Re: non-x86 runners in the Gitlab-CI (was: Re: [PATCH 12/12] configure: bump min required CLang to 7.0.0 / XCode 10.2)

[Daniel P . Berrangé]

On Wed, May 12, 2021 at 03:55:59PM +0200, Thomas Huth wrote:
> On 12/05/2021 15.47, Willian Rampazzo wrote:
> > Hi Thomas,
> > 
> > On Wed, May 12, 2021 at 8:54 AM Thomas Huth <thuth@redhat.com> wrote:
> > > 
> > > On 12/05/2021 13.44, Philippe Mathieu-Daudé wrote:
> > > > On 5/11/21 3:26 PM, Daniel P. Berrangé wrote:
> > > > > Several distros have been dropped since the last time we bumped the
> > > > > minimum required CLang version.
> > > > > 
> > > > > Per repology, currently shipping versions are:
> > > > > 
> > > > >                RHEL-8: 10.0.1
> > > > >        Debian Stretch: 7.0.1
> > > > >         Debian Buster: 7.0.1
> > > > >    openSUSE Leap 15.2: 9.0.1
> > > > >      Ubuntu LTS 18.04: 10.0.0
> > > > >      Ubuntu LTS 20.04: 11.0.0
> > > > >            FreeBSD 12: 8.0.1
> > > > >             Fedora 33: 11.0.0
> > > > >             Fedora 34: 11.1.0
> > > > > 
> > > > > With this list Debian Stretch is the constraint at 7.0.1
> > > > > 
> > > > > An LLVM version of 7.0.1 corresponds to macOS XCode version of 10.2
> > > > > which dates from March 2019.
> > > > 
> > > > But we still rely on Travis-CI (Ubuntu Bionic 18.04 LTS)
> > > > for non-x86 targets until we have figured out who is willing
> > > > to share/maintain such non-x86 native runners on Gitlab.
> > > 
> > >    Hi Cleber,
> > > 
> > > by the way, what's the status of your patch series to get the dedicated CI
> > > machines (s390x, aarch64, ...) running in our Gitlab-CI? AFAIK the last
> > > iteration of your patches has been weeks ago, so I wonder whether you 
> > > could
> > > finally send a new version with the requested fixes included? ... this 
> > > topic
> > > slowly gets more and more urgent now that our Travis-CI is in process of
> > > dying...
> > 
> > I don't know if you saw this:
> > https://docs.travis-ci.com/user/billing-overview/#partner-queue-solution.
> > 
> > tl;dr, Travis now has support from partners to run non-x86 arch. It is
> > always good to have a plan B, like qemu own CI runners, but, at least,
> > with these non-x86 arch available on Travis, we will have some time to
> > breathe.
> 
> Uh, that's what we're already using in our travis.yml ... but I guess you've
> rather missed:
> 
>  https://blog.travis-ci.com/2021-05-07-orgshutdown
> 
> and on travis-ci.com, the CI minutes are not for free anymore. At least not
> for the QEMU project. Or do you know of a sponsor who is going to pay the CI
> minutes for us there?

The link above explicitly says the non-x86 jobs cost 0 credits:


  "The build job under Partner Queue Solution costs 0 credits per 
   started minute. At the moment of introducing Partner Queue Solution
   active accounts on the Usage based Plans, including the Free Plan,
   with a balance of zero or fewer credits, balance is updated to 
   hold 1 credit. Thus everybody can use Partner Queues without 
   requesting Travis CI support to grant additional credits. If you
   run into a negative account balance after that, you still need to
   file an additional request."

IOW, anyone ought to be able to use non-x86 jobs, bt if you accidentally
run an x86 job and get into 0 (or negative) credits, then you won't even
be able to use non-x86 jobs.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|