Remove a race condition in the threaded mode. As a callback is allowed

to modify the callback array (add or remove), make sure we don't call
the same callback twice if it get remove in another thread.

This commit was SVN r18608.

opal/runtime/opal_progress.c

@@ -74,6 +74,15 @@ static int32_t num_event_users = 0;
 static int debug_output = -1;
 #endif
 
+/**
+ * Fake callback used for threading purpose when one thread
+ * progesses callbacks while another unregister somes. The root
+ * of the problem is that we allow modifications of the callback
+ * array directly from the callbacks themselves. Now if
+ * writing a pointer is atomic, we should not have any more
+ * problems.
+ */
+static int fake_cb(void) { return 0; };
 
 /* init the progress engine - called from orte_init */
 int
@@ -308,7 +317,7 @@ opal_progress_set_event_poll_rate(int polltime)
 int
 opal_progress_register(opal_progress_callback_t cb)
 {
-    int ret = OPAL_SUCCESS;
+    int ret = OPAL_SUCCESS, index;
 
 #if OMPI_HAVE_THREAD_SUPPORT
     opal_atomic_lock(&progress_lock);
@@ -322,6 +331,10 @@ opal_progress_register(opal_progress_callback_t cb)
             ret = OPAL_ERR_TEMP_OUT_OF_RESOURCE;
             goto cleanup;
         }
+        /* registering fake callbacks to fill callbacks[] */
+        for( index = callbacks_len + 1 ;  index < callbacks_size + 4 ; index++) {
+            tmp[index] = &fake_cb;
+        }
 
         callbacks = tmp;
         callbacks_size += 4;
@@ -350,7 +363,7 @@ opal_progress_unregister(opal_progress_callback_t cb)
 
     for (i = 0 ; i < callbacks_len ; ++i) {
         if (cb == callbacks[i]) {
-            callbacks[i] = NULL;
+            callbacks[i] = &fake_cb;
             ret = OPAL_SUCCESS;
             break;
         }
@@ -368,7 +381,7 @@ opal_progress_unregister(opal_progress_callback_t cb)
                 callbacks[i] = callbacks[i + 1];
             }
         }
-        callbacks[callbacks_len - 1] = NULL;
+        callbacks[callbacks_len - 1] = &fake_cb;
         callbacks_len--;
     }