On Wed, 24 Sep 2025 09:50:34 +0800 xuanqiang.luo@linux.dev wrote:

From: Xuanqiang Luo <redacted>

Since ehash lookups are lockless, if another CPU is converting sk to tw
concurrently, fetching the newly inserted tw with tw->tw_refcnt == 0 cause
lookup failure.

The call trace map is drawn as follows:
   CPU 0                                CPU 1
   -----                                -----
				     inet_twsk_hashdance_schedule()
				     spin_lock()
				     inet_twsk_add_node_rcu(tw, ...)
__inet_lookup_established()
(find tw, failure due to tw_refcnt = 0)
				     __sk_nulls_del_node_init_rcu(sk)
				     refcount_set(&tw->tw_refcnt, 3)
				     spin_unlock()

By replacing sk with tw atomically via hlist_nulls_replace_init_rcu() after
setting tw_refcnt, we ensure that tw is either fully initialized or not
visible to other CPUs, eliminating the race.

This one doesn't build cleanly

net/ipv4/inet_timewait_sock.c:116:28: warning: unused variable 'ehead' [-Wunused-variable]
  116 |         struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, sk->sk_hash);
      |                                   ^~~~~
net/ipv4/inet_timewait_sock.c:91:13: warning: unused function 'inet_twsk_add_node_rcu' [-Wunused-function]
   91 | static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
      |             ^~~~~~~~~~~~~~~~~~~~~~
-- 
pw-bot: cr