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: <31efe48a-4c68-f17c-64ee-88d45f56c438@arista.com>
Date:   Mon, 21 Nov 2022 20:31:38 +0000
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.5.0
Subject: Re: [PATCH v4 3/5] net/tcp: Disable TCP-MD5 static key on
 tcp_md5sig_info destruction
To:     Jakub Kicinski <kuba@kernel.org>
Cc:     linux-kernel@vger.kernel.org, David Ahern <dsahern@kernel.org>,
        Eric Dumazet <edumazet@google.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Ard Biesheuvel <ardb@kernel.org>,
        Bob Gilligan <gilligan@arista.com>,
        "David S. Miller" <davem@davemloft.net>,
        Dmitry Safonov <0x7f454c46@gmail.com>,
        Francesco Ruggeri <fruggeri@arista.com>,
        Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>,
        Jason Baron <jbaron@akamai.com>,
        Josh Poimboeuf <jpoimboe@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        Salam Noureddine <noureddine@arista.com>,
        Steven Rostedt <rostedt@goodmis.org>, netdev@vger.kernel.org
References: <20221115211905.1685426-1-dima@arista.com>
 <20221115211905.1685426-4-dima@arista.com>
 <20221118191809.0174f4da@kernel.org>
Content-Language: en-US
From:   Dmitry Safonov <dima@arista.com>
In-Reply-To: <20221118191809.0174f4da@kernel.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 11/19/22 03:18, Jakub Kicinski wrote:
> On Tue, 15 Nov 2022 21:19:03 +0000 Dmitry Safonov wrote:
>> +	if (!rcu_dereference_protected(tp->md5sig_info, lockdep_sock_is_held(sk))) {
>> +		if (tcp_md5sig_info_add(sk, sk_gfp_mask(sk, GFP_ATOMIC)))
>> +			return -ENOMEM;
>> +
>> +		if (!static_key_fast_inc_not_negative(&tcp_md5_needed.key.key)) {
>> +			struct tcp_md5sig_info *md5sig = tp->md5sig_info;
> 
> I don't think sparse will be able to deduce that ->md5sig_info access
> is safe here, so could you wrap it up as well?

Sure, that sounds logical to do.

> Maybe it wouldn't be 
> the worst move to provide a sk_rcu_dereference() or rcu_dereference_sk()
> or some such wrapper.
> 
> More importantly tho - was the merging part for this patches discussed?
> They don't apply to net-next.

They apply over linux-next as there's a change [1] in
linux-tip/locking/core on which the patches set based.

Could the way forward be through linux-tip tree, or that might create
net conflicts?

I'll send v5 with the trivial change to rcu_dereference_protected()
mentioned above.

[1]
https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=locking/core&id=d0c006402e7941558e5283ae434e2847c7999378

Thanks,
          Dmitry