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

tests were change appropriately. This lets one fake a Linux cluster
without all of the BGL-specific logic.

  function on BGL.

to switch_node_info_t.

to attempt scheduling every pending job in priority order. Don't
stop scheduling for a SLURM partition after first job scheduling
failure as many different job parameters are considered: geometry,
rotate, conn-type, node-use.

Return number of bytes written on sucess rather than SLURM_SUCCESS.

  switch.

switch_sprintf_node_info.

  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.

  conn_type, and rotate).
slurmctld to set min_nodes for job based upon geometry (if supplied).
  This avoid trying to schedule jobs when insufficient resources are
  available.

  or co-processor), and communications type (mesh or torus).