Peter Krempa [Tue, 24 Apr 2018 15:12:19 +0000 (17:12 +0200)]
qemu: block: Properly handle block storage in JSON generator
Block storage should actually be passed to qemu via 'host_device' or
'host_cdrom' according to the device type. There were no users of this
behaviour so we thankfully can change it.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Tue, 24 Apr 2018 14:59:00 +0000 (16:59 +0200)]
qemu: domain: Store whether a virStorageSource is a host CDROM drive
Use virFileIsCDROM to detect whether a block device is a cdrom drive and
store it in virStorageSource. This will be necessary to correctly create
the 'host_cdrom' backend in qemu when using -blockdev.
We assume that host_cdrom makes only sense when used directly as a raw
image, but if a backing chain would be put in front of it, libvirt will
use 'host_device' in that case.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Mon, 23 Apr 2018 15:41:14 +0000 (17:41 +0200)]
util: file: Add helper to determine whether a path is a CDROM
Add detection mechanism which will allow to check whether a path to a
block device is a physical CDROM drive. This will be useful once we will
need to pass it to hypervisors.
The linux implementation uses an ioctl to do the detection, while the
fallback uses a simple string prefix match.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Thu, 29 Mar 2018 06:44:02 +0000 (08:44 +0200)]
qemu: block: Add support for accessing directories via the 'vvfat' driver
Handle VIR_STORAGE_TYPE_DIR in qemuBlockStorageSourceGetBackendProps so
that a 'vvfat' driver is used, which emulates a FAT filesystem
containing the folders.
qemu requires us to add it as a storage layer, since a 'raw' layer is
usually put on top of it.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Thu, 19 Apr 2018 13:31:55 +0000 (15:31 +0200)]
utils: storage: Mark that a virStorageSource is going to be used as a floppy
Add a flag denoting that a virStorageSource is going to be used as a
floppy image. This will be useful in cases where the user passes in
files which shall be exposed as an image to the guest.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Fri, 23 Mar 2018 16:00:29 +0000 (17:00 +0100)]
qemu: domain: Carefuly transfer configuration from disk to storage source
Some properties don't make sense to be configured for every single layer
of the backing chain, but to avoid needing to pass the disk structure we
will copy them to the individual virStorageSource.
Zero detection is applied only for the top layer image, while caching
and iomode for all layers.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Tue, 5 Dec 2017 13:06:13 +0000 (14:06 +0100)]
util: storage: Add shadow copies of few disk properties to virStorageSource
Few things which are currently stored the virDomainDiskDef structure are
actually relevant for the storage source as well. Add the fields with a
note that they are just mirror of the values from the disk.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Peter Krempa [Wed, 11 Apr 2018 13:35:02 +0000 (15:35 +0200)]
storage: Properly track that backing chain members are readonly
Everything besides the top of the chain is readonly. Track this when
parsing the XML and detecting the chain from the disk. Also fix the
state when taking snapshots.
All other cases where the top image is changed already preserve the
readonly state from the original image.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Ján Tomko [Thu, 3 May 2018 08:38:42 +0000 (10:38 +0200)]
Deprecate QEMU_CAPS_NO_KVM_PIT
The -no-kvm-pit-reinjection option has been deprecated since
its introduction in QEMU 1.3. See commit <1569fa1>.
Drop the capability since all the QEMUs we support allow tuning
the kvm-pit properties via -global.
Also add the QEMU_CAPS_KVM_PIT_TICK_POLICY to the clock-catchup
tests, since expecting it to succeed with QEMU that does not
have kvm-pit makes no sense.
Signed-off-by: Ján Tomko <jtomko@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Ján Tomko [Tue, 17 Apr 2018 21:00:33 +0000 (23:00 +0200)]
storage: remove qemu-img help scraping
We have been checking whether qemu-img supports the -o compat
option by scraping the -help output.
Since we require QEMU 1.5.0 now and this option was introduced in 1.1,
assume we support it and ditch the help parsing code along with the
extra qemu-img invocation.
Signed-off-by: Ján Tomko <jtomko@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Ján Tomko [Tue, 17 Apr 2018 21:25:45 +0000 (23:25 +0200)]
tests: delete most qemu-img test cases assuming FMT_OPTIONS
We have two leftover "capabilites" for qemu-img:
QEMU_IMG_BACKING_FORMAT_OPTIONS
QEMU_IMG_BACKING_FORMAT_OPTIONS_COMPAT
The former says we are able to specify the backing format via -o
(which has been the case for a long time now) and the second one
says we can use -o compat to specify the qcow2 version.
Since we require QEMU 1.5.0, we can always assume -o compat,
which was introduced in QEMU 1.1.
Drop the test cases using FMT_OPTIONS which have a FMT_COMPAT
counterpart to prepare for deprecating FMT_OPTIONS (and these flags)
completely.
Signed-off-by: Ján Tomko <jtomko@redhat.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
John Ferlan [Mon, 23 Apr 2018 14:40:48 +0000 (10:40 -0400)]
conf: Clean up object referencing for Add and Remove
When adding a new object to the domain object list, there should
have been 2 virObjectRef calls made one for each list into which
the object was placed to match the 2 virObjectUnref calls that
would occur during Remove as part of virHashRemoveEntry when
virObjectFreeHashData is called when the element is removed from
the hash table as set up in virDomainObjListNew.
Some drivers (libxl, lxc, qemu, and vz) handled this inconsistency
by calling virObjectRef upon successful return from virDomainObjListAdd
in order to use virDomainObjEndAPI when done with the returned @vm.
While others (bhyve, openvz, test, and vmware) handled this via only
calling virObjectUnlock upon successful return from virDomainObjListAdd.
This patch will "unify" the approach to use virDomainObjEndAPI
for any @vm successfully returned from virDomainObjListAdd.
Because list removal is so tightly coupled with list addition,
this patch fixes the list removal algorithm to return the object
as entered - "locked and reffed". This way, the callers can then
decide how to uniformly handle add/remove success and failure.
This removes the onus on the caller to "specially handle" the
@vm during removal processing.
The Add/Remove logic allows for some logic simplification such
as in libxl where we can Remove the @vm directly rather than
needing to set a @remove_dom boolean and removing after the
libxlDomainObjEndJob completes as the @vm is locked/reffed.
Signed-off-by: John Ferlan <jferlan@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
John Ferlan [Mon, 23 Apr 2018 15:58:54 +0000 (11:58 -0400)]
libxl: Add refcnt for args->conn during migration
Since the @dconn reference via args->conn will be used via a thread
or callback, let's make sure memory associated with it isn't free'd
unexpectedly before we use it. The Unref will be done when the object
is Dispose'd.
Signed-off-by: John Ferlan <jferlan@redhat.com> Reviewed-by: Marc Hartmayer <mhartmay@linux.vnet.ibm.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
John Ferlan [Mon, 23 Apr 2018 15:53:16 +0000 (11:53 -0400)]
libxl: Add refcnt for args->vm during migration
When adding the @vm to the @args for usage during a thread or
callback, let's add the reference to it at the time of adding to
ensure nothing else deletes it. The corresponding Unref is then
added to the Dispose function.
Signed-off-by: John Ferlan <jferlan@redhat.com> Reviewed-by: Marc Hartmayer <mhartmay@linux.vnet.ibm.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
John Ferlan [Fri, 9 Mar 2018 12:37:14 +0000 (07:37 -0500)]
conf: Use virDomainObjListFindBy*Locked for virDomainObjListAdd
Use the FindBy{UUID|Name}Locked helpers which will return a locked
and ref counted object rather than the direct virHashLookup and
virObjectLock of the returned object. We'll need to temporarily
virObjectUnref when we assign a new domain @def, but that will
change shortly when virDomainObjListAddObjLocked returns the
correct reference counted object.
Use the virDomainObjEndAPI in the error path to Unref/Unlock for
the corresponding Unref/Unlock of either the FindBy* return or
the virDomainObjNew since both return a reffed/locked object.
Signed-off-by: John Ferlan <jferlan@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
John Ferlan [Fri, 9 Mar 2018 12:09:20 +0000 (07:09 -0500)]
conf: Split FindBy{UUID|Name} into locked helpers
Create helpers virDomainObjListFindByUUIDLocked and
virDomainObjListFindByNameLocked to avoid the need
to lock the domain object list leaving that task
for the caller.
Signed-off-by: John Ferlan <jferlan@redhat.com> Reviewed-by: Erik Skultety <eskultet@redhat.com>
<features><vmcoreinfo/> is a bare boolean XML property. We don't really
use this format anymore and instead prefer tristate <X state=on|off/>
since it's required for modeling on/off/default. If for example future
qemu started enabling vmcoreinfo by default we wouldn't have any way
for the user to turn this off.
Convert it to tristate. For writing XML this is semanticly the same,
<vmcoreinfo/> is processed as <vmcoreinfo state='on'/>.
For apps reading guest XML this is technically an API change,
as they might misinterpret <vmcoreinfo state='off'/>, however this
has only been present in libvirt since 3.10.0 and I don't think any
apps are dependent on this yet
Reviewed-by: John Ferlan <jferlan@redhat.com> Signed-off-by: Cole Robinson <crobinso@redhat.com>
Andrea Bolognani [Mon, 30 Apr 2018 15:30:12 +0000 (17:30 +0200)]
tests: Fix mode_t usage with va_arg()
Clang complains about it:
error: second argument to 'va_arg' is of promotable type
'mode_t' (aka 'unsigned short'); this va_arg has undefined
behavior because arguments will be promoted to 'int'
[-Werror,-Wvarargs]
mode = va_arg(ap, mode_t);
^~~~~~
Work around the issue by passing int to va_arg() and casting
its return value to mode_t afterwards.
Signed-off-by: Andrea Bolognani <abologna@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
all: Use virFileCanonicalizePath() instead of canonicalize_file_name()
The latter is impossible to mock on platforms that use the
gnulib implementation, such as FreeBSD, while the former
doesn't suffer from this limitation.
Signed-off-by: Andrea Bolognani <abologna@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
nwfilter: pass vm name in when instantiating filters
The vm name is not needed for any functional requirement, but it will be
useful when debugging problems to identify which VM is associated with a
filter, since UUID is not human friendly.
Reviewed-by: Jiri Denemark <jdenemar@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
nwfilter: introduce virNWFilterBinding to decouple from virDomainNet
The virDomainNet struct contains everything related to configuring a
guest network device. Out of all of this info, only 5 fields are
relevant to configuring network filters. It will be more convenient for
future changes to the nwfilter driver if the relevant fields are kept in
a dedicated struct. Thus the virNWFilterBinding struct is created to
track this information.
Reviewed-by: Jiri Denemark <jdenemar@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
nwfilter: remove obsolete code related to firewalld
There is a bunch of left over code in the nwfilter driver related to
monitoring firewalld over dbus, that is no longer used since the
conversion to use virFirewall APIs.
Reviewed-by: Jiri Denemark <jdenemar@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
nwfilter: change methods returning virNWFilterIPAddrLearnReq to use bool
Various methods return a virNWFilterIPAddrLearnReq struct, but the
callers are only interested in whether the return value is non-NULL.
It is thus preferrable to just return a bool.
Reviewed-by: Jiri Denemark <jdenemar@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
nwfilter: remove methods that are trivial wrappers for virHash APIs
This removes the virNWFilterHashTableFree, virNWFilterHashTablePut
and virNWFilterHashTableRemove methods, in favour of just calling
the virHash APIs directly.
The virNWFilterHashTablePut method was unreasonably complex because
the virHashUpdateEntry already knows how to create the entry if it
does not currently exist.
Reviewed-by: Jiri Denemark <jdenemar@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Remove pointless storage of var names in virNWFilterHashTable
Thus, this struct wrapper adds no real value over just using the
virHashTable directly, but brings the complexity of needing to derefence
the hashtable to call virHash* APIs, and adds extra memory allocation
step.
To minimize code churn this just turns virNWFilterHashTable into a
typedef aliases virHashTable.
Reviewed-by: Jiri Denemark <jdenemar@redhat.com> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Stefan Berger [Thu, 26 Apr 2018 17:42:18 +0000 (13:42 -0400)]
qemu: Add tpm-crb QEMU device to the command line
Alter qemuBuildTPMDevStr to format the tpm-crb on the command line
and use the enum range checking for valid model.
Add a test case for the formation of the tpm-crb QEMU device
command line. The qemuxml2argvtest changes cannot use the newer
DO_TEST_CAPS_LATEST since building of the command line involves
calling qemuBuildTPMBackendStr which attempts to open the
path to the device (e.g. /dev/tmp0).
Signed-off-by: Stefan Berger <stefanb@linux.vnet.ibm.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Julio Faracco [Tue, 1 May 2018 15:56:09 +0000 (12:56 -0300)]
test: avoid slash characters to the new domain name.
As QEMU driver, test driver does not accept slashes inside domain names.
This commit fixes this problem checking slashes inside the new name when
'domrename' is executed.
Signed-off-by: Julio Faracco <jcfaracco@gmail.com> Reviewed-by: John Ferlan <jferlan@redhat.com>
Julio Faracco [Tue, 1 May 2018 15:56:08 +0000 (12:56 -0300)]
qemu: avoid slash characters to the new domain name.
The 'domrename' command needs to check if the new domain name contains
the slash character. This character is not accepted by libvirt XML
definition because it is an invalid char (see Cole's commit b1fc6a7b7).
This commit enhace the 'domrename' command adding this check.
storagefile: conditional build of virStorageFileLoadBackendModule
The virStorageFileLoadBackendModule method is only used if either
fs or gluster storage is built in, which doesn't happen on mingw
leading to warning of an unused static function.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The libxlxml2domconfigtest causes a libxl-driver.log file to be created
which breaks make distchck if libxl is enabled. Delete the log file at
the end of the test.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
storage: create separate loadable modules for storage file drivers
The storage file drivers are currently loaded as a side effect of
loading the storage driver. This is a bogus dependancy because the
storage file code has no interaction with the storage drivers, and
even ultimately be running in a completely separate daemon.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
util: refactor storage file checks to allow error reporting
The virStorageFileSupportsSecurityDriver and
virStorageFileSupportsAccess currently just return a boolean
value. This is ok because they don't have any failure scenarios
but a subsequent patch is going to introduce potential failure
scenario. This changes their return type from a boolean to an
int with values -1, 0, 1.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The virStorageFileGetBackingStoreStr method has overloaded the NULL
return value to indicate both no backing available and a fatal
error dealing with it.
The caller is thus not able to correctly propagate the error
messages.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
storage: split fs storage file code from storage driver backend
The storage file code needs to be run in the hypervisor drivers, while
the storage backend code needs to be run in the storage driver. Split
the source code as a preparatory step for creating separate loadable
modules.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
storage: split gluster storage file code from storage driver backend
The storage file code needs to be run in the hypervisor drivers, while
the storage backend code needs to be run in the storage driver. Split
the source code as a preparatory step for creating separate loadable
modules.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
util: create new virmodule.{c,h} files for dlopen support code
The driver.{c,h} files are primarily targetted at loading hypervisor
drivers and some helper functions in that area. It also, however,
contains a generically useful function for loading extension modules
that is called by the storage driver. Split that functionality off
into a new virmodule.{c,h} file to isolate it.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
rpm: set wireshark plugin directory from pkg-config
The wireshark plugin directory moved again in Fedora 29, and will
move again every time wireshark do a new minor release. Call out
to pkg-config to find the right directory to use in the RPM file
list.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
rpm: libvirt-daemon-driver-libxl must obsolete libvirt-daemon-driver-xen
Snce the xen driver was deleted we need to ensure that the old
libvirt-daemon-driver-xen sub-RPM gets removed on upgrade. We
achieve this my making libvirt-daemon-driver-libxl obsolete it.
We don't add a Provides: too, because libvirt-daemon-driver-libxl
is not a functionally identical replacement, since we don't want
to satisfy deps for 3rd party apps that have a Requires on the
libvirt-daemon-driver-xen RPM.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Peter Krempa [Fri, 27 Apr 2018 14:20:15 +0000 (16:20 +0200)]
qemu: migration: Don't crash on access to 'current' job
When a VM is destroyed while being migrated (waiting in
qemuMigrationSrcWaitForCompletion) the private object cleanup code frees
the 'current' job info. Since the migration code attempts to setup
various aspects of the current job even on failure this results into a
crash.
Job data is cleared in qemuDomainObjPrivateDataClear since commit 888aa4b6b9db
Fix this by skipping all of the code which requires the qemu process to
be alive if the VM is not active any more.
Signed-off-by: Peter Krempa <pkrempa@redhat.com> Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
po: delete bogus translations from various languages
For unknown reasons about 21 languages had the same 212 msgid entries
copied into the msgstr field without having any translation applied.
This bogus non-translated data has now been purged from Zanata.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
projects / libvirt.git / log