Merge tag 'migration-20231102-pull-request' of https://gitlab.com/jua…

…n.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)