Stat the file before allocating buffer used for read rather than
  reallocating the buffer as needed. Improves performance (no
  reallocate/copy needed).

sinfo to group by like ActiveFeatures
sinfo long format option for ActiveFeatures
slurm.conf parse to accept node "Features" as well as current "Feature"
scontrol to report AcitveFeatures and AvailableFeatures on separate lines
If a user sets node available features and the specification includes "knl"
  then expand that to include all possible NUMA and MCDRAM modes, the same
  as if "knl" is specified in slurm.conf
Confirm that when AcitveFeatures is set, it is a subset of the
  AvailableFeatures

Conflicts:
	src/plugins/burst_buffer/cray/burst_buffer_cray.c

Conflicts:
	src/slurmctld/job_scheduler.c

Bug 2314.

The slurmctld behaviour has changed slightly with respect to EnforcePartLimits
  supporting new options of ANY and ALL, which broke  2 tests. This change
  modifies the tests to only run on appropriate configurations.

this happens anywhere in the code but just incase it ever does, lets fix
it.

Continuation of fa331e30.

This can be caused by a core reservation on nodes which get taken
  out of the system or fail.
bug 2296

if it is a script or not.

After that ignore them.

Print step id value as "batch" rather than 4294967294
bug 2296

Add "features_act" field (currently active features) to the node
    information. Output of scontrol, sinfo, and sview changed accordingly.
    The field previously displayed as "Features" is now "AvailableFeatures"
    while the new field is displayed as "ActiveFeatures".

commit.

partition limit.

cnodes can be reserved directly since 14.11. The plugin itself
printed warnings that it would be removed circa 15.08, following
through before 16.05.

Cray: Not running the Node Health Check after every job and step is now the
    default. Configure SelectTypeParameters with the NHC and/or NHC_STEP to
    run them.

salloc/sbatch/srun did not mention this. Also reference OverTimeLimit
as another option affecting the final run time. Bug 2309.

the job starts update the cpus_per_task appropriately.

This also moves update num_tasks to after the setting of node counts on
an update.  It didn't appear to matter, but the cpus_per_task and
pn_min_cpus had to be figured out after the cpus and nodes were set but
before tasks.

Bug 2302

Add an "scontrol top <jobid>" command to re-order the priorities of a user's
    pending jobs. May be disabled with the "disable_user_top" option in the
    SchedulerParameters configuration parameter.
bug 1133

Add some more comments to identify RPC code numeric values
Put the number to string mapping into numeric order and add some numbers in the comments

Conflicts:
	NEWS
	src/plugins/burst_buffer/cray/burst_buffer_cray.c
	src/plugins/burst_buffer/generic/burst_buffer_generic.c

Don't decrease a pool's "used_space" below the amount we believe has
  been allocated in case there is a buffer allocation in progress
  while the _load_state() function is running.
Initialize a mutex
Move loading of state_save_location earlier

burst_buffer/cray - Improve tracking of allocated resources to handle race
    condition when reading state while buffer allocation is in progress.
Also initialize a mutex

DBD for the first time.  The corruption is only noticed at shutdown.

Bug 2293

Conflicts:
	META
	NEWS
	src/slurmctld/job_scheduler.c

Set job's reason to "Priority" when higher priority job in that partition
    (or reservation) can not start rather than leaving the reason set to
    "Resources".
bug 2285