bgpd: fix incorrect keepalive timer evaluation

Incorrect check for sentinel value effectively caused peers to sometimes
use the keepalive timer value of other peers, which sometimes led to
hold timer expiry.

Signed-off-by: Quentin Young <qlyoung@qlyoung.net>

bgpd/bgp_keepalives.c

@@ -119,7 +119,7 @@ static void peer_process(struct hash_backet *hb, void *arg)
 	}
 
 	/* if calculated next update for this peer < current delay, use it */
-	if (next_update->tv_sec <= 0 || timercmp(&diff, next_update, <))
+	if (next_update->tv_sec < 0 || timercmp(&diff, next_update, <))
 		*next_update = diff;
 }