On 01/28/2014 04:46 PM, Kashyap Chamarthy wrote:
On 01/28/2014 04:53 AM, Jiri Denemark wrote:
> Jiri Denemark (4):
> tests: Better support for VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES
> cpu: Fix VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES
> cpu: Try to use source CPU model in virConnectBaselineCPU
> tests: Add more tests for virConnectBaselineCPU
>
> src/cpu/cpu_x86.c | 101 +++++++++++++++++---------
> tests/cputest.c | 27 ++++++-
> tests/cputestdata/x86-baseline-3-expanded.xml | 35 +++++++++
> tests/cputestdata/x86-baseline-3-result.xml | 34 +--------
> tests/cputestdata/x86-baseline-4-expanded.xml | 46 ++++++++++++
> tests/cputestdata/x86-baseline-4-result.xml | 14 ++++
> tests/cputestdata/x86-baseline-4.xml | 18 +++++
> tests/cputestdata/x86-baseline-5-expanded.xml | 47 ++++++++++++
> tests/cputestdata/x86-baseline-5-result.xml | 10 +++
> tests/cputestdata/x86-baseline-5.xml | 35 +++++++++
> 10 files changed, 296 insertions(+), 71 deletions(-)
> create mode 100644 tests/cputestdata/x86-baseline-3-expanded.xml
> create mode 100644 tests/cputestdata/x86-baseline-4-expanded.xml
> create mode 100644 tests/cputestdata/x86-baseline-4-result.xml
> create mode 100644 tests/cputestdata/x86-baseline-4.xml
> create mode 100644 tests/cputestdata/x86-baseline-5-expanded.xml
> create mode 100644 tests/cputestdata/x86-baseline-5-result.xml
> create mode 100644 tests/cputestdata/x86-baseline-5.xml
Thanks Jiri,
This is not a review, but a quick initial testing with these patches
applied, I ran the reproducer script[1] w/ a scratch build[2]:
$ python libvirt-test.py
VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES = 1
Seem to work fine. Still have to test w/ OpenStack set-up.
FWIW, also tested with OpenStack set-up, Nova Compute service comes up
just fine w/ these patches. Updated the bug:
https://bugzilla.redhat.com/show_bug.cgi?id=1049391#c15
--
/kashyap