netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Miller <davem@davemloft.net>
To: eric.dumazet@gmail.com
Cc: herbert@gondor.apana.org.au, thomas.jarosch@intra2net.com,
	netdev@vger.kernel.org, edumazet@google.com,
	steffen.klassert@secunet.com, bhutchings@solarflare.com
Subject: Re: tcp: Do not apply TSO segment limit to non-TSO packets
Date: Fri, 02 Jan 2015 15:36:55 -0500 (EST)	[thread overview]
Message-ID: <20150102.153655.1853692198479011402.davem@davemloft.net> (raw)
In-Reply-To: <1420223040.32621.6.camel@edumazet-glaptop2.roam.corp.google.com>

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Fri, 02 Jan 2015 10:24:00 -0800

> On Thu, 2015-01-01 at 00:42 +1100, Herbert Xu wrote:
>> Firstly not many people test non-TSO code paths anymore so bugs
>> are likely to persist for a long time there.  Perhaps it's time
>> to remove the non-TSO code path altogether? The GSO code path
>> should provide enough speed-up in terms of boosting the effective
>> MTU to offset the cost of copying.
> 
>> Secondly why are we dealing with hardware TSO segment limits
>> by limiting the size of the TSO packet in the TCP stack? Surely
>> in this case GSO is free since there won't be any copying?
> 
> It might depends on the device capabilities.
> 
> Non TSO/GSO path is known to be better for devices unable to perform TX
> checksumming, as we compute the checksum at the time we copy data from
> user to kernel (csum_and_copy_from_user() from tcp_sendmsg())).

Non-SG capable devices suffer in this scenerio as well.

  reply	other threads:[~2015-01-02 20:37 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-29 11:44 [bisected] xfrm: TCP connection initiating PMTU discovery stalls on v3.12+ Thomas Jarosch
2014-12-01 10:25 ` Herbert Xu
2014-12-01 11:20   ` Thomas Jarosch
2014-12-31 13:39   ` tcp: Do not apply TSO segment limit to non-TSO packets Herbert Xu
2014-12-31 13:42     ` Herbert Xu
2015-01-02 18:24       ` Eric Dumazet
2015-01-02 20:36         ` David Miller [this message]
2015-01-02 22:01           ` Herbert Xu
2015-01-02 22:06             ` David Miller
2015-01-02 22:09               ` Herbert Xu
2015-01-02 21:13     ` David Miller
2015-01-16 10:45     ` Thomas Jarosch
2015-01-16 10:50       ` Herbert Xu
2015-01-16 11:03         ` Thomas Jarosch
2015-01-19 13:39     ` Thomas Jarosch
     [not found]       ` <CANn89i+U-PFbuUrp08s3Ec8BmjPFq1zj8Aj2=vPVO4-iiLkTuw@mail.gmail.com>
2015-01-19 22:36         ` Herbert Xu
2015-01-19 22:38           ` Eric Dumazet
2015-01-19 22:40             ` Herbert Xu
2014-12-01 13:17 ` [bisected] xfrm: TCP connection initiating PMTU discovery stalls on v3.12+ Wolfgang Walter
2014-12-01 16:41   ` Wolfgang Walter
2014-12-05 12:09     ` Wolfgang Walter
2014-12-05 13:26       ` Eric Dumazet
2014-12-08 22:20         ` Wolfgang Walter
2014-12-09  8:54           ` Thomas Jarosch
2014-12-09 14:26             ` [bisected] xfrm: TCP connection initiating PMTU discovery stalls on v3 Eric Dumazet
2014-12-09 14:49               ` Thomas Jarosch
2014-12-09 20:36               ` Wolfgang Walter
2014-12-09 21:40                 ` Eric Dumazet
2014-12-10 18:34                   ` Wolfgang Walter
2014-12-10 19:10                     ` Eric Dumazet
2014-12-11  0:36                       ` Wolfgang Walter
2014-12-12 16:58                       ` Wolfgang Walter
2014-12-12 17:27                         ` Eric Dumazet
2014-12-12 20:31                           ` Wolfgang Walter
2014-12-12 21:30                             ` Thomas Jarosch
2014-12-12 22:31                               ` Eric Dumazet
2014-12-12 23:47                                 ` Wolfgang Walter
2014-12-13  0:15                                   ` Eric Dumazet
2014-12-13  0:43                                     ` Wolfgang Walter
2014-12-15 18:04                                     ` Wolfgang Walter

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=20150102.153655.1853692198479011402.davem@davemloft.net \
    --to=davem@davemloft.net \
    --cc=bhutchings@solarflare.com \
    --cc=edumazet@google.com \
    --cc=eric.dumazet@gmail.com \
    --cc=herbert@gondor.apana.org.au \
    --cc=netdev@vger.kernel.org \
    --cc=steffen.klassert@secunet.com \
    --cc=thomas.jarosch@intra2net.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).