Add suppoprt for --mem_bind=prefer option to prefer, but not restrict memory
    use to the identified NUMA node.
bug 3126

Modify cpu_bind and mem_bind map and mask options to accept a repetition
    count to better support large task count. For example:
    "mask_mem:0x0f*2,0xf0*2" is equivalent to "mask_mem:0x0f,0x0f,0xf0,0xf0"
bug 3065

Remove include, this was merged into xstring.h with the
rest of the string functions.

This reverts commit 8bc8e7e6.

The slurm_strcasestr() which existed in v16.05 was removed in v17.02.

Found by Coverity

Fix for possibly treating a negative number as a positive.
Problem reported by Coverity.

Change the default syscfg path to /usr/bin/syscfg
Don't load the plugin if ResumeProgram is configured
Update documentation on web page

This applies only to knl_generic plugin. The original node active
features should be the same as the features field at startup.
The logic was assuming the original value was NULL.

This was a new function added to support knl_generic and it was
  originally assigned the wrong name.

Added node_features/knl_generic plugin for KNL support on non-Cray systems.
NOTE: This plugin is still under development.

Do not propagate SLURM_UMASK environment variable to batch script.
bug 2609

task/affinity plugin: Honor a job's --ntasks-per-socket and
    --ntasks-per-core options in task binding.
bug 3118

Correct a bitmap test function (used only by the select/bluegene plugin).
  The effect of this bug is probably very limited as it will in almost
  all cases revert prematurely to a bit-by-bit test rather than using
  a full-word test.
bug 3145

Cannot use ClusterName without reading a config file
that may not exist.

Bug 3026.

This introduced an inadvertent dependency on the config file,
which does not exist when setting up a new cluster.

Bug 3026.

This reverts commit c39f9ac9.

task/affinity plugin: Honor a job's --ntasks-per-socket and
    --ntasks-per-core options in task binding.
bug 3118

Bug 3007

e.g.
brian@lappy:~/slurm/master/lappy$ scontrol show res
ReservationName=root_167 StartTime=2016-10-12T10:24:35 EndTime=2016-10-12T10:54:35 Duration=00:30:00
   Nodes=lappy[1-10] NodeCnt=10 CoreCnt=40 Features=(null) PartitionName=(null) Flags=OVERLAP,SPEC_NODES,ALL_NODES
   TRES=cpu=80
   Users=root Accounts=(null) Licenses=(null) State=ACTIVE BurstBuffer=(null) Watts=n/a

ReservationName=root_168 StartTime=2016-10-12T10:24:37 EndTime=2016-10-12T11:24:37 Duration=01:00:00
   Nodes=lappy[1-10] NodeCnt=10 CoreCnt=40 Features=(null) PartitionName=(null) Flags=OVERLAP,SPEC_NODES,ALL_NODES
   TRES=cpu=80
   Users=root Accounts=(null) Licenses=(null) State=ACTIVE BurstBuffer=(null) Watts=n/a

brian@lappy:~/slurm/master/lappy$ sbatch --wrap="sleep 60" -t1 --test-only
sbatch: Job 14435 to start at 2016-10-12T11:24:37 using 1 processors on lappy1

The for loop wasn't resetting the job_start_time to the reservation's
end_time. This was taken out in 06dc5fa7.

Changed in df70b651

Preserve non-KNL node features when updating the KNL node features for a
    multi-node job in which the non-KNL node features vary by node.

node_features/knl_cray plugin: If the reconfiguration of nodes for an
    interactive job fails, kill the job (it can't be requeued like a batch job).

Execute "capmc node_status" at more frequent intervals to handle
  nodes getting added or removed from the system using Cray tools
  (i.e. try to keep Slurm and Cray software better synchronized).

node_features/knl_cray plugin: Add separate thread to interact with capmc
    in response to unexpected node reboots.
bug 3153