Peter Krempa [Mon, 28 May 2018 09:36:47 +0000 (11:36 +0200)]
qemu: domain: Pass 'qemuCaps' to post parse callbacks when parsing status XML
When status XML was parsed the post-parse callbacks could not access
qemu caps and potentially upgrade the definition according to the
present caps. Implement the callback to pass it in.
Peter Krempa [Mon, 28 May 2018 09:36:46 +0000 (11:36 +0200)]
conf: domain: Allow passing in 'parseOpaque' for post-parse of status XML
The status XML parser function virDomainObjParseXML could not pass in
parseOpaque into the post parse callbacks. Add a callback which will
allow hypervisor drivers to fill it from the 'virDomainObj' data.
Peter Krempa [Mon, 28 May 2018 09:36:45 +0000 (11:36 +0200)]
conf: domain: Invoke post-parse callbacks after parsing private XML parts
When parsing status XML the post-parse callbacks can't access any
private data present in the status XML as the private bits were parsed
after invoking post-parse callbacks.
Move the invocation so that everything is parsed first.
Ján Tomko [Mon, 28 May 2018 12:31:50 +0000 (14:31 +0200)]
conf: introduce virDomainDefBootOrderPostParse
Move the check for boot elements into a separate function
and remove its dependency on the parser-supplied bootHash table.
Reconstructing the hash table from the domain definition
effectively duplicates the check for duplicate boot order
values, also present in virDomainDeviceBootParseXML.
Now it will also be run on domains created by other means than XML
parsing, since it will be run even for code paths that did not supply
the bootHash table before.
Ján Tomko [Mon, 28 May 2018 13:16:16 +0000 (15:16 +0200)]
vmx: add VIR_DOMAIN_DEF_FEATURE_NO_BOOT_ORDER
Further patches will introduce validation and a default setting
of def->os.bootDevs in postParse.
Introduce a feature flag to opt out of this and set it in the vmx
driver, otherwise we would be adding it <boot dev='hd'/> into every
vmx config despite having no way to change it.
Michal Privoznik [Mon, 28 May 2018 14:28:09 +0000 (16:28 +0200)]
qemu: Don't build cache= cmd line for scsi-block
Trying to set any cache for <disk device='lun'/> makes no sense.
Such disk translates into -device scsi-block on the command line
and the device lacks any cache setting because it's merely a
middle man between qemu and real SCSI device.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
The function creates a list of all (or migratable only) CPU features
supported by QEMU. It works by looking at the CPU model info returned by
query-cpu-model-expansion QMP command.
Jiri Denemark [Tue, 15 May 2018 09:57:35 +0000 (11:57 +0200)]
cpu: Add optional list of allowed features to virCPUBaseline
When computing a baseline CPU for a specific hypervisor we have to make
sure to include only CPU features supported by the hypervisor. Otherwise
the computed CPU could not be used for starting a new domain.
Jiri Denemark [Tue, 15 May 2018 09:27:20 +0000 (11:27 +0200)]
cpu_x86: Add support for passing guest CPUs to virCPUx86Baseline
Modern host CPU models from domain capabilities XMLs are reported as
guest CPU definitions with feature policies. This patch updates
virCPUx86Baseline to properly handle such CPU models.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Introduce virConnectBaselineHypervisorCPU public API
The new API computes the most feature-rich CPU which is compatible with
all given CPUs and can be provided by the specified hypervisor. It is a
more useful version of virConnectBaselineCPU, which doesn't consider any
hypervisor capabilities when computing the best CPU.
Introduce virConnectCompareHypervisorCPU public API
This new API compares the given CPU description with the CPU the
specified hypervisor is able to provide on the host. It is a more useful
version of virConnectCompareCPU, which compares the CPU definition with
the host CPU without considering any specific hypervisor and its
abilities.
virConnectGetDomainCapabilities needs to lookup QEMU capabilities
matching a specified binary, architecture, virt type, and machine type
while using default values when any of the parameters are not provided
by the user. Let's extract the lookup code into
virQEMUCapsCacheLookupDefault to make it reusable.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Jiri Denemark [Fri, 4 May 2018 21:46:55 +0000 (23:46 +0200)]
vshExtractCPUDefXML: Accept domain capabilities XML
The domain capabilities XML contains host CPU model tailored to a
specific hypervisor and since it's enclosed in <mode name='host-model'>
element rather then the required <cpu> it's impossible to directly use
the host CPU model as an input to, e.g., cpu-compare command. To make
this more convenient, vshExtractCPUDefXML now accepts full domain
capabilities XML and automatically transforms the host CPU models into
the form accepted by libvirt APIs.
virsh: Extract common code from cmdCPU{Compare,Baseline}
Both cpu-compare and cpu-baseline commands accept more that just CPU
definition XML(s). For users' convenience they are able to extract the
CPU definition(s) even from domain XML or capabilities XML. The main
differences between the two commands is in the number of CPU definitions
they expect: cpu-compare wants only one CPU definition while
cpu-baseline expects one or more CPUs.
The extracted code forms a new vshExtractCPUDefXML function.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Wed, 23 May 2018 16:01:16 +0000 (18:01 +0200)]
qemu_hotplug: Allow asynchronous detach
The virDomainDetachDeviceAlias API is designed so that it only
sends detach request to qemu. It's user's responsibility to wait
for DEVICE_DELETED event, not libvirt's. Add @async flag to
qemuDomainDetach*Device() functions so that caller can chose if
detach is semi-synchronous (old virDomainDetachDeviceFlags()) or
fully asynchronous (new virDomainDetachDeviceFlags()).
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
The fact that we are overwriting @ret multiple times makes it
difficult to see what is actually happening here. Follow our
traditional pattern where @ret is initialized to -1, and set to 0
only in case we know we succeeded.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Wed, 23 May 2018 16:36:57 +0000 (18:36 +0200)]
qemu_hotplug: Use more gotos in qemuDomainDetach*Device
We are overwriting @ret a lot. It makes hard to see what is
actually going on. Use more gotos. Two functions are fixed here:
qemuDomainDetachShmemDevice() and qemuDomainDetachWatchdog().
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Wed, 23 May 2018 16:32:26 +0000 (18:32 +0200)]
qemuDomainRemoveChrDevice: Release device address
Instead of releasing address only sometimes in
qemuDomainDetachChrDevice() let's release it whenever the device
is actually removed from the domain definition.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Mon, 21 May 2018 11:15:21 +0000 (13:15 +0200)]
Introduce virDomainDetachDeviceAlias API
When detaching a device it can be uniquely identified by its
alias. Instead of misusing virDomainDetachDeviceFlags which has
the same signature introduce new function.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Commit id 'fe2af45b' added output for logical_block_size and
num_blocks for both removeable and fixed storage, but did not
update the nodedev capability causing virt-xml-validate to fail.
It's listed as optional only because it only prints if the
sizes are > 0. For a CDROM drive the values won't be formatted.
Update the nodedevxml2xmltest in order to output the values
for storage based on the logic from udevProcessRemoveableMedia
and udevProcessSD with respect to the logical_blocksize and
num_blocks calculations.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by Michal Privoznik <mprivozn@redhat.com>
Commit id '02129b7c0' added a single pagesElem for slightly
different purposes. One usage was an output for host page size
listing and the other for NUMA supported page sizes. For the
former, only the pages unit and size are formatted, while for
the latter the pages unit, size, and availability data is formatted.
The virt-xml-validate would fail because it expected something
extra in the host page size output. So split up pagesElem a bit
and create pagesHost and pagesNuma for the differences.
Modify some capabilityschemadata output to have the output - even
though the results may not be realistic with respect to the
original incarnation of the data.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Fri, 18 May 2018 22:25:54 +0000 (18:25 -0400)]
schema: Add vzmigr for host migrate transport capability
Commit id '0eced74f3' added vzmigr as a valid option for
virCapabilitiesAddHostMigrateTransport, but didn't update
the capabilities schema resulting in possible virt-xml-validate
failure.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by Michal Privoznik <mprivozn@redhat.com>
Commit id 'b3fd95e36' added rdma as a valid option for
virCapabilitiesAddHostMigrateTransport, but didn't update
the capabilities schema resulting in possible virt-xml-validate
failure.
While at it, update the capabilityschemadata for caps-qemu-kvm
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Fri, 18 May 2018 21:47:08 +0000 (17:47 -0400)]
schema,tests: Use vpxmigr for host migrate transport capability
Commit id 'e4938ce2f' changed the esx_driver to use 'vpxmigr'
instead of esx for virCapabilitiesAddHostMigrateTransport, so
update the capabilities to allow virt-xml-validate to pass and
update the test to use the newer name.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Fri, 18 May 2018 22:05:07 +0000 (18:05 -0400)]
schema: Remove xenmigr from host migrate transport capability
Commit id '1dac5fbb' removed xenmigr as a capability option
for virCapabilitiesAddHostMigrateTransport but didn't update
the schema resulting in possible failure for virt-xml-validate.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by Michal Privoznik <mprivozn@redhat.com>
In order to avoid a possible error as a result of kernel interactions
with the partition helper, let's use virWaitForDevices to force things
to settle down before attempting to open and read the partition. This
is related to https://bugzilla.redhat.com/show_bug.cgi?id=1264719.
Although perhaps overkill to have too many places to settle, since
we know that the act of reading the partitions via libvirt_parthelper
will cause udev activity/events - we just need to ensure udev has
been settled before proceding with usage of the device.
Signed-off-by: John Ferlan <jferlan@redhat.com> ACKed-by: Michal Privoznik <mprivozn@redhat.com>
If the domain requests usage of the genid functionality,
then add the QEMU '-device vmgenid' to the command line
providing either the supplied or generated GUID value.
Add tests for both a generated and supplied GUID value.
Signed-off-by: John Ferlan <jferlan@redhat.com> ACKed-by: Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Mon, 9 Apr 2018 14:31:17 +0000 (10:31 -0400)]
qemu: Alter VM Generation ID for specific startup/launch transitions
Before we generate the command line for qemu, if the domain about to
be launched desires to utilize the VM Generation ID functionality, then
handle both the regenerating the GUID value for backup recovery (restore
operation) and the startup after snapshot as both require a new GUID to
be generated to allow the guest operating system to recognize the VM
is re-executing something that has already executed before.
Signed-off-by: John Ferlan <jferlan@redhat.com> ACKed-by: Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Mon, 9 Apr 2018 16:10:02 +0000 (12:10 -0400)]
conf: Add VM Generation ID parse/format support
The VM Generation ID is a mechanism to provide a unique 128-bit,
cryptographically random, and integer value identifier known as
the GUID (Globally Unique Identifier) to the guest OS. The value
is used to help notify the guest operating system when the virtual
machine is executed with a different configuration.
This patch adds support for a new "genid" XML element similar to
the "uuid" element. The "genid" element can have two forms "<genid/>"
or "<genid>$GUID</genid>". If the $GUID is not provided, libvirt
will generate one and save it in the XML.
Since adding support for a generated GUID (or UUID like) value to
be displayed modifying the xml2xml test to include virrandommock.so
is necessary since it will generate a "known" value.
Signed-off-by: John Ferlan <jferlan@redhat.com> ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Lin Ma [Wed, 23 May 2018 06:32:41 +0000 (14:32 +0800)]
virsh: Move vshEventCallback structure definition to virsh-domain.h
The next patch will use it in virsh-completer.c for returning the
strings of domain event name. At the same time, rename it to
avoid possible name clashing.
Signed-off-by: Lin Ma <lma@suse.com> Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Peter Krempa [Tue, 22 May 2018 11:46:05 +0000 (13:46 +0200)]
qemu: monitor: Drop JSON versions of savevm/delvm/loadvm
qemu did not QAPIfy these and the design and name will most probably
change. The replacements will not be compatible. Drop the JSON stubs and
annotate that there won't be a replacement.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 28 Feb 2018 14:20:17 +0000 (15:20 +0100)]
qemu: migration: Add support for transporting NBD over TLS
Implement the secure way to transport non-shared storage data across
migrations. The new approach uses blockdev-add to create the NBD client
so that the TLS secret object can be specified.
Peter Krempa [Thu, 22 Feb 2018 15:11:59 +0000 (16:11 +0100)]
qemu: migration: Rename NBD migration functions
Drop the mention of 'drive mirror' from the function names and mention
NBD. This will help when adding the 'blockdev mirror' migration code
which will allow using TLS.
Additionally fix some of the function comments to make more sense
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Wed, 28 Feb 2018 14:22:30 +0000 (15:22 +0100)]
qemu: domain: Add private data for NBD migration storage source definition
Allow saving various aspects necessary to do NBD migration via blockdev
by storing a 'virStorageSource' in the disk private data meant to store
the NBD target of migration. Along with this add code to parse and
format it into the status XML.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Fri, 23 Feb 2018 11:59:13 +0000 (12:59 +0100)]
qemu: block: Add helpers for hot-adding virStorageSource via blockdev
These helpers add infrastructure which simplifies adding and rolling
back virStorageSources to a running qemu instance. Using of the helper
structure and separate functions allows for a much cleaner code in the
section dealing with the monitor.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Tue, 20 Feb 2018 16:56:46 +0000 (17:56 +0100)]
qemu: monitor: Factor out and document code to format QMP command
Move formatting of the qemu command out of qemuMonitorJSONMakeCommandRaw
to qemuMonitorJSONMakeCommandInternal to allow greater reusability and
document the function better.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Thu, 29 Mar 2018 11:22:04 +0000 (13:22 +0200)]
qemu: caps: Add capability for blockdev-add/blockdev-del
The capability also represents that 'blockdev-add' is functional. It's
necessary to detect it via presence of 'blockdev-del' since blockdev-add
did not have the unsupported 'x-blockdev-add' version previously and
thus would be marked as present even if we could not use it.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Fri, 23 Feb 2018 08:03:49 +0000 (09:03 +0100)]
qemu: migration: Don't access disk members without lock
The initiation of a synchronous block job in the NBD storage migration
code was placed after entering the monitor thus after the lock on the VM
object was unlocked. Thankfully nothing bad could happen in this
situation since the migration job prevents any disk detaches or other
modifications of the domain object.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
projects / libvirt.git / log