From: Robert Latham <robl@mcs.anl.gov>
To: linux-kernel@vger.kernel.org
Subject: tigon3: 2466 and bad performance at 32bits/33mhz ?
Date: Tue, 6 Aug 2002 00:35:08 -0500 [thread overview]
Message-ID: <20020806053508.GJ25554@mcs.anl.gov> (raw)
I've got a 3com 3C996B-T (copper) and a tyan 2466 motherboard. The
kernel is 2.4.19-rc1-ac2, and i'm using the tg3 v0.99 (Jun 11, 2002).
When i run the 3C996B-T in the 64-bit/66-MHz slot, i get great
performance. When i run the card in the 32-bit/66-MHz slot, i get
pretty bad performance. Here's a netpipe graph to demonstrate:
http://www-unix.mcs.anl.gov/~robl/storage_box/tg3-32_33.2.png
The fast curve is with the 3c996B-T in the 64/66 slot. It peaks out
around 850 Mbps. The slow curve is the same card in the 32/33 slot.
It peaks around 180 Mbps.
Is a maximum throughput of ~180 MBits/sec all that can be hoped for? I
don't have the hardware to verify, but
http://www.cs.uni.edu/~gray/gig-over-copper/#3Com%203c996BT:|outline
offers the datapoint of 436 MBits/sec in an oldish dell optiplex ( old
being whenever p3 500 mhz machines were still made)
So, am i looking at a hardware limitation? driver quirk? I'm open to
any suggestions.
I'll keep an eye on the web archives, but please CC me on any replies.
thanks
==rob
--
Rob Latham
Mathematics and Computer Science Division A215 0178 EA2D B059 8CDF
Argonne National Labs, IL USA B29D F333 664A 4280 315B
next reply other threads:[~2002-08-06 5:31 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-08-06 5:35 Robert Latham [this message]
2002-08-06 8:48 ` tigon3: 2466 and bad performance at 32bits/33mhz ? Willy Tarreau
2002-08-06 16:47 ` Ben Greear
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=20020806053508.GJ25554@mcs.anl.gov \
--to=robl@mcs.anl.gov \
--cc=linux-kernel@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 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.