Add new knl.conf configuration parameter CapmcRetries
Modify capmc_suspend and capmc_resume to retry operations when
  Cray State Manager is down.
Add retry logic to node_features/knl_cray to handle Cray State
  manager being down.
bug 3100

Otherwise they'll truncate when packed into the RPC and end up
as some bizarre value at the controller.

Bug 3098.

Set completed time for pending/running runaway jobs to the max of
(start, eligible, submit) times.

Bug 3075

Added new SchedulerParameters options step_retry_count and step_retry_time
    to control scheduling behaviour of job steps waiting for resources.
bug 3121

This change can reduce RPCs sent by srun up to 50%.

Avoid using slurm_forward_data because it causes thread spawn that
introduces unwanted delays.

Bug 3102.

THis can happen due to signal/retry logic and the result is
  vestigial srun ports that are no longer valid and slurmctld
  sending messages to invalid ports to notify them about available
  resources as other steps complete

Also correct the value of NICE_OFFSET used within the perl API.

Bug 3098.

Bug 3051.

Otherwise updates would be rejected for running jobs even if there
would be no impact. Most common when the job_submit plugin is overriding
QOS/GRES values on everything; without this change an update to "comment"
or other fields would fail with ESLURM_JOB_NOT_PENDING.

Bug 3117.

Never ever run NHC, even on an edge case that NHC_NO would still launch NHC
after.

Bug 3105.

Switch to list_for_each, and check if access list actually changed
after each update before updating last_prat_update. This prevents
the backfill scheduler from resetting mid-cycle unnecessarily.

Bug 3123.

Switch to list_for_each, and check if access list actually changed
after each update before updating last_prat_update. This prevents
the backfill scheduler from resetting mid-cycle unnecessarily.

Bug 3123.

Add protocol version to slurmd startup communications for slurmstepd to
    permit changes in the protocol.

Add "flag" field to launch_tasks_request_msg. Remove the following fields
    (moved into flags): multi_prog, task_flags, user_managed_io, pty,
    buffered_stdio, and labelio. More flags to be added later.

Remove from build system, and delete L/P specific files.
Run autogen.sh as well.

Add "sbatch_wait_nodes" to SchedulerParameters to control default sbatch
    behaviour with respect to waiting for all allocated nodes to be ready for
    use. Job can override the configuration option using the --wait-all-nodes=#
    option.
bug 3120

Prior logic would treat execute line like this:
$ sbatch --wait-all-nodes -N3 tmp
with "-N3" as being the argument to the "--wait-all-nodes" option.
See bug 3120

Add salloc/sbatch/srun option --use-min-nodes to prefer smaller node counts
    when a range of node counts is specified (e.g. "-N 2-4").
bug 2996

Add salloc/sbatch/srun --priority option of "TOP" to set job priority to
    the highest possible value. This option is only available to Slurm operators
    and administrators.
bug 3115

bug 3090

Make sure no attempt is made to schedule a requeued job until all steps are
    cleaned (Node Health Check completes for all steps on a Cray).
bug 3082

Make sure no attempt is made to schedule a requeued job until all steps are
    cleaned (Node Health Check completes for all steps on a Cray).
bug 3082

Otherwise limit is checking the node count against the midplane count.

Bug 3049.

Check if node names are contiguous with respect to the node list
assigned to the partition, rather than just monotonically increasing.

Bug 3006.

Bugs 2681 and 2703

Conflicts:
	NEWS

license of a certain type.

node_features/knl_cray plugin: Increase default CapmcTimeout parameter from
    10 to 60 seconds.
bug 3100

capmc_suspend/resume - If a request modify NUMA or MCDRAM state on a set of
    nodes or reboot a set of nodes fails then just requeue the job and abort the
    entire operation rather than trying to operate on individual nodes.
bug 3100

Allow a node's PowerUp state flag to be cleared using update_node RPC.
bug 3100

When powering up a node to change it's state (e.g. KNL NUMA or MCDRAM mode)
    then pass to the ResumeProgram the job ID assigned to the nodes in the
    SLURM_JOB_ID environment variable.
bug 3100

Don't log error for job end_time being zero if node health check is still
    running.
bug 3053

capmc_suspend/resume - If a request modify NUMA or MCDRAM state on a set of
    nodes or reboot a set of nodes fails then just requeue the job and abort the
    entire operation rather than trying to operate on individual nodes.
bug 3100