diff --git a/libfdproto/rt_data.c b/libfdproto/rt_data.c
new file mode 100644
index 0000000..f205ef3
--- /dev/null
+++ b/libfdproto/rt_data.c
@@ -0,0 +1,338 @@
+/*********************************************************************************************************
+* Software License Agreement (BSD License)                                                               *
+* Author: Sebastien Decugis <sdecugis@freediameter.net>							 *
+*													 *
+* Copyright (c) 2013, WIDE Project and NICT								 *
+* All rights reserved.											 *
+* 													 *
+* Redistribution and use of this software in source and binary forms, with or without modification, are  *
+* permitted provided that the following conditions are met:						 *
+* 													 *
+* * Redistributions of source code must retain the above 						 *
+*   copyright notice, this list of conditions and the 							 *
+*   following disclaimer.										 *
+*    													 *
+* * Redistributions in binary form must reproduce the above 						 *
+*   copyright notice, this list of conditions and the 							 *
+*   following disclaimer in the documentation and/or other						 *
+*   materials provided with the distribution.								 *
+* 													 *
+* * Neither the name of the WIDE Project or NICT nor the 						 *
+*   names of its contributors may be used to endorse or 						 *
+*   promote products derived from this software without 						 *
+*   specific prior written permission of WIDE Project and 						 *
+*   NICT.												 *
+* 													 *
+* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED *
+* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A *
+* PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR *
+* ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT 	 *
+* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 	 *
+* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR *
+* TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF   *
+* ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.								 *
+*********************************************************************************************************/
+
+/* Routing module helpers.
+ * 
+ * This file provides support for the rt_data structure manipulation.
+ */
+
+#include "fdproto-internal.h"
+
+/* Structure that contains the routing data for a message */
+struct rt_data {
+	int		extracted;	/* if 0, candidates is ordered by diamid, otherwise the order is unspecified. This also counts the number of times the message was (re-)sent, as a side effect */
+	struct fd_list	candidates;	/* All the candidates. Items are struct rtd_candidate. */
+	struct fd_list	errors;		/* All errors received from other peers for this message */
+};
+
+/* Items of the errors list */
+struct rtd_error {
+	struct fd_list	chain;	/* link in the list, ordered by nexthop (fd_os_cmp) */
+	DiamId_t 	nexthop;/* the peer the message was sent to */
+	size_t		nexthoplen; /* cached string length */
+	DiamId_t	erh;	/* the origin of the error */
+	size_t		erhlen; /* cached string length */
+	uint32_t	code;	/* the error code */
+};
+
+/* Create a new structure to store routing data */
+int  fd_rtd_init(struct rt_data ** rtd)
+{
+	struct rt_data *new;
+	TRACE_ENTRY("%p", rtd);
+	CHECK_PARAMS(rtd);
+	
+	/* Alloc the structure */
+	CHECK_MALLOC( new = malloc(sizeof(struct rt_data)) );
+	memset(new, 0, sizeof(struct rt_data) );
+	fd_list_init(&new->candidates, new);
+	fd_list_init(&new->errors, new);
+	
+	*rtd = new;
+	return 0;
+}
+
+/* Destroy the routing data */
+void fd_rtd_free(struct rt_data ** rtd)
+{
+	struct rt_data *old;
+	
+	TRACE_ENTRY("%p", rtd);
+	CHECK_PARAMS_DO(rtd, return );
+	
+	old = *rtd;
+	*rtd = NULL;
+	
+	while (!FD_IS_LIST_EMPTY(&old->candidates)) {
+		struct rtd_candidate * c = (struct rtd_candidate *) old->candidates.next;
+		
+		fd_list_unlink(&c->chain);
+		free(c->diamid);
+		free(c->realm);
+		free(c);
+	}
+	
+	while (!FD_IS_LIST_EMPTY(&old->errors)) {
+		struct rtd_error * c = (struct rtd_error *) old->errors.next;
+		
+		fd_list_unlink(&c->chain);
+		free(c->nexthop);
+		free(c->erh);
+		free(c);
+	}
+	
+	free(old);
+	
+	return;
+}
+
+/* Add a peer to the candidates list. The source is our local peer list, so no need to care for the case here. */
+int  fd_rtd_candidate_add(struct rt_data * rtd, DiamId_t peerid, size_t peeridlen, DiamId_t realm, size_t realmlen)
+{
+	struct fd_list * prev;
+	struct rtd_candidate * new;
+	
+	TRACE_ENTRY("%p %p %zd %p %zd", rtd, peerid, peeridlen, realm, realmlen);
+	CHECK_PARAMS(rtd && peerid && peeridlen);
+	
+	/* Since the peers are ordered when they are added (fd_g_activ_peers) we search for the position from the end -- this should be efficient */
+	for (prev = rtd->candidates.prev; prev != &rtd->candidates; prev = prev->prev) {
+		struct rtd_candidate * cp = (struct rtd_candidate *) prev;
+		int cmp = fd_os_cmp(peerid, peeridlen, cp->diamid, cp->diamidlen);
+		if (cmp > 0)
+			break;
+		if (cmp == 0)
+			/* The candidate is already in the list */
+			return 0;
+	}
+	
+	/* Create the new entry */
+	CHECK_MALLOC( new = malloc(sizeof(struct rtd_candidate)) );
+	memset(new, 0, sizeof(struct rtd_candidate) );
+	fd_list_init(&new->chain, new);
+	CHECK_MALLOC( new->diamid = os0dup(peerid, peeridlen) )
+	new->diamidlen = peeridlen;
+	if (realm) {
+		CHECK_MALLOC( new->realm = os0dup(realm, realmlen) )
+		new->realmlen = realmlen;
+	}
+	
+	/* insert in the list at the correct position */
+	fd_list_insert_after(prev, &new->chain);
+	
+	return 0;
+}
+
+/* Remove a peer from the candidates (if it is found). Case insensitive search since the names are received from other peers */
+void fd_rtd_candidate_del(struct rt_data * rtd, uint8_t * id, size_t idsz)
+{
+	struct fd_list * li;
+	
+	TRACE_ENTRY("%p %p %zd", rtd, id, idsz);
+	CHECK_PARAMS_DO( rtd && id && idsz, return );
+	
+	if (!fd_os_is_valid_DiameterIdentity(id, idsz))
+		/* it cannot be in the list */
+		return;
+	
+	for (li = rtd->candidates.next; li != &rtd->candidates; li = li->next) {
+		struct rtd_candidate * c = (struct rtd_candidate *) li;
+		int cont;
+		int cmp = fd_os_almostcasesrch(id, idsz, c->diamid, c->diamidlen, &cont);
+		
+		if (!cmp) {
+			/* Found it! Remove it */
+			fd_list_unlink(&c->chain);
+			free(c->diamid);
+			free(c->realm);
+			free(c);
+			break;
+		}
+		
+		if (cont)
+			continue;
+		
+		/* The list is guaranteed to be ordered only if not extracted */
+		if (! rtd->extracted)
+			break;
+	}
+	
+	return;
+}
+
+/* If a peer returned a protocol error for this message, save it so that we don't try to send it there again.
+ Case insensitive search since the names are received from other peers*/
+int  fd_rtd_error_add(struct rt_data * rtd, DiamId_t sentto, size_t senttolen, uint8_t * origin, size_t originsz, uint32_t rcode, struct fd_list ** candidates, int * sendingattemtps)
+{
+	struct fd_list * li;
+	int match = 0;
+	
+	TRACE_ENTRY("%p %p %zd %p %zd %u %p %p", rtd, sentto, senttolen, origin, originsz, rcode, candidates, sendingattemtps);
+	CHECK_PARAMS( rtd && sentto && senttolen ); /* origin may be NULL */
+	
+	/* First add the new error entry */
+	for (li = rtd->errors.next; li != &rtd->errors; li = li->next) {
+		struct rtd_error * e = (struct rtd_error *) li;
+		int cmp = fd_os_cmp(sentto, senttolen, e->nexthop, e->nexthoplen);
+		if (cmp > 0)
+			continue;
+		if (!cmp)
+			match = 1;
+		break;
+	}
+	
+	/* If we already had this entry, we should not have sent the message again to this peer... anyway, let's close our eyes. */
+	/* in the normal case, we save the error */
+	if (!match) {
+		/* Add a new entry in the error list */
+		struct rtd_error * new;
+		CHECK_MALLOC( new = malloc(sizeof(struct rtd_error)) );
+		memset(new, 0, sizeof(struct rtd_error));
+		fd_list_init(&new->chain, NULL);
+
+		CHECK_MALLOC(new->nexthop = os0dup(sentto, senttolen));
+		new->nexthoplen = senttolen;
+		
+		if (origin) {
+			if (!originsz) {
+				originsz=strlen((char *)origin);
+			} else {
+				if (!fd_os_is_valid_DiameterIdentity(origin, originsz)){
+					TRACE_DEBUG(FULL, "Received error %d from peer with invalid Origin-Host AVP, not saved", rcode);
+					origin = NULL;
+					goto after_origin;
+				}
+			}
+			CHECK_MALLOC( new->erh = (DiamId_t)os0dup(origin, originsz) );
+			new->erhlen = originsz;
+		}
+after_origin:
+		new->code = rcode;
+		fd_list_insert_before(li, &new->chain);
+	}
+	
+	/* Finally, remove this (these) peers from the candidate list */
+	fd_rtd_candidate_del(rtd, (os0_t)sentto, senttolen);
+	if (origin)
+		fd_rtd_candidate_del(rtd, origin, originsz);
+	
+	if (candidates)
+		*candidates = &rtd->candidates;
+	
+	if (sendingattemtps)
+		*sendingattemtps = rtd->extracted;
+	
+	/* Done! */
+	return 0;
+}
+
+/* Only retrieve the number of times this message has been processed by the routing-out mechanism (i.e. number of times it was failed over) */
+int  fd_rtd_get_nb_attempts(struct rt_data * rtd, int * sendingattemtps)
+{
+	TRACE_ENTRY("%p %p", rtd, sendingattemtps);
+	CHECK_PARAMS( rtd && sendingattemtps );
+	
+	*sendingattemtps = rtd->extracted;
+	
+	/* Done! */
+	return 0;
+}
+
+/* Extract the list of valid candidates, and initialize their scores */
+void fd_rtd_candidate_extract(struct rt_data * rtd, struct fd_list ** candidates, int ini_score)
+{
+	struct fd_list * li;
+	
+	TRACE_ENTRY("%p %p", rtd, candidates);
+	CHECK_PARAMS_DO( candidates, return );
+	CHECK_PARAMS_DO( rtd, { *candidates = NULL; return; } );
+	
+	*candidates = &rtd->candidates;
+	
+	/* Reset all scores to INITIAL score */
+	for (li = rtd->candidates.next; li != &rtd->candidates; li = li->next) {
+		struct rtd_candidate * c = (struct rtd_candidate *) li;
+		c->score = ini_score;
+	}
+	
+	rtd->extracted += 1;
+	return;
+}
+
+/* Reorder the list of peers. If several peer have the same highest score, they are randomized. */
+int  fd_rtd_candidate_reorder(struct fd_list * candidates)
+{
+	struct fd_list unordered = FD_LIST_INITIALIZER(unordered), *li;
+	struct fd_list highest = FD_LIST_INITIALIZER(highest);
+	int hs = -1;
+	
+	TRACE_ENTRY("%p", candidates);
+	CHECK_PARAMS( candidates );
+	
+	/* First, move all items from candidates to the undordered list */
+	fd_list_move_end(&unordered, candidates);
+	
+	/* Now extract each element from unordered and add it back to list ordered by score */
+	while (!FD_IS_LIST_EMPTY(&unordered)) {
+		struct rtd_candidate * c = (struct rtd_candidate *) unordered.next;
+		
+		fd_list_unlink(&c->chain);
+		
+		/* If this candidate has a higher score than the previous ones */
+		if (c->score > hs) {
+			/* Then we move the previous high score items at end of the list */
+			fd_list_move_end(candidates, &highest);
+			
+			/* And the new high score is set */
+			hs = c->score;
+		}
+		
+		/* If this candidate equals the higher score, add it into highest list at a random place */
+		if (c->score == hs) {
+			if (rand() & 1) {
+				fd_list_insert_after(&highest, &c->chain);
+			} else {
+				fd_list_insert_before(&highest, &c->chain);
+			}
+		/* Otherwise, insert at normal place in the list */
+		} else {
+			/* Find the position in ordered candidates list */
+			for (li = candidates->next; li != candidates; li = li->next) {
+				struct rtd_candidate * cnext = (struct rtd_candidate *) li;
+				if (cnext->score >= c->score)
+					break;
+			}
+
+			/* Add the element there */
+			fd_list_insert_before(li, &c->chain);
+		}
+	}
+	
+	/* Now simply move back all the "highest" candidates at the end of the list */
+	fd_list_move_end(candidates, &highest);
+	
+	return 0;
+}
+