Thanks for filing the bug, Julien! Hopefully it's already fixed upstream, I'll marked it as closed in the next release.
The memory problem was due to a recent change to Eigen3, a cool library for matrices, but it *massively* relies on C++ templates and leads to this sort of things. I just split big .cpp files into smaller pieces and observed the memory consumption was reduced. Let's see... BTW: Do you know if those build errors are what prevents mrpt-0.9.4 to move into "testing" or it is something else? Cheers, JL On Wed, Mar 9, 2011 at 5:46 PM, Julien Cristau <jcris...@debian.org> wrote: > Package: src:mrpt > Version: 1:0.9.4-1 > Severity: serious > Justification: fails to build from source (but built successfully in the past) > > building mrpt seems to require insane amounts of memory, leading to > ftbfs on buildds: > s390 got OOM(?) killed: > https://buildd.debian.org/fetch.cgi?pkg=mrpt&arch=s390&ver=1%3A0.9.4-1&stamp=1295783490&file=log&as=raw > > mips went out of virtual memory: > https://buildd.debian.org/fetch.cgi?pkg=mrpt&arch=mips&ver=1%3A0.9.4-1&stamp=1294673927&file=log&as=raw > > mipsel didn't output anything for so long the buildd timed out: > https://buildd.debian.org/fetch.cgi?pkg=mrpt&arch=mipsel&ver=1%3A0.9.4-1&stamp=1294547471&file=log&as=raw > > Cheers, > Julien > > -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
