linux-embedded.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Johannes Stezenbach <js@sig21.net>
To: Jamie Lokier <jamie@shareable.org>
Cc: linux-embedded@vger.kernel.org, netdev@vger.kernel.org
Subject: Re: 100Mbit ethernet performance on embedded devices
Date: Fri, 28 Aug 2009 16:41:38 +0200	[thread overview]
Message-ID: <20090828144138.GB7375@sig21.net> (raw)
In-Reply-To: <20090820125649.GA29029@sig21.net>

On Thu, Aug 20, 2009 at 02:56:49PM +0200, Johannes Stezenbach wrote:
> On Wed, Aug 19, 2009 at 04:35:34PM +0100, Jamie Lokier wrote:
> > Johannes Stezenbach wrote:
> > > 
> > >   TCP RX ~70Mbit/sec  (iperf -s on SoC, iperf -c on destop PC)
> > >   TCP TX ~56Mbit/sec  (iperf -s on destop PC, iperf -c o SoC)
> > > 
> > > The CPU load during the iperf test is around
> > > 1% user, 44% system, 4% irq, 48% softirq, with 7500 irqs/sec.
> > > 
> > > The kernel used in these measurements does not have iptables
> > > support, I think packet filtering will slow it down noticably,
> > > but I didn't actually try.  The ethernet driver uses NAPI,
> > > but it doesn't seem to be a win judging from the irq/sec number.
> > 
> > You should see far fewer interrupts if NAPI was working properly.
> > Rather than NAPI not being a win, it looks like it's not active at
> > all.
> > 
> > 7500/sec is close to the packet rate, for sending TCP with
> > full-size ethernet packages over a 100Mbit ethernet link.
> 
> From debug output I can see that NAPI works in principle, however
> the timing seems to be such that ->poll() almost always completes
> before the next packet is received.  I followed the NAPI_HOWTO.txt
> which came with the 2.6.20 kernel.  The delay between irq ->
> netif_rx_schedule() -> NET_RX_SOFTIRQ ->  ->poll()  doesn't seem
> to be long enough.  But of course my understanding of NAPI is
> very limited, probably I missed something...

It would've been nice to get a comment on this.  Yeah I know,
old kernel, non-mainline driver...

On this platform NAPI seems to be a win when receiving small packets,
but not for a single max-bandwidth TCP stream.  The folks at
stlinux.com seem to be using a dedicated hw timer to delay
the NAPI poll() calls:
http://www.stlinux.com/drupal/kernel/network/stmmac-optimizations

This of course adds some latency to the packet processing,
however in the single TCP stream case this wouldn't matter.


Thanks,
Johannes

  reply	other threads:[~2009-08-28 14:41 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-08-19 14:50 100Mbit ethernet performance on embedded devices Johannes Stezenbach
2009-08-19 15:05 ` Ben Hutchings
2009-08-19 15:35 ` Jamie Lokier
2009-08-20 12:56   ` Johannes Stezenbach
2009-08-28 14:41     ` Johannes Stezenbach [this message]
2009-08-28 17:35       ` Mark Brown
2009-08-29  7:05       ` Simon Holm Thøgersen
2009-08-27 15:38 ` H M Thalib
2009-08-28 14:26   ` Johannes Stezenbach
2009-09-02  5:09 ` Aras Vaichas
2009-09-02 19:35 ` David Acker

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=20090828144138.GB7375@sig21.net \
    --to=js@sig21.net \
    --cc=jamie@shareable.org \
    --cc=linux-embedded@vger.kernel.org \
    --cc=netdev@vger.kernel.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 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).