>>>Right? If so, I don't see why you need the AC_TRY_RUN -- if RDMACM is
easily detectable as to which way it is compiled (because it has, for
example, different fields), then AC_CHECK_DECLS should be enough, right?
RDMACM API has different implementation requirements for its providers:
tcp, af_i
On Mar 13, 2014, at 4:59 AM, Mike Dubman wrote:
> >>>Right? If so, I don't see why you need the AC_TRY_RUN -- if RDMACM is
> >>>easily detectable as to which way it is compiled (because it has, for
> >>>example, different fields), then AC_CHECK_DECLS should be enough, right?
>
> RDMACM API ha
I think the critical point is this one:
> To be clear: whether AF_IB works or not is a determination to make on the
> machines on which you *run* -- NOT on the machine on which you *build*.
Many of our users compile on the frontend node of their cluster, which doesn't
even have an IB NIC instal
In the usual location:
http://www.open-mpi.org/software/ompi/v1.7/
Changes since rc2:
- MPI_Info_set now has const modifier
- Moar fixes to abnormal job termination scenarios
- Moar MPI-3 RMA fixes (including adding Fortran interfaces)
- Moar coll/ml fixes
- Change MPI_VERSION / MPI_SUBVERSI
