Merge tag 'migration-20231102-pull-request' of https://gitlab.com/juan.quintela/qemu into staging

Migration Pull request (20231102)

Hi

In this pull request:

- migration reboot mode (steve)
  * I disabled the test because our CI don't like programs using so
    much shared memory.  Searching for a fix.
- test for postcopy recover (fabiano)
- MigrateAddress QAPI (het)
- better return path error handling (peter)
- traces for downtime (peter)
- vmstate_register() check for duplicates (juan)
  thomas find better solutions for s390x and ipmi.
  now also works on s390x

Please, apply.

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEGJn/jt6/WMzuA0uC9IfvGFhy1yMFAmVDipMACgkQ9IfvGFhy
# 1yNYnQ/9E5Cywsoqljqa/9FiKBSII2qMrmkfu6JLKqePnsh5pFZiukbudYRuJCCe
# ZTDEmD0NmKRJbDx2xRU1qx/e6gKJy+gz37KP89Buuh/WwZHPboPYtxQpGvCSiH26
# J3i+1+TgaqmkLzcO35wa8tp6gneQclWeAwKgMvdb4cm2pJEhgWRKI62ccyLzxeve
# UCzFQn60t55ETyVZGnRD4YwdTQvGKH+DPlyTuJOLR3DePuvZd8EdH+ypvB4RLAy7
# 3+CuQOxmF5LRXPbpJuAeOsudbmhhHzrO/yL7ZmsiKQTthsJv+SzC1bO94jhQrawZ
# Q7GCii5KpGq0KnRTRKZRGk6XKwxcYRduXMX3R5tXuVmDmCZsjhXzziU8yEdftph8
# 5TJdk1o0Gb043EFu81mrsQYS+9yJqe6sy6m3PTJaec54cAty5ln+c17WOvpAOaSV
# +1phe05ftuVPmQ3KWhbIR/tCmavNLwEZxpVIfyaKJx04bFbtQ9gRpRyURORX4KXc
# s4WXvNirQEohxYBnP4TPvA09xBTW3V08pk/wRDwt0YDXnLiqCltOuxD8r05K8K4B
# MkCLcWj0g7he2tBkF60oz1KSIE0oTB81um9AzLIv5F2YSYLaJM5BIcoC437MR2f4
# MOR7drR1fP5GsRu/SeU5BWvhVq3IvdOxR7G2MLNRJJvl7ZtGXDc=
# =uaqL
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 02 Nov 2023 19:40:03 HKT
# gpg:                using RSA key 1899FF8EDEBF58CCEE034B82F487EF185872D723
# gpg: Good signature from "Juan Quintela <quintela@redhat.com>" [full]
# gpg:                 aka "Juan Quintela <quintela@trasno.org>" [full]
# Primary key fingerprint: 1899 FF8E DEBF 58CC EE03  4B82 F487 EF18 5872 D723

* tag 'migration-20231102-pull-request' of https://gitlab.com/juan.quintela/qemu: (40 commits)
  migration: modify test_multifd_tcp_none() to use new QAPI syntax.
  migration: Implement MigrateChannelList to hmp migration flow.
  migration: Implement MigrateChannelList to qmp migration flow.
  migration: modify migration_channels_and_uri_compatible() for new QAPI syntax
  migration: New migrate and migrate-incoming argument 'channels'
  migration: Convert the file backend to the new QAPI syntax
  migration: convert exec backend to accept MigrateAddress.
  migration: convert rdma backend to accept MigrateAddress
  migration: convert socket backend to accept MigrateAddress
  migration: convert migration 'uri' into 'MigrateAddress'
  migration: New QAPI type 'MigrateAddress'
  migration: Change ram_dirty_bitmap_reload() retval to bool
  tests/migration-test: Add a test for postcopy hangs during RECOVER
  migration: Allow network to fail even during recovery
  migration: Refactor error handling in source return path
  tests/qtest: migration: add reboot mode test
  cpr: reboot mode
  cpr: relax vhost migration blockers
  cpr: relax blockdev migration blockers
  migration: per-mode blockers
  ...

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

audio/audio.c

@@ -1781,7 +1781,7 @@ static AudioState *audio_init(Audiodev *dev, Error **errp)
 
     QTAILQ_INSERT_TAIL(&audio_states, s, list);
     QLIST_INIT (&s->card_head);
-    vmstate_register (NULL, 0, &vmstate_audio, s);
+    vmstate_register_any(NULL, &vmstate_audio, s);
     return s;
 
 out:

backends/dbus-vmstate.c

@@ -426,8 +426,7 @@ dbus_vmstate_complete(UserCreatable *uc, Error **errp)
         return;
     }
 
-    if (vmstate_register(VMSTATE_IF(self), VMSTATE_INSTANCE_ID_ANY,
-                         &dbus_vmstate, self) < 0) {
+    if (vmstate_register_any(VMSTATE_IF(self), &dbus_vmstate, self) < 0) {
         error_setg(errp, "Failed to register vmstate");
     }
 }

backends/tpm/tpm_emulator.c

@@ -975,8 +975,7 @@ static void tpm_emulator_inst_init(Object *obj)
         qemu_add_vm_change_state_handler(tpm_emulator_vm_state_change,
                                          tpm_emu);
 
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY,
-                     &vmstate_tpm_emulator, obj);
+    vmstate_register_any(NULL, &vmstate_tpm_emulator, obj);
 }
 
 /*

block/parallels.c

@@ -1369,7 +1369,7 @@ static int parallels_open(BlockDriverState *bs, QDict *options, int flags,
                bdrv_get_device_or_node_name(bs));
     bdrv_graph_rdunlock_main_loop();
 
-    ret = migrate_add_blocker(&s->migration_blocker, errp);
+    ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
     if (ret < 0) {
         goto fail;
     }

block/qcow.c

@@ -307,7 +307,7 @@ static int qcow_open(BlockDriverState *bs, QDict *options, int flags,
                bdrv_get_device_or_node_name(bs));
     bdrv_graph_rdunlock_main_loop();
 
-    ret = migrate_add_blocker(&s->migration_blocker, errp);
+    ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
     if (ret < 0) {
         goto fail;
     }

block/vdi.c

@@ -498,7 +498,7 @@ static int vdi_open(BlockDriverState *bs, QDict *options, int flags,
                bdrv_get_device_or_node_name(bs));
     bdrv_graph_rdunlock_main_loop();
 
-    ret = migrate_add_blocker(&s->migration_blocker, errp);
+    ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
     if (ret < 0) {
         goto fail_free_bmap;
     }

block/vhdx.c

@@ -1096,7 +1096,7 @@ static int vhdx_open(BlockDriverState *bs, QDict *options, int flags,
     error_setg(&s->migration_blocker, "The vhdx format used by node '%s' "
                "does not support live migration",
                bdrv_get_device_or_node_name(bs));
-    ret = migrate_add_blocker(&s->migration_blocker, errp);
+    ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
     if (ret < 0) {
         goto fail;
     }

block/vmdk.c

@@ -1386,7 +1386,7 @@ static int vmdk_open(BlockDriverState *bs, QDict *options, int flags,
     error_setg(&s->migration_blocker, "The vmdk format used by node '%s' "
                "does not support live migration",
                bdrv_get_device_or_node_name(bs));
-    ret = migrate_add_blocker(&s->migration_blocker, errp);
+    ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
     if (ret < 0) {
         goto fail;
     }

block/vpc.c

@@ -452,7 +452,7 @@ static int vpc_open(BlockDriverState *bs, QDict *options, int flags,
                bdrv_get_device_or_node_name(bs));
     bdrv_graph_rdunlock_main_loop();
 
-    ret = migrate_add_blocker(&s->migration_blocker, errp);
+    ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
     if (ret < 0) {
         goto fail;
     }

block/vvfat.c

@@ -1268,7 +1268,7 @@ static int vvfat_open(BlockDriverState *bs, QDict *options, int flags,
                    "The vvfat (rw) format used by node '%s' "
                    "does not support live migration",
                    bdrv_get_device_or_node_name(bs));
-        ret = migrate_add_blocker(&s->migration_blocker, errp);
+        ret = migrate_add_blocker_normal(&s->migration_blocker, errp);
         if (ret < 0) {
             goto fail;
         }

docs/devel/migration.rst

@@ -167,13 +167,17 @@ An example (from hw/input/pckbd.c)
       }
   };
 
-We are declaring the state with name "pckbd".
-The ``version_id`` is 3, and the fields are 4 uint8_t in a KBDState structure.
-We registered this with:
+We are declaring the state with name "pckbd".  The ``version_id`` is
+3, and there are 4 uint8_t fields in the KBDState structure.  We
+registered this ``VMSTATEDescription`` with one of the following
+functions.  The first one will generate a device ``instance_id``
+different for each registration.  Use the second one if you already
+have an id that is different for each instance of the device:
 
 .. code:: c
 
-    vmstate_register(NULL, 0, &vmstate_kbd, s);
+    vmstate_register_any(NULL, &vmstate_kbd, s);
+    vmstate_register(NULL, instance_id, &vmstate_kbd, s);
 
 For devices that are ``qdev`` based, we can register the device in the class
 init function:

hw/core/qdev-properties-system.c

@@ -673,6 +673,20 @@ const PropertyInfo qdev_prop_multifd_compression = {
     .set_default_value = qdev_propinfo_set_default_value_enum,
 };
 
+/* --- MigMode --- */
+
+QEMU_BUILD_BUG_ON(sizeof(MigMode) != sizeof(int));
+
+const PropertyInfo qdev_prop_mig_mode = {
+    .name = "MigMode",
+    .description = "mig_mode values, "
+                   "normal,cpr-reboot",
+    .enum_table = &MigMode_lookup,
+    .get = qdev_propinfo_get_enum,
+    .set = qdev_propinfo_set_enum,
+    .set_default_value = qdev_propinfo_set_default_value_enum,
+};
+
 /* --- Reserved Region --- */
 
 /*

hw/display/vmware_vga.c

@@ -1264,7 +1264,7 @@ static void vmsvga_init(DeviceState *dev, struct vmsvga_state_s *s,
 
     vga_common_init(&s->vga, OBJECT(dev), &error_fatal);
     vga_init(&s->vga, OBJECT(dev), address_space, io, true);
-    vmstate_register(NULL, 0, &vmstate_vga_common, &s->vga);
+    vmstate_register_any(NULL, &vmstate_vga_common, &s->vga);
     s->new_depth = 32;
 }
 

hw/i2c/core.c

@@ -64,7 +64,7 @@ I2CBus *i2c_init_bus(DeviceState *parent, const char *name)
     bus = I2C_BUS(qbus_new(TYPE_I2C_BUS, parent, name));
     QLIST_INIT(&bus->current_devs);
     QSIMPLEQ_INIT(&bus->pending_masters);
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY, &vmstate_i2c_bus, bus);
+    vmstate_register_any(NULL, &vmstate_i2c_bus, bus);
     return bus;
 }
 

hw/ide/isa.c

@@ -73,7 +73,7 @@ static void isa_ide_realizefn(DeviceState *dev, Error **errp)
     ide_bus_init(&s->bus, sizeof(s->bus), dev, 0, 2);
     ide_init_ioport(&s->bus, isadev, s->iobase, s->iobase2);
     ide_bus_init_output_irq(&s->bus, isa_get_irq(isadev, s->irqnum));
-    vmstate_register(VMSTATE_IF(dev), 0, &vmstate_ide_isa, s);
+    vmstate_register_any(VMSTATE_IF(dev), &vmstate_ide_isa, s);
     ide_bus_register_restart_cb(&s->bus);
 }
 

hw/input/adb.c

@@ -247,7 +247,7 @@ static void adb_bus_realize(BusState *qbus, Error **errp)
     adb_bus->autopoll_timer = timer_new_ms(QEMU_CLOCK_VIRTUAL, adb_autopoll,
                                            adb_bus);
 
-    vmstate_register(NULL, -1, &vmstate_adb_bus, adb_bus);
+    vmstate_register_any(NULL, &vmstate_adb_bus, adb_bus);
 }
 
 static void adb_bus_unrealize(BusState *qbus)

hw/input/ads7846.c

@@ -158,7 +158,7 @@ static void ads7846_realize(SSIPeripheral *d, Error **errp)
 
     ads7846_int_update(s);
 
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY, &vmstate_ads7846, s);
+    vmstate_register_any(NULL, &vmstate_ads7846, s);
 }
 
 static void ads7846_class_init(ObjectClass *klass, void *data)

hw/input/stellaris_input.c

@@ -88,6 +88,5 @@ void stellaris_gamepad_init(int n, qemu_irq *irq, const int *keycode)
     }
     s->num_buttons = n;
     qemu_add_kbd_event_handler(stellaris_gamepad_put_key, s);
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY,
-                     &vmstate_stellaris_gamepad, s);
+    vmstate_register_any(NULL, &vmstate_stellaris_gamepad, s);
 }

hw/intc/xics.c

@@ -335,8 +335,22 @@ static void icp_realize(DeviceState *dev, Error **errp)
             return;
         }
     }
-
-    vmstate_register(NULL, icp->cs->cpu_index, &vmstate_icp_server, icp);
+    /*
+     * The way that pre_2_10_icp is handling is really, really hacky.
+     * We used to have here this call:
+     *
+     * vmstate_register(NULL, icp->cs->cpu_index, &vmstate_icp_server, icp);
+     *
+     * But we were doing:
+     *     pre_2_10_vmstate_register_dummy_icp()
+     *     this vmstate_register()
+     *     pre_2_10_vmstate_unregister_dummy_icp()
+     *
+     * So for a short amount of time we had to vmstate entries with
+     * the same name.  This fixes it.
+     */
+    vmstate_replace_hack_for_ppc(NULL, icp->cs->cpu_index,
+                                 &vmstate_icp_server, icp);
 }
 
 static void icp_unrealize(DeviceState *dev)

hw/ipmi/ipmi_bmc_extern.c

@@ -453,19 +453,6 @@ static void ipmi_bmc_extern_handle_reset(IPMIBmc *b)
     continue_send(ibe);
 }
 
-static void ipmi_bmc_extern_realize(DeviceState *dev, Error **errp)
-{
-    IPMIBmcExtern *ibe = IPMI_BMC_EXTERN(dev);
-
-    if (!qemu_chr_fe_backend_connected(&ibe->chr)) {
-        error_setg(errp, "IPMI external bmc requires chardev attribute");
-        return;
-    }
-
-    qemu_chr_fe_set_handlers(&ibe->chr, can_receive, receive,
-                             chr_event, NULL, ibe, NULL, true);
-}
-
 static int ipmi_bmc_extern_post_migrate(void *opaque, int version_id)
 {
     IPMIBmcExtern *ibe = opaque;
@@ -499,12 +486,26 @@ static const VMStateDescription vmstate_ipmi_bmc_extern = {
     }
 };
 
+static void ipmi_bmc_extern_realize(DeviceState *dev, Error **errp)
+{
+    IPMIBmcExtern *ibe = IPMI_BMC_EXTERN(dev);
+
+    if (!qemu_chr_fe_backend_connected(&ibe->chr)) {
+        error_setg(errp, "IPMI external bmc requires chardev attribute");
+        return;
+    }
+
+    qemu_chr_fe_set_handlers(&ibe->chr, can_receive, receive,
+                             chr_event, NULL, ibe, NULL, true);
+
+    vmstate_register(NULL, 0, &vmstate_ipmi_bmc_extern, ibe);
+}
+
 static void ipmi_bmc_extern_init(Object *obj)
 {
     IPMIBmcExtern *ibe = IPMI_BMC_EXTERN(obj);
 
     ibe->extern_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, extern_timeout, ibe);
-    vmstate_register(NULL, 0, &vmstate_ipmi_bmc_extern, ibe);
 }
 
 static void ipmi_bmc_extern_finalize(Object *obj)

hw/ipmi/isa_ipmi_bt.c

@@ -68,6 +68,21 @@ static void isa_ipmi_bt_lower_irq(IPMIBT *ib)
     qemu_irq_lower(iib->irq);
 }
 
+static const VMStateDescription vmstate_ISAIPMIBTDevice = {
+    .name = TYPE_IPMI_INTERFACE_PREFIX "isa-bt",
+    .version_id = 2,
+    .minimum_version_id = 2,
+    /*
+     * Version 1 had messed up the array transfer, it's not even usable
+     * because it used VMSTATE_VBUFFER_UINT32, but it did not transfer
+     * the buffer length, so random things would happen.
+     */
+    .fields      = (VMStateField[]) {
+        VMSTATE_STRUCT(bt, ISAIPMIBTDevice, 1, vmstate_IPMIBT, IPMIBT),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
 static void isa_ipmi_bt_realize(DeviceState *dev, Error **errp)
 {
     Error *err = NULL;
@@ -102,30 +117,15 @@ static void isa_ipmi_bt_realize(DeviceState *dev, Error **errp)
     qdev_set_legacy_instance_id(dev, iib->bt.io_base, iib->bt.io_length);
 
     isa_register_ioport(isadev, &iib->bt.io, iib->bt.io_base);
-}
 
-static const VMStateDescription vmstate_ISAIPMIBTDevice = {
-    .name = TYPE_IPMI_INTERFACE_PREFIX "isa-bt",
-    .version_id = 2,
-    .minimum_version_id = 2,
-    /*
-     * Version 1 had messed up the array transfer, it's not even usable
-     * because it used VMSTATE_VBUFFER_UINT32, but it did not transfer
-     * the buffer length, so random things would happen.
-     */
-    .fields      = (VMStateField[]) {
-        VMSTATE_STRUCT(bt, ISAIPMIBTDevice, 1, vmstate_IPMIBT, IPMIBT),
-        VMSTATE_END_OF_LIST()
-    }
-};
+    vmstate_register(NULL, 0, &vmstate_ISAIPMIBTDevice, dev);
+}
 
 static void isa_ipmi_bt_init(Object *obj)
 {
     ISAIPMIBTDevice *iib = ISA_IPMI_BT(obj);
 
     ipmi_bmc_find_and_link(obj, (Object **) &iib->bt.bmc);
-
-    vmstate_register(NULL, 0, &vmstate_ISAIPMIBTDevice, iib);
 }
 
 static void *isa_ipmi_bt_get_backend_data(IPMIInterface *ii)

hw/ipmi/isa_ipmi_kcs.c

@@ -67,6 +67,24 @@ static void isa_ipmi_kcs_lower_irq(IPMIKCS *ik)
     qemu_irq_lower(iik->irq);
 }
 
+static bool vmstate_kcs_before_version2(void *opaque, int version)
+{
+    return version <= 1;
+}
+
+static const VMStateDescription vmstate_ISAIPMIKCSDevice = {
+    .name = TYPE_IPMI_INTERFACE,
+    .version_id = 2,
+    .minimum_version_id = 1,
+    .fields      = (VMStateField[]) {
+        VMSTATE_VSTRUCT_TEST(kcs, ISAIPMIKCSDevice, vmstate_kcs_before_version2,
+                             0, vmstate_IPMIKCS, IPMIKCS, 1),
+        VMSTATE_VSTRUCT_V(kcs, ISAIPMIKCSDevice, 2, vmstate_IPMIKCS,
+                          IPMIKCS, 2),
+        VMSTATE_END_OF_LIST()
+    }
+};
+
 static void ipmi_isa_realize(DeviceState *dev, Error **errp)
 {
     Error *err = NULL;
@@ -101,38 +119,20 @@ static void ipmi_isa_realize(DeviceState *dev, Error **errp)
     qdev_set_legacy_instance_id(dev, iik->kcs.io_base, iik->kcs.io_length);
 
     isa_register_ioport(isadev, &iik->kcs.io, iik->kcs.io_base);
-}
 
-static bool vmstate_kcs_before_version2(void *opaque, int version)
-{
-    return version <= 1;
+    /*
+     * Version 1 had an incorrect name, it clashed with the BT
+     * IPMI device, so receive it, but transmit a different
+     * version.
+     */
+    vmstate_register(NULL, 0, &vmstate_ISAIPMIKCSDevice, iik);
 }
 
-static const VMStateDescription vmstate_ISAIPMIKCSDevice = {
-    .name = TYPE_IPMI_INTERFACE,
-    .version_id = 2,
-    .minimum_version_id = 1,
-    .fields      = (VMStateField[]) {
-        VMSTATE_VSTRUCT_TEST(kcs, ISAIPMIKCSDevice, vmstate_kcs_before_version2,
-                             0, vmstate_IPMIKCS, IPMIKCS, 1),
-        VMSTATE_VSTRUCT_V(kcs, ISAIPMIKCSDevice, 2, vmstate_IPMIKCS,
-                          IPMIKCS, 2),
-        VMSTATE_END_OF_LIST()
-    }
-};
-
 static void isa_ipmi_kcs_init(Object *obj)
 {
     ISAIPMIKCSDevice *iik = ISA_IPMI_KCS(obj);
 
     ipmi_bmc_find_and_link(obj, (Object **) &iik->kcs.bmc);
-
-    /*
-     * Version 1 had an incorrect name, it clashed with the BT
-     * IPMI device, so receive it, but transmit a different
-     * version.
-     */
-    vmstate_register(NULL, 0, &vmstate_ISAIPMIKCSDevice, iik);
 }
 
 static void *isa_ipmi_kcs_get_backend_data(IPMIInterface *ii)

hw/net/eepro100.c

@@ -1883,8 +1883,7 @@ static void e100_nic_realize(PCIDevice *pci_dev, Error **errp)
 
     s->vmstate = g_memdup(&vmstate_eepro100, sizeof(vmstate_eepro100));
     s->vmstate->name = qemu_get_queue(s->nic)->model;
-    vmstate_register(VMSTATE_IF(&pci_dev->qdev), VMSTATE_INSTANCE_ID_ANY,
-                     s->vmstate, s);
+    vmstate_register_any(VMSTATE_IF(&pci_dev->qdev), s->vmstate, s);
 }
 
 static void eepro100_instance_init(Object *obj)

hw/nvram/eeprom93xx.c

@@ -321,7 +321,7 @@ eeprom_t *eeprom93xx_new(DeviceState *dev, uint16_t nwords)
     /* Output DO is tristate, read results in 1. */
     eeprom->eedo = 1;
     logout("eeprom = 0x%p, nwords = %u\n", eeprom, nwords);
-    vmstate_register(VMSTATE_IF(dev), 0, &vmstate_eeprom, eeprom);
+    vmstate_register_any(VMSTATE_IF(dev), &vmstate_eeprom, eeprom);
     return eeprom;
 }
 

hw/pci/pci.c

@@ -147,7 +147,7 @@ static void pci_bus_realize(BusState *qbus, Error **errp)
     bus->machine_done.notify = pcibus_machine_done;
     qemu_add_machine_init_done_notifier(&bus->machine_done);
 
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY, &vmstate_pcibus, bus);
+    vmstate_register_any(NULL, &vmstate_pcibus, bus);
 }
 
 static void pcie_bus_realize(BusState *qbus, Error **errp)

hw/ppc/spapr.c

@@ -143,6 +143,11 @@ static bool pre_2_10_vmstate_dummy_icp_needed(void *opaque)
 }
 
 static const VMStateDescription pre_2_10_vmstate_dummy_icp = {
+    /*
+     * Hack ahead.  We can't have two devices with the same name and
+     * instance id.  So I rename this to pass make check.
+     * Real help from people who knows the hardware is needed.
+     */
     .name = "icp/server",
     .version_id = 1,
     .minimum_version_id = 1,
@@ -155,16 +160,32 @@ static const VMStateDescription pre_2_10_vmstate_dummy_icp = {
     },
 };
 
+/*
+ * See comment in hw/intc/xics.c:icp_realize()
+ *
+ * You have to remove vmstate_replace_hack_for_ppc() when you remove
+ * the machine types that need the following function.
+ */
 static void pre_2_10_vmstate_register_dummy_icp(int i)
 {
     vmstate_register(NULL, i, &pre_2_10_vmstate_dummy_icp,
                      (void *)(uintptr_t) i);
 }
 
+/*
+ * See comment in hw/intc/xics.c:icp_realize()
+ *
+ * You have to remove vmstate_replace_hack_for_ppc() when you remove
+ * the machine types that need the following function.
+ */
 static void pre_2_10_vmstate_unregister_dummy_icp(int i)
 {
-    vmstate_unregister(NULL, &pre_2_10_vmstate_dummy_icp,
-                       (void *)(uintptr_t) i);
+    /*
+     * This used to be:
+     *
+     *    vmstate_unregister(NULL, &pre_2_10_vmstate_dummy_icp,
+     *                      (void *)(uintptr_t) i);
+     */
 }
 
 int spapr_max_server_number(SpaprMachineState *spapr)

hw/ppc/spapr_nvdimm.c

@@ -876,8 +876,7 @@ static void spapr_nvdimm_realize(NVDIMMDevice *dimm, Error **errp)
         s_nvdimm->hcall_flush_required = true;
     }
 
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY,
-                     &vmstate_spapr_nvdimm_states, dimm);
+    vmstate_register_any(NULL, &vmstate_spapr_nvdimm_states, dimm);
 }
 
 static void spapr_nvdimm_unrealize(NVDIMMDevice *dimm)

hw/s390x/s390-skeys.c

@@ -12,6 +12,7 @@
 #include "qemu/osdep.h"
 #include "qemu/units.h"
 #include "hw/boards.h"
+#include "hw/qdev-properties.h"
 #include "hw/s390x/storage-keys.h"
 #include "qapi/error.h"
 #include "qapi/qapi-commands-misc-target.h"
@@ -432,58 +433,39 @@ static int s390_storage_keys_load(QEMUFile *f, void *opaque, int version_id)
     return ret;
 }
 
-static inline bool s390_skeys_get_migration_enabled(Object *obj, Error **errp)
-{
-    S390SKeysState *ss = S390_SKEYS(obj);
-
-    return ss->migration_enabled;
-}
-
 static SaveVMHandlers savevm_s390_storage_keys = {
     .save_state = s390_storage_keys_save,
     .load_state = s390_storage_keys_load,
 };
 
-static inline void s390_skeys_set_migration_enabled(Object *obj, bool value,
-                                            Error **errp)
+static void s390_skeys_realize(DeviceState *dev, Error **errp)
 {
-    S390SKeysState *ss = S390_SKEYS(obj);
-
-    /* Prevent double registration of savevm handler */
-    if (ss->migration_enabled == value) {
-        return;
-    }
-
-    ss->migration_enabled = value;
+    S390SKeysState *ss = S390_SKEYS(dev);
 
     if (ss->migration_enabled) {
         register_savevm_live(TYPE_S390_SKEYS, 0, 1,
                              &savevm_s390_storage_keys, ss);
-    } else {
-        unregister_savevm(VMSTATE_IF(ss), TYPE_S390_SKEYS, ss);
     }
 }
 
-static void s390_skeys_instance_init(Object *obj)
-{
-    object_property_add_bool(obj, "migration-enabled",
-                             s390_skeys_get_migration_enabled,
-                             s390_skeys_set_migration_enabled);
-    object_property_set_bool(obj, "migration-enabled", true, NULL);
-}
+static Property s390_skeys_props[] = {
+    DEFINE_PROP_BOOL("migration-enabled", S390SKeysState, migration_enabled, true),
+    DEFINE_PROP_END_OF_LIST(),
+};
 
 static void s390_skeys_class_init(ObjectClass *oc, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(oc);
 
     dc->hotpluggable = false;
+    dc->realize = s390_skeys_realize;
+    device_class_set_props(dc, s390_skeys_props);
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
 }
 
 static const TypeInfo s390_skeys_info = {
     .name          = TYPE_S390_SKEYS,
     .parent        = TYPE_DEVICE,
-    .instance_init = s390_skeys_instance_init,
     .instance_size = sizeof(S390SKeysState),
     .class_init    = s390_skeys_class_init,
     .class_size    = sizeof(S390SKeysClass),

hw/s390x/s390-stattrib.c

@@ -13,6 +13,7 @@
 #include "qemu/units.h"
 #include "migration/qemu-file.h"
 #include "migration/register.h"
+#include "hw/qdev-properties.h"
 #include "hw/s390x/storage-attributes.h"
 #include "qemu/error-report.h"
 #include "exec/ram_addr.h"
@@ -330,6 +331,17 @@ static const TypeInfo qemu_s390_stattrib_info = {
 
 /* Generic abstract object: */
 
+static SaveVMHandlers savevm_s390_stattrib_handlers = {
+    .save_setup = cmma_save_setup,
+    .save_live_iterate = cmma_save_iterate,
+    .save_live_complete_precopy = cmma_save_complete,
+    .state_pending_exact = cmma_state_pending,
+    .state_pending_estimate = cmma_state_pending,
+    .save_cleanup = cmma_save_cleanup,
+    .load_state = cmma_load,
+    .is_active = cmma_active,
+};
+
 static void s390_stattrib_realize(DeviceState *dev, Error **errp)
 {
     bool ambiguous = false;
@@ -337,9 +349,18 @@ static void s390_stattrib_realize(DeviceState *dev, Error **errp)
     object_resolve_path_type("", TYPE_S390_STATTRIB, &ambiguous);
     if (ambiguous) {
         error_setg(errp, "storage_attributes device already exists");
+        return;
     }
+
+    register_savevm_live(TYPE_S390_STATTRIB, 0, 0,
+                         &savevm_s390_stattrib_handlers, dev);
 }
 
+static Property s390_stattrib_props[] = {
+    DEFINE_PROP_BOOL("migration-enabled", S390StAttribState, migration_enabled, true),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
 static void s390_stattrib_class_init(ObjectClass *oc, void *data)
 {
     DeviceClass *dc = DEVICE_CLASS(oc);
@@ -347,46 +368,13 @@ static void s390_stattrib_class_init(ObjectClass *oc, void *data)
     dc->hotpluggable = false;
     set_bit(DEVICE_CATEGORY_MISC, dc->categories);
     dc->realize = s390_stattrib_realize;
+    device_class_set_props(dc, s390_stattrib_props);
 }
 
-static inline bool s390_stattrib_get_migration_enabled(Object *obj,
-                                                       Error **errp)
-{
-    S390StAttribState *s = S390_STATTRIB(obj);
-
-    return s->migration_enabled;
-}
-
-static inline void s390_stattrib_set_migration_enabled(Object *obj, bool value,
-                                            Error **errp)
-{
-    S390StAttribState *s = S390_STATTRIB(obj);
-
-    s->migration_enabled = value;
-}
-
-static SaveVMHandlers savevm_s390_stattrib_handlers = {
-    .save_setup = cmma_save_setup,
-    .save_live_iterate = cmma_save_iterate,
-    .save_live_complete_precopy = cmma_save_complete,
-    .state_pending_exact = cmma_state_pending,
-    .state_pending_estimate = cmma_state_pending,
-    .save_cleanup = cmma_save_cleanup,
-    .load_state = cmma_load,
-    .is_active = cmma_active,
-};
-
 static void s390_stattrib_instance_init(Object *obj)
 {
     S390StAttribState *sas = S390_STATTRIB(obj);
 
-    register_savevm_live(TYPE_S390_STATTRIB, 0, 0,
-                         &savevm_s390_stattrib_handlers, sas);
-
-    object_property_add_bool(obj, "migration-enabled",
-                             s390_stattrib_get_migration_enabled,
-                             s390_stattrib_set_migration_enabled);
-    object_property_set_bool(obj, "migration-enabled", true, NULL);
     sas->migration_cur_gfn = 0;
 }
 

hw/scsi/vhost-scsi.c

@@ -210,7 +210,7 @@ static void vhost_scsi_realize(DeviceState *dev, Error **errp)
                 "When external environment supports it (Orchestrator migrates "
                 "target SCSI device state or use shared storage over network), "
                 "set 'migratable' property to true to enable migration.");
-        if (migrate_add_blocker(&vsc->migration_blocker, errp) < 0) {
+        if (migrate_add_blocker_normal(&vsc->migration_blocker, errp) < 0) {
             goto free_virtio;
         }
     }

hw/timer/arm_timer.c

@@ -181,7 +181,7 @@ static arm_timer_state *arm_timer_init(uint32_t freq)
     s->control = TIMER_CTRL_IE;
 
     s->timer = ptimer_init(arm_timer_tick, s, PTIMER_POLICY_LEGACY);
-    vmstate_register(NULL, VMSTATE_INSTANCE_ID_ANY, &vmstate_arm_timer, s);
+    vmstate_register_any(NULL, &vmstate_arm_timer, s);
     return s;
 }
 

hw/virtio/vhost.c

@@ -1527,7 +1527,7 @@ int vhost_dev_init(struct vhost_dev *hdev, void *opaque,
     }
 
     if (hdev->migration_blocker != NULL) {
-        r = migrate_add_blocker(&hdev->migration_blocker, errp);
+        r = migrate_add_blocker_normal(&hdev->migration_blocker, errp);
         if (r < 0) {
             goto fail_busyloop;
         }

hw/virtio/virtio-mem.c

@@ -1119,8 +1119,8 @@ static void virtio_mem_device_realize(DeviceState *dev, Error **errp)
     host_memory_backend_set_mapped(vmem->memdev, true);
     vmstate_register_ram(&vmem->memdev->mr, DEVICE(vmem));
     if (vmem->early_migration) {
-        vmstate_register(VMSTATE_IF(vmem), VMSTATE_INSTANCE_ID_ANY,
-                         &vmstate_virtio_mem_device_early, vmem);
+        vmstate_register_any(VMSTATE_IF(vmem),
+                             &vmstate_virtio_mem_device_early, vmem);
     }
     qemu_register_reset(virtio_mem_system_reset, vmem);
 

include/hw/qdev-properties-system.h

@@ -7,6 +7,7 @@ extern const PropertyInfo qdev_prop_chr;
 extern const PropertyInfo qdev_prop_macaddr;
 extern const PropertyInfo qdev_prop_reserved_region;
 extern const PropertyInfo qdev_prop_multifd_compression;
+extern const PropertyInfo qdev_prop_mig_mode;
 extern const PropertyInfo qdev_prop_losttickpolicy;
 extern const PropertyInfo qdev_prop_blockdev_on_error;
 extern const PropertyInfo qdev_prop_bios_chs_trans;
@@ -42,6 +43,9 @@ extern const PropertyInfo qdev_prop_cpus390entitlement;
 #define DEFINE_PROP_MULTIFD_COMPRESSION(_n, _s, _f, _d) \
     DEFINE_PROP_SIGNED(_n, _s, _f, _d, qdev_prop_multifd_compression, \
                        MultiFDCompression)
+#define DEFINE_PROP_MIG_MODE(_n, _s, _f, _d) \
+    DEFINE_PROP_SIGNED(_n, _s, _f, _d, qdev_prop_mig_mode, \
+                       MigMode)
 #define DEFINE_PROP_LOSTTICKPOLICY(_n, _s, _f, _d) \
     DEFINE_PROP_SIGNED(_n, _s, _f, _d, qdev_prop_losttickpolicy, \
                         LostTickPolicy)

include/migration/blocker.h

@@ -14,8 +14,12 @@
 #ifndef MIGRATION_BLOCKER_H
 #define MIGRATION_BLOCKER_H
 
+#include "qapi/qapi-types-migration.h"
+
+#define MIG_MODE_ALL MIG_MODE__MAX
+
 /**
- * @migrate_add_blocker - prevent migration from proceeding
+ * @migrate_add_blocker - prevent all modes of migration from proceeding
  *
  * @reasonp - address of an error to be returned whenever migration is attempted
  *
@@ -30,8 +34,8 @@
 int migrate_add_blocker(Error **reasonp, Error **errp);
 
 /**
- * @migrate_add_blocker_internal - prevent migration from proceeding without
- *                                 only-migrate implications
+ * @migrate_add_blocker_internal - prevent all modes of migration from
+ *                                 proceeding, but ignore -only-migratable
  *
  * @reasonp - address of an error to be returned whenever migration is attempted
  *
@@ -50,7 +54,7 @@ int migrate_add_blocker(Error **reasonp, Error **errp);
 int migrate_add_blocker_internal(Error **reasonp, Error **errp);
 
 /**
- * @migrate_del_blocker - remove a blocking error from migration and free it.
+ * @migrate_del_blocker - remove a migration blocker from all modes and free it.
  *
  * @reasonp - address of the error blocking migration
  *
@@ -58,4 +62,36 @@ int migrate_add_blocker_internal(Error **reasonp, Error **errp);
  */
 void migrate_del_blocker(Error **reasonp);
 
+/**
+ * @migrate_add_blocker_normal - prevent normal migration mode from proceeding
+ *
+ * @reasonp - address of an error to be returned whenever migration is attempted
+ *
+ * @errp - [out] The reason (if any) we cannot block migration right now.
+ *
+ * @returns - 0 on success, -EBUSY/-EACCES on failure, with errp set.
+ *
+ * *@reasonp is freed and set to NULL if failure is returned.
+ * On success, the caller must not free @reasonp, except by
+ *   calling migrate_del_blocker.
+ */
+int migrate_add_blocker_normal(Error **reasonp, Error **errp);
+
+/**
+ * @migrate_add_blocker_modes - prevent some modes of migration from proceeding
+ *
+ * @reasonp - address of an error to be returned whenever migration is attempted
+ *
+ * @errp - [out] The reason (if any) we cannot block migration right now.
+ *
+ * @mode - one or more migration modes to be blocked.  The list is terminated
+ *         by -1 or MIG_MODE_ALL.  For the latter, all modes are blocked.
+ *
+ * @returns - 0 on success, -EBUSY/-EACCES on failure, with errp set.
+ *
+ * *@reasonp is freed and set to NULL if failure is returned.
+ * On success, the caller must not free *@reasonp before the blocker is removed.
+ */
+int migrate_add_blocker_modes(Error **reasonp, Error **errp, MigMode mode, ...);
+
 #endif

include/migration/misc.h

@@ -15,6 +15,7 @@
 #define MIGRATION_MISC_H
 
 #include "qemu/notify.h"
+#include "qapi/qapi-types-migration.h"
 #include "qapi/qapi-types-net.h"
 
 /* migration/ram.c */

include/migration/vmstate.h

@@ -1230,6 +1230,34 @@ static inline int vmstate_register(VMStateIf *obj, int instance_id,
                                           opaque, -1, 0, NULL);
 }
 
+/**
+ * vmstate_replace_hack_for_ppc() - ppc used to abuse vmstate_register
+ *
+ * Don't even think about using this function in new code.
+ *
+ * Returns: 0 on success, -1 on failure
+ */
+int vmstate_replace_hack_for_ppc(VMStateIf *obj, int instance_id,
+                                 const VMStateDescription *vmsd,
+                                 void *opaque);
+
+/**
+ * vmstate_register_any() - legacy function to register state
+ * serialisation description and let the function choose the id
+ *
+ * New code shouldn't be using this function as QOM-ified devices have
+ * dc->vmsd to store the serialisation description.
+ *
+ * Returns: 0 on success, -1 on failure
+ */
+static inline int vmstate_register_any(VMStateIf *obj,
+                                       const VMStateDescription *vmsd,
+                                       void *opaque)
+{
+    return vmstate_register_with_alias_id(obj, VMSTATE_INSTANCE_ID_ANY, vmsd,
+                                          opaque, -1, 0, NULL);
+}
+
 void vmstate_unregister(VMStateIf *obj, const VMStateDescription *vmsd,
                         void *opaque);
 

migration/exec.c

@@ -27,7 +27,6 @@
 #include "qemu/cutils.h"
 
 #ifdef WIN32
-const char *exec_get_cmd_path(void);
 const char *exec_get_cmd_path(void)
 {
     g_autofree char *detected_path = g_new(char, MAX_PATH);
@@ -40,20 +39,51 @@ const char *exec_get_cmd_path(void)
 }
 #endif
 
-void exec_start_outgoing_migration(MigrationState *s, const char *command, Error **errp)
+/* provides the length of strList */
+static int
+str_list_length(strList *list)
+{
+    int len = 0;
+    strList *elem;
+
+    for (elem = list; elem != NULL; elem = elem->next) {
+        len++;
+    }
+
+    return len;
+}
+
+static void
+init_exec_array(strList *command, char **argv, Error **errp)
+{
+    int i = 0;
+    strList *lst;
+
+    for (lst = command; lst; lst = lst->next) {
+        argv[i++] = lst->value;
+    }
+
+    argv[i] = NULL;
+    return;
+}
+
+void exec_start_outgoing_migration(MigrationState *s, strList *command,
+                                   Error **errp)
 {
     QIOChannel *ioc;
 
-#ifdef WIN32
-    const char *argv[] = { exec_get_cmd_path(), "/c", command, NULL };
-#else
-    const char *argv[] = { "/bin/sh", "-c", command, NULL };
-#endif
+    int length = str_list_length(command);
+    g_auto(GStrv) argv = (char **) g_new0(const char *, length + 1);
 
-    trace_migration_exec_outgoing(command);
-    ioc = QIO_CHANNEL(qio_channel_command_new_spawn(argv,
-                                                    O_RDWR,
-                                                    errp));
+    init_exec_array(command, argv, errp);
+    g_autofree char *new_command = g_strjoinv(" ", (char **)argv);
+
+    trace_migration_exec_outgoing(new_command);
+    ioc = QIO_CHANNEL(
+        qio_channel_command_new_spawn(
+                            (const char * const *) g_steal_pointer(&argv),
+                            O_RDWR,
+                            errp));
     if (!ioc) {
         return;
     }
@@ -72,20 +102,22 @@ static gboolean exec_accept_incoming_migration(QIOChannel *ioc,
     return G_SOURCE_REMOVE;
 }
 
-void exec_start_incoming_migration(const char *command, Error **errp)
+void exec_start_incoming_migration(strList *command, Error **errp)
 {
     QIOChannel *ioc;
 
-#ifdef WIN32
-    const char *argv[] = { exec_get_cmd_path(), "/c", command, NULL };
-#else
-    const char *argv[] = { "/bin/sh", "-c", command, NULL };
-#endif
+    int length = str_list_length(command);
+    g_auto(GStrv) argv = (char **) g_new0(const char *, length + 1);
+
+    init_exec_array(command, argv, errp);
+    g_autofree char *new_command = g_strjoinv(" ", (char **)argv);
 
-    trace_migration_exec_incoming(command);
-    ioc = QIO_CHANNEL(qio_channel_command_new_spawn(argv,
-                                                    O_RDWR,
-                                                    errp));
+    trace_migration_exec_incoming(new_command);
+    ioc = QIO_CHANNEL(
+        qio_channel_command_new_spawn(
+                            (const char * const *) g_steal_pointer(&argv),
+                            O_RDWR,
+                            errp));
     if (!ioc) {
         return;
     }

migration/exec.h

@@ -19,8 +19,12 @@
 
 #ifndef QEMU_MIGRATION_EXEC_H
 #define QEMU_MIGRATION_EXEC_H
-void exec_start_incoming_migration(const char *host_port, Error **errp);
 
-void exec_start_outgoing_migration(MigrationState *s, const char *host_port,
+#ifdef WIN32
+const char *exec_get_cmd_path(void);
+#endif
+void exec_start_incoming_migration(strList *host_port, Error **errp);
+
+void exec_start_outgoing_migration(MigrationState *s, strList *host_port,
                                    Error **errp);
 #endif

migration/file.c

@@ -19,7 +19,7 @@
 
 /* Remove the offset option from @filespec and return it in @offsetp. */
 
-static int file_parse_offset(char *filespec, uint64_t *offsetp, Error **errp)
+int file_parse_offset(char *filespec, uint64_t *offsetp, Error **errp)
 {
     char *option = strstr(filespec, OFFSET_OPTION);
     int ret;
@@ -36,20 +36,16 @@ static int file_parse_offset(char *filespec, uint64_t *offsetp, Error **errp)
     return 0;
 }
 
-void file_start_outgoing_migration(MigrationState *s, const char *filespec,
-                                   Error **errp)
+void file_start_outgoing_migration(MigrationState *s,
+                                   FileMigrationArgs *file_args, Error **errp)
 {
-    g_autofree char *filename = g_strdup(filespec);
     g_autoptr(QIOChannelFile) fioc = NULL;
-    uint64_t offset = 0;
+    g_autofree char *filename = g_strdup(file_args->filename);
+    uint64_t offset = file_args->offset;
     QIOChannel *ioc;
 
     trace_migration_file_outgoing(filename);
 
-    if (file_parse_offset(filename, &offset, errp)) {
-        return;
-    }
-
     fioc = qio_channel_file_new_path(filename, O_CREAT | O_WRONLY | O_TRUNC,
                                      0600, errp);
     if (!fioc) {
@@ -73,19 +69,15 @@ static gboolean file_accept_incoming_migration(QIOChannel *ioc,
     return G_SOURCE_REMOVE;
 }
 
-void file_start_incoming_migration(const char *filespec, Error **errp)
+void file_start_incoming_migration(FileMigrationArgs *file_args, Error **errp)
 {
-    g_autofree char *filename = g_strdup(filespec);
+    g_autofree char *filename = g_strdup(file_args->filename);
     QIOChannelFile *fioc = NULL;
-    uint64_t offset = 0;
+    uint64_t offset = file_args->offset;
     QIOChannel *ioc;
 
     trace_migration_file_incoming(filename);
 
-    if (file_parse_offset(filename, &offset, errp)) {
-        return;
-    }
-
     fioc = qio_channel_file_new_path(filename, O_RDONLY, 0, errp);
     if (!fioc) {
         return;

migration/file.h

@@ -7,8 +7,12 @@
 
 #ifndef QEMU_MIGRATION_FILE_H
 #define QEMU_MIGRATION_FILE_H
-void file_start_incoming_migration(const char *filename, Error **errp);
 
-void file_start_outgoing_migration(MigrationState *s, const char *filename,
-                                   Error **errp);
+#include "qapi/qapi-types-migration.h"
+
+void file_start_incoming_migration(FileMigrationArgs *file_args, Error **errp);
+
+void file_start_outgoing_migration(MigrationState *s,
+                                   FileMigrationArgs *file_args, Error **errp);
+int file_parse_offset(char *filespec, uint64_t *offsetp, Error **errp);
 #endif

migration/migration-hmp-cmds.c

@@ -387,6 +387,11 @@ void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict)
         monitor_printf(mon, "%s: %" PRIu64 " MB/s\n",
             MigrationParameter_str(MIGRATION_PARAMETER_VCPU_DIRTY_LIMIT),
             params->vcpu_dirty_limit);
+
+        assert(params->has_mode);
+        monitor_printf(mon, "%s: %s\n",
+            MigrationParameter_str(MIGRATION_PARAMETER_MODE),
+            qapi_enum_lookup(&MigMode_lookup, params->mode));
     }
 
     qapi_free_MigrationParameters(params);
@@ -446,9 +451,18 @@ void hmp_migrate_incoming(Monitor *mon, const QDict *qdict)
 {
     Error *err = NULL;
     const char *uri = qdict_get_str(qdict, "uri");
+    MigrationChannelList *caps = NULL;
+    g_autoptr(MigrationChannel) channel = NULL;
+
+    if (!migrate_uri_parse(uri, &channel, &err)) {
+        goto end;
+    }
+    QAPI_LIST_PREPEND(caps, g_steal_pointer(&channel));
 
-    qmp_migrate_incoming(uri, &err);
+    qmp_migrate_incoming(NULL, true, caps, &err);
+    qapi_free_MigrationChannelList(caps);
 
+end:
     hmp_handle_error(mon, err);
 }
 
@@ -661,6 +675,10 @@ void hmp_migrate_set_parameter(Monitor *mon, const QDict *qdict)
         p->has_vcpu_dirty_limit = true;
         visit_type_size(v, param, &p->vcpu_dirty_limit, &err);
         break;
+    case MIGRATION_PARAMETER_MODE:
+        p->has_mode = true;
+        visit_type_MigMode(v, param, &p->mode, &err);
+        break;
     default:
         assert(0);
     }
@@ -744,6 +762,8 @@ void hmp_migrate(Monitor *mon, const QDict *qdict)
     bool resume = qdict_get_try_bool(qdict, "resume", false);
     const char *uri = qdict_get_str(qdict, "uri");
     Error *err = NULL;
+    MigrationChannelList *caps = NULL;
+    g_autoptr(MigrationChannel) channel = NULL;
 
     if (inc) {
         warn_report("option '-i' is deprecated;"
@@ -755,12 +775,20 @@ void hmp_migrate(Monitor *mon, const QDict *qdict)
                     " use blockdev-mirror with NBD instead");
     }
 
-    qmp_migrate(uri, !!blk, blk, !!inc, inc,
-                false, false, true, resume, &err);
+    if (!migrate_uri_parse(uri, &channel, &err)) {
+        hmp_handle_error(mon, err);
+        return;
+    }
+    QAPI_LIST_PREPEND(caps, g_steal_pointer(&channel));
+
+    qmp_migrate(NULL, true, caps, !!blk, blk, !!inc, inc,
+                 false, false, true, resume, &err);
     if (hmp_handle_error(mon, err)) {
         return;
     }
 
+    qapi_free_MigrationChannelList(caps);
+
     if (!detach) {
         HMPMigrationStatus *status;
 

migration/migration.c

@@ -66,6 +66,7 @@
 #include "sysemu/qtest.h"
 #include "options.h"
 #include "sysemu/dirtylimit.h"
+#include "qemu/sockets.h"
 
 static NotifierList migration_state_notifiers =
     NOTIFIER_LIST_INITIALIZER(migration_state_notifiers);
@@ -92,31 +93,55 @@ enum mig_rp_message_type {
 static MigrationState *current_migration;
 static MigrationIncomingState *current_incoming;
 
-static GSList *migration_blockers;
+static GSList *migration_blockers[MIG_MODE__MAX];
 
 static bool migration_object_check(MigrationState *ms, Error **errp);
 static int migration_maybe_pause(MigrationState *s,
                                  int *current_active_state,
                                  int new_state);
 static void migrate_fd_cancel(MigrationState *s);
-static int close_return_path_on_source(MigrationState *s);
+static bool close_return_path_on_source(MigrationState *s);
+
+static void migration_downtime_start(MigrationState *s)
+{
+    trace_vmstate_downtime_checkpoint("src-downtime-start");
+    s->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+}
+
+static void migration_downtime_end(MigrationState *s)
+{
+    int64_t now = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+
+    /*
+     * If downtime already set, should mean that postcopy already set it,
+     * then that should be the real downtime already.
+     */
+    if (!s->downtime) {
+        s->downtime = now - s->downtime_start;
+    }
+
+    trace_vmstate_downtime_checkpoint("src-downtime-end");
+}
 
 static bool migration_needs_multiple_sockets(void)
 {
     return migrate_multifd() || migrate_postcopy_preempt();
 }
 
-static bool uri_supports_multi_channels(const char *uri)
+static bool transport_supports_multi_channels(SocketAddress *saddr)
 {
-    return strstart(uri, "tcp:", NULL) || strstart(uri, "unix:", NULL) ||
-           strstart(uri, "vsock:", NULL);
+    return saddr->type == SOCKET_ADDRESS_TYPE_INET ||
+           saddr->type == SOCKET_ADDRESS_TYPE_UNIX ||
+           saddr->type == SOCKET_ADDRESS_TYPE_VSOCK;
 }
 
 static bool
-migration_channels_and_uri_compatible(const char *uri, Error **errp)
+migration_channels_and_transport_compatible(MigrationAddress *addr,
+                                            Error **errp)
 {
     if (migration_needs_multiple_sockets() &&
-        !uri_supports_multi_channels(uri)) {
+        (addr->transport == MIGRATION_ADDRESS_TYPE_SOCKET) &&
+        !transport_supports_multi_channels(&addr->u.socket)) {
         error_setg(errp, "Migration requires multi-channel URIs (e.g. tcp)");
         return false;
     }
@@ -131,6 +156,15 @@ static gint page_request_addr_cmp(gconstpointer ap, gconstpointer bp)
     return (a > b) - (a < b);
 }
 
+int migration_stop_vm(RunState state)
+{
+    int ret = vm_stop_force_state(state);
+
+    trace_vmstate_downtime_checkpoint("src-vm-stopped");
+
+    return ret;
+}
+
 void migration_object_init(void)
 {
     /* This can only be called once. */
@@ -423,25 +457,114 @@ void migrate_add_address(SocketAddress *address)
                       QAPI_CLONE(SocketAddress, address));
 }
 
-static void qemu_start_incoming_migration(const char *uri, Error **errp)
+bool migrate_uri_parse(const char *uri, MigrationChannel **channel,
+                       Error **errp)
 {
-    const char *p = NULL;
+    g_autoptr(MigrationChannel) val = g_new0(MigrationChannel, 1);
+    g_autoptr(MigrationAddress) addr = g_new0(MigrationAddress, 1);
+    SocketAddress *saddr = NULL;
+    InetSocketAddress *isock = &addr->u.rdma;
+    strList **tail = &addr->u.exec.args;
+
+    if (strstart(uri, "exec:", NULL)) {
+        addr->transport = MIGRATION_ADDRESS_TYPE_EXEC;
+#ifdef WIN32
+        QAPI_LIST_APPEND(tail, g_strdup(exec_get_cmd_path()));
+        QAPI_LIST_APPEND(tail, g_strdup("/c"));
+#else
+        QAPI_LIST_APPEND(tail, g_strdup("/bin/sh"));
+        QAPI_LIST_APPEND(tail, g_strdup("-c"));
+#endif
+        QAPI_LIST_APPEND(tail, g_strdup(uri + strlen("exec:")));
+    } else if (strstart(uri, "rdma:", NULL)) {
+        if (inet_parse(isock, uri + strlen("rdma:"), errp)) {
+            qapi_free_InetSocketAddress(isock);
+            return false;
+        }
+        addr->transport = MIGRATION_ADDRESS_TYPE_RDMA;
+    } else if (strstart(uri, "tcp:", NULL) ||
+                strstart(uri, "unix:", NULL) ||
+                strstart(uri, "vsock:", NULL) ||
+                strstart(uri, "fd:", NULL)) {
+        addr->transport = MIGRATION_ADDRESS_TYPE_SOCKET;
+        saddr = socket_parse(uri, errp);
+        if (!saddr) {
+            return false;
+        }
+        addr->u.socket.type = saddr->type;
+        addr->u.socket.u = saddr->u;
+    } else if (strstart(uri, "file:", NULL)) {
+        addr->transport = MIGRATION_ADDRESS_TYPE_FILE;
+        addr->u.file.filename = g_strdup(uri + strlen("file:"));
+        if (file_parse_offset(addr->u.file.filename, &addr->u.file.offset,
+                              errp)) {
+            return false;
+        }
+    } else {
+        error_setg(errp, "unknown migration protocol: %s", uri);
+        return false;
+    }
+
+    val->channel_type = MIGRATION_CHANNEL_TYPE_MAIN;
+    val->addr = g_steal_pointer(&addr);
+    *channel = g_steal_pointer(&val);
+    return true;
+}
+
+static void qemu_start_incoming_migration(const char *uri, bool has_channels,
+                                          MigrationChannelList *channels,
+                                          Error **errp)
+{
+    MigrationChannel *channel = NULL;
+    MigrationAddress *addr = NULL;
     MigrationIncomingState *mis = migration_incoming_get_current();
 
-    /* URI is not suitable for migration? */
-    if (!migration_channels_and_uri_compatible(uri, errp)) {
+    /*
+     * Having preliminary checks for uri and channel
+     */
+    if (uri && has_channels) {
+        error_setg(errp, "'uri' and 'channels' arguments are mutually "
+                   "exclusive; exactly one of the two should be present in "
+                   "'migrate-incoming' qmp command ");
+        return;
+    } else if (channels) {
+        /* To verify that Migrate channel list has only item */
+        if (channels->next) {
+            error_setg(errp, "Channel list has more than one entries");
+            return;
+        }
+        channel = channels->value;
+    } else if (uri) {
+        /* caller uses the old URI syntax */
+        if (!migrate_uri_parse(uri, &channel, errp)) {
+            return;
+        }
+    } else {
+        error_setg(errp, "neither 'uri' or 'channels' argument are "
+                   "specified in 'migrate-incoming' qmp command ");
+        return;
+    }
+    addr = channel->addr;
+
+    /* transport mechanism not suitable for migration? */
+    if (!migration_channels_and_transport_compatible(addr, errp)) {
         return;
     }
 
     migrate_set_state(&mis->state, MIGRATION_STATUS_NONE,
                       MIGRATION_STATUS_SETUP);
 
-    if (strstart(uri, "tcp:", &p) ||
-        strstart(uri, "unix:", NULL) ||
-        strstart(uri, "vsock:", NULL)) {
-        socket_start_incoming_migration(p ? p : uri, errp);
+    if (addr->transport == MIGRATION_ADDRESS_TYPE_SOCKET) {
+        SocketAddress *saddr = &addr->u.socket;
+        if (saddr->type == SOCKET_ADDRESS_TYPE_INET ||
+            saddr->type == SOCKET_ADDRESS_TYPE_UNIX ||
+            saddr->type == SOCKET_ADDRESS_TYPE_VSOCK) {
+            socket_start_incoming_migration(saddr, errp);
+        } else if (saddr->type == SOCKET_ADDRESS_TYPE_FD) {
+            fd_start_incoming_migration(saddr->u.fd.str, errp);
+        }
 #ifdef CONFIG_RDMA
-    } else if (strstart(uri, "rdma:", &p)) {
+    } else if (addr->transport == MIGRATION_ADDRESS_TYPE_RDMA) {
         if (migrate_compress()) {
             error_setg(errp, "RDMA and compression can't be used together");
             return;
@@ -454,14 +577,12 @@ static void qemu_start_incoming_migration(const char *uri, Error **errp)
             error_setg(errp, "RDMA and multifd can't be used together");
             return;
         }
-        rdma_start_incoming_migration(p, errp);
+        rdma_start_incoming_migration(&addr->u.rdma, errp);
 #endif
-    } else if (strstart(uri, "exec:", &p)) {
-        exec_start_incoming_migration(p, errp);
-    } else if (strstart(uri, "fd:", &p)) {
-        fd_start_incoming_migration(p, errp);
-    } else if (strstart(uri, "file:", &p)) {
-        file_start_incoming_migration(p, errp);
+    } else if (addr->transport == MIGRATION_ADDRESS_TYPE_EXEC) {
+        exec_start_incoming_migration(addr->u.exec.args, errp);
+    } else if (addr->transport == MIGRATION_ADDRESS_TYPE_FILE) {
+        file_start_incoming_migration(&addr->u.file, errp);
     } else {
         error_setg(errp, "unknown migration protocol: %s", uri);
     }
@@ -472,6 +593,8 @@ static void process_incoming_migration_bh(void *opaque)
     Error *local_err = NULL;
     MigrationIncomingState *mis = opaque;
 
+    trace_vmstate_downtime_checkpoint("dst-precopy-bh-enter");
+
     /* If capability late_block_activate is set:
      * Only fire up the block code now if we're going to restart the
      * VM, else 'cont' will do it.
@@ -497,6 +620,8 @@ static void process_incoming_migration_bh(void *opaque)
      */
     qemu_announce_self(&mis->announce_timer, migrate_announce_params());
 
+    trace_vmstate_downtime_checkpoint("dst-precopy-bh-announced");
+
     multifd_load_shutdown();
 
     dirty_bitmap_mig_before_vm_start();
@@ -514,6 +639,7 @@ static void process_incoming_migration_bh(void *opaque)
     } else {
         runstate_set(global_state_get_runstate());
     }
+    trace_vmstate_downtime_checkpoint("dst-precopy-bh-vm-started");
     /*
      * This must happen after any state changes since as soon as an external
      * observer sees this event they might start to prod at the VM assuming
@@ -548,6 +674,8 @@ process_incoming_migration_co(void *opaque)
     ret = qemu_loadvm_state(mis->from_src_file);
     mis->loadvm_co = NULL;
 
+    trace_vmstate_downtime_checkpoint("dst-precopy-loadvm-completed");
+
     ps = postcopy_state_get();
     trace_process_incoming_migration_co_end(ret, ps);
     if (ps != POSTCOPY_INCOMING_NONE) {
@@ -1006,7 +1134,7 @@ static void fill_source_migration_info(MigrationInfo *info)
 {
     MigrationState *s = migrate_get_current();
     int state = qatomic_read(&s->state);
-    GSList *cur_blocker = migration_blockers;
+    GSList *cur_blocker = migration_blockers[migrate_mode()];
 
     info->blocked_reasons = NULL;
 
@@ -1356,6 +1484,17 @@ bool migration_in_postcopy(void)
     }
 }
 
+bool migration_postcopy_is_alive(int state)
+{
+    switch (state) {
+    case MIGRATION_STATUS_POSTCOPY_ACTIVE:
+    case MIGRATION_STATUS_POSTCOPY_RECOVER:
+        return true;
+    default:
+        return false;
+    }
+}
+
 bool migration_in_postcopy_after_devices(MigrationState *s)
 {
     return migration_in_postcopy() && s->postcopy_after_devices;
@@ -1438,7 +1577,6 @@ int migrate_init(MigrationState *s, Error **errp)
     s->to_dst_file = NULL;
     s->state = MIGRATION_STATUS_NONE;
     s->rp_state.from_dst_file = NULL;
-    s->rp_state.error = false;
     s->mbps = 0.0;
     s->pages_per_second = 0.0;
     s->downtime = 0;
@@ -1470,44 +1608,112 @@ int migrate_init(MigrationState *s, Error **errp)
     return 0;
 }
 
-int migrate_add_blocker_internal(Error **reasonp, Error **errp)
+static bool is_busy(Error **reasonp, Error **errp)
 {
+    ERRP_GUARD();
+
     /* Snapshots are similar to migrations, so check RUN_STATE_SAVE_VM too. */
     if (runstate_check(RUN_STATE_SAVE_VM) || !migration_is_idle()) {
         error_propagate_prepend(errp, *reasonp,
                                 "disallowing migration blocker "
                                 "(migration/snapshot in progress) for: ");
         *reasonp = NULL;
-        return -EBUSY;
+        return true;
     }
-
-    migration_blockers = g_slist_prepend(migration_blockers, *reasonp);
-    return 0;
+    return false;
 }
 
-int migrate_add_blocker(Error **reasonp, Error **errp)
+static bool is_only_migratable(Error **reasonp, Error **errp, int modes)
 {
-    if (only_migratable) {
+    ERRP_GUARD();
+
+    if (only_migratable && (modes & BIT(MIG_MODE_NORMAL))) {
         error_propagate_prepend(errp, *reasonp,
                                 "disallowing migration blocker "
                                 "(--only-migratable) for: ");
         *reasonp = NULL;
+        return true;
+    }
+    return false;
+}
+
+static int get_modes(MigMode mode, va_list ap)
+{
+    int modes = 0;
+
+    while (mode != -1 && mode != MIG_MODE_ALL) {
+        assert(mode >= MIG_MODE_NORMAL && mode < MIG_MODE__MAX);
+        modes |= BIT(mode);
+        mode = va_arg(ap, MigMode);
+    }
+    if (mode == MIG_MODE_ALL) {
+        modes = BIT(MIG_MODE__MAX) - 1;
+    }
+    return modes;
+}
+
+static int add_blockers(Error **reasonp, Error **errp, int modes)
+{
+    for (MigMode mode = 0; mode < MIG_MODE__MAX; mode++) {
+        if (modes & BIT(mode)) {
+            migration_blockers[mode] = g_slist_prepend(migration_blockers[mode],
+                                                       *reasonp);
+        }
+    }
+    return 0;
+}
+
+int migrate_add_blocker(Error **reasonp, Error **errp)
+{
+    return migrate_add_blocker_modes(reasonp, errp, MIG_MODE_ALL);
+}
+
+int migrate_add_blocker_normal(Error **reasonp, Error **errp)
+{
+    return migrate_add_blocker_modes(reasonp, errp, MIG_MODE_NORMAL, -1);
+}
+
+int migrate_add_blocker_modes(Error **reasonp, Error **errp, MigMode mode, ...)
+{
+    int modes;
+    va_list ap;
+
+    va_start(ap, mode);
+    modes = get_modes(mode, ap);
+    va_end(ap);
+
+    if (is_only_migratable(reasonp, errp, modes)) {
         return -EACCES;
+    } else if (is_busy(reasonp, errp)) {
+        return -EBUSY;
     }
+    return add_blockers(reasonp, errp, modes);
+}
 
-    return migrate_add_blocker_internal(reasonp, errp);
+int migrate_add_blocker_internal(Error **reasonp, Error **errp)
+{
+    int modes = BIT(MIG_MODE__MAX) - 1;
+
+    if (is_busy(reasonp, errp)) {
+        return -EBUSY;
+    }
+    return add_blockers(reasonp, errp, modes);
 }
 
 void migrate_del_blocker(Error **reasonp)
 {
     if (*reasonp) {
-        migration_blockers = g_slist_remove(migration_blockers, *reasonp);
+        for (MigMode mode = 0; mode < MIG_MODE__MAX; mode++) {
+            migration_blockers[mode] = g_slist_remove(migration_blockers[mode],
+                                                      *reasonp);
+        }
         error_free(*reasonp);
         *reasonp = NULL;
     }
 }
 
-void qmp_migrate_incoming(const char *uri, Error **errp)
+void qmp_migrate_incoming(const char *uri, bool has_channels,
+                          MigrationChannelList *channels, Error **errp)
 {
     Error *local_err = NULL;
     static bool once = true;
@@ -1525,7 +1731,7 @@ void qmp_migrate_incoming(const char *uri, Error **errp)
         return;
     }
 
-    qemu_start_incoming_migration(uri, &local_err);
+    qemu_start_incoming_migration(uri, has_channels, channels, &local_err);
 
     if (local_err) {
         yank_unregister_instance(MIGRATION_YANK_INSTANCE);
@@ -1561,7 +1767,7 @@ void qmp_migrate_recover(const char *uri, Error **errp)
      * only re-setup the migration stream and poke existing migration
      * to continue using that newly established channel.
      */
-    qemu_start_incoming_migration(uri, errp);
+    qemu_start_incoming_migration(uri, false, NULL, errp);
 }
 
 void qmp_migrate_pause(Error **errp)
@@ -1570,8 +1776,15 @@ void qmp_migrate_pause(Error **errp)
     MigrationIncomingState *mis = migration_incoming_get_current();
     int ret = 0;
 
-    if (ms->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
+    if (migration_postcopy_is_alive(ms->state)) {
         /* Source side, during postcopy */
+        Error *error = NULL;
+
+        /* Tell the core migration that we're pausing */
+        error_setg(&error, "Postcopy migration is paused by the user");
+        migrate_set_error(ms, error);
+        error_free(error);
+
         qemu_mutex_lock(&ms->qemu_file_lock);
         if (ms->to_dst_file) {
             ret = qemu_file_shutdown(ms->to_dst_file);
@@ -1580,10 +1793,17 @@ void qmp_migrate_pause(Error **errp)
         if (ret) {
             error_setg(errp, "Failed to pause source migration");
         }
+
+        /*
+         * Kick the migration thread out of any waiting windows (on behalf
+         * of the rp thread).
+         */
+        migration_rp_kick(ms);
+
         return;
     }
 
-    if (mis->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
+    if (migration_postcopy_is_alive(mis->state)) {
         ret = qemu_file_shutdown(mis->from_src_file);
         if (ret) {
             error_setg(errp, "Failed to pause destination migration");
@@ -1592,17 +1812,19 @@ void qmp_migrate_pause(Error **errp)
     }
 
     error_setg(errp, "migrate-pause is currently only supported "
-               "during postcopy-active state");
+               "during postcopy-active or postcopy-recover state");
 }
 
 bool migration_is_blocked(Error **errp)
 {
+    GSList *blockers = migration_blockers[migrate_mode()];
+
     if (qemu_savevm_state_blocked(errp)) {
         return true;
     }
 
-    if (migration_blockers) {
-        error_propagate(errp, error_copy(migration_blockers->data));
+    if (blockers) {
+        error_propagate(errp, error_copy(blockers->data));
         return true;
     }
 
@@ -1702,17 +1924,46 @@ static bool migrate_prepare(MigrationState *s, bool blk, bool blk_inc,
     return true;
 }
 
-void qmp_migrate(const char *uri, bool has_blk, bool blk,
+void qmp_migrate(const char *uri, bool has_channels,
+                 MigrationChannelList *channels, bool has_blk, bool blk,
                  bool has_inc, bool inc, bool has_detach, bool detach,
                  bool has_resume, bool resume, Error **errp)
 {
     bool resume_requested;
     Error *local_err = NULL;
     MigrationState *s = migrate_get_current();
-    const char *p = NULL;
+    MigrationChannel *channel = NULL;
+    MigrationAddress *addr = NULL;
 
-    /* URI is not suitable for migration? */
-    if (!migration_channels_and_uri_compatible(uri, errp)) {
+    /*
+     * Having preliminary checks for uri and channel
+     */
+    if (uri && has_channels) {
+        error_setg(errp, "'uri' and 'channels' arguments are mutually "
+                   "exclusive; exactly one of the two should be present in "
+                   "'migrate' qmp command ");
+        return;
+    } else if (channels) {
+        /* To verify that Migrate channel list has only item */
+        if (channels->next) {
+            error_setg(errp, "Channel list has more than one entries");
+            return;
+        }
+        channel = channels->value;
+    } else if (uri) {
+        /* caller uses the old URI syntax */
+        if (!migrate_uri_parse(uri, &channel, errp)) {
+            return;
+        }
+    } else {
+        error_setg(errp, "neither 'uri' or 'channels' argument are "
+                   "specified in 'migrate' qmp command ");
+        return;
+    }
+    addr = channel->addr;
+
+    /* transport mechanism not suitable for migration? */
+    if (!migration_channels_and_transport_compatible(addr, errp)) {
         return;
     }
 
@@ -1729,20 +1980,23 @@ void qmp_migrate(const char *uri, bool has_blk, bool blk,
         }
     }
 
-    if (strstart(uri, "tcp:", &p) ||
-        strstart(uri, "unix:", NULL) ||
-        strstart(uri, "vsock:", NULL)) {
-        socket_start_outgoing_migration(s, p ? p : uri, &local_err);
+    if (addr->transport == MIGRATION_ADDRESS_TYPE_SOCKET) {
+        SocketAddress *saddr = &addr->u.socket;
+        if (saddr->type == SOCKET_ADDRESS_TYPE_INET ||
+            saddr->type == SOCKET_ADDRESS_TYPE_UNIX ||
+            saddr->type == SOCKET_ADDRESS_TYPE_VSOCK) {
+            socket_start_outgoing_migration(s, saddr, &local_err);
+        } else if (saddr->type == SOCKET_ADDRESS_TYPE_FD) {
+            fd_start_outgoing_migration(s, saddr->u.fd.str, &local_err);
+        }
 #ifdef CONFIG_RDMA
-    } else if (strstart(uri, "rdma:", &p)) {
-        rdma_start_outgoing_migration(s, p, &local_err);
+    } else if (addr->transport == MIGRATION_ADDRESS_TYPE_RDMA) {
+        rdma_start_outgoing_migration(s, &addr->u.rdma, &local_err);
 #endif
-    } else if (strstart(uri, "exec:", &p)) {
-        exec_start_outgoing_migration(s, p, &local_err);
-    } else if (strstart(uri, "fd:", &p)) {
-        fd_start_outgoing_migration(s, p, &local_err);
-    } else if (strstart(uri, "file:", &p)) {
-        file_start_outgoing_migration(s, p, &local_err);
+    } else if (addr->transport == MIGRATION_ADDRESS_TYPE_EXEC) {
+        exec_start_outgoing_migration(s, addr->u.exec.args, &local_err);
+    } else if (addr->transport == MIGRATION_ADDRESS_TYPE_FILE) {
+        file_start_outgoing_migration(s, &addr->u.file, &local_err);
     } else {
         error_setg(&local_err, QERR_INVALID_PARAMETER_VALUE, "uri",
                    "a valid migration protocol");
@@ -1777,19 +2031,21 @@ void qmp_migrate_continue(MigrationStatus state, Error **errp)
     qemu_sem_post(&s->pause_sem);
 }
 
-/* migration thread support */
-/*
- * Something bad happened to the RP stream, mark an error
- * The caller shall print or trace something to indicate why
- */
-static void mark_source_rp_bad(MigrationState *s)
+int migration_rp_wait(MigrationState *s)
 {
-    s->rp_state.error = true;
-}
+    /* If migration has failure already, ignore the wait */
+    if (migrate_has_error(s)) {
+        return -1;
+    }
 
-void migration_rp_wait(MigrationState *s)
-{
     qemu_sem_wait(&s->rp_state.rp_sem);
+
+    /* After wait, double check that there's no failure */
+    if (migrate_has_error(s)) {
+        return -1;
+    }
+
+    return 0;
 }
 
 void migration_rp_kick(MigrationState *s)
@@ -1817,8 +2073,9 @@ static struct rp_cmd_args {
  * We're allowed to send more than requested (e.g. to round to our page size)
  * and we don't need to send pages that have already been sent.
  */
-static void migrate_handle_rp_req_pages(MigrationState *ms, const char* rbname,
-                                       ram_addr_t start, size_t len)
+static void
+migrate_handle_rp_req_pages(MigrationState *ms, const char* rbname,
+                            ram_addr_t start, size_t len, Error **errp)
 {
     long our_host_ps = qemu_real_host_page_size();
 
@@ -1830,38 +2087,37 @@ static void migrate_handle_rp_req_pages(MigrationState *ms, const char* rbname,
      */
     if (!QEMU_IS_ALIGNED(start, our_host_ps) ||
         !QEMU_IS_ALIGNED(len, our_host_ps)) {
-        error_report("%s: Misaligned page request, start: " RAM_ADDR_FMT
-                     " len: %zd", __func__, start, len);
-        mark_source_rp_bad(ms);
+        error_setg(errp, "MIG_RP_MSG_REQ_PAGES: Misaligned page request, start:"
+                   RAM_ADDR_FMT " len: %zd", start, len);
         return;
     }
 
-    if (ram_save_queue_pages(rbname, start, len)) {
-        mark_source_rp_bad(ms);
-    }
+    ram_save_queue_pages(rbname, start, len, errp);
 }
 
-static int migrate_handle_rp_recv_bitmap(MigrationState *s, char *block_name)
+static bool migrate_handle_rp_recv_bitmap(MigrationState *s, char *block_name,
+                                          Error **errp)
 {
     RAMBlock *block = qemu_ram_block_by_name(block_name);
 
     if (!block) {
-        error_report("%s: invalid block name '%s'", __func__, block_name);
-        return -EINVAL;
+        error_setg(errp, "MIG_RP_MSG_RECV_BITMAP has invalid block name '%s'",
+                   block_name);
+        return false;
     }
 
     /* Fetch the received bitmap and refresh the dirty bitmap */
-    return ram_dirty_bitmap_reload(s, block);
+    return ram_dirty_bitmap_reload(s, block, errp);
 }
 
-static int migrate_handle_rp_resume_ack(MigrationState *s, uint32_t value)
+static bool migrate_handle_rp_resume_ack(MigrationState *s,
+                                         uint32_t value, Error **errp)
 {
     trace_source_return_path_thread_resume_ack(value);
 
     if (value != MIGRATION_RESUME_ACK_VALUE) {
-        error_report("%s: illegal resume_ack value %"PRIu32,
-                     __func__, value);
-        return -1;
+        error_setg(errp, "illegal resume_ack value %"PRIu32, value);
+        return false;
     }
 
     /* Now both sides are active. */
@@ -1871,7 +2127,7 @@ static int migrate_handle_rp_resume_ack(MigrationState *s, uint32_t value)
     /* Notify send thread that time to continue send pages */
     migration_rp_kick(s);
 
-    return 0;
+    return true;
 }
 
 /*
@@ -1919,48 +2175,46 @@ static void *source_return_path_thread(void *opaque)
     uint32_t tmp32, sibling_error;
     ram_addr_t start = 0; /* =0 to silence warning */
     size_t  len = 0, expected_len;
+    Error *err = NULL;
     int res;
 
     trace_source_return_path_thread_entry();
     rcu_register_thread();
 
-    while (!ms->rp_state.error && !qemu_file_get_error(rp) &&
-           migration_is_setup_or_active(ms->state)) {
+    while (migration_is_setup_or_active(ms->state)) {
         trace_source_return_path_thread_loop_top();
+
         header_type = qemu_get_be16(rp);
         header_len = qemu_get_be16(rp);
 
         if (qemu_file_get_error(rp)) {
-            mark_source_rp_bad(ms);
+            qemu_file_get_error_obj(rp, &err);
             goto out;
         }
 
         if (header_type >= MIG_RP_MSG_MAX ||
             header_type == MIG_RP_MSG_INVALID) {
-            error_report("RP: Received invalid message 0x%04x length 0x%04x",
-                         header_type, header_len);
-            mark_source_rp_bad(ms);
+            error_setg(&err, "Received invalid message 0x%04x length 0x%04x",
+                       header_type, header_len);
             goto out;
         }
 
         if ((rp_cmd_args[header_type].len != -1 &&
             header_len != rp_cmd_args[header_type].len) ||
             header_len > sizeof(buf)) {
-            error_report("RP: Received '%s' message (0x%04x) with"
-                         "incorrect length %d expecting %zu",
-                         rp_cmd_args[header_type].name, header_type, header_len,
-                         (size_t)rp_cmd_args[header_type].len);
-            mark_source_rp_bad(ms);
+            error_setg(&err, "Received '%s' message (0x%04x) with"
+                       "incorrect length %d expecting %zu",
+                       rp_cmd_args[header_type].name, header_type, header_len,
+                       (size_t)rp_cmd_args[header_type].len);
             goto out;
         }
 
         /* We know we've got a valid header by this point */
         res = qemu_get_buffer(rp, buf, header_len);
         if (res != header_len) {
-            error_report("RP: Failed reading data for message 0x%04x"
-                         " read %d expected %d",
-                         header_type, res, header_len);
-            mark_source_rp_bad(ms);
+            error_setg(&err, "Failed reading data for message 0x%04x"
+                       " read %d expected %d",
+                       header_type, res, header_len);
             goto out;
         }
 
@@ -1970,8 +2224,7 @@ static void *source_return_path_thread(void *opaque)
             sibling_error = ldl_be_p(buf);
             trace_source_return_path_thread_shut(sibling_error);
             if (sibling_error) {
-                error_report("RP: Sibling indicated error %d", sibling_error);
-                mark_source_rp_bad(ms);
+                error_setg(&err, "Sibling indicated error %d", sibling_error);
             }
             /*
              * We'll let the main thread deal with closing the RP
@@ -1989,7 +2242,10 @@ static void *source_return_path_thread(void *opaque)
         case MIG_RP_MSG_REQ_PAGES:
             start = ldq_be_p(buf);
             len = ldl_be_p(buf + 8);
-            migrate_handle_rp_req_pages(ms, NULL, start, len);
+            migrate_handle_rp_req_pages(ms, NULL, start, len, &err);
+            if (err) {
+                goto out;
+            }
             break;
 
         case MIG_RP_MSG_REQ_PAGES_ID:
@@ -2004,32 +2260,32 @@ static void *source_return_path_thread(void *opaque)
                 expected_len += tmp32;
             }
             if (header_len != expected_len) {
-                error_report("RP: Req_Page_id with length %d expecting %zd",
-                             header_len, expected_len);
-                mark_source_rp_bad(ms);
+                error_setg(&err, "Req_Page_id with length %d expecting %zd",
+                           header_len, expected_len);
+                goto out;
+            }
+            migrate_handle_rp_req_pages(ms, (char *)&buf[13], start, len,
+                                        &err);
+            if (err) {
                 goto out;
             }
-            migrate_handle_rp_req_pages(ms, (char *)&buf[13], start, len);
             break;
 
         case MIG_RP_MSG_RECV_BITMAP:
             if (header_len < 1) {
-                error_report("%s: missing block name", __func__);
-                mark_source_rp_bad(ms);
+                error_setg(&err, "MIG_RP_MSG_RECV_BITMAP missing block name");
                 goto out;
             }
             /* Format: len (1B) + idstr (<255B). This ends the idstr. */
             buf[buf[0] + 1] = '\0';
-            if (migrate_handle_rp_recv_bitmap(ms, (char *)(buf + 1))) {
-                mark_source_rp_bad(ms);
+            if (!migrate_handle_rp_recv_bitmap(ms, (char *)(buf + 1), &err)) {
                 goto out;
             }
             break;
 
         case MIG_RP_MSG_RESUME_ACK:
             tmp32 = ldl_be_p(buf);
-            if (migrate_handle_rp_resume_ack(ms, tmp32)) {
-                mark_source_rp_bad(ms);
+            if (!migrate_handle_rp_resume_ack(ms, tmp32, &err)) {
                 goto out;
             }
             break;
@@ -2045,13 +2301,29 @@ static void *source_return_path_thread(void *opaque)
     }
 
 out:
-    if (qemu_file_get_error(rp)) {
+    if (err) {
+        migrate_set_error(ms, err);
+        error_free(err);
         trace_source_return_path_thread_bad_end();
-        mark_source_rp_bad(ms);
+    }
+
+    if (ms->state == MIGRATION_STATUS_POSTCOPY_RECOVER) {
+        /*
+         * this will be extremely unlikely: that we got yet another network
+         * issue during recovering of the 1st network failure.. during this
+         * period the main migration thread can be waiting on rp_sem for
+         * this thread to sync with the other side.
+         *
+         * When this happens, explicitly kick the migration thread out of
+         * RECOVER stage and back to PAUSED, so the admin can try
+         * everything again.
+         */
+        migration_rp_kick(ms);
     }
 
     trace_source_return_path_thread_end();
     rcu_unregister_thread();
+
     return NULL;
 }
 
@@ -2073,12 +2345,11 @@ static int open_return_path_on_source(MigrationState *ms)
     return 0;
 }
 
-static int close_return_path_on_source(MigrationState *ms)
+/* Return true if error detected, or false otherwise */
+static bool close_return_path_on_source(MigrationState *ms)
 {
-    int ret;
-
     if (!ms->rp_state.rp_thread_created) {
-        return 0;
+        return false;
     }
 
     trace_migration_return_path_end_before();
@@ -2096,18 +2367,13 @@ static int close_return_path_on_source(MigrationState *ms)
         }
     }
 
-    trace_await_return_path_close_on_source_joining();
     qemu_thread_join(&ms->rp_state.rp_thread);
     ms->rp_state.rp_thread_created = false;
-    trace_await_return_path_close_on_source_close();
-
-    ret = ms->rp_state.error;
-    ms->rp_state.error = false;
-
     migration_release_dst_files(ms);
+    trace_migration_return_path_end_after();
 
-    trace_migration_return_path_end_after(ret);
-    return ret;
+    /* Return path will persist the error in MigrationState when quit */
+    return migrate_has_error(ms);
 }
 
 static inline void
@@ -2126,7 +2392,6 @@ static int postcopy_start(MigrationState *ms, Error **errp)
     int ret;
     QIOChannelBuffer *bioc;
     QEMUFile *fb;
-    int64_t time_at_stop = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
     uint64_t bandwidth = migrate_max_postcopy_bandwidth();
     bool restart_block = false;
     int cur_state = MIGRATION_STATUS_ACTIVE;
@@ -2148,9 +2413,11 @@ static int postcopy_start(MigrationState *ms, Error **errp)
     qemu_mutex_lock_iothread();
     trace_postcopy_start_set_run();
 
+    migration_downtime_start(ms);
+
     qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
     global_state_store();
-    ret = vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
+    ret = migration_stop_vm(RUN_STATE_FINISH_MIGRATE);
     if (ret < 0) {
         goto fail;
     }
@@ -2250,7 +2517,7 @@ static int postcopy_start(MigrationState *ms, Error **errp)
     ms->postcopy_after_devices = true;
     migration_call_notifiers(ms);
 
-    ms->downtime =  qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - time_at_stop;
+    migration_downtime_end(ms);
 
     qemu_mutex_unlock_iothread();
 
@@ -2346,13 +2613,13 @@ static int migration_completion_precopy(MigrationState *s,
     int ret;
 
     qemu_mutex_lock_iothread();
-    s->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+    migration_downtime_start(s);
     qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
 
     s->vm_old_state = runstate_get();
     global_state_store();
 
-    ret = vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
+    ret = migration_stop_vm(RUN_STATE_FINISH_MIGRATE);
     trace_migration_completion_vm_stop(ret);
     if (ret < 0) {
         goto out_unlock;
@@ -2519,7 +2786,9 @@ static int postcopy_resume_handshake(MigrationState *s)
     qemu_savevm_send_postcopy_resume(s->to_dst_file);
 
     while (s->state == MIGRATION_STATUS_POSTCOPY_RECOVER) {
-        migration_rp_wait(s);
+        if (migration_rp_wait(s)) {
+            return -1;
+        }
     }
 
     if (s->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
@@ -2703,15 +2972,8 @@ static void migration_calculate_complete(MigrationState *s)
     int64_t end_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
     int64_t transfer_time;
 
+    migration_downtime_end(s);
     s->total_time = end_time - s->start_time;
-    if (!s->downtime) {
-        /*
-         * It's still not set, so we are precopy migration.  For
-         * postcopy, downtime is calculated during postcopy_start().
-         */
-        s->downtime = end_time - s->downtime_start;
-    }
-
     transfer_time = s->total_time - s->setup_time;
     if (transfer_time) {
         s->mbps = ((double) bytes * 8.0) / transfer_time / 1000;
@@ -3130,7 +3392,7 @@ static void bg_migration_vm_start_bh(void *opaque)
     s->vm_start_bh = NULL;
 
     vm_start();
-    s->downtime = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - s->downtime_start;
+    migration_downtime_end(s);
 }
 
 /**
@@ -3197,7 +3459,7 @@ static void *bg_migration_thread(void *opaque)
     s->setup_time = qemu_clock_get_ms(QEMU_CLOCK_HOST) - setup_start;
 
     trace_migration_thread_setup_complete();
-    s->downtime_start = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+    migration_downtime_start(s);
 
     qemu_mutex_lock_iothread();
 
@@ -3210,7 +3472,7 @@ static void *bg_migration_thread(void *opaque)
 
     global_state_store();
     /* Forcibly stop VM before saving state of vCPUs and devices */
-    if (vm_stop_force_state(RUN_STATE_PAUSED)) {
+    if (migration_stop_vm(RUN_STATE_PAUSED)) {
         goto fail;
     }
     /*

migration/migration.h

@@ -308,7 +308,6 @@ struct MigrationState {
         /* Protected by qemu_file_lock */
         QEMUFile     *from_dst_file;
         QemuThread    rp_thread;
-        bool          error;
         /*
          * We can also check non-zero of rp_thread, but there's no "official"
          * way to do this, so this bool makes it slightly more elegant.
@@ -495,6 +494,7 @@ int migrate_init(MigrationState *s, Error **errp);
 bool migration_is_blocked(Error **errp);
 /* True if outgoing migration has entered postcopy phase */
 bool migration_in_postcopy(void);
+bool migration_postcopy_is_alive(int state);
 MigrationState *migrate_get_current(void);
 
 uint64_t ram_get_total_transferred_pages(void);
@@ -520,7 +520,8 @@ bool check_dirty_bitmap_mig_alias_map(const BitmapMigrationNodeAliasList *bbm,
                                       Error **errp);
 
 void migrate_add_address(SocketAddress *address);
-
+bool migrate_uri_parse(const char *uri, MigrationChannel **channel,
+                       Error **errp);
 int foreach_not_ignored_block(RAMBlockIterFunc func, void *opaque);
 
 #define qemu_ram_foreach_block \
@@ -535,8 +536,11 @@ void migration_populate_vfio_info(MigrationInfo *info);
 void migration_reset_vfio_bytes_transferred(void);
 void postcopy_temp_page_reset(PostcopyTmpPage *tmp_page);
 
-/* Migration thread waiting for return path thread. */
-void migration_rp_wait(MigrationState *s);
+/*
+ * Migration thread waiting for return path thread.  Return non-zero if an
+ * error is detected.
+ */
+int migration_rp_wait(MigrationState *s);
 /*
  * Kick the migration thread waiting for return path messages.  NOTE: the
  * name can be slightly confusing (when read as "kick the rp thread"), just
@@ -544,4 +548,6 @@ void migration_rp_wait(MigrationState *s);
  */
 void migration_rp_kick(MigrationState *s);
 
+int migration_stop_vm(RunState state);
+
 #endif

migration/options.c

@@ -176,6 +176,9 @@ Property migration_properties[] = {
     DEFINE_PROP_UINT64("vcpu-dirty-limit", MigrationState,
                        parameters.vcpu_dirty_limit,
                        DEFAULT_MIGRATE_VCPU_DIRTY_LIMIT),
+    DEFINE_PROP_MIG_MODE("mode", MigrationState,
+                      parameters.mode,
+                      MIG_MODE_NORMAL),
 
     /* Migration capabilities */
     DEFINE_PROP_MIG_CAP("x-xbzrle", MIGRATION_CAPABILITY_XBZRLE),
@@ -827,6 +830,13 @@ uint64_t migrate_max_postcopy_bandwidth(void)
     return s->parameters.max_postcopy_bandwidth;
 }
 
+MigMode migrate_mode(void)
+{
+    MigrationState *s = migrate_get_current();
+
+    return s->parameters.mode;
+}
+
 int migrate_multifd_channels(void)
 {
     MigrationState *s = migrate_get_current();
@@ -999,6 +1009,8 @@ MigrationParameters *qmp_query_migrate_parameters(Error **errp)
     params->x_vcpu_dirty_limit_period = s->parameters.x_vcpu_dirty_limit_period;
     params->has_vcpu_dirty_limit = true;
     params->vcpu_dirty_limit = s->parameters.vcpu_dirty_limit;
+    params->has_mode = true;
+    params->mode = s->parameters.mode;
 
     return params;
 }
@@ -1034,6 +1046,7 @@ void migrate_params_init(MigrationParameters *params)
     params->has_announce_step = true;
     params->has_x_vcpu_dirty_limit_period = true;
     params->has_vcpu_dirty_limit = true;
+    params->has_mode = true;
 }
 
 /*
@@ -1331,6 +1344,10 @@ static void migrate_params_test_apply(MigrateSetParameters *params,
     if (params->has_vcpu_dirty_limit) {
         dest->vcpu_dirty_limit = params->vcpu_dirty_limit;
     }
+
+    if (params->has_mode) {
+        dest->mode = params->mode;
+    }
 }
 
 static void migrate_params_apply(MigrateSetParameters *params, Error **errp)
@@ -1471,6 +1488,10 @@ static void migrate_params_apply(MigrateSetParameters *params, Error **errp)
     if (params->has_vcpu_dirty_limit) {
         s->parameters.vcpu_dirty_limit = params->vcpu_dirty_limit;
     }
+
+    if (params->has_mode) {
+        s->parameters.mode = params->mode;
+    }
 }
 
 void qmp_migrate_set_parameters(MigrateSetParameters *params, Error **errp)

migration/options.h

@@ -83,6 +83,7 @@ uint8_t migrate_max_cpu_throttle(void);
 uint64_t migrate_max_bandwidth(void);
 uint64_t migrate_avail_switchover_bandwidth(void);
 uint64_t migrate_max_postcopy_bandwidth(void);
+MigMode migrate_mode(void);
 int migrate_multifd_channels(void);
 MultiFDCompression migrate_multifd_compression(void);
 int migrate_multifd_zlib_level(void);

migration/qemu-file.c

@@ -137,7 +137,7 @@ QEMUFile *qemu_file_new_input(QIOChannel *ioc)
  *
  * If errp is specified, a verbose error message will be copied over.
  */
-static int qemu_file_get_error_obj(QEMUFile *f, Error **errp)
+int qemu_file_get_error_obj(QEMUFile *f, Error **errp)
 {
     if (!f->last_error) {
         return 0;

migration/qemu-file.h

@@ -68,6 +68,7 @@ int coroutine_mixed_fn qemu_peek_byte(QEMUFile *f, int offset);
 void qemu_file_skip(QEMUFile *f, int size);
 int qemu_file_get_error_obj_any(QEMUFile *f1, QEMUFile *f2, Error **errp);
 void qemu_file_set_error_obj(QEMUFile *f, int ret, Error *err);
+int qemu_file_get_error_obj(QEMUFile *f, Error **errp);
 void qemu_file_set_error(QEMUFile *f, int ret);
 int qemu_file_shutdown(QEMUFile *f);
 QEMUFile *qemu_file_get_return_path(QEMUFile *f);

migration/ram.c

@@ -1898,7 +1898,8 @@ static void migration_page_queue_free(RAMState *rs)
  * @start: starting address from the start of the RAMBlock
  * @len: length (in bytes) to send
  */
-int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len)
+int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len,
+                         Error **errp)
 {
     RAMBlock *ramblock;
     RAMState *rs = ram_state;
@@ -1915,7 +1916,7 @@ int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len)
              * Shouldn't happen, we can't reuse the last RAMBlock if
              * it's the 1st request.
              */
-            error_report("ram_save_queue_pages no previous block");
+            error_setg(errp, "MIG_RP_MSG_REQ_PAGES has no previous block");
             return -1;
         }
     } else {
@@ -1923,16 +1924,17 @@ int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len)
 
         if (!ramblock) {
             /* We shouldn't be asked for a non-existent RAMBlock */
-            error_report("ram_save_queue_pages no block '%s'", rbname);
+            error_setg(errp, "MIG_RP_MSG_REQ_PAGES has no block '%s'", rbname);
             return -1;
         }
         rs->last_req_rb = ramblock;
     }
     trace_ram_save_queue_pages(ramblock->idstr, start, len);
     if (!offset_in_ramblock(ramblock, start + len - 1)) {
-        error_report("%s request overrun start=" RAM_ADDR_FMT " len="
-                     RAM_ADDR_FMT " blocklen=" RAM_ADDR_FMT,
-                     __func__, start, len, ramblock->used_length);
+        error_setg(errp, "MIG_RP_MSG_REQ_PAGES request overrun, "
+                   "start=" RAM_ADDR_FMT " len="
+                   RAM_ADDR_FMT " blocklen=" RAM_ADDR_FMT,
+                   start, len, ramblock->used_length);
         return -1;
     }
 
@@ -1964,9 +1966,9 @@ int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len)
         assert(len % page_size == 0);
         while (len) {
             if (ram_save_host_page_urgent(pss)) {
-                error_report("%s: ram_save_host_page_urgent() failed: "
-                             "ramblock=%s, start_addr=0x"RAM_ADDR_FMT,
-                             __func__, ramblock->idstr, start);
+                error_setg(errp, "ram_save_host_page_urgent() failed: "
+                           "ramblock=%s, start_addr=0x"RAM_ADDR_FMT,
+                           ramblock->idstr, start);
                 ret = -1;
                 break;
             }
@@ -4097,7 +4099,9 @@ static int ram_dirty_bitmap_sync_all(MigrationState *s, RAMState *rs)
 
     /* Wait until all the ramblocks' dirty bitmap synced */
     while (qatomic_read(&rs->postcopy_bmap_sync_requested)) {
-        migration_rp_wait(s);
+        if (migration_rp_wait(s)) {
+            return -1;
+        }
     }
 
     trace_ram_dirty_bitmap_sync_complete();
@@ -4109,10 +4113,11 @@ static int ram_dirty_bitmap_sync_all(MigrationState *s, RAMState *rs)
  * Read the received bitmap, revert it as the initial dirty bitmap.
  * This is only used when the postcopy migration is paused but wants
  * to resume from a middle point.
+ *
+ * Returns true if succeeded, false for errors.
  */
-int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *block)
+bool ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *block, Error **errp)
 {
-    int ret = -EINVAL;
     /* from_dst_file is always valid because we're within rp_thread */
     QEMUFile *file = s->rp_state.from_dst_file;
     g_autofree unsigned long *le_bitmap = NULL;
@@ -4124,9 +4129,9 @@ int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *block)
     trace_ram_dirty_bitmap_reload_begin(block->idstr);
 
     if (s->state != MIGRATION_STATUS_POSTCOPY_RECOVER) {
-        error_report("%s: incorrect state %s", __func__,
-                     MigrationStatus_str(s->state));
-        return -EINVAL;
+        error_setg(errp, "Reload bitmap in incorrect state %s",
+                   MigrationStatus_str(s->state));
+        return false;
     }
 
     /*
@@ -4142,27 +4147,25 @@ int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *block)
 
     /* The size of the bitmap should match with our ramblock */
     if (size != local_size) {
-        error_report("%s: ramblock '%s' bitmap size mismatch "
-                     "(0x%"PRIx64" != 0x%"PRIx64")", __func__,
-                     block->idstr, size, local_size);
-        return -EINVAL;
+        error_setg(errp, "ramblock '%s' bitmap size mismatch (0x%"PRIx64
+                   " != 0x%"PRIx64")", block->idstr, size, local_size);
+        return false;
     }
 
     size = qemu_get_buffer(file, (uint8_t *)le_bitmap, local_size);
     end_mark = qemu_get_be64(file);
 
-    ret = qemu_file_get_error(file);
-    if (ret || size != local_size) {
-        error_report("%s: read bitmap failed for ramblock '%s': %d"
-                     " (size 0x%"PRIx64", got: 0x%"PRIx64")",
-                     __func__, block->idstr, ret, local_size, size);
-        return -EIO;
+    if (qemu_file_get_error(file) || size != local_size) {
+        error_setg(errp, "read bitmap failed for ramblock '%s': "
+                   "(size 0x%"PRIx64", got: 0x%"PRIx64")",
+                   block->idstr, local_size, size);
+        return false;
     }
 
     if (end_mark != RAMBLOCK_RECV_BITMAP_ENDING) {
-        error_report("%s: ramblock '%s' end mark incorrect: 0x%"PRIx64,
-                     __func__, block->idstr, end_mark);
-        return -EINVAL;
+        error_setg(errp, "ramblock '%s' end mark incorrect: 0x%"PRIx64,
+                   block->idstr, end_mark);
+        return false;
     }
 
     /*
@@ -4194,7 +4197,7 @@ int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *block)
      */
     migration_rp_kick(s);
 
-    return 0;
+    return true;
 }
 
 static int ram_resume_prepare(MigrationState *s, void *opaque)

migration/ram.h

@@ -50,7 +50,8 @@ uint64_t ram_bytes_total(void);
 void mig_throttle_counter_reset(void);
 
 uint64_t ram_pagesize_summary(void);
-int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len);
+int ram_save_queue_pages(const char *rbname, ram_addr_t start, ram_addr_t len,
+                         Error **errp);
 void ram_postcopy_migrated_memory_release(MigrationState *ms);
 /* For outgoing discard bitmap */
 void ram_postcopy_send_discard_bitmap(MigrationState *ms);
@@ -70,7 +71,7 @@ void ramblock_recv_bitmap_set(RAMBlock *rb, void *host_addr);
 void ramblock_recv_bitmap_set_range(RAMBlock *rb, void *host_addr, size_t nr);
 int64_t ramblock_recv_bitmap_send(QEMUFile *file,
                                   const char *block_name);
-int ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *rb);
+bool ram_dirty_bitmap_reload(MigrationState *s, RAMBlock *rb, Error **errp);
 bool ramblock_page_is_discarded(RAMBlock *rb, ram_addr_t start);
 void postcopy_preempt_shutdown_file(MigrationState *s);
 void *postcopy_preempt_thread(void *opaque);

migration/rdma.c

@@ -289,7 +289,6 @@ typedef struct RDMALocalBlocks {
 typedef struct RDMAContext {
     char *host;
     int port;
-    char *host_port;
 
     RDMAWorkRequestData wr_data[RDMA_WRID_MAX];
 
@@ -2431,9 +2430,7 @@ static void qemu_rdma_cleanup(RDMAContext *rdma)
         rdma->channel = NULL;
     }
     g_free(rdma->host);
-    g_free(rdma->host_port);
     rdma->host = NULL;
-    rdma->host_port = NULL;
 }
 
 
@@ -2723,28 +2720,16 @@ static void qemu_rdma_return_path_dest_init(RDMAContext *rdma_return_path,
     rdma_return_path->is_return_path = true;
 }
 
-static RDMAContext *qemu_rdma_data_init(const char *host_port, Error **errp)
+static RDMAContext *qemu_rdma_data_init(InetSocketAddress *saddr, Error **errp)
 {
     RDMAContext *rdma = NULL;
-    InetSocketAddress *addr;
 
     rdma = g_new0(RDMAContext, 1);
     rdma->current_index = -1;
     rdma->current_chunk = -1;
 
-    addr = g_new(InetSocketAddress, 1);
-    if (!inet_parse(addr, host_port, NULL)) {
-        rdma->port = atoi(addr->port);
-        rdma->host = g_strdup(addr->host);
-        rdma->host_port = g_strdup(host_port);
-    } else {
-        error_setg(errp, "RDMA ERROR: bad RDMA migration address '%s'",
-                   host_port);
-        g_free(rdma);
-        rdma = NULL;
-    }
-
-    qapi_free_InetSocketAddress(addr);
+    rdma->host = g_strdup(saddr->host);
+    rdma->port = atoi(saddr->port);
     return rdma;
 }
 
@@ -3353,6 +3338,7 @@ static int qemu_rdma_accept(RDMAContext *rdma)
                                             .private_data_len = sizeof(cap),
                                          };
     RDMAContext *rdma_return_path = NULL;
+    g_autoptr(InetSocketAddress) isock = g_new0(InetSocketAddress, 1);
     struct rdma_cm_event *cm_event;
     struct ibv_context *verbs;
     int ret;
@@ -3367,13 +3353,16 @@ static int qemu_rdma_accept(RDMAContext *rdma)
         goto err_rdma_dest_wait;
     }
 
+    isock->host = rdma->host;
+    isock->port = g_strdup_printf("%d", rdma->port);
+
     /*
      * initialize the RDMAContext for return path for postcopy after first
      * connection request reached.
      */
     if ((migrate_postcopy() || migrate_return_path())
         && !rdma->is_return_path) {
-        rdma_return_path = qemu_rdma_data_init(rdma->host_port, NULL);
+        rdma_return_path = qemu_rdma_data_init(isock, NULL);
         if (rdma_return_path == NULL) {
             rdma_ack_cm_event(cm_event);
             goto err_rdma_dest_wait;
@@ -4074,7 +4063,8 @@ static void rdma_accept_incoming_migration(void *opaque)
     }
 }
 
-void rdma_start_incoming_migration(const char *host_port, Error **errp)
+void rdma_start_incoming_migration(InetSocketAddress *host_port,
+                                   Error **errp)
 {
     MigrationState *s = migrate_get_current();
     int ret;
@@ -4118,13 +4108,12 @@ cleanup_rdma:
 err:
     if (rdma) {
         g_free(rdma->host);
-        g_free(rdma->host_port);
     }
     g_free(rdma);
 }
 
 void rdma_start_outgoing_migration(void *opaque,
-                            const char *host_port, Error **errp)
+                            InetSocketAddress *host_port, Error **errp)
 {
     MigrationState *s = opaque;
     RDMAContext *rdma_return_path = NULL;

migration/rdma.h

@@ -14,15 +14,17 @@
  *
  */
 
+#include "qemu/sockets.h"
+
 #ifndef QEMU_MIGRATION_RDMA_H
 #define QEMU_MIGRATION_RDMA_H
 
 #include "exec/memory.h"
 
-void rdma_start_outgoing_migration(void *opaque, const char *host_port,
+void rdma_start_outgoing_migration(void *opaque, InetSocketAddress *host_port,
                                    Error **errp);
 
-void rdma_start_incoming_migration(const char *host_port, Error **errp);
+void rdma_start_incoming_migration(InetSocketAddress *host_port, Error **errp);
 
 /*
  * Constants used by rdma return codes

migration/savevm.c

@@ -237,6 +237,8 @@ static SaveState savevm_state = {
     .global_section_id = 0,
 };
 
+static SaveStateEntry *find_se(const char *idstr, uint32_t instance_id);
+
 static bool should_validate_capability(int capability)
 {
     assert(capability >= 0 && capability < MIGRATION_CAPABILITY__MAX);
@@ -716,6 +718,18 @@ static void savevm_state_handler_insert(SaveStateEntry *nse)
 
     assert(priority <= MIG_PRI_MAX);
 
+    /*
+     * This should never happen otherwise migration will probably fail
+     * silently somewhere because we can be wrongly applying one
+     * object properties upon another one.  Bail out ASAP.
+     */
+    if (find_se(nse->idstr, nse->instance_id)) {
+        error_report("%s: Detected duplicate SaveStateEntry: "
+                     "id=%s, instance_id=0x%"PRIx32, __func__,
+                     nse->idstr, nse->instance_id);
+        exit(EXIT_FAILURE);
+    }
+
     for (i = priority - 1; i >= 0; i--) {
         se = savevm_state.handler_pri_head[i];
         if (se != NULL) {
@@ -846,6 +860,24 @@ static void vmstate_check(const VMStateDescription *vmsd)
     }
 }
 
+/*
+ * See comment in hw/intc/xics.c:icp_realize()
+ *
+ * This function can be removed when
+ * pre_2_10_vmstate_register_dummy_icp() is removed.
+ */
+int vmstate_replace_hack_for_ppc(VMStateIf *obj, int instance_id,
+                                 const VMStateDescription *vmsd,
+                                 void *opaque)
+{
+    SaveStateEntry *se = find_se(vmsd->name, instance_id);
+
+    if (se) {
+        savevm_state_handler_remove(se);
+    }
+    return vmstate_register(obj, instance_id, vmsd, opaque);
+}
+
 int vmstate_register_with_alias_id(VMStateIf *obj, uint32_t instance_id,
                                    const VMStateDescription *vmsd,
                                    void *opaque, int alias_id,
@@ -1459,6 +1491,7 @@ void qemu_savevm_state_complete_postcopy(QEMUFile *f)
 static
 int qemu_savevm_state_complete_precopy_iterable(QEMUFile *f, bool in_postcopy)
 {
+    int64_t start_ts_each, end_ts_each;
     SaveStateEntry *se;
     int ret;
 
@@ -1475,6 +1508,8 @@ int qemu_savevm_state_complete_precopy_iterable(QEMUFile *f, bool in_postcopy)
                 continue;
             }
         }
+
+        start_ts_each = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
         trace_savevm_section_start(se->idstr, se->section_id);
 
         save_section_header(f, se, QEMU_VM_SECTION_END);
@@ -1486,8 +1521,13 @@ int qemu_savevm_state_complete_precopy_iterable(QEMUFile *f, bool in_postcopy)
             qemu_file_set_error(f, ret);
             return -1;
         }
+        end_ts_each = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+        trace_vmstate_downtime_save("iterable", se->idstr, se->instance_id,
+                                    end_ts_each - start_ts_each);
     }
 
+    trace_vmstate_downtime_checkpoint("src-iterable-saved");
+
     return 0;
 }
 
@@ -1496,6 +1536,7 @@ int qemu_savevm_state_complete_precopy_non_iterable(QEMUFile *f,
                                                     bool inactivate_disks)
 {
     MigrationState *ms = migrate_get_current();
+    int64_t start_ts_each, end_ts_each;
     JSONWriter *vmdesc = ms->vmdesc;
     int vmdesc_len;
     SaveStateEntry *se;
@@ -1507,11 +1548,17 @@ int qemu_savevm_state_complete_precopy_non_iterable(QEMUFile *f,
             continue;
         }
 
+        start_ts_each = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+
         ret = vmstate_save(f, se, vmdesc);
         if (ret) {
             qemu_file_set_error(f, ret);
             return ret;
         }
+
+        end_ts_each = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+        trace_vmstate_downtime_save("non-iterable", se->idstr, se->instance_id,
+                                    end_ts_each - start_ts_each);
     }
 
     if (inactivate_disks) {
@@ -1547,6 +1594,8 @@ int qemu_savevm_state_complete_precopy_non_iterable(QEMUFile *f,
     json_writer_free(vmdesc);
     ms->vmdesc = NULL;
 
+    trace_vmstate_downtime_checkpoint("src-non-iterable-saved");
+
     return 0;
 }
 
@@ -2088,18 +2137,18 @@ static void loadvm_postcopy_handle_run_bh(void *opaque)
     Error *local_err = NULL;
     MigrationIncomingState *mis = opaque;
 
-    trace_loadvm_postcopy_handle_run_bh("enter");
+    trace_vmstate_downtime_checkpoint("dst-postcopy-bh-enter");
 
     /* TODO we should move all of this lot into postcopy_ram.c or a shared code
      * in migration.c
      */
     cpu_synchronize_all_post_init();
 
-    trace_loadvm_postcopy_handle_run_bh("after cpu sync");
+    trace_vmstate_downtime_checkpoint("dst-postcopy-bh-cpu-synced");
 
     qemu_announce_self(&mis->announce_timer, migrate_announce_params());
 
-    trace_loadvm_postcopy_handle_run_bh("after announce");
+    trace_vmstate_downtime_checkpoint("dst-postcopy-bh-announced");
 
     /* Make sure all file formats throw away their mutable metadata.
      * If we get an error here, just don't restart the VM yet. */
@@ -2110,7 +2159,7 @@ static void loadvm_postcopy_handle_run_bh(void *opaque)
         autostart = false;
     }
 
-    trace_loadvm_postcopy_handle_run_bh("after invalidate cache");
+    trace_vmstate_downtime_checkpoint("dst-postcopy-bh-cache-invalidated");
 
     dirty_bitmap_mig_before_vm_start();
 
@@ -2124,7 +2173,7 @@ static void loadvm_postcopy_handle_run_bh(void *opaque)
 
     qemu_bh_delete(mis->bh);
 
-    trace_loadvm_postcopy_handle_run_bh("return");
+    trace_vmstate_downtime_checkpoint("dst-postcopy-bh-vm-started");
 }
 
 /* After all discards we can start running and asking for pages */
@@ -2505,9 +2554,12 @@ static bool check_section_footer(QEMUFile *f, SaveStateEntry *se)
 }
 
 static int
-qemu_loadvm_section_start_full(QEMUFile *f, MigrationIncomingState *mis)
+qemu_loadvm_section_start_full(QEMUFile *f, MigrationIncomingState *mis,
+                               uint8_t type)
 {
+    bool trace_downtime = (type == QEMU_VM_SECTION_FULL);
     uint32_t instance_id, version_id, section_id;
+    int64_t start_ts, end_ts;
     SaveStateEntry *se;
     char idstr[256];
     int ret;
@@ -2556,12 +2608,23 @@ qemu_loadvm_section_start_full(QEMUFile *f, MigrationIncomingState *mis)
         return -EINVAL;
     }
 
+    if (trace_downtime) {
+        start_ts = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+    }
+
     ret = vmstate_load(f, se);
     if (ret < 0) {
         error_report("error while loading state for instance 0x%"PRIx32" of"
                      " device '%s'", instance_id, idstr);
         return ret;
     }
+
+    if (trace_downtime) {
+        end_ts = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+        trace_vmstate_downtime_load("non-iterable", se->idstr,
+                                    se->instance_id, end_ts - start_ts);
+    }
+
     if (!check_section_footer(f, se)) {
         return -EINVAL;
     }
@@ -2570,8 +2633,11 @@ qemu_loadvm_section_start_full(QEMUFile *f, MigrationIncomingState *mis)
 }
 
 static int
-qemu_loadvm_section_part_end(QEMUFile *f, MigrationIncomingState *mis)
+qemu_loadvm_section_part_end(QEMUFile *f, MigrationIncomingState *mis,
+                             uint8_t type)
 {
+    bool trace_downtime = (type == QEMU_VM_SECTION_END);
+    int64_t start_ts, end_ts;
     uint32_t section_id;
     SaveStateEntry *se;
     int ret;
@@ -2596,12 +2662,23 @@ qemu_loadvm_section_part_end(QEMUFile *f, MigrationIncomingState *mis)
         return -EINVAL;
     }
 
+    if (trace_downtime) {
+        start_ts = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+    }
+
     ret = vmstate_load(f, se);
     if (ret < 0) {
         error_report("error while loading state section id %d(%s)",
                      section_id, se->idstr);
         return ret;
     }
+
+    if (trace_downtime) {
+        end_ts = qemu_clock_get_us(QEMU_CLOCK_REALTIME);
+        trace_vmstate_downtime_load("iterable", se->idstr,
+                                    se->instance_id, end_ts - start_ts);
+    }
+
     if (!check_section_footer(f, se)) {
         return -EINVAL;
     }
@@ -2790,14 +2867,14 @@ retry:
         switch (section_type) {
         case QEMU_VM_SECTION_START:
         case QEMU_VM_SECTION_FULL:
-            ret = qemu_loadvm_section_start_full(f, mis);
+            ret = qemu_loadvm_section_start_full(f, mis, section_type);
             if (ret < 0) {
                 goto out;
             }
             break;
         case QEMU_VM_SECTION_PART:
         case QEMU_VM_SECTION_END:
-            ret = qemu_loadvm_section_part_end(f, mis);
+            ret = qemu_loadvm_section_part_end(f, mis, section_type);
             if (ret < 0) {
                 goto out;
             }

migration/socket.c

@@ -28,6 +28,8 @@
 #include "trace.h"
 #include "postcopy-ram.h"
 #include "options.h"
+#include "qapi/clone-visitor.h"
+#include "qapi/qapi-visit-sockets.h"
 
 struct SocketOutgoingArgs {
     SocketAddress *saddr;
@@ -108,19 +110,19 @@ out:
     object_unref(OBJECT(sioc));
 }
 
-static void
-socket_start_outgoing_migration_internal(MigrationState *s,
-                                         SocketAddress *saddr,
-                                         Error **errp)
+void socket_start_outgoing_migration(MigrationState *s,
+                                     SocketAddress *saddr,
+                                     Error **errp)
 {
     QIOChannelSocket *sioc = qio_channel_socket_new();
     struct SocketConnectData *data = g_new0(struct SocketConnectData, 1);
+    SocketAddress *addr = QAPI_CLONE(SocketAddress, saddr);
 
     data->s = s;
 
     /* in case previous migration leaked it */
     qapi_free_SocketAddress(outgoing_args.saddr);
-    outgoing_args.saddr = saddr;
+    outgoing_args.saddr = addr;
 
     if (saddr->type == SOCKET_ADDRESS_TYPE_INET) {
         data->hostname = g_strdup(saddr->u.inet.host);
@@ -135,18 +137,6 @@ socket_start_outgoing_migration_internal(MigrationState *s,
                                      NULL);
 }
 
-void socket_start_outgoing_migration(MigrationState *s,
-                                     const char *str,
-                                     Error **errp)
-{
-    Error *err = NULL;
-    SocketAddress *saddr = socket_parse(str, &err);
-    if (!err) {
-        socket_start_outgoing_migration_internal(s, saddr, &err);
-    }
-    error_propagate(errp, err);
-}
-
 static void socket_accept_incoming_migration(QIONetListener *listener,
                                              QIOChannelSocket *cioc,
                                              gpointer opaque)
@@ -172,9 +162,8 @@ socket_incoming_migration_end(void *opaque)
     object_unref(OBJECT(listener));
 }
 
-static void
-socket_start_incoming_migration_internal(SocketAddress *saddr,
-                                         Error **errp)
+void socket_start_incoming_migration(SocketAddress *saddr,
+                                     Error **errp)
 {
     QIONetListener *listener = qio_net_listener_new();
     MigrationIncomingState *mis = migration_incoming_get_current();
@@ -213,13 +202,3 @@ socket_start_incoming_migration_internal(SocketAddress *saddr,
     }
 }
 
-void socket_start_incoming_migration(const char *str, Error **errp)
-{
-    Error *err = NULL;
-    SocketAddress *saddr = socket_parse(str, &err);
-    if (!err) {
-        socket_start_incoming_migration_internal(saddr, &err);
-    }
-    qapi_free_SocketAddress(saddr);
-    error_propagate(errp, err);
-}

migration/socket.h

@@ -19,13 +19,14 @@
 
 #include "io/channel.h"
 #include "io/task.h"
+#include "qemu/sockets.h"
 
 void socket_send_channel_create(QIOTaskFunc f, void *data);
 QIOChannel *socket_send_channel_create_sync(Error **errp);
 int socket_send_channel_destroy(QIOChannel *send);
 
-void socket_start_incoming_migration(const char *str, Error **errp);
+void socket_start_incoming_migration(SocketAddress *saddr, Error **errp);
 
-void socket_start_outgoing_migration(MigrationState *s, const char *str,
-                                     Error **errp);
+void socket_start_outgoing_migration(MigrationState *s,
+                                     SocketAddress *saddr, Error **errp);
 #endif

migration/trace-events

@@ -17,7 +17,6 @@ loadvm_handle_recv_bitmap(char *s) "%s"
 loadvm_postcopy_handle_advise(void) ""
 loadvm_postcopy_handle_listen(const char *str) "%s"
 loadvm_postcopy_handle_run(void) ""
-loadvm_postcopy_handle_run_bh(const char *str) "%s"
 loadvm_postcopy_handle_resume(void) ""
 loadvm_postcopy_ram_handle_discard(void) ""
 loadvm_postcopy_ram_handle_discard_end(void) ""
@@ -48,6 +47,9 @@ savevm_state_cleanup(void) ""
 savevm_state_complete_precopy(void) ""
 vmstate_save(const char *idstr, const char *vmsd_name) "%s, %s"
 vmstate_load(const char *idstr, const char *vmsd_name) "%s, %s"
+vmstate_downtime_save(const char *type, const char *idstr, uint32_t instance_id, int64_t downtime) "type=%s idstr=%s instance_id=%d downtime=%"PRIi64
+vmstate_downtime_load(const char *type, const char *idstr, uint32_t instance_id, int64_t downtime) "type=%s idstr=%s instance_id=%d downtime=%"PRIi64
+vmstate_downtime_checkpoint(const char *checkpoint) "%s"
 postcopy_pause_incoming(void) ""
 postcopy_pause_incoming_continued(void) ""
 postcopy_page_req_sync(void *host_addr) "sync page req %p"
@@ -148,8 +150,6 @@ multifd_tls_outgoing_handshake_complete(void *ioc) "ioc=%p"
 multifd_set_outgoing_channel(void *ioc, const char *ioctype, const char *hostname)  "ioc=%p ioctype=%s hostname=%s"
 
 # migration.c
-await_return_path_close_on_source_close(void) ""
-await_return_path_close_on_source_joining(void) ""
 migrate_set_state(const char *new_state) "new state %s"
 migrate_fd_cleanup(void) ""
 migrate_fd_error(const char *error_desc) "error=%s"
@@ -166,7 +166,7 @@ migration_completion_postcopy_end_after_complete(void) ""
 migration_rate_limit_pre(int ms) "%d ms"
 migration_rate_limit_post(int urgent) "urgent: %d"
 migration_return_path_end_before(void) ""
-migration_return_path_end_after(int rp_error) "%d"
+migration_return_path_end_after(void) ""
 migration_thread_after_loop(void) ""
 migration_thread_file_err(void) ""
 migration_thread_setup_complete(void) ""

net/slirp.c

@@ -46,6 +46,7 @@
 #include "qapi/qmp/qdict.h"
 #include "util.h"
 #include "migration/register.h"
+#include "migration/vmstate.h"
 #include "migration/qemu-file-types.h"
 
 static int get_str_sep(char *buf, int buf_size, const char **pp, int sep)
@@ -659,8 +660,8 @@ static int net_slirp_init(NetClientState *peer, const char *model,
      * specific version?
      */
     g_assert(slirp_state_version() == 4);
-    register_savevm_live("slirp", 0, slirp_state_version(),
-                         &savevm_slirp_state, s->slirp);
+    register_savevm_live("slirp", VMSTATE_INSTANCE_ID_ANY,
+                         slirp_state_version(), &savevm_slirp_state, s->slirp);
 
     s->poll_notifier.notify = net_slirp_poll_notify;
     main_loop_poll_add_notifier(&s->poll_notifier);

qapi/migration.json

@@ -631,6 +631,28 @@
   'data': [ 'none', 'zlib',
             { 'name': 'zstd', 'if': 'CONFIG_ZSTD' } ] }
 
+##
+# @MigMode:
+#
+# @normal: the original form of migration. (since 8.2)
+#
+# @cpr-reboot: The migrate command saves state to a file, allowing one to
+#              quit qemu, reboot to an updated kernel, and restart an updated
+#              version of qemu.  The caller must specify a migration URI
+#              that writes to and reads from a file.  Unlike normal mode,
+#              the use of certain local storage options does not block the
+#              migration, but the caller must not modify guest block devices
+#              between the quit and restart.  To avoid saving guest RAM to the
+#              file, the memory backend must be shared, and the @x-ignore-shared
+#              migration capability must be set.  Guest RAM must be non-volatile
+#              across reboot, such as by backing it with a dax device, but this
+#              is not enforced.  The restarted qemu arguments must match those
+#              used to initially start qemu, plus the -incoming option.
+#              (since 8.2)
+##
+{ 'enum': 'MigMode',
+  'data': [ 'normal', 'cpr-reboot' ] }
+
 ##
 # @BitmapMigrationBitmapAliasTransform:
 #
@@ -849,6 +871,9 @@
 # @vcpu-dirty-limit: Dirtyrate limit (MB/s) during live migration.
 #     Defaults to 1.  (Since 8.1)
 #
+# @mode: Migration mode. See description in @MigMode. Default is 'normal'.
+#        (Since 8.2)
+#
 # Features:
 #
 # @deprecated: Member @block-incremental is deprecated.  Use
@@ -881,7 +906,8 @@
            'multifd-zlib-level', 'multifd-zstd-level',
            'block-bitmap-mapping',
            { 'name': 'x-vcpu-dirty-limit-period', 'features': ['unstable'] },
-           'vcpu-dirty-limit'] }
+           'vcpu-dirty-limit',
+           'mode'] }
 
 ##
 # @MigrateSetParameters:
@@ -1033,6 +1059,9 @@
 # @vcpu-dirty-limit: Dirtyrate limit (MB/s) during live migration.
 #     Defaults to 1.  (Since 8.1)
 #
+# @mode: Migration mode. See description in @MigMode. Default is 'normal'.
+#        (Since 8.2)
+#
 # Features:
 #
 # @deprecated: Member @block-incremental is deprecated.  Use
@@ -1085,7 +1114,8 @@
             '*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ],
             '*x-vcpu-dirty-limit-period': { 'type': 'uint64',
                                             'features': [ 'unstable' ] },
-            '*vcpu-dirty-limit': 'uint64'} }
+            '*vcpu-dirty-limit': 'uint64',
+            '*mode': 'MigMode'} }
 
 ##
 # @migrate-set-parameters:
@@ -1257,6 +1287,9 @@
 # @vcpu-dirty-limit: Dirtyrate limit (MB/s) during live migration.
 #     Defaults to 1.  (Since 8.1)
 #
+# @mode: Migration mode. See description in @MigMode. Default is 'normal'.
+#        (Since 8.2)
+#
 # Features:
 #
 # @deprecated: Member @block-incremental is deprecated.  Use
@@ -1306,7 +1339,8 @@
             '*block-bitmap-mapping': [ 'BitmapMigrationNodeAlias' ],
             '*x-vcpu-dirty-limit-period': { 'type': 'uint64',
                                             'features': [ 'unstable' ] },
-            '*vcpu-dirty-limit': 'uint64'} }
+            '*vcpu-dirty-limit': 'uint64',
+            '*mode': 'MigMode'} }
 
 ##
 # @query-migrate-parameters:
@@ -1550,6 +1584,91 @@
 ##
 { 'command': 'migrate-continue', 'data': {'state': 'MigrationStatus'} }
 
+##
+# @MigrationAddressType:
+#
+# The migration stream transport mechanisms.
+#
+# @socket: Migrate via socket.
+#
+# @exec: Direct the migration stream to another process.
+#
+# @rdma: Migrate via RDMA.
+#
+# @file: Direct the migration stream to a file.
+#
+# Since 8.2
+##
+{ 'enum': 'MigrationAddressType',
+  'data': [ 'socket', 'exec', 'rdma', 'file' ] }
+
+##
+# @FileMigrationArgs:
+#
+# @filename: The file to receive the migration stream
+#
+# @offset: The file offset where the migration stream will start
+#
+# Since 8.2
+##
+{ 'struct': 'FileMigrationArgs',
+  'data': { 'filename': 'str',
+            'offset': 'uint64' } }
+
+##
+# @MigrationExecCommand:
+#
+# @args: command (list head) and arguments to execute.
+#
+# Since 8.2
+##
+{ 'struct': 'MigrationExecCommand',
+  'data': {'args': [ 'str' ] } }
+
+##
+# @MigrationAddress:
+#
+# Migration endpoint configuration.
+#
+# Since 8.2
+##
+{ 'union': 'MigrationAddress',
+  'base': { 'transport' : 'MigrationAddressType'},
+  'discriminator': 'transport',
+  'data': {
+    'socket': 'SocketAddress',
+    'exec': 'MigrationExecCommand',
+    'rdma': 'InetSocketAddress',
+    'file': 'FileMigrationArgs' } }
+
+##
+# @MigrationChannelType:
+#
+# The migration channel-type request options.
+#
+# @main: Main outbound migration channel.
+#
+# Since 8.1
+##
+{ 'enum': 'MigrationChannelType',
+  'data': [ 'main' ] }
+
+##
+# @MigrationChannel:
+#
+# Migration stream channel parameters.
+#
+# @channel-type: Channel type for transfering packet information.
+#
+# @addr: Migration endpoint configuration on destination interface.
+#
+# Since 8.1
+##
+{ 'struct': 'MigrationChannel',
+  'data': {
+      'channel-type': 'MigrationChannelType',
+      'addr': 'MigrationAddress' } }
+
 ##
 # @migrate:
 #
@@ -1557,6 +1676,9 @@
 #
 # @uri: the Uniform Resource Identifier of the destination VM
 #
+# @channels: list of migration stream channels with each stream in the
+#     list connected to a destination interface endpoint.
+#
 # @blk: do block migration (full disk copy)
 #
 # @inc: incremental disk copy migration
@@ -1586,13 +1708,57 @@
 # 3. The user Monitor's "detach" argument is invalid in QMP and should
 #    not be used
 #
+# 4. The uri argument should have the Uniform Resource Identifier of
+#    default destination VM. This connection will be bound to default
+#    network.
+#
+# 5. For now, number of migration streams is restricted to one, i.e
+#    number of items in 'channels' list is just 1.
+#
+# 6. The 'uri' and 'channels' arguments are mutually exclusive;
+#    exactly one of the two should be present.
+#
 # Example:
 #
 # -> { "execute": "migrate", "arguments": { "uri": "tcp:0:4446" } }
 # <- { "return": {} }
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "socket",
+#                                    "type": "inet",
+#                                    "host": "10.12.34.9",
+#                                    "port": "1050" } } ] } }
+# <- { "return": {} }
+#
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "exec",
+#                                    "args": [ "/bin/nc", "-p", "6000",
+#                                              "/some/sock" ] } } ] } }
+# <- { "return": {} }
+#
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "rdma",
+#                                    "host": "10.12.34.9",
+#                                    "port": "1050" } } ] } }
+# <- { "return": {} }
+#
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "file",
+#                                    "filename": "/tmp/migfile",
+#                                    "offset": "0x1000" } } ] } }
+# <- { "return": {} }
+#
 ##
 { 'command': 'migrate',
   'data': {'uri': 'str',
+           '*channels': [ 'MigrationChannel' ],
            '*blk': { 'type': 'bool', 'features': [ 'deprecated' ] },
            '*inc': { 'type': 'bool', 'features': [ 'deprecated' ] },
            '*detach': 'bool', '*resume': 'bool' } }
@@ -1606,6 +1772,9 @@
 # @uri: The Uniform Resource Identifier identifying the source or
 #     address to listen on
 #
+# @channels: list of migration stream channels with each stream in the
+#     list connected to a destination interface endpoint.
+#
 # Returns: nothing on success
 #
 # Since: 2.3
@@ -1621,13 +1790,46 @@
 #
 # 3. The uri format is the same as for -incoming
 #
+# 5. For now, number of migration streams is restricted to one, i.e
+#    number of items in 'channels' list is just 1.
+#
+# 4. The 'uri' and 'channels' arguments are mutually exclusive;
+#    exactly one of the two should be present.
+#
 # Example:
 #
 # -> { "execute": "migrate-incoming",
 #      "arguments": { "uri": "tcp::4446" } }
 # <- { "return": {} }
+#
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "socket",
+#                                    "type": "inet",
+#                                    "host": "10.12.34.9",
+#                                    "port": "1050" } } ] } }
+# <- { "return": {} }
+#
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "exec",
+#                                    "args": [ "/bin/nc", "-p", "6000",
+#                                              "/some/sock" ] } } ] } }
+# <- { "return": {} }
+#
+# -> { "execute": "migrate",
+#      "arguments": {
+#          "channels": [ { "channel-type": "main",
+#                          "addr": { "transport": "rdma",
+#                                    "host": "10.12.34.9",
+#                                    "port": "1050" } } ] } }
+# <- { "return": {} }
 ##
-{ 'command': 'migrate-incoming', 'data': {'uri': 'str' } }
+{ 'command': 'migrate-incoming',
+             'data': {'*uri': 'str',
+                      '*channels': [ 'MigrationChannel' ] } }
 
 ##
 # @xen-save-devices-state:

stubs/migr-blocker.c

@@ -6,6 +6,16 @@ int migrate_add_blocker(Error **reasonp, Error **errp)
     return 0;
 }
 
+int migrate_add_blocker_normal(Error **reasonp, Error **errp)
+{
+    return 0;
+}
+
+int migrate_add_blocker_modes(Error **reasonp, Error **errp, MigMode mode, ...)
+{
+    return 0;
+}
+
 void migrate_del_blocker(Error **reasonp)
 {
 }

system/vl.c

@@ -2702,7 +2702,7 @@ void qmp_x_exit_preconfig(Error **errp)
     if (incoming) {
         Error *local_err = NULL;
         if (strcmp(incoming, "defer") != 0) {
-            qmp_migrate_incoming(incoming, &local_err);
+            qmp_migrate_incoming(incoming, false, NULL, &local_err);
             if (local_err) {
                 error_reportf_err(local_err, "-incoming %s: ", incoming);
                 exit(1);

tests/qtest/migration-test.c

@@ -728,6 +728,7 @@ typedef struct {
     /* Postcopy specific fields */
     void *postcopy_data;
     bool postcopy_preempt;
+    bool postcopy_recovery_test_fail;
 } MigrateCommon;
 
 static int test_migrate_start(QTestState **from, QTestState **to,
@@ -1309,7 +1310,12 @@ static int migrate_postcopy_prepare(QTestState **from_ptr,
 
     migrate_prepare_for_dirty_mem(from);
     qtest_qmp_assert_success(to, "{ 'execute': 'migrate-incoming',"
-                             "  'arguments': { 'uri': 'tcp:127.0.0.1:0' }}");
+                             "  'arguments': { "
+                             "      'channels': [ { 'channel-type': 'main',"
+                             "      'addr': { 'transport': 'socket',"
+                             "                'type': 'inet',"
+                             "                'host': '127.0.0.1',"
+                             "                'port': '0' } } ] } }");
 
     /* Wait for the first serial output from the source */
     wait_for_serial("src_serial");
@@ -1404,6 +1410,80 @@ static void test_postcopy_preempt_tls_psk(void)
 }
 #endif
 
+static void wait_for_postcopy_status(QTestState *one, const char *status)
+{
+    wait_for_migration_status(one, status,
+                              (const char * []) { "failed", "active",
+                                                  "completed", NULL });
+}
+
+#ifndef _WIN32
+static void postcopy_recover_fail(QTestState *from, QTestState *to)
+{
+    int ret, pair1[2], pair2[2];
+    char c;
+
+    /* Create two unrelated socketpairs */
+    ret = qemu_socketpair(PF_LOCAL, SOCK_STREAM, 0, pair1);
+    g_assert_cmpint(ret, ==, 0);
+
+    ret = qemu_socketpair(PF_LOCAL, SOCK_STREAM, 0, pair2);
+    g_assert_cmpint(ret, ==, 0);
+
+    /*
+     * Give the guests unpaired ends of the sockets, so they'll all blocked
+     * at reading.  This mimics a wrong channel established.
+     */
+    qtest_qmp_fds_assert_success(from, &pair1[0], 1,
+                                 "{ 'execute': 'getfd',"
+                                 "  'arguments': { 'fdname': 'fd-mig' }}");
+    qtest_qmp_fds_assert_success(to, &pair2[0], 1,
+                                 "{ 'execute': 'getfd',"
+                                 "  'arguments': { 'fdname': 'fd-mig' }}");
+
+    /*
+     * Write the 1st byte as QEMU_VM_COMMAND (0x8) for the dest socket, to
+     * emulate the 1st byte of a real recovery, but stops from there to
+     * keep dest QEMU in RECOVER.  This is needed so that we can kick off
+     * the recover process on dest QEMU (by triggering the G_IO_IN event).
+     *
+     * NOTE: this trick is not needed on src QEMUs, because src doesn't
+     * rely on an pre-existing G_IO_IN event, so it will always trigger the
+     * upcoming recovery anyway even if it can read nothing.
+     */
+#define QEMU_VM_COMMAND              0x08
+    c = QEMU_VM_COMMAND;
+    ret = send(pair2[1], &c, 1, 0);
+    g_assert_cmpint(ret, ==, 1);
+
+    migrate_recover(to, "fd:fd-mig");
+    migrate_qmp(from, "fd:fd-mig", "{'resume': true}");
+
+    /*
+     * Make sure both QEMU instances will go into RECOVER stage, then test
+     * kicking them out using migrate-pause.
+     */
+    wait_for_postcopy_status(from, "postcopy-recover");
+    wait_for_postcopy_status(to, "postcopy-recover");
+
+    /*
+     * This would be issued by the admin upon noticing the hang, we should
+     * make sure we're able to kick this out.
+     */
+    migrate_pause(from);
+    wait_for_postcopy_status(from, "postcopy-paused");
+
+    /* Do the same test on dest */
+    migrate_pause(to);
+    wait_for_postcopy_status(to, "postcopy-paused");
+
+    close(pair1[0]);
+    close(pair1[1]);
+    close(pair2[0]);
+    close(pair2[1]);
+}
+#endif /* _WIN32 */
+
 static void test_postcopy_recovery_common(MigrateCommon *args)
 {
     QTestState *from, *to;
@@ -1439,9 +1519,19 @@ static void test_postcopy_recovery_common(MigrateCommon *args)
      * migrate-recover command can only succeed if destination machine
      * is in the paused state
      */
-    wait_for_migration_status(to, "postcopy-paused",
-                              (const char * []) { "failed", "active",
-                                                  "completed", NULL });
+    wait_for_postcopy_status(to, "postcopy-paused");
+    wait_for_postcopy_status(from, "postcopy-paused");
+
+#ifndef _WIN32
+    if (args->postcopy_recovery_test_fail) {
+        /*
+         * Test when a wrong socket specified for recover, and then the
+         * ability to kick it out, and continue with a correct socket.
+         */
+        postcopy_recover_fail(from, to);
+        /* continue with a good recovery */
+    }
+#endif /* _WIN32 */
 
     /*
      * Create a new socket to emulate a new channel that is different
@@ -1455,9 +1545,6 @@ static void test_postcopy_recovery_common(MigrateCommon *args)
      * Try to rebuild the migration channel using the resume flag and
      * the newly created channel
      */
-    wait_for_migration_status(from, "postcopy-paused",
-                              (const char * []) { "failed", "active",
-                                                  "completed", NULL });
     migrate_qmp(from, uri, "{'resume': true}");
 
     /* Restore the postcopy bandwidth to unlimited */
@@ -1482,6 +1569,17 @@ static void test_postcopy_recovery_compress(void)
     test_postcopy_recovery_common(&args);
 }
 
+#ifndef _WIN32
+static void test_postcopy_recovery_double_fail(void)
+{
+    MigrateCommon args = {
+        .postcopy_recovery_test_fail = true,
+    };
+
+    test_postcopy_recovery_common(&args);
+}
+#endif /* _WIN32 */
+
 #ifdef CONFIG_GNUTLS
 static void test_postcopy_recovery_tls_psk(void)
 {
@@ -2026,6 +2124,31 @@ static void test_precopy_file_offset_bad(void)
     test_file_common(&args, false);
 }
 
+static void *test_mode_reboot_start(QTestState *from, QTestState *to)
+{
+    migrate_set_parameter_str(from, "mode", "cpr-reboot");
+    migrate_set_parameter_str(to, "mode", "cpr-reboot");
+
+    migrate_set_capability(from, "x-ignore-shared", true);
+    migrate_set_capability(to, "x-ignore-shared", true);
+
+    return NULL;
+}
+
+static void test_mode_reboot(void)
+{
+    g_autofree char *uri = g_strdup_printf("file:%s/%s", tmpfs,
+                                           FILE_TEST_FILENAME);
+    MigrateCommon args = {
+        .start.use_shmem = true,
+        .connect_uri = uri,
+        .listen_uri = "defer",
+        .start_hook = test_mode_reboot_start
+    };
+
+    test_file_common(&args, true);
+}
+
 static void test_precopy_tcp_plain(void)
 {
     MigrateCommon args = {
@@ -3068,6 +3191,11 @@ int main(int argc, char **argv)
             qtest_add_func("/migration/postcopy/recovery/compress/plain",
                            test_postcopy_recovery_compress);
         }
+#ifndef _WIN32
+        qtest_add_func("/migration/postcopy/recovery/double-failures",
+                       test_postcopy_recovery_double_fail);
+#endif /* _WIN32 */
+
     }
 
     qtest_add_func("/migration/bad_dest", test_baddest);
@@ -3096,6 +3224,14 @@ int main(int argc, char **argv)
     qtest_add_func("/migration/precopy/file/offset/bad",
                    test_precopy_file_offset_bad);
 
+    /*
+     * Our CI system has problems with shared memory.
+     * Don't run this test until we find a workaround.
+     */
+    if (getenv("QEMU_TEST_FLAKY_TESTS")) {
+        qtest_add_func("/migration/mode/reboot", test_mode_reboot);
+    }
+
 #ifdef CONFIG_GNUTLS
     qtest_add_func("/migration/precopy/unix/tls/psk",
                    test_precopy_unix_tls_psk);