On Fri, 14 Jun 2024 at 11:54, <r...@neoquasar.org> wrote: > > Then it's not a problem in the first place. If you can't reproduce a bug with > a reasonable effort, then it is unconfirmed and you can stop worrying about > it. A bug that can't be reproduced, effectively doesn't exist. > > That's not a reason to stop supporting an entire architecture. That's a > troubleshooting decision that you would make on any architecture.
Of course it is a reason to drop an architecture. Not the mere existence of an individual bug of course, but the fact that there is nobody who is able or willing to successfully triage and debug and fix such toolchain issues. As per RT's requirements, a release architecture requires multiple porters as one of the basic requirements to be considered, and the fact that nobody can fix issues like these means there are effectively 0 porters. Once again, maintaining a release architecture does not just mean having a machine under your desk and building trivial packages, it means maintaining a functioning toolchain, which involves triaging and fixing complex bugs.
