Now that all pieces are in place (hopefully) let's enable -blockdev.

We base the capability on presence of the fix for 'auto-read-only' on
files so that blockdev works properly, mandate that qemu supports
explicit SCSI id strings to avoid ABI regression and that the fix for
'savevm' is present so that internal snapshots work.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

The latter is deprecated and will be removed soon. The advised
replacement is '-overcommit mem-lock=on|off'.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

By default, qemu user's home dir points to '/' which shouldn't be used
at all. We therefore pass the HOME variable from the current variable
iff not running as SUID, which means that for systemd we never set it.
This patch makes sure, that for system QEMU this is always set to
libDir/<driver>, session mode is left untouched.
Signed-off-by: NErik Skultety <eskultet@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

For session mode, only XDG_CACHE_HOME is set, because we want to remain
integrating with services in user session, but for system mode, this
would have become reading/writing to '/' which carries the obvious issue
with permissions (also, '/' is the wrong location in 99.9% cases anyway).
Signed-off-by: NErik Skultety <eskultet@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

Add <controller type='scsi' model handling for virtio transitional
devices. Ex:

  <controller type='scsi' model='virtio-transitional'/>

* "virtio-transitional" maps to qemu "virtio-scsi-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-scsi-non-transitional"

The naming here doesn't match the pre-existing model=virtio-scsi.
The prescence of '-scsi' there seems kind of redundant as we have
type='scsi' already, so I decided to follow the pattern of other
patches and use virtio-transitional etc.
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add controller type='virtio-serial' model handling for virtio
transitional devices. Ex:

  <controller type='virtio-serial' model='virtio-transitional'/>

* "virtio-transitional" maps to qemu "virtio-serial-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-serial-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add <input> model handling for virtio transitional devices. Ex:

  <input type='passthrough' bus='virtio' model='virtio-transitional'>
    ...
  </input>

* "virtio-transitional" maps to qemu "virtio-input-host-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-input-host-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

<input> devices lack the model= attribute which is used by
most other device types. To eventually support
virtio-input-host-pci-{non-}traditional in qemu, let's add
a standard model= attribute. This just adds the domain_conf
wiring
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add <vsock> model handling for virtio transitional devices. Ex:

  <vsock model='virtio-transitional'>
    ...
  </vsock>

* "virtio-transitional" maps to qemu "vhost-vsock-pci-transitional"
* "virtio-non-transitional" maps to qemu "vhost-vsock-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add new <memballoon> model values for virtio transitional devices. Ex:

  <memballoon model='virtio-transitional'/>

* "virtio-transitional" maps to qemu "virtio-balloon-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-balloon-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add <filesystem> model handling for virtio transitional devices. Ex:

  <filesystem type='mount' model='virtio-transitional'>
    ...
  </filesystem

* "virtio-transitional" maps to qemu "virtio-9p-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-9p-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

<filesystem> devices lack the model= attribute which is used by
most other device types. To eventually support
virtio-9p-pci-{non-}traditional in qemu, let's add a standard
model= attribute. The accepted values are:

- virtio
- virtio-transitional
- virtio-non-transitional
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add new <rng> model values for virtio transitional devices. Ex:

  <rng model='virtio-transitional'>
    ...
  </rng>

* "virtio-transitional" maps to qemu "virtio-rng-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-rng-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add <hostdev> protocol=vhost model handling for virtio transitional
devices. Ex:

  <hostdev mode='subsystem' type='scsi_host' model='virtio-transitional'>
    <source protocol='vhost' wwpn=X/>
  </hostdev>

* "virtio-transitional" maps to qemu "vhost-scsi-pci-transitional"
* "virtio-non-transitional" maps to qemu "vhost-scsi-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

qemu vhost-scsi devices map to XML roughly like:

    <hostdev mode='subsystem' type='scsi_host'>
      <source protocol='vhost' wwpn=X/>
    </hostdev>

To support vhost-scsi-pci-{non-}traditional in qemu, we
need to to extend the SCSI Host hostdev XML to handle
model= value. This matches the XML model= format used
for mediated devices. This is just the domain_conf bits
and some XML test cases.

Use of virtio-X naming here does not match the hostdev
protocol=vhost nor does it match the qemu vhost-X device
naming, however it's more consistent with all other
model= names in this area, and also matches the
inconsistency of <vsock> devices which use model=virtio
but map to vhost-vsock on the qemu commandline
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add new <interface> model handling for virtio transitional devices. Ex:

<interface>
  <model type='virtio-transitional'/>
</interface>

* "virtio-transitional" maps to qemu "virtio-net-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-net-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Add new <disk> model values for virtio transitional devices. When
combined with bus='virtio':

* "virtio-transitional" maps to qemu "virtio-blk-pci-transitional"
* "virtio-non-transitional" maps to qemu "virtio-blk-pci-non-transitional"
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

<disk> devices lack the model= attribute which is used by
most other device types. bus= mostly acts as one, but it
serves other purposes too like determing what target=
prefix to use, and for matching against controller type=
values.

Extending bus= to handle additional virtio transitional
devices will complicate apps lives, and it isn't a clean
mapping anyways. So let's bite the bullet and add a new
<disk model=X/> attribute, and wire up common handling
for virtio and virtio-{non-}transitional
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

The default is stable per machine type so there should be no need to keep that.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1469338Signed-off-by: NMartin Kletzander <mkletzan@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

To avoid problems with test cases specifying an alias machine type which
would change once capabilities for a newer version are added strip all
alias machine types for the DO_TEST_CAPS_LATEST based tests.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

There is a race condition when spawning QEMU where libvirt has spawned
QEMU but the monitor socket is not yet open. Libvirt has to repeatedly
try to connect() to QEMU's monitor until eventually it succeeds, or
times out. We use kill() to check if QEMU is still alive so we avoid
waiting a long time if QEMU exited, but having a timeout at all is still
unpleasant.

With QEMU 2.12 we can pass in a pre-opened FD for UNIX domain or TCP
sockets. If libvirt has called bind() and listen() on this FD, then we
have a guarantee that libvirt can immediately call connect() and
succeed without any race.

Although we only really care about this for the monitor socket and agent
socket, this patch does FD passing for all UNIX socket based character
devices since there appears to be no downside to it.

We don't do FD passing for TCP sockets, however, because it is only
possible to pass a single FD, while some hostnames may require listening
on multiple FDs to cover IPv4 and IPv6 concurrently.
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

https://bugzilla.redhat.com/show_bug.cgi?id=1149445

If the domain requests usage of the genid functionality,
then add the QEMU '-device vmgenid' to the command line
providing either the supplied or generated GUID value.

Add tests for both a generated and supplied GUID value.
Signed-off-by: NJohn Ferlan <jferlan@redhat.com>
ACKed-by: NMichal Privoznik <mprivozn@redhat.com>

If QEMU uses a seccomp blacklist (since 2.11), -sandbox on
no longer tries to whitelist all the calls, but uses sets
of blacklists:
default (always blacklisted with -sandbox on)
obsolete (defaults to deny)
elevateprivileges (setuid & co, default: allow)
spawn (fork & execve, default: allow)
resourcecontrol (setaffinity, setscheduler, default: allow)

If these are supported, default to sandbox with all four
categories blacklisted.

https://bugzilla.redhat.com/show_bug.cgi?id=1492597Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

Implied by QEMU >= 1.3.0.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Implied by QEMU >= 1.2.0.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Implied by QEMU >= 1.2.0.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Implied by QEMU >= 1.2.0.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Implied by QEMU >= 1.2.0.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Implied by QEMU >= 1.2.0.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Implied by QEMU >= 1.2.0.

Delete this one first, because QEMU_CAPS_NODEFCONFIG is only used
when QEMU_CAPS_NO_USER_CONFIG is unsupported.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

We require QEMU >= 1.5.0, assume every QEMU supports it.
Sadly that does not let us trivially drop qemuMonitor's
priv->monJSON bool, because of qemuDomainQemuAttach.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Similarly to the previous commit, rename .args files.

The files were renamed using the following commands. From
qemuxml2argvdata:

  for i in qemuxml2argv-*.args; do mv $i ${i#qemuxml2argv-}; done

and then (to fix broken symlinks) from qemuxml2argvdata and
qemuxml2xmloutdata:

  for i in $(find . -xtype l); do \
      ln -sf $(readlink $i | sed 's/qemuxml2argv-//') $i;
  done
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

AFAIK there aren't any cases where we will/should hit the old code
path for our supported qemu versions, so drop the old code.

Massive test suite churn follows
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Our test data used a lot of different qemu binary paths and some
of them were based on downstream systems.

Note that there is one file where I had to add "accel=kvm" because
the qemuargv2xml code parses "/usr/bin/kvm" as virt type="kvm".
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

Support for SMP topology was added by qemu commit dc6b1c09849484fbbc50
prior to 0.12.0, our minimum supported qemu version.

$ git describe --tags dc6b1c09849484fbbc50803307e4c7a3d81eab62
v0.11.0-rc0-449-gdc6b1c0
$ git describe --tags --contains dc6b1c09849484fbbc50803307e4c7a3d81eab
v0.12.0-rc0~1477

Per-domain directories were introduced in order to be able to
completely separate security labels for each domain (commit
f1f68ca3).  However when the domain
name is long (let's say a ridiculous 110 characters), we cannot
connect to the monitor socket because on length of UNIX socket address
is limited.  In order to get around this, let's shorten it in similar
fashion and in order to avoid conflicts, throw in an ID there as well.
Also save that into the status XML and load the old status XMLs
properly (to clean up after older domains).  That way we can change it
in the future.

The shortening can be seen in qemuxml2argv tests, for example in the
hugepages-pages2 case.
Signed-off-by: NMartin Kletzander <mkletzan@redhat.com>

QEMU_CAPS_DEVICE is always enabled for qemu binaries we support.
Sync qemuxml2* to match, and regenerate all test output.

Just recently, qemu forbade specifying format for sourceless
disks (qemu commit 39c4ae941ed992a3bb5). It kind of makes sense.
If there's no file to open, why specify its format. Anyway, I
have a domain like this:

    <disk type='file' device='cdrom'>
      <driver name='qemu' type='raw'/>
      <target dev='hda' bus='ide'/>
      <readonly/>
      <address type='drive' controller='0' bus='0' target='0' unit='0'/>
    </disk>

and obviously I am unable to start it. Therefore, a fix on our
side is needed too.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

As of QEMU 0.10.0 the -drive format= parameter was added,
so the QEMU driver can assume it is always available.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

The -uuid arg was added in QEMU 0.10.0, so the QEMU driver can
assume it is always available.
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>