Protect the cpus-per-rank MCA param registration so that --without-hwloc will build

This commit was SVN r28232.

orte/mca/rmaps/base/rmaps_base_open.c

@@ -281,6 +281,7 @@ int orte_rmaps_base_open(void)
         ORTE_SET_RANKING_DIRECTIVE(orte_rmaps_base.ranking, ORTE_RANKING_GIVEN);
     }
 
+#if OPAL_HAVE_HWLOC
     /* #cpus/rank to use */
     param = mca_base_param_reg_int_name("rmaps", "base_cpus_per_proc",
                                         "Number of cpus to use for each rank [1-2**15 (default=1)]",
@@ -299,6 +300,9 @@ int orte_rmaps_base_open(void)
             OPAL_SET_BINDING_POLICY(opal_hwloc_binding_policy, OPAL_BIND_TO_CORE);
         }
     }
+#else
+    orte_rmaps_base.cpus_per_rank = 1;
+#endif
 
     /* Should we schedule on the local node or not? */
     mca_base_param_reg_int_name("rmaps", "base_no_schedule_local",