Route sources have an explicit owner

This commit prevents use-after-free of routes belonging to protocols
which have been already destroyed, delaying also all the protocols'
shutdown until all of their routes have been finally propagated through
all the pipes down to the appropriate exports.

The use-after-free was somehow hypothetic yet theoretically possible in
rare conditions, when one BGP protocol authors a lot of routes and the
user deletes that protocol by reconfiguring in the same time as next hop
update is requested, causing rte_better() to be called on a
not-yet-pruned network prefix while the owner protocol has been already
freed.

In parallel execution environments, this would happen an inter-thread
use-after-free, causing possible heisenbugs or other nasty problems.

1 files changed, 5 insertions, 0 deletions

diff --git a/proto/bgp/packets.c b/proto/bgp/packets.c
index 647551e5..d2d5b174 100644
--- a/proto/bgp/packets.c
+++ b/proto/bgp/packets.c
@@ -1339,6 +1339,8 @@ bgp_rte_update(struct bgp_parse_state *s, net_addr *n, u32 path_id, rta *a0)
 {
   if (path_id != s->last_id)
   {
+    rt_unlock_source(s->last_src);
+
     s->last_src = rt_get_source(&s->proto->p, path_id);
     s->last_id = path_id;
 
@@ -2421,6 +2423,7 @@ bgp_decode_nlri(struct bgp_parse_state *s, u32 afi, byte *nlri, uint len, ea_lis
 
   s->last_id = 0;
   s->last_src = s->proto->p.main_source;
+  rt_lock_source(s->last_src);
 
   /*
    * IPv4 BGP and MP-BGP may be used together in one update, therefore we do not
@@ -2451,6 +2454,8 @@ bgp_decode_nlri(struct bgp_parse_state *s, u32 afi, byte *nlri, uint len, ea_lis
 
   rta_free(s->cached_rta);
   s->cached_rta = NULL;
+
+  rt_unlock_source(s->last_src);
 }
 
 static void