[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a0130e21731e7b4c85a313a823f70cedbc9a1495@linux.dev>
Date: Tue, 14 Oct 2025 06:46:25 +0000
From: "Jiayuan Chen" <jiayuan.chen@...ux.dev>
To: xuanqiang.luo@...ux.dev, edumazet@...gle.com, kuniyu@...gle.com,
pabeni@...hat.com
Cc: kerneljasonxing@...il.com, davem@...emloft.net, kuba@...nel.org,
netdev@...r.kernel.org, horms@...nel.org, "Xuanqiang Luo"
<luoxuanqiang@...inos.cn>
Subject: Re: [RESEND PATCH net-next v7 3/3] inet: Avoid ehash lookup race in
inet_twsk_hashdance_schedule()
October 14, 2025 at 10:27, xuanqiang.luo@...ux.dev mailto:xuanqiang.luo@...ux.dev wrote:
>
> From: Xuanqiang Luo <luoxuanqiang@...inos.cn>
>
> 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.
>
> It's worth noting that we held lock_sock() before the replacement, so
> there's no need to check if sk is hashed. Thanks to Kuniyuki Iwashima!
>
> Fixes: 3ab5aee7fe84 ("net: Convert TCP & DCCP hash tables to use RCU / hlist_nulls")
> Reviewed-by: Kuniyuki Iwashima <kuniyu@...gle.com>
> Signed-off-by: Xuanqiang Luo <luoxuanqiang@...inos.cn>
Reviewed-by: Jiayuan Chen <jiayuan.chen@...ux.dev>
Powered by blists - more mailing lists