On Fri, 2008-11-14 at 23:12 -0600, Steve M. Robbins wrote: > Howdy, > > > On Thu, Oct 30, 2008 at 07:48:19PM -0400, Adam C Powell IV wrote: > > [Copying -beowulf as there's likely some interest there as well.] > > > > On Thu, 2008-10-30 at 15:21 +0100, Manuel Prinz wrote: > > > > When building against OpenMPI, there are a few choices: > > > > > > 1. Do not build packages using OpenMPI on the unsupported arches. > > > 2. Build against OpenMPI on the supported ones, fall back to LAM on the > > > unsupported ones. > > [ ... ] > > > As for -lam where there's no openmpi, I only know of petsc and babel. > > I have subsequently adopted this approach for minc, which uses MPI via > hdf5. I will likely adopt it for boost, too, unless someone has a > better idea. And ARPACK as well... > While reading this thread, however, I had an idle thought. Could we > prepare an "mpi-default-dev" or "sensible-mpi-dev" package for us to > build-depend on? This would be something like the gcc-defaults > package and simply depend on the appropriate -dev pacakges (OpenMPI on > some architectures, LAM on the rest). > > The idea is to put the messy details about which architectures support > OpenMPI and which use LAM in one place. That's a terrific idea. Build-Depends are not a big problem in terms of LAM vs. OpenMPI, because one can use architecture-dependent lists. But for dependencies of the -dev package (libpetsc-dev etc.), things are more tricky; you either need to make a substvar, or use things like type-handling's not+mipsel etc. Having an mpi-default-dev would make things a lot easier and clearer in both cases. Thanks! I can go ahead and take care of this later this week unless someone doesn't think it's a good idea. -Adam -- GPG fingerprint: D54D 1AEE B11C CE9B A02B C5DD 526F 01E8 564E E4B6 Engineering consulting with open source tools http://www.opennovation.com/
Attachment:
signature.asc
Description: This is a digitally signed message part