Dear members of the Mono SIG group,
I have been working in the past week on building Mono 5 for Fedora Rawhide.
This has been made possible by the work of directhex, who fixed the build for Debian.
For details on his work see https://github.com/mono/mono/issues/7445
The discussion lives at https://bugzilla.redhat.com/show_bug.cgi?id=1436896
Unfortunately we are too late for the upcoming Fedora 30, since the change affects quite a number of packages.
For Mono 4, we needed a bootstrap build, and filed it as a system wide change: https://fedoraproject.org/wiki/Changes/Mono_4
I will prepare a similar change proposal for Mono 5 for Fedora 31.
Please can you look at the code, and evaluate if the package is properly built?
The source is still here: https://github.com/tpokorra/mono-5.x-fedora/tree/master/mono-5.18
After the change proposal gets accepted, I would put it in the proper source control in https://src.fedoraproject.org/rpms/mono
The package is built here: https://copr.fedorainfracloud.org/coprs/tpokorra/mono-5.18/
Another issue I want to bring up is the state of Mono 4 in Fedora 30.
I probably committed a mistake, by disabling the ppc64 builds for Mono due to a long standing bug on that architecture.
I just did not want mono to be FTBFS for yet another Fedora release.
See the details here: https://bugzilla.redhat.com/show_bug.cgi?id=1588734
It seems, that now all packages depending on Mono cannot be built anymore, because they still have ppc64 enabled.
What do you think, is this really a problem, and if yes, how can we resolve it?
In Rawhide, things should go back to normal as soon as Mono 5 has landed.
Should we consider porting Mono 5 back to Fedora 30 after the Fedora 30 release? But that would contradict calling it a system wide change...
Or we disable ppc64 in Fedora 30 altogether? It should be available in Fedora 31 again.
Any suggestions and help in testing and evaluating are welcome!
All the best,
Timotheus
