srun wrapper for srun not getting all arguments when job allocation did not
exist at execution time.

If the srun wrapper is executed when there is no job allocation, then
create an allocation using salloc and call the srun wrapper again so
that we can configure memory limits in aprun's execute line. Without
this change, aprun would lack the memory allocation information and
the task launch would fail if the job were allocated less than the
full node.

Prevent duplicate arguments to aprun from the srun.pl wrapper.
This could happen if the command line included "--alps" arguments
plus other arguments generated by the normal srun options. For example:
srun -t 5 --alps="-t300" a.out
specifies the job time limit in two places.

Set SLURM_MEM_PER_CPU or SLURM_MEM_PER_NODE environment variables for both
interactive (salloc) and batch jobs if the job has a memory limit. For Cray
systems also set CRAY_AUTO_APRUN_OPTIONS environment variable with the
memory limit.

Clarify in the srun (aprun wrapper) which options apply to an existing
job allocation or new allocation and which are not applicable to Cray
computers.

asking for less than 1 mb per PE.

Correction to srun man page. Get SIGINT working when srun spawns salloc.

For front-end configurations (Cray and IBM BlueGene), bind each batch job to
a unique CPU to limit the damage which a single job can cause. Previously any
single job could use all CPUs causing problems for other jobs or system
daemons. This addresses a problem reported by Steve Trofinoff, CSCS.

There was a table added, but it all ran together without being put
into a table or list. I changed it to an unordered list.

enforce memory limits.

man pages. Patch by Nancy Kritkausky, Bull.

Bill Brophy, Bull.

Note the job and partition state file formats have changed and RPCs
with information for jobs and partitions have changed.

code in smap was referencing a bad symbol name. The relevant code was
commented out, but is needed for Bluegene/Q

Typo fix to sample suspend and resume scripts.

make the init_wires() definition and use consistent for BGL, BGP, and
BGQ systems. No argument is needed or used in any system.

up correctly when starting.

link to "disclaimer.html"

Fix bug in generic resource tracking of gres associated with specific CPUs.
Resources were being over-allocated.

Remove vestigial (unused) variable from smap.

Fix memory buffering bug if a AllowGroups parameter of a partition has 100
or more users. Patch by Andriy Grytsenko (Massive Solutions Limited).

Add cgroup support for device files in both the task/cgroup plugin and generic
resource (GRES) logic. Based upon patch Yiannis Georgiou.

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).