All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rick Jones <rick.jones2@hp.com>
To: Lennert Buytenhek <buytenh@wantstofly.org>
Cc: netdev@vger.kernel.org, Ashish Karkare <akarkare@marvell.com>,
	Nicolas Pitre <nico@cam.org>
Subject: Re: using software TSO on non-TSO capable netdevices
Date: Thu, 31 Jul 2008 10:00:32 -0700	[thread overview]
Message-ID: <4891EFB0.4040003@hp.com> (raw)
In-Reply-To: <20080730235004.GC21999@xi.wantstofly.org>

Lennert Buytenhek wrote:
> When using sendfile() to send a GiB worth of zeroes over a single TCP
> connection to another host on a 100 Mb/s network, with a vanilla
> 2.6.27-rc1 kernel, this runs as expected at wire speed, taking the
> following amount of CPU time per test:
> 
> 	sys     0m5.410s
> 	sys     0m5.380s
> 	sys     0m5.620s
> 	sys     0m5.360s

That's output from "time" run against your test program right?  Are 
folks confident that will account for all the CPU time consumed on 
behalf of that program?  FWIW netperf is unwilling to make such an 
assumption.  Perhaps I and it are being too paranoid but there you have 
it :)

rick jones
fwiw, there is a TCP_SENDFILE test in netperf

  parent reply	other threads:[~2008-07-31 17:00 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-07-30 23:50 using software TSO on non-TSO capable netdevices Lennert Buytenhek
2008-07-30 23:56 ` David Miller
2008-07-31  0:41   ` Lennert Buytenhek
2008-07-31  1:10     ` David Miller
2008-07-31  1:45       ` Lennert Buytenhek
2008-07-31  3:54         ` Herbert Xu
2008-07-31  9:45           ` Lennert Buytenhek
2008-07-31 10:55             ` Herbert Xu
2008-07-31 12:37               ` Lennert Buytenhek
2008-07-31 12:59                 ` Herbert Xu
2008-08-03  8:23                   ` David Miller
2008-07-31  7:34       ` Ilpo Järvinen
2008-07-31  9:50         ` Lennert Buytenhek
2008-07-31 10:27           ` Ilpo Järvinen
2008-07-31  2:29     ` Herbert Xu
2008-07-31  2:36       ` Lennert Buytenhek
2008-07-31  3:03         ` Herbert Xu
2008-07-31  6:55           ` Ilpo Järvinen
2008-07-31  9:39             ` Lennert Buytenhek
2008-07-31 10:14           ` Lennert Buytenhek
2008-07-31 10:16             ` David Miller
2008-07-31 12:25               ` Lennert Buytenhek
2008-07-31 12:35                 ` David Miller
2008-07-31 13:19                   ` Ben Hutchings
2008-07-31 13:27                     ` Herbert Xu
2008-08-03  8:19                       ` David Miller
2008-08-03  8:55                         ` Herbert Xu
2008-08-07  6:07                           ` David Miller
2008-08-07  6:15                             ` Herbert Xu
2008-09-12  4:08                               ` David Miller
2008-08-07 11:50                             ` Lennert Buytenhek
2008-08-07 20:32                             ` Rick Jones
2008-08-07 22:44                               ` David Miller
2008-07-31 17:00 ` Rick Jones [this message]
2008-07-31 17:45   ` Lennert Buytenhek

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=4891EFB0.4040003@hp.com \
    --to=rick.jones2@hp.com \
    --cc=akarkare@marvell.com \
    --cc=buytenh@wantstofly.org \
    --cc=netdev@vger.kernel.org \
    --cc=nico@cam.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.