Fix the bookmark system - we now have children being correctly spawned where they should!

Also, I am no longer seeing any issue with the child job spawning its own daemons - this appears to be fixed. We still don't reuse the existing daemons, however, but that will come.

This commit was SVN r12229.

orte/mca/rmaps/round_robin/rmaps_rr.c

@@ -231,11 +231,15 @@ static int map_app_by_slot(
             }
         }
 
-        /* if we still have procs to map, or this node is full, move to next node */
-        if (num_alloc < app->num_procs || ORTE_ERR_NODE_FULLY_USED == rc) {
-            cur_node_item = next;
+        /* we move on to the next node in all cases EXCEPT if we came
+         * out of the loop without having taken a full bite AND the
+         * node is NOT max'd out
+         *
+         */
+        if (i < (num_slots_to_take-1) && ORTE_ERR_NODE_FULLY_USED != rc) {
+            continue;
         }
-
+        cur_node_item = next;
     }
 
     return ORTE_SUCCESS;

orte/test/mpi/simple_spawn.c

@@ -1,4 +1,6 @@
 #include <stdio.h>
+#include <sys/types.h>
+
 #include <mpi.h>
 
 int main(int argc, char* argv[])
@@ -7,12 +9,14 @@ int main(int argc, char* argv[])
     MPI_Comm parent, child;
     int rank, size;
     char hostname[512];
+    pid_t pid;
 
     MPI_Init(NULL, NULL);
     MPI_Comm_get_parent(&parent);
     /* If we get COMM_NULL back, then we're the parent */
     if (MPI_COMM_NULL == parent) {
-        printf("Parent about to spawn!\n");
+        pid = getpid();
+        printf("Parent [pid %ld] about to spawn!\n", (long)pid);
         MPI_Comm_spawn(argv[0], MPI_ARGV_NULL, 1, MPI_INFO_NULL, 
                        0, MPI_COMM_WORLD, &child, MPI_ERRCODES_IGNORE);
         printf("Parent done with spawn\n");
@@ -27,7 +31,8 @@ int main(int argc, char* argv[])
         MPI_Comm_rank(MPI_COMM_WORLD, &rank);
         MPI_Comm_size(MPI_COMM_WORLD, &size);
         gethostname(hostname, 512);
-        printf("Hello from the child %d of %d on host %s\n", rank, size, hostname);
+        pid = getpid();
+        printf("Hello from the child %d of %d on host %s pid %ld\n", rank, size, hostname, (long)pid);
         MPI_Recv(&msg, 1, MPI_INT, 0, 1, parent, MPI_STATUS_IGNORE);
         printf("Child %d received msg: %d\n", rank, msg);
         MPI_Comm_disconnect(&parent);

orte/tools/orted/orted.c

@@ -365,6 +365,15 @@ int main(int argc, char *argv[])
         }
     }
 
+    /* output a message indicating we are alive, our name, and our pid
+     * for debugging purposes
+     */
+    if (orted_globals.debug_daemons) {
+        fprintf(stderr, "Daemon [%ld,%ld,%ld] checking in as pid %ld on host %s\n",
+                ORTE_NAME_ARGS(orte_process_info.my_name), (long)orte_process_info.pid,
+                orte_system_info.nodename);
+    }
+
     /* setup the thread lock and condition variables */
     OBJ_CONSTRUCT(&orted_globals.mutex, opal_mutex_t);
     OBJ_CONSTRUCT(&orted_globals.condition, opal_condition_t);