Every table has its own netindex now

It's obnoxiously slow to walk over sparse tables.

diff --git a/nest/rt-table.c b/nest/rt-table.c
index b3226a6badbb1f8b83c552610ae6e80d30a98b26..113cb1c2070a5ec689a0cb48e7f4a2de8546c9cc 100644 (file)
--- a/nest/rt-table.c
+++ b/nest/rt-table.c
@@ -122,7 +122,6 @@ pool *rt_table_pool;
 list routing_tables;
 list deleted_routing_tables;
 
-netindex_hash *rt_global_netindex_hash[NET_MAX];
 #define RT_INITIAL_ROUTES_BLOCK_SIZE   128
 
 struct rt_cork rt_cork;
@@ -3070,7 +3069,7 @@ rt_setup(pool *pp, struct rtable_config *cf)
   if (t->id >= rtable_max_id)
     rtable_max_id = t->id + 1;
 
-  t->netindex = rt_global_netindex_hash[cf->addr_type];
+  t->netindex = netindex_hash_new(birdloop_pool(t->loop), birdloop_event_list(t->loop), cf->addr_type);
   atomic_store_explicit(&t->routes, mb_allocz(p, RT_INITIAL_ROUTES_BLOCK_SIZE * sizeof(net)), memory_order_relaxed);
   atomic_store_explicit(&t->routes_block_size, RT_INITIAL_ROUTES_BLOCK_SIZE, memory_order_relaxed);
 
@@ -3218,9 +3217,6 @@ rt_init(void)
   rt_cork.run = (event) { .hook = rt_cork_release_hook };
   idm_init(&rtable_idm, rt_table_pool, 256);
 
-  for (uint i=1; i<NET_MAX; i++)
-    rt_global_netindex_hash[i] = netindex_hash_new(rt_table_pool, &global_event_list, i);
-
   ea_register_init(&ea_roa_aggregated);
 }
 
@@ -4323,6 +4319,14 @@ rt_unlock_table_priv(struct rtable_private *r, const char *file, uint line)
     ev_send_loop(r->loop, ev_new_init(r->rp, rt_shutdown, r));
 }
 
+static void
+rt_shutdown_finished(void *tab_)
+{
+  rtable *t = tab_;
+  RT_LOCK(t, tab);
+  birdloop_stop_self(t->loop, rt_delete, t);
+}
+
 static void
 rt_shutdown(void *tab_)
 {
@@ -4342,7 +4346,9 @@ rt_shutdown(void *tab_)
   rt_exporter_shutdown(&tab->export_best, NULL);
   rt_exporter_shutdown(&tab->export_all, NULL);
 
-  birdloop_stop_self(t->loop, rt_delete, t);
+  netindex_hash_delete(tab->netindex,
+      ev_new_init(tab->rp, rt_shutdown_finished, tab),
+      birdloop_event_list(tab->loop));
 }
 
 static void