Hi Adam, since it's been 10 months and I haven't heard back, I'd like to gently ping you on this?
On 11/12/2015 01:41 AM, Christian Seiler wrote: > The problem I see is that if I add the string "x32" to the list of > architectures for the udeb, src:open-iscsi will build a binary package > on an architecture that is not installable at all (because dependencies > can't be resolved). If x32 were to become an official release > architecture tomorrow, this would immediately prevent testing migration > of the open-iscsi package. Of course, it currently isn't, and solving > the question about what to do with the kernel is probably a > prerequisite to making it a release architecture, so the danger there > is probably very slim. > > Still, I'm kind of hesitant to add something to the package that could > blow up some time in the future. On the other hand, I want to support > porting efforts and do understand that sometimes things like this can > be chicken-or-egg problems (i.e. you need to have a sufficiently > diverse set of udebs to be actually able to work on d-i support for the > specific port and test things). > > Since I don't see a clear-cut answer, I'm going to defer to you here, > since you are working on the port. If you say that it helps you in > your efforts to just have this in the archive, even though on the not > yet official x32 architecture the package is not installable, I'll > add it to debian/control and close this bug. The other possibility > would be to block this bug against some other bug that tracks x32 d-i > work. Please tell me which way you'd prefer. Regards, Christian