Kautuk Consul <kcon...@linux.vnet.ibm.com> writes:

> Commit c0c8687ef0fd990db8db1655a8a6c5a5e35dd4bb disabled the
> boot_linux.py test-case due to which the code coverage for
> powerpc decreased by 2%. This patchset aims to make up this
> 2% code coverage by modifying the tuxrun_baselines.py test-case.

BTW did you see we have a basic tool to compare runs now in
scripts/coverage/compare_gcov_json.py if you want to see exactly what
the differences between a boot_linux and tuxrun_baseline test is.

Currently its only file granularity but it could be fixed up to do a
more detailed line by line breakdown.

>
> Changes since v1 and v2:
> - Modified the way and code that creates the temporary qcow2
>   image that the tuxrun_baselines.py is now using.
> - Factored out code in avocado_qemu/__init__.py to create
>   the get_qemu_img() function that will now get called from
>   tuxrun_baselines.py.
>
> Kautuk Consul (2):
>   avocado_qemu/__init__.py: factor out the qemu-img finding
>   tests/avocado/tuxrun_baselines.py: improve code coverage for ppc64
>
>  tests/avocado/avocado_qemu/__init__.py |  27 +++---
>  tests/avocado/tuxrun_baselines.py      | 120 ++++++++++++++++++++++++-
>  2 files changed, 132 insertions(+), 15 deletions(-)


-- 
Alex Bennée
Virtualisation Tech Lead @ Linaro

Reply via email to