASE_COMMAND        = 1,
+    FD_PHASE_EXECUTION      = 2,
+    FD_PHASE_RESULT         = 3,
+};
+
 #define FD_MULTI_TRACK(state) ((state) & FD_STATE_MULTI)
 #define FD_FORMAT_CMD(state) ((state) & FD_STATE_FORMAT)
 
@@ -504,6 +531,7 @@ struct FDCtrl {
     /* Controller state */
     QEMUTimer *result_timer;
     int dma_chann;
+    uint8_t phase;
     /* Controller's identification */
     uint8_t version;
     /* HW */
@@ -671,6 +699,7 @@ static const VMStateDescription vmstate_fdrive_media_changed = {
     .name = "fdrive/media_changed",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = fdrive_media_changed_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(media_changed, FDrive),
         VMSTATE_END_OF_LIST()
@@ -688,6 +717,7 @@ static const VMStateDescription vmstate_fdrive_media_rate = {
     .name = "fdrive/media_rate",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = fdrive_media_rate_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(media_rate, FDrive),
         VMSTATE_END_OF_LIST()
@@ -705,6 +735,7 @@ static const VMStateDescription vmstate_fdrive_perpendicular = {
     .name = "fdrive/perpendicular",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = fdrive_perpendicular_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(perpendicular, FDrive),
         VMSTATE_END_OF_LIST()
@@ -728,22 +759,36 @@ static const VMStateDescription vmstate_fdrive = {
         VMSTATE_UINT8(sect, FDrive),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_fdrive_media_changed,
-            .needed = &fdrive_media_changed_needed,
-        } , {
-            .vmsd = &vmstate_fdrive_media_rate,
-            .needed = &fdrive_media_rate_needed,
-        } , {
-            .vmsd = &vmstate_fdrive_perpendicular,
-            .needed = &fdrive_perpendicular_needed,
-        } , {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_fdrive_media_changed,
+        &vmstate_fdrive_media_rate,
+        &vmstate_fdrive_perpendicular,
+        NULL
     }
 };
 
+/*
+ * Reconstructs the phase from register values according to the logic that was
+ * implemented in qemu 2.3. This is the default value that is used if the phase
+ * subsection is not present on migration.
+ *
+ * Don't change this function to reflect newer qemu versions, it is part of
+ * the migration ABI.
+ */
+static int reconstruct_phase(FDCtrl *fdctrl)
+{
+    if (fdctrl->msr & FD_MSR_NONDMA) {
+        return FD_PHASE_EXECUTION;
+    } else if ((fdctrl->msr & FD_MSR_RQM) == 0) {
+        /* qemu 2.3 disabled RQM only during DMA transfers */
+        return FD_PHASE_EXECUTION;
+    } else if (fdctrl->msr & FD_MSR_DIO) {
+        return FD_PHASE_RESULT;
+    } else {
+        return FD_PHASE_COMMAND;
+    }
+}
+
 static void fdc_pre_save(void *opaque)
 {
     FDCtrl *s = opaque;
@@ -751,12 +796,24 @@ static void fdc_pre_save(void *opaque)
     s->dor_vmstate = s->dor | GET_CUR_DRV(s);
 }
 
+static int fdc_pre_load(void *opaque)
+{
+    FDCtrl *s = opaque;
+    s->phase = FD_PHASE_RECONSTRUCT;
+    return 0;
+}
+
 static int fdc_post_load(void *opaque, int version_id)
 {
     FDCtrl *s = opaque;
 
     SET_CUR_DRV(s, s->dor_vmstate & FD_DOR_SELMASK);
     s->dor = s->dor_vmstate & ~FD_DOR_SELMASK;
+
+    if (s->phase == FD_PHASE_RECONSTRUCT) {
+        s->phase = reconstruct_phase(s);
+    }
+
     return 0;
 }
 
@@ -771,6 +828,7 @@ static const VMStateDescription vmstate_fdc_reset_sensei = {
     .name = "fdc/reset_sensei",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = fdc_reset_sensei_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(reset_sensei, FDCtrl),
         VMSTATE_END_OF_LIST()
@@ -788,17 +846,37 @@ static const VMStateDescription vmstate_fdc_result_timer = {
     .name = "fdc/result_timer",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = fdc_result_timer_needed,
     .fields = (VMStateField[]) {
         VMSTATE_TIMER_PTR(result_timer, FDCtrl),
         VMSTATE_END_OF_LIST()
     }
 };
 
+static bool fdc_phase_needed(void *opaque)
+{
+    FDCtrl *fdctrl = opaque;
+
+    return reconstruct_phase(fdctrl) != fdctrl->phase;
+}
+
+static const VMStateDescription vmstate_fdc_phase = {
+    .name = "fdc/phase",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .needed = fdc_phase_needed,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT8(phase, FDCtrl),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
 static const VMStateDescription vmstate_fdc = {
     .name = "fdc",
     .version_id = 2,
     .minimum_version_id = 2,
     .pre_save = fdc_pre_save,
+    .pre_load = fdc_pre_load,
     .post_load = fdc_post_load,
     .fields = (VMStateField[]) {
         /* Controller State */
@@ -831,16 +909,11 @@ static const VMStateDescription vmstate_fdc = {
                              vmstate_fdrive, FDrive),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_fdc_reset_sensei,
-            .needed = fdc_reset_sensei_needed,
-        } , {
-            .vmsd = &vmstate_fdc_result_timer,
-            .needed = fdc_result_timer_needed,
-        } , {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_fdc_reset_sensei,
+        &vmstate_fdc_result_timer,
+        &vmstate_fdc_phase,
+        NULL
     }
 };
 
@@ -918,7 +991,7 @@ static void fdctrl_reset(FDCtrl *fdctrl, int do_irq)
     fdctrl->data_dir = FD_DIR_WRITE;
     for (i = 0; i < MAX_FD; i++)
         fd_recalibrate(&fdctrl->drives[i]);
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
     if (do_irq) {
         fdctrl->status0 |= FD_SR0_RDYCHG;
         fdctrl_raise_irq(fdctrl);
@@ -1134,17 +1207,22 @@ static uint32_t fdctrl_read_dir(FDCtrl *fdctrl)
     return retval;
 }
 
-/* FIFO state control */
-static void fdctrl_reset_fifo(FDCtrl *fdctrl)
+/* Clear the FIFO and update the state for receiving the next command */
+static void fdctrl_to_command_phase(FDCtrl *fdctrl)
 {
+    fdctrl->phase = FD_PHASE_COMMAND;
     fdctrl->data_dir = FD_DIR_WRITE;
     fdctrl->data_pos = 0;
+    fdctrl->data_len = 1; /* Accept command byte, adjust for params later */
     fdctrl->msr &= ~(FD_MSR_CMDBUSY | FD_MSR_DIO);
+    fdctrl->msr |= FD_MSR_RQM;
 }
 
-/* Set FIFO status for the host to read */
-static void fdctrl_set_fifo(FDCtrl *fdctrl, int fifo_len)
+/* Update the state to allow the guest to read out the command status.
+ * @fifo_len is the number of result bytes to be read out. */
+static void fdctrl_to_result_phase(FDCtrl *fdctrl, int fifo_len)
 {
+    fdctrl->phase = FD_PHASE_RESULT;
     fdctrl->data_dir = FD_DIR_READ;
     fdctrl->data_len = fifo_len;
     fdctrl->data_pos = 0;
@@ -1157,7 +1235,7 @@ static void fdctrl_unimplemented(FDCtrl *fdctrl, int direction)
     qemu_log_mask(LOG_UNIMP, "fdc: unimplemented command 0x%02x\n",
                   fdctrl->fifo[0]);
     fdctrl->fifo[0] = FD_SR0_INVCMD;
-    fdctrl_set_fifo(fdctrl, 1);
+    fdctrl_to_result_phase(fdctrl, 1);
 }
 
 /* Seek to next sector
@@ -1238,7 +1316,7 @@ static void fdctrl_stop_transfer(FDCtrl *fdctrl, uint8_t status0,
     fdctrl->msr |= FD_MSR_RQM | FD_MSR_DIO;
     fdctrl->msr &= ~FD_MSR_NONDMA;
 
-    fdctrl_set_fifo(fdctrl, 7);
+    fdctrl_to_result_phase(fdctrl, 7);
     fdctrl_raise_irq(fdctrl);
 }
 
@@ -1352,7 +1430,7 @@ static void fdctrl_start_transfer(FDCtrl *fdctrl, int direction)
         }
     }
     FLOPPY_DPRINTF("start non-DMA transfer\n");
-    fdctrl->msr |= FD_MSR_NONDMA;
+    fdctrl->msr |= FD_MSR_NONDMA | FD_MSR_RQM;
     if (direction != FD_DIR_WRITE)
         fdctrl->msr |= FD_MSR_DIO;
     /* IO based transfer: calculate len */
@@ -1505,9 +1583,16 @@ static uint32_t fdctrl_read_data(FDCtrl *fdctrl)
         FLOPPY_DPRINTF("error: controller not ready for reading\n");
         return 0;
     }
+
+    /* If data_len spans multiple sectors, the current position in the FIFO
+     * wraps around while fdctrl->data_pos is the real position in the whole
+     * request. */
     pos = fdctrl->data_pos;
     pos %= FD_SECTOR_LEN;
-    if (fdctrl->msr & FD_MSR_NONDMA) {
+
+    switch (fdctrl->phase) {
+    case FD_PHASE_EXECUTION:
+        assert(fdctrl->msr & FD_MSR_NONDMA);
         if (pos == 0) {
             if (fdctrl->data_pos != 0)
                 if (!fdctrl_seek_to_next_sect(fdctrl, cur_drv)) {
@@ -1523,20 +1608,28 @@ static uint32_t fdctrl_read_data(FDCtrl *fdctrl)
                 memset(fdctrl->fifo, 0, FD_SECTOR_LEN);
             }
         }
-    }
-    retval = fdctrl->fifo[pos];
-    if (++fdctrl->data_pos == fdctrl->data_len) {
-        fdctrl->data_pos = 0;
-        /* Switch from transfer mode to status mode
-         * then from status mode to command mode
-         */
-        if (fdctrl->msr & FD_MSR_NONDMA) {
+
+        if (++fdctrl->data_pos == fdctrl->data_len) {
+            fdctrl->msr &= ~FD_MSR_RQM;
             fdctrl_stop_transfer(fdctrl, 0x00, 0x00, 0x00);
-        } else {
-            fdctrl_reset_fifo(fdctrl);
+        }
+        break;
+
+    case FD_PHASE_RESULT:
+        assert(!(fdctrl->msr & FD_MSR_NONDMA));
+        if (++fdctrl->data_pos == fdctrl->data_len) {
+            fdctrl->msr &= ~FD_MSR_RQM;
+            fdctrl_to_command_phase(fdctrl);
             fdctrl_reset_irq(fdctrl);
         }
+        break;
+
+    case FD_PHASE_COMMAND:
+    default:
+        abort();
     }
+
+    retval = fdctrl->fifo[pos];
     FLOPPY_DPRINTF("data register: 0x%02x\n", retval);
 
     return retval;
@@ -1606,7 +1699,7 @@ static void fdctrl_handle_lock(FDCtrl *fdctrl, int direction)
 {
     fdctrl->lock = (fdctrl->fifo[0] & 0x80) ? 1 : 0;
     fdctrl->fifo[0] = fdctrl->lock << 4;
-    fdctrl_set_fifo(fdctrl, 1);
+    fdctrl_to_result_phase(fdctrl, 1);
 }
 
 static void fdctrl_handle_dumpreg(FDCtrl *fdctrl, int direction)
@@ -1631,20 +1724,20 @@ static void fdctrl_handle_dumpreg(FDCtrl *fdctrl, int direction)
         (cur_drv->perpendicular << 2);
     fdctrl->fifo[8] = fdctrl->config;
     fdctrl->fifo[9] = fdctrl->precomp_trk;
-    fdctrl_set_fifo(fdctrl, 10);
+    fdctrl_to_result_phase(fdctrl, 10);
 }
 
 static void fdctrl_handle_version(FDCtrl *fdctrl, int direction)
 {
     /* Controller's version */
     fdctrl->fifo[0] = fdctrl->version;
-    fdctrl_set_fifo(fdctrl, 1);
+    fdctrl_to_result_phase(fdctrl, 1);
 }
 
 static void fdctrl_handle_partid(FDCtrl *fdctrl, int direction)
 {
     fdctrl->fifo[0] = 0x41; /* Stepping 1 */
-    fdctrl_set_fifo(fdctrl, 1);
+    fdctrl_to_result_phase(fdctrl, 1);
 }
 
 static void fdctrl_handle_restore(FDCtrl *fdctrl, int direction)
@@ -1667,7 +1760,7 @@ static void fdctrl_handle_restore(FDCtrl *fdctrl, int direction)
     fdctrl->config = fdctrl->fifo[11];
     fdctrl->precomp_trk = fdctrl->fifo[12];
     fdctrl->pwrd = fdctrl->fifo[13];
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
 }
 
 static void fdctrl_handle_save(FDCtrl *fdctrl, int direction)
@@ -1697,7 +1790,7 @@ static void fdctrl_handle_save(FDCtrl *fdctrl, int direction)
     fdctrl->fifo[12] = fdctrl->pwrd;
     fdctrl->fifo[13] = 0;
     fdctrl->fifo[14] = 0;
-    fdctrl_set_fifo(fdctrl, 15);
+    fdctrl_to_result_phase(fdctrl, 15);
 }
 
 static void fdctrl_handle_readid(FDCtrl *fdctrl, int direction)
@@ -1746,7 +1839,7 @@ static void fdctrl_handle_specify(FDCtrl *fdctrl, int direction)
     else
         fdctrl->dor |= FD_DOR_DMAEN;
     /* No result back */
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
 }
 
 static void fdctrl_handle_sense_drive_status(FDCtrl *fdctrl, int direction)
@@ -1762,7 +1855,7 @@ static void fdctrl_handle_sense_drive_status(FDCtrl *fdctrl, int direction)
         (cur_drv->head << 2) |
         GET_CUR_DRV(fdctrl) |
         0x28;
-    fdctrl_set_fifo(fdctrl, 1);
+    fdctrl_to_result_phase(fdctrl, 1);
 }
 
 static void fdctrl_handle_recalibrate(FDCtrl *fdctrl, int direction)
@@ -1772,7 +1865,7 @@ static void fdctrl_handle_recalibrate(FDCtrl *fdctrl, int direction)
     SET_CUR_DRV(fdctrl, fdctrl->fifo[1] & FD_DOR_SELMASK);
     cur_drv = get_cur_drv(fdctrl);
     fd_recalibrate(cur_drv);
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
     /* Raise Interrupt */
     fdctrl->status0 |= FD_SR0_SEEK;
     fdctrl_raise_irq(fdctrl);
@@ -1788,7 +1881,7 @@ static void fdctrl_handle_sense_interrupt_status(FDCtrl *fdctrl, int direction)
         fdctrl->reset_sensei--;
     } else if (!(fdctrl->sra & FD_SRA_INTPEND)) {
         fdctrl->fifo[0] = FD_SR0_INVCMD;
-        fdctrl_set_fifo(fdctrl, 1);
+        fdctrl_to_result_phase(fdctrl, 1);
         return;
     } else {
         fdctrl->fifo[0] =
@@ -1797,7 +1890,7 @@ static void fdctrl_handle_sense_interrupt_status(FDCtrl *fdctrl, int direction)
     }
 
     fdctrl->fifo[1] = cur_drv->track;
-    fdctrl_set_fifo(fdctrl, 2);
+    fdctrl_to_result_phase(fdctrl, 2);
     fdctrl_reset_irq(fdctrl);
     fdctrl->status0 = FD_SR0_RDYCHG;
 }
@@ -1808,7 +1901,7 @@ static void fdctrl_handle_seek(FDCtrl *fdctrl, int direction)
 
     SET_CUR_DRV(fdctrl, fdctrl->fifo[1] & FD_DOR_SELMASK);
     cur_drv = get_cur_drv(fdctrl);
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
     /* The seek command just sends step pulses to the drive and doesn't care if
      * there is a medium inserted of if it's banging the head against the drive.
      */
@@ -1825,7 +1918,7 @@ static void fdctrl_handle_perpendicular_mode(FDCtrl *fdctrl, int direction)
     if (fdctrl->fifo[1] & 0x80)
         cur_drv->perpendicular = fdctrl->fifo[1] & 0x7;
     /* No result back */
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
 }
 
 static void fdctrl_handle_configure(FDCtrl *fdctrl, int direction)
@@ -1833,20 +1926,20 @@ static void fdctrl_handle_configure(FDCtrl *fdctrl, int direction)
     fdctrl->config = fdctrl->fifo[2];
     fdctrl->precomp_trk =  fdctrl->fifo[3];
     /* No result back */
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
 }
 
 static void fdctrl_handle_powerdown_mode(FDCtrl *fdctrl, int direction)
 {
     fdctrl->pwrd = fdctrl->fifo[1];
     fdctrl->fifo[0] = fdctrl->fifo[1];
-    fdctrl_set_fifo(fdctrl, 1);
+    fdctrl_to_result_phase(fdctrl, 1);
 }
 
 static void fdctrl_handle_option(FDCtrl *fdctrl, int direction)
 {
     /* No result back */
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
 }
 
 static void fdctrl_handle_drive_specification_command(FDCtrl *fdctrl, int direction)
@@ -1862,15 +1955,15 @@ static void fdctrl_handle_drive_specification_command(FDCtrl *fdctrl, int direct
             fdctrl->fifo[0] = fdctrl->fifo[1];
             fdctrl->fifo[2] = 0;
             fdctrl->fifo[3] = 0;
-            fdctrl_set_fifo(fdctrl, 4);
+            fdctrl_to_result_phase(fdctrl, 4);
         } else {
-            fdctrl_reset_fifo(fdctrl);
+            fdctrl_to_command_phase(fdctrl);
         }
     } else if (fdctrl->data_len > 7) {
         /* ERROR */
         fdctrl->fifo[0] = 0x80 |
             (cur_drv->head << 2) | GET_CUR_DRV(fdctrl);
-        fdctrl_set_fifo(fdctrl, 1);
+        fdctrl_to_result_phase(fdctrl, 1);
     }
 }
 
@@ -1887,7 +1980,7 @@ static void fdctrl_handle_relative_seek_in(FDCtrl *fdctrl, int direction)
         fd_seek(cur_drv, cur_drv->head,
                 cur_drv->track + fdctrl->fifo[2], cur_drv->sect, 1);
     }
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
     /* Raise Interrupt */
     fdctrl->status0 |= FD_SR0_SEEK;
     fdctrl_raise_irq(fdctrl);
@@ -1905,20 +1998,25 @@ static void fdctrl_handle_relative_seek_out(FDCtrl *fdctrl, int direction)
         fd_seek(cur_drv, cur_drv->head,
                 cur_drv->track - fdctrl->fifo[2], cur_drv->sect, 1);
     }
-    fdctrl_reset_fifo(fdctrl);
+    fdctrl_to_command_phase(fdctrl);
     /* Raise Interrupt */
     fdctrl->status0 |= FD_SR0_SEEK;
     fdctrl_raise_irq(fdctrl);
 }
 
-static const struct {
+/*
+ * Handlers for the execution phase of each command
+ */
+typedef struct FDCtrlCommand {
     uint8_t value;
     uint8_t mask;
     const char* name;
     int parameters;
     void (*handler)(FDCtrl *fdctrl, int direction);
     int direction;
-} handlers[] = {
+} FDCtrlCommand;
+
+static const FDCtrlCommand handlers[] = {
     { FD_CMD_READ, 0x1f, "READ", 8, fdctrl_start_transfer, FD_DIR_READ },
     { FD_CMD_WRITE, 0x3f, "WRITE", 8, fdctrl_start_transfer, FD_DIR_WRITE },
     { FD_CMD_SEEK, 0xff, "SEEK", 2, fdctrl_handle_seek },
@@ -1955,9 +2053,19 @@ static const struct {
 /* Associate command to an index in the 'handlers' array */
 static uint8_t command_to_handler[256];
 
+static const FDCtrlCommand *get_command(uint8_t cmd)
+{
+    int idx;
+
+    idx = command_to_handler[cmd];
+    FLOPPY_DPRINTF("%s command\n", handlers[idx].name);
+    return &handlers[idx];
+}
+
 static void fdctrl_write_data(FDCtrl *fdctrl, uint32_t value)
 {
     FDrive *cur_drv;
+    const FDCtrlCommand *cmd;
     uint32_t pos;
 
     /* Reset mode */
@@ -1970,12 +2078,27 @@ static void fdctrl_write_data(FDCtrl *fdctrl, uint32_t value)
         return;
     }
     fdctrl->dsr &= ~FD_DSR_PWRDOWN;
-    /* Is it write command time ? */
-    if (fdctrl->msr & FD_MSR_NONDMA) {
+
+    FLOPPY_DPRINTF("%s: %02x\n", __func__, value);
+
+    /* If data_len spans multiple sectors, the current position in the FIFO
+     * wraps around while fdctrl->data_pos is the real position in the whole
+     * request. */
+    pos = fdctrl->data_pos++;
+    pos %= FD_SECTOR_LEN;
+    fdctrl->fifo[pos] = value;
+
+    if (fdctrl->data_pos == fdctrl->data_len) {
+        fdctrl->msr &= ~FD_MSR_RQM;
+    }
+
+    switch (fdctrl->phase) {
+    case FD_PHASE_EXECUTION:
+        /* For DMA requests, RQM should be cleared during execution phase, so
+         * we would have errored out above. */
+        assert(fdctrl->msr & FD_MSR_NONDMA);
+
         /* FIFO data write */
-        pos = fdctrl->data_pos++;
-        pos %= FD_SECTOR_LEN;
-        fdctrl->fifo[pos] = value;
         if (pos == FD_SECTOR_LEN - 1 ||
             fdctrl->data_pos == fdctrl->data_len) {
             cur_drv = get_cur_drv(fdctrl);
@@ -1983,45 +2106,54 @@ static void fdctrl_write_data(FDCtrl *fdctrl, uint32_t value)
                 < 0) {
                 FLOPPY_DPRINTF("error writing sector %d\n",
                                fd_sector(cur_drv));
-                return;
+                break;
             }
             if (!fdctrl_seek_to_next_sect(fdctrl, cur_drv)) {
                 FLOPPY_DPRINTF("error seeking to next sector %d\n",
                                fd_sector(cur_drv));
-                return;
+                break;
             }
         }
-        /* Switch from transfer mode to status mode
-         * then from status mode to command mode
-         */
-        if (fdctrl->data_pos == fdctrl->data_len)
+
+        /* Switch to result phase when done with the transfer */
+        if (fdctrl->data_pos == fdctrl->data_len) {
             fdctrl_stop_transfer(fdctrl, 0x00, 0x00, 0x00);
-        return;
-    }
-    if (fdctrl->data_pos == 0) {
-        /* Command */
-        pos = command_to_handler[value & 0xff];
-        FLOPPY_DPRINTF("%s command\n", handlers[pos].name);
-        fdctrl->data_len = handlers[pos].parameters + 1;
-        fdctrl->msr |= FD_MSR_CMDBUSY;
-    }
+        }
+        break;
 
-    FLOPPY_DPRINTF("%s: %02x\n", __func__, value);
-    pos = fdctrl->data_pos++;
-    pos %= FD_SECTOR_LEN;
-    fdctrl->fifo[pos] = value;
-    if (fdctrl->data_pos == fdctrl->data_len) {
-        /* We now have all parameters
-         * and will be able to treat the command
-         */
-        if (fdctrl->data_state & FD_STATE_FORMAT) {
-            fdctrl_format_sector(fdctrl);
-            return;
+    case FD_PHASE_COMMAND:
+        assert(!(fdctrl->msr & FD_MSR_NONDMA));
+        assert(fdctrl->data_pos < FD_SECTOR_LEN);
+
+        if (pos == 0) {
+            /* The first byte specifies the command. Now we start reading
+             * as many parameters as this command requires. */
+            cmd = get_command(value);
+            fdctrl->data_len = cmd->parameters + 1;
+            if (cmd->parameters) {
+                fdctrl->msr |= FD_MSR_RQM;
+            }
+            fdctrl->msr |= FD_MSR_CMDBUSY;
+        }
+
+        if (fdctrl->data_pos == fdctrl->data_len) {
+            /* We have all parameters now, execute the command */
+            fdctrl->phase = FD_PHASE_EXECUTION;
+
+            if (fdctrl->data_state & FD_STATE_FORMAT) {
+                fdctrl_format_sector(fdctrl);
+                break;
+            }
+
+            cmd = get_command(fdctrl->fifo[0]);
+            FLOPPY_DPRINTF("Calling handler for '%s'\n", cmd->name);
+            cmd->handler(fdctrl, cmd->direction);
         }
+        break;
 
-        pos = command_to_handler[fdctrl->fifo[0] & 0xff];
-        FLOPPY_DPRINTF("treat %s command\n", handlers[pos].name);
-        (*handlers[pos].handler)(fdctrl, handlers[pos].direction);
+    case FD_PHASE_RESULT:
+    default:
+        abort();
     }
 }
 
diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
index d282695086..2ba6c77293 100644
--- a/hw/block/pflash_cfi01.c
+++ b/hw/block/pflash_cfi01.c
@@ -64,6 +64,9 @@ do {                                                        \
 #define TYPE_CFI_PFLASH01 "cfi.pflash01"
 #define CFI_PFLASH01(obj) OBJECT_CHECK(pflash_t, (obj), TYPE_CFI_PFLASH01)
 
+#define PFLASH_BE          0
+#define PFLASH_SECURE      1
+
 struct pflash_t {
     /*< private >*/
     SysBusDevice parent_obj;
@@ -75,7 +78,7 @@ struct pflash_t {
     uint8_t bank_width;
     uint8_t device_width; /* If 0, device width not specified. */
     uint8_t max_device_width;  /* max device width in bytes */
-    uint8_t be;
+    uint32_t features;
     uint8_t wcycle; /* if 0, the flash is read normally */
     int ro;
     uint8_t cmd;
@@ -235,12 +238,57 @@ static uint32_t pflash_devid_query(pflash_t *pfl, hwaddr offset)
     return resp;
 }
 
+static uint32_t pflash_data_read(pflash_t *pfl, hwaddr offset,
+                                 int width, int be)
+{
+    uint8_t *p;
+    uint32_t ret;
+
+    p = pfl->storage;
+    switch (width) {
+    case 1:
+        ret = p[offset];
+        DPRINTF("%s: data offset " TARGET_FMT_plx " %02x\n",
+                __func__, offset, ret);
+        break;
+    case 2:
+        if (be) {
+            ret = p[offset] << 8;
+            ret |= p[offset + 1];
+        } else {
+            ret = p[offset];
+            ret |= p[offset + 1] << 8;
+        }
+        DPRINTF("%s: data offset " TARGET_FMT_plx " %04x\n",
+                __func__, offset, ret);
+        break;
+    case 4:
+        if (be) {
+            ret = p[offset] << 24;
+            ret |= p[offset + 1] << 16;
+            ret |= p[offset + 2] << 8;
+            ret |= p[offset + 3];
+        } else {
+            ret = p[offset];
+            ret |= p[offset + 1] << 8;
+            ret |= p[offset + 2] << 16;
+            ret |= p[offset + 3] << 24;
+        }
+        DPRINTF("%s: data offset " TARGET_FMT_plx " %08x\n",
+                __func__, offset, ret);
+        break;
+    default:
+        DPRINTF("BUG in %s\n", __func__);
+        abort();
+    }
+    return ret;
+}
+
 static uint32_t pflash_read (pflash_t *pfl, hwaddr offset,
                              int width, int be)
 {
     hwaddr boff;
     uint32_t ret;
-    uint8_t *p;
 
     ret = -1;
 
@@ -257,43 +305,7 @@ static uint32_t pflash_read (pflash_t *pfl, hwaddr offset,
         /* fall through to read code */
     case 0x00:
         /* Flash area read */
-        p = pfl->storage;
-        switch (width) {
-        case 1:
-            ret = p[offset];
-            DPRINTF("%s: data offset " TARGET_FMT_plx " %02x\n",
-                    __func__, offset, ret);
-            break;
-        case 2:
-            if (be) {
-                ret = p[offset] << 8;
-                ret |= p[offset + 1];
-            } else {
-                ret = p[offset];
-                ret |= p[offset + 1] << 8;
-            }
-            DPRINTF("%s: data offset " TARGET_FMT_plx " %04x\n",
-                    __func__, offset, ret);
-            break;
-        case 4:
-            if (be) {
-                ret = p[offset] << 24;
-                ret |= p[offset + 1] << 16;
-                ret |= p[offset + 2] << 8;
-                ret |= p[offset + 3];
-            } else {
-                ret = p[offset];
-                ret |= p[offset + 1] << 8;
-                ret |= p[offset + 2] << 16;
-                ret |= p[offset + 3] << 24;
-            }
-            DPRINTF("%s: data offset " TARGET_FMT_plx " %08x\n",
-                    __func__, offset, ret);
-            break;
-        default:
-            DPRINTF("BUG in %s\n", __func__);
-        }
-
+        ret = pflash_data_read(pfl, offset, width, be);
         break;
     case 0x10: /* Single byte program */
     case 0x20: /* Block erase */
@@ -648,101 +660,37 @@ static void pflash_write(pflash_t *pfl, hwaddr offset,
 }
 
 
-static uint32_t pflash_readb_be(void *opaque, hwaddr addr)
-{
-    return pflash_read(opaque, addr, 1, 1);
-}
-
-static uint32_t pflash_readb_le(void *opaque, hwaddr addr)
-{
-    return pflash_read(opaque, addr, 1, 0);
-}
-
-static uint32_t pflash_readw_be(void *opaque, hwaddr addr)
-{
-    pflash_t *pfl = opaque;
-
-    return pflash_read(pfl, addr, 2, 1);
-}
-
-static uint32_t pflash_readw_le(void *opaque, hwaddr addr)
-{
-    pflash_t *pfl = opaque;
-
-    return pflash_read(pfl, addr, 2, 0);
-}
-
-static uint32_t pflash_readl_be(void *opaque, hwaddr addr)
-{
-    pflash_t *pfl = opaque;
-
-    return pflash_read(pfl, addr, 4, 1);
-}
-
-static uint32_t pflash_readl_le(void *opaque, hwaddr addr)
+static MemTxResult pflash_mem_read_with_attrs(void *opaque, hwaddr addr, uint64_t *value,
+                                              unsigned len, MemTxAttrs attrs)
 {
     pflash_t *pfl = opaque;
+    bool be = !!(pfl->features & (1 << PFLASH_BE));
 
-    return pflash_read(pfl, addr, 4, 0);
-}
-
-static void pflash_writeb_be(void *opaque, hwaddr addr,
-                             uint32_t value)
-{
-    pflash_write(opaque, addr, value, 1, 1);
-}
-
-static void pflash_writeb_le(void *opaque, hwaddr addr,
-                             uint32_t value)
-{
-    pflash_write(opaque, addr, value, 1, 0);
-}
-
-static void pflash_writew_be(void *opaque, hwaddr addr,
-                             uint32_t value)
-{
-    pflash_t *pfl = opaque;
-
-    pflash_write(pfl, addr, value, 2, 1);
-}
-
-static void pflash_writew_le(void *opaque, hwaddr addr,
-                             uint32_t value)
-{
-    pflash_t *pfl = opaque;
-
-    pflash_write(pfl, addr, value, 2, 0);
-}
-
-static void pflash_writel_be(void *opaque, hwaddr addr,
-                             uint32_t value)
-{
-    pflash_t *pfl = opaque;
-
-    pflash_write(pfl, addr, value, 4, 1);
+    if ((pfl->features & (1 << PFLASH_SECURE)) && !attrs.secure) {
+        *value = pflash_data_read(opaque, addr, len, be);
+    } else {
+        *value = pflash_read(opaque, addr, len, be);
+    }
+    return MEMTX_OK;
 }
 
-static void pflash_writel_le(void *opaque, hwaddr addr,
-                             uint32_t value)
+static MemTxResult pflash_mem_write_with_attrs(void *opaque, hwaddr addr, uint64_t value,
+                                               unsigned len, MemTxAttrs attrs)
 {
     pflash_t *pfl = opaque;
+    bool be = !!(pfl->features & (1 << PFLASH_BE));
 
-    pflash_write(pfl, addr, value, 4, 0);
+    if ((pfl->features & (1 << PFLASH_SECURE)) && !attrs.secure) {
+        return MEMTX_ERROR;
+    } else {
+        pflash_write(opaque, addr, value, len, be);
+        return MEMTX_OK;
+    }
 }
 
-static const MemoryRegionOps pflash_cfi01_ops_be = {
-    .old_mmio = {
-        .read = { pflash_readb_be, pflash_readw_be, pflash_readl_be, },
-        .write = { pflash_writeb_be, pflash_writew_be, pflash_writel_be, },
-    },
-    .endianness = DEVICE_NATIVE_ENDIAN,
-};
-
-static const MemoryRegionOps pflash_cfi01_ops_le = {
-    .old_mmio = {
-        .read = { pflash_readb_le, pflash_readw_le, pflash_readl_le, },
-        .write = { pflash_writeb_le, pflash_writew_le, pflash_writel_le, },
-    },
+static const MemoryRegionOps pflash_cfi01_ops = {
+    .read_with_attrs = pflash_mem_read_with_attrs,
+    .write_with_attrs = pflash_mem_write_with_attrs,
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
@@ -773,7 +721,8 @@ static void pflash_cfi01_realize(DeviceState *dev, Error **errp)
 
     memory_region_init_rom_device(
         &pfl->mem, OBJECT(dev),
-        pfl->be ? &pflash_cfi01_ops_be : &pflash_cfi01_ops_le, pfl,
+        &pflash_cfi01_ops,
+        pfl,
         pfl->name, total_len, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
@@ -925,7 +874,8 @@ static Property pflash_cfi01_properties[] = {
     DEFINE_PROP_UINT8("width", struct pflash_t, bank_width, 0),
     DEFINE_PROP_UINT8("device-width", struct pflash_t, device_width, 0),
     DEFINE_PROP_UINT8("max-device-width", struct pflash_t, max_device_width, 0),
-    DEFINE_PROP_UINT8("big-endian", struct pflash_t, be, 0),
+    DEFINE_PROP_BIT("big-endian", struct pflash_t, features, PFLASH_BE, 0),
+    DEFINE_PROP_BIT("secure", struct pflash_t, features, PFLASH_SECURE, 0),
     DEFINE_PROP_UINT16("id0", struct pflash_t, ident0, 0),
     DEFINE_PROP_UINT16("id1", struct pflash_t, ident1, 0),
     DEFINE_PROP_UINT16("id2", struct pflash_t, ident2, 0),
@@ -975,7 +925,7 @@ pflash_t *pflash_cfi01_register(hwaddr base,
     qdev_prop_set_uint32(dev, "num-blocks", nb_blocs);
     qdev_prop_set_uint64(dev, "sector-length", sector_len);
     qdev_prop_set_uint8(dev, "width", bank_width);
-    qdev_prop_set_uint8(dev, "big-endian", !!be);
+    qdev_prop_set_bit(dev, "big-endian", !!be);
     qdev_prop_set_uint16(dev, "id0", id0);
     qdev_prop_set_uint16(dev, "id1", id1);
     qdev_prop_set_uint16(dev, "id2", id2);
diff --git a/hw/char/parallel.c b/hw/char/parallel.c
index 4079554bb9..c2b553f0d1 100644
--- a/hw/char/parallel.c
+++ b/hw/char/parallel.c
@@ -641,28 +641,3 @@ static void parallel_register_types(void)
 }
 
 type_init(parallel_register_types)
-
-static void parallel_init(ISABus *bus, int index, CharDriverState *chr)
-{
-    DeviceState *dev;
-    ISADevice *isadev;
-
-    isadev = isa_create(bus, "isa-parallel");
-    dev = DEVICE(isadev);
-    qdev_prop_set_uint32(dev, "index", index);
-    qdev_prop_set_chr(dev, "chardev", chr);
-    qdev_init_nofail(dev);
-}
-
-void parallel_hds_isa_init(ISABus *bus, int n)
-{
-    int i;
-
-    assert(n <= MAX_PARALLEL_PORTS);
-
-    for (i = 0; i < n; i++) {
-        if (parallel_hds[i]) {
-            parallel_init(bus, i, parallel_hds[i]);
-        }
-    }
-}
diff --git a/hw/char/serial.c b/hw/char/serial.c
index 55011cfd26..513d73c27f 100644
--- a/hw/char/serial.c
+++ b/hw/char/serial.c
@@ -662,6 +662,7 @@ static const VMStateDescription vmstate_serial_thr_ipending = {
     .name = "serial/thr_ipending",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = serial_thr_ipending_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(thr_ipending, SerialState),
         VMSTATE_END_OF_LIST()
@@ -678,6 +679,7 @@ static const VMStateDescription vmstate_serial_tsr = {
     .name = "serial/tsr",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = serial_tsr_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(tsr_retry, SerialState),
         VMSTATE_UINT8(thr, SerialState),
@@ -697,6 +699,7 @@ static const VMStateDescription vmstate_serial_recv_fifo = {
     .name = "serial/recv_fifo",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = serial_recv_fifo_needed,
     .fields = (VMStateField[]) {
         VMSTATE_STRUCT(recv_fifo, SerialState, 1, vmstate_fifo8, Fifo8),
         VMSTATE_END_OF_LIST()
@@ -713,6 +716,7 @@ static const VMStateDescription vmstate_serial_xmit_fifo = {
     .name = "serial/xmit_fifo",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = serial_xmit_fifo_needed,
     .fields = (VMStateField[]) {
         VMSTATE_STRUCT(xmit_fifo, SerialState, 1, vmstate_fifo8, Fifo8),
         VMSTATE_END_OF_LIST()
@@ -729,6 +733,7 @@ static const VMStateDescription vmstate_serial_fifo_timeout_timer = {
     .name = "serial/fifo_timeout_timer",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = serial_fifo_timeout_timer_needed,
     .fields = (VMStateField[]) {
         VMSTATE_TIMER_PTR(fifo_timeout_timer, SerialState),
         VMSTATE_END_OF_LIST()
@@ -745,6 +750,7 @@ static const VMStateDescription vmstate_serial_timeout_ipending = {
     .name = "serial/timeout_ipending",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = serial_timeout_ipending_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(timeout_ipending, SerialState),
         VMSTATE_END_OF_LIST()
@@ -760,6 +766,7 @@ static bool serial_poll_needed(void *opaque)
 static const VMStateDescription vmstate_serial_poll = {
     .name = "serial/poll",
     .version_id = 1,
+    .needed = serial_poll_needed,
     .minimum_version_id = 1,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(poll_msl, SerialState),
@@ -788,31 +795,15 @@ const VMStateDescription vmstate_serial = {
         VMSTATE_UINT8_V(fcr_vmstate, SerialState, 3),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_serial_thr_ipending,
-            .needed = &serial_thr_ipending_needed,
-        } , {
-            .vmsd = &vmstate_serial_tsr,
-            .needed = &serial_tsr_needed,
-        } , {
-            .vmsd = &vmstate_serial_recv_fifo,
-            .needed = &serial_recv_fifo_needed,
-        } , {
-            .vmsd = &vmstate_serial_xmit_fifo,
-            .needed = &serial_xmit_fifo_needed,
-        } , {
-            .vmsd = &vmstate_serial_fifo_timeout_timer,
-            .needed = &serial_fifo_timeout_timer_needed,
-        } , {
-            .vmsd = &vmstate_serial_timeout_ipending,
-            .needed = &serial_timeout_ipending_needed,
-        } , {
-            .vmsd = &vmstate_serial_poll,
-            .needed = &serial_poll_needed,
-        } , {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_serial_thr_ipending,
+        &vmstate_serial_tsr,
+        &vmstate_serial_recv_fifo,
+        &vmstate_serial_xmit_fifo,
+        &vmstate_serial_fifo_timeout_timer,
+        &vmstate_serial_timeout_ipending,
+        &vmstate_serial_poll,
+        NULL
     }
 };
 
diff --git a/hw/core/machine.c b/hw/core/machine.c
index 25c45e6f9d..ac4654e9dd 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -294,6 +294,14 @@ static void machine_init_notify(Notifier *notifier, void *data)
     foreach_dynamic_sysbus_device(error_on_sysbus_device, NULL);
 }
 
+static void machine_class_init(ObjectClass *oc, void *data)
+{
+    MachineClass *mc = MACHINE_CLASS(oc);
+
+    /* Default 128 MB as guest ram size */
+    mc->default_ram_size = 128 * M_BYTE;
+}
+
 static void machine_initfn(Object *obj)
 {
     MachineState *ms = MACHINE(obj);
@@ -463,6 +471,7 @@ static const TypeInfo machine_info = {
     .parent = TYPE_OBJECT,
     .abstract = true,
     .class_size = sizeof(MachineClass),
+    .class_init    = machine_class_init,
     .instance_size = sizeof(MachineState),
     .instance_init = machine_initfn,
     .instance_finalize = machine_finalize,
diff --git a/hw/core/nmi.c b/hw/core/nmi.c
index 3dff020659..5260d6c1ec 100644
--- a/hw/core/nmi.c
+++ b/hw/core/nmi.c
@@ -21,6 +21,7 @@
 
 #include "hw/nmi.h"
 #include "qapi/qmp/qerror.h"
+#include "monitor/monitor.h"
 
 struct do_nmi_s {
     int cpu_index;
@@ -70,6 +71,25 @@ void nmi_monitor_handle(int cpu_index, Error **errp)
     }
 }
 
+void inject_nmi(void)
+{
+#if defined(TARGET_I386)
+    CPUState *cs;
+
+    CPU_FOREACH(cs) {
+        X86CPU *cpu = X86_CPU(cs);
+
+        if (!cpu->apic_state) {
+            cpu_interrupt(cs, CPU_INTERRUPT_NMI);
+        } else {
+            apic_deliver_nmi(cpu->apic_state);
+        }
+    }
+#else
+    nmi_monitor_handle(0, NULL);
+#endif
+}
+
 static const TypeInfo nmi_info = {
     .name          = TYPE_NMI,
     .parent        = TYPE_INTERFACE,
diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index c413226a97..0309fe5767 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -389,7 +389,7 @@ void qdev_set_nic_properties(DeviceState *dev, NICInfo *nd)
     nd->instantiated = 1;
 }
 
-static int qdev_add_one_global(QemuOpts *opts, void *opaque)
+static int qdev_add_one_global(void *opaque, QemuOpts *opts, Error **errp)
 {
     GlobalProperty *g;
 
@@ -404,5 +404,6 @@ static int qdev_add_one_global(QemuOpts *opts, void *opaque)
 
 void qemu_add_globals(void)
 {
-    qemu_opts_foreach(qemu_find_opts("global"), qdev_add_one_global, NULL, 0);
+    qemu_opts_foreach(qemu_find_opts("global"),
+                      qdev_add_one_global, NULL, NULL);
 }
diff --git a/hw/display/Makefile.objs b/hw/display/Makefile.objs
index 3ea106d9f3..dd8ea76d17 100644
--- a/hw/display/Makefile.objs
+++ b/hw/display/Makefile.objs
@@ -34,3 +34,7 @@ obj-$(CONFIG_CG3) += cg3.o
 obj-$(CONFIG_VGA) += vga.o
 
 common-obj-$(CONFIG_QXL) += qxl.o qxl-logger.o qxl-render.o
+
+obj-$(CONFIG_VIRTIO) += virtio-gpu.o
+obj-$(CONFIG_VIRTIO_PCI) += virtio-gpu-pci.o
+obj-$(CONFIG_VIRTIO_VGA) += virtio-vga.o
diff --git a/hw/display/cg3.c b/hw/display/cg3.c
index 1e6ff2b546..b94e5e0d78 100644
--- a/hw/display/cg3.c
+++ b/hw/display/cg3.c
@@ -106,6 +106,7 @@ static void cg3_update_display(void *opaque)
     pix = memory_region_get_ram_ptr(&s->vram_mem);
     data = (uint32_t *)surface_data(surface);
 
+    memory_region_sync_dirty_bitmap(&s->vram_mem);
     for (y = 0; y < height; y++) {
         int update = s->full_update;
 
@@ -309,6 +310,7 @@ static void cg3_realizefn(DeviceState *dev, Error **errp)
 
     memory_region_init_ram(&s->vram_mem, NULL, "cg3.vram", s->vram_size,
                            &error_abort);
+    memory_region_set_log(&s->vram_mem, true, DIRTY_MEMORY_VGA);
     vmstate_register_ram_global(&s->vram_mem);
     sysbus_init_mmio(sbd, &s->vram_mem);
 
diff --git a/hw/display/exynos4210_fimd.c b/hw/display/exynos4210_fimd.c
index 45c62afac1..603ef50568 100644
--- a/hw/display/exynos4210_fimd.c
+++ b/hw/display/exynos4210_fimd.c
@@ -337,7 +337,7 @@ static inline void fimd_swap_data(unsigned int swap_ctl, uint64_t *data)
     if (swap_ctl & FIMD_WINCON_SWAP_BITS) {
         res = 0;
         for (i = 0; i < 64; i++) {
-            if (x & (1ULL << (64 - i))) {
+            if (x & (1ULL << (63 - i))) {
                 res |= (1ULL << i);
             }
         }
@@ -1109,6 +1109,12 @@ static inline int fimd_get_buffer_id(Exynos4210fimdWindow *w)
     }
 }
 
+static void exynos4210_fimd_invalidate(void *opaque)
+{
+    Exynos4210fimdState *s = (Exynos4210fimdState *)opaque;
+    s->invalidate = true;
+}
+
 /* Updates specified window's MemorySection based on values of WINCON,
  * VIDOSDA, VIDOSDB, VIDWADDx and SHADOWCON registers */
 static void fimd_update_memory_section(Exynos4210fimdState *s, unsigned win)
@@ -1136,7 +1142,11 @@ static void fimd_update_memory_section(Exynos4210fimdState *s, unsigned win)
     /* TODO: add .exit and unref the region there.  Not needed yet since sysbus
      * does not support hot-unplug.
      */
-    memory_region_unref(w->mem_section.mr);
+    if (w->mem_section.mr) {
+        memory_region_set_log(w->mem_section.mr, false, DIRTY_MEMORY_VGA);
+        memory_region_unref(w->mem_section.mr);
+    }
+
     w->mem_section = memory_region_find(sysbus_address_space(sbd),
                                         fb_start_addr, w->fb_len);
     assert(w->mem_section.mr);
@@ -1162,6 +1172,8 @@ static void fimd_update_memory_section(Exynos4210fimdState *s, unsigned win)
         cpu_physical_memory_unmap(w->host_fb_addr, fb_mapped_len, 0, 0);
         goto error_return;
     }
+    memory_region_set_log(w->mem_section.mr, true, DIRTY_MEMORY_VGA);
+    exynos4210_fimd_invalidate(s);
     return;
 
 error_return:
@@ -1224,12 +1236,6 @@ static void exynos4210_fimd_update_irq(Exynos4210fimdState *s)
     }
 }
 
-static void exynos4210_fimd_invalidate(void *opaque)
-{
-    Exynos4210fimdState *s = (Exynos4210fimdState *)opaque;
-    s->invalidate = true;
-}
-
 static void exynos4210_update_resolution(Exynos4210fimdState *s)
 {
     DisplaySurface *surface = qemu_console_surface(s->console);
diff --git a/hw/display/framebuffer.c b/hw/display/framebuffer.c
index 4546e42654..2cabced208 100644
--- a/hw/display/framebuffer.c
+++ b/hw/display/framebuffer.c
@@ -63,6 +63,10 @@ void framebuffer_update_display(
     assert(mem_section.offset_within_address_space == base);
 
     memory_region_sync_dirty_bitmap(mem);
+    if (!memory_region_is_logging(mem, DIRTY_MEMORY_VGA)) {
+        invalidate = true;
+    }
+
     src_base = cpu_physical_memory_map(base, &src_len, 0);
     /* If we can't map the framebuffer then bail.  We could try harder,
        but it's not really worth it as dirty flag tracking will probably
diff --git a/hw/display/g364fb.c b/hw/display/g364fb.c
index 46f7b41211..52a9733bfd 100644
--- a/hw/display/g364fb.c
+++ b/hw/display/g364fb.c
@@ -260,6 +260,7 @@ static void g364fb_update_display(void *opaque)
         qemu_console_resize(s->con, s->width, s->height);
     }
 
+    memory_region_sync_dirty_bitmap(&s->mem_vram);
     if (s->ctla & CTLA_FORCE_BLANK) {
         g364fb_draw_blank(s);
     } else if (s->depth == 8) {
@@ -489,7 +490,7 @@ static void g364fb_init(DeviceState *dev, G364State *s)
     memory_region_init_ram_ptr(&s->mem_vram, NULL, "vram",
                                s->vram_size, s->vram);
     vmstate_register_ram(&s->mem_vram, dev);
-    memory_region_set_coalescing(&s->mem_vram);
+    memory_region_set_log(&s->mem_vram, true, DIRTY_MEMORY_VGA);
 }
 
 #define TYPE_G364 "sysbus-g364"
diff --git a/hw/display/qxl.c b/hw/display/qxl.c
index b220e2d5d2..722146ec3a 100644
--- a/hw/display/qxl.c
+++ b/hw/display/qxl.c
@@ -2220,6 +2220,7 @@ static VMStateDescription qxl_vmstate_monitors_config = {
     .name               = "qxl/monitors-config",
     .version_id         = 1,
     .minimum_version_id = 1,
+    .needed = qxl_monitors_config_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT64(guest_monitors_config, PCIQXLDevice),
         VMSTATE_END_OF_LIST()
@@ -2253,13 +2254,9 @@ static VMStateDescription qxl_vmstate = {
         VMSTATE_UINT64(guest_cursor, PCIQXLDevice),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &qxl_vmstate_monitors_config,
-            .needed = qxl_monitors_config_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &qxl_vmstate_monitors_config,
+        NULL
     }
 };
 
diff --git a/hw/display/sm501.c b/hw/display/sm501.c
index c72154b6f1..15a5ba8000 100644
--- a/hw/display/sm501.c
+++ b/hw/display/sm501.c
@@ -1322,6 +1322,7 @@ static void sm501_draw_crt(SM501State * s)
     }
 
     /* draw each line according to conditions */
+    memory_region_sync_dirty_bitmap(&s->local_mem_region);
     for (y = 0; y < height; y++) {
 	int update_hwc = draw_hwc_line ? within_hwc_y_range(s, y, 1) : 0;
 	int update = full_update || update_hwc;
@@ -1412,6 +1413,7 @@ void sm501_init(MemoryRegion *address_space_mem, uint32_t base,
     memory_region_init_ram(&s->local_mem_region, NULL, "sm501.local",
                            local_mem_bytes, &error_abort);
     vmstate_register_ram_global(&s->local_mem_region);
+    memory_region_set_log(&s->local_mem_region, true, DIRTY_MEMORY_VGA);
     s->local_mem = memory_region_get_ram_ptr(&s->local_mem_region);
     memory_region_add_subregion(address_space_mem, base, &s->local_mem_region);
 
diff --git a/hw/display/tc6393xb.c b/hw/display/tc6393xb.c
index 66b7ade8da..f5f3f3e69d 100644
--- a/hw/display/tc6393xb.c
+++ b/hw/display/tc6393xb.c
@@ -571,7 +571,7 @@ TC6393xbState *tc6393xb_init(MemoryRegion *sysmem, uint32_t base, qemu_irq irq)
     s->irq = irq;
     s->gpio_in = qemu_allocate_irqs(tc6393xb_gpio_set, s, TC6393XB_GPIOS);
 
-    s->l3v = *qemu_allocate_irqs(tc6393xb_l3v, s, 1);
+    s->l3v = qemu_allocate_irq(tc6393xb_l3v, s, 0);
     s->blanked = 1;
 
     s->sub_irqs = qemu_allocate_irqs(tc6393xb_sub_irq, s, TC6393XB_NR_IRQS);
diff --git a/hw/display/tcx.c b/hw/display/tcx.c
index a9f9f66d15..f3faf78bf8 100644
--- a/hw/display/tcx.c
+++ b/hw/display/tcx.c
@@ -353,6 +353,7 @@ static void tcx_update_display(void *opaque)
         return;
     }
 
+    memory_region_sync_dirty_bitmap(&ts->vram_mem);
     for (y = 0; y < ts->height; page += TARGET_PAGE_SIZE) {
         if (memory_region_get_dirty(&ts->vram_mem, page, TARGET_PAGE_SIZE,
                                     DIRTY_MEMORY_VGA)) {
@@ -446,6 +447,7 @@ static void tcx24_update_display(void *opaque)
     dd = surface_stride(surface);
     ds = 1024;
 
+    memory_region_sync_dirty_bitmap(&ts->vram_mem);
     for (y = 0; y < ts->height; page += TARGET_PAGE_SIZE,
             page24 += TARGET_PAGE_SIZE, cpage += TARGET_PAGE_SIZE) {
         if (tcx24_check_dirty(ts, page, page24, cpage)) {
@@ -1006,6 +1008,7 @@ static void tcx_realizefn(DeviceState *dev, Error **errp)
     memory_region_init_ram(&s->vram_mem, OBJECT(s), "tcx.vram",
                            s->vram_size * (1 + 4 + 4), &error_abort);
     vmstate_register_ram_global(&s->vram_mem);
+    memory_region_set_log(&s->vram_mem, true, DIRTY_MEMORY_VGA);
     vram_base = memory_region_get_ram_ptr(&s->vram_mem);
 
     /* 10/ROM : FCode ROM */
diff --git a/hw/display/vga-pci.c b/hw/display/vga-pci.c
index ff5dfb2c23..1dfa331e60 100644
--- a/hw/display/vga-pci.c
+++ b/hw/display/vga-pci.c
@@ -54,9 +54,7 @@ typedef struct PCIVGAState {
     VGACommonState vga;
     uint32_t flags;
     MemoryRegion mmio;
-    MemoryRegion ioport;
-    MemoryRegion bochs;
-    MemoryRegion qext;
+    MemoryRegion mrs[3];
 } PCIVGAState;
 
 #define TYPE_PCI_VGA "pci-vga"
@@ -76,16 +74,16 @@ static const VMStateDescription vmstate_vga_pci = {
 static uint64_t pci_vga_ioport_read(void *ptr, hwaddr addr,
                                     unsigned size)
 {
-    PCIVGAState *d = ptr;
+    VGACommonState *s = ptr;
     uint64_t ret = 0;
 
     switch (size) {
     case 1:
-        ret = vga_ioport_read(&d->vga, addr);
+        ret = vga_ioport_read(s, addr + 0x3c0);
         break;
     case 2:
-        ret  = vga_ioport_read(&d->vga, addr);
-        ret |= vga_ioport_read(&d->vga, addr+1) << 8;
+        ret  = vga_ioport_read(s, addr + 0x3c0);
+        ret |= vga_ioport_read(s, addr + 0x3c1) << 8;
         break;
     }
     return ret;
@@ -94,11 +92,11 @@ static uint64_t pci_vga_ioport_read(void *ptr, hwaddr addr,
 static void pci_vga_ioport_write(void *ptr, hwaddr addr,
                                  uint64_t val, unsigned size)
 {
-    PCIVGAState *d = ptr;
+    VGACommonState *s = ptr;
 
     switch (size) {
     case 1:
-        vga_ioport_write(&d->vga, addr + 0x3c0, val);
+        vga_ioport_write(s, addr + 0x3c0, val);
         break;
     case 2:
         /*
@@ -106,8 +104,8 @@ static void pci_vga_ioport_write(void *ptr, hwaddr addr,
          * indexed registers with a single word write because the
          * index byte is updated first.
          */
-        vga_ioport_write(&d->vga, addr + 0x3c0, val & 0xff);
-        vga_ioport_write(&d->vga, addr + 0x3c1, (val >> 8) & 0xff);
+        vga_ioport_write(s, addr + 0x3c0, val & 0xff);
+        vga_ioport_write(s, addr + 0x3c1, (val >> 8) & 0xff);
         break;
     }
 }
@@ -125,21 +123,21 @@ static const MemoryRegionOps pci_vga_ioport_ops = {
 static uint64_t pci_vga_bochs_read(void *ptr, hwaddr addr,
                                    unsigned size)
 {
-    PCIVGAState *d = ptr;
+    VGACommonState *s = ptr;
     int index = addr >> 1;
 
-    vbe_ioport_write_index(&d->vga, 0, index);
-    return vbe_ioport_read_data(&d->vga, 0);
+    vbe_ioport_write_index(s, 0, index);
+    return vbe_ioport_read_data(s, 0);
 }
 
 static void pci_vga_bochs_write(void *ptr, hwaddr addr,
                                 uint64_t val, unsigned size)
 {
-    PCIVGAState *d = ptr;
+    VGACommonState *s = ptr;
     int index = addr >> 1;
 
-    vbe_ioport_write_index(&d->vga, 0, index);
-    vbe_ioport_write_data(&d->vga, 0, val);
+    vbe_ioport_write_index(s, 0, index);
+    vbe_ioport_write_data(s, 0, val);
 }
 
 static const MemoryRegionOps pci_vga_bochs_ops = {
@@ -154,13 +152,13 @@ static const MemoryRegionOps pci_vga_bochs_ops = {
 
 static uint64_t pci_vga_qext_read(void *ptr, hwaddr addr, unsigned size)
 {
-    PCIVGAState *d = ptr;
+    VGACommonState *s = ptr;
 
     switch (addr) {
     case PCI_VGA_QEXT_REG_SIZE:
         return PCI_VGA_QEXT_SIZE;
     case PCI_VGA_QEXT_REG_BYTEORDER:
-        return d->vga.big_endian_fb ?
+        return s->big_endian_fb ?
             PCI_VGA_QEXT_BIG_ENDIAN : PCI_VGA_QEXT_LITTLE_ENDIAN;
     default:
         return 0;
@@ -170,15 +168,15 @@ static uint64_t pci_vga_qext_read(void *ptr, hwaddr addr, unsigned size)
 static void pci_vga_qext_write(void *ptr, hwaddr addr,
                                uint64_t val, unsigned size)
 {
-    PCIVGAState *d = ptr;
+    VGACommonState *s = ptr;
 
     switch (addr) {
     case PCI_VGA_QEXT_REG_BYTEORDER:
         if (val == PCI_VGA_QEXT_BIG_ENDIAN) {
-            d->vga.big_endian_fb = true;
+            s->big_endian_fb = true;
         }
         if (val == PCI_VGA_QEXT_LITTLE_ENDIAN) {
-            d->vga.big_endian_fb = false;
+            s->big_endian_fb = false;
         }
         break;
     }
@@ -206,10 +204,34 @@ static const MemoryRegionOps pci_vga_qext_ops = {
     .endianness = DEVICE_LITTLE_ENDIAN,
 };
 
+void pci_std_vga_mmio_region_init(VGACommonState *s,
+                                  MemoryRegion *parent,
+                                  MemoryRegion *subs,
+                                  bool qext)
+{
+    memory_region_init_io(&subs[0], NULL, &pci_vga_ioport_ops, s,
+                          "vga ioports remapped", PCI_VGA_IOPORT_SIZE);
+    memory_region_add_subregion(parent, PCI_VGA_IOPORT_OFFSET,
+                                &subs[0]);
+
+    memory_region_init_io(&subs[1], NULL, &pci_vga_bochs_ops, s,
+                          "bochs dispi interface", PCI_VGA_BOCHS_SIZE);
+    memory_region_add_subregion(parent, PCI_VGA_BOCHS_OFFSET,
+                                &subs[1]);
+
+    if (qext) {
+        memory_region_init_io(&subs[2], NULL, &pci_vga_qext_ops, s,
+                              "qemu extended regs", PCI_VGA_QEXT_SIZE);
+        memory_region_add_subregion(parent, PCI_VGA_QEXT_OFFSET,
+                                    &subs[2]);
+    }
+}
+
 static void pci_std_vga_realize(PCIDevice *dev, Error **errp)
 {
     PCIVGAState *d = PCI_VGA(dev);
     VGACommonState *s = &d->vga;
+    bool qext = false;
 
     /* vga + console init */
     vga_common_init(s, OBJECT(dev), true);
@@ -224,23 +246,12 @@ static void pci_std_vga_realize(PCIDevice *dev, Error **errp)
     /* mmio bar for vga register access */
     if (d->flags & (1 << PCI_VGA_FLAG_ENABLE_MMIO)) {
         memory_region_init(&d->mmio, NULL, "vga.mmio", 4096);
-        memory_region_init_io(&d->ioport, NULL, &pci_vga_ioport_ops, d,
-                              "vga ioports remapped", PCI_VGA_IOPORT_SIZE);
-        memory_region_init_io(&d->bochs, NULL, &pci_vga_bochs_ops, d,
-                              "bochs dispi interface", PCI_VGA_BOCHS_SIZE);
-
-        memory_region_add_subregion(&d->mmio, PCI_VGA_IOPORT_OFFSET,
-                                    &d->ioport);
-        memory_region_add_subregion(&d->mmio, PCI_VGA_BOCHS_OFFSET,
-                                    &d->bochs);
 
         if (d->flags & (1 << PCI_VGA_FLAG_ENABLE_QEXT)) {
-            memory_region_init_io(&d->qext, NULL, &pci_vga_qext_ops, d,
-                                  "qemu extended regs", PCI_VGA_QEXT_SIZE);
-            memory_region_add_subregion(&d->mmio, PCI_VGA_QEXT_OFFSET,
-                                        &d->qext);
+            qext = true;
             pci_set_byte(&d->dev.config[PCI_REVISION_ID], 2);
         }
+        pci_std_vga_mmio_region_init(s, &d->mmio, d->mrs, qext);
 
         pci_register_bar(&d->dev, 2, PCI_BASE_ADDRESS_SPACE_MEMORY, &d->mmio);
     }
@@ -262,6 +273,7 @@ static void pci_secondary_vga_realize(PCIDevice *dev, Error **errp)
 {
     PCIVGAState *d = PCI_VGA(dev);
     VGACommonState *s = &d->vga;
+    bool qext = false;
 
     /* vga + console init */
     vga_common_init(s, OBJECT(dev), false);
@@ -269,23 +281,12 @@ static void pci_secondary_vga_realize(PCIDevice *dev, Error **errp)
 
     /* mmio bar */
     memory_region_init(&d->mmio, OBJECT(dev), "vga.mmio", 4096);
-    memory_region_init_io(&d->ioport, OBJECT(dev), &pci_vga_ioport_ops, d,
-                          "vga ioports remapped", PCI_VGA_IOPORT_SIZE);
-    memory_region_init_io(&d->bochs, OBJECT(dev), &pci_vga_bochs_ops, d,
-                          "bochs dispi interface", PCI_VGA_BOCHS_SIZE);
-
-    memory_region_add_subregion(&d->mmio, PCI_VGA_IOPORT_OFFSET,
-                                &d->ioport);
-    memory_region_add_subregion(&d->mmio, PCI_VGA_BOCHS_OFFSET,
-                                &d->bochs);
 
     if (d->flags & (1 << PCI_VGA_FLAG_ENABLE_QEXT)) {
-        memory_region_init_io(&d->qext, NULL, &pci_vga_qext_ops, d,
-                              "qemu extended regs", PCI_VGA_QEXT_SIZE);
-        memory_region_add_subregion(&d->mmio, PCI_VGA_QEXT_OFFSET,
-                                    &d->qext);
+        qext = true;
         pci_set_byte(&d->dev.config[PCI_REVISION_ID], 2);
     }
+    pci_std_vga_mmio_region_init(s, &d->mmio, d->mrs, qext);
 
     pci_register_bar(&d->dev, 0, PCI_BASE_ADDRESS_MEM_PREFETCH, &s->vram);
     pci_register_bar(&d->dev, 2, PCI_BASE_ADDRESS_SPACE_MEMORY, &d->mmio);
diff --git a/hw/display/vga.c b/hw/display/vga.c
index d1d296c74e..b35d523e65 100644
--- a/hw/display/vga.c
+++ b/hw/display/vga.c
@@ -2035,6 +2035,7 @@ static const VMStateDescription vmstate_vga_endian = {
     .name = "vga.endian",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = vga_endian_state_needed,
     .fields = (VMStateField[]) {
         VMSTATE_BOOL(big_endian_fb, VGACommonState),
         VMSTATE_END_OF_LIST()
@@ -2078,13 +2079,9 @@ const VMStateDescription vmstate_vga_common = {
         VMSTATE_UINT32(vbe_bank_mask, VGACommonState),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_vga_endian,
-            .needed = vga_endian_state_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_vga_endian,
+        NULL
     }
 };
 
diff --git a/hw/display/vga_int.h b/hw/display/vga_int.h
index fcfcc5f431..40ba6a4207 100644
--- a/hw/display/vga_int.h
+++ b/hw/display/vga_int.h
@@ -219,4 +219,10 @@ extern const uint8_t gr_mask[16];
 
 extern const MemoryRegionOps vga_mem_ops;
 
+/* vga-pci.c */
+void pci_std_vga_mmio_region_init(VGACommonState *s,
+                                  MemoryRegion *parent,
+                                  MemoryRegion *subs,
+                                  bool qext);
+
 #endif
diff --git a/hw/display/virtio-gpu-pci.c b/hw/display/virtio-gpu-pci.c
new file mode 100644
index 0000000000..f0f25c7bc9
--- /dev/null
+++ b/hw/display/virtio-gpu-pci.c
@@ -0,0 +1,68 @@
+/*
+ * Virtio video device
+ *
+ * Copyright Red Hat
+ *
+ * Authors:
+ *  Dave Airlie
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+#include "hw/pci/pci.h"
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/virtio-bus.h"
+#include "hw/virtio/virtio-pci.h"
+#include "hw/virtio/virtio-gpu.h"
+
+static Property virtio_gpu_pci_properties[] = {
+    DEFINE_VIRTIO_GPU_PROPERTIES(VirtIOGPUPCI, vdev.conf),
+    DEFINE_VIRTIO_GPU_PCI_PROPERTIES(VirtIOPCIProxy),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_gpu_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp)
+{
+    VirtIOGPUPCI *vgpu = VIRTIO_GPU_PCI(vpci_dev);
+    DeviceState *vdev = DEVICE(&vgpu->vdev);
+
+    qdev_set_parent_bus(vdev, BUS(&vpci_dev->bus));
+    /* force virtio-1.0 */
+    vpci_dev->flags &= ~VIRTIO_PCI_FLAG_DISABLE_MODERN;
+    vpci_dev->flags |= VIRTIO_PCI_FLAG_DISABLE_LEGACY;
+    object_property_set_bool(OBJECT(vdev), true, "realized", errp);
+}
+
+static void virtio_gpu_pci_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioPCIClass *k = VIRTIO_PCI_CLASS(klass);
+    PCIDeviceClass *pcidev_k = PCI_DEVICE_CLASS(klass);
+
+    set_bit(DEVICE_CATEGORY_DISPLAY, dc->categories);
+    dc->props = virtio_gpu_pci_properties;
+    k->realize = virtio_gpu_pci_realize;
+    pcidev_k->class_id = PCI_CLASS_DISPLAY_OTHER;
+}
+
+static void virtio_gpu_initfn(Object *obj)
+{
+    VirtIOGPUPCI *dev = VIRTIO_GPU_PCI(obj);
+    object_initialize(&dev->vdev, sizeof(dev->vdev), TYPE_VIRTIO_GPU);
+    object_property_add_child(obj, "virtio-backend", OBJECT(&dev->vdev), NULL);
+}
+
+static const TypeInfo virtio_gpu_pci_info = {
+    .name = TYPE_VIRTIO_GPU_PCI,
+    .parent = TYPE_VIRTIO_PCI,
+    .instance_size = sizeof(VirtIOGPUPCI),
+    .instance_init = virtio_gpu_initfn,
+    .class_init = virtio_gpu_pci_class_init,
+};
+
+static void virtio_gpu_pci_register_types(void)
+{
+    type_register_static(&virtio_gpu_pci_info);
+}
+type_init(virtio_gpu_pci_register_types)
diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c
new file mode 100644
index 0000000000..8c109b79f4
--- /dev/null
+++ b/hw/display/virtio-gpu.c
@@ -0,0 +1,918 @@
+/*
+ * Virtio GPU Device
+ *
+ * Copyright Red Hat, Inc. 2013-2014
+ *
+ * Authors:
+ *     Dave Airlie <airlied@redhat.com>
+ *     Gerd Hoffmann <kraxel@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu-common.h"
+#include "qemu/iov.h"
+#include "ui/console.h"
+#include "trace.h"
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/virtio-gpu.h"
+#include "hw/virtio/virtio-bus.h"
+
+static struct virtio_gpu_simple_resource*
+virtio_gpu_find_resource(VirtIOGPU *g, uint32_t resource_id);
+
+static void update_cursor_data_simple(VirtIOGPU *g,
+                                      struct virtio_gpu_scanout *s,
+                                      uint32_t resource_id)
+{
+    struct virtio_gpu_simple_resource *res;
+    uint32_t pixels;
+
+    res = virtio_gpu_find_resource(g, resource_id);
+    if (!res) {
+        return;
+    }
+
+    if (pixman_image_get_width(res->image)  != s->current_cursor->width ||
+        pixman_image_get_height(res->image) != s->current_cursor->height) {
+        return;
+    }
+
+    pixels = s->current_cursor->width * s->current_cursor->height;
+    memcpy(s->current_cursor->data,
+           pixman_image_get_data(res->image),
+           pixels * sizeof(uint32_t));
+}
+
+static void update_cursor(VirtIOGPU *g, struct virtio_gpu_update_cursor *cursor)
+{
+    struct virtio_gpu_scanout *s;
+
+    if (cursor->pos.scanout_id >= g->conf.max_outputs) {
+        return;
+    }
+    s = &g->scanout[cursor->pos.scanout_id];
+
+    if (cursor->hdr.type != VIRTIO_GPU_CMD_MOVE_CURSOR) {
+        if (!s->current_cursor) {
+            s->current_cursor = cursor_alloc(64, 64);
+        }
+
+        s->current_cursor->hot_x = cursor->hot_x;
+        s->current_cursor->hot_y = cursor->hot_y;
+
+        if (cursor->resource_id > 0) {
+            update_cursor_data_simple(g, s, cursor->resource_id);
+        }
+        dpy_cursor_define(s->con, s->current_cursor);
+    }
+    dpy_mouse_set(s->con, cursor->pos.x, cursor->pos.y,
+                  cursor->resource_id ? 1 : 0);
+}
+
+static void virtio_gpu_get_config(VirtIODevice *vdev, uint8_t *config)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    memcpy(config, &g->virtio_config, sizeof(g->virtio_config));
+}
+
+static void virtio_gpu_set_config(VirtIODevice *vdev, const uint8_t *config)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    struct virtio_gpu_config vgconfig;
+
+    memcpy(&vgconfig, config, sizeof(g->virtio_config));
+
+    if (vgconfig.events_clear) {
+        g->virtio_config.events_read &= ~vgconfig.events_clear;
+    }
+}
+
+static uint64_t virtio_gpu_get_features(VirtIODevice *vdev, uint64_t features)
+{
+    return features;
+}
+
+static void virtio_gpu_notify_event(VirtIOGPU *g, uint32_t event_type)
+{
+    g->virtio_config.events_read |= event_type;
+    virtio_notify_config(&g->parent_obj);
+}
+
+static struct virtio_gpu_simple_resource *
+virtio_gpu_find_resource(VirtIOGPU *g, uint32_t resource_id)
+{
+    struct virtio_gpu_simple_resource *res;
+
+    QTAILQ_FOREACH(res, &g->reslist, next) {
+        if (res->resource_id == resource_id) {
+            return res;
+        }
+    }
+    return NULL;
+}
+
+void virtio_gpu_ctrl_response(VirtIOGPU *g,
+                              struct virtio_gpu_ctrl_command *cmd,
+                              struct virtio_gpu_ctrl_hdr *resp,
+                              size_t resp_len)
+{
+    size_t s;
+
+    if (cmd->cmd_hdr.flags & VIRTIO_GPU_FLAG_FENCE) {
+        resp->flags |= VIRTIO_GPU_FLAG_FENCE;
+        resp->fence_id = cmd->cmd_hdr.fence_id;
+        resp->ctx_id = cmd->cmd_hdr.ctx_id;
+    }
+    s = iov_from_buf(cmd->elem.in_sg, cmd->elem.in_num, 0, resp, resp_len);
+    if (s != resp_len) {
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: response size incorrect %zu vs %zu\n",
+                      __func__, s, resp_len);
+    }
+    virtqueue_push(cmd->vq, &cmd->elem, s);
+    virtio_notify(VIRTIO_DEVICE(g), cmd->vq);
+    cmd->finished = true;
+}
+
+void virtio_gpu_ctrl_response_nodata(VirtIOGPU *g,
+                                     struct virtio_gpu_ctrl_command *cmd,
+                                     enum virtio_gpu_ctrl_type type)
+{
+    struct virtio_gpu_ctrl_hdr resp;
+
+    memset(&resp, 0, sizeof(resp));
+    resp.type = type;
+    virtio_gpu_ctrl_response(g, cmd, &resp, sizeof(resp));
+}
+
+static void
+virtio_gpu_fill_display_info(VirtIOGPU *g,
+                             struct virtio_gpu_resp_display_info *dpy_info)
+{
+    int i;
+
+    for (i = 0; i < g->conf.max_outputs; i++) {
+        if (g->enabled_output_bitmask & (1 << i)) {
+            dpy_info->pmodes[i].enabled = 1;
+            dpy_info->pmodes[i].r.width = g->req_state[i].width;
+            dpy_info->pmodes[i].r.height = g->req_state[i].height;
+        }
+    }
+}
+
+void virtio_gpu_get_display_info(VirtIOGPU *g,
+                                 struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_resp_display_info display_info;
+
+    trace_virtio_gpu_cmd_get_display_info();
+    memset(&display_info, 0, sizeof(display_info));
+    display_info.hdr.type = VIRTIO_GPU_RESP_OK_DISPLAY_INFO;
+    virtio_gpu_fill_display_info(g, &display_info);
+    virtio_gpu_ctrl_response(g, cmd, &display_info.hdr,
+                             sizeof(display_info));
+}
+
+static pixman_format_code_t get_pixman_format(uint32_t virtio_gpu_format)
+{
+    switch (virtio_gpu_format) {
+#ifdef HOST_WORDS_BIGENDIAN
+    case VIRTIO_GPU_FORMAT_B8G8R8X8_UNORM:
+        return PIXMAN_b8g8r8x8;
+    case VIRTIO_GPU_FORMAT_B8G8R8A8_UNORM:
+        return PIXMAN_b8g8r8a8;
+    case VIRTIO_GPU_FORMAT_X8R8G8B8_UNORM:
+        return PIXMAN_x8r8g8b8;
+    case VIRTIO_GPU_FORMAT_A8R8G8B8_UNORM:
+        return PIXMAN_a8r8g8b8;
+    case VIRTIO_GPU_FORMAT_R8G8B8X8_UNORM:
+        return PIXMAN_r8g8b8x8;
+    case VIRTIO_GPU_FORMAT_R8G8B8A8_UNORM:
+        return PIXMAN_r8g8b8a8;
+    case VIRTIO_GPU_FORMAT_X8B8G8R8_UNORM:
+        return PIXMAN_x8b8g8r8;
+    case VIRTIO_GPU_FORMAT_A8B8G8R8_UNORM:
+        return PIXMAN_a8b8g8r8;
+#else
+    case VIRTIO_GPU_FORMAT_B8G8R8X8_UNORM:
+        return PIXMAN_x8r8g8b8;
+    case VIRTIO_GPU_FORMAT_B8G8R8A8_UNORM:
+        return PIXMAN_a8r8g8b8;
+    case VIRTIO_GPU_FORMAT_X8R8G8B8_UNORM:
+        return PIXMAN_b8g8r8x8;
+    case VIRTIO_GPU_FORMAT_A8R8G8B8_UNORM:
+        return PIXMAN_b8g8r8a8;
+    case VIRTIO_GPU_FORMAT_R8G8B8X8_UNORM:
+        return PIXMAN_x8b8g8r8;
+    case VIRTIO_GPU_FORMAT_R8G8B8A8_UNORM:
+        return PIXMAN_a8b8g8r8;
+    case VIRTIO_GPU_FORMAT_X8B8G8R8_UNORM:
+        return PIXMAN_r8g8b8x8;
+    case VIRTIO_GPU_FORMAT_A8B8G8R8_UNORM:
+        return PIXMAN_r8g8b8a8;
+#endif
+    default:
+        return 0;
+    }
+}
+
+static void virtio_gpu_resource_create_2d(VirtIOGPU *g,
+                                          struct virtio_gpu_ctrl_command *cmd)
+{
+    pixman_format_code_t pformat;
+    struct virtio_gpu_simple_resource *res;
+    struct virtio_gpu_resource_create_2d c2d;
+
+    VIRTIO_GPU_FILL_CMD(c2d);
+    trace_virtio_gpu_cmd_res_create_2d(c2d.resource_id, c2d.format,
+                                       c2d.width, c2d.height);
+
+    if (c2d.resource_id == 0) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: resource id 0 is not allowed\n",
+                      __func__);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+
+    res = virtio_gpu_find_resource(g, c2d.resource_id);
+    if (res) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: resource already exists %d\n",
+                      __func__, c2d.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+
+    res = g_new0(struct virtio_gpu_simple_resource, 1);
+
+    res->width = c2d.width;
+    res->height = c2d.height;
+    res->format = c2d.format;
+    res->resource_id = c2d.resource_id;
+
+    pformat = get_pixman_format(c2d.format);
+    if (!pformat) {
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: host couldn't handle guest format %d\n",
+                      __func__, c2d.format);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_PARAMETER;
+        return;
+    }
+    res->image = pixman_image_create_bits(pformat,
+                                          c2d.width,
+                                          c2d.height,
+                                          NULL, 0);
+
+    if (!res->image) {
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: resource creation failed %d %d %d\n",
+                      __func__, c2d.resource_id, c2d.width, c2d.height);
+        g_free(res);
+        cmd->error = VIRTIO_GPU_RESP_ERR_OUT_OF_MEMORY;
+        return;
+    }
+
+    QTAILQ_INSERT_HEAD(&g->reslist, res, next);
+}
+
+static void virtio_gpu_resource_destroy(VirtIOGPU *g,
+                                        struct virtio_gpu_simple_resource *res)
+{
+    pixman_image_unref(res->image);
+    QTAILQ_REMOVE(&g->reslist, res, next);
+    g_free(res);
+}
+
+static void virtio_gpu_resource_unref(VirtIOGPU *g,
+                                      struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_simple_resource *res;
+    struct virtio_gpu_resource_unref unref;
+
+    VIRTIO_GPU_FILL_CMD(unref);
+    trace_virtio_gpu_cmd_res_unref(unref.resource_id);
+
+    res = virtio_gpu_find_resource(g, unref.resource_id);
+    if (!res) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal resource specified %d\n",
+                      __func__, unref.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+    virtio_gpu_resource_destroy(g, res);
+}
+
+static void virtio_gpu_transfer_to_host_2d(VirtIOGPU *g,
+                                           struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_simple_resource *res;
+    int h;
+    uint32_t src_offset, dst_offset, stride;
+    int bpp;
+    pixman_format_code_t format;
+    struct virtio_gpu_transfer_to_host_2d t2d;
+
+    VIRTIO_GPU_FILL_CMD(t2d);
+    trace_virtio_gpu_cmd_res_xfer_toh_2d(t2d.resource_id);
+
+    res = virtio_gpu_find_resource(g, t2d.resource_id);
+    if (!res || !res->iov) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal resource specified %d\n",
+                      __func__, t2d.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+
+    if (t2d.r.x > res->width ||
+        t2d.r.y > res->height ||
+        t2d.r.width > res->width ||
+        t2d.r.height > res->height ||
+        t2d.r.x + t2d.r.width > res->width ||
+        t2d.r.y + t2d.r.height > res->height) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: transfer bounds outside resource"
+                      " bounds for resource %d: %d %d %d %d vs %d %d\n",
+                      __func__, t2d.resource_id, t2d.r.x, t2d.r.y,
+                      t2d.r.width, t2d.r.height, res->width, res->height);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_PARAMETER;
+        return;
+    }
+
+    format = pixman_image_get_format(res->image);
+    bpp = (PIXMAN_FORMAT_BPP(format) + 7) / 8;
+    stride = pixman_image_get_stride(res->image);
+
+    if (t2d.offset || t2d.r.x || t2d.r.y ||
+        t2d.r.width != pixman_image_get_width(res->image)) {
+        void *img_data = pixman_image_get_data(res->image);
+        for (h = 0; h < t2d.r.height; h++) {
+            src_offset = t2d.offset + stride * h;
+            dst_offset = (t2d.r.y + h) * stride + (t2d.r.x * bpp);
+
+            iov_to_buf(res->iov, res->iov_cnt, src_offset,
+                       (uint8_t *)img_data
+                       + dst_offset, t2d.r.width * bpp);
+        }
+    } else {
+        iov_to_buf(res->iov, res->iov_cnt, 0,
+                   pixman_image_get_data(res->image),
+                   pixman_image_get_stride(res->image)
+                   * pixman_image_get_height(res->image));
+    }
+}
+
+static void virtio_gpu_resource_flush(VirtIOGPU *g,
+                                      struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_simple_resource *res;
+    struct virtio_gpu_resource_flush rf;
+    pixman_region16_t flush_region;
+    int i;
+
+    VIRTIO_GPU_FILL_CMD(rf);
+    trace_virtio_gpu_cmd_res_flush(rf.resource_id,
+                                   rf.r.width, rf.r.height, rf.r.x, rf.r.y);
+
+    res = virtio_gpu_find_resource(g, rf.resource_id);
+    if (!res) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal resource specified %d\n",
+                      __func__, rf.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+
+    if (rf.r.x > res->width ||
+        rf.r.y > res->height ||
+        rf.r.width > res->width ||
+        rf.r.height > res->height ||
+        rf.r.x + rf.r.width > res->width ||
+        rf.r.y + rf.r.height > res->height) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: flush bounds outside resource"
+                      " bounds for resource %d: %d %d %d %d vs %d %d\n",
+                      __func__, rf.resource_id, rf.r.x, rf.r.y,
+                      rf.r.width, rf.r.height, res->width, res->height);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_PARAMETER;
+        return;
+    }
+
+    pixman_region_init_rect(&flush_region,
+                            rf.r.x, rf.r.y, rf.r.width, rf.r.height);
+    for (i = 0; i < VIRTIO_GPU_MAX_SCANOUT; i++) {
+        struct virtio_gpu_scanout *scanout;
+        pixman_region16_t region, finalregion;
+        pixman_box16_t *extents;
+
+        if (!(res->scanout_bitmask & (1 << i))) {
+            continue;
+        }
+        scanout = &g->scanout[i];
+
+        pixman_region_init(&finalregion);
+        pixman_region_init_rect(&region, scanout->x, scanout->y,
+                                scanout->width, scanout->height);
+
+        pixman_region_intersect(&finalregion, &flush_region, &region);
+        pixman_region_translate(&finalregion, -scanout->x, -scanout->y);
+        extents = pixman_region_extents(&finalregion);
+        /* work out the area we need to update for each console */
+        dpy_gfx_update(g->scanout[i].con,
+                       extents->x1, extents->y1,
+                       extents->x2 - extents->x1,
+                       extents->y2 - extents->y1);
+
+        pixman_region_fini(&region);
+        pixman_region_fini(&finalregion);
+    }
+    pixman_region_fini(&flush_region);
+}
+
+static void virtio_gpu_set_scanout(VirtIOGPU *g,
+                                   struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_simple_resource *res;
+    struct virtio_gpu_scanout *scanout;
+    pixman_format_code_t format;
+    uint32_t offset;
+    int bpp;
+    struct virtio_gpu_set_scanout ss;
+
+    VIRTIO_GPU_FILL_CMD(ss);
+    trace_virtio_gpu_cmd_set_scanout(ss.scanout_id, ss.resource_id,
+                                     ss.r.width, ss.r.height, ss.r.x, ss.r.y);
+
+    g->enable = 1;
+    if (ss.resource_id == 0) {
+        scanout = &g->scanout[ss.scanout_id];
+        if (scanout->resource_id) {
+            res = virtio_gpu_find_resource(g, scanout->resource_id);
+            if (res) {
+                res->scanout_bitmask &= ~(1 << ss.scanout_id);
+            }
+        }
+        if (ss.scanout_id == 0 ||
+            ss.scanout_id >= g->conf.max_outputs) {
+            qemu_log_mask(LOG_GUEST_ERROR,
+                          "%s: illegal scanout id specified %d",
+                          __func__, ss.scanout_id);
+            cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_SCANOUT_ID;
+            return;
+        }
+        dpy_gfx_replace_surface(g->scanout[ss.scanout_id].con, NULL);
+        scanout->ds = NULL;
+        scanout->width = 0;
+        scanout->height = 0;
+        return;
+    }
+
+    /* create a surface for this scanout */
+    if (ss.scanout_id >= VIRTIO_GPU_MAX_SCANOUT ||
+        ss.scanout_id >= g->conf.max_outputs) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal scanout id specified %d",
+                      __func__, ss.scanout_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_SCANOUT_ID;
+        return;
+    }
+
+    res = virtio_gpu_find_resource(g, ss.resource_id);
+    if (!res) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal resource specified %d\n",
+                      __func__, ss.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+
+    if (ss.r.x > res->width ||
+        ss.r.y > res->height ||
+        ss.r.width > res->width ||
+        ss.r.height > res->height ||
+        ss.r.x + ss.r.width > res->width ||
+        ss.r.y + ss.r.height > res->height) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal scanout %d bounds for"
+                      " resource %d, (%d,%d)+%d,%d vs %d %d\n",
+                      __func__, ss.scanout_id, ss.resource_id, ss.r.x, ss.r.y,
+                      ss.r.width, ss.r.height, res->width, res->height);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_PARAMETER;
+        return;
+    }
+
+    scanout = &g->scanout[ss.scanout_id];
+
+    format = pixman_image_get_format(res->image);
+    bpp = (PIXMAN_FORMAT_BPP(format) + 7) / 8;
+    offset = (ss.r.x * bpp) + ss.r.y * pixman_image_get_stride(res->image);
+    if (!scanout->ds || surface_data(scanout->ds)
+        != ((uint8_t *)pixman_image_get_data(res->image) + offset) ||
+        scanout->width != ss.r.width ||
+        scanout->height != ss.r.height) {
+        /* realloc the surface ptr */
+        scanout->ds = qemu_create_displaysurface_from
+            (ss.r.width, ss.r.height, format,
+             pixman_image_get_stride(res->image),
+             (uint8_t *)pixman_image_get_data(res->image) + offset);
+        if (!scanout->ds) {
+            cmd->error = VIRTIO_GPU_RESP_ERR_UNSPEC;
+            return;
+        }
+        dpy_gfx_replace_surface(g->scanout[ss.scanout_id].con, scanout->ds);
+    }
+
+    res->scanout_bitmask |= (1 << ss.scanout_id);
+    scanout->resource_id = ss.resource_id;
+    scanout->x = ss.r.x;
+    scanout->y = ss.r.y;
+    scanout->width = ss.r.width;
+    scanout->height = ss.r.height;
+}
+
+int virtio_gpu_create_mapping_iov(struct virtio_gpu_resource_attach_backing *ab,
+                                  struct virtio_gpu_ctrl_command *cmd,
+                                  struct iovec **iov)
+{
+    struct virtio_gpu_mem_entry *ents;
+    size_t esize, s;
+    int i;
+
+    if (ab->nr_entries > 16384) {
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: nr_entries is too big (%d > 16384)\n",
+                      __func__, ab->nr_entries);
+        return -1;
+    }
+
+    esize = sizeof(*ents) * ab->nr_entries;
+    ents = g_malloc(esize);
+    s = iov_to_buf(cmd->elem.out_sg, cmd->elem.out_num,
+                   sizeof(*ab), ents, esize);
+    if (s != esize) {
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "%s: command data size incorrect %zu vs %zu\n",
+                      __func__, s, esize);
+        g_free(ents);
+        return -1;
+    }
+
+    *iov = g_malloc0(sizeof(struct iovec) * ab->nr_entries);
+    for (i = 0; i < ab->nr_entries; i++) {
+        hwaddr len = ents[i].length;
+        (*iov)[i].iov_len = ents[i].length;
+        (*iov)[i].iov_base = cpu_physical_memory_map(ents[i].addr, &len, 1);
+        if (!(*iov)[i].iov_base || len != ents[i].length) {
+            qemu_log_mask(LOG_GUEST_ERROR, "%s: failed to map MMIO memory for"
+                          " resource %d element %d\n",
+                          __func__, ab->resource_id, i);
+            virtio_gpu_cleanup_mapping_iov(*iov, i);
+            g_free(ents);
+            g_free(*iov);
+            *iov = NULL;
+            return -1;
+        }
+    }
+    g_free(ents);
+    return 0;
+}
+
+void virtio_gpu_cleanup_mapping_iov(struct iovec *iov, uint32_t count)
+{
+    int i;
+
+    for (i = 0; i < count; i++) {
+        cpu_physical_memory_unmap(iov[i].iov_base, iov[i].iov_len, 1,
+                                  iov[i].iov_len);
+    }
+}
+
+static void virtio_gpu_cleanup_mapping(struct virtio_gpu_simple_resource *res)
+{
+    virtio_gpu_cleanup_mapping_iov(res->iov, res->iov_cnt);
+    g_free(res->iov);
+    res->iov = NULL;
+    res->iov_cnt = 0;
+}
+
+static void
+virtio_gpu_resource_attach_backing(VirtIOGPU *g,
+                                   struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_simple_resource *res;
+    struct virtio_gpu_resource_attach_backing ab;
+    int ret;
+
+    VIRTIO_GPU_FILL_CMD(ab);
+    trace_virtio_gpu_cmd_res_back_attach(ab.resource_id);
+
+    res = virtio_gpu_find_resource(g, ab.resource_id);
+    if (!res) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal resource specified %d\n",
+                      __func__, ab.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+
+    ret = virtio_gpu_create_mapping_iov(&ab, cmd, &res->iov);
+    if (ret != 0) {
+        cmd->error = VIRTIO_GPU_RESP_ERR_UNSPEC;
+        return;
+    }
+
+    res->iov_cnt = ab.nr_entries;
+}
+
+static void
+virtio_gpu_resource_detach_backing(VirtIOGPU *g,
+                                   struct virtio_gpu_ctrl_command *cmd)
+{
+    struct virtio_gpu_simple_resource *res;
+    struct virtio_gpu_resource_detach_backing detach;
+
+    VIRTIO_GPU_FILL_CMD(detach);
+    trace_virtio_gpu_cmd_res_back_detach(detach.resource_id);
+
+    res = virtio_gpu_find_resource(g, detach.resource_id);
+    if (!res || !res->iov) {
+        qemu_log_mask(LOG_GUEST_ERROR, "%s: illegal resource specified %d\n",
+                      __func__, detach.resource_id);
+        cmd->error = VIRTIO_GPU_RESP_ERR_INVALID_RESOURCE_ID;
+        return;
+    }
+    virtio_gpu_cleanup_mapping(res);
+}
+
+static void virtio_gpu_simple_process_cmd(VirtIOGPU *g,
+                                          struct virtio_gpu_ctrl_command *cmd)
+{
+    VIRTIO_GPU_FILL_CMD(cmd->cmd_hdr);
+
+    switch (cmd->cmd_hdr.type) {
+    case VIRTIO_GPU_CMD_GET_DISPLAY_INFO:
+        virtio_gpu_get_display_info(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_RESOURCE_CREATE_2D:
+        virtio_gpu_resource_create_2d(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_RESOURCE_UNREF:
+        virtio_gpu_resource_unref(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_RESOURCE_FLUSH:
+        virtio_gpu_resource_flush(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_TRANSFER_TO_HOST_2D:
+        virtio_gpu_transfer_to_host_2d(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_SET_SCANOUT:
+        virtio_gpu_set_scanout(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_RESOURCE_ATTACH_BACKING:
+        virtio_gpu_resource_attach_backing(g, cmd);
+        break;
+    case VIRTIO_GPU_CMD_RESOURCE_DETACH_BACKING:
+        virtio_gpu_resource_detach_backing(g, cmd);
+        break;
+    default:
+        cmd->error = VIRTIO_GPU_RESP_ERR_UNSPEC;
+        break;
+    }
+    if (!cmd->finished) {
+        virtio_gpu_ctrl_response_nodata(g, cmd, cmd->error ? cmd->error :
+                                        VIRTIO_GPU_RESP_OK_NODATA);
+    }
+}
+
+static void virtio_gpu_handle_ctrl_cb(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    qemu_bh_schedule(g->ctrl_bh);
+}
+
+static void virtio_gpu_handle_cursor_cb(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    qemu_bh_schedule(g->cursor_bh);
+}
+
+static void virtio_gpu_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    struct virtio_gpu_ctrl_command *cmd;
+
+    if (!virtio_queue_ready(vq)) {
+        return;
+    }
+
+    cmd = g_new(struct virtio_gpu_ctrl_command, 1);
+    while (virtqueue_pop(vq, &cmd->elem)) {
+        cmd->vq = vq;
+        cmd->error = 0;
+        cmd->finished = false;
+        g->stats.requests++;
+
+        virtio_gpu_simple_process_cmd(g, cmd);
+        if (!cmd->finished) {
+            QTAILQ_INSERT_TAIL(&g->fenceq, cmd, next);
+            g->stats.inflight++;
+            if (g->stats.max_inflight < g->stats.inflight) {
+                g->stats.max_inflight = g->stats.inflight;
+            }
+            fprintf(stderr, "inflight: %3d (+)\r", g->stats.inflight);
+            cmd = g_new(struct virtio_gpu_ctrl_command, 1);
+        }
+    }
+    g_free(cmd);
+}
+
+static void virtio_gpu_ctrl_bh(void *opaque)
+{
+    VirtIOGPU *g = opaque;
+    virtio_gpu_handle_ctrl(&g->parent_obj, g->ctrl_vq);
+}
+
+static void virtio_gpu_handle_cursor(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    VirtQueueElement elem;
+    size_t s;
+    struct virtio_gpu_update_cursor cursor_info;
+
+    if (!virtio_queue_ready(vq)) {
+        return;
+    }
+    while (virtqueue_pop(vq, &elem)) {
+        s = iov_to_buf(elem.out_sg, elem.out_num, 0,
+                       &cursor_info, sizeof(cursor_info));
+        if (s != sizeof(cursor_info)) {
+            qemu_log_mask(LOG_GUEST_ERROR,
+                          "%s: cursor size incorrect %zu vs %zu\n",
+                          __func__, s, sizeof(cursor_info));
+        } else {
+            update_cursor(g, &cursor_info);
+        }
+        virtqueue_push(vq, &elem, 0);
+        virtio_notify(vdev, vq);
+    }
+}
+
+static void virtio_gpu_cursor_bh(void *opaque)
+{
+    VirtIOGPU *g = opaque;
+    virtio_gpu_handle_cursor(&g->parent_obj, g->cursor_vq);
+}
+
+static void virtio_gpu_invalidate_display(void *opaque)
+{
+}
+
+static void virtio_gpu_update_display(void *opaque)
+{
+}
+
+static void virtio_gpu_text_update(void *opaque, console_ch_t *chardata)
+{
+}
+
+static int virtio_gpu_ui_info(void *opaque, uint32_t idx, QemuUIInfo *info)
+{
+    VirtIOGPU *g = opaque;
+
+    if (idx > g->conf.max_outputs) {
+        return -1;
+    }
+
+    g->req_state[idx].x = info->xoff;
+    g->req_state[idx].y = info->yoff;
+    g->req_state[idx].width = info->width;
+    g->req_state[idx].height = info->height;
+
+    if (info->width && info->height) {
+        g->enabled_output_bitmask |= (1 << idx);
+    } else {
+        g->enabled_output_bitmask &= ~(1 << idx);
+    }
+
+    /* send event to guest */
+    virtio_gpu_notify_event(g, VIRTIO_GPU_EVENT_DISPLAY);
+    return 0;
+}
+
+const GraphicHwOps virtio_gpu_ops = {
+    .invalidate = virtio_gpu_invalidate_display,
+    .gfx_update = virtio_gpu_update_display,
+    .text_update = virtio_gpu_text_update,
+    .ui_info = virtio_gpu_ui_info,
+};
+
+static void virtio_gpu_device_realize(DeviceState *qdev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(qdev);
+    VirtIOGPU *g = VIRTIO_GPU(qdev);
+    int i;
+
+    g->config_size = sizeof(struct virtio_gpu_config);
+    g->virtio_config.num_scanouts = g->conf.max_outputs;
+    virtio_init(VIRTIO_DEVICE(g), "virtio-gpu", VIRTIO_ID_GPU,
+                g->config_size);
+
+    g->req_state[0].width = 1024;
+    g->req_state[0].height = 768;
+
+    g->ctrl_vq   = virtio_add_queue(vdev, 64, virtio_gpu_handle_ctrl_cb);
+    g->cursor_vq = virtio_add_queue(vdev, 16, virtio_gpu_handle_cursor_cb);
+
+    g->ctrl_bh = qemu_bh_new(virtio_gpu_ctrl_bh, g);
+    g->cursor_bh = qemu_bh_new(virtio_gpu_cursor_bh, g);
+    QTAILQ_INIT(&g->reslist);
+    QTAILQ_INIT(&g->fenceq);
+
+    g->enabled_output_bitmask = 1;
+    g->qdev = qdev;
+
+    for (i = 0; i < g->conf.max_outputs; i++) {
+        g->scanout[i].con =
+            graphic_console_init(DEVICE(g), i, &virtio_gpu_ops, g);
+        if (i > 0) {
+            dpy_gfx_replace_surface(g->scanout[i].con, NULL);
+        }
+    }
+}
+
+static void virtio_gpu_instance_init(Object *obj)
+{
+}
+
+static void virtio_gpu_reset(VirtIODevice *vdev)
+{
+    VirtIOGPU *g = VIRTIO_GPU(vdev);
+    struct virtio_gpu_simple_resource *res, *tmp;
+    int i;
+
+    g->enable = 0;
+
+    QTAILQ_FOREACH_SAFE(res, &g->reslist, next, tmp) {
+        virtio_gpu_resource_destroy(g, res);
+    }
+    for (i = 0; i < g->conf.max_outputs; i++) {
+#if 0
+        g->req_state[i].x = 0;
+        g->req_state[i].y = 0;
+        if (i == 0) {
+            g->req_state[0].width = 1024;
+            g->req_state[0].height = 768;
+        } else {
+            g->req_state[i].width = 0;
+            g->req_state[i].height = 0;
+        }
+#endif
+        g->scanout[i].resource_id = 0;
+        g->scanout[i].width = 0;
+        g->scanout[i].height = 0;
+        g->scanout[i].x = 0;
+        g->scanout[i].y = 0;
+        g->scanout[i].ds = NULL;
+    }
+    g->enabled_output_bitmask = 1;
+}
+
+static Property virtio_gpu_properties[] = {
+    DEFINE_VIRTIO_GPU_PROPERTIES(VirtIOGPU, conf),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_gpu_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
+    vdc->realize = virtio_gpu_device_realize;
+    vdc->get_config = virtio_gpu_get_config;
+    vdc->set_config = virtio_gpu_set_config;
+    vdc->get_features = virtio_gpu_get_features;
+
+    vdc->reset = virtio_gpu_reset;
+
+    dc->props = virtio_gpu_properties;
+}
+
+static const TypeInfo virtio_gpu_info = {
+    .name = TYPE_VIRTIO_GPU,
+    .parent = TYPE_VIRTIO_DEVICE,
+    .instance_size = sizeof(VirtIOGPU),
+    .instance_init = virtio_gpu_instance_init,
+    .class_init = virtio_gpu_class_init,
+};
+
+static void virtio_register_types(void)
+{
+    type_register_static(&virtio_gpu_info);
+}
+
+type_init(virtio_register_types)
+
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_ctrl_hdr)                != 24);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_update_cursor)           != 56);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_resource_unref)          != 32);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_resource_create_2d)      != 40);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_set_scanout)             != 48);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_resource_flush)          != 48);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_transfer_to_host_2d)     != 56);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_mem_entry)               != 16);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_resource_attach_backing) != 32);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_resource_detach_backing) != 32);
+QEMU_BUILD_BUG_ON(sizeof(struct virtio_gpu_resp_display_info)       != 408);
diff --git a/hw/display/virtio-vga.c b/hw/display/virtio-vga.c
new file mode 100644
index 0000000000..94f9d0eb5a
--- /dev/null
+++ b/hw/display/virtio-vga.c
@@ -0,0 +1,175 @@
+#include "hw/hw.h"
+#include "hw/pci/pci.h"
+#include "ui/console.h"
+#include "vga_int.h"
+#include "hw/virtio/virtio-pci.h"
+
+/*
+ * virtio-vga: This extends VirtioPCIProxy.
+ */
+#define TYPE_VIRTIO_VGA "virtio-vga"
+#define VIRTIO_VGA(obj) \
+        OBJECT_CHECK(VirtIOVGA, (obj), TYPE_VIRTIO_VGA)
+
+typedef struct VirtIOVGA {
+    VirtIOPCIProxy parent_obj;
+    VirtIOGPU      vdev;
+    VGACommonState vga;
+    MemoryRegion   vga_mrs[3];
+} VirtIOVGA;
+
+static void virtio_vga_invalidate_display(void *opaque)
+{
+    VirtIOVGA *vvga = opaque;
+
+    if (vvga->vdev.enable) {
+        virtio_gpu_ops.invalidate(&vvga->vdev);
+    } else {
+        vvga->vga.hw_ops->invalidate(&vvga->vga);
+    }
+}
+
+static void virtio_vga_update_display(void *opaque)
+{
+    VirtIOVGA *vvga = opaque;
+
+    if (vvga->vdev.enable) {
+        virtio_gpu_ops.gfx_update(&vvga->vdev);
+    } else {
+        vvga->vga.hw_ops->gfx_update(&vvga->vga);
+    }
+}
+
+static void virtio_vga_text_update(void *opaque, console_ch_t *chardata)
+{
+    VirtIOVGA *vvga = opaque;
+
+    if (vvga->vdev.enable) {
+        if (virtio_gpu_ops.text_update) {
+            virtio_gpu_ops.text_update(&vvga->vdev, chardata);
+        }
+    } else {
+        if (vvga->vga.hw_ops->text_update) {
+            vvga->vga.hw_ops->text_update(&vvga->vga, chardata);
+        }
+    }
+}
+
+static int virtio_vga_ui_info(void *opaque, uint32_t idx, QemuUIInfo *info)
+{
+    VirtIOVGA *vvga = opaque;
+
+    if (virtio_gpu_ops.ui_info) {
+        return virtio_gpu_ops.ui_info(&vvga->vdev, idx, info);
+    }
+    return -1;
+}
+
+static const GraphicHwOps virtio_vga_ops = {
+    .invalidate = virtio_vga_invalidate_display,
+    .gfx_update = virtio_vga_update_display,
+    .text_update = virtio_vga_text_update,
+    .ui_info = virtio_vga_ui_info,
+};
+
+/* VGA device wrapper around PCI device around virtio GPU */
+static void virtio_vga_realize(VirtIOPCIProxy *vpci_dev, Error **errp)
+{
+    VirtIOVGA *vvga = VIRTIO_VGA(vpci_dev);
+    VirtIOGPU *g = &vvga->vdev;
+    VGACommonState *vga = &vvga->vga;
+    uint32_t offset;
+
+    /* init vga compat bits */
+    vga->vram_size_mb = 8;
+    vga_common_init(vga, OBJECT(vpci_dev), false);
+    vga_init(vga, OBJECT(vpci_dev), pci_address_space(&vpci_dev->pci_dev),
+             pci_address_space_io(&vpci_dev->pci_dev), true);
+    pci_register_bar(&vpci_dev->pci_dev, 0,
+                     PCI_BASE_ADDRESS_MEM_PREFETCH, &vga->vram);
+
+    /*
+     * Configure virtio bar and regions
+     *
+     * We use bar #2 for the mmio regions, to be compatible with stdvga.
+     * virtio regions are moved to the end of bar #2, to make room for
+     * the stdvga mmio registers at the start of bar #2.
+     */
+    vpci_dev->modern_mem_bar = 2;
+    vpci_dev->msix_bar = 4;
+    offset = memory_region_size(&vpci_dev->modern_bar);
+    offset -= vpci_dev->notify.size;
+    vpci_dev->notify.offset = offset;
+    offset -= vpci_dev->device.size;
+    vpci_dev->device.offset = offset;
+    offset -= vpci_dev->isr.size;
+    vpci_dev->isr.offset = offset;
+    offset -= vpci_dev->common.size;
+    vpci_dev->common.offset = offset;
+
+    /* init virtio bits */
+    qdev_set_parent_bus(DEVICE(g), BUS(&vpci_dev->bus));
+    /* force virtio-1.0 */
+    vpci_dev->flags &= ~VIRTIO_PCI_FLAG_DISABLE_MODERN;
+    vpci_dev->flags |= VIRTIO_PCI_FLAG_DISABLE_LEGACY;
+    object_property_set_bool(OBJECT(g), true, "realized", errp);
+
+    /* add stdvga mmio regions */
+    pci_std_vga_mmio_region_init(vga, &vpci_dev->modern_bar,
+                                 vvga->vga_mrs, true);
+
+    vga->con = g->scanout[0].con;
+    graphic_console_set_hwops(vga->con, &virtio_vga_ops, vvga);
+}
+
+static void virtio_vga_reset(DeviceState *dev)
+{
+    VirtIOVGA *vvga = VIRTIO_VGA(dev);
+    vvga->vdev.enable = 0;
+
+    vga_dirty_log_start(&vvga->vga);
+}
+
+static Property virtio_vga_properties[] = {
+    DEFINE_VIRTIO_GPU_PROPERTIES(VirtIOVGA, vdev.conf),
+    DEFINE_VIRTIO_GPU_PCI_PROPERTIES(VirtIOPCIProxy),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_vga_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioPCIClass *k = VIRTIO_PCI_CLASS(klass);
+    PCIDeviceClass *pcidev_k = PCI_DEVICE_CLASS(klass);
+
+    set_bit(DEVICE_CATEGORY_DISPLAY, dc->categories);
+    dc->props = virtio_vga_properties;
+    dc->reset = virtio_vga_reset;
+    dc->hotpluggable = false;
+
+    k->realize = virtio_vga_realize;
+    pcidev_k->romfile = "vgabios-virtio.bin";
+    pcidev_k->class_id = PCI_CLASS_DISPLAY_VGA;
+}
+
+static void virtio_vga_inst_initfn(Object *obj)
+{
+    VirtIOVGA *dev = VIRTIO_VGA(obj);
+    object_initialize(&dev->vdev, sizeof(dev->vdev), TYPE_VIRTIO_GPU);
+    object_property_add_child(obj, "virtio-backend", OBJECT(&dev->vdev), NULL);
+}
+
+static TypeInfo virtio_vga_info = {
+    .name          = TYPE_VIRTIO_VGA,
+    .parent        = TYPE_VIRTIO_PCI,
+    .instance_size = sizeof(struct VirtIOVGA),
+    .instance_init = virtio_vga_inst_initfn,
+    .class_init    = virtio_vga_class_init,
+};
+
+static void virtio_vga_register_types(void)
+{
+    type_register_static(&virtio_vga_info);
+}
+
+type_init(virtio_vga_register_types)
diff --git a/hw/display/vmware_vga.c b/hw/display/vmware_vga.c
index c17ddd1fcd..7f397d3c2e 100644
--- a/hw/display/vmware_vga.c
+++ b/hw/display/vmware_vga.c
@@ -1124,7 +1124,7 @@ static void vmsvga_update_display(void *opaque)
      * Is it more efficient to look at vram VGA-dirty bits or wait
      * for the driver to issue SVGA_CMD_UPDATE?
      */
-    if (memory_region_is_logging(&s->vga.vram)) {
+    if (memory_region_is_logging(&s->vga.vram, DIRTY_MEMORY_VGA)) {
         vga_sync_dirty_bitmap(&s->vga);
         dirty = memory_region_get_dirty(&s->vga.vram, 0,
             surface_stride(surface) * surface_height(surface),
diff --git a/hw/dma/rc4030.c b/hw/dma/rc4030.c
index af2663256e..3efa6de352 100644
--- a/hw/dma/rc4030.c
+++ b/hw/dma/rc4030.c
@@ -1,7 +1,7 @@
 /*
  * QEMU JAZZ RC4030 chipset
  *
- * Copyright (c) 2007-2009 Herve Poussineau
+ * Copyright (c) 2007-2013 Hervé Poussineau
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -24,29 +24,16 @@
 
 #include "hw/hw.h"
 #include "hw/mips/mips.h"
+#include "hw/sysbus.h"
 #include "qemu/timer.h"
-
-/********************************************************/
-/* debug rc4030 */
-
-//#define DEBUG_RC4030
-//#define DEBUG_RC4030_DMA
-
-#ifdef DEBUG_RC4030
-#define DPRINTF(fmt, ...) \
-do { printf("rc4030: " fmt , ## __VA_ARGS__); } while (0)
-static const char* irq_names[] = { "parallel", "floppy", "sound", "video",
-            "network", "scsi", "keyboard", "mouse", "serial0", "serial1" };
-#else
-#define DPRINTF(fmt, ...)
-#endif
-
-#define RC4030_ERROR(fmt, ...) \
-do { fprintf(stderr, "rc4030 ERROR: %s: " fmt, __func__ , ## __VA_ARGS__); } while (0)
+#include "exec/address-spaces.h"
+#include "trace.h"
 
 /********************************************************/
 /* rc4030 emulation                                     */
 
+#define MAX_TL_ENTRIES 512
+
 typedef struct dma_pagetable_entry {
     int32_t frame;
     int32_t owner;
@@ -63,8 +50,14 @@ typedef struct dma_pagetable_entry {
 #define DMA_FLAG_MEM_INTR   0x0200
 #define DMA_FLAG_ADDR_INTR  0x0400
 
+#define TYPE_RC4030 "rc4030"
+#define RC4030(obj) \
+    OBJECT_CHECK(rc4030State, (obj), TYPE_RC4030)
+
 typedef struct rc4030State
 {
+    SysBusDevice parent;
+
     uint32_t config; /* 0x0000: RC4030 config register */
     uint32_t revision; /* 0x0008: RC4030 Revision register */
     uint32_t invalid_address_register; /* 0x0010: Invalid Address register */
@@ -83,7 +76,7 @@ typedef struct rc4030State
     uint32_t cache_bmask; /* 0x0058: I/O Cache Byte Mask */
 
     uint32_t nmi_interrupt; /* 0x0200: interrupt source */
-    uint32_t offset210;
+    uint32_t memory_refresh_rate; /* 0x0210: memory refresh rate */
     uint32_t nvram_protect; /* 0x0220: NV ram protect register */
     uint32_t rem_speed[16];
     uint32_t imr_jazz; /* Local bus int enable mask */
@@ -96,6 +89,16 @@ typedef struct rc4030State
     qemu_irq timer_irq;
     qemu_irq jazz_bus_irq;
 
+    /* biggest translation table */
+    MemoryRegion dma_tt;
+    /* translation table memory region alias, added to system RAM */
+    MemoryRegion dma_tt_alias;
+    /* whole DMA memory region, root of DMA address space */
+    MemoryRegion dma_mr;
+    /* translation table entry aliases, added to DMA memory region */
+    MemoryRegion dma_mrs[MAX_TL_ENTRIES];
+    AddressSpace dma_as;
+
     MemoryRegion iomem_chipset;
     MemoryRegion iomem_jazzio;
 } rc4030State;
@@ -112,7 +115,7 @@ static void set_next_tick(rc4030State *s)
 }
 
 /* called for accesses to rc4030 */
-static uint32_t rc4030_readl(void *opaque, hwaddr addr)
+static uint64_t rc4030_read(void *opaque, hwaddr addr, unsigned int size)
 {
     rc4030State *s = opaque;
     uint32_t val;
@@ -220,9 +223,9 @@ static uint32_t rc4030_readl(void *opaque, hwaddr addr)
     case 0x0208:
         val = 0;
         break;
-    /* Offset 0x0210 */
+    /* Memory refresh rate */
     case 0x0210:
-        val = s->offset210;
+        val = s->memory_refresh_rate;
         break;
     /* NV ram protect register */
     case 0x0220:
@@ -238,39 +241,117 @@ static uint32_t rc4030_readl(void *opaque, hwaddr addr)
         val = 7; /* FIXME: should be read from EISA controller */
         break;
     default:
-        RC4030_ERROR("invalid read [" TARGET_FMT_plx "]\n", addr);
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "rc4030: invalid read at 0x%x", (int)addr);
         val = 0;
         break;
     }
 
     if ((addr & ~3) != 0x230) {
-        DPRINTF("read 0x%02x at " TARGET_FMT_plx "\n", val, addr);
+        trace_rc4030_read(addr, val);
     }
 
     return val;
 }
 
-static uint32_t rc4030_readw(void *opaque, hwaddr addr)
+static void rc4030_dma_as_update_one(rc4030State *s, int index, uint32_t frame)
 {
-    uint32_t v = rc4030_readl(opaque, addr & ~0x3);
-    if (addr & 0x2)
-        return v >> 16;
-    else
-        return v & 0xffff;
+    if (index < MAX_TL_ENTRIES) {
+        memory_region_set_enabled(&s->dma_mrs[index], false);
+    }
+
+    if (!frame) {
+        return;
+    }
+
+    if (index >= MAX_TL_ENTRIES) {
+        qemu_log_mask(LOG_UNIMP,
+                      "rc4030: trying to use too high "
+                      "translation table entry %d (max allowed=%d)",
+                      index, MAX_TL_ENTRIES);
+        return;
+    }
+    memory_region_set_alias_offset(&s->dma_mrs[index], frame);
+    memory_region_set_enabled(&s->dma_mrs[index], true);
 }
 
-static uint32_t rc4030_readb(void *opaque, hwaddr addr)
+static void rc4030_dma_tt_write(void *opaque, hwaddr addr, uint64_t data,
+                                unsigned int size)
 {
-    uint32_t v = rc4030_readl(opaque, addr & ~0x3);
-    return (v >> (8 * (addr & 0x3))) & 0xff;
+    rc4030State *s = opaque;
+
+    /* write memory */
+    memcpy(memory_region_get_ram_ptr(&s->dma_tt) + addr, &data, size);
+
+    /* update dma address space (only if frame field has been written) */
+    if (addr % sizeof(dma_pagetable_entry) == 0) {
+        int index = addr / sizeof(dma_pagetable_entry);
+        memory_region_transaction_begin();
+        rc4030_dma_as_update_one(s, index, (uint32_t)data);
+        memory_region_transaction_commit();
+    }
 }
 
-static void rc4030_writel(void *opaque, hwaddr addr, uint32_t val)
+static const MemoryRegionOps rc4030_dma_tt_ops = {
+    .write = rc4030_dma_tt_write,
+    .impl.min_access_size = 4,
+    .impl.max_access_size = 4,
+};
+
+static void rc4030_dma_tt_update(rc4030State *s, uint32_t new_tl_base,
+                                 uint32_t new_tl_limit)
+{
+    int entries, i;
+    dma_pagetable_entry *dma_tl_contents;
+
+    if (s->dma_tl_limit) {
+        /* write old dma tl table to physical memory */
+        memory_region_del_subregion(get_system_memory(), &s->dma_tt_alias);
+        cpu_physical_memory_write(s->dma_tl_limit & 0x7fffffff,
+                                  memory_region_get_ram_ptr(&s->dma_tt),
+                                  memory_region_size(&s->dma_tt_alias));
+    }
+    object_unparent(OBJECT(&s->dma_tt_alias));
+
+    s->dma_tl_base = new_tl_base;
+    s->dma_tl_limit = new_tl_limit;
+    new_tl_base &= 0x7fffffff;
+
+    if (s->dma_tl_limit) {
+        uint64_t dma_tt_size;
+        if (s->dma_tl_limit <= memory_region_size(&s->dma_tt)) {
+            dma_tt_size = s->dma_tl_limit;
+        } else {
+            dma_tt_size = memory_region_size(&s->dma_tt);
+        }
+        memory_region_init_alias(&s->dma_tt_alias, OBJECT(s),
+                                 "dma-table-alias",
+                                 &s->dma_tt, 0, dma_tt_size);
+        dma_tl_contents = memory_region_get_ram_ptr(&s->dma_tt);
+        cpu_physical_memory_read(new_tl_base, dma_tl_contents, dma_tt_size);
+
+        memory_region_transaction_begin();
+        entries = dma_tt_size / sizeof(dma_pagetable_entry);
+        for (i = 0; i < entries; i++) {
+            rc4030_dma_as_update_one(s, i, dma_tl_contents[i].frame);
+        }
+        memory_region_add_subregion(get_system_memory(), new_tl_base,
+                                    &s->dma_tt_alias);
+        memory_region_transaction_commit();
+    } else {
+        memory_region_init(&s->dma_tt_alias, OBJECT(s),
+                           "dma-table-alias", 0);
+    }
+}
+
+static void rc4030_write(void *opaque, hwaddr addr, uint64_t data,
+                         unsigned int size)
 {
     rc4030State *s = opaque;
+    uint32_t val = data;
     addr &= 0x3fff;
 
-    DPRINTF("write 0x%02x at " TARGET_FMT_plx "\n", val, addr);
+    trace_rc4030_write(addr, val);
 
     switch (addr & ~0x3) {
     /* Global config register */
@@ -279,11 +360,11 @@ static void rc4030_writel(void *opaque, hwaddr addr, uint32_t val)
         break;
     /* DMA transl. table base */
     case 0x0018:
-        s->dma_tl_base = val;
+        rc4030_dma_tt_update(s, val, s->dma_tl_limit);
         break;
     /* DMA transl. table limit */
     case 0x0020:
-        s->dma_tl_limit = val;
+        rc4030_dma_tt_update(s, s->dma_tl_base, val);
         break;
     /* DMA transl. table invalidated */
     case 0x0028:
@@ -371,9 +452,9 @@ static void rc4030_writel(void *opaque, hwaddr addr, uint32_t val)
             s->dma_regs[entry][idx] = val;
         }
         break;
-    /* Offset 0x0210 */
+    /* Memory refresh rate */
     case 0x0210:
-        s->offset210 = val;
+        s->memory_refresh_rate = val;
         break;
     /* Interval timer reload */
     case 0x0228:
@@ -385,48 +466,18 @@ static void rc4030_writel(void *opaque, hwaddr addr, uint32_t val)
     case 0x0238:
         break;
     default:
-        RC4030_ERROR("invalid write of 0x%02x at [" TARGET_FMT_plx "]\n", val, addr);
-        break;
-    }
-}
-
-static void rc4030_writew(void *opaque, hwaddr addr, uint32_t val)
-{
-    uint32_t old_val = rc4030_readl(opaque, addr & ~0x3);
-
-    if (addr & 0x2)
-        val = (val << 16) | (old_val & 0x0000ffff);
-    else
-        val = val | (old_val & 0xffff0000);
-    rc4030_writel(opaque, addr & ~0x3, val);
-}
-
-static void rc4030_writeb(void *opaque, hwaddr addr, uint32_t val)
-{
-    uint32_t old_val = rc4030_readl(opaque, addr & ~0x3);
-
-    switch (addr & 3) {
-    case 0:
-        val = val | (old_val & 0xffffff00);
-        break;
-    case 1:
-        val = (val << 8) | (old_val & 0xffff00ff);
-        break;
-    case 2:
-        val = (val << 16) | (old_val & 0xff00ffff);
-        break;
-    case 3:
-        val = (val << 24) | (old_val & 0x00ffffff);
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "rc4030: invalid write of 0x%02x at 0x%x",
+                      val, (int)addr);
         break;
     }
-    rc4030_writel(opaque, addr & ~0x3, val);
 }
 
 static const MemoryRegionOps rc4030_ops = {
-    .old_mmio = {
-        .read = { rc4030_readb, rc4030_readw, rc4030_readl, },
-        .write = { rc4030_writeb, rc4030_writew, rc4030_writel, },
-    },
+    .read = rc4030_read,
+    .write = rc4030_write,
+    .impl.min_access_size = 4,
+    .impl.max_access_size = 4,
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
@@ -436,22 +487,6 @@ static void update_jazz_irq(rc4030State *s)
 
     pending = s->isr_jazz & s->imr_jazz;
 
-#ifdef DEBUG_RC4030
-    if (s->isr_jazz != 0) {
-        uint32_t irq = 0;
-        DPRINTF("pending irqs:");
-        for (irq = 0; irq < ARRAY_SIZE(irq_names); irq++) {
-            if (s->isr_jazz & (1 << irq)) {
-                printf(" %s", irq_names[irq]);
-                if (!(s->imr_jazz & (1 << irq))) {
-                    printf("(ignored)");
-                }
-            }
-        }
-        printf("\n");
-    }
-#endif
-
     if (pending != 0)
         qemu_irq_raise(s->jazz_bus_irq);
     else
@@ -479,7 +514,7 @@ static void rc4030_periodic_timer(void *opaque)
     qemu_irq_raise(s->timer_irq);
 }
 
-static uint32_t jazzio_readw(void *opaque, hwaddr addr)
+static uint64_t jazzio_read(void *opaque, hwaddr addr, unsigned int size)
 {
     rc4030State *s = opaque;
     uint32_t val;
@@ -494,7 +529,6 @@ static uint32_t jazzio_readw(void *opaque, hwaddr addr)
         irq = 0;
         while (pending) {
             if (pending & 1) {
-                DPRINTF("returning irq %s\n", irq_names[irq]);
                 val = (irq + 1) << 2;
                 break;
             }
@@ -508,36 +542,25 @@ static uint32_t jazzio_readw(void *opaque, hwaddr addr)
         val = s->imr_jazz;
         break;
     default:
-        RC4030_ERROR("(jazz io controller) invalid read [" TARGET_FMT_plx "]\n", addr);
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "rc4030/jazzio: invalid read at 0x%x", (int)addr);
         val = 0;
+        break;
     }
 
-    DPRINTF("(jazz io controller) read 0x%04x at " TARGET_FMT_plx "\n", val, addr);
+    trace_jazzio_read(addr, val);
 
     return val;
 }
 
-static uint32_t jazzio_readb(void *opaque, hwaddr addr)
-{
-    uint32_t v;
-    v = jazzio_readw(opaque, addr & ~0x1);
-    return (v >> (8 * (addr & 0x1))) & 0xff;
-}
-
-static uint32_t jazzio_readl(void *opaque, hwaddr addr)
-{
-    uint32_t v;
-    v = jazzio_readw(opaque, addr);
-    v |= jazzio_readw(opaque, addr + 2) << 16;
-    return v;
-}
-
-static void jazzio_writew(void *opaque, hwaddr addr, uint32_t val)
+static void jazzio_write(void *opaque, hwaddr addr, uint64_t data,
+                         unsigned int size)
 {
     rc4030State *s = opaque;
+    uint32_t val = data;
     addr &= 0xfff;
 
-    DPRINTF("(jazz io controller) write 0x%04x at " TARGET_FMT_plx "\n", val, addr);
+    trace_jazzio_write(addr, val);
 
     switch (addr) {
     /* Local bus int enable mask */
@@ -546,43 +569,24 @@ static void jazzio_writew(void *opaque, hwaddr addr, uint32_t val)
         update_jazz_irq(s);
         break;
     default:
-        RC4030_ERROR("(jazz io controller) invalid write of 0x%04x at [" TARGET_FMT_plx "]\n", val, addr);
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "rc4030/jazzio: invalid write of 0x%02x at 0x%x",
+                      val, (int)addr);
         break;
     }
 }
 
-static void jazzio_writeb(void *opaque, hwaddr addr, uint32_t val)
-{
-    uint32_t old_val = jazzio_readw(opaque, addr & ~0x1);
-
-    switch (addr & 1) {
-    case 0:
-        val = val | (old_val & 0xff00);
-        break;
-    case 1:
-        val = (val << 8) | (old_val & 0x00ff);
-        break;
-    }
-    jazzio_writew(opaque, addr & ~0x1, val);
-}
-
-static void jazzio_writel(void *opaque, hwaddr addr, uint32_t val)
-{
-    jazzio_writew(opaque, addr, val & 0xffff);
-    jazzio_writew(opaque, addr + 2, (val >> 16) & 0xffff);
-}
-
 static const MemoryRegionOps jazzio_ops = {
-    .old_mmio = {
-        .read = { jazzio_readb, jazzio_readw, jazzio_readl, },
-        .write = { jazzio_writeb, jazzio_writew, jazzio_writel, },
-    },
+    .read = jazzio_read,
+    .write = jazzio_write,
+    .impl.min_access_size = 2,
+    .impl.max_access_size = 2,
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
-static void rc4030_reset(void *opaque)
+static void rc4030_reset(DeviceState *dev)
 {
-    rc4030State *s = opaque;
+    rc4030State *s = RC4030(dev);
     int i;
 
     s->config = 0x410; /* some boards seem to accept 0x104 too */
@@ -590,14 +594,14 @@ static void rc4030_reset(void *opaque)
     s->invalid_address_register = 0;
 
     memset(s->dma_regs, 0, sizeof(s->dma_regs));
-    s->dma_tl_base = s->dma_tl_limit = 0;
+    rc4030_dma_tt_update(s, 0, 0);
 
     s->remote_failed_address = s->memory_failed_address = 0;
     s->cache_maint = 0;
     s->cache_ptag = s->cache_ltag = 0;
     s->cache_bmask = 0;
 
-    s->offset210 = 0x18186;
+    s->memory_refresh_rate = 0x18186;
     s->nvram_protect = 7;
     for (i = 0; i < 15; i++)
         s->rem_speed[i] = 7;
@@ -631,7 +635,7 @@ static int rc4030_load(QEMUFile *f, void *opaque, int version_id)
     s->cache_ptag = qemu_get_be32(f);
     s->cache_ltag = qemu_get_be32(f);
     s->cache_bmask = qemu_get_be32(f);
-    s->offset210 = qemu_get_be32(f);
+    s->memory_refresh_rate = qemu_get_be32(f);
     s->nvram_protect = qemu_get_be32(f);
     for (i = 0; i < 15; i++)
         s->rem_speed[i] = qemu_get_be32(f);
@@ -663,7 +667,7 @@ static void rc4030_save(QEMUFile *f, void *opaque)
     qemu_put_be32(f, s->cache_ptag);
     qemu_put_be32(f, s->cache_ltag);
     qemu_put_be32(f, s->cache_bmask);
-    qemu_put_be32(f, s->offset210);
+    qemu_put_be32(f, s->memory_refresh_rate);
     qemu_put_be32(f, s->nvram_protect);
     for (i = 0; i < 15; i++)
         qemu_put_be32(f, s->rem_speed[i]);
@@ -672,44 +676,6 @@ static void rc4030_save(QEMUFile *f, void *opaque)
     qemu_put_be32(f, s->itr);
 }
 
-void rc4030_dma_memory_rw(void *opaque, hwaddr addr, uint8_t *buf, int len, int is_write)
-{
-    rc4030State *s = opaque;
-    hwaddr entry_addr;
-    hwaddr phys_addr;
-    dma_pagetable_entry entry;
-    int index;
-    int ncpy, i;
-
-    i = 0;
-    for (;;) {
-        if (i == len) {
-            break;
-        }
-
-        ncpy = DMA_PAGESIZE - (addr & (DMA_PAGESIZE - 1));
-        if (ncpy > len - i)
-            ncpy = len - i;
-
-        /* Get DMA translation table entry */
-        index = addr / DMA_PAGESIZE;
-        if (index >= s->dma_tl_limit / sizeof(dma_pagetable_entry)) {
-            break;
-        }
-        entry_addr = s->dma_tl_base + index * sizeof(dma_pagetable_entry);
-        /* XXX: not sure. should we really use only lowest bits? */
-        entry_addr &= 0x7fffffff;
-        cpu_physical_memory_read(entry_addr, &entry, sizeof(entry));
-
-        /* Read/write data at right place */
-        phys_addr = entry.frame + (addr & (DMA_PAGESIZE - 1));
-        cpu_physical_memory_rw(phys_addr, &buf[i], ncpy, is_write);
-
-        i += ncpy;
-        addr += ncpy;
-    }
-}
-
 static void rc4030_do_dma(void *opaque, int n, uint8_t *buf, int len, int is_write)
 {
     rc4030State *s = opaque;
@@ -733,32 +699,11 @@ static void rc4030_do_dma(void *opaque, int n, uint8_t *buf, int len, int is_wri
     dma_addr = s->dma_regs[n][DMA_REG_ADDRESS];
 
     /* Read/write data at right place */
-    rc4030_dma_memory_rw(opaque, dma_addr, buf, len, is_write);
+    address_space_rw(&s->dma_as, dma_addr, MEMTXATTRS_UNSPECIFIED,
+                     buf, len, is_write);
 
     s->dma_regs[n][DMA_REG_ENABLE] |= DMA_FLAG_TC_INTR;
     s->dma_regs[n][DMA_REG_COUNT] -= len;
-
-#ifdef DEBUG_RC4030_DMA
-    {
-        int i, j;
-        printf("rc4030 dma: Copying %d bytes %s host %p\n",
-            len, is_write ? "from" : "to", buf);
-        for (i = 0; i < len; i += 16) {
-            int n = 16;
-            if (n > len - i) {
-                n = len - i;
-            }
-            for (j = 0; j < n; j++)
-                printf("%02x ", buf[i + j]);
-            while (j++ < 16)
-                printf("   ");
-            printf("| ");
-            for (j = 0; j < n; j++)
-                printf("%c", isprint(buf[i + j]) ? buf[i + j] : '.');
-            printf("\n");
-        }
-    }
-#endif
 }
 
 struct rc4030DMAState {
@@ -795,31 +740,102 @@ static rc4030_dma *rc4030_allocate_dmas(void *opaque, int n)
     return s;
 }
 
-void *rc4030_init(qemu_irq timer, qemu_irq jazz_bus,
-                  qemu_irq **irqs, rc4030_dma **dmas,
-                  MemoryRegion *sysmem)
+static void rc4030_initfn(Object *obj)
 {
-    rc4030State *s;
+    DeviceState *dev = DEVICE(obj);
+    rc4030State *s = RC4030(obj);
+    SysBusDevice *sysbus = SYS_BUS_DEVICE(obj);
 
-    s = g_malloc0(sizeof(rc4030State));
+    qdev_init_gpio_in(dev, rc4030_irq_jazz_request, 16);
 
-    *irqs = qemu_allocate_irqs(rc4030_irq_jazz_request, s, 16);
-    *dmas = rc4030_allocate_dmas(s, 4);
+    sysbus_init_irq(sysbus, &s->timer_irq);
+    sysbus_init_irq(sysbus, &s->jazz_bus_irq);
 
-    s->periodic_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, rc4030_periodic_timer, s);
-    s->timer_irq = timer;
-    s->jazz_bus_irq = jazz_bus;
-
-    qemu_register_reset(rc4030_reset, s);
     register_savevm(NULL, "rc4030", 0, 2, rc4030_save, rc4030_load, s);
-    rc4030_reset(s);
+
+    sysbus_init_mmio(sysbus, &s->iomem_chipset);
+    sysbus_init_mmio(sysbus, &s->iomem_jazzio);
+}
+
+static void rc4030_realize(DeviceState *dev, Error **errp)
+{
+    rc4030State *s = RC4030(dev);
+    Object *o = OBJECT(dev);
+    int i;
+
+    s->periodic_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL,
+                                     rc4030_periodic_timer, s);
 
     memory_region_init_io(&s->iomem_chipset, NULL, &rc4030_ops, s,
                           "rc4030.chipset", 0x300);
-    memory_region_add_subregion(sysmem, 0x80000000, &s->iomem_chipset);
     memory_region_init_io(&s->iomem_jazzio, NULL, &jazzio_ops, s,
                           "rc4030.jazzio", 0x00001000);
-    memory_region_add_subregion(sysmem, 0xf0000000, &s->iomem_jazzio);
 
-    return s;
+    memory_region_init_rom_device(&s->dma_tt, o,
+                                  &rc4030_dma_tt_ops, s, "dma-table",
+                                  MAX_TL_ENTRIES * sizeof(dma_pagetable_entry),
+                                  NULL);
+    memory_region_init(&s->dma_tt_alias, o, "dma-table-alias", 0);
+    memory_region_init(&s->dma_mr, o, "dma", INT32_MAX);
+    for (i = 0; i < MAX_TL_ENTRIES; ++i) {
+        memory_region_init_alias(&s->dma_mrs[i], o, "dma-alias",
+                                 get_system_memory(), 0, DMA_PAGESIZE);
+        memory_region_set_enabled(&s->dma_mrs[i], false);
+        memory_region_add_subregion(&s->dma_mr, i * DMA_PAGESIZE,
+                                    &s->dma_mrs[i]);
+    }
+    address_space_init(&s->dma_as, &s->dma_mr, "rc4030-dma");
+}
+
+static void rc4030_unrealize(DeviceState *dev, Error **errp)
+{
+    rc4030State *s = RC4030(dev);
+    int i;
+
+    timer_free(s->periodic_timer);
+
+    address_space_destroy(&s->dma_as);
+    object_unparent(OBJECT(&s->dma_tt));
+    object_unparent(OBJECT(&s->dma_tt_alias));
+    object_unparent(OBJECT(&s->dma_mr));
+    for (i = 0; i < MAX_TL_ENTRIES; ++i) {
+        memory_region_del_subregion(&s->dma_mr, &s->dma_mrs[i]);
+        object_unparent(OBJECT(&s->dma_mrs[i]));
+    }
+}
+
+static void rc4030_class_init(ObjectClass *klass, void *class_data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+
+    dc->realize = rc4030_realize;
+    dc->unrealize = rc4030_unrealize;
+    dc->reset = rc4030_reset;
+}
+
+static const TypeInfo rc4030_info = {
+    .name = TYPE_RC4030,
+    .parent = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(rc4030State),
+    .instance_init = rc4030_initfn,
+    .class_init = rc4030_class_init,
+};
+
+static void rc4030_register_types(void)
+{
+    type_register_static(&rc4030_info);
+}
+
+type_init(rc4030_register_types)
+
+DeviceState *rc4030_init(rc4030_dma **dmas, MemoryRegion **dma_mr)
+{
+    DeviceState *dev;
+
+    dev = qdev_create(NULL, TYPE_RC4030);
+    qdev_init_nofail(dev);
+
+    *dmas = rc4030_allocate_dmas(dev, 4);
+    *dma_mr = &RC4030(dev)->dma_mr;
+    return dev;
 }
diff --git a/hw/gpio/pl061.c b/hw/gpio/pl061.c
index bd03e99975..4ba730b476 100644
--- a/hw/gpio/pl061.c
+++ b/hw/gpio/pl061.c
@@ -173,7 +173,7 @@ static uint64_t pl061_read(void *opaque, hwaddr offset,
     case 0x414: /* Raw interrupt status */
         return s->istate;
     case 0x418: /* Masked interrupt status */
-        return s->istate | s->im;
+        return s->istate & s->im;
     case 0x420: /* Alternate function select */
         return s->afsel;
     case 0x500: /* 2mA drive */
diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
index 8e88aded9f..b71e942567 100644
--- a/hw/i386/acpi-build.c
+++ b/hw/i386/acpi-build.c
@@ -615,6 +615,7 @@ static void build_append_pci_bus_devices(Aml *parent_scope, PCIBus *bus,
         }
     }
     aml_append(parent_scope, method);
+    qobject_decref(bsel);
 }
 
 /*
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index a93972fb41..3f0d435da9 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -164,27 +164,6 @@ uint64_t cpu_get_tsc(CPUX86State *env)
     return cpu_get_ticks();
 }
 
-/* SMM support */
-
-static cpu_set_smm_t smm_set;
-static void *smm_arg;
-
-void cpu_smm_register(cpu_set_smm_t callback, void *arg)
-{
-    assert(smm_set == NULL);
-    assert(smm_arg == NULL);
-    smm_set = callback;
-    smm_arg = arg;
-}
-
-void cpu_smm_update(CPUX86State *env)
-{
-    if (smm_set && smm_arg && CPU(x86_env_get_cpu(env)) == first_cpu) {
-        smm_set(!!(env->hflags & HF_SMM_MASK), smm_arg);
-    }
-}
-
-
 /* IRQ handling */
 int cpu_get_pic_interrupt(CPUX86State *env)
 {
@@ -1007,7 +986,6 @@ static X86CPU *pc_new_cpu(const char *cpu_model, int64_t apic_id,
     }
 
     qdev_set_parent_bus(DEVICE(cpu), qdev_get_child_bus(icc_bridge, "icc"));
-    object_unref(OBJECT(cpu));
 
     object_property_set_int(OBJECT(cpu), apic_id, "apic-id", &local_err);
     object_property_set_bool(OBJECT(cpu), true, "realized", &local_err);
@@ -1026,7 +1004,9 @@ static const char *current_cpu_model;
 void pc_hot_add_cpu(const int64_t id, Error **errp)
 {
     DeviceState *icc_bridge;
+    X86CPU *cpu;
     int64_t apic_id = x86_cpu_apic_id_from_index(id);
+    Error *local_err = NULL;
 
     if (id < 0) {
         error_setg(errp, "Invalid CPU id: %" PRIi64, id);
@@ -1054,7 +1034,12 @@ void pc_hot_add_cpu(const int64_t id, Error **errp)
 
     icc_bridge = DEVICE(object_resolve_path_type("icc-bridge",
                                                  TYPE_ICC_BRIDGE, NULL));
-    pc_new_cpu(current_cpu_model, apic_id, icc_bridge, errp);
+    cpu = pc_new_cpu(current_cpu_model, apic_id, icc_bridge, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+    object_unref(OBJECT(cpu));
 }
 
 void pc_cpus_init(const char *cpu_model, DeviceState *icc_bridge)
@@ -1088,6 +1073,7 @@ void pc_cpus_init(const char *cpu_model, DeviceState *icc_bridge)
             error_report_err(error);
             exit(1);
         }
+        object_unref(OBJECT(cpu));
     }
 
     /* map APIC MMIO area if CPU has APIC */
@@ -1365,9 +1351,9 @@ FWCfgState *pc_memory_init(MachineState *machine,
     return fw_cfg;
 }
 
-qemu_irq *pc_allocate_cpu_irq(void)
+qemu_irq pc_allocate_cpu_irq(void)
 {
-    return qemu_allocate_irqs(pic_irq_request, NULL, 1);
+    return qemu_allocate_irq(pic_irq_request, NULL, 0);
 }
 
 DeviceState *pc_vga_init(ISABus *isa_bus, PCIBus *pci_bus)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 0688f0c52f..e142f75649 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -52,6 +52,7 @@
 #ifdef CONFIG_XEN
 #  include <xen/hvm/hvm_info_table.h>
 #endif
+#include "migration/migration.h"
 
 #define MAX_IDE_BUS 2
 
@@ -86,10 +87,9 @@ static void pc_init1(MachineState *machine)
     ISABus *isa_bus;
     PCII440FXState *i440fx_state;
     int piix3_devfn = -1;
-    qemu_irq *cpu_irq;
     qemu_irq *gsi;
     qemu_irq *i8259;
-    qemu_irq *smi_irq;
+    qemu_irq smi_irq;
     GSIState *gsi_state;
     DriveInfo *hd[MAX_IDE_BUS * MAX_IDE_DEVS];
     BusState *idebus[MAX_IDE_BUS];
@@ -219,13 +219,13 @@ static void pc_init1(MachineState *machine)
     } else if (xen_enabled()) {
         i8259 = xen_interrupt_controller_init();
     } else {
-        cpu_irq = pc_allocate_cpu_irq();
-        i8259 = i8259_init(isa_bus, cpu_irq[0]);
+        i8259 = i8259_init(isa_bus, pc_allocate_cpu_irq());
     }
 
     for (i = 0; i < ISA_NUM_IRQS; i++) {
         gsi_state->i8259_irq[i] = i8259[i];
     }
+    g_free(i8259);
     if (pci_enabled) {
         ioapic_init_gsi(gsi_state, "i440fx");
     }
@@ -283,10 +283,10 @@ static void pc_init1(MachineState *machine)
         DeviceState *piix4_pm;
         I2CBus *smbus;
 
-        smi_irq = qemu_allocate_irqs(pc_acpi_smi_interrupt, first_cpu, 1);
+        smi_irq = qemu_allocate_irq(pc_acpi_smi_interrupt, first_cpu, 0);
         /* TODO: Populate SPD eeprom data.  */
         smbus = piix4_pm_init(pci_bus, piix3_devfn + 3, 0xb100,
-                              gsi[9], *smi_irq,
+                              gsi[9], smi_irq,
                               kvm_enabled(), &piix4_pm);
         smbus_eeprom_init(smbus, 8, NULL, 0);
 
@@ -306,6 +306,7 @@ static void pc_init1(MachineState *machine)
 
 static void pc_compat_2_3(MachineState *machine)
 {
+    savevm_skip_section_footers();
 }
 
 static void pc_compat_2_2(MachineState *machine)
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index 66220b352b..082cd93bb2 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -45,6 +45,7 @@
 #include "hw/usb.h"
 #include "hw/cpu/icc_bus.h"
 #include "qemu/error-report.h"
+#include "migration/migration.h"
 
 /* ICH9 AHCI has 6 ports */
 #define MAX_SATA_PORTS     6
@@ -79,7 +80,6 @@ static void pc_q35_init(MachineState *machine)
     GSIState *gsi_state;
     ISABus *isa_bus;
     int pci_enabled = 1;
-    qemu_irq *cpu_irq;
     qemu_irq *gsi;
     qemu_irq *i8259;
     int i;
@@ -230,8 +230,7 @@ static void pc_q35_init(MachineState *machine)
     } else if (xen_enabled()) {
         i8259 = xen_interrupt_controller_init();
     } else {
-        cpu_irq = pc_allocate_cpu_irq();
-        i8259 = i8259_init(isa_bus, cpu_irq[0]);
+        i8259 = i8259_init(isa_bus, pc_allocate_cpu_irq());
     }
 
     for (i = 0; i < ISA_NUM_IRQS; i++) {
@@ -291,6 +290,7 @@ static void pc_q35_init(MachineState *machine)
 
 static void pc_compat_2_3(MachineState *machine)
 {
+    savevm_skip_section_footers();
 }
 
 static void pc_compat_2_2(MachineState *machine)
@@ -403,6 +403,7 @@ DEFINE_Q35_MACHINE(v2_4, "pc-q35-2.4", NULL,
 static void pc_q35_2_3_machine_options(MachineClass *m)
 {
     pc_q35_2_4_machine_options(m);
+    m->no_floppy = 0;
     m->alias = NULL;
     SET_MACHINE_COMPAT(m, PC_COMPAT_2_3);
 }
diff --git a/hw/ide/core.c b/hw/ide/core.c
index fcb908061c..1efd98af63 100644
--- a/hw/ide/core.c
+++ b/hw/ide/core.c
@@ -2561,6 +2561,7 @@ static const VMStateDescription vmstate_ide_atapi_gesn_state = {
     .name ="ide_drive/atapi/gesn_state",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = ide_atapi_gesn_needed,
     .fields = (VMStateField[]) {
         VMSTATE_BOOL(events.new_media, IDEState),
         VMSTATE_BOOL(events.eject_request, IDEState),
@@ -2572,6 +2573,7 @@ static const VMStateDescription vmstate_ide_tray_state = {
     .name = "ide_drive/tray_state",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = ide_tray_state_needed,
     .fields = (VMStateField[]) {
         VMSTATE_BOOL(tray_open, IDEState),
         VMSTATE_BOOL(tray_locked, IDEState),
@@ -2585,6 +2587,7 @@ static const VMStateDescription vmstate_ide_drive_pio_state = {
     .minimum_version_id = 1,
     .pre_save = ide_drive_pio_pre_save,
     .post_load = ide_drive_pio_post_load,
+    .needed = ide_drive_pio_state_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(req_nb_sectors, IDEState),
         VMSTATE_VARRAY_INT32(io_buffer, IDEState, io_buffer_total_len, 1,
@@ -2626,19 +2629,11 @@ const VMStateDescription vmstate_ide_drive = {
         VMSTATE_UINT8_V(cdrom_changed, IDEState, 3),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_ide_drive_pio_state,
-            .needed = ide_drive_pio_state_needed,
-        }, {
-            .vmsd = &vmstate_ide_tray_state,
-            .needed = ide_tray_state_needed,
-        }, {
-            .vmsd = &vmstate_ide_atapi_gesn_state,
-            .needed = ide_atapi_gesn_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_ide_drive_pio_state,
+        &vmstate_ide_tray_state,
+        &vmstate_ide_atapi_gesn_state,
+        NULL
     }
 };
 
@@ -2646,6 +2641,7 @@ static const VMStateDescription vmstate_ide_error_status = {
     .name ="ide_bus/error",
     .version_id = 2,
     .minimum_version_id = 1,
+    .needed = ide_error_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(error_status, IDEBus),
         VMSTATE_INT64_V(retry_sector_num, IDEBus, 2),
@@ -2664,13 +2660,9 @@ const VMStateDescription vmstate_ide_bus = {
         VMSTATE_UINT8(unit, IDEBus),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_ide_error_status,
-            .needed = ide_error_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_ide_error_status,
+        NULL
     }
 };
 
diff --git a/hw/ide/macio.c b/hw/ide/macio.c
index 585a27bd6c..dd52d50732 100644
--- a/hw/ide/macio.c
+++ b/hw/ide/macio.c
@@ -51,8 +51,15 @@ static const int debug_macio = 0;
 
 #define MACIO_PAGE_SIZE 4096
 
+/*
+ * Unaligned DMA read/write access functions required for OS X/Darwin which
+ * don't perform DMA transactions on sector boundaries. These functions are
+ * modelled on bdrv_co_do_preadv()/bdrv_co_do_pwritev() and so should be
+ * easy to remove if the unaligned block APIs are ever exposed.
+ */
+
 static void pmac_dma_read(BlockBackend *blk,
-                          int64_t sector_num, int nb_sectors,
+                          int64_t offset, unsigned int bytes,
                           void (*cb)(void *opaque, int ret), void *opaque)
 {
     DBDMA_io *io = opaque;
@@ -60,76 +67,48 @@ static void pmac_dma_read(BlockBackend *blk,
     IDEState *s = idebus_active_if(&m->bus);
     dma_addr_t dma_addr, dma_len;
     void *mem;
-    int nsector, remainder;
+    int64_t sector_num;
+    int nsector;
+    uint64_t align = BDRV_SECTOR_SIZE;
+    size_t head_bytes, tail_bytes;
 
     qemu_iovec_destroy(&io->iov);
     qemu_iovec_init(&io->iov, io->len / MACIO_PAGE_SIZE + 1);
 
-    if (io->remainder_len > 0) {
-        /* Return remainder of request */
-        int transfer = MIN(io->remainder_len, io->len);
+    sector_num = (offset >> 9);
+    nsector = (io->len >> 9);
 
-        MACIO_DPRINTF("--- DMA read pop     - bounce addr: %p addr: %"
-                      HWADDR_PRIx " remainder_len: %x\n",
-                      &io->remainder + (0x200 - transfer), io->addr,
-                      io->remainder_len);
+    MACIO_DPRINTF("--- DMA read transfer (0x%" HWADDR_PRIx ",0x%x): "
+                  "sector_num: %" PRId64 ", nsector: %d\n", io->addr, io->len,
+                  sector_num, nsector);
 
-        cpu_physical_memory_write(io->addr,
-                                  &io->remainder + (0x200 - transfer),
-                                  transfer);
+    dma_addr = io->addr;
+    dma_len = io->len;
+    mem = dma_memory_map(&address_space_memory, dma_addr, &dma_len,
+                         DMA_DIRECTION_FROM_DEVICE);
 
-        io->remainder_len -= transfer;
-        io->len -= transfer;
-        io->addr += transfer;
+    if (offset & (align - 1)) {
+        head_bytes = offset & (align - 1);
 
-        s->io_buffer_index += transfer;
-        s->io_buffer_size -= transfer;
+        MACIO_DPRINTF("--- DMA unaligned head: sector %" PRId64 ", "
+                      "discarding %zu bytes\n", sector_num, head_bytes);
 
-        if (io->remainder_len != 0) {
-            /* Still waiting for remainder */
-            return;
-        }
+        qemu_iovec_add(&io->iov, &io->head_remainder, head_bytes);
 
-        if (io->len == 0) {
-            MACIO_DPRINTF("--- finished all read processing; go and finish\n");
-            cb(opaque, 0);
-            return;
-        }
+        bytes += offset & (align - 1);
+        offset = offset & ~(align - 1);
     }
 
-    if (s->drive_kind == IDE_CD) {
-        sector_num = (int64_t)(s->lba << 2) + (s->io_buffer_index >> 9);
-    } else {
-        sector_num = ide_get_sector(s) + (s->io_buffer_index >> 9);
-    }
+    qemu_iovec_add(&io->iov, mem, io->len);
 
-    nsector = ((io->len + 0x1ff) >> 9);
-    remainder = (nsector << 9) - io->len;
+    if ((offset + bytes) & (align - 1)) {
+        tail_bytes = (offset + bytes) & (align - 1);
 
-    MACIO_DPRINTF("--- DMA read transfer - addr: %" HWADDR_PRIx " len: %x\n",
-                  io->addr, io->len);
+        MACIO_DPRINTF("--- DMA unaligned tail: sector %" PRId64 ", "
+                      "discarding bytes %zu\n", sector_num, tail_bytes);
 
-    dma_addr = io->addr;
-    dma_len = io->len;
-    mem = dma_memory_map(&address_space_memory, dma_addr, &dma_len,
-                         DMA_DIRECTION_FROM_DEVICE);
-
-    if (!remainder) {
-        MACIO_DPRINTF("--- DMA read aligned - addr: %" HWADDR_PRIx
-                      " len: %x\n", io->addr, io->len);
-        qemu_iovec_add(&io->iov, mem, io->len);
-    } else {
-        MACIO_DPRINTF("--- DMA read unaligned - addr: %" HWADDR_PRIx
-                      " len: %x\n", io->addr, io->len);
-        qemu_iovec_add(&io->iov, mem, io->len);
-
-        MACIO_DPRINTF("--- DMA read push    - bounce addr: %p "
-                      "remainder_len: %x\n",
-                      &io->remainder + 0x200 - remainder, remainder);
-        qemu_iovec_add(&io->iov, &io->remainder + 0x200 - remainder,
-                       remainder);
-
-        io->remainder_len = remainder;
+        qemu_iovec_add(&io->iov, &io->tail_remainder, align - tail_bytes);
+        bytes = ROUND_UP(bytes, align);
     }
 
     s->io_buffer_size -= io->len;
@@ -137,15 +116,15 @@ static void pmac_dma_read(BlockBackend *blk,
 
     io->len = 0;
 
-    MACIO_DPRINTF("--- Block read transfer   - sector_num: %"PRIx64"  "
-                  "nsector: %x\n",
-                  sector_num, nsector);
+    MACIO_DPRINTF("--- Block read transfer - sector_num: %" PRIx64 "  "
+                  "nsector: %x\n", (offset >> 9), (bytes >> 9));
 
-    m->aiocb = blk_aio_readv(blk, sector_num, &io->iov, nsector, cb, io);
+    m->aiocb = blk_aio_readv(blk, (offset >> 9), &io->iov, (bytes >> 9),
+                             cb, io);
 }
 
 static void pmac_dma_write(BlockBackend *blk,
-                         int64_t sector_num, int nb_sectors,
+                         int64_t offset, int bytes,
                          void (*cb)(void *opaque, int ret), void *opaque)
 {
     DBDMA_io *io = opaque;
@@ -153,90 +132,80 @@ static void pmac_dma_write(BlockBackend *blk,
     IDEState *s = idebus_active_if(&m->bus);
     dma_addr_t dma_addr, dma_len;
     void *mem;
-    int nsector, remainder;
-    int extra = 0;
+    int64_t sector_num;
+    int nsector;
+    uint64_t align = BDRV_SECTOR_SIZE;
+    size_t head_bytes, tail_bytes;
+    bool unaligned_head = false, unaligned_tail = false;
 
     qemu_iovec_destroy(&io->iov);
     qemu_iovec_init(&io->iov, io->len / MACIO_PAGE_SIZE + 1);
 
-    if (io->remainder_len > 0) {
-        /* Return remainder of request */
-        int transfer = MIN(io->remainder_len, io->len);
+    sector_num = (offset >> 9);
+    nsector = (io->len >> 9);
 
-        MACIO_DPRINTF("--- processing write remainder %x\n", transfer);
-        cpu_physical_memory_read(io->addr,
-                                 &io->remainder + (0x200 - transfer),
-                                 transfer);
+    MACIO_DPRINTF("--- DMA write transfer (0x%" HWADDR_PRIx ",0x%x): "
+                  "sector_num: %" PRId64 ", nsector: %d\n", io->addr, io->len,
+                  sector_num, nsector);
 
-        io->remainder_len -= transfer;
-        io->len -= transfer;
-        io->addr += transfer;
+    dma_addr = io->addr;
+    dma_len = io->len;
+    mem = dma_memory_map(&address_space_memory, dma_addr, &dma_len,
+                         DMA_DIRECTION_TO_DEVICE);
 
-        s->io_buffer_index += transfer;
-        s->io_buffer_size -= transfer;
+    if (offset & (align - 1)) {
+        head_bytes = offset & (align - 1);
+        sector_num = ((offset & ~(align - 1)) >> 9);
 
-        if (io->remainder_len != 0) {
-            /* Still waiting for remainder */
-            return;
-        }
+        MACIO_DPRINTF("--- DMA unaligned head: pre-reading head sector %"
+                      PRId64 "\n", sector_num);
 
-        MACIO_DPRINTF("--> prepending bounce buffer with size 0x200\n");
+        blk_pread(s->blk, (sector_num << 9), &io->head_remainder, align);
 
-        /* Sector transfer complete - prepend to request */
-        qemu_iovec_add(&io->iov, &io->remainder, 0x200);
-        extra = 1;
-    }
+        qemu_iovec_add(&io->iov, &io->head_remainder, head_bytes);
+        qemu_iovec_add(&io->iov, mem, io->len);
 
-    if (s->drive_kind == IDE_CD) {
-        sector_num = (int64_t)(s->lba << 2) + (s->io_buffer_index >> 9);
-    } else {
-        sector_num = ide_get_sector(s) + (s->io_buffer_index >> 9);
+        bytes += offset & (align - 1);
+        offset = offset & ~(align - 1);
+
+        unaligned_head = true;
     }
 
-    nsector = (io->len >> 9);
-    remainder = io->len - (nsector << 9);
+    if ((offset + bytes) & (align - 1)) {
+        tail_bytes = (offset + bytes) & (align - 1);
+        sector_num = (((offset + bytes) & ~(align - 1)) >> 9);
 
-    MACIO_DPRINTF("--- DMA write transfer - addr: %" HWADDR_PRIx " len: %x\n",
-                  io->addr, io->len);
-    MACIO_DPRINTF("xxx remainder: %x\n", remainder);
-    MACIO_DPRINTF("xxx sector_num: %"PRIx64"   nsector: %x\n",
-                  sector_num, nsector);
+        MACIO_DPRINTF("--- DMA unaligned tail: pre-reading tail sector %"
+                      PRId64 "\n", sector_num);
 
-    dma_addr = io->addr;
-    dma_len = io->len;
-    mem = dma_memory_map(&address_space_memory, dma_addr, &dma_len,
-                         DMA_DIRECTION_TO_DEVICE);
+        blk_pread(s->blk, (sector_num << 9), &io->tail_remainder, align);
 
-    if (!remainder) {
-        MACIO_DPRINTF("--- DMA write aligned - addr: %" HWADDR_PRIx
-                      " len: %x\n", io->addr, io->len);
-        qemu_iovec_add(&io->iov, mem, io->len);
-    } else {
-        /* Write up to last complete sector */
-        MACIO_DPRINTF("--- DMA write unaligned - addr: %" HWADDR_PRIx
-                      " len: %x\n", io->addr, (nsector << 9));
-        qemu_iovec_add(&io->iov, mem, (nsector << 9));
+        if (!unaligned_head) {
+            qemu_iovec_add(&io->iov, mem, io->len);
+        }
+
+        qemu_iovec_add(&io->iov, &io->tail_remainder + tail_bytes,
+                       align - tail_bytes);
 
-        MACIO_DPRINTF("--- DMA write read    - bounce addr: %p "
-                      "remainder_len: %x\n", &io->remainder, remainder);
-        cpu_physical_memory_read(io->addr + (nsector << 9), &io->remainder,
-                                 remainder);
+        bytes = ROUND_UP(bytes, align);
 
-        io->remainder_len = 0x200 - remainder;
+        unaligned_tail = true;
+    }
 
-        MACIO_DPRINTF("xxx remainder_len: %x\n", io->remainder_len);
+    if (!unaligned_head && !unaligned_tail) {
+        qemu_iovec_add(&io->iov, mem, io->len);
     }
 
-    s->io_buffer_size -= ((nsector + extra) << 9);
-    s->io_buffer_index += ((nsector + extra) << 9);
+    s->io_buffer_size -= io->len;
+    s->io_buffer_index += io->len;
 
     io->len = 0;
 
-    MACIO_DPRINTF("--- Block write transfer   - sector_num: %"PRIx64"  "
-                  "nsector: %x\n", sector_num, nsector + extra);
+    MACIO_DPRINTF("--- Block write transfer - sector_num: %" PRIx64 "  "
+                  "nsector: %x\n", (offset >> 9), (bytes >> 9));
 
-    m->aiocb = blk_aio_writev(blk, sector_num, &io->iov, nsector + extra, cb,
-                              io);
+    m->aiocb = blk_aio_writev(blk, (offset >> 9), &io->iov, (bytes >> 9),
+                              cb, io);
 }
 
 static void pmac_ide_atapi_transfer_cb(void *opaque, int ret)
@@ -244,19 +213,12 @@ static void pmac_ide_atapi_transfer_cb(void *opaque, int ret)
     DBDMA_io *io = opaque;
     MACIOIDEState *m = io->opaque;
     IDEState *s = idebus_active_if(&m->bus);
-    int64_t sector_num;
-    int nsector, remainder;
+    int64_t offset;
 
-    MACIO_DPRINTF("\ns is %p\n", s);
-    MACIO_DPRINTF("io_buffer_index: %x\n", s->io_buffer_index);
-    MACIO_DPRINTF("io_buffer_size: %x   packet_transfer_size: %x\n",
-                  s->io_buffer_size, s->packet_transfer_size);
-    MACIO_DPRINTF("lba: %x\n", s->lba);
-    MACIO_DPRINTF("io_addr: %" HWADDR_PRIx "  io_len: %x\n", io->addr,
-                  io->len);
+    MACIO_DPRINTF("pmac_ide_atapi_transfer_cb\n");
 
     if (ret < 0) {
-        MACIO_DPRINTF("THERE WAS AN ERROR!  %d\n", ret);
+        MACIO_DPRINTF("DMA error: %d\n", ret);
         ide_atapi_io_error(s, ret);
         goto done;
     }
@@ -270,6 +232,7 @@ static void pmac_ide_atapi_transfer_cb(void *opaque, int ret)
     }
 
     if (s->io_buffer_size <= 0) {
+        MACIO_DPRINTF("End of IDE transfer\n");
         ide_atapi_cmd_ok(s);
         m->dma_active = false;
         goto done;
@@ -289,19 +252,13 @@ static void pmac_ide_atapi_transfer_cb(void *opaque, int ret)
         goto done;
     }
 
-    /* Calculate number of sectors */
-    sector_num = (int64_t)(s->lba << 2) + (s->io_buffer_index >> 9);
-    nsector = (io->len + 0x1ff) >> 9;
-    remainder = io->len & 0x1ff;
-
-    MACIO_DPRINTF("nsector: %d   remainder: %x\n", nsector, remainder);
-    MACIO_DPRINTF("sector: %"PRIx64"   %zx\n", sector_num, io->iov.size / 512);
+    /* Calculate current offset */
+    offset = (int64_t)(s->lba << 11) + s->io_buffer_index;
 
-    pmac_dma_read(s->blk, sector_num, nsector, pmac_ide_atapi_transfer_cb, io);
+    pmac_dma_read(s->blk, offset, io->len, pmac_ide_atapi_transfer_cb, io);
     return;
 
 done:
-    MACIO_DPRINTF("done DMA\n\n");
     block_acct_done(blk_get_stats(s->blk), &s->acct);
     io->dma_end(opaque);
 
@@ -313,16 +270,14 @@ static void pmac_ide_transfer_cb(void *opaque, int ret)
     DBDMA_io *io = opaque;
     MACIOIDEState *m = io->opaque;
     IDEState *s = idebus_active_if(&m->bus);
-    int64_t sector_num;
-    int nsector, remainder;
+    int64_t offset;
 
     MACIO_DPRINTF("pmac_ide_transfer_cb\n");
 
     if (ret < 0) {
-        MACIO_DPRINTF("DMA error\n");
+        MACIO_DPRINTF("DMA error: %d\n", ret);
         m->aiocb = NULL;
         ide_dma_error(s);
-        io->remainder_len = 0;
         goto done;
     }
 
@@ -335,7 +290,7 @@ static void pmac_ide_transfer_cb(void *opaque, int ret)
     }
 
     if (s->io_buffer_size <= 0) {
-        MACIO_DPRINTF("end of transfer\n");
+        MACIO_DPRINTF("End of IDE transfer\n");
         s->status = READY_STAT | SEEK_STAT;
         ide_set_irq(s->bus);
         m->dma_active = false;
@@ -348,24 +303,16 @@ static void pmac_ide_transfer_cb(void *opaque, int ret)
     }
 
     /* Calculate number of sectors */
-    sector_num = ide_get_sector(s) + (s->io_buffer_index >> 9);
-    nsector = (io->len + 0x1ff) >> 9;
-    remainder = io->len & 0x1ff;
-
-    s->nsector -= nsector;
-
-    MACIO_DPRINTF("nsector: %d   remainder: %x\n", nsector, remainder);
-    MACIO_DPRINTF("sector: %"PRIx64"   %x\n", sector_num, nsector);
+    offset = (ide_get_sector(s) << 9) + s->io_buffer_index;
 
     switch (s->dma_cmd) {
     case IDE_DMA_READ:
-        pmac_dma_read(s->blk, sector_num, nsector, pmac_ide_transfer_cb, io);
+        pmac_dma_read(s->blk, offset, io->len, pmac_ide_transfer_cb, io);
         break;
     case IDE_DMA_WRITE:
-        pmac_dma_write(s->blk, sector_num, nsector, pmac_ide_transfer_cb, io);
+        pmac_dma_write(s->blk, offset, io->len, pmac_ide_transfer_cb, io);
         break;
     case IDE_DMA_TRIM:
-        MACIO_DPRINTF("TRIM command issued!");
         break;
     }
 
@@ -561,15 +508,12 @@ static void ide_dbdma_start(IDEDMA *dma, IDEState *s,
                             BlockCompletionFunc *cb)
 {
     MACIOIDEState *m = container_of(dma, MACIOIDEState, dma);
-    DBDMAState *dbdma = m->dbdma;
-    DBDMA_io *io;
-    int i;
 
     s->io_buffer_index = 0;
     if (s->drive_kind == IDE_CD) {
         s->io_buffer_size = s->packet_transfer_size;
     } else {
-        s->io_buffer_size = s->nsector * 0x200;
+        s->io_buffer_size = s->nsector * BDRV_SECTOR_SIZE;
     }
 
     MACIO_DPRINTF("\n\n------------ IDE transfer\n");
@@ -578,15 +522,6 @@ static void ide_dbdma_start(IDEDMA *dma, IDEState *s,
     MACIO_DPRINTF("lba: %x    size: %x\n", s->lba, s->io_buffer_size);
     MACIO_DPRINTF("-------------------------\n");
 
-    for (i = 0; i < DBDMA_CHANNELS; i++) {
-        io = &dbdma->channels[i].io;
-
-        if (io->opaque == m) {
-            io->remainder_len = 0;
-        }
-    }
-
-    MACIO_DPRINTF("\n");
     m->dma_active = true;
     DBDMA_kick(m->dbdma);
 }
diff --git a/hw/ide/pci.c b/hw/ide/pci.c
index 1b3d1c12ad..4afd0cfe8c 100644
--- a/hw/ide/pci.c
+++ b/hw/ide/pci.c
@@ -350,6 +350,7 @@ static const VMStateDescription vmstate_bmdma_current = {
     .name = "ide bmdma_current",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = ide_bmdma_current_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT32(cur_addr, BMDMAState),
         VMSTATE_UINT32(cur_prd_last, BMDMAState),
@@ -363,6 +364,7 @@ static const VMStateDescription vmstate_bmdma_status = {
     .name ="ide bmdma/status",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = ide_bmdma_status_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(status, BMDMAState),
         VMSTATE_END_OF_LIST()
@@ -383,16 +385,10 @@ static const VMStateDescription vmstate_bmdma = {
         VMSTATE_UINT8(migration_retry_unit, BMDMAState),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_bmdma_current,
-            .needed = ide_bmdma_current_needed,
-        }, {
-            .vmsd = &vmstate_bmdma_status,
-            .needed = ide_bmdma_status_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_bmdma_current,
+        &vmstate_bmdma_status,
+        NULL
     }
 };
 
@@ -452,8 +448,6 @@ static const struct IDEDMAOps bmdma_ops = {
 
 void bmdma_init(IDEBus *bus, BMDMAState *bm, PCIIDEState *d)
 {
-    qemu_irq *irq;
-
     if (bus->dma == &bm->dma) {
         return;
     }
@@ -461,8 +455,7 @@ void bmdma_init(IDEBus *bus, BMDMAState *bm, PCIIDEState *d)
     bm->dma.ops = &bmdma_ops;
     bus->dma = &bm->dma;
     bm->irq = bus->irq;
-    irq = qemu_allocate_irqs(bmdma_irq, bm, 1);
-    bus->irq = *irq;
+    bus->irq = qemu_allocate_irq(bmdma_irq, bm, 0);
     bm->pci_dev = d;
 }
 
diff --git a/hw/input/pckbd.c b/hw/input/pckbd.c
index 9b9a7d7a8a..ddac69df6f 100644
--- a/hw/input/pckbd.c
+++ b/hw/input/pckbd.c
@@ -391,23 +391,24 @@ static int kbd_outport_post_load(void *opaque, int version_id)
     return 0;
 }
 
+static bool kbd_outport_needed(void *opaque)
+{
+    KBDState *s = opaque;
+    return s->outport != kbd_outport_default(s);
+}
+
 static const VMStateDescription vmstate_kbd_outport = {
     .name = "pckbd_outport",
     .version_id = 1,
     .minimum_version_id = 1,
     .post_load = kbd_outport_post_load,
+    .needed = kbd_outport_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(outport, KBDState),
         VMSTATE_END_OF_LIST()
     }
 };
 
-static bool kbd_outport_needed(void *opaque)
-{
-    KBDState *s = opaque;
-    return s->outport != kbd_outport_default(s);
-}
-
 static int kbd_post_load(void *opaque, int version_id)
 {
     KBDState *s = opaque;
@@ -430,12 +431,9 @@ static const VMStateDescription vmstate_kbd = {
         VMSTATE_UINT8(pending, KBDState),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_kbd_outport,
-            .needed = kbd_outport_needed,
-        },
-        VMSTATE_END_OF_LIST()
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_kbd_outport,
+        NULL
     }
 };
 
diff --git a/hw/input/ps2.c b/hw/input/ps2.c
index 4baeea2b56..fdbe565e62 100644
--- a/hw/input/ps2.c
+++ b/hw/input/ps2.c
@@ -677,6 +677,7 @@ static const VMStateDescription vmstate_ps2_keyboard_ledstate = {
     .version_id = 3,
     .minimum_version_id = 2,
     .post_load = ps2_kbd_ledstate_post_load,
+    .needed = ps2_keyboard_ledstate_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(ledstate, PS2KbdState),
         VMSTATE_END_OF_LIST()
@@ -717,13 +718,9 @@ static const VMStateDescription vmstate_ps2_keyboard = {
         VMSTATE_INT32_V(scancode_set, PS2KbdState,3),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_ps2_keyboard_ledstate,
-            .needed = ps2_keyboard_ledstate_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_ps2_keyboard_ledstate,
+        NULL
     }
 };
 
diff --git a/hw/intc/Makefile.objs b/hw/intc/Makefile.objs
index 843864a3ef..092d8a80ac 100644
--- a/hw/intc/Makefile.objs
+++ b/hw/intc/Makefile.objs
@@ -11,6 +11,7 @@ common-obj-$(CONFIG_SLAVIO) += slavio_intctl.o
 common-obj-$(CONFIG_IOAPIC) += ioapic_common.o
 common-obj-$(CONFIG_ARM_GIC) += arm_gic_common.o
 common-obj-$(CONFIG_ARM_GIC) += arm_gic.o
+common-obj-$(CONFIG_ARM_GIC) += arm_gicv2m.o
 common-obj-$(CONFIG_OPENPIC) += openpic.o
 
 obj-$(CONFIG_APIC) += apic.o apic_common.o
diff --git a/hw/intc/apic.c b/hw/intc/apic.c
index 0f97b47925..77b639cce8 100644
--- a/hw/intc/apic.c
+++ b/hw/intc/apic.c
@@ -370,13 +370,14 @@ static int apic_irq_pending(APICCommonState *s)
 static void apic_update_irq(APICCommonState *s)
 {
     CPUState *cpu;
+    DeviceState *dev = (DeviceState *)s;
 
     cpu = CPU(s->cpu);
     if (!qemu_cpu_is_self(cpu)) {
         cpu_interrupt(cpu, CPU_INTERRUPT_POLL);
     } else if (apic_irq_pending(s) > 0) {
         cpu_interrupt(cpu, CPU_INTERRUPT_HARD);
-    } else if (!apic_accept_pic_intr(&s->busdev.qdev) || !pic_get_output(isa_pic)) {
+    } else if (!apic_accept_pic_intr(dev) || !pic_get_output(isa_pic)) {
         cpu_reset_interrupt(cpu, CPU_INTERRUPT_HARD);
     }
 }
@@ -549,10 +550,12 @@ static void apic_deliver(DeviceState *dev, uint8_t dest, uint8_t dest_mode,
 
 static bool apic_check_pic(APICCommonState *s)
 {
-    if (!apic_accept_pic_intr(&s->busdev.qdev) || !pic_get_output(isa_pic)) {
+    DeviceState *dev = (DeviceState *)s;
+
+    if (!apic_accept_pic_intr(dev) || !pic_get_output(isa_pic)) {
         return false;
     }
-    apic_deliver_pic_intr(&s->busdev.qdev, 1);
+    apic_deliver_pic_intr(dev, 1);
     return true;
 }
 
diff --git a/hw/intc/apic_common.c b/hw/intc/apic_common.c
index d595d63a51..0032b97c5f 100644
--- a/hw/intc/apic_common.c
+++ b/hw/intc/apic_common.c
@@ -369,6 +369,7 @@ static const VMStateDescription vmstate_apic_common_sipi = {
     .name = "apic_sipi",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = apic_common_sipi_needed,
     .fields = (VMStateField[]) {
         VMSTATE_INT32(sipi_vector, APICCommonState),
         VMSTATE_INT32(wait_for_sipi, APICCommonState),
@@ -408,12 +409,9 @@ static const VMStateDescription vmstate_apic_common = {
                       APICCommonState), /* open-coded timer state */
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_apic_common_sipi,
-            .needed = apic_common_sipi_needed,
-        },
-        VMSTATE_END_OF_LIST()
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_apic_common_sipi,
+        NULL
     }
 };
 
diff --git a/hw/intc/arm_gic.c b/hw/intc/arm_gic.c
index c1d2e704ec..454bfd7df5 100644
--- a/hw/intc/arm_gic.c
+++ b/hw/intc/arm_gic.c
@@ -71,7 +71,7 @@ void gic_update(GICState *s)
             || !(s->cpu_ctlr[cpu] & (GICC_CTLR_EN_GRP0 | GICC_CTLR_EN_GRP1))) {
             qemu_irq_lower(s->parent_irq[cpu]);
             qemu_irq_lower(s->parent_fiq[cpu]);
-            return;
+            continue;
         }
         best_prio = 0x100;
         best_irq = 1023;
diff --git a/hw/intc/arm_gicv2m.c b/hw/intc/arm_gicv2m.c
new file mode 100644
index 0000000000..43d1976c49
--- /dev/null
+++ b/hw/intc/arm_gicv2m.c
@@ -0,0 +1,192 @@
+/*
+ *  GICv2m extension for MSI/MSI-x support with a GICv2-based system
+ *
+ * Copyright (C) 2015 Linaro, All rights reserved.
+ *
+ * Author: Christoffer Dall <christoffer.dall@linaro.org>
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* This file implements an emulated GICv2m widget as described in the ARM
+ * Server Base System Architecture (SBSA) specification Version 2.2
+ * (ARM-DEN-0029 v2.2) pages 35-39 without any optional implementation defined
+ * identification registers and with a single non-secure MSI register frame.
+ */
+
+#include "hw/sysbus.h"
+#include "hw/pci/msi.h"
+
+#define TYPE_ARM_GICV2M "arm-gicv2m"
+#define ARM_GICV2M(obj) OBJECT_CHECK(ARMGICv2mState, (obj), TYPE_ARM_GICV2M)
+
+#define GICV2M_NUM_SPI_MAX 128
+
+#define V2M_MSI_TYPER           0x008
+#define V2M_MSI_SETSPI_NS       0x040
+#define V2M_MSI_IIDR            0xFCC
+#define V2M_IIDR0               0xFD0
+#define V2M_IIDR11              0xFFC
+
+#define PRODUCT_ID_QEMU         0x51 /* ASCII code Q */
+
+typedef struct ARMGICv2mState {
+    SysBusDevice parent_obj;
+
+    MemoryRegion iomem;
+    qemu_irq spi[GICV2M_NUM_SPI_MAX];
+
+    uint32_t base_spi;
+    uint32_t num_spi;
+} ARMGICv2mState;
+
+static void gicv2m_set_irq(void *opaque, int irq)
+{
+    ARMGICv2mState *s = (ARMGICv2mState *)opaque;
+
+    qemu_irq_pulse(s->spi[irq]);
+}
+
+static uint64_t gicv2m_read(void *opaque, hwaddr offset,
+                            unsigned size)
+{
+    ARMGICv2mState *s = (ARMGICv2mState *)opaque;
+    uint32_t val;
+
+    if (size != 4) {
+        qemu_log_mask(LOG_GUEST_ERROR, "gicv2m_read: bad size %u\n", size);
+        return 0;
+    }
+
+    switch (offset) {
+    case V2M_MSI_TYPER:
+        val = (s->base_spi + 32) << 16;
+        val |= s->num_spi;
+        return val;
+    case V2M_MSI_IIDR:
+        /* We don't have any valid implementor so we leave that field as zero
+         * and we return 0 in the arch revision as per the spec.
+         */
+        return (PRODUCT_ID_QEMU << 20);
+    case V2M_IIDR0 ... V2M_IIDR11:
+        /* We do not implement any optional identification registers and the
+         * mandatory MSI_PIDR2 register reads as 0x0, so we capture all
+         * implementation defined registers here.
+         */
+        return 0;
+    default:
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gicv2m_read: Bad offset %x\n", (int)offset);
+        return 0;
+    }
+}
+
+static void gicv2m_write(void *opaque, hwaddr offset,
+                        uint64_t value, unsigned size)
+{
+    ARMGICv2mState *s = (ARMGICv2mState *)opaque;
+
+    if (size != 2 && size != 4) {
+        qemu_log_mask(LOG_GUEST_ERROR, "gicv2m_write: bad size %u\n", size);
+        return;
+    }
+
+    switch (offset) {
+    case V2M_MSI_SETSPI_NS: {
+        int spi;
+
+        spi = (value & 0x3ff) - (s->base_spi + 32);
+        if (spi >= 0 && spi < s->num_spi) {
+            gicv2m_set_irq(s, spi);
+        }
+        return;
+    }
+    default:
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gicv2m_write: Bad offset %x\n", (int)offset);
+    }
+}
+
+static const MemoryRegionOps gicv2m_ops = {
+    .read = gicv2m_read,
+    .write = gicv2m_write,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
+static void gicv2m_realize(DeviceState *dev, Error **errp)
+{
+    ARMGICv2mState *s = ARM_GICV2M(dev);
+    int i;
+
+    if (s->num_spi > GICV2M_NUM_SPI_MAX) {
+        error_setg(errp,
+                   "requested %u SPIs exceeds GICv2m frame maximum %d",
+                   s->num_spi, GICV2M_NUM_SPI_MAX);
+        return;
+    }
+
+    if (s->base_spi + 32 > 1020 - s->num_spi) {
+        error_setg(errp,
+                   "requested base SPI %u+%u exceeds max. number 1020",
+                   s->base_spi + 32, s->num_spi);
+        return;
+    }
+
+    for (i = 0; i < s->num_spi; i++) {
+        sysbus_init_irq(SYS_BUS_DEVICE(dev), &s->spi[i]);
+    }
+
+    msi_supported = true;
+    kvm_gsi_direct_mapping = true;
+    kvm_msi_via_irqfd_allowed = kvm_irqfds_enabled();
+}
+
+static void gicv2m_init(Object *obj)
+{
+    SysBusDevice *sbd = SYS_BUS_DEVICE(obj);
+    ARMGICv2mState *s = ARM_GICV2M(obj);
+
+    memory_region_init_io(&s->iomem, OBJECT(s), &gicv2m_ops, s,
+                          "gicv2m", 0x1000);
+    sysbus_init_mmio(sbd, &s->iomem);
+}
+
+static Property gicv2m_properties[] = {
+    DEFINE_PROP_UINT32("base-spi", ARMGICv2mState, base_spi, 0),
+    DEFINE_PROP_UINT32("num-spi", ARMGICv2mState, num_spi, 64),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void gicv2m_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+
+    dc->props = gicv2m_properties;
+    dc->realize = gicv2m_realize;
+}
+
+static const TypeInfo gicv2m_info = {
+    .name          = TYPE_ARM_GICV2M,
+    .parent        = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(ARMGICv2mState),
+    .instance_init = gicv2m_init,
+    .class_init    = gicv2m_class_init,
+};
+
+static void gicv2m_register_types(void)
+{
+    type_register_static(&gicv2m_info);
+}
+
+type_init(gicv2m_register_types)
diff --git a/hw/intc/exynos4210_gic.c b/hw/intc/exynos4210_gic.c
index 0590d5dfb8..b2a4950bc3 100644
--- a/hw/intc/exynos4210_gic.c
+++ b/hw/intc/exynos4210_gic.c
@@ -213,9 +213,6 @@ void exynos4210_init_board_irqs(Exynos4210Irq *s)
     uint32_t grp, bit, irq_id, n;
 
     for (n = 0; n < EXYNOS4210_MAX_EXT_COMBINER_IN_IRQ; n++) {
-        s->board_irqs[n] = qemu_irq_split(s->int_combiner_irq[n],
-                s->ext_combiner_irq[n]);
-
         irq_id = 0;
         if (n == EXYNOS4210_COMBINER_GET_IRQ_NUM(1, 4) ||
                 n == EXYNOS4210_COMBINER_GET_IRQ_NUM(12, 4)) {
@@ -230,8 +227,10 @@ void exynos4210_init_board_irqs(Exynos4210Irq *s)
         if (irq_id) {
             s->board_irqs[n] = qemu_irq_split(s->int_combiner_irq[n],
                     s->ext_gic_irq[irq_id-32]);
+        } else {
+            s->board_irqs[n] = qemu_irq_split(s->int_combiner_irq[n],
+                    s->ext_combiner_irq[n]);
         }
-
     }
     for (; n < EXYNOS4210_MAX_INT_COMBINER_IN_IRQ; n++) {
         /* these IDs are passed to Internal Combiner and External GIC */
diff --git a/hw/isa/i82378.c b/hw/isa/i82378.c
index 9da9dfc4da..fcf97d86ac 100644
--- a/hw/isa/i82378.c
+++ b/hw/isa/i82378.c
@@ -65,7 +65,6 @@ static void i82378_realize(PCIDevice *pci, Error **errp)
     uint8_t *pci_conf;
     ISABus *isabus;
     ISADevice *isa;
-    qemu_irq *out0_irq;
 
     pci_conf = pci->config;
     pci_set_word(pci_conf + PCI_COMMAND,
@@ -88,11 +87,9 @@ static void i82378_realize(PCIDevice *pci, Error **errp)
        All devices accept byte access only, except timer
      */
 
-    /* Workaround the fact that i8259 is not qdev'ified... */
-    out0_irq = qemu_allocate_irqs(i82378_request_out0_irq, s, 1);
-
     /* 2 82C59 (irq) */
-    s->i8259 = i8259_init(isabus, *out0_irq);
+    s->i8259 = i8259_init(isabus,
+                          qemu_allocate_irq(i82378_request_out0_irq, s, 0));
     isa_bus_irqs(isabus, s->i8259);
 
     /* 1 82C54 (pit) */
diff --git a/hw/isa/isa-bus.c b/hw/isa/isa-bus.c
index 825aa627df..43e0cd8ddd 100644
--- a/hw/isa/isa-bus.c
+++ b/hw/isa/isa-bus.c
@@ -21,6 +21,7 @@
 #include "hw/sysbus.h"
 #include "sysemu/sysemu.h"
 #include "hw/isa/isa.h"
+#include "hw/i386/pc.h"
 
 static ISABus *isabus;
 
@@ -178,6 +179,9 @@ ISADevice *isa_vga_init(ISABus *bus)
     case VGA_VMWARE:
         fprintf(stderr, "%s: vmware_vga: no PCI bus\n", __func__);
         return NULL;
+    case VGA_VIRTIO:
+        fprintf(stderr, "%s: virtio-vga: no PCI bus\n", __func__);
+        return NULL;
     case VGA_NONE:
     default:
         return NULL;
@@ -267,3 +271,28 @@ MemoryRegion *isa_address_space_io(ISADevice *dev)
 }
 
 type_init(isabus_register_types)
+
+static void parallel_init(ISABus *bus, int index, CharDriverState *chr)
+{
+    DeviceState *dev;
+    ISADevice *isadev;
+
+    isadev = isa_create(bus, "isa-parallel");
+    dev = DEVICE(isadev);
+    qdev_prop_set_uint32(dev, "index", index);
+    qdev_prop_set_chr(dev, "chardev", chr);
+    qdev_init_nofail(dev);
+}
+
+void parallel_hds_isa_init(ISABus *bus, int n)
+{
+    int i;
+
+    assert(n <= MAX_PARALLEL_PORTS);
+
+    for (i = 0; i < n; i++) {
+        if (parallel_hds[i]) {
+            parallel_init(bus, i, parallel_hds[i]);
+        }
+    }
+}
diff --git a/hw/isa/lpc_ich9.c b/hw/isa/lpc_ich9.c
index bc9afc60ab..b3e0b1fd52 100644
--- a/hw/isa/lpc_ich9.c
+++ b/hw/isa/lpc_ich9.c
@@ -360,11 +360,8 @@ static void ich9_set_sci(void *opaque, int irq_num, int level)
 void ich9_lpc_pm_init(PCIDevice *lpc_pci)
 {
     ICH9LPCState *lpc = ICH9_LPC_DEVICE(lpc_pci);
-    qemu_irq *sci_irq;
-
-    sci_irq = qemu_allocate_irqs(ich9_set_sci, lpc, 1);
-    ich9_pm_init(lpc_pci, &lpc->pm, sci_irq[0]);
 
+    ich9_pm_init(lpc_pci, &lpc->pm, qemu_allocate_irq(ich9_set_sci, lpc, 0));
     ich9_lpc_reset(&lpc->d.qdev);
 }
 
@@ -410,12 +407,28 @@ static void ich9_lpc_rcba_update(ICH9LPCState *lpc, uint32_t rbca_old)
     }
 }
 
+/* config:GEN_PMCON* */
+static void
+ich9_lpc_pmcon_update(ICH9LPCState *lpc)
+{
+    uint16_t gen_pmcon_1 = pci_get_word(lpc->d.config + ICH9_LPC_GEN_PMCON_1);
+    uint16_t wmask;
+
+    if (gen_pmcon_1 & ICH9_LPC_GEN_PMCON_1_SMI_LOCK) {
+        wmask = pci_get_word(lpc->d.wmask + ICH9_LPC_GEN_PMCON_1);
+        wmask &= ~ICH9_LPC_GEN_PMCON_1_SMI_LOCK;
+        pci_set_word(lpc->d.wmask + ICH9_LPC_GEN_PMCON_1, wmask);
+        lpc->pm.smi_en_wmask &= ~1;
+    }
+}
+
 static int ich9_lpc_post_load(void *opaque, int version_id)
 {
     ICH9LPCState *lpc = opaque;
 
     ich9_lpc_pmbase_update(lpc);
     ich9_lpc_rcba_update(lpc, 0 /* disabled ICH9_LPC_RBCA_EN */);
+    ich9_lpc_pmcon_update(lpc);
     return 0;
 }
 
@@ -438,6 +451,9 @@ static void ich9_lpc_config_write(PCIDevice *d,
     if (ranges_overlap(addr, len, ICH9_LPC_PIRQE_ROUT, 4)) {
         pci_bus_fire_intx_routing_notifier(lpc->d.bus);
     }
+    if (ranges_overlap(addr, len, ICH9_LPC_GEN_PMCON_1, 8)) {
+        ich9_lpc_pmcon_update(lpc);
+    }
 }
 
 static void ich9_lpc_reset(DeviceState *qdev)
@@ -634,6 +650,7 @@ static const VMStateDescription vmstate_ich9_rst_cnt = {
     .name = "ICH9LPC/rst_cnt",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = ich9_rst_cnt_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(rst_cnt, ICH9LPCState),
         VMSTATE_END_OF_LIST()
@@ -653,12 +670,9 @@ static const VMStateDescription vmstate_ich9_lpc = {
         VMSTATE_UINT32(sci_level, ICH9LPCState),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_ich9_rst_cnt,
-            .needed = ich9_rst_cnt_needed
-        },
-        { 0 }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_ich9_rst_cnt,
+        NULL
     }
 };
 
diff --git a/hw/lm32/lm32_boards.c b/hw/lm32/lm32_boards.c
index 14d0efcdd9..70f48d3b1d 100644
--- a/hw/lm32/lm32_boards.c
+++ b/hw/lm32/lm32_boards.c
@@ -78,7 +78,7 @@ static void lm32_evr_init(MachineState *machine)
     DriveInfo *dinfo;
     MemoryRegion *address_space_mem =  get_system_memory();
     MemoryRegion *phys_ram = g_new(MemoryRegion, 1);
-    qemu_irq *cpu_irq, irq[32];
+    qemu_irq irq[32];
     ResetInfo *reset_info;
     int i;
 
@@ -123,8 +123,7 @@ static void lm32_evr_init(MachineState *machine)
                           1, 2, 0x01, 0x7e, 0x43, 0x00, 0x555, 0x2aa, 1);
 
     /* create irq lines */
-    cpu_irq = qemu_allocate_irqs(cpu_irq_handler, cpu, 1);
-    env->pic_state = lm32_pic_init(*cpu_irq);
+    env->pic_state = lm32_pic_init(qemu_allocate_irq(cpu_irq_handler, cpu, 0));
     for (i = 0; i < 32; i++) {
         irq[i] = qdev_get_gpio_in(env->pic_state, i);
     }
@@ -173,7 +172,7 @@ static void lm32_uclinux_init(MachineState *machine)
     DriveInfo *dinfo;
     MemoryRegion *address_space_mem =  get_system_memory();
     MemoryRegion *phys_ram = g_new(MemoryRegion, 1);
-    qemu_irq *cpu_irq, irq[32];
+    qemu_irq irq[32];
     HWSetup *hw;
     ResetInfo *reset_info;
     int i;
@@ -225,8 +224,7 @@ static void lm32_uclinux_init(MachineState *machine)
                           1, 2, 0x01, 0x7e, 0x43, 0x00, 0x555, 0x2aa, 1);
 
     /* create irq lines */
-    cpu_irq = qemu_allocate_irqs(cpu_irq_handler, env, 1);
-    env->pic_state = lm32_pic_init(*cpu_irq);
+    env->pic_state = lm32_pic_init(qemu_allocate_irq(cpu_irq_handler, env, 0));
     for (i = 0; i < 32; i++) {
         irq[i] = qdev_get_gpio_in(env->pic_state, i);
     }
diff --git a/hw/lm32/milkymist.c b/hw/lm32/milkymist.c
index e0cec7dc41..e755f5b24f 100644
--- a/hw/lm32/milkymist.c
+++ b/hw/lm32/milkymist.c
@@ -86,7 +86,7 @@ milkymist_init(MachineState *machine)
     DriveInfo *dinfo;
     MemoryRegion *address_space_mem = get_system_memory();
     MemoryRegion *phys_sdram = g_new(MemoryRegion, 1);
-    qemu_irq irq[32], *cpu_irq;
+    qemu_irq irq[32];
     int i;
     char *bios_filename;
     ResetInfo *reset_info;
@@ -130,8 +130,7 @@ milkymist_init(MachineState *machine)
                           2, 0x00, 0x89, 0x00, 0x1d, 1);
 
     /* create irq lines */
-    cpu_irq = qemu_allocate_irqs(cpu_irq_handler, cpu, 1);
-    env->pic_state = lm32_pic_init(*cpu_irq);
+    env->pic_state = lm32_pic_init(qemu_allocate_irq(cpu_irq_handler, cpu, 0));
     for (i = 0; i < 32; i++) {
         irq[i] = qdev_get_gpio_in(env->pic_state, i);
     }
diff --git a/hw/mips/Makefile.objs b/hw/mips/Makefile.objs
index 0a652f8521..9633f3a57d 100644
--- a/hw/mips/Makefile.objs
+++ b/hw/mips/Makefile.objs
@@ -1,4 +1,5 @@
-obj-y += mips_r4k.o mips_jazz.o mips_malta.o mips_mipssim.o
+obj-y += mips_r4k.o mips_malta.o mips_mipssim.o
 obj-y += addr.o cputimer.o mips_int.o
+obj-$(CONFIG_JAZZ) += mips_jazz.o
 obj-$(CONFIG_FULONG) += mips_fulong2e.o
 obj-y += gt64xxx_pci.o
diff --git a/hw/mips/mips_jazz.c b/hw/mips/mips_jazz.c
index 2c153e092f..9d60633efb 100644
--- a/hw/mips/mips_jazz.c
+++ b/hw/mips/mips_jazz.c
@@ -135,16 +135,16 @@ static void mips_jazz_init(MachineState *machine,
     MIPSCPU *cpu;
     CPUClass *cc;
     CPUMIPSState *env;
-    qemu_irq *rc4030, *i8259;
+    qemu_irq *i8259;
     rc4030_dma *dmas;
-    void* rc4030_opaque;
+    MemoryRegion *rc4030_dma_mr;
     MemoryRegion *isa_mem = g_new(MemoryRegion, 1);
     MemoryRegion *isa_io = g_new(MemoryRegion, 1);
     MemoryRegion *rtc = g_new(MemoryRegion, 1);
     MemoryRegion *i8042 = g_new(MemoryRegion, 1);
     MemoryRegion *dma_dummy = g_new(MemoryRegion, 1);
     NICInfo *nd;
-    DeviceState *dev;
+    DeviceState *dev, *rc4030;
     SysBusDevice *sysbus;
     ISABus *isa_bus;
     ISADevice *pit;
@@ -157,12 +157,7 @@ static void mips_jazz_init(MachineState *machine,
 
     /* init CPUs */
     if (cpu_model == NULL) {
-#ifdef TARGET_MIPS64
         cpu_model = "R4000";
-#else
-        /* FIXME: All wrong, this maybe should be R3000 for the older JAZZs. */
-        cpu_model = "24Kf";
-#endif
     }
     cpu = cpu_mips_init(cpu_model);
     if (cpu == NULL) {
@@ -218,8 +213,14 @@ static void mips_jazz_init(MachineState *machine,
     cpu_mips_clock_init(env);
 
     /* Chipset */
-    rc4030_opaque = rc4030_init(env->irq[6], env->irq[3], &rc4030, &dmas,
-                                address_space);
+    rc4030 = rc4030_init(&dmas, &rc4030_dma_mr);
+    sysbus = SYS_BUS_DEVICE(rc4030);
+    sysbus_connect_irq(sysbus, 0, env->irq[6]);
+    sysbus_connect_irq(sysbus, 1, env->irq[3]);
+    memory_region_add_subregion(address_space, 0x80000000,
+                                sysbus_mmio_get_region(sysbus, 0));
+    memory_region_add_subregion(address_space, 0xf0000000,
+                                sysbus_mmio_get_region(sysbus, 1));
     memory_region_init_io(dma_dummy, NULL, &dma_dummy_ops, NULL, "dummy_dma", 0x1000);
     memory_region_add_subregion(address_space, 0x8000d000, dma_dummy);
 
@@ -246,7 +247,7 @@ static void mips_jazz_init(MachineState *machine,
         sysbus = SYS_BUS_DEVICE(dev);
         sysbus_mmio_map(sysbus, 0, 0x60080000);
         sysbus_mmio_map(sysbus, 1, 0x40000000);
-        sysbus_connect_irq(sysbus, 0, rc4030[3]);
+        sysbus_connect_irq(sysbus, 0, qdev_get_gpio_in(rc4030, 3));
         {
             /* Simple ROM, so user doesn't have to provide one */
             MemoryRegion *rom_mr = g_new(MemoryRegion, 1);
@@ -272,8 +273,17 @@ static void mips_jazz_init(MachineState *machine,
         if (!nd->model)
             nd->model = g_strdup("dp83932");
         if (strcmp(nd->model, "dp83932") == 0) {
-            dp83932_init(nd, 0x80001000, 2, get_system_memory(), rc4030[4],
-                         rc4030_opaque, rc4030_dma_memory_rw);
+            qemu_check_nic_model(nd, "dp83932");
+
+            dev = qdev_create(NULL, "dp8393x");
+            qdev_set_nic_properties(dev, nd);
+            qdev_prop_set_uint8(dev, "it_shift", 2);
+            qdev_prop_set_ptr(dev, "dma_mr", rc4030_dma_mr);
+            qdev_init_nofail(dev);
+            sysbus = SYS_BUS_DEVICE(dev);
+            sysbus_mmio_map(sysbus, 0, 0x80001000);
+            sysbus_mmio_map(sysbus, 1, 0x8000b000);
+            sysbus_connect_irq(sysbus, 0, qdev_get_gpio_in(rc4030, 4));
             break;
         } else if (is_help_option(nd->model)) {
             fprintf(stderr, "qemu: Supported NICs: dp83932\n");
@@ -287,7 +297,7 @@ static void mips_jazz_init(MachineState *machine,
     /* SCSI adapter */
     esp_init(0x80002000, 0,
              rc4030_dma_read, rc4030_dma_write, dmas[0],
-             rc4030[5], &esp_reset, &dma_enable);
+             qdev_get_gpio_in(rc4030, 5), &esp_reset, &dma_enable);
 
     /* Floppy */
     if (drive_get_max_bus(IF_FLOPPY) >= MAX_FD) {
@@ -297,7 +307,7 @@ static void mips_jazz_init(MachineState *machine,
     for (n = 0; n < MAX_FD; n++) {
         fds[n] = drive_get(IF_FLOPPY, 0, n);
     }
-    fdctrl_init_sysbus(rc4030[1], 0, 0x80003000, fds);
+    fdctrl_init_sysbus(qdev_get_gpio_in(rc4030, 1), 0, 0x80003000, fds);
 
     /* Real time clock */
     rtc_init(isa_bus, 1980, NULL);
@@ -305,23 +315,26 @@ static void mips_jazz_init(MachineState *machine,
     memory_region_add_subregion(address_space, 0x80004000, rtc);
 
     /* Keyboard (i8042) */
-    i8042_mm_init(rc4030[6], rc4030[7], i8042, 0x1000, 0x1);
+    i8042_mm_init(qdev_get_gpio_in(rc4030, 6), qdev_get_gpio_in(rc4030, 7),
+                  i8042, 0x1000, 0x1);
     memory_region_add_subregion(address_space, 0x80005000, i8042);
 
     /* Serial ports */
     if (serial_hds[0]) {
-        serial_mm_init(address_space, 0x80006000, 0, rc4030[8], 8000000/16,
+        serial_mm_init(address_space, 0x80006000, 0,
+                       qdev_get_gpio_in(rc4030, 8), 8000000/16,
                        serial_hds[0], DEVICE_NATIVE_ENDIAN);
     }
     if (serial_hds[1]) {
-        serial_mm_init(address_space, 0x80007000, 0, rc4030[9], 8000000/16,
+        serial_mm_init(address_space, 0x80007000, 0,
+                       qdev_get_gpio_in(rc4030, 9), 8000000/16,
                        serial_hds[1], DEVICE_NATIVE_ENDIAN);
     }
 
     /* Parallel port */
     if (parallel_hds[0])
-        parallel_mm_init(address_space, 0x80008000, 0, rc4030[0],
-                         parallel_hds[0]);
+        parallel_mm_init(address_space, 0x80008000, 0,
+                         qdev_get_gpio_in(rc4030, 0), parallel_hds[0]);
 
     /* FIXME: missing Jazz sound at 0x8000c000, rc4030[2] */
 
diff --git a/hw/mips/mips_malta.c b/hw/mips/mips_malta.c
index 5140882c00..786a8f0638 100644
--- a/hw/mips/mips_malta.c
+++ b/hw/mips/mips_malta.c
@@ -97,7 +97,7 @@ typedef struct {
 static ISADevice *pit;
 
 static struct _loaderparams {
-    int ram_size;
+    int ram_size, ram_low_size;
     const char *kernel_filename;
     const char *kernel_cmdline;
     const char *initrd_filename;
@@ -641,8 +641,8 @@ static void write_bootloader (CPUMIPSState *env, uint8_t *base,
     stl_p(p++, 0x34a50000 | (ENVP_ADDR & 0xffff));               /* ori a1, a1, low(ENVP_ADDR) */
     stl_p(p++, 0x3c060000 | (((ENVP_ADDR + 8) >> 16) & 0xffff)); /* lui a2, high(ENVP_ADDR + 8) */
     stl_p(p++, 0x34c60000 | ((ENVP_ADDR + 8) & 0xffff));         /* ori a2, a2, low(ENVP_ADDR + 8) */
-    stl_p(p++, 0x3c070000 | (loaderparams.ram_size >> 16));     /* lui a3, high(ram_size) */
-    stl_p(p++, 0x34e70000 | (loaderparams.ram_size & 0xffff));  /* ori a3, a3, low(ram_size) */
+    stl_p(p++, 0x3c070000 | (loaderparams.ram_low_size >> 16));     /* lui a3, high(ram_low_size) */
+    stl_p(p++, 0x34e70000 | (loaderparams.ram_low_size & 0xffff));  /* ori a3, a3, low(ram_low_size) */
 
     /* Load BAR registers as done by YAMON */
     stl_p(p++, 0x3c09b400);                                      /* lui t1, 0xb400 */
@@ -851,8 +851,10 @@ static int64_t load_kernel (void)
     }
 
     prom_set(prom_buf, prom_index++, "memsize");
-    prom_set(prom_buf, prom_index++, "%i",
-             MIN(loaderparams.ram_size, 256 << 20));
+    prom_set(prom_buf, prom_index++, "%u", loaderparams.ram_low_size);
+
+    prom_set(prom_buf, prom_index++, "ememsize");
+    prom_set(prom_buf, prom_index++, "%u", loaderparams.ram_size);
 
     prom_set(prom_buf, prom_index++, "modetty0");
     prom_set(prom_buf, prom_index++, "38400n8r");
@@ -1054,7 +1056,8 @@ void mips_malta_init(MachineState *machine)
         }
 
         /* Write a small bootloader to the flash location. */
-        loaderparams.ram_size = ram_low_size;
+        loaderparams.ram_size = ram_size;
+        loaderparams.ram_low_size = ram_low_size;
         loaderparams.kernel_filename = kernel_filename;
         loaderparams.kernel_cmdline = kernel_cmdline;
         loaderparams.initrd_filename = initrd_filename;
diff --git a/hw/misc/macio/macio.c b/hw/misc/macio/macio.c
index 063ad80412..e9037b0c39 100644
--- a/hw/misc/macio/macio.c
+++ b/hw/misc/macio/macio.c
@@ -126,17 +126,18 @@ static void macio_bar_setup(MacIOState *macio_state)
     }
 }
 
-static int macio_common_initfn(PCIDevice *d)
+static void macio_common_realize(PCIDevice *d, Error **errp)
 {
     MacIOState *s = MACIO(d);
     SysBusDevice *sysbus_dev;
-    int ret;
+    Error *err = NULL;
 
     d->config[0x3d] = 0x01; // interrupt on pin 1
 
-    ret = qdev_init(DEVICE(&s->cuda));
-    if (ret < 0) {
-        return ret;
+    object_property_set_bool(OBJECT(&s->cuda), true, "realized", &err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
     }
     sysbus_dev = SYS_BUS_DEVICE(&s->cuda);
     memory_region_add_subregion(&s->bar, 0x16000,
@@ -144,12 +145,11 @@ static int macio_common_initfn(PCIDevice *d)
 
     macio_bar_setup(s);
     pci_register_bar(d, 0, PCI_BASE_ADDRESS_SPACE_MEMORY, &s->bar);
-
-    return 0;
 }
 
-static int macio_initfn_ide(MacIOState *s, MACIOIDEState *ide, qemu_irq irq0,
-                            qemu_irq irq1, int dmaid)
+static void macio_realize_ide(MacIOState *s, MACIOIDEState *ide,
+                              qemu_irq irq0, qemu_irq irq1, int dmaid,
+                              Error **errp)
 {
     SysBusDevice *sysbus_dev;
 
@@ -157,27 +157,31 @@ static int macio_initfn_ide(MacIOState *s, MACIOIDEState *ide, qemu_irq irq0,
     sysbus_connect_irq(sysbus_dev, 0, irq0);
     sysbus_connect_irq(sysbus_dev, 1, irq1);
     macio_ide_register_dma(ide, s->dbdma, dmaid);
-    return qdev_init(DEVICE(ide));
+    object_property_set_bool(OBJECT(ide), true, "realized", errp);
 }
 
-static int macio_oldworld_initfn(PCIDevice *d)
+static void macio_oldworld_realize(PCIDevice *d, Error **errp)
 {
     MacIOState *s = MACIO(d);
     OldWorldMacIOState *os = OLDWORLD_MACIO(d);
+    Error *err = NULL;
     SysBusDevice *sysbus_dev;
     int i;
     int cur_irq = 0;
-    int ret = macio_common_initfn(d);
-    if (ret < 0) {
-        return ret;
+
+    macio_common_realize(d, &err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
     }
 
     sysbus_dev = SYS_BUS_DEVICE(&s->cuda);
     sysbus_connect_irq(sysbus_dev, 0, os->irqs[cur_irq++]);
 
-    ret = qdev_init(DEVICE(&os->nvram));
-    if (ret < 0) {
-        return ret;
+    object_property_set_bool(OBJECT(&os->nvram), true, "realized", &err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
     }
     sysbus_dev = SYS_BUS_DEVICE(&os->nvram);
     memory_region_add_subregion(&s->bar, 0x60000,
@@ -194,13 +198,12 @@ static int macio_oldworld_initfn(PCIDevice *d)
         qemu_irq irq0 = os->irqs[cur_irq++];
         qemu_irq irq1 = os->irqs[cur_irq++];
 
-        ret = macio_initfn_ide(s, &os->ide[i], irq0, irq1, 0x16 + (i * 4));
-        if (ret < 0) {
-            return ret;
+        macio_realize_ide(s, &os->ide[i], irq0, irq1, 0x16 + (i * 4), &err);
+        if (err) {
+            error_propagate(errp, err);
+            return;
         }
     }
-
-    return 0;
 }
 
 static void macio_init_ide(MacIOState *s, MACIOIDEState *ide, size_t ide_size,
@@ -268,17 +271,20 @@ static const MemoryRegionOps timer_ops = {
     .endianness = DEVICE_LITTLE_ENDIAN,
 };
 
-static int macio_newworld_initfn(PCIDevice *d)
+static void macio_newworld_realize(PCIDevice *d, Error **errp)
 {
     MacIOState *s = MACIO(d);
     NewWorldMacIOState *ns = NEWWORLD_MACIO(d);
+    Error *err = NULL;
     SysBusDevice *sysbus_dev;
     MemoryRegion *timer_memory = NULL;
     int i;
     int cur_irq = 0;
-    int ret = macio_common_initfn(d);
-    if (ret < 0) {
-        return ret;
+
+    macio_common_realize(d, &err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
     }
 
     sysbus_dev = SYS_BUS_DEVICE(&s->cuda);
@@ -294,9 +300,10 @@ static int macio_newworld_initfn(PCIDevice *d)
         qemu_irq irq0 = ns->irqs[cur_irq++];
         qemu_irq irq1 = ns->irqs[cur_irq++];
 
-        ret = macio_initfn_ide(s, &ns->ide[i], irq0, irq1, 0x16 + (i * 4));
-        if (ret < 0) {
-            return ret;
+        macio_realize_ide(s, &ns->ide[i], irq0, irq1, 0x16 + (i * 4), &err);
+        if (err) {
+            error_propagate(errp, err);
+            return;
         }
     }
 
@@ -305,8 +312,6 @@ static int macio_newworld_initfn(PCIDevice *d)
     memory_region_init_io(timer_memory, OBJECT(s), &timer_ops, NULL, "timer",
                           0x1000);
     memory_region_add_subregion(&s->bar, 0x15000, timer_memory);
-
-    return 0;
 }
 
 static void macio_newworld_init(Object *obj)
@@ -352,7 +357,7 @@ static void macio_oldworld_class_init(ObjectClass *oc, void *data)
     PCIDeviceClass *pdc = PCI_DEVICE_CLASS(oc);
     DeviceClass *dc = DEVICE_CLASS(oc);
 
-    pdc->init = macio_oldworld_initfn;
+    pdc->realize = macio_oldworld_realize;
     pdc->device_id = PCI_DEVICE_ID_APPLE_343S1201;
     dc->vmsd = &vmstate_macio_oldworld;
 }
@@ -372,7 +377,7 @@ static void macio_newworld_class_init(ObjectClass *oc, void *data)
     PCIDeviceClass *pdc = PCI_DEVICE_CLASS(oc);
     DeviceClass *dc = DEVICE_CLASS(oc);
 
-    pdc->init = macio_newworld_initfn;
+    pdc->realize = macio_newworld_realize;
     pdc->device_id = PCI_DEVICE_ID_APPLE_UNI_N_KEYL;
     dc->vmsd = &vmstate_macio_newworld;
 }
diff --git a/hw/net/Makefile.objs b/hw/net/Makefile.objs
index 7b91c4e51d..98801739ef 100644
--- a/hw/net/Makefile.objs
+++ b/hw/net/Makefile.objs
@@ -39,3 +39,4 @@ obj-$(CONFIG_ETSEC) += fsl_etsec/etsec.o fsl_etsec/registers.o \
 common-obj-$(CONFIG_ROCKER) += rocker/rocker.o rocker/rocker_fp.o \
                                rocker/rocker_desc.o rocker/rocker_world.o \
                                rocker/rocker_of_dpa.o
+obj-$(call lnot,$(CONFIG_ROCKER)) += rocker/qmp-norocker.o
diff --git a/hw/net/cadence_gem.c b/hw/net/cadence_gem.c
index dafe91421b..494a346cf6 100644
--- a/hw/net/cadence_gem.c
+++ b/hw/net/cadence_gem.c
@@ -155,7 +155,7 @@
 #define GEM_NWCFG_BCAST_REJ    0x00000020 /* Reject broadcast packets */
 #define GEM_NWCFG_PROMISC      0x00000010 /* Accept all packets */
 
-#define GEM_DMACFG_RBUFSZ_M    0x007F0000 /* DMA RX Buffer Size mask */
+#define GEM_DMACFG_RBUFSZ_M    0x00FF0000 /* DMA RX Buffer Size mask */
 #define GEM_DMACFG_RBUFSZ_S    16         /* DMA RX Buffer Size shift */
 #define GEM_DMACFG_RBUFSZ_MUL  64         /* DMA RX Buffer Size multiplier */
 #define GEM_DMACFG_TXCSUM_OFFL 0x00000800 /* Transmit checksum offload */
diff --git a/hw/net/dp8393x.c b/hw/net/dp8393x.c
index 7ce13d2b46..cd889bce86 100644
--- a/hw/net/dp8393x.c
+++ b/hw/net/dp8393x.c
@@ -17,20 +17,15 @@
  * with this program; if not, see <http://www.gnu.org/licenses/>.
  */
 
-#include "hw/hw.h"
-#include "qemu/timer.h"
+#include "hw/sysbus.h"
+#include "hw/devices.h"
 #include "net/net.h"
-#include "hw/mips/mips.h"
+#include "qemu/timer.h"
+#include <zlib.h>
 
 //#define DEBUG_SONIC
 
-/* Calculate CRCs properly on Rx packets */
-#define SONIC_CALCULATE_RXCRC
-
-#if defined(SONIC_CALCULATE_RXCRC)
-/* For crc32 */
-#include <zlib.h>
-#endif
+#define SONIC_PROM_SIZE 0x1000
 
 #ifdef DEBUG_SONIC
 #define DPRINTF(fmt, ...) \
@@ -145,9 +140,14 @@ do { printf("sonic ERROR: %s: " fmt, __func__ , ## __VA_ARGS__); } while (0)
 #define SONIC_ISR_PINT   0x0800
 #define SONIC_ISR_LCD    0x1000
 
+#define TYPE_DP8393X "dp8393x"
+#define DP8393X(obj) OBJECT_CHECK(dp8393xState, (obj), TYPE_DP8393X)
+
 typedef struct dp8393xState {
+    SysBusDevice parent_obj;
+
     /* Hardware */
-    int it_shift;
+    uint8_t it_shift;
     qemu_irq irq;
 #ifdef DEBUG_SONIC
     int irq_level;
@@ -156,8 +156,8 @@ typedef struct dp8393xState {
     int64_t wt_last_update;
     NICConf conf;
     NICState *nic;
-    MemoryRegion *address_space;
     MemoryRegion mmio;
+    MemoryRegion prom;
 
     /* Registers */
     uint8_t cam[16][6];
@@ -168,8 +168,8 @@ typedef struct dp8393xState {
     int loopback_packet;
 
     /* Memory access */
-    void (*memory_rw)(void *opaque, hwaddr addr, uint8_t *buf, int len, int is_write);
-    void* mem_opaque;
+    void *dma_mr;
+    AddressSpace as;
 } dp8393xState;
 
 static void dp8393x_update_irq(dp8393xState *s)
@@ -190,7 +190,7 @@ static void dp8393x_update_irq(dp8393xState *s)
     qemu_set_irq(s->irq, level);
 }
 
-static void do_load_cam(dp8393xState *s)
+static void dp8393x_do_load_cam(dp8393xState *s)
 {
     uint16_t data[8];
     int width, size;
@@ -201,9 +201,9 @@ static void do_load_cam(dp8393xState *s)
 
     while (s->regs[SONIC_CDC] & 0x1f) {
         /* Fill current entry */
-        s->memory_rw(s->mem_opaque,
+        address_space_rw(&s->as,
             (s->regs[SONIC_URRA] << 16) | s->regs[SONIC_CDP],
-            (uint8_t *)data, size, 0);
+            MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
         s->cam[index][0] = data[1 * width] & 0xff;
         s->cam[index][1] = data[1 * width] >> 8;
         s->cam[index][2] = data[2 * width] & 0xff;
@@ -220,9 +220,9 @@ static void do_load_cam(dp8393xState *s)
     }
 
     /* Read CAM enable */
-    s->memory_rw(s->mem_opaque,
+    address_space_rw(&s->as,
         (s->regs[SONIC_URRA] << 16) | s->regs[SONIC_CDP],
-        (uint8_t *)data, size, 0);
+        MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
     s->regs[SONIC_CE] = data[0 * width];
     DPRINTF("load cam done. cam enable mask 0x%04x\n", s->regs[SONIC_CE]);
 
@@ -232,7 +232,7 @@ static void do_load_cam(dp8393xState *s)
     dp8393x_update_irq(s);
 }
 
-static void do_read_rra(dp8393xState *s)
+static void dp8393x_do_read_rra(dp8393xState *s)
 {
     uint16_t data[8];
     int width, size;
@@ -240,9 +240,9 @@ static void do_read_rra(dp8393xState *s)
     /* Read memory */
     width = (s->regs[SONIC_DCR] & SONIC_DCR_DW) ? 2 : 1;
     size = sizeof(uint16_t) * 4 * width;
-    s->memory_rw(s->mem_opaque,
+    address_space_rw(&s->as,
         (s->regs[SONIC_URRA] << 16) | s->regs[SONIC_RRP],
-        (uint8_t *)data, size, 0);
+        MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
 
     /* Update SONIC registers */
     s->regs[SONIC_CRBA0] = data[0 * width];
@@ -272,7 +272,7 @@ static void do_read_rra(dp8393xState *s)
     s->regs[SONIC_CR] &= ~SONIC_CR_RRRA;
 }
 
-static void do_software_reset(dp8393xState *s)
+static void dp8393x_do_software_reset(dp8393xState *s)
 {
     timer_del(s->watchdog);
 
@@ -280,7 +280,7 @@ static void do_software_reset(dp8393xState *s)
     s->regs[SONIC_CR] |= SONIC_CR_RST | SONIC_CR_RXDIS;
 }
 
-static void set_next_tick(dp8393xState *s)
+static void dp8393x_set_next_tick(dp8393xState *s)
 {
     uint32_t ticks;
     int64_t delay;
@@ -296,7 +296,7 @@ static void set_next_tick(dp8393xState *s)
     timer_mod(s->watchdog, s->wt_last_update + delay);
 }
 
-static void update_wt_regs(dp8393xState *s)
+static void dp8393x_update_wt_regs(dp8393xState *s)
 {
     int64_t elapsed;
     uint32_t val;
@@ -311,33 +311,33 @@ static void update_wt_regs(dp8393xState *s)
     val -= elapsed / 5000000;
     s->regs[SONIC_WT1] = (val >> 16) & 0xffff;
     s->regs[SONIC_WT0] = (val >> 0)  & 0xffff;
-    set_next_tick(s);
+    dp8393x_set_next_tick(s);
 
 }
 
-static void do_start_timer(dp8393xState *s)
+static void dp8393x_do_start_timer(dp8393xState *s)
 {
     s->regs[SONIC_CR] &= ~SONIC_CR_STP;
-    set_next_tick(s);
+    dp8393x_set_next_tick(s);
 }
 
-static void do_stop_timer(dp8393xState *s)
+static void dp8393x_do_stop_timer(dp8393xState *s)
 {
     s->regs[SONIC_CR] &= ~SONIC_CR_ST;
-    update_wt_regs(s);
+    dp8393x_update_wt_regs(s);
 }
 
-static void do_receiver_enable(dp8393xState *s)
+static void dp8393x_do_receiver_enable(dp8393xState *s)
 {
     s->regs[SONIC_CR] &= ~SONIC_CR_RXDIS;
 }
 
-static void do_receiver_disable(dp8393xState *s)
+static void dp8393x_do_receiver_disable(dp8393xState *s)
 {
     s->regs[SONIC_CR] &= ~SONIC_CR_RXEN;
 }
 
-static void do_transmit_packets(dp8393xState *s)
+static void dp8393x_do_transmit_packets(dp8393xState *s)
 {
     NetClientState *nc = qemu_get_queue(s->nic);
     uint16_t data[12];
@@ -353,9 +353,9 @@ static void do_transmit_packets(dp8393xState *s)
                 (s->regs[SONIC_UTDA] << 16) | s->regs[SONIC_CTDA]);
         size = sizeof(uint16_t) * 6 * width;
         s->regs[SONIC_TTDA] = s->regs[SONIC_CTDA];
-        s->memory_rw(s->mem_opaque,
+        address_space_rw(&s->as,
             ((s->regs[SONIC_UTDA] << 16) | s->regs[SONIC_TTDA]) + sizeof(uint16_t) * width,
-            (uint8_t *)data, size, 0);
+            MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
         tx_len = 0;
 
         /* Update registers */
@@ -379,18 +379,18 @@ static void do_transmit_packets(dp8393xState *s)
             if (tx_len + len > sizeof(s->tx_buffer)) {
                 len = sizeof(s->tx_buffer) - tx_len;
             }
-            s->memory_rw(s->mem_opaque,
+            address_space_rw(&s->as,
                 (s->regs[SONIC_TSA1] << 16) | s->regs[SONIC_TSA0],
-                &s->tx_buffer[tx_len], len, 0);
+                MEMTXATTRS_UNSPECIFIED, &s->tx_buffer[tx_len], len, 0);
             tx_len += len;
 
             i++;
             if (i != s->regs[SONIC_TFC]) {
                 /* Read next fragment details */
                 size = sizeof(uint16_t) * 3 * width;
-                s->memory_rw(s->mem_opaque,
+                address_space_rw(&s->as,
                     ((s->regs[SONIC_UTDA] << 16) | s->regs[SONIC_TTDA]) + sizeof(uint16_t) * (4 + 3 * i) * width,
-                    (uint8_t *)data, size, 0);
+                    MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
                 s->regs[SONIC_TSA0] = data[0 * width];
                 s->regs[SONIC_TSA1] = data[1 * width];
                 s->regs[SONIC_TFS] = data[2 * width];
@@ -422,16 +422,16 @@ static void do_transmit_packets(dp8393xState *s)
         /* Write status */
         data[0 * width] = s->regs[SONIC_TCR] & 0x0fff; /* status */
         size = sizeof(uint16_t) * width;
-        s->memory_rw(s->mem_opaque,
+        address_space_rw(&s->as,
             (s->regs[SONIC_UTDA] << 16) | s->regs[SONIC_TTDA],
-            (uint8_t *)data, size, 1);
+            MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 1);
 
         if (!(s->regs[SONIC_CR] & SONIC_CR_HTX)) {
             /* Read footer of packet */
             size = sizeof(uint16_t) * width;
-            s->memory_rw(s->mem_opaque,
+            address_space_rw(&s->as,
                 ((s->regs[SONIC_UTDA] << 16) | s->regs[SONIC_TTDA]) + sizeof(uint16_t) * (4 + 3 * s->regs[SONIC_TFC]) * width,
-                (uint8_t *)data, size, 0);
+                MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
             s->regs[SONIC_CTDA] = data[0 * width] & ~0x1;
             if (data[0 * width] & 0x1) {
                 /* EOL detected */
@@ -446,12 +446,12 @@ static void do_transmit_packets(dp8393xState *s)
     dp8393x_update_irq(s);
 }
 
-static void do_halt_transmission(dp8393xState *s)
+static void dp8393x_do_halt_transmission(dp8393xState *s)
 {
     /* Nothing to do */
 }
 
-static void do_command(dp8393xState *s, uint16_t command)
+static void dp8393x_do_command(dp8393xState *s, uint16_t command)
 {
     if ((s->regs[SONIC_CR] & SONIC_CR_RST) && !(command & SONIC_CR_RST)) {
         s->regs[SONIC_CR] &= ~SONIC_CR_RST;
@@ -461,34 +461,36 @@ static void do_command(dp8393xState *s, uint16_t command)
     s->regs[SONIC_CR] |= (command & SONIC_CR_MASK);
 
     if (command & SONIC_CR_HTX)
-        do_halt_transmission(s);
+        dp8393x_do_halt_transmission(s);
     if (command & SONIC_CR_TXP)
-        do_transmit_packets(s);
+        dp8393x_do_transmit_packets(s);
     if (command & SONIC_CR_RXDIS)
-        do_receiver_disable(s);
+        dp8393x_do_receiver_disable(s);
     if (command & SONIC_CR_RXEN)
-        do_receiver_enable(s);
+        dp8393x_do_receiver_enable(s);
     if (command & SONIC_CR_STP)
-        do_stop_timer(s);
+        dp8393x_do_stop_timer(s);
     if (command & SONIC_CR_ST)
-        do_start_timer(s);
+        dp8393x_do_start_timer(s);
     if (command & SONIC_CR_RST)
-        do_software_reset(s);
+        dp8393x_do_software_reset(s);
     if (command & SONIC_CR_RRRA)
-        do_read_rra(s);
+        dp8393x_do_read_rra(s);
     if (command & SONIC_CR_LCAM)
-        do_load_cam(s);
+        dp8393x_do_load_cam(s);
 }
 
-static uint16_t read_register(dp8393xState *s, int reg)
+static uint64_t dp8393x_read(void *opaque, hwaddr addr, unsigned int size)
 {
+    dp8393xState *s = opaque;
+    int reg = addr >> s->it_shift;
     uint16_t val = 0;
 
     switch (reg) {
         /* Update data before reading it */
         case SONIC_WT0:
         case SONIC_WT1:
-            update_wt_regs(s);
+            dp8393x_update_wt_regs(s);
             val = s->regs[reg];
             break;
         /* Accept read to some registers only when in reset mode */
@@ -510,14 +512,18 @@ static uint16_t read_register(dp8393xState *s, int reg)
     return val;
 }
 
-static void write_register(dp8393xState *s, int reg, uint16_t val)
+static void dp8393x_write(void *opaque, hwaddr addr, uint64_t data,
+                          unsigned int size)
 {
-    DPRINTF("write 0x%04x to reg %s\n", val, reg_names[reg]);
+    dp8393xState *s = opaque;
+    int reg = addr >> s->it_shift;
+
+    DPRINTF("write 0x%04x to reg %s\n", (uint16_t)data, reg_names[reg]);
 
     switch (reg) {
         /* Command register */
         case SONIC_CR:
-            do_command(s, val);
+            dp8393x_do_command(s, data);
             break;
         /* Prevent write to read-only registers */
         case SONIC_CAP2:
@@ -530,37 +536,37 @@ static void write_register(dp8393xState *s, int reg, uint16_t val)
         /* Accept write to some registers only when in reset mode */
         case SONIC_DCR:
             if (s->regs[SONIC_CR] & SONIC_CR_RST) {
-                s->regs[reg] = val & 0xbfff;
+                s->regs[reg] = data & 0xbfff;
             } else {
                 DPRINTF("writing to DCR invalid\n");
             }
             break;
         case SONIC_DCR2:
             if (s->regs[SONIC_CR] & SONIC_CR_RST) {
-                s->regs[reg] = val & 0xf017;
+                s->regs[reg] = data & 0xf017;
             } else {
                 DPRINTF("writing to DCR2 invalid\n");
             }
             break;
         /* 12 lower bytes are Read Only */
         case SONIC_TCR:
-            s->regs[reg] = val & 0xf000;
+            s->regs[reg] = data & 0xf000;
             break;
         /* 9 lower bytes are Read Only */
         case SONIC_RCR:
-            s->regs[reg] = val & 0xffe0;
+            s->regs[reg] = data & 0xffe0;
             break;
         /* Ignore most significant bit */
         case SONIC_IMR:
-            s->regs[reg] = val & 0x7fff;
+            s->regs[reg] = data & 0x7fff;
             dp8393x_update_irq(s);
             break;
         /* Clear bits by writing 1 to them */
         case SONIC_ISR:
-            val &= s->regs[reg];
-            s->regs[reg] &= ~val;
-            if (val & SONIC_ISR_RBE) {
-                do_read_rra(s);
+            data &= s->regs[reg];
+            s->regs[reg] &= ~data;
+            if (data & SONIC_ISR_RBE) {
+                dp8393x_do_read_rra(s);
             }
             dp8393x_update_irq(s);
             break;
@@ -569,24 +575,32 @@ static void write_register(dp8393xState *s, int reg, uint16_t val)
         case SONIC_REA:
         case SONIC_RRP:
         case SONIC_RWP:
-            s->regs[reg] = val & 0xfffe;
+            s->regs[reg] = data & 0xfffe;
             break;
         /* Invert written value for some registers */
         case SONIC_CRCT:
         case SONIC_FAET:
         case SONIC_MPT:
-            s->regs[reg] = val ^ 0xffff;
+            s->regs[reg] = data ^ 0xffff;
             break;
         /* All other registers have no special contrainst */
         default:
-            s->regs[reg] = val;
+            s->regs[reg] = data;
     }
 
     if (reg == SONIC_WT0 || reg == SONIC_WT1) {
-        set_next_tick(s);
+        dp8393x_set_next_tick(s);
     }
 }
 
+static const MemoryRegionOps dp8393x_ops = {
+    .read = dp8393x_read,
+    .write = dp8393x_write,
+    .impl.min_access_size = 2,
+    .impl.max_access_size = 2,
+    .endianness = DEVICE_NATIVE_ENDIAN,
+};
+
 static void dp8393x_watchdog(void *opaque)
 {
     dp8393xState *s = opaque;
@@ -597,84 +611,14 @@ static void dp8393x_watchdog(void *opaque)
 
     s->regs[SONIC_WT1] = 0xffff;
     s->regs[SONIC_WT0] = 0xffff;
-    set_next_tick(s);
+    dp8393x_set_next_tick(s);
 
     /* Signal underflow */
     s->regs[SONIC_ISR] |= SONIC_ISR_TC;
     dp8393x_update_irq(s);
 }
 
-static uint32_t dp8393x_readw(void *opaque, hwaddr addr)
-{
-    dp8393xState *s = opaque;
-    int reg;
-
-    if ((addr & ((1 << s->it_shift) - 1)) != 0) {
-        return 0;
-    }
-
-    reg = addr >> s->it_shift;
-    return read_register(s, reg);
-}
-
-static uint32_t dp8393x_readb(void *opaque, hwaddr addr)
-{
-    uint16_t v = dp8393x_readw(opaque, addr & ~0x1);
-    return (v >> (8 * (addr & 0x1))) & 0xff;
-}
-
-static uint32_t dp8393x_readl(void *opaque, hwaddr addr)
-{
-    uint32_t v;
-    v = dp8393x_readw(opaque, addr);
-    v |= dp8393x_readw(opaque, addr + 2) << 16;
-    return v;
-}
-
-static void dp8393x_writew(void *opaque, hwaddr addr, uint32_t val)
-{
-    dp8393xState *s = opaque;
-    int reg;
-
-    if ((addr & ((1 << s->it_shift) - 1)) != 0) {
-        return;
-    }
-
-    reg = addr >> s->it_shift;
-
-    write_register(s, reg, (uint16_t)val);
-}
-
-static void dp8393x_writeb(void *opaque, hwaddr addr, uint32_t val)
-{
-    uint16_t old_val = dp8393x_readw(opaque, addr & ~0x1);
-
-    switch (addr & 3) {
-    case 0:
-        val = val | (old_val & 0xff00);
-        break;
-    case 1:
-        val = (val << 8) | (old_val & 0x00ff);
-        break;
-    }
-    dp8393x_writew(opaque, addr & ~0x1, val);
-}
-
-static void dp8393x_writel(void *opaque, hwaddr addr, uint32_t val)
-{
-    dp8393x_writew(opaque, addr, val & 0xffff);
-    dp8393x_writew(opaque, addr + 2, (val >> 16) & 0xffff);
-}
-
-static const MemoryRegionOps dp8393x_ops = {
-    .old_mmio = {
-        .read = { dp8393x_readb, dp8393x_readw, dp8393x_readl, },
-        .write = { dp8393x_writeb, dp8393x_writew, dp8393x_writel, },
-    },
-    .endianness = DEVICE_NATIVE_ENDIAN,
-};
-
-static int nic_can_receive(NetClientState *nc)
+static int dp8393x_can_receive(NetClientState *nc)
 {
     dp8393xState *s = qemu_get_nic_opaque(nc);
 
@@ -685,7 +629,8 @@ static int nic_can_receive(NetClientState *nc)
     return 1;
 }
 
-static int receive_filter(dp8393xState *s, const uint8_t * buf, int size)
+static int dp8393x_receive_filter(dp8393xState *s, const uint8_t * buf,
+                                  int size)
 {
     static const uint8_t bcast[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff};
     int i;
@@ -723,7 +668,8 @@ static int receive_filter(dp8393xState *s, const uint8_t * buf, int size)
     return -1;
 }
 
-static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
+static ssize_t dp8393x_receive(NetClientState *nc, const uint8_t * buf,
+                               size_t size)
 {
     dp8393xState *s = qemu_get_nic_opaque(nc);
     uint16_t data[10];
@@ -737,7 +683,7 @@ static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
     s->regs[SONIC_RCR] &= ~(SONIC_RCR_PRX | SONIC_RCR_LBK | SONIC_RCR_FAER |
         SONIC_RCR_CRCR | SONIC_RCR_LPKT | SONIC_RCR_BC | SONIC_RCR_MC);
 
-    packet_type = receive_filter(s, buf, size);
+    packet_type = dp8393x_receive_filter(s, buf, size);
     if (packet_type < 0) {
         DPRINTF("packet not for netcard\n");
         return -1;
@@ -750,7 +696,8 @@ static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
         /* Are we still in resource exhaustion? */
         size = sizeof(uint16_t) * 1 * width;
         address = ((s->regs[SONIC_URDA] << 16) | s->regs[SONIC_CRDA]) + sizeof(uint16_t) * 5 * width;
-        s->memory_rw(s->mem_opaque, address, (uint8_t*)data, size, 0);
+        address_space_rw(&s->as, address, MEMTXATTRS_UNSPECIFIED,
+                         (uint8_t *)data, size, 0);
         if (data[0 * width] & 0x1) {
             /* Still EOL ; stop reception */
             return -1;
@@ -764,18 +711,16 @@ static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
     s->regs[SONIC_TRBA0] = s->regs[SONIC_CRBA0];
 
     /* Calculate the ethernet checksum */
-#ifdef SONIC_CALCULATE_RXCRC
     checksum = cpu_to_le32(crc32(0, buf, rx_len));
-#else
-    checksum = 0;
-#endif
 
     /* Put packet into RBA */
     DPRINTF("Receive packet at %08x\n", (s->regs[SONIC_CRBA1] << 16) | s->regs[SONIC_CRBA0]);
     address = (s->regs[SONIC_CRBA1] << 16) | s->regs[SONIC_CRBA0];
-    s->memory_rw(s->mem_opaque, address, (uint8_t*)buf, rx_len, 1);
+    address_space_rw(&s->as, address,
+        MEMTXATTRS_UNSPECIFIED, (uint8_t *)buf, rx_len, 1);
     address += rx_len;
-    s->memory_rw(s->mem_opaque, address, (uint8_t*)&checksum, 4, 1);
+    address_space_rw(&s->as, address,
+        MEMTXATTRS_UNSPECIFIED, (uint8_t *)&checksum, 4, 1);
     rx_len += 4;
     s->regs[SONIC_CRBA1] = address >> 16;
     s->regs[SONIC_CRBA0] = address & 0xffff;
@@ -803,29 +748,30 @@ static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
     data[3 * width] = s->regs[SONIC_TRBA1]; /* pkt_ptr1 */
     data[4 * width] = s->regs[SONIC_RSC]; /* seq_no */
     size = sizeof(uint16_t) * 5 * width;
-    s->memory_rw(s->mem_opaque, (s->regs[SONIC_URDA] << 16) | s->regs[SONIC_CRDA], (uint8_t *)data, size, 1);
+    address_space_rw(&s->as, (s->regs[SONIC_URDA] << 16) | s->regs[SONIC_CRDA],
+        MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 1);
 
     /* Move to next descriptor */
     size = sizeof(uint16_t) * width;
-    s->memory_rw(s->mem_opaque,
+    address_space_rw(&s->as,
         ((s->regs[SONIC_URDA] << 16) | s->regs[SONIC_CRDA]) + sizeof(uint16_t) * 5 * width,
-        (uint8_t *)data, size, 0);
+        MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, size, 0);
     s->regs[SONIC_LLFA] = data[0 * width];
     if (s->regs[SONIC_LLFA] & 0x1) {
         /* EOL detected */
         s->regs[SONIC_ISR] |= SONIC_ISR_RDE;
     } else {
         data[0 * width] = 0; /* in_use */
-        s->memory_rw(s->mem_opaque,
+        address_space_rw(&s->as,
             ((s->regs[SONIC_URDA] << 16) | s->regs[SONIC_CRDA]) + sizeof(uint16_t) * 6 * width,
-            (uint8_t *)data, size, 1);
+            MEMTXATTRS_UNSPECIFIED, (uint8_t *)data, sizeof(uint16_t), 1);
         s->regs[SONIC_CRDA] = s->regs[SONIC_LLFA];
         s->regs[SONIC_ISR] |= SONIC_ISR_PKTRX;
         s->regs[SONIC_RSC] = (s->regs[SONIC_RSC] & 0xff00) | (((s->regs[SONIC_RSC] & 0x00ff) + 1) & 0x00ff);
 
         if (s->regs[SONIC_RCR] & SONIC_RCR_LPKT) {
             /* Read next RRA */
-            do_read_rra(s);
+            dp8393x_do_read_rra(s);
         }
     }
 
@@ -835,11 +781,12 @@ static ssize_t nic_receive(NetClientState *nc, const uint8_t * buf, size_t size)
     return size;
 }
 
-static void nic_reset(void *opaque)
+static void dp8393x_reset(DeviceState *dev)
 {
-    dp8393xState *s = opaque;
+    dp8393xState *s = DP8393X(dev);
     timer_del(s->watchdog);
 
+    memset(s->regs, 0, sizeof(s->regs));
     s->regs[SONIC_CR] = SONIC_CR_RST | SONIC_CR_STP | SONIC_CR_RXDIS;
     s->regs[SONIC_DCR] &= ~(SONIC_DCR_EXBUS | SONIC_DCR_LBR);
     s->regs[SONIC_RCR] &= ~(SONIC_RCR_LB0 | SONIC_RCR_LB1 | SONIC_RCR_BRD | SONIC_RCR_RNT);
@@ -862,39 +809,91 @@ static void nic_reset(void *opaque)
 static NetClientInfo net_dp83932_info = {
     .type = NET_CLIENT_OPTIONS_KIND_NIC,
     .size = sizeof(NICState),
-    .can_receive = nic_can_receive,
-    .receive = nic_receive,
+    .can_receive = dp8393x_can_receive,
+    .receive = dp8393x_receive,
 };
 
-void dp83932_init(NICInfo *nd, hwaddr base, int it_shift,
-                  MemoryRegion *address_space,
-                  qemu_irq irq, void* mem_opaque,
-                  void (*memory_rw)(void *opaque, hwaddr addr, uint8_t *buf, int len, int is_write))
+static void dp8393x_instance_init(Object *obj)
 {
-    dp8393xState *s;
+    SysBusDevice *sbd = SYS_BUS_DEVICE(obj);
+    dp8393xState *s = DP8393X(obj);
+
+    sysbus_init_mmio(sbd, &s->mmio);
+    sysbus_init_mmio(sbd, &s->prom);
+    sysbus_init_irq(sbd, &s->irq);
+}
 
-    qemu_check_nic_model(nd, "dp83932");
+static void dp8393x_realize(DeviceState *dev, Error **errp)
+{
+    dp8393xState *s = DP8393X(dev);
+    int i, checksum;
+    uint8_t *prom;
 
-    s = g_malloc0(sizeof(dp8393xState));
+    address_space_init(&s->as, s->dma_mr, "dp8393x");
+    memory_region_init_io(&s->mmio, OBJECT(dev), &dp8393x_ops, s,
+                          "dp8393x-regs", 0x40 << s->it_shift);
+
+    s->nic = qemu_new_nic(&net_dp83932_info, &s->conf,
+                          object_get_typename(OBJECT(dev)), dev->id, s);
+    qemu_format_nic_info_str(qemu_get_queue(s->nic), s->conf.macaddr.a);
 
-    s->address_space = address_space;
-    s->mem_opaque = mem_opaque;
-    s->memory_rw = memory_rw;
-    s->it_shift = it_shift;
-    s->irq = irq;
     s->watchdog = timer_new_ns(QEMU_CLOCK_VIRTUAL, dp8393x_watchdog, s);
     s->regs[SONIC_SR] = 0x0004; /* only revision recognized by Linux */
 
-    s->conf.macaddr = nd->macaddr;
-    s->conf.peers.ncs[0] = nd->netdev;
+    memory_region_init_rom_device(&s->prom, OBJECT(dev), NULL, NULL,
+                                  "dp8393x-prom", SONIC_PROM_SIZE, NULL);
+    prom = memory_region_get_ram_ptr(&s->prom);
+    checksum = 0;
+    for (i = 0; i < 6; i++) {
+        prom[i] = s->conf.macaddr.a[i];
+        checksum += prom[i];
+        if (checksum > 0xff) {
+            checksum = (checksum + 1) & 0xff;
+        }
+    }
+    prom[7] = 0xff - checksum;
+}
 
-    s->nic = qemu_new_nic(&net_dp83932_info, &s->conf, nd->model, nd->name, s);
+static const VMStateDescription vmstate_dp8393x = {
+    .name = "dp8393x",
+    .version_id = 0,
+    .minimum_version_id = 0,
+    .fields = (VMStateField []) {
+        VMSTATE_BUFFER_UNSAFE(cam, dp8393xState, 0, 16 * 6),
+        VMSTATE_UINT16_ARRAY(regs, dp8393xState, 0x40),
+        VMSTATE_END_OF_LIST()
+    }
+};
 
-    qemu_format_nic_info_str(qemu_get_queue(s->nic), s->conf.macaddr.a);
-    qemu_register_reset(nic_reset, s);
-    nic_reset(s);
+static Property dp8393x_properties[] = {
+    DEFINE_NIC_PROPERTIES(dp8393xState, conf),
+    DEFINE_PROP_PTR("dma_mr", dp8393xState, dma_mr),
+    DEFINE_PROP_UINT8("it_shift", dp8393xState, it_shift, 0),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void dp8393x_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
 
-    memory_region_init_io(&s->mmio, NULL, &dp8393x_ops, s,
-                          "dp8393x", 0x40 << it_shift);
-    memory_region_add_subregion(address_space, base, &s->mmio);
+    set_bit(DEVICE_CATEGORY_NETWORK, dc->categories);
+    dc->realize = dp8393x_realize;
+    dc->reset = dp8393x_reset;
+    dc->vmsd = &vmstate_dp8393x;
+    dc->props = dp8393x_properties;
 }
+
+static const TypeInfo dp8393x_info = {
+    .name          = TYPE_DP8393X,
+    .parent        = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(dp8393xState),
+    .instance_init = dp8393x_instance_init,
+    .class_init    = dp8393x_class_init,
+};
+
+static void dp8393x_register_types(void)
+{
+    type_register_static(&dp8393x_info);
+}
+
+type_init(dp8393x_register_types)
diff --git a/hw/net/e1000.c b/hw/net/e1000.c
index 091d61acc3..bab8e2abfb 100644
--- a/hw/net/e1000.c
+++ b/hw/net/e1000.c
@@ -1370,6 +1370,7 @@ static const VMStateDescription vmstate_e1000_mit_state = {
     .name = "e1000/mit_state",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = e1000_mit_state_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT32(mac_reg[RDTR], E1000State),
         VMSTATE_UINT32(mac_reg[RADV], E1000State),
@@ -1457,13 +1458,9 @@ static const VMStateDescription vmstate_e1000 = {
         VMSTATE_UINT32_SUB_ARRAY(mac_reg, E1000State, VFTA, 128),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_e1000_mit_state,
-            .needed = e1000_mit_state_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_e1000_mit_state,
+        NULL
     }
 };
 
diff --git a/hw/net/pcnet.c b/hw/net/pcnet.c
index bdfd38f4ca..68b9981983 100644
--- a/hw/net/pcnet.c
+++ b/hw/net/pcnet.c
@@ -1241,6 +1241,14 @@ static void pcnet_transmit(PCNetState *s)
         }
 
         bcnt = 4096 - GET_FIELD(tmd.length, TMDL, BCNT);
+
+        /* if multi-tmd packet outsizes s->buffer then skip it silently.
+           Note: this is not what real hw does */
+        if (s->xmit_pos + bcnt > sizeof(s->buffer)) {
+            s->xmit_pos = -1;
+            goto txdone;
+        }
+
         s->phys_mem_read(s->dma_opaque, PHYSADDR(s, tmd.tbadr),
                          s->buffer + s->xmit_pos, bcnt, CSR_BSWP(s));
         s->xmit_pos += bcnt;
diff --git a/hw/net/rocker/qmp-norocker.c b/hw/net/rocker/qmp-norocker.c
new file mode 100644
index 0000000000..f253747361
--- /dev/null
+++ b/hw/net/rocker/qmp-norocker.c
@@ -0,0 +1,50 @@
+/*
+ * QMP Target options - Commands handled based on a target config
+ *                      versus a host config
+ *
+ * Copyright (c) 2015 David Ahern <dsahern@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#include "qemu-common.h"
+#include "qmp-commands.h"
+#include "qapi/qmp/qerror.h"
+
+RockerSwitch *qmp_query_rocker(const char *name, Error **errp)
+{
+    error_set(errp, QERR_FEATURE_DISABLED, "rocker");
+    return NULL;
+};
+
+RockerPortList *qmp_query_rocker_ports(const char *name, Error **errp)
+{
+    error_set(errp, QERR_FEATURE_DISABLED, "rocker");
+    return NULL;
+};
+
+RockerOfDpaFlowList *qmp_query_rocker_of_dpa_flows(const char *name,
+                                                   bool has_tbl_id,
+                                                   uint32_t tbl_id,
+                                                   Error **errp)
+{
+    error_set(errp, QERR_FEATURE_DISABLED, "rocker");
+    return NULL;
+};
+
+RockerOfDpaGroupList *qmp_query_rocker_of_dpa_groups(const char *name,
+                                                     bool has_type,
+                                                     uint8_t type,
+                                                     Error **errp)
+{
+    error_set(errp, QERR_FEATURE_DISABLED, "rocker");
+    return NULL;
+};
diff --git a/hw/net/rocker/rocker.c b/hw/net/rocker/rocker.c
index 55b6c46157..4d25842509 100644
--- a/hw/net/rocker/rocker.c
+++ b/hw/net/rocker/rocker.c
@@ -94,6 +94,51 @@ World *rocker_get_world(Rocker *r, enum rocker_world_type type)
     return NULL;
 }
 
+RockerSwitch *qmp_query_rocker(const char *name, Error **errp)
+{
+    RockerSwitch *rocker = g_malloc0(sizeof(*rocker));
+    Rocker *r;
+
+    r = rocker_find(name);
+    if (!r) {
+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+                  "rocker %s not found", name);
+        return NULL;
+    }
+
+    rocker->name = g_strdup(r->name);
+    rocker->id = r->switch_id;
+    rocker->ports = r->fp_ports;
+
+    return rocker;
+}
+
+RockerPortList *qmp_query_rocker_ports(const char *name, Error **errp)
+{
+    RockerPortList *list = NULL;
+    Rocker *r;
+    int i;
+
+    r = rocker_find(name);
+    if (!r) {
+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+                  "rocker %s not found", name);
+        return NULL;
+    }
+
+    for (i = r->fp_ports - 1; i >= 0; i--) {
+        RockerPortList *info = g_malloc0(sizeof(*info));
+        info->value = g_malloc0(sizeof(*info->value));
+        struct fp_port *port = r->fp_port[i];
+
+        fp_port_get_info(port, info);
+        info->next = list;
+        list = info;
+    }
+
+    return list;
+}
+
 uint32_t rocker_fp_ports(Rocker *r)
 {
     return r->fp_ports;
@@ -238,6 +283,7 @@ static int cmd_get_port_settings(Rocker *r,
     uint8_t duplex;
     uint8_t autoneg;
     uint8_t learning;
+    char *phys_name;
     MACAddr macaddr;
     enum rocker_world_type mode;
     size_t tlv_size;
@@ -265,6 +311,7 @@ static int cmd_get_port_settings(Rocker *r,
     fp_port_get_macaddr(fp_port, &macaddr);
     mode = world_type(fp_port_get_world(fp_port));
     learning = fp_port_get_learning(fp_port);
+    phys_name = fp_port_get_name(fp_port);
 
     tlv_size = rocker_tlv_total_size(0) +                 /* nest */
                rocker_tlv_total_size(sizeof(uint32_t)) +  /*   pport */
@@ -273,7 +320,8 @@ static int cmd_get_port_settings(Rocker *r,
                rocker_tlv_total_size(sizeof(uint8_t)) +   /*   autoneg */
                rocker_tlv_total_size(sizeof(macaddr.a)) + /*   macaddr */
                rocker_tlv_total_size(sizeof(uint8_t)) +   /*   mode */
-               rocker_tlv_total_size(sizeof(uint8_t));    /*   learning */
+               rocker_tlv_total_size(sizeof(uint8_t)) +   /*   learning */
+               rocker_tlv_total_size(strlen(phys_name));
 
     if (tlv_size > desc_buf_size(info)) {
         return -ROCKER_EMSGSIZE;
@@ -290,6 +338,8 @@ static int cmd_get_port_settings(Rocker *r,
     rocker_tlv_put_u8(buf, &pos, ROCKER_TLV_CMD_PORT_SETTINGS_MODE, mode);
     rocker_tlv_put_u8(buf, &pos, ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,
                       learning);
+    rocker_tlv_put(buf, &pos, ROCKER_TLV_CMD_PORT_SETTINGS_PHYS_NAME,
+                   strlen(phys_name), phys_name);
     rocker_tlv_nest_end(buf, &pos, nest);
 
     return desc_set_buf(info, tlv_size);
@@ -1277,6 +1327,22 @@ static int pci_rocker_init(PCIDevice *dev)
         goto err_duplicate;
     }
 
+    /* Rocker name is passed in port name requests to OS with the intention
+     * that the name is used in interface names. Limit the length of the
+     * rocker name to avoid naming problems in the OS. Also, adding the
+     * port number as p# and unganged breakout b#, where # is at most 2
+     * digits, so leave room for it too (-1 for string terminator, -3 for
+     * p# and -3 for b#)
+     */
+#define ROCKER_IFNAMSIZ 16
+#define MAX_ROCKER_NAME_LEN  (ROCKER_IFNAMSIZ - 1 - 3 - 3)
+    if (strlen(r->name) > MAX_ROCKER_NAME_LEN) {
+        fprintf(stderr,
+                "rocker: name too long; please shorten to at most %d chars\n",
+                MAX_ROCKER_NAME_LEN);
+        return -EINVAL;
+    }
+
     if (memcmp(&r->fp_start_macaddr, &zero, sizeof(zero)) == 0) {
         memcpy(&r->fp_start_macaddr, &dflt, sizeof(dflt));
         r->fp_start_macaddr.a[4] += (sw_index++);
diff --git a/hw/net/rocker/rocker_fp.c b/hw/net/rocker/rocker_fp.c
index 2f1e3b348a..d8d934c396 100644
--- a/hw/net/rocker/rocker_fp.c
+++ b/hw/net/rocker/rocker_fp.c
@@ -41,11 +41,26 @@ struct fp_port {
     NICConf conf;
 };
 
+char *fp_port_get_name(FpPort *port)
+{
+    return port->name;
+}
+
 bool fp_port_get_link_up(FpPort *port)
 {
     return !qemu_get_queue(port->nic)->link_down;
 }
 
+void fp_port_get_info(FpPort *port, RockerPortList *info)
+{
+    info->value->name = g_strdup(port->name);
+    info->value->enabled = port->enabled;
+    info->value->link_up = fp_port_get_link_up(port);
+    info->value->speed = port->speed;
+    info->value->duplex = port->duplex;
+    info->value->autoneg = port->autoneg;
+}
+
 void fp_port_get_macaddr(FpPort *port, MACAddr *macaddr)
 {
     memcpy(macaddr->a, port->conf.macaddr.a, sizeof(macaddr->a));
@@ -173,8 +188,19 @@ bool fp_port_enabled(FpPort *port)
     return port->enabled;
 }
 
+static void fp_port_set_link(FpPort *port, bool up)
+{
+    NetClientState *nc = qemu_get_queue(port->nic);
+
+    if (up == nc->link_down) {
+        nc->link_down = !up;
+        nc->info->link_status_changed(nc);
+    }
+}
+
 void fp_port_enable(FpPort *port)
 {
+    fp_port_set_link(port, true);
     port->enabled = true;
     DPRINTF("port %d enabled\n", port->index);
 }
@@ -182,6 +208,7 @@ void fp_port_enable(FpPort *port)
 void fp_port_disable(FpPort *port)
 {
     port->enabled = false;
+    fp_port_set_link(port, false);
     DPRINTF("port %d disabled\n", port->index);
 }
 
@@ -201,7 +228,7 @@ FpPort *fp_port_alloc(Rocker *r, char *sw_name,
 
     /* front-panel switch port names are 1-based */
 
-    port->name = g_strdup_printf("%s.%d", sw_name, port->pport);
+    port->name = g_strdup_printf("%sp%d", sw_name, port->pport);
 
     memcpy(port->conf.macaddr.a, start_mac, sizeof(port->conf.macaddr.a));
     port->conf.macaddr.a[5] += index;
diff --git a/hw/net/rocker/rocker_fp.h b/hw/net/rocker/rocker_fp.h
index a5f28f120d..ab80fd833c 100644
--- a/hw/net/rocker/rocker_fp.h
+++ b/hw/net/rocker/rocker_fp.h
@@ -26,7 +26,9 @@ typedef struct fp_port FpPort;
 
 int fp_port_eg(FpPort *port, const struct iovec *iov, int iovcnt);
 
+char *fp_port_get_name(FpPort *port);
 bool fp_port_get_link_up(FpPort *port);
+void fp_port_get_info(FpPort *port, RockerPortList *info);
 void fp_port_get_macaddr(FpPort *port, MACAddr *macaddr);
 void fp_port_set_macaddr(FpPort *port, MACAddr *macaddr);
 uint8_t fp_port_get_learning(FpPort *port);
diff --git a/hw/net/rocker/rocker_hw.h b/hw/net/rocker/rocker_hw.h
index c9c85a75bd..fe639badd4 100644
--- a/hw/net/rocker/rocker_hw.h
+++ b/hw/net/rocker/rocker_hw.h
@@ -179,6 +179,7 @@ enum {
     ROCKER_TLV_CMD_PORT_SETTINGS_MACADDR,       /* binary */
     ROCKER_TLV_CMD_PORT_SETTINGS_MODE,          /* u8 */
     ROCKER_TLV_CMD_PORT_SETTINGS_LEARNING,      /* u8 */
+    ROCKER_TLV_CMD_PORT_SETTINGS_PHYS_NAME,     /* binary */
 
     __ROCKER_TLV_CMD_PORT_SETTINGS_MAX,
     ROCKER_TLV_CMD_PORT_SETTINGS_MAX = __ROCKER_TLV_CMD_PORT_SETTINGS_MAX - 1,
diff --git a/hw/net/rocker/rocker_of_dpa.c b/hw/net/rocker/rocker_of_dpa.c
index 1bcb7af5ef..b25a17d6d7 100644
--- a/hw/net/rocker/rocker_of_dpa.c
+++ b/hw/net/rocker/rocker_of_dpa.c
@@ -2302,6 +2302,318 @@ static void of_dpa_uninit(World *world)
     g_hash_table_destroy(of_dpa->flow_tbl);
 }
 
+struct of_dpa_flow_fill_context {
+    RockerOfDpaFlowList *list;
+    uint32_t tbl_id;
+};
+
+static void of_dpa_flow_fill(void *cookie, void *value, void *user_data)
+{
+    struct of_dpa_flow *flow = value;
+    struct of_dpa_flow_key *key = &flow->key;
+    struct of_dpa_flow_key *mask = &flow->mask;
+    struct of_dpa_flow_fill_context *flow_context = user_data;
+    RockerOfDpaFlowList *new;
+    RockerOfDpaFlow *nflow;
+    RockerOfDpaFlowKey *nkey;
+    RockerOfDpaFlowMask *nmask;
+    RockerOfDpaFlowAction *naction;
+
+    if (flow_context->tbl_id != -1 &&
+        flow_context->tbl_id != key->tbl_id) {
+        return;
+    }
+
+    new = g_malloc0(sizeof(*new));
+    nflow = new->value = g_malloc0(sizeof(*nflow));
+    nkey = nflow->key = g_malloc0(sizeof(*nkey));
+    nmask = nflow->mask = g_malloc0(sizeof(*nmask));
+    naction = nflow->action = g_malloc0(sizeof(*naction));
+
+    nflow->cookie = flow->cookie;
+    nflow->hits = flow->stats.hits;
+    nkey->priority = flow->priority;
+    nkey->tbl_id = key->tbl_id;
+
+    if (key->in_pport || mask->in_pport) {
+        nkey->has_in_pport = true;
+        nkey->in_pport = key->in_pport;
+    }
+
+    if (nkey->has_in_pport && mask->in_pport != 0xffffffff) {
+        nmask->has_in_pport = true;
+        nmask->in_pport = mask->in_pport;
+    }
+
+    if (key->eth.vlan_id || mask->eth.vlan_id) {
+        nkey->has_vlan_id = true;
+        nkey->vlan_id = ntohs(key->eth.vlan_id);
+    }
+
+    if (nkey->has_vlan_id && mask->eth.vlan_id != 0xffff) {
+        nmask->has_vlan_id = true;
+        nmask->vlan_id = ntohs(mask->eth.vlan_id);
+    }
+
+    if (key->tunnel_id || mask->tunnel_id) {
+        nkey->has_tunnel_id = true;
+        nkey->tunnel_id = key->tunnel_id;
+    }
+
+    if (nkey->has_tunnel_id && mask->tunnel_id != 0xffffffff) {
+        nmask->has_tunnel_id = true;
+        nmask->tunnel_id = mask->tunnel_id;
+    }
+
+    if (memcmp(key->eth.src.a, zero_mac.a, ETH_ALEN) ||
+        memcmp(mask->eth.src.a, zero_mac.a, ETH_ALEN)) {
+        nkey->has_eth_src = true;
+        nkey->eth_src = qemu_mac_strdup_printf(key->eth.src.a);
+    }
+
+    if (nkey->has_eth_src && memcmp(mask->eth.src.a, ff_mac.a, ETH_ALEN)) {
+        nmask->has_eth_src = true;
+        nmask->eth_src = qemu_mac_strdup_printf(mask->eth.src.a);
+    }
+
+    if (memcmp(key->eth.dst.a, zero_mac.a, ETH_ALEN) ||
+        memcmp(mask->eth.dst.a, zero_mac.a, ETH_ALEN)) {
+        nkey->has_eth_dst = true;
+        nkey->eth_dst = qemu_mac_strdup_printf(key->eth.dst.a);
+    }
+
+    if (nkey->has_eth_dst && memcmp(mask->eth.dst.a, ff_mac.a, ETH_ALEN)) {
+        nmask->has_eth_dst = true;
+        nmask->eth_dst = qemu_mac_strdup_printf(mask->eth.dst.a);
+    }
+
+    if (key->eth.type) {
+
+        nkey->has_eth_type = true;
+        nkey->eth_type = ntohs(key->eth.type);
+
+        switch (ntohs(key->eth.type)) {
+        case 0x0800:
+        case 0x86dd:
+            if (key->ip.proto || mask->ip.proto) {
+                nkey->has_ip_proto = true;
+                nkey->ip_proto = key->ip.proto;
+            }
+            if (nkey->has_ip_proto && mask->ip.proto != 0xff) {
+                nmask->has_ip_proto = true;
+                nmask->ip_proto = mask->ip.proto;
+            }
+            if (key->ip.tos || mask->ip.tos) {
+                nkey->has_ip_tos = true;
+                nkey->ip_tos = key->ip.tos;
+            }
+            if (nkey->has_ip_tos && mask->ip.tos != 0xff) {
+                nmask->has_ip_tos = true;
+                nmask->ip_tos = mask->ip.tos;
+            }
+            break;
+        }
+
+        switch (ntohs(key->eth.type)) {
+        case 0x0800:
+            if (key->ipv4.addr.dst || mask->ipv4.addr.dst) {
+                char *dst = inet_ntoa(*(struct in_addr *)&key->ipv4.addr.dst);
+                int dst_len = of_dpa_mask2prefix(mask->ipv4.addr.dst);
+                nkey->has_ip_dst = true;
+                nkey->ip_dst = g_strdup_printf("%s/%d", dst, dst_len);
+            }
+            break;
+        }
+    }
+
+    if (flow->action.goto_tbl) {
+        naction->has_goto_tbl = true;
+        naction->goto_tbl = flow->action.goto_tbl;
+    }
+
+    if (flow->action.write.group_id) {
+        naction->has_group_id = true;
+        naction->group_id = flow->action.write.group_id;
+    }
+
+    if (flow->action.apply.new_vlan_id) {
+        naction->has_new_vlan_id = true;
+        naction->new_vlan_id = flow->action.apply.new_vlan_id;
+    }
+
+    new->next = flow_context->list;
+    flow_context->list = new;
+}
+
+RockerOfDpaFlowList *qmp_query_rocker_of_dpa_flows(const char *name,
+                                                   bool has_tbl_id,
+                                                   uint32_t tbl_id,
+                                                   Error **errp)
+{
+    struct rocker *r;
+    struct world *w;
+    struct of_dpa *of_dpa;
+    struct of_dpa_flow_fill_context fill_context = {
+        .list = NULL,
+        .tbl_id = tbl_id,
+    };
+
+    r = rocker_find(name);
+    if (!r) {
+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+                  "rocker %s not found", name);
+        return NULL;
+    }
+
+    w = rocker_get_world(r, ROCKER_WORLD_TYPE_OF_DPA);
+    if (!w) {
+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+                  "rocker %s doesn't have OF-DPA world", name);
+        return NULL;
+    }
+
+    of_dpa = world_private(w);
+
+    g_hash_table_foreach(of_dpa->flow_tbl, of_dpa_flow_fill, &fill_context);
+
+    return fill_context.list;
+}
+
+struct of_dpa_group_fill_context {
+    RockerOfDpaGroupList *list;
+    uint8_t type;
+};
+
+static void of_dpa_group_fill(void *key, void *value, void *user_data)
+{
+    struct of_dpa_group *group = value;
+    struct of_dpa_group_fill_context *flow_context = user_data;
+    RockerOfDpaGroupList *new;
+    RockerOfDpaGroup *ngroup;
+    struct uint32List *id;
+    int i;
+
+    if (flow_context->type != 9 &&
+        flow_context->type != ROCKER_GROUP_TYPE_GET(group->id)) {
+        return;
+    }
+
+    new = g_malloc0(sizeof(*new));
+    ngroup = new->value = g_malloc0(sizeof(*ngroup));
+
+    ngroup->id = group->id;
+
+    ngroup->type = ROCKER_GROUP_TYPE_GET(group->id);
+
+    switch (ngroup->type) {
+    case ROCKER_OF_DPA_GROUP_TYPE_L2_INTERFACE:
+        ngroup->has_vlan_id = true;
+        ngroup->vlan_id = ROCKER_GROUP_VLAN_GET(group->id);
+        ngroup->has_pport = true;
+        ngroup->pport = ROCKER_GROUP_PORT_GET(group->id);
+        ngroup->has_out_pport = true;
+        ngroup->out_pport = group->l2_interface.out_pport;
+        ngroup->has_pop_vlan = true;
+        ngroup->pop_vlan = group->l2_interface.pop_vlan;
+        break;
+    case ROCKER_OF_DPA_GROUP_TYPE_L2_REWRITE:
+        ngroup->has_index = true;
+        ngroup->index = ROCKER_GROUP_INDEX_LONG_GET(group->id);
+        ngroup->has_group_id = true;
+        ngroup->group_id = group->l2_rewrite.group_id;
+        if (group->l2_rewrite.vlan_id) {
+            ngroup->has_set_vlan_id = true;
+            ngroup->set_vlan_id = ntohs(group->l2_rewrite.vlan_id);
+        }
+        break;
+        if (memcmp(group->l2_rewrite.src_mac.a, zero_mac.a, ETH_ALEN)) {
+            ngroup->has_set_eth_src = true;
+            ngroup->set_eth_src =
+                qemu_mac_strdup_printf(group->l2_rewrite.src_mac.a);
+        }
+        if (memcmp(group->l2_rewrite.dst_mac.a, zero_mac.a, ETH_ALEN)) {
+            ngroup->has_set_eth_dst = true;
+            ngroup->set_eth_dst =
+                qemu_mac_strdup_printf(group->l2_rewrite.dst_mac.a);
+        }
+    case ROCKER_OF_DPA_GROUP_TYPE_L2_FLOOD:
+    case ROCKER_OF_DPA_GROUP_TYPE_L2_MCAST:
+        ngroup->has_vlan_id = true;
+        ngroup->vlan_id = ROCKER_GROUP_VLAN_GET(group->id);
+        ngroup->has_index = true;
+        ngroup->index = ROCKER_GROUP_INDEX_GET(group->id);
+        for (i = 0; i < group->l2_flood.group_count; i++) {
+            ngroup->has_group_ids = true;
+            id = g_malloc0(sizeof(*id));
+            id->value = group->l2_flood.group_ids[i];
+            id->next = ngroup->group_ids;
+            ngroup->group_ids = id;
+        }
+        break;
+    case ROCKER_OF_DPA_GROUP_TYPE_L3_UCAST:
+        ngroup->has_index = true;
+        ngroup->index = ROCKER_GROUP_INDEX_LONG_GET(group->id);
+        ngroup->has_group_id = true;
+        ngroup->group_id = group->l3_unicast.group_id;
+        if (group->l3_unicast.vlan_id) {
+            ngroup->has_set_vlan_id = true;
+            ngroup->set_vlan_id = ntohs(group->l3_unicast.vlan_id);
+        }
+        if (memcmp(group->l3_unicast.src_mac.a, zero_mac.a, ETH_ALEN)) {
+            ngroup->has_set_eth_src = true;
+            ngroup->set_eth_src =
+                qemu_mac_strdup_printf(group->l3_unicast.src_mac.a);
+        }
+        if (memcmp(group->l3_unicast.dst_mac.a, zero_mac.a, ETH_ALEN)) {
+            ngroup->has_set_eth_dst = true;
+            ngroup->set_eth_dst =
+                qemu_mac_strdup_printf(group->l3_unicast.dst_mac.a);
+        }
+        if (group->l3_unicast.ttl_check) {
+            ngroup->has_ttl_check = true;
+            ngroup->ttl_check = group->l3_unicast.ttl_check;
+        }
+        break;
+    }
+
+    new->next = flow_context->list;
+    flow_context->list = new;
+}
+
+RockerOfDpaGroupList *qmp_query_rocker_of_dpa_groups(const char *name,
+                                                     bool has_type,
+                                                     uint8_t type,
+                                                     Error **errp)
+{
+    struct rocker *r;
+    struct world *w;
+    struct of_dpa *of_dpa;
+    struct of_dpa_group_fill_context fill_context = {
+        .list = NULL,
+        .type = type,
+    };
+
+    r = rocker_find(name);
+    if (!r) {
+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+                  "rocker %s not found", name);
+        return NULL;
+    }
+
+    w = rocker_get_world(r, ROCKER_WORLD_TYPE_OF_DPA);
+    if (!w) {
+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+                  "rocker %s doesn't have OF-DPA world", name);
+        return NULL;
+    }
+
+    of_dpa = world_private(w);
+
+    g_hash_table_foreach(of_dpa->group_tbl, of_dpa_group_fill, &fill_context);
+
+    return fill_context.list;
+}
+
 static WorldOps of_dpa_ops = {
     .init = of_dpa_init,
     .uninit = of_dpa_uninit,
diff --git a/hw/net/rtl8139.c b/hw/net/rtl8139.c
index f868108dfe..e0db4727ae 100644
--- a/hw/net/rtl8139.c
+++ b/hw/net/rtl8139.c
@@ -3240,6 +3240,7 @@ static const VMStateDescription vmstate_rtl8139_hotplug_ready ={
     .name = "rtl8139/hotplug_ready",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = rtl8139_hotplug_ready_needed,
     .fields = (VMStateField[]) {
         VMSTATE_END_OF_LIST()
     }
@@ -3335,13 +3336,9 @@ static const VMStateDescription vmstate_rtl8139 = {
         VMSTATE_UINT32_V(cplus_enabled, RTL8139State, 4),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_rtl8139_hotplug_ready,
-            .needed = rtl8139_hotplug_ready_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_rtl8139_hotplug_ready,
+        NULL
     }
 };
 
diff --git a/hw/net/vmxnet3.c b/hw/net/vmxnet3.c
index 34ffafd5a6..104a0f599b 100644
--- a/hw/net/vmxnet3.c
+++ b/hw/net/vmxnet3.c
@@ -2226,6 +2226,7 @@ static const VMStateDescription vmxstate_vmxnet3_mcast_list = {
     .version_id = 1,
     .minimum_version_id = 1,
     .pre_load = vmxnet3_mcast_list_pre_load,
+    .needed = vmxnet3_mc_list_needed,
     .fields = (VMStateField[]) {
         VMSTATE_VBUFFER_UINT32(mcast_list, VMXNET3State, 0, NULL, 0,
             mcast_list_buff_size),
@@ -2470,14 +2471,9 @@ static const VMStateDescription vmstate_vmxnet3 = {
 
             VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmxstate_vmxnet3_mcast_list,
-            .needed = vmxnet3_mc_list_needed
-        },
-        {
-            /* empty element. */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmxstate_vmxnet3_mcast_list,
+        NULL
     }
 };
 
diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c
index 68eff77983..88481b78c4 100644
--- a/hw/nvram/fw_cfg.c
+++ b/hw/nvram/fw_cfg.c
@@ -46,7 +46,6 @@ typedef struct FWCfgEntry {
     uint32_t len;
     uint8_t *data;
     void *callback_opaque;
-    FWCfgCallback callback;
     FWCfgReadCallback read_callback;
 } FWCfgEntry;
 
@@ -232,19 +231,7 @@ static void fw_cfg_reboot(FWCfgState *s)
 
 static void fw_cfg_write(FWCfgState *s, uint8_t value)
 {
-    int arch = !!(s->cur_entry & FW_CFG_ARCH_LOCAL);
-    FWCfgEntry *e = &s->entries[arch][s->cur_entry & FW_CFG_ENTRY_MASK];
-
-    trace_fw_cfg_write(s, value);
-
-    if (s->cur_entry & FW_CFG_WRITE_CHANNEL && e->callback &&
-        s->cur_offset < e->len) {
-        e->data[s->cur_offset++] = value;
-        if (s->cur_offset == e->len) {
-            e->callback(e->callback_opaque, e->data);
-            s->cur_offset = 0;
-        }
-    }
+    /* nothing, write support removed in QEMU v2.4+ */
 }
 
 static int fw_cfg_select(FWCfgState *s, uint16_t key)
@@ -436,6 +423,7 @@ static void fw_cfg_add_bytes_read_callback(FWCfgState *s, uint16_t key,
     key &= FW_CFG_ENTRY_MASK;
 
     assert(key < FW_CFG_MAX_ENTRY && len < UINT32_MAX);
+    assert(s->entries[arch][key].data == NULL); /* avoid key conflict */
 
     s->entries[arch][key].data = data;
     s->entries[arch][key].len = (uint32_t)len;
@@ -458,7 +446,6 @@ static void *fw_cfg_modify_bytes_read(FWCfgState *s, uint16_t key,
     s->entries[arch][key].data = data;
     s->entries[arch][key].len = len;
     s->entries[arch][key].callback_opaque = NULL;
-    s->entries[arch][key].callback = NULL;
 
     return ptr;
 }
@@ -484,6 +471,16 @@ void fw_cfg_add_i16(FWCfgState *s, uint16_t key, uint16_t value)
     fw_cfg_add_bytes(s, key, copy, sizeof(value));
 }
 
+void fw_cfg_modify_i16(FWCfgState *s, uint16_t key, uint16_t value)
+{
+    uint16_t *copy, *old;
+
+    copy = g_malloc(sizeof(value));
+    *copy = cpu_to_le16(value);
+    old = fw_cfg_modify_bytes_read(s, key, copy, sizeof(value));
+    g_free(old);
+}
+
 void fw_cfg_add_i32(FWCfgState *s, uint16_t key, uint32_t value)
 {
     uint32_t *copy;
@@ -502,23 +499,6 @@ void fw_cfg_add_i64(FWCfgState *s, uint16_t key, uint64_t value)
     fw_cfg_add_bytes(s, key, copy, sizeof(value));
 }
 
-void fw_cfg_add_callback(FWCfgState *s, uint16_t key, FWCfgCallback callback,
-                         void *callback_opaque, void *data, size_t len)
-{
-    int arch = !!(key & FW_CFG_ARCH_LOCAL);
-
-    assert(key & FW_CFG_WRITE_CHANNEL);
-
-    key &= FW_CFG_ENTRY_MASK;
-
-    assert(key < FW_CFG_MAX_ENTRY && len <= UINT32_MAX);
-
-    s->entries[arch][key].data = data;
-    s->entries[arch][key].len = (uint32_t)len;
-    s->entries[arch][key].callback_opaque = callback_opaque;
-    s->entries[arch][key].callback = callback;
-}
-
 void fw_cfg_add_file_callback(FWCfgState *s,  const char *filename,
                               FWCfgReadCallback callback, void *callback_opaque,
                               void *data, size_t len)
@@ -535,18 +515,19 @@ void fw_cfg_add_file_callback(FWCfgState *s,  const char *filename,
     index = be32_to_cpu(s->files->count);
     assert(index < FW_CFG_FILE_SLOTS);
 
-    fw_cfg_add_bytes_read_callback(s, FW_CFG_FILE_FIRST + index,
-                                   callback, callback_opaque, data, len);
-
     pstrcpy(s->files->f[index].name, sizeof(s->files->f[index].name),
             filename);
     for (i = 0; i < index; i++) {
         if (strcmp(s->files->f[index].name, s->files->f[i].name) == 0) {
-            trace_fw_cfg_add_file_dupe(s, s->files->f[index].name);
-            return;
+            error_report("duplicate fw_cfg file name: %s",
+                         s->files->f[index].name);
+            exit(1);
         }
     }
 
+    fw_cfg_add_bytes_read_callback(s, FW_CFG_FILE_FIRST + index,
+                                   callback, callback_opaque, data, len);
+
     s->files->f[index].size   = cpu_to_be32(len);
     s->files->f[index].select = cpu_to_be16(FW_CFG_FILE_FIRST + index);
     trace_fw_cfg_add_file(s, index, s->files->f[index].name, len);
diff --git a/hw/pci-host/pam.c b/hw/pci-host/pam.c
index 8272de3f28..17d826cba5 100644
--- a/hw/pci-host/pam.c
+++ b/hw/pci-host/pam.c
@@ -31,26 +31,6 @@
 #include "sysemu/sysemu.h"
 #include "hw/pci-host/pam.h"
 
-void smram_update(MemoryRegion *smram_region, uint8_t smram,
-                  uint8_t smm_enabled)
-{
-    bool smram_enabled;
-
-    smram_enabled = ((smm_enabled && (smram & SMRAM_G_SMRAME)) ||
-                        (smram & SMRAM_D_OPEN));
-    memory_region_set_enabled(smram_region, !smram_enabled);
-}
-
-void smram_set_smm(uint8_t *host_smm_enabled, int smm, uint8_t smram,
-                   MemoryRegion *smram_region)
-{
-    uint8_t smm_enabled = (smm != 0);
-    if (*host_smm_enabled != smm_enabled) {
-        *host_smm_enabled = smm_enabled;
-        smram_update(smram_region, smram, *host_smm_enabled);
-    }
-}
-
 void init_pam(DeviceState *dev, MemoryRegion *ram_memory,
               MemoryRegion *system_memory, MemoryRegion *pci_address_space,
               PAMMemoryRegion *mem, uint32_t start, uint32_t size)
diff --git a/hw/pci-host/piix.c b/hw/pci-host/piix.c
index 723836fb0e..ed2424c4cd 100644
--- a/hw/pci-host/piix.c
+++ b/hw/pci-host/piix.c
@@ -105,7 +105,7 @@ struct PCII440FXState {
     MemoryRegion *ram_memory;
     PAMMemoryRegion pam_regions[13];
     MemoryRegion smram_region;
-    uint8_t smm_enabled;
+    MemoryRegion smram, low_smram;
 };
 
 
@@ -138,18 +138,10 @@ static void i440fx_update_memory_mappings(PCII440FXState *d)
         pam_update(&d->pam_regions[i], i,
                    pd->config[I440FX_PAM + ((i + 1) / 2)]);
     }
-    smram_update(&d->smram_region, pd->config[I440FX_SMRAM], d->smm_enabled);
-    memory_region_transaction_commit();
-}
-
-static void i440fx_set_smm(int val, void *arg)
-{
-    PCII440FXState *d = arg;
-    PCIDevice *pd = PCI_DEVICE(d);
-
-    memory_region_transaction_begin();
-    smram_set_smm(&d->smm_enabled, val, pd->config[I440FX_SMRAM],
-                  &d->smram_region);
+    memory_region_set_enabled(&d->smram_region,
+                              !(pd->config[I440FX_SMRAM] & SMRAM_D_OPEN));
+    memory_region_set_enabled(&d->smram,
+                              pd->config[I440FX_SMRAM] & SMRAM_G_SMRAME);
     memory_region_transaction_commit();
 }
 
@@ -172,12 +164,13 @@ static int i440fx_load_old(QEMUFile* f, void *opaque, int version_id)
     PCII440FXState *d = opaque;
     PCIDevice *pd = PCI_DEVICE(d);
     int ret, i;
+    uint8_t smm_enabled;
 
     ret = pci_device_load(pd, f);
     if (ret < 0)
         return ret;
     i440fx_update_memory_mappings(d);
-    qemu_get_8s(f, &d->smm_enabled);
+    qemu_get_8s(f, &smm_enabled);
 
     if (version_id == 2) {
         for (i = 0; i < PIIX_NUM_PIRQS; i++) {
@@ -205,7 +198,10 @@ static const VMStateDescription vmstate_i440fx = {
     .post_load = i440fx_post_load,
     .fields = (VMStateField[]) {
         VMSTATE_PCI_DEVICE(parent_obj, PCII440FXState),
-        VMSTATE_UINT8(smm_enabled, PCII440FXState),
+        /* Used to be smm_enabled, which was basically always zero because
+         * SeaBIOS hardly uses SMM.  SMRAM is now handled by CPU code.
+         */
+        VMSTATE_UNUSED(1),
         VMSTATE_END_OF_LIST()
     }
 };
@@ -297,11 +293,7 @@ static void i440fx_pcihost_realize(DeviceState *dev, Error **errp)
 
 static void i440fx_realize(PCIDevice *dev, Error **errp)
 {
-    PCII440FXState *d = I440FX_PCI_DEVICE(dev);
-
     dev->config[I440FX_SMRAM] = 0x02;
-
-    cpu_smm_register(&i440fx_set_smm, d);
 }
 
 PCIBus *i440fx_init(PCII440FXState **pi440fx_state,
@@ -346,11 +338,23 @@ PCIBus *i440fx_init(PCII440FXState **pi440fx_state,
     pc_pci_as_mapping_init(OBJECT(f), f->system_memory,
                            f->pci_address_space);
 
+    /* if *disabled* show SMRAM to all CPUs */
     memory_region_init_alias(&f->smram_region, OBJECT(d), "smram-region",
                              f->pci_address_space, 0xa0000, 0x20000);
     memory_region_add_subregion_overlap(f->system_memory, 0xa0000,
                                         &f->smram_region, 1);
-    memory_region_set_enabled(&f->smram_region, false);
+    memory_region_set_enabled(&f->smram_region, true);
+
+    /* smram, as seen by SMM CPUs */
+    memory_region_init(&f->smram, OBJECT(d), "smram", 1ull << 32);
+    memory_region_set_enabled(&f->smram, true);
+    memory_region_init_alias(&f->low_smram, OBJECT(d), "smram-low",
+                             f->ram_memory, 0xa0000, 0x20000);
+    memory_region_set_enabled(&f->low_smram, true);
+    memory_region_add_subregion(&f->smram, 0xa0000, &f->low_smram);
+    object_property_add_const_link(qdev_get_machine(), "smram",
+                                   OBJECT(&f->smram), &error_abort);
+
     init_pam(dev, f->ram_memory, f->system_memory, f->pci_address_space,
              &f->pam_regions[0], PAM_BIOS_BASE, PAM_BIOS_SIZE);
     for (i = 0; i < 12; ++i) {
@@ -578,6 +582,7 @@ static const VMStateDescription vmstate_piix3_rcr = {
     .name = "PIIX3/rcr",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = piix3_rcr_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(rcr, PIIX3State),
         VMSTATE_END_OF_LIST()
@@ -596,12 +601,9 @@ static const VMStateDescription vmstate_piix3 = {
                               PIIX_NUM_PIRQS, 3),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_piix3_rcr,
-            .needed = piix3_rcr_needed,
-        },
-        { 0 }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_piix3_rcr,
+        NULL
     }
 };
 
diff --git a/hw/pci-host/q35.c b/hw/pci-host/q35.c
index c8827cc000..bd7409456f 100644
--- a/hw/pci-host/q35.c
+++ b/hw/pci-host/q35.c
@@ -198,6 +198,28 @@ static const TypeInfo q35_host_info = {
  * MCH D0:F0
  */
 
+static uint64_t tseg_blackhole_read(void *ptr, hwaddr reg, unsigned size)
+{
+    return 0xffffffff;
+}
+
+static void tseg_blackhole_write(void *opaque, hwaddr addr, uint64_t val,
+                                 unsigned width)
+{
+    /* nothing */
+}
+
+static const MemoryRegionOps tseg_blackhole_ops = {
+    .read = tseg_blackhole_read,
+    .write = tseg_blackhole_write,
+    .endianness = DEVICE_NATIVE_ENDIAN,
+    .valid.min_access_size = 1,
+    .valid.max_access_size = 4,
+    .impl.min_access_size = 4,
+    .impl.max_access_size = 4,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
 /* PCIe MMCFG */
 static void mch_update_pciexbar(MCHPCIState *mch)
 {
@@ -266,21 +288,70 @@ static void mch_update_pam(MCHPCIState *mch)
 static void mch_update_smram(MCHPCIState *mch)
 {
     PCIDevice *pd = PCI_DEVICE(mch);
+    bool h_smrame = (pd->config[MCH_HOST_BRIDGE_ESMRAMC] & MCH_HOST_BRIDGE_ESMRAMC_H_SMRAME);
+    uint32_t tseg_size;
+
+    /* implement SMRAM.D_LCK */
+    if (pd->config[MCH_HOST_BRIDGE_SMRAM] & MCH_HOST_BRIDGE_SMRAM_D_LCK) {
+        pd->config[MCH_HOST_BRIDGE_SMRAM] &= ~MCH_HOST_BRIDGE_SMRAM_D_OPEN;
+        pd->wmask[MCH_HOST_BRIDGE_SMRAM] = MCH_HOST_BRIDGE_SMRAM_WMASK_LCK;
+        pd->wmask[MCH_HOST_BRIDGE_ESMRAMC] = MCH_HOST_BRIDGE_ESMRAMC_WMASK_LCK;
+    }
 
     memory_region_transaction_begin();
-    smram_update(&mch->smram_region, pd->config[MCH_HOST_BRIDGE_SMRAM],
-                    mch->smm_enabled);
-    memory_region_transaction_commit();
-}
 
-static void mch_set_smm(int smm, void *arg)
-{
-    MCHPCIState *mch = arg;
-    PCIDevice *pd = PCI_DEVICE(mch);
+    if (pd->config[MCH_HOST_BRIDGE_SMRAM] & SMRAM_D_OPEN) {
+        /* Hide (!) low SMRAM if H_SMRAME = 1 */
+        memory_region_set_enabled(&mch->smram_region, h_smrame);
+        /* Show high SMRAM if H_SMRAME = 1 */
+        memory_region_set_enabled(&mch->open_high_smram, h_smrame);
+    } else {
+        /* Hide high SMRAM and low SMRAM */
+        memory_region_set_enabled(&mch->smram_region, true);
+        memory_region_set_enabled(&mch->open_high_smram, false);
+    }
+
+    if (pd->config[MCH_HOST_BRIDGE_SMRAM] & SMRAM_G_SMRAME) {
+        memory_region_set_enabled(&mch->low_smram, !h_smrame);
+        memory_region_set_enabled(&mch->high_smram, h_smrame);
+    } else {
+        memory_region_set_enabled(&mch->low_smram, false);
+        memory_region_set_enabled(&mch->high_smram, false);
+    }
+
+    if (pd->config[MCH_HOST_BRIDGE_ESMRAMC] & MCH_HOST_BRIDGE_ESMRAMC_T_EN) {
+        switch (pd->config[MCH_HOST_BRIDGE_ESMRAMC] &
+                MCH_HOST_BRIDGE_ESMRAMC_TSEG_SZ_MASK) {
+        case MCH_HOST_BRIDGE_ESMRAMC_TSEG_SZ_1MB:
+            tseg_size = 1024 * 1024;
+            break;
+        case MCH_HOST_BRIDGE_ESMRAMC_TSEG_SZ_2MB:
+            tseg_size = 1024 * 1024 * 2;
+            break;
+        case MCH_HOST_BRIDGE_ESMRAMC_TSEG_SZ_8MB:
+            tseg_size = 1024 * 1024 * 8;
+            break;
+        default:
+            tseg_size = 0;
+            break;
+        }
+    } else {
+        tseg_size = 0;
+    }
+    memory_region_del_subregion(mch->system_memory, &mch->tseg_blackhole);
+    memory_region_set_enabled(&mch->tseg_blackhole, tseg_size);
+    memory_region_set_size(&mch->tseg_blackhole, tseg_size);
+    memory_region_add_subregion_overlap(mch->system_memory,
+                                        mch->below_4g_mem_size - tseg_size,
+                                        &mch->tseg_blackhole, 1);
+
+    memory_region_set_enabled(&mch->tseg_window, tseg_size);
+    memory_region_set_size(&mch->tseg_window, tseg_size);
+    memory_region_set_address(&mch->tseg_window,
+                              mch->below_4g_mem_size - tseg_size);
+    memory_region_set_alias_offset(&mch->tseg_window,
+                                   mch->below_4g_mem_size - tseg_size);
 
-    memory_region_transaction_begin();
-    smram_set_smm(&mch->smm_enabled, smm, pd->config[MCH_HOST_BRIDGE_SMRAM],
-                    &mch->smram_region);
     memory_region_transaction_commit();
 }
 
@@ -289,7 +360,6 @@ static void mch_write_config(PCIDevice *d,
 {
     MCHPCIState *mch = MCH_PCI_DEVICE(d);
 
-    /* XXX: implement SMRAM.D_LOCK */
     pci_default_write_config(d, address, val, len);
 
     if (ranges_overlap(address, len, MCH_HOST_BRIDGE_PAM0,
@@ -329,7 +399,10 @@ static const VMStateDescription vmstate_mch = {
     .post_load = mch_post_load,
     .fields = (VMStateField[]) {
         VMSTATE_PCI_DEVICE(parent_obj, MCHPCIState),
-        VMSTATE_UINT8(smm_enabled, MCHPCIState),
+        /* Used to be smm_enabled, which was basically always zero because
+         * SeaBIOS hardly uses SMM.  SMRAM is now handled by CPU code.
+         */
+        VMSTATE_UNUSED(1),
         VMSTATE_END_OF_LIST()
     }
 };
@@ -343,6 +416,9 @@ static void mch_reset(DeviceState *qdev)
                  MCH_HOST_BRIDGE_PCIEXBAR_DEFAULT);
 
     d->config[MCH_HOST_BRIDGE_SMRAM] = MCH_HOST_BRIDGE_SMRAM_DEFAULT;
+    d->config[MCH_HOST_BRIDGE_ESMRAMC] = MCH_HOST_BRIDGE_ESMRAMC_DEFAULT;
+    d->wmask[MCH_HOST_BRIDGE_SMRAM] = MCH_HOST_BRIDGE_SMRAM_WMASK;
+    d->wmask[MCH_HOST_BRIDGE_ESMRAMC] = MCH_HOST_BRIDGE_ESMRAMC_WMASK;
 
     mch_update(mch);
 }
@@ -399,13 +475,47 @@ static void mch_realize(PCIDevice *d, Error **errp)
     pc_pci_as_mapping_init(OBJECT(mch), mch->system_memory,
                            mch->pci_address_space);
 
-    /* smram */
-    cpu_smm_register(&mch_set_smm, mch);
+    /* if *disabled* show SMRAM to all CPUs */
     memory_region_init_alias(&mch->smram_region, OBJECT(mch), "smram-region",
                              mch->pci_address_space, 0xa0000, 0x20000);
     memory_region_add_subregion_overlap(mch->system_memory, 0xa0000,
                                         &mch->smram_region, 1);
-    memory_region_set_enabled(&mch->smram_region, false);
+    memory_region_set_enabled(&mch->smram_region, true);
+
+    memory_region_init_alias(&mch->open_high_smram, OBJECT(mch), "smram-open-high",
+                             mch->ram_memory, 0xa0000, 0x20000);
+    memory_region_add_subregion_overlap(mch->system_memory, 0xfeda0000,
+                                        &mch->open_high_smram, 1);
+    memory_region_set_enabled(&mch->open_high_smram, false);
+
+    /* smram, as seen by SMM CPUs */
+    memory_region_init(&mch->smram, OBJECT(mch), "smram", 1ull << 32);
+    memory_region_set_enabled(&mch->smram, true);
+    memory_region_init_alias(&mch->low_smram, OBJECT(mch), "smram-low",
+                             mch->ram_memory, 0xa0000, 0x20000);
+    memory_region_set_enabled(&mch->low_smram, true);
+    memory_region_add_subregion(&mch->smram, 0xa0000, &mch->low_smram);
+    memory_region_init_alias(&mch->high_smram, OBJECT(mch), "smram-high",
+                             mch->ram_memory, 0xa0000, 0x20000);
+    memory_region_set_enabled(&mch->high_smram, true);
+    memory_region_add_subregion(&mch->smram, 0xfeda0000, &mch->high_smram);
+
+    memory_region_init_io(&mch->tseg_blackhole, OBJECT(mch),
+                          &tseg_blackhole_ops, NULL,
+                          "tseg-blackhole", 0);
+    memory_region_set_enabled(&mch->tseg_blackhole, false);
+    memory_region_add_subregion_overlap(mch->system_memory,
+                                        mch->below_4g_mem_size,
+                                        &mch->tseg_blackhole, 1);
+
+    memory_region_init_alias(&mch->tseg_window, OBJECT(mch), "tseg-window",
+                             mch->ram_memory, mch->below_4g_mem_size, 0);
+    memory_region_set_enabled(&mch->tseg_window, false);
+    memory_region_add_subregion(&mch->smram, mch->below_4g_mem_size,
+                                &mch->tseg_window);
+    object_property_add_const_link(qdev_get_machine(), "smram",
+                                   OBJECT(&mch->smram), &error_abort);
+
     init_pam(DEVICE(mch), mch->ram_memory, mch->system_memory,
              mch->pci_address_space, &mch->pam_regions[0],
              PAM_BIOS_BASE, PAM_BIOS_SIZE);
diff --git a/hw/pci/msi.c b/hw/pci/msi.c
index c111dbaff6..f9c0484420 100644
--- a/hw/pci/msi.c
+++ b/hw/pci/msi.c
@@ -21,10 +21,6 @@
 #include "hw/pci/msi.h"
 #include "qemu/range.h"
 
-/* Eventually those constants should go to Linux pci_regs.h */
-#define PCI_MSI_PENDING_32      0x10
-#define PCI_MSI_PENDING_64      0x14
-
 /* PCI_MSI_ADDRESS_LO */
 #define PCI_MSI_ADDRESS_LO_MASK         (~0x3)
 
diff --git a/hw/pci/pci-stub.c b/hw/pci/pci-stub.c
index 5e564c3a87..f8f237e823 100644
--- a/hw/pci/pci-stub.c
+++ b/hw/pci/pci-stub.c
@@ -29,19 +29,7 @@ PciInfoList *qmp_query_pci(Error **errp)
     return NULL;
 }
 
-static void pci_error_message(Monitor *mon)
+void hmp_pcie_aer_inject_error(Monitor *mon, const QDict *qdict)
 {
     monitor_printf(mon, "PCI devices not supported\n");
 }
-
-int hmp_pcie_aer_inject_error(Monitor *mon,
-                             const QDict *qdict, QObject **ret_data)
-{
-    pci_error_message(mon);
-    return -ENOSYS;
-}
-
-void pcie_aer_inject_error_print(Monitor *mon, const QObject *data)
-{
-    pci_error_message(mon);
-}
diff --git a/hw/pci/pci.c b/hw/pci/pci.c
index 4989408ece..21580433d4 100644
--- a/hw/pci/pci.c
+++ b/hw/pci/pci.c
@@ -147,7 +147,7 @@ static uint16_t pci_default_sub_device_id = PCI_SUBDEVICE_ID_QEMU;
 
 static QLIST_HEAD(, PCIHostState) pci_host_bridges;
 
-static int pci_bar(PCIDevice *d, int reg)
+int pci_bar(PCIDevice *d, int reg)
 {
     uint8_t type;
 
@@ -1698,6 +1698,8 @@ PCIDevice *pci_vga_init(PCIBus *bus)
         return pci_create_simple(bus, -1, "VGA");
     case VGA_VMWARE:
         return pci_create_simple(bus, -1, "vmware-svga");
+    case VGA_VIRTIO:
+        return pci_create_simple(bus, -1, "virtio-vga");
     case VGA_NONE:
     default: /* Other non-PCI types. Checking for unsupported types is already
                 done in vl.c. */
diff --git a/hw/pci/pcie_aer.c b/hw/pci/pcie_aer.c
index b48c09cd11..c8dea8ed9c 100644
--- a/hw/pci/pcie_aer.c
+++ b/hw/pci/pcie_aer.c
@@ -815,21 +815,6 @@ const VMStateDescription vmstate_pcie_aer_log = {
     }
 };
 
-void pcie_aer_inject_error_print(Monitor *mon, const QObject *data)
-{
-    QDict *qdict;
-    int devfn;
-    assert(qobject_type(data) == QTYPE_QDICT);
-    qdict = qobject_to_qdict(data);
-
-    devfn = (int)qdict_get_int(qdict, "devfn");
-    monitor_printf(mon, "OK id: %s root bus: %s, bus: %x devfn: %x.%x\n",
-                   qdict_get_str(qdict, "id"),
-                   qdict_get_str(qdict, "root_bus"),
-                   (int) qdict_get_int(qdict, "bus"),
-                   PCI_SLOT(devfn), PCI_FUNC(devfn));
-}
-
 typedef struct PCIEAERErrorName {
     const char *name;
     uint32_t val;
@@ -962,8 +947,8 @@ static int pcie_aer_parse_error_string(const char *error_name,
     return -EINVAL;
 }
 
-int hmp_pcie_aer_inject_error(Monitor *mon,
-                             const QDict *qdict, QObject **ret_data)
+static int do_pcie_aer_inject_error(Monitor *mon,
+                                    const QDict *qdict, QObject **ret_data)
 {
     const char *id = qdict_get_str(qdict, "id");
     const char *error_name;
@@ -1035,3 +1020,23 @@ int hmp_pcie_aer_inject_error(Monitor *mon,
 
     return 0;
 }
+
+void hmp_pcie_aer_inject_error(Monitor *mon, const QDict *qdict)
+{
+    QObject *data;
+    int devfn;
+
+    if (do_pcie_aer_inject_error(mon, qdict, &data) < 0) {
+        return;
+    }
+
+    assert(qobject_type(data) == QTYPE_QDICT);
+    qdict = qobject_to_qdict(data);
+
+    devfn = (int)qdict_get_int(qdict, "devfn");
+    monitor_printf(mon, "OK id: %s root bus: %s, bus: %x devfn: %x.%x\n",
+                   qdict_get_str(qdict, "id"),
+                   qdict_get_str(qdict, "root_bus"),
+                   (int) qdict_get_int(qdict, "bus"),
+                   PCI_SLOT(devfn), PCI_FUNC(devfn));
+}
diff --git a/hw/ppc/Makefile.objs b/hw/ppc/Makefile.objs
index 437955d1d5..c8ab06e7f3 100644
--- a/hw/ppc/Makefile.objs
+++ b/hw/ppc/Makefile.objs
@@ -3,7 +3,7 @@ obj-y += ppc.o ppc_booke.o
 # IBM pSeries (sPAPR)
 obj-$(CONFIG_PSERIES) += spapr.o spapr_vio.o spapr_events.o
 obj-$(CONFIG_PSERIES) += spapr_hcall.o spapr_iommu.o spapr_rtas.o
-obj-$(CONFIG_PSERIES) += spapr_pci.o spapr_rtc.o
+obj-$(CONFIG_PSERIES) += spapr_pci.o spapr_rtc.o spapr_drc.o
 ifeq ($(CONFIG_PCI)$(CONFIG_PSERIES)$(CONFIG_LINUX), yyy)
 obj-y += spapr_pci_vfio.o
 endif
diff --git a/hw/ppc/e500.c b/hw/ppc/e500.c
index c10e1b57b6..d300846c3d 100644
--- a/hw/ppc/e500.c
+++ b/hw/ppc/e500.c
@@ -1030,6 +1030,7 @@ void ppce500_init(MachineState *machine, PPCE500Params *params)
             exit(1);
         }
     }
+    g_free(filename);
 
     /* Reserve space for dtb */
     dt_base = (loadaddr + bios_size + DTC_LOAD_PAD) & ~DTC_PAD_MASK;
diff --git a/hw/ppc/mac_newworld.c b/hw/ppc/mac_newworld.c
index a365bf9223..0f3e34122a 100644
--- a/hw/ppc/mac_newworld.c
+++ b/hw/ppc/mac_newworld.c
@@ -119,7 +119,7 @@ static const MemoryRegionOps unin_ops = {
 static void fw_cfg_boot_set(void *opaque, const char *boot_device,
                             Error **errp)
 {
-    fw_cfg_add_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
+    fw_cfg_modify_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
 }
 
 static uint64_t translate_kernel_address(void *opaque, uint64_t addr)
diff --git a/hw/ppc/mac_oldworld.c b/hw/ppc/mac_oldworld.c
index f26133dedd..99879dd2d5 100644
--- a/hw/ppc/mac_oldworld.c
+++ b/hw/ppc/mac_oldworld.c
@@ -52,7 +52,7 @@
 static void fw_cfg_boot_set(void *opaque, const char *boot_device,
                             Error **errp)
 {
-    fw_cfg_add_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
+    fw_cfg_modify_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
 }
 
 static uint64_t translate_kernel_address(void *opaque, uint64_t addr)
diff --git a/hw/ppc/prep.c b/hw/ppc/prep.c
index 7f52662d76..998ee2d16b 100644
--- a/hw/ppc/prep.c
+++ b/hw/ppc/prep.c
@@ -528,7 +528,6 @@ static void ppc_prep_init(MachineState *machine)
     PCIDevice *pci;
     ISABus *isa_bus;
     ISADevice *isa;
-    qemu_irq *cpu_exit_irq;
     int ppc_boot_device;
     DriveInfo *hd[MAX_IDE_BUS * MAX_IDE_DEVS];
 
@@ -625,11 +624,11 @@ static void ppc_prep_init(MachineState *machine)
 
     /* PCI -> ISA bridge */
     pci = pci_create_simple(pci_bus, PCI_DEVFN(1, 0), "i82378");
-    cpu_exit_irq = qemu_allocate_irqs(cpu_request_exit, NULL, 1);
     cpu = POWERPC_CPU(first_cpu);
     qdev_connect_gpio_out(&pci->qdev, 0,
                           cpu->env.irq_inputs[PPC6xx_INPUT_INT]);
-    qdev_connect_gpio_out(&pci->qdev, 1, *cpu_exit_irq);
+    qdev_connect_gpio_out(&pci->qdev, 1,
+                          qemu_allocate_irq(cpu_request_exit, NULL, 0));
     sysbus_connect_irq(&pcihost->busdev, 0, qdev_get_gpio_in(&pci->qdev, 9));
     sysbus_connect_irq(&pcihost->busdev, 1, qdev_get_gpio_in(&pci->qdev, 11));
     sysbus_connect_irq(&pcihost->busdev, 2, qdev_get_gpio_in(&pci->qdev, 9));
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index a15fa3c965..f174e5a0f3 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -533,6 +533,8 @@ static void *spapr_create_fdt_skel(hwaddr initrd_base,
         refpoints, sizeof(refpoints))));
 
     _FDT((fdt_property_cell(fdt, "rtas-error-log-max", RTAS_ERROR_LOG_MAX)));
+    _FDT((fdt_property_cell(fdt, "rtas-event-scan-rate",
+                            RTAS_EVENT_SCAN_RATE)));
 
     /*
      * According to PAPR, rtas ibm,os-term does not guarantee a return
@@ -794,8 +796,8 @@ static void spapr_finalize_fdt(sPAPREnvironment *spapr,
     _FDT((fdt_pack(fdt)));
 
     if (fdt_totalsize(fdt) > FDT_MAX_SIZE) {
-        hw_error("FDT too big ! 0x%x bytes (max is 0x%x)\n",
-                 fdt_totalsize(fdt), FDT_MAX_SIZE);
+        error_report("FDT too big ! 0x%x bytes (max is 0x%x)",
+                     fdt_totalsize(fdt), FDT_MAX_SIZE);
         exit(1);
     }
 
@@ -899,7 +901,7 @@ static int spapr_check_htab_fd(sPAPREnvironment *spapr)
         spapr->htab_fd = kvmppc_get_htab_fd(false);
         if (spapr->htab_fd < 0) {
             error_report("Unable to open fd for reading hash table from KVM: "
-                    "%s", strerror(errno));
+                         "%s", strerror(errno));
             rc = -1;
         }
         spapr->htab_fd_stale = false;
@@ -1419,7 +1421,7 @@ static void ppc_spapr_init(MachineState *machine)
     rma_alloc_size = kvmppc_alloc_rma(&rma);
 
     if (rma_alloc_size == -1) {
-        hw_error("qemu: Unable to create RMA\n");
+        error_report("Unable to create RMA");
         exit(1);
     }
 
@@ -1504,6 +1506,11 @@ static void ppc_spapr_init(MachineState *machine)
         qemu_register_reset(spapr_cpu_reset, cpu);
     }
 
+    if (kvm_enabled()) {
+        /* Enable H_LOGICAL_CI_* so SLOF can talk to in-kernel devices */
+        kvmppc_enable_logical_ci_hcalls();
+    }
+
     /* allocate RAM */
     spapr->ram_limit = ram_size;
     memory_region_allocate_system_memory(ram, NULL, "ppc_spapr.ram",
@@ -1520,18 +1527,18 @@ static void ppc_spapr_init(MachineState *machine)
 
     filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, "spapr-rtas.bin");
     if (!filename) {
-        hw_error("Could not find LPAR rtas '%s'\n", "spapr-rtas.bin");
+        error_report("Could not find LPAR rtas '%s'", "spapr-rtas.bin");
         exit(1);
     }
     spapr->rtas_size = get_image_size(filename);
     spapr->rtas_blob = g_malloc(spapr->rtas_size);
     if (load_image_size(filename, spapr->rtas_blob, spapr->rtas_size) < 0) {
-        hw_error("qemu: could not load LPAR rtas '%s'\n", filename);
+        error_report("Could not load LPAR rtas '%s'", filename);
         exit(1);
     }
     if (spapr->rtas_size > RTAS_MAX_SIZE) {
-        hw_error("RTAS too big ! 0x%zx bytes (max is 0x%x)\n",
-                 (size_t)spapr->rtas_size, RTAS_MAX_SIZE);
+        error_report("RTAS too big ! 0x%zx bytes (max is 0x%x)",
+                     (size_t)spapr->rtas_size, RTAS_MAX_SIZE);
         exit(1);
     }
     g_free(filename);
@@ -1641,12 +1648,12 @@ static void ppc_spapr_init(MachineState *machine)
     }
     filename = qemu_find_file(QEMU_FILE_TYPE_BIOS, bios_name);
     if (!filename) {
-        hw_error("Could not find LPAR rtas '%s'\n", bios_name);
+        error_report("Could not find LPAR firmware '%s'", bios_name);
         exit(1);
     }
     fw_size = load_image_targphys(filename, 0, FW_MAX_SIZE);
-    if (fw_size < 0) {
-        hw_error("qemu: could not load LPAR rtas '%s'\n", filename);
+    if (fw_size <= 0) {
+        error_report("Could not load LPAR firmware '%s'", filename);
         exit(1);
     }
     g_free(filename);
@@ -1660,9 +1667,14 @@ static void ppc_spapr_init(MachineState *machine)
     /* Prepare the device tree */
     spapr->fdt_skel = spapr_create_fdt_skel(initrd_base, initrd_size,
                                             kernel_size, kernel_le,
-                                            kernel_cmdline, spapr->epow_irq);
+                                            kernel_cmdline,
+                                            spapr->check_exception_irq);
     assert(spapr->fdt_skel != NULL);
 
+    /* used by RTAS */
+    QTAILQ_INIT(&spapr->ccs_list);
+    qemu_register_reset(spapr_ccs_reset_hook, spapr);
+
     qemu_register_boot_set(spapr_boot_set, spapr);
 }
 
@@ -1794,6 +1806,7 @@ static void spapr_machine_class_init(ObjectClass *oc, void *data)
     mc->max_cpus = MAX_CPUS;
     mc->no_parallel = 1;
     mc->default_boot_order = "";
+    mc->default_ram_size = 512 * M_BYTE;
     mc->kvm_type = spapr_kvm_type;
     mc->has_dynamic_sysbus = true;
 
@@ -1816,7 +1829,12 @@ static const TypeInfo spapr_machine_info = {
 };
 
 #define SPAPR_COMPAT_2_3 \
-        HW_COMPAT_2_3
+        HW_COMPAT_2_3 \
+        {\
+            .driver   = "spapr-pci-host-bridge",\
+            .property = "dynamic-reconfiguration",\
+            .value    = "off",\
+        },
 
 #define SPAPR_COMPAT_2_2 \
         SPAPR_COMPAT_2_3 \
@@ -1905,10 +1923,15 @@ static const TypeInfo spapr_machine_2_2_info = {
 
 static void spapr_machine_2_3_class_init(ObjectClass *oc, void *data)
 {
+    static GlobalProperty compat_props[] = {
+        SPAPR_COMPAT_2_3
+        { /* end of list */ }
+    };
     MachineClass *mc = MACHINE_CLASS(oc);
 
     mc->name = "pseries-2.3";
     mc->desc = "pSeries Logical Partition (PAPR compliant) v2.3";
+    mc->compat_props = compat_props;
 }
 
 static const TypeInfo spapr_machine_2_3_info = {
diff --git a/hw/ppc/spapr_drc.c b/hw/ppc/spapr_drc.c
new file mode 100644
index 0000000000..ef985381cb
--- /dev/null
+++ b/hw/ppc/spapr_drc.c
@@ -0,0 +1,744 @@
+/*
+ * QEMU SPAPR Dynamic Reconfiguration Connector Implementation
+ *
+ * Copyright IBM Corp. 2014
+ *
+ * Authors:
+ *  Michael Roth      <mdroth@linux.vnet.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "hw/ppc/spapr_drc.h"
+#include "qom/object.h"
+#include "hw/qdev.h"
+#include "qapi/visitor.h"
+#include "qemu/error-report.h"
+
+/* #define DEBUG_SPAPR_DRC */
+
+#ifdef DEBUG_SPAPR_DRC
+#define DPRINTF(fmt, ...) \
+    do { fprintf(stderr, fmt, ## __VA_ARGS__); } while (0)
+#define DPRINTFN(fmt, ...) \
+    do { DPRINTF(fmt, ## __VA_ARGS__); fprintf(stderr, "\n"); } while (0)
+#else
+#define DPRINTF(fmt, ...) \
+    do { } while (0)
+#define DPRINTFN(fmt, ...) \
+    do { } while (0)
+#endif
+
+#define DRC_CONTAINER_PATH "/dr-connector"
+#define DRC_INDEX_TYPE_SHIFT 28
+#define DRC_INDEX_ID_MASK (~(~0 << DRC_INDEX_TYPE_SHIFT))
+
+static sPAPRDRConnectorTypeShift get_type_shift(sPAPRDRConnectorType type)
+{
+    uint32_t shift = 0;
+
+    /* make sure this isn't SPAPR_DR_CONNECTOR_TYPE_ANY, or some
+     * other wonky value.
+     */
+    g_assert(is_power_of_2(type));
+
+    while (type != (1 << shift)) {
+        shift++;
+    }
+    return shift;
+}
+
+static uint32_t get_index(sPAPRDRConnector *drc)
+{
+    /* no set format for a drc index: it only needs to be globally
+     * unique. this is how we encode the DRC type on bare-metal
+     * however, so might as well do that here
+     */
+    return (get_type_shift(drc->type) << DRC_INDEX_TYPE_SHIFT) |
+            (drc->id & DRC_INDEX_ID_MASK);
+}
+
+static int set_isolation_state(sPAPRDRConnector *drc,
+                               sPAPRDRIsolationState state)
+{
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+
+    DPRINTFN("drc: %x, set_isolation_state: %x", get_index(drc), state);
+
+    drc->isolation_state = state;
+
+    if (drc->isolation_state == SPAPR_DR_ISOLATION_STATE_ISOLATED) {
+        /* if we're awaiting release, but still in an unconfigured state,
+         * it's likely the guest is still in the process of configuring
+         * the device and is transitioning the devices to an ISOLATED
+         * state as a part of that process. so we only complete the
+         * removal when this transition happens for a device in a
+         * configured state, as suggested by the state diagram from
+         * PAPR+ 2.7, 13.4
+         */
+        if (drc->awaiting_release) {
+            if (drc->configured) {
+                DPRINTFN("finalizing device removal");
+                drck->detach(drc, DEVICE(drc->dev), drc->detach_cb,
+                             drc->detach_cb_opaque, NULL);
+            } else {
+                DPRINTFN("deferring device removal on unconfigured device\n");
+            }
+        }
+        drc->configured = false;
+    }
+
+    return 0;
+}
+
+static int set_indicator_state(sPAPRDRConnector *drc,
+                               sPAPRDRIndicatorState state)
+{
+    DPRINTFN("drc: %x, set_indicator_state: %x", get_index(drc), state);
+    drc->indicator_state = state;
+    return 0;
+}
+
+static int set_allocation_state(sPAPRDRConnector *drc,
+                                sPAPRDRAllocationState state)
+{
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+
+    DPRINTFN("drc: %x, set_allocation_state: %x", get_index(drc), state);
+
+    if (drc->type != SPAPR_DR_CONNECTOR_TYPE_PCI) {
+        drc->allocation_state = state;
+        if (drc->awaiting_release &&
+            drc->allocation_state == SPAPR_DR_ALLOCATION_STATE_UNUSABLE) {
+            DPRINTFN("finalizing device removal");
+            drck->detach(drc, DEVICE(drc->dev), drc->detach_cb,
+                         drc->detach_cb_opaque, NULL);
+        }
+    }
+    return 0;
+}
+
+static uint32_t get_type(sPAPRDRConnector *drc)
+{
+    return drc->type;
+}
+
+static const char *get_name(sPAPRDRConnector *drc)
+{
+    return drc->name;
+}
+
+static const void *get_fdt(sPAPRDRConnector *drc, int *fdt_start_offset)
+{
+    if (fdt_start_offset) {
+        *fdt_start_offset = drc->fdt_start_offset;
+    }
+    return drc->fdt;
+}
+
+static void set_configured(sPAPRDRConnector *drc)
+{
+    DPRINTFN("drc: %x, set_configured", get_index(drc));
+
+    if (drc->isolation_state != SPAPR_DR_ISOLATION_STATE_UNISOLATED) {
+        /* guest should be not configuring an isolated device */
+        DPRINTFN("drc: %x, set_configured: skipping isolated device",
+                 get_index(drc));
+        return;
+    }
+    drc->configured = true;
+}
+
+/*
+ * dr-entity-sense sensor value
+ * returned via get-sensor-state RTAS calls
+ * as expected by state diagram in PAPR+ 2.7, 13.4
+ * based on the current allocation/indicator/power states
+ * for the DR connector.
+ */
+static sPAPRDREntitySense entity_sense(sPAPRDRConnector *drc)
+{
+    sPAPRDREntitySense state;
+
+    if (drc->dev) {
+        if (drc->type != SPAPR_DR_CONNECTOR_TYPE_PCI &&
+            drc->allocation_state == SPAPR_DR_ALLOCATION_STATE_UNUSABLE) {
+            /* for logical DR, we return a state of UNUSABLE
+             * iff the allocation state UNUSABLE.
+             * Otherwise, report the state as USABLE/PRESENT,
+             * as we would for PCI.
+             */
+            state = SPAPR_DR_ENTITY_SENSE_UNUSABLE;
+        } else {
+            /* this assumes all PCI devices are assigned to
+             * a 'live insertion' power domain, where QEMU
+             * manages power state automatically as opposed
+             * to the guest. present, non-PCI resources are
+             * unaffected by power state.
+             */
+            state = SPAPR_DR_ENTITY_SENSE_PRESENT;
+        }
+    } else {
+        if (drc->type == SPAPR_DR_CONNECTOR_TYPE_PCI) {
+            /* PCI devices, and only PCI devices, use EMPTY
+             * in cases where we'd otherwise use UNUSABLE
+             */
+            state = SPAPR_DR_ENTITY_SENSE_EMPTY;
+        } else {
+            state = SPAPR_DR_ENTITY_SENSE_UNUSABLE;
+        }
+    }
+
+    DPRINTFN("drc: %x, entity_sense: %x", get_index(drc), state);
+    return state;
+}
+
+static void prop_get_index(Object *obj, Visitor *v, void *opaque,
+                                  const char *name, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(obj);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    uint32_t value = (uint32_t)drck->get_index(drc);
+    visit_type_uint32(v, &value, name, errp);
+}
+
+static void prop_get_type(Object *obj, Visitor *v, void *opaque,
+                          const char *name, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(obj);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    uint32_t value = (uint32_t)drck->get_type(drc);
+    visit_type_uint32(v, &value, name, errp);
+}
+
+static char *prop_get_name(Object *obj, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(obj);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    return g_strdup(drck->get_name(drc));
+}
+
+static void prop_get_entity_sense(Object *obj, Visitor *v, void *opaque,
+                                  const char *name, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(obj);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    uint32_t value = (uint32_t)drck->entity_sense(drc);
+    visit_type_uint32(v, &value, name, errp);
+}
+
+static void prop_get_fdt(Object *obj, Visitor *v, void *opaque,
+                        const char *name, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(obj);
+    int fdt_offset_next, fdt_offset, fdt_depth;
+    void *fdt;
+
+    if (!drc->fdt) {
+        return;
+    }
+
+    fdt = drc->fdt;
+    fdt_offset = drc->fdt_start_offset;
+    fdt_depth = 0;
+
+    do {
+        const char *name = NULL;
+        const struct fdt_property *prop = NULL;
+        int prop_len = 0, name_len = 0;
+        uint32_t tag;
+
+        tag = fdt_next_tag(fdt, fdt_offset, &fdt_offset_next);
+        switch (tag) {
+        case FDT_BEGIN_NODE:
+            fdt_depth++;
+            name = fdt_get_name(fdt, fdt_offset, &name_len);
+            visit_start_struct(v, NULL, NULL, name, 0, NULL);
+            break;
+        case FDT_END_NODE:
+            /* shouldn't ever see an FDT_END_NODE before FDT_BEGIN_NODE */
+            g_assert(fdt_depth > 0);
+            visit_end_struct(v, NULL);
+            fdt_depth--;
+            break;
+        case FDT_PROP: {
+            int i;
+            prop = fdt_get_property_by_offset(fdt, fdt_offset, &prop_len);
+            name = fdt_string(fdt, fdt32_to_cpu(prop->nameoff));
+            visit_start_list(v, name, NULL);
+            for (i = 0; i < prop_len; i++) {
+                visit_type_uint8(v, (uint8_t *)&prop->data[i], NULL, NULL);
+
+            }
+            visit_end_list(v, NULL);
+            break;
+        }
+        default:
+            error_setg(&error_abort, "device FDT in unexpected state: %d", tag);
+        }
+        fdt_offset = fdt_offset_next;
+    } while (fdt_depth != 0);
+}
+
+static void attach(sPAPRDRConnector *drc, DeviceState *d, void *fdt,
+                   int fdt_start_offset, bool coldplug, Error **errp)
+{
+    DPRINTFN("drc: %x, attach", get_index(drc));
+
+    if (drc->isolation_state != SPAPR_DR_ISOLATION_STATE_ISOLATED) {
+        error_setg(errp, "an attached device is still awaiting release");
+        return;
+    }
+    if (drc->type == SPAPR_DR_CONNECTOR_TYPE_PCI) {
+        g_assert(drc->allocation_state == SPAPR_DR_ALLOCATION_STATE_USABLE);
+    }
+    g_assert(fdt || coldplug);
+
+    /* NOTE: setting initial isolation state to UNISOLATED means we can't
+     * detach unless guest has a userspace/kernel that moves this state
+     * back to ISOLATED in response to an unplug event, or this is done
+     * manually by the admin prior. if we force things while the guest
+     * may be accessing the device, we can easily crash the guest, so we
+     * we defer completion of removal in such cases to the reset() hook.
+     */
+    if (drc->type == SPAPR_DR_CONNECTOR_TYPE_PCI) {
+        drc->isolation_state = SPAPR_DR_ISOLATION_STATE_UNISOLATED;
+    }
+    drc->indicator_state = SPAPR_DR_INDICATOR_STATE_ACTIVE;
+
+    drc->dev = d;
+    drc->fdt = fdt;
+    drc->fdt_start_offset = fdt_start_offset;
+    drc->configured = false;
+
+    object_property_add_link(OBJECT(drc), "device",
+                             object_get_typename(OBJECT(drc->dev)),
+                             (Object **)(&drc->dev),
+                             NULL, 0, NULL);
+}
+
+static void detach(sPAPRDRConnector *drc, DeviceState *d,
+                   spapr_drc_detach_cb *detach_cb,
+                   void *detach_cb_opaque, Error **errp)
+{
+    DPRINTFN("drc: %x, detach", get_index(drc));
+
+    drc->detach_cb = detach_cb;
+    drc->detach_cb_opaque = detach_cb_opaque;
+
+    if (drc->isolation_state != SPAPR_DR_ISOLATION_STATE_ISOLATED) {
+        DPRINTFN("awaiting transition to isolated state before removal");
+        drc->awaiting_release = true;
+        return;
+    }
+
+    if (drc->type != SPAPR_DR_CONNECTOR_TYPE_PCI &&
+        drc->allocation_state != SPAPR_DR_ALLOCATION_STATE_UNUSABLE) {
+        DPRINTFN("awaiting transition to unusable state before removal");
+        drc->awaiting_release = true;
+        return;
+    }
+
+    drc->indicator_state = SPAPR_DR_INDICATOR_STATE_INACTIVE;
+
+    if (drc->detach_cb) {
+        drc->detach_cb(drc->dev, drc->detach_cb_opaque);
+    }
+
+    drc->awaiting_release = false;
+    g_free(drc->fdt);
+    drc->fdt = NULL;
+    drc->fdt_start_offset = 0;
+    object_property_del(OBJECT(drc), "device", NULL);
+    drc->dev = NULL;
+    drc->detach_cb = NULL;
+    drc->detach_cb_opaque = NULL;
+}
+
+static bool release_pending(sPAPRDRConnector *drc)
+{
+    return drc->awaiting_release;
+}
+
+static void reset(DeviceState *d)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(d);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+
+    DPRINTFN("drc reset: %x", drck->get_index(drc));
+    /* immediately upon reset we can safely assume DRCs whose devices
+     * are pending removal can be safely removed, and that they will
+     * subsequently be left in an ISOLATED state. move the DRC to this
+     * state in these cases (which will in turn complete any pending
+     * device removals)
+     */
+    if (drc->awaiting_release) {
+        drck->set_isolation_state(drc, SPAPR_DR_ISOLATION_STATE_ISOLATED);
+        /* generally this should also finalize the removal, but if the device
+         * hasn't yet been configured we normally defer removal under the
+         * assumption that this transition is taking place as part of device
+         * configuration. so check if we're still waiting after this, and
+         * force removal if we are
+         */
+        if (drc->awaiting_release) {
+            drck->detach(drc, DEVICE(drc->dev), drc->detach_cb,
+                         drc->detach_cb_opaque, NULL);
+        }
+
+        /* non-PCI devices may be awaiting a transition to UNUSABLE */
+        if (drc->type != SPAPR_DR_CONNECTOR_TYPE_PCI &&
+            drc->awaiting_release) {
+            drck->set_allocation_state(drc, SPAPR_DR_ALLOCATION_STATE_UNUSABLE);
+        }
+    }
+}
+
+static void realize(DeviceState *d, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(d);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    Object *root_container;
+    char link_name[256];
+    gchar *child_name;
+    Error *err = NULL;
+
+    DPRINTFN("drc realize: %x", drck->get_index(drc));
+    /* NOTE: we do this as part of realize/unrealize due to the fact
+     * that the guest will communicate with the DRC via RTAS calls
+     * referencing the global DRC index. By unlinking the DRC
+     * from DRC_CONTAINER_PATH/<drc_index> we effectively make it
+     * inaccessible by the guest, since lookups rely on this path
+     * existing in the composition tree
+     */
+    root_container = container_get(object_get_root(), DRC_CONTAINER_PATH);
+    snprintf(link_name, sizeof(link_name), "%x", drck->get_index(drc));
+    child_name = object_get_canonical_path_component(OBJECT(drc));
+    DPRINTFN("drc child name: %s", child_name);
+    object_property_add_alias(root_container, link_name,
+                              drc->owner, child_name, &err);
+    if (err) {
+        error_report("%s", error_get_pretty(err));
+        error_free(err);
+        object_unref(OBJECT(drc));
+    }
+    DPRINTFN("drc realize complete");
+}
+
+static void unrealize(DeviceState *d, Error **errp)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(d);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    Object *root_container;
+    char name[256];
+    Error *err = NULL;
+
+    DPRINTFN("drc unrealize: %x", drck->get_index(drc));
+    root_container = container_get(object_get_root(), DRC_CONTAINER_PATH);
+    snprintf(name, sizeof(name), "%x", drck->get_index(drc));
+    object_property_del(root_container, name, &err);
+    if (err) {
+        error_report("%s", error_get_pretty(err));
+        error_free(err);
+        object_unref(OBJECT(drc));
+    }
+}
+
+sPAPRDRConnector *spapr_dr_connector_new(Object *owner,
+                                         sPAPRDRConnectorType type,
+                                         uint32_t id)
+{
+    sPAPRDRConnector *drc =
+        SPAPR_DR_CONNECTOR(object_new(TYPE_SPAPR_DR_CONNECTOR));
+
+    g_assert(type);
+
+    drc->type = type;
+    drc->id = id;
+    drc->owner = owner;
+    object_property_add_child(owner, "dr-connector[*]", OBJECT(drc), NULL);
+    object_property_set_bool(OBJECT(drc), true, "realized", NULL);
+
+    /* human-readable name for a DRC to encode into the DT
+     * description. this is mainly only used within a guest in place
+     * of the unique DRC index.
+     *
+     * in the case of VIO/PCI devices, it corresponds to a
+     * "location code" that maps a logical device/function (DRC index)
+     * to a physical (or virtual in the case of VIO) location in the
+     * system by chaining together the "location label" for each
+     * encapsulating component.
+     *
+     * since this is more to do with diagnosing physical hardware
+     * issues than guest compatibility, we choose location codes/DRC
+     * names that adhere to the documented format, but avoid encoding
+     * the entire topology information into the label/code, instead
+     * just using the location codes based on the labels for the
+     * endpoints (VIO/PCI adaptor connectors), which is basically
+     * just "C" followed by an integer ID.
+     *
+     * DRC names as documented by PAPR+ v2.7, 13.5.2.4
+     * location codes as documented by PAPR+ v2.7, 12.3.1.5
+     */
+    switch (drc->type) {
+    case SPAPR_DR_CONNECTOR_TYPE_CPU:
+        drc->name = g_strdup_printf("CPU %d", id);
+        break;
+    case SPAPR_DR_CONNECTOR_TYPE_PHB:
+        drc->name = g_strdup_printf("PHB %d", id);
+        break;
+    case SPAPR_DR_CONNECTOR_TYPE_VIO:
+    case SPAPR_DR_CONNECTOR_TYPE_PCI:
+        drc->name = g_strdup_printf("C%d", id);
+        break;
+    case SPAPR_DR_CONNECTOR_TYPE_LMB:
+        drc->name = g_strdup_printf("LMB %d", id);
+        break;
+    default:
+        g_assert(false);
+    }
+
+    /* PCI slot always start in a USABLE state, and stay there */
+    if (drc->type == SPAPR_DR_CONNECTOR_TYPE_PCI) {
+        drc->allocation_state = SPAPR_DR_ALLOCATION_STATE_USABLE;
+    }
+
+    return drc;
+}
+
+static void spapr_dr_connector_instance_init(Object *obj)
+{
+    sPAPRDRConnector *drc = SPAPR_DR_CONNECTOR(obj);
+
+    object_property_add_uint32_ptr(obj, "isolation-state",
+                                   &drc->isolation_state, NULL);
+    object_property_add_uint32_ptr(obj, "indicator-state",
+                                   &drc->indicator_state, NULL);
+    object_property_add_uint32_ptr(obj, "allocation-state",
+                                   &drc->allocation_state, NULL);
+    object_property_add_uint32_ptr(obj, "id", &drc->id, NULL);
+    object_property_add(obj, "index", "uint32", prop_get_index,
+                        NULL, NULL, NULL, NULL);
+    object_property_add(obj, "connector_type", "uint32", prop_get_type,
+                        NULL, NULL, NULL, NULL);
+    object_property_add_str(obj, "name", prop_get_name, NULL, NULL);
+    object_property_add(obj, "entity-sense", "uint32", prop_get_entity_sense,
+                        NULL, NULL, NULL, NULL);
+    object_property_add(obj, "fdt", "struct", prop_get_fdt,
+                        NULL, NULL, NULL, NULL);
+}
+
+static void spapr_dr_connector_class_init(ObjectClass *k, void *data)
+{
+    DeviceClass *dk = DEVICE_CLASS(k);
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_CLASS(k);
+
+    dk->reset = reset;
+    dk->realize = realize;
+    dk->unrealize = unrealize;
+    drck->set_isolation_state = set_isolation_state;
+    drck->set_indicator_state = set_indicator_state;
+    drck->set_allocation_state = set_allocation_state;
+    drck->get_index = get_index;
+    drck->get_type = get_type;
+    drck->get_name = get_name;
+    drck->get_fdt = get_fdt;
+    drck->set_configured = set_configured;
+    drck->entity_sense = entity_sense;
+    drck->attach = attach;
+    drck->detach = detach;
+    drck->release_pending = release_pending;
+}
+
+static const TypeInfo spapr_dr_connector_info = {
+    .name          = TYPE_SPAPR_DR_CONNECTOR,
+    .parent        = TYPE_DEVICE,
+    .instance_size = sizeof(sPAPRDRConnector),
+    .instance_init = spapr_dr_connector_instance_init,
+    .class_size    = sizeof(sPAPRDRConnectorClass),
+    .class_init    = spapr_dr_connector_class_init,
+};
+
+static void spapr_drc_register_types(void)
+{
+    type_register_static(&spapr_dr_connector_info);
+}
+
+type_init(spapr_drc_register_types)
+
+/* helper functions for external users */
+
+sPAPRDRConnector *spapr_dr_connector_by_index(uint32_t index)
+{
+    Object *obj;
+    char name[256];
+
+    snprintf(name, sizeof(name), "%s/%x", DRC_CONTAINER_PATH, index);
+    obj = object_resolve_path(name, NULL);
+
+    return !obj ? NULL : SPAPR_DR_CONNECTOR(obj);
+}
+
+sPAPRDRConnector *spapr_dr_connector_by_id(sPAPRDRConnectorType type,
+                                           uint32_t id)
+{
+    return spapr_dr_connector_by_index(
+            (get_type_shift(type) << DRC_INDEX_TYPE_SHIFT) |
+            (id & DRC_INDEX_ID_MASK));
+}
+
+/* generate a string the describes the DRC to encode into the
+ * device tree.
+ *
+ * as documented by PAPR+ v2.7, 13.5.2.6 and C.6.1
+ */
+static const char *spapr_drc_get_type_str(sPAPRDRConnectorType type)
+{
+    switch (type) {
+    case SPAPR_DR_CONNECTOR_TYPE_CPU:
+        return "CPU";
+    case SPAPR_DR_CONNECTOR_TYPE_PHB:
+        return "PHB";
+    case SPAPR_DR_CONNECTOR_TYPE_VIO:
+        return "SLOT";
+    case SPAPR_DR_CONNECTOR_TYPE_PCI:
+        return "28";
+    case SPAPR_DR_CONNECTOR_TYPE_LMB:
+        return "MEM";
+    default:
+        g_assert(false);
+    }
+
+    return NULL;
+}
+
+/**
+ * spapr_drc_populate_dt
+ *
+ * @fdt: libfdt device tree
+ * @path: path in the DT to generate properties
+ * @owner: parent Object/DeviceState for which to generate DRC
+ *         descriptions for
+ * @drc_type_mask: mask of sPAPRDRConnectorType values corresponding
+ *   to the types of DRCs to generate entries for
+ *
+ * generate OF properties to describe DRC topology/indices to guests
+ *
+ * as documented in PAPR+ v2.1, 13.5.2
+ */
+int spapr_drc_populate_dt(void *fdt, int fdt_offset, Object *owner,
+                          uint32_t drc_type_mask)
+{
+    Object *root_container;
+    ObjectProperty *prop;
+    uint32_t drc_count = 0;
+    GArray *drc_indexes, *drc_power_domains;
+    GString *drc_names, *drc_types;
+    int ret;
+
+    /* the first entry of each properties is a 32-bit integer encoding
+     * the number of elements in the array. we won't know this until
+     * we complete the iteration through all the matching DRCs, but
+     * reserve the space now and set the offsets accordingly so we
+     * can fill them in later.
+     */
+    drc_indexes = g_array_new(false, true, sizeof(uint32_t));
+    drc_indexes = g_array_set_size(drc_indexes, 1);
+    drc_power_domains = g_array_new(false, true, sizeof(uint32_t));
+    drc_power_domains = g_array_set_size(drc_power_domains, 1);
+    drc_names = g_string_set_size(g_string_new(NULL), sizeof(uint32_t));
+    drc_types = g_string_set_size(g_string_new(NULL), sizeof(uint32_t));
+
+    /* aliases for all DRConnector objects will be rooted in QOM
+     * composition tree at DRC_CONTAINER_PATH
+     */
+    root_container = container_get(object_get_root(), DRC_CONTAINER_PATH);
+
+    QTAILQ_FOREACH(prop, &root_container->properties, node) {
+        Object *obj;
+        sPAPRDRConnector *drc;
+        sPAPRDRConnectorClass *drck;
+        uint32_t drc_index, drc_power_domain;
+
+        if (!strstart(prop->type, "link<", NULL)) {
+            continue;
+        }
+
+        obj = object_property_get_link(root_container, prop->name, NULL);
+        drc = SPAPR_DR_CONNECTOR(obj);
+        drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+
+        if (owner && (drc->owner != owner)) {
+            continue;
+        }
+
+        if ((drc->type & drc_type_mask) == 0) {
+            continue;
+        }
+
+        drc_count++;
+
+        /* ibm,drc-indexes */
+        drc_index = cpu_to_be32(drck->get_index(drc));
+        g_array_append_val(drc_indexes, drc_index);
+
+        /* ibm,drc-power-domains */
+        drc_power_domain = cpu_to_be32(-1);
+        g_array_append_val(drc_power_domains, drc_power_domain);
+
+        /* ibm,drc-names */
+        drc_names = g_string_append(drc_names, drck->get_name(drc));
+        drc_names = g_string_insert_len(drc_names, -1, "\0", 1);
+
+        /* ibm,drc-types */
+        drc_types = g_string_append(drc_types,
+                                    spapr_drc_get_type_str(drc->type));
+        drc_types = g_string_insert_len(drc_types, -1, "\0", 1);
+    }
+
+    /* now write the drc count into the space we reserved at the
+     * beginning of the arrays previously
+     */
+    *(uint32_t *)drc_indexes->data = cpu_to_be32(drc_count);
+    *(uint32_t *)drc_power_domains->data = cpu_to_be32(drc_count);
+    *(uint32_t *)drc_names->str = cpu_to_be32(drc_count);
+    *(uint32_t *)drc_types->str = cpu_to_be32(drc_count);
+
+    ret = fdt_setprop(fdt, fdt_offset, "ibm,drc-indexes",
+                      drc_indexes->data,
+                      drc_indexes->len * sizeof(uint32_t));
+    if (ret) {
+        fprintf(stderr, "Couldn't create ibm,drc-indexes property\n");
+        goto out;
+    }
+
+    ret = fdt_setprop(fdt, fdt_offset, "ibm,drc-power-domains",
+                      drc_power_domains->data,
+                      drc_power_domains->len * sizeof(uint32_t));
+    if (ret) {
+        fprintf(stderr, "Couldn't finalize ibm,drc-power-domains property\n");
+        goto out;
+    }
+
+    ret = fdt_setprop(fdt, fdt_offset, "ibm,drc-names",
+                      drc_names->str, drc_names->len);
+    if (ret) {
+        fprintf(stderr, "Couldn't finalize ibm,drc-names property\n");
+        goto out;
+    }
+
+    ret = fdt_setprop(fdt, fdt_offset, "ibm,drc-types",
+                      drc_types->str, drc_types->len);
+    if (ret) {
+        fprintf(stderr, "Couldn't finalize ibm,drc-types property\n");
+        goto out;
+    }
+
+out:
+    g_array_free(drc_indexes, true);
+    g_array_free(drc_power_domains, true);
+    g_string_free(drc_names, true);
+    g_string_free(drc_types, true);
+
+    return ret;
+}
diff --git a/hw/ppc/spapr_events.c b/hw/ppc/spapr_events.c
index 283e96bca1..fda9e3590a 100644
--- a/hw/ppc/spapr_events.c
+++ b/hw/ppc/spapr_events.c
@@ -32,6 +32,9 @@
 
 #include "hw/ppc/spapr.h"
 #include "hw/ppc/spapr_vio.h"
+#include "hw/pci/pci.h"
+#include "hw/pci-host/spapr.h"
+#include "hw/ppc/spapr_drc.h"
 
 #include <libfdt.h>
 
@@ -77,6 +80,7 @@ struct rtas_error_log {
 #define   RTAS_LOG_TYPE_ECC_UNCORR              0x00000009
 #define   RTAS_LOG_TYPE_ECC_CORR                0x0000000a
 #define   RTAS_LOG_TYPE_EPOW                    0x00000040
+#define   RTAS_LOG_TYPE_HOTPLUG                 0x000000e5
     uint32_t extended_length;
 } QEMU_PACKED;
 
@@ -166,6 +170,38 @@ struct epow_log_full {
     struct rtas_event_log_v6_epow epow;
 } QEMU_PACKED;
 
+struct rtas_event_log_v6_hp {
+#define RTAS_LOG_V6_SECTION_ID_HOTPLUG              0x4850 /* HP */
+    struct rtas_event_log_v6_section_header hdr;
+    uint8_t hotplug_type;
+#define RTAS_LOG_V6_HP_TYPE_CPU                          1
+#define RTAS_LOG_V6_HP_TYPE_MEMORY                       2
+#define RTAS_LOG_V6_HP_TYPE_SLOT                         3
+#define RTAS_LOG_V6_HP_TYPE_PHB                          4
+#define RTAS_LOG_V6_HP_TYPE_PCI                          5
+    uint8_t hotplug_action;
+#define RTAS_LOG_V6_HP_ACTION_ADD                        1
+#define RTAS_LOG_V6_HP_ACTION_REMOVE                     2
+    uint8_t hotplug_identifier;
+#define RTAS_LOG_V6_HP_ID_DRC_NAME                       1
+#define RTAS_LOG_V6_HP_ID_DRC_INDEX                      2
+#define RTAS_LOG_V6_HP_ID_DRC_COUNT                      3
+    uint8_t reserved;
+    union {
+        uint32_t index;
+        uint32_t count;
+        char name[1];
+    } drc;
+} QEMU_PACKED;
+
+struct hp_log_full {
+    struct rtas_error_log hdr;
+    struct rtas_event_log_v6 v6hdr;
+    struct rtas_event_log_v6_maina maina;
+    struct rtas_event_log_v6_mainb mainb;
+    struct rtas_event_log_v6_hp hp;
+} QEMU_PACKED;
+
 #define EVENT_MASK_INTERNAL_ERRORS           0x80000000
 #define EVENT_MASK_EPOW                      0x40000000
 #define EVENT_MASK_HOTPLUG                   0x10000000
@@ -181,67 +217,105 @@ struct epow_log_full {
         }                                                          \
     } while (0)
 
-void spapr_events_fdt_skel(void *fdt, uint32_t epow_irq)
+void spapr_events_fdt_skel(void *fdt, uint32_t check_exception_irq)
 {
-    uint32_t epow_irq_ranges[] = {cpu_to_be32(epow_irq), cpu_to_be32(1)};
-    uint32_t epow_interrupts[] = {cpu_to_be32(epow_irq), 0};
+    uint32_t irq_ranges[] = {cpu_to_be32(check_exception_irq), cpu_to_be32(1)};
+    uint32_t interrupts[] = {cpu_to_be32(check_exception_irq), 0};
 
     _FDT((fdt_begin_node(fdt, "event-sources")));
 
     _FDT((fdt_property(fdt, "interrupt-controller", NULL, 0)));
     _FDT((fdt_property_cell(fdt, "#interrupt-cells", 2)));
     _FDT((fdt_property(fdt, "interrupt-ranges",
-                       epow_irq_ranges, sizeof(epow_irq_ranges))));
+                       irq_ranges, sizeof(irq_ranges))));
 
     _FDT((fdt_begin_node(fdt, "epow-events")));
-    _FDT((fdt_property(fdt, "interrupts",
-                       epow_interrupts, sizeof(epow_interrupts))));
+    _FDT((fdt_property(fdt, "interrupts", interrupts, sizeof(interrupts))));
     _FDT((fdt_end_node(fdt)));
 
     _FDT((fdt_end_node(fdt)));
 }
 
-static struct epow_log_full *pending_epow;
-static uint32_t next_plid;
+static void rtas_event_log_queue(int log_type, void *data, bool exception)
+{
+    sPAPREventLogEntry *entry = g_new(sPAPREventLogEntry, 1);
 
-static void spapr_powerdown_req(Notifier *n, void *opaque)
+    g_assert(data);
+    entry->log_type = log_type;
+    entry->exception = exception;
+    entry->data = data;
+    QTAILQ_INSERT_TAIL(&spapr->pending_events, entry, next);
+}
+
+static sPAPREventLogEntry *rtas_event_log_dequeue(uint32_t event_mask,
+                                                  bool exception)
 {
-    sPAPREnvironment *spapr = container_of(n, sPAPREnvironment, epow_notifier);
-    struct rtas_error_log *hdr;
-    struct rtas_event_log_v6 *v6hdr;
-    struct rtas_event_log_v6_maina *maina;
-    struct rtas_event_log_v6_mainb *mainb;
-    struct rtas_event_log_v6_epow *epow;
-    struct tm tm;
-    int year;
+    sPAPREventLogEntry *entry = NULL;
 
-    if (pending_epow) {
-        /* For now, we just throw away earlier events if two come
-         * along before any are consumed.  This is sufficient for our
-         * powerdown messages, but we'll need more if we do more
-         * general error/event logging */
-        g_free(pending_epow);
+    /* we only queue EPOW events atm. */
+    if ((event_mask & EVENT_MASK_EPOW) == 0) {
+        return NULL;
     }
-    pending_epow = g_malloc0(sizeof(*pending_epow));
-    hdr = &pending_epow->hdr;
-    v6hdr = &pending_epow->v6hdr;
-    maina = &pending_epow->maina;
-    mainb = &pending_epow->mainb;
-    epow = &pending_epow->epow;
 
-    hdr->summary = cpu_to_be32(RTAS_LOG_VERSION_6
-                               | RTAS_LOG_SEVERITY_EVENT
-                               | RTAS_LOG_DISPOSITION_NOT_RECOVERED
-                               | RTAS_LOG_OPTIONAL_PART_PRESENT
-                               | RTAS_LOG_TYPE_EPOW);
-    hdr->extended_length = cpu_to_be32(sizeof(*pending_epow)
-                                       - sizeof(pending_epow->hdr));
+    QTAILQ_FOREACH(entry, &spapr->pending_events, next) {
+        if (entry->exception != exception) {
+            continue;
+        }
+
+        /* EPOW and hotplug events are surfaced in the same manner */
+        if (entry->log_type == RTAS_LOG_TYPE_EPOW ||
+            entry->log_type == RTAS_LOG_TYPE_HOTPLUG) {
+            break;
+        }
+    }
+
+    if (entry) {
+        QTAILQ_REMOVE(&spapr->pending_events, entry, next);
+    }
 
+    return entry;
+}
+
+static bool rtas_event_log_contains(uint32_t event_mask, bool exception)
+{
+    sPAPREventLogEntry *entry = NULL;
+
+    /* we only queue EPOW events atm. */
+    if ((event_mask & EVENT_MASK_EPOW) == 0) {
+        return false;
+    }
+
+    QTAILQ_FOREACH(entry, &spapr->pending_events, next) {
+        if (entry->exception != exception) {
+            continue;
+        }
+
+        /* EPOW and hotplug events are surfaced in the same manner */
+        if (entry->log_type == RTAS_LOG_TYPE_EPOW ||
+            entry->log_type == RTAS_LOG_TYPE_HOTPLUG) {
+            return true;
+        }
+    }
+
+    return false;
+}
+
+static uint32_t next_plid;
+
+static void spapr_init_v6hdr(struct rtas_event_log_v6 *v6hdr)
+{
     v6hdr->b0 = RTAS_LOG_V6_B0_VALID | RTAS_LOG_V6_B0_NEW_LOG
         | RTAS_LOG_V6_B0_BIGENDIAN;
     v6hdr->b2 = RTAS_LOG_V6_B2_POWERPC_FORMAT
         | RTAS_LOG_V6_B2_LOG_FORMAT_PLATFORM_EVENT;
     v6hdr->company = cpu_to_be32(RTAS_LOG_V6_COMPANY_IBM);
+}
+
+static void spapr_init_maina(struct rtas_event_log_v6_maina *maina,
+                             int section_count)
+{
+    struct tm tm;
+    int year;
 
     maina->hdr.section_id = cpu_to_be16(RTAS_LOG_V6_SECTION_ID_MAINA);
     maina->hdr.section_length = cpu_to_be16(sizeof(*maina));
@@ -256,8 +330,37 @@ static void spapr_powerdown_req(Notifier *n, void *opaque)
                                        | (to_bcd(tm.tm_min) << 16)
                                        | (to_bcd(tm.tm_sec) << 8));
     maina->creator_id = 'H'; /* Hypervisor */
-    maina->section_count = 3; /* Main-A, Main-B and EPOW */
+    maina->section_count = section_count;
     maina->plid = next_plid++;
+}
+
+static void spapr_powerdown_req(Notifier *n, void *opaque)
+{
+    sPAPREnvironment *spapr = container_of(n, sPAPREnvironment, epow_notifier);
+    struct rtas_error_log *hdr;
+    struct rtas_event_log_v6 *v6hdr;
+    struct rtas_event_log_v6_maina *maina;
+    struct rtas_event_log_v6_mainb *mainb;
+    struct rtas_event_log_v6_epow *epow;
+    struct epow_log_full *new_epow;
+
+    new_epow = g_malloc0(sizeof(*new_epow));
+    hdr = &new_epow->hdr;
+    v6hdr = &new_epow->v6hdr;
+    maina = &new_epow->maina;
+    mainb = &new_epow->mainb;
+    epow = &new_epow->epow;
+
+    hdr->summary = cpu_to_be32(RTAS_LOG_VERSION_6
+                               | RTAS_LOG_SEVERITY_EVENT
+                               | RTAS_LOG_DISPOSITION_NOT_RECOVERED
+                               | RTAS_LOG_OPTIONAL_PART_PRESENT
+                               | RTAS_LOG_TYPE_EPOW);
+    hdr->extended_length = cpu_to_be32(sizeof(*new_epow)
+                                       - sizeof(new_epow->hdr));
+
+    spapr_init_v6hdr(v6hdr);
+    spapr_init_maina(maina, 3 /* Main-A, Main-B and EPOW */);
 
     mainb->hdr.section_id = cpu_to_be16(RTAS_LOG_V6_SECTION_ID_MAINB);
     mainb->hdr.section_length = cpu_to_be16(sizeof(*mainb));
@@ -274,7 +377,80 @@ static void spapr_powerdown_req(Notifier *n, void *opaque)
     epow->event_modifier = RTAS_LOG_V6_EPOW_MODIFIER_NORMAL;
     epow->extended_modifier = RTAS_LOG_V6_EPOW_XMODIFIER_PARTITION_SPECIFIC;
 
-    qemu_irq_pulse(xics_get_qirq(spapr->icp, spapr->epow_irq));
+    rtas_event_log_queue(RTAS_LOG_TYPE_EPOW, new_epow, true);
+
+    qemu_irq_pulse(xics_get_qirq(spapr->icp, spapr->check_exception_irq));
+}
+
+static void spapr_hotplug_req_event(sPAPRDRConnector *drc, uint8_t hp_action)
+{
+    struct hp_log_full *new_hp;
+    struct rtas_error_log *hdr;
+    struct rtas_event_log_v6 *v6hdr;
+    struct rtas_event_log_v6_maina *maina;
+    struct rtas_event_log_v6_mainb *mainb;
+    struct rtas_event_log_v6_hp *hp;
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    sPAPRDRConnectorType drc_type = drck->get_type(drc);
+
+    new_hp = g_malloc0(sizeof(struct hp_log_full));
+    hdr = &new_hp->hdr;
+    v6hdr = &new_hp->v6hdr;
+    maina = &new_hp->maina;
+    mainb = &new_hp->mainb;
+    hp = &new_hp->hp;
+
+    hdr->summary = cpu_to_be32(RTAS_LOG_VERSION_6
+                               | RTAS_LOG_SEVERITY_EVENT
+                               | RTAS_LOG_DISPOSITION_NOT_RECOVERED
+                               | RTAS_LOG_OPTIONAL_PART_PRESENT
+                               | RTAS_LOG_INITIATOR_HOTPLUG
+                               | RTAS_LOG_TYPE_HOTPLUG);
+    hdr->extended_length = cpu_to_be32(sizeof(*new_hp)
+                                       - sizeof(new_hp->hdr));
+
+    spapr_init_v6hdr(v6hdr);
+    spapr_init_maina(maina, 3 /* Main-A, Main-B, HP */);
+
+    mainb->hdr.section_id = cpu_to_be16(RTAS_LOG_V6_SECTION_ID_MAINB);
+    mainb->hdr.section_length = cpu_to_be16(sizeof(*mainb));
+    mainb->subsystem_id = 0x80; /* External environment */
+    mainb->event_severity = 0x00; /* Informational / non-error */
+    mainb->event_subtype = 0x00; /* Normal shutdown */
+
+    hp->hdr.section_id = cpu_to_be16(RTAS_LOG_V6_SECTION_ID_HOTPLUG);
+    hp->hdr.section_length = cpu_to_be16(sizeof(*hp));
+    hp->hdr.section_version = 1; /* includes extended modifier */
+    hp->hotplug_action = hp_action;
+
+
+    switch (drc_type) {
+    case SPAPR_DR_CONNECTOR_TYPE_PCI:
+        hp->drc.index = cpu_to_be32(drck->get_index(drc));
+        hp->hotplug_identifier = RTAS_LOG_V6_HP_ID_DRC_INDEX;
+        hp->hotplug_type = RTAS_LOG_V6_HP_TYPE_PCI;
+        break;
+    default:
+        /* we shouldn't be signaling hotplug events for resources
+         * that don't support them
+         */
+        g_assert(false);
+        return;
+    }
+
+    rtas_event_log_queue(RTAS_LOG_TYPE_HOTPLUG, new_hp, true);
+
+    qemu_irq_pulse(xics_get_qirq(spapr->icp, spapr->check_exception_irq));
+}
+
+void spapr_hotplug_req_add_event(sPAPRDRConnector *drc)
+{
+    spapr_hotplug_req_event(drc, RTAS_LOG_V6_HP_ACTION_ADD);
+}
+
+void spapr_hotplug_req_remove_event(sPAPRDRConnector *drc)
+{
+    spapr_hotplug_req_event(drc, RTAS_LOG_V6_HP_ACTION_REMOVE);
 }
 
 static void check_exception(PowerPCCPU *cpu, sPAPREnvironment *spapr,
@@ -282,8 +458,10 @@ static void check_exception(PowerPCCPU *cpu, sPAPREnvironment *spapr,
                             target_ulong args,
                             uint32_t nret, target_ulong rets)
 {
-    uint32_t mask, buf, len;
+    uint32_t mask, buf, len, event_len;
     uint64_t xinfo;
+    sPAPREventLogEntry *event;
+    struct rtas_error_log *hdr;
 
     if ((nargs < 6) || (nargs > 7) || nret != 1) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
@@ -298,25 +476,85 @@ static void check_exception(PowerPCCPU *cpu, sPAPREnvironment *spapr,
         xinfo |= (uint64_t)rtas_ld(args, 6) << 32;
     }
 
-    if ((mask & EVENT_MASK_EPOW) && pending_epow) {
-        if (sizeof(*pending_epow) < len) {
-            len = sizeof(*pending_epow);
-        }
+    event = rtas_event_log_dequeue(mask, true);
+    if (!event) {
+        goto out_no_events;
+    }
+
+    hdr = event->data;
+    event_len = be32_to_cpu(hdr->extended_length) + sizeof(*hdr);
+
+    if (event_len < len) {
+        len = event_len;
+    }
+
+    cpu_physical_memory_write(buf, event->data, len);
+    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
+    g_free(event->data);
+    g_free(event);
+
+    /* according to PAPR+, the IRQ must be left asserted, or re-asserted, if
+     * there are still pending events to be fetched via check-exception. We
+     * do the latter here, since our code relies on edge-triggered
+     * interrupts.
+     */
+    if (rtas_event_log_contains(mask, true)) {
+        qemu_irq_pulse(xics_get_qirq(spapr->icp, spapr->check_exception_irq));
+    }
 
-        cpu_physical_memory_write(buf, pending_epow, len);
-        g_free(pending_epow);
-        pending_epow = NULL;
-        rtas_st(rets, 0, RTAS_OUT_SUCCESS);
-    } else {
-        rtas_st(rets, 0, RTAS_OUT_NO_ERRORS_FOUND);
+    return;
+
+out_no_events:
+    rtas_st(rets, 0, RTAS_OUT_NO_ERRORS_FOUND);
+}
+
+static void event_scan(PowerPCCPU *cpu, sPAPREnvironment *spapr,
+                       uint32_t token, uint32_t nargs,
+                       target_ulong args,
+                       uint32_t nret, target_ulong rets)
+{
+    uint32_t mask, buf, len, event_len;
+    sPAPREventLogEntry *event;
+    struct rtas_error_log *hdr;
+
+    if (nargs != 4 || nret != 1) {
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
     }
+
+    mask = rtas_ld(args, 0);
+    buf = rtas_ld(args, 2);
+    len = rtas_ld(args, 3);
+
+    event = rtas_event_log_dequeue(mask, false);
+    if (!event) {
+        goto out_no_events;
+    }
+
+    hdr = event->data;
+    event_len = be32_to_cpu(hdr->extended_length) + sizeof(*hdr);
+
+    if (event_len < len) {
+        len = event_len;
+    }
+
+    cpu_physical_memory_write(buf, event->data, len);
+    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
+    g_free(event->data);
+    g_free(event);
+    return;
+
+out_no_events:
+    rtas_st(rets, 0, RTAS_OUT_NO_ERRORS_FOUND);
 }
 
 void spapr_events_init(sPAPREnvironment *spapr)
 {
-    spapr->epow_irq = xics_alloc(spapr->icp, 0, 0, false);
+    QTAILQ_INIT(&spapr->pending_events);
+    spapr->check_exception_irq = xics_alloc(spapr->icp, 0, 0, false);
     spapr->epow_notifier.notify = spapr_powerdown_req;
     qemu_register_powerdown_notifier(&spapr->epow_notifier);
     spapr_rtas_register(RTAS_CHECK_EXCEPTION, "check-exception",
                         check_exception);
+    spapr_rtas_register(RTAS_EVENT_SCAN, "event-scan", event_scan);
 }
diff --git a/hw/ppc/spapr_iommu.c b/hw/ppc/spapr_iommu.c
index f3990fdc32..8cd9dba9ac 100644
--- a/hw/ppc/spapr_iommu.c
+++ b/hw/ppc/spapr_iommu.c
@@ -41,7 +41,7 @@ enum sPAPRTCEAccess {
 
 static QLIST_HEAD(spapr_tce_tables, sPAPRTCETable) spapr_tce_tables;
 
-static sPAPRTCETable *spapr_tce_find_by_liobn(uint32_t liobn)
+sPAPRTCETable *spapr_tce_find_by_liobn(target_ulong liobn)
 {
     sPAPRTCETable *tcet;
 
@@ -52,7 +52,7 @@ static sPAPRTCETable *spapr_tce_find_by_liobn(uint32_t liobn)
     }
 
     QLIST_FOREACH(tcet, &spapr_tce_tables, list) {
-        if (tcet->liobn == liobn) {
+        if (tcet->liobn == (uint32_t)liobn) {
             return tcet;
         }
     }
@@ -126,11 +126,11 @@ static MemoryRegionIOMMUOps spapr_iommu_ops = {
 static int spapr_tce_table_realize(DeviceState *dev)
 {
     sPAPRTCETable *tcet = SPAPR_TCE_TABLE(dev);
+    uint64_t window_size = (uint64_t)tcet->nb_table << tcet->page_shift;
 
-    if (kvm_enabled()) {
+    if (kvm_enabled() && !(window_size >> 32)) {
         tcet->table = kvmppc_create_spapr_tce(tcet->liobn,
-                                              tcet->nb_table <<
-                                              tcet->page_shift,
+                                              window_size,
                                               &tcet->fd,
                                               tcet->vfio_accel);
     }
@@ -161,6 +161,7 @@ sPAPRTCETable *spapr_tce_new_table(DeviceState *owner, uint32_t liobn,
                                    bool vfio_accel)
 {
     sPAPRTCETable *tcet;
+    char tmp[64];
 
     if (spapr_tce_find_by_liobn(liobn)) {
         fprintf(stderr, "Attempted to create TCE table with duplicate"
@@ -179,7 +180,8 @@ sPAPRTCETable *spapr_tce_new_table(DeviceState *owner, uint32_t liobn,
     tcet->nb_table = nb_table;
     tcet->vfio_accel = vfio_accel;
 
-    object_property_add_child(OBJECT(owner), "tce-table", OBJECT(tcet), NULL);
+    snprintf(tmp, sizeof(tmp), "tce-table-%x", liobn);
+    object_property_add_child(OBJECT(owner), tmp, OBJECT(tcet), NULL);
 
     object_property_set_bool(OBJECT(tcet), true, "realized", NULL);
 
@@ -247,7 +249,7 @@ static target_ulong h_put_tce_indirect(PowerPCCPU *cpu,
     target_ulong ioba1 = ioba;
     target_ulong tce_list = args[2];
     target_ulong npages = args[3];
-    target_ulong ret = H_PARAMETER;
+    target_ulong ret = H_PARAMETER, tce = 0;
     sPAPRTCETable *tcet = spapr_tce_find_by_liobn(liobn);
     CPUState *cs = CPU(cpu);
     hwaddr page_mask, page_size;
@@ -267,7 +269,7 @@ static target_ulong h_put_tce_indirect(PowerPCCPU *cpu,
     for (i = 0; i < npages; ++i, ioba += page_size) {
         target_ulong off = (tce_list & ~SPAPR_TCE_RW) +
                                 i * sizeof(target_ulong);
-        target_ulong tce = ldq_phys(cs->as, off);
+        tce = ldq_be_phys(cs->as, off);
 
         ret = put_tce_emu(tcet, ioba, tce);
         if (ret) {
@@ -277,11 +279,11 @@ static target_ulong h_put_tce_indirect(PowerPCCPU *cpu,
 
     /* Trace last successful or the first problematic entry */
     i = i ? (i - 1) : 0;
-    trace_spapr_iommu_indirect(liobn, ioba1, tce_list, i,
-                               ldq_phys(cs->as,
-                               tce_list + i * sizeof(target_ulong)),
-                               ret);
-
+    if (SPAPR_IS_PCI_LIOBN(liobn)) {
+        trace_spapr_iommu_pci_indirect(liobn, ioba1, tce_list, i, tce, ret);
+    } else {
+        trace_spapr_iommu_indirect(liobn, ioba1, tce_list, i, tce, ret);
+    }
     return ret;
 }
 
@@ -315,7 +317,11 @@ static target_ulong h_stuff_tce(PowerPCCPU *cpu, sPAPREnvironment *spapr,
             break;
         }
     }
-    trace_spapr_iommu_stuff(liobn, ioba, tce_value, npages, ret);
+    if (SPAPR_IS_PCI_LIOBN(liobn)) {
+        trace_spapr_iommu_pci_stuff(liobn, ioba, tce_value, npages, ret);
+    } else {
+        trace_spapr_iommu_stuff(liobn, ioba, tce_value, npages, ret);
+    }
 
     return ret;
 }
@@ -336,7 +342,11 @@ static target_ulong h_put_tce(PowerPCCPU *cpu, sPAPREnvironment *spapr,
 
         ret = put_tce_emu(tcet, ioba, tce);
     }
-    trace_spapr_iommu_put(liobn, ioba, tce, ret);
+    if (SPAPR_IS_PCI_LIOBN(liobn)) {
+        trace_spapr_iommu_pci_put(liobn, ioba, tce, ret);
+    } else {
+        trace_spapr_iommu_put(liobn, ioba, tce, ret);
+    }
 
     return ret;
 }
@@ -376,7 +386,11 @@ static target_ulong h_get_tce(PowerPCCPU *cpu, sPAPREnvironment *spapr,
             args[0] = tce;
         }
     }
-    trace_spapr_iommu_get(liobn, ioba, ret, tce);
+    if (SPAPR_IS_PCI_LIOBN(liobn)) {
+        trace_spapr_iommu_pci_get(liobn, ioba, ret, tce);
+    } else {
+        trace_spapr_iommu_get(liobn, ioba, ret, tce);
+    }
 
     return ret;
 }
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 05f4faca6e..4df3a33db4 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -33,8 +33,11 @@
 #include <libfdt.h>
 #include "trace.h"
 #include "qemu/error-report.h"
+#include "qapi/qmp/qerror.h"
 
 #include "hw/pci/pci_bus.h"
+#include "hw/ppc/spapr_drc.h"
+#include "sysemu/device_tree.h"
 
 /* Copied from the kernel arch/powerpc/platforms/pseries/msi.c */
 #define RTAS_QUERY_FN           0
@@ -47,7 +50,15 @@
 #define RTAS_TYPE_MSI           1
 #define RTAS_TYPE_MSIX          2
 
-static sPAPRPHBState *find_phb(sPAPREnvironment *spapr, uint64_t buid)
+#define _FDT(exp) \
+    do { \
+        int ret = (exp);                                           \
+        if (ret < 0) {                                             \
+            return ret;                                            \
+        }                                                          \
+    } while (0)
+
+sPAPRPHBState *spapr_pci_find_phb(sPAPREnvironment *spapr, uint64_t buid)
 {
     sPAPRPHBState *sphb;
 
@@ -61,10 +72,10 @@ static sPAPRPHBState *find_phb(sPAPREnvironment *spapr, uint64_t buid)
     return NULL;
 }
 
-static PCIDevice *find_dev(sPAPREnvironment *spapr, uint64_t buid,
-                           uint32_t config_addr)
+PCIDevice *spapr_pci_find_dev(sPAPREnvironment *spapr, uint64_t buid,
+                              uint32_t config_addr)
 {
-    sPAPRPHBState *sphb = find_phb(spapr, buid);
+    sPAPRPHBState *sphb = spapr_pci_find_phb(spapr, buid);
     PCIHostState *phb = PCI_HOST_BRIDGE(sphb);
     int bus_num = (config_addr >> 16) & 0xFF;
     int devfn = (config_addr >> 8) & 0xFF;
@@ -95,7 +106,7 @@ static void finish_read_pci_config(sPAPREnvironment *spapr, uint64_t buid,
         return;
     }
 
-    pci_dev = find_dev(spapr, buid, addr);
+    pci_dev = spapr_pci_find_dev(spapr, buid, addr);
     addr = rtas_pci_cfgaddr(addr);
 
     if (!pci_dev || (addr % size) || (addr >= pci_config_size(pci_dev))) {
@@ -162,7 +173,7 @@ static void finish_write_pci_config(sPAPREnvironment *spapr, uint64_t buid,
         return;
     }
 
-    pci_dev = find_dev(spapr, buid, addr);
+    pci_dev = spapr_pci_find_dev(spapr, buid, addr);
     addr = rtas_pci_cfgaddr(addr);
 
     if (!pci_dev || (addr % size) || (addr >= pci_config_size(pci_dev))) {
@@ -280,9 +291,9 @@ static void rtas_ibm_change_msi(PowerPCCPU *cpu, sPAPREnvironment *spapr,
     }
 
     /* Fins sPAPRPHBState */
-    phb = find_phb(spapr, buid);
+    phb = spapr_pci_find_phb(spapr, buid);
     if (phb) {
-        pdev = find_dev(spapr, buid, config_addr);
+        pdev = spapr_pci_find_dev(spapr, buid, config_addr);
     }
     if (!phb || !pdev) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
@@ -381,9 +392,9 @@ static void rtas_ibm_query_interrupt_source_number(PowerPCCPU *cpu,
     spapr_pci_msi *msi;
 
     /* Find sPAPRPHBState */
-    phb = find_phb(spapr, buid);
+    phb = spapr_pci_find_phb(spapr, buid);
     if (phb) {
-        pdev = find_dev(spapr, buid, config_addr);
+        pdev = spapr_pci_find_dev(spapr, buid, config_addr);
     }
     if (!phb || !pdev) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
@@ -426,7 +437,7 @@ static void rtas_ibm_set_eeh_option(PowerPCCPU *cpu,
     addr = rtas_ld(args, 0);
     option = rtas_ld(args, 3);
 
-    sphb = find_phb(spapr, buid);
+    sphb = spapr_pci_find_phb(spapr, buid);
     if (!sphb) {
         goto param_error_exit;
     }
@@ -461,7 +472,7 @@ static void rtas_ibm_get_config_addr_info2(PowerPCCPU *cpu,
     }
 
     buid = ((uint64_t)rtas_ld(args, 1) << 32) | rtas_ld(args, 2);
-    sphb = find_phb(spapr, buid);
+    sphb = spapr_pci_find_phb(spapr, buid);
     if (!sphb) {
         goto param_error_exit;
     }
@@ -479,7 +490,7 @@ static void rtas_ibm_get_config_addr_info2(PowerPCCPU *cpu,
     switch (option) {
     case RTAS_GET_PE_ADDR:
         addr = rtas_ld(args, 0);
-        pdev = find_dev(spapr, buid, addr);
+        pdev = spapr_pci_find_dev(spapr, buid, addr);
         if (!pdev) {
             goto param_error_exit;
         }
@@ -516,7 +527,7 @@ static void rtas_ibm_read_slot_reset_state2(PowerPCCPU *cpu,
     }
 
     buid = ((uint64_t)rtas_ld(args, 1) << 32) | rtas_ld(args, 2);
-    sphb = find_phb(spapr, buid);
+    sphb = spapr_pci_find_phb(spapr, buid);
     if (!sphb) {
         goto param_error_exit;
     }
@@ -562,7 +573,7 @@ static void rtas_ibm_set_slot_reset(PowerPCCPU *cpu,
 
     buid = ((uint64_t)rtas_ld(args, 1) << 32) | rtas_ld(args, 2);
     option = rtas_ld(args, 3);
-    sphb = find_phb(spapr, buid);
+    sphb = spapr_pci_find_phb(spapr, buid);
     if (!sphb) {
         goto param_error_exit;
     }
@@ -596,7 +607,7 @@ static void rtas_ibm_configure_pe(PowerPCCPU *cpu,
     }
 
     buid = ((uint64_t)rtas_ld(args, 1) << 32) | rtas_ld(args, 2);
-    sphb = find_phb(spapr, buid);
+    sphb = spapr_pci_find_phb(spapr, buid);
     if (!sphb) {
         goto param_error_exit;
     }
@@ -631,7 +642,7 @@ static void rtas_ibm_slot_error_detail(PowerPCCPU *cpu,
     }
 
     buid = ((uint64_t)rtas_ld(args, 1) << 32) | rtas_ld(args, 2);
-    sphb = find_phb(spapr, buid);
+    sphb = spapr_pci_find_phb(spapr, buid);
     if (!sphb) {
         goto param_error_exit;
     }
@@ -731,6 +742,372 @@ static AddressSpace *spapr_pci_dma_iommu(PCIBus *bus, void *opaque, int devfn)
     return &phb->iommu_as;
 }
 
+/* Macros to operate with address in OF binding to PCI */
+#define b_x(x, p, l)    (((x) & ((1<<(l))-1)) << (p))
+#define b_n(x)          b_x((x), 31, 1) /* 0 if relocatable */
+#define b_p(x)          b_x((x), 30, 1) /* 1 if prefetchable */
+#define b_t(x)          b_x((x), 29, 1) /* 1 if the address is aliased */
+#define b_ss(x)         b_x((x), 24, 2) /* the space code */
+#define b_bbbbbbbb(x)   b_x((x), 16, 8) /* bus number */
+#define b_ddddd(x)      b_x((x), 11, 5) /* device number */
+#define b_fff(x)        b_x((x), 8, 3)  /* function number */
+#define b_rrrrrrrr(x)   b_x((x), 0, 8)  /* register number */
+
+/* for 'reg'/'assigned-addresses' OF properties */
+#define RESOURCE_CELLS_SIZE 2
+#define RESOURCE_CELLS_ADDRESS 3
+
+typedef struct ResourceFields {
+    uint32_t phys_hi;
+    uint32_t phys_mid;
+    uint32_t phys_lo;
+    uint32_t size_hi;
+    uint32_t size_lo;
+} QEMU_PACKED ResourceFields;
+
+typedef struct ResourceProps {
+    ResourceFields reg[8];
+    ResourceFields assigned[7];
+    uint32_t reg_len;
+    uint32_t assigned_len;
+} ResourceProps;
+
+/* fill in the 'reg'/'assigned-resources' OF properties for
+ * a PCI device. 'reg' describes resource requirements for a
+ * device's IO/MEM regions, 'assigned-addresses' describes the
+ * actual resource assignments.
+ *
+ * the properties are arrays of ('phys-addr', 'size') pairs describing
+ * the addressable regions of the PCI device, where 'phys-addr' is a
+ * RESOURCE_CELLS_ADDRESS-tuple of 32-bit integers corresponding to
+ * (phys.hi, phys.mid, phys.lo), and 'size' is a
+ * RESOURCE_CELLS_SIZE-tuple corresponding to (size.hi, size.lo).
+ *
+ * phys.hi = 0xYYXXXXZZ, where:
+ *   0xYY = npt000ss
+ *          |||   |
+ *          |||   +-- space code: 1 if IO region, 2 if MEM region
+ *          ||+------ for non-relocatable IO: 1 if aliased
+ *          ||        for relocatable IO: 1 if below 64KB
+ *          ||        for MEM: 1 if below 1MB
+ *          |+------- 1 if region is prefetchable
+ *          +-------- 1 if region is non-relocatable
+ *   0xXXXX = bbbbbbbb dddddfff, encoding bus, slot, and function
+ *            bits respectively
+ *   0xZZ = rrrrrrrr, the register number of the BAR corresponding
+ *          to the region
+ *
+ * phys.mid and phys.lo correspond respectively to the hi/lo portions
+ * of the actual address of the region.
+ *
+ * how the phys-addr/size values are used differ slightly between
+ * 'reg' and 'assigned-addresses' properties. namely, 'reg' has
+ * an additional description for the config space region of the
+ * device, and in the case of QEMU has n=0 and phys.mid=phys.lo=0
+ * to describe the region as relocatable, with an address-mapping
+ * that corresponds directly to the PHB's address space for the
+ * resource. 'assigned-addresses' always has n=1 set with an absolute
+ * address assigned for the resource. in general, 'assigned-addresses'
+ * won't be populated, since addresses for PCI devices are generally
+ * unmapped initially and left to the guest to assign.
+ *
+ * note also that addresses defined in these properties are, at least
+ * for PAPR guests, relative to the PHBs IO/MEM windows, and
+ * correspond directly to the addresses in the BARs.
+ *
+ * in accordance with PCI Bus Binding to Open Firmware,
+ * IEEE Std 1275-1994, section 4.1.1, as implemented by PAPR+ v2.7,
+ * Appendix C.
+ */
+static void populate_resource_props(PCIDevice *d, ResourceProps *rp)
+{
+    int bus_num = pci_bus_num(PCI_BUS(qdev_get_parent_bus(DEVICE(d))));
+    uint32_t dev_id = (b_bbbbbbbb(bus_num) |
+                       b_ddddd(PCI_SLOT(d->devfn)) |
+                       b_fff(PCI_FUNC(d->devfn)));
+    ResourceFields *reg, *assigned;
+    int i, reg_idx = 0, assigned_idx = 0;
+
+    /* config space region */
+    reg = &rp->reg[reg_idx++];
+    reg->phys_hi = cpu_to_be32(dev_id);
+    reg->phys_mid = 0;
+    reg->phys_lo = 0;
+    reg->size_hi = 0;
+    reg->size_lo = 0;
+
+    for (i = 0; i < PCI_NUM_REGIONS; i++) {
+        if (!d->io_regions[i].size) {
+            continue;
+        }
+
+        reg = &rp->reg[reg_idx++];
+
+        reg->phys_hi = cpu_to_be32(dev_id | b_rrrrrrrr(pci_bar(d, i)));
+        if (d->io_regions[i].type & PCI_BASE_ADDRESS_SPACE_IO) {
+            reg->phys_hi |= cpu_to_be32(b_ss(1));
+        } else {
+            reg->phys_hi |= cpu_to_be32(b_ss(2));
+        }
+        reg->phys_mid = 0;
+        reg->phys_lo = 0;
+        reg->size_hi = cpu_to_be32(d->io_regions[i].size >> 32);
+        reg->size_lo = cpu_to_be32(d->io_regions[i].size);
+
+        if (d->io_regions[i].addr == PCI_BAR_UNMAPPED) {
+            continue;
+        }
+
+        assigned = &rp->assigned[assigned_idx++];
+        assigned->phys_hi = cpu_to_be32(reg->phys_hi | b_n(1));
+        assigned->phys_mid = cpu_to_be32(d->io_regions[i].addr >> 32);
+        assigned->phys_lo = cpu_to_be32(d->io_regions[i].addr);
+        assigned->size_hi = reg->size_hi;
+        assigned->size_lo = reg->size_lo;
+    }
+
+    rp->reg_len = reg_idx * sizeof(ResourceFields);
+    rp->assigned_len = assigned_idx * sizeof(ResourceFields);
+}
+
+static int spapr_populate_pci_child_dt(PCIDevice *dev, void *fdt, int offset,
+                                       int phb_index, int drc_index,
+                                       const char *drc_name)
+{
+    ResourceProps rp;
+    bool is_bridge = false;
+    int pci_status;
+
+    if (pci_default_read_config(dev, PCI_HEADER_TYPE, 1) ==
+        PCI_HEADER_TYPE_BRIDGE) {
+        is_bridge = true;
+    }
+
+    /* in accordance with PAPR+ v2.7 13.6.3, Table 181 */
+    _FDT(fdt_setprop_cell(fdt, offset, "vendor-id",
+                          pci_default_read_config(dev, PCI_VENDOR_ID, 2)));
+    _FDT(fdt_setprop_cell(fdt, offset, "device-id",
+                          pci_default_read_config(dev, PCI_DEVICE_ID, 2)));
+    _FDT(fdt_setprop_cell(fdt, offset, "revision-id",
+                          pci_default_read_config(dev, PCI_REVISION_ID, 1)));
+    _FDT(fdt_setprop_cell(fdt, offset, "class-code",
+                          pci_default_read_config(dev, PCI_CLASS_DEVICE, 2)
+                            << 8));
+    if (pci_default_read_config(dev, PCI_INTERRUPT_PIN, 1)) {
+        _FDT(fdt_setprop_cell(fdt, offset, "interrupts",
+                 pci_default_read_config(dev, PCI_INTERRUPT_PIN, 1)));
+    }
+
+    if (!is_bridge) {
+        _FDT(fdt_setprop_cell(fdt, offset, "min-grant",
+            pci_default_read_config(dev, PCI_MIN_GNT, 1)));
+        _FDT(fdt_setprop_cell(fdt, offset, "max-latency",
+            pci_default_read_config(dev, PCI_MAX_LAT, 1)));
+    }
+
+    if (pci_default_read_config(dev, PCI_SUBSYSTEM_ID, 2)) {
+        _FDT(fdt_setprop_cell(fdt, offset, "subsystem-id",
+                 pci_default_read_config(dev, PCI_SUBSYSTEM_ID, 2)));
+    }
+
+    if (pci_default_read_config(dev, PCI_SUBSYSTEM_VENDOR_ID, 2)) {
+        _FDT(fdt_setprop_cell(fdt, offset, "subsystem-vendor-id",
+                 pci_default_read_config(dev, PCI_SUBSYSTEM_VENDOR_ID, 2)));
+    }
+
+    _FDT(fdt_setprop_cell(fdt, offset, "cache-line-size",
+        pci_default_read_config(dev, PCI_CACHE_LINE_SIZE, 1)));
+
+    /* the following fdt cells are masked off the pci status register */
+    pci_status = pci_default_read_config(dev, PCI_STATUS, 2);
+    _FDT(fdt_setprop_cell(fdt, offset, "devsel-speed",
+                          PCI_STATUS_DEVSEL_MASK & pci_status));
+
+    if (pci_status & PCI_STATUS_FAST_BACK) {
+        _FDT(fdt_setprop(fdt, offset, "fast-back-to-back", NULL, 0));
+    }
+    if (pci_status & PCI_STATUS_66MHZ) {
+        _FDT(fdt_setprop(fdt, offset, "66mhz-capable", NULL, 0));
+    }
+    if (pci_status & PCI_STATUS_UDF) {
+        _FDT(fdt_setprop(fdt, offset, "udf-supported", NULL, 0));
+    }
+
+    /* NOTE: this is normally generated by firmware via path/unit name,
+     * but in our case we must set it manually since it does not get
+     * processed by OF beforehand
+     */
+    _FDT(fdt_setprop_string(fdt, offset, "name", "pci"));
+    _FDT(fdt_setprop(fdt, offset, "ibm,loc-code", drc_name, strlen(drc_name)));
+    _FDT(fdt_setprop_cell(fdt, offset, "ibm,my-drc-index", drc_index));
+
+    _FDT(fdt_setprop_cell(fdt, offset, "#address-cells",
+                          RESOURCE_CELLS_ADDRESS));
+    _FDT(fdt_setprop_cell(fdt, offset, "#size-cells",
+                          RESOURCE_CELLS_SIZE));
+    _FDT(fdt_setprop_cell(fdt, offset, "ibm,req#msi-x",
+                          RESOURCE_CELLS_SIZE));
+
+    populate_resource_props(dev, &rp);
+    _FDT(fdt_setprop(fdt, offset, "reg", (uint8_t *)rp.reg, rp.reg_len));
+    _FDT(fdt_setprop(fdt, offset, "assigned-addresses",
+                     (uint8_t *)rp.assigned, rp.assigned_len));
+
+    return 0;
+}
+
+/* create OF node for pci device and required OF DT properties */
+static void *spapr_create_pci_child_dt(sPAPRPHBState *phb, PCIDevice *dev,
+                                       int drc_index, const char *drc_name,
+                                       int *dt_offset)
+{
+    void *fdt;
+    int offset, ret, fdt_size;
+    int slot = PCI_SLOT(dev->devfn);
+    int func = PCI_FUNC(dev->devfn);
+    char nodename[512];
+
+    fdt = create_device_tree(&fdt_size);
+    if (func != 0) {
+        sprintf(nodename, "pci@%d,%d", slot, func);
+    } else {
+        sprintf(nodename, "pci@%d", slot);
+    }
+    offset = fdt_add_subnode(fdt, 0, nodename);
+    ret = spapr_populate_pci_child_dt(dev, fdt, offset, phb->index, drc_index,
+                                      drc_name);
+    g_assert(!ret);
+
+    *dt_offset = offset;
+    return fdt;
+}
+
+static void spapr_phb_add_pci_device(sPAPRDRConnector *drc,
+                                     sPAPRPHBState *phb,
+                                     PCIDevice *pdev,
+                                     Error **errp)
+{
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    DeviceState *dev = DEVICE(pdev);
+    int drc_index = drck->get_index(drc);
+    const char *drc_name = drck->get_name(drc);
+    void *fdt = NULL;
+    int fdt_start_offset = 0;
+
+    /* boot-time devices get their device tree node created by SLOF, but for
+     * hotplugged devices we need QEMU to generate it so the guest can fetch
+     * it via RTAS
+     */
+    if (dev->hotplugged) {
+        fdt = spapr_create_pci_child_dt(phb, pdev, drc_index, drc_name,
+                                        &fdt_start_offset);
+    }
+
+    drck->attach(drc, DEVICE(pdev),
+                 fdt, fdt_start_offset, !dev->hotplugged, errp);
+    if (*errp) {
+        g_free(fdt);
+    }
+}
+
+static void spapr_phb_remove_pci_device_cb(DeviceState *dev, void *opaque)
+{
+    /* some version guests do not wait for completion of a device
+     * cleanup (generally done asynchronously by the kernel) before
+     * signaling to QEMU that the device is safe, but instead sleep
+     * for some 'safe' period of time. unfortunately on a busy host
+     * this sleep isn't guaranteed to be long enough, resulting in
+     * bad things like IRQ lines being left asserted during final
+     * device removal. to deal with this we call reset just prior
+     * to finalizing the device, which will put the device back into
+     * an 'idle' state, as the device cleanup code expects.
+     */
+    pci_device_reset(PCI_DEVICE(dev));
+    object_unparent(OBJECT(dev));
+}
+
+static void spapr_phb_remove_pci_device(sPAPRDRConnector *drc,
+                                        sPAPRPHBState *phb,
+                                        PCIDevice *pdev,
+                                        Error **errp)
+{
+    sPAPRDRConnectorClass *drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+
+    drck->detach(drc, DEVICE(pdev), spapr_phb_remove_pci_device_cb, phb, errp);
+}
+
+static sPAPRDRConnector *spapr_phb_get_pci_drc(sPAPRPHBState *phb,
+                                               PCIDevice *pdev)
+{
+    uint32_t busnr = pci_bus_num(PCI_BUS(qdev_get_parent_bus(DEVICE(pdev))));
+    return spapr_dr_connector_by_id(SPAPR_DR_CONNECTOR_TYPE_PCI,
+                                    (phb->index << 16) |
+                                    (busnr << 8) |
+                                    pdev->devfn);
+}
+
+static void spapr_phb_hot_plug_child(HotplugHandler *plug_handler,
+                                     DeviceState *plugged_dev, Error **errp)
+{
+    sPAPRPHBState *phb = SPAPR_PCI_HOST_BRIDGE(DEVICE(plug_handler));
+    PCIDevice *pdev = PCI_DEVICE(plugged_dev);
+    sPAPRDRConnector *drc = spapr_phb_get_pci_drc(phb, pdev);
+    Error *local_err = NULL;
+
+    /* if DR is disabled we don't need to do anything in the case of
+     * hotplug or coldplug callbacks
+     */
+    if (!phb->dr_enabled) {
+        /* if this is a hotplug operation initiated by the user
+         * we need to let them know it's not enabled
+         */
+        if (plugged_dev->hotplugged) {
+            error_set(errp, QERR_BUS_NO_HOTPLUG,
+                      object_get_typename(OBJECT(phb)));
+        }
+        return;
+    }
+
+    g_assert(drc);
+
+    spapr_phb_add_pci_device(drc, phb, pdev, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+    if (plugged_dev->hotplugged) {
+        spapr_hotplug_req_add_event(drc);
+    }
+}
+
+static void spapr_phb_hot_unplug_child(HotplugHandler *plug_handler,
+                                       DeviceState *plugged_dev, Error **errp)
+{
+    sPAPRPHBState *phb = SPAPR_PCI_HOST_BRIDGE(DEVICE(plug_handler));
+    PCIDevice *pdev = PCI_DEVICE(plugged_dev);
+    sPAPRDRConnectorClass *drck;
+    sPAPRDRConnector *drc = spapr_phb_get_pci_drc(phb, pdev);
+    Error *local_err = NULL;
+
+    if (!phb->dr_enabled) {
+        error_set(errp, QERR_BUS_NO_HOTPLUG,
+                  object_get_typename(OBJECT(phb)));
+        return;
+    }
+
+    g_assert(drc);
+
+    drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    if (!drck->release_pending(drc)) {
+        spapr_phb_remove_pci_device(drc, phb, pdev, &local_err);
+        if (local_err) {
+            error_propagate(errp, local_err);
+            return;
+        }
+        spapr_hotplug_req_remove_event(drc);
+    }
+}
+
 static void spapr_phb_realize(DeviceState *dev, Error **errp)
 {
     SysBusDevice *s = SYS_BUS_DEVICE(dev);
@@ -742,12 +1119,12 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
     PCIBus *bus;
     uint64_t msi_window_size = 4096;
 
-    if (sphb->index != -1) {
+    if (sphb->index != (uint32_t)-1) {
         hwaddr windows_base;
 
-        if ((sphb->buid != -1) || (sphb->dma_liobn != -1)
-            || (sphb->mem_win_addr != -1)
-            || (sphb->io_win_addr != -1)) {
+        if ((sphb->buid != (uint64_t)-1) || (sphb->dma_liobn != (uint32_t)-1)
+            || (sphb->mem_win_addr != (hwaddr)-1)
+            || (sphb->io_win_addr != (hwaddr)-1)) {
             error_setg(errp, "Either \"index\" or other parameters must"
                        " be specified for PAPR PHB, not both");
             return;
@@ -760,7 +1137,7 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
         }
 
         sphb->buid = SPAPR_PCI_BASE_BUID + sphb->index;
-        sphb->dma_liobn = SPAPR_PCI_BASE_LIOBN + sphb->index;
+        sphb->dma_liobn = SPAPR_PCI_LIOBN(sphb->index, 0);
 
         windows_base = SPAPR_PCI_WINDOW_BASE
             + sphb->index * SPAPR_PCI_WINDOW_SPACING;
@@ -768,27 +1145,27 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
         sphb->io_win_addr = windows_base + SPAPR_PCI_IO_WIN_OFF;
     }
 
-    if (sphb->buid == -1) {
+    if (sphb->buid == (uint64_t)-1) {
         error_setg(errp, "BUID not specified for PHB");
         return;
     }
 
-    if (sphb->dma_liobn == -1) {
+    if (sphb->dma_liobn == (uint32_t)-1) {
         error_setg(errp, "LIOBN not specified for PHB");
         return;
     }
 
-    if (sphb->mem_win_addr == -1) {
+    if (sphb->mem_win_addr == (hwaddr)-1) {
         error_setg(errp, "Memory window address not specified for PHB");
         return;
     }
 
-    if (sphb->io_win_addr == -1) {
+    if (sphb->io_win_addr == (hwaddr)-1) {
         error_setg(errp, "IO window address not specified for PHB");
         return;
     }
 
-    if (find_phb(spapr, sphb->buid)) {
+    if (spapr_pci_find_phb(spapr, sphb->buid)) {
         error_setg(errp, "PCI host bridges must have unique BUIDs");
         return;
     }
@@ -824,6 +1201,7 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
                            &sphb->memspace, &sphb->iospace,
                            PCI_DEVFN(0, 0), PCI_NUM_PINS, TYPE_PCI_BUS);
     phb->bus = bus;
+    qbus_set_hotplug_handler(BUS(phb->bus), DEVICE(sphb), NULL);
 
     /*
      * Initialize PHB address space.
@@ -880,6 +1258,15 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
         sphb->lsi_table[i].irq = irq;
     }
 
+    /* allocate connectors for child PCI devices */
+    if (sphb->dr_enabled) {
+        for (i = 0; i < PCI_SLOT_MAX * 8; i++) {
+            spapr_dr_connector_new(OBJECT(phb),
+                                   SPAPR_DR_CONNECTOR_TYPE_PCI,
+                                   (sphb->index << 16) | i);
+        }
+    }
+
     if (!info->finish_realize) {
         error_setg(errp, "finish_realize not defined");
         return;
@@ -893,11 +1280,11 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
 static void spapr_phb_finish_realize(sPAPRPHBState *sphb, Error **errp)
 {
     sPAPRTCETable *tcet;
+    uint32_t nb_table;
 
+    nb_table = SPAPR_PCI_DMA32_SIZE >> SPAPR_TCE_PAGE_SHIFT;
     tcet = spapr_tce_new_table(DEVICE(sphb), sphb->dma_liobn,
-                               0,
-                               SPAPR_TCE_PAGE_SHIFT,
-                               0x40000000 >> SPAPR_TCE_PAGE_SHIFT, false);
+                               0, SPAPR_TCE_PAGE_SHIFT, nb_table, false);
     if (!tcet) {
         error_setg(errp, "Unable to create TCE table for %s",
                    sphb->dtbusname);
@@ -936,6 +1323,8 @@ static Property spapr_phb_properties[] = {
     DEFINE_PROP_UINT64("io_win_addr", sPAPRPHBState, io_win_addr, -1),
     DEFINE_PROP_UINT64("io_win_size", sPAPRPHBState, io_win_size,
                        SPAPR_PCI_IO_WIN_SIZE),
+    DEFINE_PROP_BOOL("dynamic-reconfiguration", sPAPRPHBState, dr_enabled,
+                     true),
     DEFINE_PROP_END_OF_LIST(),
 };
 
@@ -1049,6 +1438,7 @@ static void spapr_phb_class_init(ObjectClass *klass, void *data)
     PCIHostBridgeClass *hc = PCI_HOST_BRIDGE_CLASS(klass);
     DeviceClass *dc = DEVICE_CLASS(klass);
     sPAPRPHBClass *spc = SPAPR_PCI_HOST_BRIDGE_CLASS(klass);
+    HotplugHandlerClass *hp = HOTPLUG_HANDLER_CLASS(klass);
 
     hc->root_bus_path = spapr_phb_root_bus_path;
     dc->realize = spapr_phb_realize;
@@ -1058,6 +1448,8 @@ static void spapr_phb_class_init(ObjectClass *klass, void *data)
     set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
     dc->cannot_instantiate_with_device_add_yet = false;
     spc->finish_realize = spapr_phb_finish_realize;
+    hp->plug = spapr_phb_hot_plug_child;
+    hp->unplug = spapr_phb_hot_unplug_child;
 }
 
 static const TypeInfo spapr_phb_info = {
@@ -1066,6 +1458,10 @@ static const TypeInfo spapr_phb_info = {
     .instance_size = sizeof(sPAPRPHBState),
     .class_init    = spapr_phb_class_init,
     .class_size    = sizeof(sPAPRPHBClass),
+    .interfaces    = (InterfaceInfo[]) {
+        { TYPE_HOTPLUG_HANDLER },
+        { }
+    }
 };
 
 PCIHostState *spapr_create_phb(sPAPREnvironment *spapr, int index)
@@ -1079,45 +1475,11 @@ PCIHostState *spapr_create_phb(sPAPREnvironment *spapr, int index)
     return PCI_HOST_BRIDGE(dev);
 }
 
-/* Macros to operate with address in OF binding to PCI */
-#define b_x(x, p, l)    (((x) & ((1<<(l))-1)) << (p))
-#define b_n(x)          b_x((x), 31, 1) /* 0 if relocatable */
-#define b_p(x)          b_x((x), 30, 1) /* 1 if prefetchable */
-#define b_t(x)          b_x((x), 29, 1) /* 1 if the address is aliased */
-#define b_ss(x)         b_x((x), 24, 2) /* the space code */
-#define b_bbbbbbbb(x)   b_x((x), 16, 8) /* bus number */
-#define b_ddddd(x)      b_x((x), 11, 5) /* device number */
-#define b_fff(x)        b_x((x), 8, 3)  /* function number */
-#define b_rrrrrrrr(x)   b_x((x), 0, 8)  /* register number */
-
-typedef struct sPAPRTCEDT {
-    void *fdt;
-    int node_off;
-} sPAPRTCEDT;
-
-static int spapr_phb_children_dt(Object *child, void *opaque)
-{
-    sPAPRTCEDT *p = opaque;
-    sPAPRTCETable *tcet;
-
-    tcet = (sPAPRTCETable *) object_dynamic_cast(child, TYPE_SPAPR_TCE_TABLE);
-    if (!tcet) {
-        return 0;
-    }
-
-    spapr_dma_dt(p->fdt, p->node_off, "ibm,dma-window",
-                 tcet->liobn, tcet->bus_offset,
-                 tcet->nb_table << tcet->page_shift);
-    /* Stop after the first window */
-
-    return 1;
-}
-
 int spapr_populate_pci_dt(sPAPRPHBState *phb,
                           uint32_t xics_phandle,
                           void *fdt)
 {
-    int bus_off, i, j;
+    int bus_off, i, j, ret;
     char nodename[256];
     uint32_t bus_range[] = { cpu_to_be32(0), cpu_to_be32(0xff) };
     const uint64_t mmiosize = memory_region_size(&phb->memwindow);
@@ -1151,6 +1513,7 @@ int spapr_populate_pci_dt(sPAPRPHBState *phb,
     uint32_t interrupt_map_mask[] = {
         cpu_to_be32(b_ddddd(-1)|b_fff(0)), 0x0, 0x0, cpu_to_be32(-1)};
     uint32_t interrupt_map[PCI_SLOT_MAX * PCI_NUM_PINS][7];
+    sPAPRTCETable *tcet;
 
     /* Start populating the FDT */
     sprintf(nodename, "pci@%" PRIx64, phb->buid);
@@ -1159,14 +1522,6 @@ int spapr_populate_pci_dt(sPAPRPHBState *phb,
         return bus_off;
     }
 
-#define _FDT(exp) \
-    do { \
-        int ret = (exp);                                           \
-        if (ret < 0) {                                             \
-            return ret;                                            \
-        }                                                          \
-    } while (0)
-
     /* Write PHB properties */
     _FDT(fdt_setprop_string(fdt, bus_off, "device_type", "pci"));
     _FDT(fdt_setprop_string(fdt, bus_off, "compatible", "IBM,Logical_PHB"));
@@ -1203,8 +1558,16 @@ int spapr_populate_pci_dt(sPAPRPHBState *phb,
     _FDT(fdt_setprop(fdt, bus_off, "interrupt-map", &interrupt_map,
                      sizeof(interrupt_map)));
 
-    object_child_foreach(OBJECT(phb), spapr_phb_children_dt,
-                         &((sPAPRTCEDT){ .fdt = fdt, .node_off = bus_off }));
+    tcet = spapr_tce_find_by_liobn(SPAPR_PCI_LIOBN(phb->index, 0));
+    spapr_dma_dt(fdt, bus_off, "ibm,dma-window",
+                 tcet->liobn, tcet->bus_offset,
+                 tcet->nb_table << tcet->page_shift);
+
+    ret = spapr_drc_populate_dt(fdt, bus_off, OBJECT(phb),
+                                SPAPR_DR_CONNECTOR_TYPE_PCI);
+    if (ret) {
+        return ret;
+    }
 
     return 0;
 }
diff --git a/hw/ppc/spapr_rtas.c b/hw/ppc/spapr_rtas.c
index 0f1ae55828..fa28d43f81 100644
--- a/hw/ppc/spapr_rtas.c
+++ b/hw/ppc/spapr_rtas.c
@@ -35,6 +35,55 @@
 #include "qapi-event.h"
 
 #include <libfdt.h>
+#include "hw/ppc/spapr_drc.h"
+
+/* #define DEBUG_SPAPR */
+
+#ifdef DEBUG_SPAPR
+#define DPRINTF(fmt, ...) \
+    do { fprintf(stderr, fmt, ## __VA_ARGS__); } while (0)
+#else
+#define DPRINTF(fmt, ...) \
+    do { } while (0)
+#endif
+
+static sPAPRConfigureConnectorState *spapr_ccs_find(sPAPREnvironment *spapr,
+                                                    uint32_t drc_index)
+{
+    sPAPRConfigureConnectorState *ccs = NULL;
+
+    QTAILQ_FOREACH(ccs, &spapr->ccs_list, next) {
+        if (ccs->drc_index == drc_index) {
+            break;
+        }
+    }
+
+    return ccs;
+}
+
+static void spapr_ccs_add(sPAPREnvironment *spapr,
+                          sPAPRConfigureConnectorState *ccs)
+{
+    g_assert(!spapr_ccs_find(spapr, ccs->drc_index));
+    QTAILQ_INSERT_HEAD(&spapr->ccs_list, ccs, next);
+}
+
+static void spapr_ccs_remove(sPAPREnvironment *spapr,
+                             sPAPRConfigureConnectorState *ccs)
+{
+    QTAILQ_REMOVE(&spapr->ccs_list, ccs, next);
+    g_free(ccs);
+}
+
+void spapr_ccs_reset_hook(void *opaque)
+{
+    sPAPREnvironment *spapr = opaque;
+    sPAPRConfigureConnectorState *ccs, *ccs_tmp;
+
+    QTAILQ_FOREACH_SAFE(ccs, &spapr->ccs_list, next, ccs_tmp) {
+        spapr_ccs_remove(spapr, ccs);
+    }
+}
 
 static void rtas_display_character(PowerPCCPU *cpu, sPAPREnvironment *spapr,
                                    uint32_t token, uint32_t nargs,
@@ -245,6 +294,308 @@ static void rtas_ibm_os_term(PowerPCCPU *cpu,
     rtas_st(rets, 0, ret);
 }
 
+static void rtas_set_power_level(PowerPCCPU *cpu, sPAPREnvironment *spapr,
+                                 uint32_t token, uint32_t nargs,
+                                 target_ulong args, uint32_t nret,
+                                 target_ulong rets)
+{
+    int32_t power_domain;
+
+    if (nargs != 2 || nret != 2) {
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+
+    /* we currently only use a single, "live insert" powerdomain for
+     * hotplugged/dlpar'd resources, so the power is always live/full (100)
+     */
+    power_domain = rtas_ld(args, 0);
+    if (power_domain != -1) {
+        rtas_st(rets, 0, RTAS_OUT_NOT_SUPPORTED);
+        return;
+    }
+
+    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
+    rtas_st(rets, 1, 100);
+}
+
+static void rtas_get_power_level(PowerPCCPU *cpu, sPAPREnvironment *spapr,
+                                  uint32_t token, uint32_t nargs,
+                                  target_ulong args, uint32_t nret,
+                                  target_ulong rets)
+{
+    int32_t power_domain;
+
+    if (nargs != 1 || nret != 2) {
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+
+    /* we currently only use a single, "live insert" powerdomain for
+     * hotplugged/dlpar'd resources, so the power is always live/full (100)
+     */
+    power_domain = rtas_ld(args, 0);
+    if (power_domain != -1) {
+        rtas_st(rets, 0, RTAS_OUT_NOT_SUPPORTED);
+        return;
+    }
+
+    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
+    rtas_st(rets, 1, 100);
+}
+
+static bool sensor_type_is_dr(uint32_t sensor_type)
+{
+    switch (sensor_type) {
+    case RTAS_SENSOR_TYPE_ISOLATION_STATE:
+    case RTAS_SENSOR_TYPE_DR:
+    case RTAS_SENSOR_TYPE_ALLOCATION_STATE:
+        return true;
+    }
+
+    return false;
+}
+
+static void rtas_set_indicator(PowerPCCPU *cpu, sPAPREnvironment *spapr,
+                               uint32_t token, uint32_t nargs,
+                               target_ulong args, uint32_t nret,
+                               target_ulong rets)
+{
+    uint32_t sensor_type;
+    uint32_t sensor_index;
+    uint32_t sensor_state;
+    sPAPRDRConnector *drc;
+    sPAPRDRConnectorClass *drck;
+
+    if (nargs != 3 || nret != 1) {
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+
+    sensor_type = rtas_ld(args, 0);
+    sensor_index = rtas_ld(args, 1);
+    sensor_state = rtas_ld(args, 2);
+
+    if (!sensor_type_is_dr(sensor_type)) {
+        goto out_unimplemented;
+    }
+
+    /* if this is a DR sensor we can assume sensor_index == drc_index */
+    drc = spapr_dr_connector_by_index(sensor_index);
+    if (!drc) {
+        DPRINTF("rtas_set_indicator: invalid sensor/DRC index: %xh\n",
+                sensor_index);
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+    drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+
+    switch (sensor_type) {
+    case RTAS_SENSOR_TYPE_ISOLATION_STATE:
+        /* if the guest is configuring a device attached to this
+         * DRC, we should reset the configuration state at this
+         * point since it may no longer be reliable (guest released
+         * device and needs to start over, or unplug occurred so
+         * the FDT is no longer valid)
+         */
+        if (sensor_state == SPAPR_DR_ISOLATION_STATE_ISOLATED) {
+            sPAPRConfigureConnectorState *ccs = spapr_ccs_find(spapr,
+                                                               sensor_index);
+            if (ccs) {
+                spapr_ccs_remove(spapr, ccs);
+            }
+        }
+        drck->set_isolation_state(drc, sensor_state);
+        break;
+    case RTAS_SENSOR_TYPE_DR:
+        drck->set_indicator_state(drc, sensor_state);
+        break;
+    case RTAS_SENSOR_TYPE_ALLOCATION_STATE:
+        drck->set_allocation_state(drc, sensor_state);
+        break;
+    default:
+        goto out_unimplemented;
+    }
+
+    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
+    return;
+
+out_unimplemented:
+    /* currently only DR-related sensors are implemented */
+    DPRINTF("rtas_set_indicator: sensor/indicator not implemented: %d\n",
+            sensor_type);
+    rtas_st(rets, 0, RTAS_OUT_NOT_SUPPORTED);
+}
+
+static void rtas_get_sensor_state(PowerPCCPU *cpu, sPAPREnvironment *spapr,
+                                  uint32_t token, uint32_t nargs,
+                                  target_ulong args, uint32_t nret,
+                                  target_ulong rets)
+{
+    uint32_t sensor_type;
+    uint32_t sensor_index;
+    sPAPRDRConnector *drc;
+    sPAPRDRConnectorClass *drck;
+    uint32_t entity_sense;
+
+    if (nargs != 2 || nret != 2) {
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+
+    sensor_type = rtas_ld(args, 0);
+    sensor_index = rtas_ld(args, 1);
+
+    if (sensor_type != RTAS_SENSOR_TYPE_ENTITY_SENSE) {
+        /* currently only DR-related sensors are implemented */
+        DPRINTF("rtas_get_sensor_state: sensor/indicator not implemented: %d\n",
+                sensor_type);
+        rtas_st(rets, 0, RTAS_OUT_NOT_SUPPORTED);
+        return;
+    }
+
+    drc = spapr_dr_connector_by_index(sensor_index);
+    if (!drc) {
+        DPRINTF("rtas_get_sensor_state: invalid sensor/DRC index: %xh\n",
+                sensor_index);
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+    drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    entity_sense = drck->entity_sense(drc);
+
+    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
+    rtas_st(rets, 1, entity_sense);
+}
+
+/* configure-connector work area offsets, int32_t units for field
+ * indexes, bytes for field offset/len values.
+ *
+ * as documented by PAPR+ v2.7, 13.5.3.5
+ */
+#define CC_IDX_NODE_NAME_OFFSET 2
+#define CC_IDX_PROP_NAME_OFFSET 2
+#define CC_IDX_PROP_LEN 3
+#define CC_IDX_PROP_DATA_OFFSET 4
+#define CC_VAL_DATA_OFFSET ((CC_IDX_PROP_DATA_OFFSET + 1) * 4)
+#define CC_WA_LEN 4096
+
+static void rtas_ibm_configure_connector(PowerPCCPU *cpu,
+                                         sPAPREnvironment *spapr,
+                                         uint32_t token, uint32_t nargs,
+                                         target_ulong args, uint32_t nret,
+                                         target_ulong rets)
+{
+    uint64_t wa_addr;
+    uint64_t wa_offset;
+    uint32_t drc_index;
+    sPAPRDRConnector *drc;
+    sPAPRDRConnectorClass *drck;
+    sPAPRConfigureConnectorState *ccs;
+    sPAPRDRCCResponse resp = SPAPR_DR_CC_RESPONSE_CONTINUE;
+    int rc;
+    const void *fdt;
+
+    if (nargs != 2 || nret != 1) {
+        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
+        return;
+    }
+
+    wa_addr = ((uint64_t)rtas_ld(args, 1) << 32) | rtas_ld(args, 0);
+
+    drc_index = rtas_ld(wa_addr, 0);
+    drc = spapr_dr_connector_by_index(drc_index);
+    if (!drc) {
+        DPRINTF("rtas_ibm_configure_connector: invalid DRC index: %xh\n",
+                drc_index);
+        rc = RTAS_OUT_PARAM_ERROR;
+        goto out;
+    }
+
+    drck = SPAPR_DR_CONNECTOR_GET_CLASS(drc);
+    fdt = drck->get_fdt(drc, NULL);
+
+    ccs = spapr_ccs_find(spapr, drc_index);
+    if (!ccs) {
+        ccs = g_new0(sPAPRConfigureConnectorState, 1);
+        (void)drck->get_fdt(drc, &ccs->fdt_offset);
+        ccs->drc_index = drc_index;
+        spapr_ccs_add(spapr, ccs);
+    }
+
+    do {
+        uint32_t tag;
+        const char *name;
+        const struct fdt_property *prop;
+        int fdt_offset_next, prop_len;
+
+        tag = fdt_next_tag(fdt, ccs->fdt_offset, &fdt_offset_next);
+
+        switch (tag) {
+        case FDT_BEGIN_NODE:
+            ccs->fdt_depth++;
+            name = fdt_get_name(fdt, ccs->fdt_offset, NULL);
+
+            /* provide the name of the next OF node */
+            wa_offset = CC_VAL_DATA_OFFSET;
+            rtas_st(wa_addr, CC_IDX_NODE_NAME_OFFSET, wa_offset);
+            rtas_st_buffer_direct(wa_addr + wa_offset, CC_WA_LEN - wa_offset,
+                                  (uint8_t *)name, strlen(name) + 1);
+            resp = SPAPR_DR_CC_RESPONSE_NEXT_CHILD;
+            break;
+        case FDT_END_NODE:
+            ccs->fdt_depth--;
+            if (ccs->fdt_depth == 0) {
+                /* done sending the device tree, don't need to track
+                 * the state anymore
+                 */
+                drck->set_configured(drc);
+                spapr_ccs_remove(spapr, ccs);
+                ccs = NULL;
+                resp = SPAPR_DR_CC_RESPONSE_SUCCESS;
+            } else {
+                resp = SPAPR_DR_CC_RESPONSE_PREV_PARENT;
+            }
+            break;
+        case FDT_PROP:
+            prop = fdt_get_property_by_offset(fdt, ccs->fdt_offset,
+                                              &prop_len);
+            name = fdt_string(fdt, fdt32_to_cpu(prop->nameoff));
+
+            /* provide the name of the next OF property */
+            wa_offset = CC_VAL_DATA_OFFSET;
+            rtas_st(wa_addr, CC_IDX_PROP_NAME_OFFSET, wa_offset);
+            rtas_st_buffer_direct(wa_addr + wa_offset, CC_WA_LEN - wa_offset,
+                                  (uint8_t *)name, strlen(name) + 1);
+
+            /* provide the length and value of the OF property. data gets
+             * placed immediately after NULL terminator of the OF property's
+             * name string
+             */
+            wa_offset += strlen(name) + 1,
+            rtas_st(wa_addr, CC_IDX_PROP_LEN, prop_len);
+            rtas_st(wa_addr, CC_IDX_PROP_DATA_OFFSET, wa_offset);
+            rtas_st_buffer_direct(wa_addr + wa_offset, CC_WA_LEN - wa_offset,
+                                  (uint8_t *)((struct fdt_property *)prop)->data,
+                                  prop_len);
+            resp = SPAPR_DR_CC_RESPONSE_NEXT_PROPERTY;
+            break;
+        case FDT_END:
+            resp = SPAPR_DR_CC_RESPONSE_ERROR;
+        default:
+            /* keep seeking for an actionable tag */
+            break;
+        }
+        if (ccs) {
+            ccs->fdt_offset = fdt_offset_next;
+        }
+    } while (resp == SPAPR_DR_CC_RESPONSE_CONTINUE);
+
+    rc = resp;
+out:
+    rtas_st(rets, 0, rc);
+}
+
 static struct rtas_call {
     const char *name;
     spapr_rtas_fn fn;
@@ -370,6 +721,16 @@ static void core_rtas_register_types(void)
                         rtas_ibm_set_system_parameter);
     spapr_rtas_register(RTAS_IBM_OS_TERM, "ibm,os-term",
                         rtas_ibm_os_term);
+    spapr_rtas_register(RTAS_SET_POWER_LEVEL, "set-power-level",
+                        rtas_set_power_level);
+    spapr_rtas_register(RTAS_GET_POWER_LEVEL, "get-power-level",
+                        rtas_get_power_level);
+    spapr_rtas_register(RTAS_SET_INDICATOR, "set-indicator",
+                        rtas_set_indicator);
+    spapr_rtas_register(RTAS_GET_SENSOR_STATE, "get-sensor-state",
+                        rtas_get_sensor_state);
+    spapr_rtas_register(RTAS_IBM_CONFIGURE_CONNECTOR, "ibm,configure-connector",
+                        rtas_ibm_configure_connector);
 }
 
 type_init(core_rtas_register_types)
diff --git a/hw/ppc/spapr_vio.c b/hw/ppc/spapr_vio.c
index 1360b97ab0..174033dd41 100644
--- a/hw/ppc/spapr_vio.c
+++ b/hw/ppc/spapr_vio.c
@@ -469,7 +469,7 @@ static void spapr_vio_busdev_realize(DeviceState *qdev, Error **errp)
     }
 
     if (pc->rtce_window_size) {
-        uint32_t liobn = SPAPR_VIO_BASE_LIOBN | dev->reg;
+        uint32_t liobn = SPAPR_VIO_LIOBN(dev->reg);
 
         memory_region_init(&dev->mrroot, OBJECT(dev), "iommu-spapr-root",
                            ram_size);
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index 8a565f657a..c574988c36 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -216,6 +216,7 @@ static void ccw_machine_class_init(ObjectClass *oc, void *data)
     mc->no_sdcard = 1;
     mc->use_sclp = 1;
     mc->max_cpus = 255;
+    mc->is_default = 1;
     nc->nmi_monitor_handler = s390_nmi;
 }
 
diff --git a/hw/s390x/s390-virtio.c b/hw/s390x/s390-virtio.c
index 59750dbfcd..00ea793651 100644
--- a/hw/s390x/s390-virtio.c
+++ b/hw/s390x/s390-virtio.c
@@ -345,7 +345,6 @@ static void s390_machine_class_init(ObjectClass *oc, void *data)
     mc->no_floppy = 1;
     mc->no_cdrom = 1;
     mc->no_sdcard = 1;
-    mc->is_default = 1;
     nc->nmi_monitor_handler = s390_nmi;
 }
 
diff --git a/hw/s390x/virtio-ccw.c b/hw/s390x/virtio-ccw.c
index 3ef0055126..e32ada9bf1 100644
--- a/hw/s390x/virtio-ccw.c
+++ b/hw/s390x/virtio-ccw.c
@@ -1,8 +1,9 @@
 /*
  * virtio ccw target implementation
  *
- * Copyright 2012,2014 IBM Corp.
+ * Copyright 2012,2015 IBM Corp.
  * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com>
+ *            Pierre Morel <pmorel@linux.vnet.ibm.com>
  *
  * This work is licensed under the terms of the GNU GPL, version 2 or (at
  * your option) any later version. See the COPYING file in the top-level
@@ -1314,6 +1315,7 @@ static void virtio_ccw_save_config(DeviceState *d, QEMUFile *f)
 {
     VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
     SubchDev *s = dev->sch;
+    VirtIODevice *vdev = virtio_ccw_get_vdev(s);
 
     subch_device_save(s, f);
     if (dev->indicators != NULL) {
@@ -1337,6 +1339,7 @@ static void virtio_ccw_save_config(DeviceState *d, QEMUFile *f)
         qemu_put_be32(f, 0);
         qemu_put_be64(f, 0UL);
     }
+    qemu_put_be16(f, vdev->config_vector);
     qemu_put_be64(f, dev->routes.adapter.ind_offset);
     qemu_put_byte(f, dev->thinint_isc);
 }
@@ -1345,6 +1348,7 @@ static int virtio_ccw_load_config(DeviceState *d, QEMUFile *f)
 {
     VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
     SubchDev *s = dev->sch;
+    VirtIODevice *vdev = virtio_ccw_get_vdev(s);
     int len;
 
     s->driver_data = dev;
@@ -1370,6 +1374,7 @@ static int virtio_ccw_load_config(DeviceState *d, QEMUFile *f)
         qemu_get_be64(f);
         dev->summary_indicator = NULL;
     }
+    qemu_get_be16s(f, &vdev->config_vector);
     dev->routes.adapter.ind_offset = qemu_get_be64(f);
     dev->thinint_isc = qemu_get_byte(f);
     if (s->thinint_active) {
@@ -1396,6 +1401,10 @@ static void virtio_ccw_device_plugged(DeviceState *d, Error **errp)
         return;
     }
 
+    if (!kvm_eventfds_enabled()) {
+        dev->flags &= ~VIRTIO_CCW_FLAG_USE_IOEVENTFD;
+    }
+
     sch->id.cu_model = virtio_bus_get_vdev_id(&dev->bus);
 
     css_generate_sch_crws(sch->cssid, sch->ssid, sch->schid,
@@ -1734,6 +1743,56 @@ static const TypeInfo virtio_ccw_bus_info = {
     .class_init = virtio_ccw_bus_class_init,
 };
 
+#ifdef CONFIG_VIRTFS
+static Property virtio_ccw_9p_properties[] = {
+    DEFINE_PROP_STRING("devno", VirtioCcwDevice, bus_id),
+    DEFINE_PROP_BIT("ioeventfd", VirtioCcwDevice, flags,
+            VIRTIO_CCW_FLAG_USE_IOEVENTFD_BIT, true),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_ccw_9p_realize(VirtioCcwDevice *ccw_dev, Error **errp)
+{
+    V9fsCCWState *dev = VIRTIO_9P_CCW(ccw_dev);
+    DeviceState *vdev = DEVICE(&dev->vdev);
+    Error *err = NULL;
+
+    qdev_set_parent_bus(vdev, BUS(&ccw_dev->bus));
+    object_property_set_bool(OBJECT(vdev), true, "realized", &err);
+    if (err) {
+        error_propagate(errp, err);
+    }
+}
+
+static void virtio_ccw_9p_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtIOCCWDeviceClass *k = VIRTIO_CCW_DEVICE_CLASS(klass);
+
+    k->exit = virtio_ccw_exit;
+    k->realize = virtio_ccw_9p_realize;
+    dc->reset = virtio_ccw_reset;
+    dc->props = virtio_ccw_9p_properties;
+    set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
+}
+
+static void virtio_ccw_9p_instance_init(Object *obj)
+{
+    V9fsCCWState *dev = VIRTIO_9P_CCW(obj);
+
+    virtio_instance_init_common(obj, &dev->vdev, sizeof(dev->vdev),
+                                TYPE_VIRTIO_9P);
+}
+
+static const TypeInfo virtio_ccw_9p_info = {
+    .name          = TYPE_VIRTIO_9P_CCW,
+    .parent        = TYPE_VIRTIO_CCW_DEVICE,
+    .instance_size = sizeof(V9fsCCWState),
+    .instance_init = virtio_ccw_9p_instance_init,
+    .class_init    = virtio_ccw_9p_class_init,
+};
+#endif
+
 static void virtio_ccw_register(void)
 {
     type_register_static(&virtio_ccw_bus_info);
@@ -1749,6 +1808,9 @@ static void virtio_ccw_register(void)
 #endif
     type_register_static(&virtio_ccw_rng);
     type_register_static(&virtual_css_bridge_info);
+#ifdef CONFIG_VIRTFS
+    type_register_static(&virtio_ccw_9p_info);
+#endif
 }
 
 type_init(virtio_ccw_register)
diff --git a/hw/s390x/virtio-ccw.h b/hw/s390x/virtio-ccw.h
index ad3af7626a..d729263960 100644
--- a/hw/s390x/virtio-ccw.h
+++ b/hw/s390x/virtio-ccw.h
@@ -1,8 +1,9 @@
 /*
  * virtio ccw target definitions
  *
- * Copyright 2012 IBM Corp.
+ * Copyright 2012,2015 IBM Corp.
  * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com>
+ *            Pierre Morel <pmorel@linux.vnet.ibm.com>
  *
  * This work is licensed under the terms of the GNU GPL, version 2 or (at
  * your option) any later version. See the COPYING file in the top-level
@@ -189,4 +190,19 @@ typedef struct VirtIORNGCcw {
 VirtualCssBus *virtual_css_bus_init(void);
 void virtio_ccw_device_update_status(SubchDev *sch);
 VirtIODevice *virtio_ccw_get_vdev(SubchDev *sch);
+
+#ifdef CONFIG_VIRTFS
+#include "hw/9pfs/virtio-9p.h"
+
+#define TYPE_VIRTIO_9P_CCW "virtio-9p-ccw"
+#define VIRTIO_9P_CCW(obj) \
+    OBJECT_CHECK(V9fsCCWState, (obj), TYPE_VIRTIO_9P_CCW)
+
+typedef struct V9fsCCWState {
+    VirtioCcwDevice parent_obj;
+    V9fsState vdev;
+} V9fsCCWState;
+
+#endif /* CONFIG_VIRTFS */
+
 #endif
diff --git a/hw/scsi/scsi-bus.c b/hw/scsi/scsi-bus.c
index bd2c0e4caa..f50b2f08af 100644
--- a/hw/scsi/scsi-bus.c
+++ b/hw/scsi/scsi-bus.c
@@ -1968,6 +1968,7 @@ static const VMStateDescription vmstate_scsi_sense_state = {
     .name = "SCSIDevice/sense",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = scsi_sense_state_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8_SUB_ARRAY(sense, SCSIDevice,
                                 SCSI_SENSE_BUF_SIZE_OLD,
@@ -1998,13 +1999,9 @@ const VMStateDescription vmstate_scsi_device = {
         },
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_scsi_sense_state,
-            .needed = scsi_sense_state_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_scsi_sense_state,
+        NULL
     }
 };
 
diff --git a/hw/sd/pxa2xx_mmci.c b/hw/sd/pxa2xx_mmci.c
index ac3ab39bea..d1fe6d58e8 100644
--- a/hw/sd/pxa2xx_mmci.c
+++ b/hw/sd/pxa2xx_mmci.c
@@ -48,7 +48,6 @@ struct PXA2xxMMCIState {
     int resp_len;
 
     int cmdreq;
-    int ac_width;
 };
 
 #define MMC_STRPCL	0x00	/* MMC Clock Start/Stop register */
@@ -215,7 +214,7 @@ static void pxa2xx_mmci_wakequeues(PXA2xxMMCIState *s)
     pxa2xx_mmci_fifo_update(s);
 }
 
-static uint32_t pxa2xx_mmci_read(void *opaque, hwaddr offset)
+static uint64_t pxa2xx_mmci_read(void *opaque, hwaddr offset, unsigned size)
 {
     PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
     uint32_t ret;
@@ -257,8 +256,8 @@ static uint32_t pxa2xx_mmci_read(void *opaque, hwaddr offset)
         return 0;
     case MMC_RXFIFO:
         ret = 0;
-        while (s->ac_width -- && s->rx_len) {
-            ret |= s->rx_fifo[s->rx_start ++] << (s->ac_width << 3);
+        while (size-- && s->rx_len) {
+            ret |= s->rx_fifo[s->rx_start++] << (size << 3);
             s->rx_start &= 0x1f;
             s->rx_len --;
         }
@@ -277,7 +276,7 @@ static uint32_t pxa2xx_mmci_read(void *opaque, hwaddr offset)
 }
 
 static void pxa2xx_mmci_write(void *opaque,
-                hwaddr offset, uint32_t value)
+                              hwaddr offset, uint64_t value, unsigned size)
 {
     PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
 
@@ -370,9 +369,9 @@ static void pxa2xx_mmci_write(void *opaque,
         break;
 
     case MMC_TXFIFO:
-        while (s->ac_width -- && s->tx_len < 0x20)
+        while (size-- && s->tx_len < 0x20)
             s->tx_fifo[(s->tx_start + (s->tx_len ++)) & 0x1f] =
-                    (value >> (s->ac_width << 3)) & 0xff;
+                    (value >> (size << 3)) & 0xff;
         s->intreq &= ~INT_TXFIFO_REQ;
         pxa2xx_mmci_fifo_update(s);
         break;
@@ -386,60 +385,9 @@ static void pxa2xx_mmci_write(void *opaque,
     }
 }
 
-static uint32_t pxa2xx_mmci_readb(void *opaque, hwaddr offset)
-{
-    PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
-    s->ac_width = 1;
-    return pxa2xx_mmci_read(opaque, offset);
-}
-
-static uint32_t pxa2xx_mmci_readh(void *opaque, hwaddr offset)
-{
-    PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
-    s->ac_width = 2;
-    return pxa2xx_mmci_read(opaque, offset);
-}
-
-static uint32_t pxa2xx_mmci_readw(void *opaque, hwaddr offset)
-{
-    PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
-    s->ac_width = 4;
-    return pxa2xx_mmci_read(opaque, offset);
-}
-
-static void pxa2xx_mmci_writeb(void *opaque,
-                hwaddr offset, uint32_t value)
-{
-    PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
-    s->ac_width = 1;
-    pxa2xx_mmci_write(opaque, offset, value);
-}
-
-static void pxa2xx_mmci_writeh(void *opaque,
-                hwaddr offset, uint32_t value)
-{
-    PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
-    s->ac_width = 2;
-    pxa2xx_mmci_write(opaque, offset, value);
-}
-
-static void pxa2xx_mmci_writew(void *opaque,
-                hwaddr offset, uint32_t value)
-{
-    PXA2xxMMCIState *s = (PXA2xxMMCIState *) opaque;
-    s->ac_width = 4;
-    pxa2xx_mmci_write(opaque, offset, value);
-}
-
 static const MemoryRegionOps pxa2xx_mmci_ops = {
-    .old_mmio = {
-        .read = { pxa2xx_mmci_readb,
-                  pxa2xx_mmci_readh,
-                  pxa2xx_mmci_readw, },
-        .write = { pxa2xx_mmci_writeb,
-                   pxa2xx_mmci_writeh,
-                   pxa2xx_mmci_writew, },
-    },
+    .read = pxa2xx_mmci_read,
+    .write = pxa2xx_mmci_write,
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
diff --git a/hw/sh4/r2d.c b/hw/sh4/r2d.c
index 4221060308..5e22ed79b2 100644
--- a/hw/sh4/r2d.c
+++ b/hw/sh4/r2d.c
@@ -127,7 +127,7 @@ static void r2d_fpga_irq_set(void *opaque, int n, int level)
     update_irl(fpga);
 }
 
-static uint32_t r2d_fpga_read(void *opaque, hwaddr addr)
+static uint64_t r2d_fpga_read(void *opaque, hwaddr addr, unsigned int size)
 {
     r2d_fpga_t *s = opaque;
 
@@ -146,7 +146,7 @@ static uint32_t r2d_fpga_read(void *opaque, hwaddr addr)
 }
 
 static void
-r2d_fpga_write(void *opaque, hwaddr addr, uint32_t value)
+r2d_fpga_write(void *opaque, hwaddr addr, uint64_t value, unsigned int size)
 {
     r2d_fpga_t *s = opaque;
 
@@ -170,10 +170,10 @@ r2d_fpga_write(void *opaque, hwaddr addr, uint32_t value)
 }
 
 static const MemoryRegionOps r2d_fpga_ops = {
-    .old_mmio = {
-        .read = { r2d_fpga_read, r2d_fpga_read, NULL, },
-        .write = { r2d_fpga_write, r2d_fpga_write, NULL, },
-    },
+    .read = r2d_fpga_read,
+    .write = r2d_fpga_write,
+    .impl.min_access_size = 2,
+    .impl.max_access_size = 2,
     .endianness = DEVICE_NATIVE_ENDIAN,
 };
 
diff --git a/hw/sparc/sun4m.c b/hw/sparc/sun4m.c
index a69bf2da45..68ac4d8bba 100644
--- a/hw/sparc/sun4m.c
+++ b/hw/sparc/sun4m.c
@@ -124,7 +124,7 @@ void DMA_register_channel (int nchan,
 static void fw_cfg_boot_set(void *opaque, const char *boot_device,
                             Error **errp)
 {
-    fw_cfg_add_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
+    fw_cfg_modify_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
 }
 
 static void nvram_init(Nvram *nvram, uint8_t *macaddr,
@@ -897,7 +897,6 @@ static void sun4m_hw_init(const struct sun4m_hwdef *hwdef,
         espdma_irq, ledma_irq;
     qemu_irq esp_reset, dma_enable;
     qemu_irq fdc_tc;
-    qemu_irq *cpu_halt;
     unsigned long kernel_size;
     DriveInfo *fd[MAX_FD];
     FWCfgState *fw_cfg;
@@ -1024,9 +1023,8 @@ static void sun4m_hw_init(const struct sun4m_hwdef *hwdef,
     escc_init(hwdef->serial_base, slavio_irq[15], slavio_irq[15],
               serial_hds[0], serial_hds[1], ESCC_CLOCK, 1);
 
-    cpu_halt = qemu_allocate_irqs(cpu_halt_signal, NULL, 1);
     if (hwdef->apc_base) {
-        apc_init(hwdef->apc_base, cpu_halt[0]);
+        apc_init(hwdef->apc_base, qemu_allocate_irq(cpu_halt_signal, NULL, 0));
     }
 
     if (hwdef->fd_base) {
@@ -1036,7 +1034,7 @@ static void sun4m_hw_init(const struct sun4m_hwdef *hwdef,
         sun4m_fdctrl_init(slavio_irq[22], hwdef->fd_base, fd,
                           &fdc_tc);
     } else {
-        fdc_tc = *qemu_allocate_irqs(dummy_fdc_tc, NULL, 1);
+        fdc_tc = qemu_allocate_irq(dummy_fdc_tc, NULL, 0);
     }
 
     slavio_misc_init(hwdef->slavio_base, hwdef->aux1_base, hwdef->aux2_base,
diff --git a/hw/sparc64/sun4u.c b/hw/sparc64/sun4u.c
index 6f34e87935..30cfa0e0a0 100644
--- a/hw/sparc64/sun4u.c
+++ b/hw/sparc64/sun4u.c
@@ -127,7 +127,7 @@ void DMA_register_channel (int nchan,
 static void fw_cfg_boot_set(void *opaque, const char *boot_device,
                             Error **errp)
 {
-    fw_cfg_add_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
+    fw_cfg_modify_i16(opaque, FW_CFG_BOOT_DEVICE, boot_device[0]);
 }
 
 static int sun4u_NVRAM_set_params(Nvram *nvram, uint16_t NVRAM_size,
diff --git a/hw/timer/arm_timer.c b/hw/timer/arm_timer.c
index 145291016b..d53f39ad62 100644
--- a/hw/timer/arm_timer.c
+++ b/hw/timer/arm_timer.c
@@ -280,14 +280,12 @@ static int sp804_init(SysBusDevice *sbd)
 {
     DeviceState *dev = DEVICE(sbd);
     SP804State *s = SP804(dev);
-    qemu_irq *qi;
 
-    qi = qemu_allocate_irqs(sp804_set_irq, s, 2);
     sysbus_init_irq(sbd, &s->irq);
     s->timer[0] = arm_timer_init(s->freq0);
     s->timer[1] = arm_timer_init(s->freq1);
-    s->timer[0]->irq = qi[0];
-    s->timer[1]->irq = qi[1];
+    s->timer[0]->irq = qemu_allocate_irq(sp804_set_irq, s, 0);
+    s->timer[1]->irq = qemu_allocate_irq(sp804_set_irq, s, 1);
     memory_region_init_io(&s->iomem, OBJECT(s), &sp804_ops, s,
                           "sp804", 0x1000);
     sysbus_init_mmio(sbd, &s->iomem);
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index b6b8a2063d..b50071ef93 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -283,6 +283,7 @@ static const VMStateDescription vmstate_hpet_rtc_irq_level = {
     .name = "hpet/rtc_irq_level",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = hpet_rtc_irq_level_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(rtc_irq_level, HPETState),
         VMSTATE_END_OF_LIST()
@@ -322,13 +323,9 @@ static const VMStateDescription vmstate_hpet = {
                                     vmstate_hpet_timer, HPETTimer),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_hpet_rtc_irq_level,
-            .needed = hpet_rtc_irq_level_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_hpet_rtc_irq_level,
+        NULL
     }
 };
 
diff --git a/hw/timer/mc146818rtc.c b/hw/timer/mc146818rtc.c
index f2b77fa118..32048258c9 100644
--- a/hw/timer/mc146818rtc.c
+++ b/hw/timer/mc146818rtc.c
@@ -733,22 +733,23 @@ static int rtc_post_load(void *opaque, int version_id)
     return 0;
 }
 
+static bool rtc_irq_reinject_on_ack_count_needed(void *opaque)
+{
+    RTCState *s = (RTCState *)opaque;
+    return s->irq_reinject_on_ack_count != 0;
+}
+
 static const VMStateDescription vmstate_rtc_irq_reinject_on_ack_count = {
     .name = "mc146818rtc/irq_reinject_on_ack_count",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = rtc_irq_reinject_on_ack_count_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT16(irq_reinject_on_ack_count, RTCState),
         VMSTATE_END_OF_LIST()
     }
 };
 
-static bool rtc_irq_reinject_on_ack_count_needed(void *opaque)
-{
-    RTCState *s = (RTCState *)opaque;
-    return s->irq_reinject_on_ack_count != 0;
-}
-
 static const VMStateDescription vmstate_rtc = {
     .name = "mc146818rtc",
     .version_id = 3,
@@ -770,13 +771,9 @@ static const VMStateDescription vmstate_rtc = {
         VMSTATE_UINT64_V(next_alarm_time, RTCState, 3),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_rtc_irq_reinject_on_ack_count,
-            .needed = rtc_irq_reinject_on_ack_count_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_rtc_irq_reinject_on_ack_count,
+        NULL
     }
 };
 
diff --git a/hw/unicore32/puv3.c b/hw/unicore32/puv3.c
index cc9a21a712..703e29d6d3 100644
--- a/hw/unicore32/puv3.c
+++ b/hw/unicore32/puv3.c
@@ -40,15 +40,15 @@ static void puv3_intc_cpu_handler(void *opaque, int irq, int level)
 
 static void puv3_soc_init(CPUUniCore32State *env)
 {
-    qemu_irq *cpu_intc, irqs[PUV3_IRQS_NR];
+    qemu_irq cpu_intc, irqs[PUV3_IRQS_NR];
     DeviceState *dev;
     MemoryRegion *i8042 = g_new(MemoryRegion, 1);
     int i;
 
     /* Initialize interrupt controller */
-    cpu_intc = qemu_allocate_irqs(puv3_intc_cpu_handler,
-                                  uc32_env_get_cpu(env), 1);
-    dev = sysbus_create_simple("puv3_intc", PUV3_INTC_BASE, *cpu_intc);
+    cpu_intc = qemu_allocate_irq(puv3_intc_cpu_handler,
+                                 uc32_env_get_cpu(env), 0);
+    dev = sysbus_create_simple("puv3_intc", PUV3_INTC_BASE, cpu_intc);
     for (i = 0; i < PUV3_IRQS_NR; i++) {
         irqs[i] = qdev_get_gpio_in(dev, i);
     }
diff --git a/hw/usb/hcd-ohci.c b/hw/usb/hcd-ohci.c
index 1a22c9c0cb..7d65818064 100644
--- a/hw/usb/hcd-ohci.c
+++ b/hw/usb/hcd-ohci.c
@@ -2034,6 +2034,7 @@ static const VMStateDescription vmstate_ohci_eof_timer = {
     .version_id = 1,
     .minimum_version_id = 1,
     .pre_load = ohci_eof_timer_pre_load,
+    .needed = ohci_eof_timer_needed,
     .fields = (VMStateField[]) {
         VMSTATE_TIMER_PTR(eof_timer, OHCIState),
         VMSTATE_END_OF_LIST()
@@ -2081,13 +2082,9 @@ static const VMStateDescription vmstate_ohci_state = {
         VMSTATE_BOOL(async_complete, OHCIState),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection []) {
-        {
-            .vmsd = &vmstate_ohci_eof_timer,
-            .needed = ohci_eof_timer_needed,
-        } , {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_ohci_eof_timer,
+        NULL
     }
 };
 
diff --git a/hw/usb/redirect.c b/hw/usb/redirect.c
index 242a654583..6b4218c037 100644
--- a/hw/usb/redirect.c
+++ b/hw/usb/redirect.c
@@ -2257,40 +2257,42 @@ static const VMStateInfo usbredir_ep_bufpq_vmstate_info = {
 
 
 /* For endp_data migration */
+static bool usbredir_bulk_receiving_needed(void *priv)
+{
+    struct endp_data *endp = priv;
+
+    return endp->bulk_receiving_started;
+}
+
 static const VMStateDescription usbredir_bulk_receiving_vmstate = {
     .name = "usb-redir-ep/bulk-receiving",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = usbredir_bulk_receiving_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(bulk_receiving_started, struct endp_data),
         VMSTATE_END_OF_LIST()
     }
 };
 
-static bool usbredir_bulk_receiving_needed(void *priv)
+static bool usbredir_stream_needed(void *priv)
 {
     struct endp_data *endp = priv;
 
-    return endp->bulk_receiving_started;
+    return endp->max_streams;
 }
 
 static const VMStateDescription usbredir_stream_vmstate = {
     .name = "usb-redir-ep/stream-state",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = usbredir_stream_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT32(max_streams, struct endp_data),
         VMSTATE_END_OF_LIST()
     }
 };
 
-static bool usbredir_stream_needed(void *priv)
-{
-    struct endp_data *endp = priv;
-
-    return endp->max_streams;
-}
-
 static const VMStateDescription usbredir_ep_vmstate = {
     .name = "usb-redir-ep",
     .version_id = 1,
@@ -2318,16 +2320,10 @@ static const VMStateDescription usbredir_ep_vmstate = {
         VMSTATE_INT32(bufpq_target_size, struct endp_data),
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &usbredir_bulk_receiving_vmstate,
-            .needed = usbredir_bulk_receiving_needed,
-        }, {
-            .vmsd = &usbredir_stream_vmstate,
-            .needed = usbredir_stream_needed,
-        }, {
-            /* empty */
-        }
+    .subsections = (const VMStateDescription*[]) {
+        &usbredir_bulk_receiving_vmstate,
+        &usbredir_stream_vmstate,
+        NULL
     }
 };
 
diff --git a/hw/vfio/Makefile.objs b/hw/vfio/Makefile.objs
index e31f30ec09..d540c9d140 100644
--- a/hw/vfio/Makefile.objs
+++ b/hw/vfio/Makefile.objs
@@ -1,4 +1,6 @@
 ifeq ($(CONFIG_LINUX), y)
 obj-$(CONFIG_SOFTMMU) += common.o
 obj-$(CONFIG_PCI) += pci.o
+obj-$(CONFIG_SOFTMMU) += platform.o
+obj-$(CONFIG_SOFTMMU) += calxeda-xgmac.o
 endif
diff --git a/hw/vfio/calxeda-xgmac.c b/hw/vfio/calxeda-xgmac.c
new file mode 100644
index 0000000000..eb914f0d0b
--- /dev/null
+++ b/hw/vfio/calxeda-xgmac.c
@@ -0,0 +1,55 @@
+/*
+ * calxeda xgmac VFIO device
+ *
+ * Copyright Linaro Limited, 2014
+ *
+ * Authors:
+ *  Eric Auger <eric.auger@linaro.org>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "hw/vfio/vfio-calxeda-xgmac.h"
+
+static void calxeda_xgmac_realize(DeviceState *dev, Error **errp)
+{
+    VFIOPlatformDevice *vdev = VFIO_PLATFORM_DEVICE(dev);
+    VFIOCalxedaXgmacDeviceClass *k = VFIO_CALXEDA_XGMAC_DEVICE_GET_CLASS(dev);
+
+    vdev->compat = g_strdup("calxeda,hb-xgmac");
+
+    k->parent_realize(dev, errp);
+}
+
+static const VMStateDescription vfio_platform_calxeda_xgmac_vmstate = {
+    .name = TYPE_VFIO_CALXEDA_XGMAC,
+    .unmigratable = 1,
+};
+
+static void vfio_calxeda_xgmac_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VFIOCalxedaXgmacDeviceClass *vcxc =
+        VFIO_CALXEDA_XGMAC_DEVICE_CLASS(klass);
+    vcxc->parent_realize = dc->realize;
+    dc->realize = calxeda_xgmac_realize;
+    dc->desc = "VFIO Calxeda XGMAC";
+    dc->vmsd = &vfio_platform_calxeda_xgmac_vmstate;
+}
+
+static const TypeInfo vfio_calxeda_xgmac_dev_info = {
+    .name = TYPE_VFIO_CALXEDA_XGMAC,
+    .parent = TYPE_VFIO_PLATFORM,
+    .instance_size = sizeof(VFIOCalxedaXgmacDevice),
+    .class_init = vfio_calxeda_xgmac_class_init,
+    .class_size = sizeof(VFIOCalxedaXgmacDeviceClass),
+};
+
+static void register_calxeda_xgmac_dev_type(void)
+{
+    type_register_static(&vfio_calxeda_xgmac_dev_info);
+}
+
+type_init(register_calxeda_xgmac_dev_type)
diff --git a/hw/vfio/platform.c b/hw/vfio/platform.c
new file mode 100644
index 0000000000..5c678b914e
--- /dev/null
+++ b/hw/vfio/platform.c
@@ -0,0 +1,615 @@
+/*
+ * vfio based device assignment support - platform devices
+ *
+ * Copyright Linaro Limited, 2014
+ *
+ * Authors:
+ *  Kim Phillips <kim.phillips@linaro.org>
+ *  Eric Auger <eric.auger@linaro.org>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ * Based on vfio based PCI device assignment support:
+ *  Copyright Red Hat, Inc. 2012
+ */
+
+#include <sys/ioctl.h>
+#include <linux/vfio.h>
+
+#include "hw/vfio/vfio-platform.h"
+#include "qemu/error-report.h"
+#include "qemu/range.h"
+#include "sysemu/sysemu.h"
+#include "exec/memory.h"
+#include "qemu/queue.h"
+#include "hw/sysbus.h"
+#include "trace.h"
+#include "hw/platform-bus.h"
+
+/*
+ * Functions used whatever the injection method
+ */
+
+/**
+ * vfio_init_intp - allocate, initialize the IRQ struct pointer
+ * and add it into the list of IRQs
+ * @vbasedev: the VFIO device handle
+ * @info: irq info struct retrieved from VFIO driver
+ */
+static VFIOINTp *vfio_init_intp(VFIODevice *vbasedev,
+                                struct vfio_irq_info info)
+{
+    int ret;
+    VFIOPlatformDevice *vdev =
+        container_of(vbasedev, VFIOPlatformDevice, vbasedev);
+    SysBusDevice *sbdev = SYS_BUS_DEVICE(vdev);
+    VFIOINTp *intp;
+
+    intp = g_malloc0(sizeof(*intp));
+    intp->vdev = vdev;
+    intp->pin = info.index;
+    intp->flags = info.flags;
+    intp->state = VFIO_IRQ_INACTIVE;
+
+    sysbus_init_irq(sbdev, &intp->qemuirq);
+
+    /* Get an eventfd for trigger */
+    ret = event_notifier_init(&intp->interrupt, 0);
+    if (ret) {
+        g_free(intp);
+        error_report("vfio: Error: trigger event_notifier_init failed ");
+        return NULL;
+    }
+
+    QLIST_INSERT_HEAD(&vdev->intp_list, intp, next);
+    return intp;
+}
+
+/**
+ * vfio_set_trigger_eventfd - set VFIO eventfd handling
+ *
+ * @intp: IRQ struct handle
+ * @handler: handler to be called on eventfd signaling
+ *
+ * Setup VFIO signaling and attach an optional user-side handler
+ * to the eventfd
+ */
+static int vfio_set_trigger_eventfd(VFIOINTp *intp,
+                                    eventfd_user_side_handler_t handler)
+{
+    VFIODevice *vbasedev = &intp->vdev->vbasedev;
+    struct vfio_irq_set *irq_set;
+    int argsz, ret;
+    int32_t *pfd;
+
+    argsz = sizeof(*irq_set) + sizeof(*pfd);
+    irq_set = g_malloc0(argsz);
+    irq_set->argsz = argsz;
+    irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD | VFIO_IRQ_SET_ACTION_TRIGGER;
+    irq_set->index = intp->pin;
+    irq_set->start = 0;
+    irq_set->count = 1;
+    pfd = (int32_t *)&irq_set->data;
+    *pfd = event_notifier_get_fd(&intp->interrupt);
+    qemu_set_fd_handler(*pfd, (IOHandler *)handler, NULL, intp);
+    ret = ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, irq_set);
+    g_free(irq_set);
+    if (ret < 0) {
+        error_report("vfio: Failed to set trigger eventfd: %m");
+        qemu_set_fd_handler(*pfd, NULL, NULL, NULL);
+    }
+    return ret;
+}
+
+/*
+ * Functions only used when eventfds are handled on user-side
+ * ie. without irqfd
+ */
+
+/**
+ * vfio_mmap_set_enabled - enable/disable the fast path mode
+ * @vdev: the VFIO platform device
+ * @enabled: the target mmap state
+ *
+ * enabled = true ~ fast path = MMIO region is mmaped (no KVM TRAP);
+ * enabled = false ~ slow path = MMIO region is trapped and region callbacks
+ * are called; slow path enables to trap the device IRQ status register reset
+*/
+
+static void vfio_mmap_set_enabled(VFIOPlatformDevice *vdev, bool enabled)
+{
+    int i;
+
+    trace_vfio_platform_mmap_set_enabled(enabled);
+
+    for (i = 0; i < vdev->vbasedev.num_regions; i++) {
+        VFIORegion *region = vdev->regions[i];
+
+        memory_region_set_enabled(&region->mmap_mem, enabled);
+    }
+}
+
+/**
+ * vfio_intp_mmap_enable - timer function, restores the fast path
+ * if there is no more active IRQ
+ * @opaque: actually points to the VFIO platform device
+ *
+ * Called on mmap timer timout, this function checks whether the
+ * IRQ is still active and if not, restores the fast path.
+ * by construction a single eventfd is handled at a time.
+ * if the IRQ is still active, the timer is re-programmed.
+ */
+static void vfio_intp_mmap_enable(void *opaque)
+{
+    VFIOINTp *tmp;
+    VFIOPlatformDevice *vdev = (VFIOPlatformDevice *)opaque;
+
+    qemu_mutex_lock(&vdev->intp_mutex);
+    QLIST_FOREACH(tmp, &vdev->intp_list, next) {
+        if (tmp->state == VFIO_IRQ_ACTIVE) {
+            trace_vfio_platform_intp_mmap_enable(tmp->pin);
+            /* re-program the timer to check active status later */
+            timer_mod(vdev->mmap_timer,
+                      qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) +
+                          vdev->mmap_timeout);
+            qemu_mutex_unlock(&vdev->intp_mutex);
+            return;
+        }
+    }
+    vfio_mmap_set_enabled(vdev, true);
+    qemu_mutex_unlock(&vdev->intp_mutex);
+}
+
+/**
+ * vfio_intp_inject_pending_lockheld - Injects a pending IRQ
+ * @opaque: opaque pointer, in practice the VFIOINTp handle
+ *
+ * The function is called on a previous IRQ completion, from
+ * vfio_platform_eoi, while the intp_mutex is locked.
+ * Also in such situation, the slow path already is set and
+ * the mmap timer was already programmed.
+ */
+static void vfio_intp_inject_pending_lockheld(VFIOINTp *intp)
+{
+    trace_vfio_platform_intp_inject_pending_lockheld(intp->pin,
+                              event_notifier_get_fd(&intp->interrupt));
+
+    intp->state = VFIO_IRQ_ACTIVE;
+
+    /* trigger the virtual IRQ */
+    qemu_set_irq(intp->qemuirq, 1);
+}
+
+/**
+ * vfio_intp_interrupt - The user-side eventfd handler
+ * @opaque: opaque pointer which in practice is the VFIOINTp handle
+ *
+ * the function is entered in event handler context:
+ * the vIRQ is injected into the guest if there is no other active
+ * or pending IRQ.
+ */
+static void vfio_intp_interrupt(VFIOINTp *intp)
+{
+    int ret;
+    VFIOINTp *tmp;
+    VFIOPlatformDevice *vdev = intp->vdev;
+    bool delay_handling = false;
+
+    qemu_mutex_lock(&vdev->intp_mutex);
+    if (intp->state == VFIO_IRQ_INACTIVE) {
+        QLIST_FOREACH(tmp, &vdev->intp_list, next) {
+            if (tmp->state == VFIO_IRQ_ACTIVE ||
+                tmp->state == VFIO_IRQ_PENDING) {
+                delay_handling = true;
+                break;
+            }
+        }
+    }
+    if (delay_handling) {
+        /*
+         * the new IRQ gets a pending status and is pushed in
+         * the pending queue
+         */
+        intp->state = VFIO_IRQ_PENDING;
+        trace_vfio_intp_interrupt_set_pending(intp->pin);
+        QSIMPLEQ_INSERT_TAIL(&vdev->pending_intp_queue,
+                             intp, pqnext);
+        ret = event_notifier_test_and_clear(&intp->interrupt);
+        qemu_mutex_unlock(&vdev->intp_mutex);
+        return;
+    }
+
+    trace_vfio_platform_intp_interrupt(intp->pin,
+                              event_notifier_get_fd(&intp->interrupt));
+
+    ret = event_notifier_test_and_clear(&intp->interrupt);
+    if (!ret) {
+        error_report("Error when clearing fd=%d (ret = %d)\n",
+                     event_notifier_get_fd(&intp->interrupt), ret);
+    }
+
+    intp->state = VFIO_IRQ_ACTIVE;
+
+    /* sets slow path */
+    vfio_mmap_set_enabled(vdev, false);
+
+    /* trigger the virtual IRQ */
+    qemu_set_irq(intp->qemuirq, 1);
+
+    /*
+     * Schedule the mmap timer which will restore fastpath when no IRQ
+     * is active anymore
+     */
+    if (vdev->mmap_timeout) {
+        timer_mod(vdev->mmap_timer,
+                  qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) +
+                      vdev->mmap_timeout);
+    }
+    qemu_mutex_unlock(&vdev->intp_mutex);
+}
+
+/**
+ * vfio_platform_eoi - IRQ completion routine
+ * @vbasedev: the VFIO device handle
+ *
+ * De-asserts the active virtual IRQ and unmasks the physical IRQ
+ * (effective for level sensitive IRQ auto-masked by the  VFIO driver).
+ * Then it handles next pending IRQ if any.
+ * eoi function is called on the first access to any MMIO region
+ * after an IRQ was triggered, trapped since slow path was set.
+ * It is assumed this access corresponds to the IRQ status
+ * register reset. With such a mechanism, a single IRQ can be
+ * handled at a time since there is no way to know which IRQ
+ * was completed by the guest (we would need additional details
+ * about the IRQ status register mask).
+ */
+static void vfio_platform_eoi(VFIODevice *vbasedev)
+{
+    VFIOINTp *intp;
+    VFIOPlatformDevice *vdev =
+        container_of(vbasedev, VFIOPlatformDevice, vbasedev);
+
+    qemu_mutex_lock(&vdev->intp_mutex);
+    QLIST_FOREACH(intp, &vdev->intp_list, next) {
+        if (intp->state == VFIO_IRQ_ACTIVE) {
+            trace_vfio_platform_eoi(intp->pin,
+                                event_notifier_get_fd(&intp->interrupt));
+            intp->state = VFIO_IRQ_INACTIVE;
+
+            /* deassert the virtual IRQ */
+            qemu_set_irq(intp->qemuirq, 0);
+
+            if (intp->flags & VFIO_IRQ_INFO_AUTOMASKED) {
+                /* unmasks the physical level-sensitive IRQ */
+                vfio_unmask_single_irqindex(vbasedev, intp->pin);
+            }
+
+            /* a single IRQ can be active at a time */
+            break;
+        }
+    }
+    /* in case there are pending IRQs, handle the first one */
+    if (!QSIMPLEQ_EMPTY(&vdev->pending_intp_queue)) {
+        intp = QSIMPLEQ_FIRST(&vdev->pending_intp_queue);
+        vfio_intp_inject_pending_lockheld(intp);
+        QSIMPLEQ_REMOVE_HEAD(&vdev->pending_intp_queue, pqnext);
+    }
+    qemu_mutex_unlock(&vdev->intp_mutex);
+}
+
+/**
+ * vfio_start_eventfd_injection - starts the virtual IRQ injection using
+ * user-side handled eventfds
+ * @intp: the IRQ struct pointer
+ */
+
+static int vfio_start_eventfd_injection(VFIOINTp *intp)
+{
+    int ret;
+
+    ret = vfio_set_trigger_eventfd(intp, vfio_intp_interrupt);
+    if (ret) {
+        error_report("vfio: Error: Failed to pass IRQ fd to the driver: %m");
+    }
+    return ret;
+}
+
+/* VFIO skeleton */
+
+static void vfio_platform_compute_needs_reset(VFIODevice *vbasedev)
+{
+    vbasedev->needs_reset = true;
+}
+
+/* not implemented yet */
+static int vfio_platform_hot_reset_multi(VFIODevice *vbasedev)
+{
+    return -1;
+}
+
+/**
+ * vfio_populate_device - Allocate and populate MMIO region
+ * and IRQ structs according to driver returned information
+ * @vbasedev: the VFIO device handle
+ *
+ */
+static int vfio_populate_device(VFIODevice *vbasedev)
+{
+    VFIOINTp *intp, *tmp;
+    int i, ret = -1;
+    VFIOPlatformDevice *vdev =
+        container_of(vbasedev, VFIOPlatformDevice, vbasedev);
+
+    if (!(vbasedev->flags & VFIO_DEVICE_FLAGS_PLATFORM)) {
+        error_report("vfio: Um, this isn't a platform device");
+        return ret;
+    }
+
+    vdev->regions = g_new0(VFIORegion *, vbasedev->num_regions);
+
+    for (i = 0; i < vbasedev->num_regions; i++) {
+        struct vfio_region_info reg_info = { .argsz = sizeof(reg_info) };
+        VFIORegion *ptr;
+
+        vdev->regions[i] = g_malloc0(sizeof(VFIORegion));
+        ptr = vdev->regions[i];
+        reg_info.index = i;
+        ret = ioctl(vbasedev->fd, VFIO_DEVICE_GET_REGION_INFO, &reg_info);
+        if (ret) {
+            error_report("vfio: Error getting region %d info: %m", i);
+            goto reg_error;
+        }
+        ptr->flags = reg_info.flags;
+        ptr->size = reg_info.size;
+        ptr->fd_offset = reg_info.offset;
+        ptr->nr = i;
+        ptr->vbasedev = vbasedev;
+
+        trace_vfio_platform_populate_regions(ptr->nr,
+                            (unsigned long)ptr->flags,
+                            (unsigned long)ptr->size,
+                            ptr->vbasedev->fd,
+                            (unsigned long)ptr->fd_offset);
+    }
+
+    vdev->mmap_timer = timer_new_ms(QEMU_CLOCK_VIRTUAL,
+                                    vfio_intp_mmap_enable, vdev);
+
+    QSIMPLEQ_INIT(&vdev->pending_intp_queue);
+
+    for (i = 0; i < vbasedev->num_irqs; i++) {
+        struct vfio_irq_info irq = { .argsz = sizeof(irq) };
+
+        irq.index = i;
+        ret = ioctl(vbasedev->fd, VFIO_DEVICE_GET_IRQ_INFO, &irq);
+        if (ret) {
+            error_printf("vfio: error getting device %s irq info",
+                         vbasedev->name);
+            goto irq_err;
+        } else {
+            trace_vfio_platform_populate_interrupts(irq.index,
+                                                    irq.count,
+                                                    irq.flags);
+            intp = vfio_init_intp(vbasedev, irq);
+            if (!intp) {
+                error_report("vfio: Error installing IRQ %d up", i);
+                goto irq_err;
+            }
+        }
+    }
+    return 0;
+irq_err:
+    timer_del(vdev->mmap_timer);
+    QLIST_FOREACH_SAFE(intp, &vdev->intp_list, next, tmp) {
+        QLIST_REMOVE(intp, next);
+        g_free(intp);
+    }
+reg_error:
+    for (i = 0; i < vbasedev->num_regions; i++) {
+        g_free(vdev->regions[i]);
+    }
+    g_free(vdev->regions);
+    return ret;
+}
+
+/* specialized functions for VFIO Platform devices */
+static VFIODeviceOps vfio_platform_ops = {
+    .vfio_compute_needs_reset = vfio_platform_compute_needs_reset,
+    .vfio_hot_reset_multi = vfio_platform_hot_reset_multi,
+    .vfio_eoi = vfio_platform_eoi,
+};
+
+/**
+ * vfio_base_device_init - perform preliminary VFIO setup
+ * @vbasedev: the VFIO device handle
+ *
+ * Implement the VFIO command sequence that allows to discover
+ * assigned device resources: group extraction, device
+ * fd retrieval, resource query.
+ * Precondition: the device name must be initialized
+ */
+static int vfio_base_device_init(VFIODevice *vbasedev)
+{
+    VFIOGroup *group;
+    VFIODevice *vbasedev_iter;
+    char path[PATH_MAX], iommu_group_path[PATH_MAX], *group_name;
+    ssize_t len;
+    struct stat st;
+    int groupid;
+    int ret;
+
+    /* name must be set prior to the call */
+    if (!vbasedev->name || strchr(vbasedev->name, '/')) {
+        return -EINVAL;
+    }
+
+    /* Check that the host device exists */
+    g_snprintf(path, sizeof(path), "/sys/bus/platform/devices/%s/",
+               vbasedev->name);
+
+    if (stat(path, &st) < 0) {
+        error_report("vfio: error: no such host device: %s", path);
+        return -errno;
+    }
+
+    g_strlcat(path, "iommu_group", sizeof(path));
+    len = readlink(path, iommu_group_path, sizeof(iommu_group_path));
+    if (len < 0 || len >= sizeof(iommu_group_path)) {
+        error_report("vfio: error no iommu_group for device");
+        return len < 0 ? -errno : -ENAMETOOLONG;
+    }
+
+    iommu_group_path[len] = 0;
+    group_name = basename(iommu_group_path);
+
+    if (sscanf(group_name, "%d", &groupid) != 1) {
+        error_report("vfio: error reading %s: %m", path);
+        return -errno;
+    }
+
+    trace_vfio_platform_base_device_init(vbasedev->name, groupid);
+
+    group = vfio_get_group(groupid, &address_space_memory);
+    if (!group) {
+        error_report("vfio: failed to get group %d", groupid);
+        return -ENOENT;
+    }
+
+    g_snprintf(path, sizeof(path), "%s", vbasedev->name);
+
+    QLIST_FOREACH(vbasedev_iter, &group->device_list, next) {
+        if (strcmp(vbasedev_iter->name, vbasedev->name) == 0) {
+            error_report("vfio: error: device %s is already attached", path);
+            vfio_put_group(group);
+            return -EBUSY;
+        }
+    }
+    ret = vfio_get_device(group, path, vbasedev);
+    if (ret) {
+        error_report("vfio: failed to get device %s", path);
+        vfio_put_group(group);
+        return ret;
+    }
+
+    ret = vfio_populate_device(vbasedev);
+    if (ret) {
+        error_report("vfio: failed to populate device %s", path);
+        vfio_put_group(group);
+    }
+
+    return ret;
+}
+
+/**
+ * vfio_map_region - initialize the 2 memory regions for a given
+ * MMIO region index
+ * @vdev: the VFIO platform device handle
+ * @nr: the index of the region
+ *
+ * Init the top memory region and the mmapped memory region beneath
+ * VFIOPlatformDevice is used since VFIODevice is not a QOM Object
+ * and could not be passed to memory region functions
+*/
+static void vfio_map_region(VFIOPlatformDevice *vdev, int nr)
+{
+    VFIORegion *region = vdev->regions[nr];
+    uint64_t size = region->size;
+    char name[64];
+
+    if (!size) {
+        return;
+    }
+
+    g_snprintf(name, sizeof(name), "VFIO %s region %d",
+               vdev->vbasedev.name, nr);
+
+    /* A "slow" read/write mapping underlies all regions */
+    memory_region_init_io(&region->mem, OBJECT(vdev), &vfio_region_ops,
+                          region, name, size);
+
+    g_strlcat(name, " mmap", sizeof(name));
+
+    if (vfio_mmap_region(OBJECT(vdev), region, &region->mem,
+                         &region->mmap_mem, &region->mmap, size, 0, name)) {
+        error_report("%s unsupported. Performance may be slow", name);
+    }
+}
+
+/**
+ * vfio_platform_realize  - the device realize function
+ * @dev: device state pointer
+ * @errp: error
+ *
+ * initialize the device, its memory regions and IRQ structures
+ * IRQ are started separately
+ */
+static void vfio_platform_realize(DeviceState *dev, Error **errp)
+{
+    VFIOPlatformDevice *vdev = VFIO_PLATFORM_DEVICE(dev);
+    SysBusDevice *sbdev = SYS_BUS_DEVICE(dev);
+    VFIODevice *vbasedev = &vdev->vbasedev;
+    VFIOINTp *intp;
+    int i, ret;
+
+    vbasedev->type = VFIO_DEVICE_TYPE_PLATFORM;
+    vbasedev->ops = &vfio_platform_ops;
+
+    trace_vfio_platform_realize(vbasedev->name, vdev->compat);
+
+    ret = vfio_base_device_init(vbasedev);
+    if (ret) {
+        error_setg(errp, "vfio: vfio_base_device_init failed for %s",
+                   vbasedev->name);
+        return;
+    }
+
+    for (i = 0; i < vbasedev->num_regions; i++) {
+        vfio_map_region(vdev, i);
+        sysbus_init_mmio(sbdev, &vdev->regions[i]->mem);
+    }
+
+    QLIST_FOREACH(intp, &vdev->intp_list, next) {
+        vfio_start_eventfd_injection(intp);
+    }
+}
+
+static const VMStateDescription vfio_platform_vmstate = {
+    .name = TYPE_VFIO_PLATFORM,
+    .unmigratable = 1,
+};
+
+static Property vfio_platform_dev_properties[] = {
+    DEFINE_PROP_STRING("host", VFIOPlatformDevice, vbasedev.name),
+    DEFINE_PROP_BOOL("x-mmap", VFIOPlatformDevice, vbasedev.allow_mmap, true),
+    DEFINE_PROP_UINT32("mmap-timeout-ms", VFIOPlatformDevice,
+                       mmap_timeout, 1100),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void vfio_platform_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+
+    dc->realize = vfio_platform_realize;
+    dc->props = vfio_platform_dev_properties;
+    dc->vmsd = &vfio_platform_vmstate;
+    dc->desc = "VFIO-based platform device assignment";
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+}
+
+static const TypeInfo vfio_platform_dev_info = {
+    .name = TYPE_VFIO_PLATFORM,
+    .parent = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(VFIOPlatformDevice),
+    .class_init = vfio_platform_class_init,
+    .class_size = sizeof(VFIOPlatformDeviceClass),
+    .abstract   = true,
+};
+
+static void register_vfio_platform_dev_type(void)
+{
+    type_register_static(&vfio_platform_dev_info);
+}
+
+type_init(register_vfio_platform_dev_type)
diff --git a/hw/virtio/dataplane/vring.c b/hw/virtio/dataplane/vring.c
index fabb8103e9..35891856ee 100644
--- a/hw/virtio/dataplane/vring.c
+++ b/hw/virtio/dataplane/vring.c
@@ -42,7 +42,7 @@ static void *vring_map(MemoryRegion **mr, hwaddr phys, hwaddr len,
     }
 
     /* Ignore regions with dirty logging, we cannot mark them dirty */
-    if (memory_region_is_logging(section.mr)) {
+    if (memory_region_get_dirty_log_mask(section.mr)) {
         goto out;
     }
 
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index b84d21cf7e..a6dcc79399 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -418,7 +418,8 @@ static void vhost_set_memory(MemoryListener *listener,
                                          memory_listener);
     hwaddr start_addr = section->offset_within_address_space;
     ram_addr_t size = int128_get64(section->size);
-    bool log_dirty = memory_region_is_logging(section->mr);
+    bool log_dirty =
+        memory_region_get_dirty_log_mask(section->mr) & ~(1 << DIRTY_MEMORY_MIGRATION);
     int s = offsetof(struct vhost_memory, regions) +
         (dev->mem->nregions + 1) * sizeof dev->mem->regions[0];
     void *ram;
@@ -677,13 +678,15 @@ static void vhost_log_global_stop(MemoryListener *listener)
 }
 
 static void vhost_log_start(MemoryListener *listener,
-                            MemoryRegionSection *section)
+                            MemoryRegionSection *section,
+                            int old, int new)
 {
     /* FIXME: implement */
 }
 
 static void vhost_log_stop(MemoryListener *listener,
-                           MemoryRegionSection *section)
+                           MemoryRegionSection *section,
+                           int old, int new)
 {
     /* FIXME: implement */
 }
diff --git a/hw/virtio/virtio-pci.h b/hw/virtio/virtio-pci.h
index d962125351..96025ca205 100644
--- a/hw/virtio/virtio-pci.h
+++ b/hw/virtio/virtio-pci.h
@@ -25,6 +25,7 @@
 #include "hw/virtio/virtio-bus.h"
 #include "hw/virtio/virtio-9p.h"
 #include "hw/virtio/virtio-input.h"
+#include "hw/virtio/virtio-gpu.h"
 #ifdef CONFIG_VIRTFS
 #include "hw/9pfs/virtio-9p.h"
 #endif
@@ -42,6 +43,7 @@ typedef struct VHostSCSIPCI VHostSCSIPCI;
 typedef struct VirtIORngPCI VirtIORngPCI;
 typedef struct VirtIOInputPCI VirtIOInputPCI;
 typedef struct VirtIOInputHIDPCI VirtIOInputHIDPCI;
+typedef struct VirtIOGPUPCI VirtIOGPUPCI;
 
 /* virtio-pci-bus */
 
@@ -261,6 +263,18 @@ struct VirtIOInputHIDPCI {
     VirtIOInputHID vdev;
 };
 
+/*
+ * virtio-gpu-pci: This extends VirtioPCIProxy.
+ */
+#define TYPE_VIRTIO_GPU_PCI "virtio-gpu-pci"
+#define VIRTIO_GPU_PCI(obj) \
+        OBJECT_CHECK(VirtIOGPUPCI, (obj), TYPE_VIRTIO_GPU_PCI)
+
+struct VirtIOGPUPCI {
+    VirtIOPCIProxy parent_obj;
+    VirtIOGPU vdev;
+};
+
 /* Virtio ABI version, if we increment this, we break the guest driver. */
 #define VIRTIO_PCI_ABI_VERSION          0
 
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index fb49ffcb2d..ee4e07c5e7 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1053,6 +1053,7 @@ static const VMStateDescription vmstate_virtio_device_endian = {
     .name = "virtio/device_endian",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = &virtio_device_endian_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT8(device_endian, VirtIODevice),
         VMSTATE_END_OF_LIST()
@@ -1063,6 +1064,7 @@ static const VMStateDescription vmstate_virtio_64bit_features = {
     .name = "virtio/64bit_features",
     .version_id = 1,
     .minimum_version_id = 1,
+    .needed = &virtio_64bit_features_needed,
     .fields = (VMStateField[]) {
         VMSTATE_UINT64(guest_features, VirtIODevice),
         VMSTATE_END_OF_LIST()
@@ -1077,16 +1079,10 @@ static const VMStateDescription vmstate_virtio = {
     .fields = (VMStateField[]) {
         VMSTATE_END_OF_LIST()
     },
-    .subsections = (VMStateSubsection[]) {
-        {
-            .vmsd = &vmstate_virtio_device_endian,
-            .needed = &virtio_device_endian_needed
-        },
-        {
-            .vmsd = &vmstate_virtio_64bit_features,
-            .needed = &virtio_64bit_features_needed
-        },
-        { 0 }
+    .subsections = (const VMStateDescription*[]) {
+        &vmstate_virtio_device_endian,
+        &vmstate_virtio_64bit_features,
+        NULL
     }
 };
 
diff --git a/hw/watchdog/Makefile.objs b/hw/watchdog/Makefile.objs
index 4b0374a555..72e3ffd93c 100644
--- a/hw/watchdog/Makefile.objs
+++ b/hw/watchdog/Makefile.objs
@@ -1,3 +1,4 @@
 common-obj-y += watchdog.o
 common-obj-$(CONFIG_WDT_IB6300ESB) += wdt_i6300esb.o
 common-obj-$(CONFIG_WDT_IB700) += wdt_ib700.o
+common-obj-$(CONFIG_WDT_DIAG288) += wdt_diag288.o
diff --git a/hw/watchdog/watchdog.c b/hw/watchdog/watchdog.c
index 54440c91c5..8d4b0eeeb0 100644
--- a/hw/watchdog/watchdog.c
+++ b/hw/watchdog/watchdog.c
@@ -27,6 +27,7 @@
 #include "sysemu/sysemu.h"
 #include "sysemu/watchdog.h"
 #include "qapi-event.h"
+#include "hw/nmi.h"
 
 /* Possible values for action parameter. */
 #define WDT_RESET        1	/* Hard reset. */
@@ -35,6 +36,7 @@
 #define WDT_PAUSE        4	/* Pause. */
 #define WDT_DEBUG        5	/* Prints a message and continues running. */
 #define WDT_NONE         6	/* Do nothing. */
+#define WDT_NMI          7	/* Inject nmi into the guest */
 
 static int watchdog_action = WDT_RESET;
 static QLIST_HEAD(watchdog_list, WatchdogTimerModel) watchdog_list;
@@ -95,6 +97,8 @@ int select_watchdog_action(const char *p)
         watchdog_action = WDT_DEBUG;
     else if (strcasecmp(p, "none") == 0)
         watchdog_action = WDT_NONE;
+    else if (strcasecmp(p, "inject-nmi") == 0)
+        watchdog_action = WDT_NMI;
     else
         return -1;
 
@@ -138,5 +142,11 @@ void watchdog_perform_action(void)
     case WDT_NONE:
         qapi_event_send_watchdog(WATCHDOG_EXPIRATION_ACTION_NONE, &error_abort);
         break;
+
+    case WDT_NMI:
+        qapi_event_send_watchdog(WATCHDOG_EXPIRATION_ACTION_INJECT_NMI,
+                                 &error_abort);
+        inject_nmi();
+        break;
     }
 }
diff --git a/hw/watchdog/wdt_diag288.c b/hw/watchdog/wdt_diag288.c
new file mode 100644
index 0000000000..1185e0681c
--- /dev/null
+++ b/hw/watchdog/wdt_diag288.c
@@ -0,0 +1,122 @@
+/*
+ * watchdog device diag288 support
+ *
+ * Copyright IBM, Corp. 2015
+ *
+ * Authors:
+ *  Xu Wang <gesaint@linux.vnet.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at your
+ * option) any later version.  See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "sysemu/watchdog.h"
+#include "hw/sysbus.h"
+#include "qemu/timer.h"
+#include "hw/watchdog/wdt_diag288.h"
+
+static WatchdogTimerModel model = {
+    .wdt_name = TYPE_WDT_DIAG288,
+    .wdt_description = "diag288 device for s390x platform",
+};
+
+static const VMStateDescription vmstate_diag288 = {
+    .name = "vmstate_diag288",
+    .version_id = 0,
+    .minimum_version_id = 0,
+    .fields = (VMStateField[]) {
+        VMSTATE_TIMER_PTR(timer, DIAG288State),
+        VMSTATE_BOOL(enabled, DIAG288State),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
+static void wdt_diag288_reset(DeviceState *dev)
+{
+    DIAG288State *diag288 = DIAG288(dev);
+
+    diag288->enabled = false;
+    timer_del(diag288->timer);
+}
+
+static void diag288_timer_expired(void *dev)
+{
+    qemu_log_mask(CPU_LOG_RESET, "Watchdog timer expired.\n");
+    watchdog_perform_action();
+    wdt_diag288_reset(dev);
+}
+
+static int wdt_diag288_handle_timer(DIAG288State *diag288,
+                                     uint64_t func, uint64_t timeout)
+{
+    switch (func) {
+    case WDT_DIAG288_INIT:
+        diag288->enabled = true;
+        /* fall through */
+    case WDT_DIAG288_CHANGE:
+        if (!diag288->enabled) {
+            return -1;
+        }
+        timer_mod(diag288->timer,
+                  qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
+                  timeout * get_ticks_per_sec());
+        break;
+    case WDT_DIAG288_CANCEL:
+        if (!diag288->enabled) {
+            return -1;
+        }
+        diag288->enabled = false;
+        timer_del(diag288->timer);
+        break;
+    default:
+        return -1;
+    }
+
+    return 0;
+}
+
+static void wdt_diag288_realize(DeviceState *dev, Error **errp)
+{
+    DIAG288State *diag288 = DIAG288(dev);
+
+    diag288->timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, diag288_timer_expired,
+                                  dev);
+}
+
+static void wdt_diag288_unrealize(DeviceState *dev, Error **errp)
+{
+    DIAG288State *diag288 = DIAG288(dev);
+
+    timer_del(diag288->timer);
+    timer_free(diag288->timer);
+}
+
+static void wdt_diag288_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    DIAG288Class *diag288 = DIAG288_CLASS(klass);
+
+    dc->realize = wdt_diag288_realize;
+    dc->unrealize = wdt_diag288_unrealize;
+    dc->reset = wdt_diag288_reset;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    dc->vmsd = &vmstate_diag288;
+    diag288->handle_timer = wdt_diag288_handle_timer;
+}
+
+static const TypeInfo wdt_diag288_info = {
+    .class_init = wdt_diag288_class_init,
+    .parent = TYPE_DEVICE,
+    .name  = TYPE_WDT_DIAG288,
+    .instance_size  = sizeof(DIAG288State),
+    .class_size = sizeof(DIAG288Class),
+};
+
+static void wdt_diag288_register_types(void)
+{
+    watchdog_add_model(&model);
+    type_register_static(&wdt_diag288_info);
+}
+
+type_init(wdt_diag288_register_types)
diff --git a/hw/xen/xen_backend.c b/hw/xen/xen_backend.c
index b2cb22b99d..2510e2e4ff 100644
--- a/hw/xen/xen_backend.c
+++ b/hw/xen/xen_backend.c
@@ -714,9 +714,7 @@ int xen_be_init(void)
         return -1;
     }
 
-    if (qemu_set_fd_handler(xs_fileno(xenstore), xenstore_update, NULL, NULL) < 0) {
-        goto err;
-    }
+    qemu_set_fd_handler(xs_fileno(xenstore), xenstore_update, NULL, NULL);
 
     if (xen_xc == XC_HANDLER_INITIAL_VALUE) {
         /* Check if xen_init() have been called */
diff --git a/hw/xen/xen_pt.c b/hw/xen/xen_pt.c
index d095c081cc..9afcda8e21 100644
--- a/hw/xen/xen_pt.c
+++ b/hw/xen/xen_pt.c
@@ -234,11 +234,12 @@ static void xen_pt_pci_write_config(PCIDevice *d, uint32_t addr,
     int index = 0;
     XenPTRegGroup *reg_grp_entry = NULL;
     int rc = 0;
-    uint32_t read_val = 0;
+    uint32_t read_val = 0, wb_mask;
     int emul_len = 0;
     XenPTReg *reg_entry = NULL;
     uint32_t find_addr = addr;
     XenPTRegInfo *reg = NULL;
+    bool wp_flag = false;
 
     if (xen_pt_pci_config_access_check(d, addr, len)) {
         return;
@@ -271,10 +272,17 @@ static void xen_pt_pci_write_config(PCIDevice *d, uint32_t addr,
     if (rc < 0) {
         XEN_PT_ERR(d, "pci_read_block failed. return value: %d.\n", rc);
         memset(&read_val, 0xff, len);
+        wb_mask = 0;
+    } else {
+        wb_mask = 0xFFFFFFFF >> ((4 - len) << 3);
     }
 
     /* pass directly to the real device for passthrough type register group */
     if (reg_grp_entry == NULL) {
+        if (!s->permissive) {
+            wb_mask = 0;
+            wp_flag = true;
+        }
         goto out;
     }
 
@@ -295,9 +303,17 @@ static void xen_pt_pci_write_config(PCIDevice *d, uint32_t addr,
             uint32_t real_offset = reg_grp_entry->base_offset + reg->offset;
             uint32_t valid_mask = 0xFFFFFFFF >> ((4 - emul_len) << 3);
             uint8_t *ptr_val = NULL;
+            uint32_t wp_mask = reg->emu_mask | reg->ro_mask;
 
             valid_mask <<= (find_addr - real_offset) << 3;
             ptr_val = (uint8_t *)&val + (real_offset & 3);
+            if (!s->permissive) {
+                wp_mask |= reg->res_mask;
+            }
+            if (wp_mask == (0xFFFFFFFF >> ((4 - reg->size) << 3))) {
+                wb_mask &= ~((wp_mask >> ((find_addr - real_offset) << 3))
+                             << ((len - emul_len) << 3));
+            }
 
             /* do emulation based on register size */
             switch (reg->size) {
@@ -339,6 +355,16 @@ static void xen_pt_pci_write_config(PCIDevice *d, uint32_t addr,
         } else {
             /* nothing to do with passthrough type register,
              * continue to find next byte */
+            if (!s->permissive) {
+                wb_mask &= ~(0xff << ((len - emul_len) << 3));
+                /* Unused BARs will make it here, but we don't want to issue
+                 * warnings for writes to them (bogus writes get dealt with
+                 * above).
+                 */
+                if (index < 0) {
+                    wp_flag = true;
+                }
+            }
             emul_len--;
             find_addr++;
         }
@@ -350,10 +376,26 @@ static void xen_pt_pci_write_config(PCIDevice *d, uint32_t addr,
     memory_region_transaction_commit();
 
 out:
-    if (!(reg && reg->no_wb)) {
+    if (wp_flag && !s->permissive_warned) {
+        s->permissive_warned = true;
+        xen_pt_log(d, "Write-back to unknown field 0x%02x (partially) inhibited (0x%0*x)\n",
+                   addr, len * 2, wb_mask);
+        xen_pt_log(d, "If the device doesn't work, try enabling permissive mode\n");
+        xen_pt_log(d, "(unsafe) and if it helps report the problem to xen-devel\n");
+    }
+    for (index = 0; wb_mask; index += len) {
         /* unknown regs are passed through */
-        rc = xen_host_pci_set_block(&s->real_device, addr,
-                                    (uint8_t *)&val, len);
+        while (!(wb_mask & 0xff)) {
+            index++;
+            wb_mask >>= 8;
+        }
+        len = 0;
+        do {
+            len++;
+            wb_mask >>= 8;
+        } while (wb_mask & 0xff);
+        rc = xen_host_pci_set_block(&s->real_device, addr + index,
+                                    (uint8_t *)&val + index, len);
 
         if (rc < 0) {
             XEN_PT_ERR(d, "pci_write_block failed. return value: %d.\n", rc);
@@ -807,6 +849,7 @@ static void xen_pt_unregister_device(PCIDevice *d)
 
 static Property xen_pci_passthrough_properties[] = {
     DEFINE_PROP_PCI_HOST_DEVADDR("hostaddr", XenPCIPassthroughState, hostaddr),
+    DEFINE_PROP_BOOL("permissive", XenPCIPassthroughState, permissive, false),
     DEFINE_PROP_END_OF_LIST(),
 };
 
diff --git a/hw/xen/xen_pt.h b/hw/xen/xen_pt.h
index 942dc60cc7..4bba559763 100644
--- a/hw/xen/xen_pt.h
+++ b/hw/xen/xen_pt.h
@@ -101,12 +101,12 @@ struct XenPTRegInfo {
     uint32_t offset;
     uint32_t size;
     uint32_t init_val;
+    /* reg reserved field mask (ON:reserved, OFF:defined) */
+    uint32_t res_mask;
     /* reg read only field mask (ON:RO/ROS, OFF:other) */
     uint32_t ro_mask;
     /* reg emulate field mask (ON:emu, OFF:passthrough) */
     uint32_t emu_mask;
-    /* no write back allowed */
-    uint32_t no_wb;
     xen_pt_conf_reg_init init;
     /* read/write function pointer
      * for double_word/word/byte size */
@@ -177,6 +177,7 @@ typedef struct XenPTMSIXEntry {
     uint32_t data;
     uint32_t vector_ctrl;
     bool updated; /* indicate whether MSI ADDR or DATA is updated */
+    bool warned;  /* avoid issuing (bogus) warning more than once */
 } XenPTMSIXEntry;
 typedef struct XenPTMSIX {
     uint32_t ctrl_offset;
@@ -196,6 +197,8 @@ struct XenPCIPassthroughState {
 
     PCIHostDeviceAddress hostaddr;
     bool is_virtfn;
+    bool permissive;
+    bool permissive_warned;
     XenHostPCIDevice real_device;
     XenPTRegion bases[PCI_NUM_REGIONS]; /* Access regions */
     QLIST_HEAD(, XenPTRegGroup) reg_grps;
diff --git a/hw/xen/xen_pt_config_init.c b/hw/xen/xen_pt_config_init.c
index 95a51dbbb1..f3cf069b60 100644
--- a/hw/xen/xen_pt_config_init.c
+++ b/hw/xen/xen_pt_config_init.c
@@ -95,6 +95,18 @@ XenPTReg *xen_pt_find_reg(XenPTRegGroup *reg_grp, uint32_t address)
     return NULL;
 }
 
+static uint32_t get_throughable_mask(const XenPCIPassthroughState *s,
+                                     const XenPTRegInfo *reg,
+                                     uint32_t valid_mask)
+{
+    uint32_t throughable_mask = ~(reg->emu_mask | reg->ro_mask);
+
+    if (!s->permissive) {
+        throughable_mask &= ~reg->res_mask;
+    }
+
+    return throughable_mask & valid_mask;
+}
 
 /****************
  * general register functions
@@ -157,14 +169,13 @@ static int xen_pt_byte_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint8_t writable_mask = 0;
-    uint8_t throughable_mask = 0;
+    uint8_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
 
     /* modify emulate register */
     writable_mask = reg->emu_mask & ~reg->ro_mask & valid_mask;
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
     *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
 
     return 0;
@@ -175,14 +186,13 @@ static int xen_pt_word_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint16_t writable_mask = 0;
-    uint16_t throughable_mask = 0;
+    uint16_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
 
     /* modify emulate register */
     writable_mask = reg->emu_mask & ~reg->ro_mask & valid_mask;
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
     *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
 
     return 0;
@@ -193,14 +203,13 @@ static int xen_pt_long_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint32_t writable_mask = 0;
-    uint32_t throughable_mask = 0;
+    uint32_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
 
     /* modify emulate register */
     writable_mask = reg->emu_mask & ~reg->ro_mask & valid_mask;
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
     *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
 
     return 0;
@@ -292,15 +301,13 @@ static int xen_pt_cmd_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint16_t writable_mask = 0;
-    uint16_t throughable_mask = 0;
+    uint16_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
 
     /* modify emulate register */
     writable_mask = ~reg->ro_mask & valid_mask;
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
-
     if (*val & PCI_COMMAND_INTX_DISABLE) {
         throughable_mask |= PCI_COMMAND_INTX_DISABLE;
     } else {
@@ -454,7 +461,6 @@ static int xen_pt_bar_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
     PCIDevice *d = &s->dev;
     const PCIIORegion *r;
     uint32_t writable_mask = 0;
-    uint32_t throughable_mask = 0;
     uint32_t bar_emu_mask = 0;
     uint32_t bar_ro_mask = 0;
     uint32_t r_size = 0;
@@ -511,8 +517,7 @@ static int xen_pt_bar_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
     }
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~bar_emu_mask & valid_mask;
-    *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
+    *val = XEN_PT_MERGE_VALUE(*val, dev_value, 0);
 
     return 0;
 }
@@ -526,9 +531,8 @@ static int xen_pt_exp_rom_bar_reg_write(XenPCIPassthroughState *s,
     XenPTRegion *base = NULL;
     PCIDevice *d = (PCIDevice *)&s->dev;
     uint32_t writable_mask = 0;
-    uint32_t throughable_mask = 0;
+    uint32_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
     pcibus_t r_size = 0;
-    uint32_t bar_emu_mask = 0;
     uint32_t bar_ro_mask = 0;
 
     r_size = d->io_regions[PCI_ROM_SLOT].size;
@@ -537,7 +541,6 @@ static int xen_pt_exp_rom_bar_reg_write(XenPCIPassthroughState *s,
     r_size = xen_pt_get_emul_size(base->bar_flag, r_size);
 
     /* set emulate mask and read-only mask */
-    bar_emu_mask = reg->emu_mask;
     bar_ro_mask = (reg->ro_mask | (r_size - 1)) & ~PCI_ROM_ADDRESS_ENABLE;
 
     /* modify emulate register */
@@ -545,7 +548,6 @@ static int xen_pt_exp_rom_bar_reg_write(XenPCIPassthroughState *s,
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~bar_emu_mask & valid_mask;
     *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
 
     return 0;
@@ -580,7 +582,7 @@ static XenPTRegInfo xen_pt_emu_reg_header0[] = {
         .offset     = PCI_COMMAND,
         .size       = 2,
         .init_val   = 0x0000,
-        .ro_mask    = 0xF880,
+        .res_mask   = 0xF880,
         .emu_mask   = 0x0743,
         .init       = xen_pt_common_reg_init,
         .u.w.read   = xen_pt_word_reg_read,
@@ -605,7 +607,8 @@ static XenPTRegInfo xen_pt_emu_reg_header0[] = {
         .offset     = PCI_STATUS,
         .size       = 2,
         .init_val   = 0x0000,
-        .ro_mask    = 0x06FF,
+        .res_mask   = 0x0007,
+        .ro_mask    = 0x06F8,
         .emu_mask   = 0x0010,
         .init       = xen_pt_status_reg_init,
         .u.w.read   = xen_pt_word_reg_read,
@@ -755,6 +758,15 @@ static XenPTRegInfo xen_pt_emu_reg_vpd[] = {
         .u.b.write  = xen_pt_byte_reg_write,
     },
     {
+        .offset     = PCI_VPD_ADDR,
+        .size       = 2,
+        .ro_mask    = 0x0003,
+        .emu_mask   = 0x0003,
+        .init       = xen_pt_common_reg_init,
+        .u.w.read   = xen_pt_word_reg_read,
+        .u.w.write  = xen_pt_word_reg_write,
+    },
+    {
         .size = 0,
     },
 };
@@ -873,7 +885,7 @@ static XenPTRegInfo xen_pt_emu_reg_pcie[] = {
         .offset     = PCI_EXP_DEVCAP,
         .size       = 4,
         .init_val   = 0x00000000,
-        .ro_mask    = 0x1FFCFFFF,
+        .ro_mask    = 0xFFFFFFFF,
         .emu_mask   = 0x10000000,
         .init       = xen_pt_common_reg_init,
         .u.dw.read  = xen_pt_long_reg_read,
@@ -890,6 +902,16 @@ static XenPTRegInfo xen_pt_emu_reg_pcie[] = {
         .u.w.read   = xen_pt_word_reg_read,
         .u.w.write  = xen_pt_word_reg_write,
     },
+    /* Device Status reg */
+    {
+        .offset     = PCI_EXP_DEVSTA,
+        .size       = 2,
+        .res_mask   = 0xFFC0,
+        .ro_mask    = 0x0030,
+        .init       = xen_pt_common_reg_init,
+        .u.w.read   = xen_pt_word_reg_read,
+        .u.w.write  = xen_pt_word_reg_write,
+    },
     /* Link Control reg */
     {
         .offset     = PCI_EXP_LNKCTL,
@@ -901,6 +923,15 @@ static XenPTRegInfo xen_pt_emu_reg_pcie[] = {
         .u.w.read   = xen_pt_word_reg_read,
         .u.w.write  = xen_pt_word_reg_write,
     },
+    /* Link Status reg */
+    {
+        .offset     = PCI_EXP_LNKSTA,
+        .size       = 2,
+        .ro_mask    = 0x3FFF,
+        .init       = xen_pt_common_reg_init,
+        .u.w.read   = xen_pt_word_reg_read,
+        .u.w.write  = xen_pt_word_reg_write,
+    },
     /* Device Control 2 reg */
     {
         .offset     = 0x28,
@@ -933,39 +964,22 @@ static XenPTRegInfo xen_pt_emu_reg_pcie[] = {
  * Power Management Capability
  */
 
-/* read Power Management Control/Status register */
-static int xen_pt_pmcsr_reg_read(XenPCIPassthroughState *s, XenPTReg *cfg_entry,
-                                 uint16_t *value, uint16_t valid_mask)
-{
-    XenPTRegInfo *reg = cfg_entry->reg;
-    uint16_t valid_emu_mask = reg->emu_mask;
-
-    valid_emu_mask |= PCI_PM_CTRL_STATE_MASK | PCI_PM_CTRL_NO_SOFT_RESET;
-
-    valid_emu_mask = valid_emu_mask & valid_mask;
-    *value = XEN_PT_MERGE_VALUE(*value, cfg_entry->data, ~valid_emu_mask);
-
-    return 0;
-}
 /* write Power Management Control/Status register */
 static int xen_pt_pmcsr_reg_write(XenPCIPassthroughState *s,
                                   XenPTReg *cfg_entry, uint16_t *val,
                                   uint16_t dev_value, uint16_t valid_mask)
 {
     XenPTRegInfo *reg = cfg_entry->reg;
-    uint16_t emu_mask = reg->emu_mask;
     uint16_t writable_mask = 0;
-    uint16_t throughable_mask = 0;
-
-    emu_mask |= PCI_PM_CTRL_STATE_MASK | PCI_PM_CTRL_NO_SOFT_RESET;
+    uint16_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
 
     /* modify emulate register */
-    writable_mask = emu_mask & ~reg->ro_mask & valid_mask;
+    writable_mask = reg->emu_mask & ~reg->ro_mask & valid_mask;
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~emu_mask & valid_mask;
-    *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
+    *val = XEN_PT_MERGE_VALUE(*val, dev_value & ~PCI_PM_CTRL_PME_STATUS,
+                              throughable_mask);
 
     return 0;
 }
@@ -999,10 +1013,11 @@ static XenPTRegInfo xen_pt_emu_reg_pm[] = {
         .offset     = PCI_PM_CTRL,
         .size       = 2,
         .init_val   = 0x0008,
-        .ro_mask    = 0xE1FC,
-        .emu_mask   = 0x8100,
+        .res_mask   = 0x00F0,
+        .ro_mask    = 0xE10C,
+        .emu_mask   = 0x810B,
         .init       = xen_pt_common_reg_init,
-        .u.w.read   = xen_pt_pmcsr_reg_read,
+        .u.w.read   = xen_pt_word_reg_read,
         .u.w.write  = xen_pt_pmcsr_reg_write,
     },
     {
@@ -1016,13 +1031,9 @@ static XenPTRegInfo xen_pt_emu_reg_pm[] = {
  */
 
 /* Helper */
-static bool xen_pt_msgdata_check_type(uint32_t offset, uint16_t flags)
-{
-    /* check the offset whether matches the type or not */
-    bool is_32 = (offset == PCI_MSI_DATA_32) && !(flags & PCI_MSI_FLAGS_64BIT);
-    bool is_64 = (offset == PCI_MSI_DATA_64) &&  (flags & PCI_MSI_FLAGS_64BIT);
-    return is_32 || is_64;
-}
+#define xen_pt_msi_check_type(offset, flags, what) \
+        ((offset) == ((flags) & PCI_MSI_FLAGS_64BIT ? \
+                      PCI_MSI_##what##_64 : PCI_MSI_##what##_32))
 
 /* Message Control register */
 static int xen_pt_msgctrl_reg_init(XenPCIPassthroughState *s,
@@ -1056,8 +1067,7 @@ static int xen_pt_msgctrl_reg_write(XenPCIPassthroughState *s,
     XenPTRegInfo *reg = cfg_entry->reg;
     XenPTMSI *msi = s->msi;
     uint16_t writable_mask = 0;
-    uint16_t throughable_mask = 0;
-    uint16_t raw_val;
+    uint16_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
 
     /* Currently no support for multi-vector */
     if (*val & PCI_MSI_FLAGS_QSIZE) {
@@ -1070,12 +1080,10 @@ static int xen_pt_msgctrl_reg_write(XenPCIPassthroughState *s,
     msi->flags |= cfg_entry->data & ~PCI_MSI_FLAGS_ENABLE;
 
     /* create value for writing to I/O device register */
-    raw_val = *val;
-    throughable_mask = ~reg->emu_mask & valid_mask;
     *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
 
     /* update MSI */
-    if (raw_val & PCI_MSI_FLAGS_ENABLE) {
+    if (*val & PCI_MSI_FLAGS_ENABLE) {
         /* setup MSI pirq for the first time */
         if (!msi->initialized) {
             /* Init physical one */
@@ -1103,10 +1111,6 @@ static int xen_pt_msgctrl_reg_write(XenPCIPassthroughState *s,
         xen_pt_msi_disable(s);
     }
 
-    /* pass through MSI_ENABLE bit */
-    *val &= ~PCI_MSI_FLAGS_ENABLE;
-    *val |= raw_val & PCI_MSI_FLAGS_ENABLE;
-
     return 0;
 }
 
@@ -1134,7 +1138,45 @@ static int xen_pt_msgdata_reg_init(XenPCIPassthroughState *s,
     uint32_t offset = reg->offset;
 
     /* check the offset whether matches the type or not */
-    if (xen_pt_msgdata_check_type(offset, flags)) {
+    if (xen_pt_msi_check_type(offset, flags, DATA)) {
+        *data = reg->init_val;
+    } else {
+        *data = XEN_PT_INVALID_REG;
+    }
+    return 0;
+}
+
+/* this function will be called twice (for 32 bit and 64 bit type) */
+/* initialize Mask register */
+static int xen_pt_mask_reg_init(XenPCIPassthroughState *s,
+                                XenPTRegInfo *reg, uint32_t real_offset,
+                                uint32_t *data)
+{
+    uint32_t flags = s->msi->flags;
+
+    /* check the offset whether matches the type or not */
+    if (!(flags & PCI_MSI_FLAGS_MASKBIT)) {
+        *data = XEN_PT_INVALID_REG;
+    } else if (xen_pt_msi_check_type(reg->offset, flags, MASK)) {
+        *data = reg->init_val;
+    } else {
+        *data = XEN_PT_INVALID_REG;
+    }
+    return 0;
+}
+
+/* this function will be called twice (for 32 bit and 64 bit type) */
+/* initialize Pending register */
+static int xen_pt_pending_reg_init(XenPCIPassthroughState *s,
+                                   XenPTRegInfo *reg, uint32_t real_offset,
+                                   uint32_t *data)
+{
+    uint32_t flags = s->msi->flags;
+
+    /* check the offset whether matches the type or not */
+    if (!(flags & PCI_MSI_FLAGS_MASKBIT)) {
+        *data = XEN_PT_INVALID_REG;
+    } else if (xen_pt_msi_check_type(reg->offset, flags, PENDING)) {
         *data = reg->init_val;
     } else {
         *data = XEN_PT_INVALID_REG;
@@ -1149,7 +1191,6 @@ static int xen_pt_msgaddr32_reg_write(XenPCIPassthroughState *s,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint32_t writable_mask = 0;
-    uint32_t throughable_mask = 0;
     uint32_t old_addr = cfg_entry->data;
 
     /* modify emulate register */
@@ -1158,8 +1199,7 @@ static int xen_pt_msgaddr32_reg_write(XenPCIPassthroughState *s,
     s->msi->addr_lo = cfg_entry->data;
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
-    *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
+    *val = XEN_PT_MERGE_VALUE(*val, dev_value, 0);
 
     /* update MSI */
     if (cfg_entry->data != old_addr) {
@@ -1177,7 +1217,6 @@ static int xen_pt_msgaddr64_reg_write(XenPCIPassthroughState *s,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint32_t writable_mask = 0;
-    uint32_t throughable_mask = 0;
     uint32_t old_addr = cfg_entry->data;
 
     /* check whether the type is 64 bit or not */
@@ -1194,8 +1233,7 @@ static int xen_pt_msgaddr64_reg_write(XenPCIPassthroughState *s,
     s->msi->addr_hi = cfg_entry->data;
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
-    *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
+    *val = XEN_PT_MERGE_VALUE(*val, dev_value, 0);
 
     /* update MSI */
     if (cfg_entry->data != old_addr) {
@@ -1217,12 +1255,11 @@ static int xen_pt_msgdata_reg_write(XenPCIPassthroughState *s,
     XenPTRegInfo *reg = cfg_entry->reg;
     XenPTMSI *msi = s->msi;
     uint16_t writable_mask = 0;
-    uint16_t throughable_mask = 0;
     uint16_t old_data = cfg_entry->data;
     uint32_t offset = reg->offset;
 
     /* check the offset whether matches the type or not */
-    if (!xen_pt_msgdata_check_type(offset, msi->flags)) {
+    if (!xen_pt_msi_check_type(offset, msi->flags, DATA)) {
         /* exit I/O emulator */
         XEN_PT_ERR(&s->dev, "the offset does not match the 32/64 bit type!\n");
         return -1;
@@ -1235,8 +1272,7 @@ static int xen_pt_msgdata_reg_write(XenPCIPassthroughState *s,
     msi->data = cfg_entry->data;
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
-    *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
+    *val = XEN_PT_MERGE_VALUE(*val, dev_value, 0);
 
     /* update MSI */
     if (cfg_entry->data != old_data) {
@@ -1266,8 +1302,9 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = {
         .offset     = PCI_MSI_FLAGS,
         .size       = 2,
         .init_val   = 0x0000,
-        .ro_mask    = 0xFF8E,
-        .emu_mask   = 0x007F,
+        .res_mask   = 0xFE00,
+        .ro_mask    = 0x018E,
+        .emu_mask   = 0x017E,
         .init       = xen_pt_msgctrl_reg_init,
         .u.w.read   = xen_pt_word_reg_read,
         .u.w.write  = xen_pt_msgctrl_reg_write,
@@ -1279,7 +1316,6 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = {
         .init_val   = 0x00000000,
         .ro_mask    = 0x00000003,
         .emu_mask   = 0xFFFFFFFF,
-        .no_wb      = 1,
         .init       = xen_pt_common_reg_init,
         .u.dw.read  = xen_pt_long_reg_read,
         .u.dw.write = xen_pt_msgaddr32_reg_write,
@@ -1291,7 +1327,6 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = {
         .init_val   = 0x00000000,
         .ro_mask    = 0x00000000,
         .emu_mask   = 0xFFFFFFFF,
-        .no_wb      = 1,
         .init       = xen_pt_msgaddr64_reg_init,
         .u.dw.read  = xen_pt_long_reg_read,
         .u.dw.write = xen_pt_msgaddr64_reg_write,
@@ -1303,7 +1338,6 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = {
         .init_val   = 0x0000,
         .ro_mask    = 0x0000,
         .emu_mask   = 0xFFFF,
-        .no_wb      = 1,
         .init       = xen_pt_msgdata_reg_init,
         .u.w.read   = xen_pt_word_reg_read,
         .u.w.write  = xen_pt_msgdata_reg_write,
@@ -1315,11 +1349,54 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = {
         .init_val   = 0x0000,
         .ro_mask    = 0x0000,
         .emu_mask   = 0xFFFF,
-        .no_wb      = 1,
         .init       = xen_pt_msgdata_reg_init,
         .u.w.read   = xen_pt_word_reg_read,
         .u.w.write  = xen_pt_msgdata_reg_write,
     },
+    /* Mask reg (if PCI_MSI_FLAGS_MASKBIT set, for 32-bit devices) */
+    {
+        .offset     = PCI_MSI_MASK_32,
+        .size       = 4,
+        .init_val   = 0x00000000,
+        .ro_mask    = 0xFFFFFFFF,
+        .emu_mask   = 0xFFFFFFFF,
+        .init       = xen_pt_mask_reg_init,
+        .u.dw.read  = xen_pt_long_reg_read,
+        .u.dw.write = xen_pt_long_reg_write,
+    },
+    /* Mask reg (if PCI_MSI_FLAGS_MASKBIT set, for 64-bit devices) */
+    {
+        .offset     = PCI_MSI_MASK_64,
+        .size       = 4,
+        .init_val   = 0x00000000,
+        .ro_mask    = 0xFFFFFFFF,
+        .emu_mask   = 0xFFFFFFFF,
+        .init       = xen_pt_mask_reg_init,
+        .u.dw.read  = xen_pt_long_reg_read,
+        .u.dw.write = xen_pt_long_reg_write,
+    },
+    /* Pending reg (if PCI_MSI_FLAGS_MASKBIT set, for 32-bit devices) */
+    {
+        .offset     = PCI_MSI_MASK_32 + 4,
+        .size       = 4,
+        .init_val   = 0x00000000,
+        .ro_mask    = 0xFFFFFFFF,
+        .emu_mask   = 0x00000000,
+        .init       = xen_pt_pending_reg_init,
+        .u.dw.read  = xen_pt_long_reg_read,
+        .u.dw.write = xen_pt_long_reg_write,
+    },
+    /* Pending reg (if PCI_MSI_FLAGS_MASKBIT set, for 64-bit devices) */
+    {
+        .offset     = PCI_MSI_MASK_64 + 4,
+        .size       = 4,
+        .init_val   = 0x00000000,
+        .ro_mask    = 0xFFFFFFFF,
+        .emu_mask   = 0x00000000,
+        .init       = xen_pt_pending_reg_init,
+        .u.dw.read  = xen_pt_long_reg_read,
+        .u.dw.write = xen_pt_long_reg_write,
+    },
     {
         .size = 0,
     },
@@ -1358,7 +1435,7 @@ static int xen_pt_msixctrl_reg_write(XenPCIPassthroughState *s,
 {
     XenPTRegInfo *reg = cfg_entry->reg;
     uint16_t writable_mask = 0;
-    uint16_t throughable_mask = 0;
+    uint16_t throughable_mask = get_throughable_mask(s, reg, valid_mask);
     int debug_msix_enabled_old;
 
     /* modify emulate register */
@@ -1366,7 +1443,6 @@ static int xen_pt_msixctrl_reg_write(XenPCIPassthroughState *s,
     cfg_entry->data = XEN_PT_MERGE_VALUE(*val, cfg_entry->data, writable_mask);
 
     /* create value for writing to I/O device register */
-    throughable_mask = ~reg->emu_mask & valid_mask;
     *val = XEN_PT_MERGE_VALUE(*val, dev_value, throughable_mask);
 
     /* update MSI-X */
@@ -1405,7 +1481,8 @@ static XenPTRegInfo xen_pt_emu_reg_msix[] = {
         .offset     = PCI_MSI_FLAGS,
         .size       = 2,
         .init_val   = 0x0000,
-        .ro_mask    = 0x3FFF,
+        .res_mask   = 0x3800,
+        .ro_mask    = 0x07FF,
         .emu_mask   = 0x0000,
         .init       = xen_pt_msixctrl_reg_init,
         .u.w.read   = xen_pt_word_reg_read,
diff --git a/hw/xen/xen_pt_msi.c b/hw/xen/xen_pt_msi.c
index 9ed9321c9f..68db6233dc 100644
--- a/hw/xen/xen_pt_msi.c
+++ b/hw/xen/xen_pt_msi.c
@@ -434,11 +434,10 @@ static void pci_msix_write(void *opaque, hwaddr addr,
     XenPCIPassthroughState *s = opaque;
     XenPTMSIX *msix = s->msix;
     XenPTMSIXEntry *entry;
-    int entry_nr, offset;
+    unsigned int entry_nr, offset;
 
     entry_nr = addr / PCI_MSIX_ENTRY_SIZE;
-    if (entry_nr < 0 || entry_nr >= msix->total_entries) {
-        XEN_PT_ERR(&s->dev, "asked MSI-X entry '%i' invalid!\n", entry_nr);
+    if (entry_nr >= msix->total_entries) {
         return;
     }
     entry = &msix->msix_entry[entry_nr];
@@ -460,8 +459,11 @@ static void pci_msix_write(void *opaque, hwaddr addr,
             + PCI_MSIX_ENTRY_VECTOR_CTRL;
 
         if (msix->enabled && !(*vec_ctrl & PCI_MSIX_ENTRY_CTRL_MASKBIT)) {
-            XEN_PT_ERR(&s->dev, "Can't update msix entry %d since MSI-X is"
-                       " already enabled.\n", entry_nr);
+            if (!entry->warned) {
+                entry->warned = true;
+                XEN_PT_ERR(&s->dev, "Can't update msix entry %d since MSI-X is"
+                           " already enabled.\n", entry_nr);
+            }
             return;
         }