If domain is killed with `xl destroy`, libvirt will not notice it and
still report the domain as running. Also trying to destroy the domain
through libvirt will fail. The only way to recover from such a situation
is to restart libvirt daemon. The problem is that even though libxl
report LIBXL_EVENT_TYPE_DOMAIN_DEATH, libvirt ignore it as all the
domain cleanup is done in a function actually destroying the domain. If
destroy is done outside of libvirt, there is no place where it would be
handled.
Fix this by doing domain cleanup in LIBXL_EVENT_TYPE_DOMAIN_DEATH too.
To avoid doing it twice, add a ignoreDeathEvent flag
libxlDomainObjPrivate, set when the domain death is triggered by libvirt
itself.
Signed-off-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jim Fehlig <jfehlig@suse.com>
Since domain was suspended before and on failed wakeup is destroyed,
send an event.
Also, add missing libxlDomainCleanup.
Signed-off-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jim Fehlig <jfehlig@suse.com>
Commit 017dfa27d changed a few switch statements in the LXC code to
have all possible enum values, and in the process changed the switch
statement in virLXCControllerGetNICIndexes() to return an error status
for unsupported interface types, but it erroneously put type='direct'
on the list of unsupported types.
type='direct' (implemented with a macvlan interface) is supported on
LXC, but it's interface shouldn't be placed on the list of interfaces
given to CreateMachineWithNetwork() because the interface is put
inside the container, while CreateMachineWithNetwork() only wants to
know about the parent veths of veth pairs (the parent veth remains on
the host side, while the child veth is put into the container).
Resolves: https://bugzilla.redhat.com/1656463
Signed-off-by: Laine Stump <laine@laine.org>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
virLXCControllerGetNICIndexes() was deciding whether or not to add the
ifindex for an interface's ifname to the list of ifindexes sent to
CreateMachineWithNetwork based on the interface type stored in the
config. This would be incorrect in the case of <interface
type='network'> where the network was giving out macvlan interfaces
tied to a physical device (i.e. when the actual interface type was
"direct").
Instead of checking the setting of "net->type", we should be checking
the setting of virDomainNetGetActualType(net).
I don't think this caused any actual misbehavior, it was just
technically wrong.
Signed-off-by: Laine Stump <laine@laine.org>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Most of the iptables APIs share code for the add/delete paths, but a
couple were separated. Merge the remaining APIs to facilitate future
changes.
Reviewed-by: Laine Stump <laine@laine.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Add support for converting openvswitch interface configuration
to/from libvirt domXML and xl.cfg(5). The xl config syntax for
virtual interfaces is described in detail in the
xl-network-configuration(5) man page. The Xen Networking wiki
also contains information and examples for using openvswitch
in xl.cfg config format
https://wiki.xenproject.org/wiki/Xen_Networking#Open_vSwitch
Tests are added to check conversions of openvswitch tagged and
trunked VLAN configuration.
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
It is currently possible to use <interface>s of type openvswitch
with the libxl driver in a non-standard way, e.g.
<interface type='bridge'>
<source bridge='ovsbr0'/>
<mac address='00:16:3e:7a:35:ce'/>
<script path='vif-openvswitch'/>
</interface>
This patch adds support for openvswitch <interface>s specified
in typical libvirt config
<interface type='bridge'>
<source bridge='ovsbr0'/>
<mac address='00:16:3e:7a:35:ce'/>
<virtualport type='openvswitch'/>
</interface>
VLAN tags and trunking are also supported using the extended
syntax for specifying an openvswitch bridge in libxl
BRIDGE_NAME[.VLAN][:TRUNK:TRUNK]
See Xen's networking wiki for more details on openvswitch support
https://wiki.xenproject.org/wiki/Xen_Networking#Open_vSwitch
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Commit 212dc9286 made a generic qemuDomainGetIOThreadsMon which
would fail if the QEMU_CAPS_OBJECT_IOTHREAD didn't exist. Then
commit d1eac927 used that helper for the collection of all domain
stats. However, if the capability doesn't exist, then the entire
stats collection fails. Since the IOThread stats were meant to be
if available only, thus rather than failing if the capability
doesn't exist, let's just not collect the stats. Restore the caps
failure logic for qemuDomainGetIOThreadsLive.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
During qemuConnectGetAllDomainStats if qemuDomainGetStats causes
a failure, then when collecting more than one domain's worth of
statistics the loop in virDomainStatsRecordListFree would call
virDomainFree which would call virResetLastError effectively wiping
out the reason we failed leaving the caller with no idea why the
collection failed.
To fix this, let's Preserve the error and Restore it prior to return
so that a caller such as 'virsh domstats' doesn't get the generic
"error: An error occurred, but the cause is unknown".
Signed-off-by: John Ferlan <jferlan@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
We already have a way stricter check in the code which is doing the
snapshot so duplicating it in the parser does not make much sense. Also
gets rid of an ugly ternary operator.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
We are preparing a certain disk source passed in as '@src' so the
individual functions should use that rather than disk->src which
corresponds to the top level element of the chain only.
Without this change TLS and persistent reservations would not work for
backing images of a chain when using -blockdev.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
The function clears and frees the passed buffers on success, but not in
one case of failure. Modify the control flow that the args are always
consumed, record it in the docs and remove few pointless cleanup paths
in callers.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1656014
An RNG device can consists of more devices than RND device
itself. For instance, in case of EGD there is a chardev that
connects to EGD daemon and feeds the qemu with random data. When
doing RNG device removal we have to remove the associated chardev
as well.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
The way that the code is currently written makes my eyes hurt.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
There are two functions called from syncNicRxFilterMultiMode:
virNetDevSetRcvAllMulti() and virNetDevSetRcvMulti(). Both of
them return 0 on success and -1 on error. However, currently
their return value is checked for != 0 which conflicts with our
assumptions on retvals: a positive value is still considered
success but with current check it would lead to failure.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Depending on whether QEMU actually supports the option, we can put the
'rendernode' on the '-display egl-headless' cmdline.
https://bugzilla.redhat.com/show_bug.cgi?id=1628892
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Just like for SPICE, we need to change the permissions on the DRI device
used as the @rendernode for egl-headless graphics type.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Just like for SPICE, we need to put the render node DRI device into the
device cgroup list so that users don't need to add it manually via
qemu.conf file.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Just like for SPICE, we need to put the DRI device into the namespace,
otherwise it will be left out from the DAC relabeling process.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Unlike with SPICE and SDL which use the <gl> subelement to enable OpenGL
acceleration, specifying egl-headless graphics in the XML has
essentially the same meaning, thus in case of egl-headless we don't have
a need for the 'enable' element attribute and we'll only be interested
in the 'rendernode' one further down the road.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Since we need to specify the rendernode option onto QEMU cmdline, we
need this union member to retain consistency in how we build the
cmdline.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Now that we have QAPI introspection of display types in QEMU upstream,
we can check whether the 'rendernode' option is supported with
egl-headless display type.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
We're going to need a bit more logic for egl-headless down the road so
prepare a helper just like for the other display types.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Up until now, we formatted 'rendernode=' onto QEMU cmdline only if the
user specified it in the XML, otherwise we let QEMU do it for us. This
causes permission issues because by default the /dev/dri/renderDX
permissions are as follows:
crw-rw----. 1 root video
There's literally no reason why it shouldn't be libvirt picking the DRM
render node instead of QEMU, that way (and because we're using
namespaces by default), we can safely relabel the device within the
namespace.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
A few simple helpers that allow us to determine whether a graphics can
and will need to make use of a DRM render node.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
This is the first step towards libvirt picking the first available
render node instead of QEMU. It also makes sense for us to be able to do
that, since we allow specifying the node directly for SPICE, so if
there's no render node specified by the user, we should pick the first
available one. The algorithm used for that is essentially the same as
the one QEMU uses.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Guest network devices can set 'overflow' when there are a number of multicast
ips configured. For virtio_net, the limit is only 64. In this case, the list
of mac addresses is empty and the 'overflow' condition is set. Thus, the guest
will currently receive no multicast traffic in this state.
When 'overflow' is set in the guest, let's turn this into ALLMULTI on the host.
Signed-off-by: Jason Baron <jbaron@akamai.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Michal Privoznik <mprivozn@redhat.com>
Support for armv6l qemu guests has been added.
Tested with arm1176 CPU on x86.
Signed-off-by: Stefan Schallenberg <infos@nafets.de>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Post-copy migration has been broken on the source since commit
v3.8.0-245-g32c29f10db which implemented support for
pause-before-switchover QEMU migration capability.
Even though the migration itself went well, the source did not really
know when it switched to the post-copy mode despite the messages logged
by MIGRATION event handler. As a result of this, the events emitted by
source libvirtd were not accurate and statistics of the completed
migration would cover only the pre-copy part of migration. Moreover, if
migration failed during the post-copy phase for some reason, the source
libvirtd would just happily resume the domain, which could lead to disk
corruption.
With the pause-before-switchover capability enabled, the order of events
emitted by QEMU changed:
pause-before-switchover
disabled enabled
MIGRATION, postcopy-active STOP
STOP MIGRATION, pre-switchover
MIGRATION, postcopy-active
The STOP even handler checks the migration status (postcopy-active) and
sets the domain state accordingly. Which is sufficient when
pause-before-switchover is disabled, but once we enable it, the
migration status is still active when we get STOP from QEMU. Thus the
domain state set in the STOP handler has to be corrected once we are
notified that migration changed to postcopy-active.
This results in two SUSPENDED events to be emitted by the source
libvirtd during post-copy migration. The first one with
VIR_DOMAIN_EVENT_SUSPENDED_MIGRATED detail, while the second one reports
the corrected VIR_DOMAIN_EVENT_SUSPENDED_POSTCOPY detail. This is
inevitable because we don't know whether migration will eventually
switch to post-copy at the time we emit the first event.
https://bugzilla.redhat.com/show_bug.cgi?id=1647365
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Both VIR_DOMAIN_FEATURE_HPT and VIR_DOMAIN_FEATURE_HTM are
handled in the exact same way, so we can remove some duplicated
code without losing any functionality.
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
The call of virResctrlMonitorGetStats will allocate the memory for
holding cache occupancy or memory bandwidth statistics.
This patch adds the function virResctrlMonitorFreeStats as the
opposing action of virResctrlMonitorGetStats to free the memory.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Return a list of virResctrlMonitorStatsPtr instead of
a virResctrlMonitorStats array in virResctrlMonitorGetStats.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Handle PVH domain type in both directions (xen-xl->xml, xml->xen-xl).
And add a test for it.
Signed-off-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jim Fehlig <jfehlig@suse.com>
builder="hvm" is deprecated since Xen 4.10, new syntax is type="hvm" (or
type="pv", which is default). Since the old one is still supported,
still use it when writing native config, so the config will work on
older Xen too (and will also not complicate tests).
Signed-off-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jim Fehlig <jfehlig@suse.com>
Since this is something between PV and HVM, it makes sense to put the
setting in place where domain type is specified.
To enable it, use <os><type machine="xenpvh">xenpvh</type></os>. It is
also included in capabilities.xml, for every supported HVM guest type - it
doesn't seems to be any other requirement (besides new enough Xen).
Signed-off-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jim Fehlig <jfehlig@suse.com>
Make it easier to share HVM and PVH code where relevant. No functional
change.
Signed-off-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jim Fehlig <jfehlig@suse.com>
The test driver state (@testDriver) uses it's own reference counting
and locking implementation. Instead of doing that, convert @testDriver
into a virObjectLockable and use the provided functionalities.
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
There are certain cases e.g. containers where the sysfs path might
exists, but might fail. Unfortunately the exact restrictions are only
known to libvirt when trying to write to it so we need to try it.
But in case it fails there is no need to fully abort, in those cases try
to fall back to the older ioctl interface which can still work.
That makes setting up a bridge in unprivileged LXD containers work.
Fixes: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1802906
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Laine Stump <laine@laine.org>
Reported-by: Brian Candler <b.candler@pobox.com>
If migration is cancelled or confirm phase fails the domain
should be kept on the source even if VIR_MIGRATE_UNDEFINE_SOURCE
was requested.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
There are some checks done when parsing a migration cookie. For
instance, one of the checks ensures that the domain is not being
migrated onto the same host. If that is the case, then we are in
big trouble because the @vm is the same domain object used by
source and it has some jobs sets and everything so recovering
from failed cookie parsing would be needlessly hard.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
The function currently takes virDomainObjPtr because it's using
both: the domain definition and domain private data.
Unfortunately, this means that in prepare phase we can't parse
migration cookie before putting incoming domain def onto domain
objects list (addressed in the very next commit). Change the
arguments so that virDomainDef and private data are passed
instead of virDomainObjPtr.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
There are several functions called in the cleanup path. Some of
them do report error (e.g. qemuDomainRemoveInactiveJob()) which
may result in overwriting an error reported earlier with some
less useful message.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
virt-aa-helper needs to grant QEMU access to VFIO MDEV devices.
This extends commit 74e86b6b which only covered PCI hostdevs for VFIO-PCI
assignment by now also covering vfio MDEVs.
It has still the same limitations regarding the device lifecycle, IOW we're
unable to predict the actual VFIO device being created, thus we need
wildcards.
Also note that the hotplug case, where apparmor is able to detect the actual
VFIO device during runtime, is already covered by commit 606afafb.
Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
The path argument of virFileIsDir should be a full name
of file, pathname and filename. Fixed it by passing the
full path name to virFileIsDir.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Since the functions only return 0 or 1, they should return bool. I missed the
change when "refactoring" the first commit.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1545732
Implement the QEMU driver mechanism in order to set the polling
parameters for an IOThread within the bounds specified by the
QEMU qapi parameter passing.
Based heavily on patches originally posted by Pavel Hrdina
<phrdina@redhat.com>, but modified to only handle alterations
for a running guest. For the most part the API names changed,
the typed parameters removed the poll enabled value, and the
capabilities check was moved to just before the live attempt
to set. Since changes are only supported for a running guest,
no guest XML alterations were kept.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Add a capability check for IOThread polling (all were added at the
same time, so only one check is necessary).
Based on code originally posted by Pavel Hrdina <phrdina@redhat.com>
with the only changes to include the more recent QEMU releases.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Rather than passing an iothread_id, let's pass a qemuMonitorIOThreadInfo
structure so that a subsequent change to modify the iothread info can
just generate and pass one.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
We're about to add a new state "modify" and thus the function
goes from just Add/Del. Use an enum to manage.
Extracted from code originally posted by Pavel Hrdina
<phrdina@redhat.com>, but placed into a separate patch.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Add functions to set the IOThreadInfo param data for the live guest.
Modify the _qemuMonitorIOThreadInfo to have a flag to indicate when
a value was set so that we don't set a value unless it was desired
to be set.
Based on code originally posted by Pavel Hrdina <phrdina@redhat.com>,
but extracted into a separate patch. Note that qapi expects to receive
integer parameters rather than unsigned long long or unsigned int's.
QEMU does save the value in larger signed 64 bit values eventually.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Create a new API that will allow an adjustment of IOThread
polling parameters for the specified IOThread. These parameters
will not be saved in the guest XML. Currently the only parameters
supported will allow the hypervisor to adjust the parameters used
to limit and alter the scope of the polling interval. The polling
interval allows the IOThread to spend more or less time processing
in the guest.
Based on code originally posted by Pavel Hrdina <phrdina@redhat.com>
to add virDomainAddIOThreadParams and virDomainModIOThreadParams.
Modification of those changes to use virDomainSetIOThreadParams
instead and remove concepts related to saving the data in guest
XML as well as the way to specifically enable the polling parameters.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Process the IOThreads polling stats if available. Generate the
output params record to be returned to the caller with the three
values - poll-max-ns, poll-grow, and poll-shrink.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Separate out the fetch of the IOThread monitor call into a separate
helper so that a subsequent domain statistics change can fetch the raw
IOThread data and parse it as it sees fit.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
If there are IOThread polling values in the query-iothreads return
buffer, then fill them in and set a bool indicating their presence.
This will allow for displaying in a domain stats output eventually.
Note that the QEMU values are managed a bit differently (as int's
stored in int64_t's) than we will manage them (as unsigned long and
int values). This is intentional to allow for value validation
checking when it comes time to provide the values to QEMU.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
virXMLFormatElement() might fail, but we were not checking
its return value.
Fixing this requires us to change virDomainDeviceInfoFormat()
so that it can report an error back to the caller.
Introduced-by: 0d6b87335c
Spotted-by: Coverity
Reported-by: John Ferlan <jferlan@redhat.com>
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
In many cases, an early exit from a function would cause
memory allocated by local virBuffer instances not to be
released.
Provide proper cleanup paths to solve the issue.
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
This avoids setting 'ret' multiple times, which will result
in errors being masked if the first operation fails but the
second one succeeds.
Introduced-by: f183b87fc1
Spotted-by: Coverity
Reported-by: John Ferlan <jferlan@redhat.com>
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Trying to use virlockd to lock metadata turns out to be too big
gun. Since we will always spawn a separate process for relabeling
we are safe to use thread unsafe POSIX locks and take out
virtlockd completely out of the picture.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
When metadata locking is enabled that means the security commit
processing will be run in a fork similar to how namespaces use fork()'s
for processing. This is done to ensure libvirt can properly and
synchronously modify the metadata to store the original owner data.
Since fork()'s (e.g. virFork) have been seen as a performance bottleneck
being able to disable them allows the admin to choose whether the
performance 'hit' is worth the extra 'security' of being able to
remember the original owner of a lock.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
For metadata locking we might need an extra fork() which given
latest attempts to do fewer fork()-s is suboptimal. Therefore,
there will be a qemu.conf knob to {en|dis}able this feature. But
since the feature is actually not metadata locking itself rather
than remembering of the original owner of the file this is named
as 'rememberOwner'. But patches for that feature are not even
posted yet so there is actually no qemu.conf entry in this patch
nor a way to enable this feature.
Even though this is effectively a dead code for now it is still
desired.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The TPM code currently accepts pointer to a domain definition.
This is okay for now, but in near future the security driver APIs
it calls will require domain object. Therefore, change the TPM
code to accept the domain object pointer.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Both virProcessRunInMountNamespace() and virProcessRunInFork()
look very similar. De-duplicate the code and make
virProcessRunInMountNamespace() call virProcessRunInFork().
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
This new helper can be used to spawn a child process and run
passed callback from it. This will come handy esp. if the
callback is not thread safe.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Add a new memoryBacking source type "memfd", supported by QEMU (when
the capability is available).
A memfd is a specialized anonymous memory kind. As such, an anonymous
source type could be automatically using a memfd. However, there are
some complications when migrating from different memory backends in
qemu (mainly due to the internal object naming at this point, but
there could be more). For now, it is simpler and safer to simply
introduce a new source type "memfd". Eventually, the "anonymous" type
could learn to use memfd transparently in a separate change.
The main benefits are that it doesn't need to create filesystem files,
and it also enforces sealing, providing a bit more safety.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
QEMU 3.1 should only expose the property if the host is actually
capable of creating hugetable-backed memfd. However, it may fail
at runtime depending on requested "hugetlbsize".
Reviewed-by: John Ferlan <jferlan@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: John Ferlan <jferlan@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Commit ("qemu_domain.c: moving maxCpu validation to
qemuDomainDefValidate") shortened the code of qemuProcessStartValidateXML.
The function is called only by qemuProcessStartValidate, in the
same file, and its code is now a single check that calls virDomainDefValidate.
Instead of leaving a function call just to execute a single check,
this patch puts the check in the body of qemuProcessStartValidate in the
place where qemuProcessStartValidateXML was being called. The function can
now be removed.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Previous patch removed the call to qemuProcessValidateCpuCount
from qemuProcessStartValidateXML, in qemu_process.c. The only
caller left is qemuDomainDefValidate, in qemu_domain.c.
Instead of having a public function declared inside qemu_process.c
that isn't used in that file, this patch moves the function to
qemu_domain.c, making in static and renaming it to
qemuDomainValidateCpuCount to be compliant with other static
functions names in the file.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Adding maxCpu validation in qemuDomainDefValidate allows the user to
spot over the board maxCpus counts at editing time, instead of
facing a runtime error when starting the domain. This check is also
arch independent.
This leaves us with 2 calls to qemuProcessValidateCpuCount: one in
qemuProcessStartValidateXML and the new one at qemuDomainDefValidate.
The call in qemuProcessStartValidateXML is redundant. Following
up in that code, there is a call to virDomainDefValidate, which
in turn will call config.domainValidateCallback. In this case, the
callback function is qemuDomainDefValidate. This means that, on startup
time, qemuProcessValidateCpuCount will be called twice.
To avoid that, let's also remove the qemuProcessValidateCpuCount call
from qemuProcessStartValidateXML.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
qemuValidateCpuCount validates the maxCpus value of a domain at
startup time, preventing it to start if the value exceeds a maximum.
This checking is also done at qemu_domain.c, qemuDomainDefValidate.
However, it is done only for x86 (and even then, in a specific
scenario). We want this check to be done for all archs.
To accomplish this, let's first make qemuValidateCpuCount public so
it can be used inside qemuDomainDefValidate. The function was renamed
to qemuProcessValidateCpuCount to be compliant with the other public
methods at qemu_process.h. The method signature was slightly adapted
to fit the const 'def' variable used in qemuDomainDefValidate. This
change has no downside in in its original usage at
qemuProcessStartValidateXML.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Adding the maxCpus value in the error message of qemuValidateCpuCount
allows the user to set an acceptable maxCpus count without knowing
QEMU internals.
x86 guests, that might have been created prior to the x86
qemuDomainDefValidate maxCpus check code (that validates the maxCpus value
in editing time), will also benefit from this change.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The commit 89563efc02 fix the
monitor error when closing the QEMU monitor. The QEMU agent
has a problem similar to QEMU monitor. So fix the QEMU agent
with the same method.
Signed-off-by: Wang Yechao <wang.yechao255@zte.com.cn>
Reviewed-by: John Ferlan <jferlan@redhat.com>
We were mistakenly skipping virZPCIDeviceAddressIsEmpty() and
virZPCIDeviceAddressIsValid() when compiling on non-Linux,
which unsurprisingly ended up causing linking failures later
in the build process.
Clue-stick-by: Peter Krempa <pkrempa@redhat.com>
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
This commit adds hotplug support for PCI devices on S390 guests.
There's no need to implement hot unplug for zPCI as QEMU implements
an unplug callback which will unplug both PCI and zPCI device in a
cascaded way.
Currently, the following PCI devices are supported:
virtio-blk-pci
virtio-net-pci
virtio-rng-pci
virtio-input-host-pci
virtio-keyboard-pci
virtio-mouse-pci
virtio-tablet-pci
vfio-pci
SCSIVhost device
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Add new functions to generate zPCI command string and append it to
QEMU command line. And the related tests are added.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
This patch adds new functions for reservation, assignment and release
to handle the uid/fid. If the uid/fid is defined in the domain XML,
they will be reserved directly in the collecting phase. If any of them
is not defined, we will find out an available value for them from the
zPCI address hashtable, and reserve them. For the hotplug case there
might not be a zPCI definition. So allocate and reserve uid/fid the
case. Assign if needed and reserve uid/fid for the defined case.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
We should ensure that QEMU supports zPCI when a zPCI address is defined
in XML and otherwise report an error. This patch introduces a generic
validation function qemuDomainDeviceDefValidateAddress() which calls
qemuDomainDeviceDefValidateZPCIAddress() if address type is PCI address.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
This patch introduces new XML parser/formatter functions. Uid is
16-bit and non-zero. Fid is 32-bit. They are the two attributes of zpci
which is introduced as PCI address element. Zpci element is parsed and
formatted along with PCI address. And add the related test cases.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
In order to add zPCI child element for PCI address, we update
virDomainDeviceInfoFormat() to format device info by helper function
virXMLFormatElement(). Then we could simply format zPCI address into
child buffer later.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
The pci-root depends on zpci capability. So autogenerate pci-root if
zpci exists.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
This patch provides a caching mechanism for the device address
extensions uid and fid on S390. For efficient sparse address allocation,
we introduce two hash tables for uid/fid which hold the address set
information per domain. Also in order to improve performance of
searching available value, we introduce our own callbacks for the two
hashtables. In this way, uid/fid is saved in hash key and hash value
could be any non-NULL pointer due to no operation on hash value. That is
also the reason why we don't introduce hash value free callback.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
This patch introduces PCI address extension flag for virDomainDeviceInfo
and virPCIDeviceAddress. The extension flag in virDomainDeviceInfo is
used internally during calculating PCI extension flag. The one in
virPCIDeviceAddress is the duplicate to indicate extension address is
being used. Currently only zPCI extension address is introduced to deal
with 'uid' and 'fid' on the S390 platform.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
QEMU on s390 supports PCI multibus since forever.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Let's introduce zPCI capability.
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Add zPCI definitions in preparation of extending the PCI address
with parameters uid (user-defined identifier) and fid (PCI function
identifier).
Signed-off-by: Yi Min Zhao <zyimin@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Stefan Zimmermann <stzi@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Support Hyper-V Enlightened VMCS in domain config. QEMU support will
be implemented in the next patch, adding interim VIR_DOMAIN_HYPERV_EVMCS
cases to src/qemu/* for now.
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
QEMU 3.1 supports Hyper-V-style PV IPIs making it cheaper for Windows
guests to send an IPI, especially when it targets many CPUs.
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Support Hyper-V PV IPI enlightenment in domain config. QEMU support will
be implemented in the next patch, adding interim VIR_DOMAIN_HYPERV_IPI
cases to src/qemu/* for now.
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
This is a simple removal of a duplicated check of the return of the
filter function. There is a nested conditional checking exactly the same
thing since commit c9ede1cf removed the (ret > 0) check condition.
Signed-off-by: Julio Faracco <jcfaracco@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The function qemuDomainGetHostdevPath() is using VIR_FREE to free the
paths stored in tmpPaths. Both syntax analyzer are reporting a warning
about this. Replacing the old method to function
virStringListFreeCount() fixes the warnings/errors.
Signed-off-by: Julio Faracco <jcfaracco@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
This patch introduce the new settings for LXC 3.0 or higher. The older
versions keep the compatibility to deprecated settings for LXC, but
after release 3.0, the compatibility was removed. This commit adds the
support to the refactored settings.
Signed-off-by: Julio Faracco <jcfaracco@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Commit 901d2b9c introduced virCgroupGetMemoryStat and replaced
the LXC virLXCCgroupGetMemStat logic in commit e634c7cd0. However,
in doing so the replacement wasn't exact as the LXC logic used
getline() to process the cgroup controller data, while the new
virCgroupGetMemoryStat used "memory.stat" manual buffer read/
processing which neglected to forward through @line in order
to read each line in the output.
To fix that, we should be sure to carry forward the @line value
for each line read updating it beyond that current @newLine value
once we've calculated the values that we want.
Signed-off-by: Peter Chubb <peter.chubb@data61.csiro.au>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1631622
If polkit authentication is enabled, an attempt to open
the connection failed during virAccessDriverPolkitGetCaller
when the call to virIdentityGetCurrent returned NULL resulting
in the errors:
virAccessDriverPolkitGetCaller:87 : access denied:
Policy kit denied action org.libvirt.api.connect.getattr from <anonymous>
Because qemuProcessReconnect runs in a thread during
daemonRunStateInit processing it doesn't have the thread
local identity. Thus when the virGetConnectNWFilter is
called as part of the qemuProcessFiltersInstantiate when
virDomainConfNWFilterInstantiate is run the attempt to get
the idenity fails and results in the anonymous error above.
To fix this, let's grab/use the virIdenityPtr of the process
that will be creating the thread, e.g. what daemonRunStateInit
has set and use that for our thread. That way any other similar
processing that uses/requires an identity for any other call
that would have previously been successfully run won't fail in
a similar manner.
Signed-off-by: John Ferlan <jferlan@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1631606
Changes made to manage and utilize a secondary connection
driver to APIs outside the scope of the primary connection
driver have resulted in some confusion processing polkit rules
since the simple "access denied" error message doesn't provide
enough of a clue when combined with the "authentication failed:
access denied by policy" as to which connection driver refused
or failed the ACL check.
In order to provide some context, let's modify the existing
"access denied" error returned from the various vir*EnsureACL
API's to provide the connection driver name that is causing
the failure. This should provide the context for writing the
polkit rules that would allow access via the driver, but yet
still adhere to the virAccessManagerSanitizeError commentary
regarding not telling the user why access was denied.
Signed-off-by: John Ferlan <jferlan@redhat.com>
This reverts commit ccc72d5cbd.
Based on upstream comment to a follow-up patch, this didn't take the
right approach and the right thing to do is revert and rework.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Missed during review and surprisingly my run through Coverity also
didn't see this. I only noticed it when reading the code while fixing
the build breaker for commit 36780a86a.
With all those continues we would leak @stats.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Refactoring qemuDomainGetStatsCpu, make it possible to add
more CPU statistics.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Add functions for creating, destroying, reconnecting resctrl
monitor in qemu according to the configuration in domain XML.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Introducing <monitor> element under <cachetune> to represent
a cache monitor.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Introduced virDomainResctrlNew to do the most part of virDomainResctrlAppend
and move the operation of appending resctrl to @def->resctrls out of
function.
Rather than rely on virDomainResctrlAppend to perform the allocation, move
the onus to the caller and make use of virBitmapNewCopy for @vcpus and
virObjectRef for @alloc, thus removing the need to set each to NULL after the
call.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Add interfaces monitor group to support operations such
as GetID, SetID, Remove, SetAlloc, etc.
Implement the internal virResctrlMonitorGetStats to fetch all
the statistical data and the virResctrlMonitorGetCacheOccupancy
in order to fetch the cache specific "llc_occupancy" value.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Refactor virResctrlAllocSetID generating an error if an attempt
is made to overwrite the existing value.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Add interface for creating the resource monitoring group according
to '@virResctrlMonitor->path'.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The code for creating resctrl allocation group could be reused
for monitoring group, refactor it for reuse in the later patch.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The code of adding PID to the allocation could be reused, refactor it
for later reuse.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Add interface for resctrl monitor to determine the path.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The code for determining resctrl allocation path could be reused
for monitor. Refactor it for reuse.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Cache Monitoring Technology (aka CMT) provides the capability
to report cache utilization information of system task.
This patch introduces the concept of resctrl monitor through
data structure virResctrlMonitor.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Refactor schemas and virresctrl to support optional <cache> element
in <cachetune>.
Later, the monitor entry will be introduced and to be placed
under <cachetune>. Either cache entry or monitor entry is
an optional element of <cachetune>.
An cachetune has no <cache> element is taking the default resource
allocating policy defined in '/sys/fs/resctrl/schemata'.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
When no server name is provided in the URI, modern versions of libxml2
will set the port to '-1'. This is a change from behaviour with earlier
versions which set it to 0.
Libvirt expects the port to be 0 in these cases and as a result we get a
bug when connecting to URIs which lack a server name:
$ virsh -c test+ssh:///default list
error: failed to connect to the hypervisor
error: Cannot recv data: Bad port '-1': Connection reset by peer
This libxml2 change was attempting to fix another bug identified by
libvirt where it didn't roundtrip URIs correctly in:
beb7281055
Essentially libxml2 was not expecting apps to look at the URI port
field when the server name is not provided. This was a reasonable
assumption, but none the less libvirt did look at it :-)
The fix is to ensure we explicitly set port to 0 when server name
is not present, avoiding undefined behaviour for the port field in
libxml2.
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
This patch introduces a new shutdown reason "daemon" in order
to indicate that the daemon needed to force shutdown the domain
as the best course of action to take at the moment.
This action would occur during reconnection when processing
encounters an error once the monitor reconnection is successful.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Return -1 and report an error message if no transaction is set and
virSecuritySELinuxTransactionCommit is called.
The function description of virSecuritySELinuxTransactionCommit says:
"Also it is considered as error if there's no transaction set and this
function is called."
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
In 4674fc6afd I've implemented transactions for selinux driver.
Well, now that I am working in this area I've noticed a subtle
bug: @ret is initialized to 0 instead of -1. Facepalm.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Marc Hartmayer <mhartmay@linux.ibm.com>
VFIO AP has a limitation on a single device per domain, however, when
commit 11708641 added the support for vfio-ap, check for this limitation
was performed as part of the post parse code. Generally, checks like that
should be performed within the driver's validation callback to eliminate
any slight chance of failing in post parse, which could potentially
result in the domain XML config vanishing.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Since we'll need to validate other models apart from VFIO PCI too,
having a helper for each model should keep the code base cleaner.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
There's a lot of stuff going on in src/conf/nodedev_conf which is
sometimes not directly related to config and we're not really consistent
with putting only parser/formatter related stuff here, e.g. like we do
for domains. So, let's start simply by adding a new module
node_device_util containing some of the helpers. Unfortunately, even
though these helpers tend to open a secondary driver connection and would
be much therefore better suited as a nodedev driver module, we can't do
that without pulling headers from the driver into conf/ and that's wrong
because we want conf/ to stay driver-agnostic.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Acked-by: Michal Privoznik <mprivozn@redhat.com>
The gotShutdown bool has been redundant since we started setting
VIR_DOMAIN_SHUTDOWN state after receiving SHUTDOWN event from QEMU.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
If gotShutdown is true, the domain state cannot be running because of
the following code in qemuProcessHandleShutdown:
priv->gotShutdown = true;
VIR_DEBUG("Transitioned guest %s to shutdown state",
vm->def->name);
virDomainObjSetState(vm,
VIR_DOMAIN_SHUTDOWN,
VIR_DOMAIN_SHUTDOWN_UNKNOWN);
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
On aarch64, lauch vm with the follow configuration:
<interface type="hostdev" managed="yes">
<mac address="fa:16:3e:14:41:00"/>
<source>
<address type="pci" domain="0x0000" bus="0x01" slot="0x0b" function="0x2"/>
</source>
</interface>
libvirtd will crash when accessing net->model.
Signed-off-by: Wang Yechao <wang.yechao255@zte.com.cn>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
If qemuDomainSnapshotDiscard() fails for any reason (rare,
but possible with an ill-timed ENOMEM or if
qemuDomainSnapshotForEachQcow2() has problems talking to the
qemu guest monitor), then an attempt to retry the snapshot
deletion API will crash because we didn't undo the effects
of virDomainSnapshotDropParent() temporarily rearranging the
internal list structures, and the second attempt to drop
parents will dereference NULL. Fix it by instead noting that
there are only two callers to qemuDomainSnapshotDiscard(),
and only one of the two callers wants the parent to be updated;
thus we can move the call to virDomainSnapshotDropParent()
into a code path that only gets executed on success.
Signed-off-by: Eric Blake <eblake@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Since commit v4.7.0-302-ge6d77a75c4 processing RESUME event is mandatory
for updating domain state. But the event handler explicitly ignored this
event in some cases. Thus the state would be wrong after a fake reboot
or when a domain was rebooted after it crashed.
BTW, the code to ignore RESUME event after SHUTDOWN didn't make sense
even before making RESUME event mandatory. Most likely it was there as a
result of careless copy&paste from qemuProcessHandleStop.
The corresponding debug message was clarified since the original state
does not have to be "paused" only and while we have a "resumed" event,
the state is called "running".
https://bugzilla.redhat.com/show_bug.cgi?id=1612943
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
The array "mount" inside lxc_container is not being checked before for
loop. Clang syntax scan is complaining about this segmentation fault.
Signed-off-by: Julio Faracco <jcfaracco@gmail.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The current qemuProcessReconnect logic paints a broad brush
determining that the shutdown reason must be crashed if it was
determined that the domain was started with -no-shutdown; however,
there's many other ways to get to the error label, so let's narrow
our reasoning window for using VIR_DOMAIN_SHUTOFF_CRASHED to the
period where we essentially know we've tried to create to the
monitor and before we were successful in opening the connection.
Failures that occur outside that window would thus be considered
as VIR_DOMAIN_SHUTOFF_UNKNOWN, at least for now.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
When qemuProcessReconnectHelper was introduced (commit d38897a5d)
reconnection failure used VIR_DOMAIN_SHUTOFF_FAILED; however, that
was changed in commit bda2f17d to either VIR_DOMAIN_SHUTOFF_CRASHED
or VIR_DOMAIN_SHUTOFF_UNKNOWN.
When QEMU_CAPS_NO_SHUTDOWN checking was removed in commit fe35b1ad6
the conditional state was just left at VIR_DOMAIN_SHUTOFF_CRASHED.
So introduce qemuDomainIsUsingNoShutdown which will manage the
condition when the domain was started with -no-shutdown so that
when/if reconnection failure occurs we can restore the decision
point used to determine whether CRASHED or UNKNOWN is provided.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
V2 of the libxl soft reset patch, which was pushed as commit da4b0fd9,
dropped the hunk that disposed of the libxl_domain_config object. Add
the missing hunk to properly dispose the object.
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
The pvops Linux kernel implements machine_ops.crash_shutdown as
static void xen_hvm_crash_shutdown(struct pt_regs *regs)
{
native_machine_crash_shutdown(regs);
xen_reboot(SHUTDOWN_soft_reset);
}
but currently the libxl driver does not handle the soft reset
shutdown event. As a result, the guest domain never proceeds
past xen_reboot(), making it impossible for HVM domains to save
a crash dump using kexec.
This patch adds support for handling the soft reset event by
calling libxl_domain_soft_reset() and re-enabling domain death
events, which is similar to the xl tool handling of soft reset
shutdown event.
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
There are too many goto labels in libxlDomainShutdownThread. Convert the
'destroy' and 'restart' labels to helper functions, leaving only the
commonly used pattern of 'endjob' and 'cleanup' labels.
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
In libxlDomainShutdownThread, virObjectEventStateQueue is needlessly
called in the destroy and restart labels. The cleanup label aready
queues whatever event was created based on libxl_shutdown_reason.
There is no need to handle destroy and restart differently.
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Our HACKING guide forbids these.
There's no point in exempting these from the spacing check
if their existence is against our coding style.
Note that the non-usage of these comments itself is not enforced
by syntax check, probably because of the need to implement a C parser.
Signed-off-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1631606
Changes made to manage and utilize a secondary connection
driver to APIs outside the scope of the primary connection
driver have resulted in some confusion processing polkit rules
since the simple "access denied" error message doesn't provide
enough of a clue when combined with the "authentication failed:
access denied by policy" as to which connection driver refused
or failed the ACL check.
In order to provide some context, let's modify the existing
"access denied" error returne from the various vir*EnsureACL
API's to provide the connection driver name that is causing
the failure. This should provide the context for writing the
polkit rules that would allow access via the driver.
Signed-off-by: John Ferlan <jferlan@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
Commit 57f5621f modified nwfilterInstantiateFilter to detect when
a filter binding was already present before attempting to add the
new binding and instantiate it. Additionally, the change to
nwfilterStateInitialize to call virNWFilterBindingObjListLoadAllConfigs
(from commit c21679fa3f) to load active domain filter bindings, but
not instantiate them eventually leads to a problem for the QEMU
driver reconnection logic after a daemon restart where the filter
bindings would no longer be instantiated.
Subsequent commit f14c37ce4c replaced the nwfilterInstantiateFilter
with virDomainConfNWFilterInstantiate which uses @ignoreExists to
detect presence of the filter and still did not restore the filter
instantiation call when making the new nwfilter bindings logic active.
Thus in order to instantiate any active domain filter, we will call
virNWFilterBuildAll with 'false' to indicate the need to go through
all the active bindings calling virNWFilterInstantiateFilter to
instantiate the filter bindings.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
Commit cdbe1332 neglected to document the API. So let's add some
details about the algorithm and why it was used to help future
readers understand the issues encountered.
NB: Management of the processing udev device notification is a
delicate balance between the udev process, the scheduler, and when
exactly the data from/for the socket is received. The balance is
particularly important for environments when multiple devices are
added into the system more or less simultaneously such as is done
for mdev or SRIOV. In these cases old libudev blocking on the udev
recv() occurs more frequently. It's expected that future devices
will follow similar algorithms. Even though the algorithm does
present some challenges for older OS's (such as Centos 6), trying
to rewrite the algorithm to fit both models would be more complex
and involve pulling the monitor object out of the private data
lockable object and would need to be guarded by a separate lock.
Devising such an algorithm to work around issues with older OS's
at the expense of more modern OS algorithms in newer event processing
code may result in unexpected issues, so the choice is to encourage
use of newer OS's with newer udev event processing code.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1524230
The qemuBuildVhostuserCommandLine builds command line for
vhostuser type interfaces. It is duplicating some code of the
function it is called from (qemuBuildInterfaceCommandLine)
because of the way it's called. If we merge it into the caller
not only we save a few lines but we also enable checks that we
would have to duplicate otherwise (e.g. QoS availability).
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
When we have variables A, B, C then there are two ways to free
them. Either in the order they are declared or the reversed one.
Any other ordering is confusing. In this commit I'm reordering
calls to VIR_FREE in the reversed order.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
The result of libssh2_userauth_password is being assigned to 'ret' in
one branch and 'rc' in the other branch. Checks are all done against the
'ret' variable, so one branch never does the correct check.
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Adjusting domain format documentation, adding device address
support and adding command line generation for vfio-ap.
Since only one mediated hostdev with model vfio-ap is supported a check
disallows to define domains with more than one such hostdev device.
Signed-off-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Chris Venteicher <cventeic@redhat.com>
Introduce vfio-ap capability.
Signed-off-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Reviewed-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Chris Venteicher <cventeic@redhat.com>
IOThread pids info will lost after libvirtd restart, then
if we call pinIOThread, sched_setaffinity will be called with
pid 0, not IOThread pid. So pinIOThread cannot work normally.
Signed-off-by: Jie Wang <wangjie88.huawei.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
virXMLFormatElement() frees attrBuf on success, but not necessarily
on failure. Most other callers of this function take the time to
reset attrBuf afterwords, but qemuDomainObjPrivateXMLFormatBlockjobs()
was relying on it succeeding, and could thus result in a memory leak.
Signed-off-by: Eric Blake <eblake@redhat.com>
ACKed-by: Michal Privoznik <mprivozn@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1640465
Weirdly enough, there can be symlinks in the path we are trying
to fix. If it is the case our clever algorithm that finds matches
against mount table won't work. Canonicalize path at the
beginning then.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
The virFileInData() function should return to the caller if the
current position the passed file is in is a data section or a
hole (and also how long the current section is). At any rate,
upon return from this function (be it successful or not) the
original position in the file is restored. This may mess up with
errno which might have been set earlier. Save the errno into a
local variable so it can be restored for the caller's sake.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
The QEMU @cfg config variable is unused in context of qemuProcessInit,
let's drop it.
Signed-off-by: Bjoern Walk <bwalk@linux.ibm.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
If the learning thread is configured to learn on all ethernet frames
(which is hardcoded) then chances are high that there is a packet on
every iteration of inspecting frames loop. As result we will hang on
shutdown because we don't check threadsTerminate if there is packet.
Let's just check termination conditions on every iteration. Since
we'll check each iteration, the check after pcap_next essentially
is unnecessary since on failure we'd loop back to the top and timeout
and then fail.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1632833
When doing a SCSI passthrough we don't put format= onto the
command line. This causes qemu to probe the format automatically
which ends up in a warning in the domain log and possible qemu
disabling writes to the first block (according to the warning
message).
Based-on-work-of: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
The @alloc object returned by virDomainResctrlVcpuMatch is not
properly referenced and un-referenced in virDomainCachetuneDefParse.
This patch fixes this problem.
Signed-off-by: Wang Huaqiang <huaqiang.wang@intel.com>
Reviewed-by: John Ferlan <jferlan@redhat.com>
The URI parser used by libvirt does not populate uri->path if the
trailing slash is missing. The code virStorageSourceParseBackingURI
would then not populate src->path.
As only NBD network disks are allowed to have the 'name' field in the
XML defining the disk source omitted we'd generate an invalid XML which
we'd not parse again.
Fix it by populating src->path with an empty string if the uri is
lacking slash.
As pointed out above NBD is special in this case since we actually allow
it being NULL. The URI path is used as export name. Since an empty
export does not make sense the new approach clears the src->path if the
trailing slash is present but nothing else.
Add test cases now to cover all the various cases for NBD and non-NBD
uris as there was to time only 1 test abusing the quirk witout slash for
NBD and all other URIs contained the slash or in case of NBD also the
export name.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
The name is misleading. Change it to 'uristr' so that 'path' can be
reused in the proper context later.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
If the same source gets built twice ('build same source on different
hosts at different times') the resulting files may differ.
Fix this by sorting the hash keys before usage.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
There are couple of things wrong with the current implementation.
The first one is that in the first loop the code tries to build a
list of fuse.glusterfs mount points. Well, since the strings are
allocated in a temporary buffer and are not duplicated this
results in wrong decision made later in the code.
The second problem is that the code does not take into account
subtree mounts. For instance, if there's a fuse.gluster mounted
at /some/path and another FS mounted at /some/path/subdir the
code would not recognize this subdir mount.
Reported-by: Han Han <hhan@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
If the given path is already a mount point (e.g. a bind mount of
a file, or simply a direct mount point of a FS), then our code
fails to detect that because the first thing it does is cutting
off part after last slash '/'.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
On s390x the struct member f_type of statsfs is hard coded to 'unsigned
int'. Change virFileIsSharedFixFUSE() to take a 'long long int' and use
a temporary to avoid pointer-casting.
This fixes the following error:
../../src/util/virfile.c:3578:38: error: cast increases required alignment of target type [-Werror=cast-align]
virFileIsSharedFixFUSE(path, (long *) &sb.f_type);
Signed-off-by: Marc Hartmayer <mhartmay@linux.ibm.com>
Signed-off-by: Bjoern Walk <bwalk@linux.ibm.com>
virFileReadValueUint does not log errors for non-existient files,
it merely returns -2.
Commit 12093f1 introduced this.
Signed-off-by: Ján Tomko <jtomko@redhat.com>
-net name= will be deprecated in QEMU 3.1:
commit 101625a4d4ac7e96227a156bc5f6d21a9cc383cd
net: Deprecate the "name" parameter of -net
git describe: v3.0.0-791-g101625a4d4
Use the id option instead, supported since QEMU 1.2:
commit 6687b79d636cd60ed9adb1177d0d946b58fa7717
convert net_client_init() to OptsVisitor
git describe: v1.0-3564-g6687b79d63 contains: v1.2.0-rc0~142^2~8
Thankfully, libvirt only uses -net for non-PCI, non-virtio NICs
on ARM.
Signed-off-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Laine Stump <laine@laine.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
We now explicitly handle media change elsewhere so we can drop the
switch statement. This will also make it more intuitive once CDROM
device hotplug might be supported.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Disk hotplug has slightly different semantics from media changing. Move
the media change code out and add proper initialization of the new
source object and proper cleanups if something fails.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
The disk hotplug code also overloads media change which is not ideal.
This will allow splitting out of the media change code.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
The disk storage source needs to be prepared if we want to use -blockdev
or secrets for the new media image. It does not hurt to do the same for
the legacy hotplug code as well.
Unfortunately helpers like qemuDomainPrepareDiskSource take
virDomainDiskDef as an argument and it would be hard to fix them to take
an explicit source, so the function also temporarily replaces disk->src
for the new source in this function.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Some functions require us to replace disk->src with the new source for
them to work properly. To avoid confusion all places which allow
explicit virStorageSource should get the appropriate definition.
The legacy code fortunately does not need anything from the old source
so that does not require modifications.
Blockdev does require the old definition so we'll pass it explicitly.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Since the code is also used when changing media we need to allow
specifying explicit source for which we are going to prepare. With this
change callers don't have to replace disk->src with the new source
definition for generating these.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
qemu media changing code tried to assume old media's format for the new
one if that was not specified. Since the format will always be present
it does not make sense to keep the code around.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Old media changing code does not bother setting up the secrets for new
media or actually removing/adding of the corresponding objects.
Additionally it uses secrets setup for the old image to be removed as
the secret for the new image which is wrong.
Remove the support for secrets while changing media for the legacy
approach. The only reasonable way to fix it is when using blockdev.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
While the idea was good the implementation not so much as we need to
take into account the old disk data and the new source. The code will be
consolidated later in a different way.
This reverts commit 663b1d55de.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Preparing the storage source prior to assigning the alias will not work
as the names of the certain objects depend on the alias for the legacy
hotplug case as we generate the object names for the secrets based on
the alias.
This reverts commit 192fdaa614.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
This enables to use both cgroup v1 and v2 at the same time together
with libvirt. It is supported by kernel and there is valid use-case,
not all controllers are implemented in cgroup v2 so there might be
configurations where administrator would enable these missing
controllers in cgroup v1.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
In order to set CPU cfs period using cgroup v2 'cpu.max' interface
we need to load the current value of CPU cfs quota first because
format of 'cpu.max' interface is '$quota $period' and in order to
change 'period' we need to write 'quota' as well. Writing only one
number changes only 'quota'.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
