Rather than numerous instances of:

emulator = vm->def->emulator;
if (!emulator)
    emulator = virDomainDefDefaultEmulator(conn, vm->def, driver->caps);
if (!emulator)
    return -1;

Set this value at XML parse time in the domain config, so we can depend on
it for all future emulator accesses. There were unchecked accesses in the
qemu driver that were tripping up on this if no emulator was specified in
the XML, see:

http://www.redhat.com/archives/libvir-list/2008-October/msg00602.html

Certain drivers always need an 'emulator' specified in the XML (qemu and lxc
at least). Store this info in capabilities.

We will eventually use this to move default emulator handling out of
drivers and into domain_conf.*

Qemu < 0.10.0 did not support it, and virt-* tools now try to add this by
default, so it's extra important we ensure the option exists.

New function qemudParseHelpStr was being called with arguments in the
wrong order, so command flags == kvm_version :/

Signed-off-by: NMark McLoughlin <markmc@redhat.com>

The KVM version string can be one of the following:

  - qemu-kvm-x.y.z in stable releases
  - kvm-XX for kvm versions up to kvm-85
  - qemu-kvm-devel-XX for kvm version kvm-86 and later

There are only a few of places where we need to detect
differences between KVM versions based on 0.9.1:

  1) VNET_HDR introduced in kvm-74

  2) -incoming tcp introduced in kvm-79

  3) -incoming exec introduced in kvm-80

  4) -incoming stdio in all earlier kvm versions

With qemu-kvm-0.10.x, we can now assume that (1) is available
if it's a KVM release, (2) and (3) is always available and
(4) is never available.

So, from now on we should only need to check the qemu version
number and the "is_kvm" flag for detecting feature availability.
We only need the KVM version number for older releases.
Signed-off-by: NMark McLoughlin <markmc@redhat.com>

This patch is purely re-factoring without any functional changes
to make way for the next patch.

The main thing achieved by the refactoring is that we now have
easier access to the parenthesised string that KVM folks seem
to delight in changing.
Signed-off-by: NMark McLoughlin <markmc@redhat.com>

* src/qemu_conf.c docs/schemas/domain.rng
  tests/capabilityschemadata/caps-qemu-kvm.xml: PPC Qemu Machine Type
  changed from g3bw to g3beige some time ago, patch by Thomas Baker
Daniel

virExec will write out the pid of the daemonized process only. Use this
in the QEMU driver, rather than QEMU's pidfile, so we can catch errors we
might miss if the emulator bails early.

* src/domain_conf.c src/domain_conf.h: parse and save multiple
  graphics elements instead of one, patch by Pritesh Kothari
* src/qemu_conf.c src/qemu_driver.c: adapt the qemu/kvm driver for
  the new structures, patch by Pritesh Kothari
* src/xend_internal.c src/xm_internal.c: same thing for Xen drivers
Daniel

Seems like a simple typo - it has been "format=" since the flag
was introduced, but we added it as "fmt=".
Signed-off-by: NMark McLoughlin <markmc@redhat.com>

* src/qemu_conf.c (CHECK_TYPE): Fix typo:
s/security_river/security_driver/

* docs/schemas/domain.rng src/domain_conf.c src/domain_conf.h
  src/qemu_conf.c
  tests/qemuxml2argvdata/qemuxml2argv-hostdev-pci-address.xml
  tests/qemuxml2argvdata/qemuxml2argv-hostdev-usb-address.xml
  tests/qemuxml2argvdata/qemuxml2argv-hostdev-usb-product.xml:
  adds a new <hostdev managed='(yes|no)'> property
  to host devices indicating whether or not we should
  automatically dettach/reset, patch by Mark McLoughlin
daniel

* qemud/qemud.c (GET_CONF_STR): Use virStrerror, not strerror.
* qemud/remote.c (remoteDispatchDomainBlockPeek): Likewise.
(remoteDispatchDomainMemoryPeek, remoteDispatchAuthSaslInit): Likewise.
(remoteDispatchAuthPolkit): Likewise.
* src/lxc_container.c (lxcContainerAvailable): Likewise.
* src/network_driver.c (networkStartNetworkDaemon): Likewise.
(networkShutdownNetworkDaemon): Likewise.
* src/qemu_conf.c (qemudExtractVersion, qemudNetworkIfaceConnect):
* src/storage_conf.c (virStoragePoolLoadAllConfigs): Likewise.
* src/storage_driver.c (storagePoolUndefine): Likewise.
* src/uml_driver.c (umlStartup, umlStartVMDaemon): Likewise.
* src/util.c (virFileReadAll): Likewise.
* src/uuid.c (virUUIDGenerate): Likewise.
* src/xen_internal.c (get_cpu_flags): Likewise.

* src/qemu_conf.c (uname_normalize): New function, factored out of...
(qemudBuildCommandLine): ...here.  Use the new function.
(qemudExtractVersion): Use it here, rather than hard-coding "i686".

* src/uml_conf.c (VIR_FROM_THIS): Define to VIR_FROM_UML.
* src/xs_internal.c (VIR_FROM_THIS): Define to VIR_FROM_XEN.
* src/xml.c (VIR_FROM_THIS): Define to VIR_FROM_XML.
* src/stats_linux.c (VIR_FROM_THIS): Define to VIR_FROM_STATS_LINUX.
* src/datatypes.c (VIR_FROM_THIS): Define to VIR_FROM_NONE.
* src/lxc_conf.c (VIR_FROM_THIS): Define to VIR_FROM_LXC.
* src/libvirt.c (VIR_FROM_THIS): Define to VIR_FROM_NONE.
* src/node_device_conf.c (VIR_FROM_THIS): Define to VIR_FROM_NODEDEV.
* src/openvz_conf.c (VIR_FROM_THIS): Define to VIR_FROM_OPENVZ.
* src/openvz_driver.c (VIR_FROM_THIS): Define to VIR_FROM_OPENVZ.
* src/conf.c (VIR_FROM_THIS): Define to VIR_FROM_CONF.
Note: this loses config_filename:config_lineno diagnostics,
but that's ok.
* src/node_device.c (VIR_FROM_THIS): Define to VIR_FROM_NODEDEV.
* src/sexpr.c (VIR_FROM_THIS): Define to VIR_FROM_SEXPR.
* po/POTFILES.in: remove src/sexpr.c and src/lxc_conf.c

IFF_VNET_HDR is a tun/tap flag that allows you to send and receive
large (i.e. GSO) packets and packets with partial checksums. Setting
the flag means that every packet is proceeded by the same header which
virtio uses to communicate GSO/csum metadata.

By enabling this flag on the tap fds we create, we greatly increase
the achievable throughput with virtio_net.

However, we need to be careful to only set the flag when a) QEMU has
support for this ABI and b) the value of the flag is queryable using
the TUNGETIFF ioctl.

It's nearly five months since kvm-74 - the first KVM release with this
feature - was released. Up until now, we've not added libvirt support
because there is no clean way to detect support for this in QEMU at
runtime. A brief attempt to add a "info capabilities" monitor command
to QEMU floundered. Perfect is the enemy of good enough. Probing the
KVM version will suffice for now.
Signed-off-by: NMark McLoughlin <markmc@redhat.com>