Yeah, that's right. A mount point doesn't have to be a directory.
It can be a file too. However, the code that tries to preserve
mount points under /dev for new namespace for qemu does not count
with that option.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
VIR_CONNECT_LIST_STORAGE_POOLS_VSTORAGE and
VIR_CONNECT_LIST_STORAGE_POOLS_ZFS were added to libvirt but the listing
API was not properly updated to use them.
bhyve supports 'gop' video device that allows clients to connect
to VMs using VNC clients. This commit adds support for that to
the bhyve driver:
- Introducr 'gop' video device type
- Add capabilities probing for the 'fbuf' device that's
responsible for graphics
- Update command builder routines to let users configure
domain's VNC via gop graphics.
Signed-off-by: Roman Bogorodskiy <bogorodskiy@gmail.com>
Extend domain capabilities XML with the information about
available UEFI firmware files. It searches in the location
that the sysutils/bhyve-firmware FreeBSD port installs
files to.
Signed-off-by: Roman Bogorodskiy <bogorodskiy@gmail.com>
Fabian Freyer [Tue, 24 May 2016 13:30:56 +0000 (15:30 +0200)]
bhyve: virBhyveProbeCaps: BHYVE_CAP_LPC_BOOTROM
Implement the BHACE_CAP_LPC_BOOTROM capability by checking the stderr
output of 'bhyve -l bootrom'. If the bootrom option is unsupported, this
will contain the following output:
bhyve: invalid lpc device configuration 'bootrom'
On newer bhyve versions that do support specifying a bootrom image, the
standard help will be printed.
Add bhyve support to virt-host-validate(1). It checks for the
essential kernel modules to be available so that user can actually
start VMs, have networking and console access.
It uses the kldnext(2)/kldstat(2) routines to retrieve modules list.
As bhyve is only available on FreeBSD and these routines were available
long before bhyve appeared, not adding any specific configure checks
for that.
Also, update tools/Makefile.am to add
virt-host-validate-$driver.[hc] to the build only if the
appropriate driver is enabled.
John Ferlan [Sun, 29 Jan 2017 16:12:48 +0000 (11:12 -0500)]
tests: Add createVHBAByStoragePool-by-parent to fchosttest
Add a new test to fchosttest in order to test creation of our vHBA
via the Storage Pool logic. Unlike the real code, we cannot yet use
the virVHBA* API's because they (currently) traverse the file system
in order to get the parent vport capable scsi_host. Besides there's
no "real" NPIV device here - so we have to take some liberties, at
least for now.
Instead, we'll follow the node device tests partially in order to
create and destroy the vHBA with the test node devices.
If a qemu process has died, we get EOF on its monitor. At this
point, since qemu process was the only one running in the
namespace kernel has already cleaned the namespace up. Any
attempt of ours to enter it has to fail.
This really happened in the bug linked above. We've tried to
attach a disk to qemu and while we were in the monitor talking to
qemu it just died. Therefore our code tried to do some roll back
(e.g. deny the device in cgroups again, restore labels, etc.).
However, during the roll back (esp. when restoring labels) we
still thought that domain has a namespace. So we used secdriver's
transactions. This failed as there is no namespace to enter.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Michal Privoznik [Fri, 10 Mar 2017 08:55:43 +0000 (09:55 +0100)]
qemuxml2argvtest: Don't overwrite driver stateDir
This is a very historic artefact. Back in the old days of 830ba76c3e when we had macros to add arguments onto qemu command
line (!) we thought it was a good idea to let qemu write out the
PID file. So we passed -pidfile $stateDir/$domName onto the
command line. Thus, in order for tests to work we needed stable
stateDir in the qemu driver. Unfortunately, after 16efa11aa696
where stateDir is mkdtemp()-d, this approach lead to a leak of
temp dir.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Peter Krempa [Fri, 3 Mar 2017 15:04:57 +0000 (16:04 +0100)]
qemu: hotplug: Reset device removal waiting code after vCPU unplug
If the delivery of the DEVICE_DELETED event for the vCPU being deleted
would time out, the code would not call 'qemuDomainResetDeviceRemoval'.
Since the waiting thread did not unregister itself prior to stopping the
waiting the monitor code would try to wake it up instead of dispatching
it to the event worker. As a result the unplug process would not be
completed and the definition would not be updated.
So our code is one big mess and we modify domain definition while
building qemu_command line and our hotplug code share only part
of the parsing and command line building code. Let's revert
that change because to fix it properly would require refactor and
move a lot of things.
configure: disable scsi stroage driver on non-Linux
Even though scsi storage driver builds fine on non-Linux, it
will not work properly because it relies on Linux procfs, so
disable that in configure if we're not building for Linux.
Pavel Hrdina [Mon, 27 Feb 2017 16:16:17 +0000 (17:16 +0100)]
conf: properly skip graphics listen element in migratable XML
We should skip <listen type='socket'/> only if the 'socket' path
is specified because if there is no 'socket' path we need to
keep that element in migratable XML.
Pavel Hrdina [Mon, 27 Feb 2017 16:00:15 +0000 (17:00 +0100)]
conf: store "autoGenerated" for graphics listen in status XML
When libvirtd is started we call qemuDomainRecheckInternalPaths
to detect whether a domain has VNC socket path generated by libvirt
based on option from qemu.conf. However if we are parsing status XML
for running domain the existing socket path can be generated also if
the config XML uses the new <listen type='socket'/> element without
specifying any socket.
The current code doesn't make difference how the socket was generated
and always marks it as "fromConfig". We need to store the
"autoGenerated" value in the status XML in order to preserve that
information.
The difference between "fromConfig" and "autoGenerated" is important
for migration, because if the socket is based on "fromConfig" we don't
print it into the migratable XML and we assume that user has properly
configured qemu.conf on both hosts. However if the socket is based
on "autoGenerated" it means that a new feature was used and therefore
we need to leave the socket in migratable XML to make sure that if
this feature is not supported on destination the migration will fail.
John Ferlan [Fri, 17 Feb 2017 15:06:14 +0000 (10:06 -0500)]
qemu: Introduce qemuDomainGetTLSObjects
Split apart and rename qemuDomainGetChardevTLSObjects in order to make a
more generic API that can create the TLS JSON prop objects (secret and
tls-creds-x509) to be used to create the objects
John Ferlan [Wed, 22 Feb 2017 17:54:10 +0000 (12:54 -0500)]
qemu: Refactor qemuDomainGetChardevTLSObjects to converge code
Create a qemuDomainAddChardevTLSObjects which will encapsulate the
qemuDomainGetChardevTLSObjects and qemuDomainAddTLSObjects so that
the callers don't need to worry about the props.
Move the dev->type and haveTLS checks in to the Add function to avoid
an unnecessary call to qemuDomainAddTLSObjects
John Ferlan [Thu, 16 Feb 2017 15:33:35 +0000 (10:33 -0500)]
qemu: Refactor hotplug to introduce qemuDomain{Add|Del}TLSObjects
Refactor the TLS object adding code to make two separate API's that will
handle the add/remove of the "secret" and "tls-creds-x509" objects including
the Enter/Exit monitor commands.
John Ferlan [Wed, 22 Feb 2017 17:39:17 +0000 (12:39 -0500)]
qemu: Move exit monitor calls in failure paths
Since qemuDomainObjExitMonitor can also generate error messages,
let's move it inside any error message saving code on error paths
for various hotplug add activities.
Some of our tests (e.g. qemuhotplugtest) call
virDomainSaveConfig(). Now the problem is, qemuTestDriverInit()
creates a fake qemu driver and fills it with some fake
configuration. At least so we hoped. The truth is, it calls
regular virQEMUDriverConfigNew() and then fix couple of paths.
Literally. Therefore our tests see regular stateDir and configDir
for the user that is running the tests. Directories, where live
domain XMLs are stored. Let's just hope our test suite hasn't
mangled any of them.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
When we are restoring a domain from a saved image, or just
updating its XML in the saved image - we have to make sure that
the ABI guests sees will not change. We have a function for that
which reports errors. But for some reason if this function fails,
we call it again with slightly different argument. Therefore it
might happen that we overwrite the original error and leave user
with less helpful one.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Thu, 2 Mar 2017 17:58:41 +0000 (12:58 -0500)]
conf: Alter coding style of nwfilter function prototypes
In an effort to be consistent with the source module, alter the function
prototypes to follow the similar style of source with the "type" on one
line followed by the function name and arguments on subsequent lines with
with argument getting it's own line.
John Ferlan [Thu, 2 Mar 2017 17:13:02 +0000 (12:13 -0500)]
conf: Adjust coding style for nwfilter conf sources
Alter the format of the code to follow more recent style guidelines of
two empty lines between functions, function decls with "[static] type"
on one line followed by function name with arguments to functions each
on one line.
John Ferlan [Tue, 7 Mar 2017 12:01:21 +0000 (07:01 -0500)]
util: Avoid possible NULL dereference in virSysinfoParsePPCProcessor
Found by Coverity. Because there's an "if ((cur = strstr(base, "revision"))
!= NULL) {" followed by a "base = cur" coverity notes that 'base' could
then be NULL causing the return to the top of the "while ((tmp_base =
strstr(base, "processor")) != NULL) {" to have strstr deref a NULL 'base'
pointer because the setting of base at the bottom of the loop is unconditional.
Alter the code to set "base = cur" after processing each key. That will
"ensure" that base doesn't get set to NULL if both "cpu" and "revision"
do no follow a "processor".
While a /proc/cpuinfo file that has a "processor" key but with neither
a "cpu" nor a "revision" doesn't seem feasible, the code is written as if
it could happen, so we have to account for it.
John Ferlan [Tue, 7 Mar 2017 11:33:38 +0000 (06:33 -0500)]
util: Resource some resource leaks in virsysinfo code
Calls to virFileReadAll after a VIR_ALLOC that return NULL all show
a memory leak since 'ret' isn't virSysinfoDefFree'd and normal path
"return ret" doesn't free outbuf.
Cole Robinson [Mon, 6 Mar 2017 21:50:53 +0000 (16:50 -0500)]
storage: Don't pass 'iso' format to qemu-img
$ virsh vol-clone /tmp/test.iso new.iso
error: Failed to clone vol from test.iso
error: internal error: Child process (/bin/qemu-img convert -f iso -O iso /tmp/test.iso /tmp/new.iso) unexpected exit status 1: qemu-img: Could not open '/tmp/test.iso': Unknown driver 'iso'
Map iso->raw before sending the format value to qemu-img
Peter Krempa [Mon, 6 Mar 2017 14:14:45 +0000 (15:14 +0100)]
util: sysinfo: Reduce amount of conditionally compiled code
Whole implementations along with helper totalling screens of code were
conditionally compiled. That made the code totally unreadable and
untestable. Rename functions to have the architecture in the name so
that all can be compiled at the same time and introduce header to allow
testing them all.
Since these are by far a minority among all struct definitions, this
patch changes all those definitions to reflect the style prefered by
the proposal so that there is 100% consistency.
Ján Tomko [Fri, 24 Jun 2016 15:41:09 +0000 (17:41 +0200)]
Cache the presence of machine1 service
After the system has been booted, it should not change.
Cache the return value of virSystemdHasMachined.
Allow starting and terminating machines with just one
DBus call, instead of three, reducing the chance of
the call timing out.
Also introduce a small function for resetting the cache
to be used in tests.
John Ferlan [Sat, 4 Mar 2017 14:34:53 +0000 (09:34 -0500)]
test: Make a common testNetworkObjFindByName
Rather than have multiple places using the same pattern to find
a network by name using virNetworkObjFindByName, create a common
helper which will provide a consistent error message as well.
Make some common helpers which will use the pattern and make a single
reference using a single common error message.
Altered for Interfaces, Storage Pools, Storage Volumes, and Node Devices.
For each the common error message can now also indicate which 'name' was
not found. For Storage Volumes, the "new" error will be more specific
rather than just invalid argument.
John Ferlan [Thu, 2 Mar 2017 17:01:29 +0000 (12:01 -0500)]
conf: Alter coding style of interface function prototypes
In an effort to be consistent with the source module, alter the function
prototypes to follow the similar style of source with the "type" on one
line followed by the function name and arguments on subsequent lines with
with argument getting it's own line.
John Ferlan [Thu, 2 Mar 2017 16:44:11 +0000 (11:44 -0500)]
conf: Adjust coding style for interface conf sources
Alter the format of the code to follow more recent style guidelines of
two empty lines between functions, function decls with "[static] type"
on one line followed by function name with arguments to functions each
on one line.
Michal Privoznik [Mon, 13 Feb 2017 13:36:53 +0000 (14:36 +0100)]
qemu: Enforce qemuSecurity wrappers
Now that we have some qemuSecurity wrappers over
virSecurityManager APIs, lets make sure everybody sticks with
them. We have them for a reason and calling virSecurityManager
API directly instead of wrapper may lead into accidentally
labelling a file on the host instead of namespace.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Sat, 4 Mar 2017 15:42:33 +0000 (10:42 -0500)]
node: Replace variable named 'system' with 'syscap'
Changes in commit id 'dec6d9df' caused a compilation failure on a RHEL6
CI build environment. So just replace 'system' with 'syscap' as a name.
cc1: warnings being treated as errors
../../src/conf/node_device_conf.c: In function 'virNodeDevCapSystemParseXML':
../../src/conf/node_device_conf.c:1415: error: declaration of 'system' shadows a global declaration [-Wshadow]
John Ferlan [Thu, 2 Mar 2017 13:50:27 +0000 (08:50 -0500)]
conf: Alter coding style of nodedev function prototypes
In an effort to be consistent with the source module, alter the function
prototypes to follow the similar style of source with the "type" on one
line followed by the function name and arguments on subsequent lines with
with argument getting it's own line.
John Ferlan [Thu, 2 Mar 2017 13:39:11 +0000 (08:39 -0500)]
conf: Adjust coding style for nodedev conf sources
Alter the format of the code to follow more recent style guidelines of
two empty lines between functions, function decls with "[static] type"
on one line followed by function name with arguments to functions each
on one line.
lxc: associate armv7l as 32-bit variant of aarch64
AArch64 kernels are technically capable of running armv7l binaries.
Though some vendors disable this feature during kernel build, we
need to allow it in LXC.
Signed-off-by: Matwey V. Kornilov <matwey.kornilov@gmail.com>
Jiri Denemark [Wed, 15 Feb 2017 14:01:40 +0000 (15:01 +0100)]
cpu_x86: Disable TSX on broken models
All Intel Haswell processors (except Xeon E7 v3 with stepping >= 4) have
TSX disabled by microcode update. As not all CPUs are guaranteed to be
patched with microcode updates we need to explicitly disable TSX on
affected CPUs to avoid its accidental usage.
Jiri Denemark [Mon, 13 Feb 2017 09:33:52 +0000 (10:33 +0100)]
cputest: Test virQEMUCapsInitCPUModel
The original test didn't use family/model numbers to make better
decisions about the CPU model and thus mis-detected the model in the two
cases which are modified in this commit. The detected CPU models now
match those obtained from raw CPUID data.
Jiri Denemark [Wed, 1 Feb 2017 16:48:41 +0000 (17:48 +0100)]
cputest: Switch host CPU data scripts to model expansion
Instantiating "host" CPU and querying it using qom-get has been the only
way of probing host CPU via QEMU until 2.9.0 implemented
query-cpu-model-expansion for x86_64. Even though libvirt never really
used the old way its result can be easily converted into the one
produced by query-cpu-model-expansion. Thus we can reuse the original
test data and possible get new data from hosts where QEMU does not
support the new QMP command.
Jiri Denemark [Thu, 23 Feb 2017 12:53:51 +0000 (13:53 +0100)]
qemu: Use full CPU model expansion on x86
The static CPU model expansion is designed to return only canonical
names of all CPU properties. To maintain backwards compatibility libvirt
is stuck with different spelling of some of the features, but we need to
use the full expansion to get the additional spellings. In addition to
returning all spelling variants for all properties the full expansion
will contain properties which are not guaranteed to be migration
compatible. Thus, we need to combine both expansions. First we need to
call the static expansion to limit the result to migratable properties.
Then we can use the result of the static expansion as an input to the
full expansion to get both canonical names and their aliases.
Jiri Denemark [Wed, 11 Jan 2017 13:36:34 +0000 (14:36 +0100)]
qemu: Get host CPU model from QEMU on x86_64
Until now host-model CPU mode tried to enable all CPU features supported
by the host CPU even if QEMU/KVM did not support them. This caused a
number of issues and made host-model quite unreliable. Asking QEMU for
the CPU it can provide and the current host makes host-model much more
robust.
Jiri Denemark [Wed, 22 Feb 2017 15:51:50 +0000 (16:51 +0100)]
qemu: Probe "max" CPU model in TCG
Querying "host" CPU model expansion only makes sense for KVM. QEMU 2.9.0
introduces a new "max" CPU model which can be used to ask QEMU what the
best CPU it can provide to a TCG domain is.
projects / libvirt.git / log