When tcp_send_synack() replaces the cloned SYN skb at the head of the
retransmit queue with a copy, it frees the original with
tcp_rtx_queue_unlink_and_free() and only repairs tp->highest_sack.
tp->retransmit_skb_hint keeps pointing at the freed
skbuff_fclone_cache object.

The dangling hint is read in tcp_verify_retransmit_hint() and used as
the root of the rbtree walk in tcp_xmit_retransmit_queue().  An
unprivileged TFO client (sendmsg(MSG_FASTOPEN)) can arm the hint with
an attacker-supplied ICMP fragmentation-needed message, after which a
simultaneous open frees the armed SYN skb:

  BUG: KASAN: slab-use-after-free in tcp_mark_skb_lost (net/ipv4/tcp_input.c:1316)
  Read of size 4 at addr ffff88800604d928 by task swapper/1/0
  Call Trace:
   tcp_mark_skb_lost (net/ipv4/tcp_input.c:1316)
   tcp_simple_retransmit (net/ipv4/tcp_input.c:3158)
   tcp_v4_err (net/ipv4/tcp_ipv4.c:587)

Sync the hint to the copy.

Fixes: c31b70c9968f ("tcp: Add logic to check for SYN w/ data in tcp_simple_retransmit")
Reported-by: Kimi Security Team <bug-report@moonshot.ai>
Tested-by: Weiming Shi <shiweiming@moonshot.ai>
Signed-off-by: Yilin Zhang <yilinzhang@moonshot.ai>
---
Changes in v2:
- Correct the Fixes: tag (Eric Dumazet).
---
 net/ipv4/tcp_output.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index d960e3de7d50..e0c392e29de5 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3886,6 +3886,7 @@ void tcp_send_active_reset(struct sock *sk, enum sk_rst_reason reason)
  */
 int tcp_send_synack(struct sock *sk)
 {
+	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
 
 	skb = tcp_rtx_queue_head(sk);
@@ -3903,6 +3904,8 @@ int tcp_send_synack(struct sock *sk)
 			if (!nskb)
 				return -ENOMEM;
 			INIT_LIST_HEAD(&nskb->tcp_tsorted_anchor);
+			if (skb == tp->retransmit_skb_hint)
+				tp->retransmit_skb_hint = nskb;
 			tcp_highest_sack_replace(sk, skb, nskb);
 			tcp_rtx_queue_unlink_and_free(skb, sk);
 			__skb_header_release(nskb);
-- 
2.34.1

On Thu, Sep 24, 2026 at 6:49 AM Yilin Zhang <yilinzhang@moonshot.ai> wrote:
>
> When tcp_send_synack() replaces the cloned SYN skb at the head of the
> retransmit queue with a copy, it frees the original with
> tcp_rtx_queue_unlink_and_free() and only repairs tp->highest_sack.
> tp->retransmit_skb_hint keeps pointing at the freed
> skbuff_fclone_cache object.
>
> The dangling hint is read in tcp_verify_retransmit_hint() and used as
> the root of the rbtree walk in tcp_xmit_retransmit_queue().  An
> unprivileged TFO client (sendmsg(MSG_FASTOPEN)) can arm the hint with
> an attacker-supplied ICMP fragmentation-needed message, after which a
> simultaneous open frees the armed SYN skb:
>
>   BUG: KASAN: slab-use-after-free in tcp_mark_skb_lost (net/ipv4/tcp_input.c:1316)
>   Read of size 4 at addr ffff88800604d928 by task swapper/1/0
>   Call Trace:
>    tcp_mark_skb_lost (net/ipv4/tcp_input.c:1316)
>    tcp_simple_retransmit (net/ipv4/tcp_input.c:3158)
>    tcp_v4_err (net/ipv4/tcp_ipv4.c:587)
>
> Sync the hint to the copy.
>
> Fixes: c31b70c9968f ("tcp: Add logic to check for SYN w/ data in tcp_simple_retransmit")
> Reported-by: Kimi Security Team <bug-report@moonshot.ai>
> Tested-by: Weiming Shi <shiweiming@moonshot.ai>
> Signed-off-by: Yilin Zhang <yilinzhang@moonshot.ai>
> ---
> Changes in v2:
> - Correct the Fixes: tag (Eric Dumazet).
> ---

Thanks for the fix.

Reviewed-by: Eric Dumazet <edumazet@google.com>