Stefano Rivera wrote:
It was not, it was triggered by me poking through buildd pages and noticing that the new version of simplejson was being blocked from building since it had picked up a dependency on pypy. I did consider filing a bug asking them to only build pypy stuff on architectures that had pypy but I decided to file a bug on pypy first and see what the response was.Hi peter (2013.11.22_20:36:12_+0200) I assume this was bug was triggered by the thread onpkg-openstack-devel.
My reply there hasn't appeared in the archives yet - so I'll repeat the relevant bits here.I appreciate that grinding swap on buildds is less than ideal but I still think it is preferable to not having the package on those architectures at all. Especially as slow architectures are where having optimised implementations of stuff is most important.Is that not a decision for those porters to make? What's the disadvantage of letting it continue to try to build there? It seems fairly harmless, to me. After installing the build-deps the package realises there isn't enough RAM, and aborts, wasting only a few minutes of buildd time.
Sorry if my post wasn't clear, what I meant to say is"I still think it is preferable to grind swap on a buildd for a few days than to not have the package built at all"
We accept build times in the days for several other packages, I don't see why we shouldn't accept them for pypy. Or do you think a pypy build with only 1GB of ram would take more than a "few days"?
I also have armhf machines with more than 1GB of ram but AIUI policy requires that packages be buildable on the autobuilders :/.I personally have a machine that's capable, and if the Debian armhf porters would like me to, I'd be quite happy to do binary uploads for it, when necessary.