Jiri Denemark [Wed, 1 Jun 2016 13:57:00 +0000 (15:57 +0200)]
tests: Add CPU detection tests
So far we only test CPUID -> CPU def conversion on artificial CPUID data
computed from another CPU def. This patch adds the infrastructure to
test this conversion on real data gathered from a host CPU and two
helper scripts for adding new test data:
- cpu-gather.sh runs cpuid tool and qemu-system-x86_64 to get CPUID data
from the host CPU; this is what users can be asked to run if they run
into an issue with host CPU detection in libvirt
- cpu-parse.sh takes the data generated by cpu-gather.sh and creates
data files for CPU detection tests
The CPUID data queried from QEMU will eventually switch to the format
used by query-host-cpu QMP command once QEMU implements it. Until then
we just spawn QEMU with -cpu host and query the guest CPU in QOM. They
should both provide the same CPUID results, but query-host-cpu does not
require any guest CPU to be created by QEMU.
Jiri Denemark [Tue, 7 Jun 2016 10:09:41 +0000 (12:09 +0200)]
cpu_x86: Refactor internal KVM features
The internal features are only used in explicit checks with
cpuHasFeature. Loading them into the CPU map is dangerous since the
features may accidentally be reported to users when decoding CPUID data.
Jiri Denemark [Tue, 7 Jun 2016 10:04:13 +0000 (12:04 +0200)]
cpu_ppc64: Avoid unnecessary pointer to virCPUppc64Data
virCPUData and struct ppc64_model structures contained a pointer to
virCPUppc64Data, which was not very nice since the real data were
accessible by yet another level of pointers from virCPUppc64Data.
Jiri Denemark [Tue, 7 Jun 2016 07:38:53 +0000 (09:38 +0200)]
cpu_x86: Avoid unnecessary pointers to virCPUx86Data
virCPUData, virCPUx86Feature, and virCPUx86Model all contained a pointer
to virCPUx86Data, which was not very nice since the real CPUID data were
accessible by yet another pointer from virCPUx86Data. Moreover, using
virCPUx86Data directly will make static definitions of internal CPU
features a bit easier.
Jiri Denemark [Mon, 6 Jun 2016 14:55:05 +0000 (16:55 +0200)]
tests: Create simple monitor in qemuMonitorTestNewFromFile
The current version uses the first JSON reply from the file as monitor
greeting. With the new parameter the caller can now request a simple
test monitor to be created, which uses an artificial greeting and uses
all JSON strings from the file as regular replies.
Jiri Denemark [Fri, 20 May 2016 07:48:21 +0000 (09:48 +0200)]
cpu_x86: Rename CPUID function to eax_in
CPUID instruction normally takes its parameter from EAX, but sometimes
ECX is used as an additional parameter. Let's rename 'function' to
'eax_in' in preparation for adding 'ecx_in'.
Jiri Denemark [Wed, 1 Jun 2016 08:55:36 +0000 (10:55 +0200)]
cpu_x86: Propagate vendor to guest's virCPUData
When computing CPU data for a given guest CPU we should set CPUID vendor
bits appropriately so that we don't lose the vendor when transforming
CPU data back to XML description.
virsh-network: Avoid possible NULL deref in cmdNetworkDHCPLeases
Problem is, localtime_r() returns a pointer to converted time or
NULL in case of an error. But checking the glibc sources, error
will occur iff a NULL has been passed as an either of arguments
the function takes. But GCC fails to see that:
../../tools/virsh-network.c: In function 'cmdNetworkDHCPLeases':
../../tools/virsh-network.c:1370:12: error: potential null pointer dereference [-Werror=null-dereference]
ts = *localtime_r(&expirytime_tmp, &ts);
~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
In Fedora >= 21, there is a new crypto priority framework
that sets TLS policies globally for all apps. To activate
this with GNUTLS we must request "@SYSTEM" instead of
the traditional "NORMAL" string. The '@' causes gnutls todo
a lookup in its config file for the 'SYSTEM' keyword entry.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Pass config file object through to driver open methods
The virConnectOpenInternal method opens the libvirt client
config file and uses it to resolve things like URI aliases.
There may be driver specific things that are useful to
store in the config file too, so rather than have them
re-parse the same file, pass the virConfPtr down to the
drivers.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Add a "tls_priority" config option to /etc/libvirt/libvirtd.conf
to allow the administrator to override the built-in default
setting. This only affects the server side configuration.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Currently libvirt calls gnutls_set_default_priority()
which on old systems resolves to "NORMAL" while new
systems it resolves to "@SYSTEM". Either way, this
is a global default that is identical across all apps.
We want to allow distros to flexibility to define a
custom default string for libvirt priority, so add
a --tls-priority=STRING flag to configure to enable
this to be set.
It is expected that distros would use this when creating
RPM/Deb/etc packages, according to their preferred crypto
handling policies.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Currently we set the gnutls log function when creating a
TLS context, however, the setting is in fact global, not
per context. So we should be setting it when we first call
gnutls_global_init() instead.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
tls: remove support for gnutls 1.x.x, require 2.2.0
We need to use the gnutls_priority_set_direct method which
was not introduced until 2.1.7, so bump version to 2.2.0
which is the first stable release with it included. This
release dates from Dec 2007 so it is reasonable to ditch
support for the 1.x.x series for gnutls releases entirely.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Pavel Hrdina [Mon, 30 May 2016 14:35:17 +0000 (16:35 +0200)]
virsh: domdisplay: if listen is 0.0.0.0 or [::] print address from URI
Currently if a guest has listen address 0.0.0.0 or [::] and you run
"virsh domdisplay $domain" you always get "spice://localhost:$port".
We want to print better address if someone is connected from a different
computer using "virsh -c qemu+ssh://some.host/system". This patch fixes the
behavior of virsh to print in this case "spice://some.host:$port".
We use libxml2 APIs in the test (e.g. xmlFreeDoc) but not link
with -lxml2 which can cause problems:
/usr/bin/ld: virschematest.o: undefined reference to symbol 'xmlFreeDoc@@LIBXML2_2.4.30'
//usr/lib/x86_64-linux-gnu/libxml2.so.2: error adding symbols: DSO missing from command line
collect2: error: ld returned 1 exit status
Makefile:4702: recipe for target 'virschematest' failed
Reported-by: Katerina Koukiou <k.koukiou@googlemail.com> Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
virschematest: Access the right directory containing XMLs
So the story goes like this. The testSchemaDirs() function is
called with: a) the schema file, b) list of the directories that
contains XMLs documents that should be checked against the schema
file from a). However, the directories in the list are really
just their names and it's up to testSchemaDirs to construct the
absolute path and call testSchemaDir() which then does the actual
validation. The absolute path is constructed, but never actually
used (maybe due to a typo). Thus a VPATH build is broken.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Peter Krempa [Thu, 2 Jun 2016 13:32:00 +0000 (15:32 +0200)]
qemu: process: Call disk startup policy check after cloning domain def
In commit 1e38ef72 the disk startup policy check was moved prior to the
call to virDomainObjSetDefTransient which dropped the disk from the
config rather than the def to be started which is a bug.
Additionally we'd not report the disk change event for this since the
disk aliases were not set at that point.
Finally 'volume' based disks would not work with startup policy too.
Fix it by moving it back after the definition is copied, aliases are
assigned and disk sources are translated.
Peter Krempa [Fri, 3 Jun 2016 13:18:48 +0000 (15:18 +0200)]
qemu: driver: Unset log file watcher after restoring a VM save file
qemuProcessStart does not unset the infrastructure that retrieves errors
from the qemu log file in case of migration. As this wasn't handled
properly in qemuDomainSaveImageStartVM we kept the logging context/fd
open for the lifetime of the VM rather than closing it after it's not
needed.
Philipp Hahn [Thu, 19 May 2016 06:24:55 +0000 (08:24 +0200)]
xen: Also add sub-type for driver=tap2 in xen-xm
tap2 only handles 'aio', but not 'raw', which must be explicitly given:
| $ virsh domxml-to-native yyy.xml > yyy.xm
| $ xm new yyy.xm
| Error: tap:/srv/xen/xxx.img not a valid disk type
| $ sed -i -e 's/tap2:/&aio:/' yyy.xm
| $ xm new yyy.xm
Fix reading and writing "xen-xm" format for "tap2" by handling it the
same as "tap".
Peter Krempa [Tue, 7 Jun 2016 14:31:15 +0000 (16:31 +0200)]
qemu: process: Append the "shutting down" message using the new APIs
Use qemuDomainLogAppendMessage rather than attempting to open a new
logging context with file descriptors. The new approach allows to log
the message even if qemu is still running at that point which appens
during migration finish phase where qemuProcessStop is killing qemu.
Peter Krempa [Tue, 7 Jun 2016 14:19:03 +0000 (16:19 +0200)]
qemu: domain: Implement helper for one-shot log entries to the VM log file
Along with the virtlogd addition of the log file appending API implement
a helper for logging one-shot entries to the log file including the
fallback approach of using direct file access.
This will be used for noting the shutdown of the qemu proces and
possibly other actions such as VM migration and other critical VM
lifecycle events.
Peter Krempa [Tue, 7 Jun 2016 14:09:09 +0000 (16:09 +0200)]
log: handler: Add new API to append to logging files
For logging one-shot entries to the VM log file it's quite a waste to
hold open the file descriptor for logging that is provided by the
current API.
This new API will be ideal for logging one-shot entries to the file
e.g. at the point when we shut the VM down rather than having to add the
whole file-descriptor infrastructure.
Additionally this will allow to add the messages even after restart of
libvirtd since virtlogd doesn't allow to obtain a regular context with
filedescriptors while the VM is still active.
Peter Krempa [Fri, 27 May 2016 11:45:05 +0000 (13:45 +0200)]
qemu: process: Convert multiple boolean args to a single flag
Validation of qemu process startup requires to know whether the process
is used for a fresh VM or whether it's reloaded from a
snapshot/migration. Pass this information in via a flag rather than
calculating it from a bunch of bools.
Peter Krempa [Fri, 27 May 2016 08:56:56 +0000 (10:56 +0200)]
conf: Add device def validation callback
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.
Peter Krempa [Thu, 26 May 2016 13:58:53 +0000 (15:58 +0200)]
conf: Add infrastructure for adding configuration validation
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.
projects / libvirt.git / log