qemu/hw/core
Markus Armbruster b69c3c21a5 qdev: Unrealize must not fail
Devices may have component devices and buses.

Device realization may fail.  Realization is recursive: a device's
realize() method realizes its components, and device_set_realized()
realizes its buses (which should in turn realize the devices on that
bus, except bus_set_realized() doesn't implement that, yet).

When realization of a component or bus fails, we need to roll back:
unrealize everything we realized so far.  If any of these unrealizes
failed, the device would be left in an inconsistent state.  Must not
happen.

device_set_realized() lets it happen: it ignores errors in the roll
back code starting at label child_realize_fail.

Since realization is recursive, unrealization must be recursive, too.
But how could a partly failed unrealize be rolled back?  We'd have to
re-realize, which can fail.  This design is fundamentally broken.

device_set_realized() does not roll back at all.  Instead, it keeps
unrealizing, ignoring further errors.

It can screw up even for a device with no buses: if the lone
dc->unrealize() fails, it still unregisters vmstate, and calls
listeners' unrealize() callback.

bus_set_realized() does not roll back either.  Instead, it stops
unrealizing.

Fortunately, no unrealize method can fail, as we'll see below.

To fix the design error, drop parameter @errp from all the unrealize
methods.

Any unrealize method that uses @errp now needs an update.  This leads
us to unrealize() methods that can fail.  Merely passing it to another
unrealize method cannot cause failure, though.  Here are the ones that
do other things with @errp:

* virtio_serial_device_unrealize()

  Fails when qbus_set_hotplug_handler() fails, but still does all the
  other work.  On failure, the device would stay realized with its
  resources completely gone.  Oops.  Can't happen, because
  qbus_set_hotplug_handler() can't actually fail here.  Pass
  &error_abort to qbus_set_hotplug_handler() instead.

* hw/ppc/spapr_drc.c's unrealize()

  Fails when object_property_del() fails, but all the other work is
  already done.  On failure, the device would stay realized with its
  vmstate registration gone.  Oops.  Can't happen, because
  object_property_del() can't actually fail here.  Pass &error_abort
  to object_property_del() instead.

* spapr_phb_unrealize()

  Fails and bails out when remove_drcs() fails, but other work is
  already done.  On failure, the device would stay realized with some
  of its resources gone.  Oops.  remove_drcs() fails only when
  chassis_from_bus()'s object_property_get_uint() fails, and it can't
  here.  Pass &error_abort to remove_drcs() instead.

Therefore, no unrealize method can fail before this patch.

device_set_realized()'s recursive unrealization via bus uses
object_property_set_bool().  Can't drop @errp there, so pass
&error_abort.

We similarly unrealize with object_property_set_bool() elsewhere,
always ignoring errors.  Pass &error_abort instead.

Several unrealize methods no longer handle errors from other unrealize
methods: virtio_9p_device_unrealize(),
virtio_input_device_unrealize(), scsi_qdev_unrealize(), ...
Much of the deleted error handling looks wrong anyway.

One unrealize methods no longer ignore such errors:
usb_ehci_pci_exit().

Several realize methods no longer ignore errors when rolling back:
v9fs_device_realize_common(), pci_qdev_unrealize(),
spapr_phb_realize(), usb_qdev_realize(), vfio_ccw_realize(),
virtio_device_realize().

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200505152926.18877-17-armbru@redhat.com>
2020-05-15 07:08:14 +02:00
..
bus.c qdev: Unrealize must not fail 2020-05-15 07:08:14 +02:00
clock-vmstate.c hw/core/clock-vmstate: define a vmstate entry for clock state 2020-04-30 15:35:40 +01:00
clock.c hw/core/clock: introduce clock object 2020-04-30 11:52:28 +01:00
cpu.c qdev: Unrealize must not fail 2020-05-15 07:08:14 +02:00
empty_slot.c Include hw/hw.h exactly where needed 2019-08-16 13:31:52 +02:00
fw-path-provider.c Include qemu/module.h where needed, drop it from qemu-common.h 2019-06-12 13:18:33 +02:00
generic-loader.c qdev: Unrealize must not fail 2020-05-15 07:08:14 +02:00
hotplug.c call HotplugHandler->plug() as the last step in device realization 2018-10-19 13:44:12 +02:00
irq.c Revert "irq: introduce qemu_irq_proxy()" 2019-11-05 23:33:12 +01:00
Kconfig hw/core: Add a config switch for the generic loader device 2019-08-20 09:11:17 +02:00
loader-fit.c hw/core: Fix fit_load_fdt() error API violations 2019-12-18 08:36:15 +01:00
loader.c hw/core: properly terminate loading .hex on EOF record 2020-04-07 16:19:49 +01:00
machine-hmp-cmds.c qapi: Only input visitors can actually fail 2020-04-30 07:26:40 +02:00
machine-qmp-cmds.c qom: Drop convenience method object_property_get_uint16List() 2020-05-15 06:26:02 +02:00
machine.c qom: Drop parameter @errp of object_property_add() & friends 2020-05-15 07:07:58 +02:00
Makefile.objs qdev: add clock input&output support to devices. 2020-04-30 15:35:40 +01:00
nmi.c Include qemu/module.h where needed, drop it from qemu-common.h 2019-06-12 13:18:33 +02:00
null-machine.c null-machine: use memdev for RAM 2020-02-19 16:49:57 +00:00
numa.c remove no longer used memory_region_allocate_system_memory() 2020-02-19 16:50:01 +00:00
or-irq.c hw/core/or-irq: Fix incorrect assert forbidding num-lines == MAX_OR_LINES 2020-01-30 16:02:01 +00:00
platform-bus.c hw/core: Let devices own the MemoryRegion they create 2020-03-17 15:18:48 +01:00
ptimer.c ptimer: Remove old ptimer_init_with_bh() API 2019-11-11 13:44:16 +00:00
qdev-clock.c qom: Drop parameter @errp of object_property_add() & friends 2020-05-15 07:07:58 +02:00
qdev-fw.c Include hw/qdev-properties.h less 2019-08-16 13:31:53 +02:00
qdev-properties-system.c audio: add audiodev properties to frontends 2019-08-21 09:13:37 +02:00
qdev-properties.c Drop more @errp parameters after previous commit 2020-05-15 07:08:14 +02:00
qdev.c qdev: Unrealize must not fail 2020-05-15 07:08:14 +02:00
register.c hw/core/register: Add register_init_block8 helper. 2020-05-05 13:37:51 -07:00
reset.c qemu/queue.h: leave head structs anonymous unless necessary 2019-01-11 15:46:55 +01:00
resettable.c hw/core: deprecate old reset functions and introduce new ones 2020-01-30 16:02:04 +00:00
split-irq.c qdev: set properties with device_class_set_props() 2020-01-24 20:59:15 +01:00
stream.c hw/core: stream: Add an end-of-packet flag 2020-05-14 13:44:35 +02:00
sysbus.c sysbus: remove unused sysbus_try_create* 2020-01-07 16:06:59 +04:00
trace-events hw/core/clock: introduce clock object 2020-04-30 11:52:28 +01:00
uboot_image.h Support u-boot noload images for arm as used by, NetBSD/evbarm GENERIC kernel. 2019-01-07 15:46:20 +00:00
vm-change-state-handler.c sysemu: Split sysemu/runstate.h off sysemu/sysemu.h 2019-08-16 13:37:36 +02:00
vmstate-if.c vmstate: add qom interface to get id 2020-01-06 18:41:32 +04:00