linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: Solio Sarabia <solio.sarabia@intel.com>
Cc: netdev@vger.kernel.org, davem@daveloft.net, kys@microsoft.com,
	shiny.sebastian@intel.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] net-sysfs: export gso_max_size attribute
Date: Thu, 23 Nov 2017 21:18:37 -0800	[thread overview]
Message-ID: <20171123211837.0f5efa8b@xeon-e3> (raw)
In-Reply-To: <1511397041-27994-1-git-send-email-solio.sarabia@intel.com>

On Wed, 22 Nov 2017 16:30:41 -0800
Solio Sarabia <solio.sarabia@intel.com> wrote:

> The netdevice gso_max_size is exposed to allow users fine-control on
> systems with multiple NICs with different GSO buffer sizes, and where
> the virtual devices like bridge and veth, need to be aware of the GSO
> size of the underlying devices.
> 
> In a virtualized environment, setting the right GSO sizes for physical
> and virtual devices makes all TSO work to be on physical NIC, improving
> throughput and reducing CPU util. If virtual devices send buffers
> greater than what NIC supports, it forces host to do TSO for buffers
> exceeding the limit, increasing CPU utilization in host.
> 
> Suggested-by: Shiny Sebastian <shiny.sebastian@intel.com>
> Signed-off-by: Solio Sarabia <solio.sarabia@intel.com>
> ---
> In one test scenario with Hyper-V host, Ubuntu 16.04 VM, with Docker
> inside VM, and NTttcp sending 40 Gbps from one container, setting the
> right gso_max_size values for all network devices in the chain, reduces
> CPU overhead about 3x (for the sender), since all TSO work is done by
> physical NIC.
> 
>  net/core/net-sysfs.c | 30 ++++++++++++++++++++++++++++++
>  1 file changed, 30 insertions(+)


You probably should expose gso_max_segs as well.

  parent reply	other threads:[~2017-11-24  5:18 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-23  0:30 [PATCH] net-sysfs: export gso_max_size attribute Solio Sarabia
2017-11-23 21:48 ` Solio Sarabia
2017-11-24  5:18 ` Stephen Hemminger [this message]
2017-11-24 17:14 ` David Ahern
2017-11-24 18:32   ` Eric Dumazet
2017-11-24 18:43     ` David Ahern
2017-11-24 18:52       ` Eric Dumazet
2017-11-27 21:47     ` Solio Sarabia

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=20171123211837.0f5efa8b@xeon-e3 \
    --to=stephen@networkplumber.org \
    --cc=davem@daveloft.net \
    --cc=kys@microsoft.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=shiny.sebastian@intel.com \
    --cc=solio.sarabia@intel.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).