To prepare for a conversion to GObject, we need virObjectUnref
to have the same API design as g_object_unref, which means it
needs to be void.
A few places do actually care about the return value though,
and in these cases a thread local flag is used to determine
if the dispose method was invoked.
Reviewed-by: Michal Privoznik <mprivozn@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Some, but not all, of the monitor event handlers check
the virObjectUnref return value to see if the domain
was disposed.
It should not be possible for this to happen, since
the function already holds a lock on the domain and
has only just acquired an extra reference on the
domain a few lines earlier.
Reviewed-by: Michal Privoznik <mprivozn@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Michal Privoznik [Tue, 26 May 2020 14:26:25 +0000 (16:26 +0200)]
qemu: Skip pre-creation of NVMe disks
Upon migration with disks, libvirt determines if each disk exists
on the destination and tries to pre-create missing ones. Well,
NVMe disks can't be pre-created, but they can be checked for
presence.
security: don't fail if built without attr support
If built without attr support removing any image will trigger
qemuBlockRemoveImageMetadata (the one that emits the warning)
-> qemuSecurityMoveImageMetadata
-> virSecurityManagerMoveImageMetadata
-> virSecurityDACMoveImageMetadata
-> virSecurityDACMoveImageMetadataHelper
-> virProcessRunInFork (spawns subprocess)
-> virSecurityMoveRememberedLabel
In there due to !HAVE_LIBATTR virFileGetXAttrQuiet will return
ENOSYS and from there the chain will error out.
That is wrong and looks like:
libvirtd[6320]: internal error: child reported (status=125):
libvirtd[6320]: Unable to remove disk metadata on vm testguest from
/var/lib/uvtool/libvirt/images/testguest.qcow (disk target vda)
This change makes virSecurityDACMoveImageMetadataHelper and
virSecuritySELinuxMoveImageMetadataHelper accept that
error code gracefully and in that sense it is an extension of: 5214b2f1a3f "security: Don't skip label restore on file systems lacking XATTRs"
which does the same for other call chains into the virFile*XAttr functions.
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com> Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Erik Skultety [Mon, 25 May 2020 12:46:26 +0000 (14:46 +0200)]
scripts: Fix E741 that pycodesyle is pointing out during syntax-check
With newer pycodestyle 2.6.0 (which is part of flake8-3.8.2) reports
the following pep violation during syntax-check:
../scripts/check-remote-protocol.py:95:9: E741 ambiguous variable name 'l'
for l in err.strip().split("\n")
On all the distros we test on, this hasn't occurred yet, but with the
future update of flake8 it likely would. The fix is easy, just name the
variable appropriately.
Signed-off-by: Erik Skultety <eskultet@redhat.com> Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Jiri Denemark [Tue, 19 May 2020 13:08:11 +0000 (15:08 +0200)]
cpu_map: Add Cooperlake x86 CPU model
The stepping range (10-11) is likely incomplete. QEMU uses 10 and the
CPUID data for Cooperlake show 11. We will update the range if needed
once more details about he CPU are available.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com> Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
qemuProcessRefreshCPU: skip 'host-model' logic for pSeries guests
Commit v3.10.0-182-g237f045d9a ("qemu: Ignore fallback CPU attribute
on reconnect") forced CPU 'fallback' to ALLOW, regardless of user
choice. This fixed a situation in which guests created with older
Libvirt versions, which used CPU mode 'host-model' in runtime, would
fail to launch in a newer Libvirt if the fallback was set to FORBID.
This would lead to a scenario where the CPU was translated to 'host-model'
to 'custom', but then the FORBID setting would make the translation
process fail.
PSeries can operate with 'host-model' in runtime due to specific PPC64
mechanics regarding compatibility mode. The update() implementation of
the cpuDriverPPC64 driver is a NO-OP if CPU mode is 'host-model', and
the driver does not implement translate(). The commit mentioned above
is causing PSeries guests to get their 'fallback' setting to ALLOW,
overwriting user choice, exposing a design problem in
qemuProcessRefreshCPU() - for PSeries guests, handling 'host-model'
as it is being done does not apply.
All other cpuArchDrivers implements update() and changes guest mode
to VIR_CPU_MODE_CUSTOM, meaning that PSeries is currently the only
exception to this logic. Let's make it official.
Jiri Denemark [Fri, 15 May 2020 20:00:29 +0000 (22:00 +0200)]
qemu: Invalidate capabilities when host CPU changes
The host CPU related info stored in the capabilities cache is no longer
valid after the host CPU changes. This is not a frequent situation in
real world, but it can easily happen in nested scenarios when a disk
image is started with various CPUs.
Jiri Denemark [Tue, 31 Mar 2020 22:44:00 +0000 (00:44 +0200)]
hostcpu: Introduce virHostCPUGetSignature
The purpose of this function is to give a short description that would
be change when a host CPU is replaced with a different model. This is
currently implemented by reading /proc/cpuinfo.
It should be implemented for all architectures for which the QEMU driver
stores host CPU data in the capabilities cache. In other words for archs
that support host-model CPUs.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
scripts: emit raw enum value in API build description
Currently the value for an enum is only emitted if it is a plain
string. If the enum is an integer or hex value, or a complex code block,
it is omitted from the API build. This fixes that by emitting the raw
value if no string value is present.
scripts: emit enum parameters in API build description
Currently the information about enums in the API document lacks any
mention of parameters, so it is impossible to tell what kind of enum
declaration is present in the libvirt API header. With this change
Jonathon Jongsma [Thu, 21 May 2020 20:09:35 +0000 (15:09 -0500)]
docs: Document full node device xml in formatnode.html.in
Some of the node device xml schema was documented in drvnodedev.html.in
rather than in formatnode.html.in. Move all of the schema documentation
to formatnode.html.in and provide reference links from the
drvnodedev.html.in page.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
Jonathon Jongsma [Thu, 21 May 2020 20:09:34 +0000 (15:09 -0500)]
docs: Fix names for PF and VF PCI device capabilities
The proper name for physical function capability is 'phys_function', not
'physical_function'. Likewise, a virtual function capability is
'virt_functions' rather than 'virtual_function'.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
Michal Privoznik [Thu, 21 May 2020 18:55:01 +0000 (20:55 +0200)]
testCompareXMLToArgvValidateSchema: Construct @vm from scratch
Currently, the @vm is passed in as an argument and
testCompareXMLToArgvCreateArgs() is called over it which means
under the hood qemuProcessPrepareDomain() is called. But at the
point where ValidateSchema() is called, the domain object is
already 'prepared', i.e. all device aliases are assigned and so
on. But our code is not prepared to 'prepare' a domain twice - it
simply overwrites all the pointers leading to a memory leak.
Fortunately, this is only the problem of this test.
Resolve this by constructing the domain object from scratch.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Peter Krempa <pkrempa@redhat.com>
logging.html.in: fix number of output formats available
There are 4 formats available (x:stderr, x:syslog:name,
x:file:file_path, x:journald), not 3. Use "the following"
instead of the actual number to avoid the need to update
the number every time a new form is added/removed.
Suggested-by: Pino Toscano <ptoscano@redhat.com> Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com> Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Peter Krempa [Fri, 15 May 2020 12:33:10 +0000 (14:33 +0200)]
qemuxml2argvtest: Add QAPI/QMP schema validation for -blockdev and -netdev
Our hotplug test cases are weak in comparison to the qemuxml2argvtest.
Use all the the input data to also validate the arguments for -netdev
and -blockdev against the appropriate commands of the QMP schema.
Note that currently it's done just for the _CAPS versions of tests but
commenting out a line in the test file allows to validate even cases
which don't use real capabilities.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Fri, 15 May 2020 12:24:21 +0000 (14:24 +0200)]
qemu: Prepare for testing of 'netdev_add' props via qemuxml2argvtest
qemuxml2argv test suite is way more comprehensive than the hotplug
suite. Since we share the code paths for monitor and command line
hotplug we can easily test the properties of devices against the QAPI
schema.
To achieve this we'll need to skip the JSON->commandline conversion for
the test run so that we can analyze the pure properties. This patch adds
flags for the comand line generator and hook them into the
JSON->commandline convertor for -netdev. An upcoming patch will make use
of this new infrastructure.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
I guess the original idea was to make it extensible while not worrying
about adding another object for it. Either way it requires us to add yet
another JSON->cmdline convertor for arrays.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Thu, 14 May 2020 20:50:59 +0000 (22:50 +0200)]
qemu: command: Generate -netdev command line via JSON->cmdline conversion
The 'netdev_add' command was recently formally described in qemu via the
QMP schema. This means that it also requires the arguments to be
properly formatted. Our current approach is to generate the command line
and then use qemuMonitorJSONKeywordStringToJSON to get the JSON
properties for the monitor. This will not work if we need to pass some
fields as numbers or booleans.
In this step we re-do internals of qemuBuildHostNetStr to format a JSON
object which is converted back via virQEMUBuildNetdevCommandlineFromJSON
to the equivalent command line. This will later allow fixing of the
monitor code to use the JSON object directly rather than rely on the
conversion.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
In preparation for converting the generator of -netdev to generate JSON
which will be used to do the command line rather than the other way
around we need to introduce a convertor which properly configures
virQEMUBuildCommandLineJSON for the quirks of -netdev.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Fri, 15 May 2020 06:50:31 +0000 (08:50 +0200)]
testCompareXMLToArgv: Split out preparation and command formatting
There are multiple steps of setting up the domain definition prior to
formatting the command line for the tests. Extract it to a separate
function so that it's self-contained and also will allow re-running the
command line formatting which will be necessary for QMP schema
validation tests.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Thu, 14 May 2020 07:41:48 +0000 (09:41 +0200)]
virQEMUBuildCommandLineJSON: Allow skipping certain keys
Allow reusing this for formatting of netdev_add arguments into -netdev.
We need to be able to skip the 'type' property as it's used without the
prefix by our generator.
Add infrastructure which allows skipping property with a specific name.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Thu, 14 May 2020 10:02:55 +0000 (12:02 +0200)]
qemuBuildHostNetStr: Stop using 'ipv6-net' convenience argument
In qemu the argument of 'ipv6-net' is split up into 'ipv6-prefix' and
'ipv6-prefixlen'. Additionally now that 'netdev_add' was qapified, only
the real properties are allowed. Switch to using them explicitly.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Thu, 14 May 2020 11:01:59 +0000 (13:01 +0200)]
qemuBuildChannelChrDeviceStr: Remove formatting of properties for -netdev
The output of the function is fed as argument to '-device' command line
argument or 'device_add' monitor command except for 'guestfwd' channels
where it needs to be fed to -netdev/netdev_add. This is confusing and
error prone. Split it up since the caller needs to know which
command/option to use anyways, so the caller can call the appropriate
function without any magic.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Thu, 14 May 2020 10:51:21 +0000 (12:51 +0200)]
qemuBuildChannelsCommandLine: Extract common formatting of 'chardev'
Both active branches create the same backend chardev. Since there is no
other case, extract it before the switch so that we don't have to
duplicate it.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Wed, 13 May 2020 18:41:45 +0000 (20:41 +0200)]
qemu: domain: Forbid unsupported 'tftp' protocol and handle tests
'tftp' storage protocol was supported by qemu until 2.7.0. Add an
interlock when blockdev is used and drop the test case for it as it's
IMO not worth adding another test file just for that.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 18:02:28 +0000 (20:02 +0200)]
qemumonitorjsontest: Mark recently deprecated migration command in our tests
"migrate_set_downtime", "migrate_set_speed", and
"query-migrate-cache-size" were marked as deprecated in the QMP schema
in qemu 5.0. Since libvirt still actively uses them we must not mark
them as okay to be missing, but still mark them as deprecated, so that
we can add tests for deprecated commands.
The replacement of the command usage in libvirt is tracked by:
https://bugzilla.redhat.com/show_bug.cgi?id=1829543
https://bugzilla.redhat.com/show_bug.cgi?id=1829544
https://bugzilla.redhat.com/show_bug.cgi?id=1829545
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 17:59:51 +0000 (19:59 +0200)]
qemumonitorjsontest: Allow use of deprecated 'cpu-add' and 'change' command
Modify the generated test cases for the 'cpu-add' and 'change' command
which are deprecated by qemu. We now use device-add and
blockdev-change-media instead so we are okay if they will be removed.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 17:57:28 +0000 (19:57 +0200)]
qemumonitorjsontest: Allow use of deprecated 'query-cpus'
The command was replaced with 'query-cpus-fast' which is always used
when detected by the capabilities so we can allow our test usage of
the deprecated command even if it will be removed from the schema.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 17:41:38 +0000 (19:41 +0200)]
qemumonitorjsontest: Add infrastructure for generated tests of deprecated commands
For sanity-chcecking of deprecated commands which are still used on some
old code paths which used the simple generated test cases add a
mechanism to mark them as deprecated so schema checking can be skipped.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 17:20:00 +0000 (19:20 +0200)]
testQemuHotplugCpuPrepare: Allow deprecated commands for non-modern cpu hotplug test
We have a few cases validating that the code behaves correctly in
pre-modern hotplug era. This is controled by the 'modern' flag for the
test. Since 'cpu-add' command is now deprecated in qemu, there is a
modern replacement for it, and the test output is checked against
expected commands we can skip schema validation for the legacy command.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 15:31:20 +0000 (17:31 +0200)]
testutilsqemuschema: Pass in 'schema' and 'debug' variables to workers in a struct
Refactor the code so that we pass in the repeated 'schema' and 'debug'
arguments via a new struct testQEMUSchemaValidateCtxt. This will
simplify adding new parameters in the future.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Upcoming patches will add validation which rejects objects with the
'deprecated' feature in the QMP schema. To support tests which deal with
legacy properties in case when a command or argument is marked as
deprecated or removed by qemu qemuMonitorTestSkipDeprecatedValidation
will allow configuring the tests to ignore such errors.
In case of commands/features which are not yet replaced, the
'allowRemoved' bool should not be set to provide a hard notification
once qemu drops the command.
Note that at this point 'allowRemoved' only includes whole commands, but
not specific properties.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Peter Krempa [Wed, 29 Apr 2020 08:38:16 +0000 (10:38 +0200)]
qemuMonitorTestProcessCommandDefaultValidate: Clean up return value use
We no longer return the error via the monitor, so the function no longer
returns '1'. Remove the mention from comment and fix callers to stop
looking for the return value of '1'.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Ján Tomko [Tue, 19 May 2020 08:56:57 +0000 (10:56 +0200)]
qemu: conf: fix stray comma
The qemu.conf change broke our augeas test:
qemu/test_libvirtd_qemu.aug:96.3-203.1:exception thrown in test
qemu/test_libvirtd_qemu.aug:96.8-.34:exception: Iterated lens matched less than it should
Lens: ../../src/qemu/libvirtd_qemu.aug:170.13-.43:
Last match: ../../src/qemu/libvirtd_qemu.aug:18.52-.113:
Not matching: ../../src/qemu/libvirtd_qemu.aug:12.19-.31:
Error encountered at 48:27 (1615 characters into string)
<\n "/dev/ptmx", "/dev/kvm"|=|,\n]\nsave_image_format = "raw>
Fixes: ab5ba57012e9e6ab4f55afdeecd1813dd3ca916b Signed-off-by: Ján Tomko <jtomko@redhat.com>
Paolo Bonzini [Mon, 18 May 2020 23:06:59 +0000 (01:06 +0200)]
qemu: do not allow /dev/rtc or /dev/hpet access via the devices cgroup
The RTC and HPET modes for the QEMU emulation tick have been dropped
almost 9 years ago, in commit 25f3151ece1d5881826232bebccc21b588d4e03e.
Do not allow them in the devices cgroup policy.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Laine Stump [Thu, 14 May 2020 18:01:47 +0000 (14:01 -0400)]
qemu: prevent attempts to detach a device on a controller with hotplug='off'
Although the original patches to support controllers with
hotplug='off' were checking during hotplug/attach requests that the
device was being plugged into a PCI controller that didn't have
hotplug disabled, but I forgot to do the same for device detach (the
main impetus for adding the feature was to prevent unplugs originating
from within the guest, so it slipped my mind). So although the guest
OS was ultimately unable to honor the unplug request, libvirt could
still be used to make such a request, and since device attach/detach
are asynchronous operations, the caller to libvirt would receive a
success status back (the device would stubbornly/correctly remain in
the domain status XML however)
This patch remedies that, by looking at the controller for the device
in the detach request, and immediately failing the operation if that
controller has hotplug=off.
Signed-off-by: Laine Stump <laine@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
Michal Privoznik [Mon, 18 May 2020 13:11:49 +0000 (15:11 +0200)]
qemuProcessStop: Reattach NVMe disks a domain is mirroring into
If the mirror destination is not a file but a NVMe disk, then
call qemuHostdevReAttachOneNVMeDisk() to reattach the NVMe back
to the host.
This would be done by blockjob code when the job finishes, but in
this case the job won't finish - QEMU is killed meanwhile.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1825785 Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Michal Privoznik [Mon, 18 May 2020 13:07:46 +0000 (15:07 +0200)]
qemuProcessStop: Use XATTRs to restore seclabels on disks a domain is mirroring into
In v5.10.0-rc1~42 (which was later fixed in v6.0.0-rc1~487) I am
removing XATTRs for a file that QEMU is mirroring a disk into but
it is killed meanwhile. Well, we can call
qemuSecurityRestoreImageLabel() which will not only remove XATTRs
but also use them to restore the original owner of the file.
This would be done by blockjob code when the job finishes, but in
this case the job won't finish - QEMU is killed meanwhile
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Michal Privoznik [Thu, 14 May 2020 11:56:54 +0000 (13:56 +0200)]
qemuBuildNumaArgStr: Use modern -numa memdev= if old -numa mem= is unsupported
In previous commit we started tracking whether QEMU supports
'-numa mem='. This is tied to the machine type because migration
from '-numa mem=' to '-numa memdev' is impossible (or vice
versa). But since it's tied to a machine type (where migration
from one to another is also unsupported) we can allow QEMU to get
rid of the deprecated command line.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1783355 Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Thu, 14 May 2020 11:59:34 +0000 (13:59 +0200)]
qemuBuildNumaArgStr: Switch order of if() and for()
When building -numa command line there is a for() loop that
builds '-numa memdev=' for each guest NUMA node. And also
records in a local variable whether any of memory-object-*
backends must be used to satisfy desired config. Well, instead of
checking in each iteration whether corresponding capabilities are
set, we can do swap if() and for() and check only once.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Thu, 14 May 2020 11:11:01 +0000 (13:11 +0200)]
qemu: Track numa-mem-supported machine attribute
There is 'numa-mem-supported' machine attribute which specifies
whether '-numa mem=' is supported. Store it in our capabilities
as it will be used in later commits when building the command
line.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Ján Tomko <jtomko@redhat.com>
Michal Privoznik [Mon, 18 May 2020 08:07:30 +0000 (10:07 +0200)]
Don't require secdrivers to implement .domainMoveImageMetadata
The AppArmor secdriver does not use labels to grant access to
resources. Therefore, it doesn't use XATTRs and hence it lacks
implementation of .domainMoveImageMetadata callback. This leads
to a harmless but needless error message appearing in the logs:
virSecurityManagerMoveImageMetadata:476 : this function is not
supported by the connection driver: virSecurityManagerMoveImageMetadata
Closes: https://gitlab.com/libvirt/libvirt/-/issues/25 Signed-off-by: Michal Privoznik <mprivozn@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
projects / libvirt.git / log