public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Benjamin LaHaise <bcrl@redhat.com>
To: "David S. Miller" <davem@redhat.com>
Cc: whitney@math.berkeley.edu, rgooch@ras.ucalgary.ca,
	linux-kernel@vger.kernel.org
Subject: Re: Broadcom 5700/5701 Gigabit Ethernet Adapters
Date: Sun, 10 Mar 2002 22:15:30 -0500	[thread overview]
Message-ID: <20020310221530.A28821@redhat.com> (raw)
In-Reply-To: <200203110205.g2B25Ar05044@adsl-209-76-109-63.dsl.snfc21.pacbell.net> <20020310.180456.91344522.davem@redhat.com> <20020310212210.A27870@redhat.com> <20020310.183033.67792009.davem@redhat.com>
In-Reply-To: <20020310.183033.67792009.davem@redhat.com>; from davem@redhat.com on Sun, Mar 10, 2002 at 06:30:33PM -0800

On Sun, Mar 10, 2002 at 06:30:33PM -0800, David S. Miller wrote:
> Syskonnect sk98 with jumbo frames gets ~107MB/sec TCP bandwidth
> without NAPI, there is no reason other cards cannot go full speed as
> well.
> 
> NAPI is really only going to help with high packet rates not with
> thinks like raw bandwidth tests.

Well, the thing that hurts the 83820 is that its interrupt 
mitigation capabilities are rather limited.  This is where napi 
helps: by turning off the rx interrupt for the duration of packet 
processing, cpu cycles aren't wasted on excess rx irqs.

As to the lack of bandwidth, it stems from far too much interrupt 
overhead and the currently braindead attempt at irq mitigation.  
Since the last time I worked on it, a number of potential techniques 
have come up that should bring it into the 100MB realm (assuming it 
doesn't get trampled on by ksoftirqd).

		-ben
-- 
"A man with a bass just walked in,
 and he's putting it down
 on the floor."

  reply	other threads:[~2002-03-11  3:15 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-02-15 12:58 Broadcom 5700/5701 Gigabit Ethernet Adapters Frank Elsner
2002-02-15 13:06 ` Jeff Garzik
2002-02-15 14:36   ` Thomas Langås
2002-02-15 14:45     ` Jeff Garzik
2002-02-15 14:55       ` Thomas Langås
2002-02-15 15:00         ` Jeff Garzik
2002-02-15 15:36           ` Thomas Langås
2002-02-15 20:20     ` David S. Miller
2002-02-19  0:18       ` Thomas Langås
2002-02-15 15:03   ` Jason Lunz
2002-02-26 20:09     ` Jes Sorensen
2002-02-15 14:43 ` J Sloan
2002-02-27 14:12   ` Stephan von Krawczynski
2002-03-10 15:33     ` Harald Welte
2002-03-10 19:10       ` Jeff Garzik
2002-03-10 21:35         ` Harald Welte
2002-03-11  0:41       ` David S. Miller
2002-03-11  0:55         ` Richard Gooch
2002-03-11  1:03           ` David S. Miller
2002-03-11  1:14             ` Richard Gooch
2002-03-11  1:31               ` David S. Miller
2002-03-11  1:31               ` Jeff Garzik
2002-03-11  2:05               ` Wayne Whitney
2002-03-11  2:04                 ` David S. Miller
2002-03-11  2:10                   ` Richard Gooch
2002-03-11  2:15                     ` David S. Miller
2002-03-11  2:22                   ` Benjamin LaHaise
2002-03-11  2:28                     ` Mike Fedyk
2002-03-11  2:32                       ` David S. Miller
2002-03-11  2:30                     ` David S. Miller
2002-03-11  3:15                       ` Benjamin LaHaise [this message]
2002-03-11  4:20                         ` Michael Clark
2002-03-11  4:28                           ` Benjamin LaHaise
2002-03-11 19:48                       ` Richard Gooch
2002-03-12  6:04                         ` David S. Miller
2002-03-12  6:20                           ` Richard Gooch
2002-03-12  5:40                       ` [patch] ns83820 0.17 (Re: Broadcom 5700/5701 Gigabit Ethernet Adapters) Benjamin LaHaise
2002-03-12 11:00                         ` Michael Clark
2002-03-12 11:15                           ` [patch] ns83820 0.17 David S. Miller
2002-03-12 13:03                             ` dean gaudet
2002-03-12 13:03                               ` David S. Miller
2002-03-12 18:12                             ` Trever L. Adams
2002-03-12 18:17                               ` David S. Miller
2002-03-12 18:31                                 ` Trever L. Adams
2002-03-12 19:52                                 ` pjd
2002-03-12 19:06                               ` Charles Cazabon
2002-03-12 19:39                               ` Pedro M. Rodrigues
2002-03-14  9:54                         ` [patch] ns83820 0.17 (Re: Broadcom 5700/5701 Gigabit Ethernet Adapters) Jeff Garzik
2002-03-14 20:37                           ` Benjamin LaHaise
2002-03-15  1:02                             ` Jeff Garzik
2002-03-15  8:56                             ` Daniel Phillips
2002-04-08  5:14                         ` Richard Gooch
2002-03-11  2:04             ` Broadcom 5700/5701 Gigabit Ethernet Adapters Ben Collins
2002-03-11  2:13               ` David S. Miller
2002-03-11  2:22                 ` Ben Collins
2002-03-11  2:31                   ` David S. Miller
2002-03-21 20:39             ` Thomas Langås
2002-03-11  6:07         ` Harald Welte

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=20020310221530.A28821@redhat.com \
    --to=bcrl@redhat.com \
    --cc=davem@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rgooch@ras.ucalgary.ca \
    --cc=whitney@math.berkeley.edu \
    /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