Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <0000000000008a7ae505aef61db1@google.com> <20200911170150.GA889@sol.localdomain>
 <c16e9ab9-13e0-b911-e33a-c9ae81e93a8d@I-love.SAKURA.ne.jp>
In-Reply-To: <c16e9ab9-13e0-b911-e33a-c9ae81e93a8d@I-love.SAKURA.ne.jp>
From:   Ard Biesheuvel <ardb@kernel.org>
Date:   Fri, 30 Jun 2023 11:36:51 +0200
Message-ID: <CAMj1kXFqYozjJ+qPeSApESb0Cb6CUaGXBrs5LP81ERRvb3+TAw@mail.gmail.com>
Subject: Re: [PATCH] net: tls: enable __GFP_ZERO upon tls_init()
To:     Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc:     Boris Pismenny <borisp@nvidia.com>,
        John Fastabend <john.fastabend@gmail.com>,
        Jakub Kicinski <kuba@kernel.org>, glider@google.com,
        herbert@gondor.apana.org.au, linux-crypto@vger.kernel.org,
        syzkaller-bugs@googlegroups.com,
        syzbot <syzbot+828dfc12440b4f6f305d@syzkaller.appspotmail.com>,
        Eric Biggers <ebiggers@kernel.org>,
        Aviad Yehezkel <aviadye@nvidia.com>,
        Daniel Borkmann <daniel@iogearbox.net>, netdev@vger.kernel.org,
        "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Paolo Abeni <pabeni@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, 28 Jun 2023 at 15:49, Tetsuo Handa
<penguin-kernel@i-love.sakura.ne.jp> wrote:
>
> syzbot is reporting uninit-value at aes_encrypt(), for block cipher assumes
> that bytes to encrypt/decrypt is multiple of block size for that cipher but
> tls_alloc_encrypted_msg() is not initializing padding bytes when
> required_size is not multiple of block cipher's block size.
>
> In order to make sure that padding bytes are automatically initialized,
> enable __GFP_ZERO flag when setsockopt(SOL_TCP, TCP_ULP, "tls") is called.
>
> Reported-by: syzbot <syzbot+828dfc12440b4f6f305d@syzkaller.appspotmail.com>
> Closes: https://syzkaller.appspot.com/bug?extid=828dfc12440b4f6f305d

Maybe I am missing something, but this syzkaller report appears to b a
huge collection of uninit-value reports, of which only one is related
to bpx_tx_verdict(), and that one has nor reproducer.

So even if this would be the right fix, I don't think closing that
issue makes sense, so? The issue refers to a couple of other
occcurences of uninit-value in aesti_encrypt, which do seem related.

> Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> ---
> According to C reproducer, this problem happens when bpf_exec_tx_verdict() is
> called with lower 4 bits of required_size being 0001 and does not happen when
> being 0100. Thus, I assumed that this problem is caused by lack of initializing
> padding bytes.
> But I couldn't figure out why KMSAN reports this problem when bpf_exec_tx_verdict()
> is called with lower 4 bits of required_size being 0001 for the second time and
> does not report this problem when bpf_exec_tx_verdict() is called with lower
> 4 bits of required_size being 0001 for the first time. More deeper problem exists?
> KMSAN reporting this problem when accessing u64 relevant?
>

As Eric pointed out as well, zeroing the allocation papers over the problem.

Why are you sending this now? Do you have a reproducer for this issue?

>  net/tls/tls_main.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
> index f2e7302a4d96..cd5366966864 100644
> --- a/net/tls/tls_main.c
> +++ b/net/tls/tls_main.c
> @@ -1025,6 +1025,7 @@ static int tls_init(struct sock *sk)
>         struct tls_context *ctx;
>         int rc = 0;
>
> +       sk->sk_allocation |= __GFP_ZERO;
>         tls_build_proto(sk);
>
>  #ifdef CONFIG_TLS_TOE
> --
> 2.34.1
>