The pack_node_list may not be ordered across multiple components, which can
  cause problems for some MPI implementations. Put the pack_node_list records
  in alphabetic order and reorder pack_task_cnts pack_tids to match

Previously the accounting sometimes recorded the job, but not the
  short lived step. It is working consistently for me now.

Problem found by test1.65

Add support for specialized cores with task/affinity plugin (previously
    only supported with task/cgroup plugin).
Add "TaskPluginParam=SlurmdOffSpec" option that will prevent the Slurm
    compute node daemons (slurmd and slurmstepd) from executing on specialized
    cores.
bug 3337

CID 177245, 177246.

Coverity CID 177244

When a job array record is split, add the new record to the
  federation job table.
bug 4033

Utility functions for new x11 forwarding implementation.

Requests the allocated port only be available to localhost.
Follow through with the locations in the API using this, which
all need to be marked as 'false'.

Grab this so we can start using it elsewhere. Rename command
from run_script to run_command (there is already a different
run_command in the slurmd/common code that lead to conflicts).

Also grab _free_script_argv while here, and make sure we can still run
the same shutdown commands the Cray plugin relied on as before.

Previously these would show as jobid and jobid.4294967295.

While here tweak the log_set_fpfx function to take in an
xmalloc'd string directly, rather than clone it again.

Revert part of commit b0e6d3c3

Replace [file exist $file_out] with [wait_for_file $file_out] so
  we not only wait for the file to appear, but for I/O to complete

Bug 4178.

SLURM_MPI_TYPE= env variable would take precedence over command line.

Debugging logic was left in place

Restores variable definitions accidentally removed
by e4eaf07f.

Bug 4174.

Removed by mistake in d8d41cff.

Problem was do to "go to proceed;" before setting "nodename"
   or "lpeerid".

Send pack job's task IDs, node count, tasks_per_node and node_list
  from srun to slurmd as needed by MPI. This information is constructed
  by srun combining information from each individual job component's
  step information.
Set global SLURM_NODELIST, SLURM_JOB_NODELIST, SLURM_STEP_NUM_NODES,
  SLURM_TASKS_PER_NODE, SLURM_NNODES, SLURM_NTAKS and SLURM_STEP_NUM_TASKS
   for pack step.
Modify pmi2 plugin to launch a single agent for pack step rather than
   one per component.
Modify pmi2 and pmix plugins to gather global step information for
   pack steps.
Make task distribtion be configurable by step component (i.e. "block"
   on one and "cyclic" on another).