On Wed, Feb 15, 2023 at 09:05:28PM +0800, Shengjing Zhu wrote: > On Wed, Feb 15, 2023 at 8:53 PM Shengjing Zhu <z...@debian.org> wrote: > > > > On Wed, Feb 15, 2023 at 7:56 PM Adrian Bunk <b...@debian.org> wrote: > > > > > > On Wed, Feb 15, 2023 at 11:58:54AM +0800, Shengjing Zhu wrote: > > > >... > > > > The package currently FTBFS on i386/experimental but it won't be > > > > problem on > > > > unstable. > > > > The dep-resolver (aspcud) in experimental chooses gccgo to bootstrap, > > > > which has a bug https://github.com/golang/go/issues/51850. > > > > But on unstable the dep-resolver is apt, and will choose old golang-go > > > > to > > > > bootstrap. > > > >... > > > > > > I gave it back on the affected architectures with an extra-depends on > > > golang-go, which confirmed that the package builds there. > > > > > > But relying on resolver behaviour is incredibly fragile, please make > > > golang-go the only (or at least the first) alternative in the build > > > dependencies. > > > > In theory gccgo should be able to bootstrap golang-go. > > And we do use gccgo to build golang-go on ppc64, and haven't met any > > problem so far. > > The new affected arch is because of lack of exercise, so nobody is > > aware of this. > > Now they are tracked at #1031349. > > > > And regarding make golang-go the first alternative, currently we have: > + Build-Depends golang-any | golang-go | gccgo > + golang-any Depends golang-go | gccgo-go > Is there anything we can improve for aspcud resolver?
The resolver is not to blame for a bug in the build dependencies. In unstable and experimental only the first alternative is considered, therefore Build-Depends: golang-go | ... would instruct the resolver to install golang-go. Safest would be to not offer any alternatives in the build dependencies, instead create a profile for bootstrapping golang-go from gccgo. > Shengjing Zhu cu Adrian