David Woodhouse wrote:
A question which had occurred to me also. We seem to be trying to
'solve' a problem which hasn't actually been demonstrated to exist yet.
While it has not happened in a while (my current issues are s390x bound,
not ppc bound) having to do a single rebuild of firefox even to add an
ExcludeArch is a huge loss when trying to get security fixes pushed
where time is critical. If for whatever reason there crops up a c++
compiler bug on the platform -- and Mozilla will exhibit it as it does
all sorts of funky c++ fu -- delaying the security fix while someone
debugs and fixes the secondary arch problem is not something I'd like to
see as a maintainer, nor something that I imagine we'd like to see in
general as a project.
I'm all for working with whomever to get secondary arches working, but I
am generally building packages for security issues, and I'd strongly
prefer to not do re-spins because of failures on arches we don't
officially support. It's hard enough getting them out on arches we do.