netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Neal Cardwell <ncardwell@google.com>
To: Per Hurtig <per.hurtig@kau.se>
Cc: "Yuchung Cheng" <ycheng@google.com>,
	netdev <netdev@vger.kernel.org>,
	"David Miller" <davem@davemloft.net>,
	"Eric Dumazet" <edumazet@google.com>,
	"Anna Brunström" <anna.brunstrom@kau.se>,
	"Andreas Petlund" <apetlund@simula.no>,
	"Michael Welzl" <michawe@ifi.uio.no>,
	"Ilpo Järvinen" <ilpo.jarvinen@helsinki.fi>,
	"Nandita Dukkipati" <nanditad@google.com>
Subject: Re: [PATCH 1/1] net: tcp: RTO restart
Date: Wed, 26 Feb 2014 09:57:39 -0500	[thread overview]
Message-ID: <CADVnQyn4Zszpc2eTg0BkJZY0+Y-LBcPHTPTjVS2o8_zZTX31jg@mail.gmail.com> (raw)
In-Reply-To: <D6FBF93A-8057-497F-A39E-DF6211999DC1@kau.se>

On Wed, Feb 26, 2014 at 4:50 AM, Per Hurtig <per.hurtig@kau.se> wrote:
> Also, as your setup includes TLP we only have the RTO restart over TLP gain to start
> with.

My gut says the next thing to try is applying the "RTO restart" idea
to the scheduling of the TLP probe (scheduling the TLP probe for
write_queue_head_skb->when + tlp_timeout). Doing an RTO earlier can be
a very bad thing, due to reseting the cwnd to 1. But doing a TLP
earlier does not have much downside, and could have significant
latency savings (though maybe Nandita already tried this at some
point)...

neal

  reply	other threads:[~2014-02-26 15:04 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-02-18 18:12 [PATCH 1/1] net: tcp: RTO restart Per Hurtig
2014-02-18 18:46 ` Yuchung Cheng
2014-02-19 10:50   ` Per Hurtig
2014-02-19 17:51     ` Yuchung Cheng
2014-02-21 10:48       ` Per Hurtig
2014-02-21 16:53         ` Yuchung Cheng
2014-02-25  0:03           ` Yuchung Cheng
2014-02-26  9:50             ` Per Hurtig
2014-02-26 14:57               ` Neal Cardwell [this message]
2014-02-26 19:52                 ` Nandita Dukkipati
2014-02-19 17:17 ` Andreas Petlund
2014-02-19 18:01   ` Eric Dumazet
2014-02-19 20:09     ` Andreas Petlund
2014-02-21 10:49     ` Per Hurtig
2014-02-21 13:14       ` Eric Dumazet
2014-02-19 18:20 ` Bill Fink
2014-02-21 10:48   ` Per Hurtig

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CADVnQyn4Zszpc2eTg0BkJZY0+Y-LBcPHTPTjVS2o8_zZTX31jg@mail.gmail.com \
    --to=ncardwell@google.com \
    --cc=anna.brunstrom@kau.se \
    --cc=apetlund@simula.no \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=ilpo.jarvinen@helsinki.fi \
    --cc=michawe@ifi.uio.no \
    --cc=nanditad@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=per.hurtig@kau.se \
    --cc=ycheng@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).