Received-SPF: pass (google.com: best guess record for domain of linux-crypto-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20190614140122.20934-1-ard.biesheuvel@linaro.org>
In-Reply-To: <20190614140122.20934-1-ard.biesheuvel@linaro.org>
From:   Eric Dumazet <edumazet@google.com>
Date:   Fri, 14 Jun 2019 08:34:16 -0700
Message-ID: <CANn89iKP2fQ6Tc0jBW_WdLq3kYQx7NsdVDB5S3y453T+6yp86g@mail.gmail.com>
Subject: Re: [PATCH v2] net: ipv4: move tcp_fastopen server side code to
 SipHash library
To:     Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc:     netdev <netdev@vger.kernel.org>, linux-crypto@vger.kernel.org,
        Herbert Xu <herbert@gondor.apana.org.au>, ebiggers@kernel.org,
        David Miller <davem@davemloft.net>,
        Jason Baron <jbaron@akamai.com>,
        Christoph Paasch <cpaasch@apple.com>,
        David Laight <David.Laight@aculab.com>,
        Yuchung Cheng <ycheng@google.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-crypto-owner@vger.kernel.org
Precedence: bulk

On Fri, Jun 14, 2019 at 7:01 AM Ard Biesheuvel
<ard.biesheuvel@linaro.org> wrote:
>
> Using a bare block cipher in non-crypto code is almost always a bad idea,
> not only for security reasons (and we've seen some examples of this in
> the kernel in the past), but also for performance reasons.
>
> In the TCP fastopen case, we call into the bare AES block cipher one or
> two times (depending on whether the connection is IPv4 or IPv6). On most
> systems, this results in a call chain such as
>
>   crypto_cipher_encrypt_one(ctx, dst, src)
>     crypto_cipher_crt(tfm)->cit_encrypt_one(crypto_cipher_tfm(tfm), ...);
>       aesni_encrypt
>         kernel_fpu_begin();
>         aesni_enc(ctx, dst, src); // asm routine
>         kernel_fpu_end();
>
> It is highly unlikely that the use of special AES instructions has a
> benefit in this case, especially since we are doing the above twice
> for IPv6 connections, instead of using a transform which can process
> the entire input in one go.
>
> We could switch to the cbcmac(aes) shash, which would at least get
> rid of the duplicated overhead in *some* cases (i.e., today, only
> arm64 has an accelerated implementation of cbcmac(aes), while x86 will
> end up using the generic cbcmac template wrapping the AES-NI cipher,
> which basically ends up doing exactly the above). However, in the given
> context, it makes more sense to use a light-weight MAC algorithm that
> is more suitable for the purpose at hand, such as SipHash.
>
> Since the output size of SipHash already matches our chosen value for
> TCP_FASTOPEN_COOKIE_SIZE, and given that it accepts arbitrary input
> sizes, this greatly simplifies the code as well.
>
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>

While the patch looks fine (I yet have to run our tests with it), it
might cause some deployment issues
for server farms.

They usually share a common fastopen key, so that clients can reuse
the same token for different sessions.

Changing some servers in the pool will lead to inconsistencies.

Probably not a too big deal, but worth mentioning.