From: Thaison Phan <thaisonphan@google.com>
To: Eric Dumazet <edumazet@kernel.org>,
Neal Cardwell <ncardwell@google.com>,
Kuniyuki Iwashima <kuniyu@google.com>,
"David S . Miller" <davem@davemloft.net>,
David Ahern <dsahern@kernel.org>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>
Cc: "Matthieu Baerts" <matttbe@kernel.org>,
"Simon Horman" <horms@kernel.org>,
"Ilpo Järvinen" <ij@kernel.org>,
"Olivier Tilmans" <olivier.tilmans@nokia.com>,
"Chia-Yu Chang" <chia-yu.chang@nokia-bell-labs.com>,
"Stan Iliev" <stani@google.com>,
netdev@vger.kernel.org, bpf@vger.kernel.org,
linux-kernel@vger.kernel.org,
"Thaison Phan" <thaisonphan@google.com>
Subject: [PATCH net v4 2/2] tcp: accecn: initialize AccECN fields in syncookie path
Date: Wed, 18 Nov 2026 21:09:33 +0000 [thread overview]
Message-ID: <20261118210933.2477423-3-thaisonphan@google.com> (raw)
In-Reply-To: <20261118210933.2477423-1-thaisonphan@google.com>
cookie_tcp_reqsk_alloc() and bpf_sk_assign_tcp_reqsk() allocate struct
request_sock without __GFP_ZERO, leaving saw_accecn_opt,
accecn_fail_mode, syn_ect_rcv, and syn_ect_snt uninitialized. When
accecn_ok is negotiated in cookie_{v4,v6}_check(),
tcp_ecn_openreq_child() reads all four fields to initialize the child
socket's AccECN state.
In addition, because the temporary request_sock used to send a cookie
SYN-ACK is immediately freed, treq->syn_ect_snt is lost. For AccECN
connections where the listener marks ECT on the SYN-ACK, leaving
syn_ect_snt as 0 causes tcp_accecn_third_ack() to see a mismatch between
the expected Not-ECT and the peer's reflected ECT in the ACE field,
erroneously flagging TCP_ACCECN_ACE_FAIL_RECV in accecn_fail_mode. Refer
to Table 4 of RFC 9768 for more details.
Initialize the AccECN fields in cookie_tcp_reqsk_init() to match
tcp_openreq_init(). For kernel-generated cookies (!req->syncookie), set
syn_ect_snt from the listener's TOS/tclass in cookie_{v4,v6}_check() to
match tcp_{v4,v6}_send_synack(); for BPF custom syncookies
(req->syncookie), leave syn_ect_snt at 0 (Not-ECT) since the SYN-ACK was
built by the BPF program rather than the kernel and bpf_tcp_req_attrs
does not report the SYN-ACK ECN codepoint.
Found by AI-assisted code inspection comparing tcp_openreq_init() and
cookie_tcp_reqsk_init() while investigating the cookie_v6_check()
LOOPBACK4_IPV6 bug. Reproduced in a QEMU VM with tcp_syncookies=2,
tcp_ecn=3, SLUB_RCU_DEBUG=y, and slub_debug=P, observing bogus
tcpi_accecn_fail_mode and tcpi_accecn_opt_seen values in TCP_INFO on the
accepted child socket (and tcpi_accecn_fail_mode=0x2 on ECT(0)
listeners), and tested with packetdrill AccECN and syncookie selftests.
Fixes: 3cae34274c79 ("tcp: accecn: AccECN negotiation")
Fixes: b40671b5ee58 ("tcp: accecn: AccECN option failure handling")
Signed-off-by: Thaison Phan <thaisonphan@google.com>
Reviewed-by: Eric Dumazet <edumazet@kernel.org>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
---
v4:
- restore treq->syn_ect_snt = 0 in cookie_tcp_reqsk_init() and guard
syn_ect_snt updates in cookie_{v4,v6}_check() with !req->syncookie
for BPF custom syncookies; document in changelog (Sashiko,
Eric Dumazet)
- add Reviewed-by tag from Matthieu Baerts
v3: https://lore.kernel.org/netdev/20261117124710.1338951-3-thaisonphan@google.com/
- clarify syn_ect_snt / tcp_accecn_third_ack() description for
AccECN connections and reference RFC 9768 Table 4 (Chia-Yu Chang)
v2: https://lore.kernel.org/netdev/20261116023709.132684-3-thaisonphan@google.com/
- drop redundant treq->syn_ect_snt = 0 in cookie_tcp_reqsk_init()
(Eric Dumazet)
- clarify how the bug was discovered and observed via TCP_INFO
(Eric Dumazet)
- add Reviewed-by tag from Eric Dumazet
v1: https://lore.kernel.org/netdev/20261114161248.3428695-3-thaisonphan@google.com/
net/ipv4/syncookies.c | 7 +++++++
net/ipv6/syncookies.c | 3 +++
2 files changed, 10 insertions(+)
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 4e0ff3b82062..89169c82431c 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -290,6 +290,10 @@ static int cookie_tcp_reqsk_init(struct sock *sk, struct sk_buff *skb,
*/
treq->txhash = treq->snt_isn;
treq->syn_tos = TCP_SKB_CB(skb)->ip_dsfield;
+ treq->saw_accecn_opt = TCP_ACCECN_OPT_NOT_SEEN;
+ treq->accecn_fail_mode = 0;
+ treq->syn_ect_rcv = 0;
+ treq->syn_ect_snt = 0;
#if IS_ENABLED(CONFIG_MPTCP)
treq->is_mptcp = sk_is_mptcp(sk);
@@ -502,6 +506,9 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
ireq->rcv_wscale = rcv_wscale;
ireq->ecn_ok &= cookie_ecn_ok(net, &rt->dst);
treq->accecn_ok = ireq->ecn_ok && cookie_accecn_ok(th);
+ /* SYN-ACK was not sent by the kernel if req->syncookie is set. */
+ if (!req->syncookie)
+ treq->syn_ect_snt = READ_ONCE(inet_sk(sk)->tos) & INET_ECN_MASK;
ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst);
/* ip_queue_xmit() depends on our flow being setup
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index b33011b71fff..e8d45a627ca2 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -276,6 +276,9 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
ireq->rcv_wscale = rcv_wscale;
ireq->ecn_ok &= cookie_ecn_ok(net, dst);
tcp_rsk(req)->accecn_ok = ireq->ecn_ok && cookie_accecn_ok(th);
+ /* SYN-ACK was not sent by the kernel if req->syncookie is set. */
+ if (!req->syncookie)
+ tcp_rsk(req)->syn_ect_snt = READ_ONCE(np->tclass) & INET_ECN_MASK;
ret = tcp_get_cookie_sock(sk, skb, req, dst);
if (!ret) {
--
2.56.1.831.g720f3f5937-goog
next prev parent reply other threads:[~2026-11-18 21:09 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-18 21:09 [PATCH net v4 0/2] tcp: initialize missing request_sock fields in syncookie path Thaison Phan
2026-11-18 21:09 ` [PATCH net v4 1/2] tcp: initialize ir_rmt_addr and ir_loc_addr in cookie_v6_check() Thaison Phan
2026-11-18 21:09 ` Thaison Phan [this message]
2026-11-19 8:48 ` [PATCH net v4 0/2] tcp: initialize missing request_sock fields in syncookie path Eric Dumazet
2026-11-20 3:20 ` patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261118210933.2477423-3-thaisonphan@google.com \
--to=thaisonphan@google.com \
--cc=bpf@vger.kernel.org \
--cc=chia-yu.chang@nokia-bell-labs.com \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=edumazet@kernel.org \
--cc=horms@kernel.org \
--cc=ij@kernel.org \
--cc=kuba@kernel.org \
--cc=kuniyu@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matttbe@kernel.org \
--cc=ncardwell@google.com \
--cc=netdev@vger.kernel.org \
--cc=olivier.tilmans@nokia.com \
--cc=pabeni@redhat.com \
--cc=stani@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.