Skip to contentSkip to conversation

tcp: add a missing nf_reset_ct() in 3WHS handling

PATCH Applied Eric Dumazet 3 participants 3 messages Jun 23, 2022 – Jun 24, 2022

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tcp: add a missing nf_reset_ct() in 3WHS handling

When the third packet of 3WHS connection establishment contains payload, it is added into socket receive queue without the XFRM check and the drop of connection tracking context.

This means that if the data is left unread in the socket receive queue, conntrack module can not be unloaded.

As most applications usually reads the incoming data immediately after accept(), bug has been hiding for quite a long time.

Commit 68822bdf76f1 ("net: generalize skb freeing deferral to per-cpu lists") exposed this bug because even if the application reads this data, the skb with nfct state could stay in a per-cpu cache for an arbitrary time, if said cpu no longer process RX softirqs.

Many thanks to Ilya Maximets for reporting this issue, and for testing various patches:
https://lore.kernel.org/netdev/[email protected]/

Note that I also added a missing xfrm4_policy_check() call, although this is probably not a big issue, as the SYN packet should have been dropped earlier.

Fixes b59c270104f0 ("[NETFILTER]: Keep conntrack reference until IPsec policy checks are done")
Cc Florian Westphal <[email protected]>
Cc Pablo Neira Ayuso <[email protected]>
Cc Steffen Klassert <[email protected]>
---
 net/ipv4/tcp_ipv4.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
net/ipv4/tcp_ipv4.c +5 -1
index fe8f23b95d32ca4a35d05166d471327bc608fa91..da5a3c44c4fb70f1d3ecc596e694a86267f1c44a 100644
@@ -1964,7 +1964,10 @@ int tcp_v4_rcv(struct sk_buff *skb)
 		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		drop_reason = tcp_inbound_md5_hash(sk, skb,
+		if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
+			drop_reason = SKB_DROP_REASON_XFRM_POLICY;
+		else
+			drop_reason = tcp_inbound_md5_hash(sk, skb,
 						   &iph->saddr, &iph->daddr,
 						   AF_INET, dif, sdif);
 		if (unlikely(drop_reason)) {
@@ -2016,6 +2019,7 @@ int tcp_v4_rcv(struct sk_buff *skb)
 			}
 			goto discard_and_relse;
 		}
+		nf_reset_ct(skb);
 		if (nsk == sk) {
 			reqsk_put(req);
 			tcp_v4_restore_cb(skb);
Signature · 1 line
2.37.0.rc0.104.g0611611a94-goog
Tested-by Reviewed-by on v1 1/1
#1 lore raw
Quoted 32 lines · Eric Dumazet

When the third packet of 3WHS connection establishment contains payload, it is added into socket receive queue without the XFRM check and the drop of connection tracking context.

This means that if the data is left unread in the socket receive queue, conntrack module can not be unloaded.

As most applications usually reads the incoming data immediately after accept(), bug has been hiding for quite a long time.

Commit 68822bdf76f1 ("net: generalize skb freeing deferral to per-cpu lists") exposed this bug because even if the application reads this data, the skb with nfct state could stay in a per-cpu cache for an arbitrary time, if said cpu no longer process RX softirqs.

Many thanks to Ilya Maximets for reporting this issue, and for testing various patches:
https://lore.kernel.org/netdev/[email protected]/

Note that I also added a missing xfrm4_policy_check() call, although this is probably not a big issue, as the SYN packet should have been dropped earlier.

Fixes b59c270104f0 ("[NETFILTER]: Keep conntrack reference until IPsec policy checks are done")
Cc Florian Westphal <[email protected]>
Cc Pablo Neira Ayuso <[email protected]>
Cc Steffen Klassert <[email protected]>
Quoted 3 lines
---
 net/ipv4/tcp_ipv4.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

Thanks! I re-tested this change with the OVS testsuite and it works fine. It can successfully reload ntfilter modules now. So, for the nf_reset_ct part of the fix:

XFRM part seems correct to me, but I didn't test it.

#2 lore raw
In reply to Eric Dumazet
This patch was applied to netdev/net.git (master) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net.git (master)
by Jakub Kicinski <[email protected]>:

Quoted 9 lines · On Thu, 23 Jun 2022 05:04:36 +0000 you wrote:

When the third packet of 3WHS connection establishment contains payload, it is added into socket receive queue without the XFRM check and the drop of connection tracking context.

This means that if the data is left unread in the socket receive queue, conntrack module can not be unloaded.

[...]

Here is the summary with links:
- [net] tcp: add a missing nf_reset_ct() in 3WHS handling

    https://git.kernel.org/netdev/net/c/6f0012e35160

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html