Can set limits based off of TRESBillingWeights value.

Bug 4162.

Before:
$ sbatch --wrap="sleep 300"
Submitted batch job 228
$ squeue
             JOBID PARTITION     NAME     USER ST       TIME   CPUS NODELIST(REASON)
               228     debug     wrap    brian PD       0:00      1 (AssocMaxUnknownPerNode)

Fixed:
$ squeue
             JOBID PARTITION     NAME     USER ST       TIME   CPUS NODELIST(REASON)
               229     debug     wrap    brian PD       0:00      1 (AssocMaxCpuPerNode)

$ sacctmgr mod account stuff set maxtrespernode=cpu=-1,mem=1
$ squeue
             JOBID PARTITION     NAME     USER ST       TIME   CPUS NODELIST(REASON)
               229     debug     wrap    brian PD       0:00      1 (AssocMaxMemPerNode)

$ sbatch --wrap="sleep 300" --gres=blah:2 -pgpu
Submitted batch job 235
$ squeue
             JOBID PARTITION     NAME     USER ST       TIME   CPUS NODELIST(REASON)
               235       gpu     wrap    brian PD       0:00      1 (AssocMaxGRESPerNode)

bug 3673

burst_buffer/cray plugin modified to work with changes in Cray UP06
   software release.
Specific changes: Cray software now returns an error if a state_in
   or stage_out script is processed that doesn't actually request a
   stage in or out (previously silently ignored).
Also the warning message about tearing down a buffer that is already
   gone changed.

Need to resolve some build issues

User-submitted batch scripts cannot be safely escaped. This functionality
will be replaced with a way to retrieve the script directly into a file
for a given jobid.

This also removes the 'script' command from scontrol's interactive
cli mode.

Writes to slurm-<jobid>.sh if no (optional) filename is specified.

Bug 4199.

Move the check up a bit more where it'll do some good.

Bug 4184.

Bug 4146.

Bug 3467

Bug 3567

Add ability to prevent suspension of some count of nodes in a specified
    range using the SuspendExcNodes configuration parameter.

instead of native Slurm.

gres listed in your slurm.conf but some in gres.conf.

Bug 3974

'type' but no file defined.

Add support for specialized cores with task/affinity plugin (previously
    only supported with task/cgroup plugin).
Add "TaskPluginParam=SlurmdOffSpec" option that will prevent the Slurm
    compute node daemons (slurmd and slurmstepd) from executing on specialized
    cores.
bug 3337

in salloc, sbatch, srun before process the command line options.
Bug 3499

plugin when constraining devices.

correctly in sacct.

A second PMI2_Init() within the same step is invalid, and cannot succeed.

Return an error code back to the client end, and close the fd to force the
step to terminate immediately.

Due to a bug in our libpmi code, just returning a cmd=response_to_init with
an appropriate rc number will not tear down the connection properly, so
send back something else that will trigger the error path.

Bug 3520.

A request to cancel a pack step leader will result in that step
  being cancelled on all pack job components. Needed by MPI.

mpi/pmi2 plugin - vestigial pointer could be referenced at shutdown with
    invalid memory reference resulting.

Bug 4154.

Use the partition specified in the reservation, rather than
using the cluster's default.

Bug 4030.

default path.

This makes it so you don't always have to put AllowedDevicesFile in your
cgroup.conf file if your etc dir is anything other than /etc/slurm.

Don't flag a job as "SPAWNED" for debugger until all process
  information is available for all pack job components.

Adding a newline prevents this error:
conftest.c:154:8: error: if statement has empty body [-Werror,-Wempty-body]

remote cluster correctly determine the select type.

Bug 2329