netdev - Re: Linux ECN Handling

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 for Android: free password hash cracker in your pocket

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

Message-ID: <CADVnQynQq6MWLXTaugV-rJq1MBmCrmL4bWNupWC+-uZw0Dz84w@mail.gmail.com>
Date:   Tue, 19 Dec 2017 19:08:50 -0500
From:   Neal Cardwell <ncardwell@...gle.com>
To:     Steve Ibanez <sibanez@...nford.edu>
Cc:     Eric Dumazet <edumazet@...gle.com>,
        Yuchung Cheng <ycheng@...gle.com>,
        Daniel Borkmann <daniel@...earbox.net>,
        Netdev <netdev@...r.kernel.org>, Florian Westphal <fw@...len.de>,
        Mohammad Alizadeh <alizadeh@...il.mit.edu>,
        Lawrence Brakmo <brakmo@...com>
Subject: Re: Linux ECN Handling

On Tue, Dec 19, 2017 at 5:00 PM, Steve Ibanez <sibanez@...nford.edu> wrote:
> Hi Neal,
>
> I managed to track down the code path that the unACKed CWR packet is
> taking. The tcp_rcv_established() function calls tcp_ack_snd_check()
> at the end of step5 and then the return statement indicated below is
> invoked, which prevents the __tcp_ack_snd_check() function from
> running.
>
> static inline void tcp_ack_snd_check(struct sock *sk)
> {
>         if (!inet_csk_ack_scheduled(sk)) {
>                 /* We sent a data segment already. */
>                 return;   /* <=== here */
>         }
>         __tcp_ack_snd_check(sk, 1);
> }
>
> So somehow tcp_ack_snd_check() thinks that a data segment was already
> sent when in fact it wasn't. Do you see a way around this issue?

Thanks for tracking that down! AFAICT in this case the call chain we
are trying to achieve is as follows:

tcp_rcv_established()
 -> tcp_data_queue()
 -> tcp_event_data_recv()
 -> inet_csk_schedule_ack()

The only think I can think of would be to add printks that fire for
CWR packets, to isolate why the code bails out before it reaches those
calls...

thanks,
neal