netdev - Re: [PATCH net-next v2] tcp: avoid reducing cwnd when ACK+DSACK is received

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: <1420732772.5947.56.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Thu, 08 Jan 2015 07:59:32 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Sébastien Barré <sebastien.barre@...ouvain.be>
Cc:	David Miller <davem@...emloft.net>, netdev@...r.kernel.org,
	Gregory Detal <gregory.detal@...ouvain.be>,
	Nandita Dukkipati <nanditad@...gle.com>,
	Yuchung Cheng <ycheng@...gle.com>,
	Neal Cardwell <ncardwell@...gle.com>
Subject: Re: [PATCH net-next v2] tcp: avoid reducing cwnd when ACK+DSACK is
 received

On Thu, 2015-01-08 at 16:24 +0100, Sébastien Barré wrote:
> Hi Eric,
> 
> Le 08/01/2015 16:07, Eric Dumazet a écrit :
> >
> >>   In the current code,
> >> such ACK+DSACK will be missed and only at next, higher ack will the TLP
> >> episode be considered done. Since the DSACK is not present anymore,
> >> this will cost a cwnd reduction.
> >>
> >> This patch ensures that this scenario does not cause a cwnd reduction, since
> >> receiving an ACK+DSACK indicates that both the initial segment and the probe
> >> have been received by the peer.
> > Do you have at hand a packetdrill test to demonstrate that the patch
> > works ?
> Not currently, but that would be very good indeed.
> I will prepare one and send it.

Here is a skeleton you might adapt :

// TLP test when a sender does not have any new segments for transmission.
// In the absence of ACKs, the sender retransmits an old segment as a probe
// after 2 RTTs.

// Tolerate RTO variations
--tolerance_usecs=15000

// Set up production config.
`../common/defaults.sh
sysctl -q net.ipv4.tcp_early_retrans=3
`

// Establish a connection.
0.000 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
0.000 setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
0.000 bind(3, ..., ...) = 0
0.000 listen(3, 1) = 0

0.100  < S 0:0(0) win 32792 <mss 1000,sackOK,nop,nop,nop,wscale 7>
+0     > S. 0:0(0) ack 1 <mss 1460,nop,nop,sackOK,nop,wscale 6>
+0.100 < . 1:1(0) ack 1 win 257
+0     accept(3, ..., ...) = 4

// Send 10 MSS.
+0.100 write(4, ..., 10000) = 10000
+0 > . 1:5001(5000) ack 1
+0 > P. 5001:10001(5000) ack 1

// TLP retransmission in 2 RTTs.
+0.200 > P. 9001:10001(1000) ack 1
+0 %{
assert tcpi_snd_cwnd == 10, 'tcpi_snd_cwnd=%d' % tcpi_snd_cwnd
assert tcpi_unacked == 10
}%

+0.100 < . 1:1(0) ack 10001 win 257 <sack 9001:1001,nop,nop>

+0 %{
assert tcpi_snd_cwnd == 20, 'tcpi_snd_cwnd=%d' % tcpi_snd_cwnd
assert tcpi_unacked == 0, 'tcpi_unacked=%d' % tcpi_unacked
}%



--
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