0x000000008fffffff is wrong and produces some weird side-effects.

Move to xstring.[ch], and use sed to swap out all references.

Will handle NO_VAL/INFINITE <-> NO_VAL64/INFINITE64 conversion,
as well as MEM_PER_CPU flag.

MaxMemPerNode, MaxMemPerCPU, DefMemPerNode, DefMemPerCPU

Used to pack/unpack new uint64_t arrays into older uint32_t
array format for RPCs.

Change RPCs and add encode/decode logic for past version.

Change variables and structures to uint64_t:
def_mem_per_cpu, max_mem_per_cpu, free_mem, real_memory, actual_real_mem,
job_mem_lim, step_mem_limit, pn_min_memory, job_mem.

Change MEM_PER_CPU flag to 64th bit instead of 32nd bit.

Note that additional work is required to address local variables.

Also need to add table conversion logic to MySQL, as mem_req column
needs to change type to 'bigint unsigned' from 'int unsigned'.

Copy previous format, but do not change yet. Update RPCs involving:
def_mem_per_cpu
max_mem_per_cpu
free_mem
real_memory
actual_real_mem
job_mem_lim

bug 3065

Add Slurm overview by Alex on first day, move KNL to second morning,
shift roadmap to right after lunch.

bug 3065

Previous cap was 2 sec (default TCP timeout) times the node count
  and divided by 1000. A 9000 node job would have the messages
  spread out over 18 seconds. This change caps the spread at
  5 seconds and assumes the normal TCP logic can handle the rest
bug 3044

If the overhead of determining the hostlist for a given task list
is too high, then report a hostlist of "Unknown" instead. If the
overhead is too high, then srun will become unresponsive and communications
will timeout/fail.
bug 3044

Modify srun task completion handling to only build the task/node string for
    logging purposes if it is needed. Modified for performance purposes.
bug 3044

Add get_log_level() function to return the highest LOG_LEVEL_* used
for any logging mechanism.

This reverts commit 1ec2a4ae.

Conflicts:
	src/api/step_launch.c

Bug 3063.

In scontrol show nodes.

Restructure srun command locking for task_exit processing logic for improved
  parallelism. This change decreases the amount of time consumed by serial
  logic by 2 orders of magnitude.
bug 3044

Grab federations from db at startup instead of waiting for db_update and
load from state if the db is down.

Instead of making a separate copy. All of the cluster_recs are now in
the federation_rec with a pointer to the local cluster rec.