netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Tore Anderson <tore@fud.no>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "Maciej Żenczykowski" <maze@google.com>,
	"David Miller" <davem@davemloft.net>,
	netdev <netdev@vger.kernel.org>,
	"Tom Herbert" <therbert@google.com>
Subject: Re: [PATCH net-next] ipv6: RTAX_FEATURE_ALLFRAG causes inefficient TCP segment sizing
Date: Thu, 26 Apr 2012 12:32:17 +0200	[thread overview]
Message-ID: <0e135d034e75f56cb6d5318595c17023@greed.fud.no> (raw)
In-Reply-To: <1335350667.3274.53.camel@edumazet-glaptop>

* Eric Dumazet

> Yes, in fact we are going to keep the allfrag feature, and arm it if 
> we
> receive an mtu < 512.

One thing to think about if you go down this path is to keep track of 
the PMTU received in the ICMPv6 PTB. If you activate the allfrag feature 
on a destination because you received a PTB with PMTU=400, it's not 
useful to include the Fragmentation header in packets <400 bytes in 
size, for example.

With the current code, even tiny packets without any L4 payload (e.g. 
TCP SYN/ACK) will have the Fragmentation header added. That's not 
optimal, but necessary since the current code does not keep track of the 
actual PMTU if it is <1280 as far as I can tell (instead it discards the 
received PMTU value, clamps the effective PMTU to 1280, and enables 
allfrag).

Tore

  reply	other threads:[~2012-04-26 10:32 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-04-24 17:37 [PATCH net-next] ipv6: RTAX_FEATURE_ALLFRAG causes inefficient TCP segment sizing Eric Dumazet
2012-04-24 19:49 ` Maciej Żenczykowski
2012-04-24 20:10   ` Eric Dumazet
2012-04-24 21:50     ` Maciej Żenczykowski
2012-04-24 21:51       ` Maciej Żenczykowski
2012-04-25  5:32       ` Eric Dumazet
2012-04-25  7:34         ` Maciej Żenczykowski
2012-04-25  9:20           ` Tore Anderson
2012-04-25  9:38             ` Eric Dumazet
2012-04-25  9:51               ` Tore Anderson
2012-04-25  9:52               ` Maciej Żenczykowski
2012-04-25 10:02               ` Eric Dumazet
2012-04-25 18:39                 ` David Miller
2012-04-25  9:48             ` Maciej Żenczykowski
2012-04-25 10:04               ` Tore Anderson
2012-04-25 10:15                 ` Eric Dumazet
2012-04-25 10:30                 ` Maciej Żenczykowski
2012-04-25 10:44                   ` Eric Dumazet
2012-04-26 10:32                     ` Tore Anderson [this message]
2012-04-25 10:45                   ` Tore Anderson
2012-04-25 11:02                     ` Maciej Żenczykowski
2012-04-25 11:49                       ` Tore Anderson
2012-04-25 11:55                         ` Maciej Żenczykowski
2012-04-27  4:03 ` David Miller

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=0e135d034e75f56cb6d5318595c17023@greed.fud.no \
    --to=tore@fud.no \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=maze@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=therbert@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).