fix(rtable): Prevent direct send hairpins

If the current node is listed as an endpoint for a message type
that this process would add to the route table a hair-pin loop
back to the node can happen. This isn't desireable and this
change will prevent the current node from being added as a
recipient endpoint to any route table entry.

Signed-off-by: E. Scott Daniels <daniels@research.att.com>
Change-Id: I5a1b2bdaad2eab499ae5a1c7430238c1da2f3256

Add latency call/receive test support

Signed-off-by: E. Scott Daniels <daniels@research.att.com>
Change-Id: I70be670d063adfdf6505431a4a2ce72e846df0a4

Actual mods to implement hairpin loop avodiance

Signed-off-by: E. Scott Daniels <daniels@research.att.com>
Change-Id: Ieead4fbf69ade58db1f37d1949cf0f03683c64de
Signed-off-by: E. Scott Daniels <daniels@research.att.com>
diff --git a/test/app_test/Makefile b/test/app_test/Makefile
index 07748ee..67acb9f 100644
--- a/test/app_test/Makefile
+++ b/test/app_test/Makefile
@@ -55,6 +55,9 @@
 mt_receiver: receiver.c
 	gcc -I $${C_INCLUDE_PATH:-.} -DMTC $< -g -o $@ -lrmr_nng -lnng -lpthread -lm
 
+lreceiver: lreceiver.c
+	gcc -I $${C_INCLUDE_PATH:-.} $< -g -o $@ -lrmr_nng -lnng -lpthread -lm
+
 sender_nano: sender.c
 	gcc -I $${C_INCLUDE_PATH:-.} $< -g -o $@ -lrmr -lnanomsg -lpthread -lm
 
@@ -64,6 +67,10 @@
 caller: caller.c
 	gcc -I $${C_INCLUDE_PATH:-.} $< -g -o $@ -lrmr_nng -lnng -lpthread -lm
 
+lcaller: lcaller.c
+	gcc -I $${C_INCLUDE_PATH:-.} $< -g -o $@  -lrmr_nng -lnng -lpthread -lm
+
+
 
 # clean removes intermediates; nuke removes everything that can be built
 .PHONY: clean nuke