bgpd, lib: memory cleanups for valgrind, plus debug changes

Description:
    We use valgrind memcheck quite a bit to spot leaks in
    our work with bgpd. In order to eliminate false positives,
    we added code in the exit path to release the remaining
    allocated memory.

    Bgpd startup log message now includes pid.

Some little tweaks by Paul Jakma <paul.jakma@hpe.com>:

* bgp_mplsvpn.c: (str2prefix_rd) do the cleanup in common code at the end
  and goto it.

7 files changed, 96 insertions, 42 deletions

diff --git a/bgpd/bgp_aspath.c b/bgpd/bgp_aspath.c
index 0aec3ef1..9d49f343 100644
--- a/bgpd/bgp_aspath.c
+++ b/bgpd/bgp_aspath.c
@@ -98,6 +98,12 @@ assegment_data_new (int num)
   return (XMALLOC (MTYPE_AS_SEG_DATA, ASSEGMENT_DATA_SIZE (num, 1)));
 }
 
+static void
+assegment_data_free (as_t *asdata)
+{
+  XFREE (MTYPE_AS_SEG_DATA, asdata);
+}
+
 /* Get a new segment. Note that 0 is an allowed length,
  * and will result in a segment with no allocated data segment.
  * the caller should immediately assign data to the segment, as the segment
@@ -126,7 +132,7 @@ assegment_free (struct assegment *seg)
     return;
   
   if (seg->as)
-    XFREE (MTYPE_AS_SEG_DATA, seg->as);
+    assegment_data_free (seg->as);
   memset (seg, 0xfe, sizeof(struct assegment));
   XFREE (MTYPE_AS_SEG, seg);
   
@@ -194,13 +200,14 @@ assegment_prepend_asns (struct assegment *seg, as_t asnum, int num)
   if (num >= AS_SEGMENT_MAX)
     return seg; /* we don't do huge prepends */
   
-  newas = assegment_data_new (seg->length + num);
-
+  if ((newas = assegment_data_new (seg->length + num)) == NULL)
+    return seg;
+  
   for (i = 0; i < num; i++)
     newas[i] = asnum;
 
   memcpy (newas + num, seg->as, ASSEGMENT_DATA_SIZE (seg->length, 1));
-  XFREE (MTYPE_AS_SEG_DATA, seg->as);
+  assegment_data_free (seg->as);
   seg->as = newas;
   seg->length += num;
 
@@ -1879,6 +1886,7 @@ aspath_init (void)
 void
 aspath_finish (void)
 {
+  hash_clean (ashash, (void (*)(void *))aspath_free);
   hash_free (ashash);
   ashash = NULL;
   
diff --git a/bgpd/bgp_attr.c b/bgpd/bgp_attr.c
index 5c832edc..d413e5b2 100644
--- a/bgpd/bgp_attr.c
+++ b/bgpd/bgp_attr.c
@@ -203,6 +203,7 @@ cluster_init (void)
 static void
 cluster_finish (void)
 {
+  hash_clean (cluster_hash, (void (*)(void *))cluster_free);
   hash_free (cluster_hash);
   cluster_hash = NULL;
 }
@@ -279,6 +280,7 @@ transit_init (void)
 static void
 transit_finish (void)
 {
+  hash_clean (transit_hash, (void (*)(void *))transit_free);
   hash_free (transit_hash);
   transit_hash = NULL;
 }
@@ -452,9 +454,20 @@ attrhash_init (void)
   attrhash = hash_create (attrhash_key_make, attrhash_cmp);
 }
 
+/*
+ * special for hash_clean below
+ */
+static void
+attr_vfree (void *attr)
+{
+  bgp_attr_extra_free ((struct attr *)attr);
+  XFREE (MTYPE_ATTR, attr);
+}
+
 static void
 attrhash_finish (void)
 {
+  hash_clean(attrhash, attr_vfree);
   hash_free (attrhash);
   attrhash = NULL;
 }
diff --git a/bgpd/bgp_main.c b/bgpd/bgp_main.c
index cacff234..591a6f93 100644
--- a/bgpd/bgp_main.c
+++ b/bgpd/bgp_main.c
@@ -37,6 +37,7 @@ Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
 #include "plist.h"
 #include "stream.h"
 #include "vrf.h"
+#include "workqueue.h"
 
 #include "bgpd/bgpd.h"
 #include "bgpd/bgp_attr.h"
@@ -196,10 +197,12 @@ sigint (void)
 {
   zlog_notice ("Terminating on signal");
 
-  if (! retain_mode)
-    bgp_terminate ();
+  if (! retain_mode) 
+    {
+      bgp_terminate ();
+      zprivs_terminate (&bgpd_privs);
+    }
 
-  zprivs_terminate (&bgpd_privs);
   bgp_exit (0);
 }
 
@@ -234,7 +237,27 @@ bgp_exit (int status)
   for (ALL_LIST_ELEMENTS (bm->bgp, node, nnode, bgp))
     bgp_delete (bgp);
   list_free (bm->bgp);
-
+  bm->bgp = NULL;
+  
+  /*
+   * bgp_delete can re-allocate the process queues after they were
+   * deleted in bgp_terminate. delete them again.
+   *
+   * It might be better to ensure the RIBs (including static routes)
+   * are cleared by bgp_terminate() during its call to bgp_cleanup_routes(),
+   * which currently only deletes the kernel routes.
+   */
+  if (bm->process_main_queue)
+    {
+     work_queue_free (bm->process_main_queue);
+     bm->process_main_queue = NULL;
+    }
+  if (bm->process_rsclient_queue)
+    {
+      work_queue_free (bm->process_rsclient_queue);
+      bm->process_rsclient_queue = NULL;
+    }
+  
   /* reverse bgp_master_init */
   for (ALL_LIST_ELEMENTS_RO(bm->listen_sockets, node, socket))
     {
@@ -447,10 +470,11 @@ main (int argc, char **argv)
   vty_serv_sock (vty_addr, vty_port, BGP_VTYSH_PATH);
 
   /* Print banner. */
-  zlog_notice ("BGPd %s starting: vty@%d, bgp@%s:%d", QUAGGA_VERSION,
+  zlog_notice ("BGPd %s starting: vty@%d, bgp@%s:%d pid %d", QUAGGA_VERSION,
 	       vty_port, 
 	       (bm->address ? bm->address : "<all>"),
-	       bm->port);
+	       bm->port,
+	       getpid ());
 
   /* Start finite state machine, here we go! */
   while (thread_fetch (bm->master, &thread))
diff --git a/bgpd/bgp_mplsvpn.c b/bgpd/bgp_mplsvpn.c
index 8a1ed70e..a72d5ede 100644
--- a/bgpd/bgp_mplsvpn.c
+++ b/bgpd/bgp_mplsvpn.c
@@ -170,11 +170,12 @@ bgp_nlri_parse_vpnv4 (struct peer *peer, struct attr *attr,
 int
 str2prefix_rd (const char *str, struct prefix_rd *prd)
 {
-  int ret;
+  int ret; /* ret of called functions */
+  int lret; /* local ret, of this func */
   char *p;
   char *p2;
-  struct stream *s;
-  char *half;
+  struct stream *s = NULL;
+  char *half = NULL;
   struct in_addr addr;
 
   s = stream_new (8);
@@ -182,12 +183,13 @@ str2prefix_rd (const char *str, struct prefix_rd *prd)
   prd->family = AF_UNSPEC;
   prd->prefixlen = 64;
 
+  lret = 0;
   p = strchr (str, ':');
   if (! p)
-    return 0;
+    goto out;
 
   if (! all_digit (p + 1))
-    return 0;
+    goto out;
 
   half = XMALLOC (MTYPE_TMP, (p - str) + 1);
   memcpy (half, str, (p - str));
@@ -198,10 +200,8 @@ str2prefix_rd (const char *str, struct prefix_rd *prd)
   if (! p2)
     {
       if (! all_digit (half))
-	{
-	  XFREE (MTYPE_TMP, half);
-	  return 0;
-	}
+        goto out;
+      
       stream_putw (s, RD_TYPE_AS);
       stream_putw (s, atoi (half));
       stream_putl (s, atol (p + 1));
@@ -210,18 +210,21 @@ str2prefix_rd (const char *str, struct prefix_rd *prd)
     {
       ret = inet_aton (half, &addr);
       if (! ret)
-	{
-	  XFREE (MTYPE_TMP, half);
-	  return 0;
-	}
+        goto out;
+      
       stream_putw (s, RD_TYPE_IP);
       stream_put_in_addr (s, &addr);
       stream_putw (s, atol (p + 1));
     }
   memcpy (prd->val, s->data, 8);
-
-  XFREE(MTYPE_TMP, half);
-  return 1;
+  lret = 1;
+
+out:
+  if (s)
+    stream_free (s);
+  if (half)
+    XFREE(MTYPE_TMP, half);
+  return lret;
 }
 
 int
diff --git a/bgpd/bgp_network.c b/bgpd/bgp_network.c
index 8621854b..3c5e6c5d 100644
--- a/bgpd/bgp_network.c
+++ b/bgpd/bgp_network.c
@@ -276,6 +276,7 @@ bgp_bind (struct peer *peer)
 #ifdef SO_BINDTODEVICE
   int ret;
   struct ifreq ifreq;
+  int myerrno;
 
   if (! peer->ifname)
     return 0;
@@ -287,13 +288,15 @@ bgp_bind (struct peer *peer)
   
   ret = setsockopt (peer->fd, SOL_SOCKET, SO_BINDTODEVICE, 
 		    &ifreq, sizeof (ifreq));
-
+  myerrno = errno;
+  
   if (bgpd_privs.change (ZPRIVS_LOWER) )
     zlog_err ("bgp_bind: could not lower privs");
 
   if (ret < 0)
     {
-      zlog (peer->log, LOG_INFO, "bind to interface %s failed", peer->ifname);
+      zlog (peer->log, LOG_INFO, "bind to interface %s failed, errno=%d",
+            peer->ifname, myerrno);
       return ret;
     }
 #endif /* SO_BINDTODEVICE */
diff --git a/bgpd/bgp_nexthop.c b/bgpd/bgp_nexthop.c
index bb658afb..145a1d88 100644
--- a/bgpd/bgp_nexthop.c
+++ b/bgpd/bgp_nexthop.c
@@ -1439,29 +1439,29 @@ bgp_scan_init (void)
 void
 bgp_scan_finish (void)
 {
-  /* Only the current one needs to be reset. */
-  bgp_nexthop_cache_reset (bgp_nexthop_cache_table[AFI_IP]);
-
-  bgp_table_unlock (cache1_table[AFI_IP]);
+  if (cache1_table[AFI_IP])
+    bgp_table_unlock (cache1_table[AFI_IP]);
   cache1_table[AFI_IP] = NULL;
 
-  bgp_table_unlock (cache2_table[AFI_IP]);
+  if (cache2_table[AFI_IP])
+    bgp_table_unlock (cache2_table[AFI_IP]);
   cache2_table[AFI_IP] = NULL;
-
-  bgp_table_unlock (bgp_connected_table[AFI_IP]);
+  
+  if (bgp_connected_table[AFI_IP])
+    bgp_table_unlock (bgp_connected_table[AFI_IP]);
   bgp_connected_table[AFI_IP] = NULL;
 
 #ifdef HAVE_IPV6
-  /* Only the current one needs to be reset. */
-  bgp_nexthop_cache_reset (bgp_nexthop_cache_table[AFI_IP6]);
-
-  bgp_table_unlock (cache1_table[AFI_IP6]);
+  if (cache1_table[AFI_IP6])
+    bgp_table_unlock (cache1_table[AFI_IP6]);
   cache1_table[AFI_IP6] = NULL;
 
-  bgp_table_unlock (cache2_table[AFI_IP6]);
+  if (cache2_table[AFI_IP6])
+    bgp_table_unlock (cache2_table[AFI_IP6]);
   cache2_table[AFI_IP6] = NULL;
 
-  bgp_table_unlock (bgp_connected_table[AFI_IP6]);
+  if (bgp_connected_table[AFI_IP6])
+    bgp_table_unlock (bgp_connected_table[AFI_IP6]);
   bgp_connected_table[AFI_IP6] = NULL;
 #endif /* HAVE_IPV6 */
 }
diff --git a/bgpd/bgpd.c b/bgpd/bgpd.c
index 3fdd9ff6..90e77f21 100644
--- a/bgpd/bgpd.c
+++ b/bgpd/bgpd.c
@@ -737,6 +737,9 @@ peer_free (struct peer *peer)
   if (peer->clear_node_queue)
     work_queue_free (peer->clear_node_queue);
   
+  if (peer->notify.data)
+    XFREE(MTYPE_TMP, peer->notify.data);
+  
   bgp_sync_delete (peer);
   memset (peer, 0, sizeof (struct peer));
   
@@ -1988,7 +1991,7 @@ bgp_create (as_t *as, const char *name)
 struct bgp *
 bgp_get_default (void)
 {
-  if (bm->bgp->head)
+  if (bm && bm->bgp && bm->bgp->head)
     return (listgetdata (listhead (bm->bgp)));
   return NULL;
 }