Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <d3fccbd0-c92e-9aff-8c32-48c1171746c3@ispras.ru>
Date:   Fri, 26 May 2023 19:41:07 +0300
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.10.0
Subject: Re: [PATCH] udp6: Fix race condition in udp6_sendmsg & connect
Content-Language: ru
To:     Eric Dumazet <edumazet@google.com>
Cc:     Paolo Abeni <pabeni@redhat.com>,
        Willem de Bruijn <willemdebruijn.kernel@gmail.com>,
        "David S. Miller" <davem@davemloft.net>,
        David Ahern <dsahern@kernel.org>,
        Jakub Kicinski <kuba@kernel.org>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, lvc-project@linuxtesting.org
References: <20230526150806.1457828-1-VEfanov@ispras.ru>
 <27614af23cd7ae4433b909194062c553a6ae16ac.camel@redhat.com>
 <027d28a0-b31b-ab42-9eb6-2826c04c9364@ispras.ru>
 <CANn89iLGOVwW-KHBuJ94E+QoVARWw5EBKyfh0mPkOT+5ws31Fw@mail.gmail.com>
From:   Vlad Efanov <vefanov@ispras.ru>
In-Reply-To: <CANn89iLGOVwW-KHBuJ94E+QoVARWw5EBKyfh0mPkOT+5ws31Fw@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

sk_dst_set() is called by sk_setup_caps().

sk_dst_set() replaces dst in socket using xchg() call and we still have 
two tasks use one socket but expect different dst in sk_dst_cache.


__sk_dst_set() is rcu protected, but it checks for socket lock.


static inline void
__sk_dst_set(struct sock *sk, struct dst_entry *dst)
{
     struct dst_entry *old_dst;

     sk_tx_queue_clear(sk);
     sk->sk_dst_pending_confirm = 0;
     old_dst = rcu_dereference_protected(sk->sk_dst_cache,
                         lockdep_sock_is_held(sk));
     rcu_assign_pointer(sk->sk_dst_cache, dst);
     dst_release(old_dst);
}


Best regards.

Vlad.


On 26.05.2023 19:00, Eric Dumazet wrote:
> On Fri, May 26, 2023 at 5:58 PM Ефанов Владислав Александрович
> <vefanov@ispras.ru> wrote:
>> Paolo,
>>
>>
>> I don't think that we can just move sk_dst_set() call.
>>
>> I think we can destroy dst of sendmsg task in this case.
>>
> dst are RCU protected, it should be easy to make sure we respect all the rules.