netdev - Re: [PATCH] tcp: fixing TLP's FIN recovery

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]

Date:	Mon, 9 Jun 2014 00:02:45 -0700
From:	Nandita Dukkipati <nanditad@...gle.com>
To:	Per Hurtig <per.hurtig@....se>
Cc:	Eric Dumazet <eric.dumazet@...il.com>,
	Netdev <netdev@...r.kernel.org>,
	Anna Brunström <anna.brunstrom@....se>,
	mohammad.rajiullah@....se, Neal Cardwell <ncardwell@...gle.com>,
	sergei.shtylyov@...entembedded.com
Subject: Re: [PATCH] tcp: fixing TLP's FIN recovery

On Sun, Jun 8, 2014 at 12:41 AM, Per Hurtig <per.hurtig@....se> wrote:
>
>
> On sön  8 jun 2014 04:58:25, Eric Dumazet wrote:
>>
>> On Sat, 2014-06-07 at 16:34 +0200, Per Hurtig wrote:
>>>
>>> Fix to a problem observed when losing a FIN segment that does not
>>> contain data.  In such situations, TLP is unable to recover from
>>> *any* tail loss and instead adds at least PTO ms to the
>>> retransmission process, i.e., RTO = RTO + PTO.
>>>
>>> Signed-off-by: Per Hurtig <per.hurtig@....se>
>>> ---
>>>   net/ipv4/tcp_output.c | 6 ++++--
>>>   1 file changed, 4 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
>>> index d463c35..6573765 100644
>>> --- a/net/ipv4/tcp_output.c
>>> +++ b/net/ipv4/tcp_output.c
>>> @@ -2130,8 +2130,10 @@ void tcp_send_loss_probe(struct sock *sk)
>>>         if (WARN_ON(!skb || !tcp_skb_pcount(skb)))
>>>                 goto rearm_timer;
>>>
>>> -       /* Probe with zero data doesn't trigger fast recovery. */
>>> -       if (skb->len > 0)
>>> +       /* Probe with zero data doesn't trigger fast recovery, if FIN
>>> +        * flag is not set.
>>> +        */
>>> +       if ((skb->len > 0) || (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN))
>>>                 err = __tcp_retransmit_skb(sk, skb);
>>>
>>>         /* Record snd_nxt for loss detection. */
>>
>>
>>
>> You know, I believe the test was exactly to avoid sending data less FIN
>> packets.
>>
>> If you write :
>>
>>      if (A  || !A)
>>
>> Better remove the condition, completely ;)
>>
> Obviously, but I don't think that FINs are the only segments
> who are targeted by this condition (or targeted at all given
> the implications of this statement). Furthermore, the comment above
> the if statement would probably have mentioned FINs explicity
> and not zero sized segments in general if this were the case.
>
>
>
>>
>> Nandita, why FIN packet wont trigger fast retransnmits ?
>>
>
> They do, that's the whole thing with this patch.
>
>
>> It sounds like if the timer is the issue you want to fix, you might
>> simply rearm a timer with RTO-PTO instead of RTO ?
>>
>>
> No I want to enable TLP for tail loss where an empty FIN is involved,
> this does not work now.

I understand the tail loss case you want to solve - essentially when a
tail loss occurs that involves data segments as well as that of an
empty FIN. However, have you verified that re-sending an empty FIN
triggers fast recovery? I would be surprised if it did, because I
think the sender needs to receive a SACK of at least 1-byte of data
before sender can trigger FACK based fast recovery.

If you have verified that a pure FIN does indeed trigger recovery, can
you tell me what part of the code makes that happen?

Nandita
--
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