summary refs log tree commit diff stats
path: root/hw
diff options
context:
space:
mode:
Diffstat (limited to 'hw')
-rw-r--r--hw/9pfs/xen-9pfs.h4
-rw-r--r--hw/arm/Kconfig14
-rw-r--r--hw/arm/Makefile.objs1
-rw-r--r--hw/arm/aspeed.c76
-rw-r--r--hw/arm/aspeed_soc.c268
-rw-r--r--hw/arm/boot.c3
-rw-r--r--hw/arm/fsl-imx7.c11
-rw-r--r--hw/arm/msf2-som.c1
-rw-r--r--hw/arm/sbsa-ref.c806
-rw-r--r--hw/arm/virt.c1
-rw-r--r--hw/block/dataplane/xen-block.c44
-rw-r--r--hw/block/dataplane/xen-block.h3
-rw-r--r--hw/block/pflash_cfi01.c11
-rw-r--r--hw/block/pflash_cfi02.c707
-rw-r--r--hw/block/trace-events10
-rw-r--r--hw/block/xen-block.c40
-rw-r--r--hw/block/xen_blkif.h5
-rw-r--r--hw/char/xen_console.c2
-rw-r--r--hw/core/Makefile.objs4
-rw-r--r--hw/core/machine-hmp-cmds.c164
-rw-r--r--hw/core/machine-qmp-cmds.c328
-rw-r--r--hw/core/numa.c605
-rw-r--r--hw/core/qdev.c2
-rw-r--r--hw/display/Kconfig2
-rw-r--r--hw/display/ati.c98
-rw-r--r--hw/display/ati_2d.c26
-rw-r--r--hw/display/ati_dbg.c2
-rw-r--r--hw/display/ati_int.h6
-rw-r--r--hw/display/ati_regs.h6
-rw-r--r--hw/display/virtio-gpu.c9
-rw-r--r--hw/display/xenfb.c7
-rw-r--r--hw/dma/rc4030.c21
-rw-r--r--hw/i2c/bitbang_i2c.c49
-rw-r--r--hw/i2c/bitbang_i2c.h12
-rw-r--r--hw/i2c/ppc4xx_i2c.c7
-rw-r--r--hw/i2c/versatile_i2c.c10
-rw-r--r--hw/i386/xen/xen-hvm.c16
-rw-r--r--hw/i386/xen/xen-mapcache.c2
-rw-r--r--hw/intc/armv7m_nvic.c54
-rw-r--r--hw/intc/aspeed_vic.c105
-rw-r--r--hw/intc/pnv_xive.c145
-rw-r--r--hw/intc/spapr_xive.c38
-rw-r--r--hw/intc/spapr_xive_kvm.c55
-rw-r--r--hw/intc/xics.c40
-rw-r--r--hw/intc/xics_kvm.c117
-rw-r--r--hw/intc/xics_spapr.c55
-rw-r--r--hw/intc/xive.c72
-rw-r--r--hw/mips/Kconfig30
-rw-r--r--hw/mips/gt64xxx_pci.c494
-rw-r--r--hw/mips/trace-events4
-rw-r--r--hw/misc/Makefile.objs1
-rw-r--r--hw/misc/aspeed_xdma.c165
-rw-r--r--hw/misc/trace-events3
-rw-r--r--hw/net/ftgmac100.c2
-rw-r--r--hw/net/sunhme.c29
-rw-r--r--hw/net/trace-events2
-rw-r--r--hw/net/virtio-net.c4
-rw-r--r--hw/net/xen_nic.c2
-rw-r--r--hw/pci-host/designware.c18
-rw-r--r--hw/ppc/mac_newworld.c4
-rw-r--r--hw/ppc/mac_oldworld.c2
-rw-r--r--hw/ppc/pnv.c34
-rw-r--r--hw/ppc/pnv_xscom.c17
-rw-r--r--hw/ppc/ppc.c7
-rw-r--r--hw/ppc/prep.c2
-rw-r--r--hw/ppc/spapr_irq.c39
-rw-r--r--hw/ppc/spapr_pci.c28
-rw-r--r--hw/ppc/spapr_rtc.c2
-rw-r--r--hw/riscv/Makefile.objs1
-rw-r--r--hw/riscv/boot.c105
-rw-r--r--hw/riscv/sifive_e.c30
-rw-r--r--hw/riscv/sifive_prci.c49
-rw-r--r--hw/riscv/sifive_u.c54
-rw-r--r--hw/riscv/spike.c21
-rw-r--r--hw/riscv/virt.c79
-rw-r--r--hw/s390x/css.c27
-rw-r--r--hw/s390x/s390-ccw.c20
-rw-r--r--hw/s390x/s390-skeys.c2
-rw-r--r--hw/sparc/sun4m.c9
-rw-r--r--hw/ssi/aspeed_smc.c1
-rw-r--r--hw/timer/Makefile.objs2
-rw-r--r--hw/timer/armv7m_systick.c26
-rw-r--r--hw/timer/aspeed_rtc.c180
-rw-r--r--hw/timer/aspeed_timer.c78
-rw-r--r--hw/timer/mc146818rtc.c4
-rw-r--r--hw/timer/trace-events4
-rw-r--r--hw/usb/xen-usb.c3
-rw-r--r--hw/vfio/ccw.c162
-rw-r--r--hw/vfio/pci.c7
-rw-r--r--hw/watchdog/wdt_aspeed.c20
-rw-r--r--hw/xen/xen-bus.c92
-rw-r--r--hw/xen/xen-legacy-backend.c2
92 files changed, 4663 insertions, 1268 deletions
diff --git a/hw/9pfs/xen-9pfs.h b/hw/9pfs/xen-9pfs.h
index fbdee3d843..241e2216a4 100644
--- a/hw/9pfs/xen-9pfs.h
+++ b/hw/9pfs/xen-9pfs.h
@@ -13,8 +13,8 @@
 #ifndef HW_9PFS_XEN_9PFS_H
 #define HW_9PFS_XEN_9PFS_H
 
-#include <xen/io/protocols.h>
-#include "hw/xen/io/ring.h"
+#include "hw/xen/interface/io/protocols.h"
+#include "hw/xen/interface/io/ring.h"
 
 /*
  * Do not merge into xen-9p-backend.c: clang doesn't allow unused static
diff --git a/hw/arm/Kconfig b/hw/arm/Kconfig
index 9aced9d54d..ab65ecd216 100644
--- a/hw/arm/Kconfig
+++ b/hw/arm/Kconfig
@@ -184,6 +184,20 @@ config REALVIEW
     select DS1338 # I2C RTC+NVRAM
     select USB_OHCI
 
+config SBSA_REF
+    bool
+    imply PCI_DEVICES
+    select AHCI
+    select ARM_SMMUV3
+    select GPIO_KEY
+    select PCI_EXPRESS
+    select PCI_EXPRESS_GENERIC_BRIDGE
+    select PFLASH_CFI01
+    select PL011 # UART
+    select PL031 # RTC
+    select PL061 # GPIO
+    select USB_EHCI_SYSBUS
+
 config SABRELITE
     bool
     select FSL_IMX6
diff --git a/hw/arm/Makefile.objs b/hw/arm/Makefile.objs
index 994e67dd0d..43ce8d5b19 100644
--- a/hw/arm/Makefile.objs
+++ b/hw/arm/Makefile.objs
@@ -19,6 +19,7 @@ obj-$(CONFIG_SPITZ) += spitz.o
 obj-$(CONFIG_TOSA) += tosa.o
 obj-$(CONFIG_Z2) += z2.o
 obj-$(CONFIG_REALVIEW) += realview.o
+obj-$(CONFIG_SBSA_REF) += sbsa-ref.o
 obj-$(CONFIG_STELLARIS) += stellaris.o
 obj-$(CONFIG_COLLIE) += collie.o
 obj-$(CONFIG_VERSATILE) += versatilepb.o
diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c
index d2ad2da24b..8b6d304247 100644
--- a/hw/arm/aspeed.c
+++ b/hw/arm/aspeed.c
@@ -22,17 +22,18 @@
 #include "hw/misc/tmp105.h"
 #include "qemu/log.h"
 #include "sysemu/block-backend.h"
+#include "sysemu/sysemu.h"
 #include "hw/loader.h"
 #include "qemu/error-report.h"
 #include "qemu/units.h"
 
 static struct arm_boot_info aspeed_board_binfo = {
     .board_id = -1, /* device-tree-only board */
-    .nb_cpus = 1,
 };
 
 struct AspeedBoardState {
     AspeedSoCState soc;
+    MemoryRegion ram_container;
     MemoryRegion ram;
     MemoryRegion max_ram;
 };
@@ -72,6 +73,17 @@ struct AspeedBoardState {
         SCU_AST2500_HW_STRAP_ACPI_ENABLE |                              \
         SCU_HW_STRAP_SPI_MODE(SCU_HW_STRAP_SPI_MASTER))
 
+/* Swift hardware value: 0xF11AD206 */
+#define SWIFT_BMC_HW_STRAP1 (                                           \
+        AST2500_HW_STRAP1_DEFAULTS |                                    \
+        SCU_AST2500_HW_STRAP_SPI_AUTOFETCH_ENABLE |                     \
+        SCU_AST2500_HW_STRAP_GPIO_STRAP_ENABLE |                        \
+        SCU_AST2500_HW_STRAP_UART_DEBUG |                               \
+        SCU_AST2500_HW_STRAP_DDR4_ENABLE |                              \
+        SCU_H_PLL_BYPASS_EN |                                           \
+        SCU_AST2500_HW_STRAP_ACPI_ENABLE |                              \
+        SCU_HW_STRAP_SPI_MODE(SCU_HW_STRAP_SPI_MASTER))
+
 /* Witherspoon hardware value: 0xF10AD216 (but use romulus definition) */
 #define WITHERSPOON_BMC_HW_STRAP1 ROMULUS_BMC_HW_STRAP1
 
@@ -159,6 +171,10 @@ static void aspeed_board_init(MachineState *machine,
     ram_addr_t max_ram_size;
 
     bmc = g_new0(AspeedBoardState, 1);
+
+    memory_region_init(&bmc->ram_container, NULL, "aspeed-ram-container",
+                       UINT32_MAX);
+
     object_initialize_child(OBJECT(machine), "soc", &bmc->soc,
                             (sizeof(bmc->soc)), cfg->soc_name, &error_abort,
                             NULL);
@@ -171,6 +187,8 @@ static void aspeed_board_init(MachineState *machine,
                             &error_abort);
     object_property_set_int(OBJECT(&bmc->soc), cfg->num_cs, "num-cs",
                             &error_abort);
+    object_property_set_int(OBJECT(&bmc->soc), smp_cpus, "num-cpus",
+                            &error_abort);
     if (machine->kernel_filename) {
         /*
          * When booting with a -kernel command line there is no u-boot
@@ -191,18 +209,16 @@ static void aspeed_board_init(MachineState *machine,
                                         &error_abort);
 
     memory_region_allocate_system_memory(&bmc->ram, NULL, "ram", ram_size);
-    memory_region_add_subregion(get_system_memory(), sc->info->sdram_base,
-                                &bmc->ram);
-    object_property_add_const_link(OBJECT(&bmc->soc), "ram", OBJECT(&bmc->ram),
-                                   &error_abort);
+    memory_region_add_subregion(&bmc->ram_container, 0, &bmc->ram);
+    memory_region_add_subregion(get_system_memory(),
+                                sc->info->memmap[ASPEED_SDRAM],
+                                &bmc->ram_container);
 
     max_ram_size = object_property_get_uint(OBJECT(&bmc->soc), "max-ram-size",
                                             &error_abort);
     memory_region_init_io(&bmc->max_ram, NULL, &max_ram_ops, NULL,
                           "max_ram", max_ram_size  - ram_size);
-    memory_region_add_subregion(get_system_memory(),
-                                sc->info->sdram_base + ram_size,
-                                &bmc->max_ram);
+    memory_region_add_subregion(&bmc->ram_container, ram_size, &bmc->max_ram);
 
     aspeed_board_init_flashes(&bmc->soc.fmc, cfg->fmc_model, &error_abort);
     aspeed_board_init_flashes(&bmc->soc.spi[0], cfg->spi_model, &error_abort);
@@ -229,7 +245,8 @@ static void aspeed_board_init(MachineState *machine,
     aspeed_board_binfo.initrd_filename = machine->initrd_filename;
     aspeed_board_binfo.kernel_cmdline = machine->kernel_cmdline;
     aspeed_board_binfo.ram_size = ram_size;
-    aspeed_board_binfo.loader_start = sc->info->sdram_base;
+    aspeed_board_binfo.loader_start = sc->info->memmap[ASPEED_SDRAM];
+    aspeed_board_binfo.nb_cpus = bmc->soc.num_cpus;
 
     if (cfg->i2c_init) {
         cfg->i2c_init(bmc);
@@ -286,6 +303,35 @@ static void romulus_bmc_i2c_init(AspeedBoardState *bmc)
     i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 11), "ds1338", 0x32);
 }
 
+static void swift_bmc_i2c_init(AspeedBoardState *bmc)
+{
+    AspeedSoCState *soc = &bmc->soc;
+
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 3), "pca9552", 0x60);
+
+    /* The swift board expects a TMP275 but a TMP105 is compatible */
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 7), "tmp105", 0x48);
+    /* The swift board expects a pca9551 but a pca9552 is compatible */
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 7), "pca9552", 0x60);
+
+    /* The swift board expects an Epson RX8900 RTC but a ds1338 is compatible */
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 8), "ds1338", 0x32);
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 8), "pca9552", 0x60);
+
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 9), "tmp423", 0x4c);
+    /* The swift board expects a pca9539 but a pca9552 is compatible */
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 9), "pca9552", 0x74);
+
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 10), "tmp423", 0x4c);
+    /* The swift board expects a pca9539 but a pca9552 is compatible */
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 10), "pca9552",
+                     0x74);
+
+    /* The swift board expects a TMP275 but a TMP105 is compatible */
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 12), "tmp105", 0x48);
+    i2c_create_slave(aspeed_i2c_get_bus(DEVICE(&soc->i2c), 12), "tmp105", 0x4a);
+}
+
 static void witherspoon_bmc_i2c_init(AspeedBoardState *bmc)
 {
     AspeedSoCState *soc = &bmc->soc;
@@ -326,7 +372,7 @@ static void aspeed_machine_class_init(ObjectClass *oc, void *data)
 
     mc->desc = board->desc;
     mc->init = aspeed_machine_init;
-    mc->max_cpus = 1;
+    mc->max_cpus = ASPEED_CPUS_NUM;
     mc->no_sdcard = 1;
     mc->no_floppy = 1;
     mc->no_cdrom = 1;
@@ -377,6 +423,16 @@ static const AspeedBoardConfig aspeed_boards[] = {
         .i2c_init  = romulus_bmc_i2c_init,
         .ram       = 512 * MiB,
     }, {
+        .name      = MACHINE_TYPE_NAME("swift-bmc"),
+        .desc      = "OpenPOWER Swift BMC (ARM1176)",
+        .soc_name  = "ast2500-a1",
+        .hw_strap1 = SWIFT_BMC_HW_STRAP1,
+        .fmc_model = "mx66l1g45g",
+        .spi_model = "mx66l1g45g",
+        .num_cs    = 2,
+        .i2c_init  = swift_bmc_i2c_init,
+        .ram       = 512 * MiB,
+    }, {
         .name      = MACHINE_TYPE_NAME("witherspoon-bmc"),
         .desc      = "OpenPOWER Witherspoon BMC (ARM1176)",
         .soc_name  = "ast2500-a1",
diff --git a/hw/arm/aspeed_soc.c b/hw/arm/aspeed_soc.c
index a2ea8c7484..c6fb3700f2 100644
--- a/hw/arm/aspeed_soc.c
+++ b/hw/arm/aspeed_soc.c
@@ -19,36 +19,99 @@
 #include "hw/char/serial.h"
 #include "qemu/log.h"
 #include "qemu/module.h"
+#include "qemu/error-report.h"
 #include "hw/i2c/aspeed_i2c.h"
 #include "net/net.h"
 
-#define ASPEED_SOC_UART_5_BASE      0x00184000
 #define ASPEED_SOC_IOMEM_SIZE       0x00200000
-#define ASPEED_SOC_IOMEM_BASE       0x1E600000
-#define ASPEED_SOC_FMC_BASE         0x1E620000
-#define ASPEED_SOC_SPI_BASE         0x1E630000
-#define ASPEED_SOC_SPI2_BASE        0x1E631000
-#define ASPEED_SOC_VIC_BASE         0x1E6C0000
-#define ASPEED_SOC_SDMC_BASE        0x1E6E0000
-#define ASPEED_SOC_SCU_BASE         0x1E6E2000
-#define ASPEED_SOC_SRAM_BASE        0x1E720000
-#define ASPEED_SOC_TIMER_BASE       0x1E782000
-#define ASPEED_SOC_WDT_BASE         0x1E785000
-#define ASPEED_SOC_I2C_BASE         0x1E78A000
-#define ASPEED_SOC_ETH1_BASE        0x1E660000
-#define ASPEED_SOC_ETH2_BASE        0x1E680000
-
-static const int uart_irqs[] = { 9, 32, 33, 34, 10 };
-static const int timer_irqs[] = { 16, 17, 18, 35, 36, 37, 38, 39, };
-
-#define AST2400_SDRAM_BASE       0x40000000
-#define AST2500_SDRAM_BASE       0x80000000
-
-static const hwaddr aspeed_soc_ast2400_spi_bases[] = { ASPEED_SOC_SPI_BASE };
-static const char *aspeed_soc_ast2400_typenames[] = { "aspeed.smc.spi" };
 
-static const hwaddr aspeed_soc_ast2500_spi_bases[] = { ASPEED_SOC_SPI_BASE,
-                                                       ASPEED_SOC_SPI2_BASE};
+static const hwaddr aspeed_soc_ast2400_memmap[] = {
+    [ASPEED_IOMEM]  = 0x1E600000,
+    [ASPEED_FMC]    = 0x1E620000,
+    [ASPEED_SPI1]   = 0x1E630000,
+    [ASPEED_VIC]    = 0x1E6C0000,
+    [ASPEED_SDMC]   = 0x1E6E0000,
+    [ASPEED_SCU]    = 0x1E6E2000,
+    [ASPEED_XDMA]   = 0x1E6E7000,
+    [ASPEED_ADC]    = 0x1E6E9000,
+    [ASPEED_SRAM]   = 0x1E720000,
+    [ASPEED_GPIO]   = 0x1E780000,
+    [ASPEED_RTC]    = 0x1E781000,
+    [ASPEED_TIMER1] = 0x1E782000,
+    [ASPEED_WDT]    = 0x1E785000,
+    [ASPEED_PWM]    = 0x1E786000,
+    [ASPEED_LPC]    = 0x1E789000,
+    [ASPEED_IBT]    = 0x1E789140,
+    [ASPEED_I2C]    = 0x1E78A000,
+    [ASPEED_ETH1]   = 0x1E660000,
+    [ASPEED_ETH2]   = 0x1E680000,
+    [ASPEED_UART1]  = 0x1E783000,
+    [ASPEED_UART5]  = 0x1E784000,
+    [ASPEED_VUART]  = 0x1E787000,
+    [ASPEED_SDRAM]  = 0x40000000,
+};
+
+static const hwaddr aspeed_soc_ast2500_memmap[] = {
+    [ASPEED_IOMEM]  = 0x1E600000,
+    [ASPEED_FMC]    = 0x1E620000,
+    [ASPEED_SPI1]   = 0x1E630000,
+    [ASPEED_SPI2]   = 0x1E631000,
+    [ASPEED_VIC]    = 0x1E6C0000,
+    [ASPEED_SDMC]   = 0x1E6E0000,
+    [ASPEED_SCU]    = 0x1E6E2000,
+    [ASPEED_XDMA]   = 0x1E6E7000,
+    [ASPEED_ADC]    = 0x1E6E9000,
+    [ASPEED_SRAM]   = 0x1E720000,
+    [ASPEED_GPIO]   = 0x1E780000,
+    [ASPEED_RTC]    = 0x1E781000,
+    [ASPEED_TIMER1] = 0x1E782000,
+    [ASPEED_WDT]    = 0x1E785000,
+    [ASPEED_PWM]    = 0x1E786000,
+    [ASPEED_LPC]    = 0x1E789000,
+    [ASPEED_IBT]    = 0x1E789140,
+    [ASPEED_I2C]    = 0x1E78A000,
+    [ASPEED_ETH1]   = 0x1E660000,
+    [ASPEED_ETH2]   = 0x1E680000,
+    [ASPEED_UART1]  = 0x1E783000,
+    [ASPEED_UART5]  = 0x1E784000,
+    [ASPEED_VUART]  = 0x1E787000,
+    [ASPEED_SDRAM]  = 0x80000000,
+};
+
+static const int aspeed_soc_ast2400_irqmap[] = {
+    [ASPEED_UART1]  = 9,
+    [ASPEED_UART2]  = 32,
+    [ASPEED_UART3]  = 33,
+    [ASPEED_UART4]  = 34,
+    [ASPEED_UART5]  = 10,
+    [ASPEED_VUART]  = 8,
+    [ASPEED_FMC]    = 19,
+    [ASPEED_SDMC]   = 0,
+    [ASPEED_SCU]    = 21,
+    [ASPEED_ADC]    = 31,
+    [ASPEED_GPIO]   = 20,
+    [ASPEED_RTC]    = 22,
+    [ASPEED_TIMER1] = 16,
+    [ASPEED_TIMER2] = 17,
+    [ASPEED_TIMER3] = 18,
+    [ASPEED_TIMER4] = 35,
+    [ASPEED_TIMER5] = 36,
+    [ASPEED_TIMER6] = 37,
+    [ASPEED_TIMER7] = 38,
+    [ASPEED_TIMER8] = 39,
+    [ASPEED_WDT]    = 27,
+    [ASPEED_PWM]    = 28,
+    [ASPEED_LPC]    = 8,
+    [ASPEED_IBT]    = 8, /* LPC */
+    [ASPEED_I2C]    = 12,
+    [ASPEED_ETH1]   = 2,
+    [ASPEED_ETH2]   = 3,
+    [ASPEED_XDMA]   = 6,
+};
+
+#define aspeed_soc_ast2500_irqmap aspeed_soc_ast2400_irqmap
+
+static const char *aspeed_soc_ast2400_typenames[] = { "aspeed.smc.spi" };
 static const char *aspeed_soc_ast2500_typenames[] = {
     "aspeed.smc.ast2500-spi1", "aspeed.smc.ast2500-spi2" };
 
@@ -57,57 +120,71 @@ static const AspeedSoCInfo aspeed_socs[] = {
         .name         = "ast2400-a0",
         .cpu_type     = ARM_CPU_TYPE_NAME("arm926"),
         .silicon_rev  = AST2400_A0_SILICON_REV,
-        .sdram_base   = AST2400_SDRAM_BASE,
         .sram_size    = 0x8000,
         .spis_num     = 1,
-        .spi_bases    = aspeed_soc_ast2400_spi_bases,
         .fmc_typename = "aspeed.smc.fmc",
         .spi_typename = aspeed_soc_ast2400_typenames,
         .wdts_num     = 2,
+        .irqmap       = aspeed_soc_ast2400_irqmap,
+        .memmap       = aspeed_soc_ast2400_memmap,
+        .num_cpus     = 1,
     }, {
         .name         = "ast2400-a1",
         .cpu_type     = ARM_CPU_TYPE_NAME("arm926"),
         .silicon_rev  = AST2400_A1_SILICON_REV,
-        .sdram_base   = AST2400_SDRAM_BASE,
         .sram_size    = 0x8000,
         .spis_num     = 1,
-        .spi_bases    = aspeed_soc_ast2400_spi_bases,
         .fmc_typename = "aspeed.smc.fmc",
         .spi_typename = aspeed_soc_ast2400_typenames,
         .wdts_num     = 2,
+        .irqmap       = aspeed_soc_ast2400_irqmap,
+        .memmap       = aspeed_soc_ast2400_memmap,
+        .num_cpus     = 1,
     }, {
         .name         = "ast2400",
         .cpu_type     = ARM_CPU_TYPE_NAME("arm926"),
         .silicon_rev  = AST2400_A0_SILICON_REV,
-        .sdram_base   = AST2400_SDRAM_BASE,
         .sram_size    = 0x8000,
         .spis_num     = 1,
-        .spi_bases    = aspeed_soc_ast2400_spi_bases,
         .fmc_typename = "aspeed.smc.fmc",
         .spi_typename = aspeed_soc_ast2400_typenames,
         .wdts_num     = 2,
+        .irqmap       = aspeed_soc_ast2400_irqmap,
+        .memmap       = aspeed_soc_ast2400_memmap,
+        .num_cpus     = 1,
     }, {
         .name         = "ast2500-a1",
         .cpu_type     = ARM_CPU_TYPE_NAME("arm1176"),
         .silicon_rev  = AST2500_A1_SILICON_REV,
-        .sdram_base   = AST2500_SDRAM_BASE,
         .sram_size    = 0x9000,
         .spis_num     = 2,
-        .spi_bases    = aspeed_soc_ast2500_spi_bases,
         .fmc_typename = "aspeed.smc.ast2500-fmc",
         .spi_typename = aspeed_soc_ast2500_typenames,
         .wdts_num     = 3,
+        .irqmap       = aspeed_soc_ast2500_irqmap,
+        .memmap       = aspeed_soc_ast2500_memmap,
+        .num_cpus     = 1,
     },
 };
 
+static qemu_irq aspeed_soc_get_irq(AspeedSoCState *s, int ctrl)
+{
+    AspeedSoCClass *sc = ASPEED_SOC_GET_CLASS(s);
+
+    return qdev_get_gpio_in(DEVICE(&s->vic), sc->info->irqmap[ctrl]);
+}
+
 static void aspeed_soc_init(Object *obj)
 {
     AspeedSoCState *s = ASPEED_SOC(obj);
     AspeedSoCClass *sc = ASPEED_SOC_GET_CLASS(s);
     int i;
 
-    object_initialize_child(obj, "cpu", OBJECT(&s->cpu), sizeof(s->cpu),
-                            sc->info->cpu_type, &error_abort, NULL);
+    for (i = 0; i < sc->info->num_cpus; i++) {
+        object_initialize_child(obj, "cpu[*]", OBJECT(&s->cpu[i]),
+                                sizeof(s->cpu[i]), sc->info->cpu_type,
+                                &error_abort, NULL);
+    }
 
     sysbus_init_child_obj(obj, "scu", OBJECT(&s->scu), sizeof(s->scu),
                           TYPE_ASPEED_SCU);
@@ -123,6 +200,9 @@ static void aspeed_soc_init(Object *obj)
     sysbus_init_child_obj(obj, "vic", OBJECT(&s->vic), sizeof(s->vic),
                           TYPE_ASPEED_VIC);
 
+    sysbus_init_child_obj(obj, "rtc", OBJECT(&s->rtc), sizeof(s->rtc),
+                          TYPE_ASPEED_RTC);
+
     sysbus_init_child_obj(obj, "timerctrl", OBJECT(&s->timerctrl),
                           sizeof(s->timerctrl), TYPE_ASPEED_TIMER);
     object_property_add_const_link(OBJECT(&s->timerctrl), "scu",
@@ -155,10 +235,17 @@ static void aspeed_soc_init(Object *obj)
                               sizeof(s->wdt[i]), TYPE_ASPEED_WDT);
         qdev_prop_set_uint32(DEVICE(&s->wdt[i]), "silicon-rev",
                                     sc->info->silicon_rev);
+        object_property_add_const_link(OBJECT(&s->wdt[i]), "scu",
+                                       OBJECT(&s->scu), &error_abort);
     }
 
-    sysbus_init_child_obj(obj, "ftgmac100", OBJECT(&s->ftgmac100),
-                          sizeof(s->ftgmac100), TYPE_FTGMAC100);
+    for (i = 0; i < ASPEED_MACS_NUM; i++) {
+        sysbus_init_child_obj(obj, "ftgmac100[*]", OBJECT(&s->ftgmac100[i]),
+                              sizeof(s->ftgmac100[i]), TYPE_FTGMAC100);
+    }
+
+    sysbus_init_child_obj(obj, "xdma", OBJECT(&s->xdma), sizeof(s->xdma),
+                          TYPE_ASPEED_XDMA);
 }
 
 static void aspeed_soc_realize(DeviceState *dev, Error **errp)
@@ -169,14 +256,22 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
     Error *err = NULL, *local_err = NULL;
 
     /* IO space */
-    create_unimplemented_device("aspeed_soc.io",
-                                ASPEED_SOC_IOMEM_BASE, ASPEED_SOC_IOMEM_SIZE);
+    create_unimplemented_device("aspeed_soc.io", sc->info->memmap[ASPEED_IOMEM],
+                                ASPEED_SOC_IOMEM_SIZE);
+
+    if (s->num_cpus > sc->info->num_cpus) {
+        warn_report("%s: invalid number of CPUs %d, using default %d",
+                    sc->info->name, s->num_cpus, sc->info->num_cpus);
+        s->num_cpus = sc->info->num_cpus;
+    }
 
     /* CPU */
-    object_property_set_bool(OBJECT(&s->cpu), true, "realized", &err);
-    if (err) {
-        error_propagate(errp, err);
-        return;
+    for (i = 0; i < s->num_cpus; i++) {
+        object_property_set_bool(OBJECT(&s->cpu[i]), true, "realized", &err);
+        if (err) {
+            error_propagate(errp, err);
+            return;
+        }
     }
 
     /* SRAM */
@@ -186,8 +281,8 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, err);
         return;
     }
-    memory_region_add_subregion(get_system_memory(), ASPEED_SOC_SRAM_BASE,
-                                &s->sram);
+    memory_region_add_subregion(get_system_memory(),
+                                sc->info->memmap[ASPEED_SRAM], &s->sram);
 
     /* SCU */
     object_property_set_bool(OBJECT(&s->scu), true, "realized", &err);
@@ -195,7 +290,7 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->scu), 0, ASPEED_SOC_SCU_BASE);
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->scu), 0, sc->info->memmap[ASPEED_SCU]);
 
     /* VIC */
     object_property_set_bool(OBJECT(&s->vic), true, "realized", &err);
@@ -203,29 +298,39 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->vic), 0, ASPEED_SOC_VIC_BASE);
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->vic), 0, sc->info->memmap[ASPEED_VIC]);
     sysbus_connect_irq(SYS_BUS_DEVICE(&s->vic), 0,
                        qdev_get_gpio_in(DEVICE(&s->cpu), ARM_CPU_IRQ));
     sysbus_connect_irq(SYS_BUS_DEVICE(&s->vic), 1,
                        qdev_get_gpio_in(DEVICE(&s->cpu), ARM_CPU_FIQ));
 
+    /* RTC */
+    object_property_set_bool(OBJECT(&s->rtc), true, "realized", &err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
+    }
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->rtc), 0, sc->info->memmap[ASPEED_RTC]);
+    sysbus_connect_irq(SYS_BUS_DEVICE(&s->rtc), 0,
+                       aspeed_soc_get_irq(s, ASPEED_RTC));
+
     /* Timer */
     object_property_set_bool(OBJECT(&s->timerctrl), true, "realized", &err);
     if (err) {
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->timerctrl), 0, ASPEED_SOC_TIMER_BASE);
-    for (i = 0; i < ARRAY_SIZE(timer_irqs); i++) {
-        qemu_irq irq = qdev_get_gpio_in(DEVICE(&s->vic), timer_irqs[i]);
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->timerctrl), 0,
+                    sc->info->memmap[ASPEED_TIMER1]);
+    for (i = 0; i < ASPEED_TIMER_NR_TIMERS; i++) {
+        qemu_irq irq = aspeed_soc_get_irq(s, ASPEED_TIMER1 + i);
         sysbus_connect_irq(SYS_BUS_DEVICE(&s->timerctrl), i, irq);
     }
 
     /* UART - attach an 8250 to the IO space as our UART5 */
     if (serial_hd(0)) {
-        qemu_irq uart5 = qdev_get_gpio_in(DEVICE(&s->vic), uart_irqs[4]);
-        serial_mm_init(get_system_memory(),
-                       ASPEED_SOC_IOMEM_BASE + ASPEED_SOC_UART_5_BASE, 2,
+        qemu_irq uart5 = aspeed_soc_get_irq(s, ASPEED_UART5);
+        serial_mm_init(get_system_memory(), sc->info->memmap[ASPEED_UART5], 2,
                        uart5, 38400, serial_hd(0), DEVICE_LITTLE_ENDIAN);
     }
 
@@ -235,21 +340,27 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->i2c), 0, ASPEED_SOC_I2C_BASE);
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->i2c), 0, sc->info->memmap[ASPEED_I2C]);
     sysbus_connect_irq(SYS_BUS_DEVICE(&s->i2c), 0,
-                       qdev_get_gpio_in(DEVICE(&s->vic), 12));
+                       aspeed_soc_get_irq(s, ASPEED_I2C));
 
     /* FMC, The number of CS is set at the board level */
+    object_property_set_int(OBJECT(&s->fmc), sc->info->memmap[ASPEED_SDRAM],
+                            "sdram-base", &err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
+    }
     object_property_set_bool(OBJECT(&s->fmc), true, "realized", &err);
     if (err) {
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->fmc), 0, ASPEED_SOC_FMC_BASE);
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->fmc), 0, sc->info->memmap[ASPEED_FMC]);
     sysbus_mmio_map(SYS_BUS_DEVICE(&s->fmc), 1,
                     s->fmc.ctrl->flash_window_base);
     sysbus_connect_irq(SYS_BUS_DEVICE(&s->fmc), 0,
-                       qdev_get_gpio_in(DEVICE(&s->vic), 19));
+                       aspeed_soc_get_irq(s, ASPEED_FMC));
 
     /* SPI */
     for (i = 0; i < sc->info->spis_num; i++) {
@@ -261,7 +372,8 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
             error_propagate(errp, err);
             return;
         }
-        sysbus_mmio_map(SYS_BUS_DEVICE(&s->spi[i]), 0, sc->info->spi_bases[i]);
+        sysbus_mmio_map(SYS_BUS_DEVICE(&s->spi[i]), 0,
+                        sc->info->memmap[ASPEED_SPI1 + i]);
         sysbus_mmio_map(SYS_BUS_DEVICE(&s->spi[i]), 1,
                         s->spi[i].ctrl->flash_window_base);
     }
@@ -272,7 +384,7 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->sdmc), 0, ASPEED_SOC_SDMC_BASE);
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->sdmc), 0, sc->info->memmap[ASPEED_SDMC]);
 
     /* Watch dog */
     for (i = 0; i < sc->info->wdts_num; i++) {
@@ -282,23 +394,42 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)
             return;
         }
         sysbus_mmio_map(SYS_BUS_DEVICE(&s->wdt[i]), 0,
-                        ASPEED_SOC_WDT_BASE + i * 0x20);
+                        sc->info->memmap[ASPEED_WDT] + i * 0x20);
     }
 
     /* Net */
-    qdev_set_nic_properties(DEVICE(&s->ftgmac100), &nd_table[0]);
-    object_property_set_bool(OBJECT(&s->ftgmac100), true, "aspeed", &err);
-    object_property_set_bool(OBJECT(&s->ftgmac100), true, "realized",
-                             &local_err);
-    error_propagate(&err, local_err);
+    for (i = 0; i < nb_nics; i++) {
+        qdev_set_nic_properties(DEVICE(&s->ftgmac100[i]), &nd_table[i]);
+        object_property_set_bool(OBJECT(&s->ftgmac100[i]), true, "aspeed",
+                                 &err);
+        object_property_set_bool(OBJECT(&s->ftgmac100[i]), true, "realized",
+                                 &local_err);
+        error_propagate(&err, local_err);
+        if (err) {
+            error_propagate(errp, err);
+           return;
+        }
+        sysbus_mmio_map(SYS_BUS_DEVICE(&s->ftgmac100[i]), 0,
+                        sc->info->memmap[ASPEED_ETH1 + i]);
+        sysbus_connect_irq(SYS_BUS_DEVICE(&s->ftgmac100[i]), 0,
+                           aspeed_soc_get_irq(s, ASPEED_ETH1 + i));
+    }
+
+    /* XDMA */
+    object_property_set_bool(OBJECT(&s->xdma), true, "realized", &err);
     if (err) {
         error_propagate(errp, err);
         return;
     }
-    sysbus_mmio_map(SYS_BUS_DEVICE(&s->ftgmac100), 0, ASPEED_SOC_ETH1_BASE);
-    sysbus_connect_irq(SYS_BUS_DEVICE(&s->ftgmac100), 0,
-                       qdev_get_gpio_in(DEVICE(&s->vic), 2));
+    sysbus_mmio_map(SYS_BUS_DEVICE(&s->xdma), 0,
+                    sc->info->memmap[ASPEED_XDMA]);
+    sysbus_connect_irq(SYS_BUS_DEVICE(&s->xdma), 0,
+                       aspeed_soc_get_irq(s, ASPEED_XDMA));
 }
+static Property aspeed_soc_properties[] = {
+    DEFINE_PROP_UINT32("num-cpus", AspeedSoCState, num_cpus, 0),
+    DEFINE_PROP_END_OF_LIST(),
+};
 
 static void aspeed_soc_class_init(ObjectClass *oc, void *data)
 {
@@ -309,6 +440,7 @@ static void aspeed_soc_class_init(ObjectClass *oc, void *data)
     dc->realize = aspeed_soc_realize;
     /* Reason: Uses serial_hds and nd_table in realize() directly */
     dc->user_creatable = false;
+    dc->props = aspeed_soc_properties;
 }
 
 static const TypeInfo aspeed_soc_type_info = {
diff --git a/hw/arm/boot.c b/hw/arm/boot.c
index b2f93f6bef..1fb24fbef2 100644
--- a/hw/arm/boot.c
+++ b/hw/arm/boot.c
@@ -1109,10 +1109,11 @@ static void arm_setup_direct_kernel_boot(ARMCPU *cpu,
                              info->initrd_filename);
                 exit(1);
             }
-            if (info->initrd_start + initrd_size > info->ram_size) {
+            if (info->initrd_start + initrd_size > ram_end) {
                 error_report("could not load initrd '%s': "
                              "too big to fit into RAM after the kernel",
                              info->initrd_filename);
+                exit(1);
             }
         } else {
             initrd_size = 0;
diff --git a/hw/arm/fsl-imx7.c b/hw/arm/fsl-imx7.c
index b7e3526b4f..2eddf3f25c 100644
--- a/hw/arm/fsl-imx7.c
+++ b/hw/arm/fsl-imx7.c
@@ -526,6 +526,17 @@ static void fsl_imx7_realize(DeviceState *dev, Error **errp)
      */
     create_unimplemented_device("lcdif", FSL_IMX7_LCDIF_ADDR,
                                 FSL_IMX7_LCDIF_SIZE);
+
+    /*
+     * DMA APBH
+     */
+    create_unimplemented_device("dma-apbh", FSL_IMX7_DMA_APBH_ADDR,
+                                FSL_IMX7_DMA_APBH_SIZE);
+    /*
+     * PCIe PHY
+     */
+    create_unimplemented_device("pcie-phy", FSL_IMX7_PCIE_PHY_ADDR,
+                                FSL_IMX7_PCIE_PHY_SIZE);
 }
 
 static void fsl_imx7_class_init(ObjectClass *oc, void *data)
diff --git a/hw/arm/msf2-som.c b/hw/arm/msf2-som.c
index 8c550a8bdd..2c9984bb3b 100644
--- a/hw/arm/msf2-som.c
+++ b/hw/arm/msf2-som.c
@@ -53,6 +53,7 @@ static void emcraft_sf2_s2s010_init(MachineState *machine)
     if (strcmp(machine->cpu_type, mc->default_cpu_type) != 0) {
         error_report("This board can only be used with CPU %s",
                      mc->default_cpu_type);
+        exit(1);
     }
 
     memory_region_init_ram(ddr, NULL, "ddr-ram", DDR_SIZE,
diff --git a/hw/arm/sbsa-ref.c b/hw/arm/sbsa-ref.c
new file mode 100644
index 0000000000..ee53f0ff60
--- /dev/null
+++ b/hw/arm/sbsa-ref.c
@@ -0,0 +1,806 @@
+/*
+ * ARM SBSA Reference Platform emulation
+ *
+ * Copyright (c) 2018 Linaro Limited
+ * Written by Hongbo Zhang <hongbo.zhang@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 "qemu/osdep.h"
+#include "qemu-common.h"
+#include "qapi/error.h"
+#include "qemu/error-report.h"
+#include "qemu/units.h"
+#include "sysemu/device_tree.h"
+#include "sysemu/numa.h"
+#include "sysemu/sysemu.h"
+#include "exec/address-spaces.h"
+#include "exec/hwaddr.h"
+#include "kvm_arm.h"
+#include "hw/arm/boot.h"
+#include "hw/block/flash.h"
+#include "hw/boards.h"
+#include "hw/ide/internal.h"
+#include "hw/ide/ahci_internal.h"
+#include "hw/intc/arm_gicv3_common.h"
+#include "hw/loader.h"
+#include "hw/pci-host/gpex.h"
+#include "hw/usb.h"
+#include "net/net.h"
+
+#define RAMLIMIT_GB 8192
+#define RAMLIMIT_BYTES (RAMLIMIT_GB * GiB)
+
+#define NUM_IRQS        256
+#define NUM_SMMU_IRQS   4
+#define NUM_SATA_PORTS  6
+
+#define VIRTUAL_PMU_IRQ        7
+#define ARCH_GIC_MAINT_IRQ     9
+#define ARCH_TIMER_VIRT_IRQ    11
+#define ARCH_TIMER_S_EL1_IRQ   13
+#define ARCH_TIMER_NS_EL1_IRQ  14
+#define ARCH_TIMER_NS_EL2_IRQ  10
+
+enum {
+    SBSA_FLASH,
+    SBSA_MEM,
+    SBSA_CPUPERIPHS,
+    SBSA_GIC_DIST,
+    SBSA_GIC_REDIST,
+    SBSA_SMMU,
+    SBSA_UART,
+    SBSA_RTC,
+    SBSA_PCIE,
+    SBSA_PCIE_MMIO,
+    SBSA_PCIE_MMIO_HIGH,
+    SBSA_PCIE_PIO,
+    SBSA_PCIE_ECAM,
+    SBSA_GPIO,
+    SBSA_SECURE_UART,
+    SBSA_SECURE_UART_MM,
+    SBSA_SECURE_MEM,
+    SBSA_AHCI,
+    SBSA_EHCI,
+};
+
+typedef struct MemMapEntry {
+    hwaddr base;
+    hwaddr size;
+} MemMapEntry;
+
+typedef struct {
+    MachineState parent;
+    struct arm_boot_info bootinfo;
+    int smp_cpus;
+    void *fdt;
+    int fdt_size;
+    int psci_conduit;
+    PFlashCFI01 *flash[2];
+} SBSAMachineState;
+
+#define TYPE_SBSA_MACHINE   MACHINE_TYPE_NAME("sbsa-ref")
+#define SBSA_MACHINE(obj) \
+    OBJECT_CHECK(SBSAMachineState, (obj), TYPE_SBSA_MACHINE)
+
+static const MemMapEntry sbsa_ref_memmap[] = {
+    /* 512M boot ROM */
+    [SBSA_FLASH] =              {          0, 0x20000000 },
+    /* 512M secure memory */
+    [SBSA_SECURE_MEM] =         { 0x20000000, 0x20000000 },
+    /* Space reserved for CPU peripheral devices */
+    [SBSA_CPUPERIPHS] =         { 0x40000000, 0x00040000 },
+    [SBSA_GIC_DIST] =           { 0x40060000, 0x00010000 },
+    [SBSA_GIC_REDIST] =         { 0x40080000, 0x04000000 },
+    [SBSA_UART] =               { 0x60000000, 0x00001000 },
+    [SBSA_RTC] =                { 0x60010000, 0x00001000 },
+    [SBSA_GPIO] =               { 0x60020000, 0x00001000 },
+    [SBSA_SECURE_UART] =        { 0x60030000, 0x00001000 },
+    [SBSA_SECURE_UART_MM] =     { 0x60040000, 0x00001000 },
+    [SBSA_SMMU] =               { 0x60050000, 0x00020000 },
+    /* Space here reserved for more SMMUs */
+    [SBSA_AHCI] =               { 0x60100000, 0x00010000 },
+    [SBSA_EHCI] =               { 0x60110000, 0x00010000 },
+    /* Space here reserved for other devices */
+    [SBSA_PCIE_PIO] =           { 0x7fff0000, 0x00010000 },
+    /* 32-bit address PCIE MMIO space */
+    [SBSA_PCIE_MMIO] =          { 0x80000000, 0x70000000 },
+    /* 256M PCIE ECAM space */
+    [SBSA_PCIE_ECAM] =          { 0xf0000000, 0x10000000 },
+    /* ~1TB PCIE MMIO space (4GB to 1024GB boundary) */
+    [SBSA_PCIE_MMIO_HIGH] =     { 0x100000000ULL, 0xFF00000000ULL },
+    [SBSA_MEM] =                { 0x10000000000ULL, RAMLIMIT_BYTES },
+};
+
+static const int sbsa_ref_irqmap[] = {
+    [SBSA_UART] = 1,
+    [SBSA_RTC] = 2,
+    [SBSA_PCIE] = 3, /* ... to 6 */
+    [SBSA_GPIO] = 7,
+    [SBSA_SECURE_UART] = 8,
+    [SBSA_SECURE_UART_MM] = 9,
+    [SBSA_AHCI] = 10,
+    [SBSA_EHCI] = 11,
+};
+
+/*
+ * Firmware on this machine only uses ACPI table to load OS, these limited
+ * device tree nodes are just to let firmware know the info which varies from
+ * command line parameters, so it is not necessary to be fully compatible
+ * with the kernel CPU and NUMA binding rules.
+ */
+static void create_fdt(SBSAMachineState *sms)
+{
+    void *fdt = create_device_tree(&sms->fdt_size);
+    const MachineState *ms = MACHINE(sms);
+    int cpu;
+
+    if (!fdt) {
+        error_report("create_device_tree() failed");
+        exit(1);
+    }
+
+    sms->fdt = fdt;
+
+    qemu_fdt_setprop_string(fdt, "/", "compatible", "linux,sbsa-ref");
+    qemu_fdt_setprop_cell(fdt, "/", "#address-cells", 0x2);
+    qemu_fdt_setprop_cell(fdt, "/", "#size-cells", 0x2);
+
+    if (have_numa_distance) {
+        int size = nb_numa_nodes * nb_numa_nodes * 3 * sizeof(uint32_t);
+        uint32_t *matrix = g_malloc0(size);
+        int idx, i, j;
+
+        for (i = 0; i < nb_numa_nodes; i++) {
+            for (j = 0; j < nb_numa_nodes; j++) {
+                idx = (i * nb_numa_nodes + j) * 3;
+                matrix[idx + 0] = cpu_to_be32(i);
+                matrix[idx + 1] = cpu_to_be32(j);
+                matrix[idx + 2] = cpu_to_be32(numa_info[i].distance[j]);
+            }
+        }
+
+        qemu_fdt_add_subnode(fdt, "/distance-map");
+        qemu_fdt_setprop(fdt, "/distance-map", "distance-matrix",
+                         matrix, size);
+        g_free(matrix);
+    }
+
+    qemu_fdt_add_subnode(sms->fdt, "/cpus");
+
+    for (cpu = sms->smp_cpus - 1; cpu >= 0; cpu--) {
+        char *nodename = g_strdup_printf("/cpus/cpu@%d", cpu);
+        ARMCPU *armcpu = ARM_CPU(qemu_get_cpu(cpu));
+        CPUState *cs = CPU(armcpu);
+
+        qemu_fdt_add_subnode(sms->fdt, nodename);
+
+        if (ms->possible_cpus->cpus[cs->cpu_index].props.has_node_id) {
+            qemu_fdt_setprop_cell(sms->fdt, nodename, "numa-node-id",
+                ms->possible_cpus->cpus[cs->cpu_index].props.node_id);
+        }
+
+        g_free(nodename);
+    }
+}
+
+#define SBSA_FLASH_SECTOR_SIZE (256 * KiB)
+
+static PFlashCFI01 *sbsa_flash_create1(SBSAMachineState *sms,
+                                        const char *name,
+                                        const char *alias_prop_name)
+{
+    /*
+     * Create a single flash device.  We use the same parameters as
+     * the flash devices on the Versatile Express board.
+     */
+    DeviceState *dev = qdev_create(NULL, TYPE_PFLASH_CFI01);
+
+    qdev_prop_set_uint64(dev, "sector-length", SBSA_FLASH_SECTOR_SIZE);
+    qdev_prop_set_uint8(dev, "width", 4);
+    qdev_prop_set_uint8(dev, "device-width", 2);
+    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);
+    qdev_prop_set_uint16(dev, "id3", 0x00);
+    qdev_prop_set_string(dev, "name", name);
+    object_property_add_child(OBJECT(sms), name, OBJECT(dev),
+                              &error_abort);
+    object_property_add_alias(OBJECT(sms), alias_prop_name,
+                              OBJECT(dev), "drive", &error_abort);
+    return PFLASH_CFI01(dev);
+}
+
+static void sbsa_flash_create(SBSAMachineState *sms)
+{
+    sms->flash[0] = sbsa_flash_create1(sms, "sbsa.flash0", "pflash0");
+    sms->flash[1] = sbsa_flash_create1(sms, "sbsa.flash1", "pflash1");
+}
+
+static void sbsa_flash_map1(PFlashCFI01 *flash,
+                            hwaddr base, hwaddr size,
+                            MemoryRegion *sysmem)
+{
+    DeviceState *dev = DEVICE(flash);
+
+    assert(size % SBSA_FLASH_SECTOR_SIZE == 0);
+    assert(size / SBSA_FLASH_SECTOR_SIZE <= UINT32_MAX);
+    qdev_prop_set_uint32(dev, "num-blocks", size / SBSA_FLASH_SECTOR_SIZE);
+    qdev_init_nofail(dev);
+
+    memory_region_add_subregion(sysmem, base,
+                                sysbus_mmio_get_region(SYS_BUS_DEVICE(dev),
+                                                       0));
+}
+
+static void sbsa_flash_map(SBSAMachineState *sms,
+                           MemoryRegion *sysmem,
+                           MemoryRegion *secure_sysmem)
+{
+    /*
+     * Map two flash devices to fill the SBSA_FLASH space in the memmap.
+     * sysmem is the system memory space. secure_sysmem is the secure view
+     * of the system, and the first flash device should be made visible only
+     * there. The second flash device is visible to both secure and nonsecure.
+     * If sysmem == secure_sysmem this means there is no separate Secure
+     * address space and both flash devices are generally visible.
+     */
+    hwaddr flashsize = sbsa_ref_memmap[SBSA_FLASH].size / 2;
+    hwaddr flashbase = sbsa_ref_memmap[SBSA_FLASH].base;
+
+    sbsa_flash_map1(sms->flash[0], flashbase, flashsize,
+                    secure_sysmem);
+    sbsa_flash_map1(sms->flash[1], flashbase + flashsize, flashsize,
+                    sysmem);
+}
+
+static bool sbsa_firmware_init(SBSAMachineState *sms,
+                               MemoryRegion *sysmem,
+                               MemoryRegion *secure_sysmem)
+{
+    int i;
+    BlockBackend *pflash_blk0;
+
+    /* Map legacy -drive if=pflash to machine properties */
+    for (i = 0; i < ARRAY_SIZE(sms->flash); i++) {
+        pflash_cfi01_legacy_drive(sms->flash[i],
+                                  drive_get(IF_PFLASH, 0, i));
+    }
+
+    sbsa_flash_map(sms, sysmem, secure_sysmem);
+
+    pflash_blk0 = pflash_cfi01_get_blk(sms->flash[0]);
+
+    if (bios_name) {
+        char *fname;
+        MemoryRegion *mr;
+        int image_size;
+
+        if (pflash_blk0) {
+            error_report("The contents of the first flash device may be "
+                         "specified with -bios or with -drive if=pflash... "
+                         "but you cannot use both options at once");
+            exit(1);
+        }
+
+        /* Fall back to -bios */
+
+        fname = qemu_find_file(QEMU_FILE_TYPE_BIOS, bios_name);
+        if (!fname) {
+            error_report("Could not find ROM image '%s'", bios_name);
+            exit(1);
+        }
+        mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(sms->flash[0]), 0);
+        image_size = load_image_mr(fname, mr);
+        g_free(fname);
+        if (image_size < 0) {
+            error_report("Could not load ROM image '%s'", bios_name);
+            exit(1);
+        }
+    }
+
+    return pflash_blk0 || bios_name;
+}
+
+static void create_secure_ram(SBSAMachineState *sms,
+                              MemoryRegion *secure_sysmem)
+{
+    MemoryRegion *secram = g_new(MemoryRegion, 1);
+    hwaddr base = sbsa_ref_memmap[SBSA_SECURE_MEM].base;
+    hwaddr size = sbsa_ref_memmap[SBSA_SECURE_MEM].size;
+
+    memory_region_init_ram(secram, NULL, "sbsa-ref.secure-ram", size,
+                           &error_fatal);
+    memory_region_add_subregion(secure_sysmem, base, secram);
+}
+
+static void create_gic(SBSAMachineState *sms, qemu_irq *pic)
+{
+    DeviceState *gicdev;
+    SysBusDevice *gicbusdev;
+    const char *gictype;
+    uint32_t redist0_capacity, redist0_count;
+    int i;
+
+    gictype = gicv3_class_name();
+
+    gicdev = qdev_create(NULL, gictype);
+    qdev_prop_set_uint32(gicdev, "revision", 3);
+    qdev_prop_set_uint32(gicdev, "num-cpu", smp_cpus);
+    /*
+     * Note that the num-irq property counts both internal and external
+     * interrupts; there are always 32 of the former (mandated by GIC spec).
+     */
+    qdev_prop_set_uint32(gicdev, "num-irq", NUM_IRQS + 32);
+    qdev_prop_set_bit(gicdev, "has-security-extensions", true);
+
+    redist0_capacity =
+                sbsa_ref_memmap[SBSA_GIC_REDIST].size / GICV3_REDIST_SIZE;
+    redist0_count = MIN(smp_cpus, redist0_capacity);
+
+    qdev_prop_set_uint32(gicdev, "len-redist-region-count", 1);
+    qdev_prop_set_uint32(gicdev, "redist-region-count[0]", redist0_count);
+
+    qdev_init_nofail(gicdev);
+    gicbusdev = SYS_BUS_DEVICE(gicdev);
+    sysbus_mmio_map(gicbusdev, 0, sbsa_ref_memmap[SBSA_GIC_DIST].base);
+    sysbus_mmio_map(gicbusdev, 1, sbsa_ref_memmap[SBSA_GIC_REDIST].base);
+
+    /*
+     * Wire the outputs from each CPU's generic timer and the GICv3
+     * maintenance interrupt signal to the appropriate GIC PPI inputs,
+     * and the GIC's IRQ/FIQ/VIRQ/VFIQ interrupt outputs to the CPU's inputs.
+     */
+    for (i = 0; i < smp_cpus; i++) {
+        DeviceState *cpudev = DEVICE(qemu_get_cpu(i));
+        int ppibase = NUM_IRQS + i * GIC_INTERNAL + GIC_NR_SGIS;
+        int irq;
+        /*
+         * Mapping from the output timer irq lines from the CPU to the
+         * GIC PPI inputs used for this board.
+         */
+        const int timer_irq[] = {
+            [GTIMER_PHYS] = ARCH_TIMER_NS_EL1_IRQ,
+            [GTIMER_VIRT] = ARCH_TIMER_VIRT_IRQ,
+            [GTIMER_HYP]  = ARCH_TIMER_NS_EL2_IRQ,
+            [GTIMER_SEC]  = ARCH_TIMER_S_EL1_IRQ,
+        };
+
+        for (irq = 0; irq < ARRAY_SIZE(timer_irq); irq++) {
+            qdev_connect_gpio_out(cpudev, irq,
+                                  qdev_get_gpio_in(gicdev,
+                                                   ppibase + timer_irq[irq]));
+        }
+
+        qdev_connect_gpio_out_named(cpudev, "gicv3-maintenance-interrupt", 0,
+                                    qdev_get_gpio_in(gicdev, ppibase
+                                                     + ARCH_GIC_MAINT_IRQ));
+        qdev_connect_gpio_out_named(cpudev, "pmu-interrupt", 0,
+                                    qdev_get_gpio_in(gicdev, ppibase
+                                                     + VIRTUAL_PMU_IRQ));
+
+        sysbus_connect_irq(gicbusdev, i, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ));
+        sysbus_connect_irq(gicbusdev, i + smp_cpus,
+                           qdev_get_gpio_in(cpudev, ARM_CPU_FIQ));
+        sysbus_connect_irq(gicbusdev, i + 2 * smp_cpus,
+                           qdev_get_gpio_in(cpudev, ARM_CPU_VIRQ));
+        sysbus_connect_irq(gicbusdev, i + 3 * smp_cpus,
+                           qdev_get_gpio_in(cpudev, ARM_CPU_VFIQ));
+    }
+
+    for (i = 0; i < NUM_IRQS; i++) {
+        pic[i] = qdev_get_gpio_in(gicdev, i);
+    }
+}
+
+static void create_uart(const SBSAMachineState *sms, qemu_irq *pic, int uart,
+                        MemoryRegion *mem, Chardev *chr)
+{
+    hwaddr base = sbsa_ref_memmap[uart].base;
+    int irq = sbsa_ref_irqmap[uart];
+    DeviceState *dev = qdev_create(NULL, "pl011");
+    SysBusDevice *s = SYS_BUS_DEVICE(dev);
+
+    qdev_prop_set_chr(dev, "chardev", chr);
+    qdev_init_nofail(dev);
+    memory_region_add_subregion(mem, base,
+                                sysbus_mmio_get_region(s, 0));
+    sysbus_connect_irq(s, 0, pic[irq]);
+}
+
+static void create_rtc(const SBSAMachineState *sms, qemu_irq *pic)
+{
+    hwaddr base = sbsa_ref_memmap[SBSA_RTC].base;
+    int irq = sbsa_ref_irqmap[SBSA_RTC];
+
+    sysbus_create_simple("pl031", base, pic[irq]);
+}
+
+static DeviceState *gpio_key_dev;
+static void sbsa_ref_powerdown_req(Notifier *n, void *opaque)
+{
+    /* use gpio Pin 3 for power button event */
+    qemu_set_irq(qdev_get_gpio_in(gpio_key_dev, 0), 1);
+}
+
+static Notifier sbsa_ref_powerdown_notifier = {
+    .notify = sbsa_ref_powerdown_req
+};
+
+static void create_gpio(const SBSAMachineState *sms, qemu_irq *pic)
+{
+    DeviceState *pl061_dev;
+    hwaddr base = sbsa_ref_memmap[SBSA_GPIO].base;
+    int irq = sbsa_ref_irqmap[SBSA_GPIO];
+
+    pl061_dev = sysbus_create_simple("pl061", base, pic[irq]);
+
+    gpio_key_dev = sysbus_create_simple("gpio-key", -1,
+                                        qdev_get_gpio_in(pl061_dev, 3));
+
+    /* connect powerdown request */
+    qemu_register_powerdown_notifier(&sbsa_ref_powerdown_notifier);
+}
+
+static void create_ahci(const SBSAMachineState *sms, qemu_irq *pic)
+{
+    hwaddr base = sbsa_ref_memmap[SBSA_AHCI].base;
+    int irq = sbsa_ref_irqmap[SBSA_AHCI];
+    DeviceState *dev;
+    DriveInfo *hd[NUM_SATA_PORTS];
+    SysbusAHCIState *sysahci;
+    AHCIState *ahci;
+    int i;
+
+    dev = qdev_create(NULL, "sysbus-ahci");
+    qdev_prop_set_uint32(dev, "num-ports", NUM_SATA_PORTS);
+    qdev_init_nofail(dev);
+    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base);
+    sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irq]);
+
+    sysahci = SYSBUS_AHCI(dev);
+    ahci = &sysahci->ahci;
+    ide_drive_get(hd, ARRAY_SIZE(hd));
+    for (i = 0; i < ahci->ports; i++) {
+        if (hd[i] == NULL) {
+            continue;
+        }
+        ide_create_drive(&ahci->dev[i].port, 0, hd[i]);
+    }
+}
+
+static void create_ehci(const SBSAMachineState *sms, qemu_irq *pic)
+{
+    hwaddr base = sbsa_ref_memmap[SBSA_EHCI].base;
+    int irq = sbsa_ref_irqmap[SBSA_EHCI];
+
+    sysbus_create_simple("platform-ehci-usb", base, pic[irq]);
+}
+
+static void create_smmu(const SBSAMachineState *sms, qemu_irq *pic,
+                        PCIBus *bus)
+{
+    hwaddr base = sbsa_ref_memmap[SBSA_SMMU].base;
+    int irq =  sbsa_ref_irqmap[SBSA_SMMU];
+    DeviceState *dev;
+    int i;
+
+    dev = qdev_create(NULL, "arm-smmuv3");
+
+    object_property_set_link(OBJECT(dev), OBJECT(bus), "primary-bus",
+                             &error_abort);
+    qdev_init_nofail(dev);
+    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, base);
+    for (i = 0; i < NUM_SMMU_IRQS; i++) {
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]);
+    }
+}
+
+static void create_pcie(SBSAMachineState *sms, qemu_irq *pic)
+{
+    hwaddr base_ecam = sbsa_ref_memmap[SBSA_PCIE_ECAM].base;
+    hwaddr size_ecam = sbsa_ref_memmap[SBSA_PCIE_ECAM].size;
+    hwaddr base_mmio = sbsa_ref_memmap[SBSA_PCIE_MMIO].base;
+    hwaddr size_mmio = sbsa_ref_memmap[SBSA_PCIE_MMIO].size;
+    hwaddr base_mmio_high = sbsa_ref_memmap[SBSA_PCIE_MMIO_HIGH].base;
+    hwaddr size_mmio_high = sbsa_ref_memmap[SBSA_PCIE_MMIO_HIGH].size;
+    hwaddr base_pio = sbsa_ref_memmap[SBSA_PCIE_PIO].base;
+    int irq = sbsa_ref_irqmap[SBSA_PCIE];
+    MemoryRegion *mmio_alias, *mmio_alias_high, *mmio_reg;
+    MemoryRegion *ecam_alias, *ecam_reg;
+    DeviceState *dev;
+    PCIHostState *pci;
+    int i;
+
+    dev = qdev_create(NULL, TYPE_GPEX_HOST);
+    qdev_init_nofail(dev);
+
+    /* Map ECAM space */
+    ecam_alias = g_new0(MemoryRegion, 1);
+    ecam_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
+    memory_region_init_alias(ecam_alias, OBJECT(dev), "pcie-ecam",
+                             ecam_reg, 0, size_ecam);
+    memory_region_add_subregion(get_system_memory(), base_ecam, ecam_alias);
+
+    /* Map the MMIO space */
+    mmio_alias = g_new0(MemoryRegion, 1);
+    mmio_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 1);
+    memory_region_init_alias(mmio_alias, OBJECT(dev), "pcie-mmio",
+                             mmio_reg, base_mmio, size_mmio);
+    memory_region_add_subregion(get_system_memory(), base_mmio, mmio_alias);
+
+    /* Map the MMIO_HIGH space */
+    mmio_alias_high = g_new0(MemoryRegion, 1);
+    memory_region_init_alias(mmio_alias_high, OBJECT(dev), "pcie-mmio-high",
+                             mmio_reg, base_mmio_high, size_mmio_high);
+    memory_region_add_subregion(get_system_memory(), base_mmio_high,
+                                mmio_alias_high);
+
+    /* Map IO port space */
+    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 2, base_pio);
+
+    for (i = 0; i < GPEX_NUM_IRQS; i++) {
+        sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]);
+        gpex_set_irq_num(GPEX_HOST(dev), i, irq + i);
+    }
+
+    pci = PCI_HOST_BRIDGE(dev);
+    if (pci->bus) {
+        for (i = 0; i < nb_nics; i++) {
+            NICInfo *nd = &nd_table[i];
+
+            if (!nd->model) {
+                nd->model = g_strdup("e1000e");
+            }
+
+            pci_nic_init_nofail(nd, pci->bus, nd->model, NULL);
+        }
+    }
+
+    pci_create_simple(pci->bus, -1, "VGA");
+
+    create_smmu(sms, pic, pci->bus);
+}
+
+static void *sbsa_ref_dtb(const struct arm_boot_info *binfo, int *fdt_size)
+{
+    const SBSAMachineState *board = container_of(binfo, SBSAMachineState,
+                                                 bootinfo);
+
+    *fdt_size = board->fdt_size;
+    return board->fdt;
+}
+
+static void sbsa_ref_init(MachineState *machine)
+{
+    SBSAMachineState *sms = SBSA_MACHINE(machine);
+    MachineClass *mc = MACHINE_GET_CLASS(machine);
+    MemoryRegion *sysmem = get_system_memory();
+    MemoryRegion *secure_sysmem = NULL;
+    MemoryRegion *ram = g_new(MemoryRegion, 1);
+    bool firmware_loaded;
+    const CPUArchIdList *possible_cpus;
+    int n, sbsa_max_cpus;
+    qemu_irq pic[NUM_IRQS];
+
+    if (strcmp(machine->cpu_type, ARM_CPU_TYPE_NAME("cortex-a57"))) {
+        error_report("sbsa-ref: CPU type other than the built-in "
+                     "cortex-a57 not supported");
+        exit(1);
+    }
+
+    if (kvm_enabled()) {
+        error_report("sbsa-ref: KVM is not supported for this machine");
+        exit(1);
+    }
+
+    /*
+     * The Secure view of the world is the same as the NonSecure,
+     * but with a few extra devices. Create it as a container region
+     * containing the system memory at low priority; any secure-only
+     * devices go in at higher priority and take precedence.
+     */
+    secure_sysmem = g_new(MemoryRegion, 1);
+    memory_region_init(secure_sysmem, OBJECT(machine), "secure-memory",
+                       UINT64_MAX);
+    memory_region_add_subregion_overlap(secure_sysmem, 0, sysmem, -1);
+
+    firmware_loaded = sbsa_firmware_init(sms, sysmem,
+                                         secure_sysmem ?: sysmem);
+
+    if (machine->kernel_filename && firmware_loaded) {
+        error_report("sbsa-ref: No fw_cfg device on this machine, "
+                     "so -kernel option is not supported when firmware loaded, "
+                     "please load OS from hard disk instead");
+        exit(1);
+    }
+
+    /*
+     * This machine has EL3 enabled, external firmware should supply PSCI
+     * implementation, so the QEMU's internal PSCI is disabled.
+     */
+    sms->psci_conduit = QEMU_PSCI_CONDUIT_DISABLED;
+
+    sbsa_max_cpus = sbsa_ref_memmap[SBSA_GIC_REDIST].size / GICV3_REDIST_SIZE;
+
+    if (max_cpus > sbsa_max_cpus) {
+        error_report("Number of SMP CPUs requested (%d) exceeds max CPUs "
+                     "supported by machine 'sbsa-ref' (%d)",
+                     max_cpus, sbsa_max_cpus);
+        exit(1);
+    }
+
+    sms->smp_cpus = smp_cpus;
+
+    if (machine->ram_size > sbsa_ref_memmap[SBSA_MEM].size) {
+        error_report("sbsa-ref: cannot model more than %dGB RAM", RAMLIMIT_GB);
+        exit(1);
+    }
+
+    possible_cpus = mc->possible_cpu_arch_ids(machine);
+    for (n = 0; n < possible_cpus->len; n++) {
+        Object *cpuobj;
+        CPUState *cs;
+
+        if (n >= smp_cpus) {
+            break;
+        }
+
+        cpuobj = object_new(possible_cpus->cpus[n].type);
+        object_property_set_int(cpuobj, possible_cpus->cpus[n].arch_id,
+                                "mp-affinity", NULL);
+
+        cs = CPU(cpuobj);
+        cs->cpu_index = n;
+
+        numa_cpu_pre_plug(&possible_cpus->cpus[cs->cpu_index], DEVICE(cpuobj),
+                          &error_fatal);
+
+        if (object_property_find(cpuobj, "reset-cbar", NULL)) {
+            object_property_set_int(cpuobj,
+                                    sbsa_ref_memmap[SBSA_CPUPERIPHS].base,
+                                    "reset-cbar", &error_abort);
+        }
+
+        object_property_set_link(cpuobj, OBJECT(sysmem), "memory",
+                                 &error_abort);
+
+        object_property_set_link(cpuobj, OBJECT(secure_sysmem),
+                                 "secure-memory", &error_abort);
+
+        object_property_set_bool(cpuobj, true, "realized", &error_fatal);
+        object_unref(cpuobj);
+    }
+
+    memory_region_allocate_system_memory(ram, NULL, "sbsa-ref.ram",
+                                         machine->ram_size);
+    memory_region_add_subregion(sysmem, sbsa_ref_memmap[SBSA_MEM].base, ram);
+
+    create_fdt(sms);
+
+    create_secure_ram(sms, secure_sysmem);
+
+    create_gic(sms, pic);
+
+    create_uart(sms, pic, SBSA_UART, sysmem, serial_hd(0));
+    create_uart(sms, pic, SBSA_SECURE_UART, secure_sysmem, serial_hd(1));
+    /* Second secure UART for RAS and MM from EL0 */
+    create_uart(sms, pic, SBSA_SECURE_UART_MM, secure_sysmem, serial_hd(2));
+
+    create_rtc(sms, pic);
+
+    create_gpio(sms, pic);
+
+    create_ahci(sms, pic);
+
+    create_ehci(sms, pic);
+
+    create_pcie(sms, pic);
+
+    sms->bootinfo.ram_size = machine->ram_size;
+    sms->bootinfo.kernel_filename = machine->kernel_filename;
+    sms->bootinfo.nb_cpus = smp_cpus;
+    sms->bootinfo.board_id = -1;
+    sms->bootinfo.loader_start = sbsa_ref_memmap[SBSA_MEM].base;
+    sms->bootinfo.get_dtb = sbsa_ref_dtb;
+    sms->bootinfo.firmware_loaded = firmware_loaded;
+    arm_load_kernel(ARM_CPU(first_cpu), &sms->bootinfo);
+}
+
+static uint64_t sbsa_ref_cpu_mp_affinity(SBSAMachineState *sms, int idx)
+{
+    uint8_t clustersz = ARM_DEFAULT_CPUS_PER_CLUSTER;
+    return arm_cpu_mp_affinity(idx, clustersz);
+}
+
+static const CPUArchIdList *sbsa_ref_possible_cpu_arch_ids(MachineState *ms)
+{
+    SBSAMachineState *sms = SBSA_MACHINE(ms);
+    int n;
+
+    if (ms->possible_cpus) {
+        assert(ms->possible_cpus->len == max_cpus);
+        return ms->possible_cpus;
+    }
+
+    ms->possible_cpus = g_malloc0(sizeof(CPUArchIdList) +
+                                  sizeof(CPUArchId) * max_cpus);
+    ms->possible_cpus->len = max_cpus;
+    for (n = 0; n < ms->possible_cpus->len; n++) {
+        ms->possible_cpus->cpus[n].type = ms->cpu_type;
+        ms->possible_cpus->cpus[n].arch_id =
+            sbsa_ref_cpu_mp_affinity(sms, n);
+        ms->possible_cpus->cpus[n].props.has_thread_id = true;
+        ms->possible_cpus->cpus[n].props.thread_id = n;
+    }
+    return ms->possible_cpus;
+}
+
+static CpuInstanceProperties
+sbsa_ref_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
+{
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+    const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(ms);
+
+    assert(cpu_index < possible_cpus->len);
+    return possible_cpus->cpus[cpu_index].props;
+}
+
+static int64_t
+sbsa_ref_get_default_cpu_node_id(const MachineState *ms, int idx)
+{
+    return idx % nb_numa_nodes;
+}
+
+static void sbsa_ref_instance_init(Object *obj)
+{
+    SBSAMachineState *sms = SBSA_MACHINE(obj);
+
+    sbsa_flash_create(sms);
+}
+
+static void sbsa_ref_class_init(ObjectClass *oc, void *data)
+{
+    MachineClass *mc = MACHINE_CLASS(oc);
+
+    mc->init = sbsa_ref_init;
+    mc->desc = "QEMU 'SBSA Reference' ARM Virtual Machine";
+    mc->default_cpu_type = ARM_CPU_TYPE_NAME("cortex-a57");
+    mc->max_cpus = 512;
+    mc->pci_allow_0_address = true;
+    mc->minimum_page_bits = 12;
+    mc->block_default_type = IF_IDE;
+    mc->no_cdrom = 1;
+    mc->default_ram_size = 1 * GiB;
+    mc->default_cpus = 4;
+    mc->possible_cpu_arch_ids = sbsa_ref_possible_cpu_arch_ids;
+    mc->cpu_index_to_instance_props = sbsa_ref_cpu_index_to_props;
+    mc->get_default_cpu_node_id = sbsa_ref_get_default_cpu_node_id;
+}
+
+static const TypeInfo sbsa_ref_info = {
+    .name          = TYPE_SBSA_MACHINE,
+    .parent        = TYPE_MACHINE,
+    .instance_init = sbsa_ref_instance_init,
+    .class_init    = sbsa_ref_class_init,
+    .instance_size = sizeof(SBSAMachineState),
+};
+
+static void sbsa_ref_machine_init(void)
+{
+    type_register_static(&sbsa_ref_info);
+}
+
+type_init(sbsa_ref_machine_init);
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 431e2900fd..ed009fa447 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -176,6 +176,7 @@ static const int a15irqmap[] = {
 };
 
 static const char *valid_cpus[] = {
+    ARM_CPU_TYPE_NAME("cortex-a7"),
     ARM_CPU_TYPE_NAME("cortex-a15"),
     ARM_CPU_TYPE_NAME("cortex-a53"),
     ARM_CPU_TYPE_NAME("cortex-a57"),
diff --git a/hw/block/dataplane/xen-block.c b/hw/block/dataplane/xen-block.c
index f7ad452bbd..0f200c5fb0 100644
--- a/hw/block/dataplane/xen-block.c
+++ b/hw/block/dataplane/xen-block.c
@@ -58,6 +58,7 @@ struct XenBlockDataPlane {
     int requests_inflight;
     unsigned int max_requests;
     BlockBackend *blk;
+    unsigned int sector_size;
     QEMUBH *bh;
     IOThread *iothread;
     AioContext *ctx;
@@ -167,7 +168,7 @@ static int xen_block_parse_request(XenBlockRequest *request)
         goto err;
     }
 
-    request->start = request->req.sector_number * XEN_BLKIF_SECTOR_SIZE;
+    request->start = request->req.sector_number * dataplane->sector_size;
     for (i = 0; i < request->req.nr_segments; i++) {
         if (i == BLKIF_MAX_SEGMENTS_PER_REQUEST) {
             error_report("error: nr_segments too big");
@@ -177,14 +178,14 @@ static int xen_block_parse_request(XenBlockRequest *request)
             error_report("error: first > last sector");
             goto err;
         }
-        if (request->req.seg[i].last_sect * XEN_BLKIF_SECTOR_SIZE >=
+        if (request->req.seg[i].last_sect * dataplane->sector_size >=
             XC_PAGE_SIZE) {
             error_report("error: page crossing");
             goto err;
         }
 
         len = (request->req.seg[i].last_sect -
-               request->req.seg[i].first_sect + 1) * XEN_BLKIF_SECTOR_SIZE;
+               request->req.seg[i].first_sect + 1) * dataplane->sector_size;
         request->size += len;
     }
     if (request->start + request->size > blk_getlength(dataplane->blk)) {
@@ -218,17 +219,17 @@ static int xen_block_copy_request(XenBlockRequest *request)
         if (to_domain) {
             segs[i].dest.foreign.ref = request->req.seg[i].gref;
             segs[i].dest.foreign.offset = request->req.seg[i].first_sect *
-                XEN_BLKIF_SECTOR_SIZE;
+                dataplane->sector_size;
             segs[i].source.virt = virt;
         } else {
             segs[i].source.foreign.ref = request->req.seg[i].gref;
             segs[i].source.foreign.offset = request->req.seg[i].first_sect *
-                XEN_BLKIF_SECTOR_SIZE;
+                dataplane->sector_size;
             segs[i].dest.virt = virt;
         }
         segs[i].len = (request->req.seg[i].last_sect -
                        request->req.seg[i].first_sect + 1) *
-                      XEN_BLKIF_SECTOR_SIZE;
+                      dataplane->sector_size;
         virt += segs[i].len;
     }
 
@@ -317,7 +318,9 @@ static void xen_block_complete_aio(void *opaque, int ret)
     }
     xen_block_release_request(request);
 
-    qemu_bh_schedule(dataplane->bh);
+    if (dataplane->more_work) {
+        qemu_bh_schedule(dataplane->bh);
+    }
 
 done:
     aio_context_release(dataplane->ctx);
@@ -336,12 +339,12 @@ static bool xen_block_split_discard(XenBlockRequest *request,
 
     /* Wrap around, or overflowing byte limit? */
     if (sec_start + sec_count < sec_count ||
-        sec_start + sec_count > INT64_MAX / XEN_BLKIF_SECTOR_SIZE) {
+        sec_start + sec_count > INT64_MAX / dataplane->sector_size) {
         return false;
     }
 
-    byte_offset = sec_start * XEN_BLKIF_SECTOR_SIZE;
-    byte_remaining = sec_count * XEN_BLKIF_SECTOR_SIZE;
+    byte_offset = sec_start * dataplane->sector_size;
+    byte_remaining = sec_count * dataplane->sector_size;
 
     do {
         byte_chunk = byte_remaining > BDRV_REQUEST_MAX_BYTES ?
@@ -514,12 +517,13 @@ static int xen_block_get_request(XenBlockDataPlane *dataplane,
  */
 #define IO_PLUG_THRESHOLD 1
 
-static void xen_block_handle_requests(XenBlockDataPlane *dataplane)
+static bool xen_block_handle_requests(XenBlockDataPlane *dataplane)
 {
     RING_IDX rc, rp;
     XenBlockRequest *request;
     int inflight_atstart = dataplane->requests_inflight;
     int batched = 0;
+    bool done_something = false;
 
     dataplane->more_work = 0;
 
@@ -551,6 +555,7 @@ static void xen_block_handle_requests(XenBlockDataPlane *dataplane)
         }
         xen_block_get_request(dataplane, request, rc);
         dataplane->rings.common.req_cons = ++rc;
+        done_something = true;
 
         /* parse them */
         if (xen_block_parse_request(request) != 0) {
@@ -602,10 +607,7 @@ static void xen_block_handle_requests(XenBlockDataPlane *dataplane)
         blk_io_unplug(dataplane->blk);
     }
 
-    if (dataplane->more_work &&
-        dataplane->requests_inflight < dataplane->max_requests) {
-        qemu_bh_schedule(dataplane->bh);
-    }
+    return done_something;
 }
 
 static void xen_block_dataplane_bh(void *opaque)
@@ -617,21 +619,23 @@ static void xen_block_dataplane_bh(void *opaque)
     aio_context_release(dataplane->ctx);
 }
 
-static void xen_block_dataplane_event(void *opaque)
+static bool xen_block_dataplane_event(void *opaque)
 {
     XenBlockDataPlane *dataplane = opaque;
 
-    qemu_bh_schedule(dataplane->bh);
+    return xen_block_handle_requests(dataplane);
 }
 
 XenBlockDataPlane *xen_block_dataplane_create(XenDevice *xendev,
-                                              BlockConf *conf,
+                                              BlockBackend *blk,
+                                              unsigned int sector_size,
                                               IOThread *iothread)
 {
     XenBlockDataPlane *dataplane = g_new0(XenBlockDataPlane, 1);
 
     dataplane->xendev = xendev;
-    dataplane->blk = conf->blk;
+    dataplane->blk = blk;
+    dataplane->sector_size = sector_size;
 
     QLIST_INIT(&dataplane->inflight);
     QLIST_INIT(&dataplane->freelist);
@@ -803,7 +807,7 @@ void xen_block_dataplane_start(XenBlockDataPlane *dataplane,
     }
 
     dataplane->event_channel =
-        xen_device_bind_event_channel(xendev, event_channel,
+        xen_device_bind_event_channel(xendev, dataplane->ctx, event_channel,
                                       xen_block_dataplane_event, dataplane,
                                       &local_err);
     if (local_err) {
diff --git a/hw/block/dataplane/xen-block.h b/hw/block/dataplane/xen-block.h
index d6fa6d26dd..76dcd51c3d 100644
--- a/hw/block/dataplane/xen-block.h
+++ b/hw/block/dataplane/xen-block.h
@@ -15,7 +15,8 @@
 typedef struct XenBlockDataPlane XenBlockDataPlane;
 
 XenBlockDataPlane *xen_block_dataplane_create(XenDevice *xendev,
-                                              BlockConf *conf,
+                                              BlockBackend *blk,
+                                              unsigned int sector_size,
                                               IOThread *iothread);
 void xen_block_dataplane_destroy(XenBlockDataPlane *dataplane);
 void xen_block_dataplane_start(XenBlockDataPlane *dataplane,
diff --git a/hw/block/pflash_cfi01.c b/hw/block/pflash_cfi01.c
index 35080d915f..db4a246b22 100644
--- a/hw/block/pflash_cfi01.c
+++ b/hw/block/pflash_cfi01.c
@@ -248,7 +248,6 @@ static uint32_t pflash_data_read(PFlashCFI01 *pfl, hwaddr offset,
     switch (width) {
     case 1:
         ret = p[offset];
-        trace_pflash_data_read8(offset, ret);
         break;
     case 2:
         if (be) {
@@ -258,7 +257,6 @@ static uint32_t pflash_data_read(PFlashCFI01 *pfl, hwaddr offset,
             ret = p[offset];
             ret |= p[offset + 1] << 8;
         }
-        trace_pflash_data_read16(offset, ret);
         break;
     case 4:
         if (be) {
@@ -272,12 +270,12 @@ static uint32_t pflash_data_read(PFlashCFI01 *pfl, hwaddr offset,
             ret |= p[offset + 2] << 16;
             ret |= p[offset + 3] << 24;
         }
-        trace_pflash_data_read32(offset, ret);
         break;
     default:
         DPRINTF("BUG in %s\n", __func__);
         abort();
     }
+    trace_pflash_data_read(offset, width << 1, ret);
     return ret;
 }
 
@@ -288,7 +286,6 @@ static uint32_t pflash_read(PFlashCFI01 *pfl, hwaddr offset,
     uint32_t ret;
 
     ret = -1;
-    trace_pflash_read(offset, pfl->cmd, width, pfl->wcycle);
     switch (pfl->cmd) {
     default:
         /* This should never happen : reset state & treat it as a read */
@@ -391,6 +388,8 @@ static uint32_t pflash_read(PFlashCFI01 *pfl, hwaddr offset,
 
         break;
     }
+    trace_pflash_io_read(offset, width, width << 1, ret, pfl->cmd, pfl->wcycle);
+
     return ret;
 }
 
@@ -414,7 +413,7 @@ static inline void pflash_data_write(PFlashCFI01 *pfl, hwaddr offset,
 {
     uint8_t *p = pfl->storage;
 
-    trace_pflash_data_write(offset, value, width, pfl->counter);
+    trace_pflash_data_write(offset, width << 1, value, pfl->counter);
     switch (width) {
     case 1:
         p[offset] = value;
@@ -453,7 +452,7 @@ static void pflash_write(PFlashCFI01 *pfl, hwaddr offset,
 
     cmd = value;
 
-    trace_pflash_write(offset, value, width, pfl->wcycle);
+    trace_pflash_io_write(offset, width, width << 1, value, pfl->wcycle);
     if (!pfl->wcycle) {
         /* Set the device in I/O access mode */
         memory_region_rom_device_set_romd(&pfl->mem, false);
diff --git a/hw/block/pflash_cfi02.c b/hw/block/pflash_cfi02.c
index eb106f4996..5392290c72 100644
--- a/hw/block/pflash_cfi02.c
+++ b/hw/block/pflash_cfi02.c
@@ -29,10 +29,7 @@
  * - CFI queries
  *
  * It does not support flash interleaving.
- * It does not implement boot blocs with reduced size
  * It does not implement software data protection as found in many real chips
- * It does not implement erase suspend/resume commands
- * It does not implement multiple sectors erase
  */
 
 #include "qemu/osdep.h"
@@ -40,6 +37,7 @@
 #include "hw/block/block.h"
 #include "hw/block/flash.h"
 #include "qapi/error.h"
+#include "qemu/bitmap.h"
 #include "qemu/timer.h"
 #include "sysemu/block-backend.h"
 #include "qemu/host-utils.h"
@@ -47,26 +45,40 @@
 #include "hw/sysbus.h"
 #include "trace.h"
 
-//#define PFLASH_DEBUG
-#ifdef PFLASH_DEBUG
+#define PFLASH_DEBUG false
 #define DPRINTF(fmt, ...)                                  \
 do {                                                       \
-    fprintf(stderr, "PFLASH: " fmt , ## __VA_ARGS__);       \
+    if (PFLASH_DEBUG) {                                    \
+        fprintf(stderr, "PFLASH: " fmt, ## __VA_ARGS__);   \
+    }                                                      \
 } while (0)
-#else
-#define DPRINTF(fmt, ...) do { } while (0)
-#endif
 
 #define PFLASH_LAZY_ROMD_THRESHOLD 42
 
+/*
+ * The size of the cfi_table indirectly depends on this and the start of the
+ * PRI table directly depends on it. 4 is the maximum size (and also what
+ * seems common) without changing the PRT table address.
+ */
+#define PFLASH_MAX_ERASE_REGIONS 4
+
+/* Special write cycles for CFI queries. */
+enum {
+    WCYCLE_CFI              = 7,
+    WCYCLE_AUTOSELECT_CFI   = 8,
+};
+
 struct PFlashCFI02 {
     /*< private >*/
     SysBusDevice parent_obj;
     /*< public >*/
 
     BlockBackend *blk;
-    uint32_t sector_len;
-    uint32_t nb_blocs;
+    uint32_t uniform_nb_blocs;
+    uint32_t uniform_sector_len;
+    uint32_t total_sectors;
+    uint32_t nb_blocs[PFLASH_MAX_ERASE_REGIONS];
+    uint32_t sector_len[PFLASH_MAX_ERASE_REGIONS];
     uint32_t chip_len;
     uint8_t mappings;
     uint8_t width;
@@ -83,7 +95,7 @@ struct PFlashCFI02 {
     uint16_t ident3;
     uint16_t unlock_addr0;
     uint16_t unlock_addr1;
-    uint8_t cfi_table[0x52];
+    uint8_t cfi_table[0x4d];
     QEMUTimer timer;
     /* The device replicates the flash memory across its memory space.  Emulate
      * that by having a container (.mem) filled with an array of aliases
@@ -94,11 +106,63 @@ struct PFlashCFI02 {
     MemoryRegion orig_mem;
     int rom_mode;
     int read_counter; /* used for lazy switch-back to rom mode */
+    int sectors_to_erase;
+    uint64_t erase_time_remaining;
+    unsigned long *sector_erase_map;
     char *name;
     void *storage;
 };
 
 /*
+ * Toggle status bit DQ7.
+ */
+static inline void toggle_dq7(PFlashCFI02 *pfl)
+{
+    pfl->status ^= 0x80;
+}
+
+/*
+ * Set status bit DQ7 to bit 7 of value.
+ */
+static inline void set_dq7(PFlashCFI02 *pfl, uint8_t value)
+{
+    pfl->status &= 0x7F;
+    pfl->status |= value & 0x80;
+}
+
+/*
+ * Toggle status bit DQ6.
+ */
+static inline void toggle_dq6(PFlashCFI02 *pfl)
+{
+    pfl->status ^= 0x40;
+}
+
+/*
+ * Turn on DQ3.
+ */
+static inline void assert_dq3(PFlashCFI02 *pfl)
+{
+    pfl->status |= 0x08;
+}
+
+/*
+ * Turn off DQ3.
+ */
+static inline void reset_dq3(PFlashCFI02 *pfl)
+{
+    pfl->status &= ~0x08;
+}
+
+/*
+ * Toggle status bit DQ2.
+ */
+static inline void toggle_dq2(PFlashCFI02 *pfl)
+{
+    pfl->status ^= 0x04;
+}
+
+/*
  * Set up replicated mappings of the same region.
  */
 static void pflash_setup_mappings(PFlashCFI02 *pfl)
@@ -121,13 +185,63 @@ static void pflash_register_memory(PFlashCFI02 *pfl, int rom_mode)
     pfl->rom_mode = rom_mode;
 }
 
-static void pflash_timer (void *opaque)
+static size_t pflash_regions_count(PFlashCFI02 *pfl)
+{
+    return pfl->cfi_table[0x2c];
+}
+
+/*
+ * Returns the time it takes to erase the number of sectors scheduled for
+ * erasure based on CFI address 0x21 which is "Typical timeout per individual
+ * block erase 2^N ms."
+ */
+static uint64_t pflash_erase_time(PFlashCFI02 *pfl)
+{
+    /*
+     * If there are no sectors to erase (which can happen if all of the sectors
+     * to be erased are protected), then erase takes 100 us. Protected sectors
+     * aren't supported so this should never happen.
+     */
+    return ((1ULL << pfl->cfi_table[0x21]) * pfl->sectors_to_erase) * SCALE_US;
+}
+
+/*
+ * Returns true if the device is currently in erase suspend mode.
+ */
+static inline bool pflash_erase_suspend_mode(PFlashCFI02 *pfl)
+{
+    return pfl->erase_time_remaining > 0;
+}
+
+static void pflash_timer(void *opaque)
 {
     PFlashCFI02 *pfl = opaque;
 
     trace_pflash_timer_expired(pfl->cmd);
+    if (pfl->cmd == 0x30) {
+        /*
+         * Sector erase. If DQ3 is 0 when the timer expires, then the 50
+         * us erase timeout has expired so we need to start the timer for the
+         * sector erase algorithm. Otherwise, the erase completed and we should
+         * go back to read array mode.
+         */
+        if ((pfl->status & 0x08) == 0) {
+            assert_dq3(pfl);
+            uint64_t timeout = pflash_erase_time(pfl);
+            timer_mod(&pfl->timer,
+                      qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + timeout);
+            DPRINTF("%s: erase timeout fired; erasing %d sectors\n",
+                    __func__, pfl->sectors_to_erase);
+            return;
+        }
+        DPRINTF("%s: sector erase complete\n", __func__);
+        bitmap_zero(pfl->sector_erase_map, pfl->total_sectors);
+        pfl->sectors_to_erase = 0;
+        reset_dq3(pfl);
+    }
+
     /* Reset flash */
-    pfl->status ^= 0x80;
+    toggle_dq7(pfl);
     if (pfl->bypass) {
         pfl->wcycle = 2;
     } else {
@@ -137,15 +251,63 @@ static void pflash_timer (void *opaque)
     pfl->cmd = 0;
 }
 
-static uint32_t pflash_read(PFlashCFI02 *pfl, hwaddr offset,
-                            int width, int be)
+/*
+ * Read data from flash.
+ */
+static uint64_t pflash_data_read(PFlashCFI02 *pfl, hwaddr offset,
+                                 unsigned int width)
+{
+    uint8_t *p = (uint8_t *)pfl->storage + offset;
+    uint64_t ret = pfl->be ? ldn_be_p(p, width) : ldn_le_p(p, width);
+    trace_pflash_data_read(offset, width << 1, ret);
+    return ret;
+}
+
+typedef struct {
+    uint32_t len;
+    uint32_t num;
+} SectorInfo;
+
+/*
+ * offset should be a byte offset of the QEMU device and _not_ a device
+ * offset.
+ */
+static SectorInfo pflash_sector_info(PFlashCFI02 *pfl, hwaddr offset)
 {
+    assert(offset < pfl->chip_len);
+    hwaddr addr = 0;
+    uint32_t sector_num = 0;
+    for (int i = 0; i < pflash_regions_count(pfl); ++i) {
+        uint64_t region_size = (uint64_t)pfl->nb_blocs[i] * pfl->sector_len[i];
+        if (addr <= offset && offset < addr + region_size) {
+            return (SectorInfo) {
+                .len = pfl->sector_len[i],
+                .num = sector_num + (offset - addr) / pfl->sector_len[i],
+            };
+        }
+        sector_num += pfl->nb_blocs[i];
+        addr += region_size;
+    }
+    abort();
+}
+
+/*
+ * Returns true if the offset refers to a flash sector that is currently being
+ * erased.
+ */
+static bool pflash_sector_is_erasing(PFlashCFI02 *pfl, hwaddr offset)
+{
+    long sector_num = pflash_sector_info(pfl, offset).num;
+    return test_bit(sector_num, pfl->sector_erase_map);
+}
+
+static uint64_t pflash_read(void *opaque, hwaddr offset, unsigned int width)
+{
+    PFlashCFI02 *pfl = opaque;
     hwaddr boff;
-    uint32_t ret;
-    uint8_t *p;
+    uint64_t ret;
 
     ret = -1;
-    trace_pflash_read(offset, pfl->cmd, width, pfl->wcycle);
     /* Lazy reset to ROMD mode after a certain amount of read accesses */
     if (!pfl->rom_mode && pfl->wcycle == 0 &&
         ++pfl->read_counter > PFLASH_LAZY_ROMD_THRESHOLD) {
@@ -153,10 +315,9 @@ static uint32_t pflash_read(PFlashCFI02 *pfl, hwaddr offset,
     }
     offset &= pfl->chip_len - 1;
     boff = offset & 0xFF;
-    if (pfl->width == 2)
+    if (pfl->width == 2) {
         boff = boff >> 1;
-    else if (pfl->width == 4)
-        boff = boff >> 2;
+    }
     switch (pfl->cmd) {
     default:
         /* This should never happen : reset state & treat it as a read*/
@@ -164,45 +325,22 @@ static uint32_t pflash_read(PFlashCFI02 *pfl, hwaddr offset,
         pfl->wcycle = 0;
         pfl->cmd = 0;
         /* fall through to the read code */
-    case 0x80:
+    case 0x80: /* Erase (unlock) */
         /* We accept reads during second unlock sequence... */
     case 0x00:
-    flash_read:
-        /* Flash area read */
-        p = pfl->storage;
-        switch (width) {
-        case 1:
-            ret = p[offset];
-            trace_pflash_data_read8(offset, ret);
-            break;
-        case 2:
-            if (be) {
-                ret = p[offset] << 8;
-                ret |= p[offset + 1];
-            } else {
-                ret = p[offset];
-                ret |= p[offset + 1] << 8;
-            }
-            trace_pflash_data_read16(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;
-            }
-            trace_pflash_data_read32(offset, ret);
+        if (pflash_erase_suspend_mode(pfl) &&
+            pflash_sector_is_erasing(pfl, offset)) {
+            /* Toggle bit 2, but not 6. */
+            toggle_dq2(pfl);
+            /* Status register read */
+            ret = pfl->status;
+            DPRINTF("%s: status %" PRIx64 "\n", __func__, ret);
             break;
         }
+        /* Flash area read */
+        ret = pflash_data_read(pfl, offset, width);
         break;
-    case 0x90:
-        /* flash ID read */
+    case 0x90: /* flash ID read */
         switch (boff) {
         case 0x00:
         case 0x01:
@@ -214,23 +352,25 @@ static uint32_t pflash_read(PFlashCFI02 *pfl, hwaddr offset,
         case 0x0E:
         case 0x0F:
             ret = boff & 0x01 ? pfl->ident3 : pfl->ident2;
-            if (ret == (uint8_t)-1) {
-                goto flash_read;
+            if (ret != (uint8_t)-1) {
+                break;
             }
-            break;
+            /* Fall through to data read. */
         default:
-            goto flash_read;
+            ret = pflash_data_read(pfl, offset, width);
         }
-        DPRINTF("%s: ID " TARGET_FMT_plx " %x\n", __func__, boff, ret);
+        DPRINTF("%s: ID " TARGET_FMT_plx " %" PRIx64 "\n", __func__, boff, ret);
         break;
-    case 0xA0:
-    case 0x10:
-    case 0x30:
+    case 0x10: /* Chip Erase */
+    case 0x30: /* Sector Erase */
+        /* Toggle bit 2 during erase, but not program. */
+        toggle_dq2(pfl);
+    case 0xA0: /* Program */
+        /* Toggle bit 6 */
+        toggle_dq6(pfl);
         /* Status register read */
         ret = pfl->status;
-        DPRINTF("%s: status %x\n", __func__, ret);
-        /* Toggle bit 6 */
-        pfl->status ^= 0x40;
+        DPRINTF("%s: status %" PRIx64 "\n", __func__, ret);
         break;
     case 0x98:
         /* CFI query mode */
@@ -241,13 +381,13 @@ static uint32_t pflash_read(PFlashCFI02 *pfl, hwaddr offset,
         }
         break;
     }
+    trace_pflash_io_read(offset, width, width << 1, ret, pfl->cmd, pfl->wcycle);
 
     return ret;
 }
 
 /* update flash content on disk */
-static void pflash_update(PFlashCFI02 *pfl, int offset,
-                          int size)
+static void pflash_update(PFlashCFI02 *pfl, int offset, int size)
 {
     int offset_end;
     if (pfl->blk) {
@@ -260,31 +400,56 @@ static void pflash_update(PFlashCFI02 *pfl, int offset,
     }
 }
 
-static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
-                         uint32_t value, int width, int be)
+static void pflash_sector_erase(PFlashCFI02 *pfl, hwaddr offset)
 {
+    SectorInfo sector_info = pflash_sector_info(pfl, offset);
+    uint64_t sector_len = sector_info.len;
+    offset &= ~(sector_len - 1);
+    DPRINTF("%s: start sector erase at %0*" PRIx64 "-%0*" PRIx64 "\n",
+            __func__, pfl->width * 2, offset,
+            pfl->width * 2, offset + sector_len - 1);
+    if (!pfl->ro) {
+        uint8_t *p = pfl->storage;
+        memset(p + offset, 0xff, sector_len);
+        pflash_update(pfl, offset, sector_len);
+    }
+    set_dq7(pfl, 0x00);
+    ++pfl->sectors_to_erase;
+    set_bit(sector_info.num, pfl->sector_erase_map);
+    /* Set (or reset) the 50 us timer for additional erase commands.  */
+    timer_mod(&pfl->timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + 50000);
+}
+
+static void pflash_write(void *opaque, hwaddr offset, uint64_t value,
+                         unsigned int width)
+{
+    PFlashCFI02 *pfl = opaque;
     hwaddr boff;
     uint8_t *p;
     uint8_t cmd;
 
+    trace_pflash_io_write(offset, width, width << 1, value, pfl->wcycle);
     cmd = value;
-    if (pfl->cmd != 0xA0 && cmd == 0xF0) {
-#if 0
-        DPRINTF("%s: flash reset asked (%02x %02x)\n",
-                __func__, pfl->cmd, cmd);
-#endif
-        goto reset_flash;
+    if (pfl->cmd != 0xA0) {
+        /* Reset does nothing during chip erase and sector erase. */
+        if (cmd == 0xF0 && pfl->cmd != 0x10 && pfl->cmd != 0x30) {
+            if (pfl->wcycle == WCYCLE_AUTOSELECT_CFI) {
+                /* Return to autoselect mode. */
+                pfl->wcycle = 3;
+                pfl->cmd = 0x90;
+                return;
+            }
+            goto reset_flash;
+        }
     }
-    trace_pflash_write(offset, value, width, pfl->wcycle);
     offset &= pfl->chip_len - 1;
 
-    DPRINTF("%s: offset " TARGET_FMT_plx " %08x %d\n", __func__,
-            offset, value, width);
-    boff = offset & (pfl->sector_len - 1);
-    if (pfl->width == 2)
+    boff = offset;
+    if (pfl->width == 2) {
         boff = boff >> 1;
-    else if (pfl->width == 4)
-        boff = boff >> 2;
+    }
+    /* Only the least-significant 11 bits are used in most cases. */
+    boff &= 0x7FF;
     switch (pfl->wcycle) {
     case 0:
         /* Set the device in I/O access mode if required */
@@ -294,12 +459,30 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
         /* We're in read mode */
     check_unlock0:
         if (boff == 0x55 && cmd == 0x98) {
-        enter_CFI_mode:
             /* Enter CFI query mode */
-            pfl->wcycle = 7;
+            pfl->wcycle = WCYCLE_CFI;
             pfl->cmd = 0x98;
             return;
         }
+        /* Handle erase resume in erase suspend mode, otherwise reset. */
+        if (cmd == 0x30) { /* Erase Resume */
+            if (pflash_erase_suspend_mode(pfl)) {
+                /* Resume the erase. */
+                timer_mod(&pfl->timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
+                          pfl->erase_time_remaining);
+                pfl->erase_time_remaining = 0;
+                pfl->wcycle = 6;
+                pfl->cmd = 0x30;
+                set_dq7(pfl, 0x00);
+                assert_dq3(pfl);
+                return;
+            }
+            goto reset_flash;
+        }
+        /* Ignore erase suspend. */
+        if (cmd == 0xB0) { /* Erase Suspend */
+            return;
+        }
         if (boff != pfl->unlock_addr0 || cmd != 0xAA) {
             DPRINTF("%s: unlock0 failed " TARGET_FMT_plx " %02x %04x\n",
                     __func__, boff, cmd, pfl->unlock_addr0);
@@ -328,9 +511,9 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
         case 0x20:
             pfl->bypass = 1;
             goto do_bypass;
-        case 0x80:
-        case 0x90:
-        case 0xA0:
+        case 0x80: /* Erase */
+        case 0x90: /* Autoselect */
+        case 0xA0: /* Program */
             pfl->cmd = cmd;
             DPRINTF("%s: starting command %02x\n", __func__, cmd);
             break;
@@ -341,57 +524,54 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
         break;
     case 3:
         switch (pfl->cmd) {
-        case 0x80:
+        case 0x80: /* Erase */
             /* We need another unlock sequence */
             goto check_unlock0;
-        case 0xA0:
-            trace_pflash_data_write(offset, value, width, 0);
-            p = pfl->storage;
+        case 0xA0: /* Program */
+            if (pflash_erase_suspend_mode(pfl) &&
+                pflash_sector_is_erasing(pfl, offset)) {
+                /* Ignore writes to erasing sectors. */
+                if (pfl->bypass) {
+                    goto do_bypass;
+                }
+                goto reset_flash;
+            }
+            trace_pflash_data_write(offset, width << 1, value, 0);
             if (!pfl->ro) {
-                switch (width) {
-                case 1:
-                    p[offset] &= value;
-                    pflash_update(pfl, offset, 1);
-                    break;
-                case 2:
-                    if (be) {
-                        p[offset] &= value >> 8;
-                        p[offset + 1] &= value;
-                    } else {
-                        p[offset] &= value;
-                        p[offset + 1] &= value >> 8;
-                    }
-                    pflash_update(pfl, offset, 2);
-                    break;
-                case 4:
-                    if (be) {
-                        p[offset] &= value >> 24;
-                        p[offset + 1] &= value >> 16;
-                        p[offset + 2] &= value >> 8;
-                        p[offset + 3] &= value;
-                    } else {
-                        p[offset] &= value;
-                        p[offset + 1] &= value >> 8;
-                        p[offset + 2] &= value >> 16;
-                        p[offset + 3] &= value >> 24;
-                    }
-                    pflash_update(pfl, offset, 4);
-                    break;
+                p = (uint8_t *)pfl->storage + offset;
+                if (pfl->be) {
+                    uint64_t current = ldn_be_p(p, width);
+                    stn_be_p(p, width, current & value);
+                } else {
+                    uint64_t current = ldn_le_p(p, width);
+                    stn_le_p(p, width, current & value);
                 }
+                pflash_update(pfl, offset, width);
             }
-            pfl->status = 0x00 | ~(value & 0x80);
+            /*
+             * While programming, status bit DQ7 should hold the opposite
+             * value from how it was programmed.
+             */
+            set_dq7(pfl, ~value);
             /* Let's pretend write is immediate */
             if (pfl->bypass)
                 goto do_bypass;
             goto reset_flash;
-        case 0x90:
+        case 0x90: /* Autoselect */
             if (pfl->bypass && cmd == 0x00) {
                 /* Unlock bypass reset */
                 goto reset_flash;
             }
-            /* We can enter CFI query mode from autoselect mode */
-            if (boff == 0x55 && cmd == 0x98)
-                goto enter_CFI_mode;
+            /*
+             * We can enter CFI query mode from autoselect mode, but we must
+             * return to autoselect mode after a reset.
+             */
+            if (boff == 0x55 && cmd == 0x98) {
+                /* Enter autoselect CFI query mode */
+                pfl->wcycle = WCYCLE_AUTOSELECT_CFI;
+                pfl->cmd = 0x98;
+                return;
+            }
             /* No break here */
         default:
             DPRINTF("%s: invalid write for command %02x\n",
@@ -400,11 +580,11 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
         }
     case 4:
         switch (pfl->cmd) {
-        case 0xA0:
+        case 0xA0: /* Program */
             /* Ignore writes while flash data write is occurring */
             /* As we suppose write is immediate, this should never happen */
             return;
-        case 0x80:
+        case 0x80: /* Erase */
             goto check_unlock1;
         default:
             /* Should never happen */
@@ -414,8 +594,12 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
         }
         break;
     case 5:
+        if (pflash_erase_suspend_mode(pfl)) {
+            /* Erasing is not supported in erase suspend mode. */
+            goto reset_flash;
+        }
         switch (cmd) {
-        case 0x10:
+        case 0x10: /* Chip Erase */
             if (boff != pfl->unlock_addr0) {
                 DPRINTF("%s: chip erase: invalid address " TARGET_FMT_plx "\n",
                         __func__, offset);
@@ -424,28 +608,16 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
             /* Chip erase */
             DPRINTF("%s: start chip erase\n", __func__);
             if (!pfl->ro) {
-                memset(pfl->storage, 0xFF, pfl->chip_len);
+                memset(pfl->storage, 0xff, pfl->chip_len);
                 pflash_update(pfl, 0, pfl->chip_len);
             }
-            pfl->status = 0x00;
-            /* Let's wait 5 seconds before chip erase is done */
+            set_dq7(pfl, 0x00);
+            /* Wait the time specified at CFI address 0x22. */
             timer_mod(&pfl->timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
-                      (NANOSECONDS_PER_SECOND * 5));
+                      (1ULL << pfl->cfi_table[0x22]) * SCALE_MS);
             break;
-        case 0x30:
-            /* Sector erase */
-            p = pfl->storage;
-            offset &= ~(pfl->sector_len - 1);
-            DPRINTF("%s: start sector erase at " TARGET_FMT_plx "\n", __func__,
-                    offset);
-            if (!pfl->ro) {
-                memset(p + offset, 0xFF, pfl->sector_len);
-                pflash_update(pfl, offset, pfl->sector_len);
-            }
-            pfl->status = 0x00;
-            /* Let's wait 1/2 second before sector erase is done */
-            timer_mod(&pfl->timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
-                      (NANOSECONDS_PER_SECOND / 2));
+        case 0x30: /* Sector erase */
+            pflash_sector_erase(pfl, offset);
             break;
         default:
             DPRINTF("%s: invalid command %02x (wc 5)\n", __func__, cmd);
@@ -455,11 +627,47 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
         break;
     case 6:
         switch (pfl->cmd) {
-        case 0x10:
+        case 0x10: /* Chip Erase */
             /* Ignore writes during chip erase */
             return;
-        case 0x30:
-            /* Ignore writes during sector erase */
+        case 0x30: /* Sector erase */
+            if (cmd == 0xB0) {
+                /*
+                 * If erase suspend happens during the erase timeout (so DQ3 is
+                 * 0), then the device suspends erasing immediately. Set the
+                 * remaining time to be the total time to erase. Otherwise,
+                 * there is a maximum amount of time it can take to enter
+                 * suspend mode. Let's ignore that and suspend immediately and
+                 * set the remaining time to the actual time remaining on the
+                 * timer.
+                 */
+                if ((pfl->status & 0x08) == 0) {
+                    pfl->erase_time_remaining = pflash_erase_time(pfl);
+                } else {
+                    int64_t delta = timer_expire_time_ns(&pfl->timer) -
+                        qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
+                    /* Make sure we have a positive time remaining. */
+                    pfl->erase_time_remaining = delta <= 0 ? 1 : delta;
+                }
+                reset_dq3(pfl);
+                timer_del(&pfl->timer);
+                pfl->wcycle = 0;
+                pfl->cmd = 0;
+                return;
+            }
+            /*
+             * If DQ3 is 0, additional sector erase commands can be
+             * written and anything else (other than an erase suspend) resets
+             * the device.
+             */
+            if ((pfl->status & 0x08) == 0) {
+                if (cmd == 0x30) {
+                    pflash_sector_erase(pfl, offset);
+                } else {
+                    goto reset_flash;
+                }
+            }
+            /* Ignore writes during the actual erase. */
             return;
         default:
             /* Should never happen */
@@ -468,7 +676,9 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
             goto reset_flash;
         }
         break;
-    case 7: /* Special value for CFI queries */
+    /* Special values for CFI queries */
+    case WCYCLE_CFI:
+    case WCYCLE_AUTOSELECT_CFI:
         DPRINTF("%s: invalid write in CFI query mode\n", __func__);
         goto reset_flash;
     default:
@@ -493,39 +703,10 @@ static void pflash_write(PFlashCFI02 *pfl, hwaddr offset,
     pfl->cmd = 0;
 }
 
-static uint64_t pflash_be_readfn(void *opaque, hwaddr addr, unsigned size)
-{
-    return pflash_read(opaque, addr, size, 1);
-}
-
-static void pflash_be_writefn(void *opaque, hwaddr addr,
-                              uint64_t value, unsigned size)
-{
-    pflash_write(opaque, addr, value, size, 1);
-}
-
-static uint64_t pflash_le_readfn(void *opaque, hwaddr addr, unsigned size)
-{
-    return pflash_read(opaque, addr, size, 0);
-}
-
-static void pflash_le_writefn(void *opaque, hwaddr addr,
-                              uint64_t value, unsigned size)
-{
-    pflash_write(opaque, addr, value, size, 0);
-}
-
-static const MemoryRegionOps pflash_cfi02_ops_be = {
-    .read = pflash_be_readfn,
-    .write = pflash_be_writefn,
-    .valid.min_access_size = 1,
-    .valid.max_access_size = 4,
-    .endianness = DEVICE_NATIVE_ENDIAN,
-};
-
-static const MemoryRegionOps pflash_cfi02_ops_le = {
-    .read = pflash_le_readfn,
-    .write = pflash_le_writefn,
+static const MemoryRegionOps pflash_cfi02_ops = {
+    .read = pflash_read,
+    .write = pflash_write,
+    .impl.max_access_size = 2,
     .valid.min_access_size = 1,
     .valid.max_access_size = 4,
     .endianness = DEVICE_NATIVE_ENDIAN,
@@ -534,15 +715,14 @@ static const MemoryRegionOps pflash_cfi02_ops_le = {
 static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
 {
     PFlashCFI02 *pfl = PFLASH_CFI02(dev);
-    uint32_t chip_len;
     int ret;
     Error *local_err = NULL;
 
-    if (pfl->sector_len == 0) {
+    if (pfl->uniform_sector_len == 0 && pfl->sector_len[0] == 0) {
         error_setg(errp, "attribute \"sector-length\" not specified or zero.");
         return;
     }
-    if (pfl->nb_blocs == 0) {
+    if (pfl->uniform_nb_blocs == 0 && pfl->nb_blocs[0] == 0) {
         error_setg(errp, "attribute \"num-blocks\" not specified or zero.");
         return;
     }
@@ -551,18 +731,64 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
         return;
     }
 
-    chip_len = pfl->sector_len * pfl->nb_blocs;
+    int nb_regions;
+    pfl->chip_len = 0;
+    pfl->total_sectors = 0;
+    for (nb_regions = 0; nb_regions < PFLASH_MAX_ERASE_REGIONS; ++nb_regions) {
+        if (pfl->nb_blocs[nb_regions] == 0) {
+            break;
+        }
+        pfl->total_sectors += pfl->nb_blocs[nb_regions];
+        uint64_t sector_len_per_device = pfl->sector_len[nb_regions];
+
+        /*
+         * The size of each flash sector must be a power of 2 and it must be
+         * aligned at the same power of 2.
+         */
+        if (sector_len_per_device & 0xff ||
+            sector_len_per_device >= (1 << 24) ||
+            !is_power_of_2(sector_len_per_device))
+        {
+            error_setg(errp, "unsupported configuration: "
+                       "sector length[%d] per device = %" PRIx64 ".",
+                       nb_regions, sector_len_per_device);
+            return;
+        }
+        if (pfl->chip_len & (sector_len_per_device - 1)) {
+            error_setg(errp, "unsupported configuration: "
+                       "flash region %d not correctly aligned.",
+                       nb_regions);
+            return;
+        }
+
+        pfl->chip_len += (uint64_t)pfl->sector_len[nb_regions] *
+                          pfl->nb_blocs[nb_regions];
+    }
+
+    uint64_t uniform_len = (uint64_t)pfl->uniform_nb_blocs *
+                           pfl->uniform_sector_len;
+    if (nb_regions == 0) {
+        nb_regions = 1;
+        pfl->nb_blocs[0] = pfl->uniform_nb_blocs;
+        pfl->sector_len[0] = pfl->uniform_sector_len;
+        pfl->chip_len = uniform_len;
+        pfl->total_sectors = pfl->uniform_nb_blocs;
+    } else if (uniform_len != 0 && uniform_len != pfl->chip_len) {
+        error_setg(errp, "\"num-blocks\"*\"sector-length\" "
+                   "different from \"num-blocks0\"*\'sector-length0\" + ... + "
+                   "\"num-blocks3\"*\"sector-length3\"");
+        return;
+    }
 
-    memory_region_init_rom_device(&pfl->orig_mem, OBJECT(pfl), pfl->be ?
-                                  &pflash_cfi02_ops_be : &pflash_cfi02_ops_le,
-                                  pfl, pfl->name, chip_len, &local_err);
+    memory_region_init_rom_device(&pfl->orig_mem, OBJECT(pfl),
+                                  &pflash_cfi02_ops, pfl, pfl->name,
+                                  pfl->chip_len, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         return;
     }
 
     pfl->storage = memory_region_get_ram_ptr(&pfl->orig_mem);
-    pfl->chip_len = chip_len;
 
     if (pfl->blk) {
         uint64_t perm;
@@ -577,13 +803,20 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     }
 
     if (pfl->blk) {
-        if (!blk_check_size_and_read_all(pfl->blk, pfl->storage, chip_len,
-                                         errp)) {
+        if (!blk_check_size_and_read_all(pfl->blk, pfl->storage,
+                                         pfl->chip_len, errp)) {
             vmstate_unregister_ram(&pfl->orig_mem, DEVICE(pfl));
             return;
         }
     }
 
+    /* Only 11 bits are used in the comparison. */
+    pfl->unlock_addr0 &= 0x7FF;
+    pfl->unlock_addr1 &= 0x7FF;
+
+    /* Allocate memory for a bitmap for sectors being erased. */
+    pfl->sector_erase_map = bitmap_new(pfl->total_sectors);
+
     pflash_setup_mappings(pfl);
     pfl->rom_mode = 1;
     sysbus_init_mmio(SYS_BUS_DEVICE(dev), &pfl->mem);
@@ -592,7 +825,9 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     pfl->wcycle = 0;
     pfl->cmd = 0;
     pfl->status = 0;
+
     /* Hardcoded CFI table (mostly from SG29 Spansion flash) */
+    const uint16_t pri_ofs = 0x40;
     /* Standard "QRY" string */
     pfl->cfi_table[0x10] = 'Q';
     pfl->cfi_table[0x11] = 'R';
@@ -601,8 +836,8 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     pfl->cfi_table[0x13] = 0x02;
     pfl->cfi_table[0x14] = 0x00;
     /* Primary extended table address */
-    pfl->cfi_table[0x15] = 0x31;
-    pfl->cfi_table[0x16] = 0x00;
+    pfl->cfi_table[0x15] = pri_ofs;
+    pfl->cfi_table[0x16] = pri_ofs >> 8;
     /* Alternate command set (none) */
     pfl->cfi_table[0x17] = 0x00;
     pfl->cfi_table[0x18] = 0x00;
@@ -617,7 +852,7 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     pfl->cfi_table[0x1D] = 0x00;
     /* Vpp max (no Vpp pin) */
     pfl->cfi_table[0x1E] = 0x00;
-    /* Reserved */
+    /* Timeout per single byte/word write (128 ms) */
     pfl->cfi_table[0x1F] = 0x07;
     /* Timeout for min size buffer write (NA) */
     pfl->cfi_table[0x20] = 0x00;
@@ -634,7 +869,7 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     /* Max timeout for chip erase */
     pfl->cfi_table[0x26] = 0x0D;
     /* Device size */
-    pfl->cfi_table[0x27] = ctz32(chip_len);
+    pfl->cfi_table[0x27] = ctz32(pfl->chip_len);
     /* Flash device interface (8 & 16 bits) */
     pfl->cfi_table[0x28] = 0x02;
     pfl->cfi_table[0x29] = 0x00;
@@ -643,37 +878,60 @@ static void pflash_cfi02_realize(DeviceState *dev, Error **errp)
     //    pfl->cfi_table[0x2A] = 0x05;
     pfl->cfi_table[0x2A] = 0x00;
     pfl->cfi_table[0x2B] = 0x00;
-    /* Number of erase block regions (uniform) */
-    pfl->cfi_table[0x2C] = 0x01;
-    /* Erase block region 1 */
-    pfl->cfi_table[0x2D] = pfl->nb_blocs - 1;
-    pfl->cfi_table[0x2E] = (pfl->nb_blocs - 1) >> 8;
-    pfl->cfi_table[0x2F] = pfl->sector_len >> 8;
-    pfl->cfi_table[0x30] = pfl->sector_len >> 16;
+    /* Number of erase block regions */
+    pfl->cfi_table[0x2c] = nb_regions;
+    /* Erase block regions */
+    for (int i = 0; i < nb_regions; ++i) {
+        uint32_t sector_len_per_device = pfl->sector_len[i];
+        pfl->cfi_table[0x2d + 4 * i] = pfl->nb_blocs[i] - 1;
+        pfl->cfi_table[0x2e + 4 * i] = (pfl->nb_blocs[i] - 1) >> 8;
+        pfl->cfi_table[0x2f + 4 * i] = sector_len_per_device >> 8;
+        pfl->cfi_table[0x30 + 4 * i] = sector_len_per_device >> 16;
+    }
+    assert(0x2c + 4 * nb_regions < pri_ofs);
 
     /* Extended */
-    pfl->cfi_table[0x31] = 'P';
-    pfl->cfi_table[0x32] = 'R';
-    pfl->cfi_table[0x33] = 'I';
-
-    pfl->cfi_table[0x34] = '1';
-    pfl->cfi_table[0x35] = '0';
-
-    pfl->cfi_table[0x36] = 0x00;
-    pfl->cfi_table[0x37] = 0x00;
-    pfl->cfi_table[0x38] = 0x00;
-    pfl->cfi_table[0x39] = 0x00;
-
-    pfl->cfi_table[0x3a] = 0x00;
-
-    pfl->cfi_table[0x3b] = 0x00;
-    pfl->cfi_table[0x3c] = 0x00;
+    pfl->cfi_table[0x00 + pri_ofs] = 'P';
+    pfl->cfi_table[0x01 + pri_ofs] = 'R';
+    pfl->cfi_table[0x02 + pri_ofs] = 'I';
+
+    /* Extended version 1.0 */
+    pfl->cfi_table[0x03 + pri_ofs] = '1';
+    pfl->cfi_table[0x04 + pri_ofs] = '0';
+
+    /* Address sensitive unlock required. */
+    pfl->cfi_table[0x05 + pri_ofs] = 0x00;
+    /* Erase suspend to read/write. */
+    pfl->cfi_table[0x06 + pri_ofs] = 0x02;
+    /* Sector protect not supported. */
+    pfl->cfi_table[0x07 + pri_ofs] = 0x00;
+    /* Temporary sector unprotect not supported. */
+    pfl->cfi_table[0x08 + pri_ofs] = 0x00;
+
+    /* Sector protect/unprotect scheme. */
+    pfl->cfi_table[0x09 + pri_ofs] = 0x00;
+
+    /* Simultaneous operation not supported. */
+    pfl->cfi_table[0x0a + pri_ofs] = 0x00;
+    /* Burst mode not supported. */
+    pfl->cfi_table[0x0b + pri_ofs] = 0x00;
+    /* Page mode not supported. */
+    pfl->cfi_table[0x0c + pri_ofs] = 0x00;
+    assert(0x0c + pri_ofs < ARRAY_SIZE(pfl->cfi_table));
 }
 
 static Property pflash_cfi02_properties[] = {
     DEFINE_PROP_DRIVE("drive", PFlashCFI02, blk),
-    DEFINE_PROP_UINT32("num-blocks", PFlashCFI02, nb_blocs, 0),
-    DEFINE_PROP_UINT32("sector-length", PFlashCFI02, sector_len, 0),
+    DEFINE_PROP_UINT32("num-blocks", PFlashCFI02, uniform_nb_blocs, 0),
+    DEFINE_PROP_UINT32("sector-length", PFlashCFI02, uniform_sector_len, 0),
+    DEFINE_PROP_UINT32("num-blocks0", PFlashCFI02, nb_blocs[0], 0),
+    DEFINE_PROP_UINT32("sector-length0", PFlashCFI02, sector_len[0], 0),
+    DEFINE_PROP_UINT32("num-blocks1", PFlashCFI02, nb_blocs[1], 0),
+    DEFINE_PROP_UINT32("sector-length1", PFlashCFI02, sector_len[1], 0),
+    DEFINE_PROP_UINT32("num-blocks2", PFlashCFI02, nb_blocs[2], 0),
+    DEFINE_PROP_UINT32("sector-length2", PFlashCFI02, sector_len[2], 0),
+    DEFINE_PROP_UINT32("num-blocks3", PFlashCFI02, nb_blocs[3], 0),
+    DEFINE_PROP_UINT32("sector-length3", PFlashCFI02, sector_len[3], 0),
     DEFINE_PROP_UINT8("width", PFlashCFI02, width, 0),
     DEFINE_PROP_UINT8("mappings", PFlashCFI02, mappings, 0),
     DEFINE_PROP_UINT8("big-endian", PFlashCFI02, be, 0),
@@ -691,6 +949,7 @@ static void pflash_cfi02_unrealize(DeviceState *dev, Error **errp)
 {
     PFlashCFI02 *pfl = PFLASH_CFI02(dev);
     timer_del(&pfl->timer);
+    g_free(pfl->sector_erase_map);
 }
 
 static void pflash_cfi02_class_init(ObjectClass *klass, void *data)
diff --git a/hw/block/trace-events b/hw/block/trace-events
index 97a17838ed..13d1b21dd4 100644
--- a/hw/block/trace-events
+++ b/hw/block/trace-events
@@ -7,13 +7,11 @@ fdc_ioport_write(uint8_t reg, uint8_t value) "write reg 0x%02x val 0x%02x"
 # pflash_cfi02.c
 # pflash_cfi01.c
 pflash_reset(void) "reset"
-pflash_read(uint64_t offset, uint8_t cmd, int width, uint8_t wcycle) "offset:0x%04"PRIx64" cmd:0x%02x width:%d wcycle:%u"
-pflash_write(uint64_t offset, uint32_t value, int width, uint8_t wcycle) "offset:0x%04"PRIx64" value:0x%03x width:%d wcycle:%u"
 pflash_timer_expired(uint8_t cmd) "command 0x%02x done"
-pflash_data_read8(uint64_t offset, uint32_t value) "data offset:0x%04"PRIx64" value:0x%02x"
-pflash_data_read16(uint64_t offset, uint32_t value) "data offset:0x%04"PRIx64" value:0x%04x"
-pflash_data_read32(uint64_t offset, uint32_t value) "data offset:0x%04"PRIx64" value:0x%08x"
-pflash_data_write(uint64_t offset, uint32_t value, int width, uint64_t counter) "data offset:0x%04"PRIx64" value:0x%08x width:%d counter:0x%016"PRIx64
+pflash_io_read(uint64_t offset, int width, int fmt_width, uint32_t value, uint8_t cmd, uint8_t wcycle) "offset:0x%04"PRIx64" width:%d value:0x%0*x cmd:0x%02x wcycle:%u"
+pflash_io_write(uint64_t offset, int width, int fmt_width, uint32_t value, uint8_t wcycle) "offset:0x%04"PRIx64" width:%d value:0x%0*x wcycle:%u"
+pflash_data_read(uint64_t offset, int width, uint32_t value) "data offset:0x%04"PRIx64" value:0x%0*x"
+pflash_data_write(uint64_t offset, int width, uint32_t value, uint64_t counter) "data offset:0x%04"PRIx64" value:0x%0*x counter:0x%016"PRIx64
 pflash_manufacturer_id(uint16_t id) "Read Manufacturer ID: 0x%04x"
 pflash_device_id(uint16_t id) "Read Device ID: 0x%04x"
 pflash_device_info(uint64_t offset) "Read Device Information offset:0x%04"PRIx64
diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c
index 4de537aef4..69d73196e2 100644
--- a/hw/block/xen-block.c
+++ b/hw/block/xen-block.c
@@ -11,7 +11,7 @@
 #include "qemu/option.h"
 #include "qapi/error.h"
 #include "qapi/qapi-commands-block-core.h"
-#include "qapi/qapi-commands-misc.h"
+#include "qapi/qapi-commands-qom.h"
 #include "qapi/qapi-visit-block-core.h"
 #include "qapi/qobject-input-visitor.h"
 #include "qapi/visitor.h"
@@ -52,11 +52,25 @@ static void xen_block_connect(XenDevice *xendev, Error **errp)
     XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev);
     const char *type = object_get_typename(OBJECT(blockdev));
     XenBlockVdev *vdev = &blockdev->props.vdev;
+    BlockConf *conf = &blockdev->props.conf;
+    unsigned int feature_large_sector_size;
     unsigned int order, nr_ring_ref, *ring_ref, event_channel, protocol;
     char *str;
 
     trace_xen_block_connect(type, vdev->disk, vdev->partition);
 
+    if (xen_device_frontend_scanf(xendev, "feature-large-sector-size", "%u",
+                                  &feature_large_sector_size) != 1) {
+        feature_large_sector_size = 0;
+    }
+
+    if (feature_large_sector_size != 1 &&
+        conf->logical_block_size != XEN_BLKIF_SECTOR_SIZE) {
+        error_setg(errp, "logical_block_size != %u not supported by frontend",
+                   XEN_BLKIF_SECTOR_SIZE);
+        return;
+    }
+
     if (xen_device_frontend_scanf(xendev, "ring-page-order", "%u",
                                   &order) != 1) {
         nr_ring_ref = 1;
@@ -150,7 +164,7 @@ static void xen_block_set_size(XenBlockDevice *blockdev)
     const char *type = object_get_typename(OBJECT(blockdev));
     XenBlockVdev *vdev = &blockdev->props.vdev;
     BlockConf *conf = &blockdev->props.conf;
-    int64_t sectors = blk_getlength(conf->blk) / XEN_BLKIF_SECTOR_SIZE;
+    int64_t sectors = blk_getlength(conf->blk) / conf->logical_block_size;
     XenDevice *xendev = XEN_DEVICE(blockdev);
 
     trace_xen_block_size(type, vdev->disk, vdev->partition, sectors);
@@ -185,6 +199,7 @@ static void xen_block_realize(XenDevice *xendev, Error **errp)
     const char *type = object_get_typename(OBJECT(blockdev));
     XenBlockVdev *vdev = &blockdev->props.vdev;
     BlockConf *conf = &blockdev->props.conf;
+    BlockBackend *blk = conf->blk;
     Error *local_err = NULL;
 
     if (vdev->type == XEN_BLOCK_VDEV_TYPE_INVALID) {
@@ -206,8 +221,8 @@ static void xen_block_realize(XenDevice *xendev, Error **errp)
      * The blkif protocol does not deal with removable media, so it must
      * always be present, even for CDRom devices.
      */
-    assert(conf->blk);
-    if (!blk_is_inserted(conf->blk)) {
+    assert(blk);
+    if (!blk_is_inserted(blk)) {
         error_setg(errp, "device needs media, but drive is empty");
         return;
     }
@@ -224,26 +239,20 @@ static void xen_block_realize(XenDevice *xendev, Error **errp)
 
     blkconf_blocksizes(conf);
 
-    if (conf->logical_block_size != XEN_BLKIF_SECTOR_SIZE) {
-        error_setg(errp, "logical_block_size != %u not supported",
-                   XEN_BLKIF_SECTOR_SIZE);
-        return;
-    }
-
     if (conf->logical_block_size > conf->physical_block_size) {
         error_setg(
             errp, "logical_block_size > physical_block_size not supported");
         return;
     }
 
-    blk_set_dev_ops(conf->blk, &xen_block_dev_ops, blockdev);
-    blk_set_guest_block_size(conf->blk, conf->logical_block_size);
+    blk_set_dev_ops(blk, &xen_block_dev_ops, blockdev);
+    blk_set_guest_block_size(blk, conf->logical_block_size);
 
     if (conf->discard_granularity == -1) {
         conf->discard_granularity = conf->physical_block_size;
     }
 
-    if (blk_get_flags(conf->blk) & BDRV_O_UNMAP) {
+    if (blk_get_flags(blk) & BDRV_O_UNMAP) {
         xen_device_backend_printf(xendev, "feature-discard", "%u", 1);
         xen_device_backend_printf(xendev, "discard-granularity", "%u",
                                   conf->discard_granularity);
@@ -260,12 +269,13 @@ static void xen_block_realize(XenDevice *xendev, Error **errp)
                                blockdev->device_type);
 
     xen_device_backend_printf(xendev, "sector-size", "%u",
-                              XEN_BLKIF_SECTOR_SIZE);
+                              conf->logical_block_size);
 
     xen_block_set_size(blockdev);
 
     blockdev->dataplane =
-        xen_block_dataplane_create(xendev, conf, blockdev->props.iothread);
+        xen_block_dataplane_create(xendev, blk, conf->logical_block_size,
+                                   blockdev->props.iothread);
 }
 
 static void xen_block_frontend_changed(XenDevice *xendev,
diff --git a/hw/block/xen_blkif.h b/hw/block/xen_blkif.h
index a353693ea0..99733529c1 100644
--- a/hw/block/xen_blkif.h
+++ b/hw/block/xen_blkif.h
@@ -1,9 +1,8 @@
 #ifndef XEN_BLKIF_H
 #define XEN_BLKIF_H
 
-#include "hw/xen/io/ring.h"
-#include <xen/io/blkif.h>
-#include <xen/io/protocols.h>
+#include "hw/xen/interface/io/blkif.h"
+#include "hw/xen/interface/io/protocols.h"
 
 /*
  * Not a real protocol.  Used to generate ring structs which contain
diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c
index 91f34ef06c..47e1092263 100644
--- a/hw/char/xen_console.c
+++ b/hw/char/xen_console.c
@@ -28,7 +28,7 @@
 #include "chardev/char-fe.h"
 #include "hw/xen/xen-legacy-backend.h"
 
-#include <xen/io/console.h>
+#include "hw/xen/interface/io/console.h"
 
 struct buffer {
     uint8_t *data;
diff --git a/hw/core/Makefile.objs b/hw/core/Makefile.objs
index a799c83815..585b734358 100644
--- a/hw/core/Makefile.objs
+++ b/hw/core/Makefile.objs
@@ -22,3 +22,7 @@ common-obj-$(CONFIG_SOFTMMU) += split-irq.o
 common-obj-$(CONFIG_PLATFORM_BUS) += platform-bus.o
 common-obj-$(CONFIG_SOFTMMU) += generic-loader.o
 common-obj-$(CONFIG_SOFTMMU) += null-machine.o
+
+obj-$(CONFIG_SOFTMMU) += machine-qmp-cmds.o
+obj-$(CONFIG_SOFTMMU) += numa.o
+common-obj-$(CONFIG_SOFTMMU) += machine-hmp-cmds.o
diff --git a/hw/core/machine-hmp-cmds.c b/hw/core/machine-hmp-cmds.c
new file mode 100644
index 0000000000..7fa6075f1e
--- /dev/null
+++ b/hw/core/machine-hmp-cmds.c
@@ -0,0 +1,164 @@
+/*
+ * HMP commands related to machines and CPUs
+ *
+ * Copyright IBM, Corp. 2011
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ * Contributions after 2012-01-13 are licensed under the terms of the
+ * GNU GPL, version 2 or (at your option) any later version.
+ */
+
+#include "qemu/osdep.h"
+#include "monitor/hmp.h"
+#include "monitor/monitor.h"
+#include "qapi/error.h"
+#include "qapi/qapi-builtin-visit.h"
+#include "qapi/qapi-commands-machine.h"
+#include "qapi/qmp/qdict.h"
+#include "qapi/string-output-visitor.h"
+#include "qemu/error-report.h"
+#include "sysemu/numa.h"
+
+void hmp_info_cpus(Monitor *mon, const QDict *qdict)
+{
+    CpuInfoFastList *cpu_list, *cpu;
+
+    cpu_list = qmp_query_cpus_fast(NULL);
+
+    for (cpu = cpu_list; cpu; cpu = cpu->next) {
+        int active = ' ';
+
+        if (cpu->value->cpu_index == monitor_get_cpu_index()) {
+            active = '*';
+        }
+
+        monitor_printf(mon, "%c CPU #%" PRId64 ":", active,
+                       cpu->value->cpu_index);
+        monitor_printf(mon, " thread_id=%" PRId64 "\n", cpu->value->thread_id);
+    }
+
+    qapi_free_CpuInfoFastList(cpu_list);
+}
+
+void hmp_cpu_add(Monitor *mon, const QDict *qdict)
+{
+    int cpuid;
+    Error *err = NULL;
+
+    error_report("cpu_add is deprecated, please use device_add instead");
+
+    cpuid = qdict_get_int(qdict, "id");
+    qmp_cpu_add(cpuid, &err);
+    hmp_handle_error(mon, &err);
+}
+
+void hmp_hotpluggable_cpus(Monitor *mon, const QDict *qdict)
+{
+    Error *err = NULL;
+    HotpluggableCPUList *l = qmp_query_hotpluggable_cpus(&err);
+    HotpluggableCPUList *saved = l;
+    CpuInstanceProperties *c;
+
+    if (err != NULL) {
+        hmp_handle_error(mon, &err);
+        return;
+    }
+
+    monitor_printf(mon, "Hotpluggable CPUs:\n");
+    while (l) {
+        monitor_printf(mon, "  type: \"%s\"\n", l->value->type);
+        monitor_printf(mon, "  vcpus_count: \"%" PRIu64 "\"\n",
+                       l->value->vcpus_count);
+        if (l->value->has_qom_path) {
+            monitor_printf(mon, "  qom_path: \"%s\"\n", l->value->qom_path);
+        }
+
+        c = l->value->props;
+        monitor_printf(mon, "  CPUInstance Properties:\n");
+        if (c->has_node_id) {
+            monitor_printf(mon, "    node-id: \"%" PRIu64 "\"\n", c->node_id);
+        }
+        if (c->has_socket_id) {
+            monitor_printf(mon, "    socket-id: \"%" PRIu64 "\"\n", c->socket_id);
+        }
+        if (c->has_core_id) {
+            monitor_printf(mon, "    core-id: \"%" PRIu64 "\"\n", c->core_id);
+        }
+        if (c->has_thread_id) {
+            monitor_printf(mon, "    thread-id: \"%" PRIu64 "\"\n", c->thread_id);
+        }
+
+        l = l->next;
+    }
+
+    qapi_free_HotpluggableCPUList(saved);
+}
+
+void hmp_info_memdev(Monitor *mon, const QDict *qdict)
+{
+    Error *err = NULL;
+    MemdevList *memdev_list = qmp_query_memdev(&err);
+    MemdevList *m = memdev_list;
+    Visitor *v;
+    char *str;
+
+    while (m) {
+        v = string_output_visitor_new(false, &str);
+        visit_type_uint16List(v, NULL, &m->value->host_nodes, NULL);
+        monitor_printf(mon, "memory backend: %s\n", m->value->id);
+        monitor_printf(mon, "  size:  %" PRId64 "\n", m->value->size);
+        monitor_printf(mon, "  merge: %s\n",
+                       m->value->merge ? "true" : "false");
+        monitor_printf(mon, "  dump: %s\n",
+                       m->value->dump ? "true" : "false");
+        monitor_printf(mon, "  prealloc: %s\n",
+                       m->value->prealloc ? "true" : "false");
+        monitor_printf(mon, "  policy: %s\n",
+                       HostMemPolicy_str(m->value->policy));
+        visit_complete(v, &str);
+        monitor_printf(mon, "  host nodes: %s\n", str);
+
+        g_free(str);
+        visit_free(v);
+        m = m->next;
+    }
+
+    monitor_printf(mon, "\n");
+
+    qapi_free_MemdevList(memdev_list);
+    hmp_handle_error(mon, &err);
+}
+
+void hmp_info_numa(Monitor *mon, const QDict *qdict)
+{
+    int i;
+    NumaNodeMem *node_mem;
+    CpuInfoList *cpu_list, *cpu;
+
+    cpu_list = qmp_query_cpus(&error_abort);
+    node_mem = g_new0(NumaNodeMem, nb_numa_nodes);
+
+    query_numa_node_mem(node_mem);
+    monitor_printf(mon, "%d nodes\n", nb_numa_nodes);
+    for (i = 0; i < nb_numa_nodes; i++) {
+        monitor_printf(mon, "node %d cpus:", i);
+        for (cpu = cpu_list; cpu; cpu = cpu->next) {
+            if (cpu->value->has_props && cpu->value->props->has_node_id &&
+                cpu->value->props->node_id == i) {
+                monitor_printf(mon, " %" PRIi64, cpu->value->CPU);
+            }
+        }
+        monitor_printf(mon, "\n");
+        monitor_printf(mon, "node %d size: %" PRId64 " MB\n", i,
+                       node_mem[i].node_mem >> 20);
+        monitor_printf(mon, "node %d plugged: %" PRId64 " MB\n", i,
+                       node_mem[i].node_plugged_mem >> 20);
+    }
+    qapi_free_CpuInfoList(cpu_list);
+    g_free(node_mem);
+}
diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
new file mode 100644
index 0000000000..1e08252af7
--- /dev/null
+++ b/hw/core/machine-qmp-cmds.c
@@ -0,0 +1,328 @@
+/*
+ * QMP commands related to machines and CPUs
+ *
+ * Copyright (C) 2014 Red Hat Inc
+ *
+ * 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 "qemu/osdep.h"
+#include "cpu.h"
+#include "hw/boards.h"
+#include "qapi/error.h"
+#include "qapi/qapi-commands-machine.h"
+#include "qapi/qmp/qerror.h"
+#include "sysemu/hostmem.h"
+#include "sysemu/hw_accel.h"
+#include "sysemu/numa.h"
+#include "sysemu/sysemu.h"
+
+CpuInfoList *qmp_query_cpus(Error **errp)
+{
+    MachineState *ms = MACHINE(qdev_get_machine());
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+    CpuInfoList *head = NULL, *cur_item = NULL;
+    CPUState *cpu;
+
+    CPU_FOREACH(cpu) {
+        CpuInfoList *info;
+#if defined(TARGET_I386)
+        X86CPU *x86_cpu = X86_CPU(cpu);
+        CPUX86State *env = &x86_cpu->env;
+#elif defined(TARGET_PPC)
+        PowerPCCPU *ppc_cpu = POWERPC_CPU(cpu);
+        CPUPPCState *env = &ppc_cpu->env;
+#elif defined(TARGET_SPARC)
+        SPARCCPU *sparc_cpu = SPARC_CPU(cpu);
+        CPUSPARCState *env = &sparc_cpu->env;
+#elif defined(TARGET_RISCV)
+        RISCVCPU *riscv_cpu = RISCV_CPU(cpu);
+        CPURISCVState *env = &riscv_cpu->env;
+#elif defined(TARGET_MIPS)
+        MIPSCPU *mips_cpu = MIPS_CPU(cpu);
+        CPUMIPSState *env = &mips_cpu->env;
+#elif defined(TARGET_TRICORE)
+        TriCoreCPU *tricore_cpu = TRICORE_CPU(cpu);
+        CPUTriCoreState *env = &tricore_cpu->env;
+#elif defined(TARGET_S390X)
+        S390CPU *s390_cpu = S390_CPU(cpu);
+        CPUS390XState *env = &s390_cpu->env;
+#endif
+
+        cpu_synchronize_state(cpu);
+
+        info = g_malloc0(sizeof(*info));
+        info->value = g_malloc0(sizeof(*info->value));
+        info->value->CPU = cpu->cpu_index;
+        info->value->current = (cpu == first_cpu);
+        info->value->halted = cpu->halted;
+        info->value->qom_path = object_get_canonical_path(OBJECT(cpu));
+        info->value->thread_id = cpu->thread_id;
+#if defined(TARGET_I386)
+        info->value->arch = CPU_INFO_ARCH_X86;
+        info->value->u.x86.pc = env->eip + env->segs[R_CS].base;
+#elif defined(TARGET_PPC)
+        info->value->arch = CPU_INFO_ARCH_PPC;
+        info->value->u.ppc.nip = env->nip;
+#elif defined(TARGET_SPARC)
+        info->value->arch = CPU_INFO_ARCH_SPARC;
+        info->value->u.q_sparc.pc = env->pc;
+        info->value->u.q_sparc.npc = env->npc;
+#elif defined(TARGET_MIPS)
+        info->value->arch = CPU_INFO_ARCH_MIPS;
+        info->value->u.q_mips.PC = env->active_tc.PC;
+#elif defined(TARGET_TRICORE)
+        info->value->arch = CPU_INFO_ARCH_TRICORE;
+        info->value->u.tricore.PC = env->PC;
+#elif defined(TARGET_S390X)
+        info->value->arch = CPU_INFO_ARCH_S390;
+        info->value->u.s390.cpu_state = env->cpu_state;
+#elif defined(TARGET_RISCV)
+        info->value->arch = CPU_INFO_ARCH_RISCV;
+        info->value->u.riscv.pc = env->pc;
+#else
+        info->value->arch = CPU_INFO_ARCH_OTHER;
+#endif
+        info->value->has_props = !!mc->cpu_index_to_instance_props;
+        if (info->value->has_props) {
+            CpuInstanceProperties *props;
+            props = g_malloc0(sizeof(*props));
+            *props = mc->cpu_index_to_instance_props(ms, cpu->cpu_index);
+            info->value->props = props;
+        }
+
+        /* XXX: waiting for the qapi to support GSList */
+        if (!cur_item) {
+            head = cur_item = info;
+        } else {
+            cur_item->next = info;
+            cur_item = info;
+        }
+    }
+
+    return head;
+}
+
+static CpuInfoArch sysemu_target_to_cpuinfo_arch(SysEmuTarget target)
+{
+    /*
+     * The @SysEmuTarget -> @CpuInfoArch mapping below is based on the
+     * TARGET_ARCH -> TARGET_BASE_ARCH mapping in the "configure" script.
+     */
+    switch (target) {
+    case SYS_EMU_TARGET_I386:
+    case SYS_EMU_TARGET_X86_64:
+        return CPU_INFO_ARCH_X86;
+
+    case SYS_EMU_TARGET_PPC:
+    case SYS_EMU_TARGET_PPC64:
+        return CPU_INFO_ARCH_PPC;
+
+    case SYS_EMU_TARGET_SPARC:
+    case SYS_EMU_TARGET_SPARC64:
+        return CPU_INFO_ARCH_SPARC;
+
+    case SYS_EMU_TARGET_MIPS:
+    case SYS_EMU_TARGET_MIPSEL:
+    case SYS_EMU_TARGET_MIPS64:
+    case SYS_EMU_TARGET_MIPS64EL:
+        return CPU_INFO_ARCH_MIPS;
+
+    case SYS_EMU_TARGET_TRICORE:
+        return CPU_INFO_ARCH_TRICORE;
+
+    case SYS_EMU_TARGET_S390X:
+        return CPU_INFO_ARCH_S390;
+
+    case SYS_EMU_TARGET_RISCV32:
+    case SYS_EMU_TARGET_RISCV64:
+        return CPU_INFO_ARCH_RISCV;
+
+    default:
+        return CPU_INFO_ARCH_OTHER;
+    }
+}
+
+static void cpustate_to_cpuinfo_s390(CpuInfoS390 *info, const CPUState *cpu)
+{
+#ifdef TARGET_S390X
+    S390CPU *s390_cpu = S390_CPU(cpu);
+    CPUS390XState *env = &s390_cpu->env;
+
+    info->cpu_state = env->cpu_state;
+#else
+    abort();
+#endif
+}
+
+/*
+ * fast means: we NEVER interrupt vCPU threads to retrieve
+ * information from KVM.
+ */
+CpuInfoFastList *qmp_query_cpus_fast(Error **errp)
+{
+    MachineState *ms = MACHINE(qdev_get_machine());
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+    CpuInfoFastList *head = NULL, *cur_item = NULL;
+    SysEmuTarget target = qapi_enum_parse(&SysEmuTarget_lookup, TARGET_NAME,
+                                          -1, &error_abort);
+    CPUState *cpu;
+
+    CPU_FOREACH(cpu) {
+        CpuInfoFastList *info = g_malloc0(sizeof(*info));
+        info->value = g_malloc0(sizeof(*info->value));
+
+        info->value->cpu_index = cpu->cpu_index;
+        info->value->qom_path = object_get_canonical_path(OBJECT(cpu));
+        info->value->thread_id = cpu->thread_id;
+
+        info->value->has_props = !!mc->cpu_index_to_instance_props;
+        if (info->value->has_props) {
+            CpuInstanceProperties *props;
+            props = g_malloc0(sizeof(*props));
+            *props = mc->cpu_index_to_instance_props(ms, cpu->cpu_index);
+            info->value->props = props;
+        }
+
+        info->value->arch = sysemu_target_to_cpuinfo_arch(target);
+        info->value->target = target;
+        if (target == SYS_EMU_TARGET_S390X) {
+            cpustate_to_cpuinfo_s390(&info->value->u.s390x, cpu);
+        }
+
+        if (!cur_item) {
+            head = cur_item = info;
+        } else {
+            cur_item->next = info;
+            cur_item = info;
+        }
+    }
+
+    return head;
+}
+
+MachineInfoList *qmp_query_machines(Error **errp)
+{
+    GSList *el, *machines = object_class_get_list(TYPE_MACHINE, false);
+    MachineInfoList *mach_list = NULL;
+
+    for (el = machines; el; el = el->next) {
+        MachineClass *mc = el->data;
+        MachineInfoList *entry;
+        MachineInfo *info;
+
+        info = g_malloc0(sizeof(*info));
+        if (mc->is_default) {
+            info->has_is_default = true;
+            info->is_default = true;
+        }
+
+        if (mc->alias) {
+            info->has_alias = true;
+            info->alias = g_strdup(mc->alias);
+        }
+
+        info->name = g_strdup(mc->name);
+        info->cpu_max = !mc->max_cpus ? 1 : mc->max_cpus;
+        info->hotpluggable_cpus = mc->has_hotpluggable_cpus;
+
+        entry = g_malloc0(sizeof(*entry));
+        entry->value = info;
+        entry->next = mach_list;
+        mach_list = entry;
+    }
+
+    g_slist_free(machines);
+    return mach_list;
+}
+
+CurrentMachineParams *qmp_query_current_machine(Error **errp)
+{
+    CurrentMachineParams *params = g_malloc0(sizeof(*params));
+    params->wakeup_suspend_support = qemu_wakeup_suspend_enabled();
+
+    return params;
+}
+
+HotpluggableCPUList *qmp_query_hotpluggable_cpus(Error **errp)
+{
+    MachineState *ms = MACHINE(qdev_get_machine());
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+
+    if (!mc->has_hotpluggable_cpus) {
+        error_setg(errp, QERR_FEATURE_DISABLED, "query-hotpluggable-cpus");
+        return NULL;
+    }
+
+    return machine_query_hotpluggable_cpus(ms);
+}
+
+void qmp_cpu_add(int64_t id, Error **errp)
+{
+    MachineClass *mc;
+
+    mc = MACHINE_GET_CLASS(current_machine);
+    if (mc->hot_add_cpu) {
+        mc->hot_add_cpu(id, errp);
+    } else {
+        error_setg(errp, "Not supported");
+    }
+}
+
+void qmp_set_numa_node(NumaOptions *cmd, Error **errp)
+{
+    if (!runstate_check(RUN_STATE_PRECONFIG)) {
+        error_setg(errp, "The command is permitted only in '%s' state",
+                   RunState_str(RUN_STATE_PRECONFIG));
+         return;
+    }
+
+    set_numa_options(MACHINE(qdev_get_machine()), cmd, errp);
+}
+
+static int query_memdev(Object *obj, void *opaque)
+{
+    MemdevList **list = opaque;
+    MemdevList *m = NULL;
+
+    if (object_dynamic_cast(obj, TYPE_MEMORY_BACKEND)) {
+        m = g_malloc0(sizeof(*m));
+
+        m->value = g_malloc0(sizeof(*m->value));
+
+        m->value->id = object_get_canonical_path_component(obj);
+        m->value->has_id = !!m->value->id;
+
+        m->value->size = object_property_get_uint(obj, "size",
+                                                  &error_abort);
+        m->value->merge = object_property_get_bool(obj, "merge",
+                                                   &error_abort);
+        m->value->dump = object_property_get_bool(obj, "dump",
+                                                  &error_abort);
+        m->value->prealloc = object_property_get_bool(obj,
+                                                      "prealloc",
+                                                      &error_abort);
+        m->value->policy = object_property_get_enum(obj,
+                                                    "policy",
+                                                    "HostMemPolicy",
+                                                    &error_abort);
+        object_property_get_uint16List(obj, "host-nodes",
+                                       &m->value->host_nodes,
+                                       &error_abort);
+
+        m->next = *list;
+        *list = m;
+    }
+
+    return 0;
+}
+
+MemdevList *qmp_query_memdev(Error **errp)
+{
+    Object *obj = object_get_objects_root();
+    MemdevList *list = NULL;
+
+    object_child_foreach(obj, query_memdev, &list);
+    return list;
+}
diff --git a/hw/core/numa.c b/hw/core/numa.c
new file mode 100644
index 0000000000..66119d181b
--- /dev/null
+++ b/hw/core/numa.c
@@ -0,0 +1,605 @@
+/*
+ * NUMA parameter parsing routines
+ *
+ * Copyright (c) 2014 Fujitsu Ltd.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "qemu/osdep.h"
+#include "sysemu/numa.h"
+#include "exec/cpu-common.h"
+#include "exec/ramlist.h"
+#include "qemu/bitmap.h"
+#include "qemu/error-report.h"
+#include "qapi/error.h"
+#include "qapi/opts-visitor.h"
+#include "qapi/qapi-visit-machine.h"
+#include "hw/mem/pc-dimm.h"
+#include "hw/mem/memory-device.h"
+#include "qemu/option.h"
+#include "qemu/config-file.h"
+#include "qemu/cutils.h"
+
+QemuOptsList qemu_numa_opts = {
+    .name = "numa",
+    .implied_opt_name = "type",
+    .head = QTAILQ_HEAD_INITIALIZER(qemu_numa_opts.head),
+    .desc = { { 0 } } /* validated with OptsVisitor */
+};
+
+static int have_memdevs = -1;
+static int max_numa_nodeid; /* Highest specified NUMA node ID, plus one.
+                             * For all nodes, nodeid < max_numa_nodeid
+                             */
+int nb_numa_nodes;
+bool have_numa_distance;
+NodeInfo numa_info[MAX_NODES];
+
+
+static void parse_numa_node(MachineState *ms, NumaNodeOptions *node,
+                            Error **errp)
+{
+    Error *err = NULL;
+    uint16_t nodenr;
+    uint16List *cpus = NULL;
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+
+    if (node->has_nodeid) {
+        nodenr = node->nodeid;
+    } else {
+        nodenr = nb_numa_nodes;
+    }
+
+    if (nodenr >= MAX_NODES) {
+        error_setg(errp, "Max number of NUMA nodes reached: %"
+                   PRIu16 "", nodenr);
+        return;
+    }
+
+    if (numa_info[nodenr].present) {
+        error_setg(errp, "Duplicate NUMA nodeid: %" PRIu16, nodenr);
+        return;
+    }
+
+    if (!mc->cpu_index_to_instance_props || !mc->get_default_cpu_node_id) {
+        error_setg(errp, "NUMA is not supported by this machine-type");
+        return;
+    }
+    for (cpus = node->cpus; cpus; cpus = cpus->next) {
+        CpuInstanceProperties props;
+        if (cpus->value >= max_cpus) {
+            error_setg(errp,
+                       "CPU index (%" PRIu16 ")"
+                       " should be smaller than maxcpus (%d)",
+                       cpus->value, max_cpus);
+            return;
+        }
+        props = mc->cpu_index_to_instance_props(ms, cpus->value);
+        props.node_id = nodenr;
+        props.has_node_id = true;
+        machine_set_cpu_numa_node(ms, &props, &err);
+        if (err) {
+            error_propagate(errp, err);
+            return;
+        }
+    }
+
+    if (node->has_mem && node->has_memdev) {
+        error_setg(errp, "cannot specify both mem= and memdev=");
+        return;
+    }
+
+    if (have_memdevs == -1) {
+        have_memdevs = node->has_memdev;
+    }
+    if (node->has_memdev != have_memdevs) {
+        error_setg(errp, "memdev option must be specified for either "
+                   "all or no nodes");
+        return;
+    }
+
+    if (node->has_mem) {
+        numa_info[nodenr].node_mem = node->mem;
+    }
+    if (node->has_memdev) {
+        Object *o;
+        o = object_resolve_path_type(node->memdev, TYPE_MEMORY_BACKEND, NULL);
+        if (!o) {
+            error_setg(errp, "memdev=%s is ambiguous", node->memdev);
+            return;
+        }
+
+        object_ref(o);
+        numa_info[nodenr].node_mem = object_property_get_uint(o, "size", NULL);
+        numa_info[nodenr].node_memdev = MEMORY_BACKEND(o);
+    }
+    numa_info[nodenr].present = true;
+    max_numa_nodeid = MAX(max_numa_nodeid, nodenr + 1);
+    nb_numa_nodes++;
+}
+
+static void parse_numa_distance(NumaDistOptions *dist, Error **errp)
+{
+    uint16_t src = dist->src;
+    uint16_t dst = dist->dst;
+    uint8_t val = dist->val;
+
+    if (src >= MAX_NODES || dst >= MAX_NODES) {
+        error_setg(errp, "Parameter '%s' expects an integer between 0 and %d",
+                   src >= MAX_NODES ? "src" : "dst", MAX_NODES - 1);
+        return;
+    }
+
+    if (!numa_info[src].present || !numa_info[dst].present) {
+        error_setg(errp, "Source/Destination NUMA node is missing. "
+                   "Please use '-numa node' option to declare it first.");
+        return;
+    }
+
+    if (val < NUMA_DISTANCE_MIN) {
+        error_setg(errp, "NUMA distance (%" PRIu8 ") is invalid, "
+                   "it shouldn't be less than %d.",
+                   val, NUMA_DISTANCE_MIN);
+        return;
+    }
+
+    if (src == dst && val != NUMA_DISTANCE_MIN) {
+        error_setg(errp, "Local distance of node %d should be %d.",
+                   src, NUMA_DISTANCE_MIN);
+        return;
+    }
+
+    numa_info[src].distance[dst] = val;
+    have_numa_distance = true;
+}
+
+void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp)
+{
+    Error *err = NULL;
+
+    switch (object->type) {
+    case NUMA_OPTIONS_TYPE_NODE:
+        parse_numa_node(ms, &object->u.node, &err);
+        if (err) {
+            goto end;
+        }
+        break;
+    case NUMA_OPTIONS_TYPE_DIST:
+        parse_numa_distance(&object->u.dist, &err);
+        if (err) {
+            goto end;
+        }
+        break;
+    case NUMA_OPTIONS_TYPE_CPU:
+        if (!object->u.cpu.has_node_id) {
+            error_setg(&err, "Missing mandatory node-id property");
+            goto end;
+        }
+        if (!numa_info[object->u.cpu.node_id].present) {
+            error_setg(&err, "Invalid node-id=%" PRId64 ", NUMA node must be "
+                "defined with -numa node,nodeid=ID before it's used with "
+                "-numa cpu,node-id=ID", object->u.cpu.node_id);
+            goto end;
+        }
+
+        machine_set_cpu_numa_node(ms, qapi_NumaCpuOptions_base(&object->u.cpu),
+                                  &err);
+        break;
+    default:
+        abort();
+    }
+
+end:
+    error_propagate(errp, err);
+}
+
+static int parse_numa(void *opaque, QemuOpts *opts, Error **errp)
+{
+    NumaOptions *object = NULL;
+    MachineState *ms = MACHINE(opaque);
+    Error *err = NULL;
+    Visitor *v = opts_visitor_new(opts);
+
+    visit_type_NumaOptions(v, NULL, &object, &err);
+    visit_free(v);
+    if (err) {
+        goto end;
+    }
+
+    /* Fix up legacy suffix-less format */
+    if ((object->type == NUMA_OPTIONS_TYPE_NODE) && object->u.node.has_mem) {
+        const char *mem_str = qemu_opt_get(opts, "mem");
+        qemu_strtosz_MiB(mem_str, NULL, &object->u.node.mem);
+    }
+
+    set_numa_options(ms, object, &err);
+
+end:
+    qapi_free_NumaOptions(object);
+    if (err) {
+        error_propagate(errp, err);
+        return -1;
+    }
+
+    return 0;
+}
+
+/* If all node pair distances are symmetric, then only distances
+ * in one direction are enough. If there is even one asymmetric
+ * pair, though, then all distances must be provided. The
+ * distance from a node to itself is always NUMA_DISTANCE_MIN,
+ * so providing it is never necessary.
+ */
+static void validate_numa_distance(void)
+{
+    int src, dst;
+    bool is_asymmetrical = false;
+
+    for (src = 0; src < nb_numa_nodes; src++) {
+        for (dst = src; dst < nb_numa_nodes; dst++) {
+            if (numa_info[src].distance[dst] == 0 &&
+                numa_info[dst].distance[src] == 0) {
+                if (src != dst) {
+                    error_report("The distance between node %d and %d is "
+                                 "missing, at least one distance value "
+                                 "between each nodes should be provided.",
+                                 src, dst);
+                    exit(EXIT_FAILURE);
+                }
+            }
+
+            if (numa_info[src].distance[dst] != 0 &&
+                numa_info[dst].distance[src] != 0 &&
+                numa_info[src].distance[dst] !=
+                numa_info[dst].distance[src]) {
+                is_asymmetrical = true;
+            }
+        }
+    }
+
+    if (is_asymmetrical) {
+        for (src = 0; src < nb_numa_nodes; src++) {
+            for (dst = 0; dst < nb_numa_nodes; dst++) {
+                if (src != dst && numa_info[src].distance[dst] == 0) {
+                    error_report("At least one asymmetrical pair of "
+                            "distances is given, please provide distances "
+                            "for both directions of all node pairs.");
+                    exit(EXIT_FAILURE);
+                }
+            }
+        }
+    }
+}
+
+static void complete_init_numa_distance(void)
+{
+    int src, dst;
+
+    /* Fixup NUMA distance by symmetric policy because if it is an
+     * asymmetric distance table, it should be a complete table and
+     * there would not be any missing distance except local node, which
+     * is verified by validate_numa_distance above.
+     */
+    for (src = 0; src < nb_numa_nodes; src++) {
+        for (dst = 0; dst < nb_numa_nodes; dst++) {
+            if (numa_info[src].distance[dst] == 0) {
+                if (src == dst) {
+                    numa_info[src].distance[dst] = NUMA_DISTANCE_MIN;
+                } else {
+                    numa_info[src].distance[dst] = numa_info[dst].distance[src];
+                }
+            }
+        }
+    }
+}
+
+void numa_legacy_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
+                                 int nb_nodes, ram_addr_t size)
+{
+    int i;
+    uint64_t usedmem = 0;
+
+    /* Align each node according to the alignment
+     * requirements of the machine class
+     */
+
+    for (i = 0; i < nb_nodes - 1; i++) {
+        nodes[i].node_mem = (size / nb_nodes) &
+                            ~((1 << mc->numa_mem_align_shift) - 1);
+        usedmem += nodes[i].node_mem;
+    }
+    nodes[i].node_mem = size - usedmem;
+}
+
+void numa_default_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
+                                  int nb_nodes, ram_addr_t size)
+{
+    int i;
+    uint64_t usedmem = 0, node_mem;
+    uint64_t granularity = size / nb_nodes;
+    uint64_t propagate = 0;
+
+    for (i = 0; i < nb_nodes - 1; i++) {
+        node_mem = (granularity + propagate) &
+                   ~((1 << mc->numa_mem_align_shift) - 1);
+        propagate = granularity + propagate - node_mem;
+        nodes[i].node_mem = node_mem;
+        usedmem += node_mem;
+    }
+    nodes[i].node_mem = size - usedmem;
+}
+
+void numa_complete_configuration(MachineState *ms)
+{
+    int i;
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+
+    /*
+     * If memory hotplug is enabled (slots > 0) but without '-numa'
+     * options explicitly on CLI, guestes will break.
+     *
+     *   Windows: won't enable memory hotplug without SRAT table at all
+     *
+     *   Linux: if QEMU is started with initial memory all below 4Gb
+     *   and no SRAT table present, guest kernel will use nommu DMA ops,
+     *   which breaks 32bit hw drivers when memory is hotplugged and
+     *   guest tries to use it with that drivers.
+     *
+     * Enable NUMA implicitly by adding a new NUMA node automatically.
+     */
+    if (ms->ram_slots > 0 && nb_numa_nodes == 0 &&
+        mc->auto_enable_numa_with_memhp) {
+            NumaNodeOptions node = { };
+            parse_numa_node(ms, &node, &error_abort);
+    }
+
+    assert(max_numa_nodeid <= MAX_NODES);
+
+    /* No support for sparse NUMA node IDs yet: */
+    for (i = max_numa_nodeid - 1; i >= 0; i--) {
+        /* Report large node IDs first, to make mistakes easier to spot */
+        if (!numa_info[i].present) {
+            error_report("numa: Node ID missing: %d", i);
+            exit(1);
+        }
+    }
+
+    /* This must be always true if all nodes are present: */
+    assert(nb_numa_nodes == max_numa_nodeid);
+
+    if (nb_numa_nodes > 0) {
+        uint64_t numa_total;
+
+        if (nb_numa_nodes > MAX_NODES) {
+            nb_numa_nodes = MAX_NODES;
+        }
+
+        /* If no memory size is given for any node, assume the default case
+         * and distribute the available memory equally across all nodes
+         */
+        for (i = 0; i < nb_numa_nodes; i++) {
+            if (numa_info[i].node_mem != 0) {
+                break;
+            }
+        }
+        if (i == nb_numa_nodes) {
+            assert(mc->numa_auto_assign_ram);
+            mc->numa_auto_assign_ram(mc, numa_info, nb_numa_nodes, ram_size);
+        }
+
+        numa_total = 0;
+        for (i = 0; i < nb_numa_nodes; i++) {
+            numa_total += numa_info[i].node_mem;
+        }
+        if (numa_total != ram_size) {
+            error_report("total memory for NUMA nodes (0x%" PRIx64 ")"
+                         " should equal RAM size (0x" RAM_ADDR_FMT ")",
+                         numa_total, ram_size);
+            exit(1);
+        }
+
+        /* QEMU needs at least all unique node pair distances to build
+         * the whole NUMA distance table. QEMU treats the distance table
+         * as symmetric by default, i.e. distance A->B == distance B->A.
+         * Thus, QEMU is able to complete the distance table
+         * initialization even though only distance A->B is provided and
+         * distance B->A is not. QEMU knows the distance of a node to
+         * itself is always 10, so A->A distances may be omitted. When
+         * the distances of two nodes of a pair differ, i.e. distance
+         * A->B != distance B->A, then that means the distance table is
+         * asymmetric. In this case, the distances for both directions
+         * of all node pairs are required.
+         */
+        if (have_numa_distance) {
+            /* Validate enough NUMA distance information was provided. */
+            validate_numa_distance();
+
+            /* Validation succeeded, now fill in any missing distances. */
+            complete_init_numa_distance();
+        }
+    }
+}
+
+void parse_numa_opts(MachineState *ms)
+{
+    qemu_opts_foreach(qemu_find_opts("numa"), parse_numa, ms, &error_fatal);
+}
+
+void numa_cpu_pre_plug(const CPUArchId *slot, DeviceState *dev, Error **errp)
+{
+    int node_id = object_property_get_int(OBJECT(dev), "node-id", &error_abort);
+
+    if (node_id == CPU_UNSET_NUMA_NODE_ID) {
+        /* due to bug in libvirt, it doesn't pass node-id from props on
+         * device_add as expected, so we have to fix it up here */
+        if (slot->props.has_node_id) {
+            object_property_set_int(OBJECT(dev), slot->props.node_id,
+                                    "node-id", errp);
+        }
+    } else if (node_id != slot->props.node_id) {
+        error_setg(errp, "invalid node-id, must be %"PRId64,
+                   slot->props.node_id);
+    }
+}
+
+static void allocate_system_memory_nonnuma(MemoryRegion *mr, Object *owner,
+                                           const char *name,
+                                           uint64_t ram_size)
+{
+    if (mem_path) {
+#ifdef __linux__
+        Error *err = NULL;
+        memory_region_init_ram_from_file(mr, owner, name, ram_size, 0, 0,
+                                         mem_path, &err);
+        if (err) {
+            error_report_err(err);
+            if (mem_prealloc) {
+                exit(1);
+            }
+            error_report("falling back to regular RAM allocation.");
+
+            /* Legacy behavior: if allocation failed, fall back to
+             * regular RAM allocation.
+             */
+            mem_path = NULL;
+            memory_region_init_ram_nomigrate(mr, owner, name, ram_size, &error_fatal);
+        }
+#else
+        fprintf(stderr, "-mem-path not supported on this host\n");
+        exit(1);
+#endif
+    } else {
+        memory_region_init_ram_nomigrate(mr, owner, name, ram_size, &error_fatal);
+    }
+    vmstate_register_ram_global(mr);
+}
+
+void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,
+                                          const char *name,
+                                          uint64_t ram_size)
+{
+    uint64_t addr = 0;
+    int i;
+
+    if (nb_numa_nodes == 0 || !have_memdevs) {
+        allocate_system_memory_nonnuma(mr, owner, name, ram_size);
+        return;
+    }
+
+    memory_region_init(mr, owner, name, ram_size);
+    for (i = 0; i < nb_numa_nodes; i++) {
+        uint64_t size = numa_info[i].node_mem;
+        HostMemoryBackend *backend = numa_info[i].node_memdev;
+        if (!backend) {
+            continue;
+        }
+        MemoryRegion *seg = host_memory_backend_get_memory(backend);
+
+        if (memory_region_is_mapped(seg)) {
+            char *path = object_get_canonical_path_component(OBJECT(backend));
+            error_report("memory backend %s is used multiple times. Each "
+                         "-numa option must use a different memdev value.",
+                         path);
+            g_free(path);
+            exit(1);
+        }
+
+        host_memory_backend_set_mapped(backend, true);
+        memory_region_add_subregion(mr, addr, seg);
+        vmstate_register_ram_global(seg);
+        addr += size;
+    }
+}
+
+static void numa_stat_memory_devices(NumaNodeMem node_mem[])
+{
+    MemoryDeviceInfoList *info_list = qmp_memory_device_list();
+    MemoryDeviceInfoList *info;
+    PCDIMMDeviceInfo     *pcdimm_info;
+    VirtioPMEMDeviceInfo *vpi;
+
+    for (info = info_list; info; info = info->next) {
+        MemoryDeviceInfo *value = info->value;
+
+        if (value) {
+            switch (value->type) {
+            case MEMORY_DEVICE_INFO_KIND_DIMM:
+            case MEMORY_DEVICE_INFO_KIND_NVDIMM:
+                pcdimm_info = value->type == MEMORY_DEVICE_INFO_KIND_DIMM ?
+                              value->u.dimm.data : value->u.nvdimm.data;
+                node_mem[pcdimm_info->node].node_mem += pcdimm_info->size;
+                node_mem[pcdimm_info->node].node_plugged_mem +=
+                    pcdimm_info->size;
+                break;
+            case MEMORY_DEVICE_INFO_KIND_VIRTIO_PMEM:
+                vpi = value->u.virtio_pmem.data;
+                /* TODO: once we support numa, assign to right node */
+                node_mem[0].node_mem += vpi->size;
+                node_mem[0].node_plugged_mem += vpi->size;
+                break;
+            default:
+                g_assert_not_reached();
+            }
+        }
+    }
+    qapi_free_MemoryDeviceInfoList(info_list);
+}
+
+void query_numa_node_mem(NumaNodeMem node_mem[])
+{
+    int i;
+
+    if (nb_numa_nodes <= 0) {
+        return;
+    }
+
+    numa_stat_memory_devices(node_mem);
+    for (i = 0; i < nb_numa_nodes; i++) {
+        node_mem[i].node_mem += numa_info[i].node_mem;
+    }
+}
+
+void ram_block_notifier_add(RAMBlockNotifier *n)
+{
+    QLIST_INSERT_HEAD(&ram_list.ramblock_notifiers, n, next);
+}
+
+void ram_block_notifier_remove(RAMBlockNotifier *n)
+{
+    QLIST_REMOVE(n, next);
+}
+
+void ram_block_notify_add(void *host, size_t size)
+{
+    RAMBlockNotifier *notifier;
+
+    QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) {
+        notifier->ram_block_added(notifier, host, size);
+    }
+}
+
+void ram_block_notify_remove(void *host, size_t size)
+{
+    RAMBlockNotifier *notifier;
+
+    QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) {
+        notifier->ram_block_removed(notifier, host, size);
+    }
+}
diff --git a/hw/core/qdev.c b/hw/core/qdev.c
index f9b6efe509..94ebc0a4a1 100644
--- a/hw/core/qdev.c
+++ b/hw/core/qdev.c
@@ -29,7 +29,7 @@
 #include "hw/qdev.h"
 #include "sysemu/sysemu.h"
 #include "qapi/error.h"
-#include "qapi/qapi-events-misc.h"
+#include "qapi/qapi-events-qdev.h"
 #include "qapi/qmp/qerror.h"
 #include "qapi/visitor.h"
 #include "qemu/error-report.h"
diff --git a/hw/display/Kconfig b/hw/display/Kconfig
index 910dccb2f7..cbdf7b1a67 100644
--- a/hw/display/Kconfig
+++ b/hw/display/Kconfig
@@ -130,3 +130,5 @@ config ATI_VGA
     default y if PCI_DEVICES
     depends on PCI
     select VGA
+    select BITBANG_I2C
+    select DDC
diff --git a/hw/display/ati.c b/hw/display/ati.c
index 76595d9511..c1d9d1518f 100644
--- a/hw/display/ati.c
+++ b/hw/display/ati.c
@@ -26,6 +26,7 @@
 #include "qapi/error.h"
 #include "hw/hw.h"
 #include "ui/console.h"
+#include "hw/display/i2c-ddc.h"
 #include "trace.h"
 
 #define ATI_DEBUG_HW_CURSOR 0
@@ -215,6 +216,24 @@ static void ati_cursor_draw_line(VGACommonState *vga, uint8_t *d, int scr_y)
     }
 }
 
+static uint64_t ati_i2c(bitbang_i2c_interface *i2c, uint64_t data, int base)
+{
+    bool c = (data & BIT(base + 17) ? !!(data & BIT(base + 1)) : 1);
+    bool d = (data & BIT(base + 16) ? !!(data & BIT(base)) : 1);
+
+    bitbang_i2c_set(i2c, BITBANG_I2C_SCL, c);
+    d = bitbang_i2c_set(i2c, BITBANG_I2C_SDA, d);
+
+    data &= ~0xf00ULL;
+    if (c) {
+        data |= BIT(base + 9);
+    }
+    if (d) {
+        data |= BIT(base + 8);
+    }
+    return data;
+}
+
 static inline uint64_t ati_reg_read_offs(uint32_t reg, int offs,
                                          unsigned int size)
 {
@@ -266,7 +285,16 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size)
     case DAC_CNTL:
         val = s->regs.dac_cntl;
         break;
-/*    case GPIO_MONID: FIXME hook up DDC I2C here */
+    case GPIO_VGA_DDC:
+        val = s->regs.gpio_vga_ddc;
+        break;
+    case GPIO_DVI_DDC:
+        val = s->regs.gpio_dvi_ddc;
+        break;
+    case GPIO_MONID ... GPIO_MONID + 3:
+        val = ati_reg_read_offs(s->regs.gpio_monid,
+                                addr - GPIO_MONID, size);
+        break;
     case PALETTE_INDEX:
         /* FIXME unaligned access */
         val = vga_ioport_read(&s->vga, VGA_PEL_IR) << 16;
@@ -391,9 +419,15 @@ static uint64_t ati_mm_read(void *opaque, hwaddr addr, unsigned int size)
         break;
     case DEFAULT_OFFSET:
         val = s->regs.default_offset;
+        if (s->dev_id != PCI_DEVICE_ID_ATI_RAGE128_PF) {
+            val >>= 10;
+            val |= s->regs.default_pitch << 16;
+            val |= s->regs.default_tile << 30;
+        }
         break;
     case DEFAULT_PITCH:
         val = s->regs.default_pitch;
+        val |= s->regs.default_tile << 16;
         break;
     case DEFAULT_SC_BOTTOM_RIGHT:
         val = s->regs.default_sc_bottom_right;
@@ -497,7 +531,33 @@ static void ati_mm_write(void *opaque, hwaddr addr,
         s->regs.dac_cntl = data & 0xffffe3ff;
         s->vga.dac_8bit = !!(data & DAC_8BIT_EN);
         break;
-/*    case GPIO_MONID: FIXME hook up DDC I2C here */
+    case GPIO_VGA_DDC:
+        if (s->dev_id != PCI_DEVICE_ID_ATI_RAGE128_PF) {
+            /* FIXME: Maybe add a property to select VGA or DVI port? */
+        }
+        break;
+    case GPIO_DVI_DDC:
+        if (s->dev_id != PCI_DEVICE_ID_ATI_RAGE128_PF) {
+            s->regs.gpio_dvi_ddc = ati_i2c(&s->bbi2c, data, 0);
+        }
+        break;
+    case GPIO_MONID ... GPIO_MONID + 3:
+        /* FIXME What does Radeon have here? */
+        if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
+            ati_reg_write_offs(&s->regs.gpio_monid,
+                               addr - GPIO_MONID, data, size);
+            /*
+             * Rage128p accesses DDC used to get EDID via these bits.
+             * Only touch i2c when write overlaps 3rd byte because some
+             * drivers access this reg via multiple partial writes and
+             * without this spurious bits would be sent.
+             */
+            if ((s->regs.gpio_monid & BIT(25)) &&
+                addr <= GPIO_MONID + 2 && addr + size > GPIO_MONID + 2) {
+                s->regs.gpio_monid = ati_i2c(&s->bbi2c, s->regs.gpio_monid, 1);
+            }
+        }
+        break;
     case PALETTE_INDEX ... PALETTE_INDEX + 3:
         if (size == 4) {
             vga_ioport_write(&s->vga, VGA_PEL_IR, (data >> 16) & 0xff);
@@ -628,22 +688,22 @@ static void ati_mm_write(void *opaque, hwaddr addr,
         break;
     case SRC_PITCH_OFFSET:
         if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
-            s->regs.src_offset = (data & 0x1fffff) << 5;
-            s->regs.src_pitch = (data >> 21) & 0x3ff;
+            s->regs.src_offset = (data & 0x1fffff) << 4;
+            s->regs.src_pitch = (data & 0x7fe00000) >> 21;
             s->regs.src_tile = data >> 31;
         } else {
-            s->regs.src_offset = (data & 0x3fffff) << 11;
+            s->regs.src_offset = (data & 0x3fffff) << 10;
             s->regs.src_pitch = (data & 0x3fc00000) >> 16;
             s->regs.src_tile = (data >> 30) & 1;
         }
         break;
     case DST_PITCH_OFFSET:
         if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
-            s->regs.dst_offset = (data & 0x1fffff) << 5;
-            s->regs.dst_pitch = (data >> 21) & 0x3ff;
+            s->regs.dst_offset = (data & 0x1fffff) << 4;
+            s->regs.dst_pitch = (data & 0x7fe00000) >> 21;
             s->regs.dst_tile = data >> 31;
         } else {
-            s->regs.dst_offset = (data & 0x3fffff) << 11;
+            s->regs.dst_offset = (data & 0x3fffff) << 10;
             s->regs.dst_pitch = (data & 0x3fc00000) >> 16;
             s->regs.dst_tile = data >> 30;
         }
@@ -723,13 +783,19 @@ static void ati_mm_write(void *opaque, hwaddr addr,
         s->regs.dp_write_mask = data;
         break;
     case DEFAULT_OFFSET:
-        data &= (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF ?
-                 0x03fffc00 : 0xfffffc00);
-        s->regs.default_offset = data;
+        if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
+            s->regs.default_offset = data & 0xfffffff0;
+        } else {
+            /* Radeon has DEFAULT_PITCH_OFFSET here like DST_PITCH_OFFSET */
+            s->regs.default_offset = (data & 0x3fffff) << 10;
+            s->regs.default_pitch = (data & 0x3fc00000) >> 16;
+            s->regs.default_tile = data >> 30;
+        }
         break;
     case DEFAULT_PITCH:
         if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
-            s->regs.default_pitch = data & 0x103ff;
+            s->regs.default_pitch = data & 0x3fff;
+            s->regs.default_tile = (data >> 16) & 1;
         }
         break;
     case DEFAULT_SC_BOTTOM_RIGHT:
@@ -788,6 +854,12 @@ static void ati_vga_realize(PCIDevice *dev, Error **errp)
         vga->cursor_draw_line = ati_cursor_draw_line;
     }
 
+    /* ddc, edid */
+    I2CBus *i2cbus = i2c_init_bus(DEVICE(s), "ati-vga.ddc");
+    bitbang_i2c_init(&s->bbi2c, i2cbus);
+    I2CSlave *i2cddc = I2C_SLAVE(qdev_create(BUS(i2cbus), TYPE_I2CDDC));
+    i2c_set_slave_address(i2cddc, 0x50);
+
     /* mmio register space */
     memory_region_init_io(&s->mm, OBJECT(s), &ati_mm_ops, s,
                           "ati.mmregs", 0x4000);
@@ -837,7 +909,7 @@ static void ati_vga_class_init(ObjectClass *klass, void *data)
     k->class_id = PCI_CLASS_DISPLAY_VGA;
     k->vendor_id = PCI_VENDOR_ID_ATI;
     k->device_id = PCI_DEVICE_ID_ATI_RAGE128_PF;
-    k->romfile = "vgabios-stdvga.bin";
+    k->romfile = "vgabios-ati.bin";
     k->realize = ati_vga_realize;
     k->exit = ati_vga_exit;
 }
diff --git a/hw/display/ati_2d.c b/hw/display/ati_2d.c
index d83c29c6d9..2dbf53f039 100644
--- a/hw/display/ati_2d.c
+++ b/hw/display/ati_2d.c
@@ -51,8 +51,9 @@ void ati_2d_blt(ATIVGAState *s)
             s->vga.vbe_start_addr, surface_data(ds), surface_stride(ds),
             surface_bits_per_pixel(ds),
             (s->regs.dp_mix & GMC_ROP3_MASK) >> 16);
-    DPRINTF("%d %d, %d %d, (%d,%d) -> (%d,%d) %dx%d\n", s->regs.src_offset,
-            s->regs.dst_offset, s->regs.src_pitch, s->regs.dst_pitch,
+    DPRINTF("%d %d %d, %d %d %d, (%d,%d) -> (%d,%d) %dx%d\n",
+            s->regs.src_offset, s->regs.dst_offset, s->regs.default_offset,
+            s->regs.src_pitch, s->regs.dst_pitch, s->regs.default_pitch,
             s->regs.src_x, s->regs.src_y, s->regs.dst_x, s->regs.dst_y,
             s->regs.dst_width, s->regs.dst_height);
     switch (s->regs.dp_mix & GMC_ROP3_MASK) {
@@ -60,10 +61,16 @@ void ati_2d_blt(ATIVGAState *s)
     {
         uint8_t *src_bits, *dst_bits, *end;
         int src_stride, dst_stride, bpp = ati_bpp_from_datatype(s);
-        src_bits = s->vga.vram_ptr + s->regs.src_offset;
-        dst_bits = s->vga.vram_ptr + s->regs.dst_offset;
-        src_stride = s->regs.src_pitch;
-        dst_stride = s->regs.dst_pitch;
+        src_bits = s->vga.vram_ptr +
+                   (s->regs.dp_gui_master_cntl & GMC_SRC_PITCH_OFFSET_CNTL ?
+                    s->regs.src_offset : s->regs.default_offset);
+        dst_bits = s->vga.vram_ptr +
+                   (s->regs.dp_gui_master_cntl & GMC_DST_PITCH_OFFSET_CNTL ?
+                    s->regs.dst_offset : s->regs.default_offset);
+        src_stride = (s->regs.dp_gui_master_cntl & GMC_SRC_PITCH_OFFSET_CNTL ?
+                      s->regs.src_pitch : s->regs.default_pitch);
+        dst_stride = (s->regs.dp_gui_master_cntl & GMC_DST_PITCH_OFFSET_CNTL ?
+                      s->regs.dst_pitch : s->regs.default_pitch);
 
         if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
             src_bits += s->regs.crtc_offset & 0x07ffffff;
@@ -111,8 +118,11 @@ void ati_2d_blt(ATIVGAState *s)
         uint8_t *dst_bits, *end;
         int dst_stride, bpp = ati_bpp_from_datatype(s);
         uint32_t filler = 0;
-        dst_bits = s->vga.vram_ptr + s->regs.dst_offset;
-        dst_stride = s->regs.dst_pitch;
+        dst_bits = s->vga.vram_ptr +
+                   (s->regs.dp_gui_master_cntl & GMC_DST_PITCH_OFFSET_CNTL ?
+                    s->regs.dst_offset : s->regs.default_offset);
+        dst_stride = (s->regs.dp_gui_master_cntl & GMC_DST_PITCH_OFFSET_CNTL ?
+                      s->regs.dst_pitch : s->regs.default_pitch);
 
         if (s->dev_id == PCI_DEVICE_ID_ATI_RAGE128_PF) {
             dst_bits += s->regs.crtc_offset & 0x07ffffff;
diff --git a/hw/display/ati_dbg.c b/hw/display/ati_dbg.c
index b045f81d06..88b3a11315 100644
--- a/hw/display/ati_dbg.c
+++ b/hw/display/ati_dbg.c
@@ -19,6 +19,8 @@ static struct ati_regdesc ati_reg_names[] = {
     {"CRTC_GEN_CNTL", 0x0050},
     {"CRTC_EXT_CNTL", 0x0054},
     {"DAC_CNTL", 0x0058},
+    {"GPIO_VGA_DDC", 0x0060},
+    {"GPIO_DVI_DDC", 0x0064},
     {"GPIO_MONID", 0x0068},
     {"I2C_CNTL_1", 0x0094},
     {"PALETTE_INDEX", 0x00b0},
diff --git a/hw/display/ati_int.h b/hw/display/ati_int.h
index 2f426064cf..31a1927b3e 100644
--- a/hw/display/ati_int.h
+++ b/hw/display/ati_int.h
@@ -10,6 +10,7 @@
 #define ATI_INT_H
 
 #include "hw/pci/pci.h"
+#include "hw/i2c/bitbang_i2c.h"
 #include "vga_int.h"
 
 /*#define DEBUG_ATI*/
@@ -35,6 +36,9 @@ typedef struct ATIVGARegs {
     uint32_t crtc_gen_cntl;
     uint32_t crtc_ext_cntl;
     uint32_t dac_cntl;
+    uint32_t gpio_vga_ddc;
+    uint32_t gpio_dvi_ddc;
+    uint32_t gpio_monid;
     uint32_t crtc_h_total_disp;
     uint32_t crtc_h_sync_strt_wid;
     uint32_t crtc_v_total_disp;
@@ -70,6 +74,7 @@ typedef struct ATIVGARegs {
     uint32_t dp_write_mask;
     uint32_t default_offset;
     uint32_t default_pitch;
+    uint32_t default_tile;
     uint32_t default_sc_bottom_right;
 } ATIVGARegs;
 
@@ -83,6 +88,7 @@ typedef struct ATIVGAState {
     uint16_t cursor_size;
     uint32_t cursor_offset;
     QEMUCursor *cursor;
+    bitbang_i2c_interface bbi2c;
     MemoryRegion io;
     MemoryRegion mm;
     ATIVGARegs regs;
diff --git a/hw/display/ati_regs.h b/hw/display/ati_regs.h
index 923bfd33ce..d7155c93d5 100644
--- a/hw/display/ati_regs.h
+++ b/hw/display/ati_regs.h
@@ -37,6 +37,8 @@
 #define CRTC_GEN_CNTL                           0x0050
 #define CRTC_EXT_CNTL                           0x0054
 #define DAC_CNTL                                0x0058
+#define GPIO_VGA_DDC                            0x0060
+#define GPIO_DVI_DDC                            0x0064
 #define GPIO_MONID                              0x0068
 #define I2C_CNTL_1                              0x0094
 #define PALETTE_INDEX                           0x00b0
@@ -368,8 +370,8 @@
 #define BRUSH_SOLIDCOLOR                        0x00000d00
 
 /* DP_GUI_MASTER_CNTL bit constants */
-#define GMC_SRC_PITCH_OFFSET_DEFAULT            0x00000000
-#define GMC_DST_PITCH_OFFSET_DEFAULT            0x00000000
+#define GMC_SRC_PITCH_OFFSET_CNTL               0x00000001
+#define GMC_DST_PITCH_OFFSET_CNTL               0x00000002
 #define GMC_SRC_CLIP_DEFAULT                    0x00000000
 #define GMC_DST_CLIP_DEFAULT                    0x00000000
 #define GMC_BRUSH_SOLIDCOLOR                    0x000000d0
diff --git a/hw/display/virtio-gpu.c b/hw/display/virtio-gpu.c
index 2b0f66b1d6..25d9e327fc 100644
--- a/hw/display/virtio-gpu.c
+++ b/hw/display/virtio-gpu.c
@@ -1002,6 +1002,11 @@ static int virtio_gpu_load(QEMUFile *f, void *opaque, size_t size,
 
     resource_id = qemu_get_be32(f);
     while (resource_id != 0) {
+        res = virtio_gpu_find_resource(g, resource_id);
+        if (res) {
+            return -EINVAL;
+        }
+
         res = g_new0(struct virtio_gpu_simple_resource, 1);
         res->resource_id = resource_id;
         res->width = qemu_get_be32(f);
@@ -1048,9 +1053,9 @@ static int virtio_gpu_load(QEMUFile *f, void *opaque, size_t size,
                 if (res->iov[i].iov_base) {
                     dma_memory_unmap(VIRTIO_DEVICE(g)->dma_as,
                                      res->iov[i].iov_base,
-                                     res->iov[i].iov_len,
+                                     len,
                                      DMA_DIRECTION_TO_DEVICE,
-                                     res->iov[i].iov_len);
+                                     0);
                 }
                 /* ...and the mappings for previous loop iterations */
                 res->iov_cnt = i;
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
index 6202f1150e..3c79913b31 100644
--- a/hw/display/xenfb.c
+++ b/hw/display/xenfb.c
@@ -32,10 +32,9 @@
 #include "ui/console.h"
 #include "hw/xen/xen-legacy-backend.h"
 
-#include <xen/event_channel.h>
-#include <xen/io/fbif.h>
-#include <xen/io/kbdif.h>
-#include <xen/io/protocols.h>
+#include "hw/xen/interface/io/fbif.h"
+#include "hw/xen/interface/io/kbdif.h"
+#include "hw/xen/interface/io/protocols.h"
 
 #include "trace.h"
 
diff --git a/hw/dma/rc4030.c b/hw/dma/rc4030.c
index 6ccafece18..155af9b26a 100644
--- a/hw/dma/rc4030.c
+++ b/hw/dma/rc4030.c
@@ -23,6 +23,7 @@
  */
 
 #include "qemu/osdep.h"
+#include "qemu/units.h"
 #include "hw/hw.h"
 #include "hw/mips/mips.h"
 #include "hw/sysbus.h"
@@ -57,8 +58,8 @@ typedef struct dma_pagetable_entry {
 
 #define TYPE_RC4030_IOMMU_MEMORY_REGION "rc4030-iommu-memory-region"
 
-typedef struct rc4030State
-{
+typedef struct rc4030State {
+
     SysBusDevice parent;
 
     uint32_t config; /* 0x0000: RC4030 config register */
@@ -151,8 +152,9 @@ static uint64_t rc4030_read(void *opaque, hwaddr addr, unsigned int size)
     case 0x0058:
         val = s->cache_bmask;
         /* HACK */
-        if (s->cache_bmask == (uint32_t)-1)
+        if (s->cache_bmask == (uint32_t)-1) {
             s->cache_bmask = 0;
+        }
         break;
     /* Remote Speed Registers */
     case 0x0070:
@@ -537,8 +539,9 @@ static void rc4030_reset(DeviceState *dev)
 
     s->memory_refresh_rate = 0x18186;
     s->nvram_protect = 7;
-    for (i = 0; i < 15; i++)
+    for (i = 0; i < 15; i++) {
         s->rem_speed[i] = 7;
+    }
     s->imr_jazz = 0x10; /* XXX: required by firmware, but why? */
     s->isr_jazz = 0;
 
@@ -550,7 +553,7 @@ static void rc4030_reset(DeviceState *dev)
 
 static int rc4030_post_load(void *opaque, int version_id)
 {
-    rc4030State* s = opaque;
+    rc4030State *s = opaque;
 
     set_next_tick(s);
     update_jazz_irq(s);
@@ -590,7 +593,8 @@ static void rc4030_do_dma(void *opaque, int n, uint8_t *buf, int len, int is_wri
     hwaddr dma_addr;
     int dev_to_mem;
 
-    s->dma_regs[n][DMA_REG_ENABLE] &= ~(DMA_FLAG_TC_INTR | DMA_FLAG_MEM_INTR | DMA_FLAG_ADDR_INTR);
+    s->dma_regs[n][DMA_REG_ENABLE] &=
+           ~(DMA_FLAG_TC_INTR | DMA_FLAG_MEM_INTR | DMA_FLAG_ADDR_INTR);
 
     /* Check DMA channel consistency */
     dev_to_mem = (s->dma_regs[n][DMA_REG_ENABLE] & DMA_FLAG_MEM_TO_DEV) ? 0 : 1;
@@ -602,8 +606,9 @@ static void rc4030_do_dma(void *opaque, int n, uint8_t *buf, int len, int is_wri
     }
 
     /* Get start address and len */
-    if (len > s->dma_regs[n][DMA_REG_COUNT])
+    if (len > s->dma_regs[n][DMA_REG_COUNT]) {
         len = s->dma_regs[n][DMA_REG_COUNT];
+    }
     dma_addr = s->dma_regs[n][DMA_REG_ADDRESS];
 
     /* Read/write data at right place */
@@ -678,7 +683,7 @@ static void rc4030_realize(DeviceState *dev, Error **errp)
 
     memory_region_init_iommu(&s->dma_mr, sizeof(s->dma_mr),
                              TYPE_RC4030_IOMMU_MEMORY_REGION,
-                             o, "rc4030.dma", UINT32_MAX);
+                             o, "rc4030.dma", 4 * GiB);
     address_space_init(&s->dma_as, MEMORY_REGION(&s->dma_mr), "rc4030-dma");
 }
 
diff --git a/hw/i2c/bitbang_i2c.c b/hw/i2c/bitbang_i2c.c
index 5dfc72d9d7..60c7a9be0b 100644
--- a/hw/i2c/bitbang_i2c.c
+++ b/hw/i2c/bitbang_i2c.c
@@ -12,7 +12,7 @@
 
 #include "qemu/osdep.h"
 #include "hw/hw.h"
-#include "bitbang_i2c.h"
+#include "hw/i2c/bitbang_i2c.h"
 #include "hw/sysbus.h"
 #include "qemu/module.h"
 
@@ -25,39 +25,6 @@ do { printf("bitbang_i2c: " fmt , ## __VA_ARGS__); } while (0)
 #define DPRINTF(fmt, ...) do {} while(0)
 #endif
 
-typedef enum bitbang_i2c_state {
-    STOPPED = 0,
-    SENDING_BIT7,
-    SENDING_BIT6,
-    SENDING_BIT5,
-    SENDING_BIT4,
-    SENDING_BIT3,
-    SENDING_BIT2,
-    SENDING_BIT1,
-    SENDING_BIT0,
-    WAITING_FOR_ACK,
-    RECEIVING_BIT7,
-    RECEIVING_BIT6,
-    RECEIVING_BIT5,
-    RECEIVING_BIT4,
-    RECEIVING_BIT3,
-    RECEIVING_BIT2,
-    RECEIVING_BIT1,
-    RECEIVING_BIT0,
-    SENDING_ACK,
-    SENT_NACK
-} bitbang_i2c_state;
-
-struct bitbang_i2c_interface {
-    I2CBus *bus;
-    bitbang_i2c_state state;
-    int last_data;
-    int last_clock;
-    int device_out;
-    uint8_t buffer;
-    int current_addr;
-};
-
 static void bitbang_i2c_enter_stop(bitbang_i2c_interface *i2c)
 {
     DPRINTF("STOP\n");
@@ -184,18 +151,12 @@ int bitbang_i2c_set(bitbang_i2c_interface *i2c, int line, int level)
     abort();
 }
 
-bitbang_i2c_interface *bitbang_i2c_init(I2CBus *bus)
+void bitbang_i2c_init(bitbang_i2c_interface *s, I2CBus *bus)
 {
-    bitbang_i2c_interface *s;
-
-    s = g_malloc0(sizeof(bitbang_i2c_interface));
-
     s->bus = bus;
     s->last_data = 1;
     s->last_clock = 1;
     s->device_out = 1;
-
-    return s;
 }
 
 /* GPIO interface.  */
@@ -207,7 +168,7 @@ typedef struct GPIOI2CState {
     SysBusDevice parent_obj;
 
     MemoryRegion dummy_iomem;
-    bitbang_i2c_interface *bitbang;
+    bitbang_i2c_interface bitbang;
     int last_level;
     qemu_irq out;
 } GPIOI2CState;
@@ -216,7 +177,7 @@ static void bitbang_i2c_gpio_set(void *opaque, int irq, int level)
 {
     GPIOI2CState *s = opaque;
 
-    level = bitbang_i2c_set(s->bitbang, irq, level);
+    level = bitbang_i2c_set(&s->bitbang, irq, level);
     if (level != s->last_level) {
         s->last_level = level;
         qemu_set_irq(s->out, level);
@@ -234,7 +195,7 @@ static void gpio_i2c_init(Object *obj)
     sysbus_init_mmio(sbd, &s->dummy_iomem);
 
     bus = i2c_init_bus(dev, "i2c");
-    s->bitbang = bitbang_i2c_init(bus);
+    bitbang_i2c_init(&s->bitbang, bus);
 
     qdev_init_gpio_in(dev, bitbang_i2c_gpio_set, 2);
     qdev_init_gpio_out(dev, &s->out, 1);
diff --git a/hw/i2c/bitbang_i2c.h b/hw/i2c/bitbang_i2c.h
deleted file mode 100644
index 9443021710..0000000000
--- a/hw/i2c/bitbang_i2c.h
+++ /dev/null
@@ -1,12 +0,0 @@
-#ifndef BITBANG_I2C_H
-#define BITBANG_I2C_H
-
-#include "hw/i2c/i2c.h"
-
-#define BITBANG_I2C_SDA 0
-#define BITBANG_I2C_SCL 1
-
-bitbang_i2c_interface *bitbang_i2c_init(I2CBus *bus);
-int bitbang_i2c_set(bitbang_i2c_interface *i2c, int line, int level);
-
-#endif
diff --git a/hw/i2c/ppc4xx_i2c.c b/hw/i2c/ppc4xx_i2c.c
index d606d3dbeb..462729db4e 100644
--- a/hw/i2c/ppc4xx_i2c.c
+++ b/hw/i2c/ppc4xx_i2c.c
@@ -30,7 +30,6 @@
 #include "cpu.h"
 #include "hw/hw.h"
 #include "hw/i2c/ppc4xx_i2c.h"
-#include "bitbang_i2c.h"
 
 #define PPC4xx_I2C_MEM_SIZE 18
 
@@ -312,9 +311,9 @@ static void ppc4xx_i2c_writeb(void *opaque, hwaddr addr, uint64_t value,
     case IIC_DIRECTCNTL:
         i2c->directcntl = value & (IIC_DIRECTCNTL_SDAC & IIC_DIRECTCNTL_SCLC);
         i2c->directcntl |= (value & IIC_DIRECTCNTL_SCLC ? 1 : 0);
-        bitbang_i2c_set(i2c->bitbang, BITBANG_I2C_SCL,
+        bitbang_i2c_set(&i2c->bitbang, BITBANG_I2C_SCL,
                         i2c->directcntl & IIC_DIRECTCNTL_MSCL);
-        i2c->directcntl |= bitbang_i2c_set(i2c->bitbang, BITBANG_I2C_SDA,
+        i2c->directcntl |= bitbang_i2c_set(&i2c->bitbang, BITBANG_I2C_SDA,
                                (value & IIC_DIRECTCNTL_SDAC) != 0) << 1;
         break;
     default:
@@ -348,7 +347,7 @@ static void ppc4xx_i2c_init(Object *o)
     sysbus_init_mmio(SYS_BUS_DEVICE(s), &s->iomem);
     sysbus_init_irq(SYS_BUS_DEVICE(s), &s->irq);
     s->bus = i2c_init_bus(DEVICE(s), "i2c");
-    s->bitbang = bitbang_i2c_init(s->bus);
+    bitbang_i2c_init(&s->bitbang, s->bus);
 }
 
 static void ppc4xx_i2c_class_init(ObjectClass *klass, void *data)
diff --git a/hw/i2c/versatile_i2c.c b/hw/i2c/versatile_i2c.c
index e07be9890c..1ac2a6f59a 100644
--- a/hw/i2c/versatile_i2c.c
+++ b/hw/i2c/versatile_i2c.c
@@ -23,7 +23,7 @@
 
 #include "qemu/osdep.h"
 #include "hw/sysbus.h"
-#include "bitbang_i2c.h"
+#include "hw/i2c/bitbang_i2c.h"
 #include "qemu/log.h"
 #include "qemu/module.h"
 
@@ -35,7 +35,7 @@ typedef struct VersatileI2CState {
     SysBusDevice parent_obj;
 
     MemoryRegion iomem;
-    bitbang_i2c_interface *bitbang;
+    bitbang_i2c_interface bitbang;
     int out;
     int in;
 } VersatileI2CState;
@@ -70,8 +70,8 @@ static void versatile_i2c_write(void *opaque, hwaddr offset,
         qemu_log_mask(LOG_GUEST_ERROR,
                       "%s: Bad offset 0x%x\n", __func__, (int)offset);
     }
-    bitbang_i2c_set(s->bitbang, BITBANG_I2C_SCL, (s->out & 1) != 0);
-    s->in = bitbang_i2c_set(s->bitbang, BITBANG_I2C_SDA, (s->out & 2) != 0);
+    bitbang_i2c_set(&s->bitbang, BITBANG_I2C_SCL, (s->out & 1) != 0);
+    s->in = bitbang_i2c_set(&s->bitbang, BITBANG_I2C_SDA, (s->out & 2) != 0);
 }
 
 static const MemoryRegionOps versatile_i2c_ops = {
@@ -88,7 +88,7 @@ static void versatile_i2c_init(Object *obj)
     I2CBus *bus;
 
     bus = i2c_init_bus(dev, "i2c");
-    s->bitbang = bitbang_i2c_init(bus);
+    bitbang_i2c_init(&s->bitbang, bus);
     memory_region_init_io(&s->iomem, obj, &versatile_i2c_ops, s,
                           "versatile_i2c", 0x1000);
     sysbus_init_mmio(sbd, &s->iomem);
diff --git a/hw/i386/xen/xen-hvm.c b/hw/i386/xen/xen-hvm.c
index 2939122e7c..469f1260a4 100644
--- a/hw/i386/xen/xen-hvm.c
+++ b/hw/i386/xen/xen-hvm.c
@@ -27,7 +27,6 @@
 #include "exec/address-spaces.h"
 
 #include <xen/hvm/ioreq.h>
-#include <xen/hvm/params.h>
 #include <xen/hvm/e820.h>
 
 //#define DEBUG_XEN_HVM
@@ -120,6 +119,8 @@ typedef struct XenIOState {
     DeviceListener device_listener;
     hwaddr free_phys_offset;
     const XenPhysmap *log_for_dirtybit;
+    /* Buffer used by xen_sync_dirty_bitmap */
+    unsigned long *dirty_bitmap;
 
     Notifier exit;
     Notifier suspend;
@@ -465,6 +466,8 @@ static int xen_remove_from_physmap(XenIOState *state,
     QLIST_REMOVE(physmap, list);
     if (state->log_for_dirtybit == physmap) {
         state->log_for_dirtybit = NULL;
+        g_free(state->dirty_bitmap);
+        state->dirty_bitmap = NULL;
     }
     g_free(physmap);
 
@@ -615,7 +618,7 @@ static void xen_sync_dirty_bitmap(XenIOState *state,
 {
     hwaddr npages = size >> TARGET_PAGE_BITS;
     const int width = sizeof(unsigned long) * 8;
-    unsigned long bitmap[DIV_ROUND_UP(npages, width)];
+    size_t bitmap_size = DIV_ROUND_UP(npages, width);
     int rc, i, j;
     const XenPhysmap *physmap = NULL;
 
@@ -627,13 +630,14 @@ static void xen_sync_dirty_bitmap(XenIOState *state,
 
     if (state->log_for_dirtybit == NULL) {
         state->log_for_dirtybit = physmap;
+        state->dirty_bitmap = g_new(unsigned long, bitmap_size);
     } else if (state->log_for_dirtybit != physmap) {
         /* Only one range for dirty bitmap can be tracked. */
         return;
     }
 
     rc = xen_track_dirty_vram(xen_domid, start_addr >> TARGET_PAGE_BITS,
-                              npages, bitmap);
+                              npages, state->dirty_bitmap);
     if (rc < 0) {
 #ifndef ENODATA
 #define ENODATA  ENOENT
@@ -647,8 +651,8 @@ static void xen_sync_dirty_bitmap(XenIOState *state,
         return;
     }
 
-    for (i = 0; i < ARRAY_SIZE(bitmap); i++) {
-        unsigned long map = bitmap[i];
+    for (i = 0; i < bitmap_size; i++) {
+        unsigned long map = state->dirty_bitmap[i];
         while (map != 0) {
             j = ctzl(map);
             map &= ~(1ul << j);
@@ -678,6 +682,8 @@ static void xen_log_stop(MemoryListener *listener, MemoryRegionSection *section,
 
     if (old & ~new & (1 << DIRTY_MEMORY_VGA)) {
         state->log_for_dirtybit = NULL;
+        g_free(state->dirty_bitmap);
+        state->dirty_bitmap = NULL;
         /* Disable dirty bit tracking */
         xen_track_dirty_vram(xen_domid, 0, 0, NULL);
     }
diff --git a/hw/i386/xen/xen-mapcache.c b/hw/i386/xen/xen-mapcache.c
index 254759f776..dc73c86c61 100644
--- a/hw/i386/xen/xen-mapcache.c
+++ b/hw/i386/xen/xen-mapcache.c
@@ -17,8 +17,6 @@
 #include "hw/xen/xen-legacy-backend.h"
 #include "qemu/bitmap.h"
 
-#include <xen/hvm/params.h>
-
 #include "sysemu/xen-mapcache.h"
 #include "trace.h"
 
diff --git a/hw/intc/armv7m_nvic.c b/hw/intc/armv7m_nvic.c
index b8ede30b3c..9f8f0d3ff5 100644
--- a/hw/intc/armv7m_nvic.c
+++ b/hw/intc/armv7m_nvic.c
@@ -812,15 +812,45 @@ void armv7m_nvic_get_pending_irq_info(void *opaque,
 int armv7m_nvic_complete_irq(void *opaque, int irq, bool secure)
 {
     NVICState *s = (NVICState *)opaque;
-    VecInfo *vec;
+    VecInfo *vec = NULL;
     int ret;
 
     assert(irq > ARMV7M_EXCP_RESET && irq < s->num_irq);
 
-    if (secure && exc_is_banked(irq)) {
-        vec = &s->sec_vectors[irq];
-    } else {
-        vec = &s->vectors[irq];
+    /*
+     * For negative priorities, v8M will forcibly deactivate the appropriate
+     * NMI or HardFault regardless of what interrupt we're being asked to
+     * deactivate (compare the DeActivate() pseudocode). This is a guard
+     * against software returning from NMI or HardFault with a corrupted
+     * IPSR and leaving the CPU in a negative-priority state.
+     * v7M does not do this, but simply deactivates the requested interrupt.
+     */
+    if (arm_feature(&s->cpu->env, ARM_FEATURE_V8)) {
+        switch (armv7m_nvic_raw_execution_priority(s)) {
+        case -1:
+            if (s->cpu->env.v7m.aircr & R_V7M_AIRCR_BFHFNMINS_MASK) {
+                vec = &s->vectors[ARMV7M_EXCP_HARD];
+            } else {
+                vec = &s->sec_vectors[ARMV7M_EXCP_HARD];
+            }
+            break;
+        case -2:
+            vec = &s->vectors[ARMV7M_EXCP_NMI];
+            break;
+        case -3:
+            vec = &s->sec_vectors[ARMV7M_EXCP_HARD];
+            break;
+        default:
+            break;
+        }
+    }
+
+    if (!vec) {
+        if (secure && exc_is_banked(irq)) {
+            vec = &s->sec_vectors[irq];
+        } else {
+            vec = &s->vectors[irq];
+        }
     }
 
     trace_nvic_complete_irq(irq, secure);
@@ -830,7 +860,19 @@ int armv7m_nvic_complete_irq(void *opaque, int irq, bool secure)
         return -1;
     }
 
-    ret = nvic_rettobase(s);
+    /*
+     * If this is a configurable exception and it is currently
+     * targeting the opposite security state from the one we're trying
+     * to complete it for, this counts as an illegal exception return.
+     * We still need to deactivate whatever vector the logic above has
+     * selected, though, as it might not be the same as the one for the
+     * requested exception number.
+     */
+    if (!exc_is_banked(irq) && exc_targets_secure(s, irq) != secure) {
+        ret = -1;
+    } else {
+        ret = nvic_rettobase(s);
+    }
 
     vec->active = 0;
     if (vec->level) {
diff --git a/hw/intc/aspeed_vic.c b/hw/intc/aspeed_vic.c
index 927638d532..266a309f3b 100644
--- a/hw/intc/aspeed_vic.c
+++ b/hw/intc/aspeed_vic.c
@@ -104,54 +104,63 @@ static void aspeed_vic_set_irq(void *opaque, int irq, int level)
 
 static uint64_t aspeed_vic_read(void *opaque, hwaddr offset, unsigned size)
 {
-    uint64_t val;
-    const bool high = !!(offset & 0x4);
-    hwaddr n_offset = (offset & ~0x4);
     AspeedVICState *s = (AspeedVICState *)opaque;
+    hwaddr n_offset;
+    uint64_t val;
+    bool high;
 
     if (offset < AVIC_NEW_BASE_OFFSET) {
-        qemu_log_mask(LOG_UNIMP, "%s: Ignoring read from legacy registers "
-                      "at 0x%" HWADDR_PRIx "[%u]\n", __func__, offset, size);
-        return 0;
+        high = false;
+        n_offset = offset;
+    } else {
+        high = !!(offset & 0x4);
+        n_offset = (offset & ~0x4);
     }
 
-    n_offset -= AVIC_NEW_BASE_OFFSET;
-
     switch (n_offset) {
-    case 0x0: /* IRQ Status */
+    case 0x80: /* IRQ Status */
+    case 0x00:
         val = s->raw & ~s->select & s->enable;
         break;
-    case 0x08: /* FIQ Status */
+    case 0x88: /* FIQ Status */
+    case 0x04:
         val = s->raw & s->select & s->enable;
         break;
-    case 0x10: /* Raw Interrupt Status */
+    case 0x90: /* Raw Interrupt Status */
+    case 0x08:
         val = s->raw;
         break;
-    case 0x18: /* Interrupt Selection */
+    case 0x98: /* Interrupt Selection */
+    case 0x0c:
         val = s->select;
         break;
-    case 0x20: /* Interrupt Enable */
+    case 0xa0: /* Interrupt Enable */
+    case 0x10:
         val = s->enable;
         break;
-    case 0x30: /* Software Interrupt */
+    case 0xb0: /* Software Interrupt */
+    case 0x18:
         val = s->trigger;
         break;
-    case 0x40: /* Interrupt Sensitivity */
+    case 0xc0: /* Interrupt Sensitivity */
+    case 0x24:
         val = s->sense;
         break;
-    case 0x48: /* Interrupt Both Edge Trigger Control */
+    case 0xc8: /* Interrupt Both Edge Trigger Control */
+    case 0x28:
         val = s->dual_edge;
         break;
-    case 0x50: /* Interrupt Event */
+    case 0xd0: /* Interrupt Event */
+    case 0x2c:
         val = s->event;
         break;
-    case 0x60: /* Edge Triggered Interrupt Status */
+    case 0xe0: /* Edge Triggered Interrupt Status */
         val = s->raw & ~s->sense;
         break;
         /* Illegal */
-    case 0x28: /* Interrupt Enable Clear */
-    case 0x38: /* Software Interrupt Clear */
-    case 0x58: /* Edge Triggered Interrupt Clear */
+    case 0xa8: /* Interrupt Enable Clear */
+    case 0xb8: /* Software Interrupt Clear */
+    case 0xd8: /* Edge Triggered Interrupt Clear */
         qemu_log_mask(LOG_GUEST_ERROR,
                       "%s: Read of write-only register with offset 0x%"
                       HWADDR_PRIx "\n", __func__, offset);
@@ -166,6 +175,8 @@ static uint64_t aspeed_vic_read(void *opaque, hwaddr offset, unsigned size)
     }
     if (high) {
         val = extract64(val, 32, 19);
+    } else {
+        val = extract64(val, 0, 32);
     }
     trace_aspeed_vic_read(offset, size, val);
     return val;
@@ -174,19 +185,18 @@ static uint64_t aspeed_vic_read(void *opaque, hwaddr offset, unsigned size)
 static void aspeed_vic_write(void *opaque, hwaddr offset, uint64_t data,
                              unsigned size)
 {
-    const bool high = !!(offset & 0x4);
-    hwaddr n_offset = (offset & ~0x4);
     AspeedVICState *s = (AspeedVICState *)opaque;
+    hwaddr n_offset;
+    bool high;
 
     if (offset < AVIC_NEW_BASE_OFFSET) {
-        qemu_log_mask(LOG_UNIMP,
-                      "%s: Ignoring write to legacy registers at 0x%"
-                      HWADDR_PRIx "[%u] <- 0x%" PRIx64 "\n", __func__, offset,
-                      size, data);
-        return;
+        high = false;
+        n_offset = offset;
+    } else {
+        high = !!(offset & 0x4);
+        n_offset = (offset & ~0x4);
     }
 
-    n_offset -= AVIC_NEW_BASE_OFFSET;
     trace_aspeed_vic_write(offset, size, data);
 
     /* Given we have members using separate enable/clear registers, deposit64()
@@ -201,7 +211,8 @@ static void aspeed_vic_write(void *opaque, hwaddr offset, uint64_t data,
     }
 
     switch (n_offset) {
-    case 0x18: /* Interrupt Selection */
+    case 0x98: /* Interrupt Selection */
+    case 0x0c:
         /* Register has deposit64() semantics - overwrite requested 32 bits */
         if (high) {
             s->select &= AVIC_L_MASK;
@@ -210,21 +221,25 @@ static void aspeed_vic_write(void *opaque, hwaddr offset, uint64_t data,
         }
         s->select |= data;
         break;
-    case 0x20: /* Interrupt Enable */
+    case 0xa0: /* Interrupt Enable */
+    case 0x10:
         s->enable |= data;
         break;
-    case 0x28: /* Interrupt Enable Clear */
+    case 0xa8: /* Interrupt Enable Clear */
+    case 0x14:
         s->enable &= ~data;
         break;
-    case 0x30: /* Software Interrupt */
+    case 0xb0: /* Software Interrupt */
+    case 0x18:
         qemu_log_mask(LOG_UNIMP, "%s: Software interrupts unavailable. "
                       "IRQs requested: 0x%016" PRIx64 "\n", __func__, data);
         break;
-    case 0x38: /* Software Interrupt Clear */
+    case 0xb8: /* Software Interrupt Clear */
+    case 0x1c:
         qemu_log_mask(LOG_UNIMP, "%s: Software interrupts unavailable. "
                       "IRQs to be cleared: 0x%016" PRIx64 "\n", __func__, data);
         break;
-    case 0x50: /* Interrupt Event */
+    case 0xd0: /* Interrupt Event */
         /* Register has deposit64() semantics - overwrite the top four valid
          * IRQ bits, as only the top four IRQs (GPIOs) can change their event
          * type */
@@ -236,15 +251,21 @@ static void aspeed_vic_write(void *opaque, hwaddr offset, uint64_t data,
                           "Ignoring invalid write to interrupt event register");
         }
         break;
-    case 0x58: /* Edge Triggered Interrupt Clear */
+    case 0xd8: /* Edge Triggered Interrupt Clear */
+    case 0x38:
         s->raw &= ~(data & ~s->sense);
         break;
-    case 0x00: /* IRQ Status */
-    case 0x08: /* FIQ Status */
-    case 0x10: /* Raw Interrupt Status */
-    case 0x40: /* Interrupt Sensitivity */
-    case 0x48: /* Interrupt Both Edge Trigger Control */
-    case 0x60: /* Edge Triggered Interrupt Status */
+    case 0x80: /* IRQ Status */
+    case 0x00:
+    case 0x88: /* FIQ Status */
+    case 0x04:
+    case 0x90: /* Raw Interrupt Status */
+    case 0x08:
+    case 0xc0: /* Interrupt Sensitivity */
+    case 0x24:
+    case 0xc8: /* Interrupt Both Edge Trigger Control */
+    case 0x28:
+    case 0xe0: /* Edge Triggered Interrupt Status */
         qemu_log_mask(LOG_GUEST_ERROR,
                       "%s: Write of read-only register with offset 0x%"
                       HWADDR_PRIx "\n", __func__, offset);
diff --git a/hw/intc/pnv_xive.c b/hw/intc/pnv_xive.c
index a55c2bbc88..4dc92ef1e3 100644
--- a/hw/intc/pnv_xive.c
+++ b/hw/intc/pnv_xive.c
@@ -169,7 +169,7 @@ static uint64_t pnv_xive_vst_addr_indirect(PnvXive *xive, uint32_t type,
     vsd = ldq_be_dma(&address_space_memory, vsd_addr);
 
     if (!(vsd & VSD_ADDRESS_MASK)) {
-        xive_error(xive, "VST: invalid %s entry %x !?", info->name, 0);
+        xive_error(xive, "VST: invalid %s entry %x !?", info->name, idx);
         return 0;
     }
 
@@ -190,7 +190,7 @@ static uint64_t pnv_xive_vst_addr_indirect(PnvXive *xive, uint32_t type,
         vsd = ldq_be_dma(&address_space_memory, vsd_addr);
 
         if (!(vsd & VSD_ADDRESS_MASK)) {
-            xive_error(xive, "VST: invalid %s entry %x !?", info->name, 0);
+            xive_error(xive, "VST: invalid %s entry %x !?", info->name, idx);
             return 0;
         }
 
@@ -294,8 +294,12 @@ static int pnv_xive_write_end(XiveRouter *xrtr, uint8_t blk, uint32_t idx,
                               word_number);
 }
 
-static int pnv_xive_end_update(PnvXive *xive, uint8_t blk, uint32_t idx)
+static int pnv_xive_end_update(PnvXive *xive)
 {
+    uint8_t  blk = GETFIELD(VC_EQC_CWATCH_BLOCKID,
+                           xive->regs[(VC_EQC_CWATCH_SPEC >> 3)]);
+    uint32_t idx = GETFIELD(VC_EQC_CWATCH_OFFSET,
+                           xive->regs[(VC_EQC_CWATCH_SPEC >> 3)]);
     int i;
     uint64_t eqc_watch[4];
 
@@ -307,6 +311,24 @@ static int pnv_xive_end_update(PnvXive *xive, uint8_t blk, uint32_t idx)
                               XIVE_VST_WORD_ALL);
 }
 
+static void pnv_xive_end_cache_load(PnvXive *xive)
+{
+    uint8_t  blk = GETFIELD(VC_EQC_CWATCH_BLOCKID,
+                           xive->regs[(VC_EQC_CWATCH_SPEC >> 3)]);
+    uint32_t idx = GETFIELD(VC_EQC_CWATCH_OFFSET,
+                           xive->regs[(VC_EQC_CWATCH_SPEC >> 3)]);
+    uint64_t eqc_watch[4] = { 0 };
+    int i;
+
+    if (pnv_xive_vst_read(xive, VST_TSEL_EQDT, blk, idx, eqc_watch)) {
+        xive_error(xive, "VST: no END entry %x/%x !?", blk, idx);
+    }
+
+    for (i = 0; i < ARRAY_SIZE(eqc_watch); i++) {
+        xive->regs[(VC_EQC_CWATCH_DAT0 >> 3) + i] = be64_to_cpu(eqc_watch[i]);
+    }
+}
+
 static int pnv_xive_get_nvt(XiveRouter *xrtr, uint8_t blk, uint32_t idx,
                             XiveNVT *nvt)
 {
@@ -320,8 +342,12 @@ static int pnv_xive_write_nvt(XiveRouter *xrtr, uint8_t blk, uint32_t idx,
                               word_number);
 }
 
-static int pnv_xive_nvt_update(PnvXive *xive, uint8_t blk, uint32_t idx)
+static int pnv_xive_nvt_update(PnvXive *xive)
 {
+    uint8_t  blk = GETFIELD(PC_VPC_CWATCH_BLOCKID,
+                           xive->regs[(PC_VPC_CWATCH_SPEC >> 3)]);
+    uint32_t idx = GETFIELD(PC_VPC_CWATCH_OFFSET,
+                           xive->regs[(PC_VPC_CWATCH_SPEC >> 3)]);
     int i;
     uint64_t vpc_watch[8];
 
@@ -333,6 +359,24 @@ static int pnv_xive_nvt_update(PnvXive *xive, uint8_t blk, uint32_t idx)
                               XIVE_VST_WORD_ALL);
 }
 
+static void pnv_xive_nvt_cache_load(PnvXive *xive)
+{
+    uint8_t  blk = GETFIELD(PC_VPC_CWATCH_BLOCKID,
+                           xive->regs[(PC_VPC_CWATCH_SPEC >> 3)]);
+    uint32_t idx = GETFIELD(PC_VPC_CWATCH_OFFSET,
+                           xive->regs[(PC_VPC_CWATCH_SPEC >> 3)]);
+    uint64_t vpc_watch[8] = { 0 };
+    int i;
+
+    if (pnv_xive_vst_read(xive, VST_TSEL_VPDT, blk, idx, vpc_watch)) {
+        xive_error(xive, "VST: no NVT entry %x/%x !?", blk, idx);
+    }
+
+    for (i = 0; i < ARRAY_SIZE(vpc_watch); i++) {
+        xive->regs[(PC_VPC_CWATCH_DAT0 >> 3) + i] = be64_to_cpu(vpc_watch[i]);
+    }
+}
+
 static int pnv_xive_get_eas(XiveRouter *xrtr, uint8_t blk, uint32_t idx,
                             XiveEAS *eas)
 {
@@ -346,12 +390,6 @@ static int pnv_xive_get_eas(XiveRouter *xrtr, uint8_t blk, uint32_t idx,
     return pnv_xive_vst_read(xive, VST_TSEL_IVT, blk, idx, eas);
 }
 
-static int pnv_xive_eas_update(PnvXive *xive, uint8_t blk, uint32_t idx)
-{
-    /* All done. */
-    return 0;
-}
-
 static XiveTCTX *pnv_xive_get_tctx(XiveRouter *xrtr, CPUState *cs)
 {
     PowerPCCPU *cpu = POWERPC_CPU(cs);
@@ -781,8 +819,7 @@ static void pnv_xive_ic_reg_write(void *opaque, hwaddr offset,
          * support recently though)
          */
         if (val & (VC_SBC_CONF_CPLX_CIST | VC_SBC_CONF_CIST_BOTH)) {
-            object_property_set_int(OBJECT(&xive->ipi_source),
-                                    XIVE_SRC_STORE_EOI, "flags", &error_fatal);
+            xive->ipi_source.esb_flags |= XIVE_SRC_STORE_EOI;
         }
         break;
 
@@ -951,28 +988,43 @@ static void pnv_xive_ic_reg_write(void *opaque, hwaddr offset,
      * XIVE PC & VC cache updates for EAS, NVT and END
      */
     case VC_IVC_SCRUB_MASK:
-        break;
     case VC_IVC_SCRUB_TRIG:
-        pnv_xive_eas_update(xive, GETFIELD(PC_SCRUB_BLOCK_ID, val),
-                            GETFIELD(VC_SCRUB_OFFSET, val));
         break;
 
-    case VC_EQC_SCRUB_MASK:
     case VC_EQC_CWATCH_SPEC:
-    case VC_EQC_CWATCH_DAT0 ... VC_EQC_CWATCH_DAT3:
+        val &= ~VC_EQC_CWATCH_CONFLICT; /* HW resets this bit */
+        break;
+    case VC_EQC_CWATCH_DAT1 ... VC_EQC_CWATCH_DAT3:
         break;
+    case VC_EQC_CWATCH_DAT0:
+        /* writing to DATA0 triggers the cache write */
+        xive->regs[reg] = val;
+        pnv_xive_end_update(xive);
+        break;
+    case VC_EQC_SCRUB_MASK:
     case VC_EQC_SCRUB_TRIG:
-        pnv_xive_end_update(xive, GETFIELD(VC_SCRUB_BLOCK_ID, val),
-                            GETFIELD(VC_SCRUB_OFFSET, val));
+        /*
+         * The scrubbing registers flush the cache in RAM and can also
+         * invalidate.
+         */
         break;
 
-    case PC_VPC_SCRUB_MASK:
     case PC_VPC_CWATCH_SPEC:
-    case PC_VPC_CWATCH_DAT0 ... PC_VPC_CWATCH_DAT7:
+        val &= ~PC_VPC_CWATCH_CONFLICT; /* HW resets this bit */
+        break;
+    case PC_VPC_CWATCH_DAT1 ... PC_VPC_CWATCH_DAT7:
         break;
+    case PC_VPC_CWATCH_DAT0:
+        /* writing to DATA0 triggers the cache write */
+        xive->regs[reg] = val;
+        pnv_xive_nvt_update(xive);
+        break;
+    case PC_VPC_SCRUB_MASK:
     case PC_VPC_SCRUB_TRIG:
-        pnv_xive_nvt_update(xive, GETFIELD(PC_SCRUB_BLOCK_ID, val),
-                           GETFIELD(PC_SCRUB_OFFSET, val));
+        /*
+         * The scrubbing registers flush the cache in RAM and can also
+         * invalidate.
+         */
         break;
 
 
@@ -1023,15 +1075,6 @@ static uint64_t pnv_xive_ic_reg_read(void *opaque, hwaddr offset, unsigned size)
     case PC_GLOBAL_CONFIG:
 
     case PC_VPC_SCRUB_MASK:
-    case PC_VPC_CWATCH_SPEC:
-    case PC_VPC_CWATCH_DAT0:
-    case PC_VPC_CWATCH_DAT1:
-    case PC_VPC_CWATCH_DAT2:
-    case PC_VPC_CWATCH_DAT3:
-    case PC_VPC_CWATCH_DAT4:
-    case PC_VPC_CWATCH_DAT5:
-    case PC_VPC_CWATCH_DAT6:
-    case PC_VPC_CWATCH_DAT7:
 
     case VC_GLOBAL_CONFIG:
     case VC_AIB_TX_ORDER_TAG2:
@@ -1044,12 +1087,6 @@ static uint64_t pnv_xive_ic_reg_read(void *opaque, hwaddr offset, unsigned size)
     case VC_IRQ_CONFIG_IPI_CASC:
 
     case VC_EQC_SCRUB_MASK:
-    case VC_EQC_CWATCH_DAT0:
-    case VC_EQC_CWATCH_DAT1:
-    case VC_EQC_CWATCH_DAT2:
-    case VC_EQC_CWATCH_DAT3:
-
-    case VC_EQC_CWATCH_SPEC:
     case VC_IVC_SCRUB_MASK:
     case VC_SBC_CONFIG:
     case VC_AT_MACRO_KILL_MASK:
@@ -1081,6 +1118,38 @@ static uint64_t pnv_xive_ic_reg_read(void *opaque, hwaddr offset, unsigned size)
     /*
      * XIVE PC & VC cache updates for EAS, NVT and END
      */
+    case VC_EQC_CWATCH_SPEC:
+        xive->regs[reg] = ~(VC_EQC_CWATCH_FULL | VC_EQC_CWATCH_CONFLICT);
+        val = xive->regs[reg];
+        break;
+    case VC_EQC_CWATCH_DAT0:
+        /*
+         * Load DATA registers from cache with data requested by the
+         * SPEC register
+         */
+        pnv_xive_end_cache_load(xive);
+        val = xive->regs[reg];
+        break;
+    case VC_EQC_CWATCH_DAT1 ... VC_EQC_CWATCH_DAT3:
+        val = xive->regs[reg];
+        break;
+
+    case PC_VPC_CWATCH_SPEC:
+        xive->regs[reg] = ~(PC_VPC_CWATCH_FULL | PC_VPC_CWATCH_CONFLICT);
+        val = xive->regs[reg];
+        break;
+    case PC_VPC_CWATCH_DAT0:
+        /*
+         * Load DATA registers from cache with data requested by the
+         * SPEC register
+         */
+        pnv_xive_nvt_cache_load(xive);
+        val = xive->regs[reg];
+        break;
+    case PC_VPC_CWATCH_DAT1 ... PC_VPC_CWATCH_DAT7:
+        val = xive->regs[reg];
+        break;
+
     case PC_VPC_SCRUB_TRIG:
     case VC_IVC_SCRUB_TRIG:
     case VC_EQC_SCRUB_TRIG:
diff --git a/hw/intc/spapr_xive.c b/hw/intc/spapr_xive.c
index 58c2e5d890..3ae311d9ff 100644
--- a/hw/intc/spapr_xive.c
+++ b/hw/intc/spapr_xive.c
@@ -194,13 +194,6 @@ void spapr_xive_pic_print_info(SpaprXive *xive, Monitor *mon)
     }
 }
 
-void spapr_xive_map_mmio(SpaprXive *xive)
-{
-    sysbus_mmio_map(SYS_BUS_DEVICE(xive), 0, xive->vc_base);
-    sysbus_mmio_map(SYS_BUS_DEVICE(xive), 1, xive->end_base);
-    sysbus_mmio_map(SYS_BUS_DEVICE(xive), 2, xive->tm_base);
-}
-
 void spapr_xive_mmio_set_enabled(SpaprXive *xive, bool enable)
 {
     memory_region_set_enabled(&xive->source.esb_mmio, enable);
@@ -305,6 +298,7 @@ static void spapr_xive_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, local_err);
         return;
     }
+    sysbus_init_mmio(SYS_BUS_DEVICE(xive), &xsrc->esb_mmio);
 
     /*
      * Initialize the END ESB source
@@ -318,6 +312,7 @@ static void spapr_xive_realize(DeviceState *dev, Error **errp)
         error_propagate(errp, local_err);
         return;
     }
+    sysbus_init_mmio(SYS_BUS_DEVICE(xive), &end_xsrc->esb_mmio);
 
     /* Set the mapping address of the END ESB pages after the source ESBs */
     xive->end_base = xive->vc_base + (1ull << xsrc->esb_shift) * xsrc->nr_irqs;
@@ -333,31 +328,18 @@ static void spapr_xive_realize(DeviceState *dev, Error **errp)
 
     qemu_register_reset(spapr_xive_reset, dev);
 
-    /* Define all XIVE MMIO regions on SysBus */
-    sysbus_init_mmio(SYS_BUS_DEVICE(xive), &xsrc->esb_mmio);
-    sysbus_init_mmio(SYS_BUS_DEVICE(xive), &end_xsrc->esb_mmio);
-    sysbus_init_mmio(SYS_BUS_DEVICE(xive), &xive->tm_mmio);
-}
-
-void spapr_xive_init(SpaprXive *xive, Error **errp)
-{
-    XiveSource *xsrc = &xive->source;
-
-    /*
-     * The emulated XIVE device can only be initialized once. If the
-     * ESB memory region has been already mapped, it means we have been
-     * through there.
-     */
-    if (memory_region_is_mapped(&xsrc->esb_mmio)) {
-        return;
-    }
-
     /* TIMA initialization */
     memory_region_init_io(&xive->tm_mmio, OBJECT(xive), &xive_tm_ops, xive,
                           "xive.tima", 4ull << TM_SHIFT);
+    sysbus_init_mmio(SYS_BUS_DEVICE(xive), &xive->tm_mmio);
 
-    /* Map all regions */
-    spapr_xive_map_mmio(xive);
+    /*
+     * Map all regions. These will be enabled or disabled at reset and
+     * can also be overridden by KVM memory regions if active
+     */
+    sysbus_mmio_map(SYS_BUS_DEVICE(xive), 0, xive->vc_base);
+    sysbus_mmio_map(SYS_BUS_DEVICE(xive), 1, xive->end_base);
+    sysbus_mmio_map(SYS_BUS_DEVICE(xive), 2, xive->tm_base);
 }
 
 static int spapr_xive_get_eas(XiveRouter *xrtr, uint8_t eas_blk,
diff --git a/hw/intc/spapr_xive_kvm.c b/hw/intc/spapr_xive_kvm.c
index b48f135838..3bf8e7a20e 100644
--- a/hw/intc/spapr_xive_kvm.c
+++ b/hw/intc/spapr_xive_kvm.c
@@ -724,12 +724,13 @@ void kvmppc_xive_connect(SpaprXive *xive, Error **errp)
     xsrc->esb_mmap = kvmppc_xive_mmap(xive, KVM_XIVE_ESB_PAGE_OFFSET, esb_len,
                                       &local_err);
     if (local_err) {
-        error_propagate(errp, local_err);
-        return;
+        goto fail;
     }
 
-    memory_region_init_ram_device_ptr(&xsrc->esb_mmio, OBJECT(xsrc),
+    memory_region_init_ram_device_ptr(&xsrc->esb_mmio_kvm, OBJECT(xsrc),
                                       "xive.esb", esb_len, xsrc->esb_mmap);
+    memory_region_add_subregion_overlap(&xsrc->esb_mmio, 0,
+                                        &xsrc->esb_mmio_kvm, 1);
 
     /*
      * 2. END ESB pages (No KVM support yet)
@@ -741,11 +742,12 @@ void kvmppc_xive_connect(SpaprXive *xive, Error **errp)
     xive->tm_mmap = kvmppc_xive_mmap(xive, KVM_XIVE_TIMA_PAGE_OFFSET, tima_len,
                                      &local_err);
     if (local_err) {
-        error_propagate(errp, local_err);
-        return;
+        goto fail;
     }
-    memory_region_init_ram_device_ptr(&xive->tm_mmio, OBJECT(xive),
+    memory_region_init_ram_device_ptr(&xive->tm_mmio_kvm, OBJECT(xive),
                                       "xive.tima", tima_len, xive->tm_mmap);
+    memory_region_add_subregion_overlap(&xive->tm_mmio, 0,
+                                        &xive->tm_mmio_kvm, 1);
 
     xive->change = qemu_add_vm_change_state_handler(
         kvmppc_xive_change_state_handler, xive);
@@ -756,24 +758,24 @@ void kvmppc_xive_connect(SpaprXive *xive, Error **errp)
 
         kvmppc_xive_cpu_connect(spapr_cpu_state(cpu)->tctx, &local_err);
         if (local_err) {
-            error_propagate(errp, local_err);
-            return;
+            goto fail;
         }
     }
 
     /* Update the KVM sources */
     kvmppc_xive_source_reset(xsrc, &local_err);
     if (local_err) {
-            error_propagate(errp, local_err);
-            return;
+        goto fail;
     }
 
     kvm_kernel_irqchip = true;
     kvm_msi_via_irqfd_allowed = true;
     kvm_gsi_direct_mapping = true;
+    return;
 
-    /* Map all regions */
-    spapr_xive_map_mmio(xive);
+fail:
+    error_propagate(errp, local_err);
+    kvmppc_xive_disconnect(xive, NULL);
 }
 
 void kvmppc_xive_disconnect(SpaprXive *xive, Error **errp)
@@ -795,21 +797,29 @@ void kvmppc_xive_disconnect(SpaprXive *xive, Error **errp)
     xsrc = &xive->source;
     esb_len = (1ull << xsrc->esb_shift) * xsrc->nr_irqs;
 
-    sysbus_mmio_unmap(SYS_BUS_DEVICE(xive), 0);
-    munmap(xsrc->esb_mmap, esb_len);
-
-    sysbus_mmio_unmap(SYS_BUS_DEVICE(xive), 1);
+    if (xsrc->esb_mmap) {
+        memory_region_del_subregion(&xsrc->esb_mmio, &xsrc->esb_mmio_kvm);
+        object_unparent(OBJECT(&xsrc->esb_mmio_kvm));
+        munmap(xsrc->esb_mmap, esb_len);
+        xsrc->esb_mmap = NULL;
+    }
 
-    sysbus_mmio_unmap(SYS_BUS_DEVICE(xive), 2);
-    munmap(xive->tm_mmap, 4ull << TM_SHIFT);
+    if (xive->tm_mmap) {
+        memory_region_del_subregion(&xive->tm_mmio, &xive->tm_mmio_kvm);
+        object_unparent(OBJECT(&xive->tm_mmio_kvm));
+        munmap(xive->tm_mmap, 4ull << TM_SHIFT);
+        xive->tm_mmap = NULL;
+    }
 
     /*
      * When the KVM device fd is closed, the KVM device is destroyed
      * and removed from the list of devices of the VM. The VCPU
      * presenters are also detached from the device.
      */
-    close(xive->fd);
-    xive->fd = -1;
+    if (xive->fd != -1) {
+        close(xive->fd);
+        xive->fd = -1;
+    }
 
     kvm_kernel_irqchip = false;
     kvm_msi_via_irqfd_allowed = false;
@@ -819,5 +829,8 @@ void kvmppc_xive_disconnect(SpaprXive *xive, Error **errp)
     kvm_cpu_disable_all();
 
     /* VM Change state handler is not needed anymore */
-    qemu_del_vm_change_state_handler(xive->change);
+    if (xive->change) {
+        qemu_del_vm_change_state_handler(xive->change);
+        xive->change = NULL;
+    }
 }
diff --git a/hw/intc/xics.c b/hw/intc/xics.c
index 29f7d39781..faa976e2f8 100644
--- a/hw/intc/xics.c
+++ b/hw/intc/xics.c
@@ -267,7 +267,14 @@ static int icp_post_load(void *opaque, int version_id)
     ICPState *icp = opaque;
 
     if (kvm_irqchip_in_kernel()) {
-        return icp_set_kvm_state(icp);
+        Error *local_err = NULL;
+        int ret;
+
+        ret = icp_set_kvm_state(icp, &local_err);
+        if (ret < 0) {
+            error_report_err(local_err);
+            return ret;
+        }
     }
 
     return 0;
@@ -300,7 +307,12 @@ static void icp_reset_handler(void *dev)
     qemu_set_irq(icp->output, 0);
 
     if (kvm_irqchip_in_kernel()) {
-        icp_set_kvm_state(ICP(dev));
+        Error *local_err = NULL;
+
+        icp_set_kvm_state(ICP(dev), &local_err);
+        if (local_err) {
+            error_report_err(local_err);
+        }
     }
 }
 
@@ -351,6 +363,7 @@ static void icp_realize(DeviceState *dev, Error **errp)
         return;
     }
 
+    /* Connect the presenter to the VCPU (required for CPU hotplug) */
     if (kvm_irqchip_in_kernel()) {
         icp_kvm_realize(dev, &err);
         if (err) {
@@ -563,7 +576,12 @@ static void ics_simple_reset(DeviceState *dev)
     icsc->parent_reset(dev);
 
     if (kvm_irqchip_in_kernel()) {
-        ics_set_kvm_state(ICS_BASE(dev));
+        Error *local_err = NULL;
+
+        ics_set_kvm_state(ICS_BASE(dev), &local_err);
+        if (local_err) {
+            error_report_err(local_err);
+        }
     }
 }
 
@@ -679,7 +697,14 @@ static int ics_base_post_load(void *opaque, int version_id)
     ICSState *ics = opaque;
 
     if (kvm_irqchip_in_kernel()) {
-        return ics_set_kvm_state(ics);
+        Error *local_err = NULL;
+        int ret;
+
+        ret = ics_set_kvm_state(ics, &local_err);
+        if (ret < 0) {
+            error_report_err(local_err);
+            return ret;
+        }
     }
 
     return 0;
@@ -765,8 +790,13 @@ void ics_set_irq_type(ICSState *ics, int srcno, bool lsi)
         lsi ? XICS_FLAGS_IRQ_LSI : XICS_FLAGS_IRQ_MSI;
 
     if (kvm_irqchip_in_kernel()) {
+        Error *local_err = NULL;
+
         ics_reset_irq(ics->irqs + srcno);
-        ics_set_kvm_state_one(ics, srcno);
+        ics_set_kvm_state_one(ics, srcno, &local_err);
+        if (local_err) {
+            error_report_err(local_err);
+        }
     }
 }
 
diff --git a/hw/intc/xics_kvm.c b/hw/intc/xics_kvm.c
index 5ba5b77561..51433b19b0 100644
--- a/hw/intc/xics_kvm.c
+++ b/hw/intc/xics_kvm.c
@@ -106,7 +106,7 @@ void icp_synchronize_state(ICPState *icp)
     }
 }
 
-int icp_set_kvm_state(ICPState *icp)
+int icp_set_kvm_state(ICPState *icp, Error **errp)
 {
     uint64_t state;
     int ret;
@@ -126,10 +126,11 @@ int icp_set_kvm_state(ICPState *icp)
         | ((uint64_t)icp->pending_priority << KVM_REG_PPC_ICP_PPRI_SHIFT);
 
     ret = kvm_set_one_reg(icp->cs, KVM_REG_PPC_ICP_STATE, &state);
-    if (ret != 0) {
-        error_report("Unable to restore KVM interrupt controller state (0x%"
-                PRIx64 ") for CPU %ld: %s", state, kvm_arch_vcpu_id(icp->cs),
-                strerror(errno));
+    if (ret < 0) {
+        error_setg_errno(errp, -ret,
+                         "Unable to restore KVM interrupt controller state (0x%"
+                         PRIx64 ") for CPU %ld", state,
+                         kvm_arch_vcpu_id(icp->cs));
         return ret;
     }
 
@@ -240,10 +241,9 @@ void ics_synchronize_state(ICSState *ics)
     ics_get_kvm_state(ics);
 }
 
-int ics_set_kvm_state_one(ICSState *ics, int srcno)
+int ics_set_kvm_state_one(ICSState *ics, int srcno, Error **errp)
 {
     uint64_t state;
-    Error *local_err = NULL;
     ICSIRQState *irq = &ics->irqs[srcno];
     int ret;
 
@@ -278,16 +278,15 @@ int ics_set_kvm_state_one(ICSState *ics, int srcno)
     }
 
     ret = kvm_device_access(kernel_xics_fd, KVM_DEV_XICS_GRP_SOURCES,
-                            srcno + ics->offset, &state, true, &local_err);
-    if (local_err) {
-        error_report_err(local_err);
+                            srcno + ics->offset, &state, true, errp);
+    if (ret < 0) {
         return ret;
     }
 
     return 0;
 }
 
-int ics_set_kvm_state(ICSState *ics)
+int ics_set_kvm_state(ICSState *ics, Error **errp)
 {
     int i;
 
@@ -297,10 +296,12 @@ int ics_set_kvm_state(ICSState *ics)
     }
 
     for (i = 0; i < ics->nr_irqs; i++) {
+        Error *local_err = NULL;
         int ret;
 
-        ret = ics_set_kvm_state_one(ics, i);
-        if (ret) {
+        ret = ics_set_kvm_state_one(ics, i, &local_err);
+        if (ret < 0) {
+            error_propagate(errp, local_err);
             return ret;
         }
     }
@@ -331,16 +332,7 @@ void ics_kvm_set_irq(ICSState *ics, int srcno, int val)
     }
 }
 
-static void rtas_dummy(PowerPCCPU *cpu, SpaprMachineState *spapr,
-                       uint32_t token,
-                       uint32_t nargs, target_ulong args,
-                       uint32_t nret, target_ulong rets)
-{
-    error_report("pseries: %s must never be called for in-kernel XICS",
-                 __func__);
-}
-
-int xics_kvm_init(SpaprMachineState *spapr, Error **errp)
+int xics_kvm_connect(SpaprMachineState *spapr, Error **errp)
 {
     int rc;
     CPUState *cs;
@@ -357,42 +349,41 @@ int xics_kvm_init(SpaprMachineState *spapr, Error **errp)
     if (!kvm_enabled() || !kvm_check_extension(kvm_state, KVM_CAP_IRQ_XICS)) {
         error_setg(errp,
                    "KVM and IRQ_XICS capability must be present for in-kernel XICS");
-        goto fail;
+        return -1;
     }
 
-    spapr_rtas_register(RTAS_IBM_SET_XIVE, "ibm,set-xive", rtas_dummy);
-    spapr_rtas_register(RTAS_IBM_GET_XIVE, "ibm,get-xive", rtas_dummy);
-    spapr_rtas_register(RTAS_IBM_INT_OFF, "ibm,int-off", rtas_dummy);
-    spapr_rtas_register(RTAS_IBM_INT_ON, "ibm,int-on", rtas_dummy);
-
     rc = kvmppc_define_rtas_kernel_token(RTAS_IBM_SET_XIVE, "ibm,set-xive");
     if (rc < 0) {
-        error_setg(errp, "kvmppc_define_rtas_kernel_token: ibm,set-xive");
+        error_setg_errno(&local_err, -rc,
+                         "kvmppc_define_rtas_kernel_token: ibm,set-xive");
         goto fail;
     }
 
     rc = kvmppc_define_rtas_kernel_token(RTAS_IBM_GET_XIVE, "ibm,get-xive");
     if (rc < 0) {
-        error_setg(errp, "kvmppc_define_rtas_kernel_token: ibm,get-xive");
+        error_setg_errno(&local_err, -rc,
+                         "kvmppc_define_rtas_kernel_token: ibm,get-xive");
         goto fail;
     }
 
     rc = kvmppc_define_rtas_kernel_token(RTAS_IBM_INT_ON, "ibm,int-on");
     if (rc < 0) {
-        error_setg(errp, "kvmppc_define_rtas_kernel_token: ibm,int-on");
+        error_setg_errno(&local_err, -rc,
+                         "kvmppc_define_rtas_kernel_token: ibm,int-on");
         goto fail;
     }
 
     rc = kvmppc_define_rtas_kernel_token(RTAS_IBM_INT_OFF, "ibm,int-off");
     if (rc < 0) {
-        error_setg(errp, "kvmppc_define_rtas_kernel_token: ibm,int-off");
+        error_setg_errno(&local_err, -rc,
+                         "kvmppc_define_rtas_kernel_token: ibm,int-off");
         goto fail;
     }
 
     /* Create the KVM XICS device */
     rc = kvm_create_device(kvm_state, KVM_DEV_TYPE_XICS, false);
     if (rc < 0) {
-        error_setg_errno(errp, -rc, "Error on KVM_CREATE_DEVICE for XICS");
+        error_setg_errno(&local_err, -rc, "Error on KVM_CREATE_DEVICE for XICS");
         goto fail;
     }
 
@@ -407,27 +398,30 @@ int xics_kvm_init(SpaprMachineState *spapr, Error **errp)
 
         icp_kvm_realize(DEVICE(spapr_cpu_state(cpu)->icp), &local_err);
         if (local_err) {
-            error_propagate(errp, local_err);
             goto fail;
         }
     }
 
     /* Update the KVM sources */
-    ics_set_kvm_state(spapr->ics);
+    ics_set_kvm_state(spapr->ics, &local_err);
+    if (local_err) {
+        goto fail;
+    }
 
     /* Connect the presenters to the initial VCPUs of the machine */
     CPU_FOREACH(cs) {
         PowerPCCPU *cpu = POWERPC_CPU(cs);
-        icp_set_kvm_state(spapr_cpu_state(cpu)->icp);
+        icp_set_kvm_state(spapr_cpu_state(cpu)->icp, &local_err);
+        if (local_err) {
+            goto fail;
+        }
     }
 
     return 0;
 
 fail:
-    kvmppc_define_rtas_kernel_token(0, "ibm,set-xive");
-    kvmppc_define_rtas_kernel_token(0, "ibm,get-xive");
-    kvmppc_define_rtas_kernel_token(0, "ibm,int-on");
-    kvmppc_define_rtas_kernel_token(0, "ibm,int-off");
+    error_propagate(errp, local_err);
+    xics_kvm_disconnect(spapr, NULL);
     return -1;
 }
 
@@ -451,13 +445,10 @@ void xics_kvm_disconnect(SpaprMachineState *spapr, Error **errp)
      * removed from the list of devices of the VM. The VCPU presenters
      * are also detached from the device.
      */
-    close(kernel_xics_fd);
-    kernel_xics_fd = -1;
-
-    spapr_rtas_unregister(RTAS_IBM_SET_XIVE);
-    spapr_rtas_unregister(RTAS_IBM_GET_XIVE);
-    spapr_rtas_unregister(RTAS_IBM_INT_OFF);
-    spapr_rtas_unregister(RTAS_IBM_INT_ON);
+    if (kernel_xics_fd != -1) {
+        close(kernel_xics_fd);
+        kernel_xics_fd = -1;
+    }
 
     kvmppc_define_rtas_kernel_token(0, "ibm,set-xive");
     kvmppc_define_rtas_kernel_token(0, "ibm,get-xive");
@@ -471,3 +462,33 @@ void xics_kvm_disconnect(SpaprMachineState *spapr, Error **errp)
     /* Clear the presenter from the VCPUs */
     kvm_disable_icps();
 }
+
+/*
+ * This is a heuristic to detect older KVMs on POWER9 hosts that don't
+ * support destruction of a KVM XICS device while the VM is running.
+ * Required to start a spapr machine with ic-mode=dual,kernel-irqchip=on.
+ */
+bool xics_kvm_has_broken_disconnect(SpaprMachineState *spapr)
+{
+    int rc;
+
+    rc = kvm_create_device(kvm_state, KVM_DEV_TYPE_XICS, false);
+    if (rc < 0) {
+        /*
+         * The error is ignored on purpose. The KVM XICS setup code
+         * will catch it again anyway. The goal here is to see if
+         * close() actually destroys the device or not.
+         */
+        return false;
+    }
+
+    close(rc);
+
+    rc = kvm_create_device(kvm_state, KVM_DEV_TYPE_XICS, false);
+    if (rc >= 0) {
+        close(rc);
+        return false;
+    }
+
+    return errno == EEXIST;
+}
diff --git a/hw/intc/xics_spapr.c b/hw/intc/xics_spapr.c
index 5a1835e8b1..7cd3c93d71 100644
--- a/hw/intc/xics_spapr.c
+++ b/hw/intc/xics_spapr.c
@@ -41,11 +41,32 @@
  * Guest interfaces
  */
 
+static bool check_emulated_xics(SpaprMachineState *spapr, const char *func)
+{
+    if (spapr_ovec_test(spapr->ov5_cas, OV5_XIVE_EXPLOIT) ||
+        kvm_irqchip_in_kernel()) {
+        error_report("pseries: %s must only be called for emulated XICS",
+                     func);
+        return false;
+    }
+
+    return true;
+}
+
+#define CHECK_EMULATED_XICS_HCALL(spapr)               \
+    do {                                               \
+        if (!check_emulated_xics((spapr), __func__)) { \
+            return H_HARDWARE;                         \
+        }                                              \
+    } while (0)
+
 static target_ulong h_cppr(PowerPCCPU *cpu, SpaprMachineState *spapr,
                            target_ulong opcode, target_ulong *args)
 {
     target_ulong cppr = args[0];
 
+    CHECK_EMULATED_XICS_HCALL(spapr);
+
     icp_set_cppr(spapr_cpu_state(cpu)->icp, cppr);
     return H_SUCCESS;
 }
@@ -56,6 +77,8 @@ static target_ulong h_ipi(PowerPCCPU *cpu, SpaprMachineState *spapr,
     target_ulong mfrr = args[1];
     ICPState *icp = xics_icp_get(XICS_FABRIC(spapr), args[0]);
 
+    CHECK_EMULATED_XICS_HCALL(spapr);
+
     if (!icp) {
         return H_PARAMETER;
     }
@@ -69,6 +92,8 @@ static target_ulong h_xirr(PowerPCCPU *cpu, SpaprMachineState *spapr,
 {
     uint32_t xirr = icp_accept(spapr_cpu_state(cpu)->icp);
 
+    CHECK_EMULATED_XICS_HCALL(spapr);
+
     args[0] = xirr;
     return H_SUCCESS;
 }
@@ -78,6 +103,8 @@ static target_ulong h_xirr_x(PowerPCCPU *cpu, SpaprMachineState *spapr,
 {
     uint32_t xirr = icp_accept(spapr_cpu_state(cpu)->icp);
 
+    CHECK_EMULATED_XICS_HCALL(spapr);
+
     args[0] = xirr;
     args[1] = cpu_get_host_ticks();
     return H_SUCCESS;
@@ -88,6 +115,8 @@ static target_ulong h_eoi(PowerPCCPU *cpu, SpaprMachineState *spapr,
 {
     target_ulong xirr = args[0];
 
+    CHECK_EMULATED_XICS_HCALL(spapr);
+
     icp_eoi(spapr_cpu_state(cpu)->icp, xirr);
     return H_SUCCESS;
 }
@@ -99,6 +128,8 @@ static target_ulong h_ipoll(PowerPCCPU *cpu, SpaprMachineState *spapr,
     uint32_t mfrr;
     uint32_t xirr;
 
+    CHECK_EMULATED_XICS_HCALL(spapr);
+
     if (!icp) {
         return H_PARAMETER;
     }
@@ -111,6 +142,14 @@ static target_ulong h_ipoll(PowerPCCPU *cpu, SpaprMachineState *spapr,
     return H_SUCCESS;
 }
 
+#define CHECK_EMULATED_XICS_RTAS(spapr, rets)          \
+    do {                                               \
+        if (!check_emulated_xics((spapr), __func__)) { \
+            rtas_st((rets), 0, RTAS_OUT_HW_ERROR);     \
+            return;                                    \
+        }                                              \
+    } while (0)
+
 static void rtas_set_xive(PowerPCCPU *cpu, SpaprMachineState *spapr,
                           uint32_t token,
                           uint32_t nargs, target_ulong args,
@@ -119,6 +158,8 @@ static void rtas_set_xive(PowerPCCPU *cpu, SpaprMachineState *spapr,
     ICSState *ics = spapr->ics;
     uint32_t nr, srcno, server, priority;
 
+    CHECK_EMULATED_XICS_RTAS(spapr, rets);
+
     if ((nargs != 3) || (nret != 1)) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
         return;
@@ -152,6 +193,8 @@ static void rtas_get_xive(PowerPCCPU *cpu, SpaprMachineState *spapr,
     ICSState *ics = spapr->ics;
     uint32_t nr, srcno;
 
+    CHECK_EMULATED_XICS_RTAS(spapr, rets);
+
     if ((nargs != 1) || (nret != 3)) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
         return;
@@ -182,6 +225,8 @@ static void rtas_int_off(PowerPCCPU *cpu, SpaprMachineState *spapr,
     ICSState *ics = spapr->ics;
     uint32_t nr, srcno;
 
+    CHECK_EMULATED_XICS_RTAS(spapr, rets);
+
     if ((nargs != 1) || (nret != 1)) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
         return;
@@ -213,6 +258,8 @@ static void rtas_int_on(PowerPCCPU *cpu, SpaprMachineState *spapr,
     ICSState *ics = spapr->ics;
     uint32_t nr, srcno;
 
+    CHECK_EMULATED_XICS_RTAS(spapr, rets);
+
     if ((nargs != 1) || (nret != 1)) {
         rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
         return;
@@ -239,14 +286,6 @@ static void rtas_int_on(PowerPCCPU *cpu, SpaprMachineState *spapr,
 
 void xics_spapr_init(SpaprMachineState *spapr)
 {
-    /* Emulated mode can only be initialized once. */
-    if (spapr->ics->init) {
-        return;
-    }
-
-    spapr->ics->init = true;
-
-    /* Registration of global state belongs into realize */
     spapr_rtas_register(RTAS_IBM_SET_XIVE, "ibm,set-xive", rtas_set_xive);
     spapr_rtas_register(RTAS_IBM_GET_XIVE, "ibm,get-xive", rtas_get_xive);
     spapr_rtas_register(RTAS_IBM_INT_OFF, "ibm,int-off", rtas_int_off);
diff --git a/hw/intc/xive.c b/hw/intc/xive.c
index 6250c0414d..cf77bdb7d3 100644
--- a/hw/intc/xive.c
+++ b/hw/intc/xive.c
@@ -132,6 +132,11 @@ static void xive_tctx_set_cppr(XiveTCTX *tctx, uint8_t ring, uint8_t cppr)
     xive_tctx_notify(tctx, ring);
 }
 
+static inline uint32_t xive_tctx_word2(uint8_t *ring)
+{
+    return *((uint32_t *) &ring[TM_WORD2]);
+}
+
 /*
  * XIVE Thread Interrupt Management Area (TIMA)
  */
@@ -150,11 +155,12 @@ static uint64_t xive_tm_ack_hv_reg(XiveTCTX *tctx, hwaddr offset, unsigned size)
 static uint64_t xive_tm_pull_pool_ctx(XiveTCTX *tctx, hwaddr offset,
                                       unsigned size)
 {
-    uint64_t ret;
+    uint32_t qw2w2_prev = xive_tctx_word2(&tctx->regs[TM_QW2_HV_POOL]);
+    uint32_t qw2w2;
 
-    ret = tctx->regs[TM_QW2_HV_POOL + TM_WORD2] & TM_QW2W2_POOL_CAM;
-    tctx->regs[TM_QW2_HV_POOL + TM_WORD2] &= ~TM_QW2W2_POOL_CAM;
-    return ret;
+    qw2w2 = xive_set_field32(TM_QW2W2_VP, qw2w2_prev, 0);
+    memcpy(&tctx->regs[TM_QW2_HV_POOL + TM_WORD2], &qw2w2, 4);
+    return qw2w2;
 }
 
 static void xive_tm_vt_push(XiveTCTX *tctx, hwaddr offset,
@@ -182,31 +188,31 @@ static uint64_t xive_tm_vt_poll(XiveTCTX *tctx, hwaddr offset, unsigned size)
  */
 
 static const uint8_t xive_tm_hw_view[] = {
-    /* QW-0 User */   3, 0, 0, 0,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-1 OS   */   3, 3, 3, 3,   3, 3, 0, 3,   3, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-2 POOL */   0, 0, 3, 3,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-3 PHYS */   3, 3, 3, 3,   0, 3, 0, 3,   3, 0, 0, 3,   3, 3, 3, 0,
+    3, 0, 0, 0,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0, /* QW-0 User */
+    3, 3, 3, 3,   3, 3, 0, 2,   3, 3, 3, 3,   0, 0, 0, 0, /* QW-1 OS   */
+    0, 0, 3, 3,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0, /* QW-2 POOL */
+    3, 3, 3, 3,   0, 3, 0, 2,   3, 0, 0, 3,   3, 3, 3, 0, /* QW-3 PHYS */
 };
 
 static const uint8_t xive_tm_hv_view[] = {
-    /* QW-0 User */   3, 0, 0, 0,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-1 OS   */   3, 3, 3, 3,   3, 3, 0, 3,   3, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-2 POOL */   0, 0, 3, 3,   0, 0, 0, 0,   0, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-3 PHYS */   3, 3, 3, 3,   0, 3, 0, 3,   3, 0, 0, 3,   0, 0, 0, 0,
+    3, 0, 0, 0,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0, /* QW-0 User */
+    3, 3, 3, 3,   3, 3, 0, 2,   3, 3, 3, 3,   0, 0, 0, 0, /* QW-1 OS   */
+    0, 0, 3, 3,   0, 0, 0, 0,   0, 3, 3, 3,   0, 0, 0, 0, /* QW-2 POOL */
+    3, 3, 3, 3,   0, 3, 0, 2,   3, 0, 0, 3,   0, 0, 0, 0, /* QW-3 PHYS */
 };
 
 static const uint8_t xive_tm_os_view[] = {
-    /* QW-0 User */   3, 0, 0, 0,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0,
-    /* QW-1 OS   */   2, 3, 2, 2,   2, 2, 0, 2,   0, 0, 0, 0,   0, 0, 0, 0,
-    /* QW-2 POOL */   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,
-    /* QW-3 PHYS */   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,
+    3, 0, 0, 0,   0, 0, 0, 0,   3, 3, 3, 3,   0, 0, 0, 0, /* QW-0 User */
+    2, 3, 2, 2,   2, 2, 0, 2,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-1 OS   */
+    0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-2 POOL */
+    0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-3 PHYS */
 };
 
 static const uint8_t xive_tm_user_view[] = {
-    /* QW-0 User */   3, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,
-    /* QW-1 OS   */   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,
-    /* QW-2 POOL */   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,
-    /* QW-3 PHYS */   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,
+    3, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-0 User */
+    0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-1 OS   */
+    0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-2 POOL */
+    0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0,   0, 0, 0, 0, /* QW-3 PHYS */
 };
 
 /*
@@ -484,11 +490,6 @@ const MemoryRegionOps xive_tm_ops = {
     },
 };
 
-static inline uint32_t xive_tctx_word2(uint8_t *ring)
-{
-    return *((uint32_t *) &ring[TM_WORD2]);
-}
-
 static char *xive_tctx_ring_print(uint8_t *ring)
 {
     uint32_t w2 = xive_tctx_word2(ring);
@@ -1229,27 +1230,16 @@ XiveTCTX *xive_router_get_tctx(XiveRouter *xrtr, CPUState *cs)
 }
 
 /*
- * By default on P9, the HW CAM line (23bits) is hardwired to :
- *
- *   0x000||0b1||4Bit chip number||7Bit Thread number.
+ * Encode the HW CAM line in the block group mode format :
  *
- * When the block grouping is enabled, the CAM line is changed to :
- *
- *   4Bit chip number||0x001||7Bit Thread number.
+ *   chip << 19 | 0000000 0 0001 thread (7Bit)
  */
-static uint32_t hw_cam_line(uint8_t chip_id, uint8_t tid)
-{
-    return 1 << 11 | (chip_id & 0xf) << 7 | (tid & 0x7f);
-}
-
-static bool xive_presenter_tctx_match_hw(XiveTCTX *tctx,
-                                         uint8_t nvt_blk, uint32_t nvt_idx)
+static uint32_t xive_tctx_hw_cam_line(XiveTCTX *tctx)
 {
     CPUPPCState *env = &POWERPC_CPU(tctx->cs)->env;
     uint32_t pir = env->spr_cb[SPR_PIR].default_value;
 
-    return hw_cam_line((pir >> 8) & 0xf, pir & 0x7f) ==
-        hw_cam_line(nvt_blk, nvt_idx);
+    return xive_nvt_cam_line((pir >> 8) & 0xf, 1 << 7 | (pir & 0x7f));
 }
 
 /*
@@ -1285,7 +1275,7 @@ static int xive_presenter_tctx_match(XiveTCTX *tctx, uint8_t format,
 
         /* PHYS ring */
         if ((be32_to_cpu(qw3w2) & TM_QW3W2_VT) &&
-            xive_presenter_tctx_match_hw(tctx, nvt_blk, nvt_idx)) {
+            cam == xive_tctx_hw_cam_line(tctx)) {
             return TM_QW3_HV_PHYS;
         }
 
diff --git a/hw/mips/Kconfig b/hw/mips/Kconfig
index cdc07e59b6..62aa01b29e 100644
--- a/hw/mips/Kconfig
+++ b/hw/mips/Kconfig
@@ -1,14 +1,44 @@
 config R4K
     bool
+    select ISA_BUS
+    select SERIAL_ISA
+    select I8259
+    select I8254
+    select MC146818RTC
+    imply VGA_ISA
+    imply NE2000_ISA
+    select IDE_ISA
+    select PCKBD
+    select PFLASH_CFI01
 
 config MALTA
     bool
 
 config MIPSSIM
     bool
+    select ISA_BUS
+    select SERIAL_ISA
+    select MIPSNET
 
 config JAZZ
     bool
+    select ISA_BUS
+    select RC4030
+    select I8259
+    select I8254
+    select I8257
+    select PCSPK
+    select VGA_ISA_MM
+    select G364FB
+    select DP8393X
+    select ESP
+    select FDC
+    select MC146818RTC
+    select PCKBD
+    select SERIAL
+    select PARALLEL
+    select DS1225Y
+    select JAZZ_LED
 
 config FULONG
     bool
diff --git a/hw/mips/gt64xxx_pci.c b/hw/mips/gt64xxx_pci.c
index f707e59c7a..2fa313f498 100644
--- a/hw/mips/gt64xxx_pci.c
+++ b/hw/mips/gt64xxx_pci.c
@@ -23,207 +23,202 @@
  */
 
 #include "qemu/osdep.h"
+#include "qemu/units.h"
+#include "qemu/log.h"
 #include "hw/hw.h"
 #include "hw/mips/mips.h"
 #include "hw/pci/pci.h"
 #include "hw/pci/pci_host.h"
 #include "hw/i386/pc.h"
 #include "exec/address-spaces.h"
+#include "trace.h"
 
-//#define DEBUG
-
-#ifdef DEBUG
-#define DPRINTF(fmt, ...) fprintf(stderr, "%s: " fmt, __func__, ##__VA_ARGS__)
-#else
-#define DPRINTF(fmt, ...)
-#endif
-
-#define GT_REGS			(0x1000 >> 2)
+#define GT_REGS                 (0x1000 >> 2)
 
 /* CPU Configuration */
-#define GT_CPU    		(0x000 >> 2)
-#define GT_MULTI    		(0x120 >> 2)
+#define GT_CPU                  (0x000 >> 2)
+#define GT_MULTI                (0x120 >> 2)
 
 /* CPU Address Decode */
-#define GT_SCS10LD    		(0x008 >> 2)
-#define GT_SCS10HD    		(0x010 >> 2)
-#define GT_SCS32LD    		(0x018 >> 2)
-#define GT_SCS32HD    		(0x020 >> 2)
-#define GT_CS20LD    		(0x028 >> 2)
-#define GT_CS20HD    		(0x030 >> 2)
-#define GT_CS3BOOTLD    	(0x038 >> 2)
-#define GT_CS3BOOTHD    	(0x040 >> 2)
-#define GT_PCI0IOLD    		(0x048 >> 2)
-#define GT_PCI0IOHD    		(0x050 >> 2)
-#define GT_PCI0M0LD    		(0x058 >> 2)
-#define GT_PCI0M0HD    		(0x060 >> 2)
-#define GT_PCI0M1LD    		(0x080 >> 2)
-#define GT_PCI0M1HD    		(0x088 >> 2)
-#define GT_PCI1IOLD    		(0x090 >> 2)
-#define GT_PCI1IOHD    		(0x098 >> 2)
-#define GT_PCI1M0LD    		(0x0a0 >> 2)
-#define GT_PCI1M0HD    		(0x0a8 >> 2)
-#define GT_PCI1M1LD    		(0x0b0 >> 2)
-#define GT_PCI1M1HD    		(0x0b8 >> 2)
-#define GT_ISD    		(0x068 >> 2)
-
-#define GT_SCS10AR    		(0x0d0 >> 2)
-#define GT_SCS32AR    		(0x0d8 >> 2)
-#define GT_CS20R    		(0x0e0 >> 2)
-#define GT_CS3BOOTR    		(0x0e8 >> 2)
-
-#define GT_PCI0IOREMAP    	(0x0f0 >> 2)
-#define GT_PCI0M0REMAP    	(0x0f8 >> 2)
-#define GT_PCI0M1REMAP    	(0x100 >> 2)
-#define GT_PCI1IOREMAP    	(0x108 >> 2)
-#define GT_PCI1M0REMAP    	(0x110 >> 2)
-#define GT_PCI1M1REMAP    	(0x118 >> 2)
+#define GT_SCS10LD              (0x008 >> 2)
+#define GT_SCS10HD              (0x010 >> 2)
+#define GT_SCS32LD              (0x018 >> 2)
+#define GT_SCS32HD              (0x020 >> 2)
+#define GT_CS20LD               (0x028 >> 2)
+#define GT_CS20HD               (0x030 >> 2)
+#define GT_CS3BOOTLD            (0x038 >> 2)
+#define GT_CS3BOOTHD            (0x040 >> 2)
+#define GT_PCI0IOLD             (0x048 >> 2)
+#define GT_PCI0IOHD             (0x050 >> 2)
+#define GT_PCI0M0LD             (0x058 >> 2)
+#define GT_PCI0M0HD             (0x060 >> 2)
+#define GT_PCI0M1LD             (0x080 >> 2)
+#define GT_PCI0M1HD             (0x088 >> 2)
+#define GT_PCI1IOLD             (0x090 >> 2)
+#define GT_PCI1IOHD             (0x098 >> 2)
+#define GT_PCI1M0LD             (0x0a0 >> 2)
+#define GT_PCI1M0HD             (0x0a8 >> 2)
+#define GT_PCI1M1LD             (0x0b0 >> 2)
+#define GT_PCI1M1HD             (0x0b8 >> 2)
+#define GT_ISD                  (0x068 >> 2)
+
+#define GT_SCS10AR              (0x0d0 >> 2)
+#define GT_SCS32AR              (0x0d8 >> 2)
+#define GT_CS20R                (0x0e0 >> 2)
+#define GT_CS3BOOTR             (0x0e8 >> 2)
+
+#define GT_PCI0IOREMAP          (0x0f0 >> 2)
+#define GT_PCI0M0REMAP          (0x0f8 >> 2)
+#define GT_PCI0M1REMAP          (0x100 >> 2)
+#define GT_PCI1IOREMAP          (0x108 >> 2)
+#define GT_PCI1M0REMAP          (0x110 >> 2)
+#define GT_PCI1M1REMAP          (0x118 >> 2)
 
 /* CPU Error Report */
-#define GT_CPUERR_ADDRLO    	(0x070 >> 2)
-#define GT_CPUERR_ADDRHI    	(0x078 >> 2)
-#define GT_CPUERR_DATALO    	(0x128 >> 2)		/* GT-64120A only  */
-#define GT_CPUERR_DATAHI    	(0x130 >> 2)		/* GT-64120A only  */
-#define GT_CPUERR_PARITY    	(0x138 >> 2)		/* GT-64120A only  */
+#define GT_CPUERR_ADDRLO        (0x070 >> 2)
+#define GT_CPUERR_ADDRHI        (0x078 >> 2)
+#define GT_CPUERR_DATALO        (0x128 >> 2)        /* GT-64120A only  */
+#define GT_CPUERR_DATAHI        (0x130 >> 2)        /* GT-64120A only  */
+#define GT_CPUERR_PARITY        (0x138 >> 2)        /* GT-64120A only  */
 
 /* CPU Sync Barrier */
-#define GT_PCI0SYNC    		(0x0c0 >> 2)
-#define GT_PCI1SYNC    		(0x0c8 >> 2)
+#define GT_PCI0SYNC             (0x0c0 >> 2)
+#define GT_PCI1SYNC             (0x0c8 >> 2)
 
 /* SDRAM and Device Address Decode */
-#define GT_SCS0LD    		(0x400 >> 2)
-#define GT_SCS0HD    		(0x404 >> 2)
-#define GT_SCS1LD    		(0x408 >> 2)
-#define GT_SCS1HD    		(0x40c >> 2)
-#define GT_SCS2LD    		(0x410 >> 2)
-#define GT_SCS2HD    		(0x414 >> 2)
-#define GT_SCS3LD    		(0x418 >> 2)
-#define GT_SCS3HD    		(0x41c >> 2)
-#define GT_CS0LD    		(0x420 >> 2)
-#define GT_CS0HD    		(0x424 >> 2)
-#define GT_CS1LD    		(0x428 >> 2)
-#define GT_CS1HD    		(0x42c >> 2)
-#define GT_CS2LD    		(0x430 >> 2)
-#define GT_CS2HD    		(0x434 >> 2)
-#define GT_CS3LD    		(0x438 >> 2)
-#define GT_CS3HD    		(0x43c >> 2)
-#define GT_BOOTLD    		(0x440 >> 2)
-#define GT_BOOTHD    		(0x444 >> 2)
-#define GT_ADERR    		(0x470 >> 2)
+#define GT_SCS0LD               (0x400 >> 2)
+#define GT_SCS0HD               (0x404 >> 2)
+#define GT_SCS1LD               (0x408 >> 2)
+#define GT_SCS1HD               (0x40c >> 2)
+#define GT_SCS2LD               (0x410 >> 2)
+#define GT_SCS2HD               (0x414 >> 2)
+#define GT_SCS3LD               (0x418 >> 2)
+#define GT_SCS3HD               (0x41c >> 2)
+#define GT_CS0LD                (0x420 >> 2)
+#define GT_CS0HD                (0x424 >> 2)
+#define GT_CS1LD                (0x428 >> 2)
+#define GT_CS1HD                (0x42c >> 2)
+#define GT_CS2LD                (0x430 >> 2)
+#define GT_CS2HD                (0x434 >> 2)
+#define GT_CS3LD                (0x438 >> 2)
+#define GT_CS3HD                (0x43c >> 2)
+#define GT_BOOTLD               (0x440 >> 2)
+#define GT_BOOTHD               (0x444 >> 2)
+#define GT_ADERR                (0x470 >> 2)
 
 /* SDRAM Configuration */
-#define GT_SDRAM_CFG    	(0x448 >> 2)
-#define GT_SDRAM_OPMODE    	(0x474 >> 2)
-#define GT_SDRAM_BM    		(0x478 >> 2)
-#define GT_SDRAM_ADDRDECODE    	(0x47c >> 2)
+#define GT_SDRAM_CFG            (0x448 >> 2)
+#define GT_SDRAM_OPMODE         (0x474 >> 2)
+#define GT_SDRAM_BM             (0x478 >> 2)
+#define GT_SDRAM_ADDRDECODE     (0x47c >> 2)
 
 /* SDRAM Parameters */
-#define GT_SDRAM_B0    		(0x44c >> 2)
-#define GT_SDRAM_B1    		(0x450 >> 2)
-#define GT_SDRAM_B2    		(0x454 >> 2)
-#define GT_SDRAM_B3    		(0x458 >> 2)
+#define GT_SDRAM_B0             (0x44c >> 2)
+#define GT_SDRAM_B1             (0x450 >> 2)
+#define GT_SDRAM_B2             (0x454 >> 2)
+#define GT_SDRAM_B3             (0x458 >> 2)
 
 /* Device Parameters */
-#define GT_DEV_B0    		(0x45c >> 2)
-#define GT_DEV_B1    		(0x460 >> 2)
-#define GT_DEV_B2    		(0x464 >> 2)
-#define GT_DEV_B3    		(0x468 >> 2)
-#define GT_DEV_BOOT    		(0x46c >> 2)
+#define GT_DEV_B0               (0x45c >> 2)
+#define GT_DEV_B1               (0x460 >> 2)
+#define GT_DEV_B2               (0x464 >> 2)
+#define GT_DEV_B3               (0x468 >> 2)
+#define GT_DEV_BOOT             (0x46c >> 2)
 
 /* ECC */
-#define GT_ECC_ERRDATALO	(0x480 >> 2)		/* GT-64120A only  */
-#define GT_ECC_ERRDATAHI	(0x484 >> 2)		/* GT-64120A only  */
-#define GT_ECC_MEM		(0x488 >> 2)		/* GT-64120A only  */
-#define GT_ECC_CALC		(0x48c >> 2)		/* GT-64120A only  */
-#define GT_ECC_ERRADDR		(0x490 >> 2)		/* GT-64120A only  */
+#define GT_ECC_ERRDATALO        (0x480 >> 2)        /* GT-64120A only  */
+#define GT_ECC_ERRDATAHI        (0x484 >> 2)        /* GT-64120A only  */
+#define GT_ECC_MEM              (0x488 >> 2)        /* GT-64120A only  */
+#define GT_ECC_CALC             (0x48c >> 2)        /* GT-64120A only  */
+#define GT_ECC_ERRADDR          (0x490 >> 2)        /* GT-64120A only  */
 
 /* DMA Record */
-#define GT_DMA0_CNT    		(0x800 >> 2)
-#define GT_DMA1_CNT    		(0x804 >> 2)
-#define GT_DMA2_CNT    		(0x808 >> 2)
-#define GT_DMA3_CNT    		(0x80c >> 2)
-#define GT_DMA0_SA    		(0x810 >> 2)
-#define GT_DMA1_SA    		(0x814 >> 2)
-#define GT_DMA2_SA    		(0x818 >> 2)
-#define GT_DMA3_SA    		(0x81c >> 2)
-#define GT_DMA0_DA    		(0x820 >> 2)
-#define GT_DMA1_DA    		(0x824 >> 2)
-#define GT_DMA2_DA    		(0x828 >> 2)
-#define GT_DMA3_DA    		(0x82c >> 2)
-#define GT_DMA0_NEXT    	(0x830 >> 2)
-#define GT_DMA1_NEXT    	(0x834 >> 2)
-#define GT_DMA2_NEXT    	(0x838 >> 2)
-#define GT_DMA3_NEXT    	(0x83c >> 2)
-#define GT_DMA0_CUR    		(0x870 >> 2)
-#define GT_DMA1_CUR    		(0x874 >> 2)
-#define GT_DMA2_CUR    		(0x878 >> 2)
-#define GT_DMA3_CUR    		(0x87c >> 2)
+#define GT_DMA0_CNT             (0x800 >> 2)
+#define GT_DMA1_CNT             (0x804 >> 2)
+#define GT_DMA2_CNT             (0x808 >> 2)
+#define GT_DMA3_CNT             (0x80c >> 2)
+#define GT_DMA0_SA              (0x810 >> 2)
+#define GT_DMA1_SA              (0x814 >> 2)
+#define GT_DMA2_SA              (0x818 >> 2)
+#define GT_DMA3_SA              (0x81c >> 2)
+#define GT_DMA0_DA              (0x820 >> 2)
+#define GT_DMA1_DA              (0x824 >> 2)
+#define GT_DMA2_DA              (0x828 >> 2)
+#define GT_DMA3_DA              (0x82c >> 2)
+#define GT_DMA0_NEXT            (0x830 >> 2)
+#define GT_DMA1_NEXT            (0x834 >> 2)
+#define GT_DMA2_NEXT            (0x838 >> 2)
+#define GT_DMA3_NEXT            (0x83c >> 2)
+#define GT_DMA0_CUR             (0x870 >> 2)
+#define GT_DMA1_CUR             (0x874 >> 2)
+#define GT_DMA2_CUR             (0x878 >> 2)
+#define GT_DMA3_CUR             (0x87c >> 2)
 
 /* DMA Channel Control */
-#define GT_DMA0_CTRL    	(0x840 >> 2)
-#define GT_DMA1_CTRL    	(0x844 >> 2)
-#define GT_DMA2_CTRL    	(0x848 >> 2)
-#define GT_DMA3_CTRL    	(0x84c >> 2)
+#define GT_DMA0_CTRL            (0x840 >> 2)
+#define GT_DMA1_CTRL            (0x844 >> 2)
+#define GT_DMA2_CTRL            (0x848 >> 2)
+#define GT_DMA3_CTRL            (0x84c >> 2)
 
 /* DMA Arbiter */
-#define GT_DMA_ARB    		(0x860 >> 2)
+#define GT_DMA_ARB              (0x860 >> 2)
 
 /* Timer/Counter */
-#define GT_TC0    		(0x850 >> 2)
-#define GT_TC1    		(0x854 >> 2)
-#define GT_TC2    		(0x858 >> 2)
-#define GT_TC3    		(0x85c >> 2)
-#define GT_TC_CONTROL    	(0x864 >> 2)
+#define GT_TC0                  (0x850 >> 2)
+#define GT_TC1                  (0x854 >> 2)
+#define GT_TC2                  (0x858 >> 2)
+#define GT_TC3                  (0x85c >> 2)
+#define GT_TC_CONTROL           (0x864 >> 2)
 
 /* PCI Internal */
-#define GT_PCI0_CMD    		(0xc00 >> 2)
-#define GT_PCI0_TOR    		(0xc04 >> 2)
-#define GT_PCI0_BS_SCS10    	(0xc08 >> 2)
-#define GT_PCI0_BS_SCS32    	(0xc0c >> 2)
-#define GT_PCI0_BS_CS20    	(0xc10 >> 2)
-#define GT_PCI0_BS_CS3BT    	(0xc14 >> 2)
-#define GT_PCI1_IACK    	(0xc30 >> 2)
-#define GT_PCI0_IACK    	(0xc34 >> 2)
-#define GT_PCI0_BARE    	(0xc3c >> 2)
-#define GT_PCI0_PREFMBR    	(0xc40 >> 2)
-#define GT_PCI0_SCS10_BAR    	(0xc48 >> 2)
-#define GT_PCI0_SCS32_BAR    	(0xc4c >> 2)
-#define GT_PCI0_CS20_BAR    	(0xc50 >> 2)
-#define GT_PCI0_CS3BT_BAR    	(0xc54 >> 2)
-#define GT_PCI0_SSCS10_BAR    	(0xc58 >> 2)
-#define GT_PCI0_SSCS32_BAR    	(0xc5c >> 2)
-#define GT_PCI0_SCS3BT_BAR    	(0xc64 >> 2)
-#define GT_PCI1_CMD    		(0xc80 >> 2)
-#define GT_PCI1_TOR    		(0xc84 >> 2)
-#define GT_PCI1_BS_SCS10    	(0xc88 >> 2)
-#define GT_PCI1_BS_SCS32    	(0xc8c >> 2)
-#define GT_PCI1_BS_CS20    	(0xc90 >> 2)
-#define GT_PCI1_BS_CS3BT    	(0xc94 >> 2)
-#define GT_PCI1_BARE    	(0xcbc >> 2)
-#define GT_PCI1_PREFMBR    	(0xcc0 >> 2)
-#define GT_PCI1_SCS10_BAR    	(0xcc8 >> 2)
-#define GT_PCI1_SCS32_BAR    	(0xccc >> 2)
-#define GT_PCI1_CS20_BAR    	(0xcd0 >> 2)
-#define GT_PCI1_CS3BT_BAR    	(0xcd4 >> 2)
-#define GT_PCI1_SSCS10_BAR    	(0xcd8 >> 2)
-#define GT_PCI1_SSCS32_BAR    	(0xcdc >> 2)
-#define GT_PCI1_SCS3BT_BAR    	(0xce4 >> 2)
-#define GT_PCI1_CFGADDR    	(0xcf0 >> 2)
-#define GT_PCI1_CFGDATA    	(0xcf4 >> 2)
-#define GT_PCI0_CFGADDR    	(0xcf8 >> 2)
-#define GT_PCI0_CFGDATA    	(0xcfc >> 2)
+#define GT_PCI0_CMD             (0xc00 >> 2)
+#define GT_PCI0_TOR             (0xc04 >> 2)
+#define GT_PCI0_BS_SCS10        (0xc08 >> 2)
+#define GT_PCI0_BS_SCS32        (0xc0c >> 2)
+#define GT_PCI0_BS_CS20         (0xc10 >> 2)
+#define GT_PCI0_BS_CS3BT        (0xc14 >> 2)
+#define GT_PCI1_IACK            (0xc30 >> 2)
+#define GT_PCI0_IACK            (0xc34 >> 2)
+#define GT_PCI0_BARE            (0xc3c >> 2)
+#define GT_PCI0_PREFMBR         (0xc40 >> 2)
+#define GT_PCI0_SCS10_BAR       (0xc48 >> 2)
+#define GT_PCI0_SCS32_BAR       (0xc4c >> 2)
+#define GT_PCI0_CS20_BAR        (0xc50 >> 2)
+#define GT_PCI0_CS3BT_BAR       (0xc54 >> 2)
+#define GT_PCI0_SSCS10_BAR      (0xc58 >> 2)
+#define GT_PCI0_SSCS32_BAR      (0xc5c >> 2)
+#define GT_PCI0_SCS3BT_BAR      (0xc64 >> 2)
+#define GT_PCI1_CMD             (0xc80 >> 2)
+#define GT_PCI1_TOR             (0xc84 >> 2)
+#define GT_PCI1_BS_SCS10        (0xc88 >> 2)
+#define GT_PCI1_BS_SCS32        (0xc8c >> 2)
+#define GT_PCI1_BS_CS20         (0xc90 >> 2)
+#define GT_PCI1_BS_CS3BT        (0xc94 >> 2)
+#define GT_PCI1_BARE            (0xcbc >> 2)
+#define GT_PCI1_PREFMBR         (0xcc0 >> 2)
+#define GT_PCI1_SCS10_BAR       (0xcc8 >> 2)
+#define GT_PCI1_SCS32_BAR       (0xccc >> 2)
+#define GT_PCI1_CS20_BAR        (0xcd0 >> 2)
+#define GT_PCI1_CS3BT_BAR       (0xcd4 >> 2)
+#define GT_PCI1_SSCS10_BAR      (0xcd8 >> 2)
+#define GT_PCI1_SSCS32_BAR      (0xcdc >> 2)
+#define GT_PCI1_SCS3BT_BAR      (0xce4 >> 2)
+#define GT_PCI1_CFGADDR         (0xcf0 >> 2)
+#define GT_PCI1_CFGDATA         (0xcf4 >> 2)
+#define GT_PCI0_CFGADDR         (0xcf8 >> 2)
+#define GT_PCI0_CFGDATA         (0xcfc >> 2)
 
 /* Interrupts */
-#define GT_INTRCAUSE    	(0xc18 >> 2)
-#define GT_INTRMASK    		(0xc1c >> 2)
-#define GT_PCI0_ICMASK    	(0xc24 >> 2)
-#define GT_PCI0_SERR0MASK    	(0xc28 >> 2)
-#define GT_CPU_INTSEL    	(0xc70 >> 2)
-#define GT_PCI0_INTSEL    	(0xc74 >> 2)
-#define GT_HINTRCAUSE    	(0xc98 >> 2)
-#define GT_HINTRMASK    	(0xc9c >> 2)
-#define GT_PCI0_HICMASK    	(0xca4 >> 2)
-#define GT_PCI1_SERR1MASK    	(0xca8 >> 2)
+#define GT_INTRCAUSE            (0xc18 >> 2)
+#define GT_INTRMASK             (0xc1c >> 2)
+#define GT_PCI0_ICMASK          (0xc24 >> 2)
+#define GT_PCI0_SERR0MASK       (0xc28 >> 2)
+#define GT_CPU_INTSEL           (0xc70 >> 2)
+#define GT_PCI0_INTSEL          (0xc74 >> 2)
+#define GT_HINTRCAUSE           (0xc98 >> 2)
+#define GT_HINTRMASK            (0xc9c >> 2)
+#define GT_PCI0_HICMASK         (0xca4 >> 2)
+#define GT_PCI1_SERR1MASK       (0xca8 >> 2)
 
 #define PCI_MAPPING_ENTRY(regname)            \
     hwaddr regname ##_start;      \
@@ -248,27 +243,34 @@ typedef struct GT64120State {
 } GT64120State;
 
 /* Adjust range to avoid touching space which isn't mappable via PCI */
-/* XXX: Hardcoded values for Malta: 0x1e000000 - 0x1f100000
-                                    0x1fc00000 - 0x1fd00000  */
-static void check_reserved_space (hwaddr *start,
-                                  hwaddr *length)
+/*
+ * XXX: Hardcoded values for Malta: 0x1e000000 - 0x1f100000
+ *                                  0x1fc00000 - 0x1fd00000
+ */
+static void check_reserved_space(hwaddr *start, hwaddr *length)
 {
     hwaddr begin = *start;
     hwaddr end = *start + *length;
 
-    if (end >= 0x1e000000LL && end < 0x1f100000LL)
+    if (end >= 0x1e000000LL && end < 0x1f100000LL) {
         end = 0x1e000000LL;
-    if (begin >= 0x1e000000LL && begin < 0x1f100000LL)
+    }
+    if (begin >= 0x1e000000LL && begin < 0x1f100000LL) {
         begin = 0x1f100000LL;
-    if (end >= 0x1fc00000LL && end < 0x1fd00000LL)
+    }
+    if (end >= 0x1fc00000LL && end < 0x1fd00000LL) {
         end = 0x1fc00000LL;
-    if (begin >= 0x1fc00000LL && begin < 0x1fd00000LL)
+    }
+    if (begin >= 0x1fc00000LL && begin < 0x1fd00000LL) {
         begin = 0x1fd00000LL;
+    }
     /* XXX: This is broken when a reserved range splits the requested range */
-    if (end >= 0x1f100000LL && begin < 0x1e000000LL)
+    if (end >= 0x1f100000LL && begin < 0x1e000000LL) {
         end = 0x1e000000LL;
-    if (end >= 0x1fd00000LL && begin < 0x1fc00000LL)
+    }
+    if (end >= 0x1fd00000LL && begin < 0x1fc00000LL) {
         end = 0x1fc00000LL;
+    }
 
     *start = begin;
     *length = end - begin;
@@ -286,9 +288,7 @@ static void gt64120_isd_mapping(GT64120State *s)
     check_reserved_space(&start, &length);
     length = 0x1000;
     /* Map new address */
-    DPRINTF("ISD: "TARGET_FMT_plx"@"TARGET_FMT_plx
-        " -> "TARGET_FMT_plx"@"TARGET_FMT_plx"\n",
-        s->ISD_length, s->ISD_start, length, start);
+    trace_gt64120_isd_remap(s->ISD_length, s->ISD_start, length, start);
     s->ISD_start = start;
     s->ISD_length = length;
     memory_region_add_subregion(get_system_memory(), s->ISD_start, &s->ISD_mem);
@@ -377,15 +377,16 @@ static const VMStateDescription vmstate_gt64120 = {
     }
 };
 
-static void gt64120_writel (void *opaque, hwaddr addr,
-                            uint64_t val, unsigned size)
+static void gt64120_writel(void *opaque, hwaddr addr,
+                           uint64_t val, unsigned size)
 {
     GT64120State *s = opaque;
     PCIHostState *phb = PCI_HOST_BRIDGE(s);
     uint32_t saddr;
 
-    if (!(s->regs[GT_CPU] & 0x00001000))
+    if (!(s->regs[GT_CPU] & 0x00001000)) {
         val = bswap32(val);
+    }
 
     saddr = (addr & 0xfff) >> 2;
     switch (saddr) {
@@ -458,12 +459,20 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_CPUERR_DATAHI:
     case GT_CPUERR_PARITY:
         /* Read-only registers, do nothing */
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gt64120: Read-only register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
 
     /* CPU Sync Barrier */
     case GT_PCI0SYNC:
     case GT_PCI1SYNC:
         /* Read-only registers, do nothing */
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gt64120: Read-only register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
 
     /* SDRAM and Device Address Decode */
@@ -502,7 +511,10 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_DEV_B3:
     case GT_DEV_BOOT:
         /* Not implemented */
-        DPRINTF ("Unimplemented device register offset 0x%x\n", saddr << 2);
+        qemu_log_mask(LOG_UNIMP,
+                      "gt64120: Unimplemented device register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
 
     /* ECC */
@@ -512,6 +524,10 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_ECC_CALC:
     case GT_ECC_ERRADDR:
         /* Read-only registers, do nothing */
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gt64120: Read-only register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
 
     /* DMA Record */
@@ -535,23 +551,20 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_DMA1_CUR:
     case GT_DMA2_CUR:
     case GT_DMA3_CUR:
-        /* Not implemented */
-        DPRINTF ("Unimplemented DMA register offset 0x%x\n", saddr << 2);
-        break;
 
     /* DMA Channel Control */
     case GT_DMA0_CTRL:
     case GT_DMA1_CTRL:
     case GT_DMA2_CTRL:
     case GT_DMA3_CTRL:
-        /* Not implemented */
-        DPRINTF ("Unimplemented DMA register offset 0x%x\n", saddr << 2);
-        break;
 
     /* DMA Arbiter */
     case GT_DMA_ARB:
         /* Not implemented */
-        DPRINTF ("Unimplemented DMA register offset 0x%x\n", saddr << 2);
+        qemu_log_mask(LOG_UNIMP,
+                      "gt64120: Unimplemented DMA register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
 
     /* Timer/Counter */
@@ -561,7 +574,10 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_TC3:
     case GT_TC_CONTROL:
         /* Not implemented */
-        DPRINTF ("Unimplemented timer register offset 0x%x\n", saddr << 2);
+        qemu_log_mask(LOG_UNIMP,
+                      "gt64120: Unimplemented timer register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
 
     /* PCI Internal */
@@ -602,6 +618,10 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_PCI1_CFGADDR:
     case GT_PCI1_CFGDATA:
         /* not implemented */
+        qemu_log_mask(LOG_UNIMP,
+                      "gt64120: Unimplemented timer register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
     case GT_PCI0_CFGADDR:
         phb->config_reg = val & 0x80fffffc;
@@ -620,19 +640,19 @@ static void gt64120_writel (void *opaque, hwaddr addr,
         /* not really implemented */
         s->regs[saddr] = ~(~(s->regs[saddr]) | ~(val & 0xfffffffe));
         s->regs[saddr] |= !!(s->regs[saddr] & 0xfffffffe);
-        DPRINTF("INTRCAUSE %" PRIx64 "\n", val);
+        trace_gt64120_write("INTRCAUSE", size << 1, val);
         break;
     case GT_INTRMASK:
         s->regs[saddr] = val & 0x3c3ffffe;
-        DPRINTF("INTRMASK %" PRIx64 "\n", val);
+        trace_gt64120_write("INTRMASK", size << 1, val);
         break;
     case GT_PCI0_ICMASK:
         s->regs[saddr] = val & 0x03fffffe;
-        DPRINTF("ICMASK %" PRIx64 "\n", val);
+        trace_gt64120_write("ICMASK", size << 1, val);
         break;
     case GT_PCI0_SERR0MASK:
         s->regs[saddr] = val & 0x0000003f;
-        DPRINTF("SERR0MASK %" PRIx64 "\n", val);
+        trace_gt64120_write("SERR0MASK", size << 1, val);
         break;
 
     /* Reserved when only PCI_0 is configured. */
@@ -650,19 +670,24 @@ static void gt64120_writel (void *opaque, hwaddr addr,
     case GT_SDRAM_B1:
     case GT_SDRAM_B2:
     case GT_SDRAM_B3:
-        /* We don't simulate electrical parameters of the SDRAM.
-           Accept, but ignore the values. */
+        /*
+         * We don't simulate electrical parameters of the SDRAM.
+         * Accept, but ignore the values.
+         */
         s->regs[saddr] = val;
         break;
 
     default:
-        DPRINTF ("Bad register offset 0x%x\n", (int)addr);
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gt64120: Illegal register write "
+                      "reg:0x03%x size:%u value:0x%0*" PRIx64 "\n",
+                      saddr << 2, size, size << 1, val);
         break;
     }
 }
 
-static uint64_t gt64120_readl (void *opaque,
-                               hwaddr addr, unsigned size)
+static uint64_t gt64120_readl(void *opaque,
+                              hwaddr addr, unsigned size)
 {
     GT64120State *s = opaque;
     PCIHostState *phb = PCI_HOST_BRIDGE(s);
@@ -674,8 +699,10 @@ static uint64_t gt64120_readl (void *opaque,
 
     /* CPU Configuration */
     case GT_MULTI:
-        /* Only one GT64xxx is present on the CPU bus, return
-           the initial value */
+        /*
+         * Only one GT64xxx is present on the CPU bus, return
+         * the initial value.
+         */
         val = s->regs[saddr];
         break;
 
@@ -685,17 +712,18 @@ static uint64_t gt64120_readl (void *opaque,
     case GT_CPUERR_DATALO:
     case GT_CPUERR_DATAHI:
     case GT_CPUERR_PARITY:
-        /* Emulated memory has no error, always return the initial
-           values */
+        /* Emulated memory has no error, always return the initial values. */
         val = s->regs[saddr];
         break;
 
     /* CPU Sync Barrier */
     case GT_PCI0SYNC:
     case GT_PCI1SYNC:
-        /* Reading those register should empty all FIFO on the PCI
-           bus, which are not emulated. The return value should be
-           a random value that should be ignored. */
+        /*
+         * Reading those register should empty all FIFO on the PCI
+         * bus, which are not emulated. The return value should be
+         * a random value that should be ignored.
+         */
         val = 0xc000ffee;
         break;
 
@@ -705,8 +733,7 @@ static uint64_t gt64120_readl (void *opaque,
     case GT_ECC_MEM:
     case GT_ECC_CALC:
     case GT_ECC_ERRADDR:
-        /* Emulated memory has no error, always return the initial
-           values */
+        /* Emulated memory has no error, always return the initial values. */
         val = s->regs[saddr];
         break;
 
@@ -785,8 +812,10 @@ static uint64_t gt64120_readl (void *opaque,
     case GT_SDRAM_B1:
     case GT_SDRAM_B2:
     case GT_SDRAM_B3:
-        /* We don't simulate electrical parameters of the SDRAM.
-           Just return the last written value. */
+        /*
+         * We don't simulate electrical parameters of the SDRAM.
+         * Just return the last written value.
+         */
         val = s->regs[saddr];
         break;
 
@@ -899,19 +928,19 @@ static uint64_t gt64120_readl (void *opaque,
     /* Interrupts */
     case GT_INTRCAUSE:
         val = s->regs[saddr];
-        DPRINTF("INTRCAUSE %x\n", val);
+        trace_gt64120_read("INTRCAUSE", size << 1, val);
         break;
     case GT_INTRMASK:
         val = s->regs[saddr];
-        DPRINTF("INTRMASK %x\n", val);
+        trace_gt64120_read("INTRMASK", size << 1, val);
         break;
     case GT_PCI0_ICMASK:
         val = s->regs[saddr];
-        DPRINTF("ICMASK %x\n", val);
+        trace_gt64120_read("ICMASK", size << 1, val);
         break;
     case GT_PCI0_SERR0MASK:
         val = s->regs[saddr];
-        DPRINTF("SERR0MASK %x\n", val);
+        trace_gt64120_read("SERR0MASK", size << 1, val);
         break;
 
     /* Reserved when only PCI_0 is configured. */
@@ -926,12 +955,16 @@ static uint64_t gt64120_readl (void *opaque,
 
     default:
         val = s->regs[saddr];
-        DPRINTF ("Bad register offset 0x%x\n", (int)addr);
+        qemu_log_mask(LOG_GUEST_ERROR,
+                      "gt64120: Illegal register read "
+                      "reg:0x03%x size:%u value:0x%0*x\n",
+                      saddr << 2, size, size << 1, val);
         break;
     }
 
-    if (!(s->regs[GT_CPU] & 0x00001000))
+    if (!(s->regs[GT_CPU] & 0x00001000)) {
         val = bswap32(val);
+    }
 
     return val;
 }
@@ -949,20 +982,20 @@ static int gt64120_pci_map_irq(PCIDevice *pci_dev, int irq_num)
     slot = (pci_dev->devfn >> 3);
 
     switch (slot) {
-      /* PIIX4 USB */
-      case 10:
+    /* PIIX4 USB */
+    case 10:
         return 3;
-      /* AMD 79C973 Ethernet */
-      case 11:
+    /* AMD 79C973 Ethernet */
+    case 11:
         return 1;
-      /* Crystal 4281 Sound */
-      case 12:
+    /* Crystal 4281 Sound */
+    case 12:
         return 2;
-      /* PCI slot 1 to 4 */
-      case 18 ... 21:
+    /* PCI slot 1 to 4 */
+    case 18 ... 21:
         return ((slot - 18) + irq_num) & 0x03;
-      /* Unknown device, don't do any translation */
-      default:
+    /* Unknown device, don't do any translation */
+    default:
         return irq_num;
     }
 }
@@ -980,12 +1013,12 @@ static void gt64120_pci_set_irq(void *opaque, int irq_num, int level)
     /* XXX: optimize */
     pic_irq = piix4_dev->config[0x60 + irq_num];
     if (pic_irq < 16) {
-        /* The pic level is the logical OR of all the PCI irqs mapped
-           to it */
+        /* The pic level is the logical OR of all the PCI irqs mapped to it. */
         pic_level = 0;
         for (i = 0; i < 4; i++) {
-            if (pic_irq == piix4_dev->config[0x60 + i])
+            if (pic_irq == piix4_dev->config[0x60 + i]) {
                 pic_level |= pci_irq_levels[i];
+            }
         }
         qemu_set_irq(pic[pic_irq], pic_level);
     }
@@ -1169,7 +1202,7 @@ PCIBus *gt64120_register(qemu_irq *pic)
     dev = qdev_create(NULL, TYPE_GT64120_PCI_HOST_BRIDGE);
     d = GT64120_PCI_HOST_BRIDGE(dev);
     phb = PCI_HOST_BRIDGE(dev);
-    memory_region_init(&d->pci0_mem, OBJECT(dev), "pci0-mem", UINT32_MAX);
+    memory_region_init(&d->pci0_mem, OBJECT(dev), "pci0-mem", 4 * GiB);
     address_space_init(&d->pci0_mem_as, &d->pci0_mem, "pci0-mem");
     phb->bus = pci_register_root_bus(dev, "pci",
                                      gt64120_pci_set_irq, gt64120_pci_map_irq,
@@ -1178,7 +1211,8 @@ PCIBus *gt64120_register(qemu_irq *pic)
                                      get_system_io(),
                                      PCI_DEVFN(18, 0), 4, TYPE_PCI_BUS);
     qdev_init_nofail(dev);
-    memory_region_init_io(&d->ISD_mem, OBJECT(dev), &isd_mem_ops, d, "isd-mem", 0x1000);
+    memory_region_init_io(&d->ISD_mem, OBJECT(dev), &isd_mem_ops, d,
+                          "isd-mem", 0x1000);
 
     pci_create_simple(phb->bus, PCI_DEVFN(0, 0), "gt64120_pci");
     return phb->bus;
diff --git a/hw/mips/trace-events b/hw/mips/trace-events
new file mode 100644
index 0000000000..75d4c73f2e
--- /dev/null
+++ b/hw/mips/trace-events
@@ -0,0 +1,4 @@
+# gt64xxx.c
+gt64120_read(const char *regname, int width, uint64_t value) "gt64120 read %s value:0x%0*" PRIx64
+gt64120_write(const char *regname, int width, uint64_t value) "gt64120 write %s value:0x%0*" PRIx64
+gt64120_isd_remap(uint64_t from_length, uint64_t from_addr, uint64_t to_length, uint64_t to_addr) "ISD: 0x%08" PRIx64 "@0x%08" PRIx64 " -> 0x%08" PRIx64 "@0x%08" PRIx64
diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs
index 77b9df9796..e9aab519a1 100644
--- a/hw/misc/Makefile.objs
+++ b/hw/misc/Makefile.objs
@@ -74,6 +74,7 @@ obj-$(CONFIG_ARMSSE_MHU) += armsse-mhu.o
 
 obj-$(CONFIG_PVPANIC) += pvpanic.o
 obj-$(CONFIG_AUX) += auxbus.o
+obj-$(CONFIG_ASPEED_SOC) += aspeed_xdma.o
 obj-$(CONFIG_ASPEED_SOC) += aspeed_scu.o aspeed_sdmc.o
 obj-$(CONFIG_MSF2) += msf2-sysreg.o
 obj-$(CONFIG_NRF51_SOC) += nrf51_rng.o
diff --git a/hw/misc/aspeed_xdma.c b/hw/misc/aspeed_xdma.c
new file mode 100644
index 0000000000..eebd4ad540
--- /dev/null
+++ b/hw/misc/aspeed_xdma.c
@@ -0,0 +1,165 @@
+/*
+ * ASPEED XDMA Controller
+ * Eddie James <eajames@linux.ibm.com>
+ *
+ * Copyright (C) 2019 IBM Corp
+ * SPDX-License-Identifer: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/log.h"
+#include "qemu/error-report.h"
+#include "hw/misc/aspeed_xdma.h"
+#include "qapi/error.h"
+
+#include "trace.h"
+
+#define XDMA_BMC_CMDQ_ADDR         0x10
+#define XDMA_BMC_CMDQ_ENDP         0x14
+#define XDMA_BMC_CMDQ_WRP          0x18
+#define  XDMA_BMC_CMDQ_W_MASK      0x0003FFFF
+#define XDMA_BMC_CMDQ_RDP          0x1C
+#define  XDMA_BMC_CMDQ_RDP_MAGIC   0xEE882266
+#define XDMA_IRQ_ENG_CTRL          0x20
+#define  XDMA_IRQ_ENG_CTRL_US_COMP BIT(4)
+#define  XDMA_IRQ_ENG_CTRL_DS_COMP BIT(5)
+#define  XDMA_IRQ_ENG_CTRL_W_MASK  0xBFEFF07F
+#define XDMA_IRQ_ENG_STAT          0x24
+#define  XDMA_IRQ_ENG_STAT_US_COMP BIT(4)
+#define  XDMA_IRQ_ENG_STAT_DS_COMP BIT(5)
+#define  XDMA_IRQ_ENG_STAT_RESET   0xF8000000
+#define XDMA_MEM_SIZE              0x1000
+
+#define TO_REG(addr) ((addr) / sizeof(uint32_t))
+
+static uint64_t aspeed_xdma_read(void *opaque, hwaddr addr, unsigned int size)
+{
+    uint32_t val = 0;
+    AspeedXDMAState *xdma = opaque;
+
+    if (addr < ASPEED_XDMA_REG_SIZE) {
+        val = xdma->regs[TO_REG(addr)];
+    }
+
+    return (uint64_t)val;
+}
+
+static void aspeed_xdma_write(void *opaque, hwaddr addr, uint64_t val,
+                              unsigned int size)
+{
+    unsigned int idx;
+    uint32_t val32 = (uint32_t)val;
+    AspeedXDMAState *xdma = opaque;
+
+    if (addr >= ASPEED_XDMA_REG_SIZE) {
+        return;
+    }
+
+    switch (addr) {
+    case XDMA_BMC_CMDQ_ENDP:
+        xdma->regs[TO_REG(addr)] = val32 & XDMA_BMC_CMDQ_W_MASK;
+        break;
+    case XDMA_BMC_CMDQ_WRP:
+        idx = TO_REG(addr);
+        xdma->regs[idx] = val32 & XDMA_BMC_CMDQ_W_MASK;
+        xdma->regs[TO_REG(XDMA_BMC_CMDQ_RDP)] = xdma->regs[idx];
+
+        trace_aspeed_xdma_write(addr, val);
+
+        if (xdma->bmc_cmdq_readp_set) {
+            xdma->bmc_cmdq_readp_set = 0;
+        } else {
+            xdma->regs[TO_REG(XDMA_IRQ_ENG_STAT)] |=
+                XDMA_IRQ_ENG_STAT_US_COMP | XDMA_IRQ_ENG_STAT_DS_COMP;
+
+            if (xdma->regs[TO_REG(XDMA_IRQ_ENG_CTRL)] &
+                (XDMA_IRQ_ENG_CTRL_US_COMP | XDMA_IRQ_ENG_CTRL_DS_COMP))
+                qemu_irq_raise(xdma->irq);
+        }
+        break;
+    case XDMA_BMC_CMDQ_RDP:
+        trace_aspeed_xdma_write(addr, val);
+
+        if (val32 == XDMA_BMC_CMDQ_RDP_MAGIC) {
+            xdma->bmc_cmdq_readp_set = 1;
+        }
+        break;
+    case XDMA_IRQ_ENG_CTRL:
+        xdma->regs[TO_REG(addr)] = val32 & XDMA_IRQ_ENG_CTRL_W_MASK;
+        break;
+    case XDMA_IRQ_ENG_STAT:
+        trace_aspeed_xdma_write(addr, val);
+
+        idx = TO_REG(addr);
+        if (val32 & (XDMA_IRQ_ENG_STAT_US_COMP | XDMA_IRQ_ENG_STAT_DS_COMP)) {
+            xdma->regs[idx] &=
+                ~(XDMA_IRQ_ENG_STAT_US_COMP | XDMA_IRQ_ENG_STAT_DS_COMP);
+            qemu_irq_lower(xdma->irq);
+        }
+        break;
+    default:
+        xdma->regs[TO_REG(addr)] = val32;
+        break;
+    }
+}
+
+static const MemoryRegionOps aspeed_xdma_ops = {
+    .read = aspeed_xdma_read,
+    .write = aspeed_xdma_write,
+    .endianness = DEVICE_NATIVE_ENDIAN,
+    .valid.min_access_size = 4,
+    .valid.max_access_size = 4,
+};
+
+static void aspeed_xdma_realize(DeviceState *dev, Error **errp)
+{
+    SysBusDevice *sbd = SYS_BUS_DEVICE(dev);
+    AspeedXDMAState *xdma = ASPEED_XDMA(dev);
+
+    sysbus_init_irq(sbd, &xdma->irq);
+    memory_region_init_io(&xdma->iomem, OBJECT(xdma), &aspeed_xdma_ops, xdma,
+                          TYPE_ASPEED_XDMA, XDMA_MEM_SIZE);
+    sysbus_init_mmio(sbd, &xdma->iomem);
+}
+
+static void aspeed_xdma_reset(DeviceState *dev)
+{
+    AspeedXDMAState *xdma = ASPEED_XDMA(dev);
+
+    xdma->bmc_cmdq_readp_set = 0;
+    memset(xdma->regs, 0, ASPEED_XDMA_REG_SIZE);
+    xdma->regs[TO_REG(XDMA_IRQ_ENG_STAT)] = XDMA_IRQ_ENG_STAT_RESET;
+
+    qemu_irq_lower(xdma->irq);
+}
+
+static const VMStateDescription aspeed_xdma_vmstate = {
+    .name = TYPE_ASPEED_XDMA,
+    .version_id = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT32_ARRAY(regs, AspeedXDMAState, ASPEED_XDMA_NUM_REGS),
+        VMSTATE_END_OF_LIST(),
+    },
+};
+
+static void aspeed_xdma_class_init(ObjectClass *classp, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(classp);
+
+    dc->realize = aspeed_xdma_realize;
+    dc->reset = aspeed_xdma_reset;
+    dc->vmsd = &aspeed_xdma_vmstate;
+}
+
+static const TypeInfo aspeed_xdma_info = {
+    .name          = TYPE_ASPEED_XDMA,
+    .parent        = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(AspeedXDMAState),
+    .class_init    = aspeed_xdma_class_init,
+};
+
+static void aspeed_xdma_register_type(void)
+{
+    type_register_static(&aspeed_xdma_info);
+}
+type_init(aspeed_xdma_register_type);
diff --git a/hw/misc/trace-events b/hw/misc/trace-events
index 47e1bccf71..c1ea1aa437 100644
--- a/hw/misc/trace-events
+++ b/hw/misc/trace-events
@@ -140,3 +140,6 @@ armsse_cpuid_write(uint64_t offset, uint64_t data, unsigned size) "SSE-200 CPU_I
 # armsse-mhu.c
 armsse_mhu_read(uint64_t offset, uint64_t data, unsigned size) "SSE-200 MHU read: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u"
 armsse_mhu_write(uint64_t offset, uint64_t data, unsigned size) "SSE-200 MHU write: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u"
+
+# aspeed_xdma.c
+aspeed_xdma_write(uint64_t offset, uint64_t data) "XDMA write: offset 0x%" PRIx64 " data 0x%" PRIx64
diff --git a/hw/net/ftgmac100.c b/hw/net/ftgmac100.c
index eb760472e5..d2cded5e94 100644
--- a/hw/net/ftgmac100.c
+++ b/hw/net/ftgmac100.c
@@ -1017,8 +1017,6 @@ static void ftgmac100_realize(DeviceState *dev, Error **errp)
     sysbus_init_irq(sbd, &s->irq);
     qemu_macaddr_default_if_unset(&s->conf.macaddr);
 
-    s->conf.peers.ncs[0] = nd_table[0].netdev;
-
     s->nic = qemu_new_nic(&net_ftgmac100_info, &s->conf,
                           object_get_typename(OBJECT(dev)), DEVICE(dev)->id,
                           s);
diff --git a/hw/net/sunhme.c b/hw/net/sunhme.c
index 1ebaee3c82..8b8603e696 100644
--- a/hw/net/sunhme.c
+++ b/hw/net/sunhme.c
@@ -44,6 +44,7 @@
 #define HME_SEBI_STAT                  0x100
 #define HME_SEBI_STAT_LINUXBUG         0x108
 #define HME_SEB_STAT_RXTOHOST          0x10000
+#define HME_SEB_STAT_NORXD             0x20000
 #define HME_SEB_STAT_MIFIRQ            0x800000
 #define HME_SEB_STAT_HOSTTOTX          0x1000000
 #define HME_SEB_STAT_TXALL             0x2000000
@@ -209,6 +210,8 @@ static void sunhme_update_irq(SunHMEState *s)
     }
 
     level = (seb ? 1 : 0);
+    trace_sunhme_update_irq(mifmask, mif, sebmask, seb, level);
+
     pci_set_irq(d, level);
 }
 
@@ -371,10 +374,20 @@ static void sunhme_mac_write(void *opaque, hwaddr addr,
                           uint64_t val, unsigned size)
 {
     SunHMEState *s = SUNHME(opaque);
+    uint64_t oldval = s->macregs[addr >> 2];
 
     trace_sunhme_mac_write(addr, val);
 
     s->macregs[addr >> 2] = val;
+
+    switch (addr) {
+    case HME_MACI_RXCFG:
+        if (!(oldval & HME_MAC_RXCFG_ENABLE) &&
+             (val & HME_MAC_RXCFG_ENABLE)) {
+            qemu_flush_queued_packets(qemu_get_queue(s->nic));
+        }
+        break;
+    }
 }
 
 static uint64_t sunhme_mac_read(void *opaque, hwaddr addr,
@@ -647,7 +660,7 @@ static int sunhme_can_receive(NetClientState *nc)
 {
     SunHMEState *s = qemu_get_nic_opaque(nc);
 
-    return s->macregs[HME_MAC_RXCFG_ENABLE >> 2] & HME_MAC_RXCFG_ENABLE;
+    return s->macregs[HME_MACI_RXCFG >> 2] & HME_MAC_RXCFG_ENABLE;
 }
 
 static void sunhme_link_status_changed(NetClientState *nc)
@@ -716,7 +729,7 @@ static ssize_t sunhme_receive(NetClientState *nc, const uint8_t *buf,
 
     /* Do nothing if MAC RX disabled */
     if (!(s->macregs[HME_MACI_RXCFG >> 2] & HME_MAC_RXCFG_ENABLE)) {
-        return -1;
+        return 0;
     }
 
     trace_sunhme_rx_filter_destmac(buf[0], buf[1], buf[2],
@@ -745,14 +758,14 @@ static ssize_t sunhme_receive(NetClientState *nc, const uint8_t *buf,
                 /* Didn't match hash filter */
                 trace_sunhme_rx_filter_hash_nomatch();
                 trace_sunhme_rx_filter_reject();
-                return 0;
+                return -1;
             } else {
                 trace_sunhme_rx_filter_hash_match();
             }
         } else {
             /* Not for us */
             trace_sunhme_rx_filter_reject();
-            return 0;
+            return -1;
         }
     } else {
         trace_sunhme_rx_filter_promisc_match();
@@ -775,6 +788,14 @@ static ssize_t sunhme_receive(NetClientState *nc, const uint8_t *buf,
     pci_dma_read(d, rb + cr * HME_DESC_SIZE, &status, 4);
     pci_dma_read(d, rb + cr * HME_DESC_SIZE + 4, &buffer, 4);
 
+    /* If we don't own the current descriptor then indicate overflow error */
+    if (!(status & HME_XD_OWN)) {
+        s->sebregs[HME_SEBI_STAT >> 2] |= HME_SEB_STAT_NORXD;
+        sunhme_update_irq(s);
+        trace_sunhme_rx_norxd();
+        return -1;
+    }
+
     rxoffset = (s->erxregs[HME_ERXI_CFG >> 2] & HME_ERX_CFG_BYTEOFFSET) >>
                 HME_ERX_CFG_BYTEOFFSET_SHIFT;
 
diff --git a/hw/net/trace-events b/hw/net/trace-events
index 3cd9e122df..58665655cc 100644
--- a/hw/net/trace-events
+++ b/hw/net/trace-events
@@ -359,6 +359,8 @@ sunhme_rx_filter_reject(void) "rejecting incoming frame"
 sunhme_rx_filter_accept(void) "accepting incoming frame"
 sunhme_rx_desc(uint32_t addr, int offset, uint32_t status, int len, int cr, int nr) "addr 0x%"PRIx32"(+0x%x) status 0x%"PRIx32 " len %d (ring %d/%d)"
 sunhme_rx_xsum_calc(uint16_t xsum) "calculated incoming xsum as 0x%x"
+sunhme_rx_norxd(void) "no free rx descriptors available"
+sunhme_update_irq(uint32_t mifmask, uint32_t mif, uint32_t sebmask, uint32_t seb, int level) "mifmask: 0x%x  mif: 0x%x  sebmask: 0x%x  seb: 0x%x  level: %d"
 
 # virtio-net.c
 virtio_net_announce_notify(void) ""
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index c3f5fccfd1..b9e1cd71cf 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -2360,7 +2360,7 @@ static int virtio_net_post_load_device(void *opaque, int version_id)
             timer_mod(n->announce_timer.tm,
                       qemu_clock_get_ms(n->announce_timer.type));
         } else {
-            qemu_announce_timer_del(&n->announce_timer);
+            qemu_announce_timer_del(&n->announce_timer, false);
         }
     }
 
@@ -2784,7 +2784,7 @@ static void virtio_net_device_unrealize(DeviceState *dev, Error **errp)
         virtio_net_del_queue(n, i);
     }
 
-    qemu_announce_timer_del(&n->announce_timer);
+    qemu_announce_timer_del(&n->announce_timer, false);
     g_free(n->vqs);
     qemu_del_nic(n->nic);
     virtio_net_rsc_cleanup(n);
diff --git a/hw/net/xen_nic.c b/hw/net/xen_nic.c
index 37cda8e4be..ffb3b5898d 100644
--- a/hw/net/xen_nic.c
+++ b/hw/net/xen_nic.c
@@ -30,7 +30,7 @@
 #include "net/util.h"
 #include "hw/xen/xen-legacy-backend.h"
 
-#include <xen/io/netif.h>
+#include "hw/xen/interface/io/netif.h"
 
 /* ------------------------------------------------------------- */
 
diff --git a/hw/pci-host/designware.c b/hw/pci-host/designware.c
index 0fdfff5784..9ae8c0deb7 100644
--- a/hw/pci-host/designware.c
+++ b/hw/pci-host/designware.c
@@ -51,6 +51,8 @@
 #define DESIGNWARE_PCIE_ATU_DEVFN(x)               (((x) >> 16) & 0xff)
 #define DESIGNWARE_PCIE_ATU_UPPER_TARGET           0x91C
 
+#define DESIGNWARE_PCIE_IRQ_MSI                    3
+
 static DesignwarePCIEHost *
 designware_pcie_root_to_host(DesignwarePCIERoot *root)
 {
@@ -67,7 +69,7 @@ static void designware_pcie_root_msi_write(void *opaque, hwaddr addr,
     root->msi.intr[0].status |= BIT(val) & root->msi.intr[0].enable;
 
     if (root->msi.intr[0].status & ~root->msi.intr[0].mask) {
-        qemu_set_irq(host->pci.irqs[0], 1);
+        qemu_set_irq(host->pci.irqs[DESIGNWARE_PCIE_IRQ_MSI], 1);
     }
 }
 
@@ -290,23 +292,19 @@ static void designware_pcie_root_config_write(PCIDevice *d, uint32_t address,
     case DESIGNWARE_PCIE_MSI_ADDR_LO:
         root->msi.base &= 0xFFFFFFFF00000000ULL;
         root->msi.base |= val;
+        designware_pcie_root_update_msi_mapping(root);
         break;
 
     case DESIGNWARE_PCIE_MSI_ADDR_HI:
         root->msi.base &= 0x00000000FFFFFFFFULL;
         root->msi.base |= (uint64_t)val << 32;
+        designware_pcie_root_update_msi_mapping(root);
         break;
 
-    case DESIGNWARE_PCIE_MSI_INTR0_ENABLE: {
-        const bool update_msi_mapping = !root->msi.intr[0].enable ^ !!val;
-
+    case DESIGNWARE_PCIE_MSI_INTR0_ENABLE:
         root->msi.intr[0].enable = val;
-
-        if (update_msi_mapping) {
-            designware_pcie_root_update_msi_mapping(root);
-        }
+        designware_pcie_root_update_msi_mapping(root);
         break;
-    }
 
     case DESIGNWARE_PCIE_MSI_INTR0_MASK:
         root->msi.intr[0].mask = val;
@@ -315,7 +313,7 @@ static void designware_pcie_root_config_write(PCIDevice *d, uint32_t address,
     case DESIGNWARE_PCIE_MSI_INTR0_STATUS:
         root->msi.intr[0].status ^= val;
         if (!root->msi.intr[0].status) {
-            qemu_set_irq(host->pci.irqs[0], 0);
+            qemu_set_irq(host->pci.irqs[DESIGNWARE_PCIE_IRQ_MSI], 0);
         }
         break;
 
diff --git a/hw/ppc/mac_newworld.c b/hw/ppc/mac_newworld.c
index 4d835f32b5..c8d3245524 100644
--- a/hw/ppc/mac_newworld.c
+++ b/hw/ppc/mac_newworld.c
@@ -437,13 +437,11 @@ static void ppc_core99_init(MachineState *machine)
     }
 
     /* The NewWorld NVRAM is not located in the MacIO device */
-#ifdef CONFIG_KVM
     if (kvm_enabled() && getpagesize() > 4096) {
         /* We can't combine read-write and read-only in a single page, so
            move the NVRAM out of ROM again for KVM */
         nvram_addr = 0xFFE00000;
     }
-#endif
     dev = qdev_create(NULL, TYPE_MACIO_NVRAM);
     qdev_prop_set_uint32(dev, "size", 0x2000);
     qdev_prop_set_uint32(dev, "it_shift", 1);
@@ -488,14 +486,12 @@ static void ppc_core99_init(MachineState *machine)
 
     fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_IS_KVM, kvm_enabled());
     if (kvm_enabled()) {
-#ifdef CONFIG_KVM
         uint8_t *hypercall;
 
         hypercall = g_malloc(16);
         kvmppc_get_hypercall(env, hypercall, 16);
         fw_cfg_add_bytes(fw_cfg, FW_CFG_PPC_KVM_HC, hypercall, 16);
         fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_KVM_PID, getpid());
-#endif
     }
     fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_TBFREQ, tbfreq);
     /* Mac OS X requires a "known good" clock-frequency value; pass it one. */
diff --git a/hw/ppc/mac_oldworld.c b/hw/ppc/mac_oldworld.c
index eddd005a7c..da751addc4 100644
--- a/hw/ppc/mac_oldworld.c
+++ b/hw/ppc/mac_oldworld.c
@@ -345,14 +345,12 @@ static void ppc_heathrow_init(MachineState *machine)
 
     fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_IS_KVM, kvm_enabled());
     if (kvm_enabled()) {
-#ifdef CONFIG_KVM
         uint8_t *hypercall;
 
         hypercall = g_malloc(16);
         kvmppc_get_hypercall(env, hypercall, 16);
         fw_cfg_add_bytes(fw_cfg, FW_CFG_PPC_KVM_HC, hypercall, 16);
         fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_KVM_PID, getpid());
-#endif
     }
     fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_TBFREQ, tbfreq);
     /* Mac OS X requires a "known good" clock-frequency value; pass it one. */
diff --git a/hw/ppc/pnv.c b/hw/ppc/pnv.c
index 9db43916ac..b87e01e5b9 100644
--- a/hw/ppc/pnv.c
+++ b/hw/ppc/pnv.c
@@ -860,6 +860,14 @@ static void pnv_chip_power8_realize(DeviceState *dev, Error **errp)
     Pnv8Psi *psi8 = &chip8->psi;
     Error *local_err = NULL;
 
+    /* XSCOM bridge is first */
+    pnv_xscom_realize(chip, PNV_XSCOM_SIZE, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+    sysbus_mmio_map(SYS_BUS_DEVICE(chip), 0, PNV_XSCOM_BASE(chip));
+
     pcc->parent_realize(dev, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
@@ -916,7 +924,6 @@ static void pnv_chip_power8e_class_init(ObjectClass *klass, void *data)
     k->isa_create = pnv_chip_power8_isa_create;
     k->dt_populate = pnv_chip_power8_dt_populate;
     k->pic_print_info = pnv_chip_power8_pic_print_info;
-    k->xscom_base = 0x003fc0000000000ull;
     dc->desc = "PowerNV Chip POWER8E";
 
     device_class_set_parent_realize(dc, pnv_chip_power8_realize,
@@ -936,7 +943,6 @@ static void pnv_chip_power8_class_init(ObjectClass *klass, void *data)
     k->isa_create = pnv_chip_power8_isa_create;
     k->dt_populate = pnv_chip_power8_dt_populate;
     k->pic_print_info = pnv_chip_power8_pic_print_info;
-    k->xscom_base = 0x003fc0000000000ull;
     dc->desc = "PowerNV Chip POWER8";
 
     device_class_set_parent_realize(dc, pnv_chip_power8_realize,
@@ -956,7 +962,6 @@ static void pnv_chip_power8nvl_class_init(ObjectClass *klass, void *data)
     k->isa_create = pnv_chip_power8nvl_isa_create;
     k->dt_populate = pnv_chip_power8_dt_populate;
     k->pic_print_info = pnv_chip_power8_pic_print_info;
-    k->xscom_base = 0x003fc0000000000ull;
     dc->desc = "PowerNV Chip POWER8NVL";
 
     device_class_set_parent_realize(dc, pnv_chip_power8_realize,
@@ -1024,6 +1029,14 @@ static void pnv_chip_power9_realize(DeviceState *dev, Error **errp)
     Pnv9Psi *psi9 = &chip9->psi;
     Error *local_err = NULL;
 
+    /* XSCOM bridge is first */
+    pnv_xscom_realize(chip, PNV9_XSCOM_SIZE, &local_err);
+    if (local_err) {
+        error_propagate(errp, local_err);
+        return;
+    }
+    sysbus_mmio_map(SYS_BUS_DEVICE(chip), 0, PNV9_XSCOM_BASE(chip));
+
     pcc->parent_realize(dev, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
@@ -1099,7 +1112,6 @@ static void pnv_chip_power9_class_init(ObjectClass *klass, void *data)
     k->isa_create = pnv_chip_power9_isa_create;
     k->dt_populate = pnv_chip_power9_dt_populate;
     k->pic_print_info = pnv_chip_power9_pic_print_info;
-    k->xscom_base = 0x00603fc00000000ull;
     dc->desc = "PowerNV Chip POWER9";
 
     device_class_set_parent_realize(dc, pnv_chip_power9_realize,
@@ -1136,11 +1148,6 @@ static void pnv_chip_core_sanitize(PnvChip *chip, Error **errp)
     }
 }
 
-static void pnv_chip_instance_init(Object *obj)
-{
-    PNV_CHIP(obj)->xscom_base = PNV_CHIP_GET_CLASS(obj)->xscom_base;
-}
-
 static void pnv_chip_core_realize(PnvChip *chip, Error **errp)
 {
     Error *error = NULL;
@@ -1206,14 +1213,6 @@ static void pnv_chip_realize(DeviceState *dev, Error **errp)
     PnvChip *chip = PNV_CHIP(dev);
     Error *error = NULL;
 
-    /* XSCOM bridge */
-    pnv_xscom_realize(chip, &error);
-    if (error) {
-        error_propagate(errp, error);
-        return;
-    }
-    sysbus_mmio_map(SYS_BUS_DEVICE(chip), 0, PNV_XSCOM_BASE(chip));
-
     /* Cores */
     pnv_chip_core_realize(chip, &error);
     if (error) {
@@ -1398,7 +1397,6 @@ static const TypeInfo types[] = {
         .name          = TYPE_PNV_CHIP,
         .parent        = TYPE_SYS_BUS_DEVICE,
         .class_init    = pnv_chip_class_init,
-        .instance_init = pnv_chip_instance_init,
         .instance_size = sizeof(PnvChip),
         .class_size    = sizeof(PnvChipClass),
         .abstract      = true,
diff --git a/hw/ppc/pnv_xscom.c b/hw/ppc/pnv_xscom.c
index 4e52885c9e..2b81c75f56 100644
--- a/hw/ppc/pnv_xscom.c
+++ b/hw/ppc/pnv_xscom.c
@@ -213,17 +213,17 @@ const MemoryRegionOps pnv_xscom_ops = {
     .endianness = DEVICE_BIG_ENDIAN,
 };
 
-void pnv_xscom_realize(PnvChip *chip, Error **errp)
+void pnv_xscom_realize(PnvChip *chip, uint64_t size, Error **errp)
 {
     SysBusDevice *sbd = SYS_BUS_DEVICE(chip);
     char *name;
 
     name = g_strdup_printf("xscom-%x", chip->chip_id);
     memory_region_init_io(&chip->xscom_mmio, OBJECT(chip), &pnv_xscom_ops,
-                          chip, name, PNV_XSCOM_SIZE);
+                          chip, name, size);
     sysbus_init_mmio(sbd, &chip->xscom_mmio);
 
-    memory_region_init(&chip->xscom, OBJECT(chip), name, PNV_XSCOM_SIZE);
+    memory_region_init(&chip->xscom, OBJECT(chip), name, size);
     address_space_init(&chip->xscom_as, &chip->xscom, name);
     g_free(name);
 }
@@ -265,12 +265,19 @@ static const char compat_p9[] = "ibm,power9-xscom\0ibm,xscom";
 
 int pnv_dt_xscom(PnvChip *chip, void *fdt, int root_offset)
 {
-    uint64_t reg[] = { cpu_to_be64(PNV_XSCOM_BASE(chip)),
-                       cpu_to_be64(PNV_XSCOM_SIZE) };
+    uint64_t reg[2];
     int xscom_offset;
     ForeachPopulateArgs args;
     char *name;
 
+    if (pnv_chip_is_power9(chip)) {
+        reg[0] = cpu_to_be64(PNV9_XSCOM_BASE(chip));
+        reg[1] = cpu_to_be64(PNV9_XSCOM_SIZE);
+    } else {
+        reg[0] = cpu_to_be64(PNV_XSCOM_BASE(chip));
+        reg[1] = cpu_to_be64(PNV_XSCOM_SIZE);
+    }
+
     name = g_strdup_printf("xscom@%" PRIx64, be64_to_cpu(reg[0]));
     xscom_offset = fdt_add_subnode(fdt, root_offset, name);
     _FDT(xscom_offset);
diff --git a/hw/ppc/ppc.c b/hw/ppc/ppc.c
index 9d91e8481b..a9e508c496 100644
--- a/hw/ppc/ppc.c
+++ b/hw/ppc/ppc.c
@@ -80,9 +80,7 @@ void ppc_set_irq(PowerPCCPU *cpu, int n_IRQ, int level)
     }
 
     if (old_pending != env->pending_interrupts) {
-#ifdef CONFIG_KVM
         kvmppc_set_interrupt(cpu, n_IRQ, level);
-#endif
     }
 
 
@@ -1036,10 +1034,7 @@ static void timebase_load(PPCTimebase *tb)
     CPU_FOREACH(cpu) {
         PowerPCCPU *pcpu = POWERPC_CPU(cpu);
         pcpu->env.tb_env->tb_offset = tb_off_adj;
-#if defined(CONFIG_KVM)
-        kvm_set_one_reg(cpu, KVM_REG_PPC_TB_OFFSET,
-                        &pcpu->env.tb_env->tb_offset);
-#endif
+        kvmppc_set_reg_tb_offset(pcpu, pcpu->env.tb_env->tb_offset);
     }
 }
 
diff --git a/hw/ppc/prep.c b/hw/ppc/prep.c
index 2a8009e20b..a248ce480d 100644
--- a/hw/ppc/prep.c
+++ b/hw/ppc/prep.c
@@ -780,7 +780,6 @@ static void ibm_40p_init(MachineState *machine)
 
     fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_IS_KVM, kvm_enabled());
     if (kvm_enabled()) {
-#ifdef CONFIG_KVM
         uint8_t *hypercall;
 
         fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_TBFREQ, kvmppc_get_tbfreq());
@@ -788,7 +787,6 @@ static void ibm_40p_init(MachineState *machine)
         kvmppc_get_hypercall(env, hypercall, 16);
         fw_cfg_add_bytes(fw_cfg, FW_CFG_PPC_KVM_HC, hypercall, 16);
         fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_KVM_PID, getpid());
-#endif
     } else {
         fw_cfg_add_i32(fw_cfg, FW_CFG_PPC_TBFREQ, NANOSECONDS_PER_SECOND);
     }
diff --git a/hw/ppc/spapr_irq.c b/hw/ppc/spapr_irq.c
index 3156daf093..ff3df0bbd8 100644
--- a/hw/ppc/spapr_irq.c
+++ b/hw/ppc/spapr_irq.c
@@ -62,7 +62,7 @@ void spapr_irq_msi_reset(SpaprMachineState *spapr)
     bitmap_clear(spapr->irq_map, 0, spapr->irq_map_nr);
 }
 
-static void spapr_irq_init_device(SpaprMachineState *spapr,
+static void spapr_irq_init_kvm(SpaprMachineState *spapr,
                                   SpaprIrq *irq, Error **errp)
 {
     MachineState *machine = MACHINE(spapr);
@@ -88,8 +88,6 @@ static void spapr_irq_init_device(SpaprMachineState *spapr,
         error_prepend(&local_err, "kernel_irqchip allowed but unavailable: ");
         warn_report_err(local_err);
     }
-
-    irq->init_emu(spapr, errp);
 }
 
 /*
@@ -114,6 +112,8 @@ static void spapr_irq_init_xics(SpaprMachineState *spapr, int nr_irqs,
     }
 
     spapr->ics = ICS_BASE(obj);
+
+    xics_spapr_init(spapr);
 }
 
 #define ICS_IRQ_FREE(ics, srcno)   \
@@ -222,7 +222,7 @@ static void spapr_irq_reset_xics(SpaprMachineState *spapr, Error **errp)
 {
     Error *local_err = NULL;
 
-    spapr_irq_init_device(spapr, &spapr_irq_xics, &local_err);
+    spapr_irq_init_kvm(spapr, &spapr_irq_xics, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         return;
@@ -234,15 +234,10 @@ static const char *spapr_irq_get_nodename_xics(SpaprMachineState *spapr)
     return XICS_NODENAME;
 }
 
-static void spapr_irq_init_emu_xics(SpaprMachineState *spapr, Error **errp)
-{
-    xics_spapr_init(spapr);
-}
-
 static void spapr_irq_init_kvm_xics(SpaprMachineState *spapr, Error **errp)
 {
     if (kvm_enabled()) {
-        xics_kvm_init(spapr, errp);
+        xics_kvm_connect(spapr, errp);
     }
 }
 
@@ -266,7 +261,6 @@ SpaprIrq spapr_irq_xics = {
     .reset       = spapr_irq_reset_xics,
     .set_irq     = spapr_irq_set_irq_xics,
     .get_nodename = spapr_irq_get_nodename_xics,
-    .init_emu    = spapr_irq_init_emu_xics,
     .init_kvm    = spapr_irq_init_kvm_xics,
 };
 
@@ -384,7 +378,7 @@ static void spapr_irq_reset_xive(SpaprMachineState *spapr, Error **errp)
         spapr_xive_set_tctx_os_cam(spapr_cpu_state(cpu)->tctx);
     }
 
-    spapr_irq_init_device(spapr, &spapr_irq_xive, &local_err);
+    spapr_irq_init_kvm(spapr, &spapr_irq_xive, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
         return;
@@ -410,11 +404,6 @@ static const char *spapr_irq_get_nodename_xive(SpaprMachineState *spapr)
     return spapr->xive->nodename;
 }
 
-static void spapr_irq_init_emu_xive(SpaprMachineState *spapr, Error **errp)
-{
-    spapr_xive_init(spapr->xive, errp);
-}
-
 static void spapr_irq_init_kvm_xive(SpaprMachineState *spapr, Error **errp)
 {
     if (kvm_enabled()) {
@@ -446,7 +435,6 @@ SpaprIrq spapr_irq_xive = {
     .reset       = spapr_irq_reset_xive,
     .set_irq     = spapr_irq_set_irq_xive,
     .get_nodename = spapr_irq_get_nodename_xive,
-    .init_emu    = spapr_irq_init_emu_xive,
     .init_kvm    = spapr_irq_init_kvm_xive,
 };
 
@@ -624,7 +612,6 @@ SpaprIrq spapr_irq_dual = {
     .reset       = spapr_irq_reset_dual,
     .set_irq     = spapr_irq_set_irq_dual,
     .get_nodename = spapr_irq_get_nodename_dual,
-    .init_emu    = NULL, /* should not be used */
     .init_kvm    = NULL, /* should not be used */
 };
 
@@ -668,6 +655,19 @@ static void spapr_irq_check(SpaprMachineState *spapr, Error **errp)
             return;
         }
     }
+
+    /*
+     * On a POWER9 host, some older KVM XICS devices cannot be destroyed and
+     * re-created. Detect that early to avoid QEMU to exit later when the
+     * guest reboots.
+     */
+    if (kvm_enabled() &&
+        spapr->irq == &spapr_irq_dual &&
+        machine_kernel_irqchip_required(machine) &&
+        xics_kvm_has_broken_disconnect(spapr)) {
+        error_setg(errp, "KVM is too old to support ic-mode=dual,kernel-irqchip=on");
+        return;
+    }
 }
 
 /*
@@ -827,6 +827,5 @@ SpaprIrq spapr_irq_xics_legacy = {
     .reset       = spapr_irq_reset_xics,
     .set_irq     = spapr_irq_set_irq_xics,
     .get_nodename = spapr_irq_get_nodename_xics,
-    .init_emu    = spapr_irq_init_emu_xics,
     .init_kvm    = spapr_irq_init_kvm_xics,
 };
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 957ae88bbd..9003fe9010 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -1343,6 +1343,7 @@ static void spapr_dt_pci_device_cb(PCIBus *bus, PCIDevice *pdev,
 static int spapr_dt_pci_bus(SpaprPhbState *sphb, PCIBus *bus,
                                void *fdt, int offset)
 {
+    Object *owner;
     PciWalkFdt cbinfo = {
         .fdt = fdt,
         .offset = offset,
@@ -1356,15 +1357,20 @@ static int spapr_dt_pci_bus(SpaprPhbState *sphb, PCIBus *bus,
     _FDT(fdt_setprop_cell(fdt, offset, "#size-cells",
                           RESOURCE_CELLS_SIZE));
 
-    if (bus) {
-        pci_for_each_device_reverse(bus, pci_bus_num(bus),
-                                    spapr_dt_pci_device_cb, &cbinfo);
-        if (cbinfo.err) {
-            return cbinfo.err;
-        }
+    assert(bus);
+    pci_for_each_device_reverse(bus, pci_bus_num(bus),
+                                spapr_dt_pci_device_cb, &cbinfo);
+    if (cbinfo.err) {
+        return cbinfo.err;
     }
 
-    ret = spapr_dt_drc(fdt, offset, OBJECT(bus->parent_dev),
+    if (pci_bus_is_root(bus)) {
+        owner = OBJECT(sphb);
+    } else {
+        owner = OBJECT(pci_bridge_get_device(bus));
+    }
+
+    ret = spapr_dt_drc(fdt, offset, owner,
                        SPAPR_DR_CONNECTOR_TYPE_PCI);
     if (ret) {
         return ret;
@@ -1782,6 +1788,12 @@ static void spapr_phb_unrealize(DeviceState *dev, Error **errp)
 
     memory_region_del_subregion(&sphb->iommu_root, &sphb->msiwindow);
 
+    /*
+     * An attached PCI device may have memory listeners, eg. VFIO PCI. We have
+     * unmapped all sections. Remove the listeners now, before destroying the
+     * address space.
+     */
+    address_space_remove_listeners(&sphb->iommu_as);
     address_space_destroy(&sphb->iommu_as);
 
     qbus_set_hotplug_handler(BUS(phb->bus), NULL, &error_abort);
@@ -1945,11 +1957,9 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
      * For KVM we want to ensure that this memory is a full page so that
      * our memory slot is of page size granularity.
      */
-#ifdef CONFIG_KVM
     if (kvm_enabled()) {
         msi_window_size = getpagesize();
     }
-#endif
 
     memory_region_init_io(&sphb->msiwindow, OBJECT(sphb), &spapr_msi_ops, spapr,
                           "msi", msi_window_size);
diff --git a/hw/ppc/spapr_rtc.c b/hw/ppc/spapr_rtc.c
index af1ef30a53..6cf0113b34 100644
--- a/hw/ppc/spapr_rtc.c
+++ b/hw/ppc/spapr_rtc.c
@@ -32,7 +32,7 @@
 #include "sysemu/sysemu.h"
 #include "hw/ppc/spapr.h"
 #include "qapi/error.h"
-#include "qapi/qapi-events-target.h"
+#include "qapi/qapi-events-misc-target.h"
 #include "qemu/cutils.h"
 #include "qemu/module.h"
 
diff --git a/hw/riscv/Makefile.objs b/hw/riscv/Makefile.objs
index a65027304a..eb9d4f9ffc 100644
--- a/hw/riscv/Makefile.objs
+++ b/hw/riscv/Makefile.objs
@@ -1,3 +1,4 @@
+obj-y += boot.o
 obj-$(CONFIG_SPIKE) += riscv_htif.o
 obj-$(CONFIG_HART) += riscv_hart.o
 obj-$(CONFIG_SIFIVE_E) += sifive_e.o
diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
new file mode 100644
index 0000000000..ff023f42d0
--- /dev/null
+++ b/hw/riscv/boot.c
@@ -0,0 +1,105 @@
+/*
+ * QEMU RISC-V Boot Helper
+ *
+ * Copyright (c) 2017 SiFive, Inc.
+ * Copyright (c) 2019 Alistair Francis <alistair.francis@wdc.com>
+ *
+ * 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 "qemu/osdep.h"
+#include "qemu/units.h"
+#include "qemu/error-report.h"
+#include "exec/cpu-defs.h"
+#include "hw/loader.h"
+#include "hw/riscv/boot.h"
+#include "hw/boards.h"
+#include "elf.h"
+
+#if defined(TARGET_RISCV32)
+# define KERNEL_BOOT_ADDRESS 0x80400000
+#else
+# define KERNEL_BOOT_ADDRESS 0x80200000
+#endif
+
+target_ulong riscv_load_firmware(const char *firmware_filename,
+                                 hwaddr firmware_load_addr)
+{
+    uint64_t firmware_entry, firmware_start, firmware_end;
+
+    if (load_elf(firmware_filename, NULL, NULL, NULL, &firmware_entry,
+                 &firmware_start, &firmware_end, 0, EM_RISCV, 1, 0) > 0) {
+        return firmware_entry;
+    }
+
+    if (load_image_targphys_as(firmware_filename, firmware_load_addr,
+                               ram_size, NULL) > 0) {
+        return firmware_load_addr;
+    }
+
+    error_report("could not load firmware '%s'", firmware_filename);
+    exit(1);
+}
+
+target_ulong riscv_load_kernel(const char *kernel_filename)
+{
+    uint64_t kernel_entry, kernel_high;
+
+    if (load_elf(kernel_filename, NULL, NULL, NULL,
+                 &kernel_entry, NULL, &kernel_high, 0, EM_RISCV, 1, 0) > 0) {
+        return kernel_entry;
+    }
+
+    if (load_uimage_as(kernel_filename, &kernel_entry, NULL, NULL,
+                       NULL, NULL, NULL) > 0) {
+        return kernel_entry;
+    }
+
+    if (load_image_targphys_as(kernel_filename, KERNEL_BOOT_ADDRESS,
+                               ram_size, NULL) > 0) {
+        return KERNEL_BOOT_ADDRESS;
+    }
+
+    error_report("could not load kernel '%s'", kernel_filename);
+    exit(1);
+}
+
+hwaddr riscv_load_initrd(const char *filename, uint64_t mem_size,
+                         uint64_t kernel_entry, hwaddr *start)
+{
+    int size;
+
+    /*
+     * We want to put the initrd far enough into RAM that when the
+     * kernel is uncompressed it will not clobber the initrd. However
+     * on boards without much RAM we must ensure that we still leave
+     * enough room for a decent sized initrd, and on boards with large
+     * amounts of RAM we must avoid the initrd being so far up in RAM
+     * that it is outside lowmem and inaccessible to the kernel.
+     * So for boards with less  than 256MB of RAM we put the initrd
+     * halfway into RAM, and for boards with 256MB of RAM or more we put
+     * the initrd at 128MB.
+     */
+    *start = kernel_entry + MIN(mem_size / 2, 128 * MiB);
+
+    size = load_ramdisk(filename, *start, mem_size - *start);
+    if (size == -1) {
+        size = load_image_targphys(filename, *start, mem_size - *start);
+        if (size == -1) {
+            error_report("could not load ramdisk '%s'", filename);
+            exit(1);
+        }
+    }
+
+    return *start + size;
+}
diff --git a/hw/riscv/sifive_e.c b/hw/riscv/sifive_e.c
index 80ac56fa7d..d27f626529 100644
--- a/hw/riscv/sifive_e.c
+++ b/hw/riscv/sifive_e.c
@@ -44,10 +44,10 @@
 #include "hw/riscv/sifive_prci.h"
 #include "hw/riscv/sifive_uart.h"
 #include "hw/riscv/sifive_e.h"
+#include "hw/riscv/boot.h"
 #include "chardev/char.h"
 #include "sysemu/arch_init.h"
 #include "exec/address-spaces.h"
-#include "elf.h"
 
 static const struct MemmapEntry {
     hwaddr base;
@@ -74,19 +74,6 @@ static const struct MemmapEntry {
     [SIFIVE_E_DTIM] =     { 0x80000000,     0x4000 }
 };
 
-static target_ulong load_kernel(const char *kernel_filename)
-{
-    uint64_t kernel_entry, kernel_high;
-
-    if (load_elf(kernel_filename, NULL, NULL, NULL,
-                 &kernel_entry, NULL, &kernel_high,
-                 0, EM_RISCV, 1, 0) < 0) {
-        error_report("could not load kernel '%s'", kernel_filename);
-        exit(1);
-    }
-    return kernel_entry;
-}
-
 static void sifive_mmio_emulate(MemoryRegion *parent, const char *name,
                              uintptr_t offset, uintptr_t length)
 {
@@ -131,7 +118,7 @@ static void riscv_sifive_e_init(MachineState *machine)
                           memmap[SIFIVE_E_MROM].base, &address_space_memory);
 
     if (machine->kernel_filename) {
-        load_kernel(machine->kernel_filename);
+        riscv_load_kernel(machine->kernel_filename);
     }
 }
 
@@ -158,17 +145,15 @@ static void riscv_sifive_e_soc_realize(DeviceState *dev, Error **errp)
 
     SiFiveESoCState *s = RISCV_E_SOC(dev);
     MemoryRegion *sys_mem = get_system_memory();
-    MemoryRegion *xip_mem = g_new(MemoryRegion, 1);
-    MemoryRegion *mask_rom = g_new(MemoryRegion, 1);
 
     object_property_set_bool(OBJECT(&s->cpus), true, "realized",
                             &error_abort);
 
     /* Mask ROM */
-    memory_region_init_rom(mask_rom, NULL, "riscv.sifive.e.mrom",
+    memory_region_init_rom(&s->mask_rom, NULL, "riscv.sifive.e.mrom",
         memmap[SIFIVE_E_MROM].size, &error_fatal);
     memory_region_add_subregion(sys_mem,
-        memmap[SIFIVE_E_MROM].base, mask_rom);
+        memmap[SIFIVE_E_MROM].base, &s->mask_rom);
 
     /* MMIO */
     s->plic = sifive_plic_create(memmap[SIFIVE_E_PLIC].base,
@@ -228,10 +213,11 @@ static void riscv_sifive_e_soc_realize(DeviceState *dev, Error **errp)
         memmap[SIFIVE_E_PWM2].base, memmap[SIFIVE_E_PWM2].size);
 
     /* Flash memory */
-    memory_region_init_ram(xip_mem, NULL, "riscv.sifive.e.xip",
+    memory_region_init_ram(&s->xip_mem, NULL, "riscv.sifive.e.xip",
         memmap[SIFIVE_E_XIP].size, &error_fatal);
-    memory_region_set_readonly(xip_mem, true);
-    memory_region_add_subregion(sys_mem, memmap[SIFIVE_E_XIP].base, xip_mem);
+    memory_region_set_readonly(&s->xip_mem, true);
+    memory_region_add_subregion(sys_mem, memmap[SIFIVE_E_XIP].base,
+        &s->xip_mem);
 }
 
 static void riscv_sifive_e_machine_init(MachineClass *mc)
diff --git a/hw/riscv/sifive_prci.c b/hw/riscv/sifive_prci.c
index fa136b5a9f..f406682c91 100644
--- a/hw/riscv/sifive_prci.c
+++ b/hw/riscv/sifive_prci.c
@@ -24,15 +24,18 @@
 #include "target/riscv/cpu.h"
 #include "hw/riscv/sifive_prci.h"
 
-/* currently implements enough to mock freedom-e-sdk BSP clock programming */
-
 static uint64_t sifive_prci_read(void *opaque, hwaddr addr, unsigned int size)
 {
-    if (addr == 0 /* PRCI_HFROSCCFG */) {
-        return 1 << 31; /* ROSC_RDY */
-    }
-    if (addr == 8 /* PRCI_PLLCFG    */) {
-        return 1 << 31; /* PLL_LOCK */
+    SiFivePRCIState *s = opaque;
+    switch (addr) {
+    case SIFIVE_PRCI_HFROSCCFG:
+        return s->hfrosccfg;
+    case SIFIVE_PRCI_HFXOSCCFG:
+        return s->hfxosccfg;
+    case SIFIVE_PRCI_PLLCFG:
+        return s->pllcfg;
+    case SIFIVE_PRCI_PLLOUTDIV:
+        return s->plloutdiv;
     }
     hw_error("%s: read: addr=0x%x\n", __func__, (int)addr);
     return 0;
@@ -41,7 +44,30 @@ static uint64_t sifive_prci_read(void *opaque, hwaddr addr, unsigned int size)
 static void sifive_prci_write(void *opaque, hwaddr addr,
            uint64_t val64, unsigned int size)
 {
-    /* discard writes */
+    SiFivePRCIState *s = opaque;
+    switch (addr) {
+    case SIFIVE_PRCI_HFROSCCFG:
+        s->hfrosccfg = (uint32_t) val64;
+        /* OSC stays ready */
+        s->hfrosccfg |= SIFIVE_PRCI_HFROSCCFG_RDY;
+        break;
+    case SIFIVE_PRCI_HFXOSCCFG:
+        s->hfxosccfg = (uint32_t) val64;
+        /* OSC stays ready */
+        s->hfxosccfg |= SIFIVE_PRCI_HFXOSCCFG_RDY;
+        break;
+    case SIFIVE_PRCI_PLLCFG:
+        s->pllcfg = (uint32_t) val64;
+        /* PLL stays locked */
+        s->pllcfg |= SIFIVE_PRCI_PLLCFG_LOCK;
+        break;
+    case SIFIVE_PRCI_PLLOUTDIV:
+        s->plloutdiv = (uint32_t) val64;
+        break;
+    default:
+        hw_error("%s: bad write: addr=0x%x v=0x%x\n",
+                 __func__, (int)addr, (int)val64);
+    }
 }
 
 static const MemoryRegionOps sifive_prci_ops = {
@@ -61,6 +87,13 @@ static void sifive_prci_init(Object *obj)
     memory_region_init_io(&s->mmio, obj, &sifive_prci_ops, s,
                           TYPE_SIFIVE_PRCI, 0x8000);
     sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->mmio);
+
+    s->hfrosccfg = (SIFIVE_PRCI_HFROSCCFG_RDY | SIFIVE_PRCI_HFROSCCFG_EN);
+    s->hfxosccfg = (SIFIVE_PRCI_HFROSCCFG_RDY | SIFIVE_PRCI_HFROSCCFG_EN);
+    s->pllcfg = (SIFIVE_PRCI_PLLCFG_REFSEL | SIFIVE_PRCI_PLLCFG_BYPASS |
+                SIFIVE_PRCI_PLLCFG_LOCK);
+    s->plloutdiv = SIFIVE_PRCI_PLLOUTDIV_DIV1;
+
 }
 
 static const TypeInfo sifive_prci_info = {
diff --git a/hw/riscv/sifive_u.c b/hw/riscv/sifive_u.c
index 5ecc47cea3..4208671552 100644
--- a/hw/riscv/sifive_u.c
+++ b/hw/riscv/sifive_u.c
@@ -41,11 +41,11 @@
 #include "hw/riscv/sifive_uart.h"
 #include "hw/riscv/sifive_prci.h"
 #include "hw/riscv/sifive_u.h"
+#include "hw/riscv/boot.h"
 #include "chardev/char.h"
 #include "sysemu/arch_init.h"
 #include "sysemu/device_tree.h"
 #include "exec/address-spaces.h"
-#include "elf.h"
 
 #include <libfdt.h>
 
@@ -65,19 +65,6 @@ static const struct MemmapEntry {
 
 #define GEM_REVISION        0x10070109
 
-static target_ulong load_kernel(const char *kernel_filename)
-{
-    uint64_t kernel_entry, kernel_high;
-
-    if (load_elf(kernel_filename, NULL, NULL, NULL,
-                 &kernel_entry, NULL, &kernel_high,
-                 0, EM_RISCV, 1, 0) < 0) {
-        error_report("could not load kernel '%s'", kernel_filename);
-        exit(1);
-    }
-    return kernel_entry;
-}
-
 static void create_fdt(SiFiveUState *s, const struct MemmapEntry *memmap,
     uint64_t mem_size, const char *cmdline)
 {
@@ -86,7 +73,7 @@ static void create_fdt(SiFiveUState *s, const struct MemmapEntry *memmap,
     uint32_t *cells;
     char *nodename;
     char ethclk_names[] = "pclk\0hclk\0tx_clk";
-    uint32_t plic_phandle, ethclk_phandle;
+    uint32_t plic_phandle, ethclk_phandle, phandle = 1;
 
     fdt = s->fdt = create_device_tree(&s->fdt_size);
     if (!fdt) {
@@ -121,6 +108,7 @@ static void create_fdt(SiFiveUState *s, const struct MemmapEntry *memmap,
     qemu_fdt_setprop_cell(fdt, "/cpus", "#address-cells", 0x1);
 
     for (cpu = s->soc.cpus.num_harts - 1; cpu >= 0; cpu--) {
+        int cpu_phandle = phandle++;
         nodename = g_strdup_printf("/cpus/cpu@%d", cpu);
         char *intc = g_strdup_printf("/cpus/cpu@%d/interrupt-controller", cpu);
         char *isa = riscv_isa_string(&s->soc.cpus.harts[cpu]);
@@ -134,8 +122,8 @@ static void create_fdt(SiFiveUState *s, const struct MemmapEntry *memmap,
         qemu_fdt_setprop_cell(fdt, nodename, "reg", cpu);
         qemu_fdt_setprop_string(fdt, nodename, "device_type", "cpu");
         qemu_fdt_add_subnode(fdt, intc);
-        qemu_fdt_setprop_cell(fdt, intc, "phandle", 1);
-        qemu_fdt_setprop_cell(fdt, intc, "linux,phandle", 1);
+        qemu_fdt_setprop_cell(fdt, intc, "phandle", cpu_phandle);
+        qemu_fdt_setprop_cell(fdt, intc, "linux,phandle", cpu_phandle);
         qemu_fdt_setprop_string(fdt, intc, "compatible", "riscv,cpu-intc");
         qemu_fdt_setprop(fdt, intc, "interrupt-controller", NULL, 0);
         qemu_fdt_setprop_cell(fdt, intc, "#interrupt-cells", 1);
@@ -167,6 +155,7 @@ static void create_fdt(SiFiveUState *s, const struct MemmapEntry *memmap,
     g_free(cells);
     g_free(nodename);
 
+    plic_phandle = phandle++;
     cells =  g_new0(uint32_t, s->soc.cpus.num_harts * 4);
     for (cpu = 0; cpu < s->soc.cpus.num_harts; cpu++) {
         nodename =
@@ -192,20 +181,21 @@ static void create_fdt(SiFiveUState *s, const struct MemmapEntry *memmap,
     qemu_fdt_setprop_string(fdt, nodename, "reg-names", "control");
     qemu_fdt_setprop_cell(fdt, nodename, "riscv,max-priority", 7);
     qemu_fdt_setprop_cell(fdt, nodename, "riscv,ndev", 0x35);
-    qemu_fdt_setprop_cells(fdt, nodename, "phandle", 2);
-    qemu_fdt_setprop_cells(fdt, nodename, "linux,phandle", 2);
+    qemu_fdt_setprop_cells(fdt, nodename, "phandle", plic_phandle);
+    qemu_fdt_setprop_cells(fdt, nodename, "linux,phandle", plic_phandle);
     plic_phandle = qemu_fdt_get_phandle(fdt, nodename);
     g_free(cells);
     g_free(nodename);
 
+    ethclk_phandle = phandle++;
     nodename = g_strdup_printf("/soc/ethclk");
     qemu_fdt_add_subnode(fdt, nodename);
     qemu_fdt_setprop_string(fdt, nodename, "compatible", "fixed-clock");
     qemu_fdt_setprop_cell(fdt, nodename, "#clock-cells", 0x0);
     qemu_fdt_setprop_cell(fdt, nodename, "clock-frequency",
         SIFIVE_U_GEM_CLOCK_FREQ);
-    qemu_fdt_setprop_cell(fdt, nodename, "phandle", 3);
-    qemu_fdt_setprop_cell(fdt, nodename, "linux,phandle", 3);
+    qemu_fdt_setprop_cell(fdt, nodename, "phandle", ethclk_phandle);
+    qemu_fdt_setprop_cell(fdt, nodename, "linux,phandle", ethclk_phandle);
     ethclk_phandle = qemu_fdt_get_phandle(fdt, nodename);
     g_free(nodename);
 
@@ -279,8 +269,12 @@ static void riscv_sifive_u_init(MachineState *machine)
     /* create device tree */
     create_fdt(s, memmap, machine->ram_size, machine->kernel_cmdline);
 
+    if (machine->firmware) {
+        riscv_load_firmware(machine->firmware, memmap[SIFIVE_U_DRAM].base);
+    }
+
     if (machine->kernel_filename) {
-        load_kernel(machine->kernel_filename);
+        riscv_load_kernel(machine->kernel_filename);
     }
 
     /* reset vector */
@@ -341,6 +335,8 @@ static void riscv_sifive_u_soc_realize(DeviceState *dev, Error **errp)
     MemoryRegion *system_memory = get_system_memory();
     MemoryRegion *mask_rom = g_new(MemoryRegion, 1);
     qemu_irq plic_gpios[SIFIVE_U_PLIC_NUM_SOURCES];
+    char *plic_hart_config;
+    size_t plic_hart_config_len;
     int i;
     Error *err = NULL;
     NICInfo *nd = &nd_table[0];
@@ -354,9 +350,21 @@ static void riscv_sifive_u_soc_realize(DeviceState *dev, Error **errp)
     memory_region_add_subregion(system_memory, memmap[SIFIVE_U_MROM].base,
                                 mask_rom);
 
+    /* create PLIC hart topology configuration string */
+    plic_hart_config_len = (strlen(SIFIVE_U_PLIC_HART_CONFIG) + 1) * smp_cpus;
+    plic_hart_config = g_malloc0(plic_hart_config_len);
+    for (i = 0; i < smp_cpus; i++) {
+        if (i != 0) {
+            strncat(plic_hart_config, ",", plic_hart_config_len);
+        }
+        strncat(plic_hart_config, SIFIVE_U_PLIC_HART_CONFIG,
+                plic_hart_config_len);
+        plic_hart_config_len -= (strlen(SIFIVE_U_PLIC_HART_CONFIG) + 1);
+    }
+
     /* MMIO */
     s->plic = sifive_plic_create(memmap[SIFIVE_U_PLIC].base,
-        (char *)SIFIVE_U_PLIC_HART_CONFIG,
+        plic_hart_config,
         SIFIVE_U_PLIC_NUM_SOURCES,
         SIFIVE_U_PLIC_NUM_PRIORITIES,
         SIFIVE_U_PLIC_PRIORITY_BASE,
diff --git a/hw/riscv/spike.c b/hw/riscv/spike.c
index 5b33d4be3b..e68be00a5f 100644
--- a/hw/riscv/spike.c
+++ b/hw/riscv/spike.c
@@ -36,12 +36,12 @@
 #include "hw/riscv/riscv_hart.h"
 #include "hw/riscv/sifive_clint.h"
 #include "hw/riscv/spike.h"
+#include "hw/riscv/boot.h"
 #include "chardev/char.h"
 #include "sysemu/arch_init.h"
 #include "sysemu/device_tree.h"
 #include "sysemu/qtest.h"
 #include "exec/address-spaces.h"
-#include "elf.h"
 
 #include <libfdt.h>
 
@@ -54,19 +54,6 @@ static const struct MemmapEntry {
     [SPIKE_DRAM] =     { 0x80000000,        0x0 },
 };
 
-static target_ulong load_kernel(const char *kernel_filename)
-{
-    uint64_t kernel_entry, kernel_high;
-
-    if (load_elf_ram_sym(kernel_filename, NULL, NULL, NULL,
-            &kernel_entry, NULL, &kernel_high, 0, EM_RISCV, 1, 0,
-            NULL, true, htif_symbol_callback) < 0) {
-        error_report("could not load kernel '%s'", kernel_filename);
-        exit(1);
-    }
-    return kernel_entry;
-}
-
 static void create_fdt(SpikeState *s, const struct MemmapEntry *memmap,
     uint64_t mem_size, const char *cmdline)
 {
@@ -199,7 +186,7 @@ static void spike_board_init(MachineState *machine)
                                 mask_rom);
 
     if (machine->kernel_filename) {
-        load_kernel(machine->kernel_filename);
+        riscv_load_kernel(machine->kernel_filename);
     }
 
     /* reset vector */
@@ -287,7 +274,7 @@ static void spike_v1_10_0_board_init(MachineState *machine)
                                 mask_rom);
 
     if (machine->kernel_filename) {
-        load_kernel(machine->kernel_filename);
+        riscv_load_kernel(machine->kernel_filename);
     }
 
     /* reset vector */
@@ -372,7 +359,7 @@ static void spike_v1_09_1_board_init(MachineState *machine)
                                 mask_rom);
 
     if (machine->kernel_filename) {
-        load_kernel(machine->kernel_filename);
+        riscv_load_kernel(machine->kernel_filename);
     }
 
     /* reset vector */
diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index 84d94d0c42..d8181a4ff1 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -34,13 +34,13 @@
 #include "hw/riscv/sifive_clint.h"
 #include "hw/riscv/sifive_test.h"
 #include "hw/riscv/virt.h"
+#include "hw/riscv/boot.h"
 #include "chardev/char.h"
 #include "sysemu/arch_init.h"
 #include "sysemu/device_tree.h"
 #include "exec/address-spaces.h"
 #include "hw/pci/pci.h"
 #include "hw/pci-host/gpex.h"
-#include "elf.h"
 
 #include <libfdt.h>
 
@@ -61,47 +61,6 @@ static const struct MemmapEntry {
     [VIRT_PCIE_ECAM] =   { 0x30000000,    0x10000000 },
 };
 
-static target_ulong load_kernel(const char *kernel_filename)
-{
-    uint64_t kernel_entry, kernel_high;
-
-    if (load_elf(kernel_filename, NULL, NULL, NULL,
-                 &kernel_entry, NULL, &kernel_high,
-                 0, EM_RISCV, 1, 0) < 0) {
-        error_report("could not load kernel '%s'", kernel_filename);
-        exit(1);
-    }
-    return kernel_entry;
-}
-
-static hwaddr load_initrd(const char *filename, uint64_t mem_size,
-                          uint64_t kernel_entry, hwaddr *start)
-{
-    int size;
-
-    /* We want to put the initrd far enough into RAM that when the
-     * kernel is uncompressed it will not clobber the initrd. However
-     * on boards without much RAM we must ensure that we still leave
-     * enough room for a decent sized initrd, and on boards with large
-     * amounts of RAM we must avoid the initrd being so far up in RAM
-     * that it is outside lowmem and inaccessible to the kernel.
-     * So for boards with less  than 256MB of RAM we put the initrd
-     * halfway into RAM, and for boards with 256MB of RAM or more we put
-     * the initrd at 128MB.
-     */
-    *start = kernel_entry + MIN(mem_size / 2, 128 * MiB);
-
-    size = load_ramdisk(filename, *start, mem_size - *start);
-    if (size == -1) {
-        size = load_image_targphys(filename, *start, mem_size - *start);
-        if (size == -1) {
-            error_report("could not load ramdisk '%s'", filename);
-            exit(1);
-        }
-    }
-    return *start + size;
-}
-
 static void create_pcie_irq_map(void *fdt, char *nodename,
                                 uint32_t plic_phandle)
 {
@@ -191,6 +150,7 @@ static void *create_fdt(RISCVVirtState *s, const struct MemmapEntry *memmap,
 
     for (cpu = s->soc.num_harts - 1; cpu >= 0; cpu--) {
         int cpu_phandle = phandle++;
+        int intc_phandle;
         nodename = g_strdup_printf("/cpus/cpu@%d", cpu);
         char *intc = g_strdup_printf("/cpus/cpu@%d/interrupt-controller", cpu);
         char *isa = riscv_isa_string(&s->soc.harts[cpu]);
@@ -203,9 +163,12 @@ static void *create_fdt(RISCVVirtState *s, const struct MemmapEntry *memmap,
         qemu_fdt_setprop_string(fdt, nodename, "status", "okay");
         qemu_fdt_setprop_cell(fdt, nodename, "reg", cpu);
         qemu_fdt_setprop_string(fdt, nodename, "device_type", "cpu");
+        qemu_fdt_setprop_cell(fdt, nodename, "phandle", cpu_phandle);
+        qemu_fdt_setprop_cell(fdt, nodename, "linux,phandle", cpu_phandle);
+        intc_phandle = phandle++;
         qemu_fdt_add_subnode(fdt, intc);
-        qemu_fdt_setprop_cell(fdt, intc, "phandle", cpu_phandle);
-        qemu_fdt_setprop_cell(fdt, intc, "linux,phandle", cpu_phandle);
+        qemu_fdt_setprop_cell(fdt, intc, "phandle", intc_phandle);
+        qemu_fdt_setprop_cell(fdt, intc, "linux,phandle", intc_phandle);
         qemu_fdt_setprop_string(fdt, intc, "compatible", "riscv,cpu-intc");
         qemu_fdt_setprop(fdt, intc, "interrupt-controller", NULL, 0);
         qemu_fdt_setprop_cell(fdt, intc, "#interrupt-cells", 1);
@@ -214,6 +177,20 @@ static void *create_fdt(RISCVVirtState *s, const struct MemmapEntry *memmap,
         g_free(nodename);
     }
 
+    /* Add cpu-topology node */
+    qemu_fdt_add_subnode(fdt, "/cpus/cpu-map");
+    qemu_fdt_add_subnode(fdt, "/cpus/cpu-map/cluster0");
+    for (cpu = s->soc.num_harts - 1; cpu >= 0; cpu--) {
+        char *core_nodename = g_strdup_printf("/cpus/cpu-map/cluster0/core%d",
+                                              cpu);
+        char *cpu_nodename = g_strdup_printf("/cpus/cpu@%d", cpu);
+        uint32_t intc_phandle = qemu_fdt_get_phandle(fdt, cpu_nodename);
+        qemu_fdt_add_subnode(fdt, core_nodename);
+        qemu_fdt_setprop_cell(fdt, core_nodename, "cpu", intc_phandle);
+        g_free(core_nodename);
+        g_free(cpu_nodename);
+    }
+
     cells =  g_new0(uint32_t, s->soc.num_harts * 4);
     for (cpu = 0; cpu < s->soc.num_harts; cpu++) {
         nodename =
@@ -298,7 +275,7 @@ static void *create_fdt(RISCVVirtState *s, const struct MemmapEntry *memmap,
     qemu_fdt_setprop_string(fdt, nodename, "device_type", "pci");
     qemu_fdt_setprop_cell(fdt, nodename, "linux,pci-domain", 0);
     qemu_fdt_setprop_cells(fdt, nodename, "bus-range", 0,
-                           memmap[VIRT_PCIE_ECAM].base /
+                           memmap[VIRT_PCIE_ECAM].size /
                                PCIE_MMCFG_SIZE_MIN - 1);
     qemu_fdt_setprop(fdt, nodename, "dma-coherent", NULL, 0);
     qemu_fdt_setprop_cells(fdt, nodename, "reg", 0, memmap[VIRT_PCIE_ECAM].base,
@@ -421,14 +398,18 @@ static void riscv_virt_board_init(MachineState *machine)
     memory_region_add_subregion(system_memory, memmap[VIRT_MROM].base,
                                 mask_rom);
 
+    if (machine->firmware) {
+        riscv_load_firmware(machine->firmware, memmap[VIRT_DRAM].base);
+    }
+
     if (machine->kernel_filename) {
-        uint64_t kernel_entry = load_kernel(machine->kernel_filename);
+        uint64_t kernel_entry = riscv_load_kernel(machine->kernel_filename);
 
         if (machine->initrd_filename) {
             hwaddr start;
-            hwaddr end = load_initrd(machine->initrd_filename,
-                                     machine->ram_size, kernel_entry,
-                                     &start);
+            hwaddr end = riscv_load_initrd(machine->initrd_filename,
+                                           machine->ram_size, kernel_entry,
+                                           &start);
             qemu_fdt_setprop_cell(fdt, "/chosen",
                                   "linux,initrd-start", start);
             qemu_fdt_setprop_cell(fdt, "/chosen", "linux,initrd-end",
diff --git a/hw/s390x/css.c b/hw/s390x/css.c
index ad310b9f94..b92395f165 100644
--- a/hw/s390x/css.c
+++ b/hw/s390x/css.c
@@ -22,6 +22,7 @@
 #include "trace.h"
 #include "hw/s390x/s390_flic.h"
 #include "hw/s390x/s390-virtio-ccw.h"
+#include "hw/s390x/s390-ccw.h"
 
 typedef struct CrwContainer {
     CRW crw;
@@ -1205,6 +1206,26 @@ static void sch_handle_start_func_virtual(SubchDev *sch)
 
 }
 
+static void sch_handle_halt_func_passthrough(SubchDev *sch)
+{
+    int ret;
+
+    ret = s390_ccw_halt(sch);
+    if (ret == -ENOSYS) {
+        sch_handle_halt_func(sch);
+    }
+}
+
+static void sch_handle_clear_func_passthrough(SubchDev *sch)
+{
+    int ret;
+
+    ret = s390_ccw_clear(sch);
+    if (ret == -ENOSYS) {
+        sch_handle_clear_func(sch);
+    }
+}
+
 static IOInstEnding sch_handle_start_func_passthrough(SubchDev *sch)
 {
     SCHIB *schib = &sch->curr_status;
@@ -1244,11 +1265,9 @@ IOInstEnding do_subchannel_work_passthrough(SubchDev *sch)
     SCHIB *schib = &sch->curr_status;
 
     if (schib->scsw.ctrl & SCSW_FCTL_CLEAR_FUNC) {
-        /* TODO: Clear handling */
-        sch_handle_clear_func(sch);
+        sch_handle_clear_func_passthrough(sch);
     } else if (schib->scsw.ctrl & SCSW_FCTL_HALT_FUNC) {
-        /* TODO: Halt handling */
-        sch_handle_halt_func(sch);
+        sch_handle_halt_func_passthrough(sch);
     } else if (schib->scsw.ctrl & SCSW_FCTL_START_FUNC) {
         return sch_handle_start_func_passthrough(sch);
     }
diff --git a/hw/s390x/s390-ccw.c b/hw/s390x/s390-ccw.c
index 8403f0e3e9..22c6878b84 100644
--- a/hw/s390x/s390-ccw.c
+++ b/hw/s390x/s390-ccw.c
@@ -30,6 +30,26 @@ IOInstEnding s390_ccw_cmd_request(SubchDev *sch)
     return cdc->handle_request(sch);
 }
 
+int s390_ccw_halt(SubchDev *sch)
+{
+    S390CCWDeviceClass *cdc = S390_CCW_DEVICE_GET_CLASS(sch->driver_data);
+
+    if (!cdc->handle_halt) {
+        return -ENOSYS;
+    }
+    return cdc->handle_halt(sch);
+}
+
+int s390_ccw_clear(SubchDev *sch)
+{
+    S390CCWDeviceClass *cdc = S390_CCW_DEVICE_GET_CLASS(sch->driver_data);
+
+    if (!cdc->handle_clear) {
+        return -ENOSYS;
+    }
+    return cdc->handle_clear(sch);
+}
+
 static void s390_ccw_get_dev_info(S390CCWDevice *cdev,
                                   char *sysfsdev,
                                   Error **errp)
diff --git a/hw/s390x/s390-skeys.c b/hw/s390x/s390-skeys.c
index daac936698..e5bd92c0c7 100644
--- a/hw/s390x/s390-skeys.c
+++ b/hw/s390x/s390-skeys.c
@@ -14,7 +14,7 @@
 #include "hw/boards.h"
 #include "hw/s390x/storage-keys.h"
 #include "qapi/error.h"
-#include "qapi/qapi-commands-target.h"
+#include "qapi/qapi-commands-misc-target.h"
 #include "qapi/qmp/qdict.h"
 #include "qemu/error-report.h"
 #include "sysemu/kvm.h"
diff --git a/hw/sparc/sun4m.c b/hw/sparc/sun4m.c
index 7e4f61fc3e..99f53e87f7 100644
--- a/hw/sparc/sun4m.c
+++ b/hw/sparc/sun4m.c
@@ -1406,6 +1406,7 @@ static void ss5_class_init(ObjectClass *oc, void *data)
     mc->is_default = 1;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("Fujitsu-MB86904");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo ss5_type = {
@@ -1424,6 +1425,7 @@ static void ss10_class_init(ObjectClass *oc, void *data)
     mc->max_cpus = 4;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("TI-SuperSparc-II");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo ss10_type = {
@@ -1442,6 +1444,7 @@ static void ss600mp_class_init(ObjectClass *oc, void *data)
     mc->max_cpus = 4;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("TI-SuperSparc-II");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo ss600mp_type = {
@@ -1460,6 +1463,7 @@ static void ss20_class_init(ObjectClass *oc, void *data)
     mc->max_cpus = 4;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("TI-SuperSparc-II");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo ss20_type = {
@@ -1477,6 +1481,7 @@ static void voyager_class_init(ObjectClass *oc, void *data)
     mc->block_default_type = IF_SCSI;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("Fujitsu-MB86904");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo voyager_type = {
@@ -1494,6 +1499,7 @@ static void ss_lx_class_init(ObjectClass *oc, void *data)
     mc->block_default_type = IF_SCSI;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("TI-MicroSparc-I");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo ss_lx_type = {
@@ -1511,6 +1517,7 @@ static void ss4_class_init(ObjectClass *oc, void *data)
     mc->block_default_type = IF_SCSI;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("Fujitsu-MB86904");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo ss4_type = {
@@ -1528,6 +1535,7 @@ static void scls_class_init(ObjectClass *oc, void *data)
     mc->block_default_type = IF_SCSI;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("TI-MicroSparc-I");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo scls_type = {
@@ -1545,6 +1553,7 @@ static void sbook_class_init(ObjectClass *oc, void *data)
     mc->block_default_type = IF_SCSI;
     mc->default_boot_order = "c";
     mc->default_cpu_type = SPARC_CPU_TYPE_NAME("TI-MicroSparc-I");
+    mc->default_display = "tcx";
 }
 
 static const TypeInfo sbook_type = {
diff --git a/hw/ssi/aspeed_smc.c b/hw/ssi/aspeed_smc.c
index 9eda0d720b..81f2fb7f70 100644
--- a/hw/ssi/aspeed_smc.c
+++ b/hw/ssi/aspeed_smc.c
@@ -913,6 +913,7 @@ static const VMStateDescription vmstate_aspeed_smc = {
 
 static Property aspeed_smc_properties[] = {
     DEFINE_PROP_UINT32("num-cs", AspeedSMCState, num_cs, 1),
+    DEFINE_PROP_UINT64("sdram-base", AspeedSMCState, sdram_base, 0),
     DEFINE_PROP_END_OF_LIST(),
 };
 
diff --git a/hw/timer/Makefile.objs b/hw/timer/Makefile.objs
index 0e9a4530f8..123d92c969 100644
--- a/hw/timer/Makefile.objs
+++ b/hw/timer/Makefile.objs
@@ -41,7 +41,7 @@ obj-$(CONFIG_MC146818RTC) += mc146818rtc.o
 obj-$(CONFIG_ALLWINNER_A10_PIT) += allwinner-a10-pit.o
 
 common-obj-$(CONFIG_STM32F2XX_TIMER) += stm32f2xx_timer.o
-common-obj-$(CONFIG_ASPEED_SOC) += aspeed_timer.o
+common-obj-$(CONFIG_ASPEED_SOC) += aspeed_timer.o aspeed_rtc.o
 
 common-obj-$(CONFIG_SUN4V_RTC) += sun4v-rtc.o
 common-obj-$(CONFIG_CMSDK_APB_TIMER) += cmsdk-apb-timer.o
diff --git a/hw/timer/armv7m_systick.c b/hw/timer/armv7m_systick.c
index a17317ce2f..94640743b5 100644
--- a/hw/timer/armv7m_systick.c
+++ b/hw/timer/armv7m_systick.c
@@ -75,11 +75,17 @@ static void systick_timer_tick(void *opaque)
     }
 }
 
-static uint64_t systick_read(void *opaque, hwaddr addr, unsigned size)
+static MemTxResult systick_read(void *opaque, hwaddr addr, uint64_t *data,
+                                unsigned size, MemTxAttrs attrs)
 {
     SysTickState *s = opaque;
     uint32_t val;
 
+    if (attrs.user) {
+        /* Generate BusFault for unprivileged accesses */
+        return MEMTX_ERROR;
+    }
+
     switch (addr) {
     case 0x0: /* SysTick Control and Status.  */
         val = s->control;
@@ -121,14 +127,21 @@ static uint64_t systick_read(void *opaque, hwaddr addr, unsigned size)
     }
 
     trace_systick_read(addr, val, size);
-    return val;
+    *data = val;
+    return MEMTX_OK;
 }
 
-static void systick_write(void *opaque, hwaddr addr,
-                          uint64_t value, unsigned size)
+static MemTxResult systick_write(void *opaque, hwaddr addr,
+                                 uint64_t value, unsigned size,
+                                 MemTxAttrs attrs)
 {
     SysTickState *s = opaque;
 
+    if (attrs.user) {
+        /* Generate BusFault for unprivileged accesses */
+        return MEMTX_ERROR;
+    }
+
     trace_systick_write(addr, value, size);
 
     switch (addr) {
@@ -172,11 +185,12 @@ static void systick_write(void *opaque, hwaddr addr,
         qemu_log_mask(LOG_GUEST_ERROR,
                       "SysTick: Bad write offset 0x%" HWADDR_PRIx "\n", addr);
     }
+    return MEMTX_OK;
 }
 
 static const MemoryRegionOps systick_ops = {
-    .read = systick_read,
-    .write = systick_write,
+    .read_with_attrs = systick_read,
+    .write_with_attrs = systick_write,
     .endianness = DEVICE_NATIVE_ENDIAN,
     .valid.min_access_size = 4,
     .valid.max_access_size = 4,
diff --git a/hw/timer/aspeed_rtc.c b/hw/timer/aspeed_rtc.c
new file mode 100644
index 0000000000..19f061c846
--- /dev/null
+++ b/hw/timer/aspeed_rtc.c
@@ -0,0 +1,180 @@
+/*
+ * ASPEED Real Time Clock
+ * Joel Stanley <joel@jms.id.au>
+ *
+ * Copyright 2019 IBM Corp
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "qemu-common.h"
+#include "hw/timer/aspeed_rtc.h"
+#include "qemu/log.h"
+#include "qemu/timer.h"
+
+#include "trace.h"
+
+#define COUNTER1        (0x00 / 4)
+#define COUNTER2        (0x04 / 4)
+#define ALARM           (0x08 / 4)
+#define CONTROL         (0x10 / 4)
+#define ALARM_STATUS    (0x14 / 4)
+
+#define RTC_UNLOCKED    BIT(1)
+#define RTC_ENABLED     BIT(0)
+
+static void aspeed_rtc_calc_offset(AspeedRtcState *rtc)
+{
+    struct tm tm;
+    uint32_t year, cent;
+    uint32_t reg1 = rtc->reg[COUNTER1];
+    uint32_t reg2 = rtc->reg[COUNTER2];
+
+    tm.tm_mday = (reg1 >> 24) & 0x1f;
+    tm.tm_hour = (reg1 >> 16) & 0x1f;
+    tm.tm_min = (reg1 >> 8) & 0x3f;
+    tm.tm_sec = (reg1 >> 0) & 0x3f;
+
+    cent = (reg2 >> 16) & 0x1f;
+    year = (reg2 >> 8) & 0x7f;
+    tm.tm_mon = ((reg2 >>  0) & 0x0f) - 1;
+    tm.tm_year = year + (cent * 100) - 1900;
+
+    rtc->offset = qemu_timedate_diff(&tm);
+}
+
+static uint32_t aspeed_rtc_get_counter(AspeedRtcState *rtc, int r)
+{
+    uint32_t year, cent;
+    struct tm now;
+
+    qemu_get_timedate(&now, rtc->offset);
+
+    switch (r) {
+    case COUNTER1:
+        return (now.tm_mday << 24) | (now.tm_hour << 16) |
+            (now.tm_min << 8) | now.tm_sec;
+    case COUNTER2:
+        cent = (now.tm_year + 1900) / 100;
+        year = now.tm_year % 100;
+        return ((cent & 0x1f) << 16) | ((year & 0x7f) << 8) |
+            ((now.tm_mon + 1) & 0xf);
+    default:
+        g_assert_not_reached();
+    }
+}
+
+static uint64_t aspeed_rtc_read(void *opaque, hwaddr addr,
+                                unsigned size)
+{
+    AspeedRtcState *rtc = opaque;
+    uint64_t val;
+    uint32_t r = addr >> 2;
+
+    switch (r) {
+    case COUNTER1:
+    case COUNTER2:
+        if (rtc->reg[CONTROL] & RTC_ENABLED) {
+            rtc->reg[r] = aspeed_rtc_get_counter(rtc, r);
+        }
+        /* fall through */
+    case CONTROL:
+        val = rtc->reg[r];
+        break;
+    case ALARM:
+    case ALARM_STATUS:
+    default:
+        qemu_log_mask(LOG_UNIMP, "%s: 0x%" HWADDR_PRIx "\n", __func__, addr);
+        return 0;
+    }
+
+    trace_aspeed_rtc_read(addr, val);
+
+    return val;
+}
+
+static void aspeed_rtc_write(void *opaque, hwaddr addr,
+                             uint64_t val, unsigned size)
+{
+    AspeedRtcState *rtc = opaque;
+    uint32_t r = addr >> 2;
+
+    switch (r) {
+    case COUNTER1:
+    case COUNTER2:
+        if (!(rtc->reg[CONTROL] & RTC_UNLOCKED)) {
+            break;
+        }
+        /* fall through */
+    case CONTROL:
+        rtc->reg[r] = val;
+        aspeed_rtc_calc_offset(rtc);
+        break;
+    case ALARM:
+    case ALARM_STATUS:
+    default:
+        qemu_log_mask(LOG_UNIMP, "%s: 0x%" HWADDR_PRIx "\n", __func__, addr);
+        break;
+    }
+    trace_aspeed_rtc_write(addr, val);
+}
+
+static void aspeed_rtc_reset(DeviceState *d)
+{
+    AspeedRtcState *rtc = ASPEED_RTC(d);
+
+    rtc->offset = 0;
+    memset(rtc->reg, 0, sizeof(rtc->reg));
+}
+
+static const MemoryRegionOps aspeed_rtc_ops = {
+    .read = aspeed_rtc_read,
+    .write = aspeed_rtc_write,
+    .endianness = DEVICE_NATIVE_ENDIAN,
+};
+
+static const VMStateDescription vmstate_aspeed_rtc = {
+    .name = TYPE_ASPEED_RTC,
+    .version_id = 1,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT32_ARRAY(reg, AspeedRtcState, 0x18),
+        VMSTATE_INT32(offset, AspeedRtcState),
+        VMSTATE_INT32(offset, AspeedRtcState),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
+static void aspeed_rtc_realize(DeviceState *dev, Error **errp)
+{
+    SysBusDevice *sbd = SYS_BUS_DEVICE(dev);
+    AspeedRtcState *s = ASPEED_RTC(dev);
+
+    sysbus_init_irq(sbd, &s->irq);
+
+    memory_region_init_io(&s->iomem, OBJECT(s), &aspeed_rtc_ops, s,
+                          "aspeed-rtc", 0x18ULL);
+    sysbus_init_mmio(sbd, &s->iomem);
+}
+
+static void aspeed_rtc_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+
+    dc->realize = aspeed_rtc_realize;
+    dc->vmsd = &vmstate_aspeed_rtc;
+    dc->reset = aspeed_rtc_reset;
+}
+
+static const TypeInfo aspeed_rtc_info = {
+    .name          = TYPE_ASPEED_RTC,
+    .parent        = TYPE_SYS_BUS_DEVICE,
+    .instance_size = sizeof(AspeedRtcState),
+    .class_init    = aspeed_rtc_class_init,
+};
+
+static void aspeed_rtc_register_types(void)
+{
+    type_register_static(&aspeed_rtc_info);
+}
+
+type_init(aspeed_rtc_register_types)
diff --git a/hw/timer/aspeed_timer.c b/hw/timer/aspeed_timer.c
index 2c3a4d0fe7..29cc5e8070 100644
--- a/hw/timer/aspeed_timer.c
+++ b/hw/timer/aspeed_timer.c
@@ -107,39 +107,49 @@ static inline uint64_t calculate_time(struct AspeedTimer *t, uint32_t ticks)
     return t->start + delta_ns;
 }
 
+static inline uint32_t calculate_match(struct AspeedTimer *t, int i)
+{
+    return t->match[i] < t->reload ? t->match[i] : 0;
+}
+
 static uint64_t calculate_next(struct AspeedTimer *t)
 {
-    uint64_t next = 0;
-    uint32_t rate = calculate_rate(t);
+    uint64_t now = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
+    uint64_t next;
 
-    while (!next) {
-        /* We don't know the relationship between the values in the match
-         * registers, so sort using MAX/MIN/zero. We sort in that order as the
-         * timer counts down to zero. */
-        uint64_t seq[] = {
-            calculate_time(t, MAX(t->match[0], t->match[1])),
-            calculate_time(t, MIN(t->match[0], t->match[1])),
-            calculate_time(t, 0),
-        };
-        uint64_t reload_ns;
-        uint64_t now = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
-
-        if (now < seq[0]) {
-            next = seq[0];
-        } else if (now < seq[1]) {
-            next = seq[1];
-        } else if (now < seq[2]) {
-            next = seq[2];
-        } else if (t->reload) {
-            reload_ns = muldiv64(t->reload, NANOSECONDS_PER_SECOND, rate);
-            t->start = now - ((now - t->start) % reload_ns);
-        } else {
-            /* no reload value, return 0 */
-            break;
-        }
+    /*
+     * We don't know the relationship between the values in the match
+     * registers, so sort using MAX/MIN/zero. We sort in that order as
+     * the timer counts down to zero.
+     */
+
+    next = calculate_time(t, MAX(calculate_match(t, 0), calculate_match(t, 1)));
+    if (now < next) {
+        return next;
+    }
+
+    next = calculate_time(t, MIN(calculate_match(t, 0), calculate_match(t, 1)));
+    if (now < next) {
+        return next;
     }
 
-    return next;
+    next = calculate_time(t, 0);
+    if (now < next) {
+        return next;
+    }
+
+    /* We've missed all deadlines, fire interrupt and try again */
+    timer_del(&t->timer);
+
+    if (timer_overflow_interrupt(t)) {
+        t->level = !t->level;
+        qemu_set_irq(t->irq, t->level);
+    }
+
+    next = MAX(MAX(calculate_match(t, 0), calculate_match(t, 1)), 0);
+    t->start = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
+
+    return calculate_time(t, next);
 }
 
 static void aspeed_timer_mod(AspeedTimer *t)
@@ -184,7 +194,11 @@ static uint64_t aspeed_timer_get_value(AspeedTimer *t, int reg)
 
     switch (reg) {
     case TIMER_REG_STATUS:
-        value = calculate_ticks(t, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL));
+        if (timer_enabled(t)) {
+            value = calculate_ticks(t, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL));
+        } else {
+            value = t->reload;
+        }
         break;
     case TIMER_REG_RELOAD:
         value = t->reload;
@@ -261,7 +275,11 @@ static void aspeed_timer_set_value(AspeedTimerCtrlState *s, int timer, int reg,
             int64_t delta = (int64_t) value - (int64_t) calculate_ticks(t, now);
             uint32_t rate = calculate_rate(t);
 
-            t->start += muldiv64(delta, NANOSECONDS_PER_SECOND, rate);
+            if (delta >= 0) {
+                t->start += muldiv64(delta, NANOSECONDS_PER_SECOND, rate);
+            } else {
+                t->start -= muldiv64(-delta, NANOSECONDS_PER_SECOND, rate);
+            }
             aspeed_timer_mod(t);
         }
         break;
diff --git a/hw/timer/mc146818rtc.c b/hw/timer/mc146818rtc.c
index 0d79e000d2..ce4550b6f2 100644
--- a/hw/timer/mc146818rtc.c
+++ b/hw/timer/mc146818rtc.c
@@ -33,8 +33,8 @@
 #include "sysemu/replay.h"
 #include "hw/timer/mc146818rtc.h"
 #include "qapi/error.h"
-#include "qapi/qapi-commands-target.h"
-#include "qapi/qapi-events-target.h"
+#include "qapi/qapi-commands-misc-target.h"
+#include "qapi/qapi-events-misc-target.h"
 #include "qapi/visitor.h"
 #include "exec/address-spaces.h"
 
diff --git a/hw/timer/trace-events b/hw/timer/trace-events
index dcaf3d6da6..db02a9142c 100644
--- a/hw/timer/trace-events
+++ b/hw/timer/trace-events
@@ -66,6 +66,10 @@ cmsdk_apb_dualtimer_read(uint64_t offset, uint64_t data, unsigned size) "CMSDK A
 cmsdk_apb_dualtimer_write(uint64_t offset, uint64_t data, unsigned size) "CMSDK APB dualtimer write: offset 0x%" PRIx64 " data 0x%" PRIx64 " size %u"
 cmsdk_apb_dualtimer_reset(void) "CMSDK APB dualtimer: reset"
 
+# hw/timer/aspeed-rtc.c
+aspeed_rtc_read(uint64_t addr, uint64_t value) "addr 0x%02" PRIx64 " value 0x%08" PRIx64
+aspeed_rtc_write(uint64_t addr, uint64_t value) "addr 0x%02" PRIx64 " value 0x%08" PRIx64
+
 # sun4v-rtc.c
 sun4v_rtc_read(uint64_t addr, uint64_t value) "read: addr 0x%" PRIx64 " value 0x%" PRIx64
 sun4v_rtc_write(uint64_t addr, uint64_t value) "write: addr 0x%" PRIx64 " value 0x%" PRIx64
diff --git a/hw/usb/xen-usb.c b/hw/usb/xen-usb.c
index b20d0cfadf..dfbb418e77 100644
--- a/hw/usb/xen-usb.c
+++ b/hw/usb/xen-usb.c
@@ -32,8 +32,7 @@
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qstring.h"
 
-#include "hw/xen/io/ring.h"
-#include <xen/io/usbif.h>
+#include "hw/xen/interface/io/usbif.h"
 
 /*
  * Check for required support of usbif.h: USBIF_SHORT_NOT_OK was the last
diff --git a/hw/vfio/ccw.c b/hw/vfio/ccw.c
index 03a2becb3e..6d0296fe4d 100644
--- a/hw/vfio/ccw.c
+++ b/hw/vfio/ccw.c
@@ -2,9 +2,12 @@
  * vfio based subchannel assignment support
  *
  * Copyright 2017 IBM Corp.
+ * Copyright 2019 Red Hat, Inc.
+ *
  * Author(s): Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com>
  *            Xiao Feng Ren <renxiaof@linux.vnet.ibm.com>
  *            Pierre Morel <pmorel@linux.vnet.ibm.com>
+ *            Cornelia Huck <cohuck@redhat.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
@@ -33,6 +36,9 @@ struct VFIOCCWDevice {
     uint64_t io_region_size;
     uint64_t io_region_offset;
     struct ccw_io_region *io_region;
+    uint64_t async_cmd_region_size;
+    uint64_t async_cmd_region_offset;
+    struct ccw_cmd_region *async_cmd_region;
     EventNotifier io_notifier;
     bool force_orb_pfch;
     bool warned_orb_pfch;
@@ -115,6 +121,87 @@ again:
     }
 }
 
+static int vfio_ccw_handle_clear(SubchDev *sch)
+{
+    S390CCWDevice *cdev = sch->driver_data;
+    VFIOCCWDevice *vcdev = DO_UPCAST(VFIOCCWDevice, cdev, cdev);
+    struct ccw_cmd_region *region = vcdev->async_cmd_region;
+    int ret;
+
+    if (!vcdev->async_cmd_region) {
+        /* Async command region not available, fall back to emulation */
+        return -ENOSYS;
+    }
+
+    memset(region, 0, sizeof(*region));
+    region->command = VFIO_CCW_ASYNC_CMD_CSCH;
+
+again:
+    ret = pwrite(vcdev->vdev.fd, region,
+                 vcdev->async_cmd_region_size, vcdev->async_cmd_region_offset);
+    if (ret != vcdev->async_cmd_region_size) {
+        if (errno == EAGAIN) {
+            goto again;
+        }
+        error_report("vfio-ccw: write cmd region failed with errno=%d", errno);
+        ret = -errno;
+    } else {
+        ret = region->ret_code;
+    }
+    switch (ret) {
+    case 0:
+    case -ENODEV:
+    case -EACCES:
+        return 0;
+    case -EFAULT:
+    default:
+        sch_gen_unit_exception(sch);
+        css_inject_io_interrupt(sch);
+        return 0;
+    }
+}
+
+static int vfio_ccw_handle_halt(SubchDev *sch)
+{
+    S390CCWDevice *cdev = sch->driver_data;
+    VFIOCCWDevice *vcdev = DO_UPCAST(VFIOCCWDevice, cdev, cdev);
+    struct ccw_cmd_region *region = vcdev->async_cmd_region;
+    int ret;
+
+    if (!vcdev->async_cmd_region) {
+        /* Async command region not available, fall back to emulation */
+        return -ENOSYS;
+    }
+
+    memset(region, 0, sizeof(*region));
+    region->command = VFIO_CCW_ASYNC_CMD_HSCH;
+
+again:
+    ret = pwrite(vcdev->vdev.fd, region,
+                 vcdev->async_cmd_region_size, vcdev->async_cmd_region_offset);
+    if (ret != vcdev->async_cmd_region_size) {
+        if (errno == EAGAIN) {
+            goto again;
+        }
+        error_report("vfio-ccw: write cmd region failed with errno=%d", errno);
+        ret = -errno;
+    } else {
+        ret = region->ret_code;
+    }
+    switch (ret) {
+    case 0:
+    case -EBUSY:
+    case -ENODEV:
+    case -EACCES:
+        return 0;
+    case -EFAULT:
+    default:
+        sch_gen_unit_exception(sch);
+        css_inject_io_interrupt(sch);
+        return 0;
+    }
+}
+
 static void vfio_ccw_reset(DeviceState *dev)
 {
     CcwDevice *ccw_dev = DO_UPCAST(CcwDevice, parent_obj, dev);
@@ -198,9 +285,8 @@ static void vfio_ccw_register_io_notifier(VFIOCCWDevice *vcdev, Error **errp)
 {
     VFIODevice *vdev = &vcdev->vdev;
     struct vfio_irq_info *irq_info;
-    struct vfio_irq_set *irq_set;
     size_t argsz;
-    int32_t *pfd;
+    int fd;
 
     if (vdev->num_irqs < VFIO_CCW_IO_IRQ_INDEX + 1) {
         error_setg(errp, "vfio: unexpected number of io irqs %u",
@@ -224,56 +310,32 @@ static void vfio_ccw_register_io_notifier(VFIOCCWDevice *vcdev, Error **errp)
         goto out_free_info;
     }
 
-    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 = VFIO_CCW_IO_IRQ_INDEX;
-    irq_set->start = 0;
-    irq_set->count = 1;
-    pfd = (int32_t *) &irq_set->data;
-
-    *pfd = event_notifier_get_fd(&vcdev->io_notifier);
-    qemu_set_fd_handler(*pfd, vfio_ccw_io_notifier_handler, NULL, vcdev);
-    if (ioctl(vdev->fd, VFIO_DEVICE_SET_IRQS, irq_set)) {
-        error_setg(errp, "vfio: Failed to set up io notification");
-        qemu_set_fd_handler(*pfd, NULL, NULL, vcdev);
+    fd = event_notifier_get_fd(&vcdev->io_notifier);
+    qemu_set_fd_handler(fd, vfio_ccw_io_notifier_handler, NULL, vcdev);
+
+    if (vfio_set_irq_signaling(vdev, VFIO_CCW_IO_IRQ_INDEX, 0,
+                               VFIO_IRQ_SET_ACTION_TRIGGER, fd, errp)) {
+        qemu_set_fd_handler(fd, NULL, NULL, vcdev);
         event_notifier_cleanup(&vcdev->io_notifier);
     }
 
-    g_free(irq_set);
-
 out_free_info:
     g_free(irq_info);
 }
 
 static void vfio_ccw_unregister_io_notifier(VFIOCCWDevice *vcdev)
 {
-    struct vfio_irq_set *irq_set;
-    size_t argsz;
-    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 = VFIO_CCW_IO_IRQ_INDEX;
-    irq_set->start = 0;
-    irq_set->count = 1;
-    pfd = (int32_t *) &irq_set->data;
-    *pfd = -1;
+    Error *err = NULL;
 
-    if (ioctl(vcdev->vdev.fd, VFIO_DEVICE_SET_IRQS, irq_set)) {
-        error_report("vfio: Failed to de-assign device io fd: %m");
+    vfio_set_irq_signaling(&vcdev->vdev, VFIO_CCW_IO_IRQ_INDEX, 0,
+                           VFIO_IRQ_SET_ACTION_TRIGGER, -1, &err);
+    if (err) {
+        error_reportf_err(err, VFIO_MSG_PREFIX, vcdev->vdev.name);
     }
 
     qemu_set_fd_handler(event_notifier_get_fd(&vcdev->io_notifier),
                         NULL, NULL, vcdev);
     event_notifier_cleanup(&vcdev->io_notifier);
-
-    g_free(irq_set);
 }
 
 static void vfio_ccw_get_region(VFIOCCWDevice *vcdev, Error **errp)
@@ -288,9 +350,13 @@ static void vfio_ccw_get_region(VFIOCCWDevice *vcdev, Error **errp)
         return;
     }
 
+    /*
+     * We always expect at least the I/O region to be present. We also
+     * may have a variable number of regions governed by capabilities.
+     */
     if (vdev->num_regions < VFIO_CCW_CONFIG_REGION_INDEX + 1) {
-        error_setg(errp, "vfio: Unexpected number of the I/O region %u",
-                   vdev->num_regions);
+        error_setg(errp, "vfio: too few regions (%u), expected at least %u",
+                   vdev->num_regions, VFIO_CCW_CONFIG_REGION_INDEX + 1);
         return;
     }
 
@@ -310,11 +376,27 @@ static void vfio_ccw_get_region(VFIOCCWDevice *vcdev, Error **errp)
     vcdev->io_region_offset = info->offset;
     vcdev->io_region = g_malloc0(info->size);
 
+    /* check for the optional async command region */
+    ret = vfio_get_dev_region_info(vdev, VFIO_REGION_TYPE_CCW,
+                                   VFIO_REGION_SUBTYPE_CCW_ASYNC_CMD, &info);
+    if (!ret) {
+        vcdev->async_cmd_region_size = info->size;
+        if (sizeof(*vcdev->async_cmd_region) != vcdev->async_cmd_region_size) {
+            error_setg(errp, "vfio: Unexpected size of the async cmd region");
+            g_free(vcdev->io_region);
+            g_free(info);
+            return;
+        }
+        vcdev->async_cmd_region_offset = info->offset;
+        vcdev->async_cmd_region = g_malloc0(info->size);
+    }
+
     g_free(info);
 }
 
 static void vfio_ccw_put_region(VFIOCCWDevice *vcdev)
 {
+    g_free(vcdev->async_cmd_region);
     g_free(vcdev->io_region);
 }
 
@@ -487,6 +569,8 @@ static void vfio_ccw_class_init(ObjectClass *klass, void *data)
     dc->reset = vfio_ccw_reset;
 
     cdc->handle_request = vfio_ccw_handle_request;
+    cdc->handle_halt = vfio_ccw_handle_halt;
+    cdc->handle_clear = vfio_ccw_handle_clear;
 }
 
 static const TypeInfo vfio_ccw_info = {
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index ce3fe96efe..d7a4e1875c 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -551,9 +551,12 @@ static void vfio_msix_vector_release(PCIDevice *pdev, unsigned int nr)
      */
     if (vector->virq >= 0) {
         int32_t fd = event_notifier_get_fd(&vector->interrupt);
+        Error *err = NULL;
 
-        vfio_set_irq_signaling(&vdev->vbasedev, VFIO_PCI_MSIX_IRQ_INDEX, nr,
-                               VFIO_IRQ_SET_ACTION_TRIGGER, fd, NULL);
+        if (vfio_set_irq_signaling(&vdev->vbasedev, VFIO_PCI_MSIX_IRQ_INDEX, nr,
+                                   VFIO_IRQ_SET_ACTION_TRIGGER, fd, &err)) {
+            error_reportf_err(err, VFIO_MSG_PREFIX, vdev->vbasedev.name);
+        }
     }
 }
 
diff --git a/hw/watchdog/wdt_aspeed.c b/hw/watchdog/wdt_aspeed.c
index 4a8409f0da..57fe24ae6b 100644
--- a/hw/watchdog/wdt_aspeed.c
+++ b/hw/watchdog/wdt_aspeed.c
@@ -44,6 +44,9 @@
 
 #define WDT_RESTART_MAGIC               0x4755
 
+#define SCU_RESET_CONTROL1              (0x04 / 4)
+#define    SCU_RESET_SDRAM              BIT(0)
+
 static bool aspeed_wdt_is_enabled(const AspeedWDTState *s)
 {
     return s->regs[WDT_CTRL] & WDT_CTRL_ENABLE;
@@ -222,6 +225,13 @@ static void aspeed_wdt_timer_expired(void *dev)
 {
     AspeedWDTState *s = ASPEED_WDT(dev);
 
+    /* Do not reset on SDRAM controller reset */
+    if (s->scu->regs[SCU_RESET_CONTROL1] & SCU_RESET_SDRAM) {
+        timer_del(s->timer);
+        s->regs[WDT_CTRL] = 0;
+        return;
+    }
+
     qemu_log_mask(CPU_LOG_RESET, "Watchdog timer expired.\n");
     watchdog_perform_action();
     timer_del(s->timer);
@@ -233,6 +243,16 @@ static void aspeed_wdt_realize(DeviceState *dev, Error **errp)
 {
     SysBusDevice *sbd = SYS_BUS_DEVICE(dev);
     AspeedWDTState *s = ASPEED_WDT(dev);
+    Error *err = NULL;
+    Object *obj;
+
+    obj = object_property_get_link(OBJECT(dev), "scu", &err);
+    if (!obj) {
+        error_propagate(errp, err);
+        error_prepend(errp, "required link 'scu' not found: ");
+        return;
+    }
+    s->scu = ASPEED_SCU(obj);
 
     if (!is_supported_silicon_rev(s->silicon_rev)) {
         error_setg(errp, "Unknown silicon revision: 0x%" PRIx32,
diff --git a/hw/xen/xen-bus.c b/hw/xen/xen-bus.c
index a4416d0bcf..7503eea9e9 100644
--- a/hw/xen/xen-bus.c
+++ b/hw/xen/xen-bus.c
@@ -924,23 +924,35 @@ done:
 }
 
 struct XenEventChannel {
+    QLIST_ENTRY(XenEventChannel) list;
+    AioContext *ctx;
+    xenevtchn_handle *xeh;
     evtchn_port_t local_port;
     XenEventHandler handler;
     void *opaque;
-    Notifier notifier;
 };
 
-static void event_notify(Notifier *n, void *data)
+static bool xen_device_poll(void *opaque)
+{
+    XenEventChannel *channel = opaque;
+
+    return channel->handler(channel->opaque);
+}
+
+static void xen_device_event(void *opaque)
 {
-    XenEventChannel *channel = container_of(n, XenEventChannel, notifier);
-    unsigned long port = (unsigned long)data;
+    XenEventChannel *channel = opaque;
+    unsigned long port = xenevtchn_pending(channel->xeh);
 
     if (port == channel->local_port) {
-        channel->handler(channel->opaque);
+        xen_device_poll(channel);
+
+        xenevtchn_unmask(channel->xeh, port);
     }
 }
 
 XenEventChannel *xen_device_bind_event_channel(XenDevice *xendev,
+                                               AioContext *ctx,
                                                unsigned int port,
                                                XenEventHandler handler,
                                                void *opaque, Error **errp)
@@ -948,24 +960,40 @@ XenEventChannel *xen_device_bind_event_channel(XenDevice *xendev,
     XenEventChannel *channel = g_new0(XenEventChannel, 1);
     xenevtchn_port_or_error_t local_port;
 
-    local_port = xenevtchn_bind_interdomain(xendev->xeh,
+    channel->xeh = xenevtchn_open(NULL, 0);
+    if (!channel->xeh) {
+        error_setg_errno(errp, errno, "failed xenevtchn_open");
+        goto fail;
+    }
+
+    local_port = xenevtchn_bind_interdomain(channel->xeh,
                                             xendev->frontend_id,
                                             port);
     if (local_port < 0) {
         error_setg_errno(errp, errno, "xenevtchn_bind_interdomain failed");
-
-        g_free(channel);
-        return NULL;
+        goto fail;
     }
 
     channel->local_port = local_port;
     channel->handler = handler;
     channel->opaque = opaque;
-    channel->notifier.notify = event_notify;
 
-    notifier_list_add(&xendev->event_notifiers, &channel->notifier);
+    channel->ctx = ctx;
+    aio_set_fd_handler(channel->ctx, xenevtchn_fd(channel->xeh), true,
+                       xen_device_event, NULL, xen_device_poll, channel);
+
+    QLIST_INSERT_HEAD(&xendev->event_channels, channel, list);
 
     return channel;
+
+fail:
+    if (channel->xeh) {
+        xenevtchn_close(channel->xeh);
+    }
+
+    g_free(channel);
+
+    return NULL;
 }
 
 void xen_device_notify_event_channel(XenDevice *xendev,
@@ -977,7 +1005,7 @@ void xen_device_notify_event_channel(XenDevice *xendev,
         return;
     }
 
-    if (xenevtchn_notify(xendev->xeh, channel->local_port) < 0) {
+    if (xenevtchn_notify(channel->xeh, channel->local_port) < 0) {
         error_setg_errno(errp, errno, "xenevtchn_notify failed");
     }
 }
@@ -991,12 +1019,16 @@ void xen_device_unbind_event_channel(XenDevice *xendev,
         return;
     }
 
-    notifier_remove(&channel->notifier);
+    QLIST_REMOVE(channel, list);
+
+    aio_set_fd_handler(channel->ctx, xenevtchn_fd(channel->xeh), true,
+                       NULL, NULL, NULL, NULL);
 
-    if (xenevtchn_unbind(xendev->xeh, channel->local_port) < 0) {
+    if (xenevtchn_unbind(channel->xeh, channel->local_port) < 0) {
         error_setg_errno(errp, errno, "xenevtchn_unbind failed");
     }
 
+    xenevtchn_close(channel->xeh);
     g_free(channel);
 }
 
@@ -1005,6 +1037,7 @@ static void xen_device_unrealize(DeviceState *dev, Error **errp)
     XenDevice *xendev = XEN_DEVICE(dev);
     XenDeviceClass *xendev_class = XEN_DEVICE_GET_CLASS(xendev);
     const char *type = object_get_typename(OBJECT(xendev));
+    XenEventChannel *channel, *next;
 
     if (!xendev->name) {
         return;
@@ -1021,15 +1054,14 @@ static void xen_device_unrealize(DeviceState *dev, Error **errp)
         xendev_class->unrealize(xendev, errp);
     }
 
+    /* Make sure all event channels are cleaned up */
+    QLIST_FOREACH_SAFE(channel, &xendev->event_channels, list, next) {
+        xen_device_unbind_event_channel(xendev, channel, NULL);
+    }
+
     xen_device_frontend_destroy(xendev);
     xen_device_backend_destroy(xendev);
 
-    if (xendev->xeh) {
-        qemu_set_fd_handler(xenevtchn_fd(xendev->xeh), NULL, NULL, NULL);
-        xenevtchn_close(xendev->xeh);
-        xendev->xeh = NULL;
-    }
-
     if (xendev->xgth) {
         xengnttab_close(xendev->xgth);
         xendev->xgth = NULL;
@@ -1046,16 +1078,6 @@ static void xen_device_exit(Notifier *n, void *data)
     xen_device_unrealize(DEVICE(xendev), &error_abort);
 }
 
-static void xen_device_event(void *opaque)
-{
-    XenDevice *xendev = opaque;
-    unsigned long port = xenevtchn_pending(xendev->xeh);
-
-    notifier_list_notify(&xendev->event_notifiers, (void *)port);
-
-    xenevtchn_unmask(xendev->xeh, port);
-}
-
 static void xen_device_realize(DeviceState *dev, Error **errp)
 {
     XenDevice *xendev = XEN_DEVICE(dev);
@@ -1096,16 +1118,6 @@ static void xen_device_realize(DeviceState *dev, Error **errp)
     xendev->feature_grant_copy =
         (xengnttab_grant_copy(xendev->xgth, 0, NULL) == 0);
 
-    xendev->xeh = xenevtchn_open(NULL, 0);
-    if (!xendev->xeh) {
-        error_setg_errno(errp, errno, "failed xenevtchn_open");
-        goto unrealize;
-    }
-
-    notifier_list_init(&xendev->event_notifiers);
-    qemu_set_fd_handler(xenevtchn_fd(xendev->xeh), xen_device_event, NULL,
-                        xendev);
-
     xen_device_backend_create(xendev, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
diff --git a/hw/xen/xen-legacy-backend.c b/hw/xen/xen-legacy-backend.c
index 36fd1e9b09..3715c94fa6 100644
--- a/hw/xen/xen-legacy-backend.c
+++ b/hw/xen/xen-legacy-backend.c
@@ -34,8 +34,6 @@
 #include "hw/xen/xen_pvdev.h"
 #include "monitor/qdev.h"
 
-#include <xen/grant_table.h>
-
 DeviceState *xen_sysdev;
 BusState *xen_sysbus;