Use existing functions and data types to get the information

29d79ef8

same time a block is destroyed and that block just happens to be the
smallest overlapping block over the bad hardware.

Conflicts:
	src/srun/opt.c

When a new reservation is created, the code tries first with current
idle nodes:

    /* First: Try to reserve nodes that are currently IDLE */
    if (bit_overlap(avail_bitmap, idle_node_bitmap) >= node_cnt) {
        bit_and(avail_bitmap, idle_node_bitmap);
        ret_bitmap = select_g_resv_test(avail_bitmap, node_cnt,
resv_desc_ptr->core_cnt, core_bitmap);
        if (ret_bitmap)
            goto fini;
    }

    /* Second: Try to reserve nodes that are will be IDLE */

I guess there's a reason for this behaviour: performance. However, I
think it should be always the second option.

Creating a new reservation is not a usual task, so the best effort
should prevail versus performance at creation time.

The underlying problem is in the sched plugin logic in SLURM v2.4

but not node count the node count is correctly figured out.

This change permits a user to get a zero size allocation by specifying
a task count of zero with no node count specification.

Without this fix, gang scheduling mode could start without creating
a list resulting in an assert when jobs are submitted.

This change permits a user to get a zero size allocation by specifying
a task count of zero with no node count specification.