Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200610053247.1583243-58-armbru@redhat.com>
So far, qdev_realize() supports only devices that plug into a bus:
argument @bus cannot be null. Extend it to support bus-less devices,
too.
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: <20200610053247.1583243-55-armbru@redhat.com>
The "null @bus means main system bus" convenience feature is no longer
used. Drop it.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200610053247.1583243-47-armbru@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200610053247.1583243-31-armbru@redhat.com>
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: <20200610053247.1583243-9-armbru@redhat.com>
I'm going to convert device realization to qdev_realize() with the
help of Coccinelle. Convert bus realization to qbus_realize() first,
to get it out of Coccinelle's way. Readability improves.
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: <20200610053247.1583243-7-armbru@redhat.com>
We commonly plug devices into their bus right when we create them,
like this:
dev = qdev_create(bus, type_name);
Note that @dev is a weak reference. The reference from @bus to @dev
is the only strong one.
We realize at some later time, either with
object_property_set_bool(OBJECT(dev), true, "realized", errp);
or its convenience wrapper
qdev_init_nofail(dev);
If @dev still has no QOM parent then, realizing makes the
/machine/unattached/ orphanage its QOM parent.
Note that the device returned by qdev_create() is plugged into a bus,
but doesn't have a QOM parent, yet. Until it acquires one,
unrealizing the bus will hang in bus_unparent():
while ((kid = QTAILQ_FIRST(&bus->children)) != NULL) {
DeviceState *dev = kid->child;
object_unparent(OBJECT(dev));
}
object_unparent() does nothing when its argument has no QOM parent,
and the loop spins forever.
Device state "no QOM parent, but plugged into bus" is dangerous.
Paolo suggested to delay plugging into the bus until realize. We need
to plug into the parent bus before we call the device's realize
method, in case it uses the parent bus. So the dangerous state still
exists, but only within realization, where we can manage it safely.
This commit creates infrastructure to do this:
dev = qdev_new(type_name);
...
qdev_realize_and_unref(dev, bus, errp)
Note that @dev becomes a strong reference here.
qdev_realize_and_unref() drops it. There is also plain
qdev_realize(), which doesn't drop it.
The remainder of this series will convert all users to this new
interface.
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Alistair Francis <alistair@alistair23.me>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200610053247.1583243-5-armbru@redhat.com>
This would have caught some of the bugs I just fixed.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200609122339.937862-25-armbru@redhat.com>
This would have caught some of the bugs I just fixed.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-Id: <20200609122339.937862-23-armbru@redhat.com>
Same story as for object_property_add(): the only way
object_property_del() can fail is when the property with this name
does not exist. Since our property names are all hardcoded, failure
is a programming error, and the appropriate way to handle it is
passing &error_abort. Most callers do that, the commit before
previous fixed one that didn't (and got the error handling wrong), and
the two remaining exceptions ignore errors.
Drop the @errp parameter.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200505152926.18877-19-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
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>
Several functions can't fail anymore: ich9_pm_add_properties(),
device_add_bootindex_property(), ppc_compat_add_property(),
spapr_caps_add_properties(), PropertyInfo.create(). Drop their @errp
parameter.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200505152926.18877-16-armbru@redhat.com>
The only way object_property_add() can fail is when a property with
the same name already exists. Since our property names are all
hardcoded, failure is a programming error, and the appropriate way to
handle it is passing &error_abort.
Same for its variants, except for object_property_add_child(), which
additionally fails when the child already has a parent. Parentage is
also under program control, so this is a programming error, too.
We have a bit over 500 callers. Almost half of them pass
&error_abort, slightly fewer ignore errors, one test case handles
errors, and the remaining few callers pass them to their own callers.
The previous few commits demonstrated once again that ignoring
programming errors is a bad idea.
Of the few ones that pass on errors, several violate the Error API.
The Error ** argument must be NULL, &error_abort, &error_fatal, or a
pointer to a variable containing NULL. Passing an argument of the
latter kind twice without clearing it in between is wrong: if the
first call sets an error, it no longer points to NULL for the second
call. ich9_pm_add_properties(), sparc32_ledma_realize(),
sparc32_dma_realize(), xilinx_axidma_realize(), xilinx_enet_realize()
are wrong that way.
When the one appropriate choice of argument is &error_abort, letting
users pick the argument is a bad idea.
Drop parameter @errp and assert the preconditions instead.
There's one exception to "duplicate property name is a programming
error": the way object_property_add() implements the magic (and
undocumented) "automatic arrayification". Don't drop @errp there.
Instead, rename object_property_add() to object_property_try_add(),
and add the obvious wrapper object_property_add().
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200505152926.18877-15-armbru@redhat.com>
[Two semantic rebase conflicts resolved]
Both qdev_connect_gpio_out_named() and device_set_realized() put
objects without a parent into the "/machine/unattached/" orphanage.
qdev_connect_gpio_out_named() needs a lengthy comment to explain how
it works. It exploits that object_property_add_child() can fail only
when we got a parent already, and ignoring that error does what we
want. True. If it failed due to "duplicate property", we'd be in
trouble, but that would be a programming error.
device_set_realized() is cleaner: it checks whether we need a parent,
then calls object_property_add_child(), aborting on failure. No need
for a comment, and programming errors get caught.
Change qdev_connect_gpio_out_named() to match.
Cc: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200505152926.18877-14-armbru@redhat.com>
object_property_set_description() and
object_class_property_set_description() fail only when property @name
is not found.
There are 85 calls of object_property_set_description() and
object_class_property_set_description(). None of them can fail:
* 84 immediately follow the creation of the property.
* The one in spapr_rng_instance_init() refers to a property created in
spapr_rng_class_init(), from spapr_rng_properties[].
Every one of them still gets to decide what to pass for @errp.
51 calls pass &error_abort, 32 calls pass NULL, one receives the error
and propagates it to &error_abort, and one propagates it to
&error_fatal. I'm actually surprised none of them violates the Error
API.
What are we gaining by letting callers handle the "property not found"
error? Use when the property is not known to exist is simpler: you
don't have to guard the call with a check. We haven't found such a
use in 5+ years. Until we do, let's make life a bit simpler and drop
the @errp parameter.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200505152926.18877-8-armbru@redhat.com>
[One semantic rebase conflict resolved]
Add functions to easily handle clocks with devices.
Clock inputs and outputs should be used to handle clock propagation
between devices.
The API is very similar the GPIO API.
This is based on the original work of Frederic Konrad.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 20200406135251.157596-4-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Deprecate device_legacy_reset(), qdev_reset_all() and
qbus_reset_all() to be replaced by new functions
device_cold_reset() and bus_cold_reset() which uses resettable API.
Also introduce resettable_cold_reset_fn() which may be used as a
replacement for qdev_reset_all_fn and qbus_reset_all_fn().
Following patches will be needed to look at legacy reset call sites
and switch to resettable api. The legacy functions will be removed
when unused.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200123132823.1117486-9-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This commit make use of the resettable API to reset the device being
hotplugged when it is realized. Also it ensures it is put in a reset
state coherent with the parent it is plugged into.
Note that there is a difference in the reset. Instead of resetting
only the hotplugged device, we reset also its subtree (switch to
resettable API). This is not expected to be a problem because
sub-buses are just realized too. If a hotplugged device has any
sub-buses it is logical to reset them too at this point.
The recently added should_be_hidden and PCI's partially_hotplugged
mechanisms do not interfere with realize operation:
+ In the should_be_hidden use case, device creation is
delayed.
+ The partially_hotplugged mechanism prevents a device to be
unplugged and unrealized from qdev POV and unrealized.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200123132823.1117486-8-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
In qdev_set_parent_bus(), when changing the parent bus of a
realized device, if the source and destination buses are not in the
same reset state, some adaptations are required. This patch adds
needed call to resettable_change_parent() to make sure a device reset
state stays coherent with its parent bus.
The addition is a no-op if:
1. the device being parented is not realized.
2. the device is realized, but both buses are not under reset.
Case 2 means that as long as qdev_set_parent_bus() is called
during the machine realization procedure (which is before the
machine reset so nothing is in reset), it is a no op.
There are 52 call sites of qdev_set_parent_bus(). All but one fall
into the no-op case:
+ 29 trivial calls related to virtio (in hw/{s390x,display,virtio}/
{vhost,virtio}-xxx.c) to set a vdev(or vgpu) composing device
parent bus just before realizing the same vdev(vgpu).
+ hw/core/qdev.c: when creating a device in qdev_try_create()
+ hw/core/sysbus.c: when initializing a device in the sysbus
+ hw/i386/amd_iommu.c: before realizing AMDVIState/pci
+ hw/isa/piix4.c: before realizing PIIX4State/rtc
+ hw/misc/auxbus.c: when creating an AUXBus
+ hw/misc/auxbus.c: when creating an AUXBus child
+ hw/misc/macio/macio.c: when initializing a MACIOState child
+ hw/misc/macio/macio.c: before realizing NewWorldMacIOState/pmu
+ hw/misc/macio/macio.c: before realizing NewWorldMacIOState/cuda
+ hw/net/virtio-net.c: Used for migration when using the failover
mechanism to migration a vfio-pci/net. It is
a no-op because at this point the device is
already on the bus.
+ hw/pci-host/designware.c: before realizing DesignwarePCIEHost/root
+ hw/pci-host/gpex.c: before realizing GPEXHost/root
+ hw/pci-host/prep.c: when initialiazing PREPPCIState/pci_dev
+ hw/pci-host/q35.c: before realizing Q35PCIHost/mch
+ hw/pci-host/versatile.c: when initializing PCIVPBState/pci_dev
+ hw/pci-host/xilinx-pcie.c: before realizing XilinxPCIEHost/root
+ hw/s390x/event-facility.c: when creating SCLPEventFacility/
TYPE_SCLP_QUIESCE
+ hw/s390x/event-facility.c: ditto with SCLPEventFacility/
TYPE_SCLP_CPU_HOTPLUG
+ hw/s390x/sclp.c: Not trivial because it is called on a SLCPDevice
just after realizing it. Ok because at this point the destination
bus (sysbus) is not in reset; the realize step is before the
machine reset.
+ hw/sd/core.c: Not OK. Used in sdbus_reparent_card(). See below.
+ hw/ssi/ssi.c: Used to put spi slave on spi bus and connect the cs
line in ssi_auto_connect_slave(). Ok because this function is only
used in realize step in hw/ssi/aspeed_smc.ci, hw/ssi/imx_spi.c,
hw/ssi/mss-spi.c, hw/ssi/xilinx_spi.c and hw/ssi/xilinx_spips.c.
+ hw/xen/xen-legacy-backend.c: when creating a XenLegacyDevice device
+ qdev-monitor.c: in device hotplug creation procedure before realize
Note that this commit alone will have no effect, right now there is no
use of resettable API to reset anything. So a bus will never be tagged
as in-reset by this same API.
The one place where side-effect will occurs is in hw/sd/core.c in
sdbus_reparent_card(). This function is only used in the raspi machines,
including during the sysbus reset procedure. This case will be
carrefully handled when doing the multiple phase reset transition.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200123132823.1117486-7-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This commit adds support of Resettable interface to buses and devices:
+ ResettableState structure is added in the Bus/Device state
+ Resettable methods are implemented.
+ device/bus_is_in_reset function defined
This commit allows to transition the objects to the new
multi-phase interface without changing the reset behavior at all.
Object single reset method can be split into the 3 different phases
but the 3 phases are still executed in a row for a given object.
From the qdev/qbus reset api point of view, nothing is changed.
qdev_reset_all() and qbus_reset_all() are not modified as well as
device_legacy_reset().
Transition of an object must be done from parent class to child class.
Care has been taken to allow the transition of a parent class
without requiring the child classes to be transitioned at the same
time. Note that SysBus and SysBusDevice class do not need any transition
because they do not override the legacy reset method.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200123132823.1117486-5-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Adds trace events to reset procedure and when updating the parent
bus of a device.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200123132823.1117486-3-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Provide a temporary device_legacy_reset function doing what
device_reset does to prepare for the transition with Resettable
API.
All occurrence of device_reset in the code tree are also replaced
by device_legacy_reset.
The new resettable API has different prototype and semantics
(resetting child buses as well as the specified device). Subsequent
commits will make the changeover for each call site individually; once
that is complete device_legacy_reset() will be removed.
Signed-off-by: Damien Hedde <damien.hedde@greensocs.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200123132823.1117486-2-damien.hedde@greensocs.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Use class properties facilities to add properties to the class during
device_class_set_props().
qdev_property_add_static() must be adapted as PropertyInfo now
operates with classes (and not instances), so we must
set_default_value() on the ObjectProperty, before calling its init()
method on the object instance.
Also, PropertyInfo.create() is now exclusively used for class
properties. Fortunately, qdev_property_add_static() is only used in
target/arm/cpu.c so far, which doesn't use "link" properties (that
require create()).
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200110153039.1379601-22-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the one-user function to the place it is being used.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200110153039.1379601-5-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
All callers use error_abort, and even the function itself calls with
error_abort.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200110153039.1379601-4-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The function is already documented in the header.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200110153039.1379601-3-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Define the new macro VMSTATE_INSTANCE_ID_ANY for callers who wants to
auto-generate the vmstate instance ID. Previously it was hard coded
as -1 instead of this macro. It helps to change this default value in
the follow up patches. No functional change.
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Replace DeviceState dependency with VMStateIf on vmstate API.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Acked-by: Halil Pasic <pasic@linux.ibm.com>
Add an interface to get the instance id, instead of depending on
Device and qdev_get_dev_path().
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Rename Error ** parameter in check_only_migratable to common errp.
In device_set_realized:
- Move "if (local_err != NULL)" closer to error setters.
- Drop 'Error **local_errp': it doesn't save any LoCs, but it's very
unusual.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20191205174635.18758-10-vsementsov@virtuozzo.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
In "b06424de62 migration: Disable hotplug/unplug during migration" we
added a check to disable unplug for all devices until we have figured
out what works. For failover primary devices qdev_unplug() is called
from the migration handler, i.e. during migration.
This patch adds a flag to DeviceState which is set to false for all
devices and makes an exception for PCI devices that are also
primary devices in a failover pair.
Signed-off-by: Jens Freimann <jfreimann@redhat.com>
Message-Id: <20191029114905.6856-8-jfreimann@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This adds support for hiding a device to the qbus and qdev APIs. The
first user of this will be the virtio-net failover feature but the API
introduced with this patch could be used to implement other features as
well, for example hiding pci devices when a pci bus is powered off.
qdev_device_add() is modified to check for a failover_pair_id
argument in the option string. A DeviceListener callback
should_be_hidden() is added. It can be used by a standby device to
inform qdev that this device should not be added now. The standby device
handler can store the device options to plug the device in at a later
point in time.
One reason for hiding the device is that we don't want to expose both
devices to the guest kernel until the respective virtio feature bit
VIRTIO_NET_F_STANDBY was negotiated and we know that the devices will be
handled correctly by the guest.
More information on the kernel feature this is using:
https://www.kernel.org/doc/html/latest/networking/net_failover.html
An example where the primary device is a vfio-pci device and the standby
device is a virtio-net device:
A device is hidden when it has an "failover_pair_id" option, e.g.
-device virtio-net-pci,...,failover=on,...
-device vfio-pci,...,failover_pair_id=net1,...
Signed-off-by: Jens Freimann <jfreimann@redhat.com>
Message-Id: <20191029114905.6856-2-jfreimann@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Introduce this new per-machine hook to give any machine class a chance
to do a sanity check on the to-be-hotplugged device as a sanity test.
This will be used for x86 to try to detect some illegal configuration
of devices, e.g., possible conflictions between vfio-pci and x86
vIOMMU.
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20190916080718.3299-3-peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In my "build everything" tree, changing sysemu/sysemu.h triggers a
recompile of some 5400 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h).
Almost a third of its inclusions are actually superfluous. Delete
them. Downgrade two more to qapi/qapi-types-run-state.h, and move one
from char/serial.h to char/serial.c.
hw/semihosting/config.c, monitor/monitor.c, qdev-monitor.c, and
stubs/semihost.c define variables declared in sysemu/sysemu.h without
including it. The compiler is cool with that, but include it anyway.
This doesn't reduce actual use much, as it's still included into
widely included headers. The next commit will tackle that.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20190812052359.30071-27-armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
In my "build everything" tree, changing hw/qdev-properties.h triggers
a recompile of some 2700 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h).
Many places including hw/qdev-properties.h (directly or via hw/qdev.h)
actually need only hw/qdev-core.h. Include hw/qdev-core.h there
instead.
hw/qdev.h is actually pointless: all it does is include hw/qdev-core.h
and hw/qdev-properties.h, which in turn includes hw/qdev-core.h.
Replace the remaining uses of hw/qdev.h by hw/qdev-properties.h.
While there, delete a few superfluous inclusions of hw/qdev-core.h.
Touching hw/qdev-properties.h now recompiles some 1200 objects.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Daniel P. Berrangé" <berrange@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20190812052359.30071-22-armbru@redhat.com>
In my "build everything" tree, changing migration/vmstate.h triggers a
recompile of some 2700 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h).
hw/hw.h supposedly includes it for convenience. Several other headers
include it just to get VMStateDescription. The previous commit made
that unnecessary.
Include migration/vmstate.h only where it's still needed. Touching it
now recompiles only some 1600 objects.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20190812052359.30071-16-armbru@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
In my "build everything" tree, changing hw/irq.h triggers a recompile
of some 5400 out of 6600 objects (not counting tests and objects that
don't depend on qemu/osdep.h).
hw/hw.h supposedly includes it for convenience. Several other headers
include it just to get qemu_irq and.or qemu_irq_handler.
Move the qemu_irq and qemu_irq_handler typedefs from hw/irq.h to
qemu/typedefs.h, and then include hw/irq.h only where it's still
needed. Touching it now recompiles only some 500 objects.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190812052359.30071-13-armbru@redhat.com>
Move commands object-add, object-del, qom-get, qom-list,
qom-list-properties, qom-list-types, and qom-set with their types from
misc.json to new qom.json.
Move commands device-list-properties, device_add, device-del, and
event DEVICE_DELETED from misc.json to new qdev.json.
Add both new files to MAINTAINERS section QOM.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Daniel P. Berrange" <berrange@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190619201050.19040-5-armbru@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
[Straightforwardly updated for "MAINTAINERS: Make section "QOM" cover
qdev as well"]
See the previous commit for rationale.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190308131445.17502-3-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Compatibility properties started life as a qdev property thing: we
supported them only for qdev properties, and implemented them with the
machinery backing command line option -global.
Recent commit fa0cb34d22 put them to use (tacitly) with memory
backend objects (subtypes of TYPE_MEMORY_BACKEND). To make that
possible, we first moved the work of applying them from the -global
machinery into TYPE_DEVICE's .instance_post_init() method
device_post_init(), in commits ea9ce8934c and b66bbee39f, then made
it available to TYPE_MEMORY_BACKEND's .instance_post_init() method
host_memory_backend_post_init() as object_apply_compat_props(), in
commit 1c3994f6d2.
Note the code smell: we now have function name starting with object_
in hw/core/qdev.c. It has to be there rather than in qom/, because it
calls qdev_get_machine() to find the current accelerator's and
machine's compat_props.
Turns out calling qdev_get_machine() there is problematic. If we
qdev_create() from a machine's .instance_init() method, we call
device_post_init() and thus qdev_get_machine() before main() can
create "/machine" in QOM. qdev_get_machine() tries to get it with
container_get(), which "helpfully" creates it as "container" object,
and returns that. object_apply_compat_props() tries to paper over the
problem by doing nothing when the value of qdev_get_machine() isn't a
TYPE_MACHINE. But the damage is done already: when main() later
attempts to create the real "/machine", it fails with "attempt to add
duplicate property 'machine' to object (type 'container')", and
aborts.
Since no machine .instance_init() calls qdev_create() so far, the bug
is latent. But since I want to do that, I get to fix the bug first.
Observe that object_apply_compat_props() doesn't actually need the
MachineState, only its the compat_props member of its MachineClass and
AccelClass. This permits a simple fix: register MachineClass and
AccelClass compat_props with the object_apply_compat_props() machinery
right after these classes get selected.
This is actually similar to how things worked before commits
ea9ce8934c and b66bbee39f, except we now register much earlier. The
old code registered them only after the machine's .instance_init()
ran, which would've broken compatibility properties for any devices
created there.
Cc: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20190308131445.17502-2-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Let's use a wrapper instead of looking it up manually. This function can
than be reused when we explicitly want to have the bus hotplug handler
(e.g. when the bus hotplug handler was overwritten by the machine
hotplug handler).
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190228122849.4296-4-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
it will allow to return another hotplug handler than the default
one for a specific bus based device type. Which is needed to handle
non trivial plug/unplug sequences that need the access to resources
configured outside of bus where device is attached.
That will allow for returned hotplug handler to orchestrate wiring
in arbitrary order, by chaining other hotplug handlers when
it's needed.
PS:
It could be used for hybrid virtio-mem and virtio-pmem devices
where it will return machine as hotplug handler which will do
necessary wiring at machine level and then pass control down
the chain to bus specific hotplug handler.
Example of top level hotplug handler override and custom plug sequence:
some_machine_get_hotplug_handler(machine){
if (object_dynamic_cast(OBJECT(dev), TYPE_SOME_BUS_DEVICE)) {
return HOTPLUG_HANDLER(machine);
}
return NULL;
}
some_machine_device_plug(hotplug_dev, dev) {
if (object_dynamic_cast(OBJECT(dev), TYPE_SOME_BUS_DEVICE)) {
/* do machine specific initialization */
some_machine_init_special_device(dev)
/* pass control to bus specific handler */
hotplug_handler_plug(dev->parent_bus->hotplug_handler, dev)
}
}
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190228122849.4296-3-david@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
When unplugging a device, at one point the device will be destroyed
via object_unparent(). This will, one the one hand, unrealize the
removed device hierarchy, and on the other hand, destroy/free the
device hierarchy.
When chaining hotplug handlers, we want to overwrite a bus hotplug
handler by the machine hotplug handler, to be able to perform
some part of the plug/unplug and to forward the calls to the bus hotplug
handler.
For now, the bus hotplug handler would trigger an object_unparent(), not
allowing us to perform some unplug action on a device after we forwarded
the call to the bus hotplug handler. The device would be gone at that
point.
machine_unplug_handler(dev)
/* eventually do unplug stuff */
bus_unplug_handler(dev)
/* dev is gone, we can't do more unplug stuff */
So move the object_unparent() to the original caller of the unplug. For
now, keep the unrealize() at the original places of the
object_unparent(). For implicitly chained hotplug handlers (e.g. pc
code calling acpi hotplug handlers), the object_unparent() has to be
done by the outermost caller. So when calling hotplug_handler_unplug()
from inside an unplug handler, nothing is to be done.
hotplug_handler_unplug(dev) -> calls machine_unplug_handler()
machine_unplug_handler(dev) {
/* eventually do unplug stuff */
bus_unplug_handler(dev) -> calls unrealize(dev)
/* we can do more unplug stuff but device already unrealized */
}
object_unparent(dev)
In the long run, every unplug action should be factored out of the
unrealize() function into the unplug handler (especially for PCI). Then
we can get rid of the additonal unrealize() calls and object_unparent()
will properly unrealize the device hierarchy after the device has been
unplugged.
hotplug_handler_unplug(dev) -> calls machine_unplug_handler()
machine_unplug_handler(dev) {
/* eventually do unplug stuff */
bus_unplug_handler(dev) -> only unplugs, does not unrealize
/* we can do more unplug stuff */
}
object_unparent(dev) -> will unrealize
The original approach was suggested by Igor Mammedov for the PCI
part, but I extended it to all hotplug handlers. I consider this one
step into the right direction.
To summarize:
- object_unparent() on synchronous unplugs is done by common code
-- "Caller of hotplug_handler_unplug"
- object_unparent() on asynchronous unplugs ("unplug requests") has to
be done manually
-- "Caller of hotplug_handler_unplug"
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190228122849.4296-2-david@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The qbus_is_full(BusState *bus) function (qdev_monitor.c) compares the max_index
value of the BusState structure with the max_dev value of the BusClass structure
to determine whether the maximum number of children has been reached for the
bus. The problem is, the max_index field of the BusState structure does not
necessarily reflect the number of devices that have been plugged into
the bus.
Whenever a child device is plugged into the bus, the bus's max_index value is
assigned to the child device and then incremented. If the child is subsequently
unplugged, the value of the max_index does not change and no longer reflects the
number of children.
When the bus's max_index value reaches the maximum number of devices
allowed for the bus (i.e., the max_dev field in the BusClass structure),
attempts to plug another device will be rejected claiming that the bus is
full -- even if the bus is actually empty.
To resolve the problem, a new 'num_children' field is being added to the
BusState structure to keep track of the number of children plugged into the
bus. It will be incremented when a child is plugged, and decremented when a
child is unplugged.
Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
Reviewed-by: Pierre Morel<pmorel@linux.ibm.com>
Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
Message-Id: <1545062250-7573-1-git-send-email-akrowiak@linux.ibm.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Markus Armbruster
Damien Hedde
Pan Nengyuan
Marc-André Lureau
Paolo Bonzini
Peter Xu
Vladimir Sementsov-Ogievskiy
Jens Freimann
David Hildenbrand
Igor Mammedov
Tony Krowiak