This is so we can more easily change the data structures for
  burst_buffer/cray without worrying about the generic plugin,
  which is a "throw-away" at this point.

Major refactoring of the burst_buffer/cray check for available space
 (before allocating a new buffer) including preemption logic. These
 changes are needed to support multiple pools.

Interactive pools support
state save/restore fix

Remove vestigial common functions for persistent buffer management
Get utilization calculation by pool working

Add granularity info to alternate pools
Also change the scontrol show burst buffer format for bb/GRES:
  "  Gres[%d] Name=%s AvailCount=%s UsedCount=%s" to
  "  AltPoolName[%d]=%s Granularity=%s TotalSpace=%s UsedSpace=%s"
  for better consistency with default pool information
Add pool name to burst buffer creation data structures
Create persistent buffer in user-specified pool
Add pool info to allocated buffer table
Fix buffer state load logic, was creating duplicate pool/gres records
bug 2190

When extra pools were reported by Cray API, it was not multiplying
  the buffer count by granularity to get the total size, but was
  just getting the counts.

Move slurmctld mail handler to separate thread for improved performance.
Original logic did fork/exec without separate thread and if the
slurmctld memory size is huge, then the time required for fork()
to complete can be significant.
bug 2252

Conflicts:
	src/slurmctld/job_scheduler.c

Conflicts:
	src/slurmctld/job_mgr.c

Avoid hard-link/copy of script/environment files for job arrays. Use the
    master job record file for all tasks of the job array.
back port of commit 41eead64
bug 2220

Refactor logic to synchronize active bach jobs and their script/environment
    files, reducing overhead dramatically for large numbers of active jobs.
backport of commit 41eead64
bug 2220

If a job can never be started due to unsatisfied job dependencies, report
    the full original job dependency specification rather than the dependencies
    remaining to be satisfied (typically NULL).
bug 2197

in use.

Conflicts:
	src/slurmctld/proc_req.c

_slurm_rpc_complete_job_allocation.

This is a rewrite of 438365ec which didn't catch the job_ptr wasn't
in a lock so the memory issue could still of existed.  This hopefully fixes
all the spots the job_ptr wasn't in the lock.

Fixes bug 2146

This reverts commit 438365ec.

Bug 2261

15.08.5.

Bug 2243

Fix state machine to support corner cases when fan-in message for the next Fence appears before fan-out message completing the current Fence. This was observed on Orion as well as on containers environment in cases where Fences was light but the number of process was reasonably high.

Change the ordering of operations during fan-in. We will need this for the improved state machine. This changes are verified to work.

Optimization of fan-in collective stage. Sense parent by sending him a zero-sized message before sending him a junk of data. This helps to resolve cases where parent slurmstepd in collective haven't been yet bootstrapped.