Add qdisks.errors into error infos and add
steps in error.context.
Signed-off-by: NShuping Cui <scui@redhat.com>

Maybe the variable 'service_status_filter is intended to
be the pattern param for re.search(),
But re.search() adopts character string directly.
So It becomes meaningless.

Changing name of spice-gtk tar package to version 0.30

Reviewed By: Swapna Krishnan <skrishna@redhat.com>

200s is not enough to update initrd sometime, need
enlarge it.
Signed-off-by: NCong Li <coli@redhat.com>

module autotest.client.shaed is typo
change back to shared
Signed-off-by: NMike Cao <bcao@redhat.com>

RHEL5 and RHEL6 host do not support option "compat"
for qemu_disk_img.convert.
Signed-off-by: NCong Li <coli@redhat.com>

The image copy test hardly has QEMU specific tech in
it, so the correct thing to do is to have it in the
generic subdirectory, so other test providers can also
use it.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

On RHEL6, using aexpect directly causes 
ShellCmdError & ShellCmdTimeout exceptions 
to not work anymore. 

Reviewed By: Swapna Krishnan <skrishna@redhat.com>

Dumps collected in a host with an Intel Broadwell CPU.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Dumps collected in a host with an Intel Broadwell CPU.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Dumps collected in a host with an Intel Broadwell CPU.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Changing the spice-gtk tar package to be the latest upstream 
release from 0.29. 

Reviewed By: Swapna Krishnan <skrishna@redhat.com>

In RHEL6, moving to aexpect from virttest.aexpect causes 
ShellCmdError to not work anymore. Hence removing 
ShellCmdError exception and keeping a general except 
statement instead. 

Reviewed By: Swapna Krishnan <skrishna@redhat.com>

We don't know if disable_kvm=yes will really work, so don't try it if we
don't know the host system.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Those dump files were collected in an Intel host, but QEMU 1.4.0 had
"enforce" support broken. The "athlon" CPU model was not supposed to run
successfully, as the host didn't have mmxext, 3dnow, and 3dnowext
available.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Dumps generated on a Haswell host.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Dumps collected on a Broadwell machine using
qemu-kvm-rhev-2.1.2-23.el7_1.3.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

CPUID dumps collected on a Haswell host CPU.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@raisama.net>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Nested VMX is not enabled by default in the kernel and some QEMU
versions failed to report VMX as missing when using enforce mode, so
it's better to explicitly disable it when testing.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

To allow testing of CPU models that can't run in the host due to missing
hardware or kernel support, add QOM-based testing that will just check
the "feature-words" and "filtered-features" QOM properties on CPU
objects, to check if the set of CPU features matches the CPUID dump.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

This will make it easier to handle missing registers when implementing
QMP-based CPUID checks.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

We will support a test mode that uses QOM, later.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

On some systems, sgabios can be enabled and BIOS output may go to
the serial port. Be more permissive, and ignore anything outside
the ==START TEST== and ==END TEST== delimiters when parsing CPUID
dump output.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

avocado-vt will throw VMCreateError if QEMU refuses to start, so handle
both VMStartError and VMCreateError, to catch errors when a CPU model
can't run.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Makes it easier to change and to solve git conflicts.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

Better for debugging.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>

	qemu/tests/cgroup.py:Fix wrong argument passed to cgroup.set_property() in cpuset_cpus_switching();

This fixes a problem found by pylint inspection: The wrong
argument was being passed in some calls to super().
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Clean up imports across the board in tp-qemu. The criteria
followed was:

1) Remove unnecessary imports
2) Turn multi imports from a single entry point to one
   import per line. This increases both patch and
   source code readability.
3) Separate blocks of imports with a line:
   * Python standard library imports
   * Autotest imports
   * virttest imports
   * Internal tp-qemu imports
4) Remove try/except blocks refering autotest libs and
   stick with the virttest.staging copy

This paves the way for the future removal of autotest
references, replacing those references with avocado ones.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>