* [PATCH 5.15.y 0/2] netfilter: nft_ct: move custom expectation support to helper
@ 2026-11-18 0:00 Mark Bundschuh
2026-11-18 0:00 ` [PATCH 5.15.y 1/2] netfilter: nft_ct: expectation timeouts are passed in milliseconds Mark Bundschuh
2026-11-18 0:00 ` [PATCH 5.15.y 2/2] netfilter: nft_ct: move custom expectation support to helper Mark Bundschuh
0 siblings, 2 replies; 3+ messages in thread
From: Mark Bundschuh @ 2026-11-18 0:00 UTC (permalink / raw)
To: stable; +Cc: Mark Bundschuh
Also in this series:
- 6fb421bd07f1 ("netfilter: nft_ct: expectation timeouts are passed in
milliseconds") from v7.2-rc1 is included because it fixes the
timeout unit bug this branch has, and the fix's moved code uses its
jiffies timeout.
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH 5.15.y 1/2] netfilter: nft_ct: expectation timeouts are passed in milliseconds
2026-11-18 0:00 [PATCH 5.15.y 0/2] netfilter: nft_ct: move custom expectation support to helper Mark Bundschuh
@ 2026-11-18 0:00 ` Mark Bundschuh
2026-11-18 0:00 ` [PATCH 5.15.y 2/2] netfilter: nft_ct: move custom expectation support to helper Mark Bundschuh
1 sibling, 0 replies; 3+ messages in thread
From: Mark Bundschuh @ 2026-11-18 0:00 UTC (permalink / raw)
To: stable; +Cc: Florian Westphal, Pablo Neira Ayuso, Mark Bundschuh
From: Florian Westphal <fw@strlen.de>
[ Upstream commit 6fb421bd07f156cdf0cdede062d31f1c21def326 ]
Userspace passes '5000' in case user asks for 5 seconds.
Allowing for sub-second expectation lifetimes makes sense to me. so
fix up the kernel side instead of munging nft to send a value rounded
up to next second.
Also note that this violates nft convention of passing integers in
network byte order, but we can't change this anymore.
Fixes: 857b46027d6f ("netfilter: nft_ct: add ct expectations support")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Stable-dep-of: 3679da4ad8be ("netfilter: nft_ct: move custom expectation support to helper")
[ b8b09dc2bf35 ("netfilter: nf_conntrack_expect: use conntrack GC to
reap expectations") is not backported because it replaces the
expectation timer with GC reaping in 9 files, so timer_list expires is
set, and mod_timer_pending() after insert. ]
Signed-off-by: Mark Bundschuh <mkbund@amazon.com>
---
net/netfilter/nft_ct.c | 21 ++++++++++++++++++---
1 file changed, 18 insertions(+), 3 deletions(-)
diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c
index 4e723257f356..af58d86b0783 100644
--- a/net/netfilter/nft_ct.c
+++ b/net/netfilter/nft_ct.c
@@ -1184,11 +1184,23 @@ struct nft_ct_expect_obj {
u32 timeout;
};
+static int nft_ct_expect_timeout_get(const struct nlattr *attr, u32 *val)
+{
+ unsigned long jiffies_val = msecs_to_jiffies(nla_get_u32(attr));
+
+ if (jiffies_val > UINT_MAX)
+ return -ERANGE;
+
+ *val = jiffies_val;
+ return 0;
+}
+
static int nft_ct_expect_obj_init(const struct nft_ctx *ctx,
const struct nlattr * const tb[],
struct nft_object *obj)
{
struct nft_ct_expect_obj *priv = nft_obj_data(obj);
+ int err;
if (!tb[NFTA_CT_EXPECT_L4PROTO] ||
!tb[NFTA_CT_EXPECT_DPORT] ||
@@ -1224,8 +1236,11 @@ static int nft_ct_expect_obj_init(const struct nft_ctx *ctx,
return -EOPNOTSUPP;
}
+ err = nft_ct_expect_timeout_get(tb[NFTA_CT_EXPECT_TIMEOUT], &priv->timeout);
+ if (err)
+ return err;
+
priv->dport = nla_get_be16(tb[NFTA_CT_EXPECT_DPORT]);
- priv->timeout = nla_get_u32(tb[NFTA_CT_EXPECT_TIMEOUT]);
priv->size = nla_get_u8(tb[NFTA_CT_EXPECT_SIZE]);
return nf_ct_netns_get(ctx->net, ctx->family);
@@ -1245,7 +1260,7 @@ static int nft_ct_expect_obj_dump(struct sk_buff *skb,
if (nla_put_be16(skb, NFTA_CT_EXPECT_L3PROTO, htons(priv->l3num)) ||
nla_put_u8(skb, NFTA_CT_EXPECT_L4PROTO, priv->l4proto) ||
nla_put_be16(skb, NFTA_CT_EXPECT_DPORT, priv->dport) ||
- nla_put_u32(skb, NFTA_CT_EXPECT_TIMEOUT, priv->timeout) ||
+ nla_put_u32(skb, NFTA_CT_EXPECT_TIMEOUT, jiffies_to_msecs(priv->timeout)) ||
nla_put_u8(skb, NFTA_CT_EXPECT_SIZE, priv->size))
return -1;
@@ -1295,7 +1310,7 @@ static void nft_ct_expect_obj_eval(struct nft_object *obj,
&ct->tuplehash[!dir].tuple.src.u3,
&ct->tuplehash[!dir].tuple.dst.u3,
priv->l4proto, NULL, &priv->dport);
- exp->timeout.expires = jiffies + priv->timeout * HZ;
+ exp->timeout.expires = jiffies + priv->timeout;
if (nf_ct_expect_related(exp, 0) != 0)
regs->verdict.code = NF_DROP;
base-commit: 29c6cdaac04a6989a7ed4ae4275ce4244e617f58
--
2.50.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [PATCH 5.15.y 2/2] netfilter: nft_ct: move custom expectation support to helper
2026-11-18 0:00 [PATCH 5.15.y 0/2] netfilter: nft_ct: move custom expectation support to helper Mark Bundschuh
2026-11-18 0:00 ` [PATCH 5.15.y 1/2] netfilter: nft_ct: expectation timeouts are passed in milliseconds Mark Bundschuh
@ 2026-11-18 0:00 ` Mark Bundschuh
1 sibling, 0 replies; 3+ messages in thread
From: Mark Bundschuh @ 2026-11-18 0:00 UTC (permalink / raw)
To: stable; +Cc: Pablo Neira Ayuso, Jaeyeong Lee, Mark Bundschuh
From: Pablo Neira Ayuso <pablo@netfilter.org>
[ Upstream commit 3679da4ad8be84cddaf40bc307fef1fe13e051ff ]
Originally, the ct expectation support called nf_ct_helper_ext_add() for
confirmed conntracks, which is invalid, triggering a splat. This was
fixed by commit 1710eb913bdc ("netfilter: nft_ct: skip expectations for
confirmed conntrack") which restricted it to unconfirmed conntracks.
However, early insertion of expectations into the expectations list when
the conntrack is unconfirmed leads to stale entries pointing to the
wrong hlist_head through .pprev due to ct extension reallocation.
Commit 7c9664351980 ("netfilter: move nat hlist_head to nf_conn") moved
the nat hlist_head to nf_conn for this reason:
1. ...
2. When reallocation of extension area occurs we need to fixup the
bysource hash head via hlist_replace_rcu.
I'd rather not increase the size of the struct nf_conn for this feature
has very limited scope: only one expectation can be created at a time
given expect_clash() will make nf_ct_expect_related() reports EBUSY.
For this reason, relax nf_ct_expect_related() not to drop packets in
case expectation creation fails, therefore, expectation creation becomes
best effort.
To address this issue, add an internal ct helper and attach it to the
conntrack entry to streamline the custom ct expectation support with
existing ct helpers.
Expose a new nf_conntrack_helper_release() function to release the
internal helper that is allocated and attached to the conntrack entry to
create the custom expectations. The nft_ct module removal always waits
for rcu grace period, then the NULL helper callback is observed after
this.
This patch also restricts the creation of expectations to different
helpers other than this custom helper that is created for this type of
expectations.
Fixes: 857b46027d6f ("netfilter: nft_ct: add ct expectations support")
Reported-by: Jaeyeong Lee <iostreampy@proton.me>
Link: https://patch.msgid.link/20260715144755.00ea7dfcd9f@proton.me
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
[ - ac46f3f35b6e ("netfilter: nf_conntrack_helper: add refcounting from
datapath") is not backported because it reworks helper lifetime
around a datapath ct_refcnt in 9 files, so one static nft_ct_expect
helper is unhelped at nft_ct module exit.
- d4beefc90a66 ("netfilter: nft_ct: support expectation creation for
natted flows") is not backported because it adds expectation support
for NATed flows, so the moved expectation code has no NAT
follow-master part.
- b8b09dc2bf35 ("netfilter: nf_conntrack_expect: use conntrack GC to
reap expectations") is not backported because it replaces the
expectation timer with GC reaping in 9 files, so timer_list expires
is set, and mod_timer_pending() after insert.
- ace53fdc262f ("netfilter: conntrack: remove
__nf_ct_unconfirmed_destroy") is not backported because it reworks
unconfirmed conntrack teardown in conntrack core, so the release
helper keeps this branch's final synchronize_rcu().
- Context differs due to commit: 6031487d4e27 ]
Signed-off-by: Mark Bundschuh <mkbund@amazon.com>
---
include/net/netfilter/nf_conntrack_helper.h | 1 +
net/netfilter/nf_conntrack_helper.c | 20 ++--
net/netfilter/nft_ct.c | 107 +++++++++++++++-----
3 files changed, 96 insertions(+), 32 deletions(-)
diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h
index 9939c366f720..53d957ffbaa6 100644
--- a/include/net/netfilter/nf_conntrack_helper.h
+++ b/include/net/netfilter/nf_conntrack_helper.h
@@ -105,6 +105,7 @@ void nf_ct_helper_init(struct nf_conntrack_helper *helper,
int nf_conntrack_helper_register(struct nf_conntrack_helper *);
void nf_conntrack_helper_unregister(struct nf_conntrack_helper *);
+void nf_conntrack_helper_release(struct nf_conntrack_helper *);
int nf_conntrack_helpers_register(struct nf_conntrack_helper *, unsigned int);
void nf_conntrack_helpers_unregister(struct nf_conntrack_helper *,
diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c
index b5f4bfc60f89..0688428f4d2b 100644
--- a/net/netfilter/nf_conntrack_helper.c
+++ b/net/netfilter/nf_conntrack_helper.c
@@ -457,6 +457,18 @@ static bool expect_iter_me(struct nf_conntrack_expect *exp, void *data)
return this == me;
}
+void nf_conntrack_helper_release(struct nf_conntrack_helper *me)
+{
+ nf_ct_expect_iterate_destroy(expect_iter_me, me);
+ nf_ct_iterate_destroy(unhelp, me);
+
+ /* Maybe someone has gotten the helper already when unhelp above.
+ * So need to wait it.
+ */
+ synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(nf_conntrack_helper_release);
+
void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me)
{
mutex_lock(&nf_ct_helper_mutex);
@@ -469,13 +481,7 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me)
*/
synchronize_rcu();
- nf_ct_expect_iterate_destroy(expect_iter_me, me);
- nf_ct_iterate_destroy(unhelp, me);
-
- /* Maybe someone has gotten the helper already when unhelp above.
- * So need to wait it.
- */
- synchronize_rcu();
+ nf_conntrack_helper_release(me);
}
EXPORT_SYMBOL_GPL(nf_conntrack_helper_unregister);
diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c
index af58d86b0783..4a130c45d4f2 100644
--- a/net/netfilter/nft_ct.c
+++ b/net/netfilter/nft_ct.c
@@ -1195,6 +1195,72 @@ static int nft_ct_expect_timeout_get(const struct nlattr *attr, u32 *val)
return 0;
}
+struct nft_ct_expect_data {
+ struct nft_ct_expect_obj obj;
+ enum ip_conntrack_dir dir;
+};
+
+static int ct_expect_help(struct sk_buff *skb, unsigned int protoff,
+ struct nf_conn *ct, enum ip_conntrack_info ctinfo)
+{
+ enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
+ struct nft_ct_expect_data *expect_data;
+ struct nf_conntrack_expect *exp;
+ int ret = NF_ACCEPT;
+ u16 l3num;
+
+ if (nf_ct_is_confirmed(ct))
+ return NF_ACCEPT;
+
+ expect_data = nfct_help_data(ct);
+ if (!expect_data)
+ return NF_ACCEPT;
+
+ if (expect_data->dir != dir)
+ return NF_ACCEPT;
+
+ exp = nf_ct_expect_alloc(ct);
+ if (!exp)
+ return NF_DROP;
+
+ if (expect_data->obj.l3num == NFPROTO_INET)
+ l3num = nf_ct_l3num(ct);
+ else
+ l3num = expect_data->obj.l3num;
+
+ nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, l3num,
+ &ct->tuplehash[!dir].tuple.src.u3,
+ &ct->tuplehash[!dir].tuple.dst.u3,
+ expect_data->obj.l4proto, NULL, &expect_data->obj.dport);
+ /* nf_ct_expect_insert() arms the timer from the helper policy,
+ * whose timeout is zero. Set the object's timeout once inserted,
+ * with BH off so the timer cannot fire on this CPU in between.
+ */
+ local_bh_disable();
+ if (nf_ct_expect_related(exp, 0) != 0)
+ ret = NF_ACCEPT;
+ else
+ mod_timer_pending(&exp->timeout,
+ jiffies + expect_data->obj.timeout);
+ local_bh_enable();
+
+ nf_ct_expect_put(exp);
+
+ return ret;
+}
+
+static const struct nf_conntrack_expect_policy nft_ct_expect_helper_policy = {
+ .max_expected = NF_CT_EXPECT_MAX_CNT,
+};
+
+/* No need to register this helper, this is internal. */
+static struct nf_conntrack_helper nft_ct_expect_helper = {
+ .name = "nft_ct_expect",
+ .me = THIS_MODULE,
+ .expect_policy = &nft_ct_expect_helper_policy,
+ .help = ct_expect_help,
+};
+
static int nft_ct_expect_obj_init(const struct nft_ctx *ctx,
const struct nlattr * const tb[],
struct nft_object *obj)
@@ -1202,6 +1268,8 @@ static int nft_ct_expect_obj_init(const struct nft_ctx *ctx,
struct nft_ct_expect_obj *priv = nft_obj_data(obj);
int err;
+ NF_CT_HELPER_BUILD_BUG_ON(sizeof(struct nft_ct_expect_data));
+
if (!tb[NFTA_CT_EXPECT_L4PROTO] ||
!tb[NFTA_CT_EXPECT_DPORT] ||
!tb[NFTA_CT_EXPECT_TIMEOUT] ||
@@ -1247,7 +1315,7 @@ static int nft_ct_expect_obj_init(const struct nft_ctx *ctx,
}
static void nft_ct_expect_obj_destroy(const struct nft_ctx *ctx,
- struct nft_object *obj)
+ struct nft_object *obj)
{
nf_ct_netns_put(ctx->net, ctx->family);
}
@@ -1272,11 +1340,9 @@ static void nft_ct_expect_obj_eval(struct nft_object *obj,
const struct nft_pktinfo *pkt)
{
const struct nft_ct_expect_obj *priv = nft_obj_data(obj);
- struct nf_conntrack_expect *exp;
+ struct nft_ct_expect_data *expect_data;
enum ip_conntrack_info ctinfo;
struct nf_conn_help *help;
- enum ip_conntrack_dir dir;
- u16 l3num = priv->l3num;
struct nf_conn *ct;
ct = nf_ct_get(pkt->skb, &ctinfo);
@@ -1284,38 +1350,28 @@ static void nft_ct_expect_obj_eval(struct nft_object *obj,
regs->verdict.code = NFT_BREAK;
return;
}
- dir = CTINFO2DIR(ctinfo);
help = nfct_help(ct);
- if (!help)
- help = nf_ct_helper_ext_add(ct, GFP_ATOMIC);
- if (!help) {
- regs->verdict.code = NF_DROP;
- return;
- }
-
- if (help->expecting[NF_CT_EXPECT_CLASS_DEFAULT] >= priv->size) {
+ if (help) {
regs->verdict.code = NFT_BREAK;
return;
}
- if (l3num == NFPROTO_INET)
- l3num = nf_ct_l3num(ct);
- exp = nf_ct_expect_alloc(ct);
- if (exp == NULL) {
+ help = nf_ct_helper_ext_add(ct, GFP_ATOMIC);
+ if (!help) {
regs->verdict.code = NF_DROP;
return;
}
- nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, l3num,
- &ct->tuplehash[!dir].tuple.src.u3,
- &ct->tuplehash[!dir].tuple.dst.u3,
- priv->l4proto, NULL, &priv->dport);
- exp->timeout.expires = jiffies + priv->timeout;
- if (nf_ct_expect_related(exp, 0) != 0)
- regs->verdict.code = NF_DROP;
+ expect_data = nfct_help_data(ct);
+ if (!expect_data) {
+ regs->verdict.code = NFT_BREAK;
+ return;
+ }
+ expect_data->obj = *priv;
+ expect_data->dir = CTINFO2DIR(ctinfo);
- nf_ct_expect_put(exp);
+ rcu_assign_pointer(help->helper, &nft_ct_expect_helper);
}
static const struct nla_policy nft_ct_expect_policy[NFTA_CT_EXPECT_MAX + 1] = {
@@ -1395,6 +1451,7 @@ static void __exit nft_ct_module_exit(void)
nft_unregister_obj(&nft_ct_helper_obj_type);
nft_unregister_expr(&nft_notrack_type);
nft_unregister_expr(&nft_ct_type);
+ nf_conntrack_helper_release(&nft_ct_expect_helper);
}
module_init(nft_ct_module_init);
--
2.50.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-11-18 0:00 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-18 0:00 [PATCH 5.15.y 0/2] netfilter: nft_ct: move custom expectation support to helper Mark Bundschuh
2026-11-18 0:00 ` [PATCH 5.15.y 1/2] netfilter: nft_ct: expectation timeouts are passed in milliseconds Mark Bundschuh
2026-11-18 0:00 ` [PATCH 5.15.y 2/2] netfilter: nft_ct: move custom expectation support to helper Mark Bundschuh
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.