When suspending a job, wait 2 seconds instead of 1 second between sending
SIGTSTP and SIGSTOP. Some MPI implementation were not stopping within the
1 second delay.

Add contribs/arrayrun tool providing support for job arrays. Contributed by
Bjørn-Helge Mevik, University of Oslo. NOTE: Not currently packaged as RPM
and manual file editing is required.

If a job needed to preempt other jobs to start and those jobs were
not completed by the time of the next scheduling cycle, other jobs
might be selected for preemption in that next cycle resulting in
more jobs being preempted than necessary.

Previous logic reported the run time as the current time minus the job start time,
ignoring any suspended time.

Export the symbol s_p_hashtbl_destroy as slurm_s_p_hashtbl_destroy
in libslurm so that external programs can link to libslurm and be
able to reference select/cray without undefined symbols

If the gres/gpu plugin is used, then set the CUDA_VISIBLE_DEVICE environment
variable to "NoDevFiles" if the gres.conf file has no device files identified
for generic recources (GRES) of type GPU. Otherwise set the device file sequence
number(s).

Enhancements to sched/backfill performance with select/cons_res plugin.
Major improvements would be seen with large job counts. Based upon
bf_build_row_bitmaps_2.2.6.patch patch from Bjørn-Helge Mevik, University of Oslo.

Add a call to acct_policy_job_runnable() to check if a job should have been marked with
AssociationResourceLimit.  This avoids trying to backfill any job who wouldn't be allowed
to run now anyway.
Patch bf_acct_policy_check_2.2.6.patch from Bjørn-Helge Mevik, University of Oslo

Have salloc log (with verbose option) when the job is suspended or resumed,
but do no suspend or resume it's child processes when that occurs. ALPS has
been modified to do so and salloc no longer needs to do it.

association limit.

Permit scontrol to update a partition's default and maximum memory limits
per job.

Add basil_switch function to select/cray libemulate (ALPS eumulation library)
to function properly with emulated Cray system.

Add default and maximum memory limits on a per-partitiion basis. If not
specified, the system-wide memory limits will apply.

Describe alps.conf changes required for gang scheduling on a cray system

Auble and Jette listed with SchedMD rather than LLNL.

Update web page header and footer to be generic (not LLNL-centric) and
add generic disclaimer web page.

from how they are in the slurm.conf file but doesn't use the a topology
plugin.

Change use of pointer in a step launch RPC to a copy of the data structure
which will insure that the step data does not change while the RPC message
is being built. No problems have been observed, but this will be safer.

"scontrol show config" was reporting both per CPU and per node memory
limits as being per CPU limits. Change to report per node limits with
the proper key name.

Correct values mainted for suspended job count (sus_job_cnt) by node and
run job count (job_cnt_run) by front-end node when reconfiguring the slurmctld
daemon while there are suspended jobs on a front-end architecture.

scheduling

Add select_jobinfo to the task launch RPC so that all nodes have access to
the information and not job the head node.  Based upon patch by Andriy
Grytsenko (Massive Solutions Limited).