This prevents bad node index values in a job step completion record
from crashing slurmctld, as is possible if srun has bad configuration
information about a job step or other failure.

Add a basic test of Bluegene/Q job step allocations within an existing
job allocation.

on a Bluegene/Q system when srun's --test-only option is used within
an existing allocation then launch the job directly with the slurmd
daemon and do not use IBM's "runjob" command. Useful for testing.

BLUEGENE - Modify "scontrol show step" to show  I/O nodes (BGL and BGP) or
c-nodes (BGQ) allocated to each step. Change field name from "Nodes=" to
"BP_List=".

The test is now more generic to support all Bluegene system types

cannot fit into the available shape.

Modify existing tests so they all run as desired on an emulated Bluegene/Q system

Previous code would fail when trying to launch more than 4096 tasks,
which is a problem on BGQ systems where SLURM actually launches job
steps.

The SLURM_JOB_CPUS_PER_NODE and SLURM_TASKS_PER_NODE environment
variables were being improperly set for IBM Bluegene systems

or not.

This change applies only to Cray systems and only when the srun
wrapper for aprun. Map --exclusive to -F exclusive and --share to
-F share. Note this does not consider the partition's Shared
configuration, so it is an imperfect mapping of options.

On Cray systems only, the value of avail_node_bitmap was not being
properly set for non-responsive nodes.

A node DOWN to ALPS will be marked DOWN to SLURM only after reaching
SlurmdTimeout. In the interim, the node state will be NO_RESPOND. This
change makes behavior makes SLURM handling of the node DOWN state more
consistent with ALPS. This change effects only Cray systems.

Fix the node state accounting to be consistent with the node state
set by ALPS.

Split set_node_down() into two functions: set_node_down() will continue
to accept a node name as an argument and set_node_down_ptr() which is new
and accepts a node pointer as an argument and will be faster.

Test4.5 was failing due to failure to parse node count with "K" suffix and
change in case of node state name.

state of block to Free if need be instead of leaving it in Term

Modify salloc, sbatch and srun man pages to clarify how max node
count is used.

be the same.

previously marked down by alps.

previously marked down by alps.

set.

Require SchedulerTimeSlice configuration parameter to be at least 5 seconds
to avoid thrashing slurmd daemon.
Addresses Cray bug 774692

Change in GRES behavior for job steps: A job step's default generic
    resource allocation will be set to that of the job. If a job step's --gres
    value is set to "none" then none of the generic resources which have been
    allocated to the job will be allocated to the job step.
Add srun environment value of SLURM_STEP_GRES to set default --gres value
    for a job step.

Major update to src/common/node_select.h to document the API calls.
Minor updates to node select plugin web page.