| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Wed, 1 Jul 2009 11:04:07 +0000
From: Jarek Poplawski <jarkao2@...il.com>
To: Paweł Staszewski <pstaszewski@...are.pl>
Cc: Linux Network Development list <netdev@...r.kernel.org>,
Robert Olsson <robert@...ur.slu.se>
Subject: Re: [PATCH net-2.6] Re: rib_trie / Fix inflate_threshold_root.
Now=15 size=11 bits
On Wed, Jul 01, 2009 at 10:13:33AM +0000, Jarek Poplawski wrote:
> On Wed, Jul 01, 2009 at 11:43:04AM +0200, Paweł Staszewski wrote:
> ...
> > Yes on on previous patches there was / no warnings / no oopses or lockups
> >
> > But now i apply this patch and i make more testing.
> > First boot with start of bgpd and - traffic is not forwarded
> > So i start to search and make only some routes (static without bgpd)
> > thru this host
> > And all is working for this host when i make all by static routes.
> >
> > So i change a little my bgp configuration and make default route to only
> > one of my iBGP peers and start bgpd process
> > All is working and what is weird is number of routes in kernel table.
> > Kernel is learning routes from bgpd but very slowly - really very slowly.
>
> Pawel, this is really very helpful! So, this is (probably) only about
> timing, not wrong memory freeing. On the other hand this test was only
> for inserts. Btw., if you didn't start the second test, you can skip
> it. I have to rethink this.
So, after your findings I'm about to recommend sending to -stable
3 patches from net-2.6, with additional lowering of threshold_root
settings, but it would be nice if you could give it a try with
CONFIG_PREEMPT instead of CONFIG_PREEMPT_NONE (if it doesn't break
your other apps!) It is expected to work this time...;-) Maybe a
bit slower.
Thanks,
Jarek P.
--------> (all-in-one preempt fixes to apply with vanilla 2.6.29.x)
diff -Nurp a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
--- a/net/ipv4/fib_trie.c 2009-07-01 06:17:08.000000000 +0000
+++ b/net/ipv4/fib_trie.c 2009-07-01 10:43:44.000000000 +0000
@@ -123,6 +123,7 @@ struct tnode {
union {
struct rcu_head rcu;
struct work_struct work;
+ struct tnode *tnode_free;
};
struct node *child[0];
};
@@ -161,6 +162,8 @@ static void tnode_put_child_reorg(struct
static struct node *resize(struct trie *t, struct tnode *tn);
static struct tnode *inflate(struct trie *t, struct tnode *tn);
static struct tnode *halve(struct trie *t, struct tnode *tn);
+/* tnodes to free after resize(); protected by RTNL */
+static struct tnode *tnode_free_head;
static struct kmem_cache *fn_alias_kmem __read_mostly;
static struct kmem_cache *trie_leaf_kmem __read_mostly;
@@ -313,8 +316,8 @@ static inline void check_tnode(const str
static const int halve_threshold = 25;
static const int inflate_threshold = 50;
-static const int halve_threshold_root = 8;
-static const int inflate_threshold_root = 15;
+static const int halve_threshold_root = 15;
+static const int inflate_threshold_root = 25;
static void __alias_free_mem(struct rcu_head *head)
@@ -385,6 +388,24 @@ static inline void tnode_free(struct tno
call_rcu(&tn->rcu, __tnode_free_rcu);
}
+static void tnode_free_safe(struct tnode *tn)
+{
+ BUG_ON(IS_LEAF(tn));
+ tn->tnode_free = tnode_free_head;
+ tnode_free_head = tn;
+}
+
+static void tnode_free_flush(void)
+{
+ struct tnode *tn;
+
+ while ((tn = tnode_free_head)) {
+ tnode_free_head = tn->tnode_free;
+ tn->tnode_free = NULL;
+ tnode_free(tn);
+ }
+}
+
static struct leaf *leaf_new(void)
{
struct leaf *l = kmem_cache_alloc(trie_leaf_kmem, GFP_KERNEL);
@@ -495,7 +516,7 @@ static struct node *resize(struct trie *
/* No children */
if (tn->empty_children == tnode_child_length(tn)) {
- tnode_free(tn);
+ tnode_free_safe(tn);
return NULL;
}
/* One child */
@@ -509,7 +530,7 @@ static struct node *resize(struct trie *
/* compress one level */
node_set_parent(n, NULL);
- tnode_free(tn);
+ tnode_free_safe(tn);
return n;
}
/*
@@ -670,7 +691,7 @@ static struct node *resize(struct trie *
/* compress one level */
node_set_parent(n, NULL);
- tnode_free(tn);
+ tnode_free_safe(tn);
return n;
}
@@ -756,7 +777,7 @@ static struct tnode *inflate(struct trie
put_child(t, tn, 2*i, inode->child[0]);
put_child(t, tn, 2*i+1, inode->child[1]);
- tnode_free(inode);
+ tnode_free_safe(inode);
continue;
}
@@ -801,9 +822,9 @@ static struct tnode *inflate(struct trie
put_child(t, tn, 2*i, resize(t, left));
put_child(t, tn, 2*i+1, resize(t, right));
- tnode_free(inode);
+ tnode_free_safe(inode);
}
- tnode_free(oldtnode);
+ tnode_free_safe(oldtnode);
return tn;
nomem:
{
@@ -885,7 +906,7 @@ static struct tnode *halve(struct trie *
put_child(t, newBinNode, 1, right);
put_child(t, tn, i/2, resize(t, newBinNode));
}
- tnode_free(oldtnode);
+ tnode_free_safe(oldtnode);
return tn;
nomem:
{
@@ -983,12 +1004,14 @@ fib_find_node(struct trie *t, u32 key)
return NULL;
}
-static struct node *trie_rebalance(struct trie *t, struct tnode *tn)
+static void trie_rebalance(struct trie *t, struct tnode *tn)
{
int wasfull;
- t_key cindex, key = tn->key;
+ t_key cindex, key;
struct tnode *tp;
+ key = tn->key;
+
while (tn != NULL && (tp = node_parent((struct node *)tn)) != NULL) {
cindex = tkey_extract_bits(key, tp->pos, tp->bits);
wasfull = tnode_full(tp, tnode_get_child(tp, cindex));
@@ -999,6 +1022,10 @@ static struct node *trie_rebalance(struc
tp = node_parent((struct node *) tn);
if (!tp)
+ rcu_assign_pointer(t->trie, (struct node *)tn);
+
+ tnode_free_flush();
+ if (!tp)
break;
tn = tp;
}
@@ -1007,7 +1034,10 @@ static struct node *trie_rebalance(struc
if (IS_TNODE(tn))
tn = (struct tnode *)resize(t, (struct tnode *)tn);
- return (struct node *)tn;
+ rcu_assign_pointer(t->trie, (struct node *)tn);
+ tnode_free_flush();
+
+ return;
}
/* only used from updater-side */
@@ -1155,7 +1185,7 @@ static struct list_head *fib_insert_node
/* Rebalance the trie */
- rcu_assign_pointer(t->trie, trie_rebalance(t, tp));
+ trie_rebalance(t, tp);
done:
return fa_head;
}
@@ -1575,7 +1605,7 @@ static void trie_leaf_remove(struct trie
if (tp) {
t_key cindex = tkey_extract_bits(l->key, tp->pos, tp->bits);
put_child(t, (struct tnode *)tp, cindex, NULL);
- rcu_assign_pointer(t->trie, trie_rebalance(t, tp));
+ trie_rebalance(t, tp);
} else
rcu_assign_pointer(t->trie, NULL);
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists