of communication protocol over IBM Federation switch.
 -- Add new slurm.conf parameter ProctrackType (process tracking) and
    associated plugin in the slurmd module.

Major update for smap.  has new functionality to find the bp XYZ off a Rack/Midplane id or vice versa.  Plugin can work off getting the partitions from the system.  We still need to test plugin so the code is just skipped over right now.

 job steps on front-end system (e.g. Blue Gene).

    prolog is running (which can be slow on BlueGene).

Make bluegene plugin variable "bgl" local instead of global.
Remove slurm_rm_free_BGL() and directly use _rm_free_BGL() now that it works.

 don't seg fault if data is missing, free the data when the agent is
 finished with it.

This will gracefully recover from an assortment of slurmctld failure
modes.

partitions and faultly BGL driver.

inducing invalid memory reference.

prolog and epilog. Needed to synchronize slurmctld with user job execution:
Don't start script until partition is ready and clear all jobs upon
termination (offloading work from slurmctld).

RM_MESH to avoid multiple definitions for these values (one in slurm/slurm.h
and the other in bgl/rm_api.h.

 Suse Linux.

All work from <takao.hatazaki@hp.com>.

 uninitialized variables (thanks to valgrind).

  switch.

  data object in job_info records: slurm_get_select_jobinfo.

Move pretty much all BGL-specific logic into that module and the
  associated plugin and make use of an opaque data object for
  maintaining the information.

  one node. NOTE: srun uses hostlist to report task numbers.