Linux MIPS Architecture development
 help / color / mirror / Atom feed
From: "Matthew Dharm" <mdharm@momenco.com>
To: "'Samuel Agarwal'" <sagarwal10@hotmail.com>,
	<linux-mips@linux-mips.org>,
	<linuxppc-embedded@lists.linuxppc.org>
Subject: RE: Intel Pro 82546 chipset performance
Date: Wed, 14 Jan 2004 10:32:32 -0800	[thread overview]
Message-ID: <001101c3dacc$c5e182a0$7200a8c0@internal.momenco.com> (raw)
In-Reply-To: <BAY10-F117sntMQN8zy00008864@hotmail.com>

We use this chip (actually, the entire family) on many of our products (PPC,
x86, and MIPS).

Our data indicates that the chip can sustain wire-saturation with minimum
size packets.  However, many CPUs cannot keep up with the load created by
interrupts coming in at that rate.

There are probably some driver enhancements (interrupt combining, etc.)
which could be made to improve the situation.  But it's not a silicon
problem.

Matt

--
Matthew D. Dharm                            Senior Software Designer
Momentum Computer Inc.                      1815 Aston Ave.  Suite 107
(760) 431-8663 X-115                        Carlsbad, CA 92008-7310
Momentum Works For You                      www.momenco.com


> -----Original Message-----
> From: linux-mips-bounce@linux-mips.org 
> [mailto:linux-mips-bounce@linux-mips.org] On Behalf Of Samuel Agarwal
> Sent: Tuesday, January 13, 2004 10:37 PM
> To: linux-mips@linux-mips.org; linuxppc-embedded@lists.linuxppc.org
> Subject: Intel Pro 82546 chipset performance
> 
> 
> Hi
> 
> We're trying to use the Intel Pro 1000 adapters with the 
> 82546 chipset for a packet switching application. Has anyone 
> investigated performance on this chipset? With minimum sized 
> ethernet frames, the maximum throughput I can get is about 1 
> million packets per second (versus a 1.4 mill for line 
> rate
> on a gigethernet) and it looks that performance is skewed 
> towards receiving traffic (rather than sending out).
> 
> If anyone has any opinions or thoughts I'd appreciate them. 
> It's hard to get information out of Intel.
> 
> _________________________________________________________________
> Find out everything you need to know about Las Vegas here for 
> that getaway.  
> http://special.msn.com/msnbc/vivalasvegas.armx
> 
> 

WARNING: multiple messages have this Message-ID (diff)
From: "Matthew Dharm" <mdharm@momenco.com>
To: 'Samuel Agarwal' <sagarwal10@hotmail.com>,
	linux-mips@linux-mips.org, linuxppc-embedded@lists.linuxppc.org
Subject: RE: Intel Pro 82546 chipset performance
Date: Wed, 14 Jan 2004 10:32:32 -0800	[thread overview]
Message-ID: <001101c3dacc$c5e182a0$7200a8c0@internal.momenco.com> (raw)
Message-ID: <20040114183232.pZsGaa2EyebO2UaUydjOJXQVUK6Qo2sJBjsWLvDT6cI@z> (raw)
In-Reply-To: <BAY10-F117sntMQN8zy00008864@hotmail.com>

We use this chip (actually, the entire family) on many of our products (PPC,
x86, and MIPS).

Our data indicates that the chip can sustain wire-saturation with minimum
size packets.  However, many CPUs cannot keep up with the load created by
interrupts coming in at that rate.

There are probably some driver enhancements (interrupt combining, etc.)
which could be made to improve the situation.  But it's not a silicon
problem.

Matt

--
Matthew D. Dharm                            Senior Software Designer
Momentum Computer Inc.                      1815 Aston Ave.  Suite 107
(760) 431-8663 X-115                        Carlsbad, CA 92008-7310
Momentum Works For You                      www.momenco.com


> -----Original Message-----
> From: linux-mips-bounce@linux-mips.org 
> [mailto:linux-mips-bounce@linux-mips.org] On Behalf Of Samuel Agarwal
> Sent: Tuesday, January 13, 2004 10:37 PM
> To: linux-mips@linux-mips.org; linuxppc-embedded@lists.linuxppc.org
> Subject: Intel Pro 82546 chipset performance
> 
> 
> Hi
> 
> We're trying to use the Intel Pro 1000 adapters with the 
> 82546 chipset for a packet switching application. Has anyone 
> investigated performance on this chipset? With minimum sized 
> ethernet frames, the maximum throughput I can get is about 1 
> million packets per second (versus a 1.4 mill for line 
> rate
> on a gigethernet) and it looks that performance is skewed 
> towards receiving traffic (rather than sending out).
> 
> If anyone has any opinions or thoughts I'd appreciate them. 
> It's hard to get information out of Intel.
> 
> _________________________________________________________________
> Find out everything you need to know about Las Vegas here for 
> that getaway.  
> http://special.msn.com/msnbc/vivalasvegas.armx
> 
> 

  reply	other threads:[~2004-01-14 18:32 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-01-14  6:36 Intel Pro 82546 chipset performance Samuel Agarwal
2004-01-14 18:32 ` Matthew Dharm [this message]
2004-01-14 18:32   ` Matthew Dharm

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='001101c3dacc$c5e182a0$7200a8c0@internal.momenco.com' \
    --to=mdharm@momenco.com \
    --cc=linux-mips@linux-mips.org \
    --cc=linuxppc-embedded@lists.linuxppc.org \
    --cc=sagarwal10@hotmail.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