Added SchedulerParameters option of "bf_busy_nodes". When selecting
resources for pending jobs to reserve for future execution (i.e. the job
can not be started immediately), then preferentially select nodes that are
in use. This will tend to leave currently idle resources available for
backfilling longer running jobs, but may result in allocations having less
than optimal network topology. This option is currently only supported by
the select/cons_res plugin.

Improve the logic for advanced reservations with the "replace" flag
to select idle nodes for use. Preserve nodes already in the reservation
as needed (when not free idle nodes exist).

and adjust the style.

Conflicts:
	src/plugins/slurmctld/dynalloc/allocate.c
	src/plugins/slurmctld/dynalloc/msg.c

Prevent advanced reservation with "replace" flag except for whole
node reservation (not with CPU count) and not with specific node
names in the reservation.

This can happen if the specified job ID is not found.

Add advance reservation flag of "replace" that causes allocated resources
to be replaced with idle resources. This maintains a pool of available
resources that maintains a constant size (to the extent possible).

This plugin was never completed, so it is not worth the effort of
continued support at this time.

This plugin was never fully completed and should probably be removed
from v15.08

cgroup plugin.

fix some burst buffer job preemption bugs
allocate burst buffer at job submit time if possible
trigger scheduling cycle when stage in completes

logged buffer state rather than job id in 3 message
reported bad stage-out results for purged buffer

Add more error checking to burst_buffer.conf contents
Add function to provide estimated earliest start time for pending job

Improve logic for failure recover on slurmctld restart
Add burst buffer use start time for use with revoking previously allocated buffers
Revoke burst buffers already allocated to pending jobs when needed
  for a higher priority job

add timer logging
add on demand job state load
Streamline logging to make less verbose
Fix bug related to tracking space allocated to staged out buffers
Sort pending jobs in order of expected start time

impact functionality.

Conflicts:
	RELEASE_NOTES

Copy (i.e. duplicate) information about floating reservations
in the highlights section

Added new job state of STOPPED indicating processes have been stopped with a
SIGSTOP (using scancel or sview), but retain its allocated CPUs. Job state
returns to RUNNING when SIGCONT is sent (also using scancel or sview).
bug 1263

No change to logic

The xtree functions are only used in the layouts framework, which
isn't currently in use, so I am just commenting out the failing
portion of the test for now. It was failing with the message:
xtree-test.c:517:F:Core:test_xtree_walk:0: unexpected stop (data, which, level, couple index) expected: 1: 1: 0: 0, got 1: 16: 0