Re: [PATCH v3 13/24] tests/functional: Convert the s390x avocado tests into standalone tests

[Thomas Huth]

On 13/08/2024 14.04, Thomas Huth wrote:
> On 02/08/2024 15.25, Philippe Mathieu-Daudé wrote:
> > On 1/8/24 19:38, Daniel P. Berrangé wrote:
> > > On Thu, Aug 01, 2024 at 07:11:01PM +0200, Philippe Mathieu-Daudé wrote:
> > > > On 30/7/24 19:03, Daniel P. Berrangé wrote:
> > > > > From: Thomas Huth <thuth@redhat.com>
> > > > >
> > > > > These tests use archive.lzma_uncompress() from the Avocado utils,
> > > > > so provide a small helper function for this, based on the
> > > > > standard lzma module from Python instead.
> > > > >
> > > > > And while we're at it, replace the MD5 hashes in the topology test
> > > > > with proper SHA256 hashes, since MD5 should not be used anymore
> > > > > nowadays.
> > > > >
> > > > > Signed-off-by: Thomas Huth <thuth@redhat.com>
> > > > > ---
> > > > >    MAINTAINERS                                   |  4 +-
> > > > >    tests/functional/meson.build                  |  6 ++
> > > > >    tests/functional/qemu_test/utils.py           | 14 +++
> > > > >    .../test_s390x_ccw_virtio.py}                 | 79 ++++++++---------
> > > > >    .../test_s390x_topology.py}                   | 86 ++++++++-----------
> > > > >    5 files changed, 94 insertions(+), 95 deletions(-)
> > > > >    rename tests/{avocado/machine_s390_ccw_virtio.py => 
> > > > > functional/test_s390x_ccw_virtio.py} (85%)
> > > > >    mode change 100644 => 100755
> > > > >    rename tests/{avocado/s390_topology.py => 
> > > > > functional/test_s390x_topology.py} (88%)
> > > > >    mode change 100644 => 100755
> > > >
> > > > Conversion LGTM, so:
> > > >
> > > > Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> > > >
> > > > Now while testing I'm getting errors, but maybe pre-existing...
> > >
> > > What host arch / OS are you running tests from ?
> > >
> > > I got through the full set of testing on Fedora x86_64, so
> > > wondering if something differs.
> >
> > Linux s390x but I'm not in the KVM group, I need to find
> > another place to run these (or ask Alex :P).
>
> Sounds like a pre-existing problem ... I guess the test should check whether 
> it can access /dev/kvm before trying to use it...?

Ah, no, I missed your comment in your earlier patch that kvm_available() 
already checks this ... so the problem must be something else?

 Thomas