Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

virtio, pc: fixes and features

nvdimm hotplug support
virtio migration and ioeventfd rework
virtio crypto device
ipmi fixes

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

# gpg: Signature made Tue 01 Nov 2016 05:23:40 PM GMT
# gpg:                using RSA key 0x281F0DB8D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>"
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>"
# Primary key fingerprint: 0270 606B 6F3C DF3D 0B17  0970 C350 3912 AFBE 8E67
#      Subkey fingerprint: 5D09 FD08 71C8 F85B 94CA  8A0D 281F 0DB8 D28D 5469

* remotes/mst/tags/for_upstream: (47 commits)
  acpi: fix assert failure caused by commit 35c5a52d
  acpi/ipmi: Initialize the fwinfo before fetching it
  ipmi: Add graceful shutdown handling to the external BMC
  ipmi: fix build config variable name for ipmi_bmc_extern.o
  ipmi: Implement shutdown via ACPI overtemp
  ipmi: chassis poweroff should use qemu_system_shutdown_request()
  ipmi_bmc_sim: Remove an unnecessary mutex
  ipmi: Remove hotplug from IPMI BMCs
  pc: memhp: enable nvdimm device hotplug
  nvdimm acpi: introduce _FIT
  nvdimm acpi: introduce fit buffer
  nvdimm acpi: prebuild nvdimm devices for available slots
  nvdimm acpi: use common macros instead of magic names
  acpi nvdimm: rename result_size to dsm_out_buf_siz
  nvdimm acpi: compile nvdimm acpi code arch-independently
  acpi nvdimm: fix Arg6 usage
  acpi nvdimm: fix ARG3 conflict
  acpi nvdimm: fix device physical address base
  acpi nvdimm: fix OperationRegion definition
  acpi nvdimm: fix wrong buffer size returned by DSM method
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

30 files changed, 1762 insertions, 478 deletions

diff --git a/hw/acpi/Makefile.objs b/hw/acpi/Makefile.objs
index 4b7da6639f..489e63bb75 100644
--- a/hw/acpi/Makefile.objs
+++ b/hw/acpi/Makefile.objs
@@ -3,7 +3,7 @@ common-obj-$(CONFIG_ACPI_X86_ICH) += ich9.o tco.o
 common-obj-$(CONFIG_ACPI_CPU_HOTPLUG) += cpu_hotplug.o
 common-obj-$(CONFIG_ACPI_MEMORY_HOTPLUG) += memory_hotplug.o memory_hotplug_acpi_table.o
 common-obj-$(CONFIG_ACPI_CPU_HOTPLUG) += cpu.o
-obj-$(CONFIG_ACPI_NVDIMM) += nvdimm.o
+common-obj-$(CONFIG_ACPI_NVDIMM) += nvdimm.o
 common-obj-$(CONFIG_ACPI) += acpi_interface.o
 common-obj-$(CONFIG_ACPI) += bios-linker-loader.o
 common-obj-$(CONFIG_ACPI) += aml-build.o
diff --git a/hw/acpi/ipmi.c b/hw/acpi/ipmi.c
index 7e74ce4460..651e2e94ea 100644
--- a/hw/acpi/ipmi.c
+++ b/hw/acpi/ipmi.c
@@ -99,6 +99,7 @@ void build_acpi_ipmi_devices(Aml *scope, BusState *bus)
 
         ii = IPMI_INTERFACE(obj);
         iic = IPMI_INTERFACE_GET_CLASS(obj);
+        memset(&info, 0, sizeof(info));
         iic->get_fwinfo(ii, &info);
         aml_append(scope, aml_ipmi_device(&info));
     }
diff --git a/hw/acpi/memory_hotplug.c b/hw/acpi/memory_hotplug.c
index ec4e64b361..70f64517fd 100644
--- a/hw/acpi/memory_hotplug.c
+++ b/hw/acpi/memory_hotplug.c
@@ -2,6 +2,7 @@
 #include "hw/acpi/memory_hotplug.h"
 #include "hw/acpi/pc-hotplug.h"
 #include "hw/mem/pc-dimm.h"
+#include "hw/mem/nvdimm.h"
 #include "hw/boards.h"
 #include "hw/qdev-core.h"
 #include "trace.h"
@@ -232,11 +233,8 @@ void acpi_memory_plug_cb(HotplugHandler *hotplug_dev, MemHotplugState *mem_st,
                          DeviceState *dev, Error **errp)
 {
     MemStatus *mdev;
-    DeviceClass *dc = DEVICE_GET_CLASS(dev);
-
-    if (!dc->hotpluggable) {
-        return;
-    }
+    AcpiEventStatusBits event;
+    bool is_nvdimm = object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM);
 
     mdev = acpi_memory_slot_status(mem_st, dev, errp);
     if (!mdev) {
@@ -244,10 +242,23 @@ void acpi_memory_plug_cb(HotplugHandler *hotplug_dev, MemHotplugState *mem_st,
     }
 
     mdev->dimm = dev;
-    mdev->is_enabled = true;
+
+    /*
+     * do not set is_enabled and is_inserting if the slot is plugged with
+     * a nvdimm device to stop OSPM inquires memory region from the slot.
+     */
+    if (is_nvdimm) {
+        event = ACPI_NVDIMM_HOTPLUG_STATUS;
+    } else {
+        mdev->is_enabled = true;
+        event = ACPI_MEMORY_HOTPLUG_STATUS;
+    }
+
     if (dev->hotplugged) {
-        mdev->is_inserting = true;
-        acpi_send_event(DEVICE(hotplug_dev), ACPI_MEMORY_HOTPLUG_STATUS);
+        if (!is_nvdimm) {
+            mdev->is_inserting = true;
+        }
+        acpi_send_event(DEVICE(hotplug_dev), event);
     }
 }
 
@@ -262,6 +273,8 @@ void acpi_memory_unplug_request_cb(HotplugHandler *hotplug_dev,
         return;
     }
 
+    /* nvdimm device hot unplug is not supported yet. */
+    assert(!object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM));
     mdev->is_removing = true;
     acpi_send_event(DEVICE(hotplug_dev), ACPI_MEMORY_HOTPLUG_STATUS);
 }
@@ -276,6 +289,8 @@ void acpi_memory_unplug_cb(MemHotplugState *mem_st,
         return;
     }
 
+    /* nvdimm device hot unplug is not supported yet. */
+    assert(!object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM));
     mdev->is_enabled = false;
     mdev->dimm = NULL;
 }
diff --git a/hw/acpi/nvdimm.c b/hw/acpi/nvdimm.c
index e486128aa1..602ec54485 100644
--- a/hw/acpi/nvdimm.c
+++ b/hw/acpi/nvdimm.c
@@ -289,8 +289,6 @@ static void
 nvdimm_build_structure_memdev(GArray *structures, DeviceState *dev)
 {
     NvdimmNfitMemDev *nfit_memdev;
-    uint64_t addr = object_property_get_int(OBJECT(dev), PC_DIMM_ADDR_PROP,
-                                            NULL);
     uint64_t size = object_property_get_int(OBJECT(dev), PC_DIMM_SIZE_PROP,
                                             NULL);
     int slot = object_property_get_int(OBJECT(dev), PC_DIMM_SLOT_PROP,
@@ -314,7 +312,8 @@ nvdimm_build_structure_memdev(GArray *structures, DeviceState *dev)
 
     /* The memory region on the device. */
     nfit_memdev->region_len = cpu_to_le64(size);
-    nfit_memdev->region_dpa = cpu_to_le64(addr);
+    /* The device address starts from 0. */
+    nfit_memdev->region_dpa = cpu_to_le64(0);
 
     /* Only one interleave for PMEM. */
     nfit_memdev->interleave_ways = cpu_to_le16(1);
@@ -349,8 +348,9 @@ static void nvdimm_build_structure_dcr(GArray *structures, DeviceState *dev)
                                          (DSM) in DSM Spec Rev1.*/);
 }
 
-static GArray *nvdimm_build_device_structure(GSList *device_list)
+static GArray *nvdimm_build_device_structure(void)
 {
+    GSList *device_list = nvdimm_get_plugged_device_list();
     GArray *structures = g_array_new(false, true /* clear */, 1);
 
     for (; device_list; device_list = device_list->next) {
@@ -368,28 +368,58 @@ static GArray *nvdimm_build_device_structure(GSList *device_list)
         /* build NVDIMM Control Region Structure. */
         nvdimm_build_structure_dcr(structures, dev);
     }
+    g_slist_free(device_list);
 
     return structures;
 }
 
-static void nvdimm_build_nfit(GSList *device_list, GArray *table_offsets,
+static void nvdimm_init_fit_buffer(NvdimmFitBuffer *fit_buf)
+{
+    qemu_mutex_init(&fit_buf->lock);
+    fit_buf->fit = g_array_new(false, true /* clear */, 1);
+}
+
+static void nvdimm_build_fit_buffer(NvdimmFitBuffer *fit_buf)
+{
+    qemu_mutex_lock(&fit_buf->lock);
+    g_array_free(fit_buf->fit, true);
+    fit_buf->fit = nvdimm_build_device_structure();
+    fit_buf->dirty = true;
+    qemu_mutex_unlock(&fit_buf->lock);
+}
+
+void nvdimm_acpi_hotplug(AcpiNVDIMMState *state)
+{
+    nvdimm_build_fit_buffer(&state->fit_buf);
+}
+
+static void nvdimm_build_nfit(AcpiNVDIMMState *state, GArray *table_offsets,
                               GArray *table_data, BIOSLinker *linker)
 {
-    GArray *structures = nvdimm_build_device_structure(device_list);
+    NvdimmFitBuffer *fit_buf = &state->fit_buf;
     unsigned int header;
 
+    qemu_mutex_lock(&fit_buf->lock);
+
+    /* NVDIMM device is not plugged? */
+    if (!fit_buf->fit->len) {
+        goto exit;
+    }
+
     acpi_add_table(table_offsets, table_data);
 
     /* NFIT header. */
     header = table_data->len;
     acpi_data_push(table_data, sizeof(NvdimmNfitHeader));
     /* NVDIMM device structures. */
-    g_array_append_vals(table_data, structures->data, structures->len);
+    g_array_append_vals(table_data, fit_buf->fit->data, fit_buf->fit->len);
 
     build_header(linker, table_data,
                  (void *)(table_data->data + header), "NFIT",
-                 sizeof(NvdimmNfitHeader) + structures->len, 1, NULL, NULL);
-    g_array_free(structures, true);
+                 sizeof(NvdimmNfitHeader) + fit_buf->fit->len, 1, NULL, NULL);
+
+exit:
+    qemu_mutex_unlock(&fit_buf->lock);
 }
 
 struct NvdimmDsmIn {
@@ -466,6 +496,22 @@ typedef struct NvdimmFuncSetLabelDataIn NvdimmFuncSetLabelDataIn;
 QEMU_BUILD_BUG_ON(sizeof(NvdimmFuncSetLabelDataIn) +
                   offsetof(NvdimmDsmIn, arg3) > 4096);
 
+struct NvdimmFuncReadFITIn {
+    uint32_t offset; /* the offset of FIT buffer. */
+} QEMU_PACKED;
+typedef struct NvdimmFuncReadFITIn NvdimmFuncReadFITIn;
+QEMU_BUILD_BUG_ON(sizeof(NvdimmFuncReadFITIn) +
+                  offsetof(NvdimmDsmIn, arg3) > 4096);
+
+struct NvdimmFuncReadFITOut {
+    /* the size of buffer filled by QEMU. */
+    uint32_t len;
+    uint32_t func_ret_status; /* return status code. */
+    uint8_t fit[0]; /* the FIT data. */
+} QEMU_PACKED;
+typedef struct NvdimmFuncReadFITOut NvdimmFuncReadFITOut;
+QEMU_BUILD_BUG_ON(sizeof(NvdimmFuncReadFITOut) > 4096);
+
 static void
 nvdimm_dsm_function0(uint32_t supported_func, hwaddr dsm_mem_addr)
 {
@@ -486,6 +532,74 @@ nvdimm_dsm_no_payload(uint32_t func_ret_status, hwaddr dsm_mem_addr)
     cpu_physical_memory_write(dsm_mem_addr, &out, sizeof(out));
 }
 
+#define NVDIMM_QEMU_RSVD_HANDLE_ROOT 0x10000
+
+/* Read FIT data, defined in docs/specs/acpi_nvdimm.txt. */
+static void nvdimm_dsm_func_read_fit(AcpiNVDIMMState *state, NvdimmDsmIn *in,
+                                     hwaddr dsm_mem_addr)
+{
+    NvdimmFitBuffer *fit_buf = &state->fit_buf;
+    NvdimmFuncReadFITIn *read_fit;
+    NvdimmFuncReadFITOut *read_fit_out;
+    GArray *fit;
+    uint32_t read_len = 0, func_ret_status;
+    int size;
+
+    read_fit = (NvdimmFuncReadFITIn *)in->arg3;
+    le32_to_cpus(&read_fit->offset);
+
+    qemu_mutex_lock(&fit_buf->lock);
+    fit = fit_buf->fit;
+
+    nvdimm_debug("Read FIT: offset %#x FIT size %#x Dirty %s.\n",
+                 read_fit->offset, fit->len, fit_buf->dirty ? "Yes" : "No");
+
+    if (read_fit->offset > fit->len) {
+        func_ret_status = 3 /* Invalid Input Parameters */;
+        goto exit;
+    }
+
+    /* It is the first time to read FIT. */
+    if (!read_fit->offset) {
+        fit_buf->dirty = false;
+    } else if (fit_buf->dirty) { /* FIT has been changed during RFIT. */
+        func_ret_status = 0x100 /* fit changed */;
+        goto exit;
+    }
+
+    func_ret_status = 0 /* Success */;
+    read_len = MIN(fit->len - read_fit->offset,
+                   4096 - sizeof(NvdimmFuncReadFITOut));
+
+exit:
+    size = sizeof(NvdimmFuncReadFITOut) + read_len;
+    read_fit_out = g_malloc(size);
+
+    read_fit_out->len = cpu_to_le32(size);
+    read_fit_out->func_ret_status = cpu_to_le32(func_ret_status);
+    memcpy(read_fit_out->fit, fit->data + read_fit->offset, read_len);
+
+    cpu_physical_memory_write(dsm_mem_addr, read_fit_out, size);
+
+    g_free(read_fit_out);
+    qemu_mutex_unlock(&fit_buf->lock);
+}
+
+static void nvdimm_dsm_reserved_root(AcpiNVDIMMState *state, NvdimmDsmIn *in,
+                                     hwaddr dsm_mem_addr)
+{
+    switch (in->function) {
+    case 0x0:
+        nvdimm_dsm_function0(0x1 | 1 << 1 /* Read FIT */, dsm_mem_addr);
+        return;
+    case 0x1 /*Read FIT */:
+        nvdimm_dsm_func_read_fit(state, in, dsm_mem_addr);
+        return;
+    }
+
+    nvdimm_dsm_no_payload(1 /* Not Supported */, dsm_mem_addr);
+}
+
 static void nvdimm_dsm_root(NvdimmDsmIn *in, hwaddr dsm_mem_addr)
 {
     /*
@@ -643,8 +757,8 @@ static void nvdimm_dsm_set_label_data(NVDIMMDevice *nvdimm, NvdimmDsmIn *in,
         return;
     }
 
-    assert(sizeof(*in) + sizeof(*set_label_data) + set_label_data->length <=
-           4096);
+    assert(offsetof(NvdimmDsmIn, arg3) +
+           sizeof(*set_label_data) + set_label_data->length <= 4096);
 
     nvc->write_label_data(nvdimm, set_label_data->in_buf,
                           set_label_data->length, set_label_data->offset);
@@ -712,6 +826,7 @@ nvdimm_dsm_read(void *opaque, hwaddr addr, unsigned size)
 static void
 nvdimm_dsm_write(void *opaque, hwaddr addr, uint64_t val, unsigned size)
 {
+    AcpiNVDIMMState *state = opaque;
     NvdimmDsmIn *in;
     hwaddr dsm_mem_addr = val;
 
@@ -739,6 +854,11 @@ nvdimm_dsm_write(void *opaque, hwaddr addr, uint64_t val, unsigned size)
         goto exit;
     }
 
+    if (in->handle == NVDIMM_QEMU_RSVD_HANDLE_ROOT) {
+        nvdimm_dsm_reserved_root(state, in, dsm_mem_addr);
+        goto exit;
+    }
+
      /* Handle 0 is reserved for NVDIMM Root Device. */
     if (!in->handle) {
         nvdimm_dsm_root(in, dsm_mem_addr);
@@ -772,23 +892,105 @@ void nvdimm_init_acpi_state(AcpiNVDIMMState *state, MemoryRegion *io,
     acpi_data_push(state->dsm_mem, sizeof(NvdimmDsmIn));
     fw_cfg_add_file(fw_cfg, NVDIMM_DSM_MEM_FILE, state->dsm_mem->data,
                     state->dsm_mem->len);
+
+    nvdimm_init_fit_buffer(&state->fit_buf);
 }
 
-#define NVDIMM_COMMON_DSM      "NCAL"
-#define NVDIMM_ACPI_MEM_ADDR   "MEMA"
+#define NVDIMM_COMMON_DSM       "NCAL"
+#define NVDIMM_ACPI_MEM_ADDR    "MEMA"
+
+#define NVDIMM_DSM_MEMORY       "NRAM"
+#define NVDIMM_DSM_IOPORT       "NPIO"
+
+#define NVDIMM_DSM_NOTIFY       "NTFI"
+#define NVDIMM_DSM_HANDLE       "HDLE"
+#define NVDIMM_DSM_REVISION     "REVS"
+#define NVDIMM_DSM_FUNCTION     "FUNC"
+#define NVDIMM_DSM_ARG3         "FARG"
+
+#define NVDIMM_DSM_OUT_BUF_SIZE "RLEN"
+#define NVDIMM_DSM_OUT_BUF      "ODAT"
+
+#define NVDIMM_DSM_RFIT_STATUS  "RSTA"
+
+#define NVDIMM_QEMU_RSVD_UUID   "648B9CF2-CDA1-4312-8AD9-49C4AF32BD62"
 
 static void nvdimm_build_common_dsm(Aml *dev)
 {
-    Aml *method, *ifctx, *function, *handle, *uuid, *dsm_mem, *result_size;
+    Aml *method, *ifctx, *function, *handle, *uuid, *dsm_mem, *elsectx2;
     Aml *elsectx, *unsupport, *unpatched, *expected_uuid, *uuid_invalid;
-    Aml *pckg, *pckg_index, *pckg_buf;
+    Aml *pckg, *pckg_index, *pckg_buf, *field, *dsm_out_buf, *dsm_out_buf_size;
     uint8_t byte_list[1];
 
     method = aml_method(NVDIMM_COMMON_DSM, 5, AML_SERIALIZED);
     uuid = aml_arg(0);
     function = aml_arg(2);
     handle = aml_arg(4);
-    dsm_mem = aml_name(NVDIMM_ACPI_MEM_ADDR);
+    dsm_mem = aml_local(6);
+    dsm_out_buf = aml_local(7);
+
+    aml_append(method, aml_store(aml_name(NVDIMM_ACPI_MEM_ADDR), dsm_mem));
+
+    /* map DSM memory and IO into ACPI namespace. */
+    aml_append(method, aml_operation_region(NVDIMM_DSM_IOPORT, AML_SYSTEM_IO,
+               aml_int(NVDIMM_ACPI_IO_BASE), NVDIMM_ACPI_IO_LEN));
+    aml_append(method, aml_operation_region(NVDIMM_DSM_MEMORY,
+               AML_SYSTEM_MEMORY, dsm_mem, sizeof(NvdimmDsmIn)));
+
+    /*
+     * DSM notifier:
+     * NVDIMM_DSM_NOTIFY: write the address of DSM memory and notify QEMU to
+     *                    emulate the access.
+     *
+     * It is the IO port so that accessing them will cause VM-exit, the
+     * control will be transferred to QEMU.
+     */
+    field = aml_field(NVDIMM_DSM_IOPORT, AML_DWORD_ACC, AML_NOLOCK,
+                      AML_PRESERVE);
+    aml_append(field, aml_named_field(NVDIMM_DSM_NOTIFY,
+               sizeof(uint32_t) * BITS_PER_BYTE));
+    aml_append(method, field);
+
+    /*
+     * DSM input:
+     * NVDIMM_DSM_HANDLE: store device's handle, it's zero if the _DSM call
+     *                    happens on NVDIMM Root Device.
+     * NVDIMM_DSM_REVISION: store the Arg1 of _DSM call.
+     * NVDIMM_DSM_FUNCTION: store the Arg2 of _DSM call.
+     * NVDIMM_DSM_ARG3: store the Arg3 of _DSM call which is a Package
+     *                  containing function-specific arguments.
+     *
+     * They are RAM mapping on host so that these accesses never cause
+     * VM-EXIT.
+     */
+    field = aml_field(NVDIMM_DSM_MEMORY, AML_DWORD_ACC, AML_NOLOCK,
+                      AML_PRESERVE);
+    aml_append(field, aml_named_field(NVDIMM_DSM_HANDLE,
+               sizeof(typeof_field(NvdimmDsmIn, handle)) * BITS_PER_BYTE));
+    aml_append(field, aml_named_field(NVDIMM_DSM_REVISION,
+               sizeof(typeof_field(NvdimmDsmIn, revision)) * BITS_PER_BYTE));
+    aml_append(field, aml_named_field(NVDIMM_DSM_FUNCTION,
+               sizeof(typeof_field(NvdimmDsmIn, function)) * BITS_PER_BYTE));
+    aml_append(field, aml_named_field(NVDIMM_DSM_ARG3,
+         (sizeof(NvdimmDsmIn) - offsetof(NvdimmDsmIn, arg3)) * BITS_PER_BYTE));
+    aml_append(method, field);
+
+    /*
+     * DSM output:
+     * NVDIMM_DSM_OUT_BUF_SIZE: the size of the buffer filled by QEMU.
+     * NVDIMM_DSM_OUT_BUF: the buffer QEMU uses to store the result.
+     *
+     * Since the page is reused by both input and out, the input data
+     * will be lost after storing new result into ODAT so we should fetch
+     * all the input data before writing the result.
+     */
+    field = aml_field(NVDIMM_DSM_MEMORY, AML_DWORD_ACC, AML_NOLOCK,
+                      AML_PRESERVE);
+    aml_append(field, aml_named_field(NVDIMM_DSM_OUT_BUF_SIZE,
+               sizeof(typeof_field(NvdimmDsmOut, len)) * BITS_PER_BYTE));
+    aml_append(field, aml_named_field(NVDIMM_DSM_OUT_BUF,
+       (sizeof(NvdimmDsmOut) - offsetof(NvdimmDsmOut, data)) * BITS_PER_BYTE));
+    aml_append(method, field);
 
     /*
      * do not support any method if DSM memory address has not been
@@ -804,9 +1006,15 @@ static void nvdimm_build_common_dsm(Aml *dev)
                /* UUID for NVDIMM Root Device */, expected_uuid));
     aml_append(method, ifctx);
     elsectx = aml_else();
-    aml_append(elsectx, aml_store(
+    ifctx = aml_if(aml_equal(handle, aml_int(NVDIMM_QEMU_RSVD_HANDLE_ROOT)));
+    aml_append(ifctx, aml_store(aml_touuid(NVDIMM_QEMU_RSVD_UUID
+               /* UUID for QEMU internal use */), expected_uuid));
+    aml_append(elsectx, ifctx);
+    elsectx2 = aml_else();
+    aml_append(elsectx2, aml_store(
                aml_touuid("4309AC30-0D11-11E4-9191-0800200C9A66")
                /* UUID for NVDIMM Devices */, expected_uuid));
+    aml_append(elsectx, elsectx2);
     aml_append(method, elsectx);
 
     uuid_invalid = aml_lnot(aml_equal(uuid, expected_uuid));
@@ -832,9 +1040,9 @@ static void nvdimm_build_common_dsm(Aml *dev)
      * it reserves 0 for root device and is the handle for NVDIMM devices.
      * See the comments in nvdimm_slot_to_handle().
      */
-    aml_append(method, aml_store(handle, aml_name("HDLE")));
-    aml_append(method, aml_store(aml_arg(1), aml_name("REVS")));
-    aml_append(method, aml_store(aml_arg(2), aml_name("FUNC")));
+    aml_append(method, aml_store(handle, aml_name(NVDIMM_DSM_HANDLE)));
+    aml_append(method, aml_store(aml_arg(1), aml_name(NVDIMM_DSM_REVISION)));
+    aml_append(method, aml_store(aml_arg(2), aml_name(NVDIMM_DSM_FUNCTION)));
 
     /*
      * The fourth parameter (Arg3) of _DSM is a package which contains
@@ -852,24 +1060,26 @@ static void nvdimm_build_common_dsm(Aml *dev)
     pckg_buf = aml_local(3);
     aml_append(ifctx, aml_store(aml_index(pckg, aml_int(0)), pckg_index));
     aml_append(ifctx, aml_store(aml_derefof(pckg_index), pckg_buf));
-    aml_append(ifctx, aml_store(pckg_buf, aml_name("ARG3")));
+    aml_append(ifctx, aml_store(pckg_buf, aml_name(NVDIMM_DSM_ARG3)));
     aml_append(method, ifctx);
 
     /*
      * tell QEMU about the real address of DSM memory, then QEMU
      * gets the control and fills the result in DSM memory.
      */
-    aml_append(method, aml_store(dsm_mem, aml_name("NTFI")));
-
-    result_size = aml_local(1);
-    aml_append(method, aml_store(aml_name("RLEN"), result_size));
-    aml_append(method, aml_store(aml_shiftleft(result_size, aml_int(3)),
-                                 result_size));
-    aml_append(method, aml_create_field(aml_name("ODAT"), aml_int(0),
-                                        result_size, "OBUF"));
+    aml_append(method, aml_store(dsm_mem, aml_name(NVDIMM_DSM_NOTIFY)));
+
+    dsm_out_buf_size = aml_local(1);
+    /* RLEN is not included in the payload returned to guest. */
+    aml_append(method, aml_subtract(aml_name(NVDIMM_DSM_OUT_BUF_SIZE),
+               aml_int(4), dsm_out_buf_size));
+    aml_append(method, aml_store(aml_shiftleft(dsm_out_buf_size, aml_int(3)),
+                                 dsm_out_buf_size));
+    aml_append(method, aml_create_field(aml_name(NVDIMM_DSM_OUT_BUF),
+               aml_int(0), dsm_out_buf_size, "OBUF"));
     aml_append(method, aml_concatenate(aml_buffer(0, NULL), aml_name("OBUF"),
-                                       aml_arg(6)));
-    aml_append(method, aml_return(aml_arg(6)));
+                                       dsm_out_buf));
+    aml_append(method, aml_return(dsm_out_buf));
     aml_append(dev, method);
 }
 
@@ -884,12 +1094,110 @@ static void nvdimm_build_device_dsm(Aml *dev, uint32_t handle)
     aml_append(dev, method);
 }
 
-static void nvdimm_build_nvdimm_devices(GSList *device_list, Aml *root_dev)
+static void nvdimm_build_fit(Aml *dev)
 {
-    for (; device_list; device_list = device_list->next) {
-        DeviceState *dev = device_list->data;
-        int slot = object_property_get_int(OBJECT(dev), PC_DIMM_SLOT_PROP,
-                                           NULL);
+    Aml *method, *pkg, *buf, *buf_size, *offset, *call_result;
+    Aml *whilectx, *ifcond, *ifctx, *elsectx, *fit;
+
+    buf = aml_local(0);
+    buf_size = aml_local(1);
+    fit = aml_local(2);
+
+    aml_append(dev, aml_create_dword_field(aml_buffer(4, NULL),
+               aml_int(0), NVDIMM_DSM_RFIT_STATUS));
+
+    /* build helper function, RFIT. */
+    method = aml_method("RFIT", 1, AML_SERIALIZED);
+    aml_append(method, aml_create_dword_field(aml_buffer(4, NULL),
+                                              aml_int(0), "OFST"));
+
+    /* prepare input package. */
+    pkg = aml_package(1);
+    aml_append(method, aml_store(aml_arg(0), aml_name("OFST")));
+    aml_append(pkg, aml_name("OFST"));
+
+    /* call Read_FIT function. */
+    call_result = aml_call5(NVDIMM_COMMON_DSM,
+                            aml_touuid(NVDIMM_QEMU_RSVD_UUID),
+                            aml_int(1) /* Revision 1 */,
+                            aml_int(0x1) /* Read FIT */,
+                            pkg, aml_int(NVDIMM_QEMU_RSVD_HANDLE_ROOT));
+    aml_append(method, aml_store(call_result, buf));
+
+    /* handle _DSM result. */
+    aml_append(method, aml_create_dword_field(buf,
+               aml_int(0) /* offset at byte 0 */, "STAU"));
+
+    aml_append(method, aml_store(aml_name("STAU"),
+                                 aml_name(NVDIMM_DSM_RFIT_STATUS)));
+
+     /* if something is wrong during _DSM. */
+    ifcond = aml_equal(aml_int(0 /* Success */), aml_name("STAU"));
+    ifctx = aml_if(aml_lnot(ifcond));
+    aml_append(ifctx, aml_return(aml_buffer(0, NULL)));
+    aml_append(method, ifctx);
+
+    aml_append(method, aml_store(aml_sizeof(buf), buf_size));
+    aml_append(method, aml_subtract(buf_size,
+                                    aml_int(4) /* the size of "STAU" */,
+                                    buf_size));
+
+    /* if we read the end of fit. */
+    ifctx = aml_if(aml_equal(buf_size, aml_int(0)));
+    aml_append(ifctx, aml_return(aml_buffer(0, NULL)));
+    aml_append(method, ifctx);
+
+    aml_append(method, aml_store(aml_shiftleft(buf_size, aml_int(3)),
+                                 buf_size));
+    aml_append(method, aml_create_field(buf,
+                            aml_int(4 * BITS_PER_BYTE), /* offset at byte 4.*/
+                            buf_size, "BUFF"));
+    aml_append(method, aml_return(aml_name("BUFF")));
+    aml_append(dev, method);
+
+    /* build _FIT. */
+    method = aml_method("_FIT", 0, AML_SERIALIZED);
+    offset = aml_local(3);
+
+    aml_append(method, aml_store(aml_buffer(0, NULL), fit));
+    aml_append(method, aml_store(aml_int(0), offset));
+
+    whilectx = aml_while(aml_int(1));
+    aml_append(whilectx, aml_store(aml_call1("RFIT", offset), buf));
+    aml_append(whilectx, aml_store(aml_sizeof(buf), buf_size));
+
+    /*
+     * if fit buffer was changed during RFIT, read from the beginning
+     * again.
+     */
+    ifctx = aml_if(aml_equal(aml_name(NVDIMM_DSM_RFIT_STATUS),
+                             aml_int(0x100 /* fit changed */)));
+    aml_append(ifctx, aml_store(aml_buffer(0, NULL), fit));
+    aml_append(ifctx, aml_store(aml_int(0), offset));
+    aml_append(whilectx, ifctx);
+
+    elsectx = aml_else();
+
+    /* finish fit read if no data is read out. */
+    ifctx = aml_if(aml_equal(buf_size, aml_int(0)));
+    aml_append(ifctx, aml_return(fit));
+    aml_append(elsectx, ifctx);
+
+    /* update the offset. */
+    aml_append(elsectx, aml_add(offset, buf_size, offset));
+    /* append the data we read out to the fit buffer. */
+    aml_append(elsectx, aml_concatenate(fit, buf, fit));
+    aml_append(whilectx, elsectx);
+    aml_append(method, whilectx);
+
+    aml_append(dev, method);
+}
+
+static void nvdimm_build_nvdimm_devices(Aml *root_dev, uint32_t ram_slots)
+{
+    uint32_t slot;
+
+    for (slot = 0; slot < ram_slots; slot++) {
         uint32_t handle = nvdimm_slot_to_handle(slot);
         Aml *nvdimm_dev;
 
@@ -910,11 +1218,11 @@ static void nvdimm_build_nvdimm_devices(GSList *device_list, Aml *root_dev)
     }
 }
 
-static void nvdimm_build_ssdt(GSList *device_list, GArray *table_offsets,
-                              GArray *table_data, BIOSLinker *linker,
-                              GArray *dsm_dma_arrea)
+static void nvdimm_build_ssdt(GArray *table_offsets, GArray *table_data,
+                              BIOSLinker *linker, GArray *dsm_dma_arrea,
+                              uint32_t ram_slots)
 {
-    Aml *ssdt, *sb_scope, *dev, *field;
+    Aml *ssdt, *sb_scope, *dev;
     int mem_addr_offset, nvdimm_ssdt;
 
     acpi_add_table(table_offsets, table_data);
@@ -939,69 +1247,13 @@ static void nvdimm_build_ssdt(GSList *device_list, GArray *table_offsets,
      */
     aml_append(dev, aml_name_decl("_HID", aml_string("ACPI0012")));
 
-    /* map DSM memory and IO into ACPI namespace. */
-    aml_append(dev, aml_operation_region("NPIO", AML_SYSTEM_IO,
-               aml_int(NVDIMM_ACPI_IO_BASE), NVDIMM_ACPI_IO_LEN));
-    aml_append(dev, aml_operation_region("NRAM", AML_SYSTEM_MEMORY,
-               aml_name(NVDIMM_ACPI_MEM_ADDR), sizeof(NvdimmDsmIn)));
-
-    /*
-     * DSM notifier:
-     * NTFI: write the address of DSM memory and notify QEMU to emulate
-     *       the access.
-     *
-     * It is the IO port so that accessing them will cause VM-exit, the
-     * control will be transferred to QEMU.
-     */
-    field = aml_field("NPIO", AML_DWORD_ACC, AML_NOLOCK, AML_PRESERVE);
-    aml_append(field, aml_named_field("NTFI",
-               sizeof(uint32_t) * BITS_PER_BYTE));
-    aml_append(dev, field);
-
-    /*
-     * DSM input:
-     * HDLE: store device's handle, it's zero if the _DSM call happens
-     *       on NVDIMM Root Device.
-     * REVS: store the Arg1 of _DSM call.
-     * FUNC: store the Arg2 of _DSM call.
-     * ARG3: store the Arg3 of _DSM call.
-     *
-     * They are RAM mapping on host so that these accesses never cause
-     * VM-EXIT.
-     */
-    field = aml_field("NRAM", AML_DWORD_ACC, AML_NOLOCK, AML_PRESERVE);
-    aml_append(field, aml_named_field("HDLE",
-               sizeof(typeof_field(NvdimmDsmIn, handle)) * BITS_PER_BYTE));
-    aml_append(field, aml_named_field("REVS",
-               sizeof(typeof_field(NvdimmDsmIn, revision)) * BITS_PER_BYTE));
-    aml_append(field, aml_named_field("FUNC",
-               sizeof(typeof_field(NvdimmDsmIn, function)) * BITS_PER_BYTE));
-    aml_append(field, aml_named_field("ARG3",
-               (sizeof(NvdimmDsmIn) - offsetof(NvdimmDsmIn, arg3)) * BITS_PER_BYTE));
-    aml_append(dev, field);
-
-    /*
-     * DSM output:
-     * RLEN: the size of the buffer filled by QEMU.
-     * ODAT: the buffer QEMU uses to store the result.
-     *
-     * Since the page is reused by both input and out, the input data
-     * will be lost after storing new result into ODAT so we should fetch
-     * all the input data before writing the result.
-     */
-    field = aml_field("NRAM", AML_DWORD_ACC, AML_NOLOCK, AML_PRESERVE);
-    aml_append(field, aml_named_field("RLEN",
-               sizeof(typeof_field(NvdimmDsmOut, len)) * BITS_PER_BYTE));
-    aml_append(field, aml_named_field("ODAT",
-               (sizeof(NvdimmDsmOut) - offsetof(NvdimmDsmOut, data)) * BITS_PER_BYTE));
-    aml_append(dev, field);
-
     nvdimm_build_common_dsm(dev);
 
     /* 0 is reserved for root device. */
     nvdimm_build_device_dsm(dev, 0);
+    nvdimm_build_fit(dev);
 
-    nvdimm_build_nvdimm_devices(device_list, dev);
+    nvdimm_build_nvdimm_devices(dev, ram_slots);
 
     aml_append(sb_scope, dev);
     aml_append(ssdt, sb_scope);
@@ -1026,17 +1278,17 @@ static void nvdimm_build_ssdt(GSList *device_list, GArray *table_offsets,
 }
 
 void nvdimm_build_acpi(GArray *table_offsets, GArray *table_data,
-                       BIOSLinker *linker, GArray *dsm_dma_arrea)
+                       BIOSLinker *linker, AcpiNVDIMMState *state,
+                       uint32_t ram_slots)
 {
-    GSList *device_list;
+    nvdimm_build_nfit(state, table_offsets, table_data, linker);
 
-    /* no NVDIMM device is plugged. */
-    device_list = nvdimm_get_plugged_device_list();
-    if (!device_list) {
-        return;
+    /*
+     * NVDIMM device is allowed to be plugged only if there is available
+     * slot.
+     */
+    if (ram_slots) {
+        nvdimm_build_ssdt(table_offsets, table_data, linker, state->dsm_mem,
+                          ram_slots);
     }
-    nvdimm_build_nfit(device_list, table_offsets, table_data, linker);
-    nvdimm_build_ssdt(device_list, table_offsets, table_data, linker,
-                      dsm_dma_arrea);
-    g_slist_free(device_list);
 }
diff --git a/hw/block/dataplane/virtio-blk.c b/hw/block/dataplane/virtio-blk.c
index 704a763603..90ef557c8c 100644
--- a/hw/block/dataplane/virtio-blk.c
+++ b/hw/block/dataplane/virtio-blk.c
@@ -88,23 +88,28 @@ void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *conf,
 
     *dataplane = NULL;
 
-    if (!conf->iothread) {
-        return;
-    }
+    if (conf->iothread) {
+        if (!k->set_guest_notifiers || !k->ioeventfd_assign) {
+            error_setg(errp,
+                       "device is incompatible with iothread "
+                       "(transport does not support notifiers)");
+            return;
+        }
+        if (!virtio_device_ioeventfd_enabled(vdev)) {
+            error_setg(errp, "ioeventfd is required for iothread");
+            return;
+        }
 
-    /* Don't try if transport does not support notifiers. */
-    if (!k->set_guest_notifiers || !k->ioeventfd_started) {
-        error_setg(errp,
-                   "device is incompatible with dataplane "
-                   "(transport does not support notifiers)");
-        return;
+        /* If dataplane is (re-)enabled while the guest is running there could
+         * be block jobs that can conflict.
+         */
+        if (blk_op_is_blocked(conf->conf.blk, BLOCK_OP_TYPE_DATAPLANE, errp)) {
+            error_prepend(errp, "cannot start virtio-blk dataplane: ");
+            return;
+        }
     }
-
-    /* If dataplane is (re-)enabled while the guest is running there could be
-     * block jobs that can conflict.
-     */
-    if (blk_op_is_blocked(conf->conf.blk, BLOCK_OP_TYPE_DATAPLANE, errp)) {
-        error_prepend(errp, "cannot start dataplane thread: ");
+    /* Don't try if transport does not support notifiers. */
+    if (!virtio_device_ioeventfd_enabled(vdev)) {
         return;
     }
 
@@ -112,9 +117,13 @@ void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *conf,
     s->vdev = vdev;
     s->conf = conf;
 
-    s->iothread = conf->iothread;
-    object_ref(OBJECT(s->iothread));
-    s->ctx = iothread_get_aio_context(s->iothread);
+    if (conf->iothread) {
+        s->iothread = conf->iothread;
+        object_ref(OBJECT(s->iothread));
+        s->ctx = iothread_get_aio_context(s->iothread);
+    } else {
+        s->ctx = qemu_get_aio_context();
+    }
     s->bh = aio_bh_new(s->ctx, notify_guest_bh, s);
     s->batch_notify_vqs = bitmap_new(conf->num_queues);
 
@@ -124,14 +133,19 @@ void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *conf,
 /* Context: QEMU global mutex held */
 void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s)
 {
+    VirtIOBlock *vblk;
+
     if (!s) {
         return;
     }
 
-    virtio_blk_data_plane_stop(s);
+    vblk = VIRTIO_BLK(s->vdev);
+    assert(!vblk->dataplane_started);
     g_free(s->batch_notify_vqs);
     qemu_bh_delete(s->bh);
-    object_unref(OBJECT(s->iothread));
+    if (s->iothread) {
+        object_unref(OBJECT(s->iothread));
+    }
     g_free(s);
 }
 
@@ -147,17 +161,18 @@ static void virtio_blk_data_plane_handle_output(VirtIODevice *vdev,
 }
 
 /* Context: QEMU global mutex held */
-void virtio_blk_data_plane_start(VirtIOBlockDataPlane *s)
+int virtio_blk_data_plane_start(VirtIODevice *vdev)
 {
-    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s->vdev)));
+    VirtIOBlock *vblk = VIRTIO_BLK(vdev);
+    VirtIOBlockDataPlane *s = vblk->dataplane;
+    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(vblk)));
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
-    VirtIOBlock *vblk = VIRTIO_BLK(s->vdev);
     unsigned i;
     unsigned nvqs = s->conf->num_queues;
     int r;
 
     if (vblk->dataplane_started || s->starting) {
-        return;
+        return 0;
     }
 
     s->starting = true;
@@ -204,20 +219,22 @@ void virtio_blk_data_plane_start(VirtIOBlockDataPlane *s)
                 virtio_blk_data_plane_handle_output);
     }
     aio_context_release(s->ctx);
-    return;
+    return 0;
 
   fail_guest_notifiers:
     vblk->dataplane_disabled = true;
     s->starting = false;
     vblk->dataplane_started = true;
+    return -ENOSYS;
 }
 
 /* Context: QEMU global mutex held */
-void virtio_blk_data_plane_stop(VirtIOBlockDataPlane *s)
+void virtio_blk_data_plane_stop(VirtIODevice *vdev)
 {
-    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s->vdev)));
+    VirtIOBlock *vblk = VIRTIO_BLK(vdev);
+    VirtIOBlockDataPlane *s = vblk->dataplane;
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vblk));
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
-    VirtIOBlock *vblk = VIRTIO_BLK(s->vdev);
     unsigned i;
     unsigned nvqs = s->conf->num_queues;
 
diff --git a/hw/block/dataplane/virtio-blk.h b/hw/block/dataplane/virtio-blk.h
index b1f0b95b32..db3f47b173 100644
--- a/hw/block/dataplane/virtio-blk.h
+++ b/hw/block/dataplane/virtio-blk.h
@@ -23,9 +23,9 @@ void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *conf,
                                   VirtIOBlockDataPlane **dataplane,
                                   Error **errp);
 void virtio_blk_data_plane_destroy(VirtIOBlockDataPlane *s);
-void virtio_blk_data_plane_start(VirtIOBlockDataPlane *s);
-void virtio_blk_data_plane_stop(VirtIOBlockDataPlane *s);
-void virtio_blk_data_plane_drain(VirtIOBlockDataPlane *s);
 void virtio_blk_data_plane_notify(VirtIOBlockDataPlane *s, VirtQueue *vq);
 
+int virtio_blk_data_plane_start(VirtIODevice *vdev);
+void virtio_blk_data_plane_stop(VirtIODevice *vdev);
+
 #endif /* HW_DATAPLANE_VIRTIO_BLK_H */
diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
index 37fe72bdcd..0c5fd27593 100644
--- a/hw/block/virtio-blk.c
+++ b/hw/block/virtio-blk.c
@@ -611,7 +611,7 @@ static void virtio_blk_handle_output(VirtIODevice *vdev, VirtQueue *vq)
         /* Some guests kick before setting VIRTIO_CONFIG_S_DRIVER_OK so start
          * dataplane here instead of waiting for .set_status().
          */
-        virtio_blk_data_plane_start(s->dataplane);
+        virtio_device_start_ioeventfd(vdev);
         if (!s->dataplane_disabled) {
             return;
         }
@@ -687,11 +687,9 @@ static void virtio_blk_reset(VirtIODevice *vdev)
         virtio_blk_free_request(req);
     }
 
-    if (s->dataplane) {
-        virtio_blk_data_plane_stop(s->dataplane);
-    }
     aio_context_release(ctx);
 
+    assert(!s->dataplane_started);
     blk_set_enable_write_cache(s->blk, s->original_wce);
 }
 
@@ -789,9 +787,8 @@ static void virtio_blk_set_status(VirtIODevice *vdev, uint8_t status)
 {
     VirtIOBlock *s = VIRTIO_BLK(vdev);
 
-    if (s->dataplane && !(status & (VIRTIO_CONFIG_S_DRIVER |
-                                    VIRTIO_CONFIG_S_DRIVER_OK))) {
-        virtio_blk_data_plane_stop(s->dataplane);
+    if (!(status & (VIRTIO_CONFIG_S_DRIVER | VIRTIO_CONFIG_S_DRIVER_OK))) {
+        assert(!s->dataplane_started);
     }
 
     if (!(status & VIRTIO_CONFIG_S_DRIVER_OK)) {
@@ -919,7 +916,7 @@ static void virtio_blk_device_realize(DeviceState *dev, Error **errp)
     s->sector_mask = (s->conf.conf.logical_block_size / BDRV_SECTOR_SIZE) - 1;
 
     for (i = 0; i < conf->num_queues; i++) {
-        virtio_add_queue_aio(vdev, 128, virtio_blk_handle_output);
+        virtio_add_queue(vdev, 128, virtio_blk_handle_output);
     }
     virtio_blk_data_plane_create(vdev, conf, &s->dataplane, &err);
     if (err != NULL) {
@@ -1002,6 +999,8 @@ static void virtio_blk_class_init(ObjectClass *klass, void *data)
     vdc->reset = virtio_blk_reset;
     vdc->save = virtio_blk_save_device;
     vdc->load = virtio_blk_load_device;
+    vdc->start_ioeventfd = virtio_blk_data_plane_start;
+    vdc->stop_ioeventfd = virtio_blk_data_plane_stop;
 }
 
 static const TypeInfo virtio_blk_info = {
diff --git a/hw/core/hotplug.c b/hw/core/hotplug.c
index 17ac986685..ab34c19461 100644
--- a/hw/core/hotplug.c
+++ b/hw/core/hotplug.c
@@ -35,6 +35,17 @@ void hotplug_handler_plug(HotplugHandler *plug_handler,
     }
 }
 
+void hotplug_handler_post_plug(HotplugHandler *plug_handler,
+                               DeviceState *plugged_dev,
+                               Error **errp)
+{
+    HotplugHandlerClass *hdc = HOTPLUG_HANDLER_GET_CLASS(plug_handler);
+
+    if (hdc->post_plug) {
+        hdc->post_plug(plug_handler, plugged_dev, errp);
+    }
+}
+
 void hotplug_handler_unplug_request(HotplugHandler *plug_handler,
                                     DeviceState *plugged_dev,
                                     Error **errp)
diff --git a/hw/core/qdev.c b/hw/core/qdev.c
index 57834423b9..d835e6259a 100644
--- a/hw/core/qdev.c
+++ b/hw/core/qdev.c
@@ -945,10 +945,21 @@ static void device_set_realized(Object *obj, bool value, Error **errp)
                 goto child_realize_fail;
             }
         }
+
         if (dev->hotplugged) {
             device_reset(dev);
         }
         dev->pending_deleted_event = false;
+        dev->realized = value;
+
+        if (hotplug_ctrl) {
+            hotplug_handler_post_plug(hotplug_ctrl, dev, &local_err);
+        }
+
+        if (local_err != NULL) {
+            dev->realized = value;
+            goto post_realize_fail;
+        }
     } else if (!value && dev->realized) {
         Error **local_errp = NULL;
         QLIST_FOREACH(bus, &dev->child_bus, sibling) {
@@ -965,13 +976,14 @@ static void device_set_realized(Object *obj, bool value, Error **errp)
         }
         dev->pending_deleted_event = true;
         DEVICE_LISTENER_CALL(unrealize, Reverse, dev);
-    }
 
-    if (local_err != NULL) {
-        goto fail;
+        if (local_err != NULL) {
+            goto fail;
+        }
+
+        dev->realized = value;
     }
 
-    dev->realized = value;
     return;
 
 child_realize_fail:
diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
index 5cd1da9a87..ce9cc93af4 100644
--- a/hw/i386/acpi-build.c
+++ b/hw/i386/acpi-build.c
@@ -2068,6 +2068,13 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         method = aml_method("_E03", 0, AML_NOTSERIALIZED);
         aml_append(method, aml_call0(MEMORY_HOTPLUG_HANDLER_PATH));
         aml_append(scope, method);
+
+        if (pcms->acpi_nvdimm_state.is_enabled) {
+            method = aml_method("_E04", 0, AML_NOTSERIALIZED);
+            aml_append(method, aml_notify(aml_name("\\_SB.NVDR"),
+                                          aml_int(0x80)));
+            aml_append(scope, method);
+        }
     }
     aml_append(dsdt, scope);
 
@@ -2810,7 +2817,7 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)
     }
     if (pcms->acpi_nvdimm_state.is_enabled) {
         nvdimm_build_acpi(table_offsets, tables_blob, tables->linker,
-                          pcms->acpi_nvdimm_state.dsm_mem);
+                          &pcms->acpi_nvdimm_state, machine->ram_slots);
     }
 
     /* Add tables supplied by user (if any) */
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index f56ea0f87b..c011552ac4 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -1721,6 +1721,16 @@ out:
     error_propagate(errp, local_err);
 }
 
+static void pc_dimm_post_plug(HotplugHandler *hotplug_dev,
+                              DeviceState *dev, Error **errp)
+{
+    PCMachineState *pcms = PC_MACHINE(hotplug_dev);
+
+    if (object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM)) {
+        nvdimm_acpi_hotplug(&pcms->acpi_nvdimm_state);
+    }
+}
+
 static void pc_dimm_unplug_request(HotplugHandler *hotplug_dev,
                                    DeviceState *dev, Error **errp)
 {
@@ -1734,6 +1744,12 @@ static void pc_dimm_unplug_request(HotplugHandler *hotplug_dev,
         goto out;
     }
 
+    if (object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM)) {
+        error_setg(&local_err,
+                   "nvdimm device hot unplug is not supported yet.");
+        goto out;
+    }
+
     hhc = HOTPLUG_HANDLER_GET_CLASS(pcms->acpi_dev);
     hhc->unplug_request(HOTPLUG_HANDLER(pcms->acpi_dev), dev, &local_err);
 
@@ -1751,6 +1767,12 @@ static void pc_dimm_unplug(HotplugHandler *hotplug_dev,
     HotplugHandlerClass *hhc;
     Error *local_err = NULL;
 
+    if (object_dynamic_cast(OBJECT(dev), TYPE_NVDIMM)) {
+        error_setg(&local_err,
+                   "nvdimm device hot unplug is not supported yet.");
+        goto out;
+    }
+
     hhc = HOTPLUG_HANDLER_GET_CLASS(pcms->acpi_dev);
     hhc->unplug(HOTPLUG_HANDLER(pcms->acpi_dev), dev, &local_err);
 
@@ -1986,6 +2008,14 @@ static void pc_machine_device_plug_cb(HotplugHandler *hotplug_dev,
     }
 }
 
+static void pc_machine_device_post_plug_cb(HotplugHandler *hotplug_dev,
+                                           DeviceState *dev, Error **errp)
+{
+    if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
+        pc_dimm_post_plug(hotplug_dev, dev, errp);
+    }
+}
+
 static void pc_machine_device_unplug_request_cb(HotplugHandler *hotplug_dev,
                                                 DeviceState *dev, Error **errp)
 {
@@ -2290,6 +2320,7 @@ static void pc_machine_class_init(ObjectClass *oc, void *data)
     mc->reset = pc_machine_reset;
     hc->pre_plug = pc_machine_device_pre_plug_cb;
     hc->plug = pc_machine_device_plug_cb;
+    hc->post_plug = pc_machine_device_post_plug_cb;
     hc->unplug_request = pc_machine_device_unplug_request_cb;
     hc->unplug = pc_machine_device_unplug_cb;
     nc->nmi_monitor_handler = x86_nmi;
diff --git a/hw/ipmi/Makefile.objs b/hw/ipmi/Makefile.objs
index a90318d5ba..1b422bbee0 100644
--- a/hw/ipmi/Makefile.objs
+++ b/hw/ipmi/Makefile.objs
@@ -1,5 +1,5 @@
 common-obj-$(CONFIG_IPMI) += ipmi.o
 common-obj-$(CONFIG_IPMI_LOCAL) += ipmi_bmc_sim.o
-common-obj-$(CONFIG_IPMI_LOCAL) += ipmi_bmc_extern.o
+common-obj-$(CONFIG_IPMI_EXTERN) += ipmi_bmc_extern.o
 common-obj-$(CONFIG_ISA_IPMI_KCS) += isa_ipmi_kcs.o
 common-obj-$(CONFIG_ISA_IPMI_BT) += isa_ipmi_bt.o
diff --git a/hw/ipmi/ipmi.c b/hw/ipmi/ipmi.c
index f09f217e78..5cf1caa88a 100644
--- a/hw/ipmi/ipmi.c
+++ b/hw/ipmi/ipmi.c
@@ -51,7 +51,7 @@ static int ipmi_do_hw_op(IPMIInterface *s, enum ipmi_op op, int checkonly)
         if (checkonly) {
             return 0;
         }
-        qemu_system_powerdown_request();
+        qemu_system_shutdown_request();
         return 0;
 
     case IPMI_SEND_NMI:
@@ -61,9 +61,15 @@ static int ipmi_do_hw_op(IPMIInterface *s, enum ipmi_op op, int checkonly)
         qmp_inject_nmi(NULL);
         return 0;
 
+    case IPMI_SHUTDOWN_VIA_ACPI_OVERTEMP:
+        if (checkonly) {
+            return 0;
+        }
+        qemu_system_powerdown_request();
+        return 0;
+
     case IPMI_POWERCYCLE_CHASSIS:
     case IPMI_PULSE_DIAG_IRQ:
-    case IPMI_SHUTDOWN_VIA_ACPI_OVERTEMP:
     case IPMI_POWERON_CHASSIS:
     default:
         return IPMI_CC_COMMAND_NOT_SUPPORTED;
diff --git a/hw/ipmi/ipmi_bmc_extern.c b/hw/ipmi/ipmi_bmc_extern.c
index 4b310e5eff..e8e3d250b6 100644
--- a/hw/ipmi/ipmi_bmc_extern.c
+++ b/hw/ipmi/ipmi_bmc_extern.c
@@ -54,7 +54,8 @@
 #define   VM_CAPABILITIES_IRQ      0x04
 #define   VM_CAPABILITIES_NMI      0x08
 #define   VM_CAPABILITIES_ATTN     0x10
-#define VM_CMD_FORCEOFF            0x09
+#define   VM_CAPABILITIES_GRACEFUL_SHUTDOWN 0x20
+#define VM_CMD_GRACEFUL_SHUTDOWN   0x09
 
 #define TYPE_IPMI_BMC_EXTERN "ipmi-bmc-extern"
 #define IPMI_BMC_EXTERN(obj) OBJECT_CHECK(IPMIBmcExtern, (obj), \
@@ -276,8 +277,8 @@ static void handle_hw_op(IPMIBmcExtern *ibe, unsigned char hw_op)
         k->do_hw_op(s, IPMI_SEND_NMI, 0);
         break;
 
-    case VM_CMD_FORCEOFF:
-        qemu_system_shutdown_request();
+    case VM_CMD_GRACEFUL_SHUTDOWN:
+        k->do_hw_op(s, IPMI_SHUTDOWN_VIA_ACPI_OVERTEMP, 0);
         break;
     }
 }
@@ -401,6 +402,10 @@ static void chr_event(void *opaque, int event)
         if (k->do_hw_op(ibe->parent.intf, IPMI_POWEROFF_CHASSIS, 1) == 0) {
             v |= VM_CAPABILITIES_POWER;
         }
+        if (k->do_hw_op(ibe->parent.intf, IPMI_SHUTDOWN_VIA_ACPI_OVERTEMP, 1)
+            == 0) {
+            v |= VM_CAPABILITIES_GRACEFUL_SHUTDOWN;
+        }
         if (k->do_hw_op(ibe->parent.intf, IPMI_RESET_CHASSIS, 1) == 0) {
             v |= VM_CAPABILITIES_RESET;
         }
@@ -512,6 +517,7 @@ static void ipmi_bmc_extern_class_init(ObjectClass *oc, void *data)
 
     bk->handle_command = ipmi_bmc_extern_handle_command;
     bk->handle_reset = ipmi_bmc_extern_handle_reset;
+    dc->hotpluggable = false;
     dc->realize = ipmi_bmc_extern_realize;
     dc->props = ipmi_bmc_extern_properties;
 }
diff --git a/hw/ipmi/ipmi_bmc_sim.c b/hw/ipmi/ipmi_bmc_sim.c
index 17c7c0ea07..c7883d6f5e 100644
--- a/hw/ipmi/ipmi_bmc_sim.c
+++ b/hw/ipmi/ipmi_bmc_sim.c
@@ -217,7 +217,6 @@ struct IPMIBmcSim {
     /* Odd netfns are for responses, so we only need the even ones. */
     const IPMINetfn *netfns[MAX_NETFNS / 2];
 
-    QemuMutex lock;
     /* We allow one event in the buffer */
     uint8_t evtbuf[16];
 
@@ -940,7 +939,6 @@ static void get_msg(IPMIBmcSim *ibs,
 {
     IPMIRcvBufEntry *msg;
 
-    qemu_mutex_lock(&ibs->lock);
     if (QTAILQ_EMPTY(&ibs->rcvbufs)) {
         rsp_buffer_set_error(rsp, 0x80); /* Queue empty */
         goto out;
@@ -960,7 +958,6 @@ static void get_msg(IPMIBmcSim *ibs,
     }
 
 out:
-    qemu_mutex_unlock(&ibs->lock);
     return;
 }
 
@@ -1055,11 +1052,9 @@ static void send_msg(IPMIBmcSim *ibs,
  end_msg:
     msg->buf[msg->len] = ipmb_checksum(msg->buf, msg->len, 0);
     msg->len++;
-    qemu_mutex_lock(&ibs->lock);
     QTAILQ_INSERT_TAIL(&ibs->rcvbufs, msg, entry);
     ibs->msg_flags |= IPMI_BMC_MSG_FLAG_RCV_MSG_QUEUE;
     k->set_atn(s, 1, attn_irq_enabled(ibs));
-    qemu_mutex_unlock(&ibs->lock);
 }
 
 static void do_watchdog_reset(IPMIBmcSim *ibs)
@@ -1753,7 +1748,6 @@ static void ipmi_sim_realize(DeviceState *dev, Error **errp)
     unsigned int i;
     IPMIBmcSim *ibs = IPMI_BMC_SIMULATOR(b);
 
-    qemu_mutex_init(&ibs->lock);
     QTAILQ_INIT(&ibs->rcvbufs);
 
     ibs->bmc_global_enables = (1 << IPMI_BMC_EVENT_LOG_BIT);
@@ -1791,6 +1785,7 @@ static void ipmi_sim_class_init(ObjectClass *oc, void *data)
     DeviceClass *dc = DEVICE_CLASS(oc);
     IPMIBmcClass *bk = IPMI_BMC_CLASS(oc);
 
+    dc->hotpluggable = false;
     dc->realize = ipmi_sim_realize;
     bk->handle_command = ipmi_sim_handle_command;
 }
diff --git a/hw/mem/nvdimm.c b/hw/mem/nvdimm.c
index 7895805a23..db896b0bb6 100644
--- a/hw/mem/nvdimm.c
+++ b/hw/mem/nvdimm.c
@@ -148,13 +148,9 @@ static MemoryRegion *nvdimm_get_vmstate_memory_region(PCDIMMDevice *dimm)
 
 static void nvdimm_class_init(ObjectClass *oc, void *data)
 {
-    DeviceClass *dc = DEVICE_CLASS(oc);
     PCDIMMDeviceClass *ddc = PC_DIMM_CLASS(oc);
     NVDIMMClass *nvc = NVDIMM_CLASS(oc);
 
-    /* nvdimm hotplug has not been supported yet. */
-    dc->hotpluggable = false;
-
     ddc->realize = nvdimm_realize;
     ddc->get_memory_region = nvdimm_get_memory_region;
     ddc->get_vmstate_memory_region = nvdimm_get_vmstate_memory_region;
diff --git a/hw/s390x/virtio-ccw.c b/hw/s390x/virtio-ccw.c
index ee136ab940..7d7f8f6e19 100644
--- a/hw/s390x/virtio-ccw.c
+++ b/hw/s390x/virtio-ccw.c
@@ -59,38 +59,11 @@ static void virtio_ccw_stop_ioeventfd(VirtioCcwDevice *dev)
     virtio_bus_stop_ioeventfd(&dev->bus);
 }
 
-static bool virtio_ccw_ioeventfd_started(DeviceState *d)
+static bool virtio_ccw_ioeventfd_enabled(DeviceState *d)
 {
     VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
 
-    return dev->ioeventfd_started;
-}
-
-static void virtio_ccw_ioeventfd_set_started(DeviceState *d, bool started,
-                                             bool err)
-{
-    VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
-
-    dev->ioeventfd_started = started;
-    if (err) {
-        /* Disable ioeventfd for this device. */
-        dev->flags &= ~VIRTIO_CCW_FLAG_USE_IOEVENTFD;
-    }
-}
-
-static bool virtio_ccw_ioeventfd_disabled(DeviceState *d)
-{
-    VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
-
-    return dev->ioeventfd_disabled ||
-        !(dev->flags & VIRTIO_CCW_FLAG_USE_IOEVENTFD);
-}
-
-static void virtio_ccw_ioeventfd_set_disabled(DeviceState *d, bool disabled)
-{
-    VirtioCcwDevice *dev = VIRTIO_CCW_DEVICE(d);
-
-    dev->ioeventfd_disabled = disabled;
+    return (dev->flags & VIRTIO_CCW_FLAG_USE_IOEVENTFD) != 0;
 }
 
 static int virtio_ccw_ioeventfd_assign(DeviceState *d, EventNotifier *notifier,
@@ -709,6 +682,10 @@ static void virtio_ccw_device_realize(VirtioCcwDevice *dev, Error **errp)
         sch->cssid, sch->ssid, sch->schid, sch->devno,
         ccw_dev->bus_id.valid ? "user-configured" : "auto-configured");
 
+    if (!kvm_eventfds_enabled()) {
+        dev->flags &= ~VIRTIO_CCW_FLAG_USE_IOEVENTFD;
+    }
+
     if (k->realize) {
         k->realize(dev, &err);
     }
@@ -1311,10 +1288,6 @@ static void virtio_ccw_device_plugged(DeviceState *d, Error **errp)
         return;
     }
 
-    if (!kvm_eventfds_enabled()) {
-        dev->flags &= ~VIRTIO_CCW_FLAG_USE_IOEVENTFD;
-    }
-
     sch->id.cu_model = virtio_bus_get_vdev_id(&dev->bus);
 
 
@@ -1616,10 +1589,7 @@ static void virtio_ccw_bus_class_init(ObjectClass *klass, void *data)
     k->pre_plugged = virtio_ccw_pre_plugged;
     k->device_plugged = virtio_ccw_device_plugged;
     k->device_unplugged = virtio_ccw_device_unplugged;
-    k->ioeventfd_started = virtio_ccw_ioeventfd_started;
-    k->ioeventfd_set_started = virtio_ccw_ioeventfd_set_started;
-    k->ioeventfd_disabled = virtio_ccw_ioeventfd_disabled;
-    k->ioeventfd_set_disabled = virtio_ccw_ioeventfd_set_disabled;
+    k->ioeventfd_enabled = virtio_ccw_ioeventfd_enabled;
     k->ioeventfd_assign = virtio_ccw_ioeventfd_assign;
 }
 
diff --git a/hw/s390x/virtio-ccw.h b/hw/s390x/virtio-ccw.h
index 565094e4fb..77d10f1671 100644
--- a/hw/s390x/virtio-ccw.h
+++ b/hw/s390x/virtio-ccw.h
@@ -86,8 +86,6 @@ struct VirtioCcwDevice {
     int revision;
     uint32_t max_rev;
     VirtioBusState bus;
-    bool ioeventfd_started;
-    bool ioeventfd_disabled;
     uint32_t flags;
     uint8_t thinint_isc;
     AdapterRoutes routes;
diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi-dataplane.c
index 9424f0e057..f2ea29dbc3 100644
--- a/hw/scsi/virtio-scsi-dataplane.c
+++ b/hw/scsi/virtio-scsi-dataplane.c
@@ -12,6 +12,7 @@
  */
 
 #include "qemu/osdep.h"
+#include "qapi/error.h"
 #include "hw/virtio/virtio-scsi.h"
 #include "qemu/error-report.h"
 #include "sysemu/block-backend.h"
@@ -21,20 +22,30 @@
 #include "hw/virtio/virtio-access.h"
 
 /* Context: QEMU global mutex held */
-void virtio_scsi_set_iothread(VirtIOSCSI *s, IOThread *iothread)
+void virtio_scsi_dataplane_setup(VirtIOSCSI *s, Error **errp)
 {
-    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s)));
-    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
     VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(s);
+    VirtIODevice *vdev = VIRTIO_DEVICE(s);
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
+    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
 
-    assert(!s->ctx);
-    s->ctx = iothread_get_aio_context(vs->conf.iothread);
-
-    /* Don't try if transport does not support notifiers. */
-    if (!k->set_guest_notifiers || !k->ioeventfd_started) {
-        fprintf(stderr, "virtio-scsi: Failed to set iothread "
-                   "(transport does not support notifiers)");
-        exit(1);
+    if (vs->conf.iothread) {
+        if (!k->set_guest_notifiers || !k->ioeventfd_assign) {
+            error_setg(errp,
+                       "device is incompatible with iothread "
+                       "(transport does not support notifiers)");
+            return;
+        }
+        if (!virtio_device_ioeventfd_enabled(vdev)) {
+            error_setg(errp, "ioeventfd is required for iothread");
+            return;
+        }
+        s->ctx = iothread_get_aio_context(vs->conf.iothread);
+    } else {
+        if (!virtio_device_ioeventfd_enabled(vdev)) {
+            return;
+        }
+        s->ctx = qemu_get_aio_context();
     }
 }
 
@@ -105,19 +116,19 @@ static void virtio_scsi_clear_aio(VirtIOSCSI *s)
 }
 
 /* Context: QEMU global mutex held */
-void virtio_scsi_dataplane_start(VirtIOSCSI *s)
+int virtio_scsi_dataplane_start(VirtIODevice *vdev)
 {
     int i;
     int rc;
-    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s)));
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(s);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
+    VirtIOSCSI *s = VIRTIO_SCSI(vdev);
 
     if (s->dataplane_started ||
         s->dataplane_starting ||
-        s->dataplane_fenced ||
-        s->ctx != iothread_get_aio_context(vs->conf.iothread)) {
-        return;
+        s->dataplane_fenced) {
+        return 0;
     }
 
     s->dataplane_starting = true;
@@ -152,7 +163,7 @@ void virtio_scsi_dataplane_start(VirtIOSCSI *s)
     s->dataplane_starting = false;
     s->dataplane_started = true;
     aio_context_release(s->ctx);
-    return;
+    return 0;
 
 fail_vrings:
     virtio_scsi_clear_aio(s);
@@ -165,14 +176,16 @@ fail_guest_notifiers:
     s->dataplane_fenced = true;
     s->dataplane_starting = false;
     s->dataplane_started = true;
+    return -ENOSYS;
 }
 
 /* Context: QEMU global mutex held */
-void virtio_scsi_dataplane_stop(VirtIOSCSI *s)
+void virtio_scsi_dataplane_stop(VirtIODevice *vdev)
 {
-    BusState *qbus = BUS(qdev_get_parent_bus(DEVICE(s)));
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus);
-    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(s);
+    VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
+    VirtIOSCSI *s = VIRTIO_SCSI(vdev);
     int i;
 
     if (!s->dataplane_started || s->dataplane_stopping) {
@@ -186,7 +199,6 @@ void virtio_scsi_dataplane_stop(VirtIOSCSI *s)
         return;
     }
     s->dataplane_stopping = true;
-    assert(s->ctx == iothread_get_aio_context(vs->conf.iothread));
 
     aio_context_acquire(s->ctx);
     virtio_scsi_clear_aio(s);
diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index 4762f05274..3e5ae6ac0f 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -434,7 +434,7 @@ static void virtio_scsi_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
     VirtIOSCSI *s = (VirtIOSCSI *)vdev;
 
     if (s->ctx) {
-        virtio_scsi_dataplane_start(s);
+        virtio_device_start_ioeventfd(vdev);
         if (!s->dataplane_fenced) {
             return;
         }
@@ -610,7 +610,7 @@ static void virtio_scsi_handle_cmd(VirtIODevice *vdev, VirtQueue *vq)
     VirtIOSCSI *s = (VirtIOSCSI *)vdev;
 
     if (s->ctx) {
-        virtio_scsi_dataplane_start(s);
+        virtio_device_start_ioeventfd(vdev);
         if (!s->dataplane_fenced) {
             return;
         }
@@ -669,9 +669,7 @@ static void virtio_scsi_reset(VirtIODevice *vdev)
     VirtIOSCSI *s = VIRTIO_SCSI(vdev);
     VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(vdev);
 
-    if (s->ctx) {
-        virtio_scsi_dataplane_stop(s);
-    }
+    assert(!s->dataplane_started);
     s->resetting++;
     qbus_reset_all(&s->bus.qbus);
     s->resetting--;
@@ -749,7 +747,7 @@ static void virtio_scsi_handle_event(VirtIODevice *vdev, VirtQueue *vq)
     VirtIOSCSI *s = VIRTIO_SCSI(vdev);
 
     if (s->ctx) {
-        virtio_scsi_dataplane_start(s);
+        virtio_device_start_ioeventfd(vdev);
         if (!s->dataplane_fenced) {
             return;
         }
@@ -848,14 +846,10 @@ void virtio_scsi_common_realize(DeviceState *dev, Error **errp,
     s->sense_size = VIRTIO_SCSI_SENSE_DEFAULT_SIZE;
     s->cdb_size = VIRTIO_SCSI_CDB_DEFAULT_SIZE;
 
-    s->ctrl_vq = virtio_add_queue_aio(vdev, VIRTIO_SCSI_VQ_SIZE, ctrl);
-    s->event_vq = virtio_add_queue_aio(vdev, VIRTIO_SCSI_VQ_SIZE, evt);
+    s->ctrl_vq = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE, ctrl);
+    s->event_vq = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE, evt);
     for (i = 0; i < s->conf.num_queues; i++) {
-        s->cmd_vqs[i] = virtio_add_queue_aio(vdev, VIRTIO_SCSI_VQ_SIZE, cmd);
-    }
-
-    if (s->conf.iothread) {
-        virtio_scsi_set_iothread(VIRTIO_SCSI(s), s->conf.iothread);
+        s->cmd_vqs[i] = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE, cmd);
     }
 }
 
@@ -885,6 +879,8 @@ static void virtio_scsi_device_realize(DeviceState *dev, Error **errp)
             return;
         }
     }
+
+    virtio_scsi_dataplane_setup(s, errp);
 }
 
 static void virtio_scsi_instance_init(Object *obj)
@@ -957,6 +953,8 @@ static void virtio_scsi_class_init(ObjectClass *klass, void *data)
     vdc->set_config = virtio_scsi_set_config;
     vdc->get_features = virtio_scsi_get_features;
     vdc->reset = virtio_scsi_reset;
+    vdc->start_ioeventfd = virtio_scsi_dataplane_start;
+    vdc->stop_ioeventfd = virtio_scsi_dataplane_stop;
     hc->plug = virtio_scsi_hotplug;
     hc->unplug = virtio_scsi_hotunplug;
 }
diff --git a/hw/virtio/Makefile.objs b/hw/virtio/Makefile.objs
index e71630812e..95c4c30ea1 100644
--- a/hw/virtio/Makefile.objs
+++ b/hw/virtio/Makefile.objs
@@ -7,3 +7,5 @@ obj-y += virtio.o virtio-balloon.o
 obj-$(CONFIG_LINUX) += vhost.o vhost-backend.o vhost-user.o
 
 obj-$(CONFIG_VHOST_VSOCK) += vhost-vsock.o
+obj-y += virtio-crypto.o
+obj-$(CONFIG_VIRTIO_PCI) += virtio-crypto-pci.o
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index bd051ab2e1..131f1643b2 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -1190,12 +1190,13 @@ int vhost_dev_enable_notifiers(struct vhost_dev *hdev, VirtIODevice *vdev)
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(vbus);
     int i, r, e;
 
-    if (!k->ioeventfd_started) {
+    if (!k->ioeventfd_assign) {
         error_report("binding does not support host notifiers");
         r = -ENOSYS;
         goto fail;
     }
 
+    virtio_device_stop_ioeventfd(vdev);
     for (i = 0; i < hdev->nvqs; ++i) {
         r = virtio_bus_set_host_notifier(VIRTIO_BUS(qbus), hdev->vq_index + i,
                                          true);
@@ -1215,6 +1216,7 @@ fail_vq:
         }
         assert (e >= 0);
     }
+    virtio_device_start_ioeventfd(vdev);
 fail:
     return r;
 }
@@ -1237,6 +1239,7 @@ void vhost_dev_disable_notifiers(struct vhost_dev *hdev, VirtIODevice *vdev)
         }
         assert (r >= 0);
     }
+    virtio_device_start_ioeventfd(vdev);
 }
 
 /* Test and clear event pending status.
diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
index 1d77028236..cfba053280 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
@@ -394,21 +394,9 @@ static void virtio_balloon_to_target(void *opaque, ram_addr_t target)
     trace_virtio_balloon_to_target(target, dev->num_pages);
 }
 
-static void virtio_balloon_save_device(VirtIODevice *vdev, QEMUFile *f)
+static int virtio_balloon_post_load_device(void *opaque, int version_id)
 {
-    VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
-
-    qemu_put_be32(f, s->num_pages);
-    qemu_put_be32(f, s->actual);
-}
-
-static int virtio_balloon_load_device(VirtIODevice *vdev, QEMUFile *f,
-                                      int version_id)
-{
-    VirtIOBalloon *s = VIRTIO_BALLOON(vdev);
-
-    s->num_pages = qemu_get_be32(f);
-    s->actual = qemu_get_be32(f);
+    VirtIOBalloon *s = VIRTIO_BALLOON(opaque);
 
     if (balloon_stats_enabled(s)) {
         balloon_stats_change_timer(s, s->stats_poll_interval);
@@ -416,6 +404,18 @@ static int virtio_balloon_load_device(VirtIODevice *vdev, QEMUFile *f,
     return 0;
 }
 
+static const VMStateDescription vmstate_virtio_balloon_device = {
+    .name = "virtio-balloon-device",
+    .version_id = 1,
+    .minimum_version_id = 1,
+    .post_load = virtio_balloon_post_load_device,
+    .fields = (VMStateField[]) {
+        VMSTATE_UINT32(num_pages, VirtIOBalloon),
+        VMSTATE_UINT32(actual, VirtIOBalloon),
+        VMSTATE_END_OF_LIST()
+    },
+};
+
 static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)
 {
     VirtIODevice *vdev = VIRTIO_DEVICE(dev);
@@ -517,9 +517,8 @@ static void virtio_balloon_class_init(ObjectClass *klass, void *data)
     vdc->get_config = virtio_balloon_get_config;
     vdc->set_config = virtio_balloon_set_config;
     vdc->get_features = virtio_balloon_get_features;
-    vdc->save = virtio_balloon_save_device;
-    vdc->load = virtio_balloon_load_device;
     vdc->set_status = virtio_balloon_set_status;
+    vdc->vmsd = &vmstate_virtio_balloon_device;
 }
 
 static const TypeInfo virtio_balloon_info = {
diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
index 11f65bd225..bf61f66a04 100644
--- a/hw/virtio/virtio-bus.c
+++ b/hw/virtio/virtio-bus.c
@@ -147,131 +147,97 @@ void virtio_bus_set_vdev_config(VirtioBusState *bus, uint8_t *config)
     }
 }
 
-/*
- * This function handles both assigning the ioeventfd handler and
- * registering it with the kernel.
- * assign: register/deregister ioeventfd with the kernel
- * set_handler: use the generic ioeventfd handler
- */
-static int set_host_notifier_internal(DeviceState *proxy, VirtioBusState *bus,
-                                      int n, bool assign, bool set_handler)
+int virtio_bus_start_ioeventfd(VirtioBusState *bus)
 {
-    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(bus);
-    VirtQueue *vq = virtio_get_queue(vdev, n);
-    EventNotifier *notifier = virtio_queue_get_host_notifier(vq);
-    int r = 0;
+    DeviceState *proxy = DEVICE(BUS(bus)->parent);
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(vdev);
+    int r;
 
-    if (assign) {
-        r = event_notifier_init(notifier, 1);
-        if (r < 0) {
-            error_report("%s: unable to init event notifier: %s (%d)",
-                         __func__, strerror(-r), r);
-            return r;
-        }
-        virtio_queue_set_host_notifier_fd_handler(vq, true, set_handler);
-        r = k->ioeventfd_assign(proxy, notifier, n, assign);
-        if (r < 0) {
-            error_report("%s: unable to assign ioeventfd: %d", __func__, r);
-            virtio_queue_set_host_notifier_fd_handler(vq, false, false);
-            event_notifier_cleanup(notifier);
-            return r;
-        }
-    } else {
-        k->ioeventfd_assign(proxy, notifier, n, assign);
-        virtio_queue_set_host_notifier_fd_handler(vq, false, false);
-        event_notifier_cleanup(notifier);
+    if (!k->ioeventfd_assign || !k->ioeventfd_enabled(proxy)) {
+        return -ENOSYS;
     }
-    return r;
+    if (bus->ioeventfd_started) {
+        return 0;
+    }
+    r = vdc->start_ioeventfd(vdev);
+    if (r < 0) {
+        error_report("%s: failed. Fallback to userspace (slower).", __func__);
+        return r;
+    }
+    bus->ioeventfd_started = true;
+    return 0;
 }
 
-void virtio_bus_start_ioeventfd(VirtioBusState *bus)
+void virtio_bus_stop_ioeventfd(VirtioBusState *bus)
 {
-    VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(bus);
-    DeviceState *proxy = DEVICE(BUS(bus)->parent);
     VirtIODevice *vdev;
-    int n, r;
+    VirtioDeviceClass *vdc;
 
-    if (!k->ioeventfd_started || k->ioeventfd_started(proxy)) {
+    if (!bus->ioeventfd_started) {
         return;
     }
-    if (k->ioeventfd_disabled(proxy)) {
-        return;
-    }
-    vdev = virtio_bus_get_device(bus);
-    for (n = 0; n < VIRTIO_QUEUE_MAX; n++) {
-        if (!virtio_queue_get_num(vdev, n)) {
-            continue;
-        }
-        r = set_host_notifier_internal(proxy, bus, n, true, true);
-        if (r < 0) {
-            goto assign_error;
-        }
-    }
-    k->ioeventfd_set_started(proxy, true, false);
-    return;
-
-assign_error:
-    while (--n >= 0) {
-        if (!virtio_queue_get_num(vdev, n)) {
-            continue;
-        }
 
-        r = set_host_notifier_internal(proxy, bus, n, false, false);
-        assert(r >= 0);
-    }
-    k->ioeventfd_set_started(proxy, false, true);
-    error_report("%s: failed. Fallback to userspace (slower).", __func__);
+    vdev = virtio_bus_get_device(bus);
+    vdc = VIRTIO_DEVICE_GET_CLASS(vdev);
+    vdc->stop_ioeventfd(vdev);
+    bus->ioeventfd_started = false;
 }
 
-void virtio_bus_stop_ioeventfd(VirtioBusState *bus)
+bool virtio_bus_ioeventfd_enabled(VirtioBusState *bus)
 {
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(bus);
     DeviceState *proxy = DEVICE(BUS(bus)->parent);
-    VirtIODevice *vdev;
-    int n, r;
 
-    if (!k->ioeventfd_started || !k->ioeventfd_started(proxy)) {
-        return;
-    }
-    vdev = virtio_bus_get_device(bus);
-    for (n = 0; n < VIRTIO_QUEUE_MAX; n++) {
-        if (!virtio_queue_get_num(vdev, n)) {
-            continue;
-        }
-        r = set_host_notifier_internal(proxy, bus, n, false, false);
-        assert(r >= 0);
-    }
-    k->ioeventfd_set_started(proxy, false, false);
+    return k->ioeventfd_assign && k->ioeventfd_enabled(proxy);
 }
 
 /*
- * This function switches from/to the generic ioeventfd handler.
- * assign==false means 'use generic ioeventfd handler'.
+ * This function switches ioeventfd on/off in the device.
+ * The caller must set or clear the handlers for the EventNotifier.
  */
 int virtio_bus_set_host_notifier(VirtioBusState *bus, int n, bool assign)
 {
+    VirtIODevice *vdev = virtio_bus_get_device(bus);
     VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(bus);
     DeviceState *proxy = DEVICE(BUS(bus)->parent);
+    VirtQueue *vq = virtio_get_queue(vdev, n);
+    EventNotifier *notifier = virtio_queue_get_host_notifier(vq);
+    int r = 0;
 
-    if (!k->ioeventfd_started) {
+    if (!k->ioeventfd_assign) {
         return -ENOSYS;
     }
-    k->ioeventfd_set_disabled(proxy, assign);
+
     if (assign) {
-        /*
-         * Stop using the generic ioeventfd, we are doing eventfd handling
-         * ourselves below
-         *
-         * FIXME: We should just switch the handler and not deassign the
-         * ioeventfd.
-         * Otherwise, there's a window where we don't have an
-         * ioeventfd and we may end up with a notification where
-         * we don't expect one.
-         */
-        virtio_bus_stop_ioeventfd(bus);
+        assert(!bus->ioeventfd_started);
+        r = event_notifier_init(notifier, 1);
+        if (r < 0) {
+            error_report("%s: unable to init event notifier: %s (%d)",
+                         __func__, strerror(-r), r);
+            return r;
+        }
+        r = k->ioeventfd_assign(proxy, notifier, n, true);
+        if (r < 0) {
+            error_report("%s: unable to assign ioeventfd: %d", __func__, r);
+            goto cleanup_event_notifier;
+        }
+        return 0;
+    } else {
+        if (!bus->ioeventfd_started) {
+            return 0;
+        }
+        k->ioeventfd_assign(proxy, notifier, n, false);
     }
-    return set_host_notifier_internal(proxy, bus, n, assign, false);
+
+cleanup_event_notifier:
+    /* Test and clear notifier after disabling event,
+     * in case poll callback didn't have time to run.
+     */
+    virtio_queue_host_notifier_read(notifier);
+    event_notifier_cleanup(notifier);
+    return r;
 }
 
 static char *virtio_bus_get_dev_path(DeviceState *dev)
diff --git a/hw/virtio/virtio-crypto-pci.c b/hw/virtio/virtio-crypto-pci.c
new file mode 100644
index 0000000000..21d998401a
--- /dev/null
+++ b/hw/virtio/virtio-crypto-pci.c
@@ -0,0 +1,77 @@
+/*
+ * Virtio crypto device
+ *
+ * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
+ *
+ * Authors:
+ *    Gonglei <arei.gonglei@huawei.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or
+ * (at your option) any later version.  See the COPYING file in the
+ * top-level directory.
+ *
+ */
+#include "qemu/osdep.h"
+#include "hw/pci/pci.h"
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/virtio-bus.h"
+#include "hw/virtio/virtio-pci.h"
+#include "hw/virtio/virtio-crypto.h"
+#include "qapi/error.h"
+
+static Property virtio_crypto_pci_properties[] = {
+    DEFINE_PROP_BIT("ioeventfd", VirtIOPCIProxy, flags,
+                    VIRTIO_PCI_FLAG_USE_IOEVENTFD_BIT, true),
+    DEFINE_PROP_UINT32("vectors", VirtIOPCIProxy, nvectors, 2),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_crypto_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp)
+{
+    VirtIOCryptoPCI *vcrypto = VIRTIO_CRYPTO_PCI(vpci_dev);
+    DeviceState *vdev = DEVICE(&vcrypto->vdev);
+
+    qdev_set_parent_bus(vdev, BUS(&vpci_dev->bus));
+    virtio_pci_force_virtio_1(vpci_dev);
+    object_property_set_bool(OBJECT(vdev), true, "realized", errp);
+    object_property_set_link(OBJECT(vcrypto),
+                 OBJECT(vcrypto->vdev.conf.cryptodev), "cryptodev",
+                 NULL);
+}
+
+static void virtio_crypto_pci_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioPCIClass *k = VIRTIO_PCI_CLASS(klass);
+    PCIDeviceClass *pcidev_k = PCI_DEVICE_CLASS(klass);
+
+    k->realize = virtio_crypto_pci_realize;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    dc->props = virtio_crypto_pci_properties;
+
+    pcidev_k->class_id = PCI_CLASS_OTHERS;
+}
+
+static void virtio_crypto_initfn(Object *obj)
+{
+    VirtIOCryptoPCI *dev = VIRTIO_CRYPTO_PCI(obj);
+
+    virtio_instance_init_common(obj, &dev->vdev, sizeof(dev->vdev),
+                                TYPE_VIRTIO_CRYPTO);
+    object_property_add_alias(obj, "cryptodev", OBJECT(&dev->vdev),
+                              "cryptodev", &error_abort);
+}
+
+static const TypeInfo virtio_crypto_pci_info = {
+    .name          = TYPE_VIRTIO_CRYPTO_PCI,
+    .parent        = TYPE_VIRTIO_PCI,
+    .instance_size = sizeof(VirtIOCryptoPCI),
+    .instance_init = virtio_crypto_initfn,
+    .class_init    = virtio_crypto_pci_class_init,
+};
+
+static void virtio_crypto_pci_register_types(void)
+{
+    type_register_static(&virtio_crypto_pci_info);
+}
+type_init(virtio_crypto_pci_register_types)
diff --git a/hw/virtio/virtio-crypto.c b/hw/virtio/virtio-crypto.c
new file mode 100644
index 0000000000..170114f52b
--- /dev/null
+++ b/hw/virtio/virtio-crypto.c
@@ -0,0 +1,898 @@
+/*
+ * Virtio crypto Support
+ *
+ * Copyright (c) 2016 HUAWEI TECHNOLOGIES CO., LTD.
+ *
+ * Authors:
+ *    Gonglei <arei.gonglei@huawei.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or
+ * (at your option) any later version.  See the COPYING file in the
+ * top-level directory.
+ */
+#include "qemu/osdep.h"
+#include "qemu/iov.h"
+#include "hw/qdev.h"
+#include "qapi/error.h"
+#include "qemu/error-report.h"
+
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/virtio-crypto.h"
+#include "hw/virtio/virtio-access.h"
+#include "standard-headers/linux/virtio_ids.h"
+
+#define VIRTIO_CRYPTO_VM_VERSION 1
+
+/*
+ * Transfer virtqueue index to crypto queue index.
+ * The control virtqueue is after the data virtqueues
+ * so the input value doesn't need to be adjusted
+ */
+static inline int virtio_crypto_vq2q(int queue_index)
+{
+    return queue_index;
+}
+
+static int
+virtio_crypto_cipher_session_helper(VirtIODevice *vdev,
+           CryptoDevBackendSymSessionInfo *info,
+           struct virtio_crypto_cipher_session_para *cipher_para,
+           struct iovec **iov, unsigned int *out_num)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+    unsigned int num = *out_num;
+
+    info->cipher_alg = ldl_le_p(&cipher_para->algo);
+    info->key_len = ldl_le_p(&cipher_para->keylen);
+    info->direction = ldl_le_p(&cipher_para->op);
+    DPRINTF("cipher_alg=%" PRIu32 ", info->direction=%" PRIu32 "\n",
+             info->cipher_alg, info->direction);
+
+    if (info->key_len > vcrypto->conf.max_cipher_key_len) {
+        error_report("virtio-crypto length of cipher key is too big: %u",
+                     info->key_len);
+        return -VIRTIO_CRYPTO_ERR;
+    }
+    /* Get cipher key */
+    if (info->key_len > 0) {
+        size_t s;
+        DPRINTF("keylen=%" PRIu32 "\n", info->key_len);
+
+        info->cipher_key = g_malloc(info->key_len);
+        s = iov_to_buf(*iov, num, 0, info->cipher_key, info->key_len);
+        if (unlikely(s != info->key_len)) {
+            virtio_error(vdev, "virtio-crypto cipher key incorrect");
+            return -EFAULT;
+        }
+        iov_discard_front(iov, &num, info->key_len);
+        *out_num = num;
+    }
+
+    return 0;
+}
+
+static int64_t
+virtio_crypto_create_sym_session(VirtIOCrypto *vcrypto,
+               struct virtio_crypto_sym_create_session_req *sess_req,
+               uint32_t queue_id,
+               uint32_t opcode,
+               struct iovec *iov, unsigned int out_num)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(vcrypto);
+    CryptoDevBackendSymSessionInfo info;
+    int64_t session_id;
+    int queue_index;
+    uint32_t op_type;
+    Error *local_err = NULL;
+    int ret;
+
+    memset(&info, 0, sizeof(info));
+    op_type = ldl_le_p(&sess_req->op_type);
+    info.op_type = op_type;
+    info.op_code = opcode;
+
+    if (op_type == VIRTIO_CRYPTO_SYM_OP_CIPHER) {
+        ret = virtio_crypto_cipher_session_helper(vdev, &info,
+                           &sess_req->u.cipher.para,
+                           &iov, &out_num);
+        if (ret < 0) {
+            goto err;
+        }
+    } else if (op_type == VIRTIO_CRYPTO_SYM_OP_ALGORITHM_CHAINING) {
+        size_t s;
+        /* cipher part */
+        ret = virtio_crypto_cipher_session_helper(vdev, &info,
+                           &sess_req->u.chain.para.cipher_param,
+                           &iov, &out_num);
+        if (ret < 0) {
+            goto err;
+        }
+        /* hash part */
+        info.alg_chain_order = ldl_le_p(
+                                     &sess_req->u.chain.para.alg_chain_order);
+        info.add_len = ldl_le_p(&sess_req->u.chain.para.aad_len);
+        info.hash_mode = ldl_le_p(&sess_req->u.chain.para.hash_mode);
+        if (info.hash_mode == VIRTIO_CRYPTO_SYM_HASH_MODE_AUTH) {
+            info.hash_alg = ldl_le_p(&sess_req->u.chain.para.u.mac_param.algo);
+            info.auth_key_len = ldl_le_p(
+                             &sess_req->u.chain.para.u.mac_param.auth_key_len);
+            info.hash_result_len = ldl_le_p(
+                           &sess_req->u.chain.para.u.mac_param.hash_result_len);
+            if (info.auth_key_len > vcrypto->conf.max_auth_key_len) {
+                error_report("virtio-crypto length of auth key is too big: %u",
+                             info.auth_key_len);
+                ret = -VIRTIO_CRYPTO_ERR;
+                goto err;
+            }
+            /* get auth key */
+            if (info.auth_key_len > 0) {
+                DPRINTF("auth_keylen=%" PRIu32 "\n", info.auth_key_len);
+                info.auth_key = g_malloc(info.auth_key_len);
+                s = iov_to_buf(iov, out_num, 0, info.auth_key,
+                               info.auth_key_len);
+                if (unlikely(s != info.auth_key_len)) {
+                    virtio_error(vdev,
+                          "virtio-crypto authenticated key incorrect");
+                    ret = -EFAULT;
+                    goto err;
+                }
+                iov_discard_front(&iov, &out_num, info.auth_key_len);
+            }
+        } else if (info.hash_mode == VIRTIO_CRYPTO_SYM_HASH_MODE_PLAIN) {
+            info.hash_alg = ldl_le_p(
+                             &sess_req->u.chain.para.u.hash_param.algo);
+            info.hash_result_len = ldl_le_p(
+                        &sess_req->u.chain.para.u.hash_param.hash_result_len);
+        } else {
+            /* VIRTIO_CRYPTO_SYM_HASH_MODE_NESTED */
+            error_report("unsupported hash mode");
+            ret = -VIRTIO_CRYPTO_NOTSUPP;
+            goto err;
+        }
+    } else {
+        /* VIRTIO_CRYPTO_SYM_OP_NONE */
+        error_report("unsupported cipher op_type: VIRTIO_CRYPTO_SYM_OP_NONE");
+        ret = -VIRTIO_CRYPTO_NOTSUPP;
+        goto err;
+    }
+
+    queue_index = virtio_crypto_vq2q(queue_id);
+    session_id = cryptodev_backend_sym_create_session(
+                                     vcrypto->cryptodev,
+                                     &info, queue_index, &local_err);
+    if (session_id >= 0) {
+        DPRINTF("create session_id=%" PRIu64 " successfully\n",
+                session_id);
+
+        ret = session_id;
+    } else {
+        if (local_err) {
+            error_report_err(local_err);
+        }
+        ret = -VIRTIO_CRYPTO_ERR;
+    }
+
+err:
+    g_free(info.cipher_key);
+    g_free(info.auth_key);
+    return ret;
+}
+
+static uint8_t
+virtio_crypto_handle_close_session(VirtIOCrypto *vcrypto,
+         struct virtio_crypto_destroy_session_req *close_sess_req,
+         uint32_t queue_id)
+{
+    int ret;
+    uint64_t session_id;
+    uint32_t status;
+    Error *local_err = NULL;
+
+    session_id = ldq_le_p(&close_sess_req->session_id);
+    DPRINTF("close session, id=%" PRIu64 "\n", session_id);
+
+    ret = cryptodev_backend_sym_close_session(
+              vcrypto->cryptodev, session_id, queue_id, &local_err);
+    if (ret == 0) {
+        status = VIRTIO_CRYPTO_OK;
+    } else {
+        if (local_err) {
+            error_report_err(local_err);
+        } else {
+            error_report("destroy session failed");
+        }
+        status = VIRTIO_CRYPTO_ERR;
+    }
+
+    return status;
+}
+
+static void virtio_crypto_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+    struct virtio_crypto_op_ctrl_req ctrl;
+    VirtQueueElement *elem;
+    struct iovec *in_iov;
+    struct iovec *out_iov;
+    unsigned in_num;
+    unsigned out_num;
+    uint32_t queue_id;
+    uint32_t opcode;
+    struct virtio_crypto_session_input input;
+    int64_t session_id;
+    uint8_t status;
+    size_t s;
+
+    for (;;) {
+        elem = virtqueue_pop(vq, sizeof(VirtQueueElement));
+        if (!elem) {
+            break;
+        }
+        if (elem->out_num < 1 || elem->in_num < 1) {
+            virtio_error(vdev, "virtio-crypto ctrl missing headers");
+            virtqueue_detach_element(vq, elem, 0);
+            g_free(elem);
+            break;
+        }
+
+        out_num = elem->out_num;
+        out_iov = elem->out_sg;
+        in_num = elem->in_num;
+        in_iov = elem->in_sg;
+        if (unlikely(iov_to_buf(out_iov, out_num, 0, &ctrl, sizeof(ctrl))
+                    != sizeof(ctrl))) {
+            virtio_error(vdev, "virtio-crypto request ctrl_hdr too short");
+            virtqueue_detach_element(vq, elem, 0);
+            g_free(elem);
+            break;
+        }
+        iov_discard_front(&out_iov, &out_num, sizeof(ctrl));
+
+        opcode = ldl_le_p(&ctrl.header.opcode);
+        queue_id = ldl_le_p(&ctrl.header.queue_id);
+
+        switch (opcode) {
+        case VIRTIO_CRYPTO_CIPHER_CREATE_SESSION:
+            memset(&input, 0, sizeof(input));
+            session_id = virtio_crypto_create_sym_session(vcrypto,
+                             &ctrl.u.sym_create_session,
+                             queue_id, opcode,
+                             out_iov, out_num);
+            /* Serious errors, need to reset virtio crypto device */
+            if (session_id == -EFAULT) {
+                virtqueue_detach_element(vq, elem, 0);
+                break;
+            } else if (session_id == -VIRTIO_CRYPTO_NOTSUPP) {
+                stl_le_p(&input.status, VIRTIO_CRYPTO_NOTSUPP);
+            } else if (session_id == -VIRTIO_CRYPTO_ERR) {
+                stl_le_p(&input.status, VIRTIO_CRYPTO_ERR);
+            } else {
+                /* Set the session id */
+                stq_le_p(&input.session_id, session_id);
+                stl_le_p(&input.status, VIRTIO_CRYPTO_OK);
+            }
+
+            s = iov_from_buf(in_iov, in_num, 0, &input, sizeof(input));
+            if (unlikely(s != sizeof(input))) {
+                virtio_error(vdev, "virtio-crypto input incorrect");
+                virtqueue_detach_element(vq, elem, 0);
+                break;
+            }
+            virtqueue_push(vq, elem, sizeof(input));
+            virtio_notify(vdev, vq);
+            break;
+        case VIRTIO_CRYPTO_CIPHER_DESTROY_SESSION:
+        case VIRTIO_CRYPTO_HASH_DESTROY_SESSION:
+        case VIRTIO_CRYPTO_MAC_DESTROY_SESSION:
+        case VIRTIO_CRYPTO_AEAD_DESTROY_SESSION:
+            status = virtio_crypto_handle_close_session(vcrypto,
+                   &ctrl.u.destroy_session, queue_id);
+            /* The status only occupy one byte, we can directly use it */
+            s = iov_from_buf(in_iov, in_num, 0, &status, sizeof(status));
+            if (unlikely(s != sizeof(status))) {
+                virtio_error(vdev, "virtio-crypto status incorrect");
+                virtqueue_detach_element(vq, elem, 0);
+                break;
+            }
+            virtqueue_push(vq, elem, sizeof(status));
+            virtio_notify(vdev, vq);
+            break;
+        case VIRTIO_CRYPTO_HASH_CREATE_SESSION:
+        case VIRTIO_CRYPTO_MAC_CREATE_SESSION:
+        case VIRTIO_CRYPTO_AEAD_CREATE_SESSION:
+        default:
+            error_report("virtio-crypto unsupported ctrl opcode: %d", opcode);
+            memset(&input, 0, sizeof(input));
+            stl_le_p(&input.status, VIRTIO_CRYPTO_NOTSUPP);
+            s = iov_from_buf(in_iov, in_num, 0, &input, sizeof(input));
+            if (unlikely(s != sizeof(input))) {
+                virtio_error(vdev, "virtio-crypto input incorrect");
+                virtqueue_detach_element(vq, elem, 0);
+                break;
+            }
+            virtqueue_push(vq, elem, sizeof(input));
+            virtio_notify(vdev, vq);
+
+            break;
+        } /* end switch case */
+
+        g_free(elem);
+    } /* end for loop */
+}
+
+static void virtio_crypto_init_request(VirtIOCrypto *vcrypto, VirtQueue *vq,
+                                VirtIOCryptoReq *req)
+{
+    req->vcrypto = vcrypto;
+    req->vq = vq;
+    req->in = NULL;
+    req->in_iov = NULL;
+    req->in_num = 0;
+    req->in_len = 0;
+    req->flags = CRYPTODEV_BACKEND_ALG__MAX;
+    req->u.sym_op_info = NULL;
+}
+
+static void virtio_crypto_free_request(VirtIOCryptoReq *req)
+{
+    if (req) {
+        if (req->flags == CRYPTODEV_BACKEND_ALG_SYM) {
+            g_free(req->u.sym_op_info);
+        }
+        g_free(req);
+    }
+}
+
+static void
+virtio_crypto_sym_input_data_helper(VirtIODevice *vdev,
+                VirtIOCryptoReq *req,
+                uint32_t status,
+                CryptoDevBackendSymOpInfo *sym_op_info)
+{
+    size_t s, len;
+
+    if (status != VIRTIO_CRYPTO_OK) {
+        return;
+    }
+
+    len = sym_op_info->dst_len;
+    /* Save the cipher result */
+    s = iov_from_buf(req->in_iov, req->in_num, 0, sym_op_info->dst, len);
+    if (s != len) {
+        virtio_error(vdev, "virtio-crypto dest data incorrect");
+        return;
+    }
+
+    iov_discard_front(&req->in_iov, &req->in_num, len);
+
+    if (sym_op_info->op_type ==
+                      VIRTIO_CRYPTO_SYM_OP_ALGORITHM_CHAINING) {
+        /* Save the digest result */
+        s = iov_from_buf(req->in_iov, req->in_num, 0,
+                         sym_op_info->digest_result,
+                         sym_op_info->digest_result_len);
+        if (s != sym_op_info->digest_result_len) {
+            virtio_error(vdev, "virtio-crypto digest result incorrect");
+        }
+    }
+}
+
+static void virtio_crypto_req_complete(VirtIOCryptoReq *req, uint8_t status)
+{
+    VirtIOCrypto *vcrypto = req->vcrypto;
+    VirtIODevice *vdev = VIRTIO_DEVICE(vcrypto);
+
+    if (req->flags == CRYPTODEV_BACKEND_ALG_SYM) {
+        virtio_crypto_sym_input_data_helper(vdev, req, status,
+                                            req->u.sym_op_info);
+    }
+    stb_p(&req->in->status, status);
+    virtqueue_push(req->vq, &req->elem, req->in_len);
+    virtio_notify(vdev, req->vq);
+}
+
+static VirtIOCryptoReq *
+virtio_crypto_get_request(VirtIOCrypto *s, VirtQueue *vq)
+{
+    VirtIOCryptoReq *req = virtqueue_pop(vq, sizeof(VirtIOCryptoReq));
+
+    if (req) {
+        virtio_crypto_init_request(s, vq, req);
+    }
+    return req;
+}
+
+static CryptoDevBackendSymOpInfo *
+virtio_crypto_sym_op_helper(VirtIODevice *vdev,
+           struct virtio_crypto_cipher_para *cipher_para,
+           struct virtio_crypto_alg_chain_data_para *alg_chain_para,
+           struct iovec *iov, unsigned int out_num)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+    CryptoDevBackendSymOpInfo *op_info;
+    uint32_t src_len = 0, dst_len = 0;
+    uint32_t iv_len = 0;
+    uint32_t aad_len = 0, hash_result_len = 0;
+    uint32_t hash_start_src_offset = 0, len_to_hash = 0;
+    uint32_t cipher_start_src_offset = 0, len_to_cipher = 0;
+
+    size_t max_len, curr_size = 0;
+    size_t s;
+
+    /* Plain cipher */
+    if (cipher_para) {
+        iv_len = ldl_le_p(&cipher_para->iv_len);
+        src_len = ldl_le_p(&cipher_para->src_data_len);
+        dst_len = ldl_le_p(&cipher_para->dst_data_len);
+    } else if (alg_chain_para) { /* Algorithm chain */
+        iv_len = ldl_le_p(&alg_chain_para->iv_len);
+        src_len = ldl_le_p(&alg_chain_para->src_data_len);
+        dst_len = ldl_le_p(&alg_chain_para->dst_data_len);
+
+        aad_len = ldl_le_p(&alg_chain_para->aad_len);
+        hash_result_len = ldl_le_p(&alg_chain_para->hash_result_len);
+        hash_start_src_offset = ldl_le_p(
+                         &alg_chain_para->hash_start_src_offset);
+        cipher_start_src_offset = ldl_le_p(
+                         &alg_chain_para->cipher_start_src_offset);
+        len_to_cipher = ldl_le_p(&alg_chain_para->len_to_cipher);
+        len_to_hash = ldl_le_p(&alg_chain_para->len_to_hash);
+    } else {
+        return NULL;
+    }
+
+    max_len = iv_len + aad_len + src_len + dst_len + hash_result_len;
+    if (unlikely(max_len > vcrypto->conf.max_size)) {
+        virtio_error(vdev, "virtio-crypto too big length");
+        return NULL;
+    }
+
+    op_info = g_malloc0(sizeof(CryptoDevBackendSymOpInfo) + max_len);
+    op_info->iv_len = iv_len;
+    op_info->src_len = src_len;
+    op_info->dst_len = dst_len;
+    op_info->aad_len = aad_len;
+    op_info->digest_result_len = hash_result_len;
+    op_info->hash_start_src_offset = hash_start_src_offset;
+    op_info->len_to_hash = len_to_hash;
+    op_info->cipher_start_src_offset = cipher_start_src_offset;
+    op_info->len_to_cipher = len_to_cipher;
+    /* Handle the initilization vector */
+    if (op_info->iv_len > 0) {
+        DPRINTF("iv_len=%" PRIu32 "\n", op_info->iv_len);
+        op_info->iv = op_info->data + curr_size;
+
+        s = iov_to_buf(iov, out_num, 0, op_info->iv, op_info->iv_len);
+        if (unlikely(s != op_info->iv_len)) {
+            virtio_error(vdev, "virtio-crypto iv incorrect");
+            goto err;
+        }
+        iov_discard_front(&iov, &out_num, op_info->iv_len);
+        curr_size += op_info->iv_len;
+    }
+
+    /* Handle additional authentication data if exists */
+    if (op_info->aad_len > 0) {
+        DPRINTF("aad_len=%" PRIu32 "\n", op_info->aad_len);
+        op_info->aad_data = op_info->data + curr_size;
+
+        s = iov_to_buf(iov, out_num, 0, op_info->aad_data, op_info->aad_len);
+        if (unlikely(s != op_info->aad_len)) {
+            virtio_error(vdev, "virtio-crypto additional auth data incorrect");
+            goto err;
+        }
+        iov_discard_front(&iov, &out_num, op_info->aad_len);
+
+        curr_size += op_info->aad_len;
+    }
+
+    /* Handle the source data */
+    if (op_info->src_len > 0) {
+        DPRINTF("src_len=%" PRIu32 "\n", op_info->src_len);
+        op_info->src = op_info->data + curr_size;
+
+        s = iov_to_buf(iov, out_num, 0, op_info->src, op_info->src_len);
+        if (unlikely(s != op_info->src_len)) {
+            virtio_error(vdev, "virtio-crypto source data incorrect");
+            goto err;
+        }
+        iov_discard_front(&iov, &out_num, op_info->src_len);
+
+        curr_size += op_info->src_len;
+    }
+
+    /* Handle the destination data */
+    op_info->dst = op_info->data + curr_size;
+    curr_size += op_info->dst_len;
+
+    DPRINTF("dst_len=%" PRIu32 "\n", op_info->dst_len);
+
+    /* Handle the hash digest result */
+    if (hash_result_len > 0) {
+        DPRINTF("hash_result_len=%" PRIu32 "\n", hash_result_len);
+        op_info->digest_result = op_info->data + curr_size;
+    }
+
+    return op_info;
+
+err:
+    g_free(op_info);
+    return NULL;
+}
+
+static int
+virtio_crypto_handle_sym_req(VirtIOCrypto *vcrypto,
+               struct virtio_crypto_sym_data_req *req,
+               CryptoDevBackendSymOpInfo **sym_op_info,
+               struct iovec *iov, unsigned int out_num)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(vcrypto);
+    uint32_t op_type;
+    CryptoDevBackendSymOpInfo *op_info;
+
+    op_type = ldl_le_p(&req->op_type);
+
+    if (op_type == VIRTIO_CRYPTO_SYM_OP_CIPHER) {
+        op_info = virtio_crypto_sym_op_helper(vdev, &req->u.cipher.para,
+                                              NULL, iov, out_num);
+        if (!op_info) {
+            return -EFAULT;
+        }
+        op_info->op_type = op_type;
+    } else if (op_type == VIRTIO_CRYPTO_SYM_OP_ALGORITHM_CHAINING) {
+        op_info = virtio_crypto_sym_op_helper(vdev, NULL,
+                                              &req->u.chain.para,
+                                              iov, out_num);
+        if (!op_info) {
+            return -EFAULT;
+        }
+        op_info->op_type = op_type;
+    } else {
+        /* VIRTIO_CRYPTO_SYM_OP_NONE */
+        error_report("virtio-crypto unsupported cipher type");
+        return -VIRTIO_CRYPTO_NOTSUPP;
+    }
+
+    *sym_op_info = op_info;
+
+    return 0;
+}
+
+static int
+virtio_crypto_handle_request(VirtIOCryptoReq *request)
+{
+    VirtIOCrypto *vcrypto = request->vcrypto;
+    VirtIODevice *vdev = VIRTIO_DEVICE(vcrypto);
+    VirtQueueElement *elem = &request->elem;
+    int queue_index = virtio_crypto_vq2q(virtio_get_queue_index(request->vq));
+    struct virtio_crypto_op_data_req req;
+    int ret;
+    struct iovec *in_iov;
+    struct iovec *out_iov;
+    unsigned in_num;
+    unsigned out_num;
+    uint32_t opcode;
+    uint8_t status = VIRTIO_CRYPTO_ERR;
+    uint64_t session_id;
+    CryptoDevBackendSymOpInfo *sym_op_info = NULL;
+    Error *local_err = NULL;
+
+    if (elem->out_num < 1 || elem->in_num < 1) {
+        virtio_error(vdev, "virtio-crypto dataq missing headers");
+        return -1;
+    }
+
+    out_num = elem->out_num;
+    out_iov = elem->out_sg;
+    in_num = elem->in_num;
+    in_iov = elem->in_sg;
+    if (unlikely(iov_to_buf(out_iov, out_num, 0, &req, sizeof(req))
+                != sizeof(req))) {
+        virtio_error(vdev, "virtio-crypto request outhdr too short");
+        return -1;
+    }
+    iov_discard_front(&out_iov, &out_num, sizeof(req));
+
+    if (in_iov[in_num - 1].iov_len <
+            sizeof(struct virtio_crypto_inhdr)) {
+        virtio_error(vdev, "virtio-crypto request inhdr too short");
+        return -1;
+    }
+    /* We always touch the last byte, so just see how big in_iov is. */
+    request->in_len = iov_size(in_iov, in_num);
+    request->in = (void *)in_iov[in_num - 1].iov_base
+              + in_iov[in_num - 1].iov_len
+              - sizeof(struct virtio_crypto_inhdr);
+    iov_discard_back(in_iov, &in_num, sizeof(struct virtio_crypto_inhdr));
+
+    /*
+     * The length of operation result, including dest_data
+     * and digest_result if exists.
+     */
+    request->in_num = in_num;
+    request->in_iov = in_iov;
+
+    opcode = ldl_le_p(&req.header.opcode);
+    session_id = ldq_le_p(&req.header.session_id);
+
+    switch (opcode) {
+    case VIRTIO_CRYPTO_CIPHER_ENCRYPT:
+    case VIRTIO_CRYPTO_CIPHER_DECRYPT:
+        ret = virtio_crypto_handle_sym_req(vcrypto,
+                         &req.u.sym_req,
+                         &sym_op_info,
+                         out_iov, out_num);
+        /* Serious errors, need to reset virtio crypto device */
+        if (ret == -EFAULT) {
+            return -1;
+        } else if (ret == -VIRTIO_CRYPTO_NOTSUPP) {
+            virtio_crypto_req_complete(request, VIRTIO_CRYPTO_NOTSUPP);
+            virtio_crypto_free_request(request);
+        } else {
+            sym_op_info->session_id = session_id;
+
+            /* Set request's parameter */
+            request->flags = CRYPTODEV_BACKEND_ALG_SYM;
+            request->u.sym_op_info = sym_op_info;
+            ret = cryptodev_backend_crypto_operation(vcrypto->cryptodev,
+                                    request, queue_index, &local_err);
+            if (ret < 0) {
+                status = -ret;
+                if (local_err) {
+                    error_report_err(local_err);
+                }
+            } else { /* ret == VIRTIO_CRYPTO_OK */
+                status = ret;
+            }
+            virtio_crypto_req_complete(request, status);
+            virtio_crypto_free_request(request);
+        }
+        break;
+    case VIRTIO_CRYPTO_HASH:
+    case VIRTIO_CRYPTO_MAC:
+    case VIRTIO_CRYPTO_AEAD_ENCRYPT:
+    case VIRTIO_CRYPTO_AEAD_DECRYPT:
+    default:
+        error_report("virtio-crypto unsupported dataq opcode: %u",
+                     opcode);
+        virtio_crypto_req_complete(request, VIRTIO_CRYPTO_NOTSUPP);
+        virtio_crypto_free_request(request);
+    }
+
+    return 0;
+}
+
+static void virtio_crypto_handle_dataq(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+    VirtIOCryptoReq *req;
+
+    while ((req = virtio_crypto_get_request(vcrypto, vq))) {
+        if (virtio_crypto_handle_request(req) < 0) {
+            virtqueue_detach_element(req->vq, &req->elem, 0);
+            virtio_crypto_free_request(req);
+            break;
+        }
+    }
+}
+
+static void virtio_crypto_dataq_bh(void *opaque)
+{
+    VirtIOCryptoQueue *q = opaque;
+    VirtIOCrypto *vcrypto = q->vcrypto;
+    VirtIODevice *vdev = VIRTIO_DEVICE(vcrypto);
+
+    /* This happens when device was stopped but BH wasn't. */
+    if (!vdev->vm_running) {
+        return;
+    }
+
+    /* Just in case the driver is not ready on more */
+    if (unlikely(!(vdev->status & VIRTIO_CONFIG_S_DRIVER_OK))) {
+        return;
+    }
+
+    virtio_crypto_handle_dataq(vdev, q->dataq);
+    virtio_queue_set_notification(q->dataq, 1);
+}
+
+static void
+virtio_crypto_handle_dataq_bh(VirtIODevice *vdev, VirtQueue *vq)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+    VirtIOCryptoQueue *q =
+         &vcrypto->vqs[virtio_crypto_vq2q(virtio_get_queue_index(vq))];
+
+    /* This happens when device was stopped but VCPU wasn't. */
+    if (!vdev->vm_running) {
+        return;
+    }
+    virtio_queue_set_notification(vq, 0);
+    qemu_bh_schedule(q->dataq_bh);
+}
+
+static uint64_t virtio_crypto_get_features(VirtIODevice *vdev,
+                                           uint64_t features,
+                                           Error **errp)
+{
+    return features;
+}
+
+static void virtio_crypto_reset(VirtIODevice *vdev)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+    /* multiqueue is disabled by default */
+    vcrypto->curr_queues = 1;
+    if (!vcrypto->cryptodev->ready) {
+        vcrypto->status &= ~VIRTIO_CRYPTO_S_HW_READY;
+    } else {
+        vcrypto->status |= VIRTIO_CRYPTO_S_HW_READY;
+    }
+}
+
+static void virtio_crypto_init_config(VirtIODevice *vdev)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(vdev);
+
+    vcrypto->conf.crypto_services =
+                     vcrypto->conf.cryptodev->conf.crypto_services;
+    vcrypto->conf.cipher_algo_l =
+                     vcrypto->conf.cryptodev->conf.cipher_algo_l;
+    vcrypto->conf.cipher_algo_h =
+                     vcrypto->conf.cryptodev->conf.cipher_algo_h;
+    vcrypto->conf.hash_algo = vcrypto->conf.cryptodev->conf.hash_algo;
+    vcrypto->conf.mac_algo_l = vcrypto->conf.cryptodev->conf.mac_algo_l;
+    vcrypto->conf.mac_algo_h = vcrypto->conf.cryptodev->conf.mac_algo_h;
+    vcrypto->conf.aead_algo = vcrypto->conf.cryptodev->conf.aead_algo;
+    vcrypto->conf.max_cipher_key_len =
+                  vcrypto->conf.cryptodev->conf.max_cipher_key_len;
+    vcrypto->conf.max_auth_key_len =
+                  vcrypto->conf.cryptodev->conf.max_auth_key_len;
+    vcrypto->conf.max_size = vcrypto->conf.cryptodev->conf.max_size;
+}
+
+static void virtio_crypto_device_realize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(dev);
+    int i;
+
+    vcrypto->cryptodev = vcrypto->conf.cryptodev;
+    if (vcrypto->cryptodev == NULL) {
+        error_setg(errp, "'cryptodev' parameter expects a valid object");
+        return;
+    }
+
+    vcrypto->max_queues = MAX(vcrypto->cryptodev->conf.peers.queues, 1);
+    if (vcrypto->max_queues + 1 > VIRTIO_QUEUE_MAX) {
+        error_setg(errp, "Invalid number of queues (= %" PRIu32 "), "
+                   "must be a postive integer less than %d.",
+                   vcrypto->max_queues, VIRTIO_QUEUE_MAX);
+        return;
+    }
+
+    virtio_init(vdev, "virtio-crypto", VIRTIO_ID_CRYPTO, vcrypto->config_size);
+    vcrypto->curr_queues = 1;
+    vcrypto->vqs = g_malloc0(sizeof(VirtIOCryptoQueue) * vcrypto->max_queues);
+    for (i = 0; i < vcrypto->max_queues; i++) {
+        vcrypto->vqs[i].dataq =
+                 virtio_add_queue(vdev, 1024, virtio_crypto_handle_dataq_bh);
+        vcrypto->vqs[i].dataq_bh =
+                 qemu_bh_new(virtio_crypto_dataq_bh, &vcrypto->vqs[i]);
+        vcrypto->vqs[i].vcrypto = vcrypto;
+    }
+
+    vcrypto->ctrl_vq = virtio_add_queue(vdev, 64, virtio_crypto_handle_ctrl);
+    if (!vcrypto->cryptodev->ready) {
+        vcrypto->status &= ~VIRTIO_CRYPTO_S_HW_READY;
+    } else {
+        vcrypto->status |= VIRTIO_CRYPTO_S_HW_READY;
+    }
+
+    virtio_crypto_init_config(vdev);
+}
+
+static void virtio_crypto_device_unrealize(DeviceState *dev, Error **errp)
+{
+    VirtIODevice *vdev = VIRTIO_DEVICE(dev);
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(dev);
+    VirtIOCryptoQueue *q;
+    int i, max_queues;
+
+    max_queues = vcrypto->multiqueue ? vcrypto->max_queues : 1;
+    for (i = 0; i < max_queues; i++) {
+        virtio_del_queue(vdev, i);
+        q = &vcrypto->vqs[i];
+        qemu_bh_delete(q->dataq_bh);
+    }
+
+    g_free(vcrypto->vqs);
+
+    virtio_cleanup(vdev);
+}
+
+static const VMStateDescription vmstate_virtio_crypto = {
+    .name = "virtio-crypto",
+    .minimum_version_id = VIRTIO_CRYPTO_VM_VERSION,
+    .version_id = VIRTIO_CRYPTO_VM_VERSION,
+    .fields = (VMStateField[]) {
+        VMSTATE_VIRTIO_DEVICE,
+        VMSTATE_END_OF_LIST()
+    },
+};
+
+static Property virtio_crypto_properties[] = {
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void virtio_crypto_get_config(VirtIODevice *vdev, uint8_t *config)
+{
+    VirtIOCrypto *c = VIRTIO_CRYPTO(vdev);
+    struct virtio_crypto_config crypto_cfg;
+
+    /*
+     * Virtio-crypto device conforms to VIRTIO 1.0 which is always LE,
+     * so we can use LE accessors directly.
+     */
+    stl_le_p(&crypto_cfg.status, c->status);
+    stl_le_p(&crypto_cfg.max_dataqueues, c->max_queues);
+    stl_le_p(&crypto_cfg.crypto_services, c->conf.crypto_services);
+    stl_le_p(&crypto_cfg.cipher_algo_l, c->conf.cipher_algo_l);
+    stl_le_p(&crypto_cfg.cipher_algo_h, c->conf.cipher_algo_h);
+    stl_le_p(&crypto_cfg.hash_algo, c->conf.hash_algo);
+    stl_le_p(&crypto_cfg.mac_algo_l, c->conf.mac_algo_l);
+    stl_le_p(&crypto_cfg.mac_algo_h, c->conf.mac_algo_h);
+    stl_le_p(&crypto_cfg.aead_algo, c->conf.aead_algo);
+    stl_le_p(&crypto_cfg.max_cipher_key_len, c->conf.max_cipher_key_len);
+    stl_le_p(&crypto_cfg.max_auth_key_len, c->conf.max_auth_key_len);
+    stq_le_p(&crypto_cfg.max_size, c->conf.max_size);
+
+    memcpy(config, &crypto_cfg, c->config_size);
+}
+
+static void virtio_crypto_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
+
+    dc->props = virtio_crypto_properties;
+    dc->vmsd = &vmstate_virtio_crypto;
+    set_bit(DEVICE_CATEGORY_MISC, dc->categories);
+    vdc->realize = virtio_crypto_device_realize;
+    vdc->unrealize = virtio_crypto_device_unrealize;
+    vdc->get_config = virtio_crypto_get_config;
+    vdc->get_features = virtio_crypto_get_features;
+    vdc->reset = virtio_crypto_reset;
+}
+
+static void virtio_crypto_instance_init(Object *obj)
+{
+    VirtIOCrypto *vcrypto = VIRTIO_CRYPTO(obj);
+
+    /*
+     * The default config_size is sizeof(struct virtio_crypto_config).
+     * Can be overriden with virtio_crypto_set_config_size.
+     */
+    vcrypto->config_size = sizeof(struct virtio_crypto_config);
+
+    object_property_add_link(obj, "cryptodev",
+                             TYPE_CRYPTODEV_BACKEND,
+                             (Object **)&vcrypto->conf.cryptodev,
+                             qdev_prop_allow_set_link_before_realize,
+                             OBJ_PROP_LINK_UNREF_ON_RELEASE, NULL);
+}
+
+static const TypeInfo virtio_crypto_info = {
+    .name = TYPE_VIRTIO_CRYPTO,
+    .parent = TYPE_VIRTIO_DEVICE,
+    .instance_size = sizeof(VirtIOCrypto),
+    .instance_init = virtio_crypto_instance_init,
+    .class_init = virtio_crypto_class_init,
+};
+
+static void virtio_register_types(void)
+{
+    type_register_static(&virtio_crypto_info);
+}
+
+type_init(virtio_register_types)
diff --git a/hw/virtio/virtio-mmio.c b/hw/virtio/virtio-mmio.c
index 13798b3cb8..a30270f902 100644
--- a/hw/virtio/virtio-mmio.c
+++ b/hw/virtio/virtio-mmio.c
@@ -89,38 +89,12 @@ typedef struct {
     uint32_t guest_page_shift;
     /* virtio-bus */
     VirtioBusState bus;
-    bool ioeventfd_disabled;
-    bool ioeventfd_started;
     bool format_transport_address;
 } VirtIOMMIOProxy;
 
-static bool virtio_mmio_ioeventfd_started(DeviceState *d)
+static bool virtio_mmio_ioeventfd_enabled(DeviceState *d)
 {
-    VirtIOMMIOProxy *proxy = VIRTIO_MMIO(d);
-
-    return proxy->ioeventfd_started;
-}
-
-static void virtio_mmio_ioeventfd_set_started(DeviceState *d, bool started,
-                                              bool err)
-{
-    VirtIOMMIOProxy *proxy = VIRTIO_MMIO(d);
-
-    proxy->ioeventfd_started = started;
-}
-
-static bool virtio_mmio_ioeventfd_disabled(DeviceState *d)
-{
-    VirtIOMMIOProxy *proxy = VIRTIO_MMIO(d);
-
-    return !kvm_eventfds_enabled() || proxy->ioeventfd_disabled;
-}
-
-static void virtio_mmio_ioeventfd_set_disabled(DeviceState *d, bool disabled)
-{
-    VirtIOMMIOProxy *proxy = VIRTIO_MMIO(d);
-
-    proxy->ioeventfd_disabled = disabled;
+    return kvm_eventfds_enabled();
 }
 
 static int virtio_mmio_ioeventfd_assign(DeviceState *d,
@@ -557,10 +531,7 @@ static void virtio_mmio_bus_class_init(ObjectClass *klass, void *data)
     k->save_config = virtio_mmio_save_config;
     k->load_config = virtio_mmio_load_config;
     k->set_guest_notifiers = virtio_mmio_set_guest_notifiers;
-    k->ioeventfd_started = virtio_mmio_ioeventfd_started;
-    k->ioeventfd_set_started = virtio_mmio_ioeventfd_set_started;
-    k->ioeventfd_disabled = virtio_mmio_ioeventfd_disabled;
-    k->ioeventfd_set_disabled = virtio_mmio_ioeventfd_set_disabled;
+    k->ioeventfd_enabled = virtio_mmio_ioeventfd_enabled;
     k->ioeventfd_assign = virtio_mmio_ioeventfd_assign;
     k->has_variable_vring_alignment = true;
     bus_class->max_dev = 1;
diff --git a/hw/virtio/virtio-pci.c b/hw/virtio/virtio-pci.c
index 06831de5ff..62001b46d7 100644
--- a/hw/virtio/virtio-pci.c
+++ b/hw/virtio/virtio-pci.c
@@ -262,34 +262,11 @@ static int virtio_pci_load_queue(DeviceState *d, int n, QEMUFile *f)
     return 0;
 }
 
-static bool virtio_pci_ioeventfd_started(DeviceState *d)
+static bool virtio_pci_ioeventfd_enabled(DeviceState *d)
 {
     VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
 
-    return proxy->ioeventfd_started;
-}
-
-static void virtio_pci_ioeventfd_set_started(DeviceState *d, bool started,
-                                             bool err)
-{
-    VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
-
-    proxy->ioeventfd_started = started;
-}
-
-static bool virtio_pci_ioeventfd_disabled(DeviceState *d)
-{
-    VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
-
-    return proxy->ioeventfd_disabled ||
-        !(proxy->flags & VIRTIO_PCI_FLAG_USE_IOEVENTFD);
-}
-
-static void virtio_pci_ioeventfd_set_disabled(DeviceState *d, bool disabled)
-{
-    VirtIOPCIProxy *proxy = to_virtio_pci_proxy(d);
-
-    proxy->ioeventfd_disabled = disabled;
+    return (proxy->flags & VIRTIO_PCI_FLAG_USE_IOEVENTFD) != 0;
 }
 
 #define QEMU_VIRTIO_PCI_QUEUE_MEM_MULT 0x1000
@@ -1719,10 +1696,6 @@ static void virtio_pci_device_plugged(DeviceState *d, Error **errp)
         pci_register_bar(&proxy->pci_dev, proxy->legacy_io_bar_idx,
                          PCI_BASE_ADDRESS_SPACE_IO, &proxy->bar);
     }
-
-    if (!kvm_has_many_ioeventfds()) {
-        proxy->flags &= ~VIRTIO_PCI_FLAG_USE_IOEVENTFD;
-    }
 }
 
 static void virtio_pci_device_unplugged(DeviceState *d)
@@ -1751,6 +1724,10 @@ static void virtio_pci_realize(PCIDevice *pci_dev, Error **errp)
     bool pcie_port = pci_bus_is_express(pci_dev->bus) &&
                      !pci_bus_is_root(pci_dev->bus);
 
+    if (!kvm_has_many_ioeventfds()) {
+        proxy->flags &= ~VIRTIO_PCI_FLAG_USE_IOEVENTFD;
+    }
+
     /*
      * virtio pci bar layout used by default.
      * subclasses can re-arrange things if needed.
@@ -2539,10 +2516,7 @@ static void virtio_pci_bus_class_init(ObjectClass *klass, void *data)
     k->device_plugged = virtio_pci_device_plugged;
     k->device_unplugged = virtio_pci_device_unplugged;
     k->query_nvectors = virtio_pci_query_nvectors;
-    k->ioeventfd_started = virtio_pci_ioeventfd_started;
-    k->ioeventfd_set_started = virtio_pci_ioeventfd_set_started;
-    k->ioeventfd_disabled = virtio_pci_ioeventfd_disabled;
-    k->ioeventfd_set_disabled = virtio_pci_ioeventfd_set_disabled;
+    k->ioeventfd_enabled = virtio_pci_ioeventfd_enabled;
     k->ioeventfd_assign = virtio_pci_ioeventfd_assign;
 }
 
diff --git a/hw/virtio/virtio-pci.h b/hw/virtio/virtio-pci.h
index b4edea6987..b2a996fa83 100644
--- a/hw/virtio/virtio-pci.h
+++ b/hw/virtio/virtio-pci.h
@@ -25,6 +25,8 @@
 #include "hw/virtio/virtio-bus.h"
 #include "hw/virtio/virtio-input.h"
 #include "hw/virtio/virtio-gpu.h"
+#include "hw/virtio/virtio-crypto.h"
+
 #ifdef CONFIG_VIRTFS
 #include "hw/9pfs/virtio-9p.h"
 #endif
@@ -48,6 +50,7 @@ typedef struct VirtIOInputHIDPCI VirtIOInputHIDPCI;
 typedef struct VirtIOInputHostPCI VirtIOInputHostPCI;
 typedef struct VirtIOGPUPCI VirtIOGPUPCI;
 typedef struct VHostVSockPCI VHostVSockPCI;
+typedef struct VirtIOCryptoPCI VirtIOCryptoPCI;
 
 /* virtio-pci-bus */
 
@@ -158,8 +161,6 @@ struct VirtIOPCIProxy {
     uint32_t guest_features[2];
     VirtIOPCIQueue vqs[VIRTIO_QUEUE_MAX];
 
-    bool ioeventfd_disabled;
-    bool ioeventfd_started;
     VirtIOIRQFD *vector_irqfd;
     int nvqs_with_notifiers;
     VirtioBusState bus;
@@ -352,6 +353,18 @@ struct VHostVSockPCI {
 };
 #endif
 
+/*
+ * virtio-crypto-pci: This extends VirtioPCIProxy.
+ */
+#define TYPE_VIRTIO_CRYPTO_PCI "virtio-crypto-pci"
+#define VIRTIO_CRYPTO_PCI(obj) \
+        OBJECT_CHECK(VirtIOCryptoPCI, (obj), TYPE_VIRTIO_CRYPTO_PCI)
+
+struct VirtIOCryptoPCI {
+    VirtIOPCIProxy parent_obj;
+    VirtIOCrypto vdev;
+};
+
 /* Virtio ABI version, if we increment this, we break the guest driver. */
 #define VIRTIO_PCI_ABI_VERSION          0
 
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index d48d1a98a7..bcbcfe063c 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -97,7 +97,6 @@ struct VirtQueue
     uint16_t vector;
     VirtIOHandleOutput handle_output;
     VirtIOHandleOutput handle_aio_output;
-    bool use_aio;
     VirtIODevice *vdev;
     EventNotifier guest_notifier;
     EventNotifier host_notifier;
@@ -1287,9 +1286,8 @@ void virtio_queue_set_vector(VirtIODevice *vdev, int n, uint16_t vector)
     }
 }
 
-static VirtQueue *virtio_add_queue_internal(VirtIODevice *vdev, int queue_size,
-                                            VirtIOHandleOutput handle_output,
-                                            bool use_aio)
+VirtQueue *virtio_add_queue(VirtIODevice *vdev, int queue_size,
+                            VirtIOHandleOutput handle_output)
 {
     int i;
 
@@ -1306,28 +1304,10 @@ static VirtQueue *virtio_add_queue_internal(VirtIODevice *vdev, int queue_size,
     vdev->vq[i].vring.align = VIRTIO_PCI_VRING_ALIGN;
     vdev->vq[i].handle_output = handle_output;
     vdev->vq[i].handle_aio_output = NULL;
-    vdev->vq[i].use_aio = use_aio;
 
     return &vdev->vq[i];
 }
 
-/* Add a virt queue and mark AIO.
- * An AIO queue will use the AioContext based event interface instead of the
- * default IOHandler and EventNotifier interface.
- */
-VirtQueue *virtio_add_queue_aio(VirtIODevice *vdev, int queue_size,
-                                VirtIOHandleOutput handle_output)
-{
-    return virtio_add_queue_internal(vdev, queue_size, handle_output, true);
-}
-
-/* Add a normal virt queue (on the contrary to the AIO version above. */
-VirtQueue *virtio_add_queue(VirtIODevice *vdev, int queue_size,
-                            VirtIOHandleOutput handle_output)
-{
-    return virtio_add_queue_internal(vdev, queue_size, handle_output, false);
-}
-
 void virtio_del_queue(VirtIODevice *vdev, int n)
 {
     if (n < 0 || n >= VIRTIO_QUEUE_MAX) {
@@ -1635,6 +1615,10 @@ void virtio_save(VirtIODevice *vdev, QEMUFile *f)
         vdc->save(vdev, f);
     }
 
+    if (vdc->vmsd) {
+        vmstate_save_state(f, vdc->vmsd, vdev, NULL);
+    }
+
     /* Subsections */
     vmstate_save_state(f, &vmstate_virtio, vdev, NULL);
 }
@@ -1781,6 +1765,13 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
         }
     }
 
+    if (vdc->vmsd) {
+        ret = vmstate_load_state(f, vdc->vmsd, vdev, version_id);
+        if (ret) {
+            return ret;
+        }
+    }
+
     /* Subsections */
     ret = vmstate_load_state(f, &vmstate_virtio, vdev, 1);
     if (ret) {
@@ -2051,7 +2042,7 @@ void virtio_queue_aio_set_host_notifier_handler(VirtQueue *vq, AioContext *ctx,
     }
 }
 
-static void virtio_queue_host_notifier_read(EventNotifier *n)
+void virtio_queue_host_notifier_read(EventNotifier *n)
 {
     VirtQueue *vq = container_of(n, VirtQueue, host_notifier);
     if (event_notifier_test_and_clear(n)) {
@@ -2059,32 +2050,6 @@ static void virtio_queue_host_notifier_read(EventNotifier *n)
     }
 }
 
-void virtio_queue_set_host_notifier_fd_handler(VirtQueue *vq, bool assign,
-                                               bool set_handler)
-{
-    AioContext *ctx = qemu_get_aio_context();
-    if (assign && set_handler) {
-        if (vq->use_aio) {
-            aio_set_event_notifier(ctx, &vq->host_notifier, true,
-                                   virtio_queue_host_notifier_read);
-        } else {
-            event_notifier_set_handler(&vq->host_notifier, true,
-                                       virtio_queue_host_notifier_read);
-        }
-    } else {
-        if (vq->use_aio) {
-            aio_set_event_notifier(ctx, &vq->host_notifier, true, NULL);
-        } else {
-            event_notifier_set_handler(&vq->host_notifier, true, NULL);
-        }
-    }
-    if (!assign) {
-        /* Test and clear notifier before after disabling event,
-         * in case poll callback didn't have time to run. */
-        virtio_queue_host_notifier_read(&vq->host_notifier);
-    }
-}
-
 EventNotifier *virtio_queue_get_host_notifier(VirtQueue *vq)
 {
     return &vq->host_notifier;
@@ -2118,6 +2083,9 @@ static void virtio_device_realize(DeviceState *dev, Error **errp)
     VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(dev);
     Error *err = NULL;
 
+    /* Devices should either use vmsd or the load/save methods */
+    assert(!vdc->vmsd || !vdc->load);
+
     if (vdc->realize != NULL) {
         vdc->realize(dev, &err);
         if (err != NULL) {
@@ -2158,15 +2126,102 @@ static Property virtio_properties[] = {
     DEFINE_PROP_END_OF_LIST(),
 };
 
+static int virtio_device_start_ioeventfd_impl(VirtIODevice *vdev)
+{
+    VirtioBusState *qbus = VIRTIO_BUS(qdev_get_parent_bus(DEVICE(vdev)));
+    int n, r, err;
+
+    for (n = 0; n < VIRTIO_QUEUE_MAX; n++) {
+        VirtQueue *vq = &vdev->vq[n];
+        if (!virtio_queue_get_num(vdev, n)) {
+            continue;
+        }
+        r = virtio_bus_set_host_notifier(qbus, n, true);
+        if (r < 0) {
+            err = r;
+            goto assign_error;
+        }
+        event_notifier_set_handler(&vq->host_notifier, true,
+                                   virtio_queue_host_notifier_read);
+    }
+
+    for (n = 0; n < VIRTIO_QUEUE_MAX; n++) {
+        /* Kick right away to begin processing requests already in vring */
+        VirtQueue *vq = &vdev->vq[n];
+        if (!vq->vring.num) {
+            continue;
+        }
+        event_notifier_set(&vq->host_notifier);
+    }
+    return 0;
+
+assign_error:
+    while (--n >= 0) {
+        VirtQueue *vq = &vdev->vq[n];
+        if (!virtio_queue_get_num(vdev, n)) {
+            continue;
+        }
+
+        event_notifier_set_handler(&vq->host_notifier, true, NULL);
+        r = virtio_bus_set_host_notifier(qbus, n, false);
+        assert(r >= 0);
+    }
+    return err;
+}
+
+int virtio_device_start_ioeventfd(VirtIODevice *vdev)
+{
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
+    VirtioBusState *vbus = VIRTIO_BUS(qbus);
+
+    return virtio_bus_start_ioeventfd(vbus);
+}
+
+static void virtio_device_stop_ioeventfd_impl(VirtIODevice *vdev)
+{
+    VirtioBusState *qbus = VIRTIO_BUS(qdev_get_parent_bus(DEVICE(vdev)));
+    int n, r;
+
+    for (n = 0; n < VIRTIO_QUEUE_MAX; n++) {
+        VirtQueue *vq = &vdev->vq[n];
+
+        if (!virtio_queue_get_num(vdev, n)) {
+            continue;
+        }
+        event_notifier_set_handler(&vq->host_notifier, true, NULL);
+        r = virtio_bus_set_host_notifier(qbus, n, false);
+        assert(r >= 0);
+    }
+}
+
+void virtio_device_stop_ioeventfd(VirtIODevice *vdev)
+{
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
+    VirtioBusState *vbus = VIRTIO_BUS(qbus);
+
+    virtio_bus_stop_ioeventfd(vbus);
+}
+
 static void virtio_device_class_init(ObjectClass *klass, void *data)
 {
     /* Set the default value here. */
+    VirtioDeviceClass *vdc = VIRTIO_DEVICE_CLASS(klass);
     DeviceClass *dc = DEVICE_CLASS(klass);
 
     dc->realize = virtio_device_realize;
     dc->unrealize = virtio_device_unrealize;
     dc->bus_type = TYPE_VIRTIO_BUS;
     dc->props = virtio_properties;
+    vdc->start_ioeventfd = virtio_device_start_ioeventfd_impl;
+    vdc->stop_ioeventfd = virtio_device_stop_ioeventfd_impl;
+}
+
+bool virtio_device_ioeventfd_enabled(VirtIODevice *vdev)
+{
+    BusState *qbus = qdev_get_parent_bus(DEVICE(vdev));
+    VirtioBusState *vbus = VIRTIO_BUS(qbus);
+
+    return virtio_bus_ioeventfd_enabled(vbus);
 }
 
 static const TypeInfo virtio_device_info = {