netdev - Re: TCP NewReno and single retransmit

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <CAK6E8=dYfJw16Q0D40QD7RLr=wq=y+5W59zHmZ24L49OPS9O5A@mail.gmail.com>
Date:	Fri, 31 Oct 2014 11:51:56 +0800
From:	Yuchung Cheng <ycheng@...gle.com>
To:	Marcelo Ricardo Leitner <mleitner@...hat.com>
Cc:	Neal Cardwell <ncardwell@...gle.com>,
	netdev <netdev@...r.kernel.org>,
	Eric Dumazet <edumazet@...gle.com>
Subject: Re: TCP NewReno and single retransmit

On Thu, Oct 30, 2014 at 7:24 PM, Marcelo Ricardo Leitner
<mleitner@...hat.com> wrote:
> On 30-10-2014 00:03, Neal Cardwell wrote:
>>
>> On Mon, Oct 27, 2014 at 2:49 PM, Marcelo Ricardo Leitner
>> <mleitner@...hat.com> wrote:
>>>
>>> Hi,
>>>
>>> We have a report from a customer saying that on a very calm connection,
>>> like
>>> having only a single data packet within some minutes, if this packet gets
>>> to
>>> be re-transmitted, retrans_stamp is only cleared when the next acked
>>> packet
>>> is received. But this may make we abort the connection too soon if this
>>> next
>>> packet also gets lost, because the reference for the initial loss is
>>> still
>>> for a big while ago..
>>
>> ...
>>>
>>> @@ -2382,31 +2382,32 @@ static inline bool tcp_may_undo(const struct
>>> tcp_sock *tp)
>>>   static bool tcp_try_undo_recovery(struct sock *sk)
>>
>> ...
>>>
>>>          if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) {
>>>                  /* Hold old state until something *above* high_seq
>>>                   * is ACKed. For Reno it is MUST to prevent false
>>>                   * fast retransmits (RFC2582). SACK TCP is safe. */
Or we can just remove this strange state-holding logic?

I couldn't find such a "MUST" statement in RFC2582. RFC2582 section 3
step 5 suggests exiting the recovery procedure when an ACK acknowledges
the "recover" variable (== tp->high_seq - 1).

Since we've called tcp_reset_reno_sack() before tcp_try_undo_recovery(),
I couldn't see how false fast retransmits can be triggered without
this state-holding.

Any insights?



>>>                  tcp_moderate_cwnd(tp);
>>> +               tp->retrans_stamp = 0;
>>>                  return true;
>>>          }
>>>          tcp_set_ca_state(sk, TCP_CA_Open);
>>>          return false;
>>>   }
>>>
>>> We would still hold state, at least part of it.. WDYT?
>>
>>
>> This approach sounds OK to me as long as we include a check of
>> tcp_any_retrans_done(), as we do in the similar code paths (for
>> motivation, see the comment above tcp_any_retrans_done()).
>
>
> Yes, okay. I thought that this would be taken care of already by then but
> reading the code again now after your comment, I can see what you're saying.
> Thanks.
>
>> So it sounds fine to me if you change that one new line to the following
>> 2:
>>
>> +  if (!tcp_any_retrans_done(sk))
>> +    tp->retrans_stamp = 0;
>
>
> Will do.
>
>> Nice catch!
>
>
> A good part of it (including the diagram) was done by customer. :)
> I'll post the patch as soon as we sync with them (credits).
>
> Marcelo
>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html