VNC graphics already supports sockets but only via 'socket' attribute.
This patch coverts that attribute into listen type 'socket'.
For backward compatibility we need to handle listen type 'socket' and 'socket'
attribute properly to support old XMLs and new XMLs. If both are provided they
have to match, if only one of them is provided we need to be able to parse that
configuration too.
To not break migration back to old libvirt if the socket is provided by user we
need to generate migratable XML without the listen element and use only 'socket'
attribute.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Even though it's auto-generated it's based on qemu.conf option and listen type
address already uses "fromConfig" to carry this information. Following commits
will convert the socket to listen element so this rename is required because
there will be also an option to get socket auto-generated independently on the
qemu.conf option.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Since it will not be called from outside of conf we can unexport it too
if we move it to the appropriate place.
Test suite change is necessary since the error will be reported sooner
now.
Similarly to the domain definition validator add a device validator. The
change to the prototype of the domain validator is necessary as
virDomainDeviceInfoIterateInternal requires a non-const pointer.
Until now we weren't able to add checks that would reject configuration
once accepted by the parser. This patch adds a new callback and
infrastructure to add such checks. In this patch all the places where
rejecting a now-invalid configuration wouldn't be a good idea are marked
with a new parser flag.
Remove the live attribute and mark the definition as transient
whether the domain is runing or not.
There were only two callers left calling with live=false:
* testDomainStartState, where the domain already is active
because we assigned vm->def->id just a few lines above the call
* virDomainObjGetPersistentDef, which now only calls
virDomainObjSetDefTransient for an active domain
Calling virDomainObjSetDefTransient with live=false is a no-op
on an inactive domain.
Only call it on an active domain, since this is the only place using
the live bool.
There's this problem on the recent gcc-6.1:
In file included from conf/domain_conf.c:37:0:
conf/domain_conf.c: In function 'virDomainChrPreAlloc':
conf/domain_conf.c:14109:35: error: potential null pointer dereference [-Werror=null-dereference]
return VIR_REALLOC_N(*arrPtr, *cntPtr + 1);
^~
./util/viralloc.h:158:73: note: in definition of macro 'VIR_REALLOC_N'
# define VIR_REALLOC_N(ptr, count) virReallocN(&(ptr), sizeof(*(ptr)), (count), \
^~~~~
conf/domain_conf.c: In function 'virDomainChrRemove':
conf/domain_conf.c:14133:21: error: potential null pointer dereference [-Werror=null-dereference]
for (i = 0; i < *cntPtr; i++) {
^~~~~~~
GCC basically fails to see, that the
virDomainChrGetDomainPtrsInternal will never actually return NULL
because it's never called over a domain char device with _LAST
type. But to make it shut up, lets turn this function into
returning an integer and check in the callers if a zero value
value was returned.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Okay, I admit that our code here is complex. It's not easy to
spot that NULL deref can't really happen here. So it's no wonder
that a dumb compiler fails to see all the connections and
produces the following errors:
CC conf/libvirt_conf_la-domain_conf.lo
conf/domain_conf.c: In function 'virDomainDefFormatInternal':
conf/domain_conf.c:22162:22: error: potential null pointer dereference [-Werror=null-dereference]
if (sched->policy == i)
~~~~~^~~~~~~~
<snip/>
cc1: all warnings being treated as errors
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Hand-entering indexes for 20 PCI controllers is not as tedious as
manually determining and entering their PCI addresses, but it's still
annoying, and the algorithm for determining the proper index is
incredibly simple (in all cases except one) - just pick the lowest
unused index.
The one exception is USB2 controllers because multiple controllers in
the same group have the same index. For these we look to see if 1) the
most recently added USB controller is also a USB2 controller, and 2)
the group *that* controller belongs to doesn't yet have a controller
of the exact model we're just now adding - if both are true, the new
controller gets the same index, but in all other cases we just assign
the lowest unused index.
With this patch in place and combined with the automatic PCI address
assignment, we can define a PCIe switch with several ports like this:
<controller type='pci' model='pcie-root-port'/>
<controller type='pci' model='pcie-switch-upstream-port'/>
<controller type='pci' model='pcie-switch-downstream-port'/>
<controller type='pci' model='pcie-switch-downstream-port'/>
<controller type='pci' model='pcie-switch-downstream-port'/>
<controller type='pci' model='pcie-switch-downstream-port'/>
<controller type='pci' model='pcie-switch-downstream-port'/>
...
These will each get a unique index, and PCI addresses that connect
them together appropriately with no pesky numbers required.
Make virDomainControllerFindUnusedIndex() a global function so that it
can be used outside domain_conf.c (as well as higher up in
domain_conf.c itself)/ Also make its DomainDef arg a const* so that
functions which only have a const* to the domain can use it.
IS_USB2_CONTROLLER() is useful in more places aside from just when
assigning PCI addresses in QEMU, and is checking for enum values that
are all defined in conf/domain_conf.h anyway, so define it there
instead.
Add a new element to <domain> XML:
<os>
<acpi>
<table type="slic">/path/to/acpi/table/file</table>
</acpi>
</os>
To supply a path to a SLIC (Software Licensing) ACPI
table blob.
https://bugzilla.redhat.com/show_bug.cgi?id=1327537
Instead of setting the flag before parsing the PCI address, set
it afterwards. This ensure we can never end up in a situation
where the flag has been set but pci_dev.physical_function has
not been filled in.
Rather than only assigning a PCI address when no address is given at
all, also do it when the config says that the address type is 'pci',
but it gives no address (virDeviceInfoPCIAddressWanted()).
There are also several places after parsing but prior to address
assignment where code previously expected that any info with address
type='pci' would have a *valid* PCI address, which isn't always the
case - now we check not only for type='pci', but also for a valid
address (virDeviceInfoPCIAddressPresent()).
The test case added in this patch was directly copied from Cole's patch titled:
qemu: Wire up address type=pci auto_allocate
Prior to this, <address type='pci'/> wasn't allowed when parsing
(domain+bus+slot+function needed to be a "valid" PCI address, meaning
that at least one of domain/bus/slot had to be non-0), the RNG
required bus to be specified, and if type was set to PCI when
formatting, domain+bus+slot+function would always be output.
This makes all the address attributes optional during parse and RNG
validation, and suppresses domain+bus+slot+function if domain+bus+slot
are all 0 (NB: if d+b+s are all 0, any value for function is
nonsensical as that will never happen in the real world, and after
the next patch we will always assign a real working address to any
empty PCI address before it is ever output to anywhere).
Note that explicitly setting all attributes to 0 is equivalent to
setting none of them, which is okay, since 0000:00:00 is reserved in
any PCI bus setup, and can't be used anyway.
In order to allow <address type='pci'/> with no other attributes to
mean "I want a PCI address, but any PCI address will do" (just as
having no <address> at all usually indicates), we will need to change
several places in the code from a simple "info->type == (or !=)
VIR_DOMAIN_DEVICE_ADDRESS_TYPE_(PCI|NONE)" into something slightly
more complex, this patch adds to new functions that take a
virDomainDeviceInfoPtr and return true/false depending on 1) whether
the current state of the info indicates that we "want" a PCI address
for this device (virDeviceInfoPCIAddressWanted()) and 2) whether this
device already has a valid PCI address
(virDeviceInfoPCIAddressPresent()).
Both of these functions required the simpler check for whether a pci
address is "empty" (i.e. all of its attributes are 0, which can never
happen in a real PCI address, since slot 0 of bus 0 of domain 0 is
always reserved), so that function is also added.
Also moves all the subordinate structs. This is necessary due to a new
inline function that will be defined in device_conf.h, and also makes
sense, because it is the *device* info that's in the struct. (Actually
a lot more stuff from domain_conf.h could move to this newer file, but
I didn't want to disturb any more than necessary).
Move code that decide whether we print the 'listen' attribute or not
into virDomainGraphicsListenDefFormatAddr() function.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Name the validation function distinctively since it's called in the
parser. Later patches will add function that will validate disk
definitions that are invalid but need to be parsed to avoid losing
domains.
We support omitting listen attribute of graphics element so we should
also support omitting address attribute of listen element. This patch
also updates libvirt to always add a listen element into domain XML
except for VNC graphics if socket attribute is specified.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Move the compatibility code out of virDomainGraphicsListensParseXML()
into virDomainGraphicsListenDefParseXML(). This also fixes a small
inconsistency between the code and error message itself.
Before this patch we would search first listen element that is
type='address' to validate listen and address attributes. After this
patch we always take the first listen element regardless of the type.
This shouldn't break anything since all drivers supports only one
listen.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
If socket attribute is present we start VNC that listens only on that
unix socket. This makes the parser behave the same way as we actually
use the socket attribute.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Commit 82ba41108a made possible to use direct mapped iSCSI
volumes in qemu as disk sources but didn't remove the define time check.
Rework the check by simplifying the condition and allow any volumes to
be used with disk type='lun'.
Move filling out the default video (v)ram to DeviceDefPostParse.
This means it can be removed from virDomainVideoDefParseXML
and qemuParseCommandLine. Also, we no longer need to special case
VIR_DOMAIN_VIRT_XEN, since the per-driver callback gets called
before the generic one.
Commit 6879be48 moved adding of an implicit video device after XML
parsing. As a result, libxlDomainDeviceDefPostParse() is no longer
called to set the default vram when adding an implicit device.
Commit 6879be48 assumes virDomainVideoDefaultRAM() will set the
default vram, but it returns 0 if the domain virtType is
VIR_DOMAIN_VIRT_XEN. Attempting to start an HVM domain with vram=0
results in
error: unsupported configuration: videoram must be at least 4MB for CIRRUS
The default vram setting for Xen HVM domains depends on the device
model used (qemu-xen vs qemu-traditional), hence setting the
default is deferred to libxlDomainDeviceDefPostParse().
Call the device post-parse callback even for implicit video,
to fill out the default vram even for VIR_DOMAIN_VIRT_XEN.
https://bugzilla.redhat.com/show_bug.cgi?id=1334557
Most-of-commit-message-by: Jim Fehlig <jfehlig@suse.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1318993
Commit id 'dd519a294' caused a regression cloning a volume into a
logical pool by removing just the 'allocation' adjustment during
storageVolCreateXMLFrom. Combined with the change to not require the
new volume input XML to have a capacity listed (commit id 'e3f1d2a8')
left the possibility that a zero allocation value (e.g., not provided)
would create a thin/sparse logical volume. When a thin lv becomes fully
populated, then LVM sets the partition 'inactive' and the subsequent
fdatasync() fails.
Add a new 'has_allocation' flag to be set at XML parse time to indicate
that allocation was provided. This is done so that if it's not provided
the create-from code uses the capacity value since we document that if
omitted, the volume will be fully allocated at time of creation.
For a logical backend, that creation time is 'createVol', while for a
file backend, creation doesn't set the size, but the 'createRaw' called
during buildVolFrom will decide whether the file is sparse or not based
on the provided capacity and allocation value.
For volume clones that provide different allocation and capacity values
to allow for sparse files, there is no change.
Usage of this keyword in front of function declaration that is exported via a
header file is unnecessary, since internally, this has been the default for most
compilers for quite some time.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Commit 5ed235c6 added unnecessary redifinition of
virDomainCapsDeviceHostdev in conf/domain_capabilities.h. This breaks
build with clang 3.4:
In file included from conf/domain_capabilities.c:25:
conf/domain_capabilities.h:88:44: error: redefinition of typedef
'virDomainCapsDeviceHostdev' is a C11 feature
[-Werror,-Wtypedef-redefinition]
typedef struct _virDomainCapsDeviceHostdev virDomainCapsDeviceHostdev;
^
conf/domain_capabilities.h:86:44: note: previous definition is here
typedef struct _virDomainCapsDeviceHostdev virDomainCapsDeviceHostdev;
So drop one of those.
If the call to virXPathNodeSet to set naddresses fails, Coverity notes
that the subsequent VIR_ALLOC_N cannot have a negative value (well it
probably wouldn't be negative per se).
Signed-off-by: John Ferlan <jferlan@redhat.com>
Requires adding the plumbing for <device><video>
The value is <enum name='modelType'> to match the associated domain
XML of <video><model type='XXX'/>
Wire it up for qemu too
Commin 36785c7e refactored the code for input devices but introduced a
bug where we removed all keyboard from migratable XML. We have to
remove only implicit keyboards like PS2 or XEN.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Add the ability to add an 'iothread' to the controller which will be how
virtio-scsi-pci and virtio-scsi-ccw iothreads have been implemented in qemu.
Describe the new functionality and add tests to parse/validate that the
new attribute can be added.
This adds a ports= attribute to usb controller XML, like
<controller type='usb' model='nec-xhci' ports='8'/>
This maps to:
qemu -device nec-usb-xhci,p2=8,p3=8
Meaning, 8 ports that support both usb2 and usb3 devices. Gerd
suggested to just expose them as one knob.
https://bugzilla.redhat.com/show_bug.cgi?id=1271408
If a panic device is being defined without a model in a domain
the default value is always overwritten with model ISA. An ISA
bus does not exist on S390 and therefore specifying a panic device
results in an unsupported configuration.
Since the S390 architecture inherently provides a crash detection
capability the panic device should be defined in the domain xml.
This patch adds an s390 panic device model and prevents setting a
device address on it.
Signed-off-by: Boris Fiuczynski <fiuczy@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
libvirt-daemon-config-nwfilter will put a bunch of xml configs
into /etc/libvirt/nwfilter. These configs don't hardcode a UUID
and depends on libvirt to generate one. However the generated UUID
is never saved to disk, unless the user manually calls Define.
This makes daemon reload quite noisy with many errors like:
error : virNWFilterObjAssignDef:3101 : operation failed: filter 'allow-incoming-ipv4' already exists with uuid 50def3b5-48d6-46a3-b005-cc22df4e5c5c
Because a new UUID is generated every time the config is read from
disk, so libvirt constantly thinks it's finding a new nwfilter.
Detect if we generated a UUID when the config file is loaded; if so,
resave the new contents to disk to ensure the UUID is persisteny.
This is similar to what was done in commit a47ae7c0 with virtual
networks and generated MAC addresses
In virNWFilterObjLoad we can still fail after virNWFilterObjAssignDef,
but we don't unlock and free the created virNWFilterObjPtr in the
cleanup path.
The bit we are trying to do after AssignDef is just STRDUP in the
configFile path. However caching the configFile in the NWFilterObj
is largely redundant and doesn't follow the same pattern we use
for domain and network objects.
So just remove all the configFile caching which fixes the latent
bug as a side effect.
We historically format runtime seclabel selinux/apparmor values,
however we skip formatting runtime DAC values. This was added in
commit 990e46c454
Author: Marcelo Cerri <mhcerri@linux.vnet.ibm.com>
Date: Fri Aug 31 13:40:41 2012 +0200
conf: Avoid formatting auto-generated DAC labels
to maintain migration compatibility with libvirt < 0.10.0.
However the formatting was skipped unconditionally. Instead only
skip formatting in the VIR_DOMAIN_DEF_FORMAT_MIGRATABLE case.
https://bugzilla.redhat.com/show_bug.cgi?id=1215833
Trying to define a pool name containing an embedded '/'
will immediately fail when trying to write the XML to disk.
This patch explicitly rejects names containing a '/'
Besides our stateful driver, there are two other storage impls:
esx and phyp. esx doesn't support pool creation, so this should
doesn't apply.
phyp does support pool creation, and the name is passed to the
'mksp' tool, which google doesn't reveal whether it accepts '/'
or not. IMO the likeliness of this impacting any users is near zero
Trying to define a network name containing an embedded '/'
will immediately fail when trying to write the XML to disk.
This patch explicitly rejects names containing a '/'
Besides the network bridge driver, the only other network
implementation is a very thin one for virtualbox, which seems to
use the network name as a host interface name, which won't
accept '/' anyways, so I think this is fine to do unconitionally.
https://bugzilla.redhat.com/show_bug.cgi?id=787604
Trying to define a domain name containing an embedded '/'
will immediately fail when trying to write the XML to disk for
our stateful drivers. This patch explicitly rejects names
containing a '/', and provides an xmlopt feature for drivers
to avoid this validation check, which is enabled in every
non-stateful driver that already has xmlopt handling wired up.
(Technically this could reject a previously accepted vmname like
'/foo', however at least for the qemu driver that falls over
later when starting qemu)
https://bugzilla.redhat.com/show_bug.cgi?id=639923
We were lacking tests that are checking for the completeness of our
nodedev XMLs and also whether we output properly formatted ones. This
patch adds parsing for the capability elements inside the <capability
type='pci'> element. Also bunch of tests are added to show everything
works properly.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
We had both and the only difference was that the latter also included
information about multifunction setting. The problem with that was that
we couldn't use functions made for only one of the structs (e.g.
parsing). To consolidate those two structs, use the one in virpci.h,
include that in domain_conf.h and add the multifunction member in it.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Modeled after the qemuDomainDiskPrivatePtr logic, create a privateData
pointer in the _virDomainHostdevDef to allow storage of private data
for a hypervisor in order to at least temporarily store auth/secrets
data for usage during qemuBuildCommandLine.
NB: Since the qemu_parse_command (qemuParseCommandLine) code is not
expecting to restore the auth/secret data, there's no need to add
code to handle this new structure there.
Updated copyrights for modules touched. Some didn't have updates in a
couple years even though changes have been made.
Signed-off-by: John Ferlan <jferlan@redhat.com>
This reverts commit 690969af9c, which
added the domain config parts to support a "peer" attribute in domain
interface <ip> elements.
It's being removed temporarily for the release of libvirt 1.3.4
because the feature doesn't work, and there are concerns that it may
need to be modified in an externally visible manner which could create
backward compatibility problems.
Similarly to what commit 7140807917 did with some internal paths,
clear vnc socket paths that were generated by us. Having such path in
the definition can cause trouble when restoring the domain. The path is
generated to the per-domain directory that contains the domain ID.
However, that ID will be different upon restoration, so qemu won't be
able to create that socket because the directory will not be prepared.
To be able to migrate to older libvirt, skip formatting the socket path
in migratable XML if it was autogenerated. And mark it as autogenerated
if it already exists and we're parsing live XML.
Best viewed with '-C'.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1326270
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Add virDomainObjGetShortName() and use it. For now that's used in one
place, but we should expose it so that future patches can use it.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Currently we only allow /dev/random and /dev/hwrng as host input
for <rng><backend model='random'/> device. This was added after
various upstream discussions in commit 4932ef45
However this restriction has generated quite a few complaints over
the years, so a new discussion was initiated:
http://www.redhat.com/archives/libvir-list/2016-April/msg00987.html
Several people suggested removing the restriction, and nobody really
spoke up to defend it. So this patch drops the path restriction
entirely
https://bugzilla.redhat.com/show_bug.cgi?id=1074464
Introduce the final accessor's to _virSecretObject data and move the
structure from virsecretobj.h to virsecretobj.c
The virSecretObjSetValue logic will handle setting both the secret
value and the value_size. Some slight adjustments to the error path
over what was in secretSetValue were made.
Additionally, a slight logic change in secretGetValue where we'll
check for the internalFlags and error out before checking for
and erroring out for a NULL secret->value. That way, it won't be
obvious to anyone that the secret value wasn't set rather they'll
just know they cannot get the secret value since it's private.
Move and rename the secretRewriteFile, secretSaveDef, and secretSaveValue
from secret_driver to virsecretobj
Need to make some slight adjustments since the secretSave* functions
called secretEnsureDirectory, but otherwise mostly just a move of code.
Move and rename secretDeleteSaved from secret_driver into virsecretobj and
split it up into two parts since there is error path code that looks to
just delete the secret data file
Move to secret_conf.c and rename to virSecretLoadAllConfigs. Also includes
moving/renaming the supporting virSecretLoad, virSecretLoadValue, and
virSecretLoadValidateUUID.
This patch replaces most of the guts of secret_driver.c with recently
added secret_conf.c APIs in order manage secret lists and objects
using the hashed virSecretObjList* lookup API's.
Add function to return a "match" filtered list of secret objects. This
function replaces the guts of secretConnectListAllSecrets.
Need to also move and make global virSecretUsageIDForDef since it'll
be used by both secret_driver.c and secret_conf.c
Add the functions to add/remove elements from the hashed secret obj list.
These will replace secret_driver functions secretAssignDef and secretObjRemove.
The virSecretObjListAddLocked will perform the necessary lookups and
decide whether to replace an existing hash entry or create a new one.
This includes setting up the configPath and base64Path as well as being
able to support the caller's need to restore from a previous definition
in case something goes wrong in the caller.
New API's including unlocked and Locked versions in order to be able
to use in either manner.
Support for searching hash object lists instead of linked lists will
replace existing secret_driver functions secretFindByUUID and
secretFindByUsage
Move virSecretObj from secret_driver.c to virsecretobj.h
To support being able to create a hashed secrets list, move the
virSecretObj to virsecretobj.h so that the code can at least find
the definition.
This should be a temporary situation while the virsecretobj.c code
is patched in order to support a hashed secret object while still
having the linked list support in secret_driver.c. Eventually, the
goal is to move the virSecretObj into virsecretobj.c, although it
is notable that the existing model from which virSecretObj was
derived has virDomainObj in src/conf/domain_conf.h and virNetworkObj
in src/conf/network_conf.h, so virSecretObj wouldn't be unique if
it were to remain in virsecretobj.h Still adding accessors to fetch
and store hashed object data will be the end goal.
Add definitions and infrastucture in virsecretobj.c to create and
handle a hashed virSecretObj and virSecretObjList including the class,
object, lock setup, and disposal API's. Nothing will call these yet.
This infrastructure will replace the forward linked list logic
within the secret_driver, eventually.
VIR_ERR_NO_SUPPORT maps to the error string
this function is not supported by the connection driver
and is largely only used for when a driver doesn't have any
implementation for a public API. So its usage with invalid
net-update requests is a bit out of place. Instead use
VIR_ERR_OPERATION_UNSUPPORTED which maps to:
Operation not supported
And is what qemu's hotplug routines use in similar scenarios
The struct contains a single boolean field, 'supported':
the meaning of this field is too generic to be limited to
devices only, and in fact it's already being used for
other things like loaders and OSs.
Instead of trying to come up with a more generic name just
get rid of the struct altogether.
Prior to this patch we didn't make any attempt to prevent two entries
in the array of interfaces/PCI devices from pointing to the same
device.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1002423
So in glibc-2.23 sys/sysmacros.h is no longer included from sys/types.h
and we don't build because of the usage of major/minor/makedev macros.
Autoconf already has AC_HEADER_MAJOR macro that check where exactly
these functions/macros are defined, so let's use that.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Ploop image consists of directory with two files: ploop image itself,
called root.hds and DiskDescriptor.xml that contains information about
ploop device: https://openvz.org/Ploop/format.
Such volume are difficult to manipulate in terms of existing volume types
because they are neither a single files nor a directory.
This patch introduces new volume type - ploop. This volume type is used
by ploop volume's exclusively.
Signed-off-by: Olga Krishtal <okrishtal@virtuozzo.com>
Signed-off-by: Ján Tomko <jtomko@redhat.com>
This controller provides a single PCIe port on a new root. It is
similar to pci-expander-bus, intended to provide a bus that can be
associated with a guest-identifiable NUMA node, but is for
machinetypes with PCIe rather than PCI (e.g. q35-based machinetypes).
Aside from PCIe vs. PCI, the other main difference is that a
pci-expander-bus has a companion pci-bridge that is automatically
attached along with it, but pcie-expander-bus has only a single port,
and that port will only connect to a pcie-root-port, or to a
pcie-switch-upstream-port. In order for the bus to be of any use in
the guest, it must have either a pcie-root-port or a
pcie-switch-upstream-port attached (and one or more
pcie-switch-downstream-ports attached to the
pcie-switch-upstream-port).
This is a standard PCI root bus (not a bridge) that can be added to a
440fx-based domain. Although it uses a PCI slot, this is *not* how it
is connected into the PCI bus hierarchy, but is only used for
control. Each pci-expander-bus provides 32 slots (0-31) that can
accept hotplug of standard PCI devices.
The usefulness of pci-expander-bus relative to a pci-bridge is that
the NUMA node of the bus can be specified with the <node> subelement
of <target>. This gives guest-side visibility to the NUMA node of
attached devices (presuming that management apps only assign a device
to a bus that has a NUMA node number matching the node number of the
device on the host).
Each pci-expander-bus also has a "busNr" attribute. The expander-bus
itself will take the busNr specified, and all buses that are connected
to this bus (including the pci-bridge that is automatically added to
any expander bus of model "pxb" (see the next commit)) will use
busNr+1, busNr+2, etc, and the pci-root (or the expander-bus with next
lower busNr) will use bus numbers lower than busNr.
There are two places in qemu_domain_address.c where we have a switch
statement to convert PCI controller models
(VIR_DOMAIN_CONTROLLER_MODEL_PCI*) into the connection type flag that
is matched when looking for an upstream connection for that model of
controller (VIR_PCI_CONNECT_TYPE_*). This patch makes a utility
function in conf/domain_addr.c to do that, so that when a new PCI
controller is added, we only need to add the new model-->connect-type
in a single place.
The flags used to determine which devices could be plugged into which
controllers were quite confusing, as they tried to create classes of
connections, then put particular devices into possibly multiple
classes, while sometimes setting multiple flags for the controllers
themselves. The attempt to have a single flag indicate, e.g. that a
root-port or a switch-downstream-port could connect was not only
confusing, it was leading to a situation where it would be impossible
to specify exactly the right combinations for a new controller.
The solution is for the VIR_PCI_CONNECT_TYPE_* flags to have a 1:1
correspondence with each type of PCI controller, plus a flag for a PCI
endpoint device and another for a PCIe endpoint device (the only
exception to this is that pci-bridge and pcie-expander-bus controllers
have their upstream connection classified as
VIR_PCI_CONNECT_TYPE_PCI_DEVICE since they can be plugged into
*exactly* the same ports as any endpoint device). Each device then
has a single flag for connect type (plus the HOTPLUG flag if that
device can e hotplugged), and each controller sets the CONNECT bits
for all controllers that can be plugged into it, as well as for either
type of endpoint device that can be plugged in (and the HOTPLUG flag
if it can accept hotplugged devices).
With this change, it is *slightly* easier to understand the matching
of connections (as long as you remember that the flag for a
device/upstream-facing connection of a controller is the same as that
device's type, while the flags for a controller's downstream
connections is the OR of all device types that can be plugged into
that controller). More importantly, it will be possible to correctly
specify what can be plugged into a pcie-switch-expander-bus, when
support for it is added.
When support for dmi-to-pci-bridge was added, it was assumed that,
just as with the pci-root bus, slot 0 was reserved. This is not the
case - it can be used to connect a device just like any other slot, so
remove the restriction and update the test cases that auto-assign an
address on a dmi-to-pci-bridge.
Every other maxSlot was either set to 0 or to
VIR_PCI_ADDRESS_SLOT_LAST, but this one was for some reason set to the
literal value 31 (which is the same as VIR_PCI_ADDRESS_SLOT_LAST).
This makes them all consistent.
GCC in RHEL-6 complains about listen:
../../src/conf/domain_conf.c:23718: error: declaration of 'listen' shadows a global declaration [-Wshadow]
/usr/include/sys/socket.h:204: error: shadowed declaration is here [-Wshadow]
This renames all the listen to gListen.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Since we didn't opt to use one single event for device lifecycle for a
VM we are missing one last event if the device removal failed. This
event will be emitted once we asked to eject the device but for some
reason it is not possible.
Instead of calling the virDomainGraphicsListensParseXML function for all
graphics types and ignore the wrong ones move the call only to graphics
types where we supports listen elements.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Those are the last two places that uses the getter functions. Use a
direct access instead and remove those getters.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Removes the check for graphics type, it's not a public API and developer
know what he's doing and this check makes no sense. It also removes
the ability to allocate a new array if there is none. This was used by
the virDomainGraphicsListenAdd* functions and isn't used anymore.
This is now a simple getter with simple check for listens array presence
and whether the index in out of bounds.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
This effectively removes virDomainGraphicsListenSetAddress which was
used only to change the address of listen structure and possible change
the listen type. The new function will auto-expand the listens array
and append a new listen.
The old function was used on pre-allocated array of listens and in most
cases it only "add" a new listen. The two remaining uses can access the
listen structure directly.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Rest of the fields of the iotune data structure did not check for
malformed integers. Use the previously defined macro to extract them
which will simplify the code and add error reporting.
Since the structure was pre-initialized to 0 we don't need to set every
single member to 0 if it's not present in the XML. Additionally if we
put the name of the field into the error message the code can be
simplified using a macro to parse the members.
If we encounter a video device with primary=yes, we insert it
at def->videos[0].
There is no need to record this in a separate variable,
just check if there already is a primary video at def->videos[0].
We call VIR_INSERT_ELEMENT_INPLACE either with 0 (for primary video)
or def->nvideos (for the rest).
Use a variable with more semantic name, since j is usually used
for iterating.
We start with both i and def->nvideos at 0 and increment both
after every successful iteration.
Use i directly, instead of passing the def->nvideos value through j.
Commit 119cd06 started setting the primary bool for the first
user-specified video even if user omitted the 'primary' attribute.
However this was done before the addition of the implicit device.
This broke startup of transient qemu domains with no <video>:
https://bugzilla.redhat.com/show_bug.cgi?id=1325757
Move this default to virDomainDefPostParseInternal,
after the addition of the implicit video device, to catch the implicit
video as well.
Commit dc98a5bc refactored the code a lot and forget about checking if
listen attribute is specified. This ensures that listen attribute and
first listen element are compared only if both exist.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Commit d77ffb6876 added not only reporting of the PCI header type, but
also parsing of that information. However, because there was no parsing
done for the other sub-PCI capabilities, if there was any other
capability then a valid header type name (like phys_function or
virt_functions) the parsing would fail. This prevented passing node
device XMLs that we generated into our own functions when dealing with,
e.g. with SRIOV cards.
Instead of reworking the whole parsing, just fix this one occurence and
remove a test for it for the time being. Future patches will deal with
the rest.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Create a bitmap of iothreads that have scheduler info set so that the
transformation algorithm does not have to iterate the empty bitmap many
times. By reusing self-expanding bitmaps the bitmap size does not need
to be pre-calculated.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1264008
This patch adds new xml element, and so we can have the option of
also having perf events enabled immediately at startup.
Signed-off-by: Qiaowei Ren <qiaowei.ren@intel.com>
Message-id: 1459171833-26416-6-git-send-email-qiaowei.ren@intel.com
This patch adds support for "vpindex", "runtime", "synic",
"stimer", and "vendor_id" features available in qemu 2.5+.
- When Hyper-V "vpindex" is on, guest can use MSR HV_X64_MSR_VP_INDEX
to get virtual processor ID.
- Hyper-V "runtime" enlightement feature allows to use MSR
HV_X64_MSR_VP_RUNTIME to get the time the virtual processor consumes
running guest code, as well as the time the hypervisor spends running
code on behalf of that guest.
- Hyper-V "synic" stands for Synthetic Interrupt Controller, which is
lapic extension controlled via MSRs.
- Hyper-V "stimer" switches on Hyper-V SynIC timers MSR's support.
Guest can setup and use fired by host events (SynIC interrupt and
appropriate timer expiration message) as guest clock events
- Hyper-V "reset" allows guest to reset VM.
- Hyper-V "vendor_id" exposes hypervisor vendor id to guest.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Signed-off-by: John Ferlan <jferlan@redhat.com>
1. All hyperv features are tristate ones. So make tristate generating part common.
2. Reduce nesting on spinlocks.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Signed-off-by: John Ferlan <jferlan@redhat.com>
1. All hyperv features are tristate ones. So make tristate parsing code common.
2. Reindent switch statement.
3. Reduce nesting in spinlocks parsing.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Signed-off-by: John Ferlan <jferlan@redhat.com>
When reading in an XML definition for a SCSI target device, the name
property of struct scsi_target refers to the @target element.
Let's fix this obvious typo and also extend the XML schema to provide
validation.
Signed-off-by: Bjoern Walk <bwalk@linux.vnet.ibm.com>
Most hypervisors use Hardware Assisted Paging by default and don't
require specifying the feature in domain conf. But some hypervisors
support disabling HAP on a per-domain basis. To enable HAP by default
yet provide a knob to disable it, extend the <hap> feature with a
'state=on|off' attribute, similar to <pvspinlock> and <vmport> features.
In the absence of <hap>, the hypervisor default (on) is used. <hap>
without the state attribute would be the same as <hap state='on'/> for
backwards compatibility. And of course <hap state='off'/> disables hap.
Signed-off-by: Jim Fehlig <jfehlig@suse.com>
VIR_DOMAIN_EVENT_SUSPENDED_POSTCOPY and VIR_DOMAIN_PAUSED_POSTCOPY are
used on the source host once migration enters post-copy mode (which
means the domain gets paused on the source. After the destination host
takes over the execution of the domain, its virtual CPUs are resumed and
the domain enters VIR_DOMAIN_RUNNING_POSTCOPY state and
VIR_DOMAIN_EVENT_RESUMED_POSTCOPY event is emitted.
In case migration fails during post-copy mode and none of the hosts have
complete state of the domain, both domains will remain paused with
VIR_DOMAIN_PAUSED_POSTCOPY_FAILED reason and an upper layer may decide
what to do.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
It's just a combination of AddImplicitControllers, and AddConsoleCompat.
Every caller that wants ImplicitControllers also wants the ConsoleCompat
AFAICT, so lump them together. We also need it for future patches.
Judging by how the whitelist has skewed quite far from the original
error message, I think it's better to just drop these.
If someone wants to revive this check I suggest implementing it on
a per-HV driver basis with PostParse callbacks.
If we expose this information, which is one byte in every PCI config
file, we let all mgmt apps know whether the device itself is an endpoint
or not so it's easier for them to decide whether such device can be
passed through into a VM (endpoint) or not (*-bridge).
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1317531
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Commit id '4f846170' added printing of a new field 'part_separator';
however, neglected to do so when there was an "freeExtent" defined
for the device (as there would be when the disk pool was started).
This patch adjusts the logic to appropriately format the device path and
if there the part_separator attribute.
Just a cleanup I stumbled upon in one of my older branches I did when
browsing through some code and forgot to send it.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Add new function to manage adding the disk -drive options to the
command line removing that task from the mainline qemuBuildCommandLine.
Also since using const virDomainDef in new function, that means other
functions called needed to change their usage.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Add new function to manage adding the controller -device options to the
command line removing that task from the mainline qemuBuildCommandLine.
Also adjust to using const virDomainDef instead of virDomainDefPtr.
This causes collateral damage in order to modify called APIs to use
the const virDomainDef instead as well.
Signed-off-by: John Ferlan <jferlan@redhat.com>
Extend the chardev source XML so that there is a new optional
<log/> element, which is applicable to all character device
backend types. For example, to log output of a TCP backed
serial port
<serial type='tcp'>
<source mode='connect' host='127.0.0.1' service='9999'/>
<protocol type='raw'/>
<log file='/var/log/libvirt/qemu/demo-serial0.log' append='on'/>
<target port='0'/>
</serial>
Not all hypervisors will support use of logfiles.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Introduce VIR_DOMAIN_DEF_FEATURE_OFFLINE_VCPUPIN domain feature flag
whcih will allow to skip ignoring of the pinning information for
hypervisor drivers which will want to implement forward-pinning of
vcpus.
Introduce a helper to check supported device and domain config and move
the memory hotplug checks to it.
The advantage of this approach is that by default all new features are
considered unsupported by all hypervisors unless specifically changed
rather than the previous approach where every hypervisor would need to
declare that a given feature is unsupported.
To avoid having to forbid new features added to domain XML in post parse
callbacks for individual hypervisor drivers the feature flag mechanism
will allow to add a central check that will be disabled for the drivers
that will add support.
As a first example flag, the 'hasWideSCSIBus' is converted to the new
bitmap.
The VIR_DOMAIN_EVENT_ID_JOB_COMPLETED event will be triggered once a job
(such as migration) finishes and it will contain statistics for the job
as one would get by calling virDomainGetJobStats. Thanks to this event
it is now possible to get statistics of a completed migration of a
transient domain on the source host.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
While trying to build with -Os couple of compile errors showed
up.
conf/domain_conf.c: In function 'virDomainChrRemove':
conf/domain_conf.c:13666:24: error: 'ret' may be used uninitialized in this function [-Werror=maybe-uninitialized]
virDomainChrDefPtr ret, **arrPtr = NULL;
^
Compiler fails to see that @ret is used only if set in the loop,
but whatever, there's no harm in initializing the variable.
In vboxAttachDrivesNew and _vboxAttachDrivesOld compiler thinks
that @rc may be used uninitialized. Well, not directly, but maybe
after some optimization. Yet again, no harm in initializing a
variable.
In file included from ./util/virthread.h:26:0,
from ./datatypes.h:28,
from vbox/vbox_tmpl.c:43,
from vbox/vbox_V3_1.c:37:
vbox/vbox_tmpl.c: In function '_vboxAttachDrivesOld':
./util/virerror.h:181:5: error: 'rc' may be used uninitialized in this function [-Werror=maybe-uninitialized]
virReportErrorHelper(VIR_FROM_THIS, code, __FILE__, \
^
In file included from vbox/vbox_V3_1.c:37:0:
vbox/vbox_tmpl.c:1041:14: note: 'rc' was declared here
nsresult rc;
^
Yet again, one uninitialized variable:
qemu/qemu_driver.c: In function 'qemuDomainBlockCommit':
qemu/qemu_driver.c:17194:9: error: 'baseSource' may be used uninitialized in this function [-Werror=maybe-uninitialized]
qemuDomainPrepareDiskChainElement(driver, vm, baseSource,
^
And another one:
storage/storage_backend_logical.c: In function 'virStorageBackendLogicalMatchPoolSource.isra.2':
storage/storage_backend_logical.c:618:33: error: 'thisSource' may be used uninitialized in this function [-Werror=maybe-uninitialized]
thisSource->devices[j].path))
^
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
While trying to build with -Os I've encountered some build
failures.
util/vircommand.c: In function 'virCommandAddEnvFormat':
util/vircommand.c:1257:1: error: inlining failed in call to 'virCommandAddEnv': call is unlikely and code size would grow [-Werror=inline]
virCommandAddEnv(virCommandPtr cmd, char *env)
^
util/vircommand.c:1308:5: error: called from here [-Werror=inline]
virCommandAddEnv(cmd, env);
^
This function is big enough for the compiler to be not inlined.
This is the error message I'm seeing:
Then virDomainNumatuneNodeSpecified is exported and called from
other places. It shouldn't be inlined then.
In file included from network/bridge_driver_platform.h:30:0,
from network/bridge_driver_platform.c:26:
network/bridge_driver_linux.c: In function 'networkRemoveRoutingFirewallRules':
./conf/network_conf.h:350:1: error: inlining failed in call to 'virNetworkDefForwardIf.constprop': call is unlikely and code size would grow [-Werror=inline]
virNetworkDefForwardIf(const virNetworkDef *def, size_t n)
^
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Prior to commit id '3d021381' virDomainObjUpdateModificationImpact was
part of virDomainLiveConfigHelperMethod and the *flags if condition
VIR_DOMAIN_AFFECT_CONFIG checked the ->persistent boolean and made the
virDomainObjGetPersistentDef call.
Since the functions were split the ->persistent check is all that remained
and thus could be combined into one if statement.
Signed-off-by: Nikolay Shirokovskiy <nshirokovskiy@virtuozzo.com>
Signed-off-by: John Ferlan <jferlan@redhat.com>
This parameter represents top level period cgroup
that limits whole domain enforcement period for a quota
Signed-off-by: Alexander Burluka <aburluka@virtuozzo.com>
This attribute is used to extend secondary PCI bar and expose it to the
guest as 64bit memory. It works like this: attribute vram is there to
set size of secondary PCI bar and guest sees it as 32bit memory,
attribute vram64 can extend this secondary PCI bar. If both attributes
are used, guest sees two memory bars, both address the same memory, with
the difference that the 32bit bar can address only the first part of the
whole memory.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1260749
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
We always place primary video device at first place, to make it easier
to create a qemu command or format an xml, but we should also set the
primary boolean for primary video device to 'true'.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Add Spice graphics gl attribute. qemu 2.6 should have -spice gl=on argument to
enable opengl rendering context (patches on the ML). This is necessary to
actually enable virgl rendering.
Add a qemuxml2argv test for virtio-gpu + spice with virgl.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
While reviewing how storage driver used ObjListPtr's for reference
in some recent secret driver patches to use the same mechanism, I came
across an instance where the wrong API was called for error paths after
successfully allocating the storage pool pointer and inserting into
the driver pool list.
The path is after virStoragePoolObjAssignDef succeeds - the 'def' passed
in is assigned to pool->def (or newDef) so it shouldn't be the only thing
deleted. The pool is now part of driver->pools.objs, so it would need to
be removed (as happens in the storagePoolCreateXML error paths).
Rather than calling virStoragePoolDefFree to free the def which is now
assigned to the pool, call virStoragePoolObjRemove to ensure the pool
element is removed from the driver list and that anything stored in pool
is properly handled by virStoragePoolObjFree including the call to
virStoragePoolDefFree for the pool->{def|newDef} element.
Checking whether x > 0 before looping over [0..x] items doesn't make
sense and multi-line body must have curly brackets around it.
Best viewed with '-w'.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
There's a check if a domain definition has any graphics card and
if so, we iterate over each one of them. This makes no sense,
because even if it has none we can still iterate over.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Seems like the natural fit, since we are already adding other XML bits
in the PostParse routine.
Previously AddImplicitControllers was only called at the end of XML
parsing, meaning code that builds a DomainDef by hand had to manually
call it. Now those PostParse callers get it for free.
There's some test churn here; xen xm and sexpr test suite bits weren't
calling this before, but now they are, so you'll see new IDE controllers.
I don't think this will cause problems in practice, since the code already
needs to handle these implicit controllers like in the case when a user
defines their own XML.
virDomainObjWait is designed to be called in a loop. Make sure we break
the loop in case the domain dies to avoid waiting for an event which
will never happen.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Commit f1a89a8 allowed parsing configs from /etc/libvirt
without validating the emulator capabilities.
Check for the presence of os->type.machine even if the
VIR_DOMAIN_DEF_PARSE_SKIP_OSTYPE_CHECKS flag is set,
otherwise the daemon can crash on carelessly crafted input
in the config directory.
https://bugzilla.redhat.com/show_bug.cgi?id=1267256
Our existing virHashForEach method iterates through all items disregarding the
fact, that some of the iterators might have actually failed. Errors are usually
dispatched through an error element in opaque data which then causes the
original caller of virHashForEach to return -1. In that case, virHashForEach
could return as soon as one of the iterators fail. This patch changes the
iterator return type and adjusts all of its instances accordingly, so the
actual refactor of virHashForEach method can be dealt with later.
Signed-off-by: Erik Skultety <eskultet@redhat.com>
Since no value in the virGICVersion enumeration is negative, a clever
enough compiler can report an error such as
src/conf/domain_conf.c:15337:75: error: comparison of unsigned enum
expression < 0 is always false [-Werror,-Wtautological-compare]
if ((def->gic_version = virGICVersionTypeFromString(tmp)) < 0 ||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ ~
virGICVersionTypeFromString() can, however, return a negative value if
the input string is not part of the enumeration, so we definitely need
that check.
Work around the problem by storing the return value in a temporary int
variable.
The snapshot name generator truncates the original file name after a '.'
and replaces the suffix with the snapshot name. If two disks source
images differ only in the suffix portion, the generated name will be
duplicate.
Since this is a corner case just error out stating that a duplicate name
was generated. The user can work around this situation by providing
the file names explicitly.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1283085
Older gcc fails to see that the variable is set iff @hasPriority
== true in which case the former is set a value. Initialize the
value while declaring it to make the compiler shut up.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Due to bad design the vcpu sched element is orthogonal to the way how
the data belongs to the corresponding objects. Now that vcpus are a
struct that allow to store other info too, let's convert the data to the
sane structure.
The helpers for the conversion are made universal so that they can be
reused for iothreads too.
This patch also resolves https://bugzilla.redhat.com/show_bug.cgi?id=1235180
since with the correct storage approach you can't have dangling data.
