The test7.3 program was run with CPUs_per_task == 0, which the
task distribution logic failed with (tried to bind the task to
no CPUs).

Add some white space and other minor reformatting
Change some print formats from %d to %u since the data is unsigned
Correct an incorrect xfree that should have been bit_free() or
  even better FREE_NULL_BITMAP()

Add some white space and other minor reformatting

The logic would previously send either configured or found counts,
but they should never differ, so remove some redundant logic.

Print the nodes core specialization details only if there is
something available. Prior code might print blank line.

Minor format changes and added commas.

If shutdown of the slurmctld daemon is in progress, then stop
trying to schedule jobs or process reconfigure requests. These
are the only operations that take a significant amount of time
and only service to slow down the shutdown process. We want the
daemon to stop processing incoming RPCs and save state as soon
as possible.

Previous limit was 4 secs, raised to 10.

Increase delays if memory leak test
Disable one test if tracking lustre file system use (leaves open file)

Eliminate one-time memory leak by freeing RPC statistics table at
slurmctld shutdown

Eliminate a one-time memory leack of the assocation hash table at shutdown.

select/cons_res plugin: Fix memory leak related to job preemption.
bug 837

Conflicts:
	src/slurmd/slurmd/req.c

d75bcaa5

Show exactly waht the bad ProfileHDF5Default value is.

ProfileHDF5Default=Filesystem should be
ProfileHDF5Default=Lustre

Added double brackets so brackets could be used within test program

Fix for commit 3f7bd421

Replace AC_TRY_RUN with more modern AC_RUN_IFELSE and assume
printf of NULL does not seg fault

This give system administrators the option on AMD Opteron 6000
series processors of either considering each NUMA node on a socket
as a separate socket (resulting in some incorrect logging of socket
count information) or not (resulting in sub-optimal job allocations
since each core in the socket will be considered equivalent, even
if on different NUMA nodes within the socket).
bug 838