Brice Goglin wrote:
> A couple other things that I am not sure about in the API:
Everything is done, so I am ok with releasing something now (once the
dynamic cpusets will be discussed).
We talked about adding the physical distances into the XML output of
lstopo, but it requires some non-trivial changes (and some thinking), so
I'd rather not put it in 0.9.1. XML backward compatibility shouldn't be
too bad anyway (maybe a warning if an unknown/expected node is found,