[PATCH net v3] Use consistent ct id hash calculation

[PATCH net v3] Use consistent ct id hash calculation

[Dirk Morris]

Change ct id hash calculation to only use invariants.

Currently the ct id hash calculation is based on some fields that can
change in the lifetime on a conntrack entry in some corner cases. The
current hash uses the whole tuple which contains an hlist pointer
which will change when the conntrack is placed on the dying list
resulting in a ct id change.

This patch also removes the reply-side tuple and extension pointer
from the hash calculation so that the ct id will will not change from
initialization until confirmation.

Fixes: 3c79107631db1f7 ("netfilter: ctnetlink: don't use conntrack/expect object addresses as id")
Signed-off-by: Dirk Morris <dmorris@metaloft.com>
---
  nf_conntrack_core.c |   16 ++++++++--------
  1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index ab73c5f..f86625f 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -302,13 +302,12 @@ EXPORT_SYMBOL_GPL(nf_ct_invert_tuple);
   * table location, we assume id gets exposed to userspace.
   *
   * Following nf_conn items do not change throughout lifetime
- * of the nf_conn after it has been committed to main hash table:
+ * of the nf_conn:
   *
   * 1. nf_conn address
- * 2. nf_conn->ext address
- * 3. nf_conn->master address (normally NULL)
- * 4. tuple
- * 5. the associated net namespace
+ * 2. nf_conn->master address (normally NULL)
+ * 3. the associated net namespace
+ * 4. the original direction tuple
   */
  u32 nf_ct_get_id(const struct nf_conn *ct)
  {
@@ -318,9 +317,10 @@ u32 nf_ct_get_id(const struct nf_conn *ct)
  	net_get_random_once(&ct_id_seed, sizeof(ct_id_seed));
  
  	a = (unsigned long)ct;
-	b = (unsigned long)ct->master ^ net_hash_mix(nf_ct_net(ct));
-	c = (unsigned long)ct->ext;
-	d = (unsigned long)siphash(&ct->tuplehash, sizeof(ct->tuplehash),
+	b = (unsigned long)ct->master;
+	c = (unsigned long)nf_ct_net(ct);
+	d = (unsigned long)siphash(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple,
+				   sizeof(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple),
  				   &ct_id_seed);
  #ifdef CONFIG_64BIT
  	return siphash_4u64((u64)a, (u64)b, (u64)c, (u64)d, &ct_id_seed);

Re: [PATCH net v3] Use consistent ct id hash calculation

[Florian Westphal]

Dirk Morris <dmorris@metaloft.com> wrote:
> Change ct id hash calculation to only use invariants.
> 
> Currently the ct id hash calculation is based on some fields that can
> change in the lifetime on a conntrack entry in some corner cases. The
> current hash uses the whole tuple which contains an hlist pointer
> which will change when the conntrack is placed on the dying list
> resulting in a ct id change.
> 
> This patch also removes the reply-side tuple and extension pointer
> from the hash calculation so that the ct id will will not change from
> initialization until confirmation.
> 
> Fixes: 3c79107631db1f7 ("netfilter: ctnetlink: don't use conntrack/expect object addresses as id")
> Signed-off-by: Dirk Morris <dmorris@metaloft.com>

Looks good, thanks Dirk.

Acked-by: Florian Westphal <fw@strlen.de>

Re: [PATCH net v3] Use consistent ct id hash calculation

[Pablo Neira Ayuso]

On Thu, Aug 08, 2019 at 01:57:51PM -0700, Dirk Morris wrote:
> Change ct id hash calculation to only use invariants.
> 
> Currently the ct id hash calculation is based on some fields that can
> change in the lifetime on a conntrack entry in some corner cases. The
> current hash uses the whole tuple which contains an hlist pointer
> which will change when the conntrack is placed on the dying list
> resulting in a ct id change.
> 
> This patch also removes the reply-side tuple and extension pointer
> from the hash calculation so that the ct id will will not change from
> initialization until confirmation.

Applied, thanks Dirk.