netdev - Re: [PATCH net-next v12 11/22] ovpn: implement TCP transport

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <af88a027-61ca-44f2-8d93-43d06076e2a6@openvpn.net>
Date: Wed, 4 Dec 2024 22:37:56 +0100
From: Antonio Quartulli <antonio@...nvpn.net>
To: Paolo Abeni <pabeni@...hat.com>, Eric Dumazet <edumazet@...gle.com>,
 Jakub Kicinski <kuba@...nel.org>, Donald Hunter <donald.hunter@...il.com>,
 Shuah Khan <shuah@...nel.org>, sd@...asysnail.net, ryazanov.s.a@...il.com,
 Andrew Lunn <andrew@...n.ch>
Cc: Simon Horman <horms@...nel.org>, netdev@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-kselftest@...r.kernel.org
Subject: Re: [PATCH net-next v12 11/22] ovpn: implement TCP transport

On 04/12/2024 12:15, Antonio Quartulli wrote:
[...]
>>> +static void ovpn_tcp_close(struct sock *sk, long timeout)
>>> +{
>>> +    struct ovpn_socket *sock;
>>> +
>>> +    rcu_read_lock();
>>> +    sock = rcu_dereference_sk_user_data(sk);
>>> +
>>> +    strp_stop(&sock->peer->tcp.strp);
>>> +    barrier();
>>
>> Again, is not clear to me the role of the above barrier, please 
>> document it.
> 
> Also taken from espintcp_close(), with the idea to avoid reordering 
> during the shut down sequence.
> 
> Will add a comment here too.

Actually, after checking this specific barrier(), I realized this is not 
needed, because we are doing the socket detach later, after calling the 
ovpn_peer_del() (which is different from what is happening in espintcp).

So I'll just drop this barrier here.

Regards,

-- 
Antonio Quartulli
OpenVPN Inc.