* src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h: Add new
  qemuMonitorSetBalloon() based on existing code in
  qemudDomainSetMemoryBalloon
* src/qemu/qemu_driver.c: Remove use of qemudDomainSetMemoryBalloon()
  in favour of qemuMonitorSetBalloon(). Fix error code when balloon
  is not supported

* src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h: Pull old
  qemudDomainGetMemoryBalloon() code into a new method called
  qemuMonitorGetBalloonInfo()
* src/qemu/qemu_driver.c: Update to call qemuMonitorGetBalloonInfo()
  and remove qemudDomainGetMemoryBalloon().

* src/qemu/qemu_driver.c: Remove use of 'system_powerdown'
* src/qemu/qemu_monitor.h, src/qemu/qemu_monitor.c: Add a new
  qemuMonitorSystemPowerdown() api call

* src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h: Add a new
  qemuMonitorStopCPUs() API
* src/qemu/qemu_driver.c: Replace direct monitor commands for 'stop'
  with qemuMonitorStopCPUs()

* src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h: Rename
  Rename qemudMonitorSendCont to qemuMonitorStartCPUs
* src/qemu/qemu_driver.c: Update callers for new name

* src/qemu/qemu_monitor.c, src/qemu/qemu_monitor.h: Add a
  new qemuMonitorSetVNCPassword() API
* src/qemu/qemu_driver.c: Refactor qemudInitPasswords to
  call qemuMonitorSetVNCPassword()

* src/qemu/qemu_monitor.h, src/qemu/qemu_monitor.c: Add a new
  qemuMonitorGetCPUInfo() command
* src/qemu/qemu_driver.c: Refactor qemudDetectVcpuPIDs to
  use qemuMonitorGetCPUInfo()

Pull out all the QEMU monitor interaction code to a separate
file. This will make life easier when we need to drop in a
new implementation for the forthcoming QMP machine friendly
monitor support.

Next step is to add formal APIs for each monitor command,
and remove direct commands for sending/receiving generic
data.

* src/Makefile.am: Add qemu_monitor.c to build
* src/qemu/qemu_driver.c: Remove code for monitor interaction
* src/qemu/qemu_monitor_text.c, src/qemu/qemu_monitor_text.h: New
  file for monitor interaction
* po/POTFILES.in: Add src/qemu/qemu_monitor_text.c

* src/conf/interface_conf.c: This was causing subsequent calls to
  virXPathxxx() to fail, since ctxt->node was left pointing at the
  dhcp node, rather than the protocol node.

* src/util/xml.c: The virXPath... function take extra care to preserve
  the XPath context node (ctxt->node) but in the case of virXPathString
  and virXPathBoolean they forgot to do this on the error path. This
  patch fixes this and move all ctxt->node = relnode instuctions just
  after the xmlXPathEval() to make sure this doesn't happen if this code
  is modified.

Seems standard to include internal.h in order to pull in libvirt.h

* src/util/util.h: include internal.h

* libvirt.spec.in: libvirt-python only needs to depend on
  libvirt-clients

The python method help docs are copied across from the C
funtion comments, but in the process all line breaks and
indentation was being lost. This made the resulting text
and code examples completely unreadable. Both the API
doc extractor and the python generator were destroying
whitespace & this fixes them to preserve it exactly.

* docs/apibuild.py: Preserve all whitespace when extracting
  function comments. Print function comment inside a <![CDATA[
  section to fully preserve all whitespace. Look for the
  word 'returns' to describe return values, instead of 'return'
  to avoid getting confused with code examples including the
  C 'return' statement.
* python/generator.py: Preserve all whitespace when printing
  function help docs
* src/libvirt.c: Change any return parameter indicated by
  'return' to be 'returns', to avoid confusing the API extractor
* docs/libvirt-api.xml: Re-build for fixed descriptions

Set def to NULL in secretLoad(), otherwise we can access a
random pointer on error.
Signed-off-by: NChris Lalancette <clalance@redhat.com>

* src/qemu/qemu_driver.c: initialize detach var

* src/conf/domain_conf.h: check domain/bus/slot, not domain/domain/slot

Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

An inaccessible datastore has no valid URL property so don't
access its URI property.

* src/esx/esx_vi.c: esxVI_LookupDatastoreByName(): check if datastore is
  accessible before accessing its URL property
* src/esx/esx_vmx.c: update to changed datastore properties

* src/esx/esx_driver.c: remove phantom mode

* src/esx/esx_util.[ch]: remove esxUtil_EqualSuffix()
* src/esx/esx_driver.c, src/esx/esx_vmx.c: replace esxUtil_EqualSuffix()
  with virFileHasSuffix()

A given domain XML gets converted to a VMX config, uploaded to the host
and registered as new virtual machine.

* src/esx/esx_driver.c: refactor datastore related path parsing into
  esxUtil_ParseDatastoreRelatedPath()
* src/esx/esx_util.[ch]: add esxUtil_ParseDatastoreRelatedPath()
* src/esx/esx_vi.[ch]: add esxVI_Context_UploadFile(), add datastores to
  the traversal in esxVI_BuildFullTraversalSpecList(), add
  esxVI_LookupDatastoreByName()
* src/esx/esx_vi_methods.[ch]: add esxVI_RegisterVM_Task()
* src/esx/esx_vi_types.c: make some error message more verbose
* src/esx/esx_vmx.[ch]: add esxVMX_AbsolutePathToDatastoreRelatedPath()
  to convert a path into a datastore related path, add esxVMX_ParseFileName()
  to convert from VMX path format to domain XML path format, extend the other
  parsing function to be datastore aware, add esxVMX_FormatFileName() to
  convert from domain XML path format to VMX path format, fix VMX ethernet
  entry formating
* tests/esxutilstest.c: add test for esxUtil_ParseDatastoreRelatedPath()
* tests/vmx2xmldata/*: update domain XML files to use datastore related paths
* tests/xml2vmxdata/*: update domain XML files to use datastore related paths,
  update VMX files to use absolute paths

Add esxVI_Occurence enum to describe expected occurence of items
* src/esx/esx_driver.c: update the use of esxVI_LookupVirtualMachineByUuid()
* src/esx/esx_vi.c: add an esxVI_Occurence parameter to
  esxVI_LookupVirtualMachineByUuid() and take care if esxVI_FindByUuid()
  can't find anything for a given uuid
* src/esx/esx_vi.h: add esxVI_Occurence enum
* src/esx/esx_vi_methods.c: expect null or more items to be returned
  from esxVI_FindByUuid()

* src/esx/esx_driver.c: add esxSupportsLongMode() and update esxCapsInit()
* src/esx/esx_vi.[ch]: Add AnyType handling for lists
* src/esx/esx_vi_types.c: bind VI type HostCpuIdInfo

* tests/.gitignore: ignore new test binaries
* tests/Makefile.am: add new tests
* tests/esxutilstest.c: test esxVMX_IndexToDiskName()
* tests/vmx2xmldata/*: config files for the VMX to domain XML test
* tests/vmx2xmltest.c: test the VMX to domain XML mapping
* tests/xml2vmxdata/*: config files for the domain XML to VMX test
* tests/xml2vmxtest.c: test the domain XML to VMX mapping

Extend and cleanup the VMX to domain XML mapping. Add the domain XML to
VMX mapping functions.

* src/esx/esx_driver.c: add esxDomainXMLToNative()
* src/esx/esx_vmx.[ch]: add esxVMX_SCSIDiskNameToControllerAndID(),
  esxVMX_IDEDiskNameToControllerAndID(), esxVMX_FloppyDiskNameToController(),
  esxVMX_GatherSCSIControllers(), add basic handling for the VMX guestOS entry
  to distinguish between i686 and x86_64, make SCSI virtualDev VMX entry
  optional as it should be, map the VMX networkName entry to the domain XML
  interface bridge name, add basic mapping for serial devices in pipe mode,
  add several esxVMX_Format*() functions

This enables the auth callback to automatically distinguish between
requests for ESX host and vCenter credentials.

* src/esx/esx_util.[ch]: set challenge for auth callback to hostname

* src/esx/esx_driver.c: add esxNodeGetFreeMemory(), cache IP address
* src/esx/esx_vi.[ch]: refactor resource pool query into esxVI_GetResourcePool()
* src/esx/esx_vi_types.[ch]: bind VI type ResourcePoolResourceUsage

Signed-off-by: NChris Lalancette <clalance@redhat.com>

Currently, libvirtd will start a dnsmasq process for the virtual
network, but (aside from killing the dnsmasq process and replacing it),
there's no way to define tftp boot options.

This change introduces the appropriate tags to the dhcp configuration:

 <network>
   <name>default</name>
   <bridge name="virbr%d" />
   <forward/>
   <ip address="192.168.122.1" netmask="255.255.255.0">
     <tftp root="/var/lib/tftproot" />
     <dhcp>
       <range start="192.168.122.2" end="192.168.122.254" />
       <bootp file="pxeboot.img"/>
     </dhcp>
   </ip>
 </network>

When the attributes are present, these are passed to the
arguments to dnsmasq:

 dnsmasq [...] --enable-tftp --tftp-root /srv/tftp --dhcp-boot pxeboot.img
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^
                      from <tftp />                     from <bootp />

At present, only local tftp servers are supported (ie, dnsmasq runs as
the tftp server), but we could improve this in future by adding a
server= attribute.
Signed-off-by: NJeremy Kerr <jk@ozlabs.org>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

2009-09-21  Paolo Bonzini  <pbonzini@redhat.com>
            Jeremy Kerr  <jk@ozlabs.org>

	* docs/formatnetwork.html.in: Document new tags.
	* docs/formatnetwork.html: Regenerate.
	* docs/schemas/network.rng: Update.
	* src/network_conf.c (virNetworkDefFree): Free new fields.
	(virNetworkDHCPRangeDefParseXML): Parse <bootp>.
	(virNetworkIPParseXML): New, parsing <dhcp> and <tftp>.
	(virNetworkDefParseXML): Use virNetworkIPParseXML instead of
	virNetworkDHCPRangeDefParseXML.
	(virNetworkDefFormat): Pretty print new fields.
	* src/network_conf.h (struct _virNetworkDef): Add netboot fields.
	* src/network_driver.c (networkBuildDnsmasqArgv): Add
	TFTP and BOOTP arguments.

	* tests/Makefile.am (EXTRA_DIST): Add networkschemadata.
	* tests/networkschematest: Look in networkschemadata.
	* tests/networkschemadata/netboot-network.xml: New.

Add the virStrncpy function, which takes a dst string, source string,
the number of bytes to copy and the number of bytes available in the
dest string.  If the source string is too large to fit into the
destination string, including the \0 byte, then no data is copied and
the function returns NULL.  Otherwise, this function copies n bytes
from source into dst, including the \0, and returns a pointer to the
dst string.  This function is intended to replace all unsafe uses
of strncpy in the code base, since strncpy does *not* guarantee that
the buffer terminates with a \0.
Signed-off-by: NChris Lalancette <clalance@redhat.com>

Latest upstream QEMU can be built with Xen support, which introduces
a -xen-domid argument. This was  mistakenly detected as -domid due
to old Xenner support. Adapt to cope with both syntax. Also only
set domid if the virt type is xen, or the guest type is xen

* src/qemu_conf.c, src/qemu_conf.h: Detect new -xen-domid flag in
  preference to -domid.
* tests/qemuxml2argvdata/qemuxml2argv-bootloader.args,
  tests/qemuxml2argvdata/qemuxml2argv-input-xen.args: Add missing
  -domid param
* tests/qemuxml2argvdata/qemuxml2argv-misc-uuid.args: Remove bogus
  -boot param.
* tests/qemuxml2argvtest.c: Add missing QEMUD_CMD_FLAG_DOMID params

* include/libvirt/.gitignore: Ignore libvirt.h

The commit cb51aa48 "Fix up connection
reference counting." changed the driver closing and virConnectPtr
unref-logic in virConnectClose().

Before this commit virConnectClose() closed all drivers of the given
virConnectPtr and virUnrefConnect()'ed it afterwards. After this
commit the driver-closing is done in virUnrefConnect() if and only if
the ref-count of the virConnectPtr dropped to zero.

This change in execution order leads to a virConnectPtr leak, at least
for connections to Xen.

The relevant call sequences:

virConnectOpen() -> xenUnifiedOpen() ...

... xenInotifyOpen() -> virConnectRef(conn)

... xenStoreOpen() -> xenStoreAddWatch() -> conn->refs++

virConnectClose() -> xenUnifiedClose() ...

... xenInotifyClose() -> virUnrefConnect(conn)

... xenStoreClose() -> xenStoreRemoveWatch() -> virUnrefConnect(conn)

Before the commit this additional virConnectRef/virUnrefConnect calls
where no problem, because virConnectClose() closed the drivers
explicitly and the additional refs added by the Xen subdrivers were
removed properly. After the commit this additional refs result in a
virConnectPtr leak (including a leak of the hypercall file handle;
that's how I noticed this problem), because now the drivers are only
close if and only if the ref-count drops to zero, but this cannot
happen anymore, because the additional refs from the Xen subdrivers
would only be removed if the drivers get closed, but that doesn't
happen because the ref-count cannot drop to zero.

The fix for this problem is simple: remove the
virConnectRef/virUnrefConnect calls from the Xen subdrivers (see
attached patch). Maybe someone could explain why the Xen Inotify and
Xen Store driver do this extra ref-counting, but none of the other Xen
subdrivers. It seems unnecessary to me and can be removed without
problems.
Signed-off-by: NChris Lalancette <clalance@redhat.com>

* src/conf/domain_conf.c: Don't assume all virDomainObjPtr have
  a non-NULL monitor_chr field in virDomainObjFormat.
* src/lxc/lxc_driver.c: Implement suspend/resume driver APis
* src/util/cgroup.c, src/util/cgroup.h: Support the 'freezer'
  cgroup controller
* src/libvirt_private.syms: Export virCgroupSetFreezerState
  and virCgroupGetFreezerState

* src/openvz/openvz_driver.c: try to autogenerate a non-conflicting
  id for the domain and use the name field as the name not as id.

* docs/formatdomain.html.in: explain the new syntax of the video and
  graphic elements in the domain XML