On Mon, Mar 06, 2023 at 04:41:28AM -0800, Andrea Bolognani wrote:
On Mon, Mar 06, 2023 at 12:02:03PM +0100, Martin Kletzander wrote:
> On Mon, Mar 06, 2023 at 11:10:33AM +0100, Andrea Bolognani wrote:
> > Move more settings to common code and more closely match
> > real-world configurations.
> >
> > Andrea Bolognani (14):
> > tests: Poison more XDG variables
> > tests: Drop unnecessary configuration overrides
> > tests: Drop unnecessary free
> > tests: Increase scope for SASLdirs
> > tests: Limit use of SPICE TLS
>
> The changes to driver.config->spiceTLS look very arbitrary, but that's
> no biggie.
I feel that it's consistent with how we set all the other *TLS
variables to 1 right before the tests that are specifically about
testing TLS with vnc/chardev/whatever, and back to 0 immediately
afterwards.
A test called "controller-order" is definitely unrelated to SPICE
TLS. For "q35-virt-manager-basic", I guess you could argue either
way? It feels more unrelated than related to me.
No, of course, that's not the point, it is not trying to test spice with
tls. But because the spice-tls tests are not next to each other (and
would probably make a bigger mess in the order anyway) it *feels*
arbitrary when the spiceTLS is set and unset, it's fine the way it is.
Or were you referring to the various VIR_FREE(*TLSx509secretUUID)
calls that I'm adding? Because that was indeed intended to be a
separate commit, which I seem to have accidentally squashed in O:-)
--
Andrea Bolognani / Red Hat / Virtualization