When the MX MTL complete a request, force a return from the progress function.

Decrease the latency by about 0.3 microseconds.

This commit was SVN r12454.

ompi/mca/mtl/mx/mtl_mx.c

@@ -214,6 +214,7 @@ int ompi_mtl_mx_progress( void ) {
                         MPI_ERR_INTERN;
                 }
                 mtl_mx_request->super.completion_callback(&mtl_mx_request->super);
+                return completed;
             }
             if(OMPI_MTL_MX_IRECV == mtl_mx_request->type) { 
                 
@@ -242,6 +243,7 @@ int ompi_mtl_mx_progress( void ) {
                         MPI_ERR_INTERN;
                 }
                 mtl_mx_request->super.completion_callback(&mtl_mx_request->super);
+                return completed;
             }
             
         } else {