Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1572614

[PATCH 4.4 15/20] tcp: initialize max window for a new fastopen socket

Show all headers | View raw

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Alexey Kodanev <alexey.kodanev@oracle.com>


[ Upstream commit 0dbd7ff3ac5017a46033a9d0a87a8267d69119d9 ]

Found that if we run LTP netstress test with large MSS (65K),
the first attempt from server to send data comparable to this
MSS on fastopen connection will be delayed by the probe timer.

Here is an example:

     < S  seq 0:0 win 43690 options [mss 65495 wscale 7 tfo cookie] length 32
     > S. seq 0:0 ack 1 win 43690 options [mss 65495 wscale 7] length 0
     < .  ack 1 win 342 length 0

Inside tcp_sendmsg(), tcp_send_mss() returns max MSS in 'mss_now',
as well as in 'size_goal'. This results the segment not queued for
transmition until all the data copied from user buffer. Then, inside
__tcp_push_pending_frames(), it breaks on send window test and
continues with the check probe timer.

Fragmentation occurs in tcp_write_wakeup()...

+0.2 > P. seq 1:43777 ack 1 win 342 length 43776
     < .  ack 43777, win 1365 length 0
     > P. seq 43777:65001 ack 1 win 342 options [...] length 21224
     ...

This also contradicts with the fact that we should bound to the half
of the window if it is large.

Fix this flaw by correctly initializing max_window. Before that, it
could have large values that affect further calculations of 'size_goal'.

Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/ipv4/tcp_fastopen.c |    1 +
 1 file changed, 1 insertion(+)

--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -161,6 +161,7 @@ static struct sock *tcp_fastopen_create_
 	 * scaled. So correct it appropriately.
 	 */
 	tp->snd_wnd = ntohs(tcp_hdr(skb)->window);
+	tp->max_window = tp->snd_wnd;
 
 	/* Activate the retrans timer so that SYNACK can be retransmitted.
 	 * The request socket is not added to the ehash

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 4.4 00/20] 4.4.47-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 15/20] tcp: initialize max window for a new fastopen socket Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 12/20] net: fix harmonize_features() vs NETIF_F_HIGHDMA Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 02/20] mlxsw: spectrum: Fix memory leak at skb reallocation Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 20/20] net: dsa: Bring back device detaching in dsa_slave_suspend() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 05/20] net: lwtunnel: Handle lwtunnel_fill_encap failure Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 14/20] ipv6: addrconf: Avoid addrconf_disable_change() using RCU read-side lock Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 13/20] net: phy: bcm63xx: Utilize correct config_intr function Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  [PATCH 4.4 06/20] net: ipv4: fix table id in getroute response Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-02-02 19:40 +0100
  Re: [PATCH 4.4 00/20] 4.4.47-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2017-02-02 21:40 +0100
  Re: [PATCH 4.4 00/20] 4.4.47-stable review Guenter Roeck <linux@roeck-us.net> - 2017-02-03 06:20 +0100

csiph-web