lkml.org 
[lkml]   [2015]   [May]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] net: tcp: Fix a PTO timing granularity issue
Hi Eric,

On Tue, May 26, 2015 at 11:25:21AM -0700, Eric Dumazet wrote:
> On Tue, 2015-05-26 at 13:55 -0400, Ido Yariv wrote:
>
> >
> > The platform this was tested on was an embedded platform with a wifi
> > module (11n, 20MHZ). The other end was a computer running Windows, and
> > the benchmarking software was IxChariot.
> > The whole setup was running in a shielded box with minimal
> > interferences.
> >
> > As it seems, the throughput was limited by the congestion window.
> > Further analysis led to TLP - the fact that its timer was expiring
> > prematurely impacted cwnd, which in turn prevented the wireless driver
> > from having enough skbs to buffer and send.
> >
> > Increasing the size of the chunks being sent had a similar impact on
> > throughput, presumably because the congestion window had enough time to
> > increase.
> >
> > Changing the congestion window to Westwood from cubic/reno also had a
> > similar impact on throughput.
> >
>
> Have you tested what results you had by completely disabling TLP ?
>
> Maybe a timer of 10 to 20ms is too short anyway in your testbed.

Yes, I have (by writing 2 to /proc/sys/net/ipv4/tcp_early_retrans), and
it also had a similar effect. That was actually the first workaround for
this issue, before the issue in TLP was traced.

With 10ms to 20ms timers the throughput was just the same as disabling
TLP altogether, so it seems it was just enough to handle.

Cheers,
Ido.


\
 
 \ /
  Last update: 2015-05-26 22:21    [W:0.066 / U:2.352 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site